Ionic liquids on desulfurization of fuel oils

https://doi.org/10.1016/j.fluid.2009.12.020Get rights and content

Abstract

In this paper, a review about the role of Ionic Liquids on desulfurization of fuel oils has been done. From these salts, the pyridiniums are showing the most promising results on sulfur reduction by simple liquid–liquid extraction. For this reason, in this work the suitability of a new pyridinium ionic liquid as solvent in the extraction of sulfur- and nitrogen-containing compounds from fuels has been analysed. Liquid + liquid equilibrium data for 1-hexyl-3,5-dimethyl pyridinium {bis[trifluoromethylsulfonyl]imide} + thiophene + n-hexane or n-dodecane or n-hexadecane and 1-hexyl-3,5-dimethyl pyridinium {bis[trifluoromethylsulfonyl]imide} + pyridine + hexane ternary systems have been determined at 298.15 K and atmospheric pressure. High solubility of thiophene and pyridine in ionic liquid but also of toluene, has been found, being this salt practically inmiscible with linear hydrocarbons. Equilibrium data of these systems have been correlated with UNIQUAC and NRTL models driving to high deviations at high solute concentrations. Three steps extraction experiments with simulated gasoline and diesel and also with real samples, previous to refinery desulfurization process, confirmed ability of ionic liquid [hmmpy][Ntf2] as solvent for extractive desulfurization of transportation fuels.

Introduction

Emissions of sulfur gases from both natural and anthropogenic sources strongly influence the chemistry of the atmosphere. Sulfur dioxide is emitted largely from the excessive burning of fossil fuels, petroleum refineries, chemical and coal burning power plants etc. Sulfur dioxide and nitrogen dioxide can suffer a harmful reaction in the atmosphere that cause acid rain. Over the last decade, environmental regulations focus attention on reduction of emissions from the transport sector in order to improve air quality and welfare [1]. According to the Directive of the European Union, gasoline and diesel fuels in Europe should not exceed 10 ppm of total sulfur content starting from 2010 [2]. US regulations established a maximum of 15 ppm for diesel starting from 2006 and 30 ppm for gasoline starting from 2005 [3].

Conventionally, hydrodesulfurization (HDS) has been employed by refineries to remove organic sulfur from fuels. This widely applied process easily allows the elimination of aliphatic and alicyclic sulfur compounds, but removal of many thiophene and dibenzothiophene derivatives turns into a difficult task due to several catalyst surface interactions and stereo hindrance [4], [5], [6], [7]. Due to these difficulties to remove some sulfur-containing aromatic compounds highly resistant to hydrotreatment, several attempts to change operation conditions of HDS are under way [8]. Nonetheless, to carry out the deep HDS processes, the energy and hydrogen consumption should be evidently increased and these severe operation conditions inevitably lead to high increase in capital expenditure.

The production of transportation fuels having very low content of sulfur became one of the priority actual challenges for oil refineries worldwide forced by new strict regulatory requirements. Under this context, many researchers have been developing new innovative methods to meet these sulfur regulations based on improving the HDS current process in refineries (high active catalysts, reactor improvements…) [9], [10] being stringently necessary to explore the alternative desulfurization approaches as reactive distillation, several precipitation techniques, reactive alkylation, complexation, selective oxidation, adsorption, biodesulfurization, photochemical desulfurization, several utltrasonic treatments extraction… and their combinations already reviewed in literature [11], [12], [13], [14], [15], [16], [17].

Among all these new processes, as clean, cheap because of low energy cost, mild operation conditions and facile operation option, extractive desulfurization appears to be especially promising and researching efforts over the last years are being focused on finding the most suitable solvent for desulfurization purposes. Apart from other desired features, an extracting solvent must primarily exhibit high capacity, high solute distribution ratio and high selectivity. A lot of organic solvents such as polyalkyleneglycol, polyalkyleneglycol ether, pyrrolidones, imidazolidinones, dimethyl sulfoxide and pyrimidinones were tried as extractants for the removal of sulfur compounds from fuels, but none of them obtaining satisfactory performances.

Over the past years Ionic Liquids (ILs) became a promising family of substances that have been object of burgeoning research. From the beginning of this decade, they have been defined as salts with melting point below 100 °C [18], [19], [20]. ILs are promising replacements for conventional non-desirable organic solvents [21] to act as entrainers in liquid–liquid extraction. Most of ionic liquids have many attractive properties such as chemical and thermal stability, nonflammability, high ionic conductivity, wide range of potential window and they posses a very low vapour pressure, which may be the single most attractive property for their use as solvents [23]. A key feature of ionic liquids is that their properties can be tailored by judicious selection of cation, anion and substituents so they can be designed to perform selective liquid–liquid separations [22], [23], [24]. It may be apparent that ionic liquids may in themselves be suitable and indeed favourable media for the design of novel liquid–liquid extraction systems [25], [26].

The use of ionic liquids for the selective extraction of sulfur compounds from diesel fuel was described by Bössman et al. for the first time [27]. Keeping in mind that the most popular ionic liquids are undoubtedly the di-alkylimidazolium salts, due to their easy synthesis and attractive physical properties, and based on the initial idea of the extraction of sulfur compounds by chemical interaction, these authors investigated the extraction with Lewis- and Brønsted-acidic ionic liquids employing mixtures of n-butyl-3-methylimidazolium chloride and n-ethyl-3-methylimidazolium chloride with AlCl3. Even though AlCl3 and AlCl4 based ionic liquids are effective for the removal of S-containing compounds [28], they often form dark precipitates and they are sensitive to water presence [29] generating HCl and making them unstable in air and of limited practical utility.

Ionic liquid extraction power for DBT has been proved to be not uniquely based on chemical interactions involving the acid proton, so different cation/anion combinations of neutral ILs were tested [27], showing that desulfurization is hardly affected by the chemical nature of the anion and that the size of ions is important for the extraction effect. A possible explanation for this behaviour may be that solubility of sulfur-containing compounds (DBT and derivates) is dependent on steric factors in the IL.

Several authors studied the use of PF6 and BF4 anions for imidazolium-based ionic liquids [30], [31], [32] significantly more stable to hydrolysis. Sulfur was preferentially removed from gasoline samples but also a change in the content of aromatics was observed, which can be problematic for maintaining octane number in gasoline. Several liquid–liquid equilibrium studies reinforce this conclusion [33], [34], [35] and coulombian interactions between these ionic liquids and S-containing compounds were also studied [36]. Removal of sulfur-containing molecules with high density of aromatic π electrons is favoured due to stronger interactions with this kind of ILs. The extractive removal efficiency for aromatic nitrogen compounds is also high and extraction results indicated that these ILs are particularly selective for aromatic N-containing compounds from fuels. But the use of these PF6 and BF4 ILs as large scale extracting agents is not optimal because of the relative high price of starting material and additionally, the formation of hydrolysis products, especially HF, is observed at elevated temperatures in presence of water [37].

To avoid these stability and corrosion problems, the use of completely halogen-free ILs was suggested [38], [39]. N-alkyl-3-methylimidazolium alkylsulfates were tried as readily accessible from cheap starting materials. Several studies of liquid–liquid equilibrium data of different fuel components and S-containing compounds were done strengthening conclusions of suitability of [EMIM][EtSO4] as solvent extractant [40], [41]. It might look inconsistent to use S-containing ILs for desulfurization but leaching of IL into the oil phase is in any case unwanted. Nevertheless, certain degree of cross-solubility of hydrocarbons in the IL was observed. In case of [BMIM][OcSO4] distribution coefficients for S-compounds is two times higher than for [EMIM][EtSO4] but the cross-solubility is by a factor 4–16 times higher.

Also, N-alkyl-3-methylimidazolium bis(trifluoromethylsulfonyl)imide ionic liquids were studied and compared to tetrafluoborate or alkylsulfate ones [42], [43], [44], [45], not showing definitive conclusions about the better suitability of these ILs for this application. Searching in a different way, Huang et al. [46] proposed the use of CuCl-based ionic liquids, also moisture-insensitive and stable in air. These ILs showed remarkable desulfurization ability attributed to the π-complexation of Cu(I) with aromatic S-containing compounds.

As cheaper option for real industrial application, the use of imidazolium-based phosphoric ionic liquids was also analysed [37], [47], [48], [49] with encouraging results. It was showed that desulfurization ability of the studied ILs varies following the order [BMIM][DBP] > [EMIM][DEP]  [MMIM][DMP], however fuel solubility in ILs is noticeable following the same tendency. [EMIM][DEP] was the recommended IL. Later on the same authors suggest the use of [EEIM][DEP] [50]. Nie et al. [51] studied the effect of the length of the alkyl group of these cations for alkylphosphate ionic liquids, showing as expected that the longer the alkyl substitute of the anion or cation, the higher the partition coefficient value for the studied ionic liquid.

Within all cation–anion combination options which offer the huge family of ionic liquids Holbrey et al. [52] investigated the DBT extraction power from dodecane for a range of ionic liquids varying cation and anion classes. Polyaromatic quinolinum-based ionic liquids were found to show the best extractive ability, nevertheless, these and other polyaromatic cations, such as 1,3-dibenzimidazolium are technically limited by their high melting points and only tend to form low-melting point ionic liquids when combined with highly flexible perfluorinated anions such as bis(trifluoromethylsulfonyl)amide. This studio ranks ionic liquids desulfurization ability by cation following the secuence methylpiridinium  pyridinium  imidazolium  pyrrolidinium with much less significant variation with anion type. Following this line, Gao et al. [53], [54] perform several extraction experiments with alkyl- and alkylmethyl-pyridinium-based ionic liquids, but best results were found for [alkyl-dmpy]+ ionic liquids. With these cations, ethanoate and thiocyanate anions gave the best extraction performance (consistent with Domanska et al. [55] conclusions in their equilibrium studies). In this way, these more benign and cheaper anions could be used as alternative to the perfluorinated and tetrafluoborate ones.

Continuum solvation models based on the conductor like screening model (COSMO) along with its extension to real solvents (COSMO-RS) also reinforce the idea of suitability of methylpyridinium ionic liquids as solvents for desulfurization [56].

A more complex option is the use of Oxidative Desulfurization (ODS), which involves in a first step oxidizing sulfur compounds, such as DBT and its derivatives, to transform them into sulfones and sulfoxides [57]. In a second step these oxidized products are removed by selective extraction with polar solvents, such as dimethyl sulfoxide.

Taking a step forward, Lu et al. [58] described in their work a combination of ODS and extraction with BF4 ILs. It was proved that desulfurization yield is increased by about one order of magnitude compared with merely extracting with ILs. Several authors had already tried this idea adding H2O2 as oxidant with several catalysts as acetic acid [59], phosphotungstenic acid [60], decatungstates [61], V2O5 [62], and several different ionic liquids (quaternary ammonium coordinated ionic liquids [63], n-methyl pyrrolidinium ionic liquids [64] or even task-specific ionic liquids [65]) as extractants. Zhu et al. [64] demonstrated using peroxotungsten and peroxomolybdenum complexes in ILs that when H2O2 and catalyst are introduced together, the removal of sulfur increased sharply. Zhao et al. [63], [66], [67], [68] studied ODS using [HNMP][BF4] and [HNMP][H2PO4] which act also themselves as catalysts. However, imidazolium or pyrrolidinium-based ionic liquids are relatively expensive and impossible to use in ODS for industry. Lewis acidic ionic liquids containing metal halide anions such as AlCl4 [31], [69], FeCl4 [70] and CuCl2 [46] or ZnCl2 [71] are being studied.

In this work, suitability of a new pyridinium ionic liquid as solvent in extraction of sulfur- and nitrogen-containing compounds from fuels is analysed. Liquid + liquid equilibrium data for ternary systems composed by 1-hexyl-3,5-dimethyl pyridinium {bis[trifluoromethylsulfonyl]imide}, thiophene and three hydrocarbons representative of fuel (n-hexane, n-dodecane and n-hexadecane), and for the ternary system composed by 1-hexyl-3,5-dimethyl pyridinium {bis[trifluoromethylsulfonyl]imide}, pyridine and hexane were determined at 298.15 K and atmospheric pressure. Experimental data were correlated with UNIQUAC and NRTL models. Three steps extraction experiments with simulated and real, previous to refinery desulfurization process, fuel oils were carried out. Results were compared with previously obtained by using other ILs [35], [72]. Study was not focused just on the desulfurization yield but also in the effect of using this ionic liquid, as solvent for extraction, on the rest of the fuel constituents, pointing attention to aromatic compounds.

Section snippets

Chemicals

n-Hexane, n-dodecane, n-hexadecane, and thiophene were supplied by the Aldrich Chemical Company with quoted purities of 0.995, 0.99 and 0.99 mass fraction; pyridine was supplied by Riedel-de Haën with quoted purity of 0.995 mass fraction. Purities were tested by gas chromatography (GC) analysis which did not detect any appreciable peaks of impurities and products were used as received from supplier without further purification.

The ionic liquid [hmmpy][Ntf2] was synthesised in our laboratory

Experimental LLE data

The LLE data for [hmmpy][Ntf2] + thiophene + hexane, [hmmpy][Ntf2] + thiophene + dodecane, [hmmpy][Ntf2] + thiophene + hexadecane and [hmmpy][Ntf2] + pyridine + hexane ternary systems at 298.15 K are reported in Table 3, Table 4, Table 5, Table 6, showing also values of the solute distribution ratio (β) and selectivity (S) defined by following equations:β=x2IIx2IS=x2IIx1Ix2Ix1IIwhere x represents mole fraction, subscripts 1 and 2 refer to hydrocarbon and solute (thiophene or pyridine) and superscripts I and

Conclusions

A large amount of research has been done about the use of ILs for fuel-oils desulfurization. Bibliography focuses on alkylmethyl-pyridinium-based ILs as possible extractant agents to afford legal requirements without the use of more complex techniques as oxidative extraction. In this work suitability of the use of ionic liquid [hmmpy][NTf2] as solvent for extraction of sulfur and nitrogen compounds from fuel oils was analysed.

Liquid + liquid equilibrium data for ternary systems composed by

Acknowledgments

The authors are grateful to the Ministerio de Educación y Ciencia (Spain) for financial support under project CTQ2006-07687. M. Francisco also wants to thank them for the award of the FPI grant with reference BES-2007-16693 through the same project.

References (81)

  • R. Shafi et al.

    Catal. Today

    (2000)
  • C. Song

    Catal. Today

    (2003)
  • I.V. Babich et al.

    Fuel

    (2003)
  • C. Song et al.

    Appl. Catal. B

    (2003)
  • S. Brunet et al.

    Appl. Catal. A

    (2005)
  • E. Ito et al.

    Catal. Today

    (2006)
  • T. Viveros-García et al.

    Chem. Eng. J.

    (2005)
  • L. Alonso et al.

    J. Chem. Thermodyn.

    (2008)
  • J. Zhou et al.

    Fuel Process. Technol.

    (2008)
  • L. Alonso et al.

    Fluid Phase Equilib.

    (2008)
  • L. Alonso et al.

    Fluid Phase Equilib.

    (2008)
  • X. Jiang et al.

    Fuel

    (2008)
  • Y. Nie et al.

    Fuel Process. Technol.

    (2008)
  • U. Domanska et al.

    J. Chem. Thermodyn.

    (2009)
  • A.A.P. Kumar et al.

    Fluid Phase Equilib.

    (2009)
  • D. Zhao et al.

    J. Fuel Chem. Technol.

    (2009)
  • F. Li et al.

    Green Chem.

    (2009)
  • J. Crosthwaite

    J. Chem. Thermodyn.

    (2005)
  • European Directive for emissions of light duty vehicles (70/220/CE amended by 99/69/EC as Euro3 and 2002/80/EC as...
  • Directive of the European Parliament and of the Council, Brussels COM (11.05.2001) 241 final (BS EN 590-2004/DIN EN...
  • EPA-Diesel RIA, United States Environmental Protection Agency, Air and Radiation, EPA 420-R-00-026, December 2000,...
  • A. Rothlishberger et al.

    J. Catal.

    (2005)
  • Z. Jiang et al.

    Langmuir

    (2003)
  • X. Ma et al.

    Ind. Eng. Chem. Res.

    (1994)
  • S.K. Beg et al.

    Energy Fuels

    (2004)
  • I. Mochida et al.

    J. Jpn. Petroleum Inst.

    (2004)
  • H. Rang et al.

    Oil Shale

    (2006)
  • Y. Shiraishi et al.

    Ind. Eng. Chem. Res.

    (2001)
  • J.F. Brennecke et al.

    AIChE J.

    (2001)
  • J.S. Wilkes

    Green Chem.

    (2002)
  • R.D. Rogers et al.

    Science

    (2003)
  • J.L. Anderson et al.

    J. Am. Chem. Soc.

    (2002)
  • J.D. Holbrey et al.

    Clean Prod. Proc.

    (1999)
  • K.R. Seddon

    J. Chem. Technol. Biotechnol.

    (1997)
  • M.J. Earle et al.

    Pure Appl. Chem.

    (2000)
  • J.G. Huddleston et al.

    Chem. Commun.

    (1998)
  • H. Zhao et al.

    J. Chem. Technol. Biotechnol.

    (2005)
  • A. Bösmann et al.

    Chem. Commun.

    (2001)
  • C. Phillips

    Energy Fuels

    (2008)
  • L. Shyu, Z. Zhang, Q. Zhang, International PCT Publication, WO 01/40150 A1,...
  • Cited by (0)

    View full text