(389d) Electrocatalytic Biomass Valorization: Efficient Strategies for Understanding Surface Coverage Effects

Electrocatalytic hydrogenation of biologically-derived platform intermediates is an emerging strategy to synthesize commodity and specialty chemicals for a range of applications. Successful implementation of such technologies will revolutionize sustainable chemistry; however, a fundamental understanding of the complex processes at electrocatalyst surfaces is lacking. One key aspect governing reactivity is surface coverage by reactive intermediates, which has been shown to significantly impact reaction selectivities and activities.^1-2 The ability to quantify coverage experimentally remains limited due to the complexity of in situ methods, while the computational burden for explicitly evaluating reaction energetics at varying coverages is immense. Numerous efforts have been made to predict interactions between small molecules on metal surfaces through interaction counting and cluster expansion methods.^3-4 While these methods are accurate (0.001 – 0.10 eV) for small molecules, extension to multifunctional oxygenates is nontrivial.

Here, we present a computationally efficient scheme for calculating interaction energies between long-chain oxygenates involved in the electrocatalytic hydrogenation of cis,cis-muconic acid (ccMA), which we have recently found to bind strongly to transition metal surfaces. Our model leverages an advanced interaction-counting approach for parameterizing interactions between various functionalities, (e.g., H–O, O–H, H–H). By applying machine learning techniques (gaussian process regression), we accurately predict adsorbate interactions to within 0.05 eV. Model parameters are sensitive to binding motifs of these adsorbates at a catalyst surface, offering a spatially-resolved and efficient strategy to predict coverage effects on reaction energies.

References:

(1) Chadderdon, X. H. et al. Green Chem. 21, 6210-6219 (2019).

(2) Patel, D. M. et al. Green Chem. Advance Article (2024).

(3) Patel, D. M. et al. J. Phys. Chem C 127, 21085-21096 (2023).

(4) Schmidt, D. J. et al. J. Chem. Theory Comput. 8, 264–273 (2011).