(748c) Propene Ammoxidation over Doped Bismuth Molybdates

Since their discovery in the 1950s by SOHIO, bismuth molybdate based catalysts have been widely used to selectively catalyze a variety of important partial oxidation and ammoxidation reactions of light olefins. The most industrially relevant of the reactions catalyzed by a bismuth molybdate derivative is that of propene, ammonia and oxygen to acrylonitrile, which is performed on the scale of 5 billion kg/year. Acrylonitrile is a main component of a number of commodity products, including acrylic fiber and polymers such as ABS, NBR, and SAN. Pure bismuth molybdenum oxides will catalyze this reaction with about 60% selectivity to acrylonitrile, with the major byproducts of acetonitrile and hydrogen cyanide also having significant industrial applications. The patented industrial catalysts for this reaction are multiphase, multicomponent mixed metal oxides that contain on the order of ten elements, including both bismuth and molybdenum, and have improved yields of acrylonitrile.[1] In this work, we investigate individual dopants that can integrate into alpha bismuth molybdate to produce a single-phase tertiary mixed metal oxide in order to understand how each element influences the  catalytic performance for propene ammoxidation.

There is a significant body of research devoted to elucidating the role of various dopants that substitute for bismuth, including lead and cerium, or for molybdenum, including vanadium, tungsten, and iron.[2] However, most of that research has focused on testing doped bismuth molybdates for the oxidation reaction of propene to acrolein. This simpler, two reactant oxidation reaction has been used as a probe because it has the same rate determining step and a related mechanism as the ammoxidation of propene to acrylonitrile, but unfortunately it does not allow for an evaluation of the interaction with ammonia.[3] During our recent investigation of the kinetics of propene ammoxidation to acrylonitrile over pure alpha bismuth molybdate, we demonstrated via XPS and EDS that nitrogen groups insert into the surface of the catalyst. This process requires at least a day of ammonia exposure at reaction temperatures, however once it is complete, the resultant catalyst has an approximately 1.5 times higher activity for propene oxidation to acrolein, and it never returns to the lower activity it had before ammonia exposure. The effect of ammonia treatment is permanent because the hydrocarbon intermediate does not pull nitrogen out of the catalyst to produce acrylonitrile as been previously hypothesized, but instead reacts with a gas phase ammonia during ammoxidation. However, this gas phase ammonia can also react with itself to form molecular nitrogen in an undesirable side reaction. These results highlight the complex behavior of ammonia that must be accounted for when evaluating potential catalysts for this reaction.  In this work, we explain how a dopant affects the interactions among the metal oxide, propene and ammonia in order to guide the design of improved catalysts for the ammoxidation of propene to acrylonitrile.

[1] Grasselli, R.K. Topics in Catalysis 2002, 21, 79-88.

[2] Getsoian, A.B.; Zhai, Z.; Bell, A.T. JACS 2014, 136, 13684-13697.

[3] Adams, C.R.; Jennings, T.J. J. Catal. 1963, 2, 63-68.