Fundamental research, using a huge arsenal of analytical and material-characterization tools, is yielding new and improved catalyst supports
Readers of this magazine certainly understand the importance of catalysts in the chemical process industries (CPI), where they are used for speeding reactions, moderating operating conditions and minimizing the formation of byproducts. Although some types of catalysts provide their own mechanical strength (the so-called bulk catalysts, such as metal gauzes and foams and amorphous oxides), a large class of catalysts are dispersed onto the surface of a support material, and the support is as important as the catalyst itself. Therefore, making a catalyst system for a commercial production process requires the development of the complete package — catalyst plus support. For this reason, catalyst suppliers are actively working on both simultaneously, tailoring the catalyst system for each and every application. And this entails fundamental research and modeling of surface chemistry coupled with advanced analytical tools for measuring physical properties (Figure 1), such as porosity, surface area, particle size, and more. Presented here are some examples of recent developments in catalyst supports.
Types of supports
“The term ‘catalyst support’ is maybe a bit misleading and for sure underrating its role for a catalytic reaction,” says Marvin Estenfelder, head of R&D at Clariant’s BU Catalysts (Munich, Germany; www.clariant.com). Generally, the term “support” is used with two very different meanings, explains Estenfelder: “In some reactions and applications — like selective oxidations — a non-active support is used to provide a geometrical structure for the active material that is coated in a thin layer onto it. In such cases, very often ceramic materials like steatite or cordierite are used and the support typically has very low porosity and surface area. It is nevertheless required since a fully body of the active material would be too active and too much heat would be released during reaction, leading to a dangerous ‘runaway.’ Due to a lack of reactive surface groups, such supports require specific binders to attach the active layer tightly onto its surface. A specific example is the selective oxidation of o-xylene to phthalic anhydride. Interestingly, the ceramic structural support applied is coated with an active layer which itself consists of a titania support on which the active species vanadia is spread. Clariant’s PhthaliMax series combines high physical strength and adherence of the active layer to the support, as well as a tailored activity pattern providing highest selectivity and yield.”
“In the second group of reactions, the catalyst support offers a high surface area to which the catalytically active material is affixed,” Estenfelder continues. “It is important to mention that the role of the support in those cases goes beyond the deposition of the active material on it. The catalyst support plays a pivotal role for most catalytic transformations itself by either adsorption of reactants or intermediates, including spillover effects from or to the active sites, by activation of the substrates by, for example, polarization or (de)protonation or by inducing unwanted side-reactions. Keeping this in mind, two things become quite clear: 1) each development of a catalyst is inseparably connected to the development or at least tailoring of the support for the specific reaction; and 2) as there are many ways a catalyst support may influence the catalytic reaction, there is no main characteristic of a support material but all of them have to be optimized at the same time,” says Estenfelder. “The support material is one integral part of the catalyst and with each new catalyst developed, a new or modified support tailored to the specific application is used.”
“It is essential that the catalyst support is compatible with the reaction conditions for the particular catalytic process so that the support does not undergo any kind of chemical change,” says Michael Brorson, senior expert at at Haldor Topsøe A/S (Lyngby, Denmark; www.topsoe.com). “Such change could be at the surface (for example, sulfatization of alumina in contact with SOx atmospheres), and this would affect the catalytically active phase dispersed on the surface. The change could also be in the bulk if the support material is not thermodynamically stable at process conditions; this will eventually lead to complete loss of integrity of the catalytic pellet or extrudate that could pulverize and lead to dramatic pressure drop increase in the reactor,” says Brorson.
Topsøe has developed sour shift catalysts — water-gas-shift catalysts that can be operated in high-sulfur atmospheres, which poison the usual copper- and iron-based shift catalysts. Such atmospheres are encountered when processing downstream gases from coal or petcoke gasification. The active phase in these sour shift catalysts is cobalt-molybdenum-sulfide and is similar to the active phase of hydrodesulfurization catalysts used in oil refineries, explains Brorson. But while the hydrodesulfurization catalysts have aluminum oxide as support, this material is not compatible with the steam content of gasification gas. Topsøe’s sour shift catalysts are therefore supported on high-surface-area magnesium aluminum oxide, which is stable in steam.
For a given support material, there are typically three parameters to optimize when it is shaped into pellets or extrudates (Figure 2): surface area, pore size and pore volume, says Brorson. These parameters are geometrically interlinked, so the optimal combination will always be a compromise. For example, a very high surface area is good for dispersion, but tends to imply small pores and consequently diffusion restrictions, explains Brorson. Also, while a low pore volume will lead to high packing density of the catalytic pellets or extrudates, which is good for increasing catalyst volume activity or reducing reactor volume, it may have disadvantages in the catalyst production process.
“In recent years, Topsøe’s continuous R&D efforts have enabled us to modify our in-house manufactured aluminum oxide precursor materials for producing aluminum oxide supports,” says Brorson. “This has allowed us to increase porosity and, with that, improve the impregnation with active metals.”
While some supports must remain inert, others can play an important role by participating in the reaction. A good example of this is the catalyst system used for controlling emissions from gasoline engines. Gasoline emissions control requires the simultaneous conversion of hydrocarbons (HCs), CO and NOx under fluctuating conditions, explains Nicola Collis, principal scientist at Johnson Matthey Plc (JM; London, U.K.; www.matthey.com). Unlike diesel, which is always lean (oxidizing), gasoline is a stoichiometric exhaust with fluctuations between rich (reducing) and lean conditions. Therefore, we use a ceria zirconia support, where the ceria also serves as an oxygen-storage material. Under difficult reducing conditions, the Ce2O4 releases O2 for oxidizing the HCs and CO; under difficult oxidizing conditions, the Ce2O3 captures O2, enabling the destruction of NO, says Collis. Although this system has been commercial for a long time, there continues to be work on improvement, she says.
Another example of the support playing a role in the catalyst system was described in Nature Communications, published in February. There, researchers from the Ruhr-Universität Bochum (RUB; www.ruhr-uni-bochum.de), the Max Planck-Institute for Energy Conversion (Mülheim, both Germany) and Pacific Northwest National Laboratories (Richland, Wash.) describe a DuBois-type-complex fuel-cell catalyst they developed that features a self-defense mechanism against O2 — a deactivating catalyst poison.
To do this, the scientists introduced a hydrophobic and redox-inactive polymer as immobilization matrix for the nickel-complex (DuBois-type) catalyst. By embedding the catalyst into the polymer matrix, two separated reaction layers form: a catalytically active layer close to the electrode surface and a protection layer at the polymer-electrolyte interface. The first layer allows for an efficient conversion of H2 at the electrode surface and the second layer removes incoming O2 at the interface and thus protects the active layer from oxygen damage.
The century-old vanadium-based SO2-to-SO3 oxidation catalysts used for manufacture of sulfuric acid contain a supported liquid consisting of vanadium oxide dissolved in alkali metal pyrosulfates, says Topsøe’s Brorson. The liquid is dispersed in the pore system of silicon dioxide support pellets. Reactions take place by diffusion of SO2 and O2 into the melt, where they catalytically react to form SO3, which subsequently diffuses out of the melt.
Using in situ transmission electron microscopy, Topsoe researchers have shown that the melt dynamically redistributes over the silica carrier surface in ways depending on temperature and precise melt composition. The latter depends on both the fixed metal concentrations and the more variable composition of the gaseous atmosphere that partially dissolves into the melt, as reported in a 2016 issue of J. Phys. Chem.
On convex carrier surfaces in the pore system, the melt sometimes partially crystallizes while concave surfaces accumulate liquid by melt migration. “This knowledge has been used for the development of an improved carrier (Figure 3) used for Topsoe’s new VK-711 Leap5 catalyst, says Brorson.
At Clariant, “we have been working for quite a while on the very promising concept of supported ionic liquids,” says Estenfelder. “Both sub-concepts SILP (supported ionic liquid phase — supported homogeneous catalysts dissolved in an ionic liquid dispersed on a support) and SCILL (solid catalyst with an ionic liquid layer) — heterogeneous catalyst with an ionic liquid as reaction promotor) show great promise for industrial application,” he says. “We are working on various applications together with partners and customers, for example, on the water-gas-shift reaction. For the final commercialization, some important questions still have to be answered, he says. “We have made good progress and we are in the final stage of evaluation for one development.”
Researchers at Evonik Industries AG (Essen, Germany; www.evonil.com) are also developing SILP catalyst technology for hydroformylation reactions (see Chem. Eng., October 2015, pp. 18–24).
There are some reactions for which only carbon can be used as a support, says Franchessa Sayler, CEO at ThruPore Technologies, Inc. (New Castle, Del.; www.thrupore.com). “Hydrogenation reactions use carbon — there is no alternative,” she says. Conventional carbon supports are made by mixing carbon particles with a binder and then extruding to form pellets. Such carbon supports have a high surface area, as measured by traditional methods, but when viewed under a microscope, the active metal tends to simply sit on the outside, so only the outer surface of the surface is used. Such pellets also suffer from attrition issues, explains Sayler. “We have taken synthetic carbon and introduced a high degree of porosity, via a proprietary process,” says Sayler.
Normal carbons are either natural-product based or residues from petroleum refineries. Traditionally, the best source of carbon material had been coconut shells, which have more porosity than alternatives, explains Sayler. But because they are derived from biomass, that carbon also contains traces of nitrogen, phosphorous and sulfur, which can poison a catalyst, she says. “We use a chemical method to get carbon with a very low ash content and no sulfur or other catalyst poisons.”
ThruPore Technologies was cofounded by Sayler and professor Martin Bakker in 2012 at the University of Alabama. As a graduate student, Sayler figured out how to control the pore size of a catalyst’s support structures, creating 50% more accessible surface area. “We can put any precious metal catalyst on the support, but we first started with palladium on carbon, because this is an important industrial catalyst for hydrogenation reactions,” she says. ThruPore’s C/Pd catalyst is six times more active than existing catalysts, and requires 30–50% less precious metal to achieve the same performance, she says.
ThruPore has formed a manufacturing partnership with Inventure Renewables, Inc. (Tuscaloosa Ala.). Current production is approximately 20 ton/yr, and the next step will be 100–200 ton/yr, says Sayler.
Control of particle size
In slurry-type polymerization reactions, the size of the catalyst support can be an important parameter. Evonik has recently introduced its latest generation of Catylen Ziegler and Ziegler-Natta catalyst, Catylen S 300, which features a magnesium ethanoate support material that is produced in precisely the required particle size. The size can be selected within a range of 18 to 80µm — up to 20 times smaller than the support material of its predecessor, Catylen S 100, according to Budo Richter, head of R&D Olefin Polymerization Catalysts. The particle size distribution can be controlled to within±1µm.
For the last few years, Johnson Matthey has been developing 3-D printing (additive manufacturing) as a forming procedure to create experimental ceramic shapes that were previously unobtainable using traditional forming methods, like extrusion or pelletization, says Alex Munnoch, a research scientist at JM. “With this technology, we can control internal features and spatial distribution through a 3-D printed structure; such structures could then be utilized in combination with active catalyst materials, which could be incorporated through a variety of traditional methods, such as impregnation, precipitation, and coating,” he says.
“At the moment, we are focusing our research on the powder bed binder-jetting method, where we start with a 3-D model using CAD, [computer-aided design] and then use an ink-jet type printer to print 2-D slices of the shape onto successive thin layers of ceramic powder to create the final 3-D structure of the support precursors, layer by layer. The subsequently sintered structures have a complex shape with a high geometrical surface area, making them industrially important for mass-transfer-limited surface reactions, where the amount of active material matters,” says Munnoch. “Notably, the heat/mass-transfer know-how developed as part of our heat-exchanging structured Catacel technology complements our additive manufacturing R&D.”
“Although the printed materials are very open and look fragile, these sintered ceramics have comparable strength to existing supports,” he adds. “The combination of our R&D printers and our demonstration plant allows us to produce multi-tonn quantities whilst maintaining flexibility of materials and designs, he adds.
Hydrotreating is an important petroleum refining process in which sulfur and other contaminants are removed in the manufacture of transportation…
In a project of the New Energy and Industrial Technology Development Organization (NEDO; Kawasaki, Japan; www.nedo.go.jp), a collaboration of six…
Using a simple, ligand-free palladium catalyst system, researchers at McGill University (Montreal, Que., Canada; www.mcgill.ca) have demonstrated that metal carbonylation…
Researchers led by professor Hideo Honoso at Tokyo Institute of Technology (Yokohama City, Japan; www.titech.ac.jp) have discovered a new catalyst…
The research group of professor Masafumi Hirando at Tokyo University of Agriculture and Technology (TUAT; Koganei City, Japan; www.tuat.ac.jp) has…