Nanocatalyst for refinery use

Dr. Neil Canter, Contributing Editor | TLT Tech Beat December 2010

Researchers have developed a nanocatalyst with tungsten oxide particles on zirconia that delivers good results. 

KEY CONCEPTS 
• Isomerization of n-pentane to isopentane is an important process needed to boost the octane rating of gasoline.
• A new nanocatalyst based on the placement of tungsten oxide nanoparticles on zirconia provides optimum performance at a density of 5 +/- 0.5 tungsten sites per square nanometer.
• The key catalyst driving the isomerization is a subnanometer zirconium-tungsten oxide cluster.

A good deal of attention has been written in this column on presenting information about new catalyst systems for reducing automotive emissions. The main technology used is a three-way catalyst based on palladium, platinum and rhodium that simultaneously removes carbon monoxide, hydrocarbons and No_x.

The size of the catalyst seems to matter when it comes to its activity. In a previous TLT article, research was described that showed there is a relationship between performance and the number of palladium atoms in a cluster (1). The researchers found that maximum performance was seen for an emission catalyst with clusters of 20 palladium atoms.

Catalysts play an important role in the manufacture of base oils and fuels. A key process used to boost the octane rating of gasoline is the isomerization of npentane to the branched isomer, isopentane.

Michael Wong, professor of chemical and biomolecular engineering and chemistry at Rice University in Houston, says, “Pentane isomerization is an acidic process that requires a strongly acidic catalyst to work. Chlorinated alumina and zeolites are used in oil refineries as the acid catalyst, with platinum and hydrogen added to help maintain catalyst lifetime performance. A zirconia-supported tungsten oxide type is a well-known candidate material for alkane isomerization, but it still is not used in commercial applications as far as I know.”

There is a lot of potential in using nanosized catalysts to improve the efficiency of commercial processes. These catalysts rely on the placement of nanoparticles on a support.

But controlling how the nanoparticles are situated on the support is difficult, according to Wong. This greatly affects the performance of the catalyst. Less is known about the placement of metal oxide nanoparticles on a different metal oxide support.

Wong and his research team initiated a project four years ago to find the optimum zirconia-supported tungsten oxide that can catalyze the isomerization of pentane. He says, “Two key discoveries that supported our approach were that small nanoparticles, which are one nanometer in diameter, were able to be situated on top of zirconia, and there was indirect proof that tungsten oxide particles can function as catalysts.

After many experiments, a new catalyst has been developed that provides good results.

ZIRCONIA-TUNGSTEN OXIDE CLUSTERS
The researchers studied an extensive number of catalysts with varying concentrations of tungsten oxide on zirconia and found a proper mix that afforded excellent results. Graduate student Nikos Soultanidis and lead author on a recent paper (2), says, “We prepared the catalysts by addition of an aqueous solution of ammonium metatungstate to amorphous zirconium oxyhydroxide. This was followed by drying at 70 C, crushing, sieving and then calcination at elevated temperature for three hours.”

After pelletizing, further crushing and sieving, catalysts with particle sizes between 300 and 600 microns were ready for evaluation in the pentane isomerization reaction.

The reaction with pentane occurred in the gas phase at an isothermal temperature under inert conditions in a quartz tube. The catalyst was pretreated with ultrahigh purity air for one hour prior to each experiment. Then 1% n-pentane in an argon/helium flow was passed over the catalyst.

Wong says, “The n-pentane used was diluted in helium and argon to ensure the highest possible quality of the analyzed data. N-pentane can contain trace amounts of pentenes, which can affect the rate of isomerization.”

The researchers deliberately developed a low conversion process in order to avoid side reactions. Wong adds, “We wanted to avoid coke formation, which is a killer. By running at a low concentration, we felt that n-pentane isomerization and the formation of side products such as propanes, butanes and hexanes could be readily detected.”

Catalysts prepared with different amounts of tungsten oxides on the surface of the zirconia were prepared. The tungsten oxide was quantified in units of tungsten sites per square nanometer.

Wong says, “We varied the concentration of tungsten sites from 0 up to 10 tungsten sites per square nanometer. Catalyst performance was found to form a bell shape curve over that range with a peak at a density of 5 +/- 0.5 tungsten sites per square nanometer. At higher levels, the tungsten oxide was present at too high a density and could not adequately catalyze the reaction. At lower densities, there was not enough tungsten oxide available on the surface of the zirconia to provide adequate conversion of n-pentane to isopentane.”

The surface of each of the catalysts was studied by using high-angle annual dark field, scanning transmission electron microscopy. Figure 2 shows an image of a surface displaying three of the catalyst species detected.


Figure 2. High-Angle Annular Dark Field, Scanning Transmission Electron Microscopy of the catalyst species detects montung-state species (circled in blue), polytung-state species (circled in green) and subnanometer zirconium-tungsten oxide clusters (circled in red). The latter species is believed to be responsible for catalyzing the isomerization of n-pentane. (Courtesy of Rice University)

Wong says, “We found three types of catalysts on the surface, which have been identified as monotungstate species (circled in blue), polytungstate species (circled in green) and subnanometer zirconium–tungsten oxide clusters (circled in red).”

The researchers found a direct correlation among the concentration of these species and the calcination temperature. When the catalyst was calcined at 500 C, a higher concentration of montungstate and polytungstate species were detected which prefer cracking of n-pentane instead of isomerization. This process leads to the formation of propane and isobutane.

At a higher calcination temperature of 700 C, the subnanometer zirconiumtungsten oxide cluster is the dominant species, and the researchers believe they are the catalysts responsible for facilitating isomerization.

Wong says, “The zirconium-tungsten oxide clusters can be thought of as nanocrystals that contain one or two zirconium atoms in each crystal. They are strongly acidic and possibly react with a hydroxyl group on the zirconia support to form an intermediate responsible for the catalysis.” These clusters have a diameter of one nanometer.

Wong speculates that the mechanism for the n-pentane isomerization probably occurs through a bimolecular pathway due to the detection of hexanes and other higher molecular weight hydrocarbons.

Future work involves trying to better understand the mechanism. This catalyst is now known, and Wong indicates it should be straightforward for industry to utilize it in a refinery.

With this development, the prospect grows that other nanocatalysts can be prepared that may be used to facilitate the preparation of lubricant base oils. Further information can be found in a recent publication (2) or by contacting Wong at mswong@rice.edu. 

REFERENCES
1. Canter, N. (2010), “Catalyst Effectiveness: Size Does Matter,” TLT, 66 (2), pp. 14–15.
2. Soultanidis, N., Zhou, W., Psarras, A., Gonzalez, A., Iliopoulou, E. , Kiely, C., Wachs, I. and Wong, M. (2010), “Relating n-Pentane Isomerization Activity to the Tungsten Surface Density of Wo_x/ZrO₂,” J. Am. Chem. Soc., 132, pp. 13462–13471.

Neil Canter heads his own consulting company, Chemical Solutions, in Willow Grove, Pa. Ideas for Tech Beat items can be sent to him at neilcanter@comcast.net.