Geosynthetics and their applications
Edited by
Sanjay Kumar Shukla
.... ThomasTelford
'-I
Published by Thomas Telford Publishing, Thomas Telford Ltd , I Heron Quay, London E I4 4JD URL: http://www.thomastelford.com Distributors for Thomas Telford books are USA: ASCE Press, 180 I Alexander Bell Drive, Reston , VA 20 191-4400, USA Japan: Maruzen Co. Ltd, Book Department, 3- 10 Nihonbashi 2-chome, Chuo-ku, Tokyo 103 Australia: DA Books and Journals, 648 Whitehorse Road, Mitcham 3132, Victoria First published 2002
A catalogue record for this book is available from the British Library ISBN : 0 7277 3117 3
© Authors
and Thomas Telford Limited, 2002
All rights, including translation , reserved. Except as permitted by the Copyright, Designs and Patents Act 1988, no part of this publication may be reproduced , sto red in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying or otherwise, without the prior written permission of the Publishing Director, Thomas Telford Publishing, Thomas Telford Ltd, I Heron Quay, London E I4 4JD. This book is published o n the understanding that the author is solely responsible for the statements made and opinions expressed in it and that its publication does not necessarily imply that such statements and/or opin ions are or reflect the views or opinions of the publishers. While every effort has been made to ensure that the statements made and the opinions expressed in this publication provide a safe and accurate guide, no liability or responsibility can be accepted in this respect by the authors or publisher. Typeset by Academic + Technical, Bristol Printed and bound in Great Britain by MPG Books, Bodmin, Cornwa ll
Biographies
Dr Sanjay Kumar Shukla is currently on the faculty of the Department of Civil Engineering, Harcourt Butler Technological Institute, Kanpur, India. Dr Shukla received his BSc Engineering (Civil Engineering) degree in 1988 from the Bihar Institute of Technology, Sindri, India, and his MTech (Civil Engineering) and PhD (Civil Engineering) degrees in 1992 and 1995, respectively, from the Indian Institute of Technology, Kanpur, India. Previously, Dr Shukla served on the faculty of North Eastern Regional Institute of Science and Technology, Ita nagar, India, and the Bihar Institute of Technology, Sindri, India. He also worked in the Department of Civil Engineering, Indian Institute of Technology, Kanpur, India, as a Senior Project Associate. He is the author of several papers published in many reputed journals and presented in conferences and symposia. Dr Shukla is the recipient of the Indra Joshi Best Paper Award for 1995 from the Indian Geotechnical Society for the paper entitled ' Effect of prestressing on settlement of geosynthetic-reinforced foundation soil'. He is also an active geotechnical consultant and has been giving expert advice to many organizations, especially in the areas of ground improvement, geosynthetic applications, slope stability and landslide control, and the design and construction of pavements, retaining structures and foundations. In addition to these areas, Dr Shukla's areas of interest include environmental geotechnique, soil- structure interaction, foundation modelling, mineralogical analysis and engineering geology. In the past, he has organized many short-term courses for practising engineers. Dr Shukla is a life member of the Institution of Engineers (India), the Indian Geotechnical Society, the Indian Society for Rock Mechanics and Tunnelling Technology, the Indian Society for Technical Education, and the Coal Ash Institute of India. Dr M-Lurdes Lopes (born in Portugal in 1955), civil engineer from the Faculty of Engineering of Oporto University since 1977, carried out her graduate work at the New University of Lisbon (MSc in 1986) and at Oporto University (PhD in 1992). She joined the Polytechnic School of Oporto in 1982 and the Faculty of Engineering of Oporto University in 1986, where she is currently responsible for the subjects entitled 'Geosynthetics in Civil Engineering' and 'Earth Works' for the civil engineering course and for the laboratory of geosynthetics. She has been teaching geosynthetics in five different Portuguese Masters courses and has been responsible for short courses on geosynthetics in the Oporto University. M-Lurdes Lopes' research interests have been geosynthetics (mainly as environmental materials) and environmental geotechnics. She has been involved in nwnerous research projects, often as the principal investigator. To date, she has 65 scientific publications (12 publications for students and 40 consultancy reports) to her name. She is a member
iv
Geosynthetics and their applications
of the Council of the IGS (International Geosynthetics Society), and the representative from Portugal at the CEN/TCI89 (European Committee for Standardization - Geotextiles and Geotextiles Related Products), the EAC (European Activities Committee) of the IGS and TC5 (Environmental Geotechnics) of the ISSMGE. Dr Braja M. Das received his BSc degree in civil engineering from Utkal University, India, and his PhD in the area of geotechnical engineering from the University of Wisconsin, Madison, in 1972. He has more than 25 years of teaching experience in the United States. At present, he is a professor of civil engineering, and the Dean of the College of Engineering and Computer Science at California State University, Sacramento, USA. He has written more than 200 technical journal and proceeding papers, and 12 text and reference books. His areas of interest are shallow foundations, earth anchors, soil stabilization, and geosynthetics. Dr Ennio M. Palmeira is Associate Professor at the University of Brasilia, Brazil. He received his BSc and MSc in civil engineering in 1977 and 1981 from the Federal University of Rio de Janeiro . He holds a PhD degree in civil engineering from Oxford University (1987). Both his master and doctoral degrees dealt with soil reinforcement and the use of geosynthetics in geotechnical engineering. His research interests include soil reinforcement and geosynthetic applications in geotechnical and environmental engineering, and he has written over 100 publications in this field. In 1996 he was awarded an IGS Award, by the International Geosynthetics Society, for his contributions to the study and applications of geosynthetics in South America. Professor Palmeira has also acted as a consultant in geotechnical engineering. He is a council member of the IGS and a member of the Brazilian Society of Soil Mechanics, the British Geotechnical Society, and the International Society of Soil Mechanics and Geotechnical Engineering. Dr Philippe L. Bourdeau studied civil engineering at the Swiss Federal Institute of Technology in Lausanne (EPFL) . He graduated in 1976 as a civil engineer, in 1978 he received a postgraduate certificate in geotechnical engineering and in 1986 a doctorate in engineering sciences from EPFL. From 1976 to 1979, he gained experience in France as a structural and foundation engineer. He also served for one year as the City Engineer of Meknes (Morocco). In 1986, he was appointed Senior Research Associate in the Soil Mechanics Laboratory of EPFL. He joined the Civil Engineering Faculty of Purdue University in 1988 while also being appointed as a lecturer at EPFL. He is currently an Associate Professor of Civil Engineering at Purdue. Dr Bourdeau's research and teaching activities are broad but can be divided into three main subject areas: probabilistic modelling of geotechnical and geoenvironmental systems, mechanics of particulate media, and the use of geosynthetics. He has written or co-authored over 60 scientific publications. He currently serves on the Board of Editors for the journal Geotextiles & Geomembranes, on the American Society of Civil Engineers, GeoInstitute Committee on Risk Assessment and Management, and on the International Society of Soil Mechanics and Geotechnical Engineering Committee TC32 on Risk Assessment and Management. Dr Alaa K. Ashmawy is Assistant Professor in the Department of Civil and Environmental Engineering at the University of South Florida. He holds a BSCE degree (Honours) from Alexandria University and
Biographies
v
MSCE and PhD degrees from Purdue University. Prior to his current appointment, Dr Ashmawy worked as a researcher at the Georgia Institute of Technology. He is the recipient of the 1997 Research and Creative Scholarshjp Award and the 2000 Outstanding Undergraduate Teaching Award at his current institution. His teaching and research are in the areas of geosynthetics, experimental geomechanics, and GIS applications. He has published more than 20 refereed papers in his field. Dr Ashmawy is also a registered professional engineer in the State of Florida. Dr Steve W. Perkins has served as an associate professor of civil engineering at Montana State University in Bozeman, Montana, for the past ten years, where he teaches courses on geotechillcal engineering, geosynthetics and roadway engineering. Prior to this, Dr Perkins practised as a consultant for several geotechnical firms in Colorado and California. Dr Perkins has conducted research and published articles and reports on geosynthetics for the reinforcement of roadways for several US state transportation agencies, the US Federal Highway Admirustration and several geosynthetic manufacturers. Dr Perkins has conducted collaborative research on geosynthetic reinforcement offlexible pavements with the Norwegian Foundation for Techrucal and Industrial Research at the Norwegian University of Science and Technology. Dr Perkins has participated in the development of state-of-the-practice and state-of-the-art documents that have been used by the Geosynthetics Materials Association, NCHRP and AASHTO committees. Dr Perkins earned his BSCE from Virginia Tech., and his MSCE and PhD from the University of Colorado. He is a registered professional engineer in Montana and is a member of the American Society of Testing and Materials, American Society of Civil Engineers, International Geosynthetics Society, North American Geosynthetics Society, and the Transportation Research Board. Mr Ryan R. Berg is a professional engineer and a geotechnical engineering consultant specializing in geosynthetic applications, project peer review, failure investigations, and design and specification of earth structures. He has over 20 years of experience in designing, specifying and construction with geosynthetics. He has written numerous technical papers on designing with, and testing of, geosynthetics, and is the author or co-author of the Geosynthetic Engineering textbook; NCM A SR W (Segmental Retaining Wall) Design Manual, first edition; and the US DOT Federal Highway Administrations' Geosynthetic Design and Construction Guidelines; Guidelines for Design, Specification, & Contracting of Geosynthetic Mechanically Stabilized Earth Slopes on Firm Foundations; and Mechanically Stabilized Earth Walls and R einforced Soil Slopes. He is a Fellow of the American Society of Civil Engineers and a member of ASTM, the Transportation Research Board , the International Geosynthetics Society, the North American Geosynthetics Society, and the Minnesota Geotechnical Society. Mr Berg earned his Master's degree at Oregon State University and his Bachelor's degree at the University of Wisconsin at Platteville. Dr Barry R . Christopher is an independent geotechnical engineering consultant specializing in reinforced soil and other ground improvement technologies, geosynthetics application and design, and geotechnical/ geosynthetics testing and instrumentation. He has written numerous technical papers on these subjects, including five design manuals for the US Federal Highway Administration, a textbook on Geosynthetic Engineering and recently two National Cooperative Highway Research
vi
Geosynthetics and their applications
Program syntheses on pavement subsurface drainage systems and the maintenance of highway edgedrains. Dr Christopher has over 23 years of geotechnical engineering experience, much of which was gained from his previous work as a principal engineer for a major geotechnical consulting firm and as the technical director for a geosynthetics manufacturer. He has a BSCE from the University of North Carolina at Charlotte, a MSCE from Northwestern University, and a PhD from Purdue University. He is a registered professional engineer in six states. He has chaired several national and international professional committees and is currently active in the American Society of Testing and Materials, the American Society of Civil Engineers, the International Geosynthetics Society, the North American Geosynthetics Society, the International Standards Organization, and the Transportation Research Board. Dr Siew Ann ( Harry ) Tan has been teaching at the National University of Singapore (NUS), with active research interests in geosysnthetics, geotechnical, asphalt and highway materials, since 1985. He was the top engineering graduate from Auckland University in 1977, and received his MEng from NUS in 1982, and his MSc and PhD from Berkeley in 1981 and 1985, respectively. He has been a professional engineer in Singapore since 1994, and a member of the Institution of Engineers Singapore (rES) and the American Society of Civil Engineers (ASCE) since 1992. He has been involved in many consulting works for industry using geosynthetics, including the geotechnical design of the Semakau Offshore Waste facility and the reclamation works for the Second Causeway link to Malaysia at Tuas. He has published over 100 technical papers in the areas of geosynthetics, geotechnics and pavement materials testing and was co-recipient of the Katahira Award for the best technical paper in the 8th Road Engineering Association of Asia and Australia Conference on Road Engineering held in Taipei, April 1995. He currently serves on the Editorial Board of Geotextiles & Geomembranes and Geotechnical Engineering - Journal of the SE Asian Geo technical Society. Professor T . S. Ingold graduated with an honours degree in civil engineering from the University of London and, following several years with major consulting firms and contractors, a masters degree in soil mechanics from Imperial College. From 1974 to 1985 he was Chief Engineer of Ground Engineering Limited during which time his research in reinforced soil was awarded a PhD by the University of Surrey and he was appointed Visiting Professor of Civil Engineering at the Queens University, Belfast. In 1985, he set up in private practice as well as taking on posts as editor of Geotextiles & Geomembranes and , subsequently, Geosynthetics International. In 1990, he was appointed as a specialist consultant in erosion control to the United Nations. Professor Ingold has written three books and over one hundred technical publications on geotechnical topics, as well as sitting on many national and international committees. In 1996, Professor Ingold was appointed Honorary Professor of Geotechnical Engineering at the U niversi ty of Birmingham. Mr H elmut Zanzinger was born in 1961. He studied civil engineering, majoring in geotechnics at the Technical University of Karlsruhe, where he received the degree Diplom-Ingenieur in 1988. He joined the staff of LGA-Geotechnical Institute at Nuremberg, a German research, testing, and consulting organization . For 13 years he has been working there as an engineer on different assignments in geotechnical engineering (earthworks, landfills and foundations). During the past II years he has
Biographies
vii
mainly been involved in research, product development, design and quality management of geosynthetics. In particular, he conducted large-scale performance tests on geosynthetic- soil systems (geogrid reinforced walls, geopipes, geodrains, protection layers, etc.). He has also developed new testing methods and directed mechanical as well as hydraulic tests on geosynthetic products. He is the author of 60 scientific and technical papers. Dr Erwin Gartung is Chief Geotechnical Engineer with the LGA at Nuremberg, Germany. He is the chairman of the committee on 'Geotechnique of Landfill Structures' of the German Geotechnical Society DGGT and co-editor of the German Recommendations on Landfills. He was educated as a civil engineer at the Technical University of Brunswick, Germany, graduated from the University of California, Berkeley, as an MS in engineering, and obtained his Doctor-Engineer degree from Stuttgart University, Germany. Dr D. N. Singh has been a faculty member of civil engineering at the Indian Institute of Technology, Bombay, since 1994. He obtained his engineering education, from Bachelor to Doctorate, from the Indian Institute of Technology, Kanpur. He works in quite diversified areas of geotechnical engineering, such as environmental geotechnology, centrifuge modelling, solid waste characterization and utilization, and modelling of contaminant migration in geomaterials. To date, he has almost 60 research papers published in national and international refereed journals and conferences. Apart from teaching and research, Dr Singh has been very actively associated with some of the most prestigious business houses, as an in-house instructor and consultant. He is the co-editor of the Indian Geotechnical Journal and was the recipient of the young teachers' award, given by the All India Council of Technical Education, New Delhi. Dr Christian Duquennoi is the Head of the Drainage and Barrier Engineering Research Unit, Cemagref, Antony, France. He obtained his PhD in civil engineering. He is a member of the French Chapter of the International Geosynthetics Society. His main research topics include geomembrane performance and durability, containment of liquid and solid waste, modelling of heat and mass transfer in porous media and barrier materials. Dr Duquennoi is the author of several articles and conference communications. Dr Richard J. Bathurst is Professor of Civil Engineering at the Royal Military College (RMC) of Canada in Kingston, Ontario, where he has taught since 1980. He also holds a cross-appointment as Professor of Civil Engineering at Queen's University at Kingston and is an Adjunct Professor at the University of Waterloo . Dr Bathurst obtained a PhD in soil mechanics from Queen's University at Kingston in 1985. Prior to RMC, Dr Bathurst worked for Golder Associates from 1978 to 1980 as a geotechnical engineer and was employed on a variety of large civil engineering projects in Canada and overseas. In 2001 he was elected as a Fellow of the Engineering Institute of Canada. Dr Bathurst has been awarded numerous research grants and has written or co-written more than 130 papers in refereed journals, conference proceedings and research monographs. Dr Bathurst has been an invited keynote speaker at international conferences and was awarded the International Geosynthetics Society Gold Medal Award in 1994 and in 1998 for his contributions to the advancement of geosynthetic-reinforced retaining wall systems. He is also
viii
Geosynthetics and their applications
the recipient of the R. M. Quigley Award, given by the Canadian Geotechnical Society, for the best paper published in the Canadian Geotechnical Journal in 1996. He has acted as a consultant to many of the major players in the geosynthetics industry in North America. Dr Bathurst is President of the International Geosynthetics Society, a past-President of the North American Geosynthetics Society, past-Chairman of the Geosynthetics Division of the Canadian Geotechnical Society and has served on several other United States and international committees devoted to geosynthetics. Dr Bathurst is editor and co-author of the chapter on geosynthetics published as part of the Canadian Foundation Engineering Manual, coauthor of the First Edition of the National Concrete Masonry Association (NCMA) manual for segmental wall design and construction, and author of the NCMA Seismic Design Supplement for Segmental Retaining Walls. Dr Bathurst has served as co-editor of the technical journal Geosynthetics International since 1995. He also serves on the editorial boards of Geotechnical Fabrics Report, International Journal of Geomechanics, Ground Improvement, and Computers and Geotechnics. Dr K. Hatami is a research associate with the Department of Civil Engineering at the Royal Military College of Canada. He received his bachelor's degree in civil engineering and his master's degree in hydraulic structures in Tehran, Iran. Dr Hatami received his PhD degree in structural engineering from McMaster University at Hamilton, Ontario, and joined the Geotechnical Research Group at the Royal Military College of Canada as a post-doctoral fellow in 1997. His research interests include the analysis of conventional and reinforced soil retaining walls subjected to static and seismic loading, dynamic response analysis of concrete dams, and subjects in the area of geotechnical earthquake engineering. Dr Hatami has been involved in various civil engineering projects in connection with hydropower generation, including dams, underground powerhouse caverns and transmission towers, as well as slope stability analyses. Dr Hatami is the author and co-author of over 20 technical publications in refereed journals and conference proceedings in the areas of concrete dams and reinforced soil retaining wall systems. He is a member of the Canadian Geotechnical Society, the International Geosynthetics Society and the Canadian Association for Earthquake Engineering. Dr Marolo C. Alfaro is Assistant Professor of Civil Engineering at the University of Manitob~ , Canada. He received a BS in civil engineering from the University of Mindanao, Philippines, an MEng in geotechnical engineering from the Asian Institute of Technology in Thailand, and a PhD in civil engineering from Saga University, Japan. Dr Alfaro received post-doctoral fellowships from the Royal Military College of Canada and the University of Calgary, Canada. His research interests are in the following areas: soil/ground improvement techniques, geosynthetics in civil engineering, embankment failures on soft ground, and the use of computer tomography (CT) scan in experimental geotechnics. Dr Alfaro has written technical papers in refereed international journals. He is a co-author of the book, Improvement Techniques of Soft Ground in Subsiding and Lowland Environment (Balkema). He is a member of the Canadian Geotechnical Society, the Canadian Society for Civil Engineering, the International Society of Soil Mechanics and Geotechnical Engineering, and the International Geosynthetics Society.
Preface
In the present-day civil engineering practice, geosynthetics are being used extensively, in several areas, to provide the most efficient and costeffective solutions to a myriad of civil engineering problems throughout the world. Rational design methods, based on sound concepts and standardized test techniques for determining properties of technical interest of geosynthetics, are now available. This places the use of geosynthetics on a firm base which is no longer empirical. Geosynthetics have the potential of functioning for hundreds of years, if properly protected . The continued growth of geosynthetic applications at a rapid pace attests to the fact that geosynthetics have arrived as a viable and widely used construction material and they can now properly be added to the list of traditional materials, such as soil, brick, timber, steel, concrete, etc. While the subject of geosynthetics is a continually growing field in civil engineering, it is presently not taught in engineering and technical colleges as a separate course like a course on soil, concrete or steel. That is why most of the students, research workers and practising engineers need information on geosynthetics and their applications in a simple presentable form , with basic definitions and concepts. Areas of geosynthetic applications, with description of case histories and practical aspects, and recent developments, are required for quick reference in connection with their study and for solving specific field and research problems. Keeping these pressing needs in view as the key features , the present book is written in a single volume. The key features decided by the potential mass users of geosynthetics make this book different from other available good books on geosynthetics.
Acknowledgments
I would like to extend special thanks and recognition to each contributing author in this book. All the contributors have worked hard in making their contributions achieve a user-based common goal for this book. I am grateful to all of them. I am grateful to Dr C. v . S. Kameswara Rao, Professor in Civil Engineering, Harcourt Butler Technological Institute, Kanpur, India, for his help and encouragement during the preparation of the manuscript of this book. I am grateful to Dr Sarvesh Chandra and Dr P. K . Basudhar, Professors in Civil Engineering, Indian Institute of Technology, Kanpur, India, for their valuable suggestions during preparation of a few chapters of this book. I wish to thank Terram Limited , Gwent, UK; Netlon Limited, Blackburn, UK; Naue Fasertechnik GmbH & Co., Liibbecke, Germany; Huesker Synthetic GmbH & Co. , Gescher, Germany; Netlon India, Vadodara, India; Archana Structural Engineering (India) Pvt. Ltd, Bhopal, India, for providing useful information and materials required for the preparation of the book. I extend special thanks to Mr Graham James, Publishing Director, Miss Maria Stewart, Commissioning Editor, and the staff of Thomas Telford Limited for their cooperation and patience at all the stages of production of this book . I also wish to thank my wife, Sharmila, for her encouragement and support throughout the preparation of the manuscript. Thanks to my daughter, Sakshi, and my son, Saket, for their patience during my absence in connection with the book-related work. I welcome suggestions from the readers and the users of this book for improving its contents in a future edition. I will be grateful to them for their suggestions and views. Sanjay Kumar Shukla
I I
I I
I I
About the book
Geosynthetics and their applications is simply what the name implies, a book
to which students (all levels) and practising engineers (who are in search of novel approaches for solutions to civil engineering problems using geosynthetics) can refer. The simple and relatively concise presentation of topics, with basic concepts, is helpful, for all who have not credited or audited any geosynthetic-related course in their academic career, in understanding what geosynthetics and their applications are. The topics presented in this book are based on major field application areas for geosynthetics in civil engineering and, therefore, the readers and users of the book may find the information related to solutions of their specific problems very easily, which is one of the most important key features of this book and rarely found in other good books on geosynthetics. The description of several case histories and practical aspects are some additional key features of this book. The inclusion of recent developments along with references will be very useful, especially for research workers. Chapter I provides basic information on geosynthetics, including definitions and classification, historical development, functions and selections, raw materials and manufacturing processes, properties and testing, areas of applications, and available standards. Chapter 2 discusses soil- geosynthetic interaction, considering only the reinforcement function of geosynthetics, complemented by the description of the methods, namely direct shear and pullout tests, for evaluation of interaction properties along with detailed discussion on parameters affecting these properties. Chapter 3 provides the general guidelines for designing retaining walls using geotextile and geogrid as reinforcing materials along with a few example problems. Chapter 4 deals with the design of embankments on soft soils using geosynthetics as basal reinforcing materials as well as drains. Chapter 5 covers various aspects of shallow footings resting on geosynthetic-reinforced foundation soil, including reinforcing mechanisms, reinforcing patterns, and modes of failure along with model test results, methods of analyses for load-bearing capacity and settlement, and selected case histories. Chapter 6 addresses the application of geosynthetics in unpaved roads, that is roadway structures that are not capped by concrete slabs or asphaltic concrete wearing courses, with a detailed discussion on the interaction between soil masses and geosynthetics, and design approaches. Chapter 7 presents material related to the use of geosynthetics in paved roads. The functions of reinforcement, separation, drainage and filtration are discussed with an emphasis placed on the application of reinforcement. A recently completed 'recommended practice' is presented as an aid for the design of base reinforcement for paved roads.
xiv
Geosynthetics and their applications
Chapter 8 introduces the components of the conventional track structures and their functions , and describes properties, design and installation of geosynthetics for stabilization and drainage of railway tracks, along with a few case histories. Chapter 9 is concerned with a brief review of erosion processes, and it focuses on surface erosion caused by wind and rain along with erosion control methods that are of particular relevance to civil engineers. Chapter 10 deals with several aspects of slopes stabilized with geosynthetics as a major component, such as types and orientations of geosynthetics, modes of failure, review of methods of slope stability analysis, model tests, and stabilization methods in practice. Chapter 11 presents the details of the basal liners and covers essential barriers of solid waste landfills, with emphasis on the German practice of design and construction of landfills. Chapter 12 discusses various features related to the use of geosynthetics in earth dams along with a brief review of conventional earth dam construction practices. Some case studies are included to highlight the construction process and efficiency of installation of the geosynthetics. Chapter 13 describes the historical background , design concepts and principles of containment ponds, reservoirs and canals, along with several case histories. Chapter 14 summarizes selected published works related to the properties of cohesionless soil, geosynthetic reinforcement and facing components under cyclic loading. The chapter highlights the important features of current analytical and numerical methods for the seismic analysis and design of geosynthetic-reinforced soil walls and slopes, along with descriptions of the behaviour of reinforced soil walls and slopes based on physical modelling and the performance offield structures during earthquakes. Chapter 15 provides information on application-related general aspects, namely general guidelines, quality control and in-situ monitoring, cost analysis, and general problems, as well as a description of selected case histories.
Contents
1
Fundamentals of geosynthetics
1
S. K. Shukla
1.1. 1.2. 1.3. 1.4. 1.5. 1.6.
2
Introduction Definitions and classification Historical development Basic functions and selection Raw materials and manufacturing processes Properties and test methods 1.6.1. Physical properties 1.6.2. Mechanical properties 1.6.3. Hydraulic properties 1.6.4. Endurance and degradation properties 1.7. Application areas 1.8. Standards 1.9. Concluding remarks References
1 8 10 13 18 19 20 28 37 43 46 50 51
Soil-geosynthetic interaction
55
M. L. Lopes 2.1. Introduction 2.2. Granular soil behaviour 2.3. Soil- geosynthetic interaction mechanisms 2.4. Soil- geosynthetic interface resistance 2.5. Factors influencing soil- geosynthetic interaction 2.5.1. Soil particle size 2.5.2. Confinement stress 2.5.3. Soil density 2.5.4. Geosynthetic structure 2.6. Laboratory tests for the quantification of soil- geosynthetic interface resistance 2.6.1. Direct shear test 2.6.2. Pullout test 2.7. Concluding remarks References
3
I
55 56 57 58 62 62 66 67 68 71 71 72
78 78
Retaining walls
81
B. M. Das 3.1. Introduction 3.2. Design considerations 3.2.l. Stability 3.2.2. Lateral earth pressure
81 81 81 81
xvi
Geosynthetics and their applications
3.3.
3.4. 3.5.
4
3.2.3. Tie force Design procedure for retaining walls with geotextile reinforcement 3.3.1. General 3.3.2. Internal stability 3.3.3. External stability Design procedure for retaining walls with geogrid reinforcement Concluding remarks References
Embankments
85 85 85 85 87 92 92 93
95
E. M. Palmeira
4.1. 4.2.
4.3.
4.4.
5
6
Introduction Geosynthetics as a basal reinforcement in embankments 4.2.1. Reinforcement roles and aspects to be considered in the analysis 4.2.2. Design approaches for reinforced embankments 4.2.3. Choice of the reinforcement 4.2.4. Anchorage length of the reinforcement 4.2.5. Additional remarks on analysis and design Geosynthetics for drainage in embankments 4.3.l. Introduction 4.3.2. Geosynthetic drainage blanket at the base of the embankment 4.3.3. Geosynthetic vertical drains Concluding remarks References
Shallow foundations
95 95 95 98 108 109 109 114 114 114 115 118 119 123
S. K. Shukla 5.1. Introduction 5.2. Functions and mechanisms 5.3. Reinforcing patterns 5.4. Modes of failure 5.5. Model tests 5.5.1. Reinforced granular soil 5.5.2. Reinforced clay 5.5.3. Reinforced granular fill - soft foundation soil system 5.6. Load-bearing capacity analysis 5.6.1. Reinforced granular fill 5.6.2. Reinforced clay 5.6.3. Reinforced granular fill - soft foundation soil system 5.7. Settlement analysis 5.8. Field applications 5.9. Concluding remarks References
143 148 153 157 158
Unpaved roads
165
P. L. Bourdeau and A . K. Ashmawy 6.l. Introduction
165
123 123 127 127 128 128 132 l34 138 l39 143
Contents
6.2.
6.3.
7
Paved roads
s. W. 7.l. 7.2. 7.3 .
7.4. 7.5. 7.6. 7.7.
8
Unpaved road reinforcement 6.2.l. Interactions under monotonic loading 6.2.2. Effect of repeated loading 6.2.3. Design for reinforcement Concluding remarks References
Perkins, R . R . Berg and B. R . Christopher Introduction Distress features and their relationship to geosynthetics Geosynthetic functions 7.3.1. Reinforcement 7.3.2. Separation 7.3.3. Filtration 7.3.4. Drainage History and experimental evidence for base reinforcement Summary of critical design variables for base reinforcemen t Design solutions and approaches for base reinforcement Concluding remarks . References
Railway tracks
s.
A. ( Harry) Tan 8.l. Introduction 8.2. Track components and substructure 8.2.1. Subgrade 8.2.2. Subballast 8.2.3. Ballast 8.3. Functions of geosynthetics 8.3.1. Separation 8.3 .2. Filtration 8.3.3. Confinemen t/ rei nforcemen t 8.3.4. Drainage 8.4. Properties of geosynthetics 8.5. Design procedure 8.6. Installation of geosynthetics 8.7. Case histories in railway track stabilization 8.7.1. Experience from Canada and the USA 8.7.2. European experience 8.7.3. Indian experience 8.8. Geosynthetic drains for track drainage applications 8.8.1. Sources of water 8.8.2. Track drainage requirements 8.8.3. Side drains 8.8.4. Drainage of subgrade seepage 8.9. Concluding remarks References
xvii
166 166 169 171 180 181
185 185 185 187 187 189 192
193 193 195 195 198 199 203 203 203 203 204 205 207 208 208 209 209 210 212 213 214 214 215 216 216 216 217 217 219 220 221
xviii
Geosynthetics and their applications
9
Slopes T. S. 9.1. 9.2. 9.3. 9.4. 9.5. 9.6. 9.7. 9.8
erosion control
Ingold Introduction Interaction of rain and river erosion Mechanics of surface erosion Classification of erosion control systems Design approach Study of short-term yield factors Results from various field and laboratory tests Concluding remarks References
10 Slopes - stabilization
223 223 223 224 225 227 228 231 234 234
237
S. K. Shukla
10.1. 10.2. 10.3. 10.4.
Introduction Types and orientations of geosynthetics Modes of failure Stability analysis of reinforced slopes 10.4.1. Limit equilibrium method 10.4.2. Limit analysis method 10.4.3. Slip line method 10.4.4. Finite element method 10.5. Model tests 10.6. Stabilization methods in practice 10.6.l. Method suggested by Broms and Wong (1986) 10.6.2. Method suggested by Koerner (1984) and Koerner and Robins (1986) 10.6.3. Methods based on the construction of reinforced soil structures 10.7. Concluding remarks References
11 Landfills H. Zanzinger and E. Gartung 11.1. Introduction 11.2. Multibarrier concept 1l.3. Landfill categories 1l.4. Basal lining systems 1l.4.l. Functional layers 11.4.2. Concept of the composite liner 1l.4.3. Alternative liners 11.5. Components of the composite liner 11.5.l. Compacted clay liner 1l.5.2. Geomembrane 11.5.3. Protective layer for the geomembrane 11.6. Construction of liners 1l.6.l. Preparations 11.6.2. General aspects of installation 11.6.3. Placement of the geomembrane 1l.6.4. Quality assurance 11.7. Leachate collection and removal 1l.7.1. Drainage blanket and filters 11.7.2. Leachate collection pipes and access shafts
237 238 238 239 239 241 242 242 242 245 245 248 250 255 255
259 259 260 261 262 262 262 263 264 264 264 266 267 267 268 268 270 271 271 271
Contents
11. 7.3. Consequences for the basal seal 11.8. Cover system 11.8.1. General 11.8.2. Regulating soil and gas venting layer 11.8.3. Mineral sealing layer 11.8.4. Geosynthetic clay liners 11.8.5. Geomembranes 11.8.6. Dewatering of cover systems 11.8.7. Drainage geocomposites 11.9. Concluding remarks References
12 Earth dams D. N. 12.1. 12.2. 12.3.
Singh and S. K. Shukla Introduction Use of conventional materials Use of geosynthetics 12.3 .1. Geosynthetics as a barrier to fluid 12.3.2. Geosynthetics as a drainage channel 12.3.3. Geosynthetics as a filter 12.3.4. Geosynthetics as a protective layer 12.3.5. Geosynthetics as a reinforcement 12.3.6. Geosynthetics as an erosion control layer 12.4. River bed and bank protection 12.5. Design considerations 12.6. Concluding remarks References
13 Containment ponds, reservoirs and canals
xix
272 272 272 273 273 274 274 275 276 277 277
281 281 282 285 285 287 289 291 291 293 295 295 296 296
299
c.
Duquennoi 13.1. Introduction 13.2. Historical background 13.3. Design of geosynthetic systems 13.3.1. Subgrade preparation 13.3.2. Underiiner drainage and protection 13.3.3. Lining systems 13.3.4. Overiiner protection and cover 13.3.5. Singularities 13.4. Case studies 13.4.1. Containment ponds 13.4.2. Reservoirs 13.4.3. Canals 13.5. Concluding remarks 13.5.1. Acknowledgements References
299 299 301 301 302 302 304 305 306 306 309 315 322 322 323
14 Geosynthetic-reinforced soil walls and slopesseismic aspects
327
R. J. Bathurst, K. Hatami and M. C. Alfaro 14.1. Introduction 14.2. Material properties under dynamic loading 14.2.1. Soil
327 328 328
xx
Geosynthetics and their applications
14.3.
14.4.
14.5. 14.6.
14.7.
14.2.2. Geosynthetic reinforcement 14.2.3 . Interface properties Seismic analysis and design of walls and slopes 14.3.1. Pseudo-static methods 14.3.2. Pseudo-dynamic methods 14.3.3. Displacement calculations 14.3.4. Dynamic analysis using numerical techniques Physical testing of model walls and slopes 14.4.l. Gravity (lg) shaking and tilt table tests 14.4.2. Centrifuge shaking table tests Seismic buffers Observed performance of reinforced soil walls and slopes during earthquakes 14.6.1. North American experience (Northridge 1994 and Loma Prieta 1989) 14.6.2. Japanese experience (Hanshin 1995) Concluding remarks 14.7.1. Acknowledgements References
15 Geosynthetic applications selected case studies S. K. 15.1. 15.2. 15.3. 15.4. 15.5. 15.6.
379 379 380 381 383 383
general aspects and
Shukla Introduction General guidelines Quality control and in-situ monitoring Cost analysis General problems Selected case studies 15.6.1. Retaining walls and steep slopes 15.6.2. Landfills 15.6.3. Pipeline and drainage systems 15.6.4. Slopes - erosion cQntrol 15.6.5. Irrigation channels and reservoirs 15.6.6. Earth dams 15.6.7. Roads 15.6.8. Tunnels 15.7. Concluding remarks References
Index
331 336 341 341 355 357 362 373 373 378 379
393 393 393 399 400 405 406 406 409 411 412 413 413 414
416 416 417
421
1
Fundamentals of geosynthetics K.
S.
S HUKLA
Department of Civil Engin eering, Harcourt Butler Te chnological Institute, Kanpur, India
1.1. Introduction
In the past three decades, geosynthetics have been used successfully worldwide in several areas of civil engineering, and are now a well-accepted construction material. Their use offers excellent economic alternatives to the conventional solutions of many civil engineering problems. Therefore, students as well as practising engineers require an exposure to the fundamentals of geosynthetics as a construction material. This chapter fulfils this requirement by providing the basic information on geosynthetics, including definitions and classification, historical development, functions and ~elections, raw materials and manufacturing processes, properties and testing, areas of applications and available standards.
1.2. Definitions and classification
Geosynthetics is a generic term for all synthetic materials used in conjunction with soil, rock and/or any other civil-engineering-related material as an integral part of a man-made project, structure or system. It includes a broad range of synthetic products; the most common ones a re: • • • • •
geotextiles geogrids geonets · geomembranes geocomposites.
These products are almost exclusively polymeric, and those based on natural fibres (jute, cotton, wool, silk, etc.) are generally not included. They are available nowadays in numerous varieties in the market, under different trade names/designations for their use mainly in geotechnical, environmental, hydraulic and transportation engineering applications. Geotextiles are permeable, polymeric textile products in the form of flexible sheets (Fig. 1.1). Currently available geotextiles are classified into the following categories based on the manufacturing process: • woven geotextiles - they are made from yarns (made of one or several fibres) by conventional weaving process with regular textile structure • non-woven geotextiles - they are made from directionally or randomly oriented fibres into a loose web by bonding with partial melting, needle punching or chemical binding agents (glue, rubber, latex, cellulose derivative, etc.) • knitted geotextiles - they are produced by interlooping one or more yarns together • stitch-bonded geotexti les - they are formed by the stitching together of fibres or yarns .
2
Geosynthetics and their applications
- -(a)
Fig . 1.1. Typical geotextiles: (a) woven; (b) non-woven; and (c) knitted
(b)
Fundamentals of geosynthetics
,...
o
I
3
,...~
~
3
4
5 em
(e)
Fig. 1.1. continued
Oeogrid is a polymeric, mesh-like planar product formed by intersecting elements, called ribs, joined at the junctions (Fig. 1.2). The ribs can be linked by extrusion, bonding or interlacing, and the resulting geogrids are called extruded geogrid, bonded geogrid and woven geogrid , respectively. Extruded geogrids are classified into the following two categories based on the direction of stretching during their manufacture: • uniaxial geogrids - they are made by the longitudinal stretching of regularly punched polymer sheets and, therefore, possess a much higher tensile strength in the longitudinal direction than in the transverse direction
Fig. 1.2. Typical geogrids: extruded ((a) uniaxial; (b) biaxial); (c) bonded; and (d) woven
"'iiiiiII
_
(a)
_
(b)
4
Geosynthetics and their applications
-- -..
(e)
Fig . 1.2. continued
(d)
Fundamentals of geosynthetics
5
Fig. 1.3. The interlocking mechanism in geogridreinforced soil
• biaxial geogrids - they are made by both the longitudinal and the transverse stretchings of regularly punched polymer sheets and, therefore, possess equal tensile strength in both the longitudinal and the transverse directions. The key feature of geogrids is that the openings between the longitudinal and transverse ribs, called ilpertures, are large enough to create interlocking with the surrounding soil particles (Fig. 1.3). The shapes of the apertur~s are either elongated ellipses, near-squares with rounded corners, squares or rectangles. The dimensions of the apertures vary from about 2· 5 to 15 cm. The ribs of geogrids are often quite stiff compared to the fibres of geotextiles. Also, the junction strength is important in the case of geogrids because, through these junctions, loads are transmitted from one type of rib to the other when placed into the soil. Geonets are extruded polymer meshes and look like geogrids (Fig. 1.4). They are different from geogrids, not in the material or configuration, but in their functions (described later in this chapter) . Geonets have generally diamond-shaped apertures that are typically 12 mm long and 8 mm wide. The resulting angles are of the order of 70° and 110°. Geomembrane is a continuous membrane type barrier/liner composed of materials of low permeability to control fluid migration (Fig. 1.5). The materials may be asphaltic or polymeric or a combination thereof. The term barrier applies when the geomembrane is used inside an earth mass . The term liner is usually reserved for the cases where the geomembrane is used as an interface or a surface revetment.
.11"'11':
• •
• ~
J I....
I
111111 11 .. 1
.. • I
• 1 A..1 I
Fig. 1.4. Typical geonets
....
I I II II
I
JI
Jl JI
I
..... ..
.. .. & JIJI JIJI I I Ii J .. I
..
... ......
.. JlJlJ
.. ........... I
.. J
................. .x r
.. JI .......... .1
.I .I .I ... .I .. .I ...
6
Geosynthetics and their applications
Fig . 1.5. Typical geomembranes
The term geocomposites is applied to products that are manufactured in laminated or composite form from two or more geosynthetic materials (geotextiles, geogrids, geonets, geomembranes, etc.) that, in combination, perform specific functions more effectively than when used separately. There can be several combinations, such as geotextile- geonet, geotextile- geogrid , geotextile- geomembrane, geonet- geomembrane, geomembrane- clay, and geomembrane- geonet- geomembrane, which are used in different civil engineering applications. Figure 1.6 shows some typical geocomposites. There are many other terms for products used in the field of geosynthe tic manufacture and applications . Some of them are explained below.
Fig . 1.6. Typical geocomposites: (a) reinforced drainage separator; (b) geosynthetic clay liner; (c) drainage composites; and (d) surface erosion control mat
(a)
Fundamentals of geosynthetics
(c)
(b)
Fig . 1.6. continued
(d)
7
8
Geosynthetics and their applications
• Geofabric - a planar flat sheet of geotextile or geotextile-related products. • Geomat - a mat with very open structures made of coarse and rigid filaments with a tortuous shape, bonded at their junctions and look like a very coarse non-woven geotextile. • Geoweb - a very coarse woven geotextile (made of strips, typically 2 to 10 cm wide), i.e. a cellular geotextile with regular hexagonal or diamond-shaped cells, all linked together. • Geocell - a three-dimensional structure assembled from geogrids and special bodkins couplings on construction site to form triangular or square cells. • Geoproducts - a term meaning geosynthetics, or geotextile-related products made from natural fibres and metals. • Geospacer - a synthetic moulded structure, consisting of cuspidated or corrugated plates (eventually perforated).
1.3. Historical development
The oldest historical examples of the use of fabrics as an aid to road construction over soft ground include the use of woven reed mats by the ancient Romans. In a style remarkably similar to our present-day techniques, they used to lay the mats over marshy ground before overlaying with stone (Rankilor, 1981). The modern concept of soil reinforcement using membrane was proposed by Casagrande, who idealized the problem in the form of a weak soil reinforced by high-strength membranes laid horizontally in layers (Westergaard, 1938). Woven cotton fabrics were used as an early form of geotextile/ geomembrane in a series of road construction field tests started in 1926 by the South Carolina Highways Department (John, 1987). In the late 1950s, Terzaghi made use of filter fabrics (today geotextiles) as flexible forms. They were filled with a cement grout, thereby making closure between steel sheet piling and rock abutments at the Mission Dam (now Terzaghi Dam) in British Columbia, Canada. During this same project, Terzaghi used pond liners (today known as geomembranes) to keep an upstream clay seepage-control liner for desiccating (Terzaghi and Lacroix, 1964). It is believed that the first applications of polymerbased geotextiles were woven industrial fabrics used beneath concrete block revetments in the late 1950s (Barratt, 1966). A PVC monofilament woven geotextile was first used in 1958 at the base of the riprap under the sea dykes of Florida, USA. A woven polyvinyl fabric was first used , . instead of straw bags, in the early 1950s in Japan (Fukuoka, 1990). Agerschou (1961) described the use of woven materials to protect coastal structures from soil migration and eventual collapse, and this is most probably the earliest published work. During the early 1970s, the Japanese-developed filter fabrics (based upon their available weaving resources) were being used in, and exerting influences on, South East Asia designs for coastal works. By the mid1970s, the UK had started to produce geotextiles and, at this time, firms such as ICI in the UK, Rhone Poulence in France, Chemie Linz in Austria and DuPont in USA, started to promote the use of nonwoven geotextiles. In 1970, for the first time, a non-woven geotextile was used in an earth dam (Valcros Dam in France) (Giroud el aI. , 1977). Around 1971 three other areas of geotextile application first appeared, namely the first fin drains (Healy and Long, 1971), the first woven geotextile basal reinforcement beneath embankments (Holtz, 1975), and the first geotextile reinforced soil wall (Puig et at., 1977). The first composite geotextiles to appear were those used in fin drain
Fundamentals of geosynthetics
9
Fig . 1.7. A cellular geotextile net (after Simon et aI. , 1982)
systems during the period from 1969 to 1974. Composite geotextiles were also developed during the 1970s as types of band drain used to accelerate the consolidation of clay deposits by providing vertical drainage. Geonets were invented by F. B. Mercer in the UK in 1958. During the late 1970s, Netlon Ltd in the UK developed a more efficient means of utilizing the basic polymer raw material to yield a product with greater strength and elastic modulus in the form of polymer grids to be used in many soil reinforcement applications. The first samples of Tensar grid were made in the Blackburn laboratories of Netlon Ltd in July 1978. Soil confinement systems based on cellular geotextile nets were first developed and evaluated in France during 1980 (Simon et at. , 1982) (Fig. 1.7) and subsequently marketed under the trade name 'Armater', to be used in the control of surface erosion as well as in temporary road bases. In fully stretched form , the cellular geotextile net forms a honeycomb structure about 200 mm deep with either hexagonal or diamond-shaped apertures. Netlon developed a similar concept, but on a larger scale, with the introduction of the Tensar Geocell Mattress in 1982 (Mercer, 1982). The geocell mattress is assembled from Tensar geogrids and special bodkins couplings on the construction site to form triangular or square cells, 1 or O' 5 m deep and are used basically as a foundation layer beneath embankments, roads and buildings constructed over soft soils. Many popular books published or revised as late as 1969 did not provide reference to the use or design of geosynthetics in soil structures. However, despite the lack of general recognition of geosynthetic technology at this time, a few papers were published somewhat sporadically, the earliest known paper being that by Agerschou (1961). The number of publications suddenly increased from 1971. The first conference on geosynthetics was held in Paris in 1977. The first book on geosynthetics was written in 1980 (Koerner and Welsh, 1980). The international technology exchange has become active after the establishment of the International Geosynthetics Society in 1983. At present, there are two specialist international journals (Geotextiles and Geomembranes and Geosynthetics International), and several magazines and newsletters. In line with many other conferences, there are regularly scheduled national, regional and international conferences on geosynthetics. The field of geosynthetics has thus established itself in civil and environmental engineering. Various geosynthetic manufacturers have consistently been involved in pushing forward the frontiers of the geosynthetics technology. Geosynthetics were introduced to Indian engineers by the Central Board of Irrigation and Power (CBIP), New Delhi, in 1985 by organizing the first National Workshop on Geomembranes and Geotextiles. The first
10
Geosynthetics and their applications
state-of-the-art volume, Use of Geosynthetics in India: Experiences and Potential, was brought out by the CBIP in 1989 (Venkatappa Rao and Saxena, 1989). This was a compilation of the field trials in the country, which helped Indian engineers to gain confidence in the use of geosynthetics.
1.4. Basic functions and selection
Geosynthetics have numerous application areas in civil engineering. They always perform at least one of the following major functions when used in conjunction with soil, rock and/or any other civil-engineering-related material: • • • • •
separation reinforcement filtration drainage (or fluid transmission) fluid barrier.
If a geosynthetic prevents intermixing of adjacent soil layers with different properties during construction and the projected service period of the geosynthetic-reinforced soil structure, it is said to have a separation function. Figure 1.8 shows that the geosynthetic layer prevents the intermixing of soft soil with granular fill, thereby maintaining the structural integrity of the granular fill. A geosynthetic shows its reinforcement function by increasing the strength of a soil mass as a result of its inclusion, thus it maintains the stability of the soil mass. In this process the geosynthetc layer carries tensile loads (Fig. 1.9). A geosynthetic may function as a filter that allows for adequate flow of fluids across its plane while preventing the migration of soil particles along with fluid flow during the projected service period of application under consideration (Fig. l.l0). If a geosynthetic allows for adequate flow offluids within its plane from surrounding soil mass to various outlets during the projected service
Granular fill
Fig. 1.B. Separation function: (a) granular fillsoft soil system without geosynthetic; and (b) granular fill - soft soil system with geosynthetic
Granular fill
(b)
(a)
Potential failure surface
. . ' . ' .' Geosynthetic layer '.:...:..; .::,·;,..·/(under tension) . ~'."
Fig. 1.9. Reinforcement function
Firm stratum
Fundamentals of geosynthetics
11
Drainage s_tones
Fig. 1.10. Filtration function
Water flow direction
Fig. 1.11. Drainage function
period of application under consideration, it is said to have a drainage or fluid transmission function. Figure 1.11 shows that the geosynthetic layer adjacent to the retaining wall collects water from the backfill and conveys it to the weep hole made in the retain ing wall. A geosynthetic may also act like an almost impermeable membrane as far as the flow of fluids is concerned. Figure 1.12 shows that the geosynthetic layer, kept at the base of a pond, prevents the infiltration of liquid waste into the natural soil. In addition to the functions described above, a geosynthetic may also perform one or more than one of the following functions in some specific field applications. • Protection - where a geosynthetic is used as a localized stress reduction layer to prevent damage to a given surface or layer (e.g. geomembrane layer), it is said to perform the protection function. • Cushion - where a geosynthetic is used to control and eventually to damp dynamic mechanical actions, it is said to perform cushion function . This function has to be emphasized particularly for the applications in canal revetments, in shore protections, and in geosynthe tic strip layers as seismic base isolation of earth structures. • Absorption - it is the process of fluid being assimilated or incorporated into a geotextile. This function may be considered for two specific environmental aspects: water absorption in erosion control applications, and the recovery of floating oil from surface waters following ecological disasters.
Fig. 1.12. Fluid barrier function
12
Geosynthetics and their applications
Table 1.1. Selection of geosynthetics based on their functions Function(s) to be served by the geosynthetic
Geosynthetics that can be used
Separation
Primary Secondary
Reinforcement
Primary Secondary Primary Secondary Primary Secondary Primary Secondary
Geotexti les , geocomposites Geotextiles, geogrids, geonets, geomembranes, geocomposites Geotextiles , geogrids, geocomposites Geotextiles , geocomposites Geotextiles , geocomposites Geotextiles, geocomposites Geotextiles, geonets, geocomposites Geotextiles , geocomposites Geomembranes, geocomposites Geocomposites
Filtration Drainage Fluid barrier
• Interlayer - it is a function performed by a geosynthetic to improve shear resistance between two layers of geosynthetic products and/or earth materials. When installed, a geosynthetic may perform more than one of the listed functions simultaneously, but generally one of them will result in the lower factor of safety, thus it becomes the primary function . The use of a geosynthetic in a specific app lication needs classification of its functions as primary or secondary. Table 1.1 shows such a classification which is useful when selecting the appropriate type of geosynthetic to solve the problem in hand. The function concept is generally used in the design with the formulation of a factor of safety, FS, in the traditional manner as: S _ value of allowable (or test) property F - value of required (or design) property
(1.1)
Factors of safety must be greater than I; the actual magnitude depending upon the implication of failure, which is always site specific. The value of allowable property is obtained from a stimulated performance test (or an index test modified by site-specific reduction factors), whereas the required property is obtained from an appropriate design model. Such models are generally modifications of existing geotechnical or hydraulic models. The entire process, generally called 'design by function' is widespread in its use. However, as might be anticipated with a young technology, universally accepted values of minimum factors of safety have not yet been established, and conservation in this regard is still warranted (Koerner, 2000). It is to be noted that only geotextiles and geocomposites perform most of the functions and, hence, they are used in many applications. Geotextiles are porous to water flow, both normal to the manufactured plane and within the plane. The degree of porosity, which may vary widely, is used to determine the selection of specific geotextiles. Geotextiles can also be used as a fluid barrier on impregnation with materials such as bitumen. The geotextiles vary with the type of polymer used, the type of fibre and the fabric style. Geogrids are used mainly for reinforcement (separation may occasionally be a function , especially when soils with very large particle sizes are involved). The performance of the geogrid for reinforcement relies on its rigidity, or high tensile modulus, and on its open geometry, which accounts for its high capacity for interlocking with soil particles (Fig. 1.3).
Fundamentals of geosynthetics
13
It has been observed that for geotextiles to function properly as reinforcement, friction must develop between the soil and the reinforcement to prevent sliding, whereas for geogrids, it is the interlocking of the soil through the apertures of the geogrid that achieves an efficient anchoring effect. In this respect, geotextiles are frictional resistancedependent reinforcement, whereas geogrids are passive resistancedependent reinforcement. The laboratory studies have shown that geogrids are a superior form of reinforcement owing to the interlocking of the soil with the grid membrane. Geotextiles may be used to serve such functions as protection, cushion , interlayer or absorption . Protection and cushion functions may also be performed by geonets. Geonets, unlike geotextiles, are relatively stiff, net-like material with large open spaces (O·9- S·0cm) between structural ribs. It should be noted that geogrids are not geonets, which are used exclusively for their in-plane drainage capability. For a fair drainage function , geonets should not be laid in contact with soils or waste material but should be used as drainage cores with geotextile, geomembrane or other materials on their upper and lower surfaces, thus avoiding the soil particles from obstructing the drainage net channels. Geonets as drainage materials, in their flow capability, fall between thick needle-punched non-woven geotextiles and drainage composites. Geomembranes are used as a fluid barrier/liner only. The permeability of a typical thermoplastic or thermoset geomembrane is 10- 13 to 10- 15 m/ s. In this regard, we speak of it as being relatively impermeable. Geocomposites can be manufactured to perform a combination of the functions described above. For example, a geomembrane- geonetgeomembrane composite can be made where the interior net acts as a drain to the leak detection system . Similarly, a geotextile- geonet composite improves the separation, filtration and drainage . Geocomposites are generally, but certainly not always, completely polymeric. Other options include using fibreglass or steel for tensile reinforcement, sand in compression or as a filler , dried clay for subsequent expansion as a liner, or bitumen as a waterproofing agent. Geomembrane- clay composites are used as the liners, where the geomembrane decreases the leakage rate while the clay layer increases the breakthrough time. In addition, the clay layer reduces the leakage rate from any holes that might develop in the geomembrane, while the geomembrane will prevent cracks in the clay layer due to changes in moisture content. Geocomposites have been used in trapping and conveying leachate in landfills and in collecting gases from beneath geomembrane liners of various types. The selection of a geosynthetic for a particular application is governed by several other factors, such as specification, durability, availability, cost, etc.
1.5. Raw materials and manufacturing processes
The polymers generally used as raw materials for geosynthetics are polyester (PET), polypropylene (PP), polyethylene (PE) (very low density polyethylene (VLDPE), medium density polyethylene (MDPE), and high density polyethylene (HDPE)), chlorinated polyethylene (CPE), chlorosulfonated polyethylene (CSPE), polyamid (PA), polyvinyl chloride (PVC), etc. Table 1.2 provides the list of raw materials used for manufacturing different geosynthetics. There are a wide number of variables that affect the material properties of these polymers, including polymer density, melt flow rate, draw ratio, polymer additives, etc. The properties of geosynthetics are governed
14
Geosynthetics and their applications
Table 1.2. Polymers used as raw materials for manufacturing geosynthetics Geosynthetics
Raw materials
Geotextiles Geogrids Geonets Geomembranes
PP , PET, PA, PE HOPE, PET, PP MOPE, HOPE PE, PVC , CPE , CSPE
by these variables and their effects have been a subject of much investigation. Most of the geotextiles are manufactured from polypropylene or polyester. The primary reason for polypropylene usage in geotextile manufacturing is its low cost. For non-critical structures, it provides an excellent, cost-effective raw material. It exhibits a second advantage in that it has excellent chemical and pH range resistance. Additives and stabilizers (such as carbon black) must be added to give PP ultraviolet light resistance. As the critical nature of the structure increases, or the long-term anticipated loads go up, PP tends to lose its effectiveness. This is because of relatively poor creep deformation characteristics under long-term sustained load . Polyester is increasingly being used to manufacture reinforcing geosynthetics, such as geogrids, because of its high strength and resistance to creep. Chemical resistance of PET is generally excellent, with the exception of very high pH environments. It is inherently stable to ultraviolet light. The properties of some of the polymers mentioned above are compared in Table 1.3. Although most of the geosynthetics are made from synthetic polymers, a few specialist geosynthetics, especially geotextiles, may also incorporate either steel wire or natural biodegradable fibres such as jute, coir, paper, cotton, wool, silk, etc. Biodegradable geotextiles are usually limited to erosion control applications where natural vegetation will replace the geotextile's role as it degrades. Jute nets are marketed under various trade names, including geojute, soil-saver and anti-wash. They are usually in the form of a woven net with a mesh open size of about II by 18 mm,
Tabl e 1.3. A comparison of properties of polymers used in manufacturing the geosyntr,etics (Adapted from John , 1987) Properties
Polymers
Strength Modulus Strain at failure Creep Unit weight Cost Resistance to ultraviolet light Resistance Res istance Resistance Resistance
to to to to
Stabilized Unstabilized
alkalis fungus , vermin , insects fuel detergents
PP
PET
PA
PE
Low Low High High Low Low High Medium High Medium Low High
High High Medium Low High High High High Low Medium Medium High
Medium Medium Medium Medium Medium Medium Medium Medium High Medium Medium High
Low Low High High Low Low High Low High High Low High
Fundamentals of geosynthetics
15
a typical thickness of about 5 mm, and an open area of about 65 % . Vegetation can easily grow through the openings and use the fabric matrix as support. The jute, which is about 80% natural cellulose, should completely degrade in about two years. An additional advantage of these biodegradable products is that the decomposed jute improves the quality of the soil for vegetation growth. The manufacturing process of a geotextile includes two steps (Giroud and Carroll, 1983). The first step consists of making linear elements such as fibres and yarns. The second step consists of combining these linear elements to make a planar structure, usually called a fabric . The basic elements of a geotextile are its fibres . There are mainly four types of synthetic fibres : the filaments (produced by extruding melted polymer through dies or spinnerets, and subsequently drawing it longitudinally), staple fibres (obtained by cutting filaments to a short length, typically 2 to 10 cm), slit films (fiat tape-like fibres , typically 1 to 3 mm wide, produced by slitting an extruded plastic fi lm with blades and subsequently drawing it), and strands (a bundle of tape-like fibres that can be partially attached to each other) . During the drawing process, the molecules become oriented in the same direction, resulting in an increase of the modulus of the fibres. A yarn is made of one or more fibres . Several types of yarn are used to construct woven geotextiles: monofilament yarn (made from a single filament) , multifilament yarn (made from fine filaments aligned together), spun yarn (made from staple fibres interlaced or twisted together), slit film yarn (made from a single slit film fibre) , and fibrillated yarn (made from strands). It should be noted that synthetic fibres are very efficient load-carrying elements, with tensile strengths equivalent to prestressing steel in some cases (e.g. in the case of polyaramid fibres). As the name implies, woven geotextiles are obtained by conventional weaving processes, using a mechanical loom (Fig. 1.13). This weaving process gives these geotextiles their charcteristic appearance of two sets of parallel yarns interlaced at right angles to each other as shown in Fig. 1.14. The terms 'warp' and 'weft' are used to distinguish between the two different directions of yarn. The yarn running along the length of the loom and hence along the length of the geotextile roll is known as the warp. The yarn running along the transverse direction, across the width of both the loom and the geotextile roll , is known as the weft. The type of weave described is plain weave, of which there are
Reeds move up and down shedding the warp threads to make a tunnel for the shuttle Shuttle containing pirn of weft thread
Fig . 1.13. Main components of a weaving loom (after Rankilor, 1981)
Woven cloth wound onto beam
Warp threads
16
Geosynthetics and their applications
Weft threads
Fig. 1.14. A typica l woven geotextile having a plain weave
Warp threads
many variations, such as twill, satin and serge; however, plain weave is the one most commonly used in geotextiles. Resulting structures are typically 1 to 2 mm thick with a comparatively regular distribution of pore or mesh openings, which vary in dimension over a reasonably small size band. Kaswell (1963) gives an excellent review of weaving technology with clear illustrations of various fabric weaves. There are no rigid criteria relating polymer type to structure; however, tapes are most commonly polypropylene and monofilaments are most commonly po lyethelene, whereas the finer multifilaments or multifilament yarns are commonly polyester (Ingold and Miller, 1988). The numerous variations of weaving structure have a major influence on the physical, mechanical and hydraulic properties of the resulting geotextile. The highly anisotropic properties shown by woven geotextiles are also the influence of the weaving structure. Non-woven geotextiles are obtained by processes other than weaving. Continuous monofilaments are usually employed; these may, however, be cut into short staple fibres before processing. The processing involves continuous laying of the fibres or filaments on to a moving conveyor belt to form a loose web slightly wider than the finished product. This passes along the conveyor to be bonded by mechanical bonding (obtained by punching thousands of small barbed needles through the loose web), thermal bonding (obtained by partial melting of the fibres), or chemical bonding (obtained by fixing the fibres with a cementing medium, such as glue, latex, cellulose derivative or synthetic resin) , resulting in three different types: mechanically bonded non-woven geotextile (or needlepunched geotextiles), thermally bonded non-woven geotextile, and chemically bonded non-woven geotextile, repectively. These geotextiles are usually relatively thick, with a typical thickness in the region of 0·5 to Smm. Knitted geotextiles are manufactured using a knitting process which involves interlocking a series of loops of one or more yarns together to form a planar structure. There is a wide range of different types of knit used, one of which is illustrated in Fig. 1.15. These geotextiles are used in very limited quantity.
Fig. 1.15. A typical knitted geotextile
Fundamentals of geosynthetics
17
Extruded geogrids are manufactured by the method of processing sheet polymer in two or three stages. The first stage involves feeding a sheet of polymer, several millimetres thick, into a punching machine, which punches out holes on a regular grid pattern. Following this, the punched sheet is heated and stretched, or drawn , in the machine direction. This distends the holes to form an elongated grid opening. In addition to changing the initial geometry of the holes, the drawing process orients the randomly oriented long-chain polymer molecules in the direction of drawing. The degree of orientation will vary along the length of the grid; however, the overall effect is an enhancement of tensile strength and tensile stiffness. The process may be halted at this stage, in which case the end product is a uniaxially oriented geogrid. Alternatively, the uniaxially oriented grid may proceed to a third stage of processing to be warm drawn in the transverse direction, in which case a biaxially oriented geogrid is obtained. Although the temperatures used in the drawing process are above ambient, this is effectively a cold drawing process, as the temperatures are significantly below the melting point of the polymer. Netlon Ltd manufactures geogrids using their patented 'Tensar' manufacturing process (Fig. 1.16). Other types of geogrids are manufactured by weaving and knitting, as well as by bonding, the mutually perpendicular high-strength strips together at their crossover points ultrasonically or thermally. The most common manufacturing technique for geonets is to extrude the molten polymer through slits in counter-rotating dies, which forms a tight net of closely spaced ribs. This net is then opened up by forcing it over a tapered mandrel until it reaches its final configurations, when it is cooled, and rolled. The resulting geonet has intersecting sets of ribs at 60° to 75° apart, with the crossover points being integrally bonded to one another. A slight variation of the above technique is to add a foaming agent to the polymer mix and then process it as just described. The foaming agent is released and forms micrometer-sized gas-filled spheres within the rib cross-sections. Geonets formed in this manner can have very high ribs (resulting in increased flow capability) in comparison to the solid-formed ribs (Koerner, 1990). Most of the geomembranes are made in a plant using one of the following manufacturing processes: extrusion, spread-coating or calendering
Uniaxial grid Punched sheet
Polymer sheet
Biaxial grid / '
Fig. 1.16. 'Tensar' manufacturing process (courtesy of Net/on Limited, UK)
18
Geosynthetics and their applications
(Giroud and Frobel, 1983). The extrusion process is a method whereby a molten polymer is extruded into an unreinforced sheet. Immediately after extrusion, when the sheet is still warm, it can be laminated with a geotextile; the geomembrane thus produced is reinforced. The spread-coating process usually consists of coating a geotextile (woven, non-woven, knitted) by spreading a polymer or asphalt compound on it. The geomembranes thus produced are therefore reinforced . Non-reinforced geomembranes can be made by spreading a polymer on a sheet of paper which is removed and discarded at the end of the manufacturing process. Calendering is the most frequently used manufacturing process in which a heated polymeric compound is passed through a series of heated rollers (calender). Typical thicknesses of geomembranes range from 0·25 to 7·5 mm (l0 to 300 mils, 1 mil = 0·00 I in.) and they are produced in rolls approximately 1· 5 to 10m. Geocomposites can be manufactured from two or more of the geosynthetic types described above. A geocomposite can therefore combine the properties of the constituent members in order to meet the needs of a specific application. Some examples of geocomposites are: sheet drains, strip (wick) drains, fin drains and geosynthetic clay liners. A strip drain usually consists of plastic fluted or nub bed cores that are surrounded by a geotextile filter. A fin drain comprises a vertical water-conducting core, i.e. drainage net, sandwiched between outer layers of the geotextile. Geotextiles are commonly used in conjunction with geomembranes for puncture protection, drainage, and improved tensile strength. A geosynthetic clay liner is used in lieu of compacted soil for the low permeability soil component of the composite liner. It consists of a thin layer of sodium bentonite (mass per unit area ::::::5 kg/m2) which is either sandwiched between two geotextiles or mixed with an adhesive and attached to a geomembrane. Geosynthetic clay liners are manufactured in panels that measure 4- 5 m in width and 30- 60 m in length, and are placed on rolls for shipment to the jobsite. When a geosynthetic clay liner comes into contact with water, the bentonite swells in the pores, thereby forming a watertight sheet or offering protection to geomembranes (Venkatappa Rao , 1996). A geosynthetic clay liner has many advantages over a compacted soil liner, including the following (Snow et al., 1994): • simple installation and lower installed cost • low water consumption, dust generation, and vehicular traffic during construction • low susceptibility to desiccation cracking • self-healing capabilities if punctured • material quality maintained in a controlled environment • lower construction quality assurance costs • tensile strength developed by the geotextiles or geomembranes • reduced loss of valuable waste disposal facility. The combination of the geotextile (filtering action), geomembranes (waterproofing properties), and geonets (drainage and load distribution) offers a complete system of filter-drainage protection, which is very compact and easy to install.
1.6. Properties and test methods
This section deals with properties of geosynthetics and highlights the basic concepts of their measurement. Geosynthetics, being polymerbased products, are viscoelastic, which means that, under working conditions, their performance is dependent on the ambient temperature, the
Fundamentals of geosynthetics
19
level of stress, the duration of the applied stress, the rate at which the stress is applied, etc. The properties of geosynthetics should therefore be used to keep these factors in view.
1.6.1. Physical properties
The physical properties of geosynthetics that are of prime interest are specific gravity, mass per unit area, thickness and stiffness . There are some more physical properties, which are important in the case of geogrids and geonets only and these are: type of structure, junction type, aperture size and shape, rib dimensions, planar angles made by intersecting ribs and vertical angles made at the junction point. The physical properties are more dependent on temperature and humidity than those of soils and rocks. In order to achieve consistent results in the laboratory, good environmental control during the testing is therefore important. Typical values of specific gravity of commonly used polymeric materials are given in Table 1.4. It is to be noted that the specific gravity of some of the polymers is less than 1'0, which is a drawback when working with geosynthetics underwater, that is, some of them will float. The mass per unit area of a geosynthetic is usually given in units of gram per square metre (g/m2). Sometimes it is referred to as 'basis weight'. It can be a good indicator of cost and several other properties such as tensile strength, tear strength, puncture strength, etc., which are defined in Section 1.6.2. It is also necessary for quality control and thus it is the most useful basic property of geosynthetics. For commonly used geosynthetics, it varies in order of magnitude from typically 100 g/m 2 to 1000 kg/ m2. For 'Tensar' SR2 and SS2 grids, the mass per unit area was estimated to be 930 g/m 2 and 345 g/m 2, respectively. The thickness of geosynthetics, particularly of geotextiles, is measured as the distance between the upper and the lower surfaces of the material at a specified normal pressure (generally 2·0 kPa). The thickness of commonly used geosynthetics ranges from 10 to 300 mils. Most geomembranes used today are 20 mils (0·50mm). The stiffness or flexibility of a geosynthetic is related to its bending under its own weight and indicates the feasibility of providing a suitable working surface for installation. It can be measured by its capacity to form a cantilever beam without exceeding a certain amount of downward bending under its own weight. It should be noted that the workability of a geosynthetic (ability of the geosynthetic to support personnel in an uncovered state and construction equipments during initial stages of cover fill placement) also depends on other factors , such as water absorption and buoyancy. When placing a geotextile or geogrid on extremely soft soils, a high stiffness is desirable. Properties such as aperture size and shape, rib dimensions, etc., can be measured directly and are relatively easily determined.
Table 1.4. Specific gravity of polymeric materials Polymers
Specific gravity
Polypropylene Polyester (Terylene) Polyamide (Nylon) Polyethylene
0,91 1·22-1'38 1'05-1-14 0·91-0·95
20
Geosynthetics and their applications
4·5 4-0 E E
en<Jl OJ
3·5 3·0
c
""" :.c: IFig . 1.17. Var iation of thickness of ge otextiles with app lied pressure (after Shamsher, 1992)
''[':~ : : : : ~
Woven geotextile
0
1
2
3
4
5
6
I
I
I
7
8
9
: 10
Pressure : kPa
1.6.2. Mechanical properties
Mechanical properties are important in those applications where a geosynthetic is required to perform a structural role, or where it is required to survive installation damage and localized stresses. There are several mechanical properties, however, some of them are only important in the case of particular geosynthetics. Compressibility of a geosynthetic is measured by the decrease in its thickness at varying applied normal pressures. This mechanical property is very important for non-woven geotextiles because they are often used to convey liquid within the plane of their structure. Figure 1.17 shows changes in thickness under pressure for typical woven and needlepunched non-woven geotextiles. For most geotextiles, except needlepunched non-woven geotextiles, the compressibility is relatively very low. Due to specific geometry and irregular cross-sectional area, the tensile strength of geosynthetics cannot be expressed conveniently in terms of stress . It is, therefore, defined as the peak load that can be applied per unit width. Tensile strength is generally determined by 11 wide-width strip tensile test on a 200 mm wide strip, because by approximating plane strain conditions, this test more closely simulates the deformation experienced by a geosynthetic embedded in soil (Fig. 1.18). The test provides parameters such as peak strength, elongation and tensile modulus. The measured strength and the rupture strain are a function of many test variables, including'sample geometry, gripping method, strain rate, temperature, initial preload, conditioning, and the amount of any normal confinement applied to the geosynthetic. Figure 1.19 shows the influence of geotextile specimen width on tensile strength. To minimize the effects of these factors , the test sample should have a width-to-gauge length ratio (aspect ratio) of at least 2 and the test should be carried out at a standard temperature. Tensile strength is closely related to mass per unit area (Fig. 1.20). Other forms of tensile tests, such as the grab test and biaxial test, are shown schematically in Fig. 1.21.
Direction of stra in
Jaws/clamps
Fig. 1.18. Wide-width strip tensile test (note B = 200mm, L = 100mm)
Fundamentals of geosynthetics
21
13
1·2
E E
1·1
Z
"" .l::
1-0
~
o·g
'c
0'8 0·7
a.
0·6
-~
"c;;
.r:;
a, c: ~
en
Fig. 1.19. Influence of geotextile sample width on its tensile strength (after Myles and Carswell, 1986)
100 200
1000
500 Sample width: mm
Previously it was pointed out that the strength of woven geotextiles is governed by the weaving structure. It has been found that the strength of a woven geotextile is higher at 45° to the warp and weft directions, but is lower parallel to the warp/weft, whereas non-woven geotextiles tend to have a lower but more uniform strength in all directions. One should obtain the minimum strength of the geosynthetic product and ensure that this stress is never exceeded in practical applications. The tensile modulus is the slope of the geosynthetic stress- strain or load- strain curve, as determined from wide width tensile test procedures. This is equivalent to Young's modulus for other construction materials, i.e. concrete, steel, timber, structural plastic, etc. It depicts the deformation required to develop a given stress (load) in the material. Figure 1.22 shows typical load- strain curves for geotextiles and interpretation methods of tensile modulus (Myles and Carswell, 1986). At the commencement of the test, the load will be zero unless a preload is used. As 80
60 E
Z
""<::
a, c: ~
1;)
40 • Woven tape
2!
'iii c:
• Thermally bonded filament
Q)
I-
• Needle-punched filament
20
Fig. 1.20. Variation of tensile strength with mass per unit area for polypropylene geotextiles (after Ingold and Miller, 1988)
o~
o
________
~
200
________- L________ 400
Mass per unit area : 91m 2
~
600
22
Geosynthetics and their applications
Geosynthetic
.
(a)
t
t Geosynthetic
t
.
Di rectio n of strai n
Hinged rods clamping geosynthetic in place with connecting pins
.
• (b)
(c)
Fig . 1.21. (a) Grab tensile test; (b) biaxial tensile test; and (c) plain strain tensile test
the test is begun, the geotextile strains without loading until it reaches the daylight point (a point where the load extension curve parts from the strain). The offset modulus (working modulus) is obtained from the slope of the linear portion of the load-extension data. An offset strain Maximum load
Breaking load
.r:::
'5
''i
~."
c:
:J
Oi
C.
"0to W
mo,",",
....J
~
15
c: Q)
E Q)
.......... Offset strain
"c: Q)
E E
0
u Strain Daylight point (a)
Maximum load
Fig . 1.22. Load-strain curves for geotextiles exhibiting: (a) linear behaviour; and (b) non-linear behaviour (after Myles and Carswell. 1986)
0 ·1 (b)
Strain
Fundamentals of geosynthetics
120
100
..§
z
-'"
80
.i:::'
c;, c
~
'iii
Fig .1.23. Typical strength properties of some geosynthetics (note overlapping zones have not been shown for clarity, s.ome non-typical geosynthetics may lie outside the zones indicated, and some geosynthetics are more sensitive to the test method) (after John , 1987)
~
[[JJ
Strips and woven multifilaments
E;J
Woven tapes
m
Geogrids
----D
Chemically bonded non-woven geotextiles
0 EJ
Thermally bonded non-woven geotextiles
23
Mechanically bonded non-woven geotextiles
60
E
::E :::J
40
20
00
10
20
30
40
50
60
70
Extension: %
is then defined by extending the linear portion of the data back to the zero load line. It is important to understand that the (unknown) strain from the indicated start of the test to the daylight point is eliminated by preloading and that the amount of offset strain is influenced by the amount of preloading. For geotextiles that do not have a linear range, the modulus is typically defined as the secant modulus at 5 or 10% strain. The designer and specifier must have a clear understanding of the interpretation of these moduli. Figure 1.23 shows typical strength properties of some geosynthetics. It is noted that woven geotextiles display generally the lowest extensibility and highest strengths of all the geotextiles. Geogrids have relatively high dimensional stability, high tensile strength and high tensile modulus at low strain levels. They develop reinforcing strength even at strain equal to 2% (Carroll, 1988). The high tensile modulus results from prestressing during manufacture, which also creates integrally formed structures without weak points either in ribs or junctions. In the case of geonets, there is a preferential direction in strength between the machine and cross-machine directions. Geonets have the greatest strength in the machine direction. The viscoelastic behaviour of geosynthetics can produce misleading results for both short-term and rapid-rate tensile tests. Tests conducted to provide design data should also consider long-term conditions and account for the effect of the surrounding soil. The geosynthetic confinement within the soil in the field and the resultant interlocking of soil particles with the geosynthetic structure are found to have a significant effect on the stress- strain properties. It is generally found that the modulus of a geosynthetic confined in soil is likely to be higher than when tested in isolation. The mechanism of this enhancement is simply the frictional force development. The deformation of a geosynthetic structure is, therefore, likely to be overestimated if the in-isolation modulus is used in the calculations (Hausmann, 1990). This fact tends to support the use of a working modulus as an appropriate interpretation method. The confined tensile test methods have been presented by McGown et at. (1982) and EI-Fermaoui and Nowatzki (1982). Due to the high costs involved, confined tensile testing is not carried out on a
24
Geosynthetics and their applications
routine basis. Keeping these facts in view, it should be noted that the wide-width tensile test is essentially an index test. At this stage, it is worthwhile mentioning index and performance tests. Index tests are carried out under standardized conditions used to compare the basic properties of geosynthetic products (e.g. wide-width tensile strength, creep under load, friction properties, etc.). They are generally used in quality control and quality assurance. They are also used to monitor changes that may occur after a geosynthetic has had some sort of exposure. Index tests generally do not reflect design features or applications. Performance tests, on the other hand, are carried out by placing the geosynthetic in contact with a soil/fill under standardized conditions in the laboratory, to provide better simulation of site conditions than index testing. Performance testing, if possible, should also be carried out at full scale at the site. It is to be noted that geosynthetics vary randomly in thickness and weight in any given sample roll due to normal manufacturing techniques. Tests must be conducted on representative samples collected as per the guidelines of available standards, which ensure that all areas of the sample roll and a full variation of the product are represented within each sample group. When two pieces of similar or dissimilar geosynthetics (or related material) are attached to each other, this is known as a 'joint', and when a geosynthetic is physically linked to, or cast into, another material (e.g. the facing panel of a retaining wall - see Chapter 3), this is known as a 'connection'. When no physical attachment is involved between two geosynthetics or a geosynthetic and another material, this is known as an 'overlap'. Where geosynthetic widths or lengths, greater than those supplied on one roll, are required, jointing becomes necessary and the same may be effected by one of the jointing methods, such as overlapping, sewing, stapling, gluing, etc. Different joints, currently in use, may be classified into prefabricated joints and joints made during field applications. In the vast majority of cases, the geosynthetic width or length is extended simply by overlapping, which is usually found to be the easiest field method (Fig. 1.24). Geotextiles may be jointed mechanically, by sewing or stapling, or chemically using an adhesive bond. Figure 1.2S(a) shows the most suitable seam configuration, known as prayer seams. Another type of seam, known as lapped ('J') seam, is shown in Fig. 1.2S(b). Depending on the critical nature of the construction, either a single or double stitch is used. For jointing the geotextiles by the stapling method, corrosion-resistant staples should be used. Figure 1.26 shows the stapled seam configuration. For geosynthetics such as geonets, and
Fig . 1.24. A simple overlap (courtesy of Terram Ltd, UK)
Fundamentals of geosynthetics
.; .;
"
..",.-//\" "j
.;
::«~,>/
Stitch line
.;
.;
25
/' .;
/'
.;
Double stitch line 5- 10 mm apart
(a-ii)
(a-i)
Fig. 1.25. (a) Face to face ('prayer') seams: (i) single stitch line; (ii) double stitch line; and (b) lapped ('J') seam (courtesy of Terram Ltd, UK)
Double stitch line 5- 10mm apart
(b)
geogrids, on the other hand, a bodkin joint may be employed, whereby two overlapping sections are coupled together using a bar passed through the apertures (Fig. 1.27). Geogrids can also be sewn using a robust cord threaded through the grid apertures. Hog rings, staples, threaded loops, wires, etc., are also used for jointing geosynthetics. The list of fieldseaming techniques for plastomeric geomembranes (made from HDPE, LDPE, PE, PP or PVC) includes: fillet extrusion, fiat extrusion, hot air, hot wedge, ultrasonic, and electric welding methods. Figure 1.28 shows some typical seams in geomembranes. Elastomeric geomembranes (made from rubbers of various types as the barrier component) require seaming by means of solution or adhesives. Geosynthetic clay liners are jointed by the application of bentonite at the panel joints. An important criterion for assessing joint performance is load transmission between the two pieces of the geosynthetics. In some applications, it
Fig. 1.26. Stapled seam (courtesy of Terram Ltd, UK)
26
Geosynthetics and their applications
Polymer or other jOint bar
Polymer grid
Fig . 1.27. A bodkin joint
, Fusion
Extrusion lap
Fig. 1.28. Some typical seams in geomembranes (after Giraud, 1994)
Extrusion fillet
may be essential that the load transfer capability is equal to that of the parent material. For other situations, a more important criterion may be the magnitude of the deformation of the joint under load. Seam strength is the load-transfer capability from one geosynthetic roll to another when ends of both the rolls are joined together by any method . The efficiency (E) of a seam joint, between geosynthetic sheets, is generally defined as the percentage of the ultimate tensile strength of the geosyn thetic, which the joint can bear before rupture. It is therefore expressed as:
E = (
T seam T geosynthetic
X
100) %
( 1.2)
where T seam is the wide-width seam strength, and T geosynthetic is the widewidth geosynthetic strength (unseamed). Ideally, the joint would be stronger than the geosynthetic being jointed and would thus never fail in tension. In practice, in the field , high efficiencies are rarely obtained. Publications generally mention that laboratory obtained efficiencies are usually higher than field efficiencies. Thus, this is of little help to the field engineer trying to meet a consultant's specification. However, efforts should be made to make seam efficiency near to 100% . As the fabric strength becomes higher, sea ms become less efficient. Above the geosynthetic strength of 44 kNlm , even the best seams have efficiency less than 100%, and beyond 440 kNlm , the best one can have approximately 50% efficiency (Koerner, 1990). Murray et af. (1986) undertook research work into the seam strengths obtained from both sewn and adhesive bonded seams. Their work was comprehensive and stated that 100% efficiency could be obtained using adhesives. With sewn joints, they described efficiencies up to 90 % but they drew attention to the large deformations that are experienced. The technique of jointing geogrids by means of a bodkin joint proved to be an effective procedure, whereby load-carrying efficiencies of about 90% were obtained . Rankilor and Heiremans (1996) reported that the use of adhesives can reduce seam extension dramatically. There are some mechanical properties of geosynthetics, which are related to geosynthetic survivability and separation function. Such tests
Fundamentals of geosynthetics
Field problem
27
Laboratory simu lation
Burst test
Subgrade
--t"f'-Pressure
Grab tensile test
Drop cone method (dynamic impact) CBR plunger test (qUaSi~stat iC)
Fig . 1.29. Laboratory tests related to geosynthetic survivability and separation function (after Hausmann , 1990)
,
i!1 D are known as integrity tests and are as follows. • Fatigue strength - ability of geosynthetics to withstand repetitive loading before undergoing failure. • Burst strength - ability of geosynthetics to withstand loading when no further deformation is possible. • Tear strength - ability of geosynthetics to withstand tearing stresses often generated during their installation. • Impact strength - ability of geosynthetics to withstand stresses generated by falling objects, such as rock pieces, tools and other construction items. • Puncture strength - ability of geosynthetics to withstand stresses generated by penetrating objects, such as pieces of rock or wood, under quasi -static condition. Figure 1.29 explains the fundamental concepts of some laboratory integrity tests . When a geosynthetic is used in reinforcing a soil mass, it is important that the bond developed between the soil and the geosynthetic is sufficient to stop the soil from sliding over the geosynthetic or the geosynthetic from pulling out of the soil when the tensile load is mobilized in the geosynthetic. The bond between the geosynthetic and the soil depends on the interaction of their contact surfaces. The soil- geosynthetic interaction or interface friction is thus the key element in the performance of the geosynthetic-reinforced soil structures. It is used to determine the bond length of the geosynthetic needed beyond the critical zone. Two test procedures, currently used to evaluate soil- geosynthetic interaction, are the direct shear test and the pullout (anchorage) test. In direct shear test, the upper half of the box is fixed, while the lower half is subjected to a horizontal force. In this case, the geosynthetic sample is anchored along the edge of the box where the tensile (horizontal) force is applied. In a pullout test, the two halves of the box are fixed and one end of the geosynthetic sample is subjected to a horizontal force . Figure 1.30 depicts both the
28
Geosynthetics and thei r applications
Geosynthetie layers
II
••• : •• " rl ' ,-:-: ' .-;-:,.•• :'-: '':' --''-:-,'-:''':''''''' :' '":''1
~3t{3!fiJ:I'~ Retain ing wall (Pullout failu re mode)
(Sliding failure mode)
Norma l force
Fig. 1.30. Comparison of pullout and direct shear tests with corresponding reinforcing app lication (after Paulson , 1987)
'. . . . . ... .' . .. . . .
Pullout test
force
Oi reet shear test
tests and the corresponding reinforcing applications with the type of failure mode. A designer of geosynthetic-reinforced soil structures must consider the potential failure mode, then the appropriate test procedure should be used to evaluate the soil- geosynthetic interaction properties. In the case of an unpaved road, the recommended test should be a combination of direct shear and pullout tests conducted simultaneously (Giroud, 1980). Ingold and Miller (1988) reported that, for most geotexti les, the ratio of the angle of shearing resistance of geotextilereinforced granular soil to the angle of shearing resistance of unreinforced granular soil, as determined from direct shear tests, rarely drops below 0·75 and is often close to unity; however, there may be exceptions.
1.6.3. Hydraulic properties
Hydraulic testing of geosynthetics is completely based on new and original concepts, methods, devices, interpretation and databases, unlike the physical and mechanical testing, as discussed in previous sections of this chapter. The reason behind this is that the traditional textile tests rarely have hydraulic applications. Porosity, permittivity and transmissivity are the most important hydraulic properties of geosynthetics, especially geotextiles, geonets and some drainage composites, which are explained below. Geosynthetic porosity is related to the ability to allow liquid to flow through it and is defined as the ratio of the void volume to the total volume. It may be indirectly calculated for geotextiles using the relationship given below (Koerner, 1990): m 1]=1 - (1.3) pt where 1] is the porosity; m is the mass per unit area; p is the overall geotextile density; t is the thickness of the geotextile. Per cent open area (POA) of a geosynthetic is the ratio of the area of its openings to its total area and is expressed in per cent. The pores in a given geosynthetic, especially in a geotextile, are not of one size but are of a range of sizes. The pore-size distribution can be represented in much the same way as the particle size distribution for a soil. Various methods
Fundamentals of geosynthetics
29
Table 1.5. Comparison of methods for determining pore-size distribution of geotextiles Test method
Test mechanism
Test material
Sample size: cm 2
Time for one test
Dry sieving Hydrodynamic sieving Wet sieving Bubble point
Sieving dry Alternating water flow Sieving wet Comparison of air flow , dry versus saturated Intrusion of a liquid in a pore Direct measurement of pore spaces in cross-section of the geotextile
Glass beads fraction Glass beads mixture Glass beads mixture Porewick
434 257 434 22·9
2h 24h 2h 20min
Mercury intrusion Image analysis
Mercury None
1'77 1·5
35min 2-3 days
are available for evaluating the pore-size distribution of geotextiles. Bhatia et al. (1994) made a comparison of six methods as presented, in Table 1.5. In the dry sieving test method , glass beads of a known size are sieved in dry condition through a screen made of the geotextile (Fig. l.31). Sieving is done by using beads of successively coarser size until 5% or less, by weight, pass through the geotextile. The hydrodynamic test method is based on hydrodynamic filtration , where a glass bead mixture is sieved with a basket with a geotextile bottom by alternating water flow that occurs as a result of the immersion and emersion of the basket several times in water. In the wet sieving method, a glass bead mixture is sieved through a screen made of a geotextile while a continuous water spray is applied . The bubble point method is based on the following: • a dry porous material will pass air through all of its pores when any amount of air pressure is applied to one side of the material , and • a saturated porous material wi ll only allow a fluid to pass when the pressure applied exceeds the capillary attraction of the fluid in the largest pore. The mercury intrusion method is based on the relationship between the pressure required to force a non-wetting fluid (mercury) into the pores of a geotextile and the radius of the pores intruded. Image analysis is a technique used for the direct measurement of pore spaces within a cross-sectional plane of a geotextile with the help of a microscope. The pore openings, which are obtained experimentally, are dependent on the technique used for their determination. It is believed that, despite some limitations, both wet and hydrodynamic sieving methods are better techniques than dry sieving. In the case of most geogrids, the open areas of the grids are greater than 50% of the total area. In this respect, a geogrid may be looked at as a highly permeable polymeric structure. Lid
Single-sized glass balls
~:
iI
~- Geotextile
'"" [r~:-;:;-;,-~:-;>::~ Fig . 1.31 . Diagram showing details of dry sieving method (courtesy of Terram Ltd, UK)
1 .". , ,,.;..:,.;,,;. ;.
<.;..
~
~
t-
Vibrating unit
30
Geosynthetics and their applications
100
80
'c"
60
Woven geotextile
If)
!':'
Heat bonded non-woven geotextile
o
c. ~
40
'" N
·iii
'"
(; c.
20
Fig. 1.32. Pore size distributions of typical geotextiles (after Ingold and Miller, 1988)
o
L -_ _ _ _
50
~L-
100
__
~
_ _LL_ _
~
____
200 500 Pore size: microns
~
1000
Figure 1.32 shows pore-size distribution curves for typical woven and non-woven geotextiles. The pore size at which 95% of the pores in the geotextile are finer, is originally termed the equivalent opening size (EOS) designated as 0 95 . If a geotextile has an 0 95 value of 300 11m, then 95% of geotextile pores are 300l1m or smaller. In other words, 95% of particles with a diameter of 300 11m are retained on the geotextile during sieving for a constant period of time. This notation is similar to that used for soil particle size distributions where, for instance, D IO is the sieve size through which 10%, by weight, of the soil passes. The apparent opening size (AOS) is equivalent to the EOS but is also quoted for other percentages retained, such as Oso or 0 90 . The EOS is used in many filter criteria established to prevent piping and erosion. It should be noted that the meaning of EOS and AOS values and their determination in the laboratory are still not uniform throughout the engineering profession and, hence, filter criteria developed in different countries may not be directly comparable. In Fig. 1.32, it is noted that the pores in a woven geotextile tend to be fairly uniform in size and regularly distributed. In general, non-woven geotextiles exhibit smaller 0 90 pore sizes than wovens; however, there is a degree of overlap in the commonly employed 0 90 sizes, which vary from approximately 50- 35011m for the non-wovens and from 150600 11m for the wovens (Ingold and Miller, 1988). For filtration application, a geotextile high in POA should be selected, with a controlled opening size to suit the soil being filtered. Most non-woven geotextiles, and some woven geotextiles, will suit this application. The permeability of a geosynthetic to water flow may be expressed by Darcy's coefficient, by permittivity (as defined below) or by a volume flow rate. The advantage in expressing geosynthetic permeability in terms of Darcy's coefficient is that it is easy to relate geosynthetic permeability directly with soil permeability. A major disadvantage is that Darcy's law assumes laminar flow , whereas geosynthetics, especially geotextiles, are often characterized as exhibiting semi-turbulent, or turbulent flows . The simplest method of describing the permeability characteristics of geosynthetics is in terms of volume flow rate at a specific constant water head (generally 10 cm) (Fig. 1.33). The advantage of this method is that it is the simplest test to carry out, it does not rely on Darcy's law for its authenticity, and it can easily be used to compare different geosynthetics used for drainage and filtration applications. The measurement of in-plane water permeability is important if the geosynthetic, such as a geotextile or a fin drain, is being used to drain
Fundamentals of geosynthetics
31
Water head cylinder
I+--#-- Loading rod
Porous loading discs Clamped flanges
Fig . 1.33. Diagram showing details of apparatus for measuring permeability of geosynthetic for water flow normal to its plane (courtesy of Terram Ltd, UK)
.~==:_
Geotextile
Outflow
Outflow
Collection reservoir
water within itself, i.e. its water transporting capability is of prime importance. The in-plane water permeability is normally described in terms of transmissivity (as defined below). The type of tests used to measure the in-plane drainage characteristics of geosynthetics are essentially the same as those used to measure water permeability normal to the plane of the geosynthetic (Fig. 1.33), except that the hydraulic gradient is applied along the length of the geosynthetic (Fig. 1.34) rather than across the thickness of the geosynthetic. Permittivity of a geosynthetic (generally geotextile) is simply the coefficient of permeability for water flow normal to its plane (Fig. 1.35(a)) divided by its thickness. This property is the preferred measure of water flow capacity across the geosynthetic plane and quite useful in filter applications. Darcy's law, in terms of permittivity, can be expressed
Inflow
~==-~~ Fig . 1.34. Diagram showing details of apparatus for measuring permeability of geosynthetic for water flow within its plane (courtesy of Terram Ltd, UK)
Appr
loodlog
l~l!l~lt~:t'i~·i,,;-~ Composite fin drain or geotextile
Outflow
32
Geosynthetics and their applications
Waterflow direction
84 Clx0c=---_ _-./ '
T i= I-======~_-=J - L - - -_ _-j (a)
Fig . 1.35. Flow of water through a geotextile strip: (a) normal flow; and (b) in-plane flow
(b)
as follows: ( 1.4) where Qn is the cross-plane volumetic rate of flow (m 3/s), i.e. volumetric rate of flow for flow across the plane of the geosynthetic, k n is the coefficient of cross-plane permeability (m/s), 6h is the head causing flow (m), 6 x is the thickness of the strip of geosynthetic measured along the flow direction under a specified normal stress (m), L is the length of the strip of geosynthetic (m), B is the width of the strip of geosynthetic (m), 'lj; = k n / 6 x, which is the permittivity of the geosynthetic (S- I), and An = LB, which is the area of cross-section of geosynthetic for crossplane flow (m 2). Permittivity may thus be defined as the volumetric rate of flow of water per unit cross-sectional area, per unit head, under laminar flow conditions in a direction normal to the plane of the geosynthetic. Transmissivity of a geosynthetic (thick non-woven geotextile, geonet or geocomposite) is simply the product of the permeability for in-plane water flow (Fig. l.35(b)) and its thickness. This property is the preferred measure of the in-plane water flow capacity of a geosynthetic and is widely used in drainage applications. D arcy's law in terms of transmissivity can be expressed as follows :
Qp
= kp ~h Ap = kp ~h (B6 x) = (JiB
( 1.5 )
where Qp is the in-plane volume tic rate of flow , i.e . volumetric rate of flow for flow within the plane of the geosynthetic (m 3 I s) , kp is the coefficient of in-plane permeability; (j = k p 6 x, which is the transmissivity of the geosynthetic (m 2 I s) , i = 6hl L , which is the hydraulic gradient, and Ap = B6x which is the area of cross-section of geosynthetic for inplane flow (m 2). Transmissivity may thus be defined as the volumetric rate of flow of water per unit width of the geosynthetic, per unit hydraulic gradient, under laminar flow conditions within the plane of the geosynthetic. To exhibit a large transmissivity, a geotextile must be thick and/or have a large permeability in its plane. Equations (1.4) and (l.5) indicate that once permittivity ('lj;) and transmissivity «(j) are successfully determined, the flow ra tes Qn and Qp do not depend on thickness of the strip of geosynthetic, 6 x, which is highly
Fundamentals of geosynthetics
33
dependent on the applied pressures and therefore, is difficult to measure accurately. Typical values of permeability are 10- 5 to 1 m/s for geotextiles and 10- 13 m/s or less for geomembranes. The permeability of geotextiles is of the same order of magnitude as the permeability of highly permeable soils, such as sand and gravel. Woven geotextiles and thermally bonded non-woven geotextiles have almost no transmissivity and cannot be used as drains. The permeability of geomembranes is much smaller than the permeability of clay, which is the least permeable soil. Needlepunched geotextiles have permeability values of the order of 10- 4 or 10- 3 m/s and geonets have permeability values of the order of 10- 2 or 10- 1 m/s. A maximum saturated hydraulic conductivity ranging from 5 x 10- 11 to 1 X 10- 12 m/s is typical of geosynthetic clay liners over the range of confining pressures typically encountered in practice. It should be noted that Darcy's law is valid only for laminar flow . This means that permeability, permittivity and transmissivity are constants, i.e. independent of the gradient only if the water flow is not turbulent. These properties are governed by several other factors , such as fibre type, size and orientation; porosity or void ratio; confining pressure; repeated loading; contamination; and ageing. When dry, some fabrics exhibit resistance to wetting. In such cases, initial permeability is low but rises until the fabric reaches saturation. Permeability may also be reduced through air bubbles trapped in the geosynthetic. This is the reason why testing standards usually require careful saturation of the geosynthetic specimens before they are subjected to water flow . In addition, permeability measurements will be more consistent with the use of de-aired water rather than tap water (Hausmann, 1990). Woven geotextiles are much less affected by stress level, but their permeability is dramatically controlled by the structure of the fabric. The common, and generally less expensive, tape-on-tape fabrics have a low open area ratio and, in consequence, exhibit water permeabilities typically in the range 10- 30 l/s per m 2 for a lOcm head. In contrast, the woven monofilament-on-monofilament geotextiles have much larger open area ratios, giving water permeabilities in the range 100-1000 I/s per m 2 for a 10cm head (Ingold and Miller, 1988). Tests performed at the University of Grenoble (France) have shown that the thickness and the permeability of needle-punched non-woven geotextiles are significantly affected by confining pressure as shown in Fig. l.36 (Giroud, 1980). In this figure , the values of kn and kp were close for the considered geotextile; therefore, only an average value, k, is presented. It may be noted that the flow in the plane of the geotextile is more affected by the confining pressure than a normal flow . When a geotextile is placed adjacent to a base soil (the soil to be filtered) , a discontinuity arises between the original soil structure and the structure of the geotextile. This discontinuity allows some soil particles to migrate through the geotextile under the influence of seepage flows . This condition is shown in an idealized manner in Fig. 1.37(a). For a geotextile to act as a filter , it is essential that a condition of equilibrium is established at the soil/geotextile interface as soon as possible after installation to prevent soil particles from being piped indefinitely through the geotextile; if this were to happen, the drain would eventually become blocked. The time taken to establish equilibrium conditions with geotextile filters varies, but is normally between one and four months. The structure, or stratification, of the soil immediately adjacent to the geotextile at the onset of equilibrium conditions dictates the filtering efficiency of the system . The stratification is dependent on the type of
34
Geosynthetics and their applications
E'" 10- 2
4
.;.:
E E
Q; iii
)(
~
",-
'c"
Q)
B
2
.~ :0
-'"
.~
a!
~
Q)
I-
E Q; a.
0
10- 1L-____- L______L -____ ______ 0·5 1·5 2 ·0 o Compressive stress : MPa (e) ~
~
2·0 Compressive stress: MPa (d)
Fig. 1.36. Influence of compressive stress on: (a) thickness; (b) permeability; (c) permittivity and (d) transmissivity of a needle-punched non-woven geotextile (after Giroud, 1980)
Soil matrix
Geotextile Soil piping
Seepage flow (a)
Seepage flow (b-i)
Soil matrix
Fig . 1.37. (a) Idealized soil-geotextile interface conditions immediately fol/owing geotextile instal/ation; (b) idealized interface conditions at equilibrium between three different soil types and geotextile filter: (i) single-sized soil and geotextile filter , (ii) weI/graded soil and geotextile filter , (iii) cohesive soil and geotextile filter (courtesy of Terram Ltd, UK)
Geotextile Seepage flow (b-ii)
l~:i:ji{j·t~~"O:-c·l;( +-+-+-+-+ Seepage flow (b-iii)
So il
Geotextile
Fundamentals of geosynthetics
Unstable conditions
35
Equilibrium conditions
~~------~~~--------------
.g=~Oilfilter forming
Soil filter formed
OJ
E
~ ~~~-------------------
OJ
iii
en'" Time "0 OJ
c. '0.
'0 Ul
Unstable conditions Soil filter
i~
Fig . 1.38. Overall requirements for optimal filter performance (after Lawson , 1986).
I.
Equilibrium conditions Soil filter formed
Time
soil being filtered , the size and frequency of the pores of geotextile, and the magnitude of the seepage forces present. Figure 1.37(b) shows typical stratification occurring with three different soil types -- single-sized soil , well-graded soil and cohesive soil. When the soil is well-graded , considerable rearrangement of the soil takes place. At equilibrium, three zones may be identified: the undisturbed soil, a 'soil filter' layer, which consists of progressively smaller particles as the distance from the geotextile increases, and a bridging layer, which is a porous, open structure . Once the stratification process is complete, it is actually the soil filter layer that actively filters the soil. If the geotextile is chosen correctly, it is possible for the soil filter layer to be more permeable than the undisturbed soil. The function of the geotextile is to ensure that the soil remains in an undisturbed state without any soil piping (Fig. 1.38). To achieve satisfactory filter performance by geosynthetics, especially geotextiles, the following functions are required during the design life of the application under consideration. (a)
(b)
(c)
Maintain adequate permeability to allow flow of water from the soil layer without significant flow impedance so as not to build up excess hydrostatic pore-water pressure behind the geosynthetic (permeability criterion). Prevent significant wash out of soil particles, i.e. soil piping (retention or soil-tightness criterion). Avoid accumulation of soil particles within the geosynthetic structure, called fabric clogging, resulting in complete shut off of water flow (clogging criterion).
It may be noted that the permeability criterion places a lower limit on the pore size of a geotextile, whereas the retention criterion places an upper limit on the pore size of a geotextile. These two criteria are, to some extent, contradictory, because the permeability of a geosynthetic filter increases with its increasing pore size. However, in the majority of cases, it is possible to find a filter that meets both the permeability criterion and the retention criterion. Several different geosynthetic filter criteria have been developed (Giroud, 1982; Lawson, 1982; Hoare, 1982; Lawson, 1986; Wang, 1994), largely based on the conventional granular filter criteria that were first formulated by Terzaghi and Peck (1948). All of these criteria use soil permeability and compare it to the geotextile permeability for establishing the permeability criterion,
36
Geosynthetics and their applications
whereas they compare soil particle size distribution to geotextile pore-size distribution for establishing the retention criterion . All these criteria are applicable for specific filter applications . It is a general misconception that the pore sizes of a filter should be smaller than the smallest particle size of the soil to be protected, because it would lead to using quasi-impermeable filters (which, of course, would not meet the permeability criterion) . In some cases, the filter openings can be larger than the largest soil particles and the filter will still retain the soil (Giroud, 1994). It should be noted that soil retention does not require that the migration of all soil particles be prevented. Soil retention simply requires that the soil behind the filter remains stable; in other words, some small particles may migrate into and/or through the filter provided that this migration does not affect the soil structure, i.e. does not cause any movement of the soil mass. At the same time, the filter and the drainage medium located downstream of the filter should be such that they can accommodate the migrating particles without clogging. The criteria of geotextile filter commonly used are III the following form: k n 2: Aks (permeability criterion) 0i
~
f3Dj (retention criterion)
( 1.6) (1.7)
where k n is the coefficient of cross-plane permeability of geotextile, k s is the coefficient of permeability of protected soil, A is the constant of permeability varying over a wide range, say 0·1 to 100, f3 is the dimensionless parameter of soil conservation, varying over a certain range, and i and j are integers. The permeability criterion, k n 2: k s, has long been advocated by many researchers on the assumption that the geotextile needs to be no more permeable than protected soil. Carroll (1983), and Christopher and Holtz (1985), recommend the criterion, k n 2: 10ks , for critical soil and hydraulic conditions in which clogging has been shown to cause, roughly, an order of magnitude decrease in the geotextile permeability. The criterion, k n 2: O·lk s , was proposed by Giroud (1982) on the premise that a geotextile with only 10% of the permeability of the soil would still have a much greater flow capacity than the soil because the length of the flow path is directly related to the flow rate through a porous media. All the existing retention criteria for geotextile filters are functions of various opening sizes of the geotextile, such as 0 9S , 0 90, Oso and OI S, and the diameter of soil particles, such as D 90, D 8S , Dso and DI S, depending mostly on the uniformity coefficient of the soil, Cu (D60 / DIO)' Most of the criteria are given in the form of the Oi / Dj ratio not exceeding a certain value or range. Typical ranges of variations of 0 9S/ Dso, 0 9S/ D8S apd 090 / D90 are, respectively, 1- 6, 1- 3 and 1- 2. The filter criteria, for a particular filter application in the field , should be developed on the basis of data obtained from detailed soil- geotextile performance. However, in the absence of such data, the following criteria can be considered for soils with predominantly granular fractions : permeability criterion: 0 90 2: DI S
(1.8)
and
0 90 2: 0.05 mm retention criterion: 0 90
( 1.9) ~
D8S
(1.10)
Fundamentals of geosynthetics
37
In the absence of suitable data, for soils with significant cohesive fractions, it is suggested that the following criteria can be used: 0 90
permeability criterion: retention criterion:
0 90
~
:::;
(1.11)
0·12 mm
0·05 mm
( 1.12)
and minimum volume water permeability = 30 11m2Is
-
10cm head (1.13)
The clogging criterion is discussed in the following section.
1.6.4. Endurance and degradation properties The endurance and degradation properties (creep behaviour, abrasion resistance, long-term flow capability, durability - construction survivability and longevity, etc.) of geosynthetics are related to their behaviour during service conditions, including time. Creep is the continued strain or elongation of a geosynthetic when subjected to a sustained load. It is an important factor in the design and performance of geosynthetic-reinforced structures having longer design life, say for about 50 years . Depending on the type of polymer and ambient temperature, creep may be significant at stress levels as low as 20% of the ultimate strength. Figure 1.39 compares strain versus time behaviour of various yarns of different polymers. As shown, both the total strain and the rate of strain differ markedly. The understanding of the geosynthetic creep helps the design engineer in the selection of allowable load to be used in designs. Two approaches to evaluate the allowable load are as follows:
(a)
(b)
Allowable load based on limiting creep strains - this requires the analysis of creep strains versus time plots for various stress levels. Details of this procedure were described by Jewell (1986) and Bonaparte and Berg (1987). Allowable loads using factors of safety - it is required to reduce the geosynthetic strength by a factor of safety corresponding to (a)
5
1 yr
1d
~:E
4 ~ c: 3
·iii
PET
~ 2
0
1h
2
0
3
4
5
6
7
Log time : 5
1 yr
(b)
~
30 0~
c:
Fig . 1.39. Results of creep tests on various yarns of different polymers: (a) creep at 20% load; (b) creep at 60 % load (after den Hoedt, 1986)
20
PA
.~
U5
10
PET 0 0
2
3
4
Log time: 5
5
6
7
38
Geosynthetics and their applications
Table 1.6. Factors of safety (after den Hoedt, 1986) Polymers Polypropylene Polyester Polyamide (nylon) Polyethylene
Factor of safety
4·0 2·0
2·5 4·0
the specific polymer type to obtain the allowable load. Values of factor of safety are given in Table 1.6. Although not as technically accurate as the previous method, this approach is sometimes the only one available to the designer. Since polymers are visco-elastic materials, strain rate and temperature are important while testing geosynthetics (Andrawes et aI. , 1986). When a low strain is applied in a wide-width tensile test, the geosynthetic sample takes longer to come to failure and, therefore, the creep strain is greater. High rates of strain (which can be as much as 100% per minute) tend to produce lower failure strains and sometimes yield higher strengths than do low rates of strain. The creep rate of geosynthetics depends on temperature. Higher creep rates are associated with higher temperatures, resulting in larger strains to rupture geosynthetics. The rate of creep is also related to the level of load to which the polymer is subjected (Greenwood and Myles, 1986; Mikki et al. , 1990). Chang et al. (1996) reported that under the same confining pressure on geotextiles, the amount of creep increases as the creep load rises; and where the creep load is the same, the increases in confining pressure decrease the amount of creep, which may even be reduced to nil. The creep is minimized by a prestretching operation of the 'Tensar' process. It has been found that the tensile and creep properties of some nonwoven geotextiles can be improved by confinement in soil (McGown et al., 1982). This has a greater effect on the tensile properties of the mechanically bonded geotextile and those of the heat-bonded geotextile. Creep reduction under confinement may be due to a significant increase in frictional resistance between the soil and the geosynthetic. However, there is little effect from confining pressure on the performance of woven geotextiles. In some applications, increases in soil strength result in the reduction of geosynthetic stresses with time. An example of this type is the foundation support for a permanent embankment over soft deposits (Fig. 1.40(a)). When a basal geosynthetic is used beneath an embankment, the strain becomes fairly constant once most of the settlement has taken place. In such a situation , there may be loss of tensile stress with time experienced by the geosynthetic. This phenomenon is called stress relaxation, which is closely related to creep. Fortunately, during this period, the underlying soil is consolidating and increasing in strength. The subsoil is therefore able to offer greater resistance to failure as time passes. The factor of safety should not be compromised unless the geosynthetic sheds load faster than the gains in soil strength. In a permanent geosynthetic-reinforced retaining wall or steep embankment, the geosynthetic load remains constant throughout the structure's life. In this case, the creep strain may be very high and, therefore, the factor of safety should not be compromised (Fig. 1.40(b)). Dynamic creep or repeated loading behaviour of geosynthetics are of paramount importance in a number of applications. These include
Fundamentals of geosynthetics
Geosynthetic
Steep slope
Embankment
~
~
Soft compressible foundation soil
,
~
Geosynthetic " layers
Factor of safety
"11 c
, ...
End of construction "-
"-
"-
"-
....
--
Factor of safety
... ...
en
c 'iii
e'"
" E
--- End of construction
.... ------------
1
Time
-- --
en
"I ! E
Time "-
en
c 'iii
't;" Ql
E Reinforcement loading (a)
-------~----.}
Firm stratum
Firm stratum
\
,,
- - - ---- ---=----,.:}~
-
en
39
!
"-
"-
"-
-------------
Reinforcement loading (b)
Fig. 1.40. Examples of: (a) time-dependent; and (b) time-independent reinforcement applications (after Paulson , 1987)
reinforcement in paved and unpaved roads, reinforced retaining structures and slopes under large repetitive live loads, such as traffic and wave action. Behaviour of the geogrid under repeated loading is generally different from those of the geotextiles (Fig. 1.41). Due care must be paid to such applications. In the recent past, constitutive models have been developed to describe direction and time-dependent, non-linear, inelastic stress- strain behaviour. More details on this type of model can be found in the works of Perkins (2000). Abrasion of a geosynthetic is defined as the wearing away of any part of it by rubbing against any surface. It is reported as the percentage weight loss or strength/elongation retained under a specified test in particular conditions. In testing the abrasion resistance of geosynthetics, it is important to simulate the actual type of abrasion which a geosynthetic would meet in the field. Van Dine et at. (1982) and Gray (1982) suggested the test procedures for evaluation of resistance to abrasion caused by different processes such as wear and impact. The long-term flow capability of geosynthetics (generally geotextiles) with respect to the hydraulic load coming from the upstream soil is of significant practical interest. The compatibility between the pore-size openings of a geotextile and retained soil particles in filtration and/or drainage applications can be assessed by the gradient ratio test. This test is basically used to evaluate the clogging resistance of geotextiles. Figure 1.42 shows the constant-head-type permeameter developed by the US Army Corps of Engineers. This permeameter allows the measurement of the head loss along a soil- geotextile system. After the test is run for some hours (or days), the piezometer readings stabilize and the
40
Geosynthetics and their applications
20 Woven polypropylene
15
W 10
o1
10
100
1000
Number 01 repetitions , N (a)
8 Geogrid
Fig . 1.41. Total strain versus lo~ N plot: (a) woven geotextile; (b) geogrid (after Kabir and Ahmed, 1994)
10
100
1000
Number 01 repetitions, N (b)
so-called gradient ratio (GR) is determined. It is defined as the hydraulic gradient through the lower 25 mm of the soil plus geotextile, divided by the hydraulic gradient through the adjacent 50 mrn of soil. The longterm flow rate behaviour of geotextile filters can also be assessed by the accelerated filtration test method (Mikki et al. , 1994). In such filtration
~ ~Di scharge
Cylinder
25mm
t
f Fig.1.42. Gradient ratio permeameter developed by US Army Corps of Engineers (after Haliburton and Wood, 1982)
Geotextile
Fundamentals of geosynthetics
41
tests, the following condition should be satisfied in order to ensure satisfactory performance in the field: GR :::::: 3·0 (clogging criterion)
( 1.14)
It is imperative that geotextile filters should be placed in close contact with the soil , so that there will be no space between the soil and the geotextile where particles could move or accumulate to form a thin layer of soil (often called soil cake) at the surface of a geotextile filter. The mechanism of the formation of a soil cake is known as blinding of the filter and this is far more detrimental than clogging. It has been reported by Giroud (1994) that geotextiles with a tortuous surface in contact with the soil , such as needle-punched non-woven geotextiles, do not favour the development of a continuous cake of the fine soil particles, whereas geotextile filters with a smooth surface may favour the development of such a cake . Furthermore, geotextiles with a tortuous surface do not favour the mechanism of blocking, because they do not have individual opemngs. Situations, such as those involving poorly-graded soils, gap-graded soils, cohesionless sands and silts, high hydraulic gradients, very high alkalinity groundwaters, etc., have been identified as creating severe clogging problems. Under such situations, one should avoid the use of a geotextile filter and should use a granular filter , or should open up the geotextile to the point where some soil loss will occur, if the upstream conditions permit such soil loss. Situations involving biological elements may also create clogging problems in filter applications. The durability of a geosynthetic may be regarded as its ability to maintain requisite properties against environmental or other influences over the selected design life. It is dependent, to a great extent, upon the composition of the polymers from which it is made. To quantify the properties of polymers, a knowledge of their structures at the chemical , molecular and supermolecular level is necessary, which was provided by Cassidy et al. (1992) . The durability of geosynthetics has traditionally been assessed on the basis of mechanical property test results, and not on the microstructural changes that cause the changes in the mechanical properties. From an engineering point of view, it may be expressed in terms of a loss of strength and studied as construction survivability and longevity. Construction survivability addresses how the geosynthetics survive during installation. Geosynthetics may suffer damage during installation due to placement and compaction of the overlying fil l. In some cases, the installation stresses might be more severe than the actual design stresses for which the geosynthetic is intended. The installation damage is taken into account during design by reducing the characteristic strength by a partial safety factor varying from 1·05 to 1·70 for thin to thick types. Longevity addresses how the geosynthetic properties change over the life of the structure. All geosynthetics are likely to be exposed to sunlight and weather during construction on site, and must therefore have a limited resistance to weathering. In service life, most of them will be covered by soil , whi le those that remain exposed during their entire life will need a far greater degree of resistance. Generally, as the ambient temperature is increased , the strength, creep, and durability characteristics of geosynthetics deteriorate. Geosynthetics are likely to encounter high temperatures only in paving applications, where they come into contact with hot bituminous materials. This application favours the use of polypropylene grids in preference to
42
Geosynthetics and their applications
polyethelene grids, because of their greater temperature resistance. Polypropylene melts at 165°C and polyester at 250°C. These high temperature extremes should be avoided. Geosynthetics may degrade when exposed to the ultraviolet (UV) component (wavelength shorter than 400 nm) of sunlight. As the geosynthetics will be buried throughout their life, in most applications, UV degradation is not a major cause of concern, provided sensible placement procedures are followed. Generally, those geosynthetics that are white or grey in colour are likely to be the most vulnerable to UV degradation . Carbon black and other stabilizers are used to provide the polymer with UV protection. Polypropylene products perform worst, with the majority having a 50% strength loss in about 4 to 24 weeks. The study of long-term performance of geosyntheics in sunlight can be carried out by exposing the geosynthetics to natural radiation or artificial radiation, such as carbon (or xenon) arc lighting in a laboratory. The artificial radiation has the advantage that, not only can testing be accelerated by increasing mean irradiance level and temperature, and eliminating the cycles of night and day, winter and summer, but equally important is that the exposure parameters can be controlled . The UV degradation test results must be analysed for practical applications, keeping in mind geographic location, radiation angle, temperature, humidity, rainfall, wind, air pollution, etc., associated with a particular construction site. Geosynthetics may come into contact with chemicals/leachates that are not normally part of the soil environment. Site-specific tests must be performed to know the chemical degradation of geosynthetics, resulting in deterioration in their engineering behaviour. Index tests are generally used in chemical-resistance studies. Chemical-resistance testing of geosynthetics should employ worst-case scenario conditions. This is necessary to ensure that when in actual use, the geosynthetics will not be subjected to conditions worse than those experienced in the testing laboratory. Chemical resistance can be evaluated, at least initially, by comparing chemicals anticipated in the application with manufacturers' published resistivity charts. Tests on geotextiles in contact with uncured concrete indicate that polypropylene products are largely unaffected, whereas polyester products can lose about 50% of their strength in two months of prolonged exposure (Wewerka, 1982). All polymeric materials have a tendency to gain moisture over time. This moisture causes some swelling but probably not enough to cause measurable changes in mechanical or hydraulic properties. All geosynthetic resins are very high in molecular weight with relatively few chain endings for the biological degradation to initiate. Biological degradation differs in that so far there has been no evidence of any such degradation in geosynthetics. Only those based on natural fibres degrade, as is the intention. Biological degradation cannot be accelerated beyond the selection of optimum soil conditions and temperature; try to accelerate it further and the microorganisms will be destroyed. It can be studied by conducting soil burial tests in which geosynthetic specimens are buried in a microbially active soil. There is no need to inoculate the soil with specific bacteria or fungi; all relevant species are assumed to be already present and those that benefit from the nutrients in the geosynthetic, if any, will multiply and accelerate the attack. The soil must be allowed to stabilize before the specimens are placed in it. A sample of untreated cotton is used to test the soil - if the tensile strength of the cotton strips is less than 25% of the original tensile strength after a seven-day exposure, the soil is regarded as biologically
Fundamentals of geosynthetics
43
active. A good-quality horticultural compost should be sufficient for soil burial tests (Greenwood et al., 1996). Ozone attack, termite attack, etc. , may also degrade geosynthetics. The research work needs more attention in this direction, as well as with regard to the combination of other degradation mechanisms. Ageing and burial test procedures and results are becoming more critical as the long-term demands on geosynthetics increase. This is an area that requires much work. The knowledge and understanding of long-term, in-service behaviour of geosynthetics are vital to the continued growth of this industry, and to the science of geosynthetic design. It must be noted that there can be many ways to perform a given test in the laboratory, depending on the case to be designed. The recommended way is the way that best simulates the actual performance of the geosynthetic at the site. Usually, a laboratory test simulates the field situation at only one point of the geosynthetic. When the whole field situation can be simulated in a laboratory test, the test results can be applied to the field situation either directly or using a minor mathematical treatment to deal with the difference in scale between the laboratory and the field. In this test, the test is a model test and an analogical method of design is used. This method of design is the simplest one but it can rarely be used, so other methods, such as the analytical method (based on mathematical theories and the basic parameters of geosynthetics) and empirical methods (based on experience and, sometimes, systematic testing, including full-scale tests) are needed (Giroud, 1980). There are presently a large number of geosynthetic products available commercially, each having different properties, and their inclusion in this chapter is beyond the scope of the book. Users can get specific values of various properties of the geosynthetic products from the manufacturers or suppliers. Some representative properties of commercially available geosynthetics are listed in Table 1.7. If the test method for determining the geosynthetic properties are not completely field -simulated, the test values must be adjusted. For example, the laboratory-generated tensile strength is usually an ultimate value, which must be reduced before being used in design. This can be carried out using the following equation suggested by Koerner (1990): Tallow = T ult [FSm
X
FSC R
~ FS
CD
X
FS BD ]
( 1.15)
where Tallow is the allowable tensile strength to be used in equation (l.l) for final design purposes, T ult is the ultimate tensile strength from the test, FS m is the factor of safety for installation damage (1·1 - 3·0 for geotextiles, 1·1 - 1·6 for geogrids), FCCR is the factor of safety for creep (1·0- 4·0 for geotextiles, 1·5- 3·5 for geogrids), FSCD is the factor of safety for chemical degradation (1 ·0- 2·0 for geotextiles,), and FS BD is the factor of safety for biological degradation (1·0- 1·3).
1.7. Application areas
Geosynthetics are versatile in use, adaptable to many field situations, and can be combined with several traditional and new building materials. They are utilized in a range of applications in many areas of civil engineering, especially geotechnical, transportation, hydraulic, and environmental engineering, in which geosynthetics are widely used for achieving technical benefits and/or economic benefits. The growth of geosynthetic applications is continuing at a rapid pace. Koerner (2000) reported the approximate growth in North America based on both quantity and sales of geosynthetics (Fig. 1.43). Estimated consumption
""'
""'
Table 1.7. A general range of some specific properties of commercially available geosynthetics (based on the information compiled by Lawson and Kempton (1995)) Type of geosynthetics
Geotexti Ies
Geogrids
Geomembranes
Geocomposites
Non-wovens Heat-bonded Needle-punched Resin-bonded Wovens Monofilament Multifilament Flat tape Knitteds Weft Warp Stitch-bonded Extruded Texti Ie-based Knitted Woven Bonded cross-laid strips Natural Reinforced (made from bitumen and non-woven geotextile) Plastomeric (made from plastomers, such as HDPE , LDPE , PP or PVC) Unreinforced Reinforced Elastomeric (made from elastomers , i.e . rubbers of various types) Reinforced Geosynthetic clay liners Linked structures (geostrip-based) *
C)
Tensile strength: kN/m
Extension at max . load: %
Apparent opening size : mm
Water flow rate (volume permeability) : 11m2I s
Mass per unit area: g/ m 2
(I)
0
III
-
'<
::l ::T
3-25 7-90 5-30
20-60 30-80 25-50
0,02-0,35 0,03-0'20 0,01-0,25
10-200 30-300 20-100
60-350 100-3000 130-800
0,07-4,0 0,05-0 ,90 0,10-0,30
80-2000 20-80 5-25
150-300 250-1500 90-250
0·20-2'0
60-2000 80-300 50-100 NA
150-300 250-1000 250-1000 200-1100
NA NA NA
150-1300 150-1100 400-800
0
0
1000-3000
50-200 15-30
0 0
0 0
400-3500 600-1200
15-30 10-30 3-15
0 0
0 0
NA
NA
500-1500 5000-8000 400-4500
20-80 40-1200 8-90
20-35 10-30 15-25
2-5 20-800 30-1000 10-200
300-600 12-30 10-30 20-30
20-400 20-250 30-200
3-20 3-20 3-15
20-60
30-60
10-50 30-60
30-60 10-20 100-1500
OAO-1 '5
0,07-0,50 15-150 20-50 20-50 50-150
*Geostrips are geocomposites having tensile strength in the range 20-200 kN and extension at max. load in the range 3-15%. Geobars are geocomposites having tensile strength in the range 20-1000 kN , if reinforced internally and in the range 20-300 kN if reinforced externally , and extension at max. load in the range 3-15% for both cases .
!2.
(:;'
III
III ::l
-... Q.
::T (I)
III
'tJ 'tJ (:;'
III
0'
::l
III
Fundamentals of geosynthetics
45
700
600
~
500
Q)
0
•c •
.. A
Q)
E Q)
0;
Geotextiles Geomembranes Geocomposites Geonets Geogrids Geosynthetic clay liners
400
::J
rr
III
'0 300
III
c:
.2 ~
200
100
0 1970
1974
1978
1982
1986
1990
1994
1998
Year (a)
1200 0
1000
•c •
800 III
A
..
Geotextiles Geomembranes Geocomposites Geonets Geogrids Geosynthetic clay liners
~
0
"0
600
0 III
c:
~ ~
400
200
Fig. 1.43. Growth of geosynthetics in North America based on quantity and sales (after Koerner, 2000)
0 1970
1974
1978
1982
1986
1990
1994
1998
Year (b)
of geotextiles, geomembranes and geogrids in Western Europe is shown in Fig. l.44, as reported by Lawson and Kempton (1995). The rapid growth in the past three decades all over the world is due mainly to the following favourable basic characteristics of geosynthetics: • • • • • • •
non-corrosiveness highly inert to biological and chemical degradation long-term durability under soil cover high flexibility minimum volume lightness robustness (geosynthetics can withstand the stresses that may be induced during installation and throughout the life of the structure)
46
Geosynthetics and their applications
400 . - - --
~
350
r
300
I-
250
r
,,
I
Geotextiles
o
200
:::>
'oc:"
150
I-
(J
100 /
Fig. 1.44. Estimated consumption of geosynthetics in Western Europe (after Lawson and Kempton, 1995)
,,
I
E
/
/
/
/
,
,."
Geogrids
o t::-::-:....-_....L...~...............
50
I
"y/'
c:
'E.
,
I
~
E~
,
- - - - - - - - - - - - - - .,.
....~..
" ........ Geomembranes
-\:!:..::--:..:."1:1.:'-:':-;';;-";"",,'-",,'t=':::::::::::::::.J
1970
1975
1980
1985
1990
1995
Year
• factory-produced to have specific quality controlled standards and they do not ex hibit the inherent variabi lity of naturally occurring materials • ease of storing and transportation • simplicity of installation, even by unskilled personnel • ease in control of execution • rapid installation, even in adverse environmental conditions and thus speeding up the construction process • useab le, even with unsuitable soils • replace soil/mineral construction materials - conserving scarce resources • cause less wear and tear on equipment • available in a wide range of products, in numerous configurations and weights, to perform a wide range of functions when placed in soils • have capacity to solve even those problems which cannot be solved by traditional techniques • make an economical and environmentally friend ly solution (the cost should be estimated to include the initial construction cost, continuing maintenance cost, cost related to production losses, in the case of roads as a result of their closure, etc.) • improved performance of structure • provide good aesthetic look to structures. The technical acceptance of geosynthetics has been achieved in a large number of application areas. Table 1.8 lists the major application areas for the geosynthetics that have been described, in detail, in latter chapters of this book. These applications are illustrated schematically in Fig. 1.45. It rarely happens, in any field application, that a geosynthetic perform's only one function. However, for simplicity, the most important function is identified as per site conditions and the same is considered in the design. Out of the reinforcement and the separation functions, the selection of a major function is governed by the ratio of the applied stress on the soft soil to its shear strength. This aspect is discussed further in Section 5.2.
1.8. Standards
The application of geosynthetics is an area where improvement in the basic concepts and the developments related to raw materials, manufacturing processes, and methods of analysis, design and construction has
Table 1.8. Major application areas for geosynthetics SI. no.
Application areas
Main purpose of geosyntheti cs
Major functions
Major geosynthetic products
Most important properties
Special consideration
Reinforce and protect backfill /soil
Reinforcement
Improve stability; provide drainage
Reinforcement Separation Drainage
3
Shallow foundations
Increase load-bearing capacity and reduce settlement
Reinforcement Separation
Strength Soil-geosynthetic friction Strength Soil-geosynthetic friction Pore size Permeabi I ity Strength Soil-geosynthetic friction Pore size
4
Unpaved roads
Increase bearing capacity and reduce degree of rutti ng
Reinforcement Separation
5
Paved roads
Inhibit crack propagation , improve cyclic fatigue behaviour
Separation Drainage
6
Railway tracks
Prevent ballast contamination ; distribute load on subgrade
Separation Filtration Drainage
Geotextiles Geogrids Geotextiles Geogrids Geocells Geocomposites Geotextiles Geogrids Geocells Geocomposites Geotextiles Geogrids Geocomposites Geotextiles Geogrids Geocomposites Geotextiles Geogrids Geocomposites
Creep
2
Retaining walls and steep-sided embankments Embankments on soft ground
7
Slopes
Protect soil slope against erosion ; reinforce soil ; provide drainage
Filtration Drainage Reinforcement
Geotexti les Geogrids Geocomposites
Pore size Permeability Strength Soil-geosynthetic friction Abrasion resistance Pore size Permeability Strength Abrasion resistance Pore size Permeability Abrasion resistance
8
9
10 11
Landfills
Dams
Containment ponds, reservoirs and canals Pipeline and drainage facilities
Extract leachate out of the waste and retain the same
Reduce seepage through the dam embankment; prevent internal erosion / piping ; provide drainage ; protect slope against erosion
Fluid barrier Drainage Filtration Reinforcement Fluid barrier Filtration Drainage
Reduce seepage of water/ liquid into ground
Fluid barrier
Protect the drainage medium ; provide drainage
Drainage Filtration
Geomembranes Geotextiles Geogrids Geocomposites Geomembranes Geotextiles Geonets Geocomposites Geogrids Geomembranes Geocomposites Geonets Geotextiles Geocomposites
Strength Soi I-geosynthetic friction Pore size Pore size Permeability Abrasion resistance Pore size Permeability
Creep/ stress relaxation
Elongation
Repeated loading Elongation Repeated loading Elongation Repeated loading Elongation Resistance to impact and wear abrasions Rapid changes in water level Clogging Construction stresses
."
c::
:::I
Co
III
Leachate characteristics Construction stresses Elongation Clogging Construction stresses
3
(I)
:::I III
Ui
a CO (I)
0
1/1
'<
:::I
Permeabi I ity Abrasion resistance Pore size Transm issivity Permittivity
Construction stresses
:T (I)
0'
1/1
Clogging
...."'"
48
Geosynthetics and their applications
Wraparound
. :;:~':.:.::..:.::.::::::.::>.: : :.: : : )+:<:J) Fill
/ /-.;.~ ·· 7:·.: .>.::.: ~:·:~.·.~ :131
/
Geotextile
(~zf:i?'Es:- /.i •. • ~t+s~:~~"
Selected backfill
7l7117ijmr/7l);;)ii;jj/l/l/»ifl'jj7/i;rj}';17717
Geotextile
(b)
(a)
=11
$fICJCJCD)
II
II
\9P
Fill
Pier
Soft foundation soil
Geotextile
•• • • •• • •• • • • • • •
Vertical drains (e)
(d) Geotextile/Geogrid
0 0
0
.D 0
•
a
"
D
0
0
0
000 c 000 0000000
o 0 ()onn(:)l")oO""
Geotextile (e)
(f)
. '.
Geotextile
.
(g)
"
~ ': '-:-.. ,',
',: .
(h)
Fig . 1.45. Typical geosynthetic applications: (a) retaining wall ; (b) steep-sided embankment; (c) embankment on soft foundation soil ; (d) bridge pier; (e) unpaved road ; (f) paved road; (g) railway track; (h) slope - erosion control ; (i) slope - stabilization ; (j) landfill ; (k) dam ; (I) liquid reservoir; (m) water channel ; and (n) trench drain
Fundamentals of geosynthetics
49
Geomembrane
Contained waste
+ Geotextile Anchored spider netting
Porous pipes
Failure plane
(j)
(i) Upstream support fill
(k)
Geomembrane
Liquid reservoir
Leakage detection drain
(I)
i-':-+-- Geotextile
~y.;o!):?-I.,l>;>- Aggregate
Geocomposite (m)
-t"""f--- Porous pipe
(n)
Fig. 1.45. continued
been continuing for the last three decades. Preparation of new standards and revision of existing standards have also been continuing. In developed countries a large number of standards has been prepared in different aspects of geosynthetics and some standards are still under preparation/ revision. The developing countries are also trying to have their own standards, so that geosynthetics can be used on a scientific basis for economical solutions to civil engineering problems. Table 1.9 provides a list of some of the standards on terminology and testing of geosynthetics from the USA, UK, India, and Switzerland.
50
Geosynthetics and their applications
Table 1.9. Some standards on testing of geosynthetics (in 2001) published by the American Society of Testing Materials (ASTM) . the British Standards Institution (BSI) . the Bureau of Indian Standards (BIS) and the International Organization for Standardization (ISO). Switzerland Topics
Designations of relevant standards ASTM
Terminology/vocabulary Sampling and preparation of test specimens Determination of mass per unit area Determination of thickness Determination of compression behaviour Determination of tensile propertieswide-width strip tensile test Determination of joint/seam strength Determination of trapezoid tearing strength Determination of puncture resistanceCBR test Determination of puncture resistancefalling cone method Determination of interface frictiondirect shear method Determination of apparent opening size Determination of water permeabilitypermittivity without load Determination of permittivity under load Determination of the (in-plane) flow rate per unit width and hydraulic transmissivity Determination of hydraulic conductivity ratio (HCR) of soil/geotextile system Determination of durability/resistance to weathering Determination of the creep properties Determination of abrasion resistance (sand paper/ sliding block method) Evaluation of sOil-geotextile system clogging potential by the gradient ratio Evaluation of biological clogging of geotextile or soil/geotextile filters Determination of resistance to the exposure of UV light and water (xenon-arc type apparatus) Evaluation of the chemical resistance Determination of microbiological resistancesoil burial test
D4439-87 D4354-99 D5261-92 (1996). D5993-99 D5199-01 . D6525-00 D6364-99. D6244-98 D4595-86 (1994) . D4885-88 (1995) D4884-96 D4533-91 (1996) D6241 -99
BS
IS
ISO
13321-1: 1992
10318: 1990 9862: 1990 9864: 1990
EN ISO 9863-2: 1996. EN 964-1 : 1995
13162-3: 1992
9863: 1990. 9863-2: 1996
EN ISO 10319: 1993
13325: 1992
10319: 1993
EN 963: 1995 EN 965: 1995
10321: 1992
EN ISO 10321 : 1996 14293: 1995 EN ISO 12236: 1996
12236: 1996
EN 918: 1996
13162-4: 1992
D5321-92 (1997)
6906-8: 1991
13326-1: 1992
D4751-99a D4491-99a
6906-2: 1989 6906-3: 1989. EN ISO 11058: 1999
14324: 1995
D5493-93 (1998) D4716-00
11058: 1999
12958: 1999
6906-7 : 1990. EN ISO 12958: 1999
D5567-94 (1999) D5819-99 D5262-97 D4886-88 (1995)e1
EN 12224: 2000, EN 12226: 2000 6906-5: 1991, EN ISO 13431: 1999 EN ISO 13427: 1998
13431 : 1999 13427: 1998
D5101-99 D1987-95 D4355-99
13162-2: 1991
D6389-99, D6213-97 , D6388-99, D5747-95a BS EN 12225: 2000
When dealing with the solution of a civil engineering problem using geosynthetics, the standards listed in this table, or other applicable standards, may be consulted.
1.9. Concluding remarks
During the past three decades, considerable work was devoted to developing geosynthetic manufacturing processes, laboratory and field-testing techniques, standards, and analysis, design and construction methods to provide safe, economical and practical solutions to civil engineering problems using geosynthetics. However, many aspects related to geosynthe tics are still not very clear or standardized. The range of tests presently available is limited, and results of many tests using even standard procedures have been found to be widely variable. In fact , the field of
Fundamentals of geosynthetics
51
geosynthetics is still young and it is hoped that future developments will continue. The design engineers should be careful when they use methods justified by common sense or imported from other disciplines of civil engineering. They should always try solutions using methods based on rational analyses or based on past experiences. The users of geosynthetics must be in constant touch with future developments, as it will be to their benefit. The basic information provided in this chapter will be useful for understanding the application-oriented concepts of geosynthetics that are discussed in detail in later chapters of this book. Many manufacturers and marketing groups designate geosynthetics with trade names, and they all have different properties that are not included in this chapter. However, users of geosynthetics may obtain a wealth of information directly from the manufacturers.
References Agerschou, H. A. (1961). Synthetic material filters in coastal protection. Journal of Waterways and Harbours Division, ASCE, 87, 111 - 124. Andrawes, K. Z., McGown, A. and Murray, R . T. (1986). The load - straintime- temperature behaviour of geotextiles and geogrids. Proceedings ol the 3rd International Conference on Geotextiles, Vienna, Austria, pp. 707- 712. Barratt, R . J. (1966). Use of plastic filte rs in coasta l structures. Proceedings of the lath International Conference on Coastal Engineering, Tokyo, pp. 1048- 1067. Bhatia, S. K. , Smith, J. L. and Christopher, B. R. (1994). Interrelationship between pore openings of geotextiles and methods of evaluation. Proceedings of the 5th International Conference on Geotextiles, Geomembranes and Related Products, Singapore, pp. 705- 7\ O. Bonaparte, R. and Berg, R. (1987). Long term allowable tension for geosynthetic reinforement. Proceedings of the Geosynthetics '87, New Orleans, Louisiana. Carroll, R. G. , Jr. (1983). Geotextile filter criteria. Transportation Research R ecord, 916, 46- 53. Carroll, R. G., Jr. (1988). Spec(lying geogrids. IFAI, St Paul, USA. Geotechnical Fabrics Report. Cassidy, P. E., Mores, M. , Kerwick, D . J ., Koeck, D. J ., Verschoor, K . L. and White, D. F. (1992). Chemical resistance of geosynthetic materials. Geotextiles and Geomembranes, 11, 61 - 98. Chang, D. T. T. , Chen, C. A . and Fu, Y. C. (1996). The creep behaviour of geotextiles under confined and unconfined conditions. Proceedings of the International Symposium on Earth Reinforcement, Fukuoka, Japan , pp. 19- 24. Christopher, B. R. and Holtz, R. D. (1985). Geotextile Engineering Manual. US Federal Highway Administration, Washington , DC. Report No. FHWA-TS-86j 203. den Hoedt, G. (1986). Creep and relaxation of geotextiles fabrics. Geotextiles and Geomembranes, 4, No.2, 83- 92. EI-Fermaoui, A. and Nowatzki, E. (1982) . Effect of confining pressure on performance of geotexti les in soils. Proceedings of the 2nd In ternational Conference on Geotextiles, Las Vegas, pp. 799- 804. Fukuoka, M. (1990). Earth reinforcement for foundations - east and west. Geotextiles and Geomembranes, 9, 3- 9. Giroud , J. P . (1980). Introduction to geotextiles and their applications. Proceedings of the 1st Canadian Symposium on Geotextiles, pp. 3- 31.
52
Geosynthetics and their applications
Giroud, J. P. (1982). Filter criteria for geotextiles. Proceedings of the 2nd International Conference on Geotextiles, Las Vegas, pp. 103- 108. Giroud , J. P. (1994). Quantification of geosynthetic behaviour. Proceedings of the 5th International Conference on Geotextiles, Geomembranes and Related Products, Singapore, pp. 1249- 1273. Giroud , J. P. and Carroll, R. G . (1983). Geotextile products. Geotechnical Fabrics Report, IFAI, St Paul, USA, pp. 12- 15. Giroud , J. P. and Frobel, R. K. (1983). Geomembrane products. Geotechnical Fabrics Report, IFAI, St Paul, USA, pp. 38- 42. Giroud , J. P., Gourc J. P. , Bally P. and Delmas, P. (1977). Behaviour of a nonwoven geotextile in an earth dam. Proceedings of the International Conference on the Use of Fabrics in Geotechnics, Paris, pp. 213 - 218. (In French.) Gray, C. G . (1982). Abrasion resistance of geotextiles fabrics. Proceedings of the 2nd International Conference on Geotextiles, Las Vegas, pp. 817- 821. Greenwood, J . H . and Myles, B. (1986). Creep and stress relaxation of geotextiles. Proceedings of the 3rd International Conference on Geotextiles, Vienna, Austria, pp. 821 - 826. Greenwood , J. H. , Trubiroha, P. , Schroder, H. F., Frank, P. and Hufenus, R. (1996). Durability standards for geosynthetics: the tests for weathering and biological resistance. Proccedings of the ist European Geosynthetics Conference. Eurogeo I, Maastricht, Netherlands, pp. 637- 641. Haliburton, T. A. and Wood, P. D. (1982). Evaluation of the US Army Corps of Engineers' gradient ratio test for geotexti le performance. Proceedings of the 2nd international Conference on Geotextiles, Las Vegas, pp. 97- 101. Hausmann, M . R . (1990). Engineering principles of ground modification. McGraw-Hill Publishing Company, Singapore. Healy, K. A. and Long, R. P. (1971). Prefabricated subsurface drains. Highway Research Record, 360. Hoare, D. J. (1982). Synthetic fabrics as soil filters. Journal of Geotechnical Engineering Division, ASCE, 108, No. GTlO, 1230- 1245. Holtz, R. D. (1975). Recent developments in reinforced earth. Proceedings of the 7th Scandinavian Geotechnical Meeting , Polyteknish Forlag, Denmark, pp. 281 291. Ingold , T. S. and Miller, K. S. (1988). Geotextiles handbook. Thomas Telford Publishing, London, UK. Jewell , R. A. (1986). Material properties for design of geotextile reinforced slopes. Geotextiles and Geomembranes, 2. John , N. W. M. (1987). Geotextiles. Blackie, London, UK. Kabir, M . H . and Ahmed , K . (1994) . Dynamic creep behaviour of geosynthetics. Proceedings of the 5th International Conference on Geotextiles, Geomembranes and Related Products, Singapore, pp. 1139- 1144. Kaswell, E. R. (1963). Handbook of industrial textiles. Industrial Fabrics Division, West Point Pepperell, New York, USA. Koerner, R. M . (1990). Designing with Geosynthetics, 2nd edition. Prentice H all, Englewood Cliffs, New Jersey, USA. Koerner, R. M. (2000). Emerging and future developments of selected geosynthetic applications. Journal of Geotechnical and Geoenvironmental Engineering, ASCE, No . 4, 293- 306. Koerner, R. M . and Welsh , J. P. (1980). Construction and geotechnical engineering using synthetic fabrics. John Wiley, New York, USA.
Fundamentals of geosynthetics
53
Lawson, C. R . (1982). Filter criteria for geotextiles: relevance and use. Journal of Geotechnical Engineering Division, ASCE, 108, No. GTlO, 1300- 1317 . Lawson, C. R . (1986). Geotextile filter criteria for tropical residual soils. Proceedings of the 3rd International Conference on Geotextiles, Vienna, pp. 557- 562. Lawson, C. R . and Kempton, G. T. (1995). Geosynthetics and their use in reinforced soil. Terram Ltd, UK. McGown, A., Andrawes, K. Z. and Kabir, M. H. (1982). Load-extension testing of geotextiles confined in soil. Proceedings of the 2nd International Conference on Geotextiles, Las Vegas, pp. 793- 796. Mercer, F. B. (1982). Retaining fill in a geotechnical structure. Brit. Pat. No. 2,078,833A, January 1982. Mikki, H ., Hayashi, Y., Yamada, K. , Takasago and Shido, H. (1990). Plane strain tensile strength and creep of spun-bonded nonwovens. Proceedings of the 4th International Conference on Geotextiles, Geomembranes and R elated Products, The Hague , pp. 667- 672. Mikki, H., Hayashi, Y. and Sato, M. (1994). Accelerated filtration test of geotextile filters. Proceedings of the 5th International Conference on Geotextiles, Geomembranes and R elated Products, Singapore, pp. 643 - 646. Murray, R . T., McGown, A., Andrawes, K. Z. and Swan, D. (1986). Testing joints in geotextiles and geogrids. Proceedings of the 3rd International Conference on Geotextiles, Vienna, Austria, pp. 731 - 736. Myles, B. and Carswell, 1. G. (1986). Tensile testing of geotextiles. Proceedings of the 3rd International Conference on Geotextiles, Vienna, Austria, pp . 713 - 718. Paulson, J. N . (1987). Geosynthetic material and physical properties relevant to soil reinforcement applications. Geotextiles and Geomembranes, 6, 211 - 223. Perkins, S. W. (2000). Constitutive modeling of geosynthetics. Geotextiles and Geomembranes, 18, 273- 292. Puig, J. , Bilvet, J. C. and Pasquet, P. (1977). Earth fill reinforced with synthetic fabric. Proceedings of the International Conference on the Use of Fabrics in Geotechnics, Paris, pp . 85- 90. (In French.) Rankilor, P. R. (1981). Membranes in ground engineering. John Wiley, Chichester, UK. Rankilor, P. R. and Heiremans, F. (1996). Properties of sewn and adhesive bonded joints between geosynthetic sheets. Proceedings of the 1st European Geosynthetics Conference. Eurogeo 1, Maastricht, Netherlands, pp. 261 - 269. Shamsher, F. H. (1992). Ground improvement with oriented geotextiles and randomly distributed geogrid micro-mesh. PhD thesis submitted to Indian Institute of Technology Delhi, New Delhi, India. Simon, A., Payany, M. and Puig, J. (1982). The use of honeycombed geotextile lap to combat erosion. Proceedings of the 2nd International Conference on Geotextiles, Las Vegas, pp . 247- 252. (In French.) Snow, M. S., Kavazanjian Jr. , E. and Sanglerat, T. R. (1994). Geosynthetic composite liner system for steep canyon landfill side slopes. Proceedings of the 5th International Conference on Geotextiles, Geomembranes and Related Products, Singapore, 1994. Terzaghi, K. and Lacroix, Y. (1964). Misson dam: an earth and rockfill dam on a highly compressible foundation. Geotechnique, 14, No. I , 13- 50. Terzaghi, K. and Peck, R. B. (1948). Soil mechanics in engineering practice. John Wiley & Sons, New York, USA.
54
Geosynthetics and their applications
Van Dine, D ., Raymond , G. and Williams, S. E. (1982). An evaluation of abrasion tests for geotextiles. Proceedings of the 2nd International Conference on Geotextiles, Las Vegas, pp. 811 - 816. Venkatappa Rao, G . (1996). Geosynthetics in the Indian environment. Indian Geotechnical Journal, 26, No. I, 3- 94. Venkatappa Rao, G . and Saxena, K. R. (eds) (1989). Use of geosynthetics in India: experiences and potential. CBIP, New Delhi. Pub. No. 209. Wang, D. W. (1994). Filter criteria of woven geotextiles for protective works. Proceedings of the 5th International Conference on Geotextiles, Geomembranes and Related Products, Singapore, pp. 763 - 766. Westergaard , H . M . (1938). A problem of elasticity suggested by a problem in soil mechanics, soft material reinforced by numerous strong horizontal sheets. Harvard University. Wewerka, M. (1982). Practical experience in the use of geotextiles. Proceedings of the Symposium on Recent Developments in Ground Engineering Techniques, Bangkok, pp. 167- 175.
2
Soil-geosynthetic interaction M. L.
LO PES
Departm en t of Civil Engin eering, University of Po rto, Po r tugal
2.1 . Introduction
Soil- geosynthetic interaction is of the utmost importance in several applications of geosynthetics, especially when they act as a soil reinforcement. Soil reinforcement consists of the placement of elements duly oriented in the soil, which, by their character, improve the mechanical properties of the new material (reinforced soil) when compared with that of the unreinforced soil. The main target of reinforcement is to inhibit the development of tensile strains in the soil and, consequently, to support the tensile stresses that the soil cannot withstand. The tensile stress supported by reinforcement improves the soil mechanical properties by reducing the shear stress that has to be carried by the soil and by increasing its available shearing resistance, as the normal stress acting on potential shear surfaces increases. The effectiveness of reinforcement depends on its alignment, it being most effective when aligned in a direction of tensile strain in the soil, so that tensile reinforcement stress develops (McGown et at., 1978; Jewell and Wroth, 1987; Jewell, 1996). The behaviour of the reinforced soil depends on several factors , such as: • • • • •
soil and reinforcement mechanical characteristics soil-reinforcement interaction mechanism and properties geometry of the reinforced system shape, number, location and alignment of reinforcements process of construction, etc.
As examples, the shear strength of the reinforced soil relies on the mobilized shear resistance in the soil and the mobilized tensile stress in the reinforcement, the relative values of these mobilized resistances are dependent on the deformation properties of the soil and reinforcement. On the other hand, the rate of mobilization of reinforcement tensile stress is determined by its stress- strain properties and, to prevent failure, the maximum mobilized tensile stress cannot exceed the reinforcement bond stress. Although all the mentioned determinant factors on soil-reinforced behaviour are important, special evidence must be given to soil-reinforcement interaction mechanisms, and to the factors that can influence them, because the effectiveness of the transference of tensile stresses from soi l to reinforcements rely on it and, so too, the behaviour of the reinforced soil system. This subject will be analysed in the following sections by considering geosynthetic materials as a reinforcement only, complemented by the study of the methods for evaluation of soil-reinforcement interaction properties .
56
Geosynthetics and their applications
2.2. Granular soil behaviour
The granular soils are widely considered in the studies of soilgeosynthetic interaction because fill materials are, usually, of that type, and their characteristics are determinant on the effectiveness of soilgeosynthetic interaction . Strength and stiffness of granular soils are extremely dependent on density. Dense soils show greater stiffness and resistance than loose soils because of greater grain interlocking. During soil shearing, when intergrain sliding starts, the mobilized forces (due to the rearrangement of grains) can be high if the soil is dense; on the other hand, inter-granular friction forces are almost independent of soil density. When the shear process starts, the void ratio of dense soils is lower than the critical value, and shear stresses induce volume increase. For small deformations, the stressstrain curve of dense soils shows a peak (maximum strength) that depends on volume increase and initial density. For large deformations, when grain interlocking is cancelled, the soil void ratio is equal to the critical value and the soil strength is constant (soil strength at a constant volume). At the beginning of shearing of loose soils, the void ratio is greater than the critical value and the shear stresses induce volume reduction. The soil stress- strain curve does not show a peak, maximum soil strength is equal to soil strength at a constant volume of dense soils and is mobilized at large deformations, when the void ratio of the soil equals its critical value. Besides the density, other factors can influence the behaviour of granular soils, such as confinement stress, grain shape and size di stribution. An increase in the confinement stress leads to a reduction in the soil critical void ratio, implying a decrease in the soil dilatant behaviour and an approach of its peak and constant volume strengths. The grain shape and size distribution can affect the soil density as denser or looser arrangements of particles are determined by them. Although grain size is not very important in relation to the behaviour of granular soils, it is of the utmost relevance in soil- geosynthetic interaction mechanism, especially when the geosynthetic is a geogrid. It must be emphasized that the characteristics that influence granular soil behaviour on reinforced soil do not change; however, its strength is improved by the presence of reinforcement (Fig. 2.1), especially when it /"'-j no =39·3%
" / ,
11
I
10
I
:'
\
\
/
:
I
! ,i
8
\
'- , ,no = 38·9%
\
\\
M
!2
': /
7
/'
: /, no = 38·6%
5
no = 43·8%
4 3
2
- \-- -..
I
:'
9
8
\\
,/
/
9
: :,
/
12
no = initial porosity
234 5 Axial strain : % (a)
\,
--,_
'----
: ______ Reinforced sand with metallic inclusions _ - . _ Reinforced sand with geotextiles _
Unreinforced sand
2
6
7
8
2345678 Axial strain: % (b)
Fig . 2.1. Stress-strain curves of a reinforced sand: (a) loose; and (b) dense (replotted from McGown et aI. , 1978)
Soil-geosynthetic interaction
57
is aligned in a direction of tensile strain in the soil, so that tensile reinforcement stress develops (McGown et at., 1978) .
2.3. Soilgeosynthetic interaction mechanisms
Although factors such as the geometry of a reinforced soil system and the process of its construction can influence the soil-reinforcement interaction properties, they are strongly determined by the interaction mechanism, the physical and mechanical properties of soil (density, grain shape and size, grain size distribution, water content, etc.), and the mechanical properties, shape and geometry of reinforcements. Three mechanisms of interaction can be identified in reinforced systems: • skin friction along the reinforcement • soil-soil friction • passive thrust on the bearing members of the reinforcement. Skin friction is the only mechanism with geotextiles and strips. In the case of geogrids, the passive thrust on the bearing members of the grids must also be considered as soil- soil friction, if relative movement occurs in the soil along the grids' apertures. Shear strength mobilization between granular soils and geotextiles is a two-dimensional phenomenon, where soil dilatance is allowed, strongly affected by the extensibility of geotextiles. In the case of strips, the phenomenon is three-dimensional and greatly dependent on the characteristics of soil dilatance and on the roughness of the reinforcement surfaces. In fact , the volume of soil shearing around the reinforcement is influenced by its geometry and roughness. With regard to geogrids, the phenomenon can also be considered three-dimensional, mobilizing skin friction for small displacements and progressively mobilizing the passive thrust on the bearing members of the grid as displacement increases. Figure 2.2 shows the stress distribution in the cases of free soil dilatance (two-dimensional phenomenon) and restricted soil dilatance (threedimensional phenomenon). Since geogrids are less extensible than geotextiles, the improvement in soil strength and the mobilization of shear resistance along the interface with the soil increase when the reinforcement used is a geogrid .
[
h {1n=Yx
h
Reinforceme~:d 1: - - - - - - - - {1n
::r ~
!l1I81111 / ~
~
Reinforcement /
T
Stress conditions during shear
Initial conditions (a)
Fig. 2.2. Stress conditions in reinforced soil: (a) free dilatance; and (b) restricted dilatance (replotted from Hayashi et aI., 1994)
Initial conditions
Stress conditions during shear
(b)
-P
58
Geosynthetics and their applications
I I
Fig. 2.3. Force distribution along the reinforcement (replotted from Jewell et al. ,1984)
2.4. Soilgeosynthetic interface resistance
Axial force
~II P,
Bond governs the rate of change of axial force
The stability of reinforced soil is strongly related to the effectiveness of stress transference from soil to reinforcement, which is dependent on the available reinforcement length to shear. In fact, as shown in Fig. 2.3 for reinforced slopes, the reinforcement length beyond the failure line must be enough to mobilize the required shear stresses to balance the maximum tensile force of reinforcement. The ratio of stress mobilization is affected by the resistance of the soil-reinforcement interface. As stated above, with geotextiles and strips only the skin friction mechanism contributes to soil- geosynthetic interface resistance, however, with geogrids two other interaction mechanisms must be added: the passive thrust mobilization on the bearing members of the grid, and the soil- soil friction , in case of relative displacement in the soil along the grid apertures. Figure 2.4 shows the soil- geogrid interaction mechanisms. Two relative movements can be responsible for the mobilization of strength in soil-reinforcement interfaces: (a)
a block of soil slides across one side of the reinforcement that is 'linked' on the other side to the other block of soil (direct sliding) (b) the reinforcement moves in relation to the surrounding soil (pullout). In the first case, when the shear strength of the soil-reinforcement interface is exceeded , the failure occurs by direct shear, and, in the second case, by pullout. In each case, the soil-reinforcement interface coefficient, f, has a different definition as will be seen below. Soil-reinforcement interface shear strength can be defined as: T = 2WL(J~f tan 1>'
(2.1)
with 0 < f < 1, f being the soil-reinforcement interface coefficient, 1>' the soil friction angle in terms of effective stresses (peak or at constant
(a)
Fig . 2.4. Soil-geogrid interaction mechanisms: (a) shear between soil and plane surfaces; and (b) soil bearing on reinforcement surfaces (replotted from Jewell et aI. , 1984)
(b)
Soil-geosynthetic interaction
59
B
it
movement
Fig. 2.5. Reinforcement dimensions (replotted from Jewell, 1996)
as = Fraction of grid surface area that is solid ab = Fraction of grid width
W available for bearing
volume, depending on the soil density), <J:, the effective normal stress in the interface, and Wand L, the width and the length of the reinforcement, respectively (Fig. 2.5). Equation (2.1) is for general application and the main problem lies with the definition off In fact, f depends on the interaction mechanism mobilized on the soil-reinforcement interface and on the relative movement that occurs on the same interface. So, if the mechanism mobilized is only skin friction , as with geotexti\es, f is very similar, if not identical, for direct sliding and pullout movements (Jewell, 1996) and is: tanD f =fds =fb = - tan ¢'
(2 .2)
where Dis the friction angle in soil-reinforcement interface, andfds andfb are the interface coefficients of direct sliding and of bond , corresponding, respectively, to the direct sliding and to the pullout movements in the soilreinforcement interface (Jewell, 1996). In the case of geogrids, the shear strength of the soil-reinforcement interface for direct sliding movement is the sum of two items, corresponding: (a) (b)
to the skin friction mechanism (Ts) to the soil-soil friction mechanism (T s/s).
The contribution of the passive thrust mobilization on the bearing members of the geogrid mechanism is almost negligible in the case of direct sliding.
T = T s + T s/s
(2.3)
with
T s = 2a s WL<J~ tanD
(2.4)
and
T s/s = 2(1 - as) WL<J:, tan ¢'
(2.5)
where as is the fraction of the geogrid surface area that is solid (Fig. 2.5). Using equations (2 .1), (2 .3), (2.4) and (2.5), the interface coefficient of direct sliding is obtained as:
(2.6) In the case of pullout movement, the contribution of a soil- soil friction mechanism on soil-reinforcement interface resistance is almost nil - that
60
Geosynthetics and their applications
resistance results from the contribution of the skin friction mechanism (T s) and from the passive thrust mobilization on the bearing members of the geogrid mechanism (T p). (2.7) where T s is expressed by equation (2.4) and T pis: Tp =
(t) ab
W
Ber~
(2.8 )
ab
S, Band are, respectively, the distance between bearing members, the thickness 'o f the geogrid bearing members, and the fraction of the width of the geogrid available for bearing (Fig. 2.5), and er~ is the effective passive stress mobilized. Using equations (2.1), (2.4), (2.7) and (2.8), the interface coefficient (coefficient of bond) is obtained as:
8) + (er~) (abB) ( er~ S
tan J = Jb = as ( tan q/
1 ) 2 tan ¢'
(2.9)
If as = 1 and ab = 0, equations (2.2) and (2.9) become equal, representing the coefficient of bond of a reinforcement where the only interaction mechanism mobilized is skin friction , as in geotextiles. If, in the soil-reinforcement interface, direct sliding and pullout movements occur, the interface coefficient that needs to be considered will be the minimum between the coefficients of direct sliding and of bond. However, the movement that starts first must be considered , especially when the soil is dense. When the soil is loose, the equations deduced above may be applicable, considering the soil strength at constant vo lume. When the soil is dense, and the first movement that occurs is direct sliding, the soil peak strength can be considered to define the coefficient of direct sliding, but on the deduction of the coefficient of bond, the strength at constant volume must be considered on the side where direct sliding occurred as well as the peak strength on the other side. When pullout is the first movement that occurs, soil peak strength can be considered in the definition of the coefficient of bond, but only the strength at constant volume in the definition of the coefficient of direct sliding. Excluding the relation er~ / er:l ' the other parameters in equation (2.9) are easy to obtain. According to Jewell et al. (1984) and Jewell (1990; 1996), the passive resistance mobilized on the bearing members of the grids is limited by the theoretical values shown in Fig. 2.6 (equations (2.13) and (2.15)). The theoretical values are defined by the general theory of bearing capacity, with the bearing members of the grids considered similar to strip footings turned 90°. The passive resistance mobilized is:
erIp
= c'Nc + ern'Nq
(2.10)
where er~ is the effective passive resistance, er~ is the effective normal stress on the interface, c' is soil cohesion, and Nc and N q are the bearing capacity factors similar to those used for footings considering the failure mechanism by bearing capacity (Peterson and Anderson, 1980). Nc and
Soil-geosynthetic interaction
61
1000
(J"t/(J"~ = N q Equation (2.13)
Equation (2.15)
Fig . 2.6. Bearing stresses on geogrids (replotted from Jewell, 1996; data from Jewell et aI. , 1984 and Palmeira and Milligan , 1989)
1
2~ 0-----3~0----~4~ 0 -----5 ~0~--~ 60
Angle of friction , $': 0
N q are given by: N = tan 2 (~+
4
q
¢2I) e
7r tan q,'
.-( ----,l,...:-) N = . .:. N--"q_ e tan ¢'
(2.11 ) (2.12)
where ¢' is the friction angle of the soil in terms 'o f effective stresses. In soils without cohesion, the passive resistance mobilized on the bearing members of the grid has an upper bound as:
O'~ = tan2 (~+ ¢ 1)e7r tan q,1 O'~
4
(2 .13)
2
For the lower bound of the passive resistance mobilized on the bearing members of the grids, Jewell et at. (1984) suggest the adoption of the shear failure mechanism by puncture on deep foundations. The bearing capacity factor , N e , is given by equation (2.12) and the bearing capacity factor, N q , by: N q
=
tan
(~ + ¢') e((7r/2)H1)tanq,1 4
2
(2.14)
In soils without cohesion, the passive resistance mobilized on the bearing members of the grid has a lower bound, as:
O'~ O'~
= tan
(~+ 4
¢2I)
e((7r/2)+q,') tan q,'
(2.15)
which is recommended for design. Ospina (1988) observed that for dry sand under small confinement stresses failure followed the mechanism underlying the lower bound , and for high confinement stresses it followed the mechanism underlying the upper bound . Palmeira and Milligan (1989) have shown, based on data from pullout tests performed with metallic grids and Leighton Buzzard sands, that when the ratio B / Dso (B is the thickness of the bearing members of the grid and Dso is the mean soil particle size) is lower than 10, the failure mechanism changes from that underlying the lower bound to that underlying the upper bound.
62
Geosynthetics and their applications
The reinforcement geometry and the soil particle size affect the passive resistance on the bearing members of grids, which can, partially, justify the scatter in the published data. In the following sections the influence of these factors will be analysed.
2.5. Factors influencing soil-geosynthetic interaction
2.5.1. Soil particle size
The soil particle size has an important role in soil- geosynthetic interaction, especially when the geosynthetic is a geogrid . By studying the influence of soil particle size on soil- geogrid interaction by direct sliding movement, Jewell et at. (1984) concluded that the coefficient of direct sliding increases with soil particle size and has its maximum value when the grain size is similar to that of the geogrid apertures. In fact , when the soil has particles of the size of silt and fine sand , the failure surface adapts to the lateral surface of the grid; when the grain size increases, but stays lower than the dimensions of the grid apertures, the failure surface is at a tangent to the bearing members of the grid; and when the grain size is similar to that of the grid apertures, the soil particles placed against the bearing members overtopping them drive away the failure surface to the soil mass. The soil- geogrid interface coefficient is maximum in this case. The lower value of that coefficient is achieved when the soil particle size is so high that it inhibits its penetration on the grid apertures, and the interface resistance is mobilized only at the points of contact between the soil and the grid. Jewell et al. (1984) recommended the ratio: Minimum aperture dimension A verage soil particle size -
- - - - - , - - - -- - > 3
(2.16)
for geogrids used as a soil reinforcement. Based on results of pullout tests carried out with metallic grids and Leighton Buzzard sands, Palmeira and Milligan (1989) showed the influence of the mean soil particle size on the mobilized bearing stress on the grid bars (Fig. 2.7). It can be seen that when the ratio, B/ Dso, between the bearing members' size of the grids and the mean soil particle size, is less than 10, the mobilized bearing stress can be improved by more than two times, depending on the bearing members' shape.
2·5
e
-be -be ~I~ ::n ::n 1·5
.!:...!:.. 0>
D
1·0
D
Square section
o -lI-
e"
Round section
~ ~EqUatiOn(2.17) D
"~D.~----~~ ~-E-qU-a-tio-n-(-2.-18-)~~~
_______________ I
.~
I
I I I I I I
(I)
(II
0·5
Fig . 2.7. Influence of particle size 8 / 0 50 on mobilized bearing stress (replotted from Jewell, 1996)
D
8
"
0"
.~
c
Palmeira and Milligan (1989)
I I
0·0
5
10 15 Size ratio, 8 1050
20
25
Soil-geosynthetic interaction
63
Based on the results of Palmeira and Milligan (1989), Jewell (1990) suggests the consideration of the influence of the soil grain size on the mobilized bearing stress, in terms of B / D so, as: when -
B
Dso
< 10
(2.17)
and
O'~ O'~
=
(O'~)
when
0':1 00
~ > 10
(2 .18)
Dso
O':Joo
In equations (2.17) and (2.18), and in Fig. 2.7, (O'~ / is the bearing stress that is mobilized where the soil particle size is unimportant, assuming for a continuum given by equation (2.15) for grid bearing members of circular cross-section . The improvement in the bearing stress when the grid bars are rectangular is about 20% greater than when they are circular. Jewell (1996) suggests that equation (2.9) can be rewritten as: f = fb = as
(t~: ; ,) + F,F2
(:D
00
(a~B)
C ¢') 1 ta n
(2.19)
where F, is the scale effect due to the soil mean particle size, D so, and F2 is the shape factor. When B/ Dso < 10:
F,
B_)
= (2 _ _
10D so
(2.20 )
when B/ D so > 10:
F, = 1·00
(2.21 )
For circular bars, F2 = 1'0, and for rectangular bars, F2 = 1·2. Lopes (1998) studied the influence of soil grain size on soil- geogrid interaction by carrying out pullout tests with geogrids in sand. Test conditions and procedures were similar all along the test program, however, different soils and geogrids were used. Figure 2.8 shows the particle size distribution of the soils (sand 1 and sand 2) used in the study and Table 2.1 presents their properties. In Table 2.1:
Cu --
D 60
C =
D 30 DIO D 60
c
(2.22)
D,o 2
ID = ( I'max I'
I' - I'min ) 100% I'max - I'min
(2.23 ) (2.24)
where D IO , D 30 and D 60 are the grain size corresponding to 10%, 30 % and 60 % of soil passed during sieving (see Fig. 2.8), I'min and I'max are the minimum and maximum soil unit weight, and I' is the soil unit weight for the relative density I D . The characteristics of the two high density polyethylene (HDPE) uniaxial geogrids (GG I and GG2) tested are presented in Table 2.2. Figure 2.9 shows the variation of the pullout force with front displacement and the displacements by strain along the geogrid GG 1 tested in sand 1 and sand 2. Similar behaviour was observed with geogrid GG2. With both geogrids, Lopes observed an increase in the soil- geogrid interface strength when they were tested with sand 2. That is, when the
64
Geosynthetics and their applications
100 90 0> C
.0;
80
(/)
'" C '"
Q.
70
Q)
0> Q)
~
Q) Q. Q)
>
~ :; E :J ()
60 50 40 30 20 10 0 0·05
0·1
0·5
5
10
5
10
Particles size: mm (a)
100 90 0> C
.0;
80
(/)
'"
Q.
70
Q)
0>
.l!! c
60
~
50
Q)
40
Q)
Q) Q.
>
:g :; 30 E :J
()
20
Fig. 2.8. Grain size distribution for: (a) sand 1; and (b) sand 2
0·1
0·5 Particles size : mm (b)
Table 2.1 . Soil characteristics Soil
° min
0 10
0 30
O SO
0 60
Cu
O max
Cc
I min
0·074 0·074
0·18 0·44
0·30 0·84
0·43 1·30
l{lo = SO% )
1/:
16·32 17·01
35·7 44·2
0
(kN/m 3 )
(mm) Sand 1 Sand 2
Imax
0·53 1-60
2·00 9·54
2·94 3·64
0·94 1·00
15·00 15·60
17·90 18'70
' Soil friction angle (at 38 kPa vertical pressure) in direct shear.
soil grain size increases and the B/ Dso ratio decreases. Table 2.3 shows, for geogrids GG I and GG2 in both sands, the ratio B/ Dso and the scale factor F) defined by equations (2.20) and (2.21). Lopes (1998) found an increase in the global strength of the soilgeogrid interface of approximately 24% and 27%, respectively, for GG 1 and GG2, which is about one half of that suggested by the values of the scale factor (due to the mean particle size effect) proposed by Table 2.2. Geogrids tested - dimensions of pullout apparatus: 1·53 m length , 1·00 m width and 0·80 m height Geogrid
Material
L
W
B
S
Tensile strength : kN/m
3·55 5'70
16 16
80 120
(mm) GG1 GG2
HDPE HDPE
960 960
330 330
Soil-geosynthetic interaction
65
60 GG1 sand 2
50 E
Z ""Qi
40
/--
GG1 sand 1 - ",_
~ 30 :::l
~ 20
,, ,,,
c..
10
100
50
150
200
Front displacement: mm (a)
80 E E
70 GG1 sand 2
C 60
.~
Fig. 2.9. Influence of soil grain size: (a) pullout force versus front displacement; and (b) displacements by strain along the geogrid (replotted from Lopes, 1998)
Cii
50
'" C Q)
40
~ ~
30
a.
'" is
20 10 O L-----L-----~----~----~~~~
1
2
4
3
5
6
Geogrids bearing members (b)
Jewell (1990; 1996) based on results of Pal me ira and Milligan (1989) for the increase in the bearing strength mobilized in the interface (48 % and 56% for GG 1 and GG2, respectively) . Among other reasons, such as different test procedures and conditions, some of the following can be given to explain the difference in the results: The geogrids tested by Lopes (1998) are in HOPE and those tested by Palmeira and Milligan (1989) are in mild and galvanized steel. (b) Inextensible materials, such as steel grids, move in relation to the surrounding soil during pullout, and the resistance is mobilized simultaneously along the reinforcement and at all the bearing members of the grids. (c) Extensible materials, such as HOPE geogrids, deform at the same time as they move in relation to the surrounding soil during pullout, owing to a different degree of strength mobilization along the reinforcement, and at the bearing members of the geogrids. (d) With extensible materials the increase in the passive thrust mobilized on the bearing members of the grid due to the soil (a)
Table 2.3. 8 / 0 50 and F1 Geogrids
GG1 GG2
Sand 1
Sand 2
8 / 0 50
F1
8 / 0 50
F1
8·26 13·26
1·17 1·00
2'73 4·38
1·73 1·56
66
Geosynthetics and their applications
(e)
grain size can be responsible for an increase in material deformation during pullout. For the reasons pointed out in (c) and (d) above, extensible materials can mobilize lower interface strength during pullout than inextensible ones.
Considering the obtained results, Lopes (1998) suggests an adoption of a scale factor Fl lower than that proposed by Jewell (1996) when grids are in extensible materials.
2.5.2. Confinement stress
Confinement stress has an important role on soil- geosynthetic interface resistance because it affects the soil friction angle, and both are directly related (see equation (2.1 )). The influence of confinement stress is even more notable when the strength mobilization in the interface is a threedimensional phenomenon, as can be considered on strips and , partially, on geogrids. In this case, an increase in the confinement stress can inhibit, more efficiently, the dilatance that tends to occur, in den se soils, in the interface, leading to a higher improvement in soil- geosynthetic interface strength. Several authors have studied the influence of confinement stress on soil- geosynthetic interaction. As an example, results from the study carried out by Lopes (1998) concerning the influence of confinement stress on soil- geogrids interaction during pullout will be presented. Lopes performed pullout tests with geogrid GG 1 (see Table 2.2) in 45
GG1 sand 1 confinement stress = 38 kPa
40
------ .......
35
.€ z ""a; 2
, ,,
30
,
,,
,
' ...
__ ...
.
' \
:' GG1 sand 1 confinement stress = 24·5 kPa
25
I
.Q
:; 20
52 'S 0..
15 10 5 00 Front displacement: mm (a)
45 E 40
E C 35 .~ 30
GG1 sand 1 confinement stress = 38 kPa
1i)
£
25
"""
VI
Fig. 2.10. Influence of the confinement stress: (a) pullout force versus front displacement; and (b) displacements by strain along the geogrid (replotted from Lopes , 1998)
C 20 Q)
~
",~,~:' ,~d ooo,'o~~"","" 1
15
al a. rn
10
i5
5 0
"
1
2
3 Geogrids bearing members (b)
24·5 kP'
Soil-geosynthetic interaction
67
sand I (see Table 2.1) with two values of confinement stress: 24·5 and 38·0 kPa. The friction angle of the sand defined in direct shear tests at 24·5 kPa was 38-4°. Figure 2.10 presents the variation of the pullout force with front displacement and the displacements by strain along the geogrid. As can be seen, an increase of about 55% on the confinement stress leads to an increase in the shear strength mobilized in the interface (Fig. 2.10(b)) and to an improvement in the soil- geogrid interface resistance of about II %. It must be emphasized that the soil's relative density (Io ) was 50% , which cannot be considered a dense state; if the sand was denser, a greater increase in the interface resistance would be expected . However, it must be kept in mind that the pulling out of geogrids leads to a different degree of interface strength mobilization along the reinforcement, with the soil, in some areas, at its strength at constant volume and in other areas at peak strength.
2.5.3. Soil density
Soil density affects soil- geosynthetic interface strength in the same way as confinement stress. Dense granular soils are more resistant and stiffer than the loose ones, presenting dilatant behaviour and inducing higher confinement stresses. Lopes and Ladeira (1996a) studied the influence of the soil density on soil- geogrid interface resistance by performing pullout tests with a uniaxial geogrid in the sand. The tests were carried out in similar conditions, except with regard to the sand relative density (Io) which was 50 % in one test and 86% in the other. Figure 2.11 shows the grain size distribution of the sand. The main characteristics of the geogrid and of the sand used during the study are presented on Tables 2.4 and 2.5, respectively. Figure 2.12(a) shows the variation of the pullout force with frontal displacement registered in the tests. It can be observed that, for the looser state of the sand , the geogrid fails by lack of adherence at a pullout force of about 32·2 kNlm , failing by lack of tensile strength for the denser state. The maximum pullout force achieved with the denser sand was about 45·3 kNlm, which would certainly have been higher if the reinforcement had not failed through lack of tensile strength. An increase higher than 40% in the strength of the soil- geogrid interface when the relative density of the sand changes from 50 % to 86 % is due to the greater soil and soil- geogrid strength in denser sand. The displacement of the geogrid decreases, increasing the interface stiffness 100 90 Ol
c 'u;
80
'"
70
Ol
~ c
60
~
50
'"c. OJ
OJ OJ
c.
>
40
:; E
30
()
20
OJ
1ii ::J
10
Fig . 2.11. Grain size distribution for sand 3
0 0·05
0·1
0·5 Particles size: mm
5
10
68
Geosynthetics and their applications
Table 2.4. Geogrid and 0·80 m height
dimensions of pullout apparatus: 1·53 m length, 1·00 m width
Geogrid
L
Material
W
B
S
Tensile strength: kN/m
Peak strain: %
16
55
11-5
(mm)
GG3
HOPE
960
2·6
330
modulus and the pullout force. As the soil density increases, the length of adherence decreases. In fact, for the higher soil density tested, only one third of the inclusion length contributes to resistance (Fig. 2.l2(b)).
2.5.4. Geosynthetic structure The distance (S) between the bearing members of grids is an important parameter with regard to soil- grids interaction . In fact, if that distance is lower than an optimum value, there is an interference between members, each one being less effective. Assuming the limit case where the strength of the soil- grid interface derives only from the passive thrust mobilized on the bearing members Table 2.5. Soil characteristics Soil
Cc
I'm;n
I'max
1'(10 = 50%)
1'(10 = 86%)
<1>(10 = 50%) ' :
18-45
35·2
<1>(10 = 86%)':
0
(mm) Sand 3 0·074 0·34
0·63 1·67 11-67 4·91
0'70 16·10 18·90 17·39
35·7
· Soil friction angle (at 46·7 kPa vertical pressure) in direct shear. 50 -/0=50%
.,../ - - - , \ \
/
1:' z
40
~
30
""
/ /
/
/
\
/
/
.E ::;
--- 10 "-
"
"-
,
/
\
I
"-
I
52 20
I
'S
a.
"\ \ I
I
I.
I I
10 O L-~
o
=86%
\ I
__L-~_ _~~~~~L-~~~~~~~
20
40
60
80
100 120
140 160 180 200 220 240
Front displacement: mm
(a)
E 35 E c
~
.0
,, ,
,,
c 20 Q)
E
Fig. 2.12. Influence of soil density: (a) pullout force versus front displacement; and (b) displacements by strain along the geogrid (replotted from Lopes and Ladeira , 1996a)
l;l
--- 10 = 86%
,,
25
J!l
Q)
-/0=50%
,,
°e
\
15
,,
,, ,, ,,
%10 Cl
,
5
'- - - - - - - -
O L------'-------'--------"==~~~--'-----'=~
2
3
4
5
6
Geogrids bearing members (b)
7
8
0
Soil-geosynthetic interaction
69
of the reinforcement and that there is an upper limit for the interface strength equivalent to that of a completely rough sheet (8 = ¢'), Jewell et at. (1984) and Jewell (1990) consider that the maximum strength on the interface is achieved for an optimum geometry of the grid (S /( abB))q/' Taking into account the authors' assumptions, from the general equation (2.9), fb is: (2.25 ) and, being (fb )max = 1.00
C:B) (:D
(2.26)
Ctaln ¢')
¢' =
From equations (2.25) and (2.26), the coefficient of bond can be expressed as:
(b) ¢, fb
(2.27)
(~)
=
abB The authors concluded that when the grid geometry is lower than the optimum, the bond strength mobilized in each single member is proportionately lower, as the coefficient of bond cannot increase above (fb)max = 1'00, as shown on Fig. 2.13. Palmeira and Milligan (1989) studied the influence of the distance between bearing members of grids on the resistance of the soil- grid interface by carrying out pullout tests with metallic grids with a different distance between the bearing members in sand. The authors concluded that, as that distance decreases, the interface resistance also decreases, thereby denoting the increase of interference of bearing members with the reduction of the distance between them. The authors suggest that the concept of interference between the grid members, as the ratio of the mobilized bearing stress to maximum possible value, should be introduced explicitly into the analysis of bond resistance in terms of the distance of interference:
(2.28)
/
I
I ___________
1·00
1_
Equation (2.27)
,.
0·75 fb
•
0·50 0·25 0
Fig. 2.13. Influence of reinforcement geometry on the coefficient of bond (replotted from Jewell et al. , 1984)
0
0·25
0·50
/
0·75
(a:B )~, (a:B )
/ 1"
/
-.
_...
/ __
~
~~~-ll--
Pullout tests ~
Hueckel and Kwasniewski (1961) D. Chang et al. (1977) o Jewell (1980)
1 1 1 1 1 1 I 1 1
Unit cell tests •
0·75 0'50
0·25
(a:B ) (a:B )~,
0
Jewell (1980)
70
Geosynthetics and their applications
Table 2.6. Geotextile Geotextile
dimensions of pullout apparatus: 1·53 m length , 1·00 m width and 0·80 m height
Type
Material
L
W
Thickness
Unit weight: g/m 2
Tensile strength : kN / m
Peak strain :
800
50
65
%
(mm) GT1
Non-woven
PP
960
330
6
with DI 2': O. In this case: (2.29)
and:
ih~(I- DJ)(bt (a:B)
(2.30)
which is another expression for equation (2.27). Palmeira and Milligan (1989) remark that the soil properties (including particle shape and surface characteristics) and the di ameter, spacing and number of bearing members are the main factors that control the interference between bearing members. Although the results obtained obey the general pattern of variation, the authors stated that deviations to that pattern are likely to occur when: • • • •
there are tangency between members the reinforcements are very long the reinforcements are extensible, and for grids in which the friction along the longitudinal members is a significant part of the soil-reinforcement interface resistance.
Lopes (1998) studied the influence of the structure of geosynthetics on soil-reinforcement interface strength by carrying out pullout tests with sand 1 (see Table 2.1). The test procedures and conditions were similar, but the reinforcements used were a uniaxial geogrid (Table 2.4) and a spunbounded non-woven geotextile. The main characteristics of the geotextile tested are presented in Table 2.6. Figure 2.14 presents the variation of the pullout force with front displacement for the two materials. It can be seen that, although both geosynthetics had similar tensile strength (50 kN/m for the non-woven 45 40 GG3 sand 1
35
.§
z 30
.><
Qi ~
25
Q
:; 20
.Q
:J Cl.
Fig. 2.14. Influence of geosynthetic structure on soil-geosynthetic interface strength (replotted from Lopes, 1998)
GT1 sand 1
15
_ - - - - - -1- ____ - - : I I
10 5 ,,
,, ,, ,,
00 Front displacement: mm
Soil-geosynthetic interaction
71
geotextile and 55 kN/m for the uniaxial geogrid), the strength mobilized on the soil- geogrid interface is about 2·6 times greater than that mobilized on the soil- geotextile interface, the first being achieved for front displacements of about 1/4 of the second . The structure of the geosynthetics and their extensibility had greater influence on the observed behaviour. The structure of the spunbounded non-woven geotextile is much more extensible than that of the geogrid (see Tables 2.4 and 2.6) . The degree of deformation of the geotextile during pullout is much higher than that of the geogrid, and the mobilization of the interface resistance is less effective. On the other hand, it must be stated that the interaction mechanisms mobilized in both geosynthetics are different, for the geotextile skin friction and for the geogrid skin friction plus the passive thrust on the bearing members, which can also contribute to the effectiveness of the interface strength mobilization when geogrids are used as reinforcement.
2.6. Laboratory tests for the quantification of soil-geosynthetic interface resistance
The laboratory tests available for the quantification of soil- geosynthetic interface strength are the direct shear and pullout tests. The adequacy of each test to the definition of interface characteristics relies on the relative movement that tends to occur in the soil- geosynthetic interface - for direct sliding movement, the direct shear test is more adequate, whereas for pullout movement, the pullout test should be conducted. Quite often the interface characteristics defined by direct shear and pullout tests are different and are even sometimes inconsistent as a result of different test procedures, stress paths, failure mechanisms and boundary conditions.
2.6.1. Direct shear test The coefficient of direct sliding is quite well defined for geotextiles and geogrids, using the modified direct shear test from that used for soil mechanics. Gourc et al. (1996) recommend the use of direct shear apparatus with 300 x 300 mm or greater. In the most common direct shear tests performed with geotextiles, these materials are fixed to a rigid plane support placed on the lower half of the apparatus (Fig. 2.15(a». This procedure models, with sufficient accuracy, the interaction mechanism that occurs in the soil- geotextile interface during direct sliding, that is skin friction between the full plan area of contact. Alternatively, the geotextiles can be supported on soil (Fig. 2.15(b». Direct shear tests with geogrids can also be performed with the geosynthetic fixed to a rigid plane support placed on the lower half of the apparatus. However, for these materials, especially for those with large apertures and a high percentage of openings, it is suggested that the
II'-------'i (a)
(b)
..
(c)
Fig. 2.15. Schematic representation: (a) for geotextiles - direct shear test with a solid block in the lower half of the apparatus ; (b) for geogrids - direct shear test with soil in the lower half of the apparatus ; and (c) pullout test
72
Geosynthetics and their applications
tests be carried out with the geogrids supported on soil placed on the lower half of the apparatus (Fig. 2.15(b)). In fact, the direct sliding strength for geogrids is generated by two mechanisms: soil- soil friction along the grid's apertures and skin friction over the geogrid itself. Soil sliding over soil through the apertures of the grid is not modelled when using the first-mentioned test procedure, and it can reach an important percentage of the interface resistance, namely, for grids with large apertures and a high percentage of openings. The direct shear tests are relatively easy to interpret, although results are dependent on some factors , such as: relative position of soil and reinforcement; methods used for normal stress control; thickness of soil layer; and roughness of the rigid plane (Gourc et at. , 1996). 2.6.2. Pullout test
The coefficient of bond can be defined by performing pullout tests (Fig. 2.15(c)). However, on the contrary to what happens with the direct shear tests, the pullout tests are difficult to interpret, as the results are greatly influenced by the boundary conditions, and test procedures and conditions (Lopes and Ladeira, 1996a; 1996b). Jewell (1996) suggests that, since the bond mechanism in both sides of geotextiles is identical to that in direct sliding (skin friction) , there is no need to perform pullout tests with these materials to define the coefficient of bond , as it is very similar to that of direct sliding obtained from the much simpler modified direct shear test. However, for geogrids, the interaction mechanisms mobilized during pullout are significantly different from those in direct shear, developing resistance by mobilizing passive thrusts on the transverse members of the grid, by soil- soil friction through the grid apertures and by skin friction over the planar geogrid surface areas. Thus, the coefficient of bond for geogrids can only be measured by pullout tests. Considerable care is required in both the execution and interpretation of pullout tests, as the results can be greatly affected by the use of different test apparatus associated with distinct boundary effects, different test procedures, different schemes of placement and compaction of the soil, etc. (Juran et at., 1988). A comprehensive study about the influence of displacement rate, specimen size, soil height, and sleeve length on the pullout behaviour of geogrids, was developed by Lopes and Ladeira (1996a; 1996b). In their studies, the pullout apparatus used was, internally, 1·53 m long, 1·00 m wide and 0·80 m high (Fig. 2.16). The use of an apparatus with large dimensions aimed to minimize the influence of the lateral, base and top boundaries. To reduce the influence of the top boundary and to achieve a uniform distribution of the applied vertical stresses on the sample top, a 0·025 m thick smooth neoprene slab was used between the top
.. :~ :
Fig. 2.16. Pullout apparatus: (a) frontal view and hydraulic system; and (b) lateral view
Soil-geosynthetic interaction
73
soil and the wood plate where the confining stress was applied. To reduce the influence of the front wall, a steel sleeve 0·20 m long was used inside the box . The soil was poured into the box from a constant height of O' 50 m and placed in 0·15 m thick layers. Each layer was levelled and compacted to the required relative density (50%) using an electric vibratory hammer. The soil density was controlled with a gammadensimeter. When the steel sleeve level was reached (located at middle height of the box), the reinforcement was placed over the compacted soil, introduced through the sleeve and fixed to the clamps outside the box. The inextensible wires used to measure the displacements along the reinforcement were then put in place and connected to the linear potentiometers at the back and at the front of the box. Six potentiometers were used in the tests. Finally, two 0·15 m thick soil layers were placed, levelled and compacted, resulting in a soil height of 0·60 m with the reinforcement at the middle. The pullout force , applied by an hydraulic system, was measured by a load cell placed in the clamping system that transmits the force to the reinforcement. The confining stress (46'7 kPa at the sample level), applied by ten small cylindrical masses, was kept constant through the test and was measured by a load cell located between one of the cylindrical masses and the wood plate below. The tests were carried out at a constant displacement rate and volume, and the results were recorded using an automatic data acquisition system. The main characteristics of the sand and of the uniaxial geogrid used in the tests are presented in Table 2.5 and Fig. 2.11 , and in Table 2.4, respectively. The test programme carried out to study the influence of the displacement rate, specimen size, soil height and sleeve length on the results of the pullout tests with the geogrids in sand is presented in Table 2.7. It is important to note the relevance of the measurement of displacements along extensible reinforcements such as geosynthetics. In fact, as mentioned before, in this type of material the displacement during pullout has two components, one corresponding to the relative movement between the reinforcement and the surrounding soil (displacement by shear strain on the soil-reinforcement interface) and another due to the Table 2.7. Testing programme Parameter under study
Sample lengthl sample width: m
Displacement rate: mmlmin
Soil height above samplelsoil height below sample : m
Sleeve length : m
Displacement rate
0'96/0'33
0·3010'30
0·20
Sample length
0·3010·30
0·20
5-4
0·3010'30
0·20
Soil height
0·8010·33 0'96/0·33 1-12/0'33 0·96/0·33 0·96/0·47 0·96/0·60 0·96/0·33
5'4 1·8 11'8 22 '0 5·4
5·4
0·20
Sleeve length
0·96/0·33
0·2010'20 0·3010·30 0·3010 ·30
Sample width
5-4
Without sleeve 0'20
74
Geosynthetics and their applications
- 1·8 mm/min - - 5·4 mm/min ---- 11·8 mm/min 22·0 mm/min
40 35 /-
.§ 30 Z
"" CD
~
---------
/ /
/
/
25
/ /
/
.E 20 '5
I
/
/
l
.2 15 'S
"
/
/
0..
10
O
L-~
o
__
20
~
__L - - L_ _
40
60
~~_ _- L_ _~~_ _~_ _L-~
80 100 120 140 160 Front displacement: mm (a)
40
E E 35 C
.~
ii)
30
, "~
~ 25 CD
Fig. 2.17. Influence of the displacement rate: (a) pullout force versus front displacement; and (b) displacements by strain along the geogrid (replotted from Lopes and Ladeira , 1996a)
E
220 240
- 1·8 - - 5-4 ---- 11 ·8 22·0
mm/min mm/min mm/min mm/min
~,> ~
(f)
C
"-
180 200
"-
20
CD
u
ro 15
C. (f)
Ci 10 5 0
.......::::_--=-_._-::;; 2
3
4 6 5 Geogrids bearing members
7
8
(b)
reinforcement elongation, this component being more significant at the front part of the specimen. The possibility of knowing each one of the displacement components of the movement separately, allows for a better understanding of the soil- geosynthetic interaction phenomenon. 2.6.2.1. Influence of displacement rate
The influence of the displacement rate on the pullout behaviour of the geogrid under study is shown in Fig. 2.17. It can be seen that the maximum pullout force of the reinforcement increases with the displacement rate, decreasing the front displacement necessary to mobilize the same pullout force (Fig. 2.17(a)). On the other hand, the displacement along the reinforcement induced by strain shows a tendency to reduce as the displacement rate increases (Fig. 2.17(b)). The observed behaviour leads to the conclusion that the increase in the pullout force with the displacement rate results, at least, partially, from the increase in the reinforcement stiffness with the velocity and not from the increase of the mobilized tangential stresses in the soilreinforcement interface. Another factor responsible for the increase of the pullout force with the displacement rate is the lower soil capacity for rearrangement with the increment of the velocity. 2.6.2.2. Influence of specimen size
The study of the influence of the specimen size on the pullout test results is important for the definition of the relation between the dimensions of the apparatus and of the specimen in order to minimize the influence of the lateral boundaries on the test results. This relation is difficult to define because it depends not only on the dimensions of the equipment
Soil-geosynthetic interaction
Length
40
_
35
""
.
/
.§ 30
z
75
v -
- - 1·12m 0·96m ---- 0·80 m
___
-- ---------
25
~
.E 20 'S
.Q
15
Cl.
10
:;
Fig. 2.18. Influence of the specimen length on the geogrid pullout behaviour (replotted from Lopes and Ladeira , 1996b)
5 0
0
20
40
60
80
100
120
140
160
180
200
220
Front displacement: mm
but also on factors such as the characteristics of the reinforcement, density and type of soil, confinement pressure, displacement rate and height of soil. The influence of the specimen length on the pullout test results is presented in Fig. 2.18 in terms of the variation of pullout force with front displacement for the three lengths tested (0'80 m, 0·96 m and 1·12 m) . It can be seen that the results are similar for the three tests in qualitative terms, but not in quantitative terms. The maximum pullout force increases from 15'3kN/m to 32'2kN/m (about 105% ) when the specimen length changes from 0·80 m to 0·96 m. However, when this parameter changes to 1·12 m, the increase in the maximum pullout force is much smaller, from 32·2 kN/m to 35·6 kN/m (about 10,6% ). The results show that there is no significant increase in the interface resistance when the length of the inclusion is greater than a certain value. In order to define the maximum specimen width for its use in the apparatus, allowing for the consideration of negligible influence of its lateral boundaries on the test results, three tests were carried out with three different specimen widths (0 '33 m, 0-47 m and 0·60 m) . The variation of pullout force with the front displacement is presented in Fig. 2.19 for the three widths tested. As the specimen width decreases, the maximum pullout force increases slightly (being 28 ·8 kN/m, 30·3 kN/m and 32·2 kN/m , respectively, for samples 0·60 m, 0-47 m and 0·33 m wide) . Taking into account the difference between the results of the present study (about 12% for the maximum pullout force) , it can be said that the influence of the specimen width on the test results is small. However, it can be seen that the reduction of the distance between the lateral boundaries of the equipment and the specimen tends to reduce the soil-reinforcement interface resistance. 35 30
// - ::--
z
20
~
15
~
Fig . 2.19. Influence of the specimen width on the geogrid pullout behaviour (replotted from Lopes and Ladeira , 1996b)
"0 'S
-
-
-
- -:..::. - --;,.., ....
.,...-; .../'l
.§ 25
""Qi
-....:::.- -
Width _ _ 0·33m ---- 0·47 m _ . - 0·60 m
>'/
!.
I
10
Cl.
5 0 0
20
40
60
80
100
120
140
160
Front displacement: mm
180
200
220
76
Geosynthetics and their applications
2.6.2.3. Influence of soil height
The distances between the upper and bottom boundaries of the apparatus and the reinforcement level can affect the resistance of the soilreinforcement interface. This influence leads to an increase of the confining stress in the reinforcement, especially when the soil height is small and soil dilatancy is forbidden . In the tests performed, the friction angle (¢) was defined as the arch of the tangent of the relation between the measured shear (t) and normal (in) stresses on the reinforcement interface (tan ¢ = t / (in) (the normal stress was considered as the sum of the applied surcharge and the pressure due to the height of the soil above the reinforcement) . In fact , the defined friction angle can be higher than the real one, as a result of the increase of the confining pressure that can occur in the pullout tests due to the restriction of soil dilatancy . Another phenomenon that can occur in these tests is the development of shear forces between the soil and the stiff horizontal boundaries, especially the bottom one (Farrag et at. , 1993). In order to study the influence of the soil height placed above and below the specimen in the pullout response of the geogrid under consideration, two tests were carried out. In one of them, the reinforcement was placed in the middle of soil 0040 m high and, in the other, the reinforcement was placed in the middle of soil 0·60 m high. Figure 2.20 shows the geogrid pullout response for both situations tested. For the test with soil 0·60 m high, the maximum pullout force was 32·2 kN/m, and for the test with soil 0040 m high, it was 33·9 kN/m (about 5·3% higher) (Fig. 2.20(a)) . The displacements by strain along Soil height
0·3010·30 m
35
,
30
00
0·2010·20 m
...••........................ .........................
20
40
60
80 100 120 140 160 180 200 Front displacement: mm
220
240
(a)
50 45 E 40 E C 35 .~
ti 30
>.c
.'!l 25 c
Fig. 2.20 . Influence of the soil height: (a) pullout force versus front displacement; and (b) displacements by strain along the geogrid (replotted from Lopes and Ladeira, 1996b)
Q) 20 E Q) <..l III 15
C. (f) (5
Soil height
.... ....
/ ......
0 ·2010·20 m 0·3010·30
m
10
5 O L--L__~~__- L__L--L__~~__~__L-~~ 7 8 6 2 3 4 5 Geogrids bearing members (b)
Soil-geosynthetic interaction
77
the reinforcement for the maximum pullout force measured was higher for the lower soil height tested (Fig. 2.20(b)). Based on the present test results, it can be said that for the soil heights considered there is no evidence of a significant influence of this parameter on the pullout response of the geogrid. However, different soil heights must be tested to confirm the present results. Unfortunately, owing to geometrical limitations of the equipment used , it was not possible to test soil heights over and below the reinforcement higher than 0·30 m. Nevertheless, published pullout test results, carried out with higher heights of soil surrounding specimens with similar dimensions of those tested in the present study, show insignificant influence of soil heights greater than 0·30 m above and 0·30 m below the reinforcement (Farrag et at. , 1993). It must be emphasized that the length of the reinforcement plays an important role with regard to the influence of the top and bottom boundaries of the apparatus. In fact , as the length of the reinforcement increases in relation to a fixed height of the box, the influence of the horizontal boundaries also increases (Palmeira and Milligan 1989). From the test results obtained it can be said that when the soil height above and below the reinforcement increases, there is a tendency for a reduction of the resistance of the soil-reinforcement interface. 2.6.2.4. Influence of sleeve
The interaction between the system soil-reinforcement and stiff frontal wall of the pullout box can affect the pullout resistance of the reinforcement, especially when friction along that wall is high (Palmeira and Milligan, 1989). When the reinforcement is pulled out inside the box, there is an increase on the soil's lateral pressures against the frontal wall that leads to an apparent increase of the pullout resistance of the reinforcement. There are some procedures, such as the lubrication of the internal face of the wall, or the placement of a sleeve inside the equipment at the level of the existing aperture in the front box wall, that can be followed in order to minimize the influence of the frontal wall of the equipment. The placement of the sleeve allows the transfer of the point of application of the pullout force away from the frontal wall, in the interior of the soil mass (Farrag et ai. , 1993). In order to study the influence of the sleeve on the pullout response of the geogrid , two tests were carried out, one without a sleeve and another with a 0·20 m long steel sleeve. The geogrid pullout responses for the two mentioned tests are presented in Fig. 2.21. The absence of a sleeve leads to an increase of the maximum pullout force by about 10% . This behaviour is due to the influence of the roughness of the frontal wall in the tests without a sleeve. In fact, the frictional forces developed on the front face of the box lead to an increase in the average vertical stresses on the reinforcement responsible for the increase of the pullout forces . The lateral pressures developed on the front wall are higher in the test without a sleeve. In this case, the restriction of the geogrid movement is more effective, resulting in a higher mobilization of shear stresses in the soil- geogrid interface near to the point of application of the pullout force (Fig. 2.21(b)) and to a lower mobilization of the stresses in the back part of the reinforcement. The results showed (Lopes and Ladeira, 1996b) a pullout movement lower in the test without a sleeve, which is in agreement with the higher mobilized resistance due to the increase in the state of stress, mainly, in the front part of the reinforcement, resulting from the influence of the stiff frontal wall of the equipment.
78
Geosynthetics and their applications
- - Without sleeve ---- 0·20 m sleeve
40 35 E
Z
-'" Q; ~
30 \
25
\ \
.2
'5 20
"S 0..
,, ......
..Q
-
..... _-- '"
15 10
5 0 0
40
20
60
80
100 120
140 160
180 200
220
240
Front displacement: mm (a) 45 - - Without sleeve - - - - 0·20 m sleeve
40 E E 35 C
.~
en
'-
30
>.0 25
Fig. 2.21. Influence of the sleeve: (a) pullout force versus front displacement; and (b) displacements by deformation along the geogrid (replotted from Lopes and Ladeira , 1996b)
'"Q) 20 C E Q) 15
"'" 0. '" i:5
'-
'-
'-
'-
'-
'-
10
'-
'-
'-
'-
'- '-
5 0 2
3
4 5 6 Geogrids bearing members (b)
7
--
8
From the obtained results it can be concluded that, for the equipment used, it is advisable to use a sleeve at least 0·20 m long in order to minimize the influence of the front wall in the pullout test results.
2.7. Concluding remarks
Soil reinforcement with geosynthetics relies on the efficacy of soilgeosynthetic interaction, which is governed largely by the properties of the soil and the geosynthetic. Soil particle size assumes special importance when the reinforcement is a geogrid. Direct shear and pullout are the most common laboratory tests available for the quantification of soilgeosynthetic interaction. Pullout tests performed in order to study the influence on the results of some test conditions and procedures, showed an important role of the displacement rate, the specimen length and the existence of a sleeve in the front box wall.
References Chang, J. c., Hannon , J. B. and Forsyth, R. A. (1977). Pull resistance and interaction of earth work reinforcement soil. Transportation Research Record, No. 640, Washington DC, USA. Farrag, K. , Acar, Y. B. and Juran , I. (1993). Pull-out resistance of geogrid reinforcements. Geotextiles and Geomembranes, 12, No.2, 33- 159. Gourc, J. P. , Lalarakotoson, S. , MUller-Rochholtz, H. and Bronstein, Z. (1996) . Friction measurement by direct shear or tilting process - development of a European Standard. Proceedings of the 1st European Conference on Geosynlhelics - EUROGEO, Maastricht, the Netherlands, pp. 1039- 1046.
Soil-geosynthetic interaction
79
Hayashi, S. , Makiuchi , K. and Ochiai, H . (1994). Testing methods for soil geosynthetic frictional behaviour - Japanese standard. Proceedings of the 5th International Conference on Geotextiles, Geomembranes and Related Products, Singapore, pp. 411 - 414. Hueckel, S. M. and Kwasniewski, 1. K. (1961). Essais sur modele reduit de la capacite d' Ancrage d'elements rigides horizontaux, enfo ui s da ns Ie sable. Proceedings of the 5th In ternational Conference on Soil Mechanics and Foundation Engineering, pp. 431 - 434. Jewell, R . A. (1980). Some effects of reinforcement on the mechanical behaviour of soils. DPhil thesis, Cambridge University, UK. Jewell, R. A. (1996) . Soil reinforcement with geotextiles, CIRIA and Thomas Telford Publishing, London . Jewell , R . A. (1990). Reinforced bond capacity. Geotechnique, 40, No.3, 513518. Jewell, R. A. and Wroth, C. P. ( 1987). Direct shear tests on reinforced sand. Geotechnique, 37, No. I, 53- 68. Jewell, R . A., Milligan, G . W. E., Sarsby, R . W. and Dubois, D. (1984). Interaction between soil and geogrids. Proceedings of the Conference on Polymer Grid Reinforcement, Thomas Telford Publishing, London , pp. 18- 30. Juran, I. , Knochenmus, G. , Acar, Y. B. and Annan, A. (1988). Pull-out response of geotexti les and geogrids. Proceedings of Symposium on Geotextiles for Soil Improvement, ASCE, Geotechical Special Publication, No . 18, pp. 92- 111. Lopes, M . J. F. P. (1998). Study of the influence of soil grain size and reinforcement structure on soil- geosynthetic interaction mechanisms. MSc thesis, University of Porto, Portugal (in Portuguese). Lopes, M . L. a nd Ladeira , M. (l996a). Influence of the confinement, soi l density and displacement rate on soil- geogrids interaction. Geotextiles and Geomembranes, 14, No . 10, 543- 554. Lopes, M . L. and Ladeira, M. ( 1996b). Role of the specimen geometry, soil height, and sleeve length on the pull-out behaviour of geogrids. Geosynthetics In ternational, 3, No.6, 701 - 719. McGown , A. , Andrawes, K. Z. and AI-Hasani, M . M. (1978). Effect of inclusion properties on the behaviour of sand. Geotechnique, 28, No .3, 327- 346. Ospina, R. I. (1988). An investigation on thefundamental interaction mechanism of non-extensible reinforcement embedded in sand. MSc thesis, Georgia Institute of Technology, USA. Palmeira, E. M. and Milligan, G. W. E. (1989). Scale and other factors affecting the results of pull-out tests of grids buried in sand. Geotechnique, 39, No.3, 511 524. Peterson, L. M. and Anderson, L. R. (1980) . Pullout resistance of Ivelded lVire mesh embedded in soil. Department of Civil Engineering, Utah State University, USA, Resea rch report.
3
Retaining walls B. M. DAs College of Engineering and Computer Science , California State University, Sacramento , USA
3.1. Introduction
Since the early 1970s, various types of geosynthetics have been used to reinforce soil in the construction of retaining walls in many parts of the world . In the early part of the 1980s, Netlon Ltd in the UK was the first to produce geogrids. In 1982, Tensar Corporation (now Tensar Earth Technologies, Inc.) introduced geogrids in the United States. Since then, geogrids have been increasingly used as a soil-reinforcement material in the construction of retaining walls. This chapter provides the general guidelines for designing retaining walls using geotextile and geogrid as reinforcing materials. These walls are flexible compared to the rigid retaining walls constructed with reinforced concrete. Figure 3.1 shows the schematic diagram of a geotextile-reinforced retaining wall. In most cases, a granular material is used as the backfill. In this type of retaining wall, the facing of the wall is formed by lapping the sheets as shown with a lap length of 11' When construction of the wall is finished , the exposed face of the wall must be covered; otherwise, the geotextile will deteriorate from exposure to ultraviolet light. Bitumen emulsion or Gunite is sprayed on the wall face. A wire mesh anchored to the geotextile facing may be necessary to keep the coating on the face of the wall. Schematic diagrams of some typical retaining walls constructed with geogrid reinforcement are shown in Fig. 3.2. Figure 3.2(a) shows a geogrid wrap-around wall. A geogrid-reinforced wall with gabion facing is shown in Fig. 3.2(b). Figure 3.2(c) shows a vertical retaining wall with precast concrete panels as the facing.
3.2. Design considerations
3.2.1. Stability
At the present time, the common practice used in designing retaining walls with geosynthetic reinforcement is the limit equilibrium analysis. The analysis consists of two major parts: (a)
(b)
Internal stability involves determining tension and pullout resistance in the reinforcing elements, length of reinforcement, and the integrity of the facing elements. Ex ternal stability involves checking the overall stability of the stabilized mass as it relates to sliding, overturning, bearing capacity failure , and deep-seated stability (Fig. 3.3).
3.2.2. Lateral earth pressure
In order to conduct the stability check described above, the lateral earth pressure behind the retaining wall must be determined. The general guidelines for determining the horizontal and vertical earth pressures used in the design of retaining walls with geosynthetic reinforcement follow .
82
Geosynthetics and their applications
Granular soil
Geotextile
Granular soil In-situ material
Fig. 3.1. Geotextilereinforced retaining wall
Figure 3.4 shows a retaining wall with a granular backfill having a unit weight of 1'1 and an effective friction angle of ¢;. Below the base of the retaining wall, the in-situ soil has been excavated and recompacted with the granular soil used for backfill. Below the backfill, the in-situ soil has a unit weight of 1'2, effective friction angle of ¢; and cohesion of A surcharge with an intensity of q per unit area lies at the top of the retaining wall. The wall has geosynthetic reinforcement ties at depths z = 0, Sy , 2Sy , .. . , NSv . The height of the wall is NS y = H . According to the Rankine active pressure theory:
e;.
(S~ = (S~Ka
-
l
2e
#a
(3.1)
where (S~ is the Rankine active pressure at any depth z. Gabion facing
(a)
(b)
Precast concrete panel
Fig. 3.2 . Typical schematic diagrams of retaining walls with geogrid reinforcement: (a) geogrid wraparound wall; (b) wall with gab ion facing; and (c) concrete panel-faced wall
Levelling pad
Retaining walls
(a)
Fig. 3.3. External stability checks: (a) sliding; (b) overturning; (c) bearing capacity; and (d) deep turning stability
83
(b)
(e)
(d)
For dry granular soils with no surcharge at the top, c' and Ka = tan 2 (45° - ¢~ /2). Thus:
= 0, O'~ = l
0'~(1) = I lzKa
i Z'
(3.2)
When a surcharge is added at the top, as shown in Fig. 3.4(a):
,
,
0' v --
= 11z
,
+
O'v( l )
r
r
Due to soil only t-- b'
r Z
(3.3)
O'v(2)
Due to the surcharge
--+- a'--I .::.~.\
I
;;:'::
I
•t'.
Ir
~.
I.
I
y, <1>;
Sv--l
I
H
I
Sv
I
,I
,,
I
1
Sv Z
:,.::.,;., ... :-'
Sand
Sv
I
.;:;:
t.4. ...
Sv
= NS v
In-situ soil Y2:
~: c~
-' ....... ~
~,~ . ..
, ••~_:-. . ,i:>- • • '
(a)
,
+
O'a
Fig. 3.4. Analysis of a reinforced earth retaining
wall
~:,:.t::y•.',.
Sv
(b)
84
Geosynthetics and their applications
• ~ ~ ~ • qlunit
L
area
~i- - - - - -~- Re-i-nf-or-ce-m-e-~-:'- ~ .~
strip
(a)
I-- b' --I-- a' --l
l qlunit area
fI
~ I ~ R : ; ~ ~ c e m e n t L
Fig . 3.5. (a) Notation for the relationship of (J~(2) equations (3.4) and (3.5); and (b) notation for the re lationship of (J~ (2) equations (3.7) and (3.8)
H
~~
strip
l~
(b)
The magnitude of (f~ (2) can be calculated by using the 2: 1 method of stress distribution, and It is shown in Fig. 3.5(a). According to Laba and Kennedy (1986): (f
,
v(2) -
-
qa
,
a' + z
and: , (fv(2)
= , a
qa z
(3.4)
(for z ::; 2b')
-
,
, (for z > 2b )
,
(3.5)
+"2+b
Also, when a surcharge is added at the top, the lateral pressure at any depth is:
,
(f'a--
= K ; '(,z
r
,
+
(fa(l)
Due to soil only
(fa(2)
(3.6)
r
Due to the surcharge
According to Laba and Kennedy (1986), (f~(2) may be expressed as (Fig. 3.5(b)):
(f~(2) =
M
[2: ((3 r
sin (3 cos
(in radians)
20:) ]
(3.7)
Retaining walls
85
where:
M
0-4b' > 1
= 1-4 _
(3.8)
0·14H -
The total active (lateral) pressure distribution on the retammg wall, calculated by using equations (3.6), (3 .7), and (3 .8), is shown in Fig. 3.4(b).
3.2.3. Tie force
Refer again to Fig. 3.4. The tie force per unit length of the wall, T , developed at any depth z can be calculated as:
T
3.3. Design procedure for retaining walls with geotextile reinforcement
=
active earth pressure at depth z x Sv = O"~Sv
(3.9)
Referring to Fig. 3.6, below is a step-by-step procedure for the design of retaining walls using geotextile as reinforcement.
3.3.1. General
1. 2. 3.
Determine the height of the wall, H, and the properties of the granular backfill, such as uni t weight (,1) and angle of friction , (¢'I ). Obtain the soil- geotextile interface friction angle, ¢~ . Obtain the in-situ soil parameters, such as unit weight (,2), effective friction angle (¢~), and cohesion (c~).
3.3.2. Internal stability 1. Determine the active pressure distribution on the wall from: (3.10)
where Ka is the Rankine earth pressure coefficient = tan 2 (45° - ¢'1/2), I I is the unit weight of the granular backfill, is the effective friction angle of the granular backfill and Select a geotextile fabric that has an allowable strength of O"G (kN/m)
¢;
2.
:>!!?'ifoI::--
-
-----;---7''------ - - Geotextile
z >.fii!<:::--------r~--------
H
-_71.rv;'---+-~--
,
'.
-.jl
- - ---l..
~O
8
Sand
1 18 v
"'(1;
4>;
~;q...----~----
v
Fig . 3.6. Retaining wall with geotextile reinforcement
Geotextile
Geotexti le
Geotextile
86
Geosynthetics and their applications
3.
Determine the vertical spacing of the layers at any depth z from: (3.11)
4.
where FS(B) is the factor of safety against ru pture of reinforcement. The magnitude of FS(B) is generally taken as 1.3- 1.5. Geotextile layers at any depth z will fai l by pu llout if the frictional resistance developed along their surfaces is less than the force to which the layers are being subjected . The effective length of the geotextile layers along which the frictional resistance is developed may be taken conservatively as the length that extends beyond the limits of the Rankine active failure zone, which is the zone ABC in Fig. 3.6. Line BC in Fig. 3.6 makes an angle of (45° + ¢/1/ 2) with the horizontal. Now, the maximum friction force , F R , per unit length of the wall that can be realized for a geotextile layer at depth z is: (3.12) Thus, the factor of safety against pullout, FS(p), at any depth z (from equations (3 .9) and (3.12)) is: FS
_ FR
_
T -
(P) -
2cr~le tan ¢~ crri Sv
(3.13)
or: I __ S_vcr_~_[F_S_(,-p~)l e 2cr'y tan 'fo'F A,I
(3.14)
cr~ = 1 1zKa
where
1
cry = l iz So: I = _Sv-::-K_a_[F---,.S7'(p,,- )1 e 2 tan ¢~
(3.15)
The magnitude of FS(p) is generally taken as !-3- 1·5. If an experimental value of ¢~ is not avai lable, it may be assumed to be about 2/ 3¢;. Now, determine the length of each layer of geotexti le as: L = Ir
+ Ie
(3.16)
where: H -z
I, ~ tan (45 + ~I)
(3.17)
0
So: L=
5.
H - z
tan (450 + ~I
SvKa [FS(p)l
+---~ )
2 tan ¢~
(3.18)
Determine the lap length, IJ, from: II
=
Sv cr~ [FS(p)l 4cr~ tan ¢~
----:--"-----'_.:...,,.c.::.
The minimum lap length should be 1 m.
(3.19)
Retaining walls
f-o!-o-------
L
87
---------;.~I
Geotextile Sand Y1 <1>;
H
...- - - x - - . . . . - j
w
...:.~~~::;:~~~~j~;.:,:.~::~~~ ;;;~~;~.:<\:.;.~~;:~.~!~~.~ :::~.:. ~~' I.'~ \ ~ ., q max
In-situ soil
Fig . 3.7. External stability check
Y2; ~ ; c~
3.3.3. External stability
The external stability check includes checks for overturning, sliding, and bearing capacity failure. They can be done as follows . 3.3.3.1. Check for overturning
1.
Referring to Fig. 3.7, calculate the Rankine active force P aper unit length of the wall as:
P a = ~'YIH2Ka
2. 3.
(3.20)
where Ka is the Rankine lateral earth pressure coefficient = tan 2 (45° - ¢;/2). The active force P a acts a distance of z' = H /3 measured from the bottom of the wall. Calculate the overturning moment, M 0, due to the active force as:
Mo = Paz' Calculate the resisting moment, M as:
R,
(3.21) due to the weight of the wall (3.22)
4.
where W = LH'Y I The factor of safety against overturning can then be calculated as: FS(overturning)
MR = Mo
(3.23)
The factor of safety against overturning should be at least 3. 3.3.3.2. Check for sliding
1.
Calculate the horizontal driving force at the bottom of the wall, F H , as FH
2.
= Pa
(3.24)
Calculate the horizontal resisting force along
FR = Wtan¢~ 3.
Be (Fig.
3.7) as: (3.25)
Check the factor of sliding against sliding as:
FR
FS(sliding)
= FH
(3.26)
88
Geosynthetics and their applications
3.3.3.3 . Check for bearing capacity failure The check against bearing capacity failure can be conducted using Meyerhof's effective area method (Meyerhof, 1953). The check can be
done as follows. 1.
Calculate the eccentricity e due to the forces on the reinforced block, or: e=
2.
MR - Mo --W---'-
(3.27)
The eccentricity should be less than L / 6. The stress q on the soil below the reinforced block can be given as: qmax qmin
3.
L
2" -
=
W(1 ± 6e) L
(3.28)
L
The ultimate bearing capacity qu of the soil with the eccentric loading can be given as (Meyerhof, 1963): (3.29)
The variation of Nc and N"( with ¢~ were given by Prandtl (1921) and Vesic (1973), respectively, by the following relationships: Nc = [tan2
(450+ ~~ )
e1Ctao <1>2 -
1] cot ¢~
(3.3 0)
and: N"( = 2 [ tan 2
4.
(450+ ~~ )
e1Ctan <1>; -
1] tan ¢~
(3.3 1)
Table 3.1 gives the variations of Nc and N"( with ¢~. Calculate the factor of safety against bearin g capacity failure as: FS(bearingcapacity ) =
~ 2: 3
(3 .32)
qmax
Table 3.1. Variations of
¢~ 0 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Nc and N-y with
¢~
Nc
N-y
¢~
Nc
N-y
¢~
Nc
N-y
5' 14 5-38 5-63 5-90 6-19 6-49 6-81 7-16 7-53 7-92 8-35 8-80 9-28 9-81 10-37 10·98 11-63
0·00 0-07 0-15 0-24 0'34 0 -45 0 -57 0-71 0 -86 1-03 1-22 1-44 1-69 1-97 2-29 2-65 3-06
17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33
12·34 13-10 13-93 14-83 15-82 16-88 18-05 19-32 20-72 22-25 23-94 25-80 27'86 30 ·14 32 ·67 35-49 38-64
3'53 4-07 4-68 5-39 6'20 7·13 8-20 9-44 10-88 12-54 14-47 16-72 19-34 22-40 25-99 30 ·22 35 -19
34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50
42-16 46 ·12 50'59 55 -63 61·35 67·87 75 -31 83 '86 93 -7 1 105-11 118-37 133-88 152-10 173·64 199-26 229'93 266'89
41 -06 48-03 56 -31 66-19 78 -03 92-25 109-41 130-22 155-55 186-54 224-64 271 '76 330 ·35 403 '67 496·01 613'16 762 ·89
Retaining walls
89
Example 3.1 A geotextile-reinforced retaining wall is 6 m high. GiYen 1'1 = 16kN/m 3 , ¢'I = 32°, 1'2 = 17·2kN/m 3 , ¢; = 20°, = 30kN/m 2 , aG = 15kN/m. Determine Sy , L , and 'I' Use FS(B) and FS(p) to be 1'5, ¢~ ~ 2/3¢;.
c;
Solution
Determination of Sy.
S _ y -
From equation (3.11):
aG
b lzKa) [FS(B )l
We will make a few trials to find Sy. At z = 2m: 15 Sy
= (16)(2)(0.307)(1.5) = 1·02 m
Atz = 4m: 15 Sy
= (16)(4)(0'307)(1'5) = 0·51 m
At z = 6m: 15 Sy = (16)(6)(0'307)(1'5) = 0·34m
= OAm from z = 0 to 4m and
For safety, use Sy
Sy
z = 4m to 6m (Fig. 3.8).
Geotextile Geotextile
•
• 4m
:- - - - L = L1 = 3·6 m - - I I
I
I I I
•
r=:-2m
1m
I I
--l
•
I
: ~
I
I
•
Geotextile
:,:;,~
II
•
J
I03m:- -----
: • r--L = L2 = 2·5m
:--
I
-
Pa
rll
I
I I
0'3m
Fig.3.B . Example 3.1
<1>; = 20° d2 = 30kN/m 2
Note: W1 = 4L'Y1 W2 = 2L 2Y1
= 0·3m from
90
Geosynthetics and their applications
Determination of L.
From equation (3.18):
H - z L=
(
tan 45°
+
+¥
( 6-
'/'1)
Z
32
0
SyKa[FS(p)l 2tan ,/,1 'f'F
+ SY(0(~7)( J
)
tan 45° + 2
2 tan
"3 x
0
~ 0554(6 _ z) + o 59Sy
5 )
32°
Now the following table can be prepared: z: m
Sv : m
L: m
0-4 2·0 4·0 6·0
0-4 0-4 0-4 0·3
3·56 2-45 1·35 0·18
For safety, use L = 3·6m for z::; 4m and L = 2·5m for z = 4m to 6m .
Determination of I,. I, =
Sy G~[FS(p) l 4G~ tan
From equation (3 .19): SyKa[FS(p)l
= ---,---'----,'--'..:.. 4 tan
At z = 0-4m: I, = (0-4)(0·307)( 1·5) ~ 0.12m 4 tan (~ x 32°)
Use I, = 1 m (see Fig. 3.8) . Example 3.2 For the retaining wall shown in Fig. 3.8, calculate the factor of safety against overturning, sliding, and bearing capacity failure. U se the soil and geotextile parameters given in Example 3.1.
So lution Factor of safety against overturning.
' H2 tan 2 (45 0 P a = 21' =
From equation (3 .20):
(~) ( 16)(62 ) tan 2 (450 _ 3~0 ) ~ 88·5 kN/ m
I H 6 z =-= - = 2m 3 3 Taking the moment about the bottom of the wall:
Mo = Paz' = (88·5)(2) = 177 kN-m/m
L,
M R = (4
X
L,
X
=(4
X
3.6
X
1',) 2
+ (2 X
L, L2 X 1',) 2
16) (3~6) + (2 X 2·5 X 16) ( 2~5)
= 514·7 kN-m/m FS(overturning) =
514·7
177 =
2·91
(Answer)
Retaining walls
91
From equation (3.25):
Factor of safety against sliding.
+ W2 ) tan(~¢D [(4 x 3·6 x 16) + (2 x 2·5
F R = (WI
=
x 16)] tan(~ x 32°)
= 121·2kNj m From equation (3.26): FS(sliding)
FR 121·2 = F H = 88 .5 ~ 1·37
(Answer)
Factor of safety against bearing capacity failure. L2
e=2-
MR -Mo WI + W 2
2·5 2
From equation (3 .27):
514·7 - 177 (4 x 3·6 x 16) + (2 x 2·5 x 16)
=0 ·16 From equation (3.28): = WI q max
(1 + 6e)
2 +W
L2
L2
= [(4 x 3·6 x 16)
+ (2
x 2·5 x 16)] (I
+ 6 x 0.16)
2·5
2·5
= 171 ·8kNj m2 From equation (3.29): qu
=
c;Nc + h '2(L 2 - 2e)N,
For ¢; = 20°, from Table 3.1 , Nc = 14·83 and N , = 5·39. qu = (30)(14·83) FS(bearing capacity)
+ HI7 '2)[2'5 -
=
545·95
17"f8 =
(2 x 0'16)](5'39) = 545·95 kN j m 2 (Answer)
3·18
Comments
In all practicality, since FS(overturning) and FS(sliding) are less than 3 and 1'5, respectively, it will be necessary to redesign the retaining wall. This can be done by making LI = L2 = 3·6 m (see Fig. 3.9). With this assumption: Mo = 177kN-m/m
MR = (6 x 3·6 x FS(overturning)
16)(3~6)
= 622kN-m/m
622 = 177 = 3· 5
(Answer)
F R = (6 x 3·6 x 16) tan(~ x 32°) = 135 kN j m FS(sliding)
3·6
e=
2 -
135 = 88.5 = 1·5 622-177 6 x 3·6 x 16 = 0·51
. = (6 x 3·6 x 16) (1 3.6
qmax
(Answer)
L
< "6
0.51) + 6 x3.6
= 177'5kNj
m
2
92
Geosynthetics and their applications
~t;
0'4m:
>-~=-----0'4 m--+---', Y, = 16 kN/ma~
!,'_ _ _ _ _ _ _ _...l-~, , I
•
I I
•
I
,
It-o-> ------- 3·6 m
<1>; = 32°
;\\t
I I
, I
I
I -----I~~I
I
, I
4m
: I
,I ,,
• •
I I
r--:-
TO'3m
I
•
'-- 1 m---t
I I
2m
: ..0 - - - - ~
3·6 m
•
- - ----.j
•
I
0·3m
<1>; = 20° c; = 30kN/m 2
Fig. 3.9. Example 3.2
q u = (30)(14·98)
+ ~ (17'2)[3·6 569
FS(bea ringcapacity)
3.4. Design procedure for retaining walls with geogrid reinforcement
= 177.6 = 3·2
(2
X
0'51)](5·39) = 569 kNj m2
(Answer)
T he design of retaining walls with geogrid reinforcement, as shown in Fig. 3.2(c), can be done using procedures similar to those described in Section 3.3 with the following modifications: (a) (b)
In equation (3.11), O'G should be replaced by the allowable tensile strength of the geogrid. In equation (3.25), ¢~ may be replaced by ¢', .
Designing geogrid-reinforced retaining walls using Rankine's active earth pressure has yielded good results. Many retaining walls withstood the 17 January 1995 Nyogoken-Nambu earthquake in Japan with minimal damage. Recent studies (FHW A, 1995) showed that the lateral earth pressure behind a geogrid-reinforced retaining wall may need to be modified for future design considerations, or: (3.33) where: K = a Ka = a tan 2 (45° - ¢;)
(3.34)
The variation of a with depth of the wall is shown in Fig. 3.10.
3.5. Concluding remarks
This chapter presents the general procedures of designing retaining walls with geotextile and geogrid as reinforcement in granular backfill. These are flexible retaining walls that are easy to construct and that can withstand earthquake forces without undergoing total coll apse. Further field and laboratory studies are now underway in many countries to
Retaining walls
93
o r----,---c...,=---C( =K-
Ka
E (ij ~ Q)
=
" c.
B Q)
:5 ~
o
6
------
Qi
.0
.r:
15. Q)
a Fig. 3.10. Variation of a with depth
evaluate the exact nature of lateral earth pressure distribution behind this type of retaining walls, which will lead to more economical designs.
References Federal Highway Administration (FHWA) (1995). Mechanically stabilized earth walls and reinforced soil slopes design and construction guidelines. FHWA Washington, D.C. No. FHWA-SA-96-071. Laba, J. T. and Kennedy, J. B. (1986). Reinforced earth retaining wall analysis and design . Canadian Geotechnical Journal, 23, No.3, 317- 326. Meyerhof, G. G. (1953). The bearing capacity of foundations under eccentric and inclined loads. Proceedings of the 3rd International Conference on Soil Mechanics and Foundation Engineering. pp. 440- 445. Meyerhof, G. G. (1963) . Some recent research on the bearing capacity of foundations. Canadian Geotechnical Journal, 1, 16- 26. Prandtl, L. , 1921. Uber die eindringungsfestigkeit (harte) plastischer baustoffe und die festigkeit von schneiden. Zeitschrift fur angewandte Mathematik und Mechanik , 1, No.1, 15- 20. Vesic, A. S. (1973). Analysis of ultimate load of shallow foundations. Journal of the Soil Mechanics and Foundations Division, ASCE, 99, No.1 , 45- 73.
4
Embankments E. M.
PALMEIRA
Department of Civil Engineering , University of Brasilia , Brazil
4.1. Introduction
Geosynthetics can be very attractive for works involving embankments built on soft foundation soils. Basically, layers of geosynthetics can serve as reinforcing materials or can accelerate the process of consolidation of the soft subgrade. The former function usually aims for a temporary increase of the safety factor of the embankment which is associated with a faster rate of construction or the use of steeper slopes that would not be possible in the absence of the reinforcement. The latter function can also be associated with the need for a more stable embankment or staged construction but also in order to accelerate the consolidation settlements. Figure 4.1 summarizes the usual functions of geosynthetics in embankments on soft soils. An additional benefit brought about by the presence of the reinforcement is to provide separation between good quality fill materials and the fine grained foundation soil, as shown in Fig. 4.2(a). This is achieved when the reinforcement is also a filter for the foundation soil, as may be the case for non-woven geotextiles, or when a geosynthetic filter is used in conjunction with the reinforcement. The presence of the reinforcement also reduces the consumption of fill material because it minimizes or avoids local failure mechanisms caused by construction equipment during transport, spreading and compaction of the fill material (Fig. 4.2(b)). Quite a few works can be found in the literature dealing with the design of embankments on soft soils using geosynthetics. In this chapter, the main contributions from these works are presented and discussed along with drainage aspects.
4.2. Geosynthetics as a basal reinforcement in embankments
4.2.1. Reinforcement roles and aspects to be considered in the analysis
The use of geosynthetic reinforcement can significantly increase the factor of safety of the embankment. Its use is particularly attractive for low ratios between foundation soil thickness and embankment base width (say, less than 0'7). For thick foundation soils, the contribution of the reinforcement can be less significant, consisting mainly of the provision of a rough boundary between the base of the embankment and the foundation soil. This boundary will only marginally increase the overall factor of safety, depending on the variation of undrained strength with depth in the subgrade. Figure 4.3 shows the possible failure mechanisms that may occur in an embankment built on soft soil. Figure 4.3(a) shows the possibility of failure inside the embankment, which may occur for very steep embankment slopes on reasonably strong subgrades. This type of mechanism has to be predicted using a stability analysis and it is not the most critical
96
Geosynthetics and their applications
Favourable stress distribution to the soft soil Factor of safety increases
Steeper slopes Reinforcement
Fig . 4.1. Contributions from geosynthetics in embankments on soft soils
Rate of consolidation increases if draining material
Geosynthetic
Fill
(a)
Fig . 4.2. Beneficial effects from the reinforcement during construction : (a) sepa ration ; and (b) reduction of local failures during construction
Geosynthetic (b)
mechanism for situations of soft soi l foundations. Figure 4.3(b) presents, schematically, a mechanism for expulsion of the soft soil laterally. This mechanism may occur for heavily reinforced embankments on thin foundation layers. The situation most commonly considered is shown schematically in Fig. 4.3(c), where the failure mechanism is characterized by a well-defined failure surface (or region) cutting the fill, the reinforcement and the soft soil. This mechanism can involve tensile failure of the reinforcement or bond failure due to insufficient anchorage of the reinforcement extremity beyond the failure surface. The usual practice for the design of reinforced embankments on soft soils is the use of limit equilibrium methods to estimate safety factors for the situations described above. Simple methods are usually preferred for preliminary routine analyses, such as the ones dealing with sliding wedges or circular fai lure surfaces. For the latter case, Fellenius and the Modified Bishop's methods are commonly used. Figure 4.4 raises some important questions to be considered when analysing the stability of reinforced embankments on soft soils. The first question relates to the magnitude and inclination of the mobilized reinforcement force. The reinforcement layer is initially placed horizontall y at the embankment base. As the embankment and the reinforcement deform, some inclination of the reinforcement force to the horizontal
Embankments
97
Geosynthetic reinforcement
(a)
Geosynthetic reinforcement
(b)
Geosynthetic reinforcement
Fig. 4.3. Instability mechanisms in an embankment on soft soil: (a) internal stability; (b) soil expulsion; and (c) overall stability
(c)
Tangential
Bisectorial
Fig. 4.4. Reinforcement tensile force orientation
direction is expected. One can find works in the literature where the inclination of the reinforcement force can be taken into account in the analysis (Low et al., 1990; Sabhahit et al., 1994; Kaniraj , 1996). Usually horizontal, tangential or bisectorial forces are considered. The consideration of this inclination in the stability analysis will affect the results of calculated safety factors. Large-scale direct shear tests on stiff granular materials (Palmeira and Milligan, 1989) suggest that, at peak strength, the contribution of the reinforcement deviation from its original direction to the stability is negligible. That is likely to be the case for stiff fill materials on soft foundations when the first sign of instability appears (cracks along the embankment surface, for instance) . The influence of the inclination of the tensile reinforcement force will be significant at large strain conditions, when usually the embankment would be considered to be compromised in operational terms. An additional difficulty, regarding the use of inclined reinforcement forces , is the impossibility of predicting the force inclination to the horizontal during embankment loading at the present stage of knowledge. Besides, the consideration of
98
Geosynthetics and their applications
a constant inclination of the reinforcement force would not be rea listic or cinematically consistent. On the other hand, the inclination of tangential and bisectorial forces to the horizontal will depend on the slip surface considered because, in both cases, the force inclination will depend on the inclination of the line tangent to the slip surface at the intersection with the reinforcement layer. At the present state of knowledge, the use of horizontal reinforcement forces in stability analyses of reinforced embankments is recommended and the accuracy of this assumption has been observed by back-analyses of reinforced embankments led to failure , and by limit analyses (palmeira et al. , 1998; Michalowski, 1998). Depending on the polymer, mobilized tensile geosynthetic reinforcement forces can be more or less affected by the rate of strain imposed during loading. This rate is a function of the speed of embankment lifting. Therefore, for those reinforcements significantly affected by the rate of strain, the tensile force- strain- time relation has to be well known for an accurate prediction of stability conditions. Another aspect to be considered in the stability analysis of reinforced or unreinforced embankments is the influence of fill material cohesion on the results of the safety factors obtained. Large fill cohesions or large slice inclinations to the horizontal inside the fill can cause negative normal forces at the base of slices in that region , which may cause misleading values of factors of safety. Discussions and ways of treating this type of problem can be found elsewhere (Whitman and Bailey, 1967; Chirapuntu and Duncan, 1975; Palmeira and Almeida, 1980). When the cohesive fill material is expected to crack throughout its entire height, it is wise to treat it simply as a surcharge on the foundation soil surface with no contribution in terms of resisting the failure mechanism. For large deformations of granular fill materials, critical state strength parameters should be used for this material in the calculations. Different methods of analysis or forms of definition of the safety factor may also affect the result obtained. For circular failure surfaces, depending on the method employed, the factor of safety may be defined from force equilibrium conditions or ratios between resisting and driving moments. In addition, for Bishop's method, for instance, the contribution from the reinforcement force can be considered in two ways - simply taking into account, in the analysis, the resisting moment of this force with respect to the centre of the circular surface, or taking the components of the reinforcement force (normal and tangential to the slice base) into the equilibrium equations for the slice. The latter approach explores the increment in frictiona l resistance caused by the increase in the normal force at the base of the slice plus the component of the reinforcement force along the slice base direction. These two different ways of considering the effect of the reinforcement force described above may yield some difference between the results for the safety factor obtained in each case. The analysis of reinforced em bankments using traditional methods for slope stability analysis consists of trying to find the critical failure surface for which the necessary force in the reinforcement is maximum. Usually, a target value for the safety factor is established and the maximum necessary reinforcement force is determined by searching the critical failure surface.
4.2.2. Design approaches for reinforced embankments 4.2.2.1. Designing against expUlsion of the foundation soil Figure 4.5 shows an approach commonly used to verify the possibility of a saturated foundation soil expulsion under undrained conditions. The
Embankments
99
Fig. 4.5. Expulsion of the soft subgrade
stability analysis consists of checking the stability of the foundation soil block below the embankment slope, as illustrated in Fig. 4.5. The factor of safety against soil expulsion can be estimated from:
F.e = Pp
+P RB + RT
(4. 1)
A
where Fe is the safety factor against foundation soil expulsion, Pp is the passive reaction force against block movement, RT is the force at the top of the soil block, RB is the force at the base of the soil block, and P Ais the active thrust on the soil block . The active and passive forces can be evaluated by earth pressures theories (Rankine's, for example), while the forces at the base and top of the soil block can be estimated as a function of the undrained strength at the bottom of the foundation soil and adherence between the reinforcement layer and the surface of the foundation soil, respectively. For a foundation soil with the undrained strength varying with depth (Fig. 4.5), equation (4.1) yields to:
F. - (0'51'r D + 2Sum )D + (>Suo + Sub )nH e (/'H + q + 0'51'r D - 2S um )D
(4.2)
where I'r is the unit weight of the foundation soil, D is the thickness of the foundation soil, Sum is the average undrained strength along the depth of the foundation soil; .A is the ratio between the shear stress mobilized at the soft soil - reinforcement interface and the undrained strength at the foundation soil surface (0 < I' ::; 1), Suo is the undrained strength at the foundation surface, Sub is the undrained strength at the bottom of the foundation soil, n is the slope inclination (Fig. 4.5), H is the embankment height, q is the uniformly distributed surcharge on the embankment platform, and I' is the unit weight of the fill material. In a real situation, the failure mechanism is rather more complex than the approach above because drainage is likely to occur at the top and bottom of the soft soil block, depending on the type of reinforcement used and on the type of soil below the soft layer. In this sense, the approach presented above is conservative. Rowe and Li (1999) discuss the effect of partial drainage of the soft soil on the stability of the embankment using the finite element method. 4.2.2.2. Generalized failure of the embankment
In this case, several approaches for the stability analysis of the embankment can be considered. Below, some of these approaches are presented and discussed. Combined failure surface (Jewel, 1987)
Jewell (1987) presents a methodology of stability analysis where the slip surface combines a circular arc in the foundation soil and a straight
100
Geosynthetics and their applications
Fig . 4.6. Combined failure mechanism (adapted from Jewell, 1987)
z
line inside the embankment, as shown in Fig. 4.6. The effect of the embankment on the stability is represented by the active thrust acting on the vertical CD in Fig. 4.6. The active thrust can be estimated by the Rankine earth pressure theory, for instance. One advantage of this approach is to allow the identification and analysis of a cracked cohesive embankment in a clearer and easier way. The depth of the crack will affect the value of the active thrust, as in usual retaining wall problems with cohesive ba~kfills . The factor of safety for the unrein forced embankment can be written as: Mr Fo= -
(4.3)
Ma
where Fo is the factor of safety of the unrein forced embankment, Mr is the sum of the moments of the resisting forces with respect to the centre of the circular portion of the slip surface, and M a is the sum of the moments of the driving forces with respect to the centre of the circular portion of the slip surface. The procedure consists in the calculation of the factor of safety for several slip surfaces in order to obtain the overall minimum value for the factor of safety. For cases with no surcharge on the embankment platform, the critical circle is located on the vertical line passing through the middle of the embankment slope (Leshchinsky, (987). For the reinforced case, the moment equilibrium equation for the sliding mass can be written as:
Mr Fr
+ Td T = M a
(4.4)
where Fr is the factor of safety for the reinforced embankment, T is the required tensile force in the reinforcement, and d T is the arm of T with respect to the circle centre (Fig. 4.6). Combining equations (4.3) and (4.4), yields:
T =
(I _FoFr ) Ma d
(4.5 )
T
Equation (4.5) is easier to use in the sense that if the value of Fr is established for the slip surface considered and the value of Fo for that surface is known , then the value of T can be directly obtained from equation (4.5), knowing the values of d T and M a As in the case for the unreinforced embankment, the values of T for several surfaces have to be calculated in order to determine the maximum required reinforcement
Embankments
101
tensile force for the critical surface in the reinforced case. It is important to point out that the critical slip circle is not necessarily the same in the reinforced and unreinforced cases. For a foundation soil with the undrained strength varying linearly with depth (Fig. 4.6), equations (4.3) and (4.5) yield to:
Fo
=
R2[a(Suo - pRcos(a/2)) + 2pRsin(a/2)] Ed + Wd + Qd E w Q
(4.6 )
and
(4.7)
where R is the radius of the circular portion of the slip surface considered, a is the internal angle defining the circular portion of the slip surface, Suo
is the undrained strength at the foundation soil surface, p is the rate of increase in the undrained strength with depth, E is the active thrust in the embankment along the vertical CD, W is the the weight of the wedge ABCD in the embankment, Q is the resultant force due to surcharge on the embankment surface (if any) along the length BC, and dE, d w and dQ are the anns of the forces E , Wand Q with respect to the centre of the circle, respectively (Fig. 4.6). Low et al. (1990) approach
Low et al. (1990) presented a methology for the design of reinforced embankments using circular failure surfaces. Mathematical expressions are presented for the determination of the relevant parameters of the problem and a chart for the determination of the value of the tensile reinforcement force required is also provided . The geometry of the problem treated by Low et al. is presented in Fig. 4.7. The maximum required tensile force in the reinforcement to guarantee a target safety factor (Fr) for the critical surface, for all the circles tangent to the horizontal line at the depth z, is given by:
T
=
(I _Fo) Fr
(4.8)
rH2 IR
where Fo is the critical factor of safety for circular surfaces tangent to the same depth z in the unreinforced case, r is the fill material unit weight, H is the embankment height, and IR is the stability number that can be obtained from Fig. 4.8, as a function of z and the inclination of the slope of the embankment (n).
Mid-slope vertical line
H
z Soft soil
Fig. 4.7. Low et al. approach (adapted from Low et aI., 1990)
/
102
Geosynthetics and their applications
1A 1·2 10
/
n
3
0·8
'R 0·6
OA Fig. 4.8. Stability number for reinforced embankments horizontal reinforcement force (adapted from Low et aI., 1990)
0·2 0 1·0
0·5
2·0
3·0
4·0
zlH
The depth z of tangency of the failure surfaces must be varied along the foundation soil thickness for the determination of the overall maximum value of T. The value of Fo at each depth considered for the analysis must be known in order to solve equation (4.8) for different failure surfaces. Low (1989) presents a solution for the determination of the minimum factor of safety for the unreinforced case among all circles reaching a established depth (z). In this case, the value of Fo is given by: Sueq
Fo = N, "(H
c + N2 ( "(H + Atan ¢ )
(4.9)
where N" N2 and A are the stability numbers obtained from Fig. 4.9 as a function of the inclination of the embankment slope (n) , height of the embankment (H) and depth reached by the circle (z), Sueq is the equivalent undrained strength of the foundation along the depth z, "( is the fill unit weight, and c and ¢ are strength parameters of the fill material.
OAO
7.0
5" 60
Nj
4_ 3
5·0
21
4·0
-
n
~ -;;--
-
0·35 ,
030
f... 0·25
l' 0·20
3·0
n
3 5 -
\ \\
2·0
1~
N2 1·0
o
0·15 0
1·0
3
4
zlH
~
o
2
---
20
3·0
4·0
5·0
r I",,_._ _, _"_~_~_, ~,,_,,_J..z
zlH
Fig. 4.9. Stability numbers for unreinforced embankments (adapted from Low, 1989)
I!:
5
Embankments
................ , ......... , .. . ............. .......... ,
.. ····· rM ,
' "'of
S'uo
,
•••••• ,' ............ ".
:/,;
/.
I"
103
"' S'uo
"I"
"I
~
z
Fig. 4.10. Typical soft soil undrained strength profile (adapted from Low, 1989)
For the common condition of undrained strength profile, as for the one shown in Fig. 4.10, the value of Sueq can be obtained by the following expression:
()
I Zc I Sueq = 0·35S uo + 0'65S uz + 0·35 ~ '" .6.Suo
(4.10)
where S~o, Suz, .6.S~o' Zc and Z are defined in Fig. 4.10. The consideration of other forms of undrained strength profiles is presented in Low (1989). The radius of the critical circle (minimum Fo) in the unrein forced case for a given depth (z) of tangency can be obtained by: Ro=
[0'1303 (;'+1) +1 .5638(~+0.5)1H 2Z+ H -+0,5
(4.11 )
H
where Ro is the radius of the critical circle tangent to the horizontal line at the depth z. For the reinforced case, the radius of the critical circle (Rr ) among all the circles tangent to the horizontal line at the depth z is given by:
3' 128(a- ~ H ~) '"'( H 2 Rr =
(-z+ 0 , 5 -T- ) H
H 2
z+ H
(4.12)
'"'( H2
with:
= ~( ~ 0'5) 2 + (n 24+1) a 2 H + 2
(4.13 )
Kaniraj (1994) approach
Kaniraj (1994) presented a methodology similar to the one presented by Low et at. (1990) . However, the presence of a drainage channel at the embankment toe, berm and cracks at the embankment surface (cohesive fills) can be considered in this case. The geometrical conditions for the problem are presented in Fig. 4.11. For the unreinforced case, the coordinates of the critical circle centre among all the circles tangent to a horizontal line at depth z are given by: Xo n . Wx - = --k,k 2 + H 2 '"'( H 2
(4.14 ) (4.15)
104
Geosynthetics and their applications
y
Q(Xo, Yo)
b
I"
Cracked zone
0Wx
N
Fig. 4.11 . Geometrical characteristcs in Kaniraj's approach (adapted from Kaniraj, 1994)
{
Xx M
X (0,0)
:::. Firm. ~.oi! ·::::::::::::::::::::::::::::::::::::::::::::: :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: :
where Xo and Yo are coordinates of the critical circle centre for the depth z in the unreinforced case, k t is the ratio between berm height and embankment height (Fig. 4.11), k2 is the ratio between berm platform width and embankment height, and Wx is the soft soi l weight removed due to the excavation of the drainage channel. The value of at in equation (4.15) is given by: I
at
=
at
(4. 16)
2
(3 - + (3 - H 2 Z
with:
(4.17 ) p, = ktk2(n
+ k2)(1
(4.18)
- kt)
and
(3 = 1 _ He (4.19) H where p, is a berm factor (p, = 0 for embankments with no berm), and (3 is the ratio between height of the non-cracked part of the embankment and the total embankment height (Fig. 4.11). The minimum factor of safety for all circles tangent to the horizontal line at the depth z in the unreinforced case is given by:
Fo
= SrN; + SeN~
(4.20)
with: _ Su Sr - "(H
Se = -
c
"(H
,\ = 0.19
(4.2 1)
+ ,\ tan ¢ +
0·02n
z/H '
(4.22) for z/ H
~
0.5
(4.23 )
Embankments
105
where Sf and Se are the normalized strength parameters for the soft soil and fill materials, respectively, Su is the soft soil undrained strength, c and ¢ are fill shear strength parameters, and N; and N~ are stability numbers given by:
(4.24)
and: (4.25 ) with:
(3 ) 0·53
]
m = 0·5 [ ( 1 + z/ H
(4.26)
- 1
For the reinforced case, Kaniraj presents solutions for horizontal, tangential and bisectorial tensile reinforcement forces. For the horizontal reinforcement force , the coordinates of the critical circle centre for a depth of tangency z are given by equation (4.14) (for Xo) and by:
(~ )'-47_3'128(z;a)(~ r-47-2' 128(~~)
= 0
(4.27)
with: (4.28 ) (4.29)
~)
Al = 3·06 (
0.53
(4.30)
B I = 1·53 [( ~+
+ -p, 2
(3 )
0.53
-
(
~
)0'53 ]
(4.31)
Wx (Xx + -n- kl k2 + -1 -Wx)] -
--?
, H-
-
H
2 , H2
2
(4.32)
where a is the elevation of the reinforcement layer with respect to the soft soil surface (Fig. 4.11) and Fr is the required safety factor for the reinforced embankment. The reinforcement tensile force required for a factor of safety equal to Fr for the reinforced embankment is given by:
T =~ H 2
(4.33)
L a/ H '
with:
(32 )Y~-~-~ (Y)I-47 ~
z
~= F ( -+(3-r H 2
Yo
z+ a
----
H
H
H
.
H
(4.34) (4.35)
106
Geosynthetics and their applications
For the reinforced case where the force in the reinforcement is considered tangent to the circle at the intersection between the failure surface and the reinforcement layer, the value of Xo is also given by equation (4.14), while Yo is given by:
(FI )0'68
Y ~ = 1·672 --.2 H F,
(4.36)
The required reinforcement force is also given by equation (4.33) but with La given by:
(4.37) After the calculations and determination of the critical circles, one has to verify if the geometrical conditions presented in Fig. 4.10 have been satisfied . These conditions are expressed mathematically by the expressions below.
Condition 1: The circle centre must be located above, or at the bottom level of, the cracked zone in the fill. So:
z H 2 (3+ H
Yo
(4.38)
Condition 2: Both the berm and the drainage channel must be entirely inside the sliding mass bounded by the circular slip surface and the soil surface boundary. So:
n Wx Xs - - k,k2 +--2 +- ) 2] ( Yo> 0.5 ~ + 2 "(H H H H r H
(4.39)
~
where Xs is the distance along the horizontal between the extremity of the channel and the Y axis (distance between points Sand E in Fig. 4.11).
Condition 3: The extremity of the failure surface (point I' in Fig. 4.11) must lie on, or beneath, the embankment platform length. So: nH ::; XI' ::; nH
+b
(4.40)
or:
(4.41 ) and: '!:.+k'k 2
Yo 1 2 -<- [( H -2
-
W~ +~)2 H
"(H
(3+~
Z + (3 +H
1
(4.42)
H
where XI' is the abscissa of point I' and b is the embankment platform width (Fig. 4.11). Analytical solution (Jewel/, 1996)
Jewell (1996) presented an analytical solution considering the equilibrium of the soft soil block below the embankment slope, as illustrated in Fig. 4.12, in a similar way to the analysis of soft soil expulsion presented
Embankments
107
n a S uo
yH
F
x
nH
Fig. 4.12. Jewel/ 's stability analysis approach (adapted from Jewel/, 1996)
z
z
2Suz F
before. The equations to be used depend on the form of variation of the undrained strength of the soft soil with depth. So, for a foundation soil with uniform strength and limited depth: Fo =
~ [8D + 2nH]
(4.43 )
~ [4 + (1 + a) nH] ,H D
(4.44)
,H
Fr =
2D + kaH
2[
T =,H
a nD ka] 4D+(I +a )nH+2
(4.45)
For a foundation soil with strength increasing with depth: Fr or Fo = T
= , H2
Zcrit
-Suo [pnH 4 + -S + 2 ,H
uo
[ans
ka]
uo + Fr, H 2
(1 + a) SuonH
=
2p
2(1
+ a) pnH] Suo
(4.46) (4.47)
(4.48)
Jewell suggests the use of the expressions above under the condition:
~~ -> 6 Su
-
(4.49 )
where Fo and Fr are the safety factors for unrein forced and reinforced embankments respectively, T is the required reinforcement force , Su is the soft soil undrained strength, n is the slope of the embankment, a is the ratio between mobilized shear stress and undrained strength at the subgrade surface (Fig. 4.12), H is the embankment height, is the Rankine's active earth pressure coefficient of the fill material, D is the thickness of the soft subgrade; , is the fill unit weight, Suo is the undrained strength at the surface of the subgrade, p is the rate of increase of undrained strength with depth, and Zc rit is the critical depth of the sliding block. The value of a for the reinforced case is established by the designer (0 < a :S 1). For the unreinforced case, the value of a is given by the expression below and the solution is obtained by trial and error in combination with equation (4.46):
ka
(4.50)
108
Geosynthetics and their applications
4.2.3. Choice of the reinforcement After the determination of the maximum required reinforcement force to guarantee the stability of the embankment, the reinforcement must be chosen carefully. Several factors have to be taken into consideration, such as embankment characteristics, consequences of an embankment failure , deformation allowances, serviceability requirements , reinforcements available, etc. With regard to emba nkments on soft soils, the following factors may be of major concern when choosing the reinforcement:
• • • • •
tensile strength and stiffness soil-reinforcement bond characteristics creep characteristics geosynthetic resistance to mechanical damage durability.
With regard to tensile strength, the reinforcement chosen has to attend to the following condition (Jewell, 1996):
T = T ref > T d
(4.51 )
f-
where Td is the reinforcement design strength, T ref is the reference tensile strength of the reinforcement obtained under appropriate testing techniques, taking into consideration the design life of the project (e.g. creep tests) , f is the reduction factor for the tensile strength, and T is the required reinforcement force. The value of f can be determined by the following expression (Jewell, 1996):
f = fm fmd fenv
(4.52)
where fm is the reduction factor to account for reinforcement material uncertainties, fmd is the reduction factor to account for mechanical damage during handling and construction, and fellV is the reduction factor to account for strength losses caused by environmental factors (chemicals, biological factors , etc.). The value of the required reinforcement stiffness (J) is still difficult to evaluate, mainly because the design methods for reinforced embankments are based on limit equilibrium analyses. Depending on the serviceability requirements for the embankment, the reinforcement tensile strain is expected to be within 2 to 10%. Therefore, a crude way to estimate the required reinforcement stiffness is to divide the required reinforcement force by an expected strain within that range. Jewell (1996) presents the expression below to estimate the reinforcement stiffness for the case of an embankment on a foundation soil with uniform strength and limited depth (for a 2: 0): G
[
.
J = (I-a)Sud kad,,/H
2( nH Lc ) 3D + 2D
(nH )2]
+ a Sud D I + D
(4.53 )
where G is the shear modulus for the foundation soil, Lc is the embankment crest width, k ad is the design active earth pressure coefficient of the fill material, and Sud is the design undrained strength of the foundation soil. For the use of equation (4.53), the value of a is established by the designer (0 < a :::; 1) and the value of Sud is equal to Su i Fn with Fr given by equation (4.44). The value of kad is obtained by the Rankine earth pressure theory using the design value for the fill material friction angle.
Embankments
109
Large bond values between soil and reinforcement are also required for an appropriate load transference between materials and to provide enough anchorage resistance for the reinforcement. Values of adhesion and friction angle between soil and reinforcement can be evaluated using appropriate testing procedures (direct shear and pullout tests, for instance). Reinforcement creep characteristics and durability may be relevant or not, depending on how long the reinforcement will be required to guarantee the embankment stability. In most cases, the reinforcement is only required during embankment construction and for a short period afterwards because the strength of the foundation soil increases with time due to consolidation. However, in cases where the reinforcement is required for long-term stability reinforcement, creep behaviour and durability must be carefully considered in the design.
4.2.4. Anchorage length of the reinforcement The methodologies presented above allow for the estimation of the required reinforcement force in order to obtain a desired factor of safety for the reinforced embankment. In addition, the anchorage conditions of the reinforcement extremities must be evaluated in order to avoid slippage of the reinforcement along the anchorage length, as presented schematically in Fig. 4.13. From this figure, the factor of safety against bond failure along the anchorage length can be estimated by: F
_
L a nch (aS uo
anch -
+ a sr ) + W tan 8sr T
(4.54)
where F anch is the factor of safety against anchorage failure, L anch is the anchorage length beyond the critical failure surface, a is the ratio between mobilized shear stress on the reinforcement and the undrained strength at the subgrade, a sr is adhesion between fill material and reinforcement, W is the fill weight above the anchorage length, 8sr is the interface friction angle between fill material and reinforcement, and T is the required reinforcement tensile force . A similar approach must also be used to verify the anchorage condition along the length AI in Fig. 4.13.
4.2.5. Additional remarks on analysis and design It is well recognized that limit equilibrium methods have important limitations with regard to model failure mechanisms realistically. Several factors are not taken into account in limit equilibrium analyses, such as stress- strain relationships for the materials involved, progressive failure and time-dependent factors. However, when used properly and with accurate input data, limit equilibrium stability analysis methods can be
Fig. 4.13. Reinforcement anchorage conditions
::: Firm soil
110
Geosynthetics and their applications
useful tools for the estimation of safety factors of reinforced embankments on soft soils. Palmeira et al. (1998) presented back-analyses of case histories of reinforced embankments on soft soils that led to failure (Volman et aI., 1977; Rowe and Soderman, 1984; Delmas et aI. , 1990; Loke et al., 1994; Rowe et al. , 1995; Schaefer and Duncan, 1988) using the Modified Bishop's method, the United States Army Corp of Engineers' method (USACE, 1970) and the method presented by Jewell (1996). These methods were used to predict the factor of safety at the failure height of the embankments in the field. Figure 4.14 presents some details of the case histories analysed . The data on the materials and reinforcements used were obtained from the original sources. Table 4.1 shows the predicted values of safety factor for the case histories at 100 m
•I ::::::::::::::::::'- 1
Geotextile
4·2m
I
Sandyfill
~~~~~~~~
Geotextile
2m 45 m
-:-_~~::HT<~!'i'~>T
j3'9m
Peat
•/
'
or~~nic, CI:
,C
y,
'
T
2·75 m
1
3 '3 m
(b)
(a)
Reinforcement
Su (kPa)
,rrr 60
. ','/
~.
-;//
Clay /:
/.
/;
18
24
z(m) (c)
Reinforcement layers
5m
.. ..... .. .. . _ Su (kPa) . . . . . .... Fill. . . .. ~ 4 2 m 0 60
.-.-.~.~~:~~
I" Om:[\ z (m)
Embankment 4RA
Embankment 4RB
(d)
Reinforcement Reinforcement
5m
Su (kPa)
~~~~~~~~~~,:~ z(m)
(e)
(f)
Fig . 4.14. Trial embankments led to failure (adapted from Palmeira et aI. , 1998): (a) case history 1, Volman et al. (1977); (b) case history 2, Rowe and Soderman (1984); (c) case history 3, Delmas et al. (1990) ; (d) case history 4, Loke et al. (1994); (e) case history 5, Rowe et al. (1995); and (f) case history 6, Schaefer and Duncan (1998)
Embankments
111
Table 4.1 . Comparisons between predicted and observed safety factors at failure for trial reinforced embankments (modified from Palmeira et aI. , 1998) Case history
MBF *
MBM*
USACE
Jewell (1996)
1t
1·012 0·810 0·977 1-181/1-116~
0·979 0·870 0·982 1-159 1·059 1·036 1·035
0·959 0·911 0·970 1·179 1-056 0'941 0'970
0·93 0'99 0·97 1·68 1·26 0·98 1·01 1-11
2 3 4 4 5 6 6
(4RA) (4RB)
1-051/1-085~ 1'045 1'037
(6RA) (6RB)
Notes: • MBF and MBM are the safety factors by Bishop 's Method, based on the force equilibrium in the slices and on the moment of the reinforcement force with respect to the circle centre only. t Reinforced embankment did not fail at final height. ~ With and without a tension crack at the embankment surface.
failure height. It can be observed that, in most cases, the predicted values are satisfactorily close to unity. The largest deviations from unity were observed for case history 4 (Loke et ai. , 1994). However, Bergado et al. (1994) reported values of predicted safety factors close to one for the case history when a site-dependent correction factor for the soft soil undrained strength (obtained by vane tests) was employed. It is important to point out that an accurate prediction of the safety factor at the failure height for case history 6 (Schaefer and Duncan, 1988) was only possible with the knowledge of the load- strain- time relation for the polymeric reinforcement used in that case history, the behaviour of which is highly dependent on the strain rate imposed. Different methods of analysis are based on different assumptions, so deviations between the results obtained are expected to occur. Figure 4.15 shows the variation of a required reinforcement force versus fill cohesion, for a safety factor (Fr) of 1,3, in a hypothetical situation of an embankment on soft soil with constant undrained strength and depth (Silva, 1996). For cohesionless fill materials the differences between predicted reinforcement forces are small, except for the case of the Fellenius method , which is significantly more conservative. As the fill cohesion increases, it affects each method differently, depending on how each of them deals with the influence of the fill cohesion . The
140 •
Combined failure mechanism
120
o
Modified Bishop's method
100
11 USA Corps of Engineers
~
z
-'"
T Fellenius
80
• Low et al. (1990)
"0
~
·s
60
~ I--
40
rr
Fig. 4.15. Comparisons between predictions of required reinforcement force from different methods (adapted from Silva , 1996)
cjl
20 0 0
5
10
Fill material cohesion : kPa
15
=35°, Y=18 kN/m 3
112
Geosynthetics and their applications
Fig. 4.16. Geosynthetic installation: (a) reinforcement inside the embankment; (b) several reinforcement layers; (c) geocell at the base of the embankment; (d) reinforcement with folded ends; (e) combination of reinforcement with berms; and (f) reinforcement and piles
(a)
(b)
(c)
(d)
.. (e)
. .. . .. "
. ..
.
(I)
combined failure surface approach tends to be the most conservative for greater cohesion values because, beyond a certain value of fill cohesion, the embankment will be considered cracked along its entire height. Therefore , users should be aware of these types of variation, depending on the method employed and the project characteristics (as mentioned earlier in this chapter). It is also important to