Carbon support effects on Pd/C catalysts for the liquid-phase hydrogenation of multifunctional chemicals
Climate change due to anthropogenic greenhouse gas emissions is a major concern for nations worldwide. In order to mitigate the challenges associated with this issue, it is imperative for industrial sectors to switch to renewable feedstocks and energy sources for decreasing CO2 emissions. Lignocellulosic biomass is one such alternative that demonstrates a promising future, specifically for the sustainable production of chemicals. However, the transition from a petroleum-based to a bio-based chemical industry poses new challenges for catalysis. In contrast to petrochemical reactions that are typically conducted in the gas phase at high temperature, biomass conversion reactions often require low temperature aqueous-phase conditions to mitigate the undesired side reactions and degradation of the feedstock molecules. Metal oxide supports that are commonly used in the petroleum industry (silica, alumina) lose their structural integrity in aqueous environments and, therefore, alternative supports and catalysts need to be developed. Carbon materials represent a suitable alternative to metal oxides as carbon-carbon bonds withstand hydrolytic attacks, under both highly acidic and basic aqueous environments, even in the presence of salts, which makes them a support material of choice for a broad range of biomass conversion reactions. However, these promising materials remain poorly understood and the structure-activity relationships required to design high-performance catalysts are still missing. The present work addresses this gap in knowledge. Specifically, we built this work on the hypothesis that structure-activity correlations can be established for carbon-supported metal catalysts by decreasing the support complexity and controlling the support properties at the nanoscale. We pursued this hypothesis and were able to deconvolve the contributions of the scaffold’s surface chemistry and electronic properties to the catalytic activity of the supported metal active phase.
In order to conduct a systematic study of support effects, we developed a synthetic platform based on stacked-cup carbon nanotubes (SCCNTs) as scaffolds. The key advantage of these SCCNTs is their dual structure consisting of a graphitic core and an amorphous carbon shell, which offers a handle on the graphitic character and surface chemistry through simple thermochemical treatments. Pd metal nanoparticles deposited on the SCCNTs were carefully characterized using advanced methods—aberration corrected transmission electron microscopy and synchrotron-based X-ray photoelectron spectroscopy—and the performance of the synthesized catalyst series was evaluated for the hydrogenation of cinnamaldehyde, an α,β-unsaturated aldehyde probe molecule. Strong correlations between the activity and selectivity of Pd/SCCNTs for the liquid phase hydrogenation reaction and the structure of the support were observed. Advanced characterization revealed that the observed trends could be assigned to electronic metal-support interaction (EMSI), resulting in a charge transfer and the formation of an electron-depleted Pdδ+ phase at the metal-carbon interface.
Once the presence of EMSI for carbon-supported catalysts was established, we attempted to quantitatively determine the contribution of the Pdδ+ phase to the overall performance of the Pd/C catalyst. Thermal annealing of the samples enabled incremental changes in the Pd particle size and Pd-C contacts. The analysis of the corresponding catalytic results revealed major differences in the selectivity and intrinsic rate of Pd0 and Pdδ+ metal atoms, and demonstrated that controlling the structure of the carbon surface offers a powerful handle for tuning the activity of Pd/C catalysts.
As the MSI effects observed in this work were electronic in nature, the catalysts were further investigated by ultraviolet photoelectron spectroscopy in an attempt to explain the origin of the charge transfer and associate it with the scaffold’s electronic properties. Linear correlations between the work function and band gap of SCCNTs and carbon nitride materials and the selectivity towards C=C bond hydrogenation were established. These results are expected to facilitate the development of strategies for rationally designing carbon supports for target biomass conversion reactions.
In summary, we developed strategies to control the properties of carbon supports at the nanoscale, we demonstrated the existence of electronic metal-support interactions for carbon-supported precious metal catalysts, and we established structure-activity correlations that may guide the rational design of next-generation hydrogenation catalysts.