Influence of Local Environments in Pores of Different Size on the Catalytic Liquid-Phase Oxidation of d-Glucose by Au Nanoparticles Supported on Nanoporous Carbon

Milena Perovic,Jan Philipp Hofmann,Nadezda V. Tarakina,Martin Oschatz

doi:10.1021/acsanm.0c01299

Abstract

Inorganic nanoparticles (NPs) dispersed on the surface of porous support materials play a dominant role in heterogeneous catalysis. Especially in liquid-phase catalysis, support materials may lead to nanoconfinement of solvent molecules, and the solute and can thus influence their local concentration and structure. This further affects the surface coverage of nanoparticles with reactants and other species, but these influences remain poorly understood—in particular under experimental conditions. Nanoconfinement effects are of particular importance in reactions such as liquid-phase oxidation with oxygen, i.e., when one reactant is a gas that has to be dissolved before reaching the NP surface. The significant influence of the pore structure of carbon materials on the catalytic activity of gold nanoparticles (AuNPs) with nearly similar size (4.1–4.7 nm) is demonstrated in this study. Experimental results on the oxidation of d-glucose with molecular oxygen in aqueous solution show that the “apparent catalytic activity” of AuNPs is a function of the carbon pore size and geometry. The architecture of the carbon pore size is determining the local concentration of reactants. Nanoconfinement of water in carbon nanopores can lead to enhanced solubility of reactants and therefore to their higher local concentration in proximity to the catalytically active sites. In contrast to purely microporous carbon support with the less wettable internal surface without any detectable catalytic activity, AuNPs supported on mesoporous carbons show a much higher metal time yield between 3.8 and 60.6 molGlc min–1 molAu–1, depending on volume and geometry of the mesopores.

Highlights

The working principles of catalysis on the surface of nanoparticles (NPs) in a given reaction are determined by the adsorption state and adsorption ratios of all different species present in the distance of a few nanometers, that is, in direct contact with the local active surface.[1−4] The intrinsic adsorption properties of NPs of given chemical composition mainly depend on the particle size and the strength of interaction with the potentially used porous support material.[5]
To point out that the catalytic activity of the gold nanoparticles of similar size and on support with similar chemical structure must be the same, we propose here to introduce the term “apparent catalytic activity” to make clear that the change in the speed of D-glucose oxidation does not arise from differences in the intrinsic properties of the AuNPs but is rather caused by changes of their local environment caused by differences in the textural properties of the carbon support materials
It has been shown that the pore architecture of carbon materials can have a crucial influence on the apparent catalytic activity of supported AuNPs

Summary

■ INTRODUCTION

The working principles of catalysis on the surface of nanoparticles (NPs) in a given reaction are determined by the adsorption state and adsorption ratios of all different species present in the distance of a few nanometers, that is, in direct contact with the local active surface.[1−4] The intrinsic adsorption properties of NPs of given chemical composition (and with that their catalytic properties) mainly depend on the particle size and the strength of interaction with the potentially used porous support material.[5]. A proof of concept is reported here by using the oxidation of D-glucose (Glc) to gluconic acid in water with molecular oxygen as a model reaction.[19−24] The catalysts used are AuNPs supported on porous carbon materials of different pore size and geometry.[25−27] the AuNP size, AuNP loadings, and oxidation state as well as the chemical properties of the support are comparable in all cases, the Au−C catalysts show significant differences in catalytic activity, depending on pore sizes and geometries of the carbon materials The influence of these parameters is clearly reflected by the metal time yield (MTY), which ranges from no detectable activity for a microporous salt-templated carbon (STC) support to 3.9 molGlc min−1 molAu−1 for the least active mesoporous catalyst and up to 60.6 molGlc min−1 molAu−1 for the most active ordered mesoporous catalyst. For 1H NMR spectra, the correlation of the signals was done by the multiplicities. 2D spectra were recorded by heteronuclear single-quantum correlation (HSQC) as well as heteronuclear multiple-bond correlation (HMBC) spectroscopy.[30]

■ RESULTS AND DISCUSSION

■ CONCLUSION

■ ACKNOWLEDGMENTS

■ REFERENCES