Applied Catalysis A, General (v.322, #C)

Preface by C. Geantet; M. Vrinat (1-2).

The Co–Mo–S model is used widely to understand and control the catalytic properties of hydrotreating catalysts. Depending on the nature of the support interactions, the Co–Mo–S structures may be present as either Type I or II which has different catalytic properties. Most current high activity industrial hydrotreating catalysts are based on Type II structures. From a fundamental point of view, it has for many years been difficult to understand in detail the catalytic properties of the different types of Co–Mo–S structures since direct atom-resolved insight has not been available. Recently, such insight has been provided by scanning tunneling microscopy (STM) and from the application of density functional theory (DFT) and together with other advances, this has contributed greatly to improve our fundamental knowledge of MoS2 and Co–Mo–S structures. Many surprising features were discovered. For example, it has been observed that key catalytic steps may involve active sites which are not sulfur vacancies. Of particular interest, it was found that fully sulfur-coordinated sites with metallic-like properties play a role in hydrogenation reactions of such catalysts. The new insight has played a role in the recent introduction of the BRIM™ family of improved industrial hydrotreating catalysts.
Keywords: Hydrodesulfurization; STM; DFT; HADF-STEM; Support interaction; Hydrogenation; Active sites; MoS2; MoS2; Brim sites;

The role of carbon in catalytically stabilized transition metal sulfides by S.P. Kelty; G. Berhault; R.R. Chianelli (9-15).
Since WWII considerable progress has been made in understanding the basis for the activity and the selectivity of molybdenum and tungsten based hydrotreating catalysts. Recently, the focus of investigation has turned to the structure of the catalytically stabilized active catalyst. The surface of the catalytically stabilized MoS2 has been shown to be carbided with the formula MoS x C y under hydrotreating conditions. In this paper we review the basis for this finding and present new data extending the concept to the promoted TMS (transition metal sulfides) systems CoMoC and NiMoC. Freshly sulfided CoMoS and NiMoS catalyst have a strong tendency to form the carbided surface phases from any available carbon source.
Keywords: Transition metal sulfides; Carbides MoS2; MoSxCy; RuS x C y ; CoMoC; NiMoC;

Insight into the formation of the active phases in supported NiW hydrotreating catalysts by E.J.M. Hensen; Y. van der Meer; J.A.R. van Veen; J.W. Niemantsverdriet (16-32).
The sulfidation mechanism of supported NiW hydrotreating catalysts is described in detail based on literature data and experiments for carbon- and alumina-supported NiW catalysts. It is more difficult to convert tungsten oxides to WS2 than molybdenum oxides to MoS2 and therefore typical sulfidation procedures tend to result in oxysulfidic tungsten phases (WO x S y ) next to the layered WS2 phase. The former phase appears to stabilize rather dispersed Ni sulfide particles (NiS–WO x S y ) which is found to be a precursor to the ‘Ni–W–S’ phase: Mössbauer spectroscopy clearly shows that the NiS particles redisperse over the WS2 edges when partly oxidic tungsten phases transform to WS2 slabs. As a consequence of very slow W sulfidation, the Ni promoter ion can end up in two phases, i.e. ‘Ni–W–S’-type and in NiS–WO x S y phases, this in contrast to Co(Ni)Mo catalysts that mainly contain ‘Co(Ni)–Mo–S’ phases after typical sulfidation procedures. The NiS–WO x S y phase is known to perform well in liquid-phase hydrodesulfurization (HDS) reactions whereas the ‘Ni–W–S’ phase is most active for gas-phase HDS. The relative speciation of these two phases for alumina-supported NiW can be controlled by parameters as the calcination temperature and the sulfidation temperature and pressure. Notably, elevated sulfidation pressures are preferred over elevated sulfidation temperatures because in the latter case the WS2 slabs tend to sinter and crystallize which negatively affects the performance. This work shows the existence of a strong support effect for carbon- and alumina-supported NiW. Similar trends were found for amorphous-silica–alumina-supported NiW which is consistent with the finding that the sulfidation mechanism is similar to that of NiW/Al2O3.Even when WS2 sulfidation is completed, as for instance found for a carbon-supported NiW catalyst at 673 K, a significant fraction of Ni is not involved in the ‘Ni–W–S’ phase. Chelating agents improve the edge occupation in NiW catalysts significantly by retarding Ni sulfidation. The result is a more efficient formation of the ‘Ni–W–S’ phase and optimized catalysts have a significantly higher activity than conventional ones.
Keywords: Hydrodesulfurization; NiWS; Sulfidation; Support effect;

The science on the preparation of the hydrodesulfurization oxidic precursors is coming of age, but the chemistry governing their preparation is not yet clearly understood. In this paper, the chemistry of alumina supported oxomolybdate preparation is revisited taking into account the dissolution/precipitation concept recently developed. Classical preparations with the ammonium heptamolybdate and Co nitrate salts will firstly be discussed. Then the use of new starting materials for the preparation of the impregnating solutions will be considered, showing that at high Mo loading the dispersion is strongly dependant on the nature of the starting salts. The use of phosphomolybdate cobalt salts will also be considered. Lastly we will discuss the improvement of the Co promoting effect using molybdocobaltate heteropolyanions as starting materials and complexing agents. This study shows that the maturation step is the determining step for the improvement of preparation of these alumina based oxidic precursor.
Keywords: Hydrotreatment; Catalysis; Heteropolyanions; Complexing agent; Thiophene;

A series of sulfided (Co)Mo/Al catalysts was examined using IR spectroscopy of CO adsorption (T ads  ∼ 100 K), thiophene hydrodesulfurization and 2,6-dimethylaniline hydrodenitrogenation. Spectroscopic analysis after treatments at H2S partial pressures and temperatures close to the working conditions allows one to establish relationship between the nature, environment and concentration of sulfide phase sites and their catalytic functionalities. Unpromoted and Co-promoted edge sites do not exhibit the same sensitivity towards change in H2S/H2 partial pressure ratio. The Mo sites created in low H2S/H2 conditions present functionalities different from those formed in sulfiding conditions. By contrast, the Co-promoted sites are easily created even in sulfiding conditions and the sites generated in high or low H2S/H2 conditions possess the same environment as well as the same reactivity. Note that CoMo catalysts prepared by classical impregnation always exhibit an incomplete promotion of the MoS2 slabs even when containing high Co/Co + Mo ratio.
Keywords: Sulfide catalysts; IR spectroscopy; CO adsorption; Hydrodesulfurization; Hydrodenitrogenation; Hydrogen treatment;

In the open literature, there are numerous papers related to the chemical and physical properties of unsupported transition metal sulfides. Besides characterization studies, the performance of a large variety of mono-, bi- and multi-metallic materials in various test reactions has been examined. The performance studies identified several unsupported materials with higher activity and/or selectivity than the traditional γ-Al2O3 supported Ni/Co–Mo/W catalysts. This has resulted in numerous patents disclosing the preparation of unsupported transition metal sulfides and their use as hydroprocessing catalysts. However, for several reasons (high price and too high activity for existing refinery process equipment), the commercial use of unsupported catalysts in refinery processes has so far been limited to the NEBULA® catalysts. Catalysts based on NEBULA technology provide unprecedented activity for hydrodesulfurization (HDS), hydrodenitrogenation (HDN) and aromatics saturation (HDA).
Keywords: Transition metal sulfides; Hydrotreating; Unsupported/bulk catalysts; Commercial application;

The three typical causes of deactivation of hydroprocessing catalysts are coke, sintering and contamination. The two first can be eliminated by regeneration, the ex situ method being the rule nowadays for a better performance recovery. Regeneration consists in a controlled oxidation which eliminates coke and converts sulfides back to oxides. The main limits to catalyst reuse are the decrease of activity and mechanical properties. Non-contaminated regenerated catalysts can recover activities rather similar to fresh ones, as assessed by a statistical study performed at Eurecat on CoMo catalysts. Nevertheless some new generation catalysts require additional treatments to recover full activity. The handling and transport of spent catalysts to an off-site regeneration facility is currently performed, but requires some precaution as the material is classified as self-heating. At the end of the cycle, spent non-reusable catalysts have to be recycled for metals reclamation. This can be performed either by hydrometallurgical or pyrometallurgical routes.
Keywords: Spent catalyst; Deactivation; Regeneration; Recycling; Hydroprocessing; Reactivation; Hydrotreating; Contamination;

This paper is a review of recent advances accomplished in the field of hydrotreatment (HDT) sulfide catalysts and using theoretical approaches based on the density functional theory (DFT) combined with thermodynamic models and microkinetic models. We illustrate first numerous concepts of modern DFT simulation for a better understanding of the industrial Co(Ni)MoS active phases: localization and role of the promoter, electronic properties and morphological changes induced by the reaction conditions or by promoter addition. Then, it is shown how support effects can be modeled by DFT to provide new insights on the local structure and energy stability of the active phase-support interface, where characterization techniques reach their limits. The comparison between γ-alumina and anatase-TiO2 supports is chosen as a relevant example. Finally, DFT simulations and microkinetic models help to rationalize “volcano-curve” type relationships between hydrodesulfurization (HDS) or hydrogenation (HYD) activities and the calculated sulfur–metal bond energy descriptor. This approach opens new routes to use systematic DFT simulations as a predictive tool. Perspectives for DFT simulations in the area of catalysis by sulfides are suggested.
Keywords: Density functional theory (DFT); Hydrodesulfurization (HDS); Hydrotreatment (HDT); Transition metal sulfides; MoS2; CoMoS; NiMoS; Supports; γ-Alumina; Anatase-TiO2; Volcano curves;

In the field of hydrotreating (HDT) catalysis, density functional theory (DFT) calculations are of great help to explore new active phases on the basis of volcano curve relationships correlating HDT activities and the sulfur–metal (S–M) bond energy chemical descriptor, calculated in transition metal sulfides catalysts. In the present study, we synthesize Mo1−x W x S2 solid solutions supported on γ-alumina. For non-promoted systems, the catalytic tests reveal that a continuous and linear evolution of the catalytic activity is obtained for solid solutions for x varying between 0 and 1. As expected from the calculated S–M bond energy values, no synergy effect is observed in that case. For Ni and Co promoted Mo1−x W x S2 active phases (ternary metal sulfides), the S–M bond energy values determined with an interpolation model of the binary sulfides predict that NiMo1−x W x S2 phases should be more active than NiMoS and NiWS ones. In contrast, CoMo1−x W x S2 phases are expected to develop weak synergetical effect with respect to CoMoS and CoWS ones. Experiments performed on Co and Ni promoted Mo1−x W x S2 active phases confirmed the DFT-volcano curve prediction and an increment of about 30% in HDS catalytic activity is obtained for NiMo0.5W0.5S catalysts in both model molecule conversion and gas oil treatment.
Keywords: Transition metal sulfides; CoMoS; NiMoS; Hydrotreatment (HDT); Hydrodesulfurization; Density functional theory (DFT); Solid solutions; Volcano curves;

Ruthenium sulfide clusters in acidic zeolites: In situ XAS characterization during sulfidation and reaction by Juliette Blanchard; Kyoko K. Bando; Takashi Matsui; Masaru Harada; Michèle Breysse; Yuji Yoshimura (98-105).
In situ XAS at the Ru K edge was used to study the reductive sulfidation (15% H2S/H2) of Ru(NH3)6 3+ supported on HYd zeolite, and the evolution of the sulfided sample under working conditions (lower H2S partial pressure, addition of toluene to the feed). Sulfidation starts at room temperature and is complete at 373 K. At that temperature, the sample is fully sulfided (5.9 S neighbours per Ru). Above 423 K a partial surface reduction (4.3 S neighbours per Ru), and the formation of metallic ruthenium (3.3 Ru neighbours per Ru) are observed. The evolution under working conditions shows that the local structure is not significantly modified when the H2S partial pressure is decreased and especially that it does not lead to an increased reduction. Upon addition of toluene the intensity of the Ru–Ru peak in the FT EXAFS spectra is modified, whereas the Ru–S peak remains unchanged. This is indicative of the chemisorption of toluene on the metallic sites and points out the prominent role of these sites for aromatics hydrogenation.
Keywords: EXAFS; XANES; RuS2; HY zeolite; Toluene hydrogenation; Operando characterization;

Ruthenium sulfide supported on alumina as hydrotreating catalyst by José Antonio De Los Reyes (106-112).
Ruthenium sulfide (either supported or unsupported) has received considerable attention since this chalcogenide displayed prominent hydrodesulfurization, hydrogenation and hydrodenitrogenation activities. Several interesting properties have been observed when RuS2 was deposited on a high surface area support of industrial relevance like alumina. This review will focus attention on studies dealing with RuS2 supported on this carrier over the last two decades. The present paper summarizes then the progress which has been made in the hydrotreatment field, focusing on synthesis methods and activation procedures and their correlation with the catalytic properties for RuS2 supported on Al2O3.
Keywords: Ruthenium sulfide; Alumina; Hydrodesulfurization; Hydrogenation; Hydrodenitrogenation;

Rhenium sulfide in hydrotreating by N. Escalona; M. Vrinat; D. Laurenti; F.J. Gil Llambías (113-120).
Has been observed that rhenium sulfides was among the most active transition metal and sometimes better than industrial catalyst for thiophene hydrodesulfurization and quinoline hydrodenitrogenation. ▪Literature data published mainly during the two last decades and dealing with the preparation, structure and catalytic activities in hydrotreating reactions with rhenium sulfide has been reviewed. These compounds have been particularly evaluated in model compounds reactions as thiophene hydrodesulfurization and quinoline hydrodenitrogenation. In both cases it has been observed that rhenium sulfide was among the most active transition metal sulfides (TMS) and sometimes better than industrial CoMo/Al2O3 and NiMo/Al2O3, the sulfiding conditions appearing to be a crucial step to get highly active catalysts as previously observed for ruthenium sulfide based catalysts. Interest in these catalysts is maintained for real feed in gas oil hydrodenitrogenation reaction (HDN), even if the promoting effect induced by Co or Ni is low compared with classical molybdenum catalysts.
Keywords: Rhenium; Sulfide; Hydrotreating;

The V deposit obtained in petroleum feedstocks hydrodemetallation can leads to significant change for the hydrotreating catalyst activity. Actually, although vanadium sulfide solids have shown only moderate activities, these are significant and higher than predicted by the more accepted theoretical models. A complete review of the activity of these sulfides, bulk or supported, pure or in presence of a promoter is proposed. A mechanism based on a charge transfer leading to change in the energy levels of the electronic bands is proposed as tentative explanation for the synergetic effect observed for the binary vanadium-transition metal sulfides.
Keywords: Hydrotreatment; Vanadium sulphide; Synergetic effect;

Ternary transition metals sulfides in hydrotreating catalysis by Pavel Afanasiev; Igor Bezverkhyy (129-141).
The present review analyses the literature of the ternary sulfides hydrotreating catalysts, with emphasize on the HDS model reactions. Different classes of mixed sulfides are considered, such as Chevrel phases, chromium thiospinels, or pentlandites. Experimental difficulties of such studies are extensively discussed. A tentative of comparison on a common basis of thiophene HDS activity at 573 K and 0.1 MPa is undertaken for the results obtained by different authors. It has been shown that the thiophene HDS activity of ternary sulfides does not correlate with the mean metal–sulfur bonding energy in them, obtained by averaging of the energies for the corresponding binary sulfides. A rough correlation can nevertheless be constructed from the available literature data, relating the mixed sulfides HDS activity to their mean electronegativity.
Keywords: Ternary sulfides; Hydrodesulfurization; Volcano plots; Electronegativity;

Literature data covering the preparation and catalytic properties of Ir sulfide catalysts during the last two decades in relation to hydrotreating reactions, mainly hydrodesulfurization (HDS) and hydrodenitrogenation (HDN) of model compounds, have been reviewed. It has been shown that Ir sulfide belongs to the most active members of the third row noble metal sulfides. The specific activity of a well dispersed Ir sulfide in reactions of model compounds (type of thiophene, indole, pyridine, quinoline) is substantially greater than that of conventional CoMo (NiMo) catalysts. Sulfided Ir in an amount of about 0.5 wt.% is an efficient promoter of the Mo/Al2O3 catalysts, yielding synergetic effects in HDS and HDN of about 3. However, this synergy diminishes at the higher Ir loadings because of the drop in Ir dispersion. Modification of the NiMo/Al2O3 catalysts by Ir usually led to the decrease of HDS activity and the increase of HDN activity.
Keywords: Iridium sulfide; IrMo catalyst; Hydrodesulfurization; Hydrodenitrogenation; Hydrotreatment;

Active phases and sulfur tolerance of bimetallic Pd–Pt catalysts used for hydrotreatment by Y. Yoshimura; M. Toba; T. Matsui; M. Harada; Y. Ichihashi; K.K. Bando; H. Yasuda; H. Ishihara; Y. Morita; T. Kameoka (152-171).
Bimetallic Pd–Pt catalysts are used industrially to saturate aromatics in industrial feedstocks under mild reaction conditions to bypass the thermodynamic limitations. A considerable amount of research effort has been focused on elucidating the structural and electronic properties of bimetallic Pd–Pt particles, generally supported on acidic supports, to correlate their properties with their sulfur tolerance as well as with their catalytic activity/selectivity. However, the properties of bimetallic Pd–Pt particles under these characterization conditions are still partly unknown, particularly what happens during hydrotreating. We therefore prepared bimetallic Pd–Pt catalysts (Pd/Pt atomic ratio of 4/1) using the same precursors of noble metals and various supports, such as acidic and non-acidic ultra-stable Y-type (USY) zeolites, SiO2–Al2O3, SiO2 and Al2O3, and then investigated the structural and electronic properties of the supported bimetallic Pd–Pt particles. These properties that appeared under liquid-phase hydrotreating conditions were correlated with reaction selectivity for tetralin hydrogenation and in 4,6-dimethyldibenzothiophene hydrodesulfurization, as well as with Fourier-transform analyses of adsorbed CO, dispersion and EXAFS data of reduced/sulfided catalysts. The effects of several parameters involved – such as calcination/activation conditions and the presence of extra-framework alumina and chlorine in the zeolite supports – on the sulfur tolerance of the bimetallic Pd–Pt catalysts were also investigated. In addition to sulfur poisoning, agglomeration of the Pd–Pt particles and inhibitory effects caused by nitrogen-containing compounds and aromatics were also investigated to develop measures to minimize the agglomeration of Pd–Pt particles under hydrotreating conditions for real feedstocks.
Keywords: Aromatics saturation; Bimetallic Pd–Pt catalyst; Sulfur tolerance; Nitrogen tolerance; USY zeolite; Hydrodesulfurization; Agglomeration; Deactivation;

Bimetallic PtPd on zirconia catalysts for hydrotreating purposes by E. Devers; C. Geantet; P. Afanasiev; M. Vrinat; M. Aouine; J.L. Zotin (172-177).
In the framework of a two stage process for the hydrotreatment of gas oil, the use of noble metal catalysts in the second stage, where the H2S partial pressure is considerably decreased, was proposed. Thus, the catalytic properties of a series of Pt x Pd100−x (x corresponding to at.% in the alloy) particles supported on zirconia were investigated in tetralin hydrogenation and decahydroquinoline hydrodenitrogenation. TEM micrographs and EDS analysis evidenced the high dispersion of the monometallic and bimetallic composition of the nanoparticles. Whereas a promoting effect is observed in the case of hydrogenation reaction, the introduction of Pd suppressed the catalytic activity in C–N bond scission reaction. This poisoning effect can be simply understood as a linear effect of the segregation of Pd at the surface of PdPt particles. The effect of H2S partial pressure on the HDN activity was investigated and showed that for this reaction thioresistance is not improved by alloying.
Keywords: Zirconia; Hydrogenation; Hydrodenitrogenation (HDN); Platinum; Palladium; Alloys; Thioresistance; Hydrogen sulfide;

Transition metal and bimetallic nitrides, especially molybdenum nitrides, have been reviewed in the areas of synthesis, structure and composition, surface properties and reactivities. The transition metal nitrides with high surface areas were synthesized by the temperature-programmed reactions of MoO3 with NH3 and a N2/H2 mixture for transformation of the MoO3 to Mo2N via MoO2 or chemical vapor deposition. The structure and composition of the Mo nitrides are discussed on the basis of the temperature-programmed desorption and reactions, X-ray photoelectron spectroscopy, and high-resolution transition electron microscopy (HR-TEM). The adsorption behavior of N2, H2, and CO on the nitrides was reported. The transition metal nitrides are effective for the hydrodenitrogenation (HDN), hydrodesulfurization (HDS), hydrogenolysis and hydrogenation. Molybdenum-containing nitrides are very active during HDN, being comparable with sulfided Ni–Mo and Co–Mo catalysts. The nitride catalysts were found to be significantly active during HDS in the initial stage, but rapidly deactivated, from which an equation for the deactivation has been developed. The active sites on the Mo nitrides for the HDN and HDS are discussed.
Keywords: Transition metal; HDN; HDS; Nitrides;

Ni2P catalysts supported on potassium ion-exchanged ultrastable Y zeolites (KUSY) were prepared by temperature-programmed reduction (TPR), and the effect of Ni2P loading and initial Ni/P ratios on the hydroprocessing performance was studied. X-ray diffraction (XRD), and extended X-ray absorption fine structure (EXAFS) were used to obtain structural parameters. Transmission electron microscopy (TEM) analysis showed that the KUSY-supported Ni2P samples consisted of nanoparticles, which were likely situated in the mesoporous cavities or the external surfaces of the zeolite crystals. The catalytic activity was measured at 613 K and 3.1 MPa in a three-phase fixed bed reactor for hydrodesulfurization (HDS) and hydrodenitrogenation (HDN) using a model liquid feed containing 500 ppm S as 4,6-dimethyldibenzothiophene (4,6-DMDBT), 500 ppm N as quinoline, and 3000–6000 ppm S as dimethyldisulfide (DMDS). Partial exchange with K enhanced the catalytic activity for the HDS of 4,6-DMDBT and resistance to N-compound inhibition. The Ni2P/KUSY had high activity with an HDS conversion of 99%, and an HDN conversion of 100%, which were much higher than those of a commercial Ni–Mo–S/Al2O3 catalyst with an HDS conversion of 80% and HDN conversion of 100%, based on equal sites (240 μmol) loaded in the reactor. The sites were counted by CO chemisorption for the phosphide and by low-temperature O2 chemisorption for the sulfide. Deficiency of P in the Ni2P resulted in deactivation, probably due to susceptibility to sulfidation. EXAFS analysis of the catalysts showed that the addition of extra P led to an increase in Ni–P coordination with lengthening of Ni–Ni bond distances, resulting in a high and stable catalytic activity.
Keywords: KUSY; Ni2P; HDS; 4,6-DMDBT;