High Pressure Food Science, Bioscience and Chemistry
High Pressure Food Science, Bioscience and Chemistry
Edited by
Neil S. Isaacs University of Reading, UK
SOCIETY OF
CHEMISTRY Information Services
The proceedings of the 35th joint meeting of the European High Pressure Research Group and Food Chemistry Group of The Royal Society of Chemistry on High Pressure Food Science, Bioscience and Chemistry held at the University of Reading Reading on 7-1 1 September 1997.
Special Publication No. 222 ISBN 0-85404-728-X
A catalogue record for this book is available from the British Library
8 The Royal Society of Chemistry 1998 All rights reserved. Apartfrom any fair dealing for the purpose of research or private study, or criticism or review as permitted under the terms of the UK Copyright, Designs and Patents Act, 1988, this publication may not be reproduced, stored or transmitted, in any form or by any means, without the prior permission in writing of The Royal Society of Chemistry, or in the case of reprographic reproduction only in accordance with the terms of the licences issued by the Copyright Licensing Agency in the UK,or in accordance with the terms of the licences issued by the appropriate Reproduction Rights Organization outside the UK. Enquiries concerning reproduction outside the terms stared here should be sent to The Royal Society of Chemistry at the address printed on this page.
Published by The Royal Society of Chemistry, Thomas Graham House, Science Park,Milton Road, Cambridge CB4 4WF, UK
For further information see our web site at www.rsc.org Rinted and bound by Bookcraft (Bath) Ltd.
Preface
It was a pleasure to welcome 125 delegates from some fifteen countries to Reading in September, 1997, for the annual presentation of research in the application of high pressures to serve scientific ends. As is now customary, the range of applications and the number of practitioners in the field having grown so large, the meeting addressed especially the topics of food science, bioscience and chemistry, in which fields major applications are now apparent and the techniques pioneered by earlier workers have assumed their recognized place. This publication records the original work presented at the meeting, and if it stimulates the reader in exploring these techniques to the extent that was evident in discussions at the conference, it will serve its purpose. Neil S. Isaacs Reading, May, 1998
Contents
Chemistry Presentations
Plenary Lecture Organic Reactions at High Pressure: The Effect of Pressure on Cyclizations and Homolytic Bond Cle:vage F.-G. Klikner , M.K. Diedrich, G. Dierkes and J . 4 . Gehrke
3
High Pressure Phases and Properties of c 6 0 B. Sundqvist
12
High Pressure Promoted [4+2]/[3+2] Tandem Cycloadditions of Nitrostyrene and En01 Ethers G.J.T. Kuster, F. Kalmoua and J.W. Scheeren
18
Plenary Lecture Application of High Pressure in Cycloaddition Reactions J.W. Scheeren* and R.W.M. Aben
22
EHPRG Award Lecture Conformational Changes in Na'/K'-ATPase Induced by Cation Binding. P. Bugnon, E. Lewitzki, E. Grell and A.E. Merbach
32
Fluorination of Carbonyl Compounds by DAST at High Pressure J. Box, L.M. Hanvood, N.S. Isaacs and R.C. Whitehead
38
The Influence of High Pressure on Reactivity and Selectivity in Transition Metal Catalyzed Reactions A. Mengel, S. Hillers, M. Glos, K. Bodmann and 0.Reiser
40
Plenary Lecture High Pressure Studies in Inorganic Chemistry: Volume Profiles for the Activation of Small Molecules by Transition Metals S. Schindle? and R. van Eldik Dense Gases as Reaction Media V. Krmelj, 2. Knez and M. Habulin
47
53
...
Vlll
High Pressure Food Science, Bioscience and Chemistry
Posters Effects of Pressure on Thermal Response of Poly-(N-vinylisobutyramide) (PNVIBA) S . Kunugi, K. Takano, N. Tanaka and M. Akashi
61
Supercritical Fluid Fractionation of Polymers S . Rey, P. Botella, Y. Garrabos, F. Cansell, J.L. Six and Y. Gnanou
67
Viscosity under High Pressure of Pure Hydrocarbons and their Mixtures: Critical Study of a Residual Viscosity Correlation (Jossi) J. Alliez, C. Boned, B. Lagourette and A. Et-Tahir
75
Calculation of Gas Solubility in Electrolyte Solutions at High Pressure, High Temperature H. Carrier, S . Ye, B. Lagourette, J. Alliez and P. Xans
81
High Pressure Ultrasonic Speed and Related Properties in Petroleum Cuts J.L. Daridon, B. Lagourette and P. Xans
88
Influence of the Pressure and Temperature on the Viscosity and Density of the Binary Water + DAA C. Boned, M. Moha-Ouchane, A. Allal, J. Jose and M. Benseddik
95
Pressure-induced Cycloadditions to Strained Arenes F.-G. Klarner, R. Ehrhardt, H. Bandmann, R. Boese, D. Blker, K.N. Houk and B.R. Beno
101
The Effect of Pressure on Retro-Diels-Alder Reactions F.-G. Klaner and V. Breitkopf
110
The Effect of Pressure on the Formation and Decay of a Furanone during the Maillard Reaction M.J. Bristow and N.S. Isaacs
120
The Hydrolysis of Lipids and Phospholipids at Atmospheric and at High Pressures N.S. Isaacs and N. Thornton-Allen
122
The Effect of High Pressure on the Degradation of Isothiocyanates C. Grupe, H. Ludwig and B. Tauscher
125
Food Science Presentations
Plenary Lecture Food Chemistry under High Hydrostatic Pressure P. Butz and B. Tauscher
133
Contents
ix
Inactivation of Microorganisms and Enzymes in Pressure-treated Raw Milk B. Rademacher. B. Pfeiffer and H.G. Kessler
145
Effect of High Pressure Processing on Properties of Emulsions made with Pure Milk Proteins E. Dickinson and J.D. James
152
Molecular and Functional Properties of HP-treated Egg Components F. Bonomi, E. Donnizzelli, S . Iametti, P. Pittia, P. Rovere, G.F. Dall’Aglio
160
Combined Application of Sub-zero Temperature and High Pressure on Biological Materials R. Hayashi, T. Kinsho and H. Ueno
166
Influence of High Pressure Treatment on P-Lactoglobulin and Bovine Serum Albumin in the Absence and Presence of Dextran Sulphate V.B. Galazka, J. Varley, D. Smith, D.A. Ledward and E. Dickinson
175
Effect of Hydrostatic Pressure on the Physicochemical Properties of Bovine Milk Fat Globules and Milk Fat Globule Membrane C. Kanno, T. Uchimura, T. Hagiwara, M.Ametani and N. Azuma
182
High-pressure Processed Apple and Strawberry Desserts 193 M. Fonberg-Broczek, J. Arabas, K. Gbrecka, A. Grochowska, K. Kariowski, E. Kostrzewa, J. Szczepek, H. Sciewska, B. Windyga, D. Zdziennicka, J. hrkowska-Beta and S . Porowski Brining of Gouda Type of Cheese Curd at High Pressure and its Effect on the Cheese Serum W. Messens and A. Huyghebaert
200
Denaturation and Functional Properties of Pressure-treated Milk Proteins J. Hinrichs and H.G. Kessler
207
Protein-Polysaccharide Interactions in Emulsions Containing High Pressure-treated Protein K. Pawlowsky and E. Dickinson
214
Formation and Syneresis of Rennet-set Gels Prepared from High Pressure Treated Milk D.E. Johnston, R.J. Murphy, J.A. Rutherford and C.A. McElhone
220
Plenary Lecture The Potential and Impact of High Pressure as Unit Operation for Food Processing D. Knorr, V. Heinz, 0. Schliiter and M. Zenker
227
X
High Pressure Food Science, Bioscience and Chemistry
New Development of High Pressure Equipment Reduces Processing Cost A. Traff
236
Acidification of Milk by Glucono-b-lactone under High Pressure M. Schwertfeger and W. Buchheim
242
Composition Changes of Strawberry Puree during High Pressure Pasteurisation D. Bnina, L. Istenesovfi, M. Voldiich and M. Ceiovskf
248
Posters Effect of High Hydrostatic Pressure in a Model Mayonnaise E. Ponce, J. Saldo, M. Capellas, B. Guamis and R. Pla
257
The Effect of High Pressure on Microorganisms and Enzymes of Ripening Cheeses A. Reps, P. Kolakowski and F. Dajnowiec
265
Diels-Alder Reactions of Food-relevant Compounds under High Pressure: 2,3-Dimethoxy-5-methylbenzoquinone and Myrcene J. Kuebel, H. Ludwig and B. Tauscher
27 1
Stability of Antimutagenic Activities in Fruit and Vegetables during High Pressure Processing P. Butz, R. Edenharder, H. Fister and B. Tauscher
277
Rheological Behaviour of Soya Protein Concentrate Following High Pressure Treatment A. Apichartsrangkoon, D.A. Ledward, A.E. Bell and S.G. Gilmour
280
Effects of High Pressure on Lipid Oxidation in Fish K. Angsupanich and D.A. Ledward
284
Impact of Combined High Pressure and Low Temperature on Enzyme Inactivation: 289 Kinetic Study of Soybean Lipoxygenase Indrawati, A. Van h e y , L. Ludikhuyze and M. Hendrickx High Pressure Improvement of the Meat Ageing Enzymes Activity S. Jung, M. de Lamballerie-Anton, Ph. Courcoux and M. Ghoul
295
High Pressure Inactivation and Survival of Pressure-resistant Escherichia coli Mutants in Fruit Juices C. Garcia-Graells, K. Hauben, C. Soontjes and C. Michiels
304
Contents
xi
Kinetics of Vitamin C Degradation under High Pressure-Moderate Temperature Processing in Model Systems and Fruit Juices P.S. Taoukis, P. Panagiotidis, N.G. Stoforos, P. Butz, H. Fister and B. Tauscher
310
Freezing of Potato Cylinders During High Pressure Treatment 0. Schluter. V. Heinz and D. Knorr
317
Molecular Modifications of Ovalbumin upon HP Treatment S. Iametti, E. Donnizzelli, P. Rovere, G.F. Dall’Aglio, G. Vecchio and F. Bonomi
325
Water Loss and Consistency Reduction in Fruits and Vegetables Treated under High Pressures H. Schoberl, W. RUB,I. Wenzel and R. Meyer-Pittroff
33 I
Biochemistry Presentations
Plenary Lecture Influence of Pressure-assisted Freezing on the Structure, Hydration and Mechanical Properties of a Protein Gel H. Barry, E.M. Dumay and J.C. Cheftel
343
Studies on Bacterial Spores by Combined High Pressure-Heat Treatments: Possibility to Sterilize Low Acid Foods P. Rovere, S. Gola, A. Maggi, N. Scaramuzza and L. Miglioli
354
Effects of Pressure on Minor Groove Binding to DNA Studied by the REPA and Fluorometric Methods G.-Q. Tang, S. Kunugi and N. Tanaka
364
Acquired Resistance of Microorganisms to Inactivation by High Hydrostatic Pressure P. Verroens. K. Hauben and C. Michiels
370
Kinetics of Refolding of P-Lactoglobulin after High-pressure Treatment Measured by Reactivity Towards Ellman’s Reagent H. Stapelfeldt, R.E. MBller and L.H. Skibsted
376
High Pressure Inactivation of Polyphenoloxidase: Effect of pH and Temperature C. Weemaes, L. Ludikhuyze, I. Van den Broeck and M. Hendrickx
38 I
Effect of High Hydrostatic Pressure on the Survival and Growth of Escherichia coli 0157:H7 M.F. Patterson, M. Linton and J.M.J. McClements
387
xii
High Pressure Food Science, Bioscience and Chemistry
Strategies for High Pressure Inactivation of Endospore-formingBacteria V. Herdegen and R.F. Vogel
394
Inactivation Kinetics of Microorganisms by High Pressure J.P.P.M. Smelt, N. Dutreux and J.C. Hellemons
403
Posters Pressure-Temperature Stability Diagrams of Proteins: a-Amylases from Bacillus Species P. Rubens, L. Smeller and K. Heremans
41 1
Pressure Induced Fluorescence Quenching in Plant Light Harvesting Complex II J.P. Connelly, A.V. Ruban and P. Horton
417
Pressure Unfolding Ribonuclease A and the Seven Mutants J.P. Connelly, J. Torrent, R. Lange, M.G. Coll, M. Rib0 and M. Vilanova
423
High Pressure Treatments of Listeria monocytogenes at pH 7 and pH 5.6, and Flow Cytometry Monitoring of Pressurized Cells M. Ritz, M.-F. Pilet and M. Federighi
429
High Pressure Germination and Inactivation Kinetics of Bacterial Spores V. Heinz and D. Knorr
435
Physiological Responses and Morphological Changes of Salmonella typhimurium and Listeria monocytogenes to High Hydrostatic Pressure Treatments J.L. Tholozan, M. Ritz, G. Delattre, M.-F. Pilet and M. Federighi
442
Peroxidase Reaction under High Pressure: Influence of Different Hydrogen Donor Molecules P. Butz,A. Fernindez Garcia and B. Tauscher
44s
Proteins under Extreme Conditions: FTIR Spectroscopy with a Cryogenic-High Pressure Cell F. Meersman, P. Rubens, L. Smeller and K. Heremans
45 1
In siru Microscopic Observation of Pressure-induced Gelatinization of Starch in the Diamond Anvil Cell J. Snauwaert, P. Rubens, G. Vermeulen, F. Hennau and K. Heremans
457
Contents
xiii
Physics Presentations Properties and Structural Peculiarities of High-impact Polystyrene obtained by Hydroextrusion under Pressure N.V. Shishkova, B.M. Efros and S.A. Tsygankov
467
Pressure Dependence of Thermal Conductivity of Rocks U. Seipold
474
Application of Natural Diamonds for Generation of Super-high Pressure B.M. Efros and N.V. Shishkova
48 1
Safety in Pressure Testing G. Saville, S.M. Richardson and B.J. Skillerne de Bristowe
486
Equipment and Systems for High Pressure M. Freeman
50 1
Commercial Pressurised Foods in Japan
506
Subject Index
508
Plenary Lecture
ORGANIC REACTIONS AT HIGH PRESSURE. THE EFFECT OF PRESSURE ON CYCLIZATIONS AND HOMOLYTIC BOND CLEAVAGE F.-G. Klhe;',
M. K. Diedrich, G. Dierkes, J . 4 . Gehrke
Institut fiir Organische Chemie, Universitat GH Essen, Universitatsstr. 5 , D-45 117 Essen, Germany Telephone: +49(0)201-183308 1, Fax: +49(0)201-1833082, E-mail: klaerneraoc1.orgchem.uni-essen.de
Pressure in the range of 1-20 kbar (units of pressure: 1 kbar 0.1 Gpa
=
=
100 Mpa
=
987 atm) strongly influences the rate and equilibrium position of
many chemical reactions. Processes accompanied by a decrease of volume (activation volume
A F < 0)
are accelerated by raising the pressure and the
equilibria are shifted toward the side of products (reaction volume
AF < 0),
while those accompanied by an increase of volume (A v' > 0) are retarded and the equilibria are shifted toward the side of reactants
( A F > 0). Many Diels-
Alder [4+2] cycloadditions are accelerated under high pressure and this effect is frequently exploited in synthetic work'). The finding, that the packing coefficient q defined as the ratio of van-der-Waals volume to molar volume (q =
Vw/V) is larger for cyclic structures than for the corresponding acyclic
structures, explains the highly negative activation volumes found for the pericyclic cycloadditions'*2),the preference of pericyclic cycloadditions over
4
High Pressure Food Science, Bioscience and Chemistry
stepwise reactions at high pressure and the negative activation volumes of many pericyclic rearrangements
2,3,4).
Recently, it has been shown that the activation
volumes of pericyclic rearrangements such as the Cope rearrangement of various substituted 1,5hexadienes or the electrocyclization of Z- 1,3,5hexatriene to 1,3-cycIohexadiene involving monocyclic transition states are in the range of A v' rz -9 to - 14 cm3mol-' whereas those of intramolecular DielsAlder reactions involving bicyclic transition states are found in the range
A f
rz
-20 to -38 cm3mol-' '). The absolute values of the latter reactions are
approximately twice as large as or even larger than those observed for the Cope rearrangements or electrocyclization. From this finding it can be estimated that each five- or six-membered ring newly formed in the rate-determining step contributes to about -10 to -15 cm3mol-' to the activation volume of these rearrangements. Here, we report on the relationship between the size of the pressure effect and the size of the newly forming rings in cyclizations. The utility of high pressure in the elucidation of reaction mechanisms will be demonstrated by the example of the racemization and isomerization in 1,3,4,6tetraphenyl- 1,5-hexadiene indicating that a pericyclic Cope rearrangement competes with a dissociative process involving free radicals. The investigation of the pressure effect on homolytic C-C bond cleavage provides a first indication that pressure may be a tool to distinguish between reactions of free radicals within and out of the cage.
Chemistry: Presentations
5
Table 1. Volumes of Reaction ( A F ) , van der Waals Volumes of Reaction (A&), Enthalpies, Entropies, and Gibbs Enthalpies of Reaction Calculated for the Hypothetical Cyclizations of 1-Alkenes to Cycloalkanes by Means of the Corresponding Thermodynamics Parameters.
AVR"
-A -0
fi
-u - -0 H
n
w
-0
'
+
-e-(+w-2n
A.S~
A C ~
-1.7
-5.5
7.86
-7.0
9.95
-2.5
-6.6
6.43
-10.3
9.50
-3.8
-14.7
-13.46
-13.1
-9.56
-4.4
-16.5
-19.47
-21.0
-13.21
-4.7
-21.2
-13.41
-19.6
-7.57
-4.9
-25.6
-9.88
-18.8
-4.28
-4.7
-30.9
-4.6
-32.3
-4.7
-32.8
-4.7
-32.3
-4.6
-27.6
-0 -c3 - 0 --0 7
mb
nzl
cm3 mol-'. V (n-alkene) calculated by the use of Exner incrementsI6). V (cycloalkane) b determined from density measurements in n-hexane. kcal mol-I. cal mol-' K-I.
a
6
High Pressure Food Science, Bioscience and Chemistry
A first indication for the ring-size dependence of volumes came from the
observation that the ring-enlargement of cis- and trans- 1,2-divinylcyclobutane leading to 1,5-~yclooctadieneor 4-vinylcyclohexene and 1,5-~yclooctadieneis accompanied by substantial decrease in volume (volumes of reaction,
A F = -12.8 cm3mol-' or -9.6
and -17.4 cm3mol*', respectively)6). The volumes
of reaction calculated for the hypothetical cyclizations of n-alkenes to the corresponding cycloalkanes by the use of experimentally observed partial molar volumes confirm this trend (Table 1). They decrease continuously from the formation of cyclopropane to that of cyclodecane and, then, seem to be constant for the larger rings, whereas the van der Waals volumes of reaction (AVw) are approximately equal, with the exceptions of the formation of the three-, four-, and five-membered rings. Therefore the ring-size dependent decrease in volume of the cyclizations results from the different packing of cyclic and acyclic states rather fiov the changes in their intrinsic molecular volumes. This may be explained by the restriction of rotational degrees of freedom during the cyclization. Provided that the activation volumes depend similarly on the ringsize, the formation of larger ring should be dramatically accelerated by pressure. The intramolecular Diels-Alder reactions of (E)-1,3,8-nonatriene and (01,3,9decatriene, in which either new five- and six-membered rings or two new sixmembered rings are formed, seem to be the first examples of the validity of this assumption. The activation volumes found for the reaction of the decatriene are, indeed, more negative by -10 and -13 cm3moT' than those found for the corresponding reactions of the nonatriene (Scheme 1). Furthermore, this ringsize effect may explain why the activation volumes observed for the formation of three-membered rings in cheleotropic reactions (A i"# m -1 5 ~ m ~ m o l - 'are )~) substantially less negative than those for the formation of five- and six-
7
Chemistry: Presentations
membered rings in 1,3-dipolar cycloadditions (A v" w -22 ~ r n ~ m o l - 'or ) ~Diels) Alder reactions (Av"
= -35 ~ m ~ m o l - ' ) ~ ) .
Scheme 1.
0-cB 153.2"C
CiS
- (24.8 & 0.3)
-32.3
1
- (24.8 f 0.3)
-27.0
1.2
- (37.6 f 1.6)
-45.4
- (35.0 f 1.3)
-37.4
3.0
..
trans
-c8 172.5"C
CiS
.. 1
trans
~
a)
~~~
the reaction volumes are extrapolated by the use of temperature dependent density measurements to each temperature of reaction.
Optically active tetraphenylhexadiene 1 obtained by the separation of the enantiomers on a chiral KPLC c01umn'~)undergoes a facile racemization at temperatures just above room temperature. At 90°C racemic 1 shows a mutual interconversion to meso-1.") Whereas the racemization of optically active 1 may be the result of the pericyclic Cope rearrangement involving the chair-like transition state 2' shown in Scheme 2, the mutual interconversion of ruc-1 into meso-1 cannot be explained by one or a sequence of Cope rearrangements. The
8
High Pressure Food Science, Bioscience and Chemistry
effect of pressure leads to an unambiguous mechanistic conclusion. The observation, that the racemization of optically active 1 is accelerated by pressure and, therewith, exhibits a negative volume of activation (A? < 0), is good evidence for a pericyclic Cope mechanism in this case. In the other case, however, the finding that the mutual interconversion of rac-1 into meso-1 is retarded by pressure (A? > 0), suggests a homolytic bond cleavage in the ratedetermining step (Scheme 2).
Scheme 2.
1
A v'
AH#
I
cm3mol-' = (21.3 ?E 0.1) kcal mol-'; A$
J
= -(7.4 k 0.4)
=-
(13.2 ?E 0.3) cal mol-' K-i
rneso-l-+ ruc-1: A v ' = + (13.5 k 0.1) cm3mol-' AH# = (30.9 k 0.2) kcal mol-I; A$ = (2.4 J. 0.5) cal mol-' K-' rac-l-+meso-1: Av'=+(11.5 f0.2)cm3mol-' AH# = (30.7 f 0.2) kcal mol-I; A$ = (1.9 k 0.5) cal mol-' K-'
enf-1
Chemistry: Presentations
9
Scheme 3. Ph
Nc*Ez
('INC
L 1300c
[
Ph 2Ph---(zy]-
Ph
Ph rac-4
5
meso-4
meso-4+ rac-4: Al/f = + (10.7 f 4.8) cm3mol-' rac-4+ meso-4: Al/f = + (8.5. f 3.4) cm3mol"
1
7
AJ'# = + (35.7 k 0.4) cm3mol-' AH# = (44.7 k 1.9) kcal mof'; Asf = (33.7 f 4.8) cal mol-' K-' 13)
A Y' = + 6 cm3mol-' 14)
Finally, the effect of pressure on rearrangements involving free radical intermediates (Scheme 2, entry 2 and Scheme 3, entry 1,3,4) shall be compared
High Pressure Food Science, Bioscience and Chemistry
10
with that of a homolytic bond cleavage of 6 in which the intermediary free radical 7 is intercepted by thiophenol as a trapping reagent (Scheme 3, entry 2).12) All reactions are decelerated by pressure. The difference between the activation volumes observed for the rearrangements (A,V" = +6 to +13 cm3mol-') and for the trapping reaction (AV = +35.7 cm3mol-') is remarkable and may be indicative for reactions of free radicals proceeding within and out of the cage.
Acknowledgment: We are grateful to the Deutsche Forschungsgemeinschaft, Ministerium f%r Wissenschaft und Forschung des Landes Nordrhein- Westfalen and Fonds der Chemischen Industrie for financial support.
References and Notes
1 ) Recent reviews: F.-G. Klarner, M. K. Diedrich, A. E. Wigger, Chapter 3 in: Chemistry Under Extreme or Non-Classical Conditions (ed. R. van Eldik, C. D. Hubbard) SpektndWiley, New York 1996; F.-G. K l h e r , M. K. Diedrich, Chapter 12 in: The Chemistry of Dienes and Polyenes, Vol.1 (ed. Z. Rappoport) Wiley 1997; M. Ciobanu, K. Matsumoto, Liebigs Ann./Recueill997,623-635.
2) Y. Yoshimura, J. Osugi, M. Nakahara, Bull. Chem. SOC.Jpn. 1983, 56,680683; Y. Yoshimura, J. Osugi, M. Nakahara, J. Am. Chem. SOC. 1983, 105, 5414-5418; R. A. Firestone, G. M. Smith, Chem. Ber. 1989, 122, 10891094; F.-G. K l h e r , Chemie in unserer Zeit, 1989,23,53-63. 3) F.-G. K l h e r , B. Krawczyk, V. Ruster, U. K. Deiters, J. Am. Chem. SOC. 1994,116, 7646-7657.; C.A Steward jr., J. Am. Chem. SOC.1971, 93,48154821; ibid. 1972,94,635-637.
Chemistry: Presentations
11
4) M. K. Diedrich, D. Hochstrate, F.-G. K l h e r , B. Zimny, Angew. Chem. Int.
Ed. Engl. 1994,33, 1079-1081. 5 ) N. S. Isaacs, P. G: van der Beeke, J. Chem. Soc., Perkin. Trans. 2, 1982,
1205; M. Buback, K. Gerke, C. Ott, L. F. Tietze, Chem. Ber. 1994, 127, 2241; M. Buback, J. Abeln, T. Hiibsch, C. Ott, L. F. Tietze, Justus Liebigs Ann. Chem., 1995,9.
6) W. v. E. Doering, L. Birladeanu, K. Sarma, J. H. Teles, F.-G. K l h e r , J.-S. Gehrke, J. Am. Chem. SOC.1994, I 16,4289-4297. 7) N. J. Turro, M. Okamoto, R. I. Gould, R. A. Moss, W. Lawrynowicz, L. M.
Hadel, ,J. Am. Chem. SOC.1987,109,4973.
8) G. Swieton, J. von Jouanne, H. Kelm, R. Huisgen, J. Org. Chem. 1983, 48, 1035; Y. Yoshimura, J. Osugi, M. Nakahara, 3: Am. Chem. SOC.1983, 10.5, 5414-5418. 9) G. Jenner, New. J. Chem. 1991,15,897. 10) We thank Prof Dr. G. Blaschke and Ms. Scheidemantel (Universitiit
Munster) for the assistance with the HPLC separation of the enantiomers. 11) Part of a collaborative project carried out together with Prof. Dr.W. von E.
Doering and Dr. L. Birladeanu (Harvard University) and Prof. Dr. R. Sustmann and Dr.H.-G. Korth (Universitat GH Essen). 12)Part of a collaborative project carried out together with Prof. Dr. C.
Ruchardt (Universitiit Freiburg). 13) G. Kratt, H.-D. Beckhaus, C. Ruchardt, Chem. Ber. 1984, I 17, 1748-1764. 14) R. C. Neumann ,M. J. Amrich, J. Org. Chem., 1980,45,4629-4636. 15) W. J. le Noble, M. R. Daka, J. Am. Chem. SOC.1978,100,5961-5962. 16) 0. Exner: “Empirical Calculations of Molar Volumes” in W . J. ie Noble, Organic High Pressure Chemistry, Elsevier, Amsterdam 1988, 19-49.
High Pressure Phases and Properties of
c60
B. Sundqvist Department of Experimental Physics, UmeA University, S-90 187 Umeb, Sweden
Abstract: Fullerenes are molecular carbon solids with very strong intramolecular bonds but very weak intermolecular ones. However, a large number of reactive O form covalent intramolecular double bonds on each molecule also enables C ~ to intermolecular bonds under pressure. T h s implies that a) pressure has a large effect on the physical properties of in its pristine zero-pressure forms, and b) treatment at high pressure and high temperature leads to the formation of new, polymeric phases with covalent intermolecular bonds, phases which have new, interesting properties and which may have practical applications in the future. I discuss here some recent experimental results on the structure and properties of C ~ under O high pressure.
The recent discovery [l] of the molecular forms of carbon collectively known as fullerenes has resulted in an almost explosive development of research in this field, and most physical and chemical properties of the most common forms, C ~ and O C ~ Ohave , already been well investigated [2]. Their physical properties often show unusual features, in most cases derived from the fact that the intramolecular bond strengths are comparable to those of diamond while the intermolecular interactions of the solid are very weak and comparable to the interlayer interaction in graphite. Even rather small applied pressures can thus have large effects on the physical properties of the material, as indeed observed even in early high pressure studies [3,4]. During the last few years high pressure studies have also revealed that the 30 double bonds on each molecule easily break up and re-form as covalent intermolecular bonds, leading to a series of transformations into new polymeric phases with increasing pressure and temperature [5-71. Pressure is thus an important parameter in the research on fullerenes and both high pressure effects in general [8] and the high pressure polymerized phases of C60[8,9] have been the subjects of recent reviews.
Chemistry: Presentations
13
High pressure effects on normal c 6 0 At ambient pressure, solid C ~ forms O a close-packed cubic (face-centered cubic, fcc) lattice where the molecules rotate (almost) freely at hlgh temperatures. With decreasing temperature the rotation slows down and at 260 K a transformation occurs into an orientationally ordered simple cubic (sc) phase. In this phase the molecules are still jumping between a small number of orientational states but on further cooling even thls movement stops near 90 K, below which the material forms an orientational glass with a small amount of orientational disorder frozen in. In the sc phase there are two possible molecular orientations with either carbon pentagons (P orientation) or hexagons (H orientation) on one molecule facing double bonds on nearest neighbours [2]. The P orientation has a slightly lower energy but gives a slightly larger intermolecular distance. Because of the small energy difference between the P and H orientations only about 60% of the molecules are in the P state near 260 K, but with decreasing temperature the fraction increases to about 83% at and below the glassy crystal transition at 90 K. When c 6 0 is submitted to high pressure the decreasing intermolecular distance leads to an increase in intermolecular interactions and molecular rotation becomes more difficult. The orientational transition near 260 K therefore shifts rapidly to higher temperatures at a rate of about 160 WGPa [8,10] such that for molecular C60 at room temperature the sc phase is the stable state at all pressures above 0.2 GPa. For the same reason the glassy crystal transition at 90 K also shfts to higher temperatures under pressure at a basic rate of 62 WGPa [ 111. However, for th~stransition the measured phase transition slope depends on the exact experimental conditions because in the sc phase the relative fractions of P and H oriented molecules depend on pressure. Since the lattice parameter of the P oriented phase is larger than that of the H oriented phase the application of pressure leads to a gradual P+H reorientation with increasing pressure [12] such that the two orientations are in equilibrium at a pressure of only about 0.2 GPa. Sundqvist et al. [13] extrapolated thls trend and predicted the formation of a completely H ordered phase at pressures above 0.6 GPa at low temperature and about 1.5-2.5 GPa at 300 K, and the existence of such a phase has later been directly verified by Raman scattering studies under pressure [14]. There is probably no sharp transition into this ordered phase, because to a first approximation the fraction fH of H oriented molecules should increase continuously with pressure according to the formula exp(-A(p)/kT))-', fH = (1 i(1) where A@) is the (linearly) pressure dependent energy difference between the orientational states. Even at very hgh pressures there should thus remain a small fraction of P oriented molecules. However, calculations by Burgos et al. [15]
High Pressure Food Science, Bioscience and Chemistry
14
show that the intermolecular potential depends strongly on the orientations of the neighbours. At ambient pressure, if all nearest neighbours are P oriented there are two minima in the potential, corresponding to the P and H orientations. However, if instead all neighbours are H oriented only one minimum, corresponding to the H orientation, is observed. As suggested by Sundqvist [8], there is thus a possibility that at some point the "P state" potential minimum crosses the level of the H state (or simply dlsappears). At this point the smooth P+H reorientation model should break down when the remaining P oriented molecules reorient more or less simultaneously into the H oriented state, gving a sharp transition. Quite well defined transition anomalies have indeed been observed in the range of pressures up to 2.5 GPa. Figure 1 shows the orientational phase diagram of C60 at low pressures. The fccwsc transition boundary is well defined and independent of external factors as long as the pressure medium does not penetrate the C60 lattice [8,10], but the glassy crystal transition near 90 K depends on the thermal history of the sample. Cooling through th~sphase line "fieezes in" the P/H fraction at the pressure used and if the pressure is changed at temperatures below this line the P/H ratio will not change. If the sample is heated at the new pressure the crystal will contain a "non-equilibrium" P/H ratio. Thls changes the actual transition temperature and gives rise to very interesting relaxation phenomena I
-
P hexagon oriented
-
100 -/---S.C. glassy crystal
-
P (GPa) Figure 1 . Low-pressure, low-temperature phase diagram of CSo(from 181). See text for details; symbols denote transition anomalies observed by measurements of the compressibility (Refs. [ 161 (w) and [ 171 (O)), by Raman (Refs. [ 181 (A), [ 191 (0)and [6] (0)), visible light (Ref [20] (*)) and IR (Ref. [21] (V)) spectroscopy, and x-ray structural studies ([22] (A)).
Chemistry: Presentations
15
[8,11] discussed in detail by Anderson at this Meeting. The P+H orientational transformation is symbolized by the hatched area in Figure 1, calculated using Eq. (1) and experimental data for A@) [ 121. The low-pressure boundary of this area corresponds to fH = 0.1 and the upper boundary to fH = 0.02, while anomalies observed by various methods are indicated by the symbols. Although the width of the "transition area" is very large at high temperature the agreement between the observed anomalies and the calculated transition pressure is good. As might be expected, the changes in volume and orientational structure brought about by applied pressure also change other physical properties of C,. For a review of the effects of pressure on the electrical, thermal, optical, and other properties Of C60 and other fullerenes the reader is referred to Ref. [8]. Polymeric high-pressure phases of c 6 0 High pressure polymerized forms of Cm were first identified by Iwasa et al. [5] after heatmg samples to above 300°C at 5 GPa. Later studies have shown that several well defined polymeric phases can be produced by high temperature treatment under pressure [6-91. In the range between about 2 and 8 GPa two well defined crystal structures are obtained depending on the temperature used [23, 241. Below about 35OOC an orthorhombic form containing linear (1D) chains of molecules running along the o r i g d (110) directions is formed, while at higher temperatures a 2D polymer consisting of polymerized close-packed 411> layers forming a rhombohedral structure is obtained. At pressures above 8 GPa less well defined, very hard quasi-cubic structures are usually found to form at "low" temperatures, while treatment at temperatures well above 5OOOC usually results in the formation of hard disordered (glass-like) phases, some of which are claimed to be harder than crystalline diamond [6,9]. Recent investigations of the specific heat capacity, hardness and acoustic properties confirm these claims [9], and the materials are thus of large potential interest for practical commercial applications in, for example, cutting tools. Since the high-pressure part of the phase diagram has recently been well reviewed elsewhere [9], I will here discuss some recent developments in the field of "low-pressure" polymerization of Ca. Marques et al. [24] have recently carried out a very extensive investigation of the phase diagram of Cm fiom 3 to 8 GPa. In addition to the orthorhombic and rhombohedral structures discussed above, they find that treatment at low pressure (< 4 GPa) and high temperatures results in the formation of a tetragonal structure, evidently formed by polymerization along the (100) directions. Interestingly, if the phase line for the orientational (fcc-sc) transition is extrapolated and drawn in the same diagram it delineates almost exactly the boundary for formation of the
16
High Pressure Food Science, Bioscience and Chemistry
tetragonal phase. As also discussed in their paper, the molecules rotate almost freely at temperatures above this line, resulting in a large probability for formation of covalent intermolecular bonds in random directions, while below this line the material should have an almost perfectly H oriented structure at the pressures used. When polymerization starts in the latter state well ordered phases (orthorhombic or rhombohedral) should easily form, but in the high temperature state the free molecular rotation should gwe a high degree of disorder which seems to result in the formation of a mixed rhombohedralltetragonal structure. It is possible that the tetragonal structure is thus basically a dsorder-induced structure. This conclusion is strengthened by the fact that such a tetragonal phase is also often observed after polymerization at lower pressures of the order of 1-2 GPa. Many studies have been carried out in this range after the early studies by Bashlun et al. [7], and it is interesting to note that there has been a considerable debate as to the structure of the material formed in this range. Neither Bashkin et al. nor Persson et al. [25] could deduce the actual structure, and later studies, for example by Rao et al. [26], have tended to show that the structure is very disordered, consisting of mixtures of all three structures discussed above. Since lowpressure polymerization is always carried out at temperatures in the fcc stability range a large random component in the polymerization reaction is probable. However, polymerization in this pressure range may under suitable conditions result in very well ordered materials, as evidenced by the fact that we have recently been able to polymerize large single crystals of c 6 0 and determine their crystal structure very accurately [27]. We found the material to be orthorhombic, as expected, but the details of the structure were different from those of the orthorhombic lattice found by Marques et al. and others above 3 GPa. A similar structure was also recently reported by Agafonov et al. [28]. Finally, while the physical properties of the disordered "orthorhombic 1D" polymers just discussed have by now been well studied by us and others [8,9], those of the 2D rhombohedral and tetragonal phases have not. We have therefore recently initiated such a study. So far, only preliminary data for the thermal expansion and compressibility are available, but these can be understood in terms of the 2D structure of the material. The covalent intermolecular bonds formed would be expected to very rigd because of their sp3 ("diamond-like") character. A material with ID chains should thus to a first approximation be incompressible along one dimension and have a thermal expansion near two thirds of that of "normal" CG0.The compressibility of c 6 0 is approximately 0.1 GPa-' [8] and we would thus expect that of a 1D polymer to be 0.07 GPa-' and that of a 2D polymer to be 0.03 GPa-'. Preliminary values for the compressibilities are smaller than but similar to those calculated. As one example, the compressibility of 2D polymer is about 0.023 GPa-'. Data for the thermal expansion coefficients of
Chemistry: Presentations
17
normal and for the 1D and 2D polymers at high temperatures also scale in approximately the 1 : 213 : 1/3 ratio expected. Acknowledgements: This work was financially supported by the Swedish Research Councils for the Natural Sciences (NFR) and Engmeering Sciences (TFR). References: [l] H.W. Kroto et al., Nature 318, 162 (1985). [2] M.S. Dresselhaus, G. Dresselhaus and P.C. Eklund, Science of Fullerenes and Carbon Nunotubes (Academic Press, San Diego 1996). [3] J.E. Fischer et al., Science 252, 1288 (1991); S.J. Duclos et al., Nature 351, 380 (1991). [4] G.A. Samara et al., Phys. Rev. Lett. 67, 3136 (1991); G. Knza et al., J. Physique I 1, 1361 (1991). [5] Y. Iwasa et al., Science 264, 1570 (1994). [6] V. Blank et al., Phys. Lett. A 188, 281 (1994). [7] 1.0.B a s h et al., J. Phys.: Condens. Matter 6,7491 (1994). [8] B. Sundqvist, Adv. Phys. (to be published). [9] V.D. Blank et al., Carbon (to be published). [lo] G.A. Samara et al., Phys. Rev. B 47,4756 (1993). [l 13 0. Andersson, A. Soldatov and B. Sundqvist, Phys. Lett. A, 206,260 (1995). [12] W.I.F. David and R.M. Ibberson, J. Phys.: Condens. Matter 5, 7923 (1993). [13] B. Sundqvist et al., Solid State Commun. 93, 109 (1995). [14] J.A. Wolk, P.J. Horoyski and M.L.W. Thewalt, Phys. Rev. Lett. 74,3483 (1995). [15] E. Burgos, E. Halac and H. Bonadeo, Phys. Rev. B 49, 15544 (1994). [16] A. Lundin and B. Sundqvist, Europhys. Lett. 27,463 (1994). [17] Z. Bao et al., Chin. Sci. Bull. 40, 898 (1995). [18] S-J. Jeon et al., J. Raman Speck. 23,311 (1992). [19] K.P. Meletov et al., Phys. Rev. B 52, 10090 (1995). [20] K.P. Meletov et al., J. Physique 12,2097 (1992). [21] Y. Huang, D.F.R. Gilson and I S . Butler, J. Phys. Chem. 95, 5723 (1991). [22] A.P. Jephcoat et al., Europhys. Lett. 25,429 (1994). [23] V.A. Davydov et al., JETP Lett. 63,818 (1996). [24] L. Marques et al., Phys. Rev. B 54, R12633 (1996). [25] P.-A. Persson et al., Chem. Phys. Lett. 258,540 (1996). [26] A.M. Rao et al., Appl. Phys. A 64, 231 (1997). [27] R. Moret et al., Europhys. Lett. (in press, 1997). [28] V. Agafonov et al., Chem. Phys. Lett. 267, 193 (1997).
High Pressure Promoted [4+2]/[3+2] Tandem Cycloadditions of Nitrostyrene and Enol Ethers G.J.T. Kuster ,F. Kalmoua, J.W. Scheeren* Department of Organic Chemistry, NSR-Center, University of Nijmegen, Toernooiveld, 6525 ED, Nijmegen, The Netherlands. E-mail: jsch@ sci.kun.nl Tandem [4+2]/[3+2] cycloadditions of nitroalkenes with electron-rich and electron-poor alkenes have been applied by Denmark et al. in the synthesis of several natural pyrrolizidine alkaloids as (-) hastanecine, (-) rosmarinecine and (-) crotanecine lP2. In these tandem cycloadditions an electron-rich alkene reacts with a nitroalkene in an inverse electron demand Diels-Alder reaction under formation of a nitronate (mono-adduct), which then reacts with a second alkene in a [3+2] cycloaddition, leading to a nitroso acetal (di-adduct): R10
)I
R2
+T8-
8
Rloy
0 8 0
p -
0 1 .
1 2 0 T
13+21
R2
heterodiene
mono -addud
di-addud
At ambient pressure this reaction needs strongly activated nitroalkenes or a stoichiometric amount of Lewis acid catalyst. Since cycloadditions are accelerated by high pressure it was tried to eliminate the use of Lewis acid catalysts or strongly activated nitroalkenes by using high pressure conditions (12- 15 kbar). Recently, we observed a strong accelerating effect of high pressure on the tandem [4+2]/[3+2] cycloaddition of nitroalkenes with enol ethers 3,4: 0
o8
R I O P $ C ?
+7 12-15 b a r [4+21
R1o’n
Ph
heterodiene
b R
i
O
~
~
12-15 kbar
Ph mono -adduct
Ph deadduct
o
R
i
19
Chemistry: Presentations
Since nitronates (mono-adducts) react faster with electron-poor alkenes than with electron-rich alkenes it was possible to perform the tandem cycloaddition in an one-pot-three-component system containing an electron-rich enol ether, nitrostyrene and an electron-poor or neutral mono-substituted alkene (ratio l/l/l). Results of this one-pot-three-component tandem cycloaddition using different dipolarophiles are shown in the following scheme.
RITQ R'F __t
R'?R2 J
12-15kbar
R2
h
h
[3+21
R2
Ph
RI = p-MeO-benzyl, R2= H
In all cases the [4+2] cycloaddition was completely regio- and stereoselective, whereas the [3+2] cycloaddition was regioselective but not completely stereoselective. By using an excess of nitrostyrene the nitroalkene reacted as well in the DielsAlder reaction (hetero-diene) as in the 1,3-dipolar cycloaddition (dipolarophile). The results are presented in the next scheme.
20
High Pressure Food Science, Bioscience and Chemistry
A
B
products Rt
RI
R,
EIO
H
n
B
A E
t
O
V
p
:
ratio AIB
E I O F J ' h z
69/31 Ph
p-MBO'
H
Ph
H
12/28
H
wpTwN12
-OCHICHl.
29/71
mr wNo2 Pfl
H Me0
Ph
no reaction
-OCH$H,CHa. -(CHlk-
O,
Ph
Ph
64/35
Ph
p-MB=p-methoxybenzyl
A mixture of regio-isomers A and B was found. The main regio-isomer A was easily transformed into a bicylic-N-alkoxy-Plactam after a base catalysed intra-molecular rearrangement:
A
B
p-lactam
C
Chemistry: Presentations
21
By performing the tandem cycloaddition with a range of enol ethers and nitrostyrene the synthesis of novel bi- and tri-cyclic p-lactams was accomplished
'.
Reduction of some cycloadducts resulted in the formation of pyrrolidine and pyrrolizidine derivatives containing the rigid aryl-ethyl-amine moiety: H O y h redudion
Ph
~
pyrmlidin
Ph
Currently we are investigating how to optimise the reduction and how to influence the regio-selectivity in the [3+2] cycloaddition. In conclusion, we have shown that the application of high pressure in this tandem [4+2]/[3+2] cycloaddition eliminates the use of stoichiometric amounts of Lewis acid catalysts or strongly activated nitroalkenes, which are necessary at ambient pressure. By the application of high pressure we were able to extend the scope of the tandem [4+2]/[3+2] cycloadditions towards the use of higher substituted nitroalkenes and enol ethers. The high pressure promoted one-pot-three-componenttandem cycloaddition opens a straightforwardroute towards a variety of pyrrolidine and pyrrolizidine derivatives, simply by varying the alkene substituents. Furthermore, the high pressure promoted synthesis of novel di- and tri-cyclic 0lactams was performed. 'Denmark S.E. JOrg. Chem. 1997,62,435-436 and ref. cited herein; Denmark S.E. et 01, Chem. Rev.( 1996) 96,137. 'Denmark, S.E.;Schnute, M.E. J. Org. Chem. 1994,59,4576 By using nitrostyrene as heterodiene, cyclic compounds containing the aryl-ethyl-amine moiety in a fixed position could be formed .Thisaryl-ethyl-amine moiety is occurring in several neurotransmitters,hormones and because of this also in CNS active drugs. Our research is focussed on the synthesis of compounds containing a rigid aryl-ethyl-amine moiety. This rigidity could express itself in a higher receptor selectivity, finally resulting in a more selective drug. Uittenboogaard, R.; Seerden, J.P.G.; Scheeren, J.W.Tetrahedron 1997,53, 11929 To be published.
'
'
Plenary Lecture Application of High Pressure in Cycloaddition Reactions J.W. Scheeren* and R.W.M. Aben Department of Organic Chemistry, NSR-Center, University of Nijmegen, Toernooiveld, 6525 ED, Nijmegen, The Netherlands. E-mail: jsch@ sci.kun.nl Cycloaddition reactions can be a straightforward and stereoselective entrance to homo cyclic and hetero cyclic compounds. Therefore they are often applied in the synthesis of cyclic natural products. Generally cycloadditions proceed well when one reaction partner is electron-rich and the other is electron-poor. For that reason the scope is often determined by the degree of electron-richness and/or electron-poorness of the used reaction partners. For example moderate electron-rich enol ethers need strong electron-poor reaction partners such as tetracyanoethylene to undergo polar [2+2] cyloaddition reactions. In several cases it has been shown that the scope of these cycloadditions can be extended by the use of Lewis acid catalysts. These catalysts usually complex with the electron-withdrawing substituent making the concomittant reaction partner more electron-poor. By the application of high pressure the scope of these cycloadditions can be extended to the use of reaction partners having fewer or weaker electrondonating or electron-withdrawing substituents. So at first sight high pressure can have the same effect as the use of a Lewis acid catalyst. It will be shown however that high pressure often gives different results from the use of Lewis acids. Further extension of the scope of a Lewis acid catalyzed cycloaddition is possible by combining high pressure and Lewis acid catalysis. Our main results from the past five years concerning the application of high pressure in polar [2+2] cycloadditions, Diels-Alder reactions and 1,3 dipolar cycloadditions will be presented. These include cycloaddition reactions of cyanoalkenes, nitroalkenes and enones with enol ethers and dienes, and of nitrones and nitronates with enol ethers. High pressure effects on the outcome and stereoselectivity of these reactions will be discussed. Diels-Alder reactions of furans and pyrroles, directed to application in the synthesis of some natural products, e.g. epibatidine, CD-ring of taxol, will be highlighted.
23
Chemistry: Presentations
Polar [2+2] and [4+2] cycloaddition reactions of alkenes, carbonyl compounds and a,p-unsatura ted carbonyl compounds. The reaction of tetracyanoethylenewith enol ethers leading to cyclobutanes proceeds via dipolar intermediates. High negative activation volumes have been determined for this reaction because of dipolar transition state.' The reaction of the much less reactive 1,1-dicyano-2-arylethenes with enol ethers which is synthetically more interesting proceeds very slowly at normal pressure.2 At 12 kbar high conversions can be achieved for a variety of enol ethers and dicyanoethenes (scheme 1). Scheme OR1 Ar M~~ R2M +
R3
R 4 H
CN
1
RqA? r$;
RqAp rJ;
CH2C12 I 12 Kbar, 25-50' C I6-48h
R3 Rz, R3, R4 = H, Alkyl A main product OR, and Ar in cis position (A) cisltrans ratio varies from 3/2 to .95/5 highest stereoselectivity for R3 = Alkyl, R2 = H
R4
R2
OR1
B
In general cyclobutanes with the aryl- and alkoxygroup in cis position are the main products. Stereochemistrydepends however on the structure of the used en01 ether3.The less reactive 1-cyanocinnamic methyl ester reacts with en01 ethers at 15 kbar and room temperature to [4+2] cycloadducts, which at 50' are converted partly into cyclobutanes ( scheme 2).
0
CN
OMe
Stereoselectivity of this cycloaddition is very high, leading mainly to the endo products. Polar [2+2] are often catalyzed by Lewis acids. The different effects of high pressure and Lewis acids are nicely illustrated in the reaction of ketene acetals
High Pressure Food Science, Bioscience and Chemistry
24
and aldehydes. In the absence of Lewis acids this cycloaddition can only be performed under high pressure. However the stereoselectivity is different leading to cis oxetanes when using Lewis acids and trans oxetanes when high pressure is applied. The stereochemical differences are explained via transoid dipolar transition states in the presence of Lewis acids and cisoid transition states under high pressure4(chartl). 0 Chart 1 II OMe o+ R,HC-C-OMe R2+o ,!)+oMeZ;FJ;dNP) A I ~
H
H
.IF R2
OMe
NP, ZnClz
R2HC-OH
stereoselectivity65-95%
H
HP
Application of high pressure allows synthesis of oxetanes from ketene acetals which polymerize under the influence of Lewis acids. Combination of high pressure and Lewis acids allows the synthesis of oxetanes from ketones and ketene acetals which is not possible at normal pressure4.High pressure and Lewis acid catalysiswere also applied in the synthesis of P-lactams from enol ethers and isocyanates derived from amino acids’. A different effect of high pressure compared with Lewis acid catalysis is also demonstrated in the reaction of ketene acetalswith a,&unsaturated aldehydes.Under high pressure circumstances cyclobutanes can be isolated as main products whereas in the presence of Lewis acids only oxetanes or dihydropyranes are found6. The less electron-rich enol ethers (compared with ketene acetals) react with a$-unsaturated aldehydes under the influence of high pressure and Lanthanide catalyst to give dihydro yrans”*.Highly substituted dihydropyrans have been synthesized in this way (scheme 3). Scheme 3
A0
+
Ro’()
R2
R4
CH3
::;b:t%..:
r
R3 CH3
25
Chemistry: Presentations
High pressure combined with Lewis acid catalysis allows even Diels-Alder reactions of the low reactive 3-methyl, cycloalk-2-ene-1-ones. No conversion is found with only Lewis acids or high pressure alone'. This cycloaddition has been applied in the regioselective construction of the steroid skeleton" (scheme 4).
+b
Scheme 4
s<
12KbarICHZCl2, 5-10% EtAICIz
:Q
K!;o & D
exo IIsyn anti endo= -1 1,7
0 Diels-Alder reactions of furans and pyrroles
Furans and pyrroles show a low reactivity in Diels-Alder reactions due to their aromatic character. Cycloaddition reactions of furan can sometimes be catalyzed by Lewis acids but they proceed generally well when performed under high pressure". The hydrogenated cycloadduct of furan and citraconic anhydride, palasonin, is a natural product of biological interest. Recently it has been synthesized from furan at normal pressure via a 12 steps synthesis12.It has been demonstrated that by application of high pressure this compound can be obtained in high yields in only two stepsI3.The investigators also showed that in this case Lewis acid catalysis cannot replace the use of high pressure. The cycloadduct decomposed quickly in the presence of Lewis acids at normal pressure. We have explored the high pressure cycloaddition of bans and citraconic anhydride as a simple entry to CD-ring precursors of the antitumor compound Taxol. In the reaction of 2-substituted furans and citraconic anhydride both regioisomers of the exo adducts were formed which were directly hydrogenated to prevent backwards reaction to starting compounds* (scheme 5).
-
yields cycloadducts 7040% ratio 314 111
High Pressure Food Science, Bioscience and Chemistry
26
The hydrogenated cycloadducts were selectively reduced to tricyclic lactones which under high pressure gave regioselective opening of the ether bridge in the reaction with acetyl bromide or the mixed anhydride of acetic acid and trifluoroacetic acid (scheme 6). Scheme 6
wo X-
CH3g-X,
X=Br, OCOF3 15 Kbar
Tricyclic lactones prepared from the high pressure cycloadducts of furfuryl acetate and citraconic anhydride were converted in 5 steps into useful precursors for the CD-ring fragment* of taxol by the sequence given in scheme 7. 0 Scheme 7 2) TosCllpyr
CH~OAC
I
3) Nal 4) Zn
*
-
KH I CISi(Et)3
cidtrans ratio determined by reaction temperature overall yield 40%
Epibatidine, an alkaloid with a bicyclic azabicyloheptane structure, isolated from the skin of an Ecuadorian poison frog, was recently reported as a novel high potent non-opioid analgesic agent and for that reason has attracted much attention of synthetic chemi~ts'~. In a straightforward synthesis of epibatidine analogs 7- azabicyclohept-2-enes have been applied in reductive Heck-arylation reactions (scheme 8)''. COOMe I
Scheme 8 Ar-X / Pd(0) /HCOO-Y+ piperidine or n-Bu4N+CIX=Br, I Y=H+, K+
COOMe I
21
Chemistry: Presentatiom
Several appoaches to 7-azabicyclohept-2-enes via Diels-Alder reactions of Ncarbomethoxypyrrolesincluding high pressure reactions are described in scheme 9. COOMe I
Scheme 9
+
COOMe I
a) H, !Pd(C)
Tos36%
I COOMe
Tos
Altenbach, H.J. et a/.Chern. 6er. 1991,124, 791
0': -
/
y
Phs02 COOMe
;?&
COOMe I
COOMe I
12Kba'/13-
6%Na(Hg)
CHsCN 80%
Sph NaHzP04
SPh
Phs02
Scheeren, J.W. eta/. Heterocycles 1989,B, 79; Tetrahedron Left 1994,35, 1299
*&
FOOMe
0
12 Kbar
f!
FOOMe
6% Na(Hg)
NaH2P04 MeOHnHF
-k
PhS02 COOMe
Seerden et al*
1:l endo/exo
Me
YOOMe N
/
fi +
Me
-
PhSo2f COOMe
Seerden et at'
r.t. 15 Kbar
CHCl3 >95%
Phs02 1:l endo/exo
KF4
COOMe I
NHPO
Me
\ Me
The results from scheme 9 show that the high pressure promoted cycloaddition of N-carbomethoxypyrrole with vinyl phenyl sulphone lead to this compound in fewer steps and in higher yields*. 1,3-dipolar cycloaddition reactions of nitrones with enol ethers
It was recently reported that the cycloaddition of nitrones with enol ethers is promoted by Lewis acids e.g., chiral oxazaborolidines and also by high pressure 16.17 . Some results with C-phenyl-N-methyl nitrone and C-phenyl-N-phenyl nitrone are presented in scheme 10.
28
High Pressure Food Science, Bioscience and Chemistry
Scheme 10
R
R
H+o-
+
ph -1 R=Me 2 R=Ph
R
+ CH2C12; r.t. and/or HP
cis
OEt 5a R=Me 6a R=Ph
trans : OEt 5a
Interestingly Dicken and De ShongI6found that the applied pressure is critical for conversion of this reaction (entries 2 and 3). No conversion was found at 4 kbar and room temperature whereas 80%yield was obtained at 2 kbar and 50’. The authors ascribe the negative results at 4 kbar to dimerisation of the nitrone under that circumstances. The regioselectivity of the reaction is high but the endo/exo stereoselectivity is low. Combination of high pressure and catalysts gave somewhat higher yields (entries 4-8)but the stereoselectivity remained low. Interestingly the highest stereoselectivity was found in the absence of a catalyst at 2 kbar, room temperature (entry 7), but conversion was low under these circumstances. are presented in Some cycloadditions of 3,4-dihydroisoquinoline-N-oxide scheme 1 1”. 3,4-Dihydroquinoline-N-oxidedid not react with ethyl vinyl ether at normal pressure and roomtemperature in dichloromethane. At 2 kbar conversion is slow but stereoselective with respect to the formation of the exo product is high (>90%).Yieldswere higher for the Lewis acid catalyzed reaction. Best yields were obtained however when high pressure was combined with Lewis acid catalysis (65%) although stereoselectivity was less. The same was found for the cycloaddition of this nitrone with dihydrofuran leading in this case exclusively to the exo adduct.
29
Chemistry: Presentations
Scheme 11
a : o . ' 7 0 E t
20 mol% cat. r.t. I2 Kbar
CHzClz 16 h; 65%
--
OEt
endo
OEt
33 : 6 7
8x0
no solvent 20 h; 91%
exo
15 Kbar, 72 h
The low reactive dihydropyran does not react in the presence of a Lewis acid catalyst at normal pressure or at 2 kbar. At 15 kbar the corresponding exo isooxazolidine was isolated in 48 % after 3 days at room temperature without solvent. Interestingly Lewis acid catalysis had no effect on the yield of this reaction. Unfortunately the chiral Lewis acid catalysts did not induce enantioselectivity in all these cycloadditions. Enantioselectivity was however found in the chiral Lewis acid catalyzed cycloaddition of pyrroline -N-oxide with dihydrohan (scheme 12). HP 20 mol% cat.
0-
exo
Tos
Combination of chiral catalysis and high pressure (2 kbar ) resulted in this case not in higher e.e's or yield (highest e.e was 38% at ambient pressure).
Tandem [4+2]/[3+2] cycloaddition reaction of nitroalkenes and enol ethers Nitroalkenes can react with enol ethers in a [4+2] cycloaddition Ieading to a nitronate. These nitronates are in fact reactive dipolar species which are able to react with a variety of as well electron-rich as electron-poor alkenes (scheme 13).
High Pressure Food Science, Bioscience and Chemistry
30
Scheme 13
General requirements for performing these reactions at normal pressure are long reaction times, use of strongly activated nitroalkenes or use of stoichiometric amounts of Lewis acid catalyst^'^"^. As can be expected, these tandem cycloadditionsare strongly promoted by high pressure. High pressure eliminates the need for a catalyst or large excess of dienophile and or dipolarophile. The effects of high pressure on the tandem cycloaddition of nitrostyrenes with ethyl vinyl ether are presented in scheme 14. Scheme 14
4
o++
l a R=H Ph 1b R=Me
10 12
Ph
I00
0.17
100
0.19
-i"~~*'") I oN ,o /
07
("1
+(O,
*xo
,.+o
- -
R I 4aR=H Ph 4b R=Me -
R=H 3b R=Me
40
15
The results at different pressure show that for nitrostyrene (R=H) the diadduct is always the main product. For the more hinderend nitrostyrene (R=Me) it is possible to obtain the monoadduct as the main product. An interesting possibilty to trap the monoadduct is to add a third alkene, which does not react with the other alkene, to the reaction mixture. This three component system offers fascinatingpossibilities for the synthesis of a large variety of bicyclic compounds as will be demonstrated in another paper presented in this conference.*O Conclusions The results obtained with a variety of cycloaddition reactions demonstrate that high pressure is often not just an alternative for the use of Lewis acid. High pressure can lead to different stereochemistry and different site selectivity. Furthermore application of high pressure in combination with Lewis acid catalysis can strongly extend the scope of cycloaddition reactions. Also cycloadducts can be obtained in cases where use of Lewis acid catalysts lead to polymerisation.
Chemistry: Presentations
31
References: 1 . G. Swieton, J.V. Jouanne, H. Kelm, Proc. Int. Conf. High Pressure, 4*, 1974,652 (1975). 2. J.P. Ortega, Mem. R. Acad. Cienc. Exactas Fis. Nat., m , 9 , 7 ; C.A. 1980, 93,45498. 3. R.W.M. Aben, J.W. Scheeren, Synthesis m , 3 7 . 4. R.W.M. Aben et a1 Tetrahedron Lett. 1983. 5. R.W.M. Aben, E.P. Limburg, J.W. Scheeren, High Pressure Reserch, 1992, 167. 6. R.W.M. Aben, J.W. Scheeren, Tetrahedron Lett. 26,1889, (1985). 7. D.A.L. Vandenput, J.W. Scheeren, Tetrahedron 5 l , 8383 (1995). 8. G. Jenner, High Pressure Research =,32 1. 9. R.W.M. Aben, L. Minuti, J.W. Scheeren, A. Taticchi, Tet. Letters 1991, 6445. 10. E.G. Baitz, L. Minuti, J.W. Scheeren, R. Selvaggi, A. Taticchi, Nat. Prod. Letters = , 9 1. 1 1 . Review: K. Matsumoto, A. Sera, Synthesis 1985,999-1027. 12. D.B. Rydberg, J. Meinwald, Tetrahedron Let. 37, 1129 (1966). 13. W.G.Dauben, J.Y. Lang,Z.R. Guo, J. Org. C h e m . m , 6 1 , 4 8 1 6 . 14. Spande, T.F; Garaffo, H.M.; Edwards, M.W.; Yeh, H.J.C.; Pannel, L,; Daly, J.W.; JACS, 1992,114,3475; Badio, B.; Daly, J.W.; Mol. Pharmacol. 1994, 45,563. 15. Clayton, S.C.; Regan, A.C., Tetrahedron Lett. 1993,34,7493; Bai, D.; Xu, R.; Chu, G.; Zhu, X., J. Org. Chem. 1996,37,4600. 16. Dicken, C.M.; Deshong, P., J. Org. Chem. 1982,47,2407-205 1 1 7. Seerden, J.P.G.; Boeren, M.M.M.; Scheeren, J.W., Tetrahedron 1997,53, 11843. 18. S.E. Denmark et al. Chem. Rev. (1996), 96, 137. 19. Uittenbogaard, R.; Seerden, J.P.G.; Scheeren, J.W., Tetrahedron 1997,53, 1 1929. 20. Kuster G.J.T. et al, this conference proceedings * to be published
E.H.P.R.G.Award Lecture Conformational Changes in Na' / K'-ATPase Induced by Cation Binding P. Bugnona, E. Lewitzkib, E. Grellb*and A. E. Merbacha* a
Institut de Chimie Minerale et Analytique, Universite de Lausanne, BCH, Universite de Lausanne, CH-1015 Lausanne, Switzerland Max-Planck-Institut f i r Biophysik, Kennedyallee 70, D-60596 Frankfurt a.M, Germany
1. INTRODUCTION The concept of protein structural transitions as an important functional element in molecular biology is widely used. Despite the importance of this concept, there is a clear lack of methods allowing to quantify the extent of conformational transitions in solution. The determination of distance changes from energy transfer studies upon suitable chemical modifications offers such a possibility to characterize structural properties of equilibrium ground states. Because conformational transitions occur often within local regions of the proteins, scattering techniques are often too insensitive to allow their detection and characterization. Another approach is based on kinetic studies. Kinetic investigations of multistep reaction systems allow, for example, to distinguish between fast initial binding and slower subsequent conformational rearrangement processes. When kinetic studies are carried out at variable pressure, it is not only possible to characterize the equilibrium ground states but also the involved transition states in terms of volume changes.
2. CATION BINDING TO Na'/K'-ATPase Some cation pumps responsible of active transport can be characterized by high alkali ion affinities as well as high cation selectivities. The integral membrane protein Na'/K'-ATPase is very similarly structured in all eucaryotic organisms. It actively transports Na' out of and K' into cells against the respective concentration gradients. This process requires catalysis of Mg2'-ATP hydrolysis and takes place until a certain high-energy state is reached. It is believed that the protein can assume at least two markedly different conformations, El and E2, which exhibit different alkali ion binding properties (Fig. 1). Besides the selective binding and
Chemistry: Presentations
33
occlusion of Na' and K' as well as of Mg'", it is assumed that several other ions, such as buffer cations and protonated amines are also bound to this enzyme [1,2]. The binding of these additional cations may affect the equilibrium between the main conformational states El and E2. By labelling the enzyme with the covalently bound fluorescent dye fluorescein 5'isothiocyanate (isomer I) (FITC), partial reactions including conformational changes of the dephosphoenzyme can be studied. Variable pressure kinetic studies of cation binding to the enzyme allow to characterize relevant protein conformational transitions occurring upon cation binding. Such structural transitions have been postulated already by Albers and Post, and are thought to regulate the direction of the sidedness and thus the function of the active alkali ion transport by this integral membrane enzyme.
.'.
3 Na'
2K'
3 Na'
Figure 1. Schematic illustration of the extended Albers-Post scheme for the cation transport by Na'/K'-ATPase
3. EXPERIMENTAL In order to compare non-selective cation binding to the selective K' binding [1,2,3,4], a variable pressure kinetic study has been performed with our high-pressure stopped-flow fluorometer [5] up to 100 MPa at 298 K for the non-selective choline (Ch') and 1,5diguanidiniumpentane (G5G2+)binding to the membrane bound preparation of the labelled enzyme, isolated and purified from pig kidney [6].The FITC-enzyme is prepared similar to
[7,8].In contrast to the selective binding of K', non-selective Ch' or G5G2+binding is indicated by a fluorescence intensity increase of the FITC enzyme (E) under the condition of low ionic strength. Under pseudo-first order conditions (large excess of cation), stopped-flow experimental data traces are single exponential (Fig. 2).
34
High Pressure Food Science, Bioscience and Chemistry
$a = 0.015 mM
C,= 4.75x1U7M'
.-
4.21
/"
'v
100 MPa
C ,
74
C,= 5x1V7 M T=298K
-04
0
20
40
0
time (sec)
= 0.6mh
D
50
time (sec)
Figure 2. Stopped-flow fluorescence experiments at two different pressures for the K' (A, in 10 mM imidazole/HCl, in 0.1 mM EDTA) and G5G2' (B, in 25 mM histidineMC1, in 0.1 mM EDTA) binding to membrane-bound FITC-Na'K500 nm. ATPase: pH 7.5,298 K, I.,,,= 466 nm, hemi>
4. RESULTS For the non-selective binding of the monovalent Ch' and divalent G5G2' cations the dependence of the observed rate constant, kobr,as a function of the total cation concentration exhibits saturation behavior (Fig. 3), as previously reported [9]. This suggests that at least a two-step process exists, based on a 1:l stoechiometry (reaction 1). Here E may represent the conformational state E l . The conformation of the state EC"' has been denoted F1 [ 1,4].
Ki E + Cn+W E-C"' fast 4
k23 W
EC"'
(reaction 1)
k32
I
cchc,'M CG5,bM ' Figure 3. Pressure dependence of the observed rate constant, kobs,for the non-selective binding of Ch' (A) and G5G2' (B) to membrane bound FITC-Na'lIC-ATPase in 25 mM histidinekIC1 containing 0.1 mM EDTA, pH 7.5, 298K; P h4Pa: 2 (+), 25 (V),50 (A),75 (*),100 (B).
35
Chemistry: Presentations
If the initial binding step (characterized by Kl) which is unresolved on the stopped-flow time scale, is fast compared to the resolvable step (characterized by h3and k3J,and as the reactions have been investigated under pseudo-first order conditions (at least a ten fold excess of cation), the observed rate constant, kobr,can be expressed by eq. (1). C"+
kobs = k23
+ k32
The equilibrium and kinetic parameters at zero pressure were fitted simultaneously (Figure 3) to eq. (l), (2) and (3), where k, (KO)is the rate (equilibrium) constant at zero pressure, AP ( A T ) the volume of activation (reaction), P the pressure, T the temperature and R the ideal gas constant.
K, = KOaexp(-;v;p) The results are the following (T = 298 K) ; A) Ch+ binding: b3,0 = 4.2 f 0.4 0.05 s-',
= 7 *2
M-', B) G5G2' binding: b3.0= 0.51
f 0.06 s-',
k32,0=0.14
(3)
d,k32p= 0.41 f 0.03 s.', Kl,o=
(1.3 f 0 . 5 ) ~lo4 M-'. The activation volumes, expressed in cm" mol-', are given in the volume
profiles shown in Figure 4.
-
B
......... ......... A
II * r AVJ
= +31M
..........-. AVV,:
+91f8
.
AV= ~ +60H
-.............
..........
I
..
--
Ava*
+93&8
>
4
v, = +37f27
.....-....... t.......................................................
reaction coordinate
i + G 5 f f E-G5GN
I
f.
EG5G" reaction coordinate
Figure 4. Volume profiles for the non-selective binding of Ch' (A) and G5GZ+(B) to FITCNa'/K'-ATPase. The symbol { $} characterizes the transition state.
36
High Pressure Food Science, Bioscience and Chemistry
5. DISCUSSION
Choline and 1,5-diguanidiniumpentanebinding are typical examples for non-selective binding of cations to a negatively charged site of Na'/K'-ATPase
[1,2]. The overall binding reactions
occurring in the time range of seconds at ambient pressure have been observed in the time range of minutes under our high pressure conditions. As for K' binding [3,9], the initial, fast binding step, is followed by a slow subsequent rearrangement process that is attributed to a protein conformational transition. The binding steps of both cations are characterized by a large increase of the reaction volume around +40 and +20 cm3 mol-'. In the case of nonselective cation binding, the subsequent rearrangement leads to a further increase of the partial molar volume around +50 to +60 cm3mol-'. Under these circumstances, the largest activation volume is expected for the reaction step characterized by
b3which is found to be unusually
high for both cations, namely around +90 cm3 mol-'. It is thus concluded that the step characterized by
h3is
likely to be associated with the rearrangement of a large protein
segment. This hypothesis is consistent with the fact that the rate constant k,, is rather low for a conformational rearrangement. The reported kinetic investigations performed at variable pressure allow to characterize a selected single process of a particular reaction sequence of a protein in terms of volume changes concerning equilibrium ground and transition states. These results offer the possibility to quantify, for example, the extent of conformational rearrangements in solution. The obtained results provide structural information that are invaluable for an interpretation of interactions and transitions at the molecular level.
6 . ACKNOWLEDGMENTS The authors wish to thank the Swiss National Foundation for financial support and Mrs. A. Schacht for expert technical assistance.
Chemistry: Presentations
31
7. REFERENCES
1
E.Grel1, R. Warmuth, E. Lewitzki, H. RufAcm Physiol. Scund. 1992,146,213 and 1993,
147, 343. 2
E. Grell, R. Warmuth, E. Lewitzki, H. Ruf The Sodium Pump: Recent Developments G.H. Kaplan, P. De Weer (Eds), The Rockfeller University Press : New York, 1991,441.
3
P. Bugnon, M. Doludda, E. Grell, A.E. Merbach High Pressure Research in the Biosciences and Biotechnology K. Heremans (Ed.), Leuven University Press : Leuven,
1997,143. 4
E. Grell, E. Lewitzki, H. Ruf, M. Doludda The Sodium Pump E. Bamberg, W. Schoner (Eds), Steinkopff : Darmstadt, 1994, 617.
5
P. Bugnon, G. Laurenczy, Y. Ducommun, P.-Y. Sauvageat, A.E. Merbach, R. Ith, R. Tschanz, M. Doludda, R. Bergbauer, E. Grell, Anal. Chem. 1996,68,3045.
6
P.L. Jorgensen Biochim. Biophys. Actu 1974,356,36.
7
S. J. D. Karlish Nu,K-ATPase: Structure and Kinetics J.C. Skou, J.G. Norby (Eds), Academic Press : New York, 1979, 1 15.
8
P. Porschke, E. Grell Biochim. Biophys. Acta 1995,1231, 181.
9
M. Doludda, E. Lewitzki, H. Ruf, E. Grell The Sodium Pump E. Bamberg, W. Schoner
(Eds), Steinkopff : Darmstadt, 1994,629.
Fluorination of Carbonyl Compounds by DAST at High Pressure J.Box, L.M.Harwood, N.S.lsaacs and R.C.Whitehead Department of Chemistry, University of Reading Reading RG6 6AD, U.K. Amine derivatives of sulphur(1V) fluorides are useful fluorinating agents, less reactive than SF4 and more selective. DAST ( diethylaminosulphurtrifluoride)is a commercially available example, widely used in synthesis whose reactions include the conversion of the carbonyl function of aldehydes and ketones to a -CF2- unit. While reactions at ambient conditions are often slow and yields poor despite the use of a threefold excess of the fluorinating agent, it has
'
recently been observed that both can be improved by the use of high pressure. Despite the use of a threefold excess of the fluorinating agent. Thus, fluorination yields of 2-bromocyclohex-2-enone rise from 25% at ambient to 44% at 10 kbar and 59% at 19 kbar. It was of interest to measure the volume
profile for the reaction and to determine the cause of the improved yields under pressure. Benzaldehyde was chosen for study since it gives a clean conversion to u,u-difluorotoluenein dichloromethane at ambient temperature.
Scheme
Fig.1 shows progress of reaction curves at 1, 500 and 1000 bar from which it can be seen that not only do rates increase with pressure but also do yields.
39
Chemistry: Presentations
1.oo
0.90 0.80 0.70
0.60 0.50 0.40
0.30 0.20 0.10
0.00
+ A
1 BAR 500 BAR
0
700BAR
0
300
200
1w
500
500
640
Fig.1 Progress of reaction between DAST and benzaldehyde, 25'.
-4'40
-5.60
'
0
100
I
I
I
I
I
I
J
200
300
400
500
600
700
800
p I bar
Fig.2 Activation volume plot
From the rates, Fig.2, a volume of activation of -20 +/- 2 cm3m01~'was obtained. The mechanism of the reaction probably proceeds according to the Scheme; The first stage, bond formation and charge separation, would be associated with a volume of activation of this magnitude and is probably rate-determining. The third step also has probably a negative volume change due to charge separation but this would be irrelevant if subsequent to the slow step. In conclusion, high pressure clearly benefits reactions of this type and we are at present exploring the further synthetic capabilities of this reagent. 1.
J.M.Box, L.M.Harwood and R.C.Whitehead, Synlett,571, (1997).
The Influence of High Pressure on Reactivity and Selectivity in Transition Metal Catalyzed Reactions Anne Mengel, Stephan Hillers, Martin Glos, Kerstin Bodmann and Oliver Reiser* Institut fur Organische Chemie, Universitat Stuttgart, Pfaffenwaldring 55, D-70569 Stuttgart. .6: 49-7 1 1-6854330; fax: 49-7 11-6854269;e-mail:
[email protected]
Transition metal catalyzed reactions have become invaluable for organic synthesis.[ll Not only offers catalysis a most ecconomic way to perform organic reactions, but many transformations would not be possible at all without the aid of a specific transition metal / ligand combination acting as a catalyst. There have been great efforts to develop new catalysts to render organic processes more efficient both in terms of reactivity as well as selectivity. In this course many general principles have been discovered how a catalyst has to be designed to perform in an optimal way. Nevertheless, many catalytic processes leave much room for further improvement. There have been fewer studies which use new reaction techniques and conditions such as ultrasound, microwaves, supercritical solvents or high pressure for catalysis. We started a general study to evaluate high pressure as a tool in transition metal catalysis. Our interest focused both on aspects of reactivity, i.e. rate and catalyst perfomance, as well as aspects of selectivity, i.e. stereo-, regio- and chemoselectivity. Moreover, we wanted to explore the question if high pressure offers the possibility to develop new catalytic reactions, which would not proceed otherwise. Given the complexity of catalytic reactions, it is not easy to predict the effect of pressure on such processes in a general way. However, based on our initial studies in the area of palladium catalyzed coupling reactions some general conclusions could be drawn which are discussed in this paper. For transition metal catalyzed reactions, it has been suggested that incorporation such as oxidative addition and complexation of substrates to a metal are favored, extrusion such as reductive elimination and decomplexation are disfavored, and intramolecular processes such as insertion, migration and deinsertion should not be influenced by pressure.r21 However, the actual situation appears to be more complicated: the most crucial point seems to be the question if ligand exchange remains possible under pressureD1 so that a catalytic cycle can occur. There are two distinct mechanistic pathways by which ligand exchange
41
Chemistry: Presentations
reactions can take place. In electronically saturated metal complexes such as Pd(O)L4 the dissociation of a ligand must occur first before a new ligand can coordinate (dissociative mechanism). The dissociation of a neutral ligand, being usually the rate determining step, should be hindered rendering the overall process unfavorable under pressure: L'#8..pd..'' L L d
'L
AVs c 0
AVs > 0 L
L
L'
L,,..pd.."L' L N
L '
WO) 16 electrons
WO) 18 electrons
However, dissociation of an ionic leaving group should be favored by pressure due to electrostriction. Moreover, in electronically unsaturated metal complexes such as Pd(II)X2L2 the additional possibility of coordination of a ligand to the metal prior or simultaneously to the extrusion of another ligand exists (associative mechanism), If the rate determining step is the initial addition of the ligand, these processes should also be favored by pressure. L'
I
L I , . . Pd..#I x
" " Pd(ll)
18 electrons
L'
A
AVscO
L I , , . pd...'x
L'
'x
Pd(ll) 16 electrons
AVs < 0
L
111,.
0
Pd(ll) 14 electrons
These principles for ligand exchange should apply also to oxidative addition and insertion and vice versa to reductive elimination and deinsertion processes. To allow an oxidative addition or an insertion to take place, the substrate in question must first be able to associate with the metal center. Subsequently, the reorganization of the substrates in the coordination sphere of the metal leading to the oxidative addition or insertion should proceed without significant changes in volume and therefore be unaffected by pressure. Consequently. a reasonable assumption appears to be that the rate of transition metal catalyzed reactions is influenced roughly on the same order of magnitude as it has been observed for simple ligand exchange reactions. We determined the activation volume for the palladium catalyzed cross coupling reaction of 2,3-dihydrofuran l a and iodobenzene to be -12 cm3/mo1,[4] which is indeed in good agreement with activation volumes for ligand exchange reactions of palladium(0) and palladium(I1) complexes.[3]
42
High Pressure Food Science, Bioscience and Chemistry
0
+ Phl pressure 2a 2b 2c
la:X=O 1b: X = NC02Me Ic: X = CH2
3a
3b 3c
In the course of this study we realized that the lifetime of the catalyst is dramatically improved by pressure. Subsequent investigation of this phenomenon revealed, that in the cross coupling of cyclic alkenes 1 with iodobenzenes at 8 kbar / 60°C turnover numbers of up to 100000 were realized compared to about 300 at normal pressure under otherwise unchanged conditions. By raising the reaction temperature to 100°C. turnover numbers of up to 770000 could be reached at a pressure of 8 kbar.[51 Similiarly, pressure was found to be beneficial in palladium catalyzed allylation reactions, while for the Suzuki coupling, pressure was found to have an adverse effect. Using pressure, milder and more economical reaction conditions can be used for certain reactions. Thus, we found that the alkenylation of aldehydes with phosphonates (HornerWadsworth-Emmons (HWE) reaction) is readily accomplished at room temperature under pressure in the presence of just triethylamine, while under normal pressure stronger bases or alternatively stochiometric amounts of a Lewis acid such as lithium chloride[6] have to be employed. Aryl aldehydes are alkenylated particulary well under these conditions, while with alkyl aldehydes double bond isomerization gives rise to side products in some cases. RCHO 4 Arl
+
+
Pd(0) I NEt3
-
8-10 kbar, 80°C
Ph
\
p-OMe-Ph H Me
[“lo]
€1
79
p-OMe-Ph 7
0
Ar
L C 0 2 M e Ar
5 ( Me0)2(0)PnC02Me
yield
R
R
73
H
80
27~73
p-OMe-Ph
90
70:30
Ph
56
s95:<5
Scheme 1. Using this alkenylation protocol, a new domino process could be designed for the synthesis of trisubstituted alkenes 7 by combining a HWE and a Heck reaction (Scheme 1). Pressure proved to be beneficial not only for the alkenylation step but also for the subsequently following arylation via a Heck reaction, since disubstituted alkenes are generally considerably
43
Chemistry: Presentations
less reactive in such coupling reactions than monosubstituted ones. However, in the case of RitAr, mixtures of (E)l(Z) isomers are formed, as was also noted in Heck reactions with cinnamic esters under normal pressure conditions.[7] Selectivity in High Pressure Reactions Equally important to the question of reactivity is how the selectivity of a reaction, and in particular of a metal catalyzed reaction, is changed under the influence of pressure. There have been very few studies which conclusively evaluate pressure as a parameter to control selectivity. In the area of cycloadditions a number of cases are reported to proceed with improved selectivity under pressure, but usually the reaction temperature could be simultaneously lowered so that it is not clear whether a temperature effect might be also responsible. Nevertheless, based on the data in the literature and our own results some general rules can be proposed: I. Selectivity which is based on favoring one of two diastereomeric transition states on steric grounds will be reversed by increasing the pressure. 1.3cm
A
1.3 cm c
I
high
CH3
H3C
3- )( normal
I
.
2.5 cm
pressure
pressure +
8
-A
I
x2
10
v 11
Scheme. 2. This rule seems to be contraintuitive since one might expect that steric repulsion under pressure is increasing thus favoring the transition state having the least steric interactions. However, it seems reasonable to assume that a given transition state of a reaction is very little influenced by pressure in respect to bond distances, and consequently the steric hindrance will not increase by raising the pressure. Nevertheless, from two different transition states the one having the smaller volume, i. e. being more compact and therefore sterically more crowded, will become more accessible under pressure, as schematically pictured in Scheme 2. In agreement with this rule we found in the palladium catalyzed telomerization[*l between butadiene (12) and benzaldehyde (13) at 7 kbar an enhanced preference for the formation of products with at least one axial positioned substituent (Scheme 3).
High Pressure Food Science, Bioscience and Chemistry
44
Pd(OAc)2I PPh3 I kPrOH PhCHO
*
1OO"C, 6h 12
13
14
15
1 bar
48
36
7
9
7 kbar
37
22
21
20
16
17
Scheme 3. Also, Yamamoto has found that the allylation of aldehydes with allylstannanes proceeds under pressure preferentially via sterically more compact transition states, i.e. chair type transitions states having the larger subsituents oriented in an axial position or alternatively via boat type transition states. Consequently, we propose that
2. A direct increase of selectivity with pressure, which is based on the differentiation of two diastereoineric transition states, can only occur if at normal pressure the sterically more dernnnding trmsition state is already preferred. A well known example for such a situation is the Diels-Alder reaction, in which often the erido-product via the sterically more hinderend transition state is preferentially formed. Consequently, enhancements of diastereo- and enantioselectivity by pressure has been demonstrated in such reactions.[9] We have evidence that the rules outlined above also apply to transition metal catalyzed reactions. However, since selectivity in metal catalyzed reactions is governed by ligands which coordinate with the metal, additional opportunities for the control of selectivity by pressure arise. An important principle which was revealed in the course of our investigation seems to be that
3. Selectivity can be altered in metal catalyzed reactions on grounds that bimolecular pathways (e.g. ligand exchange reactions) are preferred to unimolecular ones (i.e. isomerization) by pressure.
Chemistry: Presentations
45
This was illustrated again in the cross coupling reaction of 2,3-dihydrofuran and iodobenzene. It was found that at normal pressure the ratio of the regioisomers 2a to 3a is very little altered by the amount of triphenylphosphine present (2a:3a= 5:95 with no PPhg; 10:90 with Pd:PPhg = 1:60), while a reversal in regioselectivity is observed at 10 kbar (2a:3a = 10:90 with no PPhg; 75:25 with Pd:PPhg = 150). Apparently, the bimolecular Iigand displacement of 2a from the rc-complex 19 is accelerated by pressure compared to the unimolecular isomerization to 18 (Scheme 4).[Io1
2a
-'%>
3a
PPh3 +
f'dX H
w
unimolecular
bimolecular
+
HPdX(PPh3)
HPdX
18
19
Scheme 4. A similar pressure effect on regioselectivity was subsequently reported for palladium catalyzed [3+2]-~ycloadditions[~as well as for the palladium catalyzed coupling of bromobenzene and methylacrylate.[2] By the same principle, pressure should also offer the possibility to alter chemoselectivity in transition metal catalyzed reactions. While at normal pressure the palladium catalyzed, intermolecular coupling of acid chlorides with alkenes can only be accomplished using aroylchlorides (R = Ar) and vinyl ethers (X = OR),[12] under high pressure a considerable
broader range of acid chlorides and alkenes becomes available (Scheme 5).
R
1
CI
P
X
0 D
Pd(OAc),, NR,
-
R
20
R = Aryl, Adamantyl, 'Bu X = OR, COzR, Ph
R
U PdCl
8-10 kbar
up to 70% 22
21
I
AV* > 0
RPdCl
23
Scheme 5.
Our initial working hypothesis has been that after insertion of Pd(0) into the acid chloride 20 to give rise to the palladium acyl complex 21, the application of high pressure would prevent the CO extrusion to 23 but rather allow coupling with an alkene to give the a,punsaturated ketone 22. Indeed, at pressures above 7 kbar exclusively 22 is obtained, while no
High Pressure Food Science, Bioscience and Chemistry
46
coupling products arising from 23 were observed.[l3] However, it was surprising to note that acylation would always occur at the P-position of the alkene. This was especially true using vinyl ethers, contrasting the outcome of a regular Heck-reaction with such substrates, e.g. the previously discussed coupling of l a and iodobenzene. Moreover, electron rich alkenes are considerably better substrates in this reaction than electron poor alkenes, which is also not the case in Heck reactions. Further doubt on the above mechanism was shed by the fact that if catalytic amounts of triphenylphosphine are present, which should facilitate the reduction of Pd(I1) to Pd(0) and furthermore stabilize Pd(0) complexes, the reaction does not take place at all. W e therefore suggest that alternatively Pd(I1) might act as a Lewis acid which in the combination with high pressure might activate the acid chloride sufficiently to undergo an electrophilic addition to the alkene. Further corrobation of this hypothesis is found in the fact that electron rich alkenes such as 2,3-dihydrofuran (la) can be acylated with benzoyl chloride (24) at high pressure in the absence of a palladium catalyst to give 25 as the sole product.
PhCOCl
+
0
NEt3 10 kbar, 70°C
72 h 24
la
ph
0
55% 25
Acknowledgment: This work was supported by the Volkswagenstiftung (AZ 68/870), the Deutsche Forschungsgemeinschaft (RE948-2),the Fonds der Chemischen Industrie, Hoechst AG, Bayer AG and Degussa AG. OR is indebted to the Winnacker foundation for a fellowship. References [l]
[2] [3] [4] [5] [6] [7] [8] [9] [lo]
[I I] [I21 [I31
Recent Reviews: (a) W. A. Herrmann and B. Cornils, Angew. Chem. 1997,109, 1074. (b) Comprehensive Organometallic Chemistry 1996, Vol. 12, eds. G . Wilkinson. F. G . A. Stone, and E. W. Abel, 2nd Ed., (Oxford). T. Sugihara, M. Takebayashi, C . Kaneko, Tetrahedron Lett. 1995, 36,5547. c j R. v. Eldik, T. Asano, W. J. 1. Noble, Chent. Rev. 1989, 89, 549. S. Hillers, 0. Reiser, Chem. Conimun. 1996,2197. S. Hillers, S. Saratori, 0. Reiser, .I Am. . Chem. SOC. 1996, 118, 2077. W. C. Still, C . Gennari, Tetrahedron Left. 1983,24,4405. M. Moreno-Manas, M. Perez, R. Pleixats, Tetrahedron Lett. 1996,41,7449. K. Ohno, T. Mitsuyasu andJ. Tsuji, Tetrahedron 1972,3705. L. F. Tietze, T. Hubsch, E. Voss, M. Buback and W. Trost, J. Am. Chem. Soc.1988, 120,4066. S. Hillers, 0. Reiser, Tetrahedron Letr.. 1993,34,5265. B. M. Trost, J. R. Parquette, A. L. Marquart, J. Am. Chem. SOC.1995, 117,3284. C.-M. Anderson, A. Hallberg, J. Org. Chem. 1988,53,4257. cf. H.-U. Blaser, A. Spencer,J. Organomet. Chem. 1982,233, 267.
Plenary Lecture High Pressure Studies in Inorganic Chemistry: Volume Profiles for the Activation of Small Molecules by Transition Metals Siegfried Schindler* and Rudi van Eldik Institute for Inorganic Chemistry, University of Erlangen-Niirnberg , EgerlandstraBe 1 , 9 1058 Erlangen, Germany. Small molecules (dioxygen, carbon dioxide, carbon monoxide etc.) play an important role in biological reactions and in technical processes. Nature uses enzymes for very selective reactions that are usually difficult to perform with catalysts in chemical industry; for example methane monooxygenase is responsible for the selective oxidation of methane to methanol. A basic idea of bioinorganic chemistry is to exploit nature's methods for
development of new catalysts for selective reactions with small molecules in industry. To understand more of the mechanisms of these reactions two approaches are used simultaneously. First it is important to perform detailed kinetic studies on the metalloproteins themselves and second it is necessary to model the metalloenzymes with low molecular weight transition metal complexes to study their properties and their reactions with small molecules. It has been demonstrated in the past that the results of high pressure studies are important for the elucidation of reaction mechanisms in all fields of chemistry.1 Establishing a volume profile for a chemical reaction can provide detailed insight into the mechanism. In this contribution it will be demonstrated how high pressure measurements (temperature-jump, stopped-flow techniques) are used to obtain further insight into the activation of small molecules, using dioxygen as an example.
High Pressure Food Science, Bioscience and Chemistry
48
Reactions with dioxygen at the active site of a metalloprotein follow a simplified two step mechanism as outlined in equation 1. 0 2
Protein (Enzyme)
o
~
~
Substrate
-
F
~ * Product ~
(1)
In the first step dioxygen is bound reversibly to the metal center while in the second step reaction with a substrate occurs, reforming the reduced metal site. The first step, the activation of dioxygen, can be studied separately for the known dioxygen carrier proteins hemoglobin, hemerythrin and hernocyanin. It was found for these proteins that their active sites are very similar to those of many metalloenzymes involved in oxygen transfer reactions. In the following, kinetic high pressure studies on the binding of dioxygen to hemerythrinz and a model complex3 for this protein are discussed. In contrast to the large number of biological systems in which myoglobin (or its tetramer hemoglobin) and hernocyanin are used as oxygen carriers, the respiratory protein hemerythrin (Hr) is comparatively rare and found in only a few marine phyla. The active site consists of a binuclear iron center, capable of binding one oxygen molecule. The oxygenation kinetics of hemerythrin can be described by a single step process (equation 2) kon Hr + O2 L H r 0 2
koff
The rate law for this scheme is given in equation 3 -d[Hrl/dt = d[Hfl2l/dt = 16n[ml[021+ k o f f [ r n 2 1 (3) and the expression for the observed first order rate constant, determined with a T-jump relaxation method, is given in (4), where the subscript e refers to equilibrium concentrations prior to the temperature jump. kobs
= z -1 = k0" ([Hrle + [021e) + koff
(4)
49
Chemistry: Presentations
The data were plotted according to equation 4 as a function of pressure and k, could be determined rather accurately in this way. The decrease in k, with increasing pressure results in a A V O , value of +13.3
*
1.1 cm3mol-1. The
deoxygenation of oxyhemerythrin can be studied using high pressure stoppedflow techniques and an activation volume A V O @= + 52.2
* 0.7 cm3mol-1 was
obtained. Activation volumes are generally interpreted in terms of intrinsic volume changes (AV#iotr) resulting from changes in bond lengths and bond angles, and solvational changes (AVsOlv) resulting from changes in electrostriction on producing the transition state. The activation and reaction volume data can be used to construct the overall volume profile for the reversible reaction of hemerythrin with dioxygen (Figure 1). This profile is very similar to the one reported for the oxygenation of myoglobin, with the one difference that all volume parameters are almost double that found for myoglobin.4
Educts
Transition State
I
Product
Figure 1 Bond formation processes are usually characterized by a decrease in volume, i.e. a negative volume of activation. In contrast, the oxygenation of both hemerythrin and myoglobin is accompanied by positive AV# values,
50
High Pressure Food Science, Bioscience and Chemistry
although the structures of these proteins and iron active sites are completely different. The overall reaction volumes are considerably negative, in line with a bond formation process. Bond breakage in both systems is accompanied by positive A V 0 values ~ and an overall increase in volume, in line with our general expectations for such processes. It follows that the pressure dependence of the on reaction is not typical for a bond formation process and must be related to other effects. In the case of myoglobin, it was suggested that the rate-determining step for oxygenation involves the movement of
0 2
through the protein pocket,
during which desolvation of this molecule and a gate-effect of the protein, which are both accompanied by a volume increase, play an important role. If hemerythrin behaves very similar to myoglobin a possible explanation for the finding that all volume parameters for the reaction of Hr with dioxygen are twice as large as compared to Mb, may be related to the fact that the distances from the iron active site to the surface of the protein are about twice as large in Hr as in Mb (3.8
A in Mb and 6.3 A in Hr). For
Mb the large volume
collapse following the transition state was ascribed to Fe-02 bond formation, change in spin state from high to low of the iron center, and H-bonding of
0 2
with the distal histidine. It follows that conformational changes (breathing) of the protein are required during the bond formation process to account for the unusual AV'on found for myoglobin and hemerythrin. The AV'on value of +13.1
f
1 . 1 cm3mol-1 found for the oxygenation of
Hr may partly be ascribed to the desolvation of oxygen during its entrance into the protein. However, the value is too positive to be solely accounted for in this way if we compare it with the value of
+ 5 cmsmol-1 found for the oxygenation
of myoglobin. In this respect it is interesting to note that in addition to the decrease in the Fe-0 distance of the hydroxo/oxo bridge during oxygenation (from 200 to 180 pm), the Fe-Fe distance increases by 44 pm. We ascribe this
Chemistry: Presentations
51
opening up of the binuclear center to electronic effects. During the binding of 02,
electron transfer from the metal centers results in the reduction of 0 2 to
0 2 2 - and
oxidation of Fe(I1) to Fe(II1). The decrease in electron density on the
iron centers will cause a flow of electron density from the other ligands and a shortening of their bonding distances. This then causes an opening up of the binuclear center, which is accompanied by an increase in the bond angles. These effects could cause a partial overlapping of the orbitals in the transition state during the oxygenation process, which results in a shift in electron density and an increase in the Fe-Fe distance. The latter will then be accompanied by a volume increase and so account for the significantly positive value of A V O " . These effects can in principle account for the "breathing" motion of the protein. The significant volume decrease that occurs following the transition state can be ascribed to the oxidation of both Fe(I1) centers to the significantly smaller Fe(II1) species. Furthermore, the formation of the Fe-02 and 02-H bonds will also result in an overall volume collapse. These mentioned contributions are reversed during the deoxygenation reaction, since bond breakage and reduction of Fe(II1) to Fe(I1) will all contribute to the large volume increase. The fact that we are dealing with a binuclear iron center may partially account for the fact that AVlro@is ca. two times as large as that reported for the deoxygenation of the mononuclear oxymyoglobin center. The overall reaction volume of -39 rt 2 cm3mol-1 is also nearly double that reported for the oxygenation of myoglobin, and mainly results from the large contribution of AWOff.By combining this value with the partial molar volume of
0 2
it follows that V(HrO2) - v(Hr) = -11 cm3mol-1. This value
indicates that HrO2 is significantly smaller than Hr, which means that notwithstanding the increase in Fe-Fe distance during oxygenation of Hr, an overall volume decrease occurs to give HrO2 a more compact structure. This
52
High Pressure Food Science, Bioscience and Chemistry
will obviously result in a decrease in the iron-ligand bond distances, and a possible conformational change in the protein structure cannot be ruled out. A model complex for the active site of Hr and its reaction with dioxygen
is [Fe2(Et-HPTB)OBz]2+ (OBz = benzoate; Et-HPTB = N,N,N',N'-tetrakis[(N-
ethyl-2-benzimidazolyl)methyl]-2-hydroxy1,3-diaminopropane). The peroxo adduct is stable at low temperatures and its formation can be observed by stopped-flow techniques at room temperature prior
to decomposition
reactions? The effect of pressure was studied on the irreversible binding of dioxygen to [Fe2(Et-HPTB)OBz]2+between 10 and 140 MPa. The rate constant became larger with increasing pressure and an activation volume of A V = 12.8
* 0.9 cmsmol-1 was obtained at
20 "C. The activation volume for the
reaction confirms the highly structured nature of the transition state. Strongly negative volumes of activation are also found for the reaction of copper(1) complexes with dioxygen and were interpreted in terms of bond formation accompanied by intramolecular electron transfer to produce the copper(I1) peroxo species.5 The two examples shown here demonstrate that the results of kinetic high pressure studies are important for the elucidation of reaction mechanisms in the field of small molecule activation. 1) Hubbard, C. D.; van Eldik, R. Instr. Science Techn. 1995,23(1), 1. Magde, D.; van Eldik,
R. in High-pressure Techniques in Chemistry and Physics, Eds. Holzapfel, W . B., Isaacs, N . S. Oxford University Press 1997,267.
2) Projahn, H. D.; Schindler, S.; van Eldik, R.; Fortier, D. G.; Andrew, C. R.; Sykes, A. G . Inorg. Chem. 1995,34, 5935-41.
3) Feig, A. L.; Becker, M.; Schindler, S.; van Eldik, R.; Lippard, S . J. Inorg. Chem. 1996, 35, 2590.
4) F'rojahn, H. D.; Dreher, C.; van Eldik, R. J .
Am. Chem. SOC.1990,112, 17.
5) Becker, M.; Schindler, S.; van Eldik, R . Inorg. Chem. 1994,33, 5370.
Dense Gases as Reaction Media
V. Krmelj, 2. Knez* and M. Habulin
University of Maribor, Faculty of Chemistry and Chemical Engineering SI - 2000 Maribor, Smetanova 17, Slovenia
ABSTRACT Supercritical carbon dioxide (SC-C02), n-butane and mixture of n-propane:nbutane were used as reaction media for esterification of oleic acid with oleyl alcohol catalyzed by immobilized lipase from Rhizomucor miehei. Syntheses were performed in different reactors: batch stirred tank reactor and packed bed continuous operated reactor at atmospheric pressure and at high pressure. Influence of reaction parameters on initial reaction rates as well as on equilibrium conversion was studied. Results obtained in all types of reactor were compared.
54
High Pressure Food Science, Bioscience and Chemistry
INTRODUCTION Because of their excelent physico-chemical properties such as high diffusivity, low surface tension and relatively low viscosity, supercritical fluids have been used as a reaction media for enzymatic catalyzed reactions. The most used SC fluid in previous studies was CO2 (1 -4). The use of other dense gasses is not well documented.
MATERIALS
- enzyme preparation Lipozyme Ih4 (NOVO Nordisk, Denmark), activity 42 and 49 BIU/g.
- carbon dioxide 99.7 vol%, n-butane
99,9 vol%, n-propane:n-butane (70:30)
(Linde Plin, Slovenia),
- oleyl alcohol tech.65% (Aldrich Chemical, USA), oleic acid extra pure (Merck, Germany).
METHODS High pressure batch stirred tank reactor (HPBSTR) Equimolar solution of substrates was pumped into the thermostated reaction vessel. A certain amount of biocatalyst was added and fmally gas was pumped into the system until a desired pressure was reached (5,6).
Chemistry: Presentations
0
55
High pressure continuously operated reactor (HPCOR)
Gas and equimolar solution of substrates was equilibrated in the saturation column. Reaction proceeded at high pressure in the reactor which was packed with enzyme preparation. Gas was depressurized through expansion valves in two subsequent separators (5,6).
RESULTS
For lipase catalyzed synthesis of oleyl oleate in HP BSTR in SC-CO2 thermodynamic properties can be determined in a usual way. Enthalpy, entropy and free energy change were calculated on the basis of equilibrium constant (Ke) and temperature relationship. It was confirmed that enzymatic esterification does not require a high energy input (see Table 1). Table 1:Thermodynamicproperties for synthesis of oleyl oleate in BSTR in COZ.
56
High Pressure Food Science, Bioscience and Chemistry
Solubility of fatty acid compounds in n-butane are higher than in SC-CO2. Therefore in the high pressure continuous operated system test reactions at 150 bar and 40°C were performed in n-butane. The flow of the gas was 0.5 and 1L per minute. Results showed that conversion continuously decreases fiom 50% at the beginning to 22% after 300 min of the reaction operation while conversion of reaction performed in SC-CO2 at optimal conditions (T=40°C, P=80 bar) decreases for 2% in one month of continuous operation with the same enzyme preparation. Because of enzyme deactivation in HP COR further investigations were performed in high pressure batch stirred tank reactor where deactivation of Lipozyme was not detected. Esterification was done in n-butane and in the mixture of n-propane:n-butane. Initial reaction rates at 20°C (subcritical temperature for COz) are in n-butane approximately 2 times higher than in COz.and in the mixture n-propane:n-butane they are approx. 2.5-3 times higher than in C02. At 50°C the reaction rates in nbutane and in n-propane:n-butane mixture are higher in comparisson with C02 only at the pressure which is subcritical for C 0 2 (see Table 2 and 3).
Chemistry: Presentations
Table 2: Initial reaction rates (mol/(g .h)/g Lip.) in I-IP BSTR at 20°C .
Table 3: Initial reaction rates (mol/(g.h)/g Lip.) in HP BSTR at 50°C.
58
High Pressure Food Science, Bioscience and Chemistry
CONCLUSION The positive enthalpy shows that esterification is slightly endotermic. Higher positive entropy changes stimulate the synthesis of the ester and change of Gibbs free energy shows that the reaction is selfconducting while AG is less than zero. With increasing pressure AG aproaches zero what indicate a certain deactivation. COz in its supercritical state is more convenient reaction medium due to higher initial reaction rates than n-butane and the mixture of n-propane:n-butane.
REFERENCES 1. Aaltonen A., M. Rantakyla, Chemtech., 240, 1991.
2. Nakamura K., 3rd Sym.on SCF, Strassbourg, 3:121, 1994. 3. Dumont T., D. Barth, M. Perrut, J Supercrit. Fluids, 6:85, 1993. 4. Marty A., D. Combes, J.S. Condoret, Biocatalysis in Non-Conventional
Media, 425, 1992. 5 . Knez
z., V. Riher, M. Habulin, D. Bauman, JAOCS, 72(11):1345, 1995.
6. Habulin M., V. Krmelj,
Elsevier Science, 85, 1996.
2. Knez, High Pressure chemical Engineering,
Chemistry: Posters
Effects of Pressure on Thermal Response of Poly-(N-vinylisobutyramide) (PNVIBA) S.Kunug~,'*K.Takano,'N.Tanaka,,"and M.Akashib aDepartment of Polymer Science and Engineering, Kyoto Institute of Technology, Matsugasaki, Sakyo, Kyoto, 606, Japan. b Department of Applied Chemistry, Kagoshuna University, Koorimoto, Kagoshima, 890, Japan. There are several known synthetic polymers' that cany both hydrogen bonding properties and hydrophobic interactions in aqueous solutions and they show more or less distinct changes in their molecular level states (coil-to-collapse, in solution)2 and volume phase transitions (in gel Last year we showed' that the aqueous solutions of poly(N4nylisobutyramide) (PN VIBA)s*6exhibited characteristic pressure-temperature behavior, sirnilar to another thermoresponsive polymer, poly(N4sopropylacrylamide)(PNIPAM). Near atmospheric pressure, the transition temperature increased with increasing pressure, but, at pressures higher than 150 MPa, it decreased with increasing pressure to give the apparent turning point in the T-P diagrams. They were explained by the thermodynamic equation proposed by Hawley' for the pressure- and temperature-induceddenaturation of proteins and the results implied the possibility of studymg these polymers as synthetic models of socalled cold denaturation of proteins in reversible manner. Compared with PNIPAM solutions, PNVIBA shows much sharper transition.6 The much larger swelling and shrinking of the gels from PNVIBA have also been observed,' which afford us some possible applications such as drug delivery and bioseparation, as in the case of PNIPAM? but PNVIBA has much better characteristicsthan PNIPAM as for the toxicity. In the present report, we tried to certify the applicability of the Hawley's equation to the transition curves of these two polymers, by measuring the cloud point down to sub-zero temperatures. It takes advantage of the negative dP/dT of the freezing point of water due to its positive AV of fusion (Clausius-Clapeyron equation). By decreasing temperature under elevated pressure (200 to 300 MPa), the
62
High Pressure Food Science, Bioscience and Chemistry
cloud points can be measured down to ca.-20'C and we could find the turning point of AS = 0 experimentally. For this study, we have utilized fractionated samples of PNVJBA (and also of PNIPAM for the comparison) with relatively lower Mw/Mn values and examined the effect of molecular weight on the transition pressure, though several reports on PNIPAM solutions mferred very small or practically null molecular weight dependence of the transition temperature." Materials and Methods
PNVIBA was synthesized as previously reported.' N-isopropylacrylamide monomer was kindly donated by Kojin Co. (Tokyo, Japan), which was polymerized by a radical initiator after purification through precipitation fiom alcohol. They are fiactionated chromatographically and their molecular weights and molecular weight distributions were determined by GPC (Shodex AD-8OWS column in DMF solution, standardized against PEG-PEO). Three samples of PNVIBA (Mn = 1 1 kd and M w M =1.4, Mn = 66 kd and Mw/Mn =I .6, and Mn = 460 kd and Mw/Mn =2.4) and three samples of PNIPAM (Mn = 12kd and M w h h =1.5 ,Mn = 19kd and Mw/Mn =1.8 and Mn = 600kd and Mw/Mn =2.5) were used. These molecular weights are above the minimum domain size for PNIPAM to show an all-or-non transition." The cloud points for the aqueous solutions of these polymers were detennined by observing light transmission or light scattering. In the former case, a high pressure optical cell with two sapphire windows, constructed by Teramecs Co. (Kyoto, Japan), was placed between the light source (Xe lamp) and the monochrometer/photomdtipliervia optical fibers and the apparent transmittance at 500 nm was recorded, with changing either the temperature at constant pressure, or the pressure at constant temperature. The temperature of the cell was controlled by a Peltier-type thermoregulator and detected by a Pt resistance thermometer. In the latter case, another type of high pressure optical cell with three sapphire windows (constructed by Teramecs, Kyoto Japan) and an inner optical cell made of quartz was settled inside a conventional fluorophotometer (Shimadzu RF5000, Kyoto, Japan) and the apparent scattering light with the same wavelengths of excitation and emission (363 run) was recorded. In this apparatus, thermostated water containing ethyleneglycol was circulated through the cell-block and the temperature of the solutioninside the cell ('just outside of the inner cell) was detected by a Cu-constantan thermocouple.
63
Chemistry: Posters
The curve fittings of the data to the ellipsoidal curves were performed by a software Mathematica (Wolfram Research Inc. IL,USA). Results and Discussion
Hawley introduced a thermodynamic interpretation of the denaturation or absorbance change of some proteins’by considering up to the second derivatives of the free energy of transition (Eq.1). They predicted the ellipsoid diagram for the pressure-temperature dependence of protein denaturation (Eq.2) by assuming that T-TO ((T-To)/To << 1).
AG= ~p(p-P0)’/2+ A~(P-Po)(T-To) - ~Cp[T(ln(T/To)-l)+Tol + AVO(P-PO) - ASO(T-TO)+ AGO AG= A P(P-P0)’/2
+ AC~(P-PO)(T-TO) - AC~(T-TO)’/~/TO + AVO(P-PO) - ASO(T-TO) + AGO
(1)
(a,
where AP and A a are the (absolute) compressibilityand the (absolute) expansibility, ~ O The discussions by respectively. Eq.2 gves ellipsoid when A ~ A C- ACZ’>O Smeller and Herema& indicated that Eq.2 can be derived by second order Tayler’s development of the kee energy equation directly and that the consideration of higher order derivatives (i.e., compressibility, heat capacity andor expansibility are temperature and/or pressure dependent) resulted in distinct deviations from ellipsoidal features. We have shown that data for PNVIBA and PNIPAM solutions can be fitted to the ellipsoid curves’*’3and that the consideration up to the second order derivatives is sufi[icientfor the present system of the synthetic polymers. This gives a set of thermodynamicparameters, relative to AGO.When the center of the ellipsoid is in the first quadrant, then the apparent turning point of A V = 0 with increasing pressure and that of AS = 0 with increasing temperature can be observed; in other words, dP/dT > 0 near P-PO and T-TO and dP/dT
‘e
64
High Pressure Food Science, Bioscience and Chemistry
temperatures under appropriately elevated pressure. Taking advantage of these characteristicsof water here we have measured the T-P diagrams of transition down to approxiinately -20 'C. Figure 1 contains the results for three PNVIBA polymers of different molecular weights (a) and three PNIFAM polymers of different molecular weights (b). The ellipsoids reasonably explaining the obtained data are drawn by curve-fitting software. Tlie broken curyes show the freezing point of pure water to give the image of high-pressure cryotechnology affording the present measurements. The turning point of AS= 0 with increasing temperature was observed at 0-10 "c for PNVIBA and at -10 to 0 "C for PNIPAM.
> 1
Figure 1. T-P diagrams for the cloud point of PNVIBA (a) and PNIPAM (b) solutions down to sub-zero temperatures. (a)O.l wt/v%;(0)Mn=460x103; ( U ) M n = 6 6 x 1 0 3 ; ( A ) M n = 1 1 x103. (b) 0.5 wt/v %; (0)Mn = 600 x lo3; (0) Mn = 49 x lo3; (A) Mn = 12 x lo3. Open symbols, temperature scanning; closed symbols, pressure scanning. Broken curves indicate tlie freezing point of water at elevated pressures. Generally the cloud point temperature for these kinds of tliermo-responsive polymers is considered rather independent of the molecular weights; some reported tlie absence of such a dependence.'OActually transition temperature (T, ) at 0.1 MPa
65
Chemistry: Poxters
of the present polymer samples of different molecular weights differed only 2-3 "c; slightly lower Tmwas observed for higher molecular weight. On the contrary, the transition pressure varied more sensitively with molecular weight. Polymers of higher molecular weight gave lower pressure of transition and hence they gave smaller ellipsoids. The parameters calculated fiom these fitted elfipsoids are summarized in Table 1. They are given in the relative values to the standard fiee since the transition curves are too sharp to evaluate the energy of transition (AGO), independent values of the latter parameter, and an ellipsoid is uniquely determined by these relative values only. Although the axes of the ellipsoids are somehow inclined from the T-P axes, the axial lengths of the ellipsoids are mainly determined by the coefficients of the (T-Toy and (P-Poy terms,AC,and AP,and hence these two parameters seemed to take larger absolute values with increasing molecular weight, although the change in the case of AC,for PNIPAM was very small. Since the reference (standard) state of these thermodynamic analyses (0°C and 0.1 MPa) is well inside the inducted ellipsoids, the values of AGOare positive in all cases. Thus the absolute values of the parameters in Table 1 have the same signs with the relative values. The negative values of AC,have been reported for PNIPAM" and have been characterized also for PNVIBA (Akashi et al., unpublished study) by way of calorimetric measurements.
Table 1. Thermodynamicparameters explaining ellipsoid curves in Fig. 1. Polymer
Mn
AVdAGo
ASdAGo
(10") (1 02cm3cal) ( 1 OJk')
PNVIBA
ABiAGo AaiAGo ACpiAGo (1 05cm6) ( 105cm3cal K') (1 0 4 K ' )
-11.8 11 -1.76 0.887 -2.96 -6.68 66 1.36 -6.09 -9.38 460 2.48 3.54 1.08 Pl"AM 12 -2.09 49 -1.83 2.07 1.16 8.21 1.40 -1.68 600 The approximate deviations are about k 5%.
-14.1 - 7.29 10.1 5.00 4.75 10.3
-1.13
-1.52 -2.66 -1.57 -1.55 1.70
66
High Pressure Food Science, Bioscience and Chemistry
References (1) Burchard, W. in Chemistry and Technology of Water Soluble Polymers (C.A.Finch Ed.) Plenum Press, New York (1983) (2) Heskins,M.; Guillet,J.E.J.Macromol.Sci.,A2, 1441 (1 968). (3) Shibayama,M.; Tanaka,T. Adv.Polym.Sci., 109, 1 (1993). (4) Kunugi,S., Takano,K., Tanaka,N., Suwa,K., and Akashi,M. in High Pressure Research in the Biosciences and Biotechnology (K.Heremans Ed.) Leuven Univ. Press, Leuven, pp 59-62 (1997). ( 5 ) Akashi,M.; Nakano,S.; Kishida,A. .J.Polym.Sci. Part A: Polym. Chem.Ed., 34,301 (1996). (6) Suwa,K.; Wada,Y.; KikunagqY.;Morishita,K.; Kishcia,K.; AkaslqM. J.Polym. Sci.Part A: Polym.Chem.Ed,, in press. ( 7 ) Hawley,S.A.: Biochemistry, 10,2436 (1971). (8) Suwa,K., Wada,Y., Kishida,A., and Akashi,M.; J.Polym.Sci., in press. (9) Bae,Y.H., Okano,T., Hsu,R., W . S . W . ; Makromol.Chem., Rapid. Commun., 8, 481 (1987); Hofik~an.A.S. .J.Controlled Release, 6,297 (1987); Gustafsson,A., Wennerstrom,H., Tjerneld,F.; Fluid Phase Equilih., 29,365 (1 986). (10) Fujishige,S., Kubota,K., and Ando,I.; J.Phys.Chem.,93,3311 (1989); Kubota,K., Fujishge,S., and Ando,I.;.J.Phys.~hem., 94, 5154 (1990) (1 1) Tiktopu10,E.I.; Bychkova,V.E.;IZlcka,J.; Ptitsyn,O.B. Macromolecules, 27, 2879 (1 994); Tiktopulo,E.I.; Uversky,V.N.; Lushchik,B.B.; IUenin,S.I.; Bychkova,V.E.;Ptitsyq0.B. Macromolecules, 28,75 18 (1995). (12) Smeller,L.; Heremans,K. in High Pressure Research in the Biosciences and Biotechnology (K.Heremans Ed.) Leuven Univ. Press, Leven, pp 55-58 (1 997). (13) Kunugi, S., Takano, K., Tanaka, N., Suwa, K., and Akashi, M.; Macromolecules, in press.
Supercritical Fluid Fractionation of Polymers
S.Rey*, P.Botella*, Y.Garrabos*, F.Cansell*, J.L.Six** and Y.Gnanou**
* Institut de la Matiere Condensee de Bordeaux CNRS, Universite Bordeaux I avenue du Dr Schweitzer, 33608 PESSAC Cedex - France
** Laboratoire de Chimie des Polymeres Organiques, ENSCPB - Universite Bordeaux I - CNRS, 351 cours de la Liberation, 33405 TALENCE Cedex - France
The first series of experiments were meant to calibrate the fractionation apparatus and to establish the plot giving the molar mass of the extracted fiaction as a function of the pressure. Several linear PEO samples whose molar mass ranged fiom 500 to 20000 dmol were mixed so as to make a sample of very broad distribution of molar masses and then fractionated as a function of pressure. In the first time, the experiments were performed at 150°C in a pressure range varymg fiom 8 to 40 MPa with R142b (Tc=137"C, Pc4.25MPa) as solvent. The Mn value corresponding to each isolated fraction, at constant temperature, was found to increase with pressure as classically
High Pressure Food Science, Bioscience and Chemistry
68
observed for polymer chains in supercritical fluids. The maximum molar mass of the solubilized fraction, or
((
cutoff-mass
))
(Mc) for each pressure, is shown
in figure 1 . Mc corresponds to the molar mass of the slice that exhibits the lowest elution volume in the SEC eluogram. Even if the calibration curve shown in fig 1 gives the optimum pressure to apply for the extraction of species of gven molar mass, the problem becomes more complex when the sample to fractionate contains species of different architectures. The fractionation of a binary mixture (noted MI) made of three-arm stars (Mn=2800 g/mol) and of linear chains (Mn=900 g/mol) was thus attempted. In this case the contaminating chain was chosen so as to exhibit the same molar mass as that of the star branches. In the binary mixture (MI),
25000
E
-
I
I
1
I
Linear poly(ethy1ene oxide) at 150°C
h
1
zoooo
W
v
v)
s
v)
3
15000 3
r;.
8 5 V
10000
U
0
I
SO00 U
0
0
0 r
Figure 1 : Cutoff-mass of linear poly(ethy1ene oxide) in supercritical HCFC142b as a function of pressure at 150°C
Chemistry: Posters
69
the content in linear materials was purposely raised to 50% in weight so as to investigate the selectivity of this fractionation technique. As evidenced in figure 2, the separation of linear chains (Mn=900 g/mol) from the star structure (Mn=2800 g/mol) occurs as expected upon applying a pressure of 6 MPa showing that linear chains of 900 g/mol were selectively extracted. Another case involving star molecules and linear chains was also investigated. A binary mixture (M2) containing PEO star and linear samples of about the same molar mass (10000 g/mol) was then investigated. The fi-actionation using R142b as supercritical fluid was performed at 25 MPa. The value of the pressure to apply was drawn from the calibration curve (fig 1) established for linear chains. This curve shows that for the selective extraction of materials exhibiting mass lower than 10000 g/mol, pressure of 25 MPa is required. The figure 3 exhibits the SEC eluograms of recovered polymers compared with that of the parent mixture. Other experiments have been attempted to substitute the HCFC 142b by a non-toxic solvent. We have chosen to use a classical solvent as CO2 and a co-solvent as ethanol which is a good solvent of PEO samples at ambient conditions. The addition of a co-solvent in low quantities ( 5 to 10%) allows to improve the solubility of polymer. Moreover the capability to modify the cosolvent proportions allows to reach a good compromise between solubility and selectivity. So the mixture made of several linear PEO (molar masses ranged from 1000 g/mol to 35000 g/mol) has been fractionated. The experiments were prohibited in 2015. In a first approach, linear PEO samples have been
I0
High Pressure Food Science, Bioscience and Chemistry
fractionated as a function of their molar mass. Afterwards, the separation of linear PEO contaminants has been investigated with the view to obtain pure
PEO star samples.
a1
I8 MPa
W C lmcu ;hams m0verr.l u 6 MPa
, ‘ /
I
>
34
36
38
P~”,rmXrure
-
c 40
42
ve
(ml)
Figure 2 : Fractionation of binary mixture MI as a function of pressure, at 150°C by HCFC 142b
POE Stars recovered at 25 MPn
Parent mixture
Figure 3 : Fractionation of binary mixture M2 as a function of pressure, at 150°C by HCFC 142b
Chemistry: Posters
71
EXPERIMENTAL PART :
The apparatus used for SCF fiactionation, up 40 MPa and 400°C, essentially consists of a high pressure pump and a mass exchanger. The solvent or the binary mixture (solvent+co-solvent) is pressurised in liquid phase up to the required pressure then it flows through a heat exchanger which brings the fluid to the required temperature. The flow rate is adjusted by a micrometring valve located at the exit of the apparatus. The mass exchanger is a column of 0.8 cm diameter and 20 cm length. It’s filled with a glass packed bed of 0.2 cm diameter. Silica wool is placed at the inlet and outlet of the column to avoid the dragging of polymer by the flowing fluid. The characterization of linear and star PEO samples was performed by
size exclusion chromatography (SEC)with a refiactive index detector. SEC was run with either tetrahydrofuran (THF) or water solvent.
RESULTS AND DISCUSSION
Star polymers have been the subject of a keen interest over the last decade because of their peculiar behaviour both in bulk and in solution. During the synthesis, poly(ethy1ene oxide) (PEO) star samples are often contaminated with linear chains which can induce dramatic consequences on the material properties [ 1,2]. A fractionation step is also required to separate star molecules
12
High Pressure Food Science, Bioscience and Chemistry
from linear contaminants. Because fractional precipitation with a non-solvent does not satisfactorily work below the crystalline melting temperature of PEO (66"C), size exclusion chromatography and dialysis can represent alternative
approaches. Unfortunately, these techniques are not well suited to obtain large quantities of separated products. In order to improve and optimise the efficiency of the separation in time and in mass, we came to consider the use of supercritical fluids (SCF). Therefore, the SCFs have specific properties : in particular, they are characterised by a low viscosity and a large density. By small changes in pressure, it's possible to strongly modify the density, and the solubility of the solute. Indeed the SCFs are used to fiactionate a broad distribution linear polymers into narrow fractions. Fractionation of PEO has been performed in supercritical CO2 but the highest solubilized molar mass is up to 850 g/mol [3].
We have chosen to use
1-chloro-1,I-difluoroethane (HCFC 142b) as supercritical fluid following the studies of the fiactionation of the polystyrene in this solvent [4,5] and then CO2 more a co-solvent (ethanol) in order to substitute the toxic solvent which w i l l be performed with the supercritical mixture C02-ethanol (the ethanol ratio varying from 5 to 8% in mass), at 60°C and in a pressure range between 15 and 40 MPa. The critical temperature, in this case, is strongly lowered, which is important for the PEO stability. With a mixture C02-ethanol9515% in mass, the highest solubilized Mi is up to 2000 g/mol and when the composition reaches 92/8% in mass, the supercritical mixture isn't selective. So the adequate
13
Chemistry: Posters
composition, for an efficient fractionation, is 93% of COz and 7% of ethanol. The plot giving the molar mass of the extracted fraction as a function of the pressure is shown in figure 4. This calibration curve (fig 4) presents the same variations as in the first case : an increase of the cutoff-mass with the pressure. New experiments are in progress to validate results already found about separation of linear and star PEO with this supercritical mixture.
10
15
'0
25
30
35
40
P (MPa) Figure 4 : Cutoff-mass of linear poly(ethy1ene oxide) in supercritical mixture C02-ethanol (93/7% in mass) as a function of pressure at 60°C
REFERENCES 1 - Y. Gnanou, J. Macromol. Chem. Phys., C36, 1, 77 (1996). 2 - E. Cloutet, J.L. Fillant, D. Astruc, Y. Gnanou, "Macromolecular Engineering : Contemporany Themes", Ed. M. Mishra, Plenum Press, New-York, 47
(1995).
High Pressure Food Science, Bioscience and Chemistry
74
3 - M. Daneshwar, E. Gulari, J. Supercrit. Fluids, 5, 143 (1992). 4
- Ph. Desmarest, M. Hamedi, R. Tufeu, F.
Cansell, Third Inter. Symp. on
Supercritical Fluids, 3, 287, (1994). 5
- M.
Hamedi, F. Cansell, Ph. Desmaret, R. Tufeu, F.Cansell, High
Temperatures and High Pressures, 29 ( 1997) in press.
Viscosity under High Pressure of Pure Hydrocarbons and their Mixtures: Critical Study of a Residual Viscosity Correlation (Jossi)
J.Alliez, C.Boned*,B.Lagourette and A.Et-Tahir Laboratoire Haute Pression, Centre Universitaire de Recherche Scientifique, Avenue de 1' Universitk, 64000 Pau, France
Introduction: In petroleum engineering much of activities are influenced by the viscosity of the implied fluids. Models to simulate viscosity according to the pressure, the temperature and the composition of the fluids are essential. The correlation methods allow to estimate viscosity starting from another property which is often the density. Residual viscosity is defined as being the difference between viscosity T(T, P) and the viscosity q* known as of
"
diluted gas
"
generally evaluated at 1 bar, at the same temperature. The viscosity correlation most largely used in the oil models is that of the residual viscosity of Jossi et al (1) whose formulation is : [(q-q*)~+104]1'4=a,,+a,pI+~p~+a,p~+a4p~ pf=p/p, is the reduced
density, pe is the critical density and c=T,"6M'RP;m where T,, P, are the critical coordinates and M the molar mass. Stiel et a1 (2) propose for the pure substances: ~l*=3,4*10~T,0.~~/5 if TS1.5 and q' =17.78~'105(4.58T1-1.67)5m if
T,>1 (TI =T/T, is the reduced temperature). Besides a,,=O.l0230, a,=0.023364, q=0.058533, a,=-0.040758 and a4=0.0093324. These values result fiom an
High Pressure Food Science, Bioscience and Chemistry
76
adjustment carried out by Jossi et a1 (1) bearing out of eleven substances in a liquid and gas state: argon, nitrogen, oxygen, COz, SO,, methane, ethane, propane, isobutane, butane, pentane. With these coefficients q is in CPif T is in
K, P in atmospheres, p in g . ~ mand - ~ M in g.mole-I.For the mixtures one uses : q *=C%qiMi’nlC~M,’n and ~=(Cx,Tci)”6(Cx,Mi)-”2(Cx,Pci).U3
P,,, M, of the mixture are supposed to be linear forms of the characteristics of the pure substances (we noted LIN these
where pseudocritical coordinates T,
linear mixing rule). The object of this work is to test and improve the degree of reliability of this model and to determine the inaccuracy of the numerical values provided by simulation. New adjustment of coefficients a,: We extracted from our database (3) a selection of 2 148 experimental points for which one knows at the same time p and q, corresponding to twelve substances in a liquid or gas state: methane, ethane, propane, butane, isobutane, pentane, octane, decane, toluene, benzene, o-xylene, 2-2-dimethylpropane. Only the six first appear in the list of Jossi’s initial adjustment. The fitting corresponds to P400 MPa and 150Ka<520K. The new adjustment provides: a,,=O. 1019346, a,=0.024885, %=0.0507222, a,=-0.0326267 and a4=0.00758663.
By defining AAD =
-100 XINb 1- qcal(j) / qexp(j) 1 one obtains 10.6 % with Jossi j-1
and 6.0 % with the new set of coefficients a, (noted LABO). The maximum deviation is 41% with Jossi and 29.6 % with L B O . We checked that the fact of increasing the degree of the polynomial does not bring significant improvement if it is equal to or higher than 4. In certain cases one can be in the situation
where the density p is not known. Consequently, using all the pure substances
Chemistry: Posters
17
of the database of adjustment, we analyzed the results provided by the 2 sets of coefficients by generating the density with 4 equations of state : LK (4), SBRJ ( 5 , 6,), the cubic equation PR (7) and its modified form PRMOD (8).
Sometimes the use of these equations requires the knowledge of the critical coordinates (or pseudocritical for the mixtures) and of acentric factor a.Figure 1 represents the results (Em corresponds to calculations with the experimental
values of p). In a general way, it is better to use the set of coefficients L B O .
Results obtained on the mixtures: We extracted from the database 3240 experimental couples (P, T) relating to mixtures for which viscosity and the density were given. As previously P400 m a . These experimental values were separated in two groups. The first (2596 points) contains all the systems with methane, where for a very great number of couples P, T the systems are in a gas state. The second group (574 points) comprises only mixtures without methane, in the liquid state. To generate the pseudocritical coordinates we considered 10 different mixing rules :PED (9), TEJA (lo), WONG (1 l), SPDA (12), LK (4), PKP (13) and its modified form PKFVAR (3), HT (14), LIN, and LAB (3). These rules are discussed in detail in reference (3). It should be stressed here that when one generates p by an equation of state, generally it is necessary to know the pseudocritical coordinates, and thus a mixing rule appears also at this stage of calculation. The results show that when p should be calculated, the linear mixing rule LIN always gives a bad estimate of p and thus a very bad estimate of q. It is only when one uses the experimental values of p that the mixing rule LIN gives results on q comparable with the other mixing rules. In the same way, when the experimental values of p are used, the rules PED and WONG
ia
High Pressure Food Science, Bioscience and Chemistry
provide the same results for they generate the same T,
h4,,, , P,, i.e. same 5.
On the other hand, with the mixing rules PED and WONG, when one calculates initially p with an equation of state, the concerned acentric factors o are different, the calculated values of p also, and the results on viscosity are distinguished then. It is noted that for the two groups of data the results prove to be better with coefficients LABO than with the coefficients Jossi and this, almost independently of the coupling equation of state + mixing rule. It is systematically true when one uses the experimental values of p. The mixing rules PED, WONG, SPDA, LAB, are the most suitable. It is necessary to add the mixing rule LIN if one restricts to the experimental values of p (Figure 2). It should be noticed that if for binary with close components results are suitable (for example systems such as C,+C,, C,+C,, natural gaz GN...), they are poor for systems such as : cyclohexane+C,, or cyclohexane+C,,, C,o+C14, or toluene+pristane. Finally, an adequate mixing rule, with the set of coefficients
LABO and the experimental values of p , decreases the value of AAD (in the case of binary ClO+CO2: -34.9
% with TEJA and 5.6 YOwith PED).
Conclusion: In a general way, with the new set of LABO coefficients it is possible to obtain a representation more precise than with the set of coefficients of Jossi. The method of Jossi seems to be especially interesting for the restitution of viscosities of the systems containing light and close paraffins. But, punctually, for some pure substances or some mixtures, the opposite situation can occur. It appears that one should not use the cubic equation of state PR if one has to generate the density. With regard to the mixing rules, PED, WONG, SPDA and LAB rules (and LIN but only with the experimental values of p) are those which
79
Chemistry: Posters
'9 "t IS
10
L PR
EXP
LK
SBRJ
PR MOD
Fig.1 :Values of AAD versus the equation of state
(With pure compounds ;
16
9
:Jossi 0 : L B O )
d1 I 2 3c
3
I I -
Fig.2 :Valuesof AAD versus the mixing rule (With perpand the first group of data ; :Jossi 0 :LAl30)
High Pressure Food Science, Bioscience and Chemistry
80
lead to the most satisfactory results. One does not have however to expect a good systematic representation. If for light alkanes and their mixtures one can expect an average deviation lower than 10 YO,for certain pure substances (for example 2-2 dimethylpentane) one exceeds 20 % and for certain mixtures (for example some systems containing cyclohexane) the average deviations are higher than 40 %. The maximum deviations are then about 70 Yo. References 1. Jossi J.A., Stiel L.I., Thodos G., AIChE Journal, 8, 1962, 59-63 2. Stiel L.I., Thodos G., AIChE Journal, 1961,611-615
3. Et-Tahir A., Thbse de Doctorat, Universite de Pau, France, 1993 4. Lee B.I., Kesler M.G., AIChe Journal, 21, 1975,5 10-527
5. Behar E., Simonet R., Rauzy E., Fluid Phase Equilibria, 21, 1985,237-255
6. Jullian S., Thbse de Doctorat, ENSPM, France, 1988 7. Peng D.X., Robinson D.B., Industrial Engineering of Chemical Fundamental, 15, 1975,59-64
8. Peneloux A., Rauzy E., Freze R., Fluid Phase Equilibria, 8, 1982,7-23 9. Pedersen K.S., Fredenslund A., Christensen P.L., Thomassen P., Chemical
Engineering Science, 39, 1984, 101 1- 1016
10. Teja AS., AIChE Journal, 26, 1980,337-345 11. Wong D.S.H., Sandler D.I., Teja A.S., Eng. Chem. Fund. 23, 1984,38-441
12. Spencer D., Danner R.P., J. of Chem. and Eng. Data, 17, 1972,236-241 13. Plocker
U., Knapp H., Prausnitz J., Industrial and Engineering Chemistry
Process Design Development, 17, 1978,324-332 14,Hankinson R.W., Thomason G.N., AIChE, 25,1979,653-663
Calculation of Gas Solubility in Electrolyte Solutions at High Pressure,
High Temperature
H. Carrier, S . Ye, B. Lagourette, J. Alliez and P. Xans
Laboratoire Haute Pression, CURS, Av de l’universite, 64000 Pau, France
Many works have been developed to characterize liquid-vapor equilibrium (VLE) in electrolyte solutions. However, because of temperature, pressure and solution composition restrictions, specific to each component, these models do not allow to reply in a global manner to VLE determination in brines. Those restrictions constitute a limitation of size in the characterization of complex mixtures. This publication reports on calculations of gas (CH,,CO, and H,S) solubilities in brines using an homogeneous procedure based on a combination of the modified Helgeson, Kirkham and Flowers equations of state (MHKF) and
the Pitzer model. Our work puts in obviousness the thermodynamic coherence of the two models and provides a suitable procedure for the determination of
gas solubilities on a large area of pressure, temperature and composition. Basic equations of both models and thermodynamic relations used to perform VLE are presented. Solubility results calculated with this procedure are discussed and compared with literature data. MHKF equations of state
MHKF equations of state have been successfully applied to calculate the thermodynamics and transport properties of aqueous species at high pressures
82
High Pressure Food Science, Bioscience and Chemistry
and temperatures. Standard partial volume, heat capacity, entropy, apparent standard partial molal enthalpy and Gibbs free energy can be obtained for ionic species['], inorganic neutral species[21and organic speciesP1for temperature up to 1000°C and pressure up to 5000 bar. Equilibrium constant K, can be deduced from apparent standard partial molal Gibbs free energies of formation using the relation -
- AG; InK, = RT
If one considers the equilibrium between an aqueous and vapor constituent, following Shock et al. [31, the dissociation constant K, of the reaction is equal to the standard state fugacity of the neutral solute: (2)
K, =f,"
Pitzer model Numerous papers have been published in which Pitzer's ion-interaction equations are used to characterize gas solubilities in HP/HT brines. 14], ['I,
16].
Considering that, at the concentrations discussed in this article, we can neglect CO, and H,S dissociation in the liquid phase and molecule-molecule interaction, the activity coefficient of a dissolved gas given by Pitzer's model is: Inyg = 2 ~ c i m a+ 2 ~ c J mc+mamcrg,a,c
where pco) and
(3)
r stand respectively for gas-ion binary and ternary interaction
parameters. Electrical neutrality of the solution requires that all ion-neutral parameters involving one particular ion should be set to zero. In our study we set to zero the parameter involving Cl-.
83
Chemistry: Posters
Gas solubilities: Thermodynamic relations In an aqueous two-phase (liquid and vapor) system at constant temperature T and pressure P, equilibrium equation can be expressed in term of fugacities as: (4)
f i = rngygf,O
where yg is the activity coefficient on the molal scale, f i is the standard state gas fugacity and mg is the gas concentration in the liquid.
Vapor phase fugacity is calculated using a suitable equation (in this work, the Peng-Robinson cubic equation of state"]). The liquid phase is characterized using the combination of MHKF equations and the Pitzer model.
CO, and CH, solubilities in NaCl and Na2S04solutions Solubility values have been calculated by numerical fit of the above expression (Eq. 4). CO, and CH, solubilities are plotted versus NaCl or Na2S0,
concentration. For the MHKF equations of state's parameters, all the data are from referencesrz1'13]. For the Pitzer model, interaction parameters values are taken from references ['I,
and 19].
as,
CO, solubility in NaCl solution at T=40 and 160°C compared to experimental
a4
High Pressure Food Science, Bioscience and Chemistry
04
OJ
s01
01
0
w
M
60 PI..".
no
100
I b V l
CO, solubility in Na,SO, solution at T= 80, 120°C compared to experimental results[*] 05
T-1lOI
0.
03
s01
01
0
1.4
T-IMT
I
7.1 1.0
f O"
s 0s
I u ' 10
E
,,04
0.1
0.0
CH, solubility in NaCl solutions (1,2,4,6 m) at T= 150,210,240 and 270°C compared to data from Duan et a1.I9]
The (( salting out )) observed when sodium chloride or sodium sulfate is added to a mixture containing COz or CH, is well predicted by the above procedure. The deviations encountered are of the same order of magnitude as the experimental ones (8%).
85
Chemistry: Posters
H, S solubility in NaCl solution Several experimental investigations on the H2S-NaCl-H20 system have been carried out during the last decades[”]*
~
’
[I2], ~
Barta and Bradely[” fitted the
3
solubility data of Drummond[’ol to the semi-empirical interaction model of Pitzer. They introduced in their expression molecule-molecule interaction parameters. Since in the two previous cases we neglected those contributions, according to Rumpf et al. [‘I and Duan et al. f9] articles, we propose here new binary and ternary interaction parameters, obtained by regression of Drummond data without any molecule-molecule contribution. This approximation seems to be reasonable as the molality of dissolved hydrogen sulfide remains small and leads to an expression involving 12 parameters instead of a 15 one. The following expressions are proposed for the interaction parameters:
pcO) H2SNa*
= 19.754092 - 0.20387763 .T
+ 0.0008237638*T2- 1.6230109
10a.T3+ 1S594526 lOm9.P-5.8340662 10i3.T5 ‘H2S,Na*,CI-
= -4.759142
(5)
+ 0.049361213.T - 0.00020040153~T2 + 3.9781484 p+ 1.4646191 10i3,T5
10-7,T3-3.8618111
(6)
Vapor phase fbgacity is calculated using the Peng-Robinson equation as modified by Carroll et al.[I4].Tables 1 and 2 contain predicted H2S solubility values, experimental data from SuIeimenov-Krupp[”]and Drummond[’o]. A.R.D.
T
(“c)
(bar)
155.30 155.30 155.40 155.40 155.40 155.30 155.30 155.20 216.50 216.50 216.30 216.30 216.40 216.30 216.30
11.96 12.37 12.65 12.92 13.12 13.42 13.71 14.05 27.60 27.58 27.43 27.48 27.56 27.70 27.87
NBCI
0.5377 0.5641 0.5767 0.5888 0.5980 0.6103 0.6204 0.6320 0.2084 0.2189 0.2351 0.2432 0.2498 0.2597 0.2684
0.5014 1.2665 1.5652 1.8138 2.0320 2.2240 2.3911 2.5400 0.2353 0.6336 1.2445
1.4858 1.6932 1.8787 2.0434
Table 1
Ye
0.152 0.145 0.144
0.145 0.148 0.150
0.154 0.158 0.132 0.127 0.127 0.128 0.130 0.131 0.133
0.147 0.139 0.138 0.138 0.137 0.138
3.04 4.56 4.50 4.84 7.45
0.140
9.04 9.60 2.92 3.26 2.60
0.143 0.136 0.131 0.123 0.122 0.121 0.122 0.123
7.77
4.17
6.96 6.92 7.94
High Pressure Food Science, Bioscience and Chemistry
86
T (*C) 41 86.20 106.20 139.40 171.60 196.70 226.90 277.30 41 56 76.10 84.60 118.30 151.50
181.60 201.70 226.90 237 41
71.40 76.10 94.20 101.20 136.40 166.50 186.70 201.70 23 I .90
P
Y H2S
(bar)
6.969 10.404 12.136 15.934 21.881 28.617 40.61I 74.537 8.894 10.474 12.480 13.311 17.039 22.063 28.678 34.888 45.170 50.123 8.550 I 1.376 11.862 13.929 15.327 18.609 24.059 28.749 33.439 45.372
0.954 0.955 0.956 0.956 0.954 0.949 0.910 0.928 0.941 0.940 0.941 0.941 0.944 0.944 0.942 0.940 0.935 0.933 0.991 0.972 0.966 0.935 0.937 0.867 0.765 0.687 0.628 0.503
m
merp
m,,le.
NnCl 2 2 2 2 2.01 2.01 2.02 2.06 4 4 4
H2S 0.340 0.280 0.274 0.267 0.263 0.262 0.278 0.304 0.361 0.331 0.306 0.299 0.283 0.263 0.264 0.264 0.270 0.276 0.294 0.256 0.251 0.237 0.211 0.215 0.206 0.206 0.202 0.209
H2S 0.344 0.277 0.254 0.230 0.230 0.241 0.265 0.322 0.341 0.330 0.302 0.288 0.245 0.228 0.234 0.244 0.270 0.278 0.280 0.248 0.234 0.215 0.219 0.183 0.179 0.183 0.192 0.215
4
4 4.01 4.01 4.02 4.03 4.03 6 6 6 6 6 6.01 6.01 6.02 6.02 6.04 ~~
A.R.D. Oh
1.10 I .02 7.25 13.99 12.64 8.08 4.70 5.89 5.48 0.30 1.41 3.55 13.59 13.33 11.53 7.58 0.12 0.65 4.86 3.20 6.71 9.46 3.71 14.88 12.97 11.02 4.92 2.97
~
Table 2
As one can see, our calculated values compare closely with literature ones over
the range of investigation. Departures from experimental data can be compensated by regression of MHKF equations of state parameters. This remains to be undertaken but seems to be necessary to improve the previous results in the range of temperature from 100 to 180°C.
Conclusion As the close agreement between calculated and experimental data shows, the
proposed procedure, based on models which have theoretical justifications, could be used to characterize systems formed with gas, water and salts. The combination of MHKF equations and the Pitzer model, which have been also successhlly applied to the prediction of salts’ solubility in electrolyte solutions
Chemistry: Posters
87
at high pressure and high temperature conditions (reference 15) allows to process in an homogeneous and complete manner for phase equilibria prediction in aqueous solutions on a large range of temperature, pressure and compositions. References Shock E.L. and Helgeson H.C., 1988. Geochim. Cosmochim. Acta, 52:
2009-2036. Shock E.L. and Helgeson H.C., 1990. Geochim. Cosmochim. Acta, 54: 915-945. Shock E.L., Helgeson H.C. and Sverjensky D.A., 1989. Geochim. Cosmochim. Acta, 53:2157-2183. Edwards T.J., Maurer G., Newman J. and Prausnitz J.M., 1975. AIChE J., 24 : 966-975 Barta L. and Bradely J. 1985.Geochim. Cosmochim. Acta, 49 : 195-203. Rumpf B., Nicolaisen H. and Maurer G., 1994. Ber. Bunsenges Phys. Chem., 98 :1077-1081. Peng D.Y. and Robinson D.B., 1976.Ind. Eng. Chem. Fund 15 59-64. Rumpf B.and Maurer G., 1993.Ber. Bunsenges Phys. Chem., 97 : 85-97. Duan Z.,Moller N., Greenberg J. and Wear J.H., 1992. Geochim. Cosmochim. Acta, 56, 1451-1460. [lo]Dnunmond S.E., 1981. Ph. D. dissertation, The Pennsylvania States University. [ 1 13 Douabul A. A. and Riley J. P., 1979.Deep-sea Research, 26 : 259-268. [12]Barrett T.J., Andrerson G.M. and Lugowski J., 1988. Geochim. Cosmochim. Acta, 52 : 807-81 1. [13]Suleimenov O.M. and Krupp R.E., 1994.Geochim. Cosmochim. Acta, 58 : 2433-2444. [ 141 Carroll J.J. and Mather A.E., 1989.Can. J. Chem. Eng., 67 :999-1003. [ 151 Carrier H., Ye S., Vanderbeken I., Li J. and Xans P., 1997.High Pressure High Temperature (in press). [16]Nicolaisen H., 1994.Ph. D. dissertation, Technical University of Denmark LynSbY.
High Pressure Ultrasonic Speed and Related Properties in Petroleum Cuts
J.L. Daridon, B. Lagourette, P. Xans Laboratoire Haute Pression, UniversitC de Pau, Avenue de l'universite, 64000 PAU (France). Email :
[email protected]
1. Introduction
In some deep oil reservoirs, in which fluids are stored under high pressure and high temperature conditions, the light components in supercritical state are able to dissolve a significant amount of heavy components. The heavy components in large proportion may induce modifications of the thermodynamic properties of the reservoir fluids. To predict these changes, it is essential to know the properties of the heavy fraction throughout the pressure domain to which the crude oils are subjected between downhole and surface conditions. There is however, as yet, very little experimental data on these heavy fractions. In particular, thermophysical data of synthetic or real heavy fractions remain scarce under high pressure.
With a view to obtaining useful experimental data to develop or to test the abilities of thermodynamic models to predict properties of heavy fractions under pressure, we have determined, in this work, the thermophysical
Chemistry: Posters
89
properties of a real and a synthetic heavy cut with similar boiling temperatures. More precisely, ultrasonic measurements of speed of sound were carried out in the pressure range 0.1 -150 MPa and in the temperature interval 293 - 373 K. The sound speed measurements were then used to generate density data by integrating the inverse of the square sound velocity with respect to pressure. The isentropic compressibility then could be deduced from the simultaneous knowledge of ultrasonic speed and density. Finally, the acoustic measurements were used for the indirect determination of the isothermal compressibility.
2. Ultrasonic speed measurements
Ultrasonic speeds were measured by using a pulse technique in an autoclave cell closed at both ends by two identical bars which serve as a connecting medium between the piezo-electric transducers and the fluid tested. The presence of these connecting bars avoids the need for direct contact between the transducers and the fluid, and thereby isolates the transducers from the pressure constraints imposed on the fluid. The experimental apparatus and procedure have been described in an earlier article [Daridon (1 994)]. The effects of pressure and temperature on path length L were taken into account by the following relation : L(P,T) = Lo[1 + a( T - To)][1 + b( P - Po)]
(1)
where the coefficients a and b, which depend on the mechanical properties of materials composing the measuring cell, were determined by calibrating with
90
High Pressure Food Science, Bioscience and Chemistry
water. The reference data taken from literature were respectively those of Wilson (1959), Del Grosso et al. (1972) and of Petitet et al. (1983). The two systems studied were distillation cuts which have a boiling temperature around 250°C. The first system (Sl), which is a real cut, was obtained by separation of a crude oil by distillation. It has a boiling range fiom
230 to 270°C and its carbon number range from 11 to 16 (Table 1). It contains normal
is0 paraffins, naphthenes and aromatics. Whereas the second system
(S2) is a synthetic mixture composed of five components belonging to paraffins, naphthenes and aromatics with a proportion (by weight) of 40% for the components from the alkane family, 35% for the naphthene and 25% for the aromatic compounds in such a way as to correspond to the average compositions observed in real distillation cuts. Moreover, the five compounds (Table I) were chosen within each chemical family so that their boiling points are close to 250°C. Measurements in the liquid state were performed along 9 isotherms, ranging between 293.15 and 373.15 K, for pressure from atmospheric to 150 MPa in 5 MPa increments. ~
Synthetic cut (S2)
Real cut (Sl) components
mass %
02.3 1 10.55 33.86 3 1.37 17.5 1 04.40
components Tridecane Heptamethylnonane Heptylcyclohexane Heptylbenzene 1 -Methylnaphthalene
Table 1 Compositions of the systems studied
mass 'YO
20.00 20.00 35.00 15.00 10.00
Chemistry:Posters
91
Results (associated with acoustic signals of frequency 3 MHz) are plotted in Figure 1. It can be noted from this Figure which represents isothermal curves
for both cuts (real and synthetic), that the ultrasonic speedsmeasured in two systems are similar. The two sets of sound speed data deviate by 0.2% on absolute average and the maximum deviation observed does not exceed 0.4%. 2000 1800
-
0
50
100
150
P 1 MPa
Real cut :
~
Figure 1 :Ultrasonic speed measurements ; Synthetic cut : 293.15 K ; A 333.15 K ;0 373.151 K
3. Derived thermophysical properties
The ultrasonic speed, assimilated with the speed of sound, can be correlated with various thermophysical properties by means of the isentropic compressibility coefficient K~ :
and the thermodynamic relationship which links this coefficient to the isothermal compressibility coefficient K~ :
92
High Pressure Food Science, Bioscience and Chemistry
an expression in which a, designates the isobaric expansion coefficient and C, the isobaric heat capacity. Thus by replacing the product
P K ~by
l/u 2and by
integration on pressure, this relationship can be used to express density p with pressure :
%u-2dP+T
p(P,T)=p(Po,T)+
(4)
0
On the basis of p measurements carried out on the two systems at atmospheric pressure Po by means of an Anton-Paar densitometer (DMA 60 model) and on the knowledge of Cp also at atmospheric pressure [Bessiere (1997)], determination of densities could be extended from atmospheric pressure up to 150 MPa (fig. 2) by means of ultrasonic speeds using the Davis and Gordon
procedure (1 967) to calculate the right hand side of the relation (4). 900 3
,
I
I
I
880 860 840 820
.-
3
800 780 760 740
0
50
100
150
P I MPa
Figure 2 : Densities calculated from sound speed measurements Real cut : -; Synthetic cut : 293.15 K ; A 333.15 K ;0 373.151 K
93
Chemistry: Posters
In the same manner as speed of sound, the two sets of density data merge on graphic (fig. 2). Comparison qf density values, in same conditions of P and T, led to an absolute average deviation of 0.04% and to a maximum deviation of 0.1 %. Moreover the numerical procedure gives access to the derived properties K,(T,P), K~(T,P)over the entire pressure range investigated. The values
obtained for these two properties are plotted along a few isobars in figures 3 and 4. Similarly, the deviations obtained between the real and the synthetic systems during the comparative study performed on compressibilities are very small (0.4% and 0.7% for isentropic and isothermal absolute average deviations respectively).
L
1.2
k~7
I
0.8
0.4
293
313
333 T /K
353
373
Figure 3 :Isentropic compressibility Real cut : -; Synthetic cut : 0.1 MPa ; A 50 MPa ; 0 100 MPa ;+ 150 MPa
94
High Pressure Food Science, Bioscience and Chemistry
293
313
333
353
373
T K
Figure 4 :Isothermal compressibility Real cut : __ ; Synthetic cut : H 0.1 MPa ; A 50 MPa ;0 100 MPa ;+ 150 MPa 4. Conclusion In order to simulate the thermodynamic behaviour of heavy petroleum fiactions, it is common practice to reduce them to fictitious systems with a very small number of pseudo-components. The identity of the behaviour of the properties for the two systems explored in this study, throughout the pressure domain covered in the investigation, validates the choice of pseudocomponents and the proportions attributed to them. More generally, these results provide data which will facilitate the selection of pseudo-components for use in modelling heavy cuts of reservoir oils.
References : Bessieres, D. Personal communication 1997. Daridon, J.L. Acustica 1994,80,4 16-4 19. Davis, L.A.; Gordon, R.B. J. Chem. Phys. 1967,46,2650-2660. DeI Grosso, V.A. ; Mader, C.W. J.Acoust. SOC.Am. 1972,52, 1442. Petitet, J.P. ; Tufeu, R. ; Le Neidre, B. Int. J. Thermophys. 1983,4, 35-47. Wilson, W.D. J, Acoust. Soc. Am. 1959,31, 1067-1072.
Influence of the Pressure and Temperature on the Viscosity and Density
of Binary Water + DAA
C.Boned"', M.Moha-Ouchane", A.Allalb,J.Josec, M.Benseddikd
" Laboratoire Haute Pression, Centre Universitaire de Recherche Scientifique, Avenue de l'Universitk,64000 Pau, France Laboratoire de Physique des Matkriaux Industriels, Centre Universitaire de Recherche Scientifique, Avenue de l'UniversitC, 64000 Pau, France Laboratoire de Chimie Analytique I, (LICAS), Universitk Claude Bernard, Lyon I, 69622 Villeurbanne Cedex, France Dkpartement de Physique, Facultk des Sciences Dhar el Mahraz, Fes, Maroc
Introduction :
It is generally observed that the dynamic viscosity q of liquid mixtures is markedly influenced by pressure P. For many binary mixtures variations of q as a function of composition are monotonic. This occurs when interactions are weak. But in certain cases, when molecular interactions between the two solvents are stronger, non-monotonic viscosity behaviour can be noted, and the viscosity moves through a maximum or a minimum when the composition varies. The presence of the maximum or minimum depends on the P,T conditions of the study. Unfortunately non-monotonic systems have most commonly been investigated at atmospheric pressure. Studies versus pressure
%
High Pressure Food Science, Bioscience and Chemistry
are rare [ 1 to 41. It therefore seemed of interest to study the influence of P on other systems presenting a viscosity maximum or a viscosity minimum as a function of composition. We chose the water + 4-hydroxy-4-methyl-2 pentanone (or diacetone alcohol, abbreviated DAA) because at atmospheric pressure [5] it exhibits a very clearly marked maximum. Moreover another physical parameter which is important and often essential to determine (and model) viscosity q is the density p.
Experimental techniques and results: The viscosity q is determined with the aid of a falling-body viscometer, technical details of which are provided in reference [6].Values of p between 0.1 MPa and 70 MPa have been measured with an Anton-Paar DMA 60 + DMA 60 1 resonance densitometer with an additional 5 12 P cell. The values are extrapolated to P = 100 MPa according to the procedure described in reference [6]. The uncertainty of the temperature T is estimated at *0.5 K for measurements of 7,and *0.05 K the uncertainty of p. The uncertainty of pressure P is estimated to be *0.05 MPa for the measurement of p and kO.1 MPa for the measurement of q (except at P = 0.1 MPa). The uncertainty of p is less than 0.1 kg.m” (except at P = 0.1 MPa where it is estimated to be below 0.03 kg.m”) which corresponds to the estimate made by Papaioannou et al [3] with identical apparatus (with the 512 cell limited to 40 MPa, instead of the more recent 512P cell limited to 70 m a ) . The uncertainty of q is of the order of 2%, except at atmospheric pressure P = 0.1 MPa where the uncertainty is lower than 1%. The water (H20, molar mass M = 18.015 g.mole“) is distilled water. The DAA used is commercially available (CfjH1202 : Interchim, purity > 99%, molar mass M = 1 16.16 g.mole-’). The mixtures were prepared by weighing at atmospheric pressure and ambient temperature so to obtain the
molar fractions = 0.1,0.2,0.3,0.4,0.5,0.6,0.7,0.8 and 0.9.
Chemistry:Posters
91
Measurements of q and p were taken at 303.15,323.15 and 343.15 K. The reader will find the values in [7]. Measurements of the density p were carried out at pressures from atmospheric pressure 0.1 MPa to 65 MPa in progressive 5 MPa steps. There are 462 experimental values: 84 for the three pure substances and 378 for the three binary mixtures (xi# 0 and Xi# 1). Values of p extrapolated to 80 MPa and 100 MPa using a Tait type relationship [6] are also indicated. Measurements of viscosity q were made at 0.1, 20, 40, 60,80 and 100 MPa. A total of 198 experimental points is obtained: 36 for the two pure substances and 162 for the binary.
Discussion : For pure water one can refer to previous publications [8,9]. The behaviour of q of water is very particular since, depending on the value of T, either the curve representing q(P) is monotonic and rising (for T higher than approximately 303.15 K), or it presents a minimum (for T lower than approximately 303.15 K). Between 0.1 MPa and 100 MPa, q is practically constant at 303.15 K. At fixed pressure P, the viscosity q(T) of water falls when the temperature rises. In all cases, when comparison is possible, there is a good fit between our measurements and those in the literature, to within experimental error. The viscosity of D M increases with P and decreases with T. For the two pure substances within the pressure and temperature domain considered the density increases with P and decreases with T. The mixture exhibits nonmonotonic variation of the viscosity. Figure 1 shows the q(x,,P) surface at T = 303.15 K. It will also be noted that the maximum is very marked. The maximum is all the more marked as the pressure is high and the maximum moves in the direction of increasing x, when pressure decreases. At constant x, the behaviour is normal in the sense that q increases with pressure P and decreases with temperature T. It is interesting to calculate the excess activation energy of viscous flow AGEdefined by :
98
High Pressure Food Science, Bioscience and Chemistry
In(qV) = xlln(qlVl)+ x2ln(qz~,) + AG~/(RT) where V,=Mi/piis the molar volume of component i (molecular weight C xi.M, for the mixture) and R the constant of the perfect gases. This relation is a modified form of Katti and Chaudri’s equation [ 101 and is theoretically justified from Eyring’s representation of the dynamic viscosity of a pure fluid [ 113. Figure 2 shows the variations of AGE versus x, and P, at T=323.15 K The maximum value of AGEis very large, (as the uncertainty on viscosity is 2% then the uncertainty on AGE is about k100 J.mole‘’). AGE increases with P and decreases when T increases.
Figure 1 : Variations of q , at T=303.15 K, versus P and x,
Regarding the density p, its behaviour is usual: p is observed to decrease with T, increase with P, and vary monotonically with the composition Xi.
The excess volume can be calculated through VE=x,M,,(p”-pl~’)+x2M2,(p’’-
pi’). The apparatus used is sufficiently accurate to provide a satisfactory
estimate of VE.There is a marked effect. It is seen that VE is negative whatever T, P and x. At fixed composition and temperature lVEl decreases as pressure increases. At a given composition and pressure lVEldecreases when temperature
Chemistry: Posters
Figure 2 : Variations of AGEat T=323.15 K, versus P and x,
1.2
--0.30
0.2
0.4
0.6 0.8
I
XW
Figure 3 : Variations of VE(0)and p V E N(+) (in %.) at P=40 MPa and T=323.15 K , versus x,
100
High Pressure Food Science, Bioscience and Chemistry
increases. It should be emphasized here that it is necessary to be carehl with regard to the significance of the maximum or minimum of the excess volume VE.This quantity is relative to one mole of the mixture. On the other hand, the quantity p.VE/M is relative to one unit volume of the mixture. Figure 3 shows the variations of VEand p.VE/Mat T=323.15 K and P=40 MPa , versus the mole fraction x, of water. There is a notable discrepancy between the positions of the two minima. The contraction p.VE/Mis maximum (-3.06 %) at P=O.1 MPa and T=303.15 K. Which one of both quantities is really characteristic of the maximum effect of the intermolecular interactions ? From the point of view of excess volume (here contraction effect ) it seems to us that p.VE/Mis probably more significant. 1.
J.Zhang, H.Liu, J. of Chemical Industry and Engineering 3:268 (1 99 1)
2. T.Nakai, S.Sawamura, Y.Taniguchi, J. of Mol. Liquids 65/66:365 (1995)
3. D.Papaioannou, M.Bridakis, C.G.Panayiotou, J. of Chem. and Eng. Data
38:370 (1993) 4.
D.Papaioannou, C.Panayiotou, J. of Chem. and Eng. Data 40:202 (1995)
5. M.Gramajo de Doz, H.N.Solimo, Phys. and Chem. of Liquids 29: 79 (1995) 6. J.V.Sengers, J.T.R.Watson, J. Phys. Chem. Ref. Data 15: 1291 (1986) 7. A.Et-Tahir, C.Boned, B.Lagourette, P.Xans, Int. J. of Therm. 6: 1309 (1995) 8.
M.Moha-Ouchane, These de Doctorat, Fes, Maroc, (1998, in preparation)
9. J.Kestin, J.V.Sengers, B.Kamgar-Parsi, J.M.N.Levelt-Sengers, J. Phys.
Chem. Ref. Data 13: 175 (1984) 10. P.K.Katti, M.M.Chaudhri, J. of Chem. and Eng. Data 9: 442 (1964) 11. S.Glasstone, K.J.Laidler, H.Eyring, in "The theory of rate processes'; Ed.
Mc Graw-Hill, New-York (1941)
PRESSURE-INDUCED
CYCLOADDITIONS
TO
STRAINED
ARENES F.-G. K l h e r * ) "), R. Ehrhardt"), H. B a n d m a d , R. Boeseb),D. Blherb), K. N. Houk"),B. R. Beno") "'lnstitut fiir Organische Chemie, b)Institutfiir Anorganische Chemie, Universitat GH Essen, Universiatsstr. 5, D-45 117 Essen, Germany Telephone: ++49 (0) 201 183 3088, Fax: ++49 (0) 201 183 3082, e-mail: klaerneraoc 1.orgchem.uni-essen.de ')Department of Chemistry and Biochemistry, University of California, Los Angeles, CA 90095, USA
The planarization of the 1,3,5-~ycloheptatrieneskeleton, which usually exists in a boat conformation, could be achieved by fixation of one of the torsional angles to 0" by fusion of a rigid norbornane unit at C-1 and C-7 of the cycloheptatriene ring'). According to Dreiding molecular model considerations it should be possible to fix 1,3,5,7-~yclooctatetraene(COT) in a planar conformation by fusion of two norbornane units with the carbon atoms C- 1, C-2 and C-4, C-5 of the eight-membered ring. The introduction of two torsional constraints is here necessary in order to avoid that the planar eight-membered ring escapes planarization by a rapid x-bond shift, which is degenerate in derivative 1',*).Planar COT derivatives are interesting compounds with respect to the question of their antiar~maticity~'~). Hitherto, only two derivatives are
High Pressure Food Science, Bioscience and Chemistry
102
known, which according to crystal structure analyses contain planar eightmembered rings with n-bonds tending toward localization. Recently, parent planar COT has been observed by means of “transition-state ” spectroscopy. Accordingly, the singlet lies well below the triplet Dab COT in agreement with ab initio predictions that COT violates Hund‘s rule3). Here we report the synthesis and the structural properties of a dicyano-derivative of 1.
@-
+&
c _
1‘
1
The synthesis of the desired dicyanocyclooctatetraene derivative 6, shown in
Scheme 1
was
feasible
analogously
1,2-dicyanocyclooctatetraene
by
to
photolysis
the of
known the
synthesis
corresponding
2,3-di~yanobarrelene~). 1,4;5,8-Bismethano-octahydroanth1acene 3 prepared by
a highly
of
was
stereoselective Diels-Alder reaction of 2,3-
bismethylenenorbornane with norbornene followed by DDQ-oxidation of the
1,Cbenzoquinone). major adduct endo,ex0-2~)(DDQ-2,3-dichloro-5,6-dicyanoThe reaction of 3 with dicyanoacetylene (DCA) leads to the (1 :1) Diels-Alder adduct 4 and to the unexpected dark blue (2:l) adduct 5. The course of the reaction can be controlled by the concentration of DCA and by pressure to some extend. At atmospheric pressure (ca. 1 bar) the (1:l) adduct 4 is the major product after 70 % conversion of 3 whereas at high pressure (9 kbar) the (2:l) adduct 5 dominates even at 36 YOconversion of 3. The photolysis of 4 (separated from 5 by liquid chromatography, silicagel, hexane/ethylacetate 3: 1) leads to the desired COT derivative 6.
Chemistry: Posters
103
Scheme 1 :
endo,exo-2 1 bar, 160 "C, 22 h, yield 50 %
88
12
7.9 kbar, 75 "C, 6 h,conversion :30 %
96
4
DDQ, 80 "C t
endo,exo-2
o
l
u
e
n
-
e
m NC-CN
yield > 99 %
U
3
4
endo,endo-2
+
5
4
1 bar, 127 "C, 17 h, conversion : 70 %
49 %
11 %
9 kbar, 83 "C, 17 h,conversion : 36 %
8%
28 %
- p.$4
hv Pyrex ( 300 nm )
a) acetone or b)hexane \ acetone (10 : 1)
H
CN 6
cv
dihedral angle
exp:
calc.
54.2" 1.5" 59.5" 10.3" 47.5"
52.4" 3.9" 59.6" 13.9" 38.6"
e,3,4,5> <3,4,5,6> <4,5,6,7> <5,6,7, 8>
1
dihedral angle
exp.
cdc
<6,7,8, 1> <7, 8, 1,2> <8, 1,2,3> <16,6,7, 19> <9, 1,2, 12>
5.9" 43.9" 5.4" 16.7" 0.9O
0.9" 45.0" 5.9" 13.5" 0.2"
High Pressure Food Science, Bioscience and Chemistry
104
The crystal structure analysis shows, however, that the eight-membered ring in
6 is not planar contrary to the predictions. Accordingly, the distortion of the bismethylenenorbornane unit (dihedral angle <16,6,7,19>) by 16.7' is sufficient to avoid the planar COT structure. The structural parameters obtained by the crystal structure analysis agree well with a semiempirical PM3 calculation. A comparison between experimental and calculated dihedral angeles is included in
Scheme 1. The dihedral angles, responsible for the boat conformation of the eight-membered ring, do
not deviate very much from the corresponding angle
of the parent cyclooctatetraene (<1,2,3,4>
= 56 ").
Scheme 2 : Crystal structure analysis of 5
5
5
4
7 45 %
1 bar, 127 "C, benzene, 17 h, conversion: 58 % 12 kbar, 20 "C,toluene, 90 h, conversion: 51 YO
35 Yo
9 Yo
Chemistry: Posters
105
The structure of the dark blue (2:l)adduct could be elucidated by crystal structure analysis to be the cis-9,lO-dihydronaphthalenederivative 5 and the mechanism of its formation by the use of high pressure. The reaction of 4 with DCA at 1 bar and 127 "C leading only to 5 indicates the (1:l)adduct 4 to be an intermediate in the reaction 3 + DCA
5. At 12 kbar 4 reacts with DCA
readily at room temperature producing the colorless homo-Diels-Alder adduct 7 which undergoes a rearrangement (retro-Diels-Alder reaction) to the blue dihydro-naphthalenederivative 5 also at room temperature (half-life time of 7 at
20 "C, ca. 28 d). 5 undergoes a degenerate rearrangement, which is already rapid at room temperature on the NMR time scale leading to a pair-wise equilibration of the signals assigned to the four chemically nonequivalent norbornene or norbornane bridgehead hydrogen atoms. From the temperature dependent line-shape of these signals the activation parameters can be determined. The degenerate rearrangement 5 e 5' is formally a [1,5] vinyl shift6),which may occcur in a stepwise process via the diradical or dipolar intermediate 8. Analogous non-degenerate rearrangements are observed by
L. A. Paquette eta^^) and G. Maier et a1.') in the thermolysis of cis-9,lOdihydronaphthalene derivatives (cis-9,10-dimethy1-9,1O-dihydronaphthalene: M = 25.4
f
0.5 kcal-mol-', A S
=
-2.1
f
1.6 cal.mol-'-K', 61.3"C;
cis-9,1O-dicarbomethoxy-9,10dihydronaphtha1ene:AF= 25.7 f 0.4 kcal.mol-*,
A S = -6.3 f 1.2 cal-mol-**K-', 81.2 "C,and 1,2,3,4,5,6,7,8-octamethyl-cis9,1O-dimethoxymethyl-9,10-dihydro-naphthalene: AG'
fi:
25 kcal-mor', 20 "C).
The enthalpies of activation of the non-degenerate rearrangements are substantially higher by about 19 kcal-mor' than that found for the degenerate rearrangement in 5. The observed solvent dependence of the activation parameters (Figure l(c)) is not conclusive whether the reaction 5 * 5' proceeds via a diradical or a dipolar intermediate.
106
High Pressure Food Science, Bioscience and Chemistry
Figure 1 :
b) 3.3 - 2.1 ppm
CDzC12 Toluene CD30D
T= 233 K. k = 800
T=203 K, k = 75
T= 208 K, k = 100
T= 193 K, k = 20
AH', kcal.mo1-l ASf, cal.mol-'.K-' 7.7 f 0.5 -12 f 2 7.3 f 0.5 -14 f 2 12.0 f 0.4 4+2
AG', kcal.mor' 10.2 k 0.7 10.3 f 0.6 11.1 f 0.6
Figure 1 : a) Proposed mechanism of the degenerate rearrangement. b) Temperaturedependent 'H-NMR (300 MHz, CD2C12) of the bridgehead hydrogen atoms of 5. The experimental spectra and the spectra calculated by line shape analysis with the rate constants k [s-'1 at various temperatures T [K] are superimposed. c) The activation parameters derived from the temperature-dependent 'H-NMR spectra of 5 in different solvents at -60 "C
Ab initio calculations by means of DFT Becke 3LYP/6-31G* yield a transition state and an intermediate of type 8 in the reaction 5 e 5' to be higher in energy
by 12.3 and 9.5 kcal.mol-' respectively than the reactant 5 in good agreement
107
Chemistry: Posters
with the experimental findings. According to analogous calculations for the corresponding unsubstituted system the four nitrile groups lower the activation barrier by 8.5 kcal*mol-' (from 20.8 to 12.3 kcal-mol-') and stabilize the intermediate by 9.0 kcal-moil (&om 18.5 to 9.5 kcal-mof'), so that the effect of two fused norbornane units on the activation barrier can be estimated to be
about 5 kcal-mol-'. Table 1 : Solvent dependence of the UV-VIS spectrum of 5 MeOH hmax (h3 E) 267 (3.83) 281 (3.86) 305s (3.71) 339 (3.60) 399 (3.52) 545 (2.66)
EtOH &ax
(43 d
CH3CN ( k E) 263 (4.02)
&ax
280 (3.90) 341 (3.60) 384 (3.57) 399 (3.58) 547 (2.59)
307 (4.09) 337 (3.72) 368s (3.42) 399 (2.74) 539 (2.97)
CH2C12 &ax
(lg E)
CaHa &ax
(lg E)
280 (3.99)
280 (3.96)
340 (3.67)
330 (3.66)
395 (1.75) 575 (2.77)
573 (2.76)
The deep blue color of 5 is surprising, because the isolated two 1,3cyclohexadiene chromophores should absorb only ultraviolet light as they do in the case of the other 9,lO-dihydronaphthalene derivatives"*). We assume that the color of 5 comes from a charge-transfer (CT) band resulting from the interaction of the relatively electron-rich alkylsubstituted cyclohexadiene moiety with the electron-poor tetracyano-substituted cyclohexadiene moiety. The observed negative solvatochromism of the CT-band (Table l), however, indicates, that the ground state is here more polar than the excited state which means that the CT-band may be the result of the photochemical excitation of a polar intermediate of type 8. Further investigations are needed for the explanation of this intriguing problem.
108
High Pressure Food Science, Bioscience and Chemistry
On thermolysis at 80 OC in a polar solvent such as acetonitrile (half-life time : 1 10 min) or methanol (half-life time : 120 min) 5 reacts irreversibly under HCN elimination leading to the azulene derivative 11'' in quantitative yield, whereas in the less polar solvent toluene 5 is stable up to 150 "C.In order to explain this remarkable reaction we suggest a reversible disrotatory electrocyclic ringopening leading to the all-cis-cyclodecapentaene 9 followed by a disrotatory orbital symmetry-allowed electrocyclization forming the dipolar intermediate 10 in the rate-determining step (which explains the solvent dependence of the overall reaction). Finally HCN is eliminated from 10, probably in a stepwise process, to produce the observed azulene 11.
Acknowledgement : We are grateful to the Deutsche Forschungsgemeinschaft,
Ministerium fiir Wissenschafl und Forschung des Landes Nordrhein-Westfalen and Fonds der Chemischen Industrie for financial support.
References and Notes
1) Ermer, F.-G. Kliirner, M. Wette, J Am. Chem. SOC.1986,208,4908-4911
2)Review: G. I. Fray, R. G. Saxton The chemistry of cylo-octatetraene and its derivatives, 1978, Cambridge University Press
Chemistry: Posters
109
3)P. G. Wenthold, D. A. Hrovat, W. T. Borden, W. C. Lineberger, Science 1996,272, 1456-1459. 4)K. Saito and T. Mukai, Bull. Chem. SOC.Jpn., 1975,48,2334. 5) It is interesting to note that the corresponding Diels-Alder reaction between
5,6-bismethylene-2-norbornene
and
norbornadiene
proceeds
nonstereoselectively: J. Benkhoff, R Boese, F.-G. Kliirner, Liebigs
Ann.,/Recueill997,501-516. 6)A formal [1,5] vinyl shift is also observed in the thermolysis of spiro[4.4]nonatetraene by M. F. Semmelhack, H. N. Weller, J. S. Foos, J. Am. Chem. SOC.1977,99,292-294; M. F. Semmelhack, H. N. Weller, J. Clardy, J. Org. Chem. 1978,43,3791-3792. 7)A. Paquette, M. J. Carmody, J. Am. Chem. SOC.1975, 97, 5841-5850; Review: W. Alder, W.Grimme, Tetrahedron, 1981,37, 1809-1812. 8) G. Maier, H. Wiegand, S. Baum, R.Wullner, W. Mayer, R. Boese, Chem. Ber. 1989,122,767-779. 9) The spectral properties are consistent with those of 1,2,3-tricyanoazulene : S. Kuroda, M. Funamizu, Y. Kitahara, Tetrahedron Lett., 1975,1973-1974.
THE EFFECT OF PRESURE ON RETRO-DIELS-ALDER REACTIONS F.-G. Klhe;), V. Breitkopf
Institut fiir Organische Chemie, Universitat GH Essen, Universitatsstr. 5 , D-45 117 Essen, Germany Telephone: +49(0)201-183 3081, Fax: +49(0)201-183 3082, E-mail: [email protected]
Many Diels-Alder reactions show a powerful pressure-induced acceleration which is often turned to good synthetic purposes'). The activation volumes AV" resulting form the pressure-dependence of the rate constants are usually highly negative (AV" M -25 to -45 cm3.mol-') sometimes even more negative than the corresponding reaction volume A V so that the A V"/AV ratio is close to or even larger than unity (0= AV/AV> 1). Within the scope of the transition state theory, activation volumes can be considered to be a measure of the relative partial molar volume of the transition states (AV"= AV"(TS) - CV(reactants)). Accordingly, the molar volumes of transition states of many Diels-Alder reactions are approximately equal to or even smaller than those of the corresponding cycloadducts. These surprising results could be confirmed by two independent studies. In the Diels-Alder reactions of furan with acrylonitrile2) and in that of N-benzoyl-pyrrole with N-phenylmaleic imide3) the ratio AV#/AV was found to be larger than unity
(0= 1.06 and 1.37, respectively). The cycloadducts isolated from both reactions
Chemistry: Posters
111
undergo smooth retro-Diels-Alder reactions showing negative volumes of activation in agreement with the value 0 > 1 determined for the forward reactions. The result, that the transition state volume in several Diels-Alder reactions is smaller than the product volume, is surprising and not well understood. This finding seems to be contradictory to the generally accepted relation between molecular structure and its volume. In the transition state the new bonds between diene and dienophile are only partially formed (according to quantummechanical calculations4)the distances of the newly forming
Q
bonds in the
transition state are in the range between 2.1 and 2.3 A). The intrinsic molar volumes, the so called van der Waals volumes, VW, calculated for the pericyclic transition structures of Diels-Alder reactions by the use of the structural parameters obtained from quantum-mechanical calculations4)and the van der Waals radii of the different types of atoms derived from X-ray data are generally larger than those calculated analogously for the corresponding cycloadduct~~). Grieger and Eckert6' considerd two explanations of the ratio 0 > 1 in the Diels-Alder reaction of isoprene as diene with maleic anhydride as dienophile: a larger dipole moment of the transition state or secondary orbital interactions7)from which the secondary orbital interactions can only occur in endo-Diels-Alder reactions. The finding, that the difference between the activation volumes of many endo- and exo-Diels-Alder reactions is small (AV# > 1-2 cm3-mol-'), seems to rule out that secondary orbital interactions are important and induce a larger contraction of the volume of the endo-transition state. In order to gain further insight into this remarkable effect, we investigated the volume profiles of retro-Diels-Alder reactions of several dihydrobarrelene and
High Pressure Food Science, Bioscience and Chemistry
112
7-isopropylidenenorbornene systems starting from the nonpolar parent hydrocarbon l a to systems substituted by polar groups in different positions. Scheme 1.
rpcl AV.') AVT )b'
VT'~) VW') qc)
118.7 70.7 0.596
121.8 74.0 0.608
*
99.8 54.1 0.542
66.5 25.6 0.385
0' 1
+
lb
VT'~) VW') qc)
123.3 81.3 0.659
VT") Vw')
168.7 121.7
)'1
2b'
125.8 84.8 0.674
167.6 125.1 0.746
0.721
105.6
+2.5
44.8
106.3
-1.1
36.2
3b
101.5 65.0 0.640
130.1 82.7 0.636
66.6 25.6 0.384
74.0 48.4
0.654
CH,CN
w10 endo-4
VT'~) Vwa) nc) )'
169.0 121.7 0.720
cndps
*
167.9 125.1 0.745
6
130.1 82.7 0.636
74.0 48.4 0.654
[cm3*mol-'],b, extrapolated from the temperature dependence of V between TO= 20 "C and T, = 70 "C by the use of the El'yanov equation VT= VO*[ 1 + KO*(TTO)];VO[cm3.mol-'] / ~ 0 . 1 [K']: 0 ~ 110.6 / 8.86 (la); 90.0 / 13.0 (3a); 113.4 / 10.2 (lb); 100.3 / 1.37 (3b); 164.2 / 3.19 (em-4); 164.3 / 3.31 (endo-4); 122.9 / 6.83 (6); 69.3 / 7.88 (7), )' q = V W / ~ T
113
Chemistry: Posters
Figure 1.
'= 0.94
The retro-Diels-Alder reaction of dihydrobmelene l a leading to benzene 3a and ethene is decelerated by pressure and its activation volume was determined from the pressure dependence of the rate constants at 105.3 "C to be slightly positive (Scheme I ) . The partial molar volumes of l a and benzene and, hence, the volume of reaction at 105.3 "C,the temperature of reaction, were obtained from temperature- and concentration-dependent density measurements of l a and benzene, respectively, dissolved in n-hexane. The partial molar volume of gaseous ethene was extrapolated by means of Exner increments') and by the use
of temperature coefficient G, determined for benzene (Scheme 1). From these data the complete volume profile of this reaction shown in Figure 1 can be
114
High Pressure Food Science, Bioscience and Chemistry
constructed. The ratio of activation volume to reaction volume for the experimentally not accessible Diels-Alder reaction of benzene with ethene is with 0 = AV#/AV= 0.94 only a little smaller than unity as found for many other Diels-Alder reactions, too. The packing coefficient q of the pericyclic transition state 2' calculated from the ratio of its van der Waals volume to its partial molar volume, q = V,/V= 0.608, turned out to be approximately equal to or even a little larger than that of the cycloadduct la, q = 0.596. Similar results were obtained for the reaction of 2-cyanodihydrobarrelene l b included in Scheme 1. Again, the packing coefficient of the pericyclic transition state was found to be slightly larger than that of the corresponding cycloadduct
lb. The retro-Diels-Alder reactions of endo-4 and exo-4 leading both to naphthalene 6 and maleic anhydride 7 (MA), however, are slightly accelerated by pressure showing negative volumes of activation (Scheme 1). Again, the packing coefficients are calculated to be larger for the pericyclic transition state than for the corresponding cycloadducts. The effect of pressure on the Diels-Alder reaction of naphthalene 6 with MA 7 leading to a mixture of endo-4 and exo-4 is worth mentioning. At 100 "C this reaction, which is reported by Plieninger et al.9) to be the best method for the synthesis of endo-4 and exo-4, occurs in reasonable yields only at pressures higher than 7 kbar in relatively concentrated solutions. In dilute solutions the reaction stops at a relatively low pressure-dependent conversion of naphthalene (Table 1). The ratio [naphthalene]:[endo-4]:[exo-4] seems to represent the equilibrium mixture at different pressures. Obviously, the solubility of endo-4 and exo-4 is pressure-dependent, too, so that these cycloadducts precipitate from concentrated solutions at high pressure. The precipitation causes a shift of the equilibrium toward the cycloadducts. Under optimized conditions of
Chemistry: Posters
115
reaction ([6], = [7]0= 1.1 M, in CHCL at 100 "C, 12 kbar, 48 h) endo-4 and ex04 can be prepared in 94% yield. Table 1. Pressure dependence of the ratio [6]:[exo-4]:[endo-4] in the Diels-Alder reaction of 6 with 7, [6Io = [7]0= 50 mM in CDCl3 at T = 100 OC,time of reaction t = 22 h
6
7
~[kbar]
6[%]
7.0 8.0 9.0 10.0 11.0 12.0
96.80 94.38 92.77 88.74 75.07 66.65
a)
ex04 ex04 [%]
2.06 3.68 4.76 7.46 16.46 17.42
endo-4 endo-4 [%]
1.14 1.94 2.47 3.79 8.47 15.93
exo-4:endo-4
64.3:35.7 65.4:34.6 65.8:34.2 66.3:33.7 66.0:34.0 52.2:47.8a)
at 12 kbar the cycloadducts exo-4 and endo-4 are already partially precipitated. Therefore, a quantitative analysis of equilibrium mixture was no longer feasible under these conditions.
The effect of pressure on the retro-Diels-Alder reactions of endo-8 and exo-8 leading both to dimethylfblvene 10 and N-phenylmaleic imide 11 are clear-cut examples which show that secondary orbital interactions are not important for the finding of negative volumes of activation in retro-Diels-Alder reactions. The reaction of endo-8 is decelerated by pressure showing AV > 0 whereas the reaction of exo-8 is accelerated by pressure showing AV < 0 (Scheme 2). In the case of endo-9* the packing coeficient is calculated to be equal to that of endo-8 and in the case of exo-9' larger than that of exo-8.
High Pressure Food Science, Bioscience and Chemistry
116
Scheme 2.
VT'~) Vw')
242.7 165.4
238.1 162.5
uo-8
.
uo-9'
123.1 75.2
10
136.5 96.3
11
The following conclusions can be drawn from these results. The molecular packing of the entire ensemble consisting of solute and solvent molecules and its reorganization during the course of reaction and not the changes of the intrinsic molecular volumes are most important for the magnitude of activation and reaction volumes. The packing coefficients of the pericyclic transition states resemble those of the corresponding cycloadducts as already assumed for the explanation of the different activation volumes of pericyclic and stepwise cycloadditions involving diradical intermediates'*). In the retro-Diels-Alder reactions showing AV'CO the packing coefficients of transition states are calculated to be significantly larger than those of the corresponding cycloadducts, that has been particularly found in the relatively polar systems
117
Chemistry: Posters
bearing cyclic anhydride or h i d e functions. In these cases, the size of the activation volumes obviously depends not only on the effective packing around the transition states, probably caused by the restriction of vibrations and rotations, but also on the transition-state polarization enhanced by the polar groups. Blake and Jorgensen have assumed similar effects to explain the accelaration of the Diels-Alder reaction of 1,3-cyclopentadiene with methylvinylketone in water"). Figure 2. The pressure dependencies of the rate constants of the retro-Diels-Alder reaction exo-8
+ 10 + 11 at 87.0 "Cin methanol
-10.0 -10.1 0
500 l 1000 1500 2000 O P [bar1 2500 3OOO. 3500 4000 4500 O
~
The retro-Diels-Alder reaction exo-8 + 10 + 11 exhibits an unusual pressure dependence (Figure2) comparable to that observed by Asano et al. for the syn + anti isomerization in azobenzenes and N-benzylideneanilines12). The
reaction exo-8 + 10 + 11 is accelarated by raising the pressure fiom 1 to ca. 1200 bar (the activation volume given in Scheme 2 is calculated fiom these values), whereas the rate of reaction is decreased at pressures higher than 1500 bar up to 3000 bar and, then, remains almost constant between 3000 and
High Pressure Food Science, Bioscience and Chemistry
118
4000 bar. A similar but less pronounced pressure dependence is also observed
for the reaction exo-4 -+ 6 + 7.According to preliminary thermolyses of exo-8 in different solvents at different pressures there is no relationship between the
pressure dependence of the rate constant and the solvent viscosity contrary to the results obtained by Asano et al.'*) and Troe et al.13) for their systems. Further investigations are required for these remarkable effects to be elucidated. Acknowledgement: We are grateful to the Deutsche Forschungsgemeinschaft,
Ministerium fiir Wissenschaft und Forschung des Landes Nordrhein-Westfalen, and the Fonds der Chemischen Industrie for financial support.
References and Notes 1) Recent reviews: F.-G. Klimer, M. K. Diedrich, A. E. Wigger, Chapter 3 in:
Chemistry Under Extreme or Non-Classical Conditions (ed. R. van Eldik,
C. D. Hubbard), SpektrudWiley, New York 1996;M. Ciobanu,
K. Matsumoto, Liebigs Ann./Recueill997,625-635. 2) G. Jenner, M.Papadopoulos, J. Rimmelin, J. Org. Chem. 1983,48,748-749. 3) A.V.George, N. S. Isaacs, J. Chem. SOC., Perkin Trans. II1985,1845-1847.
review: K. N. 4)Y.Li, K. N. Houk, J. Am. Chem. SOC.1993,115,7478-7485; Houk, Y. Li, J. D. Evanseck, Angew. Chem. 1992,104,711-739.
5) Y.Yoshimura, J. Osugi, M. Nakahara, Bull. Chem. SOC.Jpn. 1983,56,680683;Y. Yoshimura, J. Osugi, M. Nakahara, J. Am. Chem. SOC.1983,105, 5414-5418;R.A. Firestone, G. M. Smith, Chem. Ber. 1989,122,1089-1094; F.-G. Klimer, Chemie in unserer Zeit, 1989,23,53-63. R. A. 6)R.A. Grieger, C. A. Eckert, J. Am. Chem. SOC.1970,92,2918-2919; Grieger, C. A. Eckert, J. Chem. Soc., Faraday Trans. 1970,66,2579-2584.
Chemistry: Posters
119
7) According to recent quantum mechanical calculations, the importance of
secondary orbital interactions, which have also been frequently used to explain the endo diastereoselectivity of Diels-Alder reactions, seems to be questionable and to be reserved for special cases like the addition of cyclopropene to various dienes. T. Karcher, W. Sicking, J. Sauer, R. Sustmann, Tetrahedron Lett. 1992,33,8027-8030; R. Sustmann, W. Sicking, Tetrahedron, 1992,48, 10293-10300; Y. Apeloig, E. Matzner, J. Am. Chem. SOC.1995, I1 7,5375-5376. 8) 0. Exner: "Empirical Calculations of Molar Volumes",Chapter 2, in Organic High Pressure Chemisby (ed. W. J. le Noble), Elsevier, Amsterdam 1988. 9) W. H. Jones, D. Mangold, H. Plieninger, Tetrahedron 1962,18,267-272. 10) F.-G. K l h e r , B. Krawczyk, V. Ruster, U. K. Deiters, J. Am. Chem. SOC. 1994,116,7646-7657.
11) J. F. Blake, W. L. Jorgensen, J. Am. Chem. SOC.1991,123,7430-7435. 12) K. Cosstick, T. Asano, N. Ohno, High Pressure Res. 1992, If, 37-54;
T. Asano, H. Furuta, H. Sumi,J. Am. Chem. SOC.1994,116,5545-5550; T. Asano, K. Matsuo, H. Sumi, Bull. Chem. SOC.Jpn. 1997, 70,239-244. 13) J. Schroeder, D. Schwarzer, J. Troe, F. Voa, J. Chem. Phys. 1990,93,23932404; L. Nikowa, D. Schwarzer, J. Troe, J. Schroeder, J. Chem. Phys. 1992, 97,4827-4835.
The Effect of Pressure on the Formation and Decay of a Furanone during the Maillard Reaction Mark J.Bristow and Neil S.Isaacs Department of Chemistry. University of Reading Reading, RG6 6AD, U.K.
Among the major secondary products from the Maillard reaction between lysine and xylose, 5-methyl-4-hydroxy-3(2H)-furanone(furanone 1) has been isolated and characterised. It is a flavour constituent with caramel-like odour and is found to form to a maximum concentration and then to decay as the reaction is prolonged. The route to this compound is shown in the Scheme.
NHR
//
OH OH
2 HO
OH OH
3
& H O H' H H ~ Ho % H H
O
0X
-
""I:" 0
1
The decomposition of its precurser, the Amadori rearrangement product ,2, has been shown to occur with an overall positive volume of activation. This study details the effect of pressure on the formation and decay of this product, summarised in Fig. 1 Reactions between lysine and xylose were carried out at 100" in water, analysing samples taken at time intervals by hplc.
121
Chemistry: Posters
80 80 70
'
g!
-0-4kbar -4- 2kbar
-++
40
1kbar
30 20 10 0
0
60
120
180
240
300
Timelmlnutes
Fig. 1 ;formation and decay curves of furanone 1 as a h c t i o n of time and pressure It is seen that the maximum yield of furanone 1 is obtained at 1 bar and, at 1 kbar, the compound is hardly discemable. The reason is that pressure is evidently accelerating its decomposition . This exemplifies the possibility that high pressures at temperatures at which Maillard chemistry can occur, can lead to changes in composition of the flavour and odour constituents.
Acknowledgement; This work was funded by a LINK project promoted by an industrial consortium and the Ministry of Agriculture,Fisheriesand Food.
The Hydrolysis of Lipids and Phospholipids at Atmospheric and at High Pressures. Neil Slsaacs and Neil Thornton-Allen Department of Chemistry, University of Reading, Reading RG6 6AD, U.K. It is of importance to investigate possible chemical changes which might occur to food materials and their constituents when being processed by elevated pressures. One possible reaction would be the hydrolysis of glyceride esters to the constituent fatty acids which might be accompanied by undesirable flavour changes. It is known that carboxylic esters may be hydrolysed in neutral water under the influence of high pressure with a volume of activation of around -20 cm3mol-’ .
07
0-
OH
n
R’OH
A series of triglycerides of varying water solubility was studied at 80’ and at
pressures up to 5 kbar, the course of the reactions being monitored by sampling and glc analysis. Triacetin is freely water soluble and, indeed, the rate of hydrolysis of each ester group at pH 7 is similar and greatly accelerated by pressure, Figs 1-3 rates increasing by a factor of >4 at 100 Mpa. Triacetin, however, is not a food constituent and the experiments were repeated for tributyrin, slightly water soluble, for tripalmitin, a typical vegetable oil. and totally water insoluble and also for lecithin, a phospholipid which has some solubility or, at least, considerable emulsifying properties.
Chemistry: Posters
123
60 50
TRI 40
DI
MONO
30
GLY 20 10
0 0
12
24
36
48
60
72
84
96
108
120
Time h 700
600
.s
500
L
E
c
fi i!
--m d ._
400 300
200 100
0
0
5
10
15
20
25
12
16
20
Time I h
0
4
8 Time I h
Figs.1,2,3; Composition of the mixed products during hydrolysis of triacetin at 80' ; pressures 1bar, 500 bar and 1 kbar respectively
High Pressure Food Science, Bioscience and Chemistry
124
Tributyrin was found to hydrolyse more slowly than did tr,iacetin but the pressure acceleration was similar, Figs 4 3 . However, the insoluble lipids, tripalmitin and lecithin appeared to undergo no hydrolysis at all even at pressures up to 1000 MPa and 80'.
It may be concluded that, under the
conditions to be expected for food processing - 600 Mpa, 60' and times less than 30 min, no hydrolysis of fats and oils would be expected to occur.
400
I
300 0 .e
E
L
?i .-s m
200
-aJ c
[L
100
0 0
10
20
30
40
50
60
70
80
90
100 110 120 130 140
Time /h 400 360 320
280 240 200
160 120 80
40
I
0
0
10
20
30
40
50
60
70
80
90
100
Timelh
Figs.4,5; mixed hydrolysis products of tributyrin in water at pH7,80° at 1 bar and 5 kbar, respectively Acknowledgement; This work was funded by a LINK project promoted by an industrial consortium and the Ministry of Agriculture,Fisheries and Food.
THE EFFECT OF HIGH PRESSURE ON THE DEGRADATION OF ISOTHIOCYANATES C. Grupekb,H. Ludwig and B. TauscheP' "Institute of Chemistry and Biology, Federal Research Centre for Nutrition, Engesserstr. 20, D-76 13 1 Karlsruhe; bInstituteof Pharmaceutical Technology, University of Heidelberg, INF 346, D-69 120 Heidelberg
Introduction Isothiocyanates are important for human health and for the sensory quality (pungency) of some vegetable produce. For example, isothiocyanates of broccoli had been shown to have cancer protective properties'.'). On the other hand, high concentrations of both glucosinolates and isothiocyanates may have negative physiological properties3) Isothiocyanates are formed fiom glucosinolates in an enzymatic reaction corresponding to the Lossen degradation of hydroxamic acids. They may enter into further reactions, e.g. solvolysis (figure 1). The isothiocyanate function is known to react with nucleophiles as alcohols and water, forming thiocarbamates and thiocarbamic acids. These reaction products can enter into further hydrolysis, forming amines. We investigated whether high pressure favours the degradation of allyl- and benzyl isothioyanate in aqueous ethanolic solution,
126
High Pressure Food Science, Bioscience and Chemistry
Fig. 1: Formation of isothiocyhates from glucosinolates and subsequent solvolysis. Materials and Methods : The isothiocyanates (Aldnch, D-89555 Steinheim, Germany) were applied as aqueous ethanolic solutions at a concentration of 20 mg per ml. In the case of ally1 isothiocyanate,the water-ethanol ratio was 1 to 1 by volume, in the case of the less soluble benzyl isothiocyanate the ratio was 1 to 2. The reaction was carried out in PTFE tubes (wall diameter 1 mm, interior diameter 8 mm, length 90 mm) sealed with PTFE stoppers. 2.5 ml of samples were treated at temperatures of 25,40,60 and 70 "C at atmospheric pressure and high pressure (600 Mpa). Reaction times were 30, 60, 90 and 120 minutes. The reaction mixtures were added to 20 ml of water, extracted twice with n-hexane (10 ml); the combined extracts were washed twice with water (10 ml), dried over anhydrous sodium sulfate and diluted to 50 ml with n-hexane.
Chemistry: Posters
127
Reference samples were prepared by the same extraction procedure. Concentrations of samples were determined by HPLC (Knauer Nucleosil 50-5, 100x4 mm, using n-hexane as eluent at 1 ml per min, detection by absorption at 245 nm).
Results and Discussion : As figures 2 to 5 show, application of high hydrostatic pressure (600 MPa)
increases the rate of degradation of both allyl- and benzyl isothiocyanate up to 4 times, compared with treatment under atmospheric pressure. In the case of allyl isohocyanate, a yellowish dicolouring and a significant change
in odour of the reaction mixture after treatment under high pressure conditions was also observed, whereas the changes under atmospheric pressure were much smaller. In the case of benzyl isothiocyanate, a small quantity of a white solid reaction product was isolated fiom the sample treated at 600 MPa, 70" C for 120 min. This reaction product has been shown to be 173-dibenzylthiourea,this suggests
the intermediate formation of benzylamine in the reaction pathway. The results show that under high pressure isothiocyanates are degraded by reaction with nucleophiles much faster than under atmospheric pressure. This may be of some importance for the high pressure treatment of food containing isothiocyanates, as some physiological properties and flavour may be affected.
High Pressure Food Science, Eioscience und Chemistry
128
Fig. 2: Degradation of allyl isothiocyanate in aqueous ethanol under atmospheric pressure (0.1 MPa).
0
30
60
120
timc (rain)
Fig. 3: Degradation of allyl isothiocyanate in aqueous ethanol under high pressure (600 m a ) .
129
Chemistry: Posters
-60 --
-- 60
--
40
m --
-- m
0 ,
, o
40
Fig. 4: Degradation of benzyl isothiocyanate in aqueous ethanol under atmospheric pressure (0.1 MPa).
-- 80
E3 +75’
0
80
c
120
Ume (rnln)
Fig. 5: Degradation of benzyl isothiocyanate in aqueous ethanol under high pressure (600 m a ) .
References : 1 .) Y. Zhang, T.W.Kensler, C-G. Cho, G.H.Posner, P. Talalay Proc. Nutl. Acad. Sci. U.S.A. 1992,89,2399-2403. 2.) G.H. Posner, C-G. Cho, J.V.Green, Y. Zhang, P. Talalay J. Med. Chem. 1994,37, 170-176. 3) H.L. Tookey, C.H.VanEtten, M.E.Daxenbickler in “ToxicConstuentsof Plant Foodstufs”, 2nd ed.; I.E. Liener, Ed.;Academic: New York, 1980, 103-142
Plenary Lecture
FOOD CHEMISTRY UNDER HIGH HYDROSTATIC PRESSURE P. Butz and B. Tauscher* Institute of Chemistry and Biology, Federal Research Centre for Nutrition, Engesserstr. 20, D-76131 Karlsruhe, Tel.: ++49 721 6625 116, Fax.: ++49 721 6625 167, e-mail: [email protected]
Abstract
Foods pasteurizedby hydrostatic high pressure have already been marketed in Japan. There is great interest in this method also in Europe and USA. Temperature and pressure are the essential parameters influencing the state of substances including foods. While the influence of temperature on food has been extensively investigated, effects of pressure, also in combination with temperature, are attracting increasing scientific attention now. Processes and reactions in food governed by Le Chatelier’s principle are of special interest; they include chemical reactions of both low- and macromolecular compounds. Examples of pressure affected reactions are presented.
Introduction Pasteurization of food by ultrahigh hydrostatic pressure has attracted the attention
High Pressure Food Science, Bioscience and Chemistry
134
of many disciplines. Chemical reactions of low-molecular and oligomeric compounds in food under hgh pressure have been little investigated. Generally any process and any reaction in food are of interest whch form according to the principle of Le Chatelier. Under equilibrium conditions, a process associated with
a decrease in volume is favoured by pressure, and vice versa. Pressure influences rate and equilibrium of reactions even in food (Tauscher, 1995). For any reaction in solution (Matsumoto and Acheson, 1991) between reaction partners A und B the reaction volume AV and the activation volume AVf can be
0 describes the partial volumes of the reactants or of the products. For the location of the equilibrium of a chemical reaction in solution under pressure, the reaction volume AV is decisive; it is described by: R =general gas constant
AV
=
-RT
dlnK dP
(-)T
T = absolute temperature K = equilibrium constant P =hydrostatic pressure
Pressure influences not only the location of the equilibrium of a chemical reaction in solution, but also its reaction rate. To the activation volume the following equation applies:
135
Food Science: Presentations
dlnk A V t = -RT (-)T
dP
where k =rate constant.
Examples So the question arises whether at pressure >500 MPa,which is employed for food sterilization,the chemical reactions supposed to occur are desirable or not. We investigated the autoxidation of alpha-linolenic acid (Kowalsh et al.) under pressure of up to 600 MPa. Three groups of peaks were separated by HPLC and the lunetics of their formation described. Isomeric hydroperoxyepidioxides, hydroxides and hydroperoxideswere determined by spectroscopy. Increasing pressure reduces the induction phase strongly and accelerates formation of primary oxidation products; after a maximum, it decreases relatively fast again. Shorter induction phases are explained by degradation of hydroperoxides in the original product. Propagation under elevated pressure follows the same reaction mechanism as under normal pressure. With increasing viscosity and radical concentration, reactions are stopped by recombination of radcals in a confiied area, leading to decreasing yield with increasing pressure.
Vitamin K, a denophle with a naphthoquinone system, reacts, especially at higher temperature and pressure, with a &ene such as myrcene in a Diels-Alder reaction to form a six-member ring system with a double bond (Diels-Alder-adduct) (Ludwig
136
High Pressure Food Science, Bioscience and Chemistry
et al.). Isomeric compounds formed as products were separated by HPLC, their structure explored by spectroscopic methods. Vitamin K3 and myrcene in ethanol at 70°C and 650 M a after 6 hours showed 100% yield compared with3% of the blind. Vitamin K, in ethanol at 70°C and 650 MPa formed two isomeric products
of 25% yield each after 60 hours; the blind did not react under these conditions. At 40°C vitamin K, yields only traces of Diels-Alder products whtle at 70°C yield increased significantly. Accordingly, Diels-Alder-products between food components may be expected at htgher temperatures and hgh pressure, in the case of vitamin K, and myrcene as soon as after 15 minutes. It remains to be studied whether the food matrix has catalyzing or inlubiting effect on this kind of reaction.
Ubiquinone (also: coenzyme Q,*) is an important electron carrier in oxidative phosphorylation in the respiratory chain. There are homologous compounds with shorter side chains; in mammals, however, coenzyme Ql0is found most frequently. We studied the homologous coenzyme Qo(2,3-dimethoxy-5-methylbenzoquinone) whch is not a natural substance(Ludwigeta1.,1997).Reaction of coenzyme Qowith myrcene according to the Diels-Alder mechanism may produce four isomeric compounds each existing in enantiomeric pairs. Solutions of 2,3-dimethoxy-5methylbenzoquinone and myrcene in a molar ratio of 1 :3 in ethanol, were exposed to 650 Mpa and 70°C for different times. The reaction was complete after 16 hours. Gas chromatographc separation of the reaction products gave two close peaks
Food Science: Presentations
137
which, according to their mass spectra, were identibed as 2,3-dimethoxy-8a-methyl-
7-(4-methylpent-3-enyl)-5,5a,8,8a-tetrahydro-l,4-naphthoquinone and
2,3-
dimethoxy-8a-methyl-6-(4-methylpent-3-enyl)-5,5a,8,8a-tetrahydro1,4naphthoquinone.Theresults shown were confirmed by NMR, IR and UV studies.
Isothiocyanates are important for human health and for the sensory quality (pungency) of some vegetable produce. They are formed from glucosinolates in an enzymatic reaction corresponding to the Lossen degradation of hydroxamic acids. They may enter into further reactions, e.g. solvolysis to form thiocarbamates and thtocarbamic acids. These reaction products can enter into further hydrolysis forming amines. Application of high hydrostatic pressure (600 MPa) increases the rate of degradation of both allyl- and benzyl isothiocyanate up to 4 times at elevated temperatures, compared with treatment under atmospheric pressure (Grupe et a1., 1997). In the case of benzyl isothiocyanate, a small quantity of a white solid reaction product 1,3-dibenzylthiourea was isolated. This suggests the intermediate formation of benzylamine. The stability of vitamins under high pressure conditions has been another topic.As model systems solutions of ca. 1000 mgA ascorbic acid in 0.1 movl sodium acetate buffer, pH 3.5-4,plus 10 % sucrose were used (Taoulus et a1.,1997). Three variants were studied: complete exlusion of oxygen by evacuation and rinsing with nitrogen, oxygen saturation, and solutions exposed to air. The samples were filled into teflon
138
High Pressure Food Science, Bioscience and Chemistry
bottles (10 ml) and exposed to 0.1, 300, 450 and 600 MPa for 10, 20 and 40 min. at 40,60 and 75 "C and examined by the L-ascorbic acid colour test of Boehringer Mannheim (Germany) (No. 409 677) with extended incubation time. In all model solutions with normal exposure to air vitamin C losses were greater under pressure (maximally 45 % after 40 min treatment at 600 MPa and 75 "C). In oxygen free solutions loss of Vitamin C was slowed down as expected. With 10 % sucrose a protective effect against degradation of vitamin C due to hydrostatic pressure was observed. The kmetic behaviour of solutions of vitamin A alcohol and vitamin A acetate in ethanol was recorded at hgh pressure. Vitamin A alcohol (retinol) was studied at 600 Mpa and 25/40/60/75"C for 5/10/20/40 minutes (Ludwig et al., 1997). In one exemplary case the presence of oxygen was excluded by argon bubbling the solvent and solution, respectively, and pressure processed at 600 MPa at 75°C for 40 minutes. Analysis by HPLC showed almost no loss of retinol at ambient pressure and for all temperatures employed. However, application of pressure led to a dramatic decrease of the vitamin A content. At 25°C only 50% were recovered after 40 min whereas at 40"C, 60°C and 75°C less than 40% were found remaining. After 5 min 40°C treatment still gave a good percentage of remaining retinol (80%). Still, 60°C and 75°C experiments even at short time treatment gave degradation to about 45%. The absence of oxygen gave no sigtllficant difference to the data given above.
Food Science: Presentatiom
139
The influence of pressure on the formation of the cyclization product of the dlpeptide Aspartame as a model reaction was investigated in a grape juice matrix. Aspartame has been shown to be highly sensitive to pressure e.g. in milk (Taoukis et al., 1997). 1-2 mmovl solutions of L-aspartyl-L-phenylalanine methylester (= Aspartame, Nutrasweet) in 0.05 moM Tris/HCl buffer, pH 7, and in commercial grape juice were treated at 700 MPa for 3, 10 and 30 minutes. Aspartame was very unstable in Tris buffer, but nearly unaffected in grape juice under the same conditions. Degradation products were L-aspartyl-L-phenylalanineand cis-3,6-
dioxo-5-@henylmethyl)-2S-piperazine acetic acid (=diketopiperazine, DKP, the acceptable daily intake (ADI)value of DKP is one tenth of that of Aspartame) and aspartyl-phenylalanine. Pressure treatment of fruit juices of acid pH will not lead to sipficant formation of cyclization products from Aspartame like peptides.
As a result of tissue destruction by pressure (Butz et al., 1994) enzymes are released whch hydrolyse acyl lipids. By oxidation of the resulting free fatty acids via biochemical and chemical pathways volatile C, - C,o carbonyls and furans are formed. Some of these substances smell llke oil, fat, suet or oil paint even at extremely low concentrations (e.g. threshold for the suet-like odour of nonanal in water: 0.08 ppb). From the increase of the main product, hexanal, the extent of acyl lipid oxidation and the resulting possible formation of off-flavour producing compounds may be derived. The effect of pressure on the development of hexanal
in strawberry pulp as a model matrix was investigated (Butz and Tauscher, 1997).
140
High Pressure Food Science, Bioscience and Chemistry
Pulp from frozen strawberries, cultivar Mamolada, was treated for 30 minutes at
0.1 to 600 MPa at 25, 40 and 60 "C. Volatiles were extracted from the headspace of the samples by Solid Phase Mcro Extraction (SPME) and analyzed by GC, GCMS. Of 3 1 detected volatiles (17 identified) only hexanal changed significantly
in concentration due to pressure/temperature treatment. Whlle at 25 "C hexanal almost doubles at 600 MPa, the increase is 70% at 40 "C and only 10 % at 60 "C. Pressure treatment at elevated temperatures is recommended. On heating of food containing carbohydrates and sulphur containing amino acids e.g. like milk and milk products, volatile sulphur compounds may be formed via Strecker degradation (Interaction of an amino acid with a carbonyl compound to give an aldehyde or ketone containing one less carbon atom) and follow-up reactions. The sulphur containing amino acid methonine e.g. is the precursor for mehonal and dimethyl disulphide.Volatile sulphur compounds are often implicated in off-flavor development: because of very low thresholds small quantities can have deleterious effect on organoleptic properties.The influence of pressure 0.1-600 MPa and heat on the formation of dimethyl disulphide and mehonal from a mixture of glucose, methonine, phenylalanine, proline and leucine (Chan, F., and Reineccius, G.A., 1994) in Tris buffer, pH 8, was investigated.The development of methional
was strongly irhbited by pressures of 300 MPa and above. The development of dimethyl disulphide was not mfluenced by pressure. It can be expected from the results, that pressure will not lead to increased off-flavour production by volatile
Food Science: Presentations
141
sulphur compounds. High pressure is even protective in the case of methional formation (Butz et al., unpublished results). The green colour of plants is caused by chlorophyll. Chlorophylls are known to change into phaeophytins upon heatrng in low acid solution. This is accompanied by undesirable colour changes. Little is known on the influence of pressure on such reactions. The influence of pressure and temperature on chlorophyll a in alcoholic and aqueous solutions was investigated (May and Tauscher, 1997). The absorption spectra of chlorophyll a in ethanolic solution (95%) were not influenced when the solution was exposed to 70°C aver 4 hours. Both position and intensity of the main absorption bands did not change. The curve of the absorption maximum at 665 nm suggests a monomeric form of chlorophyll a in alcoholic solution. Neither did pressure of 600 MPa applied for 40 min at 70°C influence the absorption spectrum of chlorophyll a. The magnesium obviously was not displaced from the molecular set. Completely different results were obtained with chlorophyll a in an aqueous Tris-HC1 buffer solution where the absorption spectrum changed under the d u e n c e of temperature. The solutions turned nearly colourless. Absorption at 67 1 nmdecreased continuously while in the range above 700 nm it has been found to increase. This pronounced red shift is explained by a chlorophyll-water complex. These adducts may appear as even polymeric chlorophyll species in recurring units. Aqueous buffered chlorophyll a solutions responded differently also when exposed to pressure. No change in the absorption band at 671 nm was observed at 70°C and
142
High Pressure Food Science, Bioscience and Chemistry
600 MPa over 40 min. T h s means that hydrostatic pressure prevents water from
being added to the magnesium atom of chlorophyll. The formation of larger chlorophyll aggregates is prevented already at low pressure of 200 MPa and 70°C. Higher pressure favours the monomeric form of chlorophyll a even at elevated temperatures. Consequences of pressure treatment on chlorophyll in real food systems containing natural buffers remain to be studied.
References Tauscher, B. (1995). Pasteurization of food by hydrostatic hgh pressure: chemical aspects. Z Lebensm Unters Forsch 200, 3-13.
Matsumoto, K. & Acheson, R.M. (1991) (eds.). Organic Synthesis at hgh pressure. Wiley, New York, Chchester, Brisbane, Toronto, Singapore.
Kowalslu, E., Ludwig, H. and Tauscher B. (1996). Behaviour of organic compounds in food under hgh pressure: lipid peroxidation in: High Pressure Bioscience and Biotechnology, Hagash, R. and Balny, C. (eds). Progress in Biotechnology 13, 473-479.
Ludwig H., Marx, H. and Tauscher B. (1 995). Behaviour of organic compounds in food under hgh pressure: Diels-Alder reactions of food components in: proceedings
Food Science: Presentations
143
of the 1st main meeting. Process optimization and minimal processing of foods, EC
Copemicus programm, Porto, Portugal, Vol. 4, 31-37.
Ludwig, H., Kubel J. and Tauscher, B. Diels-Alder reactions of food-relevant compounds under high pressure: 2,3- Dimethoxy-5-Methyl-P-Benzquinoneand Myrcene, submitted for publication.
Grupe, C., Ludwig, H. and Tauscher B.. The effect of h g h pressure on the degradation of isothocyanates, submitted for publication.
Taoulas, P.S., Panagiotidq P., Stoforos, H.G., Butz, P., Fister, H. and Tauscher, B. (1 997). Kinetics of vitamin C degradation under high pressure-moderate
temperature processing in model systems and fruit juices, submitted for publication.
Ludwig, H., Kubel J. and Tauscher, B. Influence of UHP on Vitamin A acetate Content in: High pressure research in the biosciences and biotechnology, K. Heremans (ed.), Leuven University Press, Leuven, Belgum, 1997, pp 33 1-334.
Butz, P. and Tauscher, B. figh pressure treatment of fruit and vegetables: problems and limitations in: hgh pressure research in the biosciences and biotechnology, K. Heremans (ed.), Leuven University Press, Leuven, Belgium, 1997, pp 435-438.
144
High Pressure Food Science, Bioscience and Chemistry
May, T. and Tauscher, B. (1997). Influence of pressure and temperature on chlorophyll a in alcoholic and aqueous solution, submitted for publication.
Inactivation of Microorganisms and Enzymes in Pressure-treated Raw Milk
B. Rademacher*, B. Pfeiffer and H.G. Kessler Institute for Food Process Engineering, Dairy and Food Research Centre Weihenstephan, Technical University of Munich,D-85350Freising, Germany
1
INTRODUCTION
For the well established thermal processes, especially the pasteurization of milk, kinetic data for the inactivation of microorganisms and enzymes are available. Less data are available for pressure treated milk. First results about the plate counts, the shelf-life and the inactivation of enzymes in pressure-treated raw milk were presented on the EHPRG conference in Leuven [l]. Aim of the present work is to specifjl pressure/time-combinations giving milk a shelf-life of 10 days, which is comparable to the shelf-life of thermal pasteurized milk. And because there w ill be a need for internal process markers if high pressure is applied, the effect of pressure on the inactivation kinetics of milk enzymes is studied.
2
MATERI.AL,ANDMETHODS
The experiments were carried out with one day old, bull< raw milk which was
High Pressure Food Science, Bioscience and Chemistry
146
pressure-treated in a pilot plant consisting of one pressure vessel of 200 ml. Pressure increase and release rates were kept constantly at 200 MPdmin. The temperature of pressure treatment was set to 20 "C by a waterbath. The total plate counts were determined after 3 days'incubation at 30 "C on PC-agar with 1 %
skimmilk powder. The end of shelf-life was defined as time when the pH-value decreases below 6.5 or sensory changes are detectable. Inactivation of all enzymes was examined as relative enzyme activity in relation to the activity of non-treated raw milk. The activity of Alkaline phosphatase (EC 3.1.3.1) was analyzed by the FluorophosB-method (Advanced Instruments) [2-4]. The method used for assaying y-Glutamyltransferase (EC 2.3.2.2) activity is based on the enzymatic cleavage of the colourless substrate y-glutamyl-p-nitroanilide and the formation of the yellow p-nitroaniline, which is determined photometrically at a wavelength of 410 nm [5]. The photometric determination of Phosphohexoseisomerase (EC 5.3.1.9) activity is based on the reaction of Glucose-6-phosphate to Fructose-6-phosphate and the subsequent formation of a red dye by a mixture of resorcinol, ethanol and hydrochloric acid after heating [6].
3
RESULTS AND DISCUSSION
Previous work has shown that the total plate count of raw milk decreases with increasing pressure in the pressure range fiom 200 to 500 MPa. Additionally the inactivation effect is dependent on the pressure holding time in the first
Food Science: Presentations
147
15 minutes of treatment. Longer holding times give no additional inactivation
effect [l] but extending the pressure to 600 and 700 MPa can improve the effect.
To prove whether fat has an effect on the inactivation of microorganisms, raw cream with a fat content of 40 % was pressure treated in the same way. The initid count of microorganisms of 3.2.104 per ml (including 40 sporedd) was only slightly different to the experiments with raw milk.
Figure 1. Total plate counts of raw cream after pressure treatment with various pressure-/time combinations Very similar results were obtained compared to the raw milk (Fig. 1). A pressure treatment of 700 MPa for a few minutes leads to a maximal effect of a log 2.5 inactivation. As expected fat does not influence the inactivation of microorganisms if these microorganisms are not entrapped in fat globules or in pure oil.
148
High Pressure Food Science, Bioscience and Chemistry
The shelf life of the pressure-treated raw milk was estimated in dependence of pressure and pressure holding time at a storage temperature of 10 "C.Resuming that pasteurized milk normally has a shelf life of about ten days, a W-treatment of 400 h4Pa and 30 min as well as a treatment at 500 MPa and 5 m h will give an equal effect of ten days shelf-life at a storage temperature of 10 "C.But the main target of pasteurization is the inactivation of pathogenic bacteria. Investigations with different strains of vegetative pathogens like Listeria mono-
cytogenes, Escherichia coli and Staphylococcus aureus in UHT-milk showed that some of these strains, especially the E. coli 0157:H7, NCTC 12079 have a very high pressure resistance. In the case of E. coli a pressure treatment of 600 MPa and 30 minutes only gave a log 2 reduction [7]. That means, that pressure/time-combinations, which would lead to a shelf life of ten days, do not fulfill the requirements of complete destruction of pathogens. Furthermore the pressure resistance of the most heat stable vegetative bacterium in milk,
Mycobacterium tuberculosis, has not been investigated yet. In order to provide an useful process marker for the pressure treatment of milk the inactivaton kinetics of milk enzymes were studied, The enzymes investigated were chosen on the basis of their sensitivity against heat treatment. All of them can be inactivated by a thermal pasteurization [2,5,6]. Figure 2 illustrates the influence of pressure and holding time on the inactivation of the Alkaline phosphatase, expressed as decrease of relative enzyme activity. Pressures above
149
Food Science: Presentations
500 MPa are necessary to inactivate the enzyme to more than 50 %. From the
curves the Activation volume AV' and the Overall rate constant
~GJcould
be
estimated by non-linear regression [8]. For the inactivation of Alkaline phosphatase at 20 "C the order of reaction was determined as n=2.4, the Activation volume amounts to about -56 mVmol and the Overall rate constant is 2-10-95-'.In further experiments pressure increase and release rates were varied &om 100 MPdmin to 500 MPdmin. The calculated inactivation effect of pressure
increase and release, based on the received kinetic data, was in conformity with measurement. Therefore it can be concluded that the inactivation of the enzymes is only determined by the effects of pressure height and holding time. There is no further effect of decompressionon the inactivation (results not shown). 1.o
-
0.8
I
Y
0.6
2 c 0
m
.9 c
-m B
0.4
0.2
0 0
20
40
60
80
100
120
140
160
Pressure holding time t H [ rnin 1
Figure 2. Influence of pressure and holding time on the inactivation of Alkaline phosphatase in raw milk
150
High Pressure Food Science, Bioscience and Chemistry
Pressure [MPa]
Figure 3. Inactivation of native milk enzymes after pressure treatment for 8 min Figure 2 reveals that the Alkaline phosphatase is relatively pressure resistant. Processes with a maximum pressure of 600 MPa would require more sensitive indicators. Therefore the suitability of Glutamyltransferase and Phosphohexoseisomerase as pressure process markers was tested. Figure 3 compares the degree of inactivation of the three enzymes, expressed in terms of relative activity,
dependent on the height of pressure applied. The pressure holding time was fixed to 8 minutes. It is visible that the most pressure sensitive enzyme is the Phosphohexoseisomerase, for which the inactivation starts at pressures of about 400 MPa. The inactivation of Glutamyltransferase to a detectable amount demands a pressure treatment above 500 MPa. Most pressure resistant is the Alkaline phosphatase. Loss of relative activity starts in the range of 600 MPa but complete inactivation requires a pressure of 800 MPa and a few minutes..
Food Science: Presentations
4
151
CONCLUSIONS
The inactivation effect of the native microorganismsin raw milk and raw cream is nearly the same. Fat does not influence the inactivation. Referring to the aspect of shelf life (at 10 "C) a pressure treatment of milk at 500 MPa for a few minutes is equivalent to a thermal pasteutiZation, which means 72 "C for 15 seconds. Inactivation of the milk enzymes Phosphohexoseisomerase, y-Glutamyltransferase and Alkaline phosphatase occurs as a resdt of UHP-treatment. Loss of enzyme activity starts at 400, 500 and 600 MPa respectively. A Combination of these enzymes may act as an internal process indicator for the intensity of pressure treatment of milk.But it is not yet known if the kinetics of enzyme inactivation are sufficiently close to the destruction of Mycobacterium tuberculosis or other
pathogens to provide a useful marker for the destruction of these organisms.
5
REFERENCES
1. B. Rademacher, H.G. Kessler (1997). High Pressure Research in the Biosciences and Biotechnology, K. Heremans (Ed.), Leuven, Leuven University Press, 291-293. 2. R.M. Rocco (1990). J. Food Protection 53,588-591. 3. E. Lechner, S. Ostertag (1993).Deutsche Milchwirtschaft 44, 1146-1 149. 4. E. Lechner, V. Regensburger (1993). Deutsche Milchwirtschaft 44, 815-819. 5. G. Zehetner, G. Bareuther, T. Henle, H. Klostermeyer (1995). Z Lebensm Unters Forsch 201,336-338. 6. G. Zehetner, G. Bareuther, T. Henle, H. Klostermeyer (1996). Netherfands Milk and Dairy Journal 50,215-226. 7. M.F. Patterson, M. Quinn, R. Simpson, A. Gilmour (1995). J. Food Protection, 58,524529. 8. J. Hinrichs, B. Rademacher, H.G. Kessler (1996). Milchwissenschd? 51, 504-509.
Effect of High Pressure Processing on Properties of Emulsions made with Pure Milk Proteins Eric Dickinson* and Jonathan D. James Procter Department of Food Science, University of Leeds, Leeds LS2 9JT, UK
The influence of high-pressure processing (up to 800 MPa for 15-120 minutes at 25 "C) on the emulsification properties of milk proteins at pH 7 has been investigated in model systems. Treatment of j3-casein or sodium caseinate before emulsification has negligible effect on average droplet size of fine oil-inwater emulsions. In contrast, the emulsifying efficiency of j3-lactoglobulin is sensitive to both treatment pressure and duration. Treatment (2400 MPa) of 0lactoglobulin-stabilizedemulsions after emulsification can induce significant levels of droplet flocculation. In concentrated systems (220 vol% oil) this leads to substantial changes in complex shear modulus associated with the conversion of a liquid-like emulsion into a viscoelastic emulsion gel. 1. INTRODUCTION
The main effect of high-pressure processing (up to 1000 MPa) on dissolved proteins is disruption of quaternary structure and unfolding of tertiary (and to a lesser extent secondary) structure [ 1-31. Hence, in milk products, one expects (and finds) a marked effect of high pressure treatment on the structure and functional properties of the large supramolecular protein aggregates ('casein micelles') and the globular whey proteins [3-51, but not on the properties of individual disordered caseins (e.g. j3-casein) or sodium caseinate [6,7]. We are concerned here with how high pressure affects the emulsification properties of j3-lactoglobulin at neutral pH. Preliminary work has shown [81 that pressure treatment (up to 800 MPa) prior to homogenization leads to loss of emulsifying efficiency as measured by change in droplet-size distribution of fine emulsions made at relatively low proteidoil ratio. Loss of emulsifying efficiency was attributed to protein aggregation following pressure-induced unfolding. It was also observed [8] that pressure treatment after emulsification leads to less significant change in average droplet size. Similar trends of
Food Science: Presentations
153
behaviour were also found [9] for analogous emulsions prepared with a commercial sample of whey protein concentrate. The new results reported here extend the previous work with P-lactoglobulin to a wider range of conditions. Comparison is made with the effects on the emulsifying efficiency of p-casein and sodium caseinate. To eliminate any complications from thermally induced structural effects, attention is particularly directed towards keeping close control of temperature (25 k 1 "C) during processing. Based on experience with other protein emulsion gel systems [lo], we study the rheology of P-lactoglobulinstabilized emulsions of high volume fraction as a way of amplifying the effects of possible changes in interdroplet interactions following pressure treatment.
2. MATERIALS AND METHODS The P-lactoglobulin (lot no. 114H7005), imidazole, sodium azide and ntetradecane were purchased from Sigma Chemicals (Poole). The p-casein was supplied by the Hannah Research Institute (Ayr), and the sodium caseinate was from DeMelkindustrie Veghel (Netherlands). High-pressure processing was carried out using a Stansted Mark I1 'Mini Food Lab' (Stansted Fluid Power, Essex) with in-built thermostatting system. Samples of protein solution (20 ml) or protein-stabilized emulsion (8 ml) were hermetically sealed in polyethylene bags and subjected to pressures of 200,400, 600 or 800 MPa for dwell periods of 15, 30, 60 or 120 min. Temperature excursions due to the effects of adiabatic heating/cooling were minimized by carefully controlling rates of compressioddecompression.In this way, except for the first 1-2 min, the temperature was kept at 25 f 1 "C during the whole of the processing cycle (compression, pressure dwell, decompression) as shown in Figure 1 for the case of a sample subjected to 400 MPa for 20 min. Oil-in-water emulsions (10, 20 or 40 vol% n-tetradecane) were prepared at ambient temperature using a laboratory-scale homogenizer operating at 40 MPa. The aqueous phase was a protein solution dissolved in 20 mM imidazole buffer (pH 7.0) containing 0.1 wt% sodium azide. Emulsions were prepared within 30 minutes of pressure treatment. Droplet-size distribution and average diameter d43 were obtained by static multi-angle light scattering (Malvern Mastersizer). Creaming was monitored by eye over a 10 day period in quiescently stored samples at 25 f 1 "C. Small-deformation rheology was studied in an oscillating concentric cyclindrical cell using a controlled stress Bohlin CS-50 rheometer. Storage and loss shear moduli were determined at 25 "C over the frequency range 10-3 to 10 Hz.
154
High Pressure Food Science, Bioscience and Chemistry
28 400
27
26 300 a,
L
a cn cn
24
200
a c.
f
23 n
t!
L
oe
25 2
100
21 20
0
0
5
10
15
20
25
30 35
40
Time (minutes)
Figure 1. Pressure/temperature profile during high-pressure treatment at 400 MPa for 20 minutes: -,pressure; ,temperature.
-
3. RESULTS AND DISCUSSION
Figure 2 shows the changes in the average droplet size d43 of emulsions made with 20 vol% oil and 0.4 wt% solutions of P-lactoglobulin subjected to various high-pressure treatments prior to homogenization. Results indicate that the treatment reduces the emulsifying capacity of the protein, and that the extent of the reduction increases with increasing pressure and dwell time. The trend is consistent with preliminary measurements made earlier with different protein sample and processing equipment, but it seems likely that the substantial difference in d43 values reported previously [8] between untreated and 200 MPa treated samples were overestimated. The larger droplets produced with the pressure-treated protein give an increased creaming rate, as noted earlier [8]. Figure 3 shows an analogous plot to that of Figure 2 except that now the plactoglobulin aqueous phase concentration is increased by 50% to 0.6 wt%, which leads to a reduction in the average droplet size for the emulsion made with the untreated sample by around 25%. Again the magnitude of the effect increases with treatment pressure and duration. The changes in emulsion droplet size in Figures 2 and 3 are explicable in terms of pressure-induced denaturation of the protein leading to protein aggregation. Despite an increase in surface hydrophobicity, aggregation involving disulphide bonds [ 11-13] causes a loss of emulsifying efficiency due to reduction in the proportion of protein available for effectively covering the oil-water interface during emulsification [8,9].
155
Food Science: Presentations
0.8
1 0
I
200
400
600
800
Pressure (MPa)
Figure 2. Average droplet size d43 of emulsions (20 vol% oil, 0.4 wt% P-lactoglobulin, pH 7) made with protein subjected to pressure treatment 15 min; 0 , 3 0 min; A, 60 min. before emulsification: 0,
0.9 10.8 y-
0
200
400
600
800
Pressure (MPa)
Figure 3. Average droplet size d43 of emulsions (20 vol% oil, 0.6 wt% P-lactoglobulin, pH 7) made with protein subjected to pressure treatment 15 min; 0 , 3 0 min; A, 60 min; A, 120 min. before emulsification: 0, High-pressure treatment of p-casein or sodium caseinate gives no discernible change in emulsifying efficiency. An emulsion prepared with 20 vol% oil + 0.4 wt% native p-casein at pH 7 was found to have exactly the same d43 value (0.84 k 0.01 pm) as one prepared with p-casein which had been treated at 800 MPa for 60 min. Similarly, emulsions made with 20 vol% oil + 2 or 10 wt% sodium caseinate at pH 7 have the same values of d43 (0.58k 0.02 or 0.81 k 0.02 pm,
156
High Pressure Food Science, Bioscience and Chemistry
respectively) as those made with sodium caseinate pre-treated at 200-800 MPa for 15-60 min. This expected insensitivity of the surface functional properties to high pressure-processing is consistent with the disordered casein molecular structure [6]. Table 1. Average droplet size d43 of P-lactoglobulin-stabilizedemulsions subjected to high-pressure treatment for 60 min after emulsification. d43 (crm)
Pressure (MPa) 10 vol% oil
-
0.65
200
0.74
400
0.90 0.95 0.99
600 800
20 vol% oil 0.71 0.70
0.95 0.98 1.01
40 vol% oil 0.73 0.77 1.02 1.19 1.29
We now turn to the effect of high-pressure treatment on P-lactoglobulin after emulsification. Table 1 lists d43 values for three sets of emulsions prepared with (i) 10 vol% oil + 0.25 wt% protein solution, (ii) 20 vol% oil + 0.5 wt% protein solution, and (iii) 40 vol% oil + 1.0 wtYo protein solution. Since whey protein emulsions exhibit time-dependent flocculation without pressure processing [ 141, the emulsion samples made with untreated f3-lactoglobulin were stored at 25 "C for the duration of the pressure treatment cycle, and droplet-size distributions were determined at the same time as for the treated samples. The results in Table 1 indicate that, at processing pressures of 2 400 MPa for 60 min, there is a significant increase in the average droplet size of emulsions following highpressure treatment. This can be attributed to pressure-induced cross-linking between adsorbed protein molecules on different droplets. Similar cross-linking occurs when P-lactoglobulin-stabilizedemulsions are heated [ 101. Figure 4 compares the droplet-size distribution in the 20 vol% emulsion treated at 800 MPa for 60 min with that for the same emulsion freshly prepared and aged without pressure processing. The results show that the degree of broadening of the distribution towards larger particle sizes, due to flocculation of emulsion droplets through interdroplet disulphide bridging [ 141, is enhanced by the high-pressure treatment. As shown in Table 1, this increase in droplet flocculation at the higher pressures (600 and 800 MPa) is particularly evident for the most concentrated emulsion (40 vol% oil).
157
Food Science: Presentations
Figure 4. Droplet-size distributions P(d) of P-lactoglobulin-stabilized emulsion (20 vol%oil, 0.5 wt% protein):-, fiesh sample, -, after ageing (- 2 h); - * -, after treatment at 800 MPa for 60 min.
10 8
6
A
i
G* (Pa)
4
2 0
0
200
400
600
800
Pressure (MPa)
Figure 5. Complex shear modulus G* (0.01 Hz,25 "C) of plactoglobulin-stabilized emulsions subjected to high-pressure treatment for 60 min after emulsification: 0, 10 vol%oil, 0.25 wtYo protein; 0 , 2 0 vol% oil, 0.5 wt% protein; A, 40 vol% oil, 1.O wt% protein.
High Pressure Food Science, Bioscience and Chemistry
158
Figure 5 shows the influence of high-pressure treatment on the emulsion rheology of the same set of P-lactoglobulin-stabilizedemulsions. Irrespective of the treatment pressure, the complex shear modulus G* at 0.01 Hz of the low viscosity 10 vol% emulsion remains < 0.04 Pa. For this dilute emulsion system, it is not surprising that the moderate degree of induced droplet flocculation is insufficient to cause any significant change in the rheological behaviour. However, for the 20 and 40 vol% emulsions, where there is a higher density of droplets and cross-links, we do observe a substantial increase in G* at treatment pressures of 400 MPa and above. Figure 6 shows the effect of high-pressure treatment on the frequency dependence of storage and loss moduli, G and G", of the 40 vol% oil emulsion (1 wt% protein). The reduction in the frequency dependence of the emulsion is indicative of an increase in solid-like character.
0.001
0.01
0.1 Frequency (Hz)
1
10
Figure 6. Influence of high-pressure treatment on the frequency dependence of storage and loss moduli, G' and G", of P-lactoglobulinstabilized emulsion (40 vol% oil, 1.O wt% protein): untreated, 0, G', 0 , G I ; 800 MPa for 60 min, A,G', A, G". Cross-linking between P-lactoglobulin layers on different droplets during prolonged high-pressure processing therefore converts the concentrated liquidlike emulsion (G' < G"at low frequencies) into a viscoelastic emulsion gel. A qualitatively similar change in rheology has been reported also by Dumay et al. [7] for 0-lactoglobulin-stabilized triglyceride oil-in-water emulsions. These results on model systems demonstrate the potential for using high-pressure technology as a processing tool for modifying the texture of food products based on milk proteins.
Food Science: Presentations
159
Acknowledgement J.D.J. acknowledges receipt of a BBSRC CASE Research Studentship in collaboration with Unigate (St. Ivel).
4. REFERENCES 1. 2.
3. 4.
J. L. Silva & G. Weber, Annu. Rev. Phys. Chem., 44 (1993) 89. K. Heremans, in High Pressure Processing of Foods (eds D. A. Ledward, D. E. Johnston, R. G. Earnshaw & A. P. M. Hastings), Nottingham University Press, Nottingham, 1995, p. 8 1. W. Messens, J. Van Camp & A. Huyghebaert, Trends Food Sci. Technol., 8 (1997) 107. D. E. Johnston, B. A. Austin & R. J. Murphy, Milchwissenschuft, 47 (1992) 760.
5.
6.
7.
8.
E. M. Dumay, M. T. Kalichevsky & J. C. Cheftel, J. Agric. Food Chem., 42 (1994) 1861. E. Dickinson, B. S. Murray & K. Pawlowsky, Food Hydrocoll., in press. E. Dumay, C. Lambert, S . Funtenberger & J. C. Cheftel, Lebensm. Wiss. Technol., 29 (1996) 606. V. B. Galazka, E. Dickinson & D. A. Ledward, Food Hydrocoll., 10 (1996) 213.
9.
10.
11.
V. B. Galazka, D. A. Ledward, E. Dickinson & K. R. Langley, J. FoodSci., 60 (1995) 1341. E. Dickinson, in Proceedings of 1st International Symposium on Food Rheology and Structure (eds E. J. Windhab and B. Wolf), Vincentz Verlag, Hannover, 1997, p. 50. V. B. Galazka, I. G. Sumner & D. A. Ledward, Food Chem., 57 (1996) 393.
12.
13.
V. B. Galazka & D. A. Ledward, in High Pressure Research in the Biosciences and Biotechnology (ed. K. Heremans), Leuven University Press, Leuven, Belgium, 1997, p. 375. S . Funtenberger, E. Dumay & J. C. Cheftel, Lebensm. Wiss. Technol., 28 (1995) 410.
14.
D. J. McClements, F. J. Monahan & J. E. Kinsella, J. Food Sci., 58 (1993) 1036.
Molecular and Functional Properties of HP-treated Egg Components
F. Bonomi"*,E. Donnizzelli", S. Iametti", P. Pittiab,P. Rovere", G.F. Dall'Aglio" "DISMA, University of Milan, Via Celoria 2,20133 Milano, Italy Dipartimento di Scienze degli Alimenti, University of Udine, Via Marangon 3 1, Udine, Italy SSICA, Via Tanara 3 1, Parma, Italy Eggs, either whole or as separated yolk and albumen, are widely used in
human nutrition both for direct consumption and as ingredients in a countless number of food preparations. Despite their versatility and their nutritional value, concerns about their cholesterol content, about their potential antinutritional effects, and about their safety from a microbiological standpoint are always looming on consumers, food authorities and food technologists. Furthermore, several of the treatments aimed at increasing the shelf-life of eggs and of eggderived products to be used as food ingredients very often yielded materials that had lost some of the properties relevant to the food technologist. Attempts at substituting high-pressure for heat treatments in stabilizing eggs and egg-derived products very often resulted in the formation of gels that, although having interesting features by themselves, hindered use of HP-treated raw eggs, yolk or albumen as a direct replacement for the fiesh materials. In this study, egg white and yolk were treated under conditions that did not result in significant macroscopic alteration of the fresh material, although they
Food Science: Presentations
161
were sufficient to reduce significantly microbial contamination. In particular, when adding 10% sucrose or 7-10% NaCl to whole eggs or to egg albumen, and then treating these materials at pressures ranging fiom 4500 to 8000 atm for times up to 5 min at nominal room temperature (=25"C), samples retained their liquid aspect, and no formation of gels was observed. To check the effects of these treatment conditions on common microbial
egg contaminants, non-sterile samples of a mixture of fresh whole egg (yolk and albumen) were inoculated with different, non-spore forming organisms (3* 10' cWml for each species). In samples containing 7% NaCl (a& 0.904) treated for 5 min at 8000 atm, reductions in bacterial counts ranged fiom 3 log units for S.
aureus and S. fecium to 6 log units for E. coli.
Table 1. Lipolyzability of yolk triglycerides as a function of pressure Yolk
separated from raw fresh eggs
(+lo%
NaCI) was treated at the given
pressures for 5 min. Lipolyzability by a commercial lipase was measured by automatic titrimetry, and is given as a percent of that of untreated yolk.
treatment pressure, MPa
lipolyzability
400
600
800
55
70
117
HP-treated yolk (+ 10% NaC1) was characterized by assessing the lipolyzability of yolk triglycerides by a tiglyceride-specific fungal lipase (Lipolase 100, Genencor, UK) [I]. As shown in Table I, lipolyzability increased
High Pressure Food Science, Bioscience and Chemistry
162
with treatment intensity, suggesting the occurrence of HP-induced modifications
in the structure of the complex associations among lipoproteins, phospholipids, and triglycerides in yolk, as reported in the literature for many lipid model systems [2,3]. Treatment of albumen in the presence of protective agents left it in liquid form, but resulted in an increase of its viscosity. As shown in Table I1 for treatment at 600 MPa, the increase in viscosity was independent of the length of the treatment and was more pronounced for samples treated in the presence of sucrose. In the absence of protein polymerization (vide infra) the increased Viscosity could be related to unfolding of proteins, as observed in model systems [41.
Table II. Viscosity of HP-treated albumen NaCl or sucrose were added to albumen separated from raw fresh eggs, and these samples were treated at 600 MPa for the time given. Viscosity (given in mPa) was determined by a stress-controlled rheometer at 20°C at a shear rate of 1.99 sec.’.
addition
freatment time, min 0
5
I0
15
10% sucrose
49
455
394
445
70% NaCI
10
158
156
174
The absence of gelation in egg albumen treated in the presence of protective agents allowed to study some molecular modifications in albumen proteins, that could have practical relevance. Among the investigated parameters
Food Science: Presentations
163
were the conversion of N-ovalbumin to the so called S-form (typical of aged egg albumen), the formation of polymeric aggregates [5-71, the sensitivity of the treated samples to proteases, and the recognition of ovalbumin in the treated egg white by suitable antibodies [8].
No alteration of the overall protein pattern was observed in HPLC anionexchange chromatography of the HP-treated albumen samples. In particular, there was no noticeable conversion of the N- to the S-form of ovalbumin. SDS-PAGE was used to monitor formation of aggregates, and showed only marginal formation of covalent ovalbumin dimers at the highest pressures, with a concomitant modest decrease in the content of native ovalbumin and of lysozyme.
Table 111. Proteolysis of treated albumen as a function of pressure
Albumen separated from raw fresh eggs (+ 10% NaCl or 10% sucrose) was treated at the given pressure for 5 min. Proteolysis by trypsin was measured spectrophotometrically after precipitation with trichloroacetic acid, and is given as a percent of that of the untreated material.
addition
treatment pressure, MPa
400
600
800
10% NaCl
246
47 1
47 1
10% sucrose
253
1527
694
Susceptibility of the treated albumen to proteolysis by trypsin increased with the seventy of the treatment. As shown in Table 111, trypsin acted more
High Pressure Food Science, Bioscience and Chemistry
164
extensively on samples treated in the presence of sucrose, as observed for isolated ovalbumin [4]. From a molecular standpoint, increased proteolysis could be related to increased protein unfolding, as suggested by the increase in Viscosity in treated albumen previously discussed in this report. Finally, recognition of ovalbumin by polyclonal antibodies in an indirect, non-competitive ELISA assay decreased markedly upon treatment, in a pressureindependent way. We found that after treating albumen for 5 min at 400-600 MPa in the presence of either sucrose or NaCl, less than 5% of ovalbumin in the treated sample was recogmzed by specific antibodies. In conclusion, HP-treatment of eggs and egg components in the presence of protective agents produced materials that remained in liquid form, and that had interesting properties from both a nutritional and a technologcal standpoint. Treatment in the presence of NaCl apparently introduced a lower extent of modification in albumen than that in the presence of sucrose. Also, the effects of treatment at the molecular level in the presence of either protective agent were different as for their dependence on combinations of time and pressure, indicating the possibility of fme-tuning HP treatment of albumen with regard to properties required from the treated material.
References 1.
Iametti S, Bonomi F, Giangiacomo R, Tragna S, Versuraro L (1997) Dairy Science Int., in press
2.
Tauc P, Mateo CR, Brochon JC (1997) In “High pressure research in the biosciences and biotechnology” (K Heremans, ed.) Leuven University Press, Leuven, Belgium pp 171-174
Food Science: Presentations
3.
165
Wolf A, Bernsdorff C, Winter, R (1997) In “High pressure research in the biosciences and biotechnology” (K Heremans, ed.) Leuven University Press, Leuven, Belgium pp 1 83- 186
4.
Iametti S, Donnizzelli E, Rovere P, Dall’Aglio GF, Vecchio G, Bonomi F, this Meeting
5.
Iametti S, Cairoli S, De Gregori B, Bonomi F (1995) J. Agric. Food Chem. 43: 53-58
6.
Iametti S, De Gregori B, Vecchio G, Bonomi F (1996) Eur. J. Biochem. 237: 106-112
7.
Iametti S, Transidico P, Bonomi F, Vecchio G, Pittia P, Rovere P, Dall’Aglio G (1997) J. Agric. Food Chem. 45: 23-29
8.
Turin L, Bonomi F (1994) J. Sci. Food Agric. 64: 39-45
Combined Application of Sub-zero Temperature and High Pressure on Biological Materials Rikimaru Hayashi*,Toshihiko Kinsho and Hiroshi Ueno Division of Applied Life Sciences, Graduate School of Agriculture, Kyoto University, Sakyo, Kyoto 606-01, JAPAN
The unique characteristics of high pressure on biological macromolecules were studied, especially when they are exposed to high pressure at low temperature, including sub-zero temperature. Our results imply useful principles in applying high pressure to food science and biotechnology. Temperature and pressure relationship of protein denaturation - Fig. 1
represents the T-P diagram of protein denaturation as described by Hawley in
1972 (1).
Protein denaturation occurs outside the elliptic curve, where the
areas A, B, and C correspond to cold-denaturation,pressure-denaturationand heat-denaturation, respectively. It is also shown in Figure l b that the upper and lower temperature limits for the cold- and heat-denaturation, respectively, approach each other when the incubation pressure is raised. Recent progress in
high pressure food science and technology has assured us that enzyme and microbial inactivation also obey the predictions of this T-P diagram. However, little study has been carried out on this important subject.
167
Food Science: Presentations
Celt mhbrane Cell wall
I
Cold denaturation
Heat denaturation
250
J
100
I . O.1MPa -20
0
20
40
-20
0
,
J
20
40
T ("C)
Fig. 1 T-P Diagram of protein denaturation Cold denaturation of carboxypeptidase Y - Carboxypeptidase Y is a
serine protease obtained from Succharomyces cerevisiae. It has a molecular weight of 61,000 of which 16% is accounted for by carbohydrate. It contains 5 disulfide bridges and is stable over a wide range of temperatures, including
sub-zero temperature, at ambient pressure. Repeated freeze and thawing do not inactivate the enzyme. Carboxypeptidase Y is inactivated when it is exposed to high pressure at sub-zero temperature (Table I). Diluted solutions
of carboxypeptidase Y were used for the study of cold inactivation. Kinetics of the inactivation at 400 Ma, which followed first order kinetics at -10 "C but did not do so at 0 "C or higher, suggested that there are both irreversible
,
168
High Pressure Food Science, Bioscience and Chemistry
and reversible processes involved for the inactivation. The reversibility was tested when activity was followed after the depressurization, where spontaneous recovery of the activity was observed for 10 "C and 0 "C-treated samples. However, no activity recovery was evident for the -20 "C-treated sample, suggesting that the cold-inactivation under sub-zero temperature was irreversible. Circular dichroism study showed that the cold-inactivated enzyme lost a significant part of its secondary structure. Table I. Cold-inactivation of carboxypeptidaseunder pressure Temperature
("C) 22 0 -10 -20 -30
Relative activity under the essure (MPa) 0.1 100 1 200
1
100.0 96.1 93.9 91.3 96.7
91.9 84.3 78.3 76.6 92.1
90.0 76.9 64.4 54.3 49.8
6.20 2.80
4.4 6.5
Effects of several additives on the pressure inactivation were examined at sub-zero temperature. Sodium chloride, sugar and glycerol protected from pressure inactivation at sub-zero temperature.
Cold and heat inactivation of CPY - Accumulating evidence has suggested that there are some differences in the process of denaturation between pressure and temperature denaturation of protein (2). CarboxypeptidaseY is inactivated either by pressure treatment at 400 MPa at -10 "C or heat treatment at 60 "C under 0.1 MPa. Both inactivations gave
Food Science: Presentations
169
first order kinetics at the same reaction rate and almost complete inactivation within 30 min. Conformational differences between cold- and heatinactivated CPY were examined with 8-anilino-1- naphthalenesulfonate (ANS), a fluorescent probe specific to hydrophobic areas. Cold inactivated enzyme exhibited an extended reaction with A N S but not the heat inactivated enzyme. Proteolytic digestibility was also different between the two denaturations: Extensive proteolysis using yeast proteinase A was evident for the heatinactivated enzyme while cold-inactivated one did not react. Based upon reactivity against ANS and protease, it is clear that the denatured protein structure is significantly different in the cold- and heat-inactivation processes. In summary, the cold-inactivated enzyme still maintains its original structure
in part to resist against proteolysis. Heat-inactivation,on the other hand, gives extensive damage to the tertiary structure and renders it totally susceptible to proteolysis. Pressure effect on protein gelation- Intermediate states of protein
denaturation due to the pressure treatment could create protein gels having unique properties. High pressure-induced gels of egg albumin were easily formed at elevated tempexature and also at subzero temperature. The properties of these high pressure gels are unique as already shown by many food scientists (3).
High Pressure Food Science, Bioscience and Chemistry
170
Microbial inactivation by pressurization at subzero-temperature -
Yeasts are inactivated effectively by pressurization at raised temperature or sub-zero temperatures when high pressure is being applied (Fig. 2) (4). The T-
P diagram of bacterial, as well as yeast, inactivation rate is drawn as elliptic curves similar to T-Pdiagram of protein denaturation. Like protein, cold and heat inactivations of yeast are also observed when the pressure is raised, and
both inactivation temperature limits approach each other when the pressure is increased.
It is important to note that more extensive pressure
A
B
2
400
.
.
,
, . , , Inactivation Rate
U Y p1
2 -2
4
-20 -20
0
20
40
Temperature (C)
0
20
Temperature (SC)
Fig. 2 T-P curves obtained by regression analysis for the inactivation of
S. cerevisiue (A) and inactivation rate (B).
(o), 0.1 MPa; curve 2 (e), 120 MPa; curve 3 (A), 150 MPa; curve 4 (A), 180 MPa; curve 5 (o), 210 MPa; curve 6 (m), 240 MPa; (A): Curve 1
curve 7 (V), 270 MPa
(B): See the text for dotted lines.
4Q
~
Food Science: Presentations
171
sterilization can be achieved under low temperature, including sub-zero temperature conditions, than at room temperature. For example, as shown by the dotted lines, only 190 MPa is needed to obtain the inactivation rate of 1 per min at -20 "C, but 320 MPa is required to reach the same rate of inactivation at 20 "C. Since low temperature treatment requires much lower pressure than room temperature treatment, the former condition gives great economical value to high pressure equipment and technology. Effects of typical food additives, i.e. sucrose, glycerol, citrate and
sodium chloride on yeast inactivation were examined by pressure treatment at 200 MPa and -20 "C or at room temperature. Sucrose, glycerol and citrate inhibited both cold and heat inactivations of yeast as shown in protein denaturation. However, sodium chloride addition protected against pressure inactivation at room temperature but enhanced pressure inactivation at -20 "C. The effects of additives on cold-inactivation were different between protein and yeast. These findings are important in two aspects. First, salt addition
may improve the sterilization effect of high pressure. Secondly, a systematic study of the salt effect may give clues for new mechanistic understanding of microbial inactivation caused by high pressure. Pressure efects on starches - Only a few studies are known for high
pressure effects on starches (5). A pressurization study was carried out on a
172
High Pressure Food Science, Bioscience and Chemistry
water suspension of corn starch. Starch granules swell by pressurization, and birefringence disappeared almost completely, a phenomenon called pressure gelatinization. Gelatinization temperature, Tm, at 0.1 MPa and gelatinization pressure, Pm, at room temperature were measured by DSC and amylase digestibility, respectively. The relationship between Tm and Pm of various starches was plotted on T-Pplane. Three types of starches designated as A, B and C, potato starch, wheat or corn starch, and intermediate type,respectively, were clearly differentiated on the T-P diagram. These differences in starch types were first recognized by this study as well as by X-ray analysis. If we measure Pm at different temperatures including subzero temperatures, we can have more information about starch structure.
Conclusion - High pressure effects on protein denaturation and protein gelation, yeast inactivation, and starch gelatinization were reported in focusing on sub-zero temperature effects. Among characteristics of high pressure on biological materials are its high reversibility, different path or process from other perturbants, i.e. heat, leading to denaturation, and easiness to observe intermediate state of conformationalchange, and cold denaturation of biological macromolecules. These unique properties of high pressure treatment can be applied not only to basic research, i.e. structural study of protein and biological materials, but also biotechnology fields, including food
Food Science: Presentations
173
science. It is desirable to expand our studies on cell membrane and cell wall for pressure effects under sub-zero temperature. For high pressure treatment of food materials such as meats and vegetables, such infomation is required
for producing better pressure-processed food products.
In order to deepen our understanding of high pressure effects on living organisms and living matter, the pressure effect on free water and bound water must be studied extensively. To know details of pressure effects, it is also important to observe the effect of salt and sugars on biological macromolecules, including cell membranes and the cell wall.
References Hawley, S.A. (1 97 1) Reversible pressure-temperature denaturation of chymotrypsinogen.Biochemistry 10,2436-2442 Hayashi, R., and Balny, C. (Eds.) (1996) High pressure bioscience and biotechnology in Progress in Biotechnology. Vol. 13, pp. 522, Elsevier, New York Dumoulin, M.,Ozawa, S., and Hayashi, R. (1997) Textural properties of pressure-induced gels of food protein obtained under different temperatures including subzero temperatures. J. Food Sci. in press Hashizume, C., Kimura, K., and Hayashi, R. (1995) Kinetic analysis of
174
High Pressure Food Science, Bioscience and Chemistry
yeast inactivation by high pressure treatment at low temperature. Biosci Biotech Biochem 59,1455-1458 5.
Hayashi, R., and Hayashida, A. (1989) Increased amylase digestibility of pressure-treated starch. Agric. Biol. Chem 53,2543-2544
Influence of High Pressure Treatment on P-Lactoglobulin and Bovine Serum Albumin in the Absence and Presence of Dextran Sulphate V.B. Galazkaa*,J. Varleya, D. Smithb,D.A. Ledwardaand E. Dickinson'
aDepartmentof Food Science and Technology, University of Reading, Whiteknights, Reading RG6 6AP, U.K. bBBSRCInstitute of Food Research, Earley Gate, Reading RG6 2EF, U.K. 'Procter Department of Food Science, University of Leeds, Leeds LS2 9JT, U.K.
1. INTRODUCTION
Proteins and polysaccharides are both present in many food products ranging fiom ice cream to mayonnaise. The overall structure - property relationship in food colloids depends not only on hnctional properties of the individual biopolymers, but also on the nature and strength of interactions of these biopolymers with each other. Depending on the biopolymers and the solution conditions (ionic strength, pH, temperature), the interactions may be specific or non specific, weak or strong, repulsive or attractive [ 13. Recently, it has been shown [2,3] that P-lactoglobulin (p-Lg) and bovine serum albumin (BSA) when subjected to high pressure processing leads to denaturation and aggregation at low ionic strength. The strong aggregation is mainly considered to be due to the formation of intermolecular disulphide bridges via -SW-S-Sinterchange. One would expect that non-covalent protein-polysaccharide interactions to be modified as a result of pressure-induced changes in protein conformation and in electrostatic interactions. One of the model systems chosen for this investigation,BSA + dextran sulphate (DS), forms an electrostatic complex at pH 7 and low ionic strength [4]. Recent studies [ 5 ] have reported that in oil-in-water emulsions containing these two biopolymers, pressurization (2 400 MPa) of BSA before homogenization leads to changes in the flocculation and rheological behaviour of the resulting BSA emulsion after DS addition.
High Pressure Food Science, Bioscience and Chemistry
176
2.MATERIALS AND METHODS BSA (product A-2934),p-Lg (product L-0130),DS (5 x lo5Da), and buffer salts were purchased from Sigma Chemical Co. (St. Louis, MO). Imidazole buffer solutions (0.02M, pH 7) for surface tension and foaming experiments were prepared ( A N S ) ammonium salt using HPLC grade water. 1-Anilinonaphthalene-8-sulfonate was obtained from SERVA (Feinbiochemica, Heidelberg). Individual proteins (0.001 - 1 wt%, pH 7) and mixtures of protein (0.001 - 1 wt%) + polysaccharide (1, 0.5 or 0.25 by weight) for structural analysis, foam stability and surface tension experiments were prepared and subjected to high pressure treatment as previously described [2,6]. Changes in protein surface hydrophobicity were estimated by reaction with A N S , and thermal stability by differential scanning calorimetry (DSC) [2].Surface tensions were measured at the air-water interface using the Wilhelmy plate technique (Kriiss K,, Processor Tensiometer) and foaming experiments using a sparging technique [6].
3. RESULTS AND DISCUSSION The probe spectrofluorimetry data for pressure treated BSA indicate a substantial reduction in protein surface hydrophobicity (Table I), which could be due to the lower number of hydrophobic groups binding to the A N S due to burying of some binding sites in the partially denatured and refolded protein or because of increased intermolecular interactions [8].Addition of the anionic polysaccharide DS to BSA greatly reduces the surface hydrophobicity for the native and pressure treated forms. Table I Effect of high pressure on the surface hydrophobicity of p-Lg, BSA and mixtures of 0-Lg + DS ( 1 :1 by weight) and BSA + DS (2:1 by weight) in aqueous solution (pH 7, 0.1 wt% protein, 4 x mol/dm3 A N S ) . Relative fluorescence intensity I is monitored at 470 nm from excitation at 350 nm. Treatment time 20 min. Quoted data are the averages of triplicate measurements. Treatment Pressure ( m a ) 0 200 350 550
P-Lg 22 39 66 85
Fluorescence Intensity (0 0-Lg + DS BSA 22 153 34 150 45 127 59 88
BSA + DS 85
78 64 48
177
Food Science: Presentations
In contrast, p-Lg shows a significant increase in surface hydrophobicity with increase in pressure in the absence and presence of DS. At the higher treatment regimes (350 and 550 MPa) the surface hydrophobicity for the p-Lg + DS mixtures does not increase to the same extent as the individual protein. ANS does not bind to the polysaccharide alone (I - 0.8). In the presence with p-Lg or BSA it is assumed that there is electrostatic repulsion between the two negatively charged molecules, and the extensive blocking to ANS of the protein's binding sites could well be due to the electrostatic protein-DS complexation. Differential scanning calorimetry data (Table 11) for BSA indicates that there is a major loss of native structure when treated at 600 or 800 MPa, which is not recovered after pressure release. The presence of DS leads to decreases in T,, and AH, in the native and pressurized samples, which suggest that complexes are formed between BSA and DS. It is thought that these may reduce the extent of BSA unfolding during heating and that pressure treatment of the mixed system can reduce this even further. Table I1 Influence of high pressure treatment on the endothermic peak temperature (Tn,)and total calorimetric enthalpies (AH) of solutions of BSA (1 wt%) and mixtures of BSA ( 1 wtYo) + DS (0.5 wt%) at neutral pH. Experiments were performed in duplicate. BSA:DS ratio (by weight) 0 0 0 0
0.5 0.5 0.5 0.5
Pressure ( m a )
Tm ("C)
AH (kcaymol)
0 300 600 800 0 300 600 800
62.4 62.2 62.4 56.7 52.6 53.2 46.9 44.6
15.1 lo4 15.6 x lo4 8.0 x 104
5.1 2.2 2.9 2.5 1.7 x
lo4 lo4 lo4 lo4 lo4
Under similar experimental conditions, DSC thermograms for p-Lg also suggest a significant loss of tertiary structure after pressure treatment (native protein T, = 73.3 "C; pressure treated at 800 MPa for 20 min T , = 38.3 "C). Addition of DS to native p-Lg (1:1 by weight) leads to no significant change in T, (73.6 "C), and the mixture denatured by pressure treatment has a T,,, of 59.3 "C, with a very low calorimetric enthalpy [2].
178
High Pressure Food Science, Bioscience and Chemisrry
Recent [7] size exclusion chromatography experiments at neutral pH for BSA and mixtures of BSA + DS suggest that the protein-protein complexation induced by pressure is influenced by disulphide bridging, whereas the BSA-DS complex is electrostatic in nature with or without pressure. This complex at low ionic strength protects the BSA against aggregation caused by -S-S- bridging during treatment. It is also inferred that this protective effect may, in part, be due to the blocking of the hydrophobic surface binding sites on the protein by the bulky poiysaccharide moieties. Circular dichroism measurements (V.B. Galazka, unpublished data) at pH 7 show that pressure treatment (800 MPa for 20 min) gives a 35 % reduction in the ahelix content [6]. The presence of DS (1 :1 by weight) with native BSA reduces the a-helical content by 15 YOand this is hrther reduced (10 YO) during pressure treatment. These results seem to agree with the size exclusion chromatography data in that they confirm that the polysaccharide has a protective effect on the protein, by reducing pressure-induced modification and aggregation.
58
0
0.5
1
1.5
2
2.5
3
W
Figure 1. Surface tension (6 h) as a finction ofpolysaccharide (DS):proteinweight ratio W: 0, native samples; (3, pressure treated samples (520 MPa for 20 min). Total BSA content 1 x 10” wt%, pH 7,20 mM; 25 “C. Surface tension (y) at 6 h for pure BSA and mixed BSA + DS solutions at various po1ysaccharide:protein weight ratios (W) before and after pressure treatment
Food Science: Presentations
179
(520 MPa) are presented in Figure 1. In this set of experiments, when the DS is mixed with the BSA at low ionic strength (20 mM), the mixture BSA + DS (1 :3 by weight) gives higher surface tension values (y = 57.39 f 0.18 mN m-') than BSA alone (y = 52.55 k 0.2 mN m"). The data indicate the formation of complexes which prevent significant reduction in surface activity after pressure processing (treated BSA + DS ( 1 :3 by weight) y = 57.39 mN m-I). Time-dependent surface tension measurements at the air-water interface (6 h) for 0-Lg give a lower surface tension after pressurization at 400 MPa for 20 min (untreated p-Lg = 5 1.73 f 0.03 mN rn-'; treated p-Lg = 50.7 k 0.02 mN rn-I), and the presence of DS leads to no change in surface activity (untreated p-Lg + DS (1 :3 by weight) y = 5 1.75 mN m-'; treated p-Lg + DS (1 :3 by weight) y = 50.72 mN rn-').
250
.-C
-EE
200
2P
150
0
100
I
i=
-m0 0
E a
50 L-
m I
0 0
0.5
1
1.5
2
2.5
3
W
Figure 2. Time for foam half volume collapse as a jkction of polysaccharide (DS):proteinweight ratio W: a, native samples; :>, pressure treated samples at 750 MPafor 20 min. Total BSA content I x I 0-2wt%, pH 7, 20 mM, 25 "C. Figure 2 shows the effect of high pressure (750 MPa) on foam stability of BSA and BSA + DS solutions at neutral pH and low ionic strength. This is a plot of time for half volume collapse Thagainst po1ysaccharide:protein weight ratio W.We note that there is a reduction in BSA foam stability after pressure treatment, which can be attributed to protein unfolding followed by aggregation due to -S-S- bridging and possible interaction of hydrophobic groups to form oligomers. Addition of DS to the
High Pressure Food Science, Bioscience and Chemistry
180
-
native BSA significantly improves foam stability at high ratios ( W 1 or 3) [9]. Pressure treated BSA + DS mixtures have lower foam stability, but at higher weight ratios ( W 1 or 3) the foam stability is higher than that for the pressure treated BSA in the absence of DS. In contrast, 0-Lg foam stability is increased after pressurization (untreated p-Lg Kh = 5 min; treated at 350 MPa for 20 min Kh = 53 min) [6], and the presence of DS at W 1 or 3 (untreated mixture Tx = 4 min) does not affect foam stability to any measurable extent [9]. In all cases pressurization of the mixtures leads to lower foam stability (0-Lg + DS (1 :3 by weight) T, = 29 min), but the foam stability is higher than that for the native p-Lg.
-
-
4.CONCLUSIONS Pressure treatment affects surface activity and the foaming properties of both proteins. There is a reduction in BSA foam stability after pressure treatment, which may be due to a decrease in surface hydrophobicity, unfolding and aggregation of the protein. Addition of DS to the native and pressure treated BSA significantly improves foam stability. This is attributed to the electrostatic complexation of BSA with DS which prevents aggregation of BSA during pressure treatment. p-Lg foam stability is increased after pressurization due to an increase in surface activity and protein surface hydrophobicity. The addition of dextran sulphate (DS) to either native and pressure treated protein does not change surface activity or foam stability to any significant extent.
5. REFERENCES 1. E. Diclunson, & S.R. Euston, in Food Polymers, Gels and Colloids (ed. E. Diclunson), Royal Society of Chemistry, Cambridge, U.K., 1991, p. 132. 2. V.B. Galazka, I.G. Suinner & D.A. Ledward, Food Chem., 57 (1996) 393. 3. E.M. Dumay, M.T. Kalichevsky & J.-C. Cheftel, J Agric. Food Chem., 42 (1994) 1861. 4. E. Dickinson & V.B. Galazka, in Gums and Stabilisersfor the Food Industry (eds. G.O. Phillips, P.A. Williams & D.J. Wedlock), IRL Press, Oxford, U.K., 1992, vol. 6, p. 35 1. 5. E. Dickinson & K. Pawlowsky, J. Agric. Food Chem., 44 (1996) 2992. 6. V.B. Galazka, D.A. Ledward & J. Varley, in Food Colloids: Proteins, Lipidr and Polysaccharides (eds. E. Dickinson & B. Bergensal), Royal Society of Chemistry, Cambridge, U.K., 1997, p. 127.
Food Science: Presentations
181
7. V.B. Galazka, D.A. Ledward, I.G. Sumner & E. Dickinson,J. Agric. Food Chem., 45( 1997) in press. 8. I. Hayakawa, J. Kajihara, K. Morikawa, M. Oda & Y .Fujio, J. Food Sci., 57 (1992) 288. 9. E. Izgi & E. Dickinson, in Food Macromolecules and Colloids (eds. E. Dickinson & D. Lorient), Royal Society of Chemistry, Cambridge, U.K., 1995, p. 312.
Effect of Hydrostatic Pressure on the Physicochemical Properties of Bovine Milk Fat Globules and the Milk Fat Globule Membrane C . Kanno, T. Uchimura, T. Hagiwara, M. Ametani and N. Azuma Utsunomiya University, Department of Applied Biochemistry, Utsunomiya 321, Japan Summary The effect of hydrostatic pressure on the physicochemical properties of
milk fat globules and the milk fat globule membrane (MFGM) was studied. The median diameter (2.6 pm) and size distribution (1 to 10 pm) of the milk fat globules were not affected by pressurization at 100-400 MPa, but were respectively increased to 4.1 pm and 1 to 18 pm at 800 MPa. Lipoprotein lipase could not attack milk fat globules pressurized even up to 800 MPa, indicatingthat hydrostatic pressure did not cause damage to the membrane enveloping the milk fat globule. About 25% protein and phosphorus in the pressurized isolated MFGM was precipitated by centrifugation at 3000 x g , indicating that MFGM was aggregated. The membrane enveloping milk fat globules ruptured once by sonication was repaired by pressurization.
Introduction 'The effect of hydrostatic pressure has been studied to better understand the structure and function of proteins and food components (Hayashi, 1992; Tauscher, 1995). 'The change in structure and/or function of proteins under hydrostatic pressure is accompanied by the formation of hydrogen bonds, the collapse of hydrophobic interaction, and the separation of ion pairs (Cheftel, 1992).
Food Science: Presentations
183
Nearly all of the fat in milk, which constitutes about 3.5% of bovine milk, is dispersed in the aqueous phase as separate milk fat globules ranging from 0.1 to 15 pm in size. The total number of milk fat globules is estimated to be roughly
3.6 x 10' and their specific surface area to be about 2 m2 per g of milk fat for a
Holstein cow. A milk lipid droplet is synthesized in the secretory epithelial cell of the mammary gland, enveloped by the apical plasma membrane and secreted into the alveolar lumen (Keenaner al., 1988; Kanno, 1990). A milk fat globule of 3.7 pm average diameter consists of core lipid and about 7 nm thickness of surface membrane. The interfacial membrane of fat globules in milk is called the milk fat globule membrane (MFGM) which consists of 44% protein and 55% lipids (Keenan et al., 1988; Kanno, 1980). Interaction between the core lipid and membrane in the milk fat globule and the protein and lipids in the membrane are mostly hydrophobic. Milk fat globules may be changed as a result of the weakening of hydrophobic interaction andor the formation of hydrogen bonds under hydrostatic pressure. The present study was undertaken to make clear the effect of hydrostatic pressure on the physicochemical properties of milk fat globules and isolated milk fat globule membrane.
Materials and Methods 1. Preparation of milk fat globules and the milk fat globule membrane
Cream was recovered by centrifuging fresh milk, which had been obtained from Holstein cows of the university herd, for 15 min at 5000 x g and 37"C, and then by washing 3 times with its 3 volumes of distilled water at 37°C. The washed cream was suspended in distilled water, and the lipid content was adjusted to 30% (w/v).
High Pressure Food Science, Bioscience and Chemistry
184
MFGM was prepared from washed cream by the method of Kanno and Kim (1990) and suspended in a 0.5 M phosphate buffer (pH 7.0). The protein concentration in the MFGM suspension was I0 mg/ml.
2. Sonication of the washed cream Washed cream in the 0.5 M phosphate buffer (pH 7.0) was sonicated for 2
rnin in an ice bath at 60 W by a 250 Sonifier (Branson). 3. Pressurization Washed cream and the MFGM suspension (1 0 mg of proteidml) were put into a Teflon tube with a screw
cap (4 or 8 ml in volume) and pressurized for
10 min at 37°C and 100-800 MPa with hand-operated oil pressure apparatus
(HRl5-EQ , Hikari Koatsu Co., Hiroshima, Japan). The required pressure was reached within 0.5-2 min and released to atmospheric pressure within 1 min. Pressurized MFGM was separated into the supernatant and precipitate by centrifuging for 15 rnin at 3000 x g. 4. Lipoprotein lipase
Pressurized milk fat globules were mixed with an equal volume of fresh skimmed milk containing indigenous milk lipoprotein lipase and then incubated for 15 or 30 min at 37"C, the reaction being stopped by heating for 2 rnin in a boiling water bath. The released free fatty acids were determined by modifications of the semi-automated method of Bowyer et a / . (1978).
5. Analyses Electrophoresis on polyacrylamide gel in the presence of sodium dodecyl sulfate was performed by the method of Laemmli ( 1970). The size distribution of the fat globules was measured at 37°C by a laser scattering particle size distribution analyzer (Horiba LA-500). Protein was determined by the method of
Food Science: Presentations
185
Markwell ef af. (1978), while phosphorus was determined by ICP apparatus (Seiko Electric Co.). The lipid content in the cream was measured by the Gerber method. Results and Discussion
1. Effect of pressure on the size of fat globules The median diameter of milk fat globules in the washed cream was about 2.6 pm, the size distribution ranging from 1.2 to 10 pm (see Fig. 2-B). Washed cream was pressurized for 10 min at hydrostatic pressures from 100 to 800 MPa, and the resulting median diameters are indicated in Fig. 1. In comparison with the size of the unpressurized globules, the median diameter of the pressurized globules hardly changed between 100 and 400 MPa, slightly increased between
500 and 600 MPa, and markedly rose to about 4.I pm at 800 MPa. These results show that the size of the milk fat globules was apparently not affected by pressures up to 400 MPa.
n
E
=L
4.2
3.8
L Q)
Y
0
5
3.2
. I
U C
m
F
. I
E
2.6
-I
-1
2.0 J C ' I I 1 . 1 0.1 200 400 600 Pressure (MPa)
I
L
800
Fig. 1. Effect of pressurization on the median diametzr of milk fat globules.
High Pressure Food Science, Bioscience and Chemistry
186
The size distribution of fat globules before and after pressurization is compared in Fig. 2, which shows the typical size distribution pattern of washed cream pressurized at 400 MPa (B) and at 800 MPa (C) compared with that without pressurizing (A). The size of the globules in the unpressurized cream ranged from 1.2 to 10 pm with a median diameter of 2.65 pm (A), while that of
Size d ist ri b ut ion 2o
A: Unpressurued
1 Size difference 1
0
15 r
1
15 I ~~
I
0
15 0.1
7.5
I
10 20 Globule size (pm) 1
0 0.1
10 20 Globule size (pm) 1
Fig. 2. Size distribution of unpressurized milk fat globules (A) and those pressurized at 400 MPa (B) and 800 MPa (C). D and E are expressed as the differences between A and B, and A and C, respectively.
Food Science: Presentations
187
the globules in pressurized cream was 1.2 to 18 pm at 400 and 800 MPa, 2.81pm at 400 MPa (B) and 4.66 pm at 800 MPa (C). The difference in the size distribution of globules between cream pressurized at 400 MPa (B) and the unpressurized cream (A) was relatively small (Fig. 2-D). The difference in the size distribution of globules between cream pressurized at 800 MPa (C) and the unpressurized cream (A) is shown in Fig. 2-E. The frequency of globules smaller than 4 pm was decreased, while the larger than 4 pm were increased by pressurizing the cream at 800 MPa. Although the size of the fat globules was not change by pressure in the 100-400 MPa range, it is considered that pressurization may have caused damage to the membrane enveloping the fat globules. If the membrane of a fat globule were partially ruptured, core lipids in the fat globule would be attacked through the ruptured membrane by milk lipoprotein lipase, and free fatty acid would be released (Shimizu et al., 1982). Bovine milk contains sufficient lipoprotein lipase to cleave the core lipid of fat globules (Bengtsson and Olivecrona, 1981). Washed cream pressurized at between 100 and 800 MPa was incubated by mixing with fresh raw skimmed milk containing indigenous milk lipoprotein lipase. The amount of free fatty acid released from the unpressurized cream was 0.42 mmol/l5 min, while no increased release of free fatty acid was apparent by the fat globules after pressurization at between 100 and 800 MPa. These results indicate that the membrane enveloping the fat globules was not ruptured by hydrostatic pressure, even up to 800 MPa. Milk fat globules were stable to hydrostatic pressure, one reason being their globular shape. Pressure is transmitted in a uniform and instantaneous manner. Volume reduction is not likely to be caused in a fat globule, since the core of the globule is lipid, of which 95% is triacylglycerol; the volume reduction of water in a milieu is close to about 15% at 600 MPa and 22°C. Another reason for stability
188
High Pressure Food Science, Bioscience and Chemistry
is that the milk fat globule is enveloped with the cell membrane derived from mammary epithelial cells. 25
Pmtein
20 n
5 15 Y
C
g
10
E
0
0.1
100 200 300 400 Pressure (MPa)
Fig. 3 . Change in the amounts of protein and phosphorus in the precipitate separated from the pressurized milk fat globule membrane. Bars indicate SD for 3 measurements.
2. Effect of pressure on the milk fat globule membrane The membrane of the fat globule was found to be stable under 100-400 MPa pressure. MFGM enveloping fat globules was isolated, pressurized at 100 to 400 MPa, and separated by centrifugation at 3000 x g for 15 min. The amount
of protein in the precipitate increased with increasing pressure and reached its maximum at 300 MPa (Fig. 3). About 25% of MFGM protein was precipitated from MFGM pressurized to 300 MPa. The amount of phosphorus in the precipitate also accounted for 25% of total phosphorus, most being due to phospholipid, and was similar to the change in protein (Fig. 3 ) . The greater amount of precipitate with slower centrifugation suggests that the aggregation of
MFGM resulted from hydrostatic pressure. An SDS-PAGE analysis of the fractions separated from pressurized and
Food Science: Presentations
189
unpressurized MFGM by centrifugation detected the major protein bands such as CB-1 of 155 kDa, PAS-5 of 55 kDa and PAS-6+7 of 47 kDa in both fractions (data not shown). No difference was found in the SDS-PAGE pattern of protein between the supernatant and precipitate, implying that no special proteins contributed to the formation of aggregated MFGM. The results suggest that MFGM was aggregated as a membrane matrix. Aggregation of MFGM seems to have been due to interaction between protein and protein of MFGM through hydrogen bonding and/or disulfide bonding by hydrostatic pressure. 3. Effect of hydrostatic pressure on sonicated milk fat globules The aggregation of isolated MFGM suggested that the ruptured membrane of fat globules could be repaired by hydrostatic pressure. Washed cream was partially ruptured by ultrasonication at 60 W for 2 min and then pressurized at between 400 and 800 MPa. Restoration of the damaged membrane was evaluated by the reaction with lipoprotein lipase, the results being shown in Table 1 and in Fig. 4.
~~~
Table 1. Effect of Sonication and Pressurization on the Size and Surface Area of Milk Fat Globules Sonication X*
O* X 0
X
0
Pressurization MPa 1 0.1 0.1 400 400 800 800
* x, untreated; 0,treated
Median diameter tbm) 2.58 2.66 2.56 2.62 3.66 4.77
Specific surface area (cm2 / cm3) 23,278 22,895 23,436 23,146 17,869 14,393
High Pressure Food Science, Bioscience and Chemistry
190
The size and specific surface area of fat globules in the cream with or without sonication are similar, and those pressurized at 400 MPa before and after sonication are also similar (Table 1). With cream pressurized at 800 MPa, however, the size of the fat globules were larger in the sonicated than in the non-sonicated sample. Furthermore, the size distribution of the non-sonicated and sonicated fat globules and that of the cream pressurized at 400 MPa after sonication was similar, ranging from 1.2 to 10 pm (Figs. 4 - 4 -B and -C), the difference between these samples being small (Fig. 4-E and F). The size distribution of fat globules pressurized at 800 MPa after sonication was broad, ranging from 1.2 to 18 pm, and was shifted in the direction of 20 pm (Figs. 4-D).Compared with the only sonicated sample (Figs. 4-H and -I), a difference between cream samples at 400 and 800 MPa was found in fat globules larger than 2.6 and 4 pm,
respectively. In comparison with the 800 MPa sample without sonication (Fig. 4-G), the sonicated and pressurized cream had more globules larger than 4 pm.
Sonication increased the amount of free fatty acids released by lipoprotein lipase from 0.72 mmol/30 min in the non-sonicated cream to 2.03 mmol/30 min in the sonicated sample. The amounts of free fatty acids in the pressurized cream at 400 and 800 MPa without sonication were similar to thoseof the control sample. On the other hand, increased pressurization of the sonicated cream markedly decreased the amounts of free fatty acids (1.56 mmoV30 rnin at 400 MPa and 1. I 0 mmolf30 min at 800 MPa). Thus, pressurization could repair the membrane
of fat globules which had ruptured once, suggesting that repair would be achieved by aggregation of protein in the surface membrane enveloping the fat globule. Hydrostatic pressure up to 400 MPa had no effect on size of the milk fat
Food Science: Presentations
191
globules. This provides an advantage for the pressure treatment of milk, because fat globules would not be destroyed.
Size distribution
Size difference
20
0 10
2
0 20 1B: Sonicated unmessurized
I
L4
10
0
ID:Sonicated & 800 MPa
1
15
7.5
0 0 0.1
1 10 20 Globule size (pm)
12 0.1
1 10 20 Globule size (pm)
Fig. 4. Size distribution of unpressurized milk fat globules (A), sonicated ones (B), and those sonicated and pressurized at 400 MPa (C) and 800 h4Pa (D). E, F, G , H and I are expressed as the differences between A and B, A and C, A and D, B and C, and B and D, respectively.
192
High Pressure Food Science, Bioscience and Chemistry
Acknowledgement This study was supported in part by a grant from National Federation of Dairy Manufacturers Cooperatives of Japan. References Bengtsson, G., and Olivecrona, T. 198 1. FEBS Lett., 128: 9- 12. Bowyer, D. E., Cridland, J. S., and King, J. P. 1978. J. Lipid. Res., 19: 274-280. Cheftel, J. C. 1992. High PressureandBiotechnology, ed. by Balny, C., Hayashi, R., Heremans, K., and Masson, P. John Libbey Eurotext, France, pp. 195-208. Hayashi, R. 1992. High Pressure andBiotechnology, ed. by Balny, C., Hayashi, R., Heremans, K., and Masson, P. John Libbey Eurotext, France, pp. 185-193. Kanno, C. 1990. Protoplasma, 159: 184-208. Kanno, C. 1980. Jpn, J. Zootech. Sci., 51: 75-88. Kanno, C, and Kim, D-H., 1990. Agric. Biol. Chem., 54: 2845-54. Keenan, T. W., Mather, I. H, and Dylewski, D. P. 1988. In Fundamentals of Duiry Chemistry (Wong, N. P. et al. eds .) third edition, pp. 5 1 1-82. New York: Van Nostrand Reinhold. Laemmli, U. K. 1970. Nature, 227: 680-685. Markwell, M. A. K., Hoos, S. M., Bieber, L. L., and Tolber, N. E. 1978. Anal. Biochem., 87: 206-2 10. Shimizu, M., Miyaji, H, and Yamauchi, K. 1982. Agric. Biol. Chem., 46: 795799.
Stamatoff, J., Guillon, D., Powers, L., Cladis, P., and Aadsen, D. 1978: Biochem. Biophys. Res. Commun., 85: 724-728. Tauscher, B. 1995. Z. Lebensm. Unters. Forsch., 200: 3-13.
High-pressure Processed Apple and Strawberry Desserts J. Arabasa, K. G6recka b, A. Grochowskab M. Fonberg-Broczeka9bY K. Karlowskib, E. Kostrzewac, J. Szczepeka, H. Scieiyriskab ,B. Windygab, D. Zdziennickac, J. Zurkowska-Betaa ,S. Porowskia a High Pressure Research Centre, Polish Academy of Sciences, UNIPRESS
Sokolowska 29/37,01-142 Warsaw, Poland, tel. 48 22 632 70 55, fax 48 3912 0331, e-mail [email protected] b National Institute of Hygiene, Chocimska 24,OO-791 Warsaw, Poland C Institute of Agricultural and Food Biotechnology, Rakowiecka 36,02-532 Warszawa, Poland SUMMARY This paper describes methods, equipment and results of high pressure treatment of h i t desserts. Pressure processing was carried out with a high pressure food processor with volume of 1.5 litre and pressure of 700 MPa, engineered by UNIPRESS. Water was used as pressure medium. Two series of experiments were performed (series A and B). In the series A, fresh strawbemes and apples of excellent quality were used as raw material. After the h i t s were crumbled, pectin, citric acid and sugar (515%) were added. The contents was mixed, warmed up to 60°C during 5 minutes and packed in 100 ml or 250 ml polystyrene packages, closed with thin metal foil covers. Then the packages were placed in high pressure vessel and subjected to pressure of 400 MPa in 15 min. at ambient temperature. After pressure processing, the h i t desserts were stored for 3 months at temperatures +3"C to + 5°C. Microbiological investigations carried out directly after processing and after 3months'storage, for contents of moulds, yeast and total count of microorganisms proved very good microbiological quality of the products. Also sensorial tests revealed high quality of the products, gelatinous consistency, almost natural colour and fresh h i t aroma. In the series Bystrawberry desserts sweetened with Aspartame (0.lg per 1OOg) were prepared and subjected to four different treatments: pre-processed at temperatures of 2OoC or 75°C and pressurised at 400 MPa and temperatures of
194
High Pressure Food Science, Bioscience und Chemistry
20°C or 50"C, for 10 minutes Directly after pressure processing, microbiological and sensorial evaluation were carried out and six physicochemical parameters were measured. It was found that sensorial quality of the strawberry desserts sweetened with Aspartame was very good for all tested samples. Insignificant losses of anthocyanin, ascorbic acid and Aspartame were found only in the samples pre-processed and pressurised at elevated temperature. INTRODUCTION Food products are excellent environment for growth of microorganisms, sometimes pathogens, which may cause food-borne diseases. Safety for consumers, quality and shelf-life of food products depend on properties of microorganisms and efficiency of food processing. A range of methods of food preservation are used for providing microbiological safety, among which the Ultra High Pressure (UHP) seems very promising because it offers opportunities of modifying plant food quality and functionality. Many experimental works [ 1,2] report that most of vegetative forms of microorganisms are pressure sensitive, especially at low pH which is characteristic for h i t products. Under these conditions, spore forms are unable to grow. But still more data are necessary defining the effect of inactivation of microorganisms in selected food products. The data published as yet are not full or not detailed. The extent of inactivation of microorganisms achieved at particular pressure depends on number of interacting factors, including pressure level and duration microbial species, processing temperature and substrate - composition of food products. Of great importance is investigation of pressure-induced changes in significant factors determining acceptability of food products, e.g. colour, flavour, consistency and taste. The aim of this experimental work was to consider the process optimisation for microbiological safety and quality of fruit desserts preserved using UHP method. MATERIALS AND METHODS Two series of investigations were conducted, (series A and B). For series A, fresh strawberries and apples of excellent quality were used as raw material. The h i t s were washed and useless parts were removed. Then the fruits were crumbled by cutting or mixing and pectin, citric acid and 5 1 0 % of sugar were added. The contents were mixed and heated only up to 60°C for 5 minutes. The warm h i t dessert was packed in commercial polystyrene packages (100 or 250 ml) and heat sealed,with thin foil cover. Then the samples were subjected to pressure of 400 MPa for 15 minutes at ambient temperature.
195
Food Science: Presentations
Sample No. 1
2 3 4
Temperature Pre-processing Pressurisation 20 "C 20 "C 20 "C 75 "C 50 "C 20 "C 50 "C 75 "C
Pressure
Time
400 MPa 400 MPa 400 MPa 400 MPa
10 min. 10 min. 10 min. 10 min.
The following parameters were determined before and after pressurisation: soluble solids (" Brix at 20°C), total titratable acidity (g citric acid100 g f.w.), pH, total anthocyanins total (ACN as pelargonidin-3-glucoside- PGN) and Degradation index (Di), using Fuleki method, Aspartame content and ascorbic acid content. For each part of raw strawberries and for all samples after pressurisation, the instrumental colour values L, a, b and dominating wave length were measured using Minolta Chroma Meter CR200. Microbiological investigations were carried out before and after pressure treatment according to IS0 methods: determination of colony count of moulds and yeasts. The appearance, colour smell, taste and consistency of samples were evaluated by trained panel (six persons) using 5-point quality scores. High pressure treatment was carried out in a high pressure vessel of pistoncylinder type, with volume of 1.5 litre, operating pressure 700 MPa, temperature from -20°C up to +8O"C. The vessel with inner diameter of I10 mm was equipped with an internal heat exchanger. Distilled water was used as pressure medium.
RESULTS AND DISCUSSION In our previous research [3, 53, pressure-induced inactivation of various microorganisms in food products was carried out using low-volume pressure vessels,
196
High Pressure Food Science, Bioscience and Chemistry
with diameter of 16 and 50 mm. The samples were packed in multilayer polypropylene bags. From those studies we could establish combination of temperature and pressure sufficient for achievement of fruit products of very good microbiological quality but only in model experiments. To start with a laboratory production of pressure processed fruit products packed in commercial polystyrene packagings it was necessary to check if the process parameters defined earlier are sufficient. For this aim we engineered new food-processor with maximum pressure 700 MPa, volume of 1.5 litre and temperature up to 80°C [3]. It was essential to investigate microbiological quality of the products defined by physico-chemical parameters. Two series of experiments (A and B) were prepared. In the series A, apple and strawberry desserts pre-processed at 60°C for 5 min., and pressurised at 400 MPa for 15 minutes were stored at temperature +3"C to + 5°C for 3 months. Refrigerator storage was essential because the applied high pressure treatment was insufficient for inactivation of natural fruit enzymes. Moulds and yeasts were examined in the samples directly after processing and after 3-months' storage. The tested UHP products were of very good microbiological quality: no moulds and no yeasts were detected. The products were characterised by lightly gelatinous consistency of fruit jelly, almost natural colour and fresh fruit aroma. They were significantly more tasteful than the products prepared by traditional food processing. In the series B, conditions of combined pressure-temperature treatment of strawberry desserts sweetened with Aspartame were studied. Strawberries were chosen due to their high sensitivity to even small deviations from process parameters (loss of taste and changes in colour due to decomposition of an thocyanins). The results are shown in Tables 11, I11 and IV. It is seen from table I1 that pressurising at 400 MPa in 10 minutes both at 20" C and 50°C does not influence pH, soluble solids and total acidity. In the samples No.1 pre-processed and pressurised at 20°C a insignificant decrease of Aspartame level (3%) and ascorbic acid (2%) were found. No influence of pressure on ACN was detected. The greatest changes of these factors were detected in the sample No.4, preprocessed at 75°C and pressurised at 400 MPa, 50°C. It is seen from Table I11 that the decrease of the instrumental colour values L, a and b (as compared with fresh raw h i t ) was the greatest for the sample No.4 (pre-processed at 75°C and pressurised at 50°C) and the lowest for the sample No. 1 (pre-processed and pressurised at 20°C). All samples of series B retained their characteristic original red colour after pressurisation.
197
Food Science: Presentations
Table 11. Physicochemical parameters of strawberry desserts before and after pressurisation, series B (arithmetic means from 5 estimations) Sample number and temperature of preprocessing and pressurisation
No.1 20°c, 20°c No.2 750c, zooc No.3 2ooc, 50°c No. 4 750c, 50°c
1
When tested: beforel after pressurisation
before after before after before after before after
-
--
PH
ACN
Di Aspar -tame
mgl lOOg
g/ lOOg
3.5 3.5 3.5 3.5 3.4 3.4 3.4 3.4 -
-0.69 6.4
8.0
31.7 1.40 31.4 1.40 28.8 1.46 28.4 1.48 36.2 1.44 36.2 1.47 32.9 1.47 31.7 1.53
-
0.100 0.097 0.100 0.095 0.100 0.094 0.100 0.092
Sample raw fruit sample No.1 sample No.2 raw fruit sample No.3 raw fruit sample No.4
L 28.60 28.95 27.37 30.06 28.38 30.64 27.20
a +30.28 +26.77 +24.40 +29.31 +29.30 +28.80 +23.09
b +17.52 +15.18 +13.27 +18.02 +13.27 +18.68 +13.86
h, nm 610 610 610 608 613 607 610
Sample No.1 No.2 No.3 No.4
colour 5.0
smell
taste 5.0 5.0 5.0
consistency 5.0 5.0 5.0
4.5
5.0
5.0 5.0
4.7
5.0 5.0
5.0 4.6
198
High Pressure Food Science, Bioscience and Chemistry
Sensorial evaluation confirmed very good quality of these pressure processed products. The appearance, colour smell, taste and consistency of samples evaluated by trained panel (six persons) using 5-point quality scores is presented in Table IV. It results from these evaluations that pressure of 400 MPa in 10 minutes did not decrease sensorial quality of strawberry desserts. Insignificant losses of anthocyanins, ascorbic acid and Aspartame seem to be caused only by elevated temperature, both in the thermal pre-processing and pressurising. All pressure-processed products were of good microbiological quality: preprocessing at 20°C subsequent pressurising at 400 MPa and 50°C for 10 minutes or thermal pre-processing at 75°C and subsequent pressurising at 400 MPa and 20°C for 10 minutes was sufficient to receive strawberry dessert of very good microbiological quality. The studies will be continued.
CONCLUSIONS 1. Pre-processing at 60°C and pressurising at 400 MPa and ambient temperature for 15 minutes appeared to be sufficient to achieve high pressure processed fruit products of very good microbiological quality characterised by gelatinous consistency, natural colour and fresh h i t aroma maintained during 3 months refrigerator storage. 2. For strawberry UHP desserts it was found that pre-processing at 20°C and subsequent pressure processing at 400 MPa, 50°C for 10 minutes or, alternatively, thermal pre-processing at 75°C with subsequent pressure processing at 400 MPa, 20°C for 10 minutes was sufficient to receive products of very good microbiological quality. 3. For all samples it was found that sensorial quality of the strawberry desserts sweetened with Aspartame was very good. Insignificant losses in contents of anthocyanins, ascorbic acid and Aspartame were found to be caused only by elevated temperature, both in the thermal pre-processing and pressurising 4. Commercial polystyrene packages heat-sealed with thin foil cover appeared to be suitable for UHP treated h i t desserts. 5. Using high hydrostatic pressure food processor with operating pressure up to 700 MPa, temperature -20°C up to 80°C and volume of 1.5 litre, for this research project engineered by UNIPRESS, it was possible to undertake laboratory production of UHP h i t products packed in commercial packaging. In the frame of this research about 200 kg apple and strawberry desserts were produced.
Food Science: Presentations
199
Acknowledgement This research was supported by the National Committee of Scientific Research Grant No P06G 005 10/1996-97.
REFERNECES 1. M. Fomberg-Broczek, B. Windyga, H. Sciernska, K. Gbrecka, A. Grochowska, B. Napiorkowska, Karlowski, J. Arabas, J. Jurczak, S Podlasin, S. Porowski, P. Salanski, J. Szczepek: The Effect of high pressure in vegetative bacteria and spores of Aspergillus Flavus and Bacillus Cereus, Proceedings of the Joint XV AIRAPT & X M I I EHPRG International Conference, Warsaw, Poland, September 11-15 1995, World Scientific Publishing Co. Re. Ltd, ed. W. A. Trzeciakowski 1996, 892 2. M. F. Patterson, M. Quinn, R. Simpson andA.Gilmour: Effects of High Pressure on Vegetative Patogenes, in: High Pressure Processing of Foods, Ed. D. A. Ledward, D.E. Johnson, R. G. Earnshaw, A.P.M. Hastings, Nottingham University Press, 1995, 47 3. J. Szczepek, J. Arabas, M. Fonberg-Broczek, J. Zurkowska-Beta, S. Porowski: Equipment for laboratory high pressure processing of fiuit products in High Pressure Research Centre, COPERNICUS PROGRAMME. Concerted action CIPA-CT94-0 195. Proceedings of Second Main Meeting, Warsaw, 13-14 December 1996 (in press). 4. J. Szczepek. J. Arabas: Experimental technique for high pressure investigation chemistry, biology and food science. XXMvth Meeting of the EHPRG, Leuven, Belgium, Sept. 1-5, 1996. Proceedings ed. K. Heremans, Leuven University Press, 1997,471 5 . J. Szczawinski, M. Szczawinska, B. Stahczak, M. Fonberg-Bmczek, J. Arabas, J. Szczepek: Effect of high pressure on survival of Listeria monocytogenes in ripened sliced cheeses at ambient temperature. XXXIVth Meeting of the EHPRG, Leuven, Belgium, Sept. 1-5, 1996. Proceedings ed. K. Heremans, Leuven University Press, 1997,295
Brining of Gouda Type of Cheese Curd at High Pressure and its Effect on the Cheese Serum
W. Messens and A. Huyghebaert*
Department of Food Technology and Nutrition, Faculty of Agricultural and Applied Biological Sciences, University of Ghent, Coupure Links 653, B-9000 Gent, Belgium
1. INTRODUCTION High pressure processing of milk is investigated by various researchers in view of its potential applications in the’dairy industry [l]. The pressure induced disintegration of the casein micelles in reconstituted skim milk started at 150 MPa and levelled off at about 400 MPa, and is accompanied by an increase in the level of both soluble calcium and phosphate [2,3] and by a pHshift. For non heat-treated pressurized milk, the pH-shift was nearly reversible in function of storage time [4]. By contrast, only a limited number of publications report on high pressure treatment of cheese curd. A patent [5] describes the ripening of Cheddar cheese in 3 days using high pressure instead of 6 months. Pressurizing Cheddar cheese curd leads to a continuous microstructure and lower microbial counts [6]. The effect of high pressure at the brining stage of small Gouda type of cheeses was studied during the ripening period. Also, the cheese serum was pressed from pressure respectively non-pressure brined cheese in order to investigate the disruption of the paracasein network in cheese by pressure.
20 1
Food Science: Presentations
2. MATERIALS AND METHODS Small Gouda type of cheese curds (450 g) from raw cow's milk (48+) were brined at high pressure and atmospheric pressure one day after manufacture by immersion in 22 % NaCl brine. The brining time was 50 min of which 30 min at high pressure. The cheeses were ripened at 14 "C and 85 % RH. UREAPAGE was applied for the separation of the polypeptide fraction non-soluble at pH 4.6 by means of PhastsystemTM (Pharmacia Biotech). The cheese serum was pressed from the cheeses [7] using a stainless steel press designed in collaboration with Laboratorium Soete (University of Ghent). The moisture, protein, fat and non-protein nitrogen and the calcium, natrium and phosphorus content were determined by resp. IDF methods and [8]. SDS-PAGE and FPLC (Superdex Peptide 10/30 HR-column, Pharmacia Biotech) were used to analyze the proteidpeptide fraction of the cheese serum. 3. RESULTS AND DISCUSSION From 200 MPa onwards, pressure brining leads to a higher pH one day after brining (Fig. 1). The pH-shift levels off from 300 MPa onwards. As ripening proceeds, the pH-shift '
gradually decreases towards zero.
,
As observed for pressurized milk
:
[4], the pH-shift of pressure brined
'
100
200 300 Pressure (MPa)
400
Fig. 1 pH-shift of pressure brined cheese at 1 ( W ) , 7 ( O ) , 14 (A) and 21 days (+) of ripening. The brining time was 50 min.
cheese could also be a consequence of the release of calcium phosphate into the cheese serum. Also, accelerated ripening could explain the pH-shift.
202
High Pressure Food Science, Bioscience and Chemistry
UREA-PAGE densitograms have indicated that the 13-casein degradation by plasmin is influenced by pressure (Fig. 2) since the concentration of y-caseins increased with increasing pressure. During the ripening time, y-caseins are accumulated (results not shown). At each ripening time, the y-casein concentration was higher for cheese brined at 300 MPa than at 0.1 MPa. In contrast, there were no significant differences in the degradation of as1-casein during maturation. This accelerated hydrolysis of 13-casein by pressure can be caused by the higher pH, by conformational changes of the paracasein gel structure leading to an increased exposure of susceptible peptide bonds from 13casein readily cleavable by plasmin andor by a potential
increase of plasmin
activities by altering its structure.
d
I
0
5
I 10
I
I5
I
20
I 25
I
M
D
M i p t i o n dlt.ncc (mm)
Fig. 2. UREA-PAGE densitograms of Gouda type of cheese brined at 0.1 (-), 100 (- - -1, 200 (- -), 300 *) and 400 MPa after 21 days of ripening. The cheese curd was brined for 50 min of which 30 min at high pressure (0
(0
0 )
Food Science: Presentorions
203
At 1 and 14 days of ripening, the cheese serum was pressed from the cheeses brined at 0.1 and 300 MPa. The pressed liquid contained a slightly opalescent aqueous solution with a yellow tint (the aqueous phase or cheese serum) and yellow liquid cheese fat. High pressure brining gives rise to a higher pH on day after the brining and a higher moisture content of the cheese (Table 1). This reflects the improved water binding characteristics of the paracasein gel structure by pressure. On dry matter basis, the composition of all cheese samples was similar. The pH-shift as observed for the cheeses was also found in the cheese serum. In accordance with cheese serum from Emmental cheese [9], the pH, total solids content and protein content of the cheese serum increased during ripening. The cheese serum obtained from pressure brined cheese had a higher dry matter content throughout the ripening, which was partly due to a higher total nitrogen content. Table 1. Composition of Gouda type of cheese and cheese serum derived from it after 1 and 14 days of ripening of cheese curd brined at 0.1 and 300 MPa for 50 mina
Cheese Cheese Cheese Cheese Cheese serum Cheese serum Cheese serum Cheese serum
RiDen. Pressure DH time (MPa) (days) 1 300 5.50 (0.01) 1 0.1 5.35 (0.02) 14 300 5.44 (0.03) 14 0.1 5.42 (0.01) 1 300 5.47 (0.01) 1 0.1 5.35 (0.03) 14 300 5.56 (0.03) 14 0.1 5.55
Moisture NaCl
TotalN Fat
409.5 (3.4) 405.8 (2.1) 331.6 (9.5) 307.7 (5.1) 911.4
34.59 (0.72) 34.45 (0.39) 39.14 (1.27) 39.22 (1.05) 3.39 (0.13) 2.56 (0.19) 10.11 (0.19) 8.20 (0.20)
(1.1)
922.4 (4.8) 849.1 (2.7) 861.6 (0.01) (2.2)
6.8 (0.7) 6.4 (0.2) 9.2 (1.4) 7.7 (1.0) 24.6 (1.9) 22.9 (1.4) 33.5
(1.8) 34.6 (2.9)
326.1 (2.7) 330.2 (1.5) 373.9 (5.8) 386.1 (5.4) -
-
Ca
P
6.87
4.02
(0.05) (0.16) 7.05 (0.10) 7.78 (0.16) 8.27 (0.06) 4.35 (0.13) 4.33 (0.20) 5.82 (0.12) 6.07 (0.15)
4.26 (0.48) 4.45 (0.14) 4.43
(0.56) 0.89 (0.01) 1.03 (0.03) 1.14 (0.03) 0.86 (0.02)
“Indicated results are expressed as g/kg cheese (serum). All determinations were performed in double on two cheeses and corresponding cheese serums. The 95 3’% probability interval is given between brackets.
204
High Pressure Food Science, Bioscience and Chemistry
The percentage of total nitrogen (TN) content [7] in the cheese serum increases during ripening. Using high pressure, the percentage of TN in the cheese serum was about 1.13units higher (35 % increase) after 1 day ripening and about 2.61 units higher (35 % increase) after 14 days ripening in comparison with cheese serum obtained from cheese at the same ripening stage brined at atmospheric pressure. This leads to the conclusion that the amount of pressure-induced nitrogen in the cheese serum keeps increasing during the ripening. This higher nitrogen content in the cheese serum of pressure brined cheese can be a result of pressure induced disruption of the paracasein network, but can also be caused by release of peptides, amino acids and further degradation products of protein hydrolysis. Pressure treatment did not significantly influence the calcium and phosphorus content in the cheese serum (Table 1). As a consequence, if disruption of the paracasein network in cheese takes place by pressure it is not accompanied by increasing levels of calcium phosphate in the cheese serum. In order to investigate the source of nitrogen content in the cheese serum,
firstly the non-protein nitrogen (NPN) content was determined (Table 2). Table 2. Total N, non-protein N (NPN) and protein N (PN) of the cheese serum after 1 and 14 days of ripening of cheese curd brined at 0.1 and 300 MPa for 50 minn Rip.
Pressure
time (days) (MPa)
I I 14 14
300 0.1 300 0.1
Total N TNwurn
~
Non-protein N
TNwum TNcheese
NPN,,,,
NPNwum TNwurn
Protein N PN,,,,
PN scnm TN senirn
(mg/lWg)
(%)
(mg/100g)
(%)
(mg/lWg)
(%)
339(13) 256(19) 1,011 (19)
4.5 3.3 10.1
97 (22) 89 (15) 519(11)
29 35 51
242(26) 167 (24) 492(22)
71 65 49
820(20)
7.5
441 (37)
54
379(42)
46
“All determinations were performed in double on two cheeses and corresponding cheese serums. The 95 % probability interval is given between brackets.
205
Food Science: Presentations
During cheese ripening, the TN content in the cheese serum increases due to an increase in both the PN and NPN content in the cheese serum. The relative amount of NPN content in the cheese serum increases during the ripening. Pressure brining gives rise to a higher TN content in the cheese serum, which is caused by both an increase in the NPN and PN content to the same amount. This is demonstrated by the non-pressure dependence of the relative amounts
of NPN and PN content in the cheese serum. The SDS-PAGE profiles of the cheese serums pressed from cheeses brined at 0.1 and 300 MPa at 14 days of ripening are the same (Fig. 3b). At I day of ripening, however, the relative peak surface of a band with a molecular weight of around 29.5 kDa was 26.6 % and 5.4 % for cheese brined at respectively
300 MPa and atmospheric pressure (Fig. 3a). This band corresponds to 13casein as confirmed by UREA-PAGE. This release of &casein from the paracasein gel structure is possibly caused by the disruption of hydrophobic bonds by pressure since B-casein is the most hydrophobic of all caseins.
0.1 MPa 300 MPa
0.1 MPa
300MPa
Fig. 3. SDS-PAGE profile of cheese serum after 1 day (a) and 14 days (b) of ripening of cheese brined at 0.1 and 300 MPa
206
High Pressure Food Science, Bioscience and Chemistry
As the ripening proceeds, more molecules with a molecular weight of 35,906; 18,165; 1,183; 491 and 106 Da are found in the cheese serum (results
of FPLC, not shown). Using pressure brining, especially a higher amount of proteins/peptides of 35,906; 18,165 and 1,183 Da are present in the cheese serum, at both ripening stages. As a consequence, pressure brining of cheese curd gives rise to a disruption of the paracasein network yielding more proteins and peptides in the cheese serum. Acknowledgements Research supported by the Flemish Institute for the promotion of scientifictechnological research in industry (IWT) and by FAIR-CT-96-1113. 4. REFERENCES 1 W. Messens, J. Van Camp & A. Huyghebaert, Trends in Food Science & Technology 8 (1997) 107. 2 Y. Shibauchi, H. Yamamoto & Y. Sagara, In High Pressure and Biotechnology; Balny, C., Hayashi, R., Heremans, K. and Masson, P. Eds., Vol. 224 (1992), 239. 3 S. Desobry-Banon, F. Richard & J. Hardy, J. Dairy Sci. 77 (1994) 3267. 4 K. Schrader & W. Buchheim, In High Pressure Research in the Biosciences and Biotechnology Proceedings of the XXXIVth Meeting of the European High Pressure Research Group; Heremans, K. Ed., (1997), 41 1. 5 H. Yokoyama, N. Sawamura & N. Motobayashi, European-PatentApplication EP 0469857 A1 (1992) 6 M.A. Torres-Mora, A. Soeldner, E.Y. Ting, A.C.O. Hawes, G.D. Aleman, G.S. Bakski, W.R. McManus, C.L. Hansen & J.A. Torres, Institute of Food Technologists annual meeting, (1996), 1082. 7 H.A. Morris, C. Holt, B.E. Brooker, J.M. Banks & W. Manson, J. Dairy Research, 55, (1988) 255. 8 R.M. Pollman, J. Assoc. Off. Anal. Chem., 74(1), (1991) 27. 9 J.A. Lucey, C. Gorry & P.F. Fox, Milchwissenschaft, 48(4), (1993) 183.
Denaturation and Functional Properties of Pressure-treated
Milk Proteins
J. Hinrichs* and H.G. Kessler
Institute for Food Process Engineering, Dairy and Food Research Centre Weihenstephan, TU-Munchen, D-853 50 Freising-Weihenstephan, Germany
[NTRODUCTION
Milk proteins basically consist of casein and whey protein. They are of special interest in food processing because of their functional properties. It is well known from heating that whey protein denaturation changes the functional properties. Although, pressure-induced denaturation of whey proteins has been recognized by several researchers, kinetic data in combination with functional properties have not been studied intensively. Therefore the objective of our work is to determine the kinetic data of whey protein denaturation and to study
High Pressure Food Science, Bioscience and Chemistry
208
how pressure treatment will influence the functional properties of milk proteins as well as the final product structure.
1 PRESSURE-INDUCED WHEY PROTEIN DENATURATION
The experiments were carried out with a whey protein isolate (BiPro, Davisco,
U S A . ) and pasteurised skim milk (Molkerei Weihenstephan, Freising, Germany) in an ultra-high pressure equipment containing ten vessels (aad, Bad Homburg, Germany). The content of the native whey protein fractions remaining after treatment was measured by Reversed Phase-High Performance Liquid Chromatography (1). a
4
2 2
Y
c -
0
/*
-21 /'
2 2
4
6
a
0
Pressure p [ 10' MPa]
Figure 1 Reaction rate constant for the denaturation P-Lactoglobulin A and B at various pressures and 30 "C The reduction of the different native protein fractions due to pressure treatment at a constant temperature depending on holding time and protein concentration
209
Food Science: Presentations
was measured in order to calculate the reaction rate constant kp,Tand the order of reaction n.
After determination, the order of reaction n, the reaction rates k,,T for different pressure levels at a constant temperature enable the pressure-independent reaction rate constant kO,T and the activation volume A.V" to be calculated by
AP
can be calculted from the slope and k , , from the intersection of the
Product
Whey protein fraction
Activation volume AP & S.D. [inI/mol]
Reaction rate Constant Ill (kn S.D.)
-,.+
Order of reaction
Correlation coefficient
n
r ~
I)
preliminary results
In Table 1 the determined data for the denaturation of the different whey protein fractions in whey protein isolate solution and skim milk at 30 "C are collected. The order of the reaction is lower in the case of skim milk compared with the
210
High Pressure Food Science, Bioscience and Chemistry
isolate, which could be due to the additional reaction of whey proteins with caseins which is known from heat treatment (2).
2 FUNCTIONAL PROPERTIES OF PRESSURE-TREATED MILK
PROTEINS Investigations were carried out with the same whey protein isolate and skim milk concentrated by reverse osmosis (RO) covering a concentration range of 2 to 20 % protein. Treatment conditions were chosen to achieve a degree of denaturation greater than D > 90 % (compare Fig. 2 right axis). The apparent viscosity was measured in a shear-controlled rheometer at a shear rate of 1000 per second at 10 "C. 1.o 0
$ 0.95
n
Pressure treated.
0 C
.-
-I I
2
0.90
a,
n 4: ol
+
Q
0.85 I"
0
2
4 6 8 Protein concentration (w/w)
10
12
[YO]
Figure 2 Apparent viscosity and degree of denaturation of pressure-treated whey protein solutions
Food Science: Presentations
21 1
The greater the protein content of the whey protein solution, the greater the increase in apparent viscosity. With protein concentrations above 10 % gel formation occurs (see Figure 2). High-pressure denaturation affects the water binding capacity of the protein due to the unfolding and aggregation of the molecules. By measuring the intrinsic viscosity it was possible to calculate that the water binding of a pressure-treated sample is about 2 times higher compared to a non-treated one. However, the water binding is still low compared to heating where, depending on heating conditions, a water binding increase of 3 to 5 times higher was found (3). In contrast to heat treatment, in pressure treatment at 30 "C fewer covalent bonds such as sulfhydryl bonds were formed which is supported by the fact that heating a high-pressure-induced gel (12 % protein) at 60 "C for 10 min reduces its strength by about 90 %. Consequently, it has to be assumed that mainly non-covalent bonds like hydrogen bonds are responsible for functional changes caused by high-pressure treatment at 30 "C of whey protein solutions. Additionally, the irreversible denaturation effect is mainly due to inter- and intramolecular reaction of the SH-groups (4). In Figure 3 the results of pressure-reated skim milk concentrate are shown. The viscosity of the non-treated skim milk concentrate is higher than that of the whey protein solution (compare Fig. 2) due to the higher water binding of the casein as well as to the minerals and lactose. Nevertheless, a lower increase in
212
High Pressure Food Science, Bioscience and Chemistry
viscosity was visible due to pressure treatment. Gel formation occurs at about 12 % protein. While in skim milk the main component is the casein, it is evident
that the whey proteins are the more reactive molecules in pressure treatment.
.
0
2
4
6
0
10
12
Protein concentration (w/w) [%]
Figure 3 Effect of high-pressure treatment on apparent viscosity of ROconcentrated skim milk
Figure 4 supports these observations in which the gel-strength is plotted against the protein concentration of the pressure-induced whey protein and skim milk concentrate gels. Due to pressure treatment a high increase of gel-strength of the whey protein isolate was visible and at a protein content of 20 % a rubber like texture was formed. In comparison, the skim milk concentrate showed a very low increase in gel-strength.
Food Science: Presentations
213
15
1
A
F'
ADlAt = 2.10' MPatmin
I
.
8
10
12 14 16 18 Protein concentration ( w h ) [%]
20
Figure 4 Effect of high-pressure treatment on gel-strength of whey protein isolate solution and RO-concentrated skim milk
REFERENCES (1)
BEYER, J. Diss. TU-Munchen Weihenstephan, 1990
(2) DEWAN, R. K.; BLOOMFIELD, V.A.; CHUDGAR, A. & M O M , C. V. 56 (6), 699-705, 1972 (3) KENNEL, R. Diss. TU-Munchen Weihenstephan, 1994
(4) TANAKA, N.; TSURUI, Y; KOBAYASHI, I. & KUNUGI, S.
International Journal of Biological Macromolecules 19, 63-68, 1996
Protein-Polysaccharide Interactions in Emulsions Containing High Pressure-treated Protein Karin Pawlowsky and Eric Dickinson Procter Department of Food Science, University of Leeds, LS2 9JT, U.K.
1. INTRODUCTION
Protein-polysaccharide interactions in emulsions influence the stability and rheology of the systems [l]. When the mixed biopolymer interaction is net attractive there is the possibility of bridging flocculation occurring. We have demonstrated previously that this mechanism takes place in bovine serum albumin (BSA) stabilized emulsions on addition of anionic polysaccharide dextran sulphate (DS) [2] or iota carrageenan (i-CAR) [3]. High-pressure treatment of the globular BSA leads to partial unfolding/aggregation [4]. Change in protein conformation is expected to alter the BSA-polysaccharide interaction which then may affect the emulsion behaviour. In this paper we compare some characteristics of BSA-stabilized oil-in-water emulsions prepared with native or pressure-treated protein and containing one of the three anionic sulphated polysaccharides: (i) DS, a model polysaccharide; and (ii) t-CAR and (iii) K-CAR, two polysaccharides more relevant for use in the food industry. 2. MATERIALS AND METHODS
Food-grade I.- and K-CAR were donated by Systems Bio Industries (Carentan, France). BSA, DS and n-tetradecane were obtained from Sigma Chemicals (St. Louis, MO). All other reagents were AnalaR grade. Buffer solutions ( 5 mM imidazole) were prepared with double-distilled water. Polysaccharide powder was dispersed in buffer and stirred continuously for 30 minutes at 70 "C. Protein (4.6 wt% for emulsion preparation, 0.25 wt% for surface tension samples) was dissolved in buffer and the pH adjusted with HCl or NaOH.
Food Science: Presentations
215
High-pressure processing was carried out using a Stansted Mark I1 Enhanced ‘Mini Food Lab’ (Stansted Fluid Power, Essex). The protein solution was vacuum sealed in polythenehylon bags and immersed in the pressure liquid. The temperature was controlled not to overshoot 30 “C and to keep to 30 f 2 “C during the 30 minutes pressure dwell. Oil-in-water emulsions (45 vol% oil) were prepared with a laboratory-scale jet homogenizer. Polysaccharide solutions were added at various concentrations to give final emulsions of 40 vol% oil for rheological experiments or 20 vol% for the particle sizing. The apparent average particle size d3** was determined using a Malvern Mastersizer S2.01. Particle electrophoretic measurements were recorded with a Malvern Zetasizer 4 with a single droplet of emulsion diluted in polysaccharide solution before measurement. The complex shear modulus G* of emulsions at 1 Hz and 30 “C was determined using a controlled stress Bohlin CS-50 rheometer with concentric cylindrical cell. The maximum strain of 0.5 % was within the linear viscoelastic regime. The solutions for surface tension measurement contained 10” wt% BSA f 4 x 10” wt% t-CAR. Time-dependent surface tension at 25 “C was monitored by the static Wilhelmy plate method using a Kriiss digital tensiometer.
3. RESULTS AND DISCUSSION To identify the presence of any protein-polysaccharide interaction between BSA and t-CAR in solution we carried out time-dependent surface tension measurements on the mixed biopolymer solutions. Figure 1 shows the adsorption curve for the 10” wt% BSA solution at pH 6. Pressure-treatment of BSA at 500 MPa leads to a decrease in surface tension y probably due to the unfolded state of the protein exposing more hydrophobic surface active residues. At the chosen conditions of low ionic strength and pH 6, the behaviour of y for the 1:4 BSA + t-CAR mixture exhibits a ‘shoulder’ at an adsorption time of 100 minutes and a higher equilibrium value. This is indicative of a macromolecular complex which we believe to be electrostatic in nature [3]. When the protein is pressure-treated prior to addition of 1-CAR there is also seen to be complexing occurring (see Figure 1). Having established the presence of an attractive BSA-I-CAR interaction in solution, we now turn to the oil-in-water emulsions to see whether there is evidence of complexing at the interface also. The emulsions were prepared with BSA as sole emulsifier and t-CAR was added to the aqueous phase after emulsification.
-
High Pressure Food Science, Bioscience and Chemistry
216
-72 'E Z
E 67
W
K
cn c
8lu v,
62
57 52
I 100
0
200
300
400
500
Adsorption Time (min)
Figure 1 Time-dependent surface tensions of biopolymer solutions (5 mM; PH 6 , oc): ~ o, I O - ~wt% BSA; m, 1o - ~wt% BSA + 4 x I o - ~wt% I-CAR; A, lo" wt% BSA treated at 600 MPa; A, wt% BSA treated at 600 MPa + 4 x wt% I-CAR
Table 1 Efect of high-pressure treatment of protein before emulsijkation on electrophoretic mobility p at 25 "C of BSA-coated emulsion droplets in pH ti solutions of I-CARof concentration c (relative to ,uoat c = 0)
c(wt%)
I
OMPa
200MPa 300MPa 400MPa 500MPa 700MPa
0
1.o
1.o
1.o
1.o
1.o
1.o
1o4
1.35
1.18
1.19
1.24
1.06
1.03
1o - ~
1.65
1.27
1.33
1.23
1.31
1.28
10-2
1.70
1.41
1.57
1.55
1.75
1.37
Food Science: Presentations
217
Table 1 summarises the electrophoretic mobilities p of emulsion particles, expressed as values relative to the mobility p,, in the absence of polysaccharide, with increasing I-CAR concentration c and for emulsions prepared with BSA treated at pressures 0-700 ma. Interfacial complexation is indicated by the substantial increase in p/p0 with increasing c. The lower values of p/po for the emulsions prepared with pressure-treated BSA may result from a weaker complex or a different macromolecular conformation in a complex carrying a lower charge.
140 120 h
m
100
e
80 60 40
20 0 0
0.05
0.1
0.15
0.2
0.25
c (wt%)
Figure 2 Eflect of added t-CAR concentration c on complex shear modulus G* at I Hz and 30 "Cfor BSA-stabilized emulsions (40 vol% oil, 2.7 wt% protein, 5 mM, pH 6). High pressure treatment: H, 0 MPa; A, 200 MPa; 0, 500 MPa.
Protein-polysaccharide interactions in emulsions can affect the rheology and stability of the system by inducing droplet flocculation. Small-deformation shear rheology measurements were carried out on concentrated BSA-stabilized emulsions containing t-CAR, and in Figure 2 the complex shear modulus G* at 1 Hz is plotted versus the t-CAR concentration c. For the emulsions prepared with native BSA, G* exhibits a maximum at 0.04 wtYo 1-CAR followed by a decrease and levelling off towards higher concentrations. This behaviour is consistent with a bridging flocculation mechanism. The anionic polysaccharide links protein-covered oil droplets by interacting with the BSA on their surface leading to a gel-like structure in the emulsion at 0.04 wt% t-CAR. Increasing the polysaccharide concentration further results in a restabilization of the
218
High Pressure Food Science, Bioscience and Chemistry
emulsion droplets as a consequence of complete coverage of the droplets with carrageenan. When BSA was high-pressure treated prior to emulsification no significant change in emulsion rheology was noticed up to pressures of 400 MPa. But at pressures 2 500 MPa the maximum in G* disappears and there is a slight increase compared with the reference system at c 2 0.15 wt%. The unfolding/aggregation of protein caused by the high-pressure processing changes the BSA-I-CAR interaction at the droplet interface. The bridging mechanism at low polysaccharide concentration is not apparent any more. At pH 7 we found a similar effect of BSA high-pressure treatment (2 400 MPa) on the rheology of concentrated emulsions containing dextran sulphate [2]. The increase in G* at the higher polysaccharide concentrations is possibly due in part to an increase in bulk viscosity caused by unbound partially denaturedaggregated BSA interacting with polysaccharide [2].
20 T h
8-r
15
5 7 10 N
5 0 0.0001
0.01 GCAR
1
(M%)
Figure 3 Efect of added (a) t-CAR or (b) K-CAR at concentration c on the apparent average droplet diameter dJ2* of BSA-stabilized emulsions (20 vol% oil, I . 7 wt% protein, 5 mM, pH 6) stored for 9 days at 25 "C.Highpressure treatment: 0,0 MPa; 500 MPa.
.,
We have also determined the apparent particle size d32* for 20 vol% oil-inwater emulsions. Figure 3(a) shows d32* versus t-CAR concentration. At c = 0.005 wt% there is a sharp increase in d32* to values above 10 pm followed by a decrease back again to the original particle sizes at c 2 0.1 wt%. This behaviour is in good agreement with the rheology of the concentrated emulsions described above. When the emulsion consists of a network of polymer-bridged oil droplets, it will tend to disperse into large flocs in the Mastersizer waterbath
Food Science: Presentations
219
(large d3**).But with restabilization, due to complete carrageenan coverage, taking place at the higher concentrations, the apparent average droplet size decreases again. The same emulsions prepared with 500 MPa treated BSA show a shift in the d32* maximum towards a higher carrageenan concentration. The interfacial BSA-t-CAR interaction has presumably been changed so that more carrageenan is necessary to induce flocculation; this is may be due to more polysaccharide attaching to unfoldedaggregated BSA in the bulk phase before bridging can occur. A similar type of behaviour is observed when t-CAR is replaced by K-CAR, as can be seen in Figure 3(b). Because of its higher molecular weight and lower charge, K-CAR is a less effective bridging polymer (flocculation occurs at a higher concentration, and the maximum value of d32* is only ca. 6 pm). Highpressure treatment of BSA here again causes a shift of the flocculation peak towards higher added polysaccharide concentration.
Acknowledgement This research was supported in part by a ROPA Award to E.D. (for K.P.) fkom the Biotechnology and Biological Sciences Research Council (UK).
4.REFERENCES 1 . E. Dickinson and D. J. McClements, ‘Advances in Food Colloids’, Blackie, Glasgow, 1995, chap. 3. 2. E. Dickinson and K.Pawlowsky, J. Agric. Food Chem., 1996,44,2992. 3. E. Dickinson and K. Pawlowsky, J. Agric. Food Chem., 1997,45, in press. 4. V. B. Galazka, I. G. Sumner and D. A. Ledward, Food Chem.,1996,57, 393.
Formation and Syneresis of Rennet-set Gels Prepared From High Pressure Treated Milk D.E.Johnstona.h*,R.J.Murphyb,J.A.Rutherford and C.A.McElhonea Department of Food Science, The Queen’s University, Newforge Lane, Belfast BT9 5PX, UK. Food Science Division, Department of Agriculture for Northern Ireland, Newforge Lane, Belfast BT9 5PX, UK. a
1. INTRODUCTION Cheese is a major dairy product found throughout the world. Therefore if high pressure treatment of milk for cheesemaking could provide a unique benefit, by enhancing the product, the process or by providing improved microbiological control, there would be a great opportunity to commercially exploit such a discovety. The transformation of milk into cheese involves several steps which are interdependent in terms of the rate and extent of the chemical and microbiological changes involved. If the final product is not to differ unacceptably from that produced by traditional methods, it is necessary to establish how far any innovation alters the rate or extent of these steps and the scope which may exist for controlling the alterations. 2. MATERIALS AND METHODS
2.1 Milks: Mdk was obtained firesh on a daily basis from a local dairy. Pastewised whole milk had a solids-not-fat (s.n.f.) content of 8.68% and was standardised to 3.60% fat. Pasteurised homogenised whole milk had a s.n.f. of 8.66% and was standardised to 3.55% fat. The homogenisation was carried out in
two stages, 3.4 MPa first stage and 13.8 h4Pa second stage.
Food Science: Presentations
22 1
2.2 Rennet: Commercial calf rennet was obtained from Chr. Hansens lab Ltd. (Reading, UK). 2.3 Hi& Pressure Treatment: Milks were pressure treated at either 200 MPa, 400
MPa, or 600 MPa for 10 minutes or 30 minutes using high pressure apparatus designed and manufactured by the Department of Mechanical and Manufacturing Engineering of the University (Crossland er al.,1971). 2.4 Rennet Coarmlation Time(RCT): Milk samples were adjusted to pH 6.45 by dropwise addition of lactic acid solution (30% w/v) and equilibrated at 29°C for 30 minutes before the addition of rennet. Quadruplicate observations of RCT
were made using the method of Berridge (1952). 2.5 Gel Formation and Svneresis: Replicate milk samples were adjusted to pH
6.45 by dropwise addition of lactic acid solution. Aliquots (25ml) were equilibrated at 29°C for 30 minutes before addition of rennet and the sample allowed to stand at 29°C for 1h for gel formation. The curds were cut into cubes using the method of Leliewe (1977) and transferred to a waterbath at 39°C. After a further 2h curds and whey were separated by a “tip and drain” method (Johnston and Murphy, 1992). 2.6 Fat Andvsis: Fat content of the curd was determined by the Werner-Schmid method. 2.7 Drv Matter(d.m.) Analvsis: Dry matter of the curd was determined by oven drylng at 102°C. 2.8 Statistical Treatment: The experiment was replicated four times and the results were examined using analysis of variance (ANOVA).
High Pressure Food Science, Bioscience and Chemistry
222
3. RESULTS HOLTER PLOT FOR HENNET COAtiLLATION OF
%>NET COhCt'WTIOh T I M E (R.C.T.) OF PRL!%C'RETREATED W L K S 0
PRESSURE TRFATED U U K S
1 O I . W
-Yd.wM
0 -1O.II.HIYH -WII.HWW
P04 0
:
1
:
2 -1
:
3
:
:
4 5 Canr (%uvv)
:
6
:
1
The coagulation times of the renneted milk samples are shown in Figure la. There was a significant effect of milk type, homogenised milks showing shorter times than non-homogenised. There was also a significant interaction between pressure and treatment time. R.C.T.of samples treated at 200 MPa decreased greatly compared to the control but 400 and 600 MPa treated samples displayed results successively closer to the control. Samples treated for 10 or 30 minutes had
R.C.T.sthat were not significantly different from each other at 200 MPa but at 400 and 600 MFa the differences between treatment times grew increasingly
greater. The effects of rennet concentration were investigated using a reciprocal plot to divide the R.C.T.into its primary, enzymatic phase and secondary, ionic phase (Figure lb). Milks treated at 200 MPa had both the primary and secondary phases significantly shortened but treatment at 600 MPa gave primary and secondary phases which were not significantly different fiom the control. The loss of whey from the cut curds prepared from the pressure treated milks is
shown in Figure 2a.
223
Food Science: Presentations
DRY MlTElt CONTENT OF RENNET CURD PREPARED FROM PRESSURE TREATED MILKS
W l i n RELEASED(Zh/J9DC) FROM RENWT 1NDITFDT.EI.S PREPmEIJ FROM P R E S S l a E TRWTED WlLKS
FAT CONTENT OF RENNET CLaO PREPARED FROM PRESSURE TREATED MILKS 0 .10dmWWM
.%-
FAT IN DRY MATTEROF RENNET CURD PREPARED FROM PRESSURE TREATED MILKS 387
0 - Y e w 0
rn
P
- N h W .Y*BlM
rn -N-BlI(
1
L
3u
200
mo
Ram W .)
28-
=
:
Colhpl
FlZIlrern)
ra
600
Syneresis was increased as a result of 200 MPa treatment but at 400 and 600 MPa the volume of whey released successively decreased. Homogenised milks released greater volumes of whey than non-homogenised and there was also a signtficant interaction between pressure and treatment time. The dry matter (d.m.) content, fat content and fat in d.m. of the curds are presented in Figure 2b, 2c and
Zd, respectively. The d.m. of the curds increased significantly as a result of treating the milk at 200 MPa but at the higher treatment pressures the effect was successively decreased. The interaction between milk type and pressure was approaching significance (p-O.52). Fat content of the curds also increased
224
High Pressure Food Science, Bioscience and Chemistry
significantly as a result of pressure treatment of the milk at 200 MPa and at higher treatment pressures the effect was successively decreased. The effects of pressure and treatment time were both significant. The fat in d.m. results showed a significant difference between the homogenised and non-homogenised control samples and a similar decreasing trend with increasing treatment pressure. Similarly treated homogenised and non-homogenised pressurised samples were not significantly different fiom each other. 4. DISCUSSION
It is clear from the results that there are two opposing effects present. A major change in all variables occurs between the controls and those samples treated at 200 MPa followed by a progressive recovery at 400 and 600 MPa. Micelle fiagmentation and transfer of micellar calcium and phosphorous to the serum phase take place very rapidly at 200 MPa (Schrader and Buchheim, 1997). On the other hand, the denaturation of whey protein requires pressures above 200 MPa (Johnston el u2.,1992).It is likely that fiagmentation and associated changes are responsible for the shortening of the primary and secondary phases of the coagulation. The increased exposure of hydrophobic regions of the protein due to micelle fiagmentation will alter the balance of attractive and repulsive forces between fragments, lowering the critical fiaction of K-casein hydrolysis required before the secondary phase can begm and the increased number of casein fragments and of hydrophobic sites would both be consistent with a faster
secondary phase. The association of denatured whey protein onto the micelles would be expected to hinder K-casein hydrolysis and screen the attractive forces to a varying extent.
Food Science: Presentations
225
From a manufacturing viewpoint, it would appear possible to achieve a curd of similar composition to a traditional process by suitable choice of pressurisation time and pressure in the range 400-600 MPa. Pressure treatment at 200 MPa results in curd compositional changes which may gve rise to unacceptable alteration to the final product. Nevertheless, this may provide the opportunity to develop a new product variation. The difference in fat globule size between homogenised and non-homogenised milks may provide some indirect evidence of altered gel structure. Retention of fat in the curd could potentially be increased by two mechanisms, improved physical entrapment due to increased numbers of network strands in the gel or by greater attractive interactions between the globules and the hydrophobic regions on the network strands. With non-pressurised milks, curd fat in d.m. is significantly lower for homogenised samples. This would suggest that the smaller fat globules can escape between the network strands. By contrast, the curd fat in d.m. of pressurised samples of homogenised milk are not sigmficantly different fiom the corresponding non-homogenised samples. This observation would be consistent with both a finer mesh network structure entrapping the smaller fat globules more efficiently and increased attractive interactions. A finer mesh network structure has been observed for acid-set gels prepared fiom pressure treated tnik (Johnston ef al., 1993) but additional direct evidence would be desirable to support this suggestion for rennet-set gels. 5 REFERENCES
Bemdge,N.J. (1952). An improved method of observing the clotting of milk containing rennin.Journal of Dairy Research, 19,328-329.
226
High Pressure Food Science, Bioscience and Chemistry
Crossland,B., Apew,R.H., Birks,A.W ., Ludlow,C.G. and Logan,J.G. ( I 97 1 ) Proceedings of the 27th Fluid Power Convention, ppl17-135, A.S.M.E., Chicago. Johnston,D.E. and Murphy,R.J. (1992). Effects of some calcium-chelating agents on the physical properties of acid-set mik gels. Journal of Dairy Research, 59, 197-208.
Johnston,D.E., Austin,B.A. and Murphy,R.J. (1992). Effects of high hydrostatic pressure on milk. Milchwissenschaji,47,760-763.
Johnston,D.E., Austin,B.A. and Murphy,R.J. (1993). Properties of acid-set gels prepared fiom high pressure treated skim milk.Milchwissenschaji,48,206-209.
Lelievre,J. (1977). Rig~d~ty modulus as a factor influencing the syneresis of renneted milk gels. Journal of Daily Research, 44,6 1 1-614.
Schrader,K. and Buchheim,W.( 1997). High pressure effects on mineral equilibria and protein interactions in milk. In “High Pressure Research in the Biosciences
and Biotechnology”, Ed. K.Heremans, Leuven University Press, Leuven, Belgium, pp 4 1 1-4 14.
Plenary Lecture
THE POTENTIAL AND IMPACT OF HIGH PRESSURE AS UNIT OPERATION FOR FOOD PROCESSING D. Knorr, V. Heinz, 0. Schluter and M. Zenker Department of Food Biotechnology and Food Process Engineering Berlin University of Technology, Konigin Luke Str.22, D- 14195 Berlin Tel:+49 30 314 71250 Fax:+49 30 832 7663
E-mail:[email protected] 1. INTRODUCTION Substantial progress has been achieved since the revival of high pressure research and development in the areas of food science and food technology and food biotechnology (Balny et a1.1992, Hayashi and Balny 1996, Heremans 1997, Ledward et al. 1995). Significant advancements have been made towards the understanding of mechanisms involved in the application of high pressure to food systems especially regarding microorganisms and proteins and kinetic data are being accumulated required to develop and monitor products and processes as well as to provide a base for regulatory approval of high pressure based processes and products. In addition, numerous publications reporting on empirical findings were accumulated. However, limited information has been provided on the impact of high pressure as a unit operation for food processing. Key advantages of high hydrostatic pressure treatment regarding food processing operations include independence of size and geometry of products, low temperature application, the potential for quality retention, modification of functionality parameters, reduction of microbial and enzyme activities and waste free processing. Major engineering issues that need to receive attention center around compressibility of water, adiabatic heating, increased reaction rates, pH changes during pressure treatment, pressure induced phase transitions and change of transition conditions as well as permeabilization of biological membranes. Unit operations based on high pressure treatment that seem feasible for food processing are high pressure blanching prior to freezing, drying or frying of foods, membrane permeabilization to aid diffusion and extraction operations, modification of biopolymers such as polysaccharides and proteins, reduction of
228
High Pressure Food Science, Bioscience and Chemistry
microbial (vegetative and spore) and enzyme activities or phase transition improvement such as pressure assisted freezing or pressure thawing. Since most of the information provided on pressure effects has been collected immediately after pressure treatment, results of shelf life studies are also required to identify quality, safety and functionality changes of treated products during storage. Some of the issues indicated above will be highlighted and exemplified by providing data on the impact of adiabatic heating, on pressure shift freezing and on shelf life changes of pressure treated plant products. 2. ADIABATIC HEATING OF REAL FOGD SYSTEMS
As an example of process considerations required during high hydrostatic pressure, processing is the increase in temperature during pressure treatment (and the connected dissipation of heat within and through the pressure vessel). A comparison of adiabatic heating of water at different pressures and various initial temperatures and adiabatic heating of a real food system containig 27% fat clearly demonstrates the impact of food composition and physico-chemical properties of the components (Fig.1). While water with an initial temperature of 40°C reaches a maximum temperature of approx. 57OC, the fatlproteidwater reaches levels of almost 65°C.
8
3
0
100
200
300
400
,
.
500
600
0
100
200
300
400 500 600
Pressure [MPa]
Figure 1 Adiabatic heating of water and a real food systems at different pressures and various initial temperatures.
229
Food Science: Presentations
Keeping the maximum product temperature during pressure treatment at 30°C the pressure, time dependence of the inactivation of Listeria innocua DSM 20649 and E. coli DH 5a could be demonstrated (Fig.%)and the three state model of microbial inactivation (Heinz and Knorr 1996) could be confirmed with two other microoorganisms. E.coli
0
2000
~
4000
~
6000
5
a
8000
Treatment Time [s]
Listeria innocua DSM 20649
0
2000
4000
0000
8000
Treatment Time [s]
Figure 2 Inactivation kinetics of Listeria innocua DSM 20649 and E. coli DH 5a in a real food system (fat 27%, protein 15%, water 55%) at a maximum product temperature of 30°C. 3. PRESSURE ASSISTED FREEZING AND THAWING OF FOOD SYSTEMS A combined phase diagram of water and protein (Chymotrypsinogen; Data from Hawley, 1978) exemplifies the possibilities for phase transitions of water, ice and proteins under pressure (Fig.3) and also illustrates the need for further research on phase diagrams of food components especially in the low temperature regions. An illustration of the numerous possibilities of of phase transitions of water and
water containing foods is provided in Figure 4.
High Pressure Food Science, Bioscience and Chemistry
230
Phase Transitions of Protein and Water
o I o 800 o "
-20
40
20
0
Temperature ["C] Figure 3 Phase diagram of protein (Chymotrypsinogen; Data: Hawley, 1978) and water
Phase Transitions ~~
0
100
200
300
400
500
Pressure [MPa]
Figure 4 Examples of possible phase transitions of water or water containing foods
23 1
Food Science: Presentarions
The regions of Figure 4 show: A,B,H.I: freezing under pressure (pressure build up, cooling to phase transition line, freezing, decompression) I,H,B,A: thawing under pressure (pressure build up to phase transition line, thawing, temperature increase, decompression) A,B,C,D,E: pressure shift freezing (pressure build up, cooling, sudden decompression) E,D,C,B,A: pressure induced thawing (pressure build up beyond phase transition line, temperature increase, decompression) A,B,C,D,G,F: freezing via Ice I11 (pressure build up, temperature decrease, freezing to Ice 111, decompression to Ice I) F,G,D,C,B,A: thawing via Ice I11 (pressure build up, recrystallization to Ice 111, thawing, temperature increase, decompression) A,B,C,K,Ice V or Ice VI: freezing at plus temperatures (pressure build up until phase transition line, freezing) A,B,C,D,C,B,A: storage under pressure at subzero temperatures without ice formation (pressure build up, cooling, holding, temperature increase, decompression Freezing of potato cylinders (diameter 28 mm, length 60 mm) under pressure (Fig.5) resulted in the expected decrease of the freezing point along the phase transition line and also revealed temperature differences between product surface (immersion medium temperature) and product center. With increasing pressure and lower temperature difference (between potato center and medium) distinct undercooling could be observed prior to initiation of crystallization. Pressure increase (due to volume increase) and constant temperature are useful indicators for the fi-eezinglcrystallizationprocess. 160
140
120
s
$
n
100
80 50
I l l
I
I I
A: Temperature Center
0
Time [rnin]
20
40
Time [min]
80
Time [min]
Figure 5 Freezing time of potato cylinders during high pressure treatment
232
High Pressure Food Science, Bioscience and Chemistry
4.SHELF LIFE BEHAVIOUR OF PRESSURE TREATED FOODS Pressure treatment of broccoli and peas and subsequent shelf life studies at 4°C and at -3OOC for 21 days and 6 months respectively were carried out using various quality indicators for product safety, quality and functionality (Koch , data). As one example the effects of pressure treatment and ~ O Kunpublished and subsequent,storage time and temperature dependent changes of peroxidase activity in broccoli is provided (Fig. 6). Activity of peroxidase, known to be one of the pressure stable enzymes, could be reduced to approx. 10% residual activity after pressure treatment at 5"C, 20 min, and 600 MPa during 14 days of storage at 4°C. However, only a reduction to 60% of the initial activity could be reached during 6 months of storage at -30°C. Another example selected is the degradation of chlorophyll of broccoli during pressure treatment and subsequent refrigerated or frozen storage (Fig. 7). Pressure treatment even at 50°C resulted in degradation comparable to changes during hot water blanching, while high pressure treatment under carbon dioxide atmosphere achieved high quality color products that could be maintained during frozen storage. During refrigerated storage the samples treated at 5 and 20°C experienced a degradation to 80%, while the hot water blanched became unacceptable and the 50°C as well as the C02 atmosphere samples reached degradation levels of over 30%.
5. CONCLUSIONS Space limitations allow only glimpses of the potential and impact of high pressure as a unit operation for preservation and modification of foods and food constituents. It is hoped that the examples provided can help to demonstrate the numerous possibilities for process and product development and can stimulate further research and developement activities in the exciting area of food application of high hydrostatic pressure.
6. ACKNOWLEDGEMENTS Parts of this work were supported by the German FEI (Forschungskreis der Emahrungsindustrie e.V.,Bonn), the AiF and the Ministery of Economics (Project No.: 9918) and by the German Research Foundation (DFG h260-6/1).
233
Food Science: Presentations
Broccoli storage 4°C: Peroxidase UHP 600 MPa, 20 min; blanching 2 min
* raw U blanch
*
UHPL' t UHPM'
*
UHPQ"
0
I0
6
IS
20
storage time [days]
Broccoli storage -3OOC: Peroxidase UHP 600 MPa, 20 min; blanching 2 min
6 * ~
raw
8 blanch 4
UHPL"
's E
f- ______.__._...-_-UHP20" *
I
UHP50'
-t ,
0
1
2
;
3 4 storage time [months]
'
I
6
'
6
Figure 6 Change in peroxidase activity in broccoli after pressure treatment or hot water blanching and subsequent storage at 4°C or -30°C
High Pressure Food Science, Bioscience and Chemistry
234
Chlorophyll degradation in broccoli during storage at 4OC after treatment
blanch UHPB" UHP20' UHP+C02
0
6
16
10
20
26
storage time at 4°C [days] UHP 600 MPa. 20 min
[blanch. 2 min. 1 0 0 ' ~
I
Chlorophyll degradation in broccoli during frozen storage after treatment
0 Iblanch: 2 min. 100°C
1
2 3 4 6 storage time at -30°C [months] U H P 600 MPa. 20 min
6
I
Figure 7 Clorophyll degradation in broccoli after pressure treatment or hot water blanching and subsequent storage at 4°C or -3OOC
Food Science: Presentations
235
7. REFERENCES Balny, C.,Hayashi, R.,Heremans, K. and Masson, P.1992. High Pressure and Biotechnology, John Libbey Eurotext Ltd., London. Hawley, S.A.1978. High pressure techniques. Meth. Enzymol. 49, 14-24. Hayashi, R. and Bzlny, C. 1996. High Pressure Bioscience and Biotechnology, Elsevier, Amsterdam. Heinz, V. und Knorr, D.1996. High pressure inactivation kinetics of Bacillus subtilis cells by a three-state-model considering distributed resistance mechanisms, Food Biotechnol. 10 (2), 149-161. Heremans, K.1997. High Pressure Research in the Biosciences and Biotechnology, Leuven University Press, Leuven. Ledward, D.A., Johnston, D.E., Earnshaw, R.G. and Hasting, A.P.M. 1995, High Pressure Processing of Foods, Nottingham University Press, Nottingham.
New Development of High Pressure Equipment Reduces Processing Cost A. Triiff
ABB Pressure Systems AB S-721 66 Vaster& Sweden
The use of high isostatic pressure, sometimes in combination with moderate heat, for the treatment of food results in a safe product of high quality especially with regard to freshness, texture and colour. New types of products can be created. The selection of the raw material, the formulation, the packaging and the distribution form are important factors to consider when using high pressure for treatment of food.
The press: The key element for high pressure treatment is the press. The pressures required for proper treatment of food products lie between 400 MPa and 800 MPa. Such pressures make extreme demands on the materials used in the press and in the pumping system. 400 MPa is too low for most applications considering there are variations in the raw material and microbiological flora. On the other hand, 800 MPa is often considered too costly for the treatment of food products. ABB has
Food Science: Presentations
237
therefore decided to focus development on equipment for 600 MPa, which is sufficiently high for treatment of all types of acidic foods and at a cost many products can afford.
Steel: When the pressure vessel is pressurised and contains these high pressures, the inner wall of the vessel will experience a stress. The stress level is so high that special designs and steels must be used to contain the pressure. The pressure vessel and pumping system are subjected to pressure/stress cycling each tim:a batch of food is processed. Such cycling will eventually fatigue the material and a probable life length can be calculated for the design. The pressure vessel and components of the pumping system can be prestressed to enhance their capacity to withstand the high stresses and prolong the life expectancy. ABB uses the wire winding technique as large pressure vessels can be given full prestress in a controlled and monitored way. The weight of a wire wound press is around 50 % of that of a press made with other prestressing techniques. For smaller parts, where these features are not critical, ABB also uses autofrettage or shrink fitting.
238
High Pressure Food Science, Bioscience and Chemistry
New ABB press concept: After extensive research in materials, designs and fatigue properties, the new
ABB QUINTUS@wire wound press designed for 600 MPa with an external, high pressure intensifier is now launched. The use of an external prestressed pump instead of internal pressure build up inside the press saves 50 % on the press weight. The energy consumption is reduced in the same order as less pressure medium need be compressed. The new ABB high pressure intensifier (patent pending) is a prestressed, through flow pump eliminating the weak points of a traditional pumping system and resulting in the same high uptime as a press with internal pressure build up. These components in the system are subjected to the same stress levels as the pressure vessel, but many times over in the course of a single press cycle since the pressure intensifier has to complete several strokes to build up full pressure in
the pressure vessel. The material fatigue problem is enhanced both due to the m y load cycles and due to stress concentrations in parts like valves and connection elements. The driving system and the water containment and filtration unit comprising
tanks,cooler, heater, filters, feed pumps, etc. are of conventional type.
Food Science: Presentations
239
The press can make 4 to 6 press cycles per hour corresponding to a capacity of 300 to 3000 kg per hour.
Material handling: The vertical arrangement of the press fits nicely into the desired process chain with preparation equipment and material handling. The press can be reloaded in 1 minute with automatic manipulators. The pressure medium remains in the press. The press’s pumping system is placed under the floor to reduce noise, facilitate cleaning and increase personnel safety.
High Pressure Food Science, Bioscience and Chemistry
240
cost: The cost of processing food with high pressure depends on several factors. The
& g key issue is to optimise the l
process chain including preparation, handling,
processing parameters, volumes and seasonal changes in demand or supply and reliability of the processing equipment. For the processing parameters it is essential to find the right combination of pressure, time and temperature, both to achieve the desired product quality and allow for variations in the raw material and to keep the cost to a minimum.
The major portion of the HPP cost based on an up-time of 90 % is the capital cost, which then amounts to around 75%. The remaining 25% is manpower, occupancy, consumables and maintenance. If the up-time falls due to failures in the pressurising system productivity goes down quickly and maintenance increases. Key issues are therefore the up-time and the utilisation of the equipment. One shift operation is 40 % more costly than two shift operation. 50 % fill density is twice as costly as 85 %, e.g. a round bottle with a long neck will fill 50 % of a press while a six-sided bottle with a short neck will fill 75 %.
Food Science: Presentations
24 1
The cost per processed v o l m is reduced with larger presses having higher capacity. The cost per kilo of processed product is 35 ?6 lower with a 400 L press than with a 150 L press. Two 400 L presses with a common hydraulic system will have the s a m production capacity as one lo00 L press. This way, the investment can be spread out in time and the capacity growth more adapted to the demand. If the micro biological results are the same, processing at 400 MPa with an 8
minute hold time will cost the s a m as using 600 MPa and a 3 minute hold time. However the 600 MPa unit has a higher capacity and a wider use for future expansion of products. Depending on fill density, processing parmters and utilisation, the cost of high pressure treatment will be between 10 to 20 pence per kg The investment can range from €400,000 to f 1,500,000.
SUmmary:
ABB’s new low-weight press with an external high pressure intensifier of proprietary design and fast and reliable material handling can handle volumes ranging from 2,000 tons up to 10,000 tons per year at a cost of between 10 to 20 pence per kilo.
Acidification of Milk by Glucono-6lactone under High Pressure
M. Schwertfeger and W. Buchheim Federal Dairy Research Centre, POB 6069,24121 Kiel, Germany Tel.: +49-43 1-6092270, Fax: +49-43 1-6092309, e-mail: buchheim@b&.de
Abstract The hydrolysis of glucono-8-lactone (GDL) is widely used for acidification processes. This reaction is well known for ambient pressure in various media.
In the present study the hydrolysis of GDL was followed in milk, in (protein-fi-ee) milk serum and in water at 0.5, 1 and 2 kbar. By measurement of pH after decompression an accelerated hydrolysis was observed. At an initial concentration of 100 mmoVl GDL, for example, the pH values reached 4.4 (milk), 3.3 (milk se-
rum)and 2.4 (water) after treatment at 2 kbar for 40 min (25"C), whereas the corresponding pH values at ambient pressure were 5.5,4.4 and 2.6. The acidification of milk by GDL under high pressure results in differently structured protein coagulates as compared to ambient pressure. Introduction Glucono-s-lactone (GDL) hydrolyses gradually to gluconic acid in aqueous solutions and is therefore widely used as an acidulant for various foods, e.g., for bakery and meat products, for tofu and for dairy products like cottage cheese. As compared to microbial acidification the use of GDL offers an improved control and reproducibility of manufacturing conditions.
Food Science: Presentations
243
The hydrolysis reaction is a complicated equilibrium among the &lactone, the ylactone and the gluconic acid (1). The kinetics and equilibria of this reaction in aqueous solutions (1-3) as well as the colloidal behaviour of milk during acidification by GDL (4) have been extensively studied but only at ambient pressure. Depending on the GDL concentration and the temperature applied, the final pH value is generally reached within approximately one hour. Since this reaction is involving changes in water structure, its sensitivity to pressure is to be expected. Taking into account the simultaneously occuning pressure-induced shifts in the mineral equilibrium of milk as well as the changes in the structural state of the milk proteins, a different acid coagulation behaviour of the proteins should occur under pressure. Experimental Glucono-ti-lactone was used as obtained from Jungbunzlauer (Ladenburg, Germany). The skim milk was prepared by reconstituting one part low-heat skim milk powder in ten parts (w/w) deionized water. Some drops of n-octanoI were added to remove air bubbles. The pH of the milk was 6.6. Protein-free milk serum (permeate) was obtained by ultrafiltering this skim milk (10 kD cut-off level). For high pressure treatment GDL was dissolved in water, permeate or milk. The
samples were processed at pressures of 0.5, 1 or 2 kbar for 5, 10, 20 or 40 minutes in small nalgene@tubes (volume 5 ml). The pressure treatment was performed with a temperature-controlled laboratory autoclave. After decompression the pH
of the samples was measured with a glass electrode. The reaction time was calculated from wetting the GDL until pH-measurement. The temperature was
25°C or 35"C, the initial concentration of GDL was 100 or 200 mmoV1. In the autoclave a constant pressure was reached within 5 seconds. The samples were kept at a somewhat lower temperature before pressurization and allowed to reach the reaction temperature as a result of adiabatic heating.
244
High Pressure Food Science, Bioscience and Chemistry
The hydrolysis at ambient pressure was studied by dissolving GDL in water (permeate or milk) at 25°C. The changes in pH were measured continuously.
Results Hvdrolvsis in water: At standard conditions (100 mmol/l, 25"C, 1 bar) the pH drops to 2.5 after a reaction time of 60 minutes. At pressures of 1 and 2 kbar the hydrolysis is accelerated and the pH reaches lower values as compared to ambient pressure. Within a reaction time of 40 minutes the hydrolysis equilibrium was not yet reached (Fig. 1). Hydrolysis in uenneate: At ambient pressure and an initial concentration of 100
mmoVl at 25 "C the pH changes towards 4 within 60 minutes. At 2 kbar the pH reaches 3.3 after 40 minutes (Fig. 2). Hydrolysis in skim milk: The hydrolysis in skim milk is significantly different from that in water or permeate. At ambient pressure the shift in pH is relatively small from -to 5.3 after 60 minutes. At pressures of 0.5 and 1 kbar the change in pH at the beginning of the reaction is similar to the change in pH at 25°C and ambient pressure. The variation in pH is greater at higher pressures and after longer reaction times (Fig. 3). After 40 minutes at 2 kbar the pH drops to 4.4. When the pH reaches 5.3 the rate of pH-shift varies (Fig. 4). The samples with a pH lower than 5.3 showed distinct coagulation. The structure of these coagulates varied depending 011 experimental conditions (Table 1). Hydrolysis at different conditions: At higher initial GDL concentration (200 mmolfl) and higher temperature (35°C) the variation of pH during the hydrolysis in water is similar to the experiment at 2 kbar, 100 mmovl and 25°C (Fig. 5 ) . During acidification of milk the shift in pH is strongly depending on the experimental conditions. In these experiments the pH of the coagulated samples was also below 5.3.
245
Food Science: Presentations
Hydrolysisal ODL In milk nrum, T = 26%, c,= 1W mmolll 6.0 8.5 5.0
f 4.0 4.0
3.6 3.0
ig. 1 Hydrolysis of ODL in sklm milk, T = ZPC, c,
'ig. 3
* I00 mmoyl
I
'ig. 4
Fig. 5
246
High Pressure Food Science, Bioscience and Chemistry
Discussion
The hydrolysis of GDL in aqueous solutions is a complex reaction (6). It leads to a carboxylic acid which should dissociate more completely at high pressures than at ambient pressure. In spite of the ring-opening the activation volume should be negative due to the formation of ions. For comparison, the slightly different compound butyrolactone shows an activation volume AV* of -9.6 mVmol for hydrolysis (7). The comparably few measurements indicate that hydrolysis of GDL in water is accelerated similarly either by an increase in temperature of 10 K or by 1 kbar in pressure. In buffered media, such as permeate and milk the pH changes are smaller than in water. In these systems the liberated, strongly hydrated hydrogen ions should be bound partially so the decrease in volume during this reaction is smaller than in water. The hydrolysis reaction in milk is more complicated because of other pressure sensitive processes occurring. In addition to the complex formation of gluconic acid with calcium ions the equilibrium should be shifted towards the products and the reaction rate should increase. Both by acidification and by pressure treatment the colloidal calcium phosphate of the casein micelles is dissolved. The larger pH-changes in penneate as compared to milk are apparently caused by the absence of colloidal calcium phosphate and the buffering capacity of the proteins. Under high pressure the casein micelles disaggregate (5). During acidification the electrosteric stabilisation of the casein micelles disappears at pH 5.3 (4). At pressures above 1.5 kbar the hydration sphere around the protein molecules starts to becomes degraded. All these processes are determining the final coagulation behaviour of the casein fraction. A complete explanation of the formation of the different products is, however, not yet possible. For the discussion of the properties of the aggregates, the classical theories of - colloid chemistry, e.g. steric sta-
247
Food Science: Presentations
bilisation, depletion flocculation and the concept of “distance of closest approach” in the DLVO theory may be helpful.
Literature 1.
Sawyer DT, Bagger JB (1959) J. Am. Chem. SOC.81: 5302-5306
2.
Skou EM, Jacobsen T (1982) Acta Chemica Scand. A36:417-422
3.
Mitchell RE, Duke FR (1970) Ann. N. Y. Acad. Sci. 172: 129-138
4.
Banon S, Hardy J (1992) J. Dairy Sci. 75: 935-941
5.
Schrader K, Buchheim W, Mom CV (1997) Nahrung 41: 133-138
6.
Ohmiya K (1 989) Agric. Biol. Chem. 53: 1-7
7.
Asano T, le Noble WJ (1978) Chem. Rev. 78: 407-489
Table 1: Acid coagulation of skim milk by GDL.
P &bar)/ co (RXIIO~/~)/ T (“C)
time of pressure treatment (min) 5
10
20
40
0.5/100/25
n. c.
n. c.
n. c.
n. c.
11100/25
n. c.
n. c.
n. c.
c. 1
2/100/25
n. c.
c.2
c.3
c.4
n. c. no visible coagulation, c.1 coarse floccules, c.2 fine floccules, c.3 fine stranded coherent coagulate, c.4 compact, fine stranded sediment
Composition Changes of Strawberry Puree during High Pressure Pasteurisation
D. Brha, L. Istenesova, M. Voldiich and M. Cefovsky Department of Food Preservation and Meat Technology, Institute of Chemical Technology, Technicka 5 , Prague 6, CZ 166 28, Czech Republic
1 . INTRODUCTION The ultra high pressure pasteurisation in combination with aseptic filling systems seems to be one of the most promising applications of high pressure technology in fruit food processing [1,2]. The high pressure treatment for 5 10 min under 300 - 600 MPa at 20 microbial cells by 4
-
-
-
50 "C allows the reduction of vegetative
5 log cycles, however some enzymes especially
polyphenoloxidase in h i t juices are more pressure resistant and their inactivation needs additional approaches [l-31. The main advantage of high pressure pasteurisation is that products retain the flavour, colour and nutrition level of fiesh fruits [ 1,4]. However, high pressure processing affects chemical reactions in food material in both, positive and negative ways. Nonenzymatic browning (Maillard reaction) as well as hydrolysis of sucrose is inhibited by pressure.
Food Science: Presentations
249
Degradation of colour and slight changes of flavour due to the higher content of dissolved oxygen in products are mentioned as an example of negative pressure effects [I ,4]. The aim of presented paper is to evaluate the effect of high pressure treatment on selected chemical changes of strawberry puree (anthocyanins content, vitamin C content) and on inactivation of yeast Cryptococcus luurentii.
2. MATERIAL AND METHODS Material. Strawberries (origin unknown) were obtained from supermarket. The puree (pH
=
3.7, 7.5"Brix) was prepared by mixing (Ultra Turrax) for
2 minutes, one part of it was inoculated by yeast Cryptococcus laurentii and both (inoculated and uninoculated) sealed into polyethylene capsules (= 2 ml), treated by pressure and analysed. The inoculated puree was analysed for yeast growth, the uninoculated one for vitamin C and anthocyanins content. Pressure treatment. Samples were pressurised in hydrostatic pressure cell of about 50 ml capacity and maximum pressure of 1000 MPa under 100 to 700 MPa at room temperature for 10, 20 and 40 minutes. A mixture of mineral oils was used as a pressure transmitting medium, no contamination of samples was detected.
High Pressure Food Science, Bioscience and Chemistry
250
Yeast. It was isolated from h i t juice and identified as ('ryp/ococcu.v lurircntii. It is osmophilic, thermoresistant yeast, which causes cloud problems in
fruit purees and juices. Yeast was incubated in shaker in liquid medium (GTK(glucose, tryptone, yeast aitolysate) + 18 % of glycerol(v/v)) at 28°C for
24 hours than inoculated into sample. After pressurisation 1 ml of sample was moved on plate and covered by liquid medium than incubated at 28°C for 5 days. Heat treatment. The inpressurized uninoculated samples of puree were also heat treated (80°C, 20 min) to compare effect of heat and pressure on vitamin C and anthocyanins content. Determination of vitamin C and colour. Vitamin C content was analysed by HPLC at conditions: the column 250 x 8 tnm Ostion LGKS 0800 H+ cycle (WATREX Prague, CZ), mobile phase 1 mM €+$SO4in water,
RI detection, the
flow rate 0.5 m l h i n , room temperature. Colour was determined as total atitliocyaiiins by spectrophotonietric method at 520 nm. [ 5 ]
3. RESULTS
The rate of yeast ( 'r);p/ococctcslaztreniii inactivation in strawberry puree under 100 MPa, 200 MPa, 300 MPa and 500 MPa at room temperature is given in Fig. I . It is obvious that high pressure accelerates inactivation of yeasts. For these data the D-p degradation curve at room temperature was constructed
25 1
Food Science: Presentarions
(Fig. 2). The D-p degradation curve means the dependence of exposition time needed to reduce the concentration of yeast of one log cycle on pressure at constant temperature. For the mentioned yeast in strawberry puree at room temperature the curve can be described by the equation:
log D = 1.3112 - 0.0004 p
I11
Pressure sensitivity of Cryptococcus laurentii in model solution of glucose under 400 MPa at room temperature was also investigated (Fig. 3). Inactivation is strongly dependent on the osmotic pressure. A similar protection effect of osmotic pressure was observed as it is known for other changes (e.g. inactivation of microorganisms generally, enzymes etc.), inactivation of yeast for 20 min at above mentioned conditions in 5% glucose solution was practically complete. In contrast to the above observation the inactivation in 12% or 50% of glucose solutions was almost negligible. I,40 ,1,30 c I
E 1,20
Y
0
40
20 time (min)
-100
MPa,*200
MPa
1,lO
- 1,oo UI 0
0,90 0
200
400
600
pmssum (MPa)
Fig. 1 Effect of high pressure treatment on CIyptococcus laurentii yeast
Fig. 2 D-p degradation curve of Ctyptococcus laurentii yeast in
inactivation in strawberrypuree at room temperature
strawbeny puree at room temperature
252
High Pressure Food Science, Bioscience and Chemistry
1,00E+06 ,I ,00E+05 1,00E+04 1,00E+03 I,00E+02 '1,00E+01 1,00E+00 0
+5% 12% x- 50% 43-
5
10
15
20
time (min)
Fig. 3 Pressure inactivation of Cryptococcus laurentii in model solution of glucose (400 MPa, room temperature)
Effect of high pressure treatment on selected chemical changes of strawbeny puree (anthocyanins and vitamin C content) is given in Fig. 4 and Fig. 5 . figh pressure treatment caused a small decrease in absorption of puree at
520 nm and in vitamin C content only. The magnitude of used pressure had no 0,40
0,55 0,50
8
oS O 3 c 8 o,30
$ 5 0,25
0,45
3 0,40 0,35
'5
& 0,20 0,15
20
0
0
40
time (min)
-+300 MPa +500 MPa +700 MPa *80°C
20
40
time (min) -o- 300 MPa +500 MPa
+-700 MPa +80°C
Fig. 4 Effect of high pressure treatment of strawberry puree on colour
Fig. 5 Effect of high pressure treatment of strawbeny puree on vitamin C
(anthocyanins content)
content
significant influence on these changes. To compare the pressure and thermal treatment the fiesh unpressurized strawberry puree was also exposed to elevated
Food Science: Presentations
253
temperature. The decrease in colour and vitamin C at heat of pasteurisation (80°C for 20 min) was much more considerable.
4.CONCLUSIONS High pressure causes inactivation of Cryptococcus laurentii in strawberry puree. Inactivation is dependent on used conditions, especially on the pressure. The effect of increasing pressure is more significant at lower osmotic pressure. The D-p of Cryptococcus luurentii for strawberry puree (pH = 3.7, 7.5"Brix) at room temperature has the equation log D = 1.3112 - 0.0004 p. Pressure treatment
had almost no effect on colour and vitamin C content.
5 . REFERENCES 1. D. Knorr, High pressure effect on plant derived foods. in High Pressure Processing of
Foods, D.A. Leward, D.E. Johnston, R.G.Earnshaw, A.P.M. Hasting (eds.), Notingham University Press (1995) 123. 2. Y. Ifuku, Y. Takahashi, S. Yamasaki, International Markets, Fruit Processing, 3(1) (1993) 19. 3. Y. Takahashi, H. Ohta, H. Yonei, Y. Ifuku, Int. J. Food Sci. and Technol., 28 (1) (1993)
95. 4. J.C. Cheftel, Effect of high hydrostartic pressure on food constituents: an overview in
Proceedings of High Pressure and Biotechnology, C. Balny, R. Hayashi, K. Heremans, P. Masson (Eds.), Colloque INSEWJohn Libbey Eurotext (1992) 195. 5. Davidek, J. et al.: Laboratorni piiruEka analyzy potravin. SNTL Prague, 1981.
Food Science: Posters
EFFECT OF HIGH HYDROSTATIC PRESSURE IN A MODEL
MAYONNAISE E. Ponce, J. Saldo, M. Capellas, B. Guamis and R Pla. Tecnologia dels Aliments, C.e.R.T.A. Facultat de Veteriduia. Universitat Autbnoma de Barcelona. 08193 Bellaterra, Spain Tel. 34-3-5811446 Fax 34-3-5812006e-mail [email protected] SUMMARY The behaviour of a model mayonnaise @H 6.0) subjected to combined treatments of pressure (300 and 450 MPa) and temperature (2 and
2OOC) for 10 min was studied. After pressurization the mayonnaise was maintained for two months at different temperatures (4,20 and 37°C)in order to examine its evolution. Emulsion stability was evaluated as the distribution of particle size by means of Coulder Counter and Confocal Scanning Laser Microscopy. Results showed that pressure maintained the particle size dispersion and, therefore the stability of the pressurised emulsions.
After one month of conservation, no important differences were observed at 4 and 25OC in treated samples, but at 37°C the control and some of the pressurised mayonnaises were destabilised Initial pH of mayonnaise was not modified by the pressure treatments. After one month of storage final pH raised to 7.4 at 3TC,but less than one unit at 4 and 25°C. Rheological and textural parameters of mayonnaise w m studied by Mettler RM180 Rheomat and TA-XT2 Textural Analyser. Immediately after the pressurisation consistency increased three-fold respect to the control in samples pressurised at 450 MPa. Mayonnaise pressurised at 2°C was more consistent than at 2OOC. Hardness was always greater in samples pressurised at 450 MPa than at 300 MPa just after the treatment and under all conditions
of storage.
258
High Pressure Food Science, Bioscience and Chemistry
INTRODUCTION Mayonnaise is a semisolid dressing with a high phase volume of oil. It is formulated to give maximum stability against coalescence. When mayonnaise is prepared, the protein particles are found at the interface between oil droplets. Some authors have atready observed that high hydrostatic pressure improves the microbiological quality of foods (at low or ambient temperature), with lack or reduction of chemical additives (Farr, 1990; Mertens, 1993). At the same time, several changes have been ob-
served in lipids under pressure as crystallisation, increasing of the melting point of tryglicerides, modifications in the phospholipid bilayer of the cell membrane of micro-organisms ... (Cheftel, 1992). The purpose of this study
was to investigate the effect of high hydrostatic pressure on mesophilic bacteria, stability properties and microstructure of mayonnaise (PH6.0).
MATERIALS AND METHODS Liquid egg yolk was prepared from one day old eggs obtained fiom the hen farm of the Universitat Authoma de Barcelona. Yolks were separated from albumen using a household separator. The yolk was gently rolled on wet cheesecloth to remove adhering albumen and chalaza. Yolk membranes were punctured, and the yolk liquid was collected, pooled and stirred gently to provide a homogeneous mixture. Mayonnaise was prepared with 10 g of egg yolk and 100 ml corn oil in a kitchen Mouliiex 170 W mixer. pH was measured with a micro pH 2001 Crison. The mayonnaise was poured into sterile polyester bottles of
40 ml (Bibby Sterilin, UK), sealed with Teflon film, and allowed to stand at 4OC for 1 day, so that the emulsion could reach the equilibrium. High pressure (350 and 450 MPa) and temperature (20 and 2OC) combinations for 10 min were assayed in a discontinuous isostatic press from ACB
259
Food Science: Posters
(Nantes, France). After pressurisation, mayonnaise was maintained for two months at different temperatures (4, 25 and 37°C). Following parameters were evaluated in order to examine its evolution. Mayonnaise stiflhess was measured using the TA-XTZ Texture Analyser. Viscosity of samples was determined with a Mettler RM180 Rheomat. Emulsion stability was evaluated by distribution of particle size by means of Coulder Counter ZM and Confocal Scanning Laser Microscopy. Mesophilic bacteria were enumerated in Plate Count Agar, after incubation, at 32°C / 48h. Sampling was done one day after pressurisation and one month later. All treatments were made twice. Analysis of variance was performed using the General Linear Models procedure of Statistical Analysis System (S.A.S, NC, USA).
RESULTS AND DISCUSSION High pressure processing produces changes in macromolecular structure in proteins, as formation of hydrogen bonds, separation of ion pairs, rupture of hydrophobic interactions, etc. to get a volume reduction when the pressure is raised (Cheftel, 1992). Not all changes are reversible when the sample is returned to atmospheric pressure. Those modifications after the pressurization could produce changes in pH. Our results show that in day 0 no important changes of pH are produced by pressure and probably during storage the increment of pH is due to microbial growth, more dramatic at room temperature and 37°C than at 4°C (Table 1). Mesophilic bacteria are not detected (< 30 ufc/g) in the control. Counts after 24 h of pressurisation are totally undetectable (0 ufc/g) for all combinations, but they increase with the storage due probably to the presence of injured cells (Table 2). The high percentage of fat (78%) and the protein macromolecules can help micro-organisms to resist the high pressure, as have already indicated some authors. (Styles et al., 1991; Gervilla et al., 1997; Ponce et al., in press).
6.4 1
6.06
450 MPa 10' 2°C
5.18
ND - Not detected
I
0.49
0.04
6.10
1 month 25°C
I
0.45
3.55
Std
-
Control
Day 0 1 month4"C
-~
Logcfdg ND
11month37"C
l
6.44
6.10
450 MPa 10' 20°C
I
I
-
6.53
5.61
1.62
0.08
0.04
1.02
10' 20°C LogcWg Std ND -
I
I
7.62
5.71
1.52
ND
0.18
0.01
0.73
10' 2°C Logcfidg Std
300 MPa
-
6.62
6.44
6.1 1
300 MPa lo' 2°C
Table 2. Mesophilic bacterial counts of mayonnaise.
6.83
6.44
6.08
300 MPa 10' 20°C
I
I
6.40
5.46
ND
0.09
0.07
-
I
I
~
~
6.21
5.42
3.14
0.07
0.12
0.97
10' 2°C LogcWg Std ND -
450MPa
7.13
7.20
7.23
6.10
7.42
1 month at 37°C
10' 20°C Logcfdg Std ND -
6.89
6.71
6.00
6.35
6.03
Control
1 month at 25°C
1 month at 4°C
Day 0
0
m N
Food Science: Posters
26 1
High hydrostatic pressure at 350 MPa (20 and 2°C) produces a slight increase of stifhess respect to the control, whereas samples treated at 450 MPa show a consistency between 2 and 2.5 times higher. Samples treated at 2OC display more rigidity than at 25°C. Storage at 25' and 37°C produce an important reduction of this rigidity, which was maintained at 4°C. However, at all storage temperatures and times tested, the stiffiess of the treated samples is always greater than control (data not shown).%
enhancement
of mayonnaise consistency is possibly due to the gelation of egg yolk pro-
teins. Stifhess reduction can be explained by coalescence among oil drop lets and microorganism action in protein matrix. The smallest possible avemge particle size of oil droplets together with a low dispersion in its distribution are necessary for an optimum mayonnaise stability. Our results indicate that these conditions can be achieved in treated samples stored at 4"C, without apparent differences with the control in same conditions. Higher temperature conditions of storage produce an important destabilisation in the control (25 and 37"C), but only samples treated at 300 MPa / 2°C and 450 MPa / 20°C and stored at 37°C show instability (Fig. I).
The photographs of Figure 2 show the changes in mi-
crostructure induced by the pressure and the storage at different temperatures.
CONCLUSIONS
Presswation could be used to keep the sta-ility and enhance the stiffiess
of mayonnaise without the use additives for a long time. However more work in this area must be done in order to improve the existing productg
and to design new ones.
High Pressure Food Science, Bioscience and Chemistry
262
Figure 1. Changes in the particle size distribution among the storage conditions Mayonnaise maintained for 1 month at 4 "C
-! 14s
Size (micrometers) Mayonnaise maintained for 1 month at room temperature
4
1
mu M 0
B a
5
5 0.
-g s 4 B
m OL '0
Size (micrometei s) Mayonnaise maintained for 1 month at 77 "C
Size (micrometers)
Food Science: Posters
263
Figure 2. Microscopic images of maionnaise among the preservation conditions Control at 37°C (1 month)
450MPa/10'/2"Cat 37°C (1 month)
450MPa/l0'/2"C at 20°C (1 month)
Control at 4°C (1 month)
450 MPa/l0'/2"C at 4°C (1 month)
264
High Pressure Food Science, Bioscience and Chemistry
REFERENCES
Fan, D. 1990. Review. High pressure technology in thefood industry. Trends Food Sci. Technol. 1: 14-16 Chehl, J.C. 1992. Efects of high hydrostatic pressure onfood constituents: un overview. p. 195-209. In C. Balny High pressure and Biotechnology. Colloque INSERM,John Libbey Eurotext Ltd.,London Gervilla, R.; Capellas, M.;Ferragut, V. and Guamis, B. 1997. Efect of high hydrostatic pressure on Listeria inocuq 910 CECT inoculated into milk J. Food Rot. 60:33-37 Mertens, B. 1993. Development in high pressurefoodprocessing 2. Lebensm Technol. 44: 100-104 Ponce, E.; Pla, R.; Mor-Mur, M.; Gervilla, R. and Guarnis, B. Inactivation of Cisteria inocw inoculated in liquid whole egg by high hyhostatic pressure. J. Food Rot. (In press) S.A.S. Institute Inc. 1982. SAS User’s Guide: Statistics, 1982 edition SAS Institute Inc. Cary NC. Styles, M., D. Hoover, and D. Farkas. 1991. Response of Listeria monoGitogeM and yibrio Darahaemolvticus to high hydrostatic pressure. J. Food Sci. 56: 1404-1407
The Effect of High Pressure on Microorganisms and Enzymes of Ripening Cheeses
A.Reps*, P.Kolakowski, F.Dajnowiec Olsztyn University of Agriculture & Technology, Institute of Food Biotechnology, Heweliusz Str. 1, 10-957 Olsztyn, Poland INTRODUCTION Among the modem technologies in the food industry, the most important are those involving non-thermal treatment of a product (Mertens & Knorr, 1992). The application of high pressure ranging from 100 to 1000 MPa, is one of the most promising methods for the food treatment and preservation at room temperature (Cheftel, 1992; Mertens, 1993). Gouda, Kurpiowski (Emmental type) and Camembert cheeses were subjected to high pressure treatment. MATERIALS AND METHODS Cyclic pressure 3 x 5 min in the range 200-1000 MPa at 200 MPa intervals, and room temperature, was applied to cheeses of different ripening stage: 2- and 6 week old Gouda, 2- and 5 week old Kurpiowski, and 5- and 10 day old Camembert. Pressure and its duration were selected on basis of the previous results ensuring highest inactivation of total microbial count in cheese (Kolakowski et al., 1995).
High Pressure Food Science, Bioscience and Chemistry
266
Microbiological analyses were performed immediately after decompression and following a 12-week cheese storage at 5°C. The proteolytic (substrate casein), aminopeptidase (substrates: L-lysine pnitroanilidine, L-y-glutamic acid p-nitroanilidine, L-leucine p-nitroanilidine, Lproline p-nitroanilidine and L-phenylalanine p-nitroanilidine) and endopeptidase (substrat N-succinylo-L-phenyalaninep-nitroanilidine) activities were studied in the protein extracts obtained by centrifugation of citric buffer-homogenized pressurized cheeses and in pressurized extracts of cheese proteins. The extent of proteolysis in cheeses pressurized in the range from 0 to 500
MPa for 240 minutes was evaluated by determining the pH, nitrogen soluble at pH 4.6, non-protein nitrogen, peptide nitrogen, amino acid nitrogen.
The pressurized cheeses were organoleptically assessed by an expert panel.
RESULTS AND DISCUSSION Infruence of high pressure on the viability of microorganisms in cheese
It was found that the degree of microorganism inactivation increased with the increase in pressure, except the spores. The type of cheese and its maturity also affected the degree of microorganism inactivation. A significant decrease of total microbial count was obtained at the pressure above 400 MPa. At 800 MPa the total count of microorganisms decreased by 4-6 log cycles, depending on the cheese type.
Food Science: Posters
267
It was found that the inactivation of microorganisms was affected more by their initial number than by the type of cheese and its maturity. Higher initial concentration of the microorganisms required higher pressure to obtain the same level of reduction. In 5 day old Camembert cheese and 800 MPa, acidic bacteria '~ to 4 . 0 ~ 1 0ckg-', ~ while in 10 day old number decreased from 2 . 8 ~ 1 0chg-' ~ ckg-*to 2% lo3 cheese the same pressure resulted in reduction fiom 3 . 9 lo9
cfug-'. Similar trends were observed in Gouda and Kurpiowski cheese. It was noted that at pressures above the 400 MPa vegetative cheese microflora were predominated by acidic and proteolytic microorganisms. Although neither the acidic nor proteolytic microorganismswere inactivated in 100% within the range of pressures applied, the degree of their reduction considerably increased with the increase of pressure, being fiom 99.687 to 99.998%for proteolytic microorganisms and from 99.943 to 99.999%for acidic bacteria at 600 MPa. Colibacilli were absent in 400 MPa-pressurized cheeses irrespective of their initial count. Enterococci were inactivated in Camembert and Kurpiowski cheeses at 400 MPa, while the pressure of 600 MPa was needed to achieve this effect in 2 week old Gouda. Yeasts and moulds in Gouda and Kurpiowski cheeses were inactivated with 200 MPa. They were absent in 10 day old Camembert exposed to 400 MPa and in 5 day old one exposed to 600 MPa.
Many other authors reported high susceptibility of yeasts and moulds to pressure (Takahashi et al. 199 1 ;Shigehisa et al. 199 1).
High Pressure Food Science, Bioscience and Chemistry
268
The spore number in the cheeses did not change much within the range of the pressures applied. The viability of the analysed microorganisms in the pressurized cheeses (3x5 min) stored at 5°C for 12 weeks and in the untreated cheeses showed similar changes. The count of microorganisms in stored cheeses usually showed a decreasing trend. It was noted that in cheeses exposed to pressure, in which the analysed microorganism groups were inactivated, these groups were absent also after storage of cheese. Influence of high pressure on the activity of proteases in cheeses
The proteolytic activity increased and the activity of endopeptidases decreased during the cheese ripening. The activity of aminopeptidases was higher in more mature cheeses. The degree of inactivation of proteases in the cheeses and their extracts increased with the pressure used. Irrespective of the substrate, the proteases did not show the enzymatic activity in 600 MPa-pressurized Kurpiowski cheese and in above 800 MPa-pressurized Camembert and Gouda cheeses. Generally, the inactivation of the proteases in cheese extract required a lower pressure compared to those in cheese. Aminopeptidases and endopeptidases of both cheese and its extract lost the catalytic abilities at 600 MPa, irrespective of the type and ripening time of cheeses. Aminopeptidases from less-matured 400 MPa-pressurized cheeses did not hydrolyze the most of the analyzed substrates.
F w d Science: Posters
269
The casein degrading proteases were the most resistant to pressure.
Influence of high pressure on cheeseproteolysis For Camembert cheese there was significant influence of high pressure on the proteolysis which was dependent on the applied pressure and the maturity of the cheese. For 10 day old Camembert, the highest degree of proteolysis was observed when a pressure of 50 MPa was appplied. This resulted in a 32.4% increase in soluble nitrogen, a 27.8% increase in NPN, a 40.2% and 31.5% increases in the peptide- and amino acid nitrogen, respectively, as compared to the control cheese. Smaller increases in nitrogen fractions were observed in 10 day old Camembert cheese pressurized at 100 MPa, and no changes were observed in the cheese pressurized at 200 MPa as compared to the control cheese. Yokoyama et al. (1993) applying high pressure to cheddar cheese with high lactic bacteria count obtained significant reduction of ripening time. The highest degree of proteolysis in 5 day old Camembert was found at the pressure 50 MPa, but the percentage differences were smaller than those for the 10 day old cheese. In the case of Gouda and Kurpiowski cheeses, there was no important influence of pressure, within the range of pressure and time applied, on proteolysis despite a substantial increase in pH of some samples.
Influence of high pressure on cheese quality The applied pressure influenced the rheological properties of the cheese.
Higk Pressure Food Science, Bioscience and Ckemistv
270
The pressurized Gouda was more elastic than the control cheese. In 5 day old Camembert being pressurized at 200 MPa for 4 h, a slight whey syneresis was observed. According to the panel of experts, the pressurized Gouda cheeses had superior taste and flavour properties than the control. In 10 day old Camembert pressurized at 50 MPa for 4h, the taste and flavour were superior to the untreated cheese of the same age and identical as those in the 14 day old untreated cheese. REFERENCES Cheftel, J.C. (1992), High Pressure & Biotechnology, C. Balny et al., eds John Libbey Eurotext, Mntrouge, pp. 195-209 Kdakowski, P., et al., (1995), High Pressure Science & Technology, ed. W. Trzeciakowski, World Sci. Pub. Co. Re. Ltd., Singapore, 898-901. Mertens, B. (1993), Int. Food Manufacture, 44,100-104 Mertens, B. & Knorr, D. (1992), Food Technol., 46, 124-133 Shigehisa, T. et al., (1991), Int. J. Food Microbiol., 12,207-216 Takahashi, K., Ishii, H., Ishikawa, H., (1991), High Pressure for Food, ed. R. Hayashi, San-Ei Pub. Co. Kyoto pp. 225-232 Yokoyama, H., Sawamura, N. & Motobayashi, N. (1993), US Pat. 005180596A
Diels-Alder Reactions of Food-relevant Compounds under High Pressure: 2,3-Dimethoxy-5-methylbenzoquinone and Myrcene
J. KuebePb', H. Ludwigband B. Tauscher" 'Institute of Chemistry and Biology, Federal Research Centre for Nutrition, Engesserstr. 20, D-76 131 Karlsruhe; bInstituteof Pharmaceutical Technology, University of Heidelberg, JNF 346, D69120 Heidelberg
Introduction
The chemical aspects of high pressure treatment of food have been paid little attention [ 1, 21 so far, even though Henri Louis Le Chatelier had described the effect of external force (here: pressure) on a system in equilibrium as early as in the late last century [3] and although the influence of pressure on chemical reactions has been thoroughly explored [4]. Among the reactions promoted by pressure are cyclization reactions which lead to covalent bonds usually associated with negative volumes of activation and reaction (AVRand AV*, resp.). The Diels-Alder reaction is a cycloaddition reaction activated by high pressure in particular with greatly negative volumes of activation and reaction as two covalent bonds are forming (AVR = AV* = -30 ml/mol). The Diels-Alder reaction, a pericyclic reaction, is since it is a [47t + 2x1 cycloaddition thermally allowed.
212
High Pressure Food Science, Bioscience and Chemistry
Materials and Methods Ubiquinone (also: coenzyme Q1,) is an important electron carrier in oxidative phosphorylationin the respiratory chain. It was named so because it is ubiquitous in biological systems. There are homologous compounds with shorter side chains; in mammals, however, coenzyme QIo is found most frequently. Because of its hydrophobic side chain it is able to difhse into the mitochondria1 membranes; in fact, it is the only permanently non-bound electron carrier of the respiration chain. In the present work the homologous coenzyme Qo (2,3-dimethoxy-5methylbenzoquinone) which does not have an isoprenoid side chain and is not a natural substance was used as the first of a series of different coenzymes of the Qseries. Partner to the Diels-Alder reaction is myrcene, a diterpene present as aroma compound in some herbs. A reaction of coenzyme Qowith myrcene according to the Diels-Alder mechanism may produce four isomeric compounds each existing in enantiomeric pairs (Fig 1). Solutions of 2,3-dimethoxy-5-methylbenzoquinoneand myrcene in a molar ratio of
1:3 in ethanol, in self-produced bags of a diffusion-tight foil (tri-laminate of PEtlAlLDPE), were exposed to 650 h4Pa und 70 OC for different times. The reaction was complete after 16 hours. The reaction mixtures were separated and analysed by 5890 I1 GC / 5985 B Quadrupole / Teknivent data system and the results shown in the form of a TIC
Food Science: Posters
273
2,3dimethoxy3-methylbenzcquinone
R-myrcene
I B
\O 0
5a,8adimethoxy-2-methyl-7-(4-methylpent-3-enyl)
5,5a,8,8a-tetrahydrc-l ,Cnaphthoquinone
+
t
0
' 0
0
2,3dimethoxy-8a-methyl-6-(4-methylpent-3enyl)-
5,5a,8,8a-tetrahydro-l ,4-naphthoquinone
Fig 1: Reaction products of the reaction of 2,3-dimethoxy-5-methylbenzoquinone with pmyrcene according to the Diels-Alder mechanism
chromatogram. A polydimethylsiloxane column (SPB 1 from Supelco) of 15 m in length, 0.2 mm in inner diameter and 0.2 pm in layer thickness had been used. Ionization was accomplished at 70 eV; the ion source temperature was 200 "C.
High Pressure Food Science, Bioscience and Chemistry
274
Results and discussion
Gas chromatographic separation gave two close peaks which, according to their mass spectra, were identified as the two products of reaction pathway A in Fig 1. (See Figure 2 and 3.) IS
Fig 2: Gas chromatogram of the two products (I, 11) and internal standards (IS)
20
50
100
150
200
Fig 3: Electron impact spectrum of peak I
250
300
340
275
Food Science: Posters
The observed mass spectrum of peak I suggests that 2,3-dimethoxy-8a-methyl-7-(4-
methylpent-3-enyl)-5,5a,8,8a-tertahydro174-naphthoquinonewas formed.
4
i'
69
2
275
20
50
100
150
200
250
300
340
Fig 4: Electron impact spectrum of peak I1
This mass spectrum indicates to be associated with 2,3-dimethoxy-8a-methyl-6-(4methylpent-3-enyl)-5,5 a, 8,8a-tetrahydro-1,4-naphthoquinone (Figure 4). The results shown were confirmed by NMR, IR and UV studies.
The results of these model experiments suggest that in pressure-treated food, compounds may be formed from valuable food components whose availability is therefore reduced. These compounds could also present a toxicological risk to be checked in each individual case.
High Pressure Food Science, Bioscience and Chemistry
216
References [ 13 B. Tauscher, Z Lebensm Unters Forsch 200,3 (1995)
[2] “H. Ludwig, H. Marx and B. Tauscher, EHPRG Proceedings of the Annual Meeting, Brno, Czech Republic, (1 994) 199; bibid,Proceedings of the
lstMain Meeting “Process Optimisation and Minimal Processing of Foods”, EC Copernicus Programme, Porto, Portugal Vol4, (1995) 3 1
[3] ‘H.L. Le Chatelier, Comptes rendus 99, 786 (1884); bibid,Annales des Mines 13 (2), 157 (1888) [4] “T. Asano and W.J. le Noble, Chem. Rev. 78,407 (1978); bR. Van Eldeik, T. Asano and W.J. le Noble, Chem. Rev. 89, 549 (1989)
Stability of Antimutagenic Activities in Fruit and Vegetables during High Pressure Processing P. Butz", R. Edenharde?, H.Fistef and B. Tauschef
'Institute of Chemistry and Biology, Federal Research Centre for Nutrition, Engesserstr. 20, D-76131 Karlsruhe bDepartmentof Hygiene and Environmental Medicine, Johannes-GutenbergUniversitat, D-55101 Maim INTRODUCTION Diets rich in fruit and vegetables have been found to be associated with a low incidence of many types of human cancer, probably because they contain inhibitors of carcinogenesis i n c l u b potent antimutagenssuch as flavonoids and coumarins. 2-Amino-3-methyl-imidazo[4,5-flquinoline (IQ), formed from high protein food under household cooking conditions is a carcinogen and highly potent mutagen in the Salmonellaheversion assay. This assay is suitable for testing fruit and vegetables for the presence and effectiveness of protective, i.e. antimutagenic, action against IQ. The antimutagenic potency of freshly squeezed juices of fruit and vegetables is in some cases reduced by heat treatment (Edenharder, 1994). In the present work the influence of hgh pressure treatment at different temperatures on the antimutagenic activity of h t and vegetables against IQ induced mutagenicity in Salmonella typhirnunurn TA 98 was investigated. MATERIALS AND METHODS
High pressure treatment: Experiments were conducted in a htgh pressure device consisting of a series of thermostated micro-autoclaves (ID 16 mm, ca. 10 mL) connected by valves (Butz et al., 1992). Preparation of juices: About 250 g of edible parts were homogenized in a home mixer with integrated centrifuge (Braun MultipressR)and the juices obtained were used for the tests without any further pretreatment. Samples to be assayed for peroxidase activity were centrifuged (10.000 x g; 20 "C; 30 min) and stored frozen until use. There was no loss in enzyme activity due to freezing. Determination of antimutagenic effect: The antimutagenic potential was tested according to Edenharder et al., 1994, who modified a method previously described by Maron and Ames (1983).
High Pressure Food Science, Bioscience and Chemistry
278
RESULTS AND DISCUSSION Figure 1 shows the results of dfierent treatments on cauliflower juice. A hgh number of revertants per plate represents low antimutagenicity or loss of protective potency. Addition of only a few pl of untreated freshly r.mbor nva(nll I p*" 3500 squeezed juice reduces the * vntrsefed 3 w tlOrnlo/lM)T number of revertants to nearly +lOrnrn 14w MPn/25 'C zero. In contrast the heat trea+I0 mm/600M P s / 2 5 ' C 2500 ted juice (10 min 100 "C) is no more able to reduce the number of revertants below 1500 1500 at any dose. In the pressurized samples the protective potency is nearly unaffected. Results of M e r experiments 0 IW 200 300 400 500 600 with fmts and vegetables are juice [ f i l / p h l compiled in table 1. AntimuFigure 1 Iiigh pre8aure treatment of cauliflower juice tagenic potency is indicated either in terms of ID,, values, which is the dose for 50 % reduction of mutagenic activity, or, where appropriate, in terms of residual mutagenic activity ("YO) at a maximum dose. Antimutagenicity of strawbemes and grapefit was not affected by any heat and pressure treatment
. - h i I
Table 1.
Antimutagenic effects of fruit and vegetable juices on mutagenicityinduced by 2amino3methylimidazo [4,5flquinoline (la)in S. typhimurium TAga aRer Various treatments ~~~~
~
dose for 50% reduction of mutagenic actiMy (IDm) [pkest] or residual mutagenic a d i y at rndmum dose tested [%I
FNI or raw Vagetable edract (batch)
10min
10min
100°C 50°C
10min 10min 10min 10min 400MPa 603MPa 603MPa 800MPa Z'C Z'C 5O'C 35%
n.d. n.d. 28 55 54% n d. n.d. 62% n.d 195 n.d 11556n.d. 250 170%' 57% n.d. 110%' n.d n.d. 160 n.d 120 80 88 €B%'n.d. n.d. 240 63 61 n.d. 235 n.d. 101 86 104 n.d. 242 n.d. *~dosstaadwas300ul n.d:nd dctclmined
80 Beets I P Beets II 98 Carrds 7 Cauliflower 46 Kohlrabi1 150 Kohlrabi II 5 Leeks 60 Spinach I Spinach I1 40 140 Tomatoes1 Tomatoes I1 2 242 Grapefruit Strawberries1 67 Strawberries11 2 3
217
40
n.d 32 n.d.
n.d. 18 n d.
42
25
n.d. 175223 n.d. n.d. 70 n.d.
n.d. 200 n.d. n.d. E.6 n.d. 150 n.d.
66 n.d 82 n.d.
80 n.d.
150 n.d 135 n.d. 66 n.d.
50 lG5 n.d.
90%' n.d. 61 n.d. 263
Food Science: Posters
219
applied. All vegetables tested responded to heat, but not to pressure. The antimutagenic activity of tomatoes and beets was inactivated by pressure, even though extremly high pressure was required (tomatoes: 600 MPd50 "C; 800 MPd35 "C, 10 min each; beets: 800 MPd35 "C 10 min). ACKNOWLEDGEMENT This work was supported by the European Community (EC) Framework Programme for Research and Technologcal Development, Contract: FAIR-CT96-1113. REFERENCES Butz, P., Koller, W.D., Tauscher, B. and Wolf, S., Ultra-High Pressure Processing of Onions: Chemical and Sensory Changes. Lebensmittel-Wissenschaft und Technologie, 27,463-467 (1994). Edenharder, R., Anthutagene Wirkung von Gemiise und Obst. XXIX.Vortragstagung DGQ, Quedlinburg, 2 1./22. March 1994. Maron, D.M., Ames B.N. Revised methods for Salmonella mutagenicity test. Mutation Research 113, 173-215 (1983).
Rheological Behaviour of Soya Protein Concentrate Following High Pressure Treatment A. Apichartsrangkoon,' D.A. Ledward,' A.E. Bell' and S.G. Gilmour' I
Department of Food Science and Technology, University of Reading, Whiteknights, PO Box 226, Reading RG6 6AP, UK 2Department of Applied Statistics, University of Reading, Early Gate, PO Box 240, Reading RG6 6FW, UK
Introduction The main soya roteins, glycinin 11S fraction) and conglycinin (the rtion of 7s f raction) are modi led by high pressure. Kajiyama et. E m d that on applying essures of less than 500 MPa to soy milk, the soy proteins dissociated an some coagulated to form a finn tofu structure. The present study describes the effect of pressure on soya protein. Materials & Methods Hydrated soya protein concentrate (moisture content 81%) was subjected to ressure treatment at 200,400,600 and 800 MPa; at 20 and 60 degree C; for O! and 50 minutes. The treated samples were analysed for storage (G') and loss (G") moduli over the frequency range 0.01-10 Hz by a controlled stress (100 Pa) rheometer. Hardness of the samples was also recorded by Texture Profile Analysis for 55% strain com ression. SDS-PAGE electrophoregrams were obtained in both non-reduce and reduced (with 2-mercaptoethanol) conditions.
r
$
B
Resuits & Discussion
0.01 I
0.1 1 FREQUENCIES (Hz)
10
Figure 1 : The initial plots of stor e and loss moduli against frequency sweep of soya samples subjected to %-SO0 MPa at ambient temperature for 50 mutes. Figure 1 shows evidence of el-like behaviour in which G' has frequency dependence than the ". The slopes for G' (0.055-0.09) are gher than for G" (0.024-0.026) and all intercepts of G' (5.45-6.25kPa) are higher than for G" (0.75-0.84 kPa). The rheological behaviours of the samples treated at the lowest pressure are similar to the untreated native material .
8
rter
u ¶a
w
(0
6.44
1
i
\o
I
I
so
I,
6.
u
--w) TEMpERA=wF) Figure 2 : Response surface analysis of the intercepts (kpa) of the storage moduli with pressure and tem rature at treatment times of 20 minute (2.1) and 50 minute (2.2). The contour plots of the Figure 2.1 (237% e contour plots of Figure 2.2 (2.4). Intercept = 6084.09 + 41.56P + 612.251; + 263.68T2 - 262.58PT + 129.61PT,(P50.05) Figures 2.1 and 2.3 show that when pressure-treated for 20 minutes at low temperature,increasing pressure leads to increasing intercepts whereas at high temperature increasing pressure yields decreasing mtercepts. Figures 2.2 and 2.4 show that when presure-treated for 50 minutes at low temperature increasmg pressure yields increasing intercepts,but at hgh temperature, increasing pressure has no significant effect on the mtefcepts. Overall figures 2.1-2.4 show that increasing temperature gives rise to larger intercepts.
M
6.17
FIGURE21
w
m
282
I
High Pressure Food Science, Bioscience and Chemistry
Food Science: Posters
FIGURE 4.1
283
kDa
kDa 205
Figure 4 : SDS-PAGE electrophoregrams of non-reduced (4.1) and reduced samples (4.2) (A; native samples, B-E; pressure treated at 200 MPa to 800 MPa at ambient temperature, F-I; pressure treated at 200 MPa to 800 MPa at 60 degree C). In the presence of 2-mercaptoethanol there is no difference in the samples but on its absence small differences are apparent, especially in the more severely treated samples. Thus although the gels have different rheological properties no non-disulphide covalent bonds are involved. Conclusion Pressure / temperature / time treatments can be manipulated to yield a range of textured soya products. Reference Kajiyama N., Isobe S., Uemura K. and Noguchi A. (1995) Inter J. Food Sci.Tech.,30,2,pp. 147-158. Acknowledgement The authors would like to thanks the Royal Thai Government for hnding this project
EFFECTS OF HIGH PRESSURE ON LIPID OXIDATION IN FISH Angsupanich,K. and Ledward, D.A. The University of Reading Department of Food Science and Technology PO Box 226, Whiteknights, Reading RG6 6AP, UK When cod muscle was subjected to pressures greater than 400 MPa for 20 min at 20°C prior to storage at 4°C marked increases in the rate of lipid oxidation, as measured by 2-thiobarbituric acid (TBA) number, were observed compared to samples treated at lower pressures. Samples treated in air had higher initial TBA numbers than those treated in nitrogen, but upon storage, both oxidised more rapidly than untreated samples. Addition of ethylenediamine tetraacetic acid (l%w/w) to minced cod totally inhibited the increased rate of oxidation induced by pressure suggesting that liberation of transition metal ions such as iron or copper, from complexes, occurs at 400 MPa and above which subsequently catalyse the reaction.
Introduction The application of high pressure processing to meat and fish based products are being increasingly studied (Cheftel & Culioli, 1997). However, the recent report on the effects of high pressure processing on meat and meat-like systems showed one of the major disadvantages was that the application of high pressure , irrespective of the presence of oxygen during treatment, led to increased rates of lipid oxidation during subsequent aerobic storage(Cheah & Ledward, 1997). The lipids in fish are more susceptible to oxidation than those
of most meat producing animals, since they have a high concentration of polyunsaturated fats and thus oxidative change induced by pressure may be very significant.
285
Food Science: Posters
To investigate the potential of high pressure processing to fish, the effects of pressure treatment on lipid stability were investigated. Materials and Methods Fresh cod (Gadus rnorhua) fillets here cut into pieces each weighing 120 g, sealed in Multivac bags (Bosley, International, NL) then pressure treated at 200 to 800 MPa at room temperature for 20 minutes in a prototype Stansted 'Food-Lab high-pressure rig (Stansted Fluid Power Ltd, Stansted, UK). Some samples were packed in nitrogen gas before pressurisation. In others, fish was minced to allow incorporation of antioxidant (1% w/w ethylenediamine tetraacetic acid disodium salt, Na2EDTA). The pressurised samples were divided and their 2-thiobarbituric acid (TBA) number (Pearson, 1976) determined either immediately or after storage in polyethylene bag at 4°C.
Results and Discussion Lipid oxidation of cod muscle packed in air after treatment at pressures below 400 MPa was limited, whereas a significant effect is seen at higher pressures (Figure 1). The accelerated oxidation may be due to the denaturation of haem protein by pressure which releases metal ions which play ah important role in promoting auto-oxidation of lipid in pressurised fish meat (Tanaka et a/., 1991; Ohshima ef a/.,1992). Other studies (Wada, 1992; Cheah and Ledward, 1996) suggest that the denatured proteins may be an important factor in catalysing lipid oxidation in pressure treated meat . Overall, we find no significant difference in TBA number in the unpressurised and pressurised samples packed in nitrogen (Figure 2). However, during subsequent storage in air for 7 days at 4°C , the TBA number increases rapidly. This suggests that catalytic
effect
pressurisation.
the
does not relate to the presence of oxygen during
286
High Pressure Food Science, Bioscience and Chemistry
3
2 2.5
i. 1
1.5
i l
E"
5
0.5 0
Figure 1 Effect of high pressum on the extent of lipid oxidation for pressure treated cod muscle in the presence of air. The TBA value is pfolted against the applied pressure for freshly treated samples(+) and samples stored for 7 days at 4' .C (m). All treatments were 20 min duration.
-
1
10.8
1
p 5
0.8
(I
0.4 0.2 I
0
200
400
800
800
PnPunlMPA
Figure 2 influence of high pressure on the extent of lipid oxidation for pressure treated cod muscle in the presence of nitrogen. The TBA value is plotted as a function of applied pressure for 1 day old samples (+) and samples stored for 7 days at 4OC (a. All treatments were 20 min duration.
Food Science: Posters
281
Figure 3 shows that addition of 1W Na2EDTAeffectively inhibited the catalysed oxidation induced by high pressure. This result confirms that the acceleration of lipid oxidation is due to the release of free metal ions (Fe and Cu) from complexes at around 400 MPa (Cheah & Ledward, 1997).
Figure 3 Effect of 1% Na2EDTA on TBA' number (mg matonaldehydelkg) in minced cod following pressure treatment for 20 min and storage at 4°C. The TBA number is plotted against the varied pressures for initial day sample and stored samples at 4 and 8 days.
288
High Pressure Food Science, Bioscience and Chemistry
References Cheah, P.B. & Ledward, D.A. (1996) . High pressure effects on Lipid oxidation in minced pork. J. Meat Sci., 43, 123-134. Cheah, P.B. & Ledward, D.A. (1997). Catalytic mechanism of lipid oxidation following high pressure treatment in pork fat and meat. J. food Sci., In
press. Cheftel, J.C. & Culioli, J. (1997). Effects of high pressure on meat : a review. Meat Sci., In press. Ohshima, T., Nakagawa, T. & Koizumi, C. (1992). Effect of high hydrostatic pressure on
the enzymatic degradation of phospholipid in fish muscle
during storage. In “Seafood Science and Technology” (E.G. Bligh, ed.), Fishing News Book, Oxford, UK, p. 64-75. Pearson, D.(1976). The chemical Analysis of Foods. 7th ed., Churchill Livingstone, Edinburgh. pp. 496-497. Tanaka, M., Xueyi, Z., Nagashima, Y. 8 Taguchi, T. (1991). Effect of high pressure on the
lipid oxidation in sardine meat. Nippon Suisan
Gakkaishi., 57, 957-963. Wada,
S. (1992). Quality and lipid change of sardine meat by high pressure treatment. In “High pressure and biotechnology” Vol. 224 (C. Balny, R. Hayashi, K.
Heremans
& P. Masson, eds.), Montrouge, Colloque
INSERMlJohn Libbies, Eurotext, Ltd. , p.235-238.
Impact of Combined High Pressure and Low Temperature on Enzyme Inactivation: Kinetic Study of Soybean Lipoxygenase Indrawati; A. Van Loey; L. Ludikhuyze and M. Hendrickx Department of Food and Microbial Technology; Laboratory of Food Technology; Katholieke Universiteit Leuven; Kardinaal Mercierlaan 92; B-3001 Heverlee-Belgium. Phone: 32- 16-32 1572, Fax: 32- 16-32 1997 and E-mail: [email protected]. be Soybean lipoxygenase (LOX) dissolved in Tris HCl buffer (0.01M; pH 9) could be irreversibly inactivated by combined pressure (up to 6.5 kbar) and low (-15 up to 35OC) temperature treatment. The enzyme inactivation followed a first order reaction and the phase transition of water did not change the kinetic inactivation behaviour.
1. INTRODUCTION
The effect of high pressure on the solid-liquid phase diagram of water, offers several potential food applications by use of high pressure and low temperature: pressure-assisted freezing, pressure-assisted thawing and non frozen storage for vegetables and fruit (1). However, hitherto, most of the kinetic inactivation studies on food related spoilage enzymes have been performed at high pressure in combination with elevated temperature but less at subzero temperature. The
290
High Pressure Food Science, Bioscience and Chemistry
objective of this study is to investigate kinetically the impact of a combined pressure (up to 6.5 kbar) and temperature treatment, especially in the low temperature area, on a food related enzyme, i.e. soybean lipoxygenase (LOX).
2. MATERIALS AND METHODS 2.1.Enzyme and activity measurement Soybean LOX type 1B (SIGMA, EC. 1.13.11.12), purchased as a dry powder, was dissolved in Tris HCI buffer (0.01M; pH 9) at a concentration of 0.4 mg/ml. The spectrophotometric (234 nm) assay of enzymic activity was based on the dienoic fatty acids formation, was modified from Axelrod et a1 (2): the substrate solution consisted of 280 mg of linoleic acid and an equal weight of Tween 20 in 25 ml oxygen-free water. The assay mixture contained 2.9 ml sodium borate buffer (0.0125M;pH 9); 0.025 ml substrate solution and 0.045 ml enzyme solution. The enzymic activity was derived from the slope of the absorbance increase at 234 nm and 25OC as a finction of reaction time.
2.2.Combined pressure and temperature treatment Pressure-temperature induced inactivation kinetics of LOX were determined on the basis of isobaric/isothermal inactivation experiments and determination of residual LOX activity. Isothermal and isobaric treatments were performed in a thermostated multivessel (8 vessels with volume of 8 ml) high pressure
Food Science: Posters
29 1
equipment (Resato, the Netherlands) with a pressure capacity in the range from
0.001 up to 10 kbar and a temperature capacity in the range from -30 up to 1OOOC. In all experiments, the pressurization rate was 0.9-0.95 kbadminute. The
enzyme solution was contained in flexible micro-cups (0.375 ml) closed by parafilm and for subzero temperature combinations, wrapped with polyethylene pockets to avoid pressure medium (oil-glycol) contamination. Afier the pressuretemperature treatment, samples were stored at 25°C for at least 20-30 minutes before activity measurement.
3. RESULTS AND DISCUSSION Isobaric and isothermal LOX inactivation could be accurately described as a first order reaction (equation 1) as can be seen from the loglinear plot of relative activity retention versus inactivation time (Figure 1). ln(A) = ln(A,) - k*t
(1)
A phase transition of water (liquid and ice V) did not change that behaviour, e.g. high pressure combined with -15°C (Figure 1). Based on the Eyring model, the pressure dependence of the rate constants is commonly expressed using an activation volume (Va) (equation 2). In( k) = In( k r e f ) +
(
:; *
- (Pref
-
292
High Pressure Food Science, Bioscience and Chemistry
In the low temperature area (-15 to 35"C), the Eyring model could be applied straightforwardly to estimate the activation volume from the slope of the
regression line of ln(k) versus P. The highest pressure dependence of k-values occurred at temperatures of 20-25°C (Figure 2).
4 3 kbar;-15"C (liquid phase) W 4 kbari-15"C (ice V phase)
4.5 kbar;-15"C (ice V phase) 0
10
20
30-
40
50
60
70
80
90
1W
Time (minute)
Figure 1. Isothermal and isobaric inactivation of soybean LOX (0.4 mg/ml) in Tris HC1 buffer (0.01M; pH 9) at different P/T combinations. Ao and A are initial LOX activity and activity after P/T treatment respectively.
Based on the Arrhenius model, the temperature dependence of the rate constants is commonly expressed using an activation energy (Ea) (equation 3). The rate constant at a certain pressure could be increased by decreasing temperature (Figure 3). However, no linear relationship between ln(k) and reciprocal absolute temperature (UT) was observed.
Food Science: Posters
293
r--
f
40
Figure 2. Evolution of pressure dependency of inactivation rate constant of soybean LOX (0.4 mg/ml) in Tris HCl buffer (O.O1M;pH9) as a function of temperature
-’1
-2 - -3 -4
~~
--
4
-6 0.0032
! 0.0034
0.0036
0.0038
l / t e m p e m h (1/K)
Figure 3. Inactivation rate constant of soybean LOX (0.4 mg/ml) in Tris HC1 buffer (0.OlM;pH 9) at 5.25 kbar as a function temperature
High Pressure Improvement of the Meat Ageing Enzymes Activity
S. Jung, M. de Lamballerie-Anton, Ph. Courcoux and M. Ghoul
Ecole Nationale d’Ingenieurs des Techmques des Industries Agncoles et Alimentaires, BP 82225,44322 Nantes Cedex 3, France
Introduction Meat tenderness is obtained by a storage period of two or three weeks at 0-5°C. During this period, several physicochemical and enzymatic modifications are involved, particularly the liberation of some enzymes as cathepsin D, by the breakdown of lysosomal membrane. However, this procedure presents a risk of microbiological contamination and is costly in time and energy. One alternative consists to liberate as fast as possible the lysosomal enzymes to reduce the storage period. High pressure technology could be a solution to attempt this goal (Ohmori et ul, 1992). The aim of this work is in one hand to improve the lysosomal enzymes’liberation and in the other hand to establish the optimal conditions allowing the higher level of enzyme liberation without drastic modifications of the main physicochemical properties particularly the colour of the meat.
296
High Pressure Food Science, Bioscience and Chemistry
Materials and methods
Animals and conditioning Muscle samples were obtained from beef (n=3, Levesque BVB, Blain, France). The muscles (t3zcep.sfemorzs) were vacuum conditioned 24h postmortem and submitted to pressure treatment.
Preparation of the beef extracts and enqyme solution Meat's cathepsin D was prepared as described by Homma et a1 (1994). Muscle (20 g) was minced and homogenised in 20 ml of cold water (4°C) using a Turrax homogeniser. Homogenisation was carried out during 30 s and 20 ml of cold water was added. The solution was then centnfuged at 10000 g for 15 min and the supernatant obtamed was filtered through glass wool and dialyzed overnight at 4°C. Samples were then centrifuged at 12000 g for 20 min and supernatants were stored at 4°C for further analysis. Commercial cathepsin D was obtained from Sigma (S' Quentin Fallavier, France) and was dissolved in 0.2M so&m acetate/acetic acid buffer (pH 3.7).
Pressurisation of muscles and entyme Muscles and commercial enzyme solution were vacuum-sealed in a polyethylene bag and pressunsed at 10°C using a reactor of 3 liters fiom GEC ALSTHOM ACB (Nantes, France). The time and pressure applied were defined by an experimental design methodology.
297
Food Science: Posters
Experimental design The simultaneous effects of applied pressure (50-600 MPa) and duration of treatment (20-300 seconds) were studied. Experiments were designed using a central composite design (Table 1). Three center points for estimation of the experimental error and the lack of fit were made. Statistical analysis was performed using Statgraphics software. Cathepsin D activity and meat colour measurements were the response values of the experimental design. Table 1 : Experimental design Experiments 1 2 3 4 5 6
7 8 9 10 11
Pressure (MPa) 325 520 600 50 130 325 520 130 325 325 325
Time (seconds) 160 60 160 160 260 160 260 60 20 300 160
BiochemiJ analysis The protein concentration of the extracts was performed by the method of bicinchoninic acid protein (Sigma, Rocedure NOTPRO-562). Bovine Serum Albumin was used as standard. Cathepsin D activity was determined with haemoglobin assubstrateaccording to Anson's method (1938). Samples of 500 p1 were added to 1 ml of 0.2M sodium
298
High Pressure Food Science, Bioscience and Chemistry
acetate/acetic acid buffer (pH 3.7) containing 2% (w:v) of haemoglobin. After incubation during 3h at 37"C, the reaction was stopped by adding 1.5 ml of trichloroacetic acid (TCA). The precipitate was removed by filtration (Whatman Cat N" 1440090) and the released trichloroacetic-soluble peptides were measured at 280 nm. The activity was determinated as absorbance/g protein. The relative activity was the ratio between treated and untreated samples. The brightness (L*) and colour (a* and b*) were determined by a ClELAB system in reflexion mode with spectrocolorimeter Uvlkon 8 1OP.
Results and discussion
Effects of pressure and time treatments on the entymatic activities The relative activity of cathepsin D extracted fiom bovine muscles is summarised in the Fig. l a and lb. These results indicated that the level of relative activity of
cathepsin D depends both on pressure and time duration. For low values of pressure and time, cathepsin D activities are lower compared to the untreated samples ( 100%). For hgher pressure and time, the activities rose gradually. The best results were obtained while set points were kept in a range of (580-600 M a , 2s-300 s). Fig.1 indicated also that the pressure has a more significant effect than
duration. It seems that for low pressure and time duration, the denaturation rate of cathepsin D is hgher than the lysosomal membrane destruction rate. To elucidate
299
Food Science: Posters
the effect of (P, t) on the kinetics of cathepsin D, a commercial enzyme was treated in the same manner as beef extracts. The obtained results are given in Fig. 2a and 2b. The commercial cathepsin D activities gradually decrease with increasing pressure and time, At 600 MPa and 200s, the relative activity of this enzyme is only 60% of the control. The above results indicate that the denaturation rate of cathepsin D is more drastic for high pressure values. This denaturation depends also on time duration. So the measured activities of cathepsin D of the meat (Fig. 1 ) are in fact residual activities due to an equilibrium between the liberation
and denaturation kinetics.
-1:
la
2a
lb
2b
surface (a) and contour plots (b) of the relative activity of cathepsin D.
Fig 2 : Response surface (a) and contour plots (b) of the relative activity of commercial cathepsin D.
High Pressure Food Science, Bioscience and Chemistry
300
Effects of pressure and time treatments on the colour The evolutions of L*, a* and b* values of meat submitted to different couples (P,t) are shown in Fig. 3 a,b,c. It appears that both time and pressure have a significant effect on the colour parameters. These results are in agreement with those described by Carlez et a1 (1 995)
-0
Fig.3 : Contours plots of L*, a*, b*. The mean values for the control are : L*=37.95, a*=l5.69. b*=5.98
5-E
F
m
a
n
m
a
m
w
Taking into account the evolution of cathepsin D activities and the three index of colour, mainly the a* index, the couple (P=520 MPa, t=260s) was chosen for the further investigations.
Evolution of the entymatic activity during the storage To study the behaviour of liberated cathepsin D during the storage period,
enzyme activity of treated and untreated samples was followed along three
Food Science: Posters
301
weeks. Table 2 shows the obtained results. The activity of cathepsin D of the untreated sample increased with the storage time. This phenomenon could be explained in one hand by cathepsin D liberation occurred during the ageing process and in the other hand by its activition at low pH (Chambers et ul, 1W6). For pressurised samples, after 20 days of storage period, more than 20% of
cathepsin D activity was lost. However, it remains two fold higher than the control at ths time. Time postmortem (days)
Percentage control
pressurised
100 245 28 132 07 269.81 233.96 141 51 116.98 200.00 Table 2 Evolution of cathepsin D activity of pressurised meat (520 MPa -260 s) and the control 2 9 14 20
Evolution of the colour of meat during the storage period As it has been said previously, the colour is an important parameter for
consumers. So profiles of the three index have been followed along ageing period. Fig.4 shows the evolution of the three index (L*, a*, b*) for pressurised meat and the control. For L* values the difference between control and samples remains constant during storage period. For untreated meat the values of a* and b* increased slightly with time, whle for pressurised samples a* values decreased and b* values were almost constant.
High Pressure Food Science, Bioscience and Chemistry
302
0
* $ ; i A
f
4 0 - A
cl w
:
z
.-
20
I
Fig. 4 : Evolutions of L*(o),a*(A), b* (U) during the storage. Solids symbols : non-treated samples, open symbols : pressure treated at 520 MPa 260 s
i -
4
&
-
I
I
0
S
€3
-
W
10 IS 20 Postmortem time (days)
2s
Conclusion The above resL..s show clearly that .igh pressure induces the des-uction of lysosomal membranes which leads to a high level of cathepsin D activity in the medium. The degree of lysosomal destruction depends on the couple (P, t). In our experiments, the higher activity was reached for high pressure. Colour index is
also affected by pressure but it is possible to find particularly conditions leading to minimum variation of these parameters. High pressure technology seems to be a serious alternative for meat ageing process. However further investigations are needed to establish the effect of high pressure on the textural properties for raw and cooked meat. References Anson M.L., 1938, J. Gen. Physiol., 22,79
Food Science: Posters
303
Carlez A., Veciana-Nogues T., Cheftel J.C., 1995, Lebensm.-Wiss.u.-Technol, 28,528-538 Chambers J.J., Reville W.J., Zeece M.G., 1996, Science des aliments, 14, 441457 Homma N., Ikeuchi Y., Suzuki A., 1994, Meat Science, 38,219-228
Ohmori T., Shigehisa T., Taji S., Hayashi R., 1992, Biosci. Biotech. Biochem, 56(8), 1285-1288
High Pressure Inactivation and Survival of Pressure-resistant Escherichia coli Mutants in Fruit Juices C. Garcia-Graells, K. Hauben, C. Soontjes and C. Michiels Laboratory of Food Microbiology , Katholieke Universiteit Leuven, Kardmaal Mercierlaan 92, B-3001 Heverlee (Belgum) Phone: + 32/16/32.15.79. Fax:+32/16/32.19.97
High pressure inactivation and survival of E.co1i has been investigated in h u t juices and in low pH buffers. The parent strain E.coli MG1655 and the pressureresistant mutant LMMlOlO (Hauben et ul. 1997) can survive for several weeks in fi-uit juices stored at 8°C. To test the pressure-resistance of these strains under acidic conditions we subjected the cells to treatments of 300 - 400MPa at room temperature during 15 min. in apple, orange and mango juice and in low pH buffers. The mutant survived all treatments, except in apple treated at 400MPa, and remained substantially more pressure-resistant than the parent strain.
INTRODUCTION We have reported recently the isolation of three extremely pressure-resistant mutants of E.coli MG1655 (Hauben et u1.,1997). Based on stuhes in phosphate buffer pH 7.0, we anticipated that these mutants are able to survive severe HP pasteurizations at up to 800MPa. Pressurised fruit-based products have been
Food Science: Posters
305
pasteurized at up to 800MPa. Pressurised fruit-based products have been introduced on the Japanese market since 1990 (3,4,6), and more recently also in Europe (1). Although pathogenic bacteria can not grow in fruit products, some strains of Enterobacteria like E.coli can survive for several weeks in fruit juices. Recently, apple juice and apple cider have been implicated in some outbreaks of
E.coli 0 1 57:H7 (5,7). To evaluate the implications for the safety and stability of pressure-processed foods we have studied the potential of HP-resistant E.coli mutant to survive HP treatment in different fruit juices: apple juice pH 3.24, orange juice pH 3.72 and mango juice pH 3.96.
RESULTS AND DISCUSSION Survival in different h i t iuices stored at 8 and 25°C We studied the long-term survival of our E.coli strains in the three juices at two different storage temperatures. The results are shown in Figure 1, and revealed that survival correlated well with juice pH at both temperatures and for both
strains, the juice with the lowest pH resulting in the most rapid inactivation. Second, survival was considerably enhanced at refrigeration temperature (8"C), with a decrease of only 1 log cycle in orange and 2.5 log cycle in apple juice after 30 days. Survivors of both strains were detected at 4 days but not at 10 days of storage at 25°C. The results did not show a significant difference in survival between the parental strain and the pressure-resistant mutant.
306
High Pressure Food Science, Bioscience and Chemistry
Fig 1. Survival of E.coli MG1655 (closed symbols) and the mutant LMMlOlO (open symbols) in h i t juices stored at 8°C and 25°C.
apple juice stored at 8OC
apple juice stored at 25OC
87 8 1
0
5
10 15 20 25 Days of storage
30
0
orange juice stored at 8OC
2
4 6 8 Days of storage
1
0
orange juice stored at Z5OC
8 1
,
"
0
5
10 15 20 25 Days of storage
0
30
0
2
4 6 8 Days of storage
1
0
Food Science: Posters
307
High Dressure inacb'vation in h i t iuices and buffers at low DH.
The pressureresistance of the mutant LMMlOlO has been characterized previously in phosphate buffer pH 7.0 (Hauben et 01.1997). Here, both strains were subjected to treatments of 300 and 400MPa in apple, orange and mango juice, and in Hepes pH 7.0, LMMlOlO was also treated in Hepes pH 3.0,3.5and 4.0. The results (Table 1) indicate that, also in the f i t juices the mutant is
considerably more pressure-resistant than the parent strain, although both strains are much more pressure sensitive in the juices than in the buffer pH 7.0.At
400MPa,inactivation of LMh4lOlO in apple juice was higher than in the Hepes buffers with the closest pH values, while in orange and mango juices was lower. Therefore, other juice-specific factors in addition to pH may be present that d a c e or reduce the lethal effect of pressure.
CONCLUSIONS. Although both the parent strain and the mutant showed a high ability to survive in acidic conditions at reliigeration temperature, they become more pressuresensitive as the pH decreased. Also the mutant is less pressure resistant in the juices than in the buffer pH 7.0,and this is partly due to the low pH of the juices, but additional factors such as organic acids may also significantly affect the lethal effect of high pressure on E.coli in the h i t juices. However the mutant E.coZi
High Pressure Food Science, Bioscience and Chemistry
308
LMMlOlO remains highly resistant relative to the parent strain, and has the potential to survive pressurization processes that otherwise efficiently eliminate non pressure-resistantE.coli strains fiom h i t juices.
Table 1. Logarithmicviabilityreduction of E.coli MG1655 and LMM1010, for 15 min. pressure treatment at 20°C in h i t juices and in buffers.
300MPa
400MPa
MG1655
LMMlOlO
MG1655
LMMlOlO
Hepes 3.0
N.D.
2.85
N.D.
3.32
Hepes 3.5
N.D.
1.20
N.D.
2.32
Hepes 4.0
N.D.
0.47
N.D.
2.58
Hepes 7.0
4.96
0.35
6.99
0.62
apple juice
4.38"
1.12
4.38"
4.71'
orange juice
3.54
0.80
4.39'
1.49
mango juice
N.D.
0.40
N.D.
0.94
'Minimum detection limit was 20 CFU/mL. Inactivation was expressed as log (NdN), with No and N the counts for the untreated control and the pressure-treated sample respectively. N.D.Not determinated.
Inital cell count was 1 x 10'- 3 x lo9 in buffers and mango juice and Sx lo5- 9 x lo6 in apple and orange juice.
Food Science: Posters
309
REFERENCES
1 . Anonymous (1997). Food Engineering Int. June, 30. 2. Hauben K.J., D.H. Bartlett, C. Soontjens, K. Cornelis, E. Wuytack and C. Michiels (1997). Appl. Environ. Microbiol. 63,945. 3. Ogawa H., K. Fukuhisa, Y. Kubo and H. Fukumoto (1990). Agric. Biol.
Chem. 54,1219. 4. Ogawa H., K. Fukuhisa and H. Fukumoto (1992). Colloque INSERM 224, 269. John Libbey Eurotext, Montr., France. 5. Semanchek J.J. and D.A. Golden (1996). J. Food Rot 59, 1256. 6. Takahashi Y., H. Otha, H. Yonei and Y. Ifuku (1993). J. Food Sci. Technol.
28,95. 7. Zhao T., M.P. Doyle and R.E. BesseT (1993). Appl. Environ. Microbiol. 59,
2526.
Kinetics of Vitamin C Degradation under High Pressure-Moderate Temperature Processing in Model Systems and Fruit Juices P.S.Taoukis",, P.Panagiotidis", N.G. Stoforos", P.Butzb, H. Fisterband B.Tauscherb National Technical University of Athens, Dept. of Chemical Engineering, Lab. of Food Chemistry and Technology, Iroon Polytechniou 5 , Zografou 15780, Athens, Greece b Institute of Chemistry and Biology, Federal Research Centre for Nutrition, Engesserstr. 20, D-76 131 Karlsruhe, Germany a
Abstract Kinetic parameters of vitamin C degradation during processing under different combinations of isobaric (0.1 to 600 MPa) and isothermal (40°C to 75°C) conditions were determined in a model system (1000 mg/l ascorbic acid and 10% sucrose in 0.1N sodium acetate buffer, pH of 3.5-4) and h i t juices (pineapple and grapefruit). Experiments were conducted in a high pressure micro-autoclave of approximately 5 ml volume and a maximum operating pressure of 1000 MPa. Initial and final vitamin C concentration data, obtained at several prespecified processing times, were used in an Arrhenius type model to estimate the order of the degradation reaction, the reaction rate constant (k), the activation energy (E,,), and the activation volume (E,) for each product. These parameters were estimated simultaneously using a non linear fitting procedure and treating the data for each product as a single data set. For all products, the degradation of vitamin C followed the kinetics of a fvst order reaction. The values for k (at 50°C and 500 MPa, in mid), E, (in kJ/mol), and E, (in ~m-~/mol) were, respectively, 0.00801, 19.8 and -3.54 for the model system, 0.00935, 58.5 and -4.75 for the pineapple juice, and 0.00999,21.9, and -4.80 for the grapefruit juice.
Introduction Food processing under high pressure (up to 1000 MPa) and low to moderate temperature (of less than 100OC) conditions has been recently introduced as an alternative to high temperature preservation. By avoiding the detrimental effects
Food Science: Posters
311
of elevated temperatures on the various food quality attributes, high pressure processing CM offer a distinct advantage, over traditional thermal processed foods, as far as product quality is concerned (Fan,1990; Knorr, 1993). High pressure processing can be also considered for the preservation of food items that normally can not be thermally treated (due to the similar kinetics of thermal destruction of the spoilage and the quality factors for the product under consideration). Food quality degradation depends upon a variety of extrinsic (i.e., environmental, e.g., temperature, pressure, presence of oxygen) and intrinsic (i.e., compositional, e.g., pH, water content) factors. So, vitamin C losses in peas (a low-acid food) during thermal processing (over a temperature range of 1 10°C to 132°C) was characterized by an E, value of 164 kl/mol and a klzl.lT value of 0.0025 m i d , while vitamin C losses in grapefruit juice (acid food) during thermal concentration (over a temperature range of 61°C to 96OC) were characterized by value of 0.00128 min'' for an11.2"Brix an E, value of 21 kJ/mol and a bloc concentrate (Villota and Hawkes, 1986). In general, vitamin C degradation was described by a first order reaction. Unfortunately, though, kinetic information on quality losses during processing (and storage) are hgmentary in the literature, even for traditional (e.g.,thermal) processes (Villota and Hawkes, 1986). The objective of this work was to study vitamin C degradation during high pressure-moderate temperature processing under various isobaric and isothermal conditions, in a model food system and fruit (pineapple and grapefruit) juices. Our approach consisted of determination of the kinetic parameters (i.e., order of reaction, reaction rate constant, activation energy, and activation volume) of the degradation reaction involved.
Materials and Methods Vitamin C degradation during processing under different combinations of isobaric (0.1 -i.e., atmospheric pressure-, 300,450,and 600 MPa) and isothermal (40,60,and 75°C)conditions was measured in a model system (1000 mg/l ascorbic acid p.A. (Merck 100127) and 10% sucrose in 0.1N sodium acetate buffer, pH of 3.5-4) and pineapple and grapefruit juices. The juices were extracted from fresh grapefruits and pineapples and stored at -50°C until treatment. Experiments were conducted in a high pressure micro-autoclave of approximately 5 ml volume (ID 8 mm) and a maximum operating pressure of 1000 MPa. Pressure was generated using a hydraulic pump in combination with a pressure intensifier. The pressure transmitting medium was glycol based (HFU7000-1,aad Hochdrucktechnik,Bad Homburg, Germany).
u+)-
312
High Pressure Food Science, Bioscience and Chemistry
For pressure treatment, samples were filled into 1 ml Teflon tubes (ID 6 mm; OD 8 mm) and sealed with silicon stoppers. Initial and remaining vitamin C concentrations, obtained at several pre-set processing times (of 10, 20, and 40 min), were determined by an enzymatic test kit (Boehringer Mannheim Nr. 409677, Germany). Experiments were performed, at least, in triplicate. For data analysis, vitamin C concentration values for each of the multiple experiments were used as following: Assuming that vitamin C degradation during high pressure-moderate temperature follows the kinetics of an apparent n order reaction, dC dt
- -= kCn
(1)
Integrating Eq. (1) for constant pressure and temperature conditions (Van Boekel, 1996) and further assuming that the effect of temperature (under constant pressure) as well as the effect of pressure (under constant temperature) on k is given by an Arrhenius type relationship (Johnson and Eyring, 1970)
k = -kTref exp[-
E,
1
-(-- -
R T
-)I 1 Tref
and
one obtains: For n=l - k,f
E, 1 1 exp[--(- -) R T Tref
-
R
T
and for n#l 1
Ea 1 1 1 + (n - l)kref exp[--(- -) R T Tref
E (P-Pref)lCon-l R T
-2
For each food product studied ( i e . , model system, pineapple and grapehit juices) the experimental data (vitamin C concentration vs time) were fitted through Eq. (4) or (5), using the SAWSTAT multiple nonlinear (NLIN) procedure, and the appropriate kinetic parameters ( i e . , n, kref,E,, and E,) were estimated. These parameters were estimated simultaneously, treating the data for each product as a single data set. Reference conditions of 323 K temperature and of 500 MPa pressure were used. Initial guesses for the kinetic parameters were obtained using a trial and error procedure.
Food Science: Posiers
313
Results and Discussion Results on vitamin C degradation at the end of the treatment (i.e.,at 40 min of processing time), for all products and at all conditions investigated, are presented on Table 1. The results are indicative of the detrimental effect of increasing processing temperature and pressure on vitamin C. The degradation of vitamin C in pineapple juice shows a stronger temperature dependence, compared to the model system and the grapefruit juice; the latter two products gave comparative responses. Based on Eq. (4) and (5), for processes characterized by constant E, and E, values, one should expect a monotonous effect of increasing temperature and pressure on vitamin C degradation. With a few exceptions, for the model system and the pineapple juice at the higher temperature studied, this was observed in the experimental data gathered. These exceptions might be attributed to experimental variation, as well as to the fact that the high vitamin C destruction rates, observed at high temperatures, might have masked the effect of pressure. Replicate experiments for each processing condition shown on Fig 1, where the evolution of vitamin C degradation over processing time for selected processing conditions is illustrated, are indicative of the experimental variation encountered. Moreover, a possible temperature dependence of the activation energy or a pressure dependence of the activation volume of the degradation destruction, might have been the cause of a non monotonic behavior. As mentioned earlier, the experimental vitamin C concentration data, as a function of time, for each of the multiple experiments were used with Eq. (4) or (5) in order to simultaneously estimate the appropriate kinetic parameters ( i e . , n, Table 1 . Percent vitamin C degradation (* one standard deviation) after 40 min of treatment at the indicated P and T conditions.
P (MPa) 0.1 0.1 0.1 300 450 600 600 600
T (“C) 40 60 75 75 75 40 60 75
Model System 85.8*0.6 78.8h3.3 82.3*3.1 70.W8.1 76.1*1.8 78.6*2.6 70.4*4.4 54.750.6
Pineapple Juice 93.31t1.6 90.H2.9 39.4h7.4 30.2h2.0 35.9*2.2 78.31t1.4 54.8h5.5 38.3k4.1
Grapehit Juice 93.0h2.0 89.2h1.7 67.81t5.4 61.0h4.3 59.4h5.6 66.81t6.9 61.4h15.3 59.3k1.8
High Pressure Food Science, Bioscience and Chemistry
314
0
-0.05
A
-0.1 n
u"
-
-0.15
M 0
-0.2 -0.25
0
10
20
30
40
Processing time (min) Figure 1. Vitamin C degradation for the model system. Straight lines represent individual (first order reaction) fitting for each processing condition. k,f, E, and E,) for each product studied. When the order of reaction was presumed different than one ( i e . , n#l) convergence of the model ( i e . , Eq. 5), was either approaching to n=l or was unsatisfactory. The kinetic parameters obtained through Eq. (4), that is, for n=l, are presented on Table 2. The k,,f and E, values were comparable for all the three products studied. However, the E, value for the pineapple juice was almost triple the value of the model system and the grapehit juice. Higher Ea values are indicative of stronger temperature effects on the reaction rate. This was also shown in the results presented on Table 1. Using the estimated kinetic parameters, the model (Ed. 4) predicted the vitamin C remaining concentrations with a relatively small error. The comparison between predicted and experimental concentration data was satisfactory (Fig. 2). According to previous experience, predictions could be improved if an Ea and an E, dependence on pressure and on temperature, respectively, is included in Eq. (4). To achieve this, an appropriate experimental design should include a larger number of processing conditions; there should be sufficient pressure combinations for every temperature to be studied and vice versa. For each processing condition, the reaction rate constant, k, can then be evaluated (fiom plots as illustrated on Figure 1) and the form of E, and E, dependence on pressure
315
Food Science: Posters
Estimated Parameters kref
E,(J/mol) E,(cm3/mol)
Model 95% Con. System Interval 0.00801 M.00263 19812 f13968 -3.54 .283
Pineapple Juice 0.00935 58517 -4.75
95% Con. Interval M.00293 f13316 f1.37
Grapefruit 95%Con. Juice Interval 0.00999 fl.00149 21902 f7170 -4.80 fl .34
and temperature can be determined. In order to achieve more accurate estimations of the kinetic parameters involved in individual experiments, this procedure might necessitate data collection over longer processing times. Ending our discussion, we should mention that the kinetic parameters obtained based on isothermal and isobaric processing conditions, will have to be validated in experiments under controlled, dynamic, ie., variable pressure andor temperature conditions that normally will be encountered in a practical application of the high pressure processing technology. Kinetics of vitamin C degradation (or of other product quality attributes) coupled with the destruction kinetics of microbial and other spoilage agents of concern, under high pressuremoderate temperature processing conditions, can be used for process optimization, leading to safe product with maximum quality retention. 1
0.9
% c
z 0.8 i% 6 0.7 3 0.6 0.5
0.5
0.6
0.7
0.8
0.9
1
C/C,experimental Figure 2. Comparison between predicted and experimental remaining vitamin C concentration values for the model system and the h i t juices studied.
316
High Pressure Food Science, Bioscience and Chemistry
Nomenclature Latin letters C EL3 E“ k n P R T t
vitamin C concentration, mg/l activation energy, J/mol activation volume, cm”/mol (or kT, or kp) reaction rate constant, min-’ for first order reactions order of reaction, dimensionless pressure, MPa universal gas constant, 8.3 14 J/(mol.K) temperature, K (unless otherwise explicitly stated) time, min
Subscripts 0
ref
initial condition reference value (at 323 K and 500 MPa)
Acknowledgment This research was supported by the European Commission (project FAIRCT96-1175).
References Farr, D. 1990. High pressure technology in the food industry. Trends in Food Science & Technology, July 1990, 14-16. Johnson F.H. and Eyring H. 1970. The kinetic basis of pressure effects in biology and chemistry. Ch. 1 in High Pressure Effects on Cellular Processes, A.M. Zimmerman (Ed.), p. 1-44. Academic Press, New York. Knorr, D. 1993. Effects of high-hydrostatic-pressure processes on food safety and quality. Food Technol. 47(6): 156-161. Van Boekel, M.A.J.S. 1996. Statistical aspects of kinetic modeling for food science problems. J. Food Sci. 61(3): 477-485,489. Villota, R. and Hawkes, J.G. 1986. Kinetics of nutrients and organoleptic changes in foods during processing. In Physical and Chemical Properties of Food, M.R. Okos (Ed.), p. 266-366. ASAE, St. Joseph, MI.
Freezing of Potato Cylinders during High Pressure Treatment 0. Schltiter, V. Heinz and D. Knorr Department of Food Biotechnology and Food Process Engineering Berlin University of Technology, Kcinigin Luke Str. 22,D- 14195 Berlin Tel: +49 30 314 71250 Fax: +49 30 832 7663 E-mail: [email protected]
1. INTRODUCTION Freezing of food provides a safe and convenient way of shelf life extension without negative effects on the nutritional quality. Water as the major constituent of most food materials undergoes phase transition in a way which can be influenced by increased hydrostatic pressure (Fig. 1 - data fkom [l]). Beside a depression of the freezing-point a reduced enthalpy of crystalhation can be observed. Therefore, in a range up to 200 MPa fiwzing times of pure water were markedly reduced. Plant tissue like potato cylinders showed a similar behaviour.
Figure 1: Latent heat, volume change and phase transition temperature as hctions of the pressure.
318
High Pressure Food Science, Bioscience and Chemistry
2. MATERIAL AND METHODS
Experimental Setup: Equal sized cylinders (diameter: 28mm, length: 6Omm) were cut from a potato (type Spunta) (Fig. 2). A Pt100-thermocouple was inserted into the center of the cyliider. A second Pt100-thermocouple monitored the temperature at the edge of the cyliider. The complete setup was placed inside of an insulated pressure vessel filled with a mixture of ethanol and glycol . Precooling of the vessel was performed by flexible tubes connected with a thermostat at the outer surface. After sealing of the vessel, pressurisation was started. The samples were frozen during pressure treatment, or after initial pressure treatment at 20°C. The temperature and the pressure were recorded digitally. The flow diagram of the complete unit is presented in Fig. 3.
Figure 2: Schematical drawing of the experimental arrangement. 1: potato cyliider; 2, 3: thermocouples (Pt 100); 4: insulation; 5 : pressure vessel; 6: pressure medium (20% ethanol / 80% glycol, v/v); 7: cooling tubes; 8: upper seal
Food Science: Posters
319
Figure 3: Flow diagram of the complete experimental unit. Freezing Time Prediction: Experimental fieezing curves (Fig. 5 ) were recalculated by a finite difference method [2]. Radial symmetrical one dimensional heat conduction was assumed to describe correctly the situation inside the middle part of an extended potato cylinder (Fig. 4). Additionally, a heat balance was implemented at each volume element to adjust the time dependent temperature field. Convective heat transfer at the surface was calculated by using a transfer coefficient of k = 96 W/mzK. The correct description of the moving hezing-fiont resulted fiom the strong discontinuity of the temperature dependent apparent specific heat capacity cp and thermal conductivity 3L in the vicinity of the fieezing point. Data for temperature dependent cp and h of potato tissue were indicated by [3]. Modifications of these data (Fig. 5 ) were used to fit correctly the experimental fieezing curves. 3. RESULTS AND DISCUSSION
Fitting the experimental hezing curves obtained during high pressure treatment at 50 MPa (B), 100 MPa (C) or 150 MPa @) made it necessary to modify both the literature values [3] of the heat capacity cp and the thermal conductivity A of potato tissue (Fig. 5). It was observed that a 15 min high pressure treatment (p > 50 MPa) at 2OoC produced irreversible changes in thermal conductivity A. In consequence, it was possible to reduce the treatment times during subsequent hezing (Fig. 6). Regressively derived A fiom these fieezing curves were used for fitting the data fiom direct high pressure fieezing experiments.
1
central element:
A V = % . Ay
volume:
Explicite Finite Difference Scheme
Qx
intermediate elements:
'
4=
2 x ' ( x - 0,5'Ax)
!
:f j I x - direction
'
Figure 4: Sectional view of potato cylinder. One dimensional incrementation of concentric volume elements was performed in the half-height of the cylinder. Heat conduction equation was approximated by an explicite finite difference scheme.
Qx
-
boundary element:
Heat Balances
y direction
- -
intermediate elements:
boundary element:
-Ay
Cylindrical Volume Elements
Ay
0 h)
W
Food Science: Posters
321
n 0
B
7 0
5 In 5
9 n Y n 0
9 0
5
n-
5,o
F5 e! 3
n
n
n
0
0
7
7
0
0
5 n 5
. ;
9
9 O r
b f0 r
d
v:
5 n
0
n
U
L
r
O
E
z
4
n
322
High Pressure Food Science, Bioscience and Chemistry
Additionally, apparent specific heat capacity cp had to be corrected as presented (Fig. 5). Whereas the peak value of the cp function was markedly reduced, the general shape of the applied distribution (Weibulf)could be retained. These thermophysical properties were used in the normalized diagram (Fig. 6) where AT between hezing point and medium temperature was set constant. A distinct acceleration of the hezing process was achieved at elevated pressure.
0
500
1000 1500 2000 2500 3000 3500
Freezing Time [s] Figure 6: Fitted thermophysical properties of potato tissue (cp, A) were used to calculate fleezing curves during pressure treatment with equal AT (left). Freezing of potato cylinders afler UHP-treatment (p > 100 m a ) also produced reduced hezing times (right).
Compared to the phase diagram of water (Fig. 7) the crystallization to ice I of the tissue water seems to take place approximately at the same temperature levels (see curve B, C, D). The expected solidification to ice I11 (curve E) could not be
Food Science: Posters
323
Figure 7:Phase diagram of pure water and the fieezing curves of potato cylinders at different pressure levels. identified on the direct way by cooling the potato cylinder to -34°C at 325 MPa. But on the following indirect path we could observe the crystallization of tissue water to ice 111: cooling under pressure (-34"C, 325 MPa), pressure release with sudden temperature increase to phase transition line (ice I, liquid), pressurization with temperature decrease and abrupt temperature increase to phase transition line (ice 111, liquid), fieezing with pressure decrease to ice 111, undercooling, pressure release with recrystallization to ice I (indicated by the jumping of pressure and temperature cause of the increasing specific volume and the latent heat released by recrystallization). 4. CONCLUSION The impact of high hydrostatic pressure on the fkezing behaviour of potato tissues could be demonstrated. Beside the depression of the fieezing-point pressure induced changes in t h e m 1 conductivity h and specific heat capacity cp. As a
324
High Pressure Food Science, Bioscience and Chemistry
result (of these changes and a reduced enthalpy of crystallization) decreased fieezing times during high pressure treatment were detected. By a finite difference method using modified thermophysical data an adaption of the calculated and the experimental freezing curves was possible. Finally the possibility of hezing the tissue water to ice III was demonstrated. To use the advantages of high hydrostatic pressure at subzero temperatures further investigations are underway to improve the knowledge of the complex interrelationships in biological tissue occurring during phase transition of H20.
5.REFERENCES [l] Bridgman, P.W. (1911) Water in the liquid and five solid forms, under pressure, Proc. Am. Acad. A r t s Sci. 47,441-558. [2] Marek, R. and G&, W. (1995) Numerische Msung von partiellen Differentialgleichungenmit finiten Differenzen, Moreno-Verlag, Buchloe. [3]Cleland, A.C. and Earle, R.L. (1984)Assessment of hezing time prediction methods, 3. Food Sci. 49, 1034-1042.
Water Loss and Consistency Reduction in Fruits and Vegetables Treated under High Pressures H. Schoberl, W. Rufl, 1. Wenzel, R. Meyer-Pittroff Lehrstuhl fur Energie- und Umwelttechnik der Lebensmittelindustrie Technische Universitat Munchen in Freising-Weihenstephan 85350 Freising Tel.: +49(0)8161714362 Fax.:+49(0)8161 71 4415 E-mail: [email protected]
Introduction Much has been written on the topic of enzymatic reactions, aromatic substances, chemical reactions and microbiological research in the high pressure treatment of fresh fruits and vegetables. An important part of the effects of hydrostatic pressure has so far remained unrecognized: water loss and the ensuing softening of the products. In this work, the water loss of various types of fruits and vegetables such as apples, potatoes, carrots, asparagus, green peppers, hot peppers, onions and radishes was examined.
332
High Pressure Food Science, Bioscience and Chemistry
Material and Methods Fresh fruits and vegetables (apples, potatoes, carrots, asparagus, green peppers, hot peppers, onions and radishes) served as sample material. The samples were
cut into pieces weighing approximately 50 grams, vacuum-sealed in PE-bags, and then subjected to high pressures. The raising and sinking of the pressure occurred at a linear rate of 200 MPa/minute. The pressure was held for 600 seconds, at temperatures of 20 and 40 "C.
In order to determine the water loss the samples were weighed before the high pressure treatment. After the treatment the free surface water was removed with filter paper and the samples were weighed again. The water loss corresponds to the weight loss,
The loss of solidity was measurec. with a penetrometer, TAXT2, made by Ste e Micro Systems. The measurements themselves were made using a cylindrical piston with a diameter of 5 mm. The rate of measuring was set at 2 mm/s and the depth of penetration was set at 5 mm. The resisting force was measured by the penetration depth and expressed as the slope of the compensating straight line in a range from 0,s mm to
15 mm penetration, in N/mm.
333
Food Science: Posters
Results: Apples Figure 1 shows the water loss of apples treated at high pressures measured directly after the treatment. The water loss rises at both temperatures, set for the experiments, to a pressure of 400 MPa, but sinks again at higher pressures. It is also remarkable that water loss is less at 40°C than at 20 "C. This effect can be attributed to the binding of released water by pectins during the gelatination process.
7
200
400
800
Pressure [MPa]
Fig. 1: Water loss in mass-% in apples treated at high pressures at 20 and 40"C, 1hour after treatment
High Pressure Food Science, Eioscience and Chemistry
334
The fact that pectins play a significant role is also shown by solidity trials conducted on apples (Fig. 2). The solidity decreased with rising temperatures until a pressure of 400 MPa was reached. Turgid pressure, the pressure in the vacuoles of plant cells, is responsible for the solidity of plant matter. With a loss of liquids the turgid pressure also falls, decreasing the solidity. This effect is later surpassed by the pectin gelatination at a pressure of 400MPa. The solidity increased again slightly in the sample treated at 600 MPa.
,
(
5
P n u n [MPa]
Fig. 2: Solidity of apples treated at high pressures (40°C, pressure holding time 600 s), 1hour after treatment
After being cold stored for 14 days the water loss increased yet again, almost entirely correcting the difference caused by the pressure differences. Even the
Food Science: Posters
335
sample treated at 200 MPa, with only slight water loss immediately after treatment, showed strong losses after a 14 day storage period (Fig. 3).
Fig. 3: Water loss in mass-% in apples treated at high pressures at 20 and 40 "C, 14 days after treatment
A comparison with analyses of solidity taken after 14 days of cold storage
(Fig. 4) also showed a large change in the samples treated at 200 MPa. The samples treated at 400,600 and 800 MPa showed no significant changes after 14 days compared to the samples measured directly after treatment.
High Pressure Food Science, Bioscience and Chemistry
336
4,5
,
I
4
-
33
E
3
t0
2,5 2
-g 1,5 1
0,s
0
Fig. 4: Solidity of apples treated at high pressures (40°C, pressure holding time 600 s) 14 days after treatment
Potato: The water loss in potatoes after high pressure treatment is shown in Figure 5.
14
200
400
600
800
Pntasure [MPa]
Fig. 5: Water loss in mass-% in potatoes treated at high pressures at 20 and 40°C, 1 hour after treatment
Food Science: Posters
331
The water loss at 20 "C rises from almost 0 % at 200 MPa to 12 % at 400 MPa. At higher temperatures the water loss is significantly less and falls with increasing pressures, indicating a starch gelatinitation, which binds the water being released.
Carrots: Figure 6 shows the water loss in carrots treated at high pressures. The water loss rises to a pressure of 400 MPa. At higher pressures no further significant changes occur.
12
Fig. 6: Water loss in mass-% of carrots treated at high pressures at 20 and 40 "C, 1 hour after treatment
338
High Pressure Food Science, Bioscience and Chemistry
Other Fruits and Vegetables: An increase in water loss up to 400 MPa, as in carrots, with a levelling out at
this niveau, was also shown in other fruits and vegetables such as asparagus, green peppers, hot peppers, onions and radishes. An overview of the water losses of all examined fruits and vegetables at 400MPa and 20 OC is given in Figure 7.
Fig. 7: Water loss in mass-% of different fruits and vegetables at 400MPa, 20°C and a pressure holding time of 600s, 1 hour after treatment
Food Science: Posters
339
Conclusion: The water loss of various types of fruits and vegetables such as apples, potatoes, carrots, asparagus, green peppers, hot peppers and radishes was examined.
All examined fruits and vegetables showed the highest loss at 400 MPa. At higher pressures most samples showed no changes in water loss. Exceptions were found in potatoes and apples, which showed a lower water loss at higher pressures. The cause for this can be found in the gelatination of pectins in apples and of starch in potatoes. Storage trials in the packaging for a period of two weeks showed an increase in the water loss.
Acknowledgements: This
research
was
supported
by
"Bayerischer
Abfallforschung und Reststoffverwertung" (Bay FORREST).
Forschungsverbund
Molecular Modifications of Ovalbumin upon HP Treatment
S. Iametti", E. Donnizzelli", P. RoverebyG. F. Dall'Agliob, G, Vecchio', F. Bonomi"'
" DISMA, University of Milan, Via Celona 2,201 33 Milano, Italy SSICA, Via Tanara 3 1,Panna, Italy Istituto di Chimica degli Ormoni, CNR,Via Bianco, 9, Milano, Italy
Ovalbumin is the most abundant protein in eggs, and it is among the proteins having the greatest relevance both as a common food ingredient and as a common cause of several forms food intolerance in humans. Its structure is well known [l], and it represents one of the best studied models for understanding
functionality of food proteins in molecular terms [2, 31. Modifications in the different levels of protein structural organization were studied on ovalbumin after treatment at the nominal temperature of 25"C, with pressures ranging from 450 to 900 MPa and for times up to 10 min. Treatments were carried out on the pure protein at different protein concentrations (0.1-10
gA), both at pH 5 (50 mh4 acetate buffer) and at pH 6.8 (50 m M phosphate buffer), and in the presence or in the absence of protecting agents (10% sucrose, 10% NaCl) representing common food ingredients.
Structural modifications of the treated proteins were studied by monitoring the formation of insoluble aggregates and the formation of transiently modified
High Pressure Food Science, Bioscience and Chemistry
326
forms through the measurement of the loss of structure, monitored by CD and fluorescence spectroscopy [4-71.Also studied were the sensitivity to proteases of the treated protein, and the recognition of this latter by suitable antibodies in ELISA-format assays [83. As shown in Fig. 1, solubility of the protein treated in the absence of
protective agents decreased when increasing the length of the treatment, the protein concentration, and the applied pressure. Solubility also decreased at low pH. Addition of either protective agent allowed the protein to retain almost complete solubility even upon the harshest treatment in the least favorable conditions (high protein concentrations, low pH). Fig. 1: Solubility of HP-treated ovalbumin as a function of pressure and treatment time
n
Pressure (MPa)
---
Formation of insoluble aggregates through protein polymerization was completely prevented by blocking the free -SH groups in the protein with a thiol reagent (dithio(bis)2,4-nitrobenzoate [9J) added before the HP treatment [5, 6 1.
Food Science: Posters
321
On the basis of the absorbance readings from the colored anion formed by the thiol reagent upon reaction with thiols (&12=13,600),
2.3 thiol groups became
accessible after treatment at 600-800 MPa. Only 0.5 were accessible after 10 min at 450 MPa, where no aggregates were formed even in the absence of protective agents. These data suggest that polymerization of ovalbumin upon HP treatment likely involved a disulfide-exchange mechanism, in which the free protein thiols made available by transient denaturation played a prominent role. This confirms some similarity in the denaturation mechanism in Hp and thermal treatments, as reported for other food proteins [3,.5,lo]. Fig. 2: Loss of secondary structure in ovalbumin treated for 5 rnin at 600 MPa as a function of added protective agents and of protein concentration
Loss of structural features in the protein fraction that remained soluble after treatment in the absence of protective agents was quite independent of the length
of the treatment, and depended on pH, protein concentration and pressure with a complex pattern. Fig. 2 presents the effects of different pressure on the secondary structure elements in ovalbumin. The two protective agents affected the intensity
High Pressure Food Science, Bioscience and Chemistry
328
of these modifications in a different way. Apparently, limited water activity (as provided by 10% sucrose (-0.3 M)) could prevent extensive protein denaturation less effectively than the ionic strength effects in 10% NaCl (- 1.3 M; a, I0.904), and protein concentration is of paramount relevance to structural modifications occuning in the presence of sucrose. The different conditions used for HP treatment also resulted in pronounced differences in the susceptibility of the soluble fraction of the treated protein to tryptic hydrolysis. Samples treated in the presence of protective agents showed a higher sensitivity to proteases, likely as a consequence of the non-polymeric nature of the protein in these samples. The highest digestibility figures were obtained for ovalbumin (10 mg/mL) treated at 800 MPa for 10 min in the presence of 10% sucrose, likely as a consequence of the higher extent of protein unfolding in these conditions. ~
Table I Recognition of treated ovalbumin by anti-ovalbumin antibodies Protein was treated at 800 MPa for 10 min in the presence of the given additives. Results are given as percent response with respect to the untreated protein. ovalbumin
additives
concentration, gA none
10% NaCl
10% sucrose
2
20
40
23
10
12
32
8
Food Science: Posters
329
The treated protein samples also gave markedly different responses in immunochemical response tests carried out using an indirect, non-competitive ELISA format and antibodies raised against the native protein [ 5 ] . As shown in
Table 1, no difference can be envisioned in the loss of recognition by the same antibodies among ovalbumin samples treated in the absence or in the presence of protective agents. On the contrary, at 10 mg/mL, the sample treated in the presence of 10% sucrose had the lowest recognition by the antiserum used in our studies. This suggests that structural modifications ensuing fiom Hp treatment in the presence of protective agents addressed antibody-recognition sites without leading to pronounced changes in macroscopic properties (e.g., loss of solubility through formation of aggregates) as observed for treatment in the absence of protective agents.
In conclusion, treatment at 600-800 MPa of ovalbumin solutions up to 10 m g / d in the presence of 10% sucrose or NaCl for 5 minutes gave a product that, despite relevant modifications of the some structural features of the protein, retained solubility and had some features of interest in practical terms. Particularly noteworthy were the increased sensitivity of the treated protein to tryptic digestion and the greatly reduced recognition of ovalbumin by specific antibodies. Such features appear promising in view of the use of HF-treated egg proteins as food ingredients for health or specialty foods.
References 1.
Stein PE, Leslie AGW, Finch JT, Carrel RW (1991) J. Biol. Chem. 221: 94 1-959
2.
Yoshinori M (1995) Trends Food Sci. Technol. 6: 225-232
High Pressure Food Science, Bioscience and Chemistry
330
3.
Yoshinori M (1996) J. Agric. Food Chem. 44: 2086-2090
4.
Cairoli S, Iametti S, Bonomi F (1994) J. Protein Chem. 13: 347-354
5.
Iametti S, De Gregon B, Vecchio G, Bonomi F (1996) Eur. J. Biochem. 237: 106-112
6.
Iametti S, Transidico P, Bonomi F, Vecchio G, Pittia P, Rovere P, Dall’Aglio G (1997) J. Agric. Food Chem. 45: 23-29
7.
Iametti S, Transidico P, Bonomi F, Vecchio G, Pittia P, Rovere P, Dall’Aglio (1997) In “High pressure research in the biosciences and biotechnology” (K. Heremans, ed.) Leuven University Press, Belgium pp 415-418
8.
Turin L, Bonomi F (1994) J. Sci. Food Agric. 64: 39-45
9.
Ellman G, (1 959) Arch. Biochem. Biophys. 82: 70-77
10. Roefs SPFM, De Kruif KG (1994) Eur. J. Biochem. 226: 883-889
Leuven,
Plenary Lecture Influence of Pressure-assisted Freezing on the Structure, Hydration and Mechanical Properties of a Protein Gel H. Barry,E. M. h a y and J. C. Cheftel* Unit6 de Biachimie-Technologie Alimentaires, Universitd des Sciences et Techniques, F-34095 Montpellier cedex 05, France
INTRODUCTION Due to renewed interest in high pressure technology (2), the phase diagram of water under pressure (1) has been reexamined in view of potential food applications (4, 8). Since water remains liquid until -22°C at a pressure of 207.5 MPa, pressureshift bzing, pressure-assisted thawing and non frozen storage under pressure are all possible in the 0 to -22°C range. For pressure-shift hezing, a biological sample is cooled under pressure to a temperature just above the melting temperature of ice at this pressure. Upon sudden pressure release, supercooling of water present in the sample enhances heterogeneous ice nucleation (12), inducing the formation of a large number of small stable ice nuclei. These give rise, after crystal growth, to a large number of small ice crystals, thus avoiding the damage to tissue and cell structures caused by large crystals with resulting drip and altered texture after thawing. Pressureshift fieezing may also lead to a uniform size distribution of ice crystals throughout the sample, since the initial supercooling is quasi-dorm throughout sample depth. To maintain these advantages, however, the latent heat of crystallization should be removed quickly, and this requires a low temperature or a stirred cookg medium and/or a small sample size. Low temperature h z e n storage would be mandatory, to avoid ice recrystallization. Some investigators of pressure-shift fieezing have determined b z i n g curves as a function of process parameters (3, 11). Tofu frozen by pressure release from 200 MPa (at -18OC) to 0.1 MPa contained smaller ice crystals than tofu frozen by air blast (9). Upon thawiug there was no drip, and the resulting tofu kept its initial shape and texture. In another study (7), pieces of tofu were maintained under pressure at 1%l0C for 45 or 90 min. No ice formed at 200 MPa, while type I ice formed at 100 MPa, ice I11 at 340 ma, ice V at 400,500 or 600 MPa, and ice VI at 700 MPa. Upon pressure release, the sample brought to 200 Mpa underwent nucleation and fkezing, while samples bm above 200 MPa were probably converted to ice I. Control samples were b z e n in air at 0.1 MPa and -20,-30 or -8O"C, or pressurized without hezing.
344
High Pressure Food Science, Bioscience and Chemistry
After thawing at 2OoC, samples differed fiom controls (increased f m e s s and strain, altered texture) in the following decreasing order : air blast at -20°C > 100 MPa > air blast at -3OOC > 700 MPa > 600 MPa > 500 MPa. Samples fiozen at 340 MPa, or by pressure release fiom 200 to 0.1 MPa were almost the same as untreated tofu. CryoSEM indicated that samples fiozen at 200, 340,400, 500 or 600 MPa contained the smallest ice crystals (10-20 pm). Raw or blanched carrot slices were fiozen under the same pressure and temperature conditions (6). As for tofu, firmness, strain and structure remained closer to those of control samples after fieezing at 200 MPa (pressure-shift), 340 MPa or 400 MPa than after fieezing at other pressure levels. Potato cubes (8 cm3) were cooled down to -12°C at 400 MPa, avoiding the formation of type V ice (10). Upon sudden pressure release, the potato cube briefly decreased to -17.5"C before reaching -2.5"C due to the latent heat of crystallization. The cooling medium (glycoYethano1 80/20 v/v) also decreased fiom -14 to -30°C. This cooling effect of decompression enhanced ice nucleation and further freezing. Upon thawing, pressure-shift fiozen potato cubes displayed less drip loss, browning, texture change or cell damage (SEM) than controls that had been fiozen in air at -3OOC. In the present study, a heat-set gel of &lactoglobulin was used as a model food and subjected to different freezing processes, including pressure-shift fieezing from 207 to 0.1 MPa. Freezing kinetics were assessed. Since gels are often destabilized by fieezing and thawing, the effects of the different freezing processes on the structure, hydration and mechanical characteristicsof thawed gels were compared.
MATERIALS AND METHODS Preparation of 0-Lg isolate solutions and of heat-set gels The 0-lactoglobulin (D-Lg) isolate fiom sweet whey (Besnier-Bridel) was the same as in previous studies (5). It contained 58 g kg' moisture, and per kg of dry solids, 5.6 g nonprotein nitrogen (NPN), 859 f 4 g protein [(total N - NPN) x 6.38)], 50 g ash, 0.4 g calcium, < 10 g fat, and ca. 40 g lactose. It also contained 890 g native D-Lg and 20 g native a-ladbumin per kg protein. A solution of D-Lg isolate, pH 7.0, was repeatedly prepared in degassed deionized water, that contained 14.1 g dry solids and 12.0 g of protein per 100 mL. For preparing heat-set gels, 60-75 mL of 0-Lg solution were placed in a polyvinylidene chloride tubing (d = 26 mm; thickness = 50 mm; Krehalon). Several tubings were heated in a water bath at 87OC for 45 min, cooled and kept 14 h at 4°C. High pressure (HP) vessel and cooliig system The HP vessel (stainless steel with internal cooling circuit ; 350 MPa maximum; internal volume = 1 L ; internal diameter = 80 mm) was fiom ACB. It was cooled by circulation of an ethanoYwater mix, itself cooled with an external cryostat (RK8Ks, Lauda). The pressure-transmitting and cooliig medium inside the HP vessel was a 55/45 v/v 1,2-propanedioVwatermix maintained at -33°C. The HPvessel wasequipped
Biochemistry: Presentations
345
with a 0-500 MPa pressure gauge (PR-811-F, Asco Instruments) and 3 thermocouples (J type). One of these indicated the temperature of the pressure-transmittingmedium, while the 2 others, used to measure gel temperature, were connected to a recorder. Thermocouples were calibrated and their precision was estimated at f 0.5"C. Pressure-shiftfreezing of gels A BLg gel prepared in a PVDC tubing (diameter a 25 mm ; length = 120 rnm ; weight 60-75 g) was equilibrated at 20°C. An upper silicon stopper was equiped with two thermowells (stainless steel ; external diameter = 2 mm ; i n t d diameter = 1.2 mm ; length = 75 mm ; model GKM-13009-CO75, Ellab) in which thennocouples were inserted. The tips of the thermocouples were located 45 mm down the gel, one at gel center (i.e. = 12.5 mm ftom gel surface), and the other near (2-5 mm) gel surface. The gel cylinder was immersed into the propanedioYwatermix (at -33°C) and the pressure was raised to 207 MPa within 80 s. The gel was kept for about 17 min at 207 MPa in the liquid at -33°C until its near surface temperature reached -19 or -20°C. Pressure was then released to 0.1 MPa within 6-8 s. The gel was further kept for about 9 min in the propanedioVwatermix until &zing was completed and the near surf'ace temperature reached -21°C (-20°C at center). Immediately afterwwds (to exclude ice recrystallization during storage), the gel was thawed by immersion in stirred water at 2OoC during 2.5 4 and stored at 4°C overnight. Freezing of gels by other processes Some 8-Lg gels were b z e n as above in the propanedioYwater mix at -33"C, except that pressure was brought to 200 MPa and maintained at this value for about 70 min, until the gel center first reached a fhezing plateau at -21.5"C, then decreased to -30°C. Pressure was released at this point. Such fhezing conditions lead to the formation of type I ice (1). B-Lg gels were also frozen by each one of the following processes at 0.1 MPa : in still air at -43OC for 24 h ; in gaseous N2at -80°C for about 20 min, by immersion in a propanedioYwater mix at -33°C for about 17 min. All h z e n gels were immediately thawed as indicated above. Control gels Non pressurized non fkozen heat-set gels of RLg were stored at 4°C for 40 h and used as controls. Pressurized non ftozen control gels were also prepared by subjecting l3-Lg gels to 207 MPa for 17 min at 20°C. Gel analysis Gels were equilibrated at 20°C in their tubing before analysis. Spontaneous exudation was determined and expressed as g liquid per 100 g of initial gel. Dry solib were determined on 1 g of drained gel. Results represent the mean value (iSJftom 3 determinations and are expressed as g of dry solids per 100 g of drained gel. Wuter holding capacity was measured on pieces of drained gel slices. The weight of gel pieces was measured before and after centrifugation at 370 x g for 5 min on a polyethylene grid disk (20 pm pores, Phannacia). Results represent the mean value
High Pressure Food Science. Bioscience and Chemistry
346
(*Q from 4 determinations and are expressed as g of water remaining in the gel per 100 g of water initially present in the drained gel. Mechanical characteristics of gels were determined with an LFRA Texture Analyzer (Stevens) using a flat cylindrical Gel cyliinders compression probe (d = 51 mm) at a displacement speed of 0.2 mm it. (d = 25 mm and h = 20 mm) were prepared with a razor blade. Gel rigidity was defined as the force F per cm2measured at 10% compression. Relaxation time (min) and an elasticity index (expressed as elasticity) were also measured (13). For each gel fkom measurementson 3-4gel cylinders. sample, results are the mean value The structure of gels (thawed in stirred water at 2OOC) was examined by cutting slices about 0.5 mm thick, placing them on a glass slide, and observing immediately with a stereo-micrOscope (model SZ-4045,Olympus) and at magnifiing ratios of 9-15.
(a
Statistical analysis
Each type of fkezing process was assessed on 3 4 gels fiozen and thawed independently on 3-4 different days. The hydration or mechanical characteristics of these 3-4 gels were calculated as mean values (~th). The characteristics of an o v e d non pressurized non h z e n control gel were calculated as the mean value (*Q of 8 control gets prepared and analyzed at the same time as fiozen gels. The pressurized (207 MPa, 17 mh, 20°C) non h z e n control gel (mean values) was statistidy compared to the overall non pressurized non b z e n control gel.
RESULTS AND DISCUSSION Kinetics of gel freezing Three Werent fleezing processes have been studied by immersing gel samples in a propanediollwater mix at -33"C, and measuring temperatures at gel center and near gel surface. The fkeezing curve at atmospheric pressure is given in Fig. la and that for pressure-shift hezing in Fig. lb. Freezing under pressure is reported in Table 1. In Fig. la, the gel center reached -2°C 9-10 min after gel immersion. This was followed by heterogeneous nucleation (at -2"C), and by a 6 min freezing plateau at -1 to -2°C. The hezing point depression is due to the presence of small amounts of lactose and salt, and to a slight fieeze-concentration effect. The gel center then reached -20°C in about 3 min. Near gel surface, the propanediollwater mix fully absorbed the latent heat of crystallization, and no freezing plateau was observed. Pressure-shift freezing is shown on Fig. lb. The small temperature increase observed at gel center upon reaching 207 MPa was due to the quasi-adiabatic compression. A cooling period of about 16 min was necessary to reach -2OOC near gel surface (2 mm in this case). This did not induce significant freezing since hezing of water under 207 MPa takes place at -22°C. After reaching -2OOC near gel surface (-18°C at gel center), pressure was released fkom 207 to 0.1 MPa within 6-8 s. This had a marked cooling effect on the propanediollwatermix, which reached -38°C (increasing slowly afterwards). Ice
347
Biochemistry: Presentations
-
25 20
-20 -_ -25
0
2
25 20 15 10 5
$
0
5
10
15
20
25
30 Time(min) 250
2oo 150
$ w
5
100 2
-5 -10 -15 + -20 -25 Q)
50
0
5 10 15 20 25 30 rime(min) Figure 1. Freezing kinetics of D-Lg gels in propanediol/water at -33°C. (a) Freezing at atmospheric pressure. Gel weight = 63 g ; (b) pressure-shift freezing from 207 MPa (-20°C near gel surface) to 0.1 MPa. Gcl weight = 69 g. Time 0 is counted about 20 s after gel immersion in the propanediollwatermix at -33°C.
0
nucleation and the resulting release of latent heat brought the temperature at gel center and near the surface to -1°C 6-8 s after the start of pressure release (Fig. Ib). It is likely that the high degree of supercooling of gel water upon pressure release enhanced heterogeneous ice (type I) nucleation throughout gel depth. Assuming adiabatic nucleation at -20°C (and a latent heat of water crystallization at -20°C of 24 1 J g'),the latent heat released by nucleation can be taken as equal to the sensible heat absorbed by ice crystals and by the remaining liquid water as they pass from -20 to -1°C : m x kvs,= [m x C,ice x AT] + [(I-m) x C,water x AT] where m is the weight proportion of gel water converted to ice during pressure release, C,ice (at -20°C) = 1.95 J g 1 K-I and C,water (at 0°C) = 4.22 J $' K-'. With a AT of 19"C, m = 0.28. Other investigators (1 l), however, suggest that quasistatic adiabatic expansion of liquid water from 210 MPa and -22°C to 0 MPa and 0°C should result in m = 0.36.This significant amount of ice may conespond to a very large number of very small ice nuclei, or to a smaller number of ice nuclei that have already undergone some crystal growth. The crystal size distribution throughout gel depth is not known.
High Pressure Food Science, Bioscience and Chemistry
348
Table 1. Duration of successive cooling/freezing steps for the different gel freezing processesa
cooling Undercooling
I
Freezing plateau
Further cooliig
0 20°C -1°C
19.0
Gaseous N2at -8OOC
I
20°C
63.1-
-1°C
-1°C
-20OC
6.4
20°C -1°C -I "C -2°C -20°C PropanedioVwater 2.7 2.4 5.7 62.6 6.5 at -33OC & 0.1 MPa 20°C -1°C -21.5" -30°C PropanedioVwater at 64.4 4.2 -33OC & 200 MPab -1°C -20°C 20°C -1°C Propanediohater at 4.3 -33OC & 207 to 0.1 MPa' 66.8 a Time values and weights represent the average of 3 to 7 experiments I type of fkeezing process Temperature at gel center at the beginning and end of each step Freezing under pressure ' Pressure-shift freezing from 207 to 0.1 MPa (pressure was released at -19°C near gel surface)
The exact path followed on the pressure/ temperature phase diagram of water is also unknown, although non equilibrium conditions probably prevail and do not preclude that the gel comes near atmospheric pressure while being still close to -20°C (9, 12). Once the gel reached -l0C, additional crystal growth took place, the latent heat of crystallization being progressively removed by the propanedioY water mix at = -38°C. An = 8 min freezing plateau at -1°C was observed at gel center, while the temperature near gel surface decreased progressively (Fig. lb). From the end of the freezing plateau, gel center and near surface reached -20°C in 4 and 2 min, respectively. A comparison of Fig. l b and l a indicates that the freezing plateau (measured here from initial freezing at gel surface till the end of the plateau) was slightly shorter after pressure release (= 8 min plateau) than for freezing at a constant pressure of 0.1 MPa (= 10 min plateau). To interpret this difference, it is necessuy to take into account : I ) the weight difference between the two gels (69 vs 63 g) ; 2) the AT between the propanedioYwater mix and the gel at the freezing plateau (37°C vs = 31"C, due to cooling upon pressure release) ; 3) the time interval between initial freezing at gel surface and at gel center (At = 0 vs 4 inin) ;4) the fact that = 28% of the initial liquid water already freezes during pressure release. The kinetics of gel freezing in propanedioVwaterat -33°C under a pressure of 200 MPa maintained throughout the whole process are not shown.About 20 min were required to cool the near surface of the gel to -22.5"C. This reveals some undercooling, since equilibrium freezing of gel water at 200 MPa took place at -21.5"C. The freezing plateau was = 30 min at gel center, due to the small AT (about 11SoC) between
Biochemistry: Presentations
349
propanediollwater and the gel. After the end of the freezing plateau at gel center, 18.5 min were necessary to reach -30°C at gel center. Pressure was then released to atmospheric within 6-8 s : the 1-2°C decrease in gel temperature due to decompression indicated that most of the freezable water was frozen at this point. Table 1 summarizes the successive cooling and freezing steps for the different processes studed. Values represent the average of 3 to 7 experiments. The weight of gels should be taken into account when comparing cooling or freezing rates. Freezing in still air at -43°C and in gaseous N, at -80°C were the slowest and fastest processes, respectively. While freezing under pressure (200 MPa) was relatively slow, as indicated above, cooling under pressure (from 20 to -1"C, at 207 MPa) was systematically faster than cooling at atmospheric pressure. This may be due to an increased density and heat conductivity of the gel (and of the propanollwater mix) under pressure. Structure of thawed gels Low resolution photographs of gels sliced soon after thawing are presented in Fig. 2. The non pressurized non frozen control gels had a smooth, white, translucid and glossy appearance, and a uniform structure (Fig. 2a). Their smooth texture resembled that of heat-gelled egg white. Pressure-shift frozen gels were quite similar to non pressurized non frozen control gels, except that the gel matrix was sectioned by a pattern of cracks evenly distributed throughout the gel (Fig. 2b). These cracks appear to correspond to the marks of aligned ice crystals (diameter = 20-70 p).Sections between cracks apparently contained much smaller ice crystals (d = 3-7 pm). The mechanism of formation of these cracks is not known, but may be related to gel expansion when pressure was released. Gels frozen in propanediollwater at -33°C either at 0.1 or 200 MPa, and gels frozen in gaseous N,, differed from control gels by a spongious and grating behavior when handled for slicing, and a dull and granular appearance once sliced. Their texture was clearly not smooth. The gel matrix contained numerous round or flattened alveoles of varying size (Fig. 2c, d, e). It is likely that these corresponded to melted ice crystals. They were set out in concentric layers (Fig. 2 4 e), probably due to temperature gradients during freezing. No concentric arrangement was visible in gels frozen under pressure (Fig. 2c). Gels frozen in still air at -43OC had a dull and opaque appearance, and the most heterogeneous structure (Fig. 2f) of all gels, at the magnification used. The gel matrix was partitioned into zones, containing large parallel andor concentric alveoles. Open cavities, I 2 mm in diameter, were also present. This structure appeared to reflect the direction of heat removal and the marks of large ice crystals, and was probably responsible for the strong exudation and poor water holding capacity of these gels (see next section). Hydration and mechanical characteristics of control and thawed gels The hydration and mechanical characteristics of the overall non pressurized non frozen control gel and of gels frozen by 5 distinct processes are given in Fig. 3.
350
High Pressure Food Science, Bioscience and Chemistry
Figure 2. Structure of l3-Lg gels frozen by various processes, as examined with binocular lenses after thawing at 2OOC. (a) non pressurized non frozen control gel ; (b) pressure-shift freezing in propanedioV water at -33°C from 207 MPa (-19OC near gel surface) to 0.1 MPa ; (c) freezing in propanedioV water at -33°C and 200 MPa ; (d) in propanedioV water at -33OC and 0.1 MPa ;(e) in gaseous N, at -8OOC ;( f ) in still air at -43OC. Bars equal 1 mm.
35 1
Biochemistry: Presentations
The overall control gel underwent very little exudation (= 1% w/w) and had a high water holding capacity. Non h z e n control gels pressurized at 20°C (207 MPa, 17 min) displayed no significant difference in water holding capacity or mechanical properties, compared to the overall control gel, but revealed a higher exudation (2.1 g per 100 g) and dry solid content (1 5.1 %) (data not shown). In most cases, the hydration characteristics of thawed gels differed fiom those of the overall control gel (Fig. 3a). Spontaneous gel exudation varied fiom 2 to 8 g per 100 g of initial gel and appeared to reflect the degree of gel destructuration by fieezing and/or thawing (Fig. 2). Freezing in still air at -43°C induced most exudation, while fieezing in propanediollwaterat -33°C and 0.1 MPa was less damaging (Fig. 3a). The exudate always contained about 5% dry solids, and therefore the dry solid (and protein) content of drained gels increased in relation to the amount of exudate. The water holding capacity (WHC), i.e. the percent of water in drained gel retained after centrifugation,should be high when prior exudation is strong. This was not the case
N2/ PGI PGI PGI AkI -8O"CI -33"CI -33"CI207 -33"CI -43"CI 0.1 MPa 0.1 MPa toO.l MPa 200MPa 0.1 MPa capacity (g water in centrifiged geV100 g water in drained gel) ~ r solids y ($1100 g +d.ge!) OSpontaneous exudation (g liquid/lOOg initial gel)
Overall control
Overall control
j
N21 Akl PGI PGI PGI -8O"CI -33"CI -33W207 -33"CI -43"CI 0.1 MPa 0.1 MPa toO.l MPa 200MPa 0.1 MPa 0Rigidity (glcm2) Elasticity 0Relaxation time (min)
1
Figure 3. Influence of various fieezing processes on (a) hydration, and (b) mechanical characteristics of B-lactoglobulin gels. Mean values from 3 to 8 independent gels.
352
High Pressure Food Science, Bioscience and Chemistry
for gels fiozen in still air at -43"C, which had a WHC well below that of the control gel. This low WHC was due to major structural damage (Fig. 2f, Fig. 3a). A pair comparison between fkezing processes according to Student's test for each hydration characteristics made it possible to assign these processes to sigdicantly different classes (p 5 0.05) by order of increasing exudation (or dry solids of drained gels), and decreasing WHC. Overall, three classes can be differenciated : 1) gels fiozen in still air at -43°C were the most different fiom the control gels. Their marked exudation and low WHC probably resulted fiom slow fieezing (Table l), formation of large ice crystals, and increased protein aggregation through freezeconcentration.Upon thawing, an altered gel matrix with large cavities (Fig. 3f) did not retain water ; 2) gels frozen at 0.1 MPa in the propanediollwater mix at -33"C, or in gaseous N2 at -80°C. Due to fast freezing, their structure (Fig. 2 4 e) and gel matrix were less modified, and exudation upon thawing was moderate. However, their WHC was somewhat reduced ; 3) gels frozen at 200 MPa or by pressure release displayed a moderate to strong exudation with a resulting increase in the dry solid (and protein) content of drained gels. This, and the better preserved gel structure (Fig. 2b, c) may explain why their residual water was well retained after centrifugation(Fig. 3a). The measurement of the mechanical characteristics of thawed gels was not independent fiom exudation, since gels with an increased dry solid content tend to have higher rigidity. In addition, the large liquid cavities (Fig. 2f) of gels fiozen in air at -43°C certainly disturbed behavior under compression. The mechanical characteristics of gels were not modified as systematically by fieezinglthawingas their hydration characteristics (Fig. 3b). Gels h z e n in still air at -43°C displayed an increased rigidity (after thawing) partly because of a higher dry solid content. The decreased gel elasticity and relaxation time were most probably due to the extensive structure modifications and cavities. Gels fiozen in gaseous N2 at -8OOC were not significantly different fiom the overall control gel (Fig. 3b). The same can be said for gels frozen in propanedioYwater at -33°C and 0.1 MPa. Gels fiozen by sudden pressure release from 207 to 0.1 MPa, or frozen under pressure (at 200 m a ) , displayed after thawing significantly higher rigidity than the overall control gel. The elasticity of gels fiozen under pressure was significanfly lower than that of the overall control gel. Both the elasticity and the relaxation time of pressure-shift fiozen gels were similar to those of this control gel (Fig. 3b). CONCLUSIONS For cylindrical gels (d = 25 mm, weight = 67 or 63 g) immersed in a pressureand heat-transmitting medium at -33"C, pressure-shift fieezing fiom 207 MPa (and -19OC) to 0.1 MPa required more time than freezing at atmospheric pressure. However, quasi-instantaneous nucleation taking place upon pressure release appeared to induce the formation of small ice crystals throughout gel depth, in contrast to all
Biochemistry: Presentations
353
other freezing processes studied. This was observed by examination of thawed gel slices. Calculations indicated that, during pressure release, the initial gel supercooling (AT = 19'C) absorbed the latent heat of crystallization of about 28% of the gel water. Pressure-shift freezing caused more gel exudation and increased more gel rigdity after thawing than freezing in the same cooling medium at atmospheric pressure. However, the residual water-holding capacity, the elasticity and the relaxation time of pressureshift frozen gels remained close to those of non frozen control gels. In spite of these contrasted results, subjective texture assessement indicated that pressure-shift fiozen gels were the closest to the non frozen control gels, compared to gels frozen by all other processes studied. This points to the need for systematic sensorial texture evaluation to further investigate the possible advantages of pressure-shift fieezing for freeze-sensitive foods. REFERENCES 1. Bridgman, P.W. 1912. Water, in the liquid and five solid forms, under pressure. Proc. Am. Acad. Arts Sci., 47: 439-558. 2. Cheftel, J.C. and Dumay, E. 1997. Les hautes pressions: principes et potentialit&. In La Conservation des Aliments, p. 197-215. Lavoisier Tec et Doc, Paris, France. 3. Denys, S., Van Loey, A,, De Cordt, S., Hendrickx, M. and Tobback, P. 1997. Modelling of high pressure assisted freezing and thawing. In High Pressure Research in the Biosciences and Biotechnology, K. Heremans (Ed.), p. 351-354. Leuven University Press, Leuven, Belgium. 4. Deuchi, T. and Hayashi, R. 1992. High pressure treatments at subzero temperature: application to preservation, rapid freezing and rapid thawing of foods.In High Pressure and Biotechnology, C . Balny, R.Hayashi and K. Heremans (Eds), p. 353-355. John Libbey, Montrouge, France. 5. Dumay, E. M., Kalichevsky, M.T. and Cheftel, J. C. 1997. Characteristics of pressure-inducedgels of B-lactoglobulin at various times after pressure release. Lebensm. Wiss. Technol., in press. 6. Fuchigami, M., Katoh, N. and Teramoto, A. 1996. Effects of pressure-shift freezing on texture, pectic composition and histological structure of carmts. In High Pressure Bioscience and Biotechnology, R.Hayashi and C. Balny (Eds), p. 379-386. Elsevier, Amstardam, The Netherlands. 7. Fuchigami, M. and Teramoto, A. 1996. Texture and cryo-scanning electron micrographs of pressure-shift f r o m tofu. In High Pressure Bioscience and BiorechnoZogv, R. Hayashi and C. Balny (Eds), p. 41 1-414. Elsevier, hmterdam, The Netherlands. 8. Kalichevsky, M.T., Knorr, D. and Lillfird, P.J. 1995. Potential food applications of high-pressure effects on ice-water transitions.Trends Food Sci. Technol., 6: 253-259. 9. Kanda, Y., Aoki, M. and Kosugi, T. 1992. Freezing of tofu (soybean curd) by pressure-shift freezing and its structure.Nippon Shokuhin Gakkaishi, 38: 608-614. lO.Koch, H.,Seydehelm, I., Wille, P., Kalichevsky, M.T. and Knorr, D. 1996. Pressure-shift freezing and its influence on texture, color, microstructure and rehydration behavior of potato cubes. Nahmg, 40: 125-131. 1l.Otero, L., Sanz, P.D., De Elvira, C. and Carrasco, J.A. 1997. Modelling thermodynamic properties of water in the high-pressure-assisted freezing process. In High Pressure Research in the Biosciences and Biotechnology, K. Heremans (Ed.), p. 347-350. Leuven University Press,Belgium 12.Sahagian, M.E. and Goff,H.D. 1996. FundamWl aspects of the freezing process. In Freezing Efects on Food Qualify, L.E. Jeremiah (Ed.), p. 1-50. Marcel Dekker, Inc., New York. 13.Zasypkin, D.V., Dumay, E. and Cheftel, J.C. 1996. Pressure- and heat-induced gelation of mixed D-lactoglobulin/xanthan solutions.Food Hydrmlloids, 10 : 203-211.
Studies on Bacterial Spores by Combined High Pressure-Heat Treatments: Possibility to Sterilize Low Acid Foods
P.Rover$, S.Golab', A.Maggib, N.Scaramuzzaband L.Migliolib
"ABB Industria, v.le Edison 50, Sesto San Giovanni, Milano (Italy) bStazione Sperimentale per l'hdustria delle Conserve Alimentari, v.le Tanara 3 1/A, Parma (Italy).
INTRODUCTION
High pressure technology can be applied to treatment of acid foods (ex: h i t juices, fruit in pieces) and of foods having characteristics similar to IMF (ex:
fish cream, vegetable cream). In the first case it is necessary to destroy only microorganisms spoiling acid foods (lactic acid bacteria and mycetes) because the low pH of these products
(with the exception of tomato juices and peeled tomatoes) prevents spore germination; normally 500 MPa treatments for some minutes at ambient temperature are sufficient to destroy these spoilage microorganisms. In the second case the
f d q
have %-pH combinations inhibiting bacterial
spores; 600-700 MPa for some minutes at ambient temperature are necessary to stabilize this kind of foods. In low-acid foods, however, the complete inactivation of spores is necessary to
produce sterile foods at ambient temperature. Some studies on spore inactivation
355
Biochemistry: Presentations
by HP-heat treatments were performed (1-7) but kinetics inactivation and heat parameters (D and z values) under pressure are not always defined. We carried out some combined treatments on Bacillus and Clostridium spores suspended in model systems and in foods to give a contribution to these studies.
Fig. 1: SEM image of B.cereus spores treated at 500 MPa for 5 min at 20°C in phosphate buffer
MATERIALS AND METHODS Microbial Strains The following microbial strains were used: 1. Bacillus cereus SSICA isolated from wheat flour; 2. Bacillus lichenifomis SSICA isolated from spices; 3. Bacillus coagulans SSICA isolated from tomato-based tuna sauce; 4. Bacillus stearothemophilus SSICA isolated from canned peas;
5. Clostridium botulimm type A, ATCC 19397; 6. Clostridium spomgenes P A 3679, ATCC 7955. The spores were suspended in distilled water before performing the trials.
High Pressure Food Science, Bioscience and Chemistry
356
Substrates Distilled water (for C.botulinum strain), phosphate buffer at pH 7 (for Bacillus strains) and beef broth, (for C.sporogenes strain) were used. The beef broth was prepared in our laboratory with the following composition: beef extract 5%, yeast extract 1%, bacto peptone 2%, pH 7. Each substrate was inoculated with 105-106spores/ml for each strain.
No = initial spore alncentration Ns= HP w i v i n g spores
Tab. 1: Effect of HPP + heat on some Bacillus sp. spores Culture media Surviving Bacillus spores were counted in Tryptone Soy Agar (Biogenetics) after incubation at 30°C for 3 days (for the two mesophilic strains), at 45°C for 4
357
Biochemistry: Presentations
days (for the B.coagu1ans strain) and at 55°C for 3 days (for the
Bsteamthennophiius strain). Clostridium spores were counted in M5 medium (8) incubated at 30°C for 5 days. HP eauipments
An isostatic press mod. QFP-6 (ABB-Pressure System) working up to 900 MPa and another isostatic press (ABB- Pressure Systems) able to reach 1500 MPa were used. The second plant was equipped with a suitable temperature control system able to keep it constant, at a preset value, throughout all the treatment. The trials were performed at: a- 700,800 and 900 MPa for 1,3,5 and 10 min each at 20,50,60 and 70"C, for
Bacillus spores; b- 800 MPa for 1,3,6 and 9 rnin at 65°C for C.botulmum spores;
c- 1000, 1200, 1400 and 1500 MPa for 5 min each at 20°C; 600, 800, 1000, 1200 and 1400 MPa for 5 rnin each at 80°C; 600,800, 1000, 1200, 1400 and 1500 MPa for 5 rnin each at 90°C for C.spomgenes spores (first repetition);
-
d- 800 MPa - 75°C for 2,4,6,8 and 10 min, 900 MPa 72°C for 2,4,6,8 and 10 min, for C.spomgenes spores (second repetition); e- 600 MPa-90°C for 5,10,15 and 20 min, 600 MPa-100°C for 2 , 4 and 6 min, 600 MPa-108°C for 0.5,2 and 4 min; f- MPa-93°C for 3,6,9 and 12 min,700 MPa-100°C for 0.5, 2 and 4 min, 700 MPa-108°C for 0.5,2 and 4 min; g- MPa-93°C for 1, 2, 4 and 6 min, 800 MPa-98°C for 0.5, 2 and 4 min, 800 MPa-108°C for 0.5,2 and 4 rnin (third repetition).
358
High Pressure Food Science, Bioscience and Chemistry
In the first and in the second trials on Csporogenes the temperatures considered are the initial ones inside the sample basket put into the press, while the temperatures of the third test are those reached in treatments under pressure. RESULTS AND DISCUSSION
In the tests performed at 20°C only B.cereus spores were sensitive to HP: the 900 MPa treatment for 10 min reduced the number of viable spores from 6x105
to 2x1o2/ml.The same treatment was ineffectiveon the spores of the other three strains.
D
-
3.45045
m
-0,2898
ds
0,05316
R
0.90832
r
0,95306
2 --
0
2
4
6
8
10
12
time (min)
Fig. 2: C.botulinum spores in distilled water treated at 800 MPa - 88°C The simultaneous increase of pressure and temperature allowed all spores of B.cereus, B.lichenifomis and B.stearothemophilus strains to be inactivated. B.cereus and B.lichenifomis needed 800 MPa for 5 min at 6 0 T , while B.stearothemophilus needed 700 MPa for 5 min at 70°C. B.cougu1un.s proved to be the most pressure resistant strain, since four decimal reductions were obtained with 900 MPa for 5 rnin at 70°C (Tab.1). SEM image
359
Biochemistry: Presentations
of B.cereus spores treated at 500 MPa for 5 min at 20°C in phosphate buffer shows a broken exosporium (Fig. 1). In these trials increase of the temperature applied was obtained by thmostating
the sample at the same temperature as that of the press vessel; adiabatic heat was then let to be exchanged with the vessel wall during the processing period. C.botulinum spores in distilled water were preheated at 65°C and treated at 800
MPa with the ABB QFP-6 press. The press vessel was preheated at 88°C (the same temperature as that reached by the sample at 800 m a ) . In these working conditions a decimal reduction time value @) of 3.4 min at 88°C was obtained
pressure@lPa)
HPvessel temp. ("C)
initialsamp. temp. ("C)
fdsamp. temp. ("C)
nD-lsec.
nD-30m.
800
60
60
53
4.3
4.5
800
84
60
66
%.7
%.7
500
70
70
64
0
0.3
500
84
70
71
2.1
4.7
-
360
High Pressure Food Science, Bioscience and Chemistry
obtained 4.3 decimal reductions while in the second total inactivation (about 7 decimal reductions) was reached (Tab. 2). It is important to notice that substrate temperature is inmased by pressure (about 3°C every 100 MPa for water) but , at the same time, this heat is easily lost by the product because of heat transfer to the press walls, if these are not properly heated. This phenomenon could lower the effect of the treatment for
Fig. 3: Example of a HP process (1200 MPa-80°C for 5 min) performed on controlled conditions (a= pressure; b=intmal heater temp.; c-meat broth samples temp)
small presses and results in lack of uniformity in the big ones. Because of this reason we performed the following trials in a new experimental press where it was possible to adequately control the pressure and the temperature values of the sample during the process. An example of a HP process (1200 MPa-80°C for 5 min) performed under these conditions is
Biochemisrry: Presentations
361
reported in the diagram (a= Pressure; b= internal heater temp.; c=meat broth sample temp) (Fig. 3). In this case the temperature of meat broth was increased by about 4°C every 100 MPa by the pressure.
During the Same trials we observed that Cspmgenes spores were not destroyed MEAT BROTH initial FO temperature (sterilizing of the sample
sample 1 (cfdml)
1500
'
lo00 1200 1400 1500
800 lo00
1
++-+ ++-p IOCI
0.1
sample 2 (cfdml)
3x10:
0.5~10
60
I
1.3~10:
1.4 x 10
1.4 x 10
0
0
0.0052
0.0102
80
I
0.61
1200 1400
800 lo00
6.74
1200
0
1400
0
1
0 0
Tab. 3: C. sporogenes spores HPP+heat exposed. by high pressure alone: in fact, 1500 MPa at 20°C for 5 min did not affect the initial number of viable spores. Nevertheless combined treatments (HP-T) determined inactivation of spores: 5 min of exposition at 800 MPa-9O0C, 1000 MPa-80°C or 1400 MPa-60°C caused destruction of lo5 sporedml in the meat broth (Tab 3). After all the previous processing combinations the samples were sterile.
362
High Pressure Food Science, Bioscience and Chemistry
z
17,926OC
0,004
0,996
Tab. 4: Csporogenes spores: thermal parameters under HP conditions. Other similar treatments were performed on Csporogenes spores: 600, 700 and 800 MPa at 93, 100 and 108°C ( values under pressure); kinetics of spore
inactivation at 600,700 and 800 MPa were determined. The thermal parameters (D and z) were changed by pressure levels. In fact, Dl10 was equal to 13.3 min at 0.1 MPa while Dlloat 800 MPa was equal to 0.54 min.
Biochemistry: Presentations
363
In another set of treatments performed at 800 MPa- llO°C (temperature under pressure) we confinned that D value under pressure was lower (DIlo=0.8) than at ambient pressure.
CONCLUSIONS These preliminary tests enabled us to assume the use of heat treatment under pressure conditions as a technique more effective than the conventional retort one because it minimized thermal damage. BIBLIOGRAPHY
1- A.J.H. Sale, G.W. Godd, W.A. Hamilton, J. Gen. Micmbiol.,60,323 (1970). 2-G.W..Gould, A.J.H. Sale,J. Gen. Micmbiol.,60,335 (1970). 3-D.G. Hoover, C. Metrick, A.M. Papineau, D.F. Farkas, D. Knon, Food Technol., 43(3), 99 (1989). 4- I. Seyderhelm, D. Knom, ZFX (J.FoodZndustry),43(4), 17 (1992). 5- I. Hayakawa, T. Kanno, M. Tomita, Y.Fujio, JFoodSci., 59,159 (1994).
6-M.F. Patterson, D.M. Margey, G. Mills, R. Simpson, A. Gilmour, in “High Pressure Research in the Biosciences and Biotechnology’’ ed. K. Heremans, Leuven University Press, Leuven, Belgium, p. 269, (1997).
7- A. Maggi, S. Gola, P. Rovere, L. Miglioli, G.Dall’Aglio, N.G Lonneborg, Industria Conserve, 71,8 (1996).
8-A. Casolari, in “Proceedings of the 4th International Congress of Science and Technoloav)’, vol. 3, p.86, Valencia (Spain), 1974. 9-A. Maggi, N. Scaramuzza, P. Rovere, S. Gola, Industria Conserve, 72, 49 (1997).
Effects of Pressure on Minor Groove Binding to DNA Studied by REPA and Fluorometric Methods G.-Q. Tang, S. Kunug~*and N. Tanaka Laboratory for BioPolymer Physics, Department of Polymer Science and Engineering, Kyoto Institute of Technology, Sakyo, Kyoto 606, Japan.
As well characterized, perturbation by high hydrostatic pressure is a very useful tool to study the mechanisms of the interactions between biological molecules. For enzymes-substratesand protein-protein interactions, many examples have been reported in the series of the EHPRG meetings and other occasions. One of the key processes very important in the elucidation of the biologcal functions is the recognition of nucleic acid sequences and structures by low and high molecular weight substances, fiom antibioticsto regulation proteins, and rather limited number of examples of hgh pressure studies have been reported for the processes including nucleic acids. We have published that the high pressure perturbation resulted in 11 recovery of the relaxed sequence recognition by restriction endonucleases, caused by solvent or ionic strength perturbation, which implicated the strengthening of thc enzyme-DNA interactions under high pressure.* To know the details of such an idea, here we have studied the DNA-low molecular weight ligand interactions under hgh pressure. For DNA-binding ligands, some studles have been done for intercalating reagents such as ethidium bromides, and here we focused on the behavior of DNA-minor groove binding reagents, which recognize successive base sequences. For example, dlstamycin A (Dst) binds to at least four successive N T bps in the B-DNA minor groove, without causing peat distortion of the DNA structure. At least two bindmg modes as 1:1 (at four bp A.T sites) and side by side 2: 1 (at five 01six bp A.T sites) ligand: DNA complex have been clearly resolved. Close van deiWaal's contacts with the groove floor, created by its crescent shape (Fig.1) antl structural torsional freedom, and hydrogen bonds between its amides and N-3 of A and 0-2 of T in the minor groove primarily contribute to their binding specificity antl af€in~ty. The negative charge potential in A.T-rich minor groove is probably a factol-
365
Biochemistry: Presentations
in attracting drugs like distamycin A (Dst), as well as 4',6-diamidino-2phenylindole (DAPI)and Hoechst 33258 (Hst), which have positively charged ends. DAPI was found to bind to G.C region versus intercalation, also, with relatively weak binding aflinity. One of the main difficulties of studying DNA-antibiotics interaction is in the signal collectingprocess. Except for some limited number of the cases, interactions cause no optical mformation. To overcome this we have applied so-called "restriction endonuclease protection assay (REPA). "his methodology has been used to study DNA minor groove binding since the 1970s. It has contributed to the rational design of drug analogous with great sequence specificities. A principle governing REPA is that the simultaneous occlusion of cleavage site by ligands inhibits an attack by restriction endonuclease. For Hoechst 33258, its fluorescence is sensitive for DNA sequences and the binding of this drug to synthetic DNAs, namely poly[d(A)].poly[d(T)] and poly[d(A-T)] .poly[d(A-T)], was studied by fluorometry.
d: y*
DisIamycin
60
7-f~~ Figure 1 Three DNA minor groove binding drugs with site preferrence for dA.dT stretches.
Materials and Methods pBR322 DNA was purchased from Boehnnger and binding drugs were from Sigma or Wako. Restriction enzymes (EcoRI, BglI, or HindIII ) were obtained from Toyobo. Poly [d(A)].poly [d(T)] and poly [d(A-T)].poly[d(A-T)] were purchased fi-om Sigma and Pharmacia, respectively. Hoechst 33258 was from Wako. Other reagents were commercial and used without further purifications.
366
High Pressure Food Science, Bioscience and Chemistry
pBR322 DNA was incubated with corresponding drugs at various molar ratios of drug to DNA bps (r') for 30 minutes at room temperature; then digested by enzymes in a reaction buffer until enough EDTA was added to end the digestion after 1 or 2 hrs. The digestion product was assayed in a 1.0% agarose gel. Gel stained with ethidium bromide was photographed; analyzed by using NIH Image 1.59 software. High pressure vessel constructed by Yamamoto Suiatsu Co. (Toyonaka, Japan) was used for high pressure incubation. The time to increase the pressure to the final value was about 2 min, relatively short coinpared with the reaction time (normally 1 or 2 hs.) Except for the extraneous pressure, other conditions were the same as for O.1MPa. Titration of Hst in Tris-C1 buffer (pH 7.2) containing 100 mh4 NaCl and 1 mM EDTA were processed by additions of concentrated solutions of poly [d (A)J.poly [d(T)] or poly [d(A-T)].poly[d(A-T)]. Increases in the 460 nm-fluorescence intensity ( A.ex = 360 nm) of Hoechst upon the DNA-association in ambient and elevated pressures are monitored in an optical pressure vessel (Teramecs, Kyoto) equipped in the sample chamber of a spectrofluorometer (RF5000, Shimadzu) with a circulating temperature bath. The exclusion-site model was used for data simulation (binding size n= 3.2 or 4.0, cooperative parametero= 0.5 or 0.7). The solution volume changes induced by the compression were corrected. Results REPA Method --- The digestion of supercoiled pBR 322 DNA at GAATTC by the EcoRI enzyme directly produces full-length linear DNA. Under 120MPa, the drug-induced enzyme mhibition showed dose-dependence (given in serial r'= [drug]/[DNA bps]), as in controls, while the extent was increased significantly. Pressure caused decreases of two r' limits at whch Dst provided the least and complete protection by a magnitude of 2 and 1 at least within 2 hr, respectively. Experiments at other pressures also found similar enhancements in EcoRI enzyme protection. The drug concentration needed to produce 50% protection (Pc50)was notably reduced fiom about 60 p M at O.1MPa to 0.8 p M at 200 MPa. In the absence of Dst, we did not find the measurable EcoRI enzyme protection by pressure alone, which coincides with a previous study, in which the pressure of 200 MPa led to no observable differences in the k,ffi of the EcoRI cleavage of several DNA samples including pBR 322 DNA.Z The generality of the pressure effect observed above was also examined on
367
Biochemistry: Presentations
other systems with structurally different drugs including Hst, DAPI and ActD. The pressure dependence of protection against the EcoRI enzyme by DAPI and Hst was weaker than by Dst, coinciding with previous observations and has no clear correlations with their apparent binding constants. Pressure also induced a moderate protection increment when intercalator ActD penetrated to GpC of the HindIII enzyme cognate sequence (AAGCTT). Similar pressure effects were also found for the EcoRV and BglI enzymes when Dst, Hst and DAPI were used. Fluommetric Method Fluorescence titration processes of Hst with the two types of dA.dT polymers showed opposite pressure effects on the binding of this drug. Pressure was shown to decrease the binding affinity of Hoechst for the homopolymer while it was shown to increase the affinity for the alternating polymer. This may reflect their distinct structural and hydration characteristics.The latter trend has the similarity to the plasmid DNA, shown above. The alternating polymer is considered to act more like a normal DNA while the homopolymer acts some abnormally. According to the standard thermodynamic relationship between pressure and equilibrium binding constant (( d lnIGsd d P). =- AVo/RT), the molar volume change (AVO) upon the DNA-ligand association was calculated. A positive AVO value (+6.1k 1.2 cm3/mol ) was obtained for the drug binding to the poly[d(A)].poly[d(T)] duplex, while a negative AVO (-3.4 f0.3 cm3/mol) was obtained for poly[d(A-T)].poly[d(A-T)]. The reaction volumes determined by these two methods are summarized in Table 1.
-
Table 1. Summary of reaction volumes upon drug binding to DNA. determined by two methods.
DNA
method Dst
A V - ~(mVmol) Act -lo* -50*
DAPI
**
pBR322
F2EPA
p(dA.dT)
Fluorescence
--
--
--
+3
Fluorescence
--
--
--
-
(dT)
-55
* Roughly estimated from the data at 0.1 and 120 MPa. ** Too small to be determined by REPA method.
368
High Pressure Food Science, Bioscience and Chemistry
Discussions Plasmid/Drug --- The present results suggest that some vital factors/steps in this ternary system (arbitrarily not including Mg2') induced negative volume changes thus were pressure sensitive. Assuming the existence of a simplified association between the EcoRl enzyme inhibition and the competitive DNA-binding of Dst to the EcoRI cognate site, as well as the unchanged DNA-binding mechanism under pressure, we have estimated the binding free energy (AG,,,) of -12.4 kcdmol at atmospheric pressure. The present result was comparable to previously reported data measured. Such an association also allows us to calculate an apparent volume change (AV,,) of about -55 ml /mol in the binding process. It corresponds with a negative volume change for netropsin binding to synthetic A.T copolymer as obtained by a density measurement. The fact that no direct pressure-induced protection from enzyme activity (including EcoRI, EcoRV, I-hdIII and BglI) occurred within the used pressure ranges allows us to consider two other possible pressure effects definitely on this system: the improvement of the drug-DNA complex stability andor the inhibition of enzyme activity for the DNA-drug complex. Since drug agents were introduced to form a ternary system, the competition from Dst for the EcoRl cognate site and further changes of local DNA conformation created the strong possibility of new pressure liability of enzyme activity (e.g., an unstable enzyme complex with drug-DNA or a pressure- unfavored catalflc step). The main factors involved in the Dst interaction with DNA (van der Waal's contacts and hydrogen bonds) are among the candidates that are responsible for the hgher pressure stability of the Dst-DNA complex. At low r' values or at high ionic strength, strong and specific binding to an NT-affluent sequence is predominately achieved by nonionic forces. At high r' values or at low ionic strengths, charge interactions also mediate the weaker and unspecific binding to regions with fewer A.T bases. The former two nonionic factors usually lead near zero and small negative volume changes thus are favored by pressure elevations only slightly. In contrast, the ionic interaction is strongly unfavored by an increase in pressure. However, the effects of DNA screening of small cations may be more liable for pressure elevations in h s system. The role of hydrostatic pressure in enhancing molecular solvation is probably applicable to the DNA-drug complex formation. There have been examples, including of several restriction enzymes, where the pressure sensitivity was related to changes in molecular solvation. As is known,the hydration extent of DNA plays a key role to form a distinct DNA structure.Unlike the inaccessible hydration sites
369
Biochemistry: Presentations
in protein interiors, DNA-hydration water is accessible and forms H-bonds with bulk water. Such hydration variants have been widely used to explain differences in thermodynamic observations of the interaction of ligands and DNA with simple sequence.It was shown that the spine of hydration in the minor groove of synthetic DNA fiagments was replaced by bound netropsin, and hydration changes with the binding (magnitude and direction) were strongly dependent upon DNA sequences. A study using natural DNA suggested that 50 extra bound water molecules occurred in a complex formation with a netropsin analogue. This means that the binding increases the general exposure extent of DNA-drug complex. Great similarities between netropsin and Dst in molecular structures and DNA binding mechanisms made it reasonable to believe that pressure elevations would favor the Dst binding to the pBR322 DNA areas in hydration, which is compatible with large and negative AVlpp observed. Hst/dA.dT polymer Different results for two polymen suggest that the binding of Hst on alternative polymer leads to a less hydrated drug-DNA complex, mainly as a result of a general loss of bound water or a decompression of the hydration shell around the DNA lattice upon association, and that on homopolymer duplicate is associated with an overall uptake of water molecules. There are two contradictory views concerningthe molecular origins of volume changes of minor groove drug binding between the two dA.dT polymers. One view primarily considers the inherent hydration state differences of the DNA lattices, and the other arguably weighs more of the differences concerningthe drug-DNA complexes. Despite such contradictory explanations, a comparative value (AAVO) fiom theAVO data of the two dA.dT polymers under the same conditions is unanimously agreed upon to reflect their inherent hydration distinctions more reasonably. Our result of 9.5 ml/mol for hydration change differences between the homopolymer and the alternating polymer corresponds to a much larger hydration capacity of the dA.dT homopolymer. The value is also in agreement with literature A AVO data derived from the binding of netropsin with the two dA.dT polymers in comparable conditions probed by different methods @om 6.9 to 11.5 cm3/mol).
-
References 1 S.Kunugi, ProgPolyrnSci. 18,805-838 (1993). 2 HKabata, A.Nomura, N.Shimamoto, and S.Kunugi, ,J.Mol.Recogn. 7, 25-30 (1 994).
Acquired Resistance of Microorganisms to Inactivation by High Hydrostatic Pressure
P. Verroens, K. Hauben and C. Michiels Laboratory of Food Microbiology, Katholieke Universiteit Leuven, Heverlee, Belgium Tel :+32/16/32.15.79 Fax:+32/16/32 19 97 E-mail : [email protected]
Introduction
Recently we demonstrated that the non-pathogenic Escherichiu coli strain MG1655 can develop high levels of barotolerance by spontaneous mutations
'.
We anticipated that this phenomenon could form a serious threat for highpressure processing of foods, assuming that pathogens and/or spoilage organisms different fiom MG1655 can also acquire resistance.
In this study, we investigated the potential to develop resistance against pressureinactivation under laboratory conditions for a wider range of microorganisms including some pathogens.
' Haulyen et a/, 1997, wid and Environmental Microlnology63(3) : 945-950.
Biochemistry: Presentations
37i
Materials and methods
Strains and media : Stationary phase cultures of the following organisms were obtained in 20 ml Tryptic Soy Broth (TSB) at 37°C with shaking (200rpm) : Escherichia coli ABG4-1, a chloramphenicol resistant derivative of MG1655, the non-sporulating Bacillus subtilis strain CW274, Yersinia enterocolitica, Salmonella typhimurium
and S. enteritidis, Shigellaflexneri and S. sonnei and Escherichiafergusonii and E. hermanii. Saccharomyces cerevisiae was grown in YPD-medium.
Selection : Cells were harvested by centrifugation (4,000 x g, 5 min), and resuspended in the same volume of potasium phosphate buffer (10 mM, pH 7). The suspension was packed in a heat-sealed sterile polyethylene bag and subjected to the desired pressure for 15 minutes at 20°C. The treated suspension was used for viability determination and for inoculation in a fresh medium (at 1/100 dilution) and grown up for the next pressure cycle. After the last selection step, a single colony was isolated and designated the selected strain. Biochemical tests (Enterotube, Becton Dickinson, Meylan, France) were performed to verig that no pressureresistant contamination was selected.
372
High Pressure Food Science, Bioscience and Chemistry
Results
Selection for different high pressure resistant microorganisms For E. coli ABG4-1, pressure resistance increased after each step, so that pressure could be gradually increased in steps of about 50 MPa. The final population obtained after 9 selection steps showed a survival rate of 1 % after 15 minutes at 600 MPa compared to
% for the parent strain. However, for some
tested organisms (S. cerevisiae, B. subtillis and Yersinia enterocolitica), no increase in pressure resistance could be selected after six or more rounds (Fig. 1). Escherichia coli ABG4-1
Yersinia enterocolitica
1
8M)U
8
moo
8000
8
7
moo 6000
c 5 E
c
k 5000
4
E3
4WO
2
P
3000
1
2000
0 1
2
3
4
5
6
7
8
9
cycle
Bacillus subtilis CW214
Saccharomyces cerevisiae myc2 8000
7
7000
8
m
z5
WOO
Y 43
8
4000
3000
2
2000
1
1
2
3
4
5
8
cycle
Figure 1 : Stepwise selection for pressure resistance in Escherichia coli ABG4-I, Yersinia enterocolitica, Bacillus subtilis and Saccharomyces cerevisiae myc 2. Points on the dotted line indicate the applied pressure (20 "C, 15 min). Bars indicate the corresponding reduction factor. "NA" indicates that the reduction factor was not determined.
313
Biochemistry: Presentations
Some other selected organisms that were more closely related to E. coli (Salmonella typhimurium and S. enteritidis, Shigella jlexneri and S. sonnei, E. fergusonii and E. hermanii) survived slightly (50-100 MPa) higher pressures than their parent strains after also 9 selection rounds (data not shown). For Salmonella typhimurium and Shigella flexneri, further selection was applied and resistance could be slowly but significantly increased as shown in Figure 2.
cycle
Figure 2 : Selection for pressure-resistant mutants of Salmonella typhimurium and Shigella Jexneri. Points on the dotted line indicate the applied pressure (20 "C,15 i n ) . Bars indicate the corresponding reduction factor. V N A " indicates that the reduction factor was not analysed.
High Pressure Food Science, Bioscience and Chemistry
314
After 25 selection cylces for Shigella and 32 for Salmonella, the selected strain was compared to the parental for high pressure inactivation at different pressures (Fig. 3)
pressure (MI'.)
pressure (MP.)
Figure 3 : High-pressure inactivation (20 "C, 15 min) of the parental strain Shigellaj7exneri (left) and salmonella typhimurium (right) compared to the corresponding selected strain at different pressures.
To achieve a similar reduction in 15 minutes at ambient temperature, the selected resistant mutants required a pressure that was at least 250 MPa (for Salmonella) and 300 MPa (for Shigellu) higher than their respective parent strains. Finally, we attempted to select pressure resistance in two adhtional E. coli
strains (Fig. 4).
While strain 1A acquired very rapidly extreme levels of
resistance, similar to strain AE3G4-1, repeated attempts to select for a pressureresistant mutant of the strain ATCC 11775 remained Without succes.
375
Biochemistry: Presentations
Eschenchia coli A TCCl1775 BOO0
0000
moo 6000 ....
moo 4000
4000
3000
moo
2000 000 (
2
3
4
5
0
7
1
2
3
4
5
0
7
Figure 4 : Selection for high-pressure resistant mutants of Escherichzu coli 1A (left) and ATCC 11775 (right). “NA” indicates that the reduction factor was not analysed.
Difference in mutation 6equency of both strain AJ3G4-1 and ATCC11775 was estimated by comparing the occurrence of rifampicin and streptomycin resistance in both strains. The mutation 6equency in strain 1A was 3 to 8 times higher than in strain ATCC 11775(data not shown). It will be further investigated wheter the higher mutation frequency can explain the more rapid selection of high pressure resistance in strain 1A and ABG 4-1.
Conclusions We conclude that also enteric pathogens like Salmonella and Shigella can develop high levels of barotolerance, that could form a safety problem for high pressure pasteurisation. However, the risk that such selection would take place in a high pressure processing plant is difficult to quantify. There seem to exist large differences down to the strain level in the ability of different bacteria to rapidly acquire pressure resistance under selection.
Kinetics of Refolding of P-Lactoglobulin after High-pressure Treatment Measured by Reactivity Towards Ellman’s Reagent H.Stapelfeldt*, R.E.Marller and L.H.Skibsted Food Chemistry, Department of Dairy and Food Science, Royal Veterinary and Agricultural University, DK- 1958 Frederiksberg C, Denmark INTRODUCTION P-lactoglobulin,the major whey protein of bovine milk, is very sensitive to high hydrostatic pressure. Based on intrinsic fluorescence data, we have determined the reaction volume for the process P-lactoglobulin (N) * P-lactoglobulin (D) to be -98 ml-mol“,corresponding to a half denaturation pressure of 1 10 MPa (Stapelfeldt et al., 1996). This places P-lactoglobulin among the most pressure-sensitive food proteins yielding a large potential for pressure-modifying this protein in milkbased systems while leaving other proteins unaffected. For P-lactoglobulin, high pressure has mainly been investigated as an alternative to heat-setting of gels, and for equal protein concentrations, pressure-induced gels have been shown to be more firm than heat-set gels (Van Camp & Huyghebaert, 1995),while increasing temperature for the pressure treatment decreases the temperature needed for gelsetting (Van Camp et al., 1996).These results are partly explained by the reactivity of the “fiee” thiol group (Cys-121;Monaco et al., 1987), which is largely increased by unfolding of the protein by heat (Sawyer, 1968)and pressure (Tanaka et al., 1996). Pressure-induced aggregation has recently been shown to be caused by thiol /disulfide interchange reactions (Funtenberger et al., 1997). The effect of high pressure has generally been measured in situ or at one fixed time after pressure release. We have shown that the reactivity of trypsin towards pressure-treated P-lactoglobulin persists for several hours after pressure release (Stapelfeldt et al., I996), which indicates slow refolding of the trypsin-labile domains of the protein. Also tyrosine in pressurized P-lactoglobulin is more susceptible to oxidative dimerization following pressure treatment (0stdal et al., 1996). Ellman’s reagent (5,5’-dithiobis(2-nitrobenzoic acid); Ellman, 1959) is widely used for quantification of thiol-reactivity, and it is as such used to characterize heat treatment (i.e. heat denaturation of P-lactoglobulin) of milk. We have therefore applied Ellman’s reagent for comparing thiol reactivity in heat- or
311
Biochemistry: Presentations
pressure-treated P-lactoglobulin solutions at fixed times after these treatments and for monitoring loss of thiol-reactivity with time after pressure treatment (Mnrller et al., 1997).
MATERIALS AND METHODS Solutions of P-lactoglobulin (0.1 1 mM in 50 mM Tris buffer, pH 7.6 1, I=O. 16 (NaC1)) from MD Foods Ingredients was used without further purification for pressure and heat treatments. Pressure treatment was carried out in closed polyethylene tubes in a thermostatted PSIKA pressure cell for 30 min at 15 "C, and heat treatments were carried out in a water thermostat. Static measurements of thiol groups were determined as described in Stapelfeldt et al. (1997), while dynamic measurements were performed with a DX- 17MV stopped-flow spectrofluorimeter (Applied Photophysics Ltd., London, U.K.) using 412 nm monitoring wavelength.
RESULTS AND DISCUSSION In native P-lactoglobulin, Cys-121 is buried in the hydrophobic interior of the protein and displays very limited reactivity towards Ellman's reagent. As may be seen from Figure 1, pressure causes unfolding of P-lactoglobulin which increases the reactivity of the aforementioned thiol group measured as absorbance at 412 nm of the yellow-coloured product of the disulfide interchange reaction between Ellman's reagent and P-lactoglobulin. 1.50 1.25 1 1.00 1 N
8
0.75
-
0.50 0.25 1
0.00 0.110
0.165
0.220
0.275
Concentration of p-lactoglobulin (mM)
Figure 1. Increase in Ad,*following reaction between Ellman's reagent and three different concentrations of P-lactoglobulin pressurized for 30 min at 15 "C.
High Pressure Food Science, Bioscience and Chemistry
378
Interestingly, the results in Figure 1 shows significant deviation from LambertBeer's law which may indicate denatured states at higher pressures at which the accessibility of Ellman's reagent to the thiol is lower than at intermediate pressures. Static measurement of reactivity towards Ellman's reagent of pressurized solutions of P-lactoglobulin showed unfolding, as displayed in Figure 2A. The pressure-dependence of denaturation revealed by static measurements was, however, not parallelled by the dynamic measurements, as may be seen from Figure 2B.
0
100
200
300
Pressure (MPa)
400
0
100
200
300
400
Pressure (MPa)
Figure 2. A. Content of reactive thiol groups of pressure-treated P-lactoglobulin (at 15 "C for 30 min) measured relative to an L-cysteine standard by Ellman's reagent. B. Relative rate of reaction between pressure-treated P-lactoglobulin (at 15 "C for 30 min) and Ellman's reagent for conditions of excess of the latter using stopped-flow kinetic measurements of A4,*. As may be seen from Figure 2, the transition in the rate by which the thiol reacts with Ellman's reagent, measured by stopped-flow absorption spectrometry, occurs at approximately 50 MPa, in contrast to the transition in exposure of the thiol group, measured statically as degree of reaction with Ellman's reagent, occurring around 120 m a . Applying the simple two-step denaturation model for pressure denaturation and using the data from the static measurements, a reaction volume of -61 f 3 ml-mol-', corresponding to a half-denaturation pressure of 140 MPa, may be calculated. Notably, a comparison of static and dynamic measurements of thiol reactivity for P-lactoglobulin subjected to different 5-min heat treatments showed half-denaturation at 60-65 "C, as determined by reaction rate, but at 75-80 "C as determined by degree of exposure.
319
Biochemistry: Presentations
The pressure unfolding and increase in thiol reactivity for pressurized samples was found to be reversible and refolding followed first-order kinetics, resulting in less reactivity with time after pressure treatment, as may be seen from Figure 3.
0.4
0.3 0.2
0.1
0
5
10
15
20
25
30
Time (h)
Figure 3. Refolding of a 0.1 1 mM solution (Tris buffer, pH 7.61, P0.16) of plactoglobulin pressurized at 200 MPa at 15 OC for 30 min, measured as thiol exposure by reaction with Ellman’s reagent at 20 “C at regular intervals after pressure release. The full line is calculated according to A(t) = a + b-exp(-l&.t) using non-linear regression analysis. The observed decrease in reactivity and refolding of p-lactoglobulin was only slightly dependent on the temperature, thus having an apparent energy of activation of 20 f 2 kl-mol-’ for p-lactoglobulin pressurized at 200 MPa. Repeated highpressure treatment of samples resulted in increasing thiol reactivity and subsequent loss of reactivity at ambient pressure for at least two pressure cycles. This indicates that the decreased reactivity towards Ellman’s reagent with time after pressure treatment, at least for the protein concentration used, was due to refolding and not caused by thiol oxidation forming disulfide bonds.
Acknowledgements Our work has been sponsored by the F0TEK programme of the Danish Ministry of Research through LMC Center for Advanced Food Studies.
380
High Pressure Food Science, Bioscience and Chemistry
REFERENCES Ellman, G.L. (1959) Arch. Biochem. Biophys. 82,70-77. Funtenberger, S.; Dumay, E. & Cheftel, J.-C. (1997) J. Agric. Food Chem. 45,912921. Monaco, H.L.; Zanotti, G.; Spadon, P.; Bolognesi, M.; Sawyer, L.; Eliopoulos, E.E. (1987) J. Mol. Biol. 197,695-706. Msller, R.E.; Stapelfeldt, H. & Skibsted. L.H. (1997) J. Agric Food Chem., submitted. Sawyer, W.H. (1968) J. Dairy Sci. 51,323-329. Stapelfeldt, H.; Petersen, P.; Kristiansen, K.R.; Qvist, K.B. & Skibsted, L.H. (1996) J. Dairy Res. 63, 111-118. Stapelfeldt, H.; Bjermm, K. & Skibsted, L.H. (1997) Milchwissenschaft 52, 146149. Tanaka, N.; Tsurui, Y; Kobayashi, I. & Kunugi, S. (1996) Int. J. Biol. Macromol. 19,63-68. Van Camp, J. & Huyghebaert, A. (1995) Lebensm. Wiss. Technol. 28,111-1 17. Van Camp, J.; Feys, G. & Huyghebaert, A. (1996) Lebensm. Wiss. Technol. 29, 49-57. 0stda1, H.; Daneshvar, B. & Skibsted, L.H. (1996) Free Rad. Res. 24,429-438.
High Pressure Inactivation of Polyphenoloxidase: Effect of pH and Temperature
C. Weemaes, L. Ludikhuyze, I. Van den Broeck and M. Hendrickx* Laboratory of Food Technology, Katholieke Universiteit Leuven, Kardinaal Mercierlaan 92, B-300 1 Heverlee, Belgium
1. INTRODUCTION
Polyphenoloxidase (PPO) is responsible for browning of damaged h i t s and vegetables. The enzyme catalyses the conversion of phenolic compounds to o-quinones, which subsequently polymerize non-enzymatically to brown pigments (1-2). Apart from this color deterioration, enzymatic browning results in the
development of off-flavours and a loss of nutritional value (1-3). Inactivation of PPO seems to be the most effective method to inhibit enzymatic browning (2-3). The objective of this study was to characterize the pressure inactivation of polyphenoloxidases from different origins and to determine whether the pressure stability of PPO can be altered by changing the pH of the surrounding medium. Moreover, the combined pressure-temperature inactivation of PPO was studied and a mathematical pressure-temperature inactivation model was formulated.
382
High Pressure Food Science, Bioscience and Chemistry
2. MATERIALS AND METHODS 2.1. ENZYME AND MEDIA Polyphenoloxidases were extracted from apples, avocados, grapes, pears and plums and were partially purified and lyophilized. The obtained powders were dissolved in McIlvaine (‘Mb’, pH 4-6) or 0.1 M phosphate buffer (‘Pb’, pH 6-8).
2.2. PRESSURE TREATMENT Isobaric experiments were carried out in a thermostated multi-vessel high pressure apparatus (Resato, The Netherlands). Enzyme samples, contained in micro centrifuge tubes, were pressure treated for pre-set times (t).
2.3. ACTIVITY MEASUREMENT Enzyme activity (AOD/min) was measured spectrophotometrically and was calculated from the linear part of the absorption curve by linear regression.
2.4. DATA-ANALYSIS The inactivation data were processed into kinetic parameters, i.e. inactivation rate constants (k), activation energies (E,) and activation volumes (AV*).
383
Biochemistry: Presentations
3. RESULTS AND DISCUSSION
3.1. PRESSURE STABILITY OF POLYPHENOLOXIDASES AT THEIR PH-OPTIMUM Pressure processing (0.1-900 MPa) at room temperature caused inactivation of apple, avocado, grape and pear PPO, but no inactivation of plum PPO. Pressure -2
induced inactivation of apple, avocado, grape and pear PPO at room temperature
-3
could be described by first-order decay processes. From Fig. 1 it is clear that the minimum pressure needed for inactivation of PPO depends on the origin of 550
650
750
850
950
the enzyme and that the activation
pessure (MPa)
volumes for apple, avocado and grape Fig I: Pressure inactivation of apple (*j, avocado (o), grape (A) and pear (#) PPO
PPO are in the range -22 to -33 cm3/mol.
3.2. EFFECT OF P H ON THE PRESSURE STABILITY OF POLYPHENOLOXIDASE
For one of the polyphenoloxidases studied, namely avocado PPO (3 mg/ml), the effect of pH on the pressure stability was evaluated. At pH 5-8, the pressure inactivation of avocado PPO followed first-order kinetics. At pH 4, however, deviation from first-order kinetics was noticed. The biphasic curves were
384
High Pressure Food Science, Bioscience and Chemistry
considered as the result of two first-loglinear decreases, which
were
both
-3
analysed in terms of k and AV*. From
p -
h
-4
Fig. 2 it can be concluded that the pressure inactivation is dependent on the
-5
pH of the surrounding medium. The -6
enzyme is more pressure sensitive at 450
550
650
750
850
950
Pressure (Mh)
Fig. 2 : Pressure inactivation of avocado PPO : pH 4 (1st (0)and 2nd (*) phase), 5 (A), 6, (Mb (0); Pb (0)), 7 (#) and 8 (e)
lower pH. From Fig. 2 it is also clear that for pH values between 5 and 8
h V * I decreases with increasing pH.
3.3. EFFECTOF TEMPERATURE ON THE PRESSURE STABILITY OF POLYPHENOLOXIDASE For one selected system (avocado PPO; 0.5 mg/ml; pH 7 ) the effect of temperature on the pressure stability was studied. Hereto, about hundred combinations of pressure (P) and temperature (T) (0.1-900 MPa; 25-77.5 "C) were investigated as a hnction of time. It was noticed that, for all P-T combinations studied, the inactivation of PPO could be described by first-order kinetics and that there is an antagonistic effect of pressure and temperature when 'low' pressures (5 250 Mpa)
385
Biochemistry: Presentations
are combined with high temperatures (2
62.5 "C). This antagonistic effect is also apparent from the shape of the P-T kinetic diagram, which indicates P-T combinations, characterized by the same inactivation rate constant (Fig. 3). O L -
'
At all pressures studied, the temperature
7-
2 0 3 0 4 0 5 0 6 0 7 0 8 0 Temperature
("c,
Fig. 3 : P-T kinetic diagram for avocado PPO (k = 0.005 and 0.015 min-' for inner and outer line)
dependence of the inactivation rate constant could be expressed by the Arrhenius equation.
The activation energies, derived from this equation, decrease with increasing pressure (from about 320 at 0.1 MPa to about 60 kJ/mol at 900 Mpa). The pressure dependence of the inactivation rate constant could not always be described by the Eyring equation. Only for pressures exceeding 300 MPa, the activation volume was constant with pressure. Based on the P-T inactivation data, a mathematical model for the P-T inactivation of avocado PPO was formulated.
386
High Pressure Food Science, Bioscience and Chemistry
EZ, n, a, b, c and d were respectively estimated as 324.302 kJ/mol, 1 6 . 8 ~lo4 MPa-’, -2.42, -17.2 x 10” MPa-’, 41.1 x
MPa-’ and -23.3 x
MPa-3.
4. ACKNOWLEDGMENT This research has been supported by the Flemish Institute for the promotion of scientifictechnological research in industry (IWT), KULeuven Research Council (project OT/94/19), National Fund for Scientific Research (project G.0189.95) and the European Commission (project FAIR-CT96- 1 175).
5. REFERENCES 1. Vamos-Vigybo, L. 1981. Polyphenol oxidase and peroxidase in h i t s and vegetables. Crit. Rev. Food Sci. Technol., 11,341-346.
2. McEvily, A.J.; Iyengar, R.; Otwell, W.S. 1992. Inhibition of enzymatic browning in foods and beverages. Crit. Rev. Food Sci. Technol., 32,253-273. 3. Golan-Goldhirsh, A.; Whitaker, J.R.; Kahn, V. 1984. Relation between structure of polyphenol oxidase and prevention of browning. Adv. Exp. Med. Biol., 177,437-456.
Effect of High Hydrostatic Pressure on the Survival and Growth of
Escherichia coli 0157:H7
M. F. Patterson,* M. Linton and J. M. J. McClements Food Science Division (Food Microbiology), Department of Agriculture for Northern Ireland, Newforge Lane, Belfast BT9 5PX, Northern Ireland, UK.
It is well established that many factors interact to influence the pressure sensitivity of micro-organisms including species, strain within a species and substrate. One of the most pressure resistant vegetative pathogens identified to date is Escherichia coli 0157:H7 NCTC 12079. This strain is more pressure resistant than other gramnegative pathogens. For example, at 20°C a 15 min treatment with 700 Mega Pascals (MPa) in UHT milk resulted in < loglo 2 reduction in numbers [l]. The same treatment conditions resulted in > loglo 8 reduction with Salmonella
typhimurium. Subsequent work at 20°C in phosphate buffered saline (PBS), poultry meat and UHT milk has shown that below 300 MPa there is no inactivation during a
15 min. treatment. As pressure is increased, substrate has a significant effect on survival, with most inactivation found in PBS and least in UHT milk. There is evidence, assessed by differential plating using trypticase soy agar containing 6% yeast extract (TSAYE) with and without additional NaCl, that sublethally injured cells are present at 200 MPa, even though there was no significant decrease in numbers [2]. The survival and growth of these surviving pathogens are currently being investigated in our laboratory. These studies include (i) the recovery and
High Pressure Food Science, Bioscience and Chemistry
388
growth of the E. coli 0157:H7 after pressure treatment, (ii) the heat resistance of the cells after pressure treatment and (iii) the ability of pressure to induce stress proteins in E . coli 0157:H7.
1.
Effect of high pressure treatment on recovery and growth of Escherichia coli 0157:H7 NCTC 12079.
A stationary phase culture of E. coli 0157:H7 (108/ml) in trypticase soy broth containing 6% yeast extract (TSBYE) was pressurised for 15 min at 300 MPa (20°C). The culture was then inoculated into fresh TSB to a level of approximately
lo4surviving cells per ml and incubated for up to 22 h at 37°C. Samples were taken at intervals and plated on to TSAYE to determine the recovery and growth of the surviving cells.
Control samples of non-pressurised cells were inoculated into
TSBYE at a level of approximately 104/ml and enumerated at intervals during storage at 37°C as described above. The control, non-pressure treated culture had a short lag phase (< 3 h) followed by a time of rapid exponential growth, to reach stationary phase after approximately 10 h (fig. 1). The pressure-treated culture, however, stayed in the lag phase for 10 h indicating that the cells were injured. This was supported by the fact that the colonies on the TSAYE plates were much smaller than those of the controls. The culture reached stationary phase after 15 h by which time colony size had returned to normal. E. coli 0157:H7 NCTC 12079 is known to produce verotoxins VT1 and VT2. It is intended to analyse cultures for their ability to produce these toxins during recovery and storage at various temperatures in
laboratory media. Further work is also planned on the ability of the pathogen to produce toxins during prolonged refrigerated storage of pressure-treated foods such as meat and milk.
389
Biochemistry: Presentations
Fig.1. Effect of pressure (300 MPa/l5 min) on the growth of
f.coli 0757:H7
at 37OC in tryptic soy broth
10 1
6 H Control
0 300MPa/'15 min
4
0
5
10
15
20
25
Time (hours)
2.
Effect of high pressure treatment on hea. resistance of Escherichtia coli 0 157:H7.
A 24h culluic of Esclzcr-iclzia coli 0157:l-17 (NCI'C 12075)) was inoculated into
methioninc assay medium and incubated for 1811 at 30°C. This log phase culture was then heated at three different temperatures; SS'C, 5H"C and 60°C. Cells were enumerated at intervals using I'SAYE. Heat resistance was assessed, as a Dlo value,
for each temperature. This heat treatment procedure was repeated €or cultures which had, less than 30 niiii previously, been subjected to pressures of 100 MPa, 200 MPa and 300 MI% for 30 min at 30%. The results are given in fig. 2 and table 1. The
Dlo value oblained at 55°C for the 100 MPa pre-treatment was significantly greater
High Pressure Food Science, Bioscience a d Chemistry
390
-5
I 0 5 10 15202530
-4 -5
1m 0 5 10 152025 30
control 0 100 MPa
A 200 MPa
-I
v
300 MPa
0 5 10 15202530 Time (minutes) N = surviving bacteria
No = original number of bacteria
Fig.2. Effect of pressure (30 min) on heat resistance of E. coli 0 157:H7.
39 1
Biochemistry: Presentations
than the control. However, pre-treatment at pressures of 200 and 300 MPa resulted in significantly lower Dlo values indicating that the cells were more heat sensitive. This effect was particularly noticeable at 55°C. These preliminary results, in a laboratory medium, indicate that sublethal pressure injury may also sensitise the cells to heat. Further work is needed to establish if a similar sensitising effect is found in foods such as meats. Also, the duration of the sensitising effect needs to be established. This is of relevance as many foods, particularly those of animal origin, may be stored at refrigeration temperatures after pressure treatment before being heated and consumed.
Table 1.
Effect of pressure on subsequent heat resistance (as Dlo values) of E. coli 0157:H7.
Temperature
Pressure treatment
ec> Control
100 MPa/30 min
200 MPa/30 min
300 MPa/30 min
55
18.0*
26.5
9.0
0.9
58
3.2
3.1
1.7
0.5
60
0.9
1.2
0.8
0.3
*D1Ovalue = time (min) required to give 1 loglo inactivation of numbers.
3.
Ability of pressure to induce stress proteins in E. coli 0157:H7.
Cultures of E. coli 0157:H7 in methionine assay medium were subjected to high pressure (300 MPa/ 30 min) and then examined for the presence of pressure shock proteins. Cultures were radiolabelled with 3sS Methionine, cells were lysed by
392
High Pressure Food Science, Bioscience and Chemistry
sonication and 2-D electrophoresis carried out on the lysate [3]. The resulting gels were dried and autoradiograms produced. Image analysis software was used to compare pressurised samples with control samples. Preliminary results suggest that pressure can induce stress proteins in this pathogen. It is not yet known if these proteins are unique to pressure stress, although Welch et a1 [4] reported that E. coli grown under 55 MPa did produce unique stress proteins.
Further studies are
required to determine the significance of these proteins in food processing e.g. their ability to confer heat resistance.
Conclusions
These preliminary results suggest that although certain resistant strains of E. coli
0157:H7 may survive pressure treatment, they are injured, as evidenced by an increased length of lag phase and lower recovery on selective media. The surviving cells are often more sensitive to subsequent heat treatment. The surviving cells appear to produce stress proteins after a sublethal pressure treatment. Further work is needed to determine the true extent of the injury. For example, if it affects the ability of the cells to produce verotoxins.
Biochemistry: Presentations
393
References Patterson, M.F., Quinn, M., Simpson, R and Gilmour, A. (1995) Sensitivity of vegetative pathogens to high hydrostatic pressure in phosphate buffered saline and foods. J. Food Protection 58: 524-529.
Patterson, M.F., Quinn, M., Simpson, R and Gilmour, A. (1996) High pressure inactivation in foods of animal origin. In: R Hayashi and C. Balny (eds.) High Pressure Bioscience and Biotechnology. Elsevier Science B.V. The Netherlands. pp267-272.
O’Farrell, P.H. (1975) High resolution two-dimensional electrophoresis of proteins. J. Biological Chemistry 250: 4007-4021.
Welch, T.J, Farewell, A,, Neidhardt, F.C. and Bartlett, D.H. (1993) Stress response of Escherichia coli to elevated hydrostatic pressure. J. Bacteriol. 175: 7170-7177.
Strategies for High Pressure Inactivation of Endospore-forming Bacteria
Herdegen V. and R. F. Vogel
Lehrstuhl fiir Technische Mikrobiologie, Technische Universitiit Munchen, 85350 Freising-Weihenstephan, Germany. Tel. :4 9 - 8 161-713663, Fax: 4 9 - 8 161-7 13327 E-mail: [email protected]
SUMMARY
The major barrier for the effective use of high hydrostatic pressure for sterilization of products is the strong resistance of bacterial endospores to pressurization. In this study the behaviour of a strain of Bacillus subfilis as representative for endospore forming bacteria was used to develop a strategy for their inactivation. Therefore various heat and pressure treatments were investigated with respect to their potential to produce pressure sensitive vegetative cells. It could be shown that both increasing temperatures or pressures enhanced the germination of endospores. For example with the following treatment a successful sterilization has been achieved: Induction at 500 MPa for 15 min, followed by a 30 min incubation at 37°C and atmospheric pressure, and a subsequent second
Biochemistry: Presentations
395
pressurization for 15 min at 600 MPa. It was possible to perform a sterilization of the endospores in one hour only.
1. INTRODUCTION
Endospore forming bacteria are the microorganisms most diflicult to inactivate by high hydrostatic pressure. The mhanism of the strong resistance of bacterial spores is still not completely elucidated. In general, low pressures up to 200 MPa can enhance the germination of
endospores whereas higher pressures lead to an inactivation. The initial germination of endospores of Bacillus subtilis and Bacillus stearothennophilus during pressurization could be masured by the release of dipicolinic acid and Ca”. It could be shown that a pretreatmnt at relatively low pressures between
60 and 100 MPa accelerated inactivation of endospores at subsequent high pressures [l]. In addition to those results an ideal germination of Bacillus subtilis
spores has been achieved between 60 and 150 MPa. Only an alternating pressure between 60 and 500 MPa in 30 min intervals for an overall time of 180 min at 50°C led to a complete inactivation of the spores [Z]. The effects of a
combination of high hydrostatic pressure with such as alternating currents [3], bacteriocins [4] or other preservatives have also been studied. Regarding c o m r c i a l aspects of the high pressure technology these parameters would need overall processing times which am much too long to be cost effective. Therefore, it is necessary to find a treatment of high hydrostatic pressure to inactivate endospore forming bacteria with very short processing
times and also a very low use of energy and no other additives. In this work all
High Pressure Food Science, Bioscience and Chemistry
396
treatments were carried out at room temperature for the whole inactivation process aiming at a maximum of one hour.
2. MATERIAL AND METHODS
For all experiments endospores of a strain of Bacillus subtilis (Merck, No. 1.10649) were used. 1 ml of spore suspension was diluted with 100 ml of sterile peptone water, pH 7,5. For all thermal experiments this solution was aliquoted in sterile tubes and subsequently treated with temperatures between 40 and 80°C for 5 or 15 min. Finally all samples were incubated at 37°C for
8 hours. Samples which were used for pressure induced germinations were filled
into sterile plastic tubes, vacuum packed, and pressurized. The “high pressure-tyndallisation”was performed with several combinations of a 15 min germination induction at various pressures followed by a 30 min incubation at 37°C and atmospheric pressure and subsequent pressurization for 15 min.
In all samples the number of endospores surviving upon different treatments was evaluated by pasteurization at 80°C for 10 min followed by serial dilutions and subsequent incubation on nutrient agar at 37°C for 24 h.
3. RESULTS 3.1 Heat induced germinations The number of endospores could be reduced in all experiments. It could be shown that both, increasing temperatures or pressures enhanced the germination of endospores.
Biochemistry: Presentations
397
Fig. 1 shows the correlation of the number of surviving spores and incubation times at 37°C after various heat induced germinations for 5 min. A treatment for 5 min without incubation did not affect germination of endospores. On the other
hand vegetative cells upon heat induced germinations of 15 min emerged without induction (Fig. 2). The induction rate was mre effective at lower temperatures than at higher ones. Therefore, it can be assumed that endospores need a certain time of adaptation before the germination starts. After one hour incubation both 5 min and 15 min treatments had nearly the same number of surviving spores at
all applied temperatures.
3.2 Pressure induced germinations Initially it was necessary to find the most sensitive pressures to enhance the pressure induced germination. Therefore, various treatments between 50 and 700 MPa of 15 min each were performed (Fig. 3). An increase of pressure up to
500 MPa resulted in a decrease in surviving endospores. When the pressure was further increased to 600 and 700 MPa the opposite effect was observed.
In Fig. 4 the correlations between a combination of pressure induced germination followed by incubation at 37°C are shown. These experiments confirmed the highest germination rates at pressures of 400 and 500 MPa after 8 hours incubation.
High Pressure Food Science, Bioscience and Chemistry
398
em(r0l
0.0
OS
1.o
2.0
4.0
8.0
LoCub.Lb0 tlmc st 37% Ialbrllgb a t iudndam PI
Fig. 1: Decrease of Bacillus subtilis endospores after various heat treatments of 5 min followed by incubation at 37°C
- 1
I
" 7
cmtd
0.0
0.5
1.o
2.0
4.0
8.0
Incubdon tlmc at 37°C Idlowlug h a t lodudlam PI
Fig. 2: Decrease of Bacillus subtilis endospores after various heat treatments of 15 min followed by incubation at 37°C
399
Biochemistry: Presentations
6
-1
50
100
200
300
400
5M)
fm
7w
p-nF8.1
Fig. 3: Reduction of Bacillus subtilis endospores after various single pressure treatments for 15 min
Fig. 4 Combination of pressure induced germination (1 5 min) of BaciZEus subtilis endospores followed by an incubation at 37°C
400
High Pressure Food Science, Bioscience and Chemistry
According to these results an experiment combining a pressure induced germination followed by an incubation phase and a second pressure treatment was set up. This “high pressure-tyndallisation”was successfi~lupon induction at 500 MPa, 15 min followed by inactivation at 600 MPa, 15 min to effective
sterilization of a sample (Fig. 5). In all experiments with higher pressures than 200 MPa the number of endospores could be remarkable reduced. In some
samples complete sterilization could be achieved. As an example the treatment was successful upon a 1“ pressure of 500 MPa, incubation, and a 2”dpressure of 600 MPa.
w I
I
--I--+-
Fig. 5: “High pressure-tyndallisation”: 1. pressure treatment of 15 min,
incubation for 30 min, 37°C and atmospheric pressure, 2. pressure treatment
401
Biochemistry: Presentations
4. CONCLUSIONS
The pressure dependent behaviour of endospore forming bacteria was used to develop a strategy for their control. It could be shown that it is possible to obtain a complete inactivation of endospores for a strain of Bacillus subtilis by
the use of high hydrostatic pressure. The total processing time was one hour only. No heating of the samples was required.
ACKNOWLEDGMENTS
This research was supported by “Bayerischer Forschungsverbund Abfallforschung und Reststoffverwertung”(BayFORREST)
5. REFERENCES
[l] Butz,
P.,
Ries,
J.,
Traugott,
U.,
Weber,
H.,
Ludwig,
H.
Hochdruckinaktivierung von Bakterien und Bakteriensporen, Pharm. Ind. 52, 487-491 (1990).
[2] Sojka, B., Ludwig, H. Pressure-induced germination and inactivation of Bacillus subtitlis spores, Pharm. Ind. 56,660-663 (1994). [ 3 ] Shimada, K . Effect of combination treatment with high pressure and alternating current on the lethal damage of Escherichiu coli cells and Bacillus subtilis spores, High Pressure and Biotechnology, John Libbey Eurotext Ltd. Vol. 224,49-51 (1992).
402
High Pressure Food Science, Bioscience and Chemistry
[4] Roberts, C.M., Hoover, D.G.Sensitivity of Bacillus couguluns spores to
combinations of high hydrostatic pressure, heat, acidity and nisin, J. Appl. Bact. 81 (4), 363-368 (1996).
Inactivation Kinetics of Microorganisms by High Pressure J.P.P.M.Smelt', N. Dutreuxb, J.C. Hellemons'. 'Department of Microbiology and Preservation, Unilever Research Laboratorium, Olivier van Noortlaan 120, 3133 AT Vlaardingen, The Netherlands bISIM Section Sciences et Technologies des Industries Alimentaires Universite Montpellier I1 place Eugene Bataillon 34095 Montpellier In many instances microbiological safety and stability of food are assured when the microorganisms are sublethally injured. This can be achieved by relatively mild pressures. A model has been proposed that describes the lethal and sublethal injury of food borne pathogenic and spoilage microorganisms. The proposed model is based on the assumption that the physiological state of microorganisms is distributed in a certain way and that stress factors such as high pressure affect the location of the distribution. The first results of inactivation experiments show that the model can describe light injury to complete kill. INTRODUCTION Since 5 -10 years Ultra High Pressure (UHP) treatment as a tool for pasteurisation of food has become a feasible alternative to heat pasteurisation. When considering microbiological safety aspects of UHP treatment one should bear in mind that physical decontamination is just one step in determining the microbiological safety risk. This risk is not only determined by the lethality of the treatment, but also by the quality of the raw materials, the distribution conditions and the use by the consumer. Infectious pathogens such as Listeria, Salmonella and verotoxigenic E. coli should be absent in foods. Consequently, inactivation of these organisms is a prerequisite to microbiological safety. Toxigenic pathogens such as Stapbyloccus aureiis and Clostridium botulinum are not infectious themselves and, if they are prevented from growing, they do not form a toxin in food. Contrary to untreated cells, sublethally injured cells cannot grow under suboptimal conditions such as reduced pH or increased salt content. Hence sublethal injury brought about by relatively low pressure, can be a feasible option in achieving microbiological safety. Non pathogenic spoilage organisms such as yeasts or lactobacilli, must reach a certain cell concentration (e.g. above lo6-10' for lactobacilli) before signs of spoilage occur. If cells are injured, they time for adaptation (lag time) before they resume growth. Extension of lag time will result in a considerable delay in time to spoilage and hence the shelf life - before opening the pack - will be extended. Again, sublethal injury can be a tool in preservation of microorganisms. Sublethal damage and lethal damage are mostly treated as two separate events Sublethal damage as reflected by a prolonged lag time is mostly modelled in growth models. Traditionally, lethal inactivation is described as first order kinetics, inspite of the many deviations from first order kinetics. Here we propose one general model which can describe both sublethal and lethal injury. MODELLING INJURY The model is based on the assumption that the physiological state of a cell population is not identical, but is distributed i n acertain way There are reasons to assume that injury o f living
High Pressure Food Science, Bioscience and Chemistry
404
cells are lognormally distributed. (Koch, 1966). This is illustrated in Fig. 1. When the cell population is subjected to heat or pressure the physiological state of the population impairs and resulting in an increased sensitivity to low pH or a longer lag time. A very small part of the population is heavily injured and these cells are no longer able to multiply. They are considered as 'dead'. These cells, however may still be able to perform certain physiological functions such as acid production. When the treatments are more severe, a larger part of the population will be 'dead' and finally the capacity to perform physiological functions is lost. Here we have applied the model both on sublethal and lethal inactivation. Fig. I . Distribution of physiological state of microorganisms frequency of damaged cells
(1-1
nuwber d cells wifh increasing degree of lnjuw
___C_
MATERIAL AND METHODS Microorganisms and culture conditions L. plantarum La 10-11 isolated from an onion ketchup and identified as L. plantarum by ATCC (American Type Culture Collection), no ATCC number. The organism was stored in milk at -20°C until use. Cells were grown in MRS broth (de Man e f al. 1960). Listena monocytogenes NCTC 11994. Methods for cultivation and recovery have been described by Patterson et al. (1995) Pressure treatments Cell suspensions of La 10-11 (in buffer or milk or carbohydrate solutions) were filled into plastic pouches (Nasco Whirl-Park ) and pressurised at 25 "C Pressure treatments were conducted i n a vessel with under isostatic pressure (National Forge St Niklaas, Belgium) Temperature and pressure were recorded during treatment Immediately after pressure treatment the pouches were put on ice and experiments such as counting were done as quickly as possible Pressure treatments of Listerra tnonocyfogenes has also been conducted i n an hydrostatic pressure vessel (Stansted Ltd) as described by Patterson et al (1995) 1)eterminatron of /he lag time by OD mcasirretnent From treated and non-treated cell suspensions. appropriate fourfold serial dilutions were made in MRS broth 126 replicates were made 0 2 ml of each dilution was inoculated in a well of a microtiter plate (Corning (96-well plate with Lid, flat bottom, well diameter 6 4 mm. polystyrene) To estimate the number of colonyforniing units per well the number of organisms
405
Biochemistry: Presentations
was simultaneously plated on MRS agar. The inoculated microtiter plates were incubated at 30 "C. To ensure that wells with one cell per well, the suspensions were diluted in such a way that finally no cells were present. The lag time was estimated according to the following formula: Lag time = (time to turbidity) (generation time)
-
((log106 4) - (log[initial number of cells per well])/logZ)
*
is the number of cells per well at the point when change in optical density can be observed As we only used results from wells with one colony forming unit, the ~) time) formula becomes Lag = (time to turbidity) - ( ( l 0 g 1 0 ~ /log2)*(generatton
Modelling lethal inactivation Lethal inactivation was modelled with a distribution model, assuming lognormal distribution of pressure resistance in a population. On the Y axis the frequency of death is plotted against the log number of time on the X axis.
0
20
time / min 450
Pressure / MPa
Fig 2 Inacllvatton hinetics of Lisfrrra rnonucyfugrncs i n 0 I M phosphate buffer pH 7 0 by various pressures (No=number or colony forming units pcr ml b e h e treatment, N, number of colony Corming untls after treatment)
High Pressure Food Science, Bioscience and Chemistry
406
RESULTS Inactivation Fig 2 illustrates that the inactivation kinetics were in line with the model presented in Fig 1. It illustrates that more cells of the population are killed after longer treatment time and that the assumption of lognormality can be used to describe inactivation.
Sublethal inactivation Sublethal inactivation is often reflected by an increase in lag time. As shown in Fig. 3.1, the distribution of lag times of untreated cells is quite close to a log normal distribution. Distribution of lag times of treated cells, however, showed a different pattern as shown in Fig 3.2 It should be borne in mind that 99 % of the number of colony forming units of the treated population was reduced after treatment. As shown only the fit with log-normality is
here quite poor. Again, both observations are in line with the hypothesis as illustrated in Fig. 1. 20 20-
16
E
-
ng. 3.2
flg. 3 t
'
15
-
a
.
$10-
;lo-
F .
L
b
5 -
6 -
dl 0.1
0.3 0.6 0.7 0.9 1.1 log lag tinv I hours. (at 30 C) H 0bSeN.d 0 pr.dicted
1.3
0
1 0.5
2.5 4.5 6.5 8.5 10.5 log lag time / hours. (at 30 C)
12.5
H ObleNed 0 predicted
Figs 3 . 1 . and 3.2. Distribution of lag times of untreated and treated (10 rnin 250 MPa) cells of Lactobacillus plantarum.Predicted means fitted to log-normality.
DISCUSSION Physical inactivation of bacterial cells is most likely to be a continuous process, ranging from light injury reflected by a small increase in lag time, to inability of perform any physiological function. An intermediate stage is the inability to multiply. Here an example is given of the effect on the population on sublethal injury as reflected by lag time and an example of lethal injury reflected by the inability of the cells to grow after treatment. REFERENCES Koch, A.L. (1966) The logarithm in biology 1. Mechanisms generating the log-normal distribution exactly. J. Theoret. Biol. 12: 276 - 290.
Biochemistry: Presentations
407
Man, J.C. de, Rogosa. M. and Sharpe, M.E.(1960). A medium for the cultivation of lactobacilli. J. Appl. Bacteriol. 23: 130 -135.
Patterson, M. F., Quinn, M, Simpson, R. and Gilmour. (1995) A. Sensitivity of vegetative pathogens to high hydrostatic pressure treatment in phosphate-buffered saline and foods. J. Food Prot. 58: 524-529.
Biochemistry: Posters
Pressure-Temperature Stability Diagrams of Proteins: a-Amylases from
Bacillus Species
P.Rubens', L.Smelle? & K.Heremans'*
'Department of Chemistry, Katholieke Universiteit Leuven, B-300 1 Leuven, Belgium. E-mail: [email protected] 21nstituteof Biophysics, Semmelweis University of Medicine, Budapest, Hungary
The stability of three a-amylases fiom different Bacillus species was examined under different conditions of pressure and temperature. Changes in protein structure were followed with FT-IR spectroscopy. The spectra of these proteins show a typical behaviour under extreme conditions. Chemical modifications may play an role in the stability of these proteins at high pressure and temperature. The stability lines were plotted in a pressure temperature plane and showed an elliptical outline. Stabilization against temperature denaturation at low pressure was observed.
412
High Pressure Food Science, Bioscience and Chemistry
1 .INTRODUCTION The most industrially sigmficant and best studied a-amylases are fiom the genus Bacillus. The pressure stability at room temperature and the temperature stability
was reported recently (Weemaes et al., 1996; Rubens et al., 1997). The aim of thls study was to examine the stability of three different naturally occurring a-amylases fiom Bacillus species in a broad pressure-temperature region. The differences between the temperature and pressure induced changes of the secondary structure during the denaturation process are studied with Fourier transform infrared spectroscopy (FTIR).
Examination of different
components in and around the Amide I’ region (1700-1500 cm-I) gives information of the secondary structure and chemical modifications in the protein. It is shown that at high temperature, chemical modrfications play an inportant role in the stability of proteins. 2. MATERIALS AND METHODS The a-amylases of Bacillus subtilis (BSA), Bacillus amyloliquefaciens (BAA) and Bacillus lichenlformis (BLA) were dissolved in D20 with lOmM TRIS-DCl buffer at pD 8.6. L-asparagine-t-butyl-ester was dissolved in D20 in order to examine chemical modification in the protein due to deamidation of Asn-residues. The pH was adjusted at 8.6 by adding NaOD.
The samples were stored
overnight to ensure that the WD-exchange has reached a static regme. Infiared
Biochemistry: Posters
413
spectra were obtained with a Bruker IFS66 FTIR spectrometer equipped with a liquid nitrogen cooled broad band mercury-cadmium-telluride solid state detector. 250 Interferograms were coadded after regstration at a resolution of 2 cm-'.
Temperature scans were performed with a Graseby/Specac low voltage heating system at a rate of 0.2 "Urnin. The cell was fitted with CaFz windows and has a constant pathlength of 0.025 mm. For the pressure measurments, the solution is placed in a stainless steel gasket of a Diamond Anvil Cell (Diacell Products, Leicester, UK).
The pathlength is 0.05 mm at ambient pressure.
The
experiments at hgh pressure and temperature were done in a temperature controlled DAC. The infrared light was focussed on the sample by a NaCl lens. The rate of pressure increase is about 0.03 kbar/min. A small amount of Bas04 is added to determine the pressure (Wong & Moffat, 1989). 3 . RESULTS ANJI DISCUSSION
The behaviour of the proteins under extreme conditions was followed by examination of changes in the Amide I' region. Analysis of the frequency of the h d e I' maximum allows the characterisation of the pressure and temperature denaturation. These data were plotted in a pressure temperature plane (fig. 1) and show an elliptical outline.
Above 70°C, BSA seems to show a decreased
temperature stability at low pressures. In contrast BLA and BAA are stabilized by pressure against heat denaturation. The deconvoluted spectra reveals different
High Pressure Food Science, Bioscience and Chemistry
414
bands which can be assigned to different secondary structures (Byler & Susi, 1986). Figure 2 shows the spectra of BSA at
different conditions.
pressure
experiment
at
The room
temperature (B) results in a broad band with an absorption maximum
A b
;i
t
,
,
,
,
0
20
40
Eu
;A
::,
0 -20
80
iw
1
at 1642 cm-’ which can be assigned to the unordered structure. At high temperature and ambient
120
Temperature (OC)
pressure (C) the h i d e I’ band
Fig.] : Stability diagram of BSA (.>,
shows the formation of two distinct
buffer at pD=8.6
These bands are assigned to the
formation of intermolecular P-sheet which indicates aggregation. (Mozhaev et al., 1996). This is in contrast with the pressure induced changes where only a broadening of the h i d e I’ band is observed. The spectrum at high pressure and temperature (D) reveals a clear band at 1615 cm-’. Due to the absence of the
band at 1680 cm-‘, it can not be assigned to intermolecular P-sheet aggregation.
Biochemistry; Posters
415
A similar behaviour is observed for the other enzymes. The formation of the band at 1615 cm-I starts at 50°C for BSA and BAA while for BLA, it only appears at 70°C. Thls band is also accompanied with a 1600
1620
1640
1660
1680
1700
Wavenumber (cm-')
Fig 2 : deconvoluted spectra of BSA in 10 mh4 Tris-buffer at pD=8.6. (A)
ambient conditions.
@) 3OoC, 10.5
kbar. (C) atmospheric pressure, 100°C.
(D) 75"C, 5.5 kbar
band at 1580 cm-' which can be assigned to aspartic acid (Chirgadze et al., 1975). Tomazic et al. (1988) have
shown
that
at
high
temperatures the deamidation of asparagine residues becomes the
cause of irreversible loss in enzymatic activity of BLA and BAA. In order to correlate these observations with the changes in the IR spectrum, the temperature dependence of Asn-t-butylester (Sigma A0502) was examined. At elevated temperatures one can observe the formation of two bands at 1619 and 1650 cm-'.
It may be assumed that the formation of the strong band at 1615 cm-' can be correlated with a chemical modification of the asparagme residues in the protein. Further experiments will have to show whch role the deamidation plays on the stability of these proteins.
High Pressure Food Science, Bioscience and Chemistry
416
4,ACKNOWLEDGMENTS This research was supported with the grant of the European Community, project number FAIR-CT96- 1 175. 5.REFERENCES
Byler, D.M. & Susi, H., Biopolym., 25, 469-487 (1986) Chirgadze, Y.N., Fedorov, O.V. & Trushina, N.P., Biopolym., 14, 679-694 (1975) Mozhaev, V.V., Heremans, K., Frank, J., Masson, P. & Balny, C., Proteins, 24, 81-91 (1996) Rubens, P., Goossens, K., Heremans, K., Weemaes, C., Hendrickx, M. & Tobback, P., In High Pressure Research in the Rio-Sciences and Rio-
Technology, (Ed. K. Heremans) Leuven University Press, 327-330 (1997) Tomazic, S.J. & Klibanov, A.M., J. Biol. Chem., 263, 3086-3091 (1 988) Tomazic, S.J. & Klibanov, A.M., J. Biol. Chem., 263, 3092-3096 (1988) Weemaes, C., De Cordt, S., Goossens, K., Hendrickx, M., Heremans, K. & Tobback, P., Biotechnology & Bioengineering, 50, 49-56 (1 996) Wong, P.T.T. & Moffat, D.,J., Appl. Spectrosc., 43, 1279-1281 (1989)
Pressure Induced Fluorescence Quenching in Plant Light Harvesting Complex I1
J.P. Connelly"*,A.V. Rubanband P. Hortonb 'INSERM U128, 1919 Route de Mende (CNRS), F-34293 Montpellier, CEDEX 5, France. bDepartment of Molecular Biology and Biotechnology, Sheffield University, Sheffield, UK. Pressure induces similar effects in LHC II, a plant pigment-protein complex, to those involved in photoprotection mechanisms in vivo and aggregation in vitro. Here we assess the potential of pressure techniques to probe the interactions and conformational changes involved. 1. INTRODUCTION
Under excess light conditions, plants activate several protective mechanisms to prevent overloading the photosynthetic reaction centres[11. The main mechanism is associated with dissipation of chlorophyll (Chl) excitation energy monitored by fluorescence quenching, called qE. This process is triggered by a drop in pH leading to aggregation of the peripheral light harvesting complexes (LHCs) and formation of a quenching species that dissipate the excess energy as heat, a process that is well documented both for leaves and isolated LHCs of photosystem II
(LHC II) [l]. Close parallels are observed between intact thylakoid
membranes and detergent solubilized and aggregated LHCs [2], which are assumed to be reasonable models of effects in vivo. Here we report that analogous quenching effects in LHC
II are induced by pressure, apparently without aggregation, which makes pressure a powerful tool to probe the conformational changes accompanying quenching.
The most abundant peripheral pigment-protein complex, LHC 11 b, is trimeric where each monomeric 27 kDa apoprotein binds at least 8 Chl a, 6 Chl b, 2 lutein, a neoxanthin and substoichiometric amounts of xanthophyll [ 1,3]. Aggregation initiates protein conformational change and changes pigment interactions resulting in quenching. Peripheral xanthophyll molecules regulate the degree of quenching, most probably by modifying the characteristics of
418
High Pressure Food Science, Bioscience and Chemistry
aggregation; zeaxanthin increases quenching and can be converted to its antiquenching antagonist, violaxanthin by the xanthophyll cycle. It is not known to what degree the quenching effects are due to aggregation directly or to aggregation induced protein conformational change. Here we report the effects of pressure on isolated LHC II complexes and assess the potential of pressure methods to probe the underlying conformational flexibility resulting in quenching.
2. MATERIALS AND METHODS LHC 11 complexes were isolated from dark adapted spinach leaves and purified using the isoelectric focusing procedure detailed elsewhere
[4].
LHC 11 monomers were prepared by
phospholipase treatment of the trimers followed by centrifugation on a sucrose gradient. Monomers and trimers were employed in a 0.01% DM, 20 mM Hepes buffer or 60% glyceroYwater buffer adjusted to pH 7.6 and at a concentration of 5pM Chl giving an OD of 0.5 /cm at 680 nm for absorption spectra or 0.9 mM (OD 0.08 /cm at 680 nm) for fluorescence. Aggregates were prepared by diluting the trimers with 20 mh4 hepes buffer to yield a concentration of 5 p M Chl. To investigate the possibility that pressure induced aggregation and to hinder diffusion, LHC Il trimers were also prepared in thermosetting agarose gel (Sigma type 7) by dissolving the complexes in 1% gelhuffer at 30°C and then cooling rapidly to 10°C in the pressure cuvette. Fluorescence emission and excitation up to 600 MPa were recorded using an Aminco
SLM spectrofluorimeter using an analogous pressure system to that used for absorption described elsewhere
[5].
To account for the compressibility of water, each spectrum was
corrected for the pressure-dependent change in volume and consequently LHC II concentration. With each pressure increment, typically in steps of 20-50 MPa, the system was allowed to equilibrate for 5 minutes before measurement. A cold denaturation experiment at
200 MPa allowed spectra down to -20°C to be measured. Dried nitrogen gas was blown over the windows to prevent condensation.
3. RESULTS Emission spectra between 600 and 740 nm for LHC II trimers at different pressures, exciting the Chl a Soret band at 435 nm, are illustrated in figure 1 and reflect the Chl a emission.
419
Biochemistry: Posters
Plotting the position and height of the main 680 nm band against pressure gives a smooth, red shift of -0.15 cm-'/MPa. Using the same analysis for the emission spectra of monomers, trimers under various conditions and aggregates, a very similar shift with pressure is obtained and the results given in table 1. The shift proves to be entirely reversible in all cases.
I
640
660
680 700 wavelength (nm)
720
740
Figure 1. Fluorescence emission spectra for LHC II trimers as a function of pressure at 5°C. Spectra at 100 MPa increments in pressure are shown. With pressure the fluorescence intensity of monomer and aggregates was also observed to decrease (table 1) and release of pressure, recovery of fluorescence was accompanied by hysteresis for all samples. A very weakly sigmoidal fluorecence intensity dependence was observed in some cases. This effect was absent in the aggregated complexes but increased using a higher detergent concentration and appears to correspond with a phase transition of the detergent, observed to occur at 250 MPa in the absence of buffer and protein. The large decrease in fluorescence is not accompanied by a corresponding decrease in the
420
High Pressure Food Science, Bioscience and Chemistry
absorption at 435 nm (excitation wavelength) so the effect is due to quenching. Experiments using degassed solvent gave the same results. Emission spectra recorded exciting Chl b at 472
nm do
not lead to appreciably different results. The degree of quenching at 600 MPa can be
calculated according to Ruban et d. using
(FIMPa-FrjwMPa)/F6wMPa
[3], where F is the
maximum fluorescence intensity at the given pressure. Values for the degree of quenching at
600 MPa for each sample are listed in table 1. Quenching is also accompanied by a slow relative broadening of the main band due to increase in intensity around 700 nm. Sample
Shift cm-'/MPa
Quenching (FI MPa-F600MPa)E600MPa
monomers trimers 0.0 1 %DM trimers 0.13%DM trimers in 60% glycerol/water small aggregates
-0.154(8) -0.133(2)
1.32 1.17
-0.126(3)
1.12
-0.140(2)
2.99
cold denaturation
-0.15l(2)
0.56
cm-'/"C
(FlsT-F-17"C)h"c
-0.17(1)
-0.12
Table 1. Characteristic changes in the fluorescence emission spectra between 600 MPa and 1 MPa. Cold denaturation leads to the reverse effect to pressure. On cooling from +15 to -16.8"C at
200 MPa, a 0.17crn-'/"C shift to the blue accompanies a small increase in fluorescence which is reversible. In contrast, for trirners in 60% glycerol/water solvent the degree of quenching is dramatically increased and there is only 60% recovery of the original signal, suggesting the formation of aggregates.
4.DISCUSSION The pressure induced fluorescence quenching and spectral changes bear a striking resemblance to the effects on aggregation although the size of the effects are smaller with pressure. Changes in the absorption spectra reinforced this observation (Connelly et al. unpublished results). Analogous changes in the carotenoid absorption around 500 nm are
Biochemistry: Posters
42 1
induced by pressure or aggregation. This raises the question as to whether pressure simply induces aggregation. To probe this possibility, trimers were prepared in 1% agarose gel to prevent diffusion leading to aggregation. Almost identical changes in the absorption spectrum with pressure as for the trimer in solution were observed and there was little evidence to suggest the gel became less viscous with pressure. In addition, there was no increase in scattering as might be expected on formation of oligomers or aggregates. High detergent also has little effect, other than the detergent phase transition becomes more apparent. In contrast, use of a glycerol/water buffer, known to induce oligomerization, results in a dramatic increase in quenching (table 1) and loss of reversibility with pressure. This only provides tenuous evidence that aggregation is not the primary effect on increasing pressure and more quantitative measurements using fluorescence anisotropy and light scattering are in progress. The general red-shift with pressure may be due to compression of the environment surrounding the pigments or reflect changes in the conformation or coupling between pigments. Related pressure measurements on bacterial light harvesting complexes [ 6 ] , showing similar red-shifts and fluorescence quenching have been interpreted in terms of compression of the surrounding protein. However, given the similarity between the effects of aggregation and pressure, this suggests that pigment conformation or coupling changes are the predominant effect. In the case of aggregates, resonance Raman spectroscopy has shown that at least one carotenoid undergoes a torsion and a Chl b forms a new hydrogen bond when LHC
II aggregates [7].It is not clear from those measurements if this is the direct result of
aggregation, i.e. closer packing and interaction among peripheral pigments, or the result of aggregate induced conformational change within trimers. Since similar effects are observed on applying pressure and to greatest degree in monomers, and tentatively postulating that the same structural effects are responsible for the pressure effect in the absence of aggregation, this suggests a conformational effect localized on the monomer. This postulate could be tested in the future using high pressure resonance Raman spectroscopy. Decrease in temperature tends to favour formation of hydration of hydrophobic interactions for soluble proteins [ 8 ] . With LHC 11 a decrease in temperature reverses the changes induced by pressure which suggests that increased hydrophobic interaction is responsible for the pressure and aggregation effects. LHC II is a membrane protein however, and it is not clear what role the detergent and residual lipid may play. Some idea of the role of
422
High Pressure Food Science, Bioscience and Chemistry
water can be obtained by using osmolytes to modulate the activity of water. Glycerol is known to induce LHC
II oligomerization (Dekker & Ruban unpublished results) although this may
result from specific interactions, since sucrose appears to have a different effect and both would be expected to show an osmotic effect. In conclusion, it appears that judicious use of pressure, temperature, osmotic and viscosity effects, variable concentrations of detergent and degrees of aggregation should yield specific information on the structural changes involved in non-photochemical quenching in LHC 11.
5. ACKNOWLEDGEMENT J.P.C. thanks INSERh4 for a Poste Vert fellowship and Dr C. Balny and R. Lange for supporting this project. A.V.R. and P.H. acknowledge support from UK BBSRC.
5. REFERENCES 1) P. Horton, A.V.Ruban, R.G. Walters, Annu. Rev. Plant Physiol. Plant Mol. Biol. 47, 655 ( 1996).
2) A.V. Ruban, A. J. Young, P. Horton, Biochemistry, 35,674-678 (1996). 3) W. Kuhlbrandt, D.N. Wang, Y. Fujiyoshi, Nature, 367,614-621 (1994). 4) A.V. Ruban, A.J.Young, A.A. Pascal, P. Horton, Plant Physiol. 104,227-234 (1994).
5) R. Lange, J. Frank, J.L. Saldana, C. Balny, Eur. Biophys. J. 24,277-283 (1996). 6) A. Freiburg, Ellervee, P. Kukk, A. Laisaar, M. Tars, K Timpmann, Chem. Phys. Lett, 214, 10-16 (1993).
7) A.V. Ruban, P. Horton, B. Robert, Biochemistry, 34,2333-2337 (1995). 8) D. Foguel, J.L. Silva, Proc. Natl. Acad. Sci. USA, 91, 8244-8247 (1994).
Pressure Unfolding Ribonuclease A and the Seven Mutants
J.P. Connellya*,J. Torrentb,R.Langea, M. G. Collb,M. Riboband M. Vilanovab aINSERM U128, 1919 Route de Mende, F-34293 Montpellier, CEDEX 5, France bArea de bioquimica i biologia molecular. Department de Biologia. Facultat de Ciencies. Universitat de Girona. Spain. Mutations in the ribonuclease A Chain Folding Initiation Site (CFIS) extending from residue 106 to 118 (of 124) have been made in the buried hydrophobic region at I106 and 1107. Replacement of these residues by Ala, Leu or Val were designed to gauge the effects of side chain hydrophobicity and size on stability on the region. Their stability to pressure has been studied by 4th derivative absorption spectroscopy. Replacement of Tyr 115 by Trp has been tested for its potential as a minimally destabilizing fluorescence probe of the CFIS. The results are rationalized in terms of residue hydrophobicity, size and water exposure. 1. INTRODUCTION
Ribonuclease A is an extensively studied enzyme with a well known structure and molecular biology, and a developed background of folding kinetics and dynamics[l]. Being relatively small, (14 kDa) with a mixture of secondary structural features it is an obvious candidate for study of the factors affecting the stability, and the principles underlying structure and function. The processes of unfolding RNase A under various conditions of pressure, temperature and chemical denaturant have been studied using many techniques, showing that the process is not concerted and proceeds via formation of intermediates[Z]. Particular regions of the protein are conferred with relatively greater stability and probably play important roles in folding and the dynamic integrity of the protein. The chain folding initiation site (CFIS) extending from residue 106 to 118 (of 124) is one such region observed to be a nucleus for protein folding [3]. The CFIS comprises a hairpin-like antiparallel p sheet structure near the C terminus. It is partially solvent exposed (around the type VI beta tum at G112 to Y115) and hydrophobic
High Pressure Food Science, Bioscience and Chemistry
424
interactions play an important role in its stability. To probe the factors affecting the CFIS stability, Ile 106 and 107, which occupy positions in the hydrophobic core of the
p
sheet
structure, have been selectively mutated to 1106AILN and 1107AILN to give series graded in hydrophobicity and residue size. Pressure has been demonstrated to disrupt hydrophobic interactions[4] and provides, together with thermally induced unfolding, a means to test the outcome of mutation. In contrast, Tyrl15 is solvent exposed on the exterior of the CFIS
turn
and has been replaced with Trp. This mutation is more remote from the CFIS hydrophobic core and is expected to affect stability far less. It may therefore potentially serve as a nondestabilizing local fluorescence label[5]. Here we demonstrate that site directed mutagenesis and pressure induced unfolding provide a useful means to explore the hydrophobic properties of the CFIS.
2. MATERIALS AND METHODS The gene coding for RNase A [6] was used as a template for the mutagenesis by PCR following the method of Juncosa et al. [7]. Mutants and wild type enzymes were expressed in
E. CoZi using the T7 expression system. Proteins were purified using cationic exchange chromatography [6]. About 40 mg of recombinant protein per litre of culture were obtained. The enzymes were dissolved to a concentration of 1 mg/ml in 50 mM MES buffer at pH 5 for pressure experiments and 50 mM sodium acetate buffer pH 5 for thermal experiments. Absorption spectra between 250 and 310 nm under pressures up to 4.5 kbar at 40°C were recorded using the modified absorption spectrometer described elsewhere [8]. To account for the compressibility of water, each spectrum was corrected for the pressuredependent change in volume and RNase A concentration. With each pressure increment, typically in steps of 20-50 MPa, the system was allowed to equilibrate for 5 minutes before measurement. Spectra were analysed using the 4th derivative [8].
3. RESULTS The 4'h derivative spectra are optimized for effects in the tyrosine spectrum, of which there are six in RNase A. Two of these Y73 andY76 are close to the CFIS and should be quite sensitive
425
Biochemistry: Posters
to local changes in conformation. Typical 4th derivative spectra following the increase of pressure on the wild type are shown in figure 1.
... - 0.6
0.1 -
-5 U
- 0.4
g
- 0.2
0’
\
6 0.0
Tf
U)
% 0)
-
(D
260
-
- v
-0.1 -
270
280
290
0.0
300
wavelength Figure 1. Effect of pressure on the 4th derivative UV spectra of RNase A wild type at pH 5 and 55°C. The zero order spectrum is shown as a dotted line. Pressure induces a blue shift and change in intensity with clear isosbestic points allowing a simple two state model to be used. The blue shift reflects the change in tyrosine environment polarity as the protein unfolds and can be compared with a polarity scale based on N-acetyl-Ltyrosine ethyl ester (ATEE) in various solvents of known dielectric constant [8], to judge the degree of the exposure to water. The transition from native to unfolded state is analysed by plotting 4th derivative absorption changes at 282.6 nm against pressure. The sigmoidal curve so obtained can then be fitted to a two state thermodynamic model as illustrated in figure 2,
where simulated normalized curves based on fits of the thermodynamic data are plotted. Not all unfolding processes were observed to be immediately reversible. I106A and the entire I107 series appeared to show very slow recovery kinetics on decrease in pressure.
High Pressure Food Science, Bioscience and Chemistry
426
1.o
0
100200300400500600
0
100200300400500600
0.5
0.0
% 0.75
B fil 3 0.25 5
.- 1.0
H
LL
0.5
0.0
Pressure (MPa)
Figure 2. Simulated, normalized pressure unfolding curves for RNase A wild type and mutants based on thermodynamic parameters fitted assuming a two state transition. All measurements were made at 4OoCand pH 5.
427
Biochemistry: Posters
It can be seen that the positions of the curves give a consistent idea of the relative stabilities. The recombinant and wild type are virtually identical in behaviour, so the process of overexpression in E. coli is not significantly affecting the stability of the RNase A. The I106 mutants are less stable than their I107 analogues. Within each series, the stability follows the order I>V>L>A although the relative stabilities within each series varies a little; I106L appears to be comparatively more stable than I107L within their respective series. Compared to the other mutants, the Y 115W mutant is less destabilized suggesting that this mutation may indeed serve as a useful local fluorescence label. Fluorescence gives the same thermodynamical parameters for the Y 115W pressure unfolding transition as 4th derivative absorption. 4. DISCUSSION Mutations employed were intended to probe the role of hydrophobicity. The order of stability accords with the hydropathic scale of Kyte [9]. However, the differences in behaviour at I106 and I107 suggest that other factors such as changes in side-chain volume, position and water accessibility are important. The 106 and 107 sites are buried in a hydrophobic region so
Van der Waals interactions are the predominant interaction involved with the side-chains. For these interactions, with a strong dependence on proximity, the packing density is the primary factor determining the degree of stabilization. I107 appears to have the higher packing density (based on the WT structure) and consequently probably contributes more to the stability of the region than 1106. However, the I106 mutants are more destabilized than I107 which may reflect that the I107 position is less buried and has greater flexibility to compensate for changes in the side-chain interactions. Replacement of Ile by Ala or Val reduces the side chain volume[lO], probably leading to some rearrangement in the packing of side chains but with relatively non-disruptive effect leading only to a weakening or loss of interactions relative to the WT. Replacement of Ile by Leu involves a change in shape and is more disruptive leading to new types of interaction relative to the WT. Being in a region of higher packing density, I107L might be expected to be relatively more destabilized than I106L within their respective mutant series, as observed. Also important in pressure induced unfolding is the role of water. I107 has a solvent accessible surface (23.3
A2) five times larger than I106 (4.8 A2). Water may
stabilize the
428
High Pressure Food Science, Bioscience and Chemistry
nominal cavity created by mutation and to greater effect in the more accessible 107 series. However, all mutants appear to be slightly blue shifted i.e. the tyrosines and probably those nearest and most affected by the mutation, are in more polar environments compared with the wild type. Only in the case of the Ala mutants is the shift significant suggesting a structural change. Otherwise there is no significant difference between the 106 and 107 series. Use of osmotic pressure techniques [l 11 may reveal if water plays a significant stabilizing role. Pressure often leads to unfolded states retaining significant secondary structure. Mutation may result in different unfolded states for the two mutant series although this is hard to judge from the spectral shifts in the absence of structural data.
5. ACKNOWLEDGEMENT The authors thank INSERM (JPC, Poste Vert fellowship), COST (J.T. collaboration costs) and acknowledge grant PB93-0872 from DGICYT of the Ministerio de Educacion y Ciencia (Spain) and the Fundacio M.F. de Roviralta (Barcelona, Spain) for equipment purchasing grants. 5. REFERENCES 1) M.V. Nogues, M. Vilanova, C.M. Cuchillo, Biochim. Biophys. Acta 1253 16-25 (1995). 2) C.R. Matthews, Annu. Rev. Biochem. 62,653-683 (1993). 3) M. Gross, R. Jaenicke, Eur. J. Biochem. 221,617-630 (1994). 4) R.W. Dodge, H.A. Scheraga, Biochemistry 35,1548-1559 (1996). 5) R.A. Send&, D.M. Rothwarf, W.J.Wedemeyer, W.A. H o w , H.A. Scheraga, Biochemistry 35, 12978-12996(1996). 6) S.B. del Cardayre, M. Ribo, E.M. Yokel, D.J. Quirk, W.J. Rutter, R.T. Raines, Protein Eng. 8,261-273 (1995). 7) M. Juncosa, J. Pons, A. Planas, E. Querol, Biotechniques, 16,820-824 (1994). 8) R. Lange, J. Frank, J.L. Saldana, C. Balny, Eur. Biophys. J. 24,277-283 (1996). 9) J. Kyte, R.F. Doolittle, J. Mol. Biol. 157, 105-132 (1982).
10) Y. Harpaz, M. Gerstein, C. Chothia, Structure, 2,641-649 (1994). 11) V.A. Parsegian, R.P. Rand, D.C. Rau, Meth. In Enzymology, 259,43-94 (1995).
High Pressure Treatments of Listeria monocytogenes at pH 7 and
pH 5.6, and Flow Cytometry Monitoring of Pressurized Cells
Magali RITZ'", Marie-France PILE"@)'b'*, Michel FEDERIGHI'"
'"Unite associee INRA d'Hygihe Alimentaire
- Ecole Nationale V6tCrinaire
de Nantes - BP 40706 - 44307 NANTES CEDEX 03. (b)Laboratoirede Microbiologie Alinientaire et lndustrielle - ENITIAA - Domaine de la Geraudiere - 13P 82225 - 44322 NANTES CEDEX 03 FRANCE.
I NTKODUCTION High hydrostatic pressure is a new technology that can be used in food industry for microbial destruction. Several studies have shown that the eficacy of high pressures is influenced by the type of microorganism, the exposition time, the temperature, the pil. In this work, we studied the activity of high pressures on the inactivation of Lzsleria munocyiogenes in neutral and acid pH. When complete
430
High Pressure Food Science, Bioscience and Chemistry
inactivation of the cells determined by viable plate count method was observed, the possibility that some cells remain viable was also studied by fluorescent staining and ilow cytomctry monitoring.
EFFECT OF 1’11 Cells of Listeria nionocylogenes Scott A (108 CFU/ml) were diluted in phosphate buffer pH 7 or citrale buffer pM 5.6. Ten ml of these suspensions were submitted to high pressures for 10 min at 200 to 650 MPa at 20°C (Gec Alsthom,
3 1 pressure vessel, 0-690 MPa). Cells were plated on PCA for viable cell counts. No significant inactivation was recorded below 300 MPa at pH 7 or 5.6. When considering maximal inactivation (8-L0g), 600 MPa were required at pH 7 and only 400 MPa at pH 5.6 (Fig.1). At 350 MPa a difference of 5-Log was recorded between the efficacy at pki 7 and the efficacy at p1-I 5.6. These observations suggest that high pressures and pH could have a synergic effect on the destruction of Listeria monocytogenes.
43 1
Biochemistry: Posters
.*
OL 0
W
200
330
. 40
5oD
aa
.7a
-(yprp
ESTIMATION OF CELL VIABILITY
For flow cytometric studies, non-treated cells, pressurized cells (10 min-400
MPa, pH 5.6) and heat-treated cells (10 min-121°C) were resuspended in PBS afler centrifugation (I 2000 g, 5 min). Cells suspensions (1 ml) were stained with 30 p1 carboxy fluorescein diacetate (cFDA, 1 mM) or 10 pl propidium iodide (PI, 1 mg/ml) before flow cytometry monitoring.
Viable cells of Lisferiu monocyfogenes (non-treated cells) exhibit a maximal fluorescence intensity after cFDA staining (Fig. 2a) and do not take PI (Fig. 3a). Killed cells of Listeriu monocytogenes are not stained by cFDA (Fig. 2b) and take PI (Fig. 3b).
432
High Pressure Food Science, Bioscience and Chemistry
C
Figure 2 : 1;luorescence ititensity at 530 nm (cFDA) of Lisferiu rmmocyfogenes cells motiitored by flow cytometry. (a) iion treated cells, (b) heat-treated cells - 15
min- I2 I "C, (c) high-pressures treated cells - I0 rnin-400 MPa, pl I 5.6.
Biochemistry: Posters
433
8
-. m n.
CELLIJLES ?IIURIIIES
CELLLILES MORIES
rn
IZI
1
1
I
C
Figure 3 : 1;luorcscciice irilcrisily at 650
tiin
monitored by Llow cytomctry. (a)
treated cells, (b) lieat-treated cells
noti
(PI) of Lisferiu nto/tocyfogenes cells
inin- I2 I “C, (c) high pressures Irealed cells - 10 min-400 MI’a, pl1 5.6.
-
15
High Pressure Food Science, Bioscience and Chemistry
434
Afler high pressure treatment (10 min-400
ma, p1-I 5.6) no viable cells were
recovered by plate count method. These cells were stained with cFDA at a lower intensity than viable cells (Fig. 2c) but the major part of the population did not take PI (Fig. 3c). These observations show that a part of the bacterial population kept enzymatic activity and membrane integrity afler high pressure treatments. Further studies will show whether high pressure treated cells are capable of revivification as it has been related by Raffali et al. (1 994).
CONCLUSlON
The effect of high hydrostatic pressures on Lisferiu monocyfogenes is strongly pH-dependent. In further studies we will show the interaction between pH and other factors on the effects of high pressures on cell inactivation. Flow cytometric studies on Listeriu monocytogenes suggest that high pressure
treatment at 400 MPa could lead to reversible cell-damage. Cell constants measures are in progress to determine the physiological mechanisms that are involved in the action of high pressures on microbial cells.
(1) : RAFFALI J., ROSEC P., CARLEZ A., DUMAY E., RICHARD N.,
CI-IEFTEL J.C. 1994
-
Stress et inactivation par hautes pressions de Lisferiu
innocua introduites dans une crknie laitikre. Sci.Afim.- 14, (349-358).
High Pressure Germination and Inactivation Kinetics of Bacterial Spores V. Heinz and D. Knorr
Department of Food Biotechnology and Food Process Engineering Berlin University of Technology, Kiinigin Luise Str.22, D-14195 Berlin Tel:+49 30 314 71441 Fax:+49 30 832 7663 E-mail: heinl33 [email protected]
1. INTRODUCTION
Based on the kinetic data accumulated in our laboratory regarding the combined action of pressure and temperature on bacterial spores (Fig. 1) a model has been put forward [13 assuming enzyme activities of the germination pathway as the key element (Fig. 2). /'
_-
Core
Plasma Membrane Cortex Outer Membrane Coat
Figure 1 Schematic diagram of the spore structure.
00 pressure 003
R
R'
R"
0 0 temDerature 0 0 3 CORTEX:
-3.-
LJ
s""" WGSLE~
-
.< ..I.
.I
Figure 2 Proposed mechanism of pressure induced spore germination, adopted and modified &om [Z].
436
High Pressure Food Science, Bioscience and Chemisrzy
Similar to the alanin induced germination, an activated proteolytic enzyme R cleaves a cortex associated 'gemination specific lytic enzyme' GSLE resulting in a successive degradation of the spore cortex peptidoglycan and a loss of typical dormant spore properties [2]. There is increasing evidence that the unusual pressure-temperature behaviour of germination kinetics is due to the complex interaction of pressure-temperature activated and/or inactivated enzyme systems involved in triggering the irreversible germination pathway. Instead of allosteric activation as the first step during the Foster-JohnstonePathway, pressure-temperature effect is assumed to initialize the germination. This is consistent with the observation of accelerated enzymatic reactions under pressure not higher than 300 MPa.
"
AND METHODS 2. MATERIALS
Cultures of Bacillus subtilis ATCC were obtained from DSM (Deutsche Sammlung von Mikroorganismen, Braunschweig - Gemany. Spore production was done in the following way: The vegetative form was cultured at 30°C in Standard-I nutrient broth (Merck 7882, Darmstadt - Germany) in two steps until the stationary state of growing was reached. This suspension was used to inoculate manganese containing agar plates (5.0g Peptone from meat, trypsin-digested (Merck 7214, Darmstadt - Germany); 3.0g Extract of meat dry (Merck 3979, Darmstadt - Germany); 12g Agar Bacteriological (Oxoid Code L1 1, Basingstoke - England); deionized water 1000 mL; 10 mgL MnSO4 ' 2H20). The plates were cultured for at least 3 days until 90 % phase bright spores were visible by phase-contrast microscopy. After harvesting, the spores washed and centrihgated (5000 x g for 20 min) 2x in deionized water, l x in ethanol (70%) and again 3x in distilled water. The concentrated suspension ( 1'109spores per mL) was stored at -20°C. Absorbance measurements The concentrated spore suspension was diluted with Ringer's solution to a concentration of approximately 1O7 spores per IT& and transferred into a pressure cell with sapphire windows. After removal of the air and closing the pressure unit, a spectrophotometer was connected to the cell by fibre-optics. Absorbance measurements (h = 580 nm; against air) were performed with an aquisition rate of 1s-'. Absorbance is indicated relative to the initial value, measured before pressure build-up started. The rate of pressure increase was 20 MPa s-' . Viable counts The number of spores surviving compression and the remaining heat-resistant spores were estimated from colony counts. Immediately after the treatments,
Biochemistry: Posters
431
serial dilutions in Ringer's solution were prepared. The "drop plate" counting method was employed using nutrient agar (5.0g Peptone Erom meat, trypsindigested (Merck 7214, Darmstadt - Germany); 3.0g Extract of meat dry (Merck 3979, Darmstadt - Germany); 12g Agar Bacteriological (Oxoid Code L11, Basingstoke - England); deionized water 1000 mL). After 48 h of incubation at 30°C the colonies were counted. Measurement of the remaining heat-resistant spores (ungerminated spores) After the pressurization treatment, samples (0.1 mL) were added in 10 mL preheated Ringer's solution (70 "C). The surviving heat-resistant spores were enumerated after 15 min of treatment at that temperature. It was predetermined that this treatment was sufficient to kill the germinated, but not the ungerminated spores. Determination of the efflux during germination Release of DPA (2,6-pyridinedicarboxylic acid), a typical mass transport process occurring during germination, was detected by the kinetics of accumulation in the medium. Immediately after pressurization, the spore solution was passed through a syringe filter (0.22 pm), in order to separate the germinating spores fiom the solution. DPA concentration of the filtrate was detected using a counter-ion HPLC method: coloumn: RP-I 8; eluent: phosphate buffer pH 7.3 /methanole (15%)/ 2.3 g L tetrabutylammonium hydrogensulfate; detection: UV at 271 nm. The DPA concentrations are indicated relative to the suspension's spore concentration. Dimensionless description was possible by connecting the results with the average DPA content of single spores, determined previously by a complete mechanical disrupture of the spores in a ball mill (Heinz, 1997). The germination index G takes values between 1 (ungerminated spore suspension) and 0 (completely germinated spore suspension).
3. RESULTSAND DISCUSSION Besides the other indicators like the release of dipicolinic acid (DPA) and heatresistance, the in situ measured change in optical density during pressure treatment (Fig. 3) shows clear germination time optima when plotted versus pressure (Fig. 4). Increasing temperature extremely speeds up the germination rate of Bacillus subtilis ATCC 9372 by accelerating the mass transfer from the core of the spore into the medium. At 38OC no further acceleration is observed suggesting that an effective transport limitation occurred in the surrounding layers of the core.
High Pressure Food Science, Bioscience and Chemistry
438
Figure 3: In situ scanning of the optical density OD (rel. absorbance at 580 nm) of suspended B. subtilis ATCC 9372. ioooo
Maximum Germination Rate Absorbance Measurement (58Onm)
0
sapphire windows
Bacillus subtilis ATCC 9372 0
50
100 150 200 250 300
Pressure [MPa]
Figure 4: Temperature-pressure characteritics of OD-data from Fig. 3. The ordinate represents the time coordinate where the highest germination rates were detected.
439
Biochemistry: Posters
More quantitatively, DPA release - monitored as the remaining interior concentration (DPA, located inside the core)- showed a similar behaviour (Fig. 5 , 6 ) . Taking into account the population’s diversity described by a distributional hnction and considering changing mass transport limitations, a combined model was used for quantitative analysis. The discontinuity of the germinative effect of high pressure at 38°C could be confirmed.
08
08
8 4
0.6
08
06
0.4
04
04
m 5 0.2
02
0.2
00
00
n
150 MPa
10
* 1.o C P) * c 0.8
10
CI
Q)
.-*>
K 0.0
10
10
E08
08
0.6
08
.-cP)> 0 4
04
5 K
02
*
0
100 200 300 400 500
0
100 200 300 400 SO0
0
Q
n
a Q
0.2
00
00 0 400 500
0
100 200 300 400 SO0
Treatment time [s]
Figure 5 : Release of dipicolinic acid (DPA) from the core of B. subtilis ATCC 9372 into the medium. The ordinate represents the DPA concentration inside the core relative to its initial value. At low temperatures pressure induced germination detected as a release of DPA always was observed initial to inactivation. (Fig. 7). Pressure treatment at 50°C or more seems to cause direct inactivation without complete germination. At 70°C and 150 MPa the kinetics of total survivors and heat-resistant survivors take the same shape (Fig. 8). It is assumed that vital parts of the germination system are affected by the combined action of temperature and pressure. Reduced or inhibited cortex degradation may yield in a delayed germination or a complete loss in viability.
440
High Pressure Food Science, Bioscience and Chemistry
10‘
10’
10’
0.08 0.06 0.04 0.02 0.00
10 20 30 40 50 60 70
Temperature [“ C]
Figure 6: Model of DPA-release of B. subtilis ATCC 9372 during pressure induced germination.
O
0
% % %
200
4m
em 10
08
06
04
02
00 0
200
400
800
200
400
200
(100
Treatment Time
400
600
[s]
Figure 7: Temperature effect on germination (0 DPA-release; heat resistant survivors) and inactivation (A)of B. subtilis ATCC 9372 during pressure treatment (150 MPa). The strong correlation between the heat resistant survivors and the remaining DPA content suggests that the DPA release of the suspended spore population is discontinuous.
441
Biochemistry: Posters
Survivors [cfulmL] after Treatment: 150 MPa
150 MPa + [15 min I IO'C IO.1 MPa]
f-
10'
10'
10'
10'
10'
10'
1o6
I 0'
10'
10'
10'
10'
10'
10'
10'
10'
0
1om
10'
10'
10'
10'
10 '
1on
10'
10'
10'
10'
to'
1o1
10'
1o2
10'
0
400 800 1200 1600 2000
0
400 800 1200 1800 2000
Treatment Time
0
400 800 1200 1800 2000
1.
Figure 8: Temperature effect on inactivation. Separation of still dormant spores after pressure treatment was done by an additional heat treatment (70°C / 15 min) prior to incubation. 4. CONCLUSION Whereas the lethal effect of high pressure on bacterial spores at ambient temperatures seems to be small, germination can be triggered even at pressures as low as 150 MPa. Increasing process temperature strongly accelerates the detectable degradative events. With Bacillus subtilis pressure-temperatureconditions (150 MPa, 70°C) could be identified were 6 log-cycles reduction in spore count were reached. Further work is underway to detect in vivo the germinative effect of pressure and temperature on the frequency dependent electrical conductivity of bacterial spores. 5. REFERENCES [ 11 Heinz, V. (1997) Wirkung hoher hydrostatischer Driicke auf das Absterbe-
und Keimungsverhalten sporenbildender Bakterien am Beispiel von Bacillus subtilis ATCC 9372, PhD-Thesis, TU-Berlin. [2] Johnstone, K. (1994) The trigger mechanism of spore germination: current concepts, J. Appl. Bacteriol. (Symposium Supplement) 76, 17s-24s.
Physiological Responses and Morphological Changes of Salmonella typhimurium and Listeria monocytogenes to High Hydrostatic Pressure Treatments J.L. Tholozan8*,M. Ritzb,G. Delattre', M.F. Piletband M. Federighib
'Food Process Engineering and Food Technology Laboratory, National Agronomic Research Institute (INRA), P.O. Box 39, F-59651 Villeneuve d'Ascq Cedex, France National Veterinary School at Nantes, Route de Gachet, Case Postale 3013, F-44087 Nantes cedex 03, France
Salmonella sp. represents one of the leading causes of food spoilage outbreaks in the world. In the U.S.A., the total costs of food contamination by this genus include approximately 60% of the costs of food spoilage due to bacteria, and represent 3.99 lo9US $ per year (Anonymous 1995). Spoilage by Listeria is the third cause of spoilage with only 5% of the spoilage costs due to bacteria. In France, food spoilage outbreaks with Salmonella represent 83% of food contamination in collective food preparation plants, but only 50% of illnesses, while Listeria food spoilage represents 10 cases per lo6inhabitants, i. e. much higher percentage than in the USA or other European countries. Two main species of Salmonella are involved in food contamination : S. typhimurium, and S. enteritidis, most frequently (Lepoutre et al. 1995). Eggs and derived products are often concerned with such a contamination. These bacteria are ubiquitous, and are the most frequent aetiological agent of gastroenteritis in man. Bacteria from genus Salmonella are straight rods, Gram negative staining, non sporing, with 0.7-1.5 pm in diameter, and 2.0-5.0 pm in length. These bacteria are facultatively anaerobic (Le Minor 1986). The physiology of this genus demonstrates a growth from 5 to 47"C, with an optimum at 35-37°C. The genus is not heat resistant, and killed by heating at 72'C for 15s. It survives at 2'C. though some psychotrophic strains have been isolated, and demonstrated a slow growth at 2'C. The pH range for development is 4.5 to 9.0, with an optimum at pH 6.5 to 7.5. A food spoilage by Salmonella indices high fever, severe abdominal pain, diarrhoea. These symptoms often disappear within a week, with almost never a fatal issue (Gledell996). Food spoilage by Listeria appeared after the fifties, while Salmonella was known long ago (Frazier and Westhoff 1978). This is also a widely distributed bacterium, with only one species which seems to be pathogenic for man : L.
rnonocytugenes(Catteau 1996). It has been isolated from sewage, soil, and healthy
Biochemistry: Posters
443
human and animals. A food contamination by Listeria results in severe diseases such as septicemia, meningitis, encephalitis in animal, and in man, elderly humans and new-born children are the most exposed to listeriosis. The genus is represented by regular short rods from 0.4-0.5 pm in diameter and 0.5-2.0 pm in length. The bacterium is Gram positive staining, and asporulated.The cells grow from 1 to 4 5 T , with an optimum growth temperature at 30-37'C. Numerous psycrotrophic strains have been described. It is killed by 30 min heating at W C , but survives at -1'C. Good growth is described between pH 6 and pH 9, with an optimum pH range near pH 7. All strains grow in complex media, containing up to 10% NaCl (w/v). The non thermal processes for food preservation have been used since 1899 (Mertensand Knorr 1992). However, a systematicapproachof this type of treatment to stabilize food has been described recently, and effect of a high isobaric treatment on food microorganisms has been extensively reviewed (Hoover et al. 1989). An important decrease in cell counts has been measured as well in yeasts,as inmoulds, bacteria, and spores. Previous results on bacteria demonstrated that Gram positive microorganisms often displayed a higher resistance to pressure (Cheftel 1995). Scanning electron micrographs have shown a mechanical disruption in Succharornyces cerevisiae after a 500 MPa pressure treatment, and the formation of vacuoles and holes in the cytoplasm of L. monocytogenes and Salmonella after a 250 MPa treatment have been demonstrated by Mackey et al. (1994) with transmission electron microscopy. The commercialuse of high pressure technology for food stabilization is routinely found in Japan with friut juices, cooked fish fillets and rice cakes. In the laboratory scale, different types of milk, meat and derived meat products have been stabilized, as well as fatty goose or duck liver in France (Shigehisa et al. 1991,Gaucheronet al. 1997). We have worked on the physiological effects of high pressure treatments on Salmonella typhimurium and Listeria rnonocytogenes Scott A. Late log-phase cells have been resuspended in phosphate (pH 7.0) and citrate-phosphatebuffer (pH 5.6) before treatment to demonstrate an effect of the pH value of the external medium on cell sensitivity to pressure. The effect of pressure treatment on cell integrity has been followed by scanning electron microscopy at different pressure values. In addition, cell size modifications after the pressure treatment have been assayed by flow cytometry on a Facscan (Becton Dickinson) by side scattering measurements. A simultaneous measurement of internal pH and potassium content of the cells has been performed with radio-labeIled probes and absorption atomic spectrophotometry, respectively. The high pressure apparatuswas from GEC Alsthom ACB at Nantes, and each750 mlsample has been placed in sterile cap sealed polyethylene bags (Gualapack, Italy), and treated in a 3 1vessel with a pressure range from 1 to 6,000am. Pressure treatment effects were assayed by standard plate counts in TSA and TSB culture media.
444
High Pressure Food Science, Bioscience and Chemistry
Pressure treatments were performed during 10 min at 20°C. The pressures of treatments were selected to produce a moderate inactivation of the cells (less than 2 logs), a medium (approximately 4 logs), and a strong inactivation of bacteria (higher than 6 logs). Results are presented below for each buffer and bacterium (Fig. 1).
B
Figure 1. Effect of a pressure treatment in (A)SuZmonellufyphimurium resting cell suspension, and (B)Listeriu monocyrogenes resting cell suspension. Both cell suspensions were treated in phosphate (pH 7.0), or in citrate-phosphatebuffer (pH 5.6).
Biochemistry: Posters
445
Cell counts were lower for a same pressure treatment in citrate-phosphate buffer. In addition, for a similar decrease in cfu/ml in both Salmonella and Listeria cultures, a lower pressure treatment was needed in S. typhimurium. Scanning electron microscopy revealed a progressive modification of the cell shape when increasing the pressure value. Important invaginations were demonstrated with the pressure increase in Salmonella typhimurium. Highest pressure treatments in Listeria monocytogenes led to the formation of evaginations on the cells (Fig. 2 and Fig. 3).
Figure 2. Reference Salmonella ryphimurium suspension in phosphate buffer pH 7.0 without a pressure treatment
446
High Pressure Food Science, Bioscience and Chemistry
Figure 3. Salmonella typhimurium resting cell suspension in phosphate buffer after a 325 MPa pressure treatment (10 min) in phosphate buffer Flow cytometry did not demonstrate a change in cell size, neither in S. typhimurium nor in L. monocyfogenes.The cell size remained constant whatever the pressure treatment. However, a progressive decrease of the ApH between external and internal pH was measured. A medium or a high pressure treatment led to a small if any pH difference. In a same manner, the treated cells were not able to maintain a high internal potassium content with a moderate pressure treatment. A medium pressure treatment (300 MPa) led to an almost complete efflux of potassium, suggesting cell death for most of the cells.
In conclusion, Gram negative cells are most sensitive to pressure treatments. The moderate pressure treatments correspond rather to a cell inactivation than to a cell death as demonstrated by the results above. References Anonymous (1995) Food poisoning, an overview. Int. Poultry Prod. 3:20-21 Catteau M (1996) Listeria. In Microbiologie Alimentaire, Tome 1,Bourgeois CM,
Biochemistry: Posters
441
Mescle JF, Zucca J (eds), Tec et Doc Lavoisier, Paris, p. 89-103 Cheftel JC (1995) Review : High-pressure, microbial inactivation and food preservation. Food Sci. Technol Int. 1:75-90 Frazier WC, Westhoff DC (1978) Foods in relation to disease. In Food Microbiology, McGraw-Hill Book Compagny, New York, p. 417-472 Gaucheron F, Famelart MH, Mariette F, Raulot K, Michel F, Le Graet Y (1997) Combined effects of temperature and high-pressure treatments on physicochemical characteristics of skim milk. Food Chem. 59:439-447 Gledel J (1996) Le genre Salmonella. In Microbiologie Alimentaire, Tome 1, Bourgeois CM, Mescle JF, Zucca J (eds), Tec et Doc Lavoisier, Paris, p. 62-79 Hoover DG, Metrick C, Papineau AM, Farkas DF, Knorr D (1989) Biological effects of high hydrostatic- pressure on food microorganisms. Food Techno1 43:99- 107 Le Minor L (1986) Salmonella. In Bergey’s Manual of Systematic Bacteriology, vol. 2, Krieg NR,Holt JG (eds), Williams and Wilkins, Baltimore, p. 427-448 Lepoutre A, Salomon J, Charley C, Le Querrec F (1995) Les toxi-infections alimentaires collectives en 1993. Bull. Epidemiol. Hebdo. France 52:245-247 Mackey BM, Forestihe K, Isaacs NS, Stewing R, Brooker B (1994) The effect of high hydrostatic pressure on Salmonella thompson and Listeria monocytogenes examined by electron microscopy. Lett. Appl. Microbiol. 19:429-432 Mertens B, Knorr D (1992) Development of nonthennal processes for food preservation. Food Technol 46:124-133 Seeliger HPR, Jones D (1986) Listeria. In Bergey’s Manual of Systematic Bacteriology, vol. 2, Krieg NR, Holt JG (eds), Williams and Wilkins, Baltimore, p. 1235-1245 ShigehisaT, OhmoriT,SaitoA,Taji S,HayashiR (1991)Effectsofhighhydrostatic pressure inactivation on characteristics of pork slurries and inactivation of microorganisms asociated with meat and meat products. Int. J. Food Microbiol. 121207-216
Peroxidase Reaction under High Pressure: Influence of Different Hydrogen Donor Molecules
P. Butz, A. Fernhdez Garcia, B. Tauscher
Institute of Chemistry and Biology, Federal Research Centre for Nutrition, Engesserstr. 20, D-7613 1 Karlsnihe, Germany.
INTRODUCTION As a consequence of cell destruction during hgh pressure treatment, enzymes of
h i t are no longer separated &om their substrates. Pressure conditions may lead to
an acceleration of enzyme reactions or a different affhty to different substrates due to changes in the structure of proteins, so those in the tertiary structure of enzymes could affect their specifity, which may form or increase concentration of undesirable compounds. Effects of hgh pressure between 0.1 and 550 MPa on commercial horse-radish peroxidase reaction velocity were measured by spectrophotometric methods. The reactions tested have been peroxidations taking place when hydroquinone or guajacol as hydrogen donors were in the reaction mixture.
449
Biochemistry: Posters
MATERIAL AND METHODS Saturated solutions of horse-radish peroxidase in TRIS buffer, pH7, with two different substrates (guajacol and hydroquinone) were pressurized in the range from 50 MPa up to 550 m a . H,O, was added to start the reaction. The changes in
absorbance of both samples (pressurized and unpressurized control) were measured in a spectrophotometer at 288 nm for hydroquinone and 435 nm for guajacol.
RESULTS AND DISCUSSION As Fig 1. shows, the velocity
J
I."
1.8
of the reaction taking place
with guajacol as hydrogen donor, decreased when rising pressure to about 0 at 550 0
54
I00
150
MQ
250
3M
350
400
450
500
554
Pre.."rr (MPn)
Fig 1 Relative concentration of product in the peroudase reaction after pressure treatment with two different hydrogen donors (guajacol0 and hydroquinone V)
MPa (final concentration of product similar to the one in the minute 2 of the reaction,
just before pressure treatment). When hydroquinone was taken as substrate, the velocity was not influenced by pressure, and the concentration of the product was
450
High Pressure Food Science, Bioscience and Chemistry
similar to the one in the unpressurized control. So pressure effects on enzyme reactions are also a fiinction of the nature of the substrate.
REFERENCES ANESE, M. et al. (1995): Effect of high presstire treatments on peroxidase and polyphenoloxidase activities. Journal of Food Biochemistry, 18,285-293. BUTZ, P. et al. (1994): IJltra-high pressure processing qfonions: Chemical and sensory changes. Lebensmittel-Wissenschafl und Technologie, 27,463-467.
GOMES, M.R.A.; LEDWARD, D.A. (1996): l$ect of high-presstire treatment on the activity of some polyphenoloxidases. Food Chemistry, 56, No. 1 , 1-5. SEYDERHELM, I. et al. (1 996): Pressure induced inactivation of selected food enzynies. Journal of Food Science, 61, No. 2,308-310. V h O S - V I Z G Y k O , L. (198 1): Polyphenot oxidase and peroxidase in Jriiits and
vegetables. CRC Critical Reviews in Food Science and Nutrition, 15,49-127.
Proteins under Extreme Conditions: FTIR Spectroscopy with a Cryogenic-High Pressure Cell
F. Meersman', P. Rubens', L. Smelle? & K. Heremans'.
'Department of Chemistry, Katholieke Universiteit Leuven, B-300 1 Leuven, Belgium. Email : [email protected] 21nstituteof Biophysics, Semmelweis University of Medicine, Budapest, Hungary
In this paper we present the results of a study in which we have submitted horse
skeletal metmyoglobin to a stepwise low temperature and high pressure treatment. The changes in conformation were followed with FTIR. It seems that there occurs no intermolecular P-sheet aggregation at low temperature, while previous studies have shown that it does happen at high temperature. No P-sheet aggregation can be observed at hgh pressure.
We tentatively
conclude that there is a difference between cold, heat and pressure denaturation of metmyoglobin.
452
High Pressure Food Science, Bioscience and Chemistry
1 . INTRODUCTION
Proteins show an elliptic phase diagram in the temperature pressure plane (Smeller & Heremans, 1997). Consequently this means that proteins can be denaturated not only at hgh pressure or temperature, but also at low temperature.
In thls study we investigate the appearance of the phenomenon of intermolecular P-sheet aggregation at low temperature and high pressure in metmyoglobin. Intermolecular P-sheet aggregation is characterized by the appearance of two distinct peaks near 1615 cm-' and 1680 cm-' respectively (Jackson et al., 1991). T h s aggregation is held responsible for the fact that temperature denaturation is irreversible in most cases. To examine thls phenomenon we have chosen the oxygen-carrying protein myoglobin.
2. MATERIALS AND METHODS
Metmyoglobin from the horse skeletal muscle (Sigma M-0630) was dissolved in lOmh4 TRIS-DCl buffer pD 7.6. The samples were stored overnight to ensure that the WD-exchange could reach a static regime. The solution was mounted in a stainless steel gasket of a Diamond Anvil Cell (Diacell Products, Leicester, UK). A small amount of BaS04 was added to determine the pressure, which is
Biochemistry: Posters
453
obtained from the position of the sulphate peak near 983 cm-I in the deconvoluted spectrum (Wong & Moffat, 1989). The DAC was then placed in the cryocell. The pressure was built up by a helium driven membrane. The liquid nitrogen cooling system, that is a part of the cryocell, allowed us to lower the temperature down to 170K. A pressure build up was still possible at this low temperature. Infixed spectra were obtained with a Bruker IFS66 FTIR spectrometer equipped with a liquid nitrogen cooled broad band MCT solid state detector.
250 Interferograms were coadded after
registration at a resolution of 2 cm" .
3. RESULTS AND DISCUSSION
Metmyogobin was submitted to two cycles in the temperature pressure plane. In the first cycle the temperature was lowered to -1OO"C, then the pressure was increased to 9 kbar and finally both temperature and pressure were brought back to their initial conltions (lbar, 25°C). In the second cycle we first increased the pressure to 9 kbar and then lowered the temperature to -100°C. Figure 1 shows the results of the cold and the heat denaturation, which has been studied earlier. It seems there appear two distinct peaks near 1683 cm-' and 1615 cm-' due to heat denaturation. The combination of these two peaks is characteristic for intermolecular P-sheet aggregation. No formation of such peaks can be observed
454
High Pressure Food Science, Bioscience and Chemistry
in the case of cold denaturation. These results lead us to the conclusion that there is a difference between the mechanism of cold and that of heat denaturation.
1700 1680 1660 1640 1620 1600
Wavenumber / cm-1 Figure 1 : h d e I' region of cold (broken line) and heat (full line) denaturated metmyoglobin. The cold denaturation spectrum shown here was taken at -1 OOOC, i.e. in the ice state. The combination of the two peaks near 1680 cm-' and 1615 cm-' is characteristic for P-aggregation. The dotted line represents pressure denaturated metmyoglobin at 9 kbar.
Furthermore there is a difference between temperature and pressure denaturation. The latter also does not lead to P-aggregation (figure l), although in both cases a gel can be seen at the end of the experiment. This means that intermolecular
Biochemistry: Posters
455
interactions also occur due to pressure denaturation, but that they are of a different, weaker nature (Heremans et al., 1997) than those that are a consequence of temperature denaturation. The formation of ice at lower temperatures was visible as spectral changes in the absorption regons of water between 3100 cm-' and 3700 cm". These changes include a narrowing of the absorption peaks, due to the fact that the influence of the surrounding on the molecules becomes more uniform, as well as a shift towards lower wavenumbers. In the amide I' region we observed a sudden increase in peak intensity upon ice formation. Strambini and Gabellieri (1996) followed the phosphorescence lifetime of tryptophan residues in liver alcohol dehydrogenase in aqueous solutions upon freezing. They suggested that the protein adsorbs to the growing ice crystal and consequently partially unfolds. The formation of ice, however, is undesired in a cold denaturation experiment. Privalov (1990) discussed some solutions for this problem. Among these were the use of denaturants, which allow the observation of cold denaturation above the freezing point of the solvent. On the other hand, if one does not want to use a denaturant one can always use pressure or add an antifreeze such as glycerol to avoid solidification of the solvent. The problem that arises here is that these factors also effect the stability of the protein and therefore complicate the
456
High Pressure Food Science, Bioscience and Chemistry
experiment. Unfortunately no method that allows the study of cold denaturation without complications has been found up to this moment.
4. ACKNOWLEDGMENTS This research was supported with a grant of the European Community, project number FAIR-CT96-1 175.
5. REFERENCES
Wong, P.T.T. & Moffat, D.J., Appl. Spectrosc., 43, 1279-1281 (1989) Smeller, L. & Heremans, K., in High Pressure Research in the Biosciences and Biotechnology (Ed. K. Heremans), Leuven University Press, 55-58 (1997) Jackson, M., Harris, P.I. & Chapman, D., Biochemistry 30,9681-9686 (1991) Heremans, K., Van Camp, J. & Huygebaert, A,, in Food Proteins And Their Applications (Ed. Damodaran, S. & Paraf, A.), Marcel Dekker Inc., New York, 473-502 (1997) Strambini, G.B., Gabellieri, E., Biophys. J. 70,971-976 (1996) Privalov, P.L., Crit. Rev. in Biochem. and Mol. Bio. 25,281-305 (1990)
In situ Microscopic Observation of Pressure-induced Gelatinization of Starch in the Diamond Anvil Cell
J. Snauwaert., P. Rubens, G. Vermeulen, F. Hennau and K. Heremans Department of Chemistry, Katholieke Universiteit Leuven, 300 1 Leuven, Belgium. Email: [email protected]
The pressure induced gelatinization of aqueous suspensions of starch has been studied by phase contrast microscopy and DSC (l), in sifu microscopic observation (2) and Fourier Transform Infiared (FTIR) spectroscopy in the diamond anvil cell (3).
We combined the diamond anvil cell and the
microscope for in sifu observations up to 10 kbar. The gelatinization process resulted in an intensity increase and band narrowing of the IR spectrum. No swelling or gelatinization could be detected with the microscope below 5 kbar for undamaged potato starch granules.
458
High Pressure Food Science, Bioscience and Chemistry
1. INTRODUCTION High pressure treatment of aqueous starch suspensions at moderated temperature activates the gelatinization process whereby the highly oriented amylose and amylopectin polymers loose their birefringence by imbibition of water. This process is generally interpreted as a transition process from an initially crystalline state to an amorphous state. The amylose and amylopectin separates into two phases, each phase enriched up to 70-80% (4). This high pressure induced demixing process can be followed in situ by measuring the swelling of the starch granules and observing the change of the polarization cross of the starch granules in the microscope. It has been shown that infiared spectroscopy is very usehl to monitoring the gelatinization process (5). In this work we combined the in situ microscopic observations with the in situ FTIR measurements for studying the pressure induced gelatinization process of potato starch granules. These starch granules have a considerably higher potential for swelling than cereal starch granules (6,7).
MATERIALS AND METHODS Potato starch granules from Sigma were suspended in water in a diamond anvil cell (Diacell Products, Leicester, UK) mounted with a 50 pm thick
Biochemistry: Posters
459
stainless steel gasket. The pressure was determined by the ruby technique, which consists of measuring the shift of the ruby fluorescence with a Spex Raman spectrometer. Infiared spectra were obtained with a Bruker IFS66 FTIR spectrometer equipped with a broadband MCT detector. The infrared beam was focused on the sample by a NaCl lens and the average of 250 interferograms at a
resolution of 2 cm-' was registrated. The Fourier self-deconvolution, a
mathematical technique for band narrowing was performed with the BNker software. The experimental setup consists of a diamond anvil cell placed on a light microscope (Olympus BH2) with a long focus objective (x20). Images are recorded by a JVC CCD-camera and transferred to a Snappy video digitizer (Play Incorporated), connected to a parallel port of an IBM-compatible computer. The Snappy video digitizer records pictures in 16.5 million colours with a resolution of 1500 x 1125. The pictures are processed with Coreldraw 4.0 to determine the diameter of the starch granules.
RESULTS AND DISCUSSION Previous infrared spectra studies of aqueous solutions of polysaccharides showed that the region 1400-800 cm-' is sensitive to polymer conformation (8).
460
High Pressure Food Science, Bioscience and Chemistry
Most bands in this region are coupled C-0and C-C stretching modes of the polysaccharide backbone. The infrared spectrum of the potato starch granules suspensions at ambient conditions (30°C and low pressure) shows several bands in that region. Band narrowing and increasing absorbance can be observed with increasing pressure. The ratio of the heights of the bands at 1047 and 1022 cm” expresses according to van Soest et al. (9) the amount of ordered to amorphous starch. In that region we observe two bands, 1040 and 1012 cm-’ which saturated very quickly with increasing pressure. The evolution of the ratio of the intensity bands at 1157 and 1127 cm-’ are shown in Fig. 1. A transition of the intensity ratio is observable at pressures higher that 5 kbar.
No
gelatinization could be observed in the microscope up to 5 kbar.
i 0
1
2
3
4
5
6
7
8
Pressure (kbar)
Fig. 1 : The evolution of the ratio of the intensity bands at 1157 and 1127 cm-’ as a fbnction of pressure.
Biochemistry; Posters
461
Destruction of granules larger than 50 pm in diameter can occur because the thickness of the diamond cell gasket is 50 pm and decreases with increasing pressure.
Initially damaged potato starch granules with cracks in their
amylopectin shell are marked by mows in Fig. 2al. Visualization of the microstructure of the gelatinized starch in the diamond anvil cell are done as described with an optical microscope. No swelling of undamaged potato starch granules, which are around 30 Fm in diameter, was observed up to 5 kbar. Birefkingence of the granules remained constant up to a pressure of 5 kbar (Fig. 2b1, 2b2). The gelatinization of these small granules takes place above 6 kbar. No cracks could be observed during the gelation of these granules. Larger granules (> 60 pm) have initially already cracks in their shell, presumably due to shearing forces which appear on mounting the diamond anvil cell. According to the severity of the suffered damage the gelatinization process of these granules starts at lower pressures than their smaller analogues (Fig. 2c 1,2c2). Since no swelling could be observed below the transition pressure, we investigated if the pressure induced gelatinization process could kinetically be determined. To our surprise we found that after 91 hours at 4 kbar and 30°C no swelling of potato starch granules was observable for granules with a diameter
462
High Pressure Food Science, Bioscience and Chemistry
a2
Fig. 2 : Photographs with normal (1) and polarized (2) light of a suspension of potato starch granules in the diamond anvil cell at 30°C and at: a l ) and a2) 0.7 kbar, b l ) and b2) 4.9 kbar, c l ) and c2) 6.4 kbar. Note that some granules may be displaced due to the volume decrease in the diamond cell at high pressure. The polarized pictures are negative images. The arrows indicate granules with damage.
Biochemistry: Posters
463
less than 60 pm. Damaged granules with a diameter around 75 pm swelled only 5 % and the 90 pm granules swelled 25 %.
ACKNOWLEDGEMENT This research was supported with a grant of the European Community, project number Fair-CT96- 1 175.
REFERENCES 1. S. Ezaki & R. Hayashi in: High Pressure and Biotechnology, (Eds.C.Balny,
R. Hayashi, K. Heremans & P. Masson) INSEMJohn Libbey Eurotext Ltd., 163- 165 (1 992). 2. J.P. Douzals, P.A. Marechal, J.C. Coquille & P. Gervais, J. Agric. Food
Chem., 44,1403-1408 (1996). 3. P. Rubens, K. Goossens & K. Heremans in: High Pressure Research in the
Bioscience and Biotechnology (Ed. KHeremans), Leuven University Press, 191-194 (1997). 4. M.T Kalichevsky & S.G. Ring, Carbohydr. Res., 162,323-328 (1987). 5. R.H.Wilson & P.S. Belton, Carbohydr. Res., 180,339 (1988)
464
High Pressure Food Science, Bioscience and Chemistry
6. H.W. Leach, L.D. McCowen, T.J. Schoch, Cereal Chemistry, 36,534-544 (1959) 7. M.R. Williams & P. Boyler, StarcWStiirke, 34,221-223 (1982) 8. P.S. Belton, R.H. Wilson & D.H. Chenery, Int. 3. Biol. Macromol., 8,247251 (1986) 9. J.J.G. van Soest, H. Tournois, D. de Wit & J.F.G. Vliegenthart, Carbohydr. Res., 279,201-214 (1995)
Physics: Presentations
Properties and Structure Peculiarities of High-impact Polystyrene obtained by Hydroextrusion under Pressure N.V.Shshkova*, B.M.Efrosand S.A.Tsygankov Donetsk Physics & Technology Institute of the National Ukrainian Academy of Sciences, 72 R.Luxemburg St., 3401 14, Donetsk, Ukraine Phone,fax: 3 80 +(622) 5 5 7462; e-mail:[email protected]
I.Introduction Brittleness is a substantial limitation (drawback) for the amorphous polymers, which restricts the field of their application. Some physical and chemical methods for modification of these materials were developed to overcome the above-mentioned limitation. The increasing of molecular mass, addition of different alloying elements, copolymerization with raw rubber are the chemical methods. During the last years some works concerning the polymer modlfication by physical methods were done. In particular, works on formation of the oriented state in brittle amorphous polymer have shown that it is possible to improve strongly its strain characteristics too [l]. There are some methods for the creation
of oriented state in polymer materials [2]. Hydrostatic extrusion is a promising method for amorphous polymers allowing to carry out the oriented deformation under high hydrostatic pressure [3,4]. As is well known, the hydrostatic pressure suppresses the fi-acture
High Pressure Food Science, Bioscience and Chemistry
468
formation at deformation of brittle materials to a great extent and thereby it promotes the realization of a hgh deformation without fracture . 2.Experimental results and discussion The aim of h s work is the study of structure and properties of high- impact polystyrene (HIPS) subjected to the hydroextrusion. The experiments were carried out with the help of equipment and by the method described in detail in Ref.[5]. Commercial hgh-impact polystyrene (HIPS) was used as a subject of inquny. The characteristics were &=20%, ot=30MPa, E=2000MPa, a=35kJ/m2. The mechanical characteristics were determined for circular section specimens with diameter of 6 mm. Intermedlate products were obtained by the method of casting under pressure. Next the specimens were made by turning treatment to the required sizes. Hydrostatic extrusion was carried out at different pressure, temperature (350 ...413K) and strain rate (rate of the specimen advance) up to 20 cmlmin. We obtained a stabilization of the process and high quality press-specimens at the strain rate of the specimen advance up to 10 cdmin. The temperature range that allows one to obtain the maximum effective strain (deformation) degree (R=6) is 373...393 K. The interconnection of the optimal technological parameters of the process can be analyzed according to the Table. Press-specimens look as the bars with diameter of 6 mm. The characteristic features of obtained specimens were the elevated transparency and rigdlty. The specimen's surface was very smooth and without roughness. These specimens failed under the alternating-sing loads only after some cycles of the loading. The results of physico-mechanicalcharacteristics of strain specimens have been shown in the Table too. For strain specimens the
469
Physics: Presentations
value of elastic modulus (Young’s)is 1.5 times hgher than those for initial state, tensile strength is higher (40%) too. Table. The values of hydroextrusion parameters and the main properties and structure characteristics of HIPS specimens.
T,K
I
R
I PJ”a
1 E,MPa lot MPa 2340
373
393
413
h,
h2
f
35
0.15
0.10
0.10
40
0.35
0.30
0.15
45
0.40
0.40
0.20
2.0
10
2620
35
38
0.35
0.30
0.15
3.0
24
2950
42
45
0.40
0.45
0.20
3.5
35
3100
45
60
0.50
0.50
0.25
2.0
5
2600
38
50
0.35
0.30
0.20
3.0
6
2500
38
75
0.50
0.45
0.25
4.0
12
3500
50
*
0.60
0.60
0.40
6.0
16
3800
57
**
0.85
0.80
0.45
2.0
5
2540
35
*
0.30
0.30
0.20
3.0
5
3080
38
0.50
0.45
0.30
4.0
5
3400
50
0.60
0.55
0.35
0.80
0.80
0.45
5
6.0 1
a,kJ/m2
I
3910 I
50 I
** ** ** I
- transparency is lost (partially), **-transparencyis lost (we&
1
R=In D?/Df2, where Di-and Df -initial and final diameters of the specimen
Impact strength has grown sharply and this points to the plastic characteristics increasing. For a number of the specimens obtained at the high strain degrees the dehtion of the impact strength values was a failure (for
High Pressure Food Science, Bioscience and Chemistry
470
specimens marked with *and
**,
see the Table) at the testing of the specimens
without the notch, because they have not failed at the maximum applied loading yet. At the impact testing it was found that the development of main crack takes place both in transverse and longtudinal direction. At the increase of strain degree the development of the main crack occurs mainly in the longitudinal direction mainly and at considerable loading the crack origmates and develops only in the transverse duection. It is important to draw the attention to the fact that the extruded specimens have got a hgher value of impact strength at the elevated temperature. Thus unique combination of the mechanical properties, as judging fkom hfferent references, was not reached with the use of other modification methods for these materials. Structure of the HIPS specimen after plastic deformation under hydrostatic pressure conditions was studied by TEM and X-ray analysis. For the TEMs studies the ultrathin-microscopic sections were prepared. They were contracted by the osmium acid according to the methods [ 6 ] .In the Figure the microstructures of the HIPS hydroextrusion specimens are shown. Structure of the initial state material was as that of two-phase system, where elastomer rubber phase with the polystyrene (PS) accommodated into it was dispersed into the continuous polystyrene matrix. At the increase of the strain degree one can observe the deformation of elastomer particles and their transformation into ellipsoid. Substantial structure transformation of the HIPS during the hydroextrusion process, and this is very well seen in the Figure, occurs without failure opposite to the deformation process under atmospheric pressure [4]. The evaluation of form change of elastomer phase inclusions
(A,)and polystyrene
(A*), contained in it, were carried out according to ratio of elliptic
Physics: Presentations
47 1
particle axes by method in Ref. [7]. The numerical values of mentioned magnitudes are shown in the Table. The increase of the strain degree gives rise to sharp change of anisodiametricity of elliptic particles of rubber phase and accommodated particles of the PS. Moreover their relative change was the same. The orientation of rubber particles corresponds to the extrusion duection.
a
b
C
d
Figure. Microstructure HIPS specimens after hydroextrusion (T=393 K), x 40 000: a, b- longitudinal direction structure: a-R=3; b-R=6: c, d - transverse direction structure: c- R=3, d- R=6.
High Pressure Food Science, Bioscience and Chemistry
412
Simultaneously with modification of the elastomer phase form and accommodated PS particle the molecular orientation of the PS matrix takes place. The numerical values of matrix orientation degree (f) are shown in the Table. They were determined with X-ray dtfliaction data at a large angle by the change of form and intensities of reflex [8]. The analysis of these data has shown that the increase of Young's modulus is connected with the increase of orientation degree
of the PS matrix. However it is not large, thereby we cannot observe the sharp increase of Young's modulus and tensile strength. The sharp increase of impact strength can be explained by changing the mechanism of deformation of oriented impact strength polystyrene. As is known in the material with unoriented matrix at any applied loadtng direction the multicrack formation takes place at the boundary with rubber phase in the perpendicular direction to applied loadtng i.e. in the more dangerous one. And
thls makes condition for the fast nucleation of main crack in the same direction. For the HIPS with oriented matrix the primary crack formation takes place always in the direction of orientation which is more energy-advantageous. If the expanding loadmg is realized in direction of orientation axis (usually it takes place for the profile products), the appearing microcracks do not initiate the crack nucleation in dangerous direction. In this case the nucleation of the main crack resulting in fracture at places of discontinuity of separate microfibrilar formation and, consequently, in the sigmficant higher levels of stress and deformation. The higher deformation characteristics (in the orientation direction) of the HIPS subjected to hydroextrusion are a consequence of the above-mentioned fracture mechanism.
3 .Conclusion Thus, the hydroextrusion of amorphous polymers and composites on their basis can be used with a high success for the obtaining of profile type products
Physics: Presentations
413
with unique combination of physico-mechanical properties, which cannot be achieved by other methods. References: [ 11 Appelt B., Porter R.S. J.App1.Polym.Scz., v.26, N9,2841( 1981).
[2] Ultra-high modulus polymer, EdA.Chiffery & 1. Ward, Leningrad.:Khimiya, 1983. [31 B.1 Beresnev, N.S.Enikolopov, S.A.Tsygankov., N.V.Shshkova. Dokl.
ANUkrSSR, Seria B, N4,47(1985). [4] S.B Iynbinder, E.L.Tyunina, K.I.Tsyrule. Polymer properties under different stressed state. M.: Chemistry, 1981. [5] S.A.Tsygankov., N.V.Shishkova, B.I.Beresnev. Physics & Technics of High Pressure, N17, 77 (1984). [6] Kato Koichi. Polymer Eng. & Sci., V.7, N1,38 (1967) [7] GiufiiaR. J.Appl.Polym.Sci.,V.7,N1,333 (1963). [8] May M. J. Polymer Sci. Polymer Symposium, V.15,N1,23 (1977).
Pressure Dependence of Thermal Conductivity of Rocks
U. Seipold
GeoForschungsZentrum Potsdam, Telegrafenberg A1 7, 14473 Potsdam, Germany
Introduction At the first glance, the different physical properties of rocks show a similar pressure dependence. The petrophysical parameters rise sharply with pressure and then asymptotically approach a straight line with a smaller slope. A typical example is reproduced in Figure 1. However, in a detailed consideration, the various petrophysical parameters have different underlying physical mechanisms and therefore, one can expect different reactions against increasing pressure. Pnmarly, in the low pressure range open pores and cracks are closed by the compaction. This process will influence the physical properties.
475
Physics: Presentations
Experimental procedure For the investigations of the thermal properties a pulse method was applied. This enabled us to determine simultaneously both thermal conductivity and thermal difisivity. For details of the method see Seipold (1988).
16
3
l25.
5.
Y
5' 2
10 O, u) 39
x
Thermal Conductivity __ 3.82(10,.0864exp@,76p))(1+0.0157p)
a
figh pressures up to 10 B a r were generated by means of a 400 ton hydraulic ram which pushed the piston into a hgh-pressure vessel of a piston-cylinder apparatus. The cylindrical rock samples with 43 mm length and 27 mm diameter were protected by jacketing with polyurethene against the penetration by the pressure transmitting liquid. The investigations were carried out on magmatic and high grade metamorphic rocks with porosities smaller than one percent.
High Pressure Food Science, Bioscience and Chemistry
476
Results The measurements using a high pressure-resolution clearly show that for the thermal parameters the range of linear pressure dependence is reached at
5,O
-
I
I
>
3-
4-5:
$
4,o-
I
m
,'* 1
e
*
*
I"s
aI
c/) 1
.s" .-
**
3,5-
*
*
*
0
O
0
9
8-
7P - cn.2
0
5.
0
0
- 6 %
0
3'0:
1:
1
0
--cJ 0
Roo'
e@ 0
175
9
I
0
0
0
.;=
0
O
- 5
ShearVelocity Thermal Condudidty
1
I
0
Cornpresjonalveioaty Thermal affuavity
I
I
8
4
4
pressures smaller than 0.5 Bar. For the elastic velocities the linear range begms at much higher pressures (see Figure 2). In order to quantify this result, the measured values were fitted to the function
K = &*(.l + b * exp( -c * p))*( 1 + a * p) with K = thermal conductivity,
p = pressure,
& = thermal conductivity,extrapolatedto zero pressure, b,c = constants, describing the low pressure range, and a = slope in the linear pressure range.
Physics: Presentations
477
While the constants a and b for elastic and thermal properties are in the same order of magnitude, the value of c is about three times greater for the thermal transport parameters than for the elastic wave velocities. The inverse value l/c is a measure for the width of the low pressure range. Averaging over the studied sample collection yielded c=(5.26 f 1.71) Bar-1 for the thermal parameters and c=(1.67 f 0.29) Bar-1 for the elastic wave velocities. For pressures higher than about 0.5 B a r the investigated magmatic and metamorphic rocks showed a relatively small increase of thermal transport parameters with rising pressure in the order of 0.3 to 1.2 percentkBar. A higher slope seems to be connected with the content of quartz. Susaki and Horai (1989) obtained a pressure coefficient of 2.95 percent/kBar for quartz in the crystallographc c-direction and 5 percentkBar perpendicular to the c-axis. We found a value of 3.77 percent/kBar for quartzitic sandstone in the pressure range up to 10 kBar which excludes the effect of porosity. Thermal cracking caused by high-temperature treatment provides the possibility to create adhtional homogeneously distributed cracks within a rock sample. In Figure 3 results are shown for a granite sample which was heated up to 300 "C and then cooled down to room temperature before the high-pressure investigation was carried out. Obviously, also the effect of artificially created cracks is limited to pressures smaller than 0.5 B ar.Similar results were obtained for samples heated up to 600 "C.
High Pressure Food Science, Bioscience and Chemistry
478
l,O
!
I
I
I
I
I
I
0
2
4
6
8
10
I
Pressure inkBar
Discussion of results We have a completely different situation for the thermal transport parameters than for elastic and electrical properties. The thermal resistivity is influenced by macroscopic as well as microscopic effects. Macroscopic effects are made by open airfilled cracks and pores, representing a very low conducting phase within the solid rock matrix. These effects can be described by the application of mixture theories. On the other hand, an important part of the total thermal resistivity is caused by the scattering of heat carrying phonons by hfferent types
of phonon interactions. These microscopic phenomena include the interaction of phonons with various lattice defects including grain boundaries, and also the various types of phonon-phonon interaction caused by the nonlinearity of the
Physics: Presentations
419
lattice potential. When a crack is closed by the application of high pressure, the macroscopic view has to be changed to a microscopical point of view. With this view, one has to take in mind that a closed crack remains as a scattering centre. This also occurs at very high pressures in so far as it represents an interruption of the crystal lattice. At room temperature and higher , the mean free path of phonons in a rock sample is in the order of lo-* cm and smalIer.This distance is small compared with the distance between adjacent cracks. This means that only a relatively small number of scattering processes is caused by closed cracks. Therefore, they have little influence on the heat conduction process. For the explanation of the dfferent behaviour of elastic properties, the propagation of elastic waves can be considered in classical physics. The transmission of elastic energy becomes better and better if the faces of a closed crack are pressed together by increasing pressure. At sufficient high pressure the cracked rock sample behaves llke a compact solid. This state is reached at a few kilobars. Contrary, the closure of open cracks up to about 0.5 mar is the reason for the low lying upper limit of the low pressure range for the thermal transport properties. Conclusion
In the discussion of the pressure dependence of the thermal transport properties one has to take into account peculiarities caused by the mechanism of heat propagation in rocks.
480
High Pressure Food Science, Bioscience and Chemistry
Acknowledgement The author wishes to thank Dr. H.-J. Mueller, GFZ Potsdam, who has carried out the high-pressure investigations of the elastic wave velocities. References Horai, K. and Susaki, J., 1989. The effect of pressure on the thermal conductivity of silicate rocks up to 12 kbar. Phys. Earth Planether. 55:292-305 Seipold, U., 1988. Simultaneous measurements of thermal difisivity and thermal conductivity under high pressure using thermal pulses of fitllte length. High Temp.-High Pressure 20: 609-613
Application of Natural Diamonds for Generation of Super-high Pressure B.M.Efros & N.V.Shishkova Donetsk Physics & Technology Institute of the National Ukrainian Academy of Sciences, 72, R.Luxemburg St. 3401 14 Donetsk, Ukraine Phone, fax:+380 (622) 557462; e-mail: [email protected] 1.Introduction
The diamond anvil became a widely used tool due to the application of the ruby method for pressure determination and the possibility of obtaining quasistatic megabar pressures. The application of gem quality single crystals for anvil manufacturing permits due to their transparency into a large range of the lengths
of electromagnetic waves to carry out the optical, X-ray and spectroscopic studies of researched substances at high and super high pressure. Choice of the diamonds and cutting of them are the essential factors for achievement of superhigh static pressure. The presence of nitrogen plate formations in diamonds influences substantially on the anvil resistance to the deformation. 2. Results and discussion
In this paper we have tried to estimate the dependence of an ultimate state of a diamond anvil on a sort of pressure distribution on the workmg surface and
mechanical properties of a diamond. We have considered shape and geometry parameters of an anvil according to recommendation [l]. The ratio of the anvil
High Pressure Food Science, Bioscience and Chemistry
482
height to the gndle dlameter is 0.3 and the ration of the anvil height to the basis diameter is 0.7. The analysis of this problem is divided in two steps. At the first step we defined the stress- strain state of a diamond anvil. The solution is obtained by the method of final elements . The dlamond is considered as an isotropic body with the Poison’s ratio of v=0.7 and elastic modulus E=ll41GPa. The four different models of loading are chosen to define the boundary condltions on the working surface of the dlamond anvil. These are models with uniform, trapezoidal and trapezoidal distributed pressures with a different depth of lateral support (Figurel). A uniform distributed loading is applied at the anvil basis. The Pisarenko-Lebedev’s criterion is used for determination of a dangerous zone locations and analysis of ultimate state of a diamond anvil. xai+(l-x)ol%t , where
x=
o&b~ - factor brittleness; o1 -the intensity of stresses, oh -the
compressive strength, obt-the tensile strength. The theoretical evaluation of tensile strengths was made by M.Ruoff at the compression and tension for the different crystallographic directions of perfect dlamond crystals. Using these data we evaluated the values of a brittleness factor
(x)for different crystallographc directions. It is determined that x changes within the interval of 0.1-0.4. It is clear that the brittleness factor can depend on the defect density and their distribution in the volume of real crystal. Since the experimental data are absent we analysed the interval,whch was defined using the Ruoff s data. The diagrams of dlstribution of equivalent stresses are shown in the Figure 2. In the case of the uniform distributed pressure on the anvil working surface for all four values of x the geometrical place of the point of maximum equivalent
483
Physics: Presentations
50
50
\
1
50 50
I
I
I I I
C
I I--------
\
I
Figurel. The different sorts of pressure distribution (GPa) on the anvil working surface: a-uniform distribution ; b- trapezoidal distribution; c, d - trapezoidal
with a different depth of lateral support. stress (cr-
) coincides with the point of the maximum value of r, and location
of the dangerous section does not depend on the x value. There is one dangerous zone allocated yet at an anvil axis at the depth of 0.1 of the anvil height. The value of an equivalent stress decreases and it equals to 0.09-0.14 ,,a
close to
the basis. There are two dangerous zones near the working surface of an anvil
High Pressure Food Science, Bioscience and Chemistry
484
y: 0.f
3 = 0.2
X r 0.I
Yz0.2
x=0.4
x=
0.2
Xz0.3
xr
X = 0.4
0.3
X = 0.4
Y z0.3
I=0 . 4
Figre 2. Distribution of equivalent stresses (GPa) into a dlamond anvil (the domains of compressive stresses are drew):
- 0.5 cmx;
-.-.-. - O.1cmx; -x-x-x- - O.OlOmx
Physics: Presentations
485
too. That is why on the other diagrams we have shown a top part of an anvil only. The results of calculations for other loading models have revealed the
existence of dependence of location of dangerous section from value of a brittleness factor x. With the growth of x the value of ,,a
moves to an axis and
at lateral support the maximum value is at the axis. Thus, we can see a role of lateral support: lateral side is protected fiom appearance of the critical stresses. The oemax decreases on 26-39% in comparison with other sort of loading. 3.Conclusion
The different sort of inclusions, containing of nitrogen plate formation, present in the real materials. And it impossible to predict where will be the more dangerous zone if we do not know the sort and the orientation of inclusion and its influence on the stress distribution. Therefore it should be taken into account the existence of such the two dangerous zones on the axis of an anvil and close to a lateral surface. References [ 1 ] H.K. Mao, P.M. Bell. Curnegze Ins?. Wash. Yearb.N7,646 (1 977).
121 B.Beresnev, Ya.Beigelzimer, SStishov and B.E&os, High pressure. phys.
and Technol.15,39.(1984).
Safety in Pressure Testing G. Saville and S.M.Richardson, Imperial College, London; and,
B.J. Skillerne de Bristowe, BJS Research, Holly Cottage, Clay Lane, Beenham, Reading, RG 75PA Tel: 44-1 18-9713722 Email: [email protected]
Abstract Quantitative determinationof the consequencesof a pressure vessel failing should form part of the hazardassessment conducted prior to the testing of pressure vessels and experimemtal work. A method has been developed for the detemhation of possible fragment and supersonic shock generation during pressure vessel failure and the protection needed to contain them. The work here was underfaken to provide advice to the UK Health and Safety Executive and will form the basis for the next edition of the High Pressure Technology Association's, High Pressure Safety Code.
Introduction The objective of thii paper is to provide information relevant to the performance of the hazard assessment and to indicate the measures which can be taken to reduce a hazardous operation to a low risk one. Issues which will be considered include the determination of the energy stored in a pressurised system, how that energy will be released in a failure and how it can be contained.
Hazard Assessment We divide pressure tests into the following types, in order of decreasing hazard: 0 Research proof test: one carried out on a new design of vessel, or where metallurgical data such as creep rates and fatigue life are to be determined, or where autofiettage is carried out. 0 Roof test: one carried out when a vessel is built to an established design and applied when the vessel is first fabricated or significantlymodified or repaired. 0 Leak test or hnction test: one carried out to test the correct hctioning of the equipment after it has first undergone a p m f test.
Physics: Presentations
481
Most pressure vessels used for laboratory work will fall in the research proof test -gw. The two main hazards during pnssun testing are the formation of missiles and the generation of a shock wave. The failure can be of the item being tested,clamping equipment holding it in place or the source of pressurising fluid. Experimental evidence', indicates that shock waves am not formed when the pressurising medium is a liquid. On the other hand, shock waves are the norm when pressuring with a gas or a saturated volatile liquid. In contrast, missiles can be generated whatever the psurising medium.
Stored Energy The damage caused by a pressure vessel failure is largely detennined by the amount of energy stored at the time of failure. In the absence of chemical reactions,the main source of energy is the fluid expansion energy, &. We estimate thii energy by assuming that the expansion process ish -t ' Ily revmibk and Surscientlyrapid for t h e to be negligible heat transfer to the surroundings. The process is, therefore, isentropic and
Figures 1 to 4 show E, for nitrogen and various liquids for a range of pressures up to l 0 k h when expansion is to atmosphericpresswe.
Fragmentation The size and initial speed of fiagments f m e d by a pressure vessel when it fails depend on the mode of failure. We can classify the relevant modes of failure as: 0 complete bgmentation due to brittle fracture,usually into a large number of thpents; 0 complete destruction of the vessel in a ductile manner, usually into a small number of rather large fiagments: a typical example would be a large welded vessel where failure begins in a longitudinal seam, extends from end to end of the vessel and then runs mund the circumfexmttial seamsjoining the cyliider to the domed end caps; 0 loss of a major section of the vessel in a ductile manner a typical example w ould be failure of the circumferentialweld between the cylinder and one end cap; 0 loss of a closure, such as a manhole in a large welded vessel or an end closure in a vessel where the closuregives essentially full-born access to the vessel; loss of a plug or other small closure, which gives only limited access to the vessel.
High Pressure Food Science, Bioscience and Chemistry
488
Figure 2 Fluid expnnsion energy for nltrogan
Figve 1 Fluid e x p ~ s l o nenergy for nitrogen
126
.
. .. . .... . ... .... . . .. ..... . .. ... .. . .... . .... . ... . .... . ... .... . ..
: : . : : : : : : : : : " : : ............................................. . ... . .... . ... . .... .... . .... .... . .... .... . .... .... ... . .. ..... .. .. .. ..
26
E '
f3
1
. . . . . . . . . . . ._..................... .. ;.;.I.; .:. . . .:. .:_ .:. .:. ;. ;. I.;. . . . .. ........................................... . . . .-. ................ . . . . . .-. . .-. ...... I
0
c
.v. .I.
1.............I
5 5
100
L
I
.c
.. .. .. .. .. .. .. . . . . . . . . . . . . . .. ..
,6
...........................................
10
.; . _
?,/ .:_.I 1 .
7
-.. - . - . - .
.c
. ... . ... . .... .... ......... . . . . .........
. .. . .. . .. . ... . . . . . 76 ........ ........
........
;.
.> .:..:. .:. .;. .:. ;. ......... . . .. .. .. .. .. .. .. .. .. .. .. .. .... .. .................................... ........ . . . . .\ . ................................. . . . . . . . . . . .,. .-.:
. . . . I : .. . I /
. . . . ..... - .- ................
. ... . ... . ... . .... ... . . ..- .._ .- ................ . . . .-..- . .--. .. . .................... . .. . .. . .. . .. . ..... . . . . . . . . . .. . .. . .. . ._ .
. .. . ... .... .... . ... .... .... . ... . .... .... . ... . .... ... . ... . .... ..
60
. .. ... ... .. . . . :. . . . . , ....... . ........ . . . . . . . ........ . . . ........ . . . . . ..... . .. L .... . . . . .............. . . . I.. -..-.... -. . i
I
i
i .
. , . I .
26
0
I .
1000
0
2000
SO00
4000
Preeeure / bar
Figure s Fluid expanoion energy for Iiqulde
...
..
.. .
..
.
.
...
. ..
.
.
.
. . . . . . .. .. . . .. . . . .. . . . ........ . . . . ....
..
..
..
..
..
..
..
.. ..
..
;/
.
.
.
Flgure 4 Fluid expansion energy for liquids
. . . . . . . . . . ... . . . .... . . .... . . ... . . . . . . . .. . . . .... . . .... . . ..
100
.
... .
. . . . . . . . . . .;. . . . ;. . . . . . . . . . . . . . . . . . . . . . ... . . . ... . . ... . . . . : .......:2:. . .
i.
p .... .
;g
... .
76
... .
I j ; J.1
. . ... . . . .... . . .... . . .... . . . . . . . ... . . . .... . . .... . . ..
:. . . .c .
.
..
..
..
..
* .
. . . . . . . .... . . . . .... . . . . . ... . I .. .................... . . .. . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . : : ; ; ....................................
1 ; q :I
50
26
0
0
600 Pressure I bar
1000
0
6
Pressure I kbar
10
Physics: Presentations
489
In the last three cases, failure leads to production of two hgments, one small and one large. If the large hgment is sufficiently massive or is adequately anchored, it may not move at all, so that only one genuine hgment is produced. On the other hand, an inadequately anchored larger Sagment may have characteristics of a rocket. The most common pressurisiing fluids are water, hydraulic oil and a gas such as air or nitrogen. Figures 1 to 4 show that very much more energy is contained in a vessel pressurised with gas than one pressurised with liquid.
Total fallun, due to britue fracture Experimental evidence, Baum’ and Chri~topherson~, points to all fhgments having the same initial speed when a brittle vessel of uniform wall thickness breaks up into a large number of hgments. Thus the fragment speed is obtained from: E, =+mVz where El, denotes the total kinetic energy (40% of the stored energy), m the mass of the vessel and V the initial speed of each fragment. The situation with regard to liquid-filled vessels is less satisfactory in that no experimental information is available. Accordingly, the only conservative solution is to assume that for the purposes of estimating fragment speeds, all available energy is communicatedto the hgments. As before, we assume that all s t o d have the same Speed.
The number and size of fiqpents produced during brittle failure is in general unknown, apart from the general observationthat the number of hgments pmduced is usually large. We postulate., as a working procedure when assessing the requirements for protection, that one assumes the largest fragment is 20% and the smallest 1% of the shell. In addition, there are the identifiable items which may be ejected intact: end caps,end closums, manhole covers, n o d e s and so on. Each of these items should be identifled and the mass, size, shape and speed of each determined.
490
High Pressure Food Science, Bioscience and Chemistry
Loss of a major secfion In all cases, the expected scenario for a gas-filled vessel involves acceleration of the closure in a two-stage process. In stage 1 gas escapes through the developing circumferentialgap between the closure and the vessel, with choking taking place on the cylindrical b t thus created. This stage persists until the fragment displacementreaches D/4, where D is the diameter of the breach, at which point, the limiting flow area becomes the circular hole in the vessel itself and choking cakes place on this plane. To a first approximation, one can treat the force on the frasmentas constant during stage t and equal to the initial pressure in the vessel acting over the area of the vessel opening (assuming that the volume of gas in the vessel is sufficiently large for this to be reasonable). Duringthe second stage, the situation is more complex since thejet leaving the choked vessel opening will undergo expansiondownstream and exert a varying force on the fr-agmentwhich will depend on the relative dimensions of the vessel opening and the w e n t as well as properties of the fluid.
Moore' proposed that the initial closure acceleration should be treated as beiig maintained until the w e n t is clear of the vessel by an amount equal to the diameter of the hole left in the vessel, that is, the initial pressure within the vessel is maintained on the underside of the closure hgment until this clearance is reached. The total distmce over which this force would operate would therefore be the diameter of the hole plus any additional travel which the closurewould have to make before a clear leakage gap was formed (this latter would thus be very important for screwed plug closures but irrelevant for domed ends on a welded vessel). Experiments' suggest that this is a reasonabk approximation, even if conservative in some cases. In the case of a liquid-filledvessel, the fluid pressure within the vessel will decay vcry much more rapidly than when it is gas-filled. Therefore, the assumptionthat the accelerationof the closure rcmains constant until a gap equal to one diameter has opened up may no longer be true. The assumptionwill be closest to being valid when the volume of the vessel is very large compared with the volume swept out by the closure during its escape. This suggests the foUowing procedurt to evaluate the speed of the closure: 0 evaluate the speed assuming the material within the vessel is 8 gas; 0 assume that the whole of the stored energy within the vessel (based on a liquid-filled vessel) is converted into kinetic energy of the closure and hence determine the speed; 0 take the actual speed as the smaller of the above two speeds.
Physics: Presentations
49 1
Speed of a rocketing fragment A n analysis ofthe rocketing motion ofa vessel which has lost its end can be undertaken
by assuming a two-stage process: in shge I, which incorpordles the period when expansion waves move up and down
the vessel: assume that conditions in the vessel do not dilTer significantly from h e initial conditions; this stage lasts until the vessel has lifted by a height equal to a quarter of its diameter: thereafter the area for efflux becomes constant at the crossscctiotial arca of tlic vcsscl; in stage 1 I, assume that there is reversible adiabatic (isentropic) decay o f pressure of a perfect pax inside the vessel; this stage lasts until the acceleration of the ves.wl reaches Lero and its velocity is a maximum. lrin addition it is assumed that gravitational eKecls are negligible and flow out of he vessel is choked, it can be shown that the maximum velocity Vmax of the vessel is given by:
?tDL 2 Here: a = po__ (mls )
h
pa is the initial pressure in the vessel (Pa);
pa is the ambient pressure (Pa); '1'" is the initial temperature (K);
L is the length of the vessel (m); D is the inside diameter of the vessel (m); m is the mass of the vessel (kg); C,, is thc dischargc cocficicnt (a valuc of unity is suegcstcd); y is the ratio ofheat capacities for the gas;
M is the molar mass of the gas (kflmole); and,
It is the universal gas constant (8.3 I43 J/K.mol).
492
High Pressure Food Science, Bioscience and Chemistry
Loss of a plug or small closure The situation is similar to that for loss of a major section except that the volume of the vessel is so large that there is negligible loss of pressure until the plug is well removed fiom the vessel. The force on the plug is therefore maintained at a higher level for a rather longer period of time. Experiments conducted into this situation, Baum6,are consistent with the assumption that the full system pressure remains acting on the plug until the clearance between the tail end of the plug and the breach in the vessel reaches twice the diameter of the plug. In the absence of any information to the contrary, we propose using this same procedure for liquid filled vessels as well but with the proviso that the total kinetic e n e w of the missile should not exceed the stored energy of the fluid in the vessel.
Protection Against Fragment Perforation Most work on missile perforation of targets is of ordnance origin, although in recent years the importance of ‘industrial missiles‘ and the damage which they can cause to plant has been recognised and some research has been directed to this, mostly by the nuclear industry. This distinction is important since hgments (missiles) produced by failure of pressurised equipment tend to have speeds which rarely exceed 500 m/s, whereas ordnance missiles tend to have speeds in excess of 500 mls. The mechanism of penetration and perforation at lower speeds (a few hundreds of d s ) is not the same as at nearer lOOOmls,thus makiig the extrapolation of ordnance results to lower speeds a somewhat uncertain operation. Protection provided by steel A number of empirical correlations can be found on the penetration and perforation of steel by pmjectiles. Each covers a limited range of variables and use outside the range of fitting is dangerous in the extreme. However, given sufficient of these correlations, it is possible to develop an overall safety envelope around them. The actual development of such an envelope is complicated by paucity of published experimental results and the incomplete knowledge of the range of applicability of the correlations available. We have tested a number of these correlations. In general we have found that the BRL correlation’, produced by the Ballistics Research Laboratory (BRL),always produced the largest perforation thickness over a very wide range of variables and was usually closest to the experimental values where these were available. The BRL correlation in SI units is:
493
Physics: Presentations
10667
t = 4 . 9 10.' ~ (mV-) '
/d
whcre t is thc thickness of the shield for pedoration by SOYOof missiles (m); m is the mass of the missile (kg); V is the speed of the missile (mk); d is the diameter if the missile (m). RRI, suggcst increavingthis thickness by 25% in order to stop pertbration by all missiles, an adjustment which we believe should be made. Also, in order to avoid a rather p w r perlbrmance Tor thin shields we put a lower limit of 3mm on all shield thicknesses.
Blast and Impulse Loading of Structures During Failure of a gas-filled pressure vessel, as much a$60% of the s t o d energy is converted into a blast (or shock) wave. However, the impact of a missile with a wall also results in the transference of kinetic energy of missile to the wall in he form of an impulse and this has a very similar effect to the impingement of a blast wave. Our knowledge ofthe generation af hlaq waves by a hiling pressure vessel and their ctlicct on structures is very limited. Most ofthe available information is for military explosives. conventional and nuclear, and f i r chemical explosions such as gas explosions. In order to provide a link to the stored energy in a pressure vessel, we use: I kg afTNT is equivalent to a stored energy o f 4 5 MJ.
Although the blast characteristics o f a TNT explosion and a pressure vessel fnilure are quitc ditlierent in ficx air, it is fortunate that due to the numerous reflections of the shock on the walls, the diflerences are very much less marked when the explosions take place in an enclosed structure such as a cubical. This means h a t experimenls conducled with high explosives can be used to predict how a pressure vessel failure within a protective cubical will behave. An additional factor. when compared with an explosion in the open, is the rclcasc of gas within thc cubical. If thc vcnt arca of thc cubical is suficicntly small, this gaq will build up an internal pressure, which decays relatively slowly. Bake? has shown that this complex behaviour can be approximated by a simple model. I le assumed that the overall shock pressure on each wall of the cubical could be treated it5 a triangular pulse of duration typically I millisecond. Meanwhile, gas pressure builds up,
494
High Pressure Food Science, Bioscience and Chemistry
reaching a maximum in a time comparable with the shock pulse and is then followed by an exponentialdecay to atmospheric pressure. Thus each wall of the enclosure is subjected to two pulses, one which is intense but of short duration, and a longer-lived gas load which has a very much lower peak pressure. To simpli& the design calculations, we require the structure to withstand each of these pulses separately rather than cumulatively. Thus we seek an 'engineering' solution commensuratewith our relatively imprecise knowledge of the nature of a given pressure vessel failure. protective 'cubicles' can range from a single cantilevered wall at one end of the scale to a fully enciosed 4-wall plus roof cubicle at the other (all cubicles are assumed to have a floor). In all cases except the fully enclosed cubicle, the 'missing' walls provide sufficient venting to prevent any build up of gas following the explosion. Thus the slow release of gas referred to in the previous paragraph only takes place when a hlly enclosed cubicle with a limited vent area is used. If the vent area is very small, this release can take several seconds.
Internal shock loading of protective cubicles Based on Ayvazyan et a19we have derived charis for the average shock loading of the walls of a cubical chamber where the explosive source is placed at the centre of the cube (or at the centre of the notional cube in the case of a single wall). Figure 5 gives the average peak reflected pressure and Figum 6 the average reflected impulse on each wall. The letter codes refer to : A. a single vertical cantilever wall; B. either wall of a 2-wall cubicle or the side wall of a 3-wall cubicle or the roof of a 2wall cubicle with toof C.the back wall of a 3-wall cubicle or any wall of a 4-wall cubicle or either wall of a 2wall cubicle with roof or the side wall or roof of a 3-wall cubicle with roof; D. the back wall of a 3-wall cubicle with roofor any walf or roofof a 4-wall cubicle with roof. The scaled distance used in both fippres is the ratio of the distance of the explosive source &om the wall (one half of the internal length of the cube in our case)to the cube root of the energy released during the explosion. The same scaling, by the cube root of the energy release, is also used for the impulse. Although the curves are strictly applicable only to high explosives, we presume their validity for shocksproduced by failing pressure vessels.
495
Physics: Presentations
. ......... . . . . ......... . , . . ......... . .
. . . ...
............. 1 I;;/;. . . . .; . . I::::::: . . . . . . . . . .I::::: ... . . .. .. .. .. .. ... .. .. .. .. .. .. . . :
. . . .
......................._.. ..*... ......." ...................... * . * ...............
. .
. . ......... .........
.. .. .. .. .. .. . .. .. .. .. .. .. .. .. .. .. . . . :......... ........ . . ............................. 0 v-
High Pressure Food Science, Bioscience and Chemistry
496
Gas impulse loading of protective cubicles As indicated above, gas loading of a cubicle is only relevant when the cubicle totally or almost totally encloses the explosive source. An open wall is sufficient to reduce gas loading to essentially zero. However, walls of very light construction are often used to keep out the elements on nominally open walls, usually termed ‘blow-outpanels’. Ayvazyan’ considers these blow-out panels to have negligible effect provided they will be displaced by a pressure which does not exceed 1 kPa (0.01 bar). The computation of the gas load on a cubicle wall involves the determination of the initial peak gas pressure, at time zero, and its rate of decay. In the case of failure of a gas filled pressure vessel, the temperature of the gas released will be very much lower than ambient. However, the gas released will mix with the ambient air in the cubicle and interchange heat with the walls, so we propose, as a slightly conservative solution, to assume that the gas released from the vessel instantaneously achieves ambient temperature. The calculation of the gaseous over-pressureat time zero is then quite straightforward. The rate of decay of gas pressure depends on the vent area of the cubicle. If there is no vent, this initial over-pressure must be assumed to persist indefmitely.If a vent is present, an approximate equation describing the pressure time history8 is:
p(t) =
i ,ad-2.137) p
where p = p ( t ) / po ;
p$ is the peak gas pressure (bag); po is the ambient pressure (bara);
a,is the ratio of the vent area to the wall area; 4 is the internal surface area (rn’); V is the internal volume (m’); and,
a,is the speed of sound in air at ambient conditions (approx. 34Ms).
Physics: Presentations
491
The duration of the impulse due to the gas release is the time taken for the pressure to fall fiom the initial absolute pressure of pi (=p* + pa) to ambient pressure po.
and the impulse from
where:
c=
“ I
2.13a,A,a,
Missile impact loading The impact of a missile with a wall imparts a transient impulse which the wall, and most importantly, its anchorage, must withstand, otherwise the wall may become a missile itself due to this cause alone. This impact is usually highly localiied. It is this which results in penetration or perforation, but the absorption of the energy/momentumof impact is spread over a much wider area. Ln order to maintain a consistency of approach to the effect of transient loading on protective walls, we assume that this loading takes place uniformly over the whole wall. On this assumption, the impulse experienced by the wall is
ii = mV/A where m is the mass of the missile, (kg); V is the speed of the missile, ( d s ) ; and, A is the area of the wall which is impacted by the missile, (m’). The impulse duration may be estimated by assuming that the missile is brought to rest linearly with time, thus the duration of the impulse is: ti = 2 x N where x is the depth of penetxation of the missile into the wall.
498
High Pressure Food Science, Bioscience and Chemistry
The peak pressure is then, assuming a triangular impulse:
The expected depth of penetration of a missiie, x, is not normally known,however, a sufficiently accurate estimate can be calculated by the methods described above.
Sttuctural response to blast The response of a structureto blast depends on the relationship of the duration of the pulse to the period of vibration of the structure. At one limit, the impulse is over before the structure has had time to move (impulsive loading), at the other, a long slow pulse will cause considerabledeflection before the pulse comes to an end (quasi-static loading). A convenient way to represent loads which cause a given level of damage is as isodamage curves on a peak-pressurev. impulse diagram. A cubicle design procedure based on this concept has been derived for structures to resist the effects of pressure vessel failures.
In order to make the procedure more tractable, we make a number of assumptions: 0 the protective cubicle is cubic in shape but may have one or more sides missing; 0 the item being tested is placed at the centre of the cubicle; 0 the edges of the cube are sufficiently strong so that failure is of the walls of the cubicle and not the edges. When a flat plate, clamped or supported on all four edges, is subjected to an increasing uniform load, elastic failure first takes place at the centre of the plate where the stress is largest. The plate does not, however, totally fail at this stage, but it does start to plastically deform and. as the load is increased further. the plastic region expands until it reaches the edges and the whole plate is now completely plastic. Failure of the plate follows shortly afterwards. This elastic-plastic behaviour greatly complicates the dynamic analysis of the plate, so the engineeringsolution advocated by Baker "is to ignore the plastic behaviour of the plate and to treat it as elastic within the body of the plate and to use the elastic failure criterion only when the plastic-elastic boundary reaches the edges of the plate. The
Physics: Presentations
499
results of a dynamic analysis of a plate which behaves in this way is summarised in the pressure impulse diagram, Figure 7,which shows the iso-damage line. In this diagram
X is the length of each side of the plate (m); h is the thickness of the plate (m); p is the peak impulse pressure (Pa); i is the impulse (pas); d is the density of the plate &@m3); E is Young's modulus of elasticity of the plate (Pa); ayis the dynamic yield strength of the plate (pa); and, f is a numerical factor which equals 0.48 for a square plate.
Figure 7 Pressure-impulse diagram for plates 10 0
a
2% 7 'c a @
6
a,
' " 6
t?
.-
4
U
.I-
3
b
2
2
Z
1 0
0
1
2
3
4
6
6
Normalised pessue PX*/fup?
7
500
High Pressure Food Science, Bioscience and Chemistry
Conclusion Quantitative procedures have been developed to assess the hgments and shock produced during the failure of a pressure vessel and to determine the protection needed to contain them. It is suggested that these considerations should form part of a hazard assessment when conducting high pressure research.
Acknowledgement The information presented in this paper is taken fiom a report entitled 'Pressure test Safety' prepared by the authors for the UK, Health and Safety Executive during 1996.
References I
Esparza, E.D. and Baker, W.E., 1977, NASA Contractor Report 281 1, Contract NSG3005, NASA 2 Baum, M.R., 1984, The velocity of missiles generated by the disintegration of gaspressurisedvesselsand pipes, Trans. ASME, J. Pressure Vessel Technology, 106: 362368'. 3 Cristopherson, D.G., 1945, Structural Defence, Home Ofice, London. 4 Moore, C.V., 1967, The design of barricades for hazardous pressure systems, Nucl. Engng and design, 5: 81-97. 5 Baum, M.R., 1995, Rupture of a gas-pressurised cylindrical vessel: the velocity of a detached end-cap, J. Loss Prev. Process Ind., 8: 149-161. 6
Baum, M.J., 1993, Velocity of a single small missile ejected fiom a vessel containing high pressure gas, J. Loss Prev. Process Ind., 6: 25 1-264. 7 as quoted by Recht, R.F., 1970, Containing ballistic hgments, Third Int. Cod. on High Pressure, pg50, Inst. Mech. Engrs. 8 Baker, W.E., Cox, P.A., Westine, P.S., Kulesz, J.J., and Strehlow, RA., 1983, Explosive Hazards and Evaluation, Elsevier, Amsterdam. 9 Ayvazyan, H., Dede, M.,Dobbs, N., Whitney, M.,Bowles, P. and Baker, W.E., 1986, Structuresto resist the effects of accidental explosions, vol 11, Blast, fragment and shock loads, Special publication ARLCD-SP-84001, US Army armament research, development and engineering centre, Army engineering directorate, Dover, N.J., (NIST document No AD-A 176 673).
Equipment & Systems for High Pressure Mark Freeman - Stansted Fluid Power Ltd 70 Bentfield Road, Stansted, Essex, CM24 8HT, UK.
[email protected] http:/www .sfp-4-hp.demon.co.uk
Introduction
Stansted Fluid Power Ltd, is a company established over 25 years specialking in the design and manufacture of high pressure equipment and systems. This paper looks at the requirements of high pressure equipment operating above 70 MPa and focuses on the requirements for research systems for the emergent applications requiring operation above 400 MPa.
Content Equipment and systems for operation up to 400 MPa has become well established over recent years although takiig a general overview of pressure equipment 70 MPa is a pressure which is generally the limit of commercial industrial hydraulic operations, with most commercial gas systems being below this. The range and use for standard industrial systems tends to plateau out at 70 MPa or below due to increasing cost and diminishing reliability. There are however several major industrial processes and application areas utilising pressures higher than 70 MPa. Offshore oil and gas exploration and production currently with applications up to 200 MPa Homogenisation with application up to 200 MPa Polyethylene production with applications up to 300 MPa Isostatic compaction of advanced materials with applications up to 400 MPa Water jet cutting with application up to 400 MPa
To meet the requirements of these applications, some now established over thirty years, high pressure equipment manufacturers such as Stansted Fluid Power Ltd have developed equipment solutions that are reliable and cost effective for these industries. Now, with emergent applications of high pressure to the food and chemical industry it is necessary to have effective solutions for extreme high pressures up to 1400 MPa. Before looking at the equipment to suit high pressure research up to 1400 MPa it is important to look at the requirements and difficulties associated with such equipment, since if very high pressure applications are Figure 1 Stansted nm-GEN high to be commercially viable for other than niche products pressure homogeniser the technical difficulties must be solved, or the process adapted to the technology
502
High Pressure Food Science, Bioscience and Chemistry
As stated, typical industrial hydraulics are generally below 70 MPa, the principal reasons being that above this pressure cost and reliability become significant factors.
Cost increases because the materials of construction must change to provide improved mechanical properties to contain the pressure, also to resist the severe erosive forces that can occur in high pressure components. In addition as much high pressure equipment is operating on non corrosion inhibiting fluids materials must often be corrosion resistant. Cost increases because the constructionbecomes more complex as valve and seal arrangements must be used while in much lower pressure hydraulics small clearances can be used instead. Reliability will tend to decrease as pressures increase as seals become more prone to wear and also fatigue of materials becomes a significant wear mechanism were increased operating stresses become higher. While below 70 MPa is possible to design systems without significant fatigue problems,aspressure increases this becomes more and more difficult. At less than 70 MPa pressure erosion wear is often negligible and many components work with a constant leak, at extreme pressures even a short duration leak can wear a component to an unusable condition. As pressure increases to 400 MPa the above problems become more acute but engineering
solutions have been developed to meet these requirements that are acceptable to industries utilising them and in general the applications are high added value products.
ASpressures increase above 400 MPa the problems begin to push the existing technology to its limits and often new solutions are required or desirable.
A research system for high pressure usually comprises several elements: Pressure containment - pressure vessel Pressure generator - pump Pressure release - valve Telemetry Measurement and instrumentation associated with the experiment - pressure measurement - Temperature measurement - optical observation
-
In addition the system must be : Safe Reliable Easy to use Safety
High pressure equipment is often perceived as dangerous, and celtainly if improperly designed or operated in an unapproved manner this could be the case, but this would apply to most laboratory tools. In general most high pressure equipment is safe but it is true to say that some designs can reduce any attendant risks and reduce the sensitivity to poor operational practices.
Physics: Presentations
503
Reliability
This is one of the prime requirements of a system as good reliability is not only beneficial for the obvious reasons of less down time and lower operating costs but also the less a system requires attention the lower the risk of the poor maintenance practices and safety interlocks being by-passed.
Ease of use
it has always to be borne in mind by the manufacturers that most users of the equipment are not equipment enthusiasts so the ease of use is essential.
Equipment concerns: Pressure Containment - Pressure Vessels There are many good methods of construction for pressure vessels and it is often a matter of contention between manufacturers as to which is the best - the answer to this is often a question of application and cost, however a highly desirable feature for pressure vessel operating at very high pressures is “leak before burst” construction Pressure Generators
As most pumps for high pressure comprise a high pressure cylinder, seals and valves, the pumping system is often the weakest link in the reliability chain. Fast cycliig pumps as may be appropriate in low pressure applicationsare not suitable for high pressure operation, as pressure increases the norm is to reduce cycling rate, as with industrial high pressure intensifiersused for water jet cutting at 400 ma,which generally operate between 5 and 30 cycles a minute. As pressures increase the tendency is to reduce cycling rate for high pressure components, with the optimum being one cycle or the pump per cycle of the system. Decompression valves. If fluid is released h m high pressure through a restriction there is the potential for severe erosion h m high velocity fluid flowing through small orifices. This is severe in many typical high pressure valves as the flow is focussed on a small annular area between needle and seat. To reduce the wear long capillary restrictions can be used which spread the wear over a larger surface area and this can be coupled with the use of wear resistant materials such as advanced ceramics; however at very high pressures even these methods do not always provide very high reliability. Instrumentation & Telemetry This is often the most important area for the user, and generally there are few difficulties with the regular measurements of pressure ,temperature and leading electrical signals into and out of the vessel. However having optical viewing windows of more than a few millimetres diameter at pressures above 400 h4Pa does present a problem.
High Pressure Food Science, Bioscience and Chemistry
504
General Purpose Equipment Solution To combat the inherent problems, described above, associated with manufacturing reliable and safe high pressure systems, Stansted Fluid Power have adopted the plunger press construction as the preferred style for high pressure systems. The plunger press construction combines pressure vessel and pressure generator into one integral module. The pressure vessel construction utilised by Stansted Fluid Power utilises a two wall high pressure cylinder which is heavily autofiettaged to providing good fatigue properties with the end closures supported by a lowly stressed mantle. The complete construction being a leak before burst design. Added benefit is made of the mantle to act a a fluid gallery for the circulation of fluid for the heating and cooling of the pressure vessel.
S U L KW
li? ?WID
lUNlU
nr C L u W o C l
The high pressure barrel is divided into two sections, the upper section for the sample and the lower for the pressure generation. High pressure is achieved by U the advancing the plunger into the lower section of the high pressure barrel, SIMPLIFIED PLUNGER PRESS CROSS SECTION compressing the fluid and raising the Figure 2 pressure. The high pressure ram is driven by medium low pressure industrial hydraulics.
To decompress the system pressure is released fkom the drive hydraulic circuit which allows the high pressure plunger to retract simultaneously decompressing the high pressure chamber. The system therefore provides 1 2 3 4
leak before burst design for pressure vessel Single shot pressurisation cycling the seals only once per system cycle Typical externally pressurised systems would cycle 10 to 1000 times during pressurisation. No release of high pressure fluid, on fluid below 70 MPa, so valve wear is vitually eliminated No high pressure external to the pressure chamber thus avaoiding high pressure transmission lines and the possibility of high pressure jets. NB high pressure water can cut through carpet and plastics and can cause serious injury.
505
Physics: Presentations
5
Pressure genration and control h m standard industrial hydraulic components operating below 70 MPa which provided flexible control options and low cost maintenance.
Conclusions 1
Research should bear in mind that there is a significant cost associated with using high pressure and attention should be given to keeping pressures as low as possible.
2
Where possible pressure cycles for a high pressure component should be kept to a minium
Figure 3
900 MPa, 2 It. System
Figure 4 900 MPa, 0.3 It. System Model S-FL-850-9-W
Figure 5 Multiple vessel system for kinetic studies 6 x 3Oml vessels MWP 900 Mpa 0-900MPa 2s, 900-0 MPa 2s
Commercial Pressurised Foods in Japan (March 1997, J.C. Cheftel, Universitb des Sciences, F-34095 Montpellier)
FIRM
MEl DI-YA
PRODUCT
PROCESS
Fruit-based products (pH < 4 5 ) :
400 MPa (at 40" Brix m a )
Jams (apple, kiwi, strawb.;
Unit : Mitsubishi 50 I, basket of 150 flexible containers (now with glass bottom)
Jellies Purees "Yoghurts" Sauces Grapefruit juice POKKA Corp. (imported fruits)
10
-
30 min, 20°C
ROLE OF HP Pasteurization Improved gelation Faster sugar penetration But residual enzyme activity (pectinesterase)
Unit : Mitsubishi 40 I and/ or 210 I
Reduced bitterness Limonin destruction at low pH ?)
WAKAYAMA Food Ind.
(winter seasor only) Only = 20% of HP juice in final juice min
300-400 MPa 2-3 min
20°C ? Unit : Yamamoto Suiatsu 3 x 50 I ? Cans, glass bottles, 250 ml Tetra Pak 4 tlh (semi-
continuous) NISSHIN Fine Foods
Sugarimpregnated tropical fruits (kept at -18°C without freezing). For sherbets and ice creams
50-200 MPa
Unit : Yamamoto Suiatsu 4 x 20 I Pieces of pineapple, papayas and mangoes immersed in a fructose syrup
Since 1990, with HP advertising. strawberry puree 1 flexible cylinders, old to food firms & consumers. Shelf-life :
2 x usual price itarted August 1991, with HP advertis. Sold in supermarkets
Pasteurization and hot fill in glass bottles 600 I/ h +
Mandarin juice
2.5 x usual price
2 months at 4°C (1 week if open)
350-400 containers per hour 200 MPa 10-15 min, 5°C
SALES
Shelf-life : 3 months at 4°C educed odor of dimethyl sulfide
Reduced thermal jegradation of methyl nethionine sulfoxide qeplaces 1st thermal pasteurization (after juice extraction) =inal pasteurization Defore packing : 90°C. 3min Faster sugar penetration and water removal (effects on cell membranes and supersaturation ?)
Since October 92, with HP advertis. in special shops. Brand name : lkki Shelf-life : 3 months at 4°C 100 yens/ 250 ml
Since August 1992, without HP advertising No recent information
507
Commercial Pressurised Foods in Japan
FIRM
PRODUCT Raw pork ham
FUJI CHIKU
+
MUlTERHAM
KIBUN
r 250 MPa 3 hours
(1/4 the usual NaCl content), Unit : Sugino 15 I pink and tender slices Slices packaged in pouches under vacuum Chilled storage 1) "Shiokara": raw squid + salt + autolysis 2) Raw scallops
YAIZU Fisheries
PROCESS
Fish sausages, terrines and "puddings* (95% water, 1% NaCI) "Raw" Sake (rice wine), turbid, with specific flavor
I
Unit : Mitsubishi Chilled storage
Mitsubishi 25 liter unit Unit : Mitsubishi 8 I (400 MPa) Raw material = fish mince
Pressurization of insoluble particles, then reintroduced into the microfiltrated juice Mochi rice cake 400 MPa 10 min containing ECHIGO SElKA 40% water 45 or 70°C (depending on the Yomogi fresh raw materials) aromatic herbs Hypoallergenic precooked rice (moist)
QP corp.
EHIME Cooperative
Ice-nucleating bacteria (used for fruit juice 8 milk concentration) Japanese mandarin juice (project)
400 MPa, Unit : IshikawajimaHarima (7 liters) Unit : NKK
ROLE OF HP
SALES
Faster maturation (reduced from 2 weeks to 3 h); Faster tenderization by internal proteases. Effect on porganisms and parasites 7 Improved water retention & shelf-life
High production cost
Microbial sanitation, tenderization, control of autolysis by endogenous proteases
For sale, without HP advertising. Stopped in 1995.
Microbial sanitation (before frozen or chilled storage) Gelation. Microbial sanitation. Good texture of raw HP gel, but not after ccoking Yeast inactivation; fermentation stopped (without heating)
Confidential. No permission required for sales Shelf-life : 2 weeks at 4°C
Microbial reduction of the cake and1 or of some ingredients (seaweeds, moist aromatic herbs) Fresh flavor & taste Enhances rice porosity 8 salt-extraction of allergenic proteins. Heat-sterilized food Inactivation of
Test market only Selling permit requested but not yet given
Test market only. No sales Since 1994, with HP advertising Small quantities sold in some hotels Since 1994 Shelf-life : 3 months at 4°C Market : 100 million yens1 year Patented. Market test since 1995 Approved
Xanthomonas bacteria by pressure, without loss of iceMitsubishi continuous (multi-orifice) HP machine
Cold pasteurization
For sale to food firms Small market Being installed (1997) with Min. tgriculture funding
J.C. Cheftel, March 1997
Subject Index
Adiabatic heating, foods, 228 Ageing, meat, 295 Albumen, viscosity, 162 Albumen, proteolysis, 163 Alkenes, strained, cycloaddition, 101 Alkenes,cyclization , 6 Allylation, 42,44 Amylase, stability, 41 I Anatexite, thermal conductance, 475 Anthocyanines in strawberry, 250 Antimutagenic activity, 277 Apple dessert, 193 Ascorbic acid, analysis, 250 Ascorbic acid, degradation, 3 10 Ascorbic acid, stability, 137 Aspartame, stability, 139 Azabicycloheptenes, 27 Azulene, 108
Bismethanooctahydroanthracene,102 Bovine serum albumen, denaturation, 175 Brittle fracture, 489
Chlorodifluoroethane, supercritical, 72 Chlorophyll, 141 Chlorophyll, degradation, 234 Choline, binding, 36 Circular dichroism, proteins, 178 Citraconic anhydride, cycloaddition, 25 Clostridium spores, 355 Clostridium, inactivation, 403 Cloud point, 62 Coenzyme Q, 136 Cold denaturation, 167 Compressibility, isentropic, 93 Conformational change, 32 Cope rearrangement, 4 Commercial pressurized foods, table, 506 Cryptococcus, inactivation, 250 Cyclic pressurization, 265 Cyclization, 3 Cycloaddition, tandem, 18 Cycloaddition, strained alkenes, 101 Cycloaddition, tandem, 29 Cyclobutane, synthesis, 22 Cyclodecapentaene, 108 Cycloheptatriene, cycloaddition, 101 Cyclooctatetraene, planar, 102
Carbon dioxide, solubility, 83 Carbon dioxide, supercritical , 53 Carboxypeptidase, denaturation, 168 Carrageenan, 2 14 Casein, degradation, 202,207 Catalysis, transition metal, 40 Cathepsin, activity, 295 Cheese, brining, 200 Cheese, enzymes in, 265 Cheese, microorganisms, 265 Chelotropic reaction, 6
DAST, 38 Density, diacetonealcohol solutions, 96 Diamond, pressure generation, 48 1 Dicyanoethene, 23 Diels-Alder reaction, 4.44 Diels-Alder reaction, coenzyme Qo, 27 1 Diels-Alder reaction, intramolecular, 6 Diels-Alder reaction, vitamin K, 135 Diels-Alder reaction, retro, 1 10 Differential scanning calorimetry, 177 Dihydrofuran, coupling, 41,4.5
Bacillus spores, 355, 395,435 Binding, enzymic, 35 Binding, to DNA, 364
Subject Index
Dihydropyrans, 24 Dimethyl disulphide, 140 Dimethylfulvene, 1 15 Distamycin, binding to DNA, 364 Divinylcyclobutane , 6 DNA, minor groove binding, 364 Droplet size, emulsions, 218 EDTA antioxidant, 285 Egg components, 160 Elastic moduli, protein, 280 Electrolytes, gas solubility in, 8 1 Electrophoresis, protein, 2 16 Emulsions, milk protein, 153 Endothermicity, proteins, 177 Energy, stored, 487 Enzyme, meat ageing, 295 Enzymes, inactivation, 289 Enzymes, milk, 145 Epibatidine, 22 Equipment, food processing, 236,50 I Escherichia coli, 148, 304 Escherichia coli, survival, growth, 387 Exner volume increments, 113 Flow cytometry, 429 Fluorination, 38 Fluorometer, hp stopped flow, 33 Foam collapse, I79 Food chemistry, plenary lecture, 133 Food processing, 227 Freeze-thaw, food systems, 229 Freezing curves, potato, 3 19 Freezing, pressure-assisted, 343 Fruit juices, bacteria in, 304 Fruit, water loss, 33 1 Fullerenes. 12 Furans, Diels-Alder reactions, 25 Gelation, protein, 169 Gels, rennet-set, syneresis of, 220 Germination, spores, 435 Gluconosactone, hydrolysis in milk, 242 Gouda cheese, brining, 200 Granulite, thermal conductance, 476 Guanidiniumpentane,binding, 36 Hardness, soya protein, 282
509
Hawley equation, 61 Hazard assessment, 486 Heat capacity, potato, 323 Heck reaction, 26,42 Hexadienes, 4 Hexanal in strawberry, 139 Hexatriene, 4 High pressure batch stirred tank reactor, 54 Homolysis, 3 Homer-Wadsworth-Emmonsreaction, 42 Hydrogen sulphide, solubility. 85 Hydrophobicity, 427 Hydrophobicity, proteins, 176 Imidazoquinolines, mutagenic activity, 277 Inorganic chemistry, hp, 47 Isopropylidenenorbornene, 1 12 Isothiocyanates, 125, 137 Jossi equation, 75 Juices, antimutagenic activity, 277 Ketene acetals, cycloaddition, 23 Lactoglobulin, 344 Lactoglobulin, denaturation, 175 Lactoglobulin, refolding, 376 Lecithin, hydrolysis of, 124 Lewis acid catalysis, 18,22, 30 Ligand exchange, 41 Light harvesting complex II,417 Linolenic acid, oxidation, 135 Lipase, 53 Lipid oxidation, 284 Lipids, hydrolysis, 122 Lipoxygenase, 289 Listeria, 148 Listeria, chmges in, 442 Listeria, inactivation, 229,403,429 Low-T, high -p method, 166 Lysine, Maillard reaction of, 120 Maillard reaction, 120 Mayonnaise, stability, 257 Meat ageing, 295
510
Meat broth, spore destruction, 361 Metalloproteins, 48 Methane monooxygenase, 47 Methane, solubility, 83 Methional, 140 Methylhydroxyfuranone, 120 Metmyoglobin, spectra of, 453 MHKF equation of state, 8 1 Microorganisms, barotolerance, 370 Microorganisms, inactivation, 170,229 Microorganisms, inactivation kinetics, 403 Microorganisms, milk, 145 Milk acidification, 242 Milk fat globules, 182 Myoglobin, 48,50 Myrcene, Diels-Alder reaction, 135, 27 1 Na+/K+ATPase, 32 Nitroalkenes, 29 Nitroalkenes, cycloaddition, 22 Nitrones, 22, 27 Nitrostyrene, cycloaddition, 18 Oleic acid, esterification, 53 Ovalbumin, conformational change, 325 Oxetans, 24 Oxidation, lipid, 284 Oxyhemerythrin, 49 Packing coefficient, 114 Palladium, catalysis by, 41 Peng-Robinson equation, 85 Pericyclic reactions, 3 Peroxidase, activity, 233 Peroxidase, reactivity, 448 Petroleum, velocity of sound in, 88 Phase diagram, protein-water, 230 Phase properties, 12 Phospholipids, hydrolysis of, 122 Pitzer model, 82 Poly(ethy1ene oxide), 68 Poly(N-vinylisobutyramide), 6 1 Polymer fractionation, 67 Polymers, c60. 15 Polyphenoloxidase, inactivation of, 38 1 Polystyrene, hydroextrusion, 467 Potato, freezing, 317, 321
High Pressure Food Science, Bioscience and Chemistry
Pressure intensifier, 236 Pressure testing, safety in, 486 Proteases in cheese, 268 Protein gel, freezing, 343 Protein, denaturation, 167,207 Protein, spectroscopy of, 45 1 Protein, structure, 32 Protein-polysaccharideinteraction, 2 14 Pyrroles, Diels-Alder reactions of, 25 Pyrrolixidines, 21 Radical intermediates, 9 Rennet gels, 22 1 REPA analysis, DNA binding, 365 Rheology, soya protein, 280 Ribonuclease A mutants, unfolding, 423 Rocks, thermal conductance, 474 Saccharomyces, inactivation, 170 Safety, 486 Salmonella, changes in, 442 Salmonella, inactivation of, 403 Selectivity, 43 Shear modulus, carrageenan, 2 17 Shelf life, vegetables, 232 Solubility, gases, 8 1 Soya protein, rheology, 280 Spores, destruction of, 354 Spores, germination and inactivation, 435 Spores, inactivation strategies, 399 Staphylococcus, 148 Staphylococcus, inactivation, 403 Starch, gelatinization, 457 Starch, hydration, 17 1 Steroid synthesis, 25 Strawberry dessert, 193 Strawberry puree, composition changes, 248 Strecker reaction, 140 Sulphur volatiles, 140 Super-high pressure generation, 48 1 Supercritical fluids, 67 Surface tension, protein solutions, 178 Suzuki reaction, 42 Syneresis, gels, 220 Systems for high p, 501
Subject Index
511
T-p diagrams, 64 Taxol, 22 Telomerization, Pd catalysed, 43 Tetracyanoethene, 23 Tetraphenylhexadiene,cyclization, 7 Thermal conductivity, potato, 323 Thermal conductivity, rocks, 474 Thermodynamic parameters, 55, 65 Thiobarbituric acid, 284 Transition curves, polymer, 6 I Triglycerides, lipolysis, 153
Vegetables, water loss, 33 1 Vinyl sulphones, 27 Viscosity, aqueous solutions, 95 Viscosity, hydrocarbons at hp, 75 Vitamin A alcohol, stability, ,138 Vitamin C (see Ascorbic acid) Vitamin K, Diels-Alder reaction, 135 Vitamins, stability, 137 Volume of activation, 52 Volume of reaction, 5, 5 1 Volumes, 3
Ubiquinone, 136 Ubiquinone, Diels-Alder reactions, 272 Ultrasonic velocity, 88 UREA-PAGE densitograms, 202
Whey protein, degradation, 207 Xylose, Maillard reaction of, 120