Highly selective immobilized bimetallic Ni-Au nanoparticle catalyst for the partial hydrogenation of m-dinitrobenzene

Transition metal nanoparticles (NPs) are extensively used as catalysts for a wide and diverse range of organic transformations when immobilized on appropriate solid supports. We describe the development of a highly active and highly selective heterogeneous catalyst based on Ni-Au NPs supported on activated carbon fibers (ACFs) for the partial reduction of m-dinitrobenzene (m-DNB) to m-nitroaniline (m-NAN), an important platform chemical used in the synthesis of dyes and polymers. Initially, Ni NPs with narrow size distribution and ranging from 2 to 14 nm were prepared with poly-N-vinyl-2-pyrrolidone (PVP) as a stabilizer. Evaluation of the NPs as catalysts in the liquid-phase hydrogenation of m-dinitrobenzene led to the establishment of an antipathetic structure sensitivity, i.e. the larger NPs displayed a 6-fold higher turnover frequency than the smaller NPs. The selectivity to the target m-NAN product is independent of the size of the Ni NPs, possibly due to preferential PVP absorption of the NP edges and vertices. Consequently, Ni NPs of 2 nm were supported on ACFs and residual PVP was removed by a ultra-violet ozone (UVO) treatment, rendering a highly selective structured catalyst that affords m-NAN in almost 96% yield. A two-site (plane vs. edge Ni-atoms) Langmuir-Hinshelwood kinetic model is consistent with the experimental kinetic data confirming that low-coordination atoms (edges and vertices) are responsible for selective reaction. Consequently, we prepared bimetallic Ni-Au NPs (Ni:Au = 1:1) aiming to generate Ni surface sites mimicking the properties of edge and vertex atoms. The resulting UVO-treated Ni-Au NPs of 3 nm immobilized on ACFs afford m-NAN with a yield exceeding 98%. Such a high yield appears to be unprecedented and shows how careful nanocatalyst design, guided by detailed structural characterization and mechanistic studied, can lead to highly selective catalysts of industrial relevance