Synthesis of Biodiesel via Interesteriﬁcation Reaction of Calophyllum inophyllum Seed Oil and Ethyl Acetate over Lipase Catalyst: Experimental and Surface Response Methodology Analysis

: Biodiesel is increasingly being considered as an alternative to the fossil fuel as it is renewable, nontoxic, biodegradable, and feasible for mass production. Biodiesel can be produced from various types of vegetable oils. Calophyllum inophyllum seed oil (CSO) is among the prospective nonedible vegetable oils considered as a raw material for biodiesel synthesis. The most common process of the biodiesel manufacturing is the transesteriﬁcation of vegetable oils which results in glycerol as a by-product. Thus, product puriﬁcation is necessary. In this work, an alternative route to biodiesel synthesis through interesteriﬁcation reaction of vegetable oil and ethyl acetate was conducted. By replacing alcohol with ethyl acetate, triacetin was produced as a side product rather than glycerol. Triacetin can be used as a fuel additive to increase the octane number of the fuel. Therefore, triacetin separation from biodiesel products is needless. The interesteriﬁcation reaction is catalyzed by an alkaline catalyst or by a lipase enzyme. In this study, biodiesel synthesis was carried out using a lipase enzyme since it is a green and sustainable catalyst. The interesteriﬁcation reaction of CSO with ethyl acetate in the presence of a lipase catalyst was conducted using the molar ratio of CSO and ethyl acetate of 1:3. The reaction time, lipase catalyst concentration, and reaction temperature were varied at 1, 2, 3, 4, 5 h, 10%,15%, 20%, and 30 ◦ C, 40 ◦ C, 50 ◦ C, 60 ◦ C, respectively. The experimental results were also analyzed using response surface methodology (RSM) with the Box–Behnken design (BBD) model on Design Expert software. Data processing using RSM revealed that the highest conversion within the studied parameter range was 41.46%, obtained at a temperature reaction of 44.43 ◦ C, a reaction time of 5 h, and a lipase catalyst concentration of 20%.


Introduction
Biodiesel is an alternative fuel for diesel engines which is synthesized from vegetable oils or animal fats. Biodiesel is one of the modern bioenergies which has several advantages. It has non-toxic properties, low emission rates and no sulfur gas, just to name a few. Thus, biodiesel is a prospective alternative as an environmentally friendly diesel fuel [1]. Biodiesel is commonly derived from vegetable oil which has renewable characteristics. The main components of vegetable oils and animal fats are triglycerides, which can be converted into mono alkyl esters of long chain fatty acid or termed as fatty acid alkyl ester (biodiesel). Various types of vegetable oils can be applied as prospective raw materials for biodiesel preparation [2]. The widely used raw material for biodiesel synthesis in Indonesia is crude (1) (2) To date, the development of green energy through environmentally friendly pro- (1) Energies 2022, 15 (1) (2) To date, the development of green energy through environmentally friendly processes has become a priority. One alternative towards a greener process is reducing the use of chemical catalysts and switching to enzymatic catalysts (biocatalysts). The enzyme-catalyzed biodiesel synthesis shows superiority compared to the chemically catalyzed one, mainly in terms of the lower energy requirement and natural conservation aspect. Lipase is the most common enzyme employed for biodiesel synthesis. There are two principal classes of lipase, free enzymes and immobilized enzymes. Free lipase enzymes are cheaper and simpler, thus it is feasible for large-scale applications [27]. Besides that, the use of lipase enzymes is very promising to overcome the disadvantages of alkaline catalysts related to the occurrence of the undesired saponification reaction when high FFA oil used as raw material. Lipase enzyme catalysts can be applied with high FFA feedstocks without any necessity of FFA removal as a pretreatment step. However, lipase biocatalysts are easily deactivated in an alcoholic environment since short-chain alcohols often cause irreversible loss of enzyme activity [28]. Therefore, the non-alcoholic interesterification route is preferable in order to maintain high biocatalyst activity and stability during the reaction. In the interesterification reaction, alcohol can be replaced by methyl acetate or ethyl acetate, which is harmless for the lipase catalyst [29].
In this work, biodiesel was synthesized through lipase-catalyzed interesterification of Calopyllum inophyllum seed oil (CSO) with ethyl acetate. The lipase enzyme used was the liquid-free lipase from Novozyme. The effects of the main parameters on the reaction conversion were studied experimentally. Analysis using the response surface methodology was also carried out to investigate the best reaction conversion obtained within the values range of the studied independent variables. Process optimization was also conducted using the response surface methodology, a collection of mathematical and statistical calculation techniques based on the compatibility of the empirical model with the experimental data obtained. Generally, RSM is useful for the modeling and analysis of problems that influences the response variables with the purpose of optimizing the response. RSM is also beneficial when determining the operating conditions to achieve the desired conversion [30]. There are several experimental design models used in RSM, such as the Box-Behnken Design (BBD), control composite design (CCD) and Doehlert Design. The BBD has been broadly applied for biodiesel production optimization since it is effective, requires the smallest number of trials than the other designs [31]. In this investigation, analysis using BBD in RSM was performed to determine the best reaction operation condition which resulted in the highest reaction conversion of the interesterification reaction of CSO in the presence of lipase enzyme catalyst within the ranges of the parameter values studied. (2) To date, the development of green energy through environmentally friendly processes has become a priority. One alternative towards a greener process is reducing the use of chemical catalysts and switching to enzymatic catalysts (biocatalysts). The enzymecatalyzed biodiesel synthesis shows superiority compared to the chemically catalyzed one, mainly in terms of the lower energy requirement and natural conservation aspect. Lipase is the most common enzyme employed for biodiesel synthesis. There are two principal classes of lipase, free enzymes and immobilized enzymes. Free lipase enzymes are cheaper and simpler, thus it is feasible for large-scale applications [27]. Besides that, the use of lipase enzymes is very promising to overcome the disadvantages of alkaline catalysts related to the occurrence of the undesired saponification reaction when high FFA oil used as raw material. Lipase enzyme catalysts can be applied with high FFA feedstocks without any necessity of FFA removal as a pretreatment step. However, lipase biocatalysts are easily deactivated in an alcoholic environment since short-chain alcohols often cause irreversible loss of enzyme activity [28]. Therefore, the non-alcoholic interesterification route is preferable in order to maintain high biocatalyst activity and stability during the reaction. In the interesterification reaction, alcohol can be replaced by methyl acetate or ethyl acetate, which is harmless for the lipase catalyst [29].
In this work, biodiesel was synthesized through lipase-catalyzed interesterification of Calopyllum inophyllum seed oil (CSO) with ethyl acetate. The lipase enzyme used was the liquid-free lipase from Novozyme. The effects of the main parameters on the reaction conversion were studied experimentally. Analysis using the response surface methodology was also carried out to investigate the best reaction conversion obtained within the values range of the studied independent variables. Process optimization was also conducted using the response surface methodology, a collection of mathematical and statistical calculation techniques based on the compatibility of the empirical model with the experimental data obtained. Generally, RSM is useful for the modeling and analysis of problems that influences the response variables with the purpose of optimizing the response. RSM is also beneficial when determining the operating conditions to achieve the desired conversion [30]. There are several experimental design models used in RSM, such as the Box-Behnken Design (BBD), control composite design (CCD) and Doehlert Design. The BBD has been broadly applied for biodiesel production optimization since it is effective, requires the smallest number of trials than the other designs [31]. In this investigation, analysis using BBD in RSM was performed to determine the best reaction operation condition which resulted in the highest reaction conversion of the interesterification reaction of CSO in the presence of lipase enzyme catalyst within the ranges of the parameter values studied.

Materials
The materials used were: Calophyllum inophyllum seed oil (CSO) from UMKM Samtamanu Cilacap, Indonesia, ethyl acetate p.a (Merck), liquid lipase catalyst (Eversa Transform 2.0 from Novozyme), phosphoric acid p.a (Merck), KOH p.a (Merck), oxalic acid p.a (Merck), ethanol p.a (Merck), and aquadest from Indrasari chemicals store, Semarang, Indonesia. The catalyst used was liquid (free form) lipase Eversa Transform 2.0 (ET 2.0) from Novozyme, which is made from genetically modifed Aspergillus oryzae microorganism [32]. Commonly, lipase are active in broad range of temperature from 20-60 • C [33,34] and the ET 2.0 was reported to work satisfactorily up to the reaction temperature of 60°C [35]. Crude CSO as much as 100 mL was put into a 250 mL flat bottom flask. The oil was then heated using hot plate equipped with a magnetic stirrer at a speed of 600 rpm until the temperature reached 70 • C. Phosphoric acid was subsequently added to the hot oil with the amount of 0.3% (w/w) of CSO and heated for 25 min. The oil was cooled at room temperature and put into a 500 mL separatory funnel. Distilled water at a temperature of 40 • C was thereafter added with a volume of 5% of the CSO volume. Subsequent to the degumming process, the mixture was settled in the separatory funnel for 24 h to attain the completion of the separation between the CSO and the residue. After the 24 h decantation process, the two layers were appeared. The top layer of brownish yellow color was the refined CSO, while the dark brown bottom layer was gum, impurities, and water which should be separated. Let the mixture in the separatory funnel for to attain the completion of the separation between the oil and the residue. The degummed CSO was then separated from the residue and was heated in the oven with a temperature of 105 • C to reduce the water content [30]. CSO was then analyzed to determine its physical properties (density, viscosity, and acid number).

Interesterification Reaction
After the degumming process, CSO was used as the raw material for biodiesel synthesis via interesterification reaction. CSO as much as 100 mL was introduced into a 250 mL flat bottom three neck flask and added with 10% (w/w) liquid lipase. The mixture was heated using a hot plate equipped with a magnetic stirrer with the stirring speed of 600 rpm until it reached the desired temperature. On the other hand, ethyl acetate was warmed up separately up to the similar temperature. When the reactants attained the specified temperature, ethyl acetate was subsequently poured into the reactor and mixed with the CSO and lipase. This incident was recorded as the reaction time of zero (t = 0 h). The ratio of CSO to ethyl acetate was fixed at 1:3. The stoichiometric ratio of the reactants was employed to investigate the optimum reaction conversion that can be obtained without any excess reactant (ethyl acetate) as it was conducted by Manurung et al. [36]. The reaction temperature was varied at 30 • C, 40 • C, 50 • C and 60 • C.The reaction temperature was varied at 30 • C, 40 • C, 50 • C and 60 • C. After the specified temperature was reached, the ethyl acetate with the similar temperature was added with the ratio of CSO to ethyl acetate of 1:3 (stoichiometric ratio). The reaction time was 1-5 h. Sample was taken periodically every 1 h of the reaction time. Prior to the analysis, sample was centrifuged with the centrifugation speed of 40,000 rpm for 15 min to separate between biodiesel product and the catalyst residue. The sample analysis was carried out using Gas Chromatography-Mass Spectroscopy (GC-MS). The independent variables for the experimental work is shown in Table 1.

Gas Chromatography-Mass Spectroscopy (GC-MS)
Analysis of the fatty acid composition of the biodiesel was determined using a Gas Chromatography-Mass Spectrometer (GC-MS QP2010 SE) with a column flow rate of 1.20 mL/min, an oven column temperature of 65 • C, a pressure of 74.5 kPa, and an injecttion temperature of 250 • C with split injection mode. The internal standard used was heptadecanoic acid, methyl ester.

Response Surface Methodology Analysis
The GC-MS data was utilized to determine the reaction conversion. The results were subsequently analyzed using response surface methodology (RSM) with the Box-Behnken Design (BBD) model on Design Expert software to determine the optimum operation condition. The BBD is a vigorous and extensively applied model for biodiesel synthesis optimization and parameter analysis [31]. The levels of the tested parameters for the BBD experiment are presented in Table 2.

Effects of the Degumming Process on the CSO Properties
Naturally, crude CSO contains phospholipids, sterols, free fatty acids, waxes, oilsoluble pigments and hydrocarbons. Degumming is the initial and essential refining process of crude vegetable oil to remove phospholipids and gums [37]. To date, there are various methods that can be employed to degum vegetable oil, explicitly water, acid, enzymatic, and membrane degumming techniques. Water degumming and acid degumming are the most used methods for vegetable degumming at an industrial scale [38]. Those methods are simple, easy, and low cost. However, water degumming is less effective compared to the acid method. The gum is comprised of hydratable (HP) and non-hydratable phosphatides (NHP). HP can be eliminated using water. In contrast, NHP cannot be eradicated via water degumming [15,16]. Thus, the acid degumming method is more advantageous in that it can remove both the HP and NHP. However, water degumming is less effective compared to acid degumming. Thus, the acid degumming method is more advantageous. The most used acids for the degumming process are phosphoric or citric acid, with a suggested concentration between 0.05-2% w/w oil [39]. In this work, 0.3% phosphoric acid was applied [23,40]. After the degumming process, the black color of the CSO turns reddishyellow due to the loss of the phospholipids compound in crude CSO. Phospholipids can form dark colors in vegetable oil as a result of the autoxidation process of these compounds during storage. The condensation reaction between the amino groups in phospholipids with aldehydes will yield melanophosphatides compounds which give a dark color to the oil [41]. Thus, removal of phospholipids in CSO will reduce the occurrence of the abovementioned reaction and diminish the color intensity.
The degumming process can improve the properties of CSO. In this work, the density, viscosity, and acid number of crude and degummed CSO were tested experimentally. The effects of the degumming process on the physicochemical properties of CSO is indicated in Table 3. It was obvious that the density and viscosity of the CSO were slightly decreasing after the degumming process due to the removal of gum and other impurities. The effect of the degumming process on the CSO properties is indicated in Table 2. It was revealed that the density and viscosity of the CSO slightly decreased after the degumming process. It occurred with the removal of gum and other impurities. Besides, degumming also leads to a lessening of the acid number of the oil which is attributable to the decrease of the acid number and the existence of free fatty acids (FFA) in the oil during the degumming process. This phenomenon is in line with the data reported by Adekunle et al. [42]. The fatty acid composition of the CSO after undergoing the degumming process is presented in the Table 4, based on the interpretation of the chromatogram shown in Figure 1. sides, degumming also leads to a lessening of the acid number of the oil which is attributable to the decrease of the acid number and the existence of free fatty acids (FFA) in the oil during the degumming process. This phenomenon is in line with the data reported by Adekunle et al. [42]. The fatty acid composition of the CSO after undergoing the degumming process is presented in the Table 4, based on the interpretation of the chromatogram shown in Figure 1.

Effects of the Temperature on the Interesterification Reaction Conversion
After the degumming process, CSO directly underwent the interesterification process with ethyl acetate in the presence of the lipase catalyst. As depicted in Table 2, the crude and degummed CSO contained high FFA of 32.47% and 24.25%, respectively, which were over 2 mgKOH/g. However, the enzymatic reaction is generally insensitive to the existence of the FFA impurities in the oil feedstock. The possibility of the undesired saponification reaction can be neglected. Therefore, a specific pre-treatment for reducing the FFA content of CSO was considered unnecessary since the FFA presentation in the oil does not affect the enzymatic catalyzed reaction [22,43].
The influences of the reaction time and temperature on the reaction conversion were observed at a fixed lipase catalyst concentration of 10% (w/w). The result is exhibited in Figure 2. Basically, lipases are active in a temperature range of 20-60 • C [33,34] and the ET 2.0 lipase can work up to 60°C [35]. Moreover, the reaction time is often studied for up to 12 h, but the significant reaction rate is in the range of 0-5 h [44]. In this work, it was disclosed that the highest reaction conversion was 54.99%, obtained at a reaction temperature of 40 • C and a reaction time of 3 h. Commonly, the reaction conversion is enhanced with the increasing of the reaction temperature and reaction time, since at the higher temperature the viscosity of the liquid decreases, causing the higher solubility of reactants, mass diffusion, rate of reaction, and conversion [45]. However, in this work, the reaction conversion decreased when the reaction was performed at 50 • C for 5 h. This trend was in accordance with the results recorded by Gusniah et al. [46], where it was reported that the lipase enzyme catalyst activity in the transesterification reaction of waste cooking oil reached an optimum performance at 40°C and it declined to some extent when the temperature rose to 50 • C, which happened when the lipase catalyst began to denature, causing damage of the active part of the enzyme at the higher temperature. Beyond the optimum temperature, the lipase catalyst becomes unstable and misplaces its tertiary structure, which causes the shortfall in its activity [45]. It thus leads to the decline of the reaction rate and conversion. Generally, the lipase enzyme reached the highest activity at the temperature of 40-45 • C as stated by Murtius et al. [47], Ayinla et al. [48], and Yazid [49].
to 12 h, but the significant reaction rate is in the range of 0-5 h [44]. In this work, it was disclosed that the highest reaction conversion was 54.99%, obtained at a reaction temperature of 40 °C and a reaction time of 3 h. Commonly, the reaction conversion is enhanced with the increasing of the reaction temperature and reaction time, since at the higher temperature the viscosity of the liquid decreases, causing the higher solubility of reactants, mass diffusion, rate of reaction, and conversion [45]. However, in this work, the reaction conversion decreased when the reaction was performed at 50 °C for 5 h. This trend was in accordance with the results recorded by Gusniah et al. [46], where it was reported that the lipase enzyme catalyst activity in the transesterification reaction of waste cooking oil reached an optimum performance at 40 ℃ and it declined to some extent when the temperature rose to 50 °C , which happened when the lipase catalyst began to denature, causing damage of the active part of the enzyme at the higher temperature. Beyond the optimum temperature, the lipase catalyst becomes unstable and misplaces its tertiary structure, which causes the shortfall in its activity [45]. It thus leads to the decline of the reaction rate and conversion. Generally, the lipase enzyme reached the highest activity at the temperature of 40-45 °C as stated by Murtius et al. [47], Ayinla et al. [48], and Yazid [49].

Effects of the Catalyst Concentration E on the Interesterification Reaction Conversion
Investigation on the influence of the lipase catalyst concentration was carried out at the molar ratio of oil to ethyl acetate of 1:3, a reaction temperature of 40 • C and a reaction time of 5 h. The lipase concentrations studied were 10%, 15%, and 20%. Figure 3 has shown that the optimum catalyst concentration was 15%, resulting in the reaction conversion of 28.445%. This result was in agreement with the results obtained by Gusniah et al. [46] which indicated that the highest lipase catalyst loading was 15%. However, the reaction conversion in this work was still low compared to those obtained in other research as the reaction was conducted with the low molar ratio of oil and ethyl acetate (1:3) and a short reaction time (5 h). By comparison, Sun et al. [44] achieved a 94.2% yield for the transesterification of semen Abutili seed oil in the presence of a similar lipase catalyst but at an excess molar ratio of oil and alcohol (1:7) and a longer reaction time (11 h). lized lipase, in which the lipase was attached on the carrier substrate, which led to higher product purity. Thus, it can be deduced that the lipase-catalyzed reaction in this work resulted in a lower conversion due to a shorter reaction time and low molar ratio of the reactants, which did not exceed the stoichiometric ratio. Besides that, utilization of the liquid lipase can reduce the apparent concentration of biodiesel yield since it can act as an impurity in the final biodiesel product.

Response Surface Methodology (RSM) Analysis
Response surface methodology (RSM) is the broadest employed method of experimental design for optimization. This technique applies mathematical and statistical methods for modeling and analyzing a process which involves numerous parameters. The purpose of the RSM is to optimize the response variables [29]. The factors that influence the process are entitled dependent variables. Meanwhile, the responses are Subhedar and Gogate [29] carried out the interesterification reaction of waste cooking oil and methyl acetate using immobilized lipase obtained from Thermomyces lanuginosus (Lipozyme TLIM) as a catalyst. The optimum reaction was demonstrated at the oil: methyl acetate molar ratio of 1:12, enzyme concentration of 6% (w/v), reaction temperature of 40 • C and reaction time of 24 h. Biodiesel yield was 90.1% under these operation conditions, which was higher than the results obtained in this work. A higher reaction conversion was obtained as the reaction was conducted on an excessive molar ratio and an extensively longer reaction time (24 h). Besides, this reaction employed an immobilized lipase, in which the lipase was attached on the carrier substrate, which led to higher product purity. Thus, it can be deduced that the lipase-catalyzed reaction in this work resulted in a lower conversion due to a shorter reaction time and low molar ratio of the reactants, which did not exceed the stoichiometric ratio. Besides that, utilization of the liquid lipase can reduce the apparent concentration of biodiesel yield since it can act as an impurity in the final biodiesel product.

Response Surface Methodology (RSM) Analysis
Response surface methodology (RSM) is the broadest employed method of experimental design for optimization. This technique applies mathematical and statistical methods for modeling and analyzing a process which involves numerous parameters. The purpose of the RSM is to optimize the response variables [29]. The factors that influence the process are entitled dependent variables. Meanwhile, the responses are named as dependent variables. Therefore, analysis using RSM is useful in designing operating conditions to achieve the targeted conversion. Many studies have shown that the most applicable tools in RSM for optimization of biodiesel production process are the Box-Behnken Design (BBD) and Central Composite Design (CCD) tools [50]. BBD is favored over CCD since it provides higher efficiency but more economical [51,52]. To evade the failure and extreme reaction variables value, BBD is limited to three levels. Hence, BBD typically has no factorial points. The BBD tool was also utilized by Sharma et al. [53] and Rokni et al. [54].
In this study, BBD was applied to obtain the best operating conditions and highest conversion for biodiesel synthesis via interesterification reaction within the ranges of the parameter values studied. The Box-Behnken Design (BBD) tool in RSM was used to evaluate the effect of the independent variables (reaction time, reaction temperature, and catalyst concentration). The experimental data of the reaction conversion compared with the prediction using BBD is exhibited in Table 5. The parameters used to obtain the appropriate model to predict the response results must match the observed model to the experimental data. Analysis of variance (ANOVA) is accomplished using the most common types of polynomial models, namely linear, interactive (2FI), quadratic and cubic can be applied to predict the response variables of experimental data. Several parameters such as sequential p-value, lack of fit p-value, adjusted R¬2, predicted R2 and Adiq precision were used to conclude the most suitable type of model for the optimization of the conversion of biodiesel produced from CSO. In this study, optimization using RSM was initially performed using the quadratic model. The ANOVA quadratic model is presented in Table 6. Based on the quadratic model, it was revealed that the p-value was 0.2098, which did not meet the requirement of the p-value < 0.05 as demonstrated in Table 6. Hence, the quadratic model was not significant for this case. To overcome this obstacle, a modified model was developed to obtain the significant model [55]. Modification of the model showed that the reduced cubic model was a significant model (Table 7). It was found that the p-value of the modified cubic model was 0.0428, which met the requirement of p-value > 0.05. Furthermore, the value of lack-of-fit was examined. Lack-offit is a test that analyzes how satisfactory the full models suit with the data. Models with a significant lack-of-fit should not be applied for predictions. In this work, the lack-of-fit was not significant, which meant that the model could be employed for the prediction. The value of adeq precision is the measure of the ratio of the signal to the disturbance. The expected value of the ratio is > 4 [50]. This model provided the value of the Adeq precision of 9.6401, indicating that the model is proper. Therefore, the reduced cubic model can be used to describe the response on the reaction conversion. The empirical correlation of the reduced cubic model in the form of polynomial order for the reaction conversion in the CSO esterification is displayed in Equation (3) where, A is the reaction temperature ( • C), B is the reaction time (min) and C is the Catalyst Concentration (%). Based on the reduced cubic model, analysis on the effects and the process variables interaction on the response variable were carried out using 3D RSM graphs. The results are presented in Figure 4.   Figure 4a shows the interaction between temperature and reaction time which gives a significant change to the increase in conversion. It was found that there was an increase in conversion at the temperature of 40 °C and reaction time of 5 h. This phenomenon is in line with the investigation of Handayani et al. [55], which reported that the optimum  Figure 4a shows the interaction between temperature and reaction time which gives a significant change to the increase in conversion. It was found that there was an increase in conversion at the temperature of 40 • C and reaction time of 5 h. This phenomenon is in line with the investigation of Handayani et al. [55], which reported that the optimum conversion was obtained at the temperature of 45 • C and reaction time of 6 h. Figure 4b discloses that the longer the reaction time employed, the higher the conversion achieved, as the reaction time is directly proportional to the amount of product yielded. Li et al. [56] observed that the reaction conversion of Pseudomonas cepacia lipase was low at the reaction time of 3 h. (44%). On the contrary, it rose to 73% when the reaction time was extended to 12 h. Figure 4c bares that the lower the reaction temperature applied, the higher the conversion. This occurred because the lipase enzyme activity worked at the temperature of 30-45 • C. In the enzymatic reaction, the increase of the temperature will promote the reaction rate and accordingly the reaction conversion due to the decrease of the viscosity which enhance the solubility of oil and methanol. However, at a certain temperature, the reaction conversion declined, invoking the enzyme deactivation [57]. Meanwhile, the addition of the catalyst concentration increased the reaction conversion. This result was in line with the description of Gusniah et al. [46]. This indicated that increasing the amount of the enzyme employed will provide a higher amount of the available active sites for the reaction, leading to a higher reaction conversion.
The Derringer method is utilized to optimize the response which is characterized by the presence of a desirability function. Desirability shows how close the optimization results are to the optimum point. The desirability function (DF) is constructed on the transformation of all the acquired responses from different scales into a scale-free value. The values of DF range from 0 to 1. The value 0 means that the factors provide an undesirable response. On the other hand, the value 1 denotes the optimal performance of the evaluated factor [58]. Hence, the desirability value which is close to 1 is the expected value. However, the RSM analysis is not only to find the desirability value, which is equal to 1, but to determine the conditions that match expectations. The multiple correlation coefficient (R 2 ) was closed to 1. It denoted that the models have good predictive ability. Based on the RSM analysis results, the highest reaction conversion was 41.46%, attained at a temperature of 44.43 • C, a reaction time of 5 h, and a catalyst concentration of 20%, with a desirability value of 0.733 as shown in Figure 5. This finding can be stated as the local optimum within the certain parameter range studied, which is worthwhile as a basis for developing process intensification in order to obtain a higher conversion.

Conclusions
Based on the RSM analysis, it was revealed that the reaction time and catalyst concentration had significant effects on the conversion of the interesterification reaction of CSO with ethyl acetate in the presence of lipase catalysts for biodiesel synthesis. The re-

Conclusions
Based on the RSM analysis, it was revealed that the reaction time and catalyst concentration had significant effects on the conversion of the interesterification reaction of CSO with ethyl acetate in the presence of lipase catalysts for biodiesel synthesis. The results of the analysis showed that the highest reaction conversion was 41.46% at the reaction temperature, reaction time, and catalyst concentration of 44.40 • C, 5 h, and 20%, respectively, which can be stated as the local optimum. While the results are good, it is not economically feasible for industrial-scale biodiesel production and purposes in its current condition. Therefore, further process intensification is necessary to enhance the reaction conversion and biodiesel yield.