IDRC-TS1

Food Legume Processing and Utilization

(with special emphasis on application in developing countries)

Alvin Siegel, Program Officer

and

Brian Fawcett, Research Assistant

Agriculture, Food and Nutrition Sciences Division International Development Research Centre

Abstract

Food legumes, when combined with cereal foods, provide an almost ideal level of dietary proteins for humans and are therefore vitally important to the population of less developed countries. However, the production of food legumes has not kept pace with other food crops, and this trend must be reversed to meet future demands. Village-scale industries adapted from home processing methods provide the greatest opportunity for increas­ing the availability of high protein legume foods. New and improved legume processing technology can also be designed for use in rural areas. Legume processing and utilization also produces by-products, such as husks, starch, and unprocessed plant parts, that have an economic, social, and nutritional value. These by-products can be processed to provide nutrition for both humans or animals. There are many aspects of food legume pro­cessing and utilization that merit further research that should concentrate on developing inexpensive, acceptable, nutritious, and easily prepared food products. This will contribute to in­creased food supplies and an improved nutritional status for the large number of malnourished people in less developed countries.

Resume

Les legumineuses constituent une source presque parfaite de proteines lorsqu'elles sont associees aux cereales dans l'alimentation humaine et consequemrnent, elles prennent une importance vitale dans les pays en developpement. Malheureuse­ment, la production des legumineuses n'a pas enregistre les memes progres que les autres cultures vivrieres et cette situation doit etre corrigee imrnediatement pour repondre a la demande eventuelle. Parmi les mecanismes susceptibles d'augrnenter l'approvisionnement de legumineuses riches en proteines, le plus interessant consiste en l'etablissement de petites industries locales utilisant des techniques traditionnelles de transformation, arneliorees et adaptees. Il serait egalement utile de Creer de nouveaux precedes ameliores de traiternent des legumineuses dans les zones rurales. Peut-etre faudrait-il egalernent envisager l'utilisation des sous-produits, cosses, arnidons, feuillages, d'une valeur economique et nutritionnelle induscutable, et leur transformation en nourriture a la fois pour les humains et pour le betail. Il existe plusieurs formules de traitement et d'utilisation des legumineuses ou une recherche plus poussee s'imposerait, par exemple, la fabrication de produits alimen­taires economiques, nourrissants, de bonne qualite et faciles a preparer. Ces recherches contribueraient a augmenter l'approv­isionnement en vivres et, par la suite, a combattre la malnutri­tion chez un grand nombre de personnes dans les pays en voie de developpement.

The IDRC Technical Studies series consists of papers designed for rapid dissemination among a specialized readership

© 1976 International Development Research Centre

Postal Address: Box 8500 Ottawa, Canada

K1G3H9 Head Office: 60 Queen Street, Ottawa

ISBN: 0-88936-086-3 UDC: 633.1

Microfiche Edition $1

CONTENTS

Introduction

Traditional Processing and Utilization

Legume processing terminology

Home-scale processing

Decortication

Soaking

Pounding and grinding

Roasting, toasting, and parching

Milling

Legume food preparation

Boiling

Roasting and parching

Frying

Puffing

Steaming

Germination

Fermentation

Agglomeration

3

5

14

14

15

16

16

17

18

18

19

19

21

22

22

22

23

25

26

Commercial-scale processing

Decortication

Milling

Legume food preparation

Canning

Processed Legumes by New and Improved Technologies

Milled pulses

Decorticated legumes

Quick-cooking legumes

Legume powders

Legume protein concentrates

Agglomeration

Air classification

Slurry centrifugation

Precipitation

By-Product Utilization of Legume Processing

Summary and Conclusion

Future research needs

Tables

References

4

27

28

28

31

31

33

34

36

39

43

46

46

47

48

49

55

58

59

62

81

INTRODUCTION

Food legumes, cereal grains, and nuts provide a large part of the calories and protein for most of the people of Africa, Asia, Latin America, and the Near East (Tables 1 and 2) • There is every indication that until the end of this century and beyond, the importance of cereals and legumes in the diet, particularly of the poorest people, will increase rather than decline. World cereal production is increasing much more rapidly than the production of legumes; therefore worldwide legume production must be increased (Table 3) • The seriousness of this trend in developing countries is illustrat­ed by the fact that it is in developed countries that the greatest increase in production has occurred (Table 4).

Virtually all cereal proteins are deficient in the essential amino acid lysine. The nutritional value of a cereal grain can be therefore improved by the addition of synthetic lysine or by the addition of a food that is rich in lysine. Table 5 presents a list of protein sources that, in addition to the legumes listed in Tables 6-8, have been tested or suggested as supplements to improve the nutritional quality of cereal proteins.

Nutritional Complementarity of Cereals and Legumes

Food legumes are comparatively rich in lysine and there­fore a combination of cereal protein and legume protein comes very close to providing an ideal source of dietary proteins for human beings. The comparatively low levels of methionine and cystine in legumes is in large part offset by the higher proportions of these amino acids present in most cereals. Tables 9 and 10 show the amino acid content of wheat, chick­pea (Cicer arietinum), and rice together with what might be described as "a perfect protein" as recommended by WHO. From Table 9 it can be seen that the WHO recommended content of lysine is 340 mg/g of nitrogen. A typical wheat protein supplies only 179 whereas chickpea provides 428 mg lysine per gram of nitrogen. An almost perfect mixture of wheat and chick­pea occurs when 67% of wheat is mixed with 33% of chickpea. Column 3 shows the resulting amino acid content and what is called the Amino Acid Score, which is expressed as a percent­age of the WHO recommended level. It can be seen that wheat

5

plus chickpea provides more than 85% of all the amino acids required. Table 10 shows that when rice and chickpea are combined in the ratio of 75 to 25%, with the exception of methionine and cystine, the amino acid balance is almost perfect.

The nutritional complementarity of cereals and legumes is of extreme importance, particularly for the people of the less developed world. From Table 2 it can be seen that whereas North Americans consume nearly 100 g of protein per person per day, IIOst developing nations, on average, consume between 45 and 65 g of protein per person per day. It must be emphasized that these statistics, as with all others quoted, are averages for total populations and that the poorest members of these populations will probably consume considerably less than the national or regional averages. Table 11 suggests that in 1970 there was more than enough protein but a deficiency of calories in terms of percentage requirement for most developing regions. However, once again, one must remember that these are gross averages and take no account of a lack of uniformity in distribution.

In order to achieve optimum nutritional complementarity, cereals and legumes need to be eaten in an approximate ratio of 65 of cereal to 35 of legume. In Southeast Asia, the ratio is closer to 90 cereal to 10 legume. There is convincing evidence from South and Southeast Asia to suggest that over the past 20 years the per capita production of food legumes has declined significantly, chickpeas from 6 to 3, soybeans (Glycine max) from 1.2 to 0.9, lentils (Lens esculenta) from 0.6 to 0.4 kg per person per year.

For the world as a whole between 1952 and 1972 (Table 4), the population increased 40%, total food production 61%, and legume production 49%. For the developing countries collec­tively, population increased 53%, total food production 62%, but food legumes only 40%. In Asia and the Far East, the population increased 51%, total food 65%, but legumes only 21%. World production and trade in legumes has been reviewed recently in Hulse et al. (1975).

Legumes refer to the edible seeds of leguminous plants belonging to the Legurninosae family, one of the three largest families of flowering plants comprising nearly 700 genera and 18,000 species. This family is further divided into three subfamilies, one of which is Papilionoideae, the largest of the subfamilies widely distributed in both tropical and temperate regions. Members of this group are herbs and shrubs contain­ing pealike flowers, which give rise to pods containing one or more seeds. The pod itself is the fruit coat corresponding to the pericarp of the cereal grain. In appearance, pods may be

6

round or flat, thick or thin, woody or fleshy, or straight or coiled. In addition, there are wide variations in seed size, shape, density, and seed coat colour (Purseglove 1968).

Since each seed is attached to the fruit coat (pod) at only one point, it can be easily removed. The seed is chiefly composed of the seed coat and the cotyledon, the latter contributing approximately 85% of the total seed com­position. Legumes are dicotyledons, so named because they separate into two halves. It is the cotyledons that contain stored nutrients that feed the embryonic plant at the time of germination. No endosperm is present in legumes. The coty­ledons themselves are a pair of specialized leaves. The seed coat is either firmly attached to the cotyledons or loosely attached. If firmly attached, it is more often referred to as a skin than a husk or hull, as in the case of a more adhering seed coat.

Within the Papilionoideae family, different uses are made of the many varieties. The legumes used by humans are conuronly called food legumes or grain legumes. The food legumes can be divided into two further groups, the pulses and the oilseeds. Pulses are the dried, edible seeds of cultivated legumes, and have been eaten for as long as historical records are available. The nutritional value of pulses was recognized in biblical times, as evidenced in the book of Daniel (Daniel I, verses 12-16) (Abradale Press 1959) as follows:

"Try, I beseech thee, thy servants for ten days, and let pulse be given us to eat, and water to drink: And look upon our faces, and the faces of the children that eat of the king's meat: and as thou shalt see, deal with thy servants. And when he had heard these words, he tried them for ten days. And after ten days their faces appeared fairer and fatter than all the children that ate of the king's meat. So Malasar took their portions, and the wine that they should drink: and he gave them pulse."

The second group, the oilseeds, consists of those legumes used primarily for their oil content, which may be extracted by pressing or by solvent extraction, the residue being a high­protein "oil cake." Another category of legume crops, the pasture or forage legumes, are those used as fodder, green manure, and cover crops.

Pulses have been cultivated since neolithic times. Kidney beans have been found in caves in Mexico, which have been dated to 4000 B.C. Peas of a similar age have also been discovered in a neolithic village in Switzerland, in predynastic Egyptian tombs, and in the ruins of Troy. Archaeologic evidence suggests that chickpeas were grown in the Eastern Mediterranean

7

and Mesopotamia at least 5000 years ago; the pigeon pea was common to Southeast Asia (FAO 1969); and the cowpea appears to have originated in West Africa (FAO 1969). Lentils were common to the diets of the ancient Greeks, Jews, Egyptians, and Romans. The broad bean was also cultivated for thousands of years in various parts of the world.

The supplementation of food with legumes can be traced as far back as biblical times, for instance, in the book of Ezechiel (Ezechiel IV, verse 9) (Abradale Press 1959):

"And take to thee wheat and barley, and beans, and lentils, and millet, and fitches, and put them in one vessel, and make thee bread thereof •••• "

In Roman times, barley and bean flour were added to lower grades of wheat flour before being processed into bread (Hulse 1974). When the Romans ate barley, it was usually con­sumed in combination with lentils, beans, and other crops. Through the ages, the mixing of legume flours with other flours to improve the nutritional status of bread has been practiced. In mediaeval England, bread was made with beans and peas added to various cereal grains.

In earlier times, grain legumes provided a large portion of the diet of pastoral societies. The early European emigrants to North America added beans, which they found grow­ing on that continent, to their diet. In England, during the same era, kidney beans were considered a luxury, which only the very rich could afford. India is one of the largest con­sumers of pulses, where they are known as "grams." Chickpea (Bengal gram), pigeon pea (red gram), and mung bean (green gram) are eaten largely as dhal in which the seed coats are removed before cooking.

Unfortunately, in most societies, dried pulses have been regarded as "the meat of the poor," a stigma that is still attached to this very valuable protein source. It is the authors' belief that the image of the common legumes can be greatly enhanced with the aid of imaginative yet comparatively inexpensive processing technologies. An approach to the subject of providing TIPre legumes is directly linked to processing methods and systems, whether they be traditional, improved, or new. A review of legume processing systems in both developed and developing countries can illustrate their potential appli­cation in the rural economy of developing countries. The development and adoption of technologies can serve as a base for small-scale industries in developing countries.

8

Production of Legumes

As stated above, world cereal production is increasing much more rapidly than the production of legumes; consequently the need to increase legume production on a worldwide basis and in particular in South and Southeast Asia must be regarded as a matter of serious concern. The importance of pulse crops as a group has declined over the past decade as the share of pulses in total world agricultural production fell from 2.3 to 1.9%.

In the more developed countries, pulse production rose substantially between 1948-52 and 1968-72 as a result of the 87% increase in total production, which was well ahead of the population growth of 22%. Despite a decline in the per capita production of many pulse crops such as dry beans, broad beans, chickpeas, lentils, and other pulses, per capita production of dry peas, vetches, and lupins increased rapidly enough to offset these trends and led to the overall growth in per capita production in the developed countries.

In general, the world production of pulse crops exceeded population growth only slightly from 1952 to 1972, with product­ion increases falling below population growth in the less developed nations and rising much more rapidly than population in the more developed countries. Cowpea and dry bean production advanced most quickly in the less developed countries as a result of both higher yields and larger areas planted with these crops.

A number of factors affect legume production. First of these is "yield," which is the agronomist's term for the weight of edible seed produced per hectare or other unit area of land. Yields of pulse crops generally, and in particular those of tropical countries, have increased very slowly over the past two decades in comparison with the major cereals. With the exception of dry peas, whose yields over the past 20 years rose by 34%, the yield increase for pulses was less than 20%. In contrast, wheat yields increased by 52% and maize by 61%.

The second factor that affects total legume production is the area planted to legumes. On a worldwide basis, from 1948-52 to 1968-72, the cultivated area of a few pulse crops has increased. The acreage under cowpeas increased by 172%, and the total area planted with dry beans and dry peas rose by 50 and 34% respectively during this period. The area planted to other pulse crops remained stable or declined over the same period.

9

Oilseed Production

Although pulse production has risen slowly from 1948-52 to 1968-72, the oilseeds have been produced in increasingly large numbers (Table 12). The area planted with soybeans rose by 123% from 1948-52 to 1968-72 with production jumping almost 200% due to a 30% increase in yields. Although the production of other oilseed crops did not rise nearly as quickly as soybeans, the cultivation of groundnuts expanded by 60% while production rose by 77% as yields increased by 10%. Rapeseed was also a crop experiencing considerable growth as the planted area grew by 80% and production rose 100% assisted by a 25% yield increase. Sunflower acreage expanded by nearly 40%, yields increased by I!K)re than 100%, to give rise to an overall increase in production of close to 150%. Land planted to cottonseed grew by only 20% but production rose by 114% in the wake of a 100% yield increase. Sesame acreage increased by 20%, but total production did not rise significantly over the 20-year period as a result of a 6% decline in yields.

It is the opinion of the authors that much I!K)re attention needs to be given to increasing the production of oilseeds, particularly those such as cottonseed, groundnuts, rape, safflower, and sesame that can be grown in many developing countries.

Apart from soybeans, and to a lesser extent groundnuts, which are important in international trade largely for their oil content, oilseeds have been as badly neglected as pulses in terms of investment in their research and development. The oilseeds are particularly important in that they provide not only good quality protein but also those unsaturated fatty acids that are essential in the diets of all human beings.

Legumes have a number of agronomic advantages, the I!K)St important being their ability to fix their own nitrogen. Certain soil microorganisms in association with the legume root modules are able to reduce at!IK)spheric nitrogen to ammonia, which is absorbed and converted to protein within the legume. Consequently, legumes are much less dependent upon chemical or organic fertilizers than are cereal grains, and legumes tend to improve the soil conditions for other crops. A number of instances have been reported in which the inter­cropping of legumes and cereals led to an increased production of cereal grains and a total increase in productivity per unit of land compared with either crop grown separately.

Legume and Oilseed Prices

As Table 13 indicates, during the past 3 years legume and oilseed prices have been subject to extreme fluctuation. Until

10

early 1972, prices had remained relatively steady for many years. During the latter months of 1973, prices rose as high as six times the earlier prices. In the dry beans group, navy beans, which formerly sold for $225 to $325 per metric ton, were commanding prices of $900 by November 1973. White kidney beans, which previously were valued at $50 to $100 per metric ton, reached prices of $650 to $700 in late 1973, and red kidney beans doubled in price from $250 to $500 at the start of the 1974 trading season. Dry peas increased from the normal $150 to over $500 per metric ton. Similar price increases were recorded for other legumes and the oilseeds traded on the world markets. Although these prices receded substantially in late 1974 and in 1975, in many cases they still remain well above the earlier price levels as has been the case with most protein supplements in today's world markets.

Trade in Legumes

World trade in pulses remains small in comparison with major commodities such as cereals. World trade in pulses in 1972 amounted to only 1.4% of the volume and 3.8% of the monetary value of world trade in cereals. Over the past decade, from 1963 to 1972, the volume of pulse trade has remained relatively stable, increasing only 20% with exports rising from 1.5 million metric tons to 1.8 million metric tons (Table 14). In comparison, the world trade in cereals rose by 44%, while soybean trade rose by 70% in the period from 1967 to 1972 alone.

Despite the fact that they exported only 4% of their production, the less developed countries were responsible for 51% of the world trade in pulses in 1972. The developed countries provided 35% of the total exports, with the centrally planned economies exporting the remainder. Of the more developed countries who purchased 67% of the 1972 imports, Europe was the primary importing area, buying 54% of the total 1972 world imports.

Between 1963 and 1972, world exports of oilseeds increased from 9.5 to 19.4 million metric tons. The more developed nations are the major exporters of oilseeds mainly because soybeans represent 71% of total oilseed exports, the world trade in soybeans being dominated by the United States.

Although there are no projections available for future world trade in pulses, Table 15 indicates the demand forecasts for pulses and nuts to 1980. It is projected that world demand for pulses and nuts will climb by over 50% between 1965 and 1980. In the EEC countries, where pulse production has been declining over the years, future demand is forecast to increase.

11

Demand projections between 1970 and 1980 for edible oils and fats indicate an annual increase in total world demand of 2.7% with consumption increasing from 41 to 53 million metric tons. The less developed countries are expected to record the fastest rate of increase in total demand, growing by 4% per year. If the demand can be satisfied, it is predicted that developing countries will consume one-third of the world's oils and fats by 1980. In 1965, the less developed nations consumed barely 25% of world production. The more developed countries are forecast to increase their consumption of oils and fats at a rate of 1.6% per year.

Population and Demand

The increased world oilseeds demand will have a strong effect on world trade. Between 1970 and 1980, the volllltle of oilseeds and fats and oils exported is expected to rise from 4.6 to 13.9 million metric tons, a 4.1% annual increase. The increased demand for oilseeds and legumes is the result of several factors, not the least being world population growth. By 1980 there will be about 850 million more mouths to feed than there were in 1970.

The sheer increase in the number of humans will exert a powerful force on world markets and require production of the means to feed 23% more people in 1980 than in 1970. Increas­ing incomes among certain sectors of the world's population will probably lead to a demand for more legumes, especially among those people with improved, but still low, incomes who wish to change their consumption patterns by eating more plant proteins. For many others in the world, rising incomes will lead to an increased demand for animal proteins placing added pressure on agricultural producers to satisfy this demand through the production of more meat products, and the plant proteins needed to feed the animals required. Indeed, while per capita world income is projected to increase by 36% between 1970 and 1980, incomes in the more developed nations will probably rise by 50% with Japan's per capita income growing by 131%. As disposable personal income rises, it is customary for the consumption of animal products to rise accordingly.

Protein Supplements for Animal Feeds

The future of the world livestock and animal protein industry is thus an important determinant of the manner in which the production of pulses and protein supplements in general will develop, since animal feeds consist largely of cereals supplemented with legume or other protein sources. World meat consumption, excluding poultry, doubled during the period 1948-52 to 1970. World demand for beef and veal, mutton and lamb, pig meat and poultry meat, and their products

12

is projected to increase by an average of 3.1% per year from 1970 to 1980. Since it will be difficult for animal produc-tion to match demand, the inflationary trends in grain and legume prices are not expected to recede over the forseeable future. In the more developed countries, consumption will rise by 28% between 1970 and 1980, whereas in the less developed countries the estimated growth of meat consumption between 1970 and 1980 is 50%. In the centrally planned economies the demand for meat may rise by 40%. These comparatively rapid rates of growth will naturally have a strong effect on the market for protein supplements. As the demand for meat increases, a proportionately greater increase in the demand for plant proteins occurs, since the conversion efficiency of plant protein into animal protein is relatively low. If world demand for meat and animal products grows by a steady 3% per year during the 1970s, the demand for high protein feed supplements may well rise by 5 or 6% per annum.

During 1971-72, Europe produced only 4.3% of its total supplies of animal feed protein. The demand for high quality protein has accelerated with the trend to the utilization of mixed feeds. From 1960 to 1972, the use of vegetable protein has increased annually by 10.5%, with imports to the EEC countries jumping by 47% between 1966 and 1972. Thus, Europe promises to continue to be a major importer of pulses and oilseeds during the next decade. In Japan, total meat consump­tion is expected to grow by 7.4% annually to 1985 when the demand for meat will be three times what it was in 1970.

Bearing in mind the total world demand for edible protein and the high cost in comparative inefficiency of providing this protein from animal sources, much more attention must be given to increasing the availability and acceptability of edible protein from plant sources, particularly from combinations of cereals and food legumes.

The purpose of this report, therefore, is to present the various processing systems that have been used for food leg:umes. The botanical and common names, and major areas of consumption for the 10 varieties discussed in this report are listed in Table 8. The application of food science and technology principles to legume processing is discussed. Legume utiliza­tion is reviewed by discussing foods based on both conventional and new and improved legume processing technologies.

13

TRADITIONAL PROCESSING AND UTILIZATION

Historically, the processing of food legumes in develop­ing countries has been done in the home by women as part of their meal preparation. Both small- and large-scale industries have evolved to some extent from these traditional cooking methods; however, they have had only a limited impact and outreach. The greatest potential for providing high-protein legume food products to a large number of people living in developing areas is by means of adapting and transferring the established home-processing methods to village-scale industries. In addition, the identification and recognition of various newer and improved legume processing methods that, for the most part, have been known for some time, can provide the basis for further development of legume processing technology in rural areas.

Subsequently, this report will discuss a number of methods and techniques for processing and subsequent utilization of grain legumes. The scope of the discussion will mostly be limited to those methods that can be adapted and employed in small-scale industries that could benefit rural areas. As previously stated, both traditional and nontraditional means of processing will be covered to provide the reader with insight into their adoption in the village industry economy. Further­more, the recognition of simple means for processing legumes can result in their increased acceptability and overall con­sumption.

Legume Processing Terminology

air classification

bean

The fractionation of finely milled flour into protetin and starch con­centrates using a spiral air stream. (Heavier starch granules are separated from the finer protein-rich particles.)

Legume, primarily referring to those of the Phaseolus genus.

14

decortication

dhal

gram

husk

legume

milling

pulse

Home-Scale Processing

Removal of the hull, husk, or outer seed coat of the legume seed.

Dehusked, split grain legume (major form eaten in India).

The entire, whole legume grain or seed usually prepared into dhal.

Seed coat, hull, or covering surrounding the legume seed.

A flowering plant belonging to the Leguminosae family, subfamily Papilionoideae.

Fruit is a pod that contains the seeds.

The edible seed of leguminous plants.

The seed is separable into two parts (dicotyledon) general tenn -bean, pulse.

Process of husking and subsequent splitting of pulse or legume -usually taking place simultaneously in reference to pulse milling and dhal preparation (India).

Decortication and/or preparation of husked legumes into flour.

Dried, edible seed of a cultivated legume; usually refers to one that has been dehusked.

Several methods exist for the preparation of legumes in the home. A variety of procedures are used for the purposes of eliminating toxic substances and antinutrients, removing the seed coat, and softening the cotyledons. Initial proces­sing steps include cooking either whole beans or those that have been dehusked. In addition, raw beans can be processed into a palatable fonn without any cooking. Unhusked, whole cooked beans are sometimes eaten or they can be subjected to a grinding or mashing step.

15

Dehulling may take place with either raw or cooked beans. usually the beans are mashed or pounded after cooking to facilitate removal of the husk or skin. Fully decorticated legumes (pulses) are left whole, split, or further prepared into various food forms such as flours, grits, and pastes. The techniques employed in this operation are mostly manual ones; however, in some instances, simple mechanical equipment is used.

The various traditional processing techniques discussed in this section include boiling, grinding, parching, toasting, roasting, puffing, germination, fermentation, and agglomera­tion. These processing methods apply to both raw or cooked, husked or dehusked, and whole or ground legume varieties commonly eaten in the semi-arid tropical regions of South America, Africa, the Middle East, and Asia. The different preparatory techniques used relate to the structural and textural characteristics of the beans and the cultural food habits of the people.

In discussing various processing methods for preparing grain legumes into foods, initial consideration is given to the topic of decortication, i.e., removal of the outer cover­ing of the grain. For the most part, with the exception of certain bean varieties that possess a relatively thin skin, grain legumes are eaten after the surface layers have been removed. Certain methods of seed coat removal are used on either raw or cooked legumeso Pretreatment steps such as boiling, soaking, or roasting may be employed to facilitate the husk removal that can be accomplished by subsequent pounding, grinding, or milling. Both dry and wet methods for decortica­tion are used (see Fig. 1).

Even though a variety of legumes are eaten in their immature state, the greatest interest from a nutritional point of view is in the consumption of the matured dried grains (Patwardhan 1962). Furthermore, it follows that the proper processing of legumes for human nutrition relates to the efficient removal of the seed coats and husks surrounding the edible cotyledons. This allows for improved digestibility and increased body utilization of the legume nutrients. Some specific legume processing methods, however, do convert un­husked grains into an edible form.

Decortication

Soaking Soaking as a pretreatment to decortication facilitates removal of the husk or skin. Nondecorticated grains that are soaked in water for a short time lend them­selves to easy husk removal. In thLs instance, the husk takes up more water than the rest of the grain, whereby it becomes

16

WHOLE LEGUME SEED

,~d"f '"T'"" ~ood Preparation\...--------

F-i.g. 1. F.tow d-i.a.gJ<.a.m 6M :tJ<.a.d-i.:t-i.ona..t -legume p'<.oc.e¢¢-i.rtg.

more easily separable. In addition, legumes may be steeped, soaked in hot water that is below its boiling point, to allow the seed coat to swell and loosen from the cotyledon.

With regard to the effect of soaking on the nutritive value of navy beans (Phaseolus vulgaris), Kakade and Evans (1966) found that soaking the beans for 4 days destroyed about 28% of the trypsin-inhibitor activity and 75% of haemagglutinin activity.

It appeared, therefore, that these antinutritional factors were destroyed or inactivated or possibly leached out of the beans as a result of the soaking process. A study of toxic factors in Chilean legumes (P. vulgaris, Vicia faba, Pisum sati vum, L. es cul en ta, and C. arietinum) indicated that humi.d heat treatments over l00°c eliminated trypsin-inhibitor and haemagglutinating activities (Gallardo et alo 1974). Partial inactivation was found when dry heat was used. In the case of trypsin inhibitor inactivation using dry heat, more positive results were found when the legumes had been previously soaked for 14 hours.

Pounding and grinding The most common and simplest home-scale method for decorticating legumes and pulses is by pounding or grinding, or a combination of these two methods. This system of hull removal can be classified into two categories, namely dry method and wet method. Both methods are practiced throughout the semi-arid tropical zone, the former having

17

greater acceptance in Central and Northern India, Africa, and Central America (Aykroyd and Doughty 1964). A combined wet and dry method is also used. A less commonly used method of preparing split, husked grams in India consists of half roast­ing the grain followed by splitting. Specific legume varieties and foods prepared from them usually dictate which method, wet or dry, is used. Accordingly, the wet method and a com­bination of the wet and dry method is preferred in Nigeria for the preparation of cowpea (Vigna unguiculata) dishes. A dry method is used to prepare cowpea flour (Dovlo et al. 1975).

In the wet method, the main objective is to soften the husk prior to its removal. A preliminary water soak may be used to hydrate and swell the seed coat. In some instances the grain is merely sprinkled with water. The husk is separated from the cotyledon by the abrasive pounding or grinding action that follows. This is usually done in a mortar with a pestle, which provides a shearing action of the pestle against the grains and the grains against each other. Subsequent sun drying may take place followed by removal of the hulls from the broken grains by winnowing.

The dry method involves simply pounding the grains to loosen the husks followed by removal by winnowing. In addition, the pounded mass then can be submerged in water, and the husks removed when they float to the top of the water's surface.

Roasting, toasting, and parching Subjecting grains to heat for varying periods of time, namely, toasting and roast­ing, is widely practiced as a method of decortication. These practices are basically techniques for drying the grains. Initially, either presoaked moistened legumes or those to which no additional water has been added are used. The addition of heat to legumes by roasting, toasting, or parching renders the husks easier to remove since they become brittle and subsequently crack. When these methods are applied to moistened grains, the cotyledons have a tendency to shrink more than the husk, resulting in the husk being loosened from the cotyledon (Kurien and Parpia 1968). In addition to facilitating husk removal, heating can be effective in destroying toxic factors present in legumes (Kurien et al. 1972). Liener (1962) has reported that most antinutritional or toxic effects of legumes can be partially or wholly eliminated by the proper application of heat. Such factors include trypsin inhibitors, haernagglutinins, goiterogenic agents, cyanogenic glucosides, alkaloids, and saponins.

Milling Milling has been widely used and adapted in the legume-eating areas of Africa and Asia as a means for decorticating grain legumes, primarily those possessing a tough, fibrous husk. It is a process in which the outer

18

layers of the grain, mainly those composing the husk, are successively removed through an abrasive action or shearing force against the whole grain. Simple, hand-operated mills are used for preparation of the whole grains into decorticated, split pulses, flours, or grits.

In India, more than 75% of grain legumes are milled to produce dehusked splits (dhals) for direct consumption. The household practice of preparing dehusked, split pulses consists of loosening the husk and subsequent removal and splitting into two cotyledons. Decortication is accomplished by either a wet or dry method. In Africa, decorticated legumes prepared by pounding and winnowing are made into flours by either hand­grinding or milling on small corranercial plate mills.

The wet method used in India is one in which grains, previously steeped for a few hours, are mixed with red earth and then sun-dried for 2-4 days. After removing the red earth by sieving, subsequent milling in mortars or chakkis dehusks and splits the grain. The dry method is one in which the sun­dried pulses, sometimes after mixing with oil, are sprinkled with small amounts of water. After heating, the pulses are milled (dehusked and split) in mortars or hand-operated wooden or stone chakkis which consist of rotating plates (Kurien and Parpia 1968).

Household milling of legumes is often laborious, time­consuming, and inefficient. An important criterion for evaluating this milling system relates to the yield of split pulses, dhal. Comparative yields of dhal using the home­scale and corranercial methods of pulse milling, indicate up to an 8% decrease in total product between the two methods. Losses can be attributed to breaking and powdering by scouring. Dhal production by one labourer in 8 hours is 30-40 kg using the dry method, whereas it is 60-75 kg using the wet method (Kurien and Parpia 1968).

Legume Food Preparation

Boiling In studying the cooking quality of whole pulses, Muller (1967) found that their hardness was influenced by the interaction of the components of phytin, Ca, Mg, and free pectin. It has been reported that 99.6% of the total phytic acid of beans (P. vulgaris) is present in a soluble form, rather than as insoluble phytin (Lolas and Markakis 1975).

During cooking, softening occurs as a reaction of phytate with insoluble Ca/Mg pectate present in the cell walls to produce soluble Na/K pectate. Studies on the hardness of two pea varieties (purple-flowering, white-flowering) that

19

differed in lignin content of the seed coats, indicated that lignin, like pectin, acts as a binding substance and it is therefore responsible for the ease with which the husk can be removed from the cotyledon of some pulses (Muller 1967). The influence of the seed coat cell walls is a governing factor in the cooking quality of unhusked legumes.

Cooking whole legumes, husked or dehusked, in boiling water is the most common method of legume food preparation. In India, dhals are often cooked until soft, mashed, then mixed with water and reboiled to give a consistency of a soup or gruel (Kurien et al. 1972).

Dry whole grains are similarly boiled until soft, and eaten mashed or unmashed in areas of Africa and the Middle East. Cowpea paste is commonly prepared in West Africa in making a fried food product (Dovlo et al. 1975). Chickpeas are often boiled and eaten whole in parts of Afghanistan, Egypt, and Ethiopia. Whole broadbeans (Vicia faba) are eaten in Lebanon in an oil sauce. A favourite and common dish found in Turkey, Jordan, and Lebanon is hommos, prepared by crushing cooked chickpeas into a puree, to which lemon juice, salt and garlic are added (Hawtin 1974) •

In Central and South America, beans (Phaseolus spp.) that have been soaked overnight are boiled to be eaten whole or mashed, or used in the preparation of other dishes. Crushed beans (P. vulgaris) are commonly eaten fried (Bressani et al. 1973). In addition, mung bean (Phaseolus aureus) paste, prepared from dehulled boiled beans is used to make a variety of dessert dishes in Thailand.

Investigations into cooking legumes by hydration have been directed at factors responsible for cooking time. This is an important consideration, since certain legumes require a considerably longer cooking time to improve their nutritional properties and to render them palatable. Bressani and Elias (1974) state that a minimum of 2 hours is required to cook dry beans (P. vulgaris) at atmospheric pressure. In a report on the utilization of cowpeas in West Africa, it was illustrated that the cooking times for several cowpea varieties differ. The average cooking times ranged from 90 minutes to 150 minutes (Dovlo et al. 1975).

various investigations have suggested that beans (P. vulgaris, P. lunatus (lima beans)) stored under iDadequate conditions of temperature and humidity control, require longer cooking times as related to higher water uptake in the final stored product (Muneta 1964; Burr et al. 1968; Kon 1968). A substantial increase in the cooking time for higher moisture dry beans (P. vulgaris) was reported by Kon (1968). Those containing an initial moisture content of 13.3% required about

20

seven times more cooking time than those containing 8.1% rnoisture. Analyses of the pectin content indicated that heat treatment will convert some of the water-insoluble to water­soluble pectin; however, there were no significant differences in amounts of pectic substances between the high and low moisture beans.

As suggested by Muller (1967), the cell walls of the seed coats, and to a lesser extent those of the cotyledons, play an important role governing the cooking quality of legumes. Since excessive cooking reportedly (Kakade and Evans 1965) leads to a lowering of digestibility, possibly due to the action of the free amino groups with carbohydrates and the inactivation or destruction of certain essential amino acids, it is important that an optimum cooking time be used that gives an acceptably textured bean product possessing the highest nutritional value.

Roasting and parching Roasting as a form of preparing legumes, also referred to as toasting or parching, is practiced mainly in India and Africa. As used in the semi-arid tropics, roasting refers to the method in which usually whole, husked, or unhusked grains are exposed to dry heat. This is accomplished either directly by placing the whole grains and beans at the edge of a fire, or directly upon it, or in hot ashes or sand that are in contact with the fire.

Leguminous seeds are commonly roasted with the husk and eaten as snacks. This practice is used in Africa for chick­peas and cowpeas, and in India for various pulses. Roasted legumes are also mixed with cereal flour after pounding. The traditional Indian household practice for roasting or parch­ing pulses (chickpea, etc.) involves initially sprinkling the grains with a little water, which may or may not contain added common salt. The pulse is then mixed with four times its own volume of preheated sand. The pulse/sand mixture is contained in a frying pan kept on an open fire, the sand reaching a temperature of about 240°c. The pulse is subsequently roasted (or parched) by rapid mixing in the frying pan using a ladle. During this process, the pulse has increased in temperature from an approximate initial 26°c to 132°c in a period of 2-3 minutes. The roasted material is separated from the sand by sieving. The average biological values for parched chickpeas and mung beans were 84.6 and 70.2%, respectively, whereas unparched samples had respective values of 78.2 and 50.0% (Acharya et al. 1942).

Roasting as a legume processing method improves the flavour, texture, and nutritive value of the grain. It also can serve as a preliminary step in facilitating husk removal during wet or dry grinding. The nutritional importance of

21

parching Indian grams, including chickpeas and mung beans, was demonstrated in early studies {Acharya et al. 1942). Animal feeding studies indicated increased biological values of the proteins in the parched samples as compared to un­parched samples. The beneficial effects of heating legumes resulting in an improvement in their protein quality is well known and documented {Bressani et al. 1963; Bressani and Elias 1974). Roasting legwninous seeds has a similar effect.

Frying Frying is mainly used on previously processed legumes, which are in the form of a flour, paste, batter, or dough. Frying takes place in an open can or kettle that con­tains hot oil. The food form that is subsequently fried may be either precooked or raw. In Brazil, bean puree from decor­ticated beans is fried to make a cake {Stanton et al. 1966). Ground legume flours are commonly made into a stiff paste and fried to make a popular Nigerian, ready-to-eat snack {Dovlo et al. 1975). In India, chickpea, black gram (Phaseolus mungo), and peas (Pisum sativum) are often prepared into doughs or paste, which are deep fried into crispy products {Kurien et al. 1972). In addition, mung bean or dhal is also fried in a little fat and eaten as a snack {Kachroo 1970). In the Sudan, filafi is prepared from chickpeas paste, to which salt and seasonings have been added. After standing 2 hours, the paste is cut into small pieces and fried in boiling oil {Hawtin 1974).

Puffing Puffed grain legumes are prepared in the Indian household in a manner similar to that used for roasting. Puffing brings about a light and porous texture in split de­husked dhal. Whole unhusked grains are more commonly used for this process. For puffing, grains are soaked in water and mixed with sand, which has been heated to 25o0 c and then toasted for a short time, approximately 15-25 seconds. After the sand is sieved off, the grains are dehusked between a hot plate and rough roller. The more common legumes prepared in this manner are chickpeas and peas. Other dried legumes may be suitable for this process. These products are traditionally eaten either with parched cereals or as a snack {Kurien et al. 1972). Exploratory studies with chickpeas indicated that moisture conditioning either by soaking or by the addition of water to the grains prior to heating is responsible for good puffing. Kurien et al. (1972) reported that steaming and par­boiling did not promote good puffing. Since the process of puffing legumes has developed in the past decade from a hand­operated home-scale system to a large-scale system in which mechanized puffing machines are used, investigation into standardizing the optimal processing conditions for producing the best puffed products is needed.

Steaming Steaming is primarily used as a secondary

22

process for converting prepared legume flours and pastes into traditional foods. In West Africa, cowpea paste is steamed in the preparation of alele; cowpea flour (grits) is used to prepare cous-cous (Dovlo et al. 1975).

Germination Germination as a means for processing legumes, allows the whole bean to be eaten in a palatable form. In itself, germination is the process by which the hypocotyl or first piece of stem (sprout) is encouraged to develop and grow in length (approx 5-15 cm) (Stanton et al. 1966). In the legumes, the cotyledons contain the stored nutrients that feed the embryonic plant at the time of germination. The prepara­tion of germinated legumes is a method developed through the years, using traditional home practices. In respect to con­sumption, its importance as a processing method is recognized primarily in the Far East and to a lesser extent in Inaia. In these two regions the process is carried out both in the home and on a cottage scale.

The germination process itself as practiced in India involves initially soaking the whole unhusked grains for 24 hours, and then spreading them on a damp cloth for up to 48 hours. Under tropical conditions, sprouts up to a length of 1.0 cm appear. Sprouted grains are eaten raw with salt, or further seasoned and fried or boiled (Aykroyd and Doughty 1964).

In the Far East, sprouted legumes, mainly mung beans and peas are familiar and popular foods in the oriental diet.

Similarly, whole grains are pre-soaked, drained, and allowed to germinate in large wetted baskets for several days. The sprouted legumes can be subsequently washed and cooked in boiling water for use in traditional dishes, or first cooked, then ground into a sauce or paste. Washing and cooking the sprouted legumes enables the removal of any adhering husks, which are usually split during the germination process. (Although the discussion of legume processing has not included the leguminous oil seeds, soybeans, and groundnuts, germination as well as fermentation are important means of legume proces­sing, especially for soybeans in the Orient.)

The direct effect of sprouting on both the physical and chemical changes and the nutritive value of legumes is of great interest. As described in a report by Aykroyd and Doughty (1964), it is proposed that the seed constituents previously in an inert form organize during soaking and become nore assimilable for human nutrition. Increases in nutrients induced by germination take place when sprouts become visible, i.e. after 24-48 hours.

It has been reported (Kurien et al. 1972) that germination

23

does not reduce cooking time or improve texture. However, initial chemical changes occurring in sprouted legumes primarily involve the carbohydrate of the grain, namely the conversion of some starch to lower molecular disaccharides (maltose) and dextrins by the action of amylases. A gradual decrease in the carbohydrate content of pulses during the course of germination was reported by Chattopadhyay et al. (1950). Increases in proteases also occur during germination, causing the degradation of high molecular proteins to lower molecular ones {Hegazi 1974).

The chemical changes that occur during legume germination were supported by Hegazi (1974), who found that total sugars, reducing sugars, and nonreducing sugars had increased in broad beans (Vicia faba) germinated for 4 days, as compared with the raw beans. Starch value had markedly decreased from 37.8% in the raw beans to 30.9% in the germinated ones. Furthermore, it was reported in the same study that protein content of raw beans slightly decreased from 28.1 to 26.7% in the germinated beans, the decrease being a result of protein breakdown by the proteolytic action of enzymes {proteases). Germination led to a noticeable increase in the concentration of amino acids as a result of proteins being broken down into simpler units. Germinated beans had increased amounts of these essential amino acids: lysine (24%), threonine (19%), alanine (29%), and phenylalanine (7%). The sulfur amino acids {cystine and methionine) were slightly increased. The interesting fact in this study is the apparent formation of a food rich in nutrients and readily assimilable for human nutrition.

Other nutrient changes occurring in germinated legumes have been primarily associated with vitamins. Ascorbic acid content increased from a trace amount to 10-12 mg/100 g in legumes germinated for 48 hours {Chattopadhyay and Banerjee 1952b). The development of this antiscurvy factor was demonstrated by the use of germinated broad beans (V. faba) and sprouted chickpeas during times of war and famine. Increases in thiamine, riboflavin, niacin, pyridoxine, biotin, tocopherol, choline, and available iron have also been reported during germination of various pulses, including chickpeas, pigeon peas (Cajanus cajan), lentils, and mung beans {Chattopadhyay and Banerjee 1952a; Banerjee et al. 1954; ICAR 1970).

In a study on germinated navy beans, no significant changes were observed in haemagglutinating activity, although a con­flicting initial decrease then increase in trypsin inhibitor was reported {Kakade and Evans 1966). The fact that germina­tion has a beneficial effect on the growth-promoting properties of leguminous seeds is often cited, although the mechanism for it remains unknown {Liener 1973).

24

Fermentation Fermentation is probably one of the oldest if not the oldest method for processing food grain legumes, The most notable application of this process is in the Orient, where fermented legume foods have been eaten for centuries. Legumes are also prepared by fermentation in South Asia and Africa. In practice, the fermentation process breaks down carbohydrate {starch) to acid as the final end product by the action of microorganisms {bacteria, moulds, and yeast). In the household practice, such microorganisms in the atmosphere are the fermenting organisms, This is also true in village-scale operations, Conversely, controlled fermentation, using specific moulds and bacteria, is followed in large-scale cormnercial operations.

Fermented mixtures of legumes and rice are widely consumed in India. In this process, previously soaked (4-6 hours) mixtures of black gram dhal and rice (1:2) are mashed and left overnight to ferment. Subsequent steaming of the fermented batter produces idli, whereas baking or frying produces dhosai. The preparation of a fermented paste from cooked, decorticated, dried, and crushed legume seeds is practiced in West Africa to make ewa {Nigeria) and soumbara. The paste is subsequently cut into pieces and sun dried, result­ing in balls or sticks that are used as a condiment or relish in various dishes {Aykroyd and Doughty 1964; Kachroo 1970), A fermented food product called tutu is also a popular dish in Brazil.

The preparation of oriental foods from soybeans as de­scribed in this report has application to various legumes. Home-scale preparation of fermented soybean foods starts with boiled, whole soybeans, which are subsequently dehusked and mashed into a paste. The soybean mass is then inoculated with bacteria, yeast, or mould cultures, e.g. Bacillus natto (Bacillus subtilis) to prepare Japanese natto; Saccharomyces ronxii in the preparation of Japanese miso; and Aspergillus oryzae for preparing Indonesian tempeh and Chinese or Japanese shoyu {soy sauce) , the latter product being made from an equal mixture of soybeans and wheat. Various moulds, bacteria, or yeasts may be used to impart different desirable flavours in the final product. Fermentation proceeds for approximately 24 to 48 hours to make tempeh, whereas an initial fermentation of 72 hours is used to make soy sauce. Subsequent roasting or frying of the fermented soybean mass produces tempeh. For soy sauce production, the final fermen­tation process takes place in earthen vessels over a 3-month period {Smith 1963; Nelson and Richardson 1967).

The main effect of fermentation, regardless of the ferment­ing organism used, is to make more of the grain nutrients available for assimilation in the body. In this respect, the

25

digestibility of the legume protein is increased. Digestive enzymes produced by microorganisms during fermentation are able to break down protein into amino acids and other water-soluble products of protein decomposition (Ebine 1972; Takeuchi 1974). By this means, nore protein in the form of amino acids is readily absorbed and utilized. Robinson and Kao (1974) found that reducing sugar, soluble protein, and water-soluble vitamins increased after fermentation in the preparation of chickpea tempeh. Ebine (1972) has reported a biological value of 63% and absorption rate of 83% i~ natto. This indicates an improvement in comparison with those of raw materials. In tempeh production mould breaks down a portion of the original protein into amino acids as indicated by an increase in soluble nitrogen (Nelson and Richardson 1967).

In addition, it has been reported that the fermentation process inactivates unfavourable substances including trypsin inhibitors, haemagglutinins, and saponins that are associated with edible legumes (Liener 1962; Ebine 1972). This relates to the heating process involved in the preparation of fermented foods. The natural formation of antioxidants during the fermen­tation process has practical significance in tropical regions. In addition, fermented products have an increased storage life at room temperature, since organic acids.and the amino acids produced during the process can prevent contamination by pathogenic bacteria and microorganisms. Increases in choline and folic acid have also been observed in fermented preparations.

The development of proteinaceous fermented foods from other legumes is a potential researc~ endeavour. Robinson and Kao (1974) prepared tempeh from chickpeas and horse beans (V. faba). New advances in one such process, as reported by Ebine (1972), describe an improved method for preparing Japanese shoyu. A shorter cooking time at a higher temperature for cooking the defatted soybeans increased the protein digestibility and product yield.

Agglomeration An interesting method of preparing legume flour to a palatable form is by agglomeration. This process is widely used in North Africa for the preparation of cous-cous, a conunon cereal-based food. Traditionally, it is made from wheat, millet, and other available cereals; however, an existing shortage of cereal grains has caused the adoption of legume flours as a substitute for its preparation. Specifically, cowpea flour is used in Senegal, primarily before the harvest­ing of millet, to prepare cous-cous. The use of bean flour in place of cereal flour also is in keeping with conunon practice, that of associating bean consumption with the diet of low-income people.

At the home level, agglomeration of the flours occurs in

26

a two-step process. Initially, the dry method of preparing cowpea flour, that of roughly grinding the beans in a mortar with a pestle, followed by removing the hulls by winnowing, is used. The dehusked grains are ground into flour either by further pounding in a mortar, or more commonly, in a local hammer mill. Subsequent sieving is done to remove the black specks or "eyes."

Agglomerates are prepared from raw flour by slowly adding water with continuous stirring, using a wooden paddle. When large amounts of agglomerates appear, they are shaken in a sieve to remove any nonadhering flour particles. This. process is repeated one or two more times. The agglomerates are then placed in a metal basket that contains small holes. This basket is placed on top of a kettle that contains water that is being boiled on an open fire. When the water boils, the steam penetrates the holes in the basket containing the flour agglom­erates. The entire basket is covered with a cloth to contain the steam. During this process, the agglomerates expand in size due to water absorption. In addition, gelatinization of the starch occurs.

The method of agglomeration broadly comprises moisturizing the particles with surface moisture to form adhesive surfaces. Subsequent agitation of the mass of moistened adhesive particles enables them to randomly come in contact with each other, thus forming agglomerates. As described by Galle (1968), the agglom­eration of flour involves an initial moistening of the flour with subsequent transferral to a mixing zone (agitation, shaking) where it agglomerates. It is reported that it is desirable to increase the total moisture content of the flour to a range of 20-35%, since below 20% results in agglomerates that are too soft and fragile, and over 35% moisture in the agglomerates causes excessive stickiness.

The traditional method of agglomeration as practiced in rural areas of Africa has important application for the utilization of various legume flours. Drying the agglomerates as practiced in a commercial-scale operation for the prepara­tion of cous-cous can be used to prepare stable legume foods. Further discussion on this topic appears in this report, when new processing methods and utilization of legumes are reviewed.

Commercial-Scale Processing

The various methods of legume processing used on a com­mercial scale include milling, frying, puffing, germination, fermentation, and agglomeration. In addition, canning is a common practice used in developed countries for the preparation of beans. In this section, discussion will be limited to the

27

topic of milling, as a primary means of decortication, and canning, since other processing methods performed on the com­mercial level are similar to those of small-scale and house­hold practices.

Decortication

Milling Most large-scale commercial operations for mill­ing legumes exist in India. The commercial methods involve the same basic principles as in household methods. As described by Kurien and Parpia (1968), both a wet method and dry method of processing are used.

Comparatively smaller units practice the wet method of processing. After grains are steeped in water for 4-12 hours in cement tanks, upon draining of the water a paste containing red earth and water is mixed thoroughly with the steeped grain at a 2-3% level. The grains are then kept heaped for about 16 hours. Later, the grains are air-dried for 2-4 days in thin layers in drying yards. The red earth is removed by sieving. When dry enough, the grains are passed through a power-operated stone- or emery-coated vertical chakki (sheller). During this step, about 95% of the grain is dehusked and split. The husk is aspirated off and the dhal is separated by sieving. Residual unhusked whole grains are again passed through the chakki for complete dehusking and splitting. Reportedly, the yield of pigeon pea dhal is about 75%. The wet method has the advantage of facilitating good dehusking and splitting of the pulse; however, it adversely affects the cooking quality of the dhal, There is a good yield in this process due to less breakage, but the dhal takes a longer time to cook. A major disadvantage of this process is its laboriousness and com-plete dependence on climatic conditions. The entire process usually takes 5-7 days and mills can only process limited amounts.

With the dry method, initially, the pulse is cleaned, graded according to size in a grading sieve, and then passed through an emery-coated roller for initial "pitting" or scratching of the husk to facilitate subsequent oil penetration. Pitted grains are then thoroughly mixed with about 1% oil (linseed) in an oiling machine, which is essentially a worm mixer. The oil grains are then spread in thin layers for sun drying in drying yards for 2-5 days. Grains are heaped during the night to preserve heat. At the end of the drying period, grains are sprayed with 2-5% water, thoroughly mixed, and heaped overnight. The grains are subsequently passed through the roller for dehusking by abrasion. In the process, about 40-50% of the grains are dehusked and a major portion of these are split simultaneously. Husks are aspirated off and the mixture of grains and dhal is passed through a dhal-separat-

28

ing sieve to remove the dhal. Residual unhusked and husked whole grains are further dried in the sun for 1 day, mixed

with further amounts of water, and again passed through the roller or chakki (sheller), whereby an additional 25-30% of the grain is dehusked and split. A cone polisher, with smooth rollers, is sometimes used to polish the split dhal.

The major disadvantage of the dry method is the high mill­ing loss due to breakage and powdering. In addition, loosening of the husk in this process is not adequate. Since the various grains milled are not of uniform size, large grains may be crushed by the shelling machines, whereas small grains pass through. Kurien et al. (1972) report that traditional com­mercial methods of milling give an average yield of dhal of 73%, which is considerably less than that of the average maximum theoretical yield, 88%. This can relate directly to the method adopted for loosening of the husk. Since removal of the husk is usually completed after several passes through the rollers, large losses of product occur by scouring of the endosperm in each pass. Splitting of husked whole grain to make dhal results in further losses of the germ, which forms 2-5% of the grains.

In an investigation by Khare et al. (1966), it was reported that the total yield of pigeon pea dhal, on the basis of actual pulse milled, was 76.1%, as against an estimated yield of 84.7%. The dhal ready for marketing had 54% of kernels with chipped off edges and the resulting chippings were found to be lost in various products. This type of damage takes place both in the rollers and shellers. Loss of dhal is also experienced in various forms as flour and fine brokens. The uneven tempering and drying in the sun may also contribute to incipient cracks in the kernel, which finally leads to break­ing and chipping of edges. In addition, the loss of pulse while drying in open yards is considerable, because there usually are large numbers of birds that have access to the grains while drying. Table 16 illustrates the yield of various products during milling of pigeon peas. Flowcharts for the wet and dry milling methods are illustrated in Fig. 2 and 3, respectively.

For marketing purposes, dhal is separated into three grades. The product obtained in the first rolling, 40-50% of the total yield, is considered the best and it is the first grade. Second-grade dhal is obtained from the unsplit pulse after the first rolling operation, having been repassed through the roller after sun-drying. It represents 35-40% of the total yield. The third grade is mostly made up of immature and deformed grains and bigger brokens. It forms 10-15% of the total yield. Third grade is commonly mixed with second grade for sale (Kurien and Parpia 1968) •

29

WHOLE RAW LEGUME ~ STEEPING IN \~ATER (4-12 hr ) -7 DRAHIING ---7

MILL!NGa SUN DRYHIG AND ALTERNATE

MIXING OF RED EARTH PASTE WITH STEEPED GRAM

1 HEAPING OVERNIGHT

l HEAPING AT NIGHT

~---AD_D_I_T_IO_N_A_L_P_A_s_s _T_H_Ro_u_G_H_S_H_E_LL_E_R--=( C=H=A=KK=I"-) __________ _

HUSK SEPARATION SIEVING -----c WHOLE HUSKED AND UilHUSKED GRAMb

DHAL - SPLIT HUSKED GRAlf t HUSK

F.lg. 2. Fiowcha.Jt.t 601L ie.gume. m.lil.ing (we..t me..thod) (Ku1t.le.n a.nd PMpfo 1968) ,

aVe.hubk.lng a.nd &pi.l.t.t.ing.

bAbou.t 95-98% 06 .ln.l:t.la.i a.moun:t.

CV hat V<<ld <• aeou~ 7 5-8 0%.

WHOLE RAW LEGUME -------} SIEVlllG ------7 GRADING____,, VARIOUS GRADES ------7 PITTING

1 SPRAYING AND L- SUN DRYING AND ('t'----HEAPING OVERNIGHT ~ ALTERNATE HEAPING At NIGHT

MIXING E-- GRAM WITH PARTIAL CRACKS IN THE HUSK

T WATER

MILLINGa ----7

1 OIL

ADDITIONAL PASS THROUGH ROLLER OP SHELLER (CHAKKI )b

HUSK SEPARATION

l -----)"" SIEVING

WHOLE HUSKED OR UNHUSKED GRAM

DHAL - SPLIT HUSKED GRAM'

F.lg. 3, Flowc..ha.lt.t 601t puihe. m.li.U.ng (dlt.y me..thod) (Kutr..le.n and Pa.11.p.la. 1968).

aVe.huhfl.lng a.nd hpi.U:.t.lng - a.bou.t 40-50% 06 .ln.l:t.la.l a.moun:t 06 ie.gume. .lh de.h1.u,k.e.d a.6.te.Jt .the. 6-Uth.t m.ll .. U.ng h.te.p.

bAbou.t 25-30% 06 gJta..ln .lh de.hu.hk.e.d a.nd .!.pi.Lt.

cVha.i y.le.id i..h a.bou..t 6 5-7 5%,

30

Improved techniques for pulse milling have been developed at the Central Food Technology Research Institute (CFTRI) in Mysore, India. They will be discussed in the section that deals with new processing irethods.

An outline of the basic features of the wet and dry mill­ing methods is given in Table 17.

Legume Food Preparation

Canning Canning represents the most common method of processing legumes for human consumption, especially in the developed countries. It is done on a commercial basis since the process itself involves considerable time and the need for cooking kettles (pressure cookers) and can-sealing equipment. Commercial legume canning operations are common in North America and, to a lesser extent, in Central and South America, Africa, and Asia. In most cases, canned legume products in developing countries are consumed by the higher-income class, or are exported to developed countries, since they are higher in cost than traditionally processed legumes.

The most popular kinds of legumes used for canning are more commonly referred to as beans and belong to the Phaseolus genus, namely, navy or kidney beans (P. vulgaris) and lima or butter beans (P. lunatus). The former bean, primarily known as the common bean, is used in the preparation of canned North American-style "baked beans." Canned kidney beans are consumed as a vegetable side dish or they may be used as the basic ingredient in a salad. In addition, green or garden peas and blackeye peas (Vigna unguiculata) are canned for similar use. Pre-cooked canned beans (P. vulgaris) are consumed in parts of Latin America (Elias et al. 1973). Purseglove (1968) reported the establishment of a new canning operation for pigeon peas in Trinidad.

The canning operation initially uses dried, whole beans, which are washed in cold water. After draining, the beans are allowed to soak overnight in water, during which time the ~ois­ture content of the dried beans increases from an initial 10-12% to approximately 20%. This facilitates cooking of the whole beans, since there is a softening of the beans. This step is necessary because the beans are processed in the unhusked form. These beans, which have a softer skin than most legumes, become hydrated during the soaking step. After a weighed amount of beans are put in each can, a liquid is added to the can, this being either a thin sauce (tomato) in the case of navy beans, or water, in the case of peas. The cans are then sealed on a sealing machine and subsequently placed in a retort for cook­ing. The heat-processing step proceeds for a time that is dependent upon the temperature and pressure used. Since

31

the retort is essentially a large pressure cooker, a shorter cooking time for processing the beans can be used than is possible during cooking at atmospheric pressure. Usuall6, a 90-minute process is used at a temperature of 135°c (250 F). After retorting, the cans are cooled in the retort under pressure for approximately 15 minutes prior to their removal. This prevents overcooking of the canned beans, even though the steam processing step has been completed. In addition, the cooling step prevents the cans from bursting when they are exposed to atmospheric pressure.

Research in canning beans has concentrated on studying processing time as it relates to the final texture and consis­tency of the products and nutritional losses involved in the canning process. It is widely recognized that considerable losses of some of the water-soluble vitamins can occur during the presoaking (hydration) step, or this loss can occur if the liquid in which the beans have been cooked is not consumed. It has been reported that as much as 50% of the thiamine can be lost during canning (Miller et al. 1973). Significant amounts of riboflavin, niacin, and vitamin A have been found in the liquid medium surrounding the beans during the hydration and cooking steps. The beneficial effects of heat-processing on eliminating some of the antinutritional factors present in beans is overshadowed by the concurrent losses of protein quality in the canned beans. Hackler (1974) reported that the protein efficiency ratio (PER) of canned beans decreases approximately 40% as a result of the canning process.

Elias et al. (1973) 6 have 5eported that a cooking time beyond 30 minutes at 121 c (250 F) under 16 pounds of pressure, without a previous soaking, decreases the nutritive value of the protein. The lower availability of lysine is partly responsible for this effect. Furthermore, soaked samples have shown a reduction in the nutritive value with a cooking time higher than 10 minutes. The determination of the optimum cooking time and temperature for canning beans to eliminate or minimize protein quality loss is a continuing research effort.

subsequent research in this area has led to the develop­ment of quick-cooking bean processes. Studies have included the development of canned bean products with higher protein quality. The preparation of instant legume powders and mixes have gained recent attention. These new developments, which relate to both canned and noncanned bean products, will be discussed in a later section.

32

PROCESSED LEGUMES BY NEW AND IMPROVED TECHNOLOGIES

New systems for processing food legumes come under several main categories of investigation, namely, improved decortication and milling techniques; the development of quick-cooking methods for whole legumes; the manufacture of instant legume powders; and the preparation of legume protein concentrates and isolates. Such new and innovative processing technologies presented in this section can be applied and used in developing nations since they evolve from basic knowledge and investigations into the physical and chemical nature of legumes. For the most part, the work done in this area has either been developed in less developed countries or was initiated with an interest in its application to these countries.

Improved techniques for efficient milling of legumes into dhal (dehusked splits) is an important area of investigation. A systematic method that eliminates wastes and losses has great significance in the Indian dietary pattern. In addition, the mechanical decortication of whole legumes, using a simple modified threshing machine similar to a pearler, can eliminate drudgery in food preparation and provide an increased market for legume products.

Some legumes require considerable time for their prepara­tion into a palatable form for subsequent use in traditional dishes. This fact, coupled with a large requirement for costly fuel, emphasizes the importance of investigating quick-cooking procedures. This deserves attention from both the nutritional and practical viewpoints since it is desirable to use a cooking time that optimizes nutritional benefits of the legumes. ·

Legume powders that can be easily and rapidly prepared for use in familiar foods can increase the consumption of legumes by people of all economic statuses. There is a need to alter the image of beans and pulses as that of a "poor man's food." It was thought that this could be accomplished by increasing their overall acceptance through the introduction of legumes in a form especially desirable to upper income people, namely, as a convenient-to-use tasty powder.

The development of protein concentrates and isolates has received wide attention in recent years. The literature contains a vast amount of information that describes costly

33

complex systems for preparing such concentrates and isolates primarily from soybeans and groundnuts (peanuts) , which are leguminous oilseeds.

Until recently, researchers had overlooked simpler ways of preparing protein concentrates and isolates. Needless to say, relatively little research has been done with other crops, specifically the food legumes. Basic methods previously developed for preparing soy protein isolates have indicated a potential for use with other legume proteins. These methods involve alkali extraction followed by acid precipitation at the iso-electric point. In addition, the age-old method of agglomerating flours by water absorption and a new method for preparing protein concentrates by air classifying finely milled flours can be nodified to a point at which they can both be applied to a small-scale operation in developing nations.

From these main categories of new processing systems, the nature of the method is described in several sections on food legume products made by the various processes.

Milled Pulses

The need for improving the traditional mechanical methods of milling pulses, that of decortication with or without subsequent splitting, is an important area of applied research, as illustrated by the high consumption and utilization of legumes in these forms. Whole decorticated legumes (pulses) and split pulses (dhal) contribute a major portion of protein in the diet of individuals living in the semi-arid tropics. Table 18 shows common pulse varieties of India.

Pulse milling, in addition to being practiced as a cottage scale industry, is one of the major food processing industries in India (10,000 commercial mills; 1-50 tons per day capacity). This fact necessitated an investigation into an improved system. Research efforts have recently focused on the develop­ment of such a system that involves a systematic and standardized procedure for the overall pulse milling process.

As previously discuss~r:, the primary shortcoming of the conventional grain legume milling system lies in the low yield of head dhal due to concurrent high losses as caused by breakage (chipped edges, fines, powder). Subsequently, there are significant losses of food for human use and equally important monetary losses since broken dhal is sold at a lower price than whole first-grade dhal. Relative to these facts, Kurien et al. (1972) have estimated that an improved method that increased the average yield of dhal 10-15% over the conventional process could increase legume availability by

34

1.0-1.2 million tons, equivalent to 300,000 tons of protein.

A view of the difficulties and deficiencies in traditional milling illustrates that both the preconditioning step and milling machinery used in this process are the problem areas. The practice of presoaking the grains followed by drying needed standardization, since various factors in these processes exist. The required water uptake, dependent upon soaking time, and the required amount of water to be removed prior to milling, a factor dependent upon drying time, are traditionally not monitored in a standardized fashion. An arbitrary amount of oil, sand paste, and water is added to the grains. In the case of water, an excess leads to poor cooking and keeping quality (Morris and Wood 1956; Kurien and Parpia 1968; Elias et al. 1973). The dependence on weather conditions for drying seriously limits an efficient milling routine. Uneven drying may contribute to fissures and cracks in the kernel subsequently leading to breakage and chipped edges.

Conventional milling machinery, chakkis (stone- or emery­coated under-runner disc sheller), rollers, and plate mills (emery- or stone-coated) may be used to perfo:rra those milling operations, namely, husking and splitting. Usually the latter two processes are done simultaneously. However, when water is added to dehusked whole pulses before they are split, a separate two-step operation is established (dry method). On the other hand, mild sun drying to loosen the husk followed by roller milling causes splitting to take place at the same time (Kurien and Parpia 1968). The chakki is more suitable for splitting pulses during the wet process.

The suitability of a variable milling machine for dehusk­ing and splitting legumes had direct bearing on an investiga­tion into the deficiencies of traditional milling, which are directly related to the physical and chemical nature of pulses. In early investigations, Kurien and Parpia (1968) found that not only repeated milling of unhusked pulses led to increased breakage and food losses, but that unequal pressure exerted on individual pulses led to incomplete dehusking, increased scouring and abrasion, and chipping of the kernels. In addition, rnoisture had a deleterious effect with regard to ease of dehusk­ing. Dried grains were more easily dehusked, whereas for split­

ting, water-treated grains were advantageous. Although a certain amount of shrinkage of husk and endosperm occurs during drying, these changes are probably influenced by the nature and amount of gums and mucilages in the grains. Drying reduces the tackiness in gums and their ability to hold moisture (Kurien and Parpia 1968) •

The use of nonuniform shelling machines causes large grains to be crushed, whereas small grains pass through unhusked. The

35

lack of a size-grading system aggravated this situation. Con­sequently, the Central Food Technological Research Institute (CFTRI) in Mysore has developed improved methods and machinery for the efficient, uniform, and economic milling of grain legumes. The important pretreatment (conditioning step) is based upon the principle of removing water from the husk to facilitate its removal. :r.bisture is uniformly adjusted to a critical point by exposing the grain to heated air (approx­imately 300°c) for several minutes, thereby loosening the husk and making it friable (brittle). The second step of this process involves the use of pulse-dehusking machines that re­move the husks by abrasive (pearling) action in stone-coated rollers. Abrasion pressure, feed, and clearance can be adjusted to suit different varieties of grain legumes. Reportedly, after a single pass of properly conditioned grain through the dehusk­ing unit, an almost complete removal of the husk (99.5%) can be achieved with the least scouring of the peripheral layers of the kernel. Dehusked whole pulses (approximately 65%) can be subsequently split into dhal in a separate, controlled system. An increased yield of about 10-15% over the traditional com­mercial methods has been reported (Kurien et al. 1972). Average yield of dhal from different legumes using various methods is illustrated in Table 19.

Advantages of the improved CFTRI pulse milling method are apparent. Increased yield can provide significant amounts of needed food protein. The processing time is considerably reduced and the cost of processing is less (Pilot Plant Trial Rs. 5.06 vs. Rs. 7.00 per 100 kg}. Independence of climatic conditions is a positive advantage when considering outdoor sun drying and associated food losses due to nonuniformity of drying, excessive handling, and attacks by birds and insects.

Continuing studies have indicated that the type of gums present between the husk and endosperm may influence the adherence of the husk, and the amount of gum may influence the duration and severity of the premilling treatment (Kurien and Parpia 1968). Further investigations into the water-holding capacity of these gums and its effect on milling characteristics of the grains are warranted. The development of improved systems to provide additional higher quality milled pulses (and dhal) can have a positive impact on the supply and utilization of food legumes in developing areas.

Decorticated Legumes

The development of a simple efficient mechanical process for decorticating whole food legumes is being investigated in an IDRC-supported project on food legume processing being conducted at the National Research Centre Prairie Regional

36

Laboratory in Canada. Subsequent processing of decorticated legumes include hammer milling and sifting to produce accept­able flours for utilization in both traditional and new foods.

The nutritional implications of efficient husk removal from grain legumes is apparent since the high fibre content (cellulose) of the seed coat (approximately 50%) affects protein digestibility and utilization. It is thought that this deleter­ious effect of fibre on legume digestibility is due to either an increase in the movement of cooked legumes through the intestinal tract or resistance to protein hydrolysis by the gastrointestinal enzymes (Bressani et al. 1973).

In addition, the husk may impart a bitter taste and an undesirable colour in flours processed from the legu_mes. Accordingly, polyphenols (often called tannins) are widely distributed in plant materials. Their presence is associated with astringency of various fruit products, e.g. tea. In a recent report, it has been suggested that polyphenols are present, particularly in chickpeas possessing purple, brown, or maroon seed coats (Hulse 1975). This same report further emphasizes that evidence exists suggesting that polyphenols can correctly be called antinutrients. The biochemical mechanism by which they interfere with protein metabolism in humans has yet to be determined.

The need to establish a legume milling system was illustrated in a consumer preference study conducted in Maiduguri, Northeast State, Nigeria. As reported by Steckle and Ewanyk (1974), grains are purchased in small quantities for household processing. The dehulling of food grains is a laborious and time-consuming task, still practiced along traditional lines by women. However, sub­sequent preparation of the grains into flour is done almost exclusively by small neighbourhood mills. A trend toward urbaniza­tion as shown by an increase in the use of packaged flour for pre­paring traditional staple cereal foods indicates a shift to pro­cessed foods. In addition, nontraditional foods derived from prepared flours are gaining popularity as part of the Nigerian food pattern, an example of this being bread, which is becoming a popular breakfast food.

Therefore, the introduction of a milling system suitable for preparing legume flours is a logical step for increasing legume consumption. A shift from household processing to technological processing could have a significant influence on the future development of new legume-based foods similar to cereal-based ones, since a controlled supply of legume flour could be introduced into the market.

An investigation into the dehulling aspect of processing legumes using simple equipment was based upon the efficiency

37

of hull removal. Research concentrated on developing a mechanical system that would be effective in removing as much of the hull constituent as possible with minimum losses of the cotyledon. It was important in the process that the physical or chemical nature of the cotyledon was not adversely altered in such a way as to affect its nutritional or functional properties associated with subsequent food preparation.

An initial comparison of the three dehulling units, namely, an attrition-type mill, a barley pearler, and a grain thresher, demonstrated that the latter two pieces of equipment that operate on a pearling principle were the most efficient for dehulling sorghum and millet. Continuing studies with a George O. Hill grain thresher indicated that this machine could efficiently dehull Nigerian cowpeas.

The Hill threshing unit consists of 13 12"-diameter carborundum discs, which rotate counterclockwise at up to 900 rpm. A centrifugal force thereby throws out the added coarse grains from the centre of the stones. The rotating stones move the grain within the thresher. Dehulling is performed by the abrasive (pearling) action of the kernel against the stone. Outer layers of the husk are successively removed during the operation, the amount removed being dependent upon the throughput and retention time.

~difications made to the Hill thresher included the addition of a fan and cyclone for removing husk fines. Also, both an adjustable hopper inlet and overflow outlet were added to enable a more continuous dehulling process.

Reflectance measurements were used to quantify the efficiency of the dehulling process. This method involved the preparation of flour from cowpeas that were dehulled in the Hill thresher at different retention times. Flour reflectance measurements were determined at 450 nm against a reference standard of magnesium oxide (MgO) on a Hitachi Perkin-Elmer UV-VIS Spectrophotometer with a diffuse reflectance attachment. The flour was packed as hard as possible into glass powder cells (22 x 10 mm). Graphically plotting these measurements against the percentage of kernel removed by the mechanical dehulling process can demonstrate the rate of colour removal (Reichert et al. 1974).

Initial findings indicate that the threshing unit is capable of mechanically dehulling brown Nigerian cowpeas that can be subsequently prepared into cowpea flour. Reichert et al. (1974) reported that at an approximate throughput (650 rpm) of 400 lb per hour and 17% kernel removal, processed flour was comparable to that prepared from traditionally dehulled (manual pounding and winnowing) brown cowpeas. Reflectance measurements for the traditionally dehulled cowpeas and the

38

mechanically (thresher) dehulled ones were 0.101 and 0.090, respectively, indicating the favourable hull removal efficiency of the mechanical process. At a higher stone speed (770 rpm), an increased production capacity of 500-700 lb per hour could be realized. The production of an acceptable product would necessitate a 27% kernel removal. Singh et al. (1968) have reported that cowpea seed coats comprise about 11% of the whole grain, whereas the seed coat of the Nigerian cowpeas comprised 3-5% of the whole grain.

The adoption of a mechanical dehulling unit (Hill thresher) in a rural legume-processing system is quite feasible in develop­ing countries. It can serve as an important element in a total postharvest system with respect to legume utilization by provid­ing products (whole decorticated legumes, split decorticated legumes, flours, middlings, grits) for household consumption. Since it is a dry method, there is less risk of product losses due to microbial and insect spoilage in a tropical climate. Subsequent transformation of decorticated legumes into milled products can be monitored to produce uniform products. In addition, whole or split husks can be further processed at home into local dishes. Efficient hull removal has a further advantage of minimizing costly food losses.

In a supportive IDRC project being conducted at the University of Saskatchewan in Canada, the physical, chemical, and nutritional changes that occur by initial mechanical decor­tication for the preparation of cowpea flour are being studied. Associated factors include water and fat absorption, nitrogen solubility, and protein quality. An awareness of such changes is necessary since they may affect traditional food preparation. A comparative assessment of traditionally and mechanically prepared flours can provide important information for their acceptability and subsequent utilization in new and familiar foods.

Quick-Cooking Legumes

The importance of whole dried beans as an economical and nutritional food source has received renewed attention in re­search being conducted at the Western Regional Laboratory, USDA, California. Recent investigations have included the develop­ment of methods for preparing quick-cooking legume products. The technology of such processes has significant application to legume consumption in developing areas since their prepara­tion requires shorter cooking time. The resulting products, however, possess similar or improved physical, chemical, and nutritional properties as compared to those prepared by standard, long-time cooking processes (Rockland et al. 1974).

39

In reviewing studies on quick-cooking legume processes, it is noteworthy that initial investigations involved simple basic technologies that were the basis for further research. An early process for preparing dehydrated North American-style "baked beans," traditionally a canned food item, were reported by Esselen and Davis (1942). Its development came about during war time when a shortage of tin cans existed. The process itself followed the basic baked bean recipe and process con­sisting of presoaking the dry beans overnight, blanching (3 minutes) the rinsed beans in boiling water, adding a flavoured brine, and baking for 6 hours at 350°F. After baking, the cooled beans were placed on drying trays and dried in a forced circulation, hot air tunnel dehydrator at 140°F for 8-10 hours. For rehydration, two parts of water were added to the dried product, which was allowed to stand for l! hours. The researchers concluded that dehydration was well adapted for the preservation of baked beans.

The preparation of precooked dehydrated beans that can be rehydrated by the simple addition of hot water (190-200°F) without additional cooking was studied much later. In preliminary studies, Feldberg et al. (1956) found that 105 minutes were necessary to obtain completely cooked beans at atmospheric boiling, whereas 20 minutes cooking using steam pressure (245°F, 13 psi) produced beans exhibiting satisfactory texture character­istics. Beans had been presoaked for 8 hours. Prefreezing, though not practical in developing countries, gave acceptable bean products that had minimized bursting during low humidity dehydration (170-190°F) in an airflow dehydrator. High-humidity drying without the prefreezing lessened the bursting effect, resulting in products with a denser nature. A grainy product resulted after rehydration with near boiling water for 10 minutes.

High-humidity drying of beans prepared by the process de­scribed in the previous study was further investigated in an attempt to eliminate the costly prefreezing step (Dorsey et al. 1961). The main problem area in dehydration (hot air) was splitting or "butterflying" caused by a differential rate of drying between the skin and cotyledon. Although a high-humidity dehydration process was found to be unacceptable in sufficiently reducing "butterflying," the study did reveal that navy beans should be blanched (15 minutes) prior to soaking. The effect of blanching was to ensure complete rehydration in addition to producing a product with better storage stability than unblanched beans. The blanching operation revealed the possibility of lipoxidase activity, which caused rapid deterioration of the product during storage.

Subsequently, a new process for the preparation of quick­cooking dehydrated beans (Phaseolus spp.) (New York State pea,

40

Perry marrow, red kidney, black) involved hydration of the dry beans by soaking in water (15 minutes), precooking in steam, coating, and dehydrating (Steinkraus et al. 1964). It was found that a prior steam treatment (15 minutes) did not favour­ably affect hydration at room temperature. Hydration for 1 hour in water at 210°F was complete and equalled that obtained from hydrating in water at room temperature for 5 hours. Because of the cooking that occurred in the former hydration step, the beans continued to hold more water as heating in water continued. Hydrated beans were precooked in steam under pressure in a retort. A standard precook of 90 minutes at 250°F was used for beans soaked at room temperature; precooking time was shortened for beans hydrated at a higher temperature. After cooking, beans were dipped for 5 minutes in 20% sucrose solution at 160°F, drained, and dehydrated in a dryer (200°F, 300 fpm airflow) for 2 hours, to a final moisture of 10%. The hydration for 30 minutes in boiling water gave a product with a soft texture. A flexibility in processing conditions ranging from 30 to 90 minutes of hydration at 180°F followed by precooking at 250°F for 30-90 minutes was possible. This flexibility of hydration and precook times and temperatures was due largely to the coat­ing step, which prevented "butterflying" during subsequent pro­cessing.

A study by Subba Rao et al. (1964) emphasized the effect of pressure-cooking on considerably shortening the cooking time (approximately 70%) of pigeon pea or dhal. The amount of solids dispersed was an indicator for evaluating the degree of cooking. It was also reported that the addition of a combina­tion of salts, namely trisodium phosphate and ammonium carbonate improved the cooking quality of the dhal.

More recent studies in the area of quick-cooking dried beans included the development of an intermittent vacuum treatment (Hydravac process) for 30-60 minutes in a solution of inorganic salts (sodium chloride, tripolyphosphate, bicarbonate, and carbonate) (Rockland and Metzler 1967). This process has been adapted to the preparation of various quick-cooking legumes, including lima beans {Phaseolus lunatus), pinto beans (Phaseolus vulgaris), and peas. The process itself consists of loosening the seed coats by the vacuum treatment, or by hot­water or steam blanching. The beans are subsequently soaked in the solution of inorganic salts. Infusion of the salt solution through the hilum and fissures in the hydrophobic outer layers of the seed coat is facilitated by the Hydravac process. The inner membrane is wetted by the solution and it subsequently hydrates rapidly. The seed coat expands to its maximum dimension within a few minutes. The cotyledons become encapsu­lated in a uniform bath of hydration medium, thereby imbibing the solution rapidly. A flow diagram for this process appears in Fig. 4.

41

WHOLE DRY BEANS

Blanch Hydravac l boiling water or steam --- or----

(1 to 3 mio""' I l in salt solution at 7ooc (below 60 minutes)

Soak in hydration (salt) medium

(6 to 24 hrs )

L Drain

l Rinse

Dry~~ Cook

Fig. 4. Flowcha4t 604 the qulck-cookl~g bea~ p~ocecc. Sou~ce: Rockla~d 1974.

In determining the type of hydration media to use, Rockland and Metzler (1967) assumed the existence of a relationship between chemical composition and cooking characteristics in dry lima beans. Differences in the cooking characteristics of the major bean tissues, seed coat and cotyledon, could be related to the high amount of fibre and low amount of protein in the seed coat. Conversely, cotyledons contain an inverse proportion of these same components. A further assumption made was that legume proteins influence texture and cooking characteristics of cotyledons. To this effect, the hydration medium was designed to disperse or solubilize proteinaceous material. Metal chelat­ing agents (phosphates) helped to disassociate metal (calcium) salt-protein complexes. The cooking time of bean seed coats was reduced by sodium chloride and the phosphates. The carbonates controlled the pH of the hydration medium, which had an initial pH of 9.0. The salt solution that worked best was composed of 2.5% sodium chloride, 0.25% sodium carbonate, 0.75% sodium bicar-

42

bonate, and 1.0% sodiwn tripolyphosphate. Drying the processed beans under a low-velocity air stream at below 140°F for 24 hours resulted in a product (8.5-10.5% moisture for lima beans) with minimwn splits and "butterfly" cracks. The overall advantage of this quick-cooking process is its conversion of dry beans to rehydrated products, which cook within 15 minutes, effecting a more than 80% reduction in a cooking time of 1-3 hours. Furthermore, the cooked product has good appearance and flavour, and its nutritional value is identical to that of untreated beans (Rockland 1974).

The development of an inexpensive mechanical method for producing quick-cooking beans (California small white, sanilac, pinto) has been reported by Kon et al. (1973). A Ce-Co-Co barley and wheat polishing machine (basically a tapered carbo­rundwn rotor turning inside a slotted screen) was used to peel whole raw beans. Peeled whole beans accounted for 73% of the yield; peeled split beans accounted for 16%. Comparative results of cooking peeled, soaked (overnight) , and unsoaked beans indicated a reduction in cooking time of at least 26 and 42%, respectively. After cooking, almost all the beans were split, which contributed to the quick-cooking characteristics of the beans. Peeling did not affect the nutritional quality of the beans as indicated by both similar protein efficiency ratio (PER) values and water-soluble vitamin contents (thiamine, niacin, pyridoxine, and folic acid).

Legwne Powders

As part of the dry bean research being conducted at the u.s" Western Regional Laboratory, the development of precooked legwne flours was initiated to provide dry beans (Phaseolus spp.) in a convenient stable form for subsequent use in con­venience foods. The high protein content and relatively low cost of legwninous seeds has further emphasized in more recent times the importance of continuing an investigation into the preparation of legwne powders.

In relation to its importance to developing countries, basic methods for preparing legwne powders could initially stimulate increased conswnption of legwnes among all people, particularly the more wealthy. This is important since the stigma of food legwnes, that of being considered for only the poor, may be overcome. Although a relatively small amount of research has been done in this area, initial findings could encourage further interest in this form of legwne processing technology.

The initial development of instant bean powders followed three basic processing systems. As reviewed by Copley (1974), the first process (~ simply involved three steps, namely,

43

soaking the whole beans, cooking them to a slurry, and drying the slurry using a drum dryer. In a second process (b), whole beans were ground into a fine powder in a mill. The powder was inunediately blended with a solution of hydrochloric acid to inactivate an enzyme responsible for causing bitter flavour. The slurry was cooked for 5 minutes, neutralized by the addition of sodium hydroxide, then further cooked to a palatable form. After drtun drying, the resulting product had a bland flavour, making it suitable for use in baked goods and other food mixtures. A third method (c) of legume powder preparation has more recently evolved from a quick-cooking bean process, in which an intermediate product (blanched, soaked, whole beans) is dried, cooked, and milled to a powder. Bean powders produced by these three methods all possess the good characteristics of rapid hydration. A schematic diagram as given in Fig. 5 best illustrates these three procedures.

{b) (c)

i t grind Blanch

i j, slurry Powder~acidify Soak

(HCl) (salt solution)

l t Dry cook ~neutralize Drain

(NaOH)

t cooked mash Rinse

l . Intermed1 ate

i Dry

Prodrt

I LEGUME POt•DER I ~--J F~g. 5. Floweha~t 60~ the p~epa~at~on 06 ~n4tant legume

powde~4 u4~ng th~ee p~oee~4e~.

44

These processes have been used to prepare legume powders from pea beans (P. vulgaris), peas, and lentils. The flavour of legumes reportedly (Kon et al. 1974) depends largely on cooking procedure, as related to the retention of the integrity of the cells. In other words, cooked powders prepared so that cellular integrity is preserved retain their original flavour. Conversely, grinding treatments, which rupture nost of the cells to release cell contents of the raw legumes, prevent subsequent development of the cooked flavour associated with beans.

The development of powders having a low flavour intensity (bland) is significant for their utilization in various food products. These include traditional noodle products and baked goods, as well as new items, such as snack foods.

Kon et al. (1974) conducted an investigation into the preparation of powders from California small white and pinto beans using two different procedures that would affect the cellular nature and physiochemical properties of the final product. For the preparation of regular or "whole cell" powders, whole beans, soaked overnight, were cooked for 1 hour and unsoaked beans were cooked for 2 hours. The cooked legumes were slurried in a disintegrator and drum dried. For the preparation of acidified or "broken cell" powders, raw legumes were first finely ground in a mill. A slurry was made by the addition of acidified (HCL) water to produce a slurry of about pH 3.5. Cooking proceeded for 15 minutes with vigorous stirring after which time the slurry was neutralized by adding sodium hydroxide, then drum dried. The flakes were further ground into a powder.

The acidified powders formed stronger flakes on drying and were bulkier. This preparation contained more extra cellular pregelatinized starch. Also, the powder was completely gelatin­ized and nost of the starch was free. Grinding the flakes in­creased the density of both preparations. The acidified slurry itself possessed high viscosity even at low concentration. Such physiochemical properties have direct relation to ultimate food uses.

In this same study, an investigation of the nutritional properties of the two powders was conducted. For this test both products had equal cooking times. The nutritional qualities of both bean powders were similar with respect to PER and digestibilities. These values were significantly lower than for casein; however, the addition of methionine (0.6%) raised them to that of casein. The PER for all bean products ranged from 1.39 to 1.47. The digestibility value was 93% for all products.

As commercial products, both powders are free flowing, non­hygroscopic, and convenient to use since they easily reconstitute

45

in hot water. The acid-treated slurry requires a shorter cook-ing time (15 vs. 45 minutes), giving similar nutritional qualities. In comparison with the regular powder, the acidified powder rehydrates IlK>re slowly due to the free gelatinized starch. However, initial studies indicate it has a much longer storage life (approximately 12 IlK>nths as compared to 4 for the regular powder).

Continuing studies indicate the potential for making instant powders from an additional variety of legumes including faba beans (Vicia faba) and lima beans (Kon et al. 1974).

Additional information on the nutritional attributes of instant bean powders (pinto - P. vulgaris) was reported by Miller et al. (1973). The vitamin content of the regular and acidified bean powders was compared. The thiamine content of the acid-processed powder was about 10% lower than that of regular powder; whereas the niacin content was about 20% higher. Pyridoxine contents were about the same.

Legume Protein Concentrates

Agglomeration

The preparation of agglomerates from cereal flour has been previously introduced as a traditional basic process for making cous-cous, a familar North African cereal food. This product in its cooked form is distinguishabie as small individual grain particles, similar in appearance to rice, that have been pre­pared by steaming the initially dried flour agglomerates. The adoption of this method for preparing similar agglomerates from legume flour has recently gained attention in research endeavours aimed at increasing the nutritive value of cous-cous. A com­prehensive study (Gainor et al. 1971) was conducted at Kansas State University on the protein supplementation of cous-cous with horse bean (Vicia faba) and chickpea flours as they influenced processing and nutritional and organoleptic pro­perties. Durum wheat flours (70, 85, and 95% extraction) were the cereal base for the legume protein supplements.

In this study, the agglomeration method illustrated the basic principles of this process, namely the dispersion of powdery flour in a humid atmosphere to wet the surface and form adhesive gluten on the flour particles. Raw material was agglomerated with water in a mixer, sieved through to separate the agglomerated particles, then subsequently dried in a forced air oven at 50°c for 20 hours with or without a 7 minute pre­steaming step. It was found that the amount of water added to the flour is critical since an excess amount caused the agglomerated particles to become too large, causing them to slide together and impairing their separation by sieving. Con­versely, when less than the optimum (as determined by the development of the agglomerated particles) amount of water was

46

used, the flour would agglomerate slowly and be unable to attain the desired size. Furthermore, the rate of water addition was important, specifically if it was too rapid, since under this condition the water would not be evenly distributed throughout the flour particles. This would result in dry unagglomerated flour particles, and large, noist masses of inseparable agglomerated particles. If water was added at a slow rate, the only undesirable end result would be a need for a longer prepara­tion time to make the cous-cous. For consumption the dried cous-cous was placed in a cheesecloth and put on top of a steamer for 20 minutes. The cous-cous would then be turned and aerated and then allowed to stand for several minutes. This procedure was repeated twice more.

Galle (1968) described a large-scale process for prepar­ing a high protein fine flour (HPFF) fraction that is normally derived by the air separation or fractionation of cereal flours during milling. Since this HPFF fraction is difficult to handle due to poor flow characteristics, he suggested the prepara­tion of a concentrate from this fraction by agglomeration in a mechanical agglomerator. The agglomerated particles would then be added to milled flours in a controlled manner. In test studies, an HPFF from milling wheat contained 18-23% protein. Similarly, agglomerated high-protein legume concentrates could be prepared from fine flour fractions of milled legume flour. These concentrates could be prepared for addition to traditional foods or as a new food.

Air Classification

An investigation into the manufacturing process for prepar­ing a protein concentrate from field peas (Pisum sativum) has recently been conducted at the Prairie Regional Laboratory in Canada. Pea protein concentrate is produced from pea flour by air classification. This process results in a product that con­tains about 60% protein. The initial pea flour material is obtained by mechanically grinding whole or dehulled peas. The flour contains approximately 23% protein (Prairie Regional Laboratory 1974)"

The protein concentrate obtained from the process is of high nutritive value, both qualitatively and quantitatively. Air classification to produce a protein concentrate initially involves fine grinding of the legume flour. In the reported process, pin-milled flour is classified in a spiral air stream. Concentration of the protein by this method is possible because of the difference in size, shape, and density of the starch (carbohydrate) granules and the protein-containing particles. In this finely ground flour the starch fraction is present in the heavier particles, whereas the protein fraction is contained in the fine particles. Protein fractionation takes place when

47

opposing centrifugal and centripetal forces are employed. The centrifugal force is applied to the fine flour particles opposed by a centripetal drag. Consequently, the heavier particles (starch) move in the opposite direction to the fine protein-rich particles. Since legume flours contain relatively large starch granules, this process of protein concentration by air classification is practical. In pilot studies, pea flour containing 21% protein yielded 25% fines (pea protein concen­trate) with a protein content of 60% and a coarse fraction (starch) containing about 8%0 Repeated milling gave an additional 10% pea protein concentrate containing a somewhat lower protein content (46%). A schematic flow sheet for air classification is given in Fig. 6.

Slurry Centrifugation

Research has also been conducted at the Prairie Regional Laboratory on the preparation of pea protein concentrate from pea flour by slurry centrifugation. This process involves mix­ing the flour initially with five parts of water. Subsequently, lime is added to the slurry, increasing the pH to around 9. Centrifugation yields a high protein supernatant and starch solids. Spray or drum drying of the supernatant yields a pea protein concentrate containing about 60% protein. The concentrate can

WHOLE RAW PEAS (21% Protein) I , oo , bs

~ ~ PEA FLOUR (21% protein)

100 1 bs

PEA STARCH

(8% protein)

75 1 bs

1 PIN MILL

i PEA STARCH

75 1 bs

PEA PROTEIN CONCENTRATE

(60% protein)

25 1 bs

PEA STARCH (25% protein)

65 1 bs

PEA PROTEIN CONCENTRATE

) ( 56% protein)

35 lbs

1 ) PEA PROTEIN CONCENTRATE

(46% protein)

10 lbs

F-lg. 6. Flowcha4t 604 the p4epa4at-lon 06 pea p4ote-ln concent4ate by a-l4 cla~~-l6-lcat-lon.

48

later be ground to a desired particle size. Starch solids are prepared from the starch fraction (containing about 6% protein) , which is reslurried with 5 parts of water and then centrifuged to give starch solids (2% protein), which are dried in a forced air oven at 60°c. The slurry-centrifugation process for pre­paring pea protein concentrate is illustrated in Fig. 7.

Both the air-classification and slurry-centrifugation processes have application to other food legumes. The use of simplified equipment to prepare such concentrates according to basic established principles can provide increased utilization of food legumes. Although the latter process (slurry centri­fugation) is a wet method that can cause problems in tropical countries, continued investigation into this process and the dry method (air classification) can increase knowledge in the area of legume processing.

A variety of uses for legume protein concentrates in both traditional and new foods exists. Blends of cereals with legume protein concentrates can be formulated to prepare infant foods, pasta products, and baked goods, including bread.

Precipitation

The preparation of protein concentrates from soybeans by

WHOLE RAW PEAS (21% protein) 1 100 lbs

I PIN MILL I ~EA FLOUR (21% protein)

11 lbs

l lb lime added~ (SLURRY TAMKI ~--_............_...._ ........... .....,.._ ____ _,

J, PEA STARCH (6% protein)

72.5 lbs

PEA PROTEIN CONCENTRATE 500 lbs H20 SLURRY TANK

.J, PEA

J:.B1iill.tJ.. CONCENTRATE (60~ protein)

33 lbs

I DRUM DRYER I -!,

PEA ...£&lllllL

CONCENTRlli (60% protein)

33 lbs (sheets or flakes)

added

PEA STARCH (l .8% protein)

167 lbs

roR'i'ER"1 ~ PEA SJAB GH ~ ' (l .8% protein)

67 lbs

F1g. 7. Flowehant 6on the pnepanat1on 06 pea pnote1n eoneentnate by &lunny eentn16ugat1on.

49

protein precipitation has been practiced for centuries. Hot­water extraction of protein from ground, cooked soybeans (soy milk), followed by precipitation of the protein, using calcium, is a long-practiced method for preparing oriental soybean curd (tofu). In other words, soy protein concentrate is the modern version of soy curd, this product containing about 60% protein in the dried form.

In more recent times, research has involved the preparation of soy protein concentrates by aqueous extraction of the protein at its isoelectric point, thereby precipitating the protein (Altschul 1969). This well-established technology for prepar­ing protein concentrates from soybeans can be utilized for the preparation of such concentrates from various food legumes. r-Ddification of the process to establish it as a base for a small-scale industry in developing countries has received attention.

Initial reference to the preparation of soy protein isolates by precipitation serves to illustrate the principles of the process and associated nutritional attributes for considering its adapta­tion to grain legumes. By definition, soy protein concentrates contain 60-70% protein, whereas soy protein isolates contain not less than 90% protein. However, the basic technology employed to produce either product, that of extracting soluble proteins in an aqueous solution followed by precipitation, are similar. The purity of each product with respect to content of protein and crude fibre dictates the terminology followed. Accordingly, the name "legume protein concentrates" is given to the high-protein products prepared by this protein precipitation method of legume processing.

Cleaned dehulled soybeans are used to prepare protein isolates. After an initial extraction of the oil, the defatted soybean flakes are extracted with an aqueous medium. After extraction, the extract containing the water-soluble protein is separated from the insoluble residue by appropriate mechanical devices. The major globulins are precipitated by the addition of a food-grade acid, usually hydrochloric acid. The pH of the clarified extract is subsequently lowered to 4.5 (isoelectric point) where the solubility of the globulins is near a minimum. The major protein fraction separates out as a finely divided white curd. Soluble constituents are removed. The protein is relatively pure since these constituents, namely, oligosac­charides, peptides, and salts, are reIOC>ved. The concentrated protein is more commonly neutralized before drying, this pro­cedure yielding a proteinate form that has the advantage of being water dispersible (Meyer 1970).

Another procedure for preparing soy protein concentrate uses rooist heat to denature and insolubilize the proteins in

50

Acidification

L

Whole Legume

1-- dehu 11 i ng

Dehulled Legume

1 Legume Flour

EXTRACTJ0;1

l Protein Liquor

1 or

Precipitated Protein

(curd)

1 Washed Protein

concentrate

1

grinding

Heat Denaturatlon

~ vvashi ng

spent whey

drying

I Legume Protein Concentrate

F-<-g. 8. Flowc.ha.11..t 601t. p1t.e.pa1t.-<-ng legume. p1t.o.te.-<-n c.onc.e.n.t1t.a.te..6 by p!t.o.te.-<-n p!t.e.c.-<-p-<-.ta..t-<-on.

soy flour followed by a water wash to remove the sugars and other minor components. Wolf (1970) emphasized that the physical properties of soy protein concentrates will differ with the method of preparation. As an example, concentrates prepared by acid leaching with neutralization in the absence of heat treatment will have a higher content of soluble protein than concentrates obtained by heat treatment. This is in accordance with the recognized fact that nitrogen dispersibility and related protein solubility can be decreased by heat treatment. A schematic diagram for the preparation of protein concentrates from leguminous seeds appears in Fig. 8.

Several investigations have been conducted on the isolation of proteins from leguminous seeds. Pant and Tulsiani (1969) studied the isolation of protein from Phaseolus (var. mungo, vulgaris) to standardize simple and effective methods of protein isolation from legume seeds. A variety of extracting solvents, including sodium chloride, solubilized 74-82% of the

51

total nitrogen, which consisted mainly of globulins (approximately 60% of the total nitrogen) and albumins. The minimum percentage of total nitrogen extracted in the Phaseolus seeds was at a pH between 2.1 and 3.4 with only 15-22% of the total nitrogen (mainly albumin) being extracted. In the case of P. mungo only 35% of the total nitrogen went into solution on the alkaline side, whereas in the case of P. vulgaris seeds, about 80% of the total nitrogen went into solution.

As suggested by the authors, another method for isolating proteins from dry beans would be to extract them with an alkali solution at the pH solubilizing the maximum percentage of nitrogenous components, followed by changing the pH to that of minimum extraction (acidification) to isolate the protein. It should be emphasized, however, that treating protein fractions with strong acidic and alkaline solutions decreases their nutritive value. Accordingly, the best and simplest method for isolating proteins from dry seeds is to extract them with sodium chloride solution and precipitate them by dialysis of the extract. An alternative to the latter step could be heat denaturation of the protein.

A study of the extraction of protein from red kidney beans Phaseolus vulgaris found that nitrogen extraction could be increased with an increase in the concentration of sodium chloride in the extracting medium (Hang et al. 1970). The dispersibility and isolation of proteins from legume flours was investigated using faba bean {V. fabaJ, pea bean (P. vulgaris}, mung bean (V. radiata), field pea (P. sativum), and chick-pea (C. arietinum). From this study, Fan and Sosulski (1974) reported that protein from mung bean and field pea flours exhibited high extractability at pH 2 and above pH 7 where 90% of the nitrogen was extracted. Low nitrogen solubility did occur at the apparent isoelectric point. In comparison, the solubility of flour nitrogen was progressively lower in faba bean and chickpea, which exhibited an apparent isoelectric range of pH 4-6. In general, an alkaline extraction solvent (0.2% NaOH) was very efficient in solubilizing flour nitrogen especially in mung bean, field pea, pea bean, and lima bean. Nitrogen dispersibilities at pH 10 were essentially complete for all legumes with the exception of chickpea flour. Yields of protein concentrate (isolate) were generally proportional to the protein content of the legume flours varying from 19% in lentil to 29% in pea bean based on legume flour. Only the faba bean concentrate contained over 90% protein. However, mung bean and field pea isolates could be further purified by additional washings. Accordingly, it was felt that faba bean and mung bean were the most promising new sources of protein isolates.

In a more applied approach to the preparation of legume

52

Mung bean

Whole beans

wide adjusted stone mill

mung beans

soaking in water (8-10 hr)

mun beans

flotation

draining

mung beans

narrow adjusted stone mill

Mung bean paste (fine)

water added (paste:water ~ l :3)

Bird centrifuge

solution Starch and residue

boiling (20 min)

protein solution

addition of glacial acetic acid (adjust solution to pH 4-5)

bean rotein

filtration using fine cloth

protein concentrate

washing with water (l-2 times)

I Fresh mung bean protein concentrate I

F1g. 9. Floweha~t 60~ the p~oduetion 06 mung bean p~ote1n eoneent~ate.

Sou~ee: Bhumi~atana and Wondaouta 1972.

protein concentrates, a project was undertaken at the Institute of Food Research and Product Development in Bangkok, Thailand, to develop a mung bean (P. aureus) protein concentrate using the basic technology of protein isolation. As illustrated in the flow chart (Fig. 9) mung bean protein concentrate is pre­pared from whole mung beans in a step-by-step process using simple equipment. Mung bean paste is prepared using a tradition­al stone mill followed by centrifugation to separate out the mung bean protein solution and the starch residue. Precipitation of the protein occurs by acidification of the boiled mung bean protein solution. Fresh mung bean protein isolate is dried in an air oven to produce a stable product containing 78.9% pro­tein and 2.9% moisture. The concentrate can be reconstituted in boiling water and used as a textured, meatlike product in traditional dishes (Bhumiratana and Nondasuta 1972).

The significance of this process is twofold, namely, it is a practical method that can be incorporated into a small­scale food industry in developing countries, and it is similar to the traditional process for manufacturing mung bean starch

53

from flour by wet starch extraction. In the process for preparing mung bean protein concentrate, therefore, mung bean starch is a by-product of the process.

Mung bean starch is widely used in the preparation of traditional oriental noodles. It is prepared by grinding whole dehulled mung beans into a paste. After adding more water, the slurry is passed through a fine mesh (cloth) to separate out the cellulose. The slurry is then allowed to settle on a series of sloped starch tables for several hours. Eventually, starch settles to the bottom of the aqueous mung bean solution while water concurrently runs off the sloped tables. The starch sediment is then collected from the tables and dried. In this process, the water, which is in fact a protein liquor, can ·be recovered by the described precipitation method to make mung bean protein concentrate.

The manufacture of protein concentrates from various legumes using this or other described methods deserves atten­tion since they mostly stem from conventional legume processing methods. Legume protein concentrates have nutritional sig­nificance because they are high in protein and their physical and chemical natures make them adaptable in various food uses. Both new and traditional foods can be prepared from legume protein concentrates leading to an increased amount of protein in the diet.

54

BY-PRODUCT UTILIZATION OF LEGUME PROCESSING

By-products that are either a direct or indirect result of legume processing or utilization have economic and social as well as nutritional importance. The main by-products that come directly from legume processing are husks, obtained by milling and decortication, and starch, obtained by protein extraction from air classification or slurry centrifugation. Those by-products obtained indirectly are parts of the legume plant. To this latter group belong vines, sterns, straw, leaves, and pods, which are part of typical leguminous crop plants. They are eaten either by man or animals.

An analysis of grain fractions of several legumes showed that the seed coat or husks contain a relatively low amount of protein (7-11%) and a high amount of fibre (20-32%) (Singh et al. 1968). Nutritional importance of husks as a ruminant feed source is apparent. The husk and pods of pigeon pea (C. cajan) are fed to cattle (Kachroo 1970). In addition, milling of pigeon pea for the preparation of dhal provides as much as 15% of total yield for cattle feed (Khare et al. 1966). Husk is a major portion of this amount with dhal powder and small brokens contributing the remainder. Also, by-products of milling chickpea (C. arietinum) are fed to livestock in India (Kurien et al. 1968; Purseglove 1968).

Starch manufacture was referred to in the discussion on processed legumes by new and improved technologies. In the preparation of legume concentrates by protein extraction (and precipitation), air classification, or slurry centrifugation, the production of legume starch is a very important economic factor. Mung bean (P. aureus) starch is traditionally used in the preparation of Chinese-style noodles. Other legume starches can be used in food preparation, which ordinarily require a food starch. Knowledge of the physicochernical properties of these starches in relation to their gelatinization properties, swelling capacity, and pasting characteristics would aid in further development and utilization of these starches in food products.

Relatively few studies on the starches and low-molecular carbohydrates of legumes have been reported. Lineback and Ke (1975) found that the arnylose (linear fraction) contents of chickpea and horse bean (V. faba) starches were 20.6 and 29.3% respectively, on a dry weight basis. Corresponding amylopectin

55

(branched fraction) contents of the starches were 73.2 and 66.8%, respectively. The legume flours (68-72% extraction) contained about 40% starch. These starches exhibited single stage swell­ing during gelatinization. The Brabender hot paste viscositv patterns were relatively stable during stirring at 95°c -indicating that the granules are very resistant to swelling and fragmentation. Pasting temperatures (67-68.5°c) for both starches are higher than the 56-59°C reported for wheat starches. Lineback and Ke (1975) further reported an 8% content of low­molecular carbohydrates in the legume starches, the predominant one being sucrose. Lesser but relatively substantial amounts of stachyose and raffinose were found. A study of cowpea (i:igna sinensis) starch illustrated similar physicochemical properties associated with legume starches. Tolmasquim et al. (1971) reported that cowpea starch contains a high linear con­tent (amylose). In addition, cowpea starch granules undergo limited swelling when heated to a temperature above that of gelatinization and have great resistance against shear at 95°c. The paste, upon cooling, shows a high viscosity, indicative of retrogradation. An average amylose content of 30-40% was reported by Schoch and Maywald (1968) in the starch fraction of lentils, lima beans, chickpeas, and navy beans. (The amylose/amylopectin content (%) in wheat starch is 23/67.) Lima bean starch has properties similar to cereal starches, i.e. moderate granule swelling and a low pasting peak. In contrast, and in agreement with other studies, lentil, chickpea, and navy bean starch granules exhibited restricted swelling, reduced solubilization, stable hot paste viscosities with no pasting peak, and a resistance to mechanical fragmentation.

The physicochemical properties associated with legume starches have direct bearing on new types of legmne processing and utilization. The reported size of starch granules was 10-14 microns for cowpeas, whereas chickpeas and broad beans contained some starch granules that were 17-29 microns and 17-31 microns in size, respectively (Tolmasquim et al. 1971; Lineback and Ke 1975). Starch granule size is an important feature when considering air classification for protein-starch separation. One main property of many legume starches, as associated with food uses and product development, indicates that it is possible to obtain viscous and stable pastes even when low concentrations of the starch are used.

The various parts of leguminous plants can provide an additional source of food and feed. Accordingly, dried vines from chickpea plants are used as cattle feed and green parts of pigeon pea and black gram (P. rrrungo) plants provide excellent fodder, while broad beans and cowpeas are also grown for hay and silage (Purseglove 1968). The roots of cowpeas are eaten in the same way as sweet potatoes in East Africa (Purse­glove 1968).

56

In India, stalks from pigeon pea plants are used for various purposes, including roofing and basket making. In addition, old woody parts of this plant are used for fuel, and charcoal is made from the thick stem (Kachroo 1970).

The development of legume-processing industries in rural areas has the advantage of economically utilizing by-products as animal feed. Alternatively, food use application of other­wise feed by-products can be developed by modified microbial technology. The conversion of waste carbohydrate into protein by the growth of microorganisms, namely bacteria, moulds, and yeast, is similar to established simple processes for preparing various traditional fermented foods. The production of single­cell proteins, based on substrates of legumes processing food wastes warrants special consideration.

57

SUMMARY AND CONCLUSION

In presenting a review of legume processing and utiliza­tion, the potential for increasing the consumption of legumes, based on new and improved processing technologies is evident. Investigations into the modification of present processing systems, the adaptation of these systems to a wide variety of legumes, and the development of new and simple processing technologies, thus providing processed legumes in food forms that fit existing legume consumption patterns, are needed.

Traditional legume foods based on present processing methods are listed in Table 20. It is apparent that many of the traditional foods can be prepared from a variety of legumes. The availability of the processed legume forms, primarily whole, husked legumes, and flours can promote increased legume consumption. Mechanical decortication by abrasive milling, followed by grinding and sieving, as previously discussed in this report, can play a major role in providing an economical source of processed legumes.

Similarly, new legume foods can be developed based on new processing methods. As illustrated in Table 21, legumes in the form of instant powders, protein concentrates, and flour lend themselves to a variety of food uses. The introduction of these methods as part of the rural food industry can promote their increased use.

Various investigations have been conducted into the prepara­tion of protein-rich foods, based on cereals and legumes. These have included reconstitutable powders, for both adult and child feeding, milky beverages, soft drinks, soups, baked goods, and pasta products (Weisberg 1972).

Bal Ahar, an infant food currently being marketed in India, contains a mixture of 70% cereal flours (wheat or maize) and 30% legume flours (groundnut, chickpea, and mung beans). It has a protein content of 22% and a 2.2 (standard casein, 2.5) protein efficiency ratio. In addition, the development of high-protein low-cost items in the form of biscuits has gained considerable attention (Parpia 1969). Various blends of chickpea, broad bean, and lentil flours, in combination with a cereal staple have been used in the preparation of reconstitut­able powders for infant feeding in Algeria, Egypt, and Turkey

58

(Kapsiotis 1969}. In Lebanon, a mixture of parboiled wheat (70%} and chickpea flour (30%} was used to develop Laubina, an infant food containing over 15% protein and having a protein efficiency ratio of 2.1, compared with standard casein (Tannous et al. 1965). The prevalence of malnutrition among children in Ethiopia has prompted research into the development of protein supplements containing chickpea and pea flours, combined with cereal staples (Agren et al. 1969). Similarly, in Peru, broad bean flour has been used in the initial investiga­tion of protein supplements for use in baked goods, soups, and noodles (Bacigalupo 1969).

Future Research Needs

Specific interest in protein-food promotion has focused on the preparation of bread in which a percentage of the wheat flour is replaced by specific legume flours. This mixture of wheat flour with a legume protein source is termed a "composite flour." Various investigations into breads from composite flours have involved the use of soybean flour and concentrates. However, other food legumes have been overlooked (Hulse 1974). The increasing popularity of bread in many of the less-developed countries (FAO 1969) warrants broader research in this area.

The use of legume flours in bread-making is presently receiving support from IDRC. Research into composite flour technology, based on a low-energy mechanical dough development process, has been investigated (Bushuk and Hulse 1974; McConnell et al. 1974). The introduction of a bread-making system suitable for preparing ethnic breads containing legume flours can pro­note legume utilization in an important dietary staple in developing countries.

Future research in legume processing and utilization must reflect the need to increase their status to one of an accept­able, nutritious, and easy-to-prepare food product. Specific areas of research that have been overlooked include legume cooking quality, properties of legume constituents, biochemical factors in legumes associated with their nutritive value, and the role of legume carbohydrates in flatulence.

Studies on the chemical nature of whole legumes and husks are necessary to provide information for establishing efficient cooking methods. Procedures for objectively determining the cooking qualities of legumes should be established. These methods should be based on desirable functional properties associated with food preparation. An investigation into the cooking quality of stored beans is also needed. Bean quality for cooking decreases rapidly during storage, thereby causing a reduction in bean consumption. A suggested approach

59

(National Academy of Sciences 1974) to this problem would include studies designed to (a) correlate storage conditions (temperature, relative humidity, and time) and cooking quality of good quality seeds; and (b) detect biochemical changes in the seed after harvest. This research would identify optimum storage conditions for various bean varieties and suggest physical or chemical treatments to inhibit undesirable changes or enhance desirable ones.

In addition, work concerning the agronomic aspects of legume cooking quality is required. In such a study, the influence of soil minerals and other environmental factors upon the cooking characteristics of legumes would be monitored. M::>difications to traditional cooking practices could subsequently be undertaken.

Properties of specific legume constituents, particularly proteins, starches, and lipids, would provide information on the factors responsible for desirable sensory attributes in food products. A systematic approach to the characterization of these constituents, as well as a study of their physico­chernical properties, could be conducted on specific popular food preparations in developing countries.

Additional research is needed to identify the antinutri­tional factors present in legumes. The unavailability of many essential amino acids in legumes that have been cooked to the point at which all the known toxic factors have been destroyed suggests the presence of other antinutritional factors. As stated by the National Academy of Sciences (1974), primary con­cern should be the presence of factors that interfere with proteolysis, their identification, and the development of processing methods that inactivate or destroy them.

Flatulence, i.e. the production of gas in the intestine or stomach, has long been associated with the ingestion of different leguminous seeds. The presence of oligosaccharides of the raffinose family in these food materials are suspected of playing a major part in flatus formation. The two primary oligosaccharides, stachyose and verbascose, have been reported in a variety of legumes including chickpeas, cowpeas, field beans, lentils, pigeon peas, and lima beans. High levels of stachyose (2.5, 2.7%) and verbascose (4.1, 4.2%) have been found in chickpeas and pigeon peas, respectively (Cristofaro et al. 1974). Initial investigations into the processing of legumes to lower the oligosaccharide content has suggested soaking or blanching as possible methods. Further investigation into this and other methods of processing to eliminate flatus formation, e.g. germination, fermentation, is warranted.

Cristofaro et al. (1974) concluded that the sugars of the

60

raf finose family are not digested because of the absence of the enzyme alpha-galactosidase. These large carbohydrate I1Plecules subsequently come in contact with bacteria of the lower intestine resulting in flatus formation. Although it is clear that stachyose and verbascose are involved in flatulence, they may not be the sole responsible factor.

The role of legumes in the diet of people living in developed countries is relatively minor. These grains, however,. can provide an improved nutritional status to a large number of people living in developing countries who otherwise would be deprived of high-protein sources. At a time when population pressures are straining available food supplies, the increased availability and utilization of food legumes should be an important goal for all concerned agricultural scientists.

61

Table 1. Calories and protein supplied by cereal grains, food legumes, and nuts in diets in the developing countries (% of total calories and proteins).

Calories Protein

Africa Asia and Far East Latin America Near East All developing countries

Source: FAO (197lc).

58.5 72.1 47.0 67.0 65.2

Table 2. Protein consumption (grams per day) per capita by major food groups.

Starchy Pulses Vegetables Cereals roots & nuts & fruit

North America 15.9 2.4 4.1 4.9 Central America 31.6 0.5 11.9 2.0 Caribbean 21.4 3.3 8.3 2.2 Africa 33.3 5.2 8.5 1.9 Near East 45.1 0.6 4.7 3.3 South Asia 32.3 0.5 8.6 0.6 China 31.8 2.9 10.8 2.2

Source: Abbott (1973) •

62

61. 3 77.3 54.8 72.0 70.3

Meat, eggs, fish, & milk

70.7 22.8 22.8 12.1 12.2 6.3 8.8

Total

98.2 58.0 58.0 61.0 65.9 48.3 56.5

CJ' w

Table 3. Comparative annual average rate of growth (%) in production, selected crops, 1961-63 to 1969-71.

Total agricultural production, Total

1961-63 to 1973 Wheat Rice grains

Africa, south of the Sahara 2.7 5.1 4.4 2.5

Far East 2.9 7.5 2.5 3.1

Near East and Northwest Africa 3.2 2.0 4.5 2.4

Latin America 2.9 1.3 3.4 3.3

Total 2.9 4.4 2.7 3.0

Source: FAO (1973a); UN (1974).

Starchy roots Pulses

2.7 1.9

5.5 -0.7

2.9 3.0

4.7 3.0

3.9 0.7

Table 4.

Developed Latin America Near East Asia and Far East Africa All developing World

Percentage change in world population and pulse production, 1949-72.

Population

22 62 57 51 52 53 40

Total food

60 65 65 65 47 62 61

Food per capita

32 2 2 9

-3 6

15

Pulses

87 100

48 21 88 40 49

Source: FAO (197lb, 1973b).

Table 5. Possible protein supplements for cereal foods and feeds.

Synthetic amino acids

Lysine

Methionine

Plant proteins

Legume flours Oilseed flours Cereal germ Cereal bran

Animal proteins Fish proteins

Egg Fish meal (dried and defatted fish)

Milk Fish protein concentrate (solvent extracted fish

Whey flesh or whole fish: ca. 90% protein)

Animal blood Meat by-products Bird feathers Processed wool

Microbial proteins

Yeast )Substrates )include carbo-

Algae )hydrates, hydro-

*Cereal protein concentrate *Legume protein concentrate Leaf and grass concentrates Coffee pulp concentrates

) carbons (natural Micro fungi)gas, petroleum

)wax, etc.), Bacteria )industrial and

)agricultural )wastes, and ) by-products

* By particle size separation, aqueous, or solvent extraction.

64

Table 6. Protein content of principal calorie and protein sources.

Protein content {g/100 g)

Rice, brown or husked Maize, grain or whole meal Millet, grain Oats, meal Rye, whole meal Wheat, whole grain

germ bran

Bean (Phaseolus vulgaris) Broad bean (Vicia faba) Chickpea (Cicer arietinum) Cowpea (Vigna spp.) Groundnut (Arachis hypogaea) Lentil (Lens esculenta) Lupin (Lupinus spp.) Peas (Pisum sativum) Pigeon pea (Cajanus cajan) Soybean, seed (Glycine max)

cake

Source: FAO (1970).

65

7.5 9.5 9.7

13.0 11.0 12.2 22.9 13.6 22.1 23.4 20.1 23.4 25.6 24.2 31.2 22.5 20.9 38.0 46.0

Table 7. Proximate analyses* of some food legumes in the semi-arid tropics.

Nutrient

Legume (botanical name) Moisture Protein Ash Fibre Fat Carbohydrate % % % % % %

O'I Cajanus cajan 10.l 19.2 3.8 8.1 1.5 57.3

O'I Cicer arietinum 9.8 17.1 2.7 3.9 5.3 61.2 Lens esculenta 11.2 25.0 3.3 3.7 1.0 55.8 Phaseolus aureus 9.7 23.6 4.0 3.3 1. 2 58.2 Phaseolus lunatus 12.6 20.7 3.7 4.3 1. 3 57.3 Phaseolus mun go 9.7 23.4 4.8 3.8 1.0 57.3 Phaseolus vulgar is 11.0 22.0 3.6 4.0 1.6 57.8 Pis um sativum 10.6 22.5 3.0 4.4 1.0 58.5 Vici a faba 14.3 25.4 3.2 7.1 1.5 48.5 Vigna unguiculata 11.0 23.4 3.6 3.9 1. 3 56.8

*Whole dried seed.

Source: Purseglove (1968) •

Table 8.

Botanical name

Cajanus cajan

Cicer arietinum

Lens esculenta

Phaseolus aureus (Vigna radiatus)

Phaseolus lunatus

Phaseolus mungo (Vigna mungo)

Phaseolus vulgaris

Pisum sativum

Legumes in human nutrition.

Common name(s)*

Pigeon pea Red gram Congo bean

Chickpea Bengal gram Gram

Lentil Split pea Red dhal

Mung bean Green gram Golden gram

Lima bean Sieva bean Butter bean

Black gram

Kidney bean Pea bean Navy bean Haricot bean Pinto bean Snap bean Common bean Black bean

Green pea Garden pea Pea

67

Areas of consumption

India, Pakistan, Middle East, East Africa

India, Pakistan

Near East, North Africa, India, Central and South America

South, Southeast, and East Asia, East Africa, India

Tropical America, West Indies, Madagascar

India, Iran, East Africa, West Indies

North, Central, and South America, Mexico, East Africa

Mainly temperate zones, parts of India and Africa

(con' t)

{Table 8, con't)

Vicia faba

Vigna unguiculata (Vigna sinensis)

Broad bean Horse bean Faba bean Windsor bean

Cowpea Blackeye pea Catjan cowpea Hindu cowpea Kaffir bean

Temperate zones, Near East, North Africa, Central and South America

Asia, Tropical Africa, West Indies, China

*First listing represents the most common name for the variety in the region where it is usually consumed.

Sources: Aykroyd and Doughty (1964); Stanton et al. (1966); Purseglove (1968); Kachroo (1970).

68

Table 9. Amino acid content of wheat plus chickpea (Cicer arietinum) {mg/gN).

WHO Amino recommended

Chickpea* Wheat {67%), acid amino acid Wheat* (Cicer arietinum) chickpea { 33%) Score composition**

°' Lysine 179 428 304 89 340 ~

Threonine 185 235 209 85 250 Methionine

& cystine 253 139 196 89 220 Leucine 417 468 443 100 440 Isoleucine 204 277 241 96 250 Valine 276 284 280 90 310 Phenylalanine

& tyrosine 469 541 505 133 380 Tryptophan 68 50 59 100 60

Source: *FAO {1970) ; **WHO {1973) •

Table 10. Amino acid content of rice plus chickpea(Cicer arietinum) (mg/gN).

WHO Amino recommended

Rice (75%) , Acid amino acid Rice* Chickpea* chickpea (25%) Score composition**

Lysine 237 428 333 98 340 Threonine 244 235 240 96 250

-..i Methionine 0 & cystine 212 139 176 80 220

Leucine 514 468 491 112 440 Isoleucine 238 277 258 103 250 Valine 344 284 314 101 310 Phenylalanine

& tyrosine 540 541 540 142 380 Tryptophan 102 50 76 126 60

Source: *FAO (1970) ; **W»O (1973).

-.J ......

Table 11. Available supplies in energy and protein in 1965 and 1970 expressed as a percentage of per capita nutritional requirements.

Available supplies as % of res,uirements

Daily per capita requirements* 1965 1970

calories Protein** Calories Protein** Calories Protein**

World 2385 38.7 100 169 101

Developed regions 2560 39.5 116 221 121

Developing regions*** 2284 38.4 93 142 96

Asia and East Asia*** 2223 36.6 89 135 93 Latin America 2383 37.7 104 169 106 Africa 2335 41.5 92 140 93 Near East 2456 45.5 94 145 97

Asian centrally planned economies 2355 38.3 86 151 88

*Revised standards of average requirements (physiological requirements plus 10% for food wastage at household level).

173

229

147

141 172 141 147

153

**Provisional data expressed in grams of local protein, i.e., adjusted for protein quality difference in national diets as compared to an ideal or reference protein.

***Excluding Asian countries with centrally planned economies. Source: Kracht (1974)0

Table 12. World production of major legumes, 1972.

Botanical name

Glycine max Arachis hypogaea Phaseolus vulgaris Pisum sativum Cicer arietinum Vicia faba Cajanus cajan Vigna unguiculata

(Vigna sinensis) Lens esculenta

World total

Source: FAO (1973b).

World production (1000 metric tons)

53024 16887 10899 10218

6718 5326 1720

1260 1182

107234

72

% world legume production

49.4 15.7 10.2 9.5 6.3 5.0 1.6

1.2 1.1

100

-..J w

Table 13.

Beans Peas Broad beans Lentils* Soybeans Groundnuts Rapeseed Cottonseed Sesame*

*Metric tons.

Price fluctuations of the principal legume crops and oilseeds in international trade 1972, 1973, 1974, 1975 (for long tons (2240 lb), in U.S. dollars).

1972 1973 1974 January Jul/Aug January Jul/Aug January Jul/Aug January

336.60 269.50 270.25 465.50 1173.00 1071. 00 432.90 149.18 183.75 219.73 392.00 816.50 583.10 468.00 103.00 85.75 103.40 196.00 149.50 154.70 150.93 178.50 176.40 197.40 428.75 425.50 483.14 456.30 132.86 140.63 270.75 276.35 259. 00 300.50 251.18 267.05 324.30 529.00 655.20 126.48 135.24 177.43 419.75 458.06 97.00 106.00 225.00 235.00 260.00

330.00 327.00 310. 00 480.00 585.00 635.00

Source: The Public Ledger (1972-75).

1975 Jul/Aug

386.26 342.20 121. 52 466.55 224.50 272. 34

175.00 650.00

Table 14. Volume and value of pulse exports and imports, and as a percentage of production, 1972.

Exports Imports 1972 1972 Exports as Imports as

1000 mt $m 1000 mt $m % of produc. % of produc.

World 1893 349 1987 401 4 5

Developed 662 129 1334 268 18 36 N. America 307 63 29 9 27 26 w. Europe 311 59 1071 213 15 52 Oceania 37 5 16 5 47 20 Other 6 2 217 42 15 54

Developing 960 172 594 123 4 3 Africa 428 61 77 17 9 2 Latin America 168 40 223 55 4 5 Near East 132 29 106 19 8 6 Far East 232 41 183 31 2 1

Central planned 271 48 59 10 2 0 Asia 119 18 25 3 1 0 USSR & Europe 152 30 34 7 2 0

Source: FAO (1973c).

74

Table 15. Demand projections, pulses and nuts.

Level of total demand (1000 metric tons) % increase

1965 1975 1980* 1965-80

World 34641 45340 52615 52

Economic class I 3803 4285 4510 19 North America 857 863 914 7 Western Europe 2397 2733 2877 20 EEC 1143 1251 1320 15 Oceania 41 52 60 46 Other developed 543 626 687 27

Economic class II 23773 32440 38049 60 Africa 3332 4486 5276 58 Latin America 4478 6042 7035 57 Near East 1069 1562 1892 77 Asia and Far East 14744 20207 23812 62

Economic class III 7107 8855 9954 40

Asian centrally planned economies 5490 7032 7880 44 USSR - Eastern Europe 1563 1826 1999 28

*Trend projection.

Source: FAO (197la).

75

Table 16. Yield of various products during milling (wet method) of pigeon pea (Cajanus cajan)

Product composition

Dust and dirt Dust, shriveled grain, and

pulse (6.25% of yield) Pulse, shriveled grain,

stems, other grains

Pulse (large size) and stems

Fine husk, dhal flour, fine broken dhal

Husk - coarse Husk - fine Dhal flour and free husk Pulse with husk

Big broken dhal Dhal* - dehusked split pulse Unaccounted losses - drying,

loss of dhal flour, dirt, etc.

Yield (%)

o. 71

4.50

2.29

2.95

9.16 0.87 4.39 2.44

13.40

1.89 54.42

2.98

Remarks

Rejected

Cattle feed

Pulse (38. 75%) sieved out

Pulse (89.5%) sieved out

Cattle feed Cattle feed Cattle feed Cattle feed Recondition with water, shelling

Low-grade dhal Dhal

*Total yield of dhal on basis of actual pulse milled was 76.1%.

Source: Khare et al. (1966).

76

Table 17. Pulse milling dry vs. wet method.

Advantages

Dry method

Cooks better, cotyledon softens easily

Larger quantities of dhal can be prepared

More advantageous method of dehusking

Wet method

15-20% higher dhal yields

Better taste

Easy husk removal due to pre-soaking

Traditional chakki (sheller) is more suitable for wet­processed grains

More advantageous method of splitting

77

Disadvantages

More costly due to higher percentage of broken dhal

Requires longer processing time due to repeated drying and milling steps

Requires longer cooking time

Small depression in centre of dhal is present

Laborious and dependent on climatic conditions

Causes losses of grain from birds and insects during outdoor drying

Table 18.

English name

Bengal gram Chickpea Red gram Pigeon pea Green gram Black gram

Lentil

Pulse varieties of India.

Regional name

Channa, gram

Arhar, Tur

Mung, Mug Udid, Urid Mash, Kalai Maur, Masoor

Botanical name

Cicer arietinum

Cajanus cajan

Vigna radiatus Phaseolus mungo

Lens esculenta

source: Pulses in India, Conference on Milling of Dhal, Mysore, 17-19 Feb 1971.

Table 19. Average yield of dhal from different legumes using various methods.

Maximum Horne- Traditional Improved theoretical scale commercial CF TRI

yield, methods, methods, process, Legume % % % %

Bengal gram (chickpea) 88 75 75 84

Red gram (pigeon pea) 88 68 75 85

Black gram 87 63 71 82 Green gram

(rnung bean) 89 62 65 83

78

Table 20.

Processing method

Pounding, grinding, milling

Boiling

Roasting, parching

Frying

Puffing

Germination

Fermentation

Agglomeration

Canning, (steaming)

Traditional food legume processing and utilization.

Processed form

Meal, flour, grits, paste, dhal, whole

Whole, husked and unhusked

Whole, husked and unhusked

Whole, husked and unhusked, flour, paste, batter

Whole, unhusked

Whole, unhusked

Flour, paste, batter

Flour

Whole, unhusked

79

Food form

Unleavened breads, biscuits, cakes, noodles, porridges, gruels, stews, sauces

Stews, vegetable dishes soups, condiments

Snacks, garnish

Snacks, breads, cakes

Snacks

Bean sprouts, curry, vegetable dishes

Fried doughs, oriental foods, condiments

Cereals, cous-cous

Vegetable dishes, salads

Table 21. Legume processing and utilization by new and improved technologies.

Processing method

Mechanical decortication and milling

Precooking, drying, grinding

Agglomeration, air classification, precipi ta ti on

Processed form

Whole legumes, flour, grits, paste

Whole, dehusked legumes, quick­cooking legumes, instant legume powders

Legume protein concentrates

80

Food form

Traditional foods: porridge, stew, gruel, cous-cous, etc.

New foods: ethnic breads, pasta products, snack foods (e.g. chips), confectionaries

Traditional foods as above

New foods: beverages, snack foods and dips, soups, oriental foods

Legume-cereal mixtures: ethnic breads, baked goods, pasta products, infant foods, cakes, pancakes

Simulated foods: rice, meat analogs

Meat extenders Snack foods

REFERENCES

Abbott, J. C. Protein's progress and prospects. of feed protein conference, Quebec City, Canada, 1973, 59.

In Proceedings 8-9 November

Acharya, B. N., Niyogi, s. P., and Patwardhan, V. N. Balanced diets, part III. The effect of parching on the biological value of the proteins of some cereals and pulses. Indian Journal of Medical Research (New Delhi), 30, 1942, 73.

Abradale Press. The Holy Bible. New York, Abradale Press, 1959, 888, 940.

Agren, G., Hofvander, J., Selinus, R., and Vahlquist, B. Faffa: a supplementary cereal-based weaning food in Ethiopia. In Milner, M., ed., Protein-Enriched Cereal Foods for World Needs. St. Paul, Minnesota, American Association of Cereal Chemists, 1969, 278.

Altschul, A. M. Food: protein for humans. Chemical Engineering News (Washington), 47, 1969, 69.

Aykroyd, W. R., and Doughty, J. FAO, Rome. Legumes in human nutrition. Rome, Food and Agriculture Organization Nutritional Studies No. 19, 1964.

Bacigalupo, A. Protein rich foods in Peru. In Milner, M., ed., Protein-Enriched Cereal Foods for World Needs. St. Paul, Minnesota, American Association of Cereal Chemists, 1969, 288.

Banerjee, s., Prohatgi, K., and Lahire, s. Pantothenic acid, frolic acid, biotin, and niacin content of germinated pulses. Food Research (Champaign, Ill.), 19, 1954, 134.

Bhumiratana, A., and Nondasuta, A. Report on protein food development project. Bangkok, Nutrition Division, Health Department, Ministry of Public Health, Thailand, 1972.

Bressani, R., and Elias, L. G. Legume foods. In Altschul, A. M., ed., New Protein Foods. New York, Academic Press, 1974, 230-297.

Bressani, R., Elias, L. G., and Valiente, A. T. Effect of cooking and of amino acid supplementation on the nutritive

81

value of black beans (Phaseolus vulgaris L.). British Journal of Nutrition (London), 17, 1963, 69.

Bressani, R., Flores, M., and Elias, L. G. Acceptability and value of food legumes in the human diet. In Potentials of Field Beans and Other Food Legumes in Latin America: A Seminar. Cali, Colombia, Centro Internacional de Agricultura Tropical (CIAT), Feb-Mar 1973, 17.

Burr, H.K., Kon, s., and M:>rris, H. J. Cooking rates of dry beans as influenced by I1XJisture content and temperature and time of storage. Food Technology (Chicago), 22, 1968, 336.

Bushuk, w., and Hulse, J. H. Dough development by sheeting and its application to bread production from composite flours. Cereal Science Today (St. Paul, Minn.), 19, 1974, 424.

Chattopadhyay, H., and Banerjee, s. Effect of germination on the total tocopherol content of pulses and cereals. Food Research (Champaign, Ill.), 17, 1952a, 402.

Studies on the ascorbic acid oxidase activity of some comnvn Indian pulses during germination. Indian Journal of Medical Research (New Delhi), 46, 1952b, 439.

Chattopadhyay, H., Mandi, N., and Banerjee, s. Studies on the effect of germination on the fat and the carbohydrate content of pulses. Indian Journal of Physiology (Calcutta), 4, 1950, 66.

Copley, M. J. Dry bean research - some accomplishments and unsolved problems. In Dickson, M. H., ed., Proceedings of the Bean Improvement Co-operative and National Dry Bean Research Association Conference, Geneva, New York, New York State Agricul­tural Experiment Station, 1974, 53.

Cristofaro, E., M:>ttu, F., and Wuhrmann, J. Involvement of the raffinose family of oligo-saccharides in flatulence. In Sipple, H. L., and McNutt, K. w., eds., Sugars in Nutrition. New York, Academic Press, 1974, 313.

Dorsey, w. R., Strashun, s. I., Roberts, R. L., and Johnson, K. R. New continuous production facility for processing instant pre­cooked beans. Food Technology (Chicago), 15, 1961, 13.

Dovlo, F. E., Williams, c. E., and Zoaka, L. Cowpeas: home preparation and use in West Africa. Ottawa, International Develop­ment Research Centre, IDRC-055e, 1975.

Ebine, H. Fermented soybean foods in Japan. a Symposium on Food Legumes. Tokyo, Japan. Research Series No. 6, Sep 1972, 217.

82

In Proceedings of Tropical Agriculture

Elias, L. G., Bressani, R., and Flores, M. Problems and potentials in storage and processing of food legumes in Latin AI1Erica. In Potentials of Field Beans and Other Food Legumes in Latin America: A Seminar. Cali, Colombia, Centro Internacional de Agricultura Tropical (CIAT), Feb-Mar 1973, 52.

Esselen, W. B., and Davis, S. G. Dehydrated baked beans. Canner Packer (Barrington, Ill.), 95, 1942, 19.

Fan, T. Y., and Sosulski, F. W. Dispersibility and isolation of protein from legume flours. Canadian Institute of Food Science and Technology (Ottawa), 7, 1974, 256.

Feldberg, c., Fritzche, H. W., and Wagner, J. R. Preparation and evaluation of pre-cooked, dehydrated bean products. Food Technology (Chicago), 10, 1956, 523.

Food and Agriculture Organization of the United Nations. Composite flour program. Bread from composite flour. Rome, FAO, 1969.

Amino acid content of foods. Rome, FAO, 1970. Agricultural commodity projections 1970-1980. Vol. I and

II. Rome, FAO, 197la. Production yearbook 1970. Vol. 24, Rome, FAO, 197lb. Agricultural production in developing countries in

relation to the targets for the second United Nations development decade. r-Dnthly Bulletin of Agricultural Economics and Statistics (Rome), 22(4), 1973a, 7-8.

Production yearbook 1972. Trade yearbook. Vol. 26.

Vol. 26. Rome, FAO, 1973b. Rome, FAO, 1973c.

Gainor, V. J., Robinson, R. J., Ward, A. B., and Hoover, W. J. Effective processing and protein supplementation (horse bean flour) on cous-cous. In Improving the Nutritive Value of Cereal Based Foods. Progress Report No. 3. Manhattan, Kansas, Agency for International Development, Food and Feed Grain Institute, Kansas State University of Agriculture and Applied Science, June 1971.

Gallardo, F., Araya, H., Pak, N., and Tagle, M. A. Toxic factors in Chilean legumes, II. Trypsin inhibitor activity. (English Summary.) Archivos Latinoamericanos de Nutricion (Caracas), 24, 1974, 101.

Toxic factors, III. Haemagglutinating activity. (English Summary.) Archivos Latinoamericanos de Nutricion (Caracas), 24, 1974, 191.

Galle, E. L. Flour product and method of making. Washington, D.C., United States Patent Office, Patent No. 3,397,067, 13 Aug 1968.

Hackler, L. R. Nutritional qualities of dry beans and the

83

potential for improvement. In Dickson, M. H., ed., Proceedings of the Bean Improvement Co-operative and National Dry Bean Research Association Conference. Geneva, New York, New York State Agricultural Experiment Station, 1974, 67.

Hang, Y. D., Wilkens, W. F., Hill, A. s., Steinkraus, K. H., and Hackler, L. R. Studies on the extraction of proteins from red kidney beans. I. Improvement in filterability of aqueous bean extract. Cereal Chemistry {St. Paul, Minn.), 47, 1970, 259.

Hawtin, L. Chickpea recipes from the Near East. Nov 1974. (Unpublished)

Hegazi, s. M. Effect of germination on the carbohydrate, protein, and amino acid contents of broad beans. Zeitschrift fur Ernahrungswissenschaft (Darmstadt, W. Germany), 13, 1974.

Hulse, J. H. The protein enrichment of bread and baked products. In Altschul, A., ed., New Protein Foods, Vol. lA. New York, Academic Press, 1974, 155.

Problems of nutritional quality of pigeon pea and chickpea and prospects of research. Report to the International Crops Research Institute for the Semi-Arid Tropics, Hyderabad, Jan 1975. (Unpublished)

Hulse, J. H., Fawcett, B., and Daniels, W. D. Protein supple­ments - world production and trade. In Harapiak, J. T., ed., Oilseed and Pulse Crops in Western Canada. Calgary, Alberta, Western Co-operative Fertilizers Ltd., 1975, 1.

Kachroo, P., ed. Pulse crops in India. New Delhi, Indian Council of Agricultural Research, 1970.

Kakode, M. L., and Evans, R. J. Nutritive value of navy beans (Phaseolus vulgaris). British Journal of Nutrition (London) 19, 1965, 269.

Effect of soaking and germinating on the nutritive value of navy beans. Journal of Food Science (Chicago), 31, 1966, 781.

Kapsiotis, G. D. History and status of specific protein-rich foods - FAO/WHO/UNICEF protein food programmes and products. In Milner, M., Protein-Enriched Cereal Foods for World Needs. St. Paul, Minnesota, American Association of Cereal Chemists, 1969, 255.

Khare, R. N., Krishnamurthy, K., losses of food grains. Part I. (Cajanus cajan) during milling. (New Delhi), 4, 1966, 1250

and Pingale, s. V. Milling Studies on losses of red gram Bulletin of Grain Technology

84

Kon, S. Pectic substances of dry beans and their possible correlation with cooking time. Journal of Food Science (Chicago), 33, 1968, 437.

Kon, S., Brown, A. H., Ohanneson, J. D., and Booth, A. N. Split peeled beans: preparation and some properties. Journal of Food Science (Chicago), 38, 1973, 496.

Kon, s., Wagner, J. R., and Booth, A. N. Legume powders: preparation and some nutritional and physiochemical properties. Journal of Food Science (Chicago), 39, 1974, 897.

Kracht, u. The nutritional aspects of food production and distribution. Impact of Science and Society (Paris), 24(2), 1974, 157.

Kurien, P. P., Desikachar, H. s. R., and Parpia, H. A. B. Processing and utilisation of grain legumes in India. In Proceedings of a Symposium on Food Legumes. Tokyo, Japan, Tropical Agriculture Research Series No. 6, Sep 1972, 225.

Kurien, P. P., and Parpia, H. A. B. Pulse milling in India I - processing and milling of Fur Arahar (Cajanas cajan Linn.). Journal of Food Science and Technology (Mysore), 5, 1968, 203.

Kurien, P. P., Desikachar, H. s. R., and Parpia, H. A. B. Processing and utilisation of grain legumes in India. In Proceedings of a Symposium on Food Legumes. Tokyo, Japan, Tropical Agriculture Research Series No. 6, Sep 1972, 225.

Kurien, P. P., and Parpia, H. A. B. Pulse milling in India I - processing and milling of Fur Arahar (Cajanas cajan Linn.). Journal of Food Science and Technology (Mysore), 5, 1968, 203.

Liener, I. Toxic factors in edible legumes and their elimina­tion. American Journal of Clinical Nutrition (Bethesda), 2, 1962, 281.

Toxic factors associated with legume proteins. Indian Journal of Nutrition and Dietetics (Coimbatore), 10, 1973, 303.

Lineback, D. R., and Ke, c. H. Starches and low molecular­weight carbohydrates from chickpea and horse bean flours. Cereal Chemistry (St. Paul, Minn.), 52, 1975, 334.

Lolas, G. M., and Markakis, P. Phytic acid and other phosphorous compounds of beans (Phaseolus vulgaris L.). Journal of Agricultural and Food Chemistry (Washington), 23, 1975, 13.

McConnell, L. M., Simmonds, D. H., and Bushuk, w. High-protein b~ead from wheat/faba bean composite flours. Cereal Science Today (St. Paul, Minn.), 19, 1974, 517.

85

Meyer, E. W. ed., Proteins 1970, 346-362.

Soy protein isolates for food. In Lawrie, R. A., as Htunan Food. Westport, AVI Publishing Company,

Miller, C. F., Guadagni, D. G., and Kon, S. Vitamin retention in bean products: cooked, canned and instant bean powders. Journal of Food Science (Chicago), 38, 1973, 493.

Morris, H. J., and Wood, E. R. on keeping quality of dry beans. 10, 1956, 225.

Influence of moisture content Food Technology (Chicago),

Muller, F. Cooking quality of pulses. Journal of the Science of Food and Agriculture (London), 18, 1967, 292.

Muneta, P. Cooking time of dry beans after extended storage. Food Technology (Chicago), 18, 1964, 1240.

National Academy of Sciences, National Research Council. Food science in developing countries. Washington, D.C., National Academy of Sciences, 1974.

Nelson, J. H., and Richardson, G. H. Flavour production by mould in oriental foods. In Peppler, H. G., ed., Microbial Technology. New York, Reinhold Publishing Company, 1967, 94-102.

Pant, R., and Tulsiani, D. R. P. Solubility, amino acid com­position, and biological evaluation of proteins isolated from leguminous seeds. Journal of Agricultural and Food Chemistry (Washington), 17, 1969, 361.

Parpia, H. A. Protein foods of India based on cereals, legumes, and oilseed meals. In Milner, M., ed., Protein-Enriched Cereal Foods for World Needs. St. Paul, Minnesota, American Association of Cereal Chemists, 1969, 129.

Patwardhan, V. N. Pulses and beans in human nutrition. American Journal of Clinical Nutrition (Bethesda), 2, 1962, 12.

Prairie Regional Laboratory, National Research Council of Canada, and College of Home Economics, University of Saskatchewan. Pea flour and pea protein concentrate. Saskatoon, Saskatchewan, Jun 1974.

The Public Ledger, 1972-1975. London, U.K. Publications Ltd., 1975.

Purseglove, J. W. Tropical crops dicotyledons 1. New York, John Wiley and Sons, Inc., 1968.

Reichert, R. B., Youngs, c. G., and Lorer, E. K. Processing

86

and utilisation of cereal grains and legumes. Ottawa, International Development Research Centre, 1974. (Unpublished)

Robinson, R. J., and Kao, c. Fermented foods from chickpeas, horse beans, and soy beans. Cereal Science Today (St. Paul, Minn.), 19, 1974, 397 (Abstract).

Rockland, L. B. Opportunities for conunercial development of new and improved bean products. In Dickson, M. H., ed., Proceedings of Bean Improvement Co-operative and National Dry Bean Research Association Conference, Geneva, New York, New York State Agricultural Experiment Station, 1974, 77.

Rockland, L. B., and Metzler, E. A. Quick-cooking lima and other dry beans. Food Technology (Chicago), 21, 1967, 30.

Rockland, L. B., Zaragosa, E. M., and Hahn, D. M. New information on the chemical, physical, and biological properties of dry beans. In Dickson, M. H., ed., Proceedings of Bean Improvement Co-operative and National Dry Bean Research Association Conference. Geneva, New York, New York State Agricultural Experiment Station, 1974, 93.

Schoch, T. J., and Maywald, E. c. Preparation and properties of various legume starches. Cereal Chemistry (St. Paul, Minn.), 45, 1968, 564.

Singh, s., Singh, H. D., and Sikka, N. c. Distribution of nutrients in the anatomical parts of common Indian pulses. Indian Journal of Agricultural Science (Calcutta), 45, 1968, 13.

Smith, A. K. Foreign uses of soya bean protein foods. Cereal Science Today (St. Paul, Minn.), 8, 1963, 191.

Stanton, w. R., Doughty, J., Orraca-Tetteh, R., and Steele, W. Grain legumes in Africa. Rome, Food and Agriculture Organization of the United Nations, 1966.

Steckle, J. M., and Ewanyk, L. L. Consumer preference study in grain utilization, Maiduguri, Nigeria. Ottawa, International Development Research Centre, IDRC-022e, 1974.

Steinkraus, K. H., Van Buren, J. P., La Belle, R. L., and Hand, D. B. Some studies on the production of pre-cooked dehydrated beans. Food Technology (Chicago), 18, 1964, 121.

Subba Rao, P. v., Ananthachar, T. K., and Desikachar, H. s. R. Effect of certain chemicals and pressure on cookability of pulses. Indian Journal of Technology (New Delhi), 2, 1964, 417.

Takeuchi, T. Low molecular weight peptide in soybean miso. Journal of Fermentation Technology (Osaka), 52, 1974, 256.

87

Tannous, R. I., Cowan, J. W., Rinnu, F., Asfour, R. J., and Sabry, z. I. Protein-rich food mixtures for feeding infants and pre-school children in the Middle East. American Journal of Clinical Nutrition (Bethesda), 17, 1965, 143.

Tolmasquim, E., Correa, A. M. N., and Tolmasquim, s. T. New starches. Properties of five varieties of cowpea starch. Cereal Chemistry (St. Paul, Minn.), 48, 1971, 132.

United Nations. Preliminary assessment of the world. food situation present and future. Rome, United Nations, Preparatory Committee of the World Food Conference, 2nd Session, Apr 1974.

Weisberg, s. Developing and marketing low-cost protein foods in developing countries. Food Technology (Chicago), 26, 1972, 60.

Wolf, w. J. Soybean proteins: their function, chemical and physical properties. Journal of Agricultural Food Chemistry (Washington), 18, 1970, 969.

World Health Organiztion. Energy and protein requirements: report of a joint FAO/WHO ad hoc expert committee. Geneva, WHO Technical Report Series No. 522, FAO Nutrition Meetings Report Series No. 52, 1973.

88