React Kinet Catal Lett (2009) 97:131–139 DOI 10.1007/s11144-009-0008-2 Comparing the hydrodesulfurization reaction of thiophene on c-Al2O3 supported CoMo, NiMo and NiW sulfide catalysts Sergio L. González-Cortés Received: 12 January 2009 / Accepted: 23 February 2009 / Published online: 17 June 2009 Ó Akadémiai Kiadó, Budapest, Hungary 2009 Abstract The thiophene hydrodesulfurization (HDS) reaction on c-Al2O3 supported CoMo, NiMo and NiW sulfide catalysts was compared in order to gain insight into the promoter effect on direct desulfurization (DDS) and hydrogenation (HYD) pathways. Ni-promoted Mo (or W) sulfide catalysts favor the hydrogen transfer reactions relative to CoMo sulfide catalyst, which facilitates the direct route instead. This different performance and also the dependence of the apparent Arrhenius parameters with the catalyst formulation were attributed to the major participation of Mo (or W) edge for the Ni-containing catalysts and S edge for CoMo sulfide catalyst upon the thiophene-HDS reaction. Keywords Co (or Ni)-promoter effect
Hydrodesulfurization reaction
Thiophene
Combustion method Introduction Deep hydrodesulfurization (HDS) of transportation fuel is currently becoming more demanding due to the lower limit of sulfur content in the final products by mandatory legislations. And also the higher sulfur contents in the reserves of crude oils besides the increase in the demand of crude oils [1, 2]. Indeed, over the last few years, global ultra low sulfur diesel and gasoline regulations have induced a marked improvement in fuel transportation quality and it is expected that this trend continues until reaching sulfur level in the order of wppb (part per billion in weight). S. L. González-Cortés Laboratorio de Cinética y Catálisis, Departamento de Quı́mica, Facultad de Ciencias, Universidad de Los Andes, Mérida 5101, Venezuela S. L. González-Cortés (&) Oxford Catalysts Company, 115e Milton Park, Oxford OX14 4RZ, UK e-mail: goncor@ula.ve; sergio.gonzalez@oxfordcatalysts.com 123 132 S. L. González-Cortés Such near zero sulfur emission legislation requires a high efficiency of the hydrodesulfurization technology particularly for upgrading of heavy oils and residua [3]. The addition of promoters has been one of the main routes to improve the hydrotreating catalyst performance; considerable efforts have been devoted to understand the influence of different parameters on the origin of the synergistic effect [4]. Currently, many efforts are devoted to find out the detailed structure and morphology of the Co-Mo-S and Ni-Mo-S nanoclusters in order to enhance the catalytic activity of conventional HDS catalysts [5]. Marked differences in the morphology and atomic-scale structure of Ni-Mo-S compared with Co-Mo-S have been observed [6]. Such atomic-scale study of the addition of either Co-Ni on MoS2 or Ni on WS2 has not been addressed yet. Nevertheless, the current insights can provide a better understanding of the differences in activity and selectivity in the two promoted systems, particularly for the direct desulfurization (DDS) and hydrogenation (HYD) pathways. Despite the advances in the determination of the structure and morphology of NiMo and CoMo sulfide nanoclusters, a comparative study of the HDS catalytic routes (DDS against HYD) on c-Al2O3 supported CoMo, NiMo and NiW sulfide catalyst performance has not been addressed. In the present study, both the influence of the reaction temperature and the catalyst formulation on the DDS and HYD are investigated in order to gain insight into the promoter effect on the HDS catalytic pathways. Experimental Catalyst oxide precursors were prepared by the urea-matrix combustion (UM 9 C) synthesis with aqueous solutions containing the corresponding metal salts and urea [7]. Bimetallic CoMo, NiMo and NiW catalyst precursors containing 12 wt% MoO3 (referred to 12Mo) or 19 wt% WO3 (19 W) and variable promoter compositions supported on c-Al2O3 (164 m2/g) were prepared. It was mixed urea with ammonium heptamolybdate-4-hydrate or ammonium meta-tungstate and Co (or Ni)(II) nitrate employing a urea/metals ratio of 10. The resulting paste was ignited at 500 °C (furnace temperature) in static air for 10 min to produce the oxide catalyst precursor. The c-Al2O3-supported CoMo, NiMo and NiW oxide catalyst precursors were sulfided at 0.1 MPa in a fixed-bed tubular quartz reactor. A gas space velocity of 100 mL (cat g min)-1 (particle size below 250 lm) and thiophene vapor (ca. 11 mol% C4H4S/H2) were employed up to 450 °C for 10 h. The catalysts with optimal compositions were also sulfided employing a gas mixture of 10 mol% H2S/ H2 at 400 °C for 4 h. The sulfide catalyst was tested at atmospheric pressure (0.1 MPa). The reactants consisted of a 20 mL min-1 H2 flow saturated with thiophene vapor at 0 °C, resulting a mixture of ca. 3 mol% C4H4S/H2. The reaction products were analyzed by gas chromatography (GC) using a flame ionization detector (FID). 123 Comparing the hydrodesulfurization reaction of thiophene 133 The Raman spectra of the sulfide catalysts were recorded on Jobin Yvon Labram spectrometer with a 632 nm HeNe laser, run in a back-scattered co-focal arrangement. The samples were pressed in a microscope slide; with a 45 s scanning time and 2 cm-1 resolution. Several points of each catalyst surface were probed to explore homogeneity of the sample and reproducibility of the data. High resolution transmission electron microscopy (HRTEM) was carried out using a JEOL 4000FX electron microscope with a 400 kV accelerating voltage. CoMo sulfide catalyst was ground into a fine powder and dispersed in AR-grade chloroform. Then, it was placed in an ultrasonic bath for ca. 15 min, before a drop of the suspension was put on a lacey carbon-coated copper grid (Agar, 20 mesh) and subsequently analyzed. Results and discussion Raman band intensity / arb. units The Raman spectra for Mo-containing sulfide catalysts are quite similar to that of -1 2H-MoS2, Fig. 1a, b [8]. Two major peaks at approximately  381 and  405 cm are 1 observed, which are assigned to the Mo-S stretching mode E2g along the basal plane and the S-Mo-S bond mode along the c-axis or perpendicular to the basal plane (A1g), respectively [9]. No Raman peaks characteristic of Mo-oxo species were detected, suggesting total sulfidation of the oxide catalyst precursors. Additionally, a peak at 1,595 and a broad band at 1,355 cm-1 characteristic of carbon material were also observed [7], owing to the use of thiophene as sulfiding agent. The Raman spectrum for alumina-supported NiW sulfide catalyst showed weak Raman peak intensity and no features related with W-oxo species (Fig. 1c). Two -1 major peaks at 361 and 421 cm-1 and a weak broad band at 175  cm  are observed. 1 The first feature is assigned to the W-S stretching mode E2g along the basal 405 381 (a) (b) 361 175 421 (c) 200 300 400 500 600 700 -1 Raman shift / cm Fig. 1 Raman spectra of the c-alumina-supported 2.5 Co-12 Mo (a), 3 Ni-12 Mo (b) and 3 Ni-19 W (c) sulfide catalysts under optimal composition 123 134 S. L. González-Cortés plane, the second one corresponds to the S-W-S bond mode along the c-axis or perpendicular to the basal plane (A1g) and the latter is attributed to second order Raman peaks [10]. The close correspondence between the Raman shift of the observed peaks and the data reported in the literature for single phase WS2 [11] suggests that not strongly distorted tungsten sulfide structure was produced. Since thiophene was employed as sulfiding agent, residual graphitic carbon was also observed. However, the carbon content (0.2–0.5 wt%) was significantly lower than that presents in NiMo catalysts (1.0–2.1 wt%). HRTEM analysis of c-alumina-supported CoMo sulfide catalyst is shown in Fig. 2. Crystalline planes characteristics of MoS2 with single and multi layer spacing of ca. 6 Å are clearly illustrated, in line with the Raman spectra. The slabs are partially intercalated by another slab and bent on a longer scale to maximize the interaction with the surface. The degree of stacking and the slab length distribution are fairly large, owing to the relatively high sulfiding temperature (i.e. 450 °C). Indeed, Co-promoted MoS2/c-Al2O3 HDS catalysts prepared by impregnation and also sulfiding with thiophene show quite similar HRTEM images [7]. The above mentioned results indicate that crystalline MoS2 and WS2 are formed upon sulfidation process and also residual carbon, owing to the relatively high sulfidation temperature and the use of thiophene as sulfiding agent, respectively. The influence of the promoter loading on the HDS rate constant for c-Al2O3supported 12 Mo and 19 W catalysts is shown in Fig. 3. A strong dependence of the first-order rate constant with the promoter loading is clearly displayed. Upon increasing the promoter content, the activity significantly rises, reaching a maximum catalytic performance for promoter/Mo (or W) molar ratio between 0.41 (i.e. Fig. 2 HRTEM image of the c-Al2O3-supported 2.5 Co-12 Mo sulfide catalyst 123 Fig. 3 Influence of the Ni and Co promoter loadings on the HDS rate constant of c-Al2O3supported 12 Mo and 19 W sulfide catalysts. Tests carried out at 350 °C, 0.1 MPa and 100 mL (cat g min)-1 HDS rate constant / µmol(g.s) -1 Comparing the hydrodesulfurization reaction of thiophene 135 160 NiMo 140 NiW 120 CoMo 100 80 60 40 0.0 0.5 1.0 1.5 2.0 2.5 Promoter/Mo(or W) molar ratio 2.5 wt% CoO-12 wt% MoO3) and 0.49 (i.e. 3 wt% NiO-12 wt% MoO3 or -19 wt% WO3). At higher promoter composition, the HDS rate constant markedly decreased, likely due to the segregation of cobalt (or nickel) sulfide. The sulfide catalysts showed quite similar maximum HDS kinetic constant, however at promoter/Mo (or W) molar ratio below optimal composition (i.e. 0.4–0.5) the NiW sulfide was the most active catalyst whereas at molar ratio above 0.5, the NiMo catalyst resulted the most active series. The promoter effect curve suggests that the synergistic effect arises from the formation of a more active phase or a more effective hydrogen spillover effect than that presents in the unpromoted catalyst. Indeed, the rise in the catalytic activity is most likely due to the increase in the number of Ni- (or Co-) MoS-type active sites formed on the MoS2 (or WS2) edges [4]. Figure 4 shows the influence of the reaction temperature on the catalytic behavior of c-Al2O3-supported CoMo, NiMo and NiW sulfide catalysts with optimal composition. The supported catalyst precursors were sulfided by using 10% H2S-H2 as sulfiding agent rather than thiophene. The thiophene conversion rises with increasing reaction temperature. The CoMo catalyst displays higher HDS activity than the Ni-promoted metal sulfides (i.e. NiMo; NiW) at temperature above 275 °C. Slightly larger thiophene conversions for NiW catalyst compared with NiMo catalyst at temperature higher than 300 °C is also noticed (Fig. 4a). In Fig. 4b, the C4/C4= mole ratio for Ni-containing catalysts decreased with increasing reaction temperature. On the other hand, the C4/C4= ratio did not vary for CoMo sulfide catalyst at temperatures below 300 °C. At higher temperatures, this ratio followed a behavior comparable to Ni-containing catalysts. Under similar thiophene conversion at low reaction temperature, NiMo and NiW catalysts displayed higher C4/C4= ratios than CoMo sulfide catalyst, suggesting that the former catalysts facilitate the hydrogen transfer reactions relative to the latter one. Density functional theory (DFT) calculations carried out by Topsøe and coworkers have given insight into the HDS of thiophene, indicating that the hydrogenation reactions take place on the Mo edge of MoS2, whereas the C-S bond scission only can occur at the S edges [12]. According to Topsøe and co-workers, this different performance between CoMo and Ni-containing catalysts can be 123 136 S. L. González-Cortés C4H4S Conversion / % (a) 100 NiMo 80 NiW CoMo 60 40 20 0 200 250 300 350 400 Temperature / °C C 4 /C 4 = (b) 7 6 NiMo 5 NiW 4 CoMo 3 2 1 0 200 250 300 350 400 Temperature / °C Fig. 4 Dependence of the HDS performance of c-alumina-supported 2.5 Co-12 Mo, 3 Ni-12 Mo and 3 Ni-19 W sulfide catalysts against the reaction temperature at 0.1 MPa and 100 mL (cat g min)-1 attributed to the formation of Co-Mo-S active site on the S-edge of MoS2 [6], which favors the direct desulfurization (DDS) of thiophene because of the participation of sulfur vacancy. On the other hand, the NiMo catalyst favors the formation of NiMo-S site mainly on the Mo edge under the sulfidation condition employed in this work [6, 13], facilitating therefore the HYD route rather than the DDS pathway. Since NiW sulfide catalyst showed a similar trend to transfer hydrogen as NiMo catalyst, it is suggested that Ni-W-S phase is also formed on the W edge of WS2 in line with the DFT calculation of Sun et al. [14]. In order to find out the thermodynamic feasibility of three hydrogenation reactions probably involved in the reaction of thiophene hydrodesulfurization, thermodynamic calculations were carried out using HSC Chemistry 5.11 software. Figure 5a shows the influence of the reaction temperature on the equilibrium molar fractions of thiophene (T) to tetrahydrothiophene (THT) at 1 bar: C4 H4 SðgÞ þ 2H2 ðgÞ ! C4 H8 SðgÞ ð1Þ The hydrogenation of T to THT is thermodynamically favorable at reaction temperatures below 200 °C and nearly negligible at temperatures above 400 °C. 123 Comparing the hydrodesulfurization reaction of thiophene (a) Equilibrium molar composition / % Fig. 5 a Thermodynamic molar fractions at equilibrium for the hydrogenation of thiophene to tetrahydrothiophene at atmospheric pressure (1 bar) against the reaction temperature. b Variation of the equilibrium constant for the hydrogenation of 1,3-butadiene to 1-butene and 1-butene to n-butane at 1 bar against the reaction temperature 137 100 C 4H 8S C 4H 4S 80 60 40 20 0 0 100 200 300 400 500 Temperature /°C (b) 25 Log (K) 20 15 C4H6 + H2 ↔ C4H8 10 5 C4H8 + H2 ↔ C4H10 0 0 100 200 300 400 500 Temperature /°C Among these temperatures, the thermodynamic equilibrium is predominant. Nevertheless, tetrahydrothiophene was not detected in the gas phase during the catalytic reactions since THT desulfurize more easily than thiophene [15]. The dependence of the equilibrium constant (K) of the hydrogenation of 1,3-butadiene to 1-butene and 1-butene to n-butane at 1 bar is given in Fig. 5b. Clearly, the hydrogenation of butadiene is favored relative to 1-butene. The equilibrium constants decrease with increasing reaction temperature, owing to the exothermicity of the hydrogenation reactions. Similar behavior is shown for the C4/C4= mole ratio for the Ni-containing catalysts, suggesting that the hydrogenation reaction is the main pathway in the thiophene-HDS reaction. The unsaturated C4 molecules (C4=) can be mainly obtained through the hydrogenolysis of either thiophene or partially hydrogenated thiophene molecules (DDS route): C4 H4 SðgÞ þ 2H2 ðgÞ ! C4 H6 ðgÞ þ H2 SðgÞ ð2Þ On the other hand, butane can be obtained by hydrogenation of unsaturated C4 molecules (Eq. 3). Indeed, the hydrogenation of olefins over HDS catalysts is fast despite their low equilibrium adsorption constant [16]. Butane might be also produced through the hydrogenolysis of tetrahydrothiophene (HYD pathway): 123 138 S. L. González-Cortés C4 H6 ðgÞ þ 2H2 ðgÞ ! C4 H10 ðgÞ ð3Þ C4 H8 SðgÞ þ 2H2 ðgÞ ! C4 H10 ðgÞ þ H2 SðgÞ ð4Þ Therefore, the direct desulfurization route would favor the formation of unsaturated C4 molecules (C4=) as the hydrogenation pathway facilitates the production of butane (C4). For CoMo sulfide catalysts, on the other hand, the DDS route significantly contributes over HDS of thiophene at temperatures below 300 °C. Higher temperatures seem also to facilitate the hydrogen transfer reactions through the hydrogenation of unsaturated C4 molecules (secondary hydrogenation reactions) considering the dependence of the C4/C4= mole ratio in function of the temperature (Fig. 4b). The apparent activation energies (Eaapp) and the pre-exponential factors (A) derived from Arrhenius plot for the optimal bimetallic formulations are given in a compensation plot (Fig. 6) and compared with carbon-supported transition metal sulfide catalysts [17]. Both data set show a lineal relation between Eaapp and Ln A (viz. Ln A = mEaapp ? c) with high correlation coefficient and fairly similar slope. Since the Arrhenius parameters fall on two roughly parallel lines, a different level of activity is expected, assuming that the variation in the Arrhenius parameters is due to changes in the surface coverage [18]. These discrepancies in the catalyst performance can be attributed to either different support or reaction/catalyst treatment conditions. At high apparent activation energy and pre-exponential factor, strong interaction between thiophene and the metal sulfide (i.e. S edge for MoS2) and hence high surface coverage are envisaged. The direct desulfurization pathway predominates in the thiophene-HDS reaction. On the other hand, low Arrhenius parameters indicate that weak interaction between thiophene and the metal sulfide (i.e. Mo edge for MoS2) occurred and therefore a low surface coverage is expected. Furthermore, hydrogen transfer reactions become more important upon catalytic process. In conclusion, crystalline MoS2 and WS2 catalysts and also residual graphitic carbon are formed upon sulfidation process of HDS, owing to the relatively high sulfidation temperature and the use of thiophene as sulfiding agent, respectively. Considering the C4/C4= mole ratio and the thermodynamic calculation, Ni-promoted 25 LnA = 0.2231Ea + 0.2389 CoMo R2 = 0.9961 20 (a) NiW 15 Rh CoMo 15 (b) NiMo 10 10 Ru Mo 5 Pd LnA = 0.2688Ea - 11.497 R2 = 0.9912 5 0 0 45 65 85 105 Ea / kJ(mol)-1 123 20 125 Ln A 25 Ln A Fig. 6 Compensation plot for the hydrodesulfurization of thiophene at 0.1 MPa. a Alumina-supported bimetallic metal sulfides, catalysts sulfided at 400 °C using 10% H2S-H2 as sulfiding agent. b Carbonsupported transition metal sulfide catalysts [17] Comparing the hydrodesulfurization reaction of thiophene 139 Mo (or W) sulfide catalysts favor the hydrogen transfer reactions relative to CoMo sulfide catalyst, which facilitates the C-S bond scission (direct desulfurization route). This different performance and also the dependence of the apparent Arrhenius parameters on the catalyst formulation is attributed to the major participation of Mo (or W) edge for the Ni-containing catalysts and S edge for CoMo sulfide catalyst on thiophene-HDS reaction. References 1. Babich, I.V., Moulijn, J.A.: Science and technology of novel processes for deep desulfurization of oil refinery streams: a review. Fuel 82(6), 607–631 (2003) 2. Song, C.: An overview of new approaches to deep desulfurization for ultra-clean gasoline, diesel fuel and jet fuel. Catal. Today 86(1–4), 211–263 (2003) 3. Rana, M.S., Sámano, V., Ancheyta, J., Diaz, J.A.I.: A review of recent advances on process technologies for upgrading of heavy oils and residua. Fuel 86(9), 1216–1231 (2007) 4. Topsøe, H., Clausen, B.S., Massoth, F.E.: In: Anderson, J.R., Boudart, M. (eds.) Hydrotreating Catalysis, Science and Technology. Springer-Verlag, Berlin (1996) 5. Topsøe, H.: The role of Co–Mo–S type structures in hydrotreating catalysts. Appl. Catal. A 322, 3–8 (2007) 6. Lauritsen, J.V., Kibsgaard, J., Olesen, G.H., Moses, P.G., Hinnemann, B., Helveg, S., Nørskov, J.K., Clausen, B.S., Topsøe, H., Lægsgaard, E., Besenbacher, F.: Location and coordination of promoter atoms in Co- and Ni-promoted MoS2-based hydrotreating catalysts J. Catal. 249(2), 220–233 (2007) 7. González-Cortés, S.L., Xiao, T-C., Costa, P.M.F.J., Fontal, B., Green, M.L.H.: Urea–organic matrix method: an alternative approach to prepare Co—MoS2/c-Al2O3 HDS catalyst. Appl. Catal. A Gen. 270(1–2), 209–222 (2004) 8. Chen, J.M., Wang, C.S.: Second order Raman spectrum of MoS2. Solid State Commun. 14(9), 857– 860 (1974) 9. Wieting, T.J., Verble, J.L.: Infrared and Raman studies of long-wavelength optical phonons in hexagonal MoS2. Phys. Rev. B 3, 4285–4286 (1971) 10. González-Cortés, S.L., Rodulfo-Baechler, S.M.A., Xiao, T., Green, M.L.H.: Rationalizing the catalytic performance of c-alumina-supported Co(Ni)–Mo(W) HDS catalysts prepared by urea-matrix combustion synthesis. Catal. Lett. 111(1–2), 57–66 (2006) 11. Chung, J.-W., Dai, Z.R., Ohuchi, F.S.: WS2 thin films by metal organic chemical vapor deposition. J. Cryst. Growth 186(1–2), 137–150 (1998) 12. Moses, P.G., Hinnemann, B., Topsøe, H., Nørskov, J.K.: Corrigendum to ‘‘The hydrogenation and direct desulfurization reaction pathway in thiophene hydrodesulfurization over MoS2 catalysts at realistic conditions: A density functional study’’ [J. Catal. 248 (2007) 188]. J. Catal. 260(1), 202–203 (2008) 13. Sun, M., Nelson, A.E., Adjaye, J.: On the incorporation of nickel and cobalt into MoS2-edge structures. J. Catal. 226(1), 32–40 (2004) 14. Sun, M., Nelson, A.E., Adjaye, J.: A DFT study of WS2, NiWS, and CoWS hydrotreating catalysts: energetics and surface structures. J. Catal. 226(1), 41–53 (2004) 15. Pazos, J.M., Andréu, P.: Hydrodesulphurization mechanism of thiophene and tetrahydrothiophene on a cobalt molybdenum catalyst. Can. J. Chem. 58, 479–484 (1980) 16. Prins, R.: Energy-related catalysis. In: Ertl, G., Knözinger, H., Schüth, F., Weitkamp, J. (eds.) Handbook of Heterogeneous Catalysis, vol. 6, 2nd edn, pp. 2695–2718. Wiley-VCH, Weinheim (2008) 17. Hensen, E.J.M., Brans, H.J.A., Lardinois, G.M.H.J., de Beer, V.H.J., van Veen, J.A.R., van Santen, R.A.: Periodic trends in hydrotreating catalysis: thiophene hydrodesulfurization over carbon-supported 4d transition metal sulfides. J. Catal. 192(1), 98–107 (2000) 18. Bond, G.C., Keane, M.A., Kral, H., Lercher, J.A.: Compensation phenomena in heterogeneous catalysis: general principles and a possible explanation. Catal. Rev. Sci. Eng. 42(3), 323–383 (2000) 123