Evaluation of energy consumption in drying African oil bean seeds using response surface methodology

Abstract This research was set out to evaluate the specific energy consumption (SEC) in drying African oil bean seed using convective hot air and microwave oven dryers. The process parameters investigated in thin-layer drying were temperature (40, 50, 60, 70 °C), microwave power (450, 720, 900 W), slice thickness (2, 3, 5 mm) and treatment (untreated (UT), brine pretreated (BT) and sucrose pretreated (ST)). Response surface methodology (RSM) was used to evaluate the effect of oven temperature, microwave power, slice thickness, and treatment on the SEC of African oil bean seed. The result revealed that the SEC for African oil bean seed ranged from 21.12 to 57.33 k Wh kg−1 and 90.76 to 261.23 k Wh kg−1 for microwave and oven drying, respectively. The study also found that the SEC decreases with increase in slice thickness in both microwave and oven dryer. The microwave oven method reduced SEC of African oil bean seed by 22% compared to convective hot air oven drying method. The optimum SEC was obtained at 4 mm material thickness, 53.57 °C hot air oven temperature and 814.32 W microwave power. Among the treatments, BT sample is comparatively the best treatment for AOB in order to save drying energy consumption.


Introduction
Water molecules are attracted to many products because its hydrogen and oxygen atoms are made up of positive and negative electrical charges, respectively.The easy bonding of water to other molecules is the reason why most agricultural plant materials contain considerable amount of water at harvest.High moisture content favours the growth and thriving of micro-organisms that are responsible for product's deterioration shortly after harvest.Drying, which is one of the ways to extend shelf-life of products by systematic moisture reduction, requires a lot of energy.Energy is needed to activate the drying process.Energy is, also, required to move the water that is tightly held within the core of the material to the surface, and then to the drying environment.
Hot air oven and microwave drying are types of convective drying (Jeevanandam & Danquah, 2020;Orsat, Changrue, & Vijaya Raghavan, 2006).Microwave drying uses electromagnetic radiation (frequencies between 300 MHz and 300 GHz) to penetrate moist materials to be dried, and changes it into heat that is responsible for vaporizing the moisture from the materials.It has advantages of fast volumetric heating, fast drying rate, low drying time, reduced energy consumption, quality dried product, good sanitation, lower operating cost and ease of operation (Orsat et al., 2006).On the other hand, convective hot oven has a simple procedure, reliable result and determines the moisture content of large sample volumes.
The African oil bean seed serves as an essential food supplement in Southeastern Nigeria.The crop is rich in protein (28.25%), lipid (44.20%) and carbohydrate (21.93%) (Okwu & Aluwuo 2008).It also contains essential minerals and vitamins in appreciable quantities (Ikhuoria, Aiwonegbe, Okoli, & Idu, 2008;Okwu & Aluwuo 2008;Odoemelam 2005).Postharvest losses of the seeds are prevalent due to the seed's high moisture content at harvest and unavailability of proper storage systems.Igbozulike, Ndirika, and Simonyan (2021) suggested drying as a possible way to extend the shelf life of African oil bean seed.
Response surface method (RSM) is a reliable technique used in studying the correlation between dependent variable(s) and independent variable(s) (Kamal et al., 2021).RSM ability to optimize the response function and to predict the future responses makes it widely acceptable as a useful tool for optimization by researchers.Sarabia and Ortiz (2009) posit that RSM aids decision making in researches that involve uncertainties by eliminating or limiting them when analyzing the empirical data set.
Therefore, the objective of this work was to investigate the specific energy consumption in drying African oil bean seed slices in microwave and hotoven considering various drying process parameters using RSM.

Materials and methods
A bulk sample of African oil bean seeds was purchased from Eke Ita, a popular local foodstuff market in Imo State, Southeastern Nigeria.The location of the market is latitude 5 21'N and longitude 7 12'E.The samples were cleaned, boiled for 2 h, dehulled and sliced into 2, 3 and 5 mm thicknesses.Slice samples were boiled for another 2 h and soaked for 6 h in tap water.The boiling and soaking of samples are necessary for debittering and removal of poisonous alkaloids (Enujiugha, Badejo, Iyiola, & Oluwamukomi, 2003;Olasupo, Okorie, & Oguntoyinbo, 2016).Pretreatment of prepared samples was done by soaking in 10 g and 5 g/100 ml (%) concentrations of brine and sucrose solutions respectively, at room temperature (30 C) for 1 h.The lower concentration of sucrose solution was purposively selected because sucrose is not a preferred condiment in preparing African oil bean delicacies.After soaking, the samples were drained of the solution, blotted with paper cloth and labelled accordingly.The brine and sucrose pre-treated samples were labelled BT and ST respectively, and the untreated sample, which served as the control for the experimentation, was coded UT.

Determination of energy consumption for hot-air oven drying
The total energy consumption (kWh) for oven dried sample was calculated using Equation (1) (Nwakuba et al., 2016): where A is cross sectional area of the holder (m 2 ), q a is air density (kg/m 3 ), C a is the specific heat of air (kJ/kg/ o C), DT is the temperature difference, and D t is the total time for drying each sample (h).
The specific energy consumption (SEC) for drying 1 kg of African oil bean seed slices is given as Motevali, Minaei, Khoshtaghaza, and Amirnejat ( 2011): where E kg is the required specific energy (kWh/kg), W 0 is the mass of sample (kg), E t is the total energy requirement in each drying phase (kWh).

Determination of energy consumption for microwave oven drying
The total energy consumption in microwave oven drying was determined using Equation (3) (Afolabi et al., 2014): where E c is the total energy consumption (kWh) during the drying, P is the microwave power density (kW) and t is the drying time (h).
The specific energy consumption (kWhkg À1 ) in microwave drying is by the following equation ( 4): where E c is the total energy consumption (kWh) during microwave oven drying and M o is the total mass of evaporated water (kg).

Experimental design
The historical-data RSM design was employed using Design Expert software 10.0.1 (STAT-EASE Inc., Minneapolis, USA) to optimize the specific energy consumption (SEC) in drying African oil bean seeds using convective hot air oven and microwave oven.Box-Behnken, Central-composite, Historical-data are some of the RSM designs used in RSM modeling.However, in Historical-data, the researcher defines the design points using all the available experimental data (Jeirani et al., 2013a).Historical-data, also, has the advantage of no limitation on the number of design factors in an experiment, and the factor settings and responses of an existing data set can be imported directly to a blank design layout (Jeirani et al., 2013b).The independent variables for the experiment (Table 1) were as follows: temperature (40, 50, 60, 70 C), microwave power (450, 720, 900 W), material thickness (2, 3, 5 mm) and the dependent variable (response) is the SEC.

Optimization of processing parameters for SEC
Numerical optimization was carried out to determine the optimal specific energy consumption (SEC) for drying African oil bean seeds in hot air oven and microwave dryers.Design Expert software 10.0.1 (STAT-EASE Inc., Minneapolis, USA) was used for the design and evaluation of the optimization process under the various treatments.

Results and discussion
The AOB seed was dried from an initial moisture content of 12.3% dry basis to 4.2% dry basis.The time taken to complete the drying in microwave was 36 min, whereas convective hot air oven took 17.5 h.The Historical data RSM design and the response for convective hot air drying at different treatments are shown in Table 2.The SEC was found to be in the range of 137.83 À 261.27 kWh/kg, 93.14 À 235.44 kWh/kg and 90.16 À 244.04 kWh/kg for UT, BT and ST samples respectively in convective hot-oven drying method.The SEC was found to increase with an increase in drying air temperature of the African oil bean slices (Table 2).This is as a result of an increase in the thermal gradient between the hot drying air medium and the sample products, as well as an increase in the kinetic energy of the hot convective air to initiate rapid moisture diffusion from the samples and resistance to capillary moisture transport.Increasing the air temperature entails increasing the amount of power drawn by the heater and fan units of the oven dryer per hour.Increasing the sample thickness at any given drying air temperature reduces the SEC.This is probably a result of the gross amount of mass transfer rate since moisture diffusion increases with an increase in capillary distance (sample thickness).Since SEC is a ratio of total useful energy to the amount of moisture removed, drying thicker samples at constant air temperature diminishes the SEC value.Similar result was reported by Onwude, Hashim, Abdan, Janius, and Chen (2018) for sweet potato dried in hot air oven at 50-70 C where 6 mm slice sweet potato has SEC value range of 220.39 À 326.47 kWh/kg and 4 mm slice sweet potato have SEC value range of 227.39 À 337.79 kWh/kg.Also, the SEC of African oil bean seed sample varies with pretreatment.
The BT sample had the least mean SEC of 167.51 kWh/kg, followed by ST (168.93 kWh/kg), and the UT (211.99 kWh/kg).This indicates that the relative moisture extraction capabilities of the pretreatment solutions tend to reduce the drying time, hasten the drying rate, and reduce the specific energy consumption of the process.
The SEC of African oil bean seed in microwave oven drying was found to be in the range of 32.88 À 46.15 kWh/kg, 21.12 À 34.24 kWh/kg and 31.35À 57.33 kWh/kg for UT, BT and ST samples respectively (Table 3).The average SEC for UT, BT,and ST samples is 40.26,27.25 and 41.64 kWh/kg respectively.The SEC, also, decreased with increasing material thickness in microwave drying method.
The influence of drying process variables on the SEC is shown in Figure 1 for convective hot-oven and Figure 2 for microwave drying.
The RSM plot (Figure 1a) revealed that the process variables influenced the SEC values for the different sample treatments.At 40 C temperature, the SEC for UT sample increased from 137.83 kWh/kg to 237.56 kWh/kg as the material thickness decreased from 5 mm to 2mm.Similarly, at the same temperature, the SEC-BT increased from 93.14 to 118.5 kWh/kg (Figure 1b), and from 90.76 to 160.82 kWh/kg for ST samples (Figure 1c) as the material thickness decreased from 5 mm to 2 mm.
The analysis of variance (ANOVA) at 5% revealed that all the process parameters are significant model terms for SEC of African oil bean seed (Table 4).The Model F-value of between 23.57 and 80.28 imply that the models are significant because there is only a 0.07% to 0.22% chance that an F-value this large could occur due to noise.The temperature (T), power density (P) material thickness (t) are significant model terms (Table 4).The coefficient of determination (R 2 ) values obtained for SEC of African oil bean seed are above 0.95 in all the drying process  variables.The predicted coefficient of determination (Pred.R 2 ) for all the variables are in reasonable agreement with the adjusted coefficient of determination (Adj.R 2 ) since their differences are less than 0.2.The adequate precision ratio (Adeq.Pre.) greater than 4 is an indication of adequate signal, and the Adeq.Pre.obtained are in the range of 15.03 to 23.13.The high standard deviation observed in hot air oven drying could be as a result of heat loss during the intermittent opening of the drying oven to weigh the samples at 30 min interval.
The model equations to predict the SEC of UT, BT and ST African oil bean seed for hot-oven drying are given in Equations ( 5), ( 6) and ( 7), respectively.
where T is the oven temperature in C and t is the thickness of sample in mm.Similarly, the model equations to predict the SEC of UT, BT and ST African oil bean seed for microwave drying are given in Equations ( 8), ( 9) and (10) respectively.
where P is the microwave power density in W and t is the thickness of sample in mm.The quadratic model predictions and the actual experimental SEC values of African oil bean seed in convective hot-air oven and microwave showed good correlation (Figures 3 and 4).This implies that the model equations ( 5)-( 7) and equations ( 8)-( 10) are adequate in predicting the SEC of African oil bean seed drying in oven and microwave, respectively.Drying methods have been revealed to affect the SEC of agricultural products (Ismail et al. 2019;Ismail and Kocabay 2016;Kaveh et al., 2021;Motevali et al., 2011;Salarikia, Miraei Ashtiani, & Golzarian, 2017;Ste Rpie n, Gorzelany, Matłok, Lech, & Figiel, 2019;Zambra et al., 2021).This is true for African oil bean seed where the SEC in microwave drying is reduced by approximately 22% when compared to convective hot-oven drying.Also, it has been established that pretreatment reduces the SEC of products (Chayjan, Peyman, Esna-Ashari, & Salari, 2011;Gazor, Maadani, & Behmadi, 2014;Jahanbakhshi, Kaveh, Taghinezhad, & Rasooli Sharabiani, 2020;Orikasa et al., 2018).The average SEC of African oil bean seed is low in the pretreated samples than the untreated samples.This means that pretreatment reduces energy consumption in drying African oil bean seeds.
The optimum SEC gotten for convective hot air oven were 194.15, 179.30 and 160.12 kWh/kg for UT, BT and ST samples respectively, while the optimum input variables were temperature of 53.57C and material thickness of 4 mm with desirability index of 1.000%.In microwave drying, the optimum SEC obtained were 41.76, 29.62 and 48.83 kWh/kg respectively for UT, BT and ST samples with a desirability index of 1.000%.Microwave power of 814.32 W and material thickness of 4 mm were optimum input variables for microwave drying of AOB seeds.To validate the model results, the optimal conditions were experimentally evaluated thrice in the laboratory.The empirical data obtained gave the average SEC for hot air oven as 192.87, 179.06 and 161.66 kWh/kg for UT, BT and ST samples respectively.Whereas for microwave drying, the empirical results were 40.564, 29.755 and 48.324 for UT, BT and ST samples respectively.

Figure 1 .
Figure 1.Response surface plot of the effect of temperature and thickness on SEC for (a) UT, (b) BT, (c) ST samples.

Figure 2 .
Figure 2. Response surface plot of the effect of microwave power and thickness on SEC for (a) UT, (b) BT, (c) ST samples.

Figure 3 .
Figure 3. Plot of quadratic model predictions versus experimental values of SEC for oven drying (a) UT, (b) BT, (c) ST.

Table 2 .
Historical data experimental design of the independent variables and the observed response values for convective hot air oven drying.

Table 3 .
Historical data experimental design of the independent variables and the observed response values for microwave drying.

Table 4 .
ANOVA summary of statistics indicators for SEC.