Optimization of microwave-assisted biodiesel production from watermelon seeds oil using thermally modified kwale anthill mud as base catalyst

A heterogeneous catalyst was developed from raw Kwale red Anthill mud by thermal treatment in a muffle furnace at 900 °C for 4 h. The resulting heterogeneous catalyst was highly porous with a surface area of 42.16m2/g, possessing excellent stability as well as high catalytic activity. Central Composite Design and Machine Learning approach (Python code) were applied to model and optimize biodiesel yield from extracted watermelon oilseed. Highest biodiesel yield of 93.41 wt% was obtained under the experimental conditions of 4min duration, 350 W microwave power, 4 wt% of catalyst, and MeOH/oil ratio of 8:1 based on Central Composite Design rotatable. The optimum value of the biodiesel yield from Machine Learning was 91.7 wt%, showing a marginal performance over the Central Composite Design rotatable value (91.6 wt%) at the optimized conditions of 3 min, 280 W, 3 wt% catalyst loading and MeOH/oil molar ratio of 6:1. The correlation of the coefficient (R2) of the model was 0.9827 for Central Composite Design rotatable while the R2 of the Machine Learning model was 1.0. Thus, python coding in terms of prediction and accuracy of biodiesel yield was superior to Central Composite Design rotatable, even though both models provide a reliable response within the region of data analyzed. The Gas Chromatography-Mass Spectroscopy of the biodiesel produced revealed the presence of both saturated and unsaturated fatty acid methyl esters. Biodiesel properties from watermelon seed oil transesterification fall within the recommended standard for biodiesel fuel. This study concluded that an effective green biowaste catalyst generated from earthen waste could enhance biodiesel production from watermelon seed oil, hence, ensuring sustainability and economic feasibility for biodiesel industries.


Introduction
Nearly all human activities are driven by energy or application of energy in one form or another. Development at local and global a base catalyst, and the transesterification carried out using a hotplate with magnetic stirrer [31,32]. In another study, ultrasonication and microwave-assisted transesterification of watermelon seed oil were performed using crystalline Mn (II) carbonate as a catalyst [33]. The challenges associated with the use of mineral catalysts and conventional heating remain a matter of concern. To have a cost-effective biodiesel production process, a heterogeneous catalyst from a renewable source was applied to investigate biodiesel production from watermelon seed oil (WMSO) in a process intensification device. In this study, a green heterogeneous amphoteric-base catalyst developed from Kwale Red Anthill mud was used to generate fatty acid methyl ester from extracted watermelon seeds oil. Biodiesel production in the microwave-assisted transesterification of WMSO was modeled using CCD and the linear regression model of ML.

Collection of materials
The watermelon seeds were gathered from a restaurant at the Delta State University of Science and Technology in Ozoro, Nigeria. Red Anthill mud was gathered from a farm along the Kwale-Ogwash road, Kwale, Nigeria (5 • 42′ 27″ N). The chemicals and reagents used for the oil physicochemical determination and biodiesel characterization are all analytical grade.

Extraction of the watermelon seed oil
The watermelon seed oil utilized for biodiesel synthesis was extracted at the Petrochemical Technology laboratory, Afe-Babalola University, Ado-Ekiti. Fresh oil was obtained from pulverized oilseeds using a Soxhlet-unit method as n-hexane acts as the solvent of extraction. Typically, watermelon seeds were washed severally to remove dirt using tap-distilled water. The sample after washing was sun-dried until a steady weight was achieved. Dried watermelon seeds used in this study are displayed in Fig. 1. A hand-powered bench grinder was used to mill the seeds after they had been separated from the kernel. In the thimble of a Soxhlet extractor, a measured volume of n-hexane was introduced to solubilize a given mass of pulverized watermelon seeds. A condenser was firmly attached to the other end of the extractor unit to prevent solvent loss, a round-bottom flask containing the extraction solvent was linked to the extractor, and the hotplate was turned on. After the completion of the extraction process, fresh oil was recovered from the solvent at 65 • C. Thereafter, the oil was dried and quantified gravimetrically according to Equation (1).
where W o represents the weight of the extracted WMSO (g) while W S represents the weight of the watermelon seed used for extraction (g). The physicochemical properties (acid value, moisture content, pH, density, kinematic viscosity) of the extracted WMSO was measured using standard methods [34].
Acid value is the amount of potassium hydroxide, measured in milligrams, required to neutralize the free fatty acid present in 1 g of oil. The titration method was used to determine the acid value of WMSO. A known mass of WMSO (g) was measured and equal volume of diethyl ether and ethanol were added to the measured oil. Two drops of phenolphthalein indicator were added to the oil solvent mixture, and the solution was titrated against 0.1 M KOH in burette. The acid value of WMSO was obtained using Equation (2). . 1. Watermelon seeds.
S.C. Iweka et al. where: V = volume of KOH used, M = Molarity of KOH, W = Weight of oil.

Preparation of KRAM catalyst
The Kwale Red Anthill Mud (KRAM) used for catalyst synthesis was collected from Kwale-Ogwash road, Kwale. This catalyst precursor (Fig. 2) was an abandoned earth soil rich in minerals. The harvested red anthill mud was first sifted to remove dirt and stones. The anthill mud was broken up into little pieces, washed with distilled water, and dried at 120 • C for 4 h. Using a mortar and pestle, the dried red anthill mud was grounded into a fine powder, and was sieved to sizes ranging between 150 and 220 mm. Then, the calcination of resulting fine particles was initiated at 900 • C for 4 h using an electric furnace at a heating rate of 10 • C/min [19,21]. The heat-treated ash of red anthill mud after cooling was crushed into a fine powder and sieved. The calcined KRAM sample was kept in a screw-top bottle and used for further analysis as well as biodiesel production.

Characterization of Kwale Red Anthill Mud catalyst (KRAM)
Characterization of synthesized catalyst (KRAM) was carried out to investigate the catalytic potential of the carbonized specimen. The surface morphology and mineral composition of KRAM were examined using a high energy-dispersive X-ray analyzer linked to a scanning electron microscope (SEM, Hitachi SU 3500 scanning microscope, Tokyo, Japan). Using an X-Ray Diffractometer, the phases and crystal structures of the carbonized sample were studied (Shimadzu XDS 2400H). The diffraction pattern was recorded in the 2θ range of 10-80 with a variation of ±0.05 • , and at a scan speed of 2 • C/minute for 120 min. The surface characteristics of the synthesized catalyst were studied using the N 2 adsorption-desorption method. Before analysis, the samples were degassed using the micromeritics flow prep 067 in combination with nitrogen gas for 3 h at a temperature of 473 K. The degassed sample was reweighed, the analysis was completed in the liquid nitrogen-conditioned micromeritics Tristar 3000 V4.02, and the results were recorded. The catalyst surface area, total pore volume, and pore size of the calcined sample were obtained using N 2 an isotherm [35]. The composition of the KRAM sample was examined using an X-ray fluorescence analyzer (Shimadzu EDXRF-702HS). All measurements were taken from the instrument operating at 40 kV and 18 mA. The collimator of choice was 10 mm, with a 100-s counting period. The Shimadzu EDX software program was used to extract the number of counts per second (cps/μA) for element Kα from the sample X-ray spectrum.

Transesterification of watermelon oil
The esterification of extracted oil was performed since the watermelon oil has an acid value of 9.16 mg KOH/g. Esterification of WMSO was performed at the following conditions; molar MeOH/WMSO of 7:1, H 2 SO 4 concentration of 1 wt% for 2 h at a temperature of 60 • C. A 500 mL two-neck glass reactor was used for the transesterification of the esterified WMSO in a modified microwave (model: H20MOW Hisense). To the modified device, a reflux condenser system was attached to the glass reactor to prevent evaporation of the reaction mixture, and an external stirrer having two blades controlled by a 12v electric motor (12v HUPE DC Stabilizer) was also attached. A CCRD design was used to model WMSO transesterification. The transesterification of WMSO was carried out according to the stipulated conditions in the experimental design. After the finalization of the transesterification, the resultant biodiesel-glycerol mixture was subjected to centrifugation at 8000 rpm for 7 min. The product mixture containing biodiesel was washed with warm distilled water three times. Equation (3) was used to compute the biodiesel yield.

Biodiesel physical and chemical properties
Biodiesel synthesized from the transesterification of WMSO was tested to ascertain its purity and quality. Standard techniques were used to determine the biodiesel's physical and chemical characteristics, including its specific gravity, acid value, kinematic viscosity, density, caloric value, and cetane number [34].

Physiochemical properties of the Extracted oil from watermelon seed
The physicochemical properties of the extracted oil from watermelon seed are presented in Table 1. The oil yield (47.61 wt%) from the watermelon seeds is similar to oil recovery from Sesamum indicum seeds (48%) [36]. The acid value of the oil which is an index of free fatty acid content was low, however, transesterification of the extracted oil would require a pretreatment step to be carried out. The WMSO possessed excellent properties which makes it a potential feedstock for biodiesel production. Properties such as kinematic viscosity, density, pour point, cloud point, and pH indicate that the extracted oil could be easily converted to biofuel. The saponification value of WMSO was less than 200 mgKOH/g which indicates that the extracted oil has a low tendency to form soap during the transesterification process. The peroxide value of the oil is less than 10 mg/g which depicts that the oil has high resistance to peroxidation during storage. The iodine value of WMSO is high, indicating that the oil is semi-dry oil.

Brunauer-Emmet-Teller (BET) surface area
The surface area, pore volume, porosity, and pore diameter of calcined Kwale Red Anthill mud were ascertained using the N 2 adsorption-desorption technique as presented in Table 2. The surface area and pore volume in this study are comparable to the value obtained in Ref. [37] for bentonite clay which is 60 m 2 /g and 0.0294 cm 3 /g respectively.

EDX result of calcined KRAM
The EDX spectrum of KRAM is shown in Fig. 3. Chemical analysis of the synthesized catalyst shows that KRAM catalyst was dominated by Silicon, Aluminum, Calcium, etc. Table 3 shows that the group 1 element (Na) was present in very small amounts. The main minerals in the sample are silicon, aluminum, and calcium as they formed the majority of the KRAM catalyst composition. Similar findings have also been reported for termite hill catalyst [20], and natural clay [38,39]. Silicon a metalloid has the largest weight fraction, which is indicative of the catalyst precursor. The main metals (Al and Ca) in the catalyst can readily give electrons to other non-metals. And these metals in addition to Silicon (Si), a metalloid makes ideal catalysts.

XRD results
The crystalline phases of the synthesized catalyst are revealed using XRD analysis (Table 4). The XRD pattern indicated that three crystal structures were present, and the phase is dominated by Kaolinite. There are multiple sharp diffractions in the calcined KRAM, notably peaks at 2θ = 13.00 • , 30.00 • , 36.21 • , and 44.00 • indicates the existence of kaolinite. The peaks at 2θ = 17.00 • , 39.22 • , and 58.00 • are assigned to goethite structure, whereas the peaks at 2θ = 28.98 • , 31.15 • , 37.18 • , 50.26 • , 62.71 • , and 76.00 • are attributed to quartz structure (Shimadzu XDS 2400H). The XRD measurement and the EDX result both support the successful heat dispersion on the red anthill mud. The diffraction of XRD using Origin Software is depicted in Fig. 4.

Results of XRF analysis
XRF measurement of calcined red anthill mud shows the mineralogical components of the sample. Silica (SiO 2 ) and alumina (Al 2 O 3 ) are two of the sample's most prevalent oxides, according to Table 5. This result having SiO 2 and Al 2 O 3 as the major oxides are consistent with the chemical analysis of calcined termite hills and are close to the % composition reported by Ref. [20]. Due to the presence of amphoteric oxides, the synthesized KRAM could be assumed to have a dual-active site. The synthesized catalyst also contained strong basic oxides (MgO, Na 2 O, K 2 O, and CaO) with a total of 7.94 wt%, the presence of different oxides on the catalyst surface would provide adequate catalytic sites for triglycerides-methanol during transesterification [40].

Modeling results of WMSO transesterification
Biodiesel from watermelon seed oil and methanol in the presence of calcined Kwale Red Anthill Mud was modeled using a rotatable design of CCD. The irradiation time, irradiation power, catalyst loading, and MeOH/Oil ratio are the four design variables modeled to   investigate their effects on the biodiesel yield. The CCD model generated thirty (30) experimental runs, and the response of each run was presented in Table 6 in a microwave-assisted transesterification process. The highest biodiesel yield (93.41 wt%) was observed at the following process conditions; time of 4 min, 350 W irradiation power, catalyst amount of 4 wt%, and MeOH-oil ratio of 8:1. On the contrary, the lowest biodiesel yield (86.38 wt%) from WMSO transesterification was obtained in run 13 at the following conditions; time of 4 min, irradiation power of 210 W, catalyst amount of 4 wt%, and MeOH-oil molar ratio of 4:1. The effect of the irradiation power and methanol to oil molar ratio seems to be the major driver of this transesterification process. The significance of the process conditions investigated in this study and the degree of influence on biodiesel yield within the region considered was examined using Analysis of Variance (ANOVA). ANOVA obtained after regression analysis of the biodiesel yields from WMSO using the calcined KRAM was depicted in Table 7. F-value of 60.72 and a p-value ≤0.0001 indicate that the transesterification process model is significant. Any term having a p-value ≤0.05 is adjudged to be a significant term [6,41]. The process factors such as time, irradiation power, MeOH-oil molar ratio, and catalyst amount are all significant. Meanwhile, some interactive terms such as AD and quadratic term B 2 are insignificant, because their values are greater than 0.1. The model's lack of fit is insignificant relative to pure error and the lack of fit has an F-value of 0.49 which imply that 84.54% possibility of being caused by the noise.
The statistical parameters presented in Table 8 were used to assess the model's fitness. Standard deviation of the analyzed data is close to zero which suggests minimum variability in experimental results. Likewise, experimental results were consistent as shown by the proximity of the R 2 , Adjusted R 2 , and Predicted R 2 . Further supporting the validity of the result is the R 2 of 0.9827, which demonstrates that the model can account for 98.27% of the variability found in the model response. The Adjusted R 2 and Predicted R 2 are reasonably in agreement, with the difference between these parameters being less than 0.2, indicating that there is a strong correlation between the predicted values and the experimental values obtained for WMSO transesterification. Additionally, an adequate precision of 29.4005 indicates that the model has sufficient signals to navigate the design space, a value greater than 4 is desirable, and hence the model can be used for the optimization of WMSO transesterification.
The regression equation which relates the mathematical relationship between the model response and the four independent variables is expressed in Equation (4) in terms of coded variables.
where, BY (biodiesel yield) is the response, +108.56167 (intercept term), and 2.55417, 0.052774, 2.83417 and 1.09625 are linear term coefficients. The independent variables A, B, C, and D represent time, irradiation power, catalyst amount, and MeOH/WMSO molar ratio, respectively. The degree of influence of each independent variable on the biodiesel yield is more positive than negative. The perturbation plot (Fig. 5) shows the model response against the parameter deviation, indicating the relative degree to which the variables influence biodiesel yield in an ideal situation and away from the referenced point. It can be seen that B and A had a higher  deviation than D and C, this observation is consistent with the ANOVA results. Also, the point of intercept is the optimum yield (91.6 wt %) obtained from actual/input factors of 3 min, 280 W, 3 wt% of catalyst, and MeOH/oil ratio of 6:1 corroborated by run 3 in Table 6. Moreover, the integrity of the data obtained and the chosen model are further understood by the plot of predicted biodiesel yield against actual biodiesel yield (Fig. 6). All the data points nearly aligned on the line of fitness, showing a good approximation of the model. This further corroborates the model's high coefficient of correlation obtained in the statistical parameter of the ANOVA. The 3D surface plots generated from the regression equation are represented in Figs. 7a-f. The interaction between variables within the region considered and the optimum biodiesel yield is graphically explored using three-dimensional surfaces. In Fig. 7a, biodiesel yield increases with an increase in time and irradiation power. The increase in both irradiation power and time resulted in increased biodiesel yield. High microwave energy combined with increased time drives the transesterification in a forward direction. High   biodiesel production was boosted by microwave-assisted transesterification's fast reaction time [6]. This is corroborated by the red colour which is a sign of the highest biodiesel yield. Biodiesel production as a result of catalyst and time was average, as indicated by the green colour in Fig. 7b. The production of biodiesel does, however, only slightly rise as time and catalyst amounts are increased. In Fig. 7c, an increase in time and MeOH/oil molar ratio leads to a slight increase in biodiesel yield. And the green colour symbolizes the optimum biodiesel yield. In Fig. 7d, an increase in irradiation power and catalyst amount leads to an increase in biodiesel yield, which is better than in Figures b and c. In Fig. 7e, an increase in irradiation power and a decrease in MeOH/oil molar leads to an increase in biodiesel yield. This is corroborated by the yellow colour, which is a higher yield than green colour. Fig. 7f shows that biodiesel yield increases with decreasing catalyst amount and MeOH/oil molar. Thus, the interaction of both time and irradiation power has a positive effect on biodiesel yield as it increased with an increase in both variables. Note: Red colour denotes the highest yield, while yellow colour denotes a decent yield that is just slightly above average. Blue colour indicates low output, whereas green colour indicates average yield.

Predictive results from python coding
The machine learning techniques could monitor and control biodiesel systems in real-time to enhance production efficiency [42] superior to traditional modeling. Results obtained for biodiesel yield within the data considered for each variable from the python coding algorithms are presented in Table 9. The Python language developed to obtain biodiesel yield using the process variables (time, irradiation power, catalyst amount, and MeOH/oil molar ratio) is presented in the appendix. The predicted biodiesel yields from machine learning are compared to the actual biodiesel yields and the predicted values from the CCD model. The comparison of which modeling tool indicates that the linear regression model technique of machine learning gives a better result as compared to CCD rotatable. Thus, machine learning generates better statistical values and visualization plots mathematically than CCD. Although the linear regression model technique of ML is superior to CCD in modeling and optimizing biodiesel yield, both can accurately forecast the dynamics of biodiesel generation using microwave-assisted transesterification methods. The extra information (Appendix) contains the ML used in this study, which can be recreated by the user's actual demand.
The model performance for biodiesel yield from WMSO was validated by R 2 , MAE, and RSME. The statistical parameters generated from ML are displayed in Table 10. The chi-squared value and the associated p-value indicate that the biodiesel yield and the variables are statistically significant. The relatively low value of MAE and RSME indicate that the errors are unbiased and it follows a normal distribution. This could mean that the datasets are accurately fitted. The regression model generated as well as the significance of model terms and quadratic terms are presented in Table 12. Similar to the ANOVA results obtained from CCD rotatable modeling, the model terms are all significant. The correlation of the coefficient shows no noticeable variation between the actual and generated biodiesel yields. The R 2 and Adjusted R 2 values from the CCD rotatable are similar to the values of the OLS regression model of ML (Table 12). Also, the average yield (89.81 wt%) of the ML is the same as the mean value of the biodiesel yield from CCD rotatable as corroborated in Table 8.
Despite the fact that the ML standard deviation (1.816937) in Table 11 is higher than the CCD rotatable standard deviation Average yield = 89.81 wt%.
(0.3356) in Table 8, the ML approach has more advantages. The 3D plot generated by ML is more visually appealing than the one generated from rotatable CCD. In addition, the values of the OLS regression model of ML are more robust than the ANOVA values generated by CCD rotatable of RSM. Nevertheless, it should be noted that the skewness (− 0.425) and kurtosis (2.722) obtained from the OLS Regression of ML are within acceptable boundaries. Although the linear regression model technique of ML is superior to CCD in modeling and optimizing biodiesel yield, both can model biodiesel production with high accuracy. According to the results, the Kurtosis score is 2.722, which is considered platykurtic and is favorable because it falls within the permissible range of − 3 to +3. Kurtosis can be defined as a measurement of the probability distribution or the width of a distribution's tails. If a normal distribution's kurtosis is 3, it is said to be mesokurtic. If it is more than 3, it is referred to as platykurtic, and if it is less   than 3, it is referred to as leptokurtic kurtosis. The skew is used to describe how much a distribution leans left or right. It is a distribution's third instant. The distribution is roughly symmetrical if the skewness is between − 0.5 and+0.5 [43]. A curve with a negative skew tends to the right. As can be seen, our skew value is − 0.425, which indicates that the distribution is symmetrical, and this is a desirable outcome. From Fig. 8, the main actual/input factors considered are Time and Irradiation power while the response/output is the yield. Each colour of a triangle represents an increase in its factor as it moves toward the tip. Sky-blue represents power, blue represents time, milky white represents catalyst, light-peach represents MeOH/oil ratio, brown represents average yield, and red represents optimum yield. However, the intercept of the four input factors and the yield gave the average value which is 89.81 wt% as corroborated in Table 9. And the red colour indicates the optimum value is 91.7 wt% biodiesel yield.

Optimization and validation of experimental data
The optimization for the process input variables was investigated by solving the regression equation. The optimum values predicted were a time of 3 min, irradiation power of 280 W, catalyst amount of 3 wt%, and MeOH/oil molar ratio of 6:1, with a predicted biodiesel yield of 91.60 wt% at the desirability of 1.0. The optimal condition of the input variables was used to carry out the transesterification of WMSO to validate the predicted model. The validation experiment at the optimal condition was conducted three times, and an average of 91.60 wt% biodiesel yield was obtained. The approximation value of biodiesel yield indicates that the model can accurately represent the biodiesel response.

Results of FT-IR analysis of biodiesel produced
A Fourier transform infrared spectrometer (Infrared spectrometer Varian 660 MidIR Dual MCT/DTGS Bundle with ATR) identifies the functional groups present in the synthesized biodiesel. Fig. 9 shows the FTIR spectrum and band characteristics of the methyl ester synthesized replotted with Origin Software. The bands at 3310, 2300, and 1550 cm − 1 are overtone, fermi resonance, and stretching vibrations of the ester functional group, respectively [6]. The CH 2 group's asymmetric/symmetric stretching is attributed to the band at 2948 cm − 1 . Between 1550 cm − 1 and 675.11 cm − 1 bands is the fingerprint region identifying biodiesel. The stretching, and bending vibration of C-O and CH 2 groups is assigned to 1289.59 cm − 1 . According to Ref. [44], the band at 1050.42 cm − 1 indicates the presence of oxygen, suggesting that the biodiesel produced would lead to complete combustion when used in a diesel engine. The bands present in the spectrum of the synthesized biodiesel are common to previously reported bands [6,45].

Characterization of biodiesel produced with GCMS
The methyl esters contained in the synthesized biodiesel from WMSO are displayed in Table 13. The saturated, unsaturated, and polyunsaturated fatty acid methyl acids are present in different proportions.

Physicochemical properties of biodiesel produced
The synthesized biodiesel's physicochemical properties were identified and evaluated against standards set by the European Union, Fig. 8. 3D effect of time and irradiation power of the biodiesel yield from ML the American Society for Testing and Materials, Acacia Farnesiana oil biodiesel (AFOB) and Albizzia julibrissin oil biodiesel (AJOB). The synthesized biodiesel's (WMSO biodiesel) in this study were determined in triplicate. It can be seen that the synthesized biodiesel met the specifications stipulated by the EU and ASTM standards. However, Table 14 shows that the biodiesel produced performs better than Acacia Farnesiana oil biodiesel and Albizzia julibrissin oil biodiesel in terms of flash point, cloud point, kinematics viscosity, cetane number, etc. As a result, the synthesized biodiesel could be used in the internal combustion engine without any modification.

Conclusion
A heterogeneous base catalyst of high catalytic efficiency was synthesized from the calcination of Kwale red anthill mud (KRAM). The synthesized catalyst in this study was predominantly Si and Al oxides or compounds, these are believed to be responsible for the catalytic capability of the solid basic catalyst. Additionally, high porosity and possession of a large surface area (42.16 m 2 /g) on the synthesized catalyst signal that reacting species in the transesterification could access its active site unhindered, thereby facilitating biodiesel production. The oil used for biodiesel production was extracted from dried watermelon seeds via the Soxhlet extraction principle using n-hexane as solvent. Two modeling tools; Central Composite Design (CCD) rotatable and (Machine learning) python coding was successfully used to model and optimize biodiesel production from WMSO and methanol in the presence of a synthesized catalyst. Python coding generated a model with better performance than CCD rotatable. The model R 2 obtained using CCD rotatable was 0.9827 while the R 2 of the linear regression model of ML was 1, and the 3D graph generated by ML was more visually appealing. ML outperforms CCD rotatable of RSM in forecasting and optimizing biodiesel output, and it also provides additional statistical factors that could help with decision-making. The highest yield in this study is at 4 wt% catalyst amount, 4 min duration, methanol/oil molar 8:1, and irradiation power of 350 W with the corresponding biodiesel yield of 93.41 wt%. The biodiesel produced in this study was within the allowable limit specified by the standards. The catalyst synthesized proved to be a good heterogeneous catalyst capable of expediting biodiesel production from watermelon seed oil.

Declarations research funding
This study has not received any financial assistance.

Author contribution statement
Sunday Chukwuka Iweka: Conceived and designed the experiments; Performed the experiments; Analyzed and interpreted the data; Contributed reagents, materials, analysis tools or data; Wrote the paper.

Data availability statement
Data will be made available on request.