Reactions of Group V Metal Oxide Cluster Ions: Insight Into Heterogeneous Catalytic Mechanisms

Open Access
Justes, Dina Rodal
Graduate Program:
Doctor of Philosophy
Document Type:
Date of Defense:
March 04, 2004
Committee Members:
  • Albert Welford Castleman Jr., Committee Chair
  • Dr A D Jones, Committee Member
  • James Bernhard Anderson, Committee Member
  • Robert John Santoro, Committee Member
  • ion molecule reactions
  • catalytic mechanisms
  • mass spectrometry
  • gas phase
  • clusters
  • quadrupole
Experiments were conducted on the reactions between vanadium and niobium oxide cluster cations and various hydrocarbon gases in an effort to obtain information on the active sites on catalytic surfaces. The data acquired may be of use to determine elementary steps in a larger catalytic mechanism. This provides information on the sites necessary for the promotion of the desired process as well as on the sites which can either prohibit the desired process or encourage the generation of unwanted side products. The guided ion beam mass spectrometer used in these studies allows for the individual analysis of clusters to determine the conditions required for a desired process. These results are part of an on-going study to systematically investigate the Group V metal oxide cluster ions toward hydrocarbons of interest to the catalytic community. The experiments discussed herein include a variety of reactant gases from saturated hydrocarbons to aromatic molecules. These studies provide a sampling of the power of this technique to evaluate different systems. These will be briefly summarized below. The reactions between the Group V metal oxides with the C2 hydrocarbons, ethane and ethylene, conducted previously have shown a dramatic preference of the (V<SUB>2</SUB>O<SUB5</SUB>)<SUB>1-3</SUB><SUP>+</SUP> clusters to transfer an oxygen atom to the hydrocarbon. We were unable, however, to determine the neutral product generated by experimental means. Therefore, we formed a collaboration with a theoretical group in Berlin, to ascertain the most probable product energetically. Density functional calculations showed the structures of the two reactive clusters, V<SUB>2</SUB>O<SUB>5</SUB><SUP>+</SUP> and V<SUB>4</SUB>O<SUB>10</SUB><SUP>+</SUP>, both exhibited an oxygen centered radical moiety as part of their structure which was not observed with any other cluster studied. It was also determined that acetaldehyde was the preferred product when (V<SUB>2</SUB>O<SUB>5</SUB>)<SUB>1,2</SUB><SUP>+</SUP> reacted with ethylene. The calculations also showed that the reactions with ethane showed that it was energetically possible for V<SUB>2</SUB>O<SUB>5</SUB><SUP>+</SUP> to dehydrogenate ethane to yield ethylene and to proceed further to eventually give acetaldehyde as a neutral product. A second study discussed here involves the reactions between the vanadium and niobium oxide cluster cations with methanol. Methanol is of interest industrially as an alternate fuel source. Hence, any information one may obtain to improve the selectivity and overall efficiency is of vital interest. Certain clusters demonstrated the association of molecular hydrogen. This was determined using deuterated methanol. These products were also observed under single collision conditions with the only possible mechanism being dehydrogenation of methanol to yield H<SUB>2</SUB>, adsorbed to the cluster, and CH<SUB>2</SUB>O, the neutral product. In addition, the association of C<SUB>2</SUB>H<SUB>6</SUB>O was also observed. Among the possible structures of C<SUB>2</SUB>H<SUB>6</SUB>O is dimethyl ether. This is not the only structure though and additional experiments are required to ascertain the exact structure on the cluster. The reactions between vanadium and niobium oxide cluster cations and propylene were also investigated. The polyolefin industry is a considerable part of the economy itself as well as producing precursors to several other substances. The oxidation of propylene is also a useful process, generating products such as propylene oxide. Considering the closeness in structure from propylene to ethylene, there was hope that the Group V metal oxide clusters would also show a significant oxygen transfer pathway for the propylene. The reactions between the vanadium and niobium oxide cluster cations mainly consisted of the formation of the monomer and dimer and even went as high as the pentamer. The addition of methyl and ethyl groups to the monomer and dimer were also observed. The oxygen transfer from the metal oxide cluster to the propylene was also noted for V<SUB>2</SUB>O<SUB>5</SUB><SUP>+</SUP> and Nb<SUB>2</SUB>O<SUB>5</SUB><SUP>+</SUP> but not for V<SUB>4</SUB>O<SUB>10</SUB><SUP>+</SUP>. The reactivity was greater for the V<SUB>2</SUB>O<SUB>5</SUB><SUP>+</SUP> than for the Nb<SUB>2</SUB>O<SUB>5</SUB><SUP>+</SUP>, however this is to be expected since the V-O bond is more easily cleaved than the Nb-O bond. The last study included here involves the reactions between the vanadium and niobium oxide cluster cations with ortho-xylene. Xylene is of interest in the production of phthalic anhydride which is used to produce polyester resins and polyvinylchloride. Xylene is also worth studying to determine if the methyl groups affect the reactivity with the metal oxide cluster ions. The V<SUB>2</SUB>O<SUB>5</SUB><SUP>+</SUP> cluster was the only species to exhibit an oxygen transfer pathway. The majority of the products observed were charged hydrocarbons. Among these were the production of C<SUB>7</SUB>H<SUB>7</SUB><SUP>+</SUP>, the loss of a methyl group, C<SUB>8</SUB>H<SUB>10</SUB><SUP>+</SUP>, charge transfer, C<SUB>8</SUB>H<SUB>10</SUB>CH<SUB>3</SUB><SUP>+</SUP>, addition of a methyl group to the charged xylene molecule, and (C<SUB>8</SUB>H<SUB>10</SUB>)<SUB>2</SUB>CH<SUB>3</SUB><SUP>+</SUP> and (C<SUB>8</SUB>H<SUB>10</SUB>CH<SUB>3</SUB>)<SUB>2</SUB><SUP>+</SUP>, single and double methyl addition to the xylene dimer, respectively. Select clusters also exhibited the association of the C<SUB>7</SUB>H<SUB>7</SUB> and C<SUB>8</SUB>H<SUB>10</SUB> species.