Efficient and Stable Rice Husk Bioderived Silica Supported Cu₂S-FeS for One Pot Esterification and Transesterification of a Malaysian Palm Fatty Acid Distillate

Abstract

A novel heterogeneous catalyst composite (CuS-FeS/SiO₂) derived from rice husk silica was engineered following pyrolysis, chemical precipitation, and chemical redox technique. The resulting catalyst was applied to the conversion of palm fatty acid distillate to biodiesel. The presence of CuS and FeS on the catalyst was verified using X-ray diffraction and Fourier transform infrared spectroscopy, nitrogen physisorption, scanning electron microscopy (FESEM) with energy dispersive X-ray (EDS) spectroscopy, and temperature-programmed desorption of NH₃ (TPD-NH₃), inductively coupled plasma-atomic emission spectrometry (ICP-AES), and TGA; a specific surface area of approximately 40 m²·g⁻¹ was identified. The impact of independent variables, i.e., reaction temperature, reaction duration, methanol:oil ratio and catalyst concentration were evaluated with respect to the efficacy of the esterification reaction. The greatest efficiency of 98% with a high productivity rate of 2639.92 µmol·g⁻¹·min⁻¹ with k of 4.03 × 10⁻⁶ mole·S⁻¹ was achieved with the following parameters: temperature, 70 °C; duration, 180 min; catalyst loading, 2 wt.%; and methanol to oil ratio, 15:1. The CuS-FeS/SiO₂ catalyst showed relatively high stability indicated by its ability to be reused up to five times.

1. Introduction

Biodiesel, comprised of long-chain fatty acid methyl esters (FAME), has been drawing considerable interest in recent years as a promising option for diesel derived from petroleum [1,2,3]. Its use has the potential to counter the growing lack of energy fuel and additionally offers a more environmentally friendly and renewable solution to the pollution caused by the utilisation of its petroleum-based equivalent [4,5,6,7]. Esterification is a frequently applied chemical technique for the production of biodiesel from poor quality feedstocks; this reaction is promoted by homogeneous acid catalysts. Essentially, 1 mol methanol and 1 mol fatty acid (FA) react stoichiometrically to yield 1 mol FAME. However, owing to the costly and complex separation required, homogeneous acid catalysts are unsuitable for the upscaling of commercial biodiesel manufacturing [8]. A more appropriate option is a heterogeneous acid catalyst, which exhibits chemical stability, and can be recycled; reaction products are obtained which have a high standard of purity, lack toxicity and have negligible adverse environmental consequences [9,10,11,12]. Biodiesel is commonly a domestically produced, clean-burning, renewable substitute for petroleum diesel. Using biodiesel as a vehicle fuel increases energy security, improves the environment and air quality, and provides safety benefits [13,14]. Biodiesel which is most frequently used as a blend with regular diesel fuel can be used in several diesel vehicles without any engine modification. The most common biodiesel blend is B20, which is 6% to 20% biodiesel blended with petroleum diesel [15]. B5 (5% biodiesel, 95% diesel) is commonly used in fleets [16].

A number of solid heterogeneous catalysts have recently been advocated for use in the manufacture of biodiesel. Earlier work has documented the activity of bifunctional catalysts containing transition metals for esterification reactions, e.g., marked esterification activity was noted following the acid-functionalisation of silica catalysts, i.e., MCM-41 and SBA-15; high biodiesel yields between 94% and 96% were attained [17,18]. The immobilisation of zinc stearate on silica gel (ZS/Si) as a catalyst displayed similar characteristics with respect to waste cooking oil transesterification–esterification; this combination produced a yield of biodiesel in the region of 98% during a reaction performed at 200 °C with an alcohol:oil molar ratio of 18:1 and a catalyst concentration of 3 wt.% to the oil weight ratio. This catalyst was observed to be stable and could be reused five-fold without attenuation of its potency [19]. However, despite its operational efficacy in enhancing esterification, its potential for upscaling into the commercial sector is limited by the expense of the acid-functionalisation of silica and by its low-density porous configuration. Another form of organic waste which is readily available in vast amounts is rice husk. This is enriched with siliceous ash and thus contains a high proportion of silica, i.e., approximately 90–97% [20]. Several researchers have evaluated the use of rice husk ash (RHA) as a precursor catalyst for the production of biodiesel from a range of sources including soybean oil [21,22], palm oil [23], and waste cooking oil [24]. Its use as a support substance for nickel sulphate as a solid acid catalyst has also been studied for the transformation of palm fatty acid distillate (PFAD) to methyl ester with respect to a range of molar ratios of MeOH:oil, the quantity of catalyst, and recyclability [25]. The production of biodiesel from PFAD has additionally been assessed using a sulphonated catalyst derived from kenaf seed cake [26]. To date, there have been no papers using RHA in combination with H₂SO₄ as a solid catalyst in this reaction context.

Comprising an excess of 90% free fatty acid (FFA), PFAD is a by-product generated by the palm oil sector. It has been appraised as a potential starting substance for the synthesis of biodiesel using esterification techniques [27,28]. Its low price and the fact that it offers a solution to the treatment of waste oils make PFAD an interesting feedstock for biodiesel manufacture. Acid catalysts are necessary as they are less susceptible to FFA and water within the PFAD [29,30]. Additionally, the catalysts promote the esterification reaction of the FFA contained within PFAD to methanol during esterification to yield FAME. Previous endeavours in relation to using PFAD oils to produce biodiesel have taken advantage of acid catalysts, such as sulphonated activated carbon and those based on zeolite. However, difficulties in relation to this process remain as acid catalysts, specifically those with a greater degree of stability, are not very efficient. In order to achieve efficient esterification of PFAD into biodiesel, the current study focused on the innovation of a metal (Cu, Fe) acid functionalised Si catalyst from a rice husk source which exhibited high stability. Differing metal compositions were incorporated into this Cu₂S-FeS/SiO₂ catalyst. Power x-ray diffraction (XRD), inductively coupled plasma-atomic emission spectrometry (ICP-AES), nitrogen adsorption–desorption isotherms, thermogravimetric analysis (TG), and Fourier transform infrared spectroscopy (FTIR) were used to characterize the prepared materials. The variables studied during the esterification process encompassed the molar ratio of methanol:oil, reaction duration and temperature, and quantity of catalyst. A one-variable-at-a-time (OCAT) method was used for their optimisation. The degree of stability and potential for recyclability of the catalyst were also evaluated.

2. Results and Discussion

2.1. Catalyst Characterisation

Figure 1 shows the XRD pattern for the support and all the prepared catalyst. The XRD diffractogram of the support shows an amorphous structure at around 2θ of 26.38°, which is a typical characteristic of silica. The amorphous structure of the SiO₂ was drastically reduced after impregnation with the CuS and FeS metal solution. It was observed that the intensity of CuS peaks was low as compared to the continuous increase in Fe content (2–15 wt.%). There is a single bored pattern with 2θ at approximately 26.38°; following silica doping with CuS, Cu₂S patterns were evident. Marked crystallinity of the copper sulphide diffraction peaks was indicated—reference code: 00-001-1281, crystal system: hexagonal, space group: P63/mmc, space group number: 194 are hkl, and 2θ: 18.62°, 19.84°, (004)22.37°, (101)27.33°, (006)33.02°, (110)48.10°, (208)73.99°, and (213)79.07°. The addition of FeS to the Cu₂S gave rise to additional FeS and FeS₂ patterns, i.e.: reference code: 00-002-1204, mineral name: Troilite (FeS), with 2θ: 17.34°, 18.78°, 23.26°, 29.96°, 33.40°, 35.30°, 38.78°, 43.03°, 51.91°, 53.21°, 56.02°, and 63.20°, and reference code: 00-002-0908 with hkl, and 2θ: (110)25.87°, (011)30.16°, (020)33.15°, 35.16°, (111)37.28°, 38.95°, (021)43.03°, 44.14°, (121)46.78°, 47.56°, 48.65°, (211)51.91°, (220)53.21°, (002)54.23°, (130)54.93°, (031)57.95°, (221)60.45°, (112)60.89°, (131)61.79°, and (310)65.18°, respectively. As the FeS loading increment was increased, the intensity of these patterns was heightened. From the XRD, the widening patterns of the line with the highest peak, together with the use of Scherrer’s equation, indicated that the mean silica crystallite dimension was 13.42 nm (

Table 1. Physicochemical properties of synthesised silica and monometallic catalyst.

Catalyst	XRD a	BET b		TPD-NH3 c	Calculated Active Site (µmol·m−2)	Mass Transfer Limitations
	Crystallite Size a (nm)	Surface Area b (m2·g−1)	Pore Volume b (cm3·g−1)	Amount of NH3 Adsorbed c (µmol·g−1)		Measured Rate (umol·g−1·min−1)	Reaction Rate Constant k (mole·S−1)
SiO2	13	68	0.72	152.06	2.23	132.59	3.54 × 10−7
Cu/SiO2	17	62	0.71	198.33	3.19	293.55	2.59 × 10−7
Fe/SiO2	18	58	0.70	204.81	3.53	363.68	2.79 × 10−7
Cu2S(2%)/SiO2	26	50	0.69	252.14	5.04	407.55	2.34 × 10−7
Cu2S(5%)/SiO2	30	44	0.65	437.99	9.95	458.58	2.18 × 10−7
Cu2S(8%)/SiO2	30	42	0.62	2590.4	61.6	480. 75	3.49 × 10−7
Cu2S(10%)/SiO2	35	50	0.64	1879.68	37.5	533. 87	2.90 × 10−7
Cu2S(12%)/SiO2	38	37	0.60	1744.12	47.1	586.98	2.32 × 10−7
Cu2S(15%)/SiO2	41	33	0.57	1503.28	45.55	548. 61	2.23 × 10−7
Cu2S(8%)-FeS(2%)/SiO2	31	42	0.63	402.32	9.57	667.30	3.45 × 10−7
Cu2S(8%)-FeS(5%)/SiO2	36	42	0.60	698.86	16.63	760.735	3.82 × 10−7
Cu2S(8%)-FeS(8%)/SiO2	38	41	0.59	2999.21	73.15	813. 66	4.71 × 10−7
Cu2S(8%)-FeS(10%)/SiO2	46	40	0.57	4133.29	103.33	2639.92	4.03 × 10−6
Cu2S(8%)-FeS(12%)/SiO2	48	35	0.30	2398.64	68.53	869.41	5.77 × 10−7
Cu2S(8%)-FeS(15%)/SiO2	51	33	0.26	2782.91	84.33	910. 88	3.97 × 10−7

^a Determined by Debye–Scherrer equation. ^b BET surface area. ^c NH₃ desorption peak for all catalysts, the total amount of NH3 desorbed from 50 °C to 950 °C.

). The crystallite dimension of the hematite Fe₃O₄ in AC-Fe_x-SO₃Cl for the 8% CuS-FeS-loaded catalyst was 40.18 nm; the corresponding figure for a 15% CuS-FeS-loaded catalyst was 40.61 nm, from which it is inferred that the addition of the copper, sulphide, and iron ions into the carbon lattice augmented the dimensions of the crystallite. The XRD results show that increasing the active metals loading percentage lead to an increase in the crystallinity size due to crystal agglomeration [11,31].

The results from the nitrogen sorption indicated a decrease in the BET surface area, with values of 68 m² g⁻¹ for silica, and 62 m² g⁻¹ and 58 m² g⁻¹ for Cu/SiO₂ and Fe/SiO_2, respectively. The equivalent porous volumes were 0.72 cc/g, 0.71 cc/g, and 0.70 cc/g, respectively. This fall was identified owing to the increased proportion of doping which blocked the pores of the silica. Additionally, there was a modest loss of integrity of the structure and collapse as a result of metal oxide loading on the surface of the pore channels. The rise in Cu and Fe doping proportions also diminished the area of the surface and the pore volume; the smallest BET value obtained for the Cu₂S_(8%)-FeS_(15%)/SiO₂ catalyst was 33. This was ascribed to the Cu and Fe loading on the surface of the silica surface pore channels which, in turn, led to a further reduction in the activation of the silica [32,33].

Figure 2a illustrates the images of the silica-based rice husk acquired by FESEM. The configuration of the husk’s particles represented heterogeneous structures of micro dimensions. Figure 2b–h illustrate the images obtained from FESEM of the Cu₂S_(x)-FeS_(y)/SiO₂ catalysts, where x = 8 wt.%, and y = 2, 5, 8, 10, 12 and 15 wt.%, respectively.

Well-distributed large particles of micro dimensions were produced by the bimetallic Cu₂S_(x)-FeS_(y)/SiO₂ catalysts. It can be seen from Figure 2b–g that adding Fe-enriched moieties generated condensed agglomerates, again of micro dimensions, disseminated amongst a few needle-like configurations. Irregular particles related to the FeS species, which were white in colour, exhibited a rise in dimension as the concentration of Fe species in the catalyst rose from 2 wt.% to 15 wt.%. The biggest particles observed, i.e., 2.5 µm, were associated with 15 wt.% doped Fe (Figure 2h). These findings suggested the surplus generation of aggregates of FeS on the surface of the catalyst, and that these were linked with a loss of structural integrity of the particles. It was therefore proposed that the type of FeS used plays a significant role in the modification of the morphological characteristics of the catalyst surface.

TPD-NH₃ analysis provided data on the acid density and strength of the empirical SiO₂, Cu/SiO₂, Fe/SiO₂, CuS/SiO₂, and CuS-FeS/SiO₂ catalysts; the respective acid profiles are presented in Figure 3 and Table 1. Weak and medium acidities, i.e., T_max > 250 °C and T_max > 500 °C, were associated with each catalyst [34,35,36]. At the lower temperature, the synthesised silica exhibited a triad of desorption zeniths, i.e., 163 °C, 512 °C, and 730 °C, together with only a minimal quantity of vapour adsorption, i.e., 152.06 µmol/g. Cu and Fe doping of the support led to a rise in acidities to 198.33 µmol/g and 204.81 µmol/g, respectively. A rise in CuS loading also caused the acidity to increase, i.e., CuS loading of 2 wt.% and 8 wt.% generated acidities of 252.14 µmol/g and 2590.4 µmol/g, respectively. However, when CuS loading was above 8 wt.%, the acidity diminished to 1503.28 µmol/g. A change in acidity to 402.3206 µmol/g and 4133.29 µmol/g was noted following the addition of Cu and Fe, respectively, to the catalyst support. The acid quantity and density were also enhanced following the use of sulphur to modify the supported catalyst; this was a result of the electron engagement between the sulphur atom and the element at the active site [37]. In temperatures greater than 600 °C, a de novo desorption zenith was evident for the bimetallic catalyst; this was indicative of the high acidity and acid density of the sites, i.e., 4133.29 µmol/g. The TPD results showed that elevating the proportion of FeS from 2 wt.% to 10 wt.% heightened the density and quantity of acid owing to the potentiating effects of the intercalated metal, i.e., FeS, and the CuS and silica residues during the phase of catalyst manufacture. Conversely, a further rise in the loading proportion decreased the acid content; the addition of more elements caused the active sites to become agglomerated.

The results of the TGA analysis for silica and modified SiO₂ are presented in Figure 4a. The silica retained its structural integrity up to temperatures of approximately 600 °C with little weight reduction. By contrast, a mullite phase weight reduction was noted in the first area arising between temperatures of 90 °C and 235 °C, and again in the final area, within the range from 700 °C to 800 °C. These could represent the loss of surface molecules of ammonia and the breakdown of CuS and FeS, respectively. The catalyst proved to be extremely stable from a thermal perspective for the esterification; the latter is typically performed at a temperature of under 100 °C.

The IR spectra of silica and those of the synthesised Cu₂S-FeS/SiO₂ catalysts are illustrated in Figure 4b. The absorption bands and notable zeniths at 567 cm⁻¹ and 1580 cm⁻¹ were indicative of the metal–sulphide bond and the existence of H₂O bending, respectively. No stretching vibrations in relation to the M–S bond were evident, as these exceeded the range of measured parameters. The zenith at 538 cm⁻¹ reflected the success of the production technique. As verified by the NH₃-TPD data, the S and active element d-orbital interplay promoted the activity of the esterification reaction by enhancing the catalyst’s acid content.

2.2. One-Step Esterification of PFAD

The FFA transformations to PFAD-derived biodiesel promoted by the various SiO₂, Cu/SiO₂, Fe/SiO₂, Cu₂S_(x%)/SiO₂, and Cu₂S_(x%)-FeS_(y%)/SiO₂ catalysts containing various Cu and Fe concentrations, i.e., 2–15 wt.%, are shown in Figure S1 and Figure 5. For the reaction, the established conditions included: temperature, 60 °C; reaction duration, 60 min; catalyst loading, 3 wt.%; and methanol:oil molar ratio, 5:1. The maximum and minimum FFA conversions were 84.44 ± 1.7% and 33.64 ± 1.4%, obtained with the use of the Cu₂S_(8%)-FeS_(10%)/SiO₂ and silica catalysts, respectively. When the Fe component was elevated to above 10 wt.%, the conversion proportion was decreased to 79.63 ± 2.5%, as shown in Figure S1. It was therefore concluded that the maximal esterification reaction rate was achieved with the following parameters: maximal acid density, 4133.29 µmol; CuS, 8 wt.%; and FeS, 10 wt.%. The esterification process was therefore optimally promoted when the catalyst had an elevated acidity. In conclusion, the Cu₂S_(8%)-FeS_(10%)/SiO₂ catalyst showed the highest FAME conversion and was used for further optimizing the reaction conditions. Figure 5 shows that the increase in the number of active acid sites led to an increase in the conversion of the methyl ester because the reaction of PFAD with high acid value follows the acid esterification mechanism. In order to determine the catalyst productivity, the catalyst transport limitation was calculated per catalyst weight and per acid site (Table 2). The Cu₂S_(8%)-FeS_(10%)/SiO₂ catalyst shows a high k of 4.03 × 10⁻⁶ mole·S⁻¹ compared to other catalysts used in this study; the highest k constant means a high reaction rate due to the fact that the Cu₂S_(8%)-FeS(10%)/SiO₂ catalyst has the highest acidity and active sites (Table 1). Moreover, modification of SiO₂ with different active sites and different active element ratios leads to an increase in the productivity yield. Table 1 shows the Cu₂S_(8%)-FeS_(10%)/SiO₂ catalyst giving the highest product yield with 2639.92 µmol·g⁻¹·min⁻¹ compared to SiO₂ 132.59 µmol·g⁻¹. This is due to the fact that the Cu₂S_(8%)-FeS_(10%)/SiO₂ catalyst shows the highest acidity which is required for the esterification reaction.

2.3. Biodiesel Production Reaction Optimisation

In order to achieve the best results from the FPAD esterification process, the Cu₂S_(8%)-FeS_(10%)/SiO₂ catalyst was selected. The consequences of the catalyst loading concentration, methanol:PFAD molar ratio, and the reaction temperature and duration on biodiesel production were examined as indicated in Figure 6a–d.

In order to study the effect of reaction temperatures between 60 °C and 80 °C on the transformation of FFA, the remaining reaction parameters were fixed, i.e.,: molar ratio methanol:oil, 10:1; catalyst concentration, 3 wt.%.; reaction time, 60 min; and stirring rate, 800 rpm. A rise in temperature between 60 °C and 70 °C produced FFA conversion rates within the range of 84.44 ± 0.8%–95.94 ± 2.1%, with the most productive temperature identified as 70 °C. Reducing the temperature to 60 °C reduced the transformation of FFA as it had a negative impact on the equilibrium of the reaction. When the temperature of the reaction was increased, the reactants acquired sufficient kinetic energy to hasten the mass transfer rate between the PFAD–methanol–catalyst phases; this, therefore, optimised the conversion rate [38,39]. The maximum FFA conversion seen at 70 °C with the use of the Cu₂S_(8%)-FeS_(10%)/SiO₂ catalyst reflected the requisite for the increased reaction temperature for the acid catalysis of esterification as opposed to that required for base catalysis [40].

The results of varying the reaction duration of the PFAD process are presented in Figure 6b. The remaining conditions were set as follows: molar ratio methanol:oil, 10:1; catalyst concentration, 3 wt.%; reaction temperature, 70 °C; and stirring rate, 800 rpm. The use of the Cu₂S_(8%)-FeS_(10%)/SiO₂ catalyst after 120 min produced a slow reaction rate, potentially as a consequence of the presence of non-uniform mass transfer systems which led to the insufficient diffusion of the catalyst within the PFAD, solid catalyst, and methanol reaction components. However, as the duration of the reaction was prolonged to 3 h, the conversion rate of FFA was increased, becoming practically constant at reaction times above 180 min. The maximal FFA transformation, i.e., 96.94 ± 1%, was seen following a reaction duration of 180 min. However, increasing the reaction time beyond this failed to improve the conversion rate further. Thus, in order to conserve energy usage and the expense of the biodiesel synthesis method, a reaction duration of 180 min was deemed to be the most favourable.

The impact of the amount of catalyst used on the conversion rate of FFA is demonstrated in Figure 6c; varying proportions, i.e., between 1 and 5 wt.%, were tested. The remaining reaction conditions were set as follows: molar ratio methanol:oil, 10:1; reaction temperature, 70 °C; reaction duration, 180 min; and stirring rate, 800 rpm. An increase in catalyst concentration from 1 wt.% to 2 wt.% led to a rise in FFA transformation, i.e., from 92.14 ± 1.4% to 94.29 ± 1.1%. However, at concentrations above 2 wt.%, no further promotion of FFA transformation was evident. Thus, a relatively low Cu₂S_(8%)-FeS_(10%)/SiO₂ catalyst concentration of 2 wt.% produced the maximal FFA conversion rate of 94.29 ± 1.1% in the presence of the other described reaction conditions.

The molar ratio of methanol:oil plays an essential role in the transformation of FFA, and in particular, in the conversion of triglycerides to methyl esters. In order to shift the reaction equilibrium in the direction of increased methyl ester generation, the molar ratio has to be increased. The effect of a ratio rise from 5:1 to 25:1 was therefore evaluated. The other reaction parameters were kept constant, i.e.: reaction temperature, 70 °C; reaction duration, 180 min; catalyst concentration, 2 wt.%; and stirring rate, 800 rpm (Figure 6d). The optimum transformation rate of FFA, i.e., 99.13 ± 0.5%, was achieved using the Cu₂S_(8%)-FeS_(10%)/SiO₂ catalyst when the molar methanol:PFAD ratio was 15:1. It was proposed that the high methanol content enhanced the methanol and oil diffusion and miscibility, which therefore promoted the esterification reaction rate and shifted the reaction to favour methyl ester synthesis [41].

2.4. Catalyst Reusability

The most favourable reaction conditions, i.e.: methanol:PFAD molar ratio, 15:1; reaction temperature, 70 °C; reaction duration, 180 min; and Cu₂S_(8%)-FeS_(10%)/SiO₂ catalyst concentration, 2 wt.%, led to a maximum FFA transformation rate of 99.13 ± 0.5%. With these in place, the potential for catalyst reuse was assessed over five consecutive reaction cycles. The data are presented in Figure 7 and indicate that for the initial three cycles, the catalytic potency of the acid catalyst was retained. Nevertheless, following reuse of the catalyst over five cycles, the FFA conversion and yield rate were notably reduced, i.e., from 99.13 ± 0.5% to 82.73 ± 2% and from 93.18± 0.2% to 77.76± 0.6%, respectively, which was thought to be the result of Cu and Fe species leaching from the surface of the catalyst [42]. Thus, the leaching of Cu²⁺, Fe³⁺, and S⁶⁻ ions into the reaction mixture may have led to the loss of catalysis efficacy over the five cycles.

These findings concurred with the data obtained from the CHNS and ICP-AES investigations, which demonstrated an increase in the liquid product of Fe³⁺ and S⁶⁻ when the first and final runs were compared, i.e., 8 and 23 ppm in the initial run, and 19 and 33 ppm in the final cycle, respectively. This observation was considered to underlie the esterification reaction activity reduction. Between the initial and final reaction cycles, in excess of 52.8 ppm metal leaching into the liquid product were identified, a figure at the higher end of the spectrum for contaminants recommended by the EN 12662 Standard Specification for Diesel Fuel Oils. These data suggested that the Cu₂S_(8%)-FeS_(10%)/SiO₂ catalyst exhibited acceptable stability and demonstrated significant resistance to leaching. Since the active element and sulphur residues have hydrophilic properties, they will tend to adsorb aqueous components prior to engagement with the solid catalyst surface [43,44]. Therefore, this hastened the esterification process and consequently, promoted the rate of FFA transformation and synthesis of biodiesel.

When the concentrations of CuS and FeS were reduced, they were insufficient to prevent water molecules from interacting with the catalyst; the latter therefore obstructed the active sites and led to an attenuation of catalytic activity, which slowed the esterification rate [45]. The fact that the Cu₂S_(8%)-FeS_(10%)/SiO₂ could be recycled added to the financial sustainability of the technique. As previously mentioned, the optimum heterogeneous catalyst for usage in biodiesel synthesis should demonstrate significant recyclability, be derived from waste products, and be able to be segregated from the reaction products with ease.

The retention of the high rate of FFA transformation into biodiesel over five reaction cycles was in keeping with the data obtained from FTIR (Figure 8); a unique absorption band at 3466 cm⁻¹ was acquired from the reaction product which differed from that representing PFAD. This was ascribed to the stretching vibration of C-O in association with the methyl esters. The absorption zeniths noted at 2920 cm⁻¹ and 2860 cm⁻¹ reflected the CH-asymmetric vibration whereas the C=O absorption was indicated by an intense band at 1744 cm⁻¹. The peaks contained within the 1000–1300 cm⁻¹ were attributed to the stretching vibration of C-O, implying that the resultant liquid following the reaction was similar in character to biodiesel, being made up of long-chain fatty acid esters [46].

These data indicate that the esterification process, catalysed by the Cu₂S_(8%)-FeS_(10%)/SiO₂ catalyst, gave rise to the total transformation of carboxylic acid to ester [47]. The FTIR spectra data confirmed that this result persisted after five consecutive cycles due to the stability of the catalyst remaining debatable. Together with the fact that the catalyst is itself manufactured from waste materials, these results indicate that the Cu₂S_(8%)-FeS_(10%)/SiO₂ catalyst is a viable financial proposition. The most efficient method of separating the catalyst from the final product was via centrifugation.

Scaling up the laboratory experiments to a pilot plant or to full-scale production is very important in the industrial sector. To make this succussed fully requires an in-depth understanding of the chemical kinetics and transport limitations. From this point of view, the catalytic activity of Cu₂S_(8%)-FeS_(10%)/SiO₂ catalyst was compared with several similar catalysts [48,49,50,51] (Table 2) that were used to study the esterification–transesterification of high FFA non-edible oil (PFAD). From this summary, Cu₂S_(8%)-FeS_(10%)/SiO₂ showed high activity with a k of 4.03 × 10⁻⁶ Mole. S⁻¹, which indicated a high reaction rate with low mass transfer limitation; on the other hand, the Cu₂S_(8%)-FeS_(10%)/SiO₂ catalyst with a 2 weight percent catalyst loading, a 15:1 MeOH to oil ratio, and a reaction temperature of 70 °C within three hours, was most reactive, converting 98 percent of the FAME. This is a result of the existence of active sites with high acidity, large surface area, and high pore volume, which are favourable conditions for esterification reactions. Interestingly, the Cu₂S_(8%)-FeS_(10%)/SiO₂ catalyst was able to be reused for five runs with FAME conversion maintained at 98%, as compared with recent studies showing low reusability with a FAME yield lower than 74%. Furthermore, the Cu₂S_(8%)-FeS_(10%)/SiO₂ catalyst is also easy to regenerate by a simple washing method. Therefore, the regeneration of this catalyst is highly economical and time efficient.

Table 2. Summary of recent studies of catalysts for biodiesel production.

Entry	Catalyst	Substrates	Reaction Conditions	Rate Constant (k)	Biodiesel Yield (%)	Reusability (Yield of the Last Run (%))	Ref.
1	SO42−/ZrO2-SiO2(Et)-IL-3	Soybean oil + MeOH	T = 150 °C, CA = 5 wt.%, M:O = 18:1	-	98.99	5 (95.22)	[48]
2	SO2−4/TiO2-SiO2	waste oil + MeOH	T = 120 °C, CA = 10 wt.%, M:O = 20:1, 3 h	-	88	2 (74.3)	[49]
		Oleic acid + MeOH	T = 120 °C, CA = 10 wt.%, M:O = 20:1, 3 h	-	93.4	-
3	La-PW-SiO2/SWCNTs	Oleic acid + MeOH	T = 65 °C, CA = 1.5 wt.%, M:O = 15:1,	-	93.1	6 (88.7)	[50]
4	SO2−4/TiO2-SiO2	Waste cooking oil	T = 200 °C, CA = 3 wt.%, M:O = 9:1, 5 h	-	92	-	[51]
5	P-C-750-S-210	Waste Oil	T = 220 °C, CA = 4 wt.%, M:O = 21:1, 5 h	0.0166 (mol−1·min−1)	50% (conversion)	-	[52]
6	H2SO4 and NaOH	Jatropha curcas oil	T = 65 °C, CA = 1 wt.%, M:O = 3:10, 6 h	0.0031 (min−1) and 0.008 min−1	21.2% and 90.1%	-	[53]
7	CaO	Sunflower oil	T = 80 °C, CA = 1 wt.%, M:O = 6:1, 15 bar 5.5 h	59.22 × 10−3·min−1	91%	-	[54]
8	Mg0.81Al	Palm oil	T = 60 °C, CA = 1 wt.%, M:O = 6:1, 3 h	1.60 × 10−6	74.8	-	[55]
9	Cu2S(8%)-FeS(10%)/SiO2	PFAD	T = 70 °C, CA = 2 wt.%, M:O = 15:1, 3 h	4.03 × 10−6 (S−1)	98% (conversion)	5 (82.73)	This work

Table 2. Summary of recent studies of catalysts for biodiesel production.

Entry	Catalyst	Substrates	Reaction Conditions	Rate Constant (k)	Biodiesel Yield (%)	Reusability (Yield of the Last Run (%))	Ref.
1	SO42−/ZrO2-SiO2(Et)-IL-3	Soybean oil + MeOH	T = 150 °C, CA = 5 wt.%, M:O = 18:1	-	98.99	5 (95.22)	[48]
2	SO2−4/TiO2-SiO2	waste oil + MeOH	T = 120 °C, CA = 10 wt.%, M:O = 20:1, 3 h	-	88	2 (74.3)	[49]
		Oleic acid + MeOH	T = 120 °C, CA = 10 wt.%, M:O = 20:1, 3 h	-	93.4	-
3	La-PW-SiO2/SWCNTs	Oleic acid + MeOH	T = 65 °C, CA = 1.5 wt.%, M:O = 15:1,	-	93.1	6 (88.7)	[50]
4	SO2−4/TiO2-SiO2	Waste cooking oil	T = 200 °C, CA = 3 wt.%, M:O = 9:1, 5 h	-	92	-	[51]
5	P-C-750-S-210	Waste Oil	T = 220 °C, CA = 4 wt.%, M:O = 21:1, 5 h	0.0166 (mol−1·min−1)	50% (conversion)	-	[52]
6	H2SO4 and NaOH	Jatropha curcas oil	T = 65 °C, CA = 1 wt.%, M:O = 3:10, 6 h	0.0031 (min−1) and 0.008 min−1	21.2% and 90.1%	-	[53]
7	CaO	Sunflower oil	T = 80 °C, CA = 1 wt.%, M:O = 6:1, 15 bar 5.5 h	59.22 × 10−3·min−1	91%	-	[54]
8	Mg0.81Al	Palm oil	T = 60 °C, CA = 1 wt.%, M:O = 6:1, 3 h	1.60 × 10−6	74.8	-	[55]
9	Cu2S(8%)-FeS(10%)/SiO2	PFAD	T = 70 °C, CA = 2 wt.%, M:O = 15:1, 3 h	4.03 × 10−6 (S−1)	98% (conversion)	5 (82.73)	This work

2.5. Heterogeneous Acid Catalysts Mechanism for Biodiesel Production

Fatty acids are transformed into ester during the esterification process when any type of alcohol is present. The most popular method for lowering the viscosity of the fatty acid and producing fatty acid esters is the esterification process, also known as alcoholization (FAEs). By using an acid catalyst, the esterification process can proceed more rapidly. Esters and water are produced as a result of the interaction between FFA and alcohol. As a result, the biodiesel generated by the esterification process is appropriately referred to as fatty acid mono-alkyl ester derived from renewable feedstock, whether it is edible or non-edible. Short-chain alcohols, i.e., methanol or ethanol, are generally preferred for the esterification process. Methanol is a polar molecule, which exhibits high reactivity. Subsequently, the reaction of FFA with methanol tends to produce fatty acid methyl ester (FAME), along with water as a side product. Because they are more affordable than other high-boiling-point alcohols, methanol and ethanol are chosen as the preferred alcohol. By employing short-chain methyl and ethyl alcohol, the steric hindrance phenomena are also prevented, and the esterification process is made more effective. To reduce conversion losses in ester conversion and prevent saponification during the esterification process, an acid catalyst is typically employed in conjunction with a high FFA catalyst. Because of this, the search for suitable raw materials that can offer significant amounts of fatty materials is typically constrained by their FFA content. In the presence of a suitable catalyst, Figure 9 shows the reaction for generating methyl ester through the esterification process [10].

3. Materials and Methods

Merck (USA) was the procurement source for copper (II) nitrate hemi(pentahydrate), ACS reagent (99.99%, Cu(NO₃)₂·2.5H₂O), iron(III) nitrate nonahydrate (99.95%, Fe(NO₃)₃·9H₂O), and hydrochloric acid (HCl) ACS reagent (37%), all of which had 98% purity. Ammonia (NH₃) (35%) and methanol (99.99%) were obtained from R&M chemicals and Sigma Aldrich (USA), respectively. Ethanol, hexane, and acetone solvents were purchased from Merck & Co. (USA). Fulka Analytical supplied the FAME reference standards for methyl oleate, methyl linoleate, methyl palmitate, methyl myristate, and methyl stearate, respectively, and the methyl heptadecanoate internal standard, with a purity of over 99%. Supra Solv Merck was the source of the GC grade n-hexane used. PFAD was purchased from Jomalina R&D (Sime Darby Co., Klang, Malaysia). The American Oil Chemist’s Society and British Standard techniques were followed for the analysis of the PFAD feedstock properties. The starting silica (SiO₂) derived from the rice husk underwent preparation as reported previously [56]. For the current study, the PFAD feedstock was gathered from the Malaysian Palm Oil Board (Selangor, Malaysia); no additional treatment or purification was performed prior to its incorporation in the reaction.

The properties of the PFAD are listed in

Table 3. Chemical properties and fatty acid composition of the MPOB-PFAD.

Properties	Method	Present Study	Ref a	Ref b
FFA content (%)	AOCS Cd 3d-63	$98.9 \pm$ 0.71	86.3	90.24
Saponification value (mg KOH/g)	AOCS Tr 1a-64	$197 \pm$ 3.32	149.74	207.69
Molecular weight (g mol−1) a		193.2		270.11
Fatty acid composition (wt.%)
Myristic (C14:0)		$1.08 \pm$ 0.05	1.9	1.03
Palmitic (C16:0)		$\pm$ 0.32	45.7	48.02
Stearic (C18:0)		$3.24 \pm$ 1.01	4.3	3.42
Oleic (C18:1)		$\pm$ 1.32	40.2	41.01
Linoleic (C18:2)		$6.42 \pm$ 0.71	7.9	6.52
Σ Saturated		63.24	51.9	52.44
Σ Unsaturated		36.76	48.1	47.53

Reference a and b obtained from [57,58], respectively.

; there was a high quantity of FFA, i.e., approximately 98.9 wt.%; the acid value, saponification value, and molecular weights were 197.8 wt.%, 197 mg/KOH/g, and 193.2 g mol⁻¹, respectively. The GC-MS data demonstrated the following PFAD component percentages: myristic acid, 1.08 wt.%; palmitic acid, 58.92 wt.%; stearic acid, 3.24 wt.%; oleic acid, 30.34 wt.%; and linoleic acid, 6.42 wt.%. This analysis demonstrated the differences in carbon chain lengths present in the PFAD, as well as both saturated and unsaturated bonds.

3.1. Preparation of SiO₂

The silica was derived from rice husks sourced from Sekinjang, Malaysia using an adaptation of a previously described method [59]. The risk husks were rinsed and agitated in a solution of hydrochloric acid (1 mol) for 4 h at a temperature of 90 °C. Then, the mixture was filtered and the treated solid rice husks were washed with hot distilled water (about 4 L till pH7) then the solid treated rice husk was subjected to drying overnight at 120 °C. Calcination was performed for 6 h at a temperature of 600 °C at ATM using a tubular furnace with a heating rate of 5 °C·min⁻¹ in order to obtain the rice husk silica product.

3.2. Preparation of Cu/SiO₂ and Fe/SiO₂ Supported Catalyst

A co-precipitation technique was employed in order to engineer the Cu/SiO₂ and Fe/SiO₂ supported catalysts. An aqueous solution of the requisite quantities of 5 mmol 20 mL Cu(NO₃)₂·2.5H₂O and 5 mmol Fe(NO₃)₃·9H₂O 20 mL was made up. Silica was admixed with 50 mL deionised (DI) water; the metal solution was then delivered into the latter drop by drop (10 drops per minute), utilising a burette for accuracy. The final solution underwent magnetic stirring (900 rpm) in ambient conditions for 30 min an hour. An amount of 5.32 M ammonia was added a drop at a time with continuance stirring at 900 rpm in order to achieve a solution pH of 10 at room temperature. A further 60 min of stirring was performed using a magnetic stirrer, following which the solution was warmed to a temperature of 90 °C for 120 min, and then allowed to dry overnight at 120 °C. Calcination of the resultant 10 g of powder was undertaken for 180 min in a tubular furnace with the following conditions: temperature, 300 °C; heating rate, 5 °C·min⁻¹; and nitrogen flow, 40 cc³·min⁻¹. The furnace was then allowed to cool to ambient conditions to yield ~8.5 g of final catalyst.

3.3. Preparation of CuS-FeS/SiO₂ Supported Catalyst

A pristine supported CuS-FeS/SiO₂ catalyst with varied active metal (CuS and FeS) loading states (i.e., 2, 5, 8, 10, 12, and 15 wt.%) was prepared using a hydrothermal technique. Initially, 5 mmol Cu(NO₃)₂·2.5H₂O and Fe(NO₃)₃·9H₂O 20 mL were made up to a solution with 5 cm³ deionised water, which was then stirred for 15 min. This solution was then admixed with a combination of activated carbon and 90 cm³ 2:1 ethanol:DI water, and stirred for 30 min. Additions of 10 cm³ of 6 M ammonia aqueous solutions and subsequently, 10 mmol (1.52 g) solution of thiourea, were made, and further 30- and then 60-min periods of agitation, respectively, were carried out. Following decanting of the solution into a Teflon-lined autoclave made from stainless steel, a 12-h hydrothermal protocol was followed at a temperature of 180 °C. The resulting mixture was allowed to cool to ambient conditions, and then underwent centrifugation at 3500 rpm for 15 min in order to facilitate extraction of the precipitate. The remaining powder was rinsed with an analytical grade of ethanol and acetone. Repetition of the above procedure was then performed with the remaining loading metal concentrations (2, 5, 8, 10, 12, and 15 wt.%).

3.4. Catalyst Characterisation

A diffractometer (Shimadzu, model XRD 6000, Duisburg, Germany) was used to perform XRD and to establish the catalyst’s crystalline profile. Debye–Scherrer’s formula was used to compute the specimens’ crystallite size D (D = β_hkl × 0.9 × 1.5438/Cos (θ)) [60]. A temperature-programmed desorption (TPD) technique (ThermoScientific TPDRO 1100, Waltham, MA, USA) was used to ascertain the active site density and the catalyst acid strength; the probe molecule employed was ammonia. A reactor equipped with a thermal conductivity detector was used for the experiment, into which 0.07 g catalyst was inserted and then subjected to a temperature of 150 °C for 45 min. Subsequently, the samples were cooled down to room temperature and subjected to ammonia adsorption for 1 h, and ammonia in helium (20% NH₃ in He) gas was flowing at a rate of 20 cc³·min⁻¹. Ambient temperature purging with nitrogen was conducted over 55 min in order to eradicate any gaseous ammonia. The ammonia desorption was then analysed over a temperature range of 50–950 °C in the presence of helium flow with parameters of 10 °C·min⁻¹ and 20 cc³·min⁻¹; the thermal conductivity detector was used to record the data.

A Mettler Toledo TG-SDTA (Leicester, UK) apparatus (Pt crucibles, Pt/Pt–Rh thermocouple) was employed to assess the catalyst’s thermal stability properties under the following conditions: 18 mg of catalyst were first purged with a nitrogen gas flow rate of 30 cc³·min⁻¹; heating rate, 10 °C·min⁻¹; and range, ambient temperature to 1000 °C. FTIR analysis was performed using a Perkin–Elmer Spectrum(Germany) (PS) Spectrum 100 FT-IR spectrometer with resolution 4 cm ⁻¹ operating in the range of 300–4000 cm⁻¹ for determination of the chemical functional group biodiesel product. Structural and elemental analysis of the final product was performed by a field emission scanning electron microscope (FESEM). Samples on the sample holder were pre-coated with platinum (SB, Pt 99.9999%) using the auto-fine coater JFC-1600, JEQL at 20 mA of the magnetron-type sputtering current for one minute. The micrograph was taken using JSM 6701F SEM, JEOL (Tokyo, Japan) with the emission current at 2.00 KV and working distance (WD) at 3.0 mm. The total surface area of the catalysts was measured by using the Brunauer–Emmett–Teller (BET) method. These analyses were conducted using Thermo-Finnigan Sorpmatic 1990 series(USA) using nitrogen adsorption/desorption analysis; at least 0.5 g sample was used for analysis. Before the analysis, the sample was degassed at 150 °C for 12 h. This technique was performed by applying the nitrogen adsorption/desorption technique on the surface of the catalyst at a liquid nitrogen temperature of −196 °C and the relative pressures (P/P_o) ranging from 0.04 to 0.4, where a linear relationship was maintained. Prior to N₂ adsorption, the sample was degassed at 250 °C for 8 h to remove any adsorbed molecules from pores and surfaces. Shimadzu X-ray Fluorescence (XRF) (Rainy EDX-720 spectrometer (Duisburg, Germany)) and CHNS elemental analysis (Leco, TruSpec^® CHN (Kirchheim, Germany)) were used to obtain a breakdown of the catalyst’s elemental components. The active metal leaching from the catalyst surface was studied by ICP. The analysis was performed using a Perkin–Elmer Emission Spectrometer Model Plasma 1000. Firstly, a 1 cm³ of biodiesel product were mixed with concentrated nitric acid (65% Merk (Malaysia) under sonication (in order to transfer the leached active element from the organic layer to the inorganic layer) using Elmasonic S150 (China) for 30 min and then the sample was diluted to be 5 v/v% of the nitric acid; then, the sample was run using the 20 cm³ in the ICP and the elements concentrations were calculated according to the standard curves.

3.5. Esterification Reaction of PFAD

The esterification–transesterification reaction of PFAD was carried out using the conventional reflux method: Approximately 5 g of PFAD, a specified amount of synthesized catalyst varying from 1 to 5 wt.% catalyst loading, specified methanol to oil ratio from 5:1, 10:1, 15:1, 20:1, and 25:1, and a reaction time in the range from 1 h to 5 h using magnetic stirring at 900 rpm. The mixture was subsequently heated at a reflux temperature of 70 °C. After the completion of the transesterification–esterification, the mixture of the solid (catalyst) and liquid (mixture of biodiesel, glycerol water, and methanol) were collected and centrifuged in order to separate liquid products from the solid product. The excessive amount of methanol was removed by heating it at temperature 65–70 °C for 4 h in rotary evaporator and leaving the biodiesel and glycerol (two layers). The acid values of both the feedstock and the resulting product were established using a traditional titration technique, i.e., the standard AOAS Cd 3d-63 protocol [34]. The conversion of FFA was computed using Equation (1):

$$\mathrm{FFA}\mathrm{conversion}=\frac{A{V}_f-A{V}_p}{A{V}_f}\times 100$$

(1)

where AV_f and AV_p, indicate the feedstock and product acid values, respectively.

3.6. Mass Transfer Limitations and Mole Balance Catalyst Evaluation

The chemical kinetic and mass transfer limitations in this work were evaluated according to the following equations:

In a closed reaction system we have assumed that there is no inflow or outflow of material and that the reactor is well mixed. For most liquid-phase reactions, the density change with the reaction is usually small and can be neglected (i.e., V = V₀). Therefore, for constant-volume (V = V₀) batch reactors the mole balance will be:

$$\frac{1}{V}\left(\frac{d{N}_A}{dt}\right)=r_A$$

(2)

Assuming there is no change in the material quantity during the reaction

The mole balance can be rewritten in terms of concentration

$$\frac{1}{V} \frac{d{N}_A}{dt}=\frac{1}{V_0}\frac{d{N}_A}{dt}=\frac{d(N_A/V_0)}{dt}=\frac{d{C}_A}{dt}=r_A$$

(3)

When V₀ = V_t

$$-\frac{d{C}_A}{dt}=-r_A$$

(4)

The reaction rate for the starting material being consumed by the time the rate law can be written as

$$-r_A=k{C}_A^2$$

(5)

When we combine the rate low and the mole balance to obtain

$$-\frac{d{C}_A}{dt}=k{C}_A^2$$

(6)

And

$$-\frac{d{C}_A}{k{C}_A^2}=dt$$

(7)

Initially, C_A = C_A0 at t = 0. We integrate this equation to obtain the reactant concentration at any time t:

$$-\frac{1}{k}{{\displaystyle \int }}_{C_{A0}}^{C_A}\frac{d{C}_A}{C_A^2}={{\displaystyle \int }}_0^{t}dt$$

(8)

$$t=\frac{X}{k{C}_{A0}\left(1-X\right)}$$

(9)

where X is the conversion in percentage, t is the reaction time, k is the rate constant (s⁻¹), and C_A0 is the initial concentration in µmol·g⁻¹.

3.7. Catalyst Reusability and Leaching Test

The reusability of the catalyst in the absence of reactivation was assessed, together with its deactivation. For these tests, the following conditions were applied: Cu₂S-FeS/SiO₂ catalyst, 3 wt.%; reaction temperature, 70 °C; methanol:oil molar ratio, 15:1; duration, 2 h at 900 rpm.

Following the individual cycles, the catalyst recovered from the product admixture was rinsed with methanol (3 × 10 mL) and then hexane (3 × 10 mL) in order to extract any polar and non-polar substances from its surface [61]. For each reaction, an analysis of the products was performed in order to determine the transformation of FFA to PFAD methyl esters (Equation (1)).

Concurrently, 0.005 g catalyst was gathered from each reaction cycle in order to establish the sulphur content remaining in the catalyst as determined by CHNS. A sample of FAME was subjected to ICP analysis in order to assay the leached Cu and Fe ions.

4. Conclusions

In this study, a catalyst derived from silica was enhanced in order to optimise an esterification reaction from the synthesis of biodiesel from PFAD. The innovated silica-supported CuS-FeS catalyst demonstrated excellent surface area and thermal stability characteristics. Its efficacy in promoting the esterification reaction was ascribed to the increased active site acidity levels, which facilitated the interplay between methanol and FFA carbonyl moieties. In excess of 98% esterification transformation was achieved using a conversional reflux system with the following reaction parameters: reaction temperature, 70 °C; reaction duration, 180 min; Cu₂S_(8%)-FeS_(10%)/SiO₂ catalyst concentration, 2 wt.%; and methanol:PFAD molar ratio, 15:1 with high productivity rate of 2639.92 µmol·g⁻¹·min⁻¹ with k of 4.03 × 10⁻⁶ mole·S⁻¹. This catalyst represents a sound financial proposition, as it is produced from waste products; it has an excellent reusability profile with its potency retained over five reaction cycles, and can be extracted from the reaction products following centrifugation for 5 min at 3000 rpm.