Development of a Continuous Flow System for the Oxidative

Nov 8, 2013 - Analytical Technology, Sasol Technology (Pty) Limited, P.O. Box 1, .... was provided by Sasol Technology (Pty) Limited (South Africa) an...
0 downloads 0 Views 5MB Size
Article pubs.acs.org/EF

Development of a Continuous Flow System for the Oxidative Desulfurization of Refractory Organosulfur Compounds in Hydrotreated Diesel Adeniyi S. Ogunlaja,† Ryan S. Walmsley,‡ Carol du Sautoy,§ Nelson Torto,† and Zenixole R. Tshentu∥,* †

Department of Chemistry, Rhodes University, P.O. Box 94, Grahamstown 6139, South Africa Research & Development, Sasol Technology (Pty) Limited, P.O. Box 1, Sasolburg 1947, South Africa § Analytical Technology, Sasol Technology (Pty) Limited, P.O. Box 1, Sasolburg 1947, South Africa ∥ Department of Chemistry, Nelson Mandela Metropolitan University, P.O. Box 77000, Port Elizabeth 6031, South Africa ‡

S Supporting Information *

ABSTRACT: Two synthesized oxovanadium(IV) polymer-supported catalysts, poly[VO(sal-AHBPD)] and poly[VO(allylSBco-EGDMA)], were employed for the oxidation of thiophene (TH), benzothiophene (BTH), dibenzothiophene (DBT), and 4,6dimethyldibenzothiophene (4,6-DMDBT) under a continuous flow system at 40 °C. Maximum conversion was attained at a flow rate of 1 mL/h, with overall conversions of 71%, 89%, 99%, and 88% achieved when poly[VO(allylSB-co-EGDMA)] was employed, and conversions of 79%, 90%, 97%, and 91% were achieved when poly[VO(sal-AHBPD)] was employed for the oxidation of thiophene (TH), benzothiophene (BT), dibenzothiophene (DBT), and 4,6 dimethyldibenzothiophene (4,6DMDBT), respectively. The high organosulfur oxidation yield displayed by this method, followed by adsorption of the sulfones using chitosan nanofibers, suggests its possible application in the oil industry for oxidative desulfurization (ODS) of hydrotreated fuel.



catalyst support surface leading to the loss in catalyst activity.17 The continuous flow process is seen as an emerging alternative in developing reaction methodology due to its importance in reducing loss of catalyst activity for organic reactions.18−20 Continuous flow processes have been applied generally in biocatalysed processes21 as well as in catalyzed oxidation processes.22,23 It is also a process that can be controlled, since reactants flow in an organized direction within the reactor, unlike the batch technique in which reactants are not easily controlled. This study compared the catalytic oxidative activity of two polymer- anchored oxovanadium(IV) catalysts, poly[VO(salAHBPD)] and poly[VO(allylSB-co-EGDMA)] (Figure 1), on some organosulfur compounds present in fuels under a continuous flow process. The syntheses protocol and performance of the catalysts have been reported in our earlier publications.15,16

INTRODUCTION Oxidative desulfurization (ODS) of transport fuels has been proposed as a complementary option to hydro desulfurization (HDS) techniques in the elimination of refractory sulfur compounds from fuel.1−6 The ODS process involves the oxidation of organosulfur compounds at a relatively low temperature and at atmospheric pressure to generate sulfone compounds that can be removed from fuels through extraction or adsorption in order to meet sulfur levels of 10−15 ppm proposed by the U.S. environmental protection agency (EPA) and European committee for standardization and regulation of fuels before 2015.7 Lyondell chemicals8−10 and EniChem/UOP11,12 independently patented the ODS process. In the Lyondell process, solvent extraction was used for separation of sulfones while EniChem/ UOP process removed the sulfone species by adsorption. In both processes, tert-butylhydroperoxide was employed as the oxidant due to its solubility in fuel.13,14 Lyondell chemicals and EniChem/UOP ODS processes produced tert-butyl alcohol after the oxidation step, and this compound was employed as potential octane-booster compound for gasoline. The oxidation of organosulfur compounds occurs in the presence of catalysts, and these catalysts are mostly transition metals in high oxidation states.15 The interaction between the catalyst and peroxide results in the formation of a peroxidocomplex, which is the active species in the oxidation of organosulfur compounds.15 Immobilization of the catalysts unto a heterogeneous support promotes their reusability.15,16 Most catalyzed oxidation processes are carried out under batch process conditions involving the mechanical stirring of both the reactants and catalysts; however, such conditions damage the © 2013 American Chemical Society

Figure 1. Chemical structures of poly[VO(sal-AHBPD)] and poly[VO(allylSB-co-EGDMA)] (“P” represents cross-linked polymer beads). Received: August 26, 2013 Revised: October 24, 2013 Published: November 8, 2013 7714

dx.doi.org/10.1021/ef401708f | Energy Fuels 2013, 27, 7714−7723

Energy & Fuels

Article

Figure 2. Schematic representation of the setup of a continuous flow system for the catalyzed oxidation process (inset: picture of the setup).



(AED), and mass spectrometry (MS) were optimized for better separation. Agilent gas chromatograph (7890A) fitted with an FID detector was equipped with a polar Zebron DB-5MS (30 m × 0.25 mm × 0.25 μm) column. The analysis on model organosulfur compounds were reported by using a temperature heating ramp from 80 to 100 °C for 2 min and then increased from 100 to 320 °C at a heating rate of 30 °C·min−1. Agilent gas chromatograph (6890N) fitted with an Agilent G2350A AED was used to monitor the oxidation of sulfur in hydrotreated diesel. A Restek-Rxi 5HT (30 m × 0.25 mm × 0.25 μm) column was employed for the analysis, with an initial temperature heating ramp from 45 to 150 °C at 10 °C·min−1, which was then held for 5 min, and finally increased from 150 to 320 °C at a heating rate of 20 °C·min−1 and held for 3 min. The AED detector for sulfur analysis was measured at a wavelength of 181 nm with a transfer line temperature of 350 °C and block cavity temperature of 320 °C. An Agilent gas chromatograph (7890A) fitted with an electron impact ionization mass selective triple-axis detector (5975C VL MSD) was used to confirm the oxidized products, the chromatograph was also equipped with a Zebron DB-5MS (30 m × 0.25 mm × 0.25 μm) column. Source temperature of 250 °C, an interface temperature of 200 °C and a scan range of m/z 60 to 650 was employed on the MS. Oxidized Product Analysis. Product analysis of the oxidized hydrotreated diesel was conducted on a Bruker, Tensor 27 platinum ATR-FTIR spectrometer (4000−400 cm−1) and also on a gas chromatograph−mass spectrometer (GC-MS) using the procedure described under gas chromatography analytical conditions. Oxovanadium(IV) Leaching Determination. Vanadium that leached out from both polymer supports in poly[VO(sal-AHBPD)] and poly[VO(allylSB-co-EGDMA)] were determined by analyzing the vanadium content before and after use. A known mass of the catalysts was weighed into separate vials, after which 10 mL of TraceSelect HNO3 (69%) was added to each vial. The mixture was then heated up to a temperature of 40 °C for 12 h to leach out the vanadium. The acidleached solution was then diluted with Millipore water to 100 mL and analyzed using Thermo Electron (iCAP 6000 Series) inductively coupled plasma (ICP) spectrometer equipped with an optical emission spectrometry (OES) detector; wavelengths with minimum interferences chosen for the analysis were 290.88, 292.40, 309.31, and 311.07 nm. General Procedure for Catalyzed Oxidation of Model Organosulfur Compounds. The respective catalytic oxidation of

EXPERIMENTAL SECTION

Materials. Thiophene (Merck Chemicals, Germany), benzothiophene (Merck Chemicals, Germany), dibenzothiophene (Merck Chemicals, U.S.A.), 4,6-dimethyldibenzothiophene (Sigma-Aldrich, South Africa), tert-butylhydroperoxide (Sigma-Aldrich, Germany), and tetrabutylammonium bromide (TBAB) (98%, Sigma-Aldrich, U.S.A.) were used as received. Hydrotreated diesel containing 385 ppm sulfur was provided by Sasol Technology (Pty) Limited (South Africa) and employed for the oxidation reaction. Total sulfur in diesel standard was purchased from Matheson Tri-Gas, Texas, U.S.A. Hexane, methanol, acetonitrile, and toluene (HPLC grade) were procured from Merck chemicals (South Africa). Catalyst Synthesis. Merrifield beads were functionalized with a tetradentate N2O2-donor ligand (sal-AHBPD). After which, the resulting beads poly[(sal-AHBPD)] were further suspended into 20 mL N,N-dimethylformamide solution containing 1 g (0.0061 mol) of [VOSO4·3H2O] while stirring at 60 °C for 6 h to produce poly[VO(sal-AHBPD)] (Figure 1, Supporting Information (SI) Scheme S1).15 For poly[VO(allylSB-co-EGDMA)], 3-allyl-2-hydroxybenzaldehyde was synthesized by adding 0.86 mL (0.01 mol) of allylbromide to a solution of 0.65 mL (0.01 mol) of salicyaldehyde in acetone under reflux for 6 h. Upon refluxing, the observed intermediate product, 2-(allyloxy)benzaldehyde undergoes Clasien rearrangement reaction to form the desired product. N,N′-bis(o-hydroxybenzaldehyde)phenylenediamine momoner (allylSB) was then synthesized by reacting 1.0 g (0.0062 mol) of 3-allyl-2hydroxybenzaldehyde with 0.33 g (0.0081 mol) o-phenylenediamine at 50 °C for 3 h. Four grams (0.01 mol) of the functional monomer N,N′bis(o-hydroxybenzaldehyde)phenylenediamine (allylSB) were polymerized, via suspension polymerization, with 2 mL (0.0075 mol) of ethyleneglycol dimethacrylate (EGDMA) as a cross-linking agent, using 0.2 g (0.001 mmol) of azobis(isobutyronitrile) (AIBN) as initiator to produce poly[allylSB-co-EGDMA] beads. Two grams of poly[allylSB-coEGDMA] beads were swollen in 30 mL of DMF for 2 h in the presence of excess VOSO4.3H2O to produce poly[VO(allylSB-co-EGDMA)] (Figure 1, SI Scheme S2).16 The SEM images of poly[VO(allylSB-co-EGDMA)] and poly[VO(sal-AHBPD)] are available in the Supporting Information (Figures S1 and S2). Gas Chromatography Analytical Conditions. The GC conditions for the flame ionization detection (FID), atomic emission detection 7715

dx.doi.org/10.1021/ef401708f | Energy Fuels 2013, 27, 7714−7723

Energy & Fuels

Article

thiophene (TH), benzothiophene (BTH), dibenzothiophene (DBT), and 4,6-dimethyldibenzothiophene (4,6-DMDBT) was carried out in a 25 mL syringe by employing poly[VO(sal-AHBPD)] and poly[VO(allylSB-co-EGDMA)] as catalysts, respectively. In a typical reaction vessel, an aqueous solution of tert-butylhydroperoxide (t-BuOOH) and organosulfur compounds {thiophene (TH), benzothiophene (BTH), dibenzothiophene (DBT), and 4,6-dimethyldibenzothiophene (4,6DMDBT)} were dissolved in 10 mL solution of toluene/hexane (1:4). This was transferred to a 25 mL syringe (with 2 cm internal diameter) attached to a tip (0.5 cm internal diameter) packed with a known mass of catalyst (Figure 2). The catalyzed oxidation reaction was conducted at 40 °C by using a heating blanket. The flow-rates 1, 3, 5, 7, and 10 mL/h were used to pump the reaction mixture through the catalyst bed to get the desired product by optimizing mass transfer rates and overcoming limitations caused by the mass transfer.



RESULTS AND DISCUSSION BET Surface Measurements. The Brunauer−Emmett− Teller (BET) surface parameters of the catalysts poly[VO(salAHBPD)] and poly[VO(allylSB-co-EGDMA)] are provided in Table 1. Poly[VO(allylSB-co-EGDMA)] presents a higher surface Table 1. Particle size and BET surface parameters of poly[VO(sal-AHBPD)] and poly[VO(allylSB-co-EGDMA)] poly[VO(sal-AHBPD)] poly[VO(allylSB-co-EGDMA)] 2

surface area (m /g) pore size (Å) t-plot micropore vol. (cm3/g) particle sizes (μm)

7 181 8 × 10−4

23 135 3 × 10−3

262−306

286−308

Figure 3. 51V NMR spectrum of (a) solution of 4 mM [VIVO(sal-HBPD)] in 1 equiv. t-BuOOH in MeOH; (b) suspension of [VVO(sal-HBPD)] after addition of 3 equiv. t-BuOOH in MeOH; and (c) suspension of [VVO(sal-HBPD)] after addition of 7 equiv. t-BuOOH.

Tetrabutylammonium bromide (TBAB) (0.01 g, 0.031 mmol) was added to thiophene solution during its catalytic oxidation in order to interrupt conjugation of thiophene ring, hence making the sulfur atom within the ring reactive.16,25 Oxidation in the presence of poly[VO(sal-AHBPD)] gave higher overall conversion of 79% (Table 2, Figure 4a) at a flow-rate of 1 mL/h with a sulfone/ sulfoxide ratio of 1.3, while at a higher flow-rate of 10 mL/h the overall conversion dropped to 61% with a decrease in the sulfone/ sulfoxide ratio to 0.9. The catalyst poly[VO(allylSB-co-EGDMA)], under the same oxidation conditions as poly[VO(sal-AHBPD)], presented an overall conversion of 71% with a sulfone/sulfoxide ratio of 1.7 at a flow-rate of 1 mL/h (Table 2, Figure 4b). At a flowrate of 10 mL/h an overall conversion of 51% was recorded with a sulfone/sulfoxide ratio of 1.7. In both catalytic oxidation studies of thiophene, it was observed that overall conversions decreased as the flow-rate increased, thus leading to higher sulfoxide yield; this is due to the reduced residence time as flow-rate increases. The catalyzed oxidation of benzothiophene gave a much higher conversion when compared to thiophene owing to the higher sulfur electron density displayed by benzothiophene.16 For poly[VO(sal-AHBPD)] catalyzed oxidation, the flow-rate of 1 mL/h presented the highest overall conversion of 90% (Table 2, Figure 5a) with an observed sulfone/sulfoxide ratio of 14.0, while at a higher flow-rate of 10 mL/h an overall conversion of 83% was observed with a decrease in the sulfone/sulfoxide to 1.8. An overall conversion of 89% was obtained when poly[VO(allylSBco-EGDMA)] was employed at a flow-rate of 1 mL/h (Table 2). At a flow-rate of 1 mL/h, a sulfone/sulfoxide ratio of 16.8 was observed (Table 2, Figure 5b), but on increasing the flow-rate to 10 mL/h, sulfone/sulfoxide ratio also decreased to 4.9 with a reduction in the overall conversion to 75%. The oxidation of dibenzothiophene by poly[VO(salAHBPD)] at a flow rate of 1 mL/h gave an overall conversion of 98% (Table 2, Figure 6a), and the conversion, however,

area than poly[VO(sal-AHBPD)], the difference can be attributed to the method of synthesing the catalysts. V (wt %) of poly[VO(allylSB-co-EGDMA)] and poly[VO(sal-AHBPD)] were 2.83 and 5.64, respectively. Other catalysts characterization parameters are reported in our previous publications. 15,16 51 V NMR Studies. The 51V spectroscopic measurements were carried out to investigate the speciation of [VIVO(salHBPD)] upon interaction with peroxide by using a Bruker 400 MHz spectrometer and VOCl3 was used as a reference in MeOH as solvent (Figure 3). A 51V NMR spectrum of the oxidized [VIVO(sal-HBPD)], (ca. 4 mM) dissolved in MeOH showed a strong resonance at δ = −568 ppm. Stepwise addition of an aqueous peroxide solution to a solution of [VVO(sal-HBPD)] generated the peroxido-vanadium(V) complex species. The addition of 3.0 equivalent solution of t-BuOOH to [VVO(salHBPD)] did not give much change except for the broadening of signals around −568 ppm with a smaller resonance peak appearing around −598 ppm. At 7.0 mol equivalent of t-BuOOH, a new resonance peak appeared at −642 ppm and was assigned to VVO(O2)-species (Figure 3).24 Leaving the NMR tube open for 24 h at room temperature, the resonance at −642 ppm disappear and resonance around −570 regains intensity, indicating the reversibility of the process. The general mechanism for V(IV) oxidation is presented in the Supporting Information (Scheme S3). UV−vis titration studies further confirming the evidence of formation of dioxido and oxidoperoxido species is reported in a previous publication.15 Continuous Flow Oxidation of Organosulfur Compounds. The catalyzed oxidation of thiophene, benzothiophene, and dibenzothiophene was conducted at a stoichiometric mole ratio of 6.8:1 (t-BuOOH-to-substrates), while 4,6dimethyldibenzothiophene oxidation was carried out in a ratio of 7.6:1 (t-BuOOH-to-4,6-dimethyldibenzothiophene). 7716

dx.doi.org/10.1021/ef401708f | Energy Fuels 2013, 27, 7714−7723

Energy & Fuels

Article

Table 2. Catalysed Oxidation of Organo-sulfur Compounds Using t-BuOOH in the Presence of p[VO(sal-AHBPD) (Cat 1) and Poly[VO(allylSB-co-EGDMA)] (Cat 2) % sulfoxides

% sulfones

% conversions

sulfone/sulfoxide ratio

TOF (h−1)

substrates

flow rate (mL/h)

Cat 1

Cat 2

Cat 1

Cat 2

Cat 1

Cat 2

Cat 1

Cat 2

Cat 1

Cat 2

THa THa THa THa THa BTH BTH BTH BTH BTH DBT DBT DBT DBT DBT 4,6-DMDBT 4,6-DMDBT 4,6-DMDBT 4,6-DMDBT 4,6-DMDBT

1 3 5 7 10 1 3 5 7 10 1 3 5 7 10 1 3 5 7 10

34 36 22 26 32 6 10 20 20 30 ND ND ND ND ND 1 5 10 12 15

26 32 36 20 19 5 2 8 15 12 ND ND ND ND ND 14 14 11 15 13

45 40 53 38 29 84 78 67 66 53 98 97 97 97 96 90 70 57 52 39

45 36 29 33 32 84 84 77 66 63 99 98 98 97 93 74 67 66 58 47

79 76 75 64 61 90 88 87 86 83 98 97 97 97 96 91 75 67 64 54

71 68 65 53 51 89 86 85 81 75 99 98 98 97 93 88 81 77 73 60

1.3 1.1 2.4 1.5 0.9 14.0 7.8 3.8 3.3 1.8 98.0 97.0 97.0 97.0 96.0 90.0 14.0 5.7 4.3 2.6

1.7 1.1 0.8 1.7 1.7 16.8 42.0 9.6 4.4 4.9 99.0 98.0 98.0 97.0 93.0 5.3 4.8 6.0 3.9 3.6

1498 1441 1422 1214 1157 293 287 284 280 271 235 233 233 234 230 191 158 141 135 113

1346 1289 1233 1005 967 290 280 277 264 244 237 235 235 233 223 185 170 162 154 126

Tetrabutylammonium bromide (TBAB) (0.01 g, 0.031 mmol) added to thiophene (TH) moles of catalyst (vanadium content) employed = 1.35 × 10−5 moles. Thiophene = 0.54 g (0.0064 mol); benzothiophene = 0.15 g (0.0011 mol); dibenzothiophene = 0.15 g, (0.00081 mol); 4,6dimethyldibenzothiophene = 0.15 g(0.00071 mol). TOF (h−1) = (% conv.) × (substrate moles)/catalyst (vanadium) moles × h (time). Maximum conversion time of 1 h was employed. a

Figure 4. Graph of % conversion for the oxidation of thiophene (TH). [t-BuOOH)eq.] = 6.8; temp. = 40 °C; toluene/hexane (1:4) = 10 mL. Using (A) poly[VO(sal-AHBPD)] = 0.015 g (0.0135 mmol) and (B) poly[VO(allylSB-co-EGDMA)] = 0.024 g (0.0135 mmol). SO = sulfoxide and SOO = sulfone.

decreased to 96% when a flow rate of 10 mL/h was employed. When poly[VO(allylSB-co-EGDMA)] was employed as catalyst, an overall conversion of 99% occurred at a flow-rate of 1 mL/h (Table 2, Figure 6b), but upon increasing the flow-rate to 10 mL/h the overall oxidation reduced to 93%. The high conversion of dibenzothiophene was attributed to the high electron density of sulfur atom in dibenzothiophene, making it much more reactive compared to thiophene and benzothiophene. The catalyzed oxidation of 4,6-dimethyldibenzothiophene presented a reduction in its overall conversions as compared to dibenzothiophene due to its bulkiness, thus sterically hindering the reactive sulfur atom from being oxidized. With

poly[VO(sal-AHBPD)] as the oxidation catalyst, an overall conversion of 91% at a flow-rate of 1 mL/h was recorded with a sulfone/sulfoxide of 90.0 (Figure 7a, Table 2), while at a higher flow-rate of 10 mL/h the overall conversion decreased to 57% and the sulfone/sulfoxide ratio decreased to 2.6. Likewise, when poly[VO(allylSB-co-EGDMA)] was employed for oxidation at a flow-rate of 1 mL/h an overall conversion of 88% was achieved (Table 2, Figure 7b), while at a flow-rate of 10 mL/h the overall conversion decreased to 60%. The sulfone/sulfoxide ratio decreased from 5.3 to 3.6 from 1 mL/h to 10 mL/h. From the overall catalyzed oxidation of the organosulfur compounds, overall conversions and turnover frequencies 7717

dx.doi.org/10.1021/ef401708f | Energy Fuels 2013, 27, 7714−7723

Energy & Fuels

Article

Figure 5. Graph of % conversion for the oxidation of benzothiophene (BTH). [t-BuOOH)eq.] = 6.8; temp. = 40 °C; toluene/hexane (1:4) = 10 mL. Using (A) poly[VO(sal-AHBPD)] = 0.015 g (0.0135 mmol) and (B) poly[VO(allylSB-co-EGDMA)] = 0.024 g (0.0135 mmol). SO = sulfoxide and SOO = sulfone.

Figure 6. Graph of % conversion for the oxidation of dibenzothiophene (DBT). [t-BuOOH)eq.] = 6.8; temp. = 40 °C, toluene/hexane (1:4) = 10 mL. Using (A) poly[VO(sal-AHBPD)] = 0.015 g (0.0135 mmol) and (B) poly[VO(allylSB-co-EGDMA)] = 0.024 g (0.0135 mmol). SO = sulfoxide and SOO = sulfone.

Figure 7. Graph of % conversion for the oxidation of 4,6-dimethyldibenzothiophene (4,6-DMDBT). [t-BuOOH)eq.] = 7.6; temp. = 40 °C; toluene/ hexane (1:4) = 10 mL. Using (A) poly[VO(sal-AHBPD)] = 0.015 g (0.0135 mmol) and (B) poly[VO(allylSB-co-EGDMA)] = 0.024 g (0.0135 mmol). SO = sulfoxide and SOO = sulfone. 7718

dx.doi.org/10.1021/ef401708f | Energy Fuels 2013, 27, 7714−7723

Energy & Fuels

Article

Figure 8. GC-FID chromatograms of organosulfur mixture (benzothiophene (BTH), dibenzothiophene (DBT) and 4,6-dimethyldibenzothiophene (4,6-DMDBT)) before and after oxidation using poly[VO(sal-AHBPD)] and poly[VO(allylSB-co-EGDMA)]. Time of reaction = 10 h.

(TOFs) decreased steadily as flow-rate increased. The sulfone/ sulfoxide ratio also increased as the flow-rate decreased indicating that more sulfones were formed at lower flow-rates as a result of the longer contact time and mass transfer between the reactant mixture and catalyst. The relative role that liquid mass transfer and transport mechanisms play in the catalyzed oxidation reaction system is important, as chemical species movement from an area of high chemical potential to an area low chemical potential are encouraged via diffusion. It is to be noted that the main factors governing mass transfer are diffusion, capillary flow, surface diffusion, and hydrodynamic mechanisms.26 Vanadium(IV) Leaching Studies. Leaching of vanadium(IV) incorporated in polymer beads was investigated in the oxidation of organosulfur compounds under flow conditions. The quantity of vanadium leached from oxidation reactions was determined as described in Experimental Section. The leaching was observed to decrease as the flow rate increased. The leaching behavior was ascribed to the reactant contact time of the reacting solution with the catalyst and also to the degree of bulkiness of the organosulfur compounds. Oxidation of thiophene required less amount of oxidant as compared to 4,6-DMDBT, a sterically hindered compound. The excess oxidant (t-BuOOH) in the system during oxidation leaches out vanadium metal from its polymer support matrix, as the theoretically required substrateto-oxidant stoichiometric ratio of 1:2 is ineffective to carry-out

complete oxidation. Vanadium leaching is in the order of TH > BT > DBT > 4,6-DMDBT, order of bulkiness of the organosulfur compounds, which is related to the amount of oxidant required. Generally, low vanadium quantities were leached from both catalysts (Tables 3), and this indicated that the vanadium species within the functionalized polymer support were stable. Oxidation of a Mixture of Model Organosulfur Compounds (Model Fuel). Equivalent mass (0.15 g) of refractory sulfur containing compounds {benzothiophene (3807 ppm S), dibenzothiophene (2606 ppm S), and 4,6-methyldibenzothiophene (2239 ppm S)} was dissolved in 10 mL solution of toluene/hexane (1:4), after which an aqueous solution of tert-butyl hydroperoxide (t-BuOOH) (maximum equivalent oxidant moles to each compound) was added. The mixture was transferred into a 25 mL syringe, which was then pumped through the catalyst as illustrated in Figure 2 to achieve the desired products. Poly[VO(allylSB-co-EGDMA)] (0.072 g, 0.0405 mmol) and poly[VO(sal-AHBPD)] (0.044 g, 0.0405 mmol) were employed as catalysts, and a flow rate of 1 mL/h was maintained for the catalyzed oxidation studies. The 0.072 g of poly[VO(allylSB-co-EGDMA)] and 0.044 g of poly[VO(salAHBPD)] contain the same amount of vanadium. From the results obtained (see chromatograms, Figure 8), most of the dibenzothiophene and 4,6-dimethyldibenzothiophene was converted to their respective sulfone compounds 7719

dx.doi.org/10.1021/ef401708f | Energy Fuels 2013, 27, 7714−7723

Energy & Fuels

Article

Figure 9. GC-MS chromatograms for (A) benzothiophene sulfone (BTO2), (B) dibenzothiophene sulfone (DBTO2), and (C) 4,6dimethyldibenzothiophene sulfone (4,6-DMDBTO2).

(mainly dibenzothiophene and dibenzothiophene derivatives) was utilized for oxidation. A similar setup as described under the catalyzed oxidation for model compounds was also employed in the oxidation of hydrotreated diesel using a relative ratio (O/S ratio, 7). The feed (hydrotreated diesel and t-BuOOH in hexane) was preheated to 40 °C and then channelled into the catalystcontaining compartment with the aid of a syringe pump. It is to be noted that the catalyst poly[VO(allylSB-co-EGDMA)] (0.072 g, 0.0405 mmol) was employed for the oxidation due to the higher overall oxidation yield it provided on the oxidation of model compounds. Progress in oxidation was monitored by a gas chromatograph equipped with atomic emission detection (AED), which assists in the selective detection and tracking of elements of interest within a complex organic matrix.27−29 From the 181 nm sulfur chromatograms (Figure 10), a shift in retention time was noticed from the initial organosulfur compounds to sulfur as sulfone compounds in the oxidized hydrotreated diesel, while the carbon remained unchanged. The absence of unaltered peaks indicated that all the sulfur compounds were oxidized completely; and further confirmation by using a gas chromatograph equipped with mass spectrometer (GC-MS) indicated that organosulfur compounds such as 4,6-dimethyldibenzothiophene (m/z 212), 2,6- and/or 2,7- dimethyldibenzothiophene (m/z 212), C2dibenzothiophene (m/z 212), C3-dibenzothiophene (m/z 226),

after passing through the catalyst. The chromatogram showed that 60% of benzothiophene was oxidized to sulfone when poly[VO(allylSB-co-EGDMA)] was employed, and 45% of benzothiophene was oxidized when poly[VO(sal-AHBPD)] was employed as catalyst. A total of 98% and 96% oxidation of dibenzothiophene and 4,6-dimethyldibenzothiophene, respectively, was observed for poly[VO(allylSB-co-EGDMA)] catalyst, while a total of 86% and 73% oxidation was observed for dibenzothiophene and 4,6-dimethyldibenzothiophene, respectively, when poly[VO(salAHBPD)] was employed as catalyst. The low electron density of the benzothiophene sulfur atom was attributed to the low oxidation observed, while the higher electron sulfur density compounds, dibenzothiophene and 4,6-dimethyldibenzothiophene, gave a much higher oxidation yield. Confirmation of the respective sulfone compounds formed was carried out via the use of a gas chromatography−mass spectrometry (Figure 9). The catalytic oxidation reactivity of the sulfur-containing compounds followed the order: BTH < 4,6-DMDBT < DBT. The order of reactivity was governed by the electron density of sulfur and order of bulkiness (steric hindrance) of the organosulfur compounds. This oxidation activity order for sulfur-containing compounds agreed well with the oxidation results obtained under batch studies.15,16 Oxovanadium(IV)-Catalyzed Oxidation of Hydrotreated Diesel. Hydrotreated diesel fuel containing 385 ppm of sulfur 7720

dx.doi.org/10.1021/ef401708f | Energy Fuels 2013, 27, 7714−7723

2.113 (0.003) 1.019 (0.003) 1.029 (0.002) 0.411 (0.004)

Article

2.333 (0.002) 1.239 (0.003) 1.029 (0.003) 0.529 (0.005) 1.467 (0.002) 0.832 (0.003) 0.721 (0.003) 0.301 (0.001)

Cat 1

1.637 (0.002) 1.089 (0.003) 0.721 (0.004) 0.332 (0.003)

Cat 2

1.233 (0.006) 0.672 (0.004) 0.412 (0.002) 0.203 (0.004)

5 1.393 (0.002) 0.849 (0.003) 0.412 (0.001) 0.241 (0.004) 1.124 (0.005) 0.453 (0.002) 0.232 (0.005) 0.132 (0.003)

Cat 1 Cat 2

Figure 10. GC-AED chromatograms of carbon at 197 nm (A) and sulfur at 181 nm (B) of the hydrotreated diesel before oxidation and carbon at 197 nm (C) and sulfur at 181 nm (D) chromatograms of the hydrotreated diesel fuel after oxidation while using poly[VO(allylSB-coEGDMA)] as catalyst. The chromatogram shows the loss of sulfides and the formation of sulfones.

Standard deviations SD are given in parentheses.

TH BT DBT 4,6-DMDBT

Figure 11. FT-IR of (A) hydrotreated diesel fuel (385 ppm S) and (B) oxidized hydrotreated diesel fuel.

C4-dibenzothiophene (m/z 240), C12-benzothiophene (m/z 302) were oxidized. Oxidized hydrotreated diesel was characterized by FT-IR spectroscopy with sulfone (SO) band frequencies visible around 1400−1300 cm−1 and 1200−1100 cm−1,16,30 on the oxidized diesel as compared to the nonoxidized diesel (Figure 11).

a

1.041 (0.004) 0.432 (0.003) 0.129 (0.001) 0.110 (0.004) 1.152 (0.002) 0.529 (0.002) 0.129 (0.002) 0.122 (0.001)

1.232 (0.002) 0.699 (0.003) 0.232 (0.004) 0.221 (0.001)

7 Cat 1 Cat 2 10 Cat 1

flow rates (mL/h)

Table 3. % Vanadium Leached from p[VO(sal-AHBPD) (Cat 1) and Poly[VO(allylSB-co-EGDMA)] (Cat 2) Beadsa

3

Cat 2

Cat 1

1

Cat 2

Energy & Fuels

7721

dx.doi.org/10.1021/ef401708f | Energy Fuels 2013, 27, 7714−7723

Energy & Fuels

Article

Scheme 1. Proposed Scheme for Adsorption of Sulfones on Chitosan MIP Nanofibers (Inset: Chitosan Structure)

Figure 12. Sulfur-specific GC-AED chromatograms measured at 181 nm for (A) hydrotreated diesel, (B) hydrotreated diesel after oxidation, and (C) after desulfurization.

Figure 13. GC-AED chromatograms of carbon at 197 nm and sulfur at 181 nm showing desorbed sulfones from imprinted fiber after Soxhlet extraction with equal volume of acetonitrile and methanol.

C−S stretching bands of the hydrotreated diesel was visible around 715 cm−1, but upon oxidizing the diesel the C−S stretching band shifted to around 761 cm−1 (Figure 10). Adsorption of Sulfones from Oxidized Diesel Fuel. The sulfones in oxidized diesel have been removed by employing either a polar extractant such as 1-methyl-2-pyrrolidone31 or by adsorption over silica gel at ambient temperature.1 Preliminary studies using 300 mg of molecularly imprinted electrospun chitosan nanofibers were contacted with 2 mL of oxidized hyrotreated diesel (394 ± 4.2 ppm S) under continuous flow adsorption conditions at a flow-rate of 1 mL/h. A total of 84% sulfones were removed (Figure 12), leaving 62 ± 3.2 ppm sulfur (S) in the hydrotreated diesel. The fabrication of molecularly imprinted chitosan nanofiber is reported elsewhere.32 Adsorption was proposed to take place via hydrogen bonding (Scheme 1) as well as the sulfone imprinting properties.32 Desorption of the

adsorbed sulfones on chitosan was carried out overnight through Soxhlet extraction by using 50 mL solvent mixture ratio 1:1, acetonitrile-to-methanol and most of the absorbed sulfones were desorbed, ∼78% desorption was recorded (Figure 13).



CONCLUSION The polymer-anchored oxovanadium(IV)-based catalysts, poly[VO(sal-AHBPD)] and poly[VO(allylSB-co-EGDMA)], proved to be effective in the catalytic oxidation of organosulfur compounds. A low level of vanadium leaching observed from both catalysts after oxidation confirmed the stability of vanadium within the polymer matrix with poly[VO(sal-AHBPD)] leaching out more vanadium compared to poly[VO(allylSB-co-EGDMA)]. A combination of GC-AED and GC-MS was found to be a suitable combination of techniques for quantification and characterization of sulfur 7722

dx.doi.org/10.1021/ef401708f | Energy Fuels 2013, 27, 7714−7723

Energy & Fuels

Article

(23) Walmsley, R. S.; Hlangothi, P.; Litwinski, C.; Torto, N.; Nyokong, T.; Tshentu, Z. R. J. Appl. Polym. Sci. 2013, 127 (6), 4719−4725. (24) Maurya, M. R.; Arya, A.; Kumar, A.; Kuznetsov, M. L.; Avecilla, F.; Pessoa, J. C. Inorg. Chem. 2010, 49, 6586−6600. (25) Lanju, C.; Shaohui, G. U. O.; Dishun, Z. Chin. J. Chem. Eng. 2006, 14 (6), 835−838. (26) Tomoi, M.; Ford, W. T. J. Am. Chem. Soc. 1980, 102 (23), 7140− 7141. (27) Quimby, B. D.; Grudoski, D. A.; Giarrocco, V. J. Chromatogr. Sci. 1998, 36, 435−443. (28) Amorelli, A.; Amos, Y. D.; Halsig, C. P.; Kosman, J. J.; Jonker, R. J.; de Wind, M.; Vrieling, J. Hydrocarbon Process. 1992, 93−100. (29) Quimby, B. D.; Giarrocco, V.; Sullivan, J. J.; McCleary, K. A. J. High Resolut. Chromatogr. Chromatogr. Commun. 1993, 15, 705−709. (30) Castillo, K.; Parsons, J. G.; Chavez, D.; Chianelli, R. R. J. Catal. 2009, 268, 329−334. (31) Zongxuan, J.; Hongyinga, L.; Yongna, Z.; Can, L. Chin. J. Catal. 2011, 32, 707−715. (32) Ogunlaja, A. S.; Coombes, M. J.; Torto, N.; Tshentu, Z. R. Submitted to React. Funct. Polym. Submitted. 2013.

compounds within the oxidized fuel. The results obtained warrant the possible application of the catalyst, poly[VO(allylSB-coEGDMA)], for the oxidation of refractory sulfur compounds in oxidative desulfurization of fuels.



ASSOCIATED CONTENT

S Supporting Information *

Additional schemes and figures. This material is available free of charge via the Internet at http://pubs.acs.org.



AUTHOR INFORMATION

Corresponding Author

*Tel.: +27 41 504 2074. Fax: +27 41 504 4236. E-mail: zenixole. [email protected]. Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS We are thankful for financial support provided by Sasol and the National Research Foundation (NRF). Nelson Mandela Metropolitan University (NMMU) and Rhodes University are also acknowledged for financial support. Special thanks to Dr Pieter van Heerden, from Sasol Technology (Pty) Ltd., for supplying the hydrotreated diesel sample.



REFERENCES

(1) Ishihara, A.; Wang, D. H.; Dumeignil, F.; Amano, H.; Qian, E. W. H.; Kabe, T. Appl. Catal., A 2005, 279, 279−287. (2) Chica, A.; Gatti, G.; Moden, B.; Marchese, L.; Iglesia, E. Chem. Eur. J. 2006, 12, 1960−1967. (3) Collins, F. M.; Lucy, A. R.; Sharp, C. J. Mol. Catal. A: Chem. 1997, 117, 397−403. (4) Kuznetsova, L.; Detusheva, L.; Kuznetsov, N.; Duplyakin, V.; Likholobov, V. Kinet. Catal. 2008, 49, 644−652. (5) Prasad, V. V. D. N.; Jeong, K. E.; Chae, H. J.; Kim, C. U.; Jeong, S. Y. Catal. Commun. 2008, 9, 1966−1969. (6) González-García, O.; Cedeño-Caero, L. Catal. Today 2009, 148, 42−48. (7) Gasoline Sulfur Standards; 2009. http://www.epa.gov/oms/ standards/fuels/gas-sulfur.htm (assessed June 28, 2013). (8) Karas, L. J.; Han, Y. Z.; Leyshon, D. W. U.S. Patent No. 2004178122, 2004. (9) Han, Y. Z.; Leyshon, D. W. U.S. Patent No. 2005109677, 2005. (10) Karas, L. J.; Grey, R. A.; Lynch, M. W. U.S. Patent No. 2008047875, 2008. (11) Kocal, J.A. U.S. Patent No. 6277271, 2001. (12) Kocal, J. A.; Branvold, T. A. U.S. Patent No. 6368495, 2002. (13) Wang, D. H.; Qian, E. W.; Amano, H.; Okata, K.; Ishihara, A.; Kabe, T. Appl. Catal., A 2003, 253, 91−99. (14) Hulea, V.; Fajula, F.; Bousquet, J. J. Catal. 2001, 198, 179−186. (15) Ogunlaja, A. S.; Chidawanyika, W.; Antunes, E.; Fernandes, M. A.; Nyokong, T.; Torto, N.; Tshentu, Z. R. Dalton Trans. 2012, 41, 13908− 13918. (16) Ogunlaja, A. S.; Khene, S.; Antunes, E.; Nyokong, T.; Torto, N.; Tshentu, Z. R. Appl. Catal., A 2013, 462−463, 157−167. (17) Hartmann, M.; Jung, D. J. Mater. Chem. 2010, 20, 844−857. (18) Rosevear, A. J. Chem. Technol. Biotechnol. 1984, 34, 127−150. (19) Malcata, F. X.; Reyes, H.; Garcia, H.; Hill, C.; Amundson, C. J. Am. Oil Chem. Soc. 1990, 67, 890−910. (20) Bornscheuer, U. T. Chem. Int. Ed. 2003, 42, 3336−3337. (21) Rao, N. N.; Lütz, S.; Würges, K.; Minör, D. Org. Process Res. Dev. 2009, 13, 607−616. (22) Walmsley, R. S.; Ogunlaja, A. S.; Coombes, M. J.; Chidawanyika, W.; Litwinski, C.; Torto, N.; Nyokong, T.; Tshentu, Z. R. J. Mater. Chem. 2012, 22, 5792−5800. 7723

dx.doi.org/10.1021/ef401708f | Energy Fuels 2013, 27, 7714−7723