Boosting biodiesel production of waste frying oil using solid magnetic acid catalyst from agro-industrial waste

2.1 Materials

The RHA was acquired from a local rice production cooperative (Bélem, Pará, Brazil). Iron (III) chloride (FeCl₃, 99.8 %) and cobalt nitrate (Co(NO₃)₂·6H₂O, 99.8 %) were aquired from Sigma-Aldrich and ECIBRA, respectively. The methyl hetadecanoate (99.0 %, C₁₈H₃₆O₂) was purchased from Sigma-Aldrich, while nitric acid (37.0 % HNO₃), ethyl alcohol (99.8 %, C₂H₆O), and methyl alcohol (99.8 %, CH₃OH) were acquired from Vetec. Ammonium heptamolybdate ((NH₄)₆Mo₇O₂₄ 4⋅H₂O) and sodium hydroxide (97.0 %, NaOH) were aquired from Dinamica. In addition, heptane (99.0 %, C₇H₁₆) and hexane (99.0 %, C₆H₁₄) were purchased from Êxodo. All materials listed for the development of this study were used without prior treatment. The WFO used as feedstock for biodiesel production was purchased from a local market. The physicochemical properties of WFO are shown in Table 1.

2.2 Preparation of heterogeneous acid magnetic catalyst (MoO₃/RHA-CoFe₂O₄)

2.2.1

2.2.1 Synthesis of cobalt ferrite (CoFe₂O₄)

Cobalt ferrite was synthesized by coprecipitation method, according to the methodology adopted by Tamjidi et al. (2022), with adaptations. The synthesis process consisted of four steps, in which Step 1 was done by mixing in 100.0 mL of water from Co(NO₃)₂·6H₂O and FeCl₃ in a Fe:Co = 2 M ratio, followed by the addition of HNO₃ to pH = 3 and subsequent addition of NaOH (4.0 mol/L) to promote precipitation at pH = 12. Then, the system was kept under mechanical agitation and heating at 80.0 °C for 4.0 h. Step 2 consisted of the filtration procedure, while Step 3 was completed by washing the precipitated solid obtained until the washing water reached pH = 7, followed by drying the material in an oven at 80.0 °C for 12.0 h. Finally, in Step 4, the material was calcined at 800.0 °C for 3.0 h in order to obtain the CoFe₂O₄, as shown in Fig. 1(a).

Close

Fig. 1

Schematic diagram of (a) CoFe₂O₄ and (b) MoO₃/RHA-CoFe₂O₄ synthesis.

Export to PPT

2.3 Materials characterization techniques

The surface acidity of CoFe₂O₄, RHA, and MoO₃/RHA-CoFe₂O₄ materials were determined according to the methodology developed by Boehm (1994), with adaptations. In this method, 0.1 g of the sample was suspended in a solution of NaOH (0.1 mol/L) and stirred at 25.0 °C for 24.0 h. Then the sample is magnetically separated and the supernatant was titrated with a solution of HCl (0.1 mol/L), using phenolphthalein as an indicator.

The X-ray diffractograms (XRD) of MoO₃, CoFe₂O₄, RHA, and MoO₃/RHA-CoFe₂O₄ were obtained in a BRUKER diffractometer, model D8 ADVANCE. The scan interval was 20° < 2θ < 90°, with angular pitch of 0.02° and time per step of 0.01 s. In addition, the radiation used was Cu Kα (1.541874 Å) with 40 KV and 30 mA. The Fourier transform infrared spectroscopy (FTIR) analysis was performed in order to qualitatively investigate the functional groups present in the MoO ₃, CoFe₂O₄, RHA, and MoO₃/RHA-CoFe₂O₄ materials, using a BRUKER spectrometer, model Verterx 70 V, with analysis in the spectral range of 2000–400.0 cm⁻¹, 4.0 cm⁻¹ resolution and 32 sweeps.

The surface morphology of MoO₃, CoFe₂O₄, RHA, and MoO₃/RHA-CoFe₂O₄ was analyzed by scanning electron microscopy (SEM) using a Vega 3 LMU TESCAN microscope. The analysis of the surface elemental composition of MoO₃, CoFe₂O₄, RHA, and MoO₃/RHA-CoFe₂O₄ was performed by energy dispersion X-ray spectroscopy (EDS) using an Oxford Microanalysis system®, Aztec Energy X-Act model, with 129 eV resolution. Finally, the magnetic properties of CoFe₂O₄ and MoO₃/RHA-CoFe₂O₄ were evaluated by the technique of vibrating sample magnetometer (VSM), model Microsense magnetometer EZ9, with application of a magnetic field in the range of −20000.0 Oersted (Oe) to 20000.0 Oe, at room temperature.

2.4 Catalytic tests

In the catalytic tests, WFO, methanol, and the MoO₃/RHA-CoFe₂O₄ synthesized catalyst were used as reagents. The transesterification reactions were performed in a PARR 5000 series pressurized reactor, with constant stirring at 900 RPM. Initially, the influence of molybdenum content (10.0, 20.0, 30.0 and 40.0 %) in the final composition of the catalyst as a function of its performance in the synthesis of biodiesel was studied. Subsequently, after the selection of the catalyst with the best performance, the parameters of the transesterification reaction using the one variable at time (OVAT) technique in the following intervals: reaction temperature (130.0–170.0 °C), WFO:methanol molar ratio (10:1–50:1), catalyst amount (2.0–10.0 wt%), and reaction time (1.0–5.0 h) were studied. All ranges studied were defined in preliminary tests. In addition, the catalytic performance using only RHA and CoFe₂O₄ materials (blank reactions) in the reaction medium were also evaluated.

At the end of each catalytic test, the products obtained were separated from the catalyst magnetically and washed with 500.0 mL of distilled water at 80.0 °C to remove impurities such as unreacted methanol, glycerol residues, mono -, di -, and triacylglycerols. Then, the biodiesel sample was dried in an oven at 60.0 °C for 24.0 h and stored for subsequent determination of the ester content by gas chromatography (GC).

2.5 Determination of biodiesel properties

The biodiesels obtained in the catalytic tests were analyzed for their ester content by GC according to the methodology adapted from the European standard EN 14103 described by Dos Santos et al. (2022). The Varian gas chromatograph, model CP-3800, equipped with flame ionization detector (FID) and CP WAX 52 CB (30.0 m long, 0.32 mm diameter and 0.25 µm film) capillary column was used. Helium gas was used as a mobile phase with a flow of 1.0 mL min⁻¹ and initial oven temperature programming of 170.0 °C, with heating rate of 10.0 °C min⁻¹ up to 250.0 °C (same temperature as FID and injector). Heptane was used as a solvent, methyl heptadecanoate as internal standard and the injection volume was of 1.0 µL. The ester content (EC) was calculated according to Eq. (1):

2.6 Catalyst reuse study

At the end of the transesterification reaction, the MoO₃/RHA-CoFe₂O₄ catalyst recovered by application of the magnetic field, was washed three times with 30.0 mL of hexane and 20.0 mL of ethanol to remove impurities. Then, the catalyst was dried in an oven at 80.0 °C for a period of 12.0 h and re-employed in the catalytic tests in order to evaluate its catalytic stability in the reactional cycles.

3.1 Influence of molybdenum concentration

The influence of the active phase on the final composition of the catalyst is of paramount importance when studying supported catalysts, since it is aimed at obtaining a high performance catalytic solid from the use of the least amount of active phase impregnated in a support when compared with the use of mass catalysts, which weights on the final cost of the catalyst and the process as a whole (Mehrabadi et al., 2017). In this respect, the influence of the concentration of 10.0, 20.0, 30.0, and 40.0 % molybdenum (active phase) in relation to the ester content of biodiesels obtained in catalytic tests was studied, as shown in Fig. 2. The reactions were performed under the following non-optimized conditions: reaction temperature of 150.0 °C, WFO:methanol molar ratio of 30:1, catalyst amount of 6.0 %, and reaction time of 3.0 h.

Close

Fig. 2

Influence of molybdenium amount (reaction conditons: reaction temperature of 150 °C, molar ratio methanol:WFO of 30:1, catalyst amount of 6 wt% and reaction time of 3 h).

Export to PPT

The obtained values of ester content in biodiesels using catalysts containing 10.0, 20.0, 30.0, and 40.0 % molybdenum in their final compositions were 44.4 %, 79.4 %, 89.0 %, and 88.8 %, respectively. Accordingly, there is an increase in the ester content of biodiesels as the concentration of molybdenum is high in the catalyst composition. This occurs due to the increase of acidic sites on the surface of the catalysts, since it is at the surface that ion exchange occurs during the reaction process (Gonçalves et al., 2021b; Almeida et al., 2014). In addition, this behavior can be evidenced when comparing the activity of the 10-MoO₃/RHA-CoFe₂O₄ and 30-MoO₃/RHA-CoFe₂O₄ catalysts which led to the synthesis of biodiesels with ester contents of 44.4 % and 89.0 %, respectively, representing a gain of about 45.0 % of ester content in biodiesel when the catalyst 30-MoO₃/RHA-CoFe₂O₄ was employed.

In addition, Fig. 2 shows the determined surface acidity values for the different catalysts studied and highlights a proportional relationship between the increase in the concentration of molybdenum in the final composition of the catalysts and their determined surface acidity values. This can be observed, for example, when comparing the surface acidity values between the 10-MoO₃/RHA-CoFe₂O₄ (3.7 mmol H⁺ g⁻¹) and 40-MoO₃/RHA-CoFe₂O₄ (9.2 mmol H⁺ g⁻¹) catalysts, demonstrating a difference in surface acidity of magnitude 5.5 mmol H⁺ g⁻¹. In addition, it is observed that the surface acidity value of the catalyst 30-MoO₃/RHA-CoFe₂O₄ (8,7 mmol H⁺ g⁻¹) is relatively close to the value of the catalyst 40-MoO₃/RHA-CoFe₂O₄, which explains the fact that both catalysts lead to biodiesels with ester contents similar to 89.0 %. In addition, it should be noted that the surface acidity determined for the RHA support was 0.9 mmol H⁺ g⁻¹ and for CoFe₂O₄ was 1.2 mmol H⁺ g⁻¹. These values are considerably lower compared to those recorded for the studied catalysts, since the increase of the surface acidity in the catalysts is related to the Brønsted acid sites of MoO₃ impregnated in them (Pinto et al., 2019). Thus, the 30-MoO₃/RHA-CoFe₂O₄ catalyst was selected as the most promising for the development of this study due to its high catalytic performance in the biodiesel production process studied preliminarily.

3.2 Catalyst characterization of the catalyst 30-MoO₃/RHA-CoFe₂O₄

The X-ray diffractograms referring to MoO₃, CoFe₂O₄, RHA, and 30-MoO₃/RHA-CoFe₂O₄ are shown in Fig. 3. The diffractogram referring to MoO₃ (in green) shows the presence of peaks related to hexagonal (h-MoO₃) and orthorhombic (α-MoO₃) polymorphic structures from MoO₃. The peaks referring to the hexagonal structure are present in 36.4°, 41.4°, 46.6°, 61.9°, and 62.6°, while the peaks referring to the orthorhombic structure are present in 33.4°, 38.4°, 45.8°, 48.8°, 53.3°, 66.4°, 73.9°, and 78.5° (Pinto et al., 2019; Thangasamy et al., 2018).

Close

Fig. 3

XDR patterns of MoO₃, CoFe₂O₄, RHA and catalyst 30-MoO₃/RHA-CoFe₂O₄.

Export to PPT

The diffractogram concerning CoFe₂O₄ (in blue) showed the presence of peaks in 29.1°, 43.2°, 49.2°, 50.2°, 59.0,° and 60.3°, which confirms the spinel structure with cubic face of centered face of ferrite CoFe₂O₄ (Racik et al., 2022). In the diffractogram referring to the RHA support (in red), the predominance of the amorphous diffraction pattern is perceived, which is characteristic of amorphous silica, an abundant element in the composition of RHA (Conceição et al., 2017; Caliman et al., 2010). The diffractogram relative to the catalyst 30-MoO₃/RHA-CoFe₂O₄ (in black) shows the occurrence of peaks characteristic of both MoO₃ (33.2°, 36.2°, 38.3°, 46.5°, and 66.2°) and of CoFe₂O₄ (53.1° and 61.7°), which indicates the presence of these materials in the composition of the catalyst. In addition, it is assumed that the low intensity of the MoO₃ and CoFe₂O₄ peaks present in the catalyst can be explained by the high dispersion of the materials in the catalytic support.

The FTIR spectra of the MoO₃, CoFe₂O₄, RHA, and 30-MoO₃/RHA-CoFe₂O₄ materials are shown in Fig. 4. In the MoO₃-related FTIR spectrum (in green), there is the presence of bands at 806, 837, and 974 cm⁻¹, characteristics of Mo—o and O—Mo—O stretch bond vibrations (Ma et al., 2020). In the FTIR spectrum related to CoFe₂O₄ (in blue), there is the presence of the band at 613 cm⁻¹, suggesting the occurrence of Fe(Co)—O bonds in the octahedral and tetrahedral sites, characteristic of the spinel phase of ferrite (Asadi et al., 2022; Kumar et al., 2014). The FTIR spectrum referring to the RHA support (in red) shows the presence of bands at 445, 792, and 1074 cm⁻¹, characteristic of the Si—O—Si symmetric and asymmetric stretch bond vibration of the silica present in RHA (Atta et al., 2012; Helmiyati and Suci, 2019; Parrillo et al., 2021).

Close

Fig. 4

FTIR spectra of MoO₃, CoFe₂O₄, RHA and catalyst 30-MoO₃/RHA-CoFe₂O₄.

Export to PPT

In the FTIR spectrum for the 30-MoO₃/RHA-CoFe₂O₄ catalyst (in black), the vibrational bands identified at 445, 792, and 1074 cm⁻¹ are characteristic of the presence of silica, while the vibrational band at 613 cm⁻¹ is related to the spinel phase of ferrite. However, it is possible to observe, in a well attenuated way, the bands at 837 and 974 cm⁻¹ concerning the normal vibration modes of the Mo–O bonds present in MoO₃. This low intensity of the MoO₃ peaks present in the catalyst can be explained by the high dispersion of the active phase in the catalytic support, behavior previously observed in a similar way in the XRD analysis.

The morphologies of the MoO₃, CoFe₂O₄, RHA, and 30-MoO₃/RHA-CoFe₂O₄ materials are shown in Fig. 5. Fig. 5(a), (b), and (c) illustrate the micrographs referring to the MoO₃ magnifications of 5.0 kx, 10.0 kx, and 15.0 kx, respectively, and suggest that this material is formed by a cluster of rods with morphologies characteristic of orthorhombic structures of MoO₃, since this structure is predominant in MoO₃ synthesized at temperatures above 400.0 °C (Dhage et al., 2009), as in this study. In Fig. 5(d), (e) and (f), the micrographs referring to the RHA support are shown at magnifications of 200.0x, 1.0 kx, and 2.0 kx, respectively, and suggest the existence of a typical silica-cellulosic structure referring to the external epidermis, such as a corn cob. In addition, it is noteworthy that, in this study, RHA was used without prior treatment and the remaining organic material is due to incomplete pyrolysis of rice husk that generated the RHA (Parrillo et al., 2021; Liu et al., 2014). Fig. 5(g), (h), and (i) shows the micrographs referring to CoFe₂O₄ at magnifications of 5.0 kx, 10.0 kx, and 15.0 kx, respectively, and highlights the absence of a defined morphology, as reported by Jayalakshmi & Jeyanthi (2018). Fig. 5(j), (k), and (l) refers to the micrographs of the 30-MoO₃/RHA-CoFe₂O₄ catalyst at magnifications of 5.0 kx, 10.0 kx, and 15.0 kx, respectively, and indicates the presence of rods relative to the MoO₃ structures covering the catalyst surface, evidencing the efficiency of the impregnation process adopted in the study.

Close

Fig. 5

SEM images of (a) MoO₃ 5000x magnification, (b) MoO₃ 10000x magnification, (c) MoO₃ 15000x magnification, (d) RHA 200x magnification, (e) RHA 1000x magnification, (f) RHA 2000x magnification, (g) CoFe₂O₄ 5000x magnification, (h) CoFe₂O₄ 10000x magnification and (i) CoFe₂O₄ 15000x magnification, (j) 30-MoO₃/RHA-CoFe₂O₄ 5000x magnification, (k) 30-MoO₃/RHA-CoFe₂O₄ 10000x magnification and (l) 30-MoO₃/RHA-CoFe₂O₄ 15000x magnification,

Export to PPT

Fig. 6 shows the composition and elemental maps of the components present in the CoFe₂O₄ support and the 30-MoO₃/RHA-CoFe₂O₄ catalyst. From the CoFe₂O₄ EDS spectrum, show in Fig. 6(a), the existence of the elements Fe (25.12 wt%), Co (51.56 wt%) and O (23.31 wt%) in the CoFe₂O₄ structure surface is confirmed. The peaks located at approximately 0.5 and 6.5 Kev are related to the absorption energy dissipated by the iron, while the cobalt element shows peaks located at approximately 0.7, 7.0 and 7.6 Kev, which is the majority element because of the higher intensity of the peaks. In addition, it can be seen that oxygen showed only one high-intensity peak at 0.5 Kev, indicating that the components are in the form of oxides. Based on the percentage values of Fe and Co, the Co²⁺ to Fe³⁺ ratio is estimated at 1:2.05, extremely close to the 1:2 ratio (Co²⁺:Fe³⁺) used in the CoFe₂O₄ synthesis method. The elementary mapping of CoFe₂O₄ (Fig. 6(b)) shows a homogeneous and uniform distribution of the elements constituting the magnetic support, indicating the efficiency of the coprecipitation process employed.

Close

Fig. 6

EDS composition of (a) CoFe₂O₄, (b) elemental maps of each component of the CoFe₂O₄ (c) catalyst 30-MoO₃/RHA-CoFe₂O₄ and (d) elemental maps of each component of the catalyst 30-MoO₃/RHA-CoFe₂O₄.

Export to PPT

Fig. 6(c) shows the spectrum referring to the catalyst composition and indicates the content of 28.23 % molybdenum in its composition, a value close to the desired percentage of 30.0 %, which suggests the success of the impregnation method adopted in the study. In addition, the element with the highest percentage in the catalyst composition is oxygen with a content of 38.28 %. This data is corroborated by the fact that the catalyst is mostly composed of oxides. The percentages of cobalt (6.53 %) and iron (3.04 %) indicate the success of the desired Fe:Co = 2 ratio in the synthesis step. Finally, the presence of carbon at a content of 12.41 % can be attributed to the presence of C—C and C⚌O groups, due to the partial thermal decomposition of biomass.

Fig. 6(d) shows the elemental surface mapping of the catalyst as well as the other elemental maps of each component. From the analysis of the maps, it is realized that the elements molybdenum and oxygen are widely dispersed over the entire surface of the 30-MoO₃/RHA-CoFe₂O₄ catalyst. In addition, the maps show that the elements iron and cobalt also have wide dispersion and are related to each other in the same regions of the catalyst surface, evidencing the presence of CoFe₂O₄ in the composition of the catalyst.

The magnetic properties of CoFe₂O₄ and 30-MoO₃/RHA-CoFe₂O₄ were studied using the VSM technique, by obtaining M−H hysteresis curves (magnetization-applied (M) vs magnetic field (H)), as shown in Fig. 7. From the analysis of Fig. 7, it is realized the ferromagnetic behavior for CoFe₂O₄ and 30-MoO₃/RHA-CoFe₂O₄ with their saturation magnetization (Ms) values being 12.7 emu g⁻¹ and 7.0 emu g⁻¹, respectively. The decrease in Ms value of CoFe₂O₄ in comparison to 30-MoO₃/RHA-CoFe₂O₄ Ms value is due to the fact that the catalyst has only 30 % CoFe₂O₄ in its composition, in addition to the presence of other elements without magnetic properties, such as RHA and molybdenum. However, the Ms value obtained for the 30-MoO₃/RHA-CoFe₂O₄ catalyst is sufficient to promote its separation by applying a magnetic field, as well as is significantly higher in comparison to other catalytic magnetic solids applied to biodiesel production. In the study developed by Tang et al. (2012) the saturation magnetization of the Ca/Al/Fe₃O₄ magnetic catalyst applied to biodiesel production was 6.3 emu g⁻¹. Shi et al. (2017) reported Ms = 4.0 emu g⁻¹ for the CaO@ γ-Fe₂O₃ magnetic catalyst used in transesterification reaction. In another study, Wang et al. (2019) reported Ms = 3.7 emu g⁻¹ to the ZrFe-SA-SO₃ magnetic catalyst applied in oleic acid esterification for biodiesel synthesis.

Close

Fig. 7

VSM characterization forCoFe₂O₄ and 30-MoO₃/RHA-CoFe₂O₄.

Export to PPT

3.3 Catalytic performance

The study of optimization of transesterification reaction parameters using the catalyst 30-MoO₃/RHA-CoFe₂O₄, such as reaction temperature, methanol:WFO molar ratio, catalyst amount, and reaction time was performed using the OVAT methodology. The results obtained are shown in Fig. 8.

Close

Fig. 8

Effect of (a) reaction temperature, (b) molar ratio methanol:WFO, (c) catalyst amount and (d) reaction time on ester content.

Export to PPT

The optimization study of the reaction temperature, shown in Fig. 8 (a), was conducted in the range of 130.0 °C to 170.0 °C, keeping constant the variables methanol:WFO molar ratio, catalyst amount and reaction time in 30:1, 5.0 wt%, and 3.0 h, respectively. From the analysis of the results present in Fig. 8(a), it is possible to notice an increase in the ester content in the biodiesels obtained as the reaction temperature of the process increases, reaching the highest ester content value of 94.7 % for the biodiesel obtained in the reaction performed at 160.0 °C. This occurs due to the increase in temperature in the reaction process providing an increase in collisions of the reactant molecules, leading to an increase in the miscibility of the reaction medium, which in turn favors mass transfer, resulting in an increase in the ester content of the biodiesel produced (Ashok and Kennedy, 2019). However, when the catalyst is used in the reaction at 170.0 °C, the obtained biodiesel has an ester content of 87.5 %, indicating a decrease in ester content when compared to the synthesized biodiesel at 160.0 °C. This suggests a negative effect of the increase in temperature, indicating the reach of the reaction equilibrium in relation to the process temperature (Gardy et al., 2018). In this sense, the value of 160.0 °C was chosen as the optimized value for reaction temperature due to the fact that it provides the highest ester content value to the synthesized biodiesel.

The results of the optimization of the variable methanol:WFO molar ratio are shown in Fig. 8(b). The study was conducted in the range of 10:1 to 50:1, keeping constant the variables reaction temperature, catalyst amount, and reaction time at 160.0 °C, 6.0 wt%, and 3.0 h, respectively. The analysis of the results show that, initially, when processing the reactions by raising the methanol:WFO molar ratio values, biodiesels with high ester contents are obtained, for example biodiesel with ester content of 61.5 % obtained using the methanol:WFO molar ratio of 10:1, while the reaction using the methanol:WFO molar ratio of 30: 1 provided a biodiesel with ester content of 94.7 %. This behavior may be related to the excess of methanol used, leading to greater miscibility of the reaction medium, which in turn leads to an increase in the number of collisions between the reactant molecules (Foroutan et al., 2022; Pinto et al., 2019). However, it is observed that the use of methanol:WFO molar ratio above 30:1 in the reaction process leads to a decrease in the ester content values of biodiesels, such as those obtained using the molar ratios of 40:1 (88.4 %) and 50:1 (86.9 %). This is due to the fact that the high excess of methanol used causes saturation of the catalytic surface at a certain point, causing deactivation of the catalyst, decreasing the efficiency of the transesterification reaction (Aziz et al., 2017). In this sense, the methanol:WFO molar ratio of 20:1 was selected because the synthesized biodiesel reached an ester content of 94.6 %, a value relatively similar to biodiesel produced using the molar ratio of 30:1 (94.7 %).

The catalyst amount optimization study, shown in Fig. 8(c), was conducted in the range of 2.0 to 10.0 wt%, keeping the other variables at 160.0 °C, methanol:WFO molar ratio of 20:1, and reaction time of 3.0 h. The results show an increase in the ester content values of the synthesized biodiesels as there is an increase in the amounts of the catalysts used in the values of 2.0 wt% to 6.0 wt%. This is due to the greater amount of active sites in the reaction medium as the amount of catalyst used in the reaction increases (Mohebbi et al., 2020; Olubunmi et al., 2022). However, when catalyst amounts above 6.0 wt% are used in the reaction, there is a decrease in the ester content values of biodiesels, such as 89.9 % and 88.7 %, for catalyst amounts of 8.0 wt% and 10.0 wt%, respectively. This behavior is due to the fact that the increase in the amount of catalyst used causes limitations in mass transfer that reduces the interaction of active sites with the reactants. It can also cause problems related to catalyst diffusion in the reaction medium (Zhang et al., 2022; Olubunmi et al., 2022). Thus, the optimized value of the catalyst amount variable chosen was 6.0 wt%.

The optimization study for the reaction time variable is shown in Fig. 6(d) and was conducted in the range of 1.0 to 5.0 h, under conditions of reaction temperature of 160.0 °C, methanol:WFO molar ratio of 20:1, and catalyst amount of 6.0 wt%. The exposed results show an increase in the ester content values of the biodiesels produced as there is an increase in the reaction time. This can be visualized considering the use of reaction times of 1.0 and 3.0 h, which led to biodiesels with ester contents of 79.1 % and 94.6 %, respectively. However, reactions using reaction times from 3.0 h up do not provide an increase in ester contents in biodiesels, as can be seen in biodiesels obtained using reaction times of 4.0 and 5.0 h that led to ester contents of 94.2 % and 94.3 %, respectively. This slight decrease is probably due to the reversible nature of the transesterification reaction after reaching equilibrium (Santos et al., 2022). In this sense, the reaction time selected in the optimization of this study was 3.0 h, since it provides the highest ester content among the biodiesels studied.

3.4 Biodiesel characterization

Biodiesel produced using 30-MoO₃/RHA-CoFe₂O₄ catalyst in the optimized conditions of the transesterification reaction was evaluated in order to determine the quality of its physicochemical properties, as shown in Table 2. From the results presented, it can be observed that the kinematic viscosity parameter obtained a value of 4.33 mm² s⁻¹, while the density showed a value of 0.887 g cm⁻³. These values comprise the limits stipulated by the ASTM D6751 and EN 14214 standards and contribute fundamentally to the evaluation of biodiesel quality, since they are directly related to the fuel injection system in the engine and to problems related to impurity deposits throughout the vehicle engine (Mares et al., 2021; Mawlid et al., 2022).

The value determined for the flash point property of biodiesel was 155.0 °C. This high value represents a very important parameter in terms of safety because biodiesel can be handled, stored, and transported in ambient conditions without offering risks. The determined copper corrosivity parameter of biodiesel was 1a, which means absence of corrosion according to the ASTM Copper Strip Corrosion Standard Classification. This, together with the fact that this value complies with the ASTM D6751 and EN 14214 standards, corroborates the high quality of the synthesized biodiesel. In addition, the cold filter clogging point for biodiesel was 0.0 °C, suggesting good applicability of this biofuel even in cold climate regions (Santos et al., 2022). Another important property determined was the acidity index, for which the biodiesel obtained a value of 0.24 mg KOH g⁻¹. The monitoring of this property in biodiesel is of fundamental importance, especially during its storage, since changes in the values in this step can mean the presence of water in the biodiesel, which can trigger the formation of acids, promoting the biodiesel oxidation process (Santos et al., 2022). Thus, all the results obtained in the characterization of biodiesel proved to be adequate to international standards, suggesting that the biofuel produced will not cause problems in the engine, such as corrosion, impurity deposits, thermal instability, and problems related to its use at low temperatures (Mares et al., 2021; Mawlid et al., 2022).

The gas chromatography technique was employed in order to evaluate the ester content of biodiesel produced in this study. In this sense, the composition in methyl esters of biodiesel produced from WFO were obtained, as shown in Fig. 9. Through the analysis of the chromatogram present in Fig. 9(a), the presence of six peaks is perceived, five of which are related to the esters that make up biodiesel (1, 3, 4, 5, and 6) and one is related to the internal standard used methyl heptadecanoate (2*). In addition, through the analysis of Fig. 9(b), the predominance of methyl linoleate (44.37 %) and methyl oleate (18.23 %) esters is perceived, which is an expected fact, since they are derivatives of their respective fatty acids, linoleic and oleic, mostly present in the composition of soybean oil (Zhang et al., 2022; Gonçalves et al., 2021b).

Close

Fig. 9

(a) GC chromatogram and (b) composition for biodiesel produced using the catalyst 30-MoO₃/RHA-CoFe₂O₄.

Export to PPT

3.5 Evaluation of the recyclability

After the optimization of the transesterification reaction variables, the stability of the catalytic property of the catalyst was investigated when employing it in several reaction cycles. Fig. 10 shows the reusability study of the 30-MoO₃/RHA-CoFe₂O₄ catalyst performed under the following optimized conditions: 160.0 °C reaction temperature, methanol:WFO molar ratio of 20: 1, catalyst amount of 6.0 wt%, and reaction time of 3.0 h.

Close

Fig. 10

Reusability and recovery of the catalyst 30-MoO₃/RHA-CoFe₂O₄.

Export to PPT

From the analysis of Fig. 10, it is possible to obtain 94.6 % ester content for the biodiesel produced in the first reaction cycle and the maintenance of ester content in biodiesels above 70.0 % over 10 reaction cycles, which indicates the high catalytic stability of the catalyst studied. In addition, it is worth mentioning that the ester content value obtained for the synthesized biodiesel in the tenth reaction cycle (72.1 %) was considerably higher than the ester content of biodiesels obtained in the reactions performed using only CoFe₂O₄ (12.5 %), the RHA support (11.5 %), and without the presence of the catalyst (blank reaction) (11.3 %), thus indicating the catalytic robustness of the catalyst 30-MoO₃/RHA-CoFe₂O₄ in relation to other solid acid catalysts reported in the literature. It is also worth noting the catalyst recovery efficiency through magnetism, as shown in Fig. 10.

However, despite the high catalytic stability of the catalyst 30-MoO₃/ RHA-CoFe₂O₄, the determining factor for the decrease of ester contents in biodiesels (approximately 20.0 %) over the ten reaction cycles was studied. For this, EDS analysis was performed for the catalyst recovered after the tenth reaction cycle in order to determine its elemental composition, as shown in Fig. 11(a). The analysis of this data shows that there was an increase in the carbon content present (20.11 %) when compared to the carbon content of 12.41 % initially determined in the catalyst 30-MoO₃/RHA-CoFe₂O₄ (Fig. 6). This may be related to the probable deposit of organic materials on the surface of the catalyst, residues from unconverted reagents, products and co-products, which cause obstruction of the active sites, directly compromising the catalytic activity (Santos et al., 2022). In addition, the decrease in the molybdenum content present in the catalyst should be noted. It was initially 28.23 % and went to a content of 19.48 % in the catalyst after the tenth reaction cycle. This is probably due to the phenomenon of leaching, which is also corroborated by the decrease in the acidity of the catalyst surface of 8.7 mmol⁺ g⁻¹ to 4.1 mmol H⁺ g⁻¹ (Oliveira et al., 2019).

Close

Fig. 11

EDS composition of (a) recovered catalyst 30-MoO₃/RHA-CoFe₂O₄ and (b) VSM characterization for the recovered catalyst 30-MoO₃/RHA-CoFe₂O₄.

Export to PPT

The magnetic property of the catalyst throughout the reaction cycles was also evaluated, as shown in Fig. 11(b). Through the analysis of Fig. 11(b), a slight decrease in the value of catalyst saturation magnetization is identified after the tenth cycle, but without significant effect since it was possible to easily separate the catalyst by applying a magnetic field.

3.6 Kinetic study

The kinetic study, shown in Fig. 12, for the WFO transesterification process using the catalyst 30-MoO₃/RHA-CoFe₂O₄ was performed under optimized reaction conditions at four different temperatures: 130.0 °C, 140.0 °C, 150.0 °C, and 160.0 °C. The reaction behavior was assumed to be of (pseudo) first order due to the excess of alcohol used (Mawlid et al., 2022; Kaur et al., 2018). Thus, the pseudo-first reaction is expressed in Eq. (2), below.

Close

Fig. 12

Plot of (a) -ln(1-X) versus reaction time (t) at different temperatures, (b) Arrhenius for activation energy determination and (c) ln (K/T) versus 1/T.

Export to PPT

For the values of the 'k' rate constants due to temperature, transesterification reactions were performed at temperatures 130.0 °C, 140.0 °C, 150.0 °C, and 160.0 °C, as shown in Fig. 12 (a). From the analysis of Fig. 12(a), a linear behavior is perceived, typical of a (pseudo) first order reaction. The activation energy (Ea) for the transesterification of WFO with methanol was estimated by Eq. (3), below.

The Ea value determined was 27.3 kJ mol⁻¹, from the analysis of the plot 'ln k' vs’ 1 / T', as shown in Fig. 12(b). Ea values above 25.0 kJ mol⁻¹ indicate that the reaction is chemically controlled and has no diffusion and mass transfer limitations (Mawlid et al., 2022; Kaur et al., 2018).

The thermodynamic parameters determined for the transesterification ration were enthalpy (ΔH^#), entropy (ΔS^#) and Gibbs free energy (ΔG^#). Enthalpy and entropy were calculated using the Eyring-Polanyi Eq. (4) and the plot referring to 'In (k/T)’ versus ‘1/T', as shown in Fig. 12(c).

In addition, the value of ΔG^# was determined by Eq. (5), below.

All values of the thermodynamic parameters are shown in Table 3. The values obtained for ΔH^#, ΔS^#, and ΔG^#, for different temperatures, were 23.83 kJ mol⁻¹; −0,23 kJ mol⁻¹ K⁻¹; 116.4 kJ mol⁻¹; 118.7 kJ mol⁻¹; 120.9 kJ mol⁻¹; and 123,2 kJ/mol.

The positive value of ΔH^# indicates that for the formation of the transition state energy supply is necessary, characterizing the endothermic nature of the reaction carried out (Mawlid et al., 2022). The negative value of ΔS^# indicates a higher ordering in the transition state when compared to the reactants (Kaur et al., 2018; Beall, 1994). In addition, the positive value of ΔG^# indicates the non-spontaneity and endorgonic nature of the transesterification reaction carried out (Kaur et al., 2018; Beall, 1994).

3.7 Transesterification reaction and proposed mechanism using the 30-MoO₃/RHA-CoFe₂O₄ catalyst

The transesterification reaction is depicted by the overall reaction presented in Fig. 13(a). This process consists of three distinct and reversible phases that occur sequentially. The initial phase involves the conversion to monoacylglicerols, followed by the subsequent phase, which encompasses the transformation of monoacylglicerols into glycerol (Zhang et al., 2022; Adhikesavan et al., 2022). The proposed reaction mechanism for the action of the 30-MoO₃/RHA-CoFe₂O₄ catalyst in the transesterification process occurs essentially through a sequence of three steps (Zhang et al., 2022), as shown in Fig. 13(b). Firstly, triacylglycerol molecules are adsorbed onto the catalyst's surface, allowing for a greater interaction between the electron pair of the carbonyl oxygen and the Brønsted and Lewis acid sites present in the catalyst's structure, resulting in an increase in positive charge density on the carbon of the carbonyl (Step 1). This leads to the formation of a carbocation that undergoes a nucleophilic attack from the electron pair of the hydroxyl group in the methanol molecule, forming a tetrahedral intermediate (Step 2). Finally, the tetrahedral intermediate eliminates a molecule of methyl ester and generates a diacylglycerol molecule, along with the regeneration of the acid site of the catalyst (Step 3). This process occurs when the methyl ester is desorbed from its surface, making the catalyst's surface available for the next catalytic cycles. Through processes similar to this, monoacylglycerol and glycerol are formed.

Close

Fig. 13

(a) Transesterification reaction and (b) proposed mechanism using the 30-MoO₃/RHA-CoFe₂O₄ catalyst.

Export to PPT

3.8 Comparative study of heterogeneous acids catalysts for biodiesel production

Table 4 shows the comparison of the 30-MoO₃/RHA-CoFe₂O₄ catalyst with other solid acid catalysts applied to biodiesel production reported in the literature. Through the analysis of Table 4, it can be seen that the optimized condition obtained for the present study (160.0 °C reaction temperature, methanol:WFO molar ratio of 20:1, catalyst amount of 6.0 wt%, and reaction time of 3.0 h) presents lower values than reported at temperature, but with similar catalytic performance, which confirms the potential of the 30-MoO₃/RHA-CoFe₂O₄ catalyst, as well as indicates that the use of the catalyst can lead to a biodiesel production process with greater economic benefits.

Another important factor to be analyzed is the stability of the catalyst, since this performance affects the process's economy and is a relevant criterion for assessing the suitability of heterogeneous catalysts in industrial applications. Thus, it is worth noting that the data regarding the reuse cycles of the catalysts, presented in Table 4, show that the catalyst developed in this study stands out for its high catalytic stability. The 30-MoO₃/RHA-CoFe₂O₄ catalyst can be employed for 10 successive reaction cycles, providing the production of biodiesel with ester contents above 70 %. This represents a considerable advantage when compared to other heterogeneous acidic catalysts, even doubling the catalytic stability exhibited by the majority of the catalysts listed in Table 4. An example of this can be seen in the data concerning the reaction cycles provided by the HPMo/Nb₂O₅, HPMo/TiO₂, 5Mo/ZSM-22 and Nb₂O₅/SO₄ catalysts.