An organic–inorganic hybrid based on an Anderson-type polyoxometalate immobilized on PVA as a reusable and efficient nanocatalyst for oxidative desulphurization of gasoline

Abstract

In the present work, an organic–inorganic hybrid based on an Anderson-type polyoxometalate ((C₄H₉)₄N)₆Mo₇O₂₄ was synthesized and immobilized on poly vinyl alcohol (PVA) via a sol–gel method. This synthesized nanocomposite was characterized by FT-IR, XRD, SEM and UV-vis spectroscopy. The catalytic activity of this hybrid nanocomposite was tested on oxidative desulphurization (ODS) of model sulfur compounds and actual gasoline. The ODS was performed using different polyoxometalates as catalyst and CH₃COOH/H₂O₂ as oxidant. The experimental results show that the supported polyoxometalate ((C₄H₉)₄N)₆Mo₇O₂₄/PVA catalyst is more active than the unsupported polyoxometalate catalysts. The oxidation reactivity of the catalysts depends on the type of countercation: ((C₄H₉)₄N)⁺ > NH₄⁺ > Na⁺. The sulfur-containing compounds, such as thiophene (Th), benzothiophene (BT) and dibenzothiophene (DBT) were oxidized to the corresponding sulfones. The reactivity order of the model sulfur compounds is: DBT > BT > Th. The effects of the reaction time, reaction temperature, dosage and nature of catalyst and concentration of hydrogen peroxide were investigated. The ((C₄H₉)₄N)₆Mo₇O₂₄/PVA also shows high selectivity for DBT oxidation desulphurization. The selective desulfurization ratio reaches 94% with ((C₄H₉)₄N)₆Mo₇O₂₄/PVA under the reaction conditions of 40 °C, 2 h. The kinetic parameters of the oxidation of model sulfur compounds were studied. The oxidation rate was interpreted with first-order kinetics. The advantages of this method lie in its mild conditions, low cost, large scale, simplicity and environmentally friendly route.

---

1. Introduction

Polyoxometalates (POMs) as metal–oxygen anionic clusters can undergo a reversible reduction without changing their structure.¹ POMs are usually synthesized via condensation reactions in aqueous, acidic solution. These reactions can be influenced by careful variation of the synthesis conditions, e.g. ratio and concentration of reagents, solvent, pH, counter cations, and temperature.^2,3 They have numerous applications in a multitude of areas ranging from catalysis to pharmacology and remain of continuous interest, especially in the areas of design and synthesis of functional nanomaterials.^2–5 The application of POMs as catalytic materials is increasing continuously in the catalytic field. Because of their high thermal stability, stronger acidity, high hydrolytic stability (pH 0–12), safety, quantity of waste, reusability, corrosiveness, separability, high oxidation potential and greenness, in this context, polyoxometalates are promising catalysts.^6–11 As an important member of POMs family, Anderson-type polyoxoanions exhibit fascinating planar structures with abundant terminal oxygen atoms showing high-reactivity, which have attracted considerable attention and been employed as a synthetic source to construct metal–organic complexes with novel structures and appealing properties.¹¹ The utility of POMs for industrial application has been limited, notably due to their tendency toward low surface area (1–10 m² g⁻¹). So, many attempts have been performed to immobilize POM on various high surface area solid materials. The immobilized POM catalysts are important since they can easily be recovered from reaction mixtures and reused. Therefore, researchers have focused on immobilizing POMs onto different acidic or neutral materials such as TiO₂, zeolites, polymers and polyaniline–graphene composite.^12–17 The environmental provisions on sulfur contents in fuels became increasingly intense all around the world, which is due to increasingly strict regulations on sulfur concentration of transport fuel.¹⁸ Sulfur-containing compounds in fuel are removed by hydrodesulphurization (HDS) in petroleum refinery industry. However HDS shows high efficiency in removing aliphatic and acyclic sulfur containing compounds, but less effective for remove aromatic sulfur-containing compounds such as benzothiophene (BT), dibenzothiophene (DBT), 4,6-dimethyldibenzothiophene (4,6-DMDBT) and their derivatives because of the high steric hindrance of these compounds.^19,20 Thus, alternative desulphurization process like adsorption, biodesulphurization, extraction, oxidation, and photo-oxidation are used to eliminate remained sulfur compounds of hydrodesulphurized oil.^18–22 One of the most hopeful methods is oxidative desulphurization (ODS). Different catalytic systems, such as organic acid, heteropolyoxometalates, ionic liquid, molecular sieve, and photocatalysts have been reported for oxidative desulphurization operations (ODS).^18–24 Among all of these systems, polyoxometalates show higher sulfur removal efficiency because of their unrivaled chemical properties of polyoxometalates properties for ODS process owing to the higher structural stability.^12–17 In the ODS process, organic sulfides are converted into their corresponding sulfoxides and sulfones, that are preferentially extracted due to their increased polarities.^12–14 One of the most serious challenges to remove sulfur compounds in fuel in this years that need catalytic systems with high activity, selectivity and good stability to reduce significantly the sulfur content of the gasoline. A literature review showed that desulphurization of gasoline using the immobilized Anderson-type polyoxometalate ((C₄H₉)₄N)₆Mo₇O₂₄ on poly vinyl alcohol (PVA) was not studied.^14–24 In continuation of our group's research on the syntheses and application of polyoxometalates in organic reactions,^25–31 in this manuscript, the ((C₄H₉)₄N)₆Mo₇O₂₄ was synthesized as an Anderson-type isopolyoxometalate (IPOM) and immobilized on PVA. The synthesized nanocomposite (IPOM–PVA) was used as a nanocatalyst for the oxidative desulphurization of thiophene (Th), benzothiophene (BT) and actual gasoline. At present work, we demonstrate that the nanocomposite IPOM–PVA as a nanocatalyst can afford an excellent activity in the oxidative desulphurization of gasoline or simulated gasoline under mild conditions with CH₃COOH/H₂O₂ as oxidant. This quaternary ammonium Anderson-type isopolyoxometalate (IPOM), which has lipophilic cation can act as phase transferagent and transfer the peroxometal anion into organic phase. This nanocomposite (IPOM–PVA) was characterized by X-ray diffraction (XRD), Fourier transform infrared spectra (FT-IR), Scanning Electron Microscopy (SEM) and UV-vis spectroscopy.

2. Experimental

2.1. Materials

All solvents and reagents were commercially available and used without further purification. Model compounds and chemicals, including Th, BT and DBT and solvent (n-heptane) for experiments and analysis and hydrogen peroxide (30 vol%) were obtained from Aldrich Chemical Company. (NH₄)₆Mo₇O₂₄ was purchased from Merck Chemical Company. Several heteropolyoxometalate catalysts were prepared according to the literature procedure.^25,26 The products characterized by analytical instrument FT-IR, UV-vis, XRD and SEM. Typical actual gasoline (density 0.7987 g mL⁻¹ at 15 °C, total sulphur content 0.497 wt%) was used and details are shown in Table 1.

Table 1 Properties of actual gasoline

Entry	Properties of gasoline	Unit	Method	Before ODS	After ODS^a
a Condition for desulphurization: 50 mL of gasoline, 0.1 g nano catalyst, 3 mL oxidant, 10 mL of extraction solvent, time = 2 h, and temperature = 40 °C.b Abbreviation: API GR.; API gravity (API – American Petroleum Institute); API = 141.5/specific gravity – 131.5.
1	Specific gravity	°F	ASTM D1298	0.7991	0.7988
2	Density@15 °C	g mL^−1	ASTM D1298	0.7987	0.7984
3	API GR.^b	°F	Calculated^b	45.57	45.64
4	Water content	Vol%	ASTM D4006	Nil	Nil
5	Total sulfur content	Wt%	ASTM D4294	0.497	0.015
6	Mercaptans	ppm	ASTM D3227	87	5
7	Distillation		ASTM D86	—	—
8	I. B. P.	°C		51.4	51.2
9	F. B. P.			209.6	209.3
10	10	Vol%		70.1	70.0
11	20			116.8	116.4
12	50			190.3	190.1
13	90			207.2	206.8
14	Residue			1.1	1.1
15	Loss			1	1
16	Recovery			97.9	97.9

---

2.2. Preparation of catalysts

2.2.1. Synthesis of tetrabutyl ammonium Anderson-type isopolyoxometalate (IPOM). The [(C₄H₉)₄N]₆Mo₇O₂₄ was prepared as following: under stirring at room temperature, an ethanol solution of tetrabutyl ammonium bromide (6 mmol) was dropwise into an aqueous solution of (NH₄)₆Mo₇O₂₄ (1 mmol). Immediately, a snow white precipitate was formed. After continuous stirring for 4 h, the formed solid was filtered, washed and dried at 60 °C in a vacuum for 24 h to obtain [(C₄H₉)₄N]₆Mo₇O₂₄ and the resultant solid is designated as IPOM.

2.2.2. Preparation of nanocatalyst by immobilization of IPOM on PVA. The preparation of the nanocatalysts has been conducted according to a sol–gel procedure. In a typical synthesis, PVA (0.5 g) was added slowly to boiling (100 °C) distilled water (40 mL). When solution temperature was 65 °C then 0.5 g of ((C₄H₉)₄N)₆Mo₇O₂₄ was added slowly to the solution with vigorously stirred (700 rpm) at 65 °C for 2 hours. The white gel was observed and this gel was put in a drying oven at 50 °C for 2 h. After the reaction, the color of precipitate not changed (Scheme 1). The resultant solid ((C₄H₉)₄N)₆Mo₇O₂₄/PVA is designated as IPOM–PVA.

Scheme 1 Schematic description of the procedure to prepare IPOM–PVA.

2.3. Catalytic test

2.3.1. Oxidative desulphurization of model compound (simulated gasoline). The experiments of oxidative desulphurization were performed in a round bottom flask equipped with stirrer. For ODS of simulated gasoline, DBT, BT and thiophene was selected as model sulfur compound. To make a stock solution with a sulfur content of 500 ppm, DBT or BT was dissolved in n-heptane. In the typical run, the water bath was first heated and stabilized to a certain temperature (25–45 °C). Into the round bottom flask, 3 mL CH₃C(O)OOH (peracetic acid, acetic acid/hydrogen peroxide with mole ratio of 1 : 1) was added and mixed with 50 mL of the model sulfur compound (DBT or BT). To initiate the reaction, 0.1 g of the IPOM–PVA as a nano catalyst was added to the round bottom flask. The flask was immersed in a heating bath (25–45 °C) and stirred at 500 rpm for 2 h. The biphasic mixture was separated by decantation.

The DBT or BT content in both phases was determined. Product identification was achieved by using GC-MS (Varian cp-1200 quadruple MS), HP5890 Series II with 5972 Series MS detector. The results indicated a perfect match of the mass spectrum of the product with the standard dibenzothiophene or benzothiophene sulfone. Also the total sulfur concentration of the simulated gasoline before and after reaction was determined using a Tanaka Scientific RX-360 SH X-ray fluorescence spectrometer (ASTM D-4294 method). The concentrations of the model sulfur compounds (DBT or BT) in the treated model oil were determined from their peak areas in the GC-FID chromatograms using a calibration curve obtained with the peak areas of their standard concentrations. The change in concentration was calculated as conversion (%) using eqn (1), in which X₀ is initial concentration of BT, and X_t is the final concentration of BT after time t.

% conversion = (X₀ − X_t)/X₀ × 100% (1)

2.3.2. Oxidative desulphurization (ODS) of actual gasoline. For ODS of actual gasoline fuel, 50 mL of gasoline was added to two-necked round bottom flask. The temperature of solution was fixed at 40 °C. Then 0.1 g of nanocatalyst IPOM–PVA was added to the solution and strongly stirred by a magnetic stirrer. A mixture of CH₃COOH : H₂O₂ (3 mL) in ratio of 1 : 1 was added drop wise in 2 h, while it has been stirring vigorously. When the oxidation has been finished the mixture was cooled down to room temperature and then 10 mL of acetonitrile (MeCN) was added to extract the oxidized sulfur compounds. The acetonitrile/oil ratio used was 1/5 by volume. The biphasic mixture was separated by decantation. The oil phase was separated and weighed to calculate present of gasoline (for three times reaction: 97%, 96% and 95%). The total sulphur and mercaptan sulphur content in gasoline before and after reaction were determined using X-ray fluorescence (a TANAKA X-ray fluorescence spectrometer RX-360 SH ASTM D-4294) standard test method. This test method provides rapid and precise measurement of total sulfur in petroleum and petroleum products with a minimum of sample preparation. A typical analysis time is 1 to 5 min per sample. Results are showed in Table 2. The desulphurization efficiency of actual gasoline was calculated by the eqn (2), where TS₁ is the total sulfur content in actual gasoline fuel and TS₂ is the total sulfur content in treated gasoline fuel samples.

% desulphurization = (TS₁ − TS₂)/TS₁ × 100% (2)

Table 2 Effect of different catalysts in the ODS of gasoline, thiophene, DBT and BT^a

Entry	Catalyst	Conversion (%)
		DBT	BT	Th	Gasoline	Ref.
a Conditions for desulphurization: 50 mL of model gasoline (500 ppm S) or 50 mL of gasoline, 0.1 g catalyst, 3 mL acetic acid/H2O2, 10 mL extraction solvent, time 2 h and temperature 40 °C.
1	[(C4H9)4N]6Mo7O24/PVA	99	98	97	97	This work
2	(N(tBu)4)5PV2W10O40/PVA	98	98	98	97	28
3	[(C4H9)4N]6Mo7O24	92	91	90	91	This work
4	(N(tBu)4)5PV2W10O40	91	91	92	91	28
5	(N(tBu)4)5PV2Mo10O40	90	88	89	88	28
6	[(H4N)]6Mo7O24	87	85	84	82	This work
7	(NH4)10[P2W18Cd4]	88	81	80	82	27
8	Na6Mo7O24	85	84	82	81	This work
9	K10[P2W18Cd4]	85	80	78	81	27
10	PVA	35	33	32	31	28
11	[(C4H9)4N]Br	32	31	31	30	28
12	None	23	22	21	20	28

---

2.4. Characterization methods

Fourier transform infrared (FT-IR) spectra were recorded by Thermo-Nicolet-is 10 in the range of 400–4000 cm⁻¹. UV-visible spectra were studied with a double beam termo-heylos spectrophotometry in the range of 200–500 nm. X-ray diffraction (XRD) patterns was accomplished by D8 advance Bruker and radiation Cu kα (λ = 1.54 nm) in the range of 0–60° (2θ). Scanning electron microscope (SEM) images were obtained on a LEO 1455 VP with an accelerating voltage of 10.00 kV. ³¹P NMR spectra were recorded by Bruker Ultra-shield spectrometer.

3. Results and discussion

3.1. Catalyst characterization

FT-IR spectra of free IPOM, PVA, and IPOM–PVA are studied for structural characterization purposes (Fig. 1). In Fig. 1, FT-IR spectrum of pure PVA sample is showed. The O–H stretching of alcohol and residual water was detected at 3200–3600 cm⁻¹ that is linked to the stretching O–H from the intermolecular and intramolecular hydrogen bonds. The vibrational mode at 2800–3000 cm⁻¹ corresponds to the C–H stretching of alkyl groups and the peaks between 1750 and 1575 cm⁻¹ are due to the stretching C=O and C–O from acetate group remaining from PVA. CH₂ bending mode at 1435 cm⁻¹, and C–O stretching mode at 1096 cm⁻¹.^33–35 It has been accepted that the presence of water can be confirmed by the absorption band at 3490 cm⁻¹ associated to O–H stretching that at 1600 cm⁻¹ to H–O–H bending and part of the adsorption between 3000 and 3300 cm⁻¹ to bending movement in free water.³² For ((C₄H₉)₄N)₆Mo₇O₂₄ (Fig. 1(a)), characteristic peaks can be found at 570, 691, 833, 890, 923, 1401, 1628, 3100–3500 cm⁻¹. The pure IPOM reveals the characteristic bands of bridged Mo–O–Mo bonds in the molecule of polyoxometalate around 570 and 691 cm⁻¹. The bands at 923, 890 and 833 cm⁻¹ are attributed to the terminal group of Mo=O bond.^24,25 In addition, the other peaks at 1479, 2876 and 2976 cm⁻¹ belong to the vibrations of the quaternary ammonium cations. The characteristic absorption at 1401 cm⁻¹ can be assigned to bending (C–N–C). Based on them; IPOM–PVA has the characteristic peak at 3100–3500 cm⁻¹ associated to O–H and N–H stretching. H–O–H bending is at 1628 cm⁻¹. Fig. 1(c) shows FT-IR spectra of IPOM–PVA. The intensity of IPOM–PVA (peaks attribute to IPOM) is lower than pure IPOM and red shift (Table 3) that indicate IPOM is loaded on PVA.

Fig. 1 FT-IR of obtained powders. (a) IPOM (b) PVA and (c) IPOM–PVA.

Table 3 FT-IR related to IPOM and IPOM/PVA

Compound	ν Mo–O–Mo (cm^−1)	ν Mo–O (cm^−1)	ν H–N–H (cm^−1)	ν H–O–H (cm^−1)
IPOM	591.31	890.1, 923.43	1401.53	1628.72
	657.97
	691.30
IPOM/PVA	557.70	875.86, 1060.95	1423.72	1643.9
	617.70
	712.91

---

In Fig. 2 the reign of peaks are 208 nm for IPOM–PVA, PVA: 207 nm and IPOM: 200–255 nm. ((C₄H₉)₄N)₆Mo₇O₂₄ catalysts manifest very intense absorption in the range 200–330 nm spectral region that is the characteristic of polymer Mo–O–Mo structures originated by the charge transfer processes from O²⁻ to Mo⁶⁺ in octahedral coordination.^24–26

Fig. 2 UV-vis spectra of obtained powders. (a) PVA (b) IPOM–PVA and (c) IPOM.

The morphology of the resulting sample was investigated by SEM images. Fig. 3 shows the SEM images of the IPOM–PVA samples prepared in 2 h that include agglomerated nanoparticles which have formed plates or layers. From Fig. 3(a), the blank PVA film is observed to be relatively flat surface and Fig. 3(b) shows polyoxometalate consist of very small agglomerated nanoparticle.²⁸ Fig. 3(c) indicated water lily pond-like structure and shows accumulated nanocomposite that IPOM (with particles) loaded with highly dispersed on the PVA support. This picture (Fig. 3(c–f)) shows non regular morphology. Nanocomposite Anderson-type polyoxometalate modified PVA is created and the estimated particle sizes are achieved nano-scale.³⁵

Fig. 3 SEM images of a sample of IPOM–PVA. The images of (a) PVA film (3 μm), (b) IPOM (1 μm), (c) IPOM–PVA (10 μm), (d) IPOM–PVA (2 μm), (e) (IPOM–PVA) 1 μm and (f) IPOM–PVA (300 nm).

X-ray diffraction is a convenient method for determining the mean size of nano crystallites in nano crystalline bulk materials. The XRD patterns of IPOM–PVA metal were shown in Fig. 4 and collected in the scanning range 6° ≤ 2θ ≤ 60°. The pattern, observed for pure PVA, shows distinct crystalline sharp peak at 2θ = 20 degree, for pure IPOM 2θ = 10, 12–15, 17, 20, 23–30 and 35. The 2 theta peaks for IPOM–PVA 9–10, 12–15, 17–20, 25–30, 35. That shows IPOM preserve its structure and good loaded on PVA. In addition, the most peaks for IPOM–PVA attributed to the crystal of IPOM are observed this indicates the highly dispersed IPOM on the surfaces of PVA support, when IPOM is bound to the PVA surface, IPOM–PVA, all of signals corresponding to IPOM and the final pattern matched to IPOM, which is most likely due to IPOM forming on the PVA surface and thus the majority of the observed signals are due to the crystal phases of IPOM, which is in good agreement with the results of SEM. Scherer equation, was used to calculate the nano crystallite size by XRD radiation of wavelength from measuring full width at half maximum of peaks in radian located at any 2θ in the pattern. The mean crystal size obtained was around 67 nm.

Fig. 4 XRD patterns of obtained powders. (a) IPOM, (b) PVA and (c) IPOM–PVA.

3.2. Catalytic results

3.2.1. Desulphurization process. According to data from the results of Table 1, total sulfur content (entry 5) and content of mercaptans (entry 6) were much lower, while other properties of gasoline showed in Table 1 remained unaffected. From the results obtained in this research, it was demonstrated that the nanocomposite IPOM–PVA, can catalyze the oxidative desulphurization reaction in 2 h and can reduce total sulfur content (wt%) of actual gasoline from 0.497 wt% to 0.015 wt% and also, reduce content of mercaptans from 87 ppm to 5 ppm. It should be noted that during the ODS process, the H₂O₂ was used in the presence of acetic acid as oxidants because acetic acid as an organic acid, reacts with H₂O₂ to in situ produce peracid, which can efficiency convert organic sulfur to sulfones without forming a substantial amount of residual product. During the oxidation desulphurization process, the H₂O₂ can efficiency convert organic sulfur to sulfones without forming a substantial amount of residual product. This is followed by a liquid extraction (acetonitrile) to obtain gasoline with low sulfur.

3.2.2. Effect of the catalyst countercation. The effect of the nature of the catalyst on the oxidative desulphurization using acetic acid/H₂O₂ as the oxidant is shown in Table 2. The amount of each catalyst was constant throughout the series. The results show that the catalytic activity of [(C₄H₉)₄N]₆Mo₇O₂₄/PVA nanocomposite has presented much higher than other unsupported polyoxometalates. The polyoxometalate with a phase transfer or emulsion catalyst comprising a quaternary ammonium salt-based polyoxometalate, it is shown very active system for oxidative desulphurization. The oxidation reactivity of the catalysts depends on the type of countercation: ((C₄H₉)₄N)⁺ > NH₄⁺ > Na⁺. That is, quaternary ammonium Anderson-type polyoxometalate ((C₄H₉)₄N)₆Mo₇O₂₄ which has lipophilic cation can better act as phase transfer agent and better transfer the peroxometal anion into organic phase. This system is bi-phase without the need of using a phase transfer agent.

3.2.3. Effect of catalyst dosage. The effect of catalyst lies mainly in the formation of peroxo-POM intermediate, which possesses much higher oxidation capacity than H₂O₂, and therefore makes easy the oxidation of DBT or BT (because the concentration of the catalytically active species (Mo(O₂)_n⁻Q⁺) increased). Consequently, knowledge on the effect of catalyst dosage is very important to the process optimization. From the results of Fig. 5 and Table 4, it is evident that higher catalyst dosage leads to higher desulfurization efficiency. Hence, the catalyst dosage is another factor that should be concerned on it. It was found from the results that the catalyst dosage does have a marked influence on the process efficiency. Under otherwise identical conditions, without catalyst, 23% of the DBT, 22% of the BT, 21% of the Th and 20% of actual gasoline is removed from the n-heptane phase in 2 h by ODS; conversion percentage of actual gasoline in the presence of [(C₄H₉)₄N]₆Mo₇O₂₄/PVA was found to be 65%, 87% and 97%, corresponding to catalyst amount of 0.06, 0.08 and 0.1 g respectively. Desulphurization efficiency increased rapidly with the increase of catalyst dosage. The results are summarized in Table 4. Results revealed that the catalyst is active and selective for the oxidation of sulfur compounds and more than 97% of the total sulfur removal was obtained during the process. The catalyst molecules act as an emulsifying agent and are uniformly distributed in the oil–water interface and form a thin film around the dispersed water droplets. Consequently, the lipophilic quaternary ammonium cations of the emulsifier lie on the oil side and the hydrophilic polyoxoanions lie on the aqueous side. H₂O₂ used as an oxidant can continuously supply active oxygen to the polyoxoanions. The polyoxoanions in the oil–water interface of the emulsion droplets oxidize the DBT to benzothiophene sulfone (DBTO₂). The oxidized form of sulfur can be extracted from gasoline (lipophilic phase) using the polar solvent (acetonitrile).³⁵

Fig. 5 Effect of catalyst dosage on the ODS of BT, Th, DBT and gasoline.

Table 4 Effect of catalyst dosage in the ODS of gasoline and simulated gasoline ^a

Entry	Amount of catalyst (g)	Conversion (%)
		DBT	BT	Th	Actual gasoline
a Condition for desulphurization: 50 mL of gasoline, 3 mL oxidant, 10 mL of extraction solvent, time = 2 h, and temperature = 40 °C, catalyst = IPOM–PVA.
1	0 (none)	23	22	21	20
2	0.02	38	36	34	33
3	0.04	49	47	45	44
4	0.06	68	66	63	62
5	0.08	91	88	84	82
6	0.10	99	98	97	97
7	0.11	99	98	97	97
8	0.12	99	98	97	97

---

3.2.4. Effect of temperature on the ODS of gasoline or simulated gasoline. The reaction was carried out at different temperatures under the same conditions by [(C₄H₉)₄N]₆Mo₇O₂₄/PVA as a catalyst and CH₃COOH/H₂O₂ as oxidant. The results are shown in Table 5 and Fig. 6. Five temperatures were tested (25°, 30°, 35°, 40°, and 45 °C). The results show that yields of products are a function of temperature. The conversion of sulfur in model compound and in actual gasoline increased with temperature and time. The conversion of sulfur in actual gasoline at 40 °C is higher than that at 35° and 30 °C. The sufficiently high desulfurization rates were obtained at 40 °C. The maximum desulfurization rate was obtained at 40 °C, which 97% of sulfur could be removed from gasoline in 120 min. After 40 °C, further increase in the reaction temperature had a negligible effect on the desulfurization rates.

Table 5 Effect of different temperatures on the ODS of gasoline and simulated gasoline ^a

Entry	Temperature (°C)	Conversion (%)
		DBT	BT	Th	Actual gasoline
a Condition for desulphurization: 50 mL of gasoline, 0.1 g nanocatalyst, 3 mL oxidant, 10 mL of extraction solvent, time = 2 h.
1	25	78	76	75	76
2	30	86	83	82	82
3	35	95	92	90	91
4	40	99	98	97	97
5	45	99	98	97	97

---

Fig. 6 Effect of different temperature on the ODS of BT, Th, DBT and gasoline.

3.2.5. Effect of solvent amount. The polar solvent (acetonitrile) plays a very important role in the catalytic reactions carried out in the liquid phase (down phase). Sulfur compounds are known to be slightly more polar than hydrocarbons with similar structures. Oxidized sulfur compounds, such as sulfones (DBTO₂) and sulfoxides (DBTO) are substantially more polar than corresponding sulphides (DBT). During ODS process the products are transferred to the polar solvent, and the solvent can influence the mass transport.³² This extraction process transferred the sulfones into the solvent, which could accelerate the reaction speed and enhance the reaction fractional conversion. Also, a fraction of DBT removed from oil phase was not converted to its corresponding sulfone (DBTO₂), under these experimental conditions. They are only removed as sulfur compounds through extraction, without ODS reaction. Therefore, the polar solvent (acetonitrile) plays a very important role in the ODS reactions. In order to investigate the solvent amount effect, different reactions were carried out by raising the solvent amount from 5 to 30 mL. There was no need to use higher volume of solvent because in an industrial process, high solvent volume implies very high operation costs and higher reactor volume. The ODS conditions were as follows: BT, Th and DBT were used as the sulfur substrates, atmospheric pressure, 313 K, 0.1 g of IPOM–PVA, and CH₃COOH : H₂O₂ (3 mL) in ratio of 1 : 1. The results obtained are shown in Fig. 7. The conversion of BT, Th and DBT were increased with increasing the amount of acetonitrile (polar solvent). These results is related to the increase of the solubility of the sulfones, produced in the ODS reaction, with the amount of solvent. This means that, the rate of removal of sulfur compounds (BT, Th and DBT) depends on the extraction capability of the solvent (acetonitrile) rather than on the transfer of the oxidant from the water phase to the oil phase. The efficiency of the sulfones extraction rises in parallel to the amount of solvent.

Fig. 7 Effect of the solvent amount on the ODS of BT, Th and DBT.

3.2.6. Mechanism of the oxidative desulphurization reaction. The IPOM–PVA and H₂O₂ heterogeneous system was used for the oxidative desulphurization of sulfur compounds in gasoline. As seen in Scheme 2, firstly, in the presence of H₂O₂, the POM compounds with Anderson-type structure will convert to polyoxoperoxo species [Mo₇O₂₅]⁶⁻. The mechanism has been supposed as follows: [Mo₇O₂₄]⁶⁻ reacted with H₂O₂ to form polyoxoperoxo specie [Mo₇O₂₅]⁶⁻ (Scheme 2 (1)) and then sulfur-containing (BT, DBT, R¹SR², …) compounds oxidized to the corresponding sulfoxides (BTO, DBTO, R¹S OR², …) by polyoxoperoxo (Scheme 2 (2 and 3)), the sulfoxides again react with [Mo₇O₂₅]⁶⁻, oxygen transfers from the active Mo-peroxo species to the sulfide; and convert to sulfones (BTO₂, DBTO₂, R¹SO₂R², …) (Scheme 2 (4 and 5)).^26,28 The sulfones can be easily removal by extraction with a polar solvent (CH₃CN). The mechanism of sulfide oxidation to sulfoxides using H₂O₂–organic acid is not studied sufficiently; however, the potential mechanism is a heterolytic electrophyl interaction where H⁺X⁻ is a polar solvent. Based on this mechanism, H₂O₂ fist reacts with CH₃COOH (organic acid) quickly and generates peroxide acid (CH₃C(O)OOH), and then the acid reacts with nonpolar sulphur compounds and generates relative sulfone or sulfoxide. The role of the metal atoms in IPOM–PVA, M = Mo, is to form peroxo-metal species which are able to activate the H₂O₂ and peracid molecules. During the oxidative desulphurization process, the H₂O₂ can efficiency convert organic sulphur to sulfones without forming a substantial amount of residual product. IPOM–PVA accepted the active oxygen from the oxidant H₂O₂ to form new oxoperoxo species mediate. The cation with carbon chain transferred oxoperoxo species to the substrates (Th or BT) and made the oxidation reaction accomplish completely.

Scheme 2 Probable mechanism for the formation of polyoxo and using H₂O₂ as the oxidant for BT, DBT, RSR oxidation.

3.2.7. Kinetics of Th oxidation. For better understand the catalytic oxidation of Th, BT and DBT reaction kinetics was examined. To the apparent consumption of Th, BT and DBT the constant rate was gained from the first-order kinetic model (eqn (3)) as follows:

C = C₀ e^−kt (5)

where:

Constant rate of Th was calculated using their initial concentrations at time zero (C₀) and time t reaction (C_t) in the plot of ln(C/C₀) or C/C₀ against t with use of eqn (5), an exponential line (C/C₀) with slope k was gained (Fig. 8). The relationship between C and t can be applied to describe rate equations. With increasing reaction temperature from 25 to 40 °C removal of BT, DBT and Th and the constant rate reaction also increasing in 2 h (Table 6). The affiliation of the rate constant k on the reaction temperature can be expressed with the Arrhenius equation (eqn (6)). The A, E_a, R and T are the pre-exponential factor, the apparent activation energy, gas constant and the reaction temperature (K), respectively.²⁸ Fig. 9 shown the Arrhenius plots and the calculated E_a values for the oxidation of DBT, BT and Th were 48.27, 49.13 and 50.11 kJ mol⁻¹, respectively.

Fig. 8 Plots of C/C₀ for the oxidation of model compounds with the IPOM–PVA catalyst. (1) 25 °C, (2) 30 °C, (3) 35 °C and (4) 40 °C.

Table 6 Pseudo-first-order rate constants and correlation factors of the DBT, Th and BT

Temperature (°C)	Rate constant k (min^−1)			Correlation factor R^2
	DBT	Th	BT	DBT	Th	BT
25	0.012	0.010	0.008	0.9621	0.9505	0.8475
30	0.016	0.013	0.009	0.9646	0.9709	0.9675
35	0.022	0.021	0.016	0.9697	0.9732	0.9676
40	0.030	0.025	0.019	0.9241	0.9496	0.9735

---

Fig. 9 Arrhenius plot for the oxidation of model compounds with the IPOM–PVA catalyst.

The oxidation reactivity decreased in the order of DBT > BT > Th (Fig. 8). Th exhibited the lowest reactivity, and this was related to the different electron density on the sulfur atom.

3.2.8. Recycling of the catalyst. At the end of the oxidative desulphurization of the model sulfur compounds and gasoline, the catalyst was filtered and washed with dichloromethane. In order to determine whether the nanocatalyst (IPOM–PVA) would succumb to poisoning and lose its catalytic activity and selectivity during the reaction, the reusability of the IPOM–PVA was investigated. For this purpose, we carried out the desulphurization reaction of gasoline and model compounds in the presence of fresh and recovered catalyst.

The catalytic run was repeated with addition of substrates under optimized reaction conditions, and the selectivity of the products was comparable to that of the original one. It was observed that the activity and selectivity of the catalyst does not change significantly after five consecutive runs (Table 7).

Table 7 Reuse of the catalyst on the ODS of DBT^a

Entry	DBTO2
	Conversion^b (%)	Selectivity (%)
a Conditions for desulphurization: 50 mL of model gasoline (500 ppm S), 0.1 g catalyst, 3 mL acetic acid/H2O2, 10 mL extraction solvent, time 2 h and temperature 40 °C.b Conversion and selectivity were determined by GC.
1	97	94
2	96	92
3	95	90
4	94	89
5	93	87

---

The IPOM–PVA recovered after the reaction was characterized, in order to check the IPOM–PVA stability. Fig. 10 illustrates XRD pattern and IR spectrum of IPOM–PVA after five catalytic cycles, respectively. Even after five runs of the reaction, the catalytic activity of IPOM–PVA was almost the same as that of freshly used catalyst.

Fig. 10 Comparison of XRD pattern and IR spectra of free IPOM–PVA after first run reuse (a) after 3 rune reuse (b) after 5 rune reuse (c).

4. Conclusions

In summary, a facile method to IPOM–PVA was prepared by combining the solution of PVA and IPOM in a one-pot strategy for the generation of IPOM–PVA. The organic–inorganic hybrid based on Anderson-type polyoxometalate ((C₄H₉)₄N)₆Mo₇O₂₄/PVA presented here is unique in several aspects; this simple synthesis route demonstrates that the use of PVA molecule as organic electron donor reacting chemically with polyoxometalate anion (electron acceptor) can give rise to amorphous nanomaterials, which exhibit high catalytic activity. FT-IR, UV-vis, SEM and XRD indicate that, IPOM was loaded at the PVA. SEM images indicate that the surface is uniform, non-smooth and nanoparticles were made. ODS of actual gasoline fuel has been obtained by IPOM–PVA as nanocatalyst, CH₃COOH/H₂O₂ as oxidant. Under these conditions, the concentration of sulfur in gasoline can be reduced from 0.497 to 0.015 ppmw. It is proved that the good performance of these catalysts could be due to the Anderson-type polyoxometalate compounds, which could rapidly transform into the active polyperoxometalate in the presence of hydrogen peroxide and eco-friendly. Nanocatalyst IPOM–PVA shows excellent catalytic activity and selectivity in oxidation of DBT. For this IPOM–PVA/CH₃COOH/H₂O₂ system, oxidation reactivity decreased according to the following order: DBT > BT > Th. The percentage conversion increases when the amount of oxidant and catalyst was increased. This system provides an efficient, convenient and practical method for oxidative desulphurization of gasoline and the advantages of this method are nontoxic, mild condition and environmentally friendly.

References

1. R. G. Finke, M. W. Dreoge and P. J. Domaille, Trivacant, Inorg. Chem., 1987, 26, 3886–3896.
2. I. V. Kozhevnikov and K. I. Matveev, Appl. Catal., A, 1983, 5, 135–150.
3. B. Li, Z. Liu, J. Liu, Z. Zhou, X. Gao, X. Pang and H. Sheng, J. Colloid Interface Sci., 2011, 362, 450–456.
4. B. Li, W. Ma, J. Liu, C. Han, S. Zuo and X. Li, Catal. Commun., 2011, 13, 101–105.
5. M. Fukuma, Y. Seto and T. Yamase, Antiviral Res., 1991, 16, 327–339.
6. H. Lü, C. Deng, W. Renb and X. Yang, Fuel Process. Technol., 2014, 119, 8416–8420.
7. B. Zhang, M. D. Zhou, M. Cokoja, J. Mink, S. L. Zang and F. E. Kühn, RSC Adv., 2012, 2, 8416–8420.
8. H. Lü, S. Wang, C. Deng, W. Ren and B. Guo, J. Hazard. Mater., 2014, 279, 220–225.
9. W. S. Zhu, H. M. Li, X. Jiang, Y. S. Yan, J. D. Lu and J. X. Xia, Energy Fuels, 2007, 21, 2514–2516.
10. W. Zhu, W. Huang, H. Li, M. Zhang, W. Jiang, G. Chen and C. Han, Fuel Process. Technol., 2011, 92, 1842–1848.
11. C. Yvon, A. Macdonell, S. Buchwald, A. J. Surman, N. Follet, J. Alex, D. L. Long and L. Cronin, Chem. Sci., 2013, 4, 3810.
12. H. Zheng, Z. Sun, X. Chen, Q. Zhao, X. Wang and Z. Jiang, Appl. Catal., A, 2013, 467, 26–32.
13. Y. Chen, S. Zhao and Y.-F. Song, Appl. Catal., A, 2013, 466, 307–314.
14. X. M. Yan, P. Mei, L. Xiong, L. Gao, Q. Yang and L. Gong, Catal. Sci. Technol., 2013, 3, 1985–1992.
15. Z. E. A. Abdalla and B. Li, Chem. Eng. J., 2012, 200, 113–121.
16. Z. Cui, C. X. Guo and C. M. Li, J. Mater. Chem. A, 2013, 1, 6687–6692.
17. B. Bi, L. Xu, B. Xu and X. Liu, Appl. Clay Sci., 2011, 54, 242–247.
18. X. L. Ma, S. Velu, J. H. Kim and C. S. Song, Appl. Catal., B, 2005, 56, 137–147.
19. A. M. Dehkordi, Z. Kiaei and M. A. Sobati, Fuel Process. Technol., 2009, 90, 435–445.
20. J. F. Wishart, Energy Environ. Sci., 2009, 2, 956–961.
21. C. D. Wilfred, C. F. Kiat, Z. Man, M. A. Bustam, M. I. M. Mutalib and C. Z. Phak, Fuel Process. Technol., 2012, 93, 85–89.
22. H. Li, W. Zhu, Y. Wang, J. Zhang, J. Lu and Y. Yan, Green Chem., 2009, 11, 810–815.
23. W. Zhu, H. Li, X. Jiang, Y. Yan, J. Lu and J. Xia, Energy Fuels, 2007, 21, 2514–2516.
24. W. Zhu, J. Zhang, H. Li, Y. Chao, W. Jiang, S. Yin and H. Liu, RSC Adv., 2012, 2, 658–664.
25. M. A. Rezvani, A. F. Shojaie and M. H. Loghmani, Fuel Process. Technol., 2014, 118, 1–6.
26. M. A. Rezvani and F. Mohammadi Zonoz, Ind. Eng. Chem., 2015, 22, 83–91.
27. M. A. Rezvani, A. F. Shojaie and M. H. Loghmani, Catal. Commun., 2012, 25, 36–40.
28. M. A. Rezvani, M. Oveisi and M. Ali Nia Asli, J. Mol. Catal. A: Chem., 2015, 410, 121–132.
29. M. A. Rezvani, A. F. Shojaei and F. M. Zonoz, J. Serb. Chem. Soc., 2014, 79, 1099–1110.
30. A. F. Shojaei, M. A. Rezvani and F. M. Zonoz, J. Serb. Chem. Soc., 2013, 78, 129–136.
31. M. A. Rezvani, M. Ali Nia Asli, L. Abdollahi and M. Oveisi, J. Serb. Chem. Soc., 2016, 81, 91–101.
32. H. S. Mansur, R. L. Oréfice and A. A. P. Mansur, Polymer, 2004, 45, 7193–7202.
33. H. S. Mansur, C. M. Sadahira, A. N. Souza and A. A. P. Mansur, Mater. Sci. Eng., C, 2008, 28, 539–548.
34. L. C. Caero, E. Hernández, F. Pedraza and F. Murrieta, Catal. Today, 2005, 564, 107–108.
35. J. Qiu, G. Wang, D. Zeng, Y. Tang, M. Wang and Y. Li, Fuel Process. Technol., 2009, 90, 1538–1542.

---