In general, the reaction proceeds according to the following stoichiometric equation as follows (Fig. 8.2):

CH2-O-H

O

catalyst

O

CH2-O-C-R

Triglyceride (oil) methanol Glycerol mixture of methyl ester

Fig. 8.2 Stoichiometric equation of transesterification process for biodiesel production

Since the transesterification reaction is reversible, an increase in the amount of one of the reactants will result in higher FAME yield, and at least three molar equivalents of methanol are required for the complete conversion of the oil to its corresponding FAME.

Figure 8.3 presents the result for performance of different lipase in their free or immobilized form synthesizing biodiesel in batch process at a low temperature of 30°C. Among the lipases, Novozyme 435 (immobilized lipase from C. antarctica) displayed the highest methanol affinity because the lipase gave the highest FAME yield at 1:1-4:1 methanol to oil molar ratio. This is in agreement with the previous finding that Novozyme 435 showed high activity in methanolysis reaction com­pared to other alcoholysis reaction (Hernandez and Otero 2008). Throughout the methanol to oil molar ratio studied, Lipozyme TL IM (immobilized lipase T. lanu — ginosus) and free lipase C. antarctica achieved the second highest FAME yield, while the methanolysis reaction catalyzed by free lipase R. miehei gave the lowest FAME yield. Optimum FAME yield at 4:1 methanol to oil molar ratio was an advan­tage because high methanol to oil substrate molar ratio of more than 3:1 was required to shift irreversible transesterification for high conversion, to minimize the diffusion limitations, and to keep the glycerol formed in solution during the reaction (Noureddini et al. 2005; Hernandez and Otero 2008). The 4:1 methanol to oil molar ratio was not an inhibition factor to lipases; thus, the differences in FAME yield achieved between each lipases were due to lipase catalyzing specificity to the CPO transesterification. In addition, the -ert-butanol solvent was used to completely

dissolve methanol and CPO into the reaction mixture. This minimized the adverse effects of methanol and glycerol inhibitions to all types of lipases. Since lipase was not inhibited at 4:1 methanol to oil molar ratio, higher methanol to CPO can be used to encourage even greater FAME productivity.

Then, study of thermal stability of different lipase was carried out and presented in Table 8.4. Maximum FAME’s yield was attained at 30-40°C regardless of types of lipases used. The lipases were thermally deactivated at high temperature of 50-60°C. The immobilized lipases from T. lanuginosus, C. antarctica, and R. miehei were susceptible to thermal deactivation during the transesterification with alcohols. The catalyzing activity was the highest at the range of 25-35°C, and the lipase activity decreased at temperature above 40°C (Rodrigues et al. 2008). The results revealed that all immobilized lipases of T. lanuginosus, C. antarctica, and R. miehei had higher thermal resistant than the free lipases. The methanolysis cata­lyzed by Novozyme 435 (immobilized lipase C. antarctica) achieved high FAME yield (61.19-68.12%) under 30-60°C operating temperature, while for lipase CALB L (free lipase C. antarctica), a minimum yield of 33.22% was observed. The same trends were observed for lipase T. lanuginosus. FAME yield of 54.85-63.14% was observed within 30-60°C operating temperature for Lipozyme TL IM (immo­bilized lipase T. lanuginosus) and minimum yield of 35.38% for lipase TL 100L (free lipase T. lanuginosus). Therefore, the adverse effects due to thermal denatur — ation were less pronounced to the immobilized lipase, and the stability of immobi­lized lipase subjected to excessive heat treatment was largely preserved. The final mixture of biodiesel is usually composed of mono-alkyl esters, alcohol, and free fatty acids, and tri-, di-, and monoglycerides were then validated. Therefore, EN 14214 and ASTM D 6751 biodiesel standards are employed as references for the quality control of FAME produced from CPO transesterification.

CPO that was used as triglyceride feedstock and pure FAME that was sampled at 4 h were analyzed for chemical properties and physical performances (Table 8.5). Most of the physical and chemical properties and fatty acid compositions for both

CPO Reaction product3 at 4 h EN 14214 Description Value Description Value Limits Physical properties 1. Density 0.89 g/ml 1. Density 0.85 g/ml 0.86-0.90 g/ml 2. Kinematic 38.83 mm2/s 2. Kinematic viscosity at 40°C 7.93 mm2/s 3.5-5.0 mm2/s viscosity at 40°C Chemical properties 1. Iodine value 53.8 meq/kg 1. Iodine value 53.9 meq/kg 2. Acid value 6.4 mgKOH/g 2. Acid value 5.8 mgKOH/g 0.50 mgKOH/g (max) 3. Saponification value 191.7 3. Saponification value 184.7 — 4. Water content 0.035% v/v 4. Water content 0.050% v/v 0.050% v/v (max) 5. Free fatty acid 2.91% wt/wt oil 5. Free fatty acid 2.67% wt/wt oil — 6. Fatty acid compositions: 6. Fatty acid compositions: Myristic acid. 04:0 1.46% Myristic acid. C14:0 1.25% — Palmitic acid. 06:0 38.76% Palmitic acid. C16:0 40.92% Stearic acid. 08:0 4.00% Stearic acid. Cl8:0 6.03% Oleic acid. 08:1 43.80% Oleic acid. Cl8:1 42.88% Linoleic acid. 08:2 11.92% Linoleic acid. C18:2 8.92% 7. Acylglycerol compositions: 7. Acylglycerol compositions: Monoglyceride 0.38% wt/wt oil Monoglyceride 0.6% wt/wt oil 0.80% wt/wt oil Diglyceride 6.58% wt/wt oil Diglyceride 0.15%wt/wt oil 0.20% wt/wt oil Triglyceride 89.48% wt/wt oil Triglyceride 0% wt/wt oil 0.20% wt/wt oil (max) 8. Fatty acid methyl ester 0% wt/wt oil 1. Fatty acid methyl ester 2. Fatty acid methyl ester compositions: 96.51% wt/wt oil 96.5% wt/wt oil (min) Myristate ester. C14:0 1.19% — Palmitate ester. Cl6:0 41.2% Stearate ester. 08:0 4.43% Oleate ester. 08:1 43.3% Linoleate ester. 08:2 9.88% “Reaction product at 4 h was obtained after evaporating tof-butanol solvent at 85°C for 1 h and followed by separating glycerol from reaction mixture using centrifuge

CPO and FAME samples were at the same values except kinematic viscosity, water content, and acylglycerol compositions of tri-, di-, and monoglycerides. Kinematic viscosity for FAME sample was 80% less viscous than the CPO mainly due to the high presence of FAME yield, 96.15% in the mixture. The high viscosity observed in CPO was correlated to the content of unreacted triglycerides. Since kinematic viscosity for FAME sample was slightly higher than the EN standard, complete conversion in CPO transesterification was expected to further improve the kine­matic viscosity to fall within the limits. The water content in FAME sample that was mildly increased from 0.035 to 0.05% v/v may signify minimum level of esterifica­tion on FFA occurred in the system. The 0.05% v/v water content in FAME sample complied with EN standard. The quality control of water content was important since water could promote microbial growth, leads to tank corrosion, participates in the formation of emulsions, and causes hydrolysis or hydrolytic oxidation. Most of the acylglycerols in CPO that were converted to FAME with traces amount of di — and monoglycerides were detected in the FAME mixture, and these acylglycerol residues were within the EN quality control. Overall, the biodiesel specifications for FAME sample such as ester content, density, water content, mono-, di-, and triglyc­erides compositions with values of 96.15%, 850 kg/m3, 0.05% v/v, 0.65%, 0.15%, and 0%, respectively, were all well within the limits in standard EN. Since the hydrolysis of ester linkages was kept at the minimum level in the study and allowing the highest extent of transesterification process, FFA content in CPO was not con­verted to FAME and thus caused the acid value 5.8 mg KOH/g exceeding the EN limit of 0.5 mg KOH/g (Sim 2011).

Another type of lipid source that could be used is waste cooking palm oil (WCPO). Figure 8.4 shows the comparison of FAME’s yield between the waste cooking palm oil (WCPO) with the refine palm oil (RPO). The trend for WCPO and the RPO is almost the same. The reaction rate of WCPO was almost the same with

RCO at the beginning of reaction time up to 4 h. While at the final, the FAME yield of RCO was higher than that of WCPO. The highest FAME’s yield of 88 and 96% was achieved for WCPO and RPO, respectively.

Table 8.6 presented results obtained after conducting experimental in batch for production of biodiesel from WCPO. It showed that Novozyme 435 was used for the transesterification of waste cooking palm oil with methanol in batch system. Novozyme 435 remained active in tert-butanol as the reaction media (Halim et al. 2009). The optimum tert-butanol to oil volume ratio and methanol to oil molar ratio were achieved at 1:1 and 4:1, respectively. The optimum Novozyme quantity was 4% based on oil weight.

Lastly, Cerbera odollam oil was tested for production of biodiesel using enzy­matic reaction. The transesterification reaction was carried in batch, and optimiza­tion study was conducted by using statistical method using design of experiment (DOE) software. The optimum conditions found are 4.1 w/w% enzyme based on oil weight, 1:5.1 oil to methanol molar ratio, 0.51:1 solvent to oil volume ratio, tem­perature 40°C, and agitation speed 200 rpm. The value of FAME yield obtained from experiment was compared with the one predicted by DOE as shown in Table 8.7 . The predicted response value from DOE software is 95.09%, and the
optimized FAME yield from experimental study was 94.75%. An error of ±2.43% for FAME yield value under 95% confidence levels provides promising result of 94.75 ± 2.43% (Rahaman 2011).