Study of some factors affecting water absorption by amaranth grain ...

Journal of Food Engineering 60 (2003) 391–396 www.elsevier.com/locate/jfoodeng

Study of some factors affecting water absorption by amaranth grain during soaking Andrea N. Calzetta Resio, Roberto J. Aguerre, Constantino Su arez

*

Departamento de Industrias Tecnologia de Alimentos, Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires, Ciudad Universitaria, Nunez, Buenos Aires 1428, Argentina Received 23 September 2002; accepted 8 February 2003

Abstract Water absorption by amaranth grain in plain water was determined at 30, 40, 50 and 60 °C by recording the weight increase in grain with respect to time. Water absorption kinetics at these temperatures was described by the FickÕs second law solution for diffusion out of sphere. Effective diffusion coefficients varied between 2.63 and 8.25 1012 m2 /s for the range investigated. The activation energy for diffusion obtained was 32.1 kJ mol1 . The rates of absorption in SO2 aqueous solution (0.01%, 0.02% and 0.03%, v/v) were slightly higher than in plain water at 40 and 60 °C. However, the increase in SO2 concentration did not seem to increase significantly water absorption. Amaranth grain soaked in 0.02% (v/v) SO2 aqueous solution and variable lactic acid concentrations (0.0025% and 0.0050%, v/v) was performed at 40 °C. Absorption rates for lactic acid concentrations were 9.1% higher than those steeped only in SO2 aqueous solution; lactic acid concentration had no effect on the absorption rate. The amount of total solids leached in sulfur dioxide solution with lactic acid (200 ppm SO2 + 0.005% lactic acid) was the highest when compared with the other steeping media investigated. Ó 2003 Elsevier Ltd. All rights reserved. Keywords: Diffusion; Moisture; Saturation; Sulfur dioxide; Lactic acid; Solids

1. Introduction Different factors have contributed to increase the interest of grain amaranth as food and feed source. Among them, the composition and properties of this crop have received special attention (Teutonico & Knorr, 1985). The main component of the grain is starch, about 48– 60% (Uriyapongson & Rayas-Duarte, 1994), which is considered to have promising applications in food and non-food areas. Another components of the grain are lysine rich protein, 12–18%, and 5–8% fat that contain high level of squalene (Myers & Fox, 1994). In the past two decades substantial investigation was performed on the functional and physicochemical properties of the main components of grain amaranth, starch and protein (Calzetta Resio, Tolaba, & Suarez, 2000; Paredes-L opez, Schevenin, Hern andez-L opez, & Carabez-Trejo, 1989; Yanez, Messinger, Walker, & Rupnow, 1986; Zhao & Whistler, 1994). Despite the

*

Corresponding author. Tel./fax: +54-11-4576-3366. E-mail address: [email protected] (C. Suarez).

0260-8774/$ - see front matter Ó 2003 Elsevier Ltd. All rights reserved. doi:10.1016/S0260-8774(03)00062-1

substantial interest on both products, separation of amaranth starch from protein is somewhat difficult because of the close bounding between the two components. Currently, there is no commercial process for recovering starch from amaranth (Myers & Fox, 1994). Conventional methods such as pearling (Betschart, Irving, Shephard, & Saunders, 1981) and stone-milling (Becker, Irving, & Saunders, 1986) resulted effective in separating amaranth protein and starch fractions. Myers and Fox (1994) used a modified laboratory method for the wet milling of amaranth to recover starch from it. However, such as pointed out by Lehmann (1996), if the ring embryo conveys novel benefits in amaranth nutrition, then further wet milling techniques need to be explored. In the wet-milling process, grains are almost invariably steeped in water at temperatures below gelatinization temperatures in the presence of chemical agents. In the wet-milling of corn, the grains are steeped in aqueous solution of SO2 to facilitate the isolation of starch. The presence of lactic acid in the steeping solution was investigated by Du, Li, Lopez-Filho, Daniels, and Eckhoff (1996), who found an increase of starch yield in

392

A.N. Calzetta Resio et al. / Journal of Food Engineering 60 (2003) 391–396

comparison with the conventional corn wet-milling procedure. According to Haros and Suarez (1999) the presence of lactic acid in steep water, not only increases the rate of water absorption by corn kernels, reducing so the effective steeping time, but also improves starchprotein separation. As we are interested to determine whether a wetmilling process with separation procedures similar to those used in the wet-milling of corn could be adapted for amaranth, a preliminary study was conducted to investigate the kinetics of water absorption by amaranth grain. As it is well recognized in various investigations concerning with the wet-milling of grains, hydration is one the most important factors in producing goodquality products (Haros, Viollaz, & Suarez, 1995; Norris & Rooney, 1970; Yuan, Chung, Seib, & Wang, 1998). Many works studied the quantitative analysis of water absorption for cereal grains (Becker, 1960; Fan, Chun, & Shellenberger, 1961; Tolaba, Viollaz, & Suarez, 1990), among others. In those investigations, FickÕs law of diffusion with constant diffusion coefficient was used without introducing too much error in the results. Other investigations, however, have shown that the diffusion equation with constant diffusivity is inadequate in describing the water absorption curve (Hsu, 1983). Based on the present considerations the absorption characteristics of amaranth grain in plain water and various temperatures were investigated. Also, the effect of certain chemical agents, SO2 and lactic acid, on the water absorption was investigated as well as a means to predict, quantitatively, the absorption kinetics in amaranth grain.

2. Materials and methods 2.1. Materials Mature grains of Amaranthus cruentus were harvested in the experimental cultivars of Facultad de Agronomıa, Santa Rosa, and La Pampa. After harvesting, the grains were cleaned from foreign materials by screening and stored at room temperature in sealed containers until used. The characteristics of the grain investigated are summarized in Table 1. Moisture content, fat, ash and protein were determined following standard AOAC methods 943.01, 920.39. 923.03 and 976.05, respectively Table 1 Characteristics of Amaranthus cruentus Moisture content (% db) Protein, (N 6.25)% Fat (%) Starch (%) Ash (%) Average diameter (mm)

10.5 16.8 7.7 73.0 3.1 0.9

(AOAC, 1995). Starch content was assessed by AACC method 76–11 (AACC, 1995). The average diameter of grains was evaluated from a size distribution analysis obtained by using three metal sieves with pore diameters of 870, 920 and 1000 lm. The weighed average diameter, d, was calculated from the expression: X d¼ ðdi qi Þ=q ð1Þ i

where di is the average screen opening, qi is the weight between (i 1)th P and ith sieve starting from the largest one, and q ¼ i qi is the total sample weight. 2.2. Experimental procedure 2.2.1. Effect of temperature Water absorption data were obtained by placing 10 0.5 g of amaranth grains in 150 ml screw-cap flasks containing distilled water. The flasks were placed in constant-temperature water bath controlled within 0.5 °C of the testing temperature. At regular intervals the flasks were removed from the bath for moisture content determination. For this purpose, the grains were rapidly removed from the flasks and superficially dried on a large filter paper to eliminate the surface water. The grains were then weighed to determine the moisture uptake. Experiments were conducted at 30, 40, 50 and 60 °C and for immersion periods from several minutes to about 3 h. 2.2.2. Effect of sulfur dioxide on soaking To study the effect of sulfur dioxide on water absorption three solutions containing 0.01%, 0.02% and 0.03% of the chemical, by volume, were prepared by dissolving the appropriate amounts of NaHSO3 in distilled water. The grain samples were placed in screw-cap flasks to prevent escape of SO2 gas. The flasks were immersed in a thermostatic water bath at 40 and 60 °C for water absorption determinations. 2.2.3. Effect of sulfur dioxide and lactic acid on soaking Soaking solutions containing 0.02%, v/v, SO2 and variable concentration of lactic acid: 0.0025% and 0.005% by volume were prepared to evaluate the effect of both agents on water absorption. The soaking temperature tested for such purpose was 40 °C. 2.3. Total solids determination The mass of solids leached during water steeping in plain water, in aqueous solutions of 0.02%, v/v, SO2 and 0.02% SO2 + 0.0025% lactic acid, by volume, was determined by the method of Steinke and Johnson (1991). A 10 ml sample of the steeping water was placed in a weighed aluminum flask and air dried in an oven at 65 °C for 24 h. After that, the sample was fully dried at the


same temperature in a vacuum oven until constant weight, in presence of P2 O5 as desiccant. The percent of total solids was referred to the weight of fully dried kernels.

3. Results and discussion In order to predict water absorption during soaking, the second FickÕs law solution for diffusion out of sphere was tested. For this purpose the following assumptions were made: (i) the effective diffusion coefficient, De , is independent of moisture concentration; (ii) the volume of the grain does not change during absorption, (iii) the surface of the grain reaches moisture saturation instantaneously upon immersion in absorption media. Based on these considerations, the solution of FickÕs second law (Cussler, 1984) was used to calculate diffusivity constants: 1 w w1 6 X 1 De n2 p2 t ¼ 2 exp ð2Þ R2 w0 w1 p n¼1 n2 Where w is the moisture content at a given time t, w0 the initial moisture content, w1 the saturation moisture concentration, and R the average radius of amaranth grain. To evaluate w1 the following method was used; when time becomes large, the limiting form of Eq. (2) becomes: w w1 6 De p2 t ¼ exp 2 ð3Þ R w0 w1 p2 From this equation it results that for any set of three moisture contents taken at equally spaced time intervals of duration j, the following expression for w1 results: w1 ¼

wi wiþ2j w2iþj wi þ wiþ2j 2wiþj

393

uration moisture concentration calculated from Eq. (4). The maximum relative error in saturation values was 2.5% while the corresponding to De was almost 4.2%. The predicted water absorption curves of amaranth in plain water are shown in Fig. 1 together with the experimental data for the four soaking temperatures tested. The moisture content of amaranth grains increased rapidly during the initial stages of hydration. This may be due to characteristics of structure of the outermost layers; i.e. the seed coat and pericarp, which being relatively porous appear to quickly attain equilibrium with the soaking water by capillary imbibition (Becker, 1960). Another significant factor that contributes to the rapid initial water absorption may be the moisture absorbed in the void space between the husk and the kernel of the grain. The major discrepancies between experimental and predicted curves are evident at the highest moisture concentrations, because hydration occurs at much higher rate than predicted by Eq. (2). Such inadequacy, also observed by some investigators during water absorption process in grains and seeds (Abdel Kader, 1995; Hsu, 1983; Sopade & Obepka, 1990), might be attributed to certain dependence of the diffusion coefficient on moisture concentration.The comparison of the diffusion coefficients for water steeping obtained in this work for amaranth grain with those reported in the literature for other grains is shown in Table 3. It can be concluded that amaranth grain had diffusion coefficients only comparable to that of maize grain. In order to evaluate the goodness of the fit of Eq. (2) to the experimental water absorption, the root mean square error, RMSE, defined as: vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi u 2 N u1 X we wp RMSE ð%Þ ¼ 100 t ð5Þ N 1 we

ð4Þ

wi and wj being the moisture concentrations at time i and j. Eq. (2) was used to estimate the effective diffusion coefficient during water absorption of amaranth grain in plain water by programming this equation on a digital computer. A non-linear regression procedure was used to calculate De ; diffusivity determinations were replicated three times. The average value of these replications is given in Table 2 together with the values of the sat-

was calculated where we and wp are, respectively, the experimental and predicted moisture concentrations and N the number of experimental data. The values of RMSE (%) are reported in Table 2. If values of RMSE (%) less than 5% are generally considered as a good fit (Lomauro, Bakshi, & Labuza, 1985), the relative high value of this parameter indicates some degree of inadequacy of the model to describe the absorption process as a whole.

Table 2 Saturation moisture contents, w1 , effective diffusion coefficient, De , and RMSE (%) values for water absorption of amaranth grain in plain water Temperature (°C)

w1 (kg water/kg dry solid)a

De 1012 (m2 /s)b

RMSE (%)

30 40 50 60

0.715 0.015 0.810 0.013 0.889 0.017 0.96 0.020

2.63 0.11 3.84 0.08 5.81 0.21 8.25 0.10

4.91 3.89 5.15 6.01

a b

Mean values standard deviation (n ¼ 5). Significant at P < 0:05. Mean values standard deviation (n ¼ 22). Significant at P < 0:05.

394


for the different grains. For example, Becker (1960) found for wheat that the saturation values resulted practically independent of the soaking temperature and equal to a moisture concentration of 75% in dry basis. A similar trend was found by Haros et al. (1995) during the steeping of maize in aqueous solution of sulfur dioxide, with a saturation value of 79% for soaking temperatures ranging from 45 to 65 °C. At this point of the analysis it may be interesting to compare the activation energy resulting from the variation of the saturation moisture concentration with temperature, with the value of Ea obtained from the diffusive process. The activation energy for variation of the saturation moisture content with temperature was calculated to be 6.71 kJ mol1 . The comparison with the value of the diffusive process (32.1 kJ mol1 ) reveals a substantial difference between these values. This difference might suggest that soaking at higher steeping temperatures will affect the diffusion coefficient in the endosperm somewhat differently than in the seed coat. Considering that the seed coat is moisture saturated since the beginning of the absorption process, less variation can be expected for the value of the diffusion coefficient as temperature increases.

Fig. 1. Water absorption rates for amaranth grain in plain water. (M) 30 °C; (}) 40 °C; () 50 °C; (j) 60 °C; Eq. (2), full line.

The diffusion coefficients reported in Table 2 were correlated with the reciprocal of the absolute temperature, according to the Arrhenius type equation De ¼ D0 expðEa =RT Þ

3.1. Effect of SO2 on the absorption rate The effect of SO2 on the rate of water absorption compared with plain water is illustrated in Fig. 2 for the soaking temperature of 40 °C (similar effects not depicted in this figure were found at 60 °C). As can be seen from this figure, the difference in moisture increase between amaranth grains soaked in plain water and aqueous SO2 solution was not very marked, though significant. The effective diffusion coefficient of water into the kernel obtained for absorption in plain water at 40 °C was 3.84 1012 m2 s1 , while for SO2 (0.01%, v/v) was of 4.30 1012 m2 s1 for the same temperature. On the contrary, the variation of SO2 concentration in the soaking media did not cause a significant variation on the absorption rate: the diffusion coefficients at 0.01% and 0.03%, v/v, SO2 were, respectively, 4.30 1012 and 4.19 1012 m2 /s.

ð6Þ

where Ea is the activation energy, R the gas constant and T the absolute temperature. The value of Ea was calculated by linear regression of ln De versus 1=T being its value 32.1 kJ mol1 . This value is comparable to those reported in the literature for water absorption in some grains. So, Engels et al. (1986) found Ea values ranging from 22.5 to 64.51 kJ mol1 , depending on the moisture concentration of rice grain, while Haros et al. (1995) reported an energy value of 39.41 kJ mol1 for water absorption in corn activation. It is also observed from Table 2 that the saturation moisture content increases as the soaking temperature is also increased. Such behavior is not apparently the same

Table 3 Diffusion coefficients for steeping in plain water of amaranth grain compared with other grains Material Maize Wheat Rough rice

Soybean Amaranth

Temperature range (°C)

De (m2 /s) 11

52 40 40

4.86 10 3.63 1011 6.67 109

50 40 60 40 60

9.03 109 1.08 1010 2.00 1010 3.84 1012 8.25 1012

Reference Haros et al. (1995) Becker (1960) Engels, Hendrickx, De Samblanx, De Gryze, and Tobback (1986) Hsu (1983) This work


395

corn steeping and attributed to the reduction of disulfide bonds of corn proteins by SO2 . 3.2. Effect of lactic acid on water absorption

Fig. 2. Water absorption rates for amaranth grain in plain water (}) and SO2 solutions: (þ) 0.01% (v/v) and () 0.03% (v/v) at 40 °C.

The effect of SO2 on the variation of total solids in the steeping water, with time is shown in Fig. 3. It can be seen that the amount of solids released in the SO2 solution after a short induction period was considerably higher than in plain water, being this effect more marked with the increase of soaking time. Such difference could be due to the soluble action of SO2 on the endosperm proteins of amaranth grain. A similar behavior was observed by Neuman, Wall, and Walker (1984) during

The water absorption curves of amaranth in aqueous solution of 0.02%, v/v, SO2 and variable concentrations of lactic acid: 0.0025% and 0.005% (v/v) are shown in Fig. 4. The beneficial effect of lactic acid on water absorption rate became significant at concentrations of 0.0025% (v/v) being the diffusion coefficient 4.69 1012 m2 s1 , for the soaking temperature of 40 °C. However, this effect did not seem to increase when the concentration of lactic acid is twofold. The estimated diffusion coefficient at 40 °C for SO2 0.02% (v/v), combined with lactic acid, 0.005% (v/v), was 4.73 1012 m2 s1 . The effect of lactic acid on grain steeping is not well understood and the results available in the literature on water absorption rates are scarce and contradictory. For corn, Shandera, Packhurst, and Jackson (1995) found that the water absorption rate increased when lactic acid was added to the steeping media, whereas Haros and Suarez (1999) observed that the incorporation of lactic acid into the steeping solution slightly affected the rate of water absorption of that grain. The variation of total solids leached in steep water with time is illustrated in Fig. 3. A marked increase is observed in the amount of solids leached on adding lactic acid, compared to that found during soaking in plain water or in SO2 solution. This may be due to the softening action of lactic acid, which increases the porosity of cellular membranes, enhancing in turn, the

6

total solids (g/g dry grain)

5

4

3

2

1

0 0

50

100

150

200

250

time (min) Fig. 3. Variation with time of total solids leached at 40 °C in: plain water (M); 0.02% v/v, SO2 solution (); 0.02% v/v, SO2 solution + 0.005% v/v, lactic acid ().

Fig. 4. Water absorption curves at 40 °C of amaranth grain in 0.02% v/v, SO2 with 0.0025% v/v, lactic acid (M) and 0.005% v/v, lactic acid ().

396


dissolving action of SO2 on the proteinaceous matrix that which binds the starch granules.

Acknowledgements We acknowledge the financial support from Consejo Nacional de Investigaciones Cientificas y Tecnicas (CONICET) and Universidad de Buenos Aires. Special thanks to Ing Rosa Troiani and Facultad de Agronomia, Universidad Nacional de La Pampa, for providing grain amaranth samples. References Abdel Kader, Z. M. (1995). Study of some factors affecting water absorption by fava beans during soaking. Food Chemistry, 53, 235– 238. American Association of Cereal Chemists (AACC) (1995). Approved methods of the association of cereal chemists (9th ed.). St. Paul, MN, USA: The Association. Association of Official Analytical Chemists (AOAC) (1995). Official methods of analysis (16th ed.). Washington, DC: AOAC. Becker, H. A. (1960). On the absorption of liquid water by the wheat kernel. Cereal Chemistry, 37, 309–323. Becker, R., Irving, D. W., & Saunders, R. M. (1986). Production of debranned amaranth flour by stone milling. Lebensmittel–Wissenschaft und- Technologie, 19, 372. Betschart, A. A., Irving, D. W., Shephard, A. D., & Saunders, R. M. (1981). Amaranthus cruentus: milling characteristics, distribution of nutrients within seed components, and the effects of temperature on nutritional quality. Journal of Food Science, 46, 1175–1177. Calzetta Resio, A. N., Tolaba, M., & Suarez, C. (2000). Food Science and Technology International, 6(5), 371–378. Cussler, E. L. (1984). Diffusion: Mass Transfer in Fluid System. Cambridge: Cambridge University Press. Du, L., Li, B., Lopez-Filho, J. F., Daniels, C. R., & Eckhoff, S. R. (1996). Effect of selected organic and inorganic acids on corn wet milling yields. Cereal Chemistry, 73, 96–98. Engels, C., Hendrickx, M., De Samblanx, S., De Gryze, I., & Tobback, P. (1986). Modeling water diffusion during long-grain rice soaking. Journal of Food Engineering, 5, 55–73. Fan, L. T., Chun, D. S., & Shellenberger, J. A. (1961). Diffusion coefficients of water in wheat kernels. Cereal Chemistry, 38, 540– 548.

Haros, C. M., & Suarez, C. (1999). Effect of chemical pretreatments and lactic acid on the rate of water absorption and starch yield in corn wet-milling. Cereal Chemistry, 76, 783–787. Haros, M., Viollaz, P. E., & Suarez, C. (1995). Effect of temperature and SO2 on the rates of water absorption of three maize hybrids. Journal of Food Engineering, 25, 473–482. Hsu, H. K. (1983). A diffusion model with a concentration-dependent diffusion coefficient for describing water movement in legumes during soaking. Journal of Food Science, 48, 618–622. Lehmann, J. W. (1996). Case history of grain amaranth as an alternative crop. Cereal Foods World, 41, 399–408. Lomauro, C. J., Bakshi, A. S., & Labuza, T. P. (1985). Lebensmittel– Wissenschaft und- Technologie, 18, 111–114. Myers, D. J., & Fox, S. R. (1994). Alkali wet-milling characteristics of pearled and unpearled amaranth seed. Cereal Chemistry, 71, 96–99. Neuman, P. E., Wall, J. S., & Walker, C. E. (1984). Chemical and physical properties of proteins in wet-milling corn gluten. Cereal Chemistry, 61(4), 272–277. Norris, J. R., & Rooney, L. W. (1970). Wet-milling properties of four sorghum parents and their hybrids. Starch, 47, 64–69. Paredes-L opez, O., Schevenin, M. L., Hernandez-L opez, D., & Carabez-Trejo, A. (1989). Amaranth starch-isolation and partial characterization. Cereal Chemistry, 6, 205–207. Shandera, D. L., Packhurst, A. M., & Jackson, D. S. (1995). Interactions of sulfur dioxide, lactic acid and temperature during simulated corn wet- milling. Cereal Chemistry, 72, 371–378. Sopade, P. A., & Obepka, J. A. (1990). Modelling water absorption in soybean, cowpea and peanuts at three temperatures using PelegÕs equation. Journal of Food Science, 55(4), 1084–1087. Steinke, J. D., & Johnson, L. A. (1991). Steeping maize in the presence of multiple enzymes I. Static batchwise steeping. Cereal Chemistry, 68(1), 7–12. Teutonico, R. A., & Knorr, D. (1985). Amaranth composition, properties and applications of a rediscovered food crop. Food Technology, 52, 49–60. Tolaba, M. P., Viollaz, P. E., & Suarez, C. (1990). The use of a diffusional model in determining the permeability of corn pericarp. Journal of Food Engineering, 12, 53–66. Uriyapongson, J., & Rayas-Duarte, P. (1994). Comparison of yield and properties of amaranth starches using wet and dry-wet milling process. Cereal Chemistry, 71, 571–577. Yanez, G. A., Messinger, J. K., Walker, C. E., & Rupnow, J. H. (1986). Amaranthus hypochondriacus: starch isolation and partial characterization. Cereal Chemistry, 63, 273–276. Yuan, J., Chung, D. S., Seib, P. A., & Wang, Y. (1998). Effect of steeping conditions on wet-milling characteristics of hard red winter wheat. Cereal Chemistry, 75, 145–148. Zhao, J., & Whistler, R. L. (1994). Isolation and characterization of starch from amaranth flour. Cereal Chemistry, 71, 392–393.