Plant Biomass Conversion

The development of biofuels and green chemistry has escalated rapidly in recent years, and with increased interest there is now a great demand for scientific information on the development of biomass crops and conversion of biomass into fuels and chemicals. Plant Biomass Conversion provides coverage of a broad range of key topics that are directly tied to the sustainable and profitable development of the biofuels industry. Plant Biomass Conversion covers topics ranging from the development of dedicated biomass crops to the evolution of conversion processes. Chapters also look at sustainability issues and the economic considerations to profitably develop fuels and industrial chemicals from biomass. Bringing together contributions from scientific researchers and industry personnel, Plant Biomass Conversion will provide the reader with thorough understanding of this evolving industry. Broad ranging in scope and written in a succinct, scientific style, Plant Biomass Conversion will be an essential reference for all researchers and industrial personnel interested in the production and development of biofuels.

KEY FEATURES: • Provides an overview of plant biomass conversion for plant and crop scientists, biofuels researchers, and industry personnel • Addresses both development of biomass crops and conversion techniques • Discusses sustainability and economic issues around the development of bio-based fuels and chemicals

EDITORS: Elizabeth E. Hood is a Distinguished Professor of Agriculture at Arkansas State University. Peter Nelson is Co-Founder and Director of Business Development with BioDimensions, Inc. Memphis, Tennessee. Randall Powell is Technology Consultant and Program Manager for Sugar Platform with BioDimensions, Inc, Memphis Tennessee.

RELATED TITLES: Biofuels from Agricultural Wastes and Byproducts Editors: Hans P. Blaschek, Thaddeus Ezeji, Jurgen Sheffren 9780813802527

ISBN: 978-0-8138-1694-4

www.wiley.com/wiley-blackwell

HOOD, NELSON, & POWELL

Anaerobic Biotechnology for Bioenergy Production: Principles and Applications Samir Kumar Khanal 9780813823461

PLANT BIOMASS CONVERSION

PLANT BIOMASS CONVERSION

BIOMASS AND BIOFUELS SERIES

PLANT BIOMASS CONVERSION

ELIZABETH E. HOOD, PETER NELSON, & RANDALL POWELL

P1: SFK/UKS fm

P2: SFK/UKS

BLBS075-Hood

QC: SFK/UKS

December 18, 2010

T1: SFK 18:21

Trim: 244mm×172mm

P1: SFK/UKS fm

P2: SFK/UKS

BLBS075-Hood

QC: SFK/UKS

December 18, 2010

T1: SFK 18:21

Trim: 244mm×172mm

Plant Biomass Conversion

P1: SFK/UKS fm

P2: SFK/UKS

BLBS075-Hood

QC: SFK/UKS

December 18, 2010

T1: SFK 18:21

Trim: 244mm×172mm

P1: SFK/UKS fm

P2: SFK/UKS

BLBS075-Hood

QC: SFK/UKS

December 18, 2010

T1: SFK 18:21

Trim: 244mm×172mm

Plant Biomass Conversion Editors Elizabeth E. Hood Arkansas Biosciences Institute Arkansas State University Jonesboro, Arkansas, USA

Peter Nelson BioDimensions, Inc. Memphis, Tennessee, USA

Randall Powell BioDimensions, Inc. Memphis, Tennessee, USA

A John Wiley & Sons, Ltd., Publication

P1: SFK/UKS fm

P2: SFK/UKS

BLBS075-Hood

QC: SFK/UKS

December 18, 2010

T1: SFK 18:21

Trim: 244mm×172mm

C 2011 by John Wiley & Sons Inc. This edition first published 2011


Wiley-Blackwell is an imprint of John Wiley & Sons, formed by the merger of Wiley’s global Scientific, Technical and Medical business with Blackwell Publishing. Registered office:

John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, UK

Editorial offices:

2121 State Avenue, Ames, Iowa 50014-8300, USA The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, UK 9600 Garsington Road, Oxford, OX4 2DQ, UK

For details of our global editorial offices, for customer services and for information about how to apply for permission to reuse the copyright material in this book please see our website at www.wiley.com/wiley-blackwell. Authorization to photocopy items for internal or personal use, or the internal or personal use of specific clients, is granted by Blackwell Publishing, provided that the base fee is paid directly to the Copyright Clearance Center, 222 Rosewood Drive, Danvers, MA 01923. For those organizations that have been granted a photocopy license by CCC, a separate system of payments has been arranged. The fee codes for users of the Transactional Reporting Service are ISBN-13: 978-0-8138-1694-4/2011. Designations used by companies to distinguish their products are often claimed as trademarks. All brand names and product names used in this book are trade names, service marks, trademarks or registered trademarks of their respective owners. The publisher is not associated with any product or vendor mentioned in this book. This publication is designed to provide accurate and authoritative information in regard to the subject matter covered. It is sold on the understanding that the publisher is not engaged in rendering professional services. If professional advice or other expert assistance is required, the services of a competent professional should be sought. Library of Congress Cataloging-in-Publication Data Plant biomass conversion / editors: Elizabeth E. Hood, Peter Nelson, Randall Powell. p. cm. Includes bibliographical references and index. ISBN 978-0-8138-1694-4 (hardcover : alk. paper) 1. Plant biomass. 2. Biomass conversion. 3. Biomass conversion–Environmental aspects. 4. Biomass energy. I. Hood, Elizabeth E. II. Nelson, Peter (Peter Allan), 1974– III. Powell, Randall Worth. TP248.27.P55P554 2011 662 .88–dc22 2010040942 A catalogue record for this book is available from the British Library. This book is published in the following electronic formats: ePDF 9780470959053; Wiley Online Library 9780470959138; ePub 9780470959091 R Inc., New Delhi, India Set in 10/11.5 pt Times New Roman by Aptara


1 2011

P1: SFK/UKS fm

P2: SFK/UKS

BLBS075-Hood

QC: SFK/UKS

December 18, 2010

T1: SFK 18:21

Trim: 244mm×172mm

Contents

Contributors Preface 1 The Bioeconomy: A New Era of Products Derived from Renewable Plant-Based Feedstocks Peter Nelson, Elizabeth Hood, and Randall Powell 1.1 1.2 1.3

1.4 1.5 1.6 1.7 1.8 1.9 1.10

Introduction Market Opportunity for Biofuels and Biobased Products Feedstocks 1.3.1 Biobased Feedstock Availability and Issues 1.3.2 Characterization of Lignocellulosic Feedstocks 1.3.3 The Role of Agricultural Biotechnology 1.3.4 Biomass Agricultural Equipment Development The Biochemical Technology Platform Investment and Major Players The Role of the Farmer Opportunities for Rural Development Environmental Benefits Economic Comparison of the Biochemical and Thermochemical Technology Platforms Conclusions and Future Prospects References

2 Agricultural Residues James Hettenhaus 2.1 2.2

Introduction 2.1.1 Key Issues Feedstock Supply 2.2.1 Residue Markets 2.2.2 Harvest Window 2.2.3 Residue Removal 2.2.4 Residue Management 2.2.5 Ag Equipment Needs 2.2.6 Operating Costs 2.2.7 Residue Nutrient Value 2.2.8 Land for Energy Crops

xi xiii

3 3 5 6 6 8 9 11 11 12 14 16 17 17 18 19 21 21 22 23 26 27 27 28 29 33 33 33 v

P1: SFK/UKS fm

P2: SFK/UKS

BLBS075-Hood

QC: SFK/UKS

December 18, 2010

T1: SFK 18:21

Trim: 244mm×172mm

Contents

vi

2.3

2.4

2.2.9 Farmer Outlook 2.2.10 Crop Research and Development Feedstock Logistics 2.3.1 Bulk Density 2.3.2 Storage 2.3.3 Regional Biomass Processing Centers Conclusion Endnotes References

3 Growing Systems for Traditional and New Forest-Based Materials Randall Rousseau, Janet Hawkes, Shijie Liu, and Tom Amidon 3.1 3.2 3.3

34 34 34 35 36 43 48 49 49 51

Introduction Natural Regeneration Overall Growing Systems 3.3.1 The Beginnings of Biomass Plantation Production 3.3.2 Short Rotation Woody Crops 3.3.3 Other Types of Hardwood Plantations 3.3.4 Southern Pine New Genetic Tools Agroforestry Products from Woody Biomass 3.6.1 Hemicellulosic Products 3.6.2 Biorefineries Using Woody Biomass 3.6.3 Hot-Water Extraction of Hemicellulose 3.6.4 Wood Extracts: Processing and Conversion 3.6.5 Residual Solid Wood Biomass: Processing and Conversion of the wood mass after extraction, an example Summary References

51 54 54 55 56 59 61 62 63 67 69 71 73 75

4 Dedicated Herbaceous Energy Crops Keat (Thomas) Teoh, Shivakumar Pattada Devaiah, Deborah Vicuna Requesens, and Elizabeth E. Hood

85

3.4 3.5 3.6

3.7

4.1 4.2

4.3

4.4

Introduction Miscanthus 4.2.1 Characteristics That Make Miscanthus a Potential Biomass Crop 4.2.2 Agronomy Sweet Sorghum 4.3.1 Biology of Sweet Sorghum 4.3.2 Production 4.3.3 Potential Yields 4.3.4 Economic and Environmental Advantages of Sweet Sorghum 4.3.5 Production Challenges Switchgrass 4.4.1 Physiology 4.4.2 Switchgrass Ecotypes

78 78 78

85 85 87 87 90 92 92 94 94 96 97 97 98

P1: SFK/UKS fm

P2: SFK/UKS

BLBS075-Hood

QC: SFK/UKS

December 18, 2010

T1: SFK 18:21

Trim: 244mm×172mm

Contents

vii

4.4.3 Advantages 4.4.4 Disadvantages 4.4.5 Yields 4.4.6 Switchgrass as a Bioenergy Crop Conclusions and Future Prospects References

98 99 100 101 101 104

5 Municipal Solid Waste as a Biomass Feedstock David J. Webster

109

4.5

5.1 5.2 5.3

5.4 5.5

5.6

5.7 5.8

Introduction Definitions 5.2.1 Second-Generation Conversion Technologies for Biofuels Disposal Infrastructure and Transfer Stations 5.3.1 Collection Practices 5.3.2 Cost Parameters Waste Generation Waste Characterization 5.5.1 Composition of Generated MSW Prior to Disposal or Processing 5.5.2 Landfilled Waste Compared to Waste Generation 5.5.3 Water in MSW 5.5.4 Heavy Metals in MSW Preparing MSW for Conversion Processing—Mixed Waste Material Recovery Facilities (MRFs) 5.6.1 Presorting 5.6.2 Mechanical Sorting Operations 5.6.3 Manual Sorting Operations 5.6.4 Recovery Rates of the MRF System Cellulosic Content of MSW 5.7.1 Glucose and Ethanol Yields from MSW Framing the Potential References

6 Water Sustainability in Biomass Cropping Systems Jennifer L. Bouldin and Rodney E. Wright 6.1 6.2 6.3

6.4

Introduction Water Use in Bioenergy Production Water Quality Issues in Bioenergy Crops 6.3.1 AGNPS Watershed Model 6.3.2 Water Quality and the Gulf Hypoxic Zone Conclusions—Water Quantity and Quality References

7 Soil Sustainability Issues in Energy Crop Production V. Steven Green 7.1 7.2

Soil Sustainability Concepts Bioenergy Crops and Soil Sustainability

109 110 110 110 112 112 113 114 114 115 116 117 119 121 122 123 123 124 124 125 126 129 129 130 133 135 138 138 139 143 143 145

P1: SFK/UKS fm

P2: SFK/UKS

BLBS075-Hood

QC: SFK/UKS

December 18, 2010

T1: SFK 18:21

Trim: 244mm×172mm

Contents

viii

7.3

7.4 7.5

7.2.1 Crop Residues 7.2.2 Dedicated Energy Crops Resource Use in Biomass Production 7.3.1 Water and Soil 7.3.2 Land Use Soil Sustainability Solutions Conclusion References

8 Fermentation Organisms for 5- and 6-Carbon Sugars Nicholas Dufour, Jeffrey Swana, and Reeta P. Rao 8.1 8.2 8.3 8.4 8.5 8.6

Introduction Fermentation Metabolic Pathways Fermenting Species 8.4.1 Brief Description of Major Species Other Relevant Products Summary Endnotes References

9 Pretreatment Options Bradley A. Saville 9.1

9.2

9.3

9.4

9.5 9.6

Overview of Pretreatment Technologies 9.1.1 History 9.1.2 Mechanistic Assessment of Pretreatment 9.1.3 Severity Factor Concept Pretreatment Classification 9.2.1 Mechanical Pretreatment Processes 9.2.2 Chemical Pretreatment Processes 9.2.3 Thermochemical Pretreatment Processes 9.2.4 Impact on Moisture Content and Hydraulic Load Laboratory vs. Commercial Scale Pretreatment—What Do We Really Know? 9.3.1 Laboratory Studies 9.3.2 Pilot/Demonstration Scale Studies 9.3.3 Limitations of Laboratory-Scale Comparisons of Pretreatment Methods Process Issues and Trade-Offs 9.4.1 Inhibitors 9.4.2 Hydrolysis Efficiency and Enzyme Loadings 9.4.3 Solvent/Catalyst Recovery 9.4.4 Viscosity Reduction and Hydraulic Load Economics Conclusions References

145 146 149 149 150 150 154 154 157 157 159 160 161 175 180 183 183 184 199 199 199 200 203 205 206 206 209 210 211 211 211 214 215 215 218 218 218 220 224 224

P1: SFK/UKS fm

P2: SFK/UKS

BLBS075-Hood

QC: SFK/UKS

December 18, 2010

T1: SFK 18:21

Trim: 244mm×172mm

Contents

10 Enzyme Production Systems for Biomass Conversion John A. Howard, Zivko Nikolov, and Elizabeth E. Hood 10.1 10.2 10.3

10.4 10.5

Introduction The Challenge: Volume and Cost of Enzymes Required Theoretical Ways to Address the Challenge of Quantity of Enzyme and Cost Requirements 10.3.1 Increase Susceptibility for Biomass Deconstruction 10.3.2 Decrease Exogenous Enzyme Load 10.3.3 Increase Accumulation of Enzymes in Production Host Cost of Producing Exogenous Enzymes 10.4.1 Cost Analysis Summary and Future Prospects References

11 Fermentation-Based Biofuels Randy Kramer and Helene Belanger 11.1 11.2

11.3

11.4

11.5

11.6

Introduction First-Generation Biofuels 11.2.1 Starch-Based Ethanol—United States 11.2.2 Sugar-Based Ethanol—Brazil 11.2.3 Biodiesel Policy and Biofuel Implementation Status 11.3.1 North America 11.3.2 South America 11.3.3 Europe 11.3.4 Asia Second-Generation Biofuels 11.4.1 Cellulosic Ethanol 11.4.2 Biobutanol Issues for Biofuels Commercial Success 11.5.1 Transport by Pipeline 11.5.2 Decentralized Production and Local Distribution 11.5.3 Optimized Engine Performance 11.5.4 Value of Biorefinery Co-products Summary References

12 Biobased Chemicals and Polymers Randall W. Powell, Clare Elton, Ross Prestidge, and Helene Belanger 12.1 12.2 12.3

12.4

Introduction Biobased Feedstock Components Biomass Conversion Technologies 12.3.1 Technology Platforms Overview 12.3.2 Lignocellulose Fractionation Overview Biobased Products 12.4.1 Oil-Based Products 12.4.2 Sugar/Starch-Based Products

ix

227 227 227 228 229 231 236 240 242 245 246 255 255 256 256 257 258 260 260 262 262 263 265 265 268 269 269 270 271 272 272 272 275 275 276 277 277 279 287 287 289

P1: SFK/UKS fm

P2: SFK/UKS

BLBS075-Hood

QC: SFK/UKS

December 18, 2010

T1: SFK 18:21

Trim: 244mm×172mm

Contents

x

12.5

12.4.3 Polymer Products 12.4.4 Lignin Products Summary References

13 Carbon Offset Potential of Biomass-Based Energy Gauri-Shankar Guha 13.1

13.2

13.3

13.4

13.5

Emerging Public Interest in Carbon 13.1.1 Overview 13.1.2 Initiatives to Address Anthropogenic Climate Change 13.1.3 GHG Mitigation and Carbon Sequestration Strategies Theory of Carbon Markets 13.2.1 Tradable Permits and the Market for Emissions 13.2.2 Concept of Carbon Markets 13.2.3 Demand and Supply of Carbon Credits Creation of Carbon Markets 13.3.1 Carbon Credits 13.3.2 Global Carbon Trade 13.3.3 Carbon Trading in the United States 13.3.4 The CCX Offset Program Role of Biomass-Based Energy in Carbon Markets 13.4.1 Economic Significance of Bioenergy 13.4.2 Bioenergy Policies, Practices, and Trends 13.4.3 Carbon Offset Opportunities for Biofuels Prognosis of Carbon Markets References

14 Biofuel Economics Daniel Klein-Marcuschamer, Brad Holmes, Blake A. Simmons, and Harvey W. Blanch 14.1 14.2 14.3 14.4 14.5 14.6 14.7 14.8

14.9 14.10 14.11 14.12 14.13 Index

Introduction Production Processes Biomass Transportation and Handling Conversion of Biomass into Sugars Conversion of Sugars into Biofuels Separation and Purification Co-product Handling Major Cost Drivers 14.8.1 Biomass-Associated Costs 14.8.2 Capital Expenses 14.8.3 Operating Costs Risks Policy Support Infrastructure and Vehicle Modifications Conclusions Acknowledgments References

293 299 303 304 311 311 311 311 314 314 314 315 316 317 317 318 318 318 319 319 321 323 324 325 329

329 330 331 332 335 337 337 338 338 340 342 343 345 346 347 348 348 355

P1: SFK/UKS fm

P2: SFK/UKS

BLBS075-Hood

QC: SFK/UKS

December 18, 2010

T1: SFK 18:21

Trim: 244mm×172mm

Contributors

Tom Amidon, SUNY, Syracuse, NY, USA. Helene Belanger, Vertichem Technology Limited, Toronto, Ontario M5H 3B7, Canada. Harvey W. Blanch, Joint Bioenergy Institute, Lawrence Berkeley National Laboratory, and University of California-Berkeley; Berkeley, CA, USA. Jennifer L. Bouldin, Department of Biological Sciences, Ecotoxicology Research Facility, Arkansas State University, Jonesboro, AR, USA. Shivakumar Pattada Devaiah, Arkansas Biosciences Institute, Arkansas State University, Jonesboro, AR, USA. Nicholas Dufour, Worcester Polytechnic Institute, Worcester, MA, USA. Clare Elton, Vertichem Technology Limited, Toronto, Ontario M5H 3B7, Canada. Steven Green, Agricultural Studies, College of Agriculture and Technology, Arkansas State University, Jonesboro, AR, USA. Gauri-Shankar Guha, Economics and Finance Department, Arkansas State University, Jonesboro, AR, USA. Janet Hawkes, HD1, LLC, Ithaca, NY, USA. James Hettenhaus, Chief Executive Assistance, Inc, Charlotte, NC, USA. Brad Holmes, Joint Bioenergy Institute and Lawrence Berkeley National Laboratory, Berkeley, CA, USA. Elizabeth E. Hood, Arkansas Biosciences Institute, Arkansas State University, Jonesboro, AR, USA. John A. Howard, Applied Biotechnology Institute, San Luis Obispo, CA, USA. Daniel Klein-Marcuschamer, Joint Bioenergy Institute and Lawrence Berkeley National Laboratory, Berkeley, CA, USA. xi

P1: SFK/UKS fm

P2: SFK/UKS

BLBS075-Hood

xii

QC: SFK/UKS

December 18, 2010

T1: SFK 18:21

Trim: 244mm×172mm

Contributors

Randy Kramer, Kramer Energy Group, Rapid City, SD, USA. Shijie Liu, SUNY, Syracuse, NY, USA. Peter Nelson, BioDimensions, Inc., Memphis, TN, USA. Zivko Nikolov, Texas A&M University, College Station, TX, USA. Randall W. Powell, BioDimensions, Inc., Memphis, TN, USA. Ross Prestidge, Vertichem Technology Limited, Toronto, Ontario M5H 3B7, Canada. Reeta P. Rao, Worcester Polytechnic Institute, Worcester, MA, USA. Deborah Vicuna Requesens, Arkansas Biosciences Institute, Arkansas State University, Jonesboro, AR, USA. Randall Rousseau, Forestry Department, Mississippi State University, Starkville, MS, USA. Bradley Saville, University of Toronto, Department of Chemical Engineering and Applied Chemistry, Toronto, Canada. Blake A. Simmons, Joint Bioenergy Institute, Lawrence Berkeley National Laboratory, and Sandia National Laboratories, Livermore, CA, USA. Jeffrey Swana, Worcester Polytechnic Institute, Worcester, MA, USA. Keat (Thomas) Teoh, Arkansas Biosciences Institute, Arkansas State University, Jonesboro, AR, USA. David J. Webster, Ark Resources, LLC, Birmingham, AL, USA. Rodney E. Wright, Ecotoxicology Research Facility, Arkansas State University, Jonesboro, AR, USA.

P1: SFK/UKS fm

P2: SFK/UKS

BLBS075-Hood

QC: SFK/UKS

December 18, 2010

T1: SFK 18:21

Trim: 244mm×172mm

Preface

A host of motivations is driving the development of the “renewables” industry—a desire for energy independence in the United States, biodegradable products, global warming, and hopefully, making money. All energy utilized on the earth is ultimately derived from the sun through photosynthesis—the only truly renewable commodity. Capitalizing on this productive process in a balanced way is crucial to our survival as a species. Crude oil represents an ancient capture of carbon and sunlight by plants. However, the rate at which we are utilizing oil releases ancient fixed carbon into the atmosphere and upsets the balance of nature in a way that is unprecedented in global history. Nature’s checks and balances are not able to accommodate this huge increase in carbon from man’s activities. Thus, development of a new source of energy and products is imperative. Many models that describe processes for generating energy from biomass exist. No one book can describe them all. This work focuses on the biochemical (enzymatic) digestion of plant biomass to produce the raw materials that make up plant cell wall polymers. These raw materials can then be used as feedstocks for ethanol and other bio-based products. This volume also is focused on solving the issues for biomass conversion into ethanol and bio-based products now—not the longer-term solutions with modified microbes and modified feedstocks. The chapters review existing technologies and future expectations for those technologies. They describe multiple feedstocks, multiple pretreatment technologies, enzyme production models, fermentation models, and manufacturing of products in biorefineries. While this is a snapshot in time of the state of the industry, this volume should serve as a guide and model for describing what is possible and where the issues are, which must be solved. We hope you enjoy our book. Elizabeth E. Hood, PhD Arkansas State University

xiii

P1: SFK/UKS fm

P2: SFK/UKS

BLBS075-Hood

QC: SFK/UKS

December 18, 2010

T1: SFK 18:21

Trim: 244mm×172mm

P1: SFK/UKS fm

P2: SFK/UKS

BLBS075-Hood

QC: SFK/UKS

December 18, 2010

T1: SFK 18:21

Trim: 244mm×172mm

Plant Biomass Conversion

P1: SFK/UKS fm

P2: SFK/UKS

BLBS075-Hood

QC: SFK/UKS

December 18, 2010

T1: SFK 18:21

Trim: 244mm×172mm

P1: SFK/UKS P2: SFK c01 BLBS075-Hood December 15, 2010

20:45

Trim: 244mm×172mm

Chapter 1

The Bioeconomy: A New Era of Products Derived from Renewable Plant-Based Feedstocks Peter Nelson, Elizabeth Hood, and Randall Powell

1.1 Introduction The first two decades of the 21st century will be marked as the turning point when large investments, technology breakthroughs, and new strategic alliances set the stage for the eventual widespread replacement of fossil feedstocks with renewable, plant-based alternatives for the production of fuels, chemicals, and energy. This is not a new idea, as humankind in the pre-industrial era utilized plant-derived chemicals such as proteins, sugars, and cellulose as the primary feedstocks to make a range of necessary materials and industrial products. However, as non-renewable fossil resources now become increasingly scarce, expensive, and produce negative environmental impacts, the need has never been so great to develop and expand agriculture and forestry as the source of sustainable feedstocks to serve a growing global population. Ultimately, renewable resources must feed, clothe, shelter, fuel, and provide for material goods for the planet’s inhabitants, while also addressing vexing environmental problems including climate change, pollution, access to clean water, and long-term soil health. In the 20th century, incredible technological improvements in agriculture and forestry were made. These advances included dramatic yield increases, drought tolerance, insect resistance in agricultural crops, and new production methodologies such as conservation tillage, which builds soil health and requires less energy. These productivity and environmental improvements offer much promise for a future bioeconomy in which agriculture and forestry will provide the predominant feedstocks for much more than food, feed, and fiber. Agricultural successes such as the Green Revolution have dramatically increased global crop yields, reduced hunger in the developing world, and expanded access to nutritious foods, but are still heavily dependent upon fossil-derived energy and chemicals. Fortunately, agricultural and forestry-based companies and institutions are now collaborating in new ways with industries traditionally dependent on fossil fuels to expand the use of renewable raw materials in a range of manufactured goods.

Plant Biomass Conversion, First Edition. Edited by Elizabeth E. Hood, Peter Nelson and Randall Powell.
C 2011 John Wiley & Sons Inc.

3

P1: SFK/UKS P2: SFK c01 BLBS075-Hood December 15, 2010

4

20:45

Trim: 244mm×172mm

Plant Biomass Conversion

Over the coming decades, this will lead to a more sustainable, closed-loop-systems-based approach to the production of food, energy, and materials. This biobased transition requires development and integration of a range of technologies encompassing energy, process efficiency, environmental compatibility, and even more advanced agricultural production systems. The increasing application of biotechnology tools—previously focused on human health—to improve agricultural crops and practices, enable clean manufacturing processes, and provide sustainable products is the essential catalyst for this transition. A renewable “bioeconomy” is now starting to become a reality, but the concept is not new. In the early days of the 20th century, industrial and agricultural leaders such as Henry Ford and George Washington Carver were proponents and practitioners of the use of plant-derived materials in a range of nonfood products. These innovators demonstrated the commercial utility of renewable biobased feedstocks in hundreds of products, such as automotive composites, glues and adhesives, dyes and inks, plastics, and, of course, biofuels. Unfortunately, the rapid emergence of petroleum as an available and inexpensive feedstock, albeit with unrecognized long-term environmental consequences, drove manufacturers to develop fossil-based rather than renewable products as the initial outputs of the Industrial Revolution. A century later, as the true costs of fossil fuels are realized, renewable feedstocks are re-emerging at commercial scale, largely through innovative partnerships across the value-chain linking agriculture, biotechnology, and the chemical process industries in new ways. Evidence of this transition has become increasingly apparent over the last 30 years as some organizations began decoupling themselves from traditional businesses to focus on agricultural biotechnology. A leading example is Monsanto Company, which has aggressively divested its mainstay fossil-based chemical manufacturing business to focus on the commercial opportunity to develop new agricultural biotechnology traits in commodity crops such as corn, cotton, and soybeans. Major agricultural commodity companies such as Archer Daniels Midland (ADM) and Cargill have also expanded chemical and fuel product offerings based upon their plant-based raw material resources. More recently, a number of multinational chemical companies—notably Dow and DuPont—are pursuing biobased product platforms, with initial commercial products now entering the marketplace. Increasingly, many traditional agricultural commodity companies, fossil-based chemical companies, and newer industrial biotechnology firms are partnering to integrate knowledge of biobased feedstocks, new conversion processes, and operational expertise in order to solve future challenges related to energy and useful materials. The transition to a photosynthesis-based bioeconomy offers commercial opportunity, resource and environmental sustainability, and more equitable global economic development than has been the recent case with fossil resources. However, more intensive agricultural and forestry utilization, and the accompanying deployment of technology must be the products of clear strategic planning and sustainable development practices. If managed correctly, the transition to a renewable-based economy can create new rural and urban opportunities, offer unique environmental solutions, and create wealth. The new economy based on renewable agricultural and forestry raw materials and clean processes will also serve as a major catalyst for realignment of some of the world’s largest companies and institutions, creating many new partnerships across the value-chain. This is an exciting time in which entrepreneurial companies as well as established industries can innovate new farm-to-factory supply chains and establish an early position in the emerging bioeconomy. Although there are multiple technology platforms and an expanding portfolio of biobased raw materials, which will comprise a future biobased economy, this book will focus on biochemical processing technologies, applied primarily to sugar, starch, and lignocellulosic

P1: SFK/UKS P2: SFK c01 BLBS075-Hood December 15, 2010

20:45

Trim: 244mm×172mm

1 The Bioeconomy: A New Era of Products Derived from Renewable Plant-Based Feedstocks

5

biomass feedstocks. This platform will be a significant game changer, lend itself to a wide range of potential chemical products, and provide opportunities for new players in the supply chain. In particular, sugars derived from lignocellulosic biomass represent an abundant feedstock resource that does not compete with food and feed supplies. Commercially viable processes for converting lignocellulosic biomass to biobased products must address two overriding issues: efficient nonseasonal feedstock supply and logistics, and cost-effective deconstruction of cellulose and hemicellulose polymers to fermentable simple sugars. Feedstock supply and cost issues are being addressed by new crops, new harvesting/storage practices, and decentralized processing models. Technology issues to access sugars (and lignin) within lignocellulosic feedstocks are being addressed through new pretreatment options, genetically modified fermentation organisms, biotechnology-enhanced plants, and plant-based enzyme production systems. The development and commercialization of these new technologies and the requisite biobased supply chain will be profiled in this publication. Despite the early stage and dynamic nature of the industrial bioprocessing industry, the authors hope that the current status and perspectives presented will prove beneficial to the diverse industry stakeholders.

1.2 Market Opportunity for Biofuels and Biobased Products Liquid biofuels including corn-based ethanol, as well as advanced biofuels such as biobutanol or cellulosic ethanol, are assured of a growing market over the next 50 years. As petroleum costs escalate with diminishing supplies, liquid transportation fuels will still be preferred due to energy density, safety, and distribution infrastructure, with increasing growth of biofuels between now and 2035 (IEA, 2008). The biofuels market represents the largest and most consistent demand from which to build a strong sugar and biomass supply chain from field to factory. The global biofuels market is already estimated to be $150 billion per year (UN, 2009). According to one report focused on the United States market: Rapid growth in the consumption of renewable fuels results mainly from the implementation of the US Renewable Fuel Standard (RFS) for transportation fuels and State renewable portfolio standard (RPS) programs. Biofuels production will grow over the next two decades, though is likely to fall short of the 36 billion gallons of RFS target in 2022. However, it may exceed expectations for 2035 including fuels from cellulosic ethanol, renewable diesel, and first generation biofuels. (Newell, 2009)

Beyond biofuels, companies are increasingly targeting higher value biobased chemical products and biomaterials. Liquid fuels are the ultimate commodity chemicals, representing the highest volume, but lowest value products whether produced from fossil or renewable feedstocks. In the United States, the petroleum-based liquid fuels industry and related energy services account for approximately 67% of petroleum consumed, with an overall industry value of $350 billion dollars. In contrast to commodity fuels, the goods and services resulting from the highervalue plastics, coatings, resins, and related consumer products utilize only 7% of petroleum consumed while resulting in an approximate $255 billion impact (Frost, 2005). Cargill and McKinsey & Company estimate that there is a potential to produce up to two-thirds of chemicals from biobased materials representing over 50,000 products, a $1 trillion annual global market (Jarrel, 2009).

P1: SFK/UKS P2: SFK c01 BLBS075-Hood December 15, 2010

6

20:45

Trim: 244mm×172mm

Plant Biomass Conversion

Commercial examples of higher value biobased products are emerging with increased frequency, as profiled in Chapter 12 of this review. By 2007, internal corporate investment in research and development related to biobased chemicals and biomaterials was as much as $3.4 billion, which far outpaced biofuels. This was due primarily to internal research and development investments from a few large pharmaceutical companies, which was in contrast to the United States Government’s continued focus during the same time period on liquid transportation biofuels (Lundy et al., 2008). While some biobased products are direct replacements for fossil-derived materials, others possess novel properties unique to their biogenic origin. An interesting example is Canadian-based EcoSynthetix (www.ecosynthetix. com) that is producing a starch-based coating product for the paper industry that outperforms its competitive products by requiring less water and heat in production, while exhibiting superior ink adhesive properties. This product is competitive with its petroleum-based counterpart when the price of oil is as low as $30 per barrel. Recent grants from United States Department of Energy (DOE) to support biomass work have not just focused on biofuels. For example a sizable $600 million round of funding awarded in late 2009, included support for Myriant Technologies’ succinic acid project and Amyris Biotechnology’s process to produce a range of biobased products to complement their biofuel program. It is expected that this trend will continue with both public and private investment focused on a range of high-value biomaterials and chemicals, as opposed to exclusively on biofuels.

1.3 Feedstocks 1.3.1

Biobased Feedstock Availability and Issues

Globally, ample supplies of renewable feedstocks are available for developing a robust and profitable biobased products industry, including agricultural crops, residues, and forestry materials, as well as future sources such as algae. Lignocellulosic biomass is globally dispersed and can be found in many forms, including agricultural crop and processing residues, forestry resources, dedicated energy crops such as miscanthus and switchgrass, and municipal solid waste. In the United States alone, resources associated with agriculture and forestry were calculated at 1.3 billion dry tons per year of biomass potential (Perlack et al., 2005). There are additional chapters in this volume that provide detailed information from a variety of perspectives on the availability of lignocellulosic biomass. It is important to note that the theoretical availability of biomass does not necessarily mean that it is economically feasible or environmentally viable to collect. For example, many primary row crop regions in the United States would produce excellent yields of perennial bioenergy crops, but the economics do not currently support substitution. The availability of agricultural crop residues must also be carefully considered. Crop residues include sustainably removable materials left after harvesting primary crops such as corn and wheat. Such residue availability has often been calculated based on 1:1 corn stover-to-grain ratios provided in the “Billon-Ton Report” published by the United States Department of Agriculture (USDA) and DOE (Perlack et al., 2005). Corn stover is widely considered a prime candidate for bioprocessing, although the actual availability within a working farm system, as well as collection incentives to farmers, may not be understood adequately. Additionally, regional (and global) variables affecting crop residue supply are acknowledged in the Billion Ton Report. For example, in northern climates, corn stover, the stalks and residues left after harvesting the grain, does not degrade quickly due to the cold winter temperatures. This sometimes creates a problem in that there is too

P1: SFK/UKS P2: SFK c01 BLBS075-Hood December 15, 2010

20:45

Trim: 244mm×172mm

1 The Bioeconomy: A New Era of Products Derived from Renewable Plant-Based Feedstocks

7

much stover for field preparation activities for the next spring. As corn yields increase due to biotechnology, this problem may increase and it will be necessary to remove stover. This is not the case in southern climates, as stover degrades quickly in the wet, relatively warm winters and is counted on by the farmer as a valuable source of organic matter in the soil. Rice straw is another potential crop residue source that is widely available. In this case it would be an environmental benefit to remove the straw because currently it must be burned or otherwise disposed of every year to avoid diseases in the following year’s rice crop. It would be a great benefit for air quality and the farmer to develop a market for this straw in a biomass application. In California, there has been much work, with limited success, in trying to develop markets for rice straw as a reaction to the Rice Straw Burning Reduction Act of 1991 (AB1378). Projects included the development of construction products and packaging materials from rice straw. Unfortunately, the straw is high in silica, which damages existing harvesting and handling equipment, making it impractical to develop a widespread harvesting system. However, if alternative technologies could extract higher value silica products, economics might support the development of more robust harvesting systems. The use of wheat straw in biomass processing and biomaterials has potential, as the straw is already harvested in the United States and globally for use in animal bedding and for other applications. Over the last two decades, wheat straw has been used for composite construction materials, filler materials in plastics, and in the development of cellulosic ethanol. Iogen (www.iogen. ca) has based its cellulosic ethanol demonstration plant on wheat straw as a major raw material. The company is planning its first commercial facility in Saskatchewan that will utilize cereal straw feedstocks. In the near term, especially in the United States, corn cobs may represent the most accessible crop residue for early commercial lignocellosic processing. Harvesting of corn cobs in a onepass system is feasible and is being developed as a component of the United States’ corn ethanol industry. There is already an existing market in some regions for corn cobs at approximately $80.00 per ton, to be used in the production of chemicals such as furfural. Companies such as POET Biomass, a division of POET (www.poetenergy. com), and DuPont Danisco Cellulosic Ethanol, LLC (www.ddce. com) are developing conversion technologies specifically targeting corn cobs as feedstocks for biochemical conversion using enzymes. There is also significant work by major equipment companies, including CNH America LLC. (www.cnh. com) and Deere & Co. (www.deere. com), on one-pass harvesting systems for corn cobs. Another potential source for lignocellulosic biomass is dedicated energy crops, both perennials and annuals. Perennials include crops such as miscanthus and switchgrass which have recently been the focus of attention by crop biotechnology and cellulosic ethanol companies. Perennials offer options to farmers and land owners for use of marginal land that is currently in pasture or other use. These crops sequester carbon in their root systems, as well as utilize relatively small amounts of inputs such as fertilizers and pesticides. The economics of producing these crops does not lend itself to replacing prime row crops, but their production may be part of farm-based crop diversification strategies in the future or as part of a program to utilize marginal and unproductive farm land. Annual crops include sweet sorghum and forage sorghums, both of which require minimal inputs and produce significant biomass. In the case of sweet sorghum, a large sugar content in the crop can be easily converted to ethanol or other biobased products with current technology, while the bagasse could serve as feedstock for lignocellulosic conversion technologies. Markets for these crops are being developed for biopower applications, even as other higher value uses are being commercialized. There is a growing market in Europe and in certain

P1: SFK/UKS P2: SFK c01 BLBS075-Hood December 15, 2010

8

20:45

Trim: 244mm×172mm

Plant Biomass Conversion

regions in the United States for densified wood and energy crop pellets and briquets for home heating, industrial use, and co-firing with coal. The latter use is increasingly being driven by regulatory requirements directed toward renewable power generation and greenhouse gas reduction. Forestry and wood-processing residues and byproducts, as well as short-rotation woody crops, also represent important biomass feedstocks. Collectively referred to as “woody biomass,” these resources are often advantaged by an existing year-round harvesting and collection infrastructure. A detailed analysis of all of the crops, trees, and residues is provided in this volume.

1.3.2

Characterization of Lignocellulosic Feedstocks

Woody and herbaceous biomass, or lignocellulosic biomass, primarily comprises three major components—lignin, cellulose, and hemicellulose—along with lesser amounts of minor and trace constituents. Cellulose and hemicellulose are polysaccharides or sugar polymers composed of repeating monomer sugar units bonded together into long chains, much like rail cars are coupled together to form a train. Combined with lignin, these biopolymers comprise the structural components of plant matter and are produced by the photosynthetic process, whereby atmospheric carbon dioxide (CO2 ) is absorbed by the plant, chemically transformed, and “fixed” into these other useful chemical materials. Lignin is a natural polymer found in all plant materials, which combines with cellulose and hemicellulose to provide structural strength to the plant. It is not a sugar polymer, but rather an aromatic polymer, meaning its component phenylpropyl molecular units contain the highly stable benzene-ring chemical structure, which is also the basis for many commercially useful materials produced from petroleum. The aromatic chemical structure also imparts a high caloric value to the lignin molecule, which is valuable for combustion (heat) and also chemical transformations. The lignin polymer can have significant variability in its chemical structure, often differing based upon the biomass source. Cellulose and hemicellulose are referred to as carbohydrates because they are aliphatic polymers composed only of carbon, hydrogen, and oxygen. Cellulose is the most abundant biopolymer on earth and is made of six carbon or C-6 glucose (sugar) monomers. Cellulose obtained from wood pulp, cotton, and other plants has been used for centuries to produce paper and cardboard, as well as derivative products. Often referred to as dietary fiber, it is not digestible by humans, but with recent technology developments, it can now be commercially hydrolyzed by chemical, enzymatic, or biological processes to its monomer sugars, which can then be readily utilized as feedstocks for bioprocessing. Yeast fermentation of glucose to ethanol (mostly for beverages) has been practiced for centuries, and other natural and genetically modified organisms can convert glucose to various useful chemical molecules. Hemicellulose is a polymer primarily composed of various five carbon or C-5 sugar monomers with some C-6 sugars as well. Unlike cellulose, it is an amorphous polymer with little structural strength and is easily hydrolyzed to its monomeric sugars with acid/base or enzymes. Unfortunately, C-5 or xylose sugars cannot be fermented using natural yeasts. However, aggressive research and development programs are developing new organisms and genetically modified yeasts to utilize these readily available C-5 sugars as bioprocessing feedstocks, as described in Chapter 8 of this volume. Fractionation, or separation, of petroleum into its component constituents has been the key methodology to develop high value petrochemical end products. A similar biobased example

P1: SFK/UKS P2: SFK c01 BLBS075-Hood December 15, 2010

20:45

Trim: 244mm×172mm

1 The Bioeconomy: A New Era of Products Derived from Renewable Plant-Based Feedstocks

9

is corn wet milling, in which the corn kernel is separated into its different components, from which value-added products are produced. As noted above, lignocellulosic biomass feedstocks possess comparable compositional diversity, and several leading technology developers are pursuing fractionation, or separation, of these components in order to facilitate more efficient and targeted downstream conversion of each component to value-added products. Historically, lignocellulose fractionation originated in the pulp and paper industry, where processes were designed to remove hemicellulose and “de-lignify” wood pulp in order to obtain a purified cellulose fraction for paper manufacturing (Agenda 2020 Technology Alliance, 2006). More recent approaches have used a combination of physical and thermal pre-processing followed by aqueous and/or solvent extractions, to afford substantially purified fractions of hemicellulose, lignin, and cellulose for further processing that is specific to each component. As a supporting technology, lignocellulose fractionation may prove to be extremely valuable as an integrated component of biochemical processing, providing sugars for fermentation and also a purified lignin stream as an aromatic chemical platform feedstock. Developmental and commercial lignocellulosic fractionation technologies are fully described in Chapter 12 of this volume.

1.3.3

The Role of Agricultural Biotechnology

In order to provide sustainable food, fuel, and material needs of humans, it will be necessary to dramatically increase the yields of agricultural crops and forest resources, as well as develop crops with specific attributes for biomass feedstocks. Currently, biotechnology traits used to reduce farmers’ costs and increase profitability are widely deployed in canola, corn, cotton, soybeans, and sugarbeets, predominantly in the United States. However, more than 13 million farmers in 25 countries currently grow agricultural biotechnology crops. In 2008, the global biotechnology crop area grew by 9.4%, or 26.4 million acres, to reach a total of 309 million global acres. Between 2007 and 2008, the United States alone increased its biotechnology crop acreage from 143 million acres to 154 million, phenomenal growth considering that the first biotechnology crops were not introduced until the mid-1990 s (ISAAA, 2008). To date, the vast majority of commercialized agricultural biotechnology-derived crops have focused on genetic “input traits,” which add value to the farmer and/or environment by reducing the production costs of the farm operation. Examples include RoundupTM Ready soybeans that are resistant to the herbicide glyphosate, allowing soybean farmers to more widely adapt conservation tillage practices. There are other examples related to insect and herbicide resistance. In addition to input traits in commodity crops, plant biotechnology has developed new crops for bioenergy and pharmaceutical applications with enhanced “output traits,” which allow the crop to produce certain characteristics desired by food, health, or industrial customers. While the value proposition for input traits is directed to the farmer, output traits are directed to those making products from the crops and ultimately to the consumer. Output traits allow crops to have higher protein and other nutritional properties, stronger fibers, specific oil profiles, novel health benefits, and to produce new products within green plants. In short, output traits enhance the value of the plant as a feedstock for the production of plant-based products. Not all of these technology improvements are created through gene transfer. Some use mutation, breeding, and other novel techniques to create new crop performance. Numerous examples of enhanced crop products have been commercialized or are in development for food, feed, and industrial applications, as summarized in Table 1.1. In the mid-1990s, Monsanto Company had a designer fiber unit that attempted to match specialized end uses for

10 Low-linolenic soybeans Phytase

Mirel Vistive QuantumTM Phytase

Metabolix

Monsanto

Syngenta Seed

Amylase and cellulase (under development) New oilseeds

Syngenta Seeds

Targeted Growth Inc.

2006.

Ricinoleic acid

Linnaeus Plant Sciences Inc.

a Grooms,

Cellulase

Infinite Enzymes LLC./ Applied Biotechnology Institute

Plant-based plastic and chemicals

Improved biomass characteristics, yields

Bioenergy Seeds

Low linolenic soybeans

Mendel Biotechnology

Partnership with Bunge

DuPont/Pioneera

40% more oil than No.2 yellow corn

Plant-made industrial enzymes

Supercede HE High Energy

Dow AgroSciences (Mycogen Seeds)a

High oleic/low linolenic fatty acids

Infinite Enzymes, LLC.

Nexera

Dow AgroSciences

Improved biomass characteristics, yields

Improved biomass characteristics and plant-made enzymes

Blade

Ceres, Inc.

Trait

Camelina

Corn

Camelina

Corn

Corn

Soybeans

Switchgrass

Miscanthus and other biomass crops

Biofuels

High amylase corn for conventional ethanol production

Green chemistry/biomaterials

Plant production system for cellulase used in green chemistry/cellulosic ethanol

Animal feed ingredient

Food/feed, reduce trans fats, increase shelf life and oil stability

Biomaterials and bioenergy

Advanced biofuels and biopwoer

Green chemicals and advanced biofuels

Biomaterials and advanced biofuels

Switchgrass and corn

Corn

Food/feed, future products include high-oleic and Omega-3

Animal feed—increases energy in metabolism

Food—no trans fats and low in saturated fat

Advanced biofuels, biobased chemicals, and biopower

Applications

Soybeans

Corn

Canola and sunflower

Switchgrass, biomass sorghum, sweet sorghum and other biomass crops

Crops

20:45

Edenspace Systems Corporation

Brand

Selected plant biotechnology companies and products.

Company

Table 1.1.

P1: SFK/UKS P2: SFK c01 BLBS075-Hood December 15, 2010 Trim: 244mm×172mm

P1: SFK/UKS P2: SFK c01 BLBS075-Hood December 15, 2010

20:45

Trim: 244mm×172mm

1 The Bioeconomy: A New Era of Products Derived from Renewable Plant-Based Feedstocks

11

cotton with specific, genetically engineered characteristics in a cotton variety. Other examples include products for reagent use that ProdiGene, Inc. commercialized. Trypsin (trade name TrypZean), beta-glucuronidase, and avidin are all sold by Sigma Chemical Co. Significant research is also being conducted to improve the fatty acid profile in camelina, a crop pioneered by United States producer groups in the Great Plains and companies such as Targeted Growth, Inc. (www.targetedgrowth. com). Targeted Growth, Inc. began a breeding program for camelina in 2005, employing a three pronged approach: classical and molecular breeding, mutation breeding, and transgenics (Panter, 2008). Other companies, such as Linnaeus Plant Sciences (www.linnaeus.net), are seeking to change the oil profile of the crop for various novel biobased product applications. Researchers are also working to adapt modern biotechnology tools to the development of new bioenergy crops. Examples of these include miscanthus, switchgrass other herbaceous crops, and short rotation woody crops. Companies working in this area include Ceres, Inc. (www.ceres.net); Chromatin, Inc. (www.chromatininc.com); Edenspace Systems Corporation (www.edenspace. com); Infinite Enzymes, LLC (www.infiniteenzymes.com), Mendel Biotechnology, Inc. (www.mendelbio.com), and Metabolix (www.metabolix. com). For both large multinational biotechnology firms and small boutique trait developers, a key hurdle is navigating a confusing and costly regulatory environment. A streamlined approach to getting crops deregulated and into commercial applications will have to be coordinated to attract the significant capital needed to grow this part of the industry.

1.3.4

Biomass Agricultural Equipment Development

As discussed in this chapter and the subsequent chapters in the volume, there is significant activity in the development of biomass feedstocks, conversion processes, and end-use applications. Increasingly, there is also investment by major farm equipment manufacturers in developing biomass harvest systems such as one-pass corn cob harvesters or systems to remove tree residues in harvesting. Noteworthy projects include commercial scale harvesting demonstrations of corn cobs by POET Energy and switchgrass by Genera Energy, LLC. (www.generaenergy.net). Equipment manufacturers actively engaged in biomass development activities include AGCO (www.agcorp. com); CLAAS (www.claasofamerica. com); CNH America, LLC. (www.cnh.com); Deere & Company; and Vermeer Corporation (www.vermeer. com). The involvement of these prominent companies in both agriculture and forestry-based biomass development addresses another major link in the new supply chain and promises innovative solutions for farmers and processors.

1.4 The Biochemical Technology Platform Biomass or components of biomass can be used as feedstocks by molecular modification of the constituents, a process often referred to as bioprocessing. There are three distinct technology platforms for these molecular transformations—chemical, thermochemical, and biochemical. Each platform has specific characteristics for commercial processing, including range of feedstocks and products, co-products, cost, scale, and stage of technology development. The focus of this book is on the biochemical technology platform that utilizes enzymes and microorganisms to effect molecular transformations, often with incredible energy efficiency and product specificity. Biochemical processing, sometimes referred to as the “sugar” or

P1: SFK/UKS P2: SFK c01 BLBS075-Hood December 15, 2010

12

20:45

Trim: 244mm×172mm

Plant Biomass Conversion

carbohydrate platform, seeks to convert C-6 and C-5 sugars derived from biomass through fermentation processes to biofuel and biobased chemical products. In the United States and globally, much recent biochemical platform R&D has focused on the development of pretreatment systems and enzymes that can depolymerize cellulose and hemicellulose into monomeric sugars to allow yeast fermentation to “cellulosic” ethanol. As a second-generation biofuel, ethanol derived from lignocellulosic feedstocks eliminates the food-fuel issue associated with sugar/starch feedstocks and has also been shown to have a much more favorable net energy balance and lifecycle GHG reduction than corn (starch-based) ethanol. Also within the biochemical platform, other research is developing new bacterial organisms and genetically modified yeasts to convert both C-6 and C-5 sugars to other biofuel and chemical products (see Chapters 11 and 12 in this volume), with perhaps the most advanced efforts directed at butanol, an important industrial chemical and possible second-generation biofuel, and succinic acid, a multifunctional platform chemical. Significant progress has been made in developing biochemical technologies to use lignocellulosic feedstocks. The overall process requires several steps, including feedstock pretreatment, depolymerization, sugar fermentation, and distillation/product isolation. Several pretreatment methodologies have been developed, generally combining heat, pressure, and chemical reaction to make the cellulose and hemicellulose polymers more accessible to enzymatic and microorganism attack (as described in Chapter 9 in this volume). Pretreatment processing must be designed to minimize introduction or formation of contaminants that would be toxic to the downstream fermentation organisms. While significant hydrolysis of the hemicellulose can occur during pretreatment, cellulase and other enzymes must be added to convert the more recalcitrant cellulose to its component C-6 sugars and complete conversion of hemicellulose to C-5 sugars. Remarkable advancements in cellulase enzyme cost and effectiveness have been made in the last 5 years and are continuing (see Chapter 10 in this volume). Novel approaches to enzyme production are also being developed, for example: A major technical challenge in making cellulosic ethanol economically viable is the need to lower the costs of enzymes needed to convert biomass to fermentable sugars. The expression of cellulases and hemicellulases in crop plants and their integration with existing ethanol production systems are key technologies that will significantly improve the process economics of cellulosic ethanol production. (Sainz, 2009)

While C-6 sugars are readily fermented to ethanol by natural yeasts, current R&D programs seek to develop new organisms that can effectively convert the C-5, as well as the C-6 lignocellulosic sugars to ethanol and other chemical products. Some R&D programs are pursuing organisms that can both hydrolyze cellulose and hemicellulose and ferment the resulting mixed sugars to ethanol, referred to as consolidated bioprocessing. These efforts will be more fully described in other Chapters throughout this volume.

1.5 Investment and Major Players Despite the recent global economic downturn, $16.9 billion was invested in new biofuels in 2008 (UN, 2009) and the cleantech sector emerged as the leading investment category for venture capitalists in 2009. Recognizing the complexities of introducing new biobased

P1: SFK/UKS P2: SFK c01 BLBS075-Hood December 15, 2010

20:45

Trim: 244mm×172mm

1 The Bioeconomy: A New Era of Products Derived from Renewable Plant-Based Feedstocks

13

technologies, strategic partnerships are becoming a preferred route to commercial products in the biobased supply chain. This is leading to a business environment that includes direct investments in entrepreneurial ventures, as well as many strategic partnerships and joint ventures, often leveraging existing competencies and assets. Table 1.2 summarizes key biobased product companies and their investors/partners.

Table 1.2.

Selected investments and joint ventures in biobased product companies.

Company

Feedstock/Product

Investors/Partners

Abengoa Bioenergy

Corn ethanol and experimental lignocellulosic biomass fuels and chemicals

DOE

Amyris Biotechnology

Development of advanced biofuels and chemicals from sugar-based feedstocks including sweet sorghum and sugar cane.

DAG Ventures, Khosla Ventures, Kleiner Perkins, TPG Ventures, Total

Butamax Advanced Biofuels

Biobutanol, sugarbeets

DuPont, British Petroleum

Catchlight Energy LLC.

Feedstocks, supply chain, technology licensing and deployment

Joint ventures between Chevron & Weyerhauser

Coskata

Wood, MSW

Khosla Ventures, Great Point Ventures, Advanced technology Ventures, General Motors, Globespan

Dupont Danisco Cellulosic Ethanol, LLC.

Corn cobs, switchgrass

DuPont, Danisco, Genera Energy (University of Tennessee).

Dupont Tate&Lyle

Sugar

DuPont, Tate&Lyle

Elevance Renewable Sciences

Oilseeds

Cargill Inc., Materia Inc., California Institute of Technology, TPG Growth, TPG Biotechnology Partners

Gevo Development LLC.

Biobutanol, other biobased products.

Cargill, ICM, Khosla Ventures, Virgin Fuels, Burrill & Company, Malaysian Life Sciences Capital Fund

Iogen Corporation

Wheat straw, other feedstocks, cellulosic ethanol

Royal Dutch Shell, Goldman Sachs, Volkswagen, Petro-Canada, Government of Canada, DSM

Mascoma Corporation

Biomass feedstocks, cellulosic ethanol

Khosla Ventures, Flagship Ventures, General Catalyst, Kleiner Perkins, Vantage Point, Atlas Ventures, Pinnacle Ventures

Range Fuels

Wood and other biomass feedstocks, cellulosic ethanol

Passport Capital, BlueMountain, Khosla Ventures, Leaf Clean Energy Company, PCG Clean Energy & Technology Fund

Verenium Corporation

Wood, sugarcane bagasse

British Petroleum, Khosla Ventures, Braemar Energy Ventures, Charles River and Rho Ventures

P1: SFK/UKS P2: SFK c01 BLBS075-Hood December 15, 2010

14

20:45

Trim: 244mm×172mm

Plant Biomass Conversion

Significant investments were made in advanced biofuels and green chemistry by a range of biotechnology, petroleum, and chemical companies in 2008 and 2009. Funding directed toward developing biobased chemicals has risen steadily since 2004 and reached $3.4 billion in 2007 in the United States (Lundy et al., 2008). From 2004 to 2009, major investments were made by large multinational petroleum and chemical companies, including Chevron, DuPont, Dow, Dutch Royal Shell, and Exxon Mobil. Venture capital investment also grew substantially, with the cleantech sector moving into a leadership position through the economic downturn of 2008–2009. As an example, new funds were announced by leading cleantech investor Khosla Ventures (www.khoslaventures.com), as well as Finistere Ventures (www.finistereventures.com). Partnerships among established global producers in the agriculture, biotechnology, chemical, and petroleum sectors are becoming commonplace. An early example was the Cargill-Dow, LLC venture started in 1997, which is now NatureWorks, LLC (www.natureworksllc. com) (wholly owned by Cargill), which invested approximately $1 billion to commercialize cornbased polylactic acid (PLA). British Petroleum and DuPont have formed Butamax Advanced Biofuels, LLC (www.butamax. com) to commercialize biobutanol as an advanced biofuel, while British Petroleum is also investing in Verenium and other advanced cellulosic biofuels businesses. More recently, Exxon Mobil announced in 2009 a $600 million investment to produce biofuels from algae in a joint venture with Synthetic Genomics founded by human genome pioneer J. Craig Venter (Mouawad, 2009). Royal Dutch Shell PLC increased its investment to $60 million in 2009 in Codexis (www.codexis) to explore biofuels production (Gold, 2009) while continuing to partner with Canada-based Iogen (www.iogen. ca). DuPont has formed Dupont Danisco Cellulosic Ethanol, LLC (www.ddce. com) to commercialize cellulosic ethanol and is in a joint venture with sugar company Tate & Lyle (called Dupont Tate & Lyle Bioproducts) to produce 1,3-propanediol (www.duponttateandlyle. com). These are a few of the hundreds of new business divisions, companies, and partnerships being developed globally to pursue renewable fuels and chemicals. Major venture capital funders such as Khosla Ventures (www.khoslaventures. com) and Burrill & Company (www.burrillandco. com) announced new funds in 2009 that will focus on biofuels and related technologies, including two new funds totaling $1 billion by Khosla Ventures. Khosla Ventures has led investment in numerous advanced biofuels companies, including Amyris (www.amyris. com), Coskata (www.coskata. com), Gevo (www.gevo. com), LS9 (www.ls9. com), Mascoma (www.mascoma. com), Range Fuels (www.rangefuels. com), and Verenium (www.verenium. com), providing strong strategic direction and momentum to the industry at a crucial time. Coskata has also received investment from General Motors, another example of the cross-industry, strategic investments made in this space between 2005 and 2009. Incorporating the forestry sector, Catchlight Energy is a joint venture between Chevron and Weyerhaeuser dedicated to combining the strengths of the two organizations to commercialize biofuels (www.catchlightenergy. com). The formation of domestic and foreign strategic alliances has grown from 532 new industrial biotechnology alliances in 2004 to 1,367 new alliances in 2007. Patent and trademark activity has intensified as firms seek to protect, commercialize, and license their new discoveries and brands. Trademark registrations in particular have shown strong growth, increasing from 197 new registrations in 2004 to 1,027 in 2007, reflecting the increasing prominence of biobased brands as the field moves from early discoveries to the commercialization of innovative technologies and products (Lundy et al., 2008). The Federal Government’s significant investments since 2000 have helped many of the early companies develop technologies, improve processes, and leverage private investment, as

P1: SFK/UKS P2: SFK c01 BLBS075-Hood December 15, 2010

20:45

Trim: 244mm×172mm

1 The Bioeconomy: A New Era of Products Derived from Renewable Plant-Based Feedstocks

Table 1.3.

15

Selected USDA and DOE grants (2002–2009) for biomass projects.

Year

Program

Total Funding

Awarded Projects

Partial List of Recipients

2002

Biomass R&D Joint

$79,350,000.00 8 awards

Broin & Associates (now POET), Cargill, DuPont, Abengoa, National Corn Growers Association, Iowa Corn Promotion Board

2003

Biomass R&D Joint

$23,803,802.00 19 awards

Dartmouth (Mascoma), University of Florida (now partnering with Buckeye Technologies), Pure Vision Technology, Metabolix, Cargill, ADM

2004

Biomass R&D Joint

$26,357,056.00 13 awards

Rohm & Haas Co., Weyerhaeuser Company

2005

Biomass R&D Joint

$12,626,931.00 11 awards

Samuel Robert Noble Foundation

2006

Biomass R&D Joint

$17,492,507

17 awards

Increasing focus on feedstock development: Ceres Inc., SUNY, Edenspace Systems

2007

Biomass R&D Joint

$18,449,090.00 21 awards

GE Global Research, Ceres Inc., Agrivida Inc.

2007

DOE Commercial Scale Biorefinery

$385,000,000.00 6 awards

Abengoa Bioenergy, BlueFire Ethanol, Broin Companies (now POET), Iogen, Range Fuels

2008

DOE Small Scale Biorefinery

$200,000,000.00 7 awards

Verenium, Lignol Innovations, ICM, UT/Genera

2009

DOE Advanced Biorefinery

$564,000,000.00 19 awards

ADM, Amyris Biotechnology Inc., Elevance Renewable Sciences, BioEnergy International LLC (Myriant)

summarized in Table 1.3. For example, the recent $564 million DOE investment in advanced biorefinery projects leveraged a private investment of $1.3 billion. The Obama Administration is continuing to invest heavily in advanced biofuels, regional innovation clusters, renewable energy jobs, and biotechnology. On the feedstock side, substantial investment is occurring in the development of dedicated energy crops and crop-based synergies such as enhanced traits or downstream processability. These initiatives include the commercialization of plant-made enzymes by Syngenta and startup companies, and the commercialization of dedicated energy crops by companies such as Arborgen, Ceres, and Mendel Biotechnology.

1.6 The Role of the Farmer An essential component of the value chain for biobased products is the alignment of companies seeking to commercialize biobased products with feedstock providers, namely the farmers, logistics, and preprocessing producers. All too often biobased products industry proponents tout the ability of biorefineries to revitalize rural regions without understanding the overall value proposition or fully considering the vital linkages necessary with the farmer. Three

P1: SFK/UKS P2: SFK c01 BLBS075-Hood December 15, 2010

16

20:45

Trim: 244mm×172mm

Plant Biomass Conversion

primary models have emerged for companies and processors to access lignocellulosic biomass, as described below (Nelson, unpublished). (1) Farm gate—In this model, farmers are paid a set price and/or contracted price for biomass baled and delivered to the factory, storage site, or made available at field for pickup. Generally, this price is between $30 and $70 per dry ton according to most publications. In discussions with farmers, it is clear that this model will require a guaranteed long-term contract that includes one or more of the following: an independent ability to market carbon credits, a contract price (especially for perennial energy crops) indexed to corn or petroleum, and a guaranteed price floor. The Biomass Crop Assistance Program (BCAP) authorized in the 2008 Farm Bill was recently released, which will provide assistance to farmers producing biomass crops within this scenario. (2) Access fee/land rental—In this model, farmers are essentially land owners, similar to the pulp and paper industry, in which companies pay to have dedicated energy crops produced and the companies handle the planting and harvesting. Although the farmers may have some role in maintenance, they are essentially operating as absentee land owners. This model is not considered viable in major row crop farming regions, but may be of interest to small, part time farmers or those that own marginal land. (3) Value-added farmer participation—In this model, farmers participate in a value-added enterprise, possibly formed as a cooperative, in which their biomass production includes some component of preprocessing, logistics, and/or value-added processing or service to the end clients. As an example, a University of Tennessee-sponsored program through its subsidiary company Genera Energy, LLC (www.generaenergy.net) has formed a biomass processing cooperative to support scale up of switchgrass in East Tennessee. In the United States, the development of farmer-owned businesses to process biomass into pellets or briquettes to cofire with coal or other biopower applications may serve to establish a reliable supply chain for lignocellulosic biomass destined for future higher-value applications, other than combustion. Show Me Energy Cooperative LLC. of Centerview, Missouri (www.goshowmeenergy.com) is a great example. Show Me Energy has over 400 farmers who own part of the cooperative and are supplying waste straw and dedicated energy crops to their pellet operation. Logistics and storage considerations for commercial lignocellulosic feedstock supply are not insignificant. For example, within each of these models, some scenarios envision biomass materials being baled and stored at the fields for delivery throughout the year to the processing facility. It remains to be seen whether off-season on-field storage of lignocellulosic crops will be accepted by high production commercial row crop farming operations. For a given project or program, the successful model must involve farmers and supporting logistics providers as more than an afterthought. To attract serious, large-scale farmers who can professionally deliver large volumes of biomass as well as invest in the supporting infrastructure will require new business models that make a compelling case to each participant in the supply chain.

1.7 Opportunities for Rural Development A pressing need exists for rural development to provide jobs and long-term sustainable opportunities in rural areas in the United States and across the world. In the United States, poverty is consistently higher in rural as opposed to urban areas, with over 500 rural counties defined as being in “persistent poverty.” Agriculture is often a significant economic sector in these regions, but its role as a job and local wealth creator has declined in recent years (Cowan,

P1: SFK/UKS P2: SFK c01 BLBS075-Hood December 15, 2010

20:45

Trim: 244mm×172mm

1 The Bioeconomy: A New Era of Products Derived from Renewable Plant-Based Feedstocks

17

2002). New biobased feedstocks may offer the opportunity to grow, process, and transport lignocellulosic biomass for new products and rebuild these local agriculture-based economies. The conversion of lignocellulosic biomass to biobased products promises to have significant impact on rural communities. The low bulk density of lignocellulosic biomass dramatically changes logistics and processing dynamics. Grain crops such as corn and soybeans can be economically transported via barge and rail for remote processing and consumption, whereas at least the first phase of commercial lignocellulosic feedstock processing will necessarily be located in close proximity to the harvested biomass due to transport economics. This is especially true for herbaceous biomass such as processing and crop residues and dedicated energy crops, while forestry materials or densified feedstocks may be transported longer distances. Most models indicate that the viable transport radius of harvested herbaceous energy crops around a rural biorefinery or preprocessing facility is approximately 25–50 miles. This will necessitate smaller decentralized biorefineries across rural areas that will at least incorporate preprocessing and/or initial refining of lignocellulosic biomass substrates. To give an example of what this opportunity can mean for rural regions, the corn ethanol industry can be examined at its high growth period from 2006 to 2008. One report summarized the economic impact during this period as follows: The industry spent $12.5 billion on raw materials, other inputs, goods and services to produce an estimated 6.5 billion gallons of ethanol during 2007. An additional $1.6 billion was spent to transport grain and other inputs to production facilities. Within the corn ethanol industry, new jobs are created as a consequence of increased economic activity resulting from ongoing production and construction of new capacity supported the creation of 238,541 jobs in all sectors of the economy during 2007. These include more than 46,000 jobs in America’s manufacturing sector – American jobs making ethanol from grain produced by American farmers (Urbanchuk, 2008).

Despite its significant rural economic impact in the United States, it is generally recognized that starch-based ethanol is an important, but ultimately limited, first-generation biofuel. As such, its economic impact on rural communities may have been largely realized. Furthermore, the germplasm, production inputs, and processing of corn are controlled by a relatively small number of multinational companies such as DuPont Pioneer, Monsanto and Syngenta on the seed side, and ADM, Bunge, and Cargill on the processing side, leaving little room for entrepreneurial technology developers and farmer value participation. Fortunately, the fully realized bioeconomy will require diverse and flexible feedstocks and technologies to produce a comprehensive range of biobased products, as well as biofuels. The multiproduct “biorefinery” will increasingly utilize more globally abundant lignocellulosic feedstocks to produce both commodity and value-added products. As a result, the lignocellulosic processing opportunity is substantially larger in terms of volume, and the low bulk density will dictate that processing be located in proximity to feedstock production. In contrast to corn and other grains with fully developed food-centric supply chains, the lignocellulosic feedstock supply chain is largely undeveloped and unconsolidated. A recent comprehensive United States study concluded that in a 98-county area in the Mid-South Mississippi Delta region, lignocellulosic feedstock processing utilizing 10% of cropland, 25% of idle lands, 25% of conservation reserve program land, and 15% of pasture land would support a biomass industry valued at over $8 billion annually. This industry would create 25,000 new jobs within a decade and 50,000 jobs by 2030 in the study area alone (Tripp et al., 2009).

P1: SFK/UKS P2: SFK c01 BLBS075-Hood December 15, 2010

18

20:45

Trim: 244mm×172mm

Plant Biomass Conversion

1.8 Environmental Benefits The environmental benefits of biobased products and related technologies are just beginning to be fully understood. For example, studies have shown that the life cycle analysis for cellulosic ethanol produced from certain dedicated energy crops has a 90% reduction in greenhouse gas (GHG) emissions compared to petroleum gasoline (Farrell et al., 2006). Case studies have shown that energy and water use for biobased processes decreased 10–80%, while the use of petrochemical solvents was reduced by 90% or eliminated completely (OECD, 2001). Additional benefits can be found in product recyclability, air emissions, and reduced overall energy consumption from locally sourced foods, materials, and fuels. However, sustainable crop production is necessary to avoid soil and water depletion, as described in Chapters 6 and 7 of this volume.

1.9 Economic Comparison of the Biochemical and Thermochemical Technology Platforms According to a recent report by the International Energy Agency (IEA), the thermochemical and biochemical routes have comparable potential energy yields, converting dry biomass at about 20 GJ/ton to about 6.5 GJ/ton of biofuels, for an overall conversion efficiency of about 35%. The report further projects potential ethanol yield of about 80 gallons/dry ton from biochemical processing and a synthetic diesel yield of 53 gallons/dry ton from thermochemical conversion. Experience with each platform, utilizing biomass feedstocks, is limited to pilot and precommercial scale at present, so accurate production cost information remains to be confirmed (see Chapter 14 in this volume). Furthermore, leading private-sector technology developers do not generally publish proprietary process cost information. IEA has estimated production costs of second-generation biofuels to be in the range of $3.02–3.79/gallon for ethanol and at least $3.79/gallon for synthetic diesel, comparable to the wholesale petrochemical fuel prices when crude oil is in the range of $100–130/bbl. The IEA report concludes that there is presently not a clear commercial or technical advantage between the platforms for the production of biofuels and that widely fluctuating crude oil prices impart high risk to investment in second-generation biofuels (IEA, 2008). The IEA report does not take into account the value of biobased products and the benefit of added flexibility of diverse end product applications.

1.10 Conclusions and Future Prospects The bioeconomy represents a disruptive technological, social, and economic change that will be realized over decades, not years. The opportunities—as described in this volume—are many, as are the challenges. These will be met and exploited by a diverse combination of traditional foodbased agribusiness, the fossil-based fuel and chemical industries, entrepreneurs, financiers, and farmers. The market for biobased products is potentially higher value than for biofuels, and biochemical processing technologies likely offer more value-added specialty chemical product options than thermochemical technologies. Sugars derived from globally abundant and dispersed

P1: SFK/UKS P2: SFK c01 BLBS075-Hood December 15, 2010

20:45

Trim: 244mm×172mm

1 The Bioeconomy: A New Era of Products Derived from Renewable Plant-Based Feedstocks

19

lignocellulosic feedstocks would serve as the predominant raw materials for rural biorefineries, which in turn could transform declining rural economies and create new “local” supply chains for energy, liquid transportation fuels, and other products. Positive environmental impacts from the use of renewable feedstocks, lower intensity manufacturing, and more efficient local supply and consumption would be significant. A vision for plant-based renewable resources was published in 1998 (http://www1. eere.energy.gov/biomass/pdfs/technology roadmap.pdf). This document was the result of a workshop comprising industry and trade group representatives assembled to discuss what it would take to convert the US industrial base to a more sustainable economy. The goals were modest by most standards: 10% of bio-based products and 40% of fuels would be from plantbased sources by the year 2050. Over a decade later, progress down this path has been slow at best. More aggressive goals than those stated in the 1998 document have been set in the US 2007 Energy Independence and Security Act—36 billion gallons of renewable transportation fuels per year (∼20–30% replacement) to be reached in 2022. Because corn-starch-based ethanol is nearly maximal at 9 billion gallons per year (∼25% of the grain crop), the balance of the ethanol should be derived from lignocellulosic biomass. The ultimate requirement to replace finite fossil feedstocks with renewable resources is perhaps best described by Italian Chemist and holocaust survivor Primo Levi in his 1975 chronicle of the elements (Levi, 1975): Carbon, in fact, is a singular element: it is the only element that can bind itself in long stable chains without a great expense of energy, and for life on earth (the only one we know so far) precisely long chains are required. If the elaboration of carbon were not a common daily occurrence, on the scale of billions of tons a week, wherever the green of a leaf appears, it would by full right deserve to be called a miracle. Man has not tried until now to compete with nature on this terrain, that is, he has not striven to draw from the carbon dioxide in the air the carbon that is necessary to nourish him, clothe him, warm him, and for the hundred other more sophisticated needs of modern life. He has not done it because he has not needed to: he has found and is still finding (but for how many more decades?) gigantic reserves of carbon already organized, or at least reduced. Besides the vegetable and animal worlds, these reserves are constituted by deposits of coal and petroleum: but these too are the inheritance of photosynthetic activity carried out in distant epochs, so that one can well affirm that photosynthesis is not only the sole path by which carbon becomes living matter, but also the sole path by which the sun’s energy becomes chemically usable.

The chapters of this volume describe the current state of the transformation to a renewable and sustainable bioeconomy and suggest the opportunities that await its realization.

References Agenda 2020 Technology Alliance, American Forest & Paper Association. Integrated Forest Products Biorefinery. Available: www.agenda2020.org/PDF/IFPB Brochure.pdf. Cowan, T. 2002. Value-Added Agricultural Enterprises in Rural Development Strategies. Washington, DC: Congressional Research Service, The Library of Congress. DOE. 2007. Top Value-Added Chemicals from Biomass, Volume II—Results of Screening for Potential Candidates from Biorefinery Lignin. Available: www1.eere.energy.gov/ biomass/pdfs/pnnl-16983.pdf.

P1: SFK/UKS P2: SFK c01 BLBS075-Hood December 15, 2010

20

20:45

Trim: 244mm×172mm

Plant Biomass Conversion

Frost, J. W. 2005. Redefining chemical manufacture, replacing petroleum with plant-derived feedstocks. Industrial Biotechnology, 1(1), 23–25. Gold, R. 2009. Shell, other oil firms bolster biofuels spending. Wall Street Journal (newspaper), December 30. Grooms, L. 2006. Farm Industry News, September. Farrell, A. E., et al. 2006. Ethanol can contribute to energy and environmental goals. Science 311(5760), 506–508. IEA. 2008. 1st to 2nd Generation Biofuel Technologies: An Overview of Current Industry and RD&D Activities. Available: www.iea.org/Textbase/Publications/free new Desc. asp?PUBS ID=20742. International Service for the Acquisition of Agri-biotech Applications (ISAAA). 2008. ISAAA Report for 2008. Jarrell, K. A. 2009. Synthetic biology and the sustainable chemistry revolution. Industrial Biotechnology, 5(4), 210–212. Levi, P. (1975). The Periodic Table. Penguin Books, London, UK. Lundy, D., et al. 2008. Industrial Biotechnology: Development and Adoption by the U. S. Chemical and Biofuel Industries. Washington, DC: U. S. International Trade Commission. Mouawad, J. 2009. Exxon to invest millions to make fuel from algae. New York Times, July 13. Organisation for Economic Co-Operation and Development (OECD). 2001. The Application of Biotechnology to Industrial Sustainability. Paris. Nelson, P. 2006. Developing an Ag-based biomass supply chain: What is the role of the farmer? Presented at The World Congress on Industrial Biotechnology, Toronto, Canada, July 14. Newell, R. 2009. Annual Energy Outlook 2010, Reference Case, U. S. Energy Information Administration, Washington, DC, December 14. Panter, D. 2008. Sustainable Oils LLC, Conference Presentation, Memphis, Tennessee, November. Perlack, R. D., Wright, L. L., Turhollow, A. F., Graham, R. L., Stokes, B. J., & Erbach, D. C. 2005. Biomass as Feedstock for a Bioenergy and Bioproducts Industry: The Technical Feasibility of a Billion-Ton Annual Supply. Oak Ridge: Oak Ridge National Laboratory. Sainz, M. 2009. Commercial cellulosic ethanol: The role of plant-expressed enzymes. In Vitro Cellular & Developmental Biology, 45, 314–329. Tripp, S., Nelson, P., & Powell, R. 2009. Regional Strategy for Biobased Products in the Mississippi Delta, Report by Battelle Technology Partnership Practice. United Nations. 2009. Global Trends in Sustainable Energy Investment. Available: www.indiaenvironmentportal.org.in/node/277152. Urbanchuk, J. M. 2008. Contribution of the Ethanol Industry to the Economy of the United States. Prepared for the Renewable Fuels Association.

P1: SFK/UKS P2: SFK c02 BLBS075-Hood December 15, 2010

20:47

Trim: 244mm×172mm

Chapter 2

Agricultural Residues James Hettenhaus

2.1 Introduction The present US ethanol industry has grown rapidly using cereal grain, mainly corn, as its primary feedstock. Corn grain has been a global commodity for more than a century, with the supply and demand inextricably linked. Infrastructure is in place to grow, harvest, store, and transport corn to global markets. When the US demand for corn increased to supply newly constructed ethanol plants over the last decade, the only concern was locating those plants to economically source the corn, convert it to biofuels, and ship the products to market. The availability and quality of corn was not in question, just the price relative to other costs of goods sold. In 2007, Congress enacted the Energy Independence and Security Act, establishing a Renewable Fuel Standard (RFS) goal for advanced biofuels. Advanced biofuel is defined as the total production from cellulosic biomass and biomass-based diesel and biogas produced through the conversion of organic matter from renewable biomass. Cellulosic biofuels make up the major portion of the mandate by 2016 (Table 2.1). The basic platform for building advanced biofuels is the cellulosic biofuels segment. The RFS mandates growth from a few research and early commercialization efforts now to 16 billion gallons by 2022. Massive investment of resources across the supply chain will be required in a relatively short time frame to meet this goal. New technology must be validated, plants designed and constructed, feedstock sourced, and logistical systems in place to move cellulosic biomass feedstocks to facilities and finished biofuels to customer markets. For example, five 100 MGPY facilities could meet the cellulosic goal of 500 million gallons in 2012. An examination of the 100 MGPY corn-based facilities gives some ideas of the size and logistics of this undertaking. Each 100 MGPY corn-based facility requires about 36 million bushels of corn, supplied from about 200,000 acres. Because of improved seeds,

Plant Biomass Conversion, First Edition. Edited by Elizabeth E. Hood, Peter Nelson and Randall Powell.
C 2011 John Wiley & Sons Inc.

21

P1: SFK/UKS P2: SFK c02 BLBS075-Hood December 15, 2010

22

20:47

Trim: 244mm×172mm

Plant Biomass Conversion

Table 2.1. EISA 2007 biofuel goals.

Year

Advanced Biofuel (in Billion Gallons)

Cellulosic Biofuel (in Billion Gallons)

Cellulosic (% Advanced Biofuels)

Cellulosic Feedstock (Million Dry Tons)

2012 2014 2016 2018 2020 2022

2.0 3.8 7.2 11.0 15.0 21.0

0.5 1.6 4.2 7.0 10.5 16.0

25% 47% 59% 64% 70% 67%

500 1,600 4,200 7,000 10,500 16,000

corn yields increase 2–3 bu/ac annually. An additional 10–15 million bushels will be produced on the 5 million acres within a 50-mile radius of the facility. Depending on the relative cost, the farmer can increase the amount of corn planted or adjust the inputs to grow more corn on the same acres. The corn is usually delivered by truck over this distance. The facility may also choose to purchase the corn outside of the immediate area. Delivery could be by truck or rail, depending on the distance and relative cost. River Barge shipments are common and ocean-going shipments are also routine. The situation is somewhat different for biomass-derived products. The five cellulosic 100 MGPY facilities, necessary to reach the 2012 RFS goal, are estimated to require 1 million or more dry tons of cellulosic feedstock. Thermochemical facilities convert the lignin and the carbohydrates to fuels, producing 100–110 gallons per ton. Biochemical facilities use the cellulose and a portion of the hemicellulose for fuels, producing 80–90 gallons per ton. Their lignin is being evaluated for a multitude of uses, including supplying the process energy needs, selling the excess as power, or as raw material for value-added products (Simmons et al., 2010). Unlike corn, currently no system is in place to supply 1 million tons of feedstock to each of these facilities. In fact, this biomass supply requirement is 20 times greater than the largest corn stover or cereal straw collection supplied to a single site, currently 50,000 tons. This stover collection project for 50,000 tons required reaching agreement with 400 farmers and contracting with more than 30 custom operators for baling and hauling (Glassner et al., 1998; Hettenhaus, 2006). Thus, one can imagine the increased complexity of logistics. Cellulosic feedstock is also unlike corn and other cereal grains and sugar, i.e., no quality standards for biomass exist. Biomass is not fungible, and it has no asset value to provide source of liquidity to the owner until sold. The challenge is to overcome these and other obstacles in order to reach the goal of tripling biomass production between 2012 and 2014 and then increasing production ten times over the subsequent 8 years, between 2014 and 2022.

2.1.1

Key Issues

One of the key issues is that farmers and landowners control the cellulosic feedstock supply. They will decide when to provide their crop residues, deliver forest biomass, and make the decision to grow dedicated energy crops. Unless there is a significant benefit for farmers to change current production and crop management practices, cellulosic feedstock in large quantities from crop residues will be difficult to source for industrial use. A reliable market is needed for the farmer to commit to producing biomass, while the processor requires a reliable supply of feedstock. Unlike corn, bulky agricultural residues are

P1: SFK/UKS P2: SFK c02 BLBS075-Hood December 15, 2010

20:47

Trim: 244mm×172mm

2 Agricultural Residues

23

limited to a local market due to transport costs. A win-win relationship between the farmer and the processor is needed for the feedstock supply chain to flourish. Economics must benefit the farmer and the processor. Once cellulosic technology is proven: farmer participation in the value chain would help overcome the current “chicken or egg” obstacle of sourcing and supplying crop residues to biorefineries. The “Minnesota Model” provided incentives for Minnesota farmers to participate in the corn grain-to-ethanol value chain through offering ownership of corn ethanol plants. The State-funded program resulted in the formation of 12 farmer-owned ethanol processing cooperatives (Morris, 2008). Currently, over 30% of Minnesota farmers are investors in corn ethanol plants and a “Minnesota Model 2” is underway with Minnesota farmers collecting corn cobs for biomass gasification in their plants, while evaluating grasses, straw, and other cellulosic feedstocks (Kleinschmit, 2008). The feedstock must be competitive with natural gas pricing. Sustainable harvest systems must be in place to ensure that soil quality is maintained. Farmers are the stewards of the soil. However, processors and their customers expect validation of sustainable feedstock. This dichotomy requires a solution for moving agriculture residues ahead as feedstocks. Quality of feedstock is still loosely defined at the present stage of industry development. There is a consensus that feedstock needs to be free of dirt and relatively consistent in composition. Feedstock containing dirt accelerates equipment wear, increases maintenance cost, and causes lost production. Preferably, the material is collected while still standing in the field. During grain harvest, some residues inevitably end up on the ground as vehicles pulling trailers and grain carts drive alongside the combine to off-load the grain, knocking down a portion of the crop residues. A final issue is that feedstock composition varies, and any move towards more consistent composition would improve operating stability. Participants in stakeholder sessions considered consistency more important for biochemical processing than for thermochemical due to the more narrow selectivity of bioprocessing. Some processors said they would pay based on dry weight, but would eventually prefer to pay based on composition and “ease of processing.”

2.2 Feedstock Supply Sourcing an adequate, reliable, and economic supply of cellulosic feedstock is a major obstacle in commercializing renewable fuels. Studies at a macro level found the supply potential to be 900 million tons of agricultural biomass and nearly 400 million tons of forest biomass feedstock (Table 2.2; Perlack, 2005). These quantities of feedstock are based on a sustainable harvest using “best crop practices” for maintaining soil quality. The quantities will shift based on markets, government policies, and local economics. Crop residues are available in significant quantities now. However, removing large quantities for a biorefinery is a serious challenge to the farmer as well as the biomass buyer representing the biorefinery. Establishing sustainable crop systems for removal requires changing to new methods, acquiring new knowledge, and investing in different equipment. An estimated 900 MDTY of agricultural biomass is available now if “best crop systems” are practiced to maintain soil quality and achieve sustainable removal. Only 200 MDTY is required to meet the EISA 2022 goal of 16 billion gallons of cellulosic biofuels. The actual amount that can be removed in a sustainable and economic manner depends on cropping practices,

P1: SFK/UKS P2: SFK c02 BLBS075-Hood December 15, 2010

24

20:47

Trim: 244mm×172mm

Plant Biomass Conversion

Table 2.2.

US cellulosic biomass feedstock supply potential.

Biomass Source

MDT

Agricultural biomass Crop residues Perennial crops Others: e.g., process residues Forest biomass Forest products industry residues Logging residues Forest thinning Urban wood residues Energy crops

900 430 370 100 370 150 60 60 50 50

MDT, million dry tons.

especially till or no-till, and geophysical parameters—soil type, field slope, and length, as well as weather. The largest component of agricultural biomass is 430 MDTY of crop residues (Table 2.2), mostly corn stover. Stover production is 300 MDTY. Other residues are largely straw from cereal grains—wheat and barley. There is a 1:1 stover to grain weight ratio in most grain crops. A field producing 180 bu/ac corn leaves 5 tons/ac of stover in the field. Straw residues in dryland farming amount to 1–2 tons per acre. Some amount is often left on the field to retain soil moisture and prevent wind and water erosion. Larger residue quantities provide a harbor for fungi, weed seeds, and pests that can damage the crop. In the northern parts of the Corn Belt, the surface cover prevents the cold damp soil from warming in the spring, delaying seed germination. A rule of thumb is that each day delay in soil warming reduces the yield by 1 bu/ac. For these reasons, there is a clear advantage to the farmer for removing the stover, which will increase as corn yields increase. Continued trait improvement of corn is expected to raise corn yield 3 bu/ac/year in the near term. Seed companies project 300 bu/ac corn yields by 2030 due to advances in molecular breeding and biotechnology. The 1:1 ratio is expected to remain unchanged and stover, now at 8.4 tons/ac, becomes even more difficult to manage. Until now, there has been little market for stover. Most is on-farm use for animal bedding and cattle grazing after the grain harvest. As a result, the fields are tilled to bury most of the surface residue and compensate for soil compaction. Tilling adds $15/ac cost, increases soil erosion, and results in a loss of soil organic material, reducing soil quality. Process residues, 100 MDTY, include cotton gin trash, bagasse, and oat hulls. Available in smaller quantities at the processing plant, they enjoy an economic advantage, as they do not have to be collected and transported from the field. Forest products and energy crops are discussed in other chapters in this volume (Chapters 3 and 4) and will not be addressed here. As mentioned previously, conventional biofuel is ethanol produced from corn starch, while advanced biofuel is defined as renewable fuel that has life-cycle greenhouse gas emissions that achieve at least a 50% reduction over baseline life-cycle greenhouse gas emissions. Cellulosic biofuel is renewable fuel derived from any cellulose or lignocellulose that is derived from renewable biomass. Included within the definition of feedstocks for advanced biofuels are sugar or starch (other than corn starch), crop residues, vegetative waste material, food waste, and yard waste.

P1: SFK/UKS P2: SFK c02 BLBS075-Hood December 15, 2010

20:47

Trim: 244mm×172mm

2 Agricultural Residues

25

While the current data provide some idea of estimated feedstock, the future is not likely to maintain the agronomic system status quo. The segments in the agricultural production and processing pie will shift due to new and varied markets and technology improvement of the crops. Future feedstock availability will depend on the following: r Crops will be planted to meet the demands of new and existing markets. As markets

change, crops will be adjusted to meet customer demand.

r Cropping practices will be required to ensure soil quality maintenance. When a market

for crop residues develops, new residue management practices are likely that lead to economic and sustainable removal. ◦ No-till is assumed in most studies of crop residue management systems that will yield excess cellulosic biomass feedstock. Presently, only one out of five corn and wheat farmers use no-till today (CTIC, 2007). ◦ New cropping systems that include cover crops are needed for erosion control and to maintain soil carbon on many fields to allow for sustainable residue removal. r Accurate models for “soil quality” based on soil organic material must be developed. Several models are in development, but validation is required (See Chapter 7 in this volume). r Changes in farm and energy policies will be needed, with stronger connections between farm policy and energy policy, which can incentivize cellulosic feedstock production. Significant regional differences exist in soils, weather, and crop characteristics, as well as differences in harvesting mechanics for stover and straw. In dryer areas, more residue is currently left in the field to retain moisture in the soil. More cover must also be left on highly erodible soils and sloping fields to protect the soil from water and wind erosion. More corn stover is produced than straw, but straw is more readily removed. Corn stover yields are 3–5 times greater per acre than straw from cereal crops. Unless cereal crops are irrigated, there is little straw left to collect. For example, the average dryland wheat straw yield is 2,500 pounds per acre compared to 10,000 pounds per acre or more for corn stover. The equivalent of 1,200 pounds per acre of straw must be left on the surface to comply with erosion guidelines with no-till. The excess is less than 1 ton of straw per acre that is removable. In contrast, leaving 2,000 lbs of stover with no-till is often sufficient and the excess is 8,000 lbs or more of stover per acre. Straw collection infrastructure is generally well developed, while corn stover collection is not. When cereal grain is ready to harvest, straw usually contains 20% moisture or less, suitable for baling. In contrast, stover contains 50% moisture and must remain in the field to dry and be collected later, depending on the weather. A wet harvest season can prevent its collection entirely. Farmers generally agree that $50/acre pretax margin would raise some interest in harvesting crop residues, but only if the grain harvesting was not hindered in any way. Cash crop values—$3–$5/bu corn, $7–$10/bu for soybeans and wheat—make the relative value of residues small in comparison. Collecting the stover can add $100–$200/ac sales ($50/dry ton, collecting 2–4 dt of the 4.8 dt produced). However, some estimate $70–$100/dry ton delivered would likely be required to raise farmer interest. Most processors base their cost for biomass on $30–$50/dry ton delivered, a $40–$70 million annual difference in costs for a 100 MGPY plant. This discrepancy has not been adequately evaluated by potential purchasers of crop residues and other biomass feedstocks.

P1: SFK/UKS P2: SFK c02 BLBS075-Hood December 15, 2010

20:47

Trim: 244mm×172mm

Plant Biomass Conversion

26

Realizing this profit from residue sales requires overcoming related issues with harvesting and transporting crop residues that must be addressed on the farm level. Farmers routinely raise the following issues in evaluating this opportunity: r r r r r r r r r

Residue markets Residue harvest window Residue removal cost Corn yield and residue management Ag equipment needs Operating costs Residue nutrient value Farmer outlook Crop R&D lead times.

2.2.1

Residue Markets

Currently, there is a limited market for cellulosic crop residues and no market for energy crops. Crop residues that are collected are mostly for on-farm use—bedding and animal feed. Corn cobs are an exception. Corn cobs are 15%–20% of the stover. They are currently used for a wide variety of applications ranging from abrasives for polishing metal and wood (no silica hazard), as a carrier for pesticides and fertilizers, and as an absorbent for hazardous liquid spills (high absorbency properties). The price for a 40 lb bag of ground cobs for polishing metal is $25 FOB facility, equivalent to $1,250/ton (Soda-Blast.com, LLC). Collecting just the corn cobs during the grain harvest is less disruptive than collecting stover. The yield is about 0.5–0.6 tons per acre. Collecting 30% of all the cobs results in 10–15 million dry tons of feedstock nationwide. The cobs are discharged with the husk and usually spread over the field surface behind the combine after the grain is removed. Collecting the cobs and husks as they are discharged from the combine is one possibility. The cobs do not touch the ground, so they are free of dirt, rocks, and other foreign material, a common problem when baling material left on the ground. Because of recent government and economic incentives, several agricultural equipment suppliers and processors are evaluating two methods for cob collection: r Pulling a caddy behind the combine to collect cobs as they are discharged after shelling.

Questions have been raised concerning whether this process will slow the harvest too much. r Collecting the cob with the grain and screening the grain enroute to storage. A key question that must be addressed is whether the additional step in the logistics infrastructure will be cost effective. Trials to look at both methods are in progress. Two companies evaluating the methods for feedstock are Chippewa Valley Ethanol Co. (www.cvec.com), who is working to replace natural gas, and Poet, Inc. (www.poetenergy.com) for conversion to fuel ethanol. Residue sales are limited to local markets, typically within a 50-mile radius. The bulky, lowdensity properties make them expensive to load and transport. The biomass is not fungible, so there is little spot market. Sales are arranged a year in advance many times with no contract and only with a handshake.

P1: SFK/UKS P2: SFK c02 BLBS075-Hood December 15, 2010

20:47

Trim: 244mm×172mm

2 Agricultural Residues

27

Another market for residue is leasing the harvested corn fields for grazing cattle. Most fields can provide several months of grazing after harvest. A rule of thumb is 1 cow/acre. Typical lease rates are $7–$10/acre. The only cost involved for the leaser is fencing the fields and providing drinking water for the livestock. A one-wire electric fence is generally used to contain the cattle. The fence can be easily moved from one area to another as the cattle consume the stover and is reused from year to year. Obviously, this use of agricultural residues competes with the ethanol and biobased products market.

2.2.2

Harvest Window

The harvest window for residues is the time between the grain harvest and the next field operation. For no-till fields, this is the planting time for the next crop. For tilled fields, fall tilling is usually done immediately after the harvest of the current crop, weather permitting. Fields must be dry enough to permit good ground conditions to support the equipment operations in tilled systems. According to the last USDA survey, 85% of the wheat acres and 80% of the corn acres are tilled (CTIC, 2007). The wheat ripens in early June and is harvested much earlier in the season than corn, which is harvested in mid-August through early September. Weather conditions are more favorable for field operations in June than in the fall, and therefore, wheat straw is generally more accessible for baling. Because corn harvest is later when days are shorter and cooler, wet weather may delay fields and crop from drying adequately, extending the harvest beyond 30 days. Additionally, the physiology of wheat straw is very similar to commercial hay and there is no special equipment required, while corn stover is not a homogeneous material and presents some issues in harvesting. For stable storage, the residue needs to be below 20% moisture or above 60% moisture. Between 20% and 60% moisture, microbial activity is high, thereby digesting the cellulosic material and reducing the yield. In the worst case scenario, the heat is not dissipated from the reaction and auto ignition of the material can occur. The physical structure of straw and the timing of harvest ensure conditions will be right for collection. Corn stalks are more difficult. The stalk contains about 30% more moisture than the grain and must be left in the field after the harvest to dry. Wet weather can prevent drying well past the time when the farmer would do fall tilling. Harvesting wet corn above 20% moisture is common. Its high value justifies drying in equipment designed for that purpose. The lower value of stover and its heterogeneous character prevent economic drying. Corn cobs are the exception as whole cobs dry well and store well in the open.

2.2.3

Residue Removal

The crop residue removal cost has two components: cost per acre and cost per dry ton collected. The more collected per acre, the lower the cost per ton. The quantity of residue for harvesting after the erosion requirement is met depends on cropping practices. Tillage greatly affects availability when following USDA guidelines for erosion control. No-till practice allows most of the residue to be removed, especially when cover crops are employed. In contrast, conventional tillage leaves less than 30% of the surface covered, and there is no excess residue. Baling dry material is common practice. Collecting the stover with existing equipment requires multiple passes through the field. Conventional baling requires additional passes to

P1: SFK/UKS P2: SFK c02 BLBS075-Hood December 15, 2010

28

20:47

Trim: 244mm×172mm

Plant Biomass Conversion

Table 2.3. Custom Bale & Haul economics for corn residues.

1:1 ratio, 15% moisture, sell Sale, $70/dt P&K nutrient credit, $10/dt Reduced field operations, $14/ac Total revenue increase Less stalk chopping (corn only), $9/ac Less raking, $6/ac Less custom baling $23/dt Handling, storage $5/dt Strinkage, 10% Hauling, 23 mile radius, $10/dt Net to farmer, $/ac

130 bu/ac

170 bu/ac

200 bu/ac

2 dt/ac $140 (20) 14 $134 (9) (6) (46) (10) (14) (20) $29

3 dt/ac $210 (30) 14 $194 (9) (6) (69) (15) (21) (30) $44

3.8 dt/ac $266 (38) 14 $242 (9) (6) (87) (19) (27) (38) $56

Assumptions: $70/dry ton delivered, 1 million acres, 23 miles radius collection site, 1 dt/ac left in field.

Table 2.4. One-pass harvest and transport economics for corn residues.

1:1 ratio, 15% moisture, sell Sale, $70/dt P&K nutrient credit, $10/dt Reduced field operations, $10/ac Total revenue increase Less one-pass harvest $18/ac Field to collection site transport $10/dt Handling, storage $6/dt Shrinkage, 3% Transport from collection site $7/dt Net to farmer, $/ac

130 bu/ac

170 bu/ac

200 bu/ac

2 dt/ac $140 (20) 14 $134 (18) (20) (12) (14) (20) $56

3 dt/ac $210 (30) 14 $194 (18) (30) (18) (21) (30) $86

3.8 dt/ac $266 (38) 14 $242 (18) (38) (23) (27) (38) $109

Assumptions: $70/dry ton delivered, 1 million acres, 3–13 miles radius collection site, 1 dt/ac left in field.

chop, rake, bale, and remove the bales from the field. Estimated baling costs are summarized in Table 2.3. Baling cost is about $30/dt for round or square bales at the roadside. Storing, stacking, and transporting bales add $14–$25/dt to the cost. The example assumes $70/dry ton delivered price. Collection is within a 23-mile radius encompassing 1 million acres. Costs are based on the average value from the 2008 Iowa Farm Custom Rate Survey. To achieve $50/ac pretax income, grain yields need to be 200 bu/ac, while leaving 1 dry ton in the field. One-pass harvest, bulk storage, and transporting the feedstock from regional collection centers is estimated to reduce the costs and nearly double the margins for the farmer but remains to be demonstrated (Table 2.4).

2.2.4

Residue Management

The emerging market for cellulosic feedstock and increasing yields of corn present new opportunities for better reside management. Continued trait improvement of corn by seed companies, including Monsanto (www.Monsanto.com), Pioneer (www.pioneer.com), and Syngenta

P1: SFK/UKS P2: SFK c02 BLBS075-Hood December 15, 2010

20:47

Trim: 244mm×172mm

2 Agricultural Residues

29

(www.syngenta.com), is expected to raise the yield 3 bu/ac/year in the near term. The ratio of stover to corn grain remains unchanged, producing an additional 150 lbs/ac of stover annually. Advances in breeding will boost the average to 300 bushels per acre by 2030. For the same 90 million acres in corn now, stover increases from 3.7 to 7.4 dt/ac, doubling stover to 600 million dt of residue to be managed. A major challenge is managing residue in an economically and environmentally sustainable way. An evolution in crop tilling practices toward no-till cropping is needed in order to supply adequate feedstock while complying with erosion guidelines and maintaining soil quality. No-till cropping is increasingly practiced but not widely utilized in regions of the country with the greatest potential to supply crop residues, as calculated based on their average yields of corn. Farmers need relevant information on the effects of biomass removal to establish a better basis for sustainable removal as biorefining opportunities emerge. A series of colloquies, or informal discussions, was held to discuss needs to accelerate cellulosic biofuels commercialization. Participants were a multidisciplinary group of industry leaders in a position to influence the future direction of the industry. Topics traced the supply chain from the farmer’s field to the marketplace. Discussion focused on increasing feedstock availability, logistical challenges to transport agricultural biomass, and financing commercial scale plants with new technology supplied with a nonfungible feedstock (Hettenhaus, 2006). All participants concurred that residue management needs to extend beyond the field, considering other portions of the supply chain. Additional infrastructure in collection, storage, and transportation are also needed, including equipment for one-pass harvesting and investments for alternatives to trucking collected material, including pipelines and short-line rail.

2.2.5

Ag Equipment Needs

Several general statements were made in the colloquies describing Agricultural Equipment needs: r To supply clean feedstock to the biorefinery, the feedstock should contain little to no dirt,

preferably the material collected never touches the ground. The dirt causes severe wear on piping, conveyors, and equipment. r To not slow down the grain harvest, crop planning is premised on collecting 10–12 acres per hour, completing the harvest within 30–40 days. The acres planted, the size of the harvest crews, and supporting equipment and infrastructure are sized to operate accordingly. If the harvest is slowed, there is increasing danger of losing a portion of the crop due to high winds and wet weather, exposing the crop to fungi and risk of more deterioration of the grain in the field. The harvest window is especially important for corn, since it is harvested late in the season with cooler weather and shorter days. Several one-pass prototypes are being investigated that do not slow down the grain harvest. The base case shown (Table 2.5) assumes the stalk and ear are removed together in one truck from the field. Grain is separated at the collection center. Locating collection centers at existing grain elevators, the “grain elevator model,” makes use of existing infrastructure to store the corn. These assets are often underutilized since much of local corn is sold to nearby ethanol plants. Land may need to be acquired at some locations to accommodate the additional cellulose storage.

P1: SFK/UKS P2: SFK c02 BLBS075-Hood December 15, 2010

30

20:47

Trim: 244mm×172mm

Plant Biomass Conversion

Table 2.5. One-pass harvester requirements: 200 bu/ac example. Harvest Days Stover Collected in Million dt

30

1 2

Harvesters required 28 21 17 56 42 34

40

50

To collect 1 million dry tons of stover in the example above, 30 one-pass harvesters would be required based on the following assumptions: (1) 12 ac/h average harvest rate, (2) 200 bu/ac average yield, (3) 20 hours per day operating time, and (4) two spare harvesters for back up. The number of harvesters drops as the harvest days are increased (Table 2.5). There are no spare harvesters included in the table values. To maintain the harvest rate at 14 ac/h, a 12 head combine in a 200 bu/ac field harvests 40 bu/min, filling a trailer to legal weight (20 tons) in 15 minutes. A one-pass harvester will fill the truck in 5 minutes. The number of trucks required per harvester triples, and the trucks are volume-limited (3,000 ft3 , 8 lbs/ ft3 , 12 tons as is). Depending on yield and distance, present corn harvesters operate with a minimum of 3–4 trucks/combine. Therefore, for similar distances, 10–12 trucks will be required per one-pass harvester. One-Pass Investment The list price of a 12 row, 375 horse power (hp) rotary combine is about $300,000. Assuming trucks can be contracted with no additional investment and the same harvest cost can be achieved when the additional handling is done off site, the investment in new one-pass harvest equipment is $9 million (30 × $300,000) per 1 million dry tons of stover collected. Large operators lease their combines, and if the economics work, the opportunity would warrant serious consideration. Baling Investment The equipment investment for baling in the same example as above, i.e., 1 million dry tons of stover in 30 days, is $15 million. The capital is the same for round or square bales (Table 2.6). Cost of equipment for moving the bales from the field, storing, and transporting are not included. The baling-related units required are based on the same values used for the one-pass harvest example: (1)feedstock, 1,000,000 dt/year, (2) yield, 200 bu/ac, (3) stover (1:1 ratio), 4.8 dt/ac, (4) cover left, 1.0 dt/ac, (5) stover baled, 3.8 dt/ac, and (6) area baled, 265,000 ac. The units required are shown in Table 2.7. They are based on the September 2009 University of Minnesota Extension Machinery Estimates for Net Cost of a New Unit and Work Performed, acres per hour. The stalk chopper is widely used now to help microbes attack stover residue. For baling, the stalks are chopped immediately after the harvest to accelerate drying. After the chopped material dries to less than 18% moisture, it is raked to form a windrow. The windrow reduces the number of passes the baler makes across the field, improving the baling rate. Windrows

P1: SFK/UKS P2: SFK c02 BLBS075-Hood December 15, 2010

20:47

Trim: 244mm×172mm

2 Agricultural Residues

Table 2.6.

31

Capital equipment investment for baling. Harvest Days

Capital Equipment

Unit Cost $(000)

Stalk chopper Rake Square baler Round baler Tractor, 60 HP w/rake Tractor, 60 HP w/rnd baler Tractor, 130 HP MFWD, chop Tractor, 130 HP MFWD, sq baler Chop, rake, square bale only Chop, rake, round bale only

20 27 75 21 26 26 106 106

Table 2.7.

30

40

50

Capital required $ (Millions) 1.1 0.8 1.7 1.3 2.1 1.5 2.3 1.7 1.5 1.1 2.9 2.1 5.0 4.5 2.9 2.2 15.3 11.4 15.4 11.6

0.7 1.0 1.2 1.4 0.9 1.7 3.6 1.7 9.1 9.3

Baling equipment requirments. Harvest Days

Harvest hours Collection rate required Equipment employed Stalk chopper Rake Square baler Round baler Tractor, 130 HP MFWD Tractor, 60 HP

20 h days ac/h Equip. rate ac/h 7.8 6.8 16.0 4.0 NA NA

30

40

50

600 439 Units required 56 64 27 110 83 174

800 329

1000 263

42 48 21 82 63 130

34 39 16 66 50 105

once wet due to rain or even a heavy dew are difficult to dry back down, so raking is scheduled just prior to baling. Tractors are multipurpose. A 60 HP tractor can rake and pull the round baler. The 130 HP Mechanical Front Wheel Drive tractor can handle all the assignments, but is only required for the stalk chopper and the square baler. Harvest Choices While baling can fit other residues and energy crops, for stover, the collection choice appears to favor one-pass harvest. Several reasons drive this conclusion. First, the grain is harvested when it is mature, usually after drying below 20% moisture. However, in a wet season, the grain is harvested when field conditions permit and dried in specially designed dryers. One-pass harvests the wet stover material with the grain. The stover is stored wet, adding water as required to be at 60% or more moisture. In contrast, baling requires material to be 18% moisture or below. Wet weather keeps residue moisture too high for baling and feedstock remains in the field. One-pass harvesters replace existing combines for corn harvest. The grain and stover are collected in one pass (with more trucks). Stalk chopping in one-pass harvested

P1: SFK/UKS P2: SFK c02 BLBS075-Hood December 15, 2010

32

20:47

Trim: 244mm×172mm

Plant Biomass Conversion

Table 2.8.

Bale staffing requirment, 1 million dry tons.

Operation

Positions

Stalk choppers Rakes Rotary balers, 30% of balers Square balers, 70% of balers Baling support crew, 15% Field operators, bales left in the field

56 64 33 20 27 200

fields is eliminated. Conventional combines continue to be used when residue is baled. The baling equipment is an additional investment and its utilization is weather dependent. Second, during the seasonal harvest, operations are planned to continue around the clock, supplementing the regular workforce with all able-bodied persons including spouses, retirees, students during nonschool times, and contract labor. There is little to no slack, as 12 or more hours per day are normal. Any additional work, including harvesting, transporting, and storing residues, will require additional people. For one-pass harvesting of 1 million dry tons of stover in 30 days, farmers must plan for four more trucks per combine, 30 harvesters, and 120 additional trucks (240 drivers, as workers would rotate through positions working 24/7) to remove the grain and stover from the field to collection centers. Replacing existing combines with one-pass harvesters is a trade-off for operators. Support crews are deployed for combines now. A similar crew would support one-pass harvest. To match existing combine harvest rates, one-pass harvesters require more trucks—one loaded every 5 minutes for ear and stalk in contrast to one every 20 minutes for grain. When baling stover, 1 million dry tons produces 2 million bales. Nearly 200 additional positions are needed to operate the equipment (Table 2.8) Since the work week is 24/7, 400 people are required to staff the baling operation not including administrative needs. Three operations are employed: chop, rake, and bale, leaving the bales in the field. Stalk chopping follows the combine to hasten the stalk drying. When the residue reaches the required moisture, a baling crew is deployed. Raking and baling are done together, collecting the windrow immediately to prevent it from getting wet. Heavy dew on a cloudy day or a local shower will add moisture and delay baling. Thus, advance scheduling is difficult. In this example, 30% are round bales and 70% square bales are produced. The net wrapping protects round bales from weather. They can be left in the field for collection until the next field operation. For no-till, this is spring. Square bales are not protected and must be removed from the field and stacked. In most areas, they must be covered to protect them from the weather. This additional movement is not included in the above operations. When wet weather curtails collection, there are limited other assignments for the baling crews, and their fixed costs continue to accumulate.

Soil Quality and Compaction When making a choice to remove stover, the impact on soil quality is a serious factor. Traffic in the field can cause excessive soil compaction, damaging the soil structure. Compacted soil inhibits root growth, increases water runoff, and reduces the flow of nutrients to nourish the plant . . . all reducing crop yields. In both harvest cases, more traffic will occur on the field.

P1: SFK/UKS P2: SFK c02 BLBS075-Hood December 15, 2010

20:47

Trim: 244mm×172mm

2 Agricultural Residues

33

Compaction can be avoided in several ways, including wide tires that have better distributive weight, limiting traffic to tracks between rows and tilling the compacted area. Tilling is the least attractive, because of added cost and it releases soil organic matter.

2.2.6

Operating Costs

Farmers make annual crop planting and selling decisions based on their historical operating costs and yields, and the expected market for their crops that year. The average age for farmers is 62. Over the many crop seasons, they have accumulated much experience, and they know their operating costs well. Their focus is increasingly on cost of inputs, especially fertilizer, emphasizing costs per acre, not yield per acre. Farmers are generally conservative in their approach, diversifying their fields and crops to produce a profit across the board, in dry years and in wet years. Changes for existing crops are adjustments made in an evolutionary way, trying several rows, then several acres, increasing acreage after results meet expectations. When considering a new practice—no-till, drill planting, removing crop residues, and growing energy crops—many questions arise. The answers vary depending on many conditions, including the soil types, crop system needs, and the weather. Accumulating the information to arrive at a satisfactory answer requires years.

2.2.7

Residue Nutrient Value

The present market for residues—grazing, bedding, and animal feed supplement—recycles the nutrients back to the soil. With grazing, nutrients remain in the same field. Bedding and manure from feedlots can be recycled to local fields where needed. The major nutrient components in the residues are phosphorus, potassium, and nitrogen. The phosphorous and potassium content (P&K nutrients) in straw and stover is typically 0.1% and 1%, respectively. The composition varies depending on the soil and local conditions. Rain quickly washes out these soluble nutrients. If a dry season, the P and K value removed with the stover is $10 per dry ton, $8 for the K, and $2 for P. The nitrogen fertilizer value is more complex and depends on local conditions. The N content in the stover is 0.5%–2.0% depending on the length of time in the field after the plant has matured. However, there is conflicting information regarding its value. If the residue is plowed under the surface, microbes desire a 10/1 ratio of C/N for breaking down residue. Since the C/N ratio of straw and stover is 40–70/1, 20 lbs N fertilizer addition per ton of residue is recommended to avoid denitrification of the next crop. If left on the surface, there is some evidence that shows that the same N deficiency occurs, but the results are not conclusive.

2.2.8

Land for Energy Crops

Energy crops provide another source of feedstock. The quantity depends on the incentives for replacing the existing land use. The Corn Belt is an unlikely area for growing energy crops due to the profitability of current cash crops. See Chapter 4 for a discussion of energy crop production.

P1: SFK/UKS P2: SFK c02 BLBS075-Hood December 15, 2010

34

2.2.9

20:47

Trim: 244mm×172mm

Plant Biomass Conversion

Farmer Outlook

Farmers embrace change at a conservative pace. Will farmers bother with supplying corn stover, the most abundant cellulosic feedstock available? Moreover, how can they participate in the value-added prospects? Farming, efficiently supplying commodity crops at a low cost, planting improved seeds with better traits, and employing improved crop practices with equipment that covers more acres per hour remains the primary focus. Removing the residues will be approached cautiously. When results are proven, expansion on their lands will occur. On average, farmers own 40% of the land they farm, renting the rest. Landlord approval is important to increase supply. Agricultural equipment and seasonal demands are planned around the total acres. Changing crops or crop practices on the rented lands is usually done with the consent of the landowner. Agreement on land use is in their mutual interest—keeping the equipment utilized and the rented land farmed. Downstream Value-Added Prospects: Farmer-owned dry mill ethanol plants are a shining example of how farmers can prosper from downstream “value-added” processing. In the areas surrounding these facilities, the farmers have enjoyed a stronger market for their corn since the facilities began operation. In addition, operating margins from their ethanol plants have increased due to higher gasoline prices. The farmer-investors and others in the community are benefitting from the increased dividends and higher value of their investment. The local economy has also benefitted from the jump in local expenditures to improve and expand plant operations. The corn to ethanol process was well proven when farmers began to invest in the dry mill plants. States, especially Minnesota, added incentives for farmer ownership. Similar programs may help to include the farmer in the value-added downstream prospects.

2.2.10 Crop Research and Development To supply the cellulosic feedstock market, new crop systems are needed. Energy crops and present cash crops that have a cellulosic feedstock component offer new opportunities for growth, but bringing a new crop to the market requires 5–7 years in the United States. A “disruptive” technology takes 10 years. To improve crop breeding, regional centers that conduct parallel crop development programs are needed. In addition to switchgrass, the possibility of other new feedstocks including cover crops, forage soybeans and sorghum, high fiber sugar cane, miscanthus, willows, and poplars are some that are being investigated. Improving plant traits, including enzymes in the plants to lower the processing costs, are related projects that could accelerate commercialization.

2.3 Feedstock Logistics Cellulosic feedstocks face several logistical hurdles, including low density for transportation and storage stability. Additional trucks and drivers are required to transport the bulky material, increasing the traffic at a time already prone to congestion. Collecting the annual supply of feedstock during the fall harvest to supply the facility over the next years requires secure storage conditions.

P1: SFK/UKS P2: SFK c02 BLBS075-Hood December 15, 2010

20:47

Trim: 244mm×172mm

2 Agricultural Residues

Table 2.9.

35

Densification of dry material. Feedstock Density (dry lbs/ft3 )

2.3.1

Form

Operating Cost $/dt

Bulk Milled Baled Pellets

NA $15–$25 $25–$30 $25–$30

Corn Stover

Straw

Cobs

40

4 8 8–12 25–35

8–10

Bulk Density

Grain density is about 40 lbs/ft3 . Stover or straw bulk density is 4 lbs/ft3 , thus requiring 10 times the volume to transport and store it, unless compacted or finely milled. A trailer with the standard dimensional limits, 3,000 ft3 , holds 60 tons of grain, nearly triple the usual legal load of 20 tons of baled or ground biomass. Thus, hauling logistics will be difficult, but critical, to resolve. During harvest season, most states provide an overweight allowance for trucks since the crops must be harvested during a short time period and varying water content makes it difficult to estimate actual harvest weight. The temporary limit is based on maximum safe weights for roads, especially bridges. For example, Wisconsin permits a 15% overweight allowance between September and November. Nevertheless, biomass bulk density is too low to take advantage of the higher weight limits. Chopped corn stalks or straw is about 4 lbs/ ft3 . Without some compaction or grinding, the same trailer would contain just six (6) tons, more than tripling the transportation costs of grain. Permitting larger trailer dimensions would help, but then equipment is nonstandard, limiting the utilization after the seasonal harvest. Increasing the bulk density—incurring cost for baling, pelletizing, or compressing in other ways—can lower transport cost. The most economic trade-off depends on transportation distances, and the costs of handling, storing, and processing. Densification Processing the material removed from the field through a mill or grinder to reduce its size can increase density to 8 lbs/ft3 (Table 2.9). The same density, 8 lbs/ft3 , can be achieved with a one-pass harvest, collecting corn stalks with the ears. Baling, round or square bales, further increases density to 8–12 lbs/ ft3 . Pellets are the densest at 25–35 lbs/ft3 . However, they can be friable, i.e., easily broken apart if not processed properly or if handled in a rough manner. Pelletizing cost Transportation cost is offset by densifying the feedstock. Dry feedstock is usually baled when removed from the field after harvest. Dense square bales, pellets, and cobs readily fill trucks to capacity while remaining in legal dimension limits. Round bales are most efficiently handled by a “load and go” wagon that picks bales off the field and carries them over the road with a high-speed tractor in less than a 10-mile radius. Pelletizing occurs in series. Bales are broken and pelletized at a collection center and then transported to their final destination. Pelletizing requires significant investment and the

P1: SFK/UKS P2: SFK c02 BLBS075-Hood December 15, 2010

36

20:47

Trim: 244mm×172mm

Plant Biomass Conversion

relatively high fixed pelletizing cost depends on the volume. For example, Green Circle’s pelletizing plant in Cottondale, FL, recently began production. The $65 million plant has 13 pelletizer machines scaled to produce 550,000 tons of wood pellets per year from regionally sourced yellow pine. Extending this example, pelletizing 1 million dry tons annually would require a $100 million investment (Kotrba, 2009).

2.3.2

Storage

Storage stability is mostly a function of moisture. Many methods for storing crop residues have been investigated over the years. An early patent (Lanthrop and Munroe, 1926) claimed that one of two conditions must exist during storage for good fiber properties and minimum storage losses: either the moisture content must be below 20% during storage so that the microbial activity is nearly dormant, or the material must be kept wet, above 60% moisture. Dry storage of cellulosic feedstock is seen as the most likely practice for supplying biomass in the United States today. Baling and bale handling equipment are readily available from Ag machinery suppliers. While baling adds cost, it supplies needed compaction for economic transport. Wet storage of silage is a common practice for feeding cattle. Wet storage is also used for bagasse, the fiber residue left after removing the juice from sugar cane. Dry Storage The first industrial scale application of crop residues occurred in the sugar cane industry in the last century. Bagasse exits the sugar cane refinery at 50% moisture, too damp for stable storage. Some is burned to supply process energy. The excess is used as a component in building materials like fiberboard and as a pulp mill feedstock. Production of building materials is a dry process. Producers pursued ways to economically dry the feedstock for storage. This effort resulted in using the heat from microbial fermentation to dry bales from 50% to less than 20% moisture (Munroe and Lathrop, 1933). The bales were sized and stacked to dissipate heat and acid fumes without fiber damage (Figure 2.1). Sheltered from the weather, bales are kept for several years without serious deterioration or fiber loss (Hay and Lathrop, 1941; Lathrop and Munroe, 1926). Classical bale storage is pictured in Figures 2.2 and 2.3.

Figure 2.1.

Bale stacking, about 1930.

P1: SFK/UKS P2: SFK c02 BLBS075-Hood December 15, 2010

20:47

Trim: 244mm×172mm

2 Agricultural Residues

Figure 2.2.

Bale storage, 1930–1960.

Figure 2.3.

Bale storage, 1997 winter.

37

This original dry storage method was used for more than 40 years. However, a change to wet storage occurred in the 1960s due to increasing recognition of dry storage disadvantages, including that the bales were relatively small, weighing 250 lbs “as is,” mechanical handling was slow and costly; the bales had to be precisely stacked to vent fumes and dissipate heat; procedures were labor intensive; several months were required to dry bales from 50% to 15% moisture; and as fire loss was high, fire insurance costs were increasing. While some progress in handling has occurred, experience in baling, storing, and transporting has demonstrated that the issues above are still unresolved today. Corn stover bales harvested in 1997 for furfural feedstock are shown in Figures 2.3–2.6. The bales survived the cold weather well (Figure 2.3). However, warm weather disclosed serious decomposition of round and square bales (Figures 2.4 and 2.5). Inspection of the round bales showed they were loosely made. Evidently, that bale operator had not tightened the belt to better compact the bale. The loosely packed bale absorbed excessive water so that microbial activity was high since air was more able to diffuse into the bale. Consistent, dense bales and covered storage on well-drained pads can minimize the feedstock loss. Other hazards remain, such as fire. Once ignited, bale fires cannot be extinguished. Figure 2.6 shows the results from a fire in Harlan, IA, started when a small flame caused by a welder’s spark blew into the stacks on a windy day. The blaze destroyed much of the inventory (Figure 2.6), which burned for several weeks. The Harlan Tribune reported “by far the most notable incident during the year was the corn bale fire at Penn Chemical Company, south of Harlan, on October 7–8, 1999. Thousands of corn bales burned, keeping the department, as well as other area fire departments, on call for both days” (Harlan Tribune, 2000).

P1: SFK/UKS P2: SFK c02 BLBS075-Hood December 15, 2010

38

20:47

Trim: 244mm×172mm

Plant Biomass Conversion

Figure 2.4.

Round bale loss.

Figure 2.5.

Square bale loss.

Figure 2.6.

Storage area after bale fire.

P1: SFK/UKS P2: SFK c02 BLBS075-Hood December 15, 2010

20:47

Trim: 244mm×172mm

2 Agricultural Residues

39

Process water

Feedstock

Washshred(pump)

Wet storage pad

Cellulosic process

Circulation liquor Water management • Remove dirt • Recycle nutrients

Figure 2.7.

Wet storage platform.

Figure 2.8.

Wet storage mound with pumping arm.

Wet Storage Since pulping is a wet process, companies in the pulp and paper industry investigated wet storage of nonwood fibers for feedstock. The results were more successful than dry storage. Even companies with a dry process abandoned bales in favor of wet storage. Wet storage of sugar cane bagasse has been in wide use on a commercial scale since 1960 (Atchison and Hettenhaus, 2002). Unlike ensiling forage for animal feed, the biomass material is typically slurried to 3% solids and piped to a storage pad (Figure 2.7). The liquor drains through the pile and is recirculated until a height limited by the pump performance is reached—typically more than 100 feet (Figure 2.8). The Imperial Young Farmers and Ranchers Assn, Imperial, NE, validated wet storage for corn stover, as part of a USDA funded project (Hettenhaus and Mosier, 2010). A 700 dt pile was built in 2005. Samples over the next 8–16 months were pretreated, hydrolyzed with cellulase enzymes, and fermented to ethanol. The feedstock from wet storage showed improved results over dry material since inorganic solubles were dissolved in the circulating liquor when the pile was constructed (Figure 2.9).

P1: SFK/UKS P2: SFK c02 BLBS075-Hood December 15, 2010

40

20:47

Trim: 244mm×172mm

Plant Biomass Conversion

Figure 2.9.

Corn stover wet storage; Imperial Young Farmers & Ranchers Assn, Imperial, NE.

Adapting wet storage to a crop that has “damp” feedstock like high fiber cane, silage sorghum, and corn stover fits well into a wet biorefinery process. Harvesting when the crop is mature, including the excess wet stover, can reduce the weather risk and soil compaction in addition to lowering cost.

Storage Comparison A comparison of dry bale and wet bulk storage shows there are significant advantages for the wet method in most categories (Table 2.10).

Storage Area Wet storage density is 12.5–14 dry lbs/ft3 of average pile density (Bruijn et al., 1974; Moebius, 1966). The height of a wet pile is only limited by the pump head capacity, how high is it economic to pump? Bales require ten times or more the wet storage space due to bale stacking limits, access corridors, and a measure of fire protection. The total area required for 1 million dry tons is about 500 acres for square bales. Annual rent at $300/acre would be $150,000. Square bales, 10 dry lbs/ft3 , can be stacked about 300 lbs/ft2 or 30 feet high before the weight begins to compress the lower bales, causing the stack to shift in storage and possibly fall. Round bales, stacked three high require about 140% more storage area, 170–200 dry lbs/ft2 . Using “cotton modules” to compress the feedstock, even more area is needed for a

P1: SFK/UKS P2: SFK c02 BLBS075-Hood December 15, 2010

20:47

Trim: 244mm×172mm

2 Agricultural Residues

Table 2.10.

41

Dry and wet storage comparison.

Parameter

Dry (Bales)

Wet Storage (Ritter Method)

Dry density, lbs/ft3 Storage areas Storage loss Foreign matter and soil nutrients Solubles removal Weather risk Fire hazard Investment Storage quantity

8–12 10× >5% High Process residue Rain High Low to high Small, mostly farm use

12–14 1× <5% Low Storage liquor Extreme cold None Medium to high Large, bagasse for pulp

single layer. Following insurance underwriting guidelines, bales are limited to 4,000 tons/pile with a minimum of 30 m distance between piles. In contrast, with wet storage, a 333,000 dt pile requires 15 acres. For 1 million dt storage, just three piles or about 50 acres is needed at $15,000 annual rent at $300/acre. Storage Loss Bales are adversely affected by wet weather and without shelter can decompose and break apart (Figures 2.2 and 2.3). In dry climates, for example, Idaho, Colorado, and Oklahoma, losses can be 5% or less. However, most of the crop residues are in the Corn Belt, which has a wet climate. In Iowa, for net-wrapped round bales at 6’ × 5’, 30% of the mass is in the outer 4 inches and 25% weight loss can easily occur in one season. Stored inside barns, both round and square bales had 14% weight loss over 10 months in Eastern Canada (Billy, 2001), although the overall composition remained nearly the same. For wet storage, the major losses are the 5%–8% solubles removed during storage. Typical cellulose and hemicellulose losses reported by the pulp and paper industry are 1%–3% (Atchison, 1971; Moebius, 1966; Salaber and Maza, 1971). Surface loss is dependent on the total surface exposed relative to the stored volumes. The higher the pile is, the lesser the surface exposure and surface loss. Foreign Matter and Solubles Bales harbor foreign material that can be deleterious during storage and processing. Wet storage has proven to remove dirt, foreign matter, and solubles over time. Removing the nutrients during storage and returning them to the fields is much preferable to processing the bales and disposing of the process ash. Less solubles in the wet, stored feedstock with the absolute values of holocellulose and pentosans, unchanged increases the facility capacity up to the distillation step: 7% removal opens up 7% more pretreatment, hydrolysis, and fermentation capacity. However, more distillation equipment is required for the increased load. The process and economic impact is significant. A nominal 100 million gallon plant increases 3 million gallons to 103 million gallons annually. NREL’s Aspen model indicates a cost decrease of $0.04/gallon or $4 million annually. The improvement assumes that half of the solubles are removed in storage and an additional 3.5% of the feedstock is processed.

P1: SFK/UKS P2: SFK c02 BLBS075-Hood December 15, 2010

42

20:47

Trim: 244mm×172mm

Plant Biomass Conversion

Disposal of Nonvolatile Solubles Crop residues contain up to 12% solubles, including valuable soil nutrients. Soluble soil nutrients, especially P and K, are removed with the bale, depleting the soil. While there is considerable variation in the composition (Templeton et al., 2009), their average value is $3.20/dt (Glassner et al., 1998). For sustainable harvest, nutrients contained in the stover must be replaced. In dry climates, farmers often run the irrigation system before baling to wash some of the solubles into the field. Unless removed, they are processed, becoming part of the process residues. Presently, it is uncertain if the process residue must be landfilled or may be used as a soil conditioner. Either route is expensive. Annual landfill cost is $1.6 million for a 2,000 dt/day facility at a nominal $20/dt. In wet storage, most of the nonvolatile solubles are removed before processing. Excess liquor is high in nutrients and can more likely be returned to the soil, free of process ash. In either case, transportation can be a significant cost factor. More investigation is needed to determine the cost and disposition of each.

Storage Risk Weather and fire are hazards for dry storage. Bales must be protected from absorbing moisture to avoid loss. Depending on the local climate and length of storage, roofed shelters are generally preferred over wraps or tarps. Round bales may be mesh wrapped, but wrap disposal presents another cost. Fire is an additional hazard for dry material, in storage and in transit. Bale fires, once started, result in a total loss (Figure 2.4). They burn slowly with much smoke and can last for weeks. Nearby areas often are obscured and at a minimum inconvenienced. At worst, roads may be closed and neighborhoods evacuated. A hay bale fire shut down Interstate 25 near Colorado Springs, CO, for hours despite the efforts of the local fire departments, including the nearby Air Force Academy, to control the flames and smoke On the other hand, wet storage piles are not affected by the weather and at 75%–80% moisture and are not flammable.

Industry Proven Wet storage of cellulosic feedstock is industry proven. Bales are largely for on-farm use and have been used for small industrial applications, mostly for particleboard. Large applications that rely on bales have historically encountered problems including cost, quality, storage deterioration, fires, and adequate supply (Lengel, 2001). These same problems encouraged the nonwood fiber pulpers to move to wet storage more than 50 years ago with much proven and well-documented success. Although stover and straw are expected to perform in a similar manner, further validation is required.

Investment Required Investment for dry bale storage is well established. Wet storage investment remains to be investigated. Bale storage cost varies from purchasing tarps to cover stacked square bales or plastic wrap for round bales to covered buildings. An uncovered stack of net wrapped round

P1: SFK/UKS P2: SFK c02 BLBS075-Hood December 15, 2010

20:47

Trim: 244mm×172mm

2 Agricultural Residues

43

bales on the ground has negligible investment. Sheltered bale storage investment can be high, depending on the structure. Bale storage systems were estimated for square bales of rice straw (Huisman et al., 2005). Results showed that tarps were favored at an annual cost of $7–$10/dt for short-term storage with no investment other than tarps, which were assumed to have a three-year life. For longerterm storage, pole barns are favored at $5/dt, including the land leasing cost. Tube wrapping has costs similar to storage in buildings, but allow greater losses of biomass in longer-term storage. Tubes are not stacked and have a large land area requirement. Metal buildings give overall better protection of quality in long-term storage, but increase storage cost $1/dt–$6/dt, adding $1 million in infrastructure cost for a 100 MGPY plant. Truss arched tarps are only slightly higher cost than conventional tarping under the best conditions but offer better long-term storage options. Land rental for storage averages $300 ac/year A longer-term investment is justified by reducing yield loss of the bales otherwise exposed to the weather. Leasing buildings reduces the upfront capital and establishes a long-term relationship, typically 15 years that corresponds to the useful life of the biorefinery. At the bioprocessing facility, the bale unloading, interim storage, and handling system designed for NREL’s model has $13 million in capital cost, or $3.00/dt. The operating cost is $2.60/dt based on 12 operators 24/7 to handle the bales, adding a total of $5.60/dt to the feedstock cost (Aden et al., 2002; Harris Group, Inc., 2000). Wet storage investment cost needs to be investigated and evaluated as part of the supply chain, including the following: r r r r

Modified cropping practices compensating for the residue removal One-pass harvest equipment and logistics Transport from the field to the collection center Collection center logistics with truck unloading, separation, circulation basins, flumes, pipes, and other equipment r Disposition of the nutrient-containing liquor r Preprocessing and pretreatment r Transport to the biorefinery.

2.3.3

Regional Biomass Processing Centers

Some industrial participants in the colloquies were confident that their technology is ready for a few feedstocks, including paper mill sludge, wood chips, crop residues, and forest trimmings. Nearly a dozen commercial and semi-commercial scale projects with a mix of processes, and feedstocks are moving ahead as a result of matching government funds, grants, and loan guarantees to test conversion to renewable fuels.

Process Types Emerging cellulosic validation projects are taking three forms: first, co-locating the new technology with proven technology on an existing site, e.g., the POET-Voyager Ethanol Plant, Emmetsburg, IA; second, greenfield site, co-locating new and proven technologies, e.g., Abengoa; third, building on a greenfield site with stand-alone new technology, e.g., Iogen and Range Fuels.

P1: SFK/UKS P2: SFK c02 BLBS075-Hood December 15, 2010

44

20:47

Trim: 244mm×172mm

Plant Biomass Conversion

450,000 ac

Figure 2.10.

15 mir

Imperial, NE, collection centers

Expanding an existing site has the advantage of operating infrastructure to support the new technology start-up and operation. Co-locating the two technologies lowers the associated fixed costs by providing a larger operating platform. For example, 80% of production will be from a proven process, with the other 20% from new technology. Recognizing the challenge of obtaining a reliable, economic, and sustainable feedstock supply at least some companies are pursuing projects that use a modular or regional collection approach encompassing a 10–20-mile radius. Companies include Terrabon, Pure Vision Technology, AFEX, UOP and Lurgi. The cellulosic feedstock is preprocessed, being pretreated at the collection sites. Grain elevator locations could make great collection centers because of their rail access. For example, AFEX estimates the capital cost for processing 900 dry tons/day at $11 million (Carolan et al., 2007). In Figure 2.10, each circle has a 15-mile radius and contains 450,000 ac, comprising a total of 3 million acres in the area surrounding the biorefinery. The total amount of economic and sustainable feedstock supplied will vary depending on the local conditions. If 30% of the acres provide 2.3 dry tons feedstock per acre, 300,000 dt is collected and stored at the collection centers during harvest. During the year, the collection centers preprocess and pretreat about 900 dt/day. The centers supply the biorefinery via truck or pipeline for further processing. About 30 miles of pipe are required for each collection center to connect to the biorefinery. In this case, 2 million dt from the 3 million ac is processed, producing 27 MGPY ethanol (90 gal/dt).

P1: SFK/UKS P2: SFK c02 BLBS075-Hood December 15, 2010

20:47

Trim: 244mm×172mm

2 Agricultural Residues

45

Initial Project Financing Commercial biofuels projects cannot be debt financed due to the associated risk of new technology.1 Relying on total equity financing for a $200 million plus facility is out of the question. For biofuel projects to move ahead, some equity coupled with government support in the form of government grants and government loan guarantees is required. Participants look at leveraging government support where possible. Some examples are discussed in Chapter 11 in this volume. A much larger federal and state government effort is required to meet RFS goals. Many approaches to commercial success are needed to find “the” most suitable processing paths for the different technologies, regional feedstocks, infrastructure, and logistics. Opinions about how grants and loan guarantees should be applied differs across the spectrum of participants. Farmers in the Corn Belt view their investment as short term, the next crop planting choice. Farmers with traditional farm operations do not have a lot of cash on the balance sheet. They mainly work with annual contracts. These conflict with the biomass project investment needs, as production facilities have a time horizon of 20–30 years. The need for working capital invested in cellulosic feedstock for the 10 or 11 months between harvests seems to be taken in stride by the farmer—as long as there is a market for the material. However, not having a fungible feedstock increases risk. To mitigate the lack of a fungible feedstock and assist the farmer in gaining an equity stake in a biofuels plant, the State of IL has established a program to guarantee the value of the feedstock. The farmer can borrow against this value to take an equity interest in the project. Grain elevators, mostly organized as a farmer co-op, may provide infrastructure for establishing collection centers. They have an asset base, are located in 10–15 mile centers in grain growing areas, and have rail access. Farmer ownership in the co-op allows an affordable entry in the value chain—a possibility for smaller projects in the $10 million range needed for preprocessing and pretreatment. Agricultural equipment manufacturers are mostly spectators, waiting to see how the feedstock collection and logistics needs evolve. There are many feedstock supply options, including r r r r

Crop residues and energy crops Wet and dry feedstock collection and storage Bales and bulk storage One-pass vs. multiple-pass corn harvest.

John Deere and CNH are investigating several methods for one-pass grain and cob collection. Vermeer and others have improved baling equipment, with a focus on stover. Oxbo Corp, a specialty manufacturer is applying matching funds from the USDA to develop a prototype for one-pass corn grain and stover harvesting with the adage that the collection “will not slow down the grain harvest.” Biomass brokers and contract harvesters are in the same category as Ag equipment manufacturers, watching and waiting.

Cellulosic Process Developers Facility owners look to government grants and loan guarantees to finance the design, construction, and start-up of facilities. Tax credits are less attractive in most cases, since the benefit is on the back-end, after the project is online and profitable.

P1: SFK/UKS P2: SFK c02 BLBS075-Hood December 15, 2010

46

20:47

Trim: 244mm×172mm

Plant Biomass Conversion

Loan guarantees can be difficult to organize. Some project situations are difficult to work through successfully, especially for a partial guarantee of the total debt. The most workable government solution to funding projects is grants with matching equity. However, even with government grants, the developer carries an extra burden when dealing with new technology. Grants only fund hard assets. The development costs are usually not included in the R&D federal costs.

Sustainable Removal Erosion control and carbon sequestration models are not comprehensive enough to understand the whole story of sustainability of soil health. No processor wants to be associated with a harvesting operation without knowing the potential impact of removing the biomass, particularly corn stover. Crop residue removal was studied nearly 40 years ago after the first Arab oil embargo (Larson et al., 1972). To prevent soil erosion, some crop residue is needed on the surface. The amount required varies and depends on many factors such as degree of slope, length of slope, and management practices. These characteristics have been modeled by the USDA, are readily applied, and can be relied upon for controlling soil erosion. For soil carbon sequestration, the Century model is widely accepted. This model simulates C, N, P, and S dynamics through an annual cycle over time scales of centuries and millennia http://www.nrel.colostate.edu/projects/century5/reference/index.htm. A farmer friendly version, Comet VR, can model individual fields as a decision support tool (http://www.cometvr.colostate.edu/). The model is offered by the USDA Natural Resource Conservation Service and considers a wide range of cropping system rotations and tillage practices. It determines the effects of management and global change on productivity and sustainability of the farmer’s fields for different scenarios. The results are presented as 10-year averages of soil carbon sequestration or emissions with associated statistical uncertainty values. Estimates can be used to construct a soil carbon inventory. The results meet the requirements for voluntarily reporting GHG emissions according to Section 1605(b) of the Energy Policy Act of 1992, often referred to as the 1605(b) program. The Chicago Climate Exchange makes payments to the landowner based on the COMET VR carbon sequestration results for Agricultural land and Rangeland (CCX, 2009) These studies continue to provide a better definition of the sustainable amount that can be removed—moving past erosion and carbon sequestration just in the soil. Research needs to be extended to include emission of greenhouse gases such as nitrous oxide, methane, and carbon dioxide for various locations, crops, and management practices. For example, nitrous oxide has 320 times the impact of carbon dioxide in the atmosphere. Its significance must be accounted for along with alternative applications down the value chain. In addition, removal studies need to look at collecting the whole plant—both the grain and the remainder of the plant on the surface. Information for Life Cycle Analysis and Life Cycle Inventory are especially needed. The impact on greenhouse gas emissions from removing some of the residue is currently being evaluated, but the results are incomplete.

Plant Science Opportunities The time to market is measured in years. Perennial crops require 2 years to establish before harvesting and gaining cash flow. Thus, financial incentives are needed for the grower to plant these crops. The major seed companies have a focus on improving seeds for current cash crops:

P1: SFK/UKS P2: SFK c02 BLBS075-Hood December 15, 2010

20:47

Trim: 244mm×172mm

2 Agricultural Residues

47

corn, soybeans, and cotton, but are watching cellulosic developments. They currently offer corn seeds with enhanced ethanol yield potential. As the biofuels market develops—especially for corn stover—one expects them to screen their corn seed germplasm for phenotypes that enhance cellulosic fuel potential. The application of biotechnology to plants opens up many possibilities: increased yields of biomass per acre and per unit processed; improved composition, e.g., increased cellulose and lower lignin; and enzyme expression along with other coproducts. To realize these gains, extended time ranging from 3–10 years for research, development, and deployment is required. The tools for implementation are well proven. In a 3 year time frame, existing hybrids can be selected, which will improve field yields, and with process development support, genotypes identified, which make processing easier and lower production cost. In the midterm, enhanced processing yields could be achieved. Longer term, 7–10 years, the plant carbon can be redirected, changing the composition to favor desired products. Cellulase enzyme system expression or other coproducts could also be engineered for new products within this period. Private companies are leading this research on cash crops: corn, soybeans, and wheat. International consortia usually perform the cane research. Sugar cane is grown by vegetative reproduction, not seed, so intellectual property is difficult to control, requiring public rather than private effort. Public acceptance of engineered crops is a concern, but most believe it will dissipate, as more of the consumer benefits are demonstrated and no adverse effects come about for existing genetically modified crops. Biotech crop plantings continue to increase. In 2009, the global acreage of biotech crops continued to grow strongly, reaching 335 million acres, up from 312 million acres in 2008. From 1996 to 2009, the cumulative acres exceeded 2 billion acres (800 million hectares). It took 10 years to reach the first billion acres but only 3 years to reach the second billion acres. In 2009, an impressive 85% of the 90 million acre corn crop nationally in the USA was biotech—remarkably, 75% was hybrids with either double or triple stacked traits. Only 25% was occupied by hybrids with a single trait (ISAA, 2008).

Feedstock Business Model It is generally recognized that a successful operation requires a win-win relationship between the local farmers and the processor. Unlike corn, the biomass feedstock is bulky and transporting it long distances is not economically feasible. Local sourcing of feedstock material is needed to keep transportation costs minimal. On the basis of the colloquies participants’ comments, considerable education is required to change management practices to be consistent with stover removal and to overcome value judgments that do not match up with facts. A significant outreach effort will likely be needed to ensure collection of an adequate amount of feedstock. Some colloquy participants felt that having the farmer invest in the facility would be necessary. While desirable, others pointed out that a previous attempt to use this approach by Heartland Fibers failed. Over the 1990s, Heartland pushed the concept to farmers in MN, OH, IN, IL, and NE before closing down. No attempts to get farmers to invest in a new, yet-to-be proven venture were successful. Once the concept is proven—such as corn to ethanol—farmers are willing to invest as is shown by the proliferation of farmer-owned corn dry mill plants.

P1: SFK/UKS P2: SFK c02 BLBS075-Hood December 15, 2010

48

20:47

Trim: 244mm×172mm

Plant Biomass Conversion

With biomass, the local connection between farmer and processor is similar to the sugar cane model, simplifying identity preservation. NREL analysis shows significant variation in the stover feedstock composition, which in turn impacts processing cost up to 20% (Ruth and Thomas, 2003). It appears the economic effect will require payment based on composition, ease of processing and product yield. Complexity may increase, depending on the co-products. However, if the variation in composition is a result of environmental conditions, the farmer may not be able to make a living or profitably harvest the biomass, creating an unstable supply.

Storage and Inventory Ownership No processor wants to have a large inventory listed as an asset with just one annual turnover. Most corn millers operate with a 3–7 day supply of corn. The bulk of the inventory is expected to be carried by the farmer or a farmer-owned entity. On basis of conversations with local farmer organizations in IL, IA, and NE, the potential processors in the colloquies believed they could deal with a single supplier group as an intermediary between them and the individual farmers. Either a Co-op or other farmer-owned entity would serve as the “feedstock elevator,” storing the material until ordered by the processor and billing in the same way that grain elevators do now. Initially, this entity would likely be the vehicle to entice grower commitments to supply the feedstock, working closely with the processor and other local stakeholders such as the USDA NRCS and Extension staff. The assignment is a sizable task. About 250,000 acres are needed when using 3 t/ac collected feedstock, leaving 1 t/ac in the field for erosion control. If each grower offers 200 acres, 1,250 growers would be required.

Feedstock Supply The feedstock supply requires an inventory cost outlay of $20–$40 million for crop residues. Some of this cost is expected to be carried by the “grain elevator-like” biomass supplier. As most, if not all, of the harvesting, collection, and storage of the feedstock needs to be performed prior to the facility start-up, feedstock suppliers need to be convinced the project is real before investing resources into the venture.

2.4 Conclusion There are multiple sources of cellulosic feedstock, and corn stover offers the largest opportunity for crop residues. The stover surrounds existing corn to ethanol facilities. Expansion of these facilities to include cellulosic processing is a likely route for their growth. Their existing infrastructure facilitates expansion, debt financing, and permitting. The obstacles for a greenfield plant are greater, with no infrastructure, more capital to raise, and a longer permitting process. The feedstock is in the field, available for collection. More than 300 million tons were produced in the 2008–2009 crop year. Seeds in the pipeline will increase corn grain yields and stover more than 50% over the next decade. Removing the stover is better than plowing it under in many areas where stable soils limit erosion or when cover crops are included in the crop system.

P1: SFK/UKS P2: SFK c02 BLBS075-Hood December 15, 2010

20:47

Trim: 244mm×172mm

2 Agricultural Residues

49

When cellulosic conversion technology is proven, farmers will gradually adopt new cropping methods that reduce tillage and better manage residues through removal of the excess required for sustainable removal. A variety of methods that improve present baling practices will be used. Regional biomass processing centers will be established, building on existing grain elevators where practical. New infrastructure is required, reducing logistics and enabling the farmer the option to participate in the value chain.

Endnotes 1. Even with proven technology, debt financing is difficult without Wall Street-type investors. For projects with estimated costs of $250–$500 million, Regional Bankers usually limit investment to $20 million/project. 2. The costs are based on “as is” metric tons, equivalent to dry US tons with 11% moisture.

References Aden, A., Ruth, M., Isben, K., et al., 2002. Lignocellulosic biomass to ethanol process design and economics utilizing co-current dilute acid prehydolysis and enzymatic hydrolysis for corn stover. NREL Report No. TP-510-32438. 154 pp. Atchison, J. 1971. Review progress of bagasse for use in industry. In: International Society of Sugar Cane Technologist. Baton Rouge, LA: Franklin Press, pp. 1202–1217. Atchison, J. & Hettenhaus, J. 2002. Benefits of bagasses type storage of stover for supplying large biomass processing plants. Presented at Corn Utilization and Technology Conference, Kansas City, MO, USA. Billy, J. 2001. Corn stover in eastern Canada as raw material for the production of ethanol. Natural Resources Canada and Georges Le of “Ressources Naturelles Quebec”. Bruijn, J., Gonin, C., Mcmaster, L. & Morgan, R. 1974. Wet bulk storage of bagasse. Presented at International Society of Sugar Cane Technology (ISSCT) XV Congress. pp. 1793–1820. Carolan, J., Joshi, S. & Dale, B. 2007. Explorations in biofuels economics, policy and history: Technical and financial feasibility analysis of distributed bioprocessing using regional biomass pre-processing centers. Journal of Agriculture and Food Industrial Organization, 5. CCX, Offset Project Protocol, Sustainably Managed Rangeland Soil Carbon Sequestration, R Chicago Climate Exchange
, Page 18, September, 2009 http://www.chicagoclimatex.com/ docs/offsets/CCX Sustainably Managed Rangeland Soil Carbon Sequestration Final.pdf. CTIC Conservation Technology Information Center National Residue Survey, 2004, http:// www.ctic.purdue.edu/CRM/crm search/ (Accessed 26 April 2010). Glassner, D., Hettenhaus, J. & Schechinger, T. 1998. Corn stover potential for ethanol production. Presented at BioEnergy ‘98 Conference. Harlan Tribune. 2000. Hay, H. & Lanthrop, E. 1941. Storage of crop fibers and preservation of their properties. TAPPI Technical Association Papers. Hettenhaus, J. 2006. Biomass commercialization and agriculture residue collection. In: Kamm, B., Gruber, P. & Kamm, M. (eds.) Biorefineris, Biobased Industrial Processes and Products. Germany: Wiley-Vch Verlag GmbH & co.

P1: SFK/UKS P2: SFK c02 BLBS075-Hood December 15, 2010

50

20:47

Trim: 244mm×172mm

Plant Biomass Conversion

Hettenhaus, J. & Mosier, N. 2010. Validating reliable stover feedstock supply. Presented at 32nd Biotechnology Symposium for Fuels and Chemicals, Clear Water, FL. Huisman, W., Jenkins, B. & Summers, M. 2005. Storage Systems for Rice Straw in California. Davis, CA: Department of Agrotechnology and Food Science, University of California. Harris Group Inc., Process Design and Cost Estimate of Critical Equipment in the Biomass to Ethanol Process, Report No. 99-10600/13, Baled Feedstock Handling System, Revision 1W, October 11, 2000 http://www.p2pays.org/ref/22/21200.pdf (Accessed 26 April 2010) ISAAA. 2008. Executive Summary: Global Status of Commercialized Biotech/GM Crops: 2009—ISAAA Brief 41-2009. Kleinschmit, J. 2008. Minnesota Model 2.0—Innovation, diversification, and evolution in Minnesota’s farmer-owned biofuel sector. Presented at Third Annual Midwest Ag Energy Summit, Rosemont, IL. Kotrba, R. 2009. Closing the wood pellet gap. Biomass Magazine. Lathrop, E. & Munroe, T. 1926. Methods of preserving fiber for pulp-making purposes, US Patent 1572540. Larson, W., Clapp, C., Pierre, W. & Morachan, Y. 1972. Effects of increasing amounts of organic residues on continuous corn. Agronomy Journal, 64, 204–208. Lengel, E. 2001. A clarion call for common sense and reality in the composite panel industry, Eastern Canadian section. Presented at Forest Products Society Meeting. Moebius, J. 1966. The Storage and Preservation of Bagasse in Bulk Form, without Baling. United Nations, NY: Pulp and Paper Development in Africa and the Near East. Morris, D. 2008. Ethanol Program—Minnesota Model, The New Rules Project. Munroe, T. & Lathrop, E. 1933. Fiber storage and preservation. US Patent No. 1920129. Perlack, R. D., Wright, L. L., Turhollow, A. F., Graham, R. L., Stokes, B. J. & Erbach, D. C. 2005. Biomass as feedstock for a bioenergy and bioproducts industry: The technical feasibility of a billion-ton annual supply. http://handle.dtic.mil/100.2/ADA436753. Ruth, M. F. & Thomas, S. R. 2003. The effect of corn stover composition on ethanol process economics. Presented at The 25th Symposium on Biotechnology for Fuels and Chemicals, Breckenridge, CO; http://www1.eere.energy.gov/biomass/pdfs/34040.pdf (Accessed 26April2010). Salaber, J. & Maza F. 1971. Ritter biological treatment process for bagasse bulk storage. TAPPI Non-wood Plant Fiber Pulping Progress Report, October ed. Simmons, B. A., Loqu´e, D. & Ralph, J. 2010. Advances in modifying lignin for enhanced biofuel production. Current Opinion in Plant Biology, 13 (3), 312–319. Templeton, D., Sluiter, A., Hayward, T., Hames, B. & Thomas, S. 2009. Assessing corn stover compostion and sources of variability via NIRS. Cellulose, 16 (4), 621–639.

P1: SFK/UKS P2: SFK c03 BLBS075-Hood December 16, 2010

16:59

Trim: 244mm×172mm

Chapter 3

Growing Systems for Traditional and New Forest-Based Materials Randall Rousseau, Janet Hawkes, Shijie Liu, and Tom Amidon

3.1 Introduction A key question that must be asked as a nation is whether we can integrate our natural renewable resources in a manner that will not only be sustainable but meet the future increases in energy, transportation fuels, and food systems. Optimistically, the answer is a resounding “Yes,” but this response is one probably from the heart and not based on scientifically verifiable data. A reserved “yes” is probably more realistic, as we are facing a technological challenge that is truly a paradigm shift, moving from a predominately fossil fuel-driven economic system to one with a substantial renewable energy and fuel component. The original forest of the United States covered approximately 1 billion acres, but today approximately 740 million acres remain. The reduction of nearly 300 million acres has been fairly stable for 100 years, but growing population centers in the Southeast has begun to substantially impact forested acres. Recent predictions estimate that an additional 12 million acres will be lost in the southern United States by 2020 (Treasure et al., 2008). Forests have played a key role in the development of the United States from lumber production for towns to the production of a variety of items including chemicals, pulp and paper, and specialty products. The National Forest System has also played a critical role in the development of the United States. National Forests were initially designated in 1891 and today total approximately 193 million acres. Originally, these forests were focused primarily on timber management and timber harvests but in a fashion that was multidimensional with wood, recreation, wildlife, and water. Today, national forests are more focused on the management of non-timber values such as ecosystem services, including recreation and wildlife. State forests have also followed the trend set by the National Forest System focusing less on timber management and more on management of recreation values. More recent changes in the forest products industry have

Plant Biomass Conversion, First Edition. Edited by Elizabeth E. Hood, Peter Nelson and Randall Powell.
C 2011 John Wiley & Sons Inc.

51

P1: SFK/UKS P2: SFK c03 BLBS075-Hood December 16, 2010

52

16:59

Trim: 244mm×172mm

Plant Biomass Conversion

resulted in many companies selling their timberlands to either TIMOs (Timber Investment Management Organizations) or REITs (Real Estate Investment Trusts), and LLCs (Limited Liability Corporations). While intensive forest products management is viewed as a key component of revenue for these types of organizations, the long-term objectives vary by company. Landowner values have changed as well with forest management for timber revenue no longer being the sole priority. In addition, the increasing urban forestry interface has begun to greatly impact management techniques, especially on the East Coast. Forest fragmentation, urban expansion, the demand for increased recreation, public demand for ecosystem management, land user values, litigation, endangered species, loss of wetlands, pesticides, and air pollution are just some of the issues influencing the current debate over appropriate management of America’s forest resources. The shift in the National Forest system, traditional timber companies, landowner values, as well as public perception seems to point to the need for dedicated biomass plantations to feed the increasing bioenergy and biofuels industry in the United States and especially in the Southeast. The use of biomass for bioenergy and biofuels will play a critical role in the United States’ efforts to become energy independent. The US Department of Energy (DOE) indicated that biomass is the key component for the southeastern United States in building a renewable energy base. The agency noted the excellent growing conditions of the southeastern United States for biomass production of both annual crops as well as forest tree species. The billion ton report cited forest resources biomass such as logging residue, removal of excess biomass from forestlands, primary and secondary wood processing mill residues, pulping liquors, and urban wood residues. Although short-rotation woody crops were mentioned in the report, they were placed under the agriculture resource section. This was probably the result of the rotation length (i.e., time from planting until harvest) of these dedicated plantations, which is expected to range from 1 to 3 years. As the United States looks to move from a fossil fuel economy to that of a renewable economy, the use of woody biomass for energy and liquid fuel production will play a key role. While the use of woody biomass for energy and fuel has been discussed for a number of years, only moderate gains have been made. It is critically important that research and development in the field of biomass utilization be substantially increased as to ensure the use of this type of feedstock. This is particularly true for the southeastern United States where biomass is extremely abundant, due to the region’s expansive forests. Biomass can be defined as plant materials, such as wood or annual crops, that can be used in the production of either energy or fuel. Biomass is a renewable resource that can be considered fairly close to being carbon neutral. This near neutrality is particularly important as pending national legislation is focused on reduction of CO2 emissions. Carbon neutrality is achieved with the growth of biomass when absorption of CO2 during growth processes is equivalent to the amount of CO2 released as energy is produced by the plants. However, when fossil fuels are consumed in the establishment, management, and harvest of biomass, it often prevents biomass production systems from being completely carbon neutral. The Energy Information Administration (EIA) identified biomass, geothermal, wind, solar, and hydropower energy as renewable energy sources. Within the definition of biomass energy sources, the EIA specifically included wood and wood waste. A memorandum of understanding among the US Department of Agriculture, the US Department of the Interior, and the DOE defined woody biomass as trees and woody plants, including limbs, tops, needles, leaves, and other woody parts, grown in a forest, woodland, or rangeland environment that are the byproducts of restoration and hazardous fuel-reduction treatments. In addition, they also defined woody biomass utilization as the harvest, sale, offer, trade, and/or utilization of woody biomass

P1: SFK/UKS P2: SFK c03 BLBS075-Hood December 16, 2010

16:59

Trim: 244mm×172mm

3 Growing Systems for Traditional and New Forest-Based Materials

53

to produce the full range of wood products, including timber, engineered lumber, paper and pulp, furniture and value-added commodities, and bioenergy and/or bio-based products such as ethanol and diesel. The use of wood as a source of energy is not new and has long been practiced within the pulp and paper industry where bark, unusable chips, and black liquor have been the feedstock for energy production, thus reducing the cost of operation. However, it is mandatory that the forests be harvested in a sustainable manner. As related to forest dynamics, sustainability includes environmentally sound management and harvesting techniques that ensure that the benefits derived from today’s forests are not compromised, thus allowing the opportunities for future generations to benefit in a similar fashion. It is apparent that woody biomass will become an increasingly important renewable source for bioenergy and biofuel production. The sources of woody biomass will include dedicated plantations, logging residues, thinnings from traditional plantations, and salvaged material from disasters. Some of the benefits of woody biomass include: r Less capital-intensive conversion technologies employed for exploiting the energy potenr r r r

tial Attractive opportunity for local and regional energy self-sufficiency Viable alternative to fossil fuel use Reduction in greenhouse gas emissions Provide opportunities to local farmers, entrepreneurs, and rural populations in making use of its sustainable development potential.

Sometimes, advocates of cellulosic technologies (also termed second-generation fuels referring to the conversion of cellulose and hemicellulose into simple sugars that then can be fermented by yeasts to produce ethanol) can become over enthused with their vision of a burgeoning industry. In their fervor, they underestimate the complexity of even the most basic processes that will be required to build the industry. That includes the collecting, transporting, and storing of feedstocks. Producing billions of gallons of ethanol will require millions of tons of cellulosic feedstocks. Moving that feedstock efficiently, safely, and with as little impact on infrastructure and the environment as possible will be a major challenge in the years ahead (Kram, 2008). Research suggests development and deployment of woody biomass resources have distinct energy, economic, and environmental advantages over traditional agricultural crop sources. These advantages are: r Year-round availability and numerous sources of material r Positive net energy ratio r Physical–chemical characteristics of hardwoods consistent even when supplied from

multiple sources

r Effective technical and engineering competencies to manage woody biomass have been

developed

r Sustainably harvested forest biomass can nationally provide at least 368 million dry tons

of wood per year. The value for production and processing requirements for woody biomass has yet to be firmly established, thus alternative pricing has been used to provide a range of possible financial trajectories (Bateman and Lovett, 2000; Bowyer et al., 2001).

P1: SFK/UKS P2: SFK c03 BLBS075-Hood December 16, 2010

54

16:59

Trim: 244mm×172mm

Plant Biomass Conversion

3.2 Natural Regeneration Although the discussion to this point has centered on plantation establishment and production, it is imperative that natural regeneration of hardwoods and pines be introduced as it could be a major component of bioenergy fiber. Today, most hardwood stands have originated from natural regeneration of idle land, or following harvest of pine and hardwood sites. In 1999, it was reported that upland and bottomland hardwoods accounted for approximately 105 million acres in the southern United States (Wear and Greis, 2002). The growing costs of plantation pine are at times 6–9 times lower than plantation hardwoods. In addition, there are also a wider array of markets for pine pulp, chip and saw, and sawtimber over a shorter timeframe. Because of this, hardwoods are typically naturally regenerated. Although naturally regenerated stands are not focused on rapid growth, they will produce a considerable amount of biomass that could be used by the bioenergy market. Under natural regeneration management, techniques such as removal of the understory and mid-story are done by either mechanical or chemical means to either improve the stand or set the stage for regeneration of the desired species. During the regeneration process, removal of the overstory is delayed until there are enough young seedlings of the desired species on a per acre basis. With the removal of the overstory, full sunlight is able to reach the forest floor, allowing numerous intolerant tree species (i.e., those species that cannot grow under shade) to populate the early stages of growth. Although these intolerant species exhibit rapid growth and dominate the stand during the early stage, they are usually not the desired species of the mature stand. It is during this early stage that thinning would be extremely advantageous to those higher-valued species. Removing unwanted intolerant species would provide reduced competition to the higher-valued species and better mid-stage growth. In the past, removal of this material was problematic, as it was generally too small for pulpwood, thus becoming an additional cost to the landowner. In most cases, landowners chose to do nothing and allowed the stand to develop naturally, with the shorter-lived intolerant species eventually dying out of the stand. If smaller woody hardwood biomass proves to be feasible for the bioenergy market, this type of previously overlooked material may become a viable source of biomass. However, the landowner and logger should be reasonably compensated in order to extract this type of woody biomass across the South. Although natural regenerated hardwoods have been the focal point of this brief discussion, it should not be forgotten that there are significant acreage in naturally regenerated pine. As stated previously, pine plantation acreage in the South comprised some 32 million acres while hardwood plantations were estimated at best to be 200,000 acres. In 1950, natural pine stands in the southern United States accounted for nearly 72 million acres, but by 1999, this acreage fell to approximately 30 million acres (Wear and Greis, 2002). However, much of the decline in natural pine and hardwood stands was due to diversions to agriculture and urban land uses. Densely stocked stands of naturally regenerated could also provide a substantial amount of woody biomass.

3.3 Overall Growing Systems It has become apparent that logging residues alone will not be enough to meet the future needs of the bioenergy and biofuels industry of the southern United States. Forest plantations dedicated not only to production, but to the maximization of biomass will play a major role in meeting the future needs for woody biomass. Forest pine plantations have been sustainably grown in the southern United States for nearly 90 years, with the first large-scale plantation

P1: SFK/UKS P2: SFK c03 BLBS075-Hood December 16, 2010

16:59

Trim: 244mm×172mm

3 Growing Systems for Traditional and New Forest-Based Materials

55

being established between 1920 and 1925 on the lower Gulf Coastal Plain. The Civilian Conservation Corps planted over 1.5 million acres in the 1930s (http://www.forestencyclopedia.net). To date, forest pine plantation acreage has grown to an estimated 32 million acres in the Southeast. Unfortunately, hardwood plantations have not matched that of pine plantation acreage, with short-rotation hardwood plantations being estimated to be approximately 200,000 acres. However, since that time, significant hardwood plantations have been established under the Federal Conservation Programs, such as the Conservation Reserve Program (CRP) and the Wetlands Reserve Program (WRP), in the Lower Mississippi Alluvial Valley (LMAV). In 2007, Rousseau estimated that in a four state area (i.e., Arkansas, Louisiana, Mississippi, and Missouri) of the LMAV, there were approximately 711,000 acres in WRP hardwood plantations. However, WRP acreage is somewhat misleading in that only early age thinning would provide material for the biomass market. The total acreage of hardwood plantations solely designed for production is extremely small. The reasons for the lack of hardwood plantations includes the lack of a single species adapted to a wide variety of sites, difficulty in correctly matching the species and the site, lack of markets for intermediate size material, and the perception that cheap and plentiful hardwood fiber was virtually inexhaustible.

3.3.1

The Beginnings of Biomass Plantation Production

While wood has been used as a source of energy for as long as mankind has existed, the use of fossil fuels, such as coal and oil, as a source of energy and fuel began to steeply increase in the United States during the 1920s. Fifty years later following the Arab oil embargo, the United States began to look at alternative energy sources. The US DOE Short Rotation Woody Crops Program, now named the Bioenergy Feedstock Development Program, funded a number of projects focused on understanding and optimizing the growth of a variety of tree species under intensive plantation culture. Later, DOE settled on poplars (Populus ssp.) and willows (Salix ssp.) as their model tree species. At the same time, the pulp and paper industry in the southern United States was also examining similar systems and species as a means of providing a captured fast-growth dedicated source of hardwood fiber for their mills. The primary species being examined in the South were eastern cottonwood (Populus deltoides Bartr.), American sycamore (Platanus occiendtalis L.), and sweetgum (Liquidambar styraciflua L.). These efforts were also combined with that of the USDA Forest Service Southern Hardwood Laboratory, the North Carolina State University Hardwood Research Cooperative, and the Western Gulf Tree Improvement Program. From these efforts, new genetic planting stock was developed and silvicultural techniques refined. With the exception of eastern cottonwood, the rotation length (i.e., time of planting to time of harvest) in most cases exceeded what would be considered today as a suitable timeframe for a bioenergy crop. Certain companies, such as Westvaco and Crown Vantage, continued their research and development efforts of this concept as it proved beneficial to their respective mills (Figure 3.1). While this material was never considered for bioenergy, it did allow the development of systems that are applicable to today’s biomass production systems. If we were constructing the optimal tree for establishment and growth as a biomass crop, the traits desired would include (1) the ease of vegetative propagation so that genetically superior selections of any age could be easily cloned and quickly placed into a large-scale plantings, (2) rapid juvenile growth rates that reached expected harvest in 2–3 years, (3) ease of coppice, thus eliminating replanting costs for numerous rotations, (4) adaptability across a wide variety of sites, and (5) the ability to spray chemicals directly over the top of the trees to control herbaceous and vine competition. This list of traits is by no means all-inclusive, but it does represent those traits that would allow accelerated development of genetic material and substantially lower

P1: SFK/UKS P2: SFK c03 BLBS075-Hood December 16, 2010

56

16:59

Trim: 244mm×172mm

Plant Biomass Conversion

Figure 3.1. Excellent survival and growth of a 4-year-old eastern cottonwood plantation on a Mississippi River alluvial site in Carlisle Co., KY, established by Westvaco Central Region.

growing costs. Unfortunately, there is no single tree species that encompasses this entire list of traits. The goal is to use those species that possess some of the desired traits and then incorporate a variety of new technologies that would increase yields while lowering costs. In addition, financial modeling will be necessary across the various species so as to determine optimal numbers of trees per acre (TPA), rotation length, and harvesting systems (Gonzalez et al., 2009).

3.3.2

Short Rotation Woody Crops

Short-rotation woody crops (SRWC) were defined in the early 1970s as a means of rapidly producing biomass for both fuel and fiber (Brown, 1976; Steinbeck et al., 1972). It was the objective of the US DOE to find a way to not only improve on yields of biomass feedstock but to also develop the technology of converting the biomass to liquid transportation fuel (Ranney et al., 1987). While development of SRWCs was rapidly building and landowners were investing in this system, there were very few end users other than the wood products industry. From these early investigations, it was learned that this type of biomass supply system involved a number of components, including site selection, site preparation, improved clonal genetic stock, extensive competition control, fertilization, pest control, and efficient harvesting systems that were essential in achieving potential yields (Tuskan, 1998). Yield improvement expressed as either green or dry tons per acre per year was the primary driving force behind short-rotation woody crop systems. Species such as eastern cottonwood, American sycamore, green ash (Fraxinus pennslyvanica Marsh.), sweetgum, willows, and hybrid poplars were among those species investigated across the United States. In the South, eastern cottonwood, American sycamore, green ash, and sweetgum were the prime species investigated. A variety of non-native species such as eucalyptus (Eucalyptus spp.) and European black alder (Alnus glutinosa L. Gaertn) were also investigated by a number of agencies and companies. Black alder did not work well in the South because most sources were not well adapted to the southern climate (http://www.ncsu.edu/project/hardwood/research).

P1: SFK/UKS P2: SFK c03 BLBS075-Hood December 16, 2010

16:59

Trim: 244mm×172mm

3 Growing Systems for Traditional and New Forest-Based Materials

57

Productivity results of eucalyptus were rather dismal except when specific species were planted in Florida. Rockwood et al. (2008) indicated that SRWC systems of eucalyptus may be implemented not only in Florida but other southern states. On suitable sites and/or with intensive culture, fast-growing hardwoods such as E. grandis may reach harvestable size in only a few years. Eucalyptus grandis is now grown commercially in southern Florida and can also be grown in central Florida. ArborGen, LLC has been testing and pursuing the possibility of commercializing a genetically engineered cold-tolerant hybrid eucalyptus for use in milder climatic conditions of the lower Gulf Coastal Plain and the Atlantic Coastal Plain. Species/Site Selection In order to obtain optimal growth of the native hardwood species, they must be established on those sites that possess characteristics conducive to each respective species. Site characteristics such as unrestricted soils, good moisture availability, and inherent fertility are paramount to obtaining optimal growth of fast-growth hardwoods. In other words, these species need considerable fertility, year-round moisture availability, and unrestricted root growth to obtain their rapid growth potential and maintain greatest possible growth to harvest. When planted off-site, these species may perform well for a short period of time (i.e., one to five years), but growth typically stagnates. Extensive mortality may follow stagnation. At times, these stressed trees succumb to a variety of either insects or diseases. Since fast-growth species are highly intolerant, effective site preparation and early competition control is an integral component of SRWC culture. Unfortunately, there are very few effective herbicides that can be used with fast-growth hardwood species. Thus, it is necessary to prepare the site to near agricultural row-crop standards followed by sub-soiling for the exact spacing desired, which allows the use of mechanized equipment to control herbaceous, vine, and undesired woody competition. Without early competition control, these fast-growth hardwoods will become stressed and never express their full growth potential. Genetic Improvement The use of genetically improved planting stock has been shown to provide greater survival, increased growth, and disease resistance. In total, these improvements have resulted in increased yields and shorter rotations. Because an autonomous tree improvement program dictates considerable labor and cost, most institutions settled for programs that were built around cooperatives such as the North Carolina State University Hardwood Cooperative and the Western Gulf Tree Improvement Program. Individually, most agencies participated in pooled testing of open-pollinated selections and formation of first-generation seed orchards. Those companies that owned suitable land for eastern cottonwood primarily conducted clonal refinement to obtain greater genetic yields. Eastern cottonwood was the only native short-rotation species operationally planted as clonal material. Eastern cottonwood planting stock is usually deployed as dormant unrooted cuttings, but can also be deployed as dormant or in-leaf rooted cuttings. Other species, such as American sycamore and sweetgum, have been clonally propagated, but this technique has not been adopted operationally. Thus, most programs rely on genetically improved open-pollinated seedlings for these two species. Competition Control When hardwood plantation establishment spacing is reduced, mechanical control of herbaceous weeds and vines becomes increasingly difficult and costly due to the lack of effective

P1: SFK/UKS P2: SFK c03 BLBS075-Hood December 16, 2010

58

16:59

Trim: 244mm×172mm

Plant Biomass Conversion

chemicals. Disking has been shown to be an effective technique for obtaining post-planting competition control (Fisher, 2002). Increased survival and growth of planted hardwood species in response to disking has been documented and is considered necessary when managing some hardwood species. Eastern cottonwood and American sycamore typically require two-way cultivation through the first and second growing seasons (Kennedy Jr and Henderson, 1976; Stanturf et al., 2001). Chemical weed control is a more desirable alternative to efficiently manage competition, but few chemicals are specifically labeled for hardwood forest plantations. Because of hardwood susceptibility to some herbicides, most chemical control research has been performed out of necessity and under specific situations. Diseases Disease problems have limited the use of hybrid polar in short-rotation plantations throughout the Southeast. A wide variety of hybrid poplar taxons demonstrated susceptibility to Septoria leaf spot and stem canker (Septoria musiva Peck.) on sites in Kentucky (Rousseau et al., 2008). This particular disease, while present in the LMAV, causes little problem to native eastern cottonwood. However, there are areas within the Southeast where the disease does not currently exist. In these areas, hybrid poplars have survived and grown well. The reason for the lack of Septoria occurrence in these areas is not understood, but this may break down fairly quickly if large acreages are dedicated to hybrid poplar plantings. While Septoria leaf spot results in loss of photosynthetic area and reduction of growth, Septoria stem cankers typically result in mortality. Although eastern cottonwood has been used in various crosses to produce hybrid poplars, the resulting hybrids for the most part have shown susceptibility to Septoria stem canker (Rousseau et al. 2008). One particular hybrid poplar clone, NM-6 (which is a cross between Populus nigra L. and P. maximowiczii A. Henry), is known for its resistance to Septoria stem canker in the Midwest. Clone NM-6 showed early susceptibility to Septoria leaf spot but no stem cankers were visible through age 5. However, Septoria stem canker was present at age 10 and trees harvested at this age showed very little sound wood. The rapid growth rate, thin bark, moderate wood density level, and wider site adaptability makes American sycamore a prime candidate for biomass production in short-rotation plantations. Yet as American sycamore plantings increased, significant disease problems arose and nearly eliminated this species as a plantation species. Disease problems were initially thought to be the result of off-site planting. However, in the 1970s and 1980s, reports of mortality and dieback as a result of diseases began to be reported in sycamore plantations throughout the southeastern United States. Many of these plantations were located on optimal sites with adaptive seed source material. Several diseases, including bacterial leaf scorch of sycamore, caused by the bacterium Xylella fastidiosa, canker stain fungus Ceratocystis fimbriata f. sp. platani (Ellis and Halst.), Botryosphaeria canker (Botryosphaeria rhodina (Cooke) Arx) were noted in hardwood plantations at ages as young as 3 years. It was also noted that Xylella fastidiosa infections cause leaf scorching, branch dieback, and eventually can kill trees (Leininger et al., 1999). Clonal propagation of resistant trees would provide the ability to build a resistant population of American sycamore for the Southeast, but little work has been completed on vegetative propagation. In each case, disease problems can be overcome but would entail concerted efforts in the fields of genetics, physiology, and pathology. Hybrid poplar clones for the southeastern United States are also somewhat limited as all of the clones were developed for other geographic areas. While new hybrid poplar clones could be developed for the South this would entail extensive testing for site adaptability, growth, and disease resistance.

P1: SFK/UKS P2: SFK c03 BLBS075-Hood December 16, 2010

16:59

Trim: 244mm×172mm

3 Growing Systems for Traditional and New Forest-Based Materials

59

Insect Control Coyle and Coleman (2005) stated that often fast-growth hardwood species by their nature are nutrient-rich and may provide insects with more optimal diets. They noted that in particular Populus and Salix species are more suited for and susceptible to various pests from establishment to age 3. The reasons that short-rotation tree plantations are susceptible to insects are low biodiversity and spatial uniformity. Coyle and Coleman (2005) also provided a number of guidelines that should be considered. First, pest-resistant stock should be planted whenever possible. Second, monocultures should be avoided by planting polycultures or mixed species plantings, thereby minimizing the evolutionary capacity of the pests to overcome the plant’s defense mechanisms. Third, it is advisable to create a mosaic of small plantations. These small plantations support smaller populations of insects and allow a higher probability of being preyed upon by natural enemies. Among the species considered for the Southeast, eastern cottonwood is the most susceptible to pests. Sweetgum is the most widely adapted of the three species being compared (eastern cottonwood, sycamore, and sweetgum) and seems to be the best in terms of disease and insect resistance. The growth of sweetgum is certainly not comparable to that of the other two species when planted on a highly fertile site, but it will outperform the others on a number of sites.

3.3.3

Other Types of Hardwood Plantations

Johnson et al. (2007b) used the term woody perennial energy crops (WPEC) to describe fast-growing trees under intensive management. This term (WPEC) was used interchangeably with other terms such as short rotation woody crops (SRWC) and bioenergy tree plantations (BEP). In each case, the rapid growth rates achieved by these intensely grown, rapid growthrate trees will allow harvesting in a matter of years (i.e., between 2 and 8 years) rather than decades, considered as the norm for traditional tree plantations. This shortened timeframe falls between agricultural systems and traditional plantation forestry, and because of this, there have been some negative comments from both the agricultural sector and environmental groups. Some in the agricultural sector look at these systems as a threat to the current agricultural footprint, but these plantations have always been designed to be grown primarily on marginal agricultural land. The recent debate between energy and food was brought to the forefront when corn production dedicated to ethanol was greatly increased. Even if second-generation lignocellulosic fuels become the standard, the debate will still continue if these dedicated plantations are established on agricultural land. Some environmentalists argue that if wood is adopted as a viable bioenergy source, the results would significantly increase harvest of natural stands, thus leading to declining biodiversity. This argument is totally unfounded as dedicated bioenergy plantations, although less biologically diverse than natural stands, will be much more economically feasible in procuring the woody biomass needed for the bioenergy and biofuels market than natural stands. In this situation, there would be much less pressure on natural forests, allowing young native stands to mature. The maturation of these native stands would provide a wider base of ecosystem services, more traditional products, and increased biodiversity. It is possible to classify the various types of forest plantations by type, species, and size. Johnson et al. 2007a classified WPEC into three categories: (1) large-stature widely spaced hardwoods, (2) low-stature coppice culture, and (3) intensively managed pine. The classification of large-stature widely spaced hardwoods is difficult to consider as a bioenergy crop primarily due to the high cost of site preparation and establishment and the current low

P1: SFK/UKS P2: SFK c03 BLBS075-Hood December 16, 2010

60

16:59

Trim: 244mm×172mm

Plant Biomass Conversion

returns for biomass. The reality is that such a system is not much different than intensive pine culture with the exception of much higher growing costs. Low-stature coppice culture is again a hardwood plantation system that has been used successfully in shrub willows in the northeastern United States. Intensive pine plantation management has been successfully used in the Southeast for over 80 years. Improvements in genetics, planting stock quality, and silvicultural techniques resulted in faster growth rates and higher survival. Thinnings must be accomplished at an earlier age and the tree size is usually at best only pulpwood size. It is this material or even smaller that could be used for bioenergy or biofuels. The three most probable native hardwood species for dedicated bioenergy plantations include eastern cottonwood, American sycamore, and sweetgum. The reasons for this are shown in Table 3.1. Table 3.1. Probable hardwood bioenergy species and the associated advantages and disadvantages of each species. Species

Advantages

Negatives

Eastern cottonwood

1. 2. 3. 4. 5. 6.

Ease of vegetative propagation Coppices well Rapid growth Extensively studied Improved clonal material available Populus genome completed

1. Lack of chemicals for weed control 2. Narrow adaptability 3. Pest susceptibility

American Sycamore

1. 2. 3. 4. 5.

Rapid growth Coppices well Vegetative propagation Thin bark Medium wood density

1. Susceptible to disease complex 2. Lack of chemicals for weed control

Sweetgum

1. 2. 3. 4. 5. 6.

Coppice well Can be cloned Advanced genetics Broad adaptability Medium wood density Few diseases

1. Medium growth rates

Populus is the one genus of tree species that continues to intrigue most people in the field of bioenergy. Rapid growth rates and ease of vegetative propagation are the characteristics that continue to draw attention. While DOE recognizes the importance of poplars, this recognition has not been limited to the United States, as Populus programs are found throughout the world, though primarily in the northern hemisphere. A great majority of these programs are based on hybridization where F1 individuals are generated through interspecific crosses among the more than 29 species within the genus Populus (Stettler et al., 1996). In the southern United States, very little hybrid poplar is grown due to its susceptibility to the disease Septoria musiva. Many hybrid poplars are susceptible to this disease resulting in mortality. However, there are places in the southern United States where this disease does not occur and hybrid poplars could thus thrive in these areas. Eastern cottonwood gained popularity in the South, especially in the LMAV during the 1960s and 1970s as result of the work accomplished by the USDA Forest Service Southern Hardwoods Laboratory at Stoneville, MS (Schmidtling et al., 2004). Numerous companies installed cottonwood plantations using genetically superior cottonwood clonal stock. These

P1: SFK/UKS P2: SFK c03 BLBS075-Hood December 16, 2010

16:59

Trim: 244mm×172mm

3 Growing Systems for Traditional and New Forest-Based Materials

61

plantations, while in most cases designed specifically for pulpwood production, unknowingly laid the groundwork for current bioenergy plantations. From these beginnings, our understanding of cottonwood grown under plantation culture grew as work in silviculture, physiology, pathology, entomology, and biometrics increased. Eastern cottonwood plantations in the South are typically regenerated by dormant un-rooted cuttings. These cuttings are grown in stool beds or cutting orchards planted at a spacing of 1 × 3 feet. This type of system is closely related to bioenergy plantations. Control of herbaceous competition is the major problem with this system during the first year. Yet, if competition is adequately controlled, the cuttings will attain heights between 8 and 19 feet in the first year. Competition control in subsequent years is rather limited and as the stump sprouts are growing so rapidly, they quickly outgrow and shade out the herbaceous competition. Although most eastern cottonwood plantations were established at a spacing of 12 × 12 feet, coppice regeneration was used whenever possible. However, even though coppice regeneration works well for widely spaced cottonwood plantations, only limited acreage was coppiced because harvesting usually occurred during the summer months. Typically, height growth of cottonwood is approximately 10 feet per year, and its diameter growth is over an inch per year. This type of growth will only be obtained by selection of a fertile site, thorough competition control, and excellent genetics. Cottonwood in a bioenergy system will more than likely comprise 800 plus TPA, with rotation lengths between 1 and 5 years. There is no doubt that coppice regeneration will have to play a key role in this system as to provide lower establishment and growing costs. However, there will always be a need to re-establish plantations because coppice regeneration can only be done a finite number of times. It is also hoped that new clones will replace older clones as plantations must be re-established. Hardwood species such as American sycamore, sweetgum, red maple (Acer rubrum L.), silver maple (Acer saccharinum L.), and black locust (Robinia pseudoacacia L.) have in general been deployed in plantations as either bare-root or containerized planting stock. The use of seedlings is somewhat problematic if seedlings are necessary for replanting at the end of every rotation. With the exception of eastern cottonwood and to a lesser extent sweetgum, the genetic programs of hardwoods are rather limited. Thus, to extract genetic gains from these species, newly designed traditional improvement programs must be initiated. The one positive aspect is that these species all coppice rather easily, and as long as suitable genetic material is established initially, a number of rotations may be completed before entire replanting is necessary.

3.3.4

Southern Pine

Forest plantations in the southern United States are, in general, planted to a single tree species and can vary from hardwoods to conifers. However, the typical forest plantation in the southern United States is planted to loblolly pine (Pinus taeda L.) at a rate of approximately 400–700 seedlings per acre. To date, there are approximately 32 million acres of plantation pine located in the southern United States. Some of this acreage is greatly overstocked and early removal of trees through chopping or mowing has been practiced to lower the number of stems per acre to allow for better growth. In a few cases, this material has been harvested and used in generation of electricity through simple burning. It has also been demonstrated that young 10-year-old loblolly pine could be used in the production of various bio-oils (Hassan et al., 2009; Mitchell, 1998). Typically, pine plantations are grown from 25 to 30 years depending on site conditions, and thinned once or twice to remove pulpwood or trees that will not yield high-quality sawtimber (Gonzalez et al., 2009). Loblolly pine has proven to be widely

P1: SFK/UKS P2: SFK c03 BLBS075-Hood December 16, 2010

62

16:59

Trim: 244mm×172mm

Plant Biomass Conversion

adapted across the southern United States, easily outperforming hardwoods on marginal upland sites. Added to this is the wealth of knowledge in pine plantation systems including nutrient needs, competition control, and genetics, easily defining loblolly pine as a prime candidate for biomass production in the Southeast (McKeand et al., 2006; Schmidtling et al., 2004). While the plantation concept has been a great success for pine and can provide material for bioenergy and biofuels, a change is needed to further the progress toward dedicated bioenergy plantations. Today, there is a definite interest in characterizing the growth and production of pine genotypes to understand if it is possible to select for more intensive use in biomass systems. ArborGen has developed a hybrid type of system that they have labeled as FlexStandsTM . In this system, they have combined forest biomass production with that of a higher revenue stream resulting from sawtimber. The keys to this system are that it takes advantage of the fact that loblolly pine is widely adapted across the entire southeastern United States, the well defined and less costly silvicultural techniques, and the sophisticated genetic program. MeadWestvaco and ArborGen’s development of mass-control pollinated (MCP) and varietal genetic material are the sawtimber focus of this system. The FlexStand approach employs 535 seedlings per acre with MCP seedlings or varietal stock (i.e., clonal seedlings) planted in every other row, while less expensive open-pollinated seedlings or designated biomass trees are placed in the alternate rows. The major difference is in the within-row spacing, where the biomass trees are planted every 3–5 feet. The sawtimber designated trees are planted at a within-row spacing of 10 feet. Depending on inherent site quality, the biomass trees will be harvested when they near the carrying capacity of the site. While this approach employs forest species as both the biomass component as well as the traditional sawtimber component, it lacks the ability to provide a continual biomass component throughout the rotation. At Mississippi State University, a true hybrid system combining loblolly pine and switchgrass is the primary focus of what has been termed “Co-Culture” (See section on Agroforestry).

3.4 New Genetic Tools Increases gained from clonal forestry have been shown to be dramatic in the US Pacific Northwest, with hybrid poplars demonstrating gains of greater than 100% over the average seedling (Stettler et al., 1988). The ability to incorporate molecular genetics techniques will provide a better understanding of the processes of wood formation, diseases, drought resistance, and growth. With this knowledge, breeding and selection can be tailored toward specific techniques that would make the biomass product more efficient and productive in yields of energy and fuels. Genetic engineering or genetic transformation has been accomplished in forestry and has demonstrated great potential. This potential would be even greater for hardwoods where chemicals used for site preparation and establishment are very limited and at times ineffective (Schmidtling et al., 2004). Herbicide resistance in hardwoods would solve the primary obstacle that stands in the way of increasing the acreage of hardwood plantations and subsequently bioenergy plantations. However, while the use of genetically engineered trees holds great promise, it is unlikely that trees of this type will be commercially deployed on a large scale in the United States until effective evaluation of their impact has been conducted. A number of organizations have taken the initiative to assist in development of purposegrown forest bioenergy plantations. This is to support the dozens of new and existing users of woody biomass for renewable energy. ArborGen, LLC is developing forest tree species for traditional uses such as pulp, paper, and lumber. They are also actively engaged in developing

P1: SFK/UKS P2: SFK c03 BLBS075-Hood December 16, 2010

16:59

Trim: 244mm×172mm

3 Growing Systems for Traditional and New Forest-Based Materials

63

forest tree species for bioenergy as well as developing appropriate silviculture to support these bioenergy forest plantations. Several genera including pine, sweetgum, poplars, and eucalyptus are important in ArborGen’s southern United States efforts. The traditional tree improvement programs have led to increased yields, and when combined with appropriate silviculture of higher number of TPA and shorter rotations, large benefits for growers and energy processors will be attained. The use of biotechnology to enhance forest plantation yields and energy value of the wood will ultimately provide a cost-effective bioenergy forest plantation system. Estimated improvements from biotechnology could lead to an increase of 40%–100% in BTUs/acre/year production over current systems. Biotechnology improvements will be achieved by utilizing selected varietal genotypes from traditional improvement programs, insertion of genes, field testing to verify gene effects, regulatory compliance, and eventual commercial sales. The use of biotechnology is especially important for traits that would take considerable time using conventional genetic improvement programs. A major step in the development of genetic resources was to decode the genetic blueprint of Populus trichocarpa by sequencing its genome (Tuskan et al., 2006). This provides the unique opportunity to understand the various genetic controls over various traits that may be critically important to the production of forest plants. This understanding will allow fundamental research questions to be translated into applications in genetic improvement for growth, stress tolerance, process compatibility, or specialized chemical components needed for biorefineries (Davis, 2008). This tool is becoming more commonplace as the cost of gene sequencing continues to drop. One of the primary objectives of the sequencing work is to more efficiently and effectively identify individual genotypes in a population that possess the desirable genes of interest for the production of optimal levels of bioenergy and biofuels. This objective is currently being addressed by a number of institutions in the United States (e.g., DOE, University of California Davis) by identifying single nucleotide polymorphisms (SNPs) via gene sequencing and phenotyping in populations of poplars and loblolly pine. The resulting SNP markers can be a valuable tool for efficient and precise breeding and selection. The use of transgenic manipulation has been done in poplars and provides the primary technology in proving gene function. When combined with naturally occurring genetic variants, it is now possible to clarify the underlying causes of observed variation (Davis, 2008). These new tools have provided the geneticist with a new opportunity to accelerate, improve, and develop novel individuals for woody trees tailored to the production of bioenergy and biofuels. Combining effective breeding with the knowledge of the genome to guide selection will prove to be a powerful approach. This might be visualized through adaptive breeding, but it will be imperative to identify the correct species (parents) for hybrid combinations. Reducing the cost of testing while increasing the effectiveness of genetic testing through the use of marker-aided selection may allow us to grow poplars on what has been previously considered marginal sites.

3.5 Agroforestry Agriculture and forestry are the primary producers of bioenergy feedstocks. Agronomic crops and forests convert solar energy into chemical energy by photosynthesis. Agriculture and forestry have a unique opportunity to both provide biomass for energy and to serve as a net sink for CO2 . The current hypothesis is that if managed correctly, bioenergy can provide a C-neutral or even a C-negative feedstock. Thus, in the future we will be increasing our demands on both the agriculture and forestry sectors to provide bioenergy (Johnson et al., 2007a).

P1: SFK/UKS P2: SFK c03 BLBS075-Hood December 16, 2010

64

16:59

Trim: 244mm×172mm

Plant Biomass Conversion

Figure 3.2. Agroforestry system employed by Westvaco Central Region located in Wickliffe, KY, along the Mississippi River. Picture shows lack of sunlight resulting in the inability to grow row crops. Chufa was planted for wildlife purposes.

Agroforestry purposely integrates agriculture and forestry in a manner that provides benefits from both systems (Schoeneberger, 2009). In most cases, the objectives are to evaluate the suitability of different tree species for marginal sites in terms of their yield potential and the influence of the trees on the sustainability and yields of annual biomass crop supply. Numerous trials have been established worldwide and some have shown interesting results. However, the common thread has been the need to identify tree species that perform well on the site chosen and evaluation of the synergy from combining agriculture and forest systems (Gruenwald et al., 2007). The Central Region of Westvaco Timberlands employed an agroforestry system on bottomland sites along the Mississippi and Ohio Rivers in the 1980s. The design consisted of two rows of either eastern cottonwood or American sycamore planted at a spacing of 8 × 12 feet. The rows of planted trees were separated by 40 feet, allowing an agricultural crop to be grown. This design was intended to allow agricultural production for a couple of years, thus providing income to help defray the cost of plantation establishment. The double row of trees provided the TPA necessary for pulpwood yields and competition control in a single direction. This system worked well for the first few years, but because of the rapid growth of the trees, agricultural production was minimized by a lack of sunlight. The area initially occupied by row crops provided additional growing space, allowing the trees, especially eastern cottonwood, to reach sawtimber size. This space also allowed access by mechanized equipment, thus easily employing a select harvest of higher value individuals (Figure 3.2). Agroforestry systems have the ability to sequester large amounts of carbon while providing additional benefits such as economic diversification, biodiversity, and water quality (Montagnini and Nair, 2004; Schoeneberger and Ruark, 2003; Schroeder, 1994). This type of

P1: SFK/UKS P2: SFK c03 BLBS075-Hood December 16, 2010

16:59

Trim: 244mm×172mm

3 Growing Systems for Traditional and New Forest-Based Materials

65

hybrid system has considerable appeal as a greenhouse gas (GHG) mitigation activity. The tree component provides substantial amounts of carbon sequestered, but the bulk of the land remains in agricultural production (Ruark et al., 2003). While agroforestry on the surface seems to be a win–win situation, it does represent a change, and it is perceived by many in the agricultural community as a cost. It is evident that landowners employing such as system must realize that it is a long-term commitment as it will take time for the tree component to reach the point of functional maturity. One aspect of forestry and agroforestry is the role they will play in the reduction of GHGs. Trees sequester a tremendous amount of carbon dioxide and have been used as an offset component of the US voluntary carbon market. The key components of any carbon offset are additionality and permanence. This at times has been questioned by those who are opposed to using forests as a means of GHG reductions. But the conversion of agricultural land to forest plantations, albeit limited, in an agroforestry system does meet the requirements of additionality. Moreover, converting these plantings back to agriculture is both costly and difficult, thereby providing a degree of permanence. Loblolly pine has a history of being grown in combination with warm-season grass species. Much of this has involved silvipastoral systems combining sawtimber production with either grazing or haying operations. Clason (1998) examined four forage species established in a 20-year-old loblolly pine plantation. Under common forage management practices, including fertilization, he noted no effect of forage species on tree growth over 11 years. Burner and Brauer (2003) planted loblolly pine seedlings into established pastures and examined the effects of different row spacings and row configurations on herbage yields. They found that loblolly pine negatively affected herbage yields up to a pine spacing of 4.9 m. Wolters (1973) also found that southern pine overstories increasingly reduced herbage production as levels of pine stocking increased, as was seen by Halls and Schuster (Halls and Schuster, 1965). Loblolly pine can also be managed as a source of biomass for bioenergy production (Scott and Tiarks, 2008). However, while its biomass production potential is high, it is probably not as high as for some other tree species such as species of poplar or willow. Loblolly pine has a substantially higher value for sawtimber production in the Southeast. A slight twist on typical agroforestry systems incorporates annual bioenergy crops such as switchgrass (Panicum virgatum L.) or giant miscanthus (Miscanthus × giganteus) in place of annual row crops, while the tree plantings favor longer rotations in the production of highly valued sawtimber species and individuals. At Mississippi State University, this type of agroforestry system has been termed “Co-Culture” and incorporates switchgrass with loblolly pine varietal (i.e., clonal) planting stock. In this system, a highly selected varietal loblolly pine genotype is combined with that of switchgrass. The number of TPA is reduced to 300, and switchgrass is established prior to tree seedling establishment. The key to this system is to continue to produce an annual biomass crop of switchgrass while allowing the trees to develop their inherent characteristics, thus providing an additional source of revenue some 25–30 years following establishment. Thinning would be possible at those times when the annual biomass crop needed re-establishment, which would allow ease of access to the trees, resulting in little or no damage to the remaining stand. If the carbon market is a viable alternative in the future, the trees may provide an additional revenue stream in carbon credits. The first trial studied at Mississippi State University included two loblolly pine varietals that expressed divergent crown ideotypes (crop vs. competitor) and a typical second-generation family as well as an area of switchgrass that is 56 feet wide and 360 feet long. The two loblolly pine varieties were selected because of their crown architecture. The crown of a crop ideotype is one that has more angular branches and carries less leaf biomass. In contrast, the

P1: SFK/UKS P2: SFK c03 BLBS075-Hood December 16, 2010

66

16:59

Trim: 244mm×172mm

Plant Biomass Conversion

crown of a competitor ideotype would be wide and carry considerably more leaf biomass, thus shading out any type of competition. A selected second-generation family was added to use a baseline in determining the number of sawtimber potential trees in comparison to that of the selected varietal types. Trees were planted at a spacing of 12 × 14 feet in either four-row or two-row configurations. The objectives were to determine the effect that shading of specific tree genotypes would have on the yield of switchgrass through time and if varietal loblolly pine genotypes would exhibit the inherent traits of quality sawtimber in a more open-growth environment. More recently, a Co-Culture Study at Mississippi State University has been designed and will be implemented in the spring of 2010, where a single loblolly pine variety (i.e., clone), selected for sawtimber quality traits, will be grown in conjunction with an herbaceous biomass crop (in this case, switchgrass) (Roberts et al., 2009). The key initial component to the study is that a uniform stand of switchgrass must be in place prior to the tree component being installed. To do this, the switchgrass component will be established during the spring of the year followed by a fall planting of containerized varietal stock. To examine the competitive effect of the switchgrass on the tree performance and vice versa, specific free-to-grow zones will be applied to the trees. In this fashion, it is hoped that a clear understanding of the primary components limiting the performance of either tree growth or biomass yields of switchgrass will be gained. The short-term objectives of this study are to r Quantify competition between pine and switchgrass for soil moisture and nutrients r Quantify the effects of competition between pine and switchgrass on early growth of both

species The long-term objective is to gain a better understanding of competitive interactions between pine and switchgrass when grown in co-culture and the impacts of these competitive interactions on growth and yield of both species. Louisiana State University AgCenter has also been investigating the use of a bioenergy crop with loblolly pine at a site located at the Hill Farm Research Station in north central Louisiana. Their approach was to sow switchgrass between rows of loblolly pine in a plantation setting that was ages 2 (juvenile age), 11 (mid-rotation age), and 21 years (late-rotation age). Switchgrass was planted under relatively low and high densities of loblolly pine in each pine age group to test the effects of tree overstory on switchgrass establishment and growth. For the juvenile pine, the low and high densities of pine were created by planting trees at 225 and 450 TPA, respectively. The space between rows of trees was 16 feet for both densities, and the space between trees within rows was 12 feet for the 225 TPA density and 6 feet for the 450 TPA density. For the mid- and late-rotation pine, the low and high densities were created by a thinning operation. The mid-rotation stand was thinned to 100 and 200 TPA for the low and high densities, respectively. The low and high densities for the late-rotation pine were 25 and 50 TPA, respectively. To prepare the stands for switchgrass planting, imazapyr herbicide was applied by tractor-mounted sprayer at 48 oz per acre one year prior to planting and a 5% solution of glyphosate herbicide was applied by tractor-mounted sprayer two weeks prior to planting. Immediately prior to planting the soil was disked, and then the Alamo cultivar of switchgrass (which has shown the best growth potential among switchgrass varieties in much of the southeast United States) was broadcast seeded in April 2008 with a tractor-mounted cyclone seeder at 8 lb of pure live seed per acre. Seed was then culti-packed immediately after broadcast. To control weeds within the switchgrass, it was necessary to rotary mow the site with a tractor-drawn mower set at a 9-inch height in June 2008, September 2008, April 2009,

P1: SFK/UKS P2: SFK c03 BLBS075-Hood December 16, 2010

16:59

Trim: 244mm×172mm

3 Growing Systems for Traditional and New Forest-Based Materials

67

and June 2009. In July 2009, 0.25 oz per acre of metsulfuron methyl herbicide was applied using a tractor-mounted sprayer to further suppress the herbaceous vegetation competition. The LSU AgCenter study has revealed that switchgrass could be successfully established in rows between loblolly pine trees at juvenile, mid-rotation, and late-rotation ages at their site. By the second year of the study, switchgrass biomass covered at least 80% of ground within the alleys in all stand ages and stand densities. Switchgrass biomass by the second year was comparable across all ages and pine densities, which suggests that there may be substantial management flexibility for establishing switchgrass within alleys of loblolly pine. Shading from the pine overstory apparently enhanced switchgrass establishment success in the first year of this study because in all stand ages, switchgrass ground coverage was greater with the presence of loblolly pine than that of switchgrass grown at the same site without loblolly pine. Further evidence of the beneficial effects of shading on first-year switchgrass coverage was seen in the switchgrass coverage among the three stand ages being in the order of late-rotation > mid-rotation > juvenile. At the Hill Farm Research Station, an older similar study has shown that yields of fully mature switchgrass (greater than age 3) when grown under loblolly pine without fertilization averaged 2.5 tons per acre per year, with a maximum of 3.5 tons per acre per year (Michael Blazier, personal communication). Catchlight Energy, LLC is a joint venture between Chevron and Weyerhaeuser, whose objective is to develop the next generation of renewable transportation fuels from non-food sources. This unique venture will capitalize on the biological expertise of Weyerhaeuser and the advanced fuels expertise of Chevron. Currently, the Catchlight group is exploring the possibility of combining a bioenergy crop that can be harvested annually with trees that are managed for wood products and fiber. Their concept is very similar to that of many agroforestry systems where alternate rows of trees and an annual biomass crop are grown and used to feed a second-generation biofuels plant (http://www.catchlightenergy.com).

3.6 Products from Woody Biomass The US DOE and US Department of Agriculture estimated potential lignocellulosic biomass production for fuels in the United States at more than 1.3 billion tons annually, sufficient to replace more than 30% of current transportation fuel use (Perlack et al., 2005). Although age-old processes are available for converting the starch content of grain into sugars, which can then be fermented to ethanol, the conversion of lignocellulose to sugars is much more difficult. The recalcitrance of lignocellulose lies in its structure, which has evolved to provide long duration resistance to pests and pathogens. Thus, the development of processes for converting lignocellulosic biomass to fuels is hampered by the lack of energy-efficient and cost-effective processes for the deconstruction and conversion of this feedstock (Lynd et al., 1999). Trees and herbaceous material are made of three basic components plus many trace materials. The most abundant components are structural natural organic polymers: cellulose, hemicelluloses, and lignin. Cellulose is the long strong and flexible natural linear polymer in fibers that predominately gives trees and woods their strength. Cellulose fibers are held together in part by hydrogen bonds with itself but also by lignin, natural phenolic glue. Hemicelluloses are heteropolysaccharides of lower degree of polymerization than cellulose that are found encrusting the cellulose and lignin and also provide some bonding in the natural structure. Structural wood is simply the tree with all these components cut and shaped to the desired form. White paper is made predominantly from the cellulose fibers with most of the hemicellulose and almost all of the lignin removed.

P1: SFK/UKS P2: SFK c03 BLBS075-Hood December 16, 2010

16:59

Trim: 244mm×172mm

Plant Biomass Conversion

68

CH2OH

HO

OH

4 O

O HO

HO

O

O OH

1

1

4

CH2OH

CH2OH O

OH OH OH

n

Figure 3.3. Sterochemical structure of cellulose.

Cellulose consists of long chains of the six-carbon sugar glucose that are connected end to end as a polymer (Figure 3.3). It would seem simple to break that connection, by mechanical means or with enzymes, so that a long fiber of cellulose would become thousands of simple glucose molecules. However in practice, it is very difficult. The glucose molecules are connected by the more difficult to break β-O-4 linkages (Figure 3.3) and are arranged in regions with high crystallinity, which limits accessibility. Thus, cellulose is much more difficult to break up than starch molecules with their α-O-4 linkages that connect the glucose. Natural selection has resulted in cellulose that does not break apart easily. Fortunately, cellulose fibers are more valuable than their derived sugars because paper products can be made from them at present. However, as the demand for fuel and energy is increasing more than the demand for paper, conversion of cellulose to platform sugars will become more important as fossil energy sources diminish. Relatively pure cellulose can be hydrolyzed to glucose in formic acid media (Sun et al., 2007) and many other methods have been used in the past and are being developed now. It has been demonstrated that a woody biomass hot-water hemicellulose hydrolyzate stream can be separated into products and the sugars used for fermentation to ethanol. This sequential deconstruction of woody biomass is expected to become the dominant commercial method for production of new forest-based materials as it preserves the original value of components and allows their recovery for society’s use. The general process flow is to remove the easiest major component first, the hemicellulose, followed by application of a separate separation technique to fractionate the lignin and cellulose. Past techniques have focused on removing the lignin from cellulose, as in Kraft pulping and pulp bleaching, to provide fiber for papermaking. This approach may not be preferable to removing the cellulose from the lignin when the goal is to produce monomeric or oligomeric sugars from the cellulose. Current research is taking both approaches to continued deconstruction after hemicellulose removal without a clear winner yet clear. The method of deconstruction has significant implications for development of growing systems for traditional products integrated with new forest-based materials. The relative value of some components in traditional products may be trivial or even negative while the same component might be as valuable as cellulose, or significantly higher, as a new forest-based material. One example of this is the Ac, which stands for acetyl group, shown in the hemicelluloses galactoglucomannan and glucuronoxylan in Table 3.1. The acetyl group when recovered in an aqueous medium as acetic acid has values that can range from equal to that for relatively pure cellulose pulp to three times as valuable per unit of mass. This is important as the most dominant method of making relatively pure cellulose pulp (white pulp) is an alkaline system (using NaOH), and the acetyl groups use up alkali by neutralization while producing sodium acetate that is burned to recover the sodium. The fuel value of sodium acetate is very small, a few pennies per pound while the acetic acid could have recovered from $0.25 to $0.65 per pound.

P1: SFK/UKS P2: SFK c03 BLBS075-Hood December 16, 2010

16:59

Trim: 244mm×172mm

3 Growing Systems for Traditional and New Forest-Based Materials

Table 3.2.

69

Major components of hemicelluloses and extractives. Distribution (wt%) Type

Hemicelluloses Galactoglucomannan (1:1:3)

G−M−M−M−G−M−M−M → | | | | Ga Ac Ac Ga

(Galacto) glucomannan (0.1:1:4) Glucomannan (1:2–1:1) Arabinoglucuronoxylan

Glucuronoxylan Extractives Aliphatic and alicyclic Phenolics Carbohydrates Inorganics Others

G−M−M−G−M →
 X −X −X−X−[X]5 → | | Gu 2 A
 X− X− X −[X] → | | Ac 7 Gu

Softwoods

Hardwoods

25–30

25–35

5–8

0

10–15

0

0

2–5

7–10

Trace

Trace

15–30

5–8 2–4 Terpenes: terpenoids, esters, fatty acids, alcohols Phenols: stilbenes, lignans, isoflavones Arabinose, galactose, glucose, xylose, raffinose, starch, pectic material Ca, K, Mg, Na, Fe, SO4 2− , Cl− Cyclitols, tropolones, amino acids, protein, alkaloids

Ash

0.2–0.5

0.2–0.8

G, glucose unit; Ga, galactose unit; M, mannose unit; X, xylose unit; A, arabinose unit; Ac, acetyl group (H3 C-CO-); and Gu, 4-O-methyl-glucuronic acid unit; all the linkages shown in the table are through –O–.

Discussion of the wood components will begin with hemicelluloses as they represent a significantly underutilized component of wood, which are most frequently lost during current chemical conversion processes used to purify cellulose, most commonly the Kraft pulping process followed by a bleaching sequence. These hemicelluloses will likely become targeted for enhancement in future growing systems, as their sugar complement will eventually be equal in value to glucose and their easier disassembly and high value substituents recommend increasing their content in the raw material.

3.6.1

Hemicellulosic Products

Hemicelluloses are heteropolymers of 5- and 6-carbon sugars with side chains. Table 3.2 shows major hemicellulose constituents together with wood biomass extractives. If one considers cellulose as a one-dimensional polymer, most hemicelluloses are two-dimensional polymers. Hemicellulose oligomers can be partially extracted from hardwood chips through solution in water at high temperatures (Amidon et al., 2008; Liu and Amidon, 2007; Liu et al., 2006), and xylose or hemicellulose sugars can be fermented to ethanol by microorganisms (Jeffries and Jin, 2004; Liu and Amidon, 2007; Qureshi et al., 2006). Hemicellulosic sugars may also be used as building blocks for biodegradable plastics (Keenan et al., 2004) or other products and chemicals that are currently made from petroleum (Liu et al., 2006). The residual woodchips can be processed into pulp to make paper, burned for renewable energy, or converted to

P1: SFK/UKS P2: SFK c03 BLBS075-Hood December 16, 2010

70

16:59

Trim: 244mm×172mm

Plant Biomass Conversion

γ

CH2OH

CH2OH

CH2OH

β

α 1

2

6

3

5 4

OCH3

OH a. trans-Coniferyl alcohol

OCH3

H3CO OH

b. trans-Sinapyl alcohol

OH c. trans-p-Coumaryl alcohol

Figure 3.4. The precursors of lignin (Al´en, 2000).

reconstituted wood products such as fiberboard or pellets (Amidon, 2006; Amidon and Liu, 2009). This process would add additional products to the current paper and wood energy business starting an evolution towards wood-based biorefineries (Myerly et al., 1981). Making pulp and/or paper products from cellulose remains attractive today due to cellulose having a higher value as pulp than as sugar. However, the pulp and paper industry in the United States and Canada has actually been declining steadily in recent years as production shifts to higher growth areas such as China and Brazil. Further conversion of cellulose into sugars and/or other chemicals/liquid fuel is highly desirable because of higher future demand. A purpose-built biorefinery will include all these process alternatives, though adaptations of current processing locations may not include all alternatives. Lignin is a heteropolymer composed primarily of methoxylated phenylpropylene alcohol monomeric units interconnected by a variety of stable carbon–carbon and carbon–oxygen–carbon (ethers and esters) linkages (Dence and Lin, 1992). Structurally, lignin is a three-dimensional macromolecule. While the lignin of gymnosperms (also called softwoods, conifers, needle trees, or evergreens) is primarily an enzyme-initiated dehydrogenative polymerization product of coniferyl alcohol (Fengel and Wegener, 1984), the lignin of angiosperms (also hardwoods, deciduous or broad-leaved trees, grasses, etc.) is derived primarily from a mixture of coniferyl and sinapyl alcohols. Figure 3.4 shows the three cinnamyl alcohols, lignin precursors. The oxygen to carbon ratio in the lignin precursors (Figure 3.4) is less than 4/11. One can infer that lignin is much less oxygenated than carbohydrates, where approximately each carbon atom is accompanied by one oxygen atom. Therefore, lignin has the highest heating value in woody biomass. Lignin aromatic structure also provides excellent functional chemical sources, e.g., aromatic monomers. Lignin phenylpropane units, guaiacyl (G) and syringyl (S) derived from coniferyl and sinapyl alcohol, respectively, are linked together by different bonds (Figure 3.5) (Pu et al., 2007). The β-O-4 inter-unit linkages are the most abundant in lignin, estimated to be as high as 50% in softwoods and almost 60% in hardwoods. In general, more than two-thirds of the phenylpropane units are linked by ether bonds and the rest by carbon–carbon bonds. Depending on the hardwood species, GS-lignin as a copolymer of coniferyl and synapyl alcohols varies in the G/S ratio from 4:1 to 1:2. G-lignin occurring in almost all softwood is largely a polymerization product of coniferyl alcohol. Different contribution of p-hydroxycinnamyl alcohols in the biosynthesis of hardwood and softwood lignins causes significant differences in their structure, including the contents of different types of bonds and main functional groups (for example, methoxyl (OMe), phenolic hydroxyl (PhOH), aliphatic hydroxyl, carbonyl, and carboxyl groups). Softwood lignin is more condensed than hardwood lignin due to the difference in substitution of the lignin

P1: SFK/UKS P2: SFK c03 BLBS075-Hood December 16, 2010

16:59

Trim: 244mm×172mm

3 Growing Systems for Traditional and New Forest-Based Materials

C C O

C

C

C

C

C

C O

C

C O

C

O

O

C

C O

O

O β-O-4

C

α-O-4

71

O

β-1

dibenzodioxocin

C C

C C

O

C

C

C

C

C

C

C

C

C

C

C

C

C

C

C

C C

O O

O β-5

O β-β

O

O 5-5

O 4-O-5

Figure 3.5. Common linkages between phenylpropane units in lignin (Pu et al., 2007).

precursors; substituted C3 and unsubstituted C5 in coniferyl alcohol versus substituted C3 and C5 in synapyl alcohol. Significant differences observed between hardwood and softwood lignin structures lead to different physical and chemical properties of these two lignin types (Argyropoulos and Menachem, 1997).

3.6.2

Biorefineries Using Woody Biomass

Current technologies for the chemical utilization of wood use a “destructive strategy” to obtain a single, relatively pure, component (cellulose). Other fractions are either wasted or used for power generation that gives a low added-value. The “biorefining” philosophy provides an alternative approach for utilization of lignocellulosic biomass. Lignocellulosic biomass can be “fractionated” into its main components by sequential treatments to give separate streams used for different product applications, maximizing the benefits of a renewable and complex resource by preserving as much of the value inherent in its original structures as possible. Wide applications of the carbohydrate oligomer streams exist. For example, xylooligomers are currently used as novel sweeteners in food additives (Aoyama, 1996; Ichikawa and Mitsumura, 1996). Health applications of indigestible oligosaccharides have been recently reported: for example, Imaizumi et al. (1991) found that a xylooligosaccharide-based diet reduced the blood concentrations of sugars and lipids on diabetic rats, and Toyoda et al. (1993) reported that xylooligosaccharides improved calcium absorption by rats. The phagocytic activity of neutrophils in mice was enhanced by either oral or intraperitoneal administration

P1: SFK/UKS P2: SFK c03 BLBS075-Hood December 16, 2010

72

16:59

Trim: 244mm×172mm

Plant Biomass Conversion

of xylooligosaccharides (Aoyama, 1996); increased resistance of mice towards infection by Clostridium difficile caused by xylo- and fructo-oligosaccharides was reported by May et al. (1995). Improvements in the gastrointestinal health of rats caused by a xylooligosaccharidecontaining diet have been reported by Campbell et al. (1997). In relation to human health, xylooligosaccharides selectively enhance the growth of bifidobacteria, thus promoting a favorable intestinal environment (Okazaki et al., 1990; 1991). Sequential incremental deconstruction of wood into cellulosic, hemicellulosic, and ligninbased streams holds significant promise for industrial development. Each of these streams can be further processed to yield valuable products in a biorefinery setting. Papermaking pulps can be produced from the cellulosic components, biofuels including ethanol and bioplastics produced by fermentation from the hemicellulosic and cellulosic streams, and other chemical components from lignin. We address the fundamental challenges of biomass conversion to both conventional products—materials, chemicals and energy—and unconventional (new) ones not currently used in our daily life. In cellulose fibers, cellulose molecules fit together snugly lengthwise via hydrogen bonding. In conversion of woody biomass to chemicals, materials, and energy, systematic incremental deconstruction can lead to energy-efficient synthesis of revolutionary new forms of matter with tailored properties. For example, paper today is made of cellulose fibers. If we further split fibers into fibrils and make paper with fibrils, the outcome can be significantly different. Deconstruction of woody biomass one component at a time reduces the chances of forming undesired side products. We can more effectively control the conversion process and make better use of the released products with sufficient scientific knowledge applied. When biomass or a constituent of woody biomass is oxidized, it produces carbon monoxide, carbon dioxide, and water. The composition depends on oxygen availability. Incremental deconstruction of woody biomass requires transformation in a controlled way, far away from equilibrium. For example, partial oxidation of aromatic compounds or lignin to aldehydes requires the oxidation reaction to stop before completion, or equilibrium. Much knowledge is needed to allow the necessary control to develop effective commercial processes. The knowledge developed in the fundamental characterization of biomass and the identification of chemical linkages that make it difficult to convert lignocellulose will positively impact the development of engineered plants specifically tailored for fuel conversion. These engineered plants are envisioned to (1) contain weak links within the cell wall components that make them more amenable to deconstruction, (2) have certain chemical, genetic, and/or environmental triggers that initiate the deconstruction process, and/or (3) have been engineered so that certain structural and cross-linking elements no longer exist. The key to obtaining benefits associated with success is the development of the science underlying biomass disassembly, separation, and conversion processes. This will allow an increase in production and economic utilization of currently underutilized components of woody and agricultural biomass available in the United States. Woody biomass crops such as willow, cottonwood, aspen, alder, and Loblolly Pine could very significantly increase the amount of biomass available for processing when the science is developed that relates the different properties among species to disassembly, separation, and conversion methods. The method of growing the wood, the age of the wood, the height above ground, and the competitive situation of the tree producing the wood all affect the wood chemical, anatomical, and physical traits that affect its value and the preferred use and disassembly method (Amidon, 1975). As new forest-based materials become a larger component of the forest or purpose-grown biomass value, the agricultural and silvicultural systems employed will adapt to enhance the production of other value components or to make the disassembly easier.

P1: SFK/UKS P2: SFK c03 BLBS075-Hood December 16, 2010

16:59

Trim: 244mm×172mm

3 Growing Systems for Traditional and New Forest-Based Materials

3.6.3

73

Hot-Water Extraction of Hemicellulose

One currently used disassembly procedure that is moving to commercial use is an example to show some of the key concepts. The method is Hot-Water Extraction of Hemicellulose. The key to this hot-water extraction process is to provide a beginning for systematic sequential deconstruction of woody biomass. Fractionation of woody feedstocks into their respective process streams (cellulose, hemicelluloses, lignin, extractives, etc.) is of interest for commercial operations. The elementary processes occurring during the conversion of biomass depended on the rate and extent of treatments that affect the structure and chemical composition of the woody biomass. The use of water as the sole solvent is a strength of this approach. The major constituents of lignocellulosic biomass (cellulose, hemicelluloses and lignin) cannot be simultaneously isolated as polymers. Extraction or prehydrolysis (e.g., Amidon et al., 2008; Conner and Lorenz, 1986; Garrote et al., 1999; 2001; Liu and Wyman, 2005; Overend et al., 1987; Weil et al., 1997) has a wide range of applications, including (Heitz et al., 1991) (1) fractionation or pulping processes, in which there is removal of hemicelluloses with selectivity towards cellulose degradation and splitting the α and β aryl ether bonds of lignin (Lora and Wayman, 1978; Schultz et al., 1984); (2) defibration for fibreboard production, in processes using high pressure steam; and (3) as a pretreatment for the enzymatic hydrolysis of cellulose (Dekker and Wallis, 1983; Grethlein and Converse, 1991). Since xylan is the main component of hardwood and grass hemicelluloses, xylooligomers and xylose are the main products obtained in hydrothermal treatments of this raw material. Sugar-degradation products (such as furfural or hydroxymethylfurfural) and acetic acid (generated from acetyl groups) can also appear in the reaction media. Extraction and subsequent acid hydrolysis are commonly carried out in water media. There are many advantages of using water as the sole solvent. However, harsh reaction conditions, i.e., either high temperature or a combination of high temperature and strong acid, are required and thus limit the processes to large-scale operations for economic success. Enzymatic hydrolysis is a slow process and is still in the process of developments (Gan et al., 2009; Zhang and Lynd, 2003). In reactions involved with woody biomass, either by pulping or extraction/autohydrolysis, a homogeneous model has largely been implicitly applied. For example, the kinetics of the Kraft pulping process is expressed through the following empiricism: −

dL = (ka [OH − ]a + kb [OH − ]b [HS − ]c )L d dt

(3.1)

where L is the concentration of lignin in wood, [OH− ] is the alkali or hydroxyl ion concentration in the cooking liquor, [HS− ] is the hydrosulfide ion concentration in the cooking liquor, and ka and kb are reaction rate constants. Different values of parameters a, b, and c were found for each stage during the delignification process. The transition points between the stages depend on the lignin content and the cooking conditions and are determined empirically by plotting carbohydrate content versus lignin content. There are two imperfections in the simplistic Equation 3.1, first, the rate of reaction is represented simply as rate change of reactant, and second, the mechanism is lost. A direct challenge to this empiricism is the gradual changing of order of reaction as it progresses (Li et al. 2002; Yang and Liu 2005) despite its popularity in dealing with woody biomass. On the basis of the surface reaction theory, Liu (2004; Liu and Wyman, 2005) and Yang and Liu (2005) inserted more fundamental kinetic approaches into developing bleaching and

P1: SFK/UKS P2: SFK c03 BLBS075-Hood December 16, 2010

16:59

Trim: 244mm×172mm

Plant Biomass Conversion

74

pulping reactions involving wood and fibers. This introduction did explain why the orders of reaction in regard to the reactants in the liquid phase change with concentration range. In Yang and Liu (2005), the Kraft pulping kinetics was described by seven steps: (1) transport of hydroxide and hydrosulfide ions from the bulk liquor to the exterior surface of the chip, (2) diffusion of the chemical ions to the interior of the chip, (3) chemisorptions occur on the interior surface, (4) surface reaction between the chemical ions and lignin, (5) desorption of the soluble lignin degradation products, (6) diffusion of the soluble lignin degradation products to the chip exterior, and (7) transport of the soluble lignin degradation products in the bulk liquor. This is appropriate in kinetics of complex materials and reactions. However, the solid phase is not acting as a catalyst. The solid phase contains both reactant and the desired product. Under normal conditions, transport steps 1 and 7 are probably unimportant but the diffusion steps of 2 and 6 may play a significant role unless the effective thickness of the biomass particles is less than the critical thickness, say 2–3 mm (e.g., Hatton and Keays, 1973; Wilder and Daleski, 1965). Under 2–3 mm thickness conditions, these transport steps are negligible in effect on the overall observed pulping rate. The chemical reactions involved are said to be the rate-controlling steps, which means the overall reaction rate is dependent on steps of 3–5 directly, provided that the wood chips with the thickness of 4 mm or less are cooked. In many cases, our interests are on the intrinsic kinetics, i.e., steps 3–5, only. For the aqueous hydrolysis of woody biomass catalyzed by acid, the underlying principles of chemical kinetics are similar. However, the reactive sites are known not to be at the end of the molecular chain. The hydronium ion needs to access the O at β-1→4, forming the H. + .. complex as the first step towards the dissolution–hydrolysis reaction. Therefore, swelling −O− of the biomass matrix to expose the β-O-(1,4) linkage to the surface or solid–liquid interface is highly desirable. While the true mechanistic steps of the hydrolysis reactions catalyzed by acid can be very complex, we use a simplistic preliminary model to describe the behavior:

R1 − O − R 2 + H

H+ • R1 − O − R 2

+

H+ • + H2O R1 − O − R 2

H+

H

R1 − O

O

H

H+

H

R1 − O

O

(3.2)

H (3.3)

R2

R 1OH + R 2OH • H +

(3.4)

R2

R2 OH•H +

R 2 OH + H +

(3.5)

In the above model, Equations 3.2–3.5, the hydronium ion (H+ ) is a catalyst. Reactions 3.2 and 3.3 can be considered as adsorption steps, while reaction 3.5 is a desorption step.

P1: SFK/UKS P2: SFK c03 BLBS075-Hood December 16, 2010

16:59

Trim: 244mm×172mm

3 Growing Systems for Traditional and New Forest-Based Materials

75

R1 and R2 can be carbohydrate groups, acetyl group, or methoxyl group. If the ROH is a small enough molecule, it effectively dissolves in the solvent (water) and becomes part of the solution. Reaction 3.4 is a hydrolysis reaction. Therefore, this hydrolysis reaction is the key reaction in the woody biomass conversion in aqueous medium. After the carbohydrates dissolve in the water, the hydrolysis reaction continues. The kinetics can be again represented by reactions 3.2–3.5). The only difference is that in the “extraction” stage, the carbohydrates are in the solid phase, whereas the carbohydrates are in the liquid phase during hydrolysis or “post-hydrolysis.” The hydrolysis ends when all the carbohydrates are converted to monosaccharides. In an acid medium, carbohydrate monomers (xylose and glucose, for example) can be dehydrated and thus form undesired products if the sugars are the desired products. The most common dehydration products are hydroxylmethylfufural (HMF, from 6-carbon sugars) and furfural (from 5-carbon sugars). Many more products can form from furfural and HMF, including levulinic acid, succinic acid, etc. Furfural, HMF, and organic acids, such as acetic acid, are potent inhibitors of the subsequent fermentation processes for biofuels. In order to control the conversion process, we must understand the chemical reactions occurring in the process. Therefore, kinetic studies are essential in the commercialization of the biomass conversions. The solid residue after complete hydrolysis can be recovered by filtration (number 3 porosity glass filter) and considered to be Klason lignin. The monosaccharides, acetic acid, and other compounds contained in the hot-water extract liquor can be determined by 1 H NMR spectroscopy (Kiemle et al., 2004). Wood extracts and hydrolyzate have been characterized by 1 H NMR extensively using the method developed at ESF (e.g., Kiemle et al., 2004). While the NMR analysis is reliable when the compositions are predominantly monomers and dilute, it does present challenges to complex mixtures. These challenges have made any research program more interesting. Figure 3.6 shows a 2D HSQC NMR (Two Dimensional Heteronuclear Single Quantum Correlation Nuclear Magnetic Resonance) spectrum we have been able to obtain on a wood extract sample. On the top, it is shown a 1 H NMR spectrum. One can observe the challenge of distinguishing the different molecules as shown inside the box by this top curve. Analytical technique development will continue, as knowledge of the complex mixture is needed for fundamental understanding that can lead to expanded commercial efforts.

3.6.4

Wood Extracts: Processing and Conversion

The conversion of woody biomass to platform chemicals, materials and energy can be achieved via two main routes: biochemical and thermochemical, with some overlap. A biochemical route has been the focus of this description. When dealing with cellulose and hemicellulose utilization, water as the solvent holds advantages over all other choices because of environmental and product recovery advantages. However, ionic liquids as solvents are also showing promise and may become major methods in the future. Recent work has shown that a cellulose solubility of up to 30 wt% can be achieved in ionic liquid solvents (Heinze et al., 2005). Ionic liquids in combination with acidic catalysts could also be used to promote the hydrolytic deconstruction of cellulose and hemicelluloses to five- and six-carbon sugars (Zhao et al., 2007). Many scientific strides are needed to push the utilization opportunities enough to enable biorefinery commercialization.

P1: SFK/UKS P2: SFK c03 BLBS075-Hood December 16, 2010

76

16:59

Trim: 244mm×172mm

Plant Biomass Conversion

ppm 92 GluN Glu Gal GluA Ara Xyl

94 GluN Rha

96

Man

Man Rha

Glu

Gal

GluA

98 Ara 100

Xyl

Oligomers

β-Region

13C

102 104

α-Region Oligomers

106 108

5.6

5.5

5.4

5.3

5.2

5.1

5.0

4.9

4.8

4.7

4.6

4.5

4.4

4.3

ppm

1H

Figure 3.6. 2D HSQC NMR spectrum for a concentrated wood extract sample. Credit Dr. B. Bujanovic, SUNYESF.

Fractionation and Concentration of Wood Extracts Microfiltration of extracts is conducted to separate colloidal and particulate material in the extract. Following microfiltration, separation processes are applied for the removal of acetic acid and other small molecules from the wood hydrolyzates, leaving purified sugar streams for fermentation. These processes include nano and reverse osmosis membranes and adsorptive membrane systems. Hot-Water Wood Extract Hydrolysis One method of hydrolysis of hot-water wood extract is to add xylanase to reduce oligomeric xylose to monomers. This will not hydrolyze non-xylan/xylose-derived oligomers and can take a long time. A second hydrolysis method of hydrolysis of hot-water wood extracts in aqueous media is to use mineral acid(s). Challenges remain on the enzymatic xylan hydrolysis due to the complex mixture generated during the hot-water extraction process. Microorganism adaption/evolution is being carried out to meet the challenges. Kinetic studies of the hydrolysis have focused on the molecular level mechanistic evaluations, coupled with computational modeling (Liu, 2008). The ability to control, separate, and analyze the low molecular weight polysaccharides is of great importance. Both proton and carbon NMR can be combined (i.e., 2D NMR) to identify and characterize the polysaccharides.

P1: SFK/UKS P2: SFK c03 BLBS075-Hood December 16, 2010

16:59

Trim: 244mm×172mm

3 Growing Systems for Traditional and New Forest-Based Materials

77

P

Permeate holding tank

Feed and concentrate tank

Figure 3.7. Schematic of the Nano Filtration unit setup. Multiple passes are required for obtaining the substrate for ethanol and PHA fermentation: sugars and levulinic acid stream from the wood hydrolyzate.

Hydrolysate Fractionation and Sugar Stream Purification One critical problem in fermenting wood extracts is the toxicity of various lignin-derived components and phenolics to the microorganisms. Furthermore, small molecules such as acetic acid, furfural, and 2-hydroxy methyl-furfural are potent inhibitors affecting the commercial viability of a biorefinery processes. The success of manufacturing bioproducts is dependent upon reliable and efficient means to separate and purify these wood extract liquors to render them easily fermentable, while recovering additional products. Selective removal of various toxic compounds such as acetic acid, furfural, and other aromatics prior to fermentation is necessary. Development of a comprehensive database for the selection, design, and optimization of separation and purification processes for the extracts is ongoing. The ability to co-flocculate, adsorb, and recover inhibitory compounds is also a developing technology. The dispersed phases can then be separated by (1) sedimentation, (2) centrifugation, (3) dead-end (cake) filtration, and (4) cross flow filtration. Parameters for scale-up of these separation operations such as the specific filtration resistance, permeability, and compressibility of the sediments or dispersed phases are still being developed. For the membrane separation/purification processes, flux–transmembrane pressure maps to identify optimal operating conditions have been obtained. A schematic of the membrane unit is shown in Figure 3.7. The objective of using the membrane is to eliminate the commonly referred to detoxification steps in most cellulosic ethanol process. This is a unique approach at SUNY ESF (Liu, 2008; Amidon et al. 2008). Product recovery is promoted over the generation of a waste stream by detoxification procedures.

Hot-water Wood Hydrolysate Fractionation and Ethanol and PHA Fermentations The sugars solutions, after membrane separation as described above, have been found to be directly usable by Burkholderia cepacia, E. coli fbr5, and Pichia stipitis to convert the separated sugars to PHA and ethanol.

P1: SFK/UKS P2: SFK c03 BLBS075-Hood December 16, 2010

78

3.6.5

16:59

Trim: 244mm×172mm

Plant Biomass Conversion

Residual Solid Wood Biomass: Processing and Conversion of the wood mass after extraction, an example

Pulping Cellulose fiber currently enjoys extensive utilization by the paper industry. One of the most surprising results from recent research is the observation that when hot-water extraction is used to extract ∼15% of the chip mass or more prior to pulping, the cooking time in the Kraft process can be decreased from 120 minutes to <30 minutes at 165◦ C (Amidon et al., 2006; 2008). Further research at SUNY-ESF indicates that the cleavage of lignin carbohydrate complexes (LCCs) during the hot-water extraction may be a major contributor to subsequent leaching of lignin out of the chips under alkaline conditions. Hot-water extraction would remove a significant amount of xylan from the chips and this produces in pulps with lower tensile strength (Francis et al., 2007). The xylan in pulp acts as a binder both within and on the fiber surface. Saake et al. (Saake et al., 2005) has reviewed the literature and also present experimental results showing that strength values of paper sheets can be increased 50%–150% by addition to the papermaking process of a few percent of wood or other plant xylan along with synthetic wet strength additives. These wet strength additives (kymene for example) are already being used in papermaking (Saake et al., 2005). The residual extracted chips have also been shown to have enhanced properties for burning, pellet production, and reconstituted wood products. The improved properties are mainly a result of decreased moisture reactivity due the reduced hemicellulose content and a reduction in ash content that results from hot-water extraction under the acidic conditions that occur in the process as a result, primarily, of acetyl group liberation from the wood to make acetic acid.

3.7 Summary Civilization has been built upon the ability of humankind to exploit the material resources that nature has provided. Liquid fuel and chemicals are the “life blood” of today’s economy. As we move into 21st century, the finite and uneven distribution of natural resources is increasingly apparent, and many of our activities are adversely impacting the environment. The fossil sources can only be replenished when living plants and animals are converted to fossil deposits. This cycle finished over 200 million years ago (Liu et al., 2006). While it was once axiomatic that civilization would always advance into the future, questions of “sustainability” now arise, as fossil resources do not replenish as we use them. Sooner or later, we will have to be able to use “renewable” energy and chemical sources on an industrial scale for our civilization to continue. Forestry and woody biomass will help to achieve that goal.

References Alen, R. 2000. Determination of the main aliphatic carboxylic acids in kraft black liquors by high performance liquid chromatography-mass spectrometry. In: Stenius, P. (ed.) Forest Products Chemistry. Helsinki, Finland: Fapet Oy. Amidon, T. 2006. The biorefinery in New York: Woody biomass into commercial ethanol. Pulp and Paper Canada, 107, 47–50. Amidon, T.E. PhD Thesis, SUNY ESF, 1975.

P1: SFK/UKS P2: SFK c03 BLBS075-Hood December 16, 2010

16:59

Trim: 244mm×172mm

3 Growing Systems for Traditional and New Forest-Based Materials

79

Amidon, T. & Liu, S. 2009. Water-based woody biorefinery. Biotechnology advances, 27, 542–550. Amidon, T., Wood, C., Shupe, A., Wang, Y., Graves, M. & Liu, S. 2008. Biorefinery: Conversion of woody biomass to chemicals, energy and materials. Journal of Biobased Materials and Bioenergy, 2, 100–120. Aoyama, M. 1996. Steaming treatment of bamboo grass. II. Characterization of solubilized hemicellulose and enzymatic digestibility of water-extracted Residue. Cellulose Chemistry and Technology, 30, 385–393. Argyropoulos, D. & Menachem, S. 1997. Lignin. In: Scheper, T. (ed.) Advances in Biochemical Engineering Biotechnology. Berlin, Heidelberg: Springer-Verlag. Bateman, I. & Lovett, A. 2000. Estimating and valuing the carbon sequestered in softwood and hardwood trees, timber products and forest soils in Wales. Journal of Environmental Management, 60, 301–323. Bowyer, J., Briggs, D., Johnson, L. & Al, E. 2001. CORRIM—A report of progress and a glimpse of the future. Forest Products Journal, 51, 10–22. Brown, C. 1976. Forests as energy sources in the year 2000: What man can imagine, man can do. Journal of Forestry (USA), 74, 7–12. Burner, D. & Brauer, D. 2003. Herbage response to spacing of loblolly pine trees in a minimal management silvopasture in southeastern USA. Agroforestry Systems, 57, 69–77. Campbell, J., Fahey Jr, G. & Wolf, B. 1997. Selected indigestible oligosaccharides affect large bowel mass, cecal and fecal short-chain fatty acids, pH and microflora in rats. Journal of Nutrition, 127, 130. Clason, T. 1998. Silvopastoral practices sustain timber and forage production in commercial loblolly pine plantations of northwest Louisiana, USA. Agroforestry Systems, 44, 293–303. Conner, A. & Lorenz, L. 1986. Kinetic modeling of hardwood prehydrolysis. Part III. Water and dilute acetic acid prehydrolysis of southern red oak. Wood and Fiber Science, 18, 248–263. Coyle, D. & Coleman, M. 2005. Forest production responses to irrigation and fertilization are not explained by shifts in allocation. Forest Ecology and Management, 208, 137–152. Davis, J. 2008. Genetic improvement of poplar (Populus spp.) as a bioenergy crop. Genetic Improvement of Bioenergy Crops, 397–419. Dekker, R. & Wallis, A. 1983. Autohydrolysis-explosion as pretreatment for the enzymic saccharification of sunflower seed hulls. Biotechnology Letters, 5, 311–316. Dence, C. & Lin, S. 1992. Introduction. In: Dence, C. & Lin, S. (eds.) Methods in Lignin Chemistry. New York, NY: Springer-Verlag Berlin Heidelberg. Fengel, D. & Wegener, G. 1984. Wood: Chemistry, Ultrastructure, Reactions. Berlin, New York: Walter De Gruyter Inc. Fisher, R. K., Gardiner, E. S., Stanturf, J. A. & Portwood, C. J. 2002. Disking Effects of Fifth-Year Volume Production of Four Eastern Cottonwood Clones Established on an Afforestation Site, Sharkey County, Mississippi. Gen. Tech. Rep. SRS–48. Asheville, NC: U.S. Department of Agriculture, Forest Service, Southern Research Station; pp. 515–519. Francis, R., Amidon, T., Shin, N., Bolton, T. & Kanungo, D. 2007. Mild acid and alkaline pretreatments before kraft and SAQ pulping. Presented at 3rd Int. Coll. on Eucalyptus Pulps, Belo Horizonte, Brazil, March 4th–7th. Gan, Q., Allen, S. & Taylor, G. 2003. Kinetic dynamics in heterogeneous enzymatic hydrolysis of cellulose: An overview, an experimental study and mathematical modelling, Process Biochemistry, 38 (7), 1003–1018.

P1: SFK/UKS P2: SFK c03 BLBS075-Hood December 16, 2010

80

16:59

Trim: 244mm×172mm

Plant Biomass Conversion

Garrote, G., Dominguez, H. & Parajo, J. 2001. Generation of xylose solutions from Eucalyptus globulus wood by autohydrolysis–posthydrolysis processes: Posthydrolysis kinetics. Bioresource Technology, 79, 155–164. Garrote, G., Dom´ınguez, H. & Paraj´o, J. 1999. Mild autohydrolysis: An environmental friendly technology for xylooligosaccharide production from wood. Journal of Chemical Technology & Biotechnology, 74, 1101–1109. Gonzalez, R., Wright, J. & Saloni, D. 2009. Filling a need: Forest plantations for bioenergy in the southern United States. Forest Landowner Magazine, 68, 26–29. Grethlein, H. & Converse, A. 1991. Common aspects of acid prehydrolysis and steam explosion for pretreating wood. Bioresource Technology, 36, 77–82. Gruenwald, H., Brandt, B., Schneider, B. & Al, E. 2007. Agroforestry systems for the production of woody biomass for energy transformation purposes. Ecological Engineering, 29, 319–328. Halls, L. & Schuster, J. 1965. Tree-herbage relations in pine-hardwood forests of Texas. Journal of Forestry, 63, 282–283. Hassan, E., Steele, P. & Ingram, L. 2009. Characterization of fast pyrolysis bio-oils produced from pretreated pine wood. Applied Biochemistry and Biotechnology, 154, 3–13. Hatton, J. & Keays, J. 1973. Effect of chip geometry and moisture on yield and quality of kraft pulps from western hemlock and black spruce. Pulp and Paper Magazine of Canada, 74, T11–T19. Heinze, T., Schwikal, K. & Barthel, S. 2005. Ionic liquids as reaction medium in cellulose functionalization. Macromolecular Bioscience, 5, 520–525. Heitz, M., Capek-Menard, E., Koeberle, P., Gagne, J., Chornet, E., Overend, R., Taylor, J. & Yu, E. 1991. Fractionation of Populus tremuloides at the pilot plant scale: Optimization of steam explosion pretreatment conditions using the STAKE II technology. Bioresource Technology, 35, 23–32. Ichikawa, T. & Mitsumura, Y. 1996. Effects of the addition of new saccharide sweeteners on the properties of sponge cakes. Journal of Home Economics of Japan, 47, 445–452. Imaizumi, K., Nakatsu, Y., Sato, M., Sedarnawati, Y. & Sugano, M. 1991. Effects of xylooligosaccharides on blood glucose, serum and liver lipids and cecum short-chain fatty acids in diabetic rats. Agricultural and Biological Chemistry, 55, 199–205. Jeffries, T. & Jin, Y. 2004. Metabolic engineering for improved fermentation of pentoses by yeasts. Applied Microbiology and Biotechnology, 63, 495–509. Johnson, J., Coleman, M., Gesch, R., Jaradat, A., Mitchell, R., Reicosky, D. & Wilhelm, W. 2007a. Biomass-bioenergy crops in the United States: A changing paradigm. The Americas, 1, 1–28. Johnson, J., Franzluebbers, A., Weyers, S. & Reicosky, D. 2007b. Agricultural opportunities to mitigate greenhouse gas emissions. Environmental Pollution, 150, 107–124. Keenan, T., Tanenbaum, S., Stipanovic, A. & Nakas, J. 2004. Production and characterization of poly-beta-hydroxyalkanoate copolymers from Burkholderia cepacia utilizing xylose and levulinic Acid. Biotechnology Progress, 20, 1697–1704. Kennedy Jr., H. E. & Henderson, W. 1976. Cultivation in cottonwood plantations—Practice and equipment. In: B.A., T. & S.B., L. (eds.) Symposium on Eastern Cottonwood and Related Species., Louisiana State University, Baton Rouge, LA. Division of Continuing Education, pp. 379–384. Kiemle, D., Stipanovic, A. & Mayo, K. 2004. Proton NMR methods in the compositional characterization of hemicelluloses. ACS Symposium Series, 864, 122–139. Kram, J. 2008. Managing woody biomass. Biomass Magazine, April.

P1: SFK/UKS P2: SFK c03 BLBS075-Hood December 16, 2010

16:59

Trim: 244mm×172mm

3 Growing Systems for Traditional and New Forest-Based Materials

81

Leininger, T., Solomon, J., Wilson, A. & Al, E. 1999. A guide to major insects, diseases, air pollution injury, and chemical injury of Sycamore. General Technical Report. Asheville NC: USDA Forest Service, Southern Research Station. Li, A., Zhang, Q., Zhang, G., Chen, J., Fei, Z. & Liu, F. 2002. Adsorption of phenolic compounds from aqueous solutions by a water-compatible hypercrosslinked polymeric adsorbent. Chemosphere, 47, 981–988. Liu, C. & Wyman, C. 2005. Partial flow of compressed-hot water through corn stover to enhance hemicellulose sugar recovery and enzymatic digestibility of cellulose. Bioresource Technology, 96, 1978–1985. Liu, S. 2004. A simplistic mechanistic model and effect of consistency on alkaline peroxide brightening of mechanical pulps. Chemical Engineering Science, 59, 20. Liu, S. 2008. A kinetic model on autocatalytic reactions in woody biomass hydrolysis. Journal of Biobased Materials and Bioenergy, 2, 135–147. Liu, S. & Amidon, T. 2007. Essential components of a wood-based biorefinery. O. Papel, 68, 54–75. Liu, S., Amidon, T., Francis, R., Ramarao, B., Lai, Y. & Scott, G. 2006. From forest biomass to chemicals and energy. Biorefinery initiative in New York State. Industrial Biotechnology, 2, 113–120. Lora, J. & Wayman, M. 1978. Delignification of hardwoods by autohydrolysis and extraction. Tappi, 61, 47–50. Lynd, L., Wyman, C. & Gerngross, T. 1999. Biocommodity engineering. Biotechnology Progress, 15, 777–793. May T., Mackie R.I. and Garleb K.A. 1995. Effect of dietary oligosaccharides in the intestinal growth of and tissue damage by Clostridium difficile. Mikrooekol Ther, 23, 158–170. McKeand, S., Abt, R., Allen, H., Li, B. & Catts, G. 2006. What are the best loblolly pine genotypes worth to landowners? Journal of Forestry, 104, 352–358. Mitchell, W. 1998. Physiology of carbohydrate to solvent conversion by clostridia. Advances in Microbial Physiology, 39, 31. Montagnini, F. & Nair, P. 2004. Carbon sequestration: An underexploited environmental benefit of agroforestry systems. Agroforestry Systems, 61, 281–295. Myerly, R., Nicholson, M., Katzen, R. & Taylor, J. 1981. The forest refinery. Chemtech, 11, 186–192. Okazaki, M., Fujikawa, S. & Matsumoto, N. 1990. Effects of xylooligosaccharide on growth of bifidobacteria. Journal of Japanese Society of Nutrition and Food Science (Japan), 43, 395–401. Okazaki, M., Koda, H., Izumi, R., Fujikawa, S. & Matsumoto, N. 1991. Effect of xylooligosaccharide on growth of intestinal bacteria and putrefaction products. Journal Japanese Society Nutrition Food Science, 44, 41–44. Overend, R. P., Chornet, E. & Gascoigne, J. A. 1987. Fractionation of lignocellulosics by steam-aqueous pretreatments [and Discussion]. Philosophical Transactions of the Royal Society of London. Series A, Mathematical and Physical Sciences, 321, 523– 536. Perlack, R., Wright, L., Turhollow, A., Graham, R., Stokes, B. & Erbach, D. 2005. Biomass as feedstock for a bioenergy and bioproducts industry: The technical feasibility of a billion-ton annual supply, Technical Report A357634. Pu, Y., Zhang, D., Singh, P. & Ragauskas, A. 2007. The new forestry biofuels sector. Biofuels, Bioproducts and Biorefining, 2, 58–73.

P1: SFK/UKS P2: SFK c03 BLBS075-Hood December 16, 2010

82

16:59

Trim: 244mm×172mm

Plant Biomass Conversion

Qureshi, N., Dien, B., Nichols, N., Saha, B. & Cotta, M. 2006. Genetically engineered Escherichia coli for ethanol production from xylose substrate and product inhibition and kinetic parameters. Food and Bioproducts Processing, 84, 114–122. Ranney, J., Wright, L. & Layton, P. 1987. Hardwood energy crops: The technology of intensive culture. Journal of Forestry (USA), 85, 17–28. Roberts, S., Rousseau, R., Hatten, J. & Al, E. 2009. Interspecific competition in a loblolly pine-switchgrass co-culture system for biomass production. Sun Grant Proposal. Rockwood, D., Rudie, A., Ralph, S. & Al, E. 2008. Energy product options for Eucalyptus species grown as short rotation woody crops. International Journal of Molecular Science, 9, 1361–1378. Rousseau, R.J., Adams, J.P. & Wilkinson, D.W. 2008 Nine-year performance of a variety of populus taxons on an upland site in Western Kentucky. Presented at 15th Biennial Southern Silvicultural Research Conference. Hot Springs. AR. 11/18/2008–11/20/2008. Ruark, G., Schoeneberger, M. & Nair, P. 2003. Agroforestry—Helping to achieve sustainable forest management. Report and Papers of the UNFF Intersession Experts Meeting on the Role of Planted Forests in Sustainable Forest Management, 24–30 March 2003, Wellington, New Zealand. Saake, B., Busse, T. & Puls, J. 2005 The effect of xylan adsorption on the properties of sulfite and kraft pulps. Presented at 13th International Symposium on Wood, Fibre and Pulping Chemistry, 16–19 May, Vol. 2, pp. 141–146, Auckland, New Zealand. Schmidtling, R., Robison, T., McKeand, S., Rousseau, R., Allen, H. & Goldfarb, B. 2004. The role of genetics and tree improvement in southern forest productivity. General Technical Report SRS–75. For delivery at the UNFF Intersessional Experts Meeting on the Role of Planted Forests in Sustainable Forest Management, 24–30 March 2003, New Zealand. Schoeneberger, M. 2009. Agroforestry: Working trees for sequestering carbon on agricultural lands. Agroforestry Systems, 75, 27–37. Schoeneberger, M. & Ruark, G. 2003. Agroforestry—Helping to Achieve Sustainable Forest Management. Schroeder, P. 1994. Carbon storage benefits of agroforestry systems. Agroforestry Systems, 27, 89–97. Schultz, T., McGinnis, G. & Biermann, C. 1984. Similarities and differences in pretreating woody biomass by steam explosion, wet oxidation, autohydrolysis, and rapid steam hydrolysis/continuous extraction, Technical Report CONF-840111-4. Scott, D. & Tiarks, A. 2008. Dual-cropping loblolly pine for biomass energy and conventional wood products. Southern Journal of Applied Forestry, 32, 33–37. Stanturf, J., Van Oosten, C., Netzer, D., Coleman, M. & Portwood, C. 2001. Ecology and silviculture of poplar plantations. In: Dickman, D. & Isebrands, J. (eds.) Poplar Culture in North America. Ottawa, Ontario: NRC Research Press. Steinbeck, K., Mcalpine, R. & May, J. 1972. Short rotation culture of sycamore: A status report. Journal of Forestry, 70, 210–213. Stettler, R., Fenn, R., Heilman, P. & Al, E. 1988. Populus Trichocarpa × populus Deltoides hybrids for short-rotation culture: Variation patterns and 4-year field performance. Canadian Journal of Forestry Research, 18 (6), 745–753. Stettler, R., Zsuffa, L. & Wu, R. 1996. The role of hybridization in the genetic manipulation of Populus. In: Stettler, R. F., Bradshaw, Jr., H. D., Heilman, P. E. & Hinckley, T. M. (eds.) Biology of Populus and Its Implications for Management and Conservation. Ottawa, ON: NRC Research Press; pp. 87–112.

P1: SFK/UKS P2: SFK c03 BLBS075-Hood December 16, 2010

16:59

Trim: 244mm×172mm

3 Growing Systems for Traditional and New Forest-Based Materials

83

Sun, Y., Lin, L., Pang, C., Deng, H., Peng, H., Li, J., He, B. & Liu, S. 2007. Hydrolysis of cotton fiber cellulose in formic acid. Energy & Fuels, 21, 2386–2389. Toyoda, Y., Hatakana, Y. & Suwa, Y. 1993. Effect of xylooligosaccharides on calcium absorption. Presented at 47th Ann Meeting Japan Society for Nutrition and Food Sci, Tokyo, Japan. 100. Treasure, E., Cohen, E., McNulty, S. & Al, E. 2008. Vulnerability of the Southeastern United States to climate change. USDA Forest Service. Tuskan, G. 1998. Short-rotation woody crop supply systems in the United States: What do we know and what do we need to know? Biomass and Bioenergy, 14, 307–315. Tuskan, G., Difazio, S., Jansson, S., Bohlmann, J., Grigoriev, I., Hellsten, U., Putnam, N., Ralph, S., Rombauts, S. & Salamov, A. 2006. The genome of black cottonwood, Populus trichocarpa (Torr. & Gray). Science, 313, 1596. Wear, D. & Greis, J. 2002. Southern forest resource assessment: Summary of findings. Journal of Forestry, 100, 6–14. Weil, J., Sarikaya, A., Rau, S., Goetz, J., Ladisch, C., Brewer, M., Hendrickson, R. & Ladisch, M. 1997. Pretreatment of yellow-poplar sawdust by pressure cooking in water. Applied Biochemistry and Biotechnology, 68, 21–40. Wilder, H. & Daleski, E. J. 1965. Delignification rate studies: Part 2 of a series on kraft pulping kinetics. TAPPI Journal, 48, 293. Wolters, G. 1973. Southern pine overstories influence herbage quality. Journal of Range Management, 26, 423–426. Yang, L. & Liu, S. 2005. Kinetic model for kraft pulping process. Industrial and Engineering Chemistry Research, 44, 7078–7085. Zhang, Y. & Lynd, L. 2003. Cellodextrin preparation by mixed-acid hydrolysis and chromatographic separation. Analytical Biochemistry, 322, 225–232. Zhao, H., Holladay, J., Brown, H. & Zhang, C. 2007. Metal chlorides in ionic liquid solvents convert sugars to 5-hydroxymethylfurfural. Science, 316, 1597–1600.

P1: SFK/UKS P2: SFK c03 BLBS075-Hood December 16, 2010

16:59

Trim: 244mm×172mm

P1: SFK/UKS P2: SFK c04 BLBS075-Hood December 15, 2010

20:50

Trim: 244mm×172mm

Chapter 4

Dedicated Herbaceous Energy Crops Keat (Thomas) Teoh, Shivakumar Pattada Devaiah, Deborah Vicuna Requesens, and Elizabeth E. Hood

4.1 Introduction Early implementation of the biomass to biofuels industry will depend on available feedstocks such as agricultural residues and wood waste from mills and forests. However, long-term productivity of the industry will require dedicated energy crops with more predictable biomass production. According to the Billion-Ton Annual Supply report, 48 million dry tons of biomass would be available in 2009 and beyond, from agricultural residues at current harvest logistics for conversion into raw materials (http://www.eere.energy.gov/biomass/publications.html). With increases in productivity, this number could double for crop residues, but would still fall far short of the necessary biomass for transportation fuels. In addition, crop residues from food/feed production can be critical for soil health (see Chapter 7 in this volume). Thus, the need for high-productivity crops whether annual or perennial will be required. It is feasible to achieve over a billion tons of biomass from US forest and cropland within the next few years (Billion-Ton study). At 80 gallons per ton, this amount of biomass could produce 80 billion gallons of renewable fuels, or about half the current usage for US transportation. Dedicated energy crops can come from multiple sources. This chapter focuses on nonwoody, high-yielding annual and perennial crops—Miscanthus, switchgrass, and sweet sorghum (Figure 4.1). Although we recognize that there are many other possibilities, these crops are the top candidates in the USA currently and are the subjects of much recent research. Woody crops are the subject of Chapter 3 in this volume.

4.2 Miscanthus Miscanthus is a member of the grass family Graminae and is closely related to the genus Saccharum (which includes sugar cane), another energy crop. Miscanthus is native in Plant Biomass Conversion, First Edition. Edited by Elizabeth E. Hood, Peter Nelson and Randall Powell.
C 2011 John Wiley & Sons Inc.

85

P1: SFK/UKS P2: SFK c04 BLBS075-Hood December 15, 2010

86

20:50

Trim: 244mm×172mm

Plant Biomass Conversion

a

b

c

Figure 4.1. (a) Miscanthus × gigantueus (Photo credit: Marcia Sofonoff), (b) switchgrass, (c) sweet sorghum (left side of the field) and grain sorghum (right side of the field).

south-eastern Asia, China, Japan, and the Pacific Islands with a few species occurring in Africa. It is a C4 rhizomatous perennial grass that can grow up to 4 m tall. Miscanthus was first introduced into Denmark from Japan in 1935 as an ornamental plant and is now cultivated throughout Europe and North America primarily for energy production with other end uses being explored. Miscanthus is valued for its qualities as an ideal biomass crop and as such in recent years has received widespread attention as a bioenergy crop. Extensive evaluation of Miscanthus as a biomass crop has been going on in Europe for the last two decades and its yield potential has been shown to be substantial (Jones and Walsh, 2001). It is being tested in Europe as solid fuel for combustion in farm heating plants as well as for cocombustion with coal, straw, and wood to generate power. Evaluation of Miscanthus as a biomass crop in North America is fairly recent compared to switchgrass, another perennial grass common in the Midwest of the United States and adopted by the Department of Energy (DOE) as the model biomass crop. However, projected yields of Miscanthus in the US Midwest, based on a mathematical productivity model developed for Miscanthus from the European studies (Clifton-Brown et al., 2002) and yields recently reported from field trials in Illinois (Heaton et al., 2008), far exceed those of switchgrass. The promising yield deserves serious investigation into the potential of Miscanthus as a dedicated energy crop in North America in addition to switchgrass, to address the energy security of the country and global carbon emissions. It is estimated that about 14 species of Miscanthus are known (Hodkinson et al., 1997). The most common species investigated as a biofuel in Europe and N. America is Miscanthus × giganteus (Figure 4.1), a naturally occurring sterile triploid hybrid that has its origin in Japan (Greef and

P1: SFK/UKS P2: SFK c04 BLBS075-Hood December 15, 2010

20:50

Trim: 244mm×172mm

4

Dedicated Herbaceous Energy Crops

87

Deuter, 1993). The genetic origin of M. × giganteus is uncertain but some evidence suggests that this hybrid is a result of a cross between M. sinensis (diploid) and M. sacchariflorus (tetraploid) (Linde-Laursen, 1993). In general, M. sinensis types are well adapted for cooler climates, whereas M. sacchariflorus can provide genetic resources for warmer regions.

4.2.1

Characteristics That Make Miscanthus a Potential Biomass Crop

An ideal potential biomass crop should have a sustained capacity to capture and convert the available solar energy into harvestable biomass with highest efficiency, with lowest energy and financial inputs and with minimal environmental impacts. Miscanthus, like all C4 plants, is highly efficient in converting solar energy into biomass. The photosynthesis potential of C4 plants is estimated to be 40% greater than C3 plants (Monteith, 1978). Miscanthus × giganteus is more tolerant of low temperatures than most C4 perennials, allowing it to maintain its photosynthesis and leaf growth even at temperatures as low as 10 ◦ C (Naidu and Long, 2004; Naidu et al., 2003). This enables M. × giganteus to establish a photosynthetically active canopy earlier in spring and persist longer in the fall, thereby lengthening its productive period. As a perennial grass, Miscanthus is a low maintenance crop that can persist for 15– 20 years requiring no tillage once it is established, thus lowering cost of energy and labor and significantly reducing soil erosion and nutrient loss. An established crop of Miscanthus requires very little fertilizer to support growth because the plants are able to acquire and conserve large quantities of nutrients and translocate these nutrients into the rhizome at the end of a growing season to support regrowth in the following season. This will amount to a substantial savings in fertilizer cost and at the same time minimize environmental problems caused by potential water pollution from leachates and run-off. Miscanthus can be grown in a wide variety of soil, which means that the crop can also be grown on marginal land or land not being used for crop production. Often, a concern when a new species is introduced into an area is the risk of escape. Miscanthus × giganteus is a sterile hybrid and poses little risk as an invasive species because it does not produce viable seed, though it does have an active rhizome.

4.2.2

Agronomy

Establishment Miscanthus can be established from seeds, rhizomes, micropropagated plantlets derived from tissue culture, and stem cuttings. The short growing period and the cold winter in Europe and the Midwest of continental North America do not favor establishment by seeds. The use of micropropagated plantlets to establish Miscanthus offers potentially a much higher multiplication rate, but conventional micropropagation is considered too expensive to make commercial production economically viable. Moreover, issues still need to be resolved concerning low winter survival rate in the first year with plants established from micropropagated plantlets. Establishment by rhizomes is the preferred method for achieving high survival rate. Rhizomes or rhizome pieces are planted directly in the field. Automated precision planting of the rhizome on a commercial scale is now possible in Europe. The factors identified as important in establishing a Miscanthus crop from rhizomes are the size of rhizomes, depth of soil, and condition of the rhizomes (fresh or stored over winter). The field trials in Europe have demonstrated that the greatest success with establishment from rhizomes is achieved with

P1: SFK/UKS P2: SFK c04 BLBS075-Hood December 15, 2010

88

20:50

Trim: 244mm×172mm

Plant Biomass Conversion

fresh, healthy rhizomes approximately 200 mm in length, planted at soil depths of 200 mm (Christian and Haase, 2001). Establishing Miscanthus from stem cuttings has been tested in Europe and shown to be feasible but requires further investigation. Miscanthus has been established successfully in Europe from latitude 37◦ N to 56◦ N. In N. America, Miscanthus has been established successfully in the Midwest from latitude 38◦ N to 48◦ N covering Ohio, Michigan, Indiana, and Illinois, and expanding into other areas in the south like Georgia and Florida and as far north as Quebec, Canada.

Factors Influencing Establishment Miscanthus prefers the mild temperatures and high water availability of its natural habitats in the tropics and subtropics, but the successful establishment of Miscanthus × giganteus in Europe and the US Midwest suggests that it is relatively tolerant of temperate temperatures and water availability. Nevertheless, low temperature still limits the growth of Miscanthus. This has been observed in field trials in Europe where the winter survival rate in Northern Europe is low and harvestable yield is 40% lower than in Central and Southern Europe (CliftonBrown et al., 2001b). Similar effects, but to a lesser extent, are observed in field trials in Illinois where the yield in the north is 25% lower than in central and southern Illinois (Heaton et al., 2008). Water availability affects the success rate of establishment in the first year. In locations where precipitation is high such as northern and central Europe M. × giganteus has been successfully grown without irrigation, but in southern Europe where annual rainfall is typically below 20 inches (500 mm), irrigation is necessary for substantial growth and biomass yields (Clifton-Brown et al., 2001b). Winter loss in the first year of Miscanthus establishment has been a significant problem in northern Europe. Two factors were notably important in improving winter survival in the first year of establishment. Plants established from rhizomes have a significantly higher winter survival rate than plants established from micro-propagation. This is often attributed to the reserves stored in rhizomes. Plants established from fresh rhizomes are less likely to succumb to the adverse winter conditions (5%–30% winter loss) compared to plants established from rhizomes that have been stored prior to planting (50%–90% winter loss). Rhizomes that have been stored for a certain period of time before planting are likely to dry out and exhibit less vigor. Interestingly, field trials in Illinois did not experience any significant winter loss in the first year. The northern site experienced only a 14% loss during the first year and no losses in the central and southern sites. This is in contrast to 100% loss in some European sites even though the mean winter minimum temperature may be milder than in the US Midwest. The timing of planting also contributes to the success of crop establishment. In order to avoid the damaging effect of frost, rhizomes are normally planted after the risk of the latest spring frost is over, typically from March until May depending on the climate while micro-propagated plants or pot-grown plants are planted later (mid-April to May). No disease has been reported to date, but the crop has been known to be susceptible to Fusarium blight and Barley Yellow Dwarf Luteovirus (Walsh and McCarthy, 1998). The slow initial growth of Miscanthus means weed control is essential during the establishment phase of the crop. Fields are typically sprayed with herbicide before planting followed by at least two sprayings a year for the first 2–3 years. A range of selective herbicides containing an active ingredient appropriate for cereals (with the possible exception of some graminicides) are deemed suitable for Miscanthus. C4-specific herbicides such as atrazine could also be used (Bullard et al., 1995).

P1: SFK/UKS P2: SFK c04 BLBS075-Hood December 15, 2010

20:50

Trim: 244mm×172mm

4

Dedicated Herbaceous Energy Crops

89

Planting, Harvesting, and Storage Rhizomes of Miscanthus can be planted using existing farm equipment such as potato planters and manure spreaders, and the plants harvested by a modified forage harvester. In Europe, significantly improved and patented planting equipment designed specifically for Miscanthus has increased the speed and efficiency of the farming process. This planting equipment can establish 16 hectares of rhizomes a day compared to 6 hectares a day with traditional farm equipment (Nixon et al., 2001). Harvesting is carried out after the crop has senesced with moisture content preferably at its lowest and with nutrients translocated into the rhizomes. The transfer of nutrients from the plants to the rhizomes is important as it will ensure winter survival and support regrowth the following spring. Harvesting is timed to achieve a balance between attaining lowest moisture content and minimizing biomass losses caused by adverse winter conditions. Harvesting is typically done between late autumn and early spring before plant regrowth. Crops grown in cooler climates are typically harvested in early spring when the moisture content is lowest. Crops grown in warmer climates reach their maturity earlier and can be harvested in late autumn to avoid biomass losses caused by adverse winter conditions or in early spring when the moisture content is lowest. However, delaying harvest until early spring results in yield reduction by as much as 30%–50% due to winter losses of dead and decaying leaves and upper stem parts. Such losses are tolerated because lower moisture content (20%) improves the fuel quality, permits ease of handling, and requires little need for drying. Drying of the harvested crop is necessary to ensure stability during storage so that the biomass material is available throughout the year. A few methods are available to dry Miscanthus. Some methods employ mechanical ventilation during storage, while others employ industrial dryers, but the most cost-effective way is to leave the crop out in the field to dry using ambient air and solar radiation. Miscanthus can be stored in the open air with or without covering or in low-cost farm buildings with roof and open sides. Harvested Miscanthus can be processed into different fuel types, such as (1) chopped into specified length and bulk transported, (2) baled, or (3) formed into pellets and briquettes. The end use of the biomass ultimately determines the methods with which the harvested crop is handled as regards to harvest, storage, and transport. For a more detailed discussion of handling strategies for the Miscanthus crop, readers are referred to the following references (El Bassam and Huisman, 2001; Nixon and Bullard, 2003).

Yields The first year crop does not yield sufficient biomass to merit harvesting, hence the crop is normally harvested from the second year onwards, but yields will continue to improve after year two until they level off. Ceiling yields can be reached in two years under good growing conditions but may take up to 5 years at some locations. The ceiling yields are attained more quickly in warmer climates and those total yields are higher than in cooler climates. The yields reported for the field trials both in Europe and in the United States so far have been promising. The winter yields reported for Europe following the third growing season varied with location and ranged between 7 and 26 dry ton/ha from northern to southern Europe (Clifton-Brown et al., 2001a). These field trials were conducted in 15 locations across the European Union countries, extending as far north as Denmark (56◦ N) and as far south as southern Italy (37◦ N). In North America, Miscanthus has been successfully established in many regions, but so far extensive evaluation of its productivity has been reported in only one region. This evaluation comes from field trials conducted in Illinois by researchers at the University of Illinois, beginning in 2001 covering the northern, central, and southern plains of the state. The reported

P1: SFK/UKS P2: SFK c04 BLBS075-Hood December 15, 2010

90

20:50

Trim: 244mm×172mm

Plant Biomass Conversion

average winter yields of dry matter for the 3 years following the third growing season were 20.9, 33.4, and 34.6 dry ton/ha in the north, central, and south, respectively (Heaton et al., 2008). The average yield per year across the three sites is approximately 30 dry ton/ha. This yield is significantly higher than the yields reported in Europe and almost three times higher than for switchgrass cv. Cave-in-Rock, grown side-by-side with Miscanthus in the same field trials (Table 4.1) (Heaton et al., 2008). The 2005 DOE and USDA supported Billion-Ton study projected 377 million dry tons of plant material from perennial crops on approximately 24 million hectares of agricultural land (Perlack et al., 2005). A yield of 30 t/ha would achieve the projected target of 377 million tons on 12.6 million hectares, half the area estimated in the Billion-Ton study projections. An assessment of fertilizer requirements from field trials in Europe and from a quantitative review of literature by Heaton et al., (2004) revealed that nitrogen has only a modest influence on the yields of M. × giganteus and relatively low fertilizer application is required during the growing season. It is estimated that the annual requirement of nitrogen is 50–70 kg/ha, phosphorus is 5–10 kg/ha, and potassium is 70– 100 kg/ha (Christian and Haase, 2001; Himken et al., 1997; Lewandowski et al., 2000). The yield of Miscanthus is strongly influenced by the availability of water, especially during the growing season. It is estimated that Miscanthus requires a minimum of 20 inches (500 mm) of water per year (Long and Beale, 2001). Application of fertilizer is generally not needed in the first two years of Miscanthus establishment unless the soil is very lacking in important nutrients like phosphorus and potassium. The plant is able to acquire and conserve large quantities of nutrients; therefore, relatively low fertilizer applications are required to support growth once the crop is established.

4.3 Sweet Sorghum Sweet sorghum is a variety of sorghum, Sorghum bicolor (L.) Moench. Several common names include sorghum (English), zuckerhirse (German), durra (Africa), Jowar (India), and bachanta (Ethiopia). It belongs to the grass family, Gramineae. The plant has a high concentration of soluble sugars in the sap or juice. This crop is attractive because of the easy accessibility of readily fermentable sugars combined with very high yields of green biomass. Most countries where sweet sorghum is cultivated have their own breeding programs aimed at the development of sweet sorghum varieties adapted to different cultivation conditions or for producing higher yields (Koppen et al., 2009). However, no overview is available on the number and description of common varieties. Sweet sorghum was introduced to the United States in the early part of the 17th century and has been extensively cultivated in the United States since the 1850s for use in sweeteners, primarily in the form of sorghum syrup. By the early 1900s, the United States produced 20 million gallons of sweet sorghum syrup annually. Currently, less than 1 million gallons are produced annually in the United States, mostly in Alabama, Arkansas, Florida, Georgia, Iowa, Kansas, Kentucky, Louisiana, Mississippi, Nebraska, North Carolina, Oklahoma, and Tennessee (Holmseth, 2008). Interest in sweet sorghum to produce biofuels is not a new concept. In the mid-1970s, during the first energy crisis, significant research was carried out to explore the development of sweet sorghum as a supply for biofuels and energy production (McBee et al., 1987). Research emphasis was focused on developing high-yielding varieties (Clark, 1981; Hallam et al., 2001). When worldwide oil prices dropped in the late 1980s, demand for alternative fuels was reduced and consequently efforts to develop high-yielding varieties were concomitantly

Miscanthus (t/ha) (±1 SE) 25.1 (2.5) 31.1 (3.2) 44.1 (2.6) 33.4 (2.8)

Switchgrass (t/ha) (±1 SE) a

Miscanthus (t/ha) (±1 SE)

13.7 (1.6) 19.1 (2.3) 29.9 (3.3) 20.9 (2.4)

Year

2004 2005 2006 Average

Source: Adapted from Heaton et al. (2008). a Signifies data not available.

7.8 (0.6) 7.7 (1.0) 7.8 (0.6)

Central

North

12.8 (1.2) 10.6 (1.3) 15.6 (2.6) 13.0 (1.1)

Switchgrass (t/ha) (±1 SE) 37.3 (3.0) 27.3 (5.7) 39.2 (2.9) 34.6 (2.6)

Miscanthus (t/ha) (±1 SE)

South

7.8 (0.6) 9.1 (2.6) 6.7 (1.1)

a

Switchgrass (t/ha) (±1 SE)

25.4(3.2) 25.8 (3.8) 37.7 (2.4) 29.6 (1.8)

Miscanthus (t/ha) (±1 SE)

State Average

12.8 (1.2) 7.9 (0.8) 15.6 (2.6) 10.4 (1.0)

Switchgrass (t/ha) (±1 SE)

20:50

Table 4.1. Comparison of mean harvestable dry matter after complete plant senescence from Miscanthus and switchgrass grown at three locations in the Midwest USA during 2004–2006 (n = 4).

P1: SFK/UKS P2: SFK c04 BLBS075-Hood December 15, 2010 Trim: 244mm×172mm

91

P1: SFK/UKS P2: SFK c04 BLBS075-Hood December 15, 2010

92

20:50

Trim: 244mm×172mm

Plant Biomass Conversion

reduced. The renewed interest in bioenergy has increased research activities in sweet sorghum as a dedicated bioenergy crop for its fermentable juice as well as biomass for the production of cellulosic ethanol. Breeding and development programs for sweet sorghum germplasm and hybrids are being conducted at several universities in the United States (University of Kentucky, Purdue University, Louisiana State University, Texas A&M University, University of Florida and Oklahoma State University).

4.3.1

Biology of Sweet Sorghum

Sweet sorghum is typically an annual, but some cultivars are perennial. Propagation is accomplished through seeds. It is easily grown in areas that are too dry for maize. Sweet sorghum originated in Ethiopia, although today it has spread to other parts of Africa, India, Southeast Asia, the United States, and Europe, in semi-arid to humid climates. Plants grow in clumps and height of stalks ranges from 0.8 to 4 m (Figure 4.1). The thickness of stalks also varies, ranging between 1.25 cm and 3.75 cm. Prop roots regularly grow from the lower nodes. Seeds are produced by self-pollination from a panicle that emerges at the top of the plant and contains both the male and female inflorescences. Seeds are small, round, and may be white, yellow, brown, or red in color. Each panicle can produce up to 4,000 starch-containing grains. Sweet sorghum, a C4 crop that is very drought resistant, shows good adaptability to poor soil types, including saline soils. It has a very short vegetation period and thus is ideal for double cropping, either three crops of sweet sorghum or with an alternative crop (Koppen et al., 2009; Reddy et al., 2005; 2007). Even though sorghum is predominately self-pollinating, hybrids and crosses can be produced using male-sterile plants as the maternal parent. Sugar content in the juice increases with maturity and is low prior to seed development. In all varieties, the primary carbohydrate is sucrose, with variable amounts of reducing sugars and starch (Billa et al., 1997; Clark, 1981).

4.3.2

Production

Sweet sorghum is of interest as a dedicated agricultural energy crop because of its drought tolerance, relatively low input requirements, and ability to produce high yields under a wide range of environmental conditions (Bennett and Anex, 2009; Buxton et al., 1999; Grassi et al., 2004; Hunter, 1994; Hunter and Anderson, 1997; Miller and McBee, 1993; Rooney et al., 2007). These traits make sweet sorghum a potentially important feedstock for bioenergy production, mainly in regions where conditions are not favorable for growing starch-rich crops such as maize. The great advantage of sweet sorghum is that it can become dormant under adverse conditions and can resume growth after relatively severe drought, which has implications for crop management. Shoot removal lowers its capacity to endure drought. Early drought stops growth before floral initiation and the plant remains vegetative; it will resume leaf production and flower when conditions again become favorable for growth. Late drought stops leaf development but not floral initiation (Wilson and Whiteman, 1965; Whiteman and Wilson, 1965). Sweet sorghum can be grown in a wide variety of soil types, but yields are normally highest in deep, well-drained soils with good fertility. Sorghum grown in shallow soils or soils with very low organic matter may be more prone to drought stress. Although sorghum is more tolerant of drought stress than many other crops, ample moisture during the growing season is important for good yields of stalks and juice. Its deep root system can extract water from deep sources. Its optimal soil pH range lies between 5.0 and 8.5. Nitrogen usually has the greatest

P1: SFK/UKS P2: SFK c04 BLBS075-Hood December 15, 2010

20:50

Trim: 244mm×172mm

4

Dedicated Herbaceous Energy Crops

93

impact on yields and will likely be needed on most soils, determined by soil type and rainfall. Lime may be applied to soils with a pH below 6.0 to correct soil acidity. A basic application of NPK may be required, and the crop usually responds well to supplementary applications of nitrogen during growth, although fallowed black clay may not need fertilizer. Rotation with a leguminous crop can provide a low-cost fertility build-up, for example, use of gum arabic (Acacia senegal). Asher and Cowie (1974) showed that the consequence of nitrogen deficiency on grain yield is greatest when the deficiency occurs early in the growing season. Low grain protein results when nitrogen deficiency occurs between anthesis and maturity. In some areas of California, accumulation of salt in the soil is sufficiently high to inhibit germination of seed, and young plants are injured. The tolerance of sweet sorghum to salt, however, appears to be relatively high after the plants become well established. Sweet sorghum is typically seeded in widely spaced rows (75–100 cm) using a corn planter. The ideal seeding rate for most sweet sorghum varieties is 3–4 seeds per foot of row with a final stand of 2–3 plants per foot of row. If plant populations are too high, the canes will be spindly and contain less juice than an equal tonnage of larger diameter canes. Currently, the only commercially viable harvest method for sweet sorghum is removing the entire crop with a forage harvester and transporting it to a mill/pretreatment/ethanol facility. Sweet sorghum has found limited use because of poor post-harvest storage characteristics and short harvest window in cooler climates. A study by Bennett and Anex (2009) indicates that fermentable carbohydrates can be produced at less expense from sweet sorghum than from corn grain. Further results published by same authors on costs associated with off-farm transportation, storage, or capital costs associated with milling and energy recovery equipment that are required to provide fermentable carbohydrates suitable for biological conversion re-evaluates sweet sorghum as a biocommodity feedstock. Hunter and Anderson (1997) indicate that the sugar produced in sweet sorghum has the potential to yield up to 8000 L of ethanol per hectare or about twice the ethanol yield potential of maize grain and 30% greater than the average Brazilian sugarcane productivity of 6000 L/ha (Luhnow and Samor, 2006). Field trials in 2009 tested field–based, mobile extractors, a method that could increase commercial viability and rural development (P. Nelson, personal communication). Roughly 4000 cultivars of sweet sorghum are distributed throughout the world (Grassi et al., 2004), providing a diverse genetic base from which to develop regionally specific, highly productive cultivars. In addition to producing large amounts of sugar-rich biomass, hybrids can be developed from crosses between grain-type seed parents and sweet-type pollen parents (Hunter and Anderson, 1997). The product of these crosses typically increases biomass yields and sugar content when compared to the original grain-type seed parents. Such hybrids can co-produce grain at levels approaching the yields of the grain-type seed parent (Miller and McBee, 1993). The co-produced, protein-rich grain can be consumed as food, animal feed, or converted to bioproducts like ethanol (Hong-Tu and Xiu-Ying, 1986; Hunter, 1994; Hunter and Anderson, 1997; Rajvanshi and Nimbkar, 2004). Proper variety selection will play a large role in the success of sweet sorghum production for ethanol. The ideal variety for a particular location should produce high yields with minimal inputs, have a high percentage of high quality and easily extractable juice, be disease and insect tolerant, and tolerate both drought and wet conditions. Undersander et al. (1990) showed that different varieties of sweet sorghum differ in terms of total biomass, brix (the dissolved sugar-to-water mass ratio of a liquid), fermentable carbohydrate, and ethanol yield (Table 4.2).

P1: SFK/UKS P2: SFK c04 BLBS075-Hood December 15, 2010

20:50

Trim: 244mm×172mm

Plant Biomass Conversion

94

Table 4.2. Comparison of different varieties of sweet sorghum and corn.

Total Dry Matter (tons/acre)

Varieties

Percentage Stalk

Stalk Moisture (%)

Brix

Fermentable Carbohydrate Yield (ton/acre)

Calculated Ethanol Yield (gal/acre)

Stalk Lodging (%)

—

—

332

15

1.19

161

13 98

Corn Dekalb 524

8.2

—

46

Cargill MorCane

8.6

40

76

Sorghum

Dale

8.9

10.0

70

74

12.3

3.00

408

9.2

41

73

10.0

1.25

171

8

Keller

10.1

70

72

13.4

2.96

403

97

M81E

96

DeKail FS5

10.0

66

73

12.7

2.83

385

NC+940

9.2

51

70

13.2

1.83

248

15

Northrup King 301

9.4

52

68

13.2

1.81

247

47

Northrup King 405

11.0

60

70

7.3

1.25

170

70

Northrup King 8361

13.1

68

73

7.4

1.96

267

93

Rox Orange

10.6

46

75

10.3

1.84

250

33

Source: Undersander et al. (1990).

4.3.3

Potential Yields

Sweet sorghum yields vary considerably depending on the varieties/hybrids that are used, the location (soil, water, climate, pests, and diseases), inputs, and production practices. When considering sweet sorghum for ethanol production, the most important yield components are biomass yield, juice yield, and sugar production per acre. The concentration of soluble sugars in sorghum ranges widely depending upon variety. Biomass yields of sweet sorghums are also variable ranging from 8 to 48 tons per acre and juice content from 65% to 80% of biomass. The combined sugar content of the juice varies between 9 and 15%. Sugar yields vary from 1.6 to 6.9 tons per acre. The bagasse (crushed stalks) and leaves make up the remainder of the wet biomass. The bagasse represents approximately two-thirds of the dry matter and leaves represent the remaining portion. Fermentation of the sugar in the juice yields 400–600 gallons of ethanol per acre (Vermerris et al., 2009). Potential yield per acre in the United States depends on the region where it is grown (Table 4.3). In warmer regions, sweet sorghum yield in terms of total biomass per year and gallons of ethanol per acre is higher when compared with colder regions.

4.3.4

Economic and Environmental Advantages of Sweet Sorghum

Sweet sorghum has generated interest as a feedstock for ethanol production since the 1970s. Juice from sweet sorghum can be converted to ethanol using currently available, conventional fermentation technology. The bagasse that remains after removal of the juice can be used for

P1: SFK/UKS P2: SFK c04 BLBS075-Hood December 15, 2010

20:50

Trim: 244mm×172mm

4

Table 4.3.

Dedicated Herbaceous Energy Crops

95

Production yield per acre in the United States.

Latitudes

Up to 32◦ N

25◦ N54◦ –26◦ N12◦

Average yield biomass (tons/acre) Ethanol production (gallons/ton) Harvests/year Total biomass in tons/year Gallons/acre

36 20 2 73 1,424

36 20 3 109 2,136

Source: Holmseth (2008).

Table 4.4.

Comparison of five ethanol feedstocks.

Parameter

Sugarcane

Forage Sorghum

Sweet Sorghum

Grain Sorghum

Maize

Crop duration (irrigated crops)

12 months

2 × 4 months

3 × 4 months

2 × 4 months

4 months

Total water requirement/ha/year

36,000 m3

8,000 m3

12,000 m3

11,000 m3

8,100 m3

17

8.7

5,950

3,211

Grain yield for food (tons/ha/year)

15

Grain yield for ethanol (tons/ha/year) Green stalk yield for ethanol (tons/ha/year)

80

120

240

Total ethanol output (litres/ha/year)

6,000

6,660

12,000

Source: AgriFuels Limited.

cellulosic ethanol production. Typically, sweet sorghum varieties have low grain yield, but recently varieties with more balanced grain/sugar production have been developed in India and China (Gnansounou et al., 2005; Hong-Tu and Xiu-Ying, 1986; Rajvanshi and Nimbkar, 2004; Reddy et al., 2005; 2007). These varieties can be used as a dual-purpose crop, whereby the grain is harvested for human or animal utilization, the juice for the production of first generation ethanol and cellulosic biomass for the production of second generation ethanol. Alternatively, these varieties can be used as a dedicated bioenergy crop, where both the sugars and the grain are used for ethanol production. Being a water-use efficient crop, sweet sorghum has the potential to be a good alternative feedstock for ethanol production. It is a multipurpose crop that can be cultivated for simultaneous production of food and feed ingredients, syrup, jaggery (human dessert-type treat), ethanol, fodder for animals, organic fertilizer, or for paper manufacturing. It can be grown in soil that is not ideal for other crops. Sweet sorghum is drought resistant and its growing period and water requirement are three times lower than that of sugarcane (Table 4.4). Sweet sorghum is better than sugarcane, forage sorghum, grain sorghum and maize in terms of water usage, grain yield for food, and grain yield and biomass for ethanol production (Table 4.4). It is best suited for ethanol production because of its higher reducing sugar content as compared to other

P1: SFK/UKS P2: SFK c04 BLBS075-Hood December 15, 2010

96

20:50

Trim: 244mm×172mm

Plant Biomass Conversion

sources. These important traits, along with its suitability for mechanized crop production and seed propagation, make it a viable alternative source of raw material for ethanol production. The cost per liter of ethanol production from sweet sorghum grain and juice is lower than that from maize and sugarcane, respectively. The stillage from sweet sorghum is rich in micronutrients and minerals and can be used as fodder or for cogeneration of power. Sweet sorghum costs $1.74 to produce a gallon of ethanol, compared with $2.19 for sugarcane and $2.12 for corn (Koppen et al., 2009). It also compares well on energy balance, with 8 units of energy produced for every unit of energy invested in its cultivation and production, compared with 8.3 units for sugarcane and 1.8 for corn grain. Comparatively, only 0.8 unit of energy is produced from fossil fuel production for every unit invested (Koppen et al., 2009). Additionally, massive amounts of cellulosic ethanol could be produced from already existing sweet sorghum plantations, since the new ethanol manufacturing techniques would use only residual products. This means sweet sorghum could be adapted to biofuel use without requiring the massive carbon-releasing deforestation inextricably linked to other crops, such as cane sugar. Because the grain of sweet sorghum is not used in ethanol production, it does not have any impact on food prices and food security, in contrast to corn and sugar cane. “As an added benefit, dry land farmers in the tropics can get three sweet sorghum crops a year, versus only one for corn and sugarcane” (Table 4.4). It is also easier and cheaper to grow sweet sorghum than other biofuel crops. For example, it requires one-eighth the amount of water compared to sugarcane and about half that required to grow corn. Moreover, sorghum requires only half as much fertilizer as corn. Besides low water requirements, sorghum grows readily in saline or alkaline soils. It can withstand environmental stress and is inexpensive to cultivate. Ethanol produced from sweet sorghum, as from many renewable sources, is carbon neutral. The carbon dioxide fixed during the growing cycle offsets the carbon dioxide produced during crop production, processing, and ethanol utilization. And, since sweet sorghum thrives on marginal lands, the need to clear rainforests or use productive cropland is reduced.

4.3.5

Production Challenges

The costs associated with transportation of the crop to a mill will be the major limiting factor for where sweet sorghum can be profitably grown. Varieties that have higher sugar contents per ton of biomass will be more efficient to process and haul to the mill. Currently, there are a limited number of varieties for which seed is commercially available. If sweet sorghum is widely and rapidly adopted as an energy crop, seed may become difficult to obtain. Disease and insect problems may also limit yield potentials, suggesting that further research in this area is essential. Leaf diseases are the most troublesome for forage producers. These are anthracnose caused by Colletotrichum graminicola (which can be overcome by using resistant varieties) and leaf blight caused by Helminthosporium turcicum. Charcoal rot (Macrophomina phaseoli) causes plants to lodge badly. Grain may be affected by covered smut (Sphacelotheca sorghi) in which the seed is replaced by a sac of spores; fungicidal seed dressing before planting corrects this latter malady (Koppen et al., 2009. From a forage point of view, grasshoppers appear to be the worst pest, and feral pigs can damage the crop in some locations. Grain pests include the sorghum midge, Contarinia sorghicola, whose larvae feed on the developing seeds. Bird damage is also important with the weaver bird, Quelea quelea, causing major losses in Africa. Damage can be prevented by using awned varieties of sorghum, giving some hope of reducing losses. The high tannin

P1: SFK/UKS P2: SFK c04 BLBS075-Hood December 15, 2010

20:50

Trim: 244mm×172mm

4

Dedicated Herbaceous Energy Crops

97

content of sweet sorghum seed is another deterrent, and early harvesting for silage avoids the main problems.

4.4 Switchgrass Switchgrass (Panicum virgatum L.) from the Poaceae/Gramineae is a perennial warm-season C4 grass indigenous to the Central and North American tall-grass prairie. It is also called prairie switchgrass, tall panic grass, water panicum, wild redtop, or thatchgrass. Switchgrass is an immense biomass producer and can reach heights of 3 m (10 feet) or more in wetter areas of the country. Switchgrass is found from central Mexico to Quebec and ecotypes have been identified in regions from the Atlantic coast to the eastern Rocky Mountain area. This grass was an important part of the North American tall-grass prairie and was adapted to other regions by European settlers. Perennial grasses like switchgrass have been widely used for forage purposes in their native state prior to being established as a crop. Since the 1940s, switchgrass has been used for conservation and warm-season pasture purposes in the Great Plains and Midwest states (Stubbendieck et al., 1992; Sanderson et al., 1996; Casler et al., 2004). The different ecotypes tolerate a large variety of growing conditions, from arid sites in the shortgrass prairie to brackish marshes and open woods (Gunderson et al., 2008). Two genetically and phenotypically distinct ecotypes of switchgrass are recognized: (1) the lowland ecotypes, vigorous, tall, thick-stemmed, and generally found in wetter, and more southern habitats and (2) the upland form, typically shorter, rhizomatous, and fine-stemmed, mainly found in drier mid- and northern latitudes (Casler et al., 2004; Hultquist et al., 1997; Sanderson et al., 1996). In general, ecotypical differences are related to local soil and climatic characteristics, with eastern and southern varieties adapted to higher moisture conditions, and western and northern varieties adapted to drier conditions.

4.4.1

Physiology

Switchgrass usually grows in large clumps that can be as much as 3–10 feet (0.9–3 m) tall. The stem cross section is round and reddish. The leaf blade is flat and sturdy and can measure as much as 30 inches (76.2 cm) long and 0.5 inch (1.3 cm) wide. The inconspicuous flowers have reddish-purple anthers. Flowers are borne in open 10-inch (25.4 cm) panicles and produce shiny seeds measuring around 1/8 inch (0.3 cm) long and shaped like teardrops. Switchgrass growing on Midwestern prairies and bottomland sites develops long rhizomes that grow horizontally and interlace to form a thick sod. In the southeast and on upland sites, switchgrass grows as a bunchgrass, with roots reaching depths of more than 10 feet (3.1 m). Switchgrass is a warm season perennial. The underground stems or rhizomes grow actively from late winter through mid-spring, but the top stays dormant until the soil warms up. Switchgrass typically turns a pale yellow in the fall, but some cultivars have been selected for color, such as “Shenandoah,” for its striking burgundy foliage or “Dallas Blues” and “Heavy Metal” exhibiting a bluish color. The “Rehbraun” cultivar has foliage of 3–4 feet (0.9–1.2 m) that turns red in the late summer and “Haense Herms” has very characteristic pink flowers and dark red autumn leaves. Switchgrass growth is optimum in deep sandy loams but is able to grow on a wide variety of soils; it tolerates moderate soil salinity and pH levels ranging from about 4.5 to 7.6. Switchgrass grows best in association with site-adapted mycorrhizal fungi, can be mowed or grazed down

P1: SFK/UKS P2: SFK c04 BLBS075-Hood December 15, 2010

98

20:50

Trim: 244mm×172mm

Plant Biomass Conversion

in the winter, and does best when burned occasionally. Switchgrass can be found in dry, mesic and wet habitats, can tolerate temperatures as low as −34◦ C (−30◦ F), and will grow in partial shade. Most research on switchgrass fertility has focused on its use as a forage, and higher nitrogen applications can ensure high yields and better quality feed. Nonetheless, some researchers have considered nitrogen fertilizer recommendations for switchgrass to be higher than necessary for biomass production. Switchgrass has a remarkable ability to extract nitrogen from unfertilized soils, and as Parrish and Fike (2005) reported, fields were harvested for seven years with no fertilizer applications, and averaged only 53 pounds of nitrogen removed per year with one harvest per year. Clearly this crop has the genetic ability to survive and produce with minimal if not zero inputs (Duever, 2003; Vogel, 2004). Switchgrass should be harvested with conventional haying equipment from mid- to late October in most regions. Parrish and Fike (2005) demonstrated that a single harvest of switchgrass from late fall or early winter will result in the highest sustainable biomass yields.

4.4.2

Switchgrass Ecotypes

As switchgrass evolved across North America, different ecotypes as well as variations within an ecotype (reproductive phenology and winter-hardiness) have emerged with specific genetic and morphological characteristics that provide a good “fit” to a particular region. More southern cultivars tend to have longer season growth, flowering later in the season and allowing more time for vegetative growth and higher dry matter yield. Through natural selection, two major types have emerged: upland and lowland varieties. The upland types fare better in semi-arid climates and prefer drier soils. The lowland varieties are found in wetter and heavier soils, favoring lands where water availability is more reliable. Lowland varieties are genetically more able to produce dry matter than the upland cultivars. They have lower winter-hardiness, which may result in low winter-survival when grown too far north (Parrish and Fike, 2005). Productivity of upland ecotypes tends to exceed that of lowland ecotypes at very high latitudes and be reduced at southern locations, yet yields from upland strains adapted to relatively more southern locations may be sustained at low latitudes. Similarly, yield and winter survival declines at high latitude sites, especially for southern strains, whether of lowland (Casler et al., 2004) or upland morphology (Berdahla et al., 2005; Casler and Boe, 2003; Casler et al., 2004). Research studies have determined that in order to increase the survivability and productivity of a switchgrass, stand selection must be based on location. Parrish and Fike (2005) have found a “strong correlation between latitude of origin and yield,” and “the main factor determining adaptation of a cultivar was its latitude of origin, with southern cultivars having higher yield potentials as they are moved north.” Switchgrass varieties should therefore be chosen based upon both latitude of origin and ecotype. The cytotypic diversity existing in switchgrass explains the high degree of phenotypic and ecotypic variation. Lowland switchgrass ecotypes are predominately tetraploid, with a base chromosome number of 9 (diploid number 18), leading to 36 chromosomes (2n = 4x = 36). Upland ecotypes are octoploid (2n = 8x = 72) and less frequently, hexaploid (2n = 6x = 54) (Gunter et al., 1996; Hopkins et al., 1995).

4.4.3

Advantages

Several benefits associated with the perennial nature of switchgrass have been identified, such as less intensive agricultural management practices, reduced energy and agrochemical

P1: SFK/UKS P2: SFK c04 BLBS075-Hood December 15, 2010

20:50

Trim: 244mm×172mm

4

Dedicated Herbaceous Energy Crops

99

consumption, and positive effects on soil and wildlife quality (Dunn et al., 1993). Parrish and Fike (2005) suggest that switchgrass can be grown on soils of moderate fertility without fertilizing, or with limited additions of fertilizer, and still maintain productivity (Rinehart, 2006). Growing switchgrass could have a positive effect on erosion control, contributing to soil stabilization on strip-mine spoils, sand dunes, dikes, and other critical areas. Hohenstein and Wright (1994) estimated a 95% reduction in soil erosion rates and a 90% reduction in pesticide use for herbaceous energy crops such as switchgrass relative to annual row crops like corn and soybean. It is also suitable for short windbreak plantings in truck farm fields (Fike et al., 2006), and it has been shown to improve water and soil quality by reducing carbon emissions through carbon sequestration (Bouton et al., 1998; Duffy and Nanhou, 2002). Switchgrass has an extensive and deep root system that, added to the large amount of above-ground biomass, can increase carbon storage. It can be easily integrated into existing farming operations because conventional equipment for seeding, crop management, and harvesting can be used (Vogel et al., 2002). Thanks to its heavy growth during late spring and early summer, switchgrass provides good warm-season pasture and high quality hay for livestock. Its benefits can be extended to the wildlife surrounding it, by providing nesting and fall and winter cover for pheasants, quail, and rabbits, while its seeds provide food for birds. It is resistant to heavy snow (particularly “Shelter” and “Kanlow” cultivars) and is useful on shooting preserves (USDA NRCS Plant Fact Sheet, 2001). Switchgrass has also been tested for the ability to remediate contaminated soils. Entry and Watrud (1998) tested soils contaminated with cesium-137 and strontium-90 using Alamo switchgrass. These elements are radionuclides released during nuclear testing, nuclear reactor accidents, and weapons production. The authors reported a 36% removal of cesium and a 44% removal of strontium over a 5-month period.

4.4.4

Disadvantages

Switchgrass seeds can be very dormant, and a dormancy-breaking step is necessary in the planting process for the majority of seeds to germinate. This is thought to be one reason so many switchgrass and other native warm-season plantings fail. Dormancy-breaking can be achieved by stratification (humidity and low temperatures), drill planting in winter, or early spring or “after-ripening,” which practices storing seeds in a warm environment for several years. The establishment of switchgrass is faced with competition from weeds. Planting switchgrass into cool soils can be problematic; cool season weeds can germinate first and choke out switchgrass seedlings when the soil warms (Vogel, 2004). Perennial forbs and warm-season grasses such as crabgrass germinate in cooler soils and can have a severe negative impact on switchgrass stand establishment. This native grass may also become weedy or invasive in some regions or habitats and may displace desirable vegetation if not properly managed (NRCS, 2001). Switchgrass is generally planted as a monoculture, either as a forage crop or a feedstock. Monocultural production is thought to be problematic by many farmers and more generally by advocates of a more sustainable agriculture because they are not as resilient as polycultures. Biodiversity provides a better use of soil and water resources as well as food and cover for numerous beneficial organisms, from microbes to small mammals. A monoculture of switchgrass will never compare to the benefits obtained with a naturally diverse prairie. Nonetheless, monocultures of switchgrass can have significant benefits, especially for fields suffering from erosion and depleted soil organic matter (Rinehart, 2006; Vogel, 2004), as well as for consistent yield for fuel production.

P1: SFK/UKS P2: SFK c04 BLBS075-Hood December 15, 2010

100

20:50

Trim: 244mm×172mm

Plant Biomass Conversion

Table 4.5. Example of biomass yields from several switchgrass cultivars in the USA (Sanderson and Adler, 2008). Yields are expressed in tons dry biomass per ha. State Cultivar Cave-in-Rock Dacotah Forestburg Shawnee Sunburst Trailblazer Alamo Pathfinder Shelter Kanlow Blackwell NJ50 Summer BoMaster Performer

ND

SD

WI

4.9 5.4

3.8 2.9 3.9 5.1 4.6 4.6

14.3 7.4 9.4 11.4 11.5 11.0

5.6 7.4 6.9

IA

Mid-West

PA

Mid-Atlantic

TX

AL

9.3

9.2

8.2–8.6

10.8–15.4

2.6

6.9 8.8 6.8 7.9 12.1

8.8 15.2–16.3

15.3

12.9–23.0

10.3–13.6 15.0–16.4

11.0

11.6–18.2

NC

12.4

8.5 6.7–12.4 8.3

14.2

11.0 12.1

9.1 12.6 5.5 15.8 12.8

Another potential disadvantage of growing switchgrass is that it is slow to establish and it can take 2–3 years for productive stands to develop (Sanderson and Adler, 2008).

4.4.5

Yields

Unlike other crops such as corn, where yield data have been available for a many years, data for switchgrass as a biomass crop are rather limited and are based mainly on smallplot research. A search across the literature provides switchgrass yield estimates that vary considerably, from less than 1 ton/ha to almost 40 ton/ha. Nonetheless, the most frequently observed yield class across all ecotypes, cultivars, soils, and management practices is between 10 and 12 ton/ha (Table 4.5). This great variability in yields is explained by the wide range of ecotypes as well as the strong interactions between genotypes and the environment. Cultivar × environment interactions must be taken into account in biomass evaluations of switchgrass cultivars (Casler and Boe, 2003; Fuentes and Taliaferro, 2002). Gunderson et al. (2008) observed higher yields on average within the lowland cultivars, 13.11 ton/ha versus 9.30 ton/ha in upland cultivars with Alamo and Kanlow being the highest yielding cultivars among the lowland varieties and Cave-in-Rock from the upland ecotype (McLaughlin et al., 1999). With newer varieties of switchgrass, yields in excess of 20 ton/ha have been reported for test plots. For example, Sanderson et al. (1996) reported 15–20 ton/ha in field trials in Texas, and Thomason et al. (2004) reported yields in excess of 30 ton/ha from field work in central Oklahoma. These yields are relatively high and the site-specificity of the test plots needs to be taken into account. In a larger scale study, Parrish and Fike (2005) reported average biomass yield in a 10-year study of 14.2 ton/ha. Schmer et al. (2008) reported on-farm yields ranging from 5.2 to 11.1 ton/ha for field trials in the United States. Being a perennial, switchgrass grown for biomass can be harvested only once per year, in the winter. Under good management, a producer can expect high yields over much of the country. According to the Agricultural Research Service, growers can expect from 7 to 16

P1: SFK/UKS P2: SFK c04 BLBS075-Hood December 15, 2010

20:50

Trim: 244mm×172mm

4

Dedicated Herbaceous Energy Crops

101

tons an acre (17.3–39.5 ton/ha) in the Southeast and the western Corn Belt, to 1–4 tons (2.5– 10 ton/ha) in the northern plains (Comis, 2006).

4.4.6

Switchgrass as a Bioenergy Crop

Switchgrass has been chosen as one of the best crops for biofuels in the United States. This crop not only grows well under a wide range of conditions, but it also produces high yields over much of the country (Comis, 2006). Traditionally, switchgrass has been managed primarily as a hay crop and bred mainly to improve its value as a forage crop for livestock. It has been typically bred for high nutrient content and high leaf to stem ratio. However, the traits necessary and relevant for biofuels crops are different. High cellulose and low ash content are important for energy conversion and low contamination of combustion systems (McLaughlin et al., 1999). From the late 1980s to early 1990s, the Department of Energy initiated research on plant species with high yield quality and quantity for biofuels. They assessed 34 species of plants with trials being conducted on a wide range of soil types in 31 different sites spread over seven states. Switchgrass became a major focus of research for bioenergy production among some other grasses and legumes. In their trials, they found switchgrass to have the potential for high yields, and they were interested in its deep rooting characteristics, the potential value in carbon sequestration and the expected positive environmental impact (Wright, 2007). Many subsequent studies have focused on switchgrass as a bioenergy crop. Ethanol yields obtained from a mature bioenergy crop production system were calculated by Morrow et al. (Morrow et al., 2006) to be between 330 and 380 L of ethanol per ton of dry switchgrass. The National Renewable Energy Laboratory’s theorical yield calculator provides similar estimates: (http://www1.eere.energy.gov/biomass/ethanol yield calculator.html). Parrish and Fike (2005) obtained similar results by calculating the theoretical ethanol production from switchgrass yielding 15 ton/ha to be between 5000 and 6000 L/ha. A model was developed by David Bransby at Auburn University to estimate delivery costs of switchgrass. After 6 years, they produced up to 15 tons of dry biomass per acre (37 ton/ha), with an average of 11.5 tons per acre (28.4 ton/ha) (Bransby, 2005). With a production of roughly 100 gallons of ethanol per ton of feedstock and the yields obtained in this model, there will be enough to make 1,150 gallons of ethanol per acre each year. US cellulosic fuel production costs are now estimated at more than $2.25 per gallon, compared with $1.03–1.65 per gallon for corn ethanol (Coyle, 2007; Goldemberg, 2007). Switchgrass feedstock costs per gallon of ethanol produced would need to be low enough to at least compete with corn ethanol to make cellulosic ethanol production a cost-effective fuel. However, the market for this crop is still immature and work is currently under development to better understand the costs of conversion as well as the cost to develop processing plants and marketing outlets (Rinehart, 2006).

4.5

Conclusions and Future Prospects

Growing crops specifically for energy is a concept developed in the past few decades with the hope of developing renewable energy sources that will mitigate the effect of soaring energy costs, reduce the global carbon footprint, and reduce dependency on fossil fuel. Miscanthus possesses many qualities of an ideal biomass crop and has a great potential as a dedicated energy

P1: SFK/UKS P2: SFK c04 BLBS075-Hood December 15, 2010

102

20:50

Trim: 244mm×172mm

Plant Biomass Conversion

40 Portugal

30 20 10 0 1

5

7

Yield (t DM ha−1)

40 Early

13

Denmark

Late

30 20 10 0 1

5

7

13

Genotype

Figure 4.2. The effect of climate and genotype on yield. Genotype 1, M. giganteus; 5, M. sacchariflorus; 7 & 10, M. sinensis hybrids; 11&13 , M. sinensis collections. (Adapted from Clfton-Brown et al. (2001).)

crop. Tapping into Miscanthus as a source of renewable energy is a relatively new concept with much room for improvement. The yield reported from field trials in Europe and N. America was achieved with an unimproved Miscanthus. Therefore, yield would be expected to increase dramatically and the cost of production expected to fall with breeding improvement efforts. The two major concerns with cultivating Miscanthus have been the high cost of establishment and currently the narrow genetic base. Future development can address these concerns through crop improvement and technology advancement. Efforts have already been undertaken by the European Miscanthus Improvement (EMI) project (www.biomatnet.org/secure/Fair/F659.htm) and research institutes across N. America to broaden the genetic base of Miscanthus and maximize the productivity and adaptive range of the crop through traditional breeding as well as modern genetic engineering. The efforts of the EMI project have demonstrated that different hybrids of Miscanthus sinensis can be found for a wide range of climatic conditions in Europe. One particular hybrid combines winter hardiness with high biomass potential (Figure 4.2), a combination that will result in improved crop quality through delayed harvest without significant loss in yield (Clifton-Brown et al., 2001b). Improved farming equipment and farming practices tailored specifically to working with Miscanthus rhizomes and plants could reduce loss of harvested material, time, and labor costs. The development of new machinery to process rhizomes (that include lifting, cleaning, splitting, sorting, and boxing) and carry out precision rhizome planting has made commercial rhizome multiplication farming and commercial production of Miscanthus feasible. Improvement in the crop together with improvement in farming technology and processing of this biomass crop into fuel would undoubtedly make

P1: SFK/UKS P2: SFK c04 BLBS075-Hood December 15, 2010

20:50

Trim: 244mm×172mm

4

Dedicated Herbaceous Energy Crops

103

Sweet sorghum

Grain

Sugar juice

Biomass (bagasse & leaves) Pretreatment & cellulase treatment

Fermentation

Methane

Ethanol

Waste water treatment

Figure 4.3. Sweet sorghum energy model.

Miscanthus a very attractive crop to the biofuel industries and at the same time helping the country meets its bioenergy goals. High-biomass sweet sorghum will play an important role as the United States moves toward alternative energy sources, particularly biomass, to supply transportation and electrical energy to the country. Compared to fossil fuels, first- and second-generation ethanol from sweet sorghum holds considerable potential to help lower fossil energy use and reduce greenhouse gas emissions since most expenditures are more than compensated by credits for bioethanol and the respective by-products (Figure 4.3). Even with the use of grain as food, there is significant potential for saving fossil energy and greenhouse gas emissions with the help of first-generation ethanol produced from juice. This makes sweet sorghum an ideal crop for reducing the competition between food and fuel. If both grains and juice are used as food, energy and greenhouse gas balances are neutral. This means that under-maximized food production in the form of grains and sugar, all expenditures occurring during food production can be compensated by ethanol production from bagasse. Hence, limited negative environmental impacts occur regarding fossil energy savings and greenhouse effects. Therefore, from a climate protection point of view, efforts should be put on optimized crop varieties and cultivation methods aiming at higher total yields. Switchgrass has been considered a model for bioenergy crops mainly due to its perennial growth habit, the compatibility of its farming practices with conventional crops and the high value in improving soil conservation and quality. More importantly, switchgrass’ potential for high yields has made this warm season prairie grass of national interest in the biofuels strategy. Trials across North America have identified three distinctive high-yielding cultivars, the lowland varieties “Alamo” and “Kanlow” and “Cave-in-Rock” for the upper Midwest. Yields for switchgrass can be as high as 40 ton/ha, but the best adapted cultivars have averaged approximately 16 Mg/ha in research plots. Even though several constrains have been identified in the establishment of switchgrass, the benefits associated with the culture of this remarkably adaptable grass make it ideal as an energy crop. The US Department of Energy believes crops grown for biofuels such as switchgrass could reduce North America’s

P1: SFK/UKS P2: SFK c04 BLBS075-Hood December 15, 2010

104

20:50

Trim: 244mm×172mm

Plant Biomass Conversion

dependence on foreign oil as well as carbon dioxide emissions. Nonetheless, further research is fundamental to increase the overall production potential of this perennial grass. Studies on boosting its hardiness and yields will be necessary to improve switchgrass as a cultivated, production crop.

References Asher, C. J. & Cowie, A. M. 1974. Grain sorghum: high yield, satisfactory protein content of both? Proceedingsof Agronomy Society of New Zealand, 4, 79–82. Bennett, A. S. & Anex, R. P. 2009. Production, transportation and milling costs of sweet sorghum as a feedstock for centralized bioethanol production in the upper Midwest. Bioresource Technology, 100, 1595–1607. Berdahla, J., Franka, A., Krupinskya, J., Carrb, P., Hansona, J. & Johnsona, H. 2005. Biomass yield, phenology, and survival of diverse switchgrass cultivars and experimental strains in western North Dakota. Agronomy Journal, 97, 549–555. Billa, E., Koullas, D. P., Monties, B. & Koukios, E. G. 1997. Structure and composition of sweet sorghum stalk components. Industrial Crops and Products, 6, 297–302. Bouton, J., Bransby, D., Conger, B., McLaughlin, S., Ocumpaugh, W., Parrish, D., Taliaferro, C., Vogel, K. & Wullschleger, S. 1998. Developing switchgrass as a bioenergy crop. Presented at Fourth National New Crops Symposium, Phoenix, AZ. Bransby, D. 2005. Switchgrass Profile [Online]. Oak Ridge National Laboratory: Bioenergy Feedstock Development Program. Available: http://bioenergy.ornl.gov/papers/ misc/switchgrass-profile.html. (accessed November 30, 2010) Bullard, M. J., Nixon, P., Kilpatrickm, J., Heath, M. & Speller, C. 1995. Principles of weed control in Miscanthus spp. under contrasting field conditions. Presented at British Crop Portection Conference, Brighton, UK. Buxton, D. R., Anderson, I. C. & Hallam, A. 1999. Performance of sweet and forage sorghum grown continuously, double-cropped with winter rye, or in rotation with soybean and maize. Agronomy Journal, 91, 93–101. Casler, M. D. & Boe, A. R. 2003. Cultivar × environment interactions in switchgrass. Crop Science, 43, 2226–2233. Casler, M. D., Vogel, K. P., Taliaferro, C. M. & Wynia, R. L. 2004. Latitudinal adaptation of switchgrass populations. Crop Science, 44, 293–303. Christian, D. G. & Haase, E. 2001. Agronomy of Miscanthus. In: Jones, M. B. & Walsh, M. (eds.) Miscanthus for Energy and Fibre. London: James & James Ltd. Clark, J. 1981. Inheritance of fermentable carbohydrates in stems of sorghum bicolor Moench, PhD Dissertation, Texas A&M University. Clifton-Brown, J., Lewandowski, I., Stampfl, P. & Jones, M. B. 2002. Modelled biomass production potential of Miscanthus and actual harvestable yield as influenced by harvest time. Presented at 12th European Biomass Conference and Exhibition, Amsterdam. Clifton-Brown, J., Long, S. P. & Jorgensen, U. 2001a. Miscanthus Productivity. London: James & James Ltd. Clifton-Brown, J. C., Lewandowski, I., Andersson, B., Basch, G., Christian, D. G., Kjeldsen, J. B., Jorgensen, U., Mortensen, J. V., Riche, A. B., Schwarz, K.-U., Tayebi, K. & Teixeira, F. 2001b. Performance of 15 Miscanthus genotypes at five sites in Europe. Agronomy Journal, 93, 1013–1019.

P1: SFK/UKS P2: SFK c04 BLBS075-Hood December 15, 2010

20:50

Trim: 244mm×172mm

4

Dedicated Herbaceous Energy Crops

105

Comis, D. 2006. Switching to Switchgrass Makes Sense. USDA-ARS. Available: http://www.ars.usda.gov/is/AR/archive/jul06/grass0706.htm?pf=1. Coyle, W. 2007. The Future of Biofuels. Amber Waves, 5. Duever, L. C. 2003. Floridata: Panicum virgatum [Online]. Available: http://www.floridata. com/ref/p/pani vir.cfm (accessed October 25,2010). Duffy, M. & Nanhou, V. 2002. Costs of producing switchgrass for biomass in Southern Iowa. In: Janick, J. & Whipkey, A. (eds.) Trends in New Crops and New Uses. Alexandria, VA: ASHS Press. Dunn, C., Stearns, F., Guntenspergen, G. R. & Sharpe, D. M. 1993. Ecological benefits of the conservation reserve program. Conservation Biology, 7, 132–139. El Bassam, N. & Huisman, W. 2001. Harvesting and storage of Miscanthus. In: Jones, M. B. & Walsh, M. (eds.) Miscanthus for energy and fibre. London: James & James Ltd. Entry, J. A. & Watrud, L. S. 1998. Potential remediation of 137Cs and 90 Sr contaminated soil by accumulation in Alamo switchgrass. Water, Air, & Soil Pollution, 104, 339–352. Fike, J. H., Parrish, D. J., Wolf, D. D., Balasko, J. A., Green, J. J. T., Rasnake, M. & Reynolds, J. H. 2006. Long-term yield potential of switchgrass-for-biofuel systems. Biomass and Bioenergy, 30, 198–206. Fuentes, R. D. & Taliaferro, C. M. 2002. Biomass yield stability of switchgrass cultivars. In: Janick, J. & Whipkey, A. (eds.) Trends in New Crops and Uses. Alexandria, VA: ASHS Press. Gnansounou, E., Dauriat, A. & Wyman, C. E. 2005. Refining sweet sorghum to ethanol and sugar: Economic trade-offs in the context of North China. Bioresource Technology, 96, 985–1002. Goldemberg, J. 2007. Ethanol for a sustainable energy future. Science, 315, 808–810. Grassi, G., Tondi, G. & Helm, P. 2004. Small-Sized Commercial Bioenergy Technologies as an Instument of Rural Development. Biomass and Agriculture: Sustainability, Markets and Policies. Paris: OECD Publication Service. Greef, J. M. & Deuter, M. 1993. Syntaxonomy of Miscanthus × giganteus. Berlin, Allemagne: Blackwell. Gunderson, C. A., Davis, E., Jager, Y., West, T. O., Perlack, R. D., Brandt, C. C., Wullschleger, S. D., Baskaran, L. M., Wilkerson, E. & Downing, M. E. 2008. Exploring Potential U. S. Switchgrass Production for Lignocellulosic Ethanol. Technical Report ORNL/TM2008/103. Gunter, L. E., Tuskan, G. & Wullschleger, S. 1996. Diversity among populations of switchgrass based on RAPD markers. Crop Science, 36, 1017–1022. Hallam, A., Anderson, I. C. & Buxton, D. R. 2001. Comparative economic analysis of perennial, annual, and intercrops for biomass production. Biomass and Bioenergy, 21, 407– 424. Heaton, E., Voigt, T. & Long, S. P. 2004. A quantitative review comparing the yields of two candidate C4 perennial biomass crops in relation to nitrogen, temperature and water. Biomass and Bioenergy, 27, 21–30. Heaton, E. A., Dohleman, F. G. & Long, S. P. 2008. Meeting US biofuel goals with less land: The potential of Miscanthus. Global Change Biology, 14, 2000–2014. Himken, M., Lammel, J., Neukirchen, D., Czypionka-Krause, U. & Olfs, H. W. 1997. Cultivation of Miscanthus under West European conditions: Seasonal changes in dry matter production, nutrient uptake and remobilization. Plant and Soil, 189, 117–126. Hohenstein, W. & Wright, L. 1994. Biomass energy production in the United States: An overview. Biomass and Bioenergy, 6, 161–173.

P1: SFK/UKS P2: SFK c04 BLBS075-Hood December 15, 2010

106

20:50

Trim: 244mm×172mm

Plant Biomass Conversion

Holmseth, T. C. 2008. Sweet harvest. Ethanol Producer Magazine. Hong-Tu, M. & Xiu-Ying, H. 1986. Breeding for sweet sorghum with high grain yield and sweet stalks. Sorghum Newsletter, 29, 2–3. Hopkins, A. A., Vogel, K. P., Moore, K. J., Johnson, K. D. & Carlson, I. T. 1995. Genotype effects and genotype by environment interactions for traits of elite switchgrass populations. Crop Science, 35, 125–132. Hultquist, S. J., Vogel, K. P., Lee, D. J., Arumuganathan, K. & Kaeppler, S. 1997. DNA content and chloroplast DNA polymorphisms among switchgrasses from remnant midwestern prairies. Crop Science, 37, 595–598. Hunter, E. L. 1994. Development, sugar yield, and ethanol potential of sweet sorghum, Masters Thesis, Iowa State University. Hunter, E. L. & Anderson, I. C. 1997. Sweet sorghum. Horticulture Reviews, 21, 73–104. Jones, M. B. & Walsh, M. 2001. Miscanthus for Energy and Fibre, London: James & James Ltd. Koppen, S., Reinhardt, G. & Gartner, S. 2009. Assessment of Energy and Greenhouse Gas Inventories of Sweet Sorghum for First and Second-Generation Bioethanol. Rome: Food and Agriculture Organization of the United Nations. Lewandowski, I., Clifton-Brown, J. C., Scurlock, J. M. O. & Huisman, W. 2000. Miscanthus: European experience with a novel energy crop. Biomass and Bioenergy, 19, 209–227. Linde-Laursen, I. 1993. Cytogenetic analysis of Miscanthus “Giganteus,” an interspecific hybrid. Hereditas, 119, 297–300. Long, S. P. & Beale, C. V. 2001. Resource capture by Miscanthus. In: Jones, M. B. & Walsh, M. (eds.) Miscanthus for Energy and Fibre. London: James & James Ltd. Luhnow, D. & Samor, G. 2006. It wears off energy imports. Wall Street Journal. Mcbee, G. G., Miller, F. R., Dominy, R. E. & Monk, R. L. 1987. Quality of sorghum biomass for methanogenesis. In: Klass, D. (ed.) Energy from Biomass and Waste X. London: Elsevier. McLaughlin, S., Bouton, J., Bransby, D., Conger, B. & Al, E. 1999. Developing switchgrass as a bioenergy crop. In: Janick, J. (ed.) Perspectives on New Crops and New Uses. Alexandria, VA: ASHS Press. Miller, F. R. & Mcbee, G. G. 1993. Genetics and management of physiological systems of sorghum for biomass production. Biomass & Bioenergy, 5, 41–49. Monteith, J. L. 1978. Reassessment of maximum growth rates for C3 and C4 crops. Experimental Agriculture, 14, 1–5. Morrow, W. R., Griffin, W. M. & Matthews, H. S. 2006. Modeling switchgrass derived cellulosic ethanol distribution in the United States. Environmental Science & Technology, 40, 2877–2886. Naidu, S. L. & Long, S. P. 2004. Potential mechanisms of low-temperature tolerance of C4 photosynthesis in Miscanthus × giganteus: An in vivo analysis. Planta, 220, 145–155. Naidu, S. L., Moose, S. P., Al-Shoaibi, A. K., Raines, C. A. & Long, S. P. 2003. Cold tolerance of C4 photosynthesis in Miscanthus × giganteus: Adaptation in Amounts and Sequence of C4 Photosynthetic Enzymes. Plant Physiol., 132, 1688–1697. Nixon, P., Boocock, H. & Bullard, M. J. 2001. An evaluation of planting options for Miscanthus. Aspects of Applied Biology, 65, 123–130. Nixon, P. & Bullard, M. J. 2003. Optimisation of Miscanthus Harvesting and Storage Stategies [Online]. Available: http://www.berr.gov.uk/files/file14955.pdf (accessed October 25, 2010). NRCS. 2001. Plant Fact Sheet [Online]. USDA. Available: http://plants. usda.gov/factsheet/pdf/fs pavi2.pdf (accessed October 25, 2010).

P1: SFK/UKS P2: SFK c04 BLBS075-Hood December 15, 2010

20:50

Trim: 244mm×172mm

4

Dedicated Herbaceous Energy Crops

107

Parrish, D. J. & Fike, J. H. 2005. The biology and agronomy of switchgrass for biofuels. Critical Reviews in Plant Sciences, 24, 423–459. Perlack, R. D., Wright, L. L., Turhollow, A. F., Graham, R. L., Stokes, B. J. & Erbach, D. C. 2005. Biomass as feedstock for a bioenergy and bioproducts industry: The technical feasibility of a billion-ton annual supply. Technical Report A357634. Rajvanshi, A. K. & Nimbkar, N. 2004. Sweet sorghum R&D at the Nimbkar Agricultural Research Institute. Presented at First European Seminar on Sorghum for Energy and Industry, Toulouse, France, April 1–3. Reddy, B. V. S., Kumar, A. A. & Ramesh, S. 2007. Sweet sorghum: A water saving bioenergy crop. Presented at Linkage Between Energy and Water Management for Agriculture in Developing Countries, ICRISAT, Hyperabad, India. Reddy, B. V. S., Ramesh, S., Reddy, P. S., Ramaiah, B., Salimath, P. & Kachapur, R. 2005. Sweet sorghum—A potential alternate raw material for bio-ethanol and bio-energy. Journal of SAT Agricultural Research, 1, 8. Rinehart, L. 2006. Switchgrass as a Bioenergy Crop [Online]. ATTRA National Sustainable Agriculture Information Service. Available: www.attra.ncat.org/attra-pub/switchgrass.html (accessed October 25, 2010). Rooney, W. L., Blumenthal, J., Bean, B. & Mullet, J. E. 2007. Designing sorghum as a dedicated bioenergy feedstock. Biofuels Bioproducts & Biorefining (Biofpr), 1, 147–157. Sanderson, M. A. & Adler, P. R. 2008. Perennial forages as second-generation bioenergy crops. International Journal of Molecular Sciences, 9, 768–788. Sanderson, M. A., Reed, R. L., McLaughlin, S. B., Wullschleger, S. D., Conger, B. V., Parrish, D. J., Wolf, D. D., Taliaferro, C., Hopkins, A. A., Ocumpaugh, W. R., Hussey, M. A., Read, J. C. & Tischler, C. R. 1996. Switchgrass as a sustainable bioenergy crop. Bioresource Technology, 56, 83–93. Schmer, M. R., Vogel, K. P., Mitchell, R. B. & Perrin, R. K. 2008. Net energy of cellulosic ethanol from switchgrass. Proceedings of the National Academy of Sciences, 105, 464– 469. Stubbendieck, J. L., Hatch, S. L., Butterfield, C. H. & Jansen, B. P. 1992. North American Range Plants. Lincoln, NE: University of Nebraska Press. Thomason, E. W., Raun, R. W., Johnson, V. G., Taliaferro, M. C., Freeman, W. K., Wynn, J. K., & Mullen W. R. 2004. Switchgrass Response to Harvest Frequency and Time and Rate of Applied Nitrogen. Philadelphia, NJ: ETATS-UNIS, Taylor & Francis. Undersander, D. J., Lueschen, W. E., Smith, L. H., Kaminski, A. R., Doll, J. D., Kelling, K. A. & Oplinger, E. S. 1990. Sorghum—For Syrup. University of Wisconsin Cooperative or Extension Service. Vermerris, W., Rainbolt, C., Wright, D. & Newman, Y. 2009. Production of biofuel crops in florida: Sweet sorghum. Institute of Food and Agricultural Sciences, University of Florida. Vogel, K. 2004. Switchgrass. In: Moser, L. E., Sollenberger, L. & Burson, B. (eds.) WarmSeason Grasses. Madison WI: ASA-CSSA-SSSA Monograph. Vogel, K. P., Brejda, J. J., Walters, D. T. & Buxton, D. R. 2002. Switchgrass biomass production in the Midwest USA: Harvest and nitrogen management. Agronomy Journal, 94, 413– 420. Walsh, M. & McCarthy, S. 1998. Miscanthus Handbook. In: 10th European Bioenergy Conference. Wurzburg, Germany: CARMEN Publishers; pp. 1071–1074. Whiteman, P. C. & Wilson, G. L. 1965. Effects of water stress on the reproductive developement of Surghum vulgare. University of Queensland Department of Botany Papers, University of Queensland.

P1: SFK/UKS P2: SFK c04 BLBS075-Hood December 15, 2010

108

20:50

Trim: 244mm×172mm

Plant Biomass Conversion

Wilson, G. L. & Whiteman, P. C. 1965. The Influence of Shoot Removal on Drought Survival of Sorghums. University of Queensland Deptartment of Botany Papers. St. Lucia: University of Queensland. Wright, L. L. 2007. Historical perspective on how and why switchgrass was selected as a “model” high-potential energy crop. Presented at Oak Ridge National Laboratory, Oak Ridge, TN.

P1: SFK/UKS P2: SFK c05 BLBS075-Hood December 15, 2010

20:51

Trim: 244mm×172mm

Chapter 5

Municipal Solid Waste as a Biomass Feedstock David J. Webster

5.1 Introduction This chapter discusses municipal solid waste (MSW) as a source of cellulosic feedstock for production of biofuels. Along with forest wood, solid waste in the United States represents a major potential source of cellulosic feedstock that has today a mature collection infrastructure, cost structure, and regulatory framework. These attributes suggest that solid waste should play an increasingly important role as a primary cellulosic feedstock. MSW has valuable potential as a cellulosic feedstock and has been evaluated by state agencies and private concerns for its use in producing ethanol. Several second-generation biofuel technology applications have targeted MSW as a cellulosic feedstock. MSW differs from wood and agricultural biomass since it is a heterogeneous feedstock that varies by region, particular markets, and by time (seasonal variations and periodic collection practices). MSW also represents an available feedstock that does not require considering the impact of expanded use of agricultural land and resources for production of feedstock. Abundant public material on solid waste management, market issues, and characterization are available. Many materials were prepared by or for regulatory and public health agencies. However, a gap in information remains about potential use, pretreatment, and processing for biofuels production of MSW as a cellulosic feedstock. This chapter focuses on information published by public agencies such as the Office of Solid Waste in the US Environmental Protection Agency (USEPA) and selected regional references. Additional research and investigation, as it applies to specific project or process development, will be required by parties involved in such activities. Such parties will likely use other references targeted to their project area and will conduct, mostly likely, new investigations of the MSW stream being considered for a particular project. Detailed planning would benefit from waste composition studies that quantify variations in composition and quantities in a particular market area. Plant Biomass Conversion, First Edition. Edited by Elizabeth E. Hood, Peter Nelson and Randall Powell.
C 2011 John Wiley & Sons Inc.

109

P1: SFK/UKS P2: SFK c05 BLBS075-Hood December 15, 2010

110

20:51

Trim: 244mm×172mm

Plant Biomass Conversion

5.2 Definitions MSW does not have a rigid definition. However, most characterizations describe MSW as comprising a wide variety of material and products generated for disposal by households, commercial establishments, and institutions. MSW includes paper and packaging, green waste (e.g., yard trimmings), bulky goods (e.g., furniture), clothing, food waste, and durable goods (i.e., appliances). Household hazardous waste (HHW) is often included as a fraction of the overall MSW stream. HHW typically comprises small amounts of batteries, paint, solvents, petroleum products, pesticides, and other material. Even with HHW present, MSW is not regulated as hazardous waste. Generally excluded from MSW are construction and demolition debris, industrial nonhazardous waste and sludge, agricultural wastes, and municipal wastewater sludge. This chapter is limited to considering MSW in the United States. Examples have been selected that represent proven or promising approaches to addressing the complexities inherent in using MSW as a cellulosic feedstock, but do not include use of methane from landfill gas and anaerobic digestion of MSW. USEPA publishes periodic reports on character and source of MSW in the United States. Municipal Solid Waste in the United States: 2007 Facts and Figures (EPA530-R-08-010) is a primary source of data on solid waste generation (USEPA, 2008).

5.2.1

Second-Generation Conversion Technologies for Biofuels

Conversion technologies are divided into two broad categories by the USDOE Energy Efficiency and Renewable Energy (EERE) Biomass Program: biochemical and thermochemical. Biochemical processes break down cellulosic feedstocks into sugars using enzymes or chemicals. The sugars are then fermented into ethanol. Thermochemical processes use heat and pressure to breakdown biomass (cellulosic and noncellulosic fractions) into intermediate compounds. The intermediate compounds are then converted to fuel or chemicals using catalysts, heat, and pressure (http://www1.eere.energy.gov/biomass/biochemical conversion.html). The Biomass Program provides funding and support for “integrated biorefineries.” Biorefineries use biomass feedstock for production of fuel, chemicals, heat, and power. Integrated biorefineries may use combinations of feedstocks and conversion technologies to target production of value-added products in addition to ethanol. Value-added products can include chemicals, lignin, waxes, and heat and power (http://www1.eere.energy.gov/biomass/integrated biorefineries.html). The Biomass Program is currently supporting several demonstration projects for individual processes and for integrated biorefineries. Bluefire Renewables and Enerkem are two companies supported by the Program that intends to use solid waste as a feedstock and other companies are actively engaged in pilot and demonstration projects designed for solid waste conversion into biofuels and other products.

5.3 Disposal Infrastructure and Transfer Stations Infrastructure used in waste management in the United States may be divided into collection, transportation, recovery (for recycling), and disposal. These practices vary according to

P1: SFK/UKS P2: SFK c05 BLBS075-Hood December 15, 2010

20:51

Trim: 244mm×172mm

5

Municipal Solid Waste as a Biomass Feedstock

111

regional and subregional infrastructure, laws, regulation, and local practice. Generalities can be made at high levels, but local practice driven by applicable law and regulation varies widely across the country. Within the United States, three major handling and disposal systems are used to treat solid waste: recycling/recovery, landfills, and incineration. For the United States as a whole, recovery of materials in MSW is now estimated at 33.4% by the USEPA. Essentially, all remaining material is disposed of in landfills and incinerators. Landfills accept 54.0% of all solid waste either by discharging at the landfill or through transfer stations. Incineration—almost always with some form of energy recovery—accounts for 12.6% of all solid waste disposal. Biofuel production from MSW will likely focus on the portion of waste currently delivered to landfills. The gross amount of MSW available for processing into second-generation biofuel is the fraction of total generated waste that remains after recycling and recovery, delivery to other sources under long-term commitments (such as incinerator/waste-to-energy (WTEs) facilities), and diversion to other facilities (such as compost or biogas production plants). The convertible fraction is further reduced by reduction of water, primarily by drying, since it does not contribute to biofuel production. As the disposal sites for solid waste move farther away from urban areas, transfer stations have experienced an increasingly important role in handling waste and in the cost structure for solid waste management. Transfer stations are employed to consolidate solid waste from collection vehicles and serve to improve logistical efficiency and reduce transportation cost (Figure 5.1). Ownership of waste disposal sites and transfer stations are split between the private sector and the public sector. A profile of the spilt in ownership of waste handling and disposal systems is summarized in Table 5.1 (Repa, 2001).

Figure 5.1.

Transfer trucks queuing for disposal.

P1: SFK/UKS P2: SFK c05 BLBS075-Hood December 15, 2010

112

20:51

Trim: 244mm×172mm

Plant Biomass Conversion

Table 5.1.

Public and private sector ownership of waste facilities (about 2001).

MSW landfills C&D landfills Transfer stations Incineration/WTE MRFs

Private Sector

Public Sector

Percent Publicly Owned (%)

1,100 200 900 70 2,300

2,100 400 2,200 70 1,200

66 67 71 50 34

Rounded numbers. WTE: waste-to-energy; MRFs: material recovery facilities (include mixed waste and source-separated material MRFs) C&D: Construction and demolition waste.

5.3.1

Collection Practices

Collection by municipality or private hauler (sometimes called subscription service) of solid waste is the primary mechanism used to gather waste from the curbside and deliver it to a disposal site or a transfer station. Self-haul is largely limited to rural areas. State and local laws primarily govern collection practice. Like disposal sites and transfer stations, control of solid waste collection is split between the private and public sectors. Municipal control here includes collection by municipally owned and operated vehicles and municipal crews, municipal contracts or franchise arrangements with private collection companies, or a combination of these mechanisms. Repa (2001) estimated the split between public and private organizations in the waste industry using a comprehensive industry survey conducted by R.W. Beck Inc. Approximately 27,000 organizations in the United States are involved in the solid waste industry. Public entities comprised 56% of this group. When municipalities or local authorities control and regulate solid waste services, contracts to collection companies often extend for five years or more. Thus, municipal control of solid waste provides an opportunity for long-term commitments for the MSW feedstock stream. Smaller urban and most rural areas of the United States rely on private collection companies to collect solid waste from households and businesses. The terms of these subscription contracts vary widely but may be characterized as short-term (12 months, for example) agreements between a household and a private company for periodic collection services. Short-term contracts pose a business challenge for a long-term reliable supply of MSW as a feedstock. A much smaller portion of solid waste is handled directly by the residential or commercial waste generator, and reliable figures are lacking. This portion of the market will be omitted from this discussion since it does not appear to comprise a significant portion.

5.3.2

Cost Parameters

MSW is unique as a potential cellulosic feedstock since generators pay to dispose of their waste. Thus, MSW is a “negative cost” feedstock. Like handling any bulk material, costs are incurred in each step of the supply chain—collection, transfer, hauling, and disposal. The range of disposal cost at landfills and WTE plants across the nation vary widely. Simmons et al. (2006) noted in their annual The State of Garbage report that landfill tipping fees range from $18 (Oklahoma) to $98 (Vermont) per ton and WTE tipping fees range from $40 (North Carolina) to $98 (Washington) per ton based on values supplied during their study. These figures are limited to the gate fees charged at the disposal facilities. Local curbside collection,

P1: SFK/UKS P2: SFK c05 BLBS075-Hood December 15, 2010

20:51

Trim: 244mm×172mm

5

Municipal Solid Waste as a Biomass Feedstock

113

transfer (if needed), and hauling to disposal sites are additional. Also, many jurisdictions have incentives to offset the cost of recycling that are either a user surcharge or an additional tax levy associated with the overall cost of solid waste management.

5.4 Waste Generation The USEPA reported that the United States generated 254 million (wet) tons of MSW in 2007 (USEPA, 2008). This waste stream includes residential, commercial, and institutional generators. This tonnage corresponds to a per capita generation rate of 4.62 pounds per person per day (365 days per year). Depending upon local and regional conditions, the portion of residential waste (by weight) ranges from 55% to 65% (USEPA, 2008). Commercial and institutional generators create the remaining portion of between 35% and 45%. Comparison to annual collection of forest-derived biomass is useful for context. Hoekman (2009) discusses the United States biofuel potential and challenges for various feedstocks. He notes that the United States currently utilizes approximately 142 million dry tons per year of forest-derived biomass for fuel, pulp-and-paper, and lumber products. Significantly, Hoekman also points to projections from the US government that indicate 368 million dry tons per year of forest-derived biomass may eventually be available for biofuel production. Using the USEPA (2008) average of 54.0% of total generated waste being landfilled, approximately 137 million wet tons of MSW is potentially available annually for biofuel production (http://www.epa.gov/solidwaste/nonhaz/municipal/pubs/msw07-rpt.pdf). Assuming about one-third of this waste material is water, approximately 92 million dry tons of MSW is available for conversion, or about two-thirds of the forest biomass currently used in the United States.

5.5 Waste Characterization 5.5.1

Composition of Generated MSW Prior to Disposal or Processing

The composition of MSW needs to be examined to understand its potential as a cellulosic feedstock. For second-generation biochemical processes, the primary component in MSW for the production yield of ethanol is the carbohydrate content of solid waste. Sugars derived from the carbohydrates for subsequent fermentation to ethanol are obtained by hydrolyzing carbohydrates by enzymatic or other means, such as dilute or concentrated acid. Thermochemical processes do not rely on carbohydrate content for value-added processing, but such processes will still need to understand waste composition to design for the various material components of MSW. Both process approaches require an understanding of moisture and heavy metals in the waste feedstock stream. Paper and organics in waste are composed of carbohydrates, water, and lignin. Carbohydrates are composed of cellulose and hemicellulose. Cellulose is a stable polymer and a crystalline material made of glucose sugars. The cellulose polymer requires pretreatment to break apart the crystalline structure so glucose can be accessible to the hydrolyzing catalyst. Hemicellulose is a more easily decomposed polymer made of mixed sugars (primarily xylose, glucose, mannose) and organic acids. Hemicellulose is water soluble and noncrystalline. It almost always is the minor portion of the carbohydrate fraction in paper and organic waste.

P1: SFK/UKS P2: SFK c05 BLBS075-Hood December 15, 2010

114

20:51

Trim: 244mm×172mm

Plant Biomass Conversion

Figure 5.2.

MSW in WTE bunker.

Lignin is a phenolic compound that is not convertible to sugar and, in biochemical processes, does not contribute to ethanol production yield. Indeed, in many second-generation biofuels processes, lignin must be specifically separated to facilitate conversion of carbohydrates to sugars. However, like wood and agricultural feedstocks in many second-generation biochemical processes aimed at biofuel production, lignin separated from the MSW feedstock provides a solid fuel that is used to produce thermal power at the processing facility. Gasification processes use the combined syngas streams created from the gasified lignin and carbohydrates in waste feedstock. A majority of the materials included in MSW may be considered good sources of cellulosic feedstock. The fractions of MSW that contain cellulose may be divided into four main groups. The largest component of MSW generated by weight (and prior to recovery efforts) is paper and cardboard. Paper and paperboard represent 32.7% of MSW. The second largest component of MSW is yard trimmings at 12.8%. The third largest fraction of MSW includes food waste amounting to 12.5%. Wood is 5.6% of MSW across the United States. These four material categories represent over half of all MSW generation (63.6%) and are considered good material for cellulose, hemicellulose, and lignin although each will have contaminants and moisture that must be addressed in any process arrangement (Figures 5.2 and 5.3). The remaining fraction of MSW (36.4% by weight) comprises largely material that will contribute little to cellulosic fuels and chemicals. These materials categories include glass, metals, plastics, rubber, leather and textiles, and miscellaneous wastes. The USEPA recently reported the amount of waste generated and recovered nationally as shown in Table 5.2.

P1: SFK/UKS P2: SFK c05 BLBS075-Hood December 15, 2010

20:51

Trim: 244mm×172mm

5

Municipal Solid Waste as a Biomass Feedstock

115

Figure 5.3. Green waste prior to shredding.

5.5.2

Landfilled Waste Compared to Waste Generation

Biofuel production likely will utilize only the portion of waste destined for landfill. Two illustrations help differentiate between gross waste generation and the portion of overall generated waste sent to landfill. The case study by Chester (2009) on the potential ethanol production from exclusively landfilled MSW in California uses an annual estimate that shows about 22 million tons of wet MSW is useable for ethanol production. This is about one-half of the 44 million tons of landfilled waste he obtained from various studies of MSW in the state. After adjusting for moisture content, Chester used about 16 million dry tons per year of convertible solids to calculate a production potential of between 1 and 1.5 billion gallons using California landfilled waste. Likewise, a 2003 study in Pennsylvania conducted by R.W. Beck Inc. illustrates the variation between total waste generation in the USEPA report and waste destined for landfill in Pennsylvania (Table 5.3) (http://www.dep.state.pa.us/dep/deputate/airwaste/wm/recycle/waste comp/1 Intro.pdf). Differences in definitions and survey approaches make direct comparisons unwieldy for this discussion, but one can readily discern notable differences in waste fractions destined for recovery versus gross generation. Note that the standard deviation values more than twice the average reflect a nonnormal distributions of waste. The nonnormal distribution indicates that material comprising certain waste categories is delivered inconsistently and, when it is delivered, it arrives in large amounts.

P1: SFK/UKS P2: SFK c05 BLBS075-Hood December 15, 2010

20:51

Trim: 244mm×172mm

116

Plant Biomass Conversion

Table 5.2.

Generation and recovery of materials in MSW in the United States (2007). Weight Generated (MMTPY)

Material

Weight Recovered (MMTPY)

Recovery as Percentage of Generation

Paper and paperboard Glass Metals Steel Aluminum Other nonferrous metalsa Total metals

83.0 13.6

45.2 3.22

54.5% 23.7%

15.6 3.35 1.76 20.8

5.28 0.73 1.22 7.23

33.8% 21.8% 69.3% 34.8%

Plastics Rubber and leather Textiles Wood Other materials Total materials in products

30.7 7.48 11.9 14.2 4.43 186.1

2.09 1.10 1.90 1.32 1.16 63.3

6.8% 14.7% 15.9% 9.3% 26.2% 34.0%

31.7 32.6 3.75 68.0

0.81 20.9 Neg. 21.7

2.6% 64.1% Neg. 31.9%

85.0

33.4%

Other wastes Food, otherb Yard trimmings Miscellaneous inorganic wastes Total other wastes Total municipal solid waste

254.1

Includes waste from residential, commercial, and institutional sources. Details may not add to totals due to rounding. MMTPY, million metric tons per year; Neg., less than 5,000 tons or 0.05%. a Includes lead from lead-acid batteries. b Includes recovery of other MSW organics for composting.

Table 5.3.

Summary of landfilled MSW composition in Pennsylvania.

MSW Component

Average Composition of Disposed Material (wt %)

Standard Deviation

Paper and cardboard Plastic Glass Metals Organics Inorganics

33.3% 11.3% 3.0% 5.4% 34.2% 12.7%

20.0% 9.2% 5.3% 8.6% 21.7% 23.2%

Readers should refer to the Beck report for detailed explanation of sampling methods and statistics.

5.5.3

Water in MSW

Water content is a key concern when evaluating conversion potential of cellulosic waste and economics of conversion approaches. In many second-generation cellulosic biofuels facilities,

P1: SFK/UKS P2: SFK c05 BLBS075-Hood December 15, 2010

20:51

Trim: 244mm×172mm

5

Municipal Solid Waste as a Biomass Feedstock

117

Table 5.4. Moisture content in California MSW components. Component of Disposed MSW Biomass components of MSW Paper/cardboard Food Leaves and grass Other organics Construction and demolition lumber Prunings, trimmings, branches, and stumps Nonbiomass carbon component compounds All nonfilm plastic Film plastic Textiles

Moisture (wt %)

10 70 60 4 12 40 0.2 0.2 10

control of moisture in the processed feedstock is an important consideration for reaction and yield efficiencies and value-added processing. Water in the waste is not convertible to biofuel and often must be removed by drying to be acceptable to the conversion process. Drying MSW requires energy and adds to operating cost. Emission controls for dryers may be needed, adding to capital cost. Moisture content in MSW varies by the material within the waste. The University of California Davis reported on the moisture content in the primary material components of solid waste, summarized in Table 5.4 (Williams et al., 2003). Williams et al. (2003) report that approximately 30% of the biomass component of California waste is water. Similarly, Chester (2009) reports an aggregate water component of 32% for landfilled California waste. Although these data provide initial guidance numbers, variations in moisture content of MSW will occur seasonally in concert with changes in waste composition. For example, yard waste has moisture contents higher than average and one may reasonably expect to receive most of this material in spring, summer, and autumn. Also, periods of wet weather will temporarily increase moisture contents in solid waste especially if collection is weekly instead of semiweekly.

5.5.4

Heavy Metals in MSW

Heavy metal concentrations in raw MSW feedstock will contribute to environmental impacts requiring control measures and also will impact the quality of output from intermediate streams and waste outputs from cellulosic production processes. The presence of trace heavy metals may also influence selection of construction materials. For example, if highly acidic process conditions are used to mobilize the heavy metal fraction in the waste feedstock, then the materials of construction must accommodate this harsh environment. Metals are known to concentrate as the waste feedstock undergoes pretreatment and processing, accumulating in intermediate or recycle streams. Each second-generation biofuel conversion process using MSW must address the fate of heavy metals in the process and the process waste streams, whether liquid, solid, or vapor. Frontend classification systems (including presorting) and temporal (hourly, daily, and seasonal) variations in waste composition will influence downstream metal concentrations. Studies that generalize the metal contents in MSW are scarce and sometimes old and represent the values

P1: SFK/UKS P2: SFK c05 BLBS075-Hood December 15, 2010

20:51

Trim: 244mm×172mm

118

Plant Biomass Conversion

Table 5.5.

Heavy metal contents in MSW to illustrate variability (ppm dry solids basis).

Aluminum Antimony Arsenic Cadmium Chromium Copper Mercury Nickel Tin Zinc

TVA: TN-1a

TVA: MNb

TVA: NJc

Law and Gordon (1979)d

12,504 Not reported 0.4 1 47 32 0.1 13 4 755

1,960 <10 <10 <10 50 30 <10 <10 <10 140

20,400 <10 <10 <10 340 330 <10 <10 40 330

5,400–12,000 Not reported Not reported 2–22 20–100 80–900 0.66–1.9 9–90 Not reported 200—2,500

Values reported as <10 signify that amount of metal was less than the detection limit of the analytical equipment used in the test. a TVA: TN-1 is reported from a RDF obtained (about 1990) from Humbolt, TN, using the Lundell processing system. b TVA: MN is reported from RDF obtained (about 1990) from Eden Prairie, MN, using the Buhler-Miag processing system. c TVA: NJ is reported from a RDF sample obtained (about 1990) from K Plant in New Jersey, but the processing system was not known to TVA. d Law and Gordon (1979) show data obtained from the Bureau of Mines report published in 1979.

for a particular waste shed. When heavy metals require an environmental assessment, or pose a significant challenge to a given process technology, an investigation of heavy metal contents may be warranted. In evaluating the environmental considerations of the use of refuse-derived fuel (RDF) for cellulosic fuels and chemicals, the Tennessee Valley Authority (TVA) obtained RDF from different facilities and analyzed metal concentrations (Broder et al., 1993). In their work, TVA observed that metals contained in MSW feedstock are soluble in low-pH conditions and precipitate out of solution in high-pH conditions. Indeed, pretreatment or process conditions in this acidic environment provide a practical approach to addressing the control and removal of heavy metals from MSW feedstock. TVA observed notable interbatch variations in metal content (Broder et al., 1993). This observation is inferred by older studies conducted by the now-closed Bureau of Mines division of the US Department of Interior (Law and Gordon, 1979). Selected test results are presented in Table 5.5 to illustrate the range and variation in metal concentrations from three facilities from which TVA obtained samples. For comparison, the range of results summarized by Law and Gordon (1979) are presented beside those selected from the TVA work. Variations in metal contents have also been reported within individual components of the solid waste stream. Savage (1994) used information gathered in 1988 from Broward County, Florida. Illustrative selections of his data are presented in Tables 5.6 and 5.7. Clearly, the higher concentration of metals in noncellulosic fractions of MSW suggests that this fraction should be carefully considered when calculating the total metal content in a given waste stream when it is part of the overall feedstock stream. Variations in the metal contents between residential and industrial waste deliveries have been observed and reported in a study by the Swiss Federal Office for the Environment (FOEN,

P1: SFK/UKS P2: SFK c05 BLBS075-Hood December 15, 2010

20:51

Trim: 244mm×172mm

5

Table 5.6.

Zinc Cadmium Nickel Chromium Copper Arsenic Mercury Antimony Tin

Municipal Solid Waste as a Biomass Feedstock

119

Metal contents in cellulosic fraction of MSW (ppm dry basis). Mixed Paper

Newsprint

Corrugated

Plastic

Yard Waste

Food Waste

Wood

Other Organic

96.0 0.3 5.0 6.0 30.0 0.5 0.05 11.1 87.0

59.0 0.3 2.0 2.0 15.0 0.4 0.01 13.5 7.0

36.0 0.3 0.2 9.0 33.0 0.8 0.02 5.1 21.0

73.0 0.4 2.5 43.0 364.0 2.2 0.02 10.1 151.0

52.0 0.3 4.0 6.0 71.0 0.4 0.05 7.2 176.0

25.0 0.2 0.6 1.0 5.0 0.3 0.01 4.3 55.0

36.0 3.2 3.0 55.0 5.0 0.5 0.01 3.3 84.0

525.0 4.3 9.0 67.0 68.0 2.1 5.45 175.0 58.0

Table 5.7.

Metal content in noncellulosic materials in MSW (ppm dry basis). Metal Fraction in Broward County, FL, MSW Stream

Trace Metal Element

Ferrous Metals

Aluminum

Glass

Zinc Cadmium Nickel Chromium Copper Arsenic Mercury Antimony Tin

442.0 0.7 206.0 303.0 180.0 9.9 0.10 0.3 670.0

193.0 1.7 103.0 46.0 640.0 6.0 0.10 0.3 12.0

5.0 0.3 9.0 7.0 7.0 0.2 0.01 1.2 593.0

or the German acronym BAFU) in 2006 for the Thurgau municipal waste incinerator plant in Weinfelden, Switzerland (Morf, 2006). FOEN notes that this was the first Swiss study to examine differences between these two generators even after two decades of periodic waste studies associated with incineration infrastructure. The English summary of the report stated that heavy metals (Sn, Cu, Zn, Sb, and Ar) concentrations in industrial waste were almost uniformly higher compared to domestic waste. Only mercury was found to be about the same in the two waste streams. FOEN indicated that the findings were preliminary and that conclusions on the origins of these differences were not possible.

5.6 Preparing MSW for Conversion Processing—Mixed Waste Material Recovery Facilities (MRFs) Use of MSW as a cellulosic feedstock depends upon the ability of individual process arrangements to accept the waste, prepare it for conversion, and process the desirable materials in MSW into streams that produce value-added fuels or chemicals. As may be seen in the characterization of MSW, the heterogeneity of MSW poses a significant challenge to its use as a cellulosic feedstock.

P1: SFK/UKS P2: SFK c05 BLBS075-Hood December 15, 2010

120

20:51

Trim: 244mm×172mm

Plant Biomass Conversion

Preparation of the solid waste stream is the first step in accessing the cellulosic fraction of solid waste. Preparation involves removal of unwanted material (as much as practical) so that a cellulose-rich stream can be used in the conversion process. Some thermochemical processes claim that little or no preparation is required. Without verifying or disputing these assertions, one can reach the reasonable and practical conclusion that some preparation will be required based on the wide range of component materials in waste streams, some of which pose safety concerns or are unacceptable to the downstream conversion process. One proven approach to preparing a mixed solid waste stream involves design and construction of a mixed waste material recovery facility (MRF). Mixed waste MRFs—sometimes referred to as a mixed waste processing (MWP) plants or, unfortunately, “dirty” MRFs—may be designed to process the typical heterogeneous waste streams from residential, institutional, and commercial generators. However, experience in existing facilities indicates that special handling of some wastes is required and some contamination will remain in the exit stream of the mixed waste MRF even after presorting and preparation. Thus, the process technology that follows the MRF must be able to safely and efficiently address any contaminants that will remain. The design goal is to provide sufficient flexibility to handle the waste stream components while still producing an exit stream suitable for the downstream conversion process. Traditional recyclables can be diverted as part of the MWP, but the market value of these materials is relatively minor compared to higher value-added biofuels. A distinction should be made between mixed waste MRFs and “clean” MRFs. Although some of the mechanical and manual operations are similar between the two systems, a clean MRF is designed to sort source-separated recyclables delivered separately from solid waste. Clean MRFs are designed to produce high-quality, saleable recyclable materials such as plastics, ferrous metals, aluminum, glass, corrugated cardboard, and mixed paper. Contamination must be low to meet market requirements. Volatility of prices in the recycling markets is legendary. Generally, clean MRF facilities require a subsidy in some form—permanent or periodic—to achieve break-even or profitable operations. In the United States, source-separated MRFs far outnumber mixed waste MRFs. Mixed waste MRFs include a tipping floor presort followed by mechanical and manual operations. Although each MRF design is unique according to the desired outputs, the primary components of these systems involve waste screening and sizing, density classification, separation of ferrous and nonferrous metals, and manual sorting stations aligned along conveyors. The final steps likely will include shredding to the desired size and drying if required for the conversion process. Capital costs of MWP facilities are difficult to generalize. So-called standard designs are not available. Each facility is designed to process a given waste stream and the design also will consider local variations in waste generation, deliveries, facility capacity, labor markets, and regulations. Chester (2009) provides a range of capital costs for large capacity MWP facilities for his case study on MSW-derived ethanol in California. He suggests a capital cost range of $26,000–58,000 per short ton capacity per day (2008 dollars). For his work, Chester applied the lower value under the assumption that the MWP facilities in his analysis constituted an expansion to existing transfer stations. This author’s industry knowledge indicates that “greenfield” facilities are more likely to reach the upper limit of Chester’s range for processing facilities near the 1,000-ton per day capacity level. Operating costs, like capital costs, also depend upon several variables, including the degree of automated and manual sorting. Chester (2009) uses a value of $36 per ton, and this author’s experience suggests a range of $30–40 per ton as indicative costs.

P1: SFK/UKS P2: SFK c05 BLBS075-Hood December 15, 2010

20:51

Trim: 244mm×172mm

5

5.6.1

Municipal Solid Waste as a Biomass Feedstock

121

Presorting

Tipping floor presorting is the first preparation step and, normally, is a requirement of the operating permit of the facility. Over and above the desire to produce a cellulose-rich feedstock, presorting is an important health, safety, and operational step. After solid waste is discharged from the collection or transfer vehicle, waste is spread across the floor for inspection. Floor personnel and mobile equipment operators visually identify bulky items that are unsuitable either for feedstock or are unacceptable to the downstream MRF operation and processing system. Examples include mattresses, furniture, carpets, tires, appliances, rope and cable, and other troublesome components. Personnel are required to look for possible hazardous materials, including large batteries, paint, drums, solvent cans, gas cylinders, asbestos-containing material, and medical waste. Finally, separated loads of green waste, unrecyclable paper and cardboard, and other cellulose-rich streams may be segregated for easier handling or direct bypass to dryers and shredders located immediately downstream from the MRF operation (Figure 5.4). One should not be surprised to learn that the tipping floor presort would not identify all unacceptable components of the waste stream. Nevertheless, trained and experienced staff can accomplish the presort with good results. Studies are available on the effect of presorting on the heavy metal component of waste in MSW incinerators. In a report prepared for the USEPA, SAIC qualitatively describes the benefits of presorting for removal of aluminum, ferrous metal, batteries, glass, and grit (SAIC

Figure 5.4.

Bulky waste offloading and presorting.

P1: SFK/UKS P2: SFK c05 BLBS075-Hood December 15, 2010

122

20:51

Trim: 244mm×172mm

Plant Biomass Conversion

and SCS Engineers, 1995). They indicate that removal of these waste stream components reduced by 40% air emissions of lead, cadmium, mercury, beryllium, selenium, zinc, and arsenic. While these reductions should not be used as absolute values, they do highlight the benefit of presorting to remove significant portions of metals prior to downstream processing. After presort, only broad descriptions of individual sorting operations are possible. Each MRF design targets the particular requirements of the downstream process. Mixed waste MRFs in operation today are used to either reduce the amount of waste going to landfill or to provide a feedstock to a WTE plant. No mixed waste MRFs have been constructed specifically for cellulose conversion plants. Finally, the MRF design team or owner typically regards as proprietary their selection and configuration of equipment.

5.6.2

Mechanical Sorting Operations

Screening and size sorting are accomplished by disk screens or rotating trommels. Generally, three sizes are targeted—“overs” are larger than about 4–6 inches, “middlings” are between about 2–4 inches, or 2 inches and the lower size of “overs,” and “unders” are the fraction smaller that the “middlings” (Figure 5.5). Density sorting is accomplished to separate light materials from heavier material and is usually accomplished with air classifiers or air knives. In air classifiers, heavy and large fractions fall for collection on conveyors, while lighter material is pulled out with a controlled air stream. Parallel high-velocity sheet flows of air in air knives help separate light material with small differences in density. MWP for second-generation conversion processes may not

Figure 5.5.

Trommel screening.

P1: SFK/UKS P2: SFK c05 BLBS075-Hood December 15, 2010

20:51

Trim: 244mm×172mm

5

Figure 5.6.

Municipal Solid Waste as a Biomass Feedstock

123

Manual sorting.

utilize density separation to the same extent as processing facilities designed for recyclable material recovery since the removal of cellulose-rich materials may not be the operational objective of the system. Magnets and eddy current separators are used to separate ferrous metal and nonferrous metal, primarily aluminum. Electromagnets are positioned above or below a conveyor that moves waste, and ferrous metals are pulled out and delivered to designated containers for recycling. Eddy current separators use spinning magnetic rotors configured to repel (as opposed to attract) nonferrous metals away from the waste stream moving on a conveyor. Nonmetallic fractions remain on the conveyor for further processing.

5.6.3

Manual Sorting Operations

Manual sorting generally follows the mechanical sorting portion of the MRF. Personnel are positioned at sort stations aligned along conveyor belts. Material from the “overs” and “middling” trommel streams pass by the sort stations. Sorters remove recyclables and unacceptable waste at designated stations. These materials are dropped into containers below the sorting stations. Depending upon the design and throughput capacity, a mixed waste MRF may have a dozen or more sorters per shift (Figure 5.6).

5.6.4

Recovery Rates of the MRF System

Mixed waste MRFs are designed to accomplish removal of portions of the waste stream. Since MSW is heterogeneous, it follows that a processing facility design does not per se produce a uniform feedstock. Designers can configure the processing facility equipment to accomplish defined operations for material sizing and removal. Manual sorting can be tailored to enhance the output to reach the desired target feedstock. In the end, however, the processing facility will have variations in feedstock composition. Table 5.8 shows some typical removal efficiencies based on this author’s experience and general industry metrics.

P1: SFK/UKS P2: SFK c05 BLBS075-Hood December 15, 2010

124

20:51

Trim: 244mm×172mm

Plant Biomass Conversion

Table 5.8.

Typical removal efficiencies in a mixed waste MRF (by weight).

Component of Waste Stream

Removal Efficiencies (by weight)

Aluminum Ferrous metals Nonferrous metals Glass Plastic Waste wood (separated for chipping) Bulky waste (removed for offsite disposal)

60–75% 80–90% 50–75% Up to 85% Up to 50% 75–90% Up to 5%

Table 5.9.

Weight percent of sugars in MSW. Weight Percent Dry Basis

Raw RDF pellet Raw newsprint Cardboard Recycled paper

Glucose

Mannose

Xylose

Lignin/Ash

72.1 62.2 75.8 62.2

9.9 9.9 7.0 6.2

13.7 9.7 12.0 9.5

4.3 18.2 5.2 22.1

5.7 Cellulosic Content of MSW TVA reported results from their research in waste-derived cellulose conversion using RDF derived from MSW (Broder et al., 1993). TVA utilized a glucose–mannose–xylose (GMX) spectrophotometric procedure that employs concentrated sulfuric acid to hydrolyze the biodegradable fraction into sugars (Table 5.9). The sugars were calculated using their differing absorption in a spectrophotometer. The samples were dried and ground to pass a 40-mesh sieve screen and obvious pieces of glass, plastic, and metal were manually removed. The available cellulose and hemicellulose for conversion depends upon processing technologies such as dilute acid, concentrated acid, or enzymatic hydrolysis (either using yeast or co-fermentation enzymes).

5.7.1

Glucose and Ethanol Yields from MSW

The chemical composition of the feedstock determines the maximum theoretical yield of ethanol (WSEO, 1991). When losses in yield due to conversion inefficiencies are ignored, the maximum theoretical yield from one pound of cellulose is shown in Table 5.10. In conversion processes, typical yields in cellulose to glucose may be expected to be about 60–70%. Fermentation of glucose to ethanol has an efficiency of about 90–95%. These benchmarks are useful and practical when considering overall ethanol production potential using solid waste. For example, an MSW-to-ethanol process provider and project development firm, Masada Resource Group LLC, provided summary data obtained from pilot plant operations that summarize the glucose yield of waste sorted at a mixed waste MRF. Pilot plant trials were accomplished using vendor equipment. Masada reported an average of about 71% glucose yield for multiple samples (Masada Resource Group, LLC, Birmingham, AL).

P1: SFK/UKS P2: SFK c05 BLBS075-Hood December 15, 2010

20:51

Trim: 244mm×172mm

5

Table 5.10. cellulose.

Table 5.11.

Municipal Solid Waste as a Biomass Feedstock

125

Ethanol yield from 1 pound of

Compound

Conversions

Cellulose Glucose Ethanol Ethanol

1.00 lb 1.11 lb 0.086 gal 0.567 lb

Variations in sugar yield from various sources of MSW.

Theoretical Yield (gal/wet ton) Generated MSW Paper Wood Yard Food

44.9 4.7 4.4 —

Dilute Acid Hydrolysis Yield (gal/wet ton)

Concentrated Acid Hydrolysis Yield (gal/wet ton)

22.2 2.0 2.2 1.5

33.5 2.9 3.2 2.3

Total yield per ton

>54.1

27.9

41.9

Landfilled MSW Paper Wood Yard Food Total yield per ton

35.1 6.5 2.7

17.3 2.8 1.3 2.0 23.4

26.2 4.0 2.0 3.0 35.2

— >44.4

Sakamoto (2004) highlights the variations in yield estimates for MSW. He calculated theoretical and estimated yields of dilute and concentrated acid hydrolysis conversion processes by looking at waste composition (generated vs. landfilled) and estimated production values. Selected data presented in Table 5.11 illustrates the variations in yield. Note that TVA (Broder et al., 1993) indicated a yield using dilute acid hydrolysis of 50.4 gallons of ethanol per dry ton of RDF waste. Chester (2006), as noted earlier, presented an estimate of ethanol production using landfilled waste in California. He estimated that 22 million tons of landfilled waste would produce 1–1.5 billion gallons of ethanol. This corresponds to a yield range of 45–68 gallons of ethanol per wet ton of landfilled waste.

5.8 Framing the Potential The 2007 Energy Independence and Security Act (EISA) established the annual production goal by 2022 of 36 billion gallons of renewable fuel, meaning fuels derived from nonfossil sources. A significant portion of the renewable fuel production is to be advanced biofuels coming from nonedible plants and wastes. Advanced biofuels must also achieve at least a 50%

P1: SFK/UKS P2: SFK c05 BLBS075-Hood December 15, 2010

126

Table 5.12.

20:51

Trim: 244mm×172mm

Plant Biomass Conversion

MSW production and ethanol yield.

Generated MSW Landfilled MSW Theoretical ethanol yield Lower ethanol yield Higher ethanol estimate

Annual Production

Use

254 million wet tons 137 million wet tons 7,400 million gallons 3,430 million gallons 6,170 million gallons

54% landfilled 54 gal/wet ton 25 gal/wet ton 45 gal/wet ton

reduction in life-cycle greenhouse gas emissions (Hoekman, 2009). (For perspective, corn ethanol contributed 6.7 billion gallons in 2007, or about 4% of United States total gasoline usage.) The range of waste compositions, conversion process efficiency, waste industry dynamics, and market issues necessitates a dedicated examination to arrive at a comprehensive estimate of ethanol production using MSW feedstock. Such an examination is beyond the scope of this chapter. Nevertheless, reasonable assumptions can be made to acquire a sense of the production capacity potential. Table 5.12 provides an indicative range of ethanol production using rounded values similar to those presented earlier in this chapter. Clearly, with potential ethanol production from MSW ranging between 3.43 and 6.17 billion gallons annually, MSW is a desirable cellulosic feedstock that helps meet the EISA goal. Solid waste represents significant potential as a second-generation cellulosic feedstock. Much remains to be accomplished to demonstrate process technologies capable of safely utilizing this complex feedstock, but the potential is clear. Existing infrastructure and logistics, economic incentives, and ubiquitous cellulosic feedstock combine to make solid was an attractive source of domestic biofuels.

References Broder, J., Henson, L. & Barrier, J. 1993. Municipal Solid Waste and Waste Cellulosics Conversion to Fuels and Chemicals, April 1990–September 1992 Final Report. Muscle Shoals, AL: Tennessee Valley Authority, Biotechnical Research Department. Chester, M. & Martin, E. 2009. Cellulosic ethanol from municipal solid waste: A case study in the economic, energy and greenhouse gas impacts in California. Environmental Science and Technology, 43, 5184. Hoekman, S. 2009. Biofuels in the US—Challenges and opportunities. Renewable Energy, 34, 14–22. Law, S. & Gordon, G. 1979. Sources of metal in municipal incinerator emissions. Environmental Science and Technology, 13, 432–438. Morf, L. 2006. Chemishc Zusammensetzung verbannter Siedlungsabfalle Untersuchungen im Einzugsgebiet der KVA Thurgau Umwelt-Wissen Nr 0602, Bundesamt fur Umwelt, Bern. 104 S. http://www.bafu.admin.ch/publikationen/publikation/00019/index.html?lang=de. Repa, E. 2001. The US Solid Waste Industry: How Big is it? Available: http://wasteage.com/ mag/waste us solid waste/. SAIC and SCS Engineers. 1995. Analysis of the Potential Effects of Toxics on Municipal Solid Waste Management Options. Cincinnati, OH: EPA Office of Research and Development Risk Reduction Laboratory.

P1: SFK/UKS P2: SFK c05 BLBS075-Hood December 15, 2010

20:51

Trim: 244mm×172mm

5

Municipal Solid Waste as a Biomass Feedstock

127

Sakamoto, O. 2004. The Financial Feasibility Analysis of Municipal Solid Waste the Ethanol Conversion. East Lansing, MI: Michigan State University. Available:http://ageconsearch. umn.edu/bitstream/11179/1/pb04sa01.pdf Savage, G. 1994. The history and utility of waste characterization studies. MSW Management, May/June. Simmons, P., Goldstien, N., Kaufman, S., et al. (2006). The State of Garbage Biocycle April 2006 p. 40. http://www.envortus.com/documents/BiocyleStateofGarbageArticle06.pdf (accessed March 15, 2009). USEPA. 2008. Municipal Solid Waste in the United States: 2007 Facts and Figures (EPA530R-08-010). Washington, DC. Washington State Energy Office (WSEO). 1991. Mixed waste paper to ethanol fuel. Olympia, WA: WSEO. Williams, R., Jenkins, B. & Nguyen, D. 2003. Solid Waste Conversion: A Review and Database of Current and Emerging Technologies. Final Report, University of California—Davis.

P1: SFK/UKS P2: SFK c05 BLBS075-Hood December 15, 2010

20:51

Trim: 244mm×172mm

P1: SFK/UKS P2: SFK c06 BLBS075-Hood December 15, 2010

20:53

Trim: 244mm×172mm

Chapter 6

Water Sustainability in Biomass Cropping Systems Jennifer L. Bouldin and Rodney E. Wright

6.1 Introduction As independence from fossil fuels increases in modern societies, the drive for energy will cause a shift in land-use patterns towards bioenergy crops. Consequently, either cropping patterns will change to bioenergy crops or uncultivated land will be brought into production. Newly cultivated land may include marginal land (not typically used for production due to anticipated low crop production) and more fertile areas not previously under crop management (i.e., forested, in Federal Reserve programs, or other production such as pasturing). First-generation bioenergy crops are typically food crops used in the production of either ethanol or biodiesel through fermentation or extraction processes. The use of food crops as energy has raised issues as they compete with the nutritional supply for an increasing global human population. However, these crops currently hold the most promise in fulfilling energy needs as technology exists to produce fuels in the form of ethanol or biodiesel from these crops. Second and subsequent generation bioenergy crops are of interest, given that they do not directly compete with food. Even so, the technology is not currently at the level for mass production from these sources. It should also be noted that competition for arable land to produce these crops creates an indirect competition with food and fiber. It is estimated that 50% additional land will be needed to supply sufficient biomass for current fuel needs (Muller et al., 2007). Second-generation bioenergy crops, grown for their cellulose-to-ethanol potential, are also considered as forage crops that may supply fodder for livestock. By-products of food and lumber production (e.g., crop stover, woody residue) can also be used in second-generation or cellulosic biomass. Whichever bioenergy crop is produced and subsequently used in biofuel production, protecting water and soil quality is of utmost importance for long-term sustainable energy production.

Plant Biomass Conversion, First Edition. Edited by Elizabeth E. Hood, Peter Nelson and Randall Powell.
C 2011 John Wiley & Sons Inc.

129

P1: SFK/UKS P2: SFK c06 BLBS075-Hood December 15, 2010

130

20:53

Trim: 244mm×172mm

Plant Biomass Conversion

Water and energy are basic resources for human existence. Clean water is fundamental for healthy lives, and modern societies would face gridlock without energy for transportation and electrical power. Without both of these resources, we cannot produce, process, distribute, or prepare food. Globally, the largest agricultural production regions are those with ample water supplies in the form of precipitation, water available for supplemental irrigation (ground or surface water), and fertile soils. Producing energy from crops will support independence from fossil fuels but must be balanced with water stewardship. Care must be taken to produce these crops without stressing available water resources and protect water quality. This chapter considers the water quality and quantity issues surrounding bioenergy cropping and production; processing and water footprints are not considered. Water for processing bioenergy crops in biofuel facilities is of local concern, and facilities are increasing their water conservation by recycling and reduction measures (Water Science and Technology Board, 2007). Waste streams from these facilities are regulated as point source discharges by the US Environmental Protection Agency under the Federal Water Pollution Control Act of 1972. Point source discharges are permitted through the National Pollution Discharge Elimination System, thus water released from these plants must meet individual permitted limits.

6.2 Water Use in Bioenergy Production Agriculture is the leading user of water worldwide. In the United States, 62% of total freshwater withdrawals is used for irrigation and thermoelectric power combined; 37% of total freshwater withdrawals is used for irrigation (Kenny et al., 2009). Irrigation increases crop production, enabling available land to meet the global needs for food and fiber. It is recognized that bioenergy crops will compete with food production for land and water resources (Berndes, 2002); however, the introduction of bioenergy crops into food crop rotations is more likely to occur in locations where the change is economically feasible to the farm manager. The decision of cropping practices relies heavily on market values, but the availability of sufficient freshwater to support economically desirable crops also plays a role in determining these changes. Typically, water used for agriculture is locally supplied (either surface or groundwater), and thus the demand from agriculture cannot exceed local availability. Many heavily irrigated areas are currently experiencing groundwater deficit (withdrawals exceed aquifer recharge), and rotation of more water-demanding crops may further exacerbate water depletion; conversely, rotation of more drought-tolerant crops may lessen the stress on local water supplies. Presently, approximately 1% of total agricultural land is dedicated to bioenergy production but is predicted to increase as a result of positive externalities, such as tax incentives for reduced greenhouse gas emissions and carbon sequestration, and improved farm revenue (de Fraiture et al., 2008) especially from the second-generation (lignocellulosic) crops (McLaughlin and Walsh, 1998; Sanderson and Adler, 2008). As increased land is dedicated to bioenergy crop production, either through voluntary changes or economic incentives, care should be taken to consider freshwater resource availability in this decision-making process. Water use efficiency (WUE) is the ratio of biomass accumulation to water consumed (Sinclair et al., 1984) and has been calculated for most bioenergy crops. Sinclair et al. (1984) noted that WUE is higher in C4 than C3 crops due to differences in photosynthetic pathways, leaving more room for WUE improvement through breeding and genetic modification in C3 plants. It was noted that higher efficiency in C4 plants leaves less room for increased efficiency, while selection of more efficient cultivars in C3 plants will result in greater WUE for those

P1: SFK/UKS P2: SFK c06 BLBS075-Hood December 15, 2010

20:53

Trim: 244mm×172mm

6 Water Sustainability in Biomass Cropping Systems

131

specific crops. Studies have compared WUE among assorted bioenergy crops (Berndes, 2002; De La Torre Ugarte et al., 2008; Gerbens-Leenes et al., 2009; Reddy et al., 2007) and even among crop genotypes (Clifton-Brown and Lewandowski, 2000). Differences in water use vary among crops, but will also vary among regions, soil types, topography, and land management; thus, interpretations must be made carefully. For a sustainable future in bioenergy crops, management decisions must take into consideration the environmental consequences to the natural resources available locally. The complexity of these issues will not result in a blanket best-case scenario for cropping types, land management, and water use, but rather must be a local decision as to which bioenergy crops are best suited for a particular region. With present technology, it is more water efficient to burn total biomass than to produce biofuels (Gerbens-Leenes et al., 2009). Greater energy recovery per water use is expected as technologies improve to include large-scale cellulose-to-ethanol techniques. Early stages of the biofuel industry are improving on existing techniques for producing biodiesel and ethanol from first-generation crops while awaiting technologies for lignocellulose-to-ethanol conversion. The demand for bioenergy crop production and subsequent water availability issues mirror the available energy conversion technologies. As these techniques are developed, water use will change as cropping evolves to the next generation. An increase in corn production has paralleled present emphasis on ethanol conversion in the United States. Biorefineries are available to process corn into ethanol; consequently, crop production in the Mississippi River Basin (MRB) has changed in response to increased demand and enhanced crop revenue. Farming technology in the form of equipment and management practices is readily available for corn and grain sorghum, making cropping changes easily manageable. Many studies have examined the WUE of corn with conflicting results. In semiarid environments, greater ethanol potential was measured with the production of sweet sorghum [Sorghum bicolor (L.) Moench] than corn (Zea mays L.) (Reddy et al., 2007). WUE of sweet sorghum has been reported as approximately 84%–85% of corn and results in greater biomass production per hectare (De La Torre Ugarte et al., 2008; Reddy et al., 2007). Conversely, Gerbens-Leenes et al. (2009) calculated that the global average water footprint of sorghum is four times higher than maize, deeming it the highest water user among 12 first-generation bioenergy crops, including sugar beet (Beta vulgaris L.) with the lowest water footprint. Regional differences are recognized in a summary of previous studies reporting that in 19 US states producing corn, water use increased from east to west and midwest to southwest (Chiu et al., 2009). Dramatic regional differences in water use for corn production can be calculated from Chiu et al. (2009). The study states that corn grown for ethanol production in California used 68 times more irrigation water than corn grown in Ohio. Differences in regional rainfall, soil types, topography, and evapotranspiration rates constitute some of these dissimilarities. This study emphasizes the regional variability of WUE in cropping systems and the importance of locally or regionally available data for sustainable bioenergy production. Water use has been calculated for numerous bioenergy crops by various methods and a summary can be found in Table 6.1. As cellulose-to-ethanol techniques are developed and become more economical, a shift from first-generation bioenergy crops (e.g., corn) to lignocellulosic crops will follow. WUE for second-generation bioenergy crops is generally lower. This shift will include C4 grasses such as switchgrass (Panicum virgatum), which are perennial natives of mixed-grass prairies of North America. As native prairie grasses, they exist within plant assemblages, without irrigation water, and are drought tolerant; however, management in monoculture necessitates additional water for establishment and maximum biomass production. Tall prairie grasses (e.g., switchgrass) have been used to manage erodible soils in the Conservation Reserve

P1: SFK/UKS P2: SFK c06 BLBS075-Hood December 15, 2010

20:53

Trim: 244mm×172mm

132

Plant Biomass Conversion

Table 6.1.

A review of reported water use in bioenergy crops. Water Use Reported

Citation

Inches/gp Plant WUE Plant WUE Plant WUE WUE WF

De La Torre Ugarte et al., 2008 Clifton-Brown and Lewandowski, 2000 Clifton-Brown and Lewandowski, 2000 Clifton-Brown and Lewandowski, 2000 Berndes, 2002 Gerbens-Leenes et al., 2009

Inches/gp

De La Torre Ugarte et al., 2008

Inches/gp Plant WUE(kg) WF

De La Torre Ugarte et al., 2008 Reddy et al., 2007 Gerbens-Leenes et al., 2009

Corn (Zea mays L.)

Inches/gp WUE Plant WUE(kg) WF EWe

De La Torre Ugarte et al., 2008 Berndes, 2002 Reddy et al., 2007 Gerbens-Leenes et al., 2009 Chiu et al., 2009

Sugar beet (Beta vulgaris L.)

WUE WF

Berndes, 2002 Gerbens-Leenes et al., 2009

Energy Crop Perennial crops Switchgrass (Panicum virgatum) Miscanthus (Miscanthus × giganteus) Miscanthus (Miscanthus × sacchariflorus) Miscanthus (Miscanthus × sinensis) Sugar cane (Saccharum spp.) Alfalfa (Medicago sativa) Annual crops Biomass sorghum (Sorghum bicolor L. Moench)

WUE, kg dry mass/ha/mm ET; Plant WUE, g (kg) H2 O/kg plant dry mass; Inches/gp, inches/growing period; WF (water footprint), m3 /GJ electricity; Ewe (embedded water in ethanol), Irrigated water + process water)/ethanol production.

Program (CRP) since 1985, providing water quality protection. Tall grasses reduce runoff from agricultural land to adjacent streams by obstructing overland flow, decreasing water velocity, and allowing absorption into well-developed root zones. These attributes decrease the amount of runoff entering streams and protect water quality by reducing sediment, nutrient, and pesticide transport into receiving waters. The same attributes that make perennial prairie grasses desirable for establishment on CRP land make them good candidates for second-generation sustainable bioenergy production in many areas. Drought-tolerant species [e.g., switchgrass, big bluestem (Andropogon gerardii)] use less irrigation water following first-year establishment but can also be flood-tolerant for use near wetlands and streams. Staggering the annual establishment in production fields by a single producer will allow for moderate to small irrigation water use by a single farm manager. The need for irrigation of perennial crops at establishment and an occasional severe drought will provide a regional water savings when replacing fields previously planted with annual crops. The 10–15 year rotations of these crops also permit re-establishment in staggered years to follow similar patterns of prior plantings. Positive environmental benefits (e.g., increased soil organic matter, carbon sequestration, decreased agrochemical inputs) and greater energy production than corn (McLaughlin and Walsh, 1998) increase the attractiveness of replacing first-generation corn crops with perennial grasses. While reduction of nonpoint source runoff in agriculturally dominated areas is the goal of CRP management planning, concerns exist about the establishment of large-scale dedicated energy crops (i.e., trees). Although perennial and requiring less maintenance than annual crops,

P1: SFK/UKS P2: SFK c06 BLBS075-Hood December 15, 2010

20:53

Trim: 244mm×172mm

6 Water Sustainability in Biomass Cropping Systems

133

the deep roots of trees will increase local evapotranspiration, leaving less water to recharge groundwater aquifers as was noted in tropical hardwood plantations (Evans and Turnbull, 2004). Establishment of bioenergy plots with nonwoody (herbaceous) perennials, such as switchgrass, will use irrigation water primarily during seedling establishment, allow for soil absorption of runoff water, have lower transpiration rates, and result in long-term water (and soil) savings following crop establishment. As bioenergy crops expand, the demand on natural resources such as arable soils and available freshwater will increase. Water competition will be seen on several fronts: crop irrigation, processing of bioenergy crops, and increased evapotranspiration where these crops are grown (Berndes, 2002). Although bioenergy crops presently account for only 1% of agricultural production globally, 2% of total irrigation withdrawals are used for these crops (de Fraiture et al., 2008). With present technology for most crops, direct production of electricity by burning biomass would consume approximately half the freshwater resources as production of bioenergy (i.e., biodiesel, bioethanol) (Gerbens-Leenes et al., 2009). With a growing global population and increasing water scarcity, water efficiency must be improved for bioenergy sustainability. Water conservation during crop production and energy conversion must be a priority with future generation bioenergy crops.

6.3 Water Quality Issues in Bioenergy Crops Agricultural production comprises about 65% of the land use in the MRB and is the main contributor to nonpoint source pollution (Turner and Rabalais, 2003). Pesticides, suspended sediments, pathogens, and nutrients are listed as leading causes of surface-water impairment in the MRB (US EPA, 2003) and monitoring indicates the northern segments of the MRB as the largest contributors to the nutrient load (US EPA, 2007). This area coincides with the corn and soybean areas of the Midwest but also has a high concentration of pasture and rangeland. It is thought that nutrient contribution from pasture and rangeland may be higher than previously believed (Alexander et al., 2008). Regardless of the nutrient source, reduction of the nutrient loading in the MRB is critical and infield nutrient management efficiency will be necessary in high nutrient input watersheds (US EPA, 2007). Corn hectares have decreased slightly from the peak in 2007, which was the highest year since 1933 (NASS, 2009). Increased corn planting in 2007 was contributed to conversion of CRP land to production (Wisner, 2007) and concerns have also been raised that typical corn–soybean rotations will shift to either continuous corn or longer rotations such as corn–corn–soybean (Donner and Kucharik, 2008). This causes concern as the nutrient management plans called for by the Hypoxia in the Gulf of Mexico Science Advisory Board recommended increased land in CRP and other Federal programs (i.e., CSP, EQIP) (US EPA, 2007). Nitrogen (N) loads from the MRB to the Gulf of Mexico have nearly tripled from 1955 to 1996 and are the leading cause for the hypoxic zone (Goolsby et al., 2000). The increased production of bioenergy crops in the southern segments of the MRB may exacerbate water quality problems due to the close proximity to the Gulf of Mexico. Nutrient and pesticide transport distances are shorter in this region of the MRB and will increase the likelihood of impaired waters reaching the river outlets. In a study using the “SPARROW” computational model, the USGS determined that large metropolitan areas and agricultural land bordering major rivers increased the contribution of nutrient transport to the Gulf (Alexander et al., 2008). Furthermore, the timing differences in fertilizer applications from the northern

P1: SFK/UKS P2: SFK c06 BLBS075-Hood December 15, 2010

134

20:53

Trim: 244mm×172mm

Plant Biomass Conversion

Table 6.2. NASS harvested acres for major row crops in the 3 state region: Arkansas, Louisiana, and Mississippi. Commodity

State

1999

2004

2009

Corn for grain Cotton all Rice all Soybeans Total

AR-MS-LA AR-MS-LA AR-MS-LA AR-MS-LA AR-MS-LA

299 1,113 1,038 2,505 4,955

467 1,008 940 2,339 4,754

694 413 881 2,525 4,513

Acreage in thousand hectares.

Table 6.3.

AGNPS model output and soil test recommendations for cropping scenarios.

Scenario

Condition

Water Runoff (cm)

1 2 3 4

04-08 Landuse Cotton Rice Soybeans

28.8 31.3 26.9 29.9

0.987 0.991 1.113 0.955

5

Corn

32.0

1.549

6 7 8 9

Switchgrass Bermuda Alfalfa Tall fescue pasture

21.8 22.5 22.5 21.6

Sediment Load (t/ha/year)

0.101 0.088 0.133 0.054

N Load (kg/ha/year)

U of A: Soil Test Recsa

P Load (kg/ha/year)

kg/Nb

kg/Pc

1.424 3.248 3.065 0.000

0.018 0.021 0.010 0.024

Various 140 202 0

Various 7 4 8

10.461

0.083

371

10

0.982 8.061 0.086 4.206

0.010 0.029 0.045 0.011

67.4d 449 0 225

0.0d 19 22 18

a Espinoza

et al., 2008. recommendation for fine-textured soils. c P recommendation pro-rated by soil test P levels for the soil series. d Garland, 2008. bN

and southern MRB may lengthen the time window of impaired water entering the Gulf. Already the shift in cropping patterns is evidenced over the first decade of the 2000s in the lower delta states of Arkansas, Louisiana, and Mississippi. Table 6.2 lists the four major row crops in this region and their harvested area statistics. The dramatic 132% increase in corn in these three states reflects the mandate for firstgeneration bioenergy crops to replace traditional crops such as cotton (Gossypium hirsutum L.) and rice (Oryza sativa L.). Potentially, a switch to corn could have a negative environmental impact due primarily to its high nutrient requirements. Recommended N and P rates for corn are among the highest of the row crops. Rates are shown on the right side of Table 6.3, which includes University of Arkansas soil test recommendations for the listed crops on a clay soil in the Arkansas Delta (Espinoza et al., 2008). N rates given are generally maximum rates and recommendations include reductions depending on soil texture and residual soil N. Rice N rates vary drastically from 84 to 225 kilograms per hectare (kg/ha) depending on variety, soil texture, and previous crop. The most common varieties planted require 202 kg/ha of N on clay soils following soybean rotation

P1: SFK/UKS P2: SFK c06 BLBS075-Hood December 15, 2010

20:53

Trim: 244mm×172mm

6

Water Sustainability in Biomass Cropping Systems

135

(Espinoza et al., 2008). Corn, by comparison, requires nearly twice the amount of N typically applied to a rice crop and 2.6 times the amount applied to cotton. Increased applications will add to the total N loads already flowing to the Gulf from the MRB (Goolsby et al., 2000). Impacts from changes in cropping scenarios and nutrient loads may be estimated through the use of watershed models.

6.3.1

AGNPS Watershed Model

Numerous water quality models are available for different applications and among the choices are plot, field, basin, and regional spatial applications. The type of output parameters simulated, such as hydrology, sediment, nutrients, and pesticides, provides further differentiation of models. Water movement in and through the watershed is the primary carrier for sediment, nutrients, and pesticides exiting the watershed and entering connected rivers and streams. Therefore, precise simulation of flow is essential to accurately predict water quality parameters (Parson et al., 2004). Two of the most popular watershed models incorporate the USDA Natural Resources Conservation Service (NRCS) runoff curve number to calculate a daily water balance. In addition, these models, SWAT (soil and water assessment tool) and AGNPS (Annualized Agricultural Non-Point Source), are long-term continuous simulation models. This makes them especially useful for analyzing changes in water quality due to best management practice (BMP) implementation in agriculturally dominated watersheds (Borah et al., 2003). The AGNPS model is designed for use in ungauged watersheds to determine the effects of BMPs, develop total maximum daily loads, and for other cost analysis (Bingner and Theurer, 2005; Yuan et al., 2001). The model incorporates the USDA–NRCS Revised Universal Soil Loss Equation (RUSLE), which is the latest technology to determine erosion loss and the impact of conservation practices in the watershed (Bingner and Theurer, 2005). Soil type, digital elevation information, land use, and land management practices are input into the model. The flow network generator of the model uses digital elevation information to divide the watershed into drainage cells. These cells are then used to track the quantity and movement of water, sediment, nutrients, and pesticides. Fertilizer application rates, tillage practices, crop rotations, and other management practices are entered for evaluation. Model calculations are made on a daily basis and also account for features such as feedlots and point source discharge. Overall, the model is an excellent tool for assessing the impacts of BMPs and land use choices in a watershed. More details on the theoretical background of AGNPS are reported by Bingner and Theurer (2005). The AGNPS model is available from Agriculture Research Service homepage (http://www.ars.usda.gov/ main/main.htm). In order to estimate the effects of the aforementioned change in cropping patterns, an agriculturally dominated Mississippi River Delta watershed was selected for evaluation. The modeled area extends from north of the St. Francis River Bridge on US Highway 64 in eastern Cross County, Arkansas. The study area is a small sub-watershed (approximately 6,070 hectares) of the St. Francis River watershed (HUC 08020203) that extends from southeastern Missouri to the Mississippi River near Helena, Arkansas. Detailed soils, digital elevation, and cropping systems information were acquired. The study area is bound on the west by Crowley’s Ridge, a hilly region consisting primarily of forested land. The cultivatable portion of the St. Francis Bay study area is approximately 3,100 hectares. Historically, this area has been planted to a rice–soybean (Glycine max L. Merr.) rotation with limited areas of cotton and corn. This watershed is characteristic of delta watersheds with slow hydrological relief, including landscape features of crop and forested lands.

P1: SFK/UKS P2: SFK c06 BLBS075-Hood December 15, 2010

136

20:53

Trim: 244mm×172mm

Plant Biomass Conversion

Model results are shown in Table 6.3 as average annual stream loading over the 10-year period of the model analysis. The table presents nine different cropping scenarios on the cultivatable land in the watershed. The first scenario is the 2004–2008 historical cropping period, which consisted of 61.4% soybeans, 34.8% rice, 3.3% cotton, and 0.5% corn (NASS, 2009). To enable a direct comparison of crop model output, scenarios 2–5 simulate conditions in the watershed for each crop individually, including the first-generation biofuel crops, soybeans and corn. Scenarios 6–9 in Table 6.3 present model output for select second-generation biofuel crops. These data are presented for comparison to the first-generation biofuel crop corn (center of Table 6.3). Output from the model includes average annual values for surface water runoff, soil erosion, N loading, and P loading. It is assumed the entire cultivatable area is planted to the selected crop for each 10-year modeled period. N loading, one of the primary drivers in the formation of the hypoxic zone in the Gulf, is compared for each scenario. A conventional-tillage corn crop will produce the highest amount of N exiting the watershed at 10.5 kg/ha per year. Calculations from the AGNPS model indicates that N loading from this watershed would be seven times greater, with 100% corn production, than crops planted in the 2004–2008 historical land-use period. Comparing the traditional and first-generation biofuel crops individually, N loss from the watershed is more than three times higher when corn is planted versus cotton and rice. The N loss is considerably higher than soybeans, which have essentially no N runoff since, as a legume, they require no supplemental inorganic N applications. P loading to the MRB is potentially as detrimental as N loading. In coastal waters, N is typically the limiting factor in algal blooms, but recent studies have shown that P can be a limiting nutrient in these areas as well (Sylvan et al., 2006). Generally, corn will require increased P applications over the traditional crops, which result in further stress on the MRB system. Recommended nutrient applications for P are dependent on the soil test results for individual fields. The values included in Table 6.3 were determined using soil test P ranges for the predominant soil series in the watershed and calculated as an average application rate for the soil series. P loading to the watershed from corn is almost four times the amount associated with scenarios 1, 2, and 4 (2004–2008 historical land-use, cotton, and soybean simulations, respectively). The P loss from corn is over eight times the amount lost from rice fields in scenario 3. P losses are tied very closely to soil erosion. It has been shown that P attached to soil particles constituted 74% of total P exported from a watershed (Sharpley et al., 2008). It was also determined by Sharpley et al. (2008) that high intensity rainfall events, which mobilized soil loss, accounted for the majority of P loss. Erosion modeled from AGNPS in this watershed was highest for corn [1.5 metric tons per hectare (t/ha)], constituting a 57% increase over historical land-use. Field operation schedules for modeled crops assumed that conventional tillage practices were used; although no-till model simulations resulted in lower total soil loss, ratios of soil loss for cropping scenarios remained the same. Because of the timing of soil disturbance for corn during high spring rainfall periods, erosion can create high soil loss, especially without BMPs in place. The impact of spring soil loss can be seen in Figure 6.1 as a sediment plume entering the Gulf of Mexico at the Mississippi and Atchafalaya river outlets. It should be noted that sedimentation is the leading contributor to nonpoint source pollution and also acts as a carrier for pesticides, pathogens, and nutrients (US EPA, 2003). The influx of sediment to the Gulf of Mexico can therefore contribute to the zone of hypoxia as can N exported from contributing watersheds. As the demand for bioenergy crops continue and the technology becomes available for efficient energy extraction from second-generation (cellulosic) crops, a change in cropping scenarios will be seen. Modeled results from land use of 100% second-generation bioenergy

P1: SFK/UKS P2: SFK c06 BLBS075-Hood December 15, 2010

20:53

Trim: 244mm×172mm

6

Water Sustainability in Biomass Cropping Systems

137

Figure 6.1. March 28, 2009 (MODIS bands 7,2,1). Satellite image showing sediment plume in the Gulf of Mexico from Atchafalaya and Mississippi rivers. Light gray represents sediment plume following rain event in MRB; white is cloud cover obstructing satellite image. Image courtesy of the USDA Foreign Agricultural Service Crop Explorer at http://www.pecad.fas.usda.gov/ and NASA Global Agriculture Monitoring (GLAM) Project at http://www.pecad.fas.usda.gov/glam.cfm.

crops indicate sediment loss substantially lower than traditional crops (i.e., cotton, rice, soybeans). In these model scenarios, soil loss in traditional cropping systems was approximately 1 t/ha, while modeled second-generation cellulosic crops ranged from 0.054 to 0.133 t/ha—an average reduction of 90% in comparison. The lower half of Table 6.3 summarizes results from AGNPS model runs with four cellulosic cropping scenarios. The potential second-generation crops included in the analysis are switchgrass, alfalfa, bermuda grass (Cynodon dactylon L. Pers.) (warm-season) pasture, and tall fescue (Festuca arundinacea) (cool-season) pasture. Because of high N applications to promote maximum yield, bermuda and fescue hay crops would significantly increase N loading in the MRB over traditional crops but would still export less N from the watershed than corn. With regard to water quality issues, switchgrass and alfalfa offer promising alternative crops since N runoff is 10 and 120 times lower than corn, respectively. Switchgrass is also preferable over first-generation and other second-generation crops in that it has very low P loss. The fertile Mississippi Delta clay soils are naturally rich in P and typically require minimal additional P applications. The remaining second-generation biofuel crops modeled (i.e., alfalfa, bermuda, fescue) require the highest P inputs of the nine scenarios. However, because of the continual ground cover and lack of tillage, they also have higher soil infiltration rates

P1: SFK/UKS P2: SFK c06 BLBS075-Hood December 15, 2010

138

20:53

Trim: 244mm×172mm

Plant Biomass Conversion

and less erosion than traditional row crops. This combination results in lower P loss rates per kg applied than first-generation crops requiring annual tillage. Note that alfalfa would result in a 50% reduction of P exported from the watershed as compared to corn, yet alfalfa requires twice the P application.

6.3.2

Water Quality and the Gulf Hypoxic Zone

Water quality implications of large-scale corn production are devastating. Model results indicate that soil erosion and N and P loading to the Gulf of Mexico would exacerbate the zone of hypoxia and negate any environmental protection that has been accomplished to date. Since water quality results were modeled specifically for a typical delta watershed, they do not fully represent the entire MRB area because of varying ratios of cropland versus noncropland, soil type, and a multitude of other factors. However, the environmental impacts due to drastic changes in crops have been illustrated in these models and numerous previous studies. Other researchers have concluded that first-generation biofuel crops, specifically corn, will create problems with water quality (Costello et al., 2009; Hill et al., 2006; Pimentel, 2003; McLaughlin and Walsh, 1998). On the basis of two models using physical watershed properties in the MRB drainage area, N loading would increase by 10%–34% if corn was planted to meet the mandated renewable fuels requirement (Donner and Kucharik, 2008). Thus, to protect water quality in the MRB, it is imperative to enforce conservation practices, specifically nutrient management plans for corn production (Simpson et al., 2008). Thus far, initial crop swapping that has taken place has not affected the size of hypoxic zone in the Gulf of Mexico. According to Rabalais et al. (2007), in spite of annual fluctuations, the overall hypoxic zone in the Gulf has remained close to the same since the mid-1990s. A decrease in N flux has been measured by USGS in the first decade of the 2000s vs. the 1990s; however, low river flows for the decade may be a contributing factor to the corresponding lack of nutrient movement and resulting stability in the hypoxic zone area (Aulenbach et al., 2007). Wetter conditions and spring floods in the MRB for the years following these measurements may provide different results from the previous decade.

6.4 Conclusions—Water Quantity and Quality Water quantity and quality are of utmost importance for a sustainable and secure global future. It is imperative that freshwater resources be protected. While the subject of bioenergy crops is at the forefront of political agendas and agricultural communities, the requirement for clean freshwater for human populations and natural ecosystems must be maintained. Sustainable agriculture cannot exist without freshwater security. Protecting this vital resource by choosing bioenergy cropping scenarios that will, in turn, protect water quantity and quality are warranted. Large increases in corn hectares in response to ethanol demands will rapidly degrade water quantity and quality. To ensure global sustainability through freshwater preservation and energy independence, a rapid evolution to second-generation (cellulosic) is required. Second-generation crops, especially perennial crops with 10–15 year rotations, are very promising for water conservation and water quality protection of freshwater resources. With a relatively low environmental impact based on nutrient loading and water use, cellulosic perennial crops will provide the sustainability desired based on bioenergy and water protection.

P1: SFK/UKS P2: SFK c06 BLBS075-Hood December 15, 2010

20:53

Trim: 244mm×172mm

6 Water Sustainability in Biomass Cropping Systems

139

As technology advances and large-scale cellulose-to-ethanol techniques are developed, freshwater resources (quality and quantity) will afford greater protection from the negative effects of agricultural runoff. Future biomass production may also offer the benefits of carbon sequestration and possible “cap and trade” for greenhouse gases making bioenergy crops a more economically viable option. The additional use of freshwater in bioenergy agriculture will most certainly add to the stress on global freshwater supplies. Ensuring availability of freshwater for a worldwide population and maintaining freshwater ecosystems to preserve the services they provide are both crucial for long-term global sustainability. Integrating the need for energy, decreasing dependence on fossil fuels, reducing impacts of climate change by carbon sequestration, growing need for food and fiber, and maintaining freshwater supplies must all be incorporated into future considerations of bioenergy production.

References Alexander, R., Smith, R., Schwarz, G. & Al, E. 2008. Differences in phosphorus and nitrogen delivery to the Gulf of Mexico from the Mississippi River Basin. Environmental Science & Technology, 42, 828–830. Aulenbach, B. T., Buxton, H. T., Battaglin, W. T. & Al, E. 2007. Streamflow and Nutrient Fluxes of the Mississippi-Atchafalaya River Basin and Subbasins for the Period of Record Through 2005—Updated [Online]. US Geological Survey. Available: http://toxics.usgs.gov/ hypoxia/mississippi/flux ests/delivery/index.html (Accessed 11 december 2009). Berndes, G. 2002. Bioenergy and water—the implications of large-scale bioenergy production for water use and supply. Global Environmental Change, 12, 253–271. Bingner, R. L. & Theurer, F. D. 2005. AnnAGNPS Technical Processes Documentation, Version 3.2. Oxford, MS: USDA-ARS National Sedimentation Laboratory. Borah, D. K., & Bera, M. 2003. Watershed-scale hydrologic and nonpoint-source pollution models: Review of mathematical bases. Transactions of the ASAE, 46 (6), 1553–1566. Chiu, Y.-W., Walseth, B. & Suh, S. 2009. Water embodied in bioethanol in the United States. Environmental Science & Technology, 43, 2688–2692. Clifton-Brown, J. C. & Lewandowski, I. 2000. Water use efficiency and biomass partitioning of three different Miscanthus genotypes with limited and unlimited water supply. Annals of Botany, 86, 191–200. Costello, C., Griffin, W. M., Landis, A. E. & Matthews, H. S. 2009. Impact of biofuel crop production on the formation of hypoxia in the Gulf of Mexico. Environmental Science & Technology, 43, 7985–7991. De Fraiture, C., Giordano, M. & Laio, Y. 2008. Biofules and implications for Agricultural water use: Blue impacts of green energy. Water Policy, 10, 67–81. De La Torre Ugarte, D., He, L., Jensen, K. L. & Al, E. 2008. Estimating agricultural impacts of expanding ethanol production: Policy implications for water demand and quality. Presented at Annual Meeting of the American Agricultural Economics Association, Orlando, FL. Donner, S. D. & Kucharik, C. J. 2008. Corn-based ethanol production compromises goal of reducing nitrogen export by the Mississippi River. Proceedings of the National Academy of Sciences, 105, 4513–4518. US EPA. 2003. National Management Measures to Control Nonpoint Source Pollution from Agriculture, Polluted Runoff (Nonpoint Source Pollution). Washington, DC. US EPA. 2007. Hypoxia in the Northern Gulf of Mexico. Washington DC.

P1: SFK/UKS P2: SFK c06 BLBS075-Hood December 15, 2010

140

20:53

Trim: 244mm×172mm

Plant Biomass Conversion

Espinoza, L., Mozaffari, M. & Slaton, N. A. 2008. University of Arkansas Lime and Fertilizer Recommendations. Marianna, AR: Univeristy of Arkansas, Division of Agriculture, Soil Testing and Research Lab. Evans, J. & Turnbull, J. W. 2004. Plantation Forestry in the Tropics. New York, NY: Oxford University Press. Gerbens-Leenes, W., Hoekstra, A. Y. & Van Der Meer, T. H. 2009. The water footprint of bioenergy. Proceedings of the National Academy of Sciences, 106, 10219–10223. Goolsby, D. A., Battaglin, W. A., Aulenbach, B. T. & Al, E. 2000. Nitrogen flux and sources in the Mississippi River Basin. The Science of the Total Environment, 248, 75–86. Hill, J., Nelson, E., Tilman, D., Polasky, S. & Tiffany, D. 2006. Environmental, economic, and energetic costs and benefits of biodiesel and ethanol biofuels. Proceedings of the National Academy of Sciences, 103, 11206–11210. Kenny, J. F., Barber, N. L., Hutson, S. S. & Al, E. 2009. Estimated use of water in the United States in 2005. US Geological Survey Circular, 1344, 4–15. McLaughlin, S. B. & Walsh, M. E. 1998. Evaluating environmental consequences of producing herbaceous crops for bioenergy. Biomass and Bioenergy, 14, 317–324. Muller, A., Schmidhuber, J., Hoogeveen, J. & Al, E. 2007. Some insights in the effect of growing bioenergy demand on global food security and natural resources. Presented at Linkages Between Energy and Water Management for Agriculture in Developing Countries, January 28–31, Hyderabad, India. NASS. 2009. USDA Agricultural Statistics Service. Quick Stats Land Use 1999–2009. Available at http://www.nass.usda.gov/ USDA. Parson, J., Thomas, D. & Huffman, R. 2004. Agricultural non-point source water quality models: Their use and application. Southern Cooperative Series Bulletin, 398. Pimentel, D. 2003. Ethanol fuels: Energy balance, economics, and environmental impacts are negative. Natural Resources Research, 12, 127–134. Rabalais, N. N., Turner, B. K., Sen Gupta, D. F. & Al, E. 2007. Hypoxia in the Northern Gulf of Mexico: Does science support the plan to reduce, mitigate and control hypoxia. Estuaries and Coasts, 30, 753–772. Reddy, B. V. S., Kumar, A. A. & Ramesh, S. 2007. Sweet Sorghum: A water saving bioenergy crop. Presented at Linkage between Energy and Water Management for Agriculture in Developing Countries, Hyperabad, India. Sanderson, M. A. & Adler, P. R. 2008. Perennial forages as second generation bioenergy crops. International Journal of Molecular Sciences, 9, 768–788. Sharpley, A. N., Kleinman, P. J. A. & Heathwaite, A. L. 2008. Phosphorus loss from an gricultural watershed as a function of storm size. Journal of Environmental Quality, 37, 362–368. Simpson, T. W., Sharpley, A. N., Howarth, R. W. & Al, E. 2008. The new gold rush: Fueling ethanol production while protecting water quality. Journal of Environmental Quality, 37, 318–323. Sinclair, T. R., Tanner, C. B. & Bennett, J. M. 1984. Water-use efficiency in crop production. BioScience, 34, 36–40. Water Science and Technology Board. 2007. Water Implications of BIofuels Production in the United States. Washington, DC: National Academic Press. Sylvan, J. B., Dortch, Q., Nelson, D. M. & Al, E. 2006. Phosphorus limits phytoplankton growth on the Louisiana shelf during the period of Hypoxia formation. Environmental Science & Technology, 40, 7548–7553.

P1: SFK/UKS P2: SFK c06 BLBS075-Hood December 15, 2010

20:53

Trim: 244mm×172mm

6 Water Sustainability in Biomass Cropping Systems

141

Turner, R. E. & Rabalais, N. N. 2003. Linking landscape and water quality in the Mississippi River Basin for 200 years. BioScience, 53, 563–572. Wisner, R. 2007. Global Economic Impacts of Ethanol Industry Growth. Ames, Iowa: Iowa State Univeristy. Yuan, Y., Bingner, R. L., & Rebich, R. A. 2001. Evaluation of AnnAGNPS on Mississippi delta MSEA watersheds. Transactions of the ASAE, 44, 1183–1190.

P1: SFK/UKS P2: SFK c06 BLBS075-Hood December 15, 2010

20:53

Trim: 244mm×172mm

P1: SFK/UKS P2: SFK c07 BLBS075-Hood December 15, 2010

20:54

Trim: 244mm×172mm

Chapter 7

Soil Sustainability Issues in Energy Crop Production V. Steven Green

7.1 Soil Sustainability Concepts The concept of sustainability focuses on three main areas: environmental, economic, and cultural (Figure 7.1). Each of the three main focus areas is important and none of them can be ignored or diminished. They are all interlinked. Many definitions of sustainability have been written. A broad, global theme for sustainability is “the ability to endure.” The following definition encompasses the main ideas of sustainability and will be used as a platform for our discussion of the sustainability of energy crop production. Sustainability (n): The ability to meet present needs without compromising the ability of future generations to meet their needs. Within the environmental area of sustainable bioenergy production lays soil quality, water quality, and atmospheric quality (Figure 7.1). Each of those is critically important, none can be ignored or diminished, and they are all interlinked. Soil quality or soil health can also be divided into three areas: soil biological, soil chemical, and soil physical parameters and dynamics (Figure 7.1). Soil health is the intersection or interaction among soil biological, chemical, and physical characteristics. Once again, all three soil characteristic areas are equally important, none can be ignored or diminished, and they are all interlinked. The issues facing the sustainable production of bioenergy are varied and complicated. Sustainable agriculture, under which bioenergy crop production fits, refers to the ideology or practice of farming that attempts to actualize long-term, sustained yields by using ecologically sound management practices (Altieri, 1995). The goal is to reduce external inputs by increasingly relying on natural nutrient and energy cycles, thereby reducing carbon footprints and on- and off-farm detrimental externalities, and maintain profits, all while enhancing a quality lifestyle of the producer both now and in the future.

Plant Biomass Conversion, First Edition. Edited by Elizabeth E. Hood, Peter Nelson and Randall Powell.
C 2011 John Wiley & Sons Inc.

143

P1: SFK/UKS P2: SFK c07 BLBS075-Hood December 15, 2010

144

20:54

Trim: 244mm×172mm

Plant Biomass Conversion

Economic

Cultural

Soil

Water

Environmental

Atmosphere

Sustainable energy crop production

Environmental quality

Chemical

Physical

Biological

Soil health

Figure 7.1. Relationship among sustainable crop production, environmental quality, and soil health. Soil health is integral to a sustainable biomass crop production system.

Modern agriculture has focused on assuring that crops have adequate nutrients for maximum potential growth. The use of chemical fertilizers has greatly increased yields in modern crops, yet the inattention to the soil has caused a decline in soil health of many farming systems. The major missing link when applying chemical fertilizers is the organic matter. Organic matter affects the soil’s many physical, chemical, and biological properties and processes. Applying chemical fertilizers and then removing the crop from the field at harvest has resulted in a decline in soil organic matter and subsequently soil health. Additionally, our practice of working the land through tillage, to prepare a clean seed bed, results in optimal conditions for rapid organic matter decomposition, which causes a decline in soil organic matter. We put a lot of demand on our soils by asking this resource to fulfill many tasks. Our soils are used for agricultural, industrial, engineering, urban, wildlife, and recreational purposes. Often forgotten are all of the ecological services that soils provide. Soils sequester carbon (help mitigate climate change), provide habitat for organisms that are essential in the carbon and nutrient cycles that are essential for human survival, manage waste, act as a filter, etc. (Lal, 2007). As we demand more and more of our soils to provide the food, fiber, feed, and fuel for a potential 10 billion person population in the coming century, we must make sure that we maintain the important ecological services that the soil provides.

P1: SFK/UKS P2: SFK c07 BLBS075-Hood December 15, 2010

20:54

Trim: 244mm×172mm

7 Soil Sustainability Issues in Energy Crop Production

145

To this end, it is essential that as we develop dedicated bioenergy cropping systems, we take into account the soil health issues and develop management strategies that enhance soil health.

7.2 Bioenergy Crops and Soil Sustainability There are many energy crops that are being researched, proposed, experimented with, etc. The two main categories of cellulosic energy crops consist of using (1) crop residues or (2) growing dedicated energy crops. When growing dedicated energy crops, we must concern ourselves with the lifecycle of the crop, including the nutrient and water inputs, the nutrient exports, and the off-site and unintended exports that may consist of polluting nutrients, suspended solids, and pesticide runoff. When using agricultural residues, we must concern ourselves with the loss of use of that residue for other purposes, most notably soil cover and soil organic carbon buildup. Here, we will focus on the agricultural plant biomass possibilities for cellulosic ethanol production at the exclusion of forest products, municipal solid wastes, and other materials that are, or could be, exploited for cellulosic conversion to ethanol.

7.2.1

Crop Residues

Agricultural residues, such as stovers and straw, the non-harvested material, are often termed a “waste.” However, these leftovers are not wastes at all, but are a valuable resource that can be returned to the soil to provide nutrients, organic matter, and habitat for beneficial organisms. This material is essential for the conservation of the soil resource; it is also extremely important in terms of crop productivity. Agricultural residues are an essential key protectant of the soil resource. These agricultural residues provide cover for the soil over the winter, protecting the soil surface from the abrading impacts of rain and wind. In fact, for those crops that have a fragile residue (one that decomposes easily and rapidly), it is often recommended that an additional crop be planted in the winter (non-crop season) for the main purpose of protecting the soil from climatic perturbations such as wind and rain. These crops, called cover crops, are an added cost to the producer in the form of seed cost, time, and planting. However, the benefits are many and the producer reaps those benefits for many years (Clark, 2007). In fact, these crop residues are so beneficial to protecting the soil that keeping the residues on the soil surface through no-tillage management has greatly improved soil conservation efforts over the last decade. As of 2004, 24 million ha (hectare) of cropland (23% of total cropland) in the United States was planted using no-tillage management for corn, soybean, wheat, and cotton (Triplett and Dick, 2008). Leaving the crop residue on the soil surface has many conservation benefits. Blanco-Canqui and Lal (2007) showed that wheat residue had a substantial positive impact on water retention, bulk density, soil porosity, and earthworm populations. Other studies have shown that high residue levels increase soil organic carbon storage (Amuri et al., 2008). Green et al. (2007) also found important soil quality benefits from no-tillage management that included increased soil biological activity, a greater nitrogen pool, and greater carbon and nitrogen mineralization potentials over that of plowed soils. Both the crop residue left at the soil surface and the decreased soil disturbance contributes to the enhanced soil health properties of no-tillage management. Looking at estimates from only corn, the United States produces corn on 28,579,000 ha and produces 196,244,000 Mg (megagram) of corn stover (the residue left after grain harvest).

P1: SFK/UKS P2: SFK c07 BLBS075-Hood December 15, 2010

146

20:54

Trim: 244mm×172mm

Plant Biomass Conversion

Of the stover produced, only about 30% is harvestable if erosion control needs are taken into account. In many southern states, crop residues are needed to remain in place since much of it is decomposed too quickly to be useful in controlling erosion such that 100% of the residue is needed to remain on the soil surface for erosion control. In the western United States, much of the corn stover is needed in order to protect the soils from wind erosion. However, this does not take into account other needs for the crop residue such as organic matter accumulation and nutrient cycling in the soil. The benefits of the crop residue left on the soil also include crop yield benefits. It is imperative to remember that the soil is the most valuable resource available to agriculture. The improvements in soil health have a positive impact on crop yields. In Ohio, where soil organic carbon contents are typically greater than in southern and western state soils, BlancoCanqui et al. (2006) found that removing only 50% of the corn stover on sloping lands caused a significant decrease in both grain and stover production in subsequent years, clearly demonstrating the need to manage crop residues wisely. A study by Maskina et al. (1993) in Nebraska, USA, showed that for each Mg/ha of stover removed from a corn crop resulted in 0.13 Mg/ha decrease in grain yield the following year. This effect continued even after the end of the project when all the stover was returned to the soil, indicating that these effects on the soil are prone to be long lived, requiring many years of improved management to return the soil to its potential productivity. The yield decreases can be attributed to declining water retention and a decrease in natural nutrient cycling. In long-term studies (13 years) by Clapp et al. (2000) and by Linden et al. (2000), greater grain and stover production was observed in treatments where the residue was returned to the soil compared to treatments where residue was removed. If only 20–30% or less of the corn stover can be harvested for bioenergy (only considering the soil erosion factor), the question begs to be answered if this is economically feasible to harvest when yields have been shown to suffer when residues are harvested. Many questions still remain with regards to harvesting crop residues for bioenergy, especially due to the spatial variability in soil types and climatic conditions across the United States and the world. There is great variability in soils and climates, and these must be taken into account when determining how much of the corn stover (or any other crop residue) can sustainably be removed. This is one of the reasons that there is so much debate about if and how much corn stover can and should be harvested for cellulosic conversion to ethanol. Obviously, corn stover is easily obtained and readily available as well as not requiring any land to be diverted to some other energy crop. Corn stover has many uses, including energy, but it is essential to remember that these residues are needed to contribute to soil organic carbon status, protect the soil from erosion, supply microorganism habitats, and sustain nutrient cycling. The amount of corn stover needed to control erosion is much less than that needed to maintain soil organic carbon status. In a continuous corn conservation tillage management system, 0.65 Mg/ha of corn stover is needed to control water erosion, while 5.25 Mg/ha of corn stover is needed to maintain the soil organic carbon status (Wilhelm et al., 2007).

7.2.2

Dedicated Energy Crops

The use of dedicated energy crops has been proposed as a more sustainable approach to biomass for energy. Dedicated energy crops include both annual and perennial crops. Of great interest in the United States is the dedicated energy crop species switchgrass (Panicum virgatum; Figure 7.2). However, there are many crops that have great potential for biomass production in the United States and abroad.

P1: SFK/UKS P2: SFK c07 BLBS075-Hood December 15, 2010

20:54

Trim: 244mm×172mm

7

Soil Sustainability Issues in Energy Crop Production

147

Figure 7.2. Panicum virgatum (switchgrass) in its third year of production at the Arkansas State University Farm Complex.

Dedicated nonwoody biomass crops receiving significant research attention are included in Table 7.1 with a sampling of biomass yields from various studies around the United States. This list is by no means a complete list, and the yields listed are by no means comprehensive nor do they represent the entire United States or the world. However, the list offers a starting point for discussion. The US Department of Energy selected switchgrass as the model biomass species for the United States based on several factors: widespread adaptation in various climates and soils, native, low input, perennial, dense root system that traps nutrients, etc. (McLaughlin and Kzsos, 2005). The use of switchgrass as a model species for biomass production does not mean that switchgrass will be the only or even the main crop used in biomass production; it is used as a crop that many scientists can research and share ideas about many different aspects of a biomass production system with a common species. Many different crops will eventually be grown for biomass with certain crops being more suited to the growing conditions of different regions.

P1: SFK/UKS P2: SFK c07 BLBS075-Hood December 15, 2010

148

20:54

Trim: 244mm×172mm

Plant Biomass Conversion

Table 7.1. A sampling of dedicated biomass energy crops receiving research attention in the United States. Energy Crop

Reported Yield

Citation

Perennial crops Switchgrass (Panicum virgatum)

5–18 Mg/ha

Angima et al. (2009), Heggenstaller et al. (2009)

Miscanthus (Miscanthus × giganteus)

15–25 Mg/ha

Lewandowski et al. (2000)

Eastern gamagrass (Tripsacum dactyloides)

5–10 Mg/ha

Angima et al. (2009), Heggenstaller et al. (2009)

Big bluestem (Andropogon gerardii)

6–12 Mg/ha

Angima et al. (2009), Heggenstaller et al. (2009), Hall et al. (1982)

Little bluestem (Schizachyrium scoparium)

6 Mg/ha

Angima et al. (2009)

Indian grass (Sorghastrum nutans)

7–18 Mg/ha

Angima et al. (2009), Heggenstaller et al. (2009)

Bermudagrass (Cynodon dactylon)

7–20 Mg/ha

Angima et al. (2009), Boateng et al. (2006)

Reed canarygrass (Phalaris arundinacea)

9–16 Mg/ha

Cherney et al. (2003), Lamb et al. (2005)

Alfalfa (Medicago sativa)

7–12 Mg/ha

Lamb et al. (2003)

Annual crops Biomass sorghum (Sorghum bicolor L. Moench)

10–20 Mg/ha

Buxton et al. (1999)

Energy cane (Saccharum spp.)

Sorghum-sudangrass (Sorghum bicolor L. Moench)

As we move into a plant-based economy, perennial biomass crops may become a major player in many parts of the United States and around the world for a variety of reasons. Perennial biomass crops 1. 2. 3. 4.

Thrive for many years without need for reseeding Compete very well with weeds after established, thus reducing herbicide inputs Efficiently use nutrients, especially nitrogen Provide year round protection of the soil, thus reducing erosion and runoff water quality issues.

With proper management, perennial biomass grasses should be able to continue producing high yields for 10–20 years or more (Don Tyler, personal communication). This aspect of not having to reseed year after year reduces cost from needing to buy seed, equipment wear from planting, labor, fuel use, and field preparations for planting. Although many of the perennial grasses have a difficult time competing with weeds in the establishment year, in subsequent years, they compete very well (McLaughlin and Kzsos, 2005). In fact, by full establishment (typically year 3), many weeds are crowded out so well that herbicide use is greatly reduced compared to that used on annual crops. Perennial grasses, instead of putting all of their energy into production for a single season, translocate some of the nutrients back to the root system or rhizomes at the end of the

P1: SFK/UKS P2: SFK c07 BLBS075-Hood December 15, 2010

20:54

Trim: 244mm×172mm

7 Soil Sustainability Issues in Energy Crop Production

149

growing season, thus reducing the export of nutrients off the field. This is extremely important, as reducing nutrient inputs has positive impact on both the economic and environmental components of the sustainability issue. Typical annual crop species will use much of the nutrients and put them into the fruiting body, which is the part that is harvested and exported from the field. Those nutrients are then lost to the producer. Many of the perennial grasses are capable of translocating nutrients, especially nitrogen, back to the roots/rhizomes after the plant has received the signal to quit growing (usually by low temperatures or decrease in sunlight hours).

7.3 Resource Use in Biomass Production There are two types of grasses: warm season and cool season grasses. They are differentiated by a number of characteristics, one of which is their photosynthetic pathway. The cool season grasses are typically C3 plants, whereas the warm season grasses are typically C4 plants. The C4 plants do not lose CO2 from the leaf and thus can utilize more of the CO2 for biomass production. C4 plants also have a greater water use efficiency in terms of biomass production and are able to translocate nutrients more easily and quickly (Volenec and Nelson, 2007). The maximum efficiency for C4 plants to convert energy into biomass is 6.0%, while C3 plants have a maximum efficiency of 4.6% (Zhu et al., 2008). Thus the C4 plants have a greater potential than the C3 plants to be used as biomass producers for sustainable bioenergy production, wherever the C4 plants are well adapted (Heaton et al., 2008).

7.3.1

Water and Soil

Perennial grass species typically provide important water quality benefits. Switchgrass and other native prairie grasses have long been used in conservation projects for their soil and water quality enhancing capabilities. Perennial grasses have deep and extensive rooting systems that are capable of scavenging for excess nutrients that could otherwise leach to groundwater. These crops can translocate those subsoil nutrients to the soil surface much like a pump. The plant acts as a pump. More importantly, these grasses are bunch-type grasses that form tillers that expand the area of ground that the plant occupies. As more ground is occupied by these tillers, soil erosion from raindrop impact is reduced, even during the winter months after the plant canopy has been harvested. Additionally, these tillers at the base of the plant provide an extensive filtering mechanism that filters many pollutants, including sediments, pesticides, and excess nutrients. This is accomplished by two mechanisms: first, these plant stems slow the water flow, which causes these pollutants to drop out of suspension (in the case of sediment), and second, it increases the contact time with the plant such that the pollutants have additional time to degrade. Nelson et al. (2006) used the Soil and Water Assessment Tool (SWAT) model to predict environmental effects of switchgrass compared to other commodity cropping systems (corn, soybean, wheat systems). They predicted 99% reduction of erosion, 55% reduction of surface runoff, 34% reduction of nitrate in surface runoff, and 98% reduction of edge of field erosion. Greater protection of water quality compared to typical row crop agriculture is also enhanced by a reduction of chemical inputs. Typical recommended nitrogen (N) inputs for perennial grasses are in the range of 67.5 to 90 kg/ha, whereas corn may receive up to 270 kg/ha for optimum production. However, maximum biomass production of many of the warm season,

P1: SFK/UKS P2: SFK c07 BLBS075-Hood December 15, 2010

150

20:54

Trim: 244mm×172mm

Plant Biomass Conversion

perennial grasses has been reported to require 120–200 kg/ha (107–178 lb/acre of nitrogen) (Heggenstaller et al., 2009; Vogel et al., 2002). The 67.5 to 90 kg/ha recommendation for the warm season grasses is based on economic optimization where the additional biomass produced at greater N rates does not pay for the additional N application at 2009 prices. Nutrient use efficiency is of critical importance, especially in light of the many loss pathways for nitrogen. Nitrogen can be lost through leaching, denitrification, runoff, and atmospheric volatilization. Continued research in the nitrogen use efficiency of the various warm season grasses is of great importance for both economic and environmental sustainability. Many of the grasses being proposed for bioenergy are native grasses that have evolved pest resistance over the years. Their strong competitiveness with weeds is also naturally evolved. Thus, reduced inputs combined with pollution mitigation capabilities will make many of these grasses good candidates for bioenergy cropping systems.

7.3.2

Land Use

The land requirements for sustaining a cellulosic ethanol program must be taken into account. The environmental and social consequences of urban sprawl have long been discussed and reported. Urban sprawl causes habitat reduction, relocation of resources, and disappearance of ecological services in addition to many social and cultural problems. Anytime land use is changed, there will be consequences. One scenario that is often overlooked is the land substitution issue. For instance, land that was in soybean in the United States may be transitioned to corn for ethanol, both grain and cellulosic ethanol pathways. The soybean lost to the marketplace may need to be produced somewhere else. If so, Brazil may pick up the slack and clear more of the savannah (Cerrado) to grow soybean. But if the Cerrado was used to graze cattle, now other areas, perhaps the rainforest, might be cleared in order to graze cattle. So even though there was no net gain or loss in United States’ agricultural land in this scenario, there was a gain of agricultural land somewhere, probably coupled with the loss of habitat and thus, could produce unintended consequences because not all of the interactions were considered on the front end. Many places in the literature state that we will be able to plant switchgrass on marginal land because it can grow with minimal inputs. This is true. Switchgrass will grow on marginal land with minimal inputs. However, for switchgrass to grow well and provide a high yield, it will need some nutrient inputs. Furthermore, for a producer to make a profit, he/she will need to obtain a minimum yield or at least some minimum price for his/her product. A 3-year study in Tennessee showed how both yields and farm-gate breakeven prices vary by land suitability. Five-year projected yields for well drained soils were 16.8 Mg/ha (7.5 ton/acre), while 5-year projected yields for switchgrass grown on poorly drained soils were about half, 8.2 Mg/ha (3.7 ton/acre). Producing energy crops on marginal lands will require more nutrient inputs than those grown on better land and will still likely not reach the yield levels of the better land due to weed infestations, water logged conditions, etc. (Mooney et al., 2009).

7.4 Soil Sustainability Solutions One of the many opportunities for dedicated bioenergy crops to be more sustainable is to focus on alternative or organic nutrient sources (Figure 7.3). Organic nutrient sources provide the required nutrients as well as the additional benefits of improved soil health and increased

P1: SFK/UKS P2: SFK c07 BLBS075-Hood December 15, 2010

20:54

Trim: 244mm×172mm

7

Soil Sustainability Issues in Energy Crop Production

151

Figure 7.3. Class A municipal biosolids (top) and poultry litter (bottom) are waste materials that can be used as a resource by applying to agricultural fields as a fertilizer.

organic matter. Poultry litter (and other animal manures) is commonly used on forage crops near poultry facilities as a means of disposing this “waste material.” The forage crops then utilize the nutrients for growth. More recently, poultry litter is being used as a fertilizer on row crops to supply nitrogen either as a sole source of fertilizer or in combination with other inorganic fertilizers (Sims, 1987). As fertilizer nitrogen prices have risen, it has become more feasible to transport poultry litter farther distances. This is true with other animal manures as well, including dairy and beef manures. About 50–70% of the organic nitrogen from poultry litter is plant available nitrogen (Bitzer and Sims, 1988; Cabrera et al., 1993; Cooperband et al., 2002). Typical poultry litter nitrogen concentrations are in the range of 3–5%. If the poultry litter is to be transported, pelleted poultry litter may be more economical, due to its density and reduced moisture content compared with fresh poultry litter.

P1: SFK/UKS P2: SFK c07 BLBS075-Hood December 15, 2010

152

20:54

Trim: 244mm×172mm

Plant Biomass Conversion

(a)

(c)

(b)

(d)

Figure 7.4. Class A municipal biosolids are produced at the Paragould Light Water & Cable wastewater treatment plant for use by local crop producers. Clockwise from the upper left: (a) oxidation ditches where wastewater is treated with microorganisms to remove pollutants, (b) end section of oxidation ditch where air is churned into the wastewater, (c) digester where wasted sludge is allowed to stabilize, and (d) the dried product from the wasting process (biosolids) that crop producers apply to their fields as a soil amendment (Paragould Light, Water, and Cable.)

Municipal biosolids (Figure 7.4) are another option for sustainable organic supplies of nutrients (Pepper et al., 2008). As with poultry litter, only a portion of the total nitrogen in the biosolids would be available to the plant in the year of application. Studies have shown that typical plant-available nitrogen from biosolids is in the range of 20–40% in the year of application (Barbarick and Ippolito, 2007; Binder et al., 2002). To compensate for the lower plant-available nitrogen issue, increased rates could be recommended as long as soil phosphorus levels are not too high or recommendations can be made based on soil phosphorus. Other advantages of using poultry litter and biosolids compared to inorganic nitrogen exist. Most notably, organic matter is added to the soil. When applying inorganic nitrogen to crops, organic matter is removed from the land in the form of the crop and only inorganic material is added to the soil. When applying organic forms of fertilizer, organic matter is added to the soil, which is essential for maintaining or increasing soil health as well as sequestering carbon as a greenhouse gas mitigation tool (Pepper et al., 2008). When looking at the lifecycle analysis of various wastes, we have a great potential in using a waste product from one process as a resource for growing dedicated bioenergy crops. By taking a waste and turning it into a resource, we will save on the lifecycle analysis of energy crops. We can reduce the amount of fossil fuel use on our lifecycle balance sheet by using wastes from other industries as

P1: SFK/UKS P2: SFK c07 BLBS075-Hood December 15, 2010

20:54

Trim: 244mm×172mm

7 Soil Sustainability Issues in Energy Crop Production

153

opposed to using inorganic fertilizer that is a very intensive user of fossil fuels to produce. Of course, this balance will require supporting policies that reward this solution beyond its current economics. The ability of a bioenergy crop to sequester carbon in the soil is a very important aspect for sustainable bioenergy production. The current issue of global climate change due to increases in atmospheric greenhouse gases is of critical importance. Studies have shown that at least some of the perennial grasses being studied have a high propensity to sequester carbon in the soil (Lee et al., 2007). Switchgrass, for instance, partitions more carbon to the roots than to the shoots. Switchgrass also has a very deep rooting behavior such that the carbon from the roots is deep in the soil and is less likely to escape through decomposition than if the carbon was closer to the soil surface. Additionally, some studies have shown that using animal manures as a source of nutrients increases carbon sequestration above that of mineral fertilizers (Lee et al., 2007). Some of the carbon applied from the manures becomes stabilized in the soil for long-term sequestration since not all of the carbon in the manures is readily decomposable.

7.5 Conclusion For bioenergy crop production to be sustainable, the soil resource must be protected. The soil organic matter/carbon content is of the utmost importance to soil health. The soil organic matter affects soil erosion susceptibility, nutrient cycling, buffering capacity, water holding capacity, water filtration, aggregate stability, microorganism habitat, etc. Soil health is essential to a sustainable bioenergy production system.

References Altieri, M. A. 1995. Agroecology: The Science of Sustainable Agriculture. Boulder, CO: Westview Press, Inc. Amuri, N., Brye, K. R., Gbur, E. E., Popp, J. & Chen, P. 2008. Soil property and soybean yield trends in response to alternative wheat residue management practices in a wheat-soybean, double-crop production system in eastern Arkansas. Electronic Journal of Integrative Biosciences, 4, 64–86. Barbarick, K. A. & Ippolito, J. A. 2007. Nutrient assessment of a dryland wheat agroecosystem after 12 years of biosolids applications. Agronomy Journal, 99, 715–722. Binder, D. L., Dobermann, A., Sander, D. H. & Cassman, K. G. 2002. Biosolids as nitrogen source for irrigated maize and rainfed sorghum. Soil Science Society of America Journal, 66, 531–543. Bitzer, C. C. & Sims, J. T. 1988. Estimating the availability of nitrogen in poultry manure through laboratory and field studies. Journal of Environmental Quality, 17, 47–54. Blanco-Canqui, H. & Lal, R. 2007. Impacts of long-term wheat straw management on soil hydraulic properties under no-tillage. Soil Science Society of America Journal, 71, 1166– 1173. Blanco-Canqui, H., Lal, R., Post, W. M. & Owens, L. B. 2006. Changes in long-term notill corn growth and yield under different rates of stover mulch. Agronomy Journal, 98, 1128–1136.

P1: SFK/UKS P2: SFK c07 BLBS075-Hood December 15, 2010

154

20:54

Trim: 244mm×172mm

Plant Biomass Conversion

Cabrera, M. L., Chiang, S. C., Merka, W. C., Thompson, S. A. & Pancorbo, O. C. 1993. Nitrogen transformations in surface-applied poultry litter: Effect of litter physical characteristics. Soil Science Society of America Journal, 54, 1519–1525. Clapp, C. E., Allmaras, R. R., Layese, M. F., Linden, D. R. & Dowdy, R. H. 2000. Soil organic carbon and C-13 abundance as related to tillage, crop residue, and nitrogen fertilization under continuous corn management in Minnesota. Soil & Tillage Research, 55, 127–142. Clark, A. (ed.) 2007. Managing Cover Crops Profitably. Beltsville, MD: Sustainable Agriculture Network. Cooperband, L., Bollero, G. & Coale, F. 2002. Effect of poultry litter and composts on soil nitrogen and phosphorus availability and corn production. Nutrient Cycling in Agroecosystems, 62, 185–194. Green, V. S., Stott, D. E., Curz, J. C. & Curi, N. 2007. Tillage impacts on soil biological activity and aggregation in a Brazillian Cerrado Oxisol. Soil and Tillage Research, 92, 114–121. Heaton, E. A., Flavell, R. B., Mascia, P. N., Thomas, S. R., Dohleman, F. G. & Long, S. P. 2008. Herbaceous energy crop development: Recent progress and future prospects. Current Opinion in Biotechnology; Energy biotechnology/Environmental biotechnology, 19, 202–209. Heggenstaller, A. H., Moore, K. J., Liebman, M. & Anex, R. P. 2009. Nitrogen influences biomass and nutrient partitioning by perennial, warm-season grasses. Agronomy Journal, 101, 1363–1371. Lal, R. 2007. Soil science and the carbon civilization. Soil Science Society of America Journal, 71, 1425–1437. Lee, D. K., Owens, V. N. & Doolittle, J. J. 2007. Switchgrass and soil carbon sequestration response to ammonium nitrate, manure, and harvest frequency on conservation reserve program land. Agronomy Journal, 99, 462–468. Linden, D. R., Clapp, C. E. & Dowdy, R. H. 2000. Long-term corn grain and stover yields as a function of tillage and residue removal in east central Minnesota. Soil & Tillage Research, 56, 167–174. Maskina, M. S., Power, J. F., Doran, J. W. & Wilhelm, W. W. 1993. Residual effects of no-till crop residues on corn yield and nitrogen uptake. Soil Science Society of America Journal, 57, 1555–1560. McLaughlin, S. B. & Kzsos, L. A. 2005. Development of switchgrass (Panicum virgatum) as a bioenergy feedstock in the United States. Biomass and Bioenergy, 28, 515–535. Mooney, D. F., Roberts, R. K., English, B. C., Tyler, D. D. & Larson, J. A. 2009. Yield and breakeven price of “Alamo” switchgrass for biofuels in Tennessee. Agronomy Journal, 101, 1234–1242. Nelson, R. G., Ascough II, J. C. & Langemeier, M. R. 2006. Environmental and economic analysis of switchgrass production for water quality improvement in northeast Kansas. Journal of Environmental Management, 79, 336–347. Pepper, I. L., Zerzghi, H., Brooks, J. P. & Gerba, C. P. 2008. Sustainability of land application of class B biosolids. Journal of Environmental Quality, 37, S58–S67. Sims, J. T. 1987. Agronomic evaluation of poultry manure as a nitrogen source for conventional and no tillage corn. Agronomy Journal, 79, 563–570. Triplett, G. B., Jr. & Dick, W. A. 2008. No-tillage crop production: A revolution in agriculture! Agronomy Journal, 100, S153–S165. Vogel, K. P., Brejda, J. J., Walters, D. T. & Buxton, D. W. 2002. Switchgrass biomass production in the Midwest USA: Harvest and nitrogen management. Agronomy Journal, 94, 413– 420.

P1: SFK/UKS P2: SFK c07 BLBS075-Hood December 15, 2010

20:54

Trim: 244mm×172mm

7 Soil Sustainability Issues in Energy Crop Production

155

Volenec, J. J. & Nelson, C. J. 2007. Physiology of forage plants. In: Barnes, R. F., Nelson, C. J., Moore, K. J. & Collins, M. (eds.) Forages: The Science of Grassland Agriculture. Ames, IA: Blackwell Publishing. Wilhelm, W. W., Johnson, J. M. F., Karlen, D. L. & Lightle, D. T. 2007. Corn stover to sustain soil organic carbon further contrains biomass supply. Agronomy Journal, 99, 1665–1667. Zhu, X.-J., Long, S. P. & Ort, D. R. 2008. What is the maximum efficiency with which photosynthesis can convert solar energy into biomass? Current Opinion in Biotechnology, 19, 153–159.

P1: SFK/UKS P2: SFK c07 BLBS075-Hood December 15, 2010

20:54

Trim: 244mm×172mm

P1: SFK/UKS P2: SFK c08 BLBS075-Hood December 15, 2010

21:54

Trim: 244mm×172mm

Chapter 8

Fermentation Organisms for 5- and 6-Carbon Sugars Nicholas Dufour, Jeffrey Swana, and Reeta P. Rao

8.1 Introduction While microbe-derived biofuels holds great promise, implementation is hampered as substrate necessary for production is under simultaneous use in food production, and therefore costly. Current efforts are focused on using lignocellulosic biomass as primary material. Lignocellulosic substrate is abundant and underutilized by industry, making it an attractive option. However, employing it industrially is problematic due to the difficulty encountered in microbial fermentation. This chapter discusses biochemical pathways for degradation of lignocellulosic materials and summarizes microorganisms that are capable of fermenting the sugars derived from both lignocellulosic polymer as well as the polymer itself. Amidst ever increasing global energy demands, interest in alternative sources of fuel continues to grow. A substantial amount of research and development centers on renewable fuel sources, which are seen as a long-term solution to “peak oil” (Lovins et al., 2005)—the point at which world petroleum extraction peaks and begins an invariable decline. Biofuels are a form of renewable fuel produced directly from biomass (Giampietro et al., 1997) and are the focus of intense research and development. At the forefront of biofuels research is bioethanol. Ethanol, a simple two-carbon alcohol, is advantageous as a biofuel due to the ease with which it is synthesized by microorganisms. Numerous organisms synthesize ethanol as a natural byproduct of metabolism. This property has been known for some time: Brewer’s yeast, Saccharomyces cerevisiae, has been used by humans since prehistory for this very purpose (Maksoud et al., 1994). Beyond the relative ease of producing ethanol biologically, it is cleaner burning than gasoline, and an ethanol-based fuel economy is calculated to produce increases in air quality (Sonderegger et al., 2004a). Unfortunately, the move to ethanol as a major source of fuel is not without its own difficulties (Bastianoni and Marchettini, 1996), the greatest of which is the substrate used to produce

Plant Biomass Conversion, First Edition. Edited by Elizabeth E. Hood, Peter Nelson and Randall Powell.
C 2011 John Wiley & Sons Inc.

157

P1: SFK/UKS P2: SFK c08 BLBS075-Hood December 15, 2010

158

21:54

Trim: 244mm×172mm

Plant Biomass Conversion

ethanol. Large-scale biogenic production of ethanol currently relies on crop-based raw materials, such as corn starch and sugar cane (Sonderegger et al., 2004a). These materials are also under consumption by various sectors in agriculture and food production. This competition affects not only the price of food, but accounts for 40% of the cost of ethanol-based biofuels (Zaldivar et al., 2001). Both agriculture and biogenic ethanol production use such substrates because they primarily contain glucose. Glucose, a monosaccharide aldohexose sugar, is highly energetic and forms the cornerstone for the energy metabolism of most organisms on the planet (Sadava et al., 2006). In ethanol production, microbes are used and energetic molecules are converted into ethanol, which combusts readily yielding the ability to perform work. Aerobic cellular respiration, the mechanism of energy production used by aerobic organisms (including humans), is highly efficient. Thus, the end products of aerobic cellular respiration are inert, containing little chemical energy. Anaerobic fermentation is a related process, but is comparatively less efficient, where glucose is converted to compounds like ethanol, which are less energetic than glucose, but still contain chemical energy that may be utilized by humans. Fermentative organisms are noted for their ability to use a wide range of sugars. These alternative carbohydrates, particularly the pentose sugars xylose and arabinose, are produced by crop plants in abundance. Additionally, these sugars are of minimal use as livestock fodder and human food and hence to the agricultural industry. In plants, xylose and arabinose are found in the substance lignocellulose. Lignocellulose is made up of three polymers: cellulose, hemicellulose, and lignin. Cellulose is made up entirely of glucose, while hemicellulose has an amorphic structure of a mix of many sugars, including xylose and arabinose. Lignin also has an amorphic structure but does not contain any readily fermentable sugars (Ragauskas et al., 2006) (Figure 8.1).

Figure 8.1. The components of different feedstocks used for biofuel production, including agricultural residues and energy crops (both hardwoods and grasses). These substances are considered the leading substrate candidates for biofuels production [9]. As indicated by the graphs, glucose is the most abundant component of lignocellulosic materials making up between 32–42% of the total material. The “Other” lignocellulose portion includes ash, uronic acids, and other extracts that are not integral to the cellular structure of the material [10].

P1: SFK/UKS P2: SFK c08 BLBS075-Hood December 15, 2010

21:54

Trim: 244mm×172mm

8 Fermentation Organisms for 5- and 6-Carbon Sugars

159

Table 8.1. Lignocellulosic content of common biological wastes. Adapted from Sun and Cheng (2002a). Agricultural Residue or Waste

Lignocellulosic Content (%)

Wheat Straw Discarded Newspapers Leaves and Lawn Refuse Swine Waste Cattle Manure

95 ∼100 90 >30 15

Constituting the bulk of plant cell walls (Sun and Cheng, 2002b), lignocellulose is composed of three polymers: lignin, hemicellulose, and cellulose. As such, it is present in all vascular plants and is astonishingly abundant. It is considered the most abundant polymer in nature (Watanabe et al., 2007) and accounts for more than half the biomass on the planet (Zaldivar et al., 2001). Since it is not easily used by the agricultural industry, it is discarded and forms the primary constituent of agricultural residue streams (Sun and Cheng, 2002a). Table 8.1 details the lignocellulosic content, by percentage, of some common residual products, both agricultural and otherwise. Because of its abundance and lack of alternative uses, lignocellulose as the primary substrate for fermentation is seen as necessary to the establishment of viable bioethanol production. Further, it is calculated to be necessary for ethanol to compete viably with fossil fuels (Hinman et al., 1989). Currently, it is mostly unutilized despite being an enormous potential source of energy. Collectively termed “lignocellulosic alcohol” (Lynd, 1991), the use of lignocelluloses to produce bioethanol (along with other alcohols, such as butanol) is problematic. No known eukaryote will grow on xylose or arabinose anaerobically, and even in bacteria, the primary means of growth on either sugar is nonfermentative, instead using an aerobic metabolism (Jin, 2004). As such, research has been directed towards engineering microbes for efficient fermentation of lignocellulosic sugars. Attempts at achieving fermentation have been performed on a number of organisms, mostly focusing on the brewer’s yeast S. cerevisiae. S. cerevisiae is used extensively in industrial fermentation to produce ethanol (Watanabe et al., 2007; Jin and Al, 2002) due to its ability to efficiently ferment glucose into ethanol, its naturally high tolerance for ethanol in solution, as well as its tolerance for inhibitory compounds produced during growth. In spite of the discovery of an endogenous pathway for xylose metabolism in S. cerevisiae (Toivari, 2004), progress is hampered by a respiratory response exhibited by S. cerevisiae when exposed to xylose, resulting in growth but a distinct lack of fermentation, with xylitol accumulating within the cell (Jin, 2004), although some ethanol production was found to take place.1 Overall, the results of previous efforts in producing a viable, industrial microbial strain for the production of ethanol from lignocellulosic sources may be described, at best, as mixed.

8.2 Fermentation The term fermentation was first coined by Louis Pasteur during the 19th century and means “life without air” (la vie sans l’air) (Bentley and Bennett, 2008). This definition implies continued metabolic activity in the absence of oxygen, which corresponds to the current definition of anaerobic growth. Today, the definition of fermentation has expanded to include any microbial product produced on a large scale, otherwise known as an “industrial fermentation” (Madigan

P1: SFK/UKS P2: SFK c08 BLBS075-Hood December 15, 2010

160

21:54

Trim: 244mm×172mm

Plant Biomass Conversion

Anaerobic fermentation 2 H+

Lactic Acid + Glucose

+

2 Pyruvate +

Ethanol

2 ATP

2 CO2

Aerobic respiration

Glucose

+

6 O2

2 CO2

+

6 H2O

+

38 ATP

Anaerobic respiration

Glucose

+

3 NO3−

+

6 H2O

6 HCO3−

+

6 NH4+

+

34 ATP

Figure 8.2. Energy output of cellular processes. Anaerobic fermentation implies reduction of sugar using endogenous intermediates as the terminal electron acceptor. Aerobic respiration implies the reduction of sugars using oxygen as the terminal electron acceptor. Anaerobic respiration is the reduction of sugars using exogenous inorganic compounds as the terminal electron acceptor. Oxygen is the strongest oxidant, so it produces the most ATPs (Madigan and Martinko, 2006).

and Martinko, 2006). While the original term referred to growth without oxygen, industrial fermentations can include varying degrees of aeration, in fact, the majority of them require relatively high levels of oxygen (Madigan and Martinko, 2006). Microbial fermentations on the industrial scale produce an incredibly wide array of products, including fuels, pharmaceuticals, and more (Ragauskas et al., 2006). The terms aerobic and anaerobic metabolism are more strictly defined to mean “with oxygen” and “without oxygen,” respectively. Anaerobic growth can be further subdivided into anaerobic fermentation and anaerobic respiration. Fermentation is anaerobic metabolism, but anaerobic metabolism does not necessarily mean fermentation. All the processes described allow the cell a means to produce energy through the conversion of energy-rich carbon sources into low-energy intermediates. During aerobic respiration, the cell has oxygen available to it and uses oxygen as the terminal electron acceptor (Figure 8.2). Conversely, anaerobic fermentation is the cellular production of energy using endogenous organic compounds as the terminal electron acceptor. Anaerobic respiration is different from both other processes because it uses exogenous inorganic compounds such as ammonium as the terminal electron acceptor (Erickson and Fung, 1988; Madigan and Martinko, 2006)

8.3 Metabolic Pathways The goal of any cell is to generate energy to drive chemical reactions. One of the most fundamental of these energetic pathways is glycolysis, or the Embden-Meyerhof pathway

P1: SFK/UKS P2: SFK c08 BLBS075-Hood December 15, 2010

21:54

Trim: 244mm×172mm

8 Fermentation Organisms for 5- and 6-Carbon Sugars

161

(EMP), which involves the breakdown of glucose into pyruvate to generate ATP. During this process, for every glucose molecule, 4 molecules of ATP are generated and 2 are consumed, giving a net of 2 ATPs produced (Figure 8.3). As glycolysis is one of the best-characterized biochemical pathways known, (Garrett and Grisham, 2005; Madigan and Martinko, 2006; Walker, 1998), we will omit an in-depth discussion of the subject in favor of briefly describing the major points of the pathway. Upon entering the cell, if the glucose molecule is to be used for glycolysis, it is phosphorylated, becoming glucose-6-phosphate, expending 1 ATP molecule. This intermediate is isomerized to fructose-6-phosphate. With the hydrolysis of another ATP molecule, fructose-1,6-bisphosphate is produced in an irreversible reaction, allowing the 6-carbon sugar to be cleaved into two 3carbon intermediates: glyceraldehyde-3-phosphate (GADP) and dihydroxyacetone phosphate (DHAP). The DHAP molecule is used either for production of glycerol (Wang et al., 2001) or is isomerized into a second GADP. Until this point, the reactions have required ATP to “prepare” for the net production of further ATP. Therefore, the first five reactions of glycolysis are referred to as the “preparatory phase.” The remaining reactions allow for the “payoff ” to take place, where ATP is generated, and accordingly are referred to as they “payoff phase.” In the first reaction of the “payoff ” phase, each GADP is phosphorylated to generate 1,3-bisphosphoglycerate along with NADH from NAD+ . The conversion of the enzyme cofactor NAD+ to NADH is important to note, and its importance will be discussed later. In the next step, 1,3-bisphosphoglycerate is dephosphorylated to 3-phosphoglycerate and generates one ATP molecule per 3-carbon intermediate, meaning that for every glucose molecule metabolized, two ATP molecules are generated, replacing the ATPs used during the preparatory phase. 3-phosphoglycerate is converted to 2-phosphoglycerate by a mutase enzyme, then dehydrated to form phosphoenolpyruvate (PEP). PEP is dephosphorylated to pyruvate, generating 1 ATP molecule per reaction, and creating a net “payoff ” of two ATP molecules per glucose during glycolysis. It is important to note the generation of NADH from NAD+ during the phosphorylation of GADP. For further glycolysis reactions to take place, oxidized NAD+ must be regenerated. In the presence of oxygen, NADH is reduced to NAD+ by the electron transport chain (ETC), generating totally 38 ATP molecules from one molecule of glucose (Garrett and Grisham, 2005). In the absence of oxygen, NAD+ is replenished through the generation of fermentative end products like lactic acid, acetic acid, and ethanol (Madigan and Martinko, 2006). If these products are used as the terminal electron acceptors, the ETC no longer takes place, leaving only the EMP to produce two ATPs from one glucose molecule.

8.4 Fermenting Species The set of organisms capable of fermenting five- and six-carbon sugars is enormous, spanning all three domains of life: archaea, bacteria, and eukarya. They occupy a great variety of environmental niches, and there is great variation in growth conditions, substrates utilized for fermentation, and fermentation end products. A complete description of fermentative organisms is impractical for the purposes of this text; rather we will endeavor to provide a brief overview of significant genera and species focusing on those of particular relevance. Table 8.2 summarizes the various microbes that ferment 5- and 6-carbon sugars or utilize them for other useful products, respectively.

P1: SFK/UKS P2: SFK c08 BLBS075-Hood December 15, 2010

21:54

Trim: 244mm×172mm

glucose ATP HK ADP glucose-6-P

Galactose

PGI

Mannose Fructose

fructose-6-phosphate ATP

PFK

ADP FBP

Intermediates shared with PPP

FBPA

DHAP

[2] Glyceraldehyde-3-P

TPI

2NAD+ GAPDH 2NADH [2] BPG

2ADP PGK 2ATP [2] 3-PG PGAM [2] 2-PG ENO 2H2O [2] PEP 2ADP PYK 2ATP [2] pyruvate

Glycolysis Enzymes Abbreviation

Enzyme name

HK PGI PFK FBPA

Hexokinase Phospho -glucoisomerase Phospho -fructokinase Fructose bisphosphate aldolase Triose phosphate isomerase Glyceraldehyde-3-phosphate dehydrogenase Phosphoglycerate kinase Phosphoglycerate mutase Enolase Pyruvate kinase

TPI GAPDH PGK PGAM ENO PYK

EC# 2.7.1.1 5.3.1.9 2.7.1.11 4.1.2.13 5.3.

1.1

1.2.1.12 2.7.2.3 5.4.2.1 4.2.1.11 2.7.1.40

Intermediates Abbreviation

Intermediate name

FBP DHAP BPG 3-PG 2-PG PEP

Fructose-1,6-bisphosphate Dihydroxyacetone phosphate 1,3 Bisphospho-glycerate 3-Phosphoglycerate 2-Phosphoglycerate Phosphoenolpyruvate

Figure 8.3. Overview of metabolic interactions involving hexose and pentose sugars. Hexose sugars are initially metabolized by the EMP, commonly known as glycolysis. When oxygen is limited to the cell and respiration cannot proceed, anaerobic fermentation allows the cell to regenerate NAD+ through the production of ethanol or other endogenous electron acceptors. Pentose sugars enter the metabolism of the cell through the pentose phosphate pathway (PPP). High energy intermediates from the PPP can enter into the glycolytic pathway, allowing for further production of ATP molecules. Common intermediates between both pathways are displayed in pink; monomeric sugars are displayed in black. Solid arrows indicate direct enzymatic reaction and dashed arrows indicate multiple enzymatic steps (Arskold et al., 2008; Garrett and Grisham, 2005; Karhumaa et al., 2005; Sonderegger et al., 2004b; Wisselink et al., 2007a).

P1: SFK/UKS P2: SFK c08 BLBS075-Hood December 15, 2010

21:54

Trim: 244mm×172mm

L-arabinose AKR1 L-AOL

AKRA

AOL4D D-xylose XylA D-XU

AKR1

L-XU

XOL

L-RU

L-XUR

AraB

glucose-6-P

L-RU-5-P

D-XUR

G6PD

XUK

PGLS AraD

D-XU-5-P

6-PGN

RPE

PK

PGD D-RU-5-P

acetyl-P

RPE

PP1

PTA acetyl-CoA

RB-5-P

D-XU-5-P

AADH Tkt

acetaldehyde

glyceraldehyde-3-P

S-7-P Tal

fructose-6-P Intermediates shared with Glycolysis

E-4-P D-XU-5-P Tkt fructose-6-P

glyceraldehyde-3-P

Pentose pathways Abbreviation G6PD PGLS PGD RPE AraD AraA AKR1 AOL4D L-XUR XylA XK D- XUR AraB RpiA RPE Tkt Tal

Enzymes Enzyme name

EC#

Glucose-6-phosphate Dehydrogenase 6-phosphogluconlactonase 6-phosphogluconate Dehydrogenase Ribulose-phopshate 3-epimerase L-rubulose-5-phosphate 4-epimerase L-arabinose isomerase Aldehyde reductase L-arabinitol 4-de hydrogenase L-xylulose reductase Xylose isomerase Xylulokinase D-Xylulose reductase L-ribulokinase Ribose 5-phosphate isomerase A Ribulose-phopshate 3-epimerase Transketolase Transaldolase

1.1.1.49 3.1.1.31 1.1.1.44 5.1.3.1 5.1.3.4 5.3.1.4 1.1.1.21 1.1.1.12 1.1.1.10 5.3.1.5 2.7.1.17 1.1.1.9 2.7.1.16 5.3.1.6 5.1.3.1 2.2.1.1 2.2.1.2

Intermediates Abbreviation

Intermediate Name

L-AOL L-XU L-RU L-RU-5-P XOL D-XU D-XU-5-P 6-PGN D-RU-5-P RB- 5- P S-7-P E-4-P

L-Arabinitol L-Xylulose L-Ribulose L-Ribulose-5-phosphate Xylitol D-Xylulose D-Xylulose-5-phosphate 6-Phosphogluconate D-Ribulose-5-phosphate D-Ribose-5-phosphate Sedoheptulose-7-phosphate Erythrose-4-phosphate

Figure 8.3. (Continued) 163

164 Classification Bacteria

Bacteria, obligate anaerobe, autotroph Fungi (yeast), facultative anaerobe Bacteria, aerobe, hyperthermophile

Bacteria, anaerobe, thermophile

Bacteria, facultative anaerobe Bacteria, autotroph, anaerobe Fungi (yeast), thermotolerant Fungi (yeast) Fungi (yeast)

Acetivibrio cellulolyticus

Acetobacterium woodii

Ambrosiozyma monospora

Bacillus caldolyticus

Bacillus stearothermophilus

Bacillus subtilis

Butyribacterium methylotrophicum

Candida acidothermophilum

Candida lusitaniaexiv

Candida shehatae

Ethanol (<4.1 g/L on arabinose)

Acetate (3 mol/mol hexose), ethanol

H2 , acetate, ethanol (3.70, 0.73, 0.39 mol/mol cellulose, respectively)

Products Formed

Lactate, Acetate, 2,3-Butanediol (23.3 mM, 16.4 mM, 16.7 mM, respectivelyx ), ethanol

Ethanol (0.30 g/g sucroseviii )

Collection (2009), Lynd and Zeikus (1983), and Zeikus et al., (1980b)

60 min

1200 minxiii

∼7.0

∼7.0

65

30xi

Glu, Man, Gal, Xyl, Mal, trehalose

Glu, Xyl, Cbs

Glu

Ethanol (maximum reported value 6.6 g/L xylose)

Ethanolxv

Ethanol (80% efficiency on glucose)

30

35

40

4–5.5

6.8–7.0

5.0

7.2

Collection (2009), Cruz Ramos et al. (2000), Mindich (1970), and Nakano et al. (1997)

30 minix

6.9–7.1

70vii

H2 -CO2 , methanol, Acetate, butyric acid (25 mM), 37 Glu, Fru, lactate, H2 , butyratexii pyruvate

Glu, pyruvate

Most hexoses and pentoses, including Xyl and Ara, starch

15 min

7.0

35–37vi

Barnett et al., Du Preez et al. (1986), Jeffries and Kurtzman, and Preez et al. (1986)

Collection (2009) and Maleszka et al. (1982)

Kadam and Schmidt (1997), and Lynd and Zeikus (1983)

Atkinson et al. (1975), Hartley and Payton (1983), Martin et al. (1992), Nanmori et al. (1990), Payton (1984), Pennock and Tempest (1988), and Sharp et al. (1980),

Emanuilova and Toda (1984), Heinen and Heinen (1972), Heinen and Lauwers (1981), Sharp et al. (1980), and Wiegel et al. (1985)

Collection (2009), Dien et al. (1996), and Endoh et al. (2008),

Balch et al. (1977) and Buschhorn et al. (1989)

360 min

5.0v

Patel and Mackenzie (1982)

References

30

Gen. Time 240 min

Opt. pHiii 6.7

35

Opt. Temp. (◦ C)ii

21:54

Glu, starch, Dex, Acetic acid glycerol, laevulose, Mal, mannitol, Man, raffinose, Suc

Ara, Xyl, erythritol, xylitol

Glu, lactate, Fru, glycerate, fumarateiv , formic acid

Glu, Cbs, Cel

Fermentation Substratesi

A partial list of notable 5- and 6-carbon sugar fermentating organisms, with fermentation substrates, products, and growth conditions.

Organism

Table 8.2.

P1: SFK/UKS P2: SFK c08 BLBS075-Hood December 15, 2010 Trim: 244mm×172mm

Suc

Glu, most hexoses, starch

Bacteria, anaerobe

Bacteria, anaerobe

Bacteria, anaerobe

Bacteria, facultative anaerobe Bacteria, obligate anaerobe

Bacteria, anaerobe Bacteria, anaerobe

Clostridium acetobutylicum

Clostridium beijerinckii

Clostridium butylicumxxiv

Clostridium formicoaceticum

Clostridium kluyveri

Clostridium papyrosolvens

Clostridium pasteurianum

37

Ethanol Lactate, ethanol, butanol, 1,3-propanediol

Glycerol, CbsCbsxxxi Glu, glycerolxxxii

37

6.4xxx

35–37xxix

Caproic acid, butyrate, caproate, and hydrogen (1.04, 1.46, 1.11 mM from 4.55 mM ethanol and 2.18 mM acetate, respectively), hexanolxxviii

Requires ethanol and a lower fatty acidxxvii

7.0

7.0

8.0–8.5

Fru, hexonic & hexauronic acids, fumarate, malate

37

150 min 30 minxxxiii

George and Chen (1983), George et al. (1983), and Yan et al. (1988)

5.0–6.8xxiii

35

Acetic acid (3 mol/mol fructosexxv ), acetate, formatexxvi , succinate

Bahl et al. (1982), Collection (2009), Gunsalus and Stanier, and Zeikus (1980)

7.0xx

37

(Continued)

Collection (2009), Dabrock et al. (1992), Heyndrickx et al. (1991), and Nakas et al. (1983),

Palop et al. (1989)

Barker and Taha (1942), Bornstein and Barker (1948), Gunsalus and Stanier, Kenealy and Waselefsky (1985), and Thauer et al. (1968)

Andreesen et al. (1970) and Dorn et al. (1978)

Gunsalus and Stanier

Collection (2009), Gong et al. (1981, 1983), and Jeffries (1981)

7.0

24–26

5.0–6.8

Butanol, acetone, acetoin, acetate, butyrate, H2 (0.60, 0.22, 0.06, 0.14, 0.04, 0.78 mol/mol glucose, respectively)

N-butanol, acetone, isopropanolxxi (10.2, 69.6, 9.8 mmolxxii )

Ethanol, acetone, acetoin, acetate, butyrate, butanol, H2 (0.07, 0.22, 0.06, 0.14, 0.04, 0.56, 1.35 mol/mol glucose, respectively)xix

Xylitolxvii , ethanol (8.28 g/L xylosexviii ), arabitol

21:54

35

Most hexoses, most pentoses, starch

Fungi (yeast)

Candida tropicalis

Glu, Fru, Suc, Xyl (aerobic utilization), xylulose

see Kluveromyces marxianus

Candida pseudotropicalisxvi

P1: SFK/UKS P2: SFK c08 BLBS075-Hood December 15, 2010 Trim: 244mm×172mm

165

166 Classification Bacteria, obligate anaerobe

Bacteria, anaerobe Bacteria, anaerobe, thermophile Bacteria, anaerobe, thermophile

Bacteria, anaerobe, thermophile

Bacteria, anaerobe, thermophile

Clostridium propionicum

Clostridium sordellii

Clostridium thermoaceticum

Clostridium thermocellum

Clostridium thermohydrosulfuricum

Clostridium thermosaccharolyticum

(Continued)

Organism

Table 8.2.

Most hexoses and pentoses, including Xyl, Ara, and Cbs as well as starch, xylan, acetate, pyruvate

Most hexoses and pentoses, including Xyl, Ara, and Cbs as well as starch, xylan, pectin

4.7–7.6

58–64

Ethanol (1.1 mol/mol glucosexl ) also produces acetic acid, lactic acid, butyrate, and H2 xli

90 min

75 min

4.8–9.7xxxix 65–70

Ethanol, acetic acid, lactic acid, H2 (1.95, 0.11, 0.02, 0.11 mol/mol glucose, respectively)

Eller and Ordal (1972), Hsu and Ordal (1970), Landuyt and Hsu (1985), Lee and Ordal (1967), Ljungdahl (1979), and Sjolander (1937)

Chiang et al. (1981), Ng et al., (1979), Reilly and Hollaender (1981), Wiegel (1980), and Wiegel et al. (1979),

Avgerinos et al. (1981), Chiang et al. (1981), Enebo (1951), Fontaine et al. (1942), Lee and Blackburn (1975), Macfadyen and Blaxall (1896), Mcbee (1948), Pringsheim (1912), and Ng et al. (1979)

130 minxxxviii

7.5xxxvii

55–60

Ethanol (0.3 mol/mol cellulose), acetate, lactate (0.67, 0.71, 0.10 mol/mol glucose, respectively), H2

Andreesen et al. (1973), Fontaine et al. (1942), and Zeikus (1980)

6.0

Collection, 2009) and Corry (1978)

Akedo et al. (1983), Cardon and Barker (1946), Gunsalus and Stanier, Johns (1952), and O’brien et al. (1990)

References

60

Gen. Time

Acetate, acetic acid (2.41 mol/mol glucose), formic acid (0.4 mol/mol glucose)

6.8

7.0–7.4

Opt. pHiii

37

37

xxxv

Opt. Temp. (◦ C)ii

Ethanol (1.7 mol/mol glucose)

Propionic acid, acrylic acid

Products Formed

21:54

Most hexoses, Cbs, starch, Cel, sorbitol

Most hexoses and pentosesxxxvi , including Xyl and Ara, starch

Glu

Alanine, lactate, pyruvate, serine, threonine, cheese whey, lactic acidxxxiv

Fermentation Substratesi

P1: SFK/UKS P2: SFK c08 BLBS075-Hood December 15, 2010 Trim: 244mm×172mm

24–26

Ethyl acetate (0.85% v/vlxii from glucose), ethanol, isoamyl alcohol, isoamyl acetate

Most hexoses and pentoses, including Ara; soluble starch; sodium acetate, ethanol

Fungi (yeast), facultative anaerobe, human pathogen

Hansenula anomalalxi

24

Ethanol (4.3% v/v from glucose, 2.5% v/v from xyloselviii , 25 g/L or 0.50 g/g xyloselix ), acetic acidlx

Glu, Xyllvi , Man, Gal, Suc, xylitol, Cellvii , hemiCels, starch

Fungi (yeast)

Fusarium oxysporumlv

6.8–7.0

5.6

(Continued)

Bedford (1942), Gray (1949), Klein et al. (1988), Plata et al. (2003), Wickerham and Burton (1962), and Yoshioka and Hashimoto (1981)

Atlas (2004), Collection (2009), Chiang et al. (1981), Gibbs et al. (1954), Kuhad et al. (1994), Laskin et al. (2009), Singh and Kumar (1991), Suihko and Enari (1981), and White and Willaman (1928)

Belaich and Belaich (1976b, 1976a), Brau ¨ and Sahm (1986), Clark (1989), Fotadar et al. (2005), Gale and Epps (1942), Hernandez and Johnson (1967), Madigan et al. (2000), and Maloy et al. (1994)

Collection, (2009), Dickey (1979), Samson et al. (2005), and Tolan and Finn, (1987)

90 minxlix

20 minliv

Collection (2009), Kraght and Starr (1952), and Tolan and Finn (1987)

90 minxlvi

Collection (2009), Martinec and Kocur (1964), Sutton and Starr (1959), and Tolan and Finn (1987)

21:54

7.1–7.5liii

37lii

Ethanoll , formic acid, acetic acid, succinic acid (0.85, 1.12, 0.72, 0.12 mol/mol glucose, respectivelyli )

Glu, Gal, Fru, Lac

Bacteria, facultative anaerobe

Escherichia coli

7.0

26–30

7.0xliv

Ethanol (0.72 mol/mol xylosexlviii )

Glu, Xyl, Ara

Bacteria, facultative anaerobe

Erwinia chrysanthemixlvii

26–28xliii

7.0

Lactic acid, succinic acid, 2,3-butylene glycol, ethanol, acetic acid, formic acid

Ethanol, lactic acid, acetic acid, formic acid, succinic acid, acetoin, 2,3-butanediol (1.5, 0.18, 0.07, 0.06, 0.02, 0.01, 0.01 mol/mol glucose, respectively) 26

Glu, Xyl, Ara, galactouronic acid

Bacteria, facultative anaerobe, plant and human pathogen

Erwinia carotovoraxlv

Glu, Suc, Gal, Fru, Man, sorbitol, mannitol, raffinosexlii

Bacteria, facultative anaerobe, plant pathogen

Erwinia amylovora

P1: SFK/UKS P2: SFK c08 BLBS075-Hood December 15, 2010 Trim: 244mm×172mm

167

168 Glycerol, Xyl, citrate, Glu, Suc, Fru

Glu, Gal, Lac, inulin, capable of growth on whey

Bacteria, facultative anaerobe

Fungi (yeast), opportunistic human pathogen

Fungi (yeast)

Bacteria, anaerobe

Bacteria, anaerobe

Klebsiella pneumoniae

Kluyveromyces fragilislxviii

Kluyveromyces marxianus

Lactobacillus brevis

Lactobacillus casei

Glu, mannitol, citrate, Cbs (most hexoses and pentoses)

Most hexoses, pentoses

Lactate (1.6 mol/mol glucose), ethanol (1.29 mol/mol mannitol), acetate (1.96 mol/mol citrate), formate (1.6 mol/mol mannitol)lxxv

Acetic acid, lactic acid, ethanol, glycerol (0.15, 0.83, 0.74, 0.32 mol/mol glucose, respectively)

Ethanol (24.4 g/L on glucoselxxii , 5.6 g/L on xyloselxxiii )

Ethanol (49.0 g/L on glucoselxix , 10.6% v/v yieldlxx on whey)

1,3-propanediol (96% theoretical yield, 48.5 g/L on glycerol), 2,3-butanediol, ethanol, hydrogen gas, acetate, butyrate, butanollxvii

Ethanol (46.4 g/L/0.37 g/g on glucoselxiii , 45.2 g/L cellobioselxiv ; 0.42 g/g xylose, 0.34 g/g arabinoselxv )lxvi

Products Formed

6.3

5.5

30

37

5.8

30

6.0–6.5

7.0

37

30

7.3

Opt. pHiii

37

Opt. Temp. (◦ C)ii

60 minlxxvi

Barnett et al., Margaritis and Bajpai (1982), Marwaha and Kennedy (1984), Postma and Van Den Broek (1990), and Vallet et al. (1996)

345 minlxxiv

De Vries et al. (1970)

Collection (2009), Gunsalus and Stanier, and Man et al. (1960)

Aaron et al. (1958), Bernstein et al. (1977), Friend and Shahani (1979), Janssens et al. (1983), Lutwick et al., (1980), Sachs et al. (1981), and Vallet et al. (1996)

Biebl et al. (1998, 1999), Bott et al. (1995), Collection (2009), Kelker et al. (1970), Menzel et al. (1997), Nishikawa et al. (1988), Sprenger and Lengeler (1988), and Zeng et al. (1993)

Bothast et al. (1994), Collection (2009), Dien et al. (2003), Doten and Mortlock (1985), and Wood and Ingram (1992)

References

240 minlxxi

Gen. Time

21:54

Xyl, Glu, Fru, Gal, Mal, Suc, trehalose, Lac

Most hexoses and pentoses, including Cbs and cellotriose

Bacteria, facultative anaerobe

Klebsiella oxytoca

Fermentation Substratesi

Classification

(Continued)

Organism

Table 8.2.

P1: SFK/UKS P2: SFK c08 BLBS075-Hood December 15, 2010 Trim: 244mm×172mm

Bacteria

Fungi (yeast)

Fungi (yeast) Fungi (yeast), facultative anaerobe

Fungi (yeast)

Bacteria, obligate anaerobe

Leuconostoc mesenteroides

Pachysolen tannophilus

Pichia stipitis

Saccharomyces cerevisiae

Saccharomyces diastaticus

Sarcina maxima

4.5–6.0

5.4

30lxxxv

25

30

Ethanol (1.45, 1.25% v/v on dextrin, and starch respectively; 60 g/L max on glucoselxxxvi ) H2 (2.23 mol/mol glucose, 0.52 mol/mol pyruvate), butyric acid, acetic acid, succinic acid, formic acid (0.77, 0.40, 0.05, 0.04 mol/mol glucose, respectively), acyl phosphate (0.66 mol/mol pyruvate), lactic acidlxxxvii

Glu, pyruvate, formate

6.8–7.0

90 min

140 minlxxx

<130 min

(Continued)

Canale-Parola (1970), Kupfer and Canale-Parola (1967, 1968), and Smit (1933)

Collection (2009), D’amore et al. (1989), Demain and Solomon (1985), and Laluce and Mattoon (1984)

C¸aylak and Vardar Sukan (1998), Coulson and Richardson (1954), Da Cruz et al. (2003), Demain and Solomon (1985), Ergun and Ferda Mutlu (2000), Krishnan et al. (1999), Krouwel and Braber (1979), Kuyper et al. (2005), Nagodawithana and Steinkraus (1976), Nevoigt (2008), Nissen et al. (2000), Vallet et al. (1996), Swanson and Clifton (1948), and Wasungu and Simard (1982)

Barnett et al. Du Preez (1994), and Slininger et al. (2006)

Alexander and Usda (1985), Barnett et al., Dekker (1982), Jeffries and Kurtzman, Slininger et al. (1985), and Wickerham (1951)

Corry (1978), Gaines and Stahly (1943), and Mccleskey et al. (1947)

21:54

Glu, Fru, Gal, Mal, maltotriose, Suc, starch, Dex

Suc, Glu, Fru, Gal, Mal, maltotriose, xylulose, Dex, starch, raffinoselxxxii Ethanol (47.9 g/L on glucoselxxxiii up to 91.8 g/L on glucose, 96.7 g/L on sucrose, 40.0 g/L on galactose, 18.4 g/L on molasseslxxxiv ), glycerol, organic acids; see Nevoigt (2008)

6.5–7.0

25–30

Ethanol (61 g/L on xyloselxxxi )

Glu, Xyl, Gal, Mal, trehalose

4.5lxxix

32

Ethanol (2.1 g/L on xyloselxxviii )

6.8

Glu, Xyl, Cbslxxvii

25–30

Ethanol (1.1 mol/mol glucose)

Most hexoses and pentoses, including Ara, Xyl, and Cbs; starch

P1: SFK/UKS P2: SFK c08 BLBS075-Hood December 15, 2010 Trim: 244mm×172mm

169

170 Classification Bacteria, anaerobe

Fungi (yeast), facultative anaerobe Bacteria, facultative anaerobexc

Bacteria, obligate anaerobe

Bacteria

Sarcina ventriculi

Schizosaccharomyces pombe

Spirochaeta aurantia

Spirochaeta litoralis

Streptococcus lactis

(Continued)

Organism

Table 8.2.

Glu, Gal, Lac, citratexciv , Ara, Xyl, Mal, mannitol, salicin

30

31.5c

Ethanolxcv , lactatexcvi , acetatexcvii (2.00, 0.07 mol/mol glucose, respectively)xcviii , formatexcix

30

25–30

30–37

Opt. Temp. (◦ C)ii

Ethanol, acetate, lactate (1.40, 0.57, <0.01 mol/mol glucose), formate, pyruvatexciii

Ethanol, H2 , acetic acid, formic acid, lactic acid (1.50, 1.07, 0.69, 0.05, 0.01 mol/mol glucose), acetoin, diacetylxci

Most hexoses and pentoses, including Ara, Xyl, and Cbs; starch; glycerol Glu, pyruvate

Ethanol, acetoin, acetic acid

Ethanol, H2 , acetate, acetoin (1.71, 0.41, 0.90 0.04 mol/mol glucose, respectively), acetyl phosphate, acetaldehydelxxxviii

Products Formed

7.0

7.5–7.6

7.0–7.3

5.5–6.0

7.8

lxxxix

Opt. pHiii

42 min

230 minxcii

Gen. Time

Citti et al. (1965), Kempler and Mckay (1980), Lawrence and Thomas (1979), Lee and Collins (1976), Rahn et al. (1938), Sherman (1937), and Thomas et al. (1979, 1980)

Canale-Parola (1977) and Hespell and Canale-Parola (1970a, 1970b, 1973)

Breznak and Canale-Parola (1969) and Canale-Parola (1977)

Collection (2009), Mayer and Temperli (1963), Mcveigh and Bracken (1955),Wang et al. (1980), and Ueng et al. (1981)

Bauchop and Dawes (1968), Collection (2009), Corry (1978), Cysewski and Wilke (1978), and Stephenson and Dawes (1971)

References

21:54

Glu, malate, oxalacetate, pyruvate, xylulose, Xyl

Most hexoses

Fermentation Substratesi

P1: SFK/UKS P2: SFK c08 BLBS075-Hood December 15, 2010 Trim: 244mm×172mm

Bacteria, thermophile, obligate anaerobe Bacteria, obligate anaerobe

Thermobacteroides acetoethylicus

Zymomonas mobilis

Glu, Suc, Fru, raffinose, sorbitolciv

Glu, Man, Cbs, Lac, Mal, Suc, starch

Glu, Xyl, Mal, Fru, Man, Suc, ribose, Lac, Gal, xylulose, starch, Cbs 5.5–8.5

5.0–7.0cvi

65

30cv

Ethanol, acetate, H2 , isobutyrate, butyrate (1.39, 1.39, 2.85, 0.02, 0.09 mol/mol glucose, respectively) Ethanol (1.93 mol/mol glucose, 0.46 g/g on amylase treated starch), acetic acid, lactic acid

5.8–8.5cii

40–77ci

Ethanol (1.95 mol/mol glucose), lactic acid, acetic acid

120 min

25 min

120 minciii

Brau ¨ and Sahm (1986), Karsch et al. (1983), Millichip and Doelle (1989), Rogers et al. (1982), and Swings and De Ley (1977)

Ben-Bassat and Zeikus (1981) and Zeikus et al. (1980a)

Carreira et al. (1984), Wiegel and Puls, Ljungdahl (1979), and Wiegel and Ljungdahl (1981)

21:54

i Substrate abbreviations: Glu, glucose; Fru, fructose; Ara, arabinose; Xyl, xylose; Gal, galactose; Mal, maltose; Man, mannose; Dex, dextrin; Suc, sucrose; Cel, cellulose; Cbs, cellobiose; Lac, lactose. ii Refers to optimum growth temperature unless otherwise noted. iii Refers to optimum growth pH unless otherwise noted. iv After adaptation. v This is not necessarily the optimal pH; however, the organism is viable after several weeks at this pH. vi Optimal ethanol production. Optimal growth: 24◦ C–26◦ C vii Growth is possible up to 105◦ C. viii Data reflects a genetically modified strain. ix At 70◦ C. x After 10 h, 37◦ C, 50 mM glucose, 50 mM pyruvate. xi Range: 25–30◦ C. xii Butyrate formation: 0.3 mol/mol methanol. xiii Growth on 100 mM methanol, 50 mM sodium acetate. xiv Also known as Clavispora lusitaniae. xv Ethanol is yielded from glucose, xylose, and cellobiose, with the greatest yield coming from coferementation of xylose and cellbiose. xvi Candida pseudotropicalis is the commonly used but obsolete name for the anamorph of Kluveromyces marxianus, the term for the teleomorph and thus the term for the holomorphic form of the fungus. The currently accepted name for the anamorph is Candida kefyr. xvii Xylitol is only yielded during growth on xylose. xviii 75 g/L xylose culture. xix Fermentation products vary from acids at >5.0 pH to solvents as pH decreases. xx Grows well in media having pH values from 4.2 to 8.0. xxi Isopropanol is produced by several strains in lieu of acetone. xxii Produced from a defined medium (20 g/L glucose)

Bacteria, thermophile, anaerobe

Thermoanaerobacter ethanolicus

P1: SFK/UKS P2: SFK c08 BLBS075-Hood December 15, 2010 Trim: 244mm×172mm

171

172

(Continued)

xxiv C.

produce solvents over a wide range of pH values, including non-acidic ranges. butylicum is no longer regarded as a distinct species but rather a strain of C. beijerinckii. xxv During growth phase. xxvi During stationary phase. xxvii Such as acetate, propionate, or butyrate. Clostridium kluyveri is notably unable to ferment carbohydrates. xxviii Hexanol is yielded during fermentation with ethanol and succinate 4-hydroxybutyrate and 3-butenoate. xxix Range: <25◦ C–43◦ C. xxx pH range: 5.2–>8.0. xxxi Optimal cellobiose concentration found to be 6 g/L. xxxii Clostridium propionicum will ferment algal biomass if supplemented with glycerol. xxxiii During butyrate and acetate production growth at optimal conditions, during conditions inducing acetone production this increases to 13 hours. xxxiv Notably unable to ferment sugars. xxxv Little difference observed between 25◦ C and 37◦ C. xxxvi Clostridium thermoaceticum will preferentially use xylose, then fructose, then glucose. xxxvii This pH reflects the final pH of the defined medium produced for C. thermocellum. Substantial pH drops are observed during culturing, dropping to 6.8 after 100 hours of fermentation. xxxviii Min doubling time (2.1 hours) occurs on cellobiose, growth on cellulose results in a doubling time of 11 hours. xxxix A pH greater than 5.2 is necessary for optimal ethanol production. Strangely, ethanol production only occurs if the pH is allowed to drop during fermentation. Constant pH results in low ethanol production. xl Ethanol is produced chiefly during sporulation phase. xli Acetic acid, lactic acid, butyrate, and hydrogen gas are produced from xylose and glucose during growth. xlii Raffinose utilization, strain-specific. xliii Minimum temperature as low as 3◦ C, maximum 37◦ C. xliv Range: 4.0–8.8 xlv Now officially known as Pectobacterium carotovorum. xlvi Generation time of 90 min reflects growth on glucose. Growth on arabinose results in a generation time of 240 min and growth on xylose results in a generation time of 480 min. xlvii Erwinia chrysanthemi was reclassified as Pectobacterium chrysanthemi, and then reclassified again in 2005 as Dyckeya chrysanthemi. xlviii This value is nearly doubled to 1.45 mol/mol when an transgene from Zymomonas mobilis coding for a pyruvate decarboxylase is inserted. Another transgenic construct achieved ethanol production on xylose and arabinose at 0.45 and 0.33 g/g sugar, respectively (see Tolan & Finn, 1987). xlix Generation time of 90 min reflects growth on glucose. Growth on arabinose results in a generation time of 240 min, growth on xylose results in a generation time of 540 min. l Ethanol yield increased variously by transgenes, including a 4-fold improvement to 1.66 mol/mol glucose (0.41 g/g glucose) yield with pyruvate decarboxylase transgene from Zymomonas mobilis. li Data reflects values obtained during exponential growth in the wild-type. At glucose exhaustion, yields of ethanol, formic, acetic, and succinic acid is 0.87, 1.10, 0.73, and 0.11 mol/mol, respectively.

xxiii Will

Table 8.2.

P1: SFK/UKS P2: SFK c08 BLBS075-Hood December 15, 2010 21:54 Trim: 244mm×172mm

liii Little

occurs at temperatures as high as 45. doubling time variation between pH of 6 and 8. liv With aeration. Totally anaerobic growth has a doubling time of 50 min. lv Former species Fusarium lini is now subsumed by F. oxysporum to F. oxysporum subspecies lini. lvi Fusarium oxysporum will utilize xylose if grown in high-xylulose mixture. lvii Cellulose, hemicellulose, and starch fermentation found in Fusarium oxysporum strain DSM 841, producing 12 g/L acetic acid and 3.6 g/L ethanol from potato waste, see Singh et al. (1991). lviii From a 5% w/v xylose solution after 144 h. lix 25 g/L produced by Fusarium oxysporum strain VTT-D-80134, max productivity 0.17 g/L/h. See Suihko and Enari (1981). lx Acetic acid is fermented at an equimolar rate with xylose. lxi Also known as Candida pelliculosa, Candida beverwijkiae, Pichia anomala, Wickerhamomyces anomalus. lxii Hansenula anomala has an aeration requirement as evidenced by the effect of the ratio of culture surface area to culture volume on ethyl acetate yield. Optimal ratio is 32:100 area:volume. Anaerobic production of ethyl acetate is increased by a factor of 100 if ethanol is present in the solution. lxiii 10% glucose media. lxiv 1% cellobiose media. lxv Ethanol production on xylose and arabinose refers to data from genetically modified strains, see Bothast et al. (1994). lxvi A mutant expressing a D-arabitol permease system and D-arabitol dehydrogenase constitutively, but lacking D-xylulokinase and D-xylose isomerase was found to produce D- and L- Xylulose from D-arabitol, see Doten and Mortlock (1984). lxvii Lactate and succinate are also present as minor products. lxviii Formerly Saccharomyces fragilis. lxix Ethanol production on glucose data reflects fermentation of 120 g/L glucose media at 30◦ C for 192 h. lxx Data reflects fermentative growth on whey with a 20% lactose supplement. lxxi 180 min achieved with lipid supplementation. lxxii Ethanol production on glucose data reflects fermentation of 50 g/L glucose media at 30◦ C for 40 h. lxxiii Data reflects fermentative growth on 0.28 g/L xylose. lxxiv Growth on xylose. lxxv Lactobacillus casei produces a variety of products as the result of fermentation, which depend strongly on the substrates being used; listed here are the substrates tested to yield the maximum amount of the product in question. lxxvi Anaerobic growth on glucose. lxxvii Growth on cellobiose characterized by the presence of the Kluyver effect: growth, but no production of ethanol. lxxviii 24% efficiency producing ethanol from xylose, maximum reported production rate is 0.12 g/g/h. lxxix The pH value of 4.5 reflects the optimal pH for growth, whereas a pH of 2.5 is optimal for ethanol production. lxxx Doubling time increases to 41 h when grown on xylose. See Dekker (1982). lxxxi Data reflects growth on 150 g/L xylose media, in mineral and nitrogen optimized media. 33.2 g/L ethanol reported from 90 g/L xylose medium. lxxxii Saccharomyces cerevisiae has also been successfully engineered to use lignocellulosic sugars such as xylose and arabinose. lxxxiii Data reflects ethanol content after 36 h of fermentation. The yield reported was 0.46 g/g sugar, a 90% yield. Maximum specific productivity measured at 2.17 g/L/h. Engineered strains have been developed to produce a large diversity of more complex metabolic end products, including biopolymers and vitamins for human use.

lii Growth

P1: SFK/UKS P2: SFK c08 BLBS075-Hood December 15, 2010 21:54 Trim: 244mm×172mm

173

174

(Continued)

lxxxiv Glucose data from fermentation on 200 g/L glucose for 94 h at 30◦ C. Sucrose data from fermentation on 220 g/L sucrose for 96 h at 28◦ C. Galactose data from fermentation on 150 g/L galactose for 60 h at 30◦ C. Molasses data from fermentation on 5 g/L glucose for 24 h at 30◦ C. lxxxv Despite the optimal growth temperature of 30◦ C, growth at 39◦ C results in an improved ethanol production to biomass ratio. lxxxvi Data reflects production of ethanol from glucose fermentation at 40◦ C. A strain produced as a fusion product between S. diastaticus and S. uvarum produces more than 80 g/L ethanol maximally from glucose, see references. lxxxvii The presence of lactic acid in the fermentation products of S. maxima are substantially strain-dependent: strains tested by Smit exhibit lactic acid product, while those tested by Kupfer and Canale-Parole did not produce lactic acid in detectable amounts. lxxxviii Acetaldehyde and acetate are minor products. lxxxix The pH value of 7.8 represents the optimal pH for production of acetate from glucose, and maximal hydrogen gas production occurs at a pH of 6.9. xc Spirochaeta aurantia is the only known facultative anaerobe in the genus. xci Acetoin and diacetyl are minor products. xcii Aerobic growth. xciii Formate and pyruvate are minor products produced during glucose fermentation. xciv Citrate fermentation observed in Streptococcus lactis subspecies dacetylactis. xcv Ethanol is not formed during glucose fermentation. On alternative sugars it is produced: 0.35 mol/mol galactose, 0.68 mol/mol lactose. xcvi Lactate is also produced on alternative sugars: 1.92 mol/mol galactose, 1.99 mol/mol lactose. xcvii Acetate is also produced on alternative sugars: 0.49 mol/mol galactose, 0.83 mol/mol lactose. xcviii Considerable production variation exists between strains, and strain maxima are listed. xcix Formate is not formed during glucose fermentation. On alternative sugars it is produced: 0.47 mol/mol galactose, 0.77 mol/mol lactose. c Maximal temperature 42◦ C. ci Little variation exhibited over entire range. cii Little variation exhibited over entire range. ciii Generation time unchanged on both xylose and glucose. civ Raffinose and sorbitol fermentations are hugely strain specific in Zymomonas mobilis. cv 30◦ C is optimal for ethanol production; however, there is at least one report of fermentation taking place at 37 equally well. cvi A pH as high as 7.5 and as low as 3.85 with <10% loss of viability. cvii The annual production is as of 2006 and is not displayed if the value is not known. cviii The majority of species listed in Table 8.2 have the ability to ferment ethanol, and these species are the leading candidates for industrial bioethanol production. cix Currently, the leading technology for production of vitamin C, or L- ascorbic acid (L-AA), is the so-called Reichstein process, which includes only one biological conversion step using Gluconobacter oxydans, and the remaining six steps are a mix of chemical and enzymatic conversions. cx This species has been genetically altered and does not directly convert glucose to vitamin C, rather 2-keto-L-gulonic acid (2KLGA) is the fermentation product. Subsequent enzymatic conversion is needed to make vitamin C

Table 8.2.

P1: SFK/UKS P2: SFK c08 BLBS075-Hood December 15, 2010 21:54 Trim: 244mm×172mm

P1: SFK/UKS P2: SFK c08 BLBS075-Hood December 15, 2010

21:54

Trim: 244mm×172mm

8 Fermentation Organisms for 5- and 6-Carbon Sugars

8.4.1

175

Brief Description of Major Species

Industrial processes that employ microbes focus overwhelmingly on a relatively small group of bacteria and fungi. Each group carries with it a particular set of advantages and disadvantages. The fungi include a major group known as the yeasts, which have been in use by man for many thousands of years and which are characterized by high ethanol production and tolerance. The bacteria are highly varied2 in terms of the rate at which the products of interest are generated as well as their tolerance to inhibitory compounds. Fungi Candida The yeast genus Candida is known for the species Candida albicans, a human pathogen (Odds, 1979) that causes candidiasis and yeast infections. Candida is a large genus, with some 200 known species (Kirsch et al., 1990), many of which are of great relevance to biofuels production. Several species, including C. lusitaniae (Maleszka et al., 1982), C. shehatae (Du Preez et al., 1986; Preez et al., 1986; Toivola et al., 1984), and C. tropicalis (Gong et al., 1981; Jeffries, 1981) (C. tropicalis is another noted human pathogen, affecting immunecompromised patients (Wingard et al., 1979) can utilize xylose.3 Utilization of cellobiose (by C. tenuis (Toivola et al., 1984)), mannose, sucrose, and fructose have also been observed in species in the genus. Candida produces ethanol as the main fermentation product, with C. acidothermophilum, a thermotolerant yeast, producing ethanol at 80% stoichiometric yield from glucose (Kadam and Schmidt, 1997). Kluyveromyces Kluyveromyces is a genus of budding yeast first characterized in 1956 and is characterized by its impressive fermentative capacity (Van Der Walt, 1956), having species capable of growing on galactose, lactose, inulin, fructose, whey (in the case of K. fragilis (Aaron et al., 1958)), and xylose (K. marxianus for instance (Margaritis and Bajpai, 1982)). K. marxianus has been shown to yield up to 5.6 g/L ethanol from fermentative growth on xylose, making it of interest to industrializing lignocellulosic ethanol production. However, the slow doubling time of growth on xylose makes industrialization difficult. Additionally, as with Candida, some of the species are known opportunistic pathogens in immune-compromised patients, making achieving GRAS (generally-regarded-as-safe) status, similar to that of S. cerevisiae, difficult (Lutwick et al., 1980). Pichia The genus Pichia contains the yeast Pichia stipitis and is of interest primarily because of its endogenous xylose metabolic pathway. P. stipitis is not the first yeast discovered to have this property; for instance, Pachysolen tannophilus was found to ferment xylose to ethanol in 1981 (Schneider et al., 1981). However, P. stipitis exhibits high ethanol production from xylose, with some strains producing nearly 6 g/L (67% of the theoretical yield of 9 g/L ethanol from a 2% xylose solution after 10 days) (Toivola et al., 1984). When oxygen optimization was considered, a maximal yield of 0.48 g/g (grams ethanol and xylose, respectively) was achieved, with a specific productivity of 0.20 g/L/h (Skoog and Hahn-Hagerdal, 1990) in a continuous culture.

P1: SFK/UKS P2: SFK c08 BLBS075-Hood December 15, 2010

176

21:54

Trim: 244mm×172mm

Plant Biomass Conversion

Additionally, research identifying the optimal pH and temperature for ethanol production from xylose by P. stipitis as 4–7 and 34◦ C, respectively, observed a maximal yield of 57 g/L when grown in a culture of 150 g/L xylose (Slininger et al., 1990). This puts P. stipitis well within range of the ethanol production rate of C. shehatae, another yeast known for its ability to produce large amounts of ethanol from xylose fermentation. Further, P. stipitis exceeds C. shehatae in terms of maximal production of ethanol from xylose. A study found P. stipitis capable of utilizing all but 7 g/L xylose from a 150 g/L xylose culture at 76% of the theoretical ethanol production yield, showing that ethanol recovery from cultures of P. stipitis fermenting xylose is economically viable (Slininger et al., 1985).

Saccharomyces Famed as the genus containing the brewer’s yeast,4 also known as baker’s yeast, S. cerevisiae, the genus Saccharomyces is of immense relevance to modern industrial processes. Its usage vastly predates the industrial revolution. S. cerevisiae has been employed by man as a leavening agent in bread making for at least five millennia (Maksoud et al., 1994) and as a means of producing ethanol in fermented beverages some four thousand years before that, dating back at least to the Neolithic period (McGovern et al., 2004). S. cerevisiae possesses a high tolerance to ethanol and a great propensity for its production (Swanson and Clifton, 1948), due at least in part to humanity’s selective breeding for these traits (Piˇskur et al., 2006). The genus has the capacity to ferment a great variety of substrates, including (in addition to glucose) fructose, galactose, maltose, sucrose, xylulose, dextrin, raffinose, and starch (Lodder, 1970; Rose and Harrison, 1969; Wang et al., 1980; Wickerham, 1951) and has been engineered to metabolize lignocellulosic5 sugars, such as xylose and arabinose (Becker and Boles, 2003; Hahn-Hagerdal, 2001; Karhumaa, 2005; Kotter and Ciriacy, 1992; Kuyper et al., 2005; Sonderegger et al., 2004; Wisselink et al., 2007b). Other species in the genus Saccharomyces have similar substrate ranges and penchants for ethanol production (D’Amore et al., 1989). One species, S. diastaticus,6 is of particular interest. While certain strains of S. cerevisiae have been engineered to produce some of the enzymes necessary to digest starchy substrates (S. cerevisiae was first engineered to express, glycosylate, and secrete the starch-degrading enzyme glucoamlyase (Innis et al., 1985) over two decades ago) and more recently to produce ethanol from complex substrates like starch (Kondo et al., 2002), it is unable to match S. diastaticus in its ability for ethanol production from starch or dextrin. Starch, like lignocellulose, is actively used for bioethanol production.7 S. diastaticus yields approximately 1.50% (v/v) ethanol when grown on glucose after 12 hours, a nearly identical amount of ethanol when grown on dextrin for a similar period, and about 1.2% (v/v) ethanol after 12 hours of growth on starch (Laluce and Mattoon, 1984). Nevertheless, S. cerevisiae remains the primary focus of biofuels research—and with good reason. S. cerevisiae will tolerate up to 20% ethanol (v/v) in liquid culture (Kodama, 1993; Morais et al., 1996), and studies in optimizing the fermentative process for ethanol production have achieved (in a fed-batch culture) a maximal ethanol titer of 147 g/L after 45 hours of fermentation (Alfenore et al., 2002), tolerating 19% (v/v) ethanol, with an impressive specific production8 rate of 9.5 g/L/h. An even higher tolerance was achieved with Saccharomyces sake, which produced and tolerated 162.2 g/L (Hayashida and Ohta, 1981), albeit after 480 hours. Lignocellulosic utilization remains a focal point of yeast-centered biofuels research (Nevoigt, 2008) and includes genetic engineering that extends beyond simple sugar utilization.

P1: SFK/UKS P2: SFK c08 BLBS075-Hood December 15, 2010

21:54

Trim: 244mm×172mm

8 Fermentation Organisms for 5- and 6-Carbon Sugars

177

These include the insertion of enzymes from other organisms (like Trichoderma reesei) to permit the yeast to degrade lignocellulosic polymers, saving the costly enzymatic hydrolysis pretreatment that must be performed on the substrate before fermentation can take place (Lynd et al., 2005).

Bacteria Bacillus Bacillus is an enormous and remarkably diverse group of bacteria. They are generally benign, being largely saprophytic, and are of great industrial importance. Among the industrial uses for the organisms in the genus are the production of enzymes, insecticidal compounds, and antibiotics. Species in this genus are characterized as anaerobic (both facultative and obligate species are represented), gram-positive, and have rod-shaped cells (Harwood, 1989). Among the more well-known species are Bacillus anthracis, the causative agent of Anthrax, and B. subtilis, a highly studied model organism. The genus is also noted for its diverse fermentation products and utilizable substrates. Apart from the common hexose substrates, certain Bacilli can grow on arabinose, xylose, starch, raffinose, and trehalose (Cruz Ramos et al., 2000; Slapack et al., 1987). They have been shown to produce ethanol, acetic acid, acetate, and 2,3-butanediol at significant levels. B. subtilis has been shown to produce 23.3 mM lactate, 16.4 mM acetate, and 16.7 mM 2,3-butanediol after 10 hours growth at 37◦ C in a culture containing 50 mM glucose and 50 mM pyruvate (Nakano et al., 1997). A number of thermophilic species belong in Bacillus, which grow at temperatures far in excess of normal organisms. Thermophiles are of particular interest to industry because of their ability to remain viable at high temperatures, given that the speed with which chemical reactions occur varies directly with temperature. Additionally, thermophiles tend to be more resistant to other stresses in addition to heat (Ljungdahl et al., 1981; Wiegel and Ljungdahl, 1981; Wiegel et al., 1979), including substances inhibitory to growth and heavy metal ions. One particularly well-studied species, B. stearothermophilus,9 utilizes a wide variety of sugars to produce ethanol (Sharp et al., 1980), including starch, arabinose, xylose, and sorbitol. Unfortunately, these ethanol yields are comparatively low, far from the stoichiometric limit of 2 moles of ethanol per mole glucose in the wild type (Atkinson et al., 1975) although further work indicates that the lackluster fermentative capacity of B. stearothermophilus could be increased greatly (Payton, 1984; Pennock and Tempest, 1988). B. stearothermophilus also produces a host of lignocellulose-degrading enzymes (Hartley and Payton, 1983; Nanmori et al., 1990), which combined with the bacterium’s high growth temperature and rapid growth (having a doubling time of 30 minutes at 70◦ C) make it of considerable interest to biofuels production. Another species, B. caldolyticus, discovered in hot springs in Yellowstone National Park (Heinen and Heinen, 1972), is even more extreme, growing at temperatures high enough to be considered a hyperthermophile (Slapack et al., 1987; Wiegel et al., 1985). B. caldolyticus grows at up to 105◦ C, and at a temperature of 75◦ C has the astonishing doubling time of only 15 minutes. The bacterium produces acetic acid from substrates like glucose, starch, and glycerol (Heinen and Heinen, 1972). Like B. stearothermophilus, B. caldolyticus produces extracellular enzymes (notably an amylase) that are highly thermostable and may find uses preparing substrates for consumption in bioethanol production.

P1: SFK/UKS P2: SFK c08 BLBS075-Hood December 15, 2010

178

21:54

Trim: 244mm×172mm

Plant Biomass Conversion

Clostridium First described in 1880 (Cato et al., 1986), the genus Clostridium consists of gram-positive rod-shaped obligate anaerobes (Minton and Clarke, 1989). Like Bacillus, the genus contains some pathogenic species, including the causative agent of botulism and tetanus. Species in the genus are capable of utilizing a wide range of substrates, ranging from hexoses and pentoses, including lignocellulosic sugars arabinose, xylose, and cellobiose, to complex carbohydrates like xylan and starch (Hyun and Zeikus, 1985a, 1985b). Further, species in Clostridium are noted for possessing a number of extracellular enzymes that degrade hemicellulose and cellulose with great efficacy (Mitchell, 1998). Species of interest in the genus Clostridium10 are broadly grouped into those that participate in acetone–butanol–ethanol (ABE) fermentation (Maddox, 1989), although many other products are evolved during fermentation across the genus.11 ABE fermentation refers unsurprisingly to the varying proportions of each solvent obtained through carbohydrate fermentation, chiefly by species in the class Clostridia, which includes the Clostridium genus (Mousdale, 2008). From the biofuels perspective, the production of butanol from ABE fermentation is the most valuable. However, the ratio of products during ABE fermentation is a multivariate parameter, depending on the species, substrate, and culture conditions. Even different strains of the same species produce wildly divergent amounts of butanol as a function of total solvents produced depending on the substrate being used12 (Mishra and Singh, 1993). Nevertheless, great promise continues to emerge from research on ABE fermentations with Clostridia. A 1999 study describes a solvent “hyper-producing” strain of C. beijerinckii known as BA101 (Qureshi and Blaschek, 1999). The strain produced 24.2 g/L solvent (of which 19.6 g/L was butanol) from 57.3 g/L glucose in solution using a batch reactor. Solvents were produced at a remarkable 0.34 g/L/h. Using the fermentation product recovery integrative process developed by investigators, the solvent production was nearly doubled to 51.5 g/L. Because of the interest in lignocellulosic biomass as a substrate for biofuels production, bacteria capable of using cellulose to produce ethanol are under considerable research. The bacterium C. cellulolyticum is one such organism, capable of degrading and utilizing insoluble cellulose directly (Desvaux, 2006). The bacterium produces three chief fermentation products: acetate, ethanol, and lactate. One study, examining fermentation by C. cellulolyticum in continuous cultures found that as the dilution rate increases, the amount of consumed cellulose decreases, but the production of pyruvate (in mmols per gram of cells per hour) increased, and further that the percent of pyruvate that is converted to ethanol increases (Desvaux et al., 2006). Unfortunately, even at the maximal dilution rate, the ratio of acetate to ethanol produced was still nearly 2:1. Nevertheless, the relative ease with which Clostridium is genetically manipulated could permit ABE fermentations and single-step cellulose-to-ethanol conversion a viable possibility.

Escherichia Escherichia contains the model organism E. coli, a gut-dwelling rod-shaped bacterium whose use as a model organism is responsible for great advances in molecular biology and microbiology (Dworkin, 2006; Kellogg and Shaffer, 1993). Importantly, E. coli has a large range of usable substrates and can metabolize all major sugars present in plant biomass (Alterthum and Ingram, 1989). In an anaerobic environment, E. coli (like all members of the family Enterobacteriaceae) engages in mixed-acid fermentation, producing lactic, acetic, formic, and

P1: SFK/UKS P2: SFK c08 BLBS075-Hood December 15, 2010

21:54

Trim: 244mm×172mm

8 Fermentation Organisms for 5- and 6-Carbon Sugars

179

succinic acid along with ethanol (Moat et al., 2002). However, when grown on glucose, E. coli produces more than double the amount of lactic acid than it does on ethanol, which accounts for less than 20% of the carbon recovery. The desire to increase the amount of ethanol produced during mixed-acid fermentation along with the ease with which E. coli is genetically manipulated has led to numerous genetically modified strains with varying ethanol producing capacities as well as usable substrates (Mousdale, 2008). Using genes from Zymomonas mobilis (see below), impressive results have been obtained from transgenic E. coli. One study achieved 54.4 g/L ethanol from 10% glucose and 41.6 g/L ethanol from 8% xylose (Ohta et al., 1991). These ethanol yields are in fact beyond the theoretical limit for ethanol production. This is accounted for by considering the use of more complex nutrients in the culture as additional substrates for ethanol production. The strain, KO11, has also been subjected to continuously increasing ethanol concentrations to force ethanol tolerance adaptation (Yomano et al., 1998). Improved ethanol tolerance was taken further by using cells immobilized on glass spheres, with one study finding the method increased ethanol tolerance up to 15-fold (Zhou et al., 2008). The immobilization had the further benefit of increasing phenotypic stability, important given that KO11 has been observed to be phenotypically unstable (in terms of ethanol productivity), particularly on nonglucose substrates. Recently, a strain of KO11, carrying the casAB operon from Klebsiella oxytoca, was found to have gained the ability to efficiently ferment cellobiose spontaneously (Moniruzzaman et al., 1997), due to mutations in the casAB operon that inactivated the operator region.

Klebsiella The genus Klebsiella consists of rod-shaped gram-negative cells that lack motility. They are noted for their polysaccharide capsule, a protective barrier that exists outside the cell wall (Ryan and Ray, 2004). Species in Klebsiella can ferment a large number of both pentoses and hexoses (including arabinose, galactose, glucose, mannose, and xylose) as well as glycerol (Biebl et al., 1998), cellobiose, and cellotriose (Dien et al., 2003). During fermentation, Klebsiella produces a range of products including organic acids, hence much research has focused on shifting the products of fermentation towards ethanol. This work has met with some success. Investigators found one construct, named P2 derived from K. oxytoca strain M5A1, that upon being transformed with ethanol-producing transgenes from Zymomonas mobilis (see below) produced 46.4 g/L ethanol when grown on 10% glucose and 45.2 g/L ethanol when grown on 1% cellobiose (Wood and Ingram, 1992). Importantly, the transgenic P2 has the ability to transport cellobiose and cellotriose, meaning that (potentially costly) enzymatic treatment to further depolymerize the substrate is unnecessary.

Zymomonas Of the genus Zymomonas, the species Zymomonas mobilis is by far the most famous. In the 1950s, it was discovered that Z. mobilis employed the Entner-Duodoroff mechanism for catabolism of glucose, making it the first anaerobe discovered to do so (Swings and De Ley, 1977). However, the enduring research interest on Zymomonas is due to the impressive efficiency with which it converts carbohydrates to ethanol, which can reach 98% (Skotnicki et al., 1981), with all strains of Z. mobilis displaying extreme efficiency with ethanol production (Rogers et al., 1982), rivaling and even exceeding that of S. cerevisiae (Karsch et al., 1983).

P1: SFK/UKS P2: SFK c08 BLBS075-Hood December 15, 2010

180

21:54

Trim: 244mm×172mm

Plant Biomass Conversion

While biofuels research on Clostridium, Escherichia, and Klebsiella have all met with difficulty due to the presence of multiple fermentation products, the primary focus of genetic improvement for Zymomonas has been to improve its range of fermentable substrates: only glucose and fructose are fermented universally, with some strains capable of using sucrose and sorbitol, and rarely raffinose (Swings and De Ley, 1977). In one 1995 study, investigators transformed the strain CP4 with two operons encoding a total of four genes for xylose assimilation and the pentose phosphate pathway. The recombinant CP4 produced ethanol from xylose as the sole carbon source with a yield of 0.44 g/g, or 1.43 mol/mol (compared with the theoretical efficiencies of 0.51 g/g and 1.67 mol/mol) (Zhang et al., 1995). The genetic changes to CP4 did not appear to form a significant metabolic drag either; the recombinant strain produced ethanol at 94% efficiency from glucose in 16 hours compared with the control strain’s 97% efficiency after an equal period of time. In mixed cultures, containing both xylose and glucose, the recombinant CP4 strain was found to metabolize both sugars completely within 30 hours, with preferential glucose utilization. In 1996, shortly after the development of the xylose-utilizing recombinant CP4, another engineered strain was reported with the capacity to utilize arabinose (Deanda et al., 1996). In 1997, the National Renewable Energy Laboratory (NREL) conducted a survey and determined that Z. mobilis was the most promising organism for genetic manipulation with the aim of bioethanol production, specifically from lignocellulosic material and agricultural wastes (Lawford and Rousseau, 1997). NREL’s first generation Zymomonas strains achieved fermentation of a variety of substrates, but suffered from genetic instability in cultures that lacked pentoses as the primary carbon source (Lawford and Rousseau, 2002). Because of restrictions on massive-scale use, antibiotics could not be used to enforce genetic stability. The second-generation strains, however, achieved much greater genetic stability and substrate utilization capacity. One strain, AX101, was capable of simultaneous co-fermentation of glucose, xylose, and arabinose with high efficiency, having a yield of about 84% (Mohagheghi et al., 2002). Importantly, the genes yielding this capacity to AX101 were integrated into the genome, as opposed to existing on a plasmid, giving the strain far greater genetic stability, while possessing the same ethanol yield as the plasmid-bearing strain.

8.5 Other Relevant Products The previous portions of this chapter have provided a comprehensive description of many species that ferment 5- and 6-carbon sugars and the mix of fermentation products produced. However, it is also important to consider industrially relevant products of microbial fermentations from 5- and 6-carbon sugars. Table 8.3 includes 12 products, which are considered industrially relevant based on their gross tonnage output or overall importance in regard to the chemical or fuel industries (Corma et al., 2007; Gavrilescu and Chisti, 2005; Ji et al., 2009; Suriyamongkol et al., 2007). Since it is neither possible nor necessary to compile a list including all current or potential products made from fermentation of five- and six-carbon sugars, we have made some omissions. We provide a list of compounds that are currently produced at a high volume or have an impact on certain major markets (Gavrilescu and Chisti, 2005; Werpy et al., 2004). As can be seen in Table 8.3, glaring omissions of major product groups produced via microbial fermentation have been made. These omissions include things like antibiotics, enzymes, sweeteners, etc. These product groups were omitted due to the large range of specific products, as well as the specific purpose each individual product has. Thus, our product list

Uses Fuel source, solvent

Flavoring agent, dietary supplements, pharmaceuticals, cosmetics, fertilizers Thickener, cryoprotectant, humectant, drug carrier, adhesive, heavy metal absorbent Food additive, pharmaceuticals, detergent additive, cosmetics Food additive, preservative, petrochemical intermediate substitute Plastics(NatureworksLLC) Antioxidant, food additive, cosmetics, pharmaceuticals Construction industries, cleaning agent, pharmaceuticals, food additive

Product

Ethanolii

L-glutamic acid Poly(glutamic acid)

Citric acid

Lactic acid Poly(lactic acid)

Vitamin Ciii

Gluconic acid

Aspergillus niger

50,000

80,000

(Continued)

Gavrilescu and Chisti (2005) and Singh and Kumar (2007)

Chotani et al. (2000), Bremus et al. (2006), and Gavrilescu and Chisti (2005)

Corma et al. (2007), Gavrilescu and Chisti (2005), John et al. (2007), and Leuchtenberger et al. (2005)

250,000

Lactococcus lactis, Lactobacillus delbrueckii, Lactobacillus helveticus, Lactobacillus casei Gluconobacter spp., Erwinia herbicolaiv

Gavrilescu and Chisti (2005) and Soccol et al. (2006)

1,000,000

Corma et al. (2007), Gavrilescu and Chisti (2005), Hermann (2003), Leuchtenberger et al. (2005), and Shih and Van (2001)

1,000,000

Corynebacterium glutamicum Bacillus subtilis Bacillus Licheniformis

21:54

Aspergillus niger

Demain (2009), Gavrilescu and Chisti (2005), and Zaldivar et al. (2001)

26,000,000

References

Saccharomyces cerevisiae, Zymomonas mobilis, Escherichia coli

Organism

Annual Production (Gavrilescu and Chisti 2005) (Tons)i

Table 8.3. List of products produced by fermentation of 5 and 6 carbon sugars along with their current uses as well as the primary microorganisms used for production.

P1: SFK/UKS P2: SFK c08 BLBS075-Hood December 15, 2010 Trim: 244mm×172mm

181

182 Food additive, thickening agent, lubricant, pharmaceuticals Food additive, pharmaceuticals, surfactants, detergents, solvents, biodegradable plastics, petrochemical intermediate alternative Petrochemical intermediate substitute, plastics, adhesives, elastomers, detergents, pharmaceuticals, agriculture Plastics, elastomers, rubbers, drug delivery systems Petrochemical feedstock substitute, fuel source Fuel source, solvent

Xanthan

Succinic acid

Itaconic acid

Poly-hydroxyalkanoates (PHAs)

2,3-Butanediol

Butanol

Demain (2009)

Ji et al. (2009)

Patnaik (2005) and Suriyamongkol et al. (2007)

ii The

annual production is as of 2006 and is not displayed if the value is not known. majority of species listed in Table 8.2 have the ability to ferment ethanol, and these species are the leading candidates for industrial bioethanol production. iii Currently, the leading technology for production of vitamin C, or L-ascorbic acid (L-AA), is the so-called Reichstein process, which includes only one biological conversion step using Gluconobacter oxydans, and the remaining six steps are a mix of chemical and enzymatic conversions. iv This species has been genetically altered and does not directly convert glucose to vitamin C, rather 2-keto-L-gulonic acid (2KLGA) is the fermentation product. Subsequent enzymatic conversion is needed to make vitamin C.

Clostridium beijerinkii; Clostridium acetobutylicum

Klebsiella oxytoca

Ralstonia eutropha

Corma et al. (2007)

Aspergillus terrus, Aspergillus itaconicus

Garcia-Ochoa et al. (2000) and Gavrilescu and Chisti (2005) Corma et al. (2007) and Zeikus et al. (1999)

30,000

References

Actinobacillus succinogenes

Xanthomonas campestris

Organism

Annual Production (Gavrilescu and Chisti 2005) (Tons)i

21:54

i The

Uses

(Continued)

Product

Table 8.3.

P1: SFK/UKS P2: SFK c08 BLBS075-Hood December 15, 2010 Trim: 244mm×172mm

P1: SFK/UKS P2: SFK c08 BLBS075-Hood December 15, 2010

21:54

Trim: 244mm×172mm

8 Fermentation Organisms for 5- and 6-Carbon Sugars

183

includes major specific products produced by industrial biological fermentations of five- and six-carbon sugars and their use.

8.6 Summary This chapter provides a comprehensive survey of microbes that can utilize 5- and 6-carbon sugars to produce compounds of commercial interest. This work provides an extensive list of references to further help the reader.

Endnotes 1. It was found, however, that the strain, when made respiration-deficient via the petite mutation (ρ0), produced more ethanol and accumulated less xylitol, but were unable to form colonies. 2. The incredible diversity of bacteria capable of fermentation means that this list is necessarily limited; we have chosen to focus on species that are of interest to industry, particularly biofuels. Even the subset of bacteria under active research for industrial applications is too large to be encompassed here. Other genera not included in this list but nevertheless of great interest to industry include Acetivibrio, Acetobacter, Butyribacterium, Erwinia, Lactobacillus, Lactococcus, Leuconostoc, Ruminococcus, Sarcina, Thermoanaerobacter, and Zymobacter. 3. The ability to use the pentose sugar xylose is of special interest to the biofuels industry, as it is a major component of the plant biopolymer lignocellulose (see footnote 5). 4. The precise definition of brewer’s yeast is, in fact, substantially more complex than Saccharomyces cerevisiae alone. Lager strains were originally derived from Saccharomyces pastorianus (formerly Saccharomyces carlsbergensis). Other studies have found lineages in industrial strains pointing to still more species, such as Saccharomyces bayanus and Saccharomyces uvarum. 5. “Lignocellulosic” refers to sugars obtained by hydrolysis of the extraordinarily abundant natural polymer known as lignocellulose, derived from plants and currently under active investigation for possible biofuels utilization. 6. S. diastaticus is a close relative of S. cerevisiae, to the point where their haploid cells are capable of mating. 7. The National Renewable Energy Laboratory estimates that starch feedstock required per gallon ethanol to be approximately $0.70, though higher than the estimated feedstock cost from lignocellulose ($0.50) is more than sufficient to generate considerable interest. 8. Note that this refers to instantaneous specific production. 9. Also known as Geobacillus stearothermophilus. 10. A number of species in Clostridium are thermophilic, but are omitted from the body text because their fermentative capacities are represented by other mesophilic and more well-studied species. These include C. thermoaceticum, C. thermocellum, C. thermohydrosulfuricum, and C. thermosaccharolyticum. 11. C. butlicum yields acetone, acetoin, acetate, and butyrate, for instance. C. kluyveri produces a number of exotic products, including caproic acid and hexanol, while C. propionicum produces propionic acid and acrylic acid. Still others produce high amounts of ethanol (C. papyrosolvens) or lactate (C. pasteurianum). Indeed, most species exhibit some change in fermentation products as the culture conditions change.

P1: SFK/UKS P2: SFK c08 BLBS075-Hood December 15, 2010

184

21:54

Trim: 244mm×172mm

Plant Biomass Conversion

12. Studies indicate that alcohol production, both by the Clostridia and Bacillus bacteria are dependent on the initial production of acids, which inhibit growth, leading some to conclude that radical changes in fermentation product ratios will be difficult to produce.

References Aaron, E. W., Hopkins, W. J., & Porges, N. 1958. Whey Utilization: Growth Conditions for Saccharomyces Fragilis. Sewage and Industrial Wastes, 30(7), 913–920. Akedo, M., Cooney, C., & Sinskey, A. 1983. Direct Demonstration of Lactate–Acrylate Interconversion in Clostridium Propionicum. Nature Biotechnology, 1(9), 791–794. Alexander, N. & USDA, A. 1985. Temperature sensitivity of the induction of xylose reductase in Pachysolen tannophilus. Alfenore, S., et al. 2002. Improving ethanol production and viability of Saccharomyces cerevisiae by a vitamin feeding strategy during fed-batch process. Applied Microbiology and Biotechnology, 60(1), 67–72. Alterthum, F. & Ingram, L. 1989. Efficient Ethanol Production from Glucose, Lactose, and Xylose by Recombinant Escherchia coli. Applied and Environmental Microbiology, 55(8), 1943–1948. Andreesen, J., et al. 1973. Fermentation of glucose, fructose, and xylose by Clostridium thermoaceticum: effect of metals on growth yield, enzymes, and the synthesis of acetate from CO2. Journal of bacteriology, 114(2), 743–751. Andreesen, J., Gottschalk, G., & Schlegel, H. 1970. Clostridium formicoaceticum nov. spec. isolation, description and distinction from C. aceticum and C. thermoaceticum. Archives of Microbiology, 72(2), 154–174. Arskold, E., et al. 2008. Phosphoketolase pathway dominates in Lactobacillus reuteri ATCC 55730 containing dual pathways for glycolysis. J Bacteriol, 190(1), 206–12. Atkinson, A., et al. 1975. Production of alcohol by Bacillus stearothermophilus. Biotechnology and Bioengineering, 17(9). Atlas, R. 2004. Handbook of microbiological media. CRC press. Avgerinos, G., et al. 1981. A novel, single-step microbial conversion of cellulosic biomass to ethanol. Advances in Biotechnology, 2, 119–124. Bahl, H., et al. 1982. Effect of pH and butyrate concentration on the production of acetone and butanol by Clostridium acetobutylicum grown in continuous culture. Applied Microbiology and Biotechnology, 14(1), 17–20. Balch, W. E., et al., 1977. Acetobacterium, a New Genus of Hydrogen-Oxidizing, Carbon Dioxide-Reducing, Anaerobic Bacteria. Int J Syst Bacteriol, 27(4), 355–361. Barker, H. & Taha, S. 1942. Clostridium kluyverii, an organism concerned in the formation of caproic acid from ethyl alcohol. Journal of bacteriology, 43(3), 347–363. Barnett, J., Payne, R., & Yarrow, D. Yeasts: characteristics and identification. United States. 3, 658,647. Bastianoni, S. & Marchettini, N. 1996. Ethanol production from biomass: Analysis of process efficiency and sustainability. Biomass and Bioenergy, 11, 411–418. Bauchop, T. & Dawes, E., 1968. The metabolism of formic acid by Sarcina ventriculi. Microbiology, 52(2), 195. Becker, J. & Boles, E. 2003. A modified Saccharomyces cerevisiae strain that consumes L-Arabinose and produces ethanol. Applied and Environmental Microbiology, 69(7), 4144–4150.

P1: SFK/UKS P2: SFK c08 BLBS075-Hood December 15, 2010

21:54

Trim: 244mm×172mm

8 Fermentation Organisms for 5- and 6-Carbon Sugars

185

Bedford, C. 1942. A taxonomic study of the genus Hansenula. Mycologia, 628–649. Belaich, A. & Belaich, J. 1976. Microcalorimetric study of the anaerobic growth of Escherichia coli: measurements of the affinity of whole cells for various energy substrates. Journal of bacteriology, 125(1), 19–24. Belaich, A. & Belaich, J. 1976. Microcalorimetric study of the anaerobic growth of Escherichia coli: growth thermograms in a synthetic medium. Journal of bacteriology, 125(1), 14–18. Ben-Bassat, A. & Zeikus, J. 1981. Thermobacteroides acetoethylicus gen. nov. and spec. nov., a new chemoorganotrophic, anaerobic, thermophilic bacterium. Archives of Microbiology, 128(4), 365–370. Bentley, R., & Bennett, J. W. 2008. A Ferment of Fermentations: Reflections on the Production of Commodity Chemicals Using Microorganisms, in Advances in Applied Microbology. Bernstein, S., Tzeng, C., & Sisson, D. 1977. The commercial fermentation of cheese whey for the production of protein and/or alcohol. Biebl, H., et al. 1998. Fermentation of glycerol to 1, 3-propanediol and 2, 3-butanediol by Klebsiella pneumoniae. Applied Microbiology and Biotechnology, 50(1), 24–29. Biebl, H. et al. 1999. Microbial production of 1,3-propanediol. Applied Microbiology and Biotechnology, 52(3), 289–297. Bornstein, B. & Barker, H. 1948. The energy metabolism of Clostridium kluyveri and the synthesis of fatty acids. Journal of Biological Chemistry, 172(2), 659–669. Bothast, R., et al. 1994. Fermentation of L-arabinose, D-xylose and D-glucose by ethanologenic recombinant Klebsiella oxytoca strain P2. Biotechnology Letters, 16(4), 401–406. Bott, M., Meyer, M., & Dimroth, P. 1995. Regulation of anaerobic citrate metabolism in Klebsiella pneumoniae. Molecular Microbiology, 18(3), 533–546. Br¨au, B. & Sahm, H. 1986. Cloning and expression of the structural gene for pyruvate decarboxylase of Zymomonas mobilis in Escherichia coli. Archives of Microbiology, 144(3), 296–301. Bremus, C., et al. 2006. The use of microorganisms in L-ascorbic acid production. Journal of Biotechnology, 124(1), 196–205. Breznak, J. & Canale-Parola, E. 1969. Spirochaeta aurantia, a pigmented, facultatively anaerobic spirochete. Journal of Bacteriology, 97(1), 386–395. Buschhorn, H., Durre, P., & Gottschalk, G. 1989. Production and Utilization of Ethanol by the Homoacetogen Acetobacterium woodii. Appl Environ Microbiol, 55(7), 1835–1840. Canale-Parola, E. 1970. Biology of the sugar-fermenting Sarcinae. Microbiology and Molecular Biology Reviews, 34(1), 82–97. Canale-Parola, E. 1977. Physiology and evolution of spirochetes. Microbiology and Molecular Biology Reviews, 41(1), 181–204. Cardon, B. & Barker, H. 1946. Two new amino-acid-fermenting bacteria, Clostridium propionicum and Diplococcus glycinophilus. Journal of bacteriology, 52(6), 629–634. Carreira, L., Wiegel, J., & Ljungdahl, L. 1984. Production of ethanol from biopolymers by anaerobic, thermophilic, and extreme thermophilic bacteria: I. Regulation of carbohydrate utilization in mutants of Thermoanaerobacter ethanolicus. Cato, E., George, W., & Finegold, S. 1986. Genus Clostridium Prazmowski 1880. BERGEY’S manual of systematic Bacteriology, 2, 1141–1200. C¸aylak, B. & Vardar Sukan, F. 1998. Comparison of different production processes for bioethanol. Turkish Journal of Chemistry, 22, 351–360. Chiang, L., et al. 1981. Enzymatic and microbial preparation of D-xylulose from D-xylose. Applied and Environmental Microbiology, 42(1), 66–69.

P1: SFK/UKS P2: SFK c08 BLBS075-Hood December 15, 2010

186

21:54

Trim: 244mm×172mm

Plant Biomass Conversion

Chotani, G., et al. 2000. The commercial production of chemicals using pathway engineering. Biochimica Et Biophysica Acta-Protein Structure and Molecular Enzymology, 1543(2), 434–455. Citti, J., Sandine, W., & Elliker, P. 1965. Comparison of slow and fast acid-producing Streptococcus lactis. Journal of dairy science, 48(1), 14. Clark, D. 1989. The fermentation pathways of Escherichia coli. FEMS Microbiology Letters, 63(3), 223–234. Collection, A. T. C., 2009. American Type Culture Collection Catalogue. American Type Culture Collection. Corma, A., Iborra, S., & Velty, A. 2007. Chemical routes for the transformation of biomass into chemicals. Chem Rev, 107(6), 2411–502. Corry, J. 1978. Possible sources of ethanol ante-and post-mortem: its relationship to the biochemistry and microbiology of decomposition. Journal of Applied Microbiology, 44(1), 1–56. Coulson, J. & Richardson, J., 1954. Chemical engineering. Pergamon Press. Cruz Ramos, H., et al. 2000. Fermentative metabolism of Bacillus subtilis: physiology and regulation of gene expression. Journal of bacteriology, 182(11), 3072–3080. Cysewski, G. & Wilke, C. 1978. Process design and economic studies of alternative fermentation methods for the production of ethanol. Biotechnology and Bioengineering, 20(9). da Cruz, S., Batistote, M., & Ernandes, J. 2003. Effect of sugar catabolite repression in correlation with the structural complexity of the nitrogen source on yeast growth and fermentation. Journal of the Institute of Brewing, 109(4), 349–355. Dabrock, B., Bahl, H., & Gottschalk, G. 1992. Parameters affecting solvent production by Clostridium pasteurianum. Applied and Environmental Microbiology, 58(4), 1233–1239. D’Amore, T., Russell, I., & Stewart, G. 1989. Sugar utilization by yeast during fermentation. Journal of Industrial Microbiology and Biotechnology, 4(4), 315–323. De Vries, W., et al. 1970. Molar growth yields and fermentation balances of Lactobacillus casei L3 in batch cultures and in continuous cultures. Microbiology, 63(3), 333. Deanda, K., et al. 1996. Development of an arabinose-fermenting Zymomonas mobilis strain by metabolic pathway engineering. Applied and Environmental Microbiology, 62(12), 4465. Dekker, R. 1982. Ethanol production from D-xylose and other sugars by the yeastPachysolen tannophilus. Biotechnology Letters, 4(7), 411–416. Demain, A. L. & Solomon, N. A., 1985, Biology of industrial microorganisms. Biotechnology series. Menlo Park, Calif.: Benjamin/Cummings Pub. Co., Advanced Book Program. xviii, 573 p. Demain, A. L. 2009. Biosolutions to the energy problem. J Ind Microbiol Biotechnol, 36(3), 319–32. Desvaux, M., Guedon, E., & Petitdemange, H. 2001. Carbon flux distribution and kinetics of cellulose fermentation in steady-state continuous cultures of Clostridium cellulolyticum on a chemically defined medium. Journal of bacteriology, 183(1), 119–130. Desvaux, M. 2006. Unravelling carbon metabolism in anaerobic cellulolytic bacteria. Biotechnology Progress, 22(5). Dickey, R. 1979. Erwinia chrysanthemi: a comparative study of phenotypic properties of strains from several hosts and other Erwinia species. Phytopathology, 69(4), 324–329. Dien, B., Cotta, M., & Jeffries, T. 2003. Bacteria engineered for fuel ethanol production: current status. Applied Microbiology and Biotechnology, 63(3), 258–266. Dien, B. S., et al. 1996. Screening for L-arabinose fermenting yeasts. Appl Biochem Biotechnol, 57–58, 233–42.

P1: SFK/UKS P2: SFK c08 BLBS075-Hood December 15, 2010

21:54

Trim: 244mm×172mm

8 Fermentation Organisms for 5- and 6-Carbon Sugars

187

Dorn, M., Andreesen, J., & Gottschalk, G. 1978. Fermentation of fumarate and L-malate by Clostridium formicoaceticum. Journal of bacteriology, 133(1), 26–32. Doten, R. & Mortlock, R. 1985. Production of D-and L-xylulose by mutants of Klebsiella pneumoniae and Erwinia uredovora. Applied and Environmental Microbiology, 49(1), 158– 162. Du Preez, J., Bosch, M., & Prior, B. 1986. Xylose fermentation by Candida shehatae and Pichia stipitis: effects of pH, temperature and substrate concentration. Enzyme and Microbial Technology, 8(6), 360–364. Du Preez, J. 1994. Process parameters and environmental factors affecting D-xylose fermentation by yeasts. Enzyme and Microbial Technology, 16(11), 944–956. Dworkin, M., 2006. The Prokaryotes: Volume 1: Symbiotic Associations, Biotechnology, Applied Microbiology. Springer Verlag. Eller, C. & Ordal, Z. 1972. Interference-Contrast and Phase-Contrast Microscopy of Sporulation in Clostridium thermosaccharolyticum Grown Under Strict Anaerobiosis. Journal of bacteriology, 111(3), 674–681. Emanuilova, E. & Toda, K. 1984. Amylase production in batch and continuous cultures by Bacillus caldolyticus. Applied Microbiology and Biotechnology, 19(5), 301–305. Endoh, R., Suzuki, M., & Benno, Y. 2008. Ambrosiozyma kamigamensis sp. nov. & A. neoplatypodis sp. nov., two new ascomycetous yeasts from ambrosia beetle galleries. Antonie Van Leeuwenhoek, 94(3), 365–76. Enebo, L. 1951. On three bacteria connected with thermophilic cellulose fermentation. Physiologia Plantarum, 4(4), 652–666. Ergun, M. & Ferda Mutlu, S. 2000. Application of a statistical technique to the production of ethanol from sugar beet molasses by Saccharomyces cerevisiae. Bioresource Technology, 73(3), 251–255. Erickson, L. E., & Fung, D. Y. C., ed. 1988. In: W. C. McGregor (ed), Handbook on Anaerobic Fermentations. Bioprocess Technology, Volume 3. New York, NY: Marcel Dekker Inc.. Fontaine, F., et al. 1942. A new type of glucose fermentation by Clostridium thermoaceticum. Journal of bacteriology, 43(6), 701–715. Fotadar, U., Zaveloff, P., & Terracio, L. 2005. Growth of Escherichia coli at elevated temperatures. Journal of basic microbiology, 45(5). Friend, B. & Shahani, K. 1979. Whey fermentation. New Zealand Journal of Dairy Science and Technology (New Zealand). Gaines, S. & Stahly, G. 1943. The growth requirements of Leuconostoc mesenteroides and preliminary studies on its use as an assay agent for several members of the vitamin B complex. Journal of bacteriology, 46(5), 441–449. Gale, E. & Epps, H. 1942. The effect of the pH of the medium during growth on the enzymic activities of bacteria (Escherichia coli and Micrococcus lysodeikticus) and the biological significance of the changes produced. Biochemical journal, 36(7–9), 600. Garcia-Ochoa, F. 2000. et al., Xanthan gum: production, recovery, and properties. Biotechnology Advances, 18(7), 549–579. Garrett, R. H., & Grisham, C. M. 2005. Biochemistry, 3rd ed., Belmont, CA: Thompson. Gavrilescu, M. & Chisti, Y. 2005. Biotechnology-a sustainable alternative for chemical industry. Biotechnol Adv, 23(7–8), 471–99. George, H. & Chen, J. 1983. Acidic conditions are not obligatory for onset of butanol formation by Clostridium beijerinckii (synonym, C. butylicum). Applied and Environmental Microbiology, 46(2), 321–327.

P1: SFK/UKS P2: SFK c08 BLBS075-Hood December 15, 2010

188

21:54

Trim: 244mm×172mm

Plant Biomass Conversion

George, H., et al. 1983. Acetone, Isopropanol, and Butanol Production by Clostridium beijerinckii (syn. Clostridium butylicum) and Clostridium aurantibutyricum. Am Soc Microbiol., 1160–1163. Giampietro, M., Ulgiati, S., & Pimentel, D. 1997. Feasibility of large-scale biofuel production. BioScience, 587–600. Gibbs, M., et al. 1954. Fermentation of D-xylose-1-C14 by Fusarium lini Bolley. Archives of biochemistry and biophysics, 50(2), 237. Gong, C., et al. 1983. Conversion of pentoses by yeasts. Biotechnology and Bioengineering, 25(1), 85–102. Gong, C., Chen, L., & Tsao, G. 1981. Quantitative production of xylitol from D-xylose by a high-xylitol producing yeast mutant Candida tropicalis HXP2. Biotechnology Letters, 3(3), 125–130. Gray, W. D. 1949. Initial Studies on the Metabolism of Hansenula anomala (Hansen) Sydow. American Journal of Botany, 36(6), 475–480. Gunsalus, I. & Stanier, R. The Bacteria. a Treatise on Structure and Function. 3th. ed. Hahn-Hagerdal, B. 2001. Metabolic Engineering of Saccharomyces cerevisiae for Xylose Utilization. Advancers in Biochemical Engineering/Biotechnology, 71, 53–84. Hartley, B. & Payton, M. 1983. Industrial prospects for thermophiles and thermophilic enzymes. Harwood, C. R. 1989. Bacillus. 1 ed. Biotechnology Handbooks, Vol. 1. Springer, p. 414. Hayashida, S. & Ohta, K. 1981. Formation of high concentrations of alcohol by various yeasts[Japan: including Saccharomyces sake]. Journal of the Institute of Brewing, 87, 42–4. Heinen, U. J. & Heinen, W. 1972. Characteristics and properties of a caldo-active bacterium producing extracellular enzymes and two related strains. Arch Mikrobiol, 82(1), 1–23. Heinen, W. & Lauwers, A. 1981. Growth of bacteria at 100 C and beyond. Archives of Microbiology, 129(2), 127–128. Hermann, T. 2003. Industrial production of amino acids by coryneform bacteria. Journal of Biotechnology, 104(1–3), 155–172. Hernandez, E. & Johnson, M. 1967. Anaerobic growth yields of Aerobacter cloacae and Escherichia coli. Journal of bacteriology, 94(4), 991–995. Hespell, R. & Canale-Parola, E., Spirochaeta litoralis sp. n., a strictly anaerobic marine spirochete. Archives of Microbiology, 1970. 74(1), 1–18. Hespell, R. B. & Canale-Parola, E., Carbohydrate Metabolism in Spirochaeta stenostrepta. J. Bacteriol., 1970. 103(1), 216–226. Hespell, R. B. & Canale-Parola, E., Glucose and Pyruvate Metabolism of Spirochaeta litoralis, an Anaerobic Marine Spirochete. J. Bacteriol., 1973. 116(2), 931–937. Heyndrickx, M., et al. 1991. The Fermentation of Glycerol by Clostridium Butyricum LMG 1212 T 2 AND 1213 T 1 AND C. Pasteurianum LMG 3285. Applied Microbiology and Biotechnology, 34(5), 637–642. Hinman, N. D., et al. 1989. Xylose fermentation-an economic analysis. Applied Biochemistry and Biotechnology, 20, 391–402. Hsu, E. & Ordal, Z. 1970. Comparative metabolism of vegetative and sporulating cultures of Clostridium thermosaccharolyticum. Journal of bacteriology, 102(2), 369–376. Hyun, H. & Zeikus, J. 1985. General biochemical characterization of thermostable pullulanase and glucoamylase from Clostridium thermohydrosulfuricum. Applied and Environmental Microbiology, 49(5), 1168–1173.

P1: SFK/UKS P2: SFK c08 BLBS075-Hood December 15, 2010

21:54

Trim: 244mm×172mm

8 Fermentation Organisms for 5- and 6-Carbon Sugars

189

Hyun, H. & Zeikus, J. 1985. Simultaneous and enhanced production of thermostable amylases and ethanol from starch by cocultures of Clostridium thermosulfurogenes and Clostridium thermohydrosulfuricum. Applied and Environmental Microbiology, 49(5), 1174– 1181. Innis, M., et al. 1985. Expression, glycosylation, and secretion of an Aspergillus glucoamylase by Saccharomyces cerevisiae. Science, 228(4695), 21–26. Janssens, J. H., et al. 1983. Lipid-Enhanced Ethanol Production by Kluyveromyces fragilis. Appl. Environ. Microbiol., 45(2), 598–602. Jeffries, T. & Kurtzman, C., Strain selection, taxonomy, and genetics of xylose-fermenting yeasts. atmosphere. 1, 2. Jeffries, T. 1981. Conversion of xylose to ethanol under aerobic conditions by Candida tropicalis. Biotechnology Letters, 3(5), 213–218. Ji, X. J., et al. 2009. Enhanced 2,3-butanediol production by Klebsiella oxytoca using a twostage agitation speed control strategy. Bioresource Technology, 100(13), 3410–3414. Jin, Y.-S. et. al. 2002. Optimal Growth and Ethanol Production from Xylose by Recombinant Saccharomyces cerevisiae Require Moderate D-Xylulokinase Activity. Applied and Environmental Microbiology, 69, 495–503. Jin, Y.-S. 2004. Saccharomyces cerevisiae Engineered for Xylose Metabolism Exhibits a Respiratory Response. Applied Environmental Microbiology, 70, 6816–6825. John, R. P., Nampoothiri, K. M., & Pandey, A. 2007. Fermentative production of lactic acid from biomass: an overview on process developments and future perspectives. Appl Microbiol Biotechnol, 74(3), 524–34. Johns, A. 1952. The mechanism of propionic acid formation by Clostridium propionicum. Microbiology, 6(1–2), 123. Kadam, K. & Schmidt, S. 1997. Evaluation of Candida acidothermophilum in ethanol production from lignocellulosic biomass. Applied Microbiology and Biotechnology, 48(6), 709–713. Karhumaa, K., Hahn-Hagerdal, B., & Gorwa-Grauslund, M. F. 2005. Investigation of limiting metabolic steps in the utilization of xylose by recombinant Saccharomyces cerevisiae using metabolic engineering. Yeast, 22(5), 359–68. Karhumaa, K. 2005. Investigation of limiting metabolic steps in the utilization of xylose by recombinant Saccharomyces cerevisiae using metabolic engineering. Yeast, 22, 359– 368. Karsch, T., Stahl, U., & Esser, K. 1983. Ethanol production by Zymomonas and Saccharomyces, advantages and disadvantages. Applied Microbiology and Biotechnology, 18(6), 387–391. Kelker, N., Hanson, T., & Anderson, R. 1970. Alternate Pathways of d-Fructose Metabolism in Aerobacter aerogenes A Specific d-Fructokinase and its Preferential Role in the Metabolism of Sucrose. Journal of Biological Chemistry, 245(8), 2060–2065. Kellogg, E. & Shaffer, H. 1993. Model organisms in evolutionary studies. Systematic Biology, 42(4), 409. Kempler, G. & McKay, L. 1980. Improved medium for detection of citrate-fermenting Streptococcus lactis subsp. diacetylactis. Am Soc Microbiol., 926–927. Kenealy, W. & Waselefsky, D. 1985. Studies on the substrate range of Clostridium kluyveri; the use of propanol and succinate. Archives of Microbiology, 141(3), 187–194. Kirsch, D., Kelly, R., & Kurtz, M., The genetics of Candida. 1990: CRC. Klein, A., et al. 1988. Hansenula anomala: a new fungal pathogen. Two case reports and a review of the literature. Archives of Internal Medicine, 148(5), 1210–1213. Kodama, K. 1993. Sake-brewing yeast. The yeasts, 3, 129–168.

P1: SFK/UKS P2: SFK c08 BLBS075-Hood December 15, 2010

190

21:54

Trim: 244mm×172mm

Plant Biomass Conversion

Kondo, A., et al. 2002. High-level ethanol production from starch by a flocculent Saccharomyces cerevisiae strain displaying cell-surface glucoamylase. Applied Microbiology and Biotechnology, 58(3), 291–296. Kotter, P. & Ciriacy, M. 1992. Xylose fermentation by Saccharomyces cerevisiae. Applied Microbiology and Biotechnology, 38(6), 776–783. Kraght, A. & Starr, M. 1952. Fermentation of galacturonic acid and glucose by a strain of Erwinia carotovora. Journal of bacteriology, 64(2), 259–264. Krishnan, M., Ho, N., & Tsao, G. 1999. Fermentation kinetics of ethanol production from glucose and xylose by recombinant Saccharomyces 1400 (pLNH33). Applied biochemistry and biotechnology, 78(1), 373–388. Krouwel, P. & Braber, L. 1979. Ethanol production by yeast at supraoptimal temperatures. Biotechnology Letters, 1(10), 403–408. Kuhad, R., Kumar, M., & Singh, A. 1994. A hypercellulolytic mutant of Fusarium oxysporum. Letters in applied microbiology, 19(5), 397–400. Kupfer, D. & Canale-Parola, E., Fermentation of glucose by Sarcina maxima. Journal of bacteriology, 1968. 95(1), 247–248. Kupfer, D. & Canale-Parola, E., Pyruvate metabolism in Sarcina maxima. Journal of Bacteriology, 1967. 94(4), 984–990. Kuyper, M., et al. 2005. Evolutionary engineering of mixed-sugar utilization by a xylosefermenting Saccharomyces cerevisiae strain. FEMS yeast research, 5(10), 925– 934. Laluce, C. & Mattoon, J. 1984. Development of rapidly fermenting strains of Saccharomyces diastaticus for direct conversion of starch and dextrins to ethanol. Applied and Environmental Microbiology, 48(1), 17–25. Landuyt, S. & Hsu, E., 1985. Solvent fermentation precedes acid fermentation in elongated cells of Clostridium thermnosaccharolyticum. Fundamental and applied aspects of bacterial spores. London: Academic Press, Inc.(London), Ltd.: pp. 477–493. Laskin, A. I., Gadd, G. M., & Sariaslani, S., 2009. Advances in Applied Microbiology. Academic Press. Lawford, H. & Rousseau, J. 1997. Corn steep liquor as a cost-effective nutrition adjunct in highperformance Zymomonas ethanol fermentations. Applied biochemistry and biotechnology, 63(1), 287–304. Lawford, H. & Rousseau, J. 2002. Performance testing of Zymomonas mobilis metabolically engineered for cofermentation of glucose, xylose, and arabinose. Applied biochemistry and biotechnology, 98(1), 429–448. Lawrence, R. & Thomas, T. 1979. The fermentation of milk by lactic acid bacteria. Lee, B. & Blackburn, T. 1975. Cellulase production by a thermophilic Clostridium species. Applied and Environmental Microbiology, 30(3), 346–353. Lee, C. & Ordal, Z. 1967. Regulatory effect of pyruvate on the glucose metabolism of Clostridium thermosaccharolyticum. Journal of bacteriology, 94(3), 530–536. Lee, D. & Collins, E. 1976. Influences of temperature on growth of Streptococcus cremoris and Streptococcus lactis. Journal of dairy science, 59(3), 405. Leuchtenberger, W., Huthmacher, K., & Drauz, K. 2005. Biotechnological production of amino acids and derivatives: current status and prospects. Appl Microbiol Biotechnol, 69(1), 1–8. Ljungdahl, L., Carreira, L., & Wiegel, J. 1981. Production of ethanol from carbohydrates using anaerobic thermophilic bacteria. Ljungdahl, L. 1979. Physiology of thermophilic bacteria. Adv Microb Physiol, 19, 149–243. Lodder, J. 1970. The yeasts; a taxonomic study. 2d rev. and enl. ed. North-Holland Pub. Co.: Amsterdam, xvi, 1385 p.

P1: SFK/UKS P2: SFK c08 BLBS075-Hood December 15, 2010

21:54

Trim: 244mm×172mm

8 Fermentation Organisms for 5- and 6-Carbon Sugars

191

Lovins, A. B., et al. 2005. Winning the Oil Endgame. Rocky Mountain Institute. Lutwick, L., Phaff, H., & Stevens, D. 1980. Kluyveromyces fragilis as an opportunistic fungal pathogen in man. Medical Mycology, 18(1), 69–73. Lynd, L. & Zeikus, J. 1983. Metabolism of H2-CO2, methanol, and glucose by Butyribacterium methylotrophicum. Journal of bacteriology, 153(3), 1415–1423. Lynd, L., et al. 2005. Consolidated bioprocessing of cellulosic biomass: an update. Current opinion in biotechnology, 16(5), 577–583. Lynd, L. R. 1991. Fuel Ethanol from Cellulosic Biomass. Science, 251, 1318–1323. Macfadyen, A. & Blaxall, F. 1896. Thermophilic bacteria. The Journal of Pathology and Bacteriology, 3(1). Maddox, I. 1989. The acetone-butanol-ethanol fermentation: recent progress in technology. Biotechnology & Genetic Engineering Reviews, 7, 189–220. Madigan, M., Martinko, J., & Parker, J. 2000. Brock biology of microorganisms. Madigan, M. T., & Martinko, J. M. 2006. Biology of Microorganisms. 11th ed. Upper Saddle River, NJ: Pearson Maksoud, S. A., M. N. El Hadidi, and Amer, W. M. 1994. Beer from the early dynasties (3500–3400 cal B.C.) of Upper Egypt, detected by archaeochemical methods. Vegetation History and Archaeobotany, 3, 219–224. Maleszka, R., Wang, P., & Schneider, H. 1982. Yeasts that ferment D-cellobiose as well as D-xylose. Biotechnology Letters, 4(2), 133–136. Maloy, S., Cronan, J., & Freifelder, D. 1994. Microbial genetics. Jones & Bartlett Publishers. Man, J. C. D., Rogosa, M., & Sharpe, M. E. 1960. A Medium for the Cultivation OF Lactobacilli. Journal of Applied Microbiology, 23(1), 130–135. Margaritis, A. & Bajpai, P. 1982. Direct fermentation of D-xylose to ethanol by Kluyveromyces marxianus strains. Applied and Environmental Microbiology, 44(5), 1039–1041. Martin, R., et al. 1992. Development of a synthetic medium for continuous anaerobic growth and ethanol production with a lactate dehydrogenase mutant of Bacillus stearothermophilus. Microbiology, 138(5), 987. Martinec, T. & Kocur, M. 1964. A taxonomic study of Erwinia amylovora (Burrill 1882) Winslow et al. 1920. International Journal of Systematic and Evolutionary Microbiology, 14(1), 5. Marwaha, S. & Kennedy, J. 1984. Alcohol production from whey permeate by immobilized and free cells of Kluyveromyces marxianus NCYC 179. Process Biochemistry, 19(2), 79–80. Mayer, K. & Temperli, A. 1963. The metabolism of L-malate and other compounds by Schizosaccharomyces pombe. Archives of Microbiology, 46(3), 321–328. McBee, R. 1948. The Culture and Physiology of a Thermophilic Cellulose-fermenting Bacterium 1. Journal of bacteriology, 56(5), 653–663. McCleskey, C., Faville, L., & Barnett, R. 1947. Characteristics of Leuconostoc mesenteroides from cane juice. Journal of bacteriology, 54(6), 697–708. McGovern, P., et al. 2004. Fermented beverages of pre-and proto-historic China. Proceedings of the National Academy of Sciences, 101(51), 17593–17598. McVeigh, I. & Bracken, E., The nutrition of Schizosaccharomyces pombe. Mycologia, 1955, 13–25. Menzel, K., Zeng, A., & Deckwer, W. 1997. High concentration and productivity of 1, 3propanediol from continuous fermentation of glycerol by Klebsiella pneumoniae. Enzyme and Microbial Technology, 20(2), 82–86. Millichip, R. & Doelle, H. 1989. Large-scale ethanol production from Milo (Sorghum) using Zymomonas mobilis. Process Biochemistry, 24(4), 141–145.

P1: SFK/UKS P2: SFK c08 BLBS075-Hood December 15, 2010

192

21:54

Trim: 244mm×172mm

Plant Biomass Conversion

Mindich, L. 1970. Membrane synthesis in Bacillus subtilis. I. Isolation and properties of strains bearing mutations in glycerol metabolism. Journal of molecular biology, 49(2), 415. Minton, N. & Clarke, D. 1989. Clostridia. Plenum Pub Corp. Mishra, P. & Singh, A. 1993. Microbial pentose utilization. Advances in applied Microbiology, 39, 91–91. Mitchell, W. 1998. Physiology of carbohydrate to solvent conversion by clostridia. Advances in Microbial Physiology, 39, 31. Moat, A., Foster, J., & Spector, M. 2002 Microbial physiology. Wiley-Liss. Mohagheghi, A., et al. 2002. Cofermentation of glucose, xylose, and arabinose by genomic DNA-integrated xylose/arabinose fermenting strain of Zymomonas mobilis AX101. Applied biochemistry and biotechnology, 98(1), 885–898. Moniruzzaman, M., et al. 1997. Isolation and molecular characterization of high-performance cellobiose-fermenting spontaneous mutants of ethanologenic Escherichia coli KO11 containing the Klebsiella oxytoca casAB operon. Applied and Environmental Microbiology, 63(12), 4633. Morais, P., et al. 1996. Production of fuel alcohol by Saccharomyces strains from tropical habitats. Biotechnology Letters, 18(11), 1351–1356. Mousdale, D. 2008. Biofuels: Biotechnology, Chemistry, and Sustainable Development. CRC. Nagodawithana, T. & Steinkraus, K. 1976. Influence of the rate of ethanol production and accumulation on the viability of Saccharomyces cerevisiae in" rapid fermentation". Applied and Environmental Microbiology, 31(2), 158–162. Nakano, M., et al. 1997. Characterization of anaerobic fermentative growth of Bacillus subtilis: identification of fermentation end products and genes required for growth. Journal of bacteriology, 179(21), 6749–6755. Nakas, J., et al. 1983. System development for linked-fermentation production of solvents from algal biomass. Applied and Environmental Microbiology, 46(5), 1017–1023. Nanmori, T., et al. 1990. Purification and properties of thermostable xylanase and betaxylosidase produced by a newly isolated Bacillus stearothermophilus strain. Journal of bacteriology, 172(12), 6669–6672. NatureworksLLC. [Website] [cited 2009 28 Sept.]; Available from: http://www.natureworksllc. com/. Nevoigt, E. 2008. Progress in metabolic engineering of Saccharomyces cerevisiae. Microbiology and Molecular Biology Reviews, 72(3), 379. Ng, T., Ben-Bassat, A., & Zeikus, J. 1981. Ethanol production by thermophilic bacteria: fermentation of cellulosic substrates by cocultures of Clostridium thermocellum and Clostridium thermohydrosulfuricum. Applied and Environmental Microbiology, 41(6), 1337–1343. Ng, T., et al. 1979. Cellulose Fermentation by Thermophilic Anaerobic Bacteria. Nishikawa, N., Sutcliffe, R., & Saddler, J. 1988. The influence of lignin degradation products on xylose fermentation by Klebsiella pneumoniae. Applied Microbiology and Biotechnology, 27(5), 549–552. Nissen, T., et al. 2000. Optimization of ethanol production in Saccharomyces cerevisiae by metabolic engineering of the ammonium assimilation. Metabolic Engineering, 2(1), 69–77. O’Brien, D., Panzer, C., & Eisele, W. 1990. Biological production of acrylic acid from cheese whey by resting cells of Clostridium propionicum. Biotechnology Progress, 6(4). Odds, F. 1979. Candida and candidosis. University Park Press. Ohta, K., et al. 1991. Genetic improvement of Escherichia coli for ethanol production: chromosomal integration of Zymomonas mobilis genes encoding pyruvate decarboxylase and alcohol dehydrogenase II. Applied and Environmental Microbiology, 57(4), 893–900.

P1: SFK/UKS P2: SFK c08 BLBS075-Hood December 15, 2010

21:54

Trim: 244mm×172mm

8 Fermentation Organisms for 5- and 6-Carbon Sugars

193

Palop, M., et al. 1989. Isolation and Characterization of an Anaerobic, Cellulolytic Bacterium, Clostridium celerecrescens sp. nov. International Journal of Systematic and Evolutionary Microbiology, 39(1), 68. Patel, G. & MacKenzie, C. 1982. Metabolism of Acetivibrio cellulolyticus during optimized growth on glucose, cellobiose and cellulose. Applied Microbiology and Biotechnology, 16(4), 212–218. Patnaik, P. R. 2005. Perspectives in the modeling and optimization of PHB production by pure and mixed cultures. Crit Rev Biotechnol, 25(3), 153–71. Payton, M. 1984. Production of ethanol by thermophilic bacteria. Trends in Biotechnology, 2(6), 153–158. Pennock, J. & Tempest, D. 1988. Metabolic and energetic aspects of the growth of Bacillus stearothermophilus in glucose-limited and glucose-sufficient chemostat culture. Archives of Microbiology, 150(5), 452–459. Perlack, R. D., Wright, L. L., Turhollow, A. F., Graham, R. L., Stokes, B. J., Erbach, D. C. 2005. Biomass as Feedstock for a Bioenergy and Bioproducts Industry: The Technical Feasibility of a Billion-Ton Annual Supply, U.S.D.o.A. Department of Energy, Editor. Piˇskur, J. et al. 2006. How did Saccharomyces evolve to become a good brewer? Trends in Genetics, 22(4), 183–186. Plata, C., et al. 2003. Formation of ethyl acetate and isoamyl acetate by various species of wine yeasts. Food microbiology, 20(2), 217–224. Postma, E. & Van den Broek, P. 1990. Continuous-culture study of the regulation of glucose and fructose transport in Kluyveromyces marxianus CBS 6556. Journal of Bacteriology, 172(6), 2871–2876. Preez, J., Bosch, M., & Prior, B. 1986. The fermentation of hexose and pentose sugars by Candida shehatae and Pichia stipitis. Applied Microbiology and Biotechnology, 23(3), 228–233. Pringsheim, H. 1912. Uber den fermentativen Abbau der Cellulose. Zeitschrift fuer Physiologische Chemie, 78, 266–291. Qureshi, N. & Blaschek, H. 1999. Production of acetone butanol ethanol (ABE) by a hyperproducing mutant strain of Clostridium beijerinckii BA101 and recovery by pervaporation. Biotechnology Progress, 15(4). Ragauskas, A. J., et al. 2006. The path forward for biofuels and biomaterials. Science, 311(5760), 484–489. Rahn, O., Hegarty, C., & Deuel, R. 1938. Factors Influencing the Rate of Fermentation of Streptococcus Lactis. Journal of bacteriology, 35(5), 547–558. Reilly, P. & Hollaender, A. 1981. Trends in the Biology of Fermentations for Fuels and Chemicals. New York: Plenum Press. Rogers, P., et al. 1982. Ethanol production by Zymomonas mobilis, In: Microbial Reactions. p. 37–84. Rose, A. H. & Harrison, J. S. 1969. The yeasts. Academic P. v.: London. Ryan, K. & Ray, C. 2004. Sherris medical microbiology: an introduction to infectious diseases. McGraw-Hill Medical. Sachs, R., et al. 1981. Fuel Alcohol From Jerusalem Artichoke. California Agriculture, 35(9), 4–6. Sadava, D., et al. 2006. Life: The science of biology. WH Freeman & Co. Samson, R., et al. 2005. Transfer of Pectobacterium chrysanthemi (Burkholder et al. 1953) Brenner et al. 1973 and Brenneria paradisiaca to the genus Dickeya gen. nov. as Dickeya chrysanthemi comb. nov. and Dickeya paradisiaca comb. nov. and delineation of four novel

P1: SFK/UKS P2: SFK c08 BLBS075-Hood December 15, 2010

194

21:54

Trim: 244mm×172mm

Plant Biomass Conversion

species, Dickeya dadantii sp. nov., Dickeya dianthicola sp. nov., Dickeya dieffenbachiae sp. nov. and Dickeya zeae sp. nov. Soc General Microbiol. p. 1415–1427. Schneider, H., et al. 1981. Conversion of D-xylose into ethanol by the yeast Pachysolen tannophilus. Biotechnology Letters, 3(2), 89–92. Sharp, R., Bown, K., & Atkinson, A. 1980. Phenotypic and genotypic characterization of some thermophilic species of Bacillus. Microbiology, 117(1), 201. Sherman, J. 1937. The streptococci. Microbiology and Molecular Biology Reviews, 1(1), 3–97. Shih, I. L. & Van, Y. T. 2001. The production of poly-(gamma-glutamic acid) from microorganisms and its various applications. Bioresour Technol, 79(3), 207–25. Singh, A. & Kumar, P. 1991. Fusarium oxysporum: status in bioethanol production. Critical Reviews in Biotechnology, 11(2), 129–147. Singh, O. V. & Kumar, R. 2007. Biotechnological production of gluconic acid: future implications. Applied Microbiology and Biotechnology, 75(4), 713–722. Sjolander, N. 1937. Studies on Anaerobic Bacteria XII. The Fermentation Products of Clostridium Thermosaccharolyticum 1. Journal of bacteriology, 34(4), 419–428. Skoog, K. & Hahn-Hagerdal, B. 1990. Effect of Oxygenation on Xylose Fermentation by Pichia stipitis. Appl. Environ. Microbiol., 56(11), 3389–3394. Skotnicki, M., et al. 1981. Comparison of ethanol production by different Zymomonas strains. Applied and Environmental Microbiology, 41(4), 889–893. Slapack, G. E., Russell, I., & Stewart, G. G. 1987. Thermophilic microbes in ethanol production. CRC Press: Boca Raton, Fla. 186 p. Slininger, P. & Usda, A. 1990. Optimum pH and temperature conditions for xylose fermentation by Pichia stipitis. Slininger, P., et al. 1985. Comparative evaluation of ethanol production by xylose-fermenting yeasts presented high xylose concentrations. Biotechnology Letters, 7(6), 431–436. Slininger, P., et al. 2006. Nitrogen source and mineral optimization enhance D-xylose conversion to ethanol by the yeast Pichia stipitis NRRL Y-7124. Applied Microbiology and Biotechnology, 72(6), 1285–1296. Smit, J. 1933. The biology of the fermenting sarcinae. The Journal of Pathology and Bacteriology, 36(3). Soccol, C. R., et al. 2006. New perspectives for citric acid production and application. Food Technology and Biotechnology, 44(2), 141–149. Sonderegger, M., et al. 2004. Fermentation Performance of Engineered and Evolved XyloseFermenting Saccharomyces cerevisiae Strains. Biotechnology and Bioengineering, 87, 90–98. Sonderegger, M., Schumperli, M., & Sauer, U. 2004. Metabolic engineering of a phosphoketolase pathway for pentose catabolism in Saccharomyces cerevisiae. Appl Environ Microbiol, 70(5), 2892–7. Sprenger, G. & Lengeler, J. 1988. Analysis of sucrose catabolism in Klebsiella pneumoniae and in Scr+ derivatives of Escherichia coli K12. Microbiology, 134(6), 1635. Stephenson, M. & Dawes, E. 1971. Pyruvic acid and formic acid metabolism in Sarcina ventriculi and the role of ferredoxin. Microbiology, 69(3), 331. Suihko, M. & Enari, T. 1981. The production of ethanol from D-glucose and D-xylose by different Fusarium strains. Biotechnology Letters, 3(12), 723–728. Sun, Y. & Cheng, J. 2002. Hydrolysis of lignocellulosic material for ethanol production: A review. Bioresearch Technology, 83, 1–11. Sun, Y. & Cheng, J. 2002. Hydrolysis of lignocellulosic materials for ethanol production: a review. Bioresource Technology, 83, 1–11.

P1: SFK/UKS P2: SFK c08 BLBS075-Hood December 15, 2010

21:54

Trim: 244mm×172mm

8 Fermentation Organisms for 5- and 6-Carbon Sugars

195

Suriyamongkol, P., et al. 2007. Biotechnological approaches for the production of polyhydroxyalkanoates in microorganisms and plants - a review. Biotechnol Adv, 25(2), 148–75. Sutton, D. & Starr, M. 1959. Anaerobic Dissimilation of Glucose by Erwinia Amylovora 1. Journal of bacteriology, 78(3), 427–431. Swanson, W. & Clifton, C. 1948. Growth and Assimilation in Cultures of Saccharomyces cerevisiae 1. Journal of bacteriology, 56(1), 115–124. Swings, J. & De, J. Ley, The biology of Zymomonas. Microbiology and Molecular Biology Reviews, 1977. 41(1), 1–46. Thauer, R., et al. 1968. The energy metabolism of Clostridium kluyveri. European Journal of Biochemistry, 4(2), 173–180. Thomas, T., Ellwood, D., & Longyear, V. 1979. Change from homo-to heterolactic fermentation by Streptococcus lactis resulting from glucose limitation in anaerobic chemostat cultures. Journal of bacteriology, 138(1), 109–117. Thomas, T., Turner, K., & Crow, V. 1980. Galactose fermentation by Streptococcus lactis and Streptococcus cremoris: pathways, products, and regulation. Journal of bacteriology, 144(2), 672–682. Toivari, M. H. 2004. Endogenous Xylose Pathway in Saccharomyces cerevisiae. Applied and Environmental Microbiology, 70, 3681–3686. Toivola, A., et al. 1984. Alcoholic fermentation of D-xylose by yeasts. Applied and Environmental Microbiology, 47(6), 1221–1223. Tolan, J. & Finn, R. 1987. Fermentation of D-xylose and L-arabinose to ethanol by Erwinia chrysanthemi. Applied and Environmental Microbiology, 53(9), 2033–2038. Ueng, P., et al. 1981. D-xylulose fermentation in yeasts. Biotechnology Letters, 3(6), 315–320. USDOE, Biomass Feedstock Composition and Propery Database. Vallet, C., et al. 1996. Natural abundance isotopic fractionation in the fermentation reaction: influence of the nature of the yeast. Bioorganic Chemistry, 24(4), 319–330. Van Der Walt, J. 1956. Kluyveromyces—A new yeast genus of theEndomycetales. Antonie Van Leeuwenhoek, 22(1), 265–272. Walker, G. M. 1998. Yeast: Physiology and Biotechnology. Wiley: West Sussex, England. Wang, P., Shopsis, C., & Schneider, H. 1980. Fermentation of a pentose by yeasts. Biochemical and biophysical research communications, 94(1), 248. Wang, Z. X., et al. 2001. Glycerol production by microbial fermentation: a review. Biotechnol Adv, 19(3), 201–23. Wasungu, K. M. & Simard, R. E. 1982. Growth characteristics of bakers’ yeast in ethanol. Biotechnology and Bioengineering, 24(5), 1125–1134. Watanabe, S., et al. 2007. Ethanol production from xylose by recombinant Saccharomyces cerevisiae expressing protein engineered NADP+-dependent xylitol dehydrogenase. Journal of Biotechnology, 130, 316–319. Werpy, T., Peterson, G., Aden, A., Bozell, J., Holladay, J., White, J., Manheim, A. 2004. Top Value Added Chemicals From Biomass, in Volume I: Results of Screening for Potential Candidates from Sugars and Synthesis Gas, T. Werpy, and Peterson, G., Editor. White, M. & Willaman, J. 1928. Biochemistry of plant diseases: Fermentation of pentoses by Fusarium lini. The Biochemical journal, 22(2), 583. Wickerham, L. & Burton, K. 1962. Phylogeny and Biochemistry of the Genus Hansenula 1. Microbiology and Molecular Biology Reviews, 26(4), 382–397. Wickerham, L. 1951. Taxonomy of yeasts. Technical bulletin, 1029(56), 1–5. Wiegel, J. & Puls, J., Production of ethanol from hemicelluloses of hardwoods and annual plants.

P1: SFK/UKS P2: SFK c08 BLBS075-Hood December 15, 2010

196

21:54

Trim: 244mm×172mm

Plant Biomass Conversion

Wiegel, J. & Ljungdahl, L. 1981. Thermoanaerobacter ethanolicus gen. nov., spec. nov., a new, extreme thermophilic, anaerobic bacterium. Archives of Microbiology, 128(4), 343– 348. Wiegel, J. 1980. Formation of ethanol by bacteria. A pledge for the use of extreme thermophilic anaerobic bacteria in industrial ethanol fermentation processes. Cellular and Molecular Life Sciences (CMLS), 36(12), 1434–1446. Wiegel, J., Ljungdahl, L., & Demain, A. 1985. The importance of thermophilic bacteria in biotechnology. Critical Reviews in Biotechnology, 3(1), 39–108. Wiegel, J., Ljungdahl, L., & Rawson, J. 1979. Isolation from soil and properties of the extreme thermophile Clostridium thermohydrosulfuricum. Journal of bacteriology, 139(3), 800–810. Wingard, J., Merz, W., & Saral, R. 1979. Candida tropicalis: a major pathogen in immunocompromised patients. Annals of Internal Medicine, 91(4), 539. Wisselink, H. W., et al. 2007. Engineering of Saccharomyces cerevisiae for efficient anaerobic alcoholic fermentation of L-arabinose. Appl Environ Microbiol, 73(15), 4881– 91. Wisselink, H. W., et al. 2007. Xylose-metabolizing Saccharomyces cerevisiae strains overexpressing the TKL1 and TAL1 genes encoding the pentose phosphate pathway enzymes transketolase and transaldolase. Applied Environmental Microbiology, 73, 4881– 4891. Wood, B. & Ingram, L. 1992. Ethanol production from cellobiose, amorphous cellulose, and crystalline cellulose by recombinant Klebsiella oxytoca containing chromosomally integrated Zymomonas mobilis genes for ethanol production and plasmids expressing thermostable cellulase genes from Clostridium thermocellum. Applied and Environmental Microbiology, 58(7), 2103–2110. Yan, R., et al. 1988. Expression of Solvent-Forming Enzymes and Onset of Solvent Production in Batch Cultures of Clostridium beijerinckii ( Clostridium butylicum”). Applied and Environmental Microbiology, 54(3), 642–648. Yomano, L., York, S., & Ingram, L. 1998. Isolation and characterization of ethanol-tolerant mutants of Escherichia coli KO11 for fuel ethanol production. Journal of Industrial Microbiology and Biotechnology, 20(2), 132–138. Yoshioka, K. & Hashimoto, N. 1981. Ester Formation by Alcohol Acetyltransferase from Brewers”Yeast. Agricultural and Biological Chemistry, 45(10), 2183–2190. Zaldivar, J., Nielsen, J., & Olsson, L. 2001. Fuel ethanol production from lignocellulose: a challenge for metabolic engineering and process integration. Appl Microbiol Biotechnol, 56(1–2), 17–34. Zaldivar, J., Nielsen, J., Olsson, L. 2001. Fuel ethanol production from lignocellulose: a challenge for metabolic engineering and process integration. Applied Microbial Biotechnology, 56, 17–34. Zeikus, J., Ben-Bassat, A., & Hegge, P. 1980. Microbiology of methanogenesis in thermal, volcanic environments. Journal of bacteriology, 143(1), 432–440. Zeikus, J. 1980. Chemical and fuel production by anaerobic bacteria. Annual Reviews in Microbiology, 34(1), 423–464. Zeikus, J., et al. 1980. Isolation and characterization of a new, methylotrophic, acidogenic anaerobe, the Marburg strain. Current Microbiology, 3(6), 381–386. Zeikus, J. G., Jain, M. K., & Elankovan, P. 1999. Biotechnology of succinic acid production and markets for derived industrial products. Applied Microbiology and Biotechnology, 51(5), 545–552.

P1: SFK/UKS P2: SFK c08 BLBS075-Hood December 15, 2010

21:54

Trim: 244mm×172mm

8 Fermentation Organisms for 5- and 6-Carbon Sugars

197

Zeng, A., et al. 1993. Pathway analysis of glycerol fermentation by Klebsiella pneumoniae: regulation of reducing equivalent balance and product formation. Enzyme and Microbial Technology, 15(9), 770–779. Zhang, M., et al. 1995. Metabolic engineering of a pentose metabolism pathway in ethanologenic Zymomonas mobilis. Science, 267(5195), 240. Zhou, B., Martin, G., & Pamment, N. 2008. Increased phenotypic stability and ethanol tolerance of recombinant Escherichia coli KO11 when immobilized in continuous fluidized bed culture. Biotechnology and Bioengineering, 100(4).

P1: SFK/UKS P2: SFK c08 BLBS075-Hood December 15, 2010

21:54

Trim: 244mm×172mm

P1: SFK/UKS P2: SFK c09 BLBS075-Hood December 15, 2010

20:55

Trim: 244mm×172mm

Chapter 9

Pretreatment Options Bradley A. Saville

9.1 Overview of Pretreatment Technologies 9.1.1

History

Strategies to pretreat lignocellulosic biomass have been developed and refined for more than 100 years, originating with acid-pretreatment systems, followed by steaming and steam explosion systems for biomass fractionation developed in the 1920s. More recently, strategies involving solvents and other chemical agents have been proposed. In spite of this relatively long history, research into pretreatment technologies has increased rapidly over recent years, largely due to the fact that pretreatment holds the key to unlocking the potential for cellulosic biofuels. Unquestionably, the effects of pretreatment are pervasive, impacting overall biofuel yields, capital and operating costs, enzyme utilization, fermentation, distillation, and waste disposal. This chapter provides some insights into what is (and is not) known about pretreatment, its impact on constituents of lignocelulosic feedstocks, challenges and opportunities for scale-up, and process/economic implications. An overview of “mechanistic” effects of pretreatment on lignin, cellulose, and hemicelluloses is provided, followed by comments about methods to assess pretreatment severity, to further our understanding of how pretreatment works. Classifications of pretreatment technology and some examples are also provided. A comparison is then presented for what is known about laboratory and pilot-scale or commercial pretreatment systems and technologies, with particular attention on prospects and challenges for commercialization of pretreatment technologies. Some of the significant process issues and trade-offs associated with different pretreatment technologies are presented, with a view on process economics and prospects for commercialization.

Plant Biomass Conversion, First Edition. Edited by Elizabeth E. Hood, Peter Nelson and Randall Powell.
C 2011 John Wiley & Sons Inc.

199

P1: SFK/UKS P2: SFK c09 BLBS075-Hood December 15, 2010

200

20:55

Trim: 244mm×172mm

Plant Biomass Conversion

Ultimately, commercial scale implementation is the goal of all of the research, development, and deployment activities related to pretreatment. While we can learn a lot in the laboratory, much can be learned from the significant history of pretreatment systems evaluated at the pilot and commercial scale. There is a huge gulf between the lab and the plant—a better understanding of both practical and economic issues can thus improve the likelihood of success. It is also true, as noted by Voltaire, that “the perfect is the enemy of the good.” There may never be a “perfect” pretreatment process, given the diversity of feedstocks and process objectives that must be met. Nonetheless, some desirable features or outcomes from a pretreatment process have been identified: 1. 2. 3. 4.

Results in a high recovery of all carbohydrates, with a minimum of degradation products. Has a low capital and operating cost. Produces a highly digestible solids fraction amenable to subsequent enzyme hydrolysis. Minimizes the need for pre- or postprocessing, either in the form of mechanical size reduction or downstream detoxification. 5. Operates at a sufficiently high solids loading to avoid dilution of sugars and ethanol that would otherwise adversely affect downstream processing costs. 6. Is able to process a wide variety of lignocellulosic feedstocks. While a perfect pretreatment process would satisfy all six of the desirable traits, it is likely that achieving four or five of the right traits may enable commercialization of a pretreatment process that is simply “good enough.” Which four or five traits, of course, could be the subject of much debate. Others cite different (or additional) desirable features, such as the need to generate relatively pure and valuable coproducts, the overall yields of fermentable sugars should be close to 100%, and a sugar-rich, fermentable liquid hydrolyzate should be generated during pretreatment. While on the surface these attributes seem laudable, they may be at odds with the practical and economic realities faced when a commercial system is developed and should not be viewed as absolutely necessary. For example, production of a fermentable liquid hydrolyzate during pretreatment is unnecessary if the pretreatment facilitates enzyme hydrolysis to such an extent that the fermentable sugars can be generated during a downstream hydrolysis or simultaneous saccharification and fermentation (SSF) process. Similarly, the need to generate high yields of fermentable sugars should not over-ride the rate at which sugars are hydrolyzed, nor is it as critical if high-value coproducts are produced. On the other hand, a highly efficient pretreatment process that leads to high biofuel production rates and yields may not have the same dependence on value-added coproducts. Unquestionably, the economics of the resulting commercial scale process will dictate which features are most important.

9.1.2

Mechanistic Assessment of Pretreatment

Laboratory studies have been the basis for much of what has been learned about the effects of pretreatment of lignocellulosic feedstocks, although commercial and pilot plant operations have also contributed valuable knowledge. Properties such as cellulose crystallinity, degree of polymerization (DP), lignin structure, lignin removal, hemicellulose solubilization, cellulose digestion, and accessible surface area have been proposed as key properties influenced by pretreatment. Others also consider the degree of acetylation of hemicelluloses to be an

P1: SFK/UKS P2: SFK c09 BLBS075-Hood December 15, 2010

20:55

Trim: 244mm×172mm

9

Pretreatment Options

201

important feature for some feedstocks and pretreatment processes. In this section, the impacts of pretreatment on hemicellulose, cellulose, and lignin are examined. Effects on Hemicellulose Hemicellulose is a highly substituted polymer of various 5-carbon and 6-carbon sugars, including xylose, galactose, arabinose, mannose, and glucose. The polymer may also contain acetyl groups, galacturonic acids, and glucuronic acids, among others. The structure of hemicelluloses is typically amorphous, making it more susceptible to hydrolysis. Its DP is also lower, often 500–3,000, meaning that shorter-chain oligosaccharides can be readily produced. Hemicellulose plays a key role in cell wall structure, due to hydrogen bonds with adjacent cellulose microfibrils and lignin, resulting in a strong network that provides the structural backbone of the cell wall. The resulting matrix imposes a barrier that limits access of cellulases to the underlying cellulose, limiting hydrolysis. Consequently, one key goal of pretreatment is to sufficiently disrupt the structure of hemicellulose so that subsequent enzymatic hydrolysis can proceed in an efficient manner. Some consider complete solubilization of hemicellulose the ultimate goal of pretreatment, and there is some merit to this argument, particularly if the solubilized C5 monomers and oligomers are to be immediately isolated and fermented. However, complete solublization often requires very severe pretreatment conditions—conditions that often lead to degradation of monomers and oligomers—thus forcing a trade-off between “early” versus “late” recovery of xylose and other carbohydrates. In the latter case, a lower severity (or alkaline) pretreatment would be chosen, which minimizes the formation of degradation products, but also sufficiently disrupts the structure of the lignocellulosic feedstock to enable efficient hydrolysis with a mixture of cellulases and hemicellulases. This strategy would certainly be preferred when a combined C5+C6 fermentation is undertaken. Acetyl groups that are part of the xylan backbone can also play an important role in pretreatment efficacy and impact downstream processes. Autohydrolysis by steam explosion and liquid hot water (LHW) pretreatments rely in part on acetic acid generated when acetyl groups are liberated during pretreatment. Removal of acetyl groups has been reported to enhance enzymatic hydrolysis by improving access to cellulose and xylan (see Kumar et al. (2009), for a detailed summary). However, the resulting acetic acid can also inhibit the activity of some (but not all) cellulase and xylanase preparations. Greater deacetylation generally occurs at higher severity; however, acetyl xylan esterases (AXE) could provide a more elegant pathway to acetyl group removal, while avoiding the adverse consequences of a higher pretreatment severity. Effects on Cellulose Pretreatment affects cellulose in a variety of ways, many of which are not fully understood. The crystallinity of cellulose, its DP, particle size, and accessible surface area have all been postulated as potentially significant features influenced by pretreatment, and/or responsible for “recalcitrance” during enzymatic hydrolysis. However, many of these properties are interrelated, and establishing the dominant property is difficult. Pretreatment influences specific surface area by physical effects such as size reduction that impact fiber length and width, and by physicochemical factors that cause removal of hemicellulose and lignin, increasing the internal surface area and pore volume while rendering the fiber more accessible to hydrolytic enzymes. A complicating factor is the fact that pore

P1: SFK/UKS P2: SFK c09 BLBS075-Hood December 15, 2010

202

20:55

Trim: 244mm×172mm

Plant Biomass Conversion

volumes are radically altered by drying. Consequently, results from trials using dried pulps must be interpreted with caution. However, the beneficial impacts of pore size are supported by research with swollenins that increase swelling of pulp fibers and suggest a synergism with cellulases that lead to enhanced hydrolysis. Changes in cellulose hydrolysis rates have often been attributed to a change in the proportion of amorphous cellulose to crystalline structure. Similarly, pretreatments that reduce crystallinity are generally considered beneficial. However, many studies that suggest a cause and effect relationship between crystallinity and hydrolysis rate have been performed with ideal substrates comprised mainly of cellulose. Caution is therefore required when extrapolating these conclusions to real substrates. Furthermore, physical pretreatments such as milling influence both crystallinity and the specific surface area, both of which, in turn, are correlated with the DP. Consequently, when viewed collectively, while correlating ease of hydrolysis to any of these parameters is possible, it is also likely that the most important attribute(s) will be a function of feedstock, fiber storage conditions, and pretreatment. Among the various pretreatment methods described below, virtually all of them increase accessible surface area, and most influence the DP, albeit to varying degrees. In spite of these broad similarities, dramatic differences in ease of hydrolysis can still be observed, indicating that hydrolysis may be influenced by more subtle structural features of the cellulose. These feature include the hydrophobicity of the surface and the types of bonds present in the resulting pretreated material (Kumar et al., 2009), which may indicate changes in cellulose, but may also reflect changes in porosity and intermolecular bonds between cellulose, hemicellulose, and lignin.

Effects on Lignin Pretreatment influences lignin in a variety of ways. Some, like ammonia fiber expansion (AFEX), modify lignin, while solvent-based pretreatments typically delignify fiber, resulting in a recoverable lignin fraction. In contrast, other pretreatments are suggested to leave the lignin apparently intact within the solid residue following pretreatment. However, the multiplicity of structures that comprise lignin are unlikely to remain intact, particularly upon exposure to heat, acids, solvents, or other chemical reagents at temperatures exceeding the glass transition temperature. Indeed, artificially high lignin recoveries are possible, even likely, if a Klason lignin analysis is performed, which captures “pseudo lignin” derived from degradation of sugars and from low molecular weight phenolic compounds (Chandra et al., 2007; Glasser and Wright, 1998). The form/structure of “lignin” remaining following pretreatment can have an important effect on downstream processes. Lignin can act as a physical barrier limiting access to cellulose and can inhibit cellulases through so-called “nonproductive” binding. Lignin that is solubilized then condensed is especially problematic, blocking enzyme access to underlying substrates. Furthermore, hydrophobicity may change as the structure of lignin is altered, impacting cellulase adsorption and inhibition. Surfactants such as TWEEN and polyethylene glycol have improved enzymatic hydrolysis, likely by blocking binding sites on lignin that would otherwise strongly adsorb cellulase. As reviewed by Chandra et al. (2007), such binding interactions are influenced by the prevalence of aliphatic hydroxyl groups, phenolic hydroxyl groups, and carboxylic groups on the residual lignin. Readers interested in detailed structural/chemical effects of pretreatment on lignin may also wish to review the comprehensive data from ESCA and FTIR analyses presented by Kumar et al. (2009).

P1: SFK/UKS P2: SFK c09 BLBS075-Hood December 15, 2010

20:55

Trim: 244mm×172mm

9

Pretreatment Options

203

Release of free phenolic groups may also occur in parallel with delignification; these phenolic species are more concentrated in solubilized lignin forms and may also inhibit enzymes and fermentation organisms. Lignin-mediation inhibition is highly dependent upon the enzyme formulation and substrate: Berlin et al. (2005) observed a greater than 3-fold difference in lignin-mediated binding between different sources of cellulases. Furthermore, lignin in pretreated Douglas fir had a greater impact on hydrolytic activity than lignin in pretreated poplar. This emphasizes the clear need to tailor enzyme selection to both the type of fiber and pretreatment process. This also implies that using the same enzyme formulation to assess hydrolysis efficiency for different substrates and pretreatments can lead to misleading conclusions.

9.1.3

Severity Factor Concept

The concept of a severity factor for pretreatment has its origins in similar developments for coal liquefaction, cracking reactions, and pulp and paper processes. The original concept of a severity factor for fractionation of biomass dates to a pioneering paper by Overend and Chornet (1987), which led to the oft-cited severity parameter for lignocellulosics: Log (Ro ) = log[t ∗ exp((T − Tref )/14.75)]

(9.1)

Regrettably, this expression for a severity factor is often improperly used, applied to processes and systems that violate the underlying assumptions stated by Overend and Chornet in their original manuscript. This has led to confusion and improper comparisons between research outcomes generated using different pretreatment methods and/or with different process objectives. It is thus important to understand the limitations and assumptions underlying the severity factor approach, and the various strategies subsequently developed to address these limitations and extend the concept to different pretreatment methods. Fundamentally, the reaction ordinate (Ro ) must take into account potential nonlinearity in reaction kinetics, the temperature-dependence of the Arrhenius parameter, the presence (or absence) of exogenous catalysts such as acids or solvents, feedstock heterogeneity, and perhaps most importantly, differences that arise between batch and continuous systems. The expression for log(Ro ) cited above applies to conditions whereby the substrate temperature is constant, the underlying fractionation kinetics are linear (i.e., first order), and there are no exogenous acids or solvents. A more general treatment was presented by Montane et al. (1998), who presented successively more general forms of the severity parameter that take into account different kinetics and process conditions. For example, Equation 9.2 applies to a simple homogeneous system in which the catalyst concentration does not vary with time: 


t Ro =

exp o

1 ωo



Tref 1− T

 dt

(9.2)

where T is the time-dependent temperature of the biomass feedstock, ωo is equal to RT ref /EA = ω/T, and is thus a characteristic parameter related to the feedstock, gas constant, and reference temperature appropriate for the system. ωo also clearly depends upon the reaction kinetics (activation energy, EA ), which will depend upon the feedstock, desired reaction, and the presence of exogenous catalysts or solvents.

P1: SFK/UKS P2: SFK c09 BLBS075-Hood December 15, 2010

20:55

Trim: 244mm×172mm

Plant Biomass Conversion

204

Depolymerization of lignin and solubilization of hemicelluloses will thus lead to different values of ωo (or ω). For example, Overend and Chornet arrived at a value of ω = 14.75 to describe hemicellulose solubilization from northern hardwoods, while Chum et al. (1990) found that values of ω = 10.0 and 11.0, respectively, best described lignin and xylan solubilization during acid-catalyzed organosolv pretreatment of aspen. Changing ω from 14.75 to 11.0 increases Ro by a factor of 10, and increases logRo by 1, showing the substantial difference between steam explosion and acid-catalyzed organosolv pretreatment. Unfortunately, some researchers fail to recognize the need to adapt the value of ω to their specific feedstock and pretreatment process, leading to a misleading assessment of pretreatment severity and improper comparisons of pretreatment outcomes based on logRo . Equation 9.2 accounts for time-dependent changes in temperature, such as in a two-step batch process where wood chips are preheated with steam for a certain period of time before initiating a high pressure steam explosion. However, if temperature is invariant, a simplified form is obtained, analogous to that shown in equation 9.1 for hemicellulose solubilization of hardwoods during steam explosion. Many researchers overlook the initial fiber warm-up period before pretreatment and the postpretreatment period when fibers are cooling, in spite of the fact that solubilization, depolymerization, and degradation reactions of lignocellulosic materials will almost certainly take place during these periods. Ignoring these pre- and postpretreatment periods may only lead to minor errors in logRo for some processes and conditions, especially if the pretreatment time is long, but in others, ignoring these factors is likely to lead to significant underestimation of the severity experienced by the substrate. As noted by Heitz et al. (1987), the extent of hydrolysis during a preheating period (typically 30–60 minutes) can be significant compared to the reactions that occur during the constant temperature pretreatment period, which is often only 5 minutes in duration (or less). Unfortunately, biomass is not a homogenous substance, and if the heterogeneity of the feedstock is significant, equation 9.2 must be adapted to account for nonlinear kinetics of a distribution of species and nonisothermal behavior: 


t Ro =

exp o

1 ωo

 1−

Tref T

 t y−1 dt

(9.3)

where γ is a parameter that describes the shape of the reactivity/activation energy curves in a heterogeneous system. A value of γ = 1 would correspond to a homogeneous system, resulting in the form shown in equation 9.2. For example, Montane et al. determined that a value of γ = 0.5928 appropriately described pentosan removal from birch via dilute acid pretreatment. The form corresponding to equation 9.3 for an isothermal system is  Ro = exp

1 ωo



Tref 1− T



ty y

(9.4)

The nonlinearity of the kinetics and heterogeneity of the substrate can also lead to substantial differences in logRo . For example, with t = 4 minutes and T = 200◦ C, equation 9.1 predicts a logRo value of 3.55. Using the same value for ω (14.75) and including γ = 0.5928 leads to an Ro value nearly 4 times greater, corresponding to a logRo value of 4.13.

P1: SFK/UKS P2: SFK c09 BLBS075-Hood December 15, 2010

20:55

Trim: 244mm×172mm

9

Pretreatment Options

205

Some authors cite corrected (or combined) severity factors that include pH in an attempt to account for acid added to the system. Such a correction can be useful, as long as it is not applied to the original/simple form for Ro presented in equation 9.1, but rather, also includes a modified value for ω that accounts for the presence of the acid in the system. Unfortunately, this is often not the case. The proliferation of inappropriate applications of the severity parameter makes it difficult to use logRo to render meaningful comparisons between pretreatment systems, assess the impact of acid or base supplementation, or evaluate potential benefits of solvents. For example, although acid-catalyzed steam explosion with SO2 may lead to shorter pretreatment times or lower temperatures, the actual severity of the pretreatment may be greater due to the catalytic action of the exogenous acid. However, comparisons based upon equation 9.1 alone would lead one to conclude that the SO2 -catalyzed system generated lower severity conditions. Similar issues may arise when trying to extrapolate outcomes from batch experiments to continuous systems. Ropars et al. (1992) noted significant differences between their batch (laboratory) system and results from their pilot scale continuous Stake II steam explosion system, observing that the batch system required SO2 supplementation in order to achieve optimal sugar yields, whereas sufficient yields were achieved without acids (i.e., via autohydrolysis) when the continuous system was used. A simple comparison based on logRo was not possible. Hosseini and Shah (2009) recently examined severity parameters in the context of a firstprinciples assessment of the time required to heat biomass and the time for molecules to diffuse through the substrate. A comparison of diffusion length, treatment time, and system size imply that large concentration gradients may exist in pretreatment systems (or fiber being pretreated). Such diffusion gradients are magnified in a three-phase system (solid, liquid, and vapor) compared to a two-phase system (vapor + solid), due to solubility issues and diffusional barriers imposed by the liquid in the system. To compensate, liquid-phase systems would need to incorporate an extensive “presoaking” to ensure adequate distribution of a catalyst (acid/base) or solvent. This investigation also implies that the “effective severity” decreases as the size of the biomass feedstock is increased; a problem with “undertreated centers” may arise. An increase in temperature or residence time or a decrease in pH would be needed in order to compensate for a larger chip size. The authors ultimately proposed a modified severity index that accounts for both pretreatment time and the characteristic diffusion time, suggesting that this may be a more effective way to scale up pretreatment systems. They also concluded that optimizing chip/particle size could reduce chemical and energy demand during pretreatment, potentially increasing the energy efficiency of pretreatment by 50%.

9.2 Pretreatment Classification Pretreatment processes are generally classified into three broad categories: (1) mechanical processes that primarily reduce the size of the incoming lignocellulosic feedstock, (2) chemical pretreatment processes that rely on the presence of acids, bases, solvents, or other (bio)agents to extract select components of the feedstock, or modify its structure, and (3) thermomechanical or thermochemical processes that rely on a combination of heat, pressure, and mechanical energy to alter lignocellulosic feedstocks. In truth, there is some debate regarding classification, partly because some processes do not uniquely fall into a particular category. Some experts consider chemical pretreatment only in the context of exogenous chemical addition, while others include chemicals endogenous to the substrate that are released/generated during pretreatment. The fact that some of these

P1: SFK/UKS P2: SFK c09 BLBS075-Hood December 15, 2010

206

20:55

Trim: 244mm×172mm

Plant Biomass Conversion

endogenous chemicals may be released during mechanical or thermomechanical processing further clouds attempts at classification. Consequently, these classifications should only be viewed in the most general way. Several reviews provide a detailed assessment of many of these pretreatment processes (e.g., Chandra et al., 2007; Galbe and Zacchi, 2007; Hu et al., 2008; Mosier et al., 2005; S´anchez and Cardona, 2008; Sun and Cheng, 2002).

9.2.1

Mechanical Pretreatment Processes

Mechanical pretreatment processes include comminution methods such as ball milling, compression milling, hammer milling, szego milling, and disk refining. These comminution methods lead to biomass size reduction, the extent of which depends strongly on the level of energy input. Changes in cellulose crystallinity and DP may also occur. Many of these technologies had their origins in the pulp and paper and grain processing industries and have thus been evaluated at a large scale. Generally speaking, using comminution processes alone is not economically favorable, due to the high energy demand required in order to achieve sufficient size reduction to render the biomass amenable to further processing. Hardwoods may require in excess of 100 kWh/ton of feedstock, whereas agricultural residues are less energy intensive, requiring “only” 20–40 kWh/ton (Cadoche and L´opez, 1989). Nonetheless, comminution methods may be used in combination with some of the other pretreatment processes described below. In such a case, the energy demand is lower, requiring only modest size reductions preor postpretreatment. This category also includes irradiation strategies such as microwave, gamma ray, and electron beam pretreatments that have been tested at the laboratory scale. Although interesting outcomes have been noted, the prospects for commercial scale implementation of these technologies are minimal.

9.2.2

Chemical Pretreatment Processes

This category includes pretreatments that rely on exogenous supplementation with chemicals—either acids, bases, neutral compounds, or solvents. Each of these is discussed briefly below. Most of these processes are liquid–solid, although some are three-phase (gas–liquid–solid) and others are gas–solid. This has important implications for diffusion of the chemical catalyst; systems containing liquids will often require an extensive presoaking step in order to ensure sufficient distribution of the catalyst throughout the lignocellulosic substrate. The ratio of liquid to solid also has important implications upon economics. A high liquid:solid ratio can lead to superior and more uniform pretreatment, but can have significant adverse consequences on downstream processes and economics.

Acidic Systems Concentrated and dilute acid pretreatments have been investigated extensively and applied at a pilot and commercial scale. For economic reasons, most studies have used sulphuric acid, but nitric acid, hydrochloric acid, phosphoric acid, and others have also been explored. Sulphur dioxide also falls into this category, since it generates acid in combination with steam. Rigorously speaking, some organosolv pretreatments also fall into this category; in particular,

P1: SFK/UKS P2: SFK c09 BLBS075-Hood December 15, 2010

20:55

Trim: 244mm×172mm

9

Pretreatment Options

207

methanol, ethanol, and acetone are usually supplemented with hydrochloric acid or sulphuric acid (see e.g., Hallberg et al., 2009). In a dilute acid system, the dose typically ranges from 0.75 up to 5 wt%. Processing temperatures range from 120◦ C to 200◦ C, depending upon the solids loading and extent of “presoaking.” Single-stage and multi-stage pretreatments have been investigated. Liquid:solid ratios between 2:1 and 9:1 are common; low liquid:solids ratios require batch processing, whereas continuous processing is possible with high liquid:solid ratios. Concentrated acid systems usually operate at a liquid:solid ratio near 1.5:1, largely due to the higher concentration of acid, normally 10–30 wt%. High temperatures, high acid concentrations, and long residence times can enhance solubilization of hemicelluloses and lead to partial hydrolysis of cellulose. Indeed, near 100% hydrolysis of hemicellulose is possible. The DP and crystallinity of cellulose also decrease. However, these same conditions also encourage the conversion of the resulting sugar monomers into furfural and hydroxymethyfurfural inhibitors that will impact downstream hydrolysis and fermentation processes. Although temperature certainly contributes to sugar degradation, the rates of these processes are markedly enhanced under acidic conditions. Use of exogenous acids nearly always requires a post-pretreatment neutralization step (e.g., with calcium hydroxide) prior to enzymatic hydrolysis, resulting in gypsum as a coproduct. “Detoxification” to remove inhibitors is also common. If acid is recycled, at least 0.02 tonnes of gypsum will be generated for each tonne of feedstock processed (Hamelinck et al., 2005), corresponding to about 12,000 tonnes of gypsum in a 100 million L/y ethanol production facility. If acid recycling is not employed, gypsum production could approach 0.6–0.9 tonnes per tonne of feedstock or 550,000 tonnes per year in a 100 million L/y plant, creating a massive disposal issue. The economics of concentrated acid processes are contingent on downstream recovery and reuse of the acid, a factor that has both capital and operating cost implications. Use of acid requires corrosion-resistant materials for pretreatment process vessels, which will dramatically increase the capital cost of the pretreatment system. The impact is magnified if the process depends upon high temperatures and pressures, and under high liquid:solid ratios. An oft-unrecognized issue with sulphuric acid or sulphur-dioxide-catalyzed pretreatment is sulphur contamination of product and coproduct. Some of the exogenous sulphur will inevitably be found in the lignin, creating an emissions issue when the lignin is burned to make electricity. Perhaps even more problematic is sulphur contamination of the biofuel, especially in jurisdictions that mandate low-sulphur petroleum fuels. This latter issue alone may preclude commercial implementation of sulphuric acid and sulphur-dioxide-catalyzed pretreatment processes. Alkaline Systems Alkaline pretreatments include supplementation with lime, sodium hydroxide, and ammonia. The latter may be administered in liquid or gaseous form, in processes known as ammonia recycle percolation (ARP) and ammonia fiber expansion, respectively. Alkaline pretreatment processes generally operate at much lower temperatures, but with longer retention times compared to acid pretreatments. The lower temperature is critical, as it is often lower than the glass transition temperature for lignin. However, the longer retention times are an issue, because degradation of glucose and xylose begins at temperatures as low as 60◦ C, and the long retention times can lead to significant cumulative losses.

P1: SFK/UKS P2: SFK c09 BLBS075-Hood December 15, 2010

208

20:55

Trim: 244mm×172mm

Plant Biomass Conversion

Similar to Kraft pulping processes, alkaline pretreatments tend to have a marked effect on lignin due to disruption of ester bonds between lignin and xylan. The degree of delignification can often be tailored according to process conditions and catalyst dose. Some removal of hemicellulose also occurs, and the crystal structure of cellulose is also changed, leading to a less crystalline solid residue. Temperatures as low as 60◦ C–85◦ C have been reported with lime pretreatment, albeit with very long retention times (48–192 hours, occasionally up to 4 weeks). Shorter retention times (2–13 hours) have been reported in systems using pressurized oxygen and in systems where the temperature has been elevated to 100◦ C or 150◦ C. Nonetheless, even a retention time on the order of a few hours provides a significant practical impediment to commercial implementation, owing to the physical size of the system and demands for sufficient on-site supply of feedstock in order to achieve a satisfactory throughput. AFEX has been explored under a variety of conditions. Generally speaking, approximately 1–2 kg of ammonia is added per kg of dry biomass, which is then treated for about 30– 60 minutes at 90◦ C–100◦ C and about 1–3 MPa. These conditions minimize production of inhibitors from sugar degradation, a significant advantage of the AFEX process. The AFEX process is best suited to low lignin feedstocks such as agricultural residues and herbaceous crops; it has had limited success with softwoods, and moderate benefits have been observed with hardwood substrates, although higher pretreatment temperatures (180◦ C) have led to acceptable outcomes with some types of woody biomass. Commercial implementation of AFEX is predicated upon the cost of ammonia and the cost to recover and recycle ammonia, although low level losses are tolerable because dissolved ammonia can act as a nitrogen source for downstream fermentation. Safety issues associated with ammonia use create another obstacle to commercial implementation—seals for the high pressure reactors and recovery systems must be designed to completely prevent vapor loss, and containment systems will almost certainly be required. Ammonia recycled percolation operates in a similar fashion to AFEX, except that the ammonia is dissolved in liquid. Unfortunately, the presence of liquid creates a diffusional barrier for the catalyst, thus necessitating longer retention times, higher catalyst loadings, and/or higher temperatures in order to achieve satisfactory pretreatment. Temperatures of 160◦ C–200◦ C are often reported, using solutions containing 10–15 wt% ammonia and a 4:1 or 5:1 liquid:solid ratio. Neutral Systems LHW and controlled pH pretreatment processes operate under near-neutral conditions, although it must be recognized that hot water is an acid (pH ∼5) at high temperatures, and endogenous acids released during pretreatment may drive the pH even lower. LHW processes generally operate at pressures greater than 5 MPa, with temperatures between 170◦ C and 230◦ C. Retention times can be fairly long, typically 15 or 20 minutes, and occasionally up to 45 or 60 minutes. Continuous cocurrent and counter-current systems have been proposed, along with a semi-batch flow-through design whereby water flows continuously through a fixed bed of lignocellulosic substrates, carrying out dissolved/solubilized components with the liquid stream. The use of LHW has several key benefits, but also some limitations. Generally, size reduction of the incoming biomass is not required, nor is downstream neutralization of the pretreatment hydrolyzate. However, the use of liquids limits the solids (dry matter) content within the

P1: SFK/UKS P2: SFK c09 BLBS075-Hood December 15, 2010

20:55

Trim: 244mm×172mm

9

Pretreatment Options

209

reactor, which is typically in the range of 15% or 20% DM. The high solvent load also results in low sugar and alcohol concentrations, adversely impacting the economics of the downstream processes. The Inbicon/Dong Energy pilot plant (Denmark) is based upon LHW pretreatment; they indicate the ability to operate at dry matter loadings near ∼30%. A variation on the LHW process involves adding sufficient base (e.g., caustic, potassium hydroxide) to ensure that the pH remains between 5 and 7, which limits hydrolysis of oligomers into monosaccharides. This “controlled pH” approach has its origins in the pulp and paper industry from the 1930s. Solvent-Based Systems Lignin extraction and biomass pretreatment using organic solvents have received significant attention, building on developmental work in the pulp and paper industry, which led to the ALCELL process developed by General Electric and commercialized by Repap (Pye and Lora, 1991), among others. Methanol, ethanol, acetone, and glycols have proven effective for delignification, particularly when used in combination with acid or alkali catalysts. Although delignification can be achieved without exogenous acids, very high severities are required. Organosolv processes usually cook the substrate for 30 to 90 minutes at 1.5–3 MPa and 120◦ C–350◦ C. Shorter retention times and lower temperatures, typically 180◦ C–200◦ C, can be used if the solvent is supplemented with 1% hydrochloric or sulphuric acid. Commercial implementation of an organosolv process is contingent on cost-effective solvent recovery. Not only is this critical to minimize solvent costs, but any residual solvent may also adversely impact enzymatic hydrolysis and fermentation.

9.2.3

Thermochemical Pretreatment Processes

Uncatalyzed steam explosion (or autohydrolysis) is generally classified as the thermochemical process, although some classify it as an acid pretreatment process, partly because water is an acid at high temperatures and partly due to the fact that acetic acid released during pretreatment contributes to lignocellulose hydrolysis. Steam explosion treatment of lignocellulosic materials dates back to the early 1920s, leading to patents by Mason (1929) and by Babcock (1932) related to pretreatment of pine with saturated steam to generate soluble sugars suitable for fermentation. The so-called Masonite Steam Gun, developed based on the original Mason patent, served as the basis for patents by Foody (1984) and DeLong (1983) that describe the use of batch steam explosion in conjunction with extraction methods to separate hemicellulose, lignin, and cellulose. The batch steam explosion process continues to be used by Iogen in their Ottawa demonstration facility, incorporating several process modifications to improve upon the original Foody process. Continuous steam explosion originated in the early 1970s and is described in early literature from Stake Technologies (now SunOpta Bioprocess Inc.) and in patents issued to Brown (1980) and Brown and Bender (Brown and Bender, 1980). In its simplest form, continuous steam explosion involves no added chemicals, a process known as autohydrolysis. In contrast, in a catalyzed steam explosion process, exogenous acids or bases such as sulphuric acid, sulphur dioxide (SO2 ), or ammonia are added to the system. Acid catalysts further reduce the pH while increasing the digestion of fiber components, while alkaline catalysts typically impact

P1: SFK/UKS P2: SFK c09 BLBS075-Hood December 15, 2010

210

20:55

Trim: 244mm×172mm

Plant Biomass Conversion

(or extract) lignin and de-acetylate hemicellulose and alter the crystallinity of cellulose. Typically, the presence of an acid catalyst allows a reduction in retention time or temperature so that the apparent pretreatment severity is reduced. Operating temperatures between 160◦ C and 260◦ C have been cited (0.70–4.8 MPa), although temperatures between 190◦ C and 220◦ C are most common. Lower temperatures coupled with longer retention times tend to preserve individual components in a lignocellulosic feedstock, whereas more aggressive conditions (higher severity) tend to cause solubilization (and perhaps degradation) of hemicellulosic components. Continuous steam explosion is one of the few pretreatment technologies used at a pilot or commercial scale, including processing of oat hulls and birch fines (Finn Sugar, Finland), corn cobs (IFP), corn stover (COFCO—Chinese Oil and Foodstuffs Company, ZhaoDong City), rye grass straw (Weyerhauser, Oregon), sugar cane bagasse (Verenium, Louisiana), various cereal straws, and hardwoods, particularly poplar (Heitz et al., 1991).

9.2.4

Impact on Moisture Content and Hydraulic Load

The moisture content during pretreatment has a significant impact on downstream processes, and upon process economics. Pretreament processes that use liquids are often constrained to lower solids loadings compared to pretreatment processes that use vapors such as steam or ammonia. Most LHW, flow through, dilute acid, and ARP processes use a 25% solids loading or less (i.e., liquid:solid > 3:1), partly because the slurry must flow during these continuous processes and partly because of diffusional limitations imposed by liquids in the system. Similarly, liquid:solid ratios in organosolv processes are often 4:1 to 6:1. In contrast, the Stake steam explosion and AFEX processes can process feedstocks without addition of liquid water. These systems can directly process woody biomass with 45%–55% solids and agricultural residues with up to 85% or 90% solids. Other process steps implemented due to pretreatment can also contribute to the hydraulic load. For example, washing steps to remove inhibitors will lead to a lower solids slurry in downstream hydrolysis and fermentation operations while also diluting soluble monomers and oligosaccharides. Carrying extra liquid has several consequences: (1) a larger (and more expensive) pretreatment system, (2) lower sugar concentrations in the pretreatment hydrolyzate, due to dilution, (3) larger downstream reactors for hydrolysis and fermentation, (4) lower sugar concentrations following enzymatic hydrolysis, (5) lower ethanol concentrations following fermentation or SSF, (6) a larger distillation system and more energy for distillation, and (7) a higher load on wastewater treatment systems. Eggeman and Elander (2005), in their assessment of the economics of pretreatment processes studied under CAFI (Consortium for Applied Fundamentals and Innovation), observed that total steam demand was directly correlated to the water concentration during pretreatment. A LHW process with 16.3 wt% solids required 27% more steam than an ARP process with 25 wt% solids. Furthermore, LHW pretreatment led to an ethanol concentration of only 3.1 wt% following fermentation—further evidence of the importance of hydraulic load. Considering all of the undesirable consequences caused by a low solids/high liquid system, it seems rational to select a pretreatment process that minimizes the hydraulic load and maximizes the solids content within the system. If, however, a low solids pretreatment system is used, it is equally imperative to remove the excess liquid as quickly as possible, to minimize the economic impact on subsequent processing steps.

P1: SFK/UKS P2: SFK c09 BLBS075-Hood December 15, 2010

20:55

Trim: 244mm×172mm

9

Pretreatment Options

211

9.3 Laboratory vs. Commercial Scale Pretreatment—What Do We Really Know? 9.3.1

Laboratory Studies

Although pretreatment of lignocellulosics has been practiced for more than a century, knowledge regarding molecular level and mechanistic impacts of pretreatment is relatively young. Over the past decade, the coordinated efforts of various researchers under the CAFI study have led to standardization of methods to assess the efficacy of pretreatment, enabling comparisons that were previously impossible. These studies, while valuable, obviously neither can include the full spectrum of pretreatment technologies, nor can they objectively assess pretreatment technologies that have dramatically different underlying mechanisms. For example, in selecting a particular enzyme cocktail to evaluate the ease of hydrolysis of the pretreated residue, pretreatment methods such as AFEX that preserve hemicelluloses (or generate xylooligosaccharides) are disadvantaged compared to those that cause significant solubilization of hemicelluloses. Clearly, a cellulase cocktail with enhanced xylanase activity could convert the residual hemicellulose and xylo-oligosaccharides, and facilitate a more objective assessment of “ease of hydrolysis.” For example, Ohgren et al. (2007) noted that enzymatic hydrolysis of steam-pretreated corn stover was dramatically enhanced by supplementation of the enzyme cocktail with Multifect xylanase (Genencor); interestingly, glucose yields increased more than xylose yields. Furthermore, the degree of benefit was highly dependent upon pretreatment conditions: a 13% (absolute) increase in glucose yield was observed with stover pretreated with SO2 , while an increase of nearly 25% was observed with autohydrolysis (i.e., no SO2 ) using the same pretreatment temperature and retention time. Lignin structures in the solids residues following pretreatment depend upon both the feedstock and pretreatment process/technology. As noted by Berlin et al. (2005), interactions with these lignin structures are dependent upon the microbial source of the hydrolytic enzymes, leading to different degrees of nonspecific binding and cellulase inhibition. This provides further evidence of the need to tailor an enzyme cocktail to the feedstock and pretreatment process. Laboratory studies often compare pretreatment processes on the basis of yield—either xylose yield following pretreatment or sugar yield following enzymatic hydrolysis. These comparisons, while valuable, do not necessarily account for differences in the hydrolysis rate that may arise due to differences in particle size, crystallinity, fiber structure, or inhibitors. In addition, many of these comparisons rely on assays with low solids loadings, using “cellulase” alone on solids that are prewashed prior to enzyme hydrolysis. Such methods, while perhaps necessary or practical in the context of laboratory work or rapid throughput assays, can mask problems caused by inhibitors that would otherwise be observed with unwashed solids at high solids loadings. Consequently, outside of the valuable mechanistic and molecular level information generated, outcomes and comparisons arising from laboratory scale studies must be interpreted with some caution, because they likely do not reflect outcomes that will be observed in a pilot or commercial scale process.

9.3.2

Pilot/Demonstration Scale Studies

Among all of the pretreatment technologies developed to date, only a select few have been practiced at a large scale for biofuel production, under continuous operation: (1) dilute acid,

P1: SFK/UKS P2: SFK c09 BLBS075-Hood December 15, 2010

212

20:55

Trim: 244mm×172mm

Plant Biomass Conversion

(2) LHW, and (3) steam explosion, either in the form of autohydrolysis or catalyzed with SO2 or sulphuric acid. It should be noted that organosolv and ammonia pretreatments for lignocellulosics have also been evaluated at the pilot scale, but for different applications: organosolv by REPAP in the 1980s for pulp processing, and ammonia (AFEX) by Stake Technology and Rhone-Poulenc in the 1990s to create dissolving pulps. Among these, publications and reports about systems using steam explosion and acid hydrolysis have been most common; some of these pilot scale activities are discussed below. Muzzy et al. (1983) describe the use of a continuous Stake IITM Hydrolyzer to process yellow poplar. The system could process 1 ton of wood chips per hour (50% moisture), at a maximum pressure of ∼2.4 MPa. They explored retention times between 3 and 7 minutes, and temperatures from 215◦ C to 225◦ C. Maximum xylose recovery (as soluble sugars) occurred at 215◦ C and 3-4 minutes of retention time, which also corresponded to maximum recovery of intact cellulose. Heitz et al. (1991) evaluated the fractionation of debarked poplar in a 4 ton/h Stake II continuous steam explosion system. Different conditions corresponding to maximal recovery of cellulose, lignin, or hemicellulose were identified. The authors discussed opportunities to produce value-added coproducts derived from pentoses, along with strategies to extract relatively pure lignin and cellulose. The authors also noted the benefits of a continuous steam explosion system compared to a batch operation, describing limitations with heat transfer rates in a batch system that lead to “significant undercooking of chip centers for reaction times of less than 2 minutes at high severities.” In their trials, Heitz et al. processed 1,500 kg quantities of poplar over 45 minutes, at temperatures from 180◦ C to 230◦ C and retention times ranging from 0.7 to 4 minutes. These conditions correspond to severity factors (logRo from equation 1) between 2.20 and 4.13. Following autohydrolysis by steam explosion, the fractions collected were processed to (1) recover soluble sugars via a water wash, (2) separate lignin and cellulose via alkali delignification, and (3) precipitate and recover lignin. Limited solubilization of sugars was observed for logRo values less than 3; about 30% solubilization was observed when logRo exceeded 4.0. The fraction of pentose/xylan remaining in the fiber decreased as logRo increased, while the fraction recovered in soluble form increased with severity up to logRo = 3.8, after which it declined. Overall (total) pentose recovery declined as the severity increased, indicative of pentose degradation. Lignin recovery increased as severity increased, while the DP of the residual cellulose decreased significantly, falling below 600 when logRo exceeded 3.8. Glucose yields from pretreated poplar increased monotonically up to logRo = 3.8, after which the yield leveled off at ∼90% following a 48 hour enzymatic hydrolysis. During the 1980s and early 1990s, the Institut Francais du Petrole (IFP) conducted an extensive developmental program aimed at producing acetone and butanol from lignocellulosic biomass. Initial laboratory-scale work eventually led to the commissioning of a pilot plant in Soustons, France. This facility included a Stake II autohydrolysis system, on-site enzyme production, enzyme hydrolysis, fermenters for production of acetone and butanol from sugars generated from corn cobs, and a distillation system to purify the products. Ropars et al. (1992) describe the process outcomes arising from the Soustons pilot plant activities, while Nativel et al. (1992) describe the equipment installed at the facility. Pretreatment was performed at 1–2.5 MPa for either 3 or 5 minutes, corresponding to severity factors (logRo ) from 2.99 to 3.90. Ropars et al. noted significant differences between their batch (laboratory) system and continuous steam explosion, observing that optimal sugar yields were achieved at lower severities when the continuous system was used, and furthermore, that higher sugar yields were obtained with the Stake system compared to the batch system. They attributed the beneficial outcomes from continuous steam explosion to

P1: SFK/UKS P2: SFK c09 BLBS075-Hood December 15, 2010

20:55

Trim: 244mm×172mm

9

Pretreatment Options

213

Strong mechanical and other effects originating from compression during formation of the feed plug . . ., as well as non-equivalence of cooking time in continuous and batch pretreatments, which are not taken into account in the definition of the pretreatment severity index.

Although Ropars et al. noted that their lab scale pretreatment required acid supplementation in order to achieve sufficient carbohydrate yields, the superior efficiency of the continuous process allowed them to achieve acceptable yields without exogenous acids, i.e., continuous autohydrolysis eliminated the need for acids that were otherwise required in a batch steam explosion system. The Soustons pilot plant work demonstrated optimal glucose yields at a logRo less than 3.7, less severe than the optimal conditions for glucose production from poplar reported by Heitz et al. Optimal conditions were achieved at 1.3–1.4 MPa, a retention time of 3 min in the continuous hydrolyzer, and a feed rate of about 1.6 dry tonnes per hour. Steam consumption was reported at 2 tonnes per tonne of dry corn cobs. The pretreated solid product contained 17% lignin, 36% glucans, 27% xylans, 2% arabinans, and 2% ash, with a total yield of 92% compared to the raw feedstock. Wayman and Parekh (1988) compared several large-scale pretreatment systems for SO2 catalyzed steam pretreatment of pine, aspen, and corn stover. Systems evaluated included the Stake II continuous steam explosion system, the Wenger reactor, a Sunds Defibrator, and a Masonite batch steam explosion system. The Stake system was capable of processing about 1.5 tonnes/h of feedstock, while the Wenger reactor had a capacity of 100–150 kg/h. Both systems were adapted to deliver SO2 into the hydrolyzer, either as a liquid solution or as a gas. They noted issues with the presence of coarse fiber bundles at the outlet (or conclusion) of the pretreatment process, an outcome that they stated as “not surprising, since the Sunds Defibrator and the Masonite gun are used primarily to make fiberboard, and the products are normally put through a mechanical attrition process to separate the fiber bundles.” Ultimately, they concluded that relatively harsh pretreatment conditions would be needed in these two systems in order to produce a finely divided product suitable for further processing into biofuels. Wayman and Parekh noted that comparable outcomes were achieved in continuous trials using the Stake system at 208◦ C with 2 minutes of retention time and the Wenger reactor at 200◦ C with 45 seconds of retention time. Both systems were supplemented with 2.5% SO2 . Total solids yields using pine, aspen, and corn stover were comparable at ∼70%. Nearly complete solubilization of hemicellulose sugars was observed using the Stake system to process pine; high levels of sugar production were also noted with aspen, but soluble sugar levels were quite low following pretreatment of corn stover. The impact of SO2 was most pronounced with pine; however, ethanol yields from aspen were fairly close under both SO2 catalyzed and autohydrolysis conditions—346 vs. 320 L/dry tonne, respectively. Nonetheless, there was a clear difference in the process—using SO2 led to more sugar production prior to fermentation, while autocatalysis required more sugar conversion during the fermentation step. Ethanol yields with SO2 -catalyzed corn stover reached 388 L/dry tonne and 372 L/dry tonne with pine, using Pichia stipitis for fermentation. In an earlier study, Wayman et al. (1986) explored the use of SO2 -catalyzed steam explosion for pretreatment of mixed pine and spruce chips. They fed commercial chips at a rate of 1.5 tonnes/h to a Stake II steam explosion system, operating at 1.2–1.7 MPa with a residence time of 2 minutes and supplemented with 2.0–2.6% SO2 (dry wood basis). They also performed a control trial without SO2 addition. Autohydrolysis at 1.7 MPa (208◦ C) for 2 minutes led to 77% solubilization of hemicellulose sugars and 8% solubilization of glucan. Most of the solubilized hemicellulose was present in oligomer form. By comparison, using the same

P1: SFK/UKS P2: SFK c09 BLBS075-Hood December 15, 2010

214

20:55

Trim: 244mm×172mm

Plant Biomass Conversion

steam pressure and retention time, 89% of the hemicellulose was solubilized and 11% of the glucan was solubilized if the system also included 2.0% SO2 . Virtually identical postpretreatment yields of sugars were achieved with autohydrolysis at 1.7 MPa compared to SO2 -catalyzed steam explosion at 192◦ C, with 2.6% SO2 added to the system. The authors ultimately concluded that augmenting steam explosion with SO2 led to approximately 25% higher ethanol titers from softwoods following fermentation with P. stipitis. Nonetheless, with recent advances in commercial xylanases, it is likely that this gap could be narrowed, due to efficient conversion of xylo-oligosaccharides generated during autocatalysis. Glasser and Wright (1998) discuss the use of a Stake II system to fractionate poplar, sugar cane bagasse, and peanut hulls. Following pretreatment, samples were processed to recover water-soluble sugars, lignin, and a cellulose-rich fiber fraction. Challenges in feeding stringy/fibrous materials such as bagasse were noted; wood chips were easier to process through the feeder system. The Stake II hydrolyzer operated at pressures up to 3.0 MPa and was fed up to 1 dry tonne/h of wood chips, or 135 dry kg/h of bagasse. Values of logRo ranged from 3.6 to 4.4. Depending upon pretreatment conditions, between 10% and 20% of the incoming biomass was solubilized. Poplar showed less variation in fractionation response, particularly between logRo = 3.9 and 4.3. By comparison, bagasse fractionation was optimal at logRo = 3.6, with higher severities leading to a rapid decline in the yield thereafter. Peanut hulls proved to be resistant to fractionation by steam explosion over the entire range of severities considered. Mass balance data from Glasser and Wright suggest that with logRo = 3.9 and 4.3, between 40% and 55% of the incoming xylan in poplar was unaccounted for in the final products, likely due to degradation of xylose and xylo-oligosaccharides. Higher severity led to greater solubilization of xylan and less intact xylan remaining in the solid fiber fraction after pretreatment. Approximately 10% of the lignin was water soluble, while the alkali-soluble fraction increased from about 35% at logRo = 3.8 to about 90% at logRo = 4.40, with a sharp increase when logRo exceeded 4.20. Beyond this point, the structure of lignin changes, particularly an increasing ratio of phenolic to aliphatic OH groups consistent with lignin depolymerization. Some formation of pseudo-lignin was also suspected. Increasing the severity caused a shift from high to low molecular weight lignin. The DP of the recovered fiber fell from about 1,100 at logRo = 3.8 to about 200 at logRo = 4.4. In the 1980s, the Tennessee Valley Authority built a pilot plant that incorporated a Sunds Defibrator and a Kamyr digestor in conjunction with dilute sulphuric acid for pretreatment. The systems were constructed from zirconium to minimize corrosion. The systems operated at about 140◦ C–160◦ C and residence times from 5 to 60 minutes with about 20%–25% solids and 1–2 wt% sulphuric acid in the system. Fairly low levels of sugars were produced, leading to ethanol yields on the order of 100 L/tonne of feedstock (Bulls et al., 1991).

9.3.3

Limitations of Laboratory-Scale Comparisons of Pretreatment Methods

Recently, the CAFI group and others have attempted to standardize methods to evaluate pretreatment efficiency, e.g., by using the same enzyme cocktail, dose, retention time, and hydrolysis conditions to assess the ease of hydrolysis of different pretreated fibers. In spite of these efforts, there is still some debate about the conclusions arising from these standardized assessments. For example, Galbe and Zacchi (2007) rendered the opinion that The assessment of pretreatment has to be performed in a more rigorous way. The standard enzymatic hydrolysis at low substrate concentration may well be used to assess the maximum digestibility. However, in this case both cellulases and hemicellulases are needed. The “real” assessment should

P1: SFK/UKS P2: SFK c09 BLBS075-Hood December 15, 2010

20:55

Trim: 244mm×172mm

9

Pretreatment Options

215

be performed by optimizing the conditions for all subsequent process steps under more realistic process conditions, taking into account the special features of the pretreated material, and then comparing the production cost for the various alternatives.

Similarly, Berlin et al. (2005) commented that Screening for improved activity typically uses “ideal” cellulosic substrates, and results are not necessarily applicable to more realistic substrates such as pretreated hardwoods and softwoods.

This comment is particularly relevant in light of extensive efforts to enhance enzyme activity with tailored enzyme cocktails, new cellulolytic platforms and through genetic modification, and enzyme discovery methods that aim to identify promising new microbial sources of hydrolytic enzymes. Positive outcomes based on idealized substrates and low solids loadings may appear promising, but may not be directly applicable to the real substrates and process conditions that will be encountered in a larger scale commercial system. Furthermore, as noted below, slurry viscosity and inhibitor tolerance are important factors at the commercial scale; bench scale assessments of pretreated materials using enzyme hydrolysis need to take these factors into account. Clearly, there is compelling evidence that the enzyme cocktail used to hydrolyze pretreated fibers must be tailored to both the type of feedstock and pretreatment process and must also consider large scale process impacts, including viscosity and enzyme/fermentation inhibitors. Consequently, the most valuable information regarding pretreatment efficacy will be obtained from large scale pilot and demonstration systems that include all of the relevant process unit operations for biofuel production, including hydrolysis, fermentation, and recovery of coproducts. These systems enable a comprehensive assessment of the full impact of the feedstock, pretreatment process, and pretreatment conditions.

9.4 Process Issues and Trade-Offs Selection of a feedstock and adoption of a pretreatment technology are the two most important factors impacting the overall economics of a lignocellulosic biofuel production process. The former impacts delivered facility-gate price, theoretical yield, capital, and operating costs, while the latter dictates how much of the incoming cellulose and hemicellulose is ultimately converted into biofuel and also influences inhibitor formation, ease of downstream processing, and overall capital and operating costs. These selections must be made while being mindful of downstream process technologies and overall process objectives, and inevitably, trade-offs will be required (Table 9.1). In this section, an expansion upon Table 9.1 is undertaken to review the impacts of feedstock/pretreatment upon inhibitor generation, enzyme hydrolysis, viscosity reduction and hydraulic load, and solvent/catalyst recovery.

9.4.1

Inhibitors

Generally speaking, inhibitor production during pretreatment is a function of pH, pretreatment temperature, and retention time during the pretreatment process. Acid pretreatment processes are markedly more prone to production of degradation products than alkaline pretreatments,

P1: SFK/UKS P2: SFK c09 BLBS075-Hood December 15, 2010

20:55

Trim: 244mm×172mm

Plant Biomass Conversion

216

Table 9.1. Pretreatment trade-offs.

Pretreatment

Enzyme Use

Chemical Use

Inhibitors

Materials/ Equipment Costs

Dilute acid Ammonia Autohydrolysis (steam explosion)

Low to moderate Moderate High

High Moderate to high Nil

High Low Low to moderate

High Moderate to high Moderate

Steam explosion with SO2

Moderate to high

High

Moderate to high

Moderate to high

Solvent extraction Hot water

Low to moderate Very high

High Nil

Moderate Low

Low to moderate Low to moderate

1.2

Relative concentration

1.0

0.8

Enzyme X

0.6

Enzyme G

0.4

0.2

0.0 30

60

90 120 180 Hydrolysis time, min

240

1440

Figure 9.1. Effect of acetic acid on cellulase activity.

because acids facilitate the rapid dehydration of xylose into furfural, and glucose into hydroxymethylfurfural (HMF). The rates of these dehydration reactions increase with temperature and with acid concentration. Formation of furfural and HMF can thus be reduced by avoiding the use of acids; autohydrolysis and alkaline pretreatments have much better inhibitor profiles. Nonetheless, even when acids are avoided, it is important to select the lowest possible pretreatment temperature and retention time. This will also inevitably lead to less solubilization of hemicellulose, lower levels of deacetylation, and the production of xylo-oligosaccharides rather than xylose, all of which will necessitate a higher enzyme loading and/or an enzyme cocktail supplemented with xylanases. Chemicals released during pretreatment and solvents used during pretreatment can also contribute to inhibition during both enzyme hydrolysis and fermentation. Figure 9.1 illustrates

P1: SFK/UKS P2: SFK c09 BLBS075-Hood December 15, 2010

20:55

Trim: 244mm×172mm

9

Pretreatment Options

217

Relative glucose concentration

1.2

1.0

0.8 10 vol% 0.6 5 vol% 0.4

0.2

0.0 4

8

24

28

32

48

Hydrolysis time, h

Figure 9.2. Effect of ethanol on cellulase activity.

the impact of 5 g/L acetic acid on the activity of two different cellulase cocktails. The data show the concentration of glucose in the presence of exogenous acetic acid normalized against the concentration of glucose produced without exogenous acetic acid. Thus, an enzyme not subject to inhibition would exhibit relative concentrations close to 1, whereas inhibition would lead to relative concentrations below 1. Stronger inhibition leads to a lower relative concentration. As shown in Figure 9.1, one commercial enzyme (identified as Enzyme X) is strongly inhibited, the other (Enzyme G) exhibits minimal inhibition. Levels of acetic acid present during hydrolysis are controlled by the acetyl content of the feedstock and by the pretreatment process; clearly, these factors must be considered when selecting components in an enzyme cocktail. The ethanol used in an organosolv process or generated during SSF can also inhibit cellulase activity, as illustrated in Figure 9.2. The data in Figure 9.2 are also presented as relative concentrations, that is, the concentration of glucose produced by the enzyme when exogenous ethanol is present relative to the glucose concentration generated by the same enzyme when ethanol is not present. These trials, using 5 and 10 vol% ethanol, show that glucose production in the presence of ethanol is typically about 20%–30% lower than levels observed in the absence of ethanol. In the case of 10% ethanol, it appears that the enzymes are irreversibly denatured, whereas in the case of 5% ethanol, some recovery of relative activity is apparent during the later stages of enzyme hydrolysis. Nonetheless, these results emphasize the importance of ensuring nearly complete recovery of solvents such as ethanol after organosolv pretreatment. They also emphasize the dilemma faced between separate hydrolysis and fermentation (SHF) and SSF: the former avoids inhibition by ethanol, but is susceptible to inhibition by glucose, xylose, and cellobiose, while the latter, due to consumption of carbohydrates during fermentation, minimizes the impact of carbohydrate-mediated inhibition, but faces significant issues with ethanol-mediated enzyme inhibition. Both of these inhibition issues are magnified at high solids loadings required for commercial operation.

P1: SFK/UKS P2: SFK c09 BLBS075-Hood December 15, 2010

218

9.4.2

20:55

Trim: 244mm×172mm

Plant Biomass Conversion

Hydrolysis Efficiency and Enzyme Loadings

The pretreatment process and pretreatment conditions can have a marked effect on enzyme loadings and hydrolysis efficiency (see Table 9.1). High severity pretreatments and pretreatments using acids, which tend to solubilize higher levels of hemicellulose and lignin, invariably lead to lower enzyme loadings when washed solid residues are used. However, if the inhibitors generated during high severity or acid pretreatment are not removed, higher enzyme loads are required to compensate. Pretreated solids from alkaline pretreatment processes such as AFEX generally require a xylanase as part of the enzyme cocktail to compensate for the residual hemicellulose and xylo-oligosaccharides that remain. However, owing to their lower severity, lower levels of inhibitors and beneficial effects on lignin, enzyme loadings for alkaline pretreatments tend to be lower than enzyme loadings required for uncatalyzed pretreatment processes such as LHW and autohydrolysis, which generally demand the highest levels of hydrolytic enzymes. Ohgren et al. (2007) examined the impact of xylanase on hydrolysis of corn stover pretreated by steam explosion with and without SO2 supplementation. Using a cellulase cocktail alone (cellulase + beta glucosidase), significantly higher glucose yields were achieved with stover pretreated with SO2 . However, if the enzyme cocktail included xylanase, total sugar yields were nearly identical following autohydrolysis and SO2 -catalyzed steam explosion. Furthermore, yields from autohydrolyzed stover hydrolyzed using xylanase supplementation were greater than yields from SO2 -catalyzed stover hydrolyzed with the cellulase cocktail alone. These outcomes illustrate the quintessential trade-off of pretreatment: a less severe (or uncatalyzed process) may give equivalent or improved results compared to an acid catalyzed pretreatment if the downstream enzymatic hydrolysis is modified. Ultimately, the cost of the acid catalyst and the impact of inhibitors must be weighed against the cost of additional enzymes. Although enzyme cocktails are currently fairly expensive, it is unlikely that this will always be the case, and on a large scale, the long-term benefits are likely going to favor a process that avoids supplemental chemical costs, waste treatment costs, and minimizes inhibitor production that would otherwise require treatment before enzymatic hydrolysis and fermentation.

9.4.3

Solvent/Catalyst Recovery

Pretreatment processes that rely on significant quantities of catalyst or solvent must be coupled with a solvent/catalyst recovery system. Examples include concentrated acid pretreatment, ammonia pretreatment technologies such as AFEX, and organosolv pretreatments. Acid recovery might also be required for some dilute acid pretreatments, particularly if the acid catalyst is a more expensive inorganic catalyst such as hydrochloric, nitric, or phosphoric acids. Acid recovery may also be required in order to avoid costly disposal of large quantities of gypsum. As noted above, recovery of certain solvents and acids may also be required to minimize inhibition during enzymatic hydrolysis and fermentation. In such a case, a balance is required between the cost of additional enzymes and larger process vessels versus the incremental cost for recovery of a successively higher proportion of the acid or solvent.

9.4.4

Viscosity Reduction and Hydraulic Load

As noted previously, the hydraulic load and solids loading are key parameters impacting economics, because they dictate vessel size, sugar and ethanol concentrations, and utility

P1: SFK/UKS P2: SFK c09 BLBS075-Hood December 15, 2010

20:55

Trim: 244mm×172mm

9

Pretreatment Options

219

4 3.5

Cocktail A Cocktail F

Mixer torque

3 2.5 2 1.5 1 0.5

23:10

22:20

21:30

20:40

19:50

19:00

18:10

17:20

16:30

15:30

14:40

13:50

13:00

12:10

10:30

11:20

09:40

08:50

08:00

07:10

06:20

05:30

04:40

03:50

03:00

02:10

00:00

0

Hydrolysis time, h

Figure 9.3. Impact of cellulase cocktail on viscosity reduction: normal pretreatment severity.

demand. Continuous, large scale processing of biomass is ultimately contingent on the ability to convey or pump a slurry between process unit operations. However, the viscosity of biomass slurries increases substantially at high solids loadings, creating issues with mixing, mass transfer, and heat transfer, and may limit access of hydrolytic enzymes to cellulose and hemicellulose. Efficient and rapid viscosity reduction of the biomass slurry is thus very important, as previously demonstrated in the starch–ethanol industry. The inability to effectively reduce viscosity ultimately forces a reduction in solids loading and increases the hydraulic load. The degree of biomass size reduction and the resulting mean particle size (or distribution) following pretreatment impact initial blending of the biomass with process liquids. Mechanical size reduction can thus serve two roles—reducing the initial particle size to a manageable level and rendering the fiber more amenable to pretreatment while producing an even smaller pretreated fiber. Post-pretreatment, the enzyme cocktail can also influence viscosity reduction during enzymatic hydrolysis. Figures 9.3 and 9.4 compare the viscosity reduction achieved with a 20% poplar slurry using two different enzyme cocktails, identified as “Cocktail A” and “Cocktail F.” In Figure 9.3, the poplar was pretreated by autohydrolysis under standard severity conditions (200◦ C–205◦ C, 5–7 minutes), whereas in Figure 9.4, a low severity pretreatment was adopted. The mixer torque, which is directly correlated to slurry viscosity, demonstrates that cocktail A was able to deliver a more rapid viscosity reduction compared to enzyme cocktail F. The results also demonstrate the strong effect of pretreatment. Under standard severity conditions (Figure 9.3), both enzyme cocktails were eventually able to reduce the viscosity to an acceptable level. However, under low severity conditions (Figure 9.4), cocktail F failed to deliver a meaningful viscosity reduction, while cocktail A provided an equally efficient viscosity

P1: SFK/UKS P2: SFK c09 BLBS075-Hood December 15, 2010

20:55

Trim: 244mm×172mm

Plant Biomass Conversion

220

5.0 4.5 4.0

Mixer torque

3.5

Cocktail A Cocktail F

3.0 2.5 2.0 1.5 1.0 0.5

22:40

21:50

21:00

20:10

19:20

18:30

17:40

16:50

16:00

15:10

14:20

13:20

12:30

11:40

10:50

10:00

09:10

08:20

07:30

06:40

05:50

05:00

04:10

03:20

02:30

01:40

00:00

00:50

0.0

Hydrolysis time, h

Figure 9.4. Impact of cellulase cocktail on viscosity reduction: low severity pretreatment.

reduction under both pretreatment conditions. It is important to note that the differences between cocktail A and cocktail F were only observed at moderate to high solids loadings, further emphasizing the value of trials with real substrates at high solids loadings.

9.5 Economics Owing to the trade-offs outlined in the previous section, the economics of pretreatment cannot be discussed in isolation from other process steps, or indeed, without considering feedstock and utility costs. Such an assessment typically requires a full-scale process simulation developed using ASPEN, HYSYS, SimSci, gProms, or similar process design/simulation packages. The complexity of developing an end-to-end process design means comprehensive economics assessments are relatively rare. Among the few that have been published are 1. Eggeman and Elander (2005), who developed ASPEN models for lignocellulosic ethanol production via various pretreatment pathways, as part of an economics assessment of pretreatment processes developed under the CAFI program

P1: SFK/UKS P2: SFK c09 BLBS075-Hood December 15, 2010

20:55

Trim: 244mm×172mm

Pretreatment Options

221

Average Annunal ROI

9

0

20

40

60

80

100

Feedstock Price ($/tonne) Ag residue steam explosion

Ag residue acid

Ag residue ammonia

Figure 9.5. Impact of feedstock price on economics for various large scale pretreatment processes.

2. Hamelinck et al. (2005), who used a combination of ASPEN and spreadsheet analysis to compare near-term, medium-term, and long-term prospects for lignocellulosic ethanol production Additional information can be gleaned from conference presentations, although embedded assumptions may not always be apparent. Many economics analyses based upon lab-scale data use observed or theoretical yields, with values assigned to feedstocks, products, coproducts, and measurable inputs such as enzymes. However, at a commercial scale, production rate and feedstock throughput are just as important, if not more important, than yield, because the real costs of achieving that yield are now taken into account. If the total cost to achieve that yield is prohibitive, some of the feedstock will remain unconverted and end up in the coproduct stream (usually leading to electricity production). Generation of relatively high-value coproducts with a minimum of additional processing or capital cost can compensate for a relatively low biofuel yield and provide the key towards an economically viable biofuel production facility. This is the objective from Lignol’s organosolv process, which has lower ethanol yields than other processes, but claims improved economics due to the value of relatively pure lignin fractions extracted during the process. An assessment of various factors impacting the economics of lignocellulosic biofuel production was presented by SunOpta Bioprocess Inc. (2007; http://www.sunopta.com/ uploadedFiles/bioprocess/news and events/071017%20Platts%20Presentation.pdf). The information included a comparison of overall process economics for ethanol produced from agricultural residues based upon three different pretreatment methods: autohydrolysis (steam explosion), AFEX, and dilute acid. The assessments are particularly valuable given that SunOpta Bioprocess Inc. (formerly Stake) is the only organization that has conducted large scale pretreatment work with all three of these processes. As shown in Figure 9.5, dilute acid pretreatment generated the lowest overall return on investment (ROI) at any particular

P1: SFK/UKS P2: SFK c09 BLBS075-Hood December 15, 2010

222

20:55

Trim: 244mm×172mm

Plant Biomass Conversion

3.0

2.5

kWh/USG

2.0

1.5

1.0

0.5

0.0 Dilute acid

Hot water

AFEX

ARP

Lima

Steam explosion

Figure 9.6. Electricity demand for cellulosic ethanol production using various pretreatment processes.

feedstock price, whereas steam explosion generated the highest ROI. Ammonia pretreatment was markedly better than dilute acid, but slightly inferior to steam explosion. Figure 9.5 also compares the impact of feedstock on economics; clearly, one is better off with ammonia pretreatment with a feedstock at $45/tonne than with steam explosion and a feedstock priced at $60/tonne. This implies that a lower feedstock price can offset the additional expense of ammonia recovery under certain situations. Pretreatment technologies that can effectively process lower cost feedstocks may thus have an economic advantage, in spite of higher capital or utility costs, as long as the yield and production rate are preserved. It is also possible to compare utility requirements for processes based upon different pretreatment methodologies, derived from data presented in Eggeman and Elander and in conference presentations by Sunopta Bioprocess Inc. Figures 9.6 and 9.7 illustrate electricity and thermal energy consumption for cellulosic ethanol using different pretreatment technologies. The data for dilute acid, hot water, AFEX, ARP, and Lime were provided in (or calculated from) Eggeman and Elander (2005), while the data for steam explosion were provided by SunOpta Bioprocess Inc. The significant difference in utility demand among processes is due in part to different process yields, but is also influenced to a large degree by the hydraulic load and potential requirements for solvent/catalyst recovery. Among the systems reported, the LHW system has the lowest solids loading and highest hydraulic load. This leads to a substantially higher demand for thermal energy/steam to heat process vessels and remove excess water during distillation, and additional electricity demand to pump a large volume of slurry. The ARP and Lime processes also have a higher hydraulic load, leading to high energy demand for these processes as well. AFEX and steam explosion have much higher solids loadings; the higher steam and electricity demand for AFEX is due to process operations to recover and recycle ammonia.

P1: SFK/UKS P2: SFK c09 BLBS075-Hood December 15, 2010

20:55

Trim: 244mm×172mm

9

Pretreatment Options

223

160,000 140,000 120,000

BTU/USG

100,000 80,000 60,000 40,000 20,000 0 Dilute acid Hot water

AFEX

ARP

Lime

Steam Explosion

Figure 9.7. Thermal energy demand for cellulosic ethanol production using various pretreatment processes.

It is important to note that in all of these cases (except possibly lime), there is enough excess lignin to satisfy these energy demands. However, the amount of excess lignin available to process electricity destined for the grid is inversely correlated with process energy demand. A process using ARP or LHW will have significantly lower coproduct revenues from lignin than a process such as steam explosion that has a substantially lower electricity and thermal energy demand. Hamelinck et al. (2005) evaluated several biofuel production technologies using a combination of ASPEN Plus and spreadsheet models. They concluded that acid pretreatment would have an advantage in the near term, because the lower enzyme requirements coupled with (currently) high enzyme prices lead to a material reduction in enzyme costs per unit of ethanol produced. However, as technology evolves and enzyme prices are reduced, Hamelinck et al. concluded that steam explosion and LHW processes would be preferred, due to higher yields, lower levels of inhibitors, lower utility costs, and lower overall equipment costs. The outcome for the LHW system was contingent upon the ability to increase the solids loading (decreased hydraulic load) by operation in conjunction with a consolidated bioprocess (CBP), which, if achieved, would improve process energy efficiency compared to the current state for LHW systems. Overall, these economics comparisons need to be interpreted with caution, since full process information is not available from the references and processes are still being refined. For example, Galbe and Zacchi (2007) issued the following comment regarding the Eggeman study: The process configuration was identical for all processes except the pretreatment step . . . However, it should be emphasized that a fairer comparison would require optimization of each process alternative taking into consideration the specific features of the pretreatment method used.

P1: SFK/UKS P2: SFK c09 BLBS075-Hood December 15, 2010

224

20:55

Trim: 244mm×172mm

Plant Biomass Conversion

9.6 Conclusions Effective fiber pretreatment is essential if cellulosic ethanol is to be produced economically. Pretreatment impacts all downstream processes and also influences potential choices regarding coproducts and, thus, coproduct value. Although many pretreatment methods have been proposed and evaluated at the laboratory scale, only a select few have been practiced at a scale needed for large scale biofuel production, including steam explosion (with and without acids), LHW, and ammonia. Among pretreatment methods that have been practiced at both the laboratory and demonstration/commercial scale, well-documented differences in outcomes from bench scale batch studies and large scale continuous processes have been observed. Consequently, it is difficult to apply conclusions from bench scale work to pilot and commercial scale operations. However, bench scale work continues to provide valuable mechanistic information about the impacts of pretreatment on fiber structure, crystallinity, and bonding. These insights may ultimately lead to improvements in pretreatment that can be applied at a commercial scale. The economics of cellulosic ethanol production are strongly influenced by the pretreatment process. In particular, hydrolysis rates, fermentation rates, and yields and hydraulic load are all dictated by the feedstock and pretreatment technology. Processes such as dilute acid, organosolv, LHW, and ARP that rely on high hydraulic loads (i.e., low solids loadings) have a significant economic disadvantage, due to higher capital costs, utility costs, and low sugar/ethanol titers. Processes that rely on acid supplementation are also likely to be less favored as enzyme costs decline over time, as the extra capital costs, lower process yields, and higher inhibitor levels eventually lead producers to less aggressive pretreatment technologies. Over the long term, pretreatments such as autohydrolysis and AFEX are likely to prevail, because of low inhibitor production, higher hydrolysis and fermentation rates, and low hydraulic loads that will lead to higher ethanol concentrations at the conclusion of fermentation. Although ammonia recovery costs must be borne when implementing an AFEX process, the technology can lead to lower enzyme loadings with certain feedstocks, a trade-off that must be continually reassessed as enzyme prices decline.

References Babcock, L. 1932. Method of producing soluble sugars and alcohols from wood. US Patent 1825464. Berlin, A., Gilkes, N., Kurabi, A., Bura, R., Tu, M., Kilburn, D. & Saddler, J. 2005. Weak lignin-binding enzymes: A novel approach to improve activity of cellulases for hydrolysis of lignocellulosics. Applied Biochemistry and Biotechnology, 121, 163–170. Brown, D. 1980. Apparatus for discharge of pressure cooked particulate or fibrous material. US Patent 4211163. Brown, D. & Bender, R. 1980. Feeding natural cellulose fiber into steam pressure vessel for manufacture of fodder from wast wood chips, straw, bagasse, etc. Canada patent 1,070,537. Bulls, M., Watson, J., Lambert, R. & Barrier, J. W. 1991. Conversion of cellulosic feedstocks to ethanol and other chemicals. In: Energy from Biomass and Waste XIV. Chicago, IL: Institute of Gas Technology; pp.1167–1179. Cadoche, L. & L´opez, G. D. 1989. Assessment of size reduction as a preliminary step in the production of ethanol from lignocellulosic wastes. Biological Wastes, 30, 153–157.

P1: SFK/UKS P2: SFK c09 BLBS075-Hood December 15, 2010

20:55

Trim: 244mm×172mm

9

Pretreatment Options

225

Chandra, R., Bura, R., Mabee, W., Berlin, A., Pan, X. & Saddler, J. 2007. Substrate pretreatment: The key to effective enzymatic hydrolysis of lignocellulosics? Advanced Biochemical Engineering and BioTechnology, 108, 67–93. Chum, H., Johnson, D., Black, S. & Overend, R. 1990. Pretreatment—Catalyst effects and the combined severity parameter. Applied Biochemistry and Biotechnology, 24–25, 1–14. Delong, E. 1981 and 1983. Method of rendering lignin separable from cellulose and hemicellulose in lignocellulosic material and the product so produced. Canada Patent 1,096,374, Canadian patent 1,141,376. Eggeman, T. & Elander, R. T. 2005. Process and economic analysis of pretreatment technologies. Bioresource Technology, 96, 2019–2025. Foody, P. 1984. Method for obtaining superior yeilds of accessible cellulose and hemicellulose from lignocellulosic materials. Canada Patent 1,163,058. Galbe, M. & Zacchi, G. 2007. Pretreatment of lignocellulosic materials for efficient bioethanol production. Advances in Biochemical Engineering Biotechnology, 108, 41–65. Glasser, W. G. & Wright, R. S. 1998. Steam-assisted biomass fractionation. II. Fractionation behavior of various biomass resources. Biomass and Bioenergy, 14, 219–235. Hallberg, C., O’Conner, D., Rushton, M., et al. 2009. Continuous counter-current organosolv processing of lignocellulosic feedstocks. United States Patent Application 0117226. Hamelinck, C. N., Hooijdonk, G. V. & Faaij, A. P. C. 2005. Ethanol from lignocellulosic biomass: Techno-economic performance in short-, middle- and long-term. Biomass and Bioenergy, 28, 384–410. Heitz, M., Capek-Menard, E., Koeberle, P., Gagne, J., Chornet, E., Overend, R., Taylor, J. & Yu, E. 1991. Fractionation of Populus tremuloides at the pilot plant scale: Optimization of steam explosion pretreatment conditions using the STAKE II technology. Bioresource Technology, 35, 23–32. Heitz, M., Carrasco, F., Rubio, M., Brown, A., Chornet, E. & Overend, R. P. 1987. Physicochemical characterization of lignocellulosic substrates pretreated via autohydrolysis: An application to tropical woods. Biomass, 13, 255–273. Hosseini, S. A. & Shah, N. 2009. Multiscale modelling of biomass pretreatment for biofuels production. Chemical Engineering Research and Design, 87, 1251–1260. Hu, G. Y., Heitmann, J. & Rojas, O. 2008. Feedstock pretreatment strategies for producing ethanol from wood, bark, and forest residues. BioResouces, 3, 270–294. Kumar, R., Mago, G., Balan, V. & Wyman, C. E. 2009. Physical and chemical characterizations of corn stover and poplar solids resulting from leading pretreatment technologies. Bioresource Technology, 100, 3948–3962. Mason, W. 1929. Apparatus for and process of explosion fibration of lignocellulosic material. US Patent 1655618. Montane, D., Overend, R. P. & Chornet, E. 1998. Kinetic models for non-homogeneous complex systems with a time-dependent rate constant. Canadian Journal of Chemical Engineering, 76, 58–68. Mosier, N., Wyman, C., Dale, B., Elander, R., Lee, Y. Y., Holtzapple, M. & Ladisch, M. 2005. Features of promising technologies for pretreatment of lignocellulosic biomass. Bioresource Technology, 96, 673–686. Muzzy, J. D., Roberts, R. S., Fieber, C. A. & Al., E. 1983. Pretreatment of hardwood by continuous steam hydrolysis. In: Stoltes, J. (ed.) Wood and Agricultural Residues. New York, NY: Academic Press.

P1: SFK/UKS P2: SFK c09 BLBS075-Hood December 15, 2010

226

20:55

Trim: 244mm×172mm

Plant Biomass Conversion

Nativel, F., Pourquie, J., Ballerini, D., Vandecasteele, J. P. & Renault, P. 1992. The biotechnology facilities at Soustons for biomass conversion. International Journal of Solar Energy, 11, 219–229. Ohgren, K., Bura, R., Saddler, J. & Zacchi, G. 2007. Effect of hemicellulose and lignin removal on enzymatic hydrolysis of steam pretreated corn stover. Bioresource Technology, 98, 2503–2510. Overend, R. P., Chornet, E. & Gascoigne, J. A. 1987. Fractionation of lignocellulosics by steam-aqueous pretreatments [and Discussion]. Philosophical Transactions of the Royal Society of London. Series A, Mathematical and Physical Sciences, 321, 523–536. Pye, K. & Lora, J. H. 1991. The ALCELL process. A proven alternative to kraft pulping. Tappi Journal, 74, 113–117. Ropars, M., Marchal, R., Pourqui´e, J. & Vandecasteele, J. P. 1992. Large-scale enzymatic hydrolysis of agricultural lignocellulosic biomass. Part 1: Pretreatment procedures. Bioresource Technology, 42, 197–204. Sun, Y. & Cheng, J. 2002. Hydrolysis of lignocellulosic materials for ethanol production: A review. Bioresource Technology, 83, 1–11. ´ J. & Cardona, C. A. 2008. Trends in biotechnological production of fuel ethanol S´anchez, O. from different feedstocks. Bioresource Technology, 99, 5270–5295. Wayman, M. & Parekh, S. 1988. SO2 prehydrolysis for high yield ethanol production from biomass. Applied Biochemistry and Biotechnology, 17, 33–43. Wayman, M., Parekh, S., Chornet, E. & Overend, R. P. 1986. SO2 -catalysed prehydrolysis of coniferous wood for ethanol production. Biotechnology Letters, 8, 749–752.

P1: SFK/UKS P2: SFK c10 BLBS075-Hood December 15, 2010

20:55

Trim: 244mm×172mm

Chapter 10

Enzyme Production Systems for Biomass Conversion John A. Howard, Zivko Nikolov, and Elizabeth E. Hood

10.1 Introduction Plant biomass is a complex matrix of polymers comprising the polysaccharides cellulose and hemicellulose, as well as the phenolic polymer lignin, as the major structural components. Cellulose, the most abundant biopolymer on earth, is a simple, linear, β-1,4-linked polymer of glucose. However, its semi-crystalline structure is notoriously resistant to hydrolysis by both enzymatic and chemical means. The cellulose in secondary cell walls is embedded in a matrix of hemicellulose and lignin with lesser amounts of pectin and protein. These secondary walls are the source of biomass targeted for conversion because of their abundant dry weight at maturity when compared to herbaceous plant materials. A strategy designed to use lignocellulose for fuel or bio-products must include the ability to efficiently convert the polysaccharide and lignin components of plant cell walls to simple sugars and phenolic monomers, respectively. Deconstruction of lignocellulose can be accomplished by heat and chemical means (Taherzadeh and Karimi, 2007) but the preferred environmentally friendly method is to use enzymes.

10.2 The Challenge: Volume and Cost of Enzymes Required With the current state of technology for biomass conversion, the overwhelming enzyme requirement is for cellulases: endo-cellulase, exo-cellulase, and glucosidase (Merino and Cherry, 2007). The specific activity of most cellulases is quite low (Jorgensen et al., 2007; Sticklen, 2008b) and much effort has focused on increasing their activity levels. However, even with improved enzymes and improved methods of production, the amount of cellulase required to Plant Biomass Conversion, First Edition. Edited by Elizabeth E. Hood, Peter Nelson and Randall Powell.
C 2011 John Wiley & Sons Inc.

227

P1: SFK/UKS P2: SFK c10 BLBS075-Hood December 15, 2010

228

20:55

Trim: 244mm×172mm

Plant Biomass Conversion

Table 10.1. Estimates of enzyme amounts necessary to digest fixed amounts of biomass.

Tons of Biomass

Billions Gallons Ethanol Possible

Enzymes Per Gallon (g)

Tons of Enzymes (MMa )

Cost of Enzyme per Gallonb ($)

425 million

36

100 30

3.6 1.1

0.17 0.12

Assumptions: 85 gallons ethanol per ton; tons of enzymes required to produce that amount per year. Enzymes required per gallon will decrease with improvements. Thirty-six billion gallons will be produced by 2022. a MM, million. b NREL estimate; David Hsu.

deconstruct the volumes of biomass necessary for 30% replacement of gasoline are in the millions of tons. For the purpose of this chapter, we are using a base case of enzyme amounts and cost as shown in Table 10.1. At a loading level of 100 g enzyme per gallon of ethanol produced, and a production goal of 36 billion gallons from the Bush mandate, 3.6 million metric tons of cellulase will be required. This is the amount per year! It is assumed for this discussion that commercial scale microbial fermenters would hold 500,000 L and yield 50 g cellulase/L of culture broth per week (i.e., per batch), conditions that would allow production of 1,250 metric tons of enzyme per year per fermenter, assuming 100% recovery. Under these conditions, a minimum of 2,900 fermenters would be required to produce enough enzyme to saccharify 425 million tons of biomass. Moreover, if each fully equipped and installed fermenter costs about $10 million, the capital investment for fermenters alone would amount to $30 billion. This is an unprecedented challenge in terms of the amount of enzymes and the extremely low cost that is required for competitively priced ethanol. In addition, the amount of upfront capital required for fermenter capacity is also problematic. If the enzyme load was 100 g/gallon at a cost of $0.17/gallon, a total capital investment of over $57 billion dollars would be required to return gross annual revenues of ∼$6 billion. This situation has led many groups to investigate ways to reduce this cost burden. There is every reason to believe that the cost will go down. While we have used this previously published model as the base case, continual reports appear of improvements from many different groups on many different enzyme cost reduction technologies. This chapter will discuss the different approaches that are being considered to lower the overall cost and the impact that these approaches may have on reducing the cost burden.

10.3 Theoretical Ways to Address the Challenge of Quantity of Enzyme and Cost Requirements There are four steps to the conversion of biomass to ethanol: (1) pretreatment to remove lignin and hemicellulose, (2) hydrolysis to allow saccharification of cellulose, (3) fermentation for ethanol production, and (4) purification of ethanol and other products. The hydrolysis step using cellulase enzymes accounts for 40% of current cost input (Miyamoto, 1997). Pretreatment is another energy and cost hurdle, which can be addressed by an enzymatic or thermochemical process. The pretreatment step has an impact on the amount and type of enzymes that are

P1: SFK/UKS P2: SFK c10 BLBS075-Hood December 15, 2010

20:55

Trim: 244mm×172mm

10 Enzyme Production Systems for Biomass Conversion

229

required for saccharification (see Chapter 9, this volume). The focus of this chapter is on what can be done to each of the four steps that will have an impact on the enzyme cost.

10.3.1

Increase Susceptibility for Biomass Deconstruction

The plant cell wall is a complex network of cellulose fibrils coated and linked by hydrogen bonds to hemicellulose. Hemicelluloses are a heterogeneous group of linear and branched polysaccharides primarily composed of pentoses, predominantly xylose (Shallom and Shoham, 2003). Hemicelluloses vary in composition between different plant sources, but often contain side chains which can sterically block the accessibility of cellulose to cellulase and thereby limit digestion. The hemicellulose polymers are further covalently linked to lignin molecules, forming an insoluble and robust structure. The amount of lignin can vary with different types of tissues as well (Campbell and Sederoff, 1996). Saccharification of cellulose microfibrils prior to fermentation requires that the shielding hemicelluloses and lignin be removed to increase cellulase access. Therefore, any treatment that reduces steric hindrance and allows greater access to cellulose can theoretically reduce the amount of enzymes required for saccharification. Pretreatment of Biomass A plethora of pretreatment paradigms have been reported (Mosier et al., 2004; Sun and Cheng, 2002) and compared (Wyman et al., 2005) (see also Chapter 9 in this volume). Physical and chemical methods including milling, high-temperature treatment with steam, acid, or alkali, as well as steam or alkali explosion, and biological pretreatment have been used successfully. Protocols include a thermochemical pretreatment that uses weak acid hydrolysis at an elevated temperature to remove hemicelluloses and expose cellulose fibrils to enzymatic deconstruction. This treatment allows for limited sugar degradation and the potential for recovery of components such as lignin, polylactic acid, and hemicellulose sugars for other applications (Ragauskas et al., 2006; Zhang et al., 2007). Biologically based pretreatment protocols utilize white rot fungi of several genera (Sun and Cheng, 2002). Lignin-degrading enzymes are secreted extracellularly by many white rot fungi and fall into three groups: lignin peroxidases (EC 1.11.1.7, LiP), manganese peroxidases (EC 1.11.1.7; MnP), and laccase (EC 1.10.3.2). LiP is synthesized by a family of about ten closely related genes, lipA through lipJ, whose expression is affected by culture conditions. MnP expression is regulated by the presence of Mn in culture. Both MnP and LiP have been expressed in heterologous systems, and their activity tested following site-directed mutagenesis. Laccase genes are less well characterized, and their presence is not thought to be essential for effective lignin degradation. All three enzymes act through oxidative, rather than hydrolytic, mechanisms and have acidic optima that make them useful for biomass pretreatment under acidic conditions (Cullen, 2002). Laccases and ligninases are used in the pulp and paper industry for lignin removal. Delignification is a central process to the utilization of lignocellulose, and therefore the genome of the model lignin-degrading white rot fungus Phanerochaete chrysosporium has been sequenced and methods for its genetic manipulation by homologous recombination and transformation by auxotrophic complementation have been developed (Cullen, 2002). A strain of Sporotrichum pulverulentum lacking cellulase has been produced to prevent hydrolysis during delignification (Ander and Eriksson, 1977). Other delignifying enzymes are produced by white rot fungi that might speed up the process, but overall

P1: SFK/UKS P2: SFK c10 BLBS075-Hood December 15, 2010

230

20:55

Trim: 244mm×172mm

Plant Biomass Conversion

the process remains slow. It is, however, energetically and environmentally less demanding than thermochemical pretreatment. The enzymes that degrade hemicellulose are as heterogeneous as the substrate, but are usually modular enzymes with a carbohydrate-binding module (CBM) and a catalytic domain. Many also have dockerin domains to bind the enzyme to the microbial cell surface or to the conserved cohesin domains on cellulosomes (Bayer et al., 2007; Gilbert, 2007). The biotechnological applications of hemicellulases include biobleaching and biopulping in the pulp and paper industry, improvement of animal feed quality, as well as the processing of flour in the bakery industry (Shallom and Shoham, 2003). Currently, hemicellulases are not used for biomass degradation, but studies are being conducted on extension of their use to economic pretreatment (Galbe and Zacchi, 2002; Mielenz, 2001; Sticklen, 2008b). Further, hemicelluloses could be converted into constituent pentoses to be fermented into ethanol, which would increase overall ethanol yield. However, this requires different organisms that ferment pentoses (See Chapter 8 in this Volume). The removal of lignin and hemicellulose from corn stover has been compared in batch and flow-through systems (Yang and Wyman, 2004). The effectiveness of flow-through systems at removing lignin and hemicelluloses was generally higher than batch systems. Lignin removal generally increased cellulose digestibility, but its removal was not essential in batch systems. In summary, the more effective the pretreatment is in exposing the cellulose and removing lignin and hemicelluloses, the lower the enzyme requirement for saccharification. The most effective pretreatment today involves thermochemical processing, which requires large capital investments and produces environmental waste. At the other extreme is enzymatic pretreatment, which is environmentally friendly, but not as efficient with current technology. The dilemma is that enzymatic pretreatment represents additional cost contraints on an already cost-prohibitive amount of enzymes used in saccharification. Therefore, while the long-term hope is to have a completely enzymatic process, in the short term, it appears that thermal and chemical pretreatments will need to be optimized to be compatible with the enzymatic saccharification process (see Chapter 9).

Genetic Alteration of Feedstock Plants show varying susceptibility to deconstruction, related to the levels of obstructive lignin and hemicelluloses. A survey of the composition of cell walls across plants (Sun and Cheng, 2002) allows comparison of the best pretreatment and hydrolysis procedures to follow. Susceptibility of cell walls to degradation may vary because lignin content may vary even within the same species, e.g., in trees (Campbell and Sederoff, 1996). Experiments have indicated that rice hulls require a ten times higher concentration of a mixture of enzymes to effect breakdown as hardwood requires (Jimenez-Flores et al., 2010). This difference may reflect the presence of materials such as silica that interfere with cellulose pretreatment and hydrolysis (Gressel, 2007; Sainz, 2009; Zhu et al., 2009). Assessing digestibility based on composition will help with choosing the best substrates for hydrolysis as well as tailoring pretreatment and choosing the cocktail of hydrolytic enzymes for optimal activity. Selecting the most amenable source for enzymatic deconstruction is not always practical, especially when using agricultural by-products such as corn stover and rice hulls or wood by-products. Therefore, the potential to increase the digestibility of these biomass sources directly is desirable. Plant lines that have competitive agronomic characteristics but also possess increased digestibility of their by-products can be obtained. The two targeted areas

P1: SFK/UKS P2: SFK c10 BLBS075-Hood December 15, 2010

20:55

Trim: 244mm×172mm

10 Enzyme Production Systems for Biomass Conversion

231

where this approach would have a major impact are in reducing the costs of pretreatment and saccharification. Plants can either be selected or engineered to be more susceptible to digestion (Campbell and Sederoff, 1996; Chen and Dixon, 2007). Attempts have been made to increase the amount of lignin-degrading enzymes in vegetative tissue in order to reduce the pretreatment burden. The challenge is to obtain plants with greater susceptibility to enzyme digestion but which still perform well in the field. MnP expression in alfalfa led to stunted development and reduced yield of dry matter, although the plants did produce seed (Austin et al., 1995). In maize, the expression of MnP correlated to tissue-specific effects. MnP expressed preferentially in seed showed no signs of deleterious effects on plant health. In contrast, cytoplasmic targeting with a constitutive promoter caused lesions in older leaves, although germination and flowering were not affected (Clough et al., 2006). A decrease in lignin content was achieved in aspen by down-regulation of the coumarate: coenzyme A ligase (Pt4CL1), which led to a 45% decrease in lignin and a compensatory 15% increase in cellulose content of the modified plant. This altered ratio favors bioethanol production (Li et al., 2003). Modifying lignin biosynthetic enzymes to lower lignin in cell walls is an obvious way to reduce biomass recalcitrance, and sugar yields from modified alfalfa lines with lower lignin-forming enzymes were nearly double over wild type (Chen and Dixon, 2007). Technologies have been developed to modulate levels of lignin biosynthetic enzymes in situ using antisense constructs in alfalfa (Reddy et al., 2005) and RNAi in maize (Sticklen, 2008a). Biomass composition and enzymatic digestibility of plants can be altered by reducing the levels of synthetic enzymes and increasing the levels of degradative enzymes. Reducing the expression of poplar glycosyltransferase using RNA interference led to a reduction in the glucuronoxylan content of poplar and consequently increased its digestibility by cellulase (Lee et al., 2009). Arabidopsis plants expressing a repressor derived from a secondary cell wall thickening-promoting factor (NST1) were twice as susceptible to enzymatic hydrolysis as control plants (Iwase et al., 2009). By contrast, increasing expression in Arabidopsis of poplar cellulase caused modification of cellulose–hemicellulose crosslinks, leading to increased plant size and improved cellulose hydrolysis (Park et al., 2003). Hartati et al. (2008) expressed poplar cellulase in the leguminous plant, sengon. Transgenic plants showed increased growth and leaf size, attributed to paracrystalline cellulose disruption and less xyloglucan content in the wall. The expression of several hydrolytic enzymes in poplar made the wood more susceptible to enzymatic hydrolysis (Kaida et al., 2009). However, one must bear in mind that thermochemical pretreatment of transgenic plants containing enzymes may lead to inactivation of the endogenous enzyme, with the potential need to add exogenous cellulases following pretreatment (Oraby et al., 2007).

10.3.2

Decrease Exogenous Enzyme Load

The deconstruction of microcrystalline cellulose is far from a simple chemical reaction. The process requires the activity of several different enzymes, thereby complicating a simple one-solution method for improvement. However, this complexity also provides the potential for improvements in many different ways, each of which are discussed separately below. Ultimately, many of these approaches can be combined to provide additive or even synergistic effects to increase the overall activity of cellulase action. A minimum of three categories of enzymes in the glycosyl hydrolase superfamily are required for deconstruction of cellulose after the biomass has undergone pretreatment. They

P1: SFK/UKS P2: SFK c10 BLBS075-Hood December 15, 2010

232

20:55

Trim: 244mm×172mm

Plant Biomass Conversion

are endoglucanase (EC 3.2.1.4), exoglucanase (also called cellobiohydrolase; EC 3.2.1.91), and β-glucosidase (EC 3.2.1.21). Endoglucanases hydrolyze internal β-1,4 glycosidic bonds within a cellulose microfibril. The fibrils are then exposed to the action of exoglucanases that processively trim cellobiose units from exposed ends, again by hydrolysis of the β-1,4 glycosidic bond. Finally, β-glucosidase converts cellobiose and cellotriose units to glucose. βGlucosidase shows severe end-product inhibition and therefore optimal activities are achieved upon glucose removal either by physical methods or by fermentation to ethanol (Andric et al., 2010). These enzymes work in synergy, that is, the combined effect on cellulose deconstruction is greater than the individual effects added together. Therefore, any consideration of economical conversion will benefit from optimizing this synergistic activity. This process has been tested widely, and many studies testify to the value of using purified, reconstituted cellulase mixtures over crude extracts from a single organism (Gusakov et al., 2007; Rosgaard et al., 2007) as well as tailoring cellulase mixtures to biomass source (reviewed further below). Cellulase enzymes are modular, with domains linked by a flexible, glycosylated, hinge region, whose length is important to optimal cellulase activity. The catalytic domain varies in shape between endoglucanases and exoglucanases. Endoglucanases have an open catalytic site, and exoglucanases have a tubular catalytic groove to fit the ends of the cellulose fibril (Henrissat, 1994). All the enzymes share a carbohydrate binding domain (CBD) on the N- or Cterminus of the catalytic domain that allows the enzymes to adsorb to insoluble cellulose. Sitedirected mutagenesis has shown that surface aromatic amino acids flanked by polar residues within the binding cleft are important for tight binding (Kormos et al., 2000). Tryptophan instead of tyrosine increases affinity of cellobiohydrolase from the four families of CBDs so far identified. These differing affinities for substrate could be manipulated using site-directed mutagenesis to optimize binding properties for different substrates, i.e., different crystalline forms of cellulose as it goes through deconstruction. Some cellulases have two CBDs and this organization synergistically increases affinity of the enzyme for cellulose, and CDBs can be mutagenized to modulate broader or altered enzyme specificity (Linder and Teeri, 1997). Thus, cellulase can be modified to optimize specificity by promoting synergism between enzymes, by varying the length of the linker region, modifying amino acids in the CBD to increase substrate specificity and increasing binding strength by increasing number or affinity of CBDs. Cellulolytic enzymes are synthesized and secreted outside the cell by fungi such as Trichoderma reesei or into membrane-bound complexes called cellulosomes in bacteria such as Clostridium thermocellum (Bayer et al., 1998, 2004). Enzymes from different sources vary in their pH and temperature optima, inducibility, stability, and other parameters (Cantarel et al., 2009) (see also www.cazy.org). These factors need to be accounted for when using multiple enzymes to degrade cellulose. In some organisms, cellulolytic activity may depend on the presence of a single gene (Tolonen et al., 2009). Disruption of the cphy3367 gene for its only family 9 glycosyl hydrolase in Clostridium phytofermentans abolished cellulolytic activity despite the presence and even elevated levels of other CAZy glycosyl hydrolases, indicating that there was an obligate requirement for Cphy3367 for cellulose metabolism to even occur. Thus, it is not enough to have the best single enzyme of the class but rather the best complement of enzymes that can work together, a condition that must be established empirically.

Tailor Enzyme Cocktails to Biomass Source The variety of possible enzymes for use in deconstruction protocols is not uniform across sources of biomass, suggesting that enzyme cocktails should be varied to fit the substrate

P1: SFK/UKS P2: SFK c10 BLBS075-Hood December 15, 2010

20:55

Trim: 244mm×172mm

10 Enzyme Production Systems for Biomass Conversion

233

(Zhang and Lynd, 2004). Attention also should be paid to pretreatment and optimal cellulose adsorption by enzymes to maximize deconstruction based on the composition of the specific substrate. In other words, one can combine the optimal mixture of enzymes with the specific substrate to decrease enzyme load. For example, the composition of cellulose, hemicellulose, and lignin fractions from different sources varies significantly (Sun and Chen, 2002) and would be deconstructed with different kinetics even if the same pretreatment and enzyme mixtures were used. In addition to using different enzyme cocktails for different biomass sources, each source should be separately evaluated as to the optimal amounts of each enzyme to be used in its deconstruction. This concept has been assessed using SO2-impregnated spruce and aspen wood (Breuil et al., 1990). It was shown that although aspen wood was completely hydrolyzed by any of five commercial cellulase preparations, spruce wood contained inhibitory agents that prevented complete hydrolysis. Enzymatic pretreatment of spruce might increase the hydrolysis profile by removing some inhibitory agents, such as lignin that might competitively adsorb cellulases (Zhang and Lynd, 2004). The same lab showed previously that removal of end products of hydrolysis was important to achieving optimal substrate hydrolysis (Chan et al., 1989). On a related issue, the rate of hydrolysis decreases with time, even when taking into account the effect of end-product inhibition and enzyme deactivation (Eriksson et al., 2002). Both the surface area and cellulose adsorptive capacity have been shown to decrease with hydrolysis over time. Hydrolysis rates may be maintained by periodic addition of new substrate (Zhang and Lynd, 2004) and by varying CBD sites on enzymes (Carrard et al., 2000). Enzymes from Chrysosporium lucknowense were used to demonstrate improved cellulose hydrolysis from reconstituted purified enzymes over crude extracts (Gusakov et al., 2007). These authors also demonstrated that modifying the mixture to contain enzymes with poor CBM overcame the pseudo-inactivation of cellulases during hydrolysis, presumably due to greater ability to migrate over the surface, thus clearing obstacles on the cellulose. Zhou et al. (2009) compared a crude cellulase extract from a Trichoderma viride T 100-14 mutant strain with purified enzymes from the same strain in various combinations to identify the optimal amounts of each enzyme for glucose production from steam exploded corn stover. The optimal enzyme combination produced 2.1 times more glucose than the crude extract. This type of manipulation to obtain maximum conversion will be useful in an industrial setting to maximize efficiency and lower cost.

Obtain New Sources of Enzyme-Encoding Genes In spite of the progress in optimization of hydrolysis, cellulases constitute a relatively high cost for the bioconversion process and account for up to 40% of the total biomass conversion cost (Miyamoto, 1997; Zhang and Lynd, 2004). Since the 1970s, screening has led to the discovery of many sources of cellulase from fungi, termites, aerobic and anaerobic bacteria, and even symbiotic microorganisms from pinfish transitioning from a carnivorous to herbivorous diet (Luczkovich and Stellwag, 1993). Novel screening methods include isolation of novel cellulase enzymes from soil metagenome libraries (Rondon et al., 2000). This process makes it feasible to isolate enzymes from organisms that are refractory to cultivation. Degenerate PCR primers were used for screening DNA of fungal communities in corn stover for the presence of glycosyl hydrolase family genes (Jacobsen et al., 2005). This rationale is powerful, given that only organisms that are able to efficiently use lignocellulosics as a carbon

P1: SFK/UKS P2: SFK c10 BLBS075-Hood December 15, 2010

234

20:55

Trim: 244mm×172mm

Plant Biomass Conversion

source would grow on it. This strategy was also applied to leaf litter and soil from oakdominated and sugar maple-dominated hardwood forests (Edwards et al., 2008) to screen for organisms containing sequences similar to the exoglucanase gene, cbhI. Novel sequences were recovered, indicating that this strategy holds promise for identifying new cellulolytic genes from natural communities. Another natural cellulase digesting community is the rumen of herbivores (Coleman, 1985; Teather and Wood, 1982), and several cellulase genes have been cloned from these sources (Cappa et al., 1997; Ekinci et al., 2002). Increase the Specific Activity of Cellulolytic Enzymes An approach that has worked exceptionally well for other enzymes is to structurally engineer them to increase catalytic activity, thereby reducing the amount of the enzymes required and effectively lowering the overall cost. Approaches to increasing the specific activity of cellulase have focused on understanding the catalytic process. The CAZY database (www.cazy.org) stores data regarding structural features and carbohydrate binding domains of a number of cellulases (Cantarel et al., 2009). Several enzymes have been crystallized and the active sites examined following mutagenesis of conserved residues (Zhang et al., 2006). In addition to modification of the catalytic site, specific activity can be modified by increasing the affinity of the enzyme for the substrate through altering residues in the carbohydrate binding domain (CBD) or increasing the number of the enzyme’s CBDs (Linder and Teeri, 1997). Two specific approaches, directed evolution and rational design, are considered below. Directed Evolution DNA shuffling using PCR is a powerful way to randomly modify the structure of enzymes, which are later screened for activity (Rabinovich et al., 2002). Some authors (Kim et al., 2000) have utilized the process of DNA shuffling to create a library of cellulase genes that are expressed on the surface of bacteria as fusions to the ice nucleation protein for screening on carboxymethylcellulose (CMC) as substrate. Improvements of up to 5-fold in CMCase activity were obtained by this method, and sequencing indicated that amino acid changes in the catalytic site were responsible for increased specific activity. This approach was used (Kaper et al., 2002) to shuffle the nonhomologous sequences of two different hyperthermostable βglucosidases with temperature optima of 105◦ C and 95◦ C to obtain a more active enzyme with a more moderate optimum of 70◦ C. Another approach (Heinzelman et al., 2009) using structure-guided recombination created chimeras of three class II cellobiohydrolases. The resultant chimeras showed higher thermostability than the most stable parent, indicating that this technique could be used to generate a large combinatorial library of enzymes with desired properties. Techniques such as these may help to manipulate pH and temperature optima for enzymes for industrial applications. Another group (Toyama et al., 2003) used the alternative process of nuclear shuffling in swollen T. reesei conidia to achieve autopolyploidy using colchicine treatment. Successive treatment with benomyl and ethylmethanesulfonate produced hyperproducers, but these were slow growing compared to the original strain (Toyama and Toyama, 2001), which may be a problem for commercial production. Rational Design In contrast to directed evolution where a variety of products are generated by DNA shuffling and then screened for activity, rational design uses targeted approaches to modifying enzymes.

P1: SFK/UKS P2: SFK c10 BLBS075-Hood December 15, 2010

20:55

Trim: 244mm×172mm

10 Enzyme Production Systems for Biomass Conversion

235

Table 10.2. Improvement in enzyme specific activity of cellulolytic enzymes. Enzyme

Improvement Method

Improvement

Reference

Cel9A endoexoglucanase from Thermobifida fusca

Rational design/ computer modeling

40% activity

Escovar-Kousen et al., 2004

Endo-β -1,4-glucanase from B. subtilis BSE616

DNA shuffling

5-fold activity of fusion; 2.2-fold activity of free enzyme

Kim et al., 2000

β-D-glucosidase from Thermotoga neapolitana

Directed evolution

31% activity

McCarthy et al., 2004

Cellobiohydrolase from Talaromyces emersonii in S. cerevisiae

Rational design/ structure-guided engineering of S–S bonds

Thermostability increased by 9◦ C; improved activity at RT

Voutilainen et al., 2009

Cel B from Pyrococcus furiosus

DNA shuffling

1.5- to 3.5-fold thermostability at 70◦ C

Kaper et al., 2002

Lac S from Sulfolobus solfataricus

DNA shuffling

3.5- to 8.6-fold increased lactose hydrolysis

Kaper et al., 2002

Cellobiohydrolase Cel7B Melanocarpus albomyces in S. cerevisiae

Random mutagenesis

2-fold increased activity at 70◦ C

Voutilainen et al., 2007

B. subtilis endo-β-1,4-glucanase (Cel5A)

Directed evolution (error-prone PCR)

2.03- to 2.68-fold

Lin et al., 2009

Availability of crystallographic or site-directed mutagenesis data allows understanding of the structure of the catalytic site of cellulases (Zhang et al., 2006) and provides the basis for a rational design approach to optimize the interaction of the enzyme with the substrate. The crystal structure of the catalytic domains of several, predominantly bacterial, hemicellulases and cellulases has been achieved (Davies and Henrissat, 1995). Molecular modeling can then be used to compute and analyze modifications in structure that might improve binding or catalysis. Such an approach was utilized by Kormos et al. (2000) to identify and modify amino acids in the CBM binding groove and identified tyrosines 19 and 85 in the groove as essential for tight binding of CBDN1 of endoglucanase C from Cellulomonas fimi. Amino acids in the catalytic site of endoglucanase CelC of Clostridium thermocellum that contributed to catalytic activity were identified (Escovar-Kousen et al., 2004; Navas and B´eguin, 1992). Computer modeling was also used to identify conserved amino acids in active sites of an endo/exoglucanase that were then mutated to obtain 40% improvement in activity (Escovar-Kousen et al., 2004). Table 10.2 shows some of the enzymes whose activity was modified by directed evolution or rational design. However, the maximum increase in activity is still a relatively modest 2- to 5-fold despite the intense effort. Specific activity may be slightly more improved by optimizing the length of linker and the number of CBM domains, but if commercial production is the target, other approaches must be considered to improve the economics of biomass hydrolysis.

P1: SFK/UKS P2: SFK c10 BLBS075-Hood December 15, 2010

236

20:55

Trim: 244mm×172mm

Plant Biomass Conversion

Synergistic Proteins That Increase Cellulase Activity Increasing the efficiency of cellulases by the action of other proteins is another way by which conversion of cellulose to fermentable sugars can be increased. Expansins are a group of proteins that “loosen” cell walls during turgor-driven wall expansion, fertilization, and growth (Cosgrove, 1999, 2005) probably by weakening noncovalent tethering of cellulose fibrils through xyloglucan bridges. Cell wall loosening immediately suggests a procedure to increase cellulase access in biomass, but the mechanism of action of expansins remains unclear. The two categories of expansins, A and B, share domain structure, have acidic pH optima, and have a conserved putative CBD and an endoglucanase-like (EG-45) domain, although no hydrolytic activity of expansins has been detected. Polysaccharide hydrolases are another category of enzymes that may weaken the matrix of the plant cell wall, exposing the cellulose fibrils for deconstruction. It appears that most polysaccharide hydrolases act on the hemicellulose crosslinks of the wall, and specifically on the xylan and the mixed linkage 1,3:1,4-β-glucan seen primarily in grasses, which comprise a large portion of agricultural biomass (Cosgrove, 2005, 2000, 1999). These have an acidic pH optimum and could conceivably be used in conjunction with acid pretreatment. Other reactions that may be beneficial include utilization of xyloglucan endotransglycosylases (XETs) and ascorbate. XETs are plant enzymes that catalyze a rearrangement reaction of the xyloglucan network in the plant cell wall and are a proposed part of cell wall loosening (Cosgrove, 1999). Hydroxyl radicals produced by treatment with ascorbate reduce the viscosity of xyloglucan solutions and may expose the cellulose fibrils in cell walls by making the hemicellulose matrix less rigid (Cosgrove, 1999). This reaction may be carried out in conjunction with the xylan hydrolase reaction as part of an acid pretreatment protocol. Protease pretreatment may loosen the cell wall matrix by digesting the protein components. Large quantities of the protease enzyme trypsin are procurable at relatively low cost by extracting and purifying either from the native source or from recombinant proteins made in plants (Horn et al., 2004; Woodard et al., 2003,), but care must be taken to inactivate the enzyme to avoid digestion of cellulolytic enzymes in subsequent steps. In summary, the addition of several different types of enzymes can aid in lowering the amount of cellulase required for the conversion of biomass. The utility of this approach will depend on the cost of the additional proteins compared to the cost of the cellulase. Several commercially important enzymes have been produced in plants (Hood and Howard, 2008) and indicate a path towards expression of several cell-wall-modifying enzymes in plants for biomass utility.

10.3.3 Increase Accumulation of Enzymes in Production Host There is an inverse correlation between the expression level of enzymes in their host and the cost of production. Therefore, a key strategy is to increase the overall accumulation of the enzymes in the host tissue to decrease the overall cost. The National Renewable Energy Laboratory has a target to reduce the cost of ethanol production to $1.07/gallon (in 2002 dollars) by the year 2012—a target that is only one year away. A second target is to produce 60 billion gallons of bioethanol/year by the year 2030 (NREL/BR-510-40742). These aggressive targets demand aggressive approaches to reducing enzyme production costs. While all of the methods discussed above are likely to be helpful in achieving the goal, the overall cost of enzyme production will likely be a key factor for any solution.

P1: SFK/UKS P2: SFK c10 BLBS075-Hood December 15, 2010

20:55

Trim: 244mm×172mm

10 Enzyme Production Systems for Biomass Conversion

237

Microbial Production Systems Both prokaryotic bacteria and eukaryotic fungi produce cellulases, and the enzymes show considerable overlap in structure and function (Gilbert and Hazlewood, 1993). A primary difference exists between aerobic producers and anaerobic producers, however. Most anaerobes produce tightly associated and aggregated enzymes called cellulosomes, which are associated with membranes through dockerin molecules, whereas aerobes produce soluble, secreted enzymes. Temperature optima vary based on source: many thermophilic bacteria produce highly efficient cellulases, but their temperature optima make them uneconomical for hydrolysis. Optimization of conditions for high cellulase enzyme synthesis has been studied using various substrates, inducers, fermentation conditions, and microbial systems. One approach used Trichoderma reesei with Solid Substrate Fermentation on wheat bran and manipulated temperature and moisture conditions to obtain a 6.2-fold increase in activity over controls (Singhania et al., 2007). Other parameters that have been tested are batch vs. continuous fermentation, substrate concentration, fungal vs. bacterial systems, and shake vs. stationary cultures (Bailey and Tahtiharju, 2003; Bailey et al., 1972; Gomes et al., 1992; Jana et al., 1994; Kadam and Keutzer, 1995; Nikolov et al., 2000; Schaffner and Toledo, 1991; Singhania et al., 2007; Spear et al., 1993; Tholudur et al., 1999; Xia and Shen, 2004; Zhou et al., 2009; Zhuang et al., 2007). The enormous literature on manipulation of individual parameters makes it clear that conditions have to be established separately for each organism, and either crude or reconstituted purified mixtures of cellulases tested for activity. The production optimum is not necessarily the optimum for activity. Screening strategies for high-producing strains rely on quantitative assays based on product accumulation, substrate reduction, or change in physical properties of substrates (reviewed in Zhang et al., 2006). These may be conveniently carried out on solid media and enzyme activity inferred by substrate digestion producing haloes. Chromogenic substrates for evaluation of enzyme activity, high throughput and high-density microtiter plates, and PCR cyclers with hot lids to prevent evaporation from samples are among the facilitated screening tools that allow rapid screening of large numbers of samples (reviewed in Zhang et al., 2006). Selection strategies to identify the highest producers are slower and use methods such as cross-feeding where soluble enzymes diffuse around colonies growing in resource-limited areas exhibiting competition to identify high-producing strains (Zhang et al., 2006). A colorimetric microplate assay to screen virulent plant pathogens for cellulolytic activity allows high throughput, quantitative screening of fungal extracts, allowing more efficient bio-prospecting for enzymes (King et al., 2009). Organisms identified by such screening strategies can then be evaluated further by other methods.

Plants to Provide Supply of Enzymes Producing exogenous enzymes in microbes is efficient and is the main method used today for the production of most industrial enzymes. Microbial fermentation for the recovery of commercial amounts of cellulase requires high capital costs for setup and operational costs for running. Thus, investigation into the use of plants as an alternate production system for transgenic proteins for biomass conversion has developed. Plants need little infrastructure beyond ensuring that transgenic plants are contained appropriately, even possibly within dedicated ethanol-producing areas (Howard and Hood, 2007). There are several theoretical reasons why plants may be the best source for the long-term supply of exogenous enzymes, including (1) plants represent the least expensive method to

P1: SFK/UKS P2: SFK c10 BLBS075-Hood December 15, 2010

238

20:55

Trim: 244mm×172mm

Plant Biomass Conversion

produce protein in general, (2) plants do not require the large amount of capital for production compared to microbial fermenters, (3) plant production can be scaled-up or -down without major changes in infrastructure, (4) almost any plant can function as a production system, (5) proteins can be targeted to specific compartments, allowing for increased accumulation in the desired tissue with little interference in other tissues (reduces potential toxicity to the cell), (6) plants have convenient storage, transport, and processing of component materials, and (7) plants have the potential to have lines containing different enzymes combined through crossing (Howard and Hood, 2005; Sainz, 2009; Twyman et al., 2003). Sainz (2009) provides a table of cellulases that have been expressed so far in plants. The choice of plant species and tissue for accumulation of foreign proteins depends on the specific transgene to be expressed, production system employed, transportation needs, and processing and storage requirements. Transformation, expression, and purification technologies also affect the cost of the products produced in plants (Howard and Hood, 2005, 2007). Subcellular targeting is critical for accumulation and protein integrity of cellulase or other industrial enzymes (Clough et al., 2006; Hood et al., 2007). Other factors that affect protein accumulation include (1) the use of the catalytic domain alone, which supports a higher level of expression than the whole enzyme (Sticklen, 2006); (2) dual targeting to two compartments is more efficient in some cases than one (Hyunjong et al., 2006); and (3) optimization of codon usage (Xue et al., 2003). Although the catalytic domain alone was optimal for protein accumulation (Sticklen, 2006), for functionality, the catalytic and binding domains are both required. Targeting enzymes to subcellular compartments may overcome toxicity, permit correct folding, and enable glycosylation. The chloroplast is another desired location because of ease of containment. Using downstream box fusions of Thermobifida fusca cellulase in chloroplasts of transgenic tobacco caused increased expression up to 10.7% TSP. This is significantly higher than the 1% accumulation seen with nuclear-targeted fusions (Gray et al., 2009). However, homoplastic transformation protocols are still elusive for most plants (Grevich and Daniell, 2005; Sticklen, 2008b). Proteins produced in plants can accumulate at significant levels, and in addition, the lignocellulosic by-products can themselves be used for production of ethanol. For example, a heterologous endo-1,4 β-glucanase E1 gene from Acidothermus cellulolyticus under control of a cauliflower mosaic virus 35S promoter in maize was used to obtain active enzyme at 1.13% TSP. Total soluble protein from transgenic maize allowed AFEX-pretreated corn stover to be directly processed into glucose (Ransom et al., 2007) . Plants with underground storage tubers can be induced to accumulate cellulase in the leaf by chloroplast targeting, while the tubers can be used as food. An Acidothermus cellulolyticus endoglucanase E1 gene was placed under the control of an RbcS-3C promoter, alfalfa mosaic virus 5’-untranslated leader, and RbcS-2A signal peptide (Dai et al., 2000) and showed protein accumulation up to 2.6% TSP in the leaf. In addition to simply using plants as a production vehicle in the way microbes are used, other more integrated approaches have been developed. One attractive strategy is to have the enzymes produced from by-products or from tissue normally discarded from existing crop plants (Howard and Hood, 2007). The germ tissue from maize, a by-product of many grain ethanol facilities, was used to demonstrate high accumulation levels of E1 (16% TSP) cellulase as well as cellobiohydrolase (18% TSP) (Hood et al., 2007). Endoglucanase expression from an apoplast-directed protein, expressed from a constitutive promoter, was obtained at high levels in rice without apparent deleterious effects. Transgenic rice extracts were active in converting pretreated rice and maize biomass into ethanol at 22% and 30%, respectively, compared to 62% and 95% conversion using purified commercial enzyme (Oraby et al., 2007). These last

P1: SFK/UKS P2: SFK c10 BLBS075-Hood December 15, 2010

20:55

Trim: 244mm×172mm

10 Enzyme Production Systems for Biomass Conversion

239

several examples show great promise in that it allows the crop to be harvested for food and the rest of the plant to be used for production of cellulase enzymes after harvest. In addition to cellulases, plants have been used to produce other potentially useful proteins for biomass conversion such as ligninases (Clough et al., 2006; Hood et al., 2003) and xylanase (Borisjuk et al., 1999). Transgenic production of trypsin as a zymogen has been successfully achieved in maize tissue, a system that provides cost effective transport, storage, and use (Woodard et al., 2003). As mentioned above, in addition to being cost-effective sources for protein production, transgenic plant lines can be crossed to other transgenic lines to allow the production of several enzymes in the same plant. Another option is to express degradative enzymes in vegetative tissue to decrease the burden of adding exogenous cellulase. This would increase the effectiveness of cellulase activity by working from the inside of cell wall tissue, while the outer tissue is exposed to exogenous cellulase. Cellulose hydrolytic enzymes have been engineered into potato, tobacco, Arabidopsis, rice, barley, and maize. Many of these proteins were targeted for expression to specific tissues or sub-cellular compartments to overcome potential toxicity (Sticklen, 2006; Torney et al., 2007). In order to be of use for saccharification, these enzymes must be used either with an enzymatic pretreatment, be extracted prior to thermochemical pretreatment, or be stable under thermochemical pretreatment conditions. The required changes to integrate this approach with downstream processing limit the rapid acceptance for incremental improvements, but this may change if progress continues in the future toward less harsh pretreatment conditions, in which more of an enzymatic approach is integrated.

Supply Live Microbes Currently, enzymes from fungi (Ramamurthy et al., 1992) or bacteria (Gilbert and Hazlewood, 1993) are used in crude or purified, reconstituted mixtures for the aerobic saccharification step. Rather than add the enzymes directly, it is possible to add live cellulase-producing organisms for the initial aerobic saccharification followed by addition of yeasts for the anaerobic fermentation processes (Mamma et al., 1996). As microbes can be finicky about growing on different substrates, this approach will require more attention than the simple addition of enzymes. It will also require that the products of the aerobic saccharification reaction be compatible with the organism used for fermentation. This approach, however, does offer a way to greatly reduce the costs of making the enzymes in separate fermentation facilities if these factors can be adequately controlled.

Simultaneous Saccharification and Fermentation The process of deriving ethanol from lignocellulosic feedstock requires that carbohydrate polymers be deconstructed and fermented to ethanol. This process can be carried out under several paradigms. In Separate Hydrolysis and Fermentation (SHF), the hydrolysis step is carried out aerobically and the products separately fermented by anaerobes. This process is inefficient because of the separate chambers required for each process and also because products of hydrolysis that accumulate can be inhibitory to the hydrolytic enzymes (Taherzadeh and Karimi, 2007). Simultaneous Saccharification and Fermentation (SSF) refers to cellulose hydrolysis carried out in the presence of a fermentative organism to relieve end-product inhibition by removal of glucose through fermentation (Sun and Cheng, 2002; Taherzadeh and Karimi, 2007).

P1: SFK/UKS P2: SFK c10 BLBS075-Hood December 15, 2010

240

20:55

Trim: 244mm×172mm

Plant Biomass Conversion

The theoretical advantages of SSF over SHF are many: increase in hydrolysis rate due to removal of inhibitory sugars by fermentation, shorter process time, and less capital cost due to single reactor usage. In addition, enzyme and sterility requirements can also be reduced (Sun and Cheng, 2002). Simultaneous Saccharification and Co-Fermentation (SSCF) increases the theoretical ethanol yield by incorporating organisms that ferment pentose to ethanol. Mixed cultures of hydrolytic and fermentative strains can be enhanced by modifying hexose fermenting strains (see Chapter 8 in this volume). This has been achieved in Zymomonas mobilis and Saccharomyces cerevisiae to also ferment pentose (Hahn-H¨agerdal et al., 2007; Olofsson et al., 2008; Rudolf et al., 2008; Zhang et al., 1995), thus increasing ethanol yield from biomass. More recently, the concept of Consolidated Bioprocessing (CBP), reviewed in Lynd et al. (2005), has been garnering a lot of interest because of the potential for drastically reducing production cost. CBP incorporates biological production of cellulase, biomass hydrolysis, and sugar fermentation, all in a single step. Therefore, the basic requirement is the development of CBP-enabling microorganisms with high levels of protein expression such as S. cerevisiae or E. coli, or at least a compatible mixed culture, and engineering in cellulose utilization at high efficiencies. Many of the suggested organisms either naturally or through modification ferment sugars. Lynd et al. (2005) proposed the utilization of Cellulose–Enzyme–Microbe complexes as opposed to Cellulose-Enzyme (CE) complexes, citing the higher rates of hydrolysis with the former, but in principle, this procedure could be adapted to both, depending on the kinetics observed. In addition, using biomass feedstocks that are engineered for low recalcitrance will provide extra benefits. SSF, SSCF, and CBP (Table 10.3) can be carried out in batch, fed-batch (semi-continuous), or continuous cycles. The advantages of fed-batch and continuous cycles are the relief from buildup of sugars, which severely inhibit cellulase enzymes. However, since hydrolysis is kinetically slower than fermentation, alcohol-tolerance of strains used is also a consideration.

10.4 Cost of Producing Exogenous Enzymes There are many types of enzyme production systems in use today, including microbial cultures, animal cell cultures, submerged plant cultures, row crops, Baculovirus and many more. Each of these can play an important role for protein production and each has inherent limitations. To reach the cost targets for cellulase, only row crops and microbes are inherently at a low enough cost to be viable for cellulase production. In addition to the inherent cost of the host, the cost of producing a specific protein is inversely proportional to the ability of the host to accumulate that protein. Therefore, a host that can accumulate large amounts of the desired protein will be able to produce the protein for a lower cost. The production practices of both microbes and row crops have been optimized for decades, and while they continue to enjoy incremental improvements, it is unlikely that there will be a quantum change in their cost structure. Improvements therefore must come in either their ability to accumulate the desired protein or other efficiencies related to the overall integration of the process. In the case of microbial production systems, the lowest cost of cellulase enzymes is by extraction from fungal cultures. Years of experience using these cultures have allowed optimization for cellulase production. In the near term, we should expect that these fungal systems

P1: SFK/UKS P2: SFK c10 BLBS075-Hood December 15, 2010

20:55

Trim: 244mm×172mm

10 Enzyme Production Systems for Biomass Conversion

241

Table 10.3. Fermentation procedures. Type of Treatment

Advantages

Disadvantages

Comments

References

Simulataneous saccharification and fermentation (SSF)

Single reactor; improved rates and yields

Sub-optimal conditions; ethanol-inhibited fermentation; difficulty recycling hydrolytic enzymes and fermentative organisms; separate C5 and C6 fermentation

Conversion rates typically 30% higher in SSF than SHF using; fed batch usage limits problems with inhibitors, mass, and heat transfer

Drissen et al., 2008; Olofsson et al., 2008b

Simultaneous saccharification and co-fermentation (SSCF)

Single reactor, higher yield xylose, and glucose conversion

Suboptimal conditions; xylose uptake needs low glucose concentrations for uptake; low temperatures for fermentation do not favor hydrolytic enzymes from thermophiles; needs heat-tolerant yeast strains

78% of theoretical yield of (35 g/L) ethanol obtained with 7% solid loading in fed-batch using S. cerevisiae TMB340 strain

Olofsson et al., 2008a, 2008b

Consolidated bio processing (CBP)

Lower costs, higher efficiency

Requires engineered hydrolytic strains to be ethanol tolerant; possible nontarget metabolite production

Lynd et al., 2005

will continue to provide the lowest cost cellulase enzymes. While there are continual improvements being made as to strains that accumulate higher levels of enzymes, it is unlikely that huge improvements will be seen from the already very high current levels. Likewise, the optimization of the production practices are likely to provide incremental improvements but unlikely to have a major effect in driving the cost down because of the maturity of the system. It is more likely that microbial production systems will rely on synergistic proteins or increases in specific activity to help drive cost down by decreasing the protein load for bioconversion. In contrast to microbial production, relatively little experience has been gained with accumulating cellulase in plants and no commercial production of cellulase using plants that exists today. Plant extracts as an inexpensive enzyme formulation will require a well-conceptualized and developed bioprocess. In the near term, the most realistic way to produce and formulate biomass hydrolytic enzymes will be from crude aqueous plant extracts of leaves, seed, stalks, or their fractions (Sainz, 2009). Changes in downstream biomass processing would not be required and would be compatible with the current system of using microbial extracts. Direct addition of dry fractionated plant tissue may be the most desirable approach because of simple processing, storage stability, and low cost associated with water addition and subsequent drying of residual spent transgenic biomass. The extent of processing of

P1: SFK/UKS P2: SFK c10 BLBS075-Hood December 15, 2010

242

20:55

Trim: 244mm×172mm

Plant Biomass Conversion

plant material containing biomass-degrading enzymes will depend on whether or not the plant tissue is amenable to dry or wet fractionation (pressing fresh juice is also wet fractionation) without undergoing aqueous extraction. Direct application of fractionated transgenic plant tissue in saccharification of pretreated biomass hinges on enzyme concentration in the transgenic plant fraction. For example, at enzyme loading of 8 g/kg biomass (20 g/kg cellulose), the amount of associated plant material should not exceed more than 20% of pretreated biomass concentration because the increase of total solids in the saccharification reaction mix could adversely affect cellulose hydrolysis, viscosity, nonproductive binding of enzymes, inhibition, etc. In this case, the amount of hydrolytic enzyme(s) should be at least 4% of dry wt. (40 g/kg) of fractionated transgenic plant tissue. Note that not all plant tissues are equally practical for direct application. For example, soybeans would be expensive for direct application as soy protein (meal) is the major revenue and feed source. The same is true for whole corn seed, unless extremely high enzyme accumulation levels are reached. Thus, seed crops must be extracted to recover the enzyme from by-products, allowing the seed to be sold for its originally intended use to capture the credits. To eliminate drying and transportation cost of extracted enzyme, corn mills, corn-to ethanol plants, and soy crushers should be an integral part of biomass conversion, which may not be a trivial stipulation. The conclusion from this discussion is that expression of enzymes in tissue or residue with little or no by-product value would be the ideal candidates for direct delivery. Examples of these include rice straw, corn stover, corn germ, and switchgrass. Of course, the key is to achieve accumulation levels of at least 4% dry wt.

10.4.1 Cost Analysis Taking into account what we know about production cost, we have made some generalizations about the enzyme cost and potential cost in the future. We have used models for production cost described below, knowing that there are continual improvements in enzyme activity as mentioned in the sections above that will affect the absolute cost. Any increase in specific activity or synergism with cellulase should apply equally to both microbes and plants, therefore the hope is these numbers will show the relative merits of the two systems and be a guide to the absolute costs. To determine the amount of enzyme per gallon ethanol and unit enzyme cost ($/kg protein) required to achieve our target cost of $0.17/gal ethanol (Table 10.1), we assumed enzyme loading of 12 FPU/g cellulose. Using enzyme specific activity of 600 FPU/g, this enzyme loading is equivalent to 20 g enzyme/kg cellulose or 8 g enzyme/kg biomass, assuming an average of 40% cellulose in biomass (Mosier et al., 2004). At 85 gal ethanol/ton biomass, the necessary amount of enzyme per gallon ethanol produced is 94 g/gal (100 g/gal in Table 10.1). Now, to achieve the near-term target of $0.17, one has to be able to produce cellulases for less than $2.0/kg enzyme.

Cost of Cellulases Produced by Microbial Fermentation The best cost models for production are undoubtedly in private companies that keep this information confidential. We have used a variety of public information sources to arrive at our cost for production models, knowing these numbers will continue to change with

P1: SFK/UKS P2: SFK c10 BLBS075-Hood December 15, 2010

20:55

Trim: 244mm×172mm

10 Enzyme Production Systems for Biomass Conversion

243

Table 10.4. Cost estimate of enzyme produced by microbial fermentation.a Fermentation capacity (cu.m.) Fermentation titer (g/L) Batch cycle (days) Recovery yield (%) Product per week (ton) Number of batches/year Plant output per year Fermentation (variable and fixed) cost ($/kg) Recovery (variable and fixed) cost ($/kg) Total enzyme production cost ($/kg) a Hepner

2,600 50 7 85 111 50 5,525 6.0 4.2 10.2

and Assoc., 1995

improvements. The hope is that these models can be used as a guide to compare relative trends in the industry. A 2002 NREL report (NREL/TP-510-32438) showed that 2000 tons/day would be an optimal size of a lignocellulosic biomass conversion facility; the facility would utilize 10% of available corn acres and haul stover from within a 50 miles radius. At 8 kg enzyme loading per ton biomass, an enzyme production facility capable of supplying 16 tons of enzyme per day (5,600 tons enzyme per year) has to be built on site to eliminate transportation and minimize enzyme formulation and stabilization cost. If enzyme is produced by a fungal fermentation process that lasts 7 days and yields 50 g cellulase/L, the required fermentation capacity would be 2.2 MM L per week. A fermentation facility with 5 fermenters of 500,000 L each would cost about $50 MM. The enzyme facility (centrifugation, concentration, and polishing) to process 2.2 MM L of fermentation broth per week would probably require an additional investment of up to $40 MM. The 2022 mandate of producing 36 billion gallons of ethanol translates to (36 × 100/gal) a total of 3.6 MM metric tons of enzyme (Table 10.1). With 5,600 tons of enzyme per facility, 640 fermentation facilities have to be built; thus, the total capital investment for enzyme production would amount to approximately $57 billion (640 × 50 MM = 32 MM for fermentation + 640 × 40 MM = 25.6 MM for processing). A model of the cost breakdown of enzyme produced by fermentation is given above (Table 10.4). The total cost based on this model shows microbial production to be at approximately $10/kg with the assumptions of protein loading from above. Cost of Cellulase Produced in Transgenic Plants The bioprocessing challenge for transgenic tissue with enzymes is to deliver formulated enzyme preparations to a saccharification facility at the lowest possible cost. Protein and tissue stability, tissue fractionation, protein extraction and formulation, as well as storage and transportation all add to this cost. For example, if transportation costs of $14/ metric ton (NREL/TP-510-32438) were assumed, then transporting an enzyme formulation that contains 5% w/w enzyme would already cost $0.28/kg enzyme. In addition, any wet fractionation such as fresh tissue pressing, aqueous extraction of dry plant tissue followed by residue separation would add at least $1/kg to the bioprocessing cost. Therefore, in order to get the enzyme cost of less than $2/kg, production of transgenic plant material should not cost more than $1.00/kg enzyme, which translates to raw material cost of $100/dry metric ton for material containing 10% enzyme.

P1: SFK/UKS P2: SFK c10 BLBS075-Hood December 15, 2010

244

20:55

Trim: 244mm×172mm

Plant Biomass Conversion

Table 10.5. Cost estimate of enzyme produced from defatted corn germ. Defatted corn germ price ($/kg germ) Enzyme expression level (% dry base) Extraction and recovery yield (%) Enzyme output per year (ton) Cost of enzyme produced in corn germ ($/kg enzyme) Recovery (variable and fixed) cost ($/kg enzyme) Byproduct credits Residual germ cake at $55/ton ($/kg enzyme) Total enzyme production cost ($/kg)

0.1 1.5 85 5,600 7.8 7.0 4.0 10.8

This calculation clearly exemplifies the cost and processing constraints imposed on enzyme production from transgenic plants. Thus, to achieve production cost targets for biomass enzymes, the following considerations must be employed: (1) eliminate transportation costs by integrating enzyme processing into biomass conversion facility, (2) minimize fractionation/extraction cost of transgenic material, and (3) reduce the contribution of transgenic plant material to enzyme production cost by capturing plant biomass value through by-product credits or cellulose. One of the most crucial assumptions today for plants is how well can the tissue accumulate the cellulase enzymes. Little experience is yet achieved with plants as compared to microbial cultures to know what the uppermost accumulation levels can be. Therefore, the model used for plants is based on the level of accumulation required for plants to be at par with the cost of microbial production and then estimating the necessary accumulation level. Table 10.5 provides an example of cellulase produced in corn germ and extracted at comparable cost with microbes. In this case, the level of protein accumulation required is 1.5 % of germ dry weight. While this seems intuitively reasonable, it has also been achieved (Hood and Howard, unpublished results), making this a realistic case for maize germ as well as potentially for other row crops. There is every reason to believe that expression levels will continue to improve in plants as it is a relatively young science. For this reason, we developed a future case where the cellulase represents 7.5 % of germ dry weight or 25% of the total soluble protein. On the basis of accumulation of proteins in other systems, this seems realistic and especially feasible for seed proteins where the vast majority of these serve only to provide amino acid reservoirs for germination. At this level of accumulation, the cost of extracted enzyme drops to $2.50/kg. If instead of extracting the germ we can use it directly, the model predicts that cellulase accumulation levels of 1.5% dry weight are needed to match the comparable microbially produced cellulase. In addition, there is the benefit of a much lower capital investment. As discussed above, however, this may require addition of too many solids, and therefore even though the cost is comparable, it may not be practical and the level may need to be as high as 4% to become practical. If the level of cellulase reaches 7.5% dry weight, then it is well below the limitations for additional solids. The cost drops to $1.50/kg, making this the most economical scenario for biomass conversion and below the cost target. This level of expression has been reached in the laboratory and improvements in expression technology continue in this relatively new field. The other practical consideration is the acreage required to grow crops to produce this amount of enzyme. We modeled this concept previously, considering the proximity limitations of the lignocellulosic biomass to the ethanol facility to avoid large transportation costs. The

P1: SFK/UKS P2: SFK c10 BLBS075-Hood December 15, 2010

20:55

Trim: 244mm×172mm

10 Enzyme Production Systems for Biomass Conversion

245

Table 10.6. Comparison of production systems for 5,600 metric tons of enzyme per year. Manufacturing Cost Estimates ($/kg)

Capital Investment

Comments

System

Source

Microbial

Microbial fermentation 50 g/L fungus

$10

∼$100 MM

All aspects feasible today. Need improvements to lower protein loading to meet cost targets

Extraction from seed w/ by-product (current)

Extraction from corn germ @ 1.5% of dry wt

$10

∼$40 MM

Expression levels reached but not scaled up or commercialized

Extraction from seed w/ by-product (future)

Extraction from corn germ @7.5%of dry wt

$2.50

<$20 MM

Can approach cost targets without major improvements lowering protein loading

Seed by-product (current)

Direct use of corn germ @1.5% of dry wt

$7.8

<$2 MM

Can meet cost targets without improvements in lowering protein loading but may not be practical because of high solid percentage

Seed by-product (future)

Direct use of corn germ @ 7.5% dry wt

$1.50

<$2 MM

The lowest cost if expression levels can be raised to their potential. No other practical limitations

Extraction from vegetative tissue

Extraction from rice leaves @ 0.1% dry wt

10

∼$40 MM

Must be extracted prior to pretreatment to keep activity (Oraby et al., 2007)

study demonstrated that if cellulase levels were 0.1% of the dry weight of the seed (1% of germ dry weight), it was more than sufficient to keep the acreage of the cellulase crop less than the acreage needed to supply the lignocellulosic biomass (Howard and Hood, 2007). Table 10.6 shows the cost factors above along with that of a previously discussed rice model to illustrate cellulase coming from the vegetative crop (Oraby et al., 2007). In summary, plant production systems that accumulate cellulase in the normally unused or low value portion of the plants are now approaching competitive cost structure with microbial systems. As expression technology improves and cellulase reaches 4% of the dry weight, direct delivery of plant tissue can be the system of choice and will easily reach the current cost targets.

10.5 Summary and Future Prospects The current production of bioethanol from starch produces the ethical dilemma of “food or fuel,” but the use of agricultural residues (Chapter 2 in this volume) and other

P1: SFK/UKS P2: SFK c10 BLBS075-Hood December 15, 2010

246

20:55

Trim: 244mm×172mm

Plant Biomass Conversion

lignocellulosic biomass (Chapters 3–5 in this volume) can potentially resolve this issue. However, lignocellulosic biomass is considerably more recalcitrant to digestion than starch and requires pretreatment and large amounts of enzymes for its deconstruction, which are expensive using current production methods. The complex problem of converting biomass to fuel is still in its infancy. The near-term will clearly rely on microbial production systems. However, none of the current systems can meet the target enzyme cost today using the assumptions we employed in our models. In order to meet the cost target, improved facility design, process, organisms, enzymes, titer, quantity, recovery, and efficiency must be addressed simultaneously. Because of the magnitude of the problem, the most likely outcome will not be a single breakthrough but rather combining the outcomes of several different lines of research. As an example, increased susceptibility of feedstocks along with increased synergy of enzymes will decrease enzyme loading, as will increases in specific activity. Companies such as Novozymes and Genencor reportedly have reduced the cost of enzymes necessary for producing a gallon of ethanol from $5.40 in 2000 to $0.20 in 2005 (Moreira, 2005). Despite this reported improvement, enzymes still represent a key stumbling block to bioconversion and a practical limitation to achieving this theoretical cost because of the large capital investment needed. Microbial production of enzymes will need to rely more on decreased protein loading than on increase in accumulation to reach lower costs. In this regard, the use of plants selected to be more susceptible to digestion can also help lower the cost requirements for exogenous enzymes. In the long term, plants may also provide the lowest cost of exogenous enzymes coming either from seed by-products or from vegetative tissue. While plants can also benefit from any decreases in protein loading similarly to microbes, plants can also lower cost by increasing accumulation of the enzymes and by better integration of the overall process. One can imagine a fully integrated and self-contained system where a crop such as maize provides the stover for the cellulose biomass designed to be susceptible to digestion and the enzymes required for digestion contained in both the stover and the germ by-product. The endosperm (starch fraction) can continue to be used as it is today to produce grain ethanol. Plants have the potential to become enzyme production systems, and the ultimate one-stop shop for the efficient production of cellulase enzymes as well as the substrate for deconstruction to bioethanol.

References Ander, P. & Eriksson, K. 1977. Selective degradation of wood components by white-rot fungi. Physiologia Plantarum, 41, 239–248. Andric, P., Meyer, A., Jensen, P. & Dam-Johansen, K. 2010. Reactor design for minimizing product inhibition during enzymatic lignocellulose hydrolysis: I. Significance and mechanism of cellobiose and glucose inhibition on cellulolytic enzymes. Biotechnology Advances, 28, 308–324. Austin, S., Bingham, E., Mathews, D., Shahan, M., Will, J. & Burgess, R. 1995. Production and field performance of transgenic alfalfa (Medicago sativa L.) expressing alpha-amylase and manganese-dependent lignin peroxidase. Euphytica, 85, 381–393. Bailey, M. J. & Tahtiharju, J. 2003. Efficient cellulase production by Trichoderma reesei in continuous cultivation on lactose medium with a computer-controlled feeding strategy. Applied Microbiology and Biotechnology, 62, 156–162. Bailey, P. J., Keilich, G. & Husemann, E. 1972. Isolation and characterisation of cellulase (beta-1,4-glucan 4-glucanohydrolase) from Chaetomium globosum Kunze. Material und Organismen, 7, 29–43.

P1: SFK/UKS P2: SFK c10 BLBS075-Hood December 15, 2010

20:55

Trim: 244mm×172mm

10 Enzyme Production Systems for Biomass Conversion

247

Bayer, E., Belaich, J., Shoham, Y. & Lamed, R. 2004. The cellulosomes: Multienzyme machines for degradation of plant cell wall polysaccharides. Annual Review of Microbiology, 58, 521–554. Bayer, E. A., Chanzy, H., Lamed, R. & Shoham, Y. 1998. Cellulose, cellulases and cellulosomes. Current Opinion in Structural Biology, 8, 548–557. Borisjuk, N., Borisjuk, L., Logendra, S., Petersen, F., Gleba, Y. & Raskin, I. 1999. Production of recombinant proteins in plant root exudates. Nature Biotechnology, 17, 466– 469. Breuil, C., Chan, M. & Saddler, J. N. 1990. Comparison of the hydrolytic activity of commercial cellulase preparations. Applied Microbiology and Biotechnology, 34, 31–35. Campbell, M. & Sederoff, R. 1996. Variation in lignin content and composition (mechanisms of control and implications for the genetic improvement of plants). Plant Physiology, 110, 3. Cantarel, B., Coutinho, P., Rancurel, C., Bernard, T., Lombard, V. & Henrissat, B. 2009. The carbohydrate-active enZymes database (CAZy): An expert resource for glycogenomics. Nucleic Acids Research, 37, D233. Cappa, F., Riboli, B., Rossi, F., Callegari, M. L. & Cocconcelli, P. S. 1997. Construction of novel Ruminococcus albus, strains with improved cellulase activity by cloning of Streptomyces rochei endoglucanase gene. Biotechnology Letters, 19, 1151–1155. Carrard, G., Koivula, A., Soderlund, H. & Beguin, P. 2000. Cellulose-binding domains promote hydrolysis of different sites on crystalline cellulose. Proceedings of the National Academy of Sciences of the United States of America, 97, 10342–10347. Chan, M., Breuil, C., Schwald, W. & Saddler, J. N. 1989. Comparison of methods for quantifying the hydrolytic potential of cellulase enzymes. Applied Microbiology and Biotechnology, 31, 413–418. Chen, F. & Dixon, R. 2007. Lignin modification improves fermentable sugar yields for biofuel production. Nature Biotechnology, 25, 759–762. Clough, R. C., Pappu, K., Thompson, K., Beifuss, K., Lane, J., Delaney, D. E., Harkey, R., Drees, C., Howard, J. A. & Hood, E. E. 2006. Manganese peroxidase from the white-rot fungus Phanerochaete chrysosporium is enzymatically active and accumulates to high levels in transgenic maize seed. Plant Biotechnology Journal, 4, 53–62. Coleman, G. S. 1985. The cellulase content of 15 species of entodiniomorphid protozoa, mixed bacteria and plant debris isolated from the ovine rumen. The Journal of Agricultural Science, 104, 349–360. Cosgrove, D. J. 1999. Enzymes and other agents that enhance cell wall extensibility. Annual Review of Plant Physiology Plant Molecular Biology, 50, 391–417. Cosgrove, D. J. 2000. Expansive growth of plant cell walls. Plant Physiology and Biochemistry, 38, 109–124. Cosgrove, D. J. 2005. Growth of the plant cell wall. Nature Reviews Molecular Cell Biology, 6, 850–861. Cullen, D. 2002. Molecular genetics of lignin-degrading fungi and their applications in organopollutant degradation. In: Kempken, F. V. E. (ed.) The Mycota XI Agricultural Applications. 1st edn. Berlin, Heidelberg, New York: Springer-Verlag. Dai, Z., Hooker, B., Anderson, D. & Thomas, S. 2000. Improved plant-based production of E1 endoglucanase using potato: Expression optimization and tissue targeting. Molecular Breeding, 6, 277–285. Davies, G. & Henrissat, B. 1995. Structures and mechanisms of glycosyl hydrolases. Structure, 3, 853–859.

P1: SFK/UKS P2: SFK c10 BLBS075-Hood December 15, 2010

248

20:55

Trim: 244mm×172mm

Plant Biomass Conversion

Drissen, R., Maas, R., Tramper, J. & Beeftink, H. 2008. Modelling ethanol production from cellulose: separate hydrolysis and fermentation versus simultaneous saccharification and fermentation. Biocatalysis and Biotransformation, 27, 27–35. Edwards, I., Upchurch, R. & Zak, D. 2008. Isolation of fungal cellobiohydrolase I genes from sporocarps and forest soils by PCR. Applied and Environmental Microbiology, 74, 3481. Ekinci, M. S., Martin, J. C. & Flint, H. J. 2002. Expression of a cellulase gene, celA, from the rumen fungus Neocallimastix patriciarum in Streptococcus bovis by means of promoter fusions. Biotechnology Letters, 24, 735–741. Eriksson, T., Karlsson, J. & Tjerneld, F. 2002. A model explaining declining rate in hydrolysis of lignocellulose substrates with cellobiohydrolase I (Cel7A) and endoglucanase I (Cel7B) of Trichoderma reesei. Applied Biochemistry and Biotechnology, 101, 41–60. Escovar-Kousen, J. M., Wilson, D. & Irwin, D. 2004. Integration of computer modeling and initial studies of site-directed mutagenesis to improve cellulase activity on Cel9A from Thermobifida fusca. Applied Biochemistry and Biotechnology, 113, 287–297. Galbe, M. & Zacchi, G. 2002. A review of the production of ethanol from softwood. Applied Microbiology and Biotechnology, 59, 618–628. Gilbert, H. & Hazlewood, G. 1993. Bacterial cellulases and xylanases. Microbiology, 139, 187–194. Gilbert, H. J. 2007. Cellulosomes: Microbial nanomachines that display plasticity in quaternary structure. Molecular Microbiology, 63, 1568–1576. Gomes, I., Gomes, J., Steiner, W. & Esterbauer, H. 1992. Production of cellulase and xylanase by a wild strain of Trichoderma viride. Applied Microbiology and Biotechnology, 36, 701–707. Gray, B. N., Ahner, B. A. & Hanson, M. R. 2009. High-level bacterial cellulase accumulation in chloroplast-transformed tobacco mediated by downstream box fusions. Biotechnology and Bioengineering, 102, 1045–1054. Gressel, J. 2007. Transgenics are imperative for biofuel crops. Plant Science, 174, 246– 263. Grevich, J. & Daniell, H. 2005. Chloroplast genetic engineering: Recent advances and future perspectives. Critical Reviews in Plant Sciences, 24, 83–107. Gusakov, A., Salanovich, T., Antonov, A., Ustinov, B., Okunev, O., Burlingame, R., Emalfarb, M., Baez, M. & Sinitsyn, A. 2007. Design of highly efficient cellulase mixtures for enzymatic hydrolysis of cellulose. Biotechnology and Bioengineering, 97, 1028–1038. Hahn-H¨agerdal, B., Karhumaa, K., Fonseca, C., Spencer-Martins, I. & Gorwa-Grauslund, M. 2007. Towards industrial pentose-fermenting yeast strains. Applied Microbiology and Biotechnology, 74, 937–953. Hartati, S., Sudarmonowati, E., Park, Y. W., Kaku, T., Kaida, R., Baba, K. I. & Hayashi, T. 2008. Overexpression of Poplar Cellulase Accelerates Growth and Disturbs the Closing Movements of Leaves in Sengon. Plant Physiology, 147, 552–561. Heinzelman, P., Snow, C., Wu, I., Nguyen, C., Villalobos, A., Govindarajan, S., Minshull, J. & Arnold, F. 2009. A family of thermostable fungal cellulases created by structure-guided recombination. Proceedings of the National Academy of Sciences, 106, 5610. Henrissat, B. 1994. Cellulases and their interaction with cellulose. Cellulose, 1, 169–196. Hood, E.E. & Howard, J.A. 2008. Over-expression of novel proteins in maize. In: Kriz, A. & Larkins, B. (eds.) Molecular Genetic Approaches to Maize Improvement. Heidelberg: Springer-Berlin; pp. 91–105. Hood, E. E., Bailey, M. R., Beifuss, K., Magallanes-Lundback, M., Horn, M. E., Callaway, E., Drees, C., Delaney, D. E., Clough, R. & Howard, J. A. 2003. Criteria for high-level

P1: SFK/UKS P2: SFK c10 BLBS075-Hood December 15, 2010

20:55

Trim: 244mm×172mm

10 Enzyme Production Systems for Biomass Conversion

249

expression of a fungal laccase gene in transgenic maize. Plant Biotechnology Journal, 1, 129–140. Hood, E. E., Love, R., Lane, J., Bray, J., Clough, R., Pappu, K., Drees, C., Hood, K. R., Yoon, S., Ahmad, A. & Howard, J. A. 2007. Subcellular targeting is a key condition for high-level accumulation of cellulase protein in transgenic maize seed. Plant Biotechnology Journal, 5, 709–719. Horn, M., Woodard, S. & Howard, J. 2004. Plant molecular farming: Systems and products. Plant Cell Reports, 22, 711–720. Howard, J. A. & Hood, E. 2005. Bioindustrial and biopharmaceutical products produced in plants. Advances in Agronomy, 85, 91–124. Howard, J. A. & Hood, E. E. 2007. Methods for growing nonfood products in transgenic plants. Crop Science, 47, 1255–1262. Hyunjong, B., Lee, D. & Hwang, I. 2006. Dual targeting of xylanase to chloroplasts and peroxisomes as a means to increase protein accumulation in plant cells. Journal of Experimental Botany, 57, 161. Iwase, A., Hideno, A., Watanabe, K., Mitsuda, N. & Ohme-Takagi, M. 2009. A chimeric NST repressor has the potential to improve glucose productivity from plant cell walls. Journal of Biotechnology, 142, 279–284. Jacobsen, J., Lydolph, M. & Lange, L. 2005. Culture independent PCR: An alternative enzyme discovery strategy. Journal of Microbiological Methods, 60, 63–71. Jana, S. K., Ghosh, V. K. & Singh, A. 1994. Production and hydrolytic potential of cellulase enzymes from a mutant strain of Trichoderma reesei. Biotechnology and Applied Biochemistry, 20, 233–239. Jorgensen, H., Kristensen, J. B. & Felby, C. 2007. Enzymatic conversion of lignocellulose into fermentable sugars: Challenges and opportunities. Biofuels, Bioproducts and Biorefining, 1, 119–134. Kadam, K. L. & Keutzer, W. J. 1995. Enhancement in cellulase production by Trichoderma reesei Rut-C30 due to citric acid. Biotechnology Letters, 17, 1111–1114. Kaida, R., Kaku, T., Baba, K., Oyadomari, M., Watanabe, T., Nishida, K., Kanaya, T., Shani, Z., Shoseyov, O. & Hayashi, T. 2009. Loosening xyloglucan accelerates the enzymatic degradation of cellulose in wood. Molecular Plant, 2, 904–909. Kaper, T., Brouns, S., Geerling, A., De Vos, W. & Van Der Oost, J. 2002. DNA family shuffling of hyperthermostable beta-glycosidases. Biochemical Journal, 368, 461. Kim, Y., Jung, H. & Pan, J. 2000. Bacterial cell surface display of an enzyme library for selective screening of improved cellulase variants. Applied and Environmental Microbiology, 66, 788. King, B., Donnelly, M., Bergstrom, G., Walker, L. & Gibson, D. 2009. An optimized microplate assay system for quantitative evaluation of plant cell wall-degrading enzyme activity of fungal culture extracts. Biotechnology Bioengineering, 102, 1033–1044. Kormos, J., Johnson, P., Brun, E., Tomme, P., Mcintosh, L., Haynes, C. & Kilburn, D. 2000. Binding site analysis of cellulose binding domain CBD (N1) from endoglucanse C of Cellulomonas fimi by site-directed mutagenesis. Biochemistry, 39, 8844–8852. Lee, C., Teng, Q., Huang, W., Zhong, R. & Ye, Z. 2009. Down-regulation of PoGT47C expression in poplar results in a reduced glucuronoxylan content and an increased wood digestibility by cellulase. Plant and Cell Physiology, 50, 1075. Li, L., Zhou, Y., Cheng, X., Sun, J., Marita, J., Ralph, J. & Chiang, V. 2003. Combinatorial modification of multiple lignin traits in trees through multigene cotransformation. Proceedings of the National Academy of Sciences of the United States of America, 100, 4939.

P1: SFK/UKS P2: SFK c10 BLBS075-Hood December 15, 2010

250

20:55

Trim: 244mm×172mm

Plant Biomass Conversion

Lin, L., Meng, X., Liu, P., Hong, Y., Wu, G., Huang, X., Li, C., Dong, J., Xiao, L. & Liu, Z. 2009. Improved catalytic efficiency of Endo- -1, 4-glucanase from Bacillus subtilis BME-15 by directed evolution. Applied Microbiology and Biotechnology, 82, 671–679. Linder, M. & Teeri, T. 1997. The roles and function of cellulose-binding domains. Journal of Biotechnology, 57, 15–28. Luczkovich, J. & Stellwag, E. 1993. Isolation of cellulolytic microbes from the intestinal tract of the pinfish, Lagodon rhomboides: Size-related changes in diet and microbial abundance. Marine Biology, 116, 381–388. Lynd, L., Zyl, W., McBride, J. & Laser, M. 2005. Consolidated bioprocessing of cellulosic biomass: An update. Current Opinion in Biotechnology, 16, 577–583. Mamma, D., Koullas, D., Fountoukidis, G., Kekos, D., Macris, B. & Koukios, E. 1996. Bioethanol from sweet sorghum: Simultaneous saccharification and fermentation of carbohydrates by a mixed microbial culture. Process Biochemistry, 31, 377– 381. McCarthy, J., Uzelac, A., Davis, D. & Eveleigh, D. 2004. Improved Catalytic Efficiency and Active Site Modification of 1, 4- -d-Glucan Glucohydrolase A from Thermotoga neapolitana by Directed Evolution. Journal of Biological Chemistry, 279, 11495. Merino, S. T. & Cherry, J. 2007. Progress and challenges in enzyme development for biomass utilization. Advances in Biochemical Engineering and Biotechnology, 108, 95–120. Mielenz, J. R. 2001. Ethanol production from biomass: Technology and commercialization status. Current Opinion in Microbiology, 4, 324–329. Miyamoto, K. 1997. Renewable biological systems for alternative sustainable energy production (FAO Agricultural Services Bulletin-128). Molecular biology, 4, 3. Moreira, N. 2005. Growing Expectations. Science News, 168, 218–220. Mosier, N. S., Wilker, J. J. & Ladisch, M. R. 2004. Rapid chromatography for evaluating adsorption characteristics of cellulase-binding domain mimetics. Biotechnology and Bioengineering, 86, 756–764. Navas, J. & B´eguin, P. 1992. Site-directed mutagenesis of conserved residues of Clostridium thermocellum endoglucanase CelC. Biochemical and Biophysical Research Communications(Print), 189, 807–812. Nikolov, T., Bakalova, N., Petrova, S., Benadova, R., Spasov, S. & Kolev, D. 2000. An effective method for bioconversion of delignified waste-cellulose fibers from the paper industry with a cellulase complex. Bioresource Technology, 71, 1–4. Olofsson, K., Bertilsson, M. & Lid´en, G. 2008. A short review on SSF–an interesting process option for ethanol production from lignocellulosic feedstocks. Biotechnology for Biofuels, 1, 7. Olofsson, K., Rudolf, A. & Lid´en, G. 2008. Designing simultaneous saccharification and fermentation for improved xylose conversion by a recombinant strain of Saccharomyces cerevisiae. Journal of Biotechnology, 134, 112–120. Oraby, H., Venkatesh, B., Dale, B., Ahmad, R., Ransom, C., Oehmke, J. & Sticklen, M. 2007. Enhanced conversion of plant biomass into glucose using transgenic rice-produced endoglucanase for cellulosic ethanol. Transgenic Research, 16, 739–749. Park, Y., Tominaga, R., Sugiyama, J., Furuta, Y., Tanimoto, E., Samejima, M., Sakai, F. & Hayashi, T. 2003. Enhancement of growth by expression of poplar cellulase in Arabidopsis thaliana. Plant Journal, 33, 1099–1106. Rabinovich, M. L., Mel’nik, M. S. & Bolobova, A. V. 2002. [Cellulases from microorganisms]. Prikl Biokhim Mikrobiol, 38, 355–373.

P1: SFK/UKS P2: SFK c10 BLBS075-Hood December 15, 2010

20:55

Trim: 244mm×172mm

10 Enzyme Production Systems for Biomass Conversion

251

Ragauskas, A., Williams, C., Davison, B., Britovsek, G., Cairney, J., Eckert, C., Frederick Jr, W., Hallett, J., Leak, D. & Liotta, C. 2006. The path forward for biofuels and biomaterials. Science, 311, 484. Ramamurthy, V., Thacker, S. P. & Kothari, R. M. 1992. Optimized protocol for the pilot-scale preparation of fungal cellulase. Journal of Industrial Microbiology, 9, 121–125. Ransom, C., Balan, V., Biswas, G., Dale, B., Crockett, E., Sticklen, M. 2007. Heterologous acidothermus cellulolyticus 1, 4-beta-endoglucanase E1 produced within the corn biomass converts corn stover into glucose. Applied Biochemistry and Biotechnology, 137, 207–219. Reddy, M., Chen, F., Shadle, G., Jackson, L., Aljoe, H. & Dixon, R. 2005. Targeted downregulation of cytochrome P450 enzymes for forage quality improvement in alfalfa (Medicago sativa L.). Proceedings of the National Academy of Sciences, 102, 16573. Rondon, M., August, P., Bettermann, A., Brady, S., Grossman, T., Liles, M., Loiacono, K., Lynch, B., Macneil, I. & Minor, C. 2000. Cloning the soil metagenome: A strategy for accessing the genetic and functional diversity of uncultured microorganisms. Applied and Environmental Microbiology, 66, 2541. Rosgaard, L., Pedersen, S., Langston, J., Akerhielm, D., Cherry, J. & Meyer, A. 2007. Evaluation of minimal Trichoderma reesei cellulase mixtures on differently pretreated barley straw substrates. Biotechnology Progress, 23, 1270–1276. Rudolf, A., Baudel, H., Zacchi, G., Hahn-Hagerdal, B. & Lid´en, G. 2008. Simultaneous saccharification and fermentation of steam-pretreated bagasse using Saccharomyces cerevisiae TMB3400 and Pichia stipitis CBS6054. Biotechnology and Bioengineering, 99, 783–790. Sainz, M. 2009. Commercial cellulosic ethanol: The role of plant-expressed enzymes. In Vitro Cellular & Developmental Biology-Plant, 45, 314–329. Schaffner, D. W. & Toledo, R. T. 1991. Cellulase production by Trichoderma reesei when cultured on xylose-based media supplemented with sorbose. Biotechnology and Bioengineering, 37, 12–16. Shallom, D. & Shoham, Y. 2003. Microbial hemicellulases. Current Opinion in Microbiology, 6, 219–228. Singhania, R. R., Sukumaran, R. K. & Pandey, A. 2007. Improved cellulase production by Trichoderma reesei RUT C30 under SSF through process optimization. Applied Biochemistry and Biotechnology, 142, 60–70. Spear, L., Gallagher, J., McHale, L. & McHale, A. P. 1993. Production of cellulase and beta-glucosidase activities following growth of Streptomyces hygroscopicus on cellulose containing media. Biotechnology Letters, 15, 1265–1268. Sticklen, M. 2006. Plant genetic engineering to improve biomass characteristics for biofuels. Current Opinion in Biotechnology, 17, 315–319. Sticklen, M. 2008a. Altering regulation of maize lignin biosynthesis enzymes via RNAi technology. US Patent App. 20,080/213,871, 2007. Sticklen, M. B. 2008b. Plant genetic engineering for biofuel production: Towards affordable cellulosic ethanol. Nature Reviews Genetics, 9, 433–443. Sun, Y. & Cheng, J. 2002. Hydrolysis of lignocellulosic materials for ethanol production: A review. Bioresource Technology, 83, 1–11. Taherzadeh, M. & Karimi, K. 2007. Enzyme-based hydrolysis processes for ethanol from lignocellulosic materials: A review. BioResources, 2, 707–738. Teather, R. & Wood, P. 1982. Use of Congo red–polysaccharide interactions in enumeration and characterization of cellulolytic bacteria from the bovine rumen. Applied and Environmental Microbiology, 43, 777–780.

P1: SFK/UKS P2: SFK c10 BLBS075-Hood December 15, 2010

252

20:55

Trim: 244mm×172mm

Plant Biomass Conversion

Tholudur, A., Ramirez, W. F. & Mcmillan, J. D. 1999. Mathematical modeling and optimization of cellulase protein production using Trichoderma reesei RL-P37. Biotechnology and Bioengineering, 66, 1–16. Tolonen, A., Chilaka, A., Church, G. & Church, G. 2009. Targeted gene inactivation in Clostridium phytofermentans shows that cellulose degradation requires the family 9 hydrolase Cphy3367. Molecular Microbiology, 74, 1300–1313. Torney, F., Moeller, L., Scarpa, A. & Wang, K. 2007. Genetic engineering approaches to improve bioethanol production from maize. Current Opinion in Biotechnology, 18, 193–199. Toyama, H. & Toyama, N. 2001. The effect of additional autopolyploidization in a slow growing cellulase hyperproducer of Trichoderma. Applied Biochemistry and Biotechnology, 91/93, 787–790. Toyama, H., Yano, M., Gisushi, A., Hotta, T. & Toyama, N. 2003. Active nuclear shuffling system using a swollen conidium of Trichoderma reesei. Applied Biochemistry and Biotechnology, 108, 821–824. Twyman, R. M., Stoger, E., Schillberg, S., Christou, P. & Fischer, R. 2003. Molecular farming in plants: Host systems and expression technology. Trends in Biotechnology, 21, 570–578. Voutilainen, S., Murray, P., Tuohy, M. & Koivula, A. 2010. Expression of Talaromyces emersonii cellobiohydrolase Cel7A in Saccharomyces cerevisiae and rational mutagenesis to improve its thermostability and activity. Protein Engineering Design and Selection, 23, 69–79. Woodard, S., Mayor, J., Bailey, M., Barker, D., Love, R., Lane, J., Delaney, D., McComasWagner, J., Mallubhotla, H. & Hood, E. 2003. Maize (Zea mays)-derived bovine trypsin: Characterization of the first large-scale, commercial protein product from transgenic plants. Biotechnology and Applied Biochemistry, 38, 123–130. Wyman, C. E., Dale, B. E., Elander, R. T., Holtzapple, M., Ladisch, M. R. & Lee, Y. Y. 2005. Comparative sugar recovery data from laboratory scale application of leading pretreatment technologies to corn stover. Bioresource Technology, 96, 2026–2032. Xia, L. M. & Shen, X. L. 2004. High-yield cellulase production by Trichoderma reesei ZU-02 on corn cob residue. Bioresource Technology, 91, 259–262. Xue, G., Patel, M., Johnson, J., Smyth, D. & Vickers, C. 2003. Selectable marker-free transgenic barley producing a high level of cellulase (1, 4-ß-glucanase) in developing grains. Plant Cell Reports, 21, 1088–1094. Yang, B. & Wyman, C. 2004. Effect of xylan and lignin removal by batch and flowthrough pretreatment on the enzymatic digestibility of corn stover cellulose. Biotechnology and Bioengineering, 86, 88–98. Zhang, M., Eddy, C., Deanda, K., Finkelstein, M. & Picataggio, S. 1995. Metabolic engineering of a pentose metabolism pathway in ethanologenic Zymomonas mobilis. Science, 267, 240. Zhang, P., Himmel, M. & Mielenz, J. 2006. Outlook for cellulase improvement: Screening and selection strategies. Biotechnology Advances, 24, 452–481. Zhang, Y.-H. P., Ding, S.-Y., Mielenz, J. R., Cui, J.-B., Elander, R. T., Laser, M., Himmel, M. E., Mcmillan, J. R. & Lynd, L. R. 2007. Fractionating recalcitrant lignocellulose at modest reaction conditions. Biotechnology and Bioengineering, 97, 214–223. Zhang, Y. H. P. & Lynd, L. R. 2004. Toward an aggregated understanding of enzymatic hydrolysis of cellulose: Noncomplexed cellulase systems. Biotechnology and Bioengineering, 88, 797–824. Zhou, J., Wang, Y.-H., Chu, J., Luo, L.-Z., Zhuang, Y.-P. & Zhang, S.-L. 2009. Optimization of cellulase mixture for efficient hydrolysis of steam-exploded corn stover by statistically designed experiments. Bioresource Technology, 100, 819–825.

P1: SFK/UKS P2: SFK c10 BLBS075-Hood December 15, 2010

20:55

Trim: 244mm×172mm

10 Enzyme Production Systems for Biomass Conversion

253

Zhu, J., Wang, G., Pan, X. & Gleisner, R. 2009. The status of and key barriers in lignocellulosic ethanol production: A technological perspective. International Conference on Biomass Energy Technologies, Guangzhou, China. Zhuang, J., Marchant, M. A., Nokes, S. E. & Strobel, H. J. 2007. Economic analysis of cellulase production methods for bio-ethanol. Applied Engineering in Agriculture, 23, 679– 687.

P1: SFK/UKS P2: SFK c10 BLBS075-Hood December 15, 2010

20:55

Trim: 244mm×172mm

P1: SFK/UKS P2: SFK c11 BLBS075-Hood December 15, 2010

20:56

Trim: 244mm×172mm

Chapter 11

Fermentation-Based Biofuels Randy Kramer and Helene Belanger

11.1 Introduction First-generation biofuels produced from food crops such as corn and other grains (starch feedstocks), sugarcane or sugar beet (sucrose feedstocks), and oilseeds (triglycerides feedstocks) have utilized well-known technologies to produce liquid fuel products generally compatible with existing mature markets, namely ethanol and biodiesel (fatty acid methyl ester or FAME). After much analysis, it is generally accepted that these products afford net benefits in terms of greenhouse gas (GHG) emission reduction and energy balance relative to petroleum-based fuels. Technologies are evolving to produce second-generation biofuels from widely available and sustainable cellulosic biomass rather than starch or oil-based feedstocks. Globally, policymakers continue to balance biofuel and fossil fuel production with environmental concerns, food supplies, national security, and land use—all issues influenced by special interests. Promising examples of balanced, renewable biofuels progress and models are emerging. Mankind has been manufacturing ethanol by fermentation for thousands of years, and its suitability as a modern liquid transportation fuel is well recognized. In recent years, petroleum supply disruptions and price volatility have led many governments to implement incentive programs to promote biofuel technology development, production, and consumption. Today, as oil prices continue to rise and global oil supplies are being depleted, national policies remain a key driver for displacement of finite fossil fuels with sustainable biofuels. Global ethanol production in 2008 stood at approximately 66,000 million liters (ML), as summarized in Table 11.1. On average, 73% of the ethanol produced worldwide is used as biofuel, 17% in beverages, and 10% in industrial processes (S´anchez and Cardona, 2008). New and reinforced government programs in place in the Americas, Asia, and Europe could cause demand for ethanol fuel to exceed 125 billion liters (BL) by 2020 (Balat and Balat, 2009).

Plant Biomass Conversion, First Edition. Edited by Elizabeth E. Hood, Peter Nelson and Randall Powell.
C 2011 John Wiley & Sons Inc.

255

P1: SFK/UKS P2: SFK c11 BLBS075-Hood December 15, 2010

256

20:56

Trim: 244mm×172mm

Plant Biomass Conversion

Table 11.1. Global biobased ethanol production in 2008.a Country

Volume (ml)

Feedstock

Land Use (Mha)

United States Brazil EU China Canada Thailand Columbia India Australia

34,070 24,500 2,780 1,900 900 340 300 250 100

Corn (98%), sorghum (2%) Sugarcane (80%), molasses (20%) Wheat (48%), sugarbeet (29%) Corn (70%), wheat (30%) Corn (70%), wheat (30%) Molasses (80%), cassava (20%) Sugarcane Molasses Molasses (37%), wheat residue (63%)

8.13b 3.4c 0.65d 0.97e 0.28f 0.53g 0.20h 4.9i NA

a Source: Ethanol Statistics at the Renewable Fuels Association (RFA)website at: http://www.ethanolrfa.org/industry/statistics/). b RFA public report (2008): Policy position” food vs. fuel at: http://www.ethanolrfa.org/policy/positions/foodfuel/. c DeSousa, E. L. (2008) Leading the Way in Sustainable Biofuels: The Brazilian Approach. Report for the OECD Forum on the Theme of Climate Change, Growth, Stability. Available: http://www.oecd.org/dataoecd/26/6/40758781.pdf. d Balat and Balat (2009). e Balat and Balat (2009). f Balat and Balat (2009). g Silalertruksa and Gheewala (2009). h Colombia’s Ethanol Industry Monthly Editorial from Kingsman. Available: http://www.kingsman. com/images/SampleRpts/Ethanol/MonthlyEthanolEditorial.pdf. i Goldemberg and Guardabassi (2010).

With expanding global production of first-generation biofuels, societal, economic and political viewpoints have introduced significant subjectivity into the biofuels discussion. Technology-based and popular arguments regarding key issues such as feedstock use for food versus fuel, indirect land use changes, environmental impacts, and sustainable economics have competed to influence public perception. The multifaceted debate surrounding biofuels has arguably slowed growth and negatively affected public acceptance in some instances. On the positive side, R&D and commercial focus has now shifted toward the development and commercialization of more sustainable second-generation biofuels, which can be produced from nonfood feedstocks and which achieve more substantial environmental benefits. The US Department of Energy (DOE) has defined such “Advanced Biofuels” as fuels derived from renewable biomass other than corn kernel starch, including sugar, starch (other than ethanol derived from corn kernel starch), cellulose, hemicellulose, and lignin, among other materials. This chapter will summarize the commercial status of first-generation biofuels and the status of development and demonstration programs for second-generation biofuels. The review will focus on fermentation-derived liquid fuels, predominantly intended to supplement or displace current liquid transportation fuels derived from fossil resources.

11.2 First-Generation Biofuels 11.2.1 Starch-Based Ethanol—United States Since 2000, ethanol production in the United States from starch-based crops—overwhelmingly corn—has increased more than 300% and has established the United States as the leading

P1: SFK/UKS P2: SFK c11 BLBS075-Hood December 15, 2010

20:56

Trim: 244mm×172mm

11 Fermentation-Based Biofuels

257

global biobased ethanol producer. Construction of new starch-based facilities reached a record in 2006, with no fewer than 15 new corn-ethanol biorefineries coming online. The addition of these plants, including the completion of expansion projects, added 3.97 BL of new production capacity for the year. As of December 31, 2007, there were 135 ethanol plants in the United States with a total operating capacity of 28 BL per year, along with 65 under construction and 9 expansion projects totaling another 22.7 BL of production per year. Unfortunately, US starch-based ethanol expansion, as well as industry operating margins, experienced a sharp decline in 2008 due to dramatically escalating grain prices and a subsequent drop in crude oil prices resulting from the global economic crisis. During this time, grain market speculation caused US corn prices to triple, despite evidence of adequate supplies. According to USDA statistics, field corn used to produce ethanol increased by about 1 billion bushels in 2007, while total corn production increased by 3 billion bushels. Specifically, between March 2007 and March 2008, there was a 13% increase in stored, uncommitted surplus corn, both on and off the farm. Throughout 2009, the reality of continued abundant starch-based feedstocks placed downward pressure on grain prices. During this period, crude oil prices peaked, dropped, and then recovered as global economic conditions varied widely. During 2008–2009, the US starch-based ethanol industry experienced a significant operating margin squeeze as well as dramatic feedstock cost and product pricing uncertainty, leading to widespread financial distress and significant consolidation. Government policy related to first-generation starch-based ethanol only exacerbated the industry’s difficulties. Despite the precipitous decline in starch-based ethanol growth—and similar negative pressures on vegetable-oil biodiesel—the grain-based ethanol industry in the United States represents a critical bridge to second-generation biofuels. In 2008 then-President George W. Bush stated that the United States had not built a new petroleum refinery since 1976. However, 84 new starch-based biorefineries had been built in the previous ten years, effectively replacing the need for approximately eight new averaged-size crude oil refineries (assuming 115,000 barrels per day of crude feed). Renewable feedstocks for biorefineries are replenished every year, while crude oil will only be extracted once. Despite significant public debate, there is evidence that the US starch-based ethanol industry has actually held oil prices down (Du and Hayes, 2009). According to a Merrill Lynch researcher, oil prices would be at least 15% higher, if not for the current US output of ethanol (Carey, 2008).

11.2.2

Sugar-Based Ethanol—Brazil

With the creation of the National Alcohol Fuel Program (PROALCOOL) in 1975, Brazil has proceeded to develop a substantial ethanol industry. Government subsidies and incentives totaling over $12B were phased out by the early 2000s, when ethanol became competitive with gasoline. Brazil was the largest ethanol producer until 2006 when it was surpassed by the United States, but it still remains the world’s largest exporter of ethanol. Most of the ethanol produced in Brazil is used domestically, to substitute for 40% of Brazil’s gasoline consumption, while 20% is exported to the United States, EU, and other markets (Mabee et al., 2009). While the US ethanol industry is required to produce anhydrous ethanol (200 proof ethanol), Brazil does not require complete water dehydration. This hydrous ethanol (190 proof ethanol) is used in Brazilian flex-fuel vehicles manufactured by the major automotive companies such as Fiat, Volkswagen, Ford, GM, Toyota, Honda and Peugeot. All of these vehicles are able to run on blends ranging from 25% to 100% ethanol. Flex-fuel vehicles have specially modified

P1: SFK/UKS P2: SFK c11 BLBS075-Hood December 15, 2010

258

20:56

Trim: 244mm×172mm

Plant Biomass Conversion

fuel systems that allow for the efficient combustion of hydrous ethanol and gasoline, mainly through the use of software programming in existing fuel-injected systems. This gives the consumer complete flexibility according to the respective prices of gasoline and ethanol each time a car is fueled. Most Brazilian sugarcane mills initiated operations in the 1970s and 80s and since then have grown organically to average processing capacities of approximately 1.5 million dry metric tons (DMMT) of sugarcane per year. The majority of mills are integrated factories capable of producing both raw sugar and ethanol (Gopale and Kammen, 2009). The proportion of the harvest allocated to each market is based on global demand. In 2009, there were 421 plants in operation crushing 450 DMMT of sugarcane (Goldemberg and Guardabassi, 2010). Approximately half the extracted sugar juice was used for the production of over 2.4 BL of ethanol. Brazil has also pioneered the use of cogeneration using steam produced from the residual sugarcane bagasse to generate electricity. A sugarcane mill co-located with an ethanol plant can produce its own electrical needs and sell its surplus to the grid. In 2007, 3% of Brazil’s energy requirements were produced by cogeneration. Because of the maturity of the Brazilian sugarcane industry, the growth of sugarcanebased ethanol production in Brazil is now more dependent on the geographical expansion of cultivation area than on gains in productivity. The area of cultivation expansion over the last two years is 1.3 million hectares (MHa), which is roughly equivalent to the geographical expansion that took place between 1996 and 2006 (Saprovek et al., 2009). The Brazilian government has recently proposed a new bill based on the National Agri-Ecological Zoning for Sugarcane (“ZAE Cana”) analysis. If passed, the bill will limit the land accessible for sugarcane culture to 7.5% of Brazil’s territory. There is some concern, however, that the expansion of sugarcane cultivation will take over pastureland and soy culture, and displace cattle breeding and soy production to either the Savannah (Cerrado) or Amazon areas. The bill also proposes phase-out measures to end crop burning practices by 2017. In Brazil, petroleum-based fuels are now considered the alternative to biofuels. The country continues to pursue a balanced approach between the development of fossil fuels and biofuels, with consideration for environmental impact, feedstock production, food supply, land use, and national security.

11.2.3 Biodiesel Although not produced by fermentation technology, biodiesel is a significant first-generation commercial biofuel and will be briefly profiled here. Biodiesel refers to fatty acid alkyl esters (methyl, propyl, or ethyl) derived from plant oil or animal fat. FAME produced by base catalyzed transesterification of triglycerides with methanol is the major biodiesel available commercially today. In commercial facilities, biodiesel processes yield a volumetric product ratio to virgin oil input of about 1:1 (Thamsiriroj and Murphy, 2009). The properties of biodiesel vary according to the fatty acid profile of the feedstock. In 2008, global biodiesel production exceeded 14,000 ML with over 8,000 ML produced in the European Union (EU) countries (Lichts, 2009; http://www.ebb-eu.org/stats.php), as shown in Table 11.2. After initial implementation in the early 1990s, EU biodiesel production and consumption expanded rapidly in response to increasing crude oil price and biodiesel feedstock availability. This expansion has slowed in recent years but biodiesel remains the principal biofuel consumed in the EU with demand reaching 9,000 ML in 2008, corresponding to the displacement of 3.5% of petroleum diesel (Flach, 2009).

P1: SFK/UKS P2: SFK c11 BLBS075-Hood December 15, 2010

20:56

Trim: 244mm×172mm

11 Fermentation-Based Biofuels

259

Table 11.2. Global biodiesel production in 2008. a Country

Volume (ML)

Feedstock

EU

8.684

Rapeseed (72%), soybean (13%), tallow-recycled (8%), palm (4%), sunflower (3%)

United States

2.583

Soybean (60%), tallow (20%), palm (20%)

Argentina

840

Soybean

Brazil

863

Soybean (85%), other plant oils (10%), tallow (5%)

Malaysia

478

Palm oil

Thailand

448

Palm oil

Australia

260

Tallow-recycled vegetable oil

Canada

126

Tallow (97%), rapeseed (3%)

Indonesia

102

Palm oil

China a Source:

60

Recycled vegetable oil

Respective Gain report, 2009, USDA Foreign Agricultural Service.

Europe is the largest global producer of rapeseed and sunflower oil (9.2 and 6.9 BL respectively in 2008/09) and the fifth largest producer of soybean oil (2.3 BL in 2008/09) (USDA Foreign Argicultural Service, 2010). To meet the EU quality standard, biodiesel in Europe is manufactured from various raw materials and blends, dominated by rapeseed oil. In 2008/2009, about 70% of the annual European production of rapeseed oil was dedicated to the manufacture of biodiesel, followed by 45% of the soybean oil and 10% of the sunflower oil produced, affording approximately 7,700 ML of biodiesel (Dobresecu et al., 2009). An additional 1,000 ML of biodiesel was produced in the EU from imported palm oil and regional supplies of tallow and recycled oil. Remaining European demand was met by importing around 2,800 ML of biodiesel, mostly from the United States. Half of the European biodiesel is produced in Germany and most of the remaining production capacity is located in France, Italy, and the Benelux. Sunflower oil, tallow, and recycled oil are the primary biodiesel feedstocks in the southern European countries, while rapeseed is cultivated in northern and central Europe. Soybeans, grown primarily for high protein content, are the second largest global source of plant oil, after oil palm, and have been increasingly used as a feedstock for biodiesel. The major global producers of soybean oil in 2008/09(USDA Foreign Argicultural Service, 2010) are the United States (9.2 BL), China (7.8 BL), Argentina (6.7 BL), and Brazil (6.4 BL). In the United States, over 15% of soybean oil produced in 2008 accounted for the manufacture of 60% of the national biodiesel production. The remaining 40% of the US biodiesel output used tallow and recycled oil feedstocks, as well as imported palm or seed oil, to reach a total production of 2,600 ML. Recently, the growth of the US biodiesel industry has been curtailed by unfavorable economics due to uncertainty regarding federal incentives and renewable mandates, volatility in commodity markets, and imposition of EU tariffs on US biodiesel imports. As a result, the production of US biodiesel dropped during 2009, despite the fact that the US has a significant untapped domestic market. Brazil’s biodiesel is mostly produced for the domestic heavy transportation vehicle market, whereas most of Argentina’s biodiesel production is destined for the export market. The three largest palm oil producing countries are Indonesia (21.2 BL), Malaysia (18.8 BL), and Thailand (1.3 BL) (USDA Foreign Argicultural Service, 2010). Most of the palm

P1: SFK/UKS P2: SFK c11 BLBS075-Hood December 15, 2010

260

20:56

Trim: 244mm×172mm

Plant Biomass Conversion

oil is used in general food consumption but a steadily increasing share is also used as an oleochemical feedstock and for biodiesel production. The introduction of mandatory blends of biodiesel with petroleum diesel have stimulated biofuel production in these regions, particularly in Thailand where 25% of the 2008/09 palm oil produced was used locally for biodiesel manufacture(Prasertsri and Kunasirirat, 2009). In 2008, only 2% of the Indonesian and Malaysian annual palm oil production was domestically converted to 100 and 484 ML of biodiesel, respectively (Bromokusumo, 2009; Hoh, 2009). The majority of the biodiesel produced in these countries is exported, with only 10%–15% of the annual production retained for domestic consumption. A very small volume of biodiesel produced outside the EU is currently sourced from rapeseed. The major producers of rapeseed oil after Europe are China (5.0 BL), India (2.2 BL), and Canada (1.9 BL) (USDA Foreign Argicultural Service, 2010). In these countries, tallow and recycled oil are the preferred feedstocks for biodiesel but increased used of rapeseed oil is anticipated for future expansion of biodiesel production capacity. Russia and the Ukraine are currently expanding their rapeseed production destined for export to the EU. Future sources of biodiesel feedstock oils are non-edible oilseeds like Jatropha curcas, high erucic mustard, Indian beech (Pongamia pinnata), green seed canola, and micro algae.

11.3 Policy and Biofuel Implementation Status 11.3.1 North America The starch-based ethanol industry in the United States was previously described. According to a national poll conducted in late 2007 by the Mellman Group, 74% of Americans say they want increased production and use of domestic renewable fuels like ethanol. By even greater margins (87%), Americans said they believe that the federal government should actively support the development of a robust renewable fuels industry. US federal support of the biofuels industry includes diverse programs that incentivize feedstock providers, biofuels producers, biofuels retailers, oil industry biofuels, blenders, and the general public who purchase flexible fuel vehicles. Following is a summary of some of the relevant US federal programs. The summary does not include a multitude of state-funded programs that also incentivize the use of biofuels. A registered small ethanol producer may be eligible for a Small Ethanol Producer Tax Credit in the amount of $0.10 per gallon of ethanol. A small producer is defined as one that has, at all times during the tax year, not more than 60 million gallons (MG) of productive capacity of any type of alcohol. The incentive, which expires December 31, 2010, applies only to the first 15 MG of ethanol produced in a tax year and is allowed as a credit against the producer’s income tax liability. The US Federal Blender’s Credit is designed as an incentive for the oil industry to use biofuels. In effect, the economics of ethanol production are supported by the partial exemption of ethanol-blended fuels from the federal excise tax on motor fuels, making ethanol pricecompetitive in fuel markets. This credit is also set to expire at the end of 2010. The Alternative Fuel Infrastructure Tax Credit is available to support the cost of installing alternative fueling equipment placed into service after December 31, 2005. Qualified alternative fuels are natural gas, liquefied petroleum gas, hydrogen, electricity, gasoline blends containing 85% ethanol (E85), or diesel fuel blends containing a minimum of 20% biodiesel (B20). The Energy Policy Act of 2005 included several important provisions favorable for biofuels. The Energy Act eliminated the Reformulated Gasoline Program of the Clean Air Act of 1990

P1: SFK/UKS P2: SFK c11 BLBS075-Hood December 15, 2010

20:56

Trim: 244mm×172mm

11 Fermentation-Based Biofuels

261

and established the Renewable Fuels Standard (RFS) that mandates a minimum of 7.5 billion gallons (BG) of renewable fuels be used by 2012. The Energy Independence and Security Act (EISA) became law in December 2007. The law incorporates a more aggressive RFS with almost 60% of the new requirements directed toward a combination of advanced biofuels and cellulosic ethanol. The EISA targets biofuels use of 20.5 BG by 2015 and 36 BG by 2022. There is evidence to show that these targets may not be met in 2010. The EISA includes three categories of ethanol feedstock: (1) renewable biofuel (exclusively derived from corn), (2) advanced biofuel (derived from other-than-corn, including cellulosic ethanol, and which achieves a 50% GHG emission reduction), and (3) cellulosic biofuel (derived from cellulose, hemicellulose, or lignin, which achieves a 60% reduction in GHG). The Food, Conservation, and Energy Act (2008 Farm Bill) became law in May 2008. Provisions in the five-year farm bill package for the development of biofuels include $320M in mandatory funding for loan guarantees for commercial scale biorefineries; $35M in mandatory funding for grants to support repowering existing biorefineries with biomass energy systems; $300M in mandatory funding for payments to support the production of advanced biofuels, including biodiesel and cellulosic biofuels; and $1M/year for competitive grants to educate the public about effective biodiesel use and the benefits of biofuels. One provision of the 2008 Farm Bill called for the creation of a Biomass Crop Assistance Program (BCAP) which has already proved to be widely successful. In 2009 this program began matching payments for biomass feedstocks delivered to certified biorefineries. Some Latin American countries achieve very high annual yields of sugarcane through efficient land use (100–130 MT/Ha) (Miron, 2009). This natural advantage has resulted in expansion of the sugarcane industry for ethanol production. These regions have already received significant private investment in the biofuel area, and governments are attempting to promote ethanol production by passing laws and offering tax incentives. In Mexico, 25 to 30 sugar mills have an ethanol production capacity totaling 170 ML per year, but a smaller number are currently in operation. Ethanol production in 2007 was about 70 ML. Besides sugarcane, other crops grown in Mexico that could be used as feedstock for ethanol production include sorghum, corn, wheat, cassava, and sugar beet. In Guatemala sugarcane plantations span 215,000 Ha, with a potential for 350,000 Ha of additional planting. The sugarcane molasses that was originally exported is now used mainly for ethanol production. Despite the absence of legislation to promote national biofuel consumption, the ethanol sector in Guatemala is still expanding due to increasing yields and production capacity. Five of Guatemala’s 14 sugar mills have produced a total of 240 ML ethanol in 2008 for export to Europe and the US (Tay, 2009). El Salvador has already drafted legislation to continue developing a local ethanol market and is beginning to invest in ethanol production. An old distillery exporting its annual ethanol production of 60 ML is being upgraded to double its capacity. Central America and Caribbean countries also serve as entry points for Brazilian ethanol companies seeking to perform off-shore ethanol dehydration in order to avoid the $0.54 per gallon US import tariff (Bevill, 2009). These countries do not currently produce ethanol, but house dehydration plants that can transform hydrous ethanol imported from Brazil into anhydrous ethanol for export to the US market. This operation takes advantage of the Free Trade Agreement between both the United States and Central America, with the Dominican Republic (CAFTA-DR). Under the Caribbean Basin Initiative (CBI), US import tariff exemptions apply for ethanol made from local feedstocks with a quota limit equal to 7% of total US ethanol consumption for ethanol produced from nonregional feedstocks. This agreement has been used mostly by El Salvador, Costa Rica, and Guatemala.

P1: SFK/UKS P2: SFK c11 BLBS075-Hood December 15, 2010

262

20:56

Trim: 244mm×172mm

Plant Biomass Conversion

11.3.2 South America The mature sugarcane-based ethanol industry in Brazil was described previously. Colombia, Peru, and Paraguay, like Brazil, have increased their ethanol production level by utilizing sugar juice as a raw material instead of only relying on molasses by-products. The production of ethanol in Colombia only began in 2005 and has reached an annual production of approximately 300 ML in 2008. One-fifth of the cultivated area in Colombia is dedicated to ethanol production. Colombia’s sugar mills are currently adding ethanol production capabilities to existing facilities and are also developing new dual raw sugar and ethanol production facilities. With 1 MHa of suitable land available for sugarcane plantations, Colombia hopes to reach an annual production close to 1 BL (Toasa, 2009). It is expected that around $500 M will be invested in the Colombian ethanol sector over the next 4–5 years (Alexander and Torres, 2006). A mandate for an E10 blend is enforced in all Colombian cities with population greater than 500,000. In Peru, new ethanol production capabilities are being installed, with the first plant producing 330,000 L of ethanol per day coming on line in 2009. Despite limited government intervention in this sector, significant private investment has been made to increase ethanol production capacity in Peru. About 200,000 Ha of sugarcane is under development specifically for ethanol production (Elbehri et al., 2009). Paraguay is the world’s largest exporter of organic sugar. In 2008, 100,000 Ha were planted with sugarcane, and official studies indicate that the country has the potential to expand this to 450,000 Ha. In 2008, 90 ML of ethanol was produced in Paraguay from 450,000 dry MT of sugarcane and 30,000 dry MT of starch grains (Joseph, 2008). Sixty percent of the sugarcane was used to produce conventional and organic sugar, and the balance was used for ethanol production. A few sugar mills produced ethanol from 230 dry MT molasses. The grain feedstocks were processed to ethanol in alternation with sugarcane. One sugar mill invested in an ethanol plant that can use primarily sorghum and corn. Cassava, which is grown on 300,000 Ha, is another potential feedstock for ethanol production in Paraguay. In 2005, the Paraguayan Congress passed Biofuels Promotion legislation, establishing a 20% blending mandate for ethanol, as long as there is sufficient local supply. In 2008, the government passed a bill to eliminate import duties on Flex-fuel and E85 new and used cars. In Uruguay, plans are in place for the production of approximately 5.6 ML of ethanol from molasses, sugarcane, and sweet sorghum.

11.3.3 Europe It is not expected that the EU will achieve its target of 5.75% biobased road transport fuel by 2010. On the basis of production costs, biofuels in Europe are currently not competitive with diesel or gasoline in the EU, and production and consumption expansion largely depends on mandates and incentives. In 2008, 2.7 BL of ethanol was produced in Europe from 5,000 MT of wheat, 2,500 MT of corn, 500 MT of barley and rye and 1,500 MT of sugar beet molasses (Flach, 2009). In 2008, ethanol production expanded more than forecasted as new production facilities came on line and the availability of sugar beet-derived raw materials increased. The use of corn is expected to increase in Spain and Central Europe, while wheat is expected to remain the major feedstock in Northwestern Europe. Expansion of ethanol production from sugar beet derivatives is expected in France, Germany, and Belgium. Feedstocks required for the anticipated production of 3.0 MT of ethanol in 2010 are estimated at about 8 MT of cereals and a volume of sugar beet derivatives equivalent to 1.5 MT of refined sugar.

P1: SFK/UKS P2: SFK c11 BLBS075-Hood December 15, 2010

20:56

Trim: 244mm×172mm

11 Fermentation-Based Biofuels

263

Several European groups have initiated pilot or demonstration lignocellulosic ethanol facilities with modest capacities. These will be profiled in Section 11.4.1 of this review. The EU initiative Futurol was formed in 2008 by top-tier French research organizations, manufacturers, and financial institutions with a $118M budget. The 8 year development and marketing project will focus on integrated production processes for second-generation ethanol from lignocellulosic biomass. Russia and Africa, where current biofuel production is minimal, represent two regions with great potential for production and supply of ethanol to European countries.

11.3.4

Asia

India is the second largest sugarcane producing country in the world. In 2008, 4.36 MHa of land were under cultivation, with a total output of 280 MMT of sugarcane. Low productivity associated with inadequate selection of sugarcane varieties and poor processing methods have made land expansion the predominant factor for increasing sugarcane production in India since the mid-1990s. The single source of raw material for ethanol production, cane molasses, is not sufficient to meet 5% of domestic fuel consumption, which has increased continuously as a result of the country’s rapid economic growth. Mandatory blending of gasoline with 5% ethanol was initiated in India in 2002 and suspended in 2004 due to difficulties in maintaining an uninterrupted ethanol supply (Pohit et al., 2009). The new Indian National Policy on biofuels established a National Biofuel Fund in 2009, with a focus on planting, processing, and production technologies, including second-generation biofuels. The gasoline blend mandate was recently reset at 20% ethanol by 2017. Alternate feedstock crops such as sweet sorghum and tropical sugar beet are also being explored as raw materials for ethanol production. However, because of land requirements for these crops, lignocellulosic feedstocks are regarded as the most promising source for ethanol production in India. In 2009, a memorandum of understanding (MOU) was signed between the Indian and US governments for cooperation on the production and distribution of biofuels in a sustainable and environmentally friendly manner. With an initial investment of $4M, the state-owned Indian Oil Corporation has entered into an agreement with the US National Renewable Energy Laboratory (NREL) for a pilot project to produce second-generation cellulosic ethanol. In partnership with the Indian Council for Scientific and Industrial Research, Godavari Sugar Mills Ltd. is establishing a pilot plant at Sameerwadi in Karnataka for the processing of sugarcane cellulosic bagasse to ethanol. Praj Industries, an India-based global leader in biofuels technology, inaugurated its initial cellulosic ethanol pilot plant in 2009, with a daily processing capacity of 2 MT, at its R&D Innovation Centre near Pune. Mission NewEnergy Ltd., a biodiesel producer based in Perth, Australia, has been operating a bagasse-based ethanol plant in northern India since 2008. China is the world’s largest producer of rice and wheat, and ranks second in corn production. With a variety of land types and climate conditions, China has excellent potential for the culture of different biofuel feedstock crops. Of the 1.9 BL of ethanol produced in 2008, half was made from grain, mainly corn, but also sorghum, wheat, and rice. Cane molasses accounted for 32% of production, with root crops (cassava, sweet potatoes) at 11% and beet molasses at 3% (Wang et al., 2009). Biofuel projects based on grain have been restricted and development of “nonfood ethanol” has been supported by the Chinese government. China, a net sugar importer, has promoted the production of ethanol from sugarcane only in specific regions. As a result, the productivity of sugarcane increased faster than in India due to innovative changes in the industry structure as well as public investments in breeding programs and irrigation

P1: SFK/UKS P2: SFK c11 BLBS075-Hood December 15, 2010

264

20:56

Trim: 244mm×172mm

Plant Biomass Conversion

infrastructure. The culture and use of cassava as a raw material for ethanol production is also expanding. In China, a mandate for gasoline to contain 10% ethanol for motor vehicles has been enforced in ten provinces. China’s ethanol exports have dropped substantially since the government removed the ethanol export rebate in 2007. As in India, the rapid economic development of China has dramatically increased fuel consumption. Therefore, there is also great interest in exploiting the vast cellulosic agricultural waste resources for ethanol production. A cellulosic pilot plant located in Fengxian, a suburb of Shanghai, was developed with government support. Another pilot plant was established by China Resources Alcohol Corporation in Zhaodong Heilongjiang for the processing of wheat straw to ethanol. A larger demonstration facility to process dry straw and corn cobs to ethanol is being established by Henan Tianguan Fuel (Fang et al., 2009). The Thailand renewable energy policy started in a noncompulsory manner, promoting the use of 15% ethanol in gasoline. In 2008, a 20% ethanol blend was introduced and E85 gasohol was launched. Direct fermentation of cane juice to ethanol is marginal in Thailand, and sugarcane molasses is the dominant source of raw material for ethanol production. Half of the annual sugarcane molasses by-product (about 2,000 MT) is used by eight sugar mills with a combined ethanol capacity of 1.4 ML/day (Prasertsri and Kunasirirat, 2009). Only one ethanol plant in Thailand is based on cassava, though the majority of facilities under construction will be cassava based. Increased land area has been dedicated to cassava at the expense of corn (Silalertruksa and Gheewala, 2009). Currently, all ethanol plants are running at 60% of total capacity as Thailand currently generates an ethanol surplus that is exported (Prasertsri and Kunasirirat, 2009). Besides the expansion of cassava cultivation, the use of bagasse cellulosic feedstock is also being explored as a means to increase national ethanol production. Thailandbased sugar producer, Thai Roong Ruang Group, is currently developing a pilot-scale bagasse ethanol facility in Saraburi, Thailand, with financial support from Japanese New Energy and Industrial Technology Development Organization. Pakistan is a large producer of raw sugar from cane with 1 MHa dedicated to sugarcane cultivation. No other feedstock in Pakistan can match the economics of cane molasses for ethanol production. Sugar producers in Pakistan have started to invest in the installation of molasses-based ethanol production units after the 2006 Presidential directives to initiate the blending of 5%–10% ethanol in gasoline in pilot retail outlets (Arijan et al., 2009). However, in recent years, ethanol production in Pakistan has been greatly affected by rapid increases in molasses prices and limitations on molasses availability. Today, Pakistan’s sugar industry operates 21 molasses-based distillery units, with an annual ethanol production capacity of more than 500,000 L. E10 blends are presently being sold only in a limited number of outlets across the country, and exports to Europe, Asia, and Middle East countries have risen steadily over the past 5 years. Declines in molasses exports will increase molasses availability for ethanol production. Trial production of sugar beet ethanol in Pakistan has proven successful, but the industry is reluctant to expand due to technical and administrative challenges. Converting rice and wheat straw to ethanol has the potential to enhance the domestic energy sector in Pakistan; however, there are no direct grants or tax incentives for the production or marketing of ethanol. Rising energy prices have increased biofuel interest in Australia but no clear biofuel policy has yet been put in place by the Australian government (Darby, 2009). Mandatory ethanol blending at the state level was set at 5% by 2010 in Victoria and Western Australia, at 10% by 2011 in New South Wales, and at 5% by 2011 in Queensland. The states of Tasmania, South Australia, and the Northern Territory remain uncommitted. In Australia, the principal raw materials for ethanol production are wheat, waste wheat starch, and cane molasses. About 140 ML of ethanol was produced in Australia in 2008. Recently, plants have not operated at

P1: SFK/UKS P2: SFK c11 BLBS075-Hood December 15, 2010

20:56

Trim: 244mm×172mm

11 Fermentation-Based Biofuels

265

Table 11.3. Demonstration-stage lignocellulosic ethanol projects.a

Venture

Feedstock

Capacity (L/year)

Operation Start year

Location

Logen Verenium New ethanol (Inbicon) AE Biofuels, Inc. Abengoa Bioenergy Co Poet LLC Godavari Sugar Mills, Ltd. BioEthanol Japan Co. Etek Etholteknik AB KL Energy Group BlueFire Ethanol Mascoma

Wheat straw Bagasse Wheat straw Corn stover Wheat/barley straw Corn cobs Bagasse Wood waste Wood waste Wood waste Wood waste Wood/Ag waste

4,000,000 5,300,000 5,000,000 570,000 75,000 75,000 Unknown 1,400,000 150,000 4,000,000 83,000 380,000

2004 2008 2009 2008 2007 2008 2008 2007 2004 2008 2002 2008

CA, ON US, LA DK US, MT US, NE US, IA IN JP SW US, WY JP US, NY

a Compiled

from various sources of available public information.

full capacity, some have closed and many proposed biofuel projects have either been shelved or abandoned. The focus of the new Japanese biofuels policy has shifted toward potential cellulosic feedstocks, rather than increasing actual limited agricultural production (Iijima, 2009). The target of displacing 10% of the domestic consumption of fossil fuels by around 2030 is based on the abundant unused biomass resources in Japan, mainly crop and forestry residues. Japan was one of the first countries to showcase a demonstration plant able to process wood for ethanol production.

11.4 Second-Generation Biofuels 11.4.1

Cellulosic Ethanol

Fermentation technologies that can utilize lignocellulosic feedstocks will maximize biofuel production from residues available from existing cultivated land and allow conversion of new sugar resources available from woody biomass and municipal solid waste. In addition, dedicated cellulosic energy crops produced on abandoned and marginal lands will also increase raw material availability for biofuels production without impacting global food supply. Fermentation-based production of second-generation ethanol derived from lignocellulosic feedstocks has not yet reached commercial application. Nevertheless, significant progress has been made in the last decade by many technology research firms who are commissioning demonstration and pilot-scale facilities around the world, as summarized in Table 11.3. Of the companies using lignocellulosic agricultural waste as the primary feedstock, Iogen Corporation is at the most advanced stage, with their cellulosic ethanol made available at Shell service stations in 2009. The company has operated a demonstration wheat straw processing plant in Ottawa, Canada, since 2004 with a capacity of 4 ML ethanol per annum. In 2007, Iogen was approved for an $80M grant from the US DOE, for the establishment of a $350M plant in Idaho, US. The Idaho project was suspended in 2008, after the Canadian government

P1: SFK/UKS P2: SFK c11 BLBS075-Hood December 15, 2010

266

20:56

Trim: 244mm×172mm

Plant Biomass Conversion

offered a $500M Canadian dollar incentive package for demonstration scale projects. In 2009, Iogen signed a letter of intent with the government of Saskatchewan and Domtar Corp. to develop a $250M Canadian dollar first-of-kind cellulosic ethanol plant with a capacity of 87 ML per year at the Prince Albert Pulp Mill in Saskatchewan. Royal Dutch Shell holds a significant investment in Iogen’s technology development, along with other private groups that will accelerate cellulosic ethanol to market. Verenium Corporation, formed in 2007 through the merger of Cambridge, MA based Celunol and San Diego, CA based Diversa, achieved demonstration scale in 2008 at a facility in Jennings, LA, with a contribution from the US DOE. British Petroleum (BP) in 2008 acquired a 50% stake in the Verenium Corporation technology and has established a new joint venture called Vercipia Biofuels. Vercipia plans to build a $300M commercial scale plant in Florida in 2010 with a capacity of 136 ML ethanol per year, and has plans for further sites including the Gulf Coast. The company has also leased land to grow sugarcane and has signed longterm agreements with Lykes Brothers to provide biomass for the Florida project, which has also brought in $7M in grants from Florida’s “Farm to Fuel” Program. The US DOE is also supporting the project through its Loan Guarantee Program. Inbicon recently inaugurated a demonstration scale facility for the processing of wheat straw to ethanol via cellulose enzymatic hydrolysis, affording animal feed from C5 molasses and heat from lignin. The new facility is integrated with DONG Energy’s Asnaes Power Station in Kalundborg, Denmark. Inbicon has evolved from the technology and engineering group of DONG Energy, Denmark’s largest energy company, which has been converting biomass to energy since the 1990s, and established a wheat straw processing pilot plant in Denmark in 2003. In order to market and license its cellulosic pretreatment technology in North America, it has partnered with G-team (a US marketing company) and has recently signed an MOU with Otaka and Great River Energy for the development of a biomass refinery in North Dakota. This facility would be integrated with a 62 MW coal fired power plant, at Spiritwood Station. From 2007 to 2009, a number of cellulosic ethanol pilot plants have come online. Many of these newer plants are able to use multiple feedstocks, generate multiple products, and/or are integrated with existing grain ethanol facilities. AE Biofuels has the largest pilot plant of this group with a capacity of 570,000 L of ethanol per year, located in Butte, MT. The technology integrates cellulose and starch feedstocks using a patent pending ambient temperature starch hydrolysis technology, which converts the raw starch to sugar at ambient temperature. The sugars from both feedstocks can then be converted into ethanol via fermentation. Abengoa BioEnergy produces cellulosic ethanol from agricultural residues, with capacity of 75,000 L of ethanol per year at pilot plants in Nebraska, US, and Salamanca, Spain. The technology tested and validated at the pilot-scale will be implemented at a commercial facility scheduled for construction in 2010 alongside an established grain facility in Kansas, US. Abengoa Bioenergy received support from the US DOE, as well as state and local support. The Kansas biorefinery capacity will be in excess of 379 ML of ethanol per year, of which 44 ML ethanol per year will be generated from local lignocellulosic feedstocks. POET LLC, like Abengoa BioEnergy, is also pursuing an integrated starch and cellulose biorefinery model. Their aim is to integrate this technology along with their current 26 corn ethanol plants. At present, they have a 75,000 L ethanol per year pilot plant in operation in South Dakota, which utilizes corncobs as feedstock. This pilot is the precursor to a much larger 95 ML per year cellulosic ethanol facility to be located in Iowa, which will be integrated with an existing ethanol plant. Scheduled for completion in 2011, this “Project Liberty” facility is supported by funding from the Iowa Power Fund and the US DOE. In 2009, POET also

P1: SFK/UKS P2: SFK c11 BLBS075-Hood December 15, 2010

20:56

Trim: 244mm×172mm

11 Fermentation-Based Biofuels

267

established a division to develop systems and infrastructure for harvesting and transportation of corncobs. Godavari Sugar Mills Ltd. is operating an advanced stage pilot plant in Karnataka, India, using 3.5 dry tonnes per day of sugarcane bagasse as feedstock. This company is one of the largest producers of ethanol, and a pioneer in the manufacturing of ethanol-based chemicals in India. The company aims to scale up in the next 3–4 years to a processing capacity of 5,000 tonnes per day. The processing of woody biomass to ethanol at commercial scale via pretreatment and enzymatic hydrolysis of cellulose poses additional challenges due to the higher lignin content of woody biomass. A number of pilot-scale operations utilizing wood waste were initiated within the 2006–2008 period, utilizing diverse technologies. BioEthanol Japan is a conglomerate of five companies evaluating wood construction waste as a feedstock, with pilot plants having an annual ethanol production capacity of 1.4 ML. They are currently advancing an industrial demonstration unit, which was predicted to be online in 2009 with a capacity of 6 ML ethanol per annum. Full-scale commercial facilities are also planned, with a 60–100 ML ethanol per year capacity to be online in 2012–13. The Swedish Etek Etholtekhnik AB consortium has operated a pilot unit with a capacity of 150,000 L ethanol per year since 2004 in Ormskoldsvik, Sweden. An industrial development unit with a 6 ML ethanol capacity is planned and is slated for completion in 2010/11 and for a future commercial facility. In the United States, KL Energy Corp. has operated a 4 ML ethanol demonstration scale plant, in which the cellulose from woody biomass is pretreated by mechanical extraction and digested to glucose using proprietary enzymes. While public records indicate the company has not produced commercial-grade ethanol, the company plans to deploy this technology in Brazil with Add Blue, Ltd. The production of fermentable sugars from woody feedstocks has also been demonstrated via the acid hydrolysis of cellulose to glucose, without the use of enzymes. A pilot plant using Arkenol’s sulfuric acid process has been operating in Japan since 2002, generating 83,000 L of ethanol per year from wood waste. BlueFire Ethanol was established to deploy the proven Arkenol process technology for the conversion of wood and municipal waste material to ethanol. Bluefire received funding from a $40M DOE grant in 2007 and is planning two commercial projects. The larger 68 ML ethanol per year facility in Fulton, MS, will process nonfoodstock urban, forestry, and agricultural wastes available in the region. A smaller 15 ML ethanol per year facility in Lancaster, CA, will utilize post-sorted cellulosic wastes gathered from Southern California’s landfills. With additional funding granted in 2009 by the US DOE for the establishment of the Fulton facility, BlueFire is now acquiring permits and preparing basic preliminary engineering for this site. Mascoma Corp. has developed microbes that allow consolidated bioprocessing of lignocellulosic feedstocks, which eliminates the need for enzymes and additives. A 380,000 L ethanol per year pilot-scale plant in New York has the flexibility to run on different feedstocks and also produces a waste lignin by-product. Mascoma has announced an agreement with Chevron Technology Ventures (CTV) to process feedstocks supplied by CTV, which will include wood chips, agricultural residue, and energy crops, and supply waste lignin by-product to CTV for evaluation and testing for petroleum-based applications. Companies in the forefront of cellulosic ethanol commercial demonstration are expanding, as technologies rapidly evolve. Announced projects are always subject to both technical and commercial considerations. Additional projects at an advanced stage of development are summarized in Table 11.4.

P1: SFK/UKS P2: SFK c11 BLBS075-Hood December 15, 2010

268

20:56

Trim: 244mm×172mm

Plant Biomass Conversion

Table 11.4. Development-stage lignocellulosic ethanol projects.a

Venture

Feedstock

Capacity (L/year

Operation Start year

Location

Petrobras Thai Roong Ruang Group Qteros Dupont Danisco, LLC Ethanol Technologies, Ltd. Flambeau Rivers Futurol HM3 Ethanol Inc Zeachem NextStep Biofuel ICM Inc., Colwich, KS

Bagasse Bagasse Ag residue Ag residue Wood waste Wood waste Wheat straw Poplar Wood Ag residue Energy crops

1,000,000 400,000 Unknown 946,000 Unknown 2,3000,000 180,000 Unknown Unknown Unknown 5,700,000

2010 2010 2010 2010 2010 2010 2010 2008 2011 NA 2012

BR TH US US, TN AU US, WY FR US, OR US US, NE US, KS

a Compiled

from various sources of available public information.

11.4.2 Biobutanol Butanol is a 4-carbon alcohol, widely used as an industrial solvent and intermediate, produced primarily from fossil feedstocks by the oxo process, comprising hydroformylation of propylene with synthesis gas. Prior to development of this lower-cost petrochemical synthetic route, biobutanol was produced by bacterial fermentation of sugars using the ABE process, which yielded a mixture of acetone, butanol, and ethanol. Only a few facilities, mostly in China, currently practice the ABE process, which is plagued by low productivity and high cost for separation of the co-products. As a second-generation biofuel, butanol has several advantages over ethanol, including higher energy density, lower vapor pressure, and improved compatibility with existing distribution infrastructure (e. g., pipelines). Technology efforts are primarily focused on biochemical routes to biobutanol, with several companies and academic research programs actively pursuing enhanced bacterial organisms to improve the fermentation process productivity, selectivity, and cost. Butanol used in chemical applications such as butyl esters currently enjoys a price premium compared to fuel products, and such applications would therefore be the initial market for a biobased substitute. Total United States and European capacity for petrochemical butanol is reported to be in excess of 2.1 MMT/year, with stable or slightly growing demand (Kirschner, 2009). Technology and process optimization, coupled with fossil feedstock pricing, could result in competitively priced biobutanol as a second-generation liquid transportation fuel. Based upon anticipated process similarities, much of the infrastructure at ethanol fermentation facilities may be adaptable to produce biobutanol. Biobutanol blends of up to 11.5% with gasoline can currently be sold in the United States under a special waiver of Clean Air Act rules granted by the Environmental Protection Agency (EPA), under the presumption that the blend is similar to a 10% ethanol blend. However, mixtures with a higher butanol blend ratio would require additional testing and registration with EPA. The growing list of entities developing enhanced fermentation-based butanol technologies is summarized in Table 11.5. United States-based Dupont and United Kingdom-based BP are in the leading commercial position with the 2006 announcement of their joint development program. The planned initial biobutanol demonstration facility for this venture is a converted British Sugar bioethanol plant in Norfolk, UK. DuPont has indicated that following the scheduled 37 ML demonstration plant start-up in 2010, the first commercial plant is expected to be

P1: SFK/UKS P2: SFK c11 BLBS075-Hood December 15, 2010

20:56

Trim: 244mm×172mm

11 Fermentation-Based Biofuels

269

Table 11.5. Biobutanol technology developers. Venture

Feedstock

Capacity (L/year

Location

Dupont, BP Gevo Development, LLC ButylFuel, LLC Cobalt Biofuels Patriarch Partners TetraVitae Bioscience Green Biologics Metabolic Explorer

Sugarbeet Biomass sugar Sugars Ag wood waste Wood Unknown Unknown Unknown

37,000,000 4,000,000 65,000 40,000 8,300,000 Unknown Unknown Unknown

UK US, MI US, OH US, CA US, MN US, IL UK FR

operational in 2013. The initial feedstock was expected to be sugar beets, but plans call for commercial manufacture using a range of feedstocks including both agricultural and cellulosic materials such as energy crops and agricultural waste. In 2009, BP and Dupont announced the formation of a joint venture “Butamax Advanced Biofuels” located in Delaware, as a commercial company that will continue to develop and market their biobutanol. At least five United States-based companies are developing biobutanol technologies. Gevo Development LLC began operation of a 4 ML demonstration plant in September 2009 in Missouri. Partners in the venture include Cargill, ICM Inc, and UCLA, providing expertise in microorganisms, process engineering, and microorganism modification, respectively. Gevo is planning to convert a number of existing ethanol plants with their technology, allowing bioethanol plants to have the flexibility to produce either fuel. ButylFuel LLC has constructed a 65,000 L pilot plant in Ohio to evaluate a continuous flow tubular reactor technology developed in collaboration with Ohio State University, with support from the DOE. Cobalt Biofuels are planning a 40,000 L per year biobutanol pilot plant located in California, to demonstrate continuous fermentation with improved Clostridium strains, followed by vapor compression distillation to produce biobutanol. Patriarch Partners are a private equity firm who have purchased a pulp and chemical mill in Maine following the bankruptcy of Red Shield Environmental. The group has reopened the pulp operation as “Old Town Fuel and Fiber,” and the facility will be retrofitted to produce biobutanol using maple, birch, and beech tree chips. Construction on the plant is planned for 2010, with biobutanol production online in 2011. Illinois-based Tetravitae Bioscience is developing a robust mutant strain of Clostridium beijerinckii, which produces high levels of butanol. There are two other early start-ups based in Europe. Green Biologics is a United Kingdombased biotechnology company utilizing thermophilic microorganisms to improve hydrolysis of biomass to sugars, followed by fermentation to biobutanol. Metabolic Explorer, based in France, is developing fermentation technology to produce five biobased compounds, including butanol.

11.5 Issues for Biofuels Commercial Success 11.5.1

Transport by Pipeline

A key challenge facing biofuels—especially ethanol—is transportation to large markets, often from rural production locations. Ethanol production in the United States and Brazil is dominated by decentralized plants located in rural agricultural areas and relies on rail or truck transport to major fuel markets. Pipeline transport would be more cost-effective but dedicated

P1: SFK/UKS P2: SFK c11 BLBS075-Hood December 15, 2010

270

20:56

Trim: 244mm×172mm

Plant Biomass Conversion

biofuels pipelines are difficult to justify for an emerging industry and require a minimum “critical mass” of product volume for acceptable economics. Biofuels have not been accepted by current pipeline operators, who are generally associated with petroleum product distribution. Nevertheless, successful evaluations of biofuels in existing pipelines have been conducted. For example, Williams Energy conducted a successful test of neat (pure) ethanol in a dedicated US pipeline in the early 1980s. Williams transported 4,600 barrels of ethanol in an 8-inch line from Kansas City to Des Moines. The pipeline was constructed in 1930 and operated in multi-product service. Ten days prior to the ethanol shipment test, the pipeline changed the product being shipped in the line to gasoline. “Pigs” or product separators were used to clean the line ahead of the ethanol prior to the test. The ethanol was batch profiled and tank tested on receipt. From the test, Williams made several recommendations, including frequent dewatering of mainlines using pigs and spheres, closed floating-roof storage tanks to prevent rainwater ingestion, use of dry storage tanks for ethanol, installation of inline corrosion monitoring and filtration systems, and an ethanol quality assurance/control program. Williams’ conclusion stated Our experimental pipeline tests indicate that fuel grade ethanol can be successfully transported in a multi-products pipeline system under controlled conditions. The greater the frequency of batches through any system through any given line segment, the fewer the quality problems that we would expect to experience. (Miller, 2001)

Legislation has been introduced in the US Congress to provide funding assistance for an ethanol pipeline. The bill would authorize up to $5B for renewable fuel pipeline loan guarantees. In March 2009, a major pipeline company, Magellan Midstream Partners of Tulsa, OK, began a feasibility study to construct a biofuels pipeline. Magellan is a combination of the old Williams Pipeline System and Mid-America Pipeline Co. If the legislation passes and financing is available, the pipeline could be operating in 5 years (Flaugh, 2009).

11.5.2 Decentralized Production and Local Distribution Commercially competitive biorefining requires the establishment of efficient manufacturing facilities of sufficient scale, at locations near abundant cost-competitive feedstock supplies, with access to an efficient transportation infrastructure to markets. US corn-based biofuels facilities are concentrated in the Midwest and Great Plains where the majority of US cropland exists. Corn is transported many miles for processing at these facilities. Alternatively, in a comprehensive “carbohydrate economy,” second-generation technologies will use a wider range of plant-derived feedstocks that are more evenly distributed across any biomass-rich region being considered. Globally, abundant lignocellulosic feedstocks include agricultural residues and forest biomass, as well as wood processing residues, urban wood waste, and perennial crops. The natural rural distribution of most biomass argues for locating biorefineries in close proximity to suitable feedstocks in order to reduce inbound transportation costs. However, biofuels must also be transported to wholesale blending facilities and retail fuel outlets in major consuming markets. An optimal future model may seek to size the biorefinery to fit the local feedstock supply, with fuel output distributed in closer proximity to production. This approach not only minimizes inbound and outbound transportation costs, but also creates a truly local energy source, while promoting local economic development.

P1: SFK/UKS P2: SFK c11 BLBS075-Hood December 15, 2010

20:56

Trim: 244mm×172mm

11 Fermentation-Based Biofuels

11.5.3

271

Optimized Engine Performance

Ethanol has a long history as a fuel, as well as an additive, for internal-combustion engines. The first prototype internal-combustion engine in 1826 used alcohol and turpentine. Alcohol powered the first engine by the German inventor Nicholas August Otto, father of the fourstroke internal-combustion engines powering cars today. Henry Ford built his first car to run on what he called farm alcohol. As the automobile era rapidly evolved, scientific journals were filled with references to alcohol. Tests in 1906 by the Department of Agriculture underscored its power and economic benefits. In 1907 and 1908, the US Geological Survey and the US Navy performed 2,000 tests on alcohol and gasoline engines, concluding that higher engine compression could be achieved with alcohol than with gasoline. They noted a complete absence of smoke and disagreeable odors (Kitman, The Secret History of Lead, 2005). As late as 1925, Henry Ford predicted in an interview with the Christian Science Monitor that ethanol—“fuel from vegetation”—would be the “fuel of the future.” Four years later, early examples of his Model A car would be equipped with a dashboard knob to adjust its carburetor to run on gasoline or alcohol (Kitman, The Secret History of Lead, 2005). In the early part of the 20th Century, General Motors and The Ethyl Corporation formed a joint venture to promote the use of octane-boosting lead additives in gasoline to prevent engine knock in high compression engines. As a result, ethanol was largely displaced by tetraethyl lead (TEL) as an antiknock additive for high compression engines. When the environmental and health effects of lead additives were discovered, TEL was effectively banned in the United States and other countries in the 1970s. By 1986, low compression engines with catalytic converters replaced high compression gasoline engines, followed by computer-controlled fuel injection. With the introduction of the low compression engine to meet the requirements for unleaded gasoline, the high compression engine—better suited for ethanol—was relegated to specialized markets. When ethanol was revived in the late 1990s as a liquid transportation fuel, the complementing high compression engine was only to be found in limited markets such as high-performance race cars. E85 could not burn as efficiently in low compression engines, which resulted in lower mileage ratings and a less than favorable public image. The factor that enabled marginal growth of E85 was computerized fuel-injected technology that accounted for the oxygen in the ethanol and allowed the engine to run smoothly, but could not maximize the full energy potential of the ethanol-rich blend. The 2007 US Energy Bill calls for a study to improve the efficiency of flex-fuel vehicles. It is well-documented that ethanol burns more cleanly and at a lower temperature than gasoline. Less recognized is the fact that ethanol has the potential to burn more efficiently than gasoline, in an optimized engine, because of its high octane rating. Test trials within the American Lemans Racing Series and fuel-injected Harley-Davidson motorcycles with high compression engines demonstrate that E85 not only burns at a lower temperature and more cleanly but also provides more horsepower and increased mileage over regular gasoline when burned. Simply put, ethanol is the preferred gasoline octane booster; however, burning ethanol in currently available low-compression engines results in lower gas mileage because the high octane is not used to its full advantage. To achieve mileage efficiencies for ethanol-rich blends such as E85 and even higher, manufacturers must incorporate into their engine design two of the operating concepts that help make today’s diesel engines so efficient. First, the diesel engine is highly turbocharged, meaning that the incoming air is compressed so that more air and fuel can fit inside the cylinder. This

P1: SFK/UKS P2: SFK c11 BLBS075-Hood December 15, 2010

272

20:56

Trim: 244mm×172mm

Plant Biomass Conversion

allows an engine of a given size to produce more power. Second, the engine can be designed with a higher compression ratio (the ratio of the volume of the combustion chamber after compression to the volume before compression). As a result, the burning gases expand more in each cycle, extracting more energy from a given amount of fuel. The combined changes could increase the power of a given-sized engine by more than a factor of two. Current and past technologies exist that can provide clean burning and efficient engines that operate almost exclusively on biofuels, especially ethanol. A Saab Biopower engine, designed with a higher compression ratio, showed only a 12.5% drop in fuel efficiency instead of the 20%–30% drop typically seen with E85 in low compression engines. Saab also achieved a reported 20% extra power and 15% extra torque from this engine. Swedish automaker Scania has been producing ultra-high compression ratio engines designed for ethanol usage, and they reach engine efficiencies as high as 43%. To optimize the engine for its E95 ethanol blend (95% ethanol with 5% ignition improver), Scania raised the compression ratio from 18:1 to 28:1, added larger fuel injection nozzles, replaced gaskets and filters, and altered the injection timing (Bucksch and Egeback, 1999).

11.5.4 Value of Biorefinery Co-products As second-generation biofuels emerge, so do the various types of co-products and residuals that result from these processes. Maximum value creation from co-products will be essential for commercial biorefinery economics. An important co-product from fermentation technologies utilizing lignocellulosic feedstocks will be the aromatic natural polymer lignin. This previously underutilized biomass component, primarily available to date in crude form from the pulp and paper industry, holds great promise as a feedstock for many value-added products, rather than as a process fuel source. Chapter 12 in this volume details the technologies and product opportunities directed toward this future biorefinery co-product.

11.6 Summary Global efforts to expand upon the pioneering commercial introduction of first-generation biofuels must have the full support and will of the public. Second-generation biofuels have the potential for dramatically improved cost, performance, and environmental impact. Models such as those found in Brazil that incorporate a balance between fossil fuels, biofuels, and land use have resulted in public acceptance and energy self-sufficiency. A diversity of sustainable biomass feedstocks, broadly distributed across the globe, is available to produce second-generation renewable fuels without impacting global food supplies. These underutilized resources, if not effectively consumed, will eventually burn or naturally decay with the loss of productive energy and generation of atmospheric carbon. Most agree that solar energy, in its infinite forms, must eventually replace finite fossil fuels. An essential application of solar energy is, in fact, the collection and efficient processing of plant life to extract its photosynthetic energy. The question then becomes how aggressively and efficiently we will choose to utilize sustainable plant life—biomass—to meet our future energy needs.

References Alexander, T. & Torres, T. 2006. In Brazil’s footsteps: The ethanol boom and Latin America. Latin Lawyer, 5, 35–36.

P1: SFK/UKS P2: SFK c11 BLBS075-Hood December 15, 2010

20:56

Trim: 244mm×172mm

11 Fermentation-Based Biofuels

273

USDA Foreign Agricultural Service. 2010. Oilseeds: World market and Trade. Circular Series, FOP 2–10. Arijan, K., Memon, M., Uqaili, M.A. & Al, E. 2009. Potential contribution on ethanol fuel to the transport sector of Pakistan. Renewable and Sustainable Energy Reviews, 13, 291– 295. Balat, M. & Balat, H. 2009. Recent trends in global production and utilization of bio-ethanol fules. Applied Energy, 86, 2273–2282. Bevill, K. 2009. Central American Getaway. Ethanol Producer Magazine. Bromokusumo, A.K. 2009. Indonesia Biofuels Annual. USDA Agricultural Service. GAIN Report Number ID9017. Bucksch, S. & Egeback, K. E. 1999. The Swedish program for investigations concerning biofuels. The Science of the Total Environment, 235, 293–303. Carey, J. 2008. Is Ethanol getting a bum rap? BusinessWeek. Darby, M. 2009. Australia Biofuel Annual. USDA Foreign Agricultual Service. GAIN Report Number AS9023. Drobrescu, M., Henard, M.C., Krautgartne, R. & Al, E. 2009. EU-27 Soybean Imports from the United States still Impeded. USDA Agricultural Service. GAIN Report Number E49079. Du, X. & Hayes, D.J. 2009. The impact of ethanol production on US and regional gasoline markets. Energy Policy, 37, 3227–3234. Elbehri, A., Mcdougall, R. & Horridge, M. 2009. A global model for agriculture and bioenergy: Application to biofuel and food security in Peru and Tanzania. Presented at International Association of Agricultural Economist Conference, Beijing, China. Fang, X., Shen, Y., Zhao, J., Bao, X. & Qu, Y. 2010. Status and prospect of lignocellulosic bioethanol production in China. Bioresource Technology, 101, 4814–4819. Flach, B. 2009. Netherlands–Germany EU-27 Biofuels Annual. USDA Foreign Agricultural Service. GAIN Report Number NL9014. Flaugh, L. G. 2009. Dedicated Ethanol Pipeline Still Under Consideration [Online]. Wallace’s Farmer. Available: http://www.wallacesfarmer.com/story.aspx?s=23052&c=9 (accessed October 31, 2010). Goldemberg, J. & Guardabassi, P. 2010. The potential for first-generation ethanol production from sugarcane. Biofuels, Bioproducts and Biorefining, 4, 17–24. Gopale, A. R. & Kammen, D. M. 2009. Molasses for ethanol: The economic and environmental impacts of a new pathway for the lifecycle greenhouse gas analysis of sugarcane ethanol. Environmental Research Letters, 5, 1–5. Hoh, R. 2009. Malaysia Biofuels Annual. USDA Agricultural Service. GAIN Report Number MY9026. Iijima, M. 2009. Japan to Focus on Next Generation Biofuel. USDA Foreign Agricultural Service. GAIN Report Number JA9044. Joseph, K. 2008. Paraguay Biofuels Annual. USDA Agricultural Service. GAIN Report Number PA8005. Kirschner K. 2009. Chemical profile: 1,4-Butanediol. ICIS Chemical Business. Available: http://www.icis. com/v2/magazine/Issue.aspx?Volume=275&Issue=9 (accessed October 31, 2010). Kitman, J.L. The Secret History of Lead. 2005 [Online]. The Nation. Available: http://www.thenation. com/docprint.mhtml?i=20000320&s=kitman (accessed February 11, 2009). Lichts, F.O. 2009. Report on Biodiesel: World Production by Country. Available: http://www. agra-net. com/portal2/home.jsp?template=pubarticle&artid = 12548286 (accessed October 31, 2010).

P1: SFK/UKS P2: SFK c11 BLBS075-Hood December 15, 2010

274

20:56

Trim: 244mm×172mm

Plant Biomass Conversion

Mabee, M., Neeft, J., Van Keulen, B. & Al, E. 2009. A Review of Key Biofuel Producing countries. Presented at IEA Bioenergy Task 39, Vancouver, Canada. Miller, R. 2001. Production of Ethanol and update on Ethanol Current Events [Online]. Oakland, CA: Williams Presentation. Available: http://www-erd.llnl.gov/ ethanol/proceed/etohupd.pdf (accessed October 31, 2010). Miron, D. 2009. Guatemala in the World. Sugar Journal. Pohit, S., Biswas, P.K., Kumar, R. & Jha, J. 2009. International experiences of ethanol as transportfuel: Policy implications for India. Energy Policy, 37, 4540–4548. Prasertsri, P. & Kunasirirat, M. 2009. Thailand Biofuels Biofuel’s Impact on Food Crops. USDA Agricultural Service. GAIN Report Number TH9047. Saprovek, G., Barretto, A., Berndes, G. & Al, E. 2009. Environmental, land-use and economic implications of Brazilian sugarcane expansion 1996–2006. Mitigation and Adaptation Strategies for Global Change, 14, 285–298. Silalertruksa, T. & Gheewala, S.H. 2009. Environmental sustainability assessment of bioethanol production in Thailand. Energy, 34, 1933–1946. ´ & Cardona, C.A. 2008. Trends in biotechnological production of fuel ethanol S´anchez, O.J. from different feedstocks. Bioresource Technology, 99, 5270–5295. Tay, K. 2009. Guatemala Biofuels Annual. USDA Agricultural Service. GAIN Report Number GT9008. Thamsiriroj, T. & Murphy, J. D. 2009. Is it better to import palm oil from Thailand to produce biodiesel in Ireland than to produce biodiesel from indigenous Irish rape seed. Applied Energy, 86, 595–604. Toasa, J. 2009. Colombia: A New Ethanol Producer on the Rise? Economic Research Service, WRS-0901. Wang, F., Xiong, X. R. & Liu, C. Z. 2009. Biofuels in China: Opportunities and Challenges. In Vitro Cellular & Developmental Biology—Plant, 45, 343–349.

P1: SFK/UKS P2: SFK c12 BLBS075-Hood December 15, 2010

20:57

Trim: 244mm×172mm

Chapter 12

Biobased Chemicals and Polymers Randall W. Powell, Clare Elton, Ross Prestidge, and Helene Belanger

12.1 Introduction Throughout history, humans have recognized and exploited the capability of plants to produce a variety of useful chemical products—in addition to foodstuffs and materials—including medicines, dyes, fragrances, flavorings, resins, and polymers. In earlier times, these products were generally extracted or otherwise recovered from the whole plant and formulated into useful products without molecular characterization and with little or no further chemical transformation. More recently, advancing technologies have elucidated the composition and structures of most of these natural products, many of which are chemically complex and highly functionalized, and in the case of physiologically active materials, optically active (chiral). While numerous commercially significant plant constituents or functionally similar analogs have been synthetically prepared, some particularly complex natural products are still extracted from biomass and formulated or further modified into commercial products, most notably some pharmaceuticals and nutraceuticals. Plant-derived oils and sugars were also important early chemical feedstocks, with fermentative production of ethanol (mostly for consumption) representing one of the oldest commercial chemical processes. The dependence upon renewable plant materials as the primary source for commercial chemical products and feedstocks was dramatically reduced with the discovery of petroleum and development of refining technologies early in the last century. With the advent of catalytic cracking technologies, the petrochemical industry followed closely on the heels of petroleum-based liquid fuels, ultimately using both petroleum and lighter gaseous hydrocarbons (e.g., natural gas) as the primary feedstocks. Abundant, readily accessible fossil hydrocarbon resources were found to be geographically dispersed globally and quickly became the feedstocks of choice for both fuels and chemicals. The fossil-based chemical industry developed over several decades as technologies evolved to convert substantially purified

Plant Biomass Conversion, First Edition. Edited by Elizabeth E. Hood, Peter Nelson and Randall Powell.
C 2011 John Wiley & Sons Inc.

275

P1: SFK/UKS P2: SFK c12 BLBS075-Hood December 15, 2010

276

20:57

Trim: 244mm×172mm

Plant Biomass Conversion

hydrocarbon components through multistage synthetic reactions into increasingly complex functionalized organic chemicals that have become the basis for most of the materials of modern living. Given the finite supply of global fossil resources, recognition that the accelerated consumption of anthropogenic carbon is contributing to greenhouse gases (GHG) and global warming, and the economic and supply considerations of external sourcing of fossil feedstocks for many nations, attention has re-focused on renewable and sustainable biogenic (plant-based) feedstocks to produce the next generation of fuels, chemicals, and polymers. This chapter will examine the current state of advanced precommercial and commercial initiatives—as opposed to Research & Development programs—to introduce biobased chemical products to compete with, displace, and ultimately replace fossil-based materials. In reviewing the state of commercial deployment of biobased conversion technologies, we will focus primarily on chemical and biochemical transformations, which generally seek to preserve and elaborate much of the molecular functional complexity inherent in biobased feedstock components. This analysis will concentrate on recognized and widely available biobased chemical components that can be recovered from the whole plant for step-wise downstream elaboration into chemical derivatives, analogous to the mature petrochemical product chain. Finally, particular emphasis will be placed upon lignocellulosic feedstocks due to their relative global abundance compared to other biomass, the potential range of accessible aliphatic and aromatic derivatives, and the lack of direct competition with the food/feed supply chain.

12.2 Biobased Feedstock Components For the purposes of this review, “biomass” will be defined as renewable organic materials that may be burned directly for energy or converted into biofuels, chemicals, or other materials. Biomass is broadly considered to include wood; agricultural crops and residues, dedicated energy crops, animal fats, and animal and municipal wastes. The latter will not be considered in this review. The chemical composition of biomass is diverse and complex; however, biomass intended as a feedstock for downstream processing contains one or more of the constituents shown in Table 12.1, in commercially useful quantities. Bioprocessing technologies seek to convert these components into other useful downstream products such as fuels, chemicals, and polymers. Oils, or triglycerides, comprise various saturated and unsaturated fatty acids esterified with glycerol and available from both plant and animal (fat) sources. Monomeric 6-carbon sugars, such as glucose and fructose or the two-unit disaccharide sucrose, are produced by certain sugar crops and are readily fermentable by yeasts and organisms to ethanol and other products, without hydrolysis or other pretreatment. Table 12.1. Major biomass feedstocks. Feedstock

Key Chemical Component(s)

Crop Examples

Oils

Plant oils: triglycerides

Soybeans, canola, camelina, algae

Sugar/starch

Hexoses, di- and polysaccharides

Sugar cane, sugar beets, sweet/grain sorghum, corn, barley

Lignocellulose

Lignin, cellulose, hemicellulose

Wood, crop residues, switchgrass, miscanthus

P1: SFK/UKS P2: SFK c12 BLBS075-Hood December 15, 2010

20:57

Trim: 244mm×172mm

12 Biobased Chemicals and Polymers

277

Polysaccharides, commonly known as starch, are the primary components in the grain of crops such as corn and barley. Unless they are used to make starch-based polymers, these must generally be hydrolyzed or cleaved by enzymatic or chemical means to simple sugar molecules for downstream biochemical transformation. Woody and herbaceous biomass—often referred to as lignocellulosic biomass—is the subject of significant technology development due to its abundance and potential as a bioprocessing feedstock (Perlack et al., 2005). The components of interest as chemical feedstocks—lignin, cellulose, and hemicellulose—vary according to the type and species of biomass, but generally fall into the range of 15%–25% lignin, 23%–32% hemicellulose, and 38%–50% cellulose. Cellulose is a biopolymer composed of six-carbon (C6) sugar units; hemicellulose is principally composed of five-carbon sugars but does also contain C6, while lignin is an aromatic polymer, comprising various phenylpropenyl molecular units. Together, the lignocellulosic polymers comprise the structural components of plant matter and are produced by the photosynthetic process, whereby atmospheric carbon dioxide is absorbed by the plant, chemically transformed, and “fixed” into these and other useful chemical materials. Cellulose and hemicellulose are aliphatic polymers composed only of carbon, hydrogen, and oxygen, thus belonging to the general chemical class of carbohydrates. Cellulose is the most abundant biopolymer on earth, composed of glucose (C6) monomeric units connected by beta-1,4-glucosidic bonds, resulting in a crystalline morphology. Cellulose obtained from wood pulp, cotton, and other plants has been used for centuries to produce paper and cardboard, as well as derivative products. With recent technology developments, cellulose can now be commercially depolymerized (hydrolyzed) by enzymes to its monomer sugars, which can then be readily utilized as feedstocks for bioprocessing. Yeast fermentation of C6 sugars to ethanol has been practiced for centuries, and other natural and genetically modified organisms (profiled elsewhere in this edition) can convert these sugars to various useful chemical molecules. Hemicellulose is an amorphous biopolymer composed of various sugar monomers (principally five-carbon sugars such as xylose) and can be readily hydrolyzed to its monomeric sugars with dilute acid/base or enzymes. Unfortunately, C5 sugars such as xylose cannot be fermented using natural yeasts, although new organisms and genetically modified yeasts are now being developed to utilize xylose as a bioprocessing feedstock. Lignin is a natural polymer found in all plant materials, which combines with cellulose and hemicellulose to provide structural strength to the plant and direct water flow. It is a complex aromatic polymer composed of phenylpropenyl molecular units, imparting chemical stability and high calorific value and affording the potential for valuable chemical transformations as well as energy recovery.

12.3 Biomass Conversion Technologies 12.3.1

Technology Platforms Overview

The technologies for conversion of biomass or biomass components to chemical products can be grouped into three distinct platforms—chemical, thermochemical, and biochemical—as depicted in Figure 12.1. Each platform offers specific characteristics for commercial processing, including variety of feedstocks and products, coproducts, cost, scale, and stage of technology development. Traditional chemical processing of biomass feedstocks is often overlooked in technology discussions in preference to the biochemical and thermochemical technology platforms being

P1: SFK/UKS P2: SFK c12 BLBS075-Hood December 15, 2010

278

20:57

Trim: 244mm×172mm

Plant Biomass Conversion

CHEMICAL Oleochemical/Chemical platform

THERMOCHEMICAL Gasification platform

Fuels, chemicals, energy, and Coproducts

BIOCHEMICAL Sugar fermentation platform

Figure 12.1.

Biomass conversion technologies (Tripp et al., 2009).

developed for lignocellulosic feedstocks. Transformation of biomass materials by classical chemical and catalytic reactions has been practiced commercially for over a century, although focused on a relatively narrow range of products. The most significant processing sector is the pulp and paper industry, which uses lignocellulosic feedstocks (mostly wood) to produce paper and cardboard products. In addition to paper, cellulosic polymers obtained by chemical modification of cellulosic feedstocks also represent important commercial products and a mature industry sector, which will be described later in this chapter. The other notable industry employing chemical processing of biomass feedstocks is the oleochemical sector. Oleochemical manufacturing facilities are mature biorefineries. This industry has its roots in the saponification (base treatment) of fats and oils to produce soap. The acid- or base-catalyzed hydrolysis chemical reaction is the basis of the oleochemical industry, producing fatty acids and glycerol, which are purified and used directly or reacted further to downstream derivatives. Commercial derivatives include fatty acid methyl esters (FAME or biodiesel), fatty alcohols, fatty amines and amides, alcohol ethoxylates and sulfates, and acylglycerols. These products find many industrial and end-user applications in areas such as coatings, surfactants, lubricants, detergents, and consumer products. The industry is characterized by significant global infrastructure and highly integrated processing facilities (Bergstra, 2007; Shaine et al., 2004). Most recent commercial development of thermochemical processing of biomass has focused on conversion of lignocellulosic feedstocks primarily to liquid fuel products (biomass-to-liquid or BTL technologies). Generally speaking, the core gasification and pyrolysis technologies have been practiced in the chemical processing industries for many years, although these are still being adapted to the use of biomass rather than fossil feedstocks. Thermochemical processes have the advantage of converting all the biogenic carbon-containing components—lignin, cellulose, and hemicellulose—to products. Pyrolysis generally produces complex mixtures of partially deoxygenated aliphatic products which may require refining or purification to afford downstream fuel products or chemical intermediates. Gasification reduces the functional complexity of biogenic carbon to a simple C1 feedstock, carbon monoxide, which is produced in

P1: SFK/UKS P2: SFK c12 BLBS075-Hood December 15, 2010

20:57

Trim: 244mm×172mm

12 Biobased Chemicals and Polymers

279

combination with hydrogen and other minor components and is commonly referred to as synthesis gas. Downstream processes can then elaborate this simple feedstock to more complex products. Chemical catalytic conversion of synthesis gas “syngas” to fuels leverages existing knowledge, while conversion by bio-organisms represents novel technology being developed by several groups. Historically, because of the temperature and pressure requirements, thermochemical processes have required economy of scale for commercial viability, although several groups are attempting to develop smaller modular gasification and downstream processing units to address this issue. Thermochemical technologies are carbon efficient, converting all forms of biogenic carbon to simpler molecules, but in the process lose much of the inherent functional complexity of the biomass feedstock. In light of this characteristic, thermochemical processes would therefore appear to be particularly well-suited for commercial production of liquid fuel products, which are most commonly mixtures of simple alcohols, hydrocarbons, or other mono-functional compounds. Shorter chain aliphatics are being developed as replacements for gasoline, while the longer-chain compounds accessible by reforming could substitute for aviation and diesel fuels. Biochemical processing, sometimes referred to as the “sugar” platform, seeks to convert the six-carbon and five-carbon sugars derived from biomass to fuels and chemicals through the use of yeasts, enzymes, and microorganisms. In addition to sugar and starch feedstocks, recent R&D has focused on the development of pretreatment systems and enzymes (See Chapter 9 in this volume), which can depolymerize cellulose and hemicellulose into their monomeric sugars, for fermentation to “cellulosic” ethanol or other liquid fuels. While C6 sugars are readily fermented to ethanol by natural yeasts, current R&D programs seek to develop new organisms that can also effectively convert the C5 sugars derived from lignocellulosic biomass to ethanol and other chemical products. Fermentation processes utilizing new organisms can increasingly effect highly specific molecular transformations that can preserve much of the inherent functional complexity of sugar or carbohydrate feedstocks. Thus, biochemical transformations hold particular promise for development of a broad range of sustainable and commercially viable aliphatic functionalized chemical intermediates, which could ultimately replace petrochemical materials. In addition, polymeric and monomeric materials derived from lignin (as discussed later in this chapter) provide access to a sustainable aromatic chemicals platform based upon biomass feedstocks.

12.3.2

Lignocellulose Fractionation Overview

The mature biorefinery must optimize the use of all components of the biomass. Lignocellulosic biomass is now recognized as an immense source of cellulose-derived sugar that can be used to support the bio-production of a variety of products. Hemicellulose can serve as a unique source of pure xylose, and lignin offers great potential to substitute for many petrochemical-derived aromatic chemicals. One of the major barriers for the optimal use of lignocellulosic biomass has been the inability to efficiently separate (fractionate) and recover all the major components in usable forms. Fractionation is increasingly being viewed as a crucial processing stage for the effective use of all lignocellulose components and greatly expands the options for downstream conversion and product derivatization. In addition to the flexibility afforded, biorefinery fractionation technologies offer the potential for bolt-on or new pulping processes in the pulp and paper industry that can extract higher value from pulpwood feedstocks. These technologies, to enable an “Integrated Forest Products

P1: SFK/UKS P2: SFK c12 BLBS075-Hood December 15, 2010

280

20:57

Trim: 244mm×172mm

Plant Biomass Conversion

Table 12.2. Lignocellulosic fractionation in the pulp and paper industry. Implementation

Development

Chemistry

By-Product

Borregaard Tembec Metso Valero energy Green value Lignol Tecnaro

Chemcell Tembec Innventia American process Granit RD Repap Fraunhofer institute

Sulfite Sulfite Kraft Sulfite Organosolv Sulphite Organosolv Hydrothermal/Organosolv

Sugars, ligninosulfonates Sugars, ligninosulfonates Pure lignin Sugars, ligninosulfonates Pure lignin Pure lignin, furfural Pure lignin

Biorefinery” (IFPB), have not yet been significantly demonstrated at commercial-scale facilities. The industry is characterized by large capital-intensive facilities, with highly integrated process chemical and heat recovery systems, which complicates retrofitting for more efficient processing of additional biomass components.

Fractionation in the Pulp and Paper Industry In the pulp and paper industry, biomass fractionation technologies have been extensively developed over many decades to achieve selective delignification and recover highly conserved polymeric forms of “purified” cellulose fiber. These celluloses are then used for the production of various paper pulp grades or dissolving pulp for the production of esters (e.g., cellulose acetate) or ethers (e.g., hydroxyethyl cellulose). The sugars and lignin-rich pulping liquors, where the majority of the remaining biomass components are dissolved, are dewatered and used as boiler fuel to produce steam and electricity. Table 12.2 summarizes advanced development and commercialization efforts to recover the valuable by-products contained in pulping effluents; however, only the Tembec and Borregaard sulfite pulping mills are operating as “biorefineries” at the commercial scale. The complexity, significant modification, and degradation of the biomass components in pulp and paper effluents have made their exploitation difficult. In the sulfite salt cooking liquor, lignin is converted by direct sulfonation into water-soluble lignosulfonates, and the hemicelluloses (with some cellulose) are hydrolyzed to pentose and/or hexose monomers and oligomers (mainly xylan). The lignosulfonates are generally recovered and purified from the sulfite liquor via membrane filtration (Restolho et al., 2009). The sugar-rich liquor remaining after lignin recovery can be fermented to ethanol using adapted or modified biocatalysts (Helle et al., 2008). The ethanol yields are, however, generally low due to a low yield of fermentable sugars and the abundance of fermentation product inhibitors such as furfural, generated by extensive hemicellulose hydrolysis during the pulping process. Borregaard’s Chemcell sulfite mill in Sarpsborg, Norway, is perhaps the most advanced lignocellulosic biorefinery, producing 160,000 metric tons (MT) per year of specialty cellulose, as well as 160,000 MT of lignosulfonates, 20 million liters (ML) of industrial ethanol, and 1,200 MT of vanillin from spruce pulping black liquor. The TEMBEC Inc. industrial complex in Temiscaming, Quebec, comprises several mills and processing plants for hardwood and softwood biorefining. The complex includes a chemical plant producing about 180,000 MT of lignosulfonates and 80 MT of phenol-based resins per year, as well as a biochemical plant producing 18 ML per year of industrial ethanol from the

P1: SFK/UKS P2: SFK c12 BLBS075-Hood December 15, 2010

20:57

Trim: 244mm×172mm

12 Biobased Chemicals and Polymers

281

waste sulfite liquor. The lignosulfonates produced from the fermentation residue are sold as dispersants and binders. The black liquor produced by the Kraft pulping process contains a mixture of sulfonated and unsulfonated degraded lignin, with numerous decomposition products. Only about 0.1 percent of Kraft lignin produced in the United States is diverted for use as a chemical resource; the remainder is burned for process energy. The Kraft lignin recovered after filtration and acidic precipitation of the black liquor is water insoluble with a highly condensed and modified structure. Borregaard Ligno Tech (Sarpsborg, Norway) produces water-soluble lignosulfonates by postsulfonation of Kraft lignin. The sulfonated Kraft lignin finds similar low-grade industrial uses as sulphite process lignosulfonates. There is a continued interest in the recovery of Kraft lignin for higher-value applications, but limited commercial activities at present. Among these, Stockholm-based Innventia (formerly STFI-Pacforsk) in 2007 established a demonstration plant using the Lignoboost process, to recover the lignin from the Kraft black liquor of the Nordic Paper Mill in B¨ackhammar. The plant has a production capacity of 10,000 MT per year of pure unsulfonated Kraft lignin. The Lignoboost process using CO2 injection, precipitation, and filtration to recover lignin from the black liquor was developed in collaboration with Chalmers University of Technology (Goteborg, Sweden). Further commercialization and implementation of the technology has been undertaken by Finland-based Metso Corp., who acquired rights to the process in 2008. During alkaline pulping, the xylan hemicellulose polymers are partially hydrolyzed and often precipitate with the cellulose fraction. To address this problem, a pretreatment step (steam or mild acid catalysis) is usually applied to the biomass in order to extract the hemicellulose sugars before the pulping process (Um and van Walsum, 2009; Walton et al., 2010). For example, the Old Town kraft mill in Maine, USA, was retrofitted to extract the hemicellulose from 80 MT per year of hardwood chips. Developed with the University of Maine, the “near-neutral” pre-extraction step is performed with a proportion of pulping liquor in order to preserve the integrity of the cellulose fibers for downstream pulp making (Mao et al., 2008). The Old Town mill was acquired in 2008 by Patriarch Partners and now operates as Old Town Fuel and Fibres with the intent to install a biorefinery module to ferment the extracted sugars to 5.7 ML per year of butanol. The xylose-rich sugar liquor extracted from hardwood can also serve as a feedstock for xylitol or furfural production and other chemical and polymer products profiled later in this review. American Value Added Pulping (AVAPTM ) is a proprietary process based on an ethanol–water–sulphur dioxide cooking liquor to fractionate softwood into a cellulose pulp, lignosulfonates that are precipitated and burned, and a remaining liquid fraction containing hydrolyzed hemicelluloses used for ethanol production. American Process (Atlanta, GA), the developer of the AVAPTM process, was awarded $18M in 2009 from the US Department of Energy (“DOE”) for the establishment of a biorefinery in Alpena, MI, to produce 3.3 ML per year of ethanol fuel and 2.6 ML per year potassium acetate. The Alpena Prototype Biorefinery (APB is a partnership between American Process Inc. and San Antonio, TX, based Valero Energy Corporation) will use wood waste from an adjacent decorative wood panel plant. To reduce sulfur emissions and prevent the formation of chlorinated organic compounds in the waste effluent, alkaline pulping in the absence of sulphide is also practiced. The Soda pulping process (sodium hydroxide-catalyzed) is generally used for the production of cellulose fibers from agricultural residues and for the treatment of short-fibered hardwoods. Soda pulp

P1: SFK/UKS P2: SFK c12 BLBS075-Hood December 15, 2010

282

20:57

Trim: 244mm×172mm

Plant Biomass Conversion

is primarily mixed with sulfite pulp to make printing papers. The lignin produced by soda pulping is generally highly modified and unstable, and like Kraft lignin, difficult to recover in a usable form. Switzerland-based Granit RD and GreenValue have developed a process to extract and use the lignin from Soda pulping spent liquor. In 2005, the lignin recovery process based on precipitation and filtration was integrated to a nonwood pulp and paper mill owned by ABC paper in the Punjab region of India. Implementation of this technology reduces the effluent load of the Soda mill by 50% and has a production capacity of 5,000 MT per annum of sulfur-free lignin from wheat straw, bagasse, and sarkanda grass. The purified lignin is formulated into phenolic-based resin and marketed by ALM India Pvt. Ltd. At an early stage of development, Pure Lignin Environmental Technology based in British Columbia, Canada has developed a process using ammonium hydroxide and caustic soda treatment with acid precipitation of the black liquor to produce a strong cellulose pulp. Unsulfonated lignin and sweet liquor used as a substrate for unicellular protein production systems are recovered from the process. A pilot plant in Durango, Mexico, with a capacity of 200 kg/day of lignin from softwood waste, has operated since 2008 and has now been reconfigured to be portable for demonstration purposes. Organosolv pulping was developed as an environmentally friendly alternative to conventional chemical pulping and has been the subject of extensive research and development but limited commercial application. A wide number of pulping processes based on the use of a variety of organic solvents (ethanol, methanol, acetic and formic acid, acetone, ethylene glycol) have been tested at large scale (Muurinen, 2000). The efficient degradation and dissolution of lignin in organic solvents such as ethanol or methanol allow the highly selective delignification of wood biomass without the addition of large amounts of inorganic catalyst. Pulping with organic solvent combined with water produces good quality polymeric cellulose pulps that are easy to bleach using chlorine-free processes. Because of the milder conditions of extraction, the lignin dissolved in the Organosolv black liquor retains many of its desirable intrinsic properties. The sulfur-free Organosolv lignin is also easy to recover in a highly uniform preparation by acidic precipitation without complex purification schemes. Furfural derived from the degradation of the hemicellulose generally accumulates in the black liquor during Organosolv fractionation for the production of cellulose fibers. It can be recovered by distillation of the liquor obtained after lignin precipitation. In the late 1980s, Repap Enterprises established a facility in New Brunswick, Canada, to demonstrate and test a proprietary ethanol-based Organosolv pulping method known as the Alcell process. The mill, producing 20 tons of pulp and 4.5 tons of high quality lignin per day, ceased operation in the mid-1990s. The Alcell technology was later acquired by Lignol Innovations based in British Columbia, Canada, who have continued to develop processes for the production of cellulose for conversion to ethanol, as discussed later in this section. The Fraunhofer Institute for Chemical Technology in Germany has worked on a supercritical water/methanol-based fractionation process for the recovery of high grade lignin from a variety of feedstocks (Eisenreich et al., 2004). In this “high pressure hydrolysis” process, developed in collaboration with Kogyo Seiyake CoT (a Japanese manufacturer of surfactants, polymers, and resin additives), the wood chips are first submitted to a supercritical water treatment for the extraction and hydrolysis of the hemicellulose sugars. After separation of the hemicellulosic sugar stream, a supercritical aqueous methanol treatment is applied to the chips in order to extract and solubilize the lignin. Germany-based Tecnaro, a spinoff of the Fraunhofer Institute for Chemical Technology, is currently developing and commercializing resins based on these lignins for thermoplastic extrusion for injection molding.

P1: SFK/UKS P2: SFK c12 BLBS075-Hood December 15, 2010

20:57

Trim: 244mm×172mm

12 Biobased Chemicals and Polymers

283

Table 12.3. Llignocellulosic fractionation for the sugars platform Implementation

Development

Process

Biomass

Abengoa, Verenium Greenfield Ethanol Shell Inbicon KL Energy Biogasol MBI International Lignol Pure Power Global Blue Fire

Sun Opta

Steam Explosion

Bagasse, Wheat straw

Iogen DonG Energy CBRD Biogasol Michigan State . Lignol BioJoule Inc. Arkenol

Acid Steam explosion Hydrothermal Thermomecanical Steam Explosion/Oxidation Ammonia fiber expansion Organosolv Organosolv Concentrate acid

Wheat straw Wheat straw Wood waste Wheat straw Grasses Wood waste Wood waste Municipal solid waste

Fractionation for the Sugars Platform Lignocellulose fractionation may prove even more valuable as an integrated processing component of the optimized biorefinery, providing a high yield of sugars from both hemicellulose and cellulose, as well as a high-quality sulphur-free lignin product stream. The efficient utilization of lignocellulosic biomass as a source of fermentable sugars for the bioprocessing industry relies on the enzymatic digestion of cellulose and hemicellulose polymers into fermentable glucose and xylose monomers. For this purpose, partially degraded cellulose polymers are considered to be better suited as substrates for hydrolysis into glucose monomers. Therefore, as opposed to a paper pulp fractionation process, which seeks to remove maximum amounts of lignin while keeping the cellulose fiber intact, fractionation processes for the production of cellulose substrate for enzymatic digestion are optimized to expose and disrupt the crystalline structure of the cellulose to increase its sensitivity to enzymatic hydrolysis (Taherzadeh and Karimi, 2008). Table 12.3 summarizes leading pretreatment/fractionation methodologies that are under development for the production of cellulose substrate for enzymatic hydrolysis (see Chapter 9 in this volume). Few of these have reached the commercial demonstration stage. Generally combining heat, pressure, and chemical reactions, pretreatment processing methods directed to the biochemical platform are designed to minimize introduction or formation of contaminants that would be toxic to the downstream enzymes and bio-catalysts (Kumar et al., 2009; Sousa et al., 2009). Lignocellulosic biomass pretreatment based on autocatalytic or weak acid hydrolysis for the production of fermentable sugars are the most advanced towards commercial deployment. Often developed for integration to an existing fermentation and distillation facility, these pretreatment processes have generally been developed on low lignin feedstocks such as corn stover, wheat straw, or bagasse. During autocatalytic and weak acid hydrolysis, a large proportion of the original biomass remains with the cellulose fraction. In noncatalyzed or autocatalyzed pulping, the cooking liquor becomes acidified by the acetic acid released from the wood. Efficient hydrolysis of the cellulose produced from autocatalyzed hydrolysis pretreatments such as steam explosion or pressurized hot water treatment requires high enzyme (xylanase and cellulase) loadings, due to the presence of partially hydrolyzed hemicelluloses and recondensed lignin in the cellulose fraction. In order to increase the level of hemicellulose and lignin hydrolysis, an acid catalyst is often added. However, the addition of acid is also associated with reduced

P1: SFK/UKS P2: SFK c12 BLBS075-Hood December 15, 2010

284

20:57

Trim: 244mm×172mm

Plant Biomass Conversion

cellulose yields and formation of downstream digestion and fermentation inhibitors such as furfural, hydroxymethylfurfural (HMF), and phenolic compounds. The challenge is to obtain the optimal balance of production and decomposition of the product. One of the best-known commercial lignocellulose fractionation processes is the steam explosion-based fractionation developed and owned by Canada-based SunOpta. SunOpta’s pretreatment process has been licensed by several groups with an interest in the production of ethanol from lignocellulosic feedstocks, including Spain-based ethanol producer Abengoa, Verenium in the USA, and Greenfield Ethanol in Canada. Among these, Verenium is the most advanced, operating a pilot plant in Jennings, LA, since 2008 using weak acid-catalyzed steam explosion. Iogen has had a demonstration-scale plant in operation using its proprietary acid-catalyzed steam explosion pretreatment for the processing of wheat straw to fermentable sugar for ethanol production. With a processing capacity of 30 MT per day of wheat straw, the plant has been in operation in Ottawa, Canada, since 2004. With a fifty percent stake in Iogen, Royal Dutch Shell Group is now selling the cellulosic ethanol at its petrol station in Canada. Inbicon, which recently started operation in Denmark, is processing wheat straw using a hydrothermal (pressurized hot water) pretreatment process to produce cellulose for enzymatic digestion, C5 molasses, and lignin pellets that are burned to produce steam and electricity (Larsen et al., 2008). The facility is integrated with DONG Energy’s Asnaes Power Station in Kalundborg, Denmark. KL Energy use a proprietary thermomechanical pretreatment developed with the South Dakota Center for Bioprocessing Research and Development (CBRD) for processing wood waste from wood mill and lumber. In the KL process, the hemicellulose-derived sugars are separated and recovered as sugar-rich syrup and the lignin pelletized for burning. The KL process has been operating at a demonstration scale in Upton, WY, since 2008. BioGasol, a spin-off from the Technical University of Denmark, has operated a pilot plant in Copenhagen since 2006, using proprietary steam explosion with a wet oxidation lignocellulosic fractionation scheme. The annual capacity of the plant is 16,000 L ethanol per annum. In the process, hemicellulose sugars are sequentially extracted for fermentation to hydrogen using a thermophilic anaerobic bacterium. The cellulose fraction is enzymatically digested to glucose and the lignin is recovered for energy production, with the effluents treated by anaerobic digestion for biogas production. BioGasol is planning the construction of an ethanol facility with a capacity of 5.0 ML per year in Bornholm, Denmark. Canada-based NovaGreen Inc. was founded in 2007 to implement a steam explosion-based biofuel refining process. They have proposed to build a pilot bio-reactor in Killam, Alberta, to demonstrate the capability of separating hemicellulosic oligomers and purify high grade lignin from the cellulose of wheat, barley, oat, and flax straw using a steam explosion-based fractionation process. Construction for the initial $2.7 Million Canadian Dollar (CAD) facility was planned to start in 2009 with a $22 Million CAD expansion planned for 2010. Alkaline-based pretreatments (lime, sodium hydroxide, and ammonia) are used for lignin solubilization and the production of cellulose with lower inhibitor content. Because of inefficient hemicellulose hydrolysis under alkaline conditions, the digestion of cellulose requires higher levels of hemicellulose-specific enzymes (xylanase). The lignin dissolved in the alkaline cooking liquor is usually heterogenous, modified, and difficult to recover in a usable form. Among alkaline-based pretreatments, the ammonia fiber expansion (AFEX) method developed at Michigan State University in the United States has attracted interest due to its low water requirement. AFEX pretreatment optimized for low lignin feedstocks results in cellulose decrystallization, partial hemicellulose hydrolysis, and an alteration of lignin structure

P1: SFK/UKS P2: SFK c12 BLBS075-Hood December 15, 2010

20:57

Trim: 244mm×172mm

12 Biobased Chemicals and Polymers

285

(Murnen et al., 2007). Because AFEX is a relatively dry process, both hydrolyzed lignin and hemicellulose are retained with the cellulose. However, the solid pretreated residue remains highly digestible when supplemented with additional xylanase enzyme (Sousa et al., 2009). The development arm of Michigan State University foundation, MBI International, has the exclusive license for the AFEX pretreatment technology and has been collaborating with various partners such as the US National Renewable Energy Laboratory (NREL), North Dakota State University, Audubon Sugar Institute at Louisiana State University (USA), and United States firms Poet and Qteros (formerly Sun Ethanol) for scaling up the fractionation process. PureVision based in Fort Lupton, CO, in collaboration with the Auburn University Center for Bioenergy and Bioproducts, has developed a two-stage process using autocatalyzed or dilute acid pretreatment for hemicellulose sugar extraction followed by sodium hydroxide-catalyzed extraction to produce a cellulose slurry fraction and a liquor fraction containing lignin and other extractives (Kadam et al., 2009). The lignin is recovered from the liquor fraction by acidic precipitation. After 4 years of development using a reactor with a capacity of 50 kg per day, in 2009, PureVision scaled the process to 500 kg per day. The production of cellulose substrates for enzymatic digestion by organic solvent fractionation has also attracted significant attention, particularly for the processing of high lignin content biomass like wood (Zhao et al., 2009). The treatment of the lignocellulosic biomass with an organic solvent allows not only the production of highly digestible cellulose but also the possibility for recovery of other high-value products, notably high grade lignin. The most advanced Organosolv-based fractionation processes are mainly directed at the production of cellulose fibers for paper and dissolving pulp as described earlier. With support from the Canadian government, Lignol Innovations in 2009 commissioned a novel ethanol Organsolv processing pilot plant with a capacity of 1 dry MT per day of hardwood or softwood, at the British Columbia Institute of Technology in Burnaby, BC. The plant was established to identify key process improvements and equipment configurations for the production of ethanol from cellulose and recovery of various lignin fractions from the aqueous ethanol pulping liquor. Recent R&D activity has sought to modify organic solvent-based pulping processes to achieve higher yield of a cellulose fraction that is more reactive to enzyme digestion (Garcia et al., 2009). Process parameters are modified in order to limit the extensive degradation of the lignin and its recondensation on the cellulose fraction and to optimize the hydrolysis of the hemicellulose by limiting its degradation to furans (Xu et al., 2007). Wide ranges of catalysts, solvents, and operating conditions have been assessed on a variety of feedstocks. Catalyzed Organosolv extraction suffered the same disadvantages as water-catalyzed extraction, namely excessive degradation of carbohydrates (to furfural, HMF, levulinic and formic acids) under acidic conditions and insufficient hemicellulose hydrolysis under alkaline conditions (Del Rio et al., 2009). Pure Power Global (Singapore) is commercializing a novel fractionation process based on a sequential Organosolv and hydrothermal treatment of lignocellulosic biomass. The technology was developed by BioJoule Limited based in New Zealand to maximize the recovery of pure functional lignin and increase the yield of xylose and highly digestible cellulose while minimizing the formation of compounds inhibitory to downstream biochemical conversion. The Pure Power process has been optimized for the recovery of a uniform high molecular weight lignin that has retained the distinctive characteristics of the native phenylpropane structures. The hemicellulose polymers, minimally affected during the Organosolv extraction, are hydrolyzed in the subsequent hot water extraction step. The cellulose fraction is therefore highly sensitive to enzymatic digestion, ensuring a high rate of cellulose conversion to fermentable

P1: SFK/UKS P2: SFK c12 BLBS075-Hood December 15, 2010

286

20:57

Trim: 244mm×172mm

Plant Biomass Conversion

sugar at lower cost. In addition to producing a higher grade lignin, this two-stage process increases the yield of xylose by making the hemicellulose hydrolysis easier to control once the majority of the lignin has been removed. The Pure Power technology has been validated at the pilot scale on wood, agricultural waste, and grass feedstock and is ready for semi-commercial implementation. In another Organosolv-based fractionation process developed at Virginia Tech in Blacksburg, VA, the cellulose and hemicellulose are first hydrolyzed with a concentrated phosphoric acid solution and then precipitated during subsequent treatments with aqueous acetone. The partially hydrolyzed hemicellulose is recovered by water washes of the cellulose fraction. The insoluble lignin fragments generated during the phosphoric acid hydrolysis are further broken down and dissolved in the acetone precipitation liquor. This process was licensed to US-based Mascoma in 2006 for cellulosic ethanol production and to French biotechnology company Biom´ethodes in 2008 for the production of biohydrogen. Though not generally undertaken as a fractionation technology, chemical hydrolysis of lignocellulose is also practiced for the production of fermentable sugars instead of enzymatic hydrolysis (Taherzadeh and Karimi, 2007). More than 20 facilities utilizing acid hydrolysis for the production of fermentable sugars were operated in Europe, Russia, China, Korea, and the United States in the 1940s. Concentrated acid hydrolysis of lignocellulosic biomass gives a higher yield of sugar than enzymatic hydrolysis, but remains impractical due to the large amount of concentrated sulfuric acid (60%–90%) required to achieve these high yields. The process requires corrosion-resistant reactors and the recovery of the acid has a high energy demand. During concentrated acid hydrolysis, hemicellulose is completely degraded to furfural or 5-HMF, which will interfere with the efficiency of downstream fermentation. Despite the high investment and maintenance costs, the concentrated acid process is still of interest. The concentrated acid-based Arkenol process has been tested on wood waste at a pilot scale in Izumi, Japan, since 2004. The facility, operated by the Japanese Gasoline Company has an annual ethanol production capacity of 83,000 L per year. BlueFire, with the support of the US DOE, has two commercial Arkenol-based acid hydrolysis facilities in the advanced planning stage for locations in the United States. HCL CleanTech Ltd., an Israel-based development company, has developed a fuming hydrochloric acid pretreatment process to achieve a high rate of cellulose to glucose conversion from lignocellulosic biomass. The process was made economically attractive by improving the recovery of gaseous hydrochloric acid. Incorporated in 2007, HCL CleanTech. Ltd. received funding from Khosla Ventures and Burrill & Co. for the establishment of a 1.25 MT per day pilot facility in North Carolina, US, scheduled to go on-line by the end of 2010. In order to avoid the formation of fermentation inhibitors, two-stage dilute acid processing is more commonly used. In the first extraction step, under mild acid conditions, hemicellulose is converted to sugar monomers. After washes, the residual solid is hydrolyzed under more severe condition, allowing cellulose to glucose conversion. With this strategy, high levels of pentose sugars are recovered from hemicellulose but a lower yield of glucose is achieved due to less efficient separation of the lignin from the cellulose component. The majority of the lignin is recovered after the sugar fermentation step and is burned to produce steam. The Tennessee Valley Authority (TVA) has developed a 2 MT per day pilot plant in Muscle Shoals, AL, using a continuous two-stage sulfuric acid process. Since 1997 US-based Pure Energy has been working with TVA on their dilute sulfuric acid-based pretreatment for the conversion of hemicellulose to furfural and the separation of lignin for thermal energy. Pure Energy’s technology is being deployed by Raven Biofuel, with several projects under development in the United States and Canada.

P1: SFK/UKS P2: SFK c12 BLBS075-Hood December 15, 2010

20:57

Trim: 244mm×172mm

12 Biobased Chemicals and Polymers

287

12.4 Biobased Products 12.4.1

Oil-Based Products

Biodiesel Biodiesel (also referred to as FAME for “fatty acid methyl ester”) is produced from vegetable oils and animal fats by reaction with an alcohol (typically methanol) and a base catalyst, through a relatively simple chemical reaction known as transesterification. Approximately 1 kg of coproduct glycerin is produced for every 9 L of biodiesel. Pure biodiesel is designated as B100 (100% biodiesel) and can be burned in unmodified diesel engines; however, lower blends with petrodiesel in the range of B5 (5% biodiesel) to B20 (20% biodiesel) are most common. Biodiesel produced from soy and canola or rapeseed oil generally exhibits enhanced cold flow properties and therefore often commands a higher price than animal fat biodiesel. With the growth of the worldwide biodiesel industry, oil and fat prices have become highly correlated with crude oil prices, as feedstock cost comprises 70%–75% of biodiesel production cost. A key issue for growth of biodiesel as an alternative fuel is feedstock supply. In addition, competing technologies have emerged for production of biobased hydrocarbon fuels from oil and fat feedstocks. A recent report by the California Environmental Protection Agency has provided a comprehensive profile of these technologies (Cleary et al., 2009). Referred to as hydrogenation-derived renewable diesel (HDRD) and Fatty Acid to Hydrocarbon (FAHC-Hydrotreatment), these processes use existing petroleum refining hydrotreatment processes to convert triglycerides to so-called “renewable” diesel—a hydrocarbon, not an ester mixture—essentially identical to petroleum-based diesel. Renewable diesel from these processes has better cold weather characteristics than FAME biodiesel, and glycerin is not produced as a side product. A number of firms, including US-based ConocoPhillips and UOP, (a Honeywell subsidiary) and Brazilian oil giant Petrobras, have demonstrated the technology at existing petroleum refineries. In the United States, Syntroleum and Tyson have formed Dynamic Fuels LLC, a partnership that is building a 284 ML (75 million gallon, MG) per year facility in Geismer, LA, to produce renewable diesel and jet fuel, using animal fat feedstock. Neste Oil (Finland) has announced plans to build Europe’s largest renewable diesel plant in Rotterdam, NL, using plant oil and animal fats. Renewable diesel production facilities will likely be larger than FAME biodiesel facilities, in order to achieve production scale benefits from the more complex process technology. Chemical Products Significant opportunities exist for conversion of plant oils to biobased chemical products, which generally command higher value than liquid fuels. Within the existing oleochemical industry, commercial processes are well established for hydrolysis of the triglyceride molecule to afford glycerin and a range of fatty acids for further processing. These mature oleochemical products will not be reviewed here. Fatty acids from plant oils are generally multifunctional, containing both olefinic and acid functionality, which makes them particularly useful as chemical intermediates for new polymers and other complex molecules with desired properties. Recent growth of the biodiesel industry has also resulted in a global excess of glycerin and the opportunity to develop valueadded derivatives from this abundant biofuel coproduct. While most commercial transformations of fatty acids are directed toward the acid group (Bozell, 2006), a partnership between Cargill and Materia named Elevance Renewable Sciences

P1: SFK/UKS P2: SFK c12 BLBS075-Hood December 15, 2010

288

20:57

Trim: 244mm×172mm

Plant Biomass Conversion

(Illinios, USA) is applying new olefin metathesis technologies to the sites of unsaturation in feedstock oils to produce a range of biochemicals and waxes (Thayer, 2009). As coproduct glycerin became available with the biodiesel boom of the last 5 years, a number of commercial projects have been announced for conversion of crude glycerin to higher value products, including propylene glycol, epichlorohydrin, glycerol carbonate, propylene, and methanol. More recent uncertainty concerning the longer-term supply and pricing of coproduct glycerin has resulted in cancellation or delay of many of these projects (McCoy, 2009). Archer Daniels Midland (ADM) has announced construction of a 100,000 MT plant in Decatur, IL, which will use a proprietary hydrogenolysis process to convert glycerin and sorbitol into propylene glycol with a start-up date in the first quarter of 2010. Dutch firm BioMCN has initiated production of methanol via gasification of glycerin at the pilot scale with a 200,000 MT per year facility scheduled for start-up in 2009 in Delfzijl, Netherlands. Belgian-based Solvay Chemicals introduced its EpicerolTM glycerin-to-epichlorohydrin process with a pilot facility in Tavaux, France, in 2007 and projects start-up of a 100,000 MT facility in Thailand in 2011. A similar project for United States-based Dow Chemical is on hold. Epichlorohydrin is a key reactant in the preparation of diglycidyl ethers (e.g., diglycidyl ether of bisphenol A) for production of epoxy resins, widely used in the coatings industry. Biobased epichlorohydrin would allow incorporation of biogenic content into this important line of resins. Vegetable oils such as soybean oil can be chemically modified to produce epoxidized oils for use as plasticizers or in coatings. Epoxidized soybean oil (ESO) and epoxidized linseed oil (ELO) are well established products, supplied by firms such as Chemtura (United States) and Arkema (France), which provide heat and light stability for polymers such as polyvinyl chloride (PVC) (Bergstra, 2007). Recent regulatory pressures on synthetic phthalate plasticizers may make these biobased products even more attractive alternatives. Also it is possible to chemically introduce multiple hydroxyl groups at sites of unsaturation in fatty acids, resulting in natural oil polyols (NOPs), which can then serve as monomers to produce polymers with biogenic carbon content, the most significant of which is polyurethane (described later in this review). Several companies have commercialized biobased polyols, including United States-based Cargill, under the BiOHTM brand; Urethane Soy Systems of Volga, SD; and BioBased Technologies of Rogers, AR. Products based upon chemically unmodified natural oils are also important in some applications, especially environmentally sensitive uses, where biobased content, biodegradability, and/or low-toxicity are desired. Such vegetable-oil-based products have been developed for diverse markets, including inks, lubricants, emulsifiers, and cosmetics. In particular, vegetable oil lubricants compete with mineral-based oils in a number of application areas. The United States-based United Soybean Board has developed a Market Opportunity report that summarizes soy-based lubricant and surfactant opportunities (Rust and Wildes, 2009). The lubricant report describes vegetable oil lubricant advantages as enhanced lubricity, lower evaporation loss, and higher viscosity index, while noting performance limitations such as thermal, oxidative, and hydrolytic stability. Performance issues can be addressed by chemical modification of the feedstock oils—most commonly hydrogenation—or by formulation with additives and stabilizers. However, recent R&D efforts have also focused on development of enhanced oil output traits in several crops to provide improved feedstock chemical composition, such as high oleic acid sunflowers and soybeans and high erucic acid rapeseed. Continued genetic modification of oilseed crops to produce specific fatty acid profiles and specific levels of fatty acid unsaturation promises to produce a new generation of specialty oilseed feedstocks, specifically tailored to the downstream application (Schmidt et al., 2006).

P1: SFK/UKS P2: SFK c12 BLBS075-Hood December 15, 2010

20:57

Trim: 244mm×172mm

12 Biobased Chemicals and Polymers

289

Table 12.4. Chemical products made from carbohydrates. No. of Carbon

Products

C2 C3 C4 C5 C6

Ethanol, acetic acid, glycolic acid Lactic acid, propanediol, acrylic acid, 3-hydroxypropionic acid Aspartic acid, butanol, methyl ethyl ketone (MEK), succinic acid and butanediol Glutamic acid, monosodium glutamate, itaconic acid, levulinic acid Citric acid,hydroxymethyl furfural, gluconic acid, sorbitol, lysine, glucaric acid

12.4.2

Sugar/Starch-Based Products

The growth of the petrochemicals industry during the twentieth century was dependent on the development of a variety of chemical catalysts able to transform crude oil into diverse petrochemical intermediates. In the same way, the development of the sugar platform will depend on the application of chemical catalysts and biocatalysts that can transform sugars into useful biobased chemical intermediates and downstream products. Two comprehensive reports have detailed the potential to produce biobased chemicals predominantly from sugar feedstocks: the “Top Value-Added Chemicals Report” produced by NREL and the Pacific Northwest National Laboratory for the Department of Energy (Werpy et al., 2004); and the “BREW Project Report” produced by a collaboration of academic and private-sector partners for the European Commission (Patel et al., 2006). Both reports have documented the potential to produce important platform or building block chemicals from chemical and biochemical processing, to replace current petrochemical products. Some examples of chemical catalysis are known, such as the transformation of glucose into sorbitol and HMF. A unique technology being developed by Virent Energy Systems (Madison, WI) uses a sophisticated combination of chemical catalysis steps known as Aqueous Phase Reforming to produce hydrocarbons from mixtures of C5 and C6 sugars. Development of new biocatalysts is a very active area of R&D within academia and the private sector. While many biocatalysts have been successful in proof-of-concept experiments on a laboratory scale, commercial industrial processes are only beginning to emerge. Biocatalysts include natural and genetically modified yeasts, bacteria, and fungi, which are comprehensively reviewed elsewhere within this book (see Chapter 8). Table 12.4 summarizes key products made from sugars on a commercial or precommercial scale using chemical and/or biocatalytic processes, as well as selected products at an advanced development stage. Products are categorized by the number of carbon atoms (i.e., C2, C3, C4) with the leading technology developers described below. Commercial realities dictate the use of the most economical readily available source of carbohydrate feedstock for these processes—at present generally corn, cassava starch, or molasses. It is expected that many of these processes will transition to the use of lignocellulosicderived sugars, when they become available at a competitive price and quantity.

C2 Products Ethanol produced for liquid transportation fuel is the primary C2 biochemical product. A 2-carbon alcohol derived from sugar/starch feedstocks, ethanol has been produced by yeast fermentation technologies for thousands of years. However, during the twentieth century, this industry lost ground to the petrochemical synthesis route. With escalating crude oil prices over

P1: SFK/UKS P2: SFK c12 BLBS075-Hood December 15, 2010

290

20:57

Trim: 244mm×172mm

Plant Biomass Conversion

the last several years, a global fuel ethanol industry has emerged, notably in Brazil and the United States, providing this renewable nonpetrochemical-based liquid transportation fuel by fermentation of sugar obtained from sugar cane, corn, sugar beets, sweet sorghum, and other sugar/starch crops. Brazil has developed a mature ethanol industry based upon sugarcane, while in the United States, over 98% of fuel ethanol production is from corn (S´anchez and Cardona, 2008). It is now widely accepted that the fuel ethanol industry is not sustainable using only starch and sugar raw materials, due to demand on these crops for food supply as well as high production costs. The market is now embracing lignocellulosic feedstocks as a solution, using agricultural and wood wastes, short rotation energy crops, and waste from wood pulping industries to generate carbohydrates for second-generation ethanol production. Second-generation ethanol derived from lignocellulosic feedstocks eliminates the food-fuel issue and has also been shown to have a much more favorable net energy balance and lifecycle GHG reduction than starchbased ethanol. The sugar/starch-based fuel ethanol industry and the emerging lignocellulosic biochemical ethanol technology developers are reviewed elsewhere in this book (see Chapter 11). A second C2 product accessible by biochemical processing is glycolic acid, an α-hydroxy acid, which is used in skin care products. French-based Metabolic Explorer/Roquette is developing this product, which is still in its early stages.

C3 Products There are four main chemicals in the market today generated from carbohydrates that have three carbon atoms: lactic acid, propanediol, acrylic acid, and 3-hydroxypropionic acid. Probably, the most well known is lactic acid, which is used in the production of polylactic acid (PLA), a significant new biodegradable plastic described later in this review. A number of largescale producers of lactic acid include PURAC, NatureWorks LLC, Myriant Technologies, and BioSpark. NatureWorks (a Cargill subsidiary) has its manufacturing base in Blair, Nebraska, and uses refined dextrose obtained from corn for fermentation to lactic acid. PURAC, based in the Netherlands, has a 100,000 MT per year facility in Thailand, a demonstration lactide facility in Spain, and plans to add another lactic acid facility to their operation in the Netherlands. PURAC holds a license for biocatalysts developed by Myriant Technologies (Quincy, MA), a company created by Bioenergy International LLC in June 2009, to incorporate its biobased chemicals business and IP. Myriant Technologies LLC has developed a fermentation process that uses biocatalysts to generate D(-) Lactic acid, which enhances the thermal properties of PLA blends. PURAC licensed this technology in 2008 from Myriant and has also started production. BioSpark is a newcomer to this group formed from a Cleantech venture in 2009 between Imperative Energy Ltd. and Sustainable BioPolymers Ltd. This company is planning to construct a number of facilities to be built in Claremorris, Co. Mayo, Ireland, over the next 2 years, including a bioprocessing facility, wood pellet production plant, and a combined heating and power plant. The bioprocessing facility will use straw and wood biomass to generate lactic acid, ethanol, lignin, methane, and hydrogen. The second three-carbon chemical product—propanediol—can be produced in two isomeric forms: 1,3-propanediol (trimethylene glycol) and 1,2-propanediol (propylene glycol). 1,3propanediol is currently made in two ways—from corn syrup fermentation with a genetically modified E. coli strain and secondly from glycerin (a by-product of biodiesel production) using

P1: SFK/UKS P2: SFK c12 BLBS075-Hood December 15, 2010

20:57

Trim: 244mm×172mm

12 Biobased Chemicals and Polymers

291

the bacterium Clostridium diolis. In 2004, DuPont (Wilmington, DE) and United Kingdombased Tate & Lyle formed a joint venture to produce biobased 1,3-propanediol (PDO), trade named SusterraTM , from corn starch using an aerobic fermentation process that was developed in collaboration with Genencor (a division of Danish Danisco A/S). Start-up of a 50 M kg/year facility followed in 2006, using corn starch feedstock at the Tate & Lyle wet milling site in Loudon, Tennessee. SusterraTM is marketed as an industrial solvent and intermediate into numerous market applications. Metabolic Explorer is also developing both 1,3-propanediol (using glycerin) and 1,2-propanediol (via fermentation). The development of the last two C3 chemical products, acrylic acid and 3-hydroxypropionic acid, is still in its infancy. OPX Biotechnologies (Boulder, CO), who engineered microbes to produce acrylic acid using sugars at laboratory scale, is now scaling up the process. Cargill/Novozymes (Denmark), Arkema/hte, and Novomer Inc. (Waltham, MA) have recently made announcements that they will also move into this area. Cargill/Novozymes will produce acrylic acid via 3-hydroxypropionic acid (3-HPA) using fermentation, while Arkema/hte is screening catalysts to convert glycerin to acrylic acid and acrolein. 3-Hydroxypropionic acid is an important intermediate for generating many commodity and specialty chemicals. A collaboration between Cargill and Novozymes is seeking to develop a bioengineered organism to convert glucose and other carbohydrates to this significant intermediate chemical. Also, Novomer is researching production of 3-HPA from ethylene and recycled carbon monoxide.

C4 Products Butanol dominates the four-carbon atom product group derived from carbohydrates, with interest focused on its superior attributes as a second-generation liquid transportation fuel, compared to ethanol. It can also be used as a chemical intermediate for the production of butyl acetate and other butyl esters. Dupont and energy giant British Petroleum (BP) have collaborated on biofuels development since 2003, announcing in 2006 the joint commercialization of biobutanol using process technology licensed from Praj Industries of India. The initial plant in Norfolk, UK, has been configured to use a range of sugar feedstocks and will be followed by a larger demonstration plant that will be sited on BP’s industrial site at Kingston upon Hull in the United Kingdom. Because of the significant R&D activity spurred by potential fuel applications, a comprehensive review of biobutanol development and commercialization activities is provided separately within this edition. The important 1,4-diacid succinic acid is another C4 chemical moving rapidly to full commercial status from renewable feedstocks. It is used as an intermediate for the production of adhesives, pharmaceuticals, and other products. The development landscape for this chemical is dominated by six main companies—Myriant Technology, BioAmber, DSM/Roquette, PURAC/BASF, and Mitsubishi Chemical (Japan)—at various stages of commercial demonstration. Processes are characterized by different microorganisms and biocatalysts, and accelerating technology improvements. Myriant Technology, BioAmber, and Royal DSM/Roquette all have plants due to be finished at the end of 2009. Myriant Technology has partnered with United States-based Buckeye Technologies and the University of Florida to construct a pilot plant in Florida, USA, which will produce both cellulosic ethanol and several biobased chemicals, two of which are fourcarbon products. The first is succinic acid, to be used as an intermediate to make 1,4-butanediol. Bioamber, a joint venture of French ARD and US DNP Green Technology, commissioned its initial demonstration facility in Pomacle, France, at the end of 2009. Royal DSM/Roquette

P1: SFK/UKS P2: SFK c12 BLBS075-Hood December 15, 2010

292

20:57

Trim: 244mm×172mm

Plant Biomass Conversion

also has their demonstration facility underway in France, targeting production of succinic acid from starch using enzyme-based fermentation technology. A commercial scale facility is planned to come online in 2011/12. In 2009, PURAC/BASF and Mitsubishi Chemical have also announced development programs for the production of succinic acid. Aspartic acid is an amino acid widely used in the food industry for a synthetic sweetener (aspartame), as well as the pharmaceutical and chemical industry. It can also be converted to polyaspartic acid, which is a polymer used in numerous industrial applications. NanoChem Solutions (a subsidiary of Flexible Solutions International), which produces protein-based polymers from aspartic acid, opened a facility in Alberta, Canada, in September 2009, which ferments sugar beet juice to create aspartic acid. A sister company in Chicago converts the acid to its polymer. The plant’s capacity is 5,000 tons, but it has been equipped to expand three-fold in the future. The development of two other C4 products from sugar feedstocks, methyl ethyl ketone (MEK) and butanediol, is still at a very early stage. A leading company is Genomatica (San Diego, CA), which is working on both products and planning to commence construction on a butanediol demonstration plant to start construction at the end of 2009. MEK development work is progressing, with the potential to retrofit existing ethanol facilities for its production. C5 Products Of the five-carbon products, the most easily derived is levulinic acid as it can be made from almost all sugars. It is also a useful starting material for the generation of a wide range of compounds including butanediol, diphenolic acid, succinic acid, and tetrahydrofuran, as described in the “Top Value-added Chemicals Report” (Werpy et al., 2004). Biofine Renewables (Waltham, MA) was the first to produce levulinic acid from biomass. The company was formed in 1998 and, with support from the US DOE, constructed a demonstration plant in New York in 1994, which has operated for 10 years. The group uses a high temperature, dilute acid hydrolysis of waste cellulosic materials to produce the acid. In 2006, the company licensed its technology to “Le Calorie” a subsidiary of the Italian construction firm Immobligi, which commissioned a 3,000 MT per year facility in Caserta, Italy, in 2006. The facility uses both tobacco bagasse and paper mill sludge as feedstocks and also generates other coproducts including lignin (used for process energy), formic acid, and furfural. Xylitol, a sugar alcohol used as a low-calorie sweetener, is commercially produced by hydrogenation of xylose (obtained from birch wood sulfite pulping liquor and other xylan-rich substrates) or it can be derived from corn husk fibers. The majority of xylitol is produced by four main groups: Futaste Pharmaceutical Company, Ltd. (China), French-based Cerestar (a Cargill subsidiary), Danisco, and Roquette Frere (France), with the remaining companies distributed among Japan, China, and Switzerland. An alternative technology based on microbial reduction of xylose from xylan-rich hydrolysates is considered to be cleaner and requires less energy than the chemical route. Furfural (furan-2-carboxaldehyde) is a C5 aromatic aldehyde, which can be used as an intermediate for the production of solvents. Furfural is generated via the acid-catalyzed dehydration of xylose and so is often coproduced with other materials from biomass feedstocks. China currently is the largest furfural producer worldwide. United States-based Raven Biofuels International is notable for its intent to produce furfural alongside other products, including ethanol, lignin (for process energy), organic acids, aldehydes, and esters. Itaconic and glutamic acids are other key C5 chemical building blocks that can be used for commodity and specialty chemicals, and five-carbon polymers, respectively. Itaconic acid can

P1: SFK/UKS P2: SFK c12 BLBS075-Hood December 15, 2010

20:57

Trim: 244mm×172mm

12 Biobased Chemicals and Polymers

293

be produced by fungal fermentation and is currently supplied in small volumes at relatively high prices from various producers, with China notably moving into a leading position. Monosodium glutamate, the sodium salt of glutamic acid, is another five-carbon product that is produced by the fermentation of starch, sugar beets, sugar cane, or molasses. Japanese producer Ajinomoto dominates this industry with over 33% of the world’s production (Farrell, 2008). C6 Products Several of the C6 products have been produced from sugar feedstocks for many years, including citric acid, gluconic acid, HMF, and lysine. Citric and gluconic acids are produced by fermentation using strains of Aspergillus niger. Citric acid is the largest produced organic acid, and it is widely used as an acidulant and preservative in the food and beverage industry. Producers include ADM, Cargill, Tate & Lyle, Gadot Biochemical Industries (Israel), Anhui BBCA Biochemical (China), and PMP Fermentation. The gluconic acid market is somewhat smaller, though it is also used as an acidulant in foods. The leading producer of gluconic acid is PMP Fermentation (a subsidiary of Japan-based Fuso Chemical Company). HMF is another organic compound that has been derived via the dehydration of sugars (especially fructose). This compound is an important building block for fuels and replacement of compounds used in the production of plastics. Avantium (Amsterdam, NL), a spin-off from Royal Dutch Shell, is exploring biofuels and chemicals based on HMF derivatives (especially HMF esters). They use chemical catalysts to derive specific derivatives, termed “Furanics” (Imhof et al., 2008), using different sugars (e.g., sucrose or glucose). Lysine is an α-amino acid used as an animal feed additive and in pharmaceuticals, derived from carbohydrates via fermentation using strains of Corynebacteria (for industrial production). ADM entered lysine manufacturing in 1989. Other C6 products to highlight include sorbitol, sodium erythorbate, glucaric acid, and other aromatics such as caprolactam, phloroglucinol, shikimic acid, and resorcinol. Sorbitol is an important chemical intermediate as it can be converted to alkanes, which can be used as biofuels. It is mainly used as an artificial sweetener in the current market and in the production of vitamin C. The major producers of sorbitol include ADM, Swiss-based Lonza, and Roquette, which all have their production plants in Illinois, and SPI Polyols with a plant in Delaware. Sodium erythorbate is an antioxidant used in the food industry that is produced by the fermentation of corn sugar. PMP Fermentation (a subsidiary of Fuso Chemical Company) produces this chemical along with their other C6 products, citric acid and gluconic acid. Glucaric acid and its derivatives are also important C6 chemical intermediates. United States-based Rivertop Renewables, founded in 2008, is pursuing commercial development of this chemical. Finally, the C6 aromatics (caprolactam, phloroglucinol, shikimic acid, and resorcinol) represent an important group of intermediate chemicals derivable from fermentation technologies. Draths Corporation (Plymouth, MI) is an early-stage development company that was issued a US patent in 2008 for the conversion of lysine to caprolactam and is pursuing the production of aromatic monomers from biobased sources.

12.4.3

Polymer Products

Fossil-based polymers and plastics are ubiquitous to modern life with applications across many market segments, where performance and permanence are required. Synthetic polymers are

P1: SFK/UKS P2: SFK c12 BLBS075-Hood December 15, 2010

294

20:57

Trim: 244mm×172mm

Plant Biomass Conversion

Table 12.5. Routes to biobased polymers. Conversion Route

Key Examples

Modification of natural polymers

Starch polymers, cellulose polymers, lignin

Production by microorganisms or crops

Polyhydroxyalkanoates, transgenic enzymes

Polymerization of biobased monomers

Polylactic acid, polyamides, polyesters, polyethylene, polyurethanes

typically inexpensive, durable, and can be manufactured to exhibit a broad range of properties. Generally speaking, most polymers command a price premium compared to commodity fuels and chemicals, but represent a lower product value than specialty chemicals. Fossil-based polymers require multiple sequential processing steps that first introduce functionality (unsaturation or heteroatom functional groups) to the monomeric unit(s), followed by controlled polymerization to afford macromolecules, and finally postpolymerization processing (purification, classification, formulation, etc.) to achieve the desired product properties. Each synthetic or processing step adds cost—and value—to the ultimate polymer product. Many important fossil-based polymers were commercially introduced in the middle of the last century, post WWII, so that today’s processes and products are well-optimized and supply chains fully established. Natural polymers are an inherent component of living things. Prior to development of modern synthetic chemistry and the global petrochemical economy, man used wood, fibers, fruits and grains, and other plant and animal biopolymers for both food and materials. Today, only 3.5% of the 170 trillion tonnes of biomass produced annually on the planet is used by mankind (mostly for food, construction, and energy), with only 5% of that total—300 million tonnes—used for nonfood applications such as chemicals and fibers (Shen et al., 2009). Despite their historical precedent, natural biopolymers have been largely replaced by synthetic petrochemical-based polymers for most materials applications in developed nations. However, the supply, environmental, and economic considerations surrounding petrochemical feedstocks—coupled with advancing technologies—has prompted a renewed interest in commercially viable biobased polymers derived from sustainable feedstocks. This section will serve to describe current commercial efforts to introduce biobased polymers, broadly defined as macromolecules incorporating some biogenic carbon, to the global marketplace. These biobased polymers may be classified based upon their production by one of three distinct routes as shown in Table 12.5. A recent review (Shen et al., 2009) has provided a comprehensive summary of biobased polymer commercialization status and projections. This section will provide an overview of the key production routes and examples of the leading commercial projects.

Modification of Natural Polymers Chemically unmodified natural polymers such as fibers, wood, and natural rubber have been used by man for centuries. In more recent times, chemical and biochemical modification of biogenic macromolecules has led to biobased polymers with enhanced properties for expanded product applications. Because of their biobased polymer content, these products often contain a very high proportion of biogenic carbon. Key examples include the following.

P1: SFK/UKS P2: SFK c12 BLBS075-Hood December 15, 2010

20:57

Trim: 244mm×172mm

12 Biobased Chemicals and Polymers

295

Cellulose Polymers Chemical modification of cellulose to produce resins, plastics, and fibers is perhaps the most mature commercial biopolymer industry, dating back to the mid 19th century development of cellulose nitrate for photographic films and fibers. Today, commercial cellulose-derived polymers fall into three categories: viscose (rayon), cellulose esters, and cellulose ethers. Manmade cellulosic fibers, notably viscose and cellulose acetate, have been developed as synthetic alternatives to natural fibers such as cotton and wool and represent the largest commercial volume biobased polymers, with 2008 production of 3.5M MT by a number of global producers (http://www.ivc-ev.de/). Cellulose acetate and cellulose butyrate and propionate “mixed esters” are produced by acid-catalyzed reaction of cellulose with the respective anhydrides. Cellulose acetate fiber is produced globally for cigarette filter tow and textiles, while mixed esters for film, coatings resins and high-end plastics are specialty products with more limited producers, notably United States-based Eastman Chemical Company and Celanese. Commercial cellulose ester production requires economy of scale and efficient spent acid recovery and regeneration. Cellulose ethers are produced by reaction of alcohols with the cellulose hydroxyl groups to afford products for diverse markets including food, personal care, coatings, and construction. The industry is characterized by producers in all geographies, including Akzo Nobel (Amsterdam, NL), Dow Wolff Cellulosics (subsidiary of United States-based Dow Chemical), Shin-Etsu Chemical (Tokyo, Japan), and SE Tylose GmbH & Co (Wiesbaden, Germany subsidiary of Shin-Etsu). Starch Polymers Historically, starch has been used extensively in processed food & feed products, as a fermentation feedstock (notably for ethanol), and for such nonfood applications as textile/paper sizing, adhesives, and viscosity modification. Within the last two decades, commercial technologies have been developed and deployed to modify native starch by physical, chemical, and fermentative processes to an expanding portfolio of starch-based polymers with enhanced properties, allowing penetration of various petrochemical polymer market applications. Blends with other biopolymers or petrochemical polymers further extend the range of properties for starch polymers, which now find key applications in packaging, films, and most recently durable plastic components. Starch polymers and blends have become the largest class of new commercial biopolymers, with 2007 global capacity of 0.17M MT, 75% of which is located in Europe (Shen et al., 2009). Major producers include Novamont (Italy), Biotec (Germany), Rodenburg Biopolymers (Netherlands), Cereplast (United States), and Biograde Limited (Australia). Lignin Because of the global abundance of lignocellulosic feedstocks, the unique aromatic character of the lignin biopolymer, and the commercial potential of lignin-derived products, this unique biopolymer is discussed separately within this review. Production by Microorganisms or Crops Polyhydroxyalkanoates (PHAs) PHAs are polyesters produced naturally by bacteria from renewable sugars or fats, to store carbon and energy in their cells. Since the 1970s, a number of companies have pursued industrial processes to optimize fermentation conditions for polymer growth within the microorganism

P1: SFK/UKS P2: SFK c12 BLBS075-Hood December 15, 2010

296

20:57

Trim: 244mm×172mm

Plant Biomass Conversion

and the subsequent extraction and purification of PHA polyesters. PHAs are thermoplastic resins, can be produced with a broad range of properties, and are biodegradable, making them suitable for a large range of applications. Blends of PHAs with fillers and copolymers (biogenic and fossil-based) can further expand the properties and range of applications. Commercial supply of PHAs is at an early stage, although pilot quantities have been available from several suppliers for about 5 years. In 2007, Metabolix and ADM formed the joint venture Telles to commercialize MirelTM branded PHAs. A 50,000 MT annual capacity production facility under construction adjacent to ADM’s wet corn mill in Clinton, Iowa, was targeting initial commercial supply in late 2009. When operational, this will represent the largest installed global capacity for PHAs. Other producers include Tianan Biological Material Co. (JV with Dutch-based DSM) and Tianjin Green BioSciences in China, Kaneka Corp. in Japan, and PHB Industrial in Brazil (Shen et al., 2009). Advanced R&D efforts are focusing on genetically modified organisms to directly produce biogenic polymers, as well as small molecules. As an example, in November 2009, the Korea Advanced Institute of Science and Technology (KAIST) announced the development of an E. coli strain capable of producing PLA and copolymers through direct fermentation (Jung et al., 2009). Crop-Based Transgenic Polymers A potentially lower-cost route to PHAs is genetic modification of plants to directly produce PHA polymers. In collaborative work co-funded by the US Department of Energy, Metabolix (Cambridge, MA) is pursuing genetic modification of switchgrass to allow it to produce PHAs, which would then be extracted from the plant material and processed to obtain desired material properties (Carole et al., 2004). In addition, Metabolix has demonstrated the production of up to 5% PHA in genetically engineered tobacco. The residual plant material remaining after the PHA extraction could be used to produce fuels, power, or other products. Other leading research efforts are aggressively developing transgenic plants to produce polymeric materials, as well as highly functionalized chemicals. As an example, the production of transgenic enzymes is profiled elsewhere in this volume. Polymerization of Biobased Monomers Polylactic Acid (PLA) Polylactic acid, also referred to as polylactide, is perhaps the most commercially significant new biogenic polyester. While known for over a century, PLA has received new attention over the last two decades, partly due to its potential as a biodegradable plastic to replace more persistent fossil-based polymers. Lactic acid produced by fermentation of corn starch or other sugars is polymerized, either through the intermediate dimer lactide or directly, to obtain PLA. Since biogenic lactic acid and the resulting lactide dimer are chiral molecules, blends of enantiomeric PLA polymers can be used to extend the range of properties of the final polymers. In general, PLA plastics exhibit many properties of synthetic thermoplastics, with the notable exception of temperature stability (low glass transition temperature), which is a key characteristic for biodegradability. In 2002, a joint venture of Cargill and The Dow Chemical Company began production of PLA polymer, at a 140,000 MT plant in Blair, Nebraska. In 2005, Cargill acquired Dow’s interest and renamed the venture NatureWorks LLC, which subsequently formed a new 50/50 partnership with Teijin Limited of Japan in 2007 to supply the IngeoTM

P1: SFK/UKS P2: SFK c12 BLBS075-Hood December 15, 2010

20:57

Trim: 244mm×172mm

12 Biobased Chemicals and Polymers

297

branded polymer. In July 2009, Cargill acquired full ownership of NatureWorks from Teijin. NatureWorks uses corn grain as the glucose source for PLA production, but is also developing organisms to convert pentose sugars, derivable from lignocelluloses, into PLA (Carole et al., 2004). Dutch-based PURAC, the world’s largest producer of natural lactic acid, lactides, and lactates, is moving into downstream production of PLAs within their PURASORBTM product lines. Dutch-based Synbra has announced a 5,000 MT per year plant in the Netherlands, using technology jointly developed by PURAC and Sulzer, for a biodegradable foamed PLA product targeted for expanded polystyrene (EPS) markets. Futerro, a 50/50 joint venture between Galactic and Total Petrochemicals formed in 2007 has announced a 1,500 MT per year PLA demonstration unit in Belgium, with the mission to license and expand the resulting PLA technologies. Pyramid Bioplastics, a joint venture of German Bioplastics GmbH and Swiss Pyramid Technologies has announced a 60,000 MT/year plant in Guben, Germany, to be online by 2012. Separate from Cargill, Teijin has noted its continued interest in expanding sales of its BIOFRONTTM heat-resistant PLA fiber. Several other potential commercial producers operate or have announced pilot scale and demonstration units, including projects in China.

Polyamides Polyamides, more generally known as nylons, find commercial applications from fibers to engineered plastics. The amide functionality is incorporated within a polymer by reaction of a diamine with a dicarboxylic acid or by polymerization of amino acids or lactams (cyclic amides). A range of nylon products are available commercially from a large number of producers in all geographic regions of the world. Three products—PA11, PA610, and PA1010—are known to contain biogenic content, resulting from monomers produced from castor oil (Shen et al., 2009). PA11 is produced by Arkema from 11-aminoundecanoic acid while PA610, offered by German chemical giant BASF, Toray (Japan), and Dupont, incorporates 1,10-decanedioic acid (sebacic acid). In addition, DuPont has introduced PA1010, a condensation product of decamethylene diamine with sebacic acid, which is claimed to contain 100% renewably sourced content. The biogenic amino acid and diacid monomers are produced by conventional chemical transformations of ricinoleic acid (or its ester) derived from castor oil. In the future, the greatest potential to incorporate significant biogenic content in commercial nylons could be the development of commercially viable processes for biobased 1,6-hexanedioic acid (adipic acid) and caprolactam, both high-volume monomers for global nylon production.

Polyesters Polyesters represent an important commercial class of polymers, produced almost entirely from purified monomers (generally diols/diesters and diacids) derived from fossil feedstocks. Described earlier, PDO can be reacted as a monomer with terephthalic acid or dimethyl terephthalate, analogous to the production of polyethylene terephthalate (or PET) to produce poly(trimethylene terephthalate) or PTT, a polymer exhibiting the desirable properties of both PET and polybutylene terephthalate (PBT). In 2004, PTT Poly Canada, a joint venture including Shell Chemical, started production in Montreal, Canada, of its branded CorterraTM petrochemical-based PTT. In 2000, DuPont began production and distribution of its SoronaTM brand PTT, containing bio-PDO, resulting in a polymer containing 37% renewably sourced content by weight. PTT polymers are being marketed by Shell and DuPont for fiber, film, and engineered plastic applications.

P1: SFK/UKS P2: SFK c12 BLBS075-Hood December 15, 2010

298

20:57

Trim: 244mm×172mm

Plant Biomass Conversion

In 2005, Ashland Specialty Chemicals Co. (Dublin, OH) introduced four versions of ENVIREZTM soy-based thermoset polyester resins, which were followed in 2008 by two unsaturated polyester resins incorporating renewably sourced propanediol. Other Monomers As additional biobased monomers become commercially available, the opportunity will increase to produce polyesters or other polymers with partial or even total biogenic content. Biobased 1,4-butanediol (BDO) represents a key diol that could substitute for its petrochemical analog in several commercial polyesters. Similarly, biobased ethylene is a precursor for ethylene glycol, which could be utilized in existing polyesters, notably PET, to incorporate biogenic content. India Glycols Limited claims to be the only producer in the world of biobased ethylene glycol, from sugar-based ethylene through ethylene glycol. In May 2009, CocaCola Company announced its intention to introduce PET water bottles containing up to 30% biobased content, deriving from biogenic ethylene glycol from unnamed sources. Likewise, succinic acid and perhaps adipic acid are leading diacid candidates for commercialization, which could also be incorporated into biobased polymers in the near future. Segetis (Golden Valley, MN) is developing patented technology to combine levulinic acid esters and biobased hydroxyl compounds to create new bi-functional or “binary” monomers, as precursors to proprietary polymers. Biobased Polyethylene (PE) and PVC Polyethylene, produced by polymerization of fossil-based ethylene, is a commodity polymer that is manufactured with a wide range of physical properties for diverse end-use applications including packaging, bottles, and pipes. Over 60 MM tons/year are produced globally. In 2007, Braskem SA, Latin America’s largest petrochemical company announced plans to build a 200,000 ton PE facility in Brazil, utilizing ethylene produced from sugar-derived ethanol. This was followed by announcement of a joint venture between Dow Chemical Company and Brazilian ethanol producer Crystalsev to construct a 350,000 MT PE facility in Brazil, also using biobased ethylene. Bioethylene is produced by catalytic dehydration of ethanol, and the resulting PE has the same properties and performance as fossil-based polyethylene. In its announcement, Dow has commented that biobased PE project economics will be competitive with hydrocarbon-based projects, but is dependent upon low-cost sugar-derived ethanol feedstocks. Tetra Pak, a Swiss food packaging producer, has announced its intention to source 5,000 MT/year of “green” high-density polyethylene from Braskem for carton closures. Ethylene is also used in the production of PVC, a versatile commercial polymer with significant construction applications. PVC is produced by polymerization of vinyl chloride monomer (VCM), derived from ethylene and chlorine. Solvay Indupa is the world’s first biobased PVC facility in Santo Andre, Brazil, which will utilize ethylene produced from sugar cane ethanol. Polyurethanes Polyurethanes (PUR) are an important class of mature polymers produced by the reaction of an isocyanate and a polyol, producing a recurring urethane functional group in the polymer. PURs can be produced with a range of properties, but applications generally fall into

P1: SFK/UKS P2: SFK c12 BLBS075-Hood December 15, 2010

20:57

Trim: 244mm×172mm

12 Biobased Chemicals and Polymers

299

flexible/rigid foams and coatings/adhesives/elastomers. To date, the commercially available isocyanate monomers have all been derived from petrochemical feedstocks, with aromatics (e.g., toluene diisocyanate) dominating the industry. However, an expanding portfolio of polyols is becoming available from biobased feedstocks, notably NOPs. Several biobased polyether and polyester polyols are also used commercially. Carbohydrates such as sucrose and sorbitol, with their inherent hydroxyl functionality, can be reacted with ethylene or propylene oxide to produce polyether polyols. DuPont has recently commercialized a family of biobased polyetherdiols, under the trade name CerenolTM , produced exclusively from its biogenic 1,3-propanediol, resulting in 100% biobased carbon content (Shen et al., 2009). Polyester polyols produced by the condensation of glycol polyols with diacids could also incorporate biogenic content through the use of biobased acids and/or glycols (e.g., propylene glycol or 1,4-butanediol), as these chemicals become commercially available from biobased feedstocks.

12.4.4

Lignin Products

Lignin, accounting for about 30% of the weight of most biomass types, is an abundant polymeric raw material. However, limited product applications for commercial lignin are currently available. These applications generally fall into three categories as described in the DOE Lignin Report (Holladay et al., 2007), prepared by Pacific Northwest National Laboratory: complete destruction of the lignin via combustion or pyrolysis, isolation as oligomeric lignin fragments, or partial degradation to give low-molecular weight or monomeric aromatic molecules such as vanillin and dimethylsulfoxide (DMSO). The first category encompasses the lowest-value, and current primary use of lignin, as a carbon source for fuel. Gasification, rather than combustion, affords the potential to produce higher-value fuels and chemicals from syngas processing. The two other categories envision the production of high-value functionalized aromatic polymers and phenolic chemical products from lignin, which could displace petroleum-based products. The unique chemical nature of lignin, its low toxicity, and its abundance make it an ideal renewable feedstock substitute for a wide range of petrochemical-based materials. The rational exploitation of lignin as a chemical feedstock has been the object of numerous studies (Lindberg et al., 1989; Pye, 2008). The successful commercial utilization of oligoor monomeric lignin, however, has been hampered by the absence of a reliable commercial supply of pure functional lignin materials with uniform characteristics and composition (Rojas et al., 2006). Of the 70 and 3 million tonnes of lignin extracted by the Kraft and sulfite processes, respectively in 2005, only about 0.1% of the Kraft and up to 25% of the sulfite lignin (lignosulfonate) was recovered and sold as a chemical for industrial processes (Voitl and Rohr, 2009). Emerging lignocellulosic biorefineries have the potential to offer a new abundant source of high grade lignin targeted for wider markets and higher value applications. The physicochemical characteristics of any lignin preparation, and its corresponding suitability for use in various applications, vary greatly depending upon the lignin origin and type of delignification method (Glasser et al., 1983). The lignocellulosic biomass source (principally the type and ratio of lignin and hemicellulose subunits) is the initial determinant of the specific properties of the extracted lignin. Secondarily, the delignification chemistry greatly impacts the polydispersity (ratio of weight (Mn ) to number (Mw ) average molecular weight), the presence of functional groups, and the levels of contaminants in the final lignin preparation. In particular, polydispersity and impurities in extracted lignin can significantly influence functional group availability and thermal behavior of the lignin preparation.

P1: SFK/UKS P2: SFK c12 BLBS075-Hood December 15, 2010

300

20:57

Trim: 244mm×172mm

Plant Biomass Conversion

Lignosulfonates The vast majority of the current commercial use of lignin (other than combustion) is in the form of lignosulfonates produced as a by-product of sulfite pulping or via the postsulfonation of Kraft lignin. In 2001, the world consumption of lignosulfonates was about 1 million tonnes (Stewart, 2008). Borregaard LignoTech is the world’s leading supplier of lignosulfonates and other lignin products and has been active in this area for over 60 years. These lignosulfonates are in the form of small oligomers of lignin subunits that are modified with sulfonate groups, making the products water soluble. The largest market for lignosulfonates is as an additive to improve the properties of concrete. Smaller amounts are used in drilling mud, in animal feeds, and for spraying on roads for dust control. Mature markets for lignosulfonates are summarized in a review produced by Ligno Tech (Gargulak and Lebo, 1999).

Kraft and Soda Lignin A very large amount of Kraft lignin is potentially available for use in the manufacture of adhesives, resins, binders, or carbon fiber. However, the inherent variability, molecular complexity, and low reactivity of the lignin found in waste effluent produced by Kraft pulping industry have hindered the exploitation of this highly abundant, renewable chemical feedstock. As a result, lignins in Kraft mill effluents continue to be incinerated during the regeneration of the cooking chemicals in the recovery boiler. Lignin precipitation and recovery from the black liquor using technology such as the Lignoboost process is starting to be implemented in order to increase the capacity of Kraft mills limited by the boiler recovery system (e.g., Innventia’s Nordic Paper Mill in B¨ackhammar, Sweden). At present, such lignin precipitated from the Kraft black liquor is used as fuel in the mill’s combined heat and power (CHP) or lime kiln units. Metso Corporation (Helsinki, Finland), who acquired the Lignoboost process in 2008 will continue to work with Innventia to further the development of the technology for production of higher value lignin products. It is recognized that extending the use of lignin beyond existing pathways will be fundamental to establishing the economic viability of forest-based biorefineries (Canada, 2006). The limited use of Kraft lignin as a chemical feedstock results mainly from its heterogeneity and high level of contaminants (organic sulfur, carbohydrate decomposition products, ash). Lignin precipitated from Kraft black liquor consists of modified condensed or oligomeric lignin fragments with molecular weight varying from 100 to >10,000 Daltons (Brodin et al., 2009; Gundersen and Sj¨oblom, 1999; Pouteau et al., 2005). Soda pulping waste effluent is today the only readily available source of sulfur-free lignin. Lignins precipitated from the Soda black liquor are a complex and heterogeneous mixture with a broad molecular weight distribution (high polydispersity). Similar to Kraft lignin, Soda lignins contain high levels of carbohydrate and ash contamination, making recovery in a usable form commercially difficult. Kraft and Soda lignins exhibit high glass transition temperatures with reduced thermal mobility. The glass transition and decomposition temperatures of the precipitated Soda nonwood lignin are variable and habitually lower than for precipitated Kraft lignin (Mousavioun and Doherty, 2010; Tejado et al., 2007). Often, Kraft lignin has no detectable softening point unless it is fractionated and purified (Brodin et al., 2009). Soda lignins differ from Kraft lignins by their lower molecular weight and hydroxyl content and higher methoxyl content (Ibrahim et al., 2004).. Fractionation of Kraft and Soda black liquor by ultrafiltration has been practiced for the isolation of specific lignins with a narrower range of molecular weight and separation from

P1: SFK/UKS P2: SFK c12 BLBS075-Hood December 15, 2010

20:57

Trim: 244mm×172mm

12 Biobased Chemicals and Polymers

301

organic degradation products (J¨onsson and Wallberg, 2009; Toledano et al., 2010). Ion exchange chromatography is applied to reduce the level of inorganic impurities. The majority of the known Soda or Kraft lignin applications are derived from purified specific fractions of the black liquor. Because of the extensive chemical modification that native lignin undergoes during extraction and fractionation, purified Kraft or Soda lignins also often need to be further upgraded via alternative lignin fractionation chemistries or via modification of lignin products. Several chemical or enzymatic methods for the preparation of lower molecular weight aromatics have been investigated, notably grafting novel reactive groups to lignin or cross-linking to yield higher molecular mass lignin products (Hu, 2002; Jin et al., 2010). Many chemical derivatives of Kraft or Soda lignin have also been investigated. These derivatization schemes include reduction by sodium dithionite, methylolation, phenolation, or reaction with alkylene oxides, esterification, and alkylation (Fox, 2006). None of these lignin derivatives are currently produced on a commercial scale. Fractionated and purified Kraft and Soda lignin have been successfully formulated in phenolic resin-based wood composite adhesives. India-based ALM India Pvt Ltd. is manufacturing up to 10,000 MT per year of lignin-based phenolic resin. Derived from the purified Soda lignin produced from agricultural waste at the ABC paper mill in India, the application of these lignin-based resins include plywood adhesives, high pressure laminates, foundry sand binders, brake pads, and molding compounds. GreenValue (Lausanne, Switzerland) is also developing lignin derivatives as nutritional and health supplements for both animals and humans with successful testing as an antioxidant, an antibacterial agent, a digestibility enhancer, and an immunity booster. Polymeric and Oligomeric Lignins Organosolv-based fractionation technologies offer the most promising potential for commercial production of value-added lignin. Lignin recovered by simple precipitation from the Organosolv black liquor is relatively pure, chemically unaltered, with low polydispersity, and is highly reactive in comparison to lignin precipitated from the Soda, Kraft, or other alkaline black liquors. Organosolv lignins are characterized by good thermal flow with lower glass transition temperature than alkali lignin (Lora et al., 1989). Wood lignin precipitated from a 50% ethanol black liquor has a molecular weight varying between 1,000 and 3,000, a higher ratio of phenolic to aliphatic hydroxyl groups and a higher methoxyl content than in Soda lignin (Hage et al., 2009; Kubo and Kadla, 2004; Pan et al., 2005). A variety of relatively pure Organosolv lignins are produced when different organic solvents are used in combination with different operating conditions. When extracted using solvents with higher polarity, such as methanol, wood lignin has similar characteristics to wood ethanol Organosolv lignin (Botello et al., 1999; Chum et al., 1999). Lignin extracted from wheat straw with acetic acid, ethanol, or methanol-based solvents has a higher molecular weight (over 4,000) but quite narrow polydispersity (1–2) (Xu et al., 2006). Larger amounts of carbohydrate are found in lignin prepared from organic acid-based solvent (acetic or formic acid processing). The ethanol Organosolv Alcell lignin produced by Repap in the 1990s has been extensively characterized and was used as an Organosolv lignin reference in many investigations. Lignol Innovations is now actively involved in the development of high value applications for their HP-L, high-purity lignin, including collaborations with PPG Industries Inc., Huntsman International, Weyerhaeuser Co. BAE Systems PLC, and the Oak Ridge National Laboratory (ORNL).

P1: SFK/UKS P2: SFK c12 BLBS075-Hood December 15, 2010

302

20:57

Trim: 244mm×172mm

Plant Biomass Conversion

Increasing the severity of the Organosolv pretreatment (decreasing solvent concentration, increasing acidity, treatment time and temperature) results in a more extensive depolymerization and degradation of the lignin, producing a preparation with lower molecular weight, Mn , and higher aromatic hydroxyl functions (Pan et al., 2006b). The fractionation of lignin recovered from a 50% ethanol black liquor produced various fractions of lignin with molecular weights ranging from 436 to 8,000 and narrow polydispersity varying between 1 and 2 (Hergert et al., 2000). In addition, the reactivity of the Organosolv lignin produced under more stringent conditions is often altered by the presence of higher levels of degradation products generated by the hydrolysis of cellulose, and especially hemicellulose, initiated during the Organosolv extraction of the hydrolysis liquor (Pan et al., 2006a; Vila et al., 2003). Hydrothermal pretreatment, to extract the hemicellulose sugars prior to the Organosolv extraction, also has an impact on the yield and characteristics of the lignin extracted during the following Organosolv step. Organosolv lignin extracted after hydrothermal pretreatment of wood has a lower molecular weight and Mn , indicative of extensive depolymerization and degradation (Ehara et al., 2002; Thring et al., 1993). Regarding commercial products, Tecnaro has been marketing a lignin-based product for the production of injection-molded thermoplastic since 2000. Composed of a mixture of lignin, fibers, fatty acids, and dyes, the resin, called Arboform, is produced at a rate of 5–10 MT per month in Ilsfeld, Germany. This biodegradable lignin-based resin can be molded into figurines, speaker cabinets, auto instrument panels, and other products. As a result of the enhanced properties, lignin precipitated from Organosolv black liquor has been included without further fractionation or purification in higher ratios in a wider range of products, than is the case for Kraft lignin (Nagy et al., 2009). Lignin has been used to displace phenol in phenolic resin-based adhesives used to bind substrates such as wood, paper, fibers (e.g., fiberglass), or particles (e.g., wood flour, foundry sand), to form a highly cross-linked composite. Phenol, derived mostly from benzene and propylene, is mainly used for the production of phenolic adhesives and bis-phenol A, from which epoxy and polycarbonate resins are derived. Introduction of lignin in phenolic resin formulations has been demonstrated to reduced the curing time and cost of the resin and to yield a product with improved strength, water resistance, thermal stability, and durability (Cetin and Ozmen, 2003). Formulation of lignin in phenol-formaldehyde-based adhesives also allows displacement of formaldehyde in addition to the reduction in emissions of toxic volatile organic compounds (Senyo et al., 1996). The wood and automotive industries have used lignin-derived adhesives for the production of wood laminate and brake pads, respectively (Lora and Glasser, 2002). United States-based IBM and later Tecnaro have developed epoxy/lignin resin formulations for the fabrication of printed wiring boards to reduce the environmental concerns associated with the fabrication, assembly, and disposal of these products (Kosbar et al., 2000; N¨agele et al., 2005). The laminates formed from lignin-based resins are processed in a similar fashion to current laminates, minimizing the financial considerations of converting to this resin system. A comparison of the lignin-based resin and current petroleum-based resins through a life-cycle assessment indicated a 40% reduction in energy consumption for the lignin-based resin. Lignin has also been incorporated into the formulation of polyurethane coatings, adhesives, and foams. In these applications, lignin is used as a polyol, having multiple aromatic and aliphatic hydroxyl functional groups that are reacted with di-isocyanates to produce PUR. With its aromatic ring, lignin can act as a flame retardant (like phthalic acid-derived aromatic polyester polyols) by producing char instead of burning. The addition of lignin to polyolefin polymers has been shown to stabilize the polymer against hydrolytic and UV degradation

P1: SFK/UKS P2: SFK c12 BLBS075-Hood December 15, 2010

20:57

Trim: 244mm×172mm

12 Biobased Chemicals and Polymers

303

and to increase the rate of biodegradation of the polymer (Gosselink et al., 2004; Rusu and Tudorachi, 1999; Simionescu et al., 1996; Tudorachi et al., 2000). Carbon Fibers Lignin also represents a potential low-cost source of a carbon skeleton suitable for displacing the synthetic polymer polyacrylonitrile (PAN) in the production of carbon fiber. Purified Syringyl lignin (S-lignin) but not guaiacyl lignin (G-lignin) enriched preparation has recently been shown to be a suitable precursor for preparing carbon fibers (Kadla et al., 2002b; Kubo and Kadla, 2005). Because of the high cost of current PAN technology, the use of carbon fiber has been limited to high performance applications. Using a widely available feedstock such as lignin for the production of carbon fibers could have a significant impact on the fuel economy of motor vehicles by reducing the vehicle weight through the replacement of steel panels with a lightweight carbon fiber-reinforced plastic (Kadla et al., 2002a). As the lead laboratory for the US DOE’s Carbon Fiber research initiative, ORNL in Knoxville, TN, has developed and tested the production of carbon fiber precursors from lignin feedstocks. ORNL and their partners UT-Battelle LLC have awarded a 3-year contract to Lignol Innovations to supply Organosolv lignin for use in carbon fiber method development. Preliminary studies show that carbon fiber produced from lignin combined with the low-cost processing technologies being developed at ORNL could substantially reduce production costs. Chemical Products A small amount of lignin is currently used as a feedstock for the manufacture of dimethyl sulfide, dimethyl sulfoxide, and vanillin. These applications are sensitive to competition from their petrochemical equivalents, depending on the variation of crude oil prices. Research has focused on the possibility of deconstructing lignin as a source of aromatic chemicals. The domestic aromatic chemical supply chain comprises approximately 45 billion pounds annually of chemical intermediates (Holladay et al., 2007). With its functionalized aromatic polymer structure, lignin contains monomeric components that could serve as building blocks to eventually displace most, or all, of the petrochemical aromatic intermediates. However, technologies must still be developed to deconstruct lignin’s polymeric structure and convert the resulting components to usable chemical intermediates. These technologies would undoubtedly be advantaged by the availability of uniform higher grade lignin. The development of lignin deconstruction technologies is considered to be a longer-term opportunity by the DOE Lignin Report.

12.5 Summary Biomass, and especially lignocellulosic biomass, represents an abundant sustainable global feedstock for fuels and chemicals to replace fossil-based materials. However, liquid fuels are the ultimate commodity chemicals, representing the lowest unit value products that can be produced in the biorefinery. Biochemical processing technologies, directed primarily toward biomass-derived sugar feedstocks, can effect highly specific conversions that have the potential to preserve much of the functional complexity of the sugar substrates. As a result, biochemical technologies hold great promise to produce a broad range of higher-value multifunctional

P1: SFK/UKS P2: SFK c12 BLBS075-Hood December 15, 2010

304

20:57

Trim: 244mm×172mm

Plant Biomass Conversion

chemical products, despite the fact that most early commercial attention has been directed toward fuel alcohols. Furthermore, lignocellulosic fractionation technologies provide the opportunity to produce a range of value-added aromatic chemicals and polymers, to complement predominantly aliphatic sugar-derived products. The commercial bioprocessing industry as profiled in this review is in its infancy. This industry, characterized by start-up companies and a growing number of established agricultural and chemical industry leaders, seeks to convert abundant biomass resources to a new generation of sustainable products. Despite economic and policy considerations that will undoubtedly enable the emerging industry, biobased products must necessarily compete with their mature petrochemical predecessors for several decades to come. Recognizing this commercial reality, novel biobased chemicals and polymers and/or functional competitive products that are not direct substitutes for petrochemical products may find an early advantage in the marketplace through unique or enhanced properties, in addition to their biogenic origin. The BREW Report identifies four factors that will affect the rate of commercialization of biobased chemicals (Patel et al., 2006): 1. 2. 3. 4.

Technological breakthroughs in the bioprocessing step Major progress in downstream processing High fossil fuel prices Low fermentable sugar prices.

To this list, the authors add two additional considerations: 1. Policy initiatives to enable the incipient industry 2. Consumer demand for environmentally sustainable products. The former recognizes the role of temporary incentives to support commercial competitiveness for early-stage technologies that must compete against mature products from highly optimized processes. Furthermore, policy can also serve to more fully assign the appropriate social and environmental cost to fossil products, in light of current science and politics. The latter consideration—consumer demand—can be a powerful factor in sustainable product preference and will be impacted by growing public awareness (often driven by economics) and enhanced information and labeling to allow alternative product selection. The displacement—and ultimate replacement—of finite fossil-derived fuels, chemicals, and materials represents a disruptive social, economic, and political change. As such it presents challenges and opportunities, with progress measured in terms of years and decades, as has been the case for development of the petrochemical economy. Science and industry have sophisticated new tools and competencies to apply to bioprocessing that were not available throughout much of the petrochemical industry development over the last century, which promise to accelerate the technical transition. Political leadership and the public will must do the rest.

References Bergstra, R. 2007. Emerging opportunities for natural oil based chemicals. In: Plant BioIndustrial Workshop Saskatoon, Canada.

P1: SFK/UKS P2: SFK c12 BLBS075-Hood December 15, 2010

20:57

Trim: 244mm×172mm

12 Biobased Chemicals and Polymers

305

Botello, J. I., Gilarranz, M.A., Rodr´ıguez, F. & Oliet, M. 1999. Preliminary study on products distribution in alcohol pulping of Eucalyptus globulus. Journal of Chemical Technology & Biotechnology, 74, 141–148. Bozell, J. 2006. Oleochemicals in the biorefinery. In: Growing the Bioeconomy. Iowa State University. Brodin, I., Sjoholm, E. & Gellerstedt, G. 2009. Kraft lignin as feedstock for chemical products: The effects of membrane filtration. Holzforschung, 63, 290–297. Canada, I. 2006. Towards a technology roadmap for Canadian forest biorefineries: Capturing Canada’s natural advantage. Ottawa, ON: Public works and Government Services Canada. Carole, T. M., Pellegrino, J. & Paster, M. D. 2004. Opportunities in the industrial biobased products industry. Applied Biochemistry and Biotechnology, 113, 871–885. Cetin, N.S. & Ozmen, N. 2003. Studies on lignin-based adhesives for particleboard panels. Turkish Journal of Argiculture and Forestry, 27, 183–189. Chum, H.L., Black, S.K., Johnson, D.K., Sarkanen, K.V. & Robert, D. 1999. Organosolv pretreatment for enzymatic hydrolysis of poplars: Isolation and quantitative structural studies of lignins. Clean Technologies and Environmental Policy, 1, 187–198. Cleary, K., Duffy, J., Gallentstein, C. & Al, E. 2009. Proposed Regulation to Implement the Low Carbon Fuel Standard, Volume I: Appendices. California Environmental Protection Agency, Air Resources Board. Del Rio, L.F., Chandra, R. P. & Saddler, J.N. 2009. The effect of varying Organosolv pretreatment chemicals on the physicochemical properties and cellulolytic hydrolysis of mountain pine beetle-killed lodgepole pine. Applied Biochemistry and Biotechnology, 161, 1–21. Ehara, K., Saka, S. & Kawamoto, H. 2002. Characterization of the lignin-derived products from wood as treated in supercritical water. Journal of Wood Science, 48, 320–325. Eisenreich, N., Sakai, T., Eckla, W. & Al, E. 2004. Practical Research of Wood-Like Thermoplastic Using Lignin Extracted by High Pressure Hydrolysis Process. NEDO International Joint Research Grant Project. Farrell, R. 2008. Japanese Investment in the World Economy: A Study of Strategic Themes in the Internationalisation of Japanese Industry, Cheltenham, UK; Northampton, MA: Edward Elgar Publishing Ltd. Fox, C. 2006. Chemical and Thermal Characterization of Three Industrial Lignins and their Cooresponding Lignin Esters. University of Idaho. Garcia, A., Toledano, A., Gonzalez, M.J. & Al, E. 2009. Biorefining of lignocellulosic residues using ethanol Organosolv process. Chemical Engineering Transactions, 18, 911–916. Gargulak, J. D. & Lebo, S. E. 1999. Commercial use of Lignin. In: Lignin: Historical, Biological, and Materials Perspectives. Washington DC: ACS Symposium Series. Glasser, W. G., Barnett, C.A., Muller, P.C. & Sarkanen, K.V. 1983. The chemistry of several novel bioconversion lignins. Journal of Agricultural and Food Chemistry, 31, 921– 930. Gosselink, R.J.A., Snijder, M.H. B., Kranenbarg, A., Keijsers, E. R. P., De Jong, E. & Stigsson, L.L. 2004. Characterisation and application of NovaFiber lignin. Industrial Crops and Products, 20, 191–203. Gundersen, S.A. & Sj¨oblom, J. 1999. High- and low-molecular-weight lignosulfonates and Kraft lignins as oil/water-emulsion stabilizers studied by means of electrical conductivity. Colloid & Polymer Science, 277, 462–468. N¨agele, H., Pfitzer, J., Lehnberger, C., Landeck, H., Birkner, K., Viebahn, U., Scheel, W., Schmidt, R., Hagel¨uken, M., & Muller, J. 2005. Renewable resources for use in printed circuit boards. Circuit World, 31, 26–29.

P1: SFK/UKS P2: SFK c12 BLBS075-Hood December 15, 2010

306

20:57

Trim: 244mm×172mm

Plant Biomass Conversion

Hage, R. E., Brosse, N., Chrusciel, L. & Al, E. 2009. Characterization of milled wood lignin and ethanol Organosolv lignin from Miscanthus. Polymer Degradation and Stability, 94, 1632–1638. Helle, S.S., Lin, T. & Duff, S.J. B. 2008. Optimization of spent sulfite liquor fermentation. Enzyme and Microbial Technology, 42, 259–264. Hergert, H.L., Goyal, G.C. & Lora, J.H. 2000. Limiting molecular weight of lignin from autocatalyzed Organosolv pulping of hardwood. In: Lignin: Historical, biological and materials perspectives. Washington DC: American Chemical Society. Holladay, J. E., Bozell, J., White, K.F. & Al, E. 2007. Top Value-Added Chemicals from Biomass—Volume II: Results of Screening for Potential Candidates from Biorefinery Lignin. Richland, WA: Pacific Northwest National Laboratory. Hu, T. Q. (ed.) 2002. Chemical Modification, Properties and Usage of Lignin. New York: Academic/Plenum Publishers. Ibrahim, M.N. M., Chuah, S. B. & Rosli, W. D.W. 2004. Characterization of lignin precipitated from the soda black liquor of oil palm empty fruit bunch fibers by various mineral acids. Asian Journal on Science and Technology for Development, 21, 57–67. Imhof, P., Dias, A., Jong, E. D. & Al, E. 2008. Furanics: Versatile molecules for biofuels and bulk chemical applications. In: Renewable Resources and Biorefineries. Rotterdam, The Netherlands. Jin, Y., Cheng, X. & Zheng, Z. 2010. Preparation and characterization of phenol-formaldehyde adhesives modified with enzymatic hydrolysis lignin. Bioresource Technology, 101, 2046– 2048. Jung, YK., Kim, T. Y., Park, S.J. & Al, E. 2009. Metabolic engineering of Escherichia coli for the production of polylactic acid and its copolymers. Biotechnology and Bioengineering, 105, 161–171. J¨onsson, A.-S. & Wallberg, O. 2009. Cost estimates of kraft lignin recovery by ultrafiltration. Desalination, 237, 254–267. Kadam, K.L., Chin, C. Y. & Brown, L.W. 2009. Fractionation process for producing ethanol and low-molecular weight lignin. Environmental Progress and Sustainable Energy, 28, 89–99. Kadla, J., Satoshi, K., Gilbert, R. D. & Al, E. 2002a. Lignin-based carbon fibers. In: Hu, T. Q. (ed.) Chemical Modification, Properties and Usage of Lignin. New York: Kluwer Academic/Plenum Publishers. Kadla, J.F., Kubo, S., Venditti, R.A., Gilbert, R. D., Compere, A.L. & Griffith, W. 2002b. Lignin-based carbon fibers for composite fiber applications. Carbon, 40, 2913–2920. Kosbar, L.L., Gelorme, J. D., Japp, R. M. & Fotorny, W.T. 2000. Introducing biobased materials into the electronics industry. Journal of Industrial Ecology, 4, 93–105. Kubo, S. & Kadla, J. F. 2004. Poly(ethylene oxide)/Organosolv lignin blends: Relationship between thermal properties, chemical structure, and blend behavior. Macromolecules, 37, 6904–6911. Kubo, S. & Kadla, J. F. 2005. Lignin-based carbon fibers: effect of synthetic polymer blending on fiber properties. Journal of Polymers and the Environment, 13, 97–105. Kumar, P., Barrett, D. M., Delwiche, M.J. & Stroeve, P. 2009. Methods for pretreatment of lignocellulosic biomass for efficient hydrolysis and biofuel production. Industrial & Engineering Chemistry Research, 48, 3713–3729. Larsen, J., Petersen, M. Ø., Thirup, L., Li, H.W. & Iversen, F.K. 2008. The IBUS process—Lignocellulosic bioethanol close to a commercial reality. Chemical Engineering & Technology, 31, 765–772.

P1: SFK/UKS P2: SFK c12 BLBS075-Hood December 15, 2010

20:57

Trim: 244mm×172mm

12 Biobased Chemicals and Polymers

307

Lindberg, J.J., Kuusela, T.A. & Levon, K. 1989. Specialty polymers from lignin. In: Lignin Properties and Materials. Washington DC: ACS Publication. Lora, J.H. & Glasser, W. G. 2002. Recent industrial applications of lignin: A sustainable alternative to nonrenewable materials. Journal of Polymers and the Environment, 10, 39–48. Lora, J.H., Wu, C., Pye, E.K. & Al, E. 1989. Characteristics and potential applications of lignin produced by Organosolv pulping process. In: Lignin Properties and Materials. Washington DC: American Chemical Society. Mao, H., Genco, J. M., Yoon, S.H., Van Heiningen, A. & Pendse, H. 2008. Technical economic evaluation of a hardwood biorefinery using the “near-neutral” hemicellulose pre-extraction process. Journal of Biobased Materials and Bioenergy, 2, 177–185. McCoy, M. 2009. Glycerin gambit. Chemical and Engineering News, 87, 16–17. Mousavioun, P. & Doherty, W.O.S. 2010. Chemical and thermal properties of fractionated bagasse soda lignin. Industrial Crops and Products, 31, 52–58. Murnen, H.K., Balan, V., Chundawat, S. P.S., Bals, B., Sousa, L. D.C. & Dale, B. E. 2007. Optimization of ammonia fiber expansion (AFEX) pretreatment and enzymatic hydrolysis of Miscanthus x giganteus to fermentable sugars. Biotechnology Progress, 23, 846–850. Muurinen, E. 2000. Organosolv Pulping—A review and distillation study related to peroxyacid pulping. University of Oulu. Nagy, M., David, K., Britovsek, G.J. P. & Ragauskas, A. 2009. Catalytic hydrogenolysis of ethanol Organosolv lignin. Holzforschung, 63, 513–520. Pan, X., Arato, C., Gilkes, N., Gregg, D., Mabee, W., Pye, K., Xiao, Z., Zhang, X. & Saddler, J. 2005. Biorefining of softwoods using ethanol Organosolv pulping: Preliminary evaluation of process streams for manufacture of fuel-grade ethanol and co-products. Biotechnology and Bioengeering, 90, 473–481. Pan, X., Gilkes, N., Kadla, J., Pye, K., Saka, S., Gregg, D., Ehara, K., Xie, D., Lam, D. & Saddler, J. 2006a. Bioconversion of hybrid poplar to ethanol and co-products using an Organosolv fractionation process: Optimization of process yields. Biotechnology and Bioengeering, 94, 851–861. Pan, X., Kadla, J.F., Ehara, K., Gilkes, N. & Saddler, J.N. 2006b. Organosolv ethanol lignin from hybrid poplar as a radical scavenger: Relationship between lignin structure, extraction conditions, and antioxidant activity. Journal of Agricultural and Food Chemistry, 54, 5806–5813. Patel, M., Crank, M., Dornburg, V., Hermann, B. G., Roes, L., H¨using, B., Overbeek, L., Terragni, F. & Recchia, E. 2006. Medium and long-term opportunities and risks of the biotechnological production of bulk chemicals from renewable resources, The potential of White Biotechnology. The BREW Project; Technical report; Utrecht University: Utrecht, The Netherlands. Perlack, R. D., Wright, L.L., Turhollow, A.F., Graham, R.L., Stokes, B.J., Erbach, D.C. 2005. Biomass as Feedstock for a Bioenergy and Bioproducts Industry: The Technical Feasibility of a Billion-Ton Annual Supply. USDOA, Technical Report: A357634. Pouteau, C., Cathala, B., Dole, P., Kurek, B. & Monties, B. 2005. Structural modification of Kraft lignin after acid treatment: Characterisation of the apolar extracts and influence on the antioxidant properties in polypropylene. Industrial Crops and Products, 21, 101– 108. Pye, E.K. 2008. Industrial lignin production and application. In: Biorefineries Industrial Processes and Products. Status Quo and Future Directions (eds B. Kamm, P. R. Gruber and M. Kamm), Wiley-VCH Verlag GmbH, Weinheim, Germany.

P1: SFK/UKS P2: SFK c12 BLBS075-Hood December 15, 2010

308

20:57

Trim: 244mm×172mm

Plant Biomass Conversion

Restolho, J.A., Prates, A., De Pinho, M.N. & Afonso, M. D. 2009. Sugars and lignosulphonates recovery from eucalyptus spent sulphite liquor by membrane processes. Biomass and Bioenergy, 33, 1558–1566. Rojas, O.J., Song, J. & Argyropoulos, D.S. 2006. Lignin separation from black liquors by tangential ultrafiltration. La Chimica e l’Industria, 88, 88–95. Rust, D. & Wildes, S. 2009. Market Opportunity Study Update: Surfactants. United Soybean Board. Rusu, M. & Tudorachi, N. 1999. Biodegradable composite materials based on polyethylene and natural polymers. I: Mechanical and thermal properties. Journal of Polymer Engineering, 19, 355–369. ´ & Cardona, C.A. 2008. Trends in biotechnological production of fuel ethanol S´anchez, O.J. from different feedstocks. Bioresource Technology, 99, 5270–5295. Schmidt, M., Dietrich, C. R. & Cahoon, E. B. 2006. Biotechnological enhancement of soybean oil for lubricant applications. In: Rudnick, L. R. (ed.) Synthetics, Mineral Oils, and BioBased Lubricants: Chemistry and Technology. Boca Raton, FL: CRC Press. Senyo, W.C., Creamer, A.W., Wu, C.F., & Lora J.H. 1996. The Use of Organosolv Lignin to Reduce Press Vent Formaldehyde Emissions in the Manufacture of Wood Composites. Madison, WI: ETATS-UNIS, Forest Products Society. Shaine, K.S., Bozell, J., Wallace, R. & Al, E. 2004. Biomass Oil Analysis: Research Needs and Recommendations. US National Renewable Energy Laboratory. Shen, L., Haufe, J. & Patel, M.K. 2009. Product overview and market projection of emerging bio-based plastics. European Polysaccharide Network of Excellence and European Bioplastics. Utrecht, The Netherlands PRO-BIP 2009. Simionescu, C. I., Macoveanu, M. M., Vasile, C., Ciobanu, F., Esanu, M., Ioanid, A., Vidrascu, P. & Georgescu-Buruntea N. 1996. Polyolefins/Lignosulfonates Blends. Cellulose Chemistry and Technology, 30, 411–429. Sousa, L. D., Chundawat, S. P.S., Balan, V. & Dale, B. E. 2009. “ Cradle-to-grave” assessment of existing lignocellulose pretreatment technologies. Current Opinion in Biotechnology, 20, 339–347. Stewart, D. 2008. Lignin as a base material for materials applications: Chemistry, application and economics. Industrial Crops and Products, 27, 202–207. Taherzadeh, M.J. & Karimi, K. 2007. Acid-based hydrolysis processes for ethanol from lignocellulosic materials: A review. Bioresources, 2, 472–499. Taherzadeh, M.J. & Karimi, K. 2008. Pretreatment of lignocellulosic wastes to improve ethanol and biogas production: A review. International Journal of Molecular Sciences, 9, 1621–1651. Tejado, A., Pena, C., Labidi, J., Echeverria, J. M. & Mondragon, I. 2007. Physico-chemical characterization of lignins from different sources for use in phenol-formaldehyde resin synthesis. Bioresource Technology, 98, 1655–1663. Thayer, A. 2009. Supplanting oil. Chemical and Engineering News, 87, 22–26. Thring, R. W., Chornet, E. & Overend, R. P. 1993. Fractionation of woodmeal by prehydrolysis and thermal Organosolv. Process strategy, recovery of constituents, and solvent fractionation of lignins so produced. The Canadian Journal of Chemical Engineering, 71, 116– 123. Toledano, A., Garc´ıa, A., Mondragon, I. & Labidi, J. 2010. Lignin separation and fractionation by ultrafiltration. Separation and Purification Technology, 71, 38–43. Tripp, S., Nelson, P., & Powell, R. 2009. Regional Strategy for Biobased Products in the Mississippi Delta, Report by Battelle Technology Partnership Practice.

P1: SFK/UKS P2: SFK c12 BLBS075-Hood December 15, 2010

20:57

Trim: 244mm×172mm

12 Biobased Chemicals and Polymers

309

Tudorachi, N., Rusu, M., Cascaval, N.C. & Al, E. 2000 Biodegradable polymeric materials I: polyethylene-natural polymer blends. Cellulose Chemistry and Technology, 34, 101–111. Um, B.H. & Van Walsum, G. P. 2009. Acid hydrolysis of hemicellulose in green liquor prepulping extract of mixed northern hardwoods. Applied Biochemistry and Biotechnology, 153, 127–138. Vila, C., Santos, V. & Parajo, JC. 2003. Recovery of lignin an furfural from acetic acid–water–HCL pulping liquors. Bioresource Technology, 90, 339–344. Voitl, T. & Rohr, P. R.V. 2009. Demonstration of a process for the conversion of kraft lignin into vanillin and methyl vanillate by acidic oxidation in aqueous methanol. Industrial & Engineering Chemistry Research, 49, 520–525. Walton, S., Heiningen, A.V. & Walsum, P.V. 2010. Inhibition effects on fermentation of hardwood extracted hemicelluloses by acetic acid and sodium. Bioresource Technology, 101, 1935–1940. Werpy, T., Peterson, G., Aden, A., Bozell, J., Holladay, J., White, J., Manheim, A. 2004. Top value added chemicals from biomass. In: Werpy, T. & Peterson, G. (ed.) Volume I: Results of Screening for Potential Candidates from Sugars and Synthesis Gas. Washington DC: Department of Energy. Xu, F., Sun, J.-X., Sun, R., Fowler, P. & Baird, M.S. 2006. Comparative study of Organosolv lignins from wheat straw. Industrial Crops and Products, 23, 180–193. Xu, Y., Li, K. & Zhang, M. 2007. Lignin precipitation on the pulp fibers in the ethanol-based Organosolv pulping. Colloids and Surfaces A: Physicochemical and Engineering Aspects, 301, 255–263. Zhao, X., Cheng, K. & Liu, D. 2009. Organosolv pretreatment of lignocellulosic biomass for enzymatic hydrolysis. Applied Microbiology and Biotechnology, 82, 815–827.

P1: SFK/UKS P2: SFK c12 BLBS075-Hood December 15, 2010

20:57

Trim: 244mm×172mm

P1: SFK/UKS P2: SFK c13 BLBS075-Hood December 15, 2010

20:59

Trim: 244mm×172mm

Chapter 13

Carbon Offset Potential of Biomass-Based Energy Gauri-Shankar Guha

13.1 Emerging Public Interest in Carbon 13.1.1

Overview

The fourth assessment report (AR4) of the Intergovernmental Panel for Climate Change (IPCC) provides remarkable evidence of rapid changes in climate during the last two decades and establishes causality with anthropogenic carbon dioxide emissions. It urges immediate cutbacks of greenhouse gas (GHG) emissions through a variety of mitigation measures including global carbon trading. As well, rising prices of conventional energy sources, security threats of petroleum imports, and environmental damages from burning fossil fuels are motivating research into biomass-based fuels. Energy derived from agricultural and forestry residues, energy crops, and other forms of lignocellulosic biomass could address these issues, result in net CO2 reductions, and, most importantly, make energy use part of a carbon cycle. The Kyoto Protocol’s prescription for providing commercial carbon offsets was to plant trees that sequester carbon from the atmosphere at rates that vary with species type and age. Subsequently, other activities like alternate fuels, methane capture, and certain farming practices have qualified as offsets. Carbon trading is already mandatory in Europe, but presently voluntary in the United States. Economic models using Monte Carlo simulation can estimate the potential for carbon offsets from biofuels.

13.1.2

Initiatives to Address Anthropogenic Climate Change

The United Nations Conference on Environment and Development (Earth Summit) in Rio de Janeiro in 1992 introduced the world to the possibility of anthropogenic climate change that could have wide ranging impacts on human health and quality of life. The Earth Summit Plant Biomass Conversion, First Edition. Edited by Elizabeth E. Hood, Peter Nelson and Randall Powell.
C 2011 John Wiley & Sons Inc.

311

P1: SFK/UKS P2: SFK c13 BLBS075-Hood December 15, 2010

312

20:59

Trim: 244mm×172mm

Plant Biomass Conversion

generated new concepts like Debt–Nature Swaps (poor nations receive debt forgiveness in exchange for preserving rainforests and habitats), Conservation Easements (designated regions of biodiversity conservation are financed by an international common fund), Green accounting, and Green GDP (Gross Domestic Product). These were envisioned with the primary purpose of environmental conservation, while climate mitigation was a desirable secondary goal (Cline, 1992) The next two decades witnessed an enormous amount of research on the science and impacts of climate change, as well as mitigation and adaptation strategies. The outcome included four assessment reports presented by the IPCC that involved thousands of scientists and economists from around the world, and 15 Conferences of Parties (COP), which are meetings of policymakers. The watershed event took place in Kyoto, Japan, in 1996, where the governments of most major countries gathered to draft a protocol that would provide concrete strategies and implementable programs, consistent with the IPCC recommendations for mitigating climate change. The Kyoto Protocol set targets and timetables for a gradual cutback of CO2 emissions and proposed the concept of creating carbon offsets through various carbon sequestration projects including afforestation and reforestation (Victor, 2001). So far, opinions and reactions have been a perplexing mix, ranging from fear mongering to skepticism, and the rhetoric has been quite strident from all parties. That is not surprising, considering what is at stake. On one hand, there is the threat of a reversal in the trend of global standards of living, which have continuously increased since the industrial revolution; and, on the other hand, there is a distinct possibility that the cost of mitigation will be huge for the current generation. Whether global climate change is a credible threat or not, economists contend that the creation of a global carbon market, along the lines of a financial derivatives market, and not as a more restrictive commodity market, would allow society to set its own price for carbon (Larsen and Shah, 1994). The AR4 of the IPCC presents three serious issues. One, it documents the rapid change in global climates during the last two decades and forecasts a global mean temperature rise between 3◦ F and 7◦ F and significant melting of polar ice caps during the 21st century, drastically in contrast with the relative climatic stability of the last half million years. Two, it establishes the link between global warming and ambient concentrations of CO2 . Carbon dioxide helps trap heat within the atmosphere, while polar ice caps help reflect heat out into space. Increase in the first combined with a decrease in the second intensifies global warming. The global emission of CO2 was estimated to be 7.5 billion tons in 2006 (Pachauri and Reisinger, 2007). Three, it forecasts that these changes will continue beyond this century, even if carbon cutbacks are implemented today. Climate change implies greater variability of weather patterns, leading to uncertainties in resource supplies. It can also lead to a rise in sea levels, which is of concern to coastal regions that are home to many megalopolises, and alter the hydrological cycle, thereby increasing the incidences of droughts and floods (Barker et al., 2007). The negative impacts of climate change in the United States include thawing of the permafrost in Alaska, declining lake levels leading to reduced water supply and costly transportation in the Great Lakes region, adverse health impacts from a higher heat index in the Northeast, more intense rainfall events increasing the potential for flash floods in the Appalachians, and rising sea-levels and storm surges that will threaten natural ecosystems and human settlements specifically on the Atlantic Seaboard. Higher winter temperatures in the Mountain West will reduce snow packs and shift the peak runoff to spring, making water management difficult, while higher stream temperatures in the Pacific Northwest will stress migrating fish, thus complicating restoration efforts. Finally, human health is vulnerable to the

P1: SFK/UKS P2: SFK c13 BLBS075-Hood December 15, 2010

20:59

Trim: 244mm×172mm

13

Carbon Offset Potential of Biomass-Based Energy

313

direct impacts of climate change like extreme weather events, as well as to its indirect effects like diseases carried by water, food, insects, and other vectors (US Global Change Research Program, 2001). These symptoms of climate change impose hardships on living species and additional economic costs on production systems. Water pricing and allocation will become very important to combat the impacts of higher temperatures, which increase the need for water, and changes in the hydrological cycles, which increase the cost of planning, harvesting, and managing water resources. Survival becomes more difficult, and species that cannot adapt or migrate may perish. Human health management becomes more costly as pests and vector-borne diseases spread to new areas (Guha, 2008). Increased awareness of anthropogenic climate change has energized people in a variety of societal roles to become proactive in mitigation efforts. Such action needs to be at the 3 Cs, namely, Congress, Corporation, and Community, and fortunately, action on climate change is snowballing at all three levels. The US Congress is engaged in debating policies that mitigate carbon as well as promote carbon neutral behavior in all economic sectors. The stalled capand-trade bill in Congress may be viewed as a failure of policymaking, but is perhaps only a temporary stumbling block for mitigation efforts, given the spate of recent green energy initiatives on display in Congress. A case in point is the raising of the Corporate Average Fuel Economy (CAFE) standards to 35 mpg for all vehicles by 2020, which has the potential to reduce CO2 emissions by 79 million metric tons (USCBO, 2004). There is also a promising discussion to introduce carbon taxes to stimulate carbon neutral auto design. Giant US corporations like Alcoa, BP America, Caterpillar, Duke Energy, DuPont, FPL Group, General Electric, PG&E Corporation, and PNM Resources forged the US Climate Action Partnership (USCAP) to develop and encourage climate friendly technologies and management practices and to reduce GHG emissions in the shortest possible time. The mission statement of USCAP is: We, the members of the US Climate Action Partnership, pledge to work with the President, the Congress, and all other stakeholders to enact an environmentally effective, economically sustainable, and fair climate change program consistent with our principles at the earliest practicable date. (USCAP, 2010)

The progress of the USCAP, however, has taken a major hit in February 2010, when BP America, Caterpillar, and ConocoPhillips chose to leave, citing a variety of reasons (Morrison, 2010). Fortune 500 giants, like DuPont, IBM, and Ford, are voluntarily capping and trading their emissions (Korosec, 2010). This type of action promises to proliferate to more corporations. Moreover, community level events like Live Earth have drawn public attention to the need for recycling, reducing waste of energy and materials, supporting environment-friendly businesses and adopting local innovations that reduce societal carbon footprints. The growing awareness of the need for carbon offsets is, without doubt, an important positive issue that can safeguard the health of the planet and maintain high standards of living. But there is a grave caveat: the risk of “scams” may discourage honest efforts in achieving carbon neutrality. Growing evidence shows that several international programs, selling carbon credits to businesses and individuals, were conceptually incorrect or poorly designed and executed, thereby failing to actually deliver carbon offsets (Harvey and Fidler, 2007).

P1: SFK/UKS P2: SFK c13 BLBS075-Hood December 15, 2010

314

20:59

Trim: 244mm×172mm

Plant Biomass Conversion

13.1.3 GHG Mitigation and Carbon Sequestration Strategies The aftermath of the Kyoto Protocol has seen the emergence of innovative strategies to mitigate GHGs, including research and development of carbon neutral processes and technologies, alternate energy sources, and the concept of carbon sequestration. Following Kyoto, the tradition was to view forests as the only suitable mechanism for harvesting carbon from the atmosphere, as well as providing saleable offsets. Currently, other activities are allowed to qualify as offsets, such as carbon sequestration in soils and projects involving biomass-based fuels. This has led to a recent flurry of voluntary actions to create new forests, farm-based innovations and alternate fuels. It is estimated that the US Farm Sector alone can potentially sequester 120–270 MM tons of CO2 Equivalent (tCE) per year (Guha, 2007). GHG mitigation efforts can be categorized as energy or nonenergy projects as follows: r Nonenergy-related broad strategies ◦ Carbon capture ◦ Methane combustion ◦ Abatement devices ◦ Biomass accumulation through reforestation ◦ Carbon accumulation in agricultural soils ◦ Agricultural manure digestion r Energy related broad strategies ◦ Increase fuel use efficiency ◦ Renewable energy sources—solar, wind ◦ Low carbon fuel like natural gas ◦ Biomass-based energy/biofuels.

Energy-related strategies offer a wider range of adoption areas to society, and biomass-based energy additionally offers the dual advantage of generating carbon offsets as well as bringing energy production and consumption into a carbon cycle.

13.2 Theory of Carbon Markets 13.2.1 Tradable Permits and the Market for Emissions Economic theory explicitly recognizes the presence of trade-offs between resource use and conservation and assures that the market mechanism always prices the resource at its best value and a social optimum is reached when all markets clear. Emissions are considered negative externalities of a productive process, implying that they do not figure in the cost calculations of the producer, but impose costs on society nevertheless. Negative externalities lead to market failures since there is little incentive for producers to pay for pollution. Hence, unlike most instances involving trade-offs, the ordinary market mechanism generally fails to price pollution correctly and does not bring about a social optimum. In such cases, markets are created through the assignment of property rights and a set of market-based system of penalties and abatement incentives, known as price (emission charges) or quantity (tradable permits) instruments. The price-based instruments are emission taxes that reward cleaner processes over dirty ones, while quantity-based instruments include various types of permits and concessions.

P1: SFK/UKS P2: SFK c13 BLBS075-Hood December 15, 2010

20:59

Trim: 244mm×172mm

13

Carbon Offset Potential of Biomass-Based Energy

315

Marketable permits are allowances that define the overall pollution rights for a firm, which it may use or sell to another firm. Additionally, the regulator may impose annual cutbacks from the baseline pollution level. Europeans have traditionally favored emission taxes since they provide revenue for clean up and may also be returned to the industry as a subsidy to compensate for potential loss of competitiveness; while regulators in the United States favor tradable permits since they are more cost effective and fair (Guha and Combs, 2008). Cap-and-trade is a standard approach for implementing marketable permits and represents one of the greatest innovations of environmental economics. The regulator sets an overall limit on emissions, called the cap, and allocates divided up units of the total allowable emission called tradable permits to the group of emitters. Finally, a system is devised by which owners of permits may sell them, resulting in the trade. The tradable permit gives the owner the right to emit up to the face value of the permit. Permits may either be endowed on the basis of historical pollution levels of the firm and take the form of an entitlement (grandfathered) or may be sold (auctioned) by the regulator (Convery, 2009).

13.2.2

Concept of Carbon Markets

The market for carbon is determined by the interplay of demand and supply, like in any other commodity market. The key factors that drive market demand are domestic and international GHG policies, while supply is driven by costs of emissions reductions, opportunity costs of alternate resource use, and competing carbon sinks. A tradable permit system enables efficient allocation of mitigation resources so that the overall emission control target is achieved at the most competitive price determined by the market. So firms with the ability to mitigate at low cost may sell permits to firms with high abatement costs. As implied earlier, this process automatically accounts for differences in marginal abatement costs and provides incentives to develop low carbon technologies and carbon neutral processes. Emission trading is referred to as a cap-and-trade system because maximum emission limits or caps are set by the regulator and tradable emission permits are allocated or auctioned to individual firms. Exceeding the emission caps attracts penalties, so firms may choose to either abate or purchase permits depending on which is more cost effective. This method has been used to regulate sulfur emissions from utilities and has been a remarkable success story in the United States (USEPA, 2010). However, a critical difference exists between previous permit trading systems and carbon markets. Conventional emissions markets have been created on the basis of historical entitlements, wherein the initial stock of emission permits are either grandfathered or auctioned to the emitters, and the supply of mitigation is completely endogenous to the system. On the other hand, the bulk of carbon credit supply is likely to be exogenous as carbon offsets are typically purchased from outside the emission system. The ultimate structure of carbon markets will depend on how carbon transactions are designed. They could be based entirely on allocations of carbon emission permits by regulators, which are then traded, or they could rely on a completely exogenous offset supply, or some system in between. Carbon offsets are voluntary acts by entities that aim to sequester or remove a certain amount of carbon dioxide from the atmosphere. Carbon offsets may even be created by a variety of ordinary activities like conserving energy, driving fuel-efficient automobiles, using energy efficient appliances, abating onsite emissions in factories, etc., but are not eligible for trade. Commercial carbon offsets, accounted in CO2 equivalents, need to follow set guidelines of

P1: SFK/UKS P2: SFK c13 BLBS075-Hood December 15, 2010

316

20:59

Trim: 244mm×172mm

Plant Biomass Conversion

creation, aggregation, and verification, before they are available for purchase through trading platforms (Wilman and Mahendrarajah, 2002). Carbon credits can be traded or transferred on secondary carbon markets, but these are presently voluntary in the United States. Resources for the Future (RFF)—an organization that strives to improve the quality of environmental policymaking—proposes requiring about 2,500 companies that introduce fossil fuels into the US economy to obtain carbon emission permits. Under this proposal, permits for 1.3 billion metric tons of carbon would be auctioned every year (Stavins, 2008). Traditionally, commercial carbon offsets were created by planting trees, which sequester carbon from the atmosphere at rates that vary with species type and age. Recently, other activities like alternate fuels, methane capture and certain farming practices are allowed to qualify as offsets, provided they satisfy set standards of specificity, third party validation, verifiable quality, and are not sold to multiple buyers. Forestry carbon offset projects offer a credible low-cost option to mitigate carbon dioxide emissions. The farm sector also provides effective alternatives for sequestering carbon with an estimated potential of 120–270 million metric tons of carbon per year. On the basis of a conservative estimate of $20–$30 per ton of carbon, sequestering 200 MM tCE per year adds $4–$6 billion of gross income to the farm economy, or 10% of typical net farm income (Sandor and Skees, 1999). As well, active trading of carbon could prove an inexpensive hedge against uncertainties, raise farm incomes, and land values.

13.2.3 Demand and Supply of Carbon Credits According to the US Department of Energy, the federal government is the single largest consumer of energy in the United States, accounting for about 2% of total energy consumption. In 2005, the government consumed 1.1 quadrillion BTUs or quads of energy at a cost of $14.5 billion and emitted over 100 million tons of carbon. The daily operation of the United States House of Representatives generated 91,000 tons of GHGs in 2006—equal to the annual emissions of 17,200 cars. The Carbon-Neutral Government Act of 2007 (H.R. 2635) sets targets for GHG reduction and standards for energy efficiency for the federal government, with the ultimate goal of zero emission by 2050. Measures to achieve this goal entail energy-efficiency standards for new federal buildings, renewable energy certificates, low carbon vehicles, and the purchase of carbon offsets (Layton, 2007). Progressive cutbacks from a given base year mandated by legislature could force industry to make dramatic cutbacks at high cost. Such expectations can provide the incentive for businesses to buy offsets, since pre-emptive carbon trading can cushion industry from legislative shocks and also provide an inexpensive hedge against uncertainties. The demand side of a carbon market is driven by these forces, as well as voluntary actions by motivated individuals, conservation groups, and other high profile groups like celebrities. The supply of carbon offsets is contingent upon the price of carbon exceeding the opportunity costs of the resource, or the next best alternative in which the owner may reasonably expect to employ the resource. Consider, for example, a landowner contemplating planting trees to provide carbon offsets. The action is economically justified only when the net revenue from carbon credits (net of planting and management costs) is equal to or greater than the revenue from the best alternative use for that piece of land. The simple analytical example below serves to explain this market model formally. Suppose a landowner is offered a carbon offset contract to adopt a certain standard offset generation practice, under which soil carbon increases by c tons/year for T years for

P1: SFK/UKS P2: SFK c13 BLBS075-Hood December 15, 2010

20:59

Trim: 244mm×172mm

13

Carbon Offset Potential of Biomass-Based Energy

317

a given acreage. The landowner will enter into contract, when the following inequality is satisfied: π0 < πc + s where π 0 is the profit from conventional landuse practices (best alternative), π c is the profit from improved practices, and s is the annual contract payment ($ s/year) for t years. Now, the buyer of the contract receives carbon credit at the rate of c for T years, and an efficient market would clear when:


s=




pc

where c mt/year is the credit that the buyer receives c/year, for T years, and p is the competitive market price ($ p/mt) of carbon.

13.3 Creation of Carbon Markets 13.3.1

Carbon Credits

One carbon credit is defined as 1 tCE, while the trading instrument in the United States, known as the Carbon Financial Instrument (CFI) Contract is an aggregated block of 100 carbon credits. Tons of carbon or CO2 emissions sound fairly nebulous in total abstraction, and the following concrete examples serve to provide better clarity. A car that has an average gas mileage of 28 mpg, driven for 10,000 miles would emit 3.67 tons of CO2 or 1 ton of net carbon; and if all 10,000 (kwh) units of electricity that a typical United States household uses in a year is generated using coal, it would emit 11 tons of CO2 or 3 tons of net carbon (Nordhaus, 2008). Nordhaus (2008) estimates that the social cost of 1 ton of carbon is $30, and the average carbon emission in the United States was 5 tons per capita in 2007. If carbon emissions in the United States could be charged a carbon tax, it would generate revenue of $150 billion, which under the European style of reasoning could then subsidize carbon-reduction research, technologies, and projects. Although carbon offsets may be created through an assortment of projects in a variety of locations worldwide, they are not all created equal, and there are significant quality variations. For example, the Clean Development Mechanism (CDM) creates credits from carbon abatement projects in developing countries. These credits are known as Certified Emission Reductions (CER) and are more widely accepted for compliance, for example in the European Union (EU) system, as well as in Canada and Japan. The Joint Implementation (JI) program of the Kyoto Protocol also creates carbon credits known as Emission Reduction Units (ERU) wherein the projects are mostly situated in the former soviet republics (Matthews et al., 2008). While selecting prospective projects, it is important to ascertain whether the offset generating activity will occur exclusively because of the carbon trade. For example, the Academy Awards of 2007 as well as the city of Seattle discovered, separately and post facto, that the vast amount of money they spent to become carbon neutral was indeed a wash. Their carbon offsets were in fact associated with projects that would have taken place regardless (Elgin, 2007). To that extent, the money only served to purchase moral satisfaction, but did not stimulate any incremental carbon sequestration.

P1: SFK/UKS P2: SFK c13 BLBS075-Hood December 15, 2010

318

20:59

Trim: 244mm×172mm

Plant Biomass Conversion

13.3.2 Global Carbon Trade The EU has been operating a thriving carbon market through the EU Emission Trading System (EU–ETS) since January 2005. Phase I of the EU–ETS ran from 2005–2007 in which all 15 members (at the time) complied. In its first year, 362 million tCE was traded for a market value of €7.2 billion or $9.72 billion. Carbon credit prices peaked to about €30, or about $40 per tCE in mid 2006, but have steadily declined ever since (Ellerman and Buchner, 2007). Phase II of the EU–ETS runs from 2008–2012, and all 27 members are likely to participate. The US carbon market has been running on a voluntary basis, but the stage is set for a formal market following the creation or consideration of several regional ones. As well, Australia, Canada, Japan, and New Zealand look poised to set up national carbon markets, while China is considering a voluntary market (Brinkman, 2010). The total global carbon market in 2009 rose by about 67% to 8.2 billion tCE. Unfortunately, with a proportionate decline in prices, the total market value of trade stagnated at about $135 billion. Carbon credit prices had peaked during the mid-2000s, reaching €32/tCE in Europe and about $3.50 in the United States. The US prices are typically low, due to both demand and supply factors, and trades at the Chicago Climate Exchange (CCX) have ranged between $1.00 (2003) and $3.50 (2007) in the past. The prices of carbon credits have bottomed out in 2010 in all markets, and carbon credits are currently trading at about €14.95/tCE in the Euro-ETS and about $0.10/tCE in the CCX. Theoretical estimates of carbon credits during a cap-and-trade regime range from $20 to $100 (Point Carbon, 2010). McKinsey estimates that the global market for carbon will grow to about €800 billion or just above $1 trillion by 2020 (Brinkman, 2010).

13.3.3 Carbon Trading in the United States The size of the US market cannot be accurately projected from the average annual 8–10 million tCE traded in the CCX, due to its purely voluntary nature. The first mandatory cap-and-trade program, called the Regional GHG Initiative, was implemented in northeastern United States during 2009, trading carbon credits for a total value of $2.5 billion (Point Carbon, 2010). California is likely to follow suit in 2011, but the US carbon market is so far mostly voluntary and dominated by the CCX. GHG emissions have been actively traded in the United States since December 2003 through the CCX, despite the absence of cap–and-trade legislation or a regulatory regime mandating GHG emission reductions. In 2005, the CCX traded around 1.4 million metric tons of CO2 and over 8.2 million metric tons up to September 2006 (Point Carbon, 2006). A highly documented success story of voluntary carbon trading in the CCX is that of the North Dakota Farmers Union’s Carbon Credit Program, which pays producers to capture CO2 in soils through no–till crop management. This has resulted in 652,200 metric tons of carbon being sequestered annually from 830,000 acres, which is equivalent to offsetting the annual carbon emissions from 130,440 automobiles. The total carbon credit revenue generated, net of transaction costs, in 2007 was $2,065,077 (Enerson, 2007).

13.3.4 The CCX Offset Program The CCX is the pioneering trading system for all six GHG emissions in North America, offering integrated GHG reduction with offset projects worldwide. Although membership of CCX is

P1: SFK/UKS P2: SFK c13 BLBS075-Hood December 15, 2010

20:59

Trim: 244mm×172mm

13

Carbon Offset Potential of Biomass-Based Energy

319

voluntary at this time, the CCX contracts, once entered, are legally binding and enforceable. The emitting members of the CCX make voluntary but legally binding commitments of annual GHG abatement targets. The commodity traded at CCX is the CFI contract, each representing 100 metric tCE. Members who exceed abatement targets may choose to either sell or bank the excess as carbon credits, while those who emit above those targets comply by purchasing CFI contracts from CCX. The CCX traded around 1.4 million tCE in 2005, and over 8.2 million tCE in September 2006. The carbon offset program of the CCX comprises enrollment, verification, and aggregation, and trade with registered CCX members. The proceeds from carbon sales are remitted to the offset supplier through the aggregator (Exchange, 2010). r The process of enrollment as an offset supplier starts with gathering landowner and tract

descriptions to identify eligible offset projects that can sequester or eliminate GHGs. This is done by the various aggregators registered with the CCX, who typically operate and focus on different regions of the United States. Aggregators scout their region for eligible projects and are responsible for imparting technical information and education to the prospective offset provider. At present, eligible project categories include landfill methane destruction, agricultural (livestock) methane destruction, no-till agriculture, new grass, rangeland, and forestry projects (new tree plantings), as well as fuel switching and renewable energy. (Note that methane is an important GHG and its destruction is considered carbon equivalent.) Estimates of carbon offset volumes and expected revenues are provided to the landowner, who has the choice of entering into contract with the aggregator to let their carbon offsets be combined into CFIs that can be sold to CCX members. r The next stage is verification and aggregation of offsets. CCX requires a third party verifier, vetted by the CCX, to review the project. Aggregators carry out verification, enrollment, certification, and credit marketing on behalf of the offset provider. r Offset Contracts are then drawn up and registered with the CCX. Once committed, the landowner’s property is verified annually. The aggregators are responsible for accumulating the individual contracts into CFIs (100 mtCE each), which are the tradable units at the CCX. Credits are put in queue in the order they are received for verification and sale, and payments are made within 14 days of credit sale. Aggregators pay the offset providers the proceeds from sale of their offsets, minus service and transaction fees, at previously agreed upon periods.

13.4 Role of Biomass-Based Energy in Carbon Markets 13.4.1

Economic Significance of Bioenergy

Modern societies have become increasingly dependent on energy, the bulk of which is obtained from fossil fuels. The vital transportation sector, in particular, relies heavily on petroleum resources. But it is unrealistic to presume that fossil fuel reserves will continue to meet the rising energy demands indefinitely, since this is a nonrenewable resource pool with a limited horizon. In fact, energy forecasting models of the 1970s had predicted that oil reserves would only last for about 50 years given the current rate of consumption, an eventuality that was averted through new discoveries and exploration technologies. The high petroleum-intensity of the energy sector also poses a national security concern, since a large portion (57% in 2008) of the US crude oil demand is met through imports (EIA, 2010). Besides, scientific evidence

P1: SFK/UKS P2: SFK c13 BLBS075-Hood December 15, 2010

320

20:59

Trim: 244mm×172mm

Plant Biomass Conversion

now links fossil fuel consumption to a host of increasing environmental problems. These concerns have set the stage for alternate energy from biomass, wind, solar, and geothermal sources. Most of these are currently priced much higher than conventional energy resources, but serve to provide future hope and a backstop against a plausible economic chaos in the event of fossil fuel exhaustion. These new energy sources promise energy abundance, sustainable concentrations of ambient CO2 , and technological solutions for rural economic development. The US government expects that renewable biomass-based fuels will supply 20% of the total energy demand by 2020 and 25% by the year 2025 (House, 2007). Concerns about future supplies and environmental costs of fossil fuels have prompted the search for clean burning fuels that can be produced in the United States from renewable sources. Ethanol from corn and biomass has been gaining currency since its production can reduce dependence on imports and other supply uncertainties. Biomass is converted into ethanol through industrial fermentation, chemical processing, and distillation. Approximately 2.8 gallons (10 L) of ethanol can be produced from one bushel of corn (56 lbs). Ethanol contains 35% oxygen, which reduces the harmful constituents of gasoline when burned as a fuel, is water soluble, and biodegradable. The annual US production of ethanol has grown over a thousandfold in the last 30 years, from about 10 million gallons in 1979 to 10.7 billion gallons in 2009, across more than 100 facilities around the nation. Ethanol is subsequently blended into unleaded gasoline in varying percentages (Association, 2010). Ethanol prices in the United States are higher at present given the state of technology, but are predicted to fall below gasoline prices with the advent of the second generation of technologies. In countries where ethanol is produced from other crops such as sugarcane, ethanol prices are already well below conventional gasoline. However, ethanol has 33% less energy content compared to gasoline, resulting in lower gas mileage, and hence, prices need to be even more competitive to be economically relevant. Biodiesel is another alternative biomass-based fuel that is attractive, partly because it can be also produced from almost any type of oil or animal fat, although the prevalent source is soybeans. The chemical makeup of biodiesel is similar to gasoline or diesel, and it can be used in existing diesel engines with only slight modifications. Input costs are prohibitive in plant biomass-based biodiesel production. One gallon of soy biodiesel requires 7.3 pounds of soybean oil, and at $0.20 per pound, the input costs are over $1.50 per gallon, whereas, biodiesel can be produced from fats and greases for about $1.00 per gallon. Biodiesel processed from animal waste and residual cooking oils from restaurants has the promise to contribute significantly to energy supply while reducing GHG emissions and pollution. Corporate endorsements of biodiesel include Wal-Mart, headquartered in Arkansas, which plans to run its fleet of trucks on biodiesel. Also Tyson Foods, the largest chicken processor in the world, also based in Arkansas, is collaborating with Conoco to produce biodiesel from chicken fat (Guha, 2008). The fourth assessment report of the IPCC predicts that bioenergy is poised for substantial growth by 2030, with the development of biomass capacity (energy crops), logistic capacity and markets, and the commercialization of second-generation biofuel production. Currently, the cost and performance in terms of CO2 emissions avoided is unfavorable, except for ethanol production from sugarcane in low wage countries. The share of biofuels use, for instance, by transport could increase to about 10% by 2030 depending on production efficiency and the development of advanced techniques such as conversion of cellulose by enzymatic processes or by gasification and synthesis (Pachauri and Reisinger, 2007).

P1: SFK/UKS P2: SFK c13 BLBS075-Hood December 15, 2010

20:59

Trim: 244mm×172mm

13

13.4.2

Carbon Offset Potential of Biomass-Based Energy

321

Bioenergy Policies, Practices, and Trends

The 1990 Iraqi invasion of Kuwait focused attention on the US oil supplies and underlined the importance of ethanol production for the future. The 1990 amendment of the Clean Air Act championed ethanol production to improve air quality, leading to an increase in corn consumption for ethanol. Several states promote ethanol production with incentives like direct payments of state funds on a per gallon basis, income tax credits, transferable tax credits, direct grants, low-interest loans, etc. Iowa, a traditional corn producing state, has widely adopted ethanol production and consumption. In 1978, ethanol was first introduced at five farm cooperatives, and by 1979, ethanol-blended gasoline was available at 650 gas stations across Iowa. In 1980, Chrysler, Ford, and General Motors issued warranty statements that covered the use of a 10% ethanol blend in their vehicles. In 1985, Ford was the first automobile manufacturer to release a flexible fuel vehicle, which ran on conventional gasoline, an 85% blended ethanol, or an 85% blended methanol. In 1988, the Iowa Corn Promotion Board began to advertise ethanol blended fuels, raising market share in Iowa drastically. E85 fuel is a blend of 85% ethanol and only 15% petroleum. According to the American Lung Association, E85 reduces ozone-forming pollution by 20% and GHG emissions by nearly 30%. However, gas retailers are hesitant to begin selling E85 because of the high installation and transportation costs, which increase with distance from the Midwest Corn Belt. The federal government provides infrastructure grants to E85 retailers and biodiesel wholesale distributors to help share in the cost of installing or upgrading E85 equipment. A retailer could receive 50% of the total cost up to a maximum of $30,000. Both federal and state tax credits also exist for retail service stations. A federal tax credit of 30% of the cost of installing clean-fuel vehicle refueling equipment, for e.g., E85 ethanol pumping stations, with the maximum tax credit of $30,000, is available through December 31, 2010. Iowa gives a tax credit when more than 60% of the total gallons of gasoline sold through metered pumps are blended with ethanol. Once station owners surpass the 60% threshold, they receive a tax credit of $0.025 for every additional gallon of gasoline blended with ethanol and sold during the tax year, until December 31, 2008 (EIA, 2010). In 2002, Congress enacted the nation’s first Energy Title in the Farm Security and Rural Investment Act (aka, the Farm bill). Loans are available to current and new biofuels plants for cellulosic biofuels and for ethanol production. Another initiative included in the Farm Bill is a transportation study to improve biofuels access to the market, including the use of existing pipelines for ethanol transport and the creation of dedicated ethanol pipelines. The Energy Title of the 2002 Farm Bill granted $23 million of funding for Section 9006 to help farmers offset some of the costs of renewable energy projects. The fact that farmers benefit from corn-based ethanol production is evident from the 25% increase in corn production and almost 5-fold increase in the percentage of corn converted into ethanol, in the last decade alone, as shown in Table 13.1. However, it would take 97% of the United States to grow enough corn to eliminate the dependence on foreign oil (Scientists, 2006). This has led to research on producing ethanol from cellulosic biomass like crop residues as an alternative to corn. Since 1978, the United States government has maintained national tax incentives to promote ethanol production, primarily applied to sales of gasoline that is blended with 10% ethanol. The current excise tax on gasoline is over 18 cents per gallon; however, it is 5 cents per gallon for gasoline mixed with 10% ethanol. In 1980, the excise tax provision was adjusted, allowing marketers of ethanol to claim a federal income tax credit in the amount of 52 cents per gallon

P1: SFK/UKS P2: SFK c13 BLBS075-Hood December 15, 2010

322

20:59

Trim: 244mm×172mm

Plant Biomass Conversion

Table 13.1. US corn production and use for ethanol: 1980–2009.

Year

Corn Production (Million Tons)

Corn Used for Ethanol (Million Tons)

Percentage of Corn Used for Ethanol

1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009

169 206 209 106 195 225 209 181 125 191 202 190 241 161 255 188 235 234 248 240 252 241 228 256 300 282 268 331 307 312

0.9 2.2 3.6 4.1 5.9 6.9 7.4 7.1 7.3 8.2 8.9 10.1 10.8 11.6 13.5 10.1 10.9 12.2 13.4 14.4 15.9 17.9 25.3 29.7 33.6 40.7 53.8 76.9 92.7 104.1

0.3 0.8 1.1 1.2 2.9 2.2 2.1 2.3 2.6 4.0 3.1 3.3 3.9 3.3 5.3 2.8 4.0 3.7 4.0 4.1 4.8 5.3 7.9 10.1 9.7 10.6 14.8 22.9 22.5 26.0

Source: Earth Policy Institute, compiled from USDA electronic database, 2009.

of ethanol used. Legislation was also passed to impose an ethanol import tariff to ensure that only domestic, US-produced ethanol received the ethanol tax incentives. The only imports that are exempt are those from countries that are covered by the Caribbean Basin Initiative. An important aspect of bioenergy is the relative abundance and variety of sources and the somewhat easily accessible production technologies. Biodiesel, for instance, can be processed from most plant-based oils by removing the fatty acids from the glycerol backbone with the help of a catalyst. Theoretically, this technology is extendable to any waste oil (e.g., cooking, animal fats, etc.) with some additional cleaning and refining steps. The most accessible plant oils are soybeans, canola, sunflower, and corn. The argument against making biodiesel from soybeans is that it introduces a price distorting competitor to the food and feed chains, which may deprive many communities of a viable source of protein. However, current usage is limited by costs and technology and does not threaten soybean’s status as a food crop. In the future, waste and marginal lands could be used for cultivating soybean as an energy crop, when either technology drives down production costs or fossil fuels reach backstop pricing, with the by-products from the process, like glycerin and solids containing highly concentrated proteins, adding further societal value.

P1: SFK/UKS P2: SFK c13 BLBS075-Hood December 15, 2010

20:59

Trim: 244mm×172mm

13

Carbon Offset Potential of Biomass-Based Energy

323

Over the past few years, many states have also taken steps to increase the use and production of biodiesel. For example, biodiesel suppliers in Arkansas are eligible to receive tax refunds of $0.50 per gallon of biodiesel fuel as well as income tax credit for up to 5% of the costs of the facilities and equipment that is used in the distribution of biodiesel. The Alternative Fuels Commission provides grants of up to $0.10 per gallon for the production of biodiesel, up to 5 million gallons per producer per year, for no longer than 5 years. Minnesota has enacted legislation requiring diesel fuel sold in the state to include at least 2% biodiesel. Michigan and other states have also created incentives for the production of biodiesel and have taken steps to mandate its use in their own motor vehicles. A predominantly agricultural state like Arkansas can reap major economic benefits from biofuel product development along with advances in conversion technology. It may be noted that sourcing of biofuels will have positive multiplier effects for any regional economy. The most recent development in the field is biodiesel from algae, which is arguably the most efficient way to make biodiesel fuel, because the land required is very small. Algae are capable of producing 30 times more oil per acre than the current crops now utilized for producing biofuels. Some species of algae are ideally suited to biodiesel production due to their high oil content, in excess of 50%, and extremely rapid growth rates. Many companies have already embarked on the research and development of algae cultivation as a commercial source for producing biodiesel. Substituting fossil fuel imports with domestic alternative energy sources like biofuels can create new indigenous industries as in the Brazil ethanol program and hence generate employment. Planting crops or trees exclusively for bioenergy requires that adequate agricultural land and labor is available to avoid competition with food production, thus creating rural employment (Pachauri and Reisinger, 2007).

13.4.3

Carbon Offset Opportunities for Biofuels

Biomass-based energy promises to be the norm for the future, once conversion technologies increase efficiency and lower costs of production. Bioenergy is renewable, relatively abundant and environment-friendly. It provides the added advantages of carbon banking via woody crops and also captures thermal energy that would have been lost by on-field burning. Its political rationale is based on national security. Significantly more amounts of carbon can be sequestered by using biomass-based fuels instead of coal or gasoline than from sequestering carbon in standing trees. Since biomass-based technologies like conversion to ethanol have the potential for widespread adoption in the near future, the cultivation of biomass as energy crops may become a potential weapon to combat climate change (Baral and Guha, 2004). Biofuels are also consistent with the natural carbon cycle: carbon dioxide and solar energy are absorbed by biomass and then transferred when it is distilled into ethanol, and finally burning ethanol releases the traces of CO2 into the atmosphere, which are again absorbed by plants, thus beginning a new cycle. In 2008, ethanol production of 8 billion gallons in the United States reduced CO2 emissions by approximately 14 million tons, which is the equivalent of removing 2.1 million cars from the road that use traditional gasoline (Association, 2010). Bioenergy emits 29% less CO2 compared to petroleum fuels, which is significant at costs of $25/tCE (EIA, 2010). Two plentiful biofuel sources that are often overlooked are field crop residues and nontimber forest produce. Field crop residues are usually disposed of through tillage or burn-off. These methods fail to capture the energy content of the biomass, and the second additionally produces

P1: SFK/UKS P2: SFK c13 BLBS075-Hood December 15, 2010

324

20:59

Trim: 244mm×172mm

Plant Biomass Conversion

Table 13.2. Projected mitigation potential of bioenergy in 2030 (gigatons of CO2 –equivalent). Economic Group

GtCE

Organization for economic co-operation and development (OECD)

0.20

Economies in transition Non-OECD World

0.07 0.95 1.22

Source: IPCC, 2007.

GHG. Any innovation in the cost-effective recovery and transformation of these residues will be a major contribution towards a sustainable society (Archer et al., 2008). Introducing biofuel policies coterminous with national emission abatement strategies are more effective in climate mitigation. This is because the CO2 abatement benefit of biofuels is mainly from the well-to-tank part, and incentives for biofuels are more effective climate policies if they are tied to entire well-to-wheels CO2 efficiencies. Thus, preferential tax rates, subsidies, and quotas for fuel blending should be calibrated to the benefits in terms of net CO2 savings over the entire well-to-wheel cycle associated with each fuel (Pachauri and Reisinger, 2007). The Working Group 3 of IPCC is charged with developing adaptation and mitigation strategies to combat climate change. This group predicts the mitigation potential of bioenergy in gigatons of CO2 –equivalent (GtCE) of emissions averted in the year 2030 as shown in Table 13.2 (Pachauri and Reisinger, 2007).

13.5 Prognosis of Carbon Markets A substantial and growing body of initiatives has been taken by US corporations to enhance the level of preparedness for likely future carbon trading. For example, the Carbon Disclosure Project, with $3 trillion in assets under management, has solicited and disseminated information on climate practices from the top 500 global corporations; the Investor Network on Climate Risk has filed some 30 shareholder proxy resolutions asking for disclosure or action on climate risk and some of these have received significant favorable votes. Most modern corporations recognize that they may lose business resources in the event of significant climate change, and hence, carbon neutrality and mitigation may be a very real prospect for the near future. For example, a company that uses freshwater in production may face aquifer salinization due to sea-level rise. Timber companies may lose millions of plantation acres due to forest migration and invasive species of pests. Biodiversity losses can hurt the sourcing of materials for pharmaceutical companies. Proactive investors may also force businesses to adopt carbon neutrality as the new corporate vision, both in order to gain access to markets and to avoid potentially expensive climate-related litigation. These signs augur well for a viable US carbon market, and during the next economic boom could well usher in a cap-and-trade regime. Cutting back on energy consumption and carbon emissions, directly or indirectly through a modified portfolio of input and process choices, is now being advocated as beneficial to the

P1: SFK/UKS P2: SFK c13 BLBS075-Hood December 15, 2010

20:59

Trim: 244mm×172mm

13

Carbon Offset Potential of Biomass-Based Energy

325

bottom-line by some major corporations. BP, perhaps the best known example of pioneering carbon trade, committed in 1997 to bring down its emissions to 10% below 1990 levels—a goal that was being hyped by the Kyoto Protocol. BP implemented an internal cap-and-trade system and reported a saving of $650 million by 2001. Bank of America is another leader in carbon mitigation, having recently launched a 10-year, $20 billion initiative to finance carbon reduction projects and green technologies (Korosec, 2010). Some recent events have given cause for cap-and-trade to proceed with caution. The progress of an important corporate climate initiative, the USCAP, took a likely hit in February 2010, when corporate giants, BP America, Caterpillar, and ConocoPhillips departed because recent legislative proposals had disadvantaged the transportation sector against international competition. The shift in congressional power blocks in early 2010 has also slowed the momentum of the cap-and-trade movement, and instead the US Congress appears to be working on a complex climate bill that includes a carbon tax. It appears that the weight of a long recession and strong political lobbying against cap-and-trade has weakened the resolve of proponents and strengthened the opposition, at least for the short term. Even the European Commission is working on a carbon tax proposal that will impact auto fuel, coal, and natural gas (Morrison, 2010). The prognosis for biofuels is very positive in either event. While a cap-and-trade regime would allow bioenergy to generate value through offsets, the sector will directly benefit from carbon taxes imposed on competing fossil fuels.

References Archer, A., Self, J., Guha, G. & Engleken, R. 2008. Cost and carbon savings from an innovative conversion of agricultural field wastes—A case study of Arkansas. Energy Sources, B: Economics, Planning, and Policy, 3, 103–108. Baral, A. & Guha, G. 2004. Trees for carbon sequestration or fossil fuel substitution: The issue of cost vs. carbon benefit. Biomass and Bioenergy, 27, 41–55. Barker, T., Bashmakov, I., Bernstein, L., Bogner, J., Bosch, P., Dave, R., Davidson, O., Fisher, B., Gupta, S. & Halsnæs, K. 2007. Technical summary. Journal of Geophysical Research. D, Atmospheres, 113, D12119. Brinkman, M. 2010. A New Look at Carbon Offsets [Online]. McKinsey Quarterly. Available: http://www.mckinseyquarterly.com (accessed February 2010). Chicago Climate Exchange. 2010. Overview of CCX [Online]. Available: http://www. chicagoclimatex.com/content.jsf?id=821 (accessed November 8, 2010). Cline, W. 1992. The Economics of Global Warming. Washington, DC: Institute for International Economics. Convery, F. 2009. Reflections—The emerging literature on emissions trading in Europe. Review of Environmental Economics and Policy, 3, 121–137. EIA (Energy Information Administration) Official Energy Statistics from the U.S. Government. Elgin, B. 2007. Another inconvenient truth. Business Week, 4027. Ellerman, A. & Buchner, B. 2007. The European Union emissions trading scheme: Origins, allocation, and early results. Review of Environmental Economics and Policy, 1, 66. Enerson, D. 2007. Farmers Union sells Carbon Credits on CCX; Nets $2 Million for Farmers and Ranchers [Online]. North Dakota Farmers Union News Release. Available: http://www. ndfu.org/data/upfiles/pressreleases/CarbonProgramIncome.pdf (accessed November 8, 2010).

P1: SFK/UKS P2: SFK c13 BLBS075-Hood December 15, 2010

326

20:59

Trim: 244mm×172mm

Plant Biomass Conversion

Guha, G. 2007. Simulation of carbon credits in Arkansas from different climate and economic scenarios. The Missouri Valley Economic Association, 44th Annual Conference, Kansas City, MO. Guha, G. 2008. Affirmative climate action in the United States: Emerging issues, adaptation and mitigation initiatives. Forum on Publice Policy, 3, 313–324. Guha, G. & Combs, S. 2008. Global perspectives on pollution control. The Southwestern Society of Economists, Federation of Business Disciplines, Houston. Harvey, F. & Fidler, S. 2007. Industry caught in carbon “smokescreen.” Financial Times, April, 25. Historical Prices. 2010. Point Carbon. Historic Price and Volume Data. Available: http://www. pointcarbon.com/news/promopages/onlinepricedata (accessed November 8, 2010). House, W. 2007. Fact Sheet: A New International Climate Change Framework [Online]. Available: http://georgewbush-whitehouse.archives.gov/news/releases/2007/05/2007053113.html (accessed November 8, 2010). Korosec, K. 2010. Cautionary Carbon Tale: Hackers Cash-In on cap-and-trade [Online]. BNET. Available: http://www.bnet.com (accessed February 5, 2010). Larsen, B. & Shah, A. 1994. Global tradeable carbon permits, participation incentives, and transfers. Oxford Economic Papers, 46, 841–856. Layton, L. 2007. A carbon neutral house? Washington Post. Matthews, H., Hendrickson, C. & Weber, C. 2008. The importance of carbon footprint estimation boundaries. Environmental Science & Technology, 42, 5839. Morrison, C. 2010. Big Oil Pulls Out: Is It RIP for Cap-and Trade? [Online]. Available: http:// www.bnet.com/blog/energy/big-oil-pulls-out-is-it-rip-for-cap-and-trade/3056 (accessed November 8, 2010). Nordhaus, W. 2008. A Question of Balance Weighing the Options on Global Wanning Policies. New Haven, CT: Yale University Press. Pachauri, R. & Reisinger, A. 2007. Climate change 2007: Synthesis report. Contribution of Working Groups I, II and III to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, Geneva, Switzerland. Point Carbon. 2006. Carbon Market Analyst: Carbon trading in the US: The Hibernating Giant, 13 September 2006 [Online]. Available: http://int.pointcarbon.com/getfile.php/ fileelement 86516/CMA US ETS Sept06 hkh9gtpd 1f.pdf (accessed November 8, 2010). RFA (Renewable Fuels Association). 2010. Resource Center—Ethanol Facts—Environment [Online]. Available: http://www.ethanolrfa.org/pages/ethanol-facts-environment (accessed November 8, 2010). Stavins, R. 2008. Addressing climate change with a comprehensive US cap-and-trade system. Oxford Review of Economic Policy, 24, 298. Union of Concerned Scientists. 2006. Cleaner Yards, Dirtier Air? Greetips. US Global Change Research Program. 2001. Climate Change Impacts on the United States—The Potential Consequences of Climate Varability and Change. Cambridge, UK: The Press Syndicate of the University of Cambridge. Sandor, R. & Skees, J. 1999. Creating A Market For Carbon Emissions: Opportunities For US Farmers. Magazine of the American Agricultural Economics Association. Available: http:// ageconsearch.umn.edu/bitstream/32957/1/fo99sa01.pdf (accessed November 8, 2010). USCAP. 2010. Policy Statements [Online]. Available: http://www.us-cap.org/policystatements/ (accessed November 8, 2010). USCBO. 2004. Fuel Economy Standards Versus a Gasoline Tax [Online]. Available: http://www.cbo.gov/ftpdocs/51xx/doc5159/03-09-CAFEbrief.pdf (accessed November 8, 2010).

P1: SFK/UKS P2: SFK c13 BLBS075-Hood December 15, 2010

20:59

Trim: 244mm×172mm

13

Carbon Offset Potential of Biomass-Based Energy

327

USEPA. 2010. SO2 Reductions and Allowance Trading under the Acid Rain Program [Online]. Available: http://www.epa.gov/airmarkets/progsregs/arp/s02.html (accessed November 8, 2010). Victor, D. 2001. The Collapse of the Kyoto Protocol and the Struggle to Slow Global Warming. Princeton, NJ: Princeton University Press. Wilman, E. & Mahendrarajah, M. 2002. Carbon offsets. Land Economics, 78, 405.

P1: SFK/UKS P2: SFK c13 BLBS075-Hood December 15, 2010

20:59

Trim: 244mm×172mm

P1: SFK/UKS P2: SFK c14 BLBS075-Hood December 15, 2010

20:59

Trim: 244mm×172mm

Chapter 14

Biofuel Economics Daniel Klein-Marcuschamer, Brad Holmes, Blake A. Simmons, and Harvey W. Blanch

14.1 Introduction For better or worse, it is impossible to determine the potential impact of a technology in everyday life without assessing its economic viability. As is the case for any product, biofuels will not become a sustainable solution if they do not emerge as an economically viable alternative to fossil fuels, no matter how elegant or admirable the methods used to produce them happen to be. Although such economic assessment may distract from the scientific questions and may—if asked too soon—give a false sense of hope or disappointment, successive evaluation of the progress through economic estimations can help focus the research efforts where they are most needed. Economic questions are particularly important in the case of biofuels, or any novel energy technology, for several reasons. First, energy products (be they liquid fuels, gaseous fuels, heat, or electricity) are used for every human activity. Therefore, increases or declines in prices impact everyone (we implicitly assume the present discussion to be about economies in which energy products are traded, not locally produced). Second, there is already a market for energy products, and there are multiple, mostly nonrenewable, production sources to supply it. This implies that new technologies will have to compete with the existing means of production, even if fostered by incentives such as carbon taxes and subsidies. Third, the scale of the energy market, and thus the investment needed to transform it, is so large that committing to any energy technology without estimating its economic viability would have dire financial consequences. Fourth, energy products made by different technologies are almost indistinguishable to the consumer. Therefore, energy technologies are evaluated and must be able to compete almost exclusively on economic grounds (as opposed to products where image and other aspects play a role in consumer choice). Determining whether biofuels are or will be economically viable can be thought of as a systems question, since several factors interact at different scales to give rise to the observed Plant Biomass Conversion, First Edition. Edited by Elizabeth E. Hood, Peter Nelson and Randall Powell.
C 2011 John Wiley & Sons Inc.

329

P1: SFK/UKS P2: SFK c14 BLBS075-Hood December 15, 2010

330

20:59

Trim: 244mm×172mm

Plant Biomass Conversion

market prices. The focus that appears most relevant here is that which lies closest to the technology, that is, the biofuel production facility. Considering this level of detail allows us to evaluate how changes in the technical parameters alter the economic appeal of the biofuel. It also allows us to prioritize our research efforts and can also serve as a means to quantify the impact of innovation. Even at the level of a production facility, in which we take global supply and demand as relatively constant values, the level of interactions is quite complex, and it is not obvious what effect different changes in operating parameters have on the economic performance of the entire process. In this chapter, we will review how cost estimates are developed within this context and explore the results of previous studies. Some aspects of agriculture and feedstock production are also of interest, and they will be covered as well. In the present chapter, we will focus on lignocellulosic-derived liquid biofuels, and ethanol production in particular. Although such focus is admittedly narrow in scope, we have chosen it because (1) lignocellulosic ethanol is very likely to play a main role in near-future developments, (2) it has been studied in more detail than alternatives such as butanol and alkanes, (3) most methods and some of the conclusions that are applicable to this type of process can be readily adapted to other processes. We use our choice of processes as a case-in-point, not as a statement of favoritism. The chapter is organized as follows: we will first discuss the commonalities of various process configurations, will then describe some major cost drivers, and will explore how changes in process parameters and other choices may impact the economic performance of the biofuel product. Finally, we will consider, briefly, how nontechnical factors affect the economics of biofuels.

14.2 Production Processes The most common production process for biochemically converting biomass into biofuel can be summarized as follows (Figure 14.1). First, the feed is transported to the facility in Feed handling

Feed

Sugar solubilization

Water recycle

Fermentation

Biofuel separation

Biofuel

Byproduct processing

Coproducts

Figure 14.1. Flow diagram of a biofuel production process.

P1: SFK/UKS P2: SFK c14 BLBS075-Hood December 15, 2010

20:59

Trim: 244mm×172mm

14

Biofuel Economics

331

sufficient quantities to support a continuous mode of operation. The feed is treated in such a way that carbohydrates are hydrolyzed into fermentable sugars and toxic compounds are partially or completely removed. The resulting saccharide mixture is microbially converted into fuel molecules via fermentation. These fuels are then separated from the mixture in one or more unit operations, typically through distillation. The fractions of the biomass that are not fermented exit the process as co-products or by-products, generally as combustible material that can be used to produce heat and electricity for internal and external use. We now describe these steps in more detail, providing the relationship between process configuration and economic drivers.

14.3 Biomass Transportation and Handling Biomass is a feedstock and hence we ignore the technological factors that affect its price at harvest (although the economic factors are considered in the next section). The cost of transporting biomass has a direct impact on the optimal location and scale of the processing plant, and this is therefore a recurrent theme in the biofuels economics literature. Steps that take place after harvest and before transportation are needed to ensure that the biomass supply is continuous, even though its production is seasonal. Ideally, continuous operation of the biorefinery is needed to maximize the efficiency of its capital and its volumetric throughput. The types of feedstock used in a lignocellulose-based biorefinery can be generalized to two types: herbaceous and woody biomass. Herbaceous biomass includes feedstocks such as corn stover, switchgrass, and Miscanthus. Woody biomass includes forest residues (tree tops and branches), thinned trees, and proposed energy crops such as poplar and eucalyptus. The inherent properties of these two materials imply different requirements for their handling and processing. The handling of biomass and its relocation to the biorefinery can be summarized as a sequence of four steps: drying, densification, storage, and transportation. Drying is required for various reasons. First, it increases the biomass density prior to formal densification. Second, it makes biomass easier to shred once it reaches the biorefinery (Mani et al., 2006). Third, it protects against parasitic growth (hydrolysis of lignocellulosic components is slower at low water concentrations). Field drying is the cheapest and easiest form of moisture management, although oven drying is needed when water content must be reduced to very low levels or when air drying is inefficient. In general, air-dried biomass has moisture in the range of 15%–20%, whereas oven-dried biomass is close to 0% (Scurlock, 2005). After the biomass is dried, it is compacted to make it more amenable to subsequent operations and to maximize its cost and conversion efficiency. For the case of herbaceous feedstocks, the biomass is left in the field for drying; specialized tractors collect and compress the biomass in a single operation, arranging them in bales. This densification step allows orderly arrangement of the biomass at the storage facility and during transportation to the biorefinery. Woody forest residues would likely be chipped at the growth site and then transported. The relatively low energy density of biomass, in comparison to liquid petroleum feedstocks and coal, makes densification necessary in order to make the produced biofuel cost-effective on the basis of dollars per unit of energy stored and transported. Processing both biomass types into pelletized forms at the production site is also being investigated as a mechanism of densification (Kaliyan and Morey, 2009) and would aid in the standardization of feedstock handling at the biorefinery.

P1: SFK/UKS P2: SFK c14 BLBS075-Hood December 15, 2010

332

20:59

Trim: 244mm×172mm

Plant Biomass Conversion

Transportation is usually carried out by heavy-duty truck, though other modes have been proposed, such as rail and pipeline. Searcy and co-workers, for example, have described several configurations for transport (Searcy et al., 2007). These include truck, truck and rail, truck and ship, and pipeline. In the first instance, the supply of biomass to the biorefinery is limited by the number of vehicles that can be unloaded at the processing site without congesting the roads and the biomass handling area (Kumar et al., 2005). Typically, biomass is transported on flat bed trucks carrying ∼6,000 kg (17 bales) or on loaded logging trucks carrying ∼24,000 kg (Sokhansanj et al., 2002). Rail freight allows bigger shipment sizes and alleviates traffic, but has additional infrastructure requirements and costs (Mahmudi and Flynn, 2006). Ship transport requires vicinity to a river or seaport, which further constrains biorefinery planning and site selection. Pipeline transport of biomass encompasses shredding the biomass at the inlet of the pipe and sending it in slurry form directly to the pretreatment section of the processing facility (Kumar et al., 2005). This last strategy is still being researched and, to the best of our knowledge, has not been explored in the field. Once the biomass arrives at the biorefinery, a series of unit operations prepare the feedstock for sugar solubilization. In particular, the biomass is washed and milled prior to entering the pretreatment section of the plant. As we will see in the next section, there are several technologies available for converting the sugars in the lignocellulosic biomass into their soluble, fermentable form.

14.4 Conversion of Biomass into Sugars Biomass is composed primarily of cellulose (30%–50%), hemicellulose (20%–35%), and lignin (7%–30%), the proportions of which vary depending on the type of feedstock being used. The remaining fraction includes proteins, and ash. After the biomass arrives at the biorefinery, it must undergo a series of steps that liberate the sugars from the complex polymer matrices in which they are arranged. The sugar and lignin polymers are organized and bonded through combinations of covalent and hydrogen bonds, and Van der Waals forces; sugar solubilization entails breaking up as many of these interactions as possible. Lignocellulosic biomass is typically hydrolyzed into its components in a two-stage process. First, the biomass is “pretreated” mechanically (at the harvest site and/or at the biorefinery) and thermochemically. Then, it is subjected to enzymatic hydrolysis using mixtures of glycosyl hydrolases. There are currently several avenues of research being pursued to achieve and fine-tune this process. Pretreatment can be defined as a combination of mechanical and thermochemical processes necessary to increase the rate of enzymatic hydrolysis into its component sugars, while minimizing the formation of toxic byproducts. A thermochemical pretreatment step prior to enzymatic hydrolysis is necessary to facilitate the access of enzymes to the cellulose and to enhance the rate of hydrolysis by 3- to 10-fold (Sousa et al., 2009). The recalcitrance of biomass to enzymatic degradation is believed to be due to the crystallinity of cellulose, to physical barriers formed by lignin and hemicellulose, and to steric hindrances that limit the access of proteins to the cellulose microfibrils (Blanch and Wilke, 1983). The various pretreatment approaches have the common aim of removing or partially rearranging the lignin and hemicellulose, which results in increased enzyme accessibility to the recalcitrant cellulose core, and partial or full decrystallization of the cellulose. Ideally, a pretreatment technology would (Mosier, 2005; Overend et al., 1987) (1) be effective on a wide range of feedstocks, (2) accept minimally prepared biomass (3) obtain each

P1: SFK/UKS P2: SFK c14 BLBS075-Hood December 15, 2010

20:59

Trim: 244mm×172mm

14

Biofuel Economics

333

of the major components with maximum yield and purity, (4) avoid degradation of product sugars or the generation of compounds that can be toxic during fermentation, (5) result in a cellulosic stream that can be hydrolyzed effectively and with minimum enzyme loading, and (6) require low energy inputs and have minimal operating and capital costs. Currently, no pretreatment technology fulfills all these requirements. The greatest challenge in designing an effective process is that the impact of pretreatment on sugar yield must be balanced against its energy requirement, its operating and capital costs, and its effect on the costs of downstream and upstream operations. Current pretreatment technologies can be separated on the basis of their mode of action. Such classification includes mechanical, solvent-based, acid-based, and alkali-based systems. Each vary in their mode of action and in the compounds they produce and, therefore, in the impact they exert on downstream processes (Sousa et al., 2009). Several reports exist in the literature that examine and compare the different pretreatment methodologies currently under investigation, summarized in Table 14.1. Most of these conversion technologies are currently at the research or development stage and have not been deployed for commercial production. Exceptions include dilute acid and steam explosion pretreatments, which have been employed in pilot plants (Mielenz, 2001). Pretreatment times are typically short for high temperature and pressure processes and, except for lime and enzymatic pretreatments, are not the rate-limiting steps in the biofuelsproduction process. The typical pretreatment residence times vary from a few minutes to a few hours depending on the process and feedstock (Table 14.1). However, the effectiveness of pretreatments differ quite considerably, due to the diversity of mechanisms by which they operate (Mosier, 2005). Depending on the pretreatment process, the time required for the subsequent enzymatic hydrolysis to achieve >90% conversion of cellulose vary from 24 hours for solvent-based fractionation such as phosphoric acid and ionic liquids (Li et al., 2009; Zhang et al., 2007) known to render cellulose amorphous (Dadi et al., 2006; Zhang et al., 2006) to 72 hours for acid/base catalyzed pretreatments (Yang and Wyman, 2008). Within the pretreatment process, there is a significant trade-off between the capital cost of equipment and operating costs of chemical and energy requirements. At present, the balance of these factors results in similar overall costs for all pretreatments (Eggeman and Elander, 2005). Efforts are being made to either improve reactor residence times, energy consumption, chemical loading, or recycle configurations to reduce the overall cost of pretreatment. Another trade-off arises between the extent of pretreatment and the cost of downstream operations; in particular, harsh conditions result in higher concentrations of products that inhibit fermentation. Knowledge of the nature and mechanism of the formation of these toxic compounds is incomplete, and thus the focus of intense investigation. (See Saville chapter in this volume.) After pretreatment, the biomass is generally slurried to make it amenable to hydrolysis. The ideal hydrolysis requires enzymes with high specific activity, long half-lives, and negligible substrate and product inhibition. Improvements, either by optimizing hydrolysis directly or by improving the efficacy of pretreatment, would impact both the capital and operating costs of the hydrolysis step as explained in further detail below. Currently, enzymatic hydrolysis is the rate-limiting step of biofuels produced biochemically through fermentation (South and Lynd, 1994). While pretreatment serves to increase the rate of hydrolysis substantially, further reduction in hydrolysis times and enzyme costs remains a challenge. The activity of enzymes on biomass is affected by several key factors, such as the crystallinity and degree of polymerization of cellulose, and the available surface area for enzyme binding. In addition, product inhibition by glucose and cellobiose and cellulase deactivation by irreversible adsorption to lignin increase the operating costs of enzymatic

334 0.015 3

Chemical loading (g/g dry biomass)

Water loading (g/g biomass)

18.0

% Lignin removed

40–45

78

87

All

5

190

4

0.03

SO2

0

91

96

Grasses and wastes

5

90

0.6

1

Anhydrous ammonia

AFEXa

75–80

88

90

Grasses

10

170

2.8

0.5

Ammonia

ARPa

55–60

76

94

All

4 weeks

55

10

0.5

CaOH

Limea

AFEX, ammonia fiber expansion; ARP, ammonia recycle and percolation; ND, not determined. a From Sousa et al. (2009). b From Li et al. (2009). c From Moxley et al. (2008). d From Pan et al. (2006).

92 93

% Xylan conversion

Biomass types

% Glucan conversion

20 Grasses and hardwoods

Time (min)

160

H2 SO4

Chemicals used

Steam Explosiona

ND

81

91

Grasses

15

190

5.25

—

—

Liquid Hot Watera

All

180

120

—

9

Ionic Liquid

Ionic Liquidb

96 ND

75

72

72

82

All

25–100 Grasses and Hardwoods

90–220 60b

1.4–2.9

10–13

Ethanol in acid catalyst

Organosolvd

50

2

11.5

H3 PO4

Phosphoric Acidc

20:59

Temperature (◦ C)

Dilute Acida

Feature

Pretreatment Process

Table 14.1. Typical conditions and characteristics of various pretreatment technologies.

P1: SFK/UKS P2: SFK c14 BLBS075-Hood December 15, 2010 Trim: 244mm×172mm

P1: SFK/UKS P2: SFK c14 BLBS075-Hood December 15, 2010

20:59

Trim: 244mm×172mm

14

Biofuel Economics

335

hydrolysis. Currently, the majority of cellulases are derived from fungi, notably Trichodermia reesei, which produces a suite of cellulolytic enzymes including endocellulases, exocellulases, and β-glucosidases that synergistically depolymerize cellulose (Henrissat et al., 1985). Production from this host requires a relatively expensive process. While the reduction of cellulase production costs has recently improved the operating costs (see below), other areas are lagging. Research is being performed to optimize enzyme cocktails to curb product inhibition, engineer enzymes to be less susceptible to inhibition and deactivation (Jeoh et al., 2008), optimize enzyme production (Kubicek et al., 2009), and to discover new enzymes for cellulose degradation (Kim et al., 2009; Liang et al., 2009; Tucker et al., 1989). Other production systems such as plants can also be considered. (See Chapter 10 in this volume.)

14.5 Conversion of Sugars into Biofuels Once the sugars are freed from the biomass and are in soluble form, they can be fermented to fuels. Conversion technologies can be categorized according to two somewhat arbitrary criteria: the substrates that are converted and the fuel molecules that result from such conversion. Substrates include soluble five- and six-carbon sugars (Delgenes et al., 1996), untreated plant matter (Demain et al., 2005), glycerol (Yazdani and Gonzalez, 2007), syngas (CO + H2 ) (Henstra et al., 2007), and municipal waste streams (Lang, 1978; Li et al., 2007). Fuels include ethanol (Bothast and Schlicher, 2005), butanol (Ezeji et al., 2007), higher alcohols (Atsumi et al., 2008), and hydrocarbons (Kalscheuer et al., 2006). Most of these conversion technologies that involve biofuels other than ethanol are currently at the Research and Development stage and have not been deployed for commercial production. Each conversion technology (for a given choice of substrate and fuel) can be evaluated on different grounds that affect the overall economics. Three parameters are most frequently quoted: product concentration, product yield, and specific or volumetric productivity. Product concentration primarily affects the capital cost and energy use of the separation operations downstream of fermentation. Yield influences the cost-effectiveness of fuel production per unit of feedstock input to the conversion facility, which is a key variable given the impact biomass prices have on biofuel cost (see below). Productivity alters the capital cost of the plant, as larger (or more) vessels are needed to make up for losses in this parameter. To date, ethanol has been the biofuel of choice. Yeast such as Saccharomyces cerevisiae ferments six-carbon (but not five-carbon) sugars into ethanol with near-theoretical yields (0.51 g/g of glucose), making them ideal for production starting from grain- or sugarcane-derived sugars. In addition, centuries of selective breeding have produced strains of Saccharomyces that are quite tolerant to fermentation conditions, a trait that is difficult to replicate in other microorganisms (Fischer et al., 2008). The need for conversion technologies that would reduce some of the drawbacks of grain or sugarcane derived ethanol has ignited interest in metabolic engineering of microbes. The ideal biofuel production process would convert most of the saccharides into a water-immiscible fuel with high productivity. The feedstock would be abundant, cheap, and would not compete with food, and the fuel would have a high energy density and similar properties to their existing oil-derived counterparts. To date, most research programs in this field aim at advancing one or more aspects towards this ideal scenario. One approach involves enhancing the efficiency of ethanologenic organisms to convert lignocellulose-derived sugars. Conversion of biomass hydrolysates presents a challenge for

P1: SFK/UKS P2: SFK c14 BLBS075-Hood December 15, 2010

336

20:59

Trim: 244mm×172mm

Plant Biomass Conversion

Saccharomyces, the traditional ethanol-producing organism, as five-carbon sugars comprise a significant fraction of the fermentable compounds in most kinds of biomass. Widespread efforts in metabolic engineering have been applied to obtaining yeast strains that can ferment these sugars, in particular xylose (Jeffries, 2006; Van Vleet and Jeffries, 2009). This approach would enhance the effective yield of biofuel on biomass, mitigating the impact of feedstock expenses on production costs. Since the theoretical yield of ethanol (on a mass basis) on xylose can be as high as that on glucose, the inability to utilize xylose increases the net relative cost of feedstock proportionally to the content of this five-carbon sugar in biomass. Microorganisms that can ferment both five- and six-carbon sugars can also produce ethanol with high yields (e.g., Zymomonas mobilis or ethanologenic strains of Escherichia coli), but these tend to be less robust, with decreased product concentrations and productivities (Almeida et al., 2007). Another approach, therefore, has focused on increasing the tolerance of these organisms to the biomass hydrolyzates and the biofuel product, both of which inhibit growth (Gutierrez et al., 2006; Wang et al., 2008). Yet another approach involves organisms that can directly ferment lignocellulosic materials into fuels. For example, a few strains of the genus Clostridium are ethanologens and produce cellulolytic enzymes, needing a single step to release sugars and convert them into fuel (socalled “consolidated bioprocessing”) (Cooney et al., 1978; Madia et al., 1979). Furthermore, some species can withstand high temperatures (Timmons et al., 2009), allowing simultaneous, yet partial, removal of ethanol while minimizing the possibility of microbial contamination (Fischer et al., 2008). Operating the process at higher temperatures would theoretically reduce the capital and operating cost of separation units. It would also allow the saccharification reactions to occur at their optimal conditions during fermentation, lowering the capital cost of the saccharification reactor. These temperature-tolerant strains are, however, not naturally tolerant to ethanol and produce small quantities of the fuel (Balusu et al., 2005; Timmons et al., 2009). Using low-cost or waste materials has also been considered to improve the economics of biofuel production. For example, glycerol has been shown to be a good substrate for ethanol production, converting a by-product of the biodiesel industry into biofuel (Yazdani and Gonzalez, 2007, 2008). Municipal waste streams have also been considered, although the variability in their composition makes them harder to handle (Becker et al., 1981; Kalogo et al., 2007; Li et al., 2007). Using feedstocks that are cheap or can be obtained at minimal costs could eliminate one of the largest costs of the process, although if demand for these increases, they can be expected to become more expensive. The availability of these feedstocks, supply and demand forces, and process performance metrics, would dictate whether these can become mainstream or niche technologies. Hydrocarbon biofuel production provides a further alternative. These compounds have the advantage of being water-insoluble and are readily separated from the fermentation broth with minimal energy use. They can also be used directly as gasoline or diesel substitutes. The savings would come not only from advantages in the process itself, but also from avoiding the need to introduce new distribution and utilization infrastructure. Efficient and fast production platforms for lignocellulose-derived hydrocarbons have, therefore, become a growing target for biofuels research. Engineering microbes for hydrocarbon production can be accomplished, for instance, by introducing genes that scavenge intracellular pools of fatty acids and ethanol and condense them into ethyl esters (biodiesel) (Kalscheuer et al., 2006). Genetic manipulations that increase these pools have been additionally considered for improving the performance of these organisms (Keasling and Chou, 2008).

P1: SFK/UKS P2: SFK c14 BLBS075-Hood December 15, 2010

20:59

Trim: 244mm×172mm

14

Biofuel Economics

337

The commercial realization of one or more of these technological advances will favor the economics of the process in different ways, depending on their impact on the cost drivers to be covered in the next section.

14.6 Separation and Purification The broth that exits the fermentors would contain approximately 4%–5% v/v ethanol, extrapolating laboratory-scale results (Dien et al., 2000; Lau and Dale, 2009; Maiorella et al., 1984). This highly dilute ethanol solution must be dehydrated to near purity, most commonly through distillation. The process can be conceptually described as boiling the water out of solution, and the energy consumption of this step is closely related to the vaporization enthalpy of water. In fact, distillation represents the single largest energy-input into a corn-ethanol plant (Maiorella et al., 1984). The distillation energy would be larger for the case of lignocellulosic ethanol, as the concentration of ethanol in this process is expected to be lower than for the case of corn starch, at least with current technology because the sugar stream is more dilute. In contrast to the 4%–5% ethanol previously quoted, the input stream to the distillation section of a corn-ethanol plant is 10%–12% ethanol (Bothast and Schlicher, 2005). Distillation has several advantages. First, it has been extensively used at large scale and studied in detail for decades, particularly by the petrochemical industry. Second, it enjoys economies of scale, making the separation cost efficient in large facilities (Vane, 2006). Third, it is amenable to substantial heat integration schemes, which alleviates a fraction of the energy use (Aden et al., 2002). Other technologies have been proposed to improve upon the performance of distillationbased processes. In particular, separation of ethanol from the broth concurrently with fermentation allows relieving product inhibition of the microbes as the ethanol accumulates. Pervaporation offers one such example (others include extractive fermentation, gas stripping, and vacuum distillation (Lipnizki et al., 1999)). Pervaporation uses a semi-permeable membrane that preferentially permits diffusion of ethanol over water, producing a concentrated ethanol solution. Addition of this step eases distillation and lengthens the productive period of fermentation. For example, O’Brien and co-workers reported a fermentation in which the ethanol concentration was maintained below 25 g/L by pervaporation, producing a concentrated solution of 17% w/w ethanol (O’Brien et al., 2004). A preliminary economic analysis by the same group concluded that minor improvements to membrane performance would render the pervaporation process advantageous compared to the distillation-only scheme (O’Brien et al., 2000). Other researchers have reported a 2-fold increase in productivity of ethanol when fermentation is simultaneously operated with a pervaporation unit (Nakao et al., 1987). The realization of these and other advantages are bound to lower the overall cost of ethanol production. Let us now consider the fate of co-products that directly impact the economics of biorefineries.

14.7 Co-product Handling The co-products of the biofuel process can be as important as the biofuels themselves, since no process configuration would be economically viable today without their consideration. In addition, life-cycle analyses (LCAs) of biofuels usually consider co-products as credits when

P1: SFK/UKS P2: SFK c14 BLBS075-Hood December 15, 2010

338

20:59

Trim: 244mm×172mm

Plant Biomass Conversion

they displace energy inputs or greenhouse gas emission sources inside the system boundaries (defined by the LCAs). In the case of corn-derived ethanol, the main by-product is distillers’ grains with solubles (DGS), which comprises all the nonfermentable material present in corn plus the biocatalyst used in the conversion (usually yeast). This mixture is prepared by drying the distillation bottoms (or stillage) to a lesser or greater extent. The wet or dry DGS (DDGS) can be sold as animal feed, being rich in protein and other necessary nutrients. On average, each unit weight of corn yields about a third of a unit of DDGS. In the case of lignocellulosic ethanol, non-fermentable material is principally lignin, found in the distillation bottoms and often times burned after it is dried together with the yeast. Lignin has a high heating value (HHV) of about 21–26 kJ/g, compared to cellulose at ∼17.5 kJ/g (Energy Efficiency and Renewable Energy, Biomass Program, 2009; Demirbas, 2001). Ligninrich residues can be burned to produce electricity and steam to supply the process needs. In addition, excess electricity can be sold to the grid. Estimates have shown that burning lignin-rich residues from ethanol production could produce 458 TWh of electricity annually, or about 3.6% of world consumption at the time of the study (Kim and Dale, 2004). This lignin would be the co-product resulting from processing 500 gigaliters per year of ethanol from the entire global supply of waste biomass. At any scale, the surplus electricity subtracts from the carbon emissions balance. One analysis of ethanol fuel in Brazil showed that the use of E25 (a 25% ethanol blend in gasoline) represents a carbon credit of 62 kg CO2 -eq/m3 (Macedo et al., 2008). For all its benefits, burning lignin is a subpar fate for this residue. Electricity is relatively inexpensive, and producing and selling it does not offset the drawbacks of starting with lignocellulosic material. Ideally, the lignin fraction would be converted to commodity or even higher value-added products, making the process economically more attractive. Biomass pretreatment technologies that aid in separating the lignin before it enters the fermentor would help in this regard. A lignin-rich stream could be purified from the biomass feedstock and subsequently processed into resins, automotive products, biopolymers, and other commercial products (Lora and Glasser, 2002; Pandey et al., 2000). Such development would hasten the realization of biofuels, not only due to better balance sheets, but also through a diversification of revenue streams.

14.8 Major Cost Drivers 14.8.1 Biomass-Associated Costs Several studies indicate that biomass cost is a significant component of the total cost of producing biofuels, even for lignocellulosics. For example, Aden and Foust calculate it to be 38% of the cost of producing ethanol by fermentation when treating the biomass with dilute acid prior to enzymatic hydrolysis (Aden and Foust, 2009). A study by Wingren et al. agrees, placing this figure approximately between 30% and 40% depending on whether simultaneous or separate saccharification and fermentation are implemented (Wingren et al., 2003). These numbers are large, but appear less so when compared to equivalent figures for grain-based biofuels, where feedstocks amount to 60% or more of the total production cost (Enguidanos et al., 2002; Perlack and Turhollow, 2003). The cost of lignocellulosic biomass can be regarded as a sum of expenses, highlighted in bold in the ensuing discussion. The first such expense derives from growing the feedstock,

P1: SFK/UKS P2: SFK c14 BLBS075-Hood December 15, 2010

20:59

Trim: 244mm×172mm

14

Biofuel Economics

339

which requires fertilizer, water irrigation, and labor. This is known in the industry as the “grower payment.” Depending on the type of feedstock and the opportunity cost of using it for other purposes, the grower payment varies from $11–$44 per dry ton (Hess et al., 2009). Another study further pinpoints the figure to an average of $15.90/dry ton (at the time of publication), specifically for corn stover (Hess et al., 2009). Other contributors to the cost are harvest, collection, and baling. The main expenses during these operations are the capital cost of machinery and labor. For the case of corn stover, initial shredding, baling, loading, and stacking amounts to $24 per dry ton (2002$) (Sokhansanj and Turhollow, 2002, 2004). Some authors also include, separately or in addition, a farmer premium in the cost of biomass, which can amount to nearly 20% of the total cost of the delivered feedstock (Aden et al., 2002). As mentioned earlier, and to minimize the capital cost of the biofuel production facility, the biomass must be supplied continuously regardless of the growth cycle, which requires storing the feedstock before it is processed. For storage, estimates vary from $2 to $17 per dry ton (2007$) (America’s Energy Future Panel on Alternative Liquid Transportation Fuels, 2009). Storage prices are affected by the price of labor (for hauling and maintenance) and rent at the storage facility. Transportation expenses can be separated into distance variable and distance fixed costs (DVC and DFC, respectively). The DFC is related to the equipment and the time required to load and unload each shipment, while the DVC depends on the distance from loading to unloading sites (Kumar et al., 2005). Truck is by far the most common assumed mode of transport of biomass to biorefineries. For the case of corn stover, this mode of shipping has an estimated DVC of $0.12 per ton per km and a DFC of $4.39 per ton (both are in 2004$ and are based on actual weight, not on dry weight) (Searcy et al., 2007). The cost of transporting biomass strongly influences the optimal size of a biorefinery. In particular, a trade-off arises between the DVC and the economies of scale that can be achieved at the plant. For example, small plants demand relatively little biomass, which can be collected from a small radius with low associated DVC. This combination, however, results in inefficient use of the capital cost per unit of biofuel output, which increases the relative cost of production. Larger plants requiring large biomass quantities fall at the other end of the spectrum. Depending on the biomass availability per unit area and the factors that affect the DVC, the optimal size of a plant can be calculated (see, for example, Kumar et al. (2003)). Various studies have estimated the final price of biomass accounting for the contributing factors mentioned above. As shown in Table 14.2, there is considerable variability in the reported prices even when the same biomass feedstock is considered. Such differences stem from uncertainties in the assumptions made for estimation and from varying degrees of rigor (i.e., level of detail considered) with which the costs are calculated. More often than not, the opportunity costs of land and alternative biomass uses are not included, which may lead to cost underestimation. Preferably, biomass cost should be derived from field studies of representative scale to ensure the appropriate accounting of the pertinent factors, in addition to geographical location. Perrin and co-workers offer one such study (Perrin et al., 2008), in which they analyze the production costs of switchgrass in ten farm sites. These were on average 6.7 ha sites located in Nebraska and the Dakotas, where switchgrass was grown for a period of 5 years. The researchers recorded the actual incurred costs of production by the farmers, even when the prescribed management practices were not perfectly followed. On average, the fields yielded 5 ton/ha and produced switchgrass at $65.86 per dry ton. When only the five lowest-cost sites were

P1: SFK/UKS P2: SFK c14 BLBS075-Hood December 15, 2010

340

20:59

Trim: 244mm×172mm

Plant Biomass Conversion

Table 14.2. Feedstock costs for different crops. Biomass

Description

Cost ($/tonne)

Year of Analysis

Reference

Corn stover Corn stover

At farmgate, estimated Delivered (∼50–180 km), estimated

33 60–72

2006 2000–2004

Graham et al., 2007 Petrolia, 2008

Corn stover

Delivered (35–100 km), estimated

43.1–51.6

2003

Perlack and Turhollow, 2003

Switchgrass Switchgrass Switchgrass

Production cost, actual Production cost, projected Delivered (77 km), estimated

65.86 46.26 37–48

2000–2004 2000–2004 2006

Perrin et al., 2008 Perrin et al., 2008 Kumar and Sokhansanj, 2007

Switchgrass Switchgrass Wheat straw Miscanthus Miscanthus

At farmgate, estimated Delivered (64km), estimated Delivered (76km), estimated At farmgate, estimated Delivered (64km), estimated

56.93 64.84 37 41.67 49.58

2003 2003 2002 2003 2003

Khanna et al., 2008 Khanna et al., 2008 Hess et al., 2007 Khanna et al., 2008 Khanna et al., 2008

considered (purportedly the best managed), this number was $51.95 per dry ton, and further reduced to $46.26 per dry ton when the costs were projected to a 10-year rotation. Unfortunately, such studies are difficult to set up and oversee, and the effects of technological advances in feedstock economics will rarely be explored at this level of detail. Most frequently, cost calculations must be done through estimation. It is imperative that researchers account for all inputs and market forces at play, if their results are to be representative of real scenarios.

14.8.2 Capital Expenses The estimation of capital expenses in a biorefinery is based on several fundamental assumptions and design considerations, namely the plant size, the feedstock that will be used, and the process type and configuration (Wooley et al., 1999). As the feedstock, pretreatment and fermentation processes employed exhibit system interactions with the rest of the plant, it is necessary to perform initial design and feasibility studies of each process configuration in order to give an accurate assessment of the capital and operating costs of each configuration (Figure 14.2). For example, steam explosion and dilute acid pretreatment technologies implemented on the pilot and demonstration scale suffer from reduced enzymatic and microbial conversion efficiencies caused by inhibitors delivered from the pretreatment process (Dunnett et al., 2008). The capital costs of ethanol production facilities reported in the literature have been based primarily on models developed with process modeling software (e.g., ASPEN/ICARUS, SuperPro Designer) in conjunction with multiplication factors and cost indices for direct fixed costs. These factorial methods are relatively well established and are the basis for most cost estimations in the chemical industry today (Mosberger, 2005). More detailed studies have relied on vendor information or on specialized consulting services. For these types of feasibility analyses, design is performed on the largest unit operations in the plant, and factorial methods used to estimate the cost of ancillary equipment based on the preliminary costs for major equipment. Commonly used, though sometimes limited, treatments of these subjects can be found in reference works such as Perry’s Chemical Engineering Handbook (Green and Perry, 2007) and Ullmann’s Encyclopedia of Industrial Chemistry (Mosberger, 2005).

P1: SFK/UKS P2: SFK c14 BLBS075-Hood December 15, 2010

20:59

Trim: 244mm×172mm

14

Biofuel Economics

341

Example factors that contribute to biorefinery costs Capital cost

Operating cost

Indirect cost

• Plant size

• Feedstock price

• Capital depreciation

• Process configuration (e.g., pretreatment method, separation operations, etc.)

• Consumption of raw materials (e.g., enzymes)

• Financing interest rates

• Overall yield

• Fuel productivity

• Technological risk

• Energy requirements and energy efficiency

• Equipment materials of construction

• Coproduct revenues

• Coproduct processing infrastructure

• Inflation

• Policy support • Oil price volatility • Alternative energy technologies (opportunity cost)

Figure 14.2. Example factors that contribute to biorefinery costs.

Like all chemical industries, ethanol processing-plants are subject to economies of scale wherein increasing facility size results in decreasing manufacturing cost. This trend is no different for ethanol plants, based on the results of various studies (Dwivedi et al. 2009; Huang et al. 2009). However, and as described above, feedstock transport cost makes up a significant proportion of the feedstock net cost, and so there is a maximum output that is economical before gains in economies of scale from the refining perspective are negated by increasing costs of biomass transport. This has led others to propose models of biofuel facilities that are distributed rather than centralized (Dunnett et al., 2008). This may well be an unusual business model, but it could be worthy of consideration given the low energy density of biomass. Similarly, considerations must be made for the materials of construction of different pieces of equipment, in particular for pretreatment. Process operations that necessitate the use of high pressures or corrosive chemicals will have an impact on equipment cost. The requirement for expensive materials of construction can be significant enough to influence the preference of pretreatment method or operating conditions. Several techno-economic analyses and reports have been published on the subject of estimating the capital and operating expenses (and subsequent ethanol selling prices) of cellulosic ethanol production (Aden and Foust, 2009; Dunnett et al., 2008; Eggeman and Elander, 2005). Irrespective of the estimated price provided by these studies, most conclude that the main areas for which technological improvement would make a significant impact on capital cost and process economics are pretreatment operations, saccharification, and fermentation. Examples include reduction of residence times through either increased enzymatic kinetics, increased conversion efficiency, or possible use of simultaneous saccharification and fermentation or consolidated bioprocessing (Aden and Foust, 2009; Dunnett et al., 2008; Hamelinck et al., 2005; Sendich et al., 2008). Improvements in any of these areas would either decrease the required facility size or conversely increase product output. The costs of individual processing steps as a percentage of the total fixed capital investment are shown in Figure 14.3 (adapted from Aden and Foust (2009)).

P1: SFK/UKS P2: SFK c14 BLBS075-Hood December 15, 2010

Trim: 244mm×172mm

Plant Biomass Conversion

Pr et re at

m en

t

342

20:59

Other costs

ing

ion

n Co n/

dit

tio

za

li ra

ut

Ne

tion

enta

erm sis/F

droly e hy

ym

Enz

Dis

tila tio

na nd

soli

Boiler/Turbogenerator

Utili

ties

ds r Wa ecove ry ste w St ate ora r ge

Figure 14.3. Contribution of processing steps as a percentage of total fixed capital investment (adapted from Aden and Foust (2009)).

14.8.3 Operating Costs In order for cellulosic ethanol to be cost competitive, it is essential to minimize the operating costs of the process. There are several areas where operating costs are most significant, in particular pretreatment (chemicals and energy), enzymatic hydrolysis (chemicals), and distillation (energy). Furthermore, as the feedstock costs are predicted to comprise up to 50% of the operating cost, efficient utilization of biomass is essential. Minimizing sugar degradation during pretreatment and maximizing the efficiency of fermentation can help to reduce overall operating costs. To offset the low margins anticipated from the sale of ethanol, it is also important to effectively use lignin from the process for the co-generation of steam and electricity (with significant incurred capital costs) or for the production of value-added co-products. Waste handling can also play a large role in the success of a biorefinery process. High ash feedstocks (e.g., rice straw) and pretreatment technologies that produce significant quantities of waste solids (e.g., dilute acid pretreatment) could pose both operating and environmental problems. The waste from fermentation and distillation that is not burned for electricity and steam generation could possibly be recycled as fertilizer for crops, reducing the cost of treatment. Cellulolytic enzymes represent another significant fraction of the production cost. The large amount of cellulase enzymes required, together with their cost on a per weight basis, has constrained their use for the production of lignocellulosic biofuels (Himmel et al., 1999). (See Chapter 10 in this volume for alternatives.) The recent reduction of cellulase production costs by 20- to 30-fold (Falhout and Nedwin, 2005) by commercial cellulase manufacturers has offered hope in this regard, but further reductions are needed nonetheless. Regardless of the actual price, a trade-off exists between the capital and operating contributions to the production price. Saccharification is a lengthy process that may take up to 7 days, which translates into a requirement for numerous large vessels. Increasing the enzyme loading may reduce the residence time of saccharification, and thus the associated capital cost, but increases the operating cost. Research on faster or more active enzyme mixtures or a reduction

P1: SFK/UKS P2: SFK c14 BLBS075-Hood December 15, 2010

20:59

Trim: 244mm×172mm

14

Biofuel Economics

343

in their unit price could help ease the impact of the hydrolysis step. Developing enzymes and/or microorganisms that can perform in the same conditions could aid in combining the saccharification and fermentation steps most effectively, without sacrificing residence times of the respective operations. Energy integration is another key consideration for reducing operating costs. The heat that must be provided to the pretreatment and distillation sections can, and should, be used to supply heat for other operations, such as the evaporators that partially dry the co-products for burning or selling. Even if energy integration is used, large energy requirements for distillation of very dilute ethanol solutions result. Studies show that the cost of distillation increases markedly when the concentration in the feed to this operation is below 6% (Maiorella, 1984). Low ethanol concentration can result from inhibition during fermentation or low sugar concentrations in the feed to this operation. In this case, little or no heat excess would be available for electricity production, eroding the margins of biofuel production even further. The energy cost of separation, among other factors, is a main driver for developing biofuels or processes that are less energy intensive. For example, next-generation biofuels could benefit from the development of two-phase solvent extraction systems for simultaneous separation and fermentation (Roffler et al., 1988) or the production of fuel molecules that will phase separate from fermentation broths (Keasling and Chou, 2008).

14.9 Risks Private investment in biofuels, a prerequisite for an economically sustainable industry, will only result from the development of profitable processes. Such profitability depends on both the cost of producing the biofuel and on the price of alternatives, notably oil-derived liquid fuels. Both these parameters are plagued with uncertainty and, in combination with other factors, translate into financial risk for potential investors. Risks, in turn, translate into actual costs, as higher returns on investment are required to lure capital into unpredictable projects. One major source of risk stems from technological uncertainties. Most of the biomassderived ethanol produced commercially today is derived from corn or sugarcane, which rely on technologies that have been used for decades. The lignocellulosic-based process, on the other hand, has not been widely established, and thus, its performance at large scale remains undetermined. One way of quantifying the potential impact of this uncertainty in the economics is to perform sensitivity analysis of different process variables. A few authors have undertaken such analyses, and some of their results are summarized in Table 14.3, which gives the percent change in ethanol production cost as a function of the percent change in a given process parameter. For example, a 67% increase in feedstock cost (corn stover in that case), resulted in a ∼21% increase in production cost. As can be appreciated from the overlap of data from different sources, the technological risk is not only associated with the uncertainty in process parameters, but also with the fact that different studies predict effects of different magnitudes. Although sensitivity analysis on a single variable can uncover sources of risk, simultaneous deviations in several variables from their assumed values could potentiate the uncertainty on the economic performance metrics. To study these possible complex effects, Monte Carlo-type simulations have been undertaken in the context of biofuel technology. In this analysis, the input variables are given a probability distribution according to which their values are sampled; the process is modeled repeatedly with different sampled values of the input variables resulting

P1: SFK/UKS P2: SFK c14 BLBS075-Hood December 15, 2010

344

20:59

Trim: 244mm×172mm

Plant Biomass Conversion

Table 14.3. Sensitivity analysis for the lignocellulosic ethanol process (percent change in ethanol production cost as a function of the percent change in a given process parameter).

Input Variable

% Change in Input

% Change in Production Cost

Reference

Enzyme cost Substrate loading Saccharification residence time Feedstock cost Saccharification residence time Electricity credit Enzyme loading Enzyme loading Increase ethanol titer

29% 46% 58% 67% 100% −50% 42% −50% 40%

12.10% −6% 3.80% 20.60% 1% 4.70% 4.70% −13% −11%

Wingren et al., 2003 Wingren et al., 2003 Wingren et al., 2003 Aden et al., 2002 Aden et al., 2002 Aden et al., 2002 Aden et al., 2002 Lynd, 2008 Lynd, 2008

in a distribution for the output variables. Monte Carlo analysis has been used in the context of bioenergy, for example, to study the effect of process parameters on biofuel production cost (Aden et al., 2002), to compute the impact of energy and feedstock costs on biofuel expected profits (Rozakis and Sourie, 2005), and to model the emissions that result from biofuels processing (Fargione et al., 2008). Other causes of risk, in addition to technical uncertainty, also decentivize investment in lignocellulosic biofuels. Volatility in oil prices complicates estimating the demand for biofuels: renewable alternatives can only compete with gasoline and diesel if the prices for the fossil raw materials are above a certain threshold. This effect has been marked in the recent past, as oil spot prices surged from $60/bbl in 2007 to almost $150/bbl in mid-2008, back to $40/bbl in early 2009, and up to $70/bbl 8 months later. These fluctuations drove capital in and out of the biofuels industry, leaving many bankrupt and others disenchanted with its promise altogether. The development in biofuels has been occurring simultaneously with an impressive boom in all renewable energy technologies. It could be argued that the plethora of options has hampered investment as much as it has helped it. Investors may perceive that choosing among the different technologies is a gamble and may be uncomfortable picking winners too early into the race. Another deterrent for venturing into biofuel production is the scale of the investment needed even for a single plant (in the hundreds of millions of dollars). In the future, once the industry is established, this may serve as a barrier of entry for competitors, but has no real advantages for its development today. The large scale of investment, and the credit crunch that accompanied the recession of the past year, has forced governments to take up the challenge, many times dwarfing the enthusiasm of the private sector. There are a few strategies that can partly relieve the financial risk. Diversification of products at the biorefinery, such as that described previously, could lower the risk by allowing the shareholders to recover their investment in various markets simultaneously. If one product becomes transiently noncompetitive (e.g., if a competing technology lowers its price), earnings can rely on other products until additional advances boost profits once more. Another strategy is to reduce risk associated with factors external to the facility, for example, by connecting new facilities to old ones that rely on proven technology (e.g., adding a lignocellulosic ethanol processing-plant to a corn ethanol facility) (Schubert, 2006). Yet another strategy, though external to the decisions of investors in the private sector, is market support by government.

P1: SFK/UKS P2: SFK c14 BLBS075-Hood December 15, 2010

20:59

Trim: 244mm×172mm

14

Biofuel Economics

345

14.10 Policy Support Governmental intervention is usually justified as a form of risk management, and has been widely applied to energy technologies. For the case of biofuels, mandates and subsidies aim at ensuring they remain an attractive alternative in the face of inexpensive oil. For example, the Energy Independence and Security Act of 2007 mandated the use of 36 billion gallons of ethanol by 2022, 21 billion of which must be from “advanced technologies” that do not use grains as feedstock. In addition, subsidies between $1.1 and $1.3 per gallon have been applied to ethanol in the recent past, topping $92 billion in projected expenses in the 2006–2012 period (Koplow, 2007). These figures have ignited significant controversy, raising questions of whether biofuels are economically sustainable and/or justified. It must be noted, however, that at present most, if not all, renewable energy technologies are subsidized in one form or another. This is needed to balance the fact that carbon emissions that result from the use of fossil fuels are not priced (that is, they are unaccounted externalities). In addition, fossil fuels are incentivized in many countries with subsidies and tax credits. For the case of ethanol, and specifically for the U.S., government incentives come from several overlapping fronts including federal, state, and local measures. Figure 14.4 shows the stages of the supply chain at which the government commonly intervenes and illustrates how the financial aid is applied. In addition to subsidies, government support comes as tax credits and exemptions, import tariffs, and production-associated payments (Koplow, 2007). Governmental support reduces the risk for potential investors by increasing the economic prospects of a technology, but in order to be effective, it must be temporary. This forces the beneficiary industry to improve and innovate until it is capable of subsisting without help. If the period is too short, however, the industry can fail before it has time to become independent. A recent study estimated that if biofuel subsidies were to suddenly disappear, the ethanol industry would contract by 30% and the biodiesel industry by 50% (Westhoff et al., 2007).

• Subsidies to the supply of inputs • Crop and irrigation subsidies • Subsidies for energy sources

• Subsidies for purchasing biofuels • Subsidies for purchasing vehicles

• Subsidies for biofuel storage • Subsidies for distributuion

Process inputs

Biofuels

• Production-based payments • Tax credits and exemptions • Market price support (e.g., import tariffs) Biofuel plant

Byproducts

• Subsidies to labor • Subsidies for purchasing capital • Subsidies to land Value adding factors

Figure 14.4. Inputs and outputs flow-chart.

• Subsidies for purchasing and consumption of byproducts

P1: SFK/UKS P2: SFK c14 BLBS075-Hood December 15, 2010

346

20:59

Trim: 244mm×172mm

Plant Biomass Conversion

Governmental intervention could not only help, but also harm the development of the biofuels industry. Mandates can create oversupply, which can hurt balance sheets for both the private and public sectors. For instance, if 36 billion gallons of ethanol would be deployed today and mixed with the current gasoline use of 138 billion gallons (Energy Information Agency 2008) that would create a blend upwards of 20%v/v ethanol, even if we assume that demand grows to accommodate for the additional fuel. This blend is regarded by experts as incompatible with current infrastructure (see below). As a result, much of the ethanol would have to be exported or used for applications other than transportation fuels, its price would drop, and subsidies would likely be required to make up for the profit difference. For these reasons, internalizing the environmental cost of carbon emissions so that the benefits of biofuels are represented in their price has been argued as a market-led alternative to direct governmental intervention. Exactly how would this be executed has been the focus of much debate; proposals have ranged from a tax on emissions to internationally tradable permits. The politics of implementing these schemes are, fortunately, outside of the scope of this discussion.

14.11 Infrastructure and Vehicle Modifications A major cost driver that is not usually discussed when performing process economic analysis is that related to modifications in infrastructure and vehicle fleet that could be needed for significant market penetration. These can be considered external to the design of the biofuel process, but they affect the profitability of the product through demand forces. Consider the case of ethanol. Regarding compatibility with existing infrastructure, the main reasons for concern are the blended fuel’s corrosiveness and hygroscopicity that arise due to increased water miscibility as a function of ethanol content. Suitability in vehicle systems is further challenged by the additional oxygen present in higher ethanol blends, which affects the fuel cycle (Hammel-Smith et al., 2002). The Underwriters Laboratories does not certify blends of gasoline and ethanol beyond E10 in distribution and storage facilities (Braeutingam, 2009). Furthermore, vehicle warranties do not cover the owner if blends above that ethanol concentration are used, unless the engine is specifically designed for them (e.g., flex-fuel vehicles) (Westcott, 2009). This “blend wall” acts by effectively curbing demand at a quantity above 10% of the gasoline use. Currently (October, 2009), ethanol production capacity in the United States is such that, if blended completely with the gasoline used nationwide, it would produce an E7 fuel. This implies it will not be long before large-scale adaptations are needed. One area would be in ethanol distribution. Ethanol is commonly transported to a fuel terminal by rail, truck or barge, and then blended with gasoline to be further distributed with it. Pipelining of ethanol has also been considered as an option for transporting it, but at significant initial capital cost. Infrastructure modifications would have to be implemented at each step of the supply chain, in addition to investments in storage tanks at the terminals and retail sites. According to a recent study, the cost to the process of transporting ethanol by truck, an average distance of 100 km, varies between $0.05 and $0.15 per GJ of biomass feedstock, depending on the type of feedstock and the production scale (both in 2004$). The cost is $0.15–$0.80/GJ for the same distance if the transportation is by pipeline (Searcy et al., 2007). The difference in cost between the truck and pipeline options decreases as the production scale

P1: SFK/UKS P2: SFK c14 BLBS075-Hood December 15, 2010

20:59

Trim: 244mm×172mm

14

Biofuel Economics

347

increases, implying that the pipeline is more effective at accommodating larger volumes of production. To put these numbers in perspective, let us assume a retail price for ethanol similar to that of gasoline adjusted for energy density—which is admittedly unreal, but suffices in this analysis. In that case, the cost of transportation given in the preceding paragraph would be up to 13% of the retail price. This does not include distribution costs other than transportation, which can be significant, nor include marketing and other indirect costs. For example, the California Energy Commission estimated a cost of $60 million (1998$) for terminal modifications in that state needed to supply E17–E24 blends (Hammel-Smith, et al., 2002), which are separate from transportation costs. As a comparison, the distribution and marketing cost of gasoline in 2008 was 12% of the retail price, according to the EIA (Anonymous, 2009). Modifications in the vehicle engines would also be required. A recent study estimated the investment in infrastructure and vehicle fleet modifications to supply 150 billion gallons of ethanol per year, an amount reportedly needed to reduce a gigaton of CO2 -equivalent emissions by 2020 (Spatari et al., 2009). Their estimated infrastructure costs per gallon of ethanol supplied are as follows: $0.08/gal in regional distribution facilities for storage and shipping, $0.25/gal in transportation (rail) infrastructure, $0.05/gal in conversion of gas stations, and $0.13/gal in new vehicle conversion (their estimate is $70 per car in new vehicles, which increases to $1300 per car if they are retrofitted). Although these changes may be seen as expensive when considering the production scale needed, a series of studies from the same organization concluded biofuels to be one of the least expensive routes for CO2 abatement. It remains to be seen what modifications are actually implemented, and whether or not the estimated costs reflect the real expenses.

14.12 Conclusions The economic performance of biofuel alternatives has been the focus of several studies spanning every aspect of the supply chain, from growing energy crops to using the fuel in specially fitted vehicles. In this review, we have described the methodology and results from some of this work, in particular focusing on lignocellulosic ethanol. Despite the increasing levels of detail that these studies have achieved, it is too early to determine whether this biofuel will be economically competitive with its fossil counterparts in the near future. Technical uncertainty is merely one source of doubt, complemented by developments in competing renewable energy technologies, increasing levels of financial risk, and the drags inflicted by policy discontinuity. Even if we focus on purely technical arguments of biofuel production costs, conclusions are hard to reach (Table 14.4). In many cases, reported capital and manufacturing costs are based on assumptions of future technology, and thus, a number of technological breakthroughs are needed before projected economic performance of many of these reports is realized. Consequently, the simulated selling prices of ethanol are commonly underestimated. These studies have noticeably heightened the level of excitement in the field, but may risk disillusionment if projections are missed for too long. A more pragmatic, but obviously less popular (or easy) approach would be sensitizing the public to the fact that renewable energy technologies will be expensive in the short term, but their advantages in the long term may well surpass the transient sacrifices needed to establish them. Techno-economic studies will be needed to support this or other arguments and should be conducted when possible.

P1: SFK/UKS P2: SFK c14 BLBS075-Hood December 15, 2010

348

20:59

Trim: 244mm×172mm

Plant Biomass Conversion

Table 14.4. Estimates of minimum ethanol selling price according to various studies. Feedstock/process

MESP ($/gal)a

In 2009$b

Reference

Softwood (spruce), steam pretreatment, SSFc Softwood (spruce), steam pretreatment Corn stover, dilute acid, SSF Corn stover, AFEX,d SSF Corn stover, dilute acid Poplar, dilute acid Corn stover, dilute acid, SSF Wood, dilute acid, SSF

$2.16 $2.39 $1.34 $1.43 $2.43 $2.37 $3.30 $1.28

$2.54 $2.81 $1.48 $1.58 $2.43 $2.71 $3.30 $1.66

Wingren et al., 2003 Wingren et al., 2003 Eggeman and Elander, 2005 Eggeman and Elander, 2005 Aden and Foust, 2009 Hamelinck et al., 2005 Dutta et al., 2009 So and Brown, 1999

a Minimum

ethanol selling price. adjusted using the Consumer Price Index reported by the US Bureau of Labor Statistics. c Simultaneous saccharification and fermentation; if not indicated, separate vessels for these steps are assumed. d Ammonia fiber expansion. b Inflation

14.13 Acknowledgments This work was part of the DOE Joint BioEnergy Institute (http://www.jbei.org) supported by the U.S. Department of Energy, Office of Science, Office of Biological and Environmental Research, through contract DE-AC02-05CH11231 between Lawrence Berkeley National Laboratory and the U.S. Department of Energy. Additional funding from the U.S. Department of Energy, Office of Energy Efficiency and Renewable Energy, Commercialization and Deployment program; Statoil; Boeing; and General Motors is acknowledged. The authors declare no conflicting interests.

References Aden, A. & Foust, T. 2009. Technoeconomic analysis of the dilute sulfuric acid and enzymatic hydrolysis process for the conversion of corn stover to ethanol. Cellulose, 16, 535– 545. Aden, A., Ruth, M., Ibsen, K., Jechura, J., Neeves, K., Sheehan, J., Wallace, B., Montague, L., Slaton, A. & Lukas, J. 2002. Lignocellulosic Biomass to Ethanol Process Design and Economics Utilizing Co-Current Dilute Acid Prehydrolysis and Enzymatic Hydrolysis for Corn Stover. Technical Report: NREL/TP- 510-32438, 154. Almeida, J. R. M., Modig, T., Petersson, A., Hahn-Hagerdal, B., Liden, G. & Gorwa-Grauslund, M. F. 2007. Increased tolerance and conversion of inhibitors in lignocellulosic hydrolysates by Saccharomyces cerevisiae. Journal of Chemical Technology and Biotechnology, 82, 340–349. America’s Energy Future Panel on Alternative Liquid Transportation Fuels. 2009. Liquid Transportation Fuels from Coal and Biomass: Technological Status, Costs, and Environmental Impacts. Washington, DC: National Academies Press. Anonymous. 2009. Available: Www.Eia.Doe.Gov (accessed October 09, 2010). Atsumi, S., Hanai, T. & Liao, J. C. 2008. Non-fermentative pathways for synthesis of branchedchain higher alcohols as biofuels. Nature, 451, 86–89.

P1: SFK/UKS P2: SFK c14 BLBS075-Hood December 15, 2010

20:59

Trim: 244mm×172mm

14

Biofuel Economics

349

Balusu, R., Paduru, R., Kuravi, S., Seenayya, G. & Reddy, G. 2005. Optimization of critical medium components using response surface methodology for ethanol production from cellulosic biomass by Clostridium thermocellum SS19. Process Biochemistry, 40, 3025–3030. Becker, D., Blotkamp, P. & Emert, G. 1981. Pilot scale conversion of cellulose to ethanol. In: Klass, D. & Emert, G. (eds.) Fuels from Biomass and Waste. Ann Arbor, MI: Ann Arbor Science Publications. Blanch, H. & Wilke, C. 1983. Sugars and chemicals from cellulose. Reviews in Chemical Engineering, 1, 71–118. Bothast, R. & Schlicher, M. 2005. Biotechnological processes for conversion of corn into ethanol. Applied Microbiology and Biotechnology, 67, 19–25. Braeutingam, J. 2009. Available: Http://Www.Energy.Ca.Gov/2009 Energypolicy/Documents/ 2009-04-14-15 Workshop/Presentations/Day-1/08- (accessed October 09, 2010). Cooney, C., Wang, D., Wang, S., Gordon, J. & Jiminez, M. 1978. Simultaneous cellulose hydrolysis and ethanol production by a cellulolytic anaerobic bacterium. Biotechnology and Bioengineering Symposium, 8, 103–114. Dadi, A. P., Varanasi, S. & Schall, C. A. 2006. Enhancement of cellulose saccharification kinetics using an ionic liquid pretreatment step. Biotechnology and Bioengineering, 95, 904–910. Delgenes, J., Laplace, J., Moletta, R. & Navarro, J. 1996. Comparative study of separated fermentations and cofermentation processes to produce ethanol from hardwood derived hydrolysates. Biomass and Bioenergy, 11, 353–360. Demain, A., Newcomb, M. & Wu, J. 2005. Cellulase, clostridia, and ethanol. Microbiology and Molecular Biology Reviews, 69, 124–154. Demirbas, A. 2001. Relationships between lignin contents and heating values of biomass. Energy Conservation and Management, 42, 183–188. Dien, B., Nichols, N., O’Bryan, P. & Bothast, R. 2000. Development of new ethanologenic Escherichia coli strains for fermentation of lignocellulosic biomass. Applied Biochemistry and Biotechnology, 84–86, 181–196. Dunnett, A., Adjiman, C. & Shah, N. 2008. A spatially explicit whole-system model of the lignocellulosic bioethanol supply chain: An assessment of decentralised processing potential. Biotechnology for Biofuels, 1, 1–17. Dutta, A., Dowe, N., Ibsen, K. N., Schell, D. J. & Aden, A. 2010. An economic comparison of different fermentation configurations to convert corn stover to ethanol using Z. mobilis and Saccharomyces. Biotechnology Progress, 26, 64—72. Dwivedi, P., Alavalapati, J. R. R. & Lal, P. 2009. Cellulosic ethanol production in the United States: Conversion technologies, current production status, economics, and emerging developments. Energy for Sustainable Development, 13, 174–182. Eggeman, T. & Elander, R. 2005. Process and economic analysis of pretreatment technologies. Bioresource Technology, 96, 2019–2025. Energy Efficiency and Renewable Energy, Biomass Program. 2009. Biomass Feedstock Composition and Property Database [Online]. Available: http://www1.eere.energy.gov/biomass/ [Accessed 1 September 2009]. Enguidanos, M., Soria, A., Boyan, K. & Jensen, P. 2002. Techno-economic analysis of bioalcohol production in the EU: A short summary for decision-makers. European Comission Joint Research Centre. Report Number EUR 20280 EN. Ezeji, T. C., Qureshi, N. & Blaschek, H. P. 2007. Bioproduction of butanol from biomass: From genes to bioreactors. Current Opinion in Biotechnology, 18, 220–227. Falhout, P. & Nedwin, G. 2005. Novozymes looks at next steps for biomass ethanol. Industrial Bioprocessing, 27, 1–2.

P1: SFK/UKS P2: SFK c14 BLBS075-Hood December 15, 2010

350

20:59

Trim: 244mm×172mm

Plant Biomass Conversion

Fargione, J., Hill, J., Tilman, D., Polasky, S. & Hawthorne, P. 2008. Land clearing and the biofuel carbon debt. Science, 319, 1235–1238. Fischer, C. R., Klein-Marcuschamer, D. & Stephanopoulos, G. 2008. Selection and optimization of microbial hosts for biofuels production. Metabolic Engineering, 10, 295–304. Graham, R. L., Nelson, R., Sheehan, J. J. & Al, E. 2007. Current and potential US corn stover supplies. Agronomy Journal, 99, 1–11. Green, D. & Perry, R. 2007. Perry’s Chemical Engineers’ Handbook. 8th edn. New York: McGraw-Hill Professional. Gutierrez, T., Ingram, L. & Preston, J. 2006. Purification and characterization of a furfural reductase (FFR) from Escherichia coli strain LYO1—An enzyme important in the detoxification of furfural during ethanol production. Journal of Biotechnology, 121, 154–164. Hamelinck, C., Hooijdonk, G. & Faaij, A. 2005. Ethanol from lignocellulosic biomass: Techno-economic performance in short-, middle-and long-term. Biomass and Bioenergy, 28, 384–410. Hammel-Smith, C., Fang, J., Powders, M. & Aabakken, J. 2002. Issues Associated with the Use of Higher Ethanol Blends (E17–E24). National Renewable Energy Laboratory. Report Number NREL/TP-510-32206. Henrissat, B., Driguez, H., Viet, C. & Sch¨ulein, M. 1985. Synergism of cellulases from Trichoderma reesei in the degradation of cellulose. Nature Biotechnology, 3, 722–726. Henstra, A. M., Sipma, J., Rinzema, A. & Stams, A. J. M. 2007. Microbiology of synthesis gas fermentation for biofuel production. Current Opinion in Biotechnology, 18, 200–206. Hess, J. R., Kenney, K. L., Wright, C. T., Perlack, R. & Turhollow, A. 2009. Corn stover availability for biomass conversion: Situation analysis. Cellulose, 16, 599–619. Himmel, M., Ruth, M. & Wyman, C. 1999. Cellulase for commodity products from cellulosic biomass. Current Opinion in Biotechnology, 10, 358–364. Huang, H., Ramaswamy, S., Al-Dajani, W., Tschirner, U. & Cairncross, R. 2009. Effect of biomass species and plant size on cellulosic ethanol: A comparative process and economic analysis. Biomass and Bioenergy, 33, 234–246. Jeffries, T. W. 2006. Engineering yeasts for xylose metabolism. Current Opinion in Biotechnology, 17, 320–326. Jeoh, T., Michener, W., Himmel, M. & Decker, S. 2008. Implications of cellobiohydrolase glycosylation for use in biomass conversion. Biotechnology for Biofuels, 1, 10–22.Kaliyan, N. & Morey, R. V. 2009. Densification characteristics of corn stover and switchgrass. Transactions of the ASABE, 52, 907–920. Kalogo, Y., Habibi, S., Maclean, H. L. & Joshi, S. V. 2007. Environmental implications of municipal solid waste-derived ethanol. Environmental Science and Technology, 41, 35–41. Kalscheuer, R., Stoelting, T. & Steinbuechel, A. 2006. Microdiesel: Escherichia coli engineered for fuel production. Microbiology, 152, 2529–2536. Keasling, J. D. & Chou, H. 2008. Metabolic engineering delivers next-generation biofuels. Nature Biotechnology, 26, 298–299. Khanna, M., Dhungana, B. & Clifton-Brown, J. 2008. Costs for producing miscanthus and switchgrass for bioenergy in Illinois. Biomass and Bioenergy, 32, 482–493. Kim, D., Baik, K. S., Park, S. C., Kim, S.-J., Shin, T.-S., Jung, S.-J., Oh, M.-J. & Seong, C. N. 2009. Cellulase production from Pseudoalteromonas sp. NO3 isolated from the sea squirt Halocynthia rorentzi. Journal Of Industrial Microbiology & Biotechnology, 36, 1375– 1382. Kim, S. & Dale, B. 2004. Global potential bioethanol production from wasted crops and crop residues. Biomass and Bioenergy, 26, 361–375.

P1: SFK/UKS P2: SFK c14 BLBS075-Hood December 15, 2010

20:59

Trim: 244mm×172mm

14

Biofuel Economics

351

Koplow, D. 2007. Biofuels—At What cost? Government support for ethanol and biodiesel in the United States. Geneva, Switzerland, Global Subsidies Initiative of the International Institute for Sustainable Development. www.globalsubsidies.org/IMG/pdf/biofuels subsidies us.pdf (Accessed 20 May 2008). Westhoff, P., Meyer, S. & Thompson, W. 2007. Economic Impacts of Not Extending Biofuels Subsidies. Journal of Agrobiotechnology Management and Economics, 10, 94–193. Kubicek, C., Mikus, M., Schuster, A. & Schmoll, M. 2009. Metabolic engineering strategies for the improvement of cellulase production by Hypocrea jecorina. Biotechnology for Biofuels, 2, 1–14. Kumar, A., Cameron, J. & Flynn, P. 2003. Biomass power cost and optimum plant size in western Canada. Biomass and Bioenergy, 24, 445–464. Kumar, A., Cameron, J. & Flynn, P. 2005. Pipeline transport and simultaneous saccharification of corn stover. Bioresource Technology, 96, 819–829. Kumar, A. & Sokhansanj, S. 2007. Switchgrass (Panicum vigratum, L.) delivery to a biorefinery using integrated biomass supply analysis and logistics (IBSAL) model. Bioresource Technology, 98, 1033–1044. Lang, J. 1978. Preparation of liquid fuel and nutrients from municipal waste water. US Patent 4093516. Lau, M. W. & Dale, B. E. 2009. Cellulosic ethanol production from AFEX-treated corn stover using Saccharomyces cerevisiae 424A(LNH-ST). Proceedings of the National Academy of Sciences, 106, 1368–1373. Li, A., Antizar-Ladislao, B. & Khraisheh, M. 2007. Bioconversion of municipal solid waste to glucose for bio-ethanol production. Bioprocess And Biosystems Engineering, 30, 189– 196. Li, C., Knierim, B., Manisseri, C., Arora, R., Scheller, H. V., Auer, M., Vogel, K. P., Simmons, B. A. & Singh, S. 2009. Comparison of dilute acid and ionic liquid pretreatment of switchgrass: Biomass recalcitrance, delignification and enzymatic saccharification. Bioresource Technology, 101, 4900–4906. Liang, Y., Yesuf, J., Schmitt, S. & Bender, K. 2009. Study of cellulases from a newly isolated thermophilic and cellulolytic Brevibacillus sp. strain JXL. Journal of Industrial Microbiology and Biotechnology, 36, 961–970. Lipnizki, F., Field, R. & Ten, P. 1999. Pervaporation-based hybrid process: A review of process design, applications and economics. Journal of Membrane Science, 153, 183–210. Lora, J. & Glasser, W. 2002. Recent industrial applications of lignin: A sustainable alternative to nonrenewable materials. Journal of Polymers and the Environment, 10, 39–48. Lynd, L. R., Laser, M. S., Brandsby, D., Dale, B. E., Davison, B., Hamilton, R., Himmel, M., Keller, M., McMillan, J. D., Sheehan, J. & Wyman, C. E. 2008. How biotech can transform biofuels. Nature Biotechnology, 26, 169–172. Macedo, I. C., Seabra, J. E. A. & Silva, J. E. A. R. 2008. Green house gases emissions in the production and use of ethanol from sugarcane in Brazil: The 2005/2006 averages and a prediction for 2020. Biomass and Bioenergy, 32, 582–595. Madia, A., Townsend, J. & Demain, A. 1979. Microbial studies on cellulase production by Clostridium thermocellum. Abstracts of the Annual Meeting of the American Society for Microbiology, 21, 205. Mahmudi, H. & Flynn, P. C. 2006. Rail vs. truck transport of biomass. Applied Biochemistry and Biotechnology, 129, 88–103. Maiorella, B. 1984. Vacuum ethanol distillation technology. In: Douglas, L. (ed.) The Ethanol Separations Handbook. Golden: Solar Energy Research Institute.

P1: SFK/UKS P2: SFK c14 BLBS075-Hood December 15, 2010

352

20:59

Trim: 244mm×172mm

Plant Biomass Conversion

Maiorella, B., Blanch, H. & Wilke, C. 1984. Economic-evaluation of alternative ethanol fermentation processes. Biotechnology and Bioengineering, 26, 1003–1025. Mani, S., Tabil, L. & Sokhansanj, S. 2006. Effects of compressive force, particle size and moisture content on mechanical properties of biomass pellets from grasses. Biomass and Bioenergy, 30, 648–654. Mielenz, J. 2001. Ethanol production from biomass: Technology and commercialization status. Current Opinion in Microbiology, 4, 324–329. Mosberger, E. 2005. Chemical plant design and construction. In: Ullmann’s Encyclopedia of Industrial Chemistry. Weinheim: Wiley-VCH Verlag GmbH and Co. KGaA. Mosier, N. 2005. Features of promising technologies for pretreatment of lignocellulosic biomass. Bioresource Technology, 96, 673–686. Moxley, G., Zhu, Z. & Zhang, Y. H. 2008. Efficient sugar release by the cellulose solventbased lignocellulose fractionation technology and enzymatic cellulose hydrolysis. Journal of Agricultural and Food Chemistry, 56, 7885–7890. Nakao, S., Saitoh, F., Asakura, T., Toda, K. & Kimura, S. 1987. Continuous ethanol extraction by pervaporation from a membrane bioreactor. Journal of Membrane Science, 30, 273–287. O’Brien, D., Roth, L. & Mcaloon, A. 2000. Ethanol production by continuous fermentationpervaporation: A preliminary economic analysis. Journal of Membrane Science, 166, 105–111. O’Brien, D., Senske, G., Kurantz, M. & Craig, J. 2004. Ethanol recovery from corn fiber hydrolysate fermentations by pervaporation. Bioresource Technology, 92, 15–19. Overend, R., Chornet, E. & Gascoigne, J. 1987. Fractionation of lignocellulosics by steamaqueous pretreatments [and Discussion]. Philosophical Transactions of the Royal Society of London. Series A, Mathematical and Physical Sciences (1934–1990), 321, 523– 536. Pan, X., Gilkes, N., Kadla, J., Pye, K., Saka, S., Gregg, D., Ehara, K., Xie, D., Lam, D. & Saddler, J. 2006. Bioconversion of hybrid poplar to ethanol and co-products using an organosolv fractionation process: Optimization of process yields. Biotechnology and Bioengineering, 94, 851–861. Pandey, A., Soccol, C., Nigam, P. & Soccol, V. 2000. Biotechnological potential of agroindustrial residues. I: Sugarcane bagasse. Bioresource Technology, 74, 69–80. Perlack, R. & Turhollow, A. 2003. Feedstock cost analysis of corn stover residues for further processing. Energy, 28, 1395–1403. Perrin, R., Vogel, K., Schmer, M. & Mitchell, R. 2008. Farm-scale production cost of switchgrass for biomass. BioEnergy Research, 1, 91–97. Petrolia, D. R. 2008. The economics of harvesting and transporting corn stover for conversion to fuel ethanol: A case study for Minnesota. Biomass and Bioenergy, 32, 603–612. Roffler, S. R., Blanch, H. W. & Wilke, C. R. 1988. In situ extractive fermentation of acetone and butanol. Biotechnology and Bioengineering, 31, 135–143. Rozakis, S. & Sourie, J. 2005. Micro-economic modelling of biofuel system in France to determine tax exemption policy under uncertainty. Energ Policy, 33, 171–182. Schubert, C. 2006. Can biofuels finally take center stage? Nature Biotechnology, 24, 777–784. Scurlock, J. 2005. Bioenergy Feedstock Characteristics. Oak Ridge: Oak Ridge National Laboratory, Bioenergy Feedstock Development Programs. Searcy, E., Flynn, P., Ghafoori, E. & Kumar, A. 2007. The relative cost of biomass energy transport. Applied Biochemistry and Biotechnology, 137, 639–652. Sendich, E. N., Laser, M., Kim, S., Alizadeh, H., Laureano-Perez, L., Dale, B. & Lynd, L. 2008. Recent process improvements for the ammonia fiber expansion (AFEX) process

P1: SFK/UKS P2: SFK c14 BLBS075-Hood December 15, 2010

20:59

Trim: 244mm×172mm

14

Biofuel Economics

353

and resulting reductions in minimum ethanol selling price. Bioresource Technology, 99, 8429–8435. So, K. S. & Brown, R. C. 1999. Economic analysis of selected lignocellulose-toethanol conversion technologies. Applied Biochemistry and Biotechnology, 77–79, 633– 640. Sokhansanj, S. & Turhollow, A. 2002. Baseline cost for corn stover collection. Applied Engineering in Agriculture, 18, 525–530. Sokhansanj, S. & Turhollow, A. 2004. Biomass densification—Cubing operations and costs for corn stover. Applied Engineering in Agriculture, 20, 495–499. Sokhansanj, S., Turhollow, A., Cushman, J. & Cundiff, J. 2002. Engineering aspects of collecting corn stover for bioenergy. Biomass and Bioenergy, 23, 347–355. Sousa, L. D. C., Chundawat, S. P. S., Balan, V. & Dale, B. E. 2009. “ Cradle-to-grave” assessment of existing lignocellulose pretreatment technologies. Current Opinion in Biotechnology, 20, 339–347. South, C. & Lynd, L. 1994. Analysis of conversion of particulate biomass to ethanol in continuous solids retaining and cascade bioreactors. Applied Biochemistry and Biotechnology, 45, 467–481. Spatari, S., Tomkins, C. D. & Kammen, D. 2009. Biofuels. Gigaton Throwdown. San Francisco: Gigaton Throwdown Initiative. Timmons, M. D., Knutson, B. L., Nokes, S. E., Strobel, H. J. & Lynn, B. C. 2009. Analysis of composition and structure of Clostridium thermocellum membranes from wild-type and ethanol-adapted strains. Applied Microbiology and Biotechnology, 82, 929–939. Tucker, M., Mohagheghi, A., Grohmann, K. & Himmel, M. 1989. Ultra-thermostable cellulases from Acidothermus cellulolyticus: Comparison of temperature optima with previously reported cellulases. Nature Biotechnology, 7, 817–820. Van Vleet, J. H. & Jeffries, T. W. 2009. Yeast metabolic engineering for hemicellulosic ethanol production. Current Opinion in Biotechnology, 20, 300–306. Vane, L. M. 2006. A review of pervaporation for product recovery from biomass fermentation processes (vol. 80, pg. 603, 2005). Journal of Chemical Technology and Biotechnology, 81, 1328–1328. Wang, Z., Chen, M., Xu, Y., Li, S., Lu, W., Ping, S., Zhang, W. & Lin, M. 2008. An ethanol-tolerant recombinant Escherichia coli expressing Zymomonas mobilis pdc and adhB genes for enhanced ethanol production from xylose. Biotechnology Letters, 30, 657– 663. Westcott, P. 2009. Full Throttle U.S. Ethanol Expansion Faces Challenges. Available: http://www.Ers.Usda.Gov/ (accessed October 09, 2010). Wingren, A., Galbe, M. & Zacchi, G. 2003. Techno-economic evaluation of producing ethanol from softwood: Comparison of SSF and SHF and identification of bottlenecks. Biotechnology Progress, 19, 1109–1117. Wooley, R., Ruth, M., Glassner, D. & Sheehan, J. 1999. Process design and costing of bioethanol technology: A tool for determining the status and direction of research and development. Biotechnology Progress, 15, 794–803. Yang, B. & Wyman, C. E. 2008. Pretreatment: The key to unlocking low-cost cellulosic ethanol. Biofuels, Bioproducts and Biorefining, 2, 26–40. Yazdani, S. S. & Gonzalez, R. 2007. Anaerobic fermentation of glycerol: A path to economic viability for the biofuels industry. Current Opinion in Biotechnology, 18, 213–219. Yazdani, S. S. & Gonzalez, R. 2008. Engineering Escherichia coli for the efficient conversion of glycerol to ethanol and co-products. Metabolic Engineering, 10, 340–351.

P1: SFK/UKS P2: SFK c14 BLBS075-Hood December 15, 2010

354

20:59

Trim: 244mm×172mm

Plant Biomass Conversion

Zhang, Y.-H. P., Cui, J., Lynd, L. R. & Kuang, L. R. 2006. A transition from cellulose swelling to cellulose dissolution byo-phosphoric acid: Evidence from enzymatic hydrolysis and supramolecular structure. Biomacromolecules, 7, 644–648. Zhang, Y.-H. P., Ding, S.-Y., Mielenz, J. R., Cui, J.-B., Elander, R. T., Laser, M., Himmel, M. E., McMillan, J. R. & Lynd, L. R. 2007. Fractionating recalcitrant lignocellulose at modest reaction conditions. Biotechnology and Bioengineering, 97 (2), 214–223.

P1: OTE/PGN ind

P2: OTE/PGN

BLBS075-Hood

QC: OTE/PGN

December 22, 2010

T1: OTE 15:25

Trim: 244mm×172mm

Index

Abatement 314–315, 317, 319, 324, 347 Aerobic 158–160, 165, 174, 233, 237, 239, 291 AGNPS watershed model 135–138 Agroforestry 62–65, 67 Ammonia 202, 207–210, 212, 216, 218, 221–222, 224, 283–284, 334 Anaerobic 110, 158–160, 162, 173, 177–178, 233, 237, 239, 284 Annuals 7, 85, 92, 99, 132, 146, 148–149 Arabinose 69, 158–159, 163–164, 168, 171–173, 176–180, 201 ATP 160–162 Autohydrolysis 73, 201, 205, 209, 211–214, 216, 218–219, 221, 224 Barley 24, 88, 239, 262, 265, 276–277, 284 Biobased Products 6–8, 13, 14, 16, 17, 27, 53, 287–288, 304 Biobutanol 5, 13, 268–269, 291 Biochemical 4, 7, 11, 17, 18, 22, 23, 75, 110, 113, 157, 161, 268, 276, 333 Biodiesel 129, 131, 133, 255, 257–261, 263, 278, 287–288, 290, 320–323, 336, 345 Biogenic 6, 158, 276, 278–279, 288, 294, 296–299, 304 Biomass Crops 10, 72, 86–87, 100–102, 144, 147–148 Biomass Plantation 52, 55 Biorefinery 14–17, 23, 29, 40, 43, 44, 70, 72, 75, 77, 266, 270, 272, 279–281, 283, 303, 331, 332, 339–342, 344 Black Liquor 53, 280–282, 300–302 Brazil 70, 93, 150, 256–259, 261–262, 267, 269, 272, 287, 290, 296, 298, 323, 338 Business Model 16, 47, 341 Candida 164–165, 171, 173, 175 Canola 9, 10, 260, 276, 287, 322 Cap-and-Trade 315, 318, 324–325

Carbon Credit 15, 65, 313, 315–319, 338 Carbon Dioxide 8, 18, 46, 65, 72, 96, 277, 315–316 Carbon Emission(s) 72, 86, 99, 315–318, 324 Carbon Tax 313, 317, 325, 329 Carbon Trading 313, 317, 325 Catalytic 205, 230, 232, 234–235, 238, 271, 275, 278–279, 298 Cell wall 72, 187, 201, 227, 229–231, 236, 239 Cellulase(s) 10, 12, 39, 47, 103, 201–203, 211, 214, 216–220, 227–246, 283, 333, 335, 342 Cellulosic Biofuels 13, 21, 23, 39, 116, 199, 215, 221, 261, 321 Cereal Straw 7, 22, 210 Chicago Climate Exchange (CCX) 318–319 China 70, 86, 95, 256, 259–260, 263–264, 268, 286, 292–293, 296–297, 318 Clostridium 72, 165–166, 172, 178, 180, 182–183, 232, 235, 269, 291, 336 Corn Stover 6, 22, 24, 25, 27, 34, 35, 37, 39, 40, 46–49, 145–146, 158, 210–211, 213, 230, 233, 238, 242, 265, 283, 331, 339, 340, 343, 348 Corn cob(s) 210, 212–213, 264–265 Corn seed 242, 255–257, 261–264, 270, 276–277, 283, 289–290, 296–297 Cotton 4, 8, 9, 24, 40, 47, 135–137, 145, 277, 295 Cottonwood 55–61, 64, 72 Crop Residue 6, 7, 25, 27, 46, 85, 145–146, 276, 321, 323 Cultivar(s) 66, 92–93, 97–100, 103–104, 130 Demonstration Facility 264, 268, 291–292 Distillation 11, 41, 199, 210, 212, 222, 269, 282–283, 320, 331, 337, 338, 342, 343 355

P1: OTE/PGN ind

P2: OTE/PGN

BLBS075-Hood

356

QC: OTE/PGN

December 22, 2010

T1: OTE 15:25

Trim: 244mm×172mm

Index

E-85, E85 260, 262, 264, 271–272, 321 Electricity 61, 132, 133, 207, 221–223, 258, 260, 280, 284, 317, 329, 331, 338, 342–344 Enzymatic 8, 12, 73, 76, 113, 124, 162, 174, 177, 179, 182, 201–202, 207, 209–212, 214, 218–219, 227–231, 233, 239, 266–267, 277, 283–286, 301, 320, 332, 333, 338, 340–342 Enzymatic digestion 283–285 Enzyme(s) 5, 7, 8, 10–12,14, 20, 34, 39, 47, 68, 70, 110, 124, 161–163, 176–178, 180, 199–203, 211–212, 214–219, 221, 223–224, 227–246, 267, 277, 279, 283–285, 292, 294, 296, 332, 333, 335, 336, 341–344 Ethanol – Starch-based 256–257, 260 Ethylene 282, 291, 298–299 Europe(an) 7, 56, 86–90, 92, 97, 102, 255, 258–264, 268–269, 286–287, 289, 295, 311, 315, 317, 318, 325 Expression 12, 47, 203, 229, 231, 236, 238–240, 242, 244–245 Fermentation 5, 8–9, 11, 12, 36, 41, 68, 72, 75, 77, 94, 103, 113, 124, 129, 157–162, 164, 166, 168, 170–184, 199–201, 207–210, 213–218, 224, 254–256, 258, 264, 265–266, 268–269, 272, 277–281, 283–284, 286, 289–293, 295–296, 320, 330–331, 333, 335–338, 340–343, 348 Fermentation Organisms 5, 12, 203 Fibers 3, 8, 9, 34, 36, 39, 40, 42, 47, 67–68, 72, 74, 129, 130, 144, 153, 201–205, 207, 209, 211–215, 219, 224, 281–282, 285, 292, 294–295, 297, 302–303, 334 Forestry 3, 4, 6, 7, 11, 14, 16, 52, 59, 62–65, 78, 265, 267, 311, 316, 319 Fossil Fuel 3, 4, 51–53, 55, 96, 101, 103, 129, 130, 153, 159, 255, 258, 265, 272, 304, 311, 316, 319–320, 322–323, 325, 329, 345 Fractionation 8, 9, 73, 76–77, 199, 203, 212, 214, 242–244, 279–280, 282–284, 300–302, 304, 333 Fungi 24, 29, 96–97, 164–165, 167–170, 175, 229, 232–233, 237, 239, 289, 335 Fungus 58, 171, 229, 245 Gasoline 17, 34, 126, 157, 228, 257–258, 260, 262–264, 268, 270–271, 279, 286, 320–321, 323, 336, 338, 344, 346, 347 Genetic Improvement 57, 63, 180 Glucose, C6 8, 68–69, 75, 113, 124–125, 158–180, 182, 201, 207, 211–213, 216–218, 227, 232–233, 238–239, 240, 267, 276–277,

279, 283–284, 286, 289, 291, 293, 297, 333, 335, 336 Glycerin 287–288, 290–291, 322 Glycolysis 160–163 Graminae 85 Green Ash 56 Greenhouse Gas (GHG) Emissions 11, 17, 46, 53, 103, 126, 311, 313–316, 318–321, 324 Gulf Hypoxic Zone 138 Hardwood 53–62, 69–71, 73, 133, 204, 206, 208, 210, 215, 230, 234, 280–281, 285, 336, 332 Hemicellulose 5, 8, 11, 12, 22, 41, 53, 67–69, 73, 75, 78, 113, 124, 158, 173, 178, 199–202, 204, 207–216, 218–219, 227–231, 233, 236,256, 261, 276–286, 299, 302, 332 Hydraulic 210, 215, 218–219, 222–224 Hydrolysis 12, 41, 73–76, 124–125, 161, 177, 183, 200–205, 207, 209–221, 224, 227–233, 235, 237, 239–240, 242, 266–267, 269, 276, 278, 280, 282–286, 292, 302, 331–333, 335, 338, 338, 343 India 90, 92, 95, 256, 260, 263–264, 267, 282, 291, 298, 301 Inhibition 202–203, 211, 216–218, 232–233, 239, 242, 333, 335, 337, 343 Japan 86, 264–265, 267, 282, 286, 291–293, 295–297, 312, 317, 318 Klebsiella 168, 179, 182 Kraft pulp 68–69, 73–74, 208, 281, 300 Kyoto 311–312, 314, 317, 325 Lignin 5, 8, 9, 22, 47, 67–68, 70–75, 77–78, 110, 114, 124, 158, 199–204, 207–214, 218, 221, 223, 227–231, 233, 256, 261, 266–267, 272, 276–286, 290, 292, 294–295, 299–303, 332–334, 338, 342 Lignocellulose 8, 9, 24, 67, 72, 131, 158–159, 176–177, 183, 209, 227, 229, 276, 279, 283–284, 286, 297, 331, 335, 336 Lignocellulosic biofuels 342, 344 Lignosulfonates 280–281, 299–300 Manures 89, 151, 154, 159, 314 Methane 46, 103, 110, 290, 311, 314, 316, 319 Microbes 30, 33, 99, 157–159, 161, 175, 183, 237–240, 242, 244, 246, 267, 291, 335, 336, 337 Minnesota Model 23

P1: OTE/PGN ind

P2: OTE/PGN

BLBS075-Hood

QC: OTE/PGN

December 22, 2010

T1: OTE 15:25

Trim: 244mm×172mm

Index

Miscanthus 6, 7, 9, 10, 34, 65, 85–91, 101–103, 132, 148, 276, 331, 340 Miscanthus sinensis 102 Miscanthus x giganteus 86–88 Mississippi River Basin (MRB) 131 Mitigation 65, 150, 153, 311–315, 324–325 Molasses 169, 174, 256, 261–264, 266, 284, 289, 294 Municipal Solid Waste 6, 109–111, 113, 115–117, 119, 121, 123, 125, 145, 265, 283 Natural Regeneration 54 Natural Resource 46, 78, 131, 133, 135 Nutrient Loss 87 Organosolv 204, 206, 209–210, 212, 217–218, 221, 224, 280, 282–283, 285–286, 301–303, 334 Oxygen 8, 70, 72, 159–162, 175, 208, 271, 277–278, 320, 346 Palm 259–260 Panicum virgatum L 65, 97, 131, 132, 145–148 Perennial(s) 6, 7, 15, 24, 46, 59, 85–87, 90, 92, 97–100, 103–104,131–133, 138, 270 Petrochemical 8, 17, 18, 181–182, 268, 275–276, 279, 289–290, 294–295, 297–299, 303–304, 337 Phosphorus 33, 90, 153 Pichia stipitis 77, 169, 173, 175, 213 Pilot Facility 286, 288 Pine – Loblolly 61–63, 65–67, 72 Poplar 34, 55–56, 58, 60, 62–63, 65, 158, 203, 210, 212–214, 219, 231, 268, 341, 348 Poultry Litter 151–153 Pretreatment 5, 11, 12, 41, 43, 45, 73, 93, 103, 109, 113, 117–118, 177, 199–224, 228–231, 233, 236, 239, 245–246, 266–267, 276, 279, 281, 283–286, 302, 332–334, 338, 340–343, 348 Pretreatment – Chemical 205–206, 230 Pretreatment – Process 12, 200–201, 203–211, 213, 215, 217–218, 220–224, 283–284, 286, 333, 334, 340, 341 Pretreatment – Thermochemical 75, 110, 120, 209, 229–231, 239, 277–279, 331 Pretreatment Systems 11, 199–200, 205, 213, 279 Pulp 8, 15, 36, 39, 41, 51, 53–55, 62, 68–70, 72–74, 78, 113, 202–203, 206, 208, 212, 229–230, 266, 269, 272, 277–283, 285, 300

357

Pulping 39, 52, 68–69, 73–74, 78, 208, 230, 279–283, 285, 290, 292, 300 Pulpwood 54, 60–61, 64, 279 Pyruvate 160–162, 164, 166, 169–172, 174, 177–178 Rapeseed 259–260, 287–288 Regulation(s) 111, 120, 231 Renewable 3–7, 11, 13–15, 17, 18–20, 21, 23, 24, 51–53, 62, 67, 69, 71, 78, 85, 96, 101–102, 110, 125, 138, 157, 180, 183, 236, 255–257, 259–261, 263–264, 270, 272, 314, 316, 319–321, 323, 338, 344, 345, 347 Resin(s) 275, 280, 282, 288, 295–296, 298, 300–302 Rhizome(s) 87–89, 97, 102, 148 Saccharification 200, 228–231, 239–243, 336, 338, 341–344, 348 Severity 199, 201, 203–205, 210, 212–214, 218–220, 302 Short Rotation Woody Crops 7, 11, 52, 55–56, 59 Soil – Erosion 24, 46, 87, 99, 136, 138, 146, 148–149, 154 Soil – Nutrient Management 66, 133, 145–146, 149 Soil – Quality 23–25, 29, 32, 57, 99, 129, 143, 145, 149 Soil – Sediment Plume 136, 137 Solids – biomass 39, 86, 145, 200, 205–208, 210–211, 213–215, 217–220, 221–224, 237, 240–242, 244–245, 322 Solids – Waste 115, 118, 145, 151–153 Solubilization 200–201, 204, 207, 210–214, 216, 284, 340, 342 Sorghum 7, 34, 40, 85–86, 90, 92–97, 103, 131, 132, 256, 261–263, 276, 290 Sorghum bicolor 90, 131 Starch 4, 6, 11, 17, 18, 24, 67–69, 92, 158, 164–167, 169–171, 173, 176–178, 183, 219, 245–246, 255–257, 260, 262, 264, 266, 276–277, 279, 289–296, 337 Steam Explosion 199, 201, 204–205, 209–210, 212–214, 216, 218, 221–224, 283–284, 333, 334, 340 Stover 6, 22, 24–27, 29, 30–35, 37, 39, 40, 42, 45, 47, 49, 129, 158, 210–211, 213, 218, 230, 233, 238, 242–243, 246, 265, 283, 331, 339, 340, 343, 348 Sugar, C4 86–88, 92, 97, 130, 131, 149, 289, 291–292

P1: OTE/PGN ind

P2: OTE/PGN

BLBS075-Hood

358

QC: OTE/PGN

December 22, 2010

T1: OTE 15:25

Trim: 244mm×172mm

Index

Sugar, C5 71, 201, 241, 266, 277, 279, 284, 289, 292 Sugar, C6 201, 241, 277, 279, 289, 293 Sugarbeet 9, 13, 256, 269 Sugarcane 13, 93, 95–96, 158, 255–256, 258, 261–264, 266–267, 290, 320, 335, 343 Sustainability 4, 19, 46, 53, 64, 78, 129, 131, 133, 135, 137, 138, 139, 143, 145, 147, 149–151, 153 Sustainable 3, 5, 9, 16–19, 23, 25, 29, 42, 44, 46, 49, 51, 53, 98–99, 129, 131, 132, 138, 143–144, 146, 149–150, 152, 154, 185, 255–256, 263, 272, 276, 279, 290, 294, 303–304, 313, 320, 324, 329, 343, 345 Switchgrass 6, 7, 9–11, 13, 15, 34, 62, 65–67, 85–86, 90–91, 97–101, 103, 131–134, 137, 146–150, 154, 158, 242, 276, 296, 331, 339, 340 Sycamore 55–61, 64 Thailand 256, 259–260, 264, 288, 290 Tillage Management Systems 146 Transgenic 63, 172, 179, 231, 237–239, 241–244, 294, 296 Tree Species 52, 54–56, 60–65 United States 5–7, 9, 11, 12, 15–17, 20, 34, 36, 51–52, 54–56, 58, 60–63,66–67, 70, 72, 86,

89–90, 94–95, 100–101, 103, 109–114, 116, 120, 126, 130, 131, 145–148, 150, 256–257, 259–261, 267–269, 271, 281, 284–288, 290–293, 295, 302, 311–312, 315–321, 323, 346 Viscosity 215, 218–220, 236, 242, 288, 295 Water – Irrigation 42, 88, 130–133, 263, 339, 345 Water – Management 39, 111, 116, 284, 312 Water – Quality 64, 130, 132, 133, 135, 137, 138, 143, 148–149 Water Use Efficiency (WUE) 130, 149 Wheat 6, 7, 13, 24, 25, 27, 145, 149, 154, 158–159, 237, 256, 261–266, 268, 282–284, 301, 340 Willow 34, 55–56, 60, 65, 72 Woody Biomass 7, 52–54, 59, 62, 67–68, 70–75, 78, 147, 208, 210, 265, 267, 277, 331 Xylose 8, 69, 73, 75–76, 113, 124, 158–159, 163–165, 167–169, 171–180, 183, 201, 207, 211–212, 214, 216–217, 229, 241, 277, 279, 281, 283, 285–286, 292, 336 Yeast 8, 11, 12, 53, 124, 157, 159, 164–165, 167–170, 175–177, 183, 239, 241, 276–277, 279, 289, 335, 336, 338 Yeast fermentation 8, 11, 277, 289 Zymomonas 171–172, 174, 179–181, 240, 336

P1: OTE/PGN ind

P2: OTE/PGN

BLBS075-Hood

QC: OTE/PGN

December 22, 2010

T1: OTE 15:25

Trim: 244mm×172mm