Comparative Study on the Catalytic Performance of a 13X Zeolite and its Dealuminated Derivative for Biodiesel Production

Natural kaolin clay was used to successfully prepare 13X zeolite catalysts, which were modified by dealumination with citric acid. Acid leaching eliminates impurities and aluminum, and improves the Si/Al ratio of the zeolite framework. The X-ray diffraction (XRD) patterns of both the original and modified 13X zeolites were the same, indicating that the crystalline frameworks were not destroyed during the dealumination process. X-ray fluorescence data of the dealuminated 13X zeolite showed an improved Si/Al ratio. Also, Atomic Force Microscopy (AFM) was used for the characterization of the catalysts. The catalytic performance of the original and modified catalysts was tested in the esterification reaction of oleic acid in a batch reactor. A higher conversion of oleic acid was obtained using the modified 13X zeolite. The resulting experimental data from the esterification reactions were fitted to the heterogeneous Langmuir-Hinshelwood-Hougen-Watson (LHHW) kinetic model to determine the rates of reaction. The results of the reaction kinetics showed an increase in the rate of reaction velocity and a distinct decrease in the activation energy when using the modified zeolite, indicating that employing the modified catalyst will give a higher conversion over a shorter time through a reaction with less sensitivity to temperature. Copyright © 2021 by Authors, Published by BCREC Group. This is an open access article under the CC BY-SA License (https://creativecommons.org/licenses/by-sa/4.0).


Introduction
Energy plays an important role in transportation, industrial sectors, and power generation; therefore, energy has become an essential requirement for maintaining and improving living standards [1]. The need for fossil fuels has significantly increased over the past few decades due to rising economies and worldwide declines in fossil fuel reserves [2]. The environmental impact of fossil fuel emissions, fossil fuel depletion, rising oil prices, and growing energy demands *Corresponding Author.
Email: ammarabbas@coeng.uobaghdad.edu.iq (A.S. Abbas); Telp: +9647901307270 has stimulated interest among academics, scientists, and industrialists in developing alternative sustainable fuel. Such alternative fuels should be eco-friendly, economically feasible, efficient, and easily constructed [3]. One example is biodiesel, which is non-toxicity, biodegradability, suitable for sensitive environments, and its greenhouse gas emissions can be controlled because most of its feedstock is renewable [4]. Biodiesel is produced by trans-esterifying triglycerides that generates alkyl esters and glycerin, or by esterifying fatty acids to give alkyl esters and water. Both processes are done with short-chain alcohols, such as ethanol [5] or methanol [6], in the presence of an acid or base catalyst [7]. Biodiesel can be chemically defined as a mixture of ethyl esters with long-chain fatty acids obtained from renewable sources, such as vegetable oils [8], animal fats, sunflower seeds [9]. Such oils and fats may contain about 20% free fatty acids (FFAs), but an ideal amount should not exceed 1% because the saponification of FFAs reduces the fatty acid alkyl ester yield [10]. Therefore, when using raw materials with high FFA content, an esterification reaction is required to produce biodiesel, using supercritical, enzyme, heterogeneous, or homogeneous acid catalysts [11].
Esterification is a significant, alternative process to transesterification. Although the availability of natural, raw materials rich in triglycerides makes transesterification more viable, esterification reactions employ materials with FFA-rich residues and by-products, allowing them to deal with industrial biomass [12]. Esterification is the reaction between a fatty acid and an alcohol that occurs in the presence of an acid catalyst. Traditionally, biodiesel is produced with homogeneous catalysts (e.g. sulfuric acid), which are strong mineral acids that produce high reaction conversions [13] but erode equipment. Therefore, the homogeneous catalysts can be replaced by heterogeneous catalysts.
Heterogeneous acid catalysts produce biodiesel in an economically feasible way. They slowly convert fats or oils into biodiesel and are reusable; thus, the production cost is low and the catalysts have high activity, are stable, and it is easy to separate from the products [11]. Various heterogeneous acid catalysts are used, the conventional ones being solid resin, enzymes, and zeolites. Several types of zeolites have been used to produce biodiesel: zeolite beta [14], NaY zeolite [15], HY zeolite [16], ZSM-5 [17], mordenite [18], FAU-type zeolite [19], [20], ZSM-5 modified with citric acid [21], modified mordenite [22], and 13X zeolite [23] that gives a low conversion rate due to its low Si/Al ratio. All of these zeolites are considered heterogeneous catalysts.
Several post-synthesis treatments for zeolites were used to raise their Si/Al ratio, which would improve their catalytic properties. The dealumination processes, including acid leaching and steaming, change the quantity (and ul-timately the strength) of Brønsted acid sites, which may produce Lewis sites when the aluminum frameworks are removed. Simultaneously, meso-porosity is generated that promotes the accessibility of the acid sites remaining in the framework, which has been particularly relevant when large molecule transformations were tested over zeolite [24].
This work prepared 13X zeolite from natural kaolin clay, which was subsequently modified by dealumination with citric acid to improve its Si/Al ratio and its ability to convert oleic acid into ethyl oleate via esterification with ethanol. The influence of reaction temperature and time on oleic acid conversion is discussed herein, and the kinetics of the batch esterification reactions were studied to compare the performance of the original and modified zeolite catalysts with those of other zeolite catalysts.

Preparation of 13X Zeolite
Natural kaolin clay was converted to active kaolin (metakaolin) by calcination at 550 °C for 2 h in an electric heater [25]. Table 1 shows the chemical composition of kaolin clay. Then the metakaolin was ground and sieved until a particle size of ≤ 75 microns was achieved. The reactant mixture expressed in terms of oxide mole ratio which has a composition of [26]: (SiO2 /Al2O3)/(Na2O/SiO2)/(H2O/Na2O): 4/1/40 Metakaolin and sodium trisilicate were mixed in a two-necked round bottom flask with sodium hydroxide solution (9 wt.%). The mixture was stirred for 72 h at 70 °C with a variable-speed, stainless-steel paddle mixer equipped with a motor driven by a relatively medium revolution (Voss Instrument Ltd. Type SG/PA/ST). An electrical mantle heater was used to heat the flask, while variable input voltage was used to control the temperature of the solution. Upon completion of the reaction, the mixture was decanted and washed with deionized water to remove any excess alkali. Once a pH of 10 was reached, the product was filtered with a Buchner funnel attached to a vacuum pump ( EiM5, serial 0146 g), dried at 120 °C for 12 h in an electric oven (Tecnoformal Ltd., Max. Temp. 250 °C), and then milled using a centrifuge ball mill (Type RETCH, S2) equipped with seven steel balls to grind the zeolite into a fine powder.

Modification of 13X Zeolite
Ten grams of the original 13X zeolite was treated with 150 mL of 0.1 N citric acid solution at 75 °C for 3 h [27]. The final product was subsequently recovered by vacuum filtration, washed with deionized water to eliminate any complexes of aluminum that may have formed, and then dried in an electric oven at 100 °C. Finally, the solid sample was calcined at 500 °C for 3 h.

Characterization of Catalysts
The structural characterization of the original and modified 13X zeolite samples was carried out by X-ray diffraction (XRD) using a D2 PHASER diffractometer (Bruker AXS, Karlsruhe-Germany) using Cu-K radiation with a nickel filter (=1.541874 Å). The diffraction patterns were obtained within the 2θ range of 3° and 65° with a scan speed of 0.04 degrees per second. The X-ray fluorescence (XRF) analysis was performed on a Spectro XEPOS spectrometer (Germany) with X-LAB Pro software to determine the chemical composition of the sample, i.e., the Si/Al molar ratio. Brunauer-Emmett-Teller (BET) surface area and pore volume of the original and modified 13X zeolite were measured by a Flowing Gas Surface Area Analyzers (SA-9600 Series, HORIBA Scientific, Japan). Atomic Force Microscopy (AFM) was used to determine the particle size and morphology of the catalysts at the nano-level and was performed using an SPM-AA3000 (Angstrom Advanced Inc., Stoughton, Wisconsin, USA). The morphology of the original and modified 13X zeolite was studied by scanning electron microscopy (SEM, Inspect S50, FEI, USA).

Catalysts Testing for Oleic Acid Esterification
Esterification reaction was performed in a 3-neck 500 mL round bottom flask fitted with a stopper, water-cooled condenser, and a thermometer to measure the solution temperature. The reaction mixture was stirred and heated using an electromagnetic hot plate. An ethanol to oil molar ratio of 6:1 and 5 wt.% (relative to oleic acid) of the catalyst was used in all the experiments. The reaction temperature ranged from 30 to 70 °C [13] and the reaction time was up to 180 min.
The reactor was loaded with 150 mL of oleic acid and the desired amount of ethanol. Since ethanol and oil are immiscible, maintaining constant agitation at 300 rpm is necessary to increase the contact surface. After the reaction mixture was heated to the desired reaction temperature, the catalyst was added. At different intervals, 2 mL of the reaction mixture was withdrawn and centrifuged (Griffin and George, Loughborough, Britain) for 10 min at 3000 rpm to improve the phase separation. The top layer was titrated with 0.1 N KOH using phenolphthalein as an indicator to obtain the acid value (AV), as shown in Equation (1). (1) Oleic acid conversion for each quantity of the catalyst can be calculated using Equation (2) [13,16]. (2) where AVto is the initial acid value (i.e. the acid value of oleic acid: 196.35) and AVt is the acid value of the supernatant layer at time t.

Catalysts Characterization
XRD patterns of the original and modified 13X zeolite catalysts at diffraction peaks between 3° and 65° (2θ) are shown in Figure 1 (blue and red lines, respectively). The original 13X zeolite pattern was similar to those of standard 13X zeolites [28] and the modified 13X zeolite. All XRD patterns obtained of the  original and modified zeolite indicate that both catalysts possess highly crystalline structures, which means that the crystalline frameworks were not destroyed during the modification process [29]. However, new diffraction peaks were observed in the dealuminated catalyst pattern (Figure 1, red  (3) The Si/Al ratio for the original 13X zeolite was 1.74, which is in a good agreement with the value obtained by Breck [30]. Chemical composition changes are evident in the modified 13X zeolite, which is ascribed to its improved Si/Al ratio via dealumination. The Si/Al ratio of the modified 13X zeolite was 4.72, nearly 2.7 times that of the original 13X zeolite. An improved Si/Al ratio occurs due to the exchange of cations existing on the catalyst surface (Na + , Ca 2+ , Mg 2+ , etc.) with a proton from the solution. This exchange forms hydroxyl bridges, which are converted to SiOH as a result of the reaction between the solution H + and the framework Al [31,32].
AFM was used to determine the particle size distribution and topography of both the original and modified 13X zeolite. The AFM images in Figure 2 [33], where a highly negative skewness value implies that the surface contains numerous local maxima above the mean line [34]. If the kurtosis parameter is added, an even better representation of a rough surface can be obtained. This parameter is a description of the surface peakedness, and it was equal to 1.8 and 2.18 for the original and modified 13X zeolite, respectively. For reference, a kurtosis value <3 suggests a broad morphology [35]. The granularity accumulation distribution shown in Figure 3 (a and b) and reveals that the average particle size for the original 13X zeolite was 72.39 nm, where ≤10% of the particles had a diameter of 45 nm, ≤50% of the particles had a diameter of 70 nm, and ≤90% of the particles had a diameter of 95 nm. Meanwhile, the average particle size for the modified 13X catalyst was 113.37 nm, where ≤10% of the particles had a diameter of 50 nm, ≤50% of the particles had a diameter of 100 nm, and ≤90% of the particles had a diameter of 170 nm. This difference in diameters may be due to the processes of desilication and dissolution of the zeolite framework that begins with the Si−O−Si linkages [36], which would decrease the crystallinity. Generally, measuring particle size by AFM techniques is considered more reliable.

Influence of Reaction Temperature and Time on Oleic Acid Conversion
Esterification reactions can occur at different temperatures, and studying the effect of reaction temperature on heterogeneous catalytic reactions is important because this information helps calculate the activation energy of these reactions. In addition, the constants of the reaction rate are strong, temperature-dependent functions. Figure 4 (a and b) presents the results obtained for oleic acid conversion with time at various temperatures using 5 wt. % of either the original or modified 13X zeolite with an ethanol/oleic acid (6:1) mixture. This molar ratio and amount of catalyst are optimal values reported by Abbas and Abbas [15]. As expected, oleic acid conversions increased with reaction time at all temperatures and it can be observed that during the first 90 min of the reaction the change in the conversion is high and then it stabilizes. For the modified 13X zeolite, the change is rapid during the first 10 min. After that, the change is negligible until it stabilizes

Bulletin of Chemical Reaction Engineering & Catalysis, 16 (4), 2021, 768
Copyright © 2021, ISSN 1978-2993 ( Figure 4 (b)). Increasing the reaction temperature had a favorable effect on the conversion of oleic acid. The oleic acid highest conversion over the original 13X zeolite was only 47% (Figure 4 (a)). Improvement in the conversion of oleic acid was observed when the modified 13X zeolite was employed, exhibiting a conversion of 73%, the highest conversion achieved in this study (Figure 4 (b)). These results show that zeolites provide better conversions for esterification reactions when they are dealuminated with citric acid [37]. An increased conversion of oleic acid with the modified zeolite could be ascribed to several factors. The quantity and intensity of acid sites, which increases with increasing the ratio of Si/Al, may play an important role in oleic acid conversion. In addition, citric acid dealumination leads to a higher external surface area, which consequently increases the number of acid sites available to oleic acid molecules [38].

Heterogeneous Reaction Kinetics of Oleic Acid Esterification
Esterification of oleic acid is treated as a heterogeneous catalytic reaction because the reaction mixture contains more than one phase: the catalyst is a solid and the reactants and products are liquids. The Langmuir-Hinshelwood-Hougen-Watson (LHHW) model is used to express the esterification reaction with many heterogeneous catalysts used in this reaction [39]. This kinetic model assumes that both the oleic acid and ethanol molecules are adsorbed onto the catalyst surface before the surface reaction. The final stage of the mechanism involves desorption of both products, ethyl oleate and water. If the surface reaction is assumed to be the rate-controlling step, the LHHW model equation for the reaction of oleic acid with ethanol is first derived in terms of surface concentration of the adsorbed species and vacant sites: (5) According to the mechanism, the surface reaction takes place between adsorbed oleic acid (OA) and adsorbed ethanol (EtOH), producing adsorbed ethyl oleate (EO) and adsorbed water (W). The rate of the surface reaction is given in Equation (6), where Ks is the ratio of ks to k's (Eq. 7).
As all other steps are considered to be in balance, the concentration of each adsorbed species can therefore be obtained: (8) where i is OA, EtOH, EO, or W. Then, the adsorbed phase concentrations can be written as: Substituting all these values (Eq. (9) to (12)) into Equation (6) gives: Therefore, which can be simplified into: Finally, solving for Cv gives: For the reaction OA + EtOH ↔ EO + W, the overall equilibrium constant is: All concentrations correspond to the equilibrium conditions. Substituting Equations (9) -(12) into Equation (18) gives: (19) which can be simplified into: Since, Substituting Equations (9) -(12) into Equation (6) gives: (23) which can be rewritten as: The LHHW model (Eq. 24) was solved numerically by minimizing the Chi-squared statistic ( 2 ) and maximizing the correlation coefficient (R 2 ), as shown in Equations (25) and (26), respectively.
(25) (26) Statistical analysis for the experimental rates (Exp. rA) versus the calculated rates (Cal. rA) obtained from the LHHW model is shown in Figure 5. It can be observed that there is an excellent agreement between the calculated and experimental values of the rates for all the esterification reactions with modified 13X zeolite. This was not observed for the reactions using the original 13X zeolite.
Regression statistics for the original and modified 13X zeolites are shown in Table 2, where the mean rates for the experimental and calculated work values of each catalyst were close. However, when comparing the mean rates of the original and modified catalyst, differences are evident. The mean rate of reaction over the original catalyst was 0.0022 while the mean rate over the modified catalyst was 0.0036. This shows that the average rate was improved when using the modified 13X zeolite, resulting in a more rapid esterification reaction.
Tables 3 and 4 present the results of the analyses using the LHHW kinetic model fitting, where R 2 values are also provided. In these tables, ks represents surface reaction rate constant, KS represents the equilibrium surface  [40] were used to determine then the activation energies (E) of esterification reactions over the original and modified 13X zeolites ( Figure 6). The activation energy for the esterification reaction with the original 13X zeolite was 72.64 kJ/mol, while that of the reaction with the modified 13X zeolite was lower, at a value of 34.76 kJ/mol. This implies that the reaction became less sensitive to temperature when the modified catalyst was used, which improves the expected quantity of production. For the esterification reaction with the original and modified 13X zeolites, the activation energies were calculated to be 72.64 and 34.76 kJ/mol, respectively.

Conclusion
A 13X zeolite catalyst was successfully prepared from local, invaluable kaolin clay. XRD was used to determine its crystalline structure. The prepared catalyst had 1.74 molar ratio of Si/Al, and sodium content of 7.441, which matched well with the values reported for standard 13X zeolite. AFM tests revealed that the average diameter of the original 13X zeolite was 72.39 nm. The conversion of esterification reaction of OA with ethanol was only 47% when a 13X zeolite was used as a heterogeneous catalyst. To improve performance of the catalyst, the Si/Al ratio was increased by dealuminating the original 13X zeolite with 0.1 N citric acid at 75 °C for 3 h. The modified catalyst had the same XRD pattern as the original one, but showed a higher Si/Al ratio (4.72). The AFM tests determined the average diameter to be 113.37 nm. The performance of the modified 13X zeolite was tested via the same OA esterification reaction where it showed remarkable activity and a conversion of 73%. LHHW kinetic models were employed to fit the experimental kinetic data of the heterogeneous catalytic reactions. The activation energies were determined to be 72.64 and 34.76 kJ/mol for the esterification reaction with the original and modified 13X zeolites, respectively. The lower activation energy observed when the modified catalyst was used indicates that the reaction became less temperature sensitive. The mean rate for the esterification reaction over the original 13X zeolite was 0.0022, while a slightly better rate of 0.0036 was observed when the modified 13X zeolite was employed. Our results indicate that treating the original 13X zeolite catalyst with citric acid did not destroy its crystalline framework, but provided a modified catalyst with a higher Si/Al ratio than the original that also increased the esterification reaction rate and achieved higher OA conversion.