Contents - Previous - Next

This is the old United Nations University website. Visit the new site at

Studies on reduction in viscosity of thick rice gruels with small quantities of an amylase-rich cereal malt

Tara Gopaldas. Pallavi Mehta, Asha Patil, and Hemangini Gandhi
Maharaja Sayajirao University, Gujarat, India


In developing countries of Asia, including India, rice is widely used in gruel or cooked form as a first or early introductory food for infants and toddlers [1,2]. The starch granules in rice flour or rice grains not only swell on cooking but show a great propensity for taking up water (water-holding and binding). This contributes to the typical dietary bulk of a rice gruel with low caloric density per unit volume of food consumed. Older infants (over 6 months of age) and young toddlers (13 to 36 months of age) just cannot ingest sufficient amounts of such preparations to fulfil their daily energy requirements [3-6] . The formulation of high-energy (nutrient density) and low-bulk foods for young children utilizing simple and traditional processing methods, such as germination and malting, was recently proposed [7-11].

A series of studies conducted in our laboratory on malted and roasted mixes clearly indicated that the viscosity of a 10 per cent hot-paste slurry made of a mixture of individually malted and powdered wheat, bengal gram (Cicer arietinum), and groundnut (Arachis hypogaea), in the proportion 4:1:1, was significantly lower than that of the mixture in which each food ingredient was roasted [8,11, 12]. In a mixture consisting of cereals and legumes, the malted cereal constituent contributes substantially to the reduction in viscosity of a hot paste [9,13]. Although fully malted ready-to-mix (RTM) or ready-to-eat (RTE) mixes had decidedly higher nutrient density per unit volume of food than the roasted variety, the major and overriding constraints were the time, labour, and space required for their production. These, we felt, would make it too inappropriate and unfeasible a procedure to be readily adopted by rural, tribal, or slumd-welling mothers.

Because of its high amylase content, 5 per cent malted barley (Hordeum vulgare) flour substantially reduces the viscosity of 15 per cent hot-paste slurries of weaning foods on the Indian market, such as Nestum, Farex, Cerelac, and Balamul [9] . We therefore hypothesized that a catalytic or small amount of any amylase-rich food (ARF) should be able to decrease the viscosity of a traditional weaning rice gruel, If this could be substantiated, it would be a useful and simple way to reduce the dietary bulk of starchy gruels.

In the Gujarat region, barley is unknown. Among cereal grains, malted ragi (Eleusine coracana), bajra (Pennisetum typhoideum), sorghum (Sorghum vulgare), and maize (Zeas mays) flours have high amylase activity [9]. Since ragi (finger millet) was also unknown in this region and since bajra is a commonly consumed staple in the low-income groups, it was decided to develop an ARF from bajra and study its bulk-reducing properties on traditional weaning foods such as gruels prepared from rice flour, or flaked or puffed rice. The specific objectives of this study were: (i) to develop and standardize the method of preparation of ARF and to establish its organoleptic acceptability; (ii) to study the shelf life of the ARF; (iii) to compare the amylase activity and the viscosity-reducing properties of the ARF against those of a pure enzyme, namely, takadiastase; and (iv) to evaluate the acceptability of the rice gruels with and without ARF among mothers, and the intake of the same by infants.


Preparation of the ARF

The cleaned and picked grains of bajra (a variant locally known as maharashtra bajri) were steeped in an equal volume of tap water at room temperature for two hours. The grains were drained of water, wrapped in moist muslin cloth, and germinated for 72 hours. The well-germinated grains and shoots (5-6 cm) were spread on filter paper to remove surface moisture and were air dried for 12 hours under a ceiling fan until dry to the touch. The dried grains were roasted a small amount (100 9) at a time for about five minutes in a typical iron concave pan used for the purpose until the characteristic aroma of malting was smelled and the grains turned brown. The malted grains (both seeds and shoots) were ground to a fine powder in a local flour mill. The malt flour was sieved several times through a 50-mesh BSS sieve. This flour constituted the ARF.

TABLE 1. Comparative reduction in viscosity of 10 per cent hot-paste rice gruels by graded and equivalent amounts of ARF and takadiastase

Amount (9) of ARFa
or Tb added to 20g
rice flour in
200 ml water

Raw rice

Raw rice

Roasted rice

Roasted rice
flour (cheap)

Puffed rice flour

Flaked rice flour

0 g ARF/T
20 g rice flour







0.2 g ARF/T

+20 g rice flour
1,086 325 1,672 408 1,310 444 1,622 291 156 27 204 14
0.4 g ARF/T +
20 g rice flour
632 202 830 226 1,008 280 1,140 250 117 15 174 8
0.6 g ARF/T +
20 g rice flour
544 110 612 123 744 120 387 103 103 0 136 0
0.8 g ARF/T +
20 g rice flour
312 40 478 38 666 22 340 15 89 0 80 0

TABLE 2. The effect of different levels of ARF on a 25% hot-paste slurry made from raw rice flour

  Amount of rice flour (g) in 200 ml water Amount of
ARF added (g)
reading (CPs)
1 50 0 37,200
2 46 4 12,400
3 42 8 6,800
4 40 10 6,400
5 38 12 6,400
6 34 16 6,800
7 30 20 6,400
8 26 24 6,800

Organoleptic Testing of the ARF

Fifteen panelists were first screened using the threshold test, which was defined as the ability of the panelists consistently to identify the lowest concentration of four taste solutions: sweet, salt, sour, and bitter. The ten who qualified evaluated the ARF using the hedonic rating test for the three organoleptic attributes of taste, smell, and texture [14].

Evaluation of the Keeping Quality of the ARF

The shelf-life trials were carried out in November and December 1984 during the relatively cool and dry season with ambient temperatures in the range of 20.1C to 27.4C (mean 23.7C 0.34C). Relative humidity ranged from 27 to 89 per cent (mean 59.0 per cent 2.45 per cent). The ARF was freshly prepared, immediately distributed in nine lots of 150 g each, and packed in 200-gauge polythene bags, which were heat sealed. Lots were drawn on days 0, 7, 14, 21, and 28, and were analysed for moisture [15], free sugar [16], peroxide value [17], and bacteriological plate count [18].

Estimation of Amylase Activity

The method of Bernfield [19] was adopted to estimate amylase activity. The milligrams of maltose released by 1 g of malt or takadiastase at 37C when acted on by 1 ml of 1 per cent starch substrate for 30 minutes was recorded.

Measurement of the Thinning Effect of Equivalent Amounts of ARF and Takadiastase

The thinning or liquifying power of the ARF was compared with that of equivalent amounts of an amylase source, namely takadiastase (Taka-Amylase Fungal Origin 'Fluke' AG, manufactured by Chem. Fabrik CH-9470 Buchs Laboratory grade), on six varieties of rice flour: (i) cheap variety of raw rice flour (3 rupees/kg rice); (ii) the same in roasted form; (iii) an expensive variety of rice flour (4.50 rupees/kg rice); (iv) the same in roasted form; (v) powdered puffed rice, or murmurs (6 rupees/kg); and (vi) powdered flaked rice, or pave (5 rupees/kg).

Viscosity reduction was measured in two experiments at two levels of hot-paste slurry, that is, 10 per cent and 25 per cent. The composition of the two slurries with and without ARF or takadiastase is given in tables 1 and 2.

Viscosity of 10 Per Cent Cold-paste Slurry

A 10 per cent cold-paste slurry was obtained by making up the volume to 200 ml with water. The slurry was heated in a boiling water bath for 10 minutes with occasional stirring.

It was then heated over an open flame (over a wire gauze) for three minutes with continuous stirring. The slurry was cooled to 46C to 45C. Five continuous readings were taken at this temperature range, with spindle no. 3 at 30 rpm, using a factor finder of 40 on a Brookfield Viscometer (LVT model no. 50).

Viscosity of 25 Per Cent Hot-paste Slurry

In the second experiment only the flour from cheap rice (3 rupees/kg) was used. The intention was to obtain the thickest possible hot-paste slurry that would register a reading on the Viscometer (an RVT model of the Brookfield Viscometer using spindle no. 7, at 50 rpm, with a factor finder of 800). This was a 25 per cent hot-paste rice slurry. The amounts of rice flour and ARF were adjusted to give a final weight of solids of 50 g (table 2). The volume was increased to 200 ml with water and the hot-paste slurry was prepared and read on the Viscometer as described above.

Acceptability and Intake Trials

Preparation of the Rice Gruel in Bulk

Thirty servings of about 100 ml each, for a 10 per cent hot-paste slurry. Ingredients: 300 g rice flour, 12 g ARF, 120 g jaggery (unrefined brown sugar), 60 g oil, and 2.7 litres water. The rice flour was browned in oil. Two litres of water plus ARF were stirred in and the gruel was cooked while being constantly stirred. The grated jaggery in 700 ml water was then added and the gruel was cooked for a few more minutes. No ARF was added to the control gruel. The caloric density was about 75 kcal per 100 ml.

Thirty servings of about 100 ml each, for a 25 per cent hot-paste slurry. Ingredients: 630 g rice flour, 120 g ARF, 300 g jaggery, 150 g oil, and 2.5 litres water. The procedure was the same. The caloric value per 100 ml was about 185 kcal. In the case of control gruels, 750 g rice flour, 300 g jaggery, and 150 g oil were incorporated but no ARF was added.

Acceptability Trials among Mothers

Since the mothers were illiterate or had had only a little schooling, a simple opinion scale of "like," "dislike," and "can't say" was used to rate gruels with and without ARF (10 and 25 per cent hot-paste slurries).

Intake Trials among Infants

Thirty infants pair-matched for age were randomly assigned to a control or experimental group. In a switchover design, each subject tasted the 10 per cent rice gruel with or without ARF for three days at approximately the same time (two hours before a feeding). First, 100 ml of gruel was offered, followed by 50 ml if the child wished it. Net intake per day per subject was recorded. The infant was then switched for the next three days to the other type of gruel. The same procedure was followed with the 25 per cent gruel.


Preparation Time, Cost, and Organoleptic Acceptability

The total time taken to prepare the ARF was 86 hours, which was longer than that necessary to prepare a fully malted RTE [8,11] . In the case of the ARF, 72 hours were required for germination for maximum elaboration of amylase content in the seed. Possibly the time taken for air drying (12 hours) could be reduced by adopting sun drying or using a solar drier. However, the major advantage in the preparation of the ARF versus fully malted foods was that, relatively speaking, greatly reduced amounts of space and cooking fuel were required. Furthermore, the utensils commonly found in any home were adequate to prepare a batch of about 500 g of ARF. This was all that was required for the optimum thinning of rice gruel for two children below three years of age every day for 20 days. Since the batch quantity of ARF was small, the cost involved for the raw ingredient, whole bajra grains, was only 1 rupee (about US$.08). The fuel and milling costs in the preparation of ARF could be estimated at another 50 paise (US$.04). The comparative costs of 500 g each of takadiastase and ARF were 2,500 and 1.50 rupees respectively. Even allowing for the fact that takadiastase was about seven times as active as the ARF (see below), the cost of the ARF was about 250 times lower. The results of the organoleptic rating of the ARF on day zero (fresh batch) and day 21 are presented in table 3.

The results show that the ARF was acceptable even on day 21 after preparation, although its organoleptic rating was lower than on the day of its preparation, Our organoleptic rating results do not support those of other investigators [9,20], who reported that malt from bajra had an unpleasant smell and a very bitter taste after a few days of storage, due to its high lipase content. it is possible that the short steeping time (2 hours) used in the present study against a much longer steeping period of 16 hours employed elsewhere [9] contributed to acceptable organoleptic characteristics even on the twenty-first day of storage.

Shelf Life of the ARF

The shelf life of the ARF was evaluated against Indian standard (IS) specifications for processed cereal weaning mixes, as no specifications were available for malted powders. The results (table 4) demonstrated that ARF had a shelf life of about 21 days at a mean room temperature of approximately 25C and relative humidity of 60.

On the twenty-eighth day the ARF was within all IS specifications tested, except for the viable plate count that exceeded the limit of 50,000 per g of dry malt powder.

Amylase Activity of ARF and Takadiastase

The amylase activity of 1 g of ARF was 100 maltose units; that of 1 g takadiastase was 660 maltose units. Therefore, takadiastase was 6.6 times as active as ARF. Amylase activity varies with the base cereal used, its cultivar, and method of preparing the malt. However, some cereal grains such as ragi, bajra, and sorghum (also called jowar), have been reported to have the highest amylase activity after 72 hours of germination, namely 178, 125, and 158 maltose units respectively [9]. Such cereal grains offer good potential for ARF preparation

Reduction in Viscosity of Hot-paste Gruels by ARF and Takadiastase

Ten Per Cent Hot-paste Slurry

The relative effect of graded but equivalent amounts of ARF and takadiastase on 10 per cent slurries prepared from raw, roasted, puffed, or flaked rice is presented in table 1.

The data clearly reveal that initially viscous hot-paste slurries (10 per cent) were greatly reduced in viscosity or liquified when ARF was added at the level of 4 g per cent. The maximum thinning took place within the first five minutes of heating the slurry over boiling water. It took just 1 g per cent of takadiastase to result in thinning equivalent or superior to that achieved with 4 g per cent AR F. Powdered, puffed, or flaked rice already had relatively low hot-paste viscosity, and this was practically brought down to the viscosity of water with the addition of 4 g per cent ARF. On cooling to room temperature, from 46C to about 25 C, all the hot-paste slurries (with or without ARF or takadiastase) tended to thicken slightly. However, a hot-paste rice gruel with 4 g per cent ARF was visibly much thinner than its counterpart without ARF.

Twenty-five Per Cent Hot-paste Slurry

Although the addition of 4 g per cent ARF to a 10 per cent hot-paste slurry made from raw rice flour was encouraging and striking, it was realized that, even if a young child was able to consume 200 ml of the gruel alone at a sitting, this would contribute only 80 kcal energy. Thus, attempts were made to study maximum reduction in the viscosity of the thickest gruel (25 per cent) with the addition of proportionately higher levels of ARF (table 2). A 25 per cent hot-paste rice flour gruel without ARF made a very stiff and scarcely stirrable gruel; it was like a slippery dough. With the addition of varying amounts of ARF, some noticeable reduction in viscosity was observed (table 2).

TABLE 3. Organoleptic evaluation of the ARF on day 0 (fresh batch) and on day 21

Mean weighted scores obtained on Maximum
Percentage of
Day 0 Day 21 Day 0 Day 21
Taste 198 162 250 150 79 65
Smell 160 121 200 120 80 61
Texture 119 120 150 90 79 80
Totals 477 403 600 360 80 67

a. Since the product has a shelf life of 21 days (see table 4), organoleptic testing was stopped on day 21 of storage.

TABLE 4. Shelf life of the ARF in terms of moisture, peroxide value, free sugar, and bacterial plate count

Day of
Peroxide value
(meq/kg oil)
Free sugar content
Bacterial plate count (per g
malt powder)
0 3.47 0.95 10.31 29,166
7 5.72 2.76 8.23 37,500
14 6.75 3.56 6.18 44,166
21 7.43 3.99 5.16 46,500
28 8.10 6.27 4.78 57,500
IS specification     No IS  
10.00 10.00 specification 50,000

TABLE 5. Opinion of mothers of infants and toddlers on a 10 per cent hot-paste rice gruel with and without ARF(4 g)

Opinion Gruel with ARF Gruel without ARF
Mothers of infants (n = 30)
Like 24 (80)a 18(60)
Dislike 6 (201 12(46)
Can't say
Mothers of toddlers (n = 30)
Like 18 (60) 41(31
Dislike 7 (23) 23(771
Can't say 5(17) 3(10)

a. Percentages are given in parentheses.

TABLE 6. Mean intake of 10 per cent hot-paste rice gruel with or without ARF by 30 infants (6-12 months of age)

Type of gruel Mean intake (ml) t value Significance
With ARF 107.61.10a 1.743 Significant
at P<0.05 level
Without ARF 55.70.94

a. Mean and SE.

TABLE 7. Mean intake of 25 per cent hot-paste rice gruel with or without ARF by 30 infants (6-12 months of age)

Type of gruel Mean intake (ml) t value Significance
With ARF 947.17a 1.741 Significant
at P<0.05 level
Without ARF 565.03

a. Mean and SE.

The maximum reduction in viscosity of the 25 per cent hot-paste slurry with the minimum of ARF was at the level of 8 g fresh ARF plus 42 g rice flour. Adding higher amounts of ARF did not bring about any further reduction in viscosity. However, even at this optimum level of ARF, the gruel was thick though stirrable; but it was not thin and free-flowing, as in the case of the 10 per cent hot-paste slurry with 4 g per cent of ARF.

Acceptability and Intake Trials of 10 Per Cent Hot-paste Rice Gruels with and without ARF with Mother-Child Dyads

Thirty mothers of infants 6 to 12 months of age and 30 with toddlers 12 to 36 months of age were asked to evaluate a 10 per cent hot-paste rice gruel with ARF (4 g per cent) and one without ARF. The gruel with ARF was preferred by the large majority of mothers because it was smooth and fairly free-flowing, while the one without ARF was lumpy.

Thirty infants took part in a six-day intake trial, where each subject tested the rice gruel either with or without the ARF for three days each, in a switchover design (table 6). The trial was repeated with the 25 per cent hot-paste slurry after a break of one week.

The paired t test indicated that infants had a significantly higher intake of the gruel with ARF, the mean intake being almost double that of gruels without ARF. Out of 30 subjects, only one infant suffered from diarrhoea, but continued the test. No other ill effects were reported by the mothers.

In a second series of intake trials, infants tested a 25 per cent hot-paste rice gruel with and without ARF (table 7). The same preference was expressed in favour of the gruel with ARF.


During recent years, extensive efforts have been made by Indian and Swedish investigators to prepare low-cost, nutrient-dense, low-bulk weaning and toddler foods from locally available food commodities, utilizing culturally acceptable and traditional home-processing methods [7-13]. In this regard, germination of whole cereal grains and subsequently malting (toasting) the grains has emerged as a most promising method for greatly reducing the dietary bulk of customary high-carbohydrate weaning diets in the developing world. However, developing fully malted foods involves time, labour, and, above all, a great deal of space, which may not be readily available in most rural and slum homes.

This study made a breakthrough with respect to the overriding constraint of space, in that it has provided data to show that small quantities of any amylase-rich food can be used effectively to reduce the dietary bulk or viscosity of traditional weaning thick rice gruels. Although no short cuts are possible with regard to mode or time of preparation of the ARF (which is nothing but a malt prepared from a cereal grain), the advantage of the ARF over and above fully malted foods is the fact that only very small, if not catalytic, amounts are required to liquify dramatically a thick and viscous gruel. Thus, the amount of ARF required can easily be prepared and stored for a reasonable length of time (15 to 20 days) in the household or day-care centre, or even at the retail level.

Organoleptic acceptability and shelf-life trials have shown that the ARF developed in this study from a cheap millet of the region, namely bajra, was highly acceptable for 21 days.

The most important factor in the development of an ARF would be the selection of the most widely consumed staple of the region for its production. In short, any whole grain with its germ intact that is capable of being germinated would be a potential candidate for ARF production. Hence it could be a wheat ARF for wheat, a sorghum ARF for sorghum, a maize ARF for maize, and so on. Although cereal grains have much higher amylase activity than pulses [9], it may be necessary to turn to a pulse when only a single staple of importance is consumed by the masses, such as rice in Burma, Bangladesh, Thailand, etc. Hence mung bean (Phaseolus aureus Roxb), which is fairly widely used in these regions, especially in the form of bean sprouts, could be easily exploited for its ARF function.

The Swedish group [3] extensively studied the hot-paste characteristics of some blended supplementary weaning foods such as SEF (Sweden), faffa (Ethiopia), sekmama (Turkey), superamine (Egypt), and lisha (Tanzania). What is of interest is that both sekmama and superamine are weaning foods treated with amylase. Lisha is an extruded product making use of 65 per cent germinated white corn. Brabender amyloviscographs reveal the extraordinary liquifying effect of amylase-treated sekmama and superamine on their hot paste, in contrast to non-amylase treated faffa and SEF that have high hot-paste viscosity at 40C. The potential of pre-treating weaning foods with amylase in appropriate quantities is of great significance in a commercial setting, but does not lend itself to a simple and practical application at the household level. We feel that the concept of the ARF is more suited to the liquification of gruels in the home.

In conclusion, the ease of preparation of ARF, its acceptable shelf life, organoleptic rating, and extremely low cost in relation to pure amylase make it a quick way to thin traditionally thick rice flour gruels at the household level.


1. Indian Council of Medical Research, "Studies on Weaning and Supplementary Foods," ICMR Technical Report Series, 27 (New Delhi, 1977).

2. D. B. Jelliffe, "Infant Nutrition in the Tropics and Subtropics," WHO Monograph Series, 29 (WHO, Geneva, 1968).

3. A. Hellstrom, A. Hermansson, A. Karison, B. Ljungqvist, O. Mellander, and U. Swanberg, "Dietary Bulk as a Limiting Factor for Nutrient Intake with Reference to the Feeding of Preschool Children. II, Consistency as Related to Dietary Bulk—A Model Study," J. Trop, Pediatr., 27: 127-135 (1981).

4. M. Cameron and Y. Hofvander, Manual on Feeding of Infants and Young Children (Oxford University Press, Delhi/Nairobi, 1983), pp. 117-127.

5. I. H. E. Rutishauser, "Factors Affecting the Intake of Energy and Protein by Ugandan Preschool Children," Ecol. Food Nutr., 3: 213-222 (1974).

6. B. M. Nicol, "Protein and Calorie Concentration," Nutr. Rev., 29: 83-88 (1971).

7. H. S. R. Desikachar, "Development of Weaning Foods with High Calorie Density and Low Hot Paste Viscosity Using Traditional Technologies," Food Nutr. Pull., 2: 21-23 (1981).

8. T. Gopaldas, F. Inamdar, and J. Patel, "Malted versus Roasted Young Child Mixes: Viscosity, Storage and Acceptability Trials," Indian J. Nutr. Diet., 19: 327-336 (1982).

9. N. G. Malleshi and H. S. R. Desikachar, "Formulation of a Weaning Food with Low Hot Paste Viscosity Based on Malted Ragi (Eleusine coracana) and Green Gram (Phaseolus radiatus)," J. Food Sci, Tech., 19: 193-197(1982).

10. B. Brandtzaeg, N. G. Malleshi, U. Swanberg, H. S. R. Desikachar, and O. Mellander, "Dietary Bulk as Limiting Factor for Nutrient Intake with Special Reference to the Feeding of Preschool Children. III. Studies on Malted Flours from Ragi, Sorghum and Green Gram," J. Trop. Pediatr., 27: 184-189 (1981).

11. T. Gopaldas, "Malted versus Roasted Weaning Mixes: Development, Storage, Acceptability and Growth Trials," in K, T. Achaya, ed., Interfaces between Agriculture, Nutrition and Food Science (United Nations University, Tokyo, 1984), pp. 293-306.

12. K. M. Tajuddin, "Acceptability and Nutritional Status Trials on Preschool Children (1-5 Years) with Malted versus Roasted RTE Mixes at Baroda and Pondicherry," M.Sc. thesis (M.S. University, Baroda, 1980).

13. V. R, Nayak, "Studies on Roasted, Malted and Combinations of Roasted and Malted RTE Mixes with Respect to Viscosity, Recipe Formulation, Acceptability and Shelf Quality," M.Sc. thesis (M.S. University, Baroda, 1983).

14. M. A. Amerine, R. M. Pangborn, and E. S. Roessler, Principles of Sensory Evaluation of Foods (Academic Press, New York, 1965), pp. 367-374.

15. M. B. Jacobs, The Chemical Analysis of Foods and Food Products, 2nd ea., (D. Van Nostrand Co., New York, 1951), pp. 504-505.

16. B. L. Oser, Hawk's Physiological Chemistry, 14th ed., (Tata McGraw, New Delhi, 1971), p. 441.

17. M. B, Jacobs, The Chemical Analysis of Foods and Food Products, 2nd ed. (D. Van Nostrand Co., New York, 1951), p. 393.

18. T. J. Mackie and J. E. McCartney, "Viable Count," in R. Cruickshank, ed., Handbook of Bacteriology (E. and S, Livingston, Edinburgh, 1956), pp. 304-305.

19. S. P. Colowick and N. O. Kaplan, eds., Methods in Enzymology, vol. 1 (Academic Press, New York, 1955), p. 149.

Contents - Previous - Next