Catalysts consisting of Au, Pd and their alloys have been shown to be active oxidation catalysts. These materials can use dioxygen or hydrogen peroxide as the oxidant with CO and activated organic molecules using O2(g) while more challenging cases, such as methane to partial oxygenates, relying on H2O2. Although H2O2 is a green oxidant, the incorporation of dioxygen greatly reduces overall cost and so there is an incentive to find new ways to reduce the reliance on H2O2. In this study we use DFT calculations to discuss the direct synthesis of H2O2 from H2(g) and O2(g) and use this understanding to identify the important surface species derived from dioxygen. We cover the adsorption of oxygen, hydrogen and water to model Au and Pd nanoclusters and the oxidation of the metals, since reduction of any oxides formed will consume H2. We then turn to the production of a surface hydroperoxy species; the first step in the synthesis of H2O2. This can occur via hydrogenation of O2(ads) with H2(ads) or via protonation of O2(ads) by solvent water. Both routes are found to be energetically reasonable, but the latter is likely to be favoured under experimental conditions.

We have investigated xNi-yCu-ZrO2 catalysts for the selective synthesis of.-valerolactone from levulinic acid (LA). A series of xNi-yCu-ZrO2 catalysts with a consistent metal loading of 50% but varying Ni and Cu composition were prepared by an oxalate gel precipitation method and tested for LA hydrogenation. Ni-rich catalysts showed higher catalytic activity compared with Cu-rich formulations with a 45Ni-5Cu-ZrO2 composition yielding 76% gamma-valerolactone after a reaction time of 30 min at 200 degrees C. Characterisation of the materials by XRD, surface area measurements and TPR allow us to attribute the differences in performance seen for different compositions to particle size and nanoparticle dispersion effects. DFT calculations also showed that a shift of d-band centre to higher energies with the mole fraction of Ni in Cu-Ni alloys would be expected to lead to improved hydrogen dissociation in Ni-rich catalysts and so aid hydrogenation activity.

A series of Cu–ZrO2 catalysts prepared by a co-precipitation method were studied for the hydrogenation of levulinic acid to give γ-valerolactone (GVL).

CO vibrational spectra over catalytic nanoparticles under high coverages/pressures are discussed from a DFT perspective. Hybrid B3LYP and PBE DFT calculations of CO chemisorbed over Pd4 and Pd13 nanoclusters, and a 1.1nm Pd38 nanoparticle, have been performed in order to simulate the corresponding coverage dependent infrared (IR) absorption spectra, and hence provide a quantitative foundation for the interpretation of experimental IR spectra of CO over Pd nanocatalysts. B3LYP simulated IR intensities are used to quantify site occupation numbers through comparison with experimental DRIFTS spectra, allowing an atomistic model of CO surface coverage to be created. DFT adsorption energetics for low CO coverage (θ→0) suggest the CO binding strength follows the order hollow>bridge>linear, even for dispersion-corrected functionals for sub-nanometre Pd nanoclusters. For a Pd38 nanoparticle, hollow and bridge-bound are energetically similar (hollow≈bridge>atop). It is well known that this ordering has not been found at the high coverages used experimentally, wherein atop CO has a much higher population than observed over Pd(111), confirmed by our DRIFTS spectra for Pd nanoparticles supported on a KIT-6 silica, and hence site populations were calculated through a comparison of DFT and spectroscopic data. At high CO coverage (θ=1), all three adsorbed CO species co-exist on Pd38, and their interdiffusion is thermally feasible at STP. Under such high surface coverages, DFT predicts that bridge-bound CO chains are thermodynamically stable and isoenergetic to an entirely hollow bound Pd/CO system. The Pd38 nanoparticle undergoes a linear (3.5%), isotropic expansion with increasing CO coverage, accompanied by 63 and 30cm−1 blue-shifts of hollow and linear bound CO respectively.

We present PBE + U calculations of gold metal nanoparticles supported on the α-Fe2O3(0001) surface. We find that the periphery atoms of Au10 particles become oxidized through the dissociation of O2 at the metal–oxide interface. The presence of a metal particle is also shown to substantially lower the defect formation energy for surface oxygen vacancies in the oxide support, particularly when the nanoparticle is in this semi-oxidized state. The defect formation energy is found to be dependent on the distance of the vacancy from the metal particle so that the lowest defect formation energies are calculated for oxygen vacancies created at the under the gold nanoparticle. For Au10/α-Fe2O3(0001) creating the vacancy under the cluster requires a defect formation energy of 2.13 eV [relative to ½O2(g)] and this is lowered to 0.86 eV when the perimeter of the particle is oxidized. These values are significantly lower than that for the bare α-Fe2O3(0001) surface of 3.04 eV. Oxidation of the periphery of the Au cluster to form an Au10O6 particle leads to even lower vacancy formation energies for the surface oxides and much lower than the energy required to abstract an O ion from the base of the cluster.