James F. Brazdil

Redox kinetics of bismuth molybdate ammoxidation catalysts

Abstract The response of several bismuth molybdate-based catalysts to reduction under propylene ammoxidation conditions in the absence of gaseous oxygen, and to reoxidation by gaseous oxygen, was studied using a pulse microreactor method. The catalysts investigated were Bi2Mo3O12, Bi2Mo2O9, Bi2MoO6, Bi3FeMo2O12, and a multicomponent system (Ma2+Mb3+BixMoyOz). The unit area rates of lattice oxygen participation at 430 °C decrease in the order: multicomponent system > Bi2Mo2O9 > Bi2Mo3O12 > Bi3FeMo2O12 ≳ Bi2MoO6. Maximum selective utilization of reactive lattice oxygen occurs after partial reduction for the multicomponent system, Bi2Mo3O12 and Bi2Mo2O9. These results are consistent with a mechanism requiring coordinately unsaturated metal ions in complex shear domains for selective ammoxidation. Conversely, Bi3FeMo2O12 and Bi2MoO6 show maximum lattice oxygen activity at their highest oxidation states. The overall reoxidation rates of partially reduced catalysts at 430 °C decrease in the order: Bi2MoO6 > Bi2Mo2O9 > Bi2Mo3O12 > Bi3FeMo2O12 ≳ multicomponent system. These reoxidation rates of partially reduced catalysts are first order in oxygen vacancy concentration and half order in gaseous oxygen. A general mechanism for catalyst reoxidation is proposed based on these kinetics. This series of catalysts exhibits two reoxidation regimes. One regime is characterized by a low activation energy at low degrees of initial reduction and involves the reoxidation of surface vacancies. A second regime is observed for deeper degrees of reduction which is characterized by a higher activation energy and involves the reoxidation of anion vacancies in the bulk of the catalyst. The observed activation energies for the reoxidation of the catalyst bulk are strongly dependent upon the structure and composition of the catalyst.

An investigation of the role of bismuth and defect cation vacancies in selective oxidation and ammoxidation catalysis

The roles of bismuth and cation vacancy defects in the selective ammoxidation of propylene to acrylonitrile were investigated. Based on kinetic and spectroscopic studies of several scheelite molybdate catalysts, it is concluded that bismuth, and not cation vacancies, is responsible for the formation of the allylic intermediate from propylene. Cation vacancies generate molybdenyl-type functionalities in the vicinity of the defects which are responsible for the insertion of oxygen or nitrogen into the allylic intermediate in the selective oxidation or ammoxidation of propylene, respectively.

Mechanistic features of selective oxidation and ammoxidation catalysis

Selectivity in the oxidation of olefins over heterogeneous catalysts, an important requirement for efficient use of feed resources, has been achieved by the use of catalysts which produce an allylic intermediate via dissociative chemisorption. The development of catalysts for ammoxidation and oxidation of propylene has resulted in substantial increases in the supply of important monomers such as acrylonitrile, resulting in the discovery of new applications in the fibres and plastics industry.The best known and most studied ammoxidation and oxidation catalysts are those based on bismuth molybdate, which operate by a redox mechanism composed of a catalyst reduction cycle (olefin oxidation–selective product formation), and a catalyst reoxidation cycle (lattice-oxygen regeneration). Relative rates of reduction for a number of bismuth molybdate systems: multi-component > Bi2Mo2O9 > Bi2Mo3O12 > Bi3FeMo2O12 > Bi2MoO6, and the requirement for partial reduction to attain maximum selectivity are consistent with a mechanism in which coordinately unsaturated metal ions in complex shear domains are responsible for selective oxidation. The reoxidation process, which follows the order of decreasing rates: Bi2MoO6 > Bi2Mo2O9 > Bi2Mo3O12 > Bi3FeMo2O12≳ multicomponent system, is composed of a low-activation-energy surface reoxidation regime and a higher-activation-energy reoxidation regime involving bulk anion vacancies. The latter is strongly dependent on the structure of the catalyst. Substituent effects and allyl radical in situ studies have shown that selective oxidation occurs via rate-determining α-hydrogen-atom abstraction by oxygen associated with Bi centres to produce a radical-like π allyl molybdenum complex. For oxidation, subsequent reversible formation of a σ-O allyl molybdate (which can be formed in situ from allyl alcohol) is followed by a second hydrogen abstraction. This rate-determining step in the conversion of the σ species to acrolein is enhanced by the presence of bismuth in the catalyst. In ammoxidation, the analogous σ-N allyl molybdenum species is formed reversibly from the π complex after conversion of terminal MoO sites to MoNH via condensation with ammonia. The σ-N complex undergoes a second hydrogen abstraction to produce a reduced molybdenum 3-iminopropene complex, in a process of higher activation energy than the formation of acrolein from the corresponding σ-O complex. Reoxidation and a subsequent third hydrogen abstraction forms acrylonitrile and a reduced site. The reduced site is then reoxidized by lattice oxygen to complete the catalytic cycle.From this detailed work, the necessary components for a selective and active oxidation catalyst are identified as a bifunctional site, composed of an active α-hydrogen abstraction (Bi) and a selective, O or NH inserting (Mo) species, and favourable solid-state and chemical properties which enhance the rapid donation of lattice oxygen and reconstitution of reduced sites by molecular oxygen.

Relationship between solid state structure and catalytic activity of rare earth and bismuth-containing molybdate ammoxidation catalysts

Abstract The formation of solid solutions in rare earth and bismuth-containing molybdate catalysts plays a key role in the selective ammoxidation of propylene to acrylonitrile. Solid state structural studies of the Bi 2 − x Ce x (MoO 4 ) 3 system reveal that bismuth dissolves readily in the Ce 2 (MoO 4 ) 3 lattice which crystallizes in the low-temperature La 2 (MoO 4 ) 3 structure, yielding a single-phase material when x ≥ 1. The solubility of cerium in the Bi 2 (MoO 4 ) 3 structure is less extensive, with maximum solubility occurring at x − 0.2. Structural substitution of bismuth or cerium by rare earth cations such as La, Pr, Nd, and Y results in lattice parameter changes which indicate that bismuth and cerium cations randomly occupy equivalent positions in BiCe-molybdate solid solutions. Catalytic activity maxima correlate well with phase compositions and occur in the two single phase regions, where there is maximum solubility of bismuth in the cerium molybdate phase and maximum solubility of cerium in the bismuth molybdate phase. A third catalytic maximum is observed in the binary phase region of the two saturated solid solutions and coincides with equal solubility of Ce in the Bi-molybdate phase and Bi in the Ce-molybdate phase. At these three optimum catalyst compositions, maximum interactions exist between the three key catalytic components: Bi (α-H-abstracting element), Ce (oxygen and electron transfer element), and Mo (olefin chemisorption and nitrogen insertion element).

Mechanism for propylene oxidation to acrolein on Bi2Mo3O12: A quantum chemical study

The selective oxidation of propylene to acrolein on α-bismuth molybdate has been studied using molecular orbital theory and a surface cluster model. The overall reaction is H2CCHCH3 + O2 → H2CCHCHO + H2O. We find propylene can first coordinate to surface MoVI through a π-donation bond, and α-hydrogen abstraction proceeds to a bismuth oxygen anion with a low barrier yielding an allyl-MoV complex. The transition state involves a CHO2− σ-donation stabilization and the activation is a result of the filled CHO2− σ∗ antibonding counterpart orbital being stabilized though mixing with the π∗ orbital on the olefinic end. Allyl then forms a σ bond to a neighboring surface oxygen atom bonded to molybdenum, and the catalytic site is reduced by a second electron. It is postulated that these electrons may flow through the Mo 4d conduction band to a remote site active for O2 reduction. The σ-allyl rotates easily so that either end carbon may receive oxygen. The second dehydrogenation step is accompanied by a second two-electron reduction of the surface. It is postulated that this step may be activated by the concomitant reoxidation of the bismuth site as water desorbs. After acrolein desorption, the oxygen vacancy is postulated to migrate to the site of earlier oxygen reduction.

New materials synthesis: Characterization of some metal-doped antimony oxides

Abstract In order to understand the chemistry of altermetal dopants in antimony oxide, the detailed structural characterization of two β-Sb 2 O 4 compounds is reported, Mo-doped β-Sb 2 O 4 (1.5 metal%) and V-doped β-Sb 2 O 4 (5 metal%). The methods used to character...

Identification of active oxide ions in a bismuth molybdate selective oxidation catalyst

Direct structural identification of catalytically active oxide ions for selective olefin oxidation has been achieved using in situ Raman spectroscopy. Spectroscopic examination of Bi2MoO6 reduced with select probe molecules such as but-1-ene, propene, methanol and ammonia, when reoxidized with oxygen-18, establishes the existence of the multifunctional nature of catalyst active site for propene oxidation and ammoxidation, wherein α-H abstraction occurs by oxide ions bridging bismuth and molybdenum atoms (Bi—O—Mo), and oxygen or NH insertion into the π-allylic intermediate occurs at centres associated with molybdenum. Sites for O2 chemisorption, reduction and dissociation are likely to be associated with the directed lone pairs of electrons of the Bi—O—Bi species in the structure. Pulse kinetic experiments reveal that diffusion of the oxide ions through the bulk of the catalyst is the rate-limiting step in catalyst reoxidation. This process is rapid when an oxide ion of only one functional type, i.e.α-H abstraction or oxygen insertion, is removed. The process becomes less facile when oxide ions associated with both α-H abstraction and oxygen insertion are removed from the catalyst. The relative rates reflect the extent of active-site reconstruction achieved by diffusion of oxide ions from the bulk to the surface of the catalyst.

Phase cooperation in oxidation catalysis. Structural studies of the iron antimonate–antimony oxide system

The compositions FeSb2O6 and FeSb5O12 of the two-phase FeSbO4–α-Sb2O4 system, an active and selective catalyst for the oxidation and ammoxidation of propylene, have been structurally characterized by Rietveld analysis of powder neutron-diffraction data. Results of the analysis indicate that the presence of Sb2O4 has no effect on the bulk structural parameters of FeSbO4. Specifically (a) the unit cell of FeSbO4 does not depend upon the presence of Sb2O4 or calcination temperature, (b) antimony atoms are not found in the intersticies of the coexisting iron antimonate and (c) the apparent Sb/Fe ratio is 1 in iron antimonate. Additionally, the Sb/Fe occupancy in the rutile FeSbO4 structure is random as no supercell reflections were observed. Results of scanning electron microscope and X-ray photoelectron spectroscopy experiments have been interpreted to show that Sb enrichment occurs on coprepared samples of the two-phase mixture. Based on this evidence and the lack of alteration of the bulk structures of both phases it is suggested that surface alteration in this two-phase system is the key to enhanced selective catalytic oxidation activity.

Oxidative dimerization of methane over lead-magnesium mixed oxide catalysts

Abstract While both methane combustion and methane ammoxidation to HCN over metal catalysts are well known, selective dimerization of methane in the presence of oxygen over oxide catalysts has been achieved only recently. We report results of the oxidative coupling of methane to form ethane and ethylene over a mixed PbO-MgO catalyst. Selectivity for methane conversion to higher hydrocarbons rises to over 80% as reactor temperature is raised to about 800°C over these catalysts when oxygen/methane in the feed is limited. However, selectivity declines as oxygen partial pressure is increased. Both observations are consistent with recently published results for other catalysts containing PbO. Here, optimal reaction conditions give yields of nearly 20%. The PbO-MgO catalyst operates effectively both with and without co-fed oxygen. Productivity, an important consideration for practical processes, was found to be about forty times higher when oxygen is co-fed. The rate of oxygen disappearance over PbO-MgO had a 2 3 order dependence on oxygen pressure under conditions where selectivity was high, indicating that oxygen activation has an important role in the coupling mechanism.

The structure of ce-doped Bi2(MoO4)3 as determined by neutron profile refinement

Abstract The structure of Bi 1.8 Ce 0.2 (MoO 4 ) 3 has been refined with powder neutron diffraction data by the Rietveld method. The structure can be derived by severely distorting the scheelite structure ( AM O 4 ) and is perhaps better written A 2 3 O 1 3 M O 4 , where O = cation vacancy. Of the two bismuth atom sites, cerium preferentially occupies the more symmetric of the two (Bi(2) in the structure) with some cerium found in the scheelite subcell vacancies also. This site preference is understood by examining the symmetries of the two Bi sites. Crystal data: monoclinic, space group P2 1 c , Z = 4, a = 7.697(2), b = 11.535(3), c = 11.944(3) , β = 115.19.