UJET VOL. 2, NO. 2, DEC Page 71

Transcription

1 OPTIMIZATION OF BIODIESEL PRODUCTION FROM WASTE COOKING OIL VIA CENTRAL COMPOSITE DESIGN TOOLS * 1 Ejim, I. F, ** 2 Nwosu-Obieogu, K. and 2 Anike, E.N. 1 Chemical Engineering Department, Institute of Management and Technology, Enugu, Nigeria ABSTRACT 2 Chemical Engineering Department, Michael Okpara University of Agriculture, Umudike, Abia, Nigeria The environmental hazards of waste cooking oil (WCO) in our society has raised a lot of concern and spurred researchers to harness its reuse. This work statistically assessed the process conditions for biodiesel production from waste cooking oil (WCO) collected from restaurants in Enugu metropolis. The central composite design of Response Surface Methodology (RSM) was carried out and the results were analyzed using the MINITAB SOFTWARE A 5-level-2-factors experimental was designed which gave a 25 experimental runs with full randomization and zero centre points. Methanol-to-oil molar ratio of (1:3, 1:4, 1:6, and 1:8) and weight of catalyst ( g, 0.5g, 0.8g, and 1.0g) were used as independent variables (factors) and biodiesel yield as dependent variable (response) in transesterification reaction. Analysis of variance (ANOVA) was used to statistically analyze the experimental matrix. The variables were later fitted in a second order quadratic polynomial regression model showing the effect of catalyst and molar ratio on yield of biodiesel with the use of interaction, main effect, contour and surface plots. The experimental results suggested that a yield of 30.89% was achieved at methanol/oil molar ratio of 1:9 and catalyst weight of 0.2 while maintaining the model fit at which resulted in coefficient of determination (R 2 ) of 91.3%. In conclusion, biodiesel derived from waste cooking oil showed a closer properties relationship with EN14214 biodiesel standard. Keywords: waste cooking oil, biodiesel, transesterification, methanol, oil 1. INTRODUCTION There is serious demand for new energy resources, not only because of the growing industrialization and population, but also environmental concern caused by increasing emission of CO2 (Verhe and Stevens,2009). The greenhouse gas effects and the rather unstable political situation of the petroleum and natural gas producing regions in the world is equally contributing to major reason for searching for alternatives to petrodiesel (Chhetri et al., 2008). The global supply of oil and natural gas from the conventional sources is unlikely to meet the growth in energy demand over the next 25 years. In this perspective, considerable attention has been given towards the production of biodiesel as a diesel substitute (Lean, 2007). *Corresponding Author: Ejim F. I. **Corresponding kenenwosuobie@yahoo.com Biodiesel chemically known as monoalkyl esters of long chain fatty acids can be produced by the transesterification reaction, in which animal fat or vegetable oil is reacted with alcohol in the presence of a catalyst. Methanol is the most commonly used alcohol due to its low cost, and an alkali catalyst is more preferable than acid or enzyme catalysts due to its short reaction time (Verhe and Stevens,2009). It is non corrosive, low cost and gives high yield of biodiesel production (Tabatabaei et al., 2015). Biodiesel also has many other advantageous features when compared to petrodiesel. Those include reduction of most exhaust emissions, higher cetane number, biodegradability, lack of sulphur, inherent lubricity, positive energy balance, higher flash point, compatibility with the existing fuel distribution infrastructure, renewability, and domestic origin (Singh and Taggar, 2014). The last feature not only could potentially secure a continuous, consistent, and economic feedstock supply, but also could provide opportunities for indigenous development of especially rural and isolated regions. (Wang et al.,2006, Hajjari et al.,2014, Pinzi et al.,2014) Restaurants keeps emerging on a daily basis, also is the amount of waste cooking oil and frying oil required for disposal which poses a serious threat to its management. Waste cooking oil and fat create serious disposal problems due to its slow degradation. Frequently, these wastes are drained without any pretreatment and most times cause water pollution (Chhetri et al., 2008). The high viscosity and poor volatility are the major limitations of vegetable oils for their utilization as fuel in compression ignition engines. This UJET VOL. 2, NO. 2, DEC Page 71

2 is because high viscous vegetable oils causes evaporation and air-fuel mixture formation leading to improper combustion and higher smoke emission. Moreover this high viscosity generates operational problems like difficulty in engine starting, unreliable ignition and deterioration in thermal efficiency (Tabatabaei et al., 2015). Hence, converting it to biodiesel is one of the options to reduce the viscosity of vegetable oils. The use of these wastes as a reactant for biodiesel synthesis not only helps the disposal problem but also reduces production cost (Banerjee and Chakraborty, 2009, Paugazhabadivu and Jeyachandran, 2005). There are major barriers in the commercialization of biodiesel from lipids. First, the primary use of vegetable oils will remain as a food ingredient. Increase in world population and the growing standard of living in developing countries is leading to a higher consumption of vegetable oils. The conflict between the use of vegetable oil as human food/animal feed on one hand, and fuel use on the other hand, is already affecting the world vegetable oil market process (Hannon et al., 2010) The second obstacle for biodiesel commercialization is the high manufacturing costs which are mainly due to the higher costs of the raw materials, especially the virgin vegetable oils and to the cost of production. which are dependent on the location, plant size, the value of the by products (especially the glycerine production) and the environmental conditions. In most cases, in order to reach the standards for biodiesel, the crude oils-fats are been subjected to an intensive refining process which is an additional cost factor in the final biodiesel production. [Verhe and Stevens,2009, Wang et al.,2006) The most studied process to obtain biodiesel from feed stocks is transesterification of triglycerides with low molecular weight alcohols catalyzed by homogenous catalysts. These catalyst may cause problems due to the side saponification reaction which creates soap as a result of high content of fatty acids and water in the waste cooking oil, notwithstanding these drawbacks, transesterification process using base catalyst has some benefits like low production cost, faster reaction speed and mild reaction conditions. (Issariyakul et al., 2007, Aranda et al., 2007) Hence this research is aimed at statistically analyzing the process conditions of the biodiesel production from waste cooking oil in Enugu metropolis and studying the effect of catalyst and molar ratio on yield of biodiesel using MiniTAB 14.0 Software. 2. MATERIAL AND METHODS Materials used in this work includes: 900ml of waste cooking oil (WCO) collected from restaurants in Enugu metropolis, 2.5 litres each of methanol, KOH, HCl, NaSO3, H2SO4, acetic acid, distilled water, and phenolphthalein indicator (all Manufactured by BDH Chemicals Ltd Poole, England). Shimazu s Gas Chromatograph FID System, aluminum weighing dish, potassium iodine solution, starch indicators, thermometer, PET bottles, masking tape, separating funnel, electronic weighing machine, measuring beakers (2.5 liters), distillation flask, conical flask, stop watch, and a magnetic stirrer. 2.1 Design of Experiments A 2-factor-5-level experiment was conducted using a central composite design (CCD) to examine the effects of methanol/oil molar ratio and catalyst concentration (%) on yield of fatty acid methyl ester of waste cooking oil. Table 1: 5-level 2-factor Experimental Design Level Code Methanol/Oil Molar ratio X1 1:3 1:4 1:6 1:8 1:9 Catalyst (g) X The CCD consisted of 25 experimental runs and provided sufficient information to fit a full second-order polynomial model. The function was approximated by a second degree polynomial equation: n n 2 Y = o + i=1 i X i + i=1 ii X i + n i j 1 ij X i X j (1) Where Y is % methyl ester yield, Xi (methanol/oil molar ratio) and Xj (catalyst) are the independent study factors, and 0, i, ii, and ij are intercept, linear, quadratic, and interaction constant coefficients, respectively. An alpha ( ) level of was used to examine the statistical significance of the fitted polynomial models. MiniTAB 14.0 software was used to analyze the transesterification data for developing response equations, for analysis of variance (ANOVA), generate surface plots, main effects plots and interaction plots using its statistical toolbox. 2.2 Esterification Process Result from previous research by (Anya et al., 2012) was used for the esterification process in this experiment. 0.75% v/v of sulphuric acid volume-to-oil and 0.58:1 methanol-tooil ratio, was used in the esterification process. UJET VOL. 2, NO. 2, DEC Page 72

3 2.3 Transesterification Process Transesterification was carried out using the experimental setup in Fig ml of methanol was measured with measuring cylinder and poured in a conical flask, 0.4g of potassium hydroxide was also measured using analytical weighing balance and carefully added to the flask that contains the 12.5ml of methanol, the flask was covered with a bung and fitted on the cap and placed on an electric shaker for about 3 5minutes to enable the potassium hydroxide dissolve completely in the methanol. After mixing the catalyst and methanol, 100ml of esterified waste cooking oil was measured out with a measuring cylinder and poured in to a distillation flask and then pre-heated with a magnetic stirrer to about 60 0 C before pouring the prepared potassium methoxide solution from the conical flask. The mixture was allowed to stay 60 minutes at constant temperature of 60 0 C and 250rpm. After 60 minutes, the reaction mixture was poured into a separating funnel with a tap at the bottom and hung on a retort stand. This was allowed to stay overnight for complete separation. The tap at the bottom of the separating funnel was opened and the glycerol was decanted from separating funnel living only the biodiesel. The biodiesel produced was measured using measuring cylinder and stored in an air free tight bottle so as to avoid reacting with air or polluted air. The amount of biodiesel yield was 84ml and glycerol was 28.5ml. 2.4 Physicochemical properties The waste cooking oil and fatty acid methyl ester were analysed to determine the acid value, saponification value and iodine value at Institute of Management and Technology Enugu s physical chemistry laboratory before storage using an air free tight bottle so as to avoid reacting with air or polluted air. R denotes an observation with a large standardized residual The effect of methanol/oil molar ratio and catalyst weight on the oil yield (Y) is as shown on table 2 and this was subsequently used to fit the response equations for oil yield. Observations with large standardized residuals whose absolute values are greater than 2 or large leverage values are flagged. 3.0 RESULTS AND DISCUSSION The saponification value of waste cooking oil (WCO) was reported as (mg KOH/g). The acid value of waste cooking oil was found to be (mg KOH/g). The rest of the parameters can be found in table 5. It has been reported that transesterification would not occur if FFA content in the oil were above 3 wt%. (Canakci and VanGerpen, 2001). Runs Methanol /oil molar ratios Table 2: 2-factors-5-level Experimental Matrix Catalyst (g) Y2 (%) Fit Se fit Residue St 1 1: : : Statistical Analysis of Data and Response Equation for Transesterification The Analysis of Variance table, table 3 gives, for each term in the model, the degree of freedom, the sequential sums of squares (Seq SS), the adjusted (partial) sums of squares (Adj SS), the adjusted means squares (Adj MS), the F- statistic from the adjusted means squares, and its p-value. The sequential sums of squares are the added sums of squares given that prior terms are in the model. These values depend upon the model order. The adjusted sums of resid squares are the sums of squares given that all other terms are in the model. These values do not depend upon the model order. 4 1: UJET VOL. 5 2, 1:3 NO. 2, 0.2 DEC Page : :

4 From table 3, p-values were printed as for regression and linear models, and for squared model all meaning that they are less than This indicates significant evidence of effects since level of significance, alpha, is greater than The significant interaction effects has p- value greater than and this shows that the model is insignificant and implies that the coefficients of second order regression models of the effect of methanol/oil molar ratio and catalyst concentration does not depends on each other. The R 2 value shows that model explains 91.3% of the variance in the output, indicating that the model fits the data extremely well. Table 4, gives the estimated coefficients for factors, their interactions and quadratic coefficients, their standard errors, t-statistics, and p-values. This table was used to formulate equation (2) and it demonstrates clearly that every other term fits into the model except for the quadratic model of catalyst weight and interaction model of methanol/oil molar ratio and catalyst weight. Table 3: Analysis of Variance for Y2 (%) Term Coef SE Coef T P Constant METHANOL/OIL MOLAR RATIOS CATALYST (g) METHANOL/OIL MOLAR RATIOS 1* METHANOL/OIL MOLAR RATIOS 1 CATALYST (g)*catalyst (g) METHANOL/OIL MOLAR RATIOS 1* CATALYST (g) Source DF Seq SS Adj SS Adj MS F P Regression Linear Square Interaction Residual Error Total Figure 2: Surface plot of methanol/oil molar ratio and catalyst weight on Main Effect plot for Biodiesel from WCO S = R-Sq = 91.3% R-Sq(adj) = 88.9% Estimated Regression Coefficients Equation for Y2 (%) using data in coded units Y = X X X 1 2 (2) Interaction Plot (data means) for Y2 (%) MOLAR RATIOS Where X1 is methanol/oil molar ratio and X2 is catalyst concentration Methanol/Oil molar ratio (X1) had linear positive effect on yield while catalyst weight had negative linear effect on yield. The interaction of catalyst weight and methanol/oil molar ratio had no significant effect on the yield while methanol/oil molar ratio had significant high negative quadratic effects on yield. Table 4: Estimated Response Surface Regression Coefficients for Y2 (%) Mean CATALYST (g) Figure 3: Surface plot of methanol/oil molar ratio and catalyst weight on Interaction plot for Biodiesel from WCO Main Effect Plot for Biodiesel from Waste Cooking Oil METHANOL/OIL MOLAR RATIOS CATALYST (g) UJET VOL. 2, NO. 2, DEC Page 74 (%) 22.5

5 Figure 4: 3-D Surface plot of methanol/oil molar ratio and catalyst weight for Biodiesel from WCO Contour Plot of Y2 (%) vs CATALYST (g), METHANOL/OIL MOLAR RATIOS 1 CATALYST (g) METHANOL/OIL MOLAR RATIOS 1 Figure 5: Surface plot of methanol/oil molar ratio and catalyst C Unit Waste cooking oil Pretreated waste frying oil Y2 (%) < > 30 Biodiesel Biodiesel (EN14214) Min Max g/cm Figure 3 shows the interaction plot indicating the mean yield versus the catalyst for each of the five values of molar ratios. The legend shows which symbols and lines are assigned to the molar ratio. This plot shows apparent interaction between some lines that are not parallel for molar ratio of 1:9, 1:8 and 1:6 implying that the effect of catalyst upon yield depends upon the molar ratio in that range. While the parallel lines between values of 1:3 and 1:4 show the have no interaction with the yield. The response surface plots and the accompany contour plots shown in figure 4 and figure 5 for the chosen model equations shows the relationship between the independent and the dependent variables. From figure 4, the response surface indicates that the percentage yield of biodiesel increases as methanol/oil molar ratio increases while further increase led to decrease of percentage yield of oil. It is also clear that, the percentage yield of biodiesel decreases as weight of catalyst increases. In addition, there was mutual interaction between the methanol/oil molar ratio and catalyst weight. From figure 5, it can be seen that increase in methanol/oil molar ratio from 1:32 ended at 1:86 and showed a slight decrease in percentage yield as it advances. For Catalyst weight, weight of 0.2 to 0.55 shows an increase in catalyst with no significant increase in yield while weight of 0.55 to 1.0 shows a decrease in yield. C Acid value Saponification value mm 2 /s mg KOH/g mg KOH/g Flash point C weight on 2-D Contour plot for Biodiesel from WCO The main effects plot displays the response means for each factor level in sorted order. A horizontal line is drawn at the grand mean. The effects are the differences between the means and the reference line. In fig 2, there is a clear demonstration of increase in mean yield as molar ration increased from 1:3 to 1:6. We also noticed a slight increase in mean yield from 1:6 to 1:8 before dropping at 1:9. For the catalyst, the decrease in mean yield as the weight of catalyst increased was clearly demonstrated. Table 5: Physicochemical Properties of Waste Cooking Oil and Biodiesel from Waste Cooking Oil From table 5, it could be observed that the density of WCO reduced by 0.20 after pretreatment while that of biodiesel (0.89) was close to minimum EN14214 standard obtained in commercial diesel. However, the viscosity of biodiesel 3.10 is in agreement with the minimum EN14214 limit of biodiesel. This low viscosity of biodiesel will have lubricating effect in engines which will be an added advantage to the users, since it will reduce wear and tear in the engine. The acid value of WCO oil reduced drastically after pretreatment showing the effects of pre-treatment before UJET VOL. 2, NO. 2, DEC Page 75

6 transesterification. The low acid value of biodiesel conforms with the EN14214 standard of commercial biodiesel and shows the extent to which constituent glycerides can be decomposed by base-catalyzed reaction. The saponification value of WCO was mg KHO/g, this implies that the triglycerides of WCO have low molecular weight of fatty acids (saturated and unsaturated). The flash point of biodiesel 163 o C is close to the minimum EN14214 specification of biodiesel. Flash point helps to monitor the safe handling and storage of fuel. The higher the flash point the safer the fuel and vice versa. Therefore, it could be said that Biodiesel is safer to handle. 4.0 Conclusion It is a clear fact now that biodiesel has come to stay. It should also serve to reduce and maintain the price of petro-diesel fuel which strongly determines the economy of most developing countries like Nigeria. Waste cooking oil has proven to be an alternative and promising source for biodiesel production. The experimental results suggested that a yield of 30.89% was achieved at methanol/oil molar ratio of 1:9 and catalyst weight of 0.2 while maintaining the model fit at which resulted in R 2 of 91.3%. Hence, this research has clearly demonstrated how MiniTab 14.0 can be used to statistically analyze the process conditions of the biodiesel production from waste cooking oil collected from Enugu metropolis and studying the effect of catalyst and molar ratio on yield of biodiesel with the use of interaction, main effect, contour and surface plots. In conclusion, waste cooking oil should no longer end up in drainages as its conversion to biodiesel encourages environmental sustainability. REFERENCES Aranda D. A. G., Santos R. T. P., Tapanes N. C. O., Ramos A. L. D. and Artunes, O.C. (2007). Acidcatalyzed homogeneous esterification reaction for biodiesel production from palm fatty acids. Catalysis Letters, 122: Anya U., Nwobia N., and Ofoegbu, O. (2012). Optimized reduction of free fatty acid content on neem seed oil, for biodiesel production J. Basic. Appl. Chem., 2(4) Canakci, M. and Van Gerpen, J. (2001). Biodiesel production from oils and fats with high free fatty acids, American Society of Agricultural and Engineers, 44(6) Banerjee, A. and Chakraborty, R. (2009). Parametric sensitivity in transesterification of waste cooking oil for biodiesel production a review, Resour. Conserv. Recy. 53, Chhetri B., Chris Watts K. and Islam R. (2008). Waste Cooking Oil as an Alternate Feedstock for Biodiesel Production, Energies 1, 3-18: DOI: /en Hajjari, M., Ardjmand, M. and Tabatabaei, M. (2014). Experimental investigation of the effect of cerium oxide nanoparticles as a combustion-improving additive on biodiesel oxidative stability: mechanism. RSC Adv. 4(28), Hannon, M., Javier, G. and Mayfield, S.(2010). Biofuels from algae: challenges and potential, National Institute of Health., 1(5): Issariyakul, T., M. G. Kulkarne, L. C. Meher, A. K. Dalai and Bakhshi, N. N. (2007). Biodiesel production from mixtures of canola oil and used cooking oil. Chemical Engineering Journal (in press). Lean, G. (2007). Oil and gas may run short by The Independent, UK.. Paugazhabadivu, M. and Jeyachandran, K. (2005). Investigations on the performance and exhaust emissions of a diesel engine using preheated waste frying oil as fuel. Renewable Energy, 30: Pinzi, S., Leiva, D., López García, I., Redel Macías, M.D. and Dorado, M.P. (2014). Latest trends in feedstocks for biodiesel production. Biofuels Bioprod. Biorefin. 8(1), Singh, I. and Taggar, M.S. (2014). Recent trends in biodiesel production an overview. Int. J. Appl. Eng. Res., 9(10), Tabatabaei, M., Karimi, K., Sárvári Horváth, I. and Kumar, R. (2015) Recent trends in biodiesel production. Biofuel Research Journal, DOI: /BRJ UJET VOL. 2, NO. 2, DEC Page 76

7 Verhe, R. and Stevens, C.V. (2009). Production of biodiesel from waste lipids, in biofuels (eds W.Soetart and E.J. Vandamme), John Wiley & Sons, Ltd, Chichester, UK. Doi: / ch9. Wang Y., Ou S., Liu P., Xue F. and Tang S.((2006). Comparison of two different processes to synthesize biodiesel by waste cooking oil, J. Mol. Catal. A-Chem., UJET VOL. 2, NO. 2, DEC Page 77

8