Hydrocarbon Production from Catalytic Pyrolysis-GC / MS of Sacha Inchi Residues Using SBA-15 Derived from Coal Fly Ash

: In this work, Sacha inchi ( Plukenetia volubilis L.) residues were used as biomass feedstocks in catalytic upgrading pyrolysis with SBA-15, which is a substance synthesized from coal ﬂy ash (CFA), using alkali fusion, followed by hydrothermal treatment (SBA-15-FA). The catalytic activity of ﬂy ash-derived SBA-15 was investigated through the fast pyrolysis of Sacha inchi residues for upgrading the pyrolysis vapors using the analytical pyrolysis-GC / MS (Py-GC / MS) technique. The pyrolysis temperature was set at 500 ◦ C and held for 30 s while maintaining the Sacha inchi residues to catalyst ratios of 1:0, 1:1, 1:5, and 1:10. In addition, the SBA-15s synthesized from chemical reagent and commercial SBA-15 were evaluated for comparison. The non-catalytic fast pyrolysis of Sacha inchi (SI) mainly consisted of fatty acids (46%), including chieﬂy linoleic acid (C18:2). Other compounds present were hydrocarbon (26%) and nitrogen-containing compounds (8.7%), esters (9.0%), alcohols (6.4%), and furans (3.6%). The study results suggested that the SBA-15-FA showcased a high ability to improve aliphatic selectivity (mainly C 5 –C 20 ) and was found to be almost 80% at the biomass to catalyst ratio of 1:5. Moreover, the increase in catalyst contents a ﬀ ected the enhancement of hydrocarbons yields and tended to promote the deoxygenation reaction. Interestingly, the catalytic performance of SBA-15 derived from ﬂy ash could be compared to that of the commercial SBA-15 in terms of producing hydrocarbon compounds as well as reducing oxygenated compounds.


Introduction
The valorization of biomass residues as biofuels is an interesting phenomenon and has been gaining much attention due to the depletion of fossil fuel reserves, its similarity with traditional fossil fuels in terms of chemical composition, and low pollutant emissions. The potential biofuel resources can be classified into three groups: first-generation feedstocks (i.e., edible vegetable oils and fats), second-generation feedstocks (i.e., non-edible vegetable oils, animal fats, waste cooking oils, agricultural residues, and wastes from food crops), and third-generation feedstocks (i.e., microalgae, bacterial, and terpenes). Direct biofuel production from first-generation feedstocks is certainly uncompetitive; on the other hand, the usage of third-generation feedstocks is limited due to commercialization. Thus, in Table 1. Based on the previous studies and to our knowledge, there are no existing reports that involve the catalytic fast pyrolysis of biomass by using fly ash-derived SBA-15 catalysts. Sciubidło, A., Majchrzak-Kucęba [19] Coal fly ash Alkali fusion and hydrothermal treatment Na-X, SBA-15, MCM-41 NO 2 capture Therefore, in this work, an SBA-15 mesoporous catalyst was synthesized using an alkali fusion, followed by the hydrothermal method. After the synthesis process, the catalytic performance of fly ash-derived SBA-15 for deoxygenating pyrolysis vapors from the pyrolysis of SI residues was investigated using pyrolysis-GC/MS (Py-GC/MS). The effect of the SI residues to SBA-15 ratio (SI/SBA-15 ratios of 1:1, 1:5, and 1:10) on the pyrolytic products was studied. Additionally, SBA-15-synthesized chemicals and commercial SBA-15 catalysts were also investigated for the purpose of comparison.

Characterization of SBA-15 Derived from CFA
The small-angle XRD pattern of SBA-15 derived from CFA (SBA-15-FA) demonstrated three characteristic peaks at 2θ = 0.89 • , 1.53 • , and 1.78 • (Figure 1), which could be indexed to the 100, 110 and 200 plans of the highly ordered 2D-hexagonal mesoporous structure (p6mm) [10,11,13]. The SBA-15 prepared from chemical reagents (SBA-15 chemical) and commercial SBA-15 (SBA-15 commercial) exhibited XRD patterns similar to that in SBA-15-FA. This indicates that the supernatant extracted from fly ash can be used as a solution gel to synthesize well-ordered mesoporous materials. The N 2 adsorption-desorption isotherms and pore size distribution curves of SBA-15-FA, SBA-15-chemical, and SBA-15-commercial catalysts have been displayed in Figure 2. SBA-15-FA exhibited type IV adsorption isotherm with type H1 hysteresis loop, which indicates that the mesoporous structure and cylindrical pore channels of SBA-15 can be obtained by using fly ash as a raw material for SBA-15 synthesis. The N 2 adsorption isotherms of SBA-15-FA increased sharply over the P/P 0 range of 0.6-0.8, resulting from the capillary condensation with uniform mesopores [11,16]. For other catalysts, they illustrated the similarity in the type IV adsorption isotherm as that of SBA-15 derived from fly ash. Additionally, all SBA-15 catalysts exhibited quite narrow and monomodal pore size distributions, confirming uniform pore size with concentration at around 5-8 nm (Figure 2b). It was seen that SBA-15-FA revealed that the largest pore size was 7.4 nm. The textural properties of SBA-15 catalysts, including specific surface area, total pore volume, pore size, and pore wall thickness, have been showcased in Table 2. Interestingly, the specific surface area and pore volume at 705 m 2 /g and 1.15 cm 3 /g, respectively, were found to be greater than the SBA-15 synthesized from chemical reagents. The value obtained corresponded well to the ones presented by the findings of previous works. Li et al. [17] prepared SBA-15 from fly ash using an alkali two-step hydrothermal method with various amounts of NaOH and reported that the high surface area of 694 m 2 /g with a pore volume of 0.76 cm 3 /g was generated after using NaOH of 6.4 g in the dissolution step of fly ash. Meanwhile, Kim et al. investigated the synthesis of SBA-15 from mine tailing, which contained more than 80% of SiO 2 . After the alkali fusion and the hydrothermal treatment, the mesoporous silica SBA-15 with a surface area of 513 m 2 /g was successfully synthesized [12]. Recently, Yang et al. [13] successfully hydrothermally synthesized Al-SBA-15 from the spent fluid catalytic cracking catalyst and reported that the highly ordered Al-SBA-15 displayed a high and specific surface area of 742 m 2 /g and a large pore volume of 0.80 cm 3 /g. The pore size of SBA-15 catalysts obtained in this study was in the order of SBA-15-FA (7.4 nm) > SBA-15-chemical (6.3 nm) > SBA-15-commercial (5.8 nm), respectively. From this result, it was observed that the SBA-15-FA catalyst had the Si/Al ratio of 144.9, as shown in Table 2. The Al incorporation led to the increase in the pore size of SBA-15 due to the expansion of the aluminosilicate framework. Liang et al. suggested that the d spacing of the (100) plane (d100) changed to the higher values significantly with the initial introduction of aluminum into the SBA-15 structure, which corresponds to the longer Al-O bond compared with the Si-O bond [20]. Based on these results, it can be concluded that it is possible to use the supernatant extracted from the fused fly ash as silicate and aluminate precursors for the synthesis of mesoporous materials with high surface areas. The SEM micrographs of all the SBA-15 catalysts have been presented in Figure 3. SBA-15-FA explicitly revealed that the particles had a rod-like morphology of SBA-15 with a relatively uniform size of less than 3 microns (as shown in Figure 3a Figure 4. The pore diameter of all the samples was less than 10 nm, while the thickness of mesoporous silica walls was estimated to be 5-8 nm, which was in agreement with the N 2 -adsorption measurements and the XRD results.    Based on the ammonia temperature-programmed desorption (NH 3 -TPD), the acid site of the acid catalysts could be categorized as weak (100-250 • C), moderate (250-400 • C), or strong (>400 • C) [21]. All the SBA-15 catalysts can be observed as weak, moderate, and strong acid sites in different proportions. The SBA-15-FA catalyst had higher proportions of moderate and strong acidity as a result of the incorporation of Al ions loaded in the SBA-15 structure. Additionally, the Fourier transform infrared (FTIR) spectrums of synthesized SBA-15 catalysts with and without pyridine adsorption are given in Figure 6. Pyridine adsorption using the FTIR technique was applied to confirm the incorporation of aluminum into the SBA-15-FA structure. This technique can distinguish between the Brϕnsted and Lewis acid sites. Among SBA-15 catalysts, only the SBA-15-FA sample demonstrated the acidity characteristic peaks in the region of 1400-1700 cm −1 (Figure 6c). The peak that appeared at 1447 and 1545 cm −1 corresponded to the Lewis and Brϕnsted acid sites, respectively, whereas the peaks at 1490 cm −1 are associated with both the Lewis and Brϕnsted acid sites. [5,20,22,23]. The number of acid sites of SBA-15-FA catalyst after pyridine adsorption is shown in Table S1. In general, the main compositions of fly ash consist of silica (SiO 2 ) and alumina (Al 2 O 3 ) [16,24]. Thus, the Al ions from the fly ash should be embedded in the framework of SBA-15 derived from fly ash after the synthesis process. On the other hand, SBA-15-chem and commercial (SBA-15-com) exhibited a lower proportion of acidity.

Non-Catalytic Pyrolysis
The non-catalytic pyrolytic products of the SI residues pyrolyzed at 400-600 • C could be classified into nine organic compounds, including aliphatic hydrocarbons, aromatic hydrocarbons, carboxylic acids, ketones, aldehydes, alcohols, esters, ethers, and N-containing compounds, as shown in Figures 7 and 8. The major compounds derived from the SI residues that appeared at a retention time (RT) between 2 and 10 min under a thermal condition were light aliphatic and aromatic hydrocarbons, while the heavier compounds, including the oxygenated compounds (carboxylic acids, alcohols, ketones, esters, etc.), heavy aliphatic hydrocarbons, and N-compounds were found in the middle (at RT 10-30 min) and at the end of the chromatogram (at RT 30-50 min). The carboxylic acid with the highest peak at a retention time between 37 and 38 min can be represented as the main compound, which was mostly linoleic acid (C18:2), and it was generally found in the oil fraction of sacha inchi (32-45%) [25,26]. The main unsaturated fatty acid (C18:2) was chiefly derived as a result of the decomposition of triglycerides by β-elimination reaction [27]. When the temperature increased, acid compounds increased from 38.5% at 400 • C to 42.6% at 500 • C, and then, they significantly decreased to 29.6% at 600 • C. The unsaturated fatty acid decreased at high temperatures as a result of deoxygenation (hydrodeoxygenation, decarboxylation, decarbonylation) and breakdown of the C-C bond though β-scission reactions, producing light hydrocarbons, CO, CO 2 , and H 2 O. Consequently, hydrocarbons, particularly cyclic compounds, can be formed by means of two reaction routes, namely, the Diels-Alder reaction and intermolecular radical cyclization, in order to generate C 5 -C 6 cycloalkanes and cycloalkenes, as well as aromatic hydrocarbon compounds [27][28][29]. As a result, a high content of aliphatic compounds can be observed at a high pyrolysis temperature up to 43%. It is well-known that hydrocarbon compounds can improve the heating value of bio-oil.  A low concentration of ketones and aldehydes were detected (<2%), while the fractions of esters, ethers, and alcohols were found to be small in quantity at 3.0-9.7%, 1.8-10.5%, and 7.9-9.2%, respectively. These oxygenated compounds lead to the poor quality of bio-oil in terms of lowering heating values and decreasing its stability. Furthermore, the high content of N-containing compounds, which mainly consist of amides, nitriles, and indoles, were found to be remarkable at 11.4-25.2% depending on the pyrolysis temperature. The high level of N-containing compounds may cause environmental issues as a result of NOx emissions during the combustion process [8,24]. From the non-catalytic results, the large fractions of undesirable compounds were significantly obtained at 45.8-68.6% for the oxygenated compounds and at 11.4-25.2% for N-compounds.
The pyrolysis temperature played an important role in the non-catalytic compositions. Although the highest selectivity of aliphatic hydrocarbons could be obtained at 600 • C, the maximum liquid yield was not achieved under that pyrolysis temperature. According to previous studies, the highest yield of bio-oil was achieved at a moderate pyrolysis temperature (∼500 • C). Rising pyrolysis temperature (600-700 • C) further promoted the secondary cracking reactions of organic vapors, leading to an increase in gas proportion [30,31]. Thus, a temperature of 500 • C was selected as the pyrolysis temperature for evaluating the catalytic performance of the fly ash-derived SBA-15.

Effect of Biomass: Catalyst Weight Ratio
The catalytic evaluation of the fly ash-derived SBA-15 for deoxygenating pyrolysis vapors from the pyrolysis of the SI residues was investigated using Py-GC/MS. Moreover, the effect of the SI residues to SBA-15 ratio (SI/SBA-15 ratios of 1:1, 1:5, and 1:10) on pyrolytic products was studied. Furthermore, SBA-15 synthesized from chemicals and commercial SBA-15 catalysts were also investigated for comparison.
The catalytic cracking mechanism of oil plants can be explained by Sinh et al. and Vichaphund et al. [24,32]. Initially, the thermal cracking of heavier oxygenated species such as long-chain fatty acids, which are the main components of vegetable oils, can take place both on the external surface and in the internal pore of zeolites to obtain smaller oxygenated compounds and hydrocarbons. Then, the oxygenated species undergo deoxygenation reactions (decarboxylation, decarbonylation, and dehydration) for the removal of oxygen in the form of CO 2 , CO, and H 2 O, consequently resulting in the formation of hydrocarbons. These active hydrocarbons can be broken down into lighter hydrocarbon chains, including olefins and paraffin. In addition, this hydrocarbon with a lower molecular weight can be activated by the active site of the zeolite structure to undergo further isomerization, cyclization, and aromatization to form cyclic hydrocarbons and aromatic hydrocarbons, which is within the gasoline and diesel ranges.
In this catalytic run, the incorporation of mesoporous SBA-15 catalysts, which had been synthesized from fly ash, chemical reagent, and commercial SBA-15, enhanced the degree of deoxygenation reactions, and consequently, the quality of bio-oil improved. As shown in Figure 9, the presence of SBA-15 catalysts remarkably decreased the level of fatty acid compounds, while the fraction of aliphatic hydrocarbons significantly increased. At low catalyst content and a SI/SBA-15 ratio of 1:1 (Figure 10), SBA-15 catalysts remarkably decreased the level of carboxylic acid compounds through decarboxylation, decarbonylation, and cracking reactions due to the acidity of SBA-15 catalysts [2,33]. As a result, the formation of aliphatic hydrocarbons largely increased from 25.8% (no catalyst) to 68.9% for SBA-15-FA, 53.7% for SBA-15-chemical, and 42.2% for SBA-15-commercial. This result corresponded with the findings of a previous report that studied the catalytic conversion of waste cooking oil (WCO) (mainly palmitic (C 16 ) and linoleic fatty acid (C 18 )) for biofuel production using MeO-SBA-15. The proportion of fatty acid in WCO decreased when MeO-SBA-15 catalysts were used, as these mesoporous catalysts promote the formation of hydrocarbon through decarbonylation, cracking, and decarboxylation [2]. The ability of SBA-15 catalyst to produce aliphatic hydrocarbons was reported by Ozbay et al., who investigated the catalytic pyrolysis of banana peel waste using Al-modified SBA-15 and found that the compositions of the bio-oil catalyzed using Al-SBA-15 materials (15 wt %) contained a high aliphatic fraction up to 55.1%, while a lower fraction of aromatics at 25.1% was achieved [34]. This may be due to the larger pore characteristics of SBA-15, which facilitate the production of linear/cyclic hydrocarbons. However, SBA-15 catalysts did not reduce the proportion of other oxygenated compounds, including ketones, aldehydes, esters, and alcohols at this ratio. However, N-compounds were still found to be in the range of 12.3-14.5%. Compared to the SBA-15 synthesized from chemicals and commercials, SBA-15-FA showed better results in terms of encouraging the deoxygenation of oxygenated compounds and enhancing the production of aliphatic hydrocarbons.
When the catalyst was increased by five times (SI:SBA-15 ratio of 1:5), a higher amount of SBA-15 catalysts significantly improved the pyrolytic vapors and their selectivity compared to a lower amount of catalyst (SI:SBA-15 ratio of 1:1). The acid compounds were eliminated during the decarboxylation reaction, whereas the other oxygenated compounds slightly decreased. SBA-15-FA was more effective in reducing the oxygenated compounds (6.9%) with no significant change in the N-compounds (13.2%). As a result, SBA-15-FA promoted a higher formation of aliphatic hydrocarbons compared to the others. The production of more aliphatic compounds over SBA-15-FA, compared to those over SBA-15-chemical and SBA-15-commercial, suggested that the presence of Al embedded in the SBA-15-FA framework contributed to the increase in acidity, thus promoting cracking reactions, deoxygenation (e.g., decarboxylation, decarbonylation), and isomerization [31]. Furthermore, the effect of well-ordered mesoporous silica with a large pore size is particularly crucial, as their large pore size can cause the bulky molecules to react at the active site inside the zeolite framework, leading to greater hydrocarbon yields. Kamaruzaman et al. found that the lowest hydrocarbon yield was obtained when Co/SBA-15 was used, which had the smallest pore size (5.40 nm) [35]. A similar trend was found in this work; SBA-15-commercial with the smallest pore size (5.81 nm) produced the lowest amount of hydrocarbons, compared to SBA-15-FA and SBA-15-chemical, with pore sizes of 7.42 nm and 6.27 nm, respectively.
Thus, the aliphatic selectivity was in the following order: SBA-15-FA (79.9%) > SBA-15-chemical (69.5%) > SBA-15-commercial (57.3%). However, the mesoporous SBA-15 catalysts showed a relatively low aromatic selectivity at 0-4.0%. This is probably due to their mesopore characteristics and low amount of acidity limiting the aromatic formation. The mesopore of SBA-15 catalysts played a dominating role in the mass transfer and diffusion of large molecules into pores along with a variety of reactions including cracking, deoxygenation, isomerization, and cyclization for aliphatic hydrocarbon production. In terms of aromatization reaction, Rahman et al. suggested that the production of aromatic hydrocarbons was closely related to the accessible Brϕnsted acid sites within the framework of the zeolite catalyst [35]. Unfortunately, from our studies, the pyridine-FTIR showed that there was a small peak associated with Brϕnsted acid sites of SBA-15-FA, which had the Si/Al of 144.9. Thus, the high Si/Al ratio of this catalyst decreased the aromatic selectivity. This was in agreement with Jeon et al. [9], who used the mesoporous SBA-15-based materials (SBA-15, Al/SBA-15, (Pd, Pt)/SBA-15) in the catalytic pyrolysis of commercial cellulose and hemicellulose (xylan) and reported that a low aromatic yield (less than 5% area) was obtained when SBA-15 and Al/SBA-15 catalysts were used. The acidity limitation of Al/SBA-15 catalyst was reported by Qiang et al., who studied the catalytic fast pyrolysis of sawdust using SBA-15 and Al/SBA-15 with different Si/Al ratios (10-70). They suggested that the production of aromatic hydrocarbon including mainly toluene was detected in a small content (≤0.4%) when using Al/SBA-15 with the Si/Al of 10-70. The lowest yield was approximately 0.3% when using Al/SBA-15 with the highest Si/Al ratio of 70, which reduced the concentration of acid sites due to the lowest Al atoms in the structure and consequently, it decreased the catalytic activity [11].
At the highest catalyst loading (SI:SBA-15 ratio of 1:10), the degree of aliphatic conversion tended to significantly decrease for the highest catalyst content, which might be attributed to catalyst deactivation. Coke formation occurred due to the re-polymerization of deoxygenated intermediates and the condensation of hydrocarbons [35]. The relative amount (wt %) of coke deposited on spent SBA-15 catalysts after pyrolysis is shown in Table 3. It can be seen that the amount of catalyst had an effect on the coke formation. Increasing the catalyst content resulted in an increase in coke formation on the surface of SBA-15 catalysts. Using SBA-15 derived from fly ash (SBA-15-FA) exhibited relatively higher coke deposited when compared with other catalysts at the SI:SBA-15 ratio of 1:10. The coke formation of synthesized SBA-15 catalysts corresponded to the decrease in the aliphatic hydrocarbon, as mentioned earlier. Consequently, undesirable products were observed in a high amount at 9.2-30.9% for the oxygenated compounds and 9.2-19.1% for the N-compounds. From these results, it can be concluded that SBA-15 derived from fly ash (SBA-15-FA) showed a high efficiency that improves the hydrocarbon selectivity up to 80% at a SI:SBA-15 ratio of 1:5 with the ability to decrease the oxygenated compounds to 6.9%. Interestingly, SBA-15 derived from fly ash exhibited superior catalytic performance in terms of producing hydrocarbon compounds (80%) compared to the commercial SBA-15 (61%). Therefore, SBA-15 derived from fly ash (SBA-15-FA) can be used as an acid catalyst for a catalytic biomass pyrolysis application.

Hydrocarbon Distribution and Selectivity
During the catalytic process, the aliphatic hydrocarbons were the main compositions, which consisted of linear hydrocarbons (e.g., alkanes, alkenes, alkynes) and cyclic hydrocarbons (cycloalkanes and cycloalkenes), with a small number of aromatic hydrocarbons (monocyclic aromatic HC: MAHs). The main compounds obtained due to non-catalytic pyrolysis were alkanes (4.9%), alkenes (7.3%), cycloalkanes (8.4%), and cycloalkenes (4.3%), as shown in Figure 11. The formation of cycloalkanes (mainly cyclohexanes (C 6 )) and cycloalkenes (e.g., cyclopentenes (C 5 ), cyclohexenes (C 6 ), and cycloheptenes (C 7 )) can be formed by means of two reaction pathways, including the Diels-Alder reaction and intermolecular cyclization. The incorporation of SBA-15 catalysts significantly increased the selectivity of aliphatic hydrocarbon at SI:SBA-15 ratios of 1:1 and 1:5. In the case of linear hydrocarbon, the formation of alkenes was preferred over that of alkanes and alkynes, confirming that the decarbonylation was the dominant deoxygenation pathway [28,36]. Meanwhile, cycloalkanes and cycloalkenes were found to be the minor constituents. Moreover, it could be seen that there was a production of small amounts of monocyclic aromatic hydrocarbons such as toluene and ethylbenzenes (0.6-4.0%) when SBA-15 catalysts were used. Aromatic compounds such as monocyclic aromatic hydrocarbons can be formed through the dehydrogenation of cyclohexenes [28].
Additionally, these hydrocarbons can be classified into three groups according to the carbon lengths of ranges C 5 -C 12 , C 13 -C 20 , and C >20 , as shown in Figure 12. From this result, the predominant carbon lengths produced from SBA-15 catalysts were C 5 -C 12 and C 13 -C 20 , which were within the ranges of gasoline and diesel, respectively. SBA-15-FA at a high catalyst content (at a SI:SBA-15 ratio of 1:5) offered higher proportions of C 5 -C 12 and C 13 -C 20 , indicating that SBA-15 derived from fly ash can be utilized as a catalyst in the pyrolysis of oilseed residues for hydrocarbon conversion and consequently improve the quality of bio-oil.

Sacha Inchi (SI) Residues
The SI residues obtained from oil extraction plants in the Kamphaeng Phet province, Thailand were used as a biomass feedstock in the pyrolysis experiment ( Figure 13). Before the experiment, the SI residues were dried, ground, and sieved to achieve a particle size of 125-425 µm. The proximate and elemental analyses of SI residues have been presented in Table 4. The results of the proximate analysis showed the presence of 71.63% volatile matter, 14.58% fixed carbon, 5.63% moisture content, and an ash content of 8.16%. The elemental result revealed that the SI residues consisted of C:41.96%, H:6.97%, N:1.76%, and O:49.31 wt %. Based on this elemental composition, the approximate normalized formula of the SI residues can be written as (C 3.5 H 7.0 O 3.1 N 0.1 )n. In addition, the heating value of the SI residues was quite high at 24.9 MJ/kg, compared to other biomass residues. Figure 13. The physical appearance of (a) Sacha inchi seed and (b) the SI residues obtained after the oil extraction process.

SBA-15 from the Chemical Reagent
The mesoporous silica SBA-15 was synthesized using the hydrothermal method. The initial molar gel composition was 1SiO 2 : 0.017 P 123 : 1.35 HCl: 19.77 H 2 O. For each batch, 4.26 g of Pluronic P123 (EO20PO70EO20; MWav = 5800, Aldrich) was dissolved into a 2M HCl solution and deionized water (D.I. water) and stirred at 40 • C for 1 h. After the complete dissolution of P123, tetraethyl orthosilicate (8.5 g of TEOS, 98%, Sigma-Aldrich) was added to the mixture solution, which was continuously stirred at 40 • C for 20 h. After this homogenous mixing process, the milky solution obtained was transferred to a Teflon-lined autoclave at 100 • C for 24 h for crystallization. The white solid powder was filtered, washed, and dried overnight at 100 • C. Following this, the dry solid sample was calcined at 550 • C for 6 h in the air atmosphere.

Acid Treatment of Fly Ash
The CFA, which was used as a raw material in this experiment, was obtained from the Mae Moh electric power generation plant in Lampang, Thailand. Its chemical composition determined by XRF is given in Table 5. Due to high amounts of impurities, such as CaO and Fe 2 O 3 , in the CFA, the acid treatment was initially executed before the SBA-15 synthesis to improve the purity of the synthesized zeolite [24,37]. This acid treatment procedure was adapted from [24,38]. Then, 10 g of CFA was mixed with 100 mL of 1M HCl and stirred at 90 • C for 1 h to remove the impurities. Following this, the treated powder was washed with deionized water, filtered, and dried at 100 • C for 24 h. After the acid treatment, alkaline fusion and silicate extraction of the treated fly ash were conducted following the method suggested in the literature by Kumar et al. [16]. The treated fly ash was homogenously mixed with sodium hydroxide (NaOH) at a ratio of 1:1.2 in terms of weight. Subsequently, the solid mixture was heated in a muffle furnace at 550 • C for 1 h. After the heat treatment, the fused fly ash was mixed with deionized water at a weight ratio of 1:4 and stirred at room temperature for 24 h. Lastly, the turbid suspension was centrifuged and filtered to achieve the supernatant solution. The initial molar gel composition for synthesizing SBA-15-FA was 1SiO 2 : 0.057Al 2 O 3 : 0.025 P 123 : 2.77 HCl: 65.04H 2 O. For each experiment, 4.48 g of P123 was dissolved into a 37 wt % HCl solution and D.I. water and then stirred at 40 • C for 1 h. Tetraethyl orthosilicate (TEOS) with the SiO 2 proportion of 0.77 and the supernatant extracted from fly ash with the SiO 2 proportion of 0.23 were added to the above P123 solution. Following this, the mixed solution was continuously stirred until a milky gel formation could be seen. Subsequently, this milky solution was stirred at 40 • C for 20 h, after which it was transferred to a Teflon-lined autoclave for the hydrothermal treatment at 100 • C for 72 h. Afterward, the solid product was filtered, washed, and dried overnight at 100 • C. Finally, the dry solid sample was calcined at 550 • C for 6 h in an air atmosphere to obtain SBA-15 derived from CFA (SBA-15-FA).

Raw Material and Catalyst Characterizations
The chemical compositions of fly ash and synthesized SBA-15-FA sample were characterized by X-Ray Fluorescence (XRF: Philips PW 2404) and Micro Energy Dispersive X-ray Fluorescence Spectrometer (µ-EDXRF; Orbis PC; EDAX (AMETEK)) equipped with a rhodium X-ray source (30 kV, 1000 µA) and with a silicon drift detector. The crystal structure of SBA-15 zeolite was investigated using an X-ray diffractometer (Rigaku TTRAX III) equipped with a CuKα source at 50 kV and 300 mA. The scan range of the small angle was at 2θ = 0.5-5 • at a scan rate of 1 • C/min. The morphology of SBA-15 zeolite was determined with the help of a scanning electron microscope (SEM) (Hitachi, SU 5000). The mesoporous SBA-15 was observed through transmission electron microscopy (TEM) images using a JEOL, JEM-2100Plus operated at 200 kV. The surface area, the total pore volume, and the pore size were determined by nitrogen adsorption-desorption isotherms at 77 K in Quantachrome's Autosorb-1 analyzer. The surface area was calculated using the Brunauer-Emmett-Teller (BET) equation, while the total pore volume was measured at P/P0 0.99. Prior to the BET, the surface area measurement samples were heated at 300 • C in vacuum for 8 h. The acidity of the synthesized SBA-15 catalysts was determined by ammonia temperature-programmed desorption (NH 3 -TPD) with the help of a BELCAT instrument. The catalyst (0.3 g) was pre-treated at 500 • C for 1 h under a helium (He) atmosphere. After cooling it to 100 • C, NH 3 gas adsorption was conducted for 1 h and consequently purged with He at ambient temperature for 30 min. Next, the SBA-15 sample was heated up to 600 • C under He flow with a ramping rate of 20 • C/min. The desorbed ammonia was identified using the thermal conductivity detector. The acidity characteristics of SBA-15 samples were determined by pyridine adsorption using a FTIR spectrometer (PerkinElmer, Spectrum Spotlight 300, FTIR). The samples (1.5 mg) were pressed to form a pellet with KBr (100 mg). Pyridine adsorption was carried out by dropping a small amount of pyridine to the samples and then adsorption at 40 • C for 30 min. After adsorption, the samples were consequently evacuated at 150 • C for 1 h in the oven to characterize the acid properties of catalysts [22,23]

Catalyst Activity (Pyrolyzer-GC/MS System)
The catalytic activity of SBA-15 catalysts was performed using a single-shot micro furnace pyrolyzer (Frontier Laboratories, Cabanatuan, Philippines; PY-2020iD) equipped with an auto-shot sampler (Frontier Laboratories, Cabanatuan, Philippines; AS-1020E). The pyrolysis unit was connected to a gas chromatograph (Agilent Technologies, Santa Clara, CA, USA; GC, 6890N GC) and a mass spectrometry detector (Agilent Technologies, Santa Clara, CA, USA; 5975C MSD). For a non-catalytic run, approximately 0.4 mg of SI residues were used in an 80-µL stainless steel crucible, which was then covered with a quartz wool. In order to investigate the effect of the catalyst contents, catalysts (SBA-15-FA, SBA-15-chemical, and SBA-15-commercial) were placed above the biomass layer at the SI residues to SBA-15 ratios of 1:1, 1:5, and 1:10, respectively. The pyrolysis temperature was set at 500 • C. In order to initiate the experiment, the sample crucible was dropped into the heated reactor at the set temperature under He atmosphere (1 cm 3 /min). The sample was rapidly pyrolyzed and held for 30 s. The evolved pyrolysis vapors were automatically injected through an Ultra Alloy-5 (UA-5; 30 m × 0.25 mm and 0.25 µm of film thickness) capillary column into separate gas species. The column temperature profile entailed heating at 50 • C for 3 min, then raising the temperature to 200 • C at 5 • C/min and finally to 350 • C at 10 • C/min. The column flow and total flow were found to be 1.30 and 69.2 mL/min, respectively. The chemical speciation of gas products was performed using mass spectrometry that is based on mass to charge ratios (m/z) of 20-800, and the products were identified through a comparison with the National Institute of Standards and Technology (NIST) mass spectra database. Besides SBA-15-FA, the mesoporous SBA-15-synthesized chemicals and commercial SBA-15 catalysts were also utilized for comparison.
The direct quantitative analysis of condensable products of biomass pyrolysis could not feasibly be extracted from Py-GC/MS data due to complexity of products obtained from the process. It is not generally practical to make the direct quantification of these compounds via a traditional calibrated peak area procedure. Many studies rely on comparison among corresponding peak areas of these compounds to express the selectivity of particular products. In this work, product selectivity is based on percentages of relative peak area of each compound to the total peak areas obtained from the chromatogram as shown in Equation (1) [24]. In this study, the GC set-up did not allow for the separation and quantification of the light gases (e.g., H 2 , H 2 O, CH 4 , CO, CO 2 , C 2 , C 3 ) [39]. For each catalyst, the Py-GC/MS experiment was conducted three times to confirm the reproducibility of pyrolysis products. The relative standard deviations were not much different; generally, they were within 15%. Yu and coworkers (2012) investigated the catalytic fast pyrolysis of lignin extracted from a black liquor collected from a pulp mill and reported the reproducibility of aromatic selectivity, which was approximately 10% [40].

Product selectivity (%) =
Area o f the desired product Total area o f the products × 100% (1)

Coke Analysis
The quantitative amount of coke was determined by thermogravimetric analysis (TGA) on a MettlerToledo (TGA/SDTA 851) instrument. The spent catalyst were heated from 30 to 850 • C and held for 10 min under air atmosphere (60 mL/min) with heating rate 10 • C/min. The mass loss of coke combustion was calculated between 250 and 850 • C in order to avoid the mass loss of desorbed water at low temperature (approximately 100-250 • C) [24]. The relative amount of coke was calculated by Equation (2).

Conclusions
SBA-15 derived from fly ash was successfully synthesized via alkali fusion followed by hydrothermal treatment. The SBA-15-FA catalyst had the Si/Al ratio of 144.9 with a specific surface area and a pore volume of 705 m 2 /g and 1.15 cm 3 /g, respectively, which were greater than SBA-15 synthesized from chemical reagent and commercial SBA-15 catalyst. The incorporation of Al in the structure increased the pore size as well as acid strength, particularly the Lewis acid sites for the SBA-15-FA catalyst. After the catalytic pyrolysis, the carboxylic acids, which were mainly linoleic acid (C18:2), were eliminated through the decarboxylation reaction, while other oxygenated compounds decreased slightly to a SI:SBA-15 ratio of 1:5. As a result, the formation of aliphatic hydrocarbons largely increased in the following order: SBA-15-FA (79.9%) > SBA-15-chemical (69.5%) > SBA-15-commercial (57.3%). However, there was only a slight change in the decrease in N-compounds (9.2-12.9%). However, the mesoporous SBA-15 catalysts exhibited a relatively low aromatic selectivity at 0-4.0%. SBA-15 derived from fly ash (SBA-15-FA) showed high efficiency in terms of improvement in the hydrocarbon selectivity up to 80%, with the ability to decrease the oxygenated compounds to 6.9%.
In terms of the aliphatic compositions, the formation of alkenes was favored over that of alkanes and alkynes; on the other hand, cycloalkanes and cycloalkenes were found as the minor constituents. SBA-15-FA at a high catalyst content (SI:SBA-15 ratio of 1:5) offered higher proportions of gasoline (C 5 -C 12 ) and diesel (C 13 -C 20 ) ranges. Therefore, SBA-15 derived from fly ash was an effective catalyst in terms of improving the overall bio-oil quality through pyrolysis application.