CHEMTECH
March 1999
CHEMTECH 1999, 29(3), 26-30.
Copyright © 1999 by the American Chemical Society.


ENABLING SCIENCE

Designing sulfur-resistant, noble-metal hydrotreating catalysts

This concept is based on the use of zeolitic supports and considers the roles of shape selectivity, hydrogen spillover, and type of sulfur resistance.

Chunshan Song

A s the world continues to rely heavily on liquid transportation fuels, it has become increasingly important to make cleaner fuels that are environmentally friendly in production and use. Hydrogenation and desulfurization of distillate fuels, particularly diesel fuels, are receiving considerable attention because of the increasingly more stringent environmental regulations on the composition of transportation fuel (1-7). Hydrogenation of aromatic compounds is exothermic, and is therefore favored at lower reaction temperatures. However, conventional supported Ni-Mo and Co-Mo sulfide hydrotreating catalysts become active only at relatively high temperatures. Although noble metals are hydrogenation active at low temperatures, their use as catalysts will become attractive only if their sulfur resistance can be greatly enhanced.

In this article, I report a new approach for the design of noble-metal catalysts for hydrotreating sulfur-containing distillates to produce clean distillate fuels such as diesel fuels, jet fuels, and gasoline. This work is motivated by the need for efficient low-temperature catalytic hydrotreating processes and, in part, by a desire to apply recent experimental observations at Pennsylvania State University (8-11) on the hydrogenation of naphthalene over noble-metal catalysts that withstand added benzothiophene.

Applications of noble-metal catalysts
In current processing systems, multiple catalyst beds are used to achieve deep desulfurization and deep hydrogenation. Hydrodesulfurization occurs in the first stage over a Ni-Mo or Co-Mo catalyst, followed by intermediate removal of byproduct gas; finally, hydrogenation over the noble-metal catalyst operates in the bottom bed where the concentrations of catalyst poisons (organosulfur and H2S) are very low (5, 6). Commercial examples of this technology are the Shell Middle Distillate Hydrogenation process, and the SynSat process developed by Criterion/Lummus (5-7). There are no reports of noble-metal catalysts that can operate without such intermediate H2S removal (6). Because of its importance, sulfur resistance of noble-metal catalysts has been the subject of several recent publications (1, 8, 9, 12-15).

Figure 1 shows the SynSat process discussed in the current literature (5-7). The SynSat process is an innovation that merges the fields of catalysis and reactor engineering. SynSat uses several different catalyst beds within a single reactor shell with intermediate removal of byproduct gas (e.g., H2S), and optional countercurrent gas flow. Catalysts A and B in Figure 1 are metal sulfide catalysts such as sulfided Ni-Mo. Catalyst C is noble metal loaded on an acidic support. There is an intermediate gas removal step between the beds of catalysts A and B. Nearly all the sulfur compounds must be converted and removed as H2S on beds A and B before the fuel feed reaches noble-metal catalyst bed C.

Figure 1 thumbnail Figure 1.
New design of a sulfur-resistant, noble-metal catalyst
At Penn State, we have been exploring low-temperature hydrogenation based on noble-metal catalysts (8-11). On the basis of our preliminary results, we are proposing a new concept for designing novel sulfur-resistant noble-metal catalysts for more efficient hydrotreating of sulfur-containing distillates.

The proposed design concept uses unique zeolites as supports for noble metals and uses shape-selective exclusion, hydrogen spillover, and two types of sulfur resistance. Unique zeolite supports can be used to prepare bimodal distributions of noble-metal particles. Some metals are located in small pore openings (<5 Å), whereas others will be contained in large pore openings (>6 Å). The two pore systems interconnect or uniformly distribute so that the systems are close to each other.

Diffusion of organosulfur compounds (such as thiophenic molecules) into the small pores would be inhibited by size (shape-selective exclusion). The large pores would preferentially allow fast diffusion and reaction of bulky polycyclic aromatic and sulfur compounds. The thiophenic molecules could enter the large pores, but not the small pores. However, hydrogen molecules can readily enter both sizes of pores, dissociatively adsorb on metal contained within, and be transported between pore systems by spillover. When the metal in the large pores becomes inactivated by adsorbed sulfur, spillover hydrogen could recover the poisoned metal sites by eliminating R-S-R and R-SH. It is also of value to classify sulfur resistance as either type I, resistance to organic sulfur compounds, or type II, resistance to inorganic H2S (8). The metal species, particularly those in small pores, should have higher type II sulfur resistance. Figure 2 shows a simplified representation of the proposed concept.

Figure 2 thumbnail Figure 2.

The proposed concept for the design of noble-metal catalysts with high sulfur resistance could result in a new class of materials for low-temperature catalytic hydrotreating, which includes but is not limited to hydrogenation and hydrodesulfurization of diesel fuels. The noble-metal catalysts with substantially enhanced sulfur resistance will allow hydrotreating at substantially lower temperatures under lower pressure, which can lead to major improvements in refining efficiency and economics. These new kinds of catalysts can be used for new processing schemes or in existing processes (such as the SynSat process) in combination with Ni-Mo or Co-Mo catalysts in stacked beds within a single reactor shell.

The proposed concept could be used for making either bimodal catalysts (with small and large pore openings) or hybrid catalysts (mixture of one catalyst having mostly small pore openings with another having large pore openings). Such catalysts are expected to significantly improve refining efficiency and economics for producing clean distillate fuels from various low-quality feedstocks such as naphtha, straight-run distillate oils, fluid catalytic-cracking (FCC) naphtha, light cycle oil from FCC, gas oil, coker distillates, visbreaker distillates, and blends of two or more of these. The following is a brief summary of the experimental observations that support our proposal.

Role of acidic zeolite supports. Certain noble metals are more active at low temperatures for aromatic hydrogenation when supported on zeolites than are their counterparts supported on alumina or titania. Noble-metal catalysts loaded on zeolites or other acidic supports will provide higher hydrogenation activity and better sulfur resistance than Al2O3 or TiO2 supports. We observed this trend for naphthalene and phenanthrene hydrogenation over microporous zeolites (8-10) and mesoporous aluminosilicate molecular sieves (11). It is not difficult to imagine the better dispersion of metals on higher-surface-area zeolites than on lower-surface-area alumina or titania. The higher activity of the former may be attributable partly to the better metal dispersion on zeolites and partly to the electron deficiency of metals supported on acidic materials. The concept of electron deficiency was first proposed by Dalla Betta and Boudart (14) and has been accepted by many researchers (1, 14-17).

Some promising results were reported for supported on ultra-stable Y zeolite (12, 13) or hydrogen-Y zeolite (14) in flow reactor tests, and for platinum and palladium on HY or H-mordenite in batch tests (8). Several proposals have been given to explain the weakened metal-sulfur interaction: electron transfer from the metal to the acidic support, creating electron-deficient metal species (13, 14); formation of metal-proton adducts (17); or interaction between the metal and cations in the zeolite (18).

Effect of metal types and zeolites on hydrogenation. Activity and selectivity vary among different noble metals for hydrogenation (8-12). For hydrogenation of a model fuel containing 20% naphthalene at 200 °C over the catalysts prepared in the Penn State laboratory (Table 1), the palladium supported on Y-zeolite or mordenite is more active than platinum on the same supports at the same loading level (8). For a given metal, some zeolites are better than other zeolites as supports for achieving higher hydrogenation activity (8, 10, 12). For example, among the catalysts listed in Table 1, Pd/HM38 is more active than Pt/HM38. Several international catalyst companies are making noble-metal-based, deep hydrogenation catalysts supported on acidic supports such as amorphous silica-alumina and Y-zeolites (19).

TO SIDEBAR: Table 1.

Effect of zeolite structure and metal on sulfur resistance. Y-zeolite-supported noble metals tend to have a higher sulfur tolerance than alumina-supported ones (1). A higher sulfur tolerance in Y-zeolite-supported noble metals has been attributed to electron deficiency (1, 15); however, the commercial Y-zeolite-supported, noble-metal catalysts are not sulfur resistant enough (6). It is also known that different metals may have different sulfur resistance (1, 8, 13).

Our work revealed the importance of the zeolite structure and the metal type for improving sulfur resistance (8). Among the catalysts listed in Table 1, the mordenite-supported palladium catalyst has been found to have a greater resistance to sulfur poisoning than the Y-zeolite-supported palladium; both catalysts were prepared by the same method at the same metal-loading level. As shown in Figure 3, even when the amount of added sulfur is more than twice that of the metal atoms, higher sulfur resistance was obtained with palladium-supported H-mordenite (Pd/HM38) than with Y-zeolite (Pd/HY) (8). Therefore, palladium supported on mordenite can be more active and more sulfur resistant than the same metal supported on Y-zeolite at the same loading level, even in the presence of a large amount of added sulfur--up to a sulfur/metal ratio of 10/1. This observation shows the importance of zeolite structure on the sulfur resistance of a supported metal. Differences in mesopore surface areas between HM38- and HY-supported catalysts are shown in Table 1. Although the contribution of mesopore areas in HM38 cannot be ignored, the platinum supported on the same HM38 is much less sulfur resistant; therefore, the structure of zeolite and the type of metal are important for superior performance.

Figure 3 thumbnail Figure 3.

Two types of sulfur resistance. In batch hydrodesulfurization tests where the H2S product remains within the reactor and in contact with the catalyst (9), we observed gradual recovery of activity of some palladium catalysts poisoned by benzothiophene; the platinum catalysts did not show significant recovery in activity with increasing residence time (Figure 4). Some supported palladium catalysts recovered activity for hydrogenation when the sulfur in benzothiophene was converted to H2S. These experimental observations suggest that noble metals have a different kind of resistance to organic sulfur (type I) and inorganic sulfur (type II). The thiophenic organosulfur compounds have stronger poisoning effects than inorganic H2S.

Figure 4 thumbnail Figure 4.

Shape-selective exclusion and hydrogen spillover. The consideration of two types of sulfur resistance became more useful to us when we conceived the hypothesis of shape-selective exclusion of organic sulfur compounds from small pore openings. There are significant structural differences between mordenite and Y-zeolite. Palladium on the mordenite HM38 shows particularly good resistance to sulfur poisoning. A possible hypothesis is that a significant fraction of the Pd particles are in mordenite side pockets. Bulky organosulfur compounds such as benzothiophene cannot enter into the small-diameter side-pocket channels; consequently, metal particles contained within the side-pocket channels are protected from type I poisoning by a molecular sieving mechanism. These considerations for our experimental results (8) are incorporated into our proposed concept as follows.

Because of the pore size limitation, it is possible to prevent the thiophenic sulfur from diffusing into the small pore openings. In other words, type I sulfur resistance may be lessened if the metals can be placed into the pores that thiophenic compounds cannot enter. Transition metals can be introduced into very small pores (17). Spillover of hydrogen on the catalyst surface is a known phenomenon (19). It is possible for hydrogen generated on the metal sites in the small pore openings to spill over to surfaces in the large pore openings. However, if the metal in the pore openings is too weak to H2S, it will not be effective because hydrodesulfurization will occur to produce H2S. It would be extremely difficult to prevent H2S from entering the small pore openings. This suggests that type II sulfur resistance cannot be improved easily by altering the catalyst pore structure; however, it may be influenced by other factors, such as metal-zeolite interaction. In our design concept, the important role of the small pore openings is to have and keep certain metal sites active even when they are exposed to H2S. Fortunately, the experimental results (Figure 3) have indicated that certain metals on some supports have higher type II sulfur resistance than do others under low-temperature hydrotreating conditions.

Highly sulfur-tolerant hydrogenation with accompanying hydrodesulfurization can be realized when certain metal particles are dispersed in small and large pores. The catalyst itself can foster continuous regeneration of organosulfur-poisoned metal sites in large pores through hydrogen spillover from sites in small pores (Figure 2), thus allowing low-temperature deep hydrogenation of aromatics and hydrodesulfurization. This has been partially supported by our work on low-temperature hydrotreating (8, 9).

During naphthalene hydrogenation at 200 °C over some zeolite-supported noble-metal catalysts in the presence of added benzothiophene, unambiguous evidence on benzothiophene hydrodesulfurization has also been obtained (8), as shown in Table 2. Through detailed analysis, we have found that the added benzothiophene converts to ethylbenzene and even ethylcyclohexane during naphthalene hydrogenation over Pd/HM38 with 373 ppm sulfur. Thus hydrodesulfurization has been achieved using palladium supported on mordenite, even at 200 °C. The proper combination of noble metal (e.g., palladium) and zeolite (e.g., partially dealuminated mordenite) was important for high activity and improved sulfur resistance.

TO SIDEBAR: Table 2.

Active metal sites: monometallic or bimetallic. The proposed concept is not limited to single catalysts or monometallic catalysts. The system may consist either of a bimodal catalyst or a hybrid of two catalysts. The metal species loaded onto the zeolite support can be either a monometallic or a bimetallic species. For design of a suitable bimetallic catalyst, one feasible idea is to introduce the second metal, which is either more sulfur resistant than the first metal or can make the resulting bimetallic species more sulfur resistant. The second metal should be a transition metal that can activate H2 (20, 21); it may or may not be a noble metal. It is known that introduction of a second metal can enhance catalytic activity or sulfur resistance (1, 13, 16, 17, 22). One important issue in our proposed approach is to place the metal species in small pores that have higher type II sulfur resistance.

A promising direction
Our recent work on hydrogenation of naphthalene over mordenite- and Y-zeolite-supported, noble-metal catalysts in the presence of benzothiophene suggests that there may be at least one practical way to use the proposed concept for exploring new catalyst design: That is to load noble-metal-based monometallic or bimetallic species that have relatively higher type II sulfur tolerance into the main channel and the side pocket of a mordenite. To facilitate the diffusion and mass transfer, the mordenite should be properly dealuminated before loading the metal. The metal precursor (monometallic or bimetallic) should be properly introduced into large and small pores by careful preparation.

The proposed concept may have immediate bearing on research into sulfur-resistant catalysts in general, as indicated in our preliminary communication (24). By dividing sulfur resistance into two types, we are able to suggest that type I tolerance can be dramatically enhanced by a structural design that uses shape-selective exclusion, and type II tolerance may be improved by modifying electronic properties of metal species to weaken the metal-sulfur interaction or bonding. Application of the new concept may involve small and larger micropores or mesopores in either a single bimodal catalyst (possessing metal in both small and large pores) or a hybrid of two catalysts in which one contains metal in mostly small pores and the other possesses metals in large pores.

The proposed new design concept is still in the hypothetical stage and must be tested and verified, and many fundamental and practical questions need to be answered by future research. The proposed concept also may have some implications on the design of other catalysts, such as nitrogen-resistant hydrotreating catalysts (23).

Although the concept is not yet fully established, a promising direction of research has been identified. We hope to apply this concept in further experimental studies for developing new catalysts for low-temperature hydrogenation, desulfurization, and denitrogenation of distillate fuels.

Acknowledgments
The author is very grateful to Professor H. Schobert for his encouragement and support, and to his former co-workers A. D. Schmitz, S-D. Lin, W-C. Lai, and K. M. Reddy for helpful discussions on catalytic hydrogenation. This article is based on a lecture given at an ACS Petroleum Chemistry Division symposium during the ACS national meeting in Dallas in March 1998.

References

ACS Pubs ChemPort ChemCenter