Optimization of biodiesel production via transesterification of soybean oil using α-MoO₃ catalyst obtained by the combustion method

1 Introduction

Biodiesel is a strong candidate to replace petroleum diesel, since it is one of the green fuels widely studied around the world, in order to reduce the environmental impacts caused by burning fossil fuels, under the aspect of greenhouse gas emissions such as COx, NOx, SOx, CxHy, particulates and other organic compounds that cause global warming (Gonçalves et al., 2021; Santos et al., 2019; Silva et al., 2020a; Silva et al., 2020b). It is a biofuel chemically composed of fatty acids monoalkyl esters, which can be obtained by transesterification and/or esterification of animal or vegetable oils and fats, as well as their rejects in the presence of a catalyst (Santos et al., 2019). In biodiesel production, heterogeneous catalysis may be very important in order to optimize the production process giving greater economic viability and thus making it more sustainable (Xie and Wang, 2020). In this context, research has focused on the use of solid catalysts in order to minimize some of the problems faced by classical homogeneous catalysts (Xie and Wang, 2020; Santamaria et al., 2021; Xie and Wan, 2019).

Among the solid heterogeneous catalysts reported in the literature, MoO₃-based oxides have drawn the attention of the scientific community due to their wide application (Xie and Wang, 2020; He et al., 2018; Gadamsetti et al., 2018; Sangeetha et al., 2019), with prominence in catalysis for biodiesel production because it has Lewis and Brønsted-Lowry acid sites that can lead to high conversions (Santos et al., 2019; Xie and Wan, 2019). Recent work describes the use of MoO₃ applied in biodiesel production, generally being used as an active phase supported on other materials: MoO₃/SrFe₂O₃ (Gonçalves et al., 2021), MoO₃/Al₂O₃ (Navajas et al., 2020), MoO₃/CeO₂ (Shadidi et al., 2020), MoO₃/ZrO₂/ KIT-6 (Wang et al., 2022) and others (Xie and Wan, 2019; Chandar et al., 2019; Maniruzzaman et al., 2014; Ramos et al., 2019). However, its use as a mass active catalyst (unsupported), is poorly reported, being found only the recent works of the researchers Pinto et al. (Pinto et al., 2019) and Silva et al. (Silva et al., 2022), which report the application of MoO₃ as a mass catalyst in biodiesel production.

Regarding the synthesis methods of MoO₃, it has been obtained by several techniques: hydrothermal (Nagyne-Kovacs et al., 2020; Sen et al., 2019), microwave (Sangeetha et al., 2019), aqueous solution (Song et al., 2017), wet chemistry (Chandar et al., 2019), solid-state decomposition (Muthamizh et al., 2020), pyrolysis (Baez-Rodriguez et al., 2019), electrochemical pyrolysis (Dighore et al., 2016), sol–gel (Al-Alotaibi et al., 2021); Pechini (Pinto et al., 2019) and the combustion reaction (Silva et al., 2022; Shahsank et al., 2021), even though it is rarely reported in the literature among the existing techniques, is considered a method that has advantages. The combustion reaction is classified as a relatively simple, fast, easy, effective and low-cost method, besides being consolidated in the literature as a procedure used in the preparation of several oxides with application in biodiesel production (Silva et al., 2020a; Silva et al., 2020b; Farias et al., 2020).

In this context, obtaining and applying molybdenum trioxide-based catalysts that meet the premises of economic and environmental benefits, as already mentioned (Song et al., 2017; Muthamizh et al., 2020), the present study highlights as its objective the use of experimental planning techniques, which are widely used as optimization strategies for industrial processes, as they allow an insight into the influence of multivariable involved in industrial and research productions and experiments (Navajas et al., 2020; Farias et al., 2020; Ali et al., 2015; Mushtaq et al., 2021). The experimental planning presents itself as a valuable resource to simulate real situations of the industrial day-to-day life, enabling the projection of a significant reduction of costs and energy expenditures (Gonçalves et al., 2021; Silva et al., 2020b).

The optimization of biodiesel synthesis conditions, is a technique widely applied and reported in the literature (Silva et al., 2020a; Silva et al., 2020b; Farias et al., 2020), as reported by Gonçalves et al. (2021), who evaluated the reaction conditions of the catalytic application of MoO₃ supported on SrFe₂O₃ in the production of biodiesel from waste oil, but the authors studied severe synthetic variables and obtained ∼95.4% methyl ester conversions in 4 h of reaction, alcohol:oil molar ratio of 40:1, 10% catalyst dosage.

In this context, the objective of this work was to use a factorial design composed of central and axial points to optimize the reaction conditions and evaluate the catalytic capacity of an acid catalyst (MoO₃ obtained by pilot scale combustion reaction) in the ethyl transesterification of soybean oil.

2 Materials and methods

2.2

2.2 Methods

For the Synthesis of the MoO₃-based catalyst, by combustion reaction, high purity reagents were used with reaction initial solution composition was based on the total valence of the oxidizing and reducing reagents using concepts from the chemistry of propellants and explosives (Jain et al., 1981). In this context, ammonium heptamolybdate (NH₄)₆Mo₇O₂₄ 4H₂O) was used as a metal precursor to obtain MoO₃, being the molybdenum (Mo) and oxygen (O) were considered as oxidizing elements and urea (CO(NH₂)₂) was considered a reducing agent (fuel).

The valences of the reactive elements are C =+4; H=+1; N = 0, O = −2 and Mo = +6 (Jain et al., 1981). For a maximum energy release, the stoichiometric coefficient of the composition of the reaction mixture was considered (Φe = 1), where the entire all the oxygen content of the metal precursor must oxidize, leading to an equilibrium of zero oxygen (OB = 0) (Hwang et al., 2005).

The mixtures of precursor reagents were carried out by magnetic stirring in a stainless steel container that makes up the conical reactor belonging to the pilot plant (Costa and Kiminami, 2012), coded P08 with a production capacity of 20 g. Then the pH was controlled (2–5) and stirring was maintained for about 20 min, being later taken to the combustion system in the pilot plant, Fig. 1 (a). The synthesis of MoO₃ by combustion reaction performed on different days and times and the calculated for obtain 10 g of product. Based on chemical reaction of theoretically balanced MoO₃ synthesis, expressed by Eq. (1):

(1)

2[(NH₄)₆Mo₇O₂₄](aq) + 4H₂O(aq) + 6[CO(NH₂)₂](aq) → 14MoO₃(s) + 18NH₄(g) + 4H₂O(g) + 6CO₂(g) + 3 N₂(g)

Close

Fig. 1

(a) Combustion reaction pilot plant; (b) Heating system used in the combustion reaction Synthesis of MoO₃.

Export to PPT

So, considering a 1:1 stoichiometry (1 mol of Mo corresponds to 1 mol of MoO₃), the Synthesis requires an equivalent amount of “n” moles of fuel that the synthesis of 10 g requires for Φe = 1 was calculated, therefore: 0.005[7x(-6) + 24x(-2) + nx(+6)] = 0, n = 0.03/6 = 0.005 mol of urea).

For the synthesis in the pilot plant (Fig. 1a), two heating systems were used, differentiated by the reactor heating source (adapted System I - composed of a belt-shaped heating element attached to the stainless steel container and superimposed on a spiral heating element; and the patented System II - composed of a heating mantle on the side and at the base, to which the stainless steel container is attached). Two circular lateral resistances were added to both systems, with the purpose of keeping the temperature free from external interference, as illustrated in Fig. 1(b).

Five reactions were performed, distributed according to the use of the heating systems: Reactions (1) and (3) using System I, and reactions (2), 4, and 5 using System II. Subsequently, the synthesized samples were mixed using a pitcher mil (ACB - LABOR), and the sample designated as α-MoO₃, sieved in ABNT 200 mesh (44 μm) and then characterized structurally, morphologically and tested catalytically.

3 Results

Fig. 2 is illustrated the time and temperature measurements results of the five Synthesis performed to obtain the α-MoO₃ catalyst during the combustion reaction process. In which it is possible to observe that the Synthesis is possible to identify three distinct stages of combustion.

Close

Fig. 2

Behavior (t × T) of combustion reaction measured during the α-MoO₃ Synthesis.

Export to PPT

However, it is possible to observe that the Synthesis 1, and 3, presented similar combustion stages behavior to each other. Stage 1 is characterized by the temperature rise and the moisture evaporation followed by reactants liquefaction. In stage 2, moderate gas release was observed, followed by ignition of the combustion reagents (between 15 ∼ 45 min). At this stage, the threshold reached at maximum temperatures was instantaneous of 672 °C and 667 °C, respectively. After this stage, a temperature decay of the combustion system is observed, which corresponds to stage 3 (Fig. 2).

Distinct stages of combustion were obtained for the Synthesis 2, 4 and 5, in relation to the Synthesis 1 and 3, because the first stage is marked by a temperature rise moderate with consequent moisture evaporation and gas release. In stage 2, the maximum temperature level reached was continuous (from 41 min to about 82 min), reaching at ∼608 °C and soon after, is observed the temperature decay in stage 3.

This difference in the stages profiles and maximum temperature combustion observed in the Synthesis conducted using the systems I and II, is an indication that in system 1 there was a greater and faster heat generation than in system II, however, consequently, energy loses at same speed. At the same time, in system II there was less energy loss to the environment, keeping the heat generated for a longer time.

It was also possible to notice that since there is a heat transfer by conduction, convection and radiation to the reaction medium, considering the losses to the environment, the maximum combustion temperatures in the Synthesis are always lower than in the combustion system temperatures (Bergman et al., 2000). The reactions behavior observed is typical of combustion reaction, because it is in consonance with the behavior reported by Dantas et al. (2021), when preparing mixed iron oxides by this method. Table 2 describes the parameters: pH, total time and maximum temperature of the reaction, and the mass yield of α-MoO₃ obtained in each synthesis.

The acid average pH of the precursor solutions to the Synthesis was 3.4, used for favored for the solubilization of precursors. The average reaction time was 88 min, the average mass yield of product from the Synthesis was 9.06 g (90.60%) and the average combustion temperature was 609 °C. In summary, the reaction process for obtaining the α-MoO₃ catalyst was cost-effective, fast and reproducible, since it showed a considerable yield.

In Fig. 3, are illustrated the x-ray diffractograms of the catalyst MoO₃ obtained in the five combustion reaction Synthesis performed.

Close

Fig. 3

X-ray diffraction of the five Synthesis by combustion to obtain the catalyst based on MoO₃.

Export to PPT

According to the diffractograms (Fig. 3) it is possible to observe that the products formed in all Synthesis present similar characteristics of orthorhombic crystalline system of MoO3 with ingle phase of high crystallinity, identified by the pattern card PDF2(2019) 00-005-0508. This results also indicate that the proposed catalyst presented good reproducibility, since there was no significant variation in the structural characteristics of the catalyst obtained in different Synthesis systems. With these identifications, the five different Synthesis products were mixed and the obtained sample were named as α-MoO₃, (MoO₃ with Orthorhombic Crystal System) with the intention of simplifying the discussion. The X-ray mixture product diffractogram is shown in Fig. 4, as well as the respective diffraction planes for the α-MoO3 catalyst sample obtained by combustion reaction.

Close

Fig. 4

X-ray diffractogram and pattern card PDF00-005-0508 of the α-MoO₃ catalyst.

Export to PPT

The x-ray diffraction result of the MoO₃ Synthesis mixture (Fig. 4) confirms what has already been observed in Fig. 3, where the single-phase α-MoO₃ catalyst, was obtained with high purity and with orthorhombic configuration identified by pattern PDF2(2019) 00-005-0508. In addition, it has crystallinity and crystallite size of 90.30% and 80.76 nm, respectively.

The single-phase product obtained in this work stands out in comparison with the recent research of Al-Alotaibi et al. (2021), who in their studies obtained MoO₃ by different Synthesis methods (sol–gel and hydrothermal), using expensive reagents and with calcination after the Synthesis process, at a time of up to 12 h with temperatures ∼600 °C. Therefore, the combustion reaction method presented in this paper and in literature recently published by our group (Silva et al., 2022), confirms that the preparation of MoO₃-based catalysts by combustion reaction is promising and economical, since it is a fast process that allows to obtain a highly crystalline and pure single-phase product with proven reproducibility.

Fig. 5 illustrates the thermal events observed in the TGA/DTGA/DTA curves for the catalyst (α-MoO₃), from which the decomposition temperatures (°C) and mass losses (%) could be determined.

Close

Fig. 5

Overlapped TGA/DTA/Derivate TGA curves for the catalyst (α-MoO₃).

Export to PPT

Thermal analysis (TGA) reveals the thermal stability of the catalyst, observed up to temperature 725 °C. Thereafter, the degradation process starts through two mass loss events in the temperature range between 725 and 881 °C, with a total mass loss of 89.68%. The DTA curve for the α-MoO₃ catalyst, in this temperature range, shows a discrete endothermic peak around 788 °C, which is related to the orthorhombic crystalline phase melting point (α-MoO₃). The results commented on in this work are in according with the literature (Al-Alotaibi et al., 2021; Hou et al., 2018).

The oxides percentages present given by X-ray fluorescence (EDX) for the α-MoO₃ catalyst were (MoO₃) 99.57% and (Fe₂O₃) 0.43%. The EDX results confirm that molybdenum oxide is the majority element present in the catalyst obtained by combustion reaction, which corresponded to about 99% of the total, followed by discrete traces of iron oxide that accounted for about 0.43% possibly resulting from the preparation process by combustion reaction. The values expressed in this work corroborate with the studies of Sales et al. (2020), when they studied photocatalysts of MoO₃ obtained by a Pechini-based method and applied to effluent treatment.

The particle size analysis is graphically illustrated in Fig. 6, which express the distribution values of the equivalent particles spherical diameters as a function of the cumulative volume for the catalyst (α-MoO₃) obtained by combustion reaction.

Close

Fig. 6

Particle size distribution for the α-MoO₃ catalyst.

Export to PPT

Analyzing Fig. 6, it can be seen that the α-MoO₃ catalyst exhibits a polymodal particle size distribution curve with a wide range of size distributions. A concentration of particles between ∼0.2 and 100 μm can be observed, and an average particle diameter of 15.43 μm is obtained. The accumulated values (black curve) illustrate a particle size accumulation of 1.07 μm up to 10%, 6.63 μm up to 50%, and 42.92 μm up to 90%. The particle size values observed in this work differ from recent published literature (Shahsank et al., 2021; Prakash et al., 2018; Rammal and Omanovic, 2020), which report a nanometer scale for powders obtained by similar methods, since in the present study the integrated synthesis method provided the formation of single phase MoO₃ during the process and with micrometric characteristics.

Fig. 7 illustrates the scanning electron microscopy results for the α-MoO₃ catalyst, synthesized by combustion reaction.

Close

Fig. 7

Morphologies obtained by SEM for the α-MoO₃ catalyst, (a) 1000x; (b) 5000x.

Export to PPT

Looking at Fig. 7, the micrographs illustrate the morphological with well-defined appearance, oriented plates with directional growth preferences, what is corroborated by X-ray diffraction analysis, which showed that the catalyst is single phase with crystalline structure orthorhombic (α-MoO₃) (Rammal and Omanovic, 2020); and also having a wide distribution of sizes, as already observed by the analysis of granulometric distribution (Fig. 5). The morphological aspects of plaques obtained in this work are in agreement with the studies of Sales et al. (2020)), when they studied molybdenum trioxide with orthorhombic phase (α-MoO₃) synthesized by the Pechini method applied to photocatalysis and the research of Gangaraju et al. (2017), when they synthesized (α-MoO₃) by combustion reaction and applied to batteries.

The catalyst presented a value of 1.36 m²g⁻¹ for the surface area, a relatively low value, characteristic of MoO₃-based materials (Pinto et al., 2019), such behavior suggests that regardless of the Synthesis method, catalytic property and relatively low surface areas are an intrinsic characteristic of this material (Sales et al., 2020).

The density for the α-MoO₃ catalyst evidenced by the He pycnometry test was 4.5 g/cm³, a value that showed little deviation from the theoretical density revealed by the standard PDF2(2019) 00–005-0508, which is 4.71 g/cm³, whose calculated relative error was 3.6%, thus illustrating the good proximity of the material produced in this work to the theoretical values. Concepts such as density are important, as the measurements allow defining the mass of solid catalyst that will be used in an industrial reactor (Dantas et al., 2021).

In the Fig. 8 presents the acidity data for the α-MoO₃ catalyst, obtained by thermo-programmed desorption (TPD-NH₃).

Close

Fig. 8

TPD-NH₃ analysis of the α-MoO₃ catalyst α-MoO₃.

Export to PPT

The parameters for acid characterization of the catalyst (α-MoO₃) were obtained based on the literature (Silva et al., 2020b; Lima and Perez-Lopez, 2018; Dantas et al., 2017) and are shown in the Table 3. They indicate that NH₃ thermodesorption peaks observed in regions at temperatures below 300 °C correspond to acid sites of weak nature, peaks between 300 and 400 °C correspond to acid sites of moderate nature, and peaks observed at temperatures above 400 °C are attributed to acid sites of strong nature.

According to Table 3, the α-MoO₃ catalyst presented six peaks of thermodesorption of NH₃ characteristics of the three types of acid strength (weak, moderate and strong), with predominance of moderate and strong acid sites. However, the total acidity of the catalyst was 33.61 µmol.g⁻¹ of NH₃. The total acidity values presented in this work, are higher when compared to the work of Pinto et al. (2019), this can possibly be explained by the different synthesis method for MoO₃.

In this case, considering the surface area obtained in analysis (1.36 m₂g⁻¹) for α-MoO₃ catalyst, there is a distribution of acid sites is 24.71 µ.mol m². So, it can be concluded that the catalyst, obtained by pilot scale combustion reaction, exhibits acid sites, which can satisfactorily con-tribute to biodiesel production reactions.

3.1

3.1 Performance of catalyst α-MoO₃

Table 4 describes the planning response used to analyze the statistical data for biodiesel production.

Table 4 Planning response 2² obtained for the random variables temperature and catalyst quantity.

Experiment	(A) Temperature (°C)	(B) Catalyst (%)	Y = Average Conversion in Ester (%)
1	100	2	18.8
2	100	4	20.3
3	200	2	96.8
4	200	4	96.6
5	79.3	3	6.5
6	220.7	3	96.9
7	150	1.6	78.6
8	150	4.4	74.4
9 (C)*	150	3	80.8
10 (C)*	150	3	78.2
11 (C)*	150	3	80.0
12 (C)*	150	3	73.1

*(C) Central point, * Fixed conditions: 30 g oil mass, time 1 h, and alcohol/oil ratio (15:1).

In general, it could be observed (Table 4) that the α-MoO₃ catalyst was active and conversions into fatty acid esters ranging from 73 to 96% where reached when the temperatures used are equal to or greater than 150 °C. All ethyl ester conversion results obtained from the experimental design are also illustrated in Fig. 9.

Close

Fig. 9

Soybean oil conversion percentage results to ethyl esters obtained in the presence of the α-MoO₃ catalyst.

Export to PPT

The best catalytic activities were observed for experiments 3, 4 and 6, with emphasis on the axial point of the design (Experiment 6) with 96.9% conversion to esters when used at a temperature of 220.7 °C, amount of catalyst 3% and the fixed conditions 30 g oil mass, 1 h and alcohol/oil ratio 15:1.

It is important to emphasize that the conditions used in this work were mild compared to recent literature (Pinto et al., 2019; Xie and Zhao, 2014; Mohebbi et al., 2020). Notably, Pinto et al. (2019) used MoO₃ by mass as the active phase for biodiesel production; however, the authors also use harsh reaction conditions such as molar ratio of 90:1, 5% of catalyst and up to 8 h of reaction, obtaining conversions of up to 95% of soybean oil methyl esters.

It is also important to highlight that this work used the ethyl route, which is beneficial to the process since the alcohol used is obtained from a renewable source, such as sugar cane (Shikida and Bacha, 2019; Alves et al., 2021). The performance of the α-MoO₃ catalyst may be associated with the presence of strong acid sites (Pinto et al., 2019) (see Table 3), which gives the system great versatility, and high conversions into biodiesel. Furthermore, in previous studies the reuse of similar catalysts in the frying oil was tested, and the α-MoO₃ catalyst maintained catalytic activity for five reuse cycles, confirming the high potential of the catalyst developed for industrial applications (Silva et al., 2022). Considering the best ester conversions (Fig. 10 and Table 4), the acidity and kinematic viscosity indexes of biodiesel were obtained and the data are presented in Table 5.

Close

Fig. 10

Pareto chart resulting from the central composite factorial design 2² for the conversion of soybean oil into biodiesel.

Export to PPT

Table 5 Acid value index and kinematic viscosity of the products of experiments 3, 4 and 6 related to the production of biodiesel.

Experiment	Acidity Index (mgKOH/g)	Kinematic viscosity at 40 °C (mm2/s)
3	0.50 ± 0.15	5.94 ± 0.06
4	0.50 ± 0.01	5.88 ± 0.01
6	0.48 ± 0.01	5.84 ± 0.00

Based on the data presented in Table 5 and with the results of conversions of biodiesels obtained from experiments 3, 4 and 6 (Fig. 9), it is possible to observe that are in accordance with the biodiesel quality parameters of the Brazilian regulatory standards (ANP N° 51 DE 25/11/2015) (ANP, 2008) and European (EN 14214) (EN - 14214, 2003), in which they are established the minimum ester contents of 96.5% (quantified by gas chromatography) and kinematic viscosity (at 40 °C) of 3.0–6.0 mm²/s in addition to the maximum acidity index of 0.50 mg KOH/g.

3.2

3.2 Statistical analysis

The Pareto graph (Fig. 10) was initially used for the statistical analysis of the experimental design response for the optimization of the conditions of the transesterification of soybean oil catalyzed by α-MoO₃.

It is possible to observe in Fig. 10, that among the variables, the most significant, both linear and quadratic, was the temperature with 95% reliability (p < 0.05), with a greater interference of the temperature linear variable for larger values, i.e., the positive level, in contrast to the quadratic behavior of the same variable, which had a negative influence. The catalyst quantity variable as well as the interactions between the variables showed no statistically significant effect. These observations corroborate with the data in Table 3 and Fig. 8, suggesting that the increase in reaction temperature, significantly increases the soybean oil conversion to biodiesel.

Level curves of the experimental design response were also obtained, using the independent variables: amount of catalyst (%) and temperature (°C) and is ilustrated in Fig. 11.

Close

Fig. 11

Level curve for the soybean oil conversion to biodiesel with the interaction between (B) amount of catalyst (%) and (A) temperature (°C).

Export to PPT

The effect of reaction temperature under the conversion of ethyl esters was best evaluated and for high levels of temperature (+1) (220 °C) and amount of catalyst (4.4%), the biodiesel conversion was maximum (close to 100%), corroborating what was observed in the Pareto chart (Fig. 10).

However, it was also possible to observe that when the reaction temperatures are lowered (reaction temperature is below 150 °C) for both catalyst percentages, the conversion to esters also decreases, while still remaining below 40%. Therefore, it can be inferred that the temperature variable is the most statistically significant to the process and showing excellent catalytic behavior for the α-MoO₃ acid catalyst.

In the region investigated, the response surface is satisfactorily described by the quadratic mathematical model given by Equation (3), which presented an R² of 98% and which defines the plane represented in perspective on the contour line (Fig. 11), from the experimental planning performed and which best represents the data collected and adjusted to the data in Table 3.

(3)

$$Y\left(x,y\right)=-183.06973776469+2.5062039912853\times x-0.0058862500000001\times x^2+13.880963023967\times y-2.103125\times y^2-0.0115\times x\times y+0$$

The data presented in Figs. 10 and 11 and the quadratic equation (Equation (3)), indicate that the model represents a relatively good description of the experimental data related to the ethyl ester content at the reaction time of 1 h, alcohol-oil ratio of 1/15, 3% catalyst and temperature of 220 °C, conditions revealed as optimal by the statistical model adopted. The quadratic model fit was also tested by analysis of variance (ANOVA) according to Table 6.

Table 6 ANOVA for optimizing biodiesel production from soybean oil using an acid catalyst α-MoO₃.

Variation Source	Quadratic Sum	Degree of Freedom	Quadratic Average	FCalculed	Fcal/Ftab
Regression	11357.55	5	2271.51	69.40	15.82
Waste	196.38	6	32.73	–	–
Lack of Adjustment	160.74	3	53.58	4.51	0.49
Pure Error	35.64	3	11.88	–	–
Total	11553.93	11
Ftabulated Regression	4.39	6
Ftabulated Lack of Adjustment	9.28	9
% Mx. Explained	98.30	–
% Mx. Explainable	99.69	–
R2	0.98	–
Fit Quality	0.76	–
S (standard regression error)	5.72	–

The ANOVA results (Table 6), showed a value for Fcal / Ftab of 15.82 for the regression, indicating that the model was statistically significant, and the value for Fcal / Ftab of the fit lack of 0.49 indicating that the model was predictive. Therefore, the regression model given by Equation (2) was a reasonable predictor of the experimental results, and the affected factors were real at a 95% confidence level, as already noted in Fig. 11.

4 Conclusions

The α-MoO₃ catalyst was successfully synthesized on a pilot scale using a combustion reaction. The pilot-scale production was safe, reproducible, and efficient. The synthesized catalyst is single-phase (α-MoO₃ orthorhombic), micrometric with an average particle size of (15.43 µm) and surface area (S_BET = 1.36 m²g⁻¹). The use of central composite factorial planning made it possible to evaluate the process in a multivariate manner, leading to the identification of variables that significantly influenced the response variable (soybean oil conversion to esters). The factorial design allowed us to identify the influence of the variables (catalyst percentage and temperature) in the soybean oil transesterification. According to the statistical study, the temperature was the variable that most affected the value of the response variable, and the statistical model adopted was significant and predictive with a significance level of 95%. The catalyst was effective when the temperature was ∼150 °C, showing ester conversions up to 96%, the acidity of the biodiesel produced ranged from 0.48 to 0.50 mg KOH/g and kinematic viscosity of 5.84–5.94 mm²/s, with significantly promising results using ethanol as transesterification agent. From the results obtained, it can be concluded that the catalyst studied can be successfully applied in biodiesel production, since from its fast and easy preparation to the high conversions achieved outperform traditional methods, since this catalyst is a new material with innovative characteristics.