The intimate relationship between bulk electronic conductivity and selectivity in the catalytic oxidation of n-butane.

The efficient and direct functionalization of alkanes from natural gas or future regenerative carbon-based resources is impeded by the lack of suitable catalysts and the absence of a detailed mechanistic understanding of the few efficient alkane oxidation reactions. The selective oxidation of nbutane to maleic anhydride (MA), an important basic chemical with an annual global production of 1.4 Mt, is one of such scarce commercialized examples. The industrially used vanadium-phosphorous-oxide (VPO) catalyst enabling a maximum MA yield of 65% mainly consists of (VO)2P2O7 (vanadyl pyrophosphate, VPP). Based on experimental evidence it is generally agreed that the reaction proceeds via a two-step mechanism, in which, in Step 1: “lattice” oxygen from the catalyst (or oxygen from an active surface site) is abstracted to oxidize the alkane, and in Step 2: the catalyst is subsequently reoxidized by gas-phase O2. It is however still debated, whether the reaction proceeds on noninteracting single surface sites with the bulk being only an inert support, or if the bulk supplies charge carriers and oxygen. 11–15] The single-site concept would demand spacious active sites to provide the large number of 14 electrons and seven oxygen atoms needed per converted n-butane molecule. In contrast, an unlimited bulk–surface charge and oxygen transfer contradicts the fundamental site-isolation principle of selective oxidation catalysis, which presumes that the (stoichiometric) limitation and spatial isolation of active oxygen prevents the further oxidation of the desired product to COx. Clearly, the investigation of charge-carrier dynamics in selective catalysts is of fundamental importance to disentangle the surface and bulk influence on the catalytic performance. Unfortunately, unstable (Schottky) contact resistances between catalyst particles, electrodes, and at grain boundaries hamper quantitative and sensitive electrical-conductivity investigations by DC or AC contact methods. Although such studies have provided valuable information on the electrical properties of VPO catalysts, the direct participation of bulk charge carriers in the catalytic reaction has not been demonstrated unequivocally yet. Herein, the disadvantages of contact methods could be circumvented by using a noncontact conductivity method based on the microwave cavity perturbation technique (MCPT). MCPT is a highly sensitive technique 21] allowing the non-invasive quantitative measurement of the permittivity and electrical conductivity of polycrystalline samples in a fixed-bed flow-through reactor. The excitation of free charge carriers in the investigated sample by microwaves (at 9.2 GHz) in a calibrated resonant cavity enables the determination of absolute conductivity values. By measuring the change of the resonance frequency and the quality factor of the cavity with and without the sample, its complex permittivity e = e1 + ie2 can be deduced (Figure 1). [20–22] The imaginary part e2 is composed of the dielectric loss, ionic, and electronic conductivity. A major contribution of ionic charge carriers (e.g. O ions) is negligible because their high masses are not able to follow the high-frequency microwave excitation. In addition, no response of the real permittivity e1 of VPO to variations of the gas-phase chemical potential was observed, hence a significant influence of dielectric relaxations (through dipoles) can be excluded. Consequently, the major contribution to e2 in VPO is electronic conductivity. Moreover, pre-investigations showed that the catalyst behaves as a p-type semiconductor with an increasing conductivity in oxidizing and a decreasing conductivity in reducing mixtures (Figure S2 in the Supporting Information). These results are in agreement with contact conductivity investigations. According to X-ray powder diffractometry, the VPO catalyst was phase pure and consisted of vanadyl pyrophosphate (Supporting Information, Figure S1). The catalytic performance and conductivity of VPP were probed simultaneously with the in situ MCPT/GC setup at constant temperature, but in different gas mixtures and at different gas hourly space velocities (GHSV), i.e. at different reaction gas contact times. The catalyst was preheated in air to 400 8C (GHSV: [*] Dr. M. Eichelbaum, Dr. M. H vecker, C. Heine, Dr. A. Trunschke, Prof. Dr. R. Schlçgl Department of Inorganic Chemistry Fritz-Haber-Institut der Max-Planck-Gesellschaft Faradayweg 4–6, 14195 Berlin (Germany) Fax: (49)30-84134401 E-mail: me@fhi-berlin.mpg.de