gums and stabilisers for the food industry 14 (special publications)-royal society of chemistry...

Edited by Petcf A vvilurm and Qyn 0 PtiltpsGums and Stabilisers for the Food ndustry 14

RSCPublishing

Gums and Stabilisers for the Food Industry 14

Gums and Stabilisers for the Food Industry 14Edited by Peter A. WilliamsCentre for Water Soluble Polymers, North East Wales Institute, Wrexham, UKGlyn o. PhillipsPhillips Hydrocolloics Research Ltd, London, UK

The proceedings of the 14th Gums and Stabilisers for the Food Industry Conerence held on 18-22 June 2007 at NEWI, Wrexham, UK.Special Publication No. 316 ISBN: 978-0-85404-461-0A catalogue record for this book is available hom the British Library The Royal Society of Chemistry 2008 Al rights reservedApart from any fair deaingfor the purpose of research or private study for non- commercial purposes, or criticism or review as permitted under the terms of the K Copyright, Designs and Patents Act, 1988 and the Copyright and Reated Rights Regidations 2003, this pablication may not be reproduced, stored or transmitted, in any form or by any means, without the prior permission n writing ofThe Royal Society of Chemistry or the Copyright owner, or in the case of reprographic reproduction ony in accordance with the terms of the icences 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 reproducton outside the terms stated 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 0WF, UKRegistered Charity Number 207890For further inrmation see our web site at www.rsc.orgPreaceThe preparation of this Preace to the Proceedings of the 14th Gums and Stabilisers for the Food Industry Conerence is this time a poignant undertaking. This Conference was special for me since I was awarded the Food Hydrocolloids Trust Medal ater a session when many of my colleagues and my son Aled gave presentations. These were both personal and recalled work of by-gone days. So it is only tting that I here thank all those who were involved in organising this session and for the Trustees in giving me this very special honour.The high Standard of the presentations is evident once again in this volume. There is no better way to leam about new developments in food hydrocolloids research than to brows in these volumes as they appear every two years. As I travel from country to country and lab to lab it is gratitying to see these volumes on the shelves, and to note the constant references to the papers published. This volume will again take the subject forward.The rst section deals with Novel Hydrocolloid Functionality, which is a target for most hydrocolloid users. With the number of new food hydrocolloids not likely to increase in the Corning years because of the standstill in industrial research in this tield, more must be squeezed out of the presently available materials. It is a fascinating and innovative section. These subject areas demonstrate the diversity of the presentations: polymers versus particles visualisation of hydration and swelling swelling of calcium pectin gel beads processing-structure-property relationships rennet-induced gelation of milk in the presence of pectin perormance of resistant starch type 3 bulk deformation behaviour of gellan gum on cross-linking with mixed cations. hydration study of soy protein in the 'dry State' adhesive of gelatinised starch granule extrusion Processing of xanthan high intensity ultrasonication of pectin gel temperature of pectin and pectin-calcium-gels transitions in egg protein dispersionsThe ingenuity demonstrated in many of the papers is truly admirable as is the global nature of the presentations.The present conerence called for papers on Sensory-Texture Relationships. The contributions were varied and dealt with the eect of texture on lavour release, effect of microstructure on lavour diffusion and release and the sensory and rheology of laxsecd gum-fortified dairy beverages.Hydrocolloid Emulsifiers remain a very interesting and well studied subject. The Leeds group led by Eric Dickinson continues to unravel the complex processes in the brmation and breakdown of emulsions. Gelatine, hydroxypropyl cellulose, mannans and xylans, are now making their mark in the food emulsication area. The potential of sugarbeet pectin continues to interest but despite the research efforts the practical commercial application is still minimal.A major target of this Conference was Hydrocolloids and Health. The papers did not disappoint. We are constantly being urged to increase dietary bre, remove fat, include antioxidants, reduce calories etc. The papers cover each of these areas and show that hydrocolloids can be in the front line in the battle against obesity.The nal three sections deal with: Interactions in mixed hydrocolloid Systems Innovative applications Developments in characterisaton (of hydrocolloids)These papers form the backbone of the subject and all workers in the ield will need to scrutinize these papers for new materials, new phenomena and new techniques. It is gratifying to note that hydrocolloids too can successfully enter the new nano structure era.I am happy, therere, once again to commend the volume to the growing body of researchers in food hydrocolloids. In China recently I found a remarkable growth in interest in this subject and the conference I attended attracted more than 600 participants who traveled from all parts of China. It is ftting, therefore, that the Food Hydrocolloids Trustees have approved that the 10th International Hydrocolloids Conference should be held in Shanghai under the Chairmanship of Protcssor Hongbin Zhang of Jia Tong University where the polysaccharide eld is well and lourishing. Please note the date now - June 2010, following the 9th Conerence in Singapore in 2008.Finally, may I thank my expert Organising Committee for their constant efforts to identify subjects of current interest and point to the specialist lead speakers who can deliver these subects effectively?Glyn o. PhillipsviPreface

#Preface

Chairman, Gums and Stabilisers Conterence Organising CommitteeContentsThe Food Hydrocolloids Trust Medal LectureGiving nature a helping hand3G. o. Phillips, NEWI, Vrexham, UK1. Novel Hydrocoloid FunctionalityMixing hydrocolloids and water: polymers versus particles29J.R. Mitchell, A.L. Ferry, M. Desse, S.E. Hil, J. Hort, L. Marcinni and B. Wolf, University of Nottingham and Queens Medical Centre, Nottingham, UKDetailed microscopic visualisation of hydration and swelling in a rapidly 40 hydrating particle bed containing a cellulose etherS.R. Pygall, p. Timmins and C.D. Melia, University of Nottingham and Bristol Myers Squibb, Moreton, UKSwelling behaviour of calcium pectin gel beads47M. Iijima, M. Takahashi, T. Hatakeyama and H. Hatakeyama, Nagasaki University, Shinshu University, Lignocell Research and Fukui University, JapanProcessing-Structure-Property relationships in biopolymer gel particles53.Burey, B. Bhandari, T. Howes and M. Gidley, The University of Queensland, AustraliaDiffusing wave spectroscopy studies of rennet-induced gelation of milk in the 61 presence of pectinA. Acero Lopez, M. Corredig, M. Alexander, University of Guelph, CanadaPerrmance of resistant starch type 3 in non pre-fried battered food68T.Sanz, A. Salvador, S.M. Fiszman, ITIA, SpanTextural and colour changes during storage and sensory shelf life of muffms 73 containing resistant starchR. Baixauli, A. Savador and S.M. Fiszman, CSIC, Valencia, SpainDramatic changes in bulk debrmation behaviour of gellan gum on cross-linking 79 with mixed cationsJ.J. Harris, A.M. Smith, R.M. Shelton, University of Birmingham and Aston Universty, Birmingham, UKHydration study of soy protein in the 'dry State'87.Keaey, M. Rout, I. Appelqvist, K. Strounina, A. Whttaker, M. Gidley, E.Gilbert and p. Lillord, Australian Nuclear Science and Technology Organisation,

Food Science Australia and The Universitv of Queensland, Australia2S Soy Protein: A Misnomer hence brgotten but unctional nevertheless96s. Kasapis and Sok Li Tay, National niversty of Singapore, SingaporeAdhesive interactions between gelatinised starch granules105s. Hasan, S.M. Fitzsimons, E. 0'Neill and E.R. Morris, Omar al Mukhtar University, Libya, Teagasc, Food Research Centre and University College Cork,IrelandPhysically modiied xanthan gum prepared by extrusion Processing114Nuno M. Sereno, Sandra E. Hill, John R. Mitchell, University of Nottingham, UKEffect of high intensity ultrasonication on the rheological characteristics of 123 selected hydrocolloid SolutionsB. K. Tiawri, K. Mnthukumarappan, c.p. 0Donnell and P.J. Cullen, niversity Colege Dublin, Ireland, South Dakota State niversity, USA and Dublin Institute of Technology, IrelandPectin is an alkali scavenger: potential usage in skincare129Jens Trudso, CP Kelco, DenmarkDemethylation of a model homogalacturonan with a citrus salt-independent pectin 141 methylesterase: effect of pH on block size and number, enzyme mode of action and resulting functionalityR.G. Cameron, G.A. Luzio, K. Goodner, M.A.K. Williams, USDA, ARS, Citrus and Subtropical Products Laboratory and Massey University, New ZealandGelling temperature determination in pectin-based Systems153L. Boettger, S.H. Christensen and H. Stapelfeldt, CP Kelco, DenmarkCharacterization of pectin-calcium-gels: Inluence of pectin methoxylation 164 propertiesI. Fraeye, E. Vandevenne, T. Duvetter, A. van Loey and M. Hendrickx, Katholieke Universiteit, Leuven, BelgiumHigh pressure-induced rheological transitions in egg protein dispersions173J.M. Aguilar, F. Cordobes, c. Bengoechea and A. Guerrero, Universidad de Sevilla, Spain2. Sensory-Texture RelationshipsEffect of texture on lavour release in fruit spread applications181E. Lynenskjold, N. w. G. Young and I. Butler, Danisco, DenmarkImpact of the microstructure on lavour diffusion and release in fruit preparations 195G. Savary, J.-L. Doublier, N. Cayot, INRA-ENESAD, INRA-Nantes, FranceSensory and rheological properties of a laxseed gum-fortified dairy beverage203H. D. Goff, A.E. Muller, F. Capel, C.J. Findlay and w.s. Cui, University of Guelph, Compusense Inc. Agriculture and Agri-Food Canada, Canada and Universitat Hohenheim, Germany3. Hydrocolloid EmulsiersControlling emulsion stability: microstructural and microrheological origins of 211Aocculating SystemsB.s. Murray, Universy of Leeds, KEmulsifcation and stabilisation with protein-polysaccharide complexes 221 Eric Dickinson, University of Leeds, UKDynamic rheological properties of gelatine films at the air/water interace233s. Domenek, R. Abdeli, s. Mezdour, s. Guegj, N. Brambati, c. Ridoux and c.Michon, AgroParisTech-INRA-CNAM and Rousselot, FranceKinetics of adsorption of gelatine at the air/water interface: Effect of concentration 239 and ionic strengths. Domenek, E. Petit, A.s. Delbes, s. Mezdour, s. Guedj, N. Brambati, c. Ridoux and c. Michon, ENSIA, Massy and Rousseot, FranceHydroxypropyl cellulose as a stabilizing agent of emulsions245A. Lepine, s. Mezdour, p. Erazo-Majewicz and c. Michon, ENSIA-INRA-CNAM,Massy, France and ercules, USAMannans and xylans as stabilisers of a model oil-in-water beverage System251K.s. Mikkonen, M. Tenkanen, s. Wilfor, K.B. Hicks and M.p. Yadav University of Helsinki, Einland, Abo Akademi University, Finland and United States Department of Agricidture, USAEmulsiication Properties of Sugar Beet Pectin257Chee Kiong Siew and P.A. Williams, NEWI, Wrexham, UKEffect of thermal treatments and pH modiication on the rheological properties of 264 o/w emulsions stabilised by food proteinsc. Bengoechea, A. Romero, F. Cordobs and A. Guerrero, Universidad de Sevia,SpainStability of emulsions containing sodium caseinate and anionic polysaccharides272L. Jonrdain, M.E. Leser, c. Schmitt, E. Dickinson, University of Leeds, UK and Nestl, SwitzerlandCharacterisation of Gum Ghatti and comparison with Gum Arabic280s. Al-Assaf, V. Amar, G.o. Phillips, Phillips Hvdrocolloids Research Centre,NEWI, UK and The Gums and Colloids Group, India4. Hydrocolloids and healthThe role of hydrocolloids in the rmulation of healthy foods 293I. T. Norton, p.w. Cox andF. Spyropoulos, University of Birmingham, UKElydrocolloids in health306A. Phillips and s. Riley, University ofWales Hospital, Cardiff, UKThe effect of hydrocolloids on satiety, and weight loss: areview313T. Paeschke and W.R. Amutis, Cargill, Inc., USAUtilization of sodium caseinate nanoparticles as molecular nanocontainers for 326 delivery of bioactive lipids to food Systems: Relationship to the retention and controlled release of phospholipids in the simulated digestion conditionsM.G. Semenova, L.E. Belyakova, Y.N. Polikarpov, A.s. Antipova, and M.s. Anokhina, Russian Academv of Sciences, RussiaReal-time CSLM observations on alpha-amylase digestion of starch in isolated 334 form and within cellular integritys. Oyman, J.G.C. Blonk, H.T.W.M. van der Hijden, H.p.p. Peters and S.E. Hill, University o/Nottingham, K and Unilever R&D, The NetherlandsBiopolymer structures for novel gastro-intestinal functionality:in vitro 341characterisation and behaviour in vivo using MRI.Rayment, s. Pregent, C.L. Eoad, E. Ciampi and M.F. Butler, Unilever R&D Colworth and University of NottinghamCalcium alginate as a gastro-activated dietary bre349o. Gserod, H. Haraldsen and G. Lynch, EMC Biopolymer Nonvay and BelgiumPectin - health beneits as a dietary bre and beyond358o. Hassewander, Danisco Sweetners Ltd, UKExtraction, characterisation and anti-inlammatory bioactivity of polysaccharides 367 from boat-fruited sterculia seedsY. Wu, s. w. Cui, J. Tang, Q. Wang and X. Gu, Shanghai Jiao Tong University,China and Agriculture and Agri-Food Canada, Guelph, CanadaStructuring of low calorie food with fruit bres379Jurgen Fischer, Herbaood Ingredients, GermanyRheological behaviour of carboxymethyl cellulose dairy desserts with different fat 386 contents. Bayarri and E. Costel, CSIC, Valencia, SpainThe role of hydrocolloids in the management of dysphagia.392G. Sworn, E. Kerclavid, J. Fayos, Danisco SAS, EranceAntioxidant activity of soy protein hydrolysate and peptides402.Kasase, A. Ganeshalingam and N. Howel, University of Surrey, UK5. Interactions in mixed hydrocolloid SystemsModelling of the rheological behaviour of the temary Systems of tragacanth, guar 409gums and methylcellulose as a unction of concentration and temperatureC. A. Silva, F. Chenlo, R. Moreira and G. Pereira, Universidade de Santiago,SpainStructural properties and phase model interpretation of the tertiary System 419 comprising gelatin, agarose and a lipid phasep. Shrinivas, T. Tongdang and s. Kasapis, National University of Singapore, Singapore and Prince Songkla University, ThalandComplex coacervation between P-lactoglobulin and K-caageenan427J. Dovle, J.s. Mounsey and B.T. 0Kennedv, Moorepark Food Research Centre,IrelandViscoelasticity of starch-milk Systems with inulin added. Inluence of inulin Chain 435 length and concentrationL. Gonzalez-Tomas, J. Coll-Marques and E. Costell, CSIC, Vaencia, SpainInteraction of different gelling carrageenans with milk proteins440J. de Vries, D. Arltoft and F. Madsen, Danisco A/S and University of Copenhagen, DenmarkAFM and DSC Studies on Gelation of Methylcellulose Mixed with Sodium 446 Cellulose SulfateT. Hatakeyama, M. Dolima, T. Onishi and H. Hatakeyama, Lignocel Research and Fukui University, JapanThe effect of iller orientation on the mechanical properties of gelatin-MCC 454 compositesLee Wah Koh, s. Kasapis and D. Teck Lip Tam, National University of Singapore, SingaporeCharacterisation of rheological properties of mixtures of whey protein isolate and 461 inulinJ.T. Tohin, S.M. Fitzsimons, E.R. Morris and M.A. Fenelon,Teagasc, Food Research Center, University College, Cork, IrelandEffect of shearing on the phase diagram and rheological behaviour of an aqueous 469 whey protein isolate-K-carrageenan mixturess. Gaaoul, s. Turgeon, M. Corredig, Universit Laval, Sante-Foy and University of Gueph, CanadaPectin-protein complexes-new roles for pectin extracts477V.J. Morris, A.p. Gunning, A.R. Kirby and A.J. MacDouga, Institute of Food Research6. Innovative ApplicationsMicroalgae biomass as a novel unctional ingredient in mixed gel Systems487A.p. Batista L. Gouveia, M.c. Nunes, J.M. Franco, A. Raymundo, ISEIT de Almada, INETI-DER-Unidade Biomassa, Portugal and Universidad de Huelva,SpainCellulose gum as protective colloid in the stabilization of acidified protein drinks 495M. van cler Wielen, w. van de Hening, Y. Brouwer, CP Kelco, The NetherlandsProtein Stabilization and Particle Suspension in Acidite Protein Drinks Using a 503 Dual-function Hydrocolloid System.C.R. Yuan, M. Kazmierski-Steele and p. Jackson, CP Kelco, USANanostructures and nanoods510V.J. Morris, Institute of Food Research, Nonvich, UKHigh-pressure-induced yuzu maimalade518H. Kuwada, Y. Jibu, K. Yasukawa, s. Makio, A. Teramoto and M. Fuchigami, Okayama Pre/ectnral University and Kanto Gaknin University, Japan7. Developments in CharacterisationMolecular structures of gellan gum imaged with atomic force microscopy (AFM) 527 in relation to the rheological behavior in an aqueous Systems. Gellan gum with various acyl contents in the presence or absence of potassium T. Funami, s. Noda, s. Ishihara, M. Nakauma, R. Takahashi, s. Al-assaf, K. Nishinari, and G.o. Phillips, San-Ei Gen F.F.I. Inc., Gunma University, Japan,Phillips Hydrocolloid Research Centre, UKRapid determination of alginate monomer composition using Raman spectroscopy 543 and chemometricsT. Salomonsen, H.M. Jensen, D. Stenbaek and S.B. Engelsen, Danisco and University of Copenhagen, DenmarkPhysicochemical properties of starch isolated from sago palm (metroxylon sagu) 552 at different palm heightsB.A. Fasihuddin and P.A. Williams, Universiti Malaysia Sarawak and North East Wales Institute, UKA cutting edge technology in the rheological studies of thermal Processing of 558 polymers: measuring response to high temperature treatment using a high pressure cellM.s. Kok, Abant Izzet Baysa University, TurkeyAFM, microstructure and unction564V... Morris, Institute of Food Research, UKPhysicochemical characterisation of psyllium fibre572M.s. Temudo, M.c. Nunes, A.p. Batista, F. Carvalheiro, M.p. Esteves and A. Raymundo ISEITde Almada and INETI, PortugalxiiContents

Contentsxi

Subject Index578AcknovvledgementsThe Food Hydrocolloids Trust are indebted to the conerence organising committee;Dr p. Boulenguer, Cargill France SASDr s. Davies, ICI p!c, UKProessor E. Dickinson, University of LeedsProessor E. A. Foegeding, North Carolina State University, USADr T. J. Foster, University of Nottingham, UKDr I. Hodgson (ViceChairman), lan Hodgson AssociatesProessor D. Hovvling, David Howling AssociatesMr H. Hughes (Secretarat), North East VVales InstituteDr A. Imeson, FMC CorporationMr D.R.J. Lloyd, CargillDr M. Marrs,Dr R.G. Morley, Delphi Consultant Services Inc., USA Proesoor J.R. Mitchell, University of Nottingham Proessor E. R. Morris, University College Cork, Ireland Proessor V.J. Morris, Institute of Food Research, Norvvich Dr J.C.F. Murray (Treasurer)Proessor K. Nishinari, North East Wales InstituteProessor G.o.Phillips (Chairman), Phillips Hydrocolloids ResearchDr K. Philp, CyberColloids Ltd, IrelandDr c. Rolin, CP Kelco, DenmarkDr c. Schorsch, Danone, FranceDr c. Speirs, CCFRADr G. Sworn, Danisco, FranceDr M. Taylor Cadbury SchvveppesDr A. Tziboula-Clarke, ISP (International Specialty Products)Proessor P.A. VVilliams (Sclentiic Secretary), North East Wales Institute and also acknovvledge inancial support from Major sponsors Coca Cola LtdPhillips Hydrocolloids Research Ltd San Ei Gen F.F.F. Incother sponsorsCadbury SchweppesCargillCP KelcoDaniscoMarinalgMasterods

Unilever

The Food Hydrocolloids Trust Medal Lecture

GIVING NATURE A HELPING HANDGlyn o. PhillipsGlyn o. Phillips Hydrocolloids Research Centre The North East Wales Institute, Plas Coch, Mold Road, Wrexham, LL1 1 2AW, Wales1. INTRODUCTIONThere is a considerable appeal for the general public in the concept of natural foods. The food producer, therefore, in the cuent health conscious climate makes every effort to equate natural with fitness and good living to promote a green image without those nasty Chemicals. Into this category come the indigestible plant polysaccharides, which were included by Trowell (1) in the definition of dietary fibre. Previously the term had been used to describe the remnants of plant components that are resistant to hydrolysis by human alimentary enzymes (2-4). The impending Codex definition of dietary ibre (5) refers to edible carbohydrate polymers naturally occurring in the food as consumed, and carbohydrate polymers, which have been obtained from food raw material by physical, emymatic or Chemical means. The expectation is that these natural polymers would need to lead to a positive physiological effect, such as: decreased intestinal transit time and increase stools, bulk fermentable by colonic microAora, reduced blood total and/or LDL cholesterol levels or reduced post-prandial blood glucose and /or insulin levels. By association, thereore, food producers can imply that these natural polymers have healthy effects when eaten regularly.I do not wish to cast any doubt about the beneicial effects of non-starch polysaccharides and indeed there is ample evidence of their effectiveness in promoting a healthy life style (6). These have undoubted avantages but there are problems to integrate them into industrial production, which demands constant quality and performance Natural polymers are never uniform or simple. Their functionality depends on more than one structural eature. Extraction processes alter the macro- and micro-structures and perormance. Their most common eature is their variability which poses a big problem for both the producer and User of natural polysaccharides.This paper draws attention to this variability in three natural polysaccharide Systems: gum arabic, sugar beet pectin and gum Ghatti. The problem we have tried to solve is how can we remove this natural variability and secondly how can we enhancetheir perormance using methods which would not involve the introduction of new Chemical groups and so be acceptable to the food industry. In other words can we circumvent Nature and find ways of producing uniform/constant Products and if possible with better speciic unctionalities ?2. GUM ARABIC ( GUM ACACIA)The cuent WHO/JECFA Speciication (1998), which is intemationally accepted and has also been approved by Codex Alimentarius (INS No. 414) is: Gum arabic is a dried exudate obtainedrom the stems and branches of Acacia Senegal (L.) Wildenow or Acacia seyal (fam. Leguminosae) (7). For comparison it is noteworthy that the European Speciication (E 414) is slightly broader (2003): Acacia gum is a dried exudation obtained from the stems and branches onatural strains of Acacia Senegal (L) Willdenow or closely related species o/Acaca (amily Leguminosae)(8).This paper deals with the gum arabic ( A. Senegal (L.) Willd. var. Senegal ). This exudate gum is a complex polysaccharide consisting of D-galactopyranose (~44 %), L- arabino- pyranose and uranose (~25 %), L- rhamnopyranose (14 %), D-glucuropyranosyl uronic acid (15.5 %) and 4-O-methyl-D-glucuropyranosyl uronic acid (1.5 %). It also contains a small amount (~2 %) of protein as an integral part of the structure. The carbohydrate structure consists of a core of P-(l,3)-linked galactose units with extensive branching at the C6 position. The branches consist of D-galactose and L-arabinose and terminate with L-rhamnose and D-glucuronic acid (9). Randall et al. (10,11) ractioned A. senegal by hydrophobic affinity chromatography and showed that it consists of three components namely arabinogalactan (raction 1, AG); arabinogalactan protein (action 2, AGP) and a glycoprotein (raction 3, GP). Each action contains a range of different molecular weight components which are responsible for the polydispersity of the gum The AG action contains 88 % of the total gum with small amounts of protein 0.35 % which represents 20 % of the total protein content, while the AGP raction comprises 10 % of the total gum with 12 % protein which is 49.5 % of the total protein content. Finally the GP raction contains 1.24 % of the total gum with 50 % protein, which represents 27 % of the total protein in the whole gum (10.11).2.1 Natural variabilityThere is now clear evidence of the great variability within commercial gum arabic supplied to the market (12). 67 samples of A. Senegal var. Senegal exudate gum, supplied by primary supplier, producer and User companies were analysed using gel permeation chromatography (GPC) coupled to a multi-angle laser light scattering detector, a reractive index detector and a uv detector operated at 214 nm. A set of 5 samples were ully authenticated A. Senegal var. Senegal and used to provide norms against which the other gum samples could be assessed. A Standard unprocessed A. Senegal var. Senegal (hashab) from a 15 years old mature tree was used for comparison. A eature of the results is the extensive variation betvveen individual samples, all of which were presented to the market as gum arabic. Of the samples, 15 were outside the selected norms. Table 1 shows the variation found in 13 samples provided by one company, with each one being marketed as an identical gum arabic product. There is more than a two-fold variation in weight average molecular weight (Mw) and wide differences in molecular parameters and amount present of the ractions, including the arabinogalactan protein (AGP) component. This is the component vvhich is responsible for the emulsion capability of gum arabic, so it was inevitable that there was also a wide variation in the capability of the different samples to unction in beverage emulsions, for example(12).Table 1. Molecular weight parameters of guin arabic samples from One company .File nameMw( processed as One peakMW/Mn%massRgnmMwt processed as two peaksmw/mn%massRgnm

C3-15.83 xio5 0.131.8510126.22.01 X 10b 0.051.4215.833.6

3.19 X 105 0.051.1685.616.0

C3-27.32 xl05 0.292.1910230.43.11 X 10 0.051.7413.3836.72

3.73 X 105 0.031.258.40.0

C3-36.17 X 105 0.101.7110020.92.09 X 10 0.131.2912.726.8

3.98 X 105 0.051.2387.315.0

C3-43.37 xl03 0.091.8210521.72.32 X 106 0.131.364.233.11

2.52 X 105 0.051.411015.1

C3-56.68 X 10^ 0.142.0010828.52.44 X 10b 0.071.4815.434.9

3.71 X 105 0.051.2692.620.4

C3-57.56 X 103 0.232.3310733.33.56 X 10b 0.161 9112.740.22

3.70 X 105 0.041.2793.51.0

C3-66.19 Xl05 0.241.6210929.71.87 X 10b 0.091.3518.634.3

3.78 X 105 0.111.1496.624.6

C3-76.24 X 105 0.381.5910034.51.87 X 10b 0.121.3616.637.8

3.88 X 105 0.221.1185.531.0

C3-85.58 X 103 0.251.5810632.51.79 X 10b 0.081.3415.136.8

3 55 X 105 0.161.1191.727.9

C3-96.36 xl05 0.251.5310131.471.69 xl0b 0.071.3016.934.8

3.90 X 105 0.171.118527.4

C3-101.15 X 10b 0.093.9611044.75.90 X 10b 0.562.9616.950.12

3.95 X 105 0.161.541057.9

C3-111.20 X 10b 0.054.3910244.65.94 X 10b 0.252.9216.650.02

3.84 X 105 0.171.6197.65.3

C3-126.06 X 10'0.482.1110135.02.68 X 10b 0.131.5512.242.5

3.47 X 105 0.161.3388.325.8

C3-137.19 X 105 0.603.0010339.34.63 X 10b 0.502.5110.047.3

3.47 X 105 0.181.5793.224.5

2.2 Removing natural variability and conversion of a poor into a good emulsifying gumTable 2 shows the molecular parameters of two samples of gum arabic, one (labeled CT-3892) with Mw 8.34 and the other (labeled FR-2635) 4.24 X 105. The former gum was shown to be an excellent emulsifier whereas the latter was an extremely poor emulsiier, in accordance with the different amounts of AGP present in each - 14.4 and 9.6% respectively (13).4Gums and Stabilisers for the Food Industry 14

The Food Hydrocolloids Trust Medal Lecture#

The Food Hydrocolloids Trust Medal Lecture5

We have demonstrated (14,15) that using a newly developed process it is possible to treat the poor emulsiier and produce a new product with the molecular parameters comparable with those of the good emulsiier. This is the product labeled MI in Tabe 2 which has Mw9.46xl05 and AGP content 17%.Table 2 Molecular vveight parameters of poor emulsifier (FR-2635) sample and after maturing to obtain increasing proportions of high molecular weight polysaccharide-protein complex.Sample nameProcessingMolecular weight (Mw g/mol)Recovery(%)Polydispersity(Mw/Mn)Rg(nm)

CT-3892Whole gum8.340.26xl051062.2325

First peak (AGP)3.060.10xl06(14.4)1.2934

Second peak (AG+GP)4.730.19xl05(85.6)1.4326

FR-2635Whole gum4.240.20xl05110.61.8533

(Control)First peak (AGP)1.550.07xl06(9.6)1.1241


PR-2635/M1Whole gum9.460.50xl05117.02.9055

First peak (AGP)3.570.17xl06(17.2)1.3765


FR-2635/M2Whole gum1.730.09xl06102.53.8573

First peak (AGP)6.060.31xl06(22.1)1.6181


FR-2635/M3Whole gum1.950.09xl0677.43.7862

First peak (AGP)6.700.28xl06(22.5)1.4068

Second peak (AG+GP)5.72+0.30xl05(77.5)1.3932

FR-2635/M4Whole gum1.450.06xl0671.83.7259

First peak (AGP)4.690.18xl06(24.0)1.4365


The process can be continued (M1-M4) such that a product with a Mw of ca. 2 xio6 and AGP content of more than 20%. Using Standard emulsion formulations which have been described (16).Emulsiication effectiveness was evaluated based on the initial particle size of the emulsions which were then subjected to an acceleration testing (7 days storage at 60C). Particle size diameter of emulsion ater the acceleration test was measured using a particle size distribution analyzer. Emulsiication stability was evaluated by the change in particle size of emulsion after acceleration test. The change in particle size after the acceleration test (7 days storage at 60 C) was taken as a parameter to designate the category of the gum sample. Therefore, the gum samples which showed a change of 0.1 pm or less were given category 1 (good emulsier). A change >0.1 pm - 1.0 (nn are classed category 2. The less stable emulsions which showed a change >1.0 pm were allocated category 3 (poor emulsiier).The process is shown to have converted a poor Class 3 emulsiier into an excellent Class 1 emulsiier (Table 3).

2.3 The maturation processThe process is carried out in a dry stainless Steel Container or on a suitable surface, open to air or in a non-oxidizing environment (an atmosphere of nitrogen). The treatment involves maturation under strictly controlled conditions of temperature and humidity of the dry gum (13,14). The method is essentially one that is used in Standard food Processing and promotes the urther maturation of gum arabic in a way, which emulates and extends that which occurs naturally.As the tree grows in the Sudan the molecular weight of the exuded gum arabic increases from 250,000 (at 5 years) to a maximum of 450,000 (after 15 years) and the amount of protein and of the high molecular weight protein raction also increases with the age of the tree (17). This build-up in the tree effectively unites small molecular weight fractions, which contain a small amount of protein into the larger units, of which the ultimate is the arabinoglactan protein with molecular weight of some 2.5 X 106. The carbohydrate and amino acid composition of these smaller sub-units are identical (11,18). Thus, the biological process involves initially the rmation of the sub-units and then these are joined into larger units as the tree grows.Moreover this change of composition continues after the gum is initially harvested. Proessors J. c. Fenyo and J. Vassal studied Veshly collected gums (so-called green- gum) from A. Senegal and observed the change in the properties on maturing (19). After storage over a year the speciic rotation, nitrogen (hence protein) and intrinsic viscosity changed significantly, indicative of a continuing change in the molecular aggregation process.The process which has now been developed to produce a new series of Supergum arabic accelerates and enhances this same natural aggregation process, under strictly controlled conditions, which were worked out first at laboratory, then pilot scale and inally at plant level. The smaller arabinogalactan units containing some protein, join to form larger molecular weight arabinogalactan protein (AGP) aggregates. By monitoring the molecular architecture of the gum at all stages, speciic new Products have been produced and characterised. In all aspects this specially matured gum is chemically and molecularly identical to the base gum, but because of the diTerence in distribution of smaller units into larger aggregates, the physical ad unctional perormance is greatly enhanced. The details can be found in a series of publications (16, 20-23).SuperGum arabic produced by accelerated maturation process is chemically and immunologically identical to gum arabic as collected from the tree. Contains exactly the same sugar moieties and in the same proportions as control gum, which has not been subjected to the accelerated maturation process Contains exactly the same amino acids and in the same proportions as control gum, which has not been subjected to the accelerated maturation process Nature of structural bonding is identical with control gum, which has not been subjected to the accelerated maturation process Immunologically identical to gum, which has not been subjected to the accelerated maturation process Only the degree of the organisation of the components has been changed by the process, with the not so useul low molecular weight protein aggregating to form the essential high molecular weight protein No new Chemical groups introduced as a result of the accelerated maturation processTable 4. Molecular weight of control and matured gum arabic by GPC-MALLS analysisSample nameProcessingMolecular weight (Mw, g/mol)% Massuv peak area(%)Rg(nm)

Control gum arabicone peak6.22 X 10528.5

FR-2876two peaks (1, AGP)2.54 X 10610.628.841.1

two peaks (2, AG+GP)3.96 X 10589.471.2-

Matured gum arabic AOne peak1.23 X 10659.0

ER-2877two peaks (1, AGP)6 58 X 10613.242.968.7

two peaks (2, AG+GP)4.13 X 10586.857.1-

Matured gum arabic Bone peak1.66 X 10664.2

ER-2788two peaks (1, AGP)8.56 X 10615.352.6709

two peaks (2, AG+GP)4.16 X 10584.747.4-

Matured gum arabic cone peak2.54 X 10685.1

FR-2789two peaks (1, AGP)1.16 X 10718.653.189.8

two peaks (2, AG+GP)4 50 X 10581.446.9-

Matured gum arabic Done peak1.08 X 1063.6

Supergum EM1two peaks (1, AGP)5.98 X 10612.335.374.5

two peaks (2, AG+GP)3.90 X 10587.764.7-

Matured gum arabic EOne peak1.77 X 10668.4

Supergum EM2two peaks (1, AGP)7.84 X 10618.254.075.5

two peaks (2, AG+GP)4.16 X 10581.846.0-

Table 4 shows the change in molecular parameters for a series of gums ER-2876 to FR- 2879 and shows also the two Products which have been selected for commercial release: Supergum EM1(MW 1.08 X 106) and EM2(1.77 X 106). EM1 is produced to provide a good emulsifying gum of conventional qualities and EM2 a gum with enhanced emulsifying qualities which can operate in beverage emulsions at one third to one quarter of the concentration normally used for commercial gum arabic (20%), depending on the nature of the emulsion. The molecular aggregation processes can be illustrated at

6Gums and Stabilisers for the Food Industry 14

2.4 New Tailor - made commercial Products: Chemical nature of the process8Gums and Stabilisers for the Food Industry 14


which shows that aggregation of the high molecular weight protein is the effectivechange.Figure 1. GPC overlay chromatogram of gum arabic showing the light scattering (LS)reractive index (RI) and uv at 214nm. The data was normalized as ected mass of 0.4 mgto display on the same Y-axis scale.Elution Volume (mJ>Elution Volums (ml)Eluton VoiurElution Voiuma (ml)Elution Voiume (m I)Elution Volume (ml)2.5 Enhanced emulsification perormance in real SystemsIn orming stable emulsions it is the hydrophobic moiety of the arabinogalactanprotein (AGP) of the gum arabic which bridges the hydrophilic barrier to coat the oildroplet, with the driving force being directed by the entropic energy of the hydrophiliccarbohydrate groups residing in the water layer (Figure 3). In the matured SupergumProducts the protein is aggregated and can offer 7 or 8 greater surace dimensions withconsequent stronger binding forces. Electrostatic and steric stabilizing iactors areimproved by the treatment. This leads to the ability to stabilise smaller droplet sizes andto provide considerable greater long-term stability for the emulsions. This is illustrated inFigure 4 for the samples described in Table 4. Figure 5 shows the comparison withStandard gum arabic in a complex concentrated beverage emulsion and shows that EM2is more eriective at 4 times less concentration

molecular level using GPC-MALLS (Figure 1) and atomic force microscopy (Figure 2)The Food Hydrocolloids Trust Medal Lecture9



Figure 2 . Atomic force image of conventional gum arabc and supergum EM2Diqital In:Scan rate IMumber 600 im.However it is possible to observe that the PSDs for particles formed at IM and 0.1M CaC2 were very similar, vvhereas a signiicant change in particle size occurs between 0.1M and 0.015M CaC2. This provides a control window in which particle sizes can be modifed for various applications, and the opportunity to investigate the effects of calcium concentrations within this range.3.2.2 ota-carrageenan. Iota-carrageenan particles also showed an increase in size with a decrease in calcium chloride concentration and displayed a very round/spherical morphology (Figure 4).At a concentration of 0.00IM CaCb, similarly to alginate, no iota-carrageenan particles were visble and a PSD could not be obtained. The sample appeared solution- lke, indicating that dissolution was the dominant mechanism, with little or no gelation occuring within the tmeame of the experiment. The extent of the increase in iota- carrageenan particle size with calcium concentration was not as great as that of the aginate particles.Figure 4 Change n iota-carrageenan partice size with calcium chorde concentration, shown through PSD and light microscopy (scae bars are 10 mcrons). The PSD curves are for sampes hydrated at M CaC2 (back line), 0.M CaC2 (grey ne) and 0.0095M CaC2 (dashed ne). PSD pots represent the /requency ofpartices w thin each size range measured by the Mavern E.O.lMCaCh0.0095 M CaCl2IM CaC2

3.2.3 Pectn. Pectin get particles also showed an increase in particle size with a decrease in calcium concentration and displayed a mostly round/spherical morphology (Figure 5).

Figure 5 Change in pectn partice sze with cacium choride concentration, shown through PSD and ight microscopy (scae hars are 10 mcrons). The PSD curves are for samples hydrated at M CaC2 (back ine), O.M CaC2 (grey line) and 0.015M CaC2 (dashed ine). PSD pots represent the /requency of partices vAthn each size range measured hy the Mavern E.

Smilarly to iota-carrageenan, no pectin particles were visible at 0.001 M CaC2 and the sample appeared solution-like, indicating a dominant dissolution mechanism. There was very little difference in PSD position between IM and 0.1M CaC2, samples, however at 0.015M CaCl2there was a marked increase in particle size (Figre 5). This particle sizes achieved were larger than for iota-carrageenan particles, but not as great as that of alginate particles.3.3 Modelling and Comparison of gel particle sizeAs a comparison, the swelling ratio (Q) for each biopolymer System was determined using the following ormulaQ = D50,x/D505usp[1]Where D50x is median particle diameter at X molL"1 CaCl2 and D5o,sp is median partice diameter of spray-dried particles of the corresponding biopolymer.Q was determined for alginate, iota-carrageenan and pectin particles hydrated at concentrations including IM, 0.1M, 0.05M, 0.04M, 0.03M, 0.02M, 0.015M and 0.0095M CaCl2. Calculations revealed that each System would achieve stoichiometric balance at the following: alginate, 0.029M, iota-carrageenan, 0.022M and pectin, 0.018M CaCl2.All three biopolymers showed similar very small swelling ratios at high calcium concentratons i.e. at IM and 0.1M, indicating a dominant gelation mechanism (Figure 6a). However, despite similar hydration conditions for all three Systems, below a concentration of 0.1M CaCl2, the swelling ratios began to differentiate, which could be explained by the differences in stoichiometric ratio (SR) i.e. the ratio of mols of Ca2+ ions to mols of biopolymer expressed on a monosaccharide resdue basis (Figure 6b).

A rise in Q was initiated at 4)-D-galacturonosyl sections containing branch-points with mostly neutral side chains, and also rhamnogalacturonans. Some of the carboxyl acid groups in galacturonans can be methyl esteried and the amount that is substituted gives the pectin which is referred to as the degree of esterication. Pectins with a degree of esterification of 50% or more are classied as high methoxyl pectin (HMP), whereas pectins with a lower number of methylesters are called low methoxyl pectins (LMP).7 Pectins are often used in the food industry because of its various unique properties like viscosity improvement, gelling properties and water holding capacity.8High methoxyl pectin is used in different dairy Products because its addition prevents the aggregation of proteins and thus phase and whey separation at low pH-values.9 At pH values close to the isoelectric point, caseins micelles increase the number of positively charged patches and when the negatively charged pectin molecules are added at these pHs, they interact electrostatically with the casein particles hindering aggregation.10 However at high pH-values (6.5-6.7), the stabilizing behavior of pectin no longer occurs, and it is believed that the pectin molecules do not interact with casein micelles, as at these pH values both caseins and pectins are negatively charged, and at high enough concentrations pectins will cause phase separation.11 Nevertheless, the effect of the presence of pectin in milk and its interactions with the casein micelles at high pH-values are not fully understood.The objective of this research was to study the effect of the presence of different amounts of HMP on the stability and aggregation of casein micelles during renneting at pH 6.7. To investigate the kinetics of aggregation of this protein-polysaccharide System, a non-invasive technique such as diffusing wave specoscopy (DWS) was used. This technique provides inormation about particle size and interactions occurring in food Systems in a concentrated State; this makes it a valuable tool for the study of aggregating Systems at industrially-relevant concentrations.2 MATERIALS AND METHODS2.1 Skimmed MilkFresh milk was collected from the Elora dairy research station (Elora, ON, Canada). In order to prevent bacterial growth sodium azide was added at a concentration of 0.02% (w/v). Milk was centrifuged at 6000 rpm for 20 min at 4 c using a Beckman J2-21 centriugc and JA-10 rotor (Beckman Coulter, Mississauga, ON, Canada). Milk was then Tiltered three times through Whatman glass fiber lters (Fisher ScientiTic, Whitby, ON, Canada) and maintained in the rerigerator at 5 c until it was used.2.2 Sample PreparationSkim milk was heated to 60c in order to acilitate pectin dilution and control heat- induced aggregation of whey proteins.15 HMP (72.8%DE Random, CpKelco, San Diego, CA) was added to skim milk at concentrations of 0.04% and 0.18% and stirred for at least 30 min. The mixtures were cooled to 30c and the pH was adjusted to 6.7. The appropriate amount of rennet (Chymostar Double Strengh ,Rhodia, Cranbury, USA) was added to the milk sample and the sample was immediately placed in the DWS.62Gums and Stabilisers for the Food Industry 14

#

63

Measurements were perormed using transmission DWS. The full details of the equipment set up can be found elsewhere.14 A volume of 1.5 ml of sample was placed into a 5 mm glass cuvette (Hellma Canada Limited, Concord, Canada) and kept at a temperature of 30c by a circulating thermostatted water bath. Each run was carried out for one hour for samples with no rennet, and for two hours for renneting experiments, in all cases measurements were perrmed at intervals of 3min. Data were analyzed usin Sigma Plot 10.0 (SPSS Inc., Chicago, USA)Novel Hydrocolloid Functionality2.3 Theory of DWSDWS measures the properties of scattered light in suspensions where it is known that all photons ha ve been multiple scattered. In transmission DWS, the laser light enters the sample and the scattered light is collected after it has traversed the whole length (L) of the scattering cell. Correlations in the scattered light are characterized by means of the correlation unction 12:Eq.lg73ro1.0 - 0.5120

Figure 2 Renneting behaviour of control milk (A) and of mik containing 0.04% pectin, measured bv DWS 1/1* (), radius (T). [footnoteRef:4]droplets, there seems to be a reversal in 1/1* behavior, fst increasing from that of the conol System to then decrease with irther addition of pectin. At this moment, we are not able to fully explain this phenomenon and urther studies are being perrmed. Under certain conditions (high polysaccharide concentration, pH above the isoelectric point, etc.) any protein-polysaccharide System is spontaneously separated in two liquid phases, one of protein and the other one of polysaccharide, this is known as thermodynamic incompatibility.17,18 This can be observed when a higher pectin concentration (0.18%) is employed; high amounts of pectin induce phase separation of the System, as is shown by the signicant increase in 1/1* value at 0.18% pectin after 30 min (Fig. 1). The System separates in two phases, a translucent phase on top and a white precipitate. [4: DISCUSSIONChanges in 1/1* can be correlated with different arrangements of the particles in the System. Addition of 0.04% pectin increases the 1/1* value indicating changes in the Structure of the System although this is not visible to the naked eye, and the System is stable, at least within the length of the experiment. Previous studies in emulsion Systems16 show that the addition of pectin produces a monotonic decrease in 1/1* values as a function of pectin concentration. In milk, where casein micelles are present instead of emulsion]

The enzymatic action of rennet on casein micelles can be followed by the changes in 1/1* and particle size parameters. At the beginning of the reaction, the particle size remains constant until cleavage of nearly all K-casein19 from the casein micelles (Fig 2a). When this point is reached, it leads to an increase in particle size since the steric and electrostatic stabilization have been reduced, inducing aggregation of casein micelles. The addition of 0.04% pectin has an effect on milk renneting (Fig 2b). The 1/1* value also shows the characteristic initial plateau followed by a sudden increase albeit to a much lesser extent than in the conol case. At the point of gelation, the second plateau can also be observed. This would tend to indicate that the System reaches its nal spatial arrangement much sooner than in control skim milk samples. The fact that the 1/1* values already reach the second constant after 1 hour of renneting action indicates that even a small amount of pectin, not enough to induce visible depletion locculation or to even affect stability (Fig. 1), affects the kinetics of renneting of the System. Skim milk plus small amounts of pectin is more prone to gel. This can be further conrmed by the much more sudden arrest in movement of the casein micelles (here shown as an increase in apparent size). It could be speculated that the presence of negatively charged pectin causes a slight redistribution of the casein micelles, too small to be detected by rheology or the naked eye, but large enough at the molecular level to affect the angular scattering of light (the 1/1* parameter). This rearrangement then enhances the ability to gel, either by trapping the caseins together or othenvise. However, visual inspection of the final gel does not show differences from that of the control. Cuently, more detailed rheological measurements as well as microscopy imaging are carried out to urther clariy this effect.5 CONCLUSIONSThe interaction between casein micelles and pectin at pH 6.7, depend on the amount of pectin that is added to the System. The addition of low pectin concentration (0.04%) has an effect on the structural arrangement of the System. This is contrmed by the increase in 1/1* value; however, no visible phase separation seems to take place, in contrast to the addition of high concentrations of pectin. Low pectin concentrations also accelerate the gelation time and have a signicant increase in the Tinal particle mobility. This provides new insights on the interaction of extremely small amounts of this anionic polysaccharide with casein micelles.References1 E. Dickinson and McClements D.J, Advances in food coloids, ed. Glasgow, UK: Blackie, 1995, ch. 32 C.G. de Kruif and R. Tuiner, Food Hydrocolloids, 2001,15, 5553 P.F. Fox and P.L.H. McSweeney, Dary chemistry and bochemistry, ed., Thomson Science, 1998, p. 1464 H.s. Rollema, AdvancedDairy Chemistry, ed., Fox, 1992, p. 1115 H. Hui, Dairy Science and technoogy handbook, eds.,VCH, 1993, ch. 1, p. 96 C.G. de Kruif, C.G. and c. Holt, Advanced Dairy Chemistry, Volume 1 - Proteins, 3rd Ed., Part A, ed., P.F. Fox & P.L.H. McSweeney. Kluwer Academic/Plenum Publishers, New York, 2003, ch. 5, p.2337 s. Cui, Food Carbohydrates, eds., Taylor & Francis, 2005, ch 6, p. 2758 M. Femandez, Handbook of dietaryber, ed., Cho., 2001, p. 5839 w. Willats, J. Knox and J. Mikkelsen, Food Sci. Technology., 2006, 17, 9710 A. Syrbe, W.J. Bauer and H. Klostermeyer, Int. Dairy Journa\, 1998, 8, 17911 E. Dickinson, M. Semenova, A. Antipova and E. Pelan, Foocl Hydrocoll, 1998,12, 42512 D.A Weitz and D.J Pine, Dynamic light scattering: The method and some applications, Oxford Unversity Press, Oxord, 1993, p. 65213 M. Alexander and D. Dalgleish, Dynamic light scattering techniques and their applications in food Science, Food Biophys., 2006, 1, 214 M. Alexander and D. Dalgleish, Colloids Surf, 2004, 38, 8315 F.GuyomarcH, A. J.R. Law and D. Dalgleish, J. Agri. Food Chem., 2003, 51, 465216 c. Bonnet, M. Corredig and M. Alexander, J. Agri. Food Chem., 2005, 53, 223617 V.Ya. Grinberg and V.B. Tolstoguzov, Food Hydrocoll., 1997, 11, 14518 J.L. Doublier, c. Gamier, D. Renard, and c. Sanchez, ColloidInterf. Sci., 2000, 5, 202


Novel Hydrocolloid Functionality67

19 D. Dalgleish, J. Dairy Res., 1979, 46, 653PERFORMANCE 0F A RES1STANT STARCH TYPE 3 IN NON PRE-FRIED BATTERED FOODT. Sanz, A. Salvador and S.M. FiszmanInstituto de Agroqumica y Tecnologa de Alimentos. Apartado de Correos 73. 46100 Burjassot. Valencia. Spain.1 1NTRODUCTIONThe greater awareness by consumers of the relationship between a good nutritious diet and health and well-being has been one of the reasons for the increase in popularity of novel food with better nutritional properties.Resistant starch (RS) is deined as the sum of starch and Products of starch degradation not absorbed in the small intestine of healthy individuals.1 According to the latest denition of dietary ibre, RS is considered fibre. The fact that RS escapes digestion in the small intestine confers many positive health eects.The objective of the present investigation was to study the suitability of including RS in battered food prepared by a process without pre-frying. Due to the high temperatures necessary for rying (180-190C), a source of RS type 3 (RS3) (Novelose 330) was selected because of its expected higher thermal stability. The specic aspects investigated were the inluence of RS on the thermo rheological behaviour of the batter, its inluence on the bre content, crispness, colour and oil and moisture content of the batter-coated food and, nally, the consumer acceptability of the new battered food.2 MATERIALS AND METHODS2.1 Batter ForimiIaThe batter rmula consisted of RS3 (Novelose 330, National Starch Food Innovation) (8% moisture, pH 5.5 and 33% fibre content) (0, 10 or 20%), salt (5.5%), leavening (Na2H2P207/NaHC03) (3.1%), Methylcellulose (MC) (Methocel A15C, The Dow Chemical Co., DS: 1.6-1.99) (2%) and wheat lour (Harinera Vilaranquina, S.A., Teruel, Spain) up to 100%.The thoroughly pre-blended powders were mixed with cold water (14 C). The water/dry mix proportion was 1.5:1.

2.2 Thermorheological Properties of the BatterThe thermorheological properties of the batters were studied at 15c at 85c and during heating from 15c to 85c by small amplitude oscillatory shear in a controlled stress rheometer (RheoStress 1, Haake, Karlsruhe, Germany) using serrated plate-plate sensor geometry (35-mm diameter) with a 1-mm gap.2.3 Battered Food PreparationThe battered food was prepared according to a manufacturing process that avoids the pre- frying step.2'[footnoteRef:5] Frozen squid rings were used as the food matrix. [5: RESULTSThermorheological PropertiesThe replacement of wheat lour by RS3 in the batter did not affect the temperature at which the moduli started to increase with temperature (Figure 1-A), so the thermogelling ability of MC is not inlluenced by the presence of RS3.The inluence of the RS3 on the mechanical spectra at the beginning (T=15C) and end (T=85C) of the temperature sweep is displayed in Figures 1-B and 1-C. At T=15c, the incorporation of RS3 in the batter produced a slight increase in both viscoelastic iinctions with a slight decrease in tg (G/G) (tg 5 closer to 0), denoting a higher predominance of the elastic component. Hovvever, the requency dependence was not aTected by RS3 incorporation. At 85c, the increase in RS3 content slightly increased the values of both]

2.4 Total Dietary Fiber (TDF) of the Batter Crust after FryingThe TDF content of the batter crust was analysed after frying according to AOAC Official Method 991.43.2.5 Instrumental Measurement of Batter Crust CrispnessA TA-XT2 Texture Analyser (Stable Micro Systems, Godalming, K) was used to evaluate the crispness of the batter crusts immediately after fmal frying. A penetrometry test was performed at lmm/s using a cylinder plunger with a flat base 4 mm in diameter. The values of maximum peak force, initial slope, area below the maximum peak force, and total number of force peaks (drop in force higher than 0.05 N) were calculated.2.6 Colour Measurements of the Battered FoodThe colour of the battered food after fnal rying was measured instrumentally using a Hunter Labscan II colourimeter. The results were expressed in accordance with the CIELAB System with reference to illuminant D65 and a visual angle of 10.2.7 Consumer Sensory AnalysisNovel Hydrocolloid Functionality#


Fifty consumers between 18 and 60 years of age scored the battered samples for appearance, colour, crispness, lavour, overall acceptability and oiliness.moduli, although the tg values were not aected, shovving that the presence of RS3 did not cause any changes in viscoelastic behaviour at 85c.100000100001000100100T(C)10001000,01100(rad/s)1000000100000iiiiiii*:1000010000,01100(rad/s)

Figure 1 A: Heating curves; B: mechanica spectra at T 5C; C: mecharca spectra atsymbols, G V outine symbos.H5 c. Squares: con troi batter, no RS3; circles: 0% RS; trianges: 20% RS. G soid

3.2 Inluence of RS3 on the Properties of the Fried Battered Food3.2.1 Fibre Coitent of the Batter Crust. The percentage of fibre in the batter crust of the no-RS3 formula was 5.0% (0.3) while in the formula with 20% RS3 it was 13.2% (1.2). This implies that the source of RS3 employed (Novelose 330) resists the frying conditions and can be used successully to increase the total dietary fibre content of the fried batter food.3.2.2 Instrumental Measnrement of Batter Crist Crispness. Values of maximum peak force, slope of the curve, area under the maximum peak force and total number of peaks produced during penetration are shown in Table 1. The Progressive introduction of RS increased the maximum brce required to penetrate the crust and raised the slope of the curve, indicating an increase in hardness and fragility, especially for the highest level of replacement, 20%. No signifcant differences were found in the total number of peaks, indicating a similarity in the jaggedness of the curves.3.2.3 Coour f the BatteredFood. The intluence of RS3 incorporation on luminosity or clarity (L*), red component (a*) and yellow component b*) are presented also in Table 1. The increase in the RS3 content of the batter-coated foo produced a signiicant increase in the parameter a* and a signicant decrease in the parameter L*. Visually, the increase in parameter a* and the decrease in parameter L* translated into a more intense golden-brown colour.3.2.4 Fat and Moisture Contents of the Batter Crust. The influence of RS3 on the fat and moisture contents of the different batters is shown in Table 2. The incorporation of RS3 decreases the amount of fat in the batter crust and increases the moisture content. This implies that RS3 contributes to reinforcing the MC barrier against fat absorption and moisture loss during the rying process. This result may be related to the increase in the water retention capacity due to the presence of RS3, which was also relected by the increase in the batter's viscoelastic moduli.Tabe 1 Inhtence of RS3 incorporation on instrumenta texture and colour parameters; bmeans in the same column without a common letter dffer (P 0.05) according to the least signi/icant difference multiple range test; vaues between parentheses are the Standard deviations.RS3contentInstrumental texture parametersColour parameters

Maximum force (N)Slope(N/s)Area (N s)Number of peaksL*a*b*

No RS313.4 a9.4 a9.7 a4.6 a54.1 a9.0 a39.6 a

(4.2)(4.9)(3.1)(2.1)(3.6)(2.7)(4.9)

10% RS314.9 a9.9 ab10.7 a4.9 a50.7 b12.9 b36.3 b

(4.6)(4.7)(2.9)(2.4)(4.0)(2.0)(3.2)

20% RS318.1 b12.0 b13.0 b4.3 a44.7 c16.4 c44.2 c

(5.4)(4.3)(3.7)(2.0)(3.9)(2.9)(8.1)

Table 2 Inluence of RS3 mcorporaion on the fat and moisiure contents after /rying; abmeans in the same column without a common etter differ (P 0.05) accarding to the east signiicant dfference multipe range test; vaues hetween parentheses are the Standard deviations.RS3FatMoisture

content(%)(%)

No RS317.5 a (4.5)23.5 a (5.9)

10% RS313.6 ab (2.8)24.9 a (3.2)

20% RS310.4 b (0.6)27.9 a (1.0)

Table 3 Sensory attribute scores after rying for battered sqnid rings prepared with different evels of RS3; ahmeans in the same coumn without a common letter differ (P< 0.05) according to the east signiicant difference muhiple range test; vaues between parentheses are the Standard deviations.RS3contentAppearanceColourCrispnessOilinessFlavourOverallacceptability

No RS35.3 a (1.9)5.8 a (1.7)5.6 a (1.6)1.9 a (0.5)6.1 ab (1.6)5.6 a (1.4)

10%5.5 a5.9 a5.3 a1.7 a6.3 a5.7 a

RS3(1.3)(1.5)(1.6)(0.6)(1.3)(1.2)

20%5.5 a5.8 a5.2 a1.8 a5.7 b5.4 a

RS3(1.8)(1.8)(1.9)(0.6)(1.5)(1.5)

3.2.5. Acceptability of the Battered Food. The appearance, colour, crispness, oiliness, lavour and overall acceptability scores for all the battered foods are presented in Table 3. No signifcant diTerences were und among the diTerent batter rmulations, implying that the hedonic ratings of the MC batters were not aected by the incorporation of RS3 up to a concentration of 20%.References1 N.G. Asp, I. Bjrck, Trends UoodSci Tech, 1992, 3(5), 111.2 s. Fiszman, A. Salvador, T. Sanz, M.A. Lluch, J Castellano, J. Camps, M. Gamero, Patent W() 03/0228-AI, 2003.3 T. Sanz, A. Salvador and S.M. Fiszman, FoodHydrocoloid, 2004, 18(2), 227. Acknowledgements70Gums and Stablsers for the Food Industry 14



Special thanks to National Starch Food Innovation and The Dow Chemical Company for the supply of RS and MC, respectively, and to Conselleria de Empresa, niversidad y Ciencia of Valencia Government and to Fondo Social Europeo for inancial support.TEXTURAL AND COLOUR CHANGES DURING STORAGE AND SENSORY SHELF LIFE OF MUFFINS CONTAINING RESISTANT STARCHR. Baixauli, A. Salvador and S.M. Fiszman*Instituto de Agroqumica y Tecnologa de Alimentos (CSIC). Apartado de Correos 73, 46100 Burjassot (Valencia), Spain.1 INTRODUCTIONResistant starch (RS) became available commercially some years ago as a food ingredient with a nutritional label listing as dietary ibre (DF). One of the latest deinitions is: DF is the edible parts of plants that are resistant to digestion and absorption in the human small intestine with complete or partial fermentation in the large intestine.1 RS has been defined as the sum of starch and the Products of starch degradation not absorbed in the small intestine of healthy individuals.2 It is well-known and documented the physiological effects of RS, but RS is also important in the diet because of its interactions with other dietary components. Also, as a unctional food ingredient, its low water-holding capacity provides good handling properties during Processing.3 One way to ensure that the general population receives adequate amounts of fibre in the diet is to fortify food that would not normally be associated with bre fortification. Several ibres have been employed to produce high fibre content muffins and cakes.4-7Texture is One of the main characteristics of bakery Products that can be affected by the addition of DF; it can be determined by instrumental or sensory methods. Instrumental methods oer some advantage over sensory analysis because they are rapid and objective. Storage stability or the shelf life of baked Products could be dened as maintenance the sensory and physical characteristics associated with eshness such as crumb tendemess, compressibility and moistness by preventing alteration associated with staling during storage.8,9The objectives of this study are to compare the inluence of replacing increasing proportions of wheat lour with four dierent levels of RS on the textural properties of the muffins, freshly baked and stored for two weeks. This work is a part of a paper published in European Food Research and Technology Journal (DOI: 10.1007/s00217-007-0565-4).2 MATERIALS AND METHODS2.1 Batter and muffn preparation

Five formulations were prepared using the same quantity of all the ingredients except the flour and RS, which were 26/0, 21/5, 16/10, 11/15 and 6/20 per cent respectively. The batterrmulation (expressed as a percentage of weight) consisted of wheat lour (Harinera ViMranquina, S.A., Teruel, Spain); resistant starch (HI-MAIZE 260, National Starch Food Innovation, Manchester, United Kingdom), sugar (26%) (Azucarera Ebro, Madrid, Spain); liquid pasteurized egg white (14%) and liquid pasteurized yolk (7%) (Ovocity, Llombay, Spain); fll-fat milk (13%) (Puleva Food, Granada, Spain); refined sunflower oil (12%) (local supermarket), sodium bicarbonate (1.03%), citric acid (0.79%) and grated lemon peel (0.18%). All ingredients were mixed in a mixer (Kenwood Major Classic, UK). The batter was placed in an automatic dosing unit (Edhard Corp., FIackettstown, USA), and each paper muffin cup (50-mm diameter) was flled with 40.5 g of batter. The muins were baked in a conventional oven for 6 min at 225 c and for a urther 6 min at 175 c. The muffns from each formulation and for each storage time were prepared twice, on different days, with twenty-four muffms in each batch.After cooling, the muffins were packed in polyethylene bags that were heat-sealed and stored in an environmental chamber at room temperature (20 2 C). The mun samples were evaluated on days 0, 2, 4, 7, 9, 11, 14 and 16.2.2 Measurement of colourThe instrumental measurement of the muffin colour was carried out with a Hunter Labscan II colorimeter, the results were expressed in accordance with the CIELAB System with reference to illuminant D65 and a visual angle of 10. The measurements were performed through a 6.4-mm-diameter diaphragm containing an optical glass. The parameters determined were L*, a*, b*, c* and H*. The total colour difference (AE*) between the control mun and the muffins with RS was calculated as follows:AE* = ((AL*)2 +{Aa*f +(Ab*)22The values used to determine if the total colour difference was visually obvious were the following:I0AE* G"; little variation of G' with requency, co; steep, linear reduction in log T* on increasing log co, with a slope close to -1), but the values of G', G" and T|* for starch from the heated wheat flour are about twice those of the unheated control. This enhancement in rheology cannot be attributed to greater space-occupancy since, as reported in Section 3, both samples have virtually the same swelling volume (~9.6 ml/g).

I '10.1 1 10 100 Frequency (rad s"1)

Eigure 3 Mechancal spectra (5C; 1 % stran) showing the /requency-dependence of G' (squares), G" (circles) and ]* (triangles) for 7.4 wt % starch from heated (fed symbos) and unheated (open symbols) wheat flour after gelatinisation under low-ampltude oscllationThe moduli recorded after heating and cooling under quiescent conditions (low-amplitude oscillation at 10 rad s'1 and 1 % strain) over a range of values of starch concentration (c = 1.0 to 10.0 wt %) are also inconsistent with simple physical contacts between swollen granules. As shown in Figure 4, the plot of log G' versus log c for starch from unheated wheat lour is broadly similar to those obtained for gelling biopolymers,11 with a Progressive increase in slope as concentration is decreased towards the minimum critical gelling concentration (c0) and a limiting slope of ~2 (c2-dependence of G') at c c0. In particular, there is no evidence of any sudden increase in moduli, or other discontinuity in the concentration-dependence of G' or G", as the degree of space-occupancy approaches and exceeds the onset of close-packing (at cQ 0.65). Similar concentration-dependence of moduli was observed for starch from heated wheat flour4 and for crosslinked waxy maize starch,5 with again no discontinuity at the onset of close-packing.

As shown in Figure 5a, the mechanical spectrum obtained for the control sample of wheat starch at a concentration of 2.5 wt %, where the swollen granules occupy less than a quarter of the total volume (cQ = 0.24), is predominantly gel-like: G' is virtually independent of requcncy across the range studied (0.1 to 100 rad s"1) and at low requencies it exceeds G" by more than an order of magnitude. The uptum in G" at higher requencies can be readily explained by a signiicant "sol raction" of material that does not form part of the gel network.log (c/wt %)Figure4 Concentration-dependence of G' () and G" (o), measured (10 rad s'1;1 % strain) at 5c after heating and cooling under low-amplitude oscillation, for starch from unheated wheat four

Figure 5 Mechanica spectra showng the /requency-dependence of G' (), G" () and 7]* (A) aer gelatinsaton under low-amplitude oscillation at the same degree of space-occupancy (cQ = 0.24) for (a) starch from unheated wheat flour (c = 0.25 wt %; Q = 9.56 ml/g) and (b) crosslinked waxy maze starch (c = 0.20 wt %; Q = 11.9 ml/g)

Formation of a continuous network at concentrations far below those at which the swollen granules are forced into contact with one another cannot be explained by gelation of the amylose component of the wheat starch since, as shown in Figure 5b, a similar gel-like spectrum was obtained5 at the same low degree of space-occupany (cQ = 0.24) for crosslinked waxy maize starch, vvhich, like any waxy starch, is of course essentially devoid of amylose. Because of its somewhat higher swelling volume (11.9 ml/g in comparison with 9.56 ml/g for the wheat starch), hovvever, the concentration of the waxy maize starch at cQ = 0.24 is lower (2.0 wt % in comparison with 2.5 wt %), and the moduli are also lower (Figure 5), as would be expected for conventional gelling biopolymers.The obvious interpretation of all the rheological evidence presented so far is that starch, when gelatinised under quiescent conditions, orms continuous networks by associative interactions (i.e. by the swollen granules adhering to one another). The enhanced rheology (Figures 2 and 3) of starch from heated wheat flour indicates that pre-heating the lour increases the adhesiveness of the gelatinised granules, which is consistent with an unexpected change in their interaction with blue dextran, reported below.7 INTERACTION WITH BLUE DEXTRANWhen starch from unheated wheat flour was gelatinised in a solution of blue dextran, swelling of the granules caused the expected increase in concentration of polymer in the surrounding liquid, and the svvelling volume calculated from this increase was in good agreement with the value of cQ = 9.56 ml/g derived (Section 3) by sedimentation of the swollen granules. When the same procedure was applied to starch from heated wheat flour, however, the concentration of blue dextran in the solution phase decreased, rather than increasing, and the gelatinised granules had a deep blue colour which persisted after repeated rinsing with water.This behaviour is unlikely to have arisen from penetration of blue dextran into the granules, particularly since DSC (Figure 1) showed no evidence of any signiicant change in their intemal structure prior to gelatinisation. The most likely inteipretation is that the exterior of the granules is modied during pre-heating of the lour (possibly by changes in surface protein), with the modied surace then being capable of binding to the dye chromophores of blue dextran, and of causing stronger adhesion between gelatinised granules (with consequent enhancement of network structure).8 EFFECT OF SHEARING DURING GELATINISATIONIn Systems where interactions are limited to physical contacts (e.g. Solutions of entangled polymer coils) the requency-dependence of T|* and shear-rate dependence of viscosity (r|) superimpose closely1 when comparison is made at equivalent numerical values of equency (co/rad s'1) and shear rate ( /s'1). However, departures from this generality (which is known as the "Cox-Merz rule")12 occur for weakly-crosslinked networks that remain intact under low-amplitude oscillation but are broken down by shear,10 giving T|* > T|.In the studies reported here, comparisons were made using a requency of 10 rad s'1 and a shear rate of 10 s'1. These values were chosen as a compromise between the conlicting considerations of having a shear rate high enough to give reliable measurements and approximate to normal Processing conditions in industry and a requency low enough to prevent the elastic response of any crosslinked netvvork structure from being obscured by simple physical interactions, which make a progressively greater contribution to overall resistance as the requency of oscillation is increased10.The results obtained are illustrated in Figure 6 for 7.4 wt % starch from unheated wheat lour. The increase in T|* on gelatinisation is substantially greater than the corresponding increase in T|, and T|* then remains above r| throughout urther heating and subsequent cooling, giving a value of T|* about 5 times higher than that of r| on completion of cooling to 5c. Similar violation of the Cox-Merz rule was observed4 for starch from heated wheat lour, and, as discussed above, strongly indicates associative interactions that can survive the low-amplitude used in measurement of T|* but are disrupted by rotational measurements of T. Further Cox-Merz comparisons were made for starch from both heated and unheated wheat lour on completion of cooling to 5c after gelatinisation under unperturbed conditions and under shear at starch concentrations ranging from 1.0 to 10.0 wt %.1000 iTT,TTTTT,1

0 20406080100Temperature (C)Figure 6 Comparison of ]*from oscillatory measurements at 10 rad s'1 (open symbols) and from rotational measurements at 10 s'1 (filled symbols) during heatng (triangles) and cooling (circles) for 7.4 wt % starch from unheated wheat flourFigure 7 shows comparisons of the fmal values of r| (10 s'1) for the sheared samples with the corresponding values of r|* (10 rad s"1) for the same sheared preparations (obtained by oscillatory measurements immediately after completion of cooling) and for equivalent samples prepared by heating and cooling under oscillation (as in Figure 2). The values of r|* for the samples gelatinised under oscillation are consistently higher than the corresponding values of r| for the sheared samples, again indicating associations that can survive low-amplitude oscillation but are disrupted by shear. The sheared samples, by contrast, show reasonable Cox-Merz superposition (r|* r|), indicating that shearing during gelatinisation prevents association, leaving only physical contacts betvveen the gelatinised granules.For both sheared and unsheared preparations, low-amplitude oscillatory measurements of T|* (10 rad s'1; 1 % strain) were made over a period of 2 hours at 5c after completion of cooling. No changes in T|* were observed,4 demonstrating (i) that no urther associations are formed by the samples gelatinised under unperturbed conditions and (ii) that the samples gelatinised under shear show no recovery from the disruption caused by shearing.:1I rmrnInrTTTTT1ImTTTT1rTTTTTT11rrrrn- '' .= A:100^10cd h*p 1 0.1 0.01

: ,':A A: 1 A .''(::1^ o .'':!A- y11It -i-LUlII1 1 i-LiilIIU-LLLLlIl I I I I lll11L-Lxm0.01 0.1 1 10 100 1000T (Pa s)Figure 7 Comparison of oscillatory measurements of r* (10 rad s'1; 5C) and rotational measurements of T (10 s'1) for sheared (open symbols) and un-sheared (fed symbols) preparations (1.0 - 10.0 wt %) of starch from heated (triangles) and unheated (circles) wheat flour. The dotted line corresponds to perfect agreement between ] and T* (Cox-Merz superposition)9 CONCLUSIONSThe main conclusions from the stuies reported above can be summarised as follows. When starch is gelatinised under quiescent conditions (such as low-amplitude oscillation) the swollen granules adhere to one another, and form gel netvvorks at concentrations well below the onset of close-packing (as illustrated in Figure 5 for both wheat starch and crosslinked waxy maize starch). Pre-heating wheat lour increases the adhesiveness of the starch granules, giving higher moduli than for control starch from unheated lour, although the swelling volume remains virtually unchanged. Enhanced rheology indicates modiTication of the granule surface, which also leads to binding of blue dextran. Shearing (e.g. during rotational measurements of viscosity) prevents association of swollen granules. Disruption by shear is irreversible (sheared preparations showed no increase in T|* on holding for 2 h at 5C). Evaluation of lours and starches by conventional measurements under shear (e.g. Amylograph) is thereore appropriate for Products such as soups and sauces which are stirred during Processing (preventing association of swollen granules). The same procedures, hovvever, may be misleading for baked Products, where gelatinisation of starch occurs under quiescent conditions, allowing associations between gelatinised granules to form in the product, but not during rotational testing.

Novel Hydrocolloid Functionality ACKNOWLEDGEMENTSWe thank Professor J.D. Scholeld for helpil discussions. The results reported for crosslinked waxy maize starch emanated from research conducted with the nancial support of Science Foundation Ireland.References1 D. Cooke and M.J. Gidley, Carbohydr. Res., 1992, 227, 103.2 I.D. Evans and A. Lips, J. Text. Stud., 1992, 23, 69.3 C.EA.M. Keetels, T. van Vliet and p. Walstra, FoodHydrocolloid,, 1996,10, 355.4 s. Hasan, Heat Treatment ofWheat Flour, PhD Thesis, University College Cork, 2006.5 S.M. Fitzsimons, D.M. Mulvihill and E.R. Morris, Food Hydrocolloids, 2007, in press.6 A.c. Johnson and R.c. Hoseney, Cereal Chem., 1979, 57, 92.7 c. Kusanose, s. Noguchi, T. Yamagishi and M. Seguchi, FoodHydrocoloids, 2002,16, 73.8 M. Nayouf, c. Loisel and J.L. Doublier, J. Food Eng., 2003, 59, 209.9 J.w. Donovan, K. Lorenz and K. Kulp, Cereal Chem., 1983, 60, 381.10 S.B. Ross-Murphy, Biophysical Methods in Food Research, SCI, 1984, p. 195.11 A.H. Clark and S.B. Ross-Murphy, Brit. Polym. J., 1985, 17, 164.1000112Gums and Stabilisers for the Food Industry 14

#

113

12 w.p. Cox and E.H. Merz,y. Polym. Sci., 1958, 28, 619.Physically moditied xanthan gum prepared by extrusionProcessingNuno M. Sereno, Sandra E. Hill, John R. MitchellDivision of Food Sciences, School of Biosciences, University of Nottingham, Sutton Bonington Campus, Loughborough, LE 12 5RD, UK1 INTRODUCTIONXanthan gum, since its acceptance by the American Food and Drug Administration in 1969, has become a thickener of choice in many food applications \ Xanthan gum is a high molecular weight polysaccharide (106 g/mol2"5) which is produced by aerobic fermentation of a pure culture of the bacterium Xanthomonas campestris 1.Once in solution the xanthan gums trisaccharide side chains align with the glucose backbone, stabilising the overall structure by non-covalent interactions and giving rise to an ordered helical organization 6. This structure is disrupted when submitted to higher temperatures (heat treatment above the xanthan gum melting temperature, Tm), and goes through a conformation transition into a disordered molecular coil. The onset of the melting transition, as well as its temperature range, is mainly dependent on: the concentration of the polymer 78, the ratio of acetyl and pyruvate groups 9,10 and the ionic strength of the System H"14. On cooling to temperatures below xanthan gums melting temperature a restoration of the helical order within the System follows.A well-recognise problem with xanthan gum has been its poor dispersibility and hydration. Thereore to achieve a homogenous dispersion, xanthan gum must be hydrated under high shear mixing for a signiEcant period of time. In order to improve xanthan gum hydration properties several Solutions have been proposed. These include production of agglomerate structures, addition of suractants and crosslinking with glyoxal l5"20.

We have previously reported that by Processing xanthan by twin-screw extrusion cooking under mild conditions, it is possible to produce a material which disperses far more readily than other non-chemically modiied xanthans 21. This study investigates the solution behaviour shown by extruded xanthan gum, in order to understand the reasons behind the observed enhanced hydration, viscosity and mixing properties.2 MATERIALS AND METHODS2.1 Extrusion and Post-Processing TreatmentXanthan gum (Satiaxane cx 910, Degusa Texturant Systems, France) was extruded with a co-rotating Twin Screw Clextral BC12 Extruder (Clextral, Firmeny-Cedex, France). A slit die of 1 mm X 30 mm was fitted to the barreTs exit. The extrusion conditions were as follows: screw diameter 24 mm; screw length 400 mm; screw speed 100 rpm; solid feed rate 3.5 kg/h; water flow rate 2.14 L/hr; total amount of water inside the barrel 2.57 kg/h; temperature of barrel heating zones 85, 85, 70 c.The extruded xanthan gum material was dried in a vacuum oven (Sanyo Gallenkamp PLC) at 65c for approximately 72 hours under a pressure of 100 Pa, and milled to a particle size between 125 and 250 pm. Final water content was lower than 8% (wet basis).2.3 Viscosity MeasurementsThe flow properties of fully hydrated xanthan gum dispersions were characterised using both the Rapid Visco Analyser (RVA) (Newport Scientic, Calibre Control International, Warrington, UK) and Bohlin CVO-R rotational rheometer (Malvem Instruments Ltd, Worcestershire, UK). The samples were prepared at a range of salt and xanthan gum concentrations by adding the solid powder to distilled water or NaCl Solutions and stirring at 700 rpm for 20 minutes. The samples were allowed to settle for a minimum of 2 hours to complete hydration.2.4 MicrocalorimetryThe thermal transition of 0.75% xanthan gum dispersions at different NaCl concentrations was followed using a Micro DSC III (Setaram, France). The Micro DSC III was operated from 20 to 120c at a fix rate of 1 c/min.2.5 Optical MicroscopyDry xanthan gum particles (125-250 pm) were placed on a Standard glass microscope slide, and a solution of toluidine blue (5 mg/kg water) was careully added. An optical light microscope (Wild Leitz GmbH, Wetzlar, Gennany) was used. Depending on the required magnification a 4x or 10x objective with lx eyepiece was used.2.6 Swelling Mass Ratio DeterminationThe swelling mass ratio (Q, g swollen extrudate/g dry extrudate) of extruded xanthan gum dispersed in different NaCl concentrated Solutions (xanthan gum concentration = 0.75%) was determined following centrifugation. Centriugation of xanthan gum dispersions was perormed on a Multex centrifuge (Multex MSE P522A, Crawley, K) with a radius to inner tip of 140 mm and operating at 4000 rpm for 2 hour (2500 g).2.7 Polyelectrolyte TitrationPolyelectrolyte titration against polydimethyl diallyl ammonium chloride (0.001N), was used to assay xanthan in the supematant following centrilugation. The end point was detennined as zero streaming potential measured in a Mutek PCD 03 apparatus (Mutek,

Germany). The concentrations were calculated from xanthan gum calibration curves obtained both in distilled water, and in 0.5% NaCl. To validate this approach the results were compared with the total sugar content of the supematant extracted from a 0.75% extruded xanthan gum dispersion in 0.1% and 0.5% NaCl Solutions (%, wet weight basis). These were measured using the colorimetric method described by Dubois et al.22.2.8 Mixing Behaviour and Sodium Delivery DeterminationTwo Systems of non-processed and extraded xanthan gum were prepared in 0.25% NaCl with adequate concentrations to obtain the same final viscosity of 450mPa.s at 50s_1. An aliquot (lOmL) of each viscous dispersion was careully introduced in 40ml of water. The Systems were mixed with a spoon for 3s in a circular motion. Pictures were taken and sodium concentration in the aqueous phase measured using Na+ specihc electrodes (Microelectrodes Inc, Bedbrd, USA).3 RESULTS AND DISCUSSION3.1 Effect of Extrusion Processing on Xanthan Water Dispersibility and ViscosityThe twin-screw extrusion of xanthan gum produced a material with greatly improved dispersibility and enhanced room temperature viscosity (Figure 1).(a)(b)1E+04 A1E+03(/) 1E+011E-011E-02 0,50% 0.30%A 0.20% 0.10% A 0.05% X 0.03% 0.50% 4 0.30% 4 0.20% 0.10% A 0.05% -l 0.03%*A*AAi E,.*44V**\ 4 4 4 44"34 4Ddda X... D 4* _4a*4*^4,nDDo44Aa Dn * y.aD4 ,,.1E-041E-031E-021E-011E+00 1E+01 1E+021E-041E-031E-21E-011E+00 1E+011E+02

Shear rate (1/s)Shear rate (1/s)Figure 1. Viscosit of (a) non-processed xanthcin gum and (b) extruded xanthan gum distied water dispersions at dfferent concentrations (%, wet weight basis). Measurements were performed in the Bohlin CVO-R.As can be observed in Figure 1 both forms of xanthan gum display a clear shear-thinning behaviour, but whereas a Newtonian plateau at low shear rates is apparent for the fully

hydrated control material, this is not apparent for the extruded materia]. The lack of a Newtonian plateau and the stronger concentration dependence of the extruded material at higher concentrations, is characteristic of a suspension of swollen particles rather than that of a polymer suspension as is the case of non-processed xanthan gum.To verify this hypothesis, the hydration of particles of processed and control xanthan gum was visualized under an optical microscope. The non-processed xanthan gum particle when in contact with excess water presented a very quick initial wetting (fst 30 seconds), followed by a more gentle hydration during the next minute and a half and an eventual dissolution (Figure 2(a)). A similar behaviour has also been observed by Clark 23. Figure 2 (b) shows a different behaviour for the extruded xanthan gum particle. Again, as observed for non-processed xanthan gum, an almost immediate water penetration into the dry particle occurred and caused a rapid increase in volume (lrst 30 seconds), but now the swollen particle remained stable without a change in size over the period of observation.Figure 2 Hvdration in distiled water of (a)3.2 Impact of Salt on Xanthan Gum Souton Behaviour(b)0 seconds15 seconds

30 seconds2 minutes

5 minutes10 minutes

non-processed and (b) extruded xanthan gum.

The impact of solution ionic strength on suspension viscosity can provide insight to the reasons behind the behaviour differences observed by extruded xanthan gum. The addition of salt to non-processed xanthan gum results in a small increase in solution viscosity 24. Such response is due to the shielding the ionic groups on the xanthan side chains. This causes a reduction of individual molecular volume occupancy, but also promotes an increase in molecular entanglements in the dispersion, leading to an increase in viscosity 25. In the semi-dilute System this latter situation will be more signicant ]\ and the System viscosity increases with increasing NaCl concentration until a maximum above which, further addition of NaCl had little effect on dispersion viscosity24.118Gums and Stabilisers for the Food Industry 14


In contrast the extruded material showed a general low temperature viscosity reduction with increasing levels of NaCl (Figure 3). The addition of small amounts of salt ( 10005000 420 25 30 35 40 45 50 55 60 65 70 75 80 85 90Temperature (C)Figure 3 Viscosity measurement of non-processed xanthan gum (0.75%, wet weight bass)in distled water and extrudedxanthan gum (0.75%, wet weight basis) in NaC Solutions ofincreasng concentration. Measurements were performed in the Rapid Visco Anayser.This strong ionic strength dependence O1 viscosity of extruded xanthan gum is aconsequence of the reduction in the degree of sweling of the particles with saltconcentration. The swelling mass ratio (Q), for a dispersion of 0.75% (wet weight basis)extruded xanthan gum in distilled water was 1110 gsvveiiing/gdry xanthan, which decreased to 27gswei]m/gdry xanthan when Q was determined in the presence of a 1.0% NaCl solution.Although the above observations suggest that the dominant behaviour of extrudedxanthan gum is particulate in origin, it is important to determine the amount of polymermass retained within the particulate structure as these can have an impact in some of theSystems properties (e.g. salt delivery from viscous dispersions). As can be seen in Table 1most of the extruded material remained in the particulate form. It also appeared that theincreased ionic strength had the ability to decrease xanthan gum molecular leakage fromthe granule to the solution. This reinorces the view that increasing ionic strength promoteshigher levels of molecular interactions, which therefore allowed for less leakage from theparticles. t is also possible that the larger swollen particles at lower salt leveis were moresusceptible to disruption on centriugation. The maintenance of the majority of theprocessed xanthan gum in the particulate form explains its good dispersibility, sinceentanglements between molecules partly released from powder particles is reduced whencompared with non-processed xanthan.nonprocxanthanextrudedxanthanNaCINaCITable 1 Concentration of xanthan retained n particuate form of extruded xanthan gum dispersions (0.75%, wet weight basis) foowng centri/ugaton. Xanthan gum dispersions were prepared at different NaC eves. Two methods were used: (a) poyeectrote titration method, (b) tota sugar content method22.

3.3 Xanthan Gum Temperature Induced Conormational TransitionThe temperature and enthalpy of melting and rebrming of the xanthan gum ordered structure was investigated in order to determine its involvement in maintaining the integrity of the hydrated xanthan gum particles. As seen in Figure 3 there was a viscosity dependency with temperature for the processed material. The increase in temperature originated a transition, which was characterized by a viscosity decrease at low NaCl concentrations or as a viscosity peak at higher NaCl concentrations. The transition shited to higher temperatures as NaCl concentration in the System increased.Figure 4 (b) shows that increased NaCl concentrations altered the transition temperature of extruded xanthan gum dispersions in a similar ashion as observed by the rheological measurements. At higher salt concentrations it is possible to compare the temperature for maximum viscosity, with the peak temperature from the heating endotherms of processed xanthan gum. The similarity between the two temperatures supports the hypothesis that the particulate nature of xanth

gums and stabilisers for the food industry 14 (special publications)-royal society of chemistry...

Documents