The Inﬂuence of Si/Al Ratio on the Physicochemical and Catalytic Properties of MgO/ZSM-5 Catalyst in Transesteriﬁcation Reaction of Rapeseed Oil

: This work presents the comparative physicochemical and catalytic studies of metal oxide MgO catalysts in a transesteriﬁcation reaction. The inﬂuence of the Si/Al ratio in the catalytic material on their catalytic properties in the studied process was extensively evaluated. In addition, the effect of the type of zeolite ZSM-5 form on the catalytic reactivity of MgO based catalysts was investigated. In order to achieve the main goals of this work, a series of MgO/ZSM-5 catalysts were prepared via the impregnation method. Their physicochemical properties were studied using X-ray diffraction (XRD), BET, FTIR and TPD-CO 2 methods. The highest activity in the studied process exhibited MgO catalyst supported on ZSM-5 characterized by the highest ratio between silica and alumina. The most active catalyst system in the transesteriﬁcation reaction was 10% MgO/ZSM-5 (Si/Al = 280), which showed the highest value of higher fatty acid methyl esters (94.6%) and high yield of triglyceride conversion (92.9%). The high activity of this system is explained by the alkalinity, sorption properties in relation to methanol and its high speciﬁc surface area compared to the rest of the investigated MgO based catalysts.


Introduction
The increase in the number of people along with changes in lifestyle and rapid industrialization causes a continuous increase in energy demand in recent years [1][2][3]. A significant part of this demand is satisfied by the consumption of natural resources, the availability of which is increasingly limited, and their constant use may eventually lead to their depletion [1]. One solution to these problems is biodiesel, which is a liquid, renewable, clean alternative fuel that has received a lot of attention due to oil depletion and environmental concerns [2,[4][5][6]. Biodiesel has a number of advantages compared to traditional diesel fuel used in diesel engines, such as biodegradability, reduction of pollutant emissions to the atmosphere, water and soil, non-toxicity, no carcinogens, low sulfur content and low aromatic content [4,7]. Biodiesel is characterized by a high cetane number, high flash point and contains approximately 10% of oxygen-containing chemicals, which improves the complete combustion of the fuel and the quality of ignition, even in mixtures with petrodiesel [4]. Combustion of biodiesel is characterized by the reduction of emissions of carbon dioxide (CO 2 ), sulfur oxides (SO x ) and unburned hydrocarbons. An additional advantage is that no significant changes are required in conventional diesel engines running on biodiesel or its blends with conventional diesel fuel [3] and emit significantly less soot and carbon monoxide (CO) in the combustion gases [3,8]. Biodiesel is defined as a mixture of long chain fatty acid methyl or ethyl esters produced by the esterification and transesterification reactions of free fatty acids (FFA) and triglycerides that naturally occur in renewable biological sources such as vegetable oils and animal fats with shortchain alcohols [9][10][11]. Various primary alcohols can be used in the transesterification (see Scheme 1), e.g., methanol, ethanol, propanol and butanol, which their cost is low [4]. Due to its physical and chemical nature, methanol is most often used in transesterification [10]. Natural vegetable oils without chemical modification, which, in addition to triglycerides, also contain free fatty acids have a negative feature of high viscosity and when applied directly, they can form carbon deposits and block nozzles [4,10,12]. In this reaction, catalysts are often introduced in order to increase the rate of reaction and ensure better efficiency of the conversion of oil to biodiesel [9]. There is a possibility of transesterification without the use of a catalyst, but it requires special conditions, such as a relatively long reaction time, high temperatures and pressures [2]. Transesterification can be carried out homogeneously or heterogeneously, with acid/base chemical catalysis, ionic liquids catalysis and enzymes bio-catalysis [4,6]. Traditional industrial methods of biodiesel production use homogeneous alkaline catalysts [6], such as alkaline hydroxides or metal alkoxides [11]. Homogeneous alkali catalysis ensures fast conversion of triglycerides from oils to biodiesel [3], with high efficiency under mild reaction conditions [6]. An additional advantage of basic catalysts can provide higher reaction rates compared to acid catalysts in the production of biodiesel [4]. The problems of alkaline homogeneous catalysis are the saponification of fatty acids and alkali cations [3], corrosion of equipment [8,13] and sensitivity to free fatty acid content of oil [14], which leads to the formation of stable emulsions and soaps [6,15]. Those type of catalysts cannot be used for high content of free fatty acid lipids in feedstock [7] and has to be anhydrous [15]. It is required to purify the obtained biodiesel through the necessary neutralization [3]. The troublesome and complex separation of products and catalysts from the final reaction mixture [3,8], which makes this process less competitive and uneconomical, and in addition it leads to a number of environmental problems related to the use of large amounts of solvents, energy [11], water for washing and wastewater, which has to be purified [15]. Typical catalysts of homogeneous acid-catalyzed transesterification are hydrochloric acid (HCl), sulfuric acid (H 2 SO 4 ) [5], organic sulfonic acid [16], phosphoric acid [11]. The advantage of acid catalysts in transesterification is their insensitivity to free fatty acids and the water content of the oil. Acid-catalyzed reactions can be carried out under mild conditions, both in terms of temperature and pressure [14]. Besides being corrosive to reactor and equipment and difficulties of separation and purification from the liquid products, the drawbacks of homogeneous acid-catalyzed transesterification are longer time reaction than for alkaline-catalyzed reaction [17], much slower reaction rate [14], higher temperature, more waste from neutralization, difficult recycling of catalyst, weaker catalytic activity [16]. Due to the difficulties of homogeneous catalyzed transesterification, in order to avoid or minimize the problems related to homogeneous catalysis, interest in heterogeneous catalysis has been attracted [18]. The main advantage of heterogeneous catalysis is the ease of separation form the final product [2,12]. Heterogeneous alkaline catalysts are characterized by high possibility of reuse and regenerate the catalyst, recyclable, noncorrosive and environmentally friendly [2,16]. There are fewer disposal problems and lifetimes of this type of catalysts are longer. Heterogeneous alkaline catalysts share the same advantages as homogeneous catalysts, but also suffer from the same disadvantages, such as requirement of low FFA content in oil, requirement of anhydrous feedstock. Another demerits are requirement of higher molar ratio of alcohol to oil, high reaction temperature and pressure, diffusion limitations, poisoning of the catalyst when exposed to ambient air and high cost of synthesis. The advantages of heterogeneous acid-catalyzed transesterification are similar as for homogeneous acid-catalyzed transesterification, are insensitive to the content of FFA and water in the raw material, with a simultaneous catalysed reaction. The drawbacks of usage homogeneous acid catalysts are high cost of synthesis, which often is complicated; diffusion limitations, higher reaction temperature with a high alcohol to oil molar ratio, longer reaction time [16].
The most commonly used heterogeneous catalysts for the synthesis of biodiesel are Mg, Ca, Sr and Cs oxides, alkali hydroxides and alkali metal salts immobilized on supports and zeolites modified with K and Cs cations [19]. Zeolites are microporous aluminosilicate minerals occurring in several types [9]. Zeolites are high surface area solid acids that are characterized by homogeneous mesopores and micropores with specific pore characteristics in molecular dimensions [3,12]. Zeolites are available with different pore sizes and regulated acidity [20,21]. Their proton form is characterized by high thermal stability [20]. The acidic properties of zeolites result from the imbalance of charges between the elementary building units of zeolite, which are [SiO 4 ] 4− and [AlO 4 ] 5− tetrahedral bound in a stable, crystalline, three-dimensional, skeletal, aluminosilicate structure [3,12,22,23]. A wide variety of cations can be accommodated in the zeolite lattice, such as Na + , K + , Ca 2+ , Mg 2+ and many others are also found in the zeolite lattice [3,21]. Besides that, zeolites have shape selectivity [21], eco-friendliness, high size selectivity, high porosity and high surface area [4]. Zeolites are considered to be versatile catalysts due to the ability to control the basicity and the ability of ion exchange with alkali metal ions and the ability to trap metal ions or their oxides inside the micropores [13]. The acid-base properties of zeolites are controlled by the kinds and quantities of ion-exchanged cations, and by the Si/Al ratio of the main zeolite framework [24]. These properties allow the use of zeolites as selective heterogeneous catalysts, adsorbents, for the treatment and purification of water from the ion exchange beds and for humidity control [3,12]. In catalysis, more attention is paid to synthetic zeolites than to natural zeolites, due to better control of an undesired impurities and chemical composition as well as possession of optimal properties for catalytic applications. Zeolite catalysts have some disadvantages in biodiesel production in transesterification process of oils of either animal or vegetable origins, such as higher reaction temperature, longer reaction times, costly and/or complicated processes to prepare them and lower conversion efficiency compared to the process realized on homogenous catalysts. Therefore, it is essential to develop zeolite catalysts, which will be characterized by high activity and selectivity in order to reduce the cost of the produced biodiesel [9].
The main objective of this study was to determine the effect of the silicon to aluminum ratio in the MgO based catalysts. The physicochemical properties of the catalysts were studied by BET, CO 2 -TPD, FTIR and XRD methods. Based on the achieved results, the differences in the catalytic activity of these systems in the transesterification reaction of vegetable oils with methanol were explained.

Transesterification of Vegetable Oil with Methanol Reaction
The biodiesel production via transesterification of vegetable oil with methanol reaction was carried out at 220 • C for 2 h over zeolite supported catalysts. Magnesium oxide (MgO) was the active phase component. The prepared catalysts were calcined for 4 h in an air atmosphere at 500 • C. The activity of the reaction was expressed as triglycerides (TG) conversion values and fatty acids methyl esters (FAME) yield. The reaction was performed in an autoclave with continuous stirring of the reaction mixture. Each process was performed for 2 h in order to obtain the final product of a mixture of the fatty acid alkyl esters, known as biodiesel. The obtained results in the studied process were presented in Table 1 and Figures 1 and 2, respectively. The catalytic activity tests carried out in the biodiesel production process showed that all prepared catalysts exhibited a very high TG conversion with a high FAME yield (see Figures 1 and 2). The lowest TG conversion (88.9%) and efficiency to FAME (79.3%) was demonstrated by the 10% MgO/Al 2 O 3 catalyst. A similar triglyceride conversion was exhibited by catalysts loaded onto the hydrogen form of ZSM-5 zeolites derived from heating the ammonium form of the NH 4 -ZSM-5 zeolite at 500 • C for 15 h, but this system demonstrated a higher FAME yield. The results of the activity tests show that the increase of the silicon to aluminum ratio in the zeolite support has a positive effect on triglycerides conversion and yield of FAME production (see Figures 1 and 2).   Furthermore, there are differences in catalytic activity between the catalysts impregnated onto the hydrogen form of ZSM-5 zeolites and the catalysts impregnated directly onto the NH 4 -ZSM-5 zeolite. For the catalysts on the HZSM-5 support, the TG conversion is lower by about 5%, while the FAME yield decreased by about 8-10%, in comparison to the catalysts impregnated on the NH 4 -ZSM-5 form of zeolite. This implies that no prior calcination of the zeolite support is needed to achieve high activity in the tested process. The catalytic system having the highest activity was the one containing 10% MgO/ZSM-5 (Si/Al = 280). This catalyst exhibited the highest value of FAME yield of 94.6% with the second highest triglyceride conversion (92.9%). This catalytic system, due to its high activity and high selectivity makes it considerable to use for testing it on an industrial scale.
Qu et al. [25] investigated MgO/ZSM-5 catalysts synthesized via impregnation and ultrasonic dispersion methods in the transesterification of Spirulina platensis algae oil with ethanol. The authors achieved 92.1% yield of biodiesel after 1 h reaction at 75 • C with 3 wt.% of catalyst concentration and an alcohol-to-oil ratio of 15:1. The ZSM-5 zeolite was activated with NaOH and dried at 600 • C for 6 h in a muffle furnace with the calcination. In another article, Qu et al. [26] investigated the microwave-assisted in-situ transesterification of Spirulina platensis algae oil with PEG/MgO/ZSM-5@Fe 3 O 4 magnetic catalysts synthesized using polyethylene glycol (PEG) as a mediator. The transesterification reaction was carried out in a flask, which was placed in a microwave reactor. Before the reaction, the catalysts were calcined at 650 • C in a nitrogen atmosphere for 4 h. The authors obtained the highest biodiesel yield of 95.8% for a reaction temperature of 70 • C, a reaction time of 40 min, an ethanol/biomass mass ratio of 8:1, using a 2 wt% catalyst in which the weight percentage of PEG was 15 wt% (15PEG/MgO/ZSM-5@Fe 3 O 4 ). Li and Rudolph [27] introduced MgO into MCM-41, SBA-15, and KIT-6. All catalysts were synthesized by impregnation and by in situ coating. Prepared catalysts were applied to transesterification of commercial-ediblegrade blended vegetable oil with ethanol in a batch reactor under continuous stirring. The most effective catalyst was found to be SBA-15 impregnated with MgO, achieving a high conversion of 96% in a reaction carried out at 220 • C for 5 h. The activity of the studied zeolites can be improved by controlling the physicochemical properties through the dealumination process in various processes, not only in the biodiesel production process via transesterification of the vegetable oil and methanol. Sithole et al. [28] prepared MgO/Al 2 O 3 catalysts for biodiesel production with waste vegetable oils as a feedstock. The oil conversion of 95% was achieved for 20% MgO/Al 2 O 3 catalyst for the reaction carried out at 225 • C, for 60 min, with a methanol to oil ratio of 18:1 using 3 wt.% of catalyst. Mohadesi et al. [29] investigated alkaline earth metal oxides (MgO, CaO and BaO) supported on silica in transesterification of refined corn oil with methanol. All experiments were performed in 250mL two-necked flask equipped with a thermometer and a condenser and the reaction mixture with catalyst was stirred at 600 rpm at 60 • C for 8 h with the molar ratio of oil-to alcohol of 1:16. The amount of used catalyst was 6 wt%. Methyl ester content in the final product of reaction with 60%MgO/SiO 2 catalyst calcined at 650 • C for 3 h was 81.4%.
In order to verify the amount of MgO in the studied catalysts supported on ZSM-5 zeolites and to find out whether small differences in magnesium oxide content could be important to justify the reactivity results observed for investigated materials in the transesterification process, the catalytic material samples were measured by FAAS technique. The obtained results are shown in Table 2. The calculated MgO loading for the catalysts supported on ZSM-5 zeolites was 10%. The highest amount of magnesium oxide (9.7779%) was determined by the FAAS technique for the 10% MgO/ZSM-5 (Si/Al = 23) catalyst, which was directly impregnated on the ammonium form of ZSM-5 zeolite and exhibited the lowest TG conversion (92.7%) and the lowest FAME yield (86.1%) compared to the rest of the catalytic materials synthesized the same way, while the lowest amount of magnesium oxide (8.6448%) was detected for the 10% MgO/ZSM-5 (Si/Al = 280) catalyst, which exhibited similar TG conversion (92.9%), but the highest FAME yield (94.6%). Lower magnesium oxide content was found for catalysts impregnated directly onto ammonium forms of support characterized by a higher silicon to aluminum ratio. In the case of the magnesium oxide catalyst supported on the hydrogen form of ZSM-5 zeolite, higher MgO content (9.2732%) is found for 10% MgO/HZSM-5 (Si/Al = 50) catalyst. While FAAS analysis performed for the 10% MgO/HZSM-5 (Si/Al = 280) catalyst showed 8.9759% wt. of MgO. The conversion of triglycerides and FAME yield exhibited by MgO catalysts supported on HZSM-5 in the transesterification reaction of rapeseed oil with methanol were lower than for catalysts directly impregnated on NH 4 -ZSM-5 zeolite.
Theoretically, the higher content of MgO as the active phase in the supported catalysts should improve the activity, due to the higher number of active centers. However, the obtained activity results suggest that the reactivity of the studied systems may be affected by the properties of the support i.e., acid-base properties, specific surface area. Another possible explanation for the observed differences in the catalytic activity of the different systems could be the higher dispersion of active phase and/or better availability of active sites for the reactants in the systems where a lower MgO content was determined by the FAAS technique. This could result in higher TG conversions and FAME yields.

Specific Surface Area Measurements of the Catalytic materials
The specific surface area (SSA) measurements were done for MgO catalysts using the N 2 adsorption-desorption method. All values of the specific surface area, pore volume and average pore radius are presented in Table 3 Table 3 show that the increasing Si/Al ratio in the catalytic system results in an increase of the SSA for both NH 4 -ZSM-5 and HZSM-5 forms of the zeolites. It is also worth mentioning that the MgO catalysts supported on both forms of zeolites characterized by the highest ratio between Si/Al exhibited higher catalytic activity in the biodiesel production process compared to the rest of the investigated catalysts.

Phase Composition Studies of Catalysts
The XRD analysis was performed and also used to determine the phase composition studies of the magnesium oxides supported zeolite catalysts after their calcination process. The analysis of these measurements were given in Figure 3. Analyses of the diffraction patterns of the calcined magnesium oxide catalysts showed the respective diffraction peaks for the relevant catalysts. The characteristic diffraction peaks for magnesium oxide ( . All the MgO catalyst samples supported on the ZSM-5 zeolites display sharp intensity peaks on the XRD patterns, which can be considered good preservation of the ZSM-5 zeolite structure during sample preparation [30]. An analysis of the diffraction curve of 10% MgO/SiO 2 catalyst shows that this sample is a completely amorphous system. The pattern presents only a broad diffraction peak with a value of 2θ angle equals 23.00 • assigned to the silica crystallographic phase. Unfortunately, the amorphosity of 10% MgO/SiO 2 catalyst did not allow the crystallite size to be determined from the XRD data obtained using the Scherrer equation. The XRD diffraction curve recorded for the 10% MgO/Al 2 O 3 catalyst showed slightly better crystallinity compared to the MgO catalyst supported on silica. The XRD pattern  In addition, the phase analysis of all the prepared catalysts, the crystallite size of the active phase in the studied systems was also investigated by calculating the crystallite size of MgO based on the Scherrer equation [33]: where D is the mean dimension of the crystallite perpendicular to the plane (hkl); β is the integral, full widths at half maximum (FWHM) in radians; k = 0.89 is a constant dependent on the crystallite shape, and λ is the X-ray wavelength. The diffraction patterns were processed using HighScore Plus software by fitting to pseudo-Voigt function. Determining the size of the crystallites was quite difficult due to the large number of peaks originating from ZSM-5 zeolites and low crystallinity for the catalysts supported on alumina. The results of the calculation performed for the calcined catalysts are given in Table 4. The lowest size of magnesium oxide crystallites was observed for 10% MgO/ZSM-5 calcined catalyst (Si/Al = 23). This catalyst exhibited high TG conversion and FAME yield, however, the obtained activity results were not the best. For the catalytic systems supported on ZSM-5 zeolite supports featuring silicon/alumina ratios of 50 and 280, the size of MgO crystallites increased.

Basic Properties of the Synthesized Catalyst Systems
The CO 2 -TPD experiments were used to characterize catalytic materials' alkalinity and the results are shown in Table 5. All the catalytic materials were previously heated at 600 • C before a proper alkalinity test. The results show that all the catalysts exhibited three types of desorption centers with various strengths (weak, medium and strong) and mainly have weak desorption centers (100-300 • C). The sum of the basicity values of all materials indicates that the material with the highest alkalinity was a 10% MgO/HZSM-5 (Si/Al = 50) catalyst (0.75 mmol/g), while the 10% MgO/SiO 2 catalyst has the lowest alkalinity (0.35 mmol/g). In general, as the amount of silicon to aluminum in the catalyst support increases, a decrease in the basicity of the catalyst system is observed. A significant increase in alkalinity is seen for the catalyst loaded on the hydrogen form of the zeolite ZSM-5 (Si/Al = 50) compared to the catalyst loaded on the ammonium form of ZSM-5, while the difference in alkalinity between the MgO systems loaded on HZSM-5 (Si/Al = 280) and NH 4 -ZSM-5 (Si/Al = 280) is negligible. The oxygen-related base sites in the occurring pairs Mg 2+ -O 2− and Al 3+ -O 2− are responsible for CO 2 desorption at about 200 • C. Desorption in the temperature range of 250-300 • C can be attributed to stronger basic sites corresponding to isolated O 2− anions, while desorption at temperatures above 350 • C is attributed to corresponding isolated O 2− but localized at a specific surface site [35]. Table 5. The amount of CO 2 adsorbed on the support surface (calcined in an air atmosphere for 4 h at 500 • C), magnesium oxides catalysts supported on ZSM-5 zeolite with a different silicon to aluminum ratio, calculated from the CO 2 -TPD profiles. The ratio of silicon to aluminum is given in parentheses.

Sorption Properties of Magnesium Oxide Catalysts Supported on ZSM-5 Zeolites in Relation to Methanol
The next stage of physicochemical studies conducted on the synthesized catalytic systems investigated the sorption properties of MgO catalysts supported on ZSM-5 zeolites with different SiO 2 /Al 2 O 3 ratios calcined at 500 • C in an air atmosphere for 4 h with the methanol-argon mixture. In addition to this, MgO/Al 2 O 3 , MgO/SiO 2 and MgO/HZSM-5 systems, which were prepared and calcined in an air atmosphere for 4 h at 500 • C for comparison, were also investigated. The purpose of this study was to clarify and understand the activity of the studied systems in the biodiesel production process. The results of FTIR studies are shown in Figure 4. Several characteristic IR bands can be easily distinguished in the spectra recorded for the studied catalysts. The bands seen in the FTIR spectra above 3250 cm −1 for the MgO catalysts supported on ZSM-5 zeolites correspond to the assigned OH stretching [36]. All the catalysts in the IR spectra show specific bands in the region 2956-2948 cm −1 , which are attributed to (ν asym (CH 3 )) antisymmetric CH stretching vibrations and in the region 2859-2832 cm −1 , which corresponds to (ν sym (CH 3 )) symmetric CH stretching vibrations from methoxy groups [37,38]. It should be emphasized that the highest intensity of these bands was observed for the 10% MgO/ZSM-5(23) system, which confirms the strongest adsorption of methanol on the catalyst surface compared to the other studied catalysts. On the other hand, the intensity of these bands is smaller for the 10% MgO/ZSM-5(50) and 10% MgO/ZSM-5(280) catalysts. This result indicates that increasing the silicon to aluminum ratio in the catalytic system reduces the sorption properties of the catalytic system in relation to methanol. The FTIR analysis performed for the catalysts supported on hydrogen form of the HZSM-5(50) zeolite showed similar bands on the IR spectrum as for the corresponding catalyst impregnated directly onto the ammonium form of the ZSM-5 zeolite. While the IR bands located in the spectrum of 10% MgO/HZSM-5(280) are much more intense than those recorded for 10% MgO/ZSM-5(280) catalyst. Whereas the FTIR measurements performed for 10% MgO/HZSM-5(50), 10% MgO/HZSM-5(280), 10% MgO/ZSM-5(23) and 10% MgO/ZSM-5(50) catalysts confirmed the occurrence of additional weak IR bands located in the range 1927-1991 cm −1 on the FTIR spectrum which are corresponding to a shift of the Si-O combination bands of the adsorbent. In addition, the spectra recorded for all catalysts, with the exception of the 10% MgO/ZSM-5(23) catalyst, show strong bands located at the value of 1576-1589 cm −1 , which are characterized by deformation vibrations occurring in -OH species [36]. The negative band above 1600 cm −1 that appears on the FTIR spectra is most likely due to the removal of initially adsorbed water (δ(OH)) [39]. The intense negative bands seen in the wavenumber range of 1200-1331 cm −1 can be attributed to the δ(OH) modes of H-bonded zeolite hydroxyls [39].
Furthermore, FTIR measurements carried out for all investigated catalysts allow the conclusion that the greatest quantity of adsorbed methanol was observed for 10% MgO/ZSM-5(23) catalyst calcined in an air atmosphere for 4 h at 500 • C. The obtained FTIR spectra recorded for all studied systems indicate that the quantity of methoxide species adsorbed on the catalyst surface is one of the first steps of the transesterification process over the catalyst surface and plays an important role during the investigated reaction using the synthesized catalyst systems. Correlating the obtained experimental results with the catalytic activity for the tested systems, it can be concluded that for catalysts impregnated on various types of NH 4 -ZSM-5 zeolites, the adsorption of methanol depends on the SiO 2 /Al 2 O 3 ratio, and weaker adsorption improves the FAME yield. . FTIR difference spectra of adsorbed species taken after exposure of the catalytic system (10% MgO/Al 2 O 3 , 10% MgO/SiO 2 , 10% MgO/ZSM-5 and 10% MgO/HZSM-5) calcined in an air atmosphere at 500 • C for 4 h, to a mixture of 6 vol% methanol-94% argon mixture at 50 • C and after the evacuation of this mixture for 10 min using argon stream. The ratio of silicon to aluminum is given in parentheses.

Preparation of the Catalytic Materials
Magnesium oxide catalysts were synthesized by impregnation method using zeolite ZSM-5 as a support. The ammonium form NH 4 -ZSM-5 zeolite with different silicon to aluminum ratios of 23, 50, and 280, respectively, was used in this study and all zeolites were purchased from Zeolyst International (Kansas City, KS, USA). A part of the ZSM-5 zeolites with silicon to aluminum ratios of 50 and 280, respectively, was converted to the hydrogen form (HZSM-5) by calcination for 15 h in an air atmosphere at 500 • C. Mg(NO 3 ) 2 ·6 H 2 O (Sigma Aldrich, ACS reagent, St. Louis, MO, USA) was used as the precursor of the active phase component. The support was impregnated with an aqueous precursor salt, and the impregnation time was 24 h. The amount of aqueous precursor salt solution used during the impregnation step for each catalyst was 150 mL. The loading of the MgO catalysts was 10% wt. [40]. Magnesium oxide catalysts supported on commercial oxide supportsalumina or silica-were also synthesized for comparison purposes. All catalysts were evaporated on a rotary evaporator and calcined in an air atmosphere for 4 h at 500 • C.

Characterization of the Catalytic Material
The phase composition studies of supports and catalysts were investigated using the Xray diffraction technique (XRD). X-ray diffraction patterns were recorded on a PANalytical X'PertPro MPD diffractometer in Bragg-Brentano reflecting geometry (Malvern Panalytical Ltd., Malvern, UK). Cu Kα radiation (k = 154.05 pm) from a sealed tube was used in the 2Θ angle range 5-90 • . The TPD-CO 2 measurements were performed in a quartz microreactor using CO 2 as a probe molecule. The CO 2 was adsorbed on the catalyst surface at 50 • C for 30 min after drying in flowing helium at 500 • C for 60 min. TPD-CO 2 measurements were carried out in the temperature range 100-600 • C, after removing physisorbed CO 2 from the catalyst surface. FTIR measurements were done using a Nicolet iS50 FT-IR spectrometer (Thermo Fisher Scientific Co., Waltham, MA, USA) equipped with a liquid-nitrogen cooled MCT detector. A resolution of 4.0 cm −1 was used throughout the investigation. A total of 64 scans were taken to achieve a satisfactory signal-to-noise ratio. The background spectrum was collected at 50 • C in an argon stream. Before the measurements, argon was shifted to a mixture of 6 vol % CH 3 OH in the argon stream. The adsorption process involved exposure of the heated catalysts to 6 vol % CH 3 OH in an argon stream flowing at 40 cm 3 /min for 10 min under atmospheric pressure. After the adsorption process, the cell was evacuated for 15 min in an argon stream, and then the spectrum was collected. FTIR spectra of adsorbed species were taken after exposure of the catalysts (10% MgO/Al 2 O 3 , 10% MgO/ZSM-5 (Si/Al = 23), 10% MgO/ZSM-5 (Si/Al = 50), 10% MgO/ZSM-5 (Si/Al = 280), 10% MgO/HZSM-5 (Si/Al = 50), 10% MgO/HZSM-5 (Si/Al = 280), 10% MgO/SiO 2 ) calcined in an air atmosphere at 500 • C for 4 h, to a mixture of 6 vol % methanol-94% argon mixture at 50 • C and after the evacuation of this mixture for 15 min using an argon stream. The collected FTIR spectra were recorded in transmittance mode, which were converted to absorbance mode and subtracted from the previously collected background of the bare catalyst. Magnesium concentration in the prepared catalysts was determined by flame atomic absorption spectrometry (acetylene/air) at a wavelength of 285.2 nm on a SOLAAR M6 (Unicam Atomic Absorption, Cambridge, UK) spectrometer equipped with an HCL hollow cathode lamp for magnesium determination and a nebulizer with a platinum-iridium capillary, allowing sample aspiration at a rate of 6 mL/min. The spectral width of the slit was 0.5 nm. The gas flow rate was 0.9 L/min and the height of the burner was 14.2 mm. Approximately 0.1 g of the analyzed 10% MgO/zeolite catalyst sample was weighed on an analytical balance into disposable Teflon vessels, followed by the addition of 2 mL of 65% HNO 3 and 4 mL of HF. Then, the samples were pre-mineralized at 25 • C for 25 min in an ultrasonic bath (BANDELIN, Berlin, Germany). Next, the samples were mineralized in a closed UltraWave microwave system (Milestone, Italy) according to the procedure shown in Table 6. The dissolved samples were transferred to 25 mL volumetric flasks and filled with deionized water.

Catalytic Activity Measurements in Transesterification of the Vegetable Oil with Methanol
The transesterification process was carried out in an autoclave using a substrate mixture of rapeseed oil and CH 3 OH with a molar ratio of 9:1. In this reaction, commercial rapeseed oil was used. Methanol used in this reaction was purchased from CHEMPUR, PiekaryŚląskie, Poland. In all catalytic tests, a weight of about 0.5 g was used. The reaction was performed at 220 • C for 2 h. Before reaction tests, catalysts were calcined at 500 • C for 4 h. The obtained reaction products were analyzed by the HPLC technique (Shimadzu, Kyoto, Kyoto Prefecture, Japan). In all catalytic tests, column C-18 (5 µm, 4.6 × 250) and a mixture of 2-isopropanol-hexane (4:5) and methanol was used as eluent, all chemicals were purchased from CHEMPUR, PiekaryŚląskie, Poland. The mobile phase gradient applied during each experiment was shown in Table 7. The products of the reaction were analyzed using a DAD detector (wavelength: λ = 205 nm) (Shimadzu, Kyoto, Kyoto Prefecture, Japan) to define the triglycerides conversion and to identify of FAME yield. Table 7. Phase gradient used in the HPLC measurements.

Conclusions
In summary, magnesium oxide catalysts supported on ZSM-5 zeolites and oxide supports were prepared by an impregnation method and all materials were tested in biodiesel production reaction in order to select the best catalytic system. The physicochemical properties of the catalysts were also correlated with the reactivity results in a biodiesel production process. The physicochemical properties were tested by BET, CO 2 -TPD, FTIR and XRD techniques. The obtained activity results indicate that the reactivity of the synthesized catalytic systems in the transesterification reaction of rapeseed oil is influenced by the silicon to aluminum ratio for the support. Magnesium oxide catalyst supported on ZSM-5 zeolite with higher silicon to aluminum ratio gave higher triglyceride conversions and higher yields of FAME. The catalysts directly impregnated on the ammonium form of the NH 4 -ZSM-5 zeolite exhibited higher TG conversion and yield to FAME in the transesterification reaction of rapeseed oil with methanol than the catalysts impregnated onto the hydrogen form of the HZSM-5 zeolite obtained by long-term heating. The most active catalyst system in the transesterification reaction was the 10% MgO/ZSM-5 (Si/Al = 280) catalyst, which showed the highest value of higher fatty acid methyl esters (94.6%) and the second-highest yield of triglyceride conversion (92.9%). Its high activity is explained by the high specific surface area, alkalinity and sorption properties in relation to methanol. The obtained activity results in the transesterification reaction clearly confirmed the potential application of MgO catalytic systems supported on zeolite as active materials for biodiesel production.