Lipid Extraction Maximization and Enzymatic Synthesis of Biodiesel from Microalgae

: Microalgae has received overwhelming attention worldwide as a sustainable source for energy generation. However, the production of biofuel from microalgae biomass consists of several steps, of which lipid extraction is the most important one. Because of the nature of feedstock, extraction needs special attention. Three di ﬀ erent methods were studied to extract algal oil from two di ﬀ erent algae variant, Chlorella sp. and Spirulina sp. The highest percentage oil yield was obtained by ultrasonication (9.4% for Chlorella sp., 6.6% for Spirulina sp.) followed by the Soxhlet and solvent extraction processes. Ultrasonication and Soxhlet extraction processes were further optimized to maximize oil extraction as solvent extraction was not e ﬀ ective in extracting lipid. For ultrasonication, an amplitude of 90% recorded the highest percentage yield of oil for Spirulina sp. and a 70% amplitude recorded the highest percentage yield of oil for Chlorella sp. On the other hand, for Soxhlet extraction, a combination of chloroform, hexane, and methanol at a 1:1:1 ratio resulted in the highest yield of algal oil. Afterward, the crude algae oil from the ultrasonication process was transesteriﬁed for 5 h using an immobilized lipase (Novozyme 435) at 40 ◦ C to convert triglycerides into fatty acid methyl ester and glycerol. Thus, ultrasonic-assisted lipid extraction was successful in producing biodiesel from both the species. Contributions: Conceptualization, M.Y.N. and H.C.O.; Data curation, I.M.R.F.; Formal analysis, M.Y.N. and A.S.S.; Investigation, M.Y.N.; Methodology, M.Y.N. and H.C.O.; Resources, M.M. and A.S.S.; Software, M.M. and T.M.I.M.; Supervision, H.C.O. and T.M.I.M.; Validation, I.A.B. and T.M.I.M.; Writing—original draft, M.Y.N.; Writing—review and editing, I.M.R.F. and T.M.Y.K.


Introduction
Depleting oil reserves, increased energy dependency, and the environmental impact of fossil fuel use has led to research on sustainable energy resources with a cleaner footprint [1][2][3]. Biofuel, in particular biodiesel, is a feasible diesel fuel replacement because it can be directly employed without modification of the engine structure [4][5][6]. Biodiesel generally refers to the mixture of fatty acid methyl esters and are derived from lipid substances originated from triacylglycerol-containing feedstocks [7][8][9]. A review of life cycle analysis on liquid biofuel systems has concluded that conventional biofuels (from grains and seeds) can provide moderate greenhouse gas (GHG) reduction benefits in measures, without the presence of a catalyst [39][40][41]. Both homogeneous catalyst and heterogeneous catalyst have been studied exhaustively in the literature. On the other hand, enzymatic catalysts are gaining attention as catalysts in transesterification or alcoholysis, especially from microalgae. Immobilized lipases are extensively used in enzymatic alcoholysis of triglycerides into fatty acid methyl esters (FAMEs) [42]. Recently, Sanjib Kumar Karmee [43] reported the optimization of methanolysis of Manilkara zapota (L.) using lipases, such as Novozyme 435 (C. antarctica lipase-B immobilized on acrylic resin) and CLEA (crosslinked enzyme aggregate) of C. antarctica lipase-B as biocatalyst. The author found that under similar operating conditions, 93% biodiesel was obtained after 12 h of reaction using Novozyme 435, whereas CLEA produced 84% biodiesel. Tran et al. [44] studied biodiesel production from Chlorella vulgaris ESP-31 microalgae by enzymatic transesterification using an immobilized lipase originating from Burkholderia sp. C20. A high conversion efficiency of 97.25 wt.% oil occured when static lipase was utilized. This worked in a high water (>71.39% wt.%) level and high molar ratio settings (>67.93% wt.%).
This work aimed to compare the efficacies of three lipid extraction methods viz. solvent extraction, Soxhlet extraction, and ultrasonication, to obtain microalgal crude oil from Chlorella sp. and Spirulina sp. The maximization of lipids using Soxhlet and ultrasonic methods was attempted by varying the operating parameters to achieve maximum lipid output. Lipids were then converted to biodiesel using the enzymatic transesterification process. The immobilized lipase was used as a catalyst under constant operating conditions. Finally, the biodiesel was characterized using Fourier transform infrared spectroscopy (FTIR) and gas chromatography (GC).

Chemicals and Reagents
The dried microalgae species used in this study were Chlorella sp. and Spirulina sp., which were obtained from a local supplier. Commercial immobilized lipase (Novozyme 435) was obtained from Sigma-Aldrich. Novozym 435 is an immobilized lipase, which is based on immobilization via interfacial activation of lipase-B from Candida antarctica on a resin, Lewatit VP OC 1600. This resin is a macroporous support formed by poly (methyl methacrylate) crosslinked with divinylbenzene [45]. The Novzyme 435 is extensively used in enzymatic transesterification, converting triglycerides into FAMEs. All organic solvents used in this study were of analytical grade obtained from Fisher Scientific.

Lipid Extraction
In this study, three extraction processes were studied, including solvent extraction, Soxhlet extraction, and ultrasonic-assisted extraction.

Lipid Extraction by Ultrasonication
Ultrasonication was carried out by mixing 30 g of dried algae with distilled water, and processed with a Vibra-cell ultrasonicator (Model VCX500, Sonics & Materials, Inc., Newtown, CT, USA). The equipment has a probe that delivers ultrasound at a frequency of 40 kHz. The probe was immersed into the mixture, and ultrasonication was carried out at 70% amplitude for 15 min to disrupt the microalgae cells. Without separating the biomass from the solution phase, the sample was treated with n-hexane, which acted as an extraction solvent [46]. The suspension was thoroughly mixed and allowed to settle into phases. Centrifugation was necessary to separate the algae residue and solvents containing lipids. This was carried out in a centrifuge (Rotofix 46, Hettich Zentrifugen) at 2000 rpm (358× g) for 15 min. The resulting lipid was then placed in a beaker and dried under the hood at room temperature. Additional hexane was then added and evaporated using a rotary evaporator to remove the remaining solvent. This process was performed multiple times to produce a purified product. The crude oil was retained for further transesterification reaction. Figure 1a shows the schematic of lipid extraction using the ultrasonication method. the remaining solvent. This process was performed multiple times to produce a purified product. The crude oil was retained for further transesterification reaction. Figure 1a shows the schematic of lipid extraction using the ultrasonication method.

Lipid Extraction by Soxhlet Apparatus
Thirty grams of dried algae biomass were used to extract lipid using a Soxhlet apparatus equipped with a condenser and heating mantle. It took 8 h to extract the oil at 60 °C using n-hexane as the solvent. The extracted lipid was then filtered several times to separate the residues. The remaining solvent was entirely removed by a rotary evaporator to obtain crude microalgae oil. The crude oil was retained for further transesterification reaction. The processes were repeated to optimize the solvent concentration for crude oil extraction. Figure 1b shows the schematic of lipid extraction using the Soxhlet method.

Lipid Extraction Using the Solvent Extraction Process
In this process, 150 mL hexane was added to 30 g of dried biomass. The mixture was stirred at 300 rpm and kept at 60 °C for 8 h. After extracting the lipid, the sample was centrifuged at 2000 rpm (358× g) for 20 min to remove the residue. In order to eliminate the residual solvent, the top layer mixture was filtered and evaporated multiple times. Figure 1c shows the schematic of lipid extraction using the solvent extraction method.

Lipid Extraction Maximization from Microalgae
Maximization of the lipids was carried out with ultrasonication and the Soxhlet extraction method only as the remaining process did not lead to a high oil yield. The methods from Section 2.2.1 and 2.2.2 were repeated with different solvents or amplitude conditions depending on the process.

Ultrasonic Extraction
Different amplitudes were applied for the ultrasonication process with the methanol:chloroform (1:1, v/v) two-phase system as solvents. The amplitude ranged from a 50% up to a 90% amplitude with an interval of 10%. Ultrasonication was carried out similarly to the process described in Section 2.2.1.

Soxhlet extraction
Meanwhile, the Soxhlet extraction maximization process was carried out using a combination of three different solvents at different volumetric ratios. Solvents of chloroform (C), hexane (H), and

Lipid Extraction by Soxhlet Apparatus
Thirty grams of dried algae biomass were used to extract lipid using a Soxhlet apparatus equipped with a condenser and heating mantle. It took 8 h to extract the oil at 60 • C using n-hexane as the solvent. The extracted lipid was then filtered several times to separate the residues. The remaining solvent was entirely removed by a rotary evaporator to obtain crude microalgae oil. The crude oil was retained for further transesterification reaction. The processes were repeated to optimize the solvent concentration for crude oil extraction. Figure 1b shows the schematic of lipid extraction using the Soxhlet method.

Lipid Extraction Using the Solvent Extraction Process
In this process, 150 mL hexane was added to 30 g of dried biomass. The mixture was stirred at 300 rpm and kept at 60 • C for 8 h. After extracting the lipid, the sample was centrifuged at 2000 rpm (358× g) for 20 min to remove the residue. In order to eliminate the residual solvent, the top layer mixture was filtered and evaporated multiple times. Figure 1c shows the schematic of lipid extraction using the solvent extraction method.

Lipid Extraction Maximization from Microalgae
Maximization of the lipids was carried out with ultrasonication and the Soxhlet extraction method only as the remaining process did not lead to a high oil yield. The methods from Sections 2.2.1 and 2.2.2 were repeated with different solvents or amplitude conditions depending on the process.

Ultrasonic Extraction
Different amplitudes were applied for the ultrasonication process with the methanol:chloroform (1:1, v/v) two-phase system as solvents. The amplitude ranged from a 50% up to a 90% amplitude with an interval of 10%. Ultrasonication was carried out similarly to the process described in Section 2.2.1.

Soxhlet extraction
Meanwhile, the Soxhlet extraction maximization process was carried out using a combination of three different solvents at different volumetric ratios. Solvents of chloroform (C), hexane (H), and methanol (M) were used in this study and combined according to these four different ratios, C:H:M of 1:1:1; 2:1:3; 3:2:1; and 1:2:3. The Soxhlet extraction process was carried out for 8 h at 60 • C.

Transesterification of Microalgal Oil to Biodiesel
Traditionally, transesterification is carried out with methanol in the presence of homogeneous catalysts, such as KOH, NaOH, etc. [47,48]. In this study, microalgae oil was transesterified using Appl. Sci. 2020, 10, 6103 5 of 18 methanol in the presence of lipase-catalyst to produce FAMEs following s similar procedure reported previously [49]. In this reaction, methanol acts as the acyl acceptor. The extracted lipid was transferred into a glass container in the presence of 5% (w/w) Novozyme 435 catalyst. The solvents, including methanol and hexane, were then applied to the container to complete the transesterification process. Instead of adding all the methanol at once, it was added in steps to avoid inhibition based on the process described in the literature [50,51]. The reaction was carried out at four different temperatures from 30 • C with a stepwise increase of 5 • C for 5 h at 250 rpm. Samples were drawn out and kept for further analysis by FTIR and GC.

GC Analysis
A GC (Model: 7890A, Agilent Technologies Inc., Santa Clara, CA, USA) was used to quantify the total FAME content (C6:0 to C24:0) and methyl linolenate content (C18:3) of the methyl esters by comparison with the standard test method. An Agilent HP-INNOWax column (Length × inner diameter × film thickness: 30 m × 0.25 mm × 0.25 µm) was used. The sample was prepared for analysis by measuring 100 mg of ME sample in a 10-mL tube with 100 mg of added methyl nonadecanoate (C19). The mixture was diluted with 10 mL of toluene before transferring into a gas chromatograph vial. About 1 µL of the transferred sample was used for the GC measurements. The temperatures of the injector and the detector were set at 250 • C. In this experiment, the temperature increased at a rate of 10 • C/min until the oven temperature reached 200 • C, and continued to heat at a rate of 5 • C/min until it reached 240 • C, which was maintained for 7 min. After that, the fatty acid composition (FAC of the biodiesel was analyzed according to the EN 14103:2011 standard test method.

Analysis of Oil Yield and FAME Percentage
The method of calculating the percentage of oil yield for each extraction is as follows: The acid value of the produced oil was measured using an acid value tester (Mettler Toledo) following ASTM D664. The measurements were triplicated, and the average was taken. The produced biodiesel was analyzed using FTIR (Model: Spectrum 400, Perkin Elmer, Waltham, MA, USA), according to ASTM D7371. The flowchart of the overall process is presented in Figure 2.

Lipid Extraction by Different Methods
Choosing the right extraction method is crucial to obtain a high yield of crude oil. The efficiency of each extraction method is generally determined by the percentage of oil yield obtained after each extraction process. In this study, three different modes of microalgae biomass extraction were

Lipid Extraction by Different Methods
Choosing the right extraction method is crucial to obtain a high yield of crude oil. The efficiency of each extraction method is generally determined by the percentage of oil yield obtained after each extraction process. In this study, three different modes of microalgae biomass extraction were compared to determine the highest oil yield from two different species of microalgae viz. Spirulina sp. and Chlorella sp. The oil yields after 15 min of ultrasonication for Chlorella sp. and Spirulina sp. were 2.83 and 1.98 g, respectively. The oil yield after 8 h of Soxhlet extraction for those microalgae species was 1.05 and 1.25 g, respectively. The oil yield after 8 h of solvent extraction was 0.68 and 0.77 g, respectively. It was reported by Halim et al. [52] that chemical extraction operating in dynamic mode, i.e., the Soxhlet extraction process, resulted in a roughly 280% increase in yield compared to static mode operation, i.e., the solvent extraction process. This significant improvement can be attributed to solvent refluxing, which exposed the cellular matrix to a fresh batch of hexane constantly and enabled continuous re-establishment of mass transfer equilibria [53]. Figure 3 shows the extracted product for the different lipid extraction methods. The percentage of the oil yield of three different methods for these two species of microalgae is depicted in Figure 4. As seen in the figure, for Spirulina sp., the rate of oil yield for solvent extraction, Soxhlet extraction, and ultrasonication is 2.5%, 4.1%, and 6.6%, respectively. On the other hand, for Chlorella sp., the percentage of oil yield is 2.3%, 3.5%, and 9.4%, respectively. Ultrasonication is the most effective among the three methods. The high efficacy of this method can be attributed to the disruption of the microalgae cells, as reported in other studies [44,54,55]. Sonic waves were passed to the microalgae culture while the ultrasound was done. They made a series of microbubble cavitations, which transferred kinetic energy to the cell surface, having broken up the cells [56]. Thus, ultrasonic waves help to reduce the size of dried algal fractions, which increases the surface area and the number of active sites available for reaction participation. Lee et al. [20] stated that the utilization of microwave and autoclaves are the best technique for disrupting Chlorella vulgaris cells, both of which produce a lipid yield of 10%. On the other hand, ultrasonication only resulted in a 5% lipid yield for a 5 min reaction time. In the present study, the 15-min sonication process resulted in a 9% lipid yield for Chlorella sp., showing that the longer reaction time was conducive to disrupting the cell walls of Chlorella sp., thereby increasing the lipid yield. However, it is to be noted that the effectiveness of cell disruption by ultrasonication will depend on the species of microalgae. For instance, Lee et al. [57] reported that Botryococcus species achieved a 16% lipid yield when ultrasonication was used. Thus, it is dependent on the microalgae cell wall thickness and diameter. Table 1 provides a study of oil production by the various methods of extraction. As shown in Figure 4, ultrasonication results in superior lipid extraction compared to other extraction processes. Structural analysis using SEM was performed to study the extraction methods in-depth. The analysis can be found in Appendix A. The percentage of the oil yield of three different methods for these two species of microalgae is depicted in Figure 4. As seen in the figure, for Spirulina sp., the rate of oil yield for solvent extraction, Soxhlet extraction, and ultrasonication is 2.5%, 4.1%, and 6.6%, respectively. On the other hand, for Chlorella sp., the percentage of oil yield is 2.3%, 3.5%, and 9.4%, respectively. Ultrasonication is the most effective among the three methods. The high efficacy of this method can be attributed to the disruption of the microalgae cells, as reported in other studies [44,54,55]. Sonic waves were passed to the microalgae culture while the ultrasound was done. They made a series of microbubble cavitations, which transferred kinetic energy to the cell surface, having broken up the cells [56]. Thus, ultrasonic waves help to reduce the size of dried algal fractions, which increases the surface area and the number of active sites available for reaction participation. Lee et al. [20] stated that the utilization of microwave and autoclaves are the best technique for disrupting Chlorella vulgaris cells, both of which produce a lipid yield of 10%. On the other hand, ultrasonication only resulted in a 5% lipid yield for a 5 min reaction time. In the present study, the 15-min sonication process resulted in a 9% lipid yield for Chlorella sp., showing that the longer reaction time was conducive to disrupting the cell walls of Chlorella sp., thereby increasing the lipid yield. However, it is to be noted that the effectiveness of cell disruption by ultrasonication will depend on the species of microalgae. For instance, Lee et al. [57] reported that Botryococcus species achieved a 16% lipid yield when ultrasonication was used. Thus, it is dependent on the microalgae cell wall thickness and diameter. Table 1 provides a study of oil production by the various methods of extraction. As shown in Figure 4, ultrasonication results in superior lipid extraction compared to other extraction processes. Structural analysis using SEM was performed to study the extraction methods in-depth. The analysis can be found in Appendix A. a 5 min reaction time. In the present study, the 15-min sonication process resulted in a 9% lipid yield for Chlorella sp., showing that the longer reaction time was conducive to disrupting the cell walls of Chlorella sp., thereby increasing the lipid yield. However, it is to be noted that the effectiveness of cell disruption by ultrasonication will depend on the species of microalgae. For instance, Lee et al. [57] reported that Botryococcus species achieved a 16% lipid yield when ultrasonication was used. Thus, it is dependent on the microalgae cell wall thickness and diameter. Table 1 provides a study of oil production by the various methods of extraction. As shown in Figure 4, ultrasonication results in superior lipid extraction compared to other extraction processes. Structural analysis using SEM was performed to study the extraction methods in-depth. The analysis can be found in Appendix A.

Lipid Extraction Maximization
As discussed earlier, ultrasonication was found to be more conducive to extracting the microalgae oil compared to other processes. Both ultrasonication and the Soxhlet extraction process were further carried out to maximize the oil yield. Ultrasonication extraction was investigated with varying amplitudes to maximize the oil yield. Different amplitudes were tested, ranging from 50% to 90%. The results of this work are shown in Figure 5. It is to be noted that, for these cases, a percentage increase in the oil yield compared to baseline was presented. The percentage yield of crude oil was observed to increase continuously as the amplitude increased. A higher amplitude enhances the disruption efficiency of the cell wall for both species of microalgae. For Chlorella sp., the percentage oil yield continues to increase up to the 90% amplitude, reaching about 36%. On the other hand, at the 70% amplitude, ultrasonication of Spirulina sp. resulted in a 30% oil yield. After that, any increase in amplitude did not affect the oil yield significantly and remained nearly identical. Thus, the 70% amplitude maximizes Chlorella sp. lipid extraction without significantly affecting output, whereas a 90% amplitude is required for Spirulina sp. lipid extraction.
Lipid maximization for Soxhlet extraction was investigated with a combination of solvents at different ratios to enhance the extraction process, and the results are shown in Figure 6. Since hexane resulted in a poor oil yield of both species of microalgae, different solvents were used to extract the microalgae cells. Instead of using a single solvent, a mixture of solvents was used based on the recommendations of previous research [17,58]. The goal of using a mixture of polar and non-polar solvent was to achieve maximum extraction efficacy compared to an earlier stage where only non-polar hexane was used. As seen in Figure 6, a C:H:M ratio of 1:1:1, 2:1:3, 3:2:1, and 1:2:3 resulted in about a 15.3%, 10%, 14%, and 13.3% percentage oil yield for Chlorella sp. On the other hand, a C:H:M ratio of 1:1:1, 2:1:3, 3:2:1, and 1:2:3 resulted in about a 23.3%, 12%, 15%, and 20% percentage oil yield for Spirulina sp. respectively. The higher percentage yield of oil was extracted from both species of microalgae compared to the hexane-only extraction. The combination of chloroform, hexane, and methanol at a ratio of 1:1:1 gave a maximum yield of crude oil for both species of microalgae. When comparing the yield of two species, Spirulina sp. showed a higher yield for every solvent combination ratio of chloroform, hexane, and methanol, as shown in Figure 6. Appendix B discusses the SEM images of dried microalgae after the Soxhlet extraction maximization process with 1:1:1 C:H:M.

Lipid Extraction Maximization
As discussed earlier, ultrasonication was found to be more conducive to extracting the microalgae oil compared to other processes. Both ultrasonication and the Soxhlet extraction process were further carried out to maximize the oil yield. Ultrasonication extraction was investigated with varying amplitudes to maximize the oil yield. Different amplitudes were tested, ranging from 50% to 90%. The results of this work are shown in Figure 5. It is to be noted that, for these cases, a percentage increase in the oil yield compared to baseline was presented. The percentage yield of crude oil was observed to increase continuously as the amplitude increased. A higher amplitude enhances the disruption efficiency of the cell wall for both species of microalgae. For Chlorella sp., the percentage oil yield continues to increase up to the 90% amplitude, reaching about 36%. On the other hand, at the 70% amplitude, ultrasonication of Spirulina sp. resulted in a 30% oil yield. After that, any increase in amplitude did not affect the oil yield significantly and remained nearly identical. Thus, the 70% amplitude maximizes Chlorella sp. lipid extraction without significantly affecting output, whereas a 90% amplitude is required for Spirulina sp. lipid extraction.  Lipid maximization for Soxhlet extraction was investigated with a combination of solvents at different ratios to enhance the extraction process, and the results are shown in Figure 6. Since hexane resulted in a poor oil yield of both species of microalgae, different solvents were used to extract the microalgae cells. Instead of using a single solvent, a mixture of solvents was used based on the recommendations of previous research [17,58]. The goal of using a mixture of polar and non-polar solvent was to achieve maximum extraction efficacy compared to an earlier stage where only nonpolar hexane was used. As seen in Figure 6 Table 2 presents the acid value of extracted crude oil using three different methods. As seen from the table, the oil from the ultrasonic extraction process achieved the lowest acid value. Thus, transesterification was carried out for the sample extracted by ultrasonication only. Methanol was added at a molar ratio of methanol to the oil of 3:1 [44] (an excess amount of the stoichiometric ratio) into a mixture of fresh microalgae oil along with immobilized lipase in a stepwise manner to complete the transesterification process. In this study, one-third of methanol was added at the initial stage,  Table 2 presents the acid value of extracted crude oil using three different methods. As seen from the table, the oil from the ultrasonic extraction process achieved the lowest acid value. Thus, transesterification was carried out for the sample extracted by ultrasonication only. Methanol was added at a molar ratio of methanol to the oil of 3:1 [44] (an excess amount of the stoichiometric ratio) into a mixture of fresh microalgae oil along with immobilized lipase in a stepwise manner to complete the transesterification process. In this study, one-third of methanol was added at the initial stage, one-third at 4 h, and the rest was added at 8 h. Methanol is known as a lipase inhibitor [59]; as such, high concentrations of methanol are unfavorable to lipase, catalyzing the effect and causing reduced enzyme activity and stability [60,61]. Thus, the stepwise addition of methanol is the standard choice since this not only avoids methanol's unfavorable effect but also expands the contact area of oil and methanol to achieve a high yield [50,62].

Analysis of FTIR Peak for FAME
In the present study, the mid-FTIR region was selected to identify the functional groups in the biodiesel sample. The mid-IR range between the 4000 and 400 cm −1 wavenumbers. A summary of the IR spectroscopy band assignments for both microalgae species after ultrasonically assisted lipase-catalyzed transesterification is presented in Table 3. The bands were assigned to specific molecular groups based on biochemical standards and previous studies [63,64]. There were three significant peaks identified for characteristic bonds in the FAMEs for Chlorella sp. and Spirulina sp., which were the functional groups for alkanes, esters, and alkyls. The patterns of band distribution and peaks obtained were found to be similar for both species with closely matching peaks. The region in the spectra from 3000 to 2800 cm −1 indicates the presence of lipids in the sample and is due to the symmetrical and asymmetrical stretching vibrations of -CH 2 -groups [65]. These -CH 2groups form the backbone of the lipids. Both Chlorella sp. and Spirulina sp. showed a transmittance peak at 2923 cm −1 , indicating the presence of the lipid backbone. An absorbance peak for Spirulina sp. was shown at a similar wavenumber by Nautiyal et al. [66]. The bands of interest to locate ester bonds in the compounds can be found in two regions: 1750-1735 cm −1 and 1320-1000 cm −1 . In particular, the peaks between 1750 and 1735 cm −1 were identified as C=O bonds of ester groups primarily from lipids and fatty acids, which indicates the conversion of oil to biodiesel [67]. Ester bond peaks were found at 1735, 1740, and 1741 cm −1 for these biodiesels, and show the presence of esters in the FAME analyzed. On the other hand, asymmetric deformation vibration of C-H [68] from proteins is usually found at~1455 cm −1 , which represents the presence of alkanes.

Analysis of Ester Content by GC
The successful detection of ester bonds in the sample by FTIR analysis led to GC analysis to determine the FAME compound in the sample in detail. Table 4 shows the composition of the FAME ester content of the obtained biodiesel for each sample at different temperatures. Different concentrations of unsaturated fatty acids (FA) were observed for each biodiesel sample at different operating temperatures. It is to be noted that palmitic (C 16:0) and stearic (C 18:0) acids were the major components for both Chlorella sp. and Spirulina sp. biodiesel. The saturated FA percentage of both samples varies around 60-75%, whereas the unsaturated FA percentage varies around 25-40% of the FAME content. Palmitic and stearic acid contents were observed to be increased with the increase in temperature. The highest percentage of palmitic acid for both microalgae biodiesels as well as the highest stearic acid for Chlorella sp. was obtained at 45 • C. However, for Spirulina sp., the highest content of stearic acid was found at 40 • C, and the percentage decreased at 45 • C. Nautiyal et al. [66] also studied the ester content of Spirulina sp. and pond water algae-based biodiesel. Palmitic, linolenic, and linoleic acids were the major FA components in those samples. Adam et al. [55] studied the FAC of ultrasound-assisted N. oculata microalgae-based biodiesel. They found only three fatty acids in the biodiesel sample viz. myristic, palmitic, and stearic acids. Similar to the previous study, palmitic acid was the dominant ester content in this biodiesel.

Conclusions
The study compared the efficacy of different lipid extraction methods on two microalgae species viz: Chlorella sp. and Spirulina sp., since different species behave differently towards an extraction method based on the size, diameter, and cell walls of the microalgae itself. Ultrasonication was found to be the best method to extract lipid from both species of microalgae, with percentage oil yields of 6.6% for Spirulina sp. and 9.4% for Chlorella sp. On the other hand, the oil yields for these species were 1.05 and 1.25 g for Soxhlet extraction and 0.68 and 0.77 g for solvent extraction, respectively, when 30 g of sample was used. At maximum yield conditions, there was a 30% increase in yield for Spirulina sp. and 36% for Chlorella sp., respectively, compared to baseline conditions. The transesterification of ultrasonic-assisted extraction samples was successful for both microalgae species, indicating their suitability as a renewable energy source. As shown in Figure A1, ultrasonication resulted in superior lipid extraction compared to other extraction processes. Structural analysis using SEM was performed here to study the extraction methods in depth. The structure of dried microalgae used in this study before extraction can be observed in Figure A1. For Chlorella sp., the cells have a spherical shape with an uneven surface of the phospholipid membrane, and appear to be clotted. From the EDX analysis, Chlorella sp. consisted of 40.2% carbon, 35.7% oxygen, and 23.4% nitrogen for the area analyzed. On the other hand, Spirulina sp. contained 43.7% carbon, 30.4% oxygen, and 25.2% nitrogen (25.2%) for the region analyzed.   Figures A2 and A3 show the structure of dried Chlorella sp. and Spirulina sp. after the Soxhlet extraction processes. The cells were shrunk compared to the original ones, some cells were broken, and the surfaces of those were distorted. Although Soxhlet extraction affected the structure of some microalgal cells, others were still intact, and the membrane cell did not show any damage. This explains the reason for the lower oil yield for Soxhlet extraction using hexane as the solvent. Hexane is the most popular and economical solvent commonly used in extraction studies. However, hexane was less effective in lipid extraction from Chlorella sp. compared to Spirulina sp., as seen in Figure A1. Comparing the 1000× magnification images from Figures A2 and A3, it is visible that Spirulina sp. cells were more affected by the process than those of Chlorella sp., thereby resulting in less yield. Since solvent extraction produced an even worse percentage, the oil extraction sample was not further analyzed. Consequently, a solvent mixture of multiple components was used further to maximize oil production using the Soxhlet extraction process.
Appl. Sci. 2020, 10, x FOR PEER REVIEW 13 of 19 Figures A2 and A3 show the structure of dried Chlorella sp. and Spirulina sp. after the Soxhlet extraction processes. The cells were shrunk compared to the original ones, some cells were broken, and the surfaces of those were distorted. Although Soxhlet extraction affected the structure of some microalgal cells, others were still intact, and the membrane cell did not show any damage. This explains the reason for the lower oil yield for Soxhlet extraction using hexane as the solvent. Hexane is the most popular and economical solvent commonly used in extraction studies. However, hexane was less effective in lipid extraction from Chlorella sp. compared to Spirulina sp., as seen in Figure A1. Comparing the 1000× magnification images from Figures A2 and A3, it is visible that Spirulina sp. cells were more affected by the process than those of Chlorella sp., thereby resulting in less yield. Since solvent extraction produced an even worse percentage, the oil extraction sample was not further analyzed. Consequently, a solvent mixture of multiple components was used further to maximize oil production using the Soxhlet extraction process. Figures A2 and A3 show the structure of dried Chlorella sp. and Spirulina sp. after the Soxhlet extraction processes. The cells were shrunk compared to the original ones, some cells were broken, and the surfaces of those were distorted. Although Soxhlet extraction affected the structure of some microalgal cells, others were still intact, and the membrane cell did not show any damage. This explains the reason for the lower oil yield for Soxhlet extraction using hexane as the solvent. Hexane is the most popular and economical solvent commonly used in extraction studies. However, hexane was less effective in lipid extraction from Chlorella sp. compared to Spirulina sp., as seen in Figure A1. Comparing the 1000× magnification images from Figures A2 and A3, it is visible that Spirulina sp. cells were more affected by the process than those of Chlorella sp., thereby resulting in less yield. Since solvent extraction produced an even worse percentage, the oil extraction sample was not further analyzed. Consequently, a solvent mixture of multiple components was used further to maximize oil production using the Soxhlet extraction process.

Appendix B. SEM Analysis of Microalgae of Lipid Extraction Maximization Study
Structures of dried microalgae after Soxhlet extraction were also observed under SEM to see the difference. Figures A4 and A5 comprise the structure of microalgae of Chlorella sp. and Spirulina sp. with 1000× magnification. The cell membranes were damaged after the extraction, broken, shrunk into a smaller size, and the membrane was distorted from the original surface. The Soxhlet extraction with mixtures of chloroform, hexane, and methanol did have a significant effect on the whole structure of microalgae. The highest yield of oil was obtained from the combination of CHM at a 1:1:1 ratio.

Appendix B. SEM Analysis of Microalgae of Lipid Extraction Maximization Study
Structures of dried microalgae after Soxhlet extraction were also observed under SEM to see the difference. Figures A4 and A5 comprise the structure of microalgae of Chlorella sp. and Spirulina sp. with 1000× magnification. The cell membranes were damaged after the extraction, broken, shrunk into a smaller size, and the membrane was distorted from the original surface. The Soxhlet extraction with mixtures of chloroform, hexane, and methanol did have a significant effect on the whole structure of microalgae. The highest yield of oil was obtained from the combination of CHM at a 1:1:1 ratio.