Identifying the Optimal Pd Ensemble Size in Dilute PdAu Alloy Nanomaterials for Benzaldehyde Hydrogenation

Abstract

Unraveling metal nuclearity effects is central for active site identification and the development of high-performance heterogeneous catalysts. Herein, a platform of nanostructured palladium (Pd) in gold (Au) dilute alloy nanoparticles supported on raspberry-colloid-templated (RCT) silica was employed to systematically assess the impact of the Pd ensemble size for the low-nuclearity regime in the Au surface layer, from single atoms to clusters, on the catalytic performance in the liquid-phase hydrogenation of benzaldehyde to benzyl alcohol. Combining catalyst evaluation, detailed characterization, and mechanistic studies based on density functional theory, we show that Pd single atoms in the Au surface plane (corresponding to samples with 4 atom % Pd in Au) are virtually inactive in this reaction and benzyl alcohol production is optimal over small Pd clusters (corresponding to samples with 10-12 atom % Pd in Au) due to superior benzaldehyde adsorption and transition state stabilization for the C-H bond formation step. For larger Pd ensembles (samples with ≥10 atom % Pd in Au), C-O bond hydrogenolysis occurs, promoting toluene formation and decreasing the selectivity toward benzyl alcohol, in line with a relatively lowered C-O bond cleavage barrier. Nevertheless, the nanostructured bimetallic Pd13Au87/SiO2-RCT catalyst still outperforms monometallic Pd counterparts in terms of selectivity for benzyl alcohol over toluene at comparable conversion and rate. Furthermore, the stability is improved compared to pure Pd nanoparticles due to inhibited particle agglomeration in the RCT silica matrix.