Introduction

The field of solid state chemistry has produced numerous studies on the optical, electrical, electrochemical and catalytic properties of transition metal sulfides (TMS). In general, TMS with layered structures exhibit similar catalytic behavior in hydrotreating reactions [1]. While some preparation methods yield highly crystalline, low surface area compounds [2,3], others lead to materials with poor crystallinity and greater surface area, along with appreciable catalytic activity. Such methods include co-maceration [4], precipitation [5, 6], and thiosalt decomposition [7]. In the search for new catalysts, some in situ methods have been developed which yield high surface area sulfides. The introduction of alkyl ammonium thiometallate precursors, for example, has produced compounds with greater surface areas than earlier generations of sulfide catalysts [8,9]. The synthesis of MoS2 by a hydrothermal route also leads to the formation of a highly dispersed compound. Thus, the decomposition of ATTM in water under 20 bar N2 or H2S atmosphere has been studied at 15 min intervals for up to 4 h and at different temperatures in the region of 473–573 K, yielding catalysts with areas between 50 and 95 m2/g [10].

Increasing attention has been paid to the role of carbon in the stability of MoS2 and RuS2 catalysts, with some studies suggesting that the active phase of these compounds involves a high surface area MoS2-x C x compound. The first claims of the importance of carbon in these catalysts are made with regard to the synthesis of RuS2-x C x compounds, where carbon atoms substituting surface sulfur atoms were detected [8,9]. Catalysts obtained from carbon-containing salts of the type (NR4)2MoS4 (where R= alkyl group) have greater catalytic activities than those obtained from the ammonium thiomolybdate salt [8,11–13]. Studies also indicate a noticeable increase in the surface area of the catalyst obtained from carbon-containing ammonium thiosalts, like in the case of MoS2-x C x which has a surface area of 152 m2/g when prepared from an ethylene-diamine thiosalt, and a surface area of 243 m2/g when prepared from a tetrabutyl ammonium salt [12,14].

Prior synthesis of this type of catalyst has been done in situ by decomposing ATTM in a reactor containing sulfur and H2 [10]. In a recent study, in situ catalysts are found to yield greater surface areas than those prepared by conventional ex situ thiosalt decomposition, where the thiosalt is treated with a mixture of H2S/H2 at 673 K for 4 h [15].

A recent proposal takes aqueous solutions of ATTM, then adds reducing agents such as hydrazine (HZN) or hydroxylamine and in some cases a surfactant agent like cetyl-trimethylammonium chloride (CTAC), to produce molybdenum sulfides with surface area as high as 211 m2/g [16].

In this work, the effect of using non-stoichiometric quantity of CTAC and/or HZN to treat ATTM precursors in a variation of the above method is studied, looking to improve catalytic activity for the hydrogenation of cyclohexene, a reaction that is little studied in connection with TMS catalysts compared to HDS.

Experimental

Preparation of ATTM

The precursor thiosalt ATTM, (NH4)2MoS4, is prepared according to the method described by Berhault et al. [17]. A weight of 4 g of ammonium heptamolybdate (NH4)6Mo7O24·4H2O (Fluka, 99.0%) is dissolved in 20 mL of water. To this solution is added 38.4 mL of concentrated ammonium sulfide (Aldrich, 20%). The mixture is heated at 333 K for 1 h then cooled in an ice bath for 3 h to induce crystallization. Crystals are separated by filtration, washed with isopropylic alcohol and left to dry at room temperature. The resulting ATTM is then kept in N2 atmosphere to avoid oxidation. The purity of ATTM is confirmed by XRD.

Thermal decomposition of ATTM in the presence of a gas

After grinding, 1.0 g of ATTM is placed in a porcelain boat, which is then introduced in the alumina tube of a tubular furnace with a H2 or N2 gas flow. Temperature is increased to 723 K using a 10 K/min heating rate, then kept constant for 2 h. At the end the furnace is allowed to cool to 473 K. Immediately after, gas flow is cut and the system is allowed to cool to ambient temperature. Finally, the sample is kept in N2 atmosphere.

Molybdenum sulfide catalyst precursor modified with HZN

Since ATTM, HZN and CTAC are sensitive to an oxidizing environment, preparations are done in a N2 atmosphere glove box. A weight of 2.8 g ATTM is dissolved in 100 mL of water. Elsewhere, 2 mL of HZN (N2H4, Aldrich 35%) are dissolved in 100 mL of water. Both solutions are then mixed together. The mixture is placed in a flask fitted with a condenser, then heated in an oil bath to 333 K for 6 h with constant stirring. The resulting precipitate is filtered, washed with water, dried at 348 K and kept in N2 atmosphere.

Molybdenum sulfide catalyst precursor modified with CTAC

Working in a glove box, a weight of 2.8 g ATTM is dissolved in 100 mL of water, then 20 mL of CTAC 0.1 M (from Aldrich, 25%) are added. The resulting mixture thus contains a non-stoichiometric rate of CTAC/ATTM (0.20 mol/mol) which is placed in a flask and the rest of the procedure is the same as in section 2.3.

Molybdenum sulfide catalyst precursor modified with HZN and CTAC

A weight of 2.8 g ATTM is dissolved in 100 mL of water, to which is added a solution consisting of 2 mL HZN (35%) in 100 mL of water, then finally adding 20.0 mL of CTAC 0.1 M. The mixture is placed in a flask and the rest of the procedure is the same as in section 2.3.

Characterization of samples

The BET specific surface area of the solid catalyst samples is measured with a Micromeritics Gemini 2060 surface area analyzer using N2 as adsorbate gas and a sample mass of 0.2–0.3 g. Diffraction patterns are obtained by powder XRD diffraction with a Philips X’Pert analytical diffractometer using CuKα radiation. High resolution electron microscopy employs a JEOL 2010 transmission electron microscope; samples are suspended in isopropanol, deposited on holey carbon-coated copper grids and air-dried.

Catalytic activity

Catalytic activity for cyclohexene hydrogenation is tested in a high pressure 300 mL Parr reactor by placing 20 mL of cyclohexene with 0.3 g of catalyst. The reactor is purged of residual air, pressurized with H2 to 35.91 KPa (750 psi) then heated to the reaction temperature of 523 K in about 10 min. A stirring rate of 50 rpm is used. The advance of the reaction is monitored by GC using samples taken every 15 min during the first hour, then every 30 min for the next 4 h. Reduction of sample volume ≤ 5%. Catalytic activity is expressed in terms of percentage conversion of cyclohexene and initial reaction rates.

Gas chromatography

Samples obtained from the reactor are analyzed using a Hewlett-Packard 5890 gas chromatograph with FID detector. A low polarity J&W DB624 capillary column, 30 m long, 0.53 mm diameter, 3.0 μm thick liquid phase is employed. Column temperature is 373 K, using a N2 carrier gas flow of 3 mL/min and a split rate of 1:10.

Results and discussion

ATTM decomposition

The surface areas of the products of thermal decomposition of ATTM at different temperatures and in different gas environments are found to be in the range of 5–10 m2/g, as shown in table 1. Changes are attributed to a varying degree of crystallite synthering. Other studies have found surface areas as low as 5.5 m2/g and up to 50 m2/g for the decomposition of ATTM in the presence of H2/H2S [7,15]. Another work using the same conditions for ATTM thermal decomposition finds that the surface area of 50 m2/g, decreases to 4 m2/g when a temperature of 973 K was applied [18]. These facts show the notable influence of temperature on surface area of catalysts. The catalysts prepared using either a H2 or N2 gas environment during the thermal decomposition were found to give surface areas of the same magnitude.

Table 1 Specific surface areas corresponding to the MoS2 obtained by thermal decomposition of ATTM under different conditions
Full size table

Decomposition of chemically modified ATTM

The surface area results of MoS2 catalysts obtained by modifying ATTM with HZN and/or CTAC and later decomposing them in N2 atmosphere at 723 K are reported in table 2. The surface area of the catalyst doubles with the use of HZN and becomes over 20 times greater with the use of the surfactant, suggesting the formation of very small crystallites. Earlier work by Alvarez et al. reported that in situ catalysts yield greater surface areas than those prepared by conventional ex situ thiosalt decomposition, with reported values of 59.9 and 8.0 m2/g for in situ and ex situ prepared molybdenum sulfide, respectively [15]. The increase in surface area is attributed to the presence of carbon-containing alkyl groups in the structure of (NR4)2MeS4 (R:H or alkyl:Me:Mo or W). In a similar way, the high surface area of the catalysts here obtained from chemically modified precursors is thought to be due to surface structural carbon derived from the surfactant material.

Table 2 Surface areas of the MoS2 obtained from different precursors by thermal decomposition at 723 K in N2
Full size table

XRD patterns

The XRD pattern of the pure ATTM phase in figure 1 consists of sharp peaks, the strongest of which are identified as the (020), (111), (200), (220), (002), (310), (311), (212), (222), (330), and (142) reflections and is in good agreement with the reflections reported for orthorhombic (NH4)2MoS4 (PDF 48–1662). The XRD pattern of the material produced by thermal decomposition of ATTM + HZN + CTAC in H2 is shown in figure 2a and is very similar to that of the decomposition product of ATTM in H2 shown in figure 2b, except for the reflection at 2θ = 14°. With their broad peaks, the diffraction patterns suggest that both materials are composed of small crystals. The reflections agree with those reported for the structure of poorly crystalline MoS2–2H [18–20]. The peak at 2θ = 14° corresponds to the layered stacking of the (002) planes. The reflection corresponding to the (100) planes is less intense than in MoS2–2H crystals and the (103) planes are absent. The attenuated (100) diffraction peak, along with its asymmetric shape, is related to a loss in stacking sequence [20]. From table 1, the fact that the decomposition product of ATTM in H2 has a smaller surface area (5.5 m2/g) compared to the one obtained by using N2 (9 m2/g), along with the resemblance of the XRD spectra, suggest a greater sintering when decomposing in a H2 atmosphere. It is known that decomposition of the thiosalt at 673 K leads to the formation of MoS3 and finally MoS2 at 723 K [7].

Figure 1.
figure 1

Diffractogram of freshly prepared (NH4)2MoS4.

Full size image
Figure 2.
figure 2

Diffractogram of the MoS2 obtained by thermal decomposition in H2 of (a) ATTM + HZN + CTAC and (b) ATTM.

Full size image

The XRD patterns for the catalyst materials obtained by decomposition of modified ATTM in N2 are shown in figure 3 Analysis of the nearly identical figure 3a and b, corresponding to the decomposition products of TTMA + HZN + CTAC and TTMA + CTAC, respectively, identifies the (100), (103), (110) and (200) reflections of MoS2–2H. It is important to note the absence of the reflection corresponding to the (002) family of planes. This suggests the formation of crystallites consisting of only a few atomic layers. The pattern in figure 3c is nearly identical to figure 2b and corresponds to ATTM decomposed in N2.

Figure 3.
figure 3

Diffractogram of the MoS2 obtained by thermal decomposition in N2 of (a) ATTM + HZN + CTAC, (b) ATTM + CTAC, and (c) ATTM.

Full size image

Electron microscopy

High resolution electron micrographs of the MoS2 obtained by thermal decomposition of pure ATTM and the MoS2 obtained by thermal decomposition of ATTM + HZN + CTAC are similar to the one in figure 4. The micrographs show no long range crystalline order, but mostly different areas made up of only a few atomic layers with the 0.62 nm interplanar spacing characteristic of the (002) family of MoS2–2H planes. The MoS2 from modified ATTM seems to exhibit a greater dispersion than the one from pure MoS2.

Figure 4.
figure 4

Electron micrograph of the MoS2 obtained by the thermal decomposition of ATTM + HZN + CTAC.

Full size image

Catalytic activity

Hydrogenation activity measurements for catalyst samples are presented in figures 5 and 6. Analysis of the kinetic data by the integration method finds that the hydro-genation reaction closely follows a first-order rate law over the 5 h reaction time, with a correlation factor of over 0.98. The corresponding hydrogenation reaction rates at the initial cyclohexene concentration are then calculated. Since molybdenum disulfides are anysotropic and their activity has no direct relation with their total surface area, catalytic activity is expressed in terms of rates per catalyst weight rather than rates per catalyst surface area, as shown in table 3, in order to compare their activity with that of other catalysts reported in the literature. Thus, the initial reaction rates at 523 K for both ATTM-derived catalysts are greater than those found by Ramirez et al. for MoS2/Al2O3 catalysts, which are in the order of 8  ×   10−6 mol/gcats at 573 K [21].

Figure 5.
figure 5

Catalytic activity of the MoS2 obtained by thermal decomposition of different precursors in H2.

Full size image
Figure 6.
figure 6

Catalytic activity of the MoS2 obtained by thermal decomposition of different precursors in N2.

Full size image
Table 3 Initial reaction rates (105 mol/g cats) for cyclohexene hydrogenation with MoS2 catalysts obtained from untreated and treated ATTM
Full size table

According to figures 5 and 6, the catalyst obtained by decomposition of the ATTM in H2 yields a 13% conversion, compared to 33% for the catalyst obtained from the ATTM decomposed in N2. For samples prepared in N2, the corresponding reaction rate ratio, \({{\nu}}\) (ATTM + HZN + CTAC) : \({{\nu}}\) (ATTM + CTAC), is 1.03, while that of the samples prepared in H2 is 1.10. Thus, the combination of HZN + CTAC seems to contributes little to the enhancement of MoS2 catalytic activity over CTAC alone. The catalysts derived from ATTM + CTAC in both figures reach over 90% cyclohexene conversion after 5 h, compared to less than 35% for those derived from pure ATTM. The corresponding reaction rate ratio, \({{\nu}}\) (ATTM + CTAC) : \({{\nu}}\) (ATTM), is 14 for samples prepared in H2, and 7 for samples prepared in N2.

The improvement in activity for the catalyst prepared by decomposition of ATTM in N2 relative to decomposition in H2 [\({{\nu}}\) (ATTM, N2) : \({{\nu}}\) (ATTM, H2) is 2.8], is due to a greater amount of active sites responsible for hydrogenation. Decomposing the thiosalt in H2 may in fact induce the generation of species like Mo(II), which is not the case with the decomposition in N2. In terms of the so called “rim-edge model” used to discuss the hydrodesulfurization (HDS) and hydrogenation behavior of MoS2 catalysts [22], catalyst microcrystals consist of a stack of n monolayers of diameter d. The top and bottom layers have both border and rim sites, while the middle layers have only border sites; the top and bottom surfaces (basal planes) are inert. According to the “rim-edge” model, desulfurization reactions take place on both rim sites and border sites, while hydrogenation only occurs on rim sites. The greater activity of the catalysts obtained by thiosalt decomposition in N2 compared to those obtained in H2 would thus imply a greater number of MoS2 crystallites having rim sites independent of the number of atomic layers in-between, as suggested by the greater surface area of the catalyst obtained using N2. In these crystallites the presence of the “rim” is important, since it constitutes the site assigned for hydrogenation.

The hydrogenation activity of the MoS2 obtained from ATTM treated with CTAC yields about 95% conversion in 5 h, as seen in figure 6. The activity of the MoS2 obtained by ATTM treated with both HZN and CTAC is only slightly greater than that of the catalyst derived from ATTM + CTAC alone, observing a greater slope at the beginning of the reaction in both cases. The TEM micrograph of the MoS2 derived from ATTM treated with CTAC, figure 4, shows the typical “rag structure” described by Chianelli [23] for these sulfide catalysts.

Conclusions

ATTM treated with CTAC using a 1:0.2 molar ratio ATTM/CTAC yields MoS2 catalysts with surface areas comparable to those reported in the literature for MoS2 catalysts obtained using a 1:1 ratio of CTAC/ATTM. In other words, the 1:1 radio is not critical to attaining high surface areas.

The XRD pattern of the ATTM + CTAC system decomposed in N2 does not contain the (002) reflection, indicating that the resulting molybdenum sulfide is made of amorphous material or very small crystallites, consistent with an efficient dispersion produced by the tensoactive agent.

The use of the tensoactive agent in the ATTM + CTAC derived catalyst produces a 7-fold increase in the hydrogenation rate constant, compared to that of the ATTM derived catalyst, when the precursors are decomposed in N2, and a 14-fold increase when precursors are decomposed in H2.