1. INTRODUCTION
⌅For the past few years, advance research and development on the production of biofuels from renewable sources has been vehemently growing due to the excessive burning of fossil fuels which cause various environmental issues (Bankovicllic et al., 2012Bankovic-Llic O, Stamenkovic OS, Velikovic VB. 2012. Biodiesel production from non-edible plant oils. Renew. Sust. Ener. Ver. 16, 3621-3647. http://dx.doi.org/10.1016/j.rser.2012.03.002 ). The use of biodiesel has been widely recognized due to its significant contribution to the reduction of greenhouse gas emission, specifically in the transportation sector (Lam et al., 2019Lam MK, Jamalluddin NA, Lee KT. 2019. Production of Biodiesel Using Palm Oil. Biofuels: Alternative Feedstocks and Conversion Processes for the Production of Liquid and Gaseous Biofuels. Biomass Biofuels Biochem. 2, 539-574. https://doi.org/10.1016/B978-0-12-816856-1.00023-3 ).
According to the Brazilian National Petroleum Agency (ANP, 2015ANP, Agência Nacional de Petróleo, Gás Natural e Biocombustíveis. 2015. Boletim mensal de biodiesel.), most of the biodiesel produced in Brazil is obtained from soybeans and since this is a crop mainly grown for human consumption, research on the exploitation of other oilseed crops with the potential to produce biodiesel has been reported. The oil from the Macauba fruit (Acrocomia aculeata) stands out in this sense, due to the great potential for production, which can be from 1500 to 5000 Kg of oil per hectare (Manfio et al., 2011Manfio CE, Motoike SY, Pimentel LD, de Queiroz V, Sato AY. 2011. Repetibilidade em características biométricas do fruto de macaúba. Cienc. Rural 41, 70-76. https://doi.org/10.1590/S0103-84782011000100012 ), higher than the productivity displayed by soybeans, about ~ 576 kg of oil per hectare (Tamagno et al., 2020Tamagno S, Aznar-Moreno JA, Durrett TP, Prasad PVV, Rotundo JL, Ciampitti IA. 2020. Dynamics of oil and fatty acid accumulation during seed development in historical soybean varieties. Field Crops Res. 248, 1-10, 2020. https://doi.org/10.1016/j.fcr.2020.107719 ). The extraction of macauba oil can come from the kernel and the pulp, and the fatty acid composition of the oil extracted from the pulp consists of unsaturated (oleic and linoleic) and saturated (palmitic and stearic) fatty acids (Rosa et al., 2020Rosa ACS, Stevanato N, Garcia VAS, Silva C. 2020. Simultaneous extraction of the oil from the kernel and pulp of macauba fruit using a green solvent. J. Food Process Pres. 44, 1-14. https://doi.org/10.1111/jfpp.14855 ). The composition of kernel oil is mainly composed of saturated fatty acids (lauric, myristic and palmitic) (Trentini et al., 2018Trentini CP, Cuco RP, Cardozo-Filho L, Silva C. 2018. Extraction of macauba kernel oil using supercritical carbon dioxide and compressed propane. Can. J. Chem. Eng. 97, 785-792. https://doi.org/10.1002/cjce.23236 ; Rosa et al., 2020Rosa ACS, Stevanato N, Garcia VAS, Silva C. 2020. Simultaneous extraction of the oil from the kernel and pulp of macauba fruit using a green solvent. J. Food Process Pres. 44, 1-14. https://doi.org/10.1111/jfpp.14855 ) and therefore, because it contains a greater amount of these fatty acids, the use of macauba kernel oil confers the production of biofuels with greater oxidative stability (Trentini et al., 2018Trentini CP, Cuco RP, Cardozo-Filho L, Silva C. 2018. Extraction of macauba kernel oil using supercritical carbon dioxide and compressed propane. Can. J. Chem. Eng. 97, 785-792. https://doi.org/10.1002/cjce.23236 ). In addition, problems related to the supercooling of biodiesel from this oil were not observed (Menezes et al., 2021Menezes FAF, Rangel AB, Cordeiro TC, Vargas H, Silva EC. 2021. Investigation of phase transitions in vegetable oils through temperaturedependent optical measurements: supercooling efect. J. Therm. Anal. Calorim. 143, 27-33. https://doi.org/10.1007/s10973-019-09168-7 ).
The application of macauba oil in the synthesis of biodiesel requires its use in crude form (without refining process) in order to reduce raw material costs, which contribute with a high share in production costs. However, crude macauba oil has high acidity, with reports of 70.26% (Silva et al., 2021Silva C, Colonelli TAS, Postaue N, Zempulski D, Trentini CP, Silva EA, Cardozo Filho L. 2021. Catalyst-free production of fatty acid ethyl esters (FAEE) from macauba pulp oil. Grasas Aceites 72 (1), e398. https://doi.org/10.3989/gya.0103201 ) and 23% (Raspe et al., 2013Raspe DT, Silva C, Cardozo-Filho L. 2013. Effect of additives and process variables on enzymatic hydrolysis of macauba kernel oil (Acrocomia aculeata). Chem. Eng. J. 2013, 1-8. https://doi.org/10.1155/2013/438270 ) in pulp and kernel oil, respectively. The hydrolysis of triglycerides followed by the esterification of the obtained fatty acids has stood out in obtaining fatty acid esters for substrates with high acidity (Vescovi et al., 2016Vescovi V, Rojas MJ, Baraldo A, Botta DC, Santana FAM, Costa JP, Machado MS, Honda VK, Giordano RLC, Tardioli PW. 2016. LipaseCatalyzed Production of Biodiesel by Hydrolysis of Waste Cooking Oil Followed by Esterification of Free Fatty Acids. J. Am. Oil Chem. Soc. 93, 1615-1624. https://doi.org/10.1007/s11746-016-2901-y ), mainly in the conventional transesterification process with alkaline catalyst, which inevitably generates soap in the presence of these substrates, thus inactivating the catalyst, making separating biodiesel and glycerol expensive and affecting process productivity (Sousa et al., 2010Sousa JS, Cavalcanti-Oliveira EA, Arandab DAG, Freire DMG. 2010. Application of lipase from the physic nut (Jatropha curcas L.) to a new hybrid (enzyme/chemical) hydroesterification process for biodiesel production. J. Mol. Catal. B. Enzym. 65, 133-137. https://doi.org/10.1016/j.molcatb.2010.01.003 ). However, to make this route industrially viable, in addition to the raw material, operational parameters related to the production costs of this biofuel, such as reaction time, energy demand and catalyst performance must be considered (Wancura et al., 2021Wancura JHC, Fantinel AL, Ugalde GA, Donato FF, Oliveira JV, Tres MV, Jahn SL. (2021). Semi-continuous production of biodiesel on pilot scale via enzymatic hydroesterification of waste material: Process and economics considerations. J. Clean Prod. 285, 124838. https://doi.org/10.1016/j.jclepro.2020.124838 ).
Hydrolysis and esterification reactions have been reported using enzymatic catalysts (Santos et al., 2015Santos LDF, Coutinho JAP, Ventura SPM. 2015. From Water-in-Oil to Oil-in-Water Emulsions to Optimize the Production of Fatty Acids Using Ionic Liquids in Micellar Systems. Biotechnol. Prog. 31, 1473-1480. https://doi.org/10.1002/btpr.2156 ; Zhou et al., 2015Zhou GX, Chen GV, Yan BB. 2015. Two-step biocatalytic process using lipase and whole cell catalysts for biodiesel production from unrefined jatropha oil. Biotechnol. Lett. 37, 1959-1963. https://doi.org/10.1007/s10529-015-1883-4 ; Barbosa et al., 2019Barbosa MS, Freire CCC, Almeida LC, Freitas LS, Souza RL, Pereira EB, Mendes AA, Pereira MM, Lima AS, Soares CMF. 2019. Optimization of the enzymatic hydrolysis of Moringa oleifera Lam oil using molecular docking analysis for fatty acid specificity. Biotechnol. Appl. Biochem. 66, 823-832. https://doi.org/10.1002/bab.1793 ), mainly due to heterogeneity, the employment of soft conditions (Kabbashi et al., 2015Kabbashi NA, Mohammed NI, Alam MZ, Mirghani MES. 2015. Hydrolysis of Jatropha curcas oil for biodiesel synthesis using immobilized Candida cylindracea lipase. J. Mol. Catal. B. Enzym. 116, 95-100. https://doi.org/10.1016/j.molcatb.2015.03.009 ; Nguyen et al., 2017Nguyen VTA, Le TD, Phan HN, Tran LB. 2017. Antibacterial Activity of Free Fatty Acids from Hydrolyzed Virgin Coconut Oil Using Lipase from Candida rugosa. J. Lipids 2017, 1-7. https://doi.org/10.1155/2017/7170162 ), and the high degree of specificity of the desired substrates, which promotes reaction acceleration and biodegradability, making them less polluting compared to other catalysts and facilitating their reuse (Rodrigues and Ayub, 2011Rodrigues RC, Ayub MAZ. 2011. Effects of the combined use of Thermomyces lanuginosus and Rhizomucor miehei lipases for the transesterification and hydrolysis of soybean oil. Process Biochem 46, 682-688. https://doi.org/10.1016/j.procbio.2010.11.013 ; Vescovi et al., 2016Vescovi V, Rojas MJ, Baraldo A, Botta DC, Santana FAM, Costa JP, Machado MS, Honda VK, Giordano RLC, Tardioli PW. 2016. LipaseCatalyzed Production of Biodiesel by Hydrolysis of Waste Cooking Oil Followed by Esterification of Free Fatty Acids. J. Am. Oil Chem. Soc. 93, 1615-1624. https://doi.org/10.1007/s11746-016-2901-y ). Although the use of these catalysts still faces problems related to low reaction rates and the need for long periods of time to achieve high yields, their use in the sequential process has stood out, demonstrating its potential to overcome these drawbacks (Wancura et al., 2019Wancura, JHC, Rosset DV, Mazutti MA, Ugalde GA, Oliveira V, Tres MV, Jahn SL. 2019. Improving the soluble lipase-catalyzed biodiesel production through a two-step hydroesterification reaction system. Appl. Microbiol. Biotechnol. 103, 7805-7817. https://doi.org/10.1007/s00253-019-10075-y ).
The use of organic solvents as reaction media for enzymatic reactions provides attractive advantages over traditional systems, such as increased reaction yield over increased substrate solubility, suppression of water-dependent reactions and elimination of microbial contamination (Raspe et al., 2013Raspe DT, Silva C, Cardozo-Filho L. 2013. Effect of additives and process variables on enzymatic hydrolysis of macauba kernel oil (Acrocomia aculeata). Chem. Eng. J. 2013, 1-8. https://doi.org/10.1155/2013/438270 ), besides the influence on catalytic activity and enzyme stability caused by the nature of these solvents. In contrast, other authors report that their effect causes the inactivation of enzymes, high solvent cost, limitations in mass transfer for heterogeneous systems or systems with high viscosity solvents/substrates (Doukyu and Ogino, 2010Doukyu N, Ogino H. 2010. Organic solvent-tolerant enzymes. Biochem. Eng. J. 48, 270-282. https://doi.org/10.1016/j.bej.2009.09.009 ).
Therefore, the aim of this study was to evaluate the production of esters from the enzymatic hydroesterification of macauba oil in s two-step reaction: oil hydrolysis followed by esterification of the hydrolyzate obtained. The effects of the experimental variables (buffer solution percentage and catalyst concentration) were investigated in the hydrolysis step in order to maximize the free fatty acid (FFA) yield, and to determine the effect of reaction time. The hydrolyzate obtained (with maximum FFA content) was directed to the esterification step. In addition, the reuse of the enzymes used in the hydrolysis and esterification steps was evaluated.
2. MATERIALS AND METHODS
⌅2.1. Materials
⌅Macauba kernel oil (Cocal Brasil) was used in the reactions, and its chemical composition was previously reported by Raspe et al. (2013)Raspe DT, Silva C, Cardozo-Filho L. 2013. Effect of additives and process variables on enzymatic hydrolysis of macauba kernel oil (Acrocomia aculeata). Chem. Eng. J. 2013, 1-8. https://doi.org/10.1155/2013/438270 . Sodium phosphate buffer (Neon), enzyme Lipozyme® Rhizomucor miehei (Sigma-Aldrich) and n-hexane (Nuclear) were used in the hydrolysis step. In the esterification reactions, the hydrolyzate obtained from the hydrolysis step, methanol (Panreac, 99.9% purity) and enzyme Novozym® 435 (Candida antarctica lipase immobilized) were used. Heptane (Nuclear) and ethanol (Anidrol) were used to wash the enzymes in the catalyst reuse tests. In titration step of the samples, a solution of ethyl ether:ethanol 2:1 (v:v) (Vetec/Nuclear), potassium hydroxide (Nuclear), and phenolphthalein as indicator (Nuclear) were used.
2.2. Experimental procedure
⌅The hydrolysis reaction was carried out in a magnetically stirred, jacketed flask (40 mL) connected to a constant temperature bath (Marconi) for temperature monitoring. The reaction was conducted at 55 ºC, with agitation of 400 rpm and the reaction medium was composed of macauba kernel oil, sodium phosphate buffer solution (pH 8.0), n-hexane (oil to n-hexane mass ratio of 1:1) and Lipozyme® Rhizomucor miehei (RM IM) as catalyst (Raspe et al., 2013Raspe DT, Silva C, Cardozo-Filho L. 2013. Effect of additives and process variables on enzymatic hydrolysis of macauba kernel oil (Acrocomia aculeata). Chem. Eng. J. 2013, 1-8. https://doi.org/10.1155/2013/438270 ). The enzyme was maintained at 40 ºC for 1 hour for activation before its addition to the reaction medium. After the reaction time of 6 hours, enzymes were separated by filtration and two phases (oil + solvent and water) were separated by centrifugation and the solvent in the oil phase was dried in an oven to evaporate the excess solvent.
An experimental central composite design (with axial points) was applied to evaluate the effects of process variables on FFA yield using Statistica® 8.0 software (STATSOFTTM, Inc.). Buffer solution percentage (A) and catalyst concentration (B) were the variables investigated in the enzymatic hydrolysis, and these factors varied, as shown in Table 1. A total of 11 experiments with different combinations of levels of the variables were performed in duplicate, and the mean values ± standard deviation of the results were reported.
| Levels | |||||
|---|---|---|---|---|---|
| Factors | (-1.41) | (-1) | (0) | (+1) | (+1.41) |
| (A) Buffer (in relation to oil mass) | 21.71 | 10 | 30 | 50 | 78.28 |
| (B) Enzyme (in relation to substrate mass) | 7.92 | 5 | 7.5 | 10 | 22.07 |
A second-order polynomial model (Barbosa et al., 2019Barbosa MS, Freire CCC, Almeida LC, Freitas LS, Souza RL, Pereira EB, Mendes AA, Pereira MM, Lima AS, Soares CMF. 2019. Optimization of the enzymatic hydrolysis of Moringa oleifera Lam oil using molecular docking analysis for fatty acid specificity. Biotechnol. Appl. Biochem. 66, 823-832. https://doi.org/10.1002/bab.1793 ) was adjusted in relation to the responses obtained and the variables investigated. Analysis of variance (ANOVA) was used to evaluate the effects of operational variables and their interactions on the proposed model based on the values of p-value and F, where p < 0.05 was used as the threshold of statistical significance.
The effect of reaction time was determined from the conduction of destructive kinetics (in duplicate) in the times of 1, 2, 4, 6, 8 and 10 hours. The reactions were conducted keeping the temperature and buffer solution fixed at 55 ºC and 50 wt% (in relation to the oil mass), respectively, with the evaluation of the addition of lipase in concentrations of 10, 15 and 20 wt% (in relation to the substrate’s mass).
The reactions with the macauba oil hydrolyzate were conducted keeping the temperature fixed at 65 ºC and a methanol to FFA molar ratio of 3:1. The reaction conditions were selected based on the work of Cerveró et al. (2014)Cerveró JM, Álvarez JR, Luque S. 2014. Novozym 435-catalyzed synthesis of fatty acid ethyl esters from soybean oil for biodiesel production. Biomass Bioenerg. 61, 131-137. https://doi.org/10.1016/j.biombioe.2013.12.005 . Preliminary tests were conducted with different catalysts (Lipozyme® Rhizomucor miehei, Lipozyme® Thermomyces lanuginosus and Novozym® 435), which would indicate higher conversions with the use of Novozym® 435 with percentage in the reactions of 10 wt% (relation to substrates mass). These reactions were performed with magnetic stirrers in a batch reactor equipped with condenser and immersed in a temperature-controlled water bath. The hydrolyzate (4 g) was heated until the desired temperature was reached. At this point, methanol and the catalyst (after activation at 40 ºC for 1 hour) were added and esterification began. At the end of each reaction, the catalyst was separated by centrifugation (3000 rpm for 10 minutes), and the alcohol and water were removed from the reaction mixture using a rotary evaporator.
2.3. Analytical method
⌅The content in free fatty acids (FFA) was determined based on the method Ca 5a-40 (AOCS, 1998AOCS, American Oil Chemists’ Society. 1998. Official methods and recommended practices. 2, 4 (Ed.) Champaign.), which is based on acid-base titration using an ethanol solution of potassium hydroxide (KOH) previously standardized as the titrant. Each sample was titrated in duplicate and the FFA content was calculated from Equation 1:
where C is the concentration of sodium hydroxide (mol L−1) used as titrant, MM corresponds to the molar mass of the predominant fatty acids in the sample, v is the volume required for the titration (mL) and m is the mass of sample (g).
The FFA yield of the hydrolysis reactions was calculated from Equation 2:
where FFA corresponds to FFA content produced after the hydrolysis reaction and the CHI content in compounds present in the macauba oil that can be hydrolyzed (considering the initial content of FFA of 23.0 ± 0.4 wt%) reported by Raspe et al. (2013)Raspe DT, Silva C, Cardozo-Filho L. 2013. Effect of additives and process variables on enzymatic hydrolysis of macauba kernel oil (Acrocomia aculeata). Chem. Eng. J. 2013, 1-8. https://doi.org/10.1155/2013/438270 .
The macauba oil hydrolysate used was characterized in terms of the free fatty acid and water contents using the official methods recommended by the AOCS (1990)AOCS, American Oil Chemists’ Society. 1998. Official methods and recommended practices. 2, 4 (Ed.) Champaign.: Ca 5a40 and 984.20, respectively. The glycerol content was determined by titration, using the sodium periodate method described by Cocks and Van Rede (1996)Cocks LV, Van Rede C. 1966. Laboratory Handbook for Oil and Fat Analysis. Londres. Acad. Press..
The conversion of the esterification reaction (FFA conversion) was determined according to Equation 3:
where FFAi is the initial FFA content in the hydrolyzate and FFAf is the FFA content in the final sample of the reaction medium.
2.4. Reuse of lipase
⌅For the reuse assays of the Lipozyme® RM IM and Novozym® 435, batches of hydrolysis and esterification were repeated for 15 cycles of 6 hours and 1 hour, respectively. After each reaction, the biocatalyst was recovered by filtration, washed with heptane and ethanol to remove adsorbed products and dried in oven at 40 ºC for 1 hour, kept in a desiccator for 24 hours and reused in another batch.
3. RESULTS
⌅3.1. Enzymatic hydrolysis of macauba oil
⌅Table 2 presents the results obtained for the reactions conducted in order to evaluate the effect of the process variables for obtaining a hydrolyzate rich in FFA.
| Run | Variables1 | FFA yield2 (%) | |
|---|---|---|---|
| A | B | ||
| 1 | 30 (-1) | 10 (-1) | 69.03 ± 0.38 |
| 2 | 30 (-1) | 20 (+1) | 78.09 ± 0.29 |
| 3 | 70 (+1) | 10 (-1) | 71.08 ± 0.82 |
| 4 | 70 (+1) | 20 (+1) | 80.67 ± 0.40 |
| 5 | 21.71 (-1.41) | 15 (0) | 76.67 ± 0.90 |
| 6 | 78.28 (+1.41) | 15 (0) | 80.15 ± 0.48 |
| 7 | 50 (0) | 7.92 (-1.41) | 68.56 ± 0.36 |
| 8 | 50 (0) | 22.07 (+1.41) | 80.17 ± 0.05 |
| 9.1 | 50 (0) | 15 (0) | 77.46 ± 0.05 |
| 9.2 | 50 (0) | 15 (0) | 78.39 ± 0.27 |
| 9.3 | 50 (0) | 15 (0) | 78.49 ± 0.75 |
1(A) Buffer solution percentage (in relation to oil mass) and Enzyme concentration (in relation to substrate mass); 2 calculated according FFA content produced after the hydrolysis reaction and CHI content of compounds present in the macauba oil which can be hydrolyzed (23.0 ± 0.4 wt%), mean value (2 replicates) ± standard deviation.
The ANOVA of the quadratic model adjusted to the experimental data is presented in Table 3. Significant terms (p < 0.05) were obtained, which indicates that the experimental data can adequately describe the model proposed. The F values of 62.53 indicate that the models were significant, since these values were higher than the Fcritic value (8.89). In addition, the values for R2 and R2 adjusted, calculated considering only the significant parameters, showed that the variability of the data (> 90%) is adequately explained by the regression model, which indicates good linearity between the predicted data and the observed data.
| Sum of squares | Degree of freedom | Mean square | F | p 1 | |
|---|---|---|---|---|---|
| A (L) | 11.38 | 1 | 11.38 | 35.39 | 0.027 |
| A (Q) | 0.40 | 1 | 0.40 | 1.26 | 0.376 |
| B (L) | 153.71 | 1 | 153.71 | 478.08 | 0.002 |
| B (Q) | 29.66 | 1 | 29.66 | 92.26 | 0.010 |
| A*B | 0.07 | 1 | 0.07 | 0.22 | 0.684 |
| Lack of Fit | 6.19 | 3 | 2.06 | 6.41 | 0.137 |
| Pure Error | 0.64 | 2 | 0.32 | ||
| Total SS | 202.67 | 10 | |||
| R2 = 0.964 | |||||
| R2 adjusted = 0.949 |
1 Statistical significance (p < 0.05); L - linear effect and Q - quadratic effect.
Table 3 shows that linear and quadratic terms of all variables were significant for the adjusted model, except for the buffer percentage, which showed influence only on the linear term and the binary interaction which was not significant. From the adjusted model, the linear term of the buffer percentage was the variable that presented a higher F value and a lower p value. The polynomial model for the FFA yield (%) was regressed considering the significant terms as presented in Equation 4:
3.1.1. Effect of process variables
⌅The variables evaluated in the experimental design have a greater influence on the FFA yield (based on the experimental range considered). The greater amount of buffer solution in the reaction medium caused an increase in the interfacial area of the oil-water system, providing a greater number of bonds between the substrates to be catalyzed by the lipase (Zhou et al., 2015Zhou GX, Chen GV, Yan BB. 2015. Two-step biocatalytic process using lipase and whole cell catalysts for biodiesel production from unrefined jatropha oil. Biotechnol. Lett. 37, 1959-1963. https://doi.org/10.1007/s10529-015-1883-4 ). In addition, when the percentage of the buffer solution was increased, there was less variation in the pH of the reaction medium and less aggregation (McClements and Weiss, 2005McClements DJ, Weiss J. 2005. Em Bailey’s industrial oil and fat products. John Wiley & Sons Inc.: New York.). In addition, a higher proportion of water changed the balance in favor of the products, improving the reaction rate in each of the hydrolysis steps and accelerating their completion (Wang et al., 2012Wang WC, Turner TL, Stikeleather LF, Roberts WL. 2012. Exploration of process parameters for continuous hydrolysis of canola oil, camelina oil and algal oil. Chem. Eng. Process 57-58, 51-58. https://doi.org/10.1016/j.cep.2012.04.001 ). This was possibly because lipase, which is a surface-active enzyme, bound with the substrates at the oil-water interface and, with the increased addition of water, the amount of water available for oil to form oil-water droplets increased, thereby increasing the available interfacial area, since the lipase catalyzes the hydrolysis reaction at the interfacial area of emulsion (Nguyen et al., 2017Nguyen VTA, Le TD, Phan HN, Tran LB. 2017. Antibacterial Activity of Free Fatty Acids from Hydrolyzed Virgin Coconut Oil Using Lipase from Candida rugosa. J. Lipids 2017, 1-7. https://doi.org/10.1155/2017/7170162 ).
Santos et al. (2015)Santos LDF, Coutinho JAP, Ventura SPM. 2015. From Water-in-Oil to Oil-in-Water Emulsions to Optimize the Production of Fatty Acids Using Ionic Liquids in Micellar Systems. Biotechnol. Prog. 31, 1473-1480. https://doi.org/10.1002/btpr.2156 observed an increase in the hydrolysis yield from ~35 to 100% FFA with the use of 50 wt% and 90 wt% buffer solution in the reaction, respectively. Zhou et al. (2015)Zhou GX, Chen GV, Yan BB. 2015. Two-step biocatalytic process using lipase and whole cell catalysts for biodiesel production from unrefined jatropha oil. Biotechnol. Lett. 37, 1959-1963. https://doi.org/10.1007/s10529-015-1883-4 evaluated the hydrolysis of unrefined jatropha oil and found that the application of the highest proportion of water (relation to oil mass) (100 wt%) resulted in obtaining ~88 wt% FFA, while using the proportion of 50 wt%, ~75 wt% FFA was obtained. Barbosa et al. (2019)Barbosa MS, Freire CCC, Almeida LC, Freitas LS, Souza RL, Pereira EB, Mendes AA, Pereira MM, Lima AS, Soares CMF. 2019. Optimization of the enzymatic hydrolysis of Moringa oleifera Lam oil using molecular docking analysis for fatty acid specificity. Biotechnol. Appl. Biochem. 66, 823-832. https://doi.org/10.1002/bab.1793 reported a ~150% increase in the hydrolysis degree of Moringa oleifera Lam oil by varying the oil-to-water mass ratio from 15 to 35 wt%.
The higher catalyst concentration in the reaction medium favored the achievement of higher values for FFA yield, which is the result of increased contact between the substrate and the active sites of the lipase and the cumulative adsorption of the enzyme at the oil-water interface (Santos et al., 2015Santos LDF, Coutinho JAP, Ventura SPM. 2015. From Water-in-Oil to Oil-in-Water Emulsions to Optimize the Production of Fatty Acids Using Ionic Liquids in Micellar Systems. Biotechnol. Prog. 31, 1473-1480. https://doi.org/10.1002/btpr.2156 ), leading to increased hydrolysis rates. In general, the reaction rate increases with the greater availability of enzyme in the reaction medium. Zenevicz et al. (2016)Zenevicz MCP, Jacques A, Furigo A, Oliveira V, Oliveira D. 2016. Enzymatic hydrolysis of soybean and waste cooking oils under ultrasound system. Ind. Crops Prod. 80, 235-241. https://doi.org/10.1016/j.indcrop.2015.11.031 found a 20% increase in the FFA content obtained from the hydrolysis of soybean oil by increasing the Lipozyme® TL IM percentage from 1 to 10 wt% (based on the total mass of substrates). For the hydrolysis of soybean oil, Corradini et al. (2019)Corradini FAS, Alves ES, Kopp W, Ribeiro MPA, Mendes AA, Tardioli PW, Giordano RC, Giordano RLC. 2019. Kinetic study of soybean oil hydrolysis catalyzed by lipase from solid castor bean seeds. Chem. Eng. Res. Des. 144, 115-122. https://doi.org/10.1016/j.cherd.2019.02.008 obtained 1800 mM and ~500 mM of FFAs using 6 and 2 g of castor seed lipase, respectively.
3.1.2. Maximization of FFA yield
⌅The conditions that maximized the production of FFA from the enzymatic hydrolysis of macauba oil were 50 wt% buffer solution and 20 wt% enzyme, with predicted FFA yields of 80.2%. Verification experiments were conducted (in triplicate) and provided FFA yield of 82.65 ± 0.57%. The predicted experimental values were compared and according to the t-Student test, there was agreement between these values in a significance interval of 0.05, which shows the predictive capacity of the adjusted models.
3.1.3. Effect of reaction time
⌅Considering that for the investigated system the enzyme concentration had a greater influence on the FFA production, the kinetic of the reaction was determined keeping the temperature and buffer solution fixed at 55 ºC and 50 wt% (in relation to the oil mass), respectively, with evaluation of the addition of lipase in concentrations of 10, 15 and 20 wt% (in relation to the substrates mass).
Figure 1 presents the results obtained for the reactions carried out in the interval from 1 to 10 hours and according to this figure it can be seen that the gradual increase in the reaction time resulted in higher FFA yields, with the maximum value of ~88.90% (corresponding to FFA content of 92.08 wt%) obtained using 20 wt% catalyst and after 8 hours of reaction. For the reactions conducted with 15 and 20 wt% of catalyst, there was no increase in yield after 8 hours, indicating that the process equilibrium was reached.
Rodrigues and Ayub (2011)Rodrigues RC, Ayub MAZ. 2011. Effects of the combined use of Thermomyces lanuginosus and Rhizomucor miehei lipases for the transesterification and hydrolysis of soybean oil. Process Biochem 46, 682-688. https://doi.org/10.1016/j.procbio.2010.11.013 reported yields in the order of 95% FFA after 10 hours of reaction when investigating the hydrolysis of soybean oil using a water-to-soybean oil molar ratio of 3:1, 25 wt% (in relation to mass of oil) of the biocatalyst mixture (combination of 65% Thermomyces lanuginosus and 35% Rhizomucor miehei) at 30 °C. After 2 hours of reaction at 40 °C, with an oil-to-water molar ratio of 1:20 and 10 wt% of Lipozyme® TL IM (in relation to the substrates mass), Zenevicz et al. (2016)Zenevicz MCP, Jacques A, Furigo A, Oliveira V, Oliveira D. 2016. Enzymatic hydrolysis of soybean and waste cooking oils under ultrasound system. Ind. Crops Prod. 80, 235-241. https://doi.org/10.1016/j.indcrop.2015.11.031 obtained a maximum yield of 60% FFA in the hydrolysis of soybean oil. Vescovi et al. (2016)Vescovi V, Rojas MJ, Baraldo A, Botta DC, Santana FAM, Costa JP, Machado MS, Honda VK, Giordano RLC, Tardioli PW. 2016. LipaseCatalyzed Production of Biodiesel by Hydrolysis of Waste Cooking Oil Followed by Esterification of Free Fatty Acids. J. Am. Oil Chem. Soc. 93, 1615-1624. https://doi.org/10.1007/s11746-016-2901-y obtained 100% FFA yield in the hydrolysis of frying oil catalyzed by the immobilized lipase of Thermomyces lanuginosus, in a ratio of oil-to-water of 1:4 (v/v) and enzyme/reaction medium of 1:100 (w/v), at 30 °C and 24 h. After 40 hours of hydrolysis catalyzed by Lipozyme® RM IM, Tavares et al. (2018)Tavares F, Silva EAD, Pinzan F, Canevesi RS, Milinsk MC, Scheufele FB, Borba CE. 2018. Hydrolysis of crambe oil by enzymatic catalysis: An evaluation of the operational conditions. Biocat. Biotransf. 36, 1-14. https://doi.org/10.1080/10242422.2018.1430786 obtained a maximum FFA yield of 74% from crambe oil, under experimental conditions of 2.7 wt% of lipase (in relation to the mass of substrates) and water-to-oil molar ratio of 10:1.
3.1.4. Characterization of hydrolyzate
⌅The macauba oil hydrolyzate collected at 55 ºC, with 50 wt% buffer solution, 20 wt% lipase and 8 hours of reaction showed a free fatty acid content of 92.08 ± 0.28 wt%, water content of 0.865 ± 0.07 wt% and glycerol content of 0.64 ± 0.002 wt%, respectively.
3.1.5. Reuse of biocatalyst
⌅Figure 2 shows the evaluation of the reuse of the enzyme catalyst in reactions conducted at 55 ºC, 50 wt% buffer solution and 20 wt% lipase, evaluated for 15 cycles of 6 hours each. From the data in Figure 2, it can be seen that the efficiency of the lipase declines in the course of its reuse, obtaining ~50% lower yield after 15 cycles compared to cycle 1. The loss in activity observed may be related to the saturation of the active sites of the enzyme during the reaction, since upon reaching its maximum activity, the interfacial effects and obstacles to mass transfer imply a decrease in reaction rates, preventing the enzyme from absorbing more substrate (Corradini et al., 2019Corradini FAS, Alves ES, Kopp W, Ribeiro MPA, Mendes AA, Tardioli PW, Giordano RC, Giordano RLC. 2019. Kinetic study of soybean oil hydrolysis catalyzed by lipase from solid castor bean seeds. Chem. Eng. Res. Des. 144, 115-122. https://doi.org/10.1016/j.cherd.2019.02.008 ). Kabbashi et al. (2015)Kabbashi NA, Mohammed NI, Alam MZ, Mirghani MES. 2015. Hydrolysis of Jatropha curcas oil for biodiesel synthesis using immobilized Candida cylindracea lipase. J. Mol. Catal. B. Enzym. 116, 95-100. https://doi.org/10.1016/j.molcatb.2015.03.009 repored that the decrease in product yield may be attributed to desorption of the enzyme from the support and inactivation upon repeated reuse.
Rodrigues and Ayub (2011)Rodrigues RC, Ayub MAZ. 2011. Effects of the combined use of Thermomyces lanuginosus and Rhizomucor miehei lipases for the transesterification and hydrolysis of soybean oil. Process Biochem 46, 682-688. https://doi.org/10.1016/j.procbio.2010.11.013 verified a drop of ~80% in the hydrolysis yield of soybean oil catalyzed by the mixture of Thermomyces lanuginosus and Lipozyme® RM IM after 10 cycles of 10 hours each, relating this behavior to the lack of washing of the catalysts at the end of each process. Vescovi et al. (2016)Vescovi V, Rojas MJ, Baraldo A, Botta DC, Santana FAM, Costa JP, Machado MS, Honda VK, Giordano RLC, Tardioli PW. 2016. LipaseCatalyzed Production of Biodiesel by Hydrolysis of Waste Cooking Oil Followed by Esterification of Free Fatty Acids. J. Am. Oil Chem. Soc. 93, 1615-1624. https://doi.org/10.1007/s11746-016-2901-y when evaluating the reuse of Thermomyces lanuginosus lipase in the hydrolysis of residual cooking oil, determined that enzyme activity decreased during reuse in proportions similar to those reported in this work (~50%), although after only five cycles (10 hours each). According to the authors, this decrease in yield is due to the low pH of the reaction medium (around pH 4.6 after 10 hours of hydrolysis), which probably caused enzyme inactivation. Assessing the reuse of Lipozyme® TL IM in the enzymatic hydrolysis of soybean oil, Zenevicz et al. (2016)Zenevicz MCP, Jacques A, Furigo A, Oliveira V, Oliveira D. 2016. Enzymatic hydrolysis of soybean and waste cooking oils under ultrasound system. Ind. Crops Prod. 80, 235-241. https://doi.org/10.1016/j.indcrop.2015.11.031 observed that the process maintained the yield at ~60% FFA after 4 cycles (2 hours each).
3.2. Reaction carried out with macauba oil hydrolyzate
⌅The esterification step of the FFA in solvent medium (50 wt% of n-hexane in relation to the hydrolysate mass) was carried out at different times, as shown in Figure 3. From the data in this figure, it appears that the reaction rate is high in short reaction times (15 min), reaching equilibrium in 60 min with ~95% conversion of FFA.
When investigating the influence of operational conditions on the use of Novozym® 435 in FFA esterification with methanol, Mulalee et al. (2015)Mulalee S, Srisuwan P, Phisalaphong M. 2015. Influences of operating conditions on biocatalytic activity and reusability of Novozym 435 for esterification of free fatty acids with short-chainalcohols: A case study of palm fatty acid distillate. Chin. J. Chem. Eng. 23, 1851-1856. https://doi.org/10.1016/j.cjche.2015.08.016 reported ~ 95% conversion with 5 wt% catalyst (in relation to oleic acid), with a methanol-to-FFA molar ratio of 2:1, at 45 °C after 8 hours. ~97% FFA conversion was obtained by Teixeira et al. (2017)Teixeira DA, Motta CR, Ribeiro CMS, Castro AM. 2017. A rapid enzyme-catalyzed pretreatment of the acidic oil of macauba (Acrocomia aculeata) for chemoenzymatic biodiesel production. Process Biochem. 53, 188-193. https://doi.org/10.1016/j.procbio.2016.12.011 in the esterification of FFA of macauba oil, conducted at a methanol-to-FFA molar ratio of 2:1, 5 wt% of Lipozym® 435 (in relation to FFA mass), 30 °C and 60 min. Rosset et al. (2019)Rosset DV, Wancura JHC, Ugalde GA, Oliveira JV, Tres MV, Kuhn RC. Jahn SL. 2019. Enzyme-Catalyzed Production of FAME by Hydroesterification of Soybean Oil Using the Novel Soluble Lipase NS 40116. Biotechnol. Appl. Biochem. 188, 914-926. https://doi.org/10.1007/s12010-019-02966-7 reported 94.3% conversion in the esterification of soybean oil hydrolyzate with the lipase NS 40116 (enzyme-in-liquid formulation from genetically-modified Thermomyces lanuginosus microorganism) at 35 °C, methanol-to-oil molar ratio of 4.5:1 and 12 hours of reaction.
3.2.1. Reuse of biocatalyst
⌅The reuse of the Novozym® 435 in the esterification reaction was evaluated under the conditions reported in Figure 3 for the reaction time of 1 hour and during 15 cycles, as shown in Figure 4. It was verified from the results that the lipase maintained ~98% of its initial conversion capacity at the end of the evaluated cycles. The maintenance of the catalytic activity of the lipase is related to the low solubility of its support during the reactions (Shin et al., 2020Shin M, Seo J, Baek Y, Lee T, Jang M, Park C. 2020. Novel and Efficient Synthesis of Phenethyl Formate via Enzymatic Esterification of Formic Acid. Biomolecules 10, 1-15. https://doi.org/10.3390/biom10010070 ), promoted by operating conditions and the alcohol in the process. Adequate temperature and agitation do not weaken the enzyme support and do not promote its interfacial inactivation (Ortiz et al., 2019Ortiz C, Ferreira ML, Barbosa O, Dos Santos JCS, Rodrigues R.C., Berenguer-Murcia, A., Briand, L.E., Fernandez-Lafuente, R. 2019. Novozym 435: the “perfect” lipase immobilized biocatalyst? Catal. Sci. Technol. 9, 2380-2420. https://doi.org/10.1039/C9CY00415G ). In addition, methanolic esterification promotes less swelling in the reuse of Novozym® 435, causing less loss in activity and degree of catalytic deactivation (Mulalee et al., 2015Mulalee S, Srisuwan P, Phisalaphong M. 2015. Influences of operating conditions on biocatalytic activity and reusability of Novozym 435 for esterification of free fatty acids with short-chainalcohols: A case study of palm fatty acid distillate. Chin. J. Chem. Eng. 23, 1851-1856. https://doi.org/10.1016/j.cjche.2015.08.016 ). At the sqame time, the recovery of the enzyme by washing with heptane may also be responsible for maintaining the stability of Novozym® 435, due to the greater removal/dissolution of the constituents linked to the active sites of this enzyme (Chowdhury and Mitra, 2015Chowdhury A, Mitra D. 2015. A kinetic study on the Novozyme 435-catalyzed esterification of free fatty acids with octanol to produce octyl esters. Biotechnol. Prog. 31, 1494-1499. https://doi.org/10.1002/btpr.2165 ), restoring the catalyst activity almost completely.
Baek et al. (2020)Baek Y, Lee J, Son J, Leem T, Sobhan A, Lee J, Koo SM, Shin WH, Oh JM, Park C. 2020. Enzymatic Synthesis of Formate Ester through Immobilized Lipase and Its Reuse. Polym. 12, 1-10. https://doi.org/10.3390/polym12081802 performed the enzymatic synthesis of formate ester through immobilized lipase, and observed that Novozym® 435 could be reused for 10 cycles of 1 hour, keeping the conversions at ~92%. Moreira et al. (2020)Moreira KS, Moura Junior LS, Monteiro RRC, Oliveira ALB, Valle CP, Freire TM, Fechine PBA, Souza MCM, Fernandez-Lorente G, Guisan JM, Santos JCS. 2020. Optimization of the Production of Enzymatic Biodiesel from Residual Babassu Oil (Orbignya sp.) via RSM. Catal. 10, 1-20. https://doi.org/10.3390/catal10040414 reported that Novozyme® 435 maintained catalytic activity at the end of 10 consecutive cycles in the enzymatic esterification of babassu FFA in reactions at 48 °C and duration of 4 hours.
4. CONCLUSIONS
⌅The present study evaluated the hydroesterification of macauba oil using enzymatic catalysis, with evaluation of the processes of hydrolysis of macauba oil and esterification of the hydrolyzate. Evaluating the effects of the process variables, it was found a yield of ~ 80% in FFA, through positive and significant effects for the percentage of buffer solution and concentration of catalyst in the reaction medium. A hydrolyzate with ~92% FFA was obtained by evaluating the percentage of 20 wt% of catalyst (in relation to oil mass), temperature of 55 ºC, stirring 400 rpm and 50 wt% of buffer solution (in relation to mass of substrates) in the hydrolysis kinetics after 8 hours of reaction. The conversion of the hydrolyzate in the esterification step was evaluated at 65 ºC, methanol to FFA of 3:1 and 10 wt% (in relation to substrates mass) of catalyst, where 95% conversion of the FFA was achieved. In the reuse of the catalysts, the efficiency of the Lipozyme® RM IM lipase decreased ~50% the FFA yield in hydrolysis after 90 hours of the process, while Novozym® 435 maintained ~98% of its initial conversion capacity, at the end of the 15 cycles investigated (15 hours).