:HESli This is to certify that the thesis entitled CHARACI'ERIZATIG‘I OF NAVY, PINIOAND BLACKBEANSTARCHES presented by Naruemon Srisuma has been accepted towards fulfillment of the requirements for Master of Science degree in Food Science //7MI flL/(61é294j/ Major professor Date April 10, 1984 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution ill/lliii.iiilliililiil L 1293 7 8762 )V153I_J RETURNING MATERIALS: Place in book drop to LJBRAfiJES remove this checkout from .—c—- your record. FINES will be charged if book is returned after the date stamped below. -o".“ I .1” 2 3 fig), 3 nwmaim: CHARACTERIZATION OF NAVY, PINIO AND BLACK BEAN SEARCHES By Naruemon Srisuna A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASI‘EROF SCIENCE Department of Food Science and Hanan Nutrition 1984 ABSTRACT CliARACI'ERIZATIONOFNAVY,P1NlDANDBLACKBEANSIIARQ-IES By Naruemon Srisuma Pasting characteristics of air-classified high starch flours and purified starches from navy, pinto and black beans were determined using a Brabender Viscoamylograph. Restricted swelling characteristics , showing increased viscosity with no peak during heating and sane tendency to retrograde during cooling, were observed. Navy and black bean starch pasting curves were similar and more viscous than those of pinto starch; pH variation (4 to 8) produced only minor effects on pasting properties. Extended high temperature holding (95°C, 30 min) resulted in steady viscosity increase without breakdom. Extended low tanperature holding (50°C, 30 min) produced minimal shear breakdown of purified starches; air-classified flours were less stable. Gel permeation high performance liquid chromatography (HPLC) of bean starches produced similar elution patterns for all three bean types. After hydrolysis with beta amylase, the elution patterns of navy and black bean starches remained similar. Elution pattern differences of pinto bean starch hydrolysates supported pasting characteristic differences . To my parents and family for their love , patience and moral support ii I wish to express my deepest gratitude to my najor professor, Dr. Mark A. Uebersax, for his thoughtful guidance, constant encour- agement, and valuable suggestions throughout the course of my study at Michigan State University. Appreciation is also extended to Dr. George L. Hosfield, Dr. James F. Steffe, and Dr. Mary E. Zabik for their serving as merrbers of the guidance coumittee. I sincerely thank Ruengsakulrach Songyos and Titiranonda Araya for their help in laboratory work and thesis preparation. I also thank my Thai friends at Michigan State University, especially Arayaprecha Worraporn and Suvarnamani Anintita for their encouragement and moral support. Further, I thank friends in the Department of Food Science and Human Nutrition, particularly Donald G. Coffey, John W. Larkin, James J. Lee, Richard A. Toukinson, and Janet L. Wilson, for their assistance which made the coupletion of this work possible. I gratefully acknowledge Mr. and Mrs. Cady, my dear host family, and Dr. Beverly B. Cozort and her family for their continuing support and the beautiful experiences they provided during my stay in the United States. Above all, my special gratitude, appreciation, and love go to my wonderful parents and family for their love , understanding, and unending encouragement . iii TABLEOFCQTI'EN'I‘S LISI‘ OF TABLES LIST OF FIGURES ...................... LITERATURE REVIEW ..................... Dry Beans As A Food Resource ............... Description of Beans ................ Bean Handling and Storage .............. Moisture Content ................. Antinutritional Factors and Flatulence Production .................... Bean Flour and Bean Starch ................ Physical-Chanical Properties ............ Granule Size, Shape, and Microscopic Appearance .................... mlecular Weights ................. \OOQVO‘O‘U'I-L‘bwww Amylose and Amylopectin .............. Granule Crystallinity ............... Functional Properties ................ Swelling and Solubility .............. Gelatinization and Pasting ............ Starch Analysis ................... Extraction .................... Quantitation ................... Preparation of Samples ............... Air-Classified High Starch Flours ......... Pm'ified Starch .................. Physico—Chemical Analyses ................ Analytical Methods ................. Moisture Amylose Pasting Characteristics ............... Pblecular Structure ................. 21 22 22 22 23 26 26 26 31 31 31 31 31 33 35 35 35 35 36 36 36 39 40 40 41 Alpha and Beta Amylase Hydrolysates ....... Statistical Analyses .................. RESULTS AND DISCUSSION .................. Chanical Canposition .................. Proutimate Composition Analyses .......... Air-Classified High Starch Flours ........ Purified Starches ................ Sugar Analyses .................. Air-Classified High Starch Flours ........ Purified Starches ................ Brabender Pas ting Characteristics ........... Molecular Structure .................. Analysis of Variance of Pasting Characteristics of Air-Classified High Starch Flours and Purified Starches from Navy, Pinto and Black Beans at Various Concentrations ................. Analysis of Variance of Pasting Characteristics of Air-Classified High Starch Flours fran Navy, Pinto and Black Beans at Different pH Values of Phosphate Buffer Solution ............... Analysis of Variance of Pasting Characteristics of Purified Starches fran Navy, Pinto and Black 41 42 43 43 43 43 45 47 47 47 49 80 88 90 91 91 92 92 93 Beans at Different pH Values of Phosphate Buffer Solution ..................... APPENDIXD ........................ Analysis of Variance of Pasting Characteristics of Air-Classified High Starch Flours from Navy, Pinto and Black Beans at Different High or Low Tanperature Holding Periods ............... APPENDIX E ........................ Analysis of Variance of Pasting Characteristics of Purified Starches from Navy, Pinto and Black Beans at Different High or Low Tanperature Holding Periods ..................... vii 93 94 94 95 95 96 Table LISI‘ OF TABLES Functional and molecular properties associated with pasting characteristics of starches ...... Gelatinization temperature of legune starches Proximate canposition of air-classified high starch flom:s and purified starches fran navy, pinto and black beans ............... Percent sugar content (glucose , sucrose, inos itol , raffinose, and stachyose) of air-classified high starch flours and purified starches fran navy, pinto and black beans ............... Pasting characteristics of air-classified high starch flours and purified starches fran navy, pinto and black beans at various concmtrations in pH 5 . 3 phosphate buffer solution ........ Pasting characteristics of air-classified high starch flours (10 g/100 ml) and purified starches 8 g/100 ml) frcm navy, pinto and black beans at different pH values of phosphate buffer solution Pasting characteristics of air-classified high starch flours (10 g/100 ml) and purified starches (8 g/100 ml) fran navy, pinto and black beans at different high or low tamerature holding periods in pH 5 . 3 phosphate buffer solution ........ Analysis of variance of pasting characteristics of air-classified high starch flours and purified starches fran navy, pinto and black beans at various concentrations ............... Analysis of variance of pasting characteristics of air-classified high starch flours from navy, pinto and black beans at different pH values of phosphate buffer solution ............. viii Page 20 24 44 48 51 55 74 91 92 Table 10. 11. 12. Analysis of variance of pasting characteristics of purified starches fran navy, pinto and black beans at different pH values of phosphate buffer solution .................. Analysis of variance of pasting characteristics of air-classified high starch flours from navy, pinto and black beans at different high or low temperature holding periods ............ Analysis of variance of pasting characteristics of purified starches from navy, pinto and black beans at different high or low temperature holding periods .................. Page 93 94 95 LIST OF FIGURES Figure Schematic diagram of the particle— to-particle dry roaster .................... Flow diagram of processing navy, pinto and black beans .................... Brabender pasting curves (standard conditions) of air-classified navy, pinto and black bean flours at 10 g/100 ml in pH 5.3 phosphate buffer solution .................. Brabender pasting curves (standard conditions) of purified navy, pinto and black bean starches at 8 g/ 100 ml in pH 5.3 phosphate buffer solution ...................... Brabender pasting curves (standard conditions) of air-classified navy bean flour at various concentrations in pH 5. 3 phosphate buffer solution ...................... Brabmder pasting curves (standard conditions) of air-classified pinto bean flour at various concentrations in pH 5 . 3 phosphate buffer solution ...................... Brabender pasting curves (standard conditions) of air-classified black bean flour at various concentrations in pH 5 . 3 phosphate buffer solution ...................... Brabender pasting curves (standard conditions) of purified navy bean starch at various concentrations in pH 5 . 3 phosphate buffer solution ...................... Brabender pasting curves (standard conditions) of purified pinto bean starch at various concentrations in pH 5 . 3 phosphate buffer solution ...................... Page 32 34 50 54 59 60 61 63 64 Figmre 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. Brabender pasting curves (staidard conditions) of purified black beau starch at various concentrations in pH 5 . 3 phosphate buffer solution ...................... Brabender pasting curves (standard conditions) of air-classified navy bean flour at different pH values of phosphate buffer solution at 10 g/ 100 m1 concentration ................ Brabender pasting curves (staidard conditions) of air-classified pinto beai flour at different pH values of phosphate buffer solution at 10 g/ 100 m1 concentration ................ Brabender pasting curves (standard conditions) of air-classified black bean flour at different pH values of phosphate buffer solution at 10 g/ 100 m1 concentration ................ Brabender pasting curves (staidard conditions) of purified navy bea'i starch at different pH values of phosphate buffer solution at 8 g/ 100 m1 concentration ................ Brabender pasting curves (staidard conditions) of purified pinto bean starch at different pH values of phosphate buffer solution at 8 g/ 100 m1 concentration ................ Brabender pasting curves (standard conditions) of purified black bean starch at different pH values of phosphate buffer solution at 8 g/ 100 m1 concentration ................ Brabender pasting curves of air-classified high starch flours (10 g/100 ml) at an extended high temperature (95°C) holding period (30 min) in pH 5 . 3 phosphate buffer solution .......... Brabender pasting curves of purified beam starches (8 g/ 100 ml) at an extended high temperature (95°C) holding period (30 min) in pH 5.3 phosphate buffer solution ................... Brabender pasting curves of air-classified high starch flours (10 g/100 ml) at an extended low temperature (50°C) holding period (30 min) in pH 5 . 3 phosphate buffer solution .......... Page 65 66 67 68 70 71 72 73 77 78 Figure 20. 21. 22. 23. 24. Brabender pasting curves of purified beam starches (8 g/100 ml) at a1 extended low temperatmre (50°C) holding period (30 min) in pH 5.3 phosphate buffer solution ...................... Elution profiles of stmdard components on micro-BCl‘lDAGEL gel permeation colums ........ Elution profiles of purified navy bean starch (control) aid their products hydrolyzed by alpha and beta amylases on micro-BONDAGE. gel permeation colums ................. Elution profiles of purified pinto bea'i starch (control) aid their products hydrolyzed by alpha and beta amylases on micro-m gel permeation colums ............... Elution profiles of purified black beai starch (control) aid their products hychrolyzed by alpha and beta anylases on micro-mm. gel permeation colums ................. Page 79 81 83 84 85 INTRODUCTION Dry edible beans (Phaseolus vulgaris L.), as well as other legunes, constitute traditional foods for populations of subtropical aud tropical areas. They provide significant amunts of protein, calories, vitanins aud minerals. Their primry importance is as an inexpensive source of protein for humn consumption (FAQ, 1977) , particularly for the lower- income groups in many developing countries of the world. Althouglu they are nutritious aud inexpensive, the per capita consmiption of dry beans in the United States has declined from about 8.4 1b in 1940 to 4.1 lb in 1981 (USDA, 1981). Several reasons for declining consumption have been proposed, including unavailability of bean-based food ingredients , the presence of antinutritional factors , flatulence production, and especially, time consmn’ng preparation due to long soaking aud cooking requirements . Several researchers have studied these problems to find innovative alternatives to increase beau utilization. For example, whole beau flour was air-classified into high protein aud high starch fractions (Vose et a1. , 1976) . These fractions have unique functional aud sensory characteristics which make them useful as ingredients in specific foods or as nutritional supplements. The protein quality aud digestibility (Charley, 1970; Bressani, 1975; Tobin aud Carpenter, 1978; Patel et al. , 1980; Carpenter, 1981; Sahasrabudhe et al. , 1981), the flatulence producing factors (Galloway et al. , 1971; Olson et al. , 1975; Wagner et al. , 1976; Fleming, 1 2 1981a,b; Flaming aud Reichert, 1983; Sathe et a1. , 1983), aud the antinutritional factors (Honavar et a1. , 1962; Liener, 1962; Kakade et al., 1969; Liener, 1975; Johnson et al. , 1980; Ellenrieder et a1. , 1980 aud 1981; Chang aud Tsen, 1981) have also received considerable study in order to improve the nutritional quality of dry beaus. On the other hand, the starches have been somewhat neglected. Dry beais contain up to 602 carbohydrates (Adams, 1972), composed mainly of starch. Quly in recent years have detailed investigations of the characteristics aud functional properties of beau starch been conducted (Naivikul aud D'Appolonia, 1979; Salimath and Tharanathan, 1982; Reddy et a1. , 1984). Nevertheless, further studies on beau starch are still needed to provide more useful technical data for beau product development aud industrial applications. This study was undertaken to determine the starch characteristics of navy, pinto and black beans . The purified starches were prepared fran air-classified higlu starch bean flours. The general chemical analyses aud the sugar and amylose contents of the air-classified flours aud purified starches were determined. The pasting character- istics of these starches at varying pH values, holding periods aud concentrations were also studied. Finally, the molecular structural characteristics of bean starches were investigated by using the hychc’o- 1ytic enzymes, alpha aud beta amylases, in conjunction with gel permeation high performance liquid chranatography (HPLC) . LITERATUREREVIEW DryBeausAsAFoodResource Description of Beaus legumes are the edible dicotyledonous seed of legmuinous plauts that belong to the Legwmlnosae family. These plauts commonly grow in a wide range of conditions aud climates. The family Leguminosae is second only to Gramineae in importance as a source of food aid fodder. The Phaseoleae subfamily of the Leguminosae contains 47 genera including Phaseolus. The genus Phaseolus comprises about 200 species of which only 20 are cultivated for their edible pods aud seeds. Navy, pinto and black beans are major cannercial beans gram in Michigau aud are classified as Phaseolus vulgaris L. (Deschamps, 1958). Navy beans are a mid-season bush type plaut with white flowers. The seeds are chalky white, roundish to ovoid in shape, weighing in the range of 17 g to 19 g per 100 seeds. Prevalent varieties in Michigau are Sauilac, Gratiot, Seafarer, Kentwood, Fleetwood, Charity, Upland, Snow Flake aud Snow Bunting. Pinto beans are a vine type plaut. The seeds are medium sized, irregular and flat with brown variegated pattern against light tan background. The major variety is Pinto 114. Black beans are a high yielding mid-to—full season plaut with small seeded, purple flowers on au erect vine. Prevalent varieties are Black Turtle Soup, Sanfemaudo, Icapijao, Jamapa aud Tanazulapa. 3 Beau Handling aud Storage The handling aud storage of beaus in the field critically affect the ultimate quality of beaus aud beau products. The moisture content is the most importaut consideration following harvest. Seeds with a moisture content above 181 are subjected to excessive damage during storage aud then in the processing line, due to physical susceptibility to mechanical forces and microbial spoilage fran mold (Weston and Murris, 1954). On the other haud, seeds with a moisture content below 152 are sensitive to impact damage. Edible dry beans have a thin seed coat layer which is easily damaged in handling or drying. Belt conveyors are usually used to move beam into and out of storage in order to reduce the damage. Dry beaus cau be stored in wood, concrete or steel bins. The choice depends upon the quantity of beaus and the processing or drying that might be done in the field or in existing storage structures. Maddex (1978) suggested not using existing silos because filling equipment for loading beaus into a silo usually causes beau damage. He recommended flat—bottomed, overhead wooden bins, round steel bins (recently the most popular form of storage) or concrete bins equipped with au aeration system. During storage , seed deterioration cau be retarded by providing proper storage conditions. The two major controlling factors are moisture content and temperature . Moisture Content. Beaus stored at too low moisture exhibit clmping aud splitting due to seed coat aud cotyledon rupture, while storage at high initial moisture encourages discoloration, off-flavor development, loss of water uptake capacity aud mold growth. Morris and Wood (1956) 5 reported that beaus with moisture content above 132 deteriorated significantly in both flavor and texture after six months at 77°F and became unpalatable within 12 months. lbrris (1963 aud 1964), Burr et al. , (1968) and Bedford (1972) reported that beaus stored at higlu moisture showed a significaut increase in their required cooking time while low moisture beans did not lose their cooking quality. Temperature. Quality degradation is faster at high temperature than at low tarperature. Beans stored at high temperatures get darker in color aud require longer cooking times. The deteriorative effect of high moisture on beau quality is increased by high temperature. long cooking time of beans fram high temperature storage was observed by Dawson et al. (1952), Pbrris (1963) aud Burr et a1. (1968). Harris (1964) stated that the reduction of 15°F in storage temperature had the sane effect as the decrease of 0.67. moisture content to yield au equiva- lent short cooking time. Uebersax (1972) reported that deterioration rate both in discoloration and mold growth was minimized in beans stored at 55°F under relative humidities ranging from 752 to 86%. The influence of increased storage temperature became greater at higher relative lunidity. Vongsarrpigoon et a1. (1981) suggested that optimum beau quality was obtained from dry beaus stored at 142 moisture at 70°F. Hard-to- cook phenomenon cal develop due to improper storage conditions (Norris aud Wood, 1956; Mmeta, 1964) . Recommendations to prevent storage loss of dry beans fran hardening (Mejia, 1980) include: 1) beaus should be stored at the lowest possible moisture content aud 2) beaus should be stored in a dry aud cool environment. Aeration with the proper flow rate , relative humidity aud temperature improve the stored beau quality. 6 Aeration, which is the practice of moving air at low flow rates to cool all beans in a bin, prevents moisture migration and also reduces mold growth and development of musty odors and off flavors (Maddex, 1978). Vongsarnpigoon et al. (1981) observed that organic acid treatment provided limited mold inhibitian and resulted in beans with brown discol- oration aud firm texture , whereas NaHSO3 treatment provided limited color stability without adversely influencing processing characteristics. In addition, they reported that vacuum and CO2 storage did not significantly improve beau quality. Beau Canposition Legumeshavebeeuusedas themainsairce of proteinandcalories in many developing countries. Besides the major constituents of proteins, carbohydrates and lipids, all grain legumes contain substantial quantities of minerals and vitamins. Their chemical and nutritional composition vary with genetic as well as environmental conditions under which the crop is grown. _;Pr_c_>_t_<-2_in. Beaus have a high protein content , which is twice greater than cereals, ranging from 172 to 252 on a dry weight basis (Brenner, 1965; Goodhart and Shils, 1980). However, it is not generally recognized that beans contain proteins equivalent to milk, beef and other animal proteins in respect to their content of essential amino acids and their availability via digestion. Beau proteins are relatively rich in essential amino acids, particularly lysine and threornine (Patel et al. , 1980; Carpenter, 1981; Sahasrabudhe et al., 1981), whereas cereals are deficient in lysine (Sathe et al. , 1981). In additionu to having twice as much protein, beans are also better sarrces of isoleucine, leucine, 7 phenylalanine, and valine than are cereals (Charley, 1970; Bressani, 1975; Tobin and Carpenter, 1978). They are deficient, however, in sulfur amino acids , particularly methionine and cystine (Patel et a1. , 1980; Sahasrabudlne et al. , 1981). Tobin and Carpenter (1978) reviewed 47 papers on the amino acid composition of dry beans. Thirty-one papers describe the protein quality as a result of studies with rats and humans . This research has indicated that methiauine is always the first limiting amnino acid. In addition, this research has demonstrated the low digestibility of bean protein (762) . This was confirmed by Graham et a1. (1979) and also by other studies where beans have been fed in mixtures. Bressani (1975) and Kay (1979) reported that the predaminant class of proteins present in Phaseolus beans is salt soluble globulins of which three distinct proteins have been identified: phaseolin, phaselin and conphaseolin. Liener and Thompson (1980) , Geervani and Theophilus (1982) and Sathe et al. (1981) reported that the Great Northern (Phaseon vu Zgaris L.) proteins , albumins and protein isolates were characterized by high acidic amnino acid content , while globulins and protein concentrates had a high proportion of hydrophobic amino acids . They also found that the bean proteins were resistant to in vitro enzymatic attack; however, heating improved in vitro susceptibility to enzymatic hydrolysis . Carbol'gdrates . The total carbohydrates of dry beans range fram 242 to 682. These carbohydrates include mono- and oligosaccharides, starcln and other polysaccharides . Starch is the most abumdaut legume carbo— hydrate and varies from 247.’ in peas to 56.5% in beans. The variations observed are due to different cultivars and analytical procedures (Pritchard et al. , 1973; Cerning-Beroard et al. , 1975). 8 Total angers (mono- and oligosaccharides) represent only a small percentage of total carbohydrates in dry legume seeds . Among the sugars, oligosaccharides of the raffinose family (raffinose, stachyose, verbascose and ajugose) predominate in most legumes and accamnt for a significant percentage (31.12 to 762) of the total sugars in several others (Neue et al., 1975; Hymowitz et al. , 1972; Cerning-Beroard and Filiatre, 1976; Naivikul and D'Appolonia, 1978; Becker et al., 1974; Ron, 1979; Rocklaud et al., 1979; Akpapunamn and Parkakis, 1979; kaenyong and Borchers, 1980; Reddy and Salunkhe, 1980; Fleming, 1981 a, b; Sathe and Salunkhe, 1981a). Walker and Hymowitz (1972) famd that the sugar content ranged from 4.42 to 9.22 in the 28 varieties of Phaseolus vulgaris. Of total sugars present, sucrose accannted for 46.42; raffinose, 10.42; and stachyose, 432. The predominance of a particular oligosaccharide seems to depend on the type of legume. For example, verbascose is a major. oligosaccharide in black gram, red gram, mung bean and broad beans (faba beans), and stachyose is the major oligosaccharide in smooth and wrinkled peas, Great Northern beans, California small white beans, red kidney beans, navy beans, pinto beans, pink beans, black eye beans, Bengal gram, soybeans, lentils, cowpeas and lupine seeds. Crude fiber consists of cellulose, hemicellulose (a heterogeneous group in which peutosans usually predaminate) , lignin (an aromatic polymer), pectin and cutin substances. legumes contain appreciable amounts of crude fiber (1.22 to 13.52). Rather large variations in crude fiber content have been observed in black gram, Bengal gran, mung bean and red gram. Cellulose is the major component of crude fiber in smooth and wrinkled peas, red kidney beans, navy beans, pinto beans, pink beans and black eye beans, while in other legumes (lupine seeds, 9 lentils, broad beans, red gran, black gram) , hennicellulose is the major campanent of fiber. Deschanps (1958), Tobin and Carpenter (1978), and Kay (1979) reported that the crude fiber content of the common varieties of the genus Phaseolus was less than 62. Preliminary evidence suggests that dietary fiber may contribute positively to health and well-being (Burkitt, 1971; Painter and Burkitt, 1971; Trowell, 1972). _L_iLid_§. The highly unsaturated lipid content of dry beans is generally in the range of 12 to 22. Deschamps (1958) and Watt and Merrill (1963) reported that Phaseolus vulgaris shows a low fat content, generally below 22 with 63. 32 being unsaturated fatty acids. Vitamnins. Dry edible beans provide some water soluble vitamins: thiamine, riboflavin, niacin and folic acid, but very little ascorbic acid (Watt and Merrill, 1963; Forham et al. , 1975; Tobin and Carpenter, 1978). Cannon commercial methods of preparation of canned beans cause a significant loss of water soluble vitamins. Therefore, nnany workers have studied the retention values of water soluble vitamins in order to optimnize the quality of bean products (Augustin et al. , 1981; Carpenter, 1981) . There is no evidence in the literature which indicates that dry beans contain appreciable amounts of fat soluble vitamins. Watt and lbrrill (1963) and Kay (1979) reported that Phaseolus vulgaris provides less than 30 International Units of vitamin A per 100 grams of raw beans. Variability of vitamnin content is higln. Augustin et a1. (1981) suggested that geographic location of growth appeared to have had a significant effect on this content. Ash and Minerals. The total ash content of Phaseolus vulgaris ranges frcm 3.52 to 4.12 (Fordham et al., 1975; Tobin and Carpenter, 1978; Kay, 1979). Beaus are generally considered to be a good sarce of 10 some mninerals, such as calcium and iran, but they also contain signif- icant amournts of phosphorus and potassium. Adams (1972) and Patel et al. (1980) observed that navy beau flair had 2 to 17 times as many minerals as wheat flar. The specific nnineral content in mature, raw legumes has been reported by several researchers in recent years; however, most values show large variability. Augustin et al. (1981) pointed ant that been classes and environmental factors greatly influence this variability. Antinutritional Factors and Flatulence Production. The two primary concerns with the use of legumes are the presence of autinutritional factors and flatulence productiau. Beans contain a number of "anti- nutrients" and potentially toxic substances. Bressani ( 1975) categorized the toxic substances present in legumes into seven groups: trypsin inhibitors , hemagglutinins or lectins , goitrogeuic factors , cyanogeuic glucosides, lathyric factors, compamnds that cause favism, and other factors about which less information is known. Of these factors, trypsin inhibitors and hemagglutinins are considered primarily responsible for causing the growth retardation observed in laboratory animals fed raw beans (Honavar et al., 1962; Liener, 1962; Kakade and Evans, 1965, 1966; Jaffe, 1969; Liener and Kakade, 1969; Liener, 1975; Liener and Thompson, 1980) . Trypsin inhibitors are famd in several legumes including the species of Phaseolus. As the name implies, they are protein fractions that strongly inhibit the enzymatic activity of trypsin in the intestine (Genes et a1. , 1979) , thereby reducing the digestiau of proteins and hence, the absorption of their constituent amnino acids. Besides the inhibition of the enzyme trypsin, the poor digestibility of been proteins may result from the refractory nature of native bean protein to enzyme trypsin (Liener, 1975). Fortunately, soaking (Kakade and Evans, 1966; 11 Tyler et al. , 1981) or germnination (Gupta and Wagle, 1980) plus heating (Ellenrieder et al. , 1980 and 1981; Johnson et al. , 1980; Chang and Tsen, 1981) have been fannd to improve the digestibility of beau protein in two ways: 1) by denaturation which makes the proteins more susceptible to the action of enzymes, and 2) by destroying trypsin inhibitors. Liener (1962 and 1975) and Liener and Thanpson (1980) also reported that the protein of unheated beans resists proteolysis in the intestine; however, after heating, the true digestibility increases and trypsin inhibitory activity decreases . Hemagglutinins are also present in beans and are destroyed by heat (Thompson et a1. , 1983) . These compamds caise agglutination of red cells and impair absorption (Liener, 1975). Kakade and Evans (1966) fannd that merely soaking the beans did not result in any loss of hennagglutinating activity. Aguilera et al. (1981 and 1982) proposed that special grinding control during processing may aid inactivation and reduction of both heragglutinin and trypsin activity. Some evidence that seed coat color may affect protein digestibility has been offered (Parquardt et al. , 1978; Elias et al., 1979), such that the protein in white- coated dry beans showed greater digestibility than that of black-coated dry beans. Extensive chemical analyses, including trypsin inhibitor activity, hemagglutinin concentrationu, and amino acid analysis showed both black and white dry beans to be essentially alike. This suggested that the pigments of the seed coat were responsible for the decreased protein digestibility. These pignents are tannins, which could react with bean proteins, decreasing their digestibility. Tannins inhibit the digestive enzymes by com- plexing the enzymes (Goldstein and Swain, 1965; Haslen, 1974; Griffith and Jones, 1977) and also with protein substrates. In vitro studies have 12 indicated that isolated tannins have higlu inhibitory action against digestive enzymes (Pbrris and Wood, 1956 ; Milic et a1. , 1972). It has been shown that high tannin content reduces the nutritional value , and at enzyme inhibition, is associated with grain that has a broom or colored pericarp. Protein binding by tannins (polyphenols) is believed to be responsible for the growth depression observed in rats and chickens. In addition, legmes contain relatively large amounts of phytic acid, up to 52 by weight (DeBoland et al. , 1975). This compamd can effectively bind nutritionally important minerals, reducing their availablility for absorption (Sathe and Kristmnamurthy, 1953; O'Dell and Savage, 1960; Roberts and Yudkin, 1960; O'Dell et al. , 1972; Davies and Nightingale, 1975; Maga, 1982). The ability of legume seeds to stimulate intestinal gas formatian (flatulence production) is another factor limiting their consumption by humans (Protein Advisory Group, 1973). It has frequently been suggested that the galactose- containing oligosaccharides including raffinose, stachyose and verbascose are the components in the legmne seeds which are respansible for flatulence (Fleming, 1981a; Fleming and Reichert, 1983; Sathe et al. , 1983). These sugars have been considered fermentable since the alpha galactosidase enzyme, althangh not produced by the lnuman digestive system, is secreted by the indigenais gut microflora. Therefore, these substrates are expected to be fermented within the large intestine and hence, caise gas pro- duction (Galloway et al., 1971 ; Olson et al. , 1975 ; Wagner et al. , 1976). Fleming (1981b) studied the flatus potential of seven types of legume seeds and reported that hydrogen production was significantly and positively correlated to oligosaccharides and acid hydrolyzable 13 pentosan contents , but negatively correlated to the starch and lignin contents . In view of increasing dry beau utilization and nutritional sig- nificance, greater research efforts should be directed to understanding and minimizing the effects of these autinutritional and flatulence- prodncing factors . Bean Flour and Bean Starch Dried beans traditionally have been utilized by soaking and cooking in the heme or consumed as cannercially processed canned beans. Many researchers have searched for innovative alternatives to increase bean consummation. A shelf—stable flour has been proposed as a food ingredient to provide increased versatility and to improve bean utilization. Initial efforts to produce beau ingredients were aimed at keeping cell rupture at a minimum in order to retain the same texture, appearance and taste as conventionally prepared bean saips. Instant precooked bean powders have been prepared by soaking, cooking, slurrying, and drum or spray drying (Bakker et al. , 1973). The second generation of beau ingredients are based on fractionation of the two major canponents , starch and protein. Chang and Satterlee (1979) prodiced bean protein concentrates containing 722 to 812 protein by wet processing using water extraction techniques. Molina and Bressani (1973) prepared protein isolates containing about 902 protein and starch products with almost 502 starch. Several disadvantages exist in using wet processing for fractionation and concentrationn: 1) significant energy is used in drying the final prodicts; 2) waste by—prodncts are produced which contain significant amounts of organic matter; and 3) yields are reduced by the losses in wet by-prodnct streams. 14 In recent years, dry milling of beans into whole bean flar and air-classifying into high starch and high protein fractionus have received increased interest as efficient processing methods. Compartmeutalization of protein in friable protein bodies, and starch in denser, less brittle granules , enables size reduction and fractionation of dry beaus into a coarse fraction which contains most of the starch and a fine fraction which contains most of the protein. Because of size distribution, sieving, a simple fractionation technique, had been proposed. However, it becomes progressively more difficult as the size separation point is reduced. Very fine materials have a tendency to agglcmerate especially if oil is present. Such powders tend to build-up on the screen causing blinding of the sieve and a loss of separation efficiency. Air-classi- fication has been applied to wheat (Stringfellow and Peplinski, 1964); to corn, sorghum and soy (Pfeifer et al. , 1960); to rice (Stringfellow et al. , 1961); to barley (Pomerauz et al. , 1971); and to triticale flours (Stringfellow et al., 1976). Air-classification works on the principle of difference in terminal settling velocity anong different particle sizes. It has been observed that, when raw legumes are gramd without any pretreatments , they develOp undesirable odors and flavors . Lipoxidase has been held respansible for the appearance of off-flavors by catalyzing formation of hydrogen peroxides fram unsaturated fatty acids (Kon et a1 . , 1970) . However, treatment with dry heat for six to eight minutes at 104°C to 105°C cenpletely inactivates this enzyme (Smith and Circle, 1972) . Aguilera et al. (1982) fanud that heat treatment also destroys trypsin inhibitor and henagglutinin activities. Therefore, it is advantageaus to auply a heat treatment to beans prior to been flar production. 15 The conventional heating method, using air as the heat transfer medium, is attractive due to ease of handling and lack of product contamnination. However, air does not give a high rate of heat transfer. Khan (1972) reported higln tenperature dying of paddy rice by using heated sand. Harper and Iorenz (1974) prepared the full-fat soy flar by roasting soy beans in a bed of heated salt. Similarly, Carvalho et al. (1977) produced instant navy bean powders by roasting dry beans in a bed of salt, then grinding. They also reported a reduction of trypsin inhibitory activity by this method. In 1978, Peterson and Harper developed the new heated granular bed roaster. Based on this previaus work, Aguilera et al. (1982) developed a new particle-to-particle heat exchanger for roasting navy beans and observed the functional character- istics of several roasted bean products. These roasted products showed reduced water- soluble nitrogen content, gel forming capacity, trypsin inhibitor and hexagglutinin activities , increased water holding capacity and cold paste viscosities, and no changes in available lysine and degree of starch damage. In additiau, roasting causes fracture and separation of hulls, and facilitates their removal. Dry beaus contain up to 602 carbohydrates, most of which is starch. Chemically, bean starch is similar to starches frcm other sarces (Staudinger, 1932) . However, their physico-chemical properties and internal molecular structures are different depending on the original sarce, maturation and environmental factors (Rosenthal and Takeko, 1972) . Generally, starch is defined as a polymer of D—glucose units. It is canposed of two different polymers, a linear ccmpamd, amylose, and a branched carponent, amylopectin. In the linear fraction, the glucose units are joined exclusively by ou-l,4 glucosidic bonds. The number of 16 glucose units may range from a few hundred to several thousand. The amylose stains bright blue with iodine. Rundle and French (1943) deuonstrated that the polymer chain takes the form of a helix which may form inclusion compoumuds with a variety of materials, such as iodine (Harris et al., 1928; Hanes, 1937; Freudenberg et al., 1939). ‘ Amylo- pectin is a branched molecule cantaining both uni-1,4 aud uni-1,6 glucosidic bonds and does not stain blue with iodine. Because it is a branched polymer, various workers atteupted to describe the molecular structure of amylopectin. In 1937, I-laworth et al. proposed the "laminated" structure, i.e., the ratio of A and B chains is l:(n-Z), where A chains are attached to the macromolecule by a single linkage frcm the potential reducing-gran), Bchains are linkedto twoormoreother chains, andn isthetotalnumberofchains. Intlnesaneyear, Standingeraud Husemann suggested the "curb" structure. All of the chains except one were A chains. layer (1952) originated the "tree type" structure. Helen (1970) later revised the Mayer structure such that the A:B ratio was approximately 1:1. A cluster-type structure was originally prOposed by Nikuni (1969) for starch molecules; tlnen independently, for the amylo- pectin ccmponeut by French (1972), and later by Robin et a1. (1974). Fire recently, an extended cluster model has been proposed by Yamaguchi et al. (1979). Manners and Matheson (1981) supported the structure of the cluster-type because it was more in accord with the physical properties, the observed structural features , and node of biosynthesis within the starch granule than were other hypothesized models. The linear (amnylose) and branched (amylopectin) molecules are linked together by hydrogen bonds and form a micellar network. The starch granules are ccmpletely insoluble in cold water; however, they ednibit a limited capacity for absorbing cold water and swelling reversibly. l7 Mueu heat or appropriate chemicals are applied to disrupt the hydrogen bands that link the molecule in the presence of surplus water, the hydration of the network is encouraged, and an irreversible process of swelling is started. The critical tenperature at which this process starts is known as gelatinizatian tenperature, which is a characteristic for starches of different origins . Attaining the gelatinization tenper- ature range with subsequent swelling of the granules is accompanied by an increase in viscosity. This phenomenon may be explained by a greater chance of swollen granules coming into contact with each other, which probably cauld be considered a measure of the work required to move the granules past each other as they cantinue to swell (Schoch, 1965 and 1969). In 1973, Miller and co-workers, however, indicated that granule swelling does not account for the rapid rise in viscosity of a wheat starch suspensioun heated in excess water, but that the exudate principally causes the large increase in viscosity at higher tenperatures. They indicated that a correlation between starch paste viscosity and granule size was almost nonexistent. This exudate contains colloidally and molecularly dispersed starch molecules leached frcm the granules into the surramding aqueous phase dring the initial free swelling stage. As suggested by Payer and Gibbonns (1957) , this process of leaching starts with dissolving molecules of low molecular weight fram the surface of the granules, whereas those of higher molecular weight are still left in swollen, elastic meshes. Banks and Greenwood (1967a) and Ghiasi et a1. (1982) studied the leaching exudate during the heating of starch granules and reported that at a low temperature, the lower molecular weight linear amylose exudate is preferentially leached from the granules. As the teuperature increases , the higher molecular weight branch amnylose and amylopectin are gradually obtained. Water penetrates the granules 18 throuugh these meshes and continues to dissolve molecules of low molec- ular weight within the granules. At this point, the granules are very swollen making the paste very viscous. If this process of swelling reaches the point when the granules occuupy the entire volume, the solubles in the exudate most likely rediffuse back into the swollen granules and the system becomes a gel-like continuum (Leach, 1965). At this stage, the granules are highly susceptible to thermal or mechanical breakdown due both to the highly swollen condition and progressive weakening of bounding forces within the micellar lattice. Prolonged heating of the paste results in the disintegration of the granules, which leads to decrease in paste viscosity. Continued heating, especially while stirring, produces a paste that is a mixture of swollen granules, granule fragments, and dispersed starch molecules from the granules. If the temperature of the paste is suubsequently decreased, the elements present in the paste start to associate or retrograde, then increase in viscosity. Starches from different origins or having been pretreated exhibit a different rate of viscosity increase at this stage of the pasting cycle, reflecting their different retrogradation tendencies. The practical importance of all these viscosity changes during the above-described pasting cycle was recognized by Caesar (1932) . Caesar and more (1935), Anker and Geddes (1944) and Mazurs et al. (1957) stressed the practical significance of the following points: the gelatinization temperature, the maximum viscosity during heating, the paste viscosity at the end of cooking, the viscosity increase (firing cooling, and the viscosity of the cold paste. Several instruments and techniques have been devised for viscosity measurements at these points. The Brabender Viscoamylograph has become the most commonly used for starch and starch containing products . It measures and records the l9 gelatinizing properties with relation to time , temperature and rate of shear. It is essentially a torsion viscosimeter in which the sample is kept at a constant temperature, heated or cooled at a constant rate, and the torque continuously plotted on the strip-chart recorder. Table 1 (Mazurs et a1 . , 1957) interprets the Brabender amylogran relative to the functional properties of starch. Bean starches, which are of interest in this study, have only recently been characterized by variouus investigators . Even though the starch analysis tecfmiques may be advanced, analytical difficulties still very mucln exist. Greenwood (1979) suggested that one of the inherent difficulties of studying starch is its microheterogeneiw. Because of this carplication, it is not possible to make mauny generalizations about starch. However, this literature review has summarized the current information available on the characteristics of bean starches. Physical-Chenical Properties Granule Size, Shape, and Microscopic Appearance. Granule size of bean starches has been investigated by several researchers (Kawamura et al. , 1955; Rosenthal et al. , 1970; Hall and Sayre, 1971; meen et al., 1974; Rockland and Jones, 1974; Lineback and Ke, 1975; Lai. and Varriano-Marston, 1979; Naiviloul and D'Appolonia, 1979; Lii and Chang, 1981; Sathe and Salurnkhe, 1981a). The granule size is quite variable andthegranule dimensions range frcnnabout 1 to 80 umdependingonthe source. Most bean starch granules are slender (greater length than width), althougln spherical, ovoid, elliptical, kidney-shaped, and irregular granules are also foumd. This wide variation in granule size and shape counld be due to genetic control and seed maturity. For dry 20. Table l . Functional and molecunlar properties associated with pasting characteristics of starches Paste Properties Functional Nblecuular (experimentally determined) Properties Properties Rate of increase in Ease of cooking Rate of granule viscosity when heated to 95°C Viscosity peak Viscosity changes (after reaching maximum viscosity) during heating and 95°C holding cycles Increase in viscosity during cooling es in viscosity during holding at 50°C Final viscosity after holding at 50°C Maximum thicknness on cooking Stability during C Set-back on cooling Resistance to shear swelling Extent of granule swelling Granule fragility and degree of solubilization Regradation of linear molecuules Granule rigidity Granule rigidity extent of main- tained swelling (Mazurs et a1. , 1957) 21 beans, however, a wide variability in shape is foumd in starch granules Iran the sane source (Reddy et al. , 1984). Light microscopic studies of legume starches clearly reveal two distinct starch granule characteristics: the presence of a hylum and the presence of lamellae. The hyla have been described as furrows , grooves, cracks and stria (Naivikul and D'Appolonia, 1979), cracking dark bands (Hall and Sayre, 1971), and mnicrofibrils (Donovan, 1979). The hylum and lanellae observed umnder the light microscope are not seen, however, under a scanning electron microscope. Instead, their surfaces appear to be smootln with some occasional scar-like features. These latter structures couuld arise from adhering cell wall materials or proteins, or both. Molecular Weights. Starches are high molecuular weight compounds since they are polymeric monosaccharides. Biliaderis et al. (1979) reported that the major portion of legume starches have molecular weights higher than 2 x 106, and over 902 of then have molecular weights above 4 x 10“. Amylose and Amylopectin. In legumes, the amount of amylose, ranging frum 102 to 66%, may influence starch solubility, lipid binding, and other functional properties . Amylopectin is thought to be responsible for the structural form of starch granules . Recently, Biliaderis et a1 . (1981) fractionated legume starches into amylose and amylopectin and determnined certain molecular properties . They reported that the variabil- ity between amyloses frcm different legumes may be due to: 1) maturity of seed, 2) genetic control of amylose synthesis, 3) cultivar differences, and/ or 4) seed history. The range for degree of polymerization of amnylose is quite variable. A high degree of amylose polymerization may confer 22 structural stability on the granule and also may be partially responsible for its resistance toward in vitro alpha anylolysis . Granule Crystallinity. Starch granules contain both crystalline (ordered) and amorphous (unordered) regions. This crystallinity gives rise to the birefringent property of starch granules (Elbert, 1965). From X ray diffraction studies, most legume granules endnibit crystalline structures in the Type C pattern (Rosenthal et al. , 1974). Functional Properties Swelling and Solubility. legume starches usually have restricted swelling behavior. leach et a1. (1959) indicated that legume starch swelling power is lower than that exhibited by wheat starch. Many workers have studied the swelling patternn of various legume starches: Lineback and Ke (1975), chick pea and horse bean; Lai and Varriano- Marston (1979), black bean, yellow pea, and navy bean; Lii and Chang (1981) , red bean. They have observed the sane single-stage pattern. Lai and Varriano—Marston (1979) indicated that bean starch had lower swelling power than wheat starch within the temperature range 60°C to 74°C; above 75°C, houever, bean starch surpassed wheat starch in swelling power. The restricted swelling, plus the single-stage swelling pattern, have been interpreted to explain the crystalline and amorphous regions of starch granules and the presence of strong binding forces which relax at one temperature range. Reddy et a1. (1984) suggested that the swelling ability and solubility depend on starch source , temperature and pH. Lai and Varriano—Marston (1979), Comer and Fry (1978), and Sathe et al. (1981) reported that solubility of legume starches is less than 302. 23 Gelatinization and Pasting. Must legume starches have a gelatini- zation temperature of 60°C to 90°C, with the exception of wrinkled peas (Reddy et al. , 1984) . The gelatinization temperature ranges of various legume starches are presented in Table 2. The Brabender viscosity pattern of been starches showed restricted swelling characteristics similar to those shown by chemically cross-linked starches (Schoch and Maywald, 1968: Lineback and Re, 1975: Vose, 1977: Lai and Varriano—Marston, 1979: Naivikul and D'Appolonia, 1979: Lii and Clnang, 1981; Sathe and Saluunkhe, 1981a) . According to Schoch and Maywald (1968) , these restricted swelling pastes are classified as Type C starches which show no pasting peak, but rather a very high viscosity which remains constant or increases during cooking. Nbst bean starches show some tendency to retrograde during cooling and have relatively constant cold-paste viscosity during a holding period at 50°C. Kawanra and Fukuba (1957) classified legume starches into two categories based on hot paste characteristics: 1) those which do not have a substantial rise in viscosity during heating (25°C to 92.5°C) and cooling (92.5°C to 25°C) cycles and 2) those which show a distinct rise in viscosity during heating and cooling cycles. lai and Varriano-Marston (19 79) stuudied the gelatinization temperature range of black bean, 63.8°C to 76°C, which is considerably higher than that of wheat starch, 55.6°C to 63°C. This result is simnilar to that reported for navy and mung bean starches, according to Schoch and Maywald (1968). Rosenthal et al. (1970) studied the pasting characteristics of jack bean and guandu bean starches and reported that there was no peak at low concentrations (up to 6%) . At these low concentrations , no sensible retrogradation was observed and their viscosity curves were similar to cross-bonded starch. Vose (1977) studied the effects of 24 Table 2. Gelatinization temperature of legume starches Gelatinization Starch Source Temperature Range (°C) Reference Lima bean 70 to 85 Schoch & Maywald (1968) Lentil 64 to 74 Schoch 8: Maywald (1968) 58 to 61 Biliaderis et al. (1979) Yellow pea 63 to 73. Schoch & Maywald (1968) Navy bean 66 to 77 Schoch & Maywald (1968) 68 to 74 Biliaderis et al. (1979) Garbanzo bean 62.5 to 72 Schoch & Maywald (1968) 65 to 71 Biliaderis et a1. (1979) Mung been 60 to 78 Schoch & Maywald (1968) 63 to 69 Biliaderis et al. (1979) Wrinkled pea 69 to 83 Schoch & Maywald (1968) > 99 Biliaderis et al. (1979) Black gram 71.5 to 74 Sathe et al. (1982) Black bean 63.8 to 76 Lai & Varriano—Marston (1979) Suooth pea 65 to 69 Biliaderis et al. (1979) Red kidney been 64 to 68 Biliaderis et a1. (1979) Faba been 61 to 66 Biliaderis et a1. (1979) 61 to 69 lorenz (1979) Soybean (Amsoy 71) 73 to 81 Wilson et al. (1978) Pea 54 to 66 Coner & Fry (1978) Red bean 63 to 70 Lii & Chang (1981) Adzuki bean 83 to 89 Biliaderis et al. (1979) 25 acidic and basic conditions during pasting of double-milled air— classified pea flour. He found that at acidic pH values up to pH 4, there was a steady increase in viscosity during the pasting program, including a resistance to shear breakdown when 82 slurries were main- tained at 96°C for 15 min: however, at pH 3 a pasting peak occurred, which was followed by a decrease in viscosity and then a set-back during cooling. Lii and Chang (1981) also observed that the different steeping solutions used during starch isolation did not affect the viscosity pattern. Sathe and Saluunkhe (198la,c) indicated that the gelation of purified bean starch could yield stable a gel at concentrations of 72 or above (w/v) . In recent years , more interest has focused on the relationship between the molecular struucture and the functional properties of starch. Biliaderis et al. (1979) reported that the average molecunlar weight of legume amylopectin, determined by gel chronatography, was greater than 20 x 106, and there was significant correlation (P < 0.05) between set- back viscosities and percentage of starch components fractionated between 0.2 and 0.9 Kav (Kav = We - V0] / [Vt - Vo]), where Va is the elution volume, V0 is the exclusion volume, and Vt is the total volume. In the same year, Naivikuul and D'Appolonia observed that the intrinsic viscosity was directly proportional to the molecular weight of amylose fraction from various legume starches. In 1981, Biliaderis et al. developed an hypothesis to explain starch gelatinization and granule rigidity of legume starches based on the molecular characteristics of the branched polysaccharide fractions (amylopectin) . 0n the other hand, Suzuuki et al. (1981) suggested that the gelatinization and retrogradation properties may be influenced by the structure of amnylose. Recently, 26 Takeda et al. (1983) confirmed that the chain length distribution of amyloPectin affects the pasting properties of starch. Starch Analysis Extraction. In order to study the properties of been starches, extraction and purification are essential to isolate starch granules from beans and/or bean flours. Schoch and Maywald (1968) found that the separation of pure starch was difficuult from certain legmes because of the presence of a highly hydrated fine fiber fraction and also a high content of insoluble protein. Kawauura et al. (1955) proposed a method to isolate starch from various legmes by treatment with a 0.22 NaOH solution, washing with water, and dehydration with ethanol and ether. Schoch and Maywald (1968) simply isolated legume flour by repeated washings with distilled water and air-drying at man temperature. Recently, Sathe and Salunkhe (1981a) isolated starch from Great Northern beans (Phaseolus vulgaris L.) by extracting sequentially with different solvents: water, 2.02! NaCl, 0.1N NaOH, and aqueous ethanol. Quantitation. Various methods for quantitative determination of starch have been reported. Those based upon optical rotation were unsatisfactory because of turbidity and the relatively large quuantities of materials requuired. The methods involving acid hydrolysis followed by reducing sugar determinations were slow and limited to sauples which were free from other cellulosic materials which could also yield gluucose on treatment with boiling acid. In 1970, the AACC Starch and Pentosans Committee began an investigation into new and improved methods for determining starch in cereal produucts . After surveying the literature, 27 it was decided to examine a procedure that incorporated the use of enzymes specific in their catalytic action on starch and glucose. For instance, the use of glucoamylase to hydrolyze starch to glucose could eliminate the problem of inversion caused by acid hydrolysis (AACC Method 76-20). Also, the use of glucose oxidase wouuld reduce the uncertainties associated with empirical methods of measuring glucose, such as reducing sugars (AACC Methods 76-30A, 80-60, 80-68; Under- kofler et al. , 1943) , and certain colorimetric techniques (Dubovski, 1962: Saunders et a1. , 1970). The use of enzymes could eliminate the problem of materials other than starch contributing to the angle of rotation as measured by polarimetric procedures (AACC Method 76-20) . The use of sore pretreatments before starch determination has been proposed by many investigators. Banks et al. (1970) proposed a method of initial extraction with aquueous methanol or ethanol to remove the lov molecular weiglnt carbohydrates . Gelatinization of starch is accomplished by permitting rapid and complete enzymatic hydrolysis . This gelatinization process may be effected by autoclaving (Thivend et a1. , 1965 ; Shetty et al. , 1974) or by chemical treatments using dimethyl sulfocide (HzGuire and Erlander, 1966: Libby, 1970) or concentrated calcium chloride (Earle and Milner, 1944 : Banks et al. , 1970). Libby (1970) and Vose (1977) suggested the conbination of alpha amylase and glucoamylase enzymes for digesting the starch to yield glucose. In 1979, Bar and Alexander developed the enzymatic hydrolysis method by using an ordinary water bath for gelatinizing starcln followed by hydrolysis with both alpha amylase and glucoamylase. Fleming and Reichert (1980) reported that calcium chloride dispersion provides higher yield of starch and more reproducible analytical results than does gelatinization by autoclaving. 28 Althougln sensitive, these enzymatic methods are not sufficiently accurate for all purposes. Enzyme purity is a limiting factor. Gluco- amylase preparations are usually contaminated with beta glucanase , which can release glucose from the cell wall. Recently, Luustinec et al. (1983) presented a new, accurate and rapid method which is based on simplified extraction of the starch-iodine complex on a glass fiber disk. This method was designed to overcoue the disadvantages of enzymatic method- ology: moreover, it is faster as well as more starch-specific. Liunear starch couponents , amylose, can be determined directly fron starch materials by either of the for general techniques: 1) potentio- metric titration of the dissolved starch with standard iodine, 2) similar amperouetric titration, 3) spectrophotometric determination of intensity of blue coloration with iodine, or 4) absorption of congo red (Schoch, 1943: Larston et al. , 1953: Gilbert and Spragg, 1964: Carroll and Cheung, 1960). The first three metlnods are based on the iodine affinity of amylose: the forth procedure is based on the reduction of dye affinity due to the presence of branching. Whistler et al. (1964) suggested that the precision of absorption of the congo red method was inferior to the iodine binding methods . Therefore, the determnination of the iodine bound by amnylose is the fundamental, basic method used to determine the amount of amnylose in starches . Quantitative isolation of amylose is reported to give the more accurate results . Methods of fractionating starch have been extensively reviewed since 1945 by Schoch. The Schoch method is based on dispersing starch granules into an aqueous solution, followed by the addition of a polar, organic suubstance, resulting in the formation of an insoluble amnylose complex. Amylose is then recovered and determnined by the fundamental 29 methods. Several investigators developed and modified Schoch's method (Greenwood and Thompson, 1962: Mnntogouery and Senti, 1964: Banks and Greenwood, 1967a,b). Size exclusion chromatography is another usefuul techniquue used for fractionating starch couponents. This gel filtration tecl'mique was originally developed for the separation of biological materials in aqueous media on bead-formed gels consisting of dextran cross-linked with epichlorohydrin. A significant breakthrough in the evolution of the size exclusion chromatography was the development of polystyrene beads. These allow for the separation of polymers dissolved in organic solvents , and this method is known as gel permeation chromatography (GPC) . Sincethebasisofthetwoteclmiquesisthesame, theyareconbined uunder the name of size occlusion chromatography separation (according to molecuule size). Because of the different molecule size of amylose and amylopectin, size exclusion clromatography has been applied by many researchers to separate the starch couponents. In addition, comercially available dextrans, with known molecuular weights, can be used as size- excluusion calibration standards to obtain the molecular weight distri- bution (MWD) patterns. Barker et al. (1969) have applied chromatography over porous glass beads for the fractionation of dextran and hyaluronic acid. Patil and Kale (1973) developed a cellulose column equilibrated with ethanol-urea used for the quuantitative separation of amylose and amylopectin. Dintzis and Tobin (1974) fractionated amylose and dextran over porous glass. Ebermann and Schwarz (1975) separated the starch components by gel filtration on agarose beads. Yamada and Taki (1976) adopted perchloric acid as the solvent in a gel chromatography sys ten. Aquueous dimethyl- sulfoxide (11430) is able to dissolve carbohydrates without damaging their 30 primary structure and the activities of hydrolytic enzymes are umaffected by concentrations of D’ISO uup to 20% (Lineback and Sayeed, 1971). There- fore , D130 was suuggested to be used in gel permeation chromatography (GPC) syst-I to give better resolution, although it is a viscous solvent (Minor, 1979). Papantonakis (1980) used the aquueous D’ISO as solvent in micro-BONDAGE. GPC columns and suggested that the ratio of D480 and H20 had an effect on resolution. Because D60 is a viscous solvent, it is not particuularly well suuited for use in conventional chromatography techniques. High performance liquid chromatography (HPLC), however, can overcome the disadvantage of using D150 by supplying enough pressure to counteract the viscosity of the D430. High performance liquuid chrouatography facilitates more rapid separations than conventional column chromatography. The main difference between HPLC and conventional chromatography is that HPLC uses a pressurized eluent as a driving force for separating samples. Develop- ment of improved pumping systeue capable of maintaining a constant liquuid flow rate over a period of time and improved columns able to separate closely related conpomds in a mixture have made HPLC more valuable in the past few years , especially for the anlysis of carbo- hydrates . The corbination of size exclusion with HPLC techniques has been successfuully used and reported by many workers (Papantonakis, 1980: Kruger and Marchylo, 1982) . MATERIALS ANDMEIHODS Dry Bean Products Bean Sorce Prime, handpicked navy, black (1981 Michigan crop) and pinto (1982 Michigan crop) beans (Phaseolus vulgaris L.) were used in the production of dry roasted fractionated bean flora . Ninety-one kilogramns of each been type were shipped to the Food Protein Research and Development Center at Texas A & M University for further processing. Preparation of Samples Air-Classified High Starch Flours. Forty—five kilograms of each been type were roasted in a solid-to-solid heat exchanger as shown iun Figure 1 (custom built by Food Processes Inc. , Saginaw, Michigan). The heat ecchanger's heat transfer medium consisted of 1.6 mm (1/16 in) diameter type A 902 aluminum ocide ceramic beads (Coors Ceramic Co. , Golden City, Colorado) with a specific gravity of 3.6 g/cumz. The beads were heated to 240°C and were maintained in the chamber with the raw beans for 100 seconds in a 1:5 ratio of beans to beads. These pro- cessing conditions resulted in a bean ecit temperature of 113°C. Roasted beans were then cracked through a corruugated roller mill (Ferrell Ross, Oklahoma City, Oklahome) into six to eight pieces. The lnulls were reuoved using a zig-zag aspirator (Kice Metal Products, Wichita, Kansas). 31 32 Product hopper Beads elevator / Separation screen Arr exhaust /0’ a Gas burner discharge Scheuatic diagram of the particle- to-particle dry roaster. Points 1 to 5 indicate teuperature measurerent sites: 1) product, 2) beads out of heating drum, 3) beads into uupper drum, 4) beads return, and 5) air. Arrows indicate flow of beads. Figure 1 . 33 After huull removal was coupleted at the Food Protein Research and Development Center, Texas A & M University, the cracked cotyledons were shipped to Alpine American Corporation, Natick, Massachusetts , for milling and air-classification. The cracked cotyledons were finely ground in a Model 250 CW stud impact mill at a Speed of 11,789 rpm and a door speed of 5,647 rpm. The resulting flors were air-classified in a Model 410 MPVI air-classifier at a rotor speed of 2,200 rpuun and a brake ring setting of 3, using a 7.62 on (3 in) screw feeder operating at 25 rpm. At this point, two flor fractions were obtained: an intermediate starch fraction (coarse I) and an intermediate protein fraction (fines I) . The intermediate proteiun fraction was reclassified under the folloping conditions: rotor speed of 2,200 rpm, brake ring setting of 0, 7.62 cm (3 in) screw feeder operating at 25 rpm. As a result of this second air-classification step, a high starch fraction (coarse II) and a high protein fraction (fines II) were obtained. A flow diagram of the flor processirng scheme and the fractions produced is presented in Figure 2. All flor fractions were packaged in poly- ethylene bags and shipped to Michigan State University in fiber drums at arbient temperature. Upon arrival at Michigan State University, the flor fractions were stored at roomn terperature. The high starch flora froneachbeantypewereusedinthis study. Purified Starch. Bean starches were purified according to the method modified by Naivilcuul and D'Appolonia (1979) . Approximately five kilograms of air-classified high starch flor of each bean type were slurried in 20 Kg of 0.016 N NaOH solution with contiunuous mixing for 20 min. The starch was recovered by centrifugation (2,000 x g, 20 mnin). The lighter fine fiber and any unndissolved protein, which remained 34 whole Beans [ we... 1 _ i [ Dehulling 7] i Aspiration ‘ l Hulls . Cotyledons Grinding I Pin Milling] Hull Flour - Air Classification] ' - l _ Coarse I [Pines 1‘] Immediate ‘ Fiber Fraction [ Air Classification I Coarse 11 High Starch Flour Fines 11 High Protein Flour I Figure 2. Flow diagram of processing navy, pinto and black beans 35 suspended in the supernatant, were decanted, and the spongy gray upper layer of starch tailing was skimmed off. The dense white bottom layer was crude starch which was washed three times with distilled deionized water, two times with 702 ethanol, and two additional times with dis- tilled deionized water, respectively. The supernatant and sludge was removed and discarded after each washing. The prime starch was than air- dried at 38°C to 40°C for two days. Finally, the dry starch was ground using a Waring blended (Waring Products Division, New Hartford, Connecticut). Physico-Chemical Analyses Analytical Methods Moisture. The moisture content of samples was determined in triplicate by AACC Method 44-32 (1962). Approximately 5 g samples were weighed into previously dried, cooled and tared cruucibles. Sauples were dried uunder a partial vacuum (ca. 25 mm Hg) between 95°C and 100°C for six hors. After cooling in a desiccator, samples were reweighed and percent moisture was calcuulated based on the fresh weight , as follows: wet sauple wt (g) - dry sarple wt (g) Z Moisture = x 100 wet sample wt (g) 5333. The ash content of sarples was determined in triplicate by AACC Method 08-01 (1962). The dried samples obtained from the moisture determinations were incinerated at 525°C for 24 hours in a Barber- Coleunan muffle furnace Nbdel No. 293 C (Thermolyn Corp. , Dubuque, Iowa). The uuniform white ash was cooled to room teuperature in a desiccator 36 prior to weighing. Percent ash was calculated based on the dry weight, as follows: (cruucible wt [g] + ash wt [g]) - crucible wt (g) Z Ash = x 100 sample wt (g) Protein. The protein content of samples was determined in tripli- cate by AACC bathed 46-23 (1962), a standard micro-Kjeldahl procedre. Percent nitrogen was calcuulated based on the dry weight, as follows: (mlHCl-mlblank)xNHClxequiv.tha0H ZN = sample wt (mg) Percent nitrogen was tlnen multipled by 6.25 to obtain percent total protein. Cruude Fat. The cruude fat content of samples was determined in triplicate using a modification of AACC Method 30-25 (1962) , a Soxhlet extraction procedre. Percent cruude fat was calcuulated based on the wet weight, as follows: (fat wt [g] + flask wt [g]) x filtrate vol (ml) Z Crude Fat = x 100 sample wt (3) x 50 ml Dietary Fiber. The enzyme neutral detergent fiber (ENDF) was determnined for each sample in triplicate by the method of Robertson and Van Soest (1977). This method was modified to iunclude 1 mg of amylo- glucosidase to provide additional digestion to starch. iugars. The simple and oligosaccharide sugars (glucose, sucrose, inositol, raffinose and stachyose) coumonly found in beans (Phaseolus 37 vulgaris L.) were determined by the new method developed by Agbo (1982) using higln performance liquid chromatography (HPLC) . Sugars were analyzed by first extracting a one gram flour sample with 10 ml of 802 ethanol in an 80°C water bath shaker for 15 min. The mixture was then centrifuuged at 2,000 rpm for three minutes and the supernatant was collected in a new tuube . The extraction was repeated two more times with the same procedre except that 5 ml of 802 ethanol were used in the second extraction. The precipitate was saved for starch analysis . Two ml of lead acetate were added to the extract collected from these three extractions . The mixture was shaken and centrifuged at 2, 000 rpm for three mimutes to precipitate protein and the super- natant was transferred to another tuube. T'wo ml of 107. ocalic acid were added to the tuubeandagainthemuixturewas shaken and centrifugedto remove the residual lead acetate. The final supernatant was brought to 25 ml volume and was passed through a SEP-PAK C Cartridge (Waters Associates, Innc. , Milford, Massachuusetts) prior to analysis. The HPLC consisted of a Solvent Delivery System 6000 A, a Universal Chromato- graph Injector U6K, Differential Refractoneter R 401, and a Data Module Model 730 (Waters Associates, Inc.) . Acetonitrile-water (75:25, v/v) was used as the solvent at a flowrate of 2.0 ml/min. The carbohydrate analysis (micro-BOMTAPAK) column was used to resolve the suugars from the aqueous extract . Quantitation and identification of sugars were done using the external standard method programmed in the Data Module to calculate the absolute amounts of the components in the injected sample (25 1:1) and to obtain the percent sugar content based on dry weight: 38 Data Module ant (mg/m1) x original sample (ml) Z Suugar = x 100 sample wt (mg) M. Starch was deth by the enzymatic hydrolysis method, developed by Agbo (1982) , in conjunction with HPlC. The precipitate residue obtained from the sugar extraction was air dried to reuove etlanol. Ten ml of 0.5 N NaOH were added to the centrifuge tuube containing the pellets (approx. 1 g). With a rod, the pellets were colloidally dispersed into the 10 ml of 0. 5 N NaOH solution by continnu- ous crushing and stirring. After the pellets were completely dispersed, 10 ml of 0.5 N acetic acid (CH3CIJOH) solution were added to the tuube to neutralize the NaOH used for dispersion. The neutral colloidal suspen- sions were allowed to stand for a few mimutes in order to settle uundissolved particles , mainly cellulose, as well as pectic substances. Then, 1 ml was transferred to a test tube, and 3 m1 of amyloglucosidase enzyme solution (5 mg/ml) were added (gluucoamylase 1,4-a-D-glucan glucohydrolase from Rhizopus, Sigua Chenical Co. , St. louis, Missori, in 0.02 M acetate buffer, pH 4.9). The sauple was continuously shaken in a water bath at 55°C :I: 2°C for 30 min. After shaking, the test tube was removed from the water bath and filtered through a SEP-PAK to remove the acetate (Lester, 1980) prior to injection into the HPLC column (the sample system used in the sugar analysis system). Quantitation was the same as for sugars with exception that only glucose was used as a standard solution: 39 Data Mndule original amt (mg/ml) x 10 x sample (ml) x 0.9 7. Starch = x 100, sample wt (mg) where 0 . 9 is the factor to accoumt for the water gained dring hydrolysis and 10 is the equuation correction factor. m. Amylose content of the purified starch was determined in quadruplicate using the "Blue Value" method of Gilbert and Spragg (1964) . To a 50 ml flask 1. 0 m1 of an aqueous solution containing approximately 0.5mgof sample and0.5ml of 1.0NNa0Hwere added. Themixturewas heated iun a boiling water bath for 3 min. After cooling, 0.5 m1 of 1.0 N HCl were added, followed by 0.07 g to 0.1 g of potassium hydrogen tartrate. Water was added to a volume of about 45 ml followed by 0.5 ml of iodine solution (2 mg iodine/m1: 20 mg potassium iodine/m1). This solution was made up to 50 ml, minced, and allowed to stand 20 min at room tenperature. The absorbance was measured at 680 mu in a spectro- photometer (Bausch & loub) using a 1 on cell. For the reference solution, an iodine solution of equal concentra- tion was used. The "Blue Value" (B.V.) was calculated from the following equation: absorbance (mm) x 4 c (mg/d1) "Blue Value" = where c refers to the carbohydrate content of the solution. Blue Values of the samples were compared with those of pure amnylose (amylose from potato, type III, Sigma Clnenical Co.) to determine a measure of amylose content . 40 Pasting Characteristics A Brabender Viscoanylograph (C.W. Brabender Instruments, Inc.) equipped with a 700 cum-g sensitivity cartridge was used to determine the pasting characteristics of air-classified high starch flora and purified starches. Ten g/100 ml and 8 g/ 100 ml concentrations of starch slurry in 400 m1 of pH 5.3 phosphate buffer were used as standard conditions for air- classified and purified samples , respectively. Each ample was examined in duplicate following the standard procedre according to the AACC Method 22-10 (1962). The terperature of the ample was raised uniformly from 30°C to 95°C at a rate of l.5°C/min, held at 95°C for 15 min, and tlnen cooled uniformly to 50°C at the same rate, and held at 50°C for 15 min. Concentrations of 6, 8, 10 and 12 g/100 ml were uused for air- classified high starch flora, and concentrations of 6, 8 and 10 g/ 100 ml for purified starches. Standard conditions of amylograph oper- ation were utilized throughout these studies . Pl'osphate buffers of pH 4.0, 5.3, 6.0 and 8.0 were used for all smples . Standard conditions of amylograph operation were utilized throughout these stuudies. Ehctended high teuperature holding at 95°C for 30 min and extended low temperature holding at 50°C for 30 mnin following standard heating conditions were applied to all samples . Molecular Structure The molecuular struucture of purified starches was characterized by the application of gel permeation and enzymatic method. 41 Alpha amylase from Bacillus subtilia (Sigma Chemical Co.) and beta amylase from sweet potato (Sigua Chemical Co.) were applied to provide better structural information for this stuudy. The gel permeation chromatography (GPC) system used in this study was couprised of two 4.0 mm LB. 30 cmmicro BONDAGEL E-linear columns connected in series (Waters Associates, Inc.) . This type of GPC column contains fuully porous silica gel with a bonded organic ether as the stationary phase. Silica particles, approximately 10 mm in diameter, provide resolution for polymers ranging in molecular weight from 2 x 103 to 2 x 106. These two GPC columns were connected to the HPLC system used for sugar analysis. Dimethyl sulfoxide-water (90:10, v/v) was used as the solvent at a flow rate of 0.3 ml/min. The injected ample volume was 50 ul. Smples prepared for this stuudy included starch dissolved in aqueous D60 (90:10, v/v)‘without any enzyme treatments as control, alpha muylase hydrolysates , and beta amylase hydrolysates . _Cg_n_£r_o_l. A ample of 0.15 g purified starch was dissolved in 100 m1 aqueous D60 (90:10, v/v). The mixture was contirnuously shaken in a 55°C water bath for 24 hora (leach and Schoch, 1962). After shaking, the mnixture was centrifuged at 2,000 rpmn for 10 mnin. The supernatant was filtered through the clarification kit (Waters Associates, Inc.) prior to injection onto BONDAGEL E- linear GPC columns . Alpha and Beta Amylase Hydrolysates. A ample of 0.15 g purified starch was treated separately with 10 ml of two enzyme solutions . Alpha amylase enzyme solution contained 560 units/ml in 0. 02 M sodium acetate buffer, pH 5.8. Beta amylase enzyme solution contained 200 units/ml in 0.02 M sodiumn acetate buuffer, pH 4.8. The starch-enzyme mixture was 42 incubated in a 35°C water batln for 24 hours with continuous shaking. The digestion solution was subsequently heated in boiling water for 20 min to inactivate enzyme activity. 11430 was added to afford the desired solvent couposition (INSO:H20, 90: 10, v/v) . Centrifugation and filtering were also required to prepare the injection smnples. ' Comercial amylase and amylopectin from potato (Sigma Chemical Co. ), maltose, and glucose were used as standards to investigate their elution volumes . Sample preparation of standards followed the procedures for the control smnple. Statistical Analyses The "Statistical Package for the Social Sciences" (SPSS) conputer progrmns described by Nie et a1. (1975) for use on the CDC 6500 computer operated by Michigan State University Couputer Laboratory was used to assist statistical analyses. Multivariate analyses of variance and covariance were determnined by using subprogram ANNA. Mean squares were reported after rounding. Single classification analyses of variance and T‘ukey mean separation were determined by uusing subprogram (NEWAY. Tunkey mean separation was selected for judging the significance of observed differences between each treatment and control. Treatments not significantly different (P < 0.05) were indicated with like letters. RESULTS AND DISCUSSIG‘I Chemical Composition Proximate Composition Analyses than values and T'ukey mean separations for proximate conposition of air-classified higln starch flors and purified starches from navy, pinto and black beans are presented in Table 3. Air-Classified High Starch Flors. All tlnree flors show similar cmsition. This similarity might be a result from not only the same preparation process but also from the same general species of leguminous seeds, Phaseolus vulgaris L. Fat content of navy bean flor (1.472) was significantly higher than that of pinto (0.832) and black (0.722) bean flors. Significant differences in protein content were detected for all three bean flors . Black bean flor showed the lowest protein content{ (12.422) and pinto bean flor, the highest protein content (19.062). Conversely, the starch content of pinto bean flour was the lowest (65.432) and that of black bean flor, the highest (69.872). This inverse relationship between starch and protein content has been noticed previously by Tyler et al. (1981) for a number of other bean flora. Navy and black bean flors showed no significant differences in ash (3.632 and 3.742, 43 44 meadow wanna mo “Swag b6 unmoumo Go momma .q u a N Amoé v m .moucmummmwo ugoauflflmwm on 38.35 95.30 Sumo mafia? mumuuoa oxfl .moowumamoom 5m: 5953 m u a dag gem nN©.H¢ onm.hw me.o nom.H wHw.o mNH.o amo.© @QHMHHDQ nun num.m© QNN.H oqm.m UN¢.NH flmw.o U¢H.w wmflMHmmmHonHHm saga moo.©m omm.om mmn.o mo~.o nMQ.H mNH.o woo.m wmflmwhsa Inn www.mo 0mm.N omo.m moo.mH nNm.o omo.w @QwMAmmmHouHHm ouGHm Amqm.mm 05H.Hm mHm.o mom.o mwm.o wHH.o UNN.~ wmflMHHDQ nun mm¢.m© amn.H umo.m va.mH omm.H mmn.m @mHMHmmmHonHHm .062 E c3 N ufifimmue museum N333 H308 madman: BE anon fiufim wee a... £88m “we H963 sums am 85a .5 gum $30.33 “63%.": pg magnum mega swan Bwflmmflonha. mo 8330980 Baum .m 3an 45 respectively,) or ENDF (1.762 and 1.722, respectively) contents, but they were significantly different from those of pinto bean flor (3.052 ash, 2.532 ENDF) . The air-classified high starch flor investigated by Tyler et al. (1981), however, had a higher starch content (71.02 to 75.22)than any of these studied here. This higher starch separation efficiency might have been caused by differences in bean seeds (seasonal variations mong beans), seed hardness, or process conditions. Purified Starches. The air-classified high starch flours were purified further in order to study the characteristics of only the bean starch. The flors were purified by following the method modified by Naivikul and D'Appolonia (1979), where 0.016 N NaOH, water, and 702 ethanol were used in the purification process as extractant solvents. Dilute NaOH is used by many investigators and it is suggested as a good medium to dissolve protein, but not to gelatinize the starch (Schoch and Maywald, 1968) . In addition, dilute alkali saponifies free fatty acids which are then removed by water washing. Fiber is partially soluble in this solution. Water is used as a neutralized solvent and wash, where water soluble corponents suuch as water soluble pentosans and mono- saccharides are removed. Ethanol is used to extract sugars from starch and also to decrease the fat content. The lighter, fine fiber, some undissolved denatured protein, and other insoluble solids remain suspended. Due to differences of density, starch can be recovered by centrifugation. Two distinct layers show after centrifugation: a spongy gray top layer and a dense white bottom layer. The top layer was skimmed off; the bottom layer is the purified starch. 46 During starch purification, greater difficulties in separation of pinto starch were encountered than for navy or black bean starches. The top layer, a spongy gel-like substance, could not be separated from the starch layer, as it was with the navy and black bean starches. Thus, a higher impurities content was expected. Also as expected, pinto starch showed a significantly higher protein content (1.452) than that of navy (0.982) or black (0.812) bean starches. However, no significant differences in fat content were detected for any of the three bean starches (0.112 to 0.122). Black bean starch showed significantly higher ash content (1.562) than that of navy (0.462) or pinto (0.262), due, perhans, to the inclusion of small black particles dispersed in the purified starch fraction of the black bean after centrifugation. Naivikuul and D'Appolonia (1979) reported the chemical composition of navy and pinto bean starches isolated by this same method. Their results show higher starch contents than do the results presented in this study, possibly because they prepared starcln on a smaller scale (500 g) than this study did (7 Kg). The washing step could have been more efficient with the smaller sample than with the larger smple size. The amylase content of navy, pinto and black bean starches, determm'ned by the "Blue Value" method (Gilbert and Spragg, 1964) , were 39.2421, 36.092 and 41.62z, respectively. Among these three bean starches, the mnylose content of navy bean was not significantly different from that of pinto or black beans, but there was a significant difference between pinto and black beans . Naivikul and D'Appolonia (1979) reported much lower amylose contents of navy (22.12) and pinto (25.82) bean starches. The amylose content of black bean starch was slightly higher than the value of 38.12 reported by Lai and Varriano-Marston (1979). However, all of the amylose contents 47 from this study fall within the range reported by Reddy et a1. (1984) for legume starches. The disparity in results of anylose contents between previous reports and the present study might be explained by differences in the methods used to determine anylose levels. In addition, Duffus and Murdoch (1979) suggested that large variations in amylose contents of individual varieties of legumes depended on not only geo- graphic location of grovth but also the maturation of seed. Sugar Analyses Mean values and Tuukey mean separations for sugar analyses of air- classified high starch flors and purified starches from navy, pinto and black beans are reported in Table 4. Air-Classified High Starch Flors. All three bean flors show the sme sugar order pattern, from high to low: sucrose, stachyose, glucose, raffinose, with no inositol detection. This order of dominance agrees with mnany previous studies (Sclnweizer et al. , 1977 ; Naivikul and D'Appolonia, 1978; Fleming and Reichert, 1983); however, the sugar content levels are different. The stachyose (0.932) and raffinose (0.172) levels of navy bean flors are slightly lower than those reported by Fleming and Reichert (1983), 1.022 and 0.512, respectively. The differences may be explained by the different methods of sugar determination, as well as the variation inlnerent in beans grown in different locations. Purified Starches. As ecpected, no sugar was detected because 702 ethanol was used twice as the solvent to wash the starches during the final step of purification, thereby completely extracting the sugars with ethanol. 48 3303933838?“ x magnum\mmumoflmoa my a u .9 3830330 uoam N Gee v e .seefieeefle ugamflswwm on goofing 55.30 £08 guns mumuuoa 9H: .mooeumumoom Emma .3035 wood? 59H ea ea ez qz ea Banner s35 €25 ez nae; em; eeeflemflusas xeflm e2 92 oz oz nz Emma same and E ewes; some eeefieeflo-hs BEE oz ea nz oz oz Beater s85 s25 me some some eeefisefioées bfiz omoensumum omofimmmm HouwmoaH 08.83 800.95 pom—Boone. fiHmum «Be a mama $25.88 , Hagen x039. mam 85m :06: Sum 3305a edema: one madam £03m AWE moamwmmmau nwa mo Ammobfimum new magmmmu .Houamofi .8803 .0803va ”€380 .3me ufiuhom .q manna. 49 No comparison can be made with literature for the purified starches, since there has been no previous study of them with regard to the simple sugars. Brabender Pasting Characteristics The Brabender pasting curves of air-classified high starch flors and purified starches from navy, pinto and black beans with varying concentrations, pH values , and holding periods are shown in Figres 3 through 20. As a result from these curves, several practical signifi- cant points are expressed in tabular form according to Medcalf and Giles (1966) terminology (Tables 5, 6 and 7). The initial pasting tenperature is defined as the teuperature at which an initial increase of ten Brabender units (BU) inn viscosity is reached. Normally, peak height is the maximum viscosity obtained; hov- ever, since no definite peak was obtained in this stuudy, the values at 95°C were reported. The viscosities of smnples after the 15 min holding time at 95°C, after cooling to 50°C, and after 15 min holding time at 50°C are also reported. The Brabender pasting characteristics of air-classified navy, pinto and black bean flors at standard conditions (10 g/ 100 ml concentration in pH 5. 3 phosphate buuffer solution) are shown iun Figure 3. Individual points are recorded in Table 5 . Navy and black bean flors gave similar initial pasting teuperatures (78.5°C) which were slightly higher than that of pinto bean flor (75.5°C). However, they all showed the same viscosity pattern. This pattern showed restricted swelling, containing no pasting peak, and increasing viscosity during heating at 95°C. A tendency to retrograde during cooling was also shown on their pasting 50 8338 Hommdc. mammomono m.m mo 5 H8 oon S no whoa some x83 now 893 corms Bwfimmmaunfim mo 3.83.2080 onoEwumv mug wfinummo Begum .m shaman AEEV oE_._. o: oo 2 on. 8 n 8. 1 8n 8:... 1 can A x 8.. :08 w n H... 08 .A ) . 8 con n ( . 85 :33 u can L P L g omAllluvoui 3134 on 8 av oeaeoeooEo... 51 Ammo amaeoomss coca. 82A ass mmm w.~e OH sum mmN ewe om m.Hw m an am we a m.sm e seemesse emu nae mes oHN m.ne NH mwN com owe as n.m~ OH was sue um a o.om m NH sH s - - e eeeeeseseo-eas emcee coca. coca. see mes m.m~ ea one mmm mms omH m.ea m mm mm as m - e eeeeesse 82A ocean man one n.m~ NH mss mes sen nee m.w~ 0H nse was we we e.mm m NH we a - - e eeameeesee-sss assz Pom as 0.8 as 0.3 us 0.3 as 8.0 one. 3a 823 efifisea. fiufim fia 2 5a 2 masseuse Seesflfiosoo Bee. 88 seems seems Heeesse 8.330s you? mongoose m .m mo 5 23338080 30% um women #83 new 85o assesses Beaus as when fame ewe: Renegade mo seseefleeosssfi 333 .m flame 52 pm 0.3”." u €33.53 Revenue pom AN n 5 monomer Ems." 65 338mg 82A 83A OOOHA eem o.ee OH mms ass can “He m.we m me we mm m m.em e eeeeemse coca. 82A coed. ose m.m~ NH sme emu own emN m.m~ ea New sew «NH sN m.w~ m we mm sH e o.mm e eeeeseeseu-ess semen 0.8 us 0.2.. as 0.3 us 0.3 as 8.0 any 25 8:3 efiflseee 65m. as 2 HE D masseuse Seessefiueoo tea 53 seems seems Heeeeea A.e.eeese n seems 53 curves. Among the tlnree types of beans, black bean flor showed the greater tendency to retrograde than did either pinto or navy bean flor . later on during the 15 min holding phase at 50°C, the curves showed some decrease in viscosity. This phenomenon is most likely the case of molecular breakdom, resulting from a continuous shear force. At the same concentration, black bean flor developed a more viscous paste than did thenavyorpintobean flors. Thereasonmaynot onlybe due to the individual starch characteristics, but also to some effects from the differences in starch content (Table 3). The Brabender pasting curves of purified starches of the three bean types at standard conditions (8 g/100 ml in pH 5.3 phosphate buuffer solution) are illustrated in Figure 4 , with individual points reported in Table 6. A lower solid concentration was used for the purified starches (8.0 g/100 ml) than was used for the air-classified starches (10 g/100 ml), because purified starch contains a higher percentage of starch at the same concentration than the air-classified flor. From these three bean starches, pinto showed the highest pasting teuperature (81.5°C) followed by navy (79.3°C) and black (78.5°C), respectively. The initial pasting temperature of navy and black bean starches deter- mined by this investigation has good agreennent with previous studies: navy bean, 77.0°C (Naivikuul and D'Appolonia, 1979); black bean, 76.0°C (Lai and Varriano—Marston, 1979). However, the initial pasting temper- ature of pinto starch from this study (81.0°C) was higlner than 77.0°C reported by Naivikul and D'Appolonia (1979). All bean starches showed similar viscosity patterns as their flors. Nevertheless, there were some differences, which will be discussed later. The purified starch curves indicated a tendency for the starches to retrograde during cooling. Later on during the 15 mnin holding phase at 50°C, the viscosity of the 54 8338 883 Seasons m.m me 5 as 8% w as 888s 83. roman. o8 oufio :98 “03.38 .46 A83380 ohooafiumv 88 wfiuummo H8883 3.5 st: 09? 00 Os. 8 8 J q E 3 ir 818A] .3134 .on 8 av 0.398th .s semen 8w 8.4. 89 m x 8. 0 5. al.- 8» A ) 8 cos n ( 8s. 80 80 55 snn snn nnn on n.no o.o non non sen os n.no o.s sen nnn nnH on n.no n.n osn nnn osH on n.no o.s nonmemso non nnm nee nn o.on o.n nnn non sen ns o.on o.o non oon one as n.nn n.n non non nee as n.nn o.s oennnessno-sns ease oon non nos nan o.sn o.o non nnn oss an o.nn o.s onn nnn nus one n.on n.n sos nos nos non o.nn o.s onenesse nnn son nss son o.nn o.n oon nnn ons one n.nn o.s nss nns son non n.nn n.n nos nss snn nsn n.nn o.s eennnsssee-sns ens/oz oeon us ooon es oono es n.no es no.0 ease no posseseme_eessen see no nos nH geese sea seen seems some 383 homo nuneoosn> sequence memos; muse-semen mo 83? mo u8u0mm3 um 89 x83 o8 85d .8 83 38 ome 8 8:08am Enema one as oon o: sued senses awn: Bnhssflefie no esneefleeosssfi mated .o flan. 5m 0.2% u 833nm:V 388m new 3 u 5 wonder amaze 56 nnn nnn non one o.oo o.e nos ses onn owe n.nn o.e nos oss onn nee n.nn n.n oes ens nen one o.nn o.s eeeeeese non son sne sen n.nu o.e oen een nnn one n.nn o.e snn nnn own one n.oe n.n one see oes one o.oe o.s eeeeesesee-ses semen o.on es ooon us one so. one us 8.0 see. me assesses Essen sea ne fie ne wfiessne see 53 “sees seems eseeeee eomo nuesooseo A6388 0 0.3mm. 57 starches steadily increased which indicated that starch granules are very resistant to swelling and fragmentation. Most legune starches show these pasting characteristics (Schoch and Maywald, 1968; Kawauura, 1969; Lineback and K2, 1975; Vose, 1977; Comer and Fry, 1978; Billiaderis et al., 1979; Lai and Varriano-Marston, 1979; Naivikul and D'Appolonia, 1979; Lii and Chang, 1981; Sathe and Saltmkhe, 1981a). For the same concentration, navy bean starch was more viscous than either black or pinto bean starches. In this case, the effect of individual starch characteristics was a predaninant factor, because among three purified bean starches, there was no significant difference in starch contait (Table 3) . When the pasting characteristics of air-classified flours and purified starches were ccmpared (Figures 3 and 4), it could be seen that there were other factors affecting pasting characteristics of starch. There was no significant change in initial pasting temperature of black bean flour after purification; on the other hand, purified navy bean starch showed a small increase in this temperature, and purified pinto bean starch showed a significant increase. This significant increase in initial pasting tenperature of pinto bean starch might have bem due to the effect of its protein: the higher protein content, the lower initial pasting temperature. Because proteins can absorb water and display a tendency to swell without disintegration, this swelling fran water uptake might cause the lower initial pasting temperature and increase in viscosity (Edsall, 1965; Hermansson, 1979). Gaucher- Choquette and Boulet (1982) also reported that the concentration of protein can affect the viscosity change. From both Figures 3 and 4, it can be noticed that the flours show a higher rate of viscosity increase during the pasting cycle than do the purified starches. 58 The above results confirm that protein plays a significant role in viscosity changes. In an aqueous enviromlent, proteins are generally folded in compact state with the polar groups exposed to the aqueous phase and the hydrophobic groups located internally (Fox and Condom, 1981) . With the temperature increase and the applied shear force, the bean flour proteins may change their physical structure to an unfolding form, which is a prerequisite for the viscosity increase. As the pro- tein molecules unfold and their hydrophobic interior is exposed to the environment , they tend to associate with themselves at a rate related to the hydrophobic character of the protein (Salimath and 'Iharanathan, 1982) and thus, the viscosity steadily increases. Even thwgh the protein level of the flour affects the initial pasting temperature, it still does not cause any change in the flour viscosity pattern. The decrease in viscosity of flours during the 15 minholding phase at 50°C indicated that protein molecules were more susceptible to shear break- down than were starch granules . Figures 5, 6 and 7 show the Brabender pasting curves of air-class- ified navy, pinto and black bean flours at various concentrations with the individual points presented in Table 5 . The viscosity patterns for the various concentrations did not change . At low concentrations (6 g and 8 g/100 ml), viscosity stability was evident throughout the heating and cooling cycles. At low concentration 6 g/ 100 ml, the detected initial pasting tarperatures were rather high. At this low concentration, the starch content may be too low to pick up the early stages of gelatinization. At high concentration, an increase in viscosity during heating was evident. A higher tendency. to retrograde was also observed for the high concentration conditions , because of the greater probability for amylose molecules to associate and form hydrogen bonds among themselves. 59 8330.0. new? mongoose m . n ma n.n gwumflomoooo mag um BOG awn baa EnfimmmHouHmm mo 383.380 ohmoonmumv meg wfiummm ggm EEV 9.5... o: on on on on .18.)... I 350')... L grip J 4 '00-.in . on I 3 Al . no Allllv 3 A7 8 8 av 9338.20... .n manna 8w 8a m 8. o m. 1' 8a .A ) 8 8.. n ( CON 8. 80 60 £03300 H0mflfi 0u0fimo£a m . n ma 5" 0003035960 050g 00 “50$ :00o. 005m Enflmmfloufim m0 3:03.380 0.3.0833 00250 wfiummm 503.5 .0 0.93m 955 0EP o: 3 on on 8 3.00283 .3333 f 8. .. 8a 3.00.38. 8» A. m 03 O S. slo- 03 ..A ) 8 80 n ( . 02 3.32302 1 084 . , 80 81 on Al .H 3 13A 8 8 av 0.30.850... 61 83300 8m3n 0008008 m . m me an 083038080 0830> um 83 80h x003 0033000H0u3 no 383380 08:33 008 9.3000 H8833 3.5 we: 0: on on S on .603}... ‘ .. / 1 .6002le l 2:002:00. L loot-.0: 3 L 00 A..|.|I.v 00 A1 L 00 Al.|..lv no A 00 8 av 0.30.00E0e. .5 0.53m... 00. 000 a. m x 8. o S. 1... 000 .A \l 8 8. n I\ SN 80 000 62 Figures 8, 9 and 10 show Brabender pasting characteristics of purified navy, pinto and black bean starches at various concentrations with their specific points presented in Table 5. Their results are quite similar to those of air-classified flours, i.e. , no change in viscosity pattern, more stable paste through pasting cycle at low con- centration, and a high tendency to retrograde at higher concentration. The differmce between air-classified flours and purified starches is that the rate of increasing viscosity of purified starches is higher than that of air- classified flours at the same increasing rate of con- centration. The reason is that the purified starch contains a higher starch content than does air-classified flour (Table 3). Figures 11, 12 and 13 show the Brabender pasting curves of air- classified flours at different pH values of phosphate buffer solution, with their individual points presented in Table 6. Because most foods have pH values in the range of 4 to 8, this pH range was selected for this study. Variation within this pH range produced only minor effects on pasting characteristics and had no effect on the viscosity pattern. Most changes occurred after the tanperature reached 95°C. This minor effect of variation in pH on pasting properties might be explained by the buffer capacity property of specific type of proteins (Sathe and Salunkhe, 1981b) . Significant changes in pasting characteristics were noticed only in the pH 8 and the pH 4 solutions. There were no signif- icant changes in initial pasting tanperature of navy and black bean flours in the pH range of 4 to 8; however, one can see a trend to lower this tenperature as the pH increases . Pinto bean flour gave the opposite result: the higher the pH, the higher the initial pasting temperature. An increase in pasting viscosity was observed in navy and black bean flours with increasing pH values , while pinto bean flour showed the 63 00.3300 H033 000500050 m . n mm 5 000303000000 000305 30 A0300 003 .90: 003350 0.0 30033000 000000000 00>.H00 90.3000 H0033 «£3 05:. o: 8 2 3 3502...: I 3002...... 332.53 P b S Alliv 00 A] . no I no A 8 av 0.30.00E0... .0 9.5000 8. 8a 80 m x 8.. .O S H... 80 .A . \l .. 8 80 n ( 02. 80 8o 64 003300 H0355 000500050 m . m :0 3 08303000000 090.30> 00 50.030 0005 0030 0033.90 m0 30033000 0.000503 00350 @3300 00000503 .0 030 3.5 we: 0: 8 2 3 8 153).... I 4 8. 8a 1.8 i. I! p 80 M S s. m 3. 0|? 80 .A ) a 8 8. n (H 02 80 tootsie. . F 80 00 Allllv 00 9 a. I 00 AF 00 8 av 0030000000... 65 83300 H0355 000500050 m . m 00 3 000303000000 050.33 00 5030 0005 35 0033.30 3 30033000 00000003 00350 930000 3000503 .3 0.3me 3.5 2.: 0: on 2 3 8 2.63:...- - 8. . 8a 1 80 m x 8.. 0 m... 2.63).... 1 80 .M. ) 8 80 n ( . 80 .. 80 scoot-‘0' r . 80 00100 9 . 0010091 00 8 av 0020.00.20... 66 0000000000000 00 oon S 00 00.3300 03.05 000500050 08 035g 00 050mm? 00 0.05.0 0005 .9000 000300050000 .00 50000000000 00000003 00% E0000 00000505 5.5.5 08:. on 2 3 + 1 \ ‘ O ‘\ O O \ “ O u‘ C ‘ I “ 0 I‘ll O. I, “ ""|'-" . L SIS? .aoAll-YnaAw 8 av 0030000800. .2 005000 (08) KHSOOSM 67 0000000000000 08 80 \w 3 00 00000000 0000005 000500050 .00 00000> 00 0000000000 00 0900.0 0005 00000 00000000000000 mo 3000000000 00000003 00350 w000000 00000505 .3 00%00 0.5.5 2.0 o: .. 8. 0.08.. n 80 0.2.. 0.0... 0.3... L 8.. A w. 8.. O m. .I? 4 8... .A ) 4 8 So nu ( . 8» .. 08 k . L 80 81 on A! .34'I.v3fi a... 8 00 0500000800. 68 0000000009000 05 03% OH 00 00000000 0000005 000500050 .00 00000> 00 000000000 00 000.8 0005 050005 0000000000000 .00 00000000000 000000000 00250 w000000 00000500m .mH 000M005 2.5 2.0 0:. o. as. s 3 0| 0 . d \. q \ \\ x . 0 l 8.. \ s s s 00 n DON \ x s x . 80 x m. xx S \ o \ 8v \ O \ glo- \\ 4 80 .A . \\\ ) \ 8 \\\\\\\\\ 4 80 n ....... l\ 9...... \x . oooooooooooooooo I. 0.2... ......... 80. 0.2... 3::1: ....... 9.: ................ l 8” k 0 80 8184 P0010ko on 8 av 0500000000... 69 opposite effect. It may have been the effect of protein type or content. Edsall (1965) reported that the viscosity of a protein dispersion is a function of protein type (molecular size, shape, surface charge, concen- tration) , pH, tmerature, and shear rate. Figures 14, 15 and 16 show the Brabender pasting curves of purified starches at different pH values of phosphate buffer solution with indi— vidual points recorded in Table 6. As can be seen from these curves, changing pH values fran pH 4 to pH 8 caused only small effects on pasting properties of bean starches. There were no changes in the viscosity patterns, in the initial pasting temperature, or in the pasting proper- ties during heating up to 95°C. However, all bean starches showed the tendency of increasing viscosity with increasing pH. These characteristics indicated that bean starch granules were quite resistant to variations in pH. Vose (1977) and Comer and Fry (1978) reported a reverse viscosity pattern for pea starch: the higher pH gave the lower viscosity. When air-classified flours were coupared to their purified starches (Table 6), neither showed significant differences in viscosity pattern, pasting properties, or pH sensitivity, except for the air-classified pinto flour which showed a decreased viscosity with increasing pH values . This effect might have been due to the high protein content in pinto flour. The Braba'lder pasting curves of air-classified high starch flours of three bean types at an extended high temperature (95°C) holding period (30 min) are shown in Figure 17, with their specific points reported in Table 7. Prolonged heating tended to increase viscosity of flours. There was no change in the viscosity pattern. The curves steadily increased and showed less shear force breakdown during the 15 min holding period at 50°C as ccmpared to bean flours at standard 70 0000000000000 08 80 \m w 00 00000000 00.0005. 000500000 .00 000005. :0 000000000 00 000.000 0000 5 00000000 00 0000000000 00000003 000000 E0000 00000005 .00 00.0w0.m 0.5.5 08:. a: on - 2 on 8 1 .1 J 8. . 80 .. l 80 A x 03 O .u. m. 1 0|... 000 K ) . 8 80 n ( . 80 .. 80 L! ; 8a 0013? L. 001004 .00 8 av 050000050... 71 0000000000000 0B 80% w 00 00000000 000.000 000000000 00 00000> :0 000000000 00 000000 0000 00000 00000000 00 0080000000 000000000 00300 @0000 0000000000 :05 we: 0: ‘ “ ““‘ ‘ ““I .. 'I'-'l‘ ........ ..... .. ........ ooo .000 one. .......... . L b [P 0013A: ooTllvoaA: 00 8 av 0.30.0080... .2 95000 Sp 8a 80 M x 8. o S. cl? 8. .A. ) . 8 8.. n ( 80. 80 80 72 0000000000000 0E 80 \w w 00 00000000 0000000 000000000 00 00000> 00 000000000 00 000.000 0000 00000 00000000 00 00000000000 00000003 008 @0080 00080000 .00 000w00 0.5.5 0.0:. a: on on 00 8 1 8. .. 8« . 80 A w. 03 O 8. 1|.- 4 80 .A ) J 8 80 n (. . 80 L 000 0 80 8100A! .3130 00 8 av 0.30.0080... 73 . 00000000 .0003 330090 0.0 00 5 000 om. 0x000 0.039. 8.08 00500060 00.2 08800 um am as 8:0 o: 0080 .0030 000. B0flm$0u-8 05 0996 0309.0 0880.00 2.5 2.: ON p 00 p O. 00 0' 8118A! . ‘2' 'UV 00" 8 av 0.30.0080... .00 0.9000 8. 8.“ 80 A m... 0 8v 0 B. al.. 30 .A ) 8 80 n ( 8.. 80 8o 74 mom ohm «mm - ooH 0m - «Hm Hmu com omH mm - «mm mmm - umH am 00000050 mom owN 0m~ - 05H 00 - mfim Bum «0N m¢fi mm - mam com - owfi mg 00000mmmflo.uflm 8&0 cum «mm mmm - 5N0 NwH - can mum com mmq wmfi - omm mmm - mmq omfi 00000050 oqq owe mmq - moq «NH - mmo awe mmm 0H0 «0H - mqq muq - «mm moH umHMHmmmHo-Hflm .902 coon um coon um coon 00 00mm um comm um comm um pamaummya suumum fie om fie fl 55 om fie fl i.e. 980 00000 “0000 “mama “0000 Be .3380; H8338 0000.5 320.020 m . n ma fi $300 0863: 0.3030900 33 no swufi 050mg 00 0:000 0303 080 350 $23 Scum 38 8Q» 8 005000 0300.50 08 :5 8:0 o: 383 65m :02 8000mmflo.§ mo 83.050836 050.03 .5 0309 75 B 0.20 a 800360 055m Em a n 5 83g 50:“ 8m 03 03 .. 0% o: - a «mm 8.0 an as - 8.0 03 - 8m 5 efifiea owe 5 5 - :m a - Sm Sm 80 a emu - 5 E - 8m 03 efifimmfiofie 083 0.8 um moon em 9% um 0.3 pm 0.3 um 93 00 0888.0. fiefim fie om fie 2 fie om fie 2 ES. 980 000mm 000.00 .80me 000.00 85 30000005 70.083 m 3008 76 conditions (Figure 3) . This indicated that prolonged high tenperature heating had an effect on the protein structure such that the structures were more resistant to shear force. As discussed earlier, protein denaturation causes physical structure changes from native, folding protein to the unfolding form which results in increased viscosity. Prolonged heating may enhance an increase in hydrophobic hydration and result in a more viscous and higher shear resistant paste. The Brabender pasting curves of purified starches of three bean types at an extended high tenperature (95°C) holding period (30 min) are shown in Figure 18, with specific points presented in Table 7. They all showed a steady increasing viscosity with no shear breakdown. Black and pinto starches show a greater increase in viscosity than did the navy bean starch which was quite stable. This difference in viscosity is not explained by differences in the amylose content of starches determined by this study (Table 3) . Differences in molecular structure of starch components may be hypothesized. 'Ihe Brabender pasting curves of air-classified flours at an extended low temperature (50°C) holding period (30 min) are illustrated in Figure 19 with specific points recorded in Table 7. While carparing with the curves in Figures 3 and 17, it may be concluded that the sta- bility of the protein viscosity depended on the amount of the heat treat- ment. This conclusion agrees with a previous study by Kinsella (1982). Figure 20 shows the Brabender pasting curves of purified starches at an extended low tanperature (50°C) holding period (30 min). When canpared with Figures 4 and 18, it may be seen that there is no shear breakdown during this longer holding period at a low temperatmre. These results indicate that bean starches are quite resistant to heat and to shear force . 77 efiufiom 00003 82%20 n.n 00 fi Afie one 80.80 mfififi some 005000080 . 00.2 80830 8 e... Se 8:0 8 039300 5.8 800000 mo 0908 05080 808030 .8 88000 65.5 05... ON w 00 p O. . 8 0' ace: 5! GOA-Illa? .02! 'IVDOAI 8 av 0020.00.00... (08) AMSOOSM 78 80038 00008 80.00090 2.. 00 fi Afie one 80.80 00.6090 8.02 00020060 32 08806 00 em :e 8:0 o: mesa €030 ewes Bfifimmmfifie ma 8% 000030 0008030 3.5 9:: ON -. 8. O. OO O. N 1 d a 02-: L L :3. I. 89... L F b L can JVonAl . 3AIIIVquW on 8 av 0030000800. .00 005000 (n8) finsoosm 79 8338 .0005 820080 n.n 00 fi fie one 80.80 0.0.608 8.03 00030080 33 009000000 :0 00 38 oon 8 006.0000 9000 00000050 mo 00% 9300.00 0008005 45.5 05:. ON p OOv O. OO O? u d C. 1 IV 8.... \.\\ :08“ F n p - .8 0.9000 .. OON ODA {YOGA {DO A'quoAW 8 av 0.5.0.0900... OON OOO OOO (n8) Ausoosm 80 IVblecular Structure Because the gelatinization phenmienon and pasting characteristics of the navy, pinto and black bean starches are very complex, differences in their properties carmot be ezqalained by anylose-amylopectin content alone (Table 3). Other factors such as granule size, their molecular structure and organization are important (Banks and Greemood, 1975). In recent years, research interest has focused on the molecular structure level which can be determined by combining enzymatic methods with size exclusion chranatography techniques (Mercier, 1973; Juliane and Perdon, 1975; Biliaderis et al. , 1979 and 1981; Abdul-Hussain and Varriano-Marston, 1982) . Papantonakis (1980) and others have applied the size exclusion methodology with high pressure liquid chromatography to provide molecular weight (size) distribution (Mm) data to characterize the viscosity of paper coating starches. Starches, like polymers, contain molecules with a distribution of different molecular sizes or chain lengths. This study applied the Papantonakis (1980) tecl'miique to the bean starches that were degraded with alpha and beta amylase enzymes. In the process of investigating the molecular structure of bean starches, a mnber of elution patterns were obtained using pure can- ponents: potato amylopectin, potato amylose, maltose and glucose. The elution patterns for the different canpounds are presented in Figure 21 . Amylopectin elutes faster than amylose, maltose and glucose, respectively. This indicates that the molecular size of amylopectin is larger than that of amnylose, maltose or glucose. In addition, the potato amylopectin profile showed a bimodal distribution which indicates that the potato amylopectin may contain two ranges of ccmponents with different molecular 81 08000 800005000 00m $0805 8 008880 00.00800 00 000000.00 8003a . 0N 0053.0 2.5 0£:_o> cozew v.0 NJ. 0.0 0.? 0.6 Q.N NJ 0 q - a — q - -. I I s ‘ Hgfllzbthr‘ullflllgl Q... ‘tfl’cfliflflttqu‘lil'. “. b . I I i c :‘ . E. ‘ ’ \ I ‘ I onOO-Zu .OIOIIIO (memos SA uonngos nouns) xepu! enuoeueav 000:2: ......... one—>50 0.300 0.001010 cZoooBan 00300 82 sizes . The difference in elution volume between potato amylose and amylopectin was not large enough to show the canpletely separated peaks when potato starch (amylose and amylopectin) was applied to these coltms. To increase the resolution of the HPLC analysis , many canbinations of dimethylsulfoucide (11430) and water were tested. The best resolution was obtained using 1160:1120 (9:1, v/v). Another factor which contributed to poor separation was the colum' s lack of efficiency. The efficiency of a colurn depends on the alignment of the highly porous and delicate silica gel particles. This particle alignment is very sensitive to any physical or thermal shock which may cause the particles to shift to a more dense, but less efficient condition. Due to some conditions during previous use, the colum packing was denser than it should have been. Figures 22, 23 and 24 show elution profiles of purified navy, pinto and black bean starches and of the bean starches after being treated with either alpha or beta amylases. As expected, there was no separation of the amylose and amylopectin peaks; however, the high molecular weight elution peak of the control trailed into the higher elution volunes , which may indicate the presence of amylose. The same single elution volume peaks were obtained for all three bean types. These elution volunes were close to those exhibited by potato amylose, but larger than those of potato amylopectin. This indicates that the differences in amylose and amylopectin molecular size ofthethreebeantypeswerenot largemmlghtobeshovmby these colums, and that their amylose and amylopectin unlecular sizes were similar to standard potato amylose, but lower than standard potato amylopectin. Potato starch exhibits much higher viscosity than bean starches at similar concentrations; however, the viscosity patterns are different. 83 085000 0000008000 00w $50008 no 00000050 0000 0000 03000 08. 800803 38080 .090 20 20088 6030 :03 .90: B00000 00 $00080 80030 .8 0.8000 2.5 mE:_o> cozaw v.0 «.5 0.0 0.1 . 0.0 v.“ N.w -‘-’-"l-- -llinlilalllutllil‘ulcllliIllullllnl.lv‘p‘.0 . 0 .l 0 a (wanes '31s uoumos 001235) xepug eAg1oeJ;eav ”WW—>8“ a 0% 000_>E0 n ........ .ozcou 84 08000 8000008000 00w 809808008 8 00000.90 0000 0000 03000 .3 85000.3 89080 090 E» 20083 €03... EB 850 800.80 00 800080 80030 .00 000E 2.5 0£:_o> 88$ «.0 .0... 0.0 06 0.0 «.0 «.0 m v E .. H 0 w M. J m w 0 1. u I. A .0 a m. m. M 0. 000.300 0 06010.0. ( 000380 0 ......... .OZCOD 85 08000 8000000000 00m 500960700005 8 00000.90 0000 0000 00000 .00 8&5»: 39080 090 90 30083 608... .08 083 B00000 00 800080 80.30 1.00 cEv 0&0_o> co_0:_m «.0 0.0 0.0 0.0 0.0 «.0 000_>E0 0 000380 0 .Opacoo E§8?030 tfinlnlc’ P3,. 1 (WOMOS ’SA uonmos 00.1035) :0 00000 xepug eAnoeJlaav 86 Potato starch shows a typeAviscosity pattern, c1mracterized by starch granules swelling enormously, internal bonding forces becoming tenuous and fragile toward shear, and Brabender curves showing a high pasting peak followed by rapid and najor thinning during cooling. Potato starch has a lower (approximately 202) amylose content . Therefore , the differaices in molecular size may be one important factor to affect pasting characteristics . Unfortunately, there was no difference shown among the three bean types . When alpha amylase was applied to digest raw bean starches in pH 5.8 acetate buffer and incubated at 35°C for 24 hours, there were no differences among the elution patterns (Figures 22, 23 and 24) of the three bean types. All showed single peaks at the same elution volune of glucose with some tailings . These tailing portions may represent sane imcouplete digested oligosaccharide such as isanaltose (two glucose units linked by a 1,6 linkage), because alpha amylase can Split only a 1,4 linkages. Generally, different raw starches have different susceptibility to degradation by alpha amylase (Marshall, 1980) . 'Ihe same susceptibility to alpha amylase degradation was obtained fran all three bean starches in this study, which indicates a similarity in unlecular organizatim of the starch granules or may implicate the use of excessive incubation periods . Tne results from applying beta amylase in a pH 4.8 acetate buffer solution which was incubated at 35°C for 24 hours to navy and black bean starches show a similar peak shape, with broad shouldering with increased elution volume. However , pinto bean starch gave a sharper peak without tailing. These results agree with their pasting characteristics, because navy and black bean starches have similar pasting preperties, which differ fran those exhibited by pinto starch. Beta amylase 87 hydrolyzes al,4 glucose linkages fran the non-reducing ends of starch components to form maltose (two glucose units linked by «11,4 glucose linkage) and will cease activity at the branching points of al,6 glucose linkages. Therefore, these differaices may be associated with the different unlecular structure of amylopectin. 'lhese elution profiles are useful in hypothesis development and not confirmatory for the molecular structure of the starches. SLIMRY AND CDNCLUSIG‘IS Results of the chemical canposition study indicated that air- classified high starch flours contained a relatively high ammmt of protein (15.512 to 19.062); however, the starch isolation method modified by Naiviknl and D'Appolonia (1979) was efficient enough to prepare the high purified starches which were significantly lower in protein, ash, fat, and E'NDF. Among the three bean starches, the amylose content of navy bean starch was not significantly different from that of pinto or black bean starches, but there was a significant difference (P < 0.05) between pinto and black bean starches. As a result of sugar analysis, the same sugar order pattern, from high to low (sucrose, stachyose, glucose and raffinose, with no inositol detection) was shown in all three bean flours. After purification, there were no sugars detected. 'I'ne Brabender pasting characteristic study indicated that all flours and starches showed a type C viscosity pattern, which is restricted swelling, increasing in viscosity, with no peak during heating and tending to retrograde during cooling. The pasting curves of flours illustrated higher rates of increasing viscosity during heating and cooling cycles than those of starches . In addition, flours showed shear breakdown during a holding period at 50°C, while starches showed slightly increased viscosity. Variation in pH (range of l. to 8) showed only minor effects on pasting characteristics for both flours and starches. During an 88 89 extended high temperature holding period (95°C, 30 min), both flours and starches showed moderately increased viscosity which indicated their resistance to thermal breakdown. During an extended low tamerature holding period (50°C, 30 min), flours showed moderately decreased viscosity; whereas, starches showed slightly increased viscosity. This indicated that starch granules themselves were resistant to shear force; however, the protein in flours was not stable to shear force mder these conditions . From the molecular structure study, using gel permeation HPLC, the beta amylase hydrolysate elution profile of pinto bean starch differed frcm that of navy and black bean starches, suggesting a difference in pinto bean amylopectin molecular structure. WIG‘ISI‘URFLRIHERRESEARQ-I In order to clarify the differaices of amylopectin molecular structure, further research should be done on the degree of branching, as well as the length of branches by using pullulanase, debranching enzyme, and isoamylase combined with gel permeation HPLC to character- ize the fine structure of amylopectin. Study of amylose molecular structure is also needed because the size of this starch component l'es affected molecular organization. The relationship between molecular stmcture and pasting properties should be observed for both starch calponents , amylose and amylopectin. According to the extremely complex gelatinization phenomenon, not only the structure of a granule and its ccnponents, but also other factors such as granule size, phosphorous content, bound lipids, and protein are important (Banks and Greenwood, 1975; Biliaderis et al., 1981). 90 APPENDIXA APPENDIX A Table 8. Analysis of variance of pasting characteristics of air- classified high starch flours and purified starches fran navy, pinto and black beans at various concentratims Viscosity (BU) Source of after 15 min Variation df holding at 95°C Mean Squares Main Effects 6 649,096.69*** Bean Type 2 l38,6l3.50*** Starch Treatment 1 586,245.44” Concentration 3 l , 165 , 963 . 25*** Two-Way 9 81 , 552 . 57*** Bean Type x Starch Treatment 2 6,260.19*** Bean Type x Concentration 5 58 , 051 . 46*** Starch Treatment x Concentration 2 209,436. 69*** Three-Way Bean Type x Starch Treatment x Concentration 4 19 , 959 . 57*** Residual 20 76.90 CV (Z) 2.73 91 APPENDIX B APPENDD(D Table 11. Analysis of variance of pasting characteristics of air- classified high starch flours fran navy, pinto and black beans at different high or low temperature holding periods Viscosity (BU) Source of extended high extended low Variation df temp. holding temp. holding Mean Squares Main Effects 3 123,025.94*** 142,697.72*** Bean Type 2 157,872.25*** 213,505.08*** Time 1 53,333.33” 1,083.00** 'I‘wo-Way Bean Type x Time 2 4,804.08*** 36.75* Residual 6 27.00 38.00 CV (Z) 1.21 1.29 94 APPENDIXC APPENDR E Table 12. Analysis of variance of pasting characteristics of purified starches fran navy, pinto and black beans at different high or low temperatlue holding periods Viscosity (BU) Source of extended high extended low Variation df tanp. holding temp. holding Mean Squares Main Effects 3 61,597.16*** 58,143.42*** Bean Type 2 84,329.08*** 85,360.08*** Time 1 16 , 133 . 33*** 3 , 710 . 08* Two—Way Bean Type J: Time 2 473.08** 217.58* Residual 6 35 . 83 278 . 92 CV (Z) 1.77 3.80 95 LIST OF REFERENCES LIST OF REFERENCES AACC. 1962. "AACC Approved Methods," 7th ed. Americm Association of Cereal Chemists. St. Paul, W. Abdul-Hussain, S. and Varriano-Marston, E. 1982. Amylolysis of pearl millet starch and its fractions by pearl alpha-amylase. Cereal Chem. 59:351. Adamo, M.W. 1972. On the quest for quality in the field bem. In "Nutritional Improvement of Food legmes by Breeding . " Protein Advisory Group of the United Nations System, United Nations, NY. Agbo, N.G. 1982. Genetic, physico-chemical and structtual parameters affecting temcuue of dry edible bems. Ph.D. thesis. Michigan State University, East Lansing, MI. Aguilera, J.M., Lusas, E.W., Uebersax, M.A. and Zabik, M.E. 1981. Process development, characterization and utilization of dry heated navy bem products. Research Agreement No. 59-2481-0-2-001-0, United States Deparment of Agricultlue Scimce and Education Admini- stration. Arlington, VA. Aguilera, J.M., lusas, E.W., Uebersax, M.A. and Zabik, M.E. 1982. Development of food ingredients from navy bems (Phaseolus vulgaris) by roasting, pin milling, and air classification. J. Food Sci. 47: 1151. Akpapunan, M.A. and Markakis, P. 1979. Oligosaccharides of 13 Americm cultivars of cowpea (Vigna sinesis). J. Food Sci. 44:1317. Anker, C.A. md Geddes, W.F. 1944. Gelatinization studies upon wheat and other starches with the amylograph. Cereal Chan. 21:335. Augustin, J., Beck, C.B., Kalbfleish, G., Kagel, L.C. and Mathews, R.H. 1981 . Variation in the vitamin and mineral content of raw and cooked commercial Phaseolus vulgaris classes. J. Food Sci. 46:1701. Bakker, F.W., Patterson, R.J. and Bedford, C.L. 1973. Production of instmt bem powders. In "Nutritional Aspects of Common Bems and Other Legme Seeds as Animal and Human Foods." Jaffe, W. (ed.). Archivos Latinoamericmos de Nutricion , Venezuela. Banks, W. and Greenwood, C.T. 1967a. Physicochemical studies on starches. Part II. The incomplete beta-amylolysis of anylose: A discussion of its cause and implications. Staerke. 19:197. 96 97 Bmks, W. and Greenwood, C.T. 1967b. The fraction of laboratory- isalated cereal starches using dimethylsulfaxide. Staerke. 19:394. Banks, W. and Greenwood, C.T. 1975. "Starch and Its Cmpanents." Halsted Press, New York, NY. Banks, W., Greenwood, C.T. and Muir, D.D. 1970. The characterization of starch and its canponents. Part II . The semi-micro estimation of the starch content of cereal grains and related materials . Staerke. 22: 105. Barker, S.A., Hatt, B.W., Marsters, J.B. and Somers, P.J. 1969. Fractionation of dextrm and hyaluronic acid on porous silica beads . Carbohydrate Res. 9:373. Bau, M.C. and Alexander, R.J. 1979. Enzymatic procedure for deter- mination of starch in cereal products. Cereal Chen. 56:364. Bear, R.S. 1942. Review of studies an starch componmts: amylase and amylopectin. J. Amer. Chem. Soc. 64:1388. Becker, R., Olson, A.C., Frederick, D.P., Kan, S., Gumhnann, M.R. and Wagner, J .R. 1974 . Condition for the autolysis of alpha-galactosides and phytic acid in California Small White bems. J. Food Sci. 39:766. Bedford, C.L. 1972. Bean storage and rocessing. In "Proceedings of the International Dry Bean Symposium, ' August 22-24. Michigm State University, East Lansing, MI. Biliaderis, C.G., Grmt, D.R. and Vase, J.R. 1979. Pblecular weight dis- tributions of legume starches by gel chromatography. Cereal Chem. 58: 496. Biliaderis, C.G., Grmt, D.R. and Vase, J.R. 1981. Structtual charac- terization of legume starches. I. Studies on amylase, amylopectin, md beta-limit dextrins. Cereal Chen. 58:496. Bremer, J .M. 1965. Total nitragm. In 'Ththods of Soil Analysis. Part 11. Chemical and Microbiological Properties." Black, C.A. (ed.). Amer. Sac. Agran. Madison, WI. Bressmi, R. 1975. legumes in human diets and how they might be improved. In "Nutritional Improvement of Food legumes by Breeding . " Milner, M. (ed.) Jalm Wiley md Sons, New York, NY. BLukitt, D.P. 1971. Epidemiology of cmcer of the colon and rectum. Cmcer 28:3. Blur, H.K., Kan, S., and Morris, H.J. 1968. Cooking rates of dry bems as influenced by moisttue cantmt, temperattue, and time of storage. Food Technol. 22:336. Caesar, G.V. 1932. Consistency chmges in starch pastes. Ind. Eng Chem. 24:1432. Caesar, G.V. and More, E.E. 1935. Consistency changes in starch pastes. Ind. Eng. Chem. 27:1447. 98 Galloway, D.H., Hickey, C.A. mdMurphy, E.L. 1971. Reduction of intestinal gas-forming properties of legumes by traditional and experimental food processing methods. J. Food Sci. 36:251. Carpenter, R.J. 1981. The nutritional contribution of dry bems (PhaseoZus vulgaris) in perspective. J. Food Technol. 35:77. Carroll, B. and Clmeng, H.G. 1960. In "Methods in Carbohydrate Chemistry, Vol. IV." Whistler, R.L., Smith, R.J. and BeMiller, J.N. (eds.) Academic Press, New York, NY. Carvalho, C.C.C., Jmse1, G.R. and Harper, J.M. 1977. Protein qualiqt evaluation of an instmt bem powder produced by dry heat processing. J. Food Sci. 42:553. Cerm'ng-Beroard, J. and Filiatre, A. 1976. A cmparison of the carbohydrate composition of legume seeds: harsebems, peas, and 1Lpines. Cereal Chem. 53:968. Cerning-Beroard, J., Sapsonik, A. md Guilbat, A. 1975. Carbohydrate camposition of horsebems (Vicia faba L.) of differeit origins. Cereal Chem. 52:125. Chang, C.R. md Tsm, C.C. 1981. Characterization md heat stability of trypsin inhibitors frum rye, triticale, and wheat samples. Cereal Chem. 58:211. Chmg, K.C. aid Satterlee, L.D. 1979. Chemical, nutritional and microbiological quality of a protein concentrate fran culled dry bems. J. Food Sci. 44:1589. Charley, H. 1970. legmes. In"Foad Science." Charley, H. (ed.). John Wiley and Sons, New York, NY. Caner, F.W. and Fry, M.R. 1978. Ptuification, modification, and properties of air-classified pea starch. Cereal Chen. 55:818. Davies, N.T. and Nightingale, R. 1975. The effects of phytate on in- testinal absorption and secretion of zinc, and whole-body retention of Zn, copper, iron mdmenganese in rats. Br. J. Nutr. 34:243. Dawson, E.H., Lamb, J.C.,Toepfer, E.W. and Warren, H.W. 1952. Development of rapid methods of soaking and cooking dry bems. Bull. 1051, USDA Tech., Washington, D.C. De Balmd, A.R., Garner, G.B. and O'Dell, B.L. 1975. Idmtification md praperties of "phytate" in cereal grains and oilseed products. J. Agric. Food Chem. 23:1186. Deschamps, I. 1958. Peas and bems. In "Processed let Protein Foodstuffs." Altschul, A.M. (ed.). Academic Press, New York, NY. Dintzis, F.R. and Tobin, R. 1974. bblecular—sieve chromatography of amylase and dextran aver porous glass. J. Chromatogr. 88:77. 99 Donavm, J .W. 1979. Phase trmsitians of the starch-water system. Biopolymers 18:263. Dubowski, K.M. 1962. An o-toluidine method for body-fluid glucose determination. Clin. Chem. 8:215. Duffus, C.M. md Mudoch, S.M. 1979. Variation in starch gramle size distribution and amylase cantelt dmring wheat endosperm develop- ment. Cereal Chem. 56:427. Earle, F.R. md Milner, R.T. 1944. Improvement in the determination of starch in corn md wheat. Cereal Chen. 21:567. Ebermenn, R. md Schwarz, R. 1975. Fraction of starch by gel fil- tration on agarose beads. Staerke. 27:361. Edsall, J.T. 1965. In "Food Proteins." Fox, P.F. md Condon, J.J. (eds.) Applied Scieice Publ. , London md New York. Elbert, E.M. 1965. Starch changes during heating in the presmce of moistxue. J. Hale Econ. 57:197. Elias, L.C., De Fernmdez, D.C. md Bressmi, R. 1979. Possible effect of seed coat palyphenolics on the nutritional quality of bem protein. J. Food Sci. 44:524. Ekpe'xyong, T.E. md Borchers, R.L. 1980. Effect of cooking on the . chemical camposition of winged bems (Psophocarpus tetragona Zobus) . J. Food Sci. 45:1559. Elleuieder, G. , Blmco, S. md Bandoni, A. 1981. Therml inactivation of trypsin inhibitors in aqueous extracts of soybems. Studies on substmces that accelerate inactivation. Cereal Chem. 58: 291 . Ellerrieder, G., Geronazzo, H. md de Bojarski, A.B. 1980. Thermal inactivation of trypsin inhibitors in aqueous extracts of soybems , pearmits , md kidney bems: presence of substmces that accelerate inactivation. Cereal Chem 57:25. FAD let Production md Protection Paper. 1977. Food legumes: distribution, adaptability md biology of yield. Rare, Italy. Fleming, S.E. 1981a. A study of relationships between flatus potmtial and carbohydrate distribution in legume seeds. J. Food Sci. 46:794. Fleming, S.E. 1981b. Flatulmce activity of the smooth-seeded field pea as indicated by hydrogm production in the rat. J. Food Sci. 47:12. Fleming, S.E. md Reichert, R.D. 1980. Note an a modified method for the quantitative determination of starch. Cereal Chem. 57:153. Fleming, S.E. md Reichert, R.D. 1983. A carparison of the flatulmce potential of field pea md soybem seed fractions. Cm. Inst. Food Sci. Technol. J. 16:30. 100 Fordham, J.R., Wells, C.E. md Chen, L.H. 1975. Sprouting of seeds md nutrient cumposition of seeds md sprouts. J. Food Sci. 40:552. Fax, P.F. md Condom, J.J. 1982. ”Food Proteins." Applied Scimce Publ. , London and New York. French, D. 1972. Fine structtue of starch md its relationship to the orgmization of the granules. J. Jpn. Sac. Starch Sci. 19:8. Freudenberg, K., Schaaf, E., Dumpert, G., md Ploetz, T. 1939. Review studies on starch components. Natluwissemschaftem 27:850. Gaucher-Choquette, J. md Boulet, M. 1982. Effect of alkali treatment on viscosity md an the sulfhydryl md disulfide content of proteins. Cereal Chem. 59:453. Geervani, P. and Theaphilus, F. 1982. Influence of legume starches on protein utilization md availability of lysine and methionine to albino rats: a literatiue review. Nutr. Abstr. Rev. Ser. A. 52:1237. Ghiasi, K., Hoseney, R.C. md Varrian-Marston, E. 1982. Gelatinization of wheat starch. I. Excess-water system. Cereal Chem. 59:81. Gilbert, C.A. and Spragg, S.P. 1964. Iodimetric determination of amylase. In "Methods in Carbohydrate Chemistry, Vol. IV." Whistler, R.L. (ed.). Academic Press, Nev York, NY. Galdstein, J.L. and Swain, T. 1965. The inhibition of mzymes by tamins. Phytachem. 4:185. Gates, J.C., Koch, U., md Brumer, J.R. 1979. Isolation of a trypsin inhibitor frum navy bems by affinity chromatography. Cereal Chem. 56:525. Goodhart, R.S. md Shils, M.E. 1980. 'Madem Nutrition in Health md Disease," 6th ed. Lea & Febiger, Philadelphia, PA. Graham, G., lVbrales, E., Placka, R.P. md MacLem, W.C. 1979. Nutri- tive value of brom and black bems for infmts md small children. Amer. J. Clin. Nutr. 32:2362. Greenwood, C.T. 1979. Observation on the structure of the starch granule. In "Polysaccharides in Food. " Blmshard, J .M. md Mitchell, J.R. (eds.) Butterworth's md Compmy Ltd., London, UK. Greemaood, C.T. md Thampsan, J. 1962. Studies on the biosynthesis of starch grmules. II. The properties of the cmpmmts of starches fram smooth- aid wrinkled- seeded peas during growth. Biochem. J. 82: 156. Griffith, D.W. md Jones, D.R. 1977. Cellulase inhibition by tamins in the testa of field bems (Vicia faba). J. Sci. Food Agric. 28:983. 101 Gupta, K. md Wagle, D.S. 1980. Changes in mtinutritianal factors ' germination in Phaseolus mungoreus, a cross between Phaseolus mango L. (Ml-1) and Phaseolus aureus (T1). J. Food Sci. 45:394. Hall, D.M. and Sayre, J.G. 1971. A scanning electron microscope study of starches: III. Miscellmeous starches. Textile Res. J. 44:880. Hmes, C.S. 1937. Review studies an starch components. New Phytol— agist 36:189. Harper, J.M. md lorenz, K. 1974. Production md evaluation of salt bed roasted full-fat soy fleu. Iebmsmfi.tt-Wiss. u. — Technal. 7:268. Harris, P.M., Mack, E. md Blake, F.C. 1928. Review studies on starch campmmts. J. Amer. Chem. Soc. 50:1583. Hagglam, E. 1974. Polyphenal-protein interactions. Biachen. J. 139: 5. Haworth, W.N., Hirst, E.L. md Isherwoad, F.A. 1937. Review studies on the starch gramiles. J. Chem. Soc. 784:577. Hermanssm, A.M. 1979. Methods of studing functional characteristics of vegetable proteins. J. Amer. Oil Chem. Soc. 56:272. Hanovar, P.M., Shih, C. md Iie1er, LE. 1962. Inhibition of the growth of rats by ptuified hemgglutinin fractions isolated frum Phaseolus vulgaris. J. Nutr. 77:109. Hymowitz, T., Collins, F.I., Pmezner, J. md Walker, W.M. 1972. Relationship betweei oil, protein, md sugar in soybean seed. Agron. J. 64:613. Jaffe, W.C. 1969. Heregglutinins. In "Toxic Constituemts of let Foodstuffs." Liener, LE. (ed.). Academic Press, NewYark, NY. Johnson, L.A., Deyae, C.W., Hoover, W.J. and Schwenke, J.R. 1980. Inactivation of trypsin inhibitors in aqueous soybem extracts by direct steam infusion. Cereal Chem. 57:376. Juliana, B.O. md Perdon, A.A. 1975. Gel md molecular properties of nonwaxy rice starch. Staerke. 27:115. Kakade, M.L. md Evms, R.J. 1965. Nutritive value of navy bems (Phaseolus vulgaris). Brit. J. Nutr. 19:269. Kakade, M.L. md Evms, R.J. 1966. Effect of soaking md germinating an the nutritive value of navy bems. J. Food Sci. 31:781. Kakade, M.L., Simons, N. md Limer, T.E. 1969. An evaluation of nattual versus synthetic substrates for measuring the mtitryptic activity of soybem samples. Cereal Chem. 46:518. Kawanua, S. 1969. Studies on the starches of edible legute seeds. J. Jpn. Sac. Starch Sci. 17:19. 102 Kawanua, S. md Fukuba, H. 1957. Studies on legume starches. II. Viscosity behaviors. Tech. Bull. Fac. Agric. Kagawa Univ., Japm, 9:38. Kawamua, S., Tuboi, Y. md Huzii, T. 1955. Studies on legme starches. I. MicroscOpic observation on the grmules of starches frum some Japmese legmes. Tech. Bull. Kagawa Agric. Coll. 7:87. Kay, D.E. 1979. Haricat bem. In "Food Legumes." Kay, D.E. (ed.). Tropical Products Institute, Imdm, UK. Khm, A.V. 1972. Rice drying md processing equipment for South East Asia. ASAE Paper No. 72-339, ASAE, St. Joseph, MI. Kinsella, J .E. 1982. Relationships betwem struct1ue md functional properties of food proteins. In "Food Proteins." Fonz, P.F. md Condon, J.J. (eds.). Applied Science Publ. , Iondon md New York. Kan, S. 1979. Effect of soaking telperattue on cooking md nutritional quality of bems. J. Food Sci. 44:1329. Kan, 8., Wagner, J.R., Guadagni, D.C. md Horvat, R.J. 1970. pH adjustment control of oucidative off-flavors during grinding of raw legme seeds. J. Food Sci. 35:343. Kruger, J .E. md ‘Marchylo, B.A. 1982. High-performance aqueous gel permeation chrumatographic malysis of beta-limit dextrin hydrolysis by malt hard red spring wheat, malted dunmwheat, md fungal (Aspergillus aryzae) alpha amylases. Cereal Chem. 59:488. Iai, C.C. md Varriano-Marston, E. 1979. Studies on the character- istics of black bem starch. J. Food Sci. 44:528. Iarston, B.L., Giles, K.A. md Jamess, R. 1953. In "Methods in Carbohydrate Chemistry, Vol. IV." Whistler, R.L., finith, R.J. md BeMiller, J.N. (eds.) Academic Press, New York, NY. Teach, H.W. 1965. Gelatinization of starch. In "Starch: Chemistry and Technology, Vol. I." Whistler, R.L. and Paschall, E.F. (eds.). Academic Prass, New York, NY. Leach, H.W., McGowan, L.D. md Schoch, T.J. 1959. Structtue of the starch granule. I. Swelling md solubility patterns of various starches. Cereal Chem. 36:534. Leach, H.W. and Schoch, T.J. 1962. StructLue of starch gramle. III. Salubilities of grmular starches in dimethylsulfaxide. Cereal Chem. 39:318. Libby, R.A. 1970. Direct starch malysis using 11130 solubilization md glucoamylase. Cereal Chem. 47:273. Liener, T.E. 1962. Toxic factors in edible legumes md their elimin- ation. Amer. J. Clin. Nutr. 11:281. 103 Limer, LE. 1975. Antitryptic md other mtirmitritianal factors in legumes. In "Nutritional Improvement of Food Legures by Breeding. " Milner, M. (ed.). John Wiley md Sons, New York, NY. Limer, I.E. md Kakade, M.L. 1969. Protease inhibitors. In "Toxic Constitumts of let Foodstuffs." Limer, M.L. (ed.). Academic Press, New York, NY. Liener, T.E. dehompson, R.H. 1980. In vitro md in vivo studies on the digestibility of major storage protein of the navy bem (Phaseolus vulgaris). Qual. Plant. 30:13. Lii, C.Y. md (Ihémg, S.M. 1981. Characterization of Red Bem (Phaseolus radiatus var.Aurea) starch md its noodle quality. J. Food Sci. 46:78. Lineback, D.R. and Sayeed, A.L. 1971. Effect of dimethylsulfoxide on amylase activity. Carbohydrate Res. 17: 453. Lineback, D.R. md Ke. C.H. 1975. Starches and low-molecular weight carbohydrates from chick pea md horse bem flatus. Cereal Chem. 52:334. Lorenz, K.J. 1979. The starch of the faba bem (Vicia faba L.). Staerke. 31:181. Lustinec, J., Hadacava, V., Kaminek, M. md Prochazka, Z. 1983. Quantitative determination of starch, amylase, md amylopectin in plmt tissues using glass fiber paper. Anal. Biochem. 132:265. Maddex, R.L. 1978. Hmdling md storage. In "Dry Bem Production Principles md Practices." Bull. E-1251, Agric. Expt. Sta. , Michigm State University, East Imsing, MI. Maga, J.A. 1982. Phytate: Its chemistry, occureice, food interactions, nutritional significmce, md methods of malysis. J. Agric. Food Cheuh 30:1. Mamers, D.J. md Matheson, N.K. 1981. The fine structtue of amylo- pectin. Carbohydrate Res. 8:102. Marquardt, R.R., Ward, A.T. md Evms, L.E. 1978. Comparative prOperties of tannin-free and tamin-containing cultivars of faba bems (Vicia faba). Cm. J. let Sci. 58:753. Marshall, J .J . 1980. Enzymic degradation of starch. In "Mechmisms of Saccharide Polymerization md Depalymerization. " Academic Press , New York, NY. Mazurs, E.G., Schoch, T.J. md Kite, F.E. 1957. Graphical malysis of the Brabender viscosity curves of various starches . Cereal Chen. 34:141. PbEwm, T.J., Dronzek, B.L. md Bushuk, W. 1974. A seaming electron microscope study of faba bem seed. Cereal Chem. 51:750. 104 M:Gruire, J .P. and Erlmder, S.R. 1966. Qumtitative isolation and dispersion of starch from corn kernels without degradation. Staerke. 18:342. Madcalf, D.C. and Giles, K.A. 1966. Effect of a hyotropic ion series on the pasting characteristics of wheat and corn starches. Staerke. 4:101. Mejia, E.G. 1980. Effect of various conditions of storage in general aspects of bem hardening. Final Rpt. UHU Newsletter 4(1):Apri1. Mercier, C. 1973. The fine structtue of corn starches of various amylase-percentage: waxy, normal md amylanaize. Staerke. 25:78. Meyer, K.H. 1952. Review studies on starch granules. Experientia. 8:405. Meyer, K.H. and Gibbons, C.C. 1957. The present status of starch chemistry. Adv. Enzymol. 12:341. Milic, B.L., Stojmovic, S. and Vucurevic, N. 1972. Lucerne tannins. II. Isolation of tamins from lucerne, their nature md influence on the digestive enzymes in vitro. J. Sci. Food Agric. 23:1157. Miller, B.S., Derby, R.I. and Trimbo, H.B. 1973. A pictorial erplma— tion for the increase in viscosity of a heated wheat starch-water sus- pension. Cereal Chem. 50:271. Minor, J .L. 1979. Gel permeation chromatography of polymer frum wood. J. Liq. Chrcmatogr. 2:309. Molina, M.R. md Bressmi, R. 1973. Protein-starch ectraction and nutritive value of the jack bem md jack bem protein isolate . In "Nutritional Aspects of Guam Beans and Other LeguIe Seeds as Animal md Human Foods." Jaffe, W. (ed.). Archivos Latinoanericmos de Nutricion, Venezuela. ‘ antogamery, E.M. md Senti, F.R. 1964. Separation of amylase fram amylopectin of starch by an ectraction- sedimentation procedure . J. IkflJML Sci. 28:1. Morris, H.J. 1963. Cooking quality of dry bems. Proceedings of the 6th Armual Dry Bean Conference, January 2-4, Ios Angeles, CA. Morris, H.J. 1964. Change in cooking qualities of raw bems as influenced by moisture content and storage time. Proceedings of the 7th Res. Conf. on Dry Bems, Decerber 2-4, Ithaca and Geneva, NY. Morris, H.J. and Wood, E.R. 1956. Influe1ce of moisture content on cooking quality of dry bems. Food Technol. 10:1. Muneta, P. 1964. The cooking time of dry bems after ectended storage. Food Technol. 18:1240. Naivikul, O. and D'Appolonia, B.L. 1978. Comparison of legure and wheat fleu carbohydrates. 1. Sugar malysis. Cereal Chen. 55:913. 105 Naivikul, O. and D'Appolonia, B.L. 1979. Carbohydrates of legume fleus canpared to wheat flanu. II. Starch. Cereal Chem. 56:24. Nene, S.P., Vakil, U.K. and Sreenivasm, A. 1975. Effect of gamma radiation on physico-chemical characteristics of red gram (Caj anus cazan) stardh. J. Fbod Sci. 40:943. Nie, N.H., H1111, C.H., Jenkins, J.G., Steinbremner, K. and Bent, D.H. 1975. "Statistical Package for the Social Sciences," 2nd ed. IVbGraw-Hill, Inc., New York, NY. Nikuni, Z. 1969. Historical review of studies on starch granules. J. Jpn. Sac. Starch Sci. 22:78. O'Dell, B.L., Bnupo, C.E. and Savage, J.E. 1972. Evaluation of zinc availability in foodstuffs of plmt and mimal origin. J. Nutr. 102:653. O'Dell, B.L. and Savage, J.E. 1960. Effect of phytic acid on zinc availability. Proc. Soc. Exp. Biol. Med. 103:304. Olson, A.C., Becker, R., Miers, J.C., Gutbmann, M.R. and Wagner, J.R. 1975. Problems in the digestibility of dry bems. In "Nutrition and Clinical Nutrition, Vol. I. Protein Nutritional Quality of Foods and Feeds. Part II. Qualiqr Factors-let Breeding, Composi— tion, Processing and Antinutrients." Marcel Dekker, Inc., New York, NY. Painter, N.S. and Burkitt, D.P. 1971. Diverticular disease of the colon: a deficiency disease of western civilization. Brit. Med. J. 2:450. Papmtonakis , M. 1980. Size-etclusion characterization of paper coating starches using a high-pressure liquid chromatography. Tappi. 63:65. Patel, K.M., Bedford, C.L. and Youngs, C.W. 1980. Amino acid and mnineral profile of air-classified navy bem flonu fractions. Cereal Chem. 57:125. Patil, N.B. and Kale, N.R. 1973. Qumtitative separation of starch canpanents on cellulose column. J. Chranatogr. 84: 75. Petersen, D.C. and Harper, J.M. 1978. Granular bed roaster construc- tion. U.S. Patent 4,094,633. Pfeifer, V.F., Stringfellow, A.C. and Griffin, E.L. 1960. Fraction- ating corn, sorghum and soy flonus by fine grinding and air-class- ification. Amer. Miller Process. 88:11. Panermz, Y., Ke, H. and Ward, A.B. 1971. Composition and utilization of milled barley products. I. Gross composition of roller-milled and air separated fractions. Cereal Chem. 48:47. Pritchard, P.J., Dryburgh, B.A. and Wilson, B.J. 1973. Carbohydrates of spring and winter field bems (Vicia faba L.) J. Sci. Food Agric. 24:663. 106 Protein Advisory Group. 1973. PAG Statement 22. A call for action upgrading mutritian through the improvement of food legures. In ' 'Nutrition Improvement of Food LeguIes by Breeding. " Protein Advisory Group, United Nations, New York, NY. Reddy, N.R., Pierson, M.D., Sathe, S.K. and Salunkhe, D.K. 1984. Chemical , nutritional and physiological aspects of (by bem carbo- hydrates. J. Food Clnem. 13:25. Reddy, N.R. and Salunkhe, D.K. 1980. Changes in oligosaccharides ' germination and cooking of black gram and fermentation of black gram/rice blend. Cereal Chen. 57:356. Roberts, A.H. and Yudkin, J. 1960. Dietary phytate as a possible cause of magnesium deficiency. Nature 185:823. Robertson, J.B. and Van Soest, P.J. 1977. Dietary fiber estimation in concentrated feedstuffs. J. Anim. Sci. 45:254. Robin, J.P., Marcier, C., anarbanniere, R. and Guilbot, A. 1974. Linmerized starches. Gel filtration and enzymatic studies of insoluble residue frun prolonged acid treatment of potato starch. Cereal Chen. 51:389. Rocklmd, L.B. and Jones, F.T‘. 1974. Scanning electron microscope studies: effects of cooking on the cellular structtue of cotyledons in rehydrated large limna bems. J. Food Sci. 39:342. Rocklmd, L.B., Zaragosa, E.M. and Oracca-Tetteh, R. 1979. Quick- cooking of winged bems (Psophocarpus tetragonolobus). J. Food Sci. 44:1004. Rosenthal, F.R.T., Espindola, L. and Oliveira, S.M.G. 1970. Jack bem starch. 1. Properties of the granules and of the pastes. Staerke. 22:126. Rosenthal, F.R.T., Espindola, L., Serapiano, M.I.S. and Silva, S.M.O. 1971. Lablab bem starch. 1. Properties of its granules and pastes. Staerke. 23:18. Rosenthal, F.R.T, Nakamua, T. Espindola, A.M.C. and Jochimek, M.R. 1974. Structnue of starch granules. 3. Same consideration on leguninosae and tuberosae. Staerke. 26:50. Rosenthal, F.R.T. and Takeko, N. 1972. Studies on the structnue of Leguminosae starches. Staerke. 24:152. Ibundle, R.E. and French, D. 1943. In "Starch: Chemistry and Tech- nology." Whistler, R.L. and Paschall, E.F. (eds.) Academic Press, New York, NY. Salesrabudhe, M.R., Quinn, J.R., Paton, D., Youngs, C.C. and Sknua, B.J. 1981. Chemical canposition of White bem (PhaseoZus vulgaris L.) and functional characteristics of its air-classified protein and starch fractions. J. Food Sci. 46:1079. 107 Salimath, P.V. and Tharmathan, R.N. 1982. Carbohydrates of field bean (Dolicho Zablab). Cereal Chen. 59:430. Sathe, S.K., Deshpmde, 8.3. and Salunkhe, D.K. 1983. Dry bems of Phasealus: A review. CRC Crit. Rev. Food Sci. Nutr. (J. Food Sci.: in press). Sathe, S.K., Iyer, V. and Salunhke, D.K. 1981. annctimal properties of the Great Northern bem (PhaseoZus vulgaris) proteins. Amino acid campasition, in vitro digestibility, and application to cookies. J. Food Sci. 47:8. Sathe, S.K. and Krishnanuthy, K. 1953. Phytic acid and absorption of iron. Ind. J. Ivhd. Res. 41:453. Sathe, S.K., ngekar, P.D., Deshpmde, 8.8. and Salunkhe, D.K. 1982. Isolation and particle characterization of black gran (PhaseoZus mango L.) starch. J. Food Sci. 47:1524. Sathe, S.K. and Salunkhe, D.K. 1981a. Isolatim, partial character- ization and modification of the Great Northern bem (PhaseoZus vulgaris L.) starch. J. Food Sci. 46:617. Sathe, S.K. and Salunkhe, D.K. 1981b. Functional properties of the Great Northern bem (PhaseoZus vulgaris L.) proteins: sorption buffer, ultraviolet dielectric, and adhesive properties. J. Food Sci. 46:1910. Sathe, S.K. and Salunkhe, D.K. 1981c. Investigation of the Great Northern bem (Phasealus vulgaris L.) starch: solubility, swelling, interaction with free fatty acids , and alkali water retention capacity of blends with wheat flours. J. Food Sci. 46:1914. Saunders, R.M., Potter, A.L., Conner, M., McCready, R.M. and Walker, H.G. 1970. Analysis of starch in wheat mnilling fractions. Cereal Chem. 47:147. Schoch, T.J. 1943. In "Methods in Carbohydrate Clnenistry, Vol. IV." Mnistler, R.L., Smith, R.J. and BeMiller, J.N. (eds.) Academic Press, New York, NY. Schoch, T.J. 1945. The fraction of starch, In "Advmces in Carbo- hydrate Chemistry, Vol. I." Academic Press, New York, NY. Schoch, T.J. 1965. Starch in bakery products. Bakers Dig. 39:48. Schoch, T.J. 1969. Mechmo—chenistry of starch. Wallerstein Lab Camun. 32:149. Schoch, T.J. and Maywald, E.C. 1968. Preparation and properties of various leguIe starches. Cereal Chen. 45:564. Sclnweizer, T.F., Harman, I. and Witrsch, P. 1978. Iowmolecular weight carbohydrates frum leguninous seeds; a new disaccharide: galactapinitol. J. Sci. Food Agric. 29:148. 108 Shetty, R.M., Lineback, D.R. and Seib, P.A. 1974. Determining the degree of starch gelatinization. Cereal Chem. 51:364. Silva, H.C. and Luh, B.S. 1978. Scanning electron microscopy studies on starch granules of red kidney beans and bem sprouts. J. Food Sci. 43:1405. Smith, A.K. and Circle, S.J. 1972. "Soybems: Chemistry and Tech- nology, Vol. 1, Proteins." Avi Publishing Co., Westport, CT. Staudinger, H. 1932. In "Starch: Chemistry and Teclmnology, Vol. II." Whistler, R.L. and Pascall, E.F. (eds.). Academic Press, New York, NY. Standinger, H. and Husemann, E. 1937. Review studies an starch granules. Justus Liebigs Amer. Chem. 527:195. Stringfellow, A.C. and Peplinski, A.J. 1964. Air classification of Kmsas hard red winter wheat flatus. Northwest Miller 270:19. Stringfellow, A.C., Pfeizer, V.F. and Griffin, E.L. 1961. Air class- ification of rice flatus. Rice J. 64:30. Stringfellow, A.C., Wall, J.S., Donaldson, S.L. and Andersen, R.A. 1976. Protein and amnino canposition of ch'y-milled and air-classified fractions of triticale grain. Cereal Chem. 53:51. Suzuki, A., Hizuknui, S. and Takeda, Y. 1981. Physicochenical studies of kuzu starch. Cereal Chem. 58:286. Takeda, C., Takeda, Y. and Hizuknui, S. 1983. Physicochenical properties of lilt starch. Cereal Clnem. 60:212. T‘hivend, P. , Mercier, C. and Guilbot, A. 1965. Determination of starch in canplen media. Armn. Biol. Anim. Biochem. Biophys. 5:513. Thompson, L.U., Rea, R.L. and Jenkins, D.J.A. 1983. Effect of heat processing on hemgglutinin activity in red kidney bems. J. Food Sci. 48:235. Tobin, G. and Carpenter, K.J. 1978. The nutritional value of the dry bem (Phasealus: vulgaris). A literatnue review. Nutr. Abstr. and Rev. 48:920. Trowell, H.C. 1972. Dietary fiber and coronary heart diseses. Enu0p. J. Clin. Biol. Res. 17:345. Tyler, R.T., Youngs, C.C. and Sosulski, F.W. 1981. Air classification of legmes. I. Separation efficiency, yield and canpasition of the starch and protein fractions. Cereal Chem. 58:144. Uebersax, M.A. 1972. Effects of storage and processing parameters on quality attributes of processed navy bems. M.S. thesis. Michigm State University, East Lansing, MI. 109 USDA. 1981. Michigan, United States, and world dry bem statistics. Mich. Dry Bean Digest 8:16. Vongsarnpigoon, S., Uebersax, M.A. and Hosfield, C.L. 1981. Effects of dry storage conditions an the quality of processed edible bems. Cereal Foods World 26:503. Vase, J .R. 1977. Functional characteristics of an intermediate amylase starch from smooth-seeded, field peas campared with corn and wheat starches. Cereal Chem. 54:1141. Vase, J.R., Basterrechea, M.J., Goren, P.A.J., Finelayson, A.J. and Youngs, C.G. 1976. Air classification of field peas and harsebem flatus: chemical studies of starch and protein fractions . Cereal Chen. 53:928. Wagner, J.R., Becker, R., Gunhmamn, M.R. and Olson, A.C. 1976. Hydro- gen production in the rat following ingestion of raffinose, stachyose and oligosaccharide free bem residue. J. Nutr. 106:466. Walker, W.M. and Hymowitz, T. 1972. Simple correlatiarns between certain mineral and orgmic components of cannon bems, peanuts, cow- peas. Camun. Soil Sci. let Anal. 3:505. Watt, B.K. and Merrill, A. 1963. USDA Hmdbk. Na. 8. Agric. Res. Ser. Washington, D.C. Weston, W.J. and Harris, H.J. 1954. Hygrascopic equilibria of dry bems. Food Technol. 8:353. Whelm, W.J. 1970. Enzymatic ecplorations of the structnues of starch and glycogen. Biochem. J. 122:609. Whistler, R.L., Smith, R.J. and BeMiller, J.N. 1964. 'Methads in Carbohydrate Chemistry, Vol. IV." Academic Press, New York, NY. Wilson, L.A., Birmingham, V.A., Moon, D.P. and Snyder, R.E. 1978. Isolation and characterization of starch framn matnue soybems . Cereal Chem. 55:661. Yamada, T. and Taki, M. 1976. Fractionation of maize starch by gel- chrunatography. Staerke. 28:374. Yamaguchi, M., Kaimnma, K. and French, D. 1979. Electron microscopic observations of waxy maize starch. J. Ultrastruet. Res. 69:249. "Tiltiifli‘t'iiiiiiii