Reduction of the amount of sulfur in petroleum products and fuels is essential for environmental protection [1]. In the last decade, countries worldwide have drastically altered their regulations restricting the sulfur content in diesel fuel thus making operation units intended for deep desulfurization at petroleum refineries to be an indispensable feature [2, 3]. Considerable efforts have been made to develop effective and economical methods for the fuel desulfurization. Hydrodesulfurization (HDS) is a standard method to eliminate sulfur contaminants from fuel [4, 5]. However, HDS has several challenging drawbacks, such as high hydrogen consumption, rough operational conditions including high temperature (300–400°C) and pressure (10–130 atm), and an extensive variation in reactivity of aromatic sulfur-containing compounds such as benzothiophene (BT), dibenzothiophene (DBT), and other methyl-substituted dibenzothiophene derivatives [6, 7]. As a result, the sulfur level in a HDS product (~500 ppm) does not meet the requirements of deep desulfurization. Therefore, the further attempts are needed to overcome challenges associated with HDS and to achieve a deeper desulfurization [8, 9].

The process of a catalytic oxidation of sulfur compounds (CODS) is a reliable method for achieving a low level of sulfur [10, 11]. Even those DBT derivatives, which are difficult to eliminate through the HDS process, can be oxidized into sulfoxides and sulfones and subsequently removed by CODS under ambient conditions [12‒14]. The most widely known alternative method is oxidative desulfurization (ODS), which includes the removal of sulfur-containing compounds under the action of oxidants followed by the removal of oxidized products by oxidation methods [14‒16]. Among various oxidizers described in the literature, air oxygen is the most interesting [17‒19].

Changing the original polyoxymetalate (POM) precursors makes it possible to create new compounds with special structures and properties. These compounds are called transition metal-substituted POM (TMSPOM) and have metal ions in their structure. Researchers are interested in these structures because they can be modified in many ways including changes in their size, shape, and type of interaction with other materials. TMSPOMs are better than other metal composites because they are able to withstand oxidation. They have the following advantages over organometallic complexes: (i) they are stable under oxidation conditions, while the most of organic ligands became decomposed, (ii) their solubility is adjustable by changing counter cations, and (iii) their redox properties are adjustable by the replacement of a central (hetero)atom and an incorporated transition metal. The use of manganese-substituted polyoxometalates (Mn-POMs) as catalysts attracted a considerable attention due to such distinctive properties as variable oxidation states as well as oxidation and magnetic properties. Mn-POMs represent a diverse group of inorganic building blocks, which offer a wide range of properties, including different morphology, structural flexibility, and capabilities. Their use has a potential to significantly increase the outcome of the oxidation process [20, 21]. Nanoporous materials are structures containing nano-size pores, which recently provided conditions for the emergence of a new class of mesoporous structures called SBA [22‒24]. The mesoporous silicates can be modified by organosilanes such as aminopropyltrimethoxysilane (APTMS) to form bonds between organic compounds (NH2) and minerals (manganese-substituted polyoxometalates) [25, 26].

In this study, a new heterogeneous nanocatalyst CTAB-PMo11Mn@OM SBA-15 was synthesized by the sol-gel method and used for completely removal of sulfur compounds from petroleum products by a hybrid desulfurization method.

EXPERIMENTAL

Materials and methods. Cetyltrimethylammonium bromide (C16H33N(CH3)3 Br), tetraethyl orthosilicate (SiC8H20O4), and sodium molybdate dihydrate (Na2MoO4·2H2O) were obtained from the Merck company (Rahway, United States). DBT was obtained from the Sigma-Aldrich company (St. Louis, United States). All other chemicals used in this study were prepared by the Amertat-shimi Company (Tehran, Iran) with a confirmed analytical purity. A Binder Gmbh-ED53 electric oven (Binder Gmbh, Tuttlingen, Germany) was used for sample drying. An electric furnace for calcination was purchased from the Atbin Company (Tehran, Iran), and a YX2000A ultrasonic device (Yahun, Guangzhou, China) was used to homogenize the samples.

Field-emission scanning electron microscopy (FE-SEM) was performed on a JEOL model JSM-7500 F with an accelerating voltage of 5.0 kV and a current of 10 mA. Fourier transform infrared spectroscopy (FT-IR) studies weredone on a Thermo-Nicolet-is 10 spectrometer, using KBr disksin the range 400–4000 cm–1. Powder X-ray diffraction (PXRD) patterns of the samples were recorded on a Bruker Focus D8 diffractometer with CuKα X-ray radiation (λ = 0.154056 nm). A JEOL JEM 2010 (Accelerating voltage: 200 kV, JEOL, Japan) Transmission Electron Microscope was used to obtain TEM images. N2 adsorption–desorption isotherms were measured on a Micromeritics 2020 analyzer at 77 K, Based on Brunauer–Emmett–Teller (BET) and Barrett–Joyner–Halenda (BJH) analyses.

A gas chromatography-flame ionization detector (GC-FID) was used to measure the amount of the analyte in a sample [27, 28]. Pre-hydrotreated diesel fuel (density 0.7957 g/mL at 25°C and the total sulfur content 481.80 mg/L) was provided by the Arak Petrochemical Ltd. (Arak, Iran).

Oxidative desulfurization of model fuel. Reactions were performed in 250-mL three-necked round-bottom flasks. To prepare a model diesel sample (simulated diesel fuel), 50 mL of n-heptane solution containing 500 ppm of refractory sulfur was prepared by adding an appropriate amount of DBT to a three-necked flask equipped with a funnel, reflux condenser, thermometer, and motorized stirrer. Different dosages of CTAB-PMo11Mn@OM SBA-15 and 3 mL of hydrogen peroxide were mixed in 10 mL acetonitrile. Then, the flask with solution was heated in the water bath; the temperature and duration of heating were pre-set based on an experimental design table. All experiments were carried out at a constant agitation rate (200 rpm) and the atmospheric pressure [29‒31]. After a completion of the reaction, a certain volume of the oil phase was removed, and sulfur-containing compounds in the fuel were determined by a gas chromatography [32, 33].

ODS is a multi-step process. In this process, a CTAB-PMo11Mn@OM SBA-15 catalyst and an oxidant are present in the polar extracting phase along with DBT extracted from the fuel phase. The initial step of the oxidation reaction involves adsorption of the oxidant (H2O2) and DBT on the surface of the catalyst. This interaction between the adsorbed H2O2 and the catalyst results in the creation of intermediate peroxo-heteropoly active sites on the catalyst surface. Active peroxo sites react with adjacent adsorbed DBT to produce an intermediate (DBTO2). The proposed mechanism for the removal of sulfur-containing compounds of petroleum derivatives by CTAB-PMo11Mn@OM SBA-15 is shown in Scheme 1.

Scheme 1.
scheme 1

Proposed mechanism for the removal of sulfur-containing compounds of petroleum derivatives by CTAB-PMo11Mn@OM SBA-15.

Full size image

Oxidative desulfurization process for real diesel fuel. The study was aimed to investigate the effectiveness of a heterogeneous hybrid nanocatalyst in desulfurization of four pre-hydrotreated diesel fuel samples obtained from the Arak Petrochemical Company Co. Ltd. (Arak, Iran). The density of samples at 25°C was 0.7957 g/mL and the total sulfur content was 481.80 mg/L. All catalytic experiments were conducted at the ambient pressure within a closed 25-mL borosilicate reaction vessel equipped with a magnetic stirring bar and immersed in a thermostated oil to maintain a constant temperature. Each experiment included an addition of 10 mL of a pre-hydrotreated diesel fuel sample into the reactor. A catalyst (60 mg) was mixed with 3 mL of a 30% H2O2 solution (at a hydrogen peroxide : sulfur content ratio of 1 : 3) and 10 mL of acetonitrile to form a biphasic liquid-liquid system. The mixture was thoroughly stirred for 45 min at 60°C with a periodical measurement of a sulfur content in the fuel using a gas chromatography with a flame ionization detection (GC-FID). To do this, a 10-µL aliquot of an internal standard (tetradecane) was added to the aliquot (10 µL) of a fuel phase with the further analysis of the resulting solution. The primary purpose of the study was to determine the amount of dibenzothiophene (DBT) present in the fuel sample. Therefore, samples were chromatographed to determine the percentage of DBT removal prior and after the catalytic reaction.

Synthesis of CTAB-PMo11Mn@OM SBA-15

Synthesis of a nanocatalyst base by a sol-gel method. To synthesize SBA-15, 5 g of Pluronic P123 was dissolved in 165 ml of 2M hydrochloric acid at 35°C. Then 10 g of tetraethyl orthosilicate (TEOS) was slowly added to the solution, which was then refluxed for 48 h at 130°C. The mixture was filtered, and the precipitate was rinsed twice with a water–ethanol mix (1 : 1). Finally, the obtained sample was calcined in a furnace for 6 h at 550°C [34].

Synthesis of the APTMS/SBA-15 nanocatalyst. Calcined SBA-15 (~1 g) and 5 mmol of aminopropyltrimethoxysilane (APTMS) were mixed with 30 g of anhydrous toluene, and the mixture was refluxed for 12 h at 110°C. Then the mixture was filtered and refluxed again with 15 ml of toluene for 24 h at 110°C. The obtained product (APTMS/SBA-15) was dried at 110°C for 24 h [35, 36].

Synthesis of the CTAB-PMO11Mn@OM SBA-15 nanocatalyst. APTMS/SBA-15 was dispersed in 50 mL deionized water, and then supplemented with 9.1 mmol of disodium hydrogen phosphate (Na2HPO4), 100 mmol of dihydrate sodium molybdate (Na2MoO4‧2H2O), and 12 mmol of manganese nitrate (Mn(NO3)2 under thorough stirring. Sodium hydroxide and hydrochloric acid were used to adjust pH to 4.8. Then 20 mmol of CTAB were slowly added to the solution at 80‒85°C. The sediment was filtered and washed twice with a water-ethanol mix (1 : 1). Finally, recrystallization was done using acetonitrile. The nanocatalyst prepared at this step was called CTAB-PMo11Mn@OM SBA-15 [37‒40]. The reaction path for the catalyst synthesis and the mechanism of its oxidative action are shown in Scheme 2.

Scheme 2.
scheme 2

The reaction path for the synthesis of CTAB-PMO11Mn@OM SBA-15 nanocatalyst and the mechanism of its oxidative action.

Full size image
Fig. 1.
figure 1

FTIR (a), SEM (b), TEM (c) spectra and elemental mapping (d) of a CTAB-PMO11Mn@OM SBA-15 nanocatalyst.

Full size image

RESULTS AND DISCUSSION

Characterizations of a CTAB- PMO11Mn@OM SBA-15 nanocatalyst. The FTIR spectrum of a synthesized CTAB-PMO11Mn@OM SBA-15 nanocatalyst is shown in Fig. 1a. The peaks in the region of 800‒ 1100 cm–1 correspond to the bending vibrations of regular and dense Si‒O‒Si network bonds. The peak at 1048 cm–1 corresponds to the stretching vibrations of the Si‒OR bond. This indicates a formation of Si‒O‒Si bonds and confirms a formation of the silicate structure [41, 42]. Due to the APTMS addition at the second step, the absorption peaks at 2851 and 2922 cm–1 can be assigned to the stretching vibrations of aliphatic carbons, and the peak at 1473 cm–1 is assigned to the bending vibrations of aliphatic carbons that indicates the presence of APTMS on the surface. The peak at 1644 cm–1 corresponds to the bending vibrations of NH2. The morphology of SBA-15 was investigated by a scanning electron microscope (Fig. 1a). The mesoporous channels and structures of SBA-15 were clearly visible by SEM. In addition, a small number of fine agglomerations was also observed on the surface [43]. Figure 1c illustrates the TEM study of a synthesized CTAB-PMo11Mn@OM SBA-15 nanocatalyst. Two-dimensional periodic structures of SBA-15 were observed even after the introduction of Mo and Mn into the pores, that was consistent with the XRD results. No agglomerations (big particles) were observed in the images that indicated the high dispersion and confinement of Mo and Mn species in the channels in SBA-15 causing pores to be filled. This fact can be a reasonable explanation for the reduction of a specific surface area [44]. Based on the size of the formed pores, the synthesized nanocatalyst can be classified as mesoporous. In addition, the results of mapping of P, N, Mn, Mo, Si, O, C, and Br atoms shown in Fig. 1a confirm that Mn and Mo elements are uniformly distributed and dispersed in SBA-15 channels. This indicates that the surface of SBA-15, which represents a silicate base, is covered by a large amount of a material and in a multilayered manner. The isothermal graph of a nitrogen adsorption- desorption and the Barret Joyner-Halenda (BJH) analysis of SBA-15 showed it belongs to the type IV according to the IUPAC classification, that confirms the mesoporous nature of the porous composition (Figs. 2a and 2b). The texture properties of SBA-15 and CTAB-PMO11Mn@OM SBA-15 are shown in Table 1. The presence of a hysteresis cycle between the nitrogen absorption and desorption on the surface is explained by the presence of a capillary condensate in mesoporous cavities, which is absorbed in the cylindrical cavities due to a gas liquefaction. The delayed desorption is one of the characteristics of mesoporous materials (evaporation occurs at a lower relative pressure). As it is shown in the isothermal graph, this hysteresis cycle can be observed in SBA-15 within the relative pressure range of 0.55‒0.75. The BET and BJH analyses of the CTAB-PMo11Mn@OM SBA-15 shows the type IV isotherm [45]. According to the obtained results, the reduction of the specific surface area of the SBA-15 nanocatalyst base from 520.58 to 150.89 m2/g in the CTAB-PMo11Mn@OM SBA-15 nanocatalyst indicates the addition of APTMS and metal oxides into the pores; as a result, the pores are blocked and the specific surface area and volume in the final synthesized nanocatalyst are decreased [46].

Table 1. Textural characteristics of SBA-15 and CTAB-PMO11Mn@OM SBA-15a
Full size table

The CTAB-PMO11Mn@OM SBA-15 nanocomposite was also underwent to the X-ray diffraction (XRD) analysis to determine its structural characteristics. A wide-angle XRD pattern showed a broad band between the 2θ angle values of 15° and 35° (Fig. 2c) that is typical for amorphous SBA-15. Along with this broad band, nine additional diffraction peaks were observed for the CTAB-PMO11Mn@OM SBA-15 nanocomposite, which corresponded to reflections of the nanocomposite at 2θ values equal to 33.1°, 35.1°, 37.6°, 48.8°, 55.8°, 62.3°, 65.9°, 67.8°, and 68.8°. These reflections were assigned to the (100), (002), (101), (102), (110), (103), (200), (122), and (201) planes, respectively. According to the previously confirmed results and documents, these reflections indicate the correct synthesis and placement of polyoxomolybdate substituted with manganese (PMO11Mn) on the mesoporous silica substrate [45‒46]. As shown in Fig. 2d, the low angle XRD pattern of the as-synthesized SBA-15 materials exhibited three well-defined peaks corresponding to the (100), (110), and (200) planes, that indicates SBA-15 is characterized by the 2D hexagonal symmetry and long-range mesoporous ordering.

Fig. 2.
figure 2

(a) Adsorption/desorption isotherms, (b) dBJH plot, (c) wide angle XRD patterns, and (d) small angle XRD patterns of the SBA-15 and CTAB-PMo11Mn@OM SBA-15 nanocatalyst.

Full size image

Oxidative Desulfurization (OD) Experiments

A GC analysis was performed using an Agilent 7820A gas chromatograph equipped with the flame ionization detector and capillary column (Agilent 1909/z-530, 100 m × 250 μm × 0.5 μm). The carrier gas was nitrogen; the temperature of the injector and detector was 245 and 310°C, respectively. The temperature of the column in the beginning of the analysis was 50°C. Then it was heated from 50 to 250ºC with a 10ºC increments. Figure 3 shows chromatograms of DBT-STD and DBT-sulfone standards and the results of oxidation performed at 30 and 65°C.

Fig. 3.
figure 3

GC-FID chromatogram of (a) DBT-STD standard, (b) DBT-sulfone standard, (c) model diesel fuel at 30°C, and (d) model diesel fuel at 65°C.

Full size image

Effect of temperature on the ODS process. The temperature is considered as one of the effective parameters of the EODS process influencing on the removal of sulfur-containing compounds of DBT. The effect of a reaction temperature on the oxidative desulfurization process was investigated at the temperature of 40, 50, 60, and 70°C. The DBT removal was shown to increase with the temperature increase from 40 to 60°C (Fig. 4a). Thus, one can conclude that the temperature increase can accelerate the reaction rate that, in turn, results in an increased generation of intermediate peroxo-heteropoly active sites on the catalyst surface. In addition, an increase in the temperature causes an increase in the vapor pressure and catalytic activity and reduces the liquid viscosity. Therefore, it results in the reduction of the limitation of a mass transfer between the aqueous and organic phases. However, the temperature increase to 70°C did not provide any effect on the DBT removal [47‒50].

Fig. 4.
figure 4

Effect of (a) temperature, (b) oxidant ratio, (c) catalyst dosage, and (d) reaction time on the ODS process.

Full size image

Effect of the oxidant ratio on the ODS process. The effect of the molar ratio of an oxidant in the presence of a CTAB-PMo11Mn@OM SBA-15 nanocatalyst was studied at 60°C for the contact duration of 30 min. The results are shown in Fig. 4b. The hydrogen peroxide to sulfur ratios used in the experiment were 1 : 2, 1 : 1, 2 : 1, and 1 : 3. The DBT removal was provided by a decomposition of hydrogen peroxide to a hydroxyl radical. As it is shown in the figure, the hydrogen peroxide to sulfur content ratio equal to 1:3 provided an increased efficiency of the DBT removal from the model fuel.

Effect of a catalyst dosage on the ODS process. The effect of the CTAB-PMo11Mn@OM SBA-15 dosage on the ODS process is shown in Fig. 4c. An increase in the amount of a nanocatalyst from 40 to 60 mg resulted in an increased in the DBT removal efficiency. The reason of such behavior can be associated with an increase in the number of catalytically active sites. The further increase in the amount of the nanocatalyst to 70 mg did not provide any significant changes in the DBT removal. Therefore, the optimal amount of a CTAB-PMo11Mn@OM SBA-15 nanocatalyst for the DBT removal is 60 mg [50‒54].

Effect of the reaction time on the ODS process. The effect of the reaction time as an effective parameter on the DBT removal in the ODS process is shown in Fig. 4d. The highest DBT removal rate (98%) was registered after 45 min of the contact. The further increase of the contact time did not provide a significant effect on the removal of dibenzothiophene. At the beginning of the reaction, the number of catalytic active sites for oxidation was large, that explained a high percentage of dibenzothiophene absorption within the first 45 min. However, as the contact time increased, the adsorption sites became occupied, so the volume of a DBT adsorption by the nanocatalyst gradually decreased.

Oxidative desulfurization of diesel fuel. The results of a DBT removal are shown in Fig. 5. The DBT and DBT-sulfone peaks appeared at the level of 15 and 23 min, respectively. As shown in the figure, the nanocatalyst was able to remove 98% of the residual sulfur content in diesel and gasoline [55‒57]. In this experiment, the efficiency of removal of sulfur-containing compounds was studied at different reaction duration. According to the obtained results, the maximum sulfur removal occurred in the beginning of the process (the first 20 min), and it took 45 min for the reaching of its maximum efficiency.

Fig. 5.
figure 5

Time-dependent sulfur removal of DBT from four Diesel fuel samples.

Full size image

Comparison of the process performance achieved in the current and previous studies. The data on the performance and the yield of oxidative desulfurization provided by different catalysts reported by other researchers are summarized in Table 2. Though our study was characterized by moderate operation conditions compared to those in previous studies, the method showed a superior performance in terms of the desulfurization yield, catalyst usage, and reaction time. Furthermore, experiments described in the literature were conducted mainly at lower DBT concentrations than that used in our study (500 ppm). On the other hand, the time required for a complete oxidation of sulfur-containing compounds by our method is significantly shorter than that required for other methods that results in an increase in the rate of catalysis and provides a possibility of a process automation. Another advantage of this method is a temperature required for the oxidation of sulfur-containing compounds. Compared to other methods, the proposed method provides more mild conditions. The superior performance of the proposed ODS method in comparison with other similar desulfurization systems can be attributed to the unique characteristics of the Mn-substituted catalyst, while the use of SBA-15 as a support effectively immobilizes CTAB-PMO11Mn and promotes the mass transfer rate with large specific area and low diffusion resistance.

Table 2. Comparison of the proposed oxidative desulfurization (ODS) system with other known ODS systems
Full size table

CONCLUSIONS

In this study, a cetyltrimethylammonium bromide-manganese phosphomolybdate nanocatalyst stabilized on SBA-15 modified with organosilanes has been synthesized. The structure of the heterogeneous hybrid nanocatalyst was investigated and the catalytic oxidative desulfurization of DBT in gasoline was studied. In addition, the effect of various operating parameters such as temperature, reaction time, oxidant ratios and the amount of a nanocatalyst on the process performance was evaluated. In the structure of the nanocatalyst, CTAB was used as a transfer agent. After manganese introduction, the resulting nanocatalyst was immobilized on mesoporous silicate SBA-15. The optimum conditions for removing 98% of DBT from diesel fuel were determined; they included application of the temperature of 60°C for 45 min, use of 40 mg of nanocatalyst, and the oxidant ratio of 1 : 3.