Biodiesel Isomerization Using Sulfated Tin(IV) Oxide as a Superacid Catalyst to Improve Cold Flow Properties

Abstract

The development of alternative energies has become a concern for all countries to ensure domestic energy supply and provide environmental friendliness. One of the providential alternative energies is biodiesel. Biodiesel, commonly stated as fatty acid alkyl ester (FAAE), is a liquid fuel intended to substitute petroleum diesel. Nevertheless, implementation of pure biodiesel is not recommended for conventional diesel engines. It holds poor values of cold flow properties, as the effect of high saturated FAAE content contributes to this constraint. Several processes have been proposed to enhance cold flow properties of biodiesel, but this work focuses on the skeletal isomerization process. This process rearranges the skeletal carbon chain of straight-chain FAAE into branched isomeric products to lower the melting point, related to the good cold flow behavior. This method specifically requires an acid catalyst to elevate the isomerization reaction rate. And then, sulfated tin(IV) oxide emerged as a solid superacid catalyst due to its superiority in acidity. The results of biodiesel isomerization over this catalyst and its modification with iron had not satisfied the expectation of high isomerization yield and significant CFP improvement. However, they emphasized that the skeletal isomers demonstrated minimum impact on biodiesel oxidation stability. They also affirmed the role of an acid catalyst in the reaction mechanism in terms of protonation, isomerization, and deprotonation. Furthermore, the metal promotion was theoretically necessary to boost the catalytic activity of this material. It initiated the dehydrogenation of linear hydrocarbon before protonation and terminated the isomerization by hydrogenating the branched carbon chain after deprotonation. Finally, the overall findings indicated promising prospects for further enhancement of catalyst performance and reusability.

1. Introduction

Energy is critical for humans to carry out all their activities. Consequently, the rapid growth of global human residents over the past decade has forced the rising of energy demands [1]. Additionally, the prompt technological evolution in this era has changed world lifestyle and economic development, yielding a significant multiplication in energy requirement [2]. Nevertheless, the positive elevation in energy demand is not matched by its supply. The presence of existing energy sources, especially fossils, experiences a declining trend every year [3]. This situation strongly causes energy scarcity and unaffordability [4]. Moreover, the application of fossil fuels must be controlled to prevent negative impacts on humans and the environment because of its harmful emissions [5,6]. To respond to those conditions, the most adopted strategy is optimizing the utilization of fossil energy as well as developing alternative energies to ensure energy security and sovereignty. One of the providential alternative fuels is biodiesel.

Biodiesel, commonly stated as fatty acid esters, is synthesized from various fatty acid-rich feedstocks and short-chain alcohols by an esterification or transesterification process [7,8]. The fatty acid-rich feedstocks are increasing in variety and can be classified into five groups: safe-to-eat oils, inedible oils, animal fats, wastes, and microalgae-microorganisms [9,10,11,12,13]. Globally, total biodiesel production (Figure 1) has been recorded as increasing in the period of 2019–2024. Compared to 2019, the accumulated production in 2023 expanded more than 10 billion liters. Further, the European Union (EU), the United States (US), Indonesia, and Brazil were the four largest biodiesel producers in the world in the last five years [14]. In the future, they are predicted to still dominate the production and will gradually intensify the replacement of vegetable oils with low-cost feedstocks [15].

In application, this cleaner-combustion diesel is intended to reduce the dependence on petro-diesel [16]. Generally, biodiesel is mixed with petro-diesel at a certain volumetric percentage. Implementation of pure biodiesel or mixed fuel with high biodiesel concentration is not recommended for conventional diesel engines. Some of the considered reasons are (1) low oxidation stability [17]; (2) poor cold flow properties [18]; and (3) high rubber-dissolving properties [19]. These three points have significant negative impacts on engine performance. Based on the revealed problems, this paper is devoted to further discussing the obstacle(s) in cold flow properties (CFPs).

Cold flow properties (CFPs) are classified as physical properties measuring the fluidity characteristics of fuel in cold conditions [20]. The measurable parameters of biodiesel CFPs are cloud point, cold filter plugging point, and pour point. Cloud point (CP) is interpreted as the temperature at which the adhered solidified fuel particles during cooling begin to exhibit a cloudy emersion in the liquid bulk [20]. Cold filter plugging point (CFPP) is provided as the CFP parameter to measure the temperature at which biodiesel cannot flow properly by way of a specific filtration apparatus (4.5 × 10⁻⁵ m) for a finite time (max. 60 s) [21]. Furthermore, temperatures below the CFPP lead to massive solidification, causing loss of fluidity [22]. In that condition, the measured temperature is expressed as pour point (PP).

Basically, cold flow properties of biodiesel are specifically affected by saturation degree of fatty esters [23]. Saturated fatty ester (SFE) is regularly proven to have higher oxidation stability, but poorer cold flow properties than unsaturated fatty ester (UFE) with the same number of carbon atoms [24]. Moreover, the increasing quantity of double bonds in the carbon chain of fatty ester has the effect of decreasing oxidative stability and elevating cold fluidity characteristics [25]. As shown in Figure 2, the melting points of C-18 fatty acid methyl esters are going lower in line with alleviating saturation degree. In other words, the order of CFP from good to poor is methyl linolenate (C18:3), methyl linoleate (C18:2), methyl oleate (C18:1-cis), and methyl stearate (C18:0). Nevertheless, the order of oxidation stability from good to poor is inversely proportional to cold flow properties.

Based on these characteristics, several processes have been promoted to improve the biodiesel CFPs, including chemical insertion, fractionation, alkyl modification, isomerization, and cracking. However, this article focuses on isomerization (specifically the branching process) because this process has minimal impact on oxidation stability. Based on Figure 3, the principal target is to modify straight-chain saturated fatty esters which have higher melting points into branched forms which have better CFPs [3]. At the same time, the oxidative stability of branched-chain saturated fatty esters does not change significantly. Moreover, they have only a slight effect on cetane number and viscosity issues [25,28].

Generally, hydrocarbon branching isomerization requires an acid catalyst to accelerate the reaction [29]. To date, the isomerization catalysts were preferred in solid form. This was because they have the advantage of having a larger surface area, are easier to separate, and are more able to be reactivated than the homogeneous ones [21,25]. Based on the acidity characteristic, the solid catalysts are divided into two categories: (i) solid acid catalyst; and (ii) solid superacid catalyst. According to the basis of catalyst compound, the solid acid catalysts can be classified into zeolite, clay, and alumina. The previous isomerization results using these catalysts for short- and long-chain hydrocarbons are tabulated in Table 1. According to Table 1, the skeletal isomerization was mostly carried out under high pressure at a temperature range of 160–310 °C. The longer the carbon chain required, the higher the reaction temperature to branch the carbon chain. For fatty acid or ester raw materials, the pressurized reactions were run at a temperature of 250–285 °C and catalyst loading of 0.5–5 wt% for up to 24 h. In this condition, the isomeric product yields were reported to vary up to 85.7%. The higher isomerization temperature promoted the hydrocarbon cracking process [30].

For the second catalyst type, the solid superacid catalyst is an evolution of the solid acid catalyst, which has very strong acidity and proper stability against thermal exposures [31]. The solid catalysts are classified as superacid when their acidity is stronger than pure sulfuric acid. In other word, the Hammet indicator (Ho) of a solid superacid catalyst must be less than or equal to −12 [32]. Arata et al. [33] has comprehensively studied the acid strength of various anion-modified metal oxides, such as sulfated zirconium oxide (SO₄/ZrO₂), tin oxide (SO₄/SnO₂), titanium oxide (SO₄/TiO₂), aluminum oxide (SO₄/Al₂O₃), and hafnium oxide (SO₄/HfO₂). The highest acid strength was shown by sulfated tin oxide (calcined at 550 °C) with an Ho value of −18 [33]. Then, the performance of metal oxide-based superacid catalysts for branching isomerization is summarized in Table 2. According to data in Table 2, this catalyst type successfully isomerized hydrocarbons, varying from short to long carbon chains. Generally, superacid catalysts with sulfate anions required lower reaction temperatures than tungstated catalysts. The sulfated catalyst tended to accelerate the cracking process when the reaction was conducted at a higher temperature [34]. Later, isomerization using this superacid catalyst could be carried out at lower temperature and atmospheric pressure.

Specific to sulfated tin oxide (STO), previous studies in Table 2 reported that this superacid catalyst demonstrated catalytic activity in isomerizing several short-chain hydrocarbons. Nevertheless, only a few studies have further explored the role of this catalyst to isomerize long-chain hydrocarbon, especially biodiesel. Therefore, this research aimed to synthesize sulfated tin oxide and examine it in a catalytic isomerization reaction for improvement of biodiesel cold flow properties. This reaction was atmospherically conducted at a temperature of 200 °C to prevent evaporation of fatty esters at higher temperatures. This reason was supported by previous research presented in Table 1 and Table 2 involving long-chain hydrocarbon feeds that temperatures higher than 200 °C were mostly applied for high-pressure isomerization reactions. Meanwhile, lower temperatures tended to decline catalyst activity [34]. Additionally, temperatures higher than 235 °C potentially led to hydrocarbon cracking [35]. Moreover, the metal modification, such as iron, in STO was also carried out to investigate the effect of metal in the reaction.

Table 1. Previous studies of solid acid catalysts for hydrocarbon skeletal isomerization.

Basis of Catalyst Compound	Catalyst	Feed	Product	Catalyst Loading (wt%)	Temp. (°C)	Pressure (bar)	Time (h)	Yield (wt%)	Ref.
Zeolite	Ferrierite	Oleic and linoleic acid	Iso-stearic acid	5.0	260	9.3	8	77.3	[36]
	Ferrierite	Oleic and linoleic acid	Iso-stearic acid	5.0	260	9.3	24	83.1	[37]
	Ferrierite	Oleic and linoleic acid	Iso-stearic acid	5.0	260	5.5	4	85.7	[38]
	Mordenite	Oleic and linoleic acid	Iso-stearic acid	5.0	250	~13.8	6	70.2	[39]
	β-zeolite	Oleic and linoleic acid	Iso-stearic acid	5.0	260	9.3	24	70.5	[37]
	Pt/β-zeolite	Methyl palmitate	Methyl Iso-palmitate	0.5	285	40	16	±16.0	[40]
	ZSM-5	Oleic and linoleic acid	Iso-stearic acid	5.0	260	9.3	24	81.8	[37]
Clay	Clay	Oleic acid	Iso-stearic acid	4.3	250	~1.4	2.5	30.8	[41]
	Montmorillonite	Oleic acid	Iso-stearic acid	5.0	250	5.0	4	46.0	[42]
	Bentonite	Methyl oleate	Methyl iso-stearate	5.0	250	~13.8	24	29.6	[43]
Alumina (Al2O3)	Pt/γ-Al2O3	n-hexane	2-methyl pentane 3-methyl pentane	n.a.	180	20	2	89.7	[44]
	Pt/γ-Al2O3	n-heptane	2,2-dimethyl pentane 2,3-dimethyl pentane 3-ethyl pentane	n.a.	180	20	2	58.2	[44]
	Pt/γ-Al2O3	n-octane	2,3-dimethyl hexane 3,3-dimethyl hexane 3-ethyl hexane 3-methyl heptane	n.a.	180	20	2	41.5	[44]
	Pt/Cl/γ-Al2O3	n-hexane	2-methyl pentane 3-methyl pentane 2,3-dimethyl butane 2,2-dimethyl butane	WHSV = 1.8 h−1	160	30	0.56	66.2	[45]
	Pd/SiO2-Al2O3	n-hexadecane	iso-hexadecane	WHSV = 3 h−1	310	30	0.33	49.7	[46]

Table 1. Previous studies of solid acid catalysts for hydrocarbon skeletal isomerization.

Basis of Catalyst Compound	Catalyst	Feed	Product	Catalyst Loading (wt%)	Temp. (°C)	Pressure (bar)	Time (h)	Yield (wt%)	Ref.
Zeolite	Ferrierite	Oleic and linoleic acid	Iso-stearic acid	5.0	260	9.3	8	77.3	[36]
	Ferrierite	Oleic and linoleic acid	Iso-stearic acid	5.0	260	9.3	24	83.1	[37]
	Ferrierite	Oleic and linoleic acid	Iso-stearic acid	5.0	260	5.5	4	85.7	[38]
	Mordenite	Oleic and linoleic acid	Iso-stearic acid	5.0	250	~13.8	6	70.2	[39]
	β-zeolite	Oleic and linoleic acid	Iso-stearic acid	5.0	260	9.3	24	70.5	[37]
	Pt/β-zeolite	Methyl palmitate	Methyl Iso-palmitate	0.5	285	40	16	±16.0	[40]
	ZSM-5	Oleic and linoleic acid	Iso-stearic acid	5.0	260	9.3	24	81.8	[37]
Clay	Clay	Oleic acid	Iso-stearic acid	4.3	250	~1.4	2.5	30.8	[41]
	Montmorillonite	Oleic acid	Iso-stearic acid	5.0	250	5.0	4	46.0	[42]
	Bentonite	Methyl oleate	Methyl iso-stearate	5.0	250	~13.8	24	29.6	[43]
Alumina (Al2O3)	Pt/γ-Al2O3	n-hexane	2-methyl pentane 3-methyl pentane	n.a.	180	20	2	89.7	[44]
	Pt/γ-Al2O3	n-heptane	2,2-dimethyl pentane 2,3-dimethyl pentane 3-ethyl pentane	n.a.	180	20	2	58.2	[44]
	Pt/γ-Al2O3	n-octane	2,3-dimethyl hexane 3,3-dimethyl hexane 3-ethyl hexane 3-methyl heptane	n.a.	180	20	2	41.5	[44]
	Pt/Cl/γ-Al2O3	n-hexane	2-methyl pentane 3-methyl pentane 2,3-dimethyl butane 2,2-dimethyl butane	WHSV = 1.8 h−1	160	30	0.56	66.2	[45]
	Pd/SiO2-Al2O3	n-hexadecane	iso-hexadecane	WHSV = 3 h−1	310	30	0.33	49.7	[46]

Table 2. Previous studies of solid superacid catalysts for hydrocarbon skeletal isomerization.

Basis of Catalyst Compound	Catalyst	Feed	Product	Catalyst Loading (wt%)	Temp. (°C)	Pressure (bar)	Time (h)	Yield (wt%)	Ref.
Zirconium oxide (ZrO2)	SO4/ZrO2	Methyl oleate	Methyl iso-stearate	20	250	4.4	8	36	[47]
	Pt/SO4/ZrO2	n-heptane	3-methyl hexane 2,2-dimethyl pentane 2,3-dimethyl pentane	WHSV = 0.15 h−1	250	1.01	6.7	24.7	[34]
	Pt/SO4/ZrO2	n-hexadecane	iso-hexadecane	WHSV = 18.4 h−1	225	20.3	6	0.2	[48]
	Pt/SO4/ZrO2	n-hexadecane	iso-hexadecane	16.7	160	25.8	1	8.9	[49]
	Pt/WO3/ZrO2	n-hexadecane	iso-hexadecane	WHSV = 18.4 h−1	225	20.3	6	4.1	[48]
	Pd/WO3/ZrO2	n-hexadecane	iso-hexadecane	WHSV = 1 h−1	300	21.7	6	17.5	[50]
	Ni/WO3/ZrO2	n-hexadecane	iso-hexadecane	WHSV = 1 h−1	300	21.7	6	31.7	[50]
Tin oxide (SnO2)	SO4/SnO2	n-butane	iso-butane	0.05 g	110	1.01	0.02	39.2–44.8	[51]
	SO4/SnO2	n-pentane	2-methyl butane	0.8 g	110	0.07	3	8.0	[52]
	WO3/SnO2	cycloheptane	Methyl cyclohexane	WHSV = 5.2 h−1	140	1.01	0.19	7.6	[51]
Titanium oxide (TiO2)	SO4/TiO2	n-hexane	2-methyl pentane 3-methyl pentane	WHSV = 0.5 h−1	350	1.01	1.5	0.74	[53]
	SO4/TiO2	n-hexane	2-methyl pentane 3-methyl pentane	5.7 × 103 g.min/mol	100	n.a.	0.83	19	[54]
	Pt/SO4/TiO2	n-hexane	2-methyl pentane 3-methyl pentane	WHSV = 0.5 h−1	350	1.01	2	15.3	[53]
Aluminum oxide (Al2O3)	SO4/Al2O3	n-pentane	iso-pentane	0.8 g	50	0.07	3	3.7	[55]
Hafnium oxide (HfO2)	Pt/SO4/HfO2	n-hexadecane	iso-hexadecane	WHSV = 2 h−1	210	21.6	0.5	6.5	[56]
	Pt/WO3/HfO2	n-hexadecane	iso-hexadecane	WHSV = 0.5 h−1	300	21.6	2	65.1	[56]

Table 2. Previous studies of solid superacid catalysts for hydrocarbon skeletal isomerization.

Basis of Catalyst Compound	Catalyst	Feed	Product	Catalyst Loading (wt%)	Temp. (°C)	Pressure (bar)	Time (h)	Yield (wt%)	Ref.
Zirconium oxide (ZrO2)	SO4/ZrO2	Methyl oleate	Methyl iso-stearate	20	250	4.4	8	36	[47]
	Pt/SO4/ZrO2	n-heptane	3-methyl hexane 2,2-dimethyl pentane 2,3-dimethyl pentane	WHSV = 0.15 h−1	250	1.01	6.7	24.7	[34]
	Pt/SO4/ZrO2	n-hexadecane	iso-hexadecane	WHSV = 18.4 h−1	225	20.3	6	0.2	[48]
	Pt/SO4/ZrO2	n-hexadecane	iso-hexadecane	16.7	160	25.8	1	8.9	[49]
	Pt/WO3/ZrO2	n-hexadecane	iso-hexadecane	WHSV = 18.4 h−1	225	20.3	6	4.1	[48]
	Pd/WO3/ZrO2	n-hexadecane	iso-hexadecane	WHSV = 1 h−1	300	21.7	6	17.5	[50]
	Ni/WO3/ZrO2	n-hexadecane	iso-hexadecane	WHSV = 1 h−1	300	21.7	6	31.7	[50]
Tin oxide (SnO2)	SO4/SnO2	n-butane	iso-butane	0.05 g	110	1.01	0.02	39.2–44.8	[51]
	SO4/SnO2	n-pentane	2-methyl butane	0.8 g	110	0.07	3	8.0	[52]
	WO3/SnO2	cycloheptane	Methyl cyclohexane	WHSV = 5.2 h−1	140	1.01	0.19	7.6	[51]
Titanium oxide (TiO2)	SO4/TiO2	n-hexane	2-methyl pentane 3-methyl pentane	WHSV = 0.5 h−1	350	1.01	1.5	0.74	[53]
	SO4/TiO2	n-hexane	2-methyl pentane 3-methyl pentane	5.7 × 103 g.min/mol	100	n.a.	0.83	19	[54]
	Pt/SO4/TiO2	n-hexane	2-methyl pentane 3-methyl pentane	WHSV = 0.5 h−1	350	1.01	2	15.3	[53]
Aluminum oxide (Al2O3)	SO4/Al2O3	n-pentane	iso-pentane	0.8 g	50	0.07	3	3.7	[55]
Hafnium oxide (HfO2)	Pt/SO4/HfO2	n-hexadecane	iso-hexadecane	WHSV = 2 h−1	210	21.6	0.5	6.5	[56]
	Pt/WO3/HfO2	n-hexadecane	iso-hexadecane	WHSV = 0.5 h−1	300	21.6	2	65.1	[56]

2. Materials and Methods

2.1. Materials

The chemicals used in this research were tin tetrachloride 98% (Sigma-Aldrich Co., St. Louis, MO, USA), ammonia solution 32% (Merck KGaA, Darmstadt, Germany), ammonium acetate ≥ 98% (Smart Lab Indonesia Co., Tangerang, Indonesia), sulfuric acid 95–97% (Merck KGaA, Darmstadt, Germany), iron(III) nitrate nonahydrate ≥ 98% (Loba Chemie Pvt., Mumbai, India), n-hexane (Merck KGaA, Darmstadt, Germany), sodium hydroxide pellets (Merck KGaA, Darmstadt, Germany), and sodium chloride (Merck KGaA, Darmstadt, Germany). For quantitative analysis of the biodiesel component, the major analytical standard components used in this study were Me-palmitate (>99%), Me-stearate (>99%), Me-oleate (>99%), Me-linoleate (>99%), Me-isopalmitate (>98%), and Me-isostearate (>98%). These standards were purchased from Larodan AB Company (Solna, Sweden). Then, palm–biodiesel feedstock (Me-palmitate = 46.85 wt%; Me-stearate = 5.8 wt%; Me-oleate = 40.86 wt%; Me-linoleate = 5.5 wt%) was supplied by Sinarmas Bio Energy Co., Ltd. (Jakarta, Indonesia).

2.2. Methods

2.2.1. Catalyst Synthesis

Generally, STO was synthesized with the procedure reported by Matsuhashi et al. [52]. It was carried out through two sequential process stages: (1) tin oxide (TO) preparation; and (2) tin oxide sulfation. After sulfation in 3M sulfuric acid, the sulfate-modified solid was then dried at 105 °C for 2 h and calcined at 500 °C for 3 h. The final product was STO. For iron-modified STO (Fe-STO) catalyst, metal impregnation was conducted in the first stage. The iron(III) nitrate nonahydrate (metal precursor) was agitated with tin tetrachloride at stoichiometric iron to a total solid percentage of 3.6%wt in the hydrolysis of tin tetrachloride (Stage 1). Thus, the synthesized solid was sulfated (Stage 2) and then calcined as conducted for the STO catalyst.

2.2.2. Catalyst Characterization

The synthesized TO, STO, and Fe-STO were characterized to evaluate several characteristics of the catalyst, such as thermal decomposition, crystal structure, functional group, component, solid surface textural, and acid sites. Thermal decomposition examination of STO (25 mg) was conducted in air medium at atmospheric pressure using a Linseis STA PT-1600 thermal analyzer (Linseis Messgeraete GmbH, Selb, Germany). This analysis was executed at a temperature range of 30–1000 °C with a heating rate of 10 °C/min. Next, the crystal structure of all samples calcined at 500 °C was analyzed using a Bruker D2 Phaser X-ray Diffractometer (Bruker AXS, Karlsruhe, Germany). The scanning process was carried out with a radiation of Cu-Kα (λ = 1.5406 Å) and in the diffraction angle (2θ) range of 10–80°.

The Fourier Transform Infra-Red (FTIR) analysis was also applied to collect the transmittance spectra of the calcined solid materials. The equipment used in this analysis was a Bruker Spectrometer, Model Alpha with Platinum-ATR (Bruker Optics, Ettlingen, Germany). This analysis was run between wavelengths of 500 and 2000 cm⁻¹. Then, the composition analysis of the prepared solid materials was conducted using Rigaku XRF Equipment, Model NEXCG (Applied Rigaku Technologies, Inc., Cedar Park, TX, USA). Meanwhile, the surface properties of all calcined samples were characterized by the Brunauer–Emmett–Teller (BET) method. The BET measurement equipment was a Micromeritics TriStar II Plus (Micromeritics Instrument Corporation, Norcross, GA, USA), which was operated with isothermal nitrogen at −195.8 °C. In addition, the surface morphology and elemental mapping of the best-performing catalyst was characterized using a SEM-EDS Hitachi SU-3500 (Hitachi High-Tech, Tokyo, Japan).

Later, the acidity of all calcined materials was tested using the potentiometric titration and NH₃-Temperature Programmed Desorption (TPD) methods. For the potentiometric titration method, the titrant used in this analysis was sodium hydroxide (NaOH) with a concentration of 0.05 M in the 4% sodium chloride solution. The titration steps and acid site determination of this test followed the working procedure and calculation described by Yu et al. [57]. For the NH₃-TPD method, the acidity quantification of the samples was carried out using Microtrac MRB Equipment, Model Belcat II (MicrotracBEL Corp., Osaka, Japan). This analysis was operated using helium gas at a flow rate of 30 cm³-STP/min and heating rate of 20 °C/min.

2.2.3. Isomerization Reaction

The catalytic performance of the prepared catalysts (STO and Fe-STO) was investigated in the biodiesel isomerization. This reaction was run at a temperature of 200 °C, atmospheric pressure, stirring rate of 900 rpm, residence time of 12 h, and catalyst loading of 10 wt%. After reaction, the biodiesel product was separated from the solid catalyst. The isomerized biodiesel was then tested for its pour point (PP) and cloud point (CP) using a Phase Technology PCA-70Xi Analyzer (Phase Analyzer Company, Ltd., BC, Canada). This analyzer used an ASTM-D5949 method for PP testing and an ASTM-D5773 method for CP evaluation. Next, the investigated CP data were used to predict CFPP through Equation (1) [58].

$$\mathrm{C}\mathrm{F}\mathrm{P}\mathrm{P}=1.0\left(\mathrm{C}\mathrm{P}\right)-4.5$$

(1)

Afterwards, the composition analysis for the isomerized biodiesel was quantitatively performed using Perkin Elmer Claurus 600 Gas Chromatograph (GC) with Perkin Elmer Claurus 600T Mass Spectrometer (MS) and a Perkin Elmer Elite-5ms Capillary Column (length = 30 m; inner diameter = 0.25 mm; film thickness = 0.25 μm; max. temperature = 350 °C). This instrument set was manufactured by PerkinElmer, Inc. (Shelton, CT, USA). The component separation in the column was run at a temperature range of 180–260 °C, heating rate of 5 °C/min, injection temperature of 250 °C, and split ratio of 90:1. Based on the composition data, the yield (Y) of fatty ester products was quantified using Equation (2). In this equation, m_P,A and m_P,B were the mass of products after and before reaction (g), respectively. Then, m_R,B referred to the mass of reactants before reaction (g).

$$\mathrm{Y}\left(\%\right)={\displaystyle \frac{\left({\mathrm{m}}_{\mathrm{P},\mathrm{A}}-{\mathrm{m}}_{\mathrm{P},\mathrm{B}}\right)}{{\mathrm{m}}_{\mathrm{R},\mathrm{B}}}}\times 100\%$$

(2)

Moreover, the data series of component composition, NH₃-TPD acidity (A_TPD, mmol/g), and BET surface area (S_BET, m²/g) were applied to compute turnover frequency (TOF, s⁻¹) using Equations (3)–(5) and isomerization selectivity (S_isomer, %) using Equation (6). In these equations, R_reaction and N_s denoted the reaction rate and the mole of active sites per catalyst surface area, respectively. N_p,B and N_p,A symbolized the moles of products before and after reaction (mmol), respectively. Then, m_cat and t_reaction referred to the mass of catalyst loading (g) and the total reaction time (s), respectively.

$$\mathrm{T}\mathrm{O}\mathrm{F}\left({\mathrm{s}}^{-1}\right)={\displaystyle \frac{{\mathrm{R}}_{\mathrm{r}\mathrm{e}\mathrm{a}\mathrm{c}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}}}{{\mathrm{N}}_{\mathrm{s}}}}$$

(3)

$${\mathrm{R}}_{\mathrm{r}\mathrm{e}\mathrm{a}\mathrm{c}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}}\left({\displaystyle \frac{\mathrm{m}\mathrm{m}\mathrm{o}\mathrm{l}}{{\mathrm{m}}^2·\mathrm{s}}}\right)={\displaystyle \frac{\sum \left({\mathrm{N}}_{\mathrm{p},\mathrm{A}}-{\mathrm{N}}_{\mathrm{p},\mathrm{B}}\right)}{{\mathrm{S}}_{\mathrm{B}\mathrm{E}\mathrm{T}}\times {\mathrm{m}}_{\mathrm{c}\mathrm{a}\mathrm{t}}\times {\mathrm{t}}_{\mathrm{r}\mathrm{e}\mathrm{a}\mathrm{c}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}}}}$$

(4)

$${\mathrm{N}}_{\mathrm{s}}\left({\displaystyle \frac{\mathrm{m}\mathrm{m}\mathrm{o}\mathrm{l}}{{\mathrm{m}}^2}}\right)={\displaystyle \frac{{\mathrm{A}}_{\mathrm{T}\mathrm{P}\mathrm{D}}}{{\mathrm{S}}_{\mathrm{B}\mathrm{E}\mathrm{T}}}}$$

(5)

$${\mathrm{S}}_{\mathrm{i}\mathrm{s}\mathrm{o}\mathrm{m}\mathrm{e}\mathrm{r}}\left(\%\right)={\displaystyle \frac{{\mathrm{T}\mathrm{O}\mathrm{F}}_{\mathrm{i}\mathrm{s}\mathrm{o}\mathrm{m}\mathrm{e}\mathrm{r}}}{{\mathrm{T}\mathrm{O}\mathrm{F}}_{\mathrm{o}\mathrm{v}\mathrm{e}\mathrm{r}\mathrm{a}\mathrm{l}\mathrm{l}}}}$$

(6)

Lastly, the biodiesel feed and products were also tested for their oxidation stability by measuring the induction period. The equipment used for this investigation was a Metrohm 873 Biodiesel Rancimat Oxidative Stability Analyzer (Metrohm AG, Herisau, Switzerland). This analysis was run at a temperature of 110 °C and gas flow of 10 L/h. In addition, the measurement method was ensured to comply with international standards for biodiesel (ASTM-D6751 and EN-14112).

3. Results and Discussion

3.1. Catalyst Characterization

3.1.1. Thermal Decomposition Characterization

The thermal decomposition result in the temperature range of 25–1000 °C is illustrated in Figure 4. There were four endothermic decomposition stages detected on the TGA–DTA curve. The first stage was an endothermic curve in the range of 50–200 °C with a mass loss of 2.2 wt%, describing physical water desorption [59,60]. The initial curve of this stage demonstrated a mass gain due to water adsorption. At the beginning of heating, the surrounding water was evaporated and adsorbed onto the solid surface. As the solid temperature increased, the adsorbed water was vaporized again. This case also appeared in the research on sulfated WO₃/SnO₂ conducted by Alaya and Rabah [59]. The second stage occurred at 200–450 °C with a peak at 265 °C, indicating the release of volatile matters and the decomposition of residual sulfuric acid on the solid surface [51,59]. In this stage, the mass loss percentage was recorded at 4.2 wt%. At this temperature range, the DTA illustration graph did not display an exothermic peak point, which signified TO crystallization. It denoted that the sulfate was evenly distributed on the solid surface and inhibited the formation of TO crystal [59,61].

The third and fourth stages were identified at 450–540 and >540 °C, respectively. These stages represented sulfate decomposition [59], resulting in mass reductions of 3.8 and 7.3 wt%, successively. Overall, the total mass reduction of STO in the temperature range of 25–1000 °C was 17.5 wt%. These results implied that the calcination temperature exceeding 450 °C provided significant removal of sulfate and acid sites. As a result, the material may face challenges in maintaining the stability of catalytic performance and crystalline structure at high temperatures [62]. Consequently, more frequent reactivation may be necessary to sustain high catalytic activity for subsequent reactions. Therefore, it is advisable to calcine and utilize the STO catalysts below 450 °C.

3.1.2. X-Ray Diffraction (XRD)

The results of XRD analysis are demonstrated in Figure 5. The diffractogram of pure SnO₂ (TO) presented main intensity peaks at diffraction angles of 26.6, 33.8, and 51.8° with a crystallinity index of 72.8%. It indicated that the synthesized TO had crystal structure of tetragonal cassiterite. This pattern was also found in the pure SnO₂ (TO) reported by [59]. Then, sulfate and metal (iron) modification had almost no effect on the diffraction angles of the main peaks, representing no change in the SnO₂ crystal structure. However, these modifications hindered SnO₂ crystallization [60], so the curve peaks were wider and the crystallinity indexes (STO and Fe-STO) decreased sharply to 53–56%. Additionally, the peak of sulfate crystal was not detected in the diffractograms due to the complete dispersion of sulfate crystal on the solid surface [59].

3.1.3. Infra-Red Spectra

The results of infra-red spectra are shown in Figure 6. All investigated materials had vibration bands between 500 and 622 cm⁻¹. This band range was associated with the SnO₂ structure in the form of O–Sn–O and Sn–O [59,63]. Modification of sulfate and metal (iron) had an effect on the addition of four vibration band categories. The first category appeared in vibration bands between 890 and 980 cm⁻¹, which were related to the Sn–OH functional group [63]. This appearance was due to the presence of incompletely decomposed hydroxyl groups. The second category consisted of transmittance peaks at 1030–1040 cm⁻¹, assigned to symmetric S=O dan S–O [60]. The third additional functional group was asymmetric S=O dan S–O, proven by infra-red spectra bands at 1130–1160 cm⁻¹ [59,60]. The second and third functional groups were attributed to the dentate sulfate complex structures in the solid material [60]. The last category was the vibration bands between 1500 and 1650 cm⁻¹, denoting the stretching spectra of leftover water [63].

3.1.4. X-Ray Fluorescence (XRF)

The composition of major oxide compounds and elements is presented in

Table 3. Component composition of the synthesized materials (TO, STO, and Fe-STO).

Component	Composition, wt%
	TO	STO	Fe-STO
Oxide compound:
SnO2	93.4	90.5	85.6
SO3	0.16	5.04	6.75
Fe2O3	0.01	0.01	4.20
CaO	3.76	2.31	0.69
Element:
Sn	73.6	71.3	67.4
S	0.06	2.02	2.70
Fe	0.01	0.01	2.94
Ca	2.69	1.65	0.49

. Pure TO mostly contained SnO₂ and Sn elements with percentages of 93.4 and 73.6 wt%, respectively. The addition of sulfate progressively enhanced SO₃ content (5.04 wt%) in the solid (STO). Consequently, it reduced the SnO₂ percentage to 90.5 wt%. Moreover, metal (iron) impregnation impressively alleviated SnO₂ composition to 85.6 wt% in the material (Fe-STO). Apart from the elevation of metal (iron) content, the percentage of SO₃ component in Fe-STO (6.75 wt%) was also greater than STO. This phenomenon was due to the strengthening of sulfate binding by the impregnated metal on the solid surface. Therefore, it was hypothetically predicted to improve the acidity of the catalyst.

3.1.5. Brunauer–Emmett–Teller (BET)

The hysteresis curves of nitrogen adsorption–desorption are illustrated in Figure 7. The adsorption–desorption curves for all samples did not overlap because of pore obstruction. Based on IUPAC classification, these isotherm curves could be classified into Type IV, representing mesoporous solid materials [64]. This indication was also supported by the pore-blocking loops (Type H2) formed at relative pressures above 0.4 [59]. The sulfate and metal (iron) modification shifted the hysteresis loops to lower relative pressures, indicating smaller pore sizes. Based on the surface properties summarized in

Table 4. Solid surface properties of the synthesized materials (TO, STO, and Fe-STO).

Property	TO	STO	Fe-STO
BET surface area (m2/g)	43.5	142.3	157.7
Total pore volume (cm3/g)	0.1583	0.1242	0.1146
Mean pore diameter (nm)	14.6	3.5	2.9
BET constant	62	64	91

, STO and Fe-STO were proven to have smaller pore diameters than TO. Their pore diameters were recorded at 3.5 and 2.9 nm, respectively. These pore diameters were then compared to the molecular size of the targeted reactants and products.

In this study, the targeted reactants were methyl palmitate and methyl stearate contained in palm biodiesel. Later, the targeted products were their branched-chain isomers. For estimation, the molecular size of stearic acid was used to represent the largest size of these targeted components. As informed by Anwar and Garforth [25], the molecular size of stearic acid (C18:0) was reported at 2.45 × 0.25 nm. Hence, the pore diameters of STO and Fe-STO were larger than the reacting particle size. This condition ensured the pore diffusion of the reacting particles during the reaction. In this stage, the reactant diffuses through the pore from external surface into the active site and vice versa for the synthesized product. Hereafter, the surface area and BET constant of STO and Fe-STO were larger than the pure SnO₂ (TO). It implied that the larger surface area provided the larger place for reaction. Therefore, the combination of a larger surface area and appropriate pore diameter encouraged catalysts to have better catalytic performance.

3.1.6. Scanning Electron Microscopy-Energy Dispersive X-Ray Spectroscopy (SEM-EDS)

The material morphology of the finest-performing catalyst (Fe-STO) is demonstrated in Figure 8. The SEM image (Figure 8a) and the EDS elemental mapping (Figure 8b) indicate almost even dispersion of oxygen (O), sulfur, (S), and iron (Fe) elements on the surface. Additionally, the EDS elemental spectrum (Figure 8c) represents the weight percentages of the investigated elements. These results confirmed that the presence of sulfur (10 wt%) and oxygen (20 wt%) indicated the content of sulfate groups in the synthesized material. Sulfate group and solid tin oxide were found to form a dentate sulfate complex as an acid active site [29,60]. Subsequently, the appearance of iron (5 wt%) denoted that this metal was successfully impregnated on the solid surface as a metal active site. Theoretically, each category of these active sites has a specific role in the isomerization reaction mechanism. The details of the reaction mechanism and catalyst role are described in Section 3.3.

3.1.7. Acidity Measurement

The acid amounts of the investigated solids measured by the potentiometric titration and NH₃-TPD methods are presented in

Table 5. Acidity data of the investigated materials (TO, STO, and Fe-STO).

Material	Acid Amount (mmol/g)
	Potentiometric Titration	NH3-TPD
TO	0.19	0.47
STO	1.15	0.80
Fe-STO	1.63	1.34

. Through both methods, pure SnO₂ (TO) was noted to have acid amounts of 0.19 and 0.47 mmol/g, respectively. These values were higher than the acidity reported by Khalaf et al. [60] at 0.11 mmol/g. The sulfation significantly intensified acidity to 1.15 and 0.8 mmol/g, respectively. These results were also higher than the STO material studied by Khalaf et al. [60] at 0.4 mmol/g, SO₄/WO₃/SnO₂ explored by Alaya and Rabah [59] at 0.37 mmol/g, and SO₄/ZrO₂ synthesized by Chen et al. [65] at 0.16 mmol/g. And then, metal (iron) modification also elevated the acidity more than the basic sulfated SnO₂. The acidity values for Fe-STO (1.63 and 1.34 mmol/g) were more excellent than Fe/SO₄/ZrO₂ prepared by Chen et al. [65] at 0.3 mmol/g. This trend was positively correlated with the trend of SO₃ content resulted from the XRF analysis. Moreover, these results represented the superiority in acidity of the synthesized STO and Fe-STO.

3.2. Catalyst Performance Test

Performance investigation of the synthesized catalysts (STO and Fe-STO) was conducted in the batch biodiesel isomerization. The biodiesel feed and product were evaluated via CFPs investigation, composition analysis, TOF quantification, and oxidation stability measurement.

3.2.1. Cold Flow Properties (CFPs) Investigation

The CFPs investigation results of the isomerized biodiesel are tabulated in

Table 6. Cold flow properties data of the isomerized biodiesels and several standards (obtained from Pradana et al. [66]).

Catalyst	Biodiesel Cold Flow Properties
	PP (°C)	CP (°C)	CFPP (°C)
No catalyst	12	15.60 ± 0.10	11.1 ± 0.10
STO	11	16.45 ± 0.05	11.95 ± 0.05
Fe-STO	9	15.25 ± 0.05	10.75 ± 0.05
ASTM Standard (ASTM D-6751)	−15 ≤ PP ≤ 10	−3 ≤ CP ≤ 12	n.s.
EU Standard (EN14214)	n.s.	n.s.	(report)
Indonesian Standard (148.K/EK.05/DJE/2024)	n.s.	n.s.	≤15
Malaysian Standard (MS2008:2014)	n.s.	n.s.	≤15
Brazilian Standard (ANPN°920/2023)	n.s.	n.s.	(report)

Note: n.s. = not specified.

. The STO catalyst was known to slightly reduce biodiesel PP from 12 to 11 °C and increase CP from 15.60 ± 0.10 to 16.45 ± 0.05 °C. The more significant performance was shown by the Fe-STO catalyst. It alleviated both biodiesel PP and CP to 9 and 15.25 ± 0.05 °C, respectively. Compared to the American standard for pure biodiesel (ASTM-D6751) [66], the best PP in this study (9 °C) has complied with the standard value (−15 ≤ PP ≤ 10 °C). Nevertheless, the lowest CP value (15.2 °C) has not met this standard value (−3 ≤ CP ≤ 12 °C). For other standards (European Standard-EN14214, Indonesian Standard-148.K/EK.05/DJE/2024, Malaysian Standard-MS2008:2014, and Brazilian Standard-ANPN°920/2023) [66], there is no value limitation for PP and CP, specifically. As a replacement, these standards regulate biodiesel CFPP. For European and Brazilian Standards, biodiesel producers are only required to report the CFPP of their products. Meanwhile, Indonesian and Malaysian Standards define the maximum value of CFPP at 15 °C. According to Table 6, all calculated CFPP data complied with the standards. The findings suggest that the optimal biodiesel product identified in this study is suitable exclusively for regions that comply with EN-14214, Indonesian, Malaysian, and Brazilian standards. In simple terms, this biodiesel should be applied in areas with an environment temperature above 9 °C. Otherwise, the application of this product leads to fuel solidification, resulting in fuel line clogging [66]. Moreover, this slight improvement may reduce the formation of solidified particles in cold conditions, thus prolonging filter lifespan and lowering the risk of injector damage.

3.2.2. Composition Analysis

Afterwards, the composition analysis results of the isomerized biodiesel for C16-FAME are visualized in Figure 9. Based on this graph, the investigated catalysts (STO and Fe-STO) were proven to isomerize methyl palmitate into methyl iso-palmitate, but they were still too small. The yields of methyl iso-palmitate (based on methyl palmitate feed) for STO and Fe-STO were calculated at 0.2 ± 0.020 and 0.4 ± 0.028%, respectively. Based on these results, the higher catalyst acidity (Table 5) and surface area (Table 4) had not provided a significant effect in promoting a high yield of branched-chain fatty esters. Theoretically, the higher catalyst acidity and the larger surface area promoted more active sites for reaction, resulting in higher product yields. However, in the isomerization reaction, the activity of acid sites in isomerizing saturated carbon chains depended on the dehydrogenation stage at the beginning of the reaction [25,66]. Then, the dehydrogenation process commonly required a metal catalyst to accelerate the reaction [53]. It was proven in the Fe-STO-catalyzed reaction that the promotion of iron(III) further encouraged the dehydrogenation stage, presenting a higher isomerization yield than the STO catalyst. In other words, the presence of iron metal in the catalyst enhanced its activity in the fatty ester isomerization.

Nonetheless, the iron active sites were not yet effective enough in initiating the dehydrogenation process. According to electrode potential, iron(III) has a value of −0.04 V to form an iron component with zero oxidation number [67]. In this condition, iron(III) tends to act as a weak reducing agent, slightly hydrogenating unsaturated fatty esters into more saturated components. In addition, there were no specific oxidizing-agent metal active sites in the Fe-STO catalyst which strongly initiated the dehydrogenation stage. Because of this, the impregnated iron did not present excellent performance on isomerization. Therefore, the exploration of effective impregnated metals, either alone or in combination, still requires intensive attention to significantly improve catalyst activity at atmospheric pressure.

Then, the visualization of the biodiesel component for C18-FAME is presented in Figure 10. The examined catalysts (STO and Fe-STO) were detected to produce small amounts of methyl iso-stearate as the isomerization product of methyl stearate. The quantified yields for STO and Fe-STO were at 0.11 ± 0.003 and 0.11 ± 0.001%, respectively. The branched-chain saturated fatty esters led to the reduction in the PP of biodiesel products. Interestingly, the STO catalyst also tended to catalytically hydrogenate methyl linoleate into methyl oleate. Compared to FAME feedstock, the methyl linoleate composition in the STO-catalyzed biodiesel was reduced by 4.2 wt% and the methyl oleate content enlarged with a yield of 73%. This situation confirmed a slight elevation in the CP of biodiesel product when using an STO catalyst. A different phenomenon occurred in the isomerization reaction with the Fe-STO catalyst. The composition of methyl oleate and linoleate in the product altered insignificantly from the feedstock. For C18-FAME, iron metal modification was specifically a direct isomerization reaction to produce methyl iso-stearate. The weak reduction-oxidation properties of iron(III) promoted a minor effect on the hydrogenation–dehydrogenation process, apart from the absence of a hydrogen source in the reaction. Because of this composition, the isomerized biodiesel using the Fe-STO catalyst reduced PP and CP better than the STO catalyst.

For comparison, Zhang et al. [47] studied methyl oleate isomerization using a SO₄/ZrO₂ catalyst with the highest yield of 36%. This reaction was run at a temperature of 250 °C, pressure of 4.4 bar, catalyst loading of 20 wt%, and residence time of 8 h. Venkatesh et al. [49] reported the branching of n-hexadecane using Pt/SO₄/ZrO₂ with the best yield of 8.9%, obtained at a temperature of 160 °C, pressure of 25.8 bar, and catalyst loading of 16.7 wt% for 1 h. Busto et al. [48] prepared Pt/SO₄/ZrO₂ and Pt/WO₃/SO₄ to accelerate isomerization of n-hexadecane at a temperature of 225 °C and pressure of 20.3 bar for 6 h. The highest yields were 0.2 and 4.1%, respectively. Zhang et al. [50] manufactured Pd/WO₃/ZrO₂ and Ni/WO₃/ZrO₂ to catalyze alkane n-C₁₆H₃₄ with the isomerization yields of 17.5 and 31.7%, respectively. These yields were achieved at a temperature of 300 °C and pressure of 21.7 bar for 6 h. According to these previous studies using long-chain hydrocarbon feeds [47,48,49,50], the higher pressure and/or catalyst loading led to a higher yield than this work. The prepared STO and Fe-STO catalysts in this study only performed better when compared to the Pt/SO₄/ZrO₂ synthesized by Busto et al. [48]. It was due to the longer residence time in this research.

In addition, Hino et al. [51] tested the STO catalyst on the n-butane branching with yields of 39.2–44.8%, reacted at a temperature of 110 °C, atmospheric pressure, and catalyst loading of 0.05 g (reactor volume of 0.05 mL). Hino and coworkers [51] also modified TO with tungstate anion to isomerize cycloheptane with the highest yield of 7.6%. Matsuhashi and his team [52] published their working results on the n-pentane isomerization using an STO catalyst with the best yield of 8.0%, run under vacuum conditions at a temperature of 110 °C and reaction time of 3 h. Enriquez et al. [53] discussed n-hexane isomerization using SO₄/TiO₂ and Pt/SO₄/TiO₂ with yields of 0.74 and 15.3%, respectively. Both catalysts were performed at a temperature of 250 °C and atmospheric pressure. Matsuhashi et al. [55] worked on the vacuum isomerization of n-pentane using SO₄/Al₂O₃ with an isomerization yield of 3.7%. Based on these prior findings [51,52,53,55], isomerization of short-chain hydrocarbon presented higher yields at milder temperatures than this study. It implied that biodiesel isomerization required more energy and higher catalyst activity.

3.2.3. Turnover Frequency (TOF) Quantification

The correlation of catalyst performance, acidity, and surface area is quantitatively represented by catalyst turnover frequency (TOF). The calculation results of acid sites, reaction rates, TOFs, and isomerization selectivity for STO and Fe-STO catalysts are summarized in

Table 7. Calculated acid sites, reaction rates, turnover frequencies, and isomerization selectivity of the prepared catalysts (STO and Fe-STO).

Catalyst	No. of Acid Sites (mmol·m−2)	Reaction Rate (mmol·m−2·s−1)		Turnover Frequency (s−1)		Isomerization Selectivity (%)
		Overall	Isomerization	Overall	Isomerization
STO	5.6 × 10−3	2.7 × 10−7	5.5 × 10−9	4.9 × 10−5	1.0 × 10−6	2.0
Fe-STO	8.5 × 10−3	1.1 × 10−8	1.0 × 10−8	1.3 × 10−6	1.2 × 10−6	90.8

. The STO catalyst was recorded to have overall and isomerization TOFs of 4.9 × 10⁻⁵ and 1.0 × 10⁻⁶ s⁻¹, respectively. These values indicated that the STO catalyst had low selectivity for isomerization (2.0%). Later, the overall and isomerization TOFs for the Fe-STO catalyst were 1.3 × 10⁻⁶ and 1.2 × 10⁻⁶ s, respectively. These results denoted high isomerization selectivity at point of 90.8%. Besides that, the isomerization TOFs of the STO and Fe-STO catalysts were identified at the similar level, confirming the fundamental isomerization TOF for STO-based catalysts. The modification of metals, either alone or in combination, may alter the fundamental value of isomerization TOF, depending on the reduction-oxidation properties of metals.

The results also demonstrated that the increase in the acid sites and surface area had almost no effect on the catalyst performance for isomerization. It emphasized that the skeletal isomerization was strongly influenced by the metal catalyst initiating the dehydrogenation process in the first step reaction. And, iron(iii) has not properly enhanced the isomerization reaction rate. Hereafter, the TOF values of the STO-based catalyst in this study were still lower than the TOF values of TO and ZrO₂ catalysts for methanol redox reaction (~7.0 × 10⁻¹ and ~3.5 × 10⁻² s⁻¹, respectively) [68]. The previous work published by Badlani and Wachs [69] also reported the higher TOF values of TO and ZrO₂ catalysts for methanol oxidation reaction (1.9 × 10⁻¹ and 1.3 × 10⁻¹ s⁻¹, respectively). Even though the TOF values have not met expectation, these results implied an opportunity in the further development of STO-based isomerization catalysts.

3.2.4. Oxidation Stability Measurement

Furthermore, biodiesel feed and products were evaluated for their oxidation stability. As shown in

Table 8. Oxidation stability data of the isomerized biodiesels and several standards (obtained from Pradana et al. [66]).

Catalyst	Biodiesel Oxidation Stability (h)
No catalyst	18.0
STO	19.6
Fe-STO	17.9
ASTM Standard (ASTM D-6751)	≥3
EU Standard (EN14214)	≥8
Indonesian Standard (148.K/EK.05/DJE/2024)	≥11
Malaysian Standard (MS2008:2014)	≥10
Brazilian Standard (ANPN°920/2023)	≥13

, isomerization using the STO catalyst enhanced the oxidation stability of the biodiesel product. This result was in line with the notable increase in biodiesel saturation level, as depicted in Figure 10. The composition of methyl oleate (C18:1) increased, while the content of methyl linoleate (C18:2) declined. Based on data shown in Figure 2, the oxidation stability of methyl oleate and linoleate is 2.5 and 1 h, respectively. Thus, conversion of methyl linoleate into methyl oleate clearly encouraged improvement in oxidation stability. Subsequently, biodiesel products treated by Fe-STO run into a slight reduction in oxidation stability. This was predominantly due to the presence of branched-chain fatty esters. It affirmed data presented in Figure 3 that the skeletal isomerization products did not significantly affect biodiesel oxidation stability. This situation may not significantly interfere with the operation and storage of biodiesel. Additionally, all samples have complied with all established standards for biodiesel. Amidst the high CFP rule, the American standardization organization, through ASTM-D6751 standard, preferred to lower the limit of oxidation stability parameters (min. at 3 h) [66]. On the other hand, biodiesel standards in Europe, Indonesia, Malaysia, and Brazil have higher minimum limits of 8, 11, 10, and 13 h, respectively [66]. These standards are noted to have moderate regulation for CFP. The differences in the limits of each standard are intended to accommodate the certain application of biodiesel.

3.3. Theoretical Catalyst Role in the Reaction Mechanism

Generally, the acid-based catalyst has a principal role in the skeletal isomerization process [25]. Based on the previous studies [25,66,70] and proposed by this research, the theoretical catalyst role in the isomerization reaction is demonstrated in Figure 11. It performed protonation on the double bond site of unsaturated hydrocarbon to produce a saturated carbocation compound. Hereinafter, it positively forced the carbocation to undergo a process of carbon chain restructuring. Initially, it directed saturated carbocation to form a protonated intermediate compound. Next, the protonated carbon chain in the intermediate format was branched. In this case, the branching usually occurred at the end of the carbon chain. The branched-chain carbocation was then deprotonated, establishing unsaturated branched-chain hydrocarbon. Figure 11 also visualizes the duties of metal promotion in the catalyst for the reaction mechanism. Firstly, the promoted metal had the function of dehydrogenating saturated carbon chains and creating unstable double bond(s) in hydrocarbon. This step was necessary at the beginning, so that the three acid-catalyzed steps (protonation, branching, and deprotonation) could proceed. Later, the impregnated metal was also aimed to again hydrogenate the unsaturated branched-chain hydrocarbon after deprotonation. For simple understanding, the promoted metal had three sequential purposes: (i) taking hydrogen from straight-chain hydrocarbon; (ii) storing it during the branching process; and (iii) returning it again to the branched-chain hydrocarbon. Therefore, the presence of metal is very useful because it significantly enhanced the catalyst activity in the isomerization reaction.

4. Conclusions

This study focused on the skeletal isomerization process to improve biodiesel cold flow properties. This process was reported to require an acid catalyst to facilitate the reaction. Then, STO-based materials appeared as potential acid catalysts due to their excellence in acid strength. The isomerization of biodiesel over STO and Fe-STO catalysts had not met the target for high isomerization yield and remarkable CFPs improvement. However, it proved that the skeletal isomers offered minimum impact on biodiesel oxidation stability. It also confirmed the role of an acid catalyst in the reaction mechanism in terms of protonation, isomerization, and deprotonation. Hereinafter, the metal impregnation was theoretically required to enhance catalyst activity. It initiated the isomerization by dehydrogenating the straight-chain hydrocarbon and terminated the reaction by hydrogenating the branched-chain fatty ester. Finally, this result indicated promising prospect for further performance enhancement and reusability advancement of STO-based isomerization catalyst.