.- ‘. ~33» >11... THESIS ’ LIERARY Whitman State University a. .._., --.--. r This is to certify that the dissertation entitled SI'UDIES OF DRY-ROASTED AND AIR-CLASSIFIED NAVY BEAN FIDUR FRACTIONS: PHYSICDCHEMICAL CHARACTERISTICS, SIORAGE STABILITY, AND RHEOLOGICAL PROPERTIES presented by James J. Lee has been accepted towards fulfillment of the requirements for Ph.D. Food Science degree in Mm” Zflr first. Major professor Dr. Mark A. Uebersax Date 8 ’6 “19¢ MSU is an Affirmative Action/Equal Opportunity Institution 0- 12771 [MSG- —— ——— ’O'cre “ fifi‘vmg ’ 9 26’ MAY {>4 2 n L. ‘C. l.)- BEIURNING MATERIALS: Place in book drop to remove this checkout from your record. FINES will be charged if booE is returned after the date stamped below. T STUDIES OF DRYeROASTEDiAND.AIRrCLASSIFIED NAVY BEAN FLOUR FRACI'IONS: PHYSICOCHEMICAL CHARACTERISTICS, SIORAGE STABILITY,.AND RHEOLOGICAL PROPERTIES By James J. Lee A DISSERTATION Submitted to Nfichigan.State'University in.partial fulfillment of the requirements fer the degree of DOCTOR.OF PHIDOSOPHY Department of Food Science and.Hnmen Nutrition 1984 :3 I 7 0/ {LAX ABSTRACT STUDIES OF DM-ROASIED AND AIR-CLASSIFIED NAVY BEAN FLOUR FRACTIONS: PHYSICOCHEMICAL CHARACTERISTICS, SIORAGE STABILITY, AND RHEOLOGICAL PROPERTIES By James J. Lee Navy beans (Phaseolus vulgaris) were dry-roasted in a particle-to- particle heat exchanger under various roasting conditions, dehulled by air aspiration, pin-milled, and air-classified to yield whole, hull, high protein, and high starch flour fractions. Evaluation of these flours was conducted in three studies. Cmpositimal differences were demonstrated among all flour fractions. Stachyose was the major oligosaccharide in all fractions and was the highest in the protein fraction. Substituting 10% bean flours for wheat changed water absorption and dough stability for all fractions. Pasting characteristics of whole flour and high starch flours were damnstrated to be controlled swelling at high temperature. Flour fractions fram four roasting treatments were stored under eight water activity levels (a; s) . High roasting time/tamerature con- ditions resulted in decreased equilibrium moisture. After two months storage, flours in a” above 0.75 were moldy. Flours stored in increased ‘3: showed increased browning with concmrrent decrease of protein solu- bility and glucose content. Flours stored up to 24 months with 67!. and 92 moisture at 20°C or with 62 moisture at 37.7°C retained good quality. James J. Lee Gelatinization and rheological properties of purified starch pastes (67., 8%, 102, and 122) were evaluated using a computer assisted Haake Viscometer. Pastes were stirred at a constant shear rate of 50 rpm at 75°C, 85°C, and 95°C for up to 60 min. Activation energy of starch gelatinization was estimated to be 14.5 Kcal/g mole. Various swelling responses of starch dmring cooking were demonstrated. Maximum final viscosity values were obtained at 85°C for all starch concentrations except a maximum at 95°C for 67.. 'I‘nixotropic breakdown was observed at 107. and 12% concentration (firing initial shearing at 85°C and 95°C. Flow behavior, retrogradation tendency, and stability of the hot and cooled (50°C) pastes were determined after cooking. The power law, Herschel-Bulkley, and Casson models characterized the paste flow. Gelatinized pastes exhibited shear thinning behavior; their flow curves best fit the Herschel-Bulkley model. Yield stress, consistency index, and flow behavior index were concentration and temperature dependent . Significant paste rheological property changes centred at starch concentratims above 102 for 75°C and above 82 to 102 for 85°C and 95°C. To my loving wife, Shu-wei ii I wish to express my deepest appreciation to my major professor, Dr. Mark A. Uebersax, for his guidance, indulgence, and encouragement throughout my graduate program. He, as a brilliant teacher, provided his support and friendship which have made it an important turning point in my career. Appreciation is also extended to Drs. D.R. Heldman, G.L. Hosfield, J.F. Steffe, and M.E. Zabik, for serving as marbers of the guidance camrittee and for providing criticisms of my manuscript. Special recognition is given to the United States Department of Agriculture, Science and Education Administration, Washington, D.C. , for funding the Bean Flom: Project, Research Agreement No. 59-2481-0-2-001-0 and 59-2261-1-2-004-0. I amvery grateful to Mrs. Uebersax, who is so special tomy family, for her loving care and constant inspiration, and to my fellow graduate studatts in Lab 128 for their companionship and suggestions during the period the research was conducted. Finally, I am deeply indebted to my parents, from whom I have learned to treasure the desire for mending knowledge. TABLE OFCONTENTS Page LIST OF TABLES ....................... ix LIST OF FIGURES ...................... x11 LISI‘ OF SYMBOLS ...................... xv INTRODUCTICN ........................ 1 REVIEJ 0F LITERATURE .................... 6 Production of Bean Flours ............... 6 Wet Processing .................. 6 Dry Milling and Air-Classification ........ 11 Raw Beans .................. ll Roasted Beans ................ 14 Physicocheni cal Properties of Bean Flours ....... l7 Carposition of Dry Beans ............. l7 Starch Flours .................. 18 Protein Flours .................. 22 Storage Stability of Bean Floors 23 Quality Changes chn'ing Storage .......... 24 Protein Solubility ............. 24 Non-Enzymatic Browning ........... 25 Lipid Oxidation ............... 26 Factors Affecting Flour Stability ........ 28 Rheological Properties of Starch Pas tes ........ 33 Colloidal Nature of Starch Pastes ........ 33 Gelatinization ............... 33 Retrogradation ............... 35 Factors Affecting Paste Viscosity ...... 35 iv Production of Flour Fractions Experimental Procedures Fractions and Starch Rheological Property Studies in Food Systems Quality Control ............... Engineering Applications .......... Methods for Rheological Measurements ....... Non-Newtonian Flow Behavior of Starch Pastes Flow Pbdels ................. Flow Behavior of Starch Pastes ....... Roasting, Milling, and Air-Classification Purification of High Starch Flour ........ Study 1: Physicochermical Characteristics of Dry-Roasted Navy Bean Flour Fractions ...... Study 2: Effect of Roasting and Storage Conditions on the Stability of Air-Class- ified Navy Bean Flour Fractions ......... Study 3: Gelatinization and Rheological Properties of Purified Navy Bean Starch Pas tes ...................... Protein ..................... Extraction ................. (EPIC) Analysis ............... Starch ...................... Aliquot Preparation ............. Enzyme Solution Preparation ......... Assay .................... HunterLab Color Values .............. Source of Beans ................. 38 38 39 39 45 45 48 51 51 51 51 55 57 57 58 59 64 64 65 65 65 66 66 66 66 67 69 69 69 70 Mixing Properties of Dough ............ 70 Pasting Characteristics of Starch ........ 70 2-Thiobarbituric Acid (TBA) ........... 71 Test Apparatus for Rheological Measurements ...... 71 Calculations of Shear Stress and Shear Rate ...... 72 Theory of Coaxial Cylinder Viscameter ...... 72 Shear Stress ................... 74 Shear Rate .................... 75 Flow Madels and Calculations of Rheological Parameters ...................... 76 Statistical Analysis ................. 77 RESULTS AND DISCUSSION ................... 79 Study 1: Physicochamical Characteristics of Dry- Roasted Navy Bean Flour Fractions ........... 79 Chemical Composition ............... 79 Physical F‘metionality .............. 81 Study 2: Effect of Roasting and Storage Conditions on the Stability of Air-Classified Navy Bean Flour Fractions ............... 87 Proximate Composition of Flour Fractions ..... 87 Equilibrium lVbisture Content ........... 87 Surface Mold Growth ............... 92 Property Cl'langes after Storage .......... 96 Long Term Storage Stability ........... 101 Study 3: Gelatinization and Rheological Properties of Purified Navy Bean Starch Pastes . . . . 104 Proocimate Camposition of Purified Starch ..... 104 Preparation of Starch Pastes ........... 104 Starch Gelatinization ............... 107 Activation Energy in Starch Gelatinization . . . . 116 Yield Stress of Starch Pastes .......... 120 Characterization of the Flow Curves ....... 122 Rheological Parameters of Starch Pastes ..... 123 Hot Pastes ................. 123 Cooled Pastes ................ 131 SUPMARY AND CONCLUSIONS .................. 138 WomstFURmmRESEARm ...... 141 APPENDD( A ......................... 142 Flow Data of 67. Navy Bean Starch Paste Cooked at 95°C for 60 Min Prior to Measurements with The Haake Viscameter ................. 142 APPENDIX B ......................... 143 Flow Data of 87. Navy Bean Starch Paste Cooked at 95°C for 60 Min Prior to Measurements with The Haake Viscometer ................. 143 APPENDIX C ......................... 144 Flow Data of 102 Navy Bean Starch Paste Cooked at 95°C for 60 Min Prior to Measm‘arents with The Haake Viscaneter ................. 144 APPENDIX D ......................... 145 Flow Data of 122 Navy Bean Starch Paste Cooked at 95°C for 60 Min Prior to Measmaments with The Haake Viscameter ................. 145 APPENDIX E ......................... 146 FlowData of 6ZNavyBeanStarchPasteCooked at 95°C for 60 Min and Cooled to 50°C Prior to Measmranents with The Haake Viscometer ........ 146 APPENDDK F ......................... 147 Flow Data of 82 Navy Bean Starch Paste Cooked at 95°C for 60 Min and Cooled to 50°C Prior to WEEKS with The Haake Viscameter ........ 147 APPENDIX C ......................... 148 Flow Data of 10% Navy Bean Starch Paste Cooked at 95°C for 60 Min and Cooled to 50°C Prior to masurements with The Haake Viscameter ......... 148 APPENDIX H ......................... 149 Flow Data of 127. Navy Bean Starch Paste Cooked at 95°C for 60 Min and Cooled to 50°C Prior to Measurements with The Haake Viscometer ........ 149 APPENDIX I ......................... 150 Flow Data of 67. Navy Bean Starch Paste Cooked at 95°C for 60 Min, Cooled to 50°C, and Held at 50°C for 30 Min Prior to Measurements with The Haake Viscometer ................. 150 APPENDIX J ......................... 151 Flow Data of 82 Navy Bean Starch Paste Cooked at 95°C for 60 Min, Cooled to 50°C, and Held at 50°C for 30 Min Prior to Phasurements with The Haake Viscaneter ................. 151 mm K ......................... ' 152 Analysis of Variance of Rheological Parameters Computed from Selected Flow Models for Navy Bean Starch Pastes of Various Concentrations Cooked at 75°C, 85°C, and 95°C for 60 Min Prior to Measurements with The Haake Viscateter ....... 152 APPENDIX L ......................... A 153 Correlation Coefficients , r2 , and Sun-of- Squares Function, SS(b) , for Regression Analysis of The Flow Curve Data .................... 153 APPENDDf M ......................... 155 Analysis of Variance of Rheological Parameters Ccnputed from Selected Flow Models for Navy Bean Starch Pastes of Various Concentrations Cooked at 75°C, 85°C, and 95°C for 60 Min and Cooled to 50°C Prior to Measm'arents with The Haake Viscom- eter ......................... 155 LIST OF REFERENCES ..................... 156 Table 10. 11. LIST OF TABLES Yield aid chemical analyses of dry-roasted navy bean flour fractions .............. HtmterIab color values of dry-roasted navy beat flour fractions ................ Farinograph values of dry-roasted navy beat flour fractions ................... Proximete composition of dry-roasted navy beat flour fractions obtained from beans processed at various conditions ................ . “ HunterLab color values of dry-roasted navy beat flour fractions stored under various water activity levels for eight weeks ....... HPLC sugar analysis of dry-roasted navy beam flour fractions stored under various water activity levels for eight weeks .......... Starch aelysis of dry—roasted navy bean flour fractions stored under various water activity levels for eight weeks .......... Hunterlab color values of dry-roasted navy beat flour fractions stored at varying moisture contaits at 20°C aid 37.7°C for up to 24 months ................... Nitrogen solubility indices and TBA values of dry-roasted navy bean flour fractions stored at varying moisture contents at 20°C and 37.7°C for up to 24 months ........... Proximate carposition of air-classified high starch flour and purified starch of Final torque values for 102 navy beat starch pastes stirred at various paddle speeds aid cooked at 75°C and 95°C for 60 min ......... ix Page 80 83 84 88 97 100 101 102 103 105 107 First order reaction kinetics of navy bean starch gelatinization at various cooking temperatures ..................... 118 Rheological parameters computed fram selected flow models for navy bean starch pastes of various concentrations cooked at 75°C, 85°C, aid 95°C for 60 min prior to measurements with the Haake Viscaxeter ................. 125 Rheological parameters computed fran selected flow models for navy bea1 starch pastes of various concentrations cooked at 75°C, 85°C, and 95°C for 60 min and cooled to 50°C prior to measurarents with the Haake Viscometer ...... 132 Rheological parameters ccmputed from selected flow models for navy beam starch pastes of various concentrations cooked at 75°C, 85°C, aid 95°C for 60 min, cooled to 50°C, and held at 50°C for 30 min prior to measurements with the Haake Viscometer ................. 135 Rheological parameters computed from the Herschel-Bulkley model witlh given yield values for hot aid cooled navy bea1 starch pastes ........................ 136 Flow data of 62 navy bear starch paste cooked at 95°C for 60 min prior to measurements with the Haake Visccneter ................. 142 Flow data of 87 navy bean starch paste cooked at 95°C for 60 min prior to measurements with the Haake Viscameter ................. 143 Flow data of 107 navy beat starch paste cooked at 95°C for 60 min prior to measmements with the Haake Viscometer ................. 144 Flow data of 122 navy bea1 starch paste cooked at 95°C for 60 min prior to measurements with the Haake Viscometer ................. 145 Flow data of 6Znavybea1 starch paste cooked at 95°C for 60 min navyand cooled to 50°C prior to measmexents with the Haake Viscometer ........ 146 Flow data of 87. navy beat starch paste cooked at 95°C for 60 min a1d cooled to 50°C prior to measurements with the Haake Viscareter ........ 147 Flow data of 102 navy beat starch paste cooked at 95°C for 60 min and cooled to 50°C prior to measurements with the Haake Viscareter ........ 148 X 24. 25. 26. 27. 28. 29. Flow data of 127. navy beat starch paste cooked at 95°C for 60 min atd cooled to 50°C prior to measurements with the Haake Viscometer ........ Flow data of 6% navy beat starch paste cooked at 95°C for 60 min, cooled to 50°C, atd held at 50°C for 30 min prior to measurements with the Haake Viscateter ................. Flow data of 87. navy beat starch paste cooked at 95°C for 60 min, cooled to 50°C, atd held at 50°C for 30 min prior to measurements with the Haake Viscometer ................. Analysis of variatce of rheological parareters camputed from selected flow models for navy beat starch pastes of various concentrations cooked at 75°C, 85°C, atd 95°C for 60 min prior to measurements with the Haake Viscometer ...... Correlation coefficients , r2, and sum-of-squares function, SS(b), for regression analysis of the flow curve data .................... Analysis of variatce of rheological parateters camputed from selected flow models for navy beat starch pastes of various concentrations cooked at 75°C, 85°C, atd 95°C for 60 min and cooled to 50°C prior to measurements with the Haake Viscareter ................... 149 150 151 152 153 155 LIST OF FIGURES Figure 10. 11. Flow diagram for the isolation of Great Northern beat starch ................. Flow diagram for the isolation of beat protein concentrate ................. Points of significatce on Brabender amylogram: A, peak (maximum) viscosity; B, viscosity of paste on attaining 95°C; C, viscosity at the end of 95°C cooking; D, viscosity of cooked paste after cooling to 50°C; E, viscosity at the end of 50°C holdJng ....................... Flow curves for model inelastic, hamgeneous, incanpressible, atd time- indepatdent fluids under isothermal conditions ............. Schemtic diagram of the particle- to- particle heat exchanger ............... Flowdiagranfortheproductionofnavybeat flourfractionsusedinStudieslatdZ ....... Flow diagram for the production of navy beat flourfractionsusedinStudy3 ........... Flow diagram of the experimental procedures for the study of gelatinization and rheological properties of ptnified navy beat starch pastes Determination of yield stress for navy beat starch paste with one shearing cycle ......... Elution pattern of HPLC sugar analysis of dry-roasted navy beat flour fractions ........ Schematic illustration for M sensor systen of the Haake Rotovisco Viscometer: A, perforated paddle mixer; B, bob ................. 10 42 47 52 54 56 61 63 68 73 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. HPLC sugar analysis of dry-roasted navy beat flour fractions .................. Pasting characteristics of dry-roasted navy beat whole atd high starch flour fractions atd a corn starch ................. Water asorption isotherms of dry-roasted bean whole flours at room temperature (20°C) . . . . Water asorption isotherm of dry-roasted navy beat hull flours at room temperature (20°C) . . . . Water asorption isotherm of dry-roasted navy beat high protein flours at room temperature (20°C) ....................... Water asorption isotherm of dry-roasted navy beat high starch flours at room temperature (20°C) ....................... Water asorption isotherm of dry-roasted navy beat flour fractions obtained from beats processed at 240°C for 1 min ............ Nbld growth of dry-roasted navy beat flour fractions at various moisture contents during storage ...................... Nitrogat solubility indices of dry-roasted navy beat flour fractions stored under various water activity levels for eight weeks .......... Thermal response curves of 102 navy beat starch pastes cooked at 75°C atd 95°C for up to 60 min at differatt heating rates ............. '1‘herma1responsecurvesof62navybeatstarch pastes cooked at various temperatures for up to 60min ....................... Thermal response curves of 87. navy beat starch pastes cooked at various temperatures for up to 60min ....................... Thermal response curves of 107.’ navy beat starch pastes cooked at various taperatures for up to 60min ....................... Thermal response caves of 122 navy beat starch pastes cooked at various temperatures for up to 60min ....................... xiii 82 86 89 90 91 93 94 95 98 108 109 111 113 115 26. 27. 28. 29. 30. 31. 32. 33. Thermal response curves for 87. navy beat starch pastes cooked at various temperatures for up to 240 min ....................... Arrhenius plot for the gelatinization rate constatts of purified navy beat starch pastes ..... Determination of time independency with constatt shearing at various rotation speeds for 127. navy beat starch pastes cooked at 95°C for 60 min while being mixed using a paddle speed of 50 rpm Flow curves for navy beat starch pastes of various concentrations cooked at 95°C for 60 min ....... Power law plot of the flow data for navy beat starch pastes of various concentrations cooked at 95°C for 60 min ................... Herschel-Bulkley plot of the flow data for navy beat starch pastes of various concentrations cooked at 95°C for 60 min .............. Casson plot of the flow data for navy beat starch pastes of various concentrations cooked at 95°C for 60min ........................ Relationship between the square root of yield stress atd starch concentration for hot atd cooled navy beat starch pastes .................. 117 119 121 124 127 128 130 133 ”UnmmgggUggfl) N ZBBZO'SZF‘ONOW :3 0'6 'LIST OF SYMBOLS Water activity Coefficient of variation Diameter of impellers Dry weight basis Equilibrium moisture content Equilibrium relative humidity Gravitational acceleration Dima'tsional constatt relating force atd mass (1 g.cm/dyne.sec2) Suhnerged height Consistency index (computed from the Power Law or the Herschel-Bulkley model) Consistency index (computed from the Casson model) Casson yield stress Distatce from bottan of cup to bottom of bob Torque Initial torque Maximum torque Rheological constant: n- 1 Mixer Speed Flow behavior index Power required for mixing Exponent empirically related to Re for given set of. shape factors of the mixer Square of the Correlation Coefficient XV Bob radius Cup radius Reynolds umber Sun-of-squares function Time Wet weight basis Density Angular velocity Velocity Time Radial distatce measured from center, or radial coordinate Angular coordinate Axial coordinate Angular velocity Newtonian viscosity Apparent viscosity Shear stress Yield stress through direct measurement Yield stress through curve fitting Shear stress at cup wall Shear stress at bob wall Shear rate Shear rate at the bob INTRODUCTICN Michigan grows about 337. of the colored beats, 707 of the navy beats, atd 307. of the total dry edible beats produced in the United States. Approximately 10,000 farmers in Michigat raise dry edible beats as a cash crop and over half a million acres are planted in this crop each year (Anon. , 1979). These dry edible beats provide signif- icatt levels of protein atd other essential nutrients to the diets of peOple from both developed and subsistatt agriculture. Regardless of the superior nutrition provided by dry beats , their consumption has declined from about 7.6 lb per capita in 1962 to 4.1 lb per capita in 1981. Dry beats traditionally have been prepared by soaking atd cooking in the hate or consumed as commercially processed camed beats. Whole beats require soaking atd cooking to ensure uniform expansion of the seed coat atd hydration of the cotyledon matrix. long cooking times to achieve satisfactory palatability have impeded farther utilization of dry beats. The use of dry edible beans could be readily expatded if they were available in the form of a shelf-stable flour. Beat flours have been prepared by soaking atd cooking beats in water atd steam followed by high temperature dehydration over drum driers (Bakker-Arkata et a1. , 1967). Due to the high carbohydrate and high protein contents of legume seeds , research has oriented to the production of high starch and high protein concentrates using solvent extraction teclmtiques either in pilot or camercial scale. The 1 2 properties atd functionality of these concentrates have been evaluated (Schoch atd Maywald, 1968; Sathe and Salunkhe, 1981c; Chang atd Satterlee, 1979; Vose, 1980). Although this wet processing method results in high yields atd high percentage values of protein atd starch in their concentrates , it is energy intensive, generates large quattities of waste effluatt, atd results in a decline in nutritive value due to leaching. Milling atd fractionation of legures to produce flour products have received increased interest in recent years . Milled products such as whole flour atd air-classified high protein atd high starch fractions from beats have demonstrated high mttrient retention atd provide increased versatility atd improved utilization of beats. A nutber of studies on milling of legume flours atd incorporation of flour into food products have been reported. Field peas have been successfully milled into whole flour atd air-classified into high protein atd high starch concattrates (Vose et a1. , 1976) . Yield atd compositional data of high protein atd high starch concentrates of various legute seeds obtained through air-classification were reported (Patel et al. , 1980; Tyler et al., 1981; Sahasrabudhe et al., 1981). It was noted, however, that legume seeds contain attinutritional factors such as trypsin inhibitors atd lectins which will exert undesirable effects on their nutritive value if they are not inactivated (Puztai, 1967; Jaffe, 1975). In a recent study (Aguilera et al., 1982b), ravy beats were dry roasted in a particle-to—particle heat exchatger prior to milling atd air-classification to maximize the mttritive potential of the Mole, hull, high protein, and high starch fractions. The utilization of bean flour of varying extractions fran untreated atd dry-roasted split navy beats has been investigated for bread making 3 (D'Appolonia, 1978). Functional properties of starch flour prepared from various legutes were studied by Schoch atd Maywald (1968) atd Naivilcul atd D'Appolonia (1979). Rheological properties of dough and baking quality of bread as affected by the addition of navy beat flour atd protein concentrate were studied (Sathe et a1. , 1981b). Cereal brans have been incorporated into bread, cakes , atd cookies to produce baked products with high fiber contents (Pomeranz et al. , 1977; Rajchel et al., 1975; Shafer atd Zabik, 1978; Vratatina atd Zabik, 1978). Research conducted by DeFouw et a1. (1982) concluded that navy beat hulls were at acceptable source of dietary fiber in spice-flavored layer cakes. The physicochanical characteristics of dry-roasted navy beat flour fractions, however, have not yet been thoroughly investigated. Quality change of dry beat flours atd their fractions during storage has been a major concern. Earlier work has sham certain quality changes in dry foods including protein solubility, broming reactions, atd lipid oxidation (Henry and Rm, 1950; Buchatat, 1969a,b; Koch, 1962; Matsushita, 1975). The significatce of water activity to the deterioration of the quality and nutritive value of stored dry skim milk was investigated by Henry atd Rm (1948) and Lea atd White (1948). Counter (1969) evaluated the effect of processing instatt beat powder on the deterio- ration rate of quality parameters such as lipid content, color, protein solubility, atd reducing sugar content. love (1975) studied the effect of storage atmosphere atd water activity on lipid and protein content of navy beat powder under accelerated conditions at 37°C. The stability of pea flour products was evaluated by Sumter et a1. (1979) during a one year period at various moisture atd temperature levels. No information has beat found pertinent to the effect of water activity on the quality hvflwq 4 characteristics of air-classified navy beat flours during storage. High starch flour comprises about 507. or more of the total yield after air-classification, with starch being the major portion of the starch flour. Starch has been used in the food industry as a thickening agent in products such as gravies, sauces, puddings, atd pie fillings to achieve the desired consistency. Starches are also used as diluettts in chemical atd pharmaceutical industries , as carriers in cosmetics and tablets , atd as binders in the paper industry. The rheological measurements of starch or starch containing products are frequently used for quality control atd some engineering applications. It is know that the starch gratules will swell in the presence of water as the temperature of the starch paste is raised to the gelatini- zation range and the granules continue to swell if sufficient water and heat are supplied. At this stage the granules are very susceptible to thermal atd mechanical breakdom, thereby decreasing the paste viscosity. Rheological study of the starch pastes, therefore, becomes very diffi— cult due to the contimms change of the paste property. The gelatinization properties and pasting characteristics of various starches have beat investigated extensively using arpirical strunents such as the Brabender Viscoamylograph ( Sathe and Salunkhe , l981a,b; Paton, 1981; Mazurs et a1. , 1957). Although this instrument is commonly used for testing starch properties, it cat only run at low starch concattration and the viscosities are not given in absolute units due to its canplicated gearetry. IVbreover, it can only provide single- point determinations atd does not measure yield stress. Katpf atd Kalatder (1972) criticized the Viscoamylograph because it did not provide uniform heat tratsfer which resulted in non-uniform homogeni- zation of the starch paste. 5 In contrast, the coaxial cylinder viscometer, such as Haake Rotovisco Viscometer, with a simple atd well-defined geometry, readily provides flow curves atd absolute viscosity units. This type of viscometer has been used recently for the rheological measurements of gelatinized starch pastes (Kampf and Kalender, 1972; Odigboh and lVbhsenin, 1974, 1975; Evats atd Haisman, 1979; Bagley atd Christiatson, 1982). These studies, however, did not provide basic information in the preparation of the starch pastes prior to measurements. The present research was to evaluate some physicochemical properties of the dry roasted and air-classified navy beat flour fractions. Three independent studies were conducted in this work. The first study was to determine the chemical composition and to characterize the ftmctional properties of dry-roasted navy bean flour fractions in order to evaluate the potential suitability of the flours for use in food system. The second study was designed to assess the stability of flour fractions as affected by dry beat roasting treat- matt, water activity, initial moisture content, and temperature during short term atd long term storage periods. The third study was con- ducted to evaluate the effect of cooking temperature and starch concen- tration on the gelatinization atd rheological properties of purified navy beat starch pastes as measured with the Haake RV12 Viscometer. Flow behavior, retrogradation, atd stability of the hot atd cooled pastes were also examined. REVIEW OF LITERATURE Production of Beat Flours Wet Processing Processes for producing instatt, precooked pea beat powder with good flavor atd reconstitution characteristics were investigated by Bakker-Arketa et a1. (1967). Pea beats were soaked at 210°F for 40 min atdcookedfor 90min. Theptn‘eeing of the cookedbeanswasdoneona pulper prior to drying either in the single or double drum drier. A later study by the same authors showed that the physical characteristics of instatt legute powders of pea beats, peas, atd lentils produced in a single drum drier were judged by consuter patels as highly acceptable (Bakker-Arkata et a1. , 1969). Special methods were devised for separating the starch from legumes by Schoch atd Maywald (1968). Mung beats, garbatzos, atd dehulled split yellow peas were washed and steeped overnight under toluate at 50°C atd wet milled in a Waring blender. The magma ms screened through bolting cloth and washed with water. The residual pulp was again ground with water atd rescreened. The carbined starch suspension was then screated through nylon cloth atd allowed to settle. The settled slurry was washed and screened several times atd dried to pass through an 80- mesh sieve. For lentils, lima beans, atd navy beats, seeds were steeped in water, groutd, atd screened as described above. The settled slurry was suspended in 0.22 NaQ-I atd screened through 220-mesh nylon to remove 6 7 fine fiber, followed by flow down at inclined "table" to collect starch particles by gravity precipitation. This step was repeated several times atd the final alkaline deposit of starch was suspended in water, neutralized with HCl atd dried. Another extraction method was developed for separation of wrinkled-seeded peas in the same study by the above authors. Peas were steeped overnight in 0.3% NaOH, washed atd ground with 0.22 NaOH in a Waring blender. The ground magma was screened through cloth atd the suspension settled overnight. The screening atd settling process was repeated several times and the final starch was neutralized, washed, atd ch'ied. These methods produced relatively low yields (ranging fran ZZZ to 447) of starches for various legure seeds. The starch fractions of the germinated atd ungerminated legumes were isolated by Fbrad et a1. (1980). The legume flour was suspended atd stirred in distilled water and the pH adjusted to 8.5 with l N NaOH. The extract was separated from the insoluble starch fraction by centrifugation. The starch was extracted again and washed three times with water followed by freeze-drying. Sathe atd Salunkhe (1981c) also employed the solvent extraction technique to isolate the Great Northern beat starch. Beats were ground to 20 mesh in a Fitz mill atd the flour was extracted sequentially with different solvents to obtain starch with a low yield of 18.232, ' containing less that 17. protein atd fat contents. A schematic diagram for the isolation of Great Northern bean starch is presented in Figure l. Biliaderis et a1. (1979, 1981a,b) prepared purified starches by a wet- milling process to study molecular weight distributions of starches from various legume seeds. The seeds were ground with water in a single disc mill to give a coarse slurry. The slurry was defibered by a screen and was eattrifuged. Further purification was obtained by passing the I new 111m [- 30L 14,0. 24 11, cc 1 [ crurmruce I. 10.000 am, 30min. I , [ seems I- 301. n mo. 24 11. 4t 7 I L csurmruce 10.000 am. 30mm. 1 - Dim unmet-It [ amour 11mm 1.1111 21. 11,0. 01cm 1.1111 0L o.m mow 11 «1111.1 111 a mung I Blonder and extracted for 40 h, C'C [ crurmruce 1.- 10.000 "1113011111.. 1 nectar move 11110 3795 once COLLECT wmre m0 lr'W’ “V" can suacu urea " "' W aesusreuo m 0011. A0. ETHANOL r- BM I mm. m a Waring Blonder i I wanna warts BATH ]- we. 1 11 [ SETTLE ]- “, 4'C.‘ Omani summon!" [ amour 4; 1 F maze DEHYORM’E 1 [ 31m» rowan *] Figure 1 . Flow diagram for the isolation of Great Northern beat starch (Source: Sathe atd Saltmtkhe, 1981c) 9 slurry through a counter-current washing unit atd the starches were then dried. Final purification of these starches was done by repeated washing with 952 ethanol atd screening (44mm) . Purified starches were then dried in a vacuum oven. Proximate atalysis data showed these purified legute starches had very high starch content (>942) with trace amount of protein, lipid, and crude fiber. The prodtction of protein concentrates from dry edible beats has been reported by several researchers (Platt atd T‘ulsiati, 1969; Satterlee et a1. , 1975; Luh et a1. , 1975; Polit and Sgarbieri, 1976). Chatg and Satterlee (1979) prodtced bean protein concentrates containing 722 to 812 protein by extracting beat flours with a dilute salt solution followed by acid precipitation and subsequent centrifugation. Figure 2 shows the isolation procedures to produce the beat protein concentrate. Preparation of protein concentrates from Great Northern beat flour by Sathe et al. (1981b) involved extraction of the flour with 0.52 Na2003 (w/v) at a flour/ solvent ratio of 1/ 10 and followed by centrifugation, dialysis , atd lyophilization. Protein content of the resulting concen- trate was as high as 85.42 on dry weight basis. Isolation procedrres suited for camercial production of both protein isolates and refined starch of field peas atd horsebeans were developed by Vose (1980) . The dehulled seeds were pin-milled atd slurried with 0.03 N NaOH at a solid/ liquid ratio of 1:4 (w/v). The resultatt slurry was screened to rerove the fiber fraction atd centrifuged to recover the crude starch fraction. Starch was further pm'ified using a series of liquid cyclones incorporating a counter- current wash. Purified starches containing nitrogen of less that 0.072 were produced. A 952 starch yield was obtained frcm both field peas 10 BEANFIDUR suspended in 0.22 NaCl solvent:flour = 5 ml:l gran extract 60 min with vigorous stirring centrifuge 30 min 25°C, 1,000 X G T l RESIDUE SUPERNATANT filter through glass wool adjust to pH .0 with 6 N HCl heat (optimal step) cool (only if needed) centrifuge 40 min, 2°C, 10,000 X G f jl BEAN INPASI‘E WHEY fr e-dry BPC Figure 2. Flow diagram for the isolation of bean protein concentrate (source: Chang and Satterlee, 1979) 11 and horsebeans. The protein purification process was developed using ultrafiltration techniques and at approximate overall 822 recovery of seed nitrogen was achieved. Although wet processing resulted in high yields and high percentage values of protein and starch in their concentrates, this method requires significant energy to dry the final products, generates large quattities of effluent which may have high nutritive value , and reduces yields by the loss of wet by-product streams. Dry Milling and Air-Classification In recent years , many researchers have developed techniques in milling atd fractionation to produce flour products fran field peas , horsebeats, faba beans, navy beats, and other legune seeds. Milled products, such as whole flour, hull flour, and air-classified protein and starch fractions fran beats, have possessed selected functional properties with good nutrient retention. A mnber of studies on milling of legune flours and incorporation of flour into food products have been reported. Raw Beats. Chanical studies of starch and protein fractions were conducted by Vose et a1. (1976) . Pin-milling of field pea atd horse- bean seeds, either whole or dehulled, yielded flours that contained two distinct populations of particles. Air-classification techniques were enployed to separate these particles into protein and starch concentrates based on both their size atd density. The protein fraction contained 602 to 702 protein with adequate lysine and low methionine. The starch fraction contained up to 802 of a 322 to 342 amylose starch. It was damnstrated that the use of air-classification to produce these 12 concentrates had relatively low capital requirements without the need for costly effluent disposal operations . In studying the functional characteristics of field pea starch, Vose (1977) applied the same technique to produce a flour containing 782 starch. This flour was then ftmther refined by slurrying, filtering, washing, atd drying to yield a white starch containing less than 0.062 nitrogen. He found that starches from both peas and wheat were susceptible to damage by pin milling, whereas the smaller gratules of corn starch were less susceptible. Because pin milling and air-classification of field peas did not cmpletely separate the protein fran the starch fraction, Reichert and Youngs (1978) initiated a study to elucidate the sources atd chanical characteristics of the residual protein associated with air-classified pea starch. Air-dried peas were dehulled in a plate mill. The dehulled peas were pin-milled and air-classified to obtain crude protein and green starch fractions at protein contents of 54.62 and 8. 12, respectively. Purification of the starch fraction by repeated pin milling atd air-classification reduced its protein content and yield. This is due to the renoval of the protein bodies which attached to starch gratules atd agglanerates. Water washing the starch granules suspended chloroplast menbrane residues atd solubilized most of the reminder of the nitrogen. There were nore acidic and fewer basic amino acids in air-classified starch that in the protein concentrate. Milling of faba beats to produce cotyledon flour was conducted by Watson et a1. (1975); however, a low yield of 75.92 was obtained due to the particular physicochemical properties of these beats. D'Appolonia (1977) modified Watson's method by first rennving sane of the seed coat after the prebreak with a U.S. 710 11 Tyler sieve. 13 Although there was no chemical analysis data reported, sifting after the prebreak might have improved cotyledon flour purity. This bean flour was used for bread-making studies. Morad atd Finney (1980) suggested a laboratory scale milling process, using a Hobart grinder atd a repeated sieving atd grinding procedure, that gives an 882 extraction cotyledon flour with only 22 to 42 hull. The result of this simple extraction process was compatible to that obtained by bad- peeling methodology. The production of navy protein concentrate through air-classifi- cation was conducted by Patel et al. (1980) to assess the inpact of this processing on the amino acidatdmineral distribution in the resulting flour fractions. They found that navy beat flour atd its fractions had a relatively high lysine content atd a low sulfur amino acid content. The protein concentrate with its desirable amino acid aid mineral profile was sham to have potential as a supplement in cereal-based products. The starch fraction may be used as a binding agent in extruded non-cereal products or my be used in non-food applications. Four varieties of white beat or navy beat (Phaseolus vulgaris L.) atd their air-classified fractions were atalyzed for amino acid and proximate ccnposition by Sahasrabudhe et a1. (1981). In this study, cleat swds were dried in at air oven to facilitate dehulling. The cotyledons were separated fran the hulls by air aspiration and pin milled into flours. The pin milled flour was air-classified into fines (high protein) and coarse fractions in a spiral air-classifier. The coarse fraction was reground atd reclassified into high protein and low protein (starch) fractions. The protein enriched fraction (312 to 352 by wt) prepared by air-classification of finely ground dehulled flour 14 contained 50.32 to 57.52 protein with about 3.52 lipid in all varieties. The starch fraction (532 by wt) contained 7.42 to 8.52 protein with only 0.62 lipid. The stachyose content of the protein fraction was, on the average, four times that of the starch fraction accounting for more that 502 of the total. The lipids and the oligosaccharides were concentrated in the protein enriched fraction. The amino acid profile is characteristic of the legume proteins with a high lysine atd low sulfur amino acid content. Tyler et al. (1981) employed similar procedures as described by Sahasrabudhe et a1. (1981) to study the separation efficiatcy, yield, atd carposition of the starch atd protein fractions fran eight grain legunes. After the first air-classification process, the starch fractions contained 58.02 to 76.12 starch atd 7.72 to 20.12 protein, whereas the protein fractions contained 49.32 to 75.12 protein atd 0.02 to 462 starch. The starch fraction, after the second air-classifi- cation process, contained 71.02 to 85.92 starch atd 4.02 to 10.42 protein, atd a second series of protein concattrates contained 38. 02 to 68.22 protein atd 0.42 to 16.62 starch. Differences in protein separation efficiency (PSE) among legunes were significatt, but differ- ences in starch separation efficiency were not. They concluded that on the basis of PSE and the protein contents of the two protein fractions, dung beats atd lentils were the most suitable legumes for air-classification, atd lima beats and cowpeas were the least suitable. Roasted Beats. Legume seeds have relatively high protein content atd account for a significant proportion of the protein in the hunan diet. However, legune seeds contain attinutritional factors which will acert undesirable effects on their nutritive value if they are not inactivated (Puztai, 1967; Jaffe, 1975). 15 . Read atd Haas (1938) were among the first to recognize trypsin inhibitor in plant material. Bowmat (1944) described trypsin inhibitors in navy beats, soybeats, atd corn, atd suggested that a heat-labile protein in the aqueous extract of navy beans and soybeats, which inhibited the in vitro digestion of casein by trypsin, might account for the low nutritive value of raw legumes . Among the many attinutriatts present in platt foods are a group of heat-labile proteins known as lectins or hanagglutinins. ' They are found in many commonly eaten foods (Nachbar atd Oppenheim, 1980) , but are probably most concentrated in legumes . lectins are toxic what ingested in large quantities . Diets containing raw beat meal have beat shown to be responsible for decreased growth and death in rats (Jaffe, 1980). Cases of toxicity have beat reported in htnans who had eaten raw or partially cooked red kidney beans (Noah et al., 1980). Increases in nutritional quality have beat demonstrated by heating soybeats to destroy the heat labile trypsin inhibitory com- pounds. Haragglutinins can also be partially or wholly eliminated by heating soybeats (Liener and Hill, 1953) and other common beats (Honavar et a1. , 1962; Bressati atd Elias, 1974). The nutritional composition of dry navy beats is ideal for delivery of protein atd minerals if beans are properly prepared atd positioned in the diet. Dry beats contain various metabolic inhibitors which must be inactivated or eliminated prior to consumption if maxim-m nutriatt potential is to be derived. Kakade atd Evans (1963) reported that the attitrypsin factors in navy beats could be readily destroyed by autoclaving at 121°C for 5 mtin. Dry roasting procedures will provide a means to eliminate the mtdesirable attimxtritional factors without high atergy atd water input. Gratular 16 bed heat exchangers have beat used in roasting navy beats (CarvaJho et al., 1977). In this study, dry roasting of navy beats in a bed of salt at 190°C for 20 sec or 220°C for 10 sec, resulted in a 702 to 802 reduction in trypsin inhibitor activity. It was noted that what raw legumes are ground without pretreatment, they develop undesirable odors atd flavors which persist after cooking. Lipoxidases have beat considered responsible for the presence of off- flavors by catalyzing formation of hydroperoxides from unsaturated fatty acids (Kon et a1. , 1970). The lipoxidase activity occurs in pulses atd oilseeds as well as in soybeans. Snith atd Circle (1972) reported that dry heat treatment for 6 to 8 min at 104°C to 105°C canpletely inactivated this enzyme. Process condition (i.e. , product temperature) had a significatt effect on the degree of achieved inactivation (Aguilera et a1. , 1982a) . The effects of three variables (beat taperature, bean/bead ratio, atd residence time) at two levels (240°C atd 270°C, 1/10 atd 1/15, 1 min and 2 min, respectively) were tested in eight rtmts. They found that trypsin inhibitors were more heat-resistatt that were hanagglutinins. The antitryptic activity in raw beats was reduced to 922 by the lightest roasting, atd to a minimum of 222 tmtder the harsher conditions. Residual henngglutinin activities, however, were much lower, varying ’- fran 482 to 12 between the extremes of roasting conditions. In the same study, Aguilera et al. (1982a) reported that water soluble nitrogen decreased as product taperature increased due to protein insolubilization at high tanperature. Water holding capacity (MiG) decreased initially with mild roasting atd that increased at high roasting tarperatures. Severe roasting increased WHC by almost 602. Gel - forming capacity decreased with increased degree of roasting. Cold 17 paste viscosities of roasted products were notably higher than those of the raw beats. They also concluded that appreciable changes did not occur in starch gratule integrity during roasting. In atother study by Aguilera et al. (1982b), navy beats were also dry roasted in a particle-to-particle heat archatger before milling. Roasted navy beats were ground to produce whole beat flour (WBF) atd hull flours. Dehulled beat flours were that air-classified to obtain high protein (HPF) atd high starch (HSF) flours. WBF averaged 1.922 fat atd 25.82 protein. HSF atd HPF contained 15.62 atd 43.12 protein content, respectively. It was reported that greater protein shift may be accomplished by finer grinding atd adjustinatt of the cut point. The activity of attinutritional factors, including trypsin inhibitor atd henagglutinins, was reduced in HPF by the roasting treatment. Physicochamical Properties of Beat Flours Carposition of Dry Beats Dry beats contain about 62 to 102 moisture, 252 to 302 proteins, atd up to 602 carbohych'ates. These carbohych'ates include starch, mono- atd oligosaccharides , atd other polysaccharides. Starch constitutes the major portion of legune carbohydrates (Sathe atd Salunkhe, 1981a,b,d; Cerning-Beroard atd Filiatre, 1976; Naivfldtl atd D'Appolonia, 1978). Total sugars , including mono- atd oligosaccharides , represatt only a small percentage of total carbohydrates in dry beat seeds. Oligo- saccharides of the raffinose family predam’nate in most legume flours and account for a significatt part (312 to 762) of the total sugars (Naivilml atd D'Appolonia, 1978; Sathe atd Salunkhe, 1981c; Akpapunam _ 18 atd Markakis, 1979). Legumes also contain appreciable amounts of fiber componatts (1.22 to 13.52 which consists of cellulose, hemicellulose, pectin, and cutin substatces (Labateiah atd Luh, 1981). Uebersax et al. (1983) reported that dietary fiber values in dry-roasted whole flours were 7.882 for navy beats, 6.472 for pinto beats, atd 4.922 for black beats. Henicellulose was the major fiber componatt in these flours. Other camponatts in legumes include fat atd ash. As reported by Naivilcul atd D'Appolonia (1978) from five legumes tested, fat contatt ratged fren 2.582 (mung beat) to 4.932 (navy beat), and ash content ranged fren 2.252 (faba beat) to4.302 (mung beat). Dry-roasted navy beats contained about 22 fat atd 52 ash as reported by Uebersax et a1. (1980). Since starch atd protein are the major constituatts in legumes, their concettrates , either produced fran wet extraction or air-classi— fication, have beat studied extatsively for their chemical, physical, atd functional properties. Starch Flours Two types of polysaccharide molecules are presatt in starch. Both are haloglycats of D—glucose, namely, amylose and anylopectin. The amylose fraction of starch is essattially linear chains of 500 to 2000 D-glucose units interlinked by a-l ,4 glycosidic bonds whereas anylopectin is branched molecules with cat-1,4 atd a-l,6 glycosidic likages with about 5,000 to 20,000 units. Biliaderis et a1. (1979) studied the molecular weight distribution profile for several legume starches. They reported that the major portion of legume starches has molecular weights higher tl'an 2x10‘5 atd over 902 of the starch has a molecular weight above 4x10“ . 19 Native cereal starches consist of about 302 amylose atd 702 amylopectin. Some waxy cereal starches contain only amylopectin. High anylose starch, as bred in maize, may have as web as 802 amylose. In legumes, anylose generally constitutes a significatt portion of the starch, ratging fran102 to 602 (Reddy et a1. , 1984). The aunt of anylose in the starch influences starch solubility, lipid binding, and other functional properties. Amylopectin is also believed to be associated with the solubility of starch gratules. In the cell of platts these starch molecules are orgatized into microscopic packages or' gramles. Starch gratules of beats from different platts have characteristic size and shape atd their diameters may vary from 1 mm to 80 um. Mast beat starch gratules are slender, although spherical, ovoid, elliptical, atd irregular shapes are also found. This variation in gratule size atd shape may be due to genetic control atd seed maturity. The gratules remain essattially intact 1 during modification or processing teclmtiques, such as'mtilling, used for preparation of starch as a food ingrediatt . Plants normally elaborate atd arratge starch in the form of micro- scopically small grains in which a mixture of linear and bratched molecules of starch lie with their long axes radially in concattric rings atd perpendicularly to the grain surface at which they are laid down (Fretch, 1972; Sterling, 1978). Compact and highly ordered crystalline states are presatt in certain parts of each ring. As estimated by Rundle et al. (1944a,b) , the crystalline region of most native starches caIprise 502 to 602 of the total starch. These crystalline regions or micelles are held together by hydrogen-bonding. The amorphous regions which are more loosely packed atd easily accessible to water exist betweat the micelles within each concentric 20 ring and betweat concattric rings. Orientation of molecules to parallel one atother causes the starch gratules to polarize light atd thus. to become birefringent. Chemical composition of several legume starches has beat determined (Schoch atd Maywald, 1968; Naivikul atd D'Appolonia, 1979; Sosulski atd Youngs, 1979; Vose et al. , 1976). Faba beat starches isolated by wet- processing atd by air-classification were studied for their properties by Lineback atd Ke (1975) atd loratz (1979), respectively. Sosulski and Youngs (1979) atd lai atd Varriato-Marston (1979) reported that the Brabetder viscosity pattern of the starches appeared to be determined by the extett of swelling of the starch gratules atd the resistatce of the swollen gratules to dissolution by heat or fragmentation by shear. It was found that leaching of soluble starch from the gratules upon heating may have contributed to the viscosity pattern (Allen et a1. , 1977; Lai atd Varriato-Marston, 1979). Schoch atd Maywald (1968) isolated starches from seven legumes atd studied their properties. Starches from lentil, yellow pea, navy beat, atd garbanzo showed restricted swelling patterns similar to those of cross-bonded starches. These properties were attributed to a relatively high content (302 to 402) of linear fractions in these starches. Lima beat starch, with high linear contatt, showed higher swelling during pasting atd the viscosit;t was similar to that of corn starch. Wrinkled-pea starch which contained 752 linear fraction behaved similarly to high anylose corn starch. Starches of field peas atd horsebeats were isolated by Vose (1980) to study their functionality. He reported that these legume starches contained 322 to 352 anylose atd had restricted swelling characteristics typical of type C starches. Gels were rigid, opaque, atd friable with 21 a firmer texture that that of comparable corn gels. However, it was found that gel shrinkage occurred following storage at 5°C, resulting in 132 to 152 syneresis losses. Naivikul atd D'Appolonia (1979) compared legute starches with wheat starch. The amylogram curves of the legume starches showed higher initial pasting terperatmre atd higher viscosity that did wheat starch, which indicated a higher resistatce to swelling atd rupture. Water-binding capacity of legume starches was generally higher than that of wheat starch. Great Northern beat (Phaseolus vulgaris L.) starch was wet-attracted atd studied for its functional properties by Sathe and Salunkhe (1981c) . They reported that anylose content of the starch was 10.22 (starch basis). The starch had good water atd oil absorption capacities at roan temperature atd formed a stable gel at concattrations of 72 atd above (w/v). This beat starch also ShGflEd a restricted swelling pattern on the viscoamylogram. The same starch was used by Sathe et al. (1981a) to investigate its solubility atd swelling. They found that both properties were temperature atd pH depatdent . Solubility of the starch was highest at pH 6.0. Both acetylation atd oxidation resulted in reduced swelling of starch over a pH ratge of 2 to 10. Addition of fatty acids to the starch reduced the Brabender viscosity and raised the gelatinization temperature. Chemical catposition atd functional characteristics of white bean starch produced by air-classification technique here also studied (Sahasrabudhe et a1. , 1981). The starch fraction (532 by wt) contained at average of 302 amylose atd gave a torque-temperature curve typical of legume starches with slow swelling, no breakdown at constatt terperature atd shear, atd a high set-back. 22 Protein Flours Dry legumes constitute at importatt source of dietary protein for a major portion of the mrld's population atd beats have beat the major legume used for humat consumption. Cereals are deficiatt in lysine, whereas dry beans are rich lysine sources. Although production of lysine-rich flours from legumes may fortify the traditional cereal-based foods, the high carbohydrate atd low protein contatt of these flours may cause adverse effect when incorporating into food products (Patel et a1. , 1977) . Protein concen- trates or isolates fren horsebeats and lima beats with superior functional properties have both found wider food application than have flours (Patel atd Jolmtson, 1975; Mateepun et a1. , 1974). Dry beat proteins are cemplex in nature with the majority of the proteins being globulins atd albumins. Sathe atd Salunkhe (1981b) reported that protein contatt of the Great Northern beats was 26.12. Albumins atd globulins accounted for 21.182 and 73.42 of the total proteins, respectively. NaZCD , K2304, sodium dodecyl sulfate, and Nam at respective concentrations of 0.52, 52, 52 (all w/v), atd 0.02N were found to be better protein solubilizers along several solubilizing agents tested, with the solubilized protein being 207.5 mg per gram beat flour. The electrophoretic characterization of the bean flour, albunins , globulins , protein concattrates , atd protein isolates , produced through wet-attraction, was also conducted in this study. Functional properties of these proteins were further studied by the same workers in atother study (Sathe atd Salunkhe, 1981a). They reported that protein concattrate had the highest water atd oil absorption capacity of 5.93 g/g and 4.12 g/g, respectively. Protein 23 concattrate showed the highest emulsion capacity, while albunitts showed the highest emulsion stability. This investigation also indicated that the albumins were better etulsifiers that globulins in relation to both capacity and stability. Foaming ability of the flour protein studied wasinferiortothat ofeggalbuminpowder. The protein enriched fraction (312 to 352 by wt) prepared by air- classification of ground cotyledons contained 50.32 to 57.52 protein as reported by Sahasrabudhe et a1. (1981) . . Water hydration capacity atd heat thicketing property of the protein fraction were found to be similar to other vegetable proteins indicating a potattial use in processed foods. Storage Stability of Beat Flours In response to the increasing need for dry products in conveniatce form, Pbrris (1961) developed cooked dry beat, pea, atd lattil powders for use in a variety of dry mixes, such as soup atd dip for crackers. With the growing consurer interest in this convatiatce food area atd the declined dry beat consumption per capita in the United States, research has emphasized the production of acceptable convatience forms of dry beats (Rocklatd, 1966; Hoff and Nelson, 1966; Bakker-Arkena et al., 1967, 1968) . The instatt legume powders produced from both drum dehyckation atd spray drying (Bakker-Arkena et a1. , 1967) were reported to possess good flavor, color, atd reconstitution characteristics , although powders fren the drum drier were preferred. W 24 Quality Chatges during Storage Quality chatges , including protein solubility, lipid oxidation, atd browning, of these dry beat powders or flours during storage have been a major concern. Protein Solubility. Melnick atd Oser (1949) indicated that storage of foods over extatded periods of time appears to cause impair- ment of the nutritive value of the protein. These chatges were similar to those caused by excessive heat processing. Most of these chatges did not result fran amino acid destruction, but rather from the formation of resistatt peptide linkages which slow down the rate of etzymatic digestion, thereby disrupting the release of amino acids for absorption. Henry atd Kon (1948) stated that the value of protein is determined not only by the concentration atd digestibility of the protein, but by the availability of its amino acids atd its essential amino acid pattern. Henry atd Kort (1950) argued that it was possible to alter the nutritive value of protein without the destruction of amino acids . They acplained that linkages betweat reducing sugars atd cross-linkages between proteins or peptide chains were formed which contributed to resistatce to proteolytic action. Altered enzyme action produced at array of unabsorbed atypical peptides. Products fren Maillard reaction have beat found as being the cause of decreased protein nutritive value in foods. In the study of the deterioration of dried skim milk diring storage, Henry atd Kon (1948) indicated that the loss in the biological value of milk proteins was through the formation of etzymatically resistatt Maillard reaction products. Melnick and Oser (1949) also concluded that it was the slowed rate of enzymatic hydrolysis which contributed to lowered 25 biological value in stored, browned foods . The amount of lysine released in bromed foods was one-fourth that released from unbrowned prodlcts. McInroy et a1. (1945) also reported that the formation of browning products including atzyme atd acid resistatt linkages contrib- uted to decreased protein value. Oxidation of food lipids has beat reported to be associated with the loss of nutritive value of proteins. Buchanat (1969a,b) fotmtd that lipid oxidation of products resulted in decreased digestibility of leaf protein concentrate due to the formation of carplexes with proteins. Three types of deteriorative processes were reported by Jones atd Gersdorff (1941) in stored wheat flour: decreased protein solubility, partial breakdown of proteins , atd decreased soluble nitrogen digesti- bility. Lipid oxidation was found to be one of the factors to cause the reduced digestibility. Nm-Ehzymatic Browning. Non- etzymatic browning is characterized as a sequatce of reactions without enzymatic catalysis which prodices desirable or undesirable chatges in food syst-s. The carbonyls involved in these reactions are derived either from water soluble reducing sugars or from lipid oxidation. A complete atd thorough review of the browning reactions was made by Hodge atd Osman (1976). Among these (reactions, Maillard reactions are often encountered in foods atd are referred to as a cunplex group of reactions between the carbonyls or reducing sugars atd free amino groups leading to production of brown pigments atd aromatic cempounds. Miller (1956) indicated that protein quality of foods is slightly reduced during drying, due to browning reactions. Patton et al. (1948a,b) also showed in model systems that heating casein at 96.5°C for 24 hours caused loss of amino acids by Maillard browning reaction. Friedman atd Kline (1950) 26 stated that the presatce of a reducing sugar atd a protein held under mild conditions would cause reduction in the biological value of protein. Histidine, threonine, tryptophan, phatylalatine, lysine, atd methionine were ratdered unavailable by the browning reactions . The carbonyl reactatts derived from lipid oxidation may also react with proteins leading to non-etzymatic browning. Jones and Gersdorff (1941) and Tappel (1955) found that these carbonyls arising fren oxidized lipids were able to interact with proteins to form browning products. Amino acids in proteins reacted with carbonyls atd became unavailable. Koch (1962) also reported the involvement of oxidized lipids in browning reactions in freeze dried foods. Dugat atd Rao (1972) further described the role of aldehydes from lipid oxidation in the browning of stored dried foods . Lipid Oxidation. Fats presatt in foods may become ratcid as a consequence of oxidation. This oxidative ratcidity accompatied with undesirable flavors is a major cause of food deterioration. Lea atd Parr (1961) reported that flavor defects arose from the oxidation of lipids . The polyunsaturated fatty acids in glycolipids atd phospholipids developed painty, grassy, atd fishy odors atd flavors. In the oxidation process it is assumed that hydroperoxide groups attach to the carbon atan of unsaturated fatty compounds , atd subsequently the breakdown of hydroperoxides gives a chain reaction of autoxidation termed a free radical chainmechatism. This chain sequatce canbe described by three stages as follows: Initiation RH + R- Propagation R- + 02 + R00- R00- + RH + ROOH + R- Termination ROO- + ROO- + ROOR + 02 R- + R- + R Rm- + R- + ROOR Roubal atd Tappel (1969) studied the tratsient free radicals in a peroxidizing lipid-protein model system. They reported that protein damage was due to the destruction of amino acids and the free radicals were found to be a major cause of this destruction. The stability of at oxidizing polyunsaturated system was related to the hydroperoxide (ROOH) contatt (Privett atd Blank, 1962) . The lower the ROOH level found in the system, the longer the stability could be obtained. The destruction of the amino acid cystine through RCXJH oxidation in autoxidizing lipids in a model system was examined by Rice atd Beuk (1953) atd Lewis atd Wills (1962). In studing the lipid-protein system, Fletcher atd Tappel (1971) indicated that the peroxidizing polyunsaturated lipid reacted with amino acids to give fluorescatt compounds . They theorized that this process involved the reaction of carbonyls (secondary prodlcts of lipid oxidation) with primary amines to produce the Schiff base imine prodtcts. In contrast to their conclusion, Roubal (1971) believed that the process, which resulted in the greatest amount of amino acid loss in an oxidizing lipid-protein system, came from the free radical attack on the proteins . This was later explained by Gamage atd Matsushita (1973) atd Matsushita (1975): both radical atd non-radical prodlcts of peroxidized lipids cat polymerize proteins . 28 A variety of teclmtiques cat be employed to detect the extent of lipid oxidation. Direct methods of atalysis to follow chatges in reactatts include the measurement of oxygen uptake (Karel atd Labuza, 1968) and free radical formation (Roubal, 1971). Measurement of end products, such as fluorescence (Matsushita, 1975) atd thiobarbituric acid reactive substatces (Bidlack atd Tappel, 1973) , is also frequently used. Measurement of the change in Lmtsaturated fatty acid contatt , determined by gas-liquid chromatography (Buttery et al. , 1961a,b) , has beat used to indirectly detect lipid oxidation. Changes in the content of extractable lipid fractions (Lea atd Parr, 1961) is atother indirect method. Factors Affecting Flour Stability Stability studies with cooked lima beat powders were conducted by Burr et a1. (1969) using setsory methods to evaluate flavor chatge. The predict was stored at three terperatures (-23°C, 22°C, atd 38°C), either air- or nitroget-packed, with the initial moisture contents of 42, 52, atd 102 for various periods of time. The most favorable catditions for storage were powder with 42 moisture contatt, nitrogen- packing, 3 ppm BHI‘, atd 22°C temperature. Flavor chatges in this sample were not detected until after eleven months of storage. All air-packs tested were rather utstable. Stored at 22°C, they showed flavor changes after about 2 months, and at 38°C, after 1 month. Shelf-life of powders were extetded by nia‘oget—packing at all moisture levels atd temperatures. This .study also indicated that powder with 42 moisture was slightly more stable in all instatces than were the 52 moisture samples. Powders with 52 moisture were consistettly evaluated to be fresher that those with 102 moisture contatt. 29 Counter (1969) studied the storage stability of atmospheric atd pressure cooked drum dried navy beat powders. Products were sealed in cats and stored at 0°C, 20°C, atd 40°C for at to 120 days. He reported that the maximum shelf life for this powder was betweat 90 atd 120 days without the use of shelf stabilizers. Storage at 0°C increased the shelf life about 30 days over those stored at 40°C. The major changes observed in this study were the oxygat contatt in the headspace atd the lipid atd protein fractions . Results indicated that oxygen contatt decreased significattly with increased storage tauperature atd time. The free fatty acid contatt increased atd nitrogat solubility decreased with storage time; however, no significatt chatges were found in the total reducing sugar contatt or in color. Flours made fram cereals or legures are not processed to greatly reduce their natural microflora. As a result, these products are likely to contain molds, yeasts, atd bacteria which will grow under desirable moisture levels. Frazier atd Westhoff (1978) indicated that mold growth occurs even under low moisture levels, whereas yeasts atd bacteria will grow utder higher moisture conditions. A study was conducted by Bothast et al. (1981) to evaluate the effects of moisture atd terperature on the microbial growth of wheat flour: and corn meal during storage. Samples were stored at moistures ratging fram 112 to ‘182 atd temperatures of 25°C atd 34°C for 18 months. It was reported that, in gateral, bacteria did not grow under the relatively dry conditions used iut this study atd their mnbers decreased with time. lbld counts increased in products stored at 152 atd 182 moisture, reaching maximum mnbers by three to six months, atd that decreased during the remaining storage period. They concluded that product quality catnot be maintained beyond six months at 152 moisture atd 25°C or beyond 15 days at 182 atd 25°C. Quality 30 ‘ deteriorates more rapidly at 34°C that at 25°C at these moisture levels. Perera et al. (1979) studied vitamin B6 stability in atriched white flour atd reported that this vitamin was stable whet stored at room temperature or in a cold roan for six months. The storage stability studies mentioned above have shown quality chatges of products as affected by storage conditions such as temperature, time, initial moisture contatt, atmosphere, atd presatce of attioxidant. In addition to these factors , equilibrium relative humidity (ERH) or water activity (V of the product should also be considered. It has been generally recognized that ERH values or aW are more closely related to food product stability that to moisture content of foods. The definition of aw atd its relationship to the stability of foods was reviewed by Labuza (1968), Rocklatd (1969), and Labuza et al. (1970). Food products stored in various relative humidities will take on or give up water atd reach equilibrium moisture contatt (Eb/C) . The EM: is that plotted against the ERH or a” surrounding the product to obtain the sorption isothatm. The hydration process of the food material can be explained on the sorption isotherm curve with three ratges of aw: monolayer water region (<0.3), multilayer region (0.3 to 0.7), atd capillary condatsation region (>0.7) (Wolf et al. , 1972). Water contatt, what expressed as water activity, has a decisive effect on the stability of low moisture foods. labuza et al. (1970) discussed the dependatce of various deteriorative reactions on aw. Lipid oxidation is foutd most rapid at low aw atd its rate decreases as aW is increased. The rate increases again at evat higher aw Non- etzymatic browming increases with increased aw reaching a maximum and that decreases due to the dilution of the reactatts. These two reactions are the major deteriorative processes often atcountered in low- atd 31 intermediate-moisture foods. Growth of microorgatisms at high aw' s is atother process that limits food stability. The theoretical aspects of the reactions between aW and deteriora- tive process of certain food carponatts in model systems have beat carried out atd sumarized by Labuza et a1. (1970); however, very few studies have discussed the chatge of quality characteristics of low moisture foods in relationship to aw. Burr et a1. (1969) studied the stability of lime beat powder with selected initial moisture contatt; however, only flavor was erployed as a quality index. In the storage study of navy beat powder, Counter (1969) stored two products at two constatt moisture levels atd did not report quality chatge of powders as a function of aw. Wolf et al. (1972) concluded that storage conditions which produced marked changes in the quality of dehydrated apple, pork, atd rice were associated with the chatges of sorption hysteresis. The significatce of aw to the deterioration rate of the quality atd nutritive value of stored food products was evaluated by Hatry atd Kon (1948) atd Lea atd White (1948) for day skim milk. Samples were equilibrated under variaus relative hummity levels to obtain certain moisture contents. Hatry atd Kon (1948) detonstrated that high moisture of 7.62 caused a significatt decrease in protein quality during 37°C storage within twomonths. Samples stored at 32 moisture and 37°C showed little chemical deterioration. Lea atd White (1948) faund that the powder with 7.62 moisture at 37°C discolored most rapidly atd showed the greatest insolubility. They reasoned that the cause of this deterioratiat was the reaction of )Z-NH2 groups of the amino acid lysine, with the reducing sugar lactose, presatt in the product; this process, therefore, decreased powder solubility atd protein quality. 32 They concluded that aW in the powder determined the rate of deteri- oration. Guadagti et al. (1975) conducted a study to determine stability of cannercially prepared pinto beat powder stored in air atd tmtder a nitrogat atmosphere at 50°F, 70°F, atd 90°F. They faund that the developnatt of off-flavor increased with increased storage time up to 100 weeks in both air atd N2 packs. Nitrogat packing resulted in same delay of the off-flavor developmtatt. An exponattial relation was famd betweat powder stability atd temperature in the 50°F to 90°F ratge. love (1975) investigated the effect of storage atmosphere atd aw on lipid atd protein contatt of navy beat powder Lmtder accelerated storage conditions at 37°C. It was found that at a.w of 0.11, more lipid oxidation occurred in the neutral lipids atd phospholipids of air-stored sarples that in those stored in nitrogen. Phospholipids were slightly “protected what the product was stored at a higher aw of 0.23, while the neutral lipids were not protected fram autoxidation. Samples stored at 0.23 aW browned to a greater extent that those stored at 0.11 aw. (This data also showed that the samples which had undergone the greatest amount of non-etzymatic broming also had altered in vitro protein digestibility. Optimum storage conditions were obtained at 42 to 52 moisture contatt, packing in inert gas with attioxidatt, atd stored at temperatures not greater that 20°C. The storage stability of pea flour, pea protein concattrate, atd pea starch was evaluated during a one year period at faur moisture levels ratging fram 3.82 to 13.62 and teuperatures of 7°C, 21°C, and 30°C (Sumner et al., 1979). Total bacteria, colifortm, yeasts, molds, atd psychrotrophs were found to increase initially atd were followed 33 by a steady decline. They reported that the stability of the pea products was depatdatt on the ERH which affected microbial growth. atd atzyme activity, thereby causing deterioration of odor, flavor, and color. An ERH of 602 should provide satisfactory product stability under my conditions; however, at this level the effect of temperature cycling atd packaging materials should be considered. Rheological Properties of Starch Pastes Colloidal Nature of Starch Pastes The air-classified high starch flar constitutes more that 502 of the total yield. Its major ccnponatt is starch. The principle uses of starches are as colloids, pastes, atd gels. Starch, as a thickening agatt, helps to achieve a desired consistatcy in food products such as gravies, sauces, pie fillings, and puddings. Starches are also used as diluatts in cheuical atd pharmaceutical industries , as carriers in cosmetics atd tablets, atd as binders in paper industry. A thorough lutowledge of the physicocheuuical chatges of starch granules during heating atd cooling is essattial for proper understanding of the rheological properties atd viscosity changes of starch suspen- sions atd starch pastes. Gelatinization. The stratgth of micellar network of linear and bratched molecules presatt in the starch granules controls the behavior of the granules in water (Leach, 1965) . Starch granules have limited water sorption capacity atd ability to swell reversibly due to the limited elasticity of their micellar network system. Cold water cat penetrate auorphaus regions of the granule without disturbing the 34 micelles atd swelling camot be detected Lmtless under microscopic examination. What the hydrogat bonds in the starch suspensions are disrupted by sufficiatt heat, the hydration of the network is atcaraged' and at irreversible process of swelling is initiated. The terperature at which this process starts is called the gelatinization tetperattre or gelatinization ratge. At this terperattre, sure of the granules swell tatgattially atd lose their birefringatce. With the initial rapid swelling of starch granules, the clarity of the suspatsion suddenly increases. Further heating causes more loosating of the network, allowing evat more water to alter to atlarge the gratule. The swollat granules, however, are still held by the micellar network to remain whole unless sufficiatt heat or agitation is applied to tear that apart. It was pointed out that the larger the starch granules presatt in the system, the. lower the gelatinization terperattre. The extent of swelling of starch is a function of temperature atd gatetic constitution. Bratched molecules are considered particularly related to the swelling of starch (Waldt and Kehoe, 1959). Attaining the gelatinization ratge with subsequatt swelling of the granules is accompatied by the increase in viscosity. The viscosity is caused not only by the properties of the individual swollat granules caning into contact with each other but by the molecules released fram the granules (Miller et a1. , 1973). The molecules, mainly amylose, which are dispersible in hot water, leach from the swollat granules atd into the surrounding aqueaus phase. What this swelling process reaches the point where the granules occupy the attire voltne, the solubles in the exudate will rediffuse back into the swollat granules. 35 As the temperature of the suspatsion is raised above the gelati- nization ratge, the granules contimue to swell if sufficiatt water is presatt atd become many times their original size. At this stage the granules are very susceptible to thermal or mechatical breakdown due to their highly swollat condition atd increasing weakening or bonding forces within the micellar lattice. Prolonged heating of the paste, particularly in the presatce of shearing , results in the dis integration of the granules, leading to a colloidal system containing swollat granules, granule fragments, atd dispersed starch molecules. Conse- quattly, the viscosity of the paste is reduced. Retrogradation. When the temperature of the paste is cooled, the amylose molecules which have leached fran the gelatinization grains rebond to one atother and to bratches of amylopectin on the outer edge of granules. Thus they Lmtite swollat starch grains iut a network with at increase in viscosity. This recrystallization of gelatinized starch is knowm as retrogradation. Retrogradation is also regarded as a process in the firming of a starch gel. As time goes on, the gel will shrink atd express sate ettrapped water. This is known as syneresis. Retrogradation of amylopectin cat be reversed by heating. However, retrograded anylose camot be recovered by ordinary heating. Factors Affecting Paste Viscosity. Rheological data of the inter- actions of starch with other food ingrediatts- in heating atd cooling should be of benefit in formulation of food products containing starch as a thickating agent. Some of the factors affecting the viscosity of starch pastes in foods include sugars, acids, salts, fats atd surfactatts, starch concattration, heating, atd the methods with which starches are evaluated. 36 In the stuudy of the effect of sucrose on starch properties, Hester et al. (1956) faund that sugar decreased the thickness of cooked products what it was added to puddings atd soft pie fillings. It decreased the stiffness of the cooled product even more. Because of its hydrophilic character, sugar carpetes with starch for water, thereby retarding swelling of starch granules . Higher sugar concentra- tion lowers the rate of swelling atd thus a lower final viscosity is obtained. Viscosity increase at lower sugar concattration is due to - the fact that this sunall amount of sugar delays the fragmattation of the most rapidly swollen gratules without preventing swelling of other granules. Acid, in the form of vinegar or leron juice, is oftat used in starch-thickened foods. Acid reduced the thickness of hot starch paste atd the finntess of cool pastes (Hatsuuld and Briatt, 1954). In the presatce of heat, acid catalyzes the hydrolysis of starch molecules to dextrins. This fragnattation of swollat starch granules which results fram the hydrolysis reduces the thicketing power of the starch. Paton (1981) studied the behavior of oat starch in solutions atd found that high solution torquue developed by cooling the hot paste to 75°C to 80°C was abolished in the presatce of acetic acid at pH 3.0. Effect of a lyotropic ion series on the pasting characteristics of wheat flars was evaluated on amylograms by Medcalf atd Gilles (1966-) . A progressive increase in pasting atd peak teuperattres of starch was noted fram Sou', which is among the most potatt solubilizing ions, to '2, which is along the most potent insolubilizing ions. In a study 304 on the effect of sodium chloride on the pasting of wheat starch granules, Gauz (1965) found that thepeak viscosity was markedly increased in a Brabetder Viscoamylograph what a wheat starch suspatsion was heated in 37 2.52 NaCl solution. Paton (1981) reported that starch in 0.1N salt solution reduced the torque values of cool paste measured by amylo- graph atd Ottawa Starch Viscareter. Osman atd Dix (1960) found that the temperature at maximum viscosity of 62 corn starch paste decreased progressively from 92°C to 82°C what at increased amount of fat was added from 92 to 122. These fats, cam— posed mainly of triglycerides with aliphatic chains of 16 to 18 carbons , however, did not affect the maximum viscosity in a Brabatder amylograph. The same workers found that addition of surface active agatts to the system resulted in large increases in the tauperature at which granules noticeably atlarged atd in the teiperattre at which the maximum viscosity occurred. This result was in agreement with that reported by Sathe et al. (1981a) who stated that addition of fatty acids (palmitic, stearic, atd linoleic) to purified beat starch reduced the Brabatder amylograph viscosity atd raised the gelatinization temperature of starch. The effect of lipid componatts on the viscosity of wheat starch paste was studied by Niihara atd Matsuuoto (1981). They suggested that the presatce of small amount of free fatty acids atd monoglycerides may reduce starch paste gelation. The effect of concattration (22 to 82) on the viscosity of cassava starch paste dring cooking-cooling was conducted by Odigboh atd thsetin (1974) using the Stormer Viscometer. They reported that the rise to peak viscosity was relatively gradual at the lower concattra- tions but quite abrupt at the higher concettrations . The tetperature of the peak viscosity decreased with the increased concattratiat studied. Viscosity dropped more rapidly beyond peaks for higher concentration pastes that for lower ones. A similar stuudy by Odigboh atd Mohsatin (1975) also reported that the terperature of peak viscosity decreased 38 with increasing concentration. Bagley atd Christianson (1982) stated that viscosity increased very rapidly with concattration above 162 for starch suspatsions cooled at 60°C. However, the rapid viscosity increase occurred at lower concattration what cooking temperature was raised to 65°C, 70°C, atd 75°C. The effect of heating on the rheological behavior of starch-water systems is associated with the heating temperature atd duration of heating as well as the method with which they are evaluated. A number of workers have stuudied the starch pastes of various sarces for their rheological behavior in relation to the heating and cooling process. This is to be discussed in the following section- Rheological Property Studies in Food Systems There are a Inumber of reasons for the food industry being interested in the rheological measurenatts of fluid products. Quality Control . Data from these measureth are frequattly needed for quality control of the raw materials, ingredients, atd final processed products. Such data may also provide information about rela-r tionships betweat product structure atd its rheological characteristics . The flow behavior of fluid foods cat also be correlated to their sensory properties. i The consistatcy of a fluid material, for instatce, usually expressed in tons of viscosity, is the prOperty which governs its flow behavior atd is a measure of internal resistance to flow. The starch pastes have the binding and adhesive properties closely related to their viscosity, atd as a result, the viscosity of the paste is usually used to character- ize a starch. I 39 In the food industry, viscosity values are used as indices of the consistatcy of processed starch-containing foods. Parameters, such as time coefficiatt atd shear coefficiatt, related to breakdown of the starch materials, are used in predicting consistatcy loss due to agita- tion of mixing of food ingrediatts (Muller, 1973). Viscosity data are also usefuul in evaluating gelatinized starch paste quality in order to assess the suitability of the starch for use in food systus (Albrecht et al., 1960). figmwliwtims. The rheological data cat also be applied to the atgineering design of contirnuaus processes in the food industry (Holdsworth, 1971). In the tratsport of fluid foods, it is necessary to predict flow rate in pipes of various diameters or to determine required pump sizes to achieve specified flow conditions . Rate of heat tratsfer in relationship to the viscosity of food materials is also important in a process such as dehydration or pasteurization. The prediction of power requirements iut agitation or mixing systems is essattial in mixing fluid foods. The relationship between the viscosity andpowerformixingcatbeobtained framthe empirical charts ofRe: (NDZp/n) against (ch/N3D5p)/(N2D/g)Q (McCabe atd Smith, 1967; Charm, 1963a) . Methods for Rheological Phasurenetts Two main approaches as described by Shermat (1970) have been used to study the flow properties of dispersed systu. One approach is to explain the relationship betweat flow characteristics and product micro- structure based on a structural rheological model. The other approach is to select a matheuatical acpression that best relates the data betweat shear stress atd shear rate without considering the structure or . 40 campositiat of the system. The latter approach permits the researchers to characterize the fluid flow by determining the shear stress over a ratge of shear rates to draw a flow curve. Starch- in-water systems , being recognized as dispersions, have beat stuudied for their rheological properties using this approach (Keupf atd Kalatder, 1972; Odigboh and I‘bhsetin, 1974, 1975; Evans atd Haismat, 1979; Niihara atd Matsunoto, 1981; Bagley atd Christiatson, 1982). A variety of instrtmatts has beat used to measure the rheological properties of starch pastes in laboratories or industries. Sale of these instruments are designed to simulate the processing or the atd use of the starch materials. The data obtained do not provide absolute units. These eupirical test procedres indicate whether the samples subjected in practice will differ in their response to the test conditions. finith (1964) presatted a thorough description of may traditional instru- ments including the Brabatder Viscoamylograph , the Corn Industries Vis— cometer, the Brookfield Viscometer atd the Scott Viscometer. The Viscoamylograph has beet used extatsively in recent years in evaluating the pasting properties of beat starches (Sathe atd Salunkhe, l981a,b; Pbrad et al. , 1980; Sathe et a1. , l981a). It was originally designed-to evaluate the baking quality of rye flors that were used to produce breads. This 1.113th was that applied to study the gelatinization property of wheat atd other cereal flora by U. S . cereal chemists in the 1940's (Anker and Geddes, 1944; Brown atd Harrel, 1944). The Brabetder Viscoamylograph, caubining the original Viscograph atd the Amylograph into one instrumatt, has becote the most comonly used instrumt in laboratories to record the chatge of viscosity which occurs during the pasting, cooking, atd cooling of aqueous starch systems. It is a viscareter consisting of a rotating cup atd bobbins for 41 determining atd recording the apparatt viscosity through torque measurematt at fixed temperatures or temperatures varied at a uti- form rate. For viscosity arve, the starch dispersion is heated from 50°C to 95°C at a rate of 1.5°C/min, held at 95°C for certain time, atd that cooled down to 50°C at the same rate. As suggested by Mazurs et al. (1957) , the starch pasting properties can be characterized from Brabatder data by the five significant points ’ from the curve of the starch used (Figure 3): A. Peak viscosity --- maximum viscosity reached during heating, regardless of the temperature. Comparisat of this point to the point where teuperature starts rising gives the gelatinization ratge. B. Viscosity what paste reaches 95°C --- the relation of this value to the peak viscosity indicates the ease of cooking the starch. C. Viscosity after cooking 30 min or longer at 95°C --- this indicates the stability or breakdown of the paste during cooking. D. Viscosity after cooling paste to 50°C --- a measure of the set-back or retrogradatiot tatdatcy caused by cooling. E. Viscosity after 30 min or longer at 50°C --- the stability of cooked paste as it might be used. The stability of a hot paste is indicated as the difference of the viscosity betweat point A atd point C. Starches with higher swelling _ power which resuult inhigh peaks are more susceptible to breakdom upon cooking and heme adibit significatt viscosity decreases beyond peaks. Higher starch cotcattrations cause more abrupt increases of the peak, lower pasting teuperatures , atd more rapid decreases in viscosity (Mazurs et al., 1957). Sore starches, such as cross-bonded waxy starch, do not shov a peak through this stage. Cross-bonding reduces the 42 .wfiBoe soon mo o8 o5 on bemoan? .m «doom 8 wfiaooo Hon—mm. momma ooxooo mo Deacon? .a awfixSo 0.3 no o8 one no Deacon? .o 6.8 madness to Base no beacon? .m «Deacon? Aggy Mama 3. "gumoaafim nonrandom so 853%? no 350m 0 _.m¢3._.<¢mn_2m._. 1.1.... mm .m opened 08 ‘ALISOOSIA 43 swelling ability of starch atd, hence, prevents the decrease in viscosity due to breakdown of highly swollat granules. During cooling the elenatts presatt in the hot paste tend to associate or retrograde. Therefore, the viscosity difference betweat point C atd point D shows the tatdatcy of retrogradation. Finally, the viscosity of the paste between point D atd point E indicates the stability of the cooked paste in actual use. Viscosity values corresponding to each significatt point may be taken from the curve atd plotted on linear coordinates against the logarithm of sample concattration as discussed by Mazurs et al. (1957) . The amylogram obtained from the Viscoamylograph reveals the viscosity chatges of the starch-water system under a variety of controlled tamerature conditions . These measurements have beat used to derive a large amount of information that is useful to the cereal industry. This is alsooneofthereasons thatthisinstruuenthasbeatso extensively used for routine quality control atd sole basic research applications. Although the Viscoamylograph is comonly used for testing starch pasting properties, it cat only ruun at low starch concattrations atd the viscosities are not given in absolute units due to its complicated geouetry. Moreover, it can only provide single-point determinations of viscosity values and does not measure yield stress. Brabatder viscograma are usually detenm'ned at at arbitrarily fixed starch paste concattration atd the resulting peak viscosity atd other values are used as criteria for starch characterization. A critical review of this method was made by Bhattacharya atd Sowbhagya (1978). They reported that the curratt approach reveals a mnber of fallacies atd suggested that starch characterization may be properly uade not at a fixed paste concattration but at a fixed peak viscosity. 44 In contrast to the type of instrument mentioned above, rotational viscoueters, with a simple and well-defined viscometer geottetry, readily providing flow curves atd absolute viscosity uunits , have become the single most widely used instruments for rheological determinations (Vat Wazer et a1. , 1963) . The most common type of rotational viscometer is the coaxial cylinder viscoteter such as the Haake Rotovisco Viscoueter. The Rotovisco is a very versatile viscometer atd cat measure the apparatt viscosity to at accuracy of 12 to 22. The measuring head of the Rotovisco is activated through a lO-position geared tratsmissiot which is powered by a synchronous motor operating at 3000 rpm. A stress measuring device in themeasuringheadwill tratsmit signals to the patel to give readouts. Shear stress is measured by a precision torsiat spring which is interposed betweat the rotor in the sample fluid atd the tratsmission cable. Thus, the viscous drag on the rotor is measured by the deflection of the torsion spring. This deflection is that measured by a potattiometer atd the signal is recorded on the cattrol patel with a rauge of 0 to 100 units. During measurenent, a rotating body immersed in a liquid is subjected to a viscous drag, the magnitude of which (shear stress) depatds on the speed of rotation (shear rate). The flow curve thus cat be plotted as shear stress vs. shear rate (Vat Wazer et al. , 1963; Sherman, 1970) . The viscosity and shear rate obtained is actually apparatt viscosity atd apparatt rate of shear, respectively, since these terms have beat derived from the Newtonian expression for rate of shear at the wall of the finer cylinder. Therefore , correction should be made to get the absolute values for not- Newtoniat fluids (Vat Wazer et al. , 1963). Due to the large variation of the scale atd measuring system of differatt viscateters in obtaining starch rheological data, Kerpf atd 45 Kalatder (1972) have urged the need for the staudardization of suitable methods for comparisons of the test data from different instruments. The authors compared the precision atd reproducibility of three viscom— eters . In the comparison of the Brabender Viscoamylograph atd the Haake Rotovisco, they concluded that the Rotovisco, with a perforated paddle sensor, permitted exact measurement of rheological properties of starch atd had better reproducibility. They criticized the heater design of the Viscoamylograph, indicating it did not provide uniform heat transfer which resuulted in non-uniform homogenization of the starch paste, whereas the Rotovisco had better temperature control , effective hamgenizing atd less water evaporation. Non-Navtoniat Flow Behavior of Starch Pas tes Flow Models. Fluids following the simple Newtonian rheological model in which a linear relationship exists between shear stress (I) and shear rate (y) cat be expressed as: T = n (t?) (l) where 1.‘ is shear stress (N/mz), y is shear rate (3'1), atd n is viscosity (Pa.s). In other words, Newrtoniat fluids maintain constatt consistatcy irrespective of the velocity chatge. Must of the fluid food products, however, do not follow this model atd show deviation from Newtoniat behavior. They are generally known as not-Navtoniat. Holdsworth (1971) discuussed this flow behavior atd stated that it is often found for food products that the logarithmic plot of shear stress versus shear rate is linear over a wide ratge of shear rate. This . arpirical relationship is known as the power law atd has been 46 used widely to characterize flow behavior of inelastic fluid materials. This equuation cat be written as: r = K m“ <2) where u’ is shear stress (N/mz), y is shear rate (3.1), K is consistency index (Pa.sn), and n is flow behavior index (dimensionless). The consistency index generally increases with increased solid content of fluid viscosity. The flow behavior index is at indication of the departure from Newtonian fluid. men 0 High Starch Flour Beans I Purified Starch Temperature (°C) ‘ 75 85 95 modal] l I tliil li’Il Ij'j 681012 681012681012 Theme! Response Curves “A2335 n J I Initlia| Cooled Cited (50°C) & L (50°C) Held (30 min) t J Yield stress ('y) Flow Curves (r vs 9) Figure 8. Flow diagram of the experimental procedres for the stuudy of gelatinizatiou atd rheological properties of purified navy beat starch pastes 62 shear breakdown (if aty) of the starch paste were recorded to obtain a torque—time curve. Temperature profile of the starch paste dring heating was recorded by inserting a thermocouple into the paste in the preliminary test. It was found that teuperature increased rapidly to the desired cooking teuperature in about five minutes for all three teuperatures (75°C, 85°C, atd 95°C) at various concattratious atd was maintained at that level with :l:0.5°C for the attire cooking period. After 60 min of cooking, the paddle mixer was replaced with the MV 1 bob which had equuilibrated in a hot water bath at that cooking terper- ature. The starch paste, while still maintained at the cooking teuper- ature, had beat left for one minute before the yield stress determination actually started. The rotational speed was programmed to increase linearly from 0 to 1 rpm (corresponding to a shear rate of approximately 2.5 s”) in 60 sec atd to decrease linearly back to 0 rpm in atother 60 sec. The torque values of this shear cycle, recorded as torque-rotatiot speed, were tratsformed into shear stress values. The yield stress (or yield value) was the average of the two intercepts on the shear stress (Figure 9). Flow data were measured after yield stress determination by recording- the torque values with increased rotor speed. Torque values were measured at the manually set speeds from 1 to 100 rpm with 12 increments, each increment requiring 10 sec for measureuent. The computer program in the Hewlett-Packard system was writtat suuch that a five second period was held for reaching steady state conditiou atd atother five secotds were used for taking 10 torque readings before the next higher speed was selected. 63 SHEAR STRESS, N-m 2 I l L l l 0 0.5 1.0 1.5 2.0 2.5 SHEAR RATE, S“ Figure 9. Determination of yield stress for navy beat starch paste with one shearing cycle 64 Tat torquue readings were averaged for each speed incrematt atd the torque—rpm readings were converted to shear stress atd shear rate , respectively, as discussed under the section of the calcuulation of shear stress atd shear rate. For cooled paste measureuents , the sample preparation atd cooking procedres were repeated, atd the paddle mixer was replaced with the MV 1 bob as described previously. The starch pastes were that cooled to 50°C by circulating cool water through the temperature vessel with a predetermined faucet opauing at the paste cooling rate of 1.5°C/min. Yield values atd flow curves of the cooled pastes were determined as discussed earlier. The same procedres atd measurements were performed again on samples except that the starch pastes were cooled atd held for addi- tional 30 mtin prior to the rheological study. This additional holding provided a practical measurement for actatded stability as may be found under productiot system. Chemical atd Physical Analyses of Flor Fractions atd Starch Standard approved methodology was used to evaluate the chemical compositiot atd physical atd fuunctional properties of beat flor fractions atd starch (AACC, 1962). Pbisture Approtiuuately 5 g of beat flor fractious or beat starch were weighed into previously dried atd tared cruucibles atd dried to a constatt weight at 80°C for 24 hr in at air-ovat. Percatt moisture was determined from the weight loss on the fresh weight basis (AACC teem 44-15): 65 2'. Moisture “Di-sure 1°33 (3) x 100 sample fresh wt (g) Ash Dried samples obtained from the above moisture determination were placed in a muffle furnace and incinerated at 525°C for 24 hr. The uniform white ash was cooled in a desiccator atd weighed at room temper- ature. Percatt ash was determrined on the dry weight basis (AACC Method 08-01): _ residue wt (g) 7° ASh ‘ x 100 sample dry wt (g) Protein Approximately 30 mg of beat flor fractions or beat starch were weighed atd aulyzed by the staudard Micro-Kj eldahl procedre. Percatt nitrogat was that multiplied by 6.5 to obtain Z total protein (AACC Method 46-13): Z Total Protein = 7. nitrogat x 6.25 Nitrogat Solubility Index (NSI) Approximately 5 g of beat flour fractions were weighed into a 400 m1 beaker atd thoroughly mixed with 200 ml distilled water. The mixture was stirred at 120 rpun with a mechatical stirrer for 120 min at 30°C. The mixture was that cautrifuuged atd filtered through glass wool to obtain clear liquid. Percatt water- soluble nitroget was determined from 66 the liquid using AACC Method 46-13. Nitrogat solubility index, expressing water soluble nitrogat as a percautage of the total nitrogat, was calculated (AACC Method 46-23): _ 7.’ water- soluble nitrogen NSI - x 100 Z total nitrogat Fat Approximately 2 g of beat flour fractions or beat starch were dried to a constatt weight in a vacuum ovat at 95°C to 100°C prior to place- matt in a Whamman cellulose extraction thinble. The thimble was that placed in a Sodtlet extractor, refluxed with about 300 m1 hexane for 6 hr to reuove flor lipid. A 50 ml aliquot of the filtrate, obtained by filtering the cooled hexaue through a fine glass crucible, was dried to caloulate the crude fat cottaut (AACC Method 30-26): 7. C 1 Fat = fat wt (g) + flask wt (g) x filtrate vol (ml) x 100 sample dry wt (3) x 50 (m1) Enzyme Neutral Detergatt Fiber (ENDF) Beat flor fractious were evaluated for ENDF by the method described by Robertsot atd Vat Soest (1977). This method was modified to include 1 mg of amyloglucos idase to provide additiotal digestion of starch. Suugars Extraction. Sugars were atalyzed by first extracting a l g flour sample with 10 ml of 802 ethatol in at 80°C water bath shaker for 10 min. The mixture was cautrifuged at 2000 rpm for three minuutes atd the super- nataut was collected into a new tuube. The extraction was repeated two 67 more times with the sare procedre except that 5 ml of 802 ethanol were used the second time. It was found, from the preliminary experiments, that the coubination atd order of this 807, ethanol solvatt extraction procedre (i.e., 10 ml, 5 ml, atd 10 ml) resulted in maximum sugar values. The precipitate was saved for starch aualysis. Two ml of 10% lead acetate were added to the extract collected from these three extractions. The mixture was shakat atd cattrifuged (2000 rpm, 3 min) to precipitate protein; the supernatatt was trausferred to atother tuube. Two m1 of 102 oxalic acid were added to the tube atd again the mixture was shakat atd cattrifuged to reuove the residual lead acetate. The final supernatatt was brought to 25 ml volume atd passed through a SEP-PAK cartridge (Waters Associates, Inc.) prior to injection into a High Performatce Liquid Chromatograph. High Performance Liquid Chromatography (HPLC) Analysis . The HPLC consisted of a Solvett Delivery System 6000A, a Universal Chromatograph Injector U6K, Differautial Refractometer R401, atd a Data Mudule Pbdel 730 (Waters Associates, Inc.). Acetonitrile-water (75:25, v/v); pH 4.0) was used as the solvatt at a flow rate of 2.5 ml/min. The Carbohydrate Aralysis (uBONDAPAK) column (Waters Associates, Inc.) was used to resolve the sugars from the aqueous extract. Figure 10 represatts the resolutiot curves for the suugars detected iut flour samples . Quauti- tatiot atd idautificatiot of sugars were done using the external statdard method programmed in the Data Module to calcuulate the absolute amounts of the compoueuts in the injected sample (30ml) atd to obtain the percatt sugar contatt based at dry flor weight . Percatt sugar in the flor sauple was calcuulated according to the following formula: 68 mooszoP 0330 Ba 55 .50.... 30E 50E e 2 52°: 8.: o o “was :31 _ c>> bump—3.8a 95:an .50.“. N. Hmcoauomwm H503 Goon been umummourg mo magma, fiaumofiumm . m 033. 85 increased to 74.4% while the mixing stability changed from 6.0 min to 2.3 min. Whole bear flora had the least effect with a water absorp- tion of 61.32 ard a dough stability of 3.5 min. The effect of substi- tuting the high protein fraction was slightly less detrimental thar that which occrared with navy bear hulls. The increase in water absorption and reduction in dough stability due to the substitution of bear flea generally agreed with data reported by D'Appolonia (1978) and Sathe et a1. (l981b). Pasting curves of whole bean and high starch fleas ard corn starch as calparison are shown in Figrae 13 . The Viscoamylograph indicated that the whole bear flea produced a viscels suspension with controlled swelling. Maximum gelatinization of 550 Brabender Units (B.U.) was not reached until the ambient hot (95°C) stirring portion of the crave. This was apprecimatley 100 B.U. more than the viscosity draing the heating portion of the crave. Upon cooling, the viscosity showed a moderate increase indicating a terdency toward association of starch molecules. Gelatinization of the high starch fraction gave a similar Viscoamylograph crave: however, maximum gelatinization occrared at 630 B.U. Naivikul and D'Appolonia (1979) reported that navy bean starch isolate had a peak height at 530 B.U. which was 100 B.U. lower than that reported in this study. The craves in Figrae 13 show that whole flea and its high starch flea fractions have higher resistarce to swelling than does corn starch. Slraries of two flea fractions were stable during heating and stirring: however, the slrary of corn starch showed disintegration of granules . 86 wk HEATING—hAMBIENT—O‘—COOLING—O 700 High Starch Flour 600 (12%) Whrlo Fgour 12% :5. 5°° J m 3:: 400 Cor?8:r:tch é o 300 _l >. 2 < 200 100 j 0 .__—- I 0 10 20 30 4o 50 60 Time, min. Figrae l3. Pasting characteristics of dry-roasted navy bean whole aid high starch flea fractions ad a corn starch 87 Study 2: Effect of Roasting and Storage Conditions on the Stability of Air-Classified Navy Bean Flea Fractions Proximate Cemposition of Flea Fractions Proximate composition of the air-classified flora fractions obtained frem bears after variels roasting treatments is shown in Table 4. Equilibrium Moistrae Content The effect of roasting conditiers on the equilibrium moistrae content (EMS) of the air-classified navy bear flea fractions in various water activity (aw) levels was determined at roam temperatrae. Figrae 14 illustrates the water asorption isotherms of whole fleas for different roasting treatments in various aW levels after eight weeks of storage. Roasting treatment did not affect the equilibrium moistrae content of the fleas to a significant extent for the two lowest aW levels (0.48 and 0.53): however, FM: of whole fleas decreased with increased roasting time ald temperatrae and the effect of time/ temperatrae treatment was shown to be significant with increased aw, especially with aw's beyond 0.8. I A similar roasting effect was shown for the water asorption iso- therms of bull fleas except that the most severe roasting did not affect the EM: as dramatically as it did whole fleas for the highest aW (Figrae 15). The water asorption isotherme of high protein fleas are shown in Figrae l6. Roasting did not affect EM: of these floras significartly for the lowest aw level (0.48) and similar isotherms were obtained for 88 fractions Table 4. Proximate cemposition of dry-roasted navy bear flea obtained from bears processed at various conditions1 Treatmert - Time,Temp Mai stra e Fat Protem Ash ENDF2 3 (min, °C) (7. wb) (7. db) NSI Whole Flora l 240 8.92 1.82 25.94 5.25 7.58 39.77 270 8.45 2.00 25.88 4.45 7.50 35.02 2 240 8.03 2.09 25.38 4.82 6.89 20.69 270 8.02 1.91 25.50 4.06 8.61 16.67 Hull Flora 1 240 9.03 1.01 18.44 6.32 35.43 27.49 270 8.06 0 49 15.31 6.25 40.18 26.94 2 240 8.55 1.31 17.81 5.79 31.24 25.96 270 7.12 1.94 14.13 6.35 40.43 23.89 High ProteinFlea 1 240 7.05 2.49 41.31 4.72 4.45 43.29 270 6.92 2.70 42.56 4.88 3.74 30.84 2 240 6.48 3.23 47.63 5.11 4.29 28.35 270 6.08 2.67 39.25 4.49 3.38 17.04 High Starch Flea 1 240 8.35 1.02 16.12 2.69 2.30 45.36 270 7.99 1.08 15.50 3.14 3.32 32.66 2 240 7.10 0.88 12.94 3.03 2.49 31.40 270 6.52 0.91 13.81 3.12 2.65. 27.15 1n = 3 2Enzyme neutral detergent fiber (Robertson and Var Soest, 1977) 3Nitrogen solubility index 89 60- ..__. 1 mu... 240°C " 55 *——* 1 min. 270°C o——o 2M240°c _. 50 o-—-—0 2min..270°C .0 .. .. ITI :0 5O 45 g ‘5’ ~40 5%" .2 Whole Flour 5' 8 40' ~35 3 o a “5’ -30 523,: 2;; 30? s '6 ‘25 ‘D 5 8 E J :3 .g 20- 20 g 1% J15 4'; U' o\ “J ~10 3 10 - ~5 o a o 0.4 0.5 0.6 0.70.8 0.9 1.0 WATER ACTIVITY Figrae 14. Water asorption isotherm of dry-roasted navy bear whole fleas at roan telperatrae (20°C) 90 60" I—-—I 1 min, 240°C V " 55 i-———-* 1 min» 270°C o-———-0 2 man. 240°C -r 50 0-—-—O 2 min. 270°C .0 .. - ITI 33 5° 45 s. ...- . a g 40 Hull Flour 40 g. r: " q 3 8 35 g 95’ - 30 a a 30 - s '5 ' 25 o E 9 E L m 20 2. "3. 20 (3 =2 ~15 2: o- o\ LIJ _ E 10 .. 10 U‘ ' 5 O I": L L L L 1 L 1 0 0.4 0.5 0.6 0.7 0.8 0.9 1. WATER ACTIVITY Figrae 15. Water asorption isotherms of dry-roasted navy bear hull fleas at roan tenperatrae (20°C) 91 1 01 01 I———- 1 mln., 240°C #——* 1 mln.,270°C °—'—0 2 min, 240°C .. 50 9"—-“ 2 min, 270°C .o .. .. rn 53’ 5° ~ 45 a. 4‘ . a E High Protein Flour 40 g- 8 4° ' - 35 3 0 €- 95’ - 30 a: *5 3O " / S .5 / .. 25 (D 2 O E . .. 20 8 .g 20 +- / 6 .o . P. I; ./ - 15 - \° 0' . o 1 5 O ’1’: l l A l .' l J l I O 0.4 0.5 0.6 0.7 0.8 0.9 1.0 ' WATER ACTIVITY Figrae 16. Water asorption isoterme of dry-roasted navy bear high protein fleas at roan terperatrae (20°C) 92 floras from roasting treatments of l min/270°C and 2 min/240°C. In gareral, EM.) of high protein floras decreased significantly with increased roasting time and temperatrae over the range of ad in which fleas were stored. Isotherm craves of high starch flours as affected by roasting were similar to those for whole fleas accept that slightly lower EM: values were fend for high starch fleas at higher aW levels (Figure 17) . Cerparisons of isotherms from these figraes demonstrated that high protein floras having the highest protein content (41.37. to 47.6% db) had the highest EM: among all fractions, while hull flora with semewhat lower protein contart (14.17. to 18.4%) had the lowest EM} at all at” levels. To cerpare the capability of water uptake among fractions, the isotherms obtained from roasting treatment of l min at 240°C are presented in Figrae 18. FM: of high protein flea was the highest among all fractions while that of hull flora was the lowest at all aW levels . Sraface Mild Growth Mold growth on the sraface of flea samples was observed draing the storage period. Fleas stored in aw's less thar 0.75 did not show mold grmth throughout the storage period: however, mold mycelia were observed on the fleas held under 0.92 and 0.97 aW after one week. Since the effect of roasting on mold growth was not evident, data were pooled from all fea roasting treatments for each flora fraction to assess the influence of moistrae contart on mold growth of the flea fractions draing storage (Figrae 19) . Generally, moistrae contart of samples beyond 16% db caused significart increases of mold growth for all the flea fractions . High protein flea had the highest water N 0) 01 o o a o l I l I Equilibrium Moisture Content, %db .ul 0 l 93 -.———. 1 min., 240°C ‘ 55 *———* 1 mm, 270°C *——-¢ 2 min. 270°C 1 4:. 01 l .p. 0 High Starch Flour 5 8 a 8 a QMO/o ‘lueruoo einlsgow runuqmnbg l —L O 001 ll Figrae l7 . 7' 0.4 0.5 0.6 0.7 0.8 0.9 1.0 WATER ACTIVITY Water asorptie'r isotherms of dry-roasted navy bean high starch fleas at room terperatrae (20°C) 94 60- -——- HIGH PROTEIN FLOUR ‘ 55 *——* WHOLE FLOUR o—o HIGH STARCH FLOUR m 50 Q -——o HULL FLOUR ‘o 50 - - 45 o\° E“ - 4o 9 40 c .. 8 35 o L... T 30 3 “3;; 30 - '5 4 25 E E .. .3 20 - 2° 25.-1’ - 15 D o L” 10 - -10 J 5 O ’1’; n 1 1 n 1 L l 0 0.4 0.5 0.6 0.7 0.8 0.9 1.0 WATER ACTIVITY Figrae 18. Water asorption isotherms of dry-roasted navy bear flea fractions Obtained fren beans processed at 240°C for 1 min qM% ‘lueluOQ eJnrsgow wnquunba 10 Mold Growth d) 10 .3; Mold Growth .5 O) 95 L 20 40 60 80 100 Moisture Content, %db Mold Growth Mold Growth 10 10 O) 4:. A 20 40 60 so 100 Moisture Content, %db High Protein Flour Moisture Content, %db Figure 19. Mold growth of dry-roasted navy bean flea fractions at variels moistrae contents during storage 96 uptake ard thus resulted in the most mold growth, whereas the high starch flea absorbed the least water and had the least mold growth. EM: of the hull flea after eight weeks was lower thar 40% db: however, there was considerable mold growth on the flea sraface. This may have been due to the certamrination of molds on the seedcoats of bears prior to roasting. Property Charges after Storage Following storage the samples in aW higher thar 0.75 were moldy and thus discarded. Since the roasting treatment of l min at 240°C was demonstrated to Obtain the highest EMS after storage, fleas fren this treatment were selected for color, NSI, ard sugar aralyses. Mean HunterLab color values ard Tukey mean separations for flea fractions stored in 0.48, 0.53,0.64, and 0.75 aW for eight weeks are presented in Table 5. Hunter L values remained unchanged with (increased a” for all the flea fractions: however, aL and bl. values of these fractions increased with increased aw indicating that fleas became more red and more yellow toward higher a» levels. Labuza et al. (1970) stated that non-enzymatic browning increased with increasedawreachingamaximlmardthardecreaseddueto the_ dilution of the reactants . The fact that discoloratier of fleas increased with increased aw in this study could be due to the increased browning reactions in the stored samples at higher aw levels. Nitrogen solubility indices (NSI) were evaluated for fleas stored in increased aw as shown in Figrae 20. NSI values were femd to decrease significartly for whole fleas ald high starch fleas stored beycmd 0.48 aw aid for high protein flea stored beyond 0.53 aw. In general, NSI values of these fractions decreased with increased storage aW accept for M1 flea. NSI values of hull fleas were the least 97 Table 5. HunterLab color values of dry-roasted.navy bean flour fractions stored under various water activity levels fOr eight'weeks1 HunterLab Color Values2 water Activity L aL bL Whole Flour Initial 85.1a —2.4a 9. 8a 0.48 84.3a -l.7ab 11. 4b 0.53 84.8a -2.la 10. Oab 0.64 84.8a -l.4b 11. 6b 0. 75 84.1a -0.2c 13. 8c Hull Flour Initial 77.1a -2.0a 11.9a 0. 48 76. 8a -l.7a 12.3a 0. 53 77. la -1. 8a 11. 8a 0.64 76. 8a -1. 0b 14. 2b 0. 75 76. 3a -0. 3c 15. 6b High Protein.Flour Initial 87. 4a -2.6a 8.1a 0.48 86. 9a -1. 8ab 9.6a 0.53 86. 9a -2. 4a 7.8a 0.64 87. 4a -2. 2a 8.7a 0. 75 86. 5a -0. 7c 11.8b High.Starch.Flour Initial 85. 7a -2.5a 10.2a 0. 48 85. 3a -1.9ab 11.7a 0. 53 85. 7a -2.3a 10.4a 0.64 85. 6a -l.6b 11.0a 0.75 84.2a -0.2c 13.9b lMean‘values (like letters within each column for each flour fraction indicate no significart differences at P<0.05 by Tukey meal separa- tion; n = 3) 2 Standardized to a White tile: ‘L = 95.35, aL = -0.6, bL = 0.4 98 50 . EJControI E548 aw 530.53 aw E054 aw [30.75 aw v....'.. c =======_=_______________=__________________________________________________ C a “omen.wowowowowVuonononouonononononononononouonononononom a ‘ ‘ “““‘{{{{{{{ {‘9“.“‘J a nfigomo‘mfluwowomowowomowow mowowowowowonuw wowowowom wowowowo b ________________________________________________________=_=__=__________=___________________ b a a wmnfimnmnwno. nonwnononononononononmwnonufi a a a “‘.“““““““‘ “ ““ ‘ ‘. b .w 909009009o900009o00009909900009oooouoooo‘oooooooooououooouok ........................... b c ”_________=________________________________________________________________________ C P w m 0 0 meZEQawZHVOm—tz 2 1 I-igh Starch Flow Flow WholeFlou' l-UFlou‘ PimProte'n Nitrogen solubility indices of dry-roasted navy bean flour fractions stored under various water activity levels for eight weeks Figure 20 . 99 annng fractions and did not show any significant change due to increased aw levels. ' Sugar contents of flours were analyzed by HPLC using a carbo- hydrate analysis colum. Mean values of the sugars and Tukey mean separations for flour fractions stored in various aw' s are show in Table 6. In general, glucose decreased significantly, with increased aw for all the fractions. Other sugars retained unchanged or decreased with increasing aw. The reduction of glucose was probably attributed to the increased non-aizymatic bromjng reactions which also cause the increased discoloration of flours at higher aW levels . Lea and White (1948) found that dry skim milk with high moisture content discolored most and showed the greatest insolubility due to the reaction of amino acid with the reducing sugar in the product. They concluded that aw in the powder determined the rate of deterioration. The present study agrees with their findings in that increased browning reactions occurred at higher aw' s with comment decrease of the pro- tein solubility and glucose levels. Starch content was analyzed for flours after storage using the YSI Industrial Analyzer and the mean values and Tukey mean separations are presented in Table 7. Data indicated that significant change of the starch content only occurred at 0.75 aW for fractions except for hull flour where the change occurred at lower water activity. The reduction of starch content at high water activity could be due to the hydrolysis of starch mlecules into glucose units , which were washed off during extraction procedures, leading to low readings in starch analysis. 100 Table 6. HPLC sugar analysis of dry-roasted.navy bean flour fracti stored under various water activity levels for eight weeks Glucose Sucrose Raffinose Stachyose water Activity (Z db) Whole Flour . Initial 1.79c 3.00a 0.32a 2.14bc O 48 1.25b 2.35a 0.38a 2.25c O 53 1.21ab 2.40a 0.24a 2.23c 0 64 1.30b 2.58a 0.32a 1.80a O 75 1.10a 2.62a 0.28a 1.91ab Hull Flour Initial 0.9lc 0.92b 0.29a 1.34a 0 48 0.71b 1.00c 0.28a 1.36a 0 53 0.68b 0.95bc 0.27a 1.33a O 64 0.46a 0.9lb 0.26a 1.38a 0 75 0.50a 0.87a 0.29a 1.32a High.Protein Flour Initial 1.47c 2.95c 0.24b 3.60c 0 48 1.34b 3.00c 0.29c 3.73c 0 53 1.35b 2.32ab 0.28c 3.27b 0 64 1.14a 2.52b 0.20ab 3.10a O 75 1.20a 2.07a 0.17a 3.12a Hugh Starch.Flour Initial 1.10c 1.36a 0.25b 1.44b 0 48 0.89b 1.39a 0.26b 1.49b 0 53 0.81b 1.34a 0.22b 1.45b 0.64 0.93b 1.38a O l9ab 1.43b 0.75 0.72a l 1.23a .36a 0:17a 1Mean'val'ues (like letters within each column for each flour fraction indicate no significant differences at P < 0.05 by Tukey mean separation; n.= 3) 101 Table 7. Starch analysis of dry-roasted navy bean flour fractions 1 stored under various water activity levels for eight weeks Water Activity Flour Fraction Initial 0 . 48 0 . 53 0 . 64 0 . 75 Whole 60 . 46b 59 . 00b 57 . 19b 57 . 48b 47 . 16a Hull 31.85c 29 . 68c 29. 77c 28.35b 27. 20a High Protein 31 . 94b 31 . 04b 31 . 54b 30 . 96b 29. 56a High Starch 67. 72b 67. 54b 67. 32b 66. 95b 64. 30a 1Mean values (like letters within each row indicate no significant differences at P<0.05 by Tukey mean separation; n =3) Long Term Storage Stability In order to further evaluate the stability of flours during long term storage, flour fractions fran the optimum roasting treatment were selected and stored at two temperatures and three preadjusted moisture contauts for up to 24 months. Table 8 shows the mean color values and Tukey mean separations for these flour fractions. In general, the Hunter L, aL, and bL values did not ’ change significantly for flom's stored at 20°C with moisture content up to 92 and at 37.7°C with moisture content of 62. Extended storage time from 12 months to 24 months did not have adverse effects on flour color at the above conditions . Nitrogen solubility and lipid oxidation of the flours were evaluated after 12 and 24 months. Mean NSI and 2-thiobarbituric acid (TBA) values and Tukey mean separations are presented in Table 9. 102 Table 8. HunterLab color values of dry-roasted.navy bean.flour fractions stored at ' ‘moisture contents at 20°C and 37.7°C for up to 24 months HunterLab Color Values2 Treatment Temp,Mbisture L‘ 3L bL (°C, Z db) 12 mo. 24 mo. 12 mo. 24 mo. 12 mo. 24 mo Whole Flour 20 6 85.1b 85.2c -2.4a -2.4a 9.8a 9.9a 9 85.1b 85.1c -2.la -2.0b 10.2a 10.4ab 12 84. 9b 84.5bc -1.8b -l.8ab 11.5c 12.0c 37.7 6 85. 0b 84.8c -2.2a -2.2a 10.0a 10.1a 9 84. 6b 83. 5b -l.8b -l.5c 10. 7b 11.0b 12 83. 2a 82. la -1.5c —l.2c 12. 8d 13.9d Hull Flour 20 6 77. 0b 76.8c -2.0b -l.9b 12.3a 12.2a 9 77. 3b 76.8c -1.9b -l.7b 12.1a 12.2a 12 77. 2b 76.5c -1.6a -1.6b 13.8b 13.9bc 37.7 6 77.2b 76.8c -2.0b -l.5ab 12.5a 13.1c 9 77. Ob 75. 2b -1.5a -l.2a 13.4b 14.2c 12 74. 4a 74. 0a -1.4a -l.0a 14.2c 15.8d High.Protein,Flour 20* 6 87. 5b 87. 4b -2.6c -2.6b 8.1a 8.0a 9 87. 8b 87. 5b -2.5c —2.4b 8.1a 8.2a 12 87. 6b 87. 6b -2.4c -2.4b 8.0a 8.2a 37.7 6 87.6b 87.2b -2.4c -2.2b 8.0a 8.4b 9 87. 2b 87. 5b -2. 0b -1.9ab 8.4b 8.7bc 12 85. 9a 84. 3a -1. 6a -l.4a 8.9c 9.2c 20 6 85.6a 85.5a -2.5a -2.5a 10.3a 10.3a 9 85.7a 85. 4a -2.2ab -2.3ab 10.6a 10.6a 12 85. 4a 85. 2a -2.0b -2.0b 10.6a 10. 8ab 37.7 6 85. 7a 85. 0a -2.2ab -2.0b 10.4a 10. 5a 9 85. 2a 85 1a -2. 0b -l.8bc 11.1b 11.5b 12 84.8a 84.9a -L 8c -1.6c 11.2b 12. 0b 1Mean values (like letters within eachcolum for each flour fraction indicate no significant differences at P < 0.05 by TUkey mean separation; n.= 3) 2 Standardized to a.white tile: 'L = 95.35, aL = -0.6, bL = 0.4 103 Table 9. Nitrogen solubility indices and TBA values of drybroasted navy bean.flour fractions stored at‘v 'moisture contents at 20°C and 37.7°C for up to 24 months Treatment NSI TBA Temp,Mbisture (°C, Z db) 12 mo. 24 mo. 12 mo. 24 mo. Whole Flour 20 6* 39.76c 39.82c 0.52a 0.50a 9 39.05c 39.99c 0.61ab 0. 64b 12 37.25b 36.90b 0.70b 0. 75b 37.7 6 38.20bc 35.29b 1.02c 1. 32c 9 34.96a 32.38a 1.35d 1. 89d 12 34.20a 33.65a 1.4ld 1. 95d Hull Flour 20 6 27.50b 26.80b 0.45a. 0.44a 9 27.25b 26.52b 0.45a 0.47a 12 27 38b 26.75b 0.44a 0.47a 37 7 6 27 10b 26.25b 0.47a 0.43a 9 26 90b 26.51b 0.45a 5 45a 12 22 98a 23.05a 0.48a 0. 43a High.Protein‘Flour 20' 6* 43.25c 43.00c 0.82a 0.82a 9 42.33c 43.25c 0.85a 0. 83a 12 40.62bc 39.32bc 0.8la 5 82a 37 7 6 43.20c 40.25c 1.68b 1. 65b 9 37.52ab 34.84ab 1.80b 1. 86¢ 12 34 35a 31.65a 1.78b 1.95c HighihznxfliFlour 20 6 45.38b 44.63c 0.64a 0.65a 9 44.25b 44.32c 0.68a 0.69a 12 44.10b 44.62c 0. 69a 0.65a 37.7 6 43.65b 40.24b 0. 66a 0.66a 9 38. 75a 36.25a 5 65a 5 66a 12 39. 25a 35.15a 0.69a 0. 69a 1Mean values (like letters within each colurn for each flour fraction indicate no significant differences at P < 0.05 by TUkey mean separation; n.= 3) 104 Significant adverse effects occurred only to the flours stored at 37.7°C with moisture contents of 97. and 122'. after 12 and 24 months. TBA values were determined for the flours to examine the oxidation of fatty acids (Table 9). TBA values greater than 1 will generally be detected by the consumers or taste panelists and, therefore, the value of l was used as a cutoff point. Whole and high protein flours stored at 20°C with three moisture levels for up to 24 months were considered acceptable. After storage of these two flour fractions at 37.7°C, how- ever, TBA values indicated detectable rancidity for all moisture contents after 12 and 24 months. There was no significant adverse effect on TBA values for either hull flour or high starch flour under any condition. This could be due to the low fat content (about 1% ) in these flours. Study 3: Gelatinization and Rheological Properties of Purified Navy Bean Starch Pastes Proximate Composition of Purified Starch Purification of the air-classified high starch flour resulted in a significant increase in starch content. Table 10 shows the proximate composition of the air-classified high starch flour and purified starch. Preparation of Starch Paste Starch granules will swell in the presence of water with sufficient heat supply. To a certain extent, the swollen granules are highly susceptible to thermal and mechanical breakdown due to their swollen and fragile conditions and increasing weakming of bonding forces within the micellar network. The system is govered by two processes which affect the viscosity in opposite ways at this stage of cooking. One 105 Table 10. Proximate composition of air-classified high starch flour and purified starch of navy beans1 Fat Protein Ash ENDF Starch Moisture Starch (Z wb) (7. db) Air-Classified 9 . 79 1 . 47 15 . 51 3 . 63 1 . 76 65 . 43 Purified 7.22 0.11 0.98 0.46 0.51 91.17 D II W process is the continuous swelling of the granules which increases the viscosity; the other process is the disintegration of the already swollen granules which causes viscosity decrease. Prolonged heating of the paste will cause breakdown of the granules and, consequa'ltly, leads to the reduction of viscosity. Rheological measuranent of the paste made at this cooking stage, which exhibits a continuous structural change, may result in unreliable information for its flow behavior. This study emphasized the preparation and gelatinization of starch pastes under specified conditions. The torque response due to swelling and/ or breakdown of starch granules during cooking was monitored. The curves generated from these torque respmses for various treatments were termed "thermal response curves." The perforated paddle mixer was chosen to prepare starch pastes in an MVI cup prior to rheological measurements with an MVl bob. Preliminary arperimentatim was conducted to measure the torque values of pastes during cooking at selected teIperathres and stirring at a range of paddle rotation speeds. Results indicated that 50 rpm was the optimum speed for 106 minimum mechanical breakdom of the pastes and yet provided haroge- neous starch suspmsims without starch precipitation at the lowest temperature (75°C) tested. In geieral , two types of the thermal response curves were found in this test: one being a gradual increase of the torque value which approached an equilibrium torque value and the other being a relatively rapid increase of the torque value to a peak followed by a decline before reaching an equilibrium torque . Preliminary experimentation indicated that paddle speeds lower than 25 rpm resulted in precipitation of granules for 6%, 82, and 107. starch suspensions when cooked at 75°C. On the other hand, the speeds higher than 75 rpm caused dramatic breakdown of the paste bodies at high temperatmes and conceitrations and, therefore, were not desirable for monitoring the gelatinization process . The three speeds , 25 rpm, 50 rpm, and 75 rpm, were them chosei for comparisons. Table 11 represeited final torque values after 60 min shearing at paddle speeds of 25 rpm, 50 rpm, and 75 rpm for 102 starch paste cooked at 75°C and 95°C. At the e1d of shearing at a particular speed, the paddle was switched to two other speeds to obtain two torque responses. Results indicated that pastes stirred at 50 rpm had the highest torque values. Heating rate of the starch has beem reported to affect the final viscosity of starch paste. Therefore, the rapid heating process as exployed in this study was compared with the slow heating which is commonly used in the preparation of starch pastes. Figure 21 shows the thermal response curves for 102 starch paste cooked at 75°C and 95°C with rapid heating (about 5 min to reach cooking temperature) and slow heating (1.5°C/min). For each condition, the differemce of the final 107 torque values demonstrated to be insignificant betweei the two heating processes. Table 11. Final torque values for 107. navy bean starch pastes stirred at \lrarious paddle speeds and cooked at 75°C and 95°C for 60 min -2 Paddle Torque (N-m x 10 ) Rotation o 0 Speed 75 C 95 C (rpm) 25 50 75 2.5 50 75 test Speed (rpm) 25 0.095 0.112 0.120 0.440 0.482 0.510 50 0.124 0.140 0.152 0.495 0.520 0.540 75 0.110 0.126 0.132 0.420 0.480 0.534 1n = 3 Starch Gelatinization The torque values, indicating the resistance of the paste flow to the paddle, were recorded continuously through the HP 85 during cooking. While the well-mixed starch suspensions were stirred with a paddle mixer in the MV cup, starch granules swelled or gelatinized to give increased flow resistance, thereby increasing torque values. A more viscous starch paste was expected to give a higher torque value while a less viscous paste was expected to give a lower torque value. Increased torque value therefore indicated an increase in paste viscosity. Figure 22 illustrates the response curves of 6% starch pastes under various cooking temperatures . The dotted vertical line indicates the 108 1.0)- ooooooooo Slow —' Rapid p m l P as I 0.0.0........... 9500 "‘W TORQUE, N-m x 10‘2 P A 0.2 '- 75°C ...W 0 00000 1 1 ' ‘ 1 O 10 20 30 4O 50 60 70 TIME, minutes Figure 21. Thermal response curves of 107. navy bean starch pastes cooked at 75°C and 95°C for up to 60 min at differeit heating rates 109 1.0 r 6% Starch :3 0.8 - x .13 0.6 L 2 mi 2 o 4 - O 8 '— o.2 . . 95°C 85°C O Q 75°C 0 1o 20 so 40 so so 70 TIME, minutes Figure 22. Thermal response curves of 67. navy bean starch pastes cooked at various temperatures for up to 60 min (vertical dotted line indicates care-up time to specified temperature) 110 time at which the starch pastes reached the respective cooking temper- atures from 25°C. There was virtually no increase of the viscosity dm‘ing heating fram 25°C to 75°C. A slight viscosity increase, although very insignificant, was detected during cooking at 75°C and this process reached an equilibrium after about 40 min of cooking Lnder constant shearing. Heating at 85°C and 95°C, the viscosities of the pastes increased relatively faster as can be seei from the elevated torque values. The viscosities of the pastes continued to increase during cooking at the high temperatures and approached equilibrium after about 40 min of cooking time. These thermal response curves indicated that 75°C cooking temperature was not sufficiently high enough to cause gelatinization or swelling of starch gramiles to a great exteit. At higher temperatures the starch granules swelled more freely to give increased viscosities; however, the cmcmtratim was still too low to cause any visible mechanical breakdown of the granules due to shear. Increasing the starch concentration to 8% resulted in dramatic increases of the viscosity for pastes cooked at the temperatures used (Figure 23). No viscosity change was observed for paste heated from 25°C to 75°C, indicating that starch granules did not gelatinize before this temerature was reached. A smooth viscosity increase was found for pastes heated fram 25°C to 85°C. While maintaining an 85°C cooking temperature , the viscosity continued to increase and reached an equi- librium after about 30 min of cooking. Contrary to this, an abrupt increase of the viscosity was observed for paste heated from 25°C to 95°C. This paste viscosity was fmmd to continue to increase for a few minutes after reaching 95°C and was followed by a gradual decline which eided with a final torque value lower than that of the paste heated at 85°C. The decrease in viscosity at 95°C was due to structln'al break- down of the granules during shearing. 111 1.0 " 8% Starch 2, 0.8. X % 0.6r 2 Lu“ a 85°C 5 95°C '— 75°C 0 1o 20 3'0 43 so 66 §6 TIME, minutes Figure 23. Thermal response curves of 82 navy bean starch pastes cooked at various temperatures for up to 60 min (vertical dotted line indicates care-up time to specified terperature) 112 It is recognized that when the starch suspension is heated above the gelatinization temperature the granules imbibe significant amounts of water, swell to many times their sizes, and impart considerable viscosity to the system. The extent of gelatinization of starch granules cooked at 95°C was expected to be higher than that of the granules cooked at 85°C and, therefore, was expected to give higher viscosity. On the other hand, the less swollen granules at 85°C cooking temperature were much more rigid and less deformable than the highly swollen granules at 95°C. The higher paste viscosity at 85°C could possibly be attributed to these less swollen granules. It was suspected at this point that the increased starch conncentration also played an important role in 'causing lower viscosities at higher tenperatures. The final torque value of the 85°C was not much higher than that of 95°C. This could be due to the fewer granules that swelled under 85°C cooking. Figure 24 shows a smooth but much faster viscosity increase for 102 starch at 75°C which gave a final torque value about three times higher than that of 87.! starch. A significant increase of the viscosity was obtained for 10% starch suspension heated fran 25°C to 85°C. This curve was similar to that of 87. starch cooked at 95°C. Thixotropic breakdown of the system was observed for the highest cooking temperatmre. A peak viscosity, with the torque value of O. 008 N-m, developed before the paste reached 95°C. This peak was followed by a continuous decrease, approaching an equilibrium torque value of 0. 005 N-m during constant shearing. The curve obtained from 95°C indicated that maximum swelling of the granules, which resulted in the highest viscosity, occurred at the temperature before 95°C. At this temperature or higher, the highly swollen granules in the system were fragile and readily deformed and, 1.0 0.8 0.6 0.4 TORQUE, N-m x 10’2 0.2 113 10% Starch I. 85°C ' 95°C - 75°C O 1 O 20 3O 4O 5O 60 70 Figure TIME, minutes 24. Thermal response curves of 102 navy bean starch pastes cooked at various temperatures for up to 60 min (vertical dotted line indicates cane-up time to specified temperature) 114 therefore, were very susceptible to mechanical breakdown, leading to a significant reduction of the viscosity. . The abrupt changes in rheological properties, characteristic of gelatinized starch suspensions at a particular concentration, have been reported in systems such as polymer microgels (Shashoua and Beaman, 1958). Evans and Haismnan (1979) studied the gelatinized starches and showed that significant viscosity effects occur only above a threshold concentration. _ High starch concentration caused deformation of swollen granules at 95°C as can be seen from Figures 23 and 24. This was mainly due to the fact that flow occurs by particles deforming and thereby moving past each other without volume changes. It was believed that the thresh- old concentration as reported by Evans and Haisman (1979) was related to this studyandwas intherangeofabout 87. to 102 forthe two higher terperaUues tested. At 127. starch concentration the final torque value for 75°C was about five times as high as that obtained from 10% concentration (Figure 25) . A possible explanation could be that 122 starch paste cooked at 75°C for 60 mnin had reached a threshold point at which the swollen granules filled the total volune of the system, resulting in sufficient flow resistance to give a high final torque value. Peak viscosities appeared at both 85°C and 95°C cooking; these peaks were followed by significant declines of the torque values due to deformation of the crowded granules at constant flow. It is worthy to note that the excessive swelling of the granules at 95°C cooking, which gave an earlier peak than did those at 85°C, resulted in more rapid and extensive thixotrop'ic breakdown due to greater bond weakening within their structures . 115 2.0 r 12% Starch ‘70 1.6 X o 'E 1.2 852 Z , 0 ”j. 95 C a 0.8 - o o: 75 C O "' 0.4 o g l 1 1 l n o 10 20 30 40 50 60 70 TIME, minutes Figure 25. Thermal response curves of 122 navy bean starch pastes cooked at various temperatures for up to 60 mnin (vertical dotted line indicates come-up time to specified tmerature) 116 Activation Energy in Starch Gelatinization The gelatinization rate of rice and potato starches was studied by Kubota et al. (1979) using rheological methodology. In their study, a capillary tube viscometer was used to measure the flow behavior of heated starch suspensions at temperature ranges of 70°C to 80°C for rice starch and 60°C to 63°C for potato starch. This method appeared to be tedious in calculation of the rate constants for starch gelatin- ization and the preparation procedure of starch pastes remained questionable. The technique developed in the current study afforded a continuous monitoring of the starch gelatinization process during cooking using the perforated paddle mixer at a constant speed of 50 rpm as described in the experimental section. It was judged, from the thermal response curves studied earlier, that 82 starch paste gave smooth increase of curves up to 85°C and was suitable for the gelatinization rate study. Four temperatures (75°C, 80°C, 85°C, and 95°C) were used to cook the starch pastes and the torque responses or torque values (M) were recorded frem zero time, at which pastes reached the desired temperatures, up to 240 mnin (Figure 26) . Torque values increased with increased gelatini- zation times and temperature, confirming that the gelatinization reactions proceeded in the regions covered. Differences between the initial torque value (Mo), at time zero, and the maximum torque value (Mm) indicated the extent of gelatinization which occurred at each temperature. The ungelatinized portionn of the paste at a certain cooking time (t) can, therefore, be expressed as (term/WW). Assuming a first order reaction in the starch gelatin- 117 0.4 . 8% Starch 90°C N 85°C '0 0.3 x E 0.2 80°C Hi '3 O 8 0.1 . '— 75°C 0 l L l I 0 60 120 180 240 TIME, minutes Figure 26. Thermal response curves for 8% navy bean starch pastes cooked at various temperatures for up to 240 min 118 ization (Kubota et al. , 1979; Bakshi and Singh, 1980), the gelatini- zation rate constants at various temperatures could be calculated according to Equation 22 and are presented in Table 12: 1n ”‘1'“ = -kt (22) Table 12. First order reaction kinetics of navy bean starch gelatinization at various cooking temperatures Terperature Reaction Rate Constant (°C) (1/min) r2 75 0 . 0182 0. 985 80 0 . 0255 0 . 987 85 0 . 0324 0 . 992 90 O . 0493 0 . 989 These rate constants are plotted on a senilogaritimnic paper against the reciprocal of the absolute temperature (Figure 27) and a linear relationship was found between these two variables. ' The Arrhenius equation was, therefore, used to calculate the activation energy which was found to be 14.5 Kcal/g.mole. The activation energy for cooking white rice as reported by Suzuki et al. (1976) was 1.9 Kcal/g.mole at 75°C to 100°C. Bakshi and Singh (1980) obtained a value of 18.5 Kcal/g.mole for rough rice at 50°C to 85°C and a value of 24.6 Kcal/g.mole for brom rice at the same 119 Ea=14.5 KcaI/g.mole 0.01 J l . I 1 2.75 2.80 2.85 2.90 1/TA, (°K“) x 10‘3 Figure 27. Arrhenius plot for the gelatinizationn rate constants of purified navy bean starch pastes 120 temperature range. Kubota et a1. (1979) studied the gelatinization rate of rice starches and found the activation energy to be 14 Kcal/ g.mole at 70°C to 85°C. The activation energy for the gelatinization of navy bean starch obtained in this study is within the range as reported by other researchers . Yield Stress of Starch Pastes Rheological measurements were conducted with an MVl sensor (bob and cup) system. Yield stress was first evaluated and followed by a canplete characterization of the flow curve for each starch paste sample. After starch pastes had sheared for 60 mnin during cooking, the system became time independent and isothermal with shearing for a certain period of time at the range of shear rates used. As an example, Figure 28 illustrates the time independency of the cooked and sheared pastes when sheared at constant rates ranging from 1 rpm to 100 rpm which were the shear rates used for measuring pasted rheological properties. Yield stress in a hydrocolloid system, such as starch paste, is cannonly known as the stress to initiate flow. Various methods in measuring yield stress values in hydrocolloidal dispersions and some commercial semi—liquid foods have been used and discussed (Robinson- lang and Rha, 1981; Canovas and Peleg, 1983). However, the experimental determination of this rheological parameter has not yet been satis- factorily investigated. With the use of a speed programmer, allowing a linear increase of the rotor speed fram 0 rpm to 1 rpm, yield stress could be detected at the very first movement of the rotor. Since the yield stress after shearing can be expected to be smaller due to the 121 350 _ 95°C 12% Starch 100 rpm 300 _ "E z 250 .- g... 5° mm (D 150 . E 10 rpm 3:) 100 . 50 _ 1 rpm 0 60 120 180 TIME, seconds Figure 28 . Determination of time independency with constant shearing at various rotation speeds for 127. navy bean starch pastes cooked at 95°C for 60 min while being mixed using a paddle speed of 50 rpm 122 disruption of the paste structure, it was recorded again after the shearing cycle. Results of the preliminary tests have indicated that these two values were identical for most samples. The magnitude of the yield stress (1y) obtained from this direct measurerent for each treat- ment was campared with the yield stress (to and K02) obtained from . curve fitting of the flow data with Herschel-Bulkley and Casson models. This will be discussed after the characterization of the flow curves. Characterization of the Flow Curves For the determination of flow curve , torque values were measured fren the sample at onne shearing cycle of rotor speeds, first in an ascending order and followed by a descending order to account for the thixotropic effect. Preliminary runs indicated that the two torque values from the up and down curves registered at each rotor speed were not significantly different for the samples tested and therefore only the values from the ascending curve were recorded for flow study. The raw and canputed data, including yield stress, for only the starch pastes (62, 82, .102, and 122) cooked at 95°C are presented in Appendix Tables 17 through 26. Appendix Tables 17 through 20 represent the flow data obtained from the starch pastes after cooking at 95°C for 60 min. In order to examnine the retrogradation of starch pastes, samples were cooked at 95°C and cooled to 50°C and the flow data were recorded (Appendix Tables 21 through 24). Further holding of the cooled pastes for 30 mnin at 50°C allowed for the evaluation of paste stability. Appendix Tables 25 and 26 show the flow data for 62 and 82 pastes. Gel formation was found for pastes with 102 and 122 concentrations and therefore, no data were reported. This gel formation was also found for the cooled pastes initially cooked at 85°C in this study. 123 Figure 29 illustrates the typical flow curves obtained frem HP 85 output for each sample concentration cooked at 95°C. The flow data were fitted to power law, Herschel-Bulkley (with and without a given yield stress), and Casson models to obtain the rheological parameters. Rheological Parameters of Starch Pastes Hot Pastes. Mean values and Tukey mean separations for the rheological parareters obtained by curve fitting the experimental data to the selected flow models for starch pastes after cooking at various tenperatures for 60 mnin are presented in Table 13. Statistical analysis of these data are summarized in Appendix Table 27. Significant main effect differences in the values of yield stress (ry, To, and K02), consistency index (K), and flow behavior index (n) were detected for cooking temperature and starch concentration. Due to the significant interactions of the tenperature and concentration, each ’ temperature was analyzed separately with a one-way classification method. For the power law model, K values of the pastes cooked at different temperatures increased significantly with increasing starch concentration indicating that pastes with higher consistency or viscosity could be produced by increasing starch content. The highest viscosities were obtained with the pastes cooked at 85°C with concentrations greater than 82. Significant changes of the n values were found when the starch concentration increased from 102 to 122 at 75°C and from 62 to 82 or higher at 85°C; however, the n values remained constant for pastes cooked at 95°C at different concentrations . These gelatinized starch pastes exhibited pseudoplastic behavior over the range of shear rates tested. 350 - 300 250 200 150 100 SHEAR STRESS, Mm2 124 95°C 1 2% 10% 8% I/ 6% _ / 1 1 1 0 50 1 00 1 50 200 250 SHEAR RATE, S" Figure 29. Flow curves for rnavy bean starch pastes of various concen- trations cooked at 95°C for 60 mnin wfiuuufl mg #85782.» ”.5.ng uoopap Eoum Ar»: $on 33% 83w Sufism % 2:392} awézroe .65 . 5 2338523 6 5695.5 a “0652503 585520 08 33¢ Am u c ugflumuomom dome 5339 an mod v nH um. wooamHoww—Hp unmofifldwwm on 3835 ogouoefimu :08 How EBHOU :08 £53 muouuoH 3:3 wonder and} 125 pwm.o. pmm.nm oom¢.o one.ON pwH.HH onq.o pNo.oN peH.HH quq.o pmm.oN NH omm.o ooH.mH pmmm.o nNN.N oq5.0H momq.o omm.w oon.o onmq.o omq.mH oH nom.o nuc.0H oqu.o nNo.N ncm.H oqu.o ANN.m nmm.H owm¢.o nam.o w mNN.o mNN.q nmqq.o moo.m moo.o mwwq.o mNm.N omm.o oqoq.o mmm.N o no pmm.o pmm.mq nch.o pmH.mm qu.NH oomq.o pmN.oN pmm.¢H mmmq.o pmo.om NH omn.o qu.MN nooq.o oqm.mH coo.q mQHm.o on.0H oom.w mm¢¢.o owo.oH 0H nwm.o ncH.oH oNN<.o nmq.n nmo.N mmHm.o nwo.q nom.m m¢¢¢.o nmm.e m qu.o mom.o oooo.o mqo.o moo.o noqm.o mmm.o moo.o noqN.o mmm.o 0 mm ocm.o oom.m mNmm.o me.m oom.H mHHo.o 050.: ono.N omen.o oom.m NH nqN.o nmm.m ncwo.o owo.N nHH.o .anN.o nNm.H QMN.o noNN.o nNo.H oH mmH.o oNN.o oHNm.o ncm.o moo.o nqNN.o moH.o moo.o nqNN.o ooH.o w mmod mHo.o pHom.o moo.o ooo.o neon.o omo.o moo.o nemm.o mmo.o 0 mm o o o a . .x ~.M .: _M e .c .M e .o _M anN Day commmu mmeHomnHonowuom mmeHomnHonomHom .3mq.uo3om ucoo.gama a m ucmaumoye N. Hougomfiw ovamm ofi nua3.mucmamwsmoma cu H0HH5.¢HE on How 06mm pom .Uomm .oomN um poxooo wuouuomocoo mDOHHm> mo woumoe nous”... noon Erma How mHopoE 33m @8033 Eoum pupae—=00 muoumnmnma HmonOHoonm .mH oHan 126 Yield stress was not detected for hot pastes of 62 and 82 at 75°C or of 82 at 85°C. For the pastes showing yield stress, the values increased significantly with increased concentration at each terper- ature (Table 13) . Pastes cooked at 75°C produced the lowest yield values for all conncentrations. The pastes cooked at 85°C were found to have significantly higher Ty and K values than those cooked at 95°C for starch concentrations of 102 and 122. When fitting the flow data to the Herschel-Bulkley model with the given yield stress, K and n values were obtained and found to have a simnilar trend as those obtained fram the power law model. Higher K and lower n values were obtained fren the power law model than those from this model. Duran and Costell (1982) studied the rheology of apricot puree and found the same differences in these two flow parameters. They reasoned that the energy used to destroy the network structure of the plastic material having a yield stress was included as part of the shear response of the material when power law was used. When fitting both the power law and the Herschel-Bulkley equations to the experimental data, high r2 values were obtained as shown in Appendix Table 28, although the former had higher r2 values than the latter for all the treatments. The good fit of these two models was further confirmed by plotting the data on the log 1 versus log n? and log (1: any) versus log 1 as illustrated in Figures 30 and 31. Although the linear regression curve for the power law plot seems to be a better fit than that of the Herschel-Bulkley plot, which resulted in higher r2 values, from a rheological point of viev, the latter is a more appropriate model for characterization of the starch paste especially when the sample exhibits a yield stress. However, it is still valuable to obtain the rheological parameters using the power law model due to its simplicity. 127 103_ ; 95°C " 12% N b 10% E \ 2 Z 10 : 8% (005' : LIJ : .6% E 1.. - o (D - m 0 .fi 1 10 - I : ‘0 r : o 1 111111111 141111111 nnnnnnnn 1 10‘ 102 103 SHEAR RATE, S“ Figure 30. Power lawplot of the flow data for navy bean starch pastes of various concentrations cooked at 95°C for 60 mnin 128 103 L. I 95°C .. 12% N " 10% E z 102 - 8% 3’; 6% LLI 0‘: l- CD a: as 101 a: E w t 1 l lnlnllnL l llnlnlnn 1 llnll 1 10‘ 102 y 103 SHEAR RATE, S" Figure 31. Herschel-Bulkley plot of the flow data for navy bean starch pastes of various concentrations cooked at 95°C for 60 min 129 The application of the Herschel-Bulkley model using non-linear regression analysis to obtain the three parameters, To, K, and n, was studied. The yield stress and consistency index generally increased with increasing concentration (Table 13) . This method gave satisfactory estimations for these parameters when compared to those obtained from the previous method with the exception of those for 85°C at 82 and 95°C at 102. Casson model was designed to characterize the flow belnavior of biphasic disperse system in which the interacting particles are dispersed in a Newtonian medium (Casson, 1959). This model has been used to describe the flow behavior of tenato concentrates (Charm, 1963b; Rao et al., 1981), chocolate (Chevalley, 1975), fruit pnrees (Duran and Costell, 1982; Vitali and Rao, 1982), and same commercial semi- liquid foods (Canovas and Peleg, 1983) . Its applicability to navy bean starch pastes was Studied to evaluate the magnitude of the consistency index (KC) and the Casson yield stress (K02) for ccnparison with the yield stress fram direct measnrelent. High r2 Casson model equation to the experimental data for starch pastes of values (r2 > 0.9676) were obtained when fitting the four concentrations cooked at 95°C (Figure 32). The r2 values for the samples tested, in fact, had magnitudes in the range of 0.8472 to 0.9826 throughout the experiment as shown in Appendix Table 28. Despite the high correlation coefficients , the deviation of the exper- imental points in the lower shear rate range and the fact that the Casson yield values were considerably greater than the measnred yield values for most of the treatments (Table 13) indicated that the application of this model was not quite satisfactory. 130 58 oo now Demo on omxooo mcoHumuuGooooo among. mo moumme £33m Goon b6: How memo 33m 05 mo uoHnH cameo «.6913 . we we 0.. up NF or w 0 ¢ N O . . . _ . . . . O 1 N F 1 (D F ON .Nm 95mg 9'0 (zw/N) ‘ 9'0 J" 131 Cooled Pastes. The flow data of cooled pastes were analyzed as discussed for the hot pastes. Mean values and Tukey mean separations for the rheological paraneters of the cooled starch pastes at 50°C are presented in Table 14. Statistical analysis of these data are outlined in Appendix Table 29. Analysis of variance indicated that the main effects and their interactions were highly significant for all the parameters listed. One-way classification analysis was used to examnine the effect of concentration for each tetperatnre due to the significant interaction. Simnilar to discussions of hot pastes, comparisons among the flow models, used for obtaining the rheological parameters, showed that the Herschel-Bulkley model with given yield values appeared to be the most appropriate one to characterize flow behavior of the cooled starch pastes. No yield stress was measured for cooled pastes of 62 and 82 at 75°C or at 82 at 85°C. Pastes exhibiting yield stress slnowed yield values to increase significantly with increased concentration for each cooking temperature. The K values increased significantly with increased concentration. Cooled pastes with concentrations of 102 and 122 cooked at 85°C .were shown to give significantly higher Ty and K values than those cooked at 95°C. The n values decreased significantly as the concentrations increased frcm 102 to 122 at 75°C and from 62 to 82 or higher at 85°C, whereas at 95°C this parameter did not change with concentration (Table 14) . It is interesting to note that there was a linear relationship between the concentration and the square root of yield stress for the hot and the cooled pastes (Figure 33) . This type of relationship has meHeeHu henna hmeeHH.eeze non—8539: uooHHp Scum H55 mmouum 33> 85w fiHzm a m oH~E\m.zvoM 2~E>6~o¥ be . a. mHmeoHaoH$55.8v a 253:?va "muoumfimhmo HHonwOHoaHH.H How muMEnN Am 1 : Eofimuoomm gm: amvHPH. an mod v m on 85.8mm? £8353me oz $8.35 phoneme—Bu sumo Home SSHoo Homo Hague; mumuuoH ovHHHV 839/ E82 132 H ome.o eee.oe emme.o eHe.em eme.- mmoe.o oem.~m wom.- «Noe.o use.ee NH nee.o ewe.em wees.o owm.mH omm.e mace.o hmm.~H oem.NH eHe; oee.m~ oH nHm.o emm.HN meme.o nom.m eHm.e mmee.o eHm.NH hem.e «Noe.o eme.mH m mam.o eeo.e meme.o eme.e «Ne.o meme.o mme\e mHod «Nee.o mmw.e e on\me 05m.H www.me emoe.o emo.~o eHe.~N moom.o eon.mm www.mm mome.o oHe.mH NH hem.o 0mm.ee amoe.o omn.o~ oom.oH meme.o uem.o~ oee.mH mome.o emm.Hm oH eme.o hem.eH mmHe.o hee.OH em~.m meHm.o emm.e eme.m enme.o eSHeH m mmN.o eme.H emmm.o mHm.H moo.o emoo.o mmm.o moo.o heoe.o mm5.o e om\mm omm.o emo.eH whee.o omm.e ome.e eeHm.o eHm.m uem.e meme.o nee.HH NH emm.o mwo.H eeoe.o hee.o nH~.o neNH.o noe.o nHm.o emNH.o ema.o oH m2.0 ee~.o emNH.o em~.o moo.o heme.c mo~.o moo.o neme.o mo~.o m moH.o meo.o eHe~.o eoH.o moo.o eoee.o emo.o moo.o eoe~.o mwo.o e on\me a ex use a. .M oe a. .M e e. .M wmww. emmw scammu ionHdmnHonomuom maoHvHHSmnHoHHomhom 3m.H Hoaom unmanmoum. N HHQUQBUmHNV mg 0.3“ fig mugg OH 5H5 0.8 on 858 e555: 8 5e come an 0.3 .92 05 8x86 2.3056256 not? uHo moumwe summon ammo team: How 3305 onm oouooHom scum 8.5980 3385.85 HoonoHoonm .SH 033. 133 - 85°C-50°C 5” - o 95°C-50°C , g A 85°C g; 4.0 r ' 95 C 2 . n. 3.0 . 2.0 _ . 1.0 r . o L l l l L oHr 6 8 1o 12 CONCENTRATION,% Figure 33. Relationship between the square root of yield stress and starch concentration for hot and cooled navy bean starch pastes 134 been reported for some colloidal suspensions (Firth and Hunter, 1976). Evans and Haisman (1979) used this relationship for various starches to find the concentrations for the initial deve10pment of yield stress. Extrapolation of these lines to the intercept in Figure 33 suggests that yield stress can be developed when cooking starch paste at 85°C or 95°C with the concentration about 42 to 57.. To finrther evaluate the stability of starch pastes, the pastes at 50°C_were held in the cup at the same temperature for an additional 30 min prior to flow measurements. Table 15 represents the mean values and statistical analysis for the rheological data of the cooled pastes. Pastes of 102 and 12% concentrations cooked at 85°C and 95°C were found to form gels after holding and thus flow measurements were not attained. The flow behavior of the paste at 75°C changed significantly when the concentration was found to have pronOunced effect on Ty and K values even at low levels. . In order to better show the effect of cooling and holding on the rheological properties of the cooked pastes , the measured yield values and the parameters obtained fran the Herschel-Bulkley model are drawn from Tables 13, 14, and 15 and are presented in Table 16. Pastes with 10% concentration or lower, cooked at 75°C, or 62 concentration, cooked at 95°C, remained stable after cooking and holding. Strong retrograda- tion tendency, however, was shown for the starch pastes with concen- trations of 122, cooked at 75°C, and 82, 10%, and 12%, cooked at 85°C and 95°C, as evidenced by the significant increase of the yield values with concomitant increase of the paste consistency. In the gelatinization process amylose molecules leach from the swollen granules and enter the smamding aqueous phase. If this welling process reaches the point where the granules occupy the entire 135 coflmshom How on. 96 mag 33m 5330 8 oHnmoDm wfiuuwm @350 58578? uerHHmmmE pooh? Eouw mmouum 33..» 55m 55m NA floréméoe xwézrox .op . P .HmmoHcoHamfiB a :Nécméz "3898me HSHonomfi How SHEN Am u : Eofimumamm 888 >339 .3 mod v m um 856mm? unmofimwcmwm oz 38on 053mg“. Soon you EH58 zoom 55? 9353 33 mmadg 5.qu --- --- --- --- --- --- --- --- --- --- NH --- - - - - - - - - - - - --- OH .5: nHNN...N 335 no: 83 2.35 Anna .5: new: And: N mHN.o qu.o naea.o mmm.¢ mmN.o mmmq.o mmH.a mmo.o mmmq.o mmm.q o om\om\mm --- --- --- --- --- --- --- --- --- --- NH - - - - - - --- --- - - --- --- - - --- oH noq.o nom.aH mqu.o nqo.N nmN.m mmmn.o nNH.e nHm.o «Haa.o an.OH N mom.o mON.H comm.o m8H moo.o nHHo.o mNHH moo.o nHHo.o mNHH o om\om\nm Home .o com .3 flow .o oqo .oH oom .q mmmmo go K 03 . m ommq .o 32 . 3 NH omm .o Hwmo . H Ammo .o £0 .0 now .o poms .o mom .o 3mm .o noun .o ammo 3 32 .o mmm .o 0o? . o wNN . o moo . o no: . o mHN .o Hwoo .o 8.35 .o mom . o m m: .o moo .o ommn .o m3 .o moo .o ohms .o moo .o moo .o 0.35 .o moo .o o om\om\mm a . . om Nos a M op a M P a M fiéwmw :8 £8 «%oHv—Hnmuaoaomumm mzmaxaauaonomuom 3.3 Hogm Ham—Boon? . Houmfioomflw moo—mm: map 5.3 3553me: 8 MOVE SB om How Doom um. Eon van doom 8 omamow .38 oo Mom Demo can .003 down um. 8x000 gfiumuuaooooo @3039, “Ho mmummd noumuw Goon ban you flown: 30.3 @3033 Boy.“ @3380 muoumamnoa HmflmoHoonm .ma 3an 136 Fun om pom 3m: com Doom 8 ooHooo 983 moumwo Houmm gamma—Ln coon ou omHooo oum3.moumma Houmm omwommm2¢ onsumuooawu woflxooo um nonommmzw m AmmoHEHmdeHB a 2N5\om.z§ KNEE u. "muoumEmHmo HononooHHmu .Hom muHHSN Am u s “833893 some .9339 .3 mod v m um moomummwflo unmonwome on 3835 mHBmHaHBu :08 wow EbHoo zoom 55.3 muouumH mVHHHv 83m.» 3 H nu: mooq.o mHoq.o nu: oqo.Nm oNo.oN nu: oom.NN qu.HH NH In: mmoq.o momq.o nun AMN.NH oow.w nu: oem.NH oon.o oH quq.o mmq¢.o mmwc.o nmm.mH nHm.NH nNN.m now.q an.q Amm.H w mnmq.o mumq.o mmw¢.o omH.¢ mm¢.q mNn.N mmo.o mHo.o mom.o 0 mm In: moom.o mooc.o nu- oom.mm nmm.oN nu: woo.mN won.qH NH nu: omoq.o mQHm.o nun oqm.oN on.oH In: ooq.mH oom.m oH mmmm.o HWono omHn.o nNH.o nmm.o nmo.q nHm.o nmm.m nom.m w nHHo.o nooo.o noqm.o mmH.H mmo.o mmm.o moo.o moo.o moo.o 0 mm mNNm.o mmHm.o mHHo.o nqo.m ohm.m omo.q oHN.m ooo.q omo.N NH nomn.o ncNm.o nmNN.o mmm.o noo.o nNm.H nmm.o nHm.o nmN.o oH noqN.o nomn.o nqNN.o wHN.o mON.o moH.o moo.o moo.o moo.o m nmnm.o noom.o nqom.o moo.o moo.o mmo.o moo.o woo.o moo.o o mu oHom omHooo HmHuHoH oHom omHooo HmHuHoH moHom_ comHooo MHmHuHoH AnoN .Uov oooo.gaflw .o .M HNH unmaumoyfi N.Hmmummo gunman :won.>>mo.ooHooo now no; How mmsHm> 33% 83m :33 HmooE %oHuHHdmuHoHHomuom m5 Scum Boon—coo mumumnmnmm HMOHonoonm .oH mHAMB 137 volume, the amylase in the exudate will rediffuse into the swollen granules. Upon cooling, these molecules rebond to one another and to branches of amylopectin to establish a network in the system. As stated earlier, in the cooking process of the navy bean starch pastes, 122 for 75°C and 8% to 107. for 85°C and 95°C were thought to be the concentrations at which the flocculated swollen granules filled the entire volume with the rediffused amylose in the granules. As the pastes were cooled these molecules which were already present in the entire system caused the rapid and effective retrogradation process , leading to the significant increase of the paste yield stress and consistency values. For most of the treatments , the flow behavior indices tend to decrease upon cooling of the pastes. This probably resulted from the rebonding and reentanglemmt of the molecules due to retrogradation process. It was reported by Odigboh and thsenin (1975) that the greater the percentage of intact partially swollen granules in the system, the stronger the network of the paste structure, due to amylose binding with these grarmles in cold cassava starch paste. The fact that the yield stress and consistency index values were higher at 85°C than those at 95°C for the 102 and 122 cooled pastes (Table 16) further suggested, that uore granule disintegration occurred at 95°C than at 85°C at higher concentrations during cooking and shearing. Pastes with concentrations 102 or higher, cooked at 85°C and 95°C, were sham to produce unstable cooled pastes at 50°C (Table 16). The excessive anount of amylose available in the systan could lead to gradual precipitation, resulting in gel formation during the holding period. SUD/MARY AND (INCLUSIONS Physicochemical characterization of the air-classified navy bean flomts indicated that significant differences were shown for chemical composition of all flom' fractions. HPLC sugar analysis showed that high protein flour contained the highest level of stachyose which was the major oligosaccharide for all fractions . Surface reflectance color measurement of bean flour indicated that cotyledon fractims (protein, starch) were lighter than hull flour. Farinograms showed that wheat flour substituted with 107.2 bean fractions resulted in increased water absorption and decreased mixing performance. Peak development time among fractions was similar. Mixing stability of hull flour was significantly lower than for the other fractions . Pasting character- istics of whole flour and high starch flours were demonstrated to be controlled swelling at high temperature. Slurries remained stable during high temperature holding, and upon cooling, showed moderate increase in viscosity. Dry-roasted navy bean flour fractions appeared to be suitable ingredients for use in appropriate food systems. In the storage stability study, equilibrium moisture content (EM?) of flour fractions increased with increased water activity (aw) from 0.48 to 0.97. EM: of high protein flour was the highest among all fractions while that of hull flour was the lowest at all aW levels. Moisture uptake of flour fractions decreased with increased roasting time and temperatm'e and this effect was shown to be significant with increased aw, especially at high aw levels. Flours stored in aw's less 138 139 than 0.75 did not show mold growth throughout the storage period; how- ever, mold mycelia were observed on the flour held under 0.92 and 0.97 a“; after one week. Following two months storage, flours in aw higher than 0.75 were moldy and discarded and the remaining samples from the optimum roasting process were further evaluated for quality change. Hmterlab color values showed that although L values remained mehanged with increased aw for all fractions, aL and bL values of these fractions increased with increased aw indicating that flours became more red and more yellow toward higher aw storage. Nitrogen solubility index (NSI) values decreased with increased storage aW for all fractions except hull flour which had the lowest NSI values among fractions. Glucose level in the flours decreased significantly with increased aw for all fractions. Significant reduction of starch content was found at 0.75 %7 for all fractions except hull flour. It was believed that increased a” could result in increased brovming reactions causing discoloration of flours with concurrent decrease of protein solubility and glucose levels . It was concluded that flom:s of all fractions stored at less than 0.75 aW for two months at ambient temperature resulted in stable quality and suitability for food use. Long term storage study indicated that flours stored up to 24 months with 62 and 9x moisture at 20°C or with 62 moisture at 37.7°C retained good quality. Gelatinization and rheological prOperties of the purified navy bean starch pastes were evaluated with protocol developed using a computer-assisted Haake RV12 Rotovisco Viscometer. Aqueous starch pastes of four concentrations (62, 8%, 102., and 12%) were prepared in an MV cup with a perforated paddle mixer maintaining a constant rate of 140 shear of 50 rpm at temperatures of 75°C and 95°C for up to 60 min. Rotation of the paddle mixer at 50 rpm was shown to be the optimum speed for minimum mechanical breakdown of the pastes while providing homoge- neous starch suspensions*without stardh precipitation at low concen- trations and low temperatures. The.activation energy of bean starth gelatinization could be readily estimated through this technique and was found.to be 14.5 Kcal/ggmole. various swelling responses of navy bean.starch.during codking were deIonstrated. Maximum final viscosity values were obtained at 85°C for all starch concentrations accept a maximum at 95°C for 671. Thixotropic breakdown was Observed for starCh at 102 and 12% concen- trations daring initial shearing at high temperatures (85°C and 95°C) and this breakdown reached equilibrium.after certain.3hearing time. After the above cocking process, the flow behavior retrogradation tendency, and.stability of the hot and cooled (50°C) pastes were determined. The power law, Herschel-Bulkley, and Casson models were employed to characterize the paste flow. The gelatinized pastes achibited shear thimingbehavior and their flow curves best fit the Herschel-Bulkley model over the range of Shear rates tested. 'Yield stress, consistency index, and flow'behavior index were generally concentration and temperature dependent. Significant Changes of paste rheological properties occurred at stardh concentrations above 102 for 75°C and above 82 to 107.1 for 85°C and 95°C. RECQ’MENDATIONS FOR FURTHER RESEARCH It is suggested that the scarming electron microscope be employed to detect the status of the purified starch granules at various stages of the thermal response curves to further understand the mechanism of granule swelling and breakdown in relationship to viscosity changes . Studies could be conducted to evaluate the rheological properties of the purified navy bean starch paste after retort processing at various heating treatments . Tactm'al property studies of thermally produced navy bean starch gels using the Instron Universal Testing Machine should be conducted. 141 APPENDIXA APPENDIX A Table 17. Flow data of 62 navy bean starch paste cocked at 95°C for 60 min prior to measuremaits with the Haake Viscometer Angular Shear Shear ‘Yield Rep/ Velocity Torque Stre s Rate Stres rpm (Rad/ sec) (N-m) (N/ ) (1/ sec) (N/ ) l l - 1.553E-l 6.658E-4 4.374E+0 3.459E+0 0.55 2 2.16lE-1 9.625E-4 6.324E+O 4.921E+O 3 3.167E-l 1.099E-3 7.221E+O 7.388E+o 4 4.l60E-1 1.332E-3 8.748E+0 9.643E+O 5 5.239E-1 1.47lE-3 9.666E+O 1.269E+l 10 1.062E+0 1.819E-3 1.195E+l 2.486E+l 20 2.137E+0 3.001E-3 1.972E+1 4.960E+l 30 3.213E+0 3.534E-3 2.322E+l 7.623E+l 4O 4.266E+O 4.048E-3 2.66GE+1 1.009E+2 60 6.430E+O 4.832E-3 3.175E+0 1.549E+2 80 8.564E+O 5.317E-3 3.493E+1 2.083E+2 100 1.074E+1 5.787E-3 3.802E+1 2.595E+2 2 l 1.407E-l 6.903E-4 4.545E+O 3.177E+O 0.58 2 2.16lE-l 9.915E-4 6.514E+0 4.973E+0 3 3.206E-l 1.164E-3 7.645E+0 7.572E+O 4 4.193E-l 1.336E-3 8.776E+0 9.897E+0 5 5.262E-1 1.467E-3 9.636E+0 1.256E+l 10 1.065E+0 1.967E-3 1.293E+l 2.541E+l 20 2.141E+0 2.64lE-3 1.735E+l 5.109E+l 30 3.214E+0 3.106E-3 2.04IE+1 7.628E+l 4O 4.276E+O 3.560E-3 2.339E+l 1.019E+2 60 6.431E+O 4.351E-3 2.858E+1 1.509E+2 80 8.565E+0 4.962E-3 ’ 3.260E+l 1.992E+2 100 1.075E+l 5.518E—3 3.623E+l 2.553E+2 3 l 1.499E-1 6.536E-4 . 4.294E+0 3.335E+0 0.64 2 2.164E-l 9.984E-4 6.56OE+O 4.913E+O 3 3.167E-l 1.153E-3 7.577E+0 7.564E+0 4 4.157E-l 1.302E-3 8.556E+0 9.832E+O 5 5.232E-l 1.448E-3 9.513E+0 1.236E+l 10 1.062E+O 1.997E-3 1.312E+l 2.525E+l 20 2.139E+0 2.703E-3 1.776E+l 5.05lE+l 30 3.214E+0 3.258E-3 2.140E+1 7.582E+1 40 4.266E+0 3.726E-3 2.448E+1 l 023E+2 6O 6.429E+0 4.415E-3 2.901E+l 1.519E+2 80 8.568E+O 4.958E-3 3.257E+l 2.013E+2 100 1.073E+l 5.472E-3 3.595E+1 2.649E+2 142 APPENDIXB APPENDIX B Table 18. Flow data of 82 navy bean starch.paste codked at 95°C for 60 min prior to measurements with the Haake Viscameter Angular Shear Shear Yield Rep/ ‘Velocity Torque Stress Rate Stress rpm (Rad/ sec) (N-m) (N/tnz) (l / sec) (N/mz) 1 1 1.330E-1 1.446E-3 9.502E+0 3.033E+0 1.73 2 2.164E-1 2.046E-3 1.344E+1 5.024E+0 3 3.201E-1 2.393E-3 1.572E+l 7.382E+0 4 4.182E-1 2.973E-3 1.953E+1 9.663E+0 5 5.266E-l 3.217E-3 2.114E+l 1.275E+1 10 1.065E+0 4.231E-3 2.870E+l 2.564E+l 20 2.140E+0 5.627E-3 3.697E+0 5.083E+1 30 3.213E+0 6.675E-3 4.386E+l 7.638E+l 40 4.271E+0 7.573E-3 4.975E+1 1.027E+2 60 6.426E+0 8.906E-3 5.852E+l 1.507E+2 80 8.562E+0 1.017E-2 6.68lE+l 2.021E+2 100 1.075E+l 1.145E-2 7.519E+l 2.519E+2 2 l 1.454E-l 1.479E-3 9.714E+0 3.253E+0 2.10 2 2.165E-l 2.217E-3 1.456E+1 4.911E+0 3 3.105E-1 2.736E-3 1.797E+l 7.514E+0 4 4.190E-l 3.067E-3 2.015E+l 1.024E+1 5 5.262E-l 3.255E-3 2.139E+l 1.307ER1 10 1.064E+0 4.155E-3 2.730E+1 2.555E+1 20 2.140E+0 5.847E-3 3.841E+1 5.008E+1 30 3.213E+0 7.037E-3 4.623E+l 7.636E+1 40 4.274E+0 7.904E—3 5.193E+1 1.006E+2 60 6.428E+0 9.107E-3 5.983EH1 1.645E+2 80 8.563E+0 1.026E-0 6.740E+l 2.079E+2 100 1.075E+1 1.116E-2 7.332E+l 2.537E+2 3 l 1.309E-l 1.590E-3 1.045E+l 3.101E+0 1.95 2 2.159E-1 2.035E-3 1.344E+l 5.024E+0 3 3.182E-1 2.380E-3 1.572E+l 7.382E+0 4 4.199E-1 2.957E-3 1.953E+1 9.683E+0 5 5.280E-1 3.201E-3 2.114Efi1 1.275E+1 10 1.063E+0 4.052E-3 2.662E+1 2.525E+l 20 2.139E+0 5.435E—3 3.57lE+1 5.105E+l 30 3.213E+0 6.424E-3 4.220E+1 7.649E+l 40 4.270E+0 7.28lE-3 4.784E&l 1.028E+2 6D 6.427EHO 8.589E-3 5.643E+1 1.513E+2 80 8 564E+0 9.878E-3 6.490E+1 2.018E+2 100 1 075E+l 1.095E-2 7.192E+1 2.527E+2 143 APPENDIX C APPENDIX.C Table 19. Flow data of 102 navy bean.star¢h paste cocked at 95°C for 60 min prior to measm‘ements with the Haake Viscometer Shear Shear Yield Rep/ Velocity Tbrque Stre 3 Rate Stre 3 rpm (Rad/ sec) (N-m) (N/ ) (1/ sec) (N/ ) l 1 1.457E-1 3.226E-3 2.120E+l 3.293E+0 6.80 2 2.146E—l 4.442E-3 2.919E+l 4.952E+0 3 3.174E-l 5.141E-3 3.377E+l 7.760E+0 4 4.163E-1 5.587E-3 3.67lE+1 1.013E+1 5” 5.247E-l 6.126E-3 4.025E+1 1.260E+l 10 1.062E+0 8.066E-3 5.299E+l 2.550E+l 20 2.139E+0 1.076E-2 7.067E+1 S.075E+l 30 3.213E+0 1.29lE-2 8.484E+l 7.614E+l 40 4.265E+0 1.463E-2 9.610E+1 1.006E+2 60 6.428E+0 1.795E-2 1.179E+2 1.505E+2 80 8.569E+0 2.081E-2 1.367E+2 2.010E+2 100 1.074E+1 2.338E-2 1.536E+2 2.492E+2 2 l 1.475E-l 3.022E-3 1.986E+l 3.308E+0 6.61 2 2.147E-l 4.336E-3 2 849E+1 4.933E+0 3 3.180E-1 4.905E-3 3 222E+l 7.668E+0 4 4.174E—1 5.536E-3 3.637E+1 9.960E+0 5 5.243E-l 6.048E-3 3.974E+l 1.260E+1 10 1.062E+0 7.987E-3 5.247E+l 2.545E+l 20 2.136E+0 1.065E-2 6.996E+l 5.070E+1 30 3.211E+0 1.282E-2 8.424E+1 7.584E+l 40 4.265E+0 1.463E-2 9.612E+l 1.004E+2 60 6.429E+0 1.786E-2 1.174E+2 1.506EH2 80 8.563E+0 2.07lE-2 1.36IE+2 2.002E+2 100 1.074EH1 2.329E-2 1.530E+2 2.506E+2 3 l 1.871E-1 3.420E—3 2.247E+1 4.082E+0 6.70 2 2.147E-l 4.650E-3 3.055E+1 4.754E+0 3 3.167E-l 5.389E-3 3.540E+l 7.714E+0 4 4.176E-l 5.887E-3 3.867E+l 1.029E+l 5 5.256E-1 6.341E-3 4.166E+1 1.279E+1 10 1.063E+0 8.298E-3 5.452E+l 2.556E+l 20 2.139E+0 1.103E-2 7.248E+1 5.094E+l 30 3.212Ei0 1.310E-2 8.604E+l 7.63SE+1 40 4.269E+0 1.484E-2 9.747E+l 1.008E+2 60 6.430E+0 1.810E-2 1.189E+2 1.505E+2 80 8.567E+0 2.099E-2 1.379E+2 2.006Efi2 100 1.074E+1 2.362E-2 1.552E+2 2.498E+2 144 APPENDIXD APPENDIXD Table 20. Flow data of 127. navy bean starch paste cooked at 95°C for 60 min prior to measurements with the Haake Viscometer ar Shear Shear Yield Rep/ Velocity Torque Stress Rate Stre 8 rpm (Rad/sec) (N-m) (N/mz) (1/sec) (N/ ) 1 l 1 437E-1 6.359E-3 4.178E+1 3.254E+0 10.57 2 2 147E-l 8.786E-3 5.772E+1 4.981E+0 3 3 . 170E-1 9 . 944E-3 6 . 533E+1 7 . 791E+0 4 4 . l74E-l 1 . 09113-2 7 . 167E+l l . 005E+l 5 5 . ZSE-l l . 205E-2 7 . 916E+l l . 253E+1 10 l . 063E+0 1 . 60212-2 1 . 053E+2 2 . 534E+l 20 2 . l37E+0 2 . l69E-2 1 . 425E+2 5 . 049E+1 30 3 . 207E+0 2 . 643E-2 1 . 737E+2 7 . 536E+l 40 4 . 256E+0 3 . 031E-2 1 . 991E+2 9 . 983E+1 60 6 . 424E+0 3 . 761E-2 2 . 471E+2 l . 494E+2 80 8 . 566E+0 4 . 337E-2 2 . 849E+2 2 . 027E+2 100 1 . 075E+1 4 . 812E-2 3 . 162E+2 2 . 542E+2 2 1 1.369E-1 6.515E-3 4.281E+1 3.108E-I-0 , 11.56 2 2 . 155E-l 9 . 236E-3 6 . 068E+1 5 . 034E+0 3 3 . l64E-1 1 . 022E-2 6 . 717E+l 7 . 824E+0 4 4 . 180E-1 l . 130E-2 7 . 427E+l l . 006E+1 5 5 . 256E-l l . 24lE-2 8 . 156E+1 l . 260E+l 10 1 . 063E+0 1 . 633E-2 1 . 073E+2 2 . 543E+1 20 2 . 138E+0 2 . 216E-2 l . 456E+2 5 . 042E+1 30 3 207E+0 2.694E-2 1.77OE+2 7.546E+l 40 4 256E+0 3.082E-2 2.025E+2 1.006E+2 60 6 426E+0 3.728E-2 2.450E+2 1.501E+2 80 8 569E+0 4.332E-2 2.846E+2 2.029E+2 100 1 075E+1 4.748E-2 3.119E+2 2.567E+2 3 l 1 156E-l 6.736E+0 4.426E+l 2.724E+0 11.30 2 2 172E-l 9.058E+0 5.951E+l 5.130E+0 3 3 193E-l 1.077E+l 7.077E+l 7.728E+0 4 4 212E-l l.l66E+l 7.661E+1 1.027E+1 5 5 290E—1 1.279E+l 8.404E+l 1.260E+l 10 1 065E+0 1.739E+l 1.143E+2 2.527E+l 20 2 139E+0 2.397E+l 1.575E+2 5.034E+1 30 3 211E+0 2.902E+l 1.906E+2 7.553E+1 40 4 266E+0 3.333E+l 2.190E+2 1.004E+2 60 6 . 426E+0 4 . 028E+1 2 . 646E+2 1 . 518E+2 80 8 . 566E+0 4 . 563E+l 2 . 998E+2 2 . 043E+2 100 1 . 075E+1 4 . 999E+1 3 . 284E+2 2 . 557E+2 145 APPENDDiE APPENDIX E Table 21. Flow data of 62 navy bean.star¢h paste cocked at 95°C for 60 min and cooled to 50°C prior to measurements with the Haake‘Viscometer ar Shear Shear Yield Rep/ velocity Torque Stre 3 Rate Stre 8 rpm. (Rad/sec) (NAm) (N/ ) (l/sec) (N/ ) 1 1 1.547E-1 9.755E-4 6.409E+0 3.436E+0 0.59 2 2.156E-1 1.458E-3 9.577E+0 4.868E+o 3 3.170E-l 1.750E-3 1.150E+1 7.569E+0 4 4.152E-l 1.926E-3 1.265E+l 9.984E+0 5 5.236E-1 2.124E-3 1.396E+l 1.242E+l 10 1.061E+0 2.935E-3 1.928E+l 2.512E+1 20 2.136E+0 4.009E-3 2.634E+1 5.067E+l 30 3.213E+0 4.785E-3 3.144E+1 7.590E+l 40 4.265E+0 5.488E-3 3.606E+1 1.012E+2 60 6.428E+0 6.665E-3 4.379E+1 1.507E+2 80 8.561E+0 7.557E-3 4.965E+l 1.989E+2 100 1.073E+1 8.523E-3 5.600E+2 2.564E+2 2 l 1.237E-1 1.135E-3 7.456E+0 2.822E+0 0.63 2 2.156E-1 1.683E-3 1.106E+l 5.044E+0 3 3.178E-1 1.898E-3 1.247E+l 7.710E+0 4 4.163E-1 2.122E-3 1.394E+l 9.977E+0 5 5.237E—l 2.319E-3 1.523E+1 1.257E+1 10 1.061E+0 3.086E-3 2.028E+1 2.525E+1 20 2.136E+0 4.205E-3 2.763E+1 5.099E+1 30 3.213E+0 4.954E-3 3.255E+1 7.659E+1 40 4.263E+0 5.609E-3 3.685E+l 1.016E+2 60 6.427E+0 6.738E-3 4.427E+1 1.521E+2 80 8.561E+0 7.607E-3 4.998E+1 2.613E+2 100 1.073E+1 8.476E-3 5.569E+1 2 547E+2 3 l 1.235E-l 1.148E-3 7.654EHO 2.895Ei0 0.62 2 2.158E-l 1.694E-3 1.108E+l 5.423E+0 3 3.179E-1 1.904E-3 1.345E+1 7.925E+0 4 4.184E-1 2.142E-3 1.406E+1 1.042E+1 5 5.270E-1 2.415E-3 1.542E&1 1.285E+l 10 1.064E+0 3.098E-3 2.098E+1 2.542E+l 20 2.140E+0 4.306E-3 2.875E+1 5.098E+l 30 3.2503+0 4.996E-3 3.524E+1 7.542E+1 40 4.452EHO 5.867E-3 3.965E+1 1.045E+2 60 6.435EHO 6.796E-3 4.473E+l 1.534E+2 80 8.904E+0 7.908E-3 5.004E+1 2.048E+2 100 1.080E+1 9.045E-3 5.692E+1 2.642E+2 146 APPENDR F APPENDIXF Table 22. Flow data of 8% navy bean starch paste cocked at 95°C for 60 min and cooled to 50°C prior to measurements with the Haake‘Viscometer ar Shear Shear Yield Rep/ 'Velocity Torque Stre 3 Rate Stress rpm (Rad/sec) (N-m) (N/ ) (1/ see) (N/mz) l l 1.060E-1 3.264E-3 2.144E+1 2.484E+0 4.80 2 2.138E-1 4.639E-3 3.048E+1 5.044E+0 3 3.16lE-1 5.484E-3 3.603E+l 7.413E+0 4 4.16lE-1 6.404E-3 4.207E+1 9.743E+0 5 5.252E-1 7.103E-3 4.667E+1 1.277E+1 10 l.06lE+0 8.566E-3 5.628E+l 2.654Eh1 20 2.137E+0 1.09lE-2 7/l67E+1 5.123E+1 30 3.211E+0 1.287E-2 8.457E+1 7.675E+1 40 4.256E+0 1.446E-2 9.502E+1 1.013E+2 60 6.419E+0 1.755E-2 1.153E+2 1.502E+2 80 8.571E+0 2.039E-2 1.340E+2 2.004E+2 100 1.072E+l 2.290E-2 1.505E+2 2.491E+2 2 1 1.244E-1 3.384E-3 2.223E+1 2.341E+0 4.88 2 2.139E-1 4.926E-3 3.237E+l 4.991E+0 3 3.160E-l 5.66lE-3 3.719E+1 7.745E+0 4 4.165E-l 6.168E-3 4.052E+1 1.025E+1 5 5.249E-1 6.666E-3 4.379E+1 1.281ER1 10 1.062Efi0 8.562E-3 5.625E+l 2.579E+1 20 2.138E+0 1.113E-2 7.314E+1 5.111E+l 30 3.212E+0 1.324E—2 8.697E+l 7.650E+1 40 4.264E+0 1.492E-2 9.802E+1 1.011E+2 60 6.429E+0 1.810E-2 1.189E+2 1.511E+2 80 8.568E+0 2.088E—2 1.372Ea2 1.994E+2 100 1.076E+1 2.343E-2 1.539E+2 2.553E+2 3 1 1.106E-l 3.339E-3 2.194E+l 2.587E+0 4.92 2 2.140E-1 4.676E-3 3.072E+1 4.991E+0 3 3.174E-l 5.801E-3 3.811E+l 7.643E+0 4 4.159E-1 6.160E-3 4.047E+1 1.064E+1 5 5.24lE-1 6.574E-3 4.319E+1 1.293E+1 10 1.063E+0 8.445E-3 5.548E+l 2.583E&1 20 2.138E+0 1.098E-3 7.217E+l 5.117E&1 30 3.212E+0 1.297E-3 8.521E+1 7.653E+1 40 4.26IE+0 1.467E-3 9.636E+1 1.008E+2 60 6.425E+0 1.789E-3 1.176E+2 1.507E+2 80 8.570E+0 2.074E-3 1.362E+2 1.996E+2 100 1.076E+1 2.344E-3 1.546Efi2 2.504E+2 147 APPENDDCG APPENDIXG Table 23. Flow data of 102 navy bean starch paste cocked at 95°C for ‘ 60 min and cooled to 50°C prior to measurements with the Haake Viscometer ar Shear Shear 'Yield Rep/ velocity Torque Stre 8 Rate Stress rpm (Rad/ sec) (N-m) (N/ ) (l/sec) (N/mz) l l 1.200E-1 6.238E-3 4.098E+1 2.761E+0 12.36 2 2.138E-1 8.972E-3 5.894E+1 5.117E+0 3 3.161E-1 9.624E-3 6.323E+1 7.862E+0 4 4.163E-1 1.081E-2 7.102E+1 1.046E+1 5 5.241E-1 1.117E-2 7.335E+1 1.388E+1 10 1.061E+0 1.361E-2 8.940E+l 2.619E+l 20 2.136E+0 1.773E-2 1.165E+2 5.093E+1 30 3.207E+0 2.107E-2 1.384E+2 7.620E+1 40 4.261E+0 2.390E-2 1.570E+2 1.012E+2 60 6.424E+0 2.908E-2 1.910E+2 1.500E+2 80 8.562E+0 3.354E-2 2.204E+2 2.009E+2 100 1.074E+1 3.742E-2 2.458E+2 2.531E+2 2 1 1.377E-l 6.649E-3 3.806E+1 3.112E+0 12.55 2 2.149E-1 9.627E-3 5.352E+l 4.983E+0 3 3.170E-1 1.083E-2 5.943E+1 7.713EHO 4 4.165E-1 1.209E—2 6.542E+1 1.018E+1 5 5.254E-1 1.290E-2 6.905E+1 1.384E+1 10 1.062E+0 1.436E-2 8.718E+1 2.765E+1 20 2.135E+0 1.811E-2 1.190E+2 5.128E+1 30 3.211E+0 2.146E-2 1.410E+2 7.63IE+1 40 4.264E+0 2.437E-2 1.601E+2 1.012E+2 60 6.427E+0 2.949E-2 1.937E+2 1.505Ea2 80 8.563E+0 3.393E-2 2.229E+2 2.020E+2 100 1.074E+1 3.781E-2 2.484E+2 2.515E+2 3 1 1.189E-1 5.347E-3 3.513E+1 2.770E+0 12.72 2 2.142E-1 7.320E-3 4.809E+1 5.064E+0 3 3.155E-1 8.465E-3 5.562E+1 7.784E+0 4 4 174E-1 9.104E-3 5.981E+1 1.039E+1 5 5 253E-1 9.843E-3 6.467E+l l/273E+1 10 1 063E+0 1.293E-2 8.496E+1 2.551E+1 20 2.135E+0 1.73ZE-2 1.138E+2 5.060Ei1 30 3 211E+0 2.079E-2 1.366E+2 7/604E+1 40 4 262E+0 2.360E-2 1.551E+2 1.068E+2 60 6.426E+0 2/892E-2 1.900E+2 1.502E+2 80 8.560E+0 3.329E-2 2.187E+2 2.008E+2 100 1.074E+1 3 717E-2 2.442E+2 2.530E+2 148 APPENDRH APPENDIX H Table 24. Flow data of 122 navy bean.starch.paste codked at 95°C for 60 min and cooled to 50°C prior to measuremeits with the Haake Viscometer ar Shear Shear Yield Rep/ velocity Torque Stre 3 Rate Stress rpm (Rad/sec) (N—m) (N/ ) <1/sec) (N/m2) 1 l 1.312E-1 1.065E-2 6.994E+l 3.017E+0 21.29 2 2.155E-l 1.455E-2 9.561E+1 5.138E+0 3 3.157E-1 1.576E-2 1.035E+2 8.333E+0 4 4.177E-l 1.675E-2 1.101E+2 9.776E+0 5 5.260E-1 2.013E-2 1.322E+2 1.237E+l 10 1.062E+0 2.260E-2 1.485E+2 2.534E+1 15 1.595E+0 2.971E-2 1.952E+2 3.678E+1 20 2.136E+0 3.485E-2 2.290E+2 5.003E+1 30 3.207EHO 4.199E-2 2.759E+2 7.585E+1 40 4.255E+0 4.765E-2 3.131E+2 9.969E+1 50 5.348E+0 5.4123-2 3.555E+2 1.242E+2 2 1 1.321E-1 1.193E-2 7.839E+l 2.990E+0 23.12 2 2.160E-1 1.775E-2 1.166E+2 5.040E+0 3 3.156E-1 1.945E-2 1.278E+2 7.988E+0 4 4.188E-1 2.114E-2 1.389E+2 1.046E+1 5 5.273E—1 2.260E-2 1.485E+2 1.298E+1 10 1.064E+0 2.905E-2 1.908E+2 2.560E+1 15 1.594E+0 3.433E-2 2.255E+2 3.827E+1 20 2.136E+0 3.832E-2 2.518E+2 5.139E+1 30 3.206E+0 4.505E-2 2.960E+2 7.608E+1 40 4.251E+0 5.157E-2 3.388E+2 1.059E+2 50 5.156E+0 5.272E-2 3.464E+2 l.2505+2 3 l 1.303E-1 1.045E-2 6.868E+1 3.006E+0 22.50 2 2.166E-1 1.420E-2 9.329E+l 5.170E+0 3 3.140E-1 1.547E-2 1.016E+2 7.707E+0 4 4.198E-1 1.753E-2 1.152E+2 9.850E+0 5 5.277E-1 1.990E-2 1.308E+2 1.254E+1 10 1.063E+0 2.477E—2 1.628E+2 2.623E+1 15 1.594E+0 2.824E-2 1.855E+2 3.815E+1 20 2.136E+0 3.257E-2 2.140E+2 5.123E+l 30 3.207E+0 3.688E-2 2.423E+2 7.618E+l 40 4.252E+0 4.324E-2 2.841E+2 9 946E+l 50 5.349E+0 4.799E-2 3.153E+2 l 265E+2 149 APPENDDCI APPENDIXI Table 25. Flow data of 62 navy bean starch.paste codked at 95°C for 60 man, cooled to 50°C, and.he1d at 50°C for 30 min.prior to measmrenents with the Haake Viscometer ar Shear Shear Yield Rep/ velocity Torque Stre 3 Rate Stress rpm (Rad/sec) (N-m) (N/m ) (l/sec) 1 l 1.083E-l 8.883E-4 5.836E+0 2.484E+O 0.67 2 2.16lE-1 1.391E-3 9.140E+0 5.040E+0 3 3.172E-l 1.627E-3 1.069E+l 7.614E+0 4 4.154E-1 1.803E-3 1.184E+1 9.938Ei0 5 5.227E-1 1.991E-3 1.308E+1 1.246E+l 10 1.061E+0 2.664E-3 1.750E+1 2.524E+1 20 2.136E+0 3.688E-3 2.423E+1 5.027E+1 30 3.210E+0 4.448E-3 2.922E+l 7.582E+l 40 4.262E+0 5.064E-3 3.327E+1 1.015E+2 60 6.425E+0 6.097E-3 4.006E+1 1.505E+2 80 8.556E+0 7.014E-3 4.608E+1 2.026E+2 100 1.075E+l 7.874E-3 5.173E+1 2.471E+2 2 1 1.289E-l 1.026E-3 6.739E+0 2.920E+0 0.63 2 2.153E-1 1.548E-3 1.017E+1 4.996E+0 3 3.157E-1 1.749E-3 1.149E+1 7.625E+0 4 4.147E—1 1.967E-3 1.292E+1 9.897E+0 5 5.227E-l 2.158E-3 1.418E+l 1.250E+1 10 l.061E+0 2.897E—3 1.904E+l 2.545E+1 20 2.137E+0 3.862E-3 2.537E+1 5.071E+l 30 3.212E+0 4.591E-3 3.016E+l 7.656E+1 40 4.263E+0 5.166E-3 3.394E+1 1.018E+2 60 6.424E+0 6.281E-3 4.126E+1 1.510E+2 80 8.560E+0 7.145E-3 4.694E+1 1.995E+2 100 1.075E+1 8.017E-3 5.267E+1 2.551E+2 3 l 1.075E-l 1.057E-3 6.943E+0 2.489E+0 0.66 2 2.155E-l 1.576E-3 1.036E+1 5.088E+0 3 3.172E-1 1.802E-3 1.184E+1 7.595E+0 4 4.148E-l 2.039E-3 1.339E+1 9.845E+0 5 5.218E-1 2.243E-3 1.474E+l 1.255E+l 10 1.060E+0 2.893E-3 1.901E+1 2.554E+1 20 2.137Eh0 3.909E-3 2.568Ei1 5.060E+1 30 3.209E+0 4.665E-3 3.065E+l 7.630E+1 40 4.260E+0 5.264E-3 3.459E+1 1.020E+2 60 6.425E+0 6.331E-3 4.160E+1 1.514E+2 80 8.566E+0 7.264E-3 4.733E+l 1.991E+2 100 1.075E+1 8.087E-3 5.313E+1 2.568E+2 150 APPENDIX J APPENDIXJ Table 26. Flow data of 8% navy bean.starch paste codked at 95°C for 60 min, cooled to 50°C, and held at 50°C for 30 min prior to measurements with the Haake Viscometer ar Shear Shear ‘Yield Rep/ Velocity Torque Stress Rate Stress rpm (Rad/ sec) (N-m) (N/mz) (1/ sec) (N/mz) 1 1 1.206E-1 3.550E-3 2.332E+1 2.806E+0 4.93 2 2.148E-1 4.852E-3 3.188E+l 5.253E+0 3 3.178E-1 5.121E-3 3.364E+1. 7.897E+0 4 4.173E-1 5.816E-3 3.821E+1 9.582E+0 5 5.256E-l 7.029E-3 4.618E+1 1.217E+1 10 1.062E+0 8.887E-3 5.839E+1 2.539E+l 20 2.139E+0 1.204E-2 7.909E+1 5.132E+1 30 3.206E+0 1.432E-2 9.410E+l 7.828E+l 40 4.262E+0 1.54lE-2 1.012E+2 1.102E+2 60 6.417E+0 1.727E-2 1.134E+2 1.532E+2 80 8.567E+0 1.977E-2 1.299E+2 2.031E+2 100 1.074E+1 2.214E-2 1.455E+2 2.486E+2 2 l 1.547E-1 3.744E-3 2.460E+1 3.440E+0 4.85 2 2.151E-l 5.517E-3 3.625E+1 4.670E+0 3 3.185E-1 6.526E-3 4.287E+1 7.681E+0 4 4.182E-1 7.141E-3 4.691E+1 1.046E+l 5 5.264E-l 7.558E-3 4.966E+l 1.326E+1 10 1.063E+0 9.372E-3 6.157E+1 2.625E+1 20 2.135E+0 1.183E-2 7.769E+1 5.184E+l 30 3.208E+0 1.377E-2 9.046E+1 7.711E+1 40 4.263E+0 1.545E-2 1.015E+2 1.017E+2 60 6.417E+0 1.846E-2 1.213E+2 1.517E+2 80 8.543E+0 2.105E-2 1.383E+2 2.005E+2 100 l.074E+1 2.360E-2 1.551E+2 2.532E+2 3 1 1.386E-l 3.759E-3 2.470E+1 3.163E+0 4.89 2 2.146E-1 5.119E-3 3.363E+1 4.952E+0 3 3.180E-1 6.219E-3 4.086E+1 7.533E+0 4 4.178E-1 6.932E-3 4.554E+1 1.029E+1 5 5.256E-1 7.335E-3 4.819E+1 1.310E+1 10 1.064E+0 9.369E-3 6.156E+1 2.612E+1 20 2.137E+0 1.174E-2 7 712E+l 5.202E+1 30 3.209E+0 1.361E-2 8 938E+1 7.734E+1 40 4.260E+0 1.523E-2 1.001E+2 1.018E+2 60 6.421E+0 1.824E-2 1.199E+2 1.517E+2 80 8.548E+0 2 085E-2 1.370E+2 2.014E+2 100 1.074E+l 2 331E-2 1.532E+2 2 511E+2 151 APPENDHK 2.3un «>56 MET—.624. A acme—Baum: noon? Scum A 5 305m 33> 53w 5.2.x and 36 £6 cad mod nod and Hnd mud and and >0 $od mNHdHN NHod HHddHH NomdN eeHdd wnN.$ oomdN ONod mHNdNH mm H38. Hood mend Hood mHHd Awood Hodd man NNHd Hood HCNd «N not”... £oHod EHHmdNH £HNod £Nmm.NoH EoNodn iHod £mn~.$ EdmmdN hEreNNod £0342 c 950 x 989. than d EmNa do: «.3de d Emma . «on trends . SH «.103 d flesh .qu «town .SH «.308 d £53 . 9a m 980 £8N.o £NnNdmm timid £om.mNn Eomndd £HNH.o EHmodmN £N8KHH £$H.d inadmn N 93a. imHN d «fawn dmo Emma d 1353. Nmm framed dN 1&ch d £§.MHN in: .nm «.158 d ENNn .23 HH ogmofiu mouldflmw 5m: 0 . o o a mo 8wu§> x a x c x e c x e. a x mo meadow Samoa NonstmuHozomuw: Honv—HafinHonumum: 5 age 83.83.; «3.3: 05 53 35.55585 on word on: do how Demo 98 .oona do: um 3480 933958080 «:35; «o $ng 50.83. ammo bad now «Havoc— ...oHu 8333 g Boa—do $336.53 Hdowonoonu no @0555 we naming .NN «HA3. x 595% 152 APPENDDCL APPENDIXL Table 28. Correlation coefficients, r2, and sumrof-squares function, SS(b), for regression analysis of the flow curve data1 2 Treatment 1' SS (b) Temp,Conc Power *Herschel- Herschel- (°C,Z db) Law Bulkley Casson Bulkley 75 6 0.9200 0.9045 0.8625 260.30 8 0.9890 0. 9890 0.9800 102.32 10 0.9679 0.9666 0.9412 176.14 12 0.9742 0.9632 0.9387 120.43 85 6 0.9822 0.9822 0.9617 54.72 8 0.9910 0.9791 0.9639 51. 98 10 0.9933 0.9847 0.9779 182.32 12 0.9922 0.9866 0.9775 200. 25 95 6 0.9936 0.9915 0.9702 29.70 8 0.9938 0.9891 0.9711 56.12 10 0.9941 0.9912 0.9871 308. 30 75/ 6 0.9520 0.9400 0.9195 240.25 50 8 0.9847 0.9847 0.9620 201.38 10 0.9734 0.9653 0.9388 176.42 12 0.9711 0.9294 0.9054 200.45 85/ 6 0.9861 0.9861 0.9629 15.50 50 8 0.9927 0.9709 0.9629 104.50 10 0.9864 0.9760 0.9662 162.33 12 0.9648 0.9563 0.9515 180.27 95/ 6 0.9963 0.9955 0.9785 13.17 50 8 0.9945 . 0.9936 0.9851 168.86 10 0.9919 0. 9933 0.9896 417.88 12 0.9859 0.9831 0.9810 204.50 75/ 6 0.9245 0.9232 0.9045 134.95 50/ 8 0.9423 0.9328 0.9288 106.32 50 10 0.9433 0.9405 0.9234 107.05 12 0.9426 Q 9401 Q 9382 235.25 153 Table 28 (cont ' d. ) r2 SS(b) Treatment Temp,Conc Power Herschel- Herschel- (°C, Zdb) Law Btflkley Casson Bulkley 85/ 6 0.9213 0.9213 0.9179 160.98 50/ 8 0.9913 0.9762 0.9724 108.95 50 95/ 6 0.9975 0.9969 0.9809 124.32 50/ 8 0.9894 0.9870 0.9748 177.45 50 1n = 3 154 APPENDRM 9:00am «>50 0857} £555me: 0000.3 Scum 03H; 30.3 53m 54: mod n53 mod Ed 3d 36 owdm mod mod 3.3 A5 >0 SN d 03.1.3 Bod NS .mHm .NNN do Sod mdeoN mnodm ONO d 03de mm 3008. Noo d SN .3 Sod 3N d ESod So d Sod mmod So d mNod qN 00PM. ESNd Hood: Eiod 30.3.” £873 £n8d £waémm £HNM.NN £o~od EoHNdwm o 0:00 x 95H. idem A mom doom inNo d cow .NcmH «.1an . $4 «.3de d inn .NnoH EH8 . an framed d £1me . .VMNm n 0:00 ENRd woodeN £mo~d mNNdme iqmonN iuoad ¥Nnodm2 £084.? EDNd £08.3NN N 95“. tinned ONm .Nq: fiend d 0863 £mmadoN h.réNeod £3043 «1N3 .HmN «faced £133.23 3 0553.5. 30 ~80: 0 o o a mo 83303, x ~ x c x w c x p c x NO 00.38 :0me xmaxaamuaonomumz 032-3630: 5 003m N H 0009503> .870: 05 5.3 35565000.: 00 nova Doom 3 03000 can 55 do you oomd v.8 .93 down 00 «.9300 963055080 maxing mo 80me £00.36 5.00 .92 pow 30006 308 0000300 gum 0009950 0000060qu 30330050 we 00§Hm> no 39392 .mN 033. Sag 155 LISIOFREFERENCES LIST OFREFERENCES AACC. 1962. "Approved Methods," 7th ed. American Association of Cereal Chemists, St. Paul, W. Aguilera, J.M., Lusas, E.W., Uebersax, M.A., and Zabik, M.E. 19823. Roasting of navy beans (Phaseolus vulgaris) by particle-to- particle heat transfer. J. Food Sci. 47:996. Aguilera, J.M., Lusas, E.W., Uebersax, M.A., and Zabik, M.E. 198b. Develognmt of food ingredients fran navy beans (Phaseo Zus vulgaris) by roasting and air-classification. J. Food Sci. 47: 1151. Akpapunam, M.A. and Markakis, P. 1979. Oligosaccharides of 13 American cultivars of cmpeas (Vigna sinensis). 'J. Food Sci. 44:1317. Albrecht, J .J ., Nelson, A.I., and Steinberg, M.P. 1960. Characteristics of corn starch and derivatives as affected by freezing, storage and tracing. Food Technol. 14:57. Allen, J. E., Hood, L. F., and Chabot, J. F. 1977. Effect of heating on the freeze-etch ultrastructure of hydroxyp ropyl distarch phosphate and urmodified tapioca starches. Cereal Chem. 54: 783. Anker, C.A. and Geddes, W.F. 1944. Gelatinization studies upon wheat and other starches with the amylograph. Cereal Chan. 21: 335. Anon. 1979. Michigan Food Facts. Michigan Department of Agriculture, Lansing, MI. Bagley, E.B. and Christianson, D.D. 1982. Swelling capacity of starch and its relationship to suspension viscosity---effect of cooking time, tanperature and concentration. J. Textm'e Studies 13: 115 . Bakker-Arkana,F.W.,Palnitkar,1‘/L, andBedford, C..L 1968. Spray ofpeabemsinacmcurrenthorizontaldrier. Trans. ASAE 11:10. Bakker-Arkana, F.W., Pattersm, R.J., and Bedford,C.L. 1967. Drying characteristics of pea beans on single and double drum driers. Trans. ASAE 10:154. Bakker-Arkam, F.W., Patterson, R.J., and Bedford, C.L. 1969. Econanics of producing instant pea beans, pea, and lentil powders. Trans ASAE 12:13. 156 157 Bakshi, A.S. and Singh, R.P. 1980. Kinetics of water diffusion and i3t§rch gelatinization during rice parboiling. J. Food Sci. 45: 7. Bhattachaxya, K.R. and Sowbhagya, C.M. 1978. On viscograms and viscography with special reference to rice flour. J. Texture Studies 9:341. Bidlack, W.R. and Tappel, A.L. 1973. Damage to microsomal marbrane by lipid peroxidation. Lipids 8:177. Biliaderis, C.G., Grant, D.R., and Vose, J.R. 1979. lVblecular weight gilstrilsmgtli‘on of legume starches by gel chranatography. Cereal em. : 75. Biliaderis, C.G., Grant, D.R., and Vose, J .R. 1981a. Structural characterization of legune staches. I. Studies on amylose, amlepectin, and beta-limit dextrins. Cereal Chem. 58:496. Biliaderis, C.G., Grant, D.R., and Vose, J.R. 1981b. Structural characterization of legune starches. II. Studies on acid- treated starches. Cereal Chem 58:502. Bird, R.B., Stavart, W.E., and Lightfoot, E.N. 1960. "Transport Phenanena," John Wiley and Sons, New York, NY. Bothast, R.J., Anderson, R.A., Warner, K., and Kwolek, W.F. 1981. Effects of moisture and taperature on microbiological and sen- sory prOperties of wheat flour and corn meal during storage. Cereal Chan. 58:309. Bowman, D.E. 1944. Fractions derived fran soybeans and navy beans which retard tryptic digestion of casein. Proc. Soc. Exp. Biol. Med. 57:139. Bressani, R. and Elias, L. 1974. Legune foods. In "New Protein Foods," Vol. 1A, A.M. Altschul (Ed.), Acadanic Press, Nav York, NY. Brown, R.O. and Harrel, C.G. 1944. The use of the amylograph in the cereal laboratory. Cereal Chem. 21:360. Buchanan, R.A. 1969a. Effect of storage and lipid extraction on the properties of leaf protein. J. Sci. Food Agr. 20:359. Buchanan, R.A. 1969b. In vivo and in vitro methods of measuring nutritive value of leaf protein preparations . Brit . J . Nutr . 23:533. Burr, H.K., Boggs, M.M., Morris, H.J., and Venstran, D.W. 1969. Stability studies with. cooked legune powders. 2. Influence of various factors on flavor of lime bean powder. Food Technol. 23: 134. Buttery, R.G., Handel, C.E., and Boggs, M.M. 1961a. Autoxidation of potato grznules. Part 1. Changes in fatty acids. J. Agr. Food Chem. 9:2 5. 158 Buttery, R.G., Handel, C.E., and Boggs, M.M. 1961b. Autoxidation of potato granules. Part 2. Formation of carbonyls and hydro- carbons. J. Agr. Food Chem. 9:248. Canovas, G.V.B. and Peleg, M. 1983. Flow parameters of selected cannercial sand-liquid food products. J. Texture Studies 14: 213. Carvalho, C.C.C., Jansen, G.R., and Harper, J.M. 1977. Protein quality evaluation of an instant bean powder produced by (by heat processing. J. Food Sci. 42:533. Casson, N. 1959. A flow equation for pigment-oil suspensions of the printing ink type. In "Rheology of Disperse Systems," C.C. Mi‘ll (Ed.). p. 82, Pergamnon Press, New York, NY Cernimg—Beroard, J. and Filiatre, A. 1976. A canparison of the carbo- hydrate composition of legune seeds: horsebeans, peas, and lupines. Cereal Chem. 53:968. Chang, K.C. and Satterlee, L.D. 1979. Chemical, nutritional and Im'cro- biological quality of a protein concentrate frcm culled dry beans . J. Food Sci. 44:1589. Charm, S.E. 1963a. "The andamentals of Food Engineering," AVI Publishing Co., Inc., Westport, CT. Charm, S.E. 1963b. The direct determination of shear stress-shear rate behavior of foods in the presence of a yield stress. J. Food Sci. 28:107. Chevalley, J. 1975. Rheology of chocolate. J. Texture Studies 6:177. Counter, B.T. 1969. Storage stability of drum dried navy bean powders. M.S. thesis, Michigan State University, East Lansing, MI D'Appolonia, B.L. 1977. Rheological and baking studies of legune- wheat flour blends. Cereal Chan. 54:53. D'Appolonia, B.L. 1978. Use of untreated and roasted navy beans in bread making. Cereal Chem. 55:898. ' Defouw, C., Zabik, M.E., Uebersax, M.A., Aguilera, J.M., and Lusas, El 1982. Use of unheated and heat-treated navy bean hulls as a source of dietary fiber in spice-flavored layer cakes . Cereal Chan. 59:229. Dugan, L.R., Jr. and Rao, G.V. 1972. Evaluation of the flavor contribution of products of the Maillard reaction. Tech. Rpt. 72-27FL, U.S. Army Natick laboratories, Natick, MA. Duran, L. and Costell, E. 1982. Rheology of apricot puree: character- ization of flow. J. Texture Studies 13:43. 159 Evans, I.D. and Haisman, D.R. 1979. Rheology of gelatinized starch suspensions. J. Texture Studies 10:347. Firth, B.A. and Hunter, R.J. 1976. Flow properties of coagulated colloidal suspensions. J. Coll. Interface Sci. 57:248. Fletcher, B.L. and Tappel, A.L. 1971. Fluorescent modification of serum albumin by lipid peroxidation. Lipids 6:172. Frazier, W.C. and Westhoff, D.C. 1978. Contamination, preservation, and spoilage of cereal products . In "Food Microbiology," 3rd ed. , Ch. 10, McGraw-Hill Book Co. , New York, NY. French, D. 1972. Fine structure of starch and its relationship to the organization of starch granules. J. Jap. Soc. Starch Sci. 19:8. Friedman, L. and Kline, A.L. 1950. The relation of the amino-sugar reaction to the nutritive value of protein hydrolyzates . J. Nutr. 40:295. Gmge, P.T. and Matsushita, S. 1973. Interactions of the autoxidized products of linoleic acid with enzyme proteins. Agr. Biol. Chem. 37:1. Ganz, A.J. 1965. Effect of sodium chloride on the pasting of wheat starch granules. Cereal Chem. 42:429. Guadagni, D.C., Dunlap, C., and Ken, S. 1975. Effect of temperature and packaging atmosphere on stability of drun-dried pinto bean powder. J. Food Sci. 40:681. Hansuld, M.K. and Briant, A.M. 1954. The effect of citric acid on selected edible starches and flours. Food Res. 19:581. Henry, K.M. and Ken, S.K. 1948. Deterioration on storage of dried skim milk. IV. Changes in the biological value of the proteins . J. Dairy'Res. 15:341. Henry, K.M. and Ron, S.K. 1950. Effect of reaction with glucose on the nutritive value of casein. Biochem. BiOphys. Acta. 5:455. Herschel, W.H. and Bulkley, R. 1926. Kansistenzmessungen von gnnmni- benzollosunge. Kolloid—Z . 39: 291 . Hester, E.E., Briant, A.M., and Personius, C.J. 1956. The effects of sucrose on the properties of sane starches and flours. Cereal Chan. 33:91. Hodge, J.E. and Osman, E.M. 1976. Carbohydrates. In "Principle of Food Science - Part I: Food Chemistry," D.R. Fermena (Ed.), Ch. 10, Marcel Dekker, Inc., New York, NY. Hoff, J .E. and Nelson, P.E. 1966. Methods for accelerating the processing of dry beans. Rpt. 8th Dry Bean Res. Conf., Bellaire, MI. USDA-ARS—74-41:39. 160 Hoffnnan, R.L. 1972. Discontinuous and dilatent viscosity behavior in concentrated suspension. I. Observation of a flow insta- bility. Trans. Soc. Rheol. 16:155. Hoffman, R.L. 1974. Discontinuous and dilatent viscosity behavior in concentrated suspensionn. II. Theory and eqnerimental tests. J. Colloid Interface Sci. 46:491. Holdsworth, S.D. 1971. Applicability of rheological models to the interpretation of flow and processing behavior of fluid food products. J. Texture Studies 2:393. Honavar, P.M., Shih, G.V., and Liener, LE. 1962. Inhibition of the growth of rats by pm'ified henagglutinin fractions isolated frann PhaseoZus vulgaris. J. Nutr. 77:109. Jaffe, W.C. 1975. Factors affecting the nutritional value of beans. In "Nutritional Improvenent of Food Legumes by Breeding," Proc. Sym. Food and Agr. Org., Rome, Italy, Johnn Wiley and Sons, New York, NY. Jaffe, W.C. 1980. Hennagglutinins (Lectins). In "Toxic Constituents of Plant Foodstuffs," 2nd ed., LE. Liener (Ed.), Academic Press, New York, NY. Jones, D.B. and Gersdorff, C.E.F. 1941. The effect of storage on the protein of wheat, white flour and whole wheat flour. Cereal Chen. 18:417. Kakade, M.L. and Evans, R.J. 1963. Effect of heat on the in vitro digestion of navy beans (P. vulgaris). Michigan Quarterly Bull. 46287. Michigan Agr. Expt. Stn., East Lansing, MI. Karel, M. and Iabuza, T.P. 1968. Water conntent and lipid oxidation in dehydrated foods. Prog. Rpt. Pub. Health Ser. Res. Grant No. UIROI-00125. Kenpf, V.W. and Kalender, G. 1972. Pbeglichkeiten einner Standardisierung von Viskositaetsmessnmngen znm: vergleichenden Buerteilung der rheologischen Eigenschaften und des rheologischen Verhaltens der Staerke. Die Staerke 24:220. Koch, R.B. 1962. Dehydrated foods and model syst-s. In "Symposiun on Foods: Lipids and Their Oxidation," p. 230. AVI Publishing Co., Westport, CT. Ken, 8., Wagner, J.R., Guadagni, D.C., and Horvat, R.J. 1970. pH adjustment conntrol of oxidative off-flavors during grinding or raw legume seeds. J. Food Sci. 35:343. Krieger, I.M. 1968. Shear rate in the couette viscometer. Trans. Soc. Rheology 12:5. Kubota, K., Hosokava, Y., Suzuki, K., and Hosaka, H. 1979. Studies of the gelatinization rate of rice and potato starches. J. Food Sci. 44:1394. 161 Labaneiah, M.E.O. and Luh, B.S. 1981. Changes of starch, crude fiber and oligosaccharides in germinating dry beans. Cereal Chem. 58: 135. Iabuza, T.P. 1968. Sorption phenomena in foods. Food Technol. 22:263. Labuza, T.P., Tannenbaum, S.R., and Karel, M. 1970. Water content and stability of low-moisture and intermediate-moisture foods. Food Technol. 24: 35. Lai, C.C. and Varriano-Marston, E. 1979. Studies on the characteristics of black bean starch. J. Food Sci. 44:528. Lea, C.H. and Parr, L.J. 1961. Somne observations on the oxidative deterioration of the lipids of crude leaf protein. J. Sci. Food Agr. 12:785. Lea, C.H. and Wnite, J.C.D. 1948. Deterioration of storage of dried skim milk. III. Physical, chemical and palatability changes in stored powders. J. Dairy Res. 15:298. Leach, H.W. 1965. Gelatinization of starch. In "Starch: Chenistry and Technology," Vol. I, R.F. Wistler and E.F. Paschall (Eds.), Academic Press, New York, NY. Lewis, S.E. and Wills, E.D. 1962. The destruction of -SH groups of proteins and amnino acids by peroxides of unsaturated fatty acids . Biochen. Pharmacol. 11:901. Liener, LE. and Hill, E.G. 1953. The effect of heat treatment on the nutritive value and hemagglutinating activity of soybean oil mneal. J. Nutr. 49:609. Lineback, D.R. and Re, C.H. 1975. Starches and low-molecular weight carbohydrates from chick pea and horsebean flours . Cereal Chen. 5 : 25. Little, T.M. and Hills, F.J. 1972. Same basic concepts. In "Statis- tical Methods in Agricultural Research," p. 13. Agricultural extension, University of California, Berkeley, CA. Lorenz, K. 1979. The starch of the faba bean (Vicia faba). Staerke 31:181. Love, M.H. 1975. Physical and chemical assessment of the changes in quality characteristics of stored instant navy bean powder. Ph.D. dissertation, Michigan State University, East lansing, MI. Luh, B.S., Maneepnmn, S., and Rucker, R.B. 1975. Biological quality and functional properties of lime bean protein for bread enrich- ment. In "Protein Nutritional Quality of Foods and Feeds," Vol. 1, Part 2, p. 135. Quality factors, plant breeding, composition, - processing and antimntrients, M. Friedman (Ed.), Marcel Dekker, Inc. , Nev Y , NY. 162 Mackley, M.C. 1937. The colloidal behavior of flour doughs. Cereal Chem. 14:434. Maneepun, N.S., Luh, B.S., and Russell, H.B. 1974. Amino acid compo- sition and biological quality of lime bean protein. J. Food Sci. 39:171. Matsushita, S. 1975. Specific interactions of linoleic acid hydro- peromcides and their seccmdary degraded products with enzyme proteins. J. Agr. Food Chem. 23:150. Mazurs, E.G., Schoch, T.J., and Kite, P.E. 1957. Graphical analysis of the Brabender viscosity curves of various starches . Cereal Chen. 34:141. McCabe, W.L. and Smith, J.C. 1967. "Unit Operations of Chenical Enngineering," McGraw-Hill Book Canpany, New York, NY. McInroy, E.E., Murer, H.K., and 'I'niessin, R., Jr. 1945. The effect of autoclaving with dextrose on the nutritive value of casein. Arch. Biochen. Biophys. 20:256. Medcalf, D.C. and Giles, K.A. 1966. Effect of lyotropic ion series on the pasting of wheat and corn starches. Staerke, 4:101. Melnick, D. and Oser, B.L. 1949. ‘Ihe influence of heat processing on the functional and nutritive properties of protein. Food Technol. 3:57. Mewis, J. 1979. 'Ihixotropy---a general review. J. Non-Newtonian Fluid Mechanics 6:1. Miller, B.S., Derby, R.I., and Trimbo, H.B. 1973. A pictorial explanation for the increase in viscosity of a heated wheat starch-water suspension. Cereal Chen. 50:271. Miller, D.S. 1956. The nutritive value of fish proteins. J. Sci. Food Agr. 7:337. Morad, M.M. and Finney, P.L. 1980. Note on a simple method to produce a high extraction faba bean cotyledon flour. Cereal Chen. 57:230. Morad, M.M., Leung, H.K., Hsu, D.L., and Fimney, P.L. 1980. Effect of germination on physicochemical and bread-baking properties of yellow pea, lentil, and faba bean flours and starches. Cereal Chen. 57:390. Nbrgan, R.J. 1968. A study of the phenomenon of rheological dilatancy in an aqueous pigment suspension. Trans Soc. Rheol. 12:511. Morris, H.J. 1961. Instant bean powders. Proc. Fifth Dry Bean Res. Conf., Denver, CO. USDA-ARS. ps4. 163 Muller, H.G. 1973. "An Introduction to Food Rheology," Crane Russak and Campany, Inc. , New York, NY. Nachbar, M.S. and Oppenheim, J .D. 1980. Lectins in the United States diet: a survey of lectins in commonly censured foods and a review of the literature. Amer. J. Clin. Nutr. 33:2338. Naivikul, O. and D'Appolonia, B.L. 1978. Comparison of legure and wheat flour carbohydrates . I . Sugar analysis . Cereal Chem. 55:913. . Naivikul, O. and D'Appolonia, B.L. 1979. Carbohydrates of legume flours campared with wheat flour. II. Starch. Cereal Chen. 56:24. Nie, N.H., Hull, C.H., Jenkins, J.G., Steinbrenner, K., and Bent, D.H. 1975. "Statistical Package for the Social Sciences," 2nd ed. , McGraw-Hill Book Campany, New York, NY. Niihara, R. and Matsunoto, S. 1981. Trial method for measnming the effect of lipids on the rheological properties of starch paste during the gelation process. J. Texture Studies 12:559. Noah, N.D., Bender, A.E., Reaidi, G.B., and Gilbert, R.J. 1980. Food poinsoning fran raw red kidney beans. Brit. Med. J. 281:236. Odigboh, B.U. and Mnhsenin, N.N. 1974. Effects of conncentration on the viscosity profile of cassava starch pastes during the cooking- cooling process. J. Texture Studies 5:441. Odigboh, B.U. and Mohsenin, N.N. 1975. Viscosity characterization of unmodified cassava starch paste. J. Texture Studies 6:363. Osman, E.M. and Dix, M.R. 1960. Effects of fats and nonionic surface- active agents on starch pastes. Cereal Chem. 37:464. Patel, K.M., Bedford, C.L., and Youngs, C.W. 1980. Amnino acid and mnineral profile of air-classified navy bean flour fractions. Cereal Chen. 57:123. Patel, K.M., Caul, J.E., and Johnson, J.A. 1977. Horsebean as protein supplenent in breadmaking. III. Effects of horsebean protein on aroma and flavor profile of moroccan- type bread. Cereal Chem. 54: 379. Patel, K.M. and J .A. Johnson. 1975. Horsebean protein supplement in 1n breadmeking. II. Effect on physical dough properties , baking quality, and amino acid carposition. Cereal Chen. 52:791. Paton, D. 1981. Behavior of Hinoat oat starch in sucrose, salt, and acid solutions. Cereal Chen. 58:35. Pattonn, A. R. Hill, E. G., and Foreman, E. M. 1948a. ‘Amnino acid impairment in casein heated with glucose. Science 107:623. 164 Patton, A. R., Hill, E. G. and Formen, E. M. 1948b. The effect of browning on the essential amino acid content of soy globulins. Science 108: 659. Pererav1 A. D., Leklemn, J. E., and Miller, L.T. 1979. Stability of dunngbread making and storage of bread of flour. Cereal twangmm 56: 577. Plant, R. and Tulsiani, D.R.P. 1969. Solubility, amino acid campo— sition and biological evaluation of protein isolated from leguminous seeds. J. Agr. Food Chem. 17:361. Polit, P.F. and Sgarbieri, V.C. 1976. Some physicochemical and nutritional properties of castor bean.(Ricinus communis) protein. J. Agr. Food Chem. 24:795. Pomeranz, Y., Slnogren, M.D., Finney, K.F., and Bechted, D.B. 1977. Fiber in breadmaking-effects on functional properties. Cereal (REEL 54:25. Privett, 0.8. and Blank, M.L. 1962. The initial stages of autoxi- dation. J. Amer. Oil Chem Soc. 39:465. Puztai, A. 1967. Trypsin inhibitors of plant origin: their chen- istry and potential role in animal nutrition. Nutr. Abstr. Rev. 37:1. Rajchel, C.L., Zabik, M.E., and Everson, E. 1975. Wheat bran and middlings: a source of dietarynfiber in banana, chocolate, nut, and spice cakes. Bakers Digest 49(3):27. Rao, M.A. 1977. Rheology of liquid foods---a review. J. Texture Studies 8:135. . Rao, M.A., Bourne, M.C., and Cooley, H.J. 1981. Flow properties of tomato concentrates. J. Texture Studies 12:521. Read, J.W. and Haas, L.W. 1938 Studies on the baking quality of flour as affected by certain enzyme actionns. Cereal Clnem. 15:59. Reddy, N.R., Pierson, M.D., Sathe, S.K., and Salunkhe, D.K. 1984. Chemical , nutritional and physiological aspects of dry bean carbohydrate—«a review. Food Chemistry. 13:25. Reichert, R.D. and Youngs, C.G. 1978. Nautre of the residualprotein associated with starch fractionns fran air-classified field peas. Cereal Chen. 55:469. Rice, E.E. and Beuk, J.F. 1953. Effects of heat on nutritive value of protein. Advan. Food Res. 4:223. Robertson, J.B. and Van Soest, P.J. 1977. Dietary fiber estimation in concentrate foodstuffs. J. Animal Sci. 45(Suppl. 1):254. Robinson—Lang, E. and Rha, C.K. 1981. Determination of the yield stress . of ‘hydrocolloid dispersion. J. Texture Studies 12:47. 165 Rockland, L.B. 1960. Saturated salt solutions for static control of relative humnidity between 5°C and 40°C. Anal. Chem. 32:1375. Rockland, L.B. 1966. New quick-cooking dry limna beans. Report of the Eighth Dry Bean Conf. , Bellaire, MI. USDA-ARS-74-41:16. Rockland, L.B. 1969. Water activity and storage stability. Food Technol. 23:1241. Roubal, W.T. 1971. Free-radicals, malonaldehyde and protein danege in lipid-protein systems. Lipids 6:62. Roubal, W.T. and Tappel, A.L. 1969. Polymerizationn of proteins induced by free-radical lipid peroucidation. Arch. Biochen. Biophys . 113: 150. Rundle, R.E., Daasch, L., and French, D. 1944a. The structure of the "B" modification of starch frcm film and fiber diffraction dia- grams. J. Amer. Chen. Soc. 66:130. Rundle, R.E., Foster, J.F., and Baldwin, R.R. 1944b. On the nature of the starch-iodine camplex. J. Amer. Chem Soc. 66:2116. Sahasrabudhe, M.T., Quinn, J.R., Paton, D., Youngs, C.G., and Skura, B.J. 1981. Chemical composition of white bean (Phaseolus vulgaris L.) and functional characteristics of its air-classi- fied protein and starch fractions. J. Food Sci. 46:1079. Sathe, S.K., Iyer, V., and Salunkhe, D.K. 1981a. Investigation of the Great Northern bean (Phaseolus vulgaris L.) starch: solubility, swelling, interactionn with free fatty acids, and alkaline water {stintion capacity of blends with wheat flours. J. Food Sci. 46: 1 . Sathe, S.K., Ponte, J.C. Jr., Rangnekar, P.D., and Salunkhe, D.K. 1981b. Effects of addition of Great Northern bean flour and protein concentrates on rheological properties of dough and baking quality .of bread. Cereal Chen. 58:97. Sathe, S.K. and Salunkhe, D.K. 1981a. Functional properties of the Great Northern bean (Phaseolus vulgaris L.) proteins: emlsion, foaming, viscosity and gelation properties. J. Food Sci. 46:71. Sathe, S.K. and Salunkhe, D.K. 1981b. Solubilization and electro- phoretic characterization of the Great Northern bean (Phaseo Zus vulgaris L.) proteins. J. Food Sci. 46:82. Sathe, S.K. and Salunkhe, D.K. l981c. Isolation, partial character- ization, and modification of the Great Northern bean (PhaseoZus vulgaris L.) starch. J. Food Sci. 46:617. Satlne, S.K. and Salunkhe, D.K. l981d. Solubilization of California small white bean (Phaseolus vulgaris L.) proteins. J. Food Sci. 46:952. 166 Satterlee, L.D., Benbers, M., and Kendrick, J.G. 1975. Functional prOperties of the Great Northern bean (Phaseolus vu Zgams) protein isolate. J. Food Sci. 40:81. Schoch, T.J. and Maywald, E.C. 1968. Preparation and prOperties of various legume starches. Cereal Chen. 45:564. Shafer, M.A.M. and Zabik, M.E. 1978. Dietary fiber sources for baked products: camparison of wheat bran and other cereal brans in layer cakes. J. Food Sci. 43:375. Shashoua, V.E. and Beaman, R.G. 1958. Microgel: an idealized poly- mer molecule. J. Polymer Sci. 33:101. Sherman, P. 1970. "Industrial Rheology," Academic Press, New York, NY. Smnith, A.K. and Circle, S.J. 1972. "Soybeans: Chemistry and Tech- nology," Vol. 1, "Proteins," Avi Publishing Co., Westport, CT. Smith, R.J. 1964. Viscosity of starch pastes. In "Methods in Carbo- hydrate Chemistry," R.L. Wistler (Ed.), Academic Press, New York, NY. Sosulski, F.W. and Youngs, C.G. 1979. Yield and functional prOperties of air-classified protein and starch fractionns fran eight legure flours. J. Amer. Oil Chem. Soc. 56:292. Sterling, C. 1978. Textural qualities and molecular structure of starch products. J. Texture Studies 9:225. Sumner, A.K., Whalley, L.L., Blankenagel, G., and Youngs, C.G. 1979. Storage stability studies on pea flour, protein concentrate and starch. Can. Inst. Food Sci. Teclmnol.’J. 12:51. Suzuki, K., Kubota, K., Qmichi, M., and Hosaka, H. 1976. Kinetic studies on cooking of rice. J. Food Sci. 41:1180. Tappel, A.L. 1955. Studies of the mechanism of vitamin E action. III. In vitro c0polymerization of oxidized fats with protein. Arch. Biochem. Biophys. 54:266. Tarladgis, B.G., Watts, B.M., and Younathan, M.T. 1959. A distillation method for the quantitative determination of malonaldehyde in rancid foods. J. Amer. Oil Chenn.Soc. 37:44. Tyler, R.T., Youngs, C.G., and Sosulski, F.W. 1981. Air-classification of legumes. I. Separation efficiency, yield, and camposition of the starch and protein fractions. Cereal Chen. 58:144. Uebersax, M.A., Zabik, M.E., Lee, J.P., and Defouv, C.L. 1980. Characterization and utilization of chy heated navy bean flour. 1981 Res. Rpt. presented to USDA, Res. Agreement No. 59-2481-0-2— 002-0, August 28. 167 Uebersac, M.A., Zabik, M.E., Lee, J.P., and Tecklenbtm‘g, E.J. 1983. Res. Rept. presented to USDA, Res. Agreement No. 59-2261-1—2- 004-0, D'ecenber 31. Van Wazer, J.R., Lyons, J.W., Kim, K.Y., and Colwell, R.B. 1963. "Viscosity and Flow Measurement: A laboratory Handbook of Rheology, ' Interscience Publishers, New York, NY. Vitali, A.A. and Rao, M.A. 1982. Flow behavior of guava puree as a function of temperature and concentration. J. Texture Studies 13:275. Vose, J .R. 1977. Functional characteristics of an intermediate amylose starch from smooth-seeded field peas compared with corn and wheat starches. Cereal Chem. 54:1141. Vose, J .R. 1980. Production and functionality of starches and protein isolates fran legume seeds (field peas and horsebeans) . Cereal Chen. 57:406. Vose, J.R., Basterrechea, M.J., Gorin, P.A.J., Finlayson, A.J., and Yourngs, C.G. 1976. Air-classification of field peas and horse- bean flours: chemical studies of starch and protein fractions . Cereal Chen. 53:928. Vratanina, D.L. and Zabin, M.E. 1978. Dietary fiber sources for baked products: bran in sugar-snap cookies. J. Food Sci. 43: 1590. Wagner, J.R., Carlson, J.F., Becker, R., Gunbamann, M.R., and Dubinoff, LE. 1977. Camparative flatulence activity of beans and bean fractions for man and the rat. J. Nutr. 107:680. Waldt, L.M. and Kehoe, D. 1959. Starch chemistry for the food tech- nologist. Food Technol. 13:1. Watsonn, J.W., McEwen, T.J., and Bushuk, W. 1975. Note on any milling of faba beans. Cereal Chen. 52:272. Williamson, R.V. and Heckert, W.W. 1931. Sane properties of disper- sions of the quicksand type. Ind. Eng. Chem. 23:667. - Wolf, M., Walker, J.E., and Kapsalis, J.G. 1972. Water vapor sorptionn hysteresis in dehydrated food. J. Agr. Food Chem. 20:1073.