Biojet Fuel Production from Waste of Palm Oil Mill Effluent through Enzymatic Hydrolysis and Decarboxylation

Palm oil mill effluent (POME), wastewater discharged from the palm oil refinery industry, is classified as an environmental pollutant. In this work, a heterogeneous catalytic process for biojet fuel or green kerosene production was investigated. The enzymatic hydrolysis of POME was firstly performed in order to obtain hydrolysed POME (HPOME) rich in free fatty acid (FFA) content. The variations of the water content (30 to 50), temperature (30 to 60 °C) and agitation speed (150 to 250 rpm) were evaluated. The optimal condition for the POME hydrolysis reaction was obtained at a 50% v/v water content, 40 °C and 200 rpm. The highest FFA yield (Y FA) of 90% was obtained. Subsequently, FFA in HPOME was converted into hydrocarbon fuels via a hydrocracking reaction catalysed by Pd/Al2O3 at 400 °C, 10 bars H2 for 1 h under a high pressure autoclave reactor (HPAR). The refined-biofuel yield (94%) and the biojet selectivity (57.44%) were achieved. In this study, we are the first group to successfully demonstrate the POME waste valorisation towards renewable biojet fuel production based on biochemical and thermochemical routes. The process can be applied for the sustainable management of POME waste. It promises to be a high value-added product parallel to the alleviation of wastewater environmental issues.


Introduction
Biojet fuel, classified as green kerosene derived from biomass, is generally blended with petroleum-based kerosene. The range of the jet fuel carbon number is strictly controlled to be between 8 and 16 in order to obtain desirable fuel properties and meet the strict standard specifications of the American Society for Testing and Materials (ASTM) D1655. Since the primary function of biojet fuel is to power an aircraft, the energy content and combustion quality are significant key factors. Other important properties are stability, lubricity, fluidity, volatility, noncorrosivity and cleanliness [1]. Typically, there are four routes to produce green kerosene from bio-feedstocks: (i) Hydro-processing of fatty acids and natural triglycerides; (ii) Hydrothermal liquefaction of algal biomass and pyrolysis of lignocellulosic materials; (iii) Gas-to-jet fuel platform: the gasification reaction of biomass in order to obtain syngas, which is subsequently converted into hydrocarbon fuel via the Fischer-Tropsch reaction; and (iv) Alcohol-to-jet platform: bio-based alcohols are sequentially dehydrated, oligomerized and hydrogenated in order to generate biojet fuel. Currently, only the hydroprocessing of oil/fat is selected for renewable jet fuel production at the industrial scale [2]. The key factors to decide on suitable raw materials for this process are availability, potential yield and especially the price. It is quite cheap, classified as waste, and it also plays an important role in the total production cost. Thus, palm oil can be considered as feedstock for bio-aviation fuel production because it is the least expensive vegetable oil [3]. However, the major challenge of first-generation feedstock is food competition.
In palm oil milling plants, large quantities of palm oil mill effluent (POME) are produced during the milling process. According to its high pH values, biochemical oxygen demand (BOD) and chemical oxygen demand (COD), it was classified as an industrial waste. It was discharged as wastewater containing high BOD (>25,000 mg·L −1 ), which accounted for 100 times more pollution than municipal wastes [4,5]. Nowadays, the biological process named anaerobic digestion is the most common method for POME treatment when creating value-added products, including fertilizer and biogas. In addition, applications of biogas capture and membrane separation were reported in an integrated reactor for the production of biogas from POME, leading to a higher energy yield and lower greenhouse gas (GHG) emissions [6,7].
Considering new applications of POME, the production of hydrocarbon fuel from waste triglyceride is an interesting possible route for delivering clean fuel and as a response to environmental issues. The use of free fatty acid (FFA) as a feedstock for biojet fuel production has been reported as a cost-effective process due to a low hydrogen pressure, mild cracking reaction, short retention time and high yield, as compared to oil/fats [8][9][10][11][12]. In addition, the noble metals palladium (Pd) and platinum (Pt) were also proven to be high potential catalysts in decarboxylation reactions. They showed a high catalytic efficiency for biofuel production from triglycerides (TG) and free fatty acids (FFA) due to their high hydrogenation ability [13][14][15]. Accordingly, consecutive processes for hydrocarbon fuel have been developed, and triglycerides were hydrolysed to FFA under a high pressure and temperature, followed by the decarboxylation of FFA [16,17].
In this study, a new strategy to decrease an environmental pollutant of POME was presented. A combination of biochemical and thermochemical processes was firstly applied to generate renewable biojet fuel from POME. The enzymatic hydrolysis of POME was carried out in order to obtain FFA. After that, it was converted to hydrocarbon fuels via a hydrocracking reaction under mild conditions.

Influence of Temperature on Hydrolysis Reaction
The results showed that the hydrolysis degrees of POME increased considerably from 30 • C to 40 • C and presented a downward trend when the reaction temperature was higher than 40 • C (see Figure 1A). This can be explained by the fact that the viscosity of the substrate can be reduced by a higher temperature, resulting in a high catalytic rate of lipase. As a result, a higher FFA yield was observed in the mixture reaction. However, an excessively high temperature caused the enzyme structure to be denatured and to eventually deactivate. Thus, by increasing the reaction temperature to more than 40 • C, the hydrolysis degree of the three raw materials decreased slightly, as shown in Figure 1. Meanwhile, the highest hydrolysis degree of POME (89.51 ± 0.04%) was obtained at 40 • C. Therefore, the reaction temperature of 40 • C was selected to investigate the effect of the water content on the hydrolysis reaction in the following experiment. Furthermore, the results obtained were in agreement with the previous work [18], which reported an enzymatic POME hydrolysis. They found that the maximum hydrolytic rate of lipase was achieved at 40 • C.

Influence of Water Content on Hydrolysis Reaction
The results illustrated that the hydrolysis degree of POME kept continually rising as the water content increased from 30 to 50% v/v (see Figure 1B). This can be explained by the fact that the enzymatic hydrolysis reaction is a reversible reaction. It is controlled by the amount of water in the mixture solution. Normally, excess water affects the hydrol-ysis reaction in a positive way; however, a high water content (60% v/v) resulted in the hydrolysis degree decreasing dramatically [19]. When the water in the mixture solution was 50% v/v, the maximum FFA yield of 89.48 ± 0.05% was obtained. Therefore, the water content was fixed at 50% v/v for the following experiments. The results obtained were in agreement with the previous work [20], which reported that as the molar ratio of oil to water increased from 1:1 to 1:4, the fatty acid yield also increased continuously. However, the fatty acid yield decreased as the ratio of oil to water increased up to 1:5.

Influence of Agitation Rate on Hydrolysis Reaction
The results observed showed that the agitation rate affected the hydrolysis reaction in a positive way, as it was clearly seen that the hydrolysis degree of POME significantly increased between the agitation rates of 100 rpm and 200 rpm (see Figure 1C). This might due to a higher agitation rate creating more oil-water interface, where the catalysis of the hydrolysis reaction via lipase occurred; this led to a high FFA yield. However, increasing the agitation rate up to 250 rpm caused a lower hydrolysis degree of POME. The agitation can reduce the droplet size and increase the interface area between oil and water, leading to a higher activity of lipase. However, the enzyme activity can be gradually decreased under a high agitation rate due to the shear stress [21]. The highest FFA yield (90 ± 0.04%) was observed under an agitation rate of 200 rpm. Therefore, this rate was chosen as the optimal rate for the enzymatic hydrolysis of POME.

Characterisation of Refined Biofuels
The characteristic of refined biofuels in terms of the carbon number was analysed by Gas chromatography-Mass spectroscopy (GC-MS). According to the result in Figure 3, the carbon atom distribution of refined biofuels was mainly in the ranges of kerosene (C9-C15) and diesel (C15-C18). This was due to three major fatty acid compositions of HPOME: palmitic acid (C16:0), stearic acid (C18:0) and oleic acid (C18:1). One carbon atom of the carboxyl group (-COOH) was eliminated during decarboxylation [22]. It should be noted that the refined biofuels were produced and measured in triplicate for reproducibility.

Comparative Green Kerosene Yields Obtained from Hydrocracking Reaction of Different Waste Oils
Based on the results in Table 2, enzymatic hydrolysis coupled with the hydrocracking reaction of POME could serve the highest green kerosene yield when compared to the single hydrocracking reaction of other waste oils. The production of green kerosene from waste cooking oil [23] was operated at the lowest temperature (380 • C), and a high green kerosene yield of 40.50% was also observed. However, this took a longer reaction time than in other works. The results clearly indicated that HPOME could be efficiently transformed into hydrocarbon fuel by using only a 1% catalyst loading, low H 2 pressure (10 bar), moderate temperature (400 • C) and short reaction time (1 h). This was due to the decarboxylation of fatty acids having a higher efficiency than triglycerides [17]. Therefore, the heterogeneous catalytic process could be considered as a high potential process for renewable jet fuel production in the future.

Materials
Samples of low-cost raw material of palm oil mill effluent (POME) were collected from the waste water lagoon of E-san palm oil industry, Sakonakhorn, Thailand. A chemical catalyst, Pd/Al 2 O 3 , was purchased from Sigma (Missouri, MO, USA). An immobilised lipase of Rhizopus oryzae on sodium alginate (SAL beads) was prepared by a simple entrapment technique in a CaCl 2 solution. First, sodium alginate (2 g) was added into 100 mL of phosphate buffer (pH 7.0) solution. Then, it was heated at 40 • C until a homogeneous solution was obtained. After that, it was cooled down to room temperature. Subsequently, lipase (2 g) was mixed with alginate solution (100 mL). The mixture solution was dropped into 0.1 M CaCl 2 by a sterile syringe and stored at 4 • C for 12 h to form an alginate bead (Diameter size of 3 mm.). Finally, the immobilised bead was washed three times with distilled water and was kept in distilled water at 4 • C before being used as a biocatalyst in the hydrolysis reaction. The hydrolytic activity of the SAL beads was 360 U/mg. This was determined by the spectrophotometric method using p-nitrophenyl laurate (p-NPL) as the substrate [25].

Pretreatment of POME via Enzymatic Hydrolysis
The contaminants contained in POME, such as trunks barks and leaves, were separated by gravity method using a separating funnel [20]. Consequently, SAL beads (2 g), POME (100 mL) and distilled water (50% v/v) were loaded into a 250-mL Erlenmeyer flask. The hydrolysis reaction was conducted in an incubator shaker at 40 • C, 200 rpm for 4 h. The samples were further analysed to determine the FFA content, as described by a previous study [26]. The fatty acid yield (Y FA ) was calculated as shown in Equation (1). The fatty profile of hydrolysed POME (HPOME) was analysed by gas chromatography-spectrometry (GC-MS) following the standard method EN 1403. The HPOME sample (50 mg) and a 0.5 M methanolic sodium hydroxide solution (5 mL) were added into a reflux flask connected to the condenser. The reaction mixture in the reflux flask was heated to 140-160 • C for 5 min for a saponification reaction. Subsequently, a boron trifluoride-methanol solution (5 mL) was added and heated to 140-160 • C for 5 min for a methylation reaction. The mixture solution was cooled down to room temperature, followed by adding hexane (5 mL) and saturated sodium chloride solution (10 mL) in order to obtain fatty acid methyl ester (FAME). The bi-layer of the mixture reaction was observed. The FAME in the hexanic phase (upper layer) was dried by adding anhydrous sodium sulfate (1 g) and was filtrated through Whatman paper No. 1. Last, the samples were dissolved in heptane and filtrated through a microfilter (0.45 µm) before being injected into GC-MS equipped with a FID detector in order to identify the fatty acid composition, as described in previous studies [27]. The operating condition for the fatty acids' profile analysis was explained by Muanruksa et al. [28]. First, a 250-mg sample was filled into a 10-mL vial, followed by the addition of 5 mL of internal standard (methyl haptadecanoate solution, 10 mg/mL). Subsequently, the sample was analysed by gas chromatography-mass spectroscopy (GC-MS) (GC-2010, Shimadzu, Tokyo Japan) equipped with a 30-m long and 0.25-mm diameter capillary column lined with a 0.25-µm(Rtx-5 ms, Rextex). Samples were injected in a split/column flow ratio of 24:1. Helium was used as the carrier gas at a flow rate of 1 mL/min. The injection temperature was 250 • C, and the column oven's temperature was 250 • C (programmed to start at 120 • C, held at this temperature for 5 min and heated at a rate of 3 • C/min to 250 • C).
when FA1 is the FFA content of the oil sample before the hydrolysis reaction, and FA2 is the FFA content of the oil sample after the hydrolysis reaction.

Decarboxylation of Fatty Acid in HPOME
The hydrocarbon fuel production from hydrolysed POME (HPOME) was carried out in a high pressure batch reactor (HPAR) (See Figure 4). Raw materials of HPOME (100 mL) and Pd/Al 2 O 3 (1 g) were loaded in a chamber of the reactor. Then, hydrogen was flushed three times to remove the oxygen in the reaction. Subsequently, it was fed to the reactor until a hydrogen pressure of 10 bars was reached at the initial stage, and the reaction temperature was set at 400 • C and kept constant for 1 h. Finally, the crude biofuel was refined by using a fractional distillation technique (ASTM D86).

Fractional Distillation
A commercially manual distillation apparatus (Koehler model K45200, New York, NY, USA) was used to conduct fractional distillation in this research work. The apparatus components met the standard of the ASTM D86 specifications, including a 125-mL sidearm distillation flask (borosilicate glass), a 100-mL graduated cylinder with 1.0-mL graduation intervals and a mercury thermometer. The graduate cylinder was centered under the condenser tube exit to allow condensate drops to fall to the bottom of the cylinder. The HPOME (100 mL) was taken to be used as a liquid sample in the fractional distillation at 1 atm. The refined-biofuel products could be classified into three types based on their boiling points, including gasoline (50-150 • C), kerosene (150-280 • C) and diesel (280-360 • C), respectively. The percentages of crude-biofuel yield, refined-biofuel yield, green diesel selectivity, green kerosene selectivity and green gasoline selectivity were determined as described in Equations (2)- (6).
The liquid products were analysed by gas chromatography-mass spectroscopy (GC-MS) using a flame ionization detector (GC-FID-QP2010 Shimadzu, Tokyo, Japan) equipped with HT5 capillary columns (length of 15 m, diameter of 0.25 mm, film of 0.1 µm), as described by a previous study [29]. The liquid samples (40 mg) were diluted via 4 mL hexane (high performance liquid chromatography (HPLC) grade). Then, 0.5 µL of diluted sample was injected into the column. Helium was used as the carrier gas, and the flow rate was set at 1.5 mL/min. The oven temperature was first operated at 50 • C, held for 1 min, and then increased to 220 • C at the rate of 9 • C/min, followed by an increase of 20 • C/min to 350 • C, before being increased to 380 • C at the rate of 10 • C/min and held for 5 min. The n-alkanes and n-alkenes (C8-C20) were used as the chemical standard for the calibration curves to identify the composition of the liquid product. The internal standard was 1-bromohexane [27].

Conclusions
A combination of biochemical and thermochemical processes for the production of renewable biojet fuel from POME has successfully been performed in this study. The process provided a high efficiency in terms of a high green kerosene yield and a high selectivity of green kerosene. Additionally, the hydrocracking reaction could be carried out entirely using a low amount of catalyst, low H 2 pressure, moderate temperature and short reaction time. It was apparently indicated that the pretreatment of POME via the hydrolysis reaction was catalysed by immobilised lipase, resulting in positive effects on the hydrocracking reaction. Therefore, the POME showed a high potential use as a secondgeneration feedstock for producing biojet fuel. The heterogeneous catalytic process can be classified as a possible pathway for figuring out environmental issues related to POME, as well as for the development of a sustainable renewable energy production.
Author Contributions: This article was performed as collaborative research between P.M., J.W. and P.K.; the work included: conceptualization, methodology, resources, formal analysis, data curation, visualisation and writing-original draft preparation, P.M. and J.W. considered the software, validation and investigation, while P.K. performed the supervision, project administration, funding acquisition, review and corrections. All authors have read and agreed to the published version of the manuscript.