CATALYST FOR THE SYNTHESIS OF ALKYL MERCAPTAN AND PROCESS FOR ITS PREPARATION

Abstract

A catalyst may include a support and from 5 to 20 wt.-% of a promoter, based on the total weight of the catalyst, wherein the support may include titanium dioxide, zirconium dioxide, and/or a mixture thereof, and the promoter may be an alkali metal oxide. Processes for preparing such catalysts may include impregnating a support of titanium dioxide and/or zirconium dioxide with an aqueous solution including a preferably soluble alkali compound and calcining. Alkyl mercaptans may be prepared in the presence of such catalysts or catalysts obtained by such processes.

Claims

1. A catalyst, comprising: a support; and 5 to 20 wt.-% of a promoter, based on total catalyst weight, wherein the support comprises titanium dioxide and/or zirconium dioxide, and wherein the promoter is an alkali metal oxide.

2. The catalyst of claim 1, wherein at least a part of the support has a tetragonal phase.

3. The catalyst of claim 1, wherein the alkali metal is sodium, potassium, cesium, or rubidium.

4. The catalyst of claim 1, wherein the promoter is cesium oxide.

5. The catalyst of claim 1, wherein the catalyst is a full catalyst.

6. The catalyst of claim 1, wherein the catalyst is a core-shell catalyst.

7. A process for preparing a supported catalyst of claim 1, the process comprising: (a) impregnating a support comprising titanium dioxide and/or zirconium dioxide with an aqueous solution comprising a soluble alkali compound, to obtain an impregnated support; (b) drying the impregnated support, to obtain a dried impregnated support; and (c) calcining the dried impregnated support to provide the catalyst.

8. The process of claim 7, wherein the impregnating (a), the drying (b), and the calcining (c) are repeated at least once.

9. The process of claim 7, further comprising: (d1) shaping the catalyst obtained from the calcining (c) to give a full catalyst.

10. The process of claim 7, further comprising: d2) applying the catalyst obtained from the calcining (c) to a core to provide a core-shell catalyst.

11. A process for preparing an alkyl mercaptan, the process comprising: reacting an alkyl alcohol with hydrogen sulfide in the presence of the catalyst of claim 1.

12. The process of claim 11, wherein the alkyl alcohol to be reacted is methanol and the alkyl mercaptan to be prepared is methyl mercaptan.

13. The catalyst of claim 1, consisting of the support and the promoter.

14. The catalyst of claim 1, wherein the alkali metal comprises sodium.

15. The catalyst of claim 1, wherein the alkali metal comprises potassium.

16. The catalyst of claim 1, wherein the alkali metal comprises cesium.

17. The catalyst of claim 1, wherein the alkali metal comprises rubidium.

18. The catalyst of claim 1, wherein the support consists of titanium dioxide and/or zirconium dioxide.

Description

DESCRIPTION OF THE FIGURES

[0068] FIG. 1 shows the XRD patterns of pure metal oxides (a) and metal oxides loaded with 5 wt.-% Cs (b), 10 wt.-% Cs (c), 15 wt.-% Cs (d) and 20 wt.-% Cs (e), with y indicating characteristic signals of pure γ-Al.sub.2O.sub.3, t indicating tetragonal ZrO.sub.2, A indicating anatase (TiO.sub.2) and Cs indicating Cs.sub.2CO.sub.3.

[0069] FIG. 2 shows the subtracted IR spectra of absorbed pyridine on the pure metal oxides at 50° C., with the solid line representing the IR spectrum taken a pyridine partial pressure of 0.1 mbar and the broken line representing the IR spectrum taken after evacuation at 10.sup.−7 mbar.

[0070] FIG. 3 shows the difference spectra of the OH vibration region of γ-Al.sub.2O.sub.3, ZrO.sub.2 and TiO.sub.2, with the solid line representing the IR spectrum taken a pyridine partial pressure of 0.1 mbar and the broken line representing the IR spectrum taken after evacuation at 10.sup.−7 mbar.

[0071] FIG. 4 shows the subtracted IR spectra of metal oxides with Cs loading of 10 or 20 wt.-% at 50° C. with the solid line representing the IR spectrum taken a pyridine partial pressure of 0.1 mbar and the broken line representing the IR spectrum taken after evacuation at 10.sup.−7 mbar.

[0072] FIG. 5 shows the IR spectra of CO adsorption on γ-Al.sub.2O.sub.3 (left), ZrO.sub.2 (middle), and TiO.sub.2 (right) at a CO partial pressure of 5 mbar and −150° C., with (a) the pure metal oxides, (b) Cs loading of 10 wt.-%, and (c) Cs loading of 20 wt.-%.

[0073] FIG. 6 shows the IR spectra of methanol adsorbed on Al.sub.2O.sub.3 (left), ZrO.sub.2 (middle) and TiO.sub.2 (left) at a methanol partial pressure of 0.1 mbar and 50° C., with (a) the pure metal oxides, (b) Cs loading of 10 wt.-%, and (c) Cs loading of 20 wt.-%.

[0074] FIG. 7 shows the IR spectra of methanol adsorption on the pure metal oxides γ-Al.sub.2O.sub.3 (left), ZrO.sub.2 (middle) and TiO.sub.2 (right) at methanol partial pressures and temperature of (a) 50° C. and 0.1 mbar, (b) 50° C. and 1 mbar, (c) 100° C. and 1 mbar, (d) 150° C. and 1 mbar, (e) 200° C. and 1 bar, (f 250° C. and 1 mbar, and (g) 300° C. and 1 mbar.

[0075] FIG. 8 shows the IR spectra of methanol over γ-Al.sub.2O.sub.3 with 10 and 20 wt.-% Cs (left), ZrO.sub.2 with 10 and 20 wt.-% Cs (middle), and TiO.sub.2 with 10 and 20 wt.-% Cs (right) at methanol partial pressure and temperatures of (a) 50° C. and 0.1 mbar, (b) 50° C. and 1 mbar, (c) 100° C. and 1 mbar, (d) 150° C. and 1 mbar, (e) 200° C. and 1 mbar, (f 250° C. and 1 mbar, and (g) 300° C. and 1 mbar.

[0076] FIG. 9 shows the initial rates for methyl mercaptan formation over γ-Al.sub.2O.sub.3 (left), ZrO.sub.2 (middle) and TiO.sub.2 (right) between 300 and 360° C. for pure metal oxides (solid line with cubes), Cs loading of 10 wt.-% (dotted line with circles), and Cs loading of 20 wt.-% (broken line with triangles) FIG. 10 shows the yields for methyl mercaptan (cubes), dimethyl ether (circles) and dimethyl sulfide (triangles) as a function of methanol conversion over the pure metal oxides γ-Al.sub.2O.sub.3 (left), ZrO.sub.2 (middle) and TiO.sub.2 (right) at 360° C.

[0077] FIG. 11 shows the yields for methyl mercaptan (cubes), dimethyl ether (circles) and dimethyl sulfide (triangles) as a function of methanol conversion over γ-Al.sub.2O.sub.3 with 10 and 20 wt.-% Cs (left), ZrO.sub.2 with 10 and 20 wt.-% Cs (middle), and TiO.sub.2 with 10 and 20 wt.-% Cs (right) at temperatures between 300 and 360° C.

[0078] FIG. 12 shows the dependency of methyl mercaptan formation rates over γ-Al.sub.2O.sub.3 in methanol (cubes; y=0.3x+1.7) and hydrogen sulfide (triangles; y=0.5x+1.9), ZrO.sub.2 in methanol (cubes; y=0.2x−0.3) and hydrogen sulfide (triangles; y=0.4x−0.6) and TiO.sub.2 in methanol (cubes; y=0.3x+1.2) and hydrogen sulfide (triangles; y=0.5x+1.7) with the concentration in mol/l.

[0079] FIG. 13 shows the dependency of methyl mercaptan formation rates over catalysts with different loadings of cesium: 10 wt.-% Cs (first row) in methanol indicated by cubes (γ-Al.sub.2O.sub.3: y=0.4x+1.0; ZrO.sub.2: y=0.5x+1.5; and TiO.sub.2: y=0.6x+2.3) and in hydrogen sulfide indicated by triangles (γ-Al.sub.2O.sub.3: y=0.4x+0.6; ZrO.sub.2: y=0.3x−0.01; and TiO.sub.2: y=0.2x+0.2). 20 wt.-% Cs (second row) in methanol indicated by cubes (γ-Al.sub.2O.sub.3: y=0.3x+1.0; ZrO.sub.2: 0.6x+2.3; and TiO.sub.2: y=0.5x+2.0) and in hydrogen sulfide indicated by triangles (γ-Al.sub.2O.sub.3: y=0.5x+1.3; ZrO.sub.2: y=0.3x+0.5; and TiO.sub.2: y=0.2x+0.3), concentrations in mol/l.

[0080] FIG. 14 shows the dependency of dimethyl ether formation rate over pure γ-Al.sub.2O.sub.3 in methanol (indicated by cubes; y=1.5x+11.0) and in hydrogen sulfide (indicated by triangles y=0.0x), over pure ZrO.sub.2 in methanol (indicated by cubes; y=0.7x) and over pure TiO.sub.2 (indicated by cubes; y=0.7x+2.3), concentration in mol/l.

[0081] FIG. 15 shows the initial rates for methyl mercaptan formation over γ-Al.sub.2O.sub.3 loaded with CsWS.sub.2 (tungsten content of 5.1 wt.-%, and cesium content of 20.6 wt.-%) at temperatures of 300, 320, 340 and 360° C. (solid squares).

EXAMPLES

1. Preparation of Cs Loaded Metal Oxides According to the Invention

[0082] Catalysts with a Cs loading of 5, 10, 15 and 20 wt.-%, based on the total weight of the catalysts, were prepared by incipient wetness impregnation of the commercially available metal oxides γ-Al.sub.2O.sub.3 (Spheralite 101, Axens), TiO.sub.2 (Hombikat 100 UV, Sachtleben), and ZrO.sub.2 (SZ 61152, Norpro), each having grain sizes of 0.125-0.25 mm, with an aqueous solution of cesium acetate, added dropwise to the agitated solid. For each Cs loading different impregnation solutions were prepared containing the required amount of cesium acetate to provide the desired Cs loading. 76 mg of cesium acetate (Sigma Aldrich, >99.99%) were dissolved in 0.5 mL H.sub.2O per 1 g of support to give a Cs loading of 5 wt.-%, respectively 160.5 of cesium acetate for a Cs loading of 10 wt.-%, 255 mg of cesium acetate for a Cs loading of 15 wt.-%, and 361.0 mg of cesium acetate for a Cs loading of 20 wt.-%. The impregnated metal oxides were dried over night at 70° C., followed by calcination in flowing synthetic air with a flow rate of 100 mL/min and at 400° C. for 2 h, achieved with a temperature ramp of 0.5° C./min. Prior to their use in the catalytic testing, all samples were activated by treatment in H.sub.2S with a flow rate of 20 ml/min at 360° C. for 2 hours.

2. Characterization of the Prepared Catalysts

[0083] 2.1 Elemental Composition and Surface Area Determination

[0084] The elemental composition of the prepared catalysts according to the invention was determined by atomic absorption spectroscopy (AAS). The measurements were performed on an UNICAM 939 AA-Spectrometer. To determine the textural properties, N.sub.2 physisorption was performed on a Porous Materials Inc. BET-121 sorptometer. After activation at 250° C. for 2 h under vacuum, N.sub.2 was adsorbed at a temperature of 77.4 K. The surface area was calculated using the BET-method. The results for elemental analysis and surface determination of all prepared catalysts are summarized in table 1 below.

TABLE-US-00001 TABLE 1 Results for elemental analysis and surface determination for all prepared catalysts Support Determined Cs loading [wt.-%] material parameter 0 5 10 15 20 Al.sub.2O.sub.3 c(Cs) [mmol g.sup.−1] 0 0.3 0.7 1.2 1.4 TiO.sub.2 0 0.4 0.7 1.1 1.5 ZrO.sub.2 0 0.4 0.8 1.1 1.5 Al.sub.2O.sub.3 S.sub.BET [m.sup.2 g.sup.−1] 283 260 239 171 154 TiO.sub.2 314 243 232 195 109 ZrO.sub.2 126 123 70 62 50

[0085] The results show that comparable loadings with Cs were achieved for each of three types of support materials. In general, the specific surface area of the prepared supported catalysts decreases with increasing Cs loading. This can be attributed to the increased density of the catalyst and coverage of the surface with Cs, both leading to a loss in surface area.

[0086] The table 2 below summarizes the different masses of cesium acetate in the different impregnation solutions used in the preparation of the catalysts, the masses of Cs.sup.+ in these impregnation solutions (the weight of the acetate counter-ion was neglected), the mass of the catalysts (support+Cs.sup.+), the theoretical concentration (C.sub.Th(M(cesium)/m(catalyst)) of Cs.sup.+ in the prepared catalysts, and the concentration (C.sub.EA(M(cesium)/m(catalyst)) of Cs.sup.+ in the prepared catalysts found with elemental analysis.

TABLE-US-00002 TABLE 2 Overview over the prepared catalyst and their concentrations of cesium. Catalyst/ m(CsAc) m(Cs.sup.+) m(cat) C.sub.Th(Cs/Cat) C.sub.EA(Cs/cat) loading [g] [g] [g] [mmol/g] [mmol/g] Cs(5 wt.-%)/ 0.076 0.052 1.052 0.38 0.4 ZrO.sub.2 Cs(10 wt.-%)/ 0.160 0.110 1.100 0.75 0.8 ZrO.sub.2 Cs(15 wt.-%)/ 0.255 0.175 1.175 1.13 1.1 ZrO.sub.2 Cs(20 wt.-%)/ 0.361 0.248 1.248 1.50 1.5 ZrO.sub.2 Cs(5 wt.-%)/ 0.076 0.052 1.052 0.38 0.3 TiO.sub.2 Cs(10 wt.-%)/ 0.160 0.110 1.100 0.75 0.7 TiO.sub.2 Cs(15 wt.-%)/ 0.255 0.175 1.175 1.13 1.2 TiO.sub.2 Cs(20 wt.-%)/ 0.361 0.248 1.248 1.50 1.5 TiO.sub.2 Cs(5 wt.-%)/ 0.076 0.052 1.052 0.38 0.4 Al.sub.2O.sub.3 Cs(10 wt.-%)/ 0.160 0.110 1.100 0.75 0.7 Al.sub.2O.sub.3 Cs(15 wt.-%)/ 0.255 0.175 1.175 1.13 1.1 Al.sub.2O.sub.3 Cs(20 wt.-%)/ 0.361 0.248 1.248 1.50 1.5 Al.sub.2O.sub.3

[0087] 2.2 Crystal Structure

[0088] The crystalline structure of all support and all catalysts was determined by powder X-ray diffraction. XRD patterns were collected with a Philips X'Pert System (Cu Kα radiation, 0.1542 nm) operating at 45 kV/40 mA, using a nickel Kβ-filter and solid-state detector (X'Celerator). The measurements were carried out with a step size of 0.0170 and scan time of 0.31 s per step.

[0089] The support materials gave the expected diffraction patterns, being phase-pure in γ-Al.sub.2O.sub.3, anatase in TiO.sub.2, and tetragonal zirconia in ZrO.sub.2. The XRD patterns are shown in FIG. 1. Upon Cs addition there was no change in the crystal structure of the support material, showing the same diffraction pattern. An additional diffraction peak was observed on γ-Al.sub.2O.sub.3 and TiO.sub.2 at 45° C., indicative of Cs.sub.2CO.sub.3. The carbonate is believed to be formed by the reaction of the surface Cs species with atmospheric CO.sub.2. Upon sulfidation, the carbonate and its peaks disappeared leading to the formation of sulfur oxyanions, that were not detected by means of X-ray diffraction. No other reflections appeared. It is therefore concluded that the active Cs species is XRD amorphous.

[0090] 2.3 Characterization of Acid Base Properties

[0091] Adsorption of CO and pyridine onto the pure metal oxides and the prepared catalysts was monitored via IR spectroscopy in transmission absorption mode (samples pressed into self-supporting wafers) to measure the Lewis acidity. Before the adsorption, the samples were heated to 360° C. with a hating ramp of 10° C. per minute under a helium flow of 10 mL per minute. Subsequently, the samples were sulfided for 0.5 h at 360° C. under a flow of 10 mL per minute of 10 vol.-% of hydrogen sulfide in nitrogen. To remove physisorbed hydrogen sulfide, the sample was flushed with a He flow of 10 mL per minute for another 15 min, before it was evacuated to 10.sup.−7 mbar and cooled down to 50° C. For pyridine adsorption, the cell was cooled down to 50° C. and the sample was exposed to pyridine at a partial pressure of 1 mbar of pyridine, followed by decreasing the pyridine partial pressure. Further evacuation to 10.sup.−5 mbar resulted in no pyridine adsorbed on Cs containing samples. Thus, spectra from different catalysts were compared at 0.1 mbar, before evacuation took place. The concentrations of coordinating pyridine were calculated using the molar integrated extinction coefficient of 0.96 cm per μmol determined for the characteristic band at 1450 cm.sup.−1. CO adsorption took place by cooling down the IR cell to −150° C., using liquid nitrogen. The spectra were recorded at a CO partial pressure of 5 mbar.

[0092] Methanol was adsorbed at 50° C., while stepwise increasing the methanol partial pressure (0.1 mbar, 0.5 mbar, 1 mbar and 5 mbar) followed by an increase in temperature to 300° C. All spectra were recorded with a Nicolet 6700 FTIR spectrometer (64 scans were collected to obtain each spectrum). All spectra were subjected to a background subtraction and normalized to the mass of the wafer.

[0093] 2.4 Pyridine Absorption

[0094] The acidity of the metal oxides was measured with pyridine adsorbed via IR, FIG. 2. On pure γ-Al.sub.2O.sub.3 eight bands were observed at 1621, 1612, 1591, 1577, 1450 and 1440 cm.sup.−1. The bands at 1621 and 1612 cm.sup.−1 are assigned to the 8a vibrational mode of pyridine coordinatively bound to Lewis acid sites (LAS) of different acid strength (the wavenumber increases with acid strength), while the band at 1579 cm.sup.−1 is assigned to the 8b vibrational mode. The band at 1591 cm.sup.−1 is assigned to the 8a vibrational mode of H-bond pyridine, caused by the interaction of pyridine with weak acidic surface hydroxyl groups. The signal at 1450 cm.sup.−1 are attributed to the 9b vibration of pyridine on LAS, while the band at 1440 cm.sup.−1 is assigned again to pyridine H-bonded on hydroxyl groups. The sites assigned to pyridine coordinatively bound to LAS (1450 cm.sup.−1, 1612-1620 cm.sup.−1) were stable against evacuation, while the H-bonded pyridine bands (1440 and 1593 cm.sup.−1) disappeared after evacuation due to their weak interaction with the probe molecule. This is in line with the release of OH groups, leading to a decrease in the negative OH band around region of 3700 cm.sup.−1, as H-bond pyridine desorbed (FIG. 3).

[0095] The adsorption of pyridine on ZrO.sub.2 and TiO.sub.2 via IR gave bands at 1604, 1593, 1573 and 1445 cm.sup.−1. The 1604 cm.sup.−1 is assigned to the 8a vibrational mode of pyridine bound to LAS of ZrO.sub.2 and TiO.sub.2, while the 1573 cm.sup.−1 is assigned to the 8a vibrational mode. The 1593 cm.sup.−1 is assigned to the 8a vibrational mode of H-bond pyridine, caused by the interaction of pyridine with weak acidic surface hydroxyl groups. As in the case for γ-alumina, this signal vanished after evacuation. The signal at 1445 cm.sup.−1 is assigned to the 9b vibration of pyridine on LAS. Integrating the band at 1450 cm.sup.−1, LAS concentration on the metal oxide was determined as 454 μmol g.sup.−1 on γ-Al.sub.2O.sub.3, 220 μmol g.sup.−1 on ZrO.sub.2 and 749 μmol g.sup.−1 on TiO.sub.2. The shift of the signals of pyridine to lower wavenumbers from γ-Al.sub.2O.sub.3 (1450 cm.sup.−1), to ZrO.sub.2 and TiO.sub.2 (1445 cm.sup.−1) shows a higher Lewis acid strength of the former than the latter.

TABLE-US-00003 TABLE 3 Summary of all discussed signals of the metal and their assignments. Metal Wavenumber Surface oxide [nm.sup.−1] Vibration species y-Al.sub.2O.sub.3 1620 8a.sub.LAS,strong Al.sup.IV 1612 8a.sub.LAS,weak Al.sup.IV-Al.sup.VI 1593 8a.sub.H x-OH 1573 8b All Al + x-OH 1450 9b.sub.LAS All Al 1440 9b.sub.H x-OH ZrO.sub.2 1604 8a.sub.LAS,weak Zr.sup.IV 1593 8a.sub.H x-OH 1573 8b Zr.sup.IV and x-OH 1445 9b Zr.sup.IV TiO.sub.2 1604 8a.sub.LAS,weak Ti.sup.IV 1591 8a.sub.H x-OH 1573 8b Ti.sup.IV and x-OH 1445 9b Ti.sup.IV

[0096] The addition of Cs on the metal oxides modified the IR spectra with adsorbed pyridine (FIG. 4). On medium doped γ-Al.sub.2O.sub.3, Cs(10)γ-Al.sub.2O.sub.3, the bands assigned to the 8a vibrational mode of pyridine coordinatively bound to strong (1612 cm.sup.−1) and weak (1609 cm.sup.−1) LAS were no longer detected, as well as the signal of H-bonded pyridine. A new band appeared at 1583 cm.sup.−1, corresponding to the 8a vibrational mode of pyridine coordinatively bound to a weak Lewis acidic alkali, i.e. Cs.sup.+ with lower Lewis acid strength than those measured on γ-Al.sub.2O.sub.3. Upon addition of Cs on ZrO.sub.2 and TiO.sub.2, Cs(10)/ZrO.sub.2 and Cs(10)/TiO.sub.2, the bands assigned to the LAS of the support were not observed. As in the case of Cs/γ-Al.sub.2O.sub.3, new bands at 1600 and 1583 cm.sup.−1 appeared, corresponding to the vibrational mode of the 1+6a and 8a overtone vibrations of pyridine on Cs, respectively. After evacuation, signals of pyridine on Cs sites disappeared on all three samples, remaining partially on Cs(10)/γ-Al.sub.2O.sub.3 and Cs(10)/TiO.sub.2.

[0097] Additional Cs, Cs(20)/γ-Al.sub.2O.sub.3, led to a decrease in the band at 1612 cm.sup.−1. A new signal at 1600 cm.sup.−1 was observed, attributed to the 1+6a overtone vibration of pyridine on Cs. The already mentioned 1583 cm.sup.−1 8a vibration of pyridine on Cs sites and 1573 cm.sup.−1 8b vibration of pyridine on LAS and Cs. Thus, the gradual addition of Cs on the surface of γ-Al.sub.2O.sub.3 led to the replacement of strong LAS from γ-Al.sub.2O.sub.3 with weaker LAS from Cs. Pyridine adsorption on Cs(20)/ZrO.sub.2 and Cs(20)/TiO.sub.2 catalysts resulted only on pyridine coordinatively bound to Cs.sup.+ sites (8a, 8b and 1+6a). All adsorbed pyridine species on Cs doped ZrO.sub.2 and TiO.sub.2 desorbed under vacuum; while a minor signal of LAS on Cs(20)/γ-Al.sub.2O.sub.3 remained.

TABLE-US-00004 TABLE 4 Assignments of pyridine absorptions on Cs loaded metal oxides. Wavenumber Surface Catalysts [cm.sup.−1] Vibration species Cs(10 wt.-%)/ 1612 8a.sub.LAS,strong Al.sup.IV-Al.sup.VI γ-Al.sub.2O.sub.3 1583 8a.sub.Cs Cs.sup.+ 1573 8b Al.sup.IV-Al.sup.VI + Cs.sup.+ 1450 9b.sub.LAS Al.sup.IV-Al.sup.VI 1440 9b.sub.Cs Cs.sup.+ Cs(20 wt.-%)/ 1612 8a.sub.LAS,strong Al.sup.IV-Al.sup.VI γ-Al.sub.2O.sub.3 1602 1 + 6a.sub.Cs Cs.sup.+ 1583 8a.sub.Cs Cs.sup.+ 1573 8b Al.sup.IV-Al.sup.VI + Cs.sup.+ 1450 9b.sub.LAS Al.sup.IV-Al.sup.VI 1440 9b.sub.Cs Cs.sup.+ Cs(10 wt.-%)/ZrO.sub.2 1602 1 + 6a.sub.Cs Cs.sup.+ 1583 8a.sub.Cs Cs.sup.+ 1573 8b.sub.Cs Cs.sup.+ 1440 9b.sub.Cs Cs.sup.+ Cs(20 wt.-%)/ZrO.sub.2 1602 1 + 6a.sub.Cs Cs.sup.+ 1583 8a.sub.Cs Cs.sup.+ 1573 8b.sub.Cs Cs.sup.+ 1440 9b.sub.Cs Cs.sup.+ Cs(10 wt.-%)/TiO.sub.2 1604 1 + 6a.sub.Cs Cs.sup.+ 1583 8a.sub.Cs Cs.sup.+ 1573 8b.sub.Cs Cs.sup.+ 1440 9b.sub.Cs Cs.sup.+ Cs(20 wt.-%)/TiO.sub.2 1602 1 + 6a.sub.Cs Cs.sup.+ 1583 8a.sub.Cs Cs.sup.+ 1573 8b.sub.Cs Cs.sup.+

[0098] The titration of the acid sites with pyridine indicates a high heterogeneity of LAS sites in γ-Al.sub.2O.sub.3, with two types of LAS, while TiO.sub.2 and ZrO.sub.2 only provide one type of LAS with similar strength on both materials, being coincident with those observed in literature. The effect of Cs deposition can be rationalized as following: At medium Cs loading, Cs.sup.+ modifies the surface sites of the metal oxide by direct interaction, increasing the surface basicity, due to the lower Sanderson electronegativity. This direct interaction is done by exchange of surface protons with Cs.sup.+ cations.

[0099] At high Cs loading, the surface is dominated by Cs. As postulated for potassium on TiO.sub.2, high loading of alkali leads to a complete coverage of the metal oxide surface, leading to surface properties similar to the bulk alkaline material.

[0100] 2.5 CO Absorption

[0101] The CO adsorption on the catalysts via IR is shown in FIG. 5. The assignments for the different CO bands are in Table 4. Similar IR bands were obtained in the adsorption of CO on the different metal oxides. The bands in between 2180-2190 cm.sup.−1 are assigned to CO adsorption on LAS, while those around 2150 cm.sup.−1 are assigned to surface hydroxyl. The CO stretching vibration in γ-Al.sub.2O.sub.3 (2188 cm.sup.−1) was at higher wavenumber than ZrO.sub.2 (2177 cm.sup.−1) and TiO.sub.2 (2181 cm.sup.−1), indicative of a higher perturbation of the CO bond. This trend is the same as the one observed with pyridine, implying a higher strength of the Lewis acid sites of γ-Al.sub.2O.sub.3.

[0102] The addition of 10 wt % Cs on the three supports led to a decrease of the CO stretching vibration on LAS to lower wavenumbers (2138-2136 cm.sup.−1), corresponding to CO adsorbed on Cs.sup.+ ions. In the case of Cs(10)/γ-Al.sub.2O.sub.3, an additional band appeared at 2179 cm.sup.−1, corresponding to LAS in the γ-Al.sub.2O.sub.3 support altered by the alkali cation. No bands were observed for CO adsorbed on OH groups. On the samples with high Cs loading of 20 wt % only the signals of CO on Cs cations and physisorbed CO were detected. CO did not adsorb on the Cs(20)/ZrO.sub.2 sample.

[0103] The red shift of the CO stretching vibration on LAS with Cs on the surface of γ-Al.sub.2O.sub.3 is due to an increase in the basicity, decreasing the Sanderson electronegativity. The results for TiO.sub.2 and ZrO.sub.2 are in line with the results of pyridine adsorption, with no LAS being accessible on those materials at 10 wt % Cs loading. As in the case of the pyridine adsorption, Cs.sup.+ is the only species available for CO adsorption on the heavy Cs doped materials. The adsorption of CO via IR is in line with the pyridine adsorption; Lewis acid sites from the support are not present with 10 wt % Cs loading, with the exception of Cs(10)/γ-Al.sub.2O.sub.3.

TABLE-US-00005 TABLE 5 Assignments of CO absorptions on pure and Cs loaded metal oxides. Metal oxides Cs (10 wt.-%) Cs (20 wt.-%) CO.sub.LAS CO.sub.OH CO.sub.Cs+ CO.sub.Phys CO.sub.LAS CO.sub.OH CO.sub.Cs+ CO.sub.Phys CO.sub.LAS CO.sub.OH CO.sub.Cs+ CO.sub.Phys γ- 2188 2150 — — 2179 — 2146 2136 — — 2146 2136 Al.sub.2O.sub.3 ZrO.sub.2 2177 2154 — — — — 2144 2136 — — — — TiO.sub.2 2181 2150 — — — — 2144 2138 — — 2144 2133

[0104] 2.6 Methanol Absorption

[0105] The IR spectra of methanol adsorbed on the metal oxides and its Cs doped are shown in FIG. 6, exhibiting bands in the 3000-2750 cm.sup.−1 region (alkyl (sp.sup.3) C—H vibrations). The region in between 3000-2900 cm.sup.−1 is assigned to the asymmetric stretch of (υ.sub.as(CH.sub.3)) or its Fermi resonance with CH.sub.3 deformation vibrations (2δ.sub.s(CH.sub.3)), while lower bands are assigned to symmetric stretching vibrations (υ.sub.s(CH.sub.3)). Different intensities were observed for the IR bands assigned to the adsorption of methanol on strong Lewis acid sites and strong Lewis basic sites, for both the υ.sub.s and the υ.sub.s at 50° C. (FIGS. 6 to 8). The former site led to the formation of a bridging methoxide, known as Species I, with IR bands at higher wavenumber for both the υ.sub.as (2943, 2948 and 2944 cm.sup.−1 for γ-Al.sub.2O.sub.3, ZrO.sub.2 and TiO.sub.2) and the υ.sub.s (2845, 2852 and 2844 cm.sup.−1 for γ-Al.sub.2O.sub.3, ZrO.sub.2 and TiO.sub.2). The latter site resulted in the formation of an alcoholate (dissociation of the O—H group), known as Species II, for both the υ.sub.as (2939, 2931 and 2923 cm.sup.−1 for γ-Al.sub.2O.sub.3, ZrO.sub.2 and TiO.sub.2) and the υ.sub.s (2821, 2827 and 2821 cm.sup.−1 for γ-Al.sub.2O.sub.3, ZrO.sub.2 and TiO.sub.2). On ZrO.sub.2 a relatively higher concentration of dissociated methanol was visible, which increased further for TiO.sub.2. In the case of γ-Al.sub.2O.sub.3, heating of IR cell led to an increase in the intensity of the bridging methoxides (Species I). No major changes were observed upon heating in the other two supports. The relative intensities of methanol on the surface species directly leads to the conclusion that the general acidic character of the metal oxide to a more basic one decreases in the order γ-Al.sub.2O.sub.3>ZrO.sub.2˜TiO.sub.2.

3. Catalytic Testing of the Supported Catalysts According to the Invention

[0106] The catalytic thiolation of methanol was performed in a reaction tube with a volume of 25 mL. Before the reaction, 125.0 mg of catalyst (125-250 μm), diluted in 1 g of SiC, were sulfided in a flow of 20 mL min.sup.−1 H.sub.2S at 360° C. and 9 bar. The volume of the catalyst was almost negligible compared to the empty volume of the plug flow reactor (20 mL). This led to a relatively low Liquid Hourly Space Velocity (LHSV) based on liquid methanol (CH.sub.3OH) of only 0.054 h.sup.−1. The Gas Hourly Space Velocity (GHSV) based on the complete feed (H.sub.2S, CH.sub.3OH and N.sub.2) was 150 h.sup.−1 (based on standard conditions at 0° C. and 1.013 bar in accordance with DIN 1343). To determine activation energies, the reaction was performed with a flow of gaseous CH.sub.3OH (10 mL min.sup.−1) mixed with H.sub.2S (20 mL min.sup.−1) and N.sub.2 (20 mL min.sup.−1) at a pressure of the feed stream of 9 bar with partial pressures of 3.6 bar for N.sub.2, 3.6 bar for H.sub.2S and 1.8 bar for methanol. The reaction tube was heated via a jacket by means of a heat transfer medium to temperatures between 300 and 360° C.

[0107] Standard calculations of the Weisz-Prater modulus showed that it was <1 for all catalysts under all conditions, and, therefore, it can be concluded that the kinetic results were unaffected by internal mass transfer effects. Online analysis of the product flow was done using a Shimadzu GC-2014 equipped with a HP plot Q column (2.7 m, 2.0 mm inner diameter), using a TCD detector. Reaction rate constants were calculated using the integrated rate law for a 0.5 order reaction in CH.sub.3OH and H.sub.2S for CH.sub.3SH. To study the product distribution over the whole range of conversion, the residence time was adjusted, keeping partial pressure of CH.sub.3OH at 2.2 bar, and N.sub.2 and H.sub.2S at 3.3 bar at 360° C.

[0108] Reaction orders were determined at 360° C. For reaction orders in H.sub.2S, the partial pressure of methanol was kept constant at 2.2 bar, while the H.sub.2S partial pressure was varied between 1.1 and 5.6 bar. To measure methanol reaction orders, the H.sub.2S partial pressure was set to 4.5 bar and the CH.sub.3OH partial pressure varied from 0.6 mbar to 2.2 gaseous CH.sub.3OH. The N.sub.2 gas flow was adjusted to compensate volume flow changes and keep the total volume flow constant at 80 ml/min. The amount of catalyst used in each experiment was adjusted accordingly, to ensure CH.sub.3OH conversion below 10%. Reaction orders for cesium-modified materials were measured with 10.0 mg catalysts, while 5.0 mg for TiO.sub.2 and ZrO.sub.2 and 1.0 mg of γ-Al.sub.2O.sub.3 was sufficient. In the case of γ-Al.sub.2O.sub.3, the catalyst was physically mixed with SiO.sub.2, being known to be inactive in the studied reaction, in a ratio of 1:9, to avoid channeling effects.

[0109] 3.1 Catalytic Activity

[0110] Initial rates for methyl mercaptan (CH.sub.3SH) formation are shown in FIG. 9. The highest rate in methanol thiolation was observed for TiO.sub.2 (0.17-1.4.Math.10.sup.−6 mol.sub.CH3SH s.sup.−1 g.sub.cat.sup.−1), followed by γ-Al.sub.2O.sub.3 (0.13-9.2.Math.10.sup.−6 mol.sub.CH3SH s.sup.−1 g.sub.cat.sup.−1) and ZrO.sub.2 (0.02-0.2.Math.10.sup.−6 mol.sub.CH3SH s.sup.−1 g.sub.cat.sup.−1). For the Cs doped systems the rates for CH.sub.3SH formation decreased in the order of Cs(10 wt.-%)/γ-Al.sub.2O.sub.3 (1.8-8.7.Math.10.sup.−6 mol.sub.CH3SH s.sup.−1 g.sub.cat.sup.−1)>Cs(10 wt.-%)/ZrO.sub.2 (1.7-7.1.Math.10.sup.−6 mol.sub.CH3SH s.sup.−1 g.sub.cat.sup.−1 g.sub.cat)>Cs(10 wt.-%)/TiO.sub.2 (1.8-6.6.Math.10.sup.−6 mol.sub.CH3SH s.sup.−1 g.sub.cat.sup.−1). Higher Cs loading of 20 wt % did not lead to more active catalysts, rather the activity was slightly lower for Cs(20 wt.-%)/γ-Al.sub.2O.sub.3 (2.0-7.6.Math.10.sup.−6 mol.sub.CH3SH s.sup.−1 g.sub.cat.sup.−1), Cs(20 wt.-%)/ZrO.sub.2 (1.7-7.1.Math.10.sup.−6 mol.sub.CH3SH s.sup.−1 g.sub.cat.sup.−1) and Cs(20 wt.-%)/TiO.sub.2 (1.8-5.8.Math.10.sup.−6 mol.sub.CH3SH s.sup.−1 g.sub.cat.sup.−1). While there is one magnitude difference in the rates of CH.sub.3SH formation with the different metal oxides, the activity of the Cs systems showed only minor differences. This indicates that the overall activity is determined by the surface Cs species, which appears to be similar on all three metal oxide supports. Indeed, the CH.sub.3SH formation rate decreased slightly for all three systems.

[0111] The yields of methyl mercaptan, dimethyl sulfide (DMS), and dimethyl ether (DME) were measured as a function of the methanol conversion at 360° C. for all three metal oxides (FIG. 10). On γ-Al.sub.2O.sub.3, CH.sub.3SH and (DME) were obtained as primary products, being DME the highest primary product until 60% of methanol conversion. Overcoming 60% conversion the yield of DME decreases to 20% at 90% conversion, being CH.sub.3SH the main product. This behavior is explained by the re-adsorption of DME on the catalyst, undergoing secondary reaction to form CH.sub.3SH. Similar results were observed with ZrO.sub.2, being CH.sub.3SH the main primary product and a dimethyl ether yield below 10%. Remarkably, no DME was formed at conversion lower 10% on ZrO.sub.2. On both γ-Al.sub.2O.sub.3 and ZrO.sub.2 dimethyl sulfide was found at higher conversion levels, being a secondary product of CH.sub.3SH formation. On TiO.sub.2 no dimethyl ether was observed, being dimethyl sulfide the only byproduct. Performing the reaction without H.sub.2S on ZrO.sub.2 and TiO.sub.2 resulted in the formation of DME (FIG. 10), hinting for a competition between the reactants on ZrO.sub.2 and TiO.sub.2. The yields to CH.sub.3SH increases in the order γ-Al.sub.2O.sub.3<ZrO.sub.2<TiO.sub.2. The yields of CH.sub.3SH, DMS, and DME were measured as a function of the CH.sub.3OH conversion at 360° C. for all three metal oxides with loading of 10 and 20 wt.-% of Cs (FIG. 11). A general trend was observed for all Cs containing systems: CH.sub.3SH was obtained as main product, while the only catalyst yielding DME was Cs(10 wt.-%)/Al.sub.2O.sub.3, with a DME yield of 0.3% at 360° C. Main side product was DMS, with a maximum yield of 0.7% on Cs(10 wt.-%)/Al.sub.2O.sub.3 at 360° C. The absence of DME with Cs present on the surface was attributed to the absence of strong LAS. These results are supported with the adsorption of pyridine and CO via IR, showing a drastic decrease of Lewis acidity by Cs doping.

[0112] 3.2 Kinetics

[0113] a) Formation of Methyl Mercaptan

[0114] The dependency of the methyl mercaptan formation rates in methanol and hydrogen sulfide over the pure metal oxides in FIG. 12 and over the metal oxides with a loading of 10 and 20 wt.-% Cs in FIG. 13. The reaction orders for the formation of methyl mercaptan, with respect to CH.sub.3OH and H.sub.2S are shown in Table 6.

TABLE-US-00006 TABLE 6 Determined reaction order for methyl mercaptan formation in H.sub.2S and CH.sub.3OH on all tested systems. Support Catalyst loading material Kinetics 0 10 20 Al.sub.2O.sub.3 Reaction 0.4 0.5 0.3 TiO.sub.2 order 0.3 0.6 0.6 ZrO.sub.2 (Methanol) 0.3 0.5 0.6 Al.sub.2O.sub.3 Reaction 0.4 0.4 0.6 TiO.sub.2 order 0.5 0.2 0.2 ZrO.sub.2 (H.sub.2S) 0.4 0.2 0.2

[0115] On all metal oxides, the reaction order of 0.5 in both H.sub.2S and CH.sub.3OH in the formation of CH.sub.3SH hints for the dissociation of both reactants, prior to the bimolecular Langmuir-Hinshelwood mechanism. The rate equation for methyl mercaptan is

[00001]$r_{C{H}_{3}SH}=\frac{k_5{K}_2{{K_3\left[{\mathrm{CH}}_3\mathrm{OH}\right]}^{0.5}\left[H_{2}S\right]}^{0.5}}{a^2}\mathrm{with}$$a=\left(1+{K_2^{0.5}\left[{\mathrm{CH}}_3\mathrm{OH}\right]}^{0.5}+{K_3^{0.5}\left[H_{2}S\right]}^{0.5}+{\left[{\mathrm{CH}}_3\mathrm{SH}\right]}^{0.5}/K_6^{0.5}+{\left[H_{2}O\right]}^{0.5}/K_7^{0.5}\right).$

[0116] The rate equation for dimethyl ether is

[00002]$r_{C{H}_{3}OC{H}_3}=\frac{k_4{K_1\left[{\mathrm{CH}}_3\mathrm{OH}\right]}^{1.5}}{b}\mathrm{with}$$b=\left(1+{K_1^{0.5}\left[{\mathrm{CH}}_3\mathrm{OH}\right]}^{0.5}+{\left[H_{2}O\right]}^{0.5}/K_8^{0.5}\right).$

[0117] Hydrogen sulfide is known to adsorb dissociatively on the surface of metal oxides, while methanol also adsorbs dissociatively forming a methanolate on the Lewis acid-base pairs of the surface oxides. Thus, it is believed that both substrates dissociate on the same kind of basic sites. One could speculate that a decrease in the reaction order with partial pressure would have the effect that the substrates were competing for adsorption on the surface. However, this is not observed in the case of metal oxides.

[0118] The apparent activation energy for methyl mercaptan formation was found to be around 112 kJ mol.sup.−1 on γ-Al.sub.2O.sub.3, 115 kJ mol.sup.−1 on ZrO.sub.2 and 107 kJ mol.sup.−1 on TiO.sub.2. This is the apparent activation energy of methyl mercaptan formed over the active sites of the metal oxides, as no Cs is present.

TABLE-US-00007 TABLE 7 Determined apparent activation energy for methyl mercaptan formation. Cs loading [wt.-%] 0 10 20 Al.sub.2O.sub.3 E.sub.a,app,CH3SH 112 78 65 TiO.sub.2 [kJ mol.sup.−1 ] 107 66 59 ZrO.sub.2 115 73 64

[0119] The addition of Cs (10 wt.-%) resulted in a reaction order close to 0.5 in both reactants, hinting to the same dissociative reaction mechanism as proposed for the pure metal oxides. However, the reaction order of 0.2 on H.sub.2S suggests that the Cs/TiO.sub.2 and Cs/ZrO.sub.2 catalysts are operating under partial coverage of H.sub.2S. The apparent activation energy decreased to values in between 66 to 78 kJ mol.sup.−1. The lower activation energy of these catalysts, with respect to the metal oxides, is associated to an increase in basicity. The presence of the Cs.sup.+ cation on the support hints for the coverage of its surface hydroxyls, similarly to that observed with sodium and potassium on alumina. This was confirmed with the absence of the OH bands during adsorption of pyridine and CO via IR. On all heavy Cs doped materials (20 wt.-%), similar reaction order values were obtained as for Cs 10 wt.-%, also hinting for a dissociative mechanism. The apparent activation barrier was in between 65-59 kJ mol.sup.−1 for the three heavy doped materials. The decrease of apparent activation energy can be explained by a complete modification of the surface. As shown by pyridine adsorption, the only surface species available during pyridine adsorption was Cs, suppressing the chemical properties of the metal oxides and acting as a very weak LAS. In addition, the absence of strong Lewis acid sites resulted only in the formation of surface methanolate, as observed during methanol adsorption via IR on the heavy Cs doped catalysts.

[0120] The reaction orders for DME formation were determined for all pure metal oxides in methanol and hydrogen sulfide (table 8 and FIG. 14). The reaction order for dimethyl ether formation was 1.5 in methanol and 0 in H.sub.2S on γ-Al.sub.2O. Zero order in H.sub.2S shows that H.sub.2S does not compete with methanol on DME formation sites. The reaction order of 1.5 in methanol is explained by the partially coverage of the catalyst surface by methanol. On γ-Al.sub.2O.sub.3, the adsorption of methanol seems to be favored compared to H.sub.2S, leading to DME formation and zero order in H.sub.2S. On ZrO.sub.2 and TiO.sub.2 reaction order for DME formation (without H.sub.2S present) was found to be 0.7. It is believed that on these materials surface coverage of methanol is higher compared to γ-Al.sub.2O. The lower apparent activation energy on γ-Al.sub.2O.sub.3 compared to the other two materials is attributed to the higher Lewis acid strength, as shown by CO and pyridine adsorption, facilitating the break of the CO bond of CH.sub.3OH.

TABLE-US-00008 TABLE 8 Reaction in methanol and hydrogen sulfide and apparent activation energy for dimethyl ether formation. Reaction order Reaction order E.sub.a,app,DME Support (Methanol) (H.sub.2S) [kJ mol.sup.−1] Al.sub.2O.sub.3 1.5 0 70 TiO.sub.2 0.7 n.d. 91 ZrO.sub.2 0.7 n.d. 93

[0121] 3.3 Catalytic Selectivity

[0122] For the pure metal oxide catalysts, γ-Al.sub.2O.sub.3 had the lowest selectivity for the formation of methyl mercaptan, with dimethyl ether being the main product in the temperature range between 300 and 320° C. (with S.sub.DME, 300° C.=71.2% and S.sub.DME, 320° C.=63.4%). With increasing temperature, the selectivity to dimethyl ether decreased to S.sub.DME, 340° C.=49.8% and finally S.sub.DME, 360° C.=28.6%). As dimethyl ether selectivity decreased, the methyl mercaptan selectivity increased from 28.5% at 300° C. to 71.2% at 360° C. The selectivity for the formation of dimethyl sulfide was less than 5%.

[0123] On pure ZrO.sub.2, the selectivity for the formation of methyl mercaptan was 53.1% at 300° C. and 60% at 360° C. Again, the major side product was dimethyl ether, however, with a selectivity decreasing from 46.7% at 300° C. to 36.6% at 360° C. Like for γ-Al.sub.2O.sub.3, the selectivity to dimethyl sulfide was less than 5%.

[0124] Of all pure metal oxides, TiO.sub.2 gave the highest selectivity for the formation of methyl mercaptan: 96% at 300° C., with decreasing selectivity at higher temperatures (S.sub.DME, 360° C.=79.2%). In contrast to the other two metal oxides, the selectivity to dimethyl ether increased with increasing temperature, from S.sub.DME, 300° C.=4.0% to S.sub.DME, 360° C.=16.6%). Dimethyl sulfide was produced with a selectivity of less than 4%.

[0125] Compared to the pure metal oxide catalysts, all Cs loaded catalysts showed a dramatic increase in CH.sub.3SH selectivity. For 10 wt.-% Cs on γ-Al.sub.2O.sub.3, the selectivity for the formation of CH.sub.3SH increased to the range of from S.sub.CH3SH, 300° C.=99.7% to S.sub.CH3SH, 360° C.=98.3%. The selectivity to side product increased with increasing temperature: DME was formed with a selectivity of from S.sub.DME, 300° C.=0.1% to S.sub.DME, 360° C.=0.5%, and DMS was formed with a selectivity of from S.sub.DMS, 300° C.=0.2% to S.sub.DMS, 360° C.=1.2%. Further increase of the Cs loading to 20 wt.-% also increased the CH.sub.3SH selectivity up to a selectivity of from S.sub.CH3SH, 300° C.=99.9% to S.sub.CH3SH, 360° C.=99.1%. In this case, the only side product found was DMS with a selectivity of from S.sub.DMS, 300° C.=0.1% to S.sub.DMS, 360° C.=0.9%.

[0126] For the ZrO.sub.2 based catalysts, the selectivity to CH.sub.3SH increased to a range of from S.sub.CH3SH, 300° C.=99.9% to S.sub.CH3SH, 360° C.=99.1%. Again, the only side found was DMS with a selectivity of from S.sub.DMS, 300° C.=0.1% to S.sub.DMS, 360° C.=0.9%. Increasing Cs loading to 20 wt.-% led to an even higher selectivity between S.sub.CH3SH, 300° C.=99.9% and S.sub.CH3SH, 360° C.=99.4%.

[0127] Similar results were found for the TiO.sub.2 based catalysts: With 10 wt.-% Cs, the selectivity for CH.sub.3SH increased to a range between S.sub.CH3SH, 300° C.=99.9% to S.sub.CH3SH, 360° C.=99.4%. Again, the only side found was dimethyl sulfide with a selectivity of from S.sub.DMS, 300° C.=0.1% to S.sub.DMS, 360° C.=0.6%. Increasing Cs loading to 20 wt.-% again led to an even higher selectivity between S.sub.CH3SH, 300° C.=99.9% and S.sub.CH3SH, 360° C.=99.5%.

TABLE-US-00009 TABLE 9 Summary of the product selectivities of the prepared catalysts (n.d. = not detectable) c(Cs) T S(CH.sub.3SH) S(DME) S(DMS) Catalyst [wt.-%] [° C.] [%] [%] [%] γ-Al.sub.2O.sub.3 — 300 28.5 71.2 <5 — 360 71.2 28.6 10 300 99.7 0.1 0.2 10 360 98.3 0.5 1.2 20 300 99.9 — 0.1 20 360 99.1 — 0.9 ZrO2 — 300 53.1 46.7 <5 — 360 60 36.6 10 300 99.9 — 0.1 10 360 99.1 — 0.9 20 300 99.9 — n.d. 20 360 99.4 — n.d. TiO2 — 300 96 4.0 <4 — 360 79.2 16.6 10 300 99.9 — 0.1 10 360 99.4 — 0.6 20 300 99.9 — n.d. 20 360 99.5 — n.d.

4. Comparative Examples

[0128] Comparative examples were carried out with a catalyst comprising Cs.sub.2WS.sub.4 on γ-Al.sub.2O.sub.3. Said catalyst was prepared by two-step incipient wetness impregnation process. First, 5.0 g γ-Al.sub.2O.sub.3 (analogue to SPH 509 Axens, grain size of 150-250 μm) were impregnated with 0.64 g of cesium acetate (Sigma Aldrich, >99.99%) dissolved in 1.6 mL of H.sub.2O. The sample was dried at room temperature overnight to give Cs/Al.sub.2O.sub.3. Next, the Cs.sub.2WS.sub.4/Al.sub.2O.sub.3 system was synthesized as followed: Cs.sub.2WS.sub.4 crystals were formed by precipitation, mixing a solution of 350 mg of (NH.sub.4).sub.2WS.sub.4 in 20 ml of H.sub.2O and 325 mg of Cs.sub.2CO.sub.3 in 20 ml of H.sub.2O. A yellow precipitate was formed. These solids were filtered, washed with ice-cold water and 1-propanol. Due to the low solubility of Cs.sub.2WS.sub.4, 450 mg of these were dissolved in 150 ml of water. Then 2 g of Cs/Al.sub.2O.sub.3 were added to the solution. The water was removed by evaporation in continuous rotation, precipitating the Cs.sub.2WS.sub.4 crystals on the solid sample. The sample was dried at room temperature overnight. After drying, the sample was calcined at 455° C. for 4 h, with an increment of 5° C./min. The prepared catalyst had a tungsten content of 5.1 wt.-%, a cesium content of 20.6 wt.-%, a pore volume of 0.20 cm.sup.3 g.sup.−1, and a BET surface area of 141 m.sup.2 g.sup.−1, both measured as described above. Adsorption followed by temperature programmed desorption of H.sub.2S was performed with a pulse technique using a flow apparatus equipped with a mass spectrometer (QME 200, Pfeiffer Vacuum). A sample of catalyst was loaded in a quartz reactor and activated in situ under 4.2 vol.-% H.sub.2S/He with a flow of 6 ml/min at 360° C. for 2 h. For H.sub.2S adsorption, the temperature was set to 360° C. and the sample was flushed with He for 1 hour prior to adsorption. Pulses of 4.4 vol.-% of H.sub.2S in He were introduced every 30 min (5.0 μmol/min of H.sub.2S). The total concentration of gas adsorbed was calculated as the sum of the uptakes per pulse.

[0129] The thus obtained catalyst was tested under the same reaction condition and the same reaction tube as in example 3. Prior to testing, the catalyst was activated by treatment in H.sub.2S with a flow rate of 20 ml/min at 360° C. for 2 hours.

[0130] The MeOH conversion, yields for CH.sub.3SH, DME, and DMS and the selectivities for CH.sub.3SH, DME and DMS at temperatures of 300, 320, 340 and 360° C. are summarized in Table 10.

TABLE-US-00010 TABLE 10 Summary of the results of the comparative examples X(CH.sub.3OH) Y(CH.sub.3SH) Y(DME) Y(DMS) S(CH.sub.3SH) S(DME) S(DMS) T [° C.] [%] [%] [%] [%] [%] [%] [%] 300 10.8 10.8 0.00 0.00 100.0 0.0 0.0 320 17.4 17.4 0.00 0.02 100.0 0.0 0.1 340 28.4 28.3 0.01 0.10 99.6 0.0 0.4 360 44.3 44.0 0.03 0.20 99.3 0.1 0.5

[0131] Initial rates for CH.sub.3SH formation are shown in FIG. 15. The highest rate in methanol thiolation was observed at a temperature of 300° C. (1.34.Math.10.sup.−6 mol.sub.CH3SH s.sup.−1 g.sub.cat.sup.−1), with following rates at higher temperatures and the lowest rate at a temperature of 360° C. (6.38.Math.10.sup.−6 mol.sub.CH3SH s.sup.−1 g.sub.cat.sup.−1).