- Email: [email protected]
Influence of corn drying on its quality for the wet-milling process
Journal of Food Engineering 60 (2003) 177–184 www.elsevier.com/locate/jfoodeng
Influence of corn drying on its quality for the wet-milling process M onica Haros, Marcela P. Tolaba, Constantino Su arez
*
Departamento de Industrias, Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires, Ciudad Universitaria, 1428 Buenos Aires, Argentina Received 9 September 2002; accepted 13 January 2003
Abstract Flint and dent corn were forced air-dried at 70–110 C to a final moisture of about 12.0%. The rates of water absorption of both hybrids in the presence of 0.25% SO2 aqueous solution were evaluated in terms of the diffusion coefficient. Drying temperature affected negatively the rate of absorption of both hybrids. A laboratory wet-milling procedure was developed to evaluate starch recovery of corn samples. For flint corn starch recovery drop from 96.5% (undried) to 82% (dried at 110 C); for dent corn the drop was from 97.5% to 90%. The starch isolated from air-dried corn contained greater amounts of protein than starch originated from undried corn. The effect was more marked for flint than for dent corn and increased with the drying temperature. Sorptional characteristics of starch were practically unaffected by drying temperature. DSC transition temperatures of starch showed an increasing tendency with drying temperature. For both hybrids the gelatinization enthalpy of starch decreased with the increasing of drying temperature, the effect being more marked for flint than for dent corn. 2003 Published by Elsevier Ltd. Keywords: Corn drying; Wet-milling; Starch quality; Starch thermal properties
1. Introduction Corn wet milling is an industrial process that separates the corn kernel into its starch, protein, germ and fiber fractions. Commercially, corn kernels are steeped for about 48–50 h in SO2 aqueous solution (50–55%), which facilitates the separation of starch and gluten (Watson, 1991). As most of the starch is embedded in the cytoplasmatic matrix, the isolation and purification in the native state is rather difficult (Lorenz & Kulp, 1980). This difficulty may increase considerably by improper drying of the grain. According to Mac Masters, Finkner, Holzapfel, Ramser, and Dungan (1959) drying may impact negatively on starch recovery, creating other processing problems such as poor germ separation and low yield of oil from germ. Watson and Hitara (1962) found that corn millability may be affected when drying takes place at temperatures above 82 C. According to that authors the dried corn in that conditions releases two-thirds of the original starch quantity present in corn kernels. Mistry, Xutian, Eckhoff, and Litchfields (1993)
*
Corresponding author. Tel./fax: +54-11-4576-3366. E-mail address: [email protected] (C. Suarez).
0260-8774/03/$ - see front matter 2003 Published by Elsevier Ltd. doi:10.1016/S0260-8774(03)00038-4
analyzed the combined effect of high moisture levels and temperatures on the yields of the different fractions. Peplinski, Paulis, Bietz, and Pratt (1994) investigated the damage caused by drying to physical properties of corn kernels, such as density, breakage susceptibility and germination. Factors such as water uptake and saturation moisture level during corn steeping and how they are affected by drying conditions are not well documented. The rate of water penetration into the kernels is also an important parameter to assure an adequate separation the protein from starch granules (Shandera & Jackson, 1996). Little information exists in the literature concerning the wet-milling characteristics of flint corn and how they are affected by drying conditions. Nerying and Reilly (1984) have pointed out that the cultivar of the corn has a strong effect on yield, purity, and ease of milling. Flint corn, widely cultivated in Argentina, has hard kernels due to a large amount of corneous endosperm. Based on the present considerations, the present work was conducted to obtain the following information: • To evaluate the effect of drying temperature on the millability of flint and dent corn. • To quantify the recovery and purity of starch isolated from dried and undried flint and dent corn.
178
M. Haros et al. / Journal of Food Engineering 60 (2003) 177–184
• To evaluate the incidence of drying on the physical and thermal properties of wet-milled starch.
2. Materials and methods 2.1. Materials Two corn hybrids provided by INTA Pergamino, Buenos Aires Province, Argentina, were used in this work: flint (CARGILL T-42) and dent (PIONEER 3379). The grains were harvested with a moisture content of 0.166 and 0.168 g water per g dry solid, for flint and dent corn, respectively. Conditioning was achieved by adding a calculated quantity of distilled water in the form of a fine spray with periodic agitation of grains. The corn was then stored in sealed plastic jars for 2 weeks at 4 C to allow for moisture equilibration. By means of such procedure the moisture contents of flint corn was increased to 0.277 g water per g dry solid and the moisture content of dent corn was 0.245 g water per g dry solid. The moisture content was determined in two steps by AACC method 44-15A (1995). Previous to conditioning the grains were screened to obtain samples of uniform size, free of broken kernels and foreign material. Only the fractions retained between screens 9.0–9.5 mm were used. Equivalent spherical radius of whole kernels was determined by chlorohexene displacement; a total of 12 kernels were used in each measurement and the number of replications was three. Maximum length, including tip cap (a), thickness (b), and width (c) of 50 samples were also measured using a micrometer. Starch, protein and oil contents were determined for better characterization of the grains. Starch was measured using the Twers method (Egan, Kirk, & Sawyer, 1987). Protein (N 6:25) was determined by the macro– Kjeldahl method with B€ uchi 430 digester and B€ uchi 320 distiller (AOAC, 1980). Oil was removed by Soxhlet extraction with hexane for 24 h (Bertoni, Pereyra Gonzalez, & Cattaneo, 1994). For flint corn the starch, protein and oil contents were 0.70 0.001, 0.092 0.001, and 0.015 0.001 g/g dry basis. For dent corn the starch, protein and oil contents were, respectively, 0.76 0.02, 0.063 0.001 and 0.012 0.001 g/g dry basis. 2.2. Laboratory drier The drying equipment basically consisted of a centrifugal fan blowing air over electric bar elements into the base of a camber and then upwards through a vertical duct. The vertical duct had a flow-homogenizing section containing small glass spheres. A metal cup with a mesh base and lid served as the drying chamber and was mounted on the outlet of the duct.
Single layers of corn of 150 g were used for each drying run. An air velocity of 5 ms1 was used in the experiments. The inlet air dry bulb temperature was controlled to within 0.2 C by an electronic proportional controller. Wet and dry bulb thermometers were placed in the inlet duct to measure the relative humidity of the drying air. Three dry bulb temperatures were tested: 70, 90, 110 C. The relative humidity of drying air was maintained between 5% and 8% in all drying runs. All drying experiments were conduced in one-stage drying. The final moisture content of grains after drying were 0.125 g water per g dry solid for flint and 0.122 g water per g dry solid for dent corn. 2.3. Water absorption procedure In order to evaluate the rate of water absorption of control (undried) and dried corn samples the following procedure was performed. Control and dried corn samples (10 g) were soaked in 0.25% SO2 aqueous solution prepared by dissolving the appropriate amount of sodium bisulfite in distilled water. The grains were placed in 100 ml vessels with screw caps and gently agitated in a thermostatic bath at 52.0 0.5 C to reduce film resistance in the steeped solution. At regular time intervals, up to 48 h, the flasks were removed from the bath to determinate water uptake. The grains were quickly removed from the flasks and superficially dried by manually rolling the kernels on a large piece of filter paper until they lost the glistening appearance associated with the presence of surface film of water. The grains were then weighed to determine the moisture uptake at each time interval; water absorption curves were determined by replicate. The amount of water absorbed in each soaking experiment was determined by drying the corn kernels in two steps (Haros, Viollaz, & Suarez, 1995). The first of them was in current air at 65– 70 C overnight, then the corn kernels were dried in a vacuum oven at 70 C, in presence of P2 O5 as desiccant, to constant weight. Constant weight was attained when the weight of the kernels between successive measurements changed <0.5 mg over 24 h. 2.4. Calculation of water diffusivity To compare the rate of water absorption of undried and dried samples of flint and dent corn, the diffusion of water into the kernel was estimated. For this purpose, the analytical solution of FickÕs second law of diffusion in solids of arbitrary shape was used. According to Luikov (1968), the diffusion in a body of arbitrary shape can be reduced to the analytical solution corresponding to sphere by defining adequate shape factors. Following this idea, Tolaba, Aguerre, and Suarez (1989) found that water migration in corn kernels can be described by means of the equation:
M. Haros et al. / Journal of Food Engineering 60 (2003) 177–184
u ¼
1 ðu us Þ 6 X 1 ¼ 2 expðq2 p2 WFoÞ ðu0 us Þ p q¼1 q2
ð1Þ
where u, u0 , and us ¼ mean, initial, and saturation moisture concentrations, kg of water/kg of dry solid; Fo ¼ Dt=R2e ¼ Fourier number, dimensionless; t ¼ time, s; Re ¼ equivalent radius: radius of sphere having the same volume as the grain, m; D ¼ diffusion coefficient, m2 /s; and W ¼ shape factor, dimensionless. The equation used to calculate the shape factor was (Tolaba et al., 1989) 0 12 2 8R e A W¼@ ð2Þ 1 cd dc þ sin e e where a, b and c are the parameters, d ¼ ða þ bÞ=2 and 1=2 e ¼ ½ða2 d 2 Þ=a2 . To estimate the diffusion coefficient, Eq. (1) was programmed on a digital computer and the value of D calculated by a nonlinear regression technique that varied the value of D until the difference between experimental and predicted moisture concentrations was at a minimum. 2.5. Total solids leached determination The amount of solids leached at different time intervals during water steeping in aqueous solution of SO2 was determined as follows (Haros et al., 1995). A 10 ml sample of the steeping water was placed in a weighed glass 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 the presence of P2 O5 desiccant. The percent of total solids leached was referred to the weight of fully dried kernels. 2.6. Steeping index The steeping index, described by Brown, Fulford, Daynard, Meiering, and Otten (1979) was used here to have a visual examination of kernel section after steeping process of undried and dried corn samples. The samples used for this purpose were steeped in aqueous SO2 solution, 0.25% v/v, at 52 C for 48 h. Steeped kernels were then sectioned longitudinally (parallel to the germ) and examined visually; after that the grains were classified into one of three following categories: Category 1: horny endosperm intact and tough, protein matrix intact. Category 2: kernel softened but retains dented appearance and dry interior, protein matrix disrupted. Category 3: kernel fully swollen and wet internally, protein matrix dispersed. The steeping index, SI, was then calculated by the expression:
179
SI ¼ No: of kernels in Category 1 þ ðNo: of kernels in Category 2Þ 2 þ ðNo: of kernels in Category 3Þ 3
ð3Þ
The number of grains used for the calculation of SI was 100; so an index of 300 will indicate that the grains are totally steeped while for SI ¼ 100 they will retain intact the horny endosperm. 2.7. Starch isolation The method of isolation was based on that described by Haros and Suarez (1997). Roughly 50 g of artificially dried and undried corn samples were steeped in 250 ml of 0.25% SO2 aqueous solution for 48 h at 52 C. After that the steepwater was decanted and the excess of liquid water was removed from the corn by blotting. Steeped corn was ground in presence of 100 ml of distilled water for 3 min by means of a Waring blender (Waring Co., New Jersey, USA). The water slurry was hand-sieved through a set of stainless-steel screens of the following sized openings: 0.420, 0.074 and 0.053 mm. Germ and fiber were retained in the first screen, protein in the second and third. The starch slurry passing trough 0.074 and 0.053 mm opening was centrifuged at 1200 g for 20 min (Centrifuge Rolco Ind., Buenos Aires, Argentina). The starch fraction was resuspended at room temperature in pure acetone for 5 min for purification, the suspension filtrated and finally dried in air current at 25 C (Krueger, Knutson, Inglett, & Walker, 1987). Starch recovery was calculated as the ratio of the weight of starch recovered from wet milling to the total weight of the starch present in the corn. The nitrogen content of the recovered fraction starch was determined by the macro-Kjeldahl method (AOAC, 1980). The factor of 6.25 g of protein per gram of nitrogen was used to calculate the protein contents. 2.8. Measurement of the desorption isotherms A fraction of the isolated starch by the method above described, about 100 g, was used to evaluate the sorptional characteristics at different water activities. For this purpose samples of about 2 g were taken from the bulk of starch powder and placed in 30 mm 30 mm weighing bottles to form a layer of about 4 mm high. The bottles were placed in vacuum dessicators containing saturated salt solutions, which provided defined constant water activity inside the dessicator. Each salt solution was prepared following the specifications given by Greenspan (1977). Desorption isotherms of each starch fraction were determined at 25 C by the static method: starch samples were let to equilibrate over different salt solutions in vacuum dessicators. All the samples lost water and the equilibrium was judged to
180
M. Haros et al. / Journal of Food Engineering 60 (2003) 177–184
have been attained when the weight of the sample between successive measurements changed by less than 2 mg over 72 h. Dry weigh of corn starch was measured gravimetrically after vacuum drying over P2 O5 at 70 C until constant weight. 2.9. Thermal properties of starch Differential scanning calorimetry (DSC) was used to study the thermal properties of starch isolated from dried and undried corn samples. A PL-DSC (Polymer Laboratories, UK) calorimeter equipped with a thermal analysis data station was used. Samples (5 mg) of starch were weighed into aluminum sample pans (accuracy 0.01 mg). Deionized water was added into the pan to obtain water: starch ratio 3:1 and the pan was then hermetically sealed. Samples were heated at the rate of 10 C/min from 30 to 110 C. Heat of gelatinization DH , onset temperature T0 , peak temperature Tp and end temperature Te were computed. All samples were run in triplicate.
3. Results and discussion 3.1. Water absorption and diffusion coefficient For flint and dent corn, the equivalent spherical radii of whole kernels were 4.35 0.04 mm and 4.45 0.05 mm (mean values deviation standard), respectively. The values of maximum length, thickness, and width for flint corn were a ¼ 11:30 0:15 mm, b ¼ 9:53 0:89 mm, and c ¼ 6:01 1:75 mm; for dent corn were a ¼ 13:25 0:26 mm, b ¼ 9:39 0:32 mm, and c ¼ 4:82 0:68 mm (mean values deviation standard), respectively. The W values for flint and dent corn were 0.78 0.09 and 0.75 0.05 (mean values deviation standard), respectively. Eq. (1) was used to estimate the diffusion coefficients of water in both corn hybrids (dried and control). The saturation moisture content, us , in Eq. (1) was estimated
as follows. When the time becomes large the limiting form of Eq. (1) results u us 6 ¼ expðp2 WFoÞ u0 us p2
ð4Þ
From this equation it can be easily demonstrated that for any set of three moisture levels taken at equally spaced time intervals of duration j, the following expression results: us ¼
ui uiþ2j u2iþj ui þ uiþ2j 2uiþj
ð5Þ
ui and uj being the moisture contents of time i and j. The saturation moisture contents of control and dried corn samples are given in Table 1. These values were then used to estimate D from Eq. (1) using a nonlinear regression method. Diffusion coefficient determinations were replicated three times. It is observed from Table 1 that the values of us decrease with the increase of drying temperature, being this effect more marked for flint than for dent corn. It can also be seen that for the control samples, the value of D for flint corn is lower than for dent corn. It is probable that the main reason for such differences would be the kernel characteristics of both hybrids. For flint type kernels the concentration of starch is higher around the periphery of the endosperm than in the centre, giving the endosperm a hard, external layer. In dent kernels the ratio of hard to floury endosperm is, in general, lower than in flint kernels, and the floury tissue extends to the crown of the endosperm, providing easy paths for moisture penetration. Structural differences in the protein matrix in which starch is embedded may also account for that value (Robuti, 1980). In Table 1 it can also be observed that the values of the diffusion coefficients of the dried samples were higher than the controls. The kernels dried exhibited excessive stress cracks and fissures, which resulted in increased the water absorption rate because the barrier for water diffusion had been reduced. However, the increase of drying temperature from 70 to 110 C
Table 1 Saturation moisture contents and diffusion coefficients of undried and dried flint and dent corn samples steeped in 0.25% SO2 Corn sample
Drying temperature (C)
us (g/g dry basis)a ;b
D 1011 (m2 /s)a;c
Flint
Undried 70 90 110
0.634 0.005 0.614 0.002 0.588 0.007 0.570 0.005
3.8 0.6 5.6 0.9 5.0 0.7 4.7 0.9
Dent
Undried 70 90 110
0.734 0.010 0.715 0.025 0.714 0.005 0.689 0.002
5.7 0.6 7.2 0.2 6.8 1.1 6.5 0.3
a
Mean value standard error, significant a ¼ 0:05. Eq. (5). c Eq. (1). b
M. Haros et al. / Journal of Food Engineering 60 (2003) 177–184
181
produced a decrease in the rate of absorption; for flint corn the D values decrease above 17%, while for dent corn such decreasing was only of 10%. The high drying temperatures tend to damage the endosperm protein provoking irreversible biochemical changes reducing the beneficial effect of SO2 on the absorption rate (Watson, 1991).
3.2. Effect of drying temperature on steeping The comparative effect of drying temperature of kernels on the total solids leached during steeping is illustrated in Fig. 1A and B, from flint and dent, respectively. In these figures it can be seen that the amount of solids released increases with the increase of steeping time. This is due to the action of SO2 on the endosperm matrix of the kernels which provokes solubilization of the endosperm proteins by reduction of disulfide bonds by the SO2 (Neuman, Wall, & Walker, 1984). It is also observed that the level of solids released decreases with the drying treatment. For both hybrids, the amount of solids released in dried samples was in most cases smaller than for the control. The diminishing of endogenous enzymatic activity (Peplinski et al., 1994) together with the denaturation and chemical changes induced by the high drying temperature (French & Kingsolver, 1964; Peplinski et al., 1994) are mainly responsible for such differences.
Fig. 1. Effect of drying temperature on total solids leached during (A) flint corn steeping and (B) dent corn steeping.
Wet-milling performance was evaluated on the basis of recovery and purity of starch fraction. Both param-
Table 2 Effect of drying temperature on starch recovery, protein content of starch wet-milling fraction and steeping index of dent and flint corn Drying temperature (C)
Undried 70 90 110
Flint
Dent a;b
a ;c
Starch recovery (%)
PPS
Steeping index
Starch recoverya ;b (%)
PPSa; c
Steeping indexd
96.5 0.5 92.3 0.4 92.4 0.5 82.0 0.6
0.90 0.03 0.90 0.05 1.31 0.06 5.72 0.03
259 238 215 161
97.5 0.3 97.2 0.4 98.0 0.6 90.0 0.3
1.20 0.03 2.41 0.01 4.00 0.04 4.22 0.04
253 241 223 138
a
Average of four replicates ( standard deviation). All recoveries are expressed on a percentage dry solid. c Percent of protein in starch (mean value standard deviation). d Eq. (3). b
Table 3 Estimated GAB parameters, Eq. (6), for corn starch isolated from undried and dried flint and dent corn Material
Drying temperature
mm (dry basis)
C
k
Flint corn
Undried 70 C 90 C 110 C
0.129 0.002 0.116 0.009 0.122 0.010 0.122 0.012
32.0 1.7 27.9 0.7 19.2 1.2 18.4 0.4
0.58 0.06 0.62 0.04 0.59 0.06 0.58 0.03
Dent corn
Undried 70 C 90 C 110 C
0.118 0.007 0.129 0.011 0.121 0.008 0.132 0.014
30.3 2.8 17.1 2.5 14.9 1.4 15.3 2.4
0.60 0.03 0.54 0.03 0.57 0.04 0.54 0.05
182
M. Haros et al. / Journal of Food Engineering 60 (2003) 177–184
eters are reported in Table 2 for control and dried corn samples. High starch recovery and low protein content in starch were considered as indicators of good milling. Wet-milling results from our 48-h controls (flint and dent hybrids) are compared with other batch steeping milling procedures (Steinke & Johnson, 1991) and wetmilling results from industry (Anderson & Watson, 1982). The starch recovery of undried flint corn was 96.5% that is lower but not significantly different than the value of 97.5% corresponding to dent hybrid (relative error of starch recovery was 1.8%). Weller, Paulsen, and Steinberg (1988) found similar results working with dent corn hybrids of different vitreous-to-floury endosperm ratios. The effect of drying temperature on starch recovery had similar trends in dent and flint hybrids. The increase of temperature reduced starch recovery rates of both hybrids. The starch recoveries drop to the extent of 15.0% and 7.7% for flint and dent corn, respectively, when drying temperature was increased to 110 C. Our starch protein contents for the control samples were higher than the typical industrial practice, 0.4–0.5% (Anderson & Watson, 1982). Even though the wetmilling procedure used here reproduced relatively well, as evidenced by the relatively low standard deviations reported in Table 2, certain subjectivity in the separation protein–starch may account for those differences. We believe, however, that the wet-milling process used in this study served well as the basis for comparing the effect of drying temperature on starch isolation of grain corn. This is illustrated in Table 2 where it can be seen that the increasing of drying temperature leads to more severe reduction in recoverable starch and higher protein contamination. These results correlate well with the values of steeping index also given in Table 2. Brown et al. (1979) found that steeping index less than 200 conducts to unsatisfactory milling performance and difficulties for starch–gluten separation. This is corroborated for the drying temperature of 110 C, which conducted to the lowest starch recovery and highest protein contamination for both of the hybrids investigated. 3.3. Sorptional behaviour of starch The experimental equilibrium moisture contents of corn starch isolated from flint and dent corn at 25 C were obtained by triplicate. The maximum relative error (coefficient of variation) in equilibrium moisture values was 0.65%. For a better quantification of the treatment on the equilibrium values the monolayer moisture contents were calculated from GAB equation (Weisser, 1985): m Ckaw ¼ mm ð1 kaw Þð1 kaw þ Ckaw Þ
ð6Þ
Fig. 2. DSC thermograms of isolated starch of (A) corn (flint) drying and (B) corn (dent) drying at different temperature: (a) undried, (b) 70 C, (c) 90 C, (d) 110 C.
In this equation m and mm are the moisture and the monolayer moisture contents (kg water/kg dry solid), aw the water activity, and C and k the parameters related to the temperature effect through the equation. The
M. Haros et al. / Journal of Food Engineering 60 (2003) 177–184
183
Table 4 Thermal properties of isolated starch from flint and dent corn as affect by drying temperature Material
Endotherm temperature, C, and gelatinization enthalpy, DH (J/g)
Drying temperature Undried
70 C
90 C
110 C
Flint corn
T0 Tp Te DH
69.1 73.8 82.0 13.7
69.1 74.0 82.7 13.0
69.1 74.3 83.5 12.2
70.0 75.5 84.2 11.3
Dent corn
T0 Tp Te DH
69.2 73.2 81.6 11.6
69.2 73.5 82.7 11.3
69.5 75.1 83.2 10.4
70.4 75.6 84.3 10.0
parameters of GAB equation (mm , C and k) were estimated by nonlinear regression and are given in Table 3. The values of mm range to approximately 0.11–0.13 kg water per kg dry solid and represent the water more strongly adsorbed and immobilized at polar sorption sites, located in a fairly rigid structure of starch polymer. There is no marked difference in the monolayer values of the starch isolated from dent and flint corn (untreated) as well as of the isolated from dried corn samples. For both hybrids it is observed, however, certain differences between the C values of the dried and undried corn samples. It appears that the increase of drying temperature tends to decrease the binding energy of the first adsorbed layer, as it results from the less values of C obtained from the dried corn samples in comparison with the undried ones. On the contrary, the properties of the adsorbed water in the multilayer region do not seem to be affected by the drying temperature of the grain, such as it results from the similarity among k-values reported in Table 3.
nization; in both hybrids it is observed a decrease of DH with the increase of temperature. As drying has a negative effect on starch–protein separation (see Table 2), it is likely that the protein remaining in starch should reduce the entrance of water into the granules during gelatinization, avoiding interaction between water and starch. If this were so, a partial wetting of the starch granules would be expected during DSC runs and a loss in the enthalpy of gelatinization. It is interesting to observe that such loss was about 18% for flint corn and 13% for dent corn.
Acknowledgements We acknowledge financial support from the Consejo Nacional de Investigaciones Cientıficas y Tecnicas (CONICET) and Universidad de Buenos Aires, Rep ublica Argentina.
3.4. Thermal properties of starch Thermal transitions of milled starch (in excess water) from flint and dent corn (undried and dried) are shown in Fig. 2A and B. The parameters of DSC thermograms are reported in Table 4. It can be observed that the transition temperatures, onset (T0 ), peak (Tp ), and end (Te ) show a slight but significant tendency to increase with drying treatment. Endotherm peak widths (Te T0 ) were also influenced by drying. For flint corn peak width increased to 14.2 C for corn dried at 110 C; a similar behaviour was found for dent corn. In Table 4 it is also observed that DH -value for flint corn (undried) was higher than the corresponding to dent corn. The large proportion of vitreous endosperm of flint corn in comparison to dent corn (Watson, 1991) together with the fact that vitreous endosperm is more difficult to gelatinize than floury endosperm (Le Brass, 1982) may account for such differences. The drying temperature seems to affect also the enthalpy of gelati-
References AACC (1995). Approved methods of the American Association of Cereal Chemists (9th ed.). Method 44-15A. Anderson, R. A., & Watson, S. A. (1982). The corn of milling industry. In A. Wolf (Ed.), CRC handbook of processing and utilization in agriculture (Vol. 2, Part 1.I, pp. 31–61). Boca Raton: CRC Press. AOAC (1980). Official methods of analysis (13th ed.). Washington, DC: Association of Official Analytical Chemists. Bertoni, M. H., Pereyra Gonzalez, A., & Cattaneo, P. (1994). Estudios sobre frutos de Cucurbita moschata Duch (calabazita). II. Aceite crudo de extracci on de semilla. Caracterısticas y composici on analıtica. Anales de la Asociacion Quımica Argentina, 82, 131–133. Brown, R. B., Fulford, G. N., Daynard, T. B., Meiering, A. G., & Otten, L. (1979). Effect of drying method on grain corn quality. Cereal Chemistry, 56, 529–532. Egan, H., Kirk, R. S., & Sawyer, R. (1987). Analisis Quımico de Alimentos de Pearson (pp. 244–245). S.A. de C.V., Mexico: Compa~ nıa Editorial Continental. French, R. C., & Kingsolver, C. H. (1964). Effect of excessive heat during artificial drying of corn on reducing sugar content and diastasic activity. Cereal Chemistry, 41, 47–58.
184
M. Haros et al. / Journal of Food Engineering 60 (2003) 177–184
Greenspan, L. (1977). Humidity fixed points of binary saturated aqueous solutions. Journal of Research of the National Bureau of Standards––A. Physics and Chemistry, 81(1), 89–96. Haros, M., & Suarez, C. (1997). Effect of drying, initial moisture and variety in corn wet milling. Journal of Food Engineering, 34, 473– 481. 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. Krueger, B. R., Knutson, C. A., Inglett, G. E., & Walker, C. E. (1987). A differential scanning calorimetry study on the effect of annealing on gelatinization behaviour of corn starch. Journal of Food Science, 52, 715–718. Le Brass, A. (1982). Maize drying conditions and its resulting quality for wet-milling industry. In G. E. Inglett (Ed.), Maize: recent progress in chemistry and technology (pp. 95–127). New York: Academic Press. Lorenz, K., & Kulp, K. (1980). Steeping of starch at various temperatures––effects on functional properties. Starch/St€arke, 32, 181–186. Luikov, A. V. (1968). Analytical heat diffusion theory. London: Academic Press. Mac Masters, M. M., Finkner, M. O., Holzapfel, M. M., Ramser, J. H., & Dungan, G. H. (1959). A study of the effect of drying conditions on the suitability for starch production of corn artificially dried after shelling. Cereal Chemistry, 36, 247–259. Mistry, A. H., Xutian, W., Eckhoff, S. R., & Litchfields, J. B. (1993). Wet-milling characteristics of high-temperature, high-humidity maize. Cereal Chemistry, 70, 360–361. Nerying, A., & Reilly, P. J. (1984). Laboratory wet milling of ensiled corn kernels. Cereal Chemistry, 61, 8–14.
Neuman, P. E., Wall, J. S., & Walker, C. E. (1984). Chemical and physical properties of proteins in wet-milled corn gluten. Cereal Chemistry, 61(41), 353–356. Peplinski, A. J., Paulis, J. W., Bietz, J. A., & Pratt, R. C. (1994). Drying of high-moisture corn: changes in properties and physical quality. Cereal Chemistry, 71, 129–133. Robuti, J. L. (1980). Observaciones de microscopıa electr onica de barrido sobre endosperma de la lınea end ocrina P21 de maız. Boletın de divulgaci on tecnica No. 41. INTA, Instituto Nacional de Tecnologıa Agropecuaria, Estaci on Experimental Regional Agropecuaria Pergamino, Argentina. Shandera, D. L., & Jackson, D. S. (1996). Effect of corn wet-milling conditions (sulfur dioxide, lactid acid, and steeping temperature) on starch functionality. Cereal Chemistry, 73, 632–637. Steinke, J. D., & Johnson, L. A. (1991). Steeping maize in the resence of multiple enzymes. I. Static batchwire steeping. Cereal Chemistry, 68, 7–12. Tolaba, M. P., Aguerre, R. J., & Suarez, C. (1989). Shape characterisation for diffusional analysis of corn drying. Drying Technology, 7(2), 205–217. Watson, S. A. (1991). Structure and composition. In S. A. Watson, & P. E. Ramstad (Eds.), Corn: chemistry and technology (pp. 53–82). St Paul, MN: American Association of Cereal Chemistry. Watson, S. A., & Hitara, Y. (1962). Some wet-milling properties of artificially dried corn. Cereal Chemistry, 39, 35–44. Weisser, H. (1985). Influence of temperature on sorption equilibria. In D. Simato, & J. L. Multon (Eds.), Properties of water in foods (p. 95). Dordrecht: Martinus Nijhoff. Weller, C. L., Paulsen, M. R., & Steinberg, M. P. (1988). Correlation of starch recovery with assorted quality factors of four corn hybrids. Cereal Chemistry, 65, 392–397.
Influence of corn drying on its quality for the wet-milling process M onica Haros, Marcela P. Tolaba, Constantino Su arez
*
Departamento de Industrias, Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires, Ciudad Universitaria, 1428 Buenos Aires, Argentina Received 9 September 2002; accepted 13 January 2003
Abstract Flint and dent corn were forced air-dried at 70–110 C to a final moisture of about 12.0%. The rates of water absorption of both hybrids in the presence of 0.25% SO2 aqueous solution were evaluated in terms of the diffusion coefficient. Drying temperature affected negatively the rate of absorption of both hybrids. A laboratory wet-milling procedure was developed to evaluate starch recovery of corn samples. For flint corn starch recovery drop from 96.5% (undried) to 82% (dried at 110 C); for dent corn the drop was from 97.5% to 90%. The starch isolated from air-dried corn contained greater amounts of protein than starch originated from undried corn. The effect was more marked for flint than for dent corn and increased with the drying temperature. Sorptional characteristics of starch were practically unaffected by drying temperature. DSC transition temperatures of starch showed an increasing tendency with drying temperature. For both hybrids the gelatinization enthalpy of starch decreased with the increasing of drying temperature, the effect being more marked for flint than for dent corn. 2003 Published by Elsevier Ltd. Keywords: Corn drying; Wet-milling; Starch quality; Starch thermal properties
1. Introduction Corn wet milling is an industrial process that separates the corn kernel into its starch, protein, germ and fiber fractions. Commercially, corn kernels are steeped for about 48–50 h in SO2 aqueous solution (50–55%), which facilitates the separation of starch and gluten (Watson, 1991). As most of the starch is embedded in the cytoplasmatic matrix, the isolation and purification in the native state is rather difficult (Lorenz & Kulp, 1980). This difficulty may increase considerably by improper drying of the grain. According to Mac Masters, Finkner, Holzapfel, Ramser, and Dungan (1959) drying may impact negatively on starch recovery, creating other processing problems such as poor germ separation and low yield of oil from germ. Watson and Hitara (1962) found that corn millability may be affected when drying takes place at temperatures above 82 C. According to that authors the dried corn in that conditions releases two-thirds of the original starch quantity present in corn kernels. Mistry, Xutian, Eckhoff, and Litchfields (1993)
*
Corresponding author. Tel./fax: +54-11-4576-3366. E-mail address: [email protected] (C. Suarez).
0260-8774/03/$ - see front matter 2003 Published by Elsevier Ltd. doi:10.1016/S0260-8774(03)00038-4
analyzed the combined effect of high moisture levels and temperatures on the yields of the different fractions. Peplinski, Paulis, Bietz, and Pratt (1994) investigated the damage caused by drying to physical properties of corn kernels, such as density, breakage susceptibility and germination. Factors such as water uptake and saturation moisture level during corn steeping and how they are affected by drying conditions are not well documented. The rate of water penetration into the kernels is also an important parameter to assure an adequate separation the protein from starch granules (Shandera & Jackson, 1996). Little information exists in the literature concerning the wet-milling characteristics of flint corn and how they are affected by drying conditions. Nerying and Reilly (1984) have pointed out that the cultivar of the corn has a strong effect on yield, purity, and ease of milling. Flint corn, widely cultivated in Argentina, has hard kernels due to a large amount of corneous endosperm. Based on the present considerations, the present work was conducted to obtain the following information: • To evaluate the effect of drying temperature on the millability of flint and dent corn. • To quantify the recovery and purity of starch isolated from dried and undried flint and dent corn.
178
M. Haros et al. / Journal of Food Engineering 60 (2003) 177–184
• To evaluate the incidence of drying on the physical and thermal properties of wet-milled starch.
2. Materials and methods 2.1. Materials Two corn hybrids provided by INTA Pergamino, Buenos Aires Province, Argentina, were used in this work: flint (CARGILL T-42) and dent (PIONEER 3379). The grains were harvested with a moisture content of 0.166 and 0.168 g water per g dry solid, for flint and dent corn, respectively. Conditioning was achieved by adding a calculated quantity of distilled water in the form of a fine spray with periodic agitation of grains. The corn was then stored in sealed plastic jars for 2 weeks at 4 C to allow for moisture equilibration. By means of such procedure the moisture contents of flint corn was increased to 0.277 g water per g dry solid and the moisture content of dent corn was 0.245 g water per g dry solid. The moisture content was determined in two steps by AACC method 44-15A (1995). Previous to conditioning the grains were screened to obtain samples of uniform size, free of broken kernels and foreign material. Only the fractions retained between screens 9.0–9.5 mm were used. Equivalent spherical radius of whole kernels was determined by chlorohexene displacement; a total of 12 kernels were used in each measurement and the number of replications was three. Maximum length, including tip cap (a), thickness (b), and width (c) of 50 samples were also measured using a micrometer. Starch, protein and oil contents were determined for better characterization of the grains. Starch was measured using the Twers method (Egan, Kirk, & Sawyer, 1987). Protein (N 6:25) was determined by the macro– Kjeldahl method with B€ uchi 430 digester and B€ uchi 320 distiller (AOAC, 1980). Oil was removed by Soxhlet extraction with hexane for 24 h (Bertoni, Pereyra Gonzalez, & Cattaneo, 1994). For flint corn the starch, protein and oil contents were 0.70 0.001, 0.092 0.001, and 0.015 0.001 g/g dry basis. For dent corn the starch, protein and oil contents were, respectively, 0.76 0.02, 0.063 0.001 and 0.012 0.001 g/g dry basis. 2.2. Laboratory drier The drying equipment basically consisted of a centrifugal fan blowing air over electric bar elements into the base of a camber and then upwards through a vertical duct. The vertical duct had a flow-homogenizing section containing small glass spheres. A metal cup with a mesh base and lid served as the drying chamber and was mounted on the outlet of the duct.
Single layers of corn of 150 g were used for each drying run. An air velocity of 5 ms1 was used in the experiments. The inlet air dry bulb temperature was controlled to within 0.2 C by an electronic proportional controller. Wet and dry bulb thermometers were placed in the inlet duct to measure the relative humidity of the drying air. Three dry bulb temperatures were tested: 70, 90, 110 C. The relative humidity of drying air was maintained between 5% and 8% in all drying runs. All drying experiments were conduced in one-stage drying. The final moisture content of grains after drying were 0.125 g water per g dry solid for flint and 0.122 g water per g dry solid for dent corn. 2.3. Water absorption procedure In order to evaluate the rate of water absorption of control (undried) and dried corn samples the following procedure was performed. Control and dried corn samples (10 g) were soaked in 0.25% SO2 aqueous solution prepared by dissolving the appropriate amount of sodium bisulfite in distilled water. The grains were placed in 100 ml vessels with screw caps and gently agitated in a thermostatic bath at 52.0 0.5 C to reduce film resistance in the steeped solution. At regular time intervals, up to 48 h, the flasks were removed from the bath to determinate water uptake. The grains were quickly removed from the flasks and superficially dried by manually rolling the kernels on a large piece of filter paper until they lost the glistening appearance associated with the presence of surface film of water. The grains were then weighed to determine the moisture uptake at each time interval; water absorption curves were determined by replicate. The amount of water absorbed in each soaking experiment was determined by drying the corn kernels in two steps (Haros, Viollaz, & Suarez, 1995). The first of them was in current air at 65– 70 C overnight, then the corn kernels were dried in a vacuum oven at 70 C, in presence of P2 O5 as desiccant, to constant weight. Constant weight was attained when the weight of the kernels between successive measurements changed <0.5 mg over 24 h. 2.4. Calculation of water diffusivity To compare the rate of water absorption of undried and dried samples of flint and dent corn, the diffusion of water into the kernel was estimated. For this purpose, the analytical solution of FickÕs second law of diffusion in solids of arbitrary shape was used. According to Luikov (1968), the diffusion in a body of arbitrary shape can be reduced to the analytical solution corresponding to sphere by defining adequate shape factors. Following this idea, Tolaba, Aguerre, and Suarez (1989) found that water migration in corn kernels can be described by means of the equation:
M. Haros et al. / Journal of Food Engineering 60 (2003) 177–184
u ¼
1 ðu us Þ 6 X 1 ¼ 2 expðq2 p2 WFoÞ ðu0 us Þ p q¼1 q2
ð1Þ
where u, u0 , and us ¼ mean, initial, and saturation moisture concentrations, kg of water/kg of dry solid; Fo ¼ Dt=R2e ¼ Fourier number, dimensionless; t ¼ time, s; Re ¼ equivalent radius: radius of sphere having the same volume as the grain, m; D ¼ diffusion coefficient, m2 /s; and W ¼ shape factor, dimensionless. The equation used to calculate the shape factor was (Tolaba et al., 1989) 0 12 2 8R e A W¼@ ð2Þ 1 cd dc þ sin e e where a, b and c are the parameters, d ¼ ða þ bÞ=2 and 1=2 e ¼ ½ða2 d 2 Þ=a2 . To estimate the diffusion coefficient, Eq. (1) was programmed on a digital computer and the value of D calculated by a nonlinear regression technique that varied the value of D until the difference between experimental and predicted moisture concentrations was at a minimum. 2.5. Total solids leached determination The amount of solids leached at different time intervals during water steeping in aqueous solution of SO2 was determined as follows (Haros et al., 1995). A 10 ml sample of the steeping water was placed in a weighed glass 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 the presence of P2 O5 desiccant. The percent of total solids leached was referred to the weight of fully dried kernels. 2.6. Steeping index The steeping index, described by Brown, Fulford, Daynard, Meiering, and Otten (1979) was used here to have a visual examination of kernel section after steeping process of undried and dried corn samples. The samples used for this purpose were steeped in aqueous SO2 solution, 0.25% v/v, at 52 C for 48 h. Steeped kernels were then sectioned longitudinally (parallel to the germ) and examined visually; after that the grains were classified into one of three following categories: Category 1: horny endosperm intact and tough, protein matrix intact. Category 2: kernel softened but retains dented appearance and dry interior, protein matrix disrupted. Category 3: kernel fully swollen and wet internally, protein matrix dispersed. The steeping index, SI, was then calculated by the expression:
179
SI ¼ No: of kernels in Category 1 þ ðNo: of kernels in Category 2Þ 2 þ ðNo: of kernels in Category 3Þ 3
ð3Þ
The number of grains used for the calculation of SI was 100; so an index of 300 will indicate that the grains are totally steeped while for SI ¼ 100 they will retain intact the horny endosperm. 2.7. Starch isolation The method of isolation was based on that described by Haros and Suarez (1997). Roughly 50 g of artificially dried and undried corn samples were steeped in 250 ml of 0.25% SO2 aqueous solution for 48 h at 52 C. After that the steepwater was decanted and the excess of liquid water was removed from the corn by blotting. Steeped corn was ground in presence of 100 ml of distilled water for 3 min by means of a Waring blender (Waring Co., New Jersey, USA). The water slurry was hand-sieved through a set of stainless-steel screens of the following sized openings: 0.420, 0.074 and 0.053 mm. Germ and fiber were retained in the first screen, protein in the second and third. The starch slurry passing trough 0.074 and 0.053 mm opening was centrifuged at 1200 g for 20 min (Centrifuge Rolco Ind., Buenos Aires, Argentina). The starch fraction was resuspended at room temperature in pure acetone for 5 min for purification, the suspension filtrated and finally dried in air current at 25 C (Krueger, Knutson, Inglett, & Walker, 1987). Starch recovery was calculated as the ratio of the weight of starch recovered from wet milling to the total weight of the starch present in the corn. The nitrogen content of the recovered fraction starch was determined by the macro-Kjeldahl method (AOAC, 1980). The factor of 6.25 g of protein per gram of nitrogen was used to calculate the protein contents. 2.8. Measurement of the desorption isotherms A fraction of the isolated starch by the method above described, about 100 g, was used to evaluate the sorptional characteristics at different water activities. For this purpose samples of about 2 g were taken from the bulk of starch powder and placed in 30 mm 30 mm weighing bottles to form a layer of about 4 mm high. The bottles were placed in vacuum dessicators containing saturated salt solutions, which provided defined constant water activity inside the dessicator. Each salt solution was prepared following the specifications given by Greenspan (1977). Desorption isotherms of each starch fraction were determined at 25 C by the static method: starch samples were let to equilibrate over different salt solutions in vacuum dessicators. All the samples lost water and the equilibrium was judged to
180
M. Haros et al. / Journal of Food Engineering 60 (2003) 177–184
have been attained when the weight of the sample between successive measurements changed by less than 2 mg over 72 h. Dry weigh of corn starch was measured gravimetrically after vacuum drying over P2 O5 at 70 C until constant weight. 2.9. Thermal properties of starch Differential scanning calorimetry (DSC) was used to study the thermal properties of starch isolated from dried and undried corn samples. A PL-DSC (Polymer Laboratories, UK) calorimeter equipped with a thermal analysis data station was used. Samples (5 mg) of starch were weighed into aluminum sample pans (accuracy 0.01 mg). Deionized water was added into the pan to obtain water: starch ratio 3:1 and the pan was then hermetically sealed. Samples were heated at the rate of 10 C/min from 30 to 110 C. Heat of gelatinization DH , onset temperature T0 , peak temperature Tp and end temperature Te were computed. All samples were run in triplicate.
3. Results and discussion 3.1. Water absorption and diffusion coefficient For flint and dent corn, the equivalent spherical radii of whole kernels were 4.35 0.04 mm and 4.45 0.05 mm (mean values deviation standard), respectively. The values of maximum length, thickness, and width for flint corn were a ¼ 11:30 0:15 mm, b ¼ 9:53 0:89 mm, and c ¼ 6:01 1:75 mm; for dent corn were a ¼ 13:25 0:26 mm, b ¼ 9:39 0:32 mm, and c ¼ 4:82 0:68 mm (mean values deviation standard), respectively. The W values for flint and dent corn were 0.78 0.09 and 0.75 0.05 (mean values deviation standard), respectively. Eq. (1) was used to estimate the diffusion coefficients of water in both corn hybrids (dried and control). The saturation moisture content, us , in Eq. (1) was estimated
as follows. When the time becomes large the limiting form of Eq. (1) results u us 6 ¼ expðp2 WFoÞ u0 us p2
ð4Þ
From this equation it can be easily demonstrated that for any set of three moisture levels taken at equally spaced time intervals of duration j, the following expression results: us ¼
ui uiþ2j u2iþj ui þ uiþ2j 2uiþj
ð5Þ
ui and uj being the moisture contents of time i and j. The saturation moisture contents of control and dried corn samples are given in Table 1. These values were then used to estimate D from Eq. (1) using a nonlinear regression method. Diffusion coefficient determinations were replicated three times. It is observed from Table 1 that the values of us decrease with the increase of drying temperature, being this effect more marked for flint than for dent corn. It can also be seen that for the control samples, the value of D for flint corn is lower than for dent corn. It is probable that the main reason for such differences would be the kernel characteristics of both hybrids. For flint type kernels the concentration of starch is higher around the periphery of the endosperm than in the centre, giving the endosperm a hard, external layer. In dent kernels the ratio of hard to floury endosperm is, in general, lower than in flint kernels, and the floury tissue extends to the crown of the endosperm, providing easy paths for moisture penetration. Structural differences in the protein matrix in which starch is embedded may also account for that value (Robuti, 1980). In Table 1 it can also be observed that the values of the diffusion coefficients of the dried samples were higher than the controls. The kernels dried exhibited excessive stress cracks and fissures, which resulted in increased the water absorption rate because the barrier for water diffusion had been reduced. However, the increase of drying temperature from 70 to 110 C
Table 1 Saturation moisture contents and diffusion coefficients of undried and dried flint and dent corn samples steeped in 0.25% SO2 Corn sample
Drying temperature (C)
us (g/g dry basis)a ;b
D 1011 (m2 /s)a;c
Flint
Undried 70 90 110
0.634 0.005 0.614 0.002 0.588 0.007 0.570 0.005
3.8 0.6 5.6 0.9 5.0 0.7 4.7 0.9
Dent
Undried 70 90 110
0.734 0.010 0.715 0.025 0.714 0.005 0.689 0.002
5.7 0.6 7.2 0.2 6.8 1.1 6.5 0.3
a
Mean value standard error, significant a ¼ 0:05. Eq. (5). c Eq. (1). b
M. Haros et al. / Journal of Food Engineering 60 (2003) 177–184
181
produced a decrease in the rate of absorption; for flint corn the D values decrease above 17%, while for dent corn such decreasing was only of 10%. The high drying temperatures tend to damage the endosperm protein provoking irreversible biochemical changes reducing the beneficial effect of SO2 on the absorption rate (Watson, 1991).
3.2. Effect of drying temperature on steeping The comparative effect of drying temperature of kernels on the total solids leached during steeping is illustrated in Fig. 1A and B, from flint and dent, respectively. In these figures it can be seen that the amount of solids released increases with the increase of steeping time. This is due to the action of SO2 on the endosperm matrix of the kernels which provokes solubilization of the endosperm proteins by reduction of disulfide bonds by the SO2 (Neuman, Wall, & Walker, 1984). It is also observed that the level of solids released decreases with the drying treatment. For both hybrids, the amount of solids released in dried samples was in most cases smaller than for the control. The diminishing of endogenous enzymatic activity (Peplinski et al., 1994) together with the denaturation and chemical changes induced by the high drying temperature (French & Kingsolver, 1964; Peplinski et al., 1994) are mainly responsible for such differences.
Fig. 1. Effect of drying temperature on total solids leached during (A) flint corn steeping and (B) dent corn steeping.
Wet-milling performance was evaluated on the basis of recovery and purity of starch fraction. Both param-
Table 2 Effect of drying temperature on starch recovery, protein content of starch wet-milling fraction and steeping index of dent and flint corn Drying temperature (C)
Undried 70 90 110
Flint
Dent a;b
a ;c
Starch recovery (%)
PPS
Steeping index
Starch recoverya ;b (%)
PPSa; c
Steeping indexd
96.5 0.5 92.3 0.4 92.4 0.5 82.0 0.6
0.90 0.03 0.90 0.05 1.31 0.06 5.72 0.03
259 238 215 161
97.5 0.3 97.2 0.4 98.0 0.6 90.0 0.3
1.20 0.03 2.41 0.01 4.00 0.04 4.22 0.04
253 241 223 138
a
Average of four replicates ( standard deviation). All recoveries are expressed on a percentage dry solid. c Percent of protein in starch (mean value standard deviation). d Eq. (3). b
Table 3 Estimated GAB parameters, Eq. (6), for corn starch isolated from undried and dried flint and dent corn Material
Drying temperature
mm (dry basis)
C
k
Flint corn
Undried 70 C 90 C 110 C
0.129 0.002 0.116 0.009 0.122 0.010 0.122 0.012
32.0 1.7 27.9 0.7 19.2 1.2 18.4 0.4
0.58 0.06 0.62 0.04 0.59 0.06 0.58 0.03
Dent corn
Undried 70 C 90 C 110 C
0.118 0.007 0.129 0.011 0.121 0.008 0.132 0.014
30.3 2.8 17.1 2.5 14.9 1.4 15.3 2.4
0.60 0.03 0.54 0.03 0.57 0.04 0.54 0.05
182
M. Haros et al. / Journal of Food Engineering 60 (2003) 177–184
eters are reported in Table 2 for control and dried corn samples. High starch recovery and low protein content in starch were considered as indicators of good milling. Wet-milling results from our 48-h controls (flint and dent hybrids) are compared with other batch steeping milling procedures (Steinke & Johnson, 1991) and wetmilling results from industry (Anderson & Watson, 1982). The starch recovery of undried flint corn was 96.5% that is lower but not significantly different than the value of 97.5% corresponding to dent hybrid (relative error of starch recovery was 1.8%). Weller, Paulsen, and Steinberg (1988) found similar results working with dent corn hybrids of different vitreous-to-floury endosperm ratios. The effect of drying temperature on starch recovery had similar trends in dent and flint hybrids. The increase of temperature reduced starch recovery rates of both hybrids. The starch recoveries drop to the extent of 15.0% and 7.7% for flint and dent corn, respectively, when drying temperature was increased to 110 C. Our starch protein contents for the control samples were higher than the typical industrial practice, 0.4–0.5% (Anderson & Watson, 1982). Even though the wetmilling procedure used here reproduced relatively well, as evidenced by the relatively low standard deviations reported in Table 2, certain subjectivity in the separation protein–starch may account for those differences. We believe, however, that the wet-milling process used in this study served well as the basis for comparing the effect of drying temperature on starch isolation of grain corn. This is illustrated in Table 2 where it can be seen that the increasing of drying temperature leads to more severe reduction in recoverable starch and higher protein contamination. These results correlate well with the values of steeping index also given in Table 2. Brown et al. (1979) found that steeping index less than 200 conducts to unsatisfactory milling performance and difficulties for starch–gluten separation. This is corroborated for the drying temperature of 110 C, which conducted to the lowest starch recovery and highest protein contamination for both of the hybrids investigated. 3.3. Sorptional behaviour of starch The experimental equilibrium moisture contents of corn starch isolated from flint and dent corn at 25 C were obtained by triplicate. The maximum relative error (coefficient of variation) in equilibrium moisture values was 0.65%. For a better quantification of the treatment on the equilibrium values the monolayer moisture contents were calculated from GAB equation (Weisser, 1985): m Ckaw ¼ mm ð1 kaw Þð1 kaw þ Ckaw Þ
ð6Þ
Fig. 2. DSC thermograms of isolated starch of (A) corn (flint) drying and (B) corn (dent) drying at different temperature: (a) undried, (b) 70 C, (c) 90 C, (d) 110 C.
In this equation m and mm are the moisture and the monolayer moisture contents (kg water/kg dry solid), aw the water activity, and C and k the parameters related to the temperature effect through the equation. The
M. Haros et al. / Journal of Food Engineering 60 (2003) 177–184
183
Table 4 Thermal properties of isolated starch from flint and dent corn as affect by drying temperature Material
Endotherm temperature, C, and gelatinization enthalpy, DH (J/g)
Drying temperature Undried
70 C
90 C
110 C
Flint corn
T0 Tp Te DH
69.1 73.8 82.0 13.7
69.1 74.0 82.7 13.0
69.1 74.3 83.5 12.2
70.0 75.5 84.2 11.3
Dent corn
T0 Tp Te DH
69.2 73.2 81.6 11.6
69.2 73.5 82.7 11.3
69.5 75.1 83.2 10.4
70.4 75.6 84.3 10.0
parameters of GAB equation (mm , C and k) were estimated by nonlinear regression and are given in Table 3. The values of mm range to approximately 0.11–0.13 kg water per kg dry solid and represent the water more strongly adsorbed and immobilized at polar sorption sites, located in a fairly rigid structure of starch polymer. There is no marked difference in the monolayer values of the starch isolated from dent and flint corn (untreated) as well as of the isolated from dried corn samples. For both hybrids it is observed, however, certain differences between the C values of the dried and undried corn samples. It appears that the increase of drying temperature tends to decrease the binding energy of the first adsorbed layer, as it results from the less values of C obtained from the dried corn samples in comparison with the undried ones. On the contrary, the properties of the adsorbed water in the multilayer region do not seem to be affected by the drying temperature of the grain, such as it results from the similarity among k-values reported in Table 3.
nization; in both hybrids it is observed a decrease of DH with the increase of temperature. As drying has a negative effect on starch–protein separation (see Table 2), it is likely that the protein remaining in starch should reduce the entrance of water into the granules during gelatinization, avoiding interaction between water and starch. If this were so, a partial wetting of the starch granules would be expected during DSC runs and a loss in the enthalpy of gelatinization. It is interesting to observe that such loss was about 18% for flint corn and 13% for dent corn.
Acknowledgements We acknowledge financial support from the Consejo Nacional de Investigaciones Cientıficas y Tecnicas (CONICET) and Universidad de Buenos Aires, Rep ublica Argentina.
3.4. Thermal properties of starch Thermal transitions of milled starch (in excess water) from flint and dent corn (undried and dried) are shown in Fig. 2A and B. The parameters of DSC thermograms are reported in Table 4. It can be observed that the transition temperatures, onset (T0 ), peak (Tp ), and end (Te ) show a slight but significant tendency to increase with drying treatment. Endotherm peak widths (Te T0 ) were also influenced by drying. For flint corn peak width increased to 14.2 C for corn dried at 110 C; a similar behaviour was found for dent corn. In Table 4 it is also observed that DH -value for flint corn (undried) was higher than the corresponding to dent corn. The large proportion of vitreous endosperm of flint corn in comparison to dent corn (Watson, 1991) together with the fact that vitreous endosperm is more difficult to gelatinize than floury endosperm (Le Brass, 1982) may account for such differences. The drying temperature seems to affect also the enthalpy of gelati-
References AACC (1995). Approved methods of the American Association of Cereal Chemists (9th ed.). Method 44-15A. Anderson, R. A., & Watson, S. A. (1982). The corn of milling industry. In A. Wolf (Ed.), CRC handbook of processing and utilization in agriculture (Vol. 2, Part 1.I, pp. 31–61). Boca Raton: CRC Press. AOAC (1980). Official methods of analysis (13th ed.). Washington, DC: Association of Official Analytical Chemists. Bertoni, M. H., Pereyra Gonzalez, A., & Cattaneo, P. (1994). Estudios sobre frutos de Cucurbita moschata Duch (calabazita). II. Aceite crudo de extracci on de semilla. Caracterısticas y composici on analıtica. Anales de la Asociacion Quımica Argentina, 82, 131–133. Brown, R. B., Fulford, G. N., Daynard, T. B., Meiering, A. G., & Otten, L. (1979). Effect of drying method on grain corn quality. Cereal Chemistry, 56, 529–532. Egan, H., Kirk, R. S., & Sawyer, R. (1987). Analisis Quımico de Alimentos de Pearson (pp. 244–245). S.A. de C.V., Mexico: Compa~ nıa Editorial Continental. French, R. C., & Kingsolver, C. H. (1964). Effect of excessive heat during artificial drying of corn on reducing sugar content and diastasic activity. Cereal Chemistry, 41, 47–58.
184
M. Haros et al. / Journal of Food Engineering 60 (2003) 177–184
Greenspan, L. (1977). Humidity fixed points of binary saturated aqueous solutions. Journal of Research of the National Bureau of Standards––A. Physics and Chemistry, 81(1), 89–96. Haros, M., & Suarez, C. (1997). Effect of drying, initial moisture and variety in corn wet milling. Journal of Food Engineering, 34, 473– 481. 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. Krueger, B. R., Knutson, C. A., Inglett, G. E., & Walker, C. E. (1987). A differential scanning calorimetry study on the effect of annealing on gelatinization behaviour of corn starch. Journal of Food Science, 52, 715–718. Le Brass, A. (1982). Maize drying conditions and its resulting quality for wet-milling industry. In G. E. Inglett (Ed.), Maize: recent progress in chemistry and technology (pp. 95–127). New York: Academic Press. Lorenz, K., & Kulp, K. (1980). Steeping of starch at various temperatures––effects on functional properties. Starch/St€arke, 32, 181–186. Luikov, A. V. (1968). Analytical heat diffusion theory. London: Academic Press. Mac Masters, M. M., Finkner, M. O., Holzapfel, M. M., Ramser, J. H., & Dungan, G. H. (1959). A study of the effect of drying conditions on the suitability for starch production of corn artificially dried after shelling. Cereal Chemistry, 36, 247–259. Mistry, A. H., Xutian, W., Eckhoff, S. R., & Litchfields, J. B. (1993). Wet-milling characteristics of high-temperature, high-humidity maize. Cereal Chemistry, 70, 360–361. Nerying, A., & Reilly, P. J. (1984). Laboratory wet milling of ensiled corn kernels. Cereal Chemistry, 61, 8–14.
Neuman, P. E., Wall, J. S., & Walker, C. E. (1984). Chemical and physical properties of proteins in wet-milled corn gluten. Cereal Chemistry, 61(41), 353–356. Peplinski, A. J., Paulis, J. W., Bietz, J. A., & Pratt, R. C. (1994). Drying of high-moisture corn: changes in properties and physical quality. Cereal Chemistry, 71, 129–133. Robuti, J. L. (1980). Observaciones de microscopıa electr onica de barrido sobre endosperma de la lınea end ocrina P21 de maız. Boletın de divulgaci on tecnica No. 41. INTA, Instituto Nacional de Tecnologıa Agropecuaria, Estaci on Experimental Regional Agropecuaria Pergamino, Argentina. Shandera, D. L., & Jackson, D. S. (1996). Effect of corn wet-milling conditions (sulfur dioxide, lactid acid, and steeping temperature) on starch functionality. Cereal Chemistry, 73, 632–637. Steinke, J. D., & Johnson, L. A. (1991). Steeping maize in the resence of multiple enzymes. I. Static batchwire steeping. Cereal Chemistry, 68, 7–12. Tolaba, M. P., Aguerre, R. J., & Suarez, C. (1989). Shape characterisation for diffusional analysis of corn drying. Drying Technology, 7(2), 205–217. Watson, S. A. (1991). Structure and composition. In S. A. Watson, & P. E. Ramstad (Eds.), Corn: chemistry and technology (pp. 53–82). St Paul, MN: American Association of Cereal Chemistry. Watson, S. A., & Hitara, Y. (1962). Some wet-milling properties of artificially dried corn. Cereal Chemistry, 39, 35–44. Weisser, H. (1985). Influence of temperature on sorption equilibria. In D. Simato, & J. L. Multon (Eds.), Properties of water in foods (p. 95). Dordrecht: Martinus Nijhoff. Weller, C. L., Paulsen, M. R., & Steinberg, M. P. (1988). Correlation of starch recovery with assorted quality factors of four corn hybrids. Cereal Chemistry, 65, 392–397.