HDS catalysts

There are many possible formulations for HDS catalysts, and the selection in any one refinery is usually adapted to both the feed to be processed and to the specifications of the corresponding products. However, the most commonly used catalysts are those referred to as promoted Mo or W catalysts which are composed of two metals, predominantly Co-Mo, supported on alumina other common combinations include Ni-Mo, Ni-W, and less frequently, Co-W. The Co-Mo catalysts are excellent for HDS and less active for HDN and hydrogenation reactions which are better performed over Ni-Mo or the more expensive Ni-W (see Section 1.3 below). 

Many other metals have been shown to be active in HDS catalysis, and a number of papers have been published on the study of periodic trends in activities for transition metal sulfides [15, 37-43]. Both pure metal sulfides and supported metal sulfides have been considered and experimental studies indicate that the HDS activities for the desulfurization of dibenzothiophene [37] or of thiophene [38, 39] are related to the position of the metal in the periodic table, as exemplified in Fig. 1.2 (a), 1.2 (b), and 1.2 (c). Although minor differences can be observed from one study to another, all of them agree in that second and third row metals display a characteristic volcano-type dependence of the activity on the periodic position, and they are considerably more active than their first row counterparts. Maximum activities were invariably found around Ru, Os, Rh, Ir, and this will be important when considering organometallic chemistry related to HDS, since a good proportion of that work has been concerned with Ru, Rh, and Ir complexes, which are therefore reasonable models in this sense however, Pt and Ni complexes have also been recently shown to promote the very mild stoichiometric activation and desulfurization of substituted dibenzothiophenes (See Chapter 4). 

---

Sulfided bimetallic clusters which mimic the metal composition of commercial hydrodesulfurization (HDS) catalysts have been prepared and their homogeneous catalytic behavior studied. Reaction of thiophenol with [Mo2Co2(/z4-S)... 

More effective utilization of the metals In HDS catalysts certainly seems possible, because only a very small fraction of the metal content appears to be usefully exposed on the surface. 

It is believe that the HDS sites (rim sites and edge sites) are different than the olefin hydrogenation sites (rim sites) opening an opportunity for the development of selective HDS catalysts [45 171. Another concept to exploit in catalyst development is the competitive adsorption, by which the sulfur compounds inhibit olefins hydrogenation [48]. 

The reasons are elimination of HDS catalysts, such as Ni, Mo, etc., which result in hazardous transition metal waste, less use of energy for heating (since bioprocesses operate at room temperature or at the most 80°C), reduced green house gases such as C02, as well as less production of NOx, SOx, etc. 

SULFICAT A method for presufiding HDS catalysts. Developed by Eurocat. Piloted in 1982 and commercialized in France in 1986. 

Let us take 1978 as the starting point. Massoth [51] then published an extensive review of what was known about the structure of HDS catalysts. Characterization was essentially based on techniques such as X-ray diffraction, electron microscopy, photoelectron spectroscopy, electron spin resonance and magnetic methods. Massoth was rather unhappy with the state of affairs in 1978. He was struck by the ...diversity and apparent contradictions of results and interpretations... It almost seems as though everyone is working with a different catalyst . 

Mo Figure 9.23 Schematic representation of the different phases present in a typical sulfided, alumina-supported Co-Mo HDS catalyst (from Topspe et al. [49]). 

Catalytic and structural information has been obtained for unsupported Co-Mo hydrotreating (HDS) catalysts. 

Figure 6. Photoacoustic spectra of sulfided HDS catalysts. Frequencies (cm-1) of the most prominent absorbance bands of pyridine on the sulfided Mo/AloOj and Co-Mo/AlgOj are indicated. Only bands representative of Lewis acid sites are observed.

The proprietary HD catalyst operates at lower temperatures than incineration and other types of catalysts. This reduces the production of toxic products of incomplete combustion (PICs) to negligible levels as well as minimizing energy demands. 

HDS catalysts have been characterized extensively with a wide variety of tools, and several extensive reviews of the subject have been presented (85,88-91). Substantial effort has been aimed at relating catalytic activity and selectivity to microscopic properties such as catalyst composition, electronic structure, and geometric structure. EXAFS investigations of working catalysts have provided information about the composition, average local coordination, and interatomic distances of atoms in the catalyst clusters. It has been concluded that the active phase under operating conditions is MoS2-like particles with a dimension of 10—20 A (92-94). 

Notwithstanding the impressive number of investigations of HDS catalysts, many fundamental questions remain unanswered, and most of them are obviously related to the fact that traditional spectroscopic techniques are not capable of unequivocally mapping the real-space atomic structure. For instance, what is the preferential shape of the M0S2 crystallites The basal planes of the M0S2 slabs are chemically inert, and it is therefore clear that the HDS activity is associated with the... 

The clusters are also very similar to each other in size, with an average area of 500 A. This area corresponds to a side length of a triangle of approximately 30 A, which corresponds well to the dimensions of the active clusters in typical HDS catalysts. As a model system for the HDS catalyst, the clusters are therefore a well-characterized reference for further experiments to elucidate details of the structure of the active edges and the reactivity with adsorbed molecules. 

Gas oil refining currently operates at moderate temperatures (340-360°C) and hydrogen pressures of 3.0-5.0 MPa, usually with CoMo/A1203 catalysts. These conditions are unable to achieve the 0.05% S specification of gas oil. Increasing the temperature can achieve the S goal, but the color of the produced oil becomes degraded at the elevated temperature due to unwanted side reactions. Consequently, it is necessary to understand the HDS reactivity of refractory sulfur species in the gas oil and to clarify possible inhibition mechanisms that may limit HDS reactions. The design and development of more effective HDS catalysts that will minimize the increase of the costs to the refinery are severely needed. 

The reaction mechanisms for conversion of the refractory sulfur species such as 4,6-dimethyldibenzothiophene (4,6-DMDBT) in gas oil and the inhibition of their HDS by coexistent species need clarification. Novel mechanistic routes that enhance the reactivity or reduce the inhibition are necessary for designing more efficient deep HDS catalysts and processes. 

It is believed that conventional HDS catalysts possess two very different types of catalytic sites which contribute to the different pathways described earlier. One induces the direct extraction of sulfur (kDo), and the other catalyzes aromatic ring hydrogenation (kHs,) CO- This topic is discussed in... 

D. Composition and Structure of Present-Day HDS Catalysts 1. Structure and Classifications of the Co(Ni)-Mo-S Species... 

With improvements in the preparation of more active HDS catalysts, MoS2 crystallites became smaller, and traditional physical techniques for characterization such as X-ray diffraction (XRD), scanning electron microscopy (SEM), and transmission electron microscopy (TEM) became limited. In fact, today s best catalysts do not exhibit XRD patterns, and the active catalyst particles can no longer be observed directly by TEM. Thus, new techniques were required to provide structural information about Co(Ni)-Mo-S catalysts. As modern surface science characterization procedures evolved, they were immediately applied to the study of CoMoSx-based... 

Throughout the previous discussions, HDS catalysts were described as containing two different types of catalytic sites, one that facilitates direct sulfur extraction and another that facilitates hydrogenation. This could easily be rationalized in catalysts of a few years ago wherein the distribution of the promoter in the catalyst surface was uncertain, the crystals of MoS2 were large, and the composition of the support was variable. However, as catalysts have been improved, the crystallite sizes have been reduced to as small as seven Mo atoms in a cluster, and the stoichiometry of promoter to Mo is optimized at 1/2. The surface of the support is now carefully controlled, and the stacking of MoS2 can be dictated with reasonable accuracy. With such improved catalysts, it now becomes difficult to surmise how two different types of sites can exist, each with a different composition and function. 

---