Three samples of the maize hybrids (Abotem, Abeleehi and Dorke SR) were obtained from the Crop Research Institute’s (CRI) experimental farm situated at Fumesua, Kumasi. The samples collected were graded and only good kernels were used for the experiment. Each sample was packed separately in an air-tight polythene bag, sealed and stored at 4˚C to maintain moisture and prevent recontamination.

The initial moisture content of the sampled kernels were estimated using the Standard Oven drying method proposed by [10]. Three samples of 5 g each were measured, placed in dishes and numbered. The dishes containing the samples were heated for 4 h at a temperature of 103˚C in the oven. For each sample, the method was repeated three times and the mean value determined.

The principal dimensions of each maize kernels were determined following [11] approach with slight changes. Hundred (100) kernels were randomly selected for each sample and their dimensions taken using a digital micrometer screw gauge. Thus, the major diameter was taken as length, the intermediate diameter as width, and the minor diameter as thickness of the kernel. Length was taken on 100 kernels and width and thickness on 50 kernels as specified by [12].

The equivalent radius for each variety was measured based on the assumption that the volume of the maize kernel could be approximated to the volume of a sphere with radius being half of the diameter of the kernel. Fifty kernels of each variety were immersed in the water and volume of water displaced was recorded. The average volume obtained was equated to the volume of a spherical object

[13] and r in mm calculated from V = 4 3 π r 3 .

Thus,

r = 3 v 4 π 3 (1)

Water absorption behaviour of the hybrids was evaluated essentially by immersing the kernels into water at four different temperatures of 30˚C, 40˚C, 50˚C, and 60˚C in a thermostatic water bath. Three replicates of sample size 5 ± 0.02 g each was separately placed in nylon mosquito net, tied and labelled. The mosquito net containing the samples were separately placed into the bath and removed at predetermined time interval of 1 h. The moisture content (Mc) of the kernels at each interval, was determined from the difference between the initial weight and the final weight. The moisture content at saturation point (SMC) was calculated when there was no significant change between the initial weight and the final weight.

At constant temperature, the diffusion process was assumed to follow the Fick’s second law of diffusion, where Fick’s three-dimensional equation given by:

∂ M ∂ t = D ( ∂ 2 M ∂ x 2 + ∂ 2 M ∂ y 2 + ∂ 2 M ∂ z 2 ) (2)

where, M represent instantaneous Mc at time t and D_f the diffusion coefficient. A solution proposed for a spherical shaped object with radius r presented by [13] and [14], as

MR = M − M i M e − M i = 1 − ( 6 π 2 ) ∑ i = 1 ∞ ( 1 i 2 ) exp ( − D i 2 π 2 t r 2 ) (3)

where, MR represent moisture ratio,

M_i and M_e being the initial and the equilibrium moisture content respectively,

i the number of terms in the summation,

Di, the effective water diffusion coefficient,

r is the radius of the kernels and

t the soaking time.

The problem was then formulated as a root finding equation:

f ( D i ) = MR expt − MR calc

and searched for the Di value which made the function f(D_f) = 0. With MR_expt and MR_calc representing moisture ratio from experimental data and the model predictions respectively.

Activation energies (Ea) of the hybrids were evaluated using an Arrhenius- type equation [15]:

D f = D o e − E a / R T (4)

where:

D_f, (m²/s), D_o (m²/s), Ea (kJ/mol), R (8.314 J/mol K), T (K) represent diffusion coefficient, diffusion constant, activation energy, gas constant and absolute temperature respectively.

The D_o and the slope (E_a/R) were obtained from the linear regression analysis of D_f vs. 1/T and (Ea) of the varieties were calculated.

The physical properties, initial moisture contents of the maize hybrids investigated are presented in Table 1.

Generally, the values presented in Table 1 agree with those reported by [16]. From Table 1, the surface (SA) varied from 521.34 mm² (Dorke SR) to 537.62 mm² (Abeleehi). In general, for different cultivars of a cereal, one will expect an inverse relationship between kernel size and rate of water absorption, since a large kernel produce smaller surface area per unit mass [7].

Water absorption curves of dry kernels at four soaking temperatures are presented in Figures 1(a)-(c).

Characteristically, the samples showed a normal moisture absorption behaviour. The initial water absorption was rapid, and was followed by a slower rate before it asymptotically reached the SMC at later stage. The SMC was reached at 12 h for water temperature of 30˚C but the soaking duration was reduced to 8 h when the temperature was raised to 60˚C as shown in Table 2.

The water absorption rate of the kernels increased as soaking temperature increased. A. Addo et al. [4] generated similar curves for two varieties of maize— Obatanpa and Mamaba. Studies by [13] [17] and [18] documented similar curves when they investigated the soaking behaviour of red kidney beans, wheat, selected Turkish legumes, and barley respectively.

As the soaking temperature increased, water absorption capacity also increased for all the hybrids although the extent and rates of increment were not the same. This could be attributed to differences in chemical composition of the hybrids used. This assertion is in line with reports by other researchers such as [19] and [20] after they studies on water absorption behaviours of rice. Dorke SR recorded a relatively higher moisture gain at most of the temperatures. This may be due to higher carbohydrate as reported by [4]. Abotem recorded the lowest moisture gain at 30˚C and 40˚C whiles Abeleehi recorded the lowest moisture gain at temperature 60˚C. The rate of water uptake was influenced by temperature. The greater the soaking temperature, the higher the absorption rate due to increased rate of diffusion. The temperature effects reported in this study agrees with other published studies [4] [13] [15] [19] [21] [22]. S. M. Shafaei et al. [23] made similar observation when they studied water absorption behaviours of chickpea and soybean.

J. Tang et al. [24] indicated that small kernels have larger water absorption capacity because of an increased specific surface area available for absorption. This phenomenon was exhibited in the present study.

Dorke SR, with a smaller kernel size compared to Abeleehi, and Abotem had a higher absorption capacity of water as expected. On the contrary, Abeleehi kernel is relatively larger and should have had lower water absorption capacity, but had the highest water absorption capacity at 40˚C. The current observation may be attributed to nutrient composition differences and the severity of dis-orderness resulting from differences in levels of temperature treatment. The present

*×10⁻¹⁰ m²/s, D_f: Diffusion coefficient, M_s: Saturation moisture contents.

observation suggest that Abeleehi may possibly have higher starch content than the other two maize hybrids. A. Addo et al. [4] documented similar pattern when they reported that higher starch content in Obatanpa was responsible for its higher water absorption capacity although Obatanpa kernel is bigger than Mamaba. This observation confirms earlier report by [25] who indicated that temperature and grain characteristics are the major factors influencing SMC values.

The generated data on water soaking of the maize hybrids were plotted against the square root of soaking time. Based on Equation (4), the effective water diffusivities of the hybrids during soaking were evaluated.

The SMCs used in Equation (4) and the evaluated values of D_f at various temperatures are presented in Table 2.

Generally, the experimental data for the grain component was able to fit the diffusion equation with high accuracy. The analysis of the linear regression is also presented in Table 3. The estimated values of coefficient of determination R² of 0.93 to 0.98 suggest that the experimental data fitted into the equation very well. The D_f for the hybrids varied from (2.54 - 3.49) × 10⁻¹⁰ m²/s for Abotem; (2.64 - 3.59) × 10⁻¹⁰ m²/s for Abeleehi; and (3.21 - 4.2) × 10⁻¹⁰ m²/s for Dorke SR. The D_f values documented by [4] for Mamaba and Obatanpa varieties are lower than values obtained in this study. Kernel’s physical characteristics, internal structure, physiochemical and nutrient composition may have accounted for the differences. This observation confirms earlier findings of other researchers such as [4] and [26]. In addition, earlier report by [27] indicated that physiochemical and nutritional components as well as amylose content play key role in predicting rate of moisture gain in rice.

The rates of water absorption in Abotem, Abeleehi, and, Dorke SR are illustrated in Figures 2(a)-(c).

The figures indicate that increasing soaking temperature produces a higher water absorption rate. This phenomenon may probably be due to the increased temperature resulting in more open structures which favoures rapid hydration. Also, the figures show the existence of linear relationship between moisture increase and the square root of time (Time^1/2) during water absorption. This observation confirms earlier report by [28] that the rate of water absorption is a function of time and temperature.