Effective Coordination Based Descriptors for Rational Design of Metal Nanocatalysts

Effective Coordination based Descriptors for Rational Design of Metal Nanocatalysts

Xianfeng Ma, Siwen Wang and Hongliang Xin*

Department of Chemical Engineering, Virginia Tech,

Blacksburg, VA, 24061

hxin@vt.edu

Session: Catalysis and Reaction Engineering Division ¨C Rational Catalyst Design

One major challenge for designing metal nanocatalysts is to know a priori the appropriate surface structure for a given catalytic reaction. Understanding the effects of particle morphology and composition on the interaction of surface atoms with adsorbates is of pivotal importance for identifying optimal active sites and engineering nanoparticles with maximized fraction of such sites. Many efforts have been made aiming to predict the binding energy of an adsorbate at metal surfaces using the electronic and/or geometric factors of the adsorption site (termed descriptor).^2,3 While these models have been successfully used as reactivity descriptors for either extended-metal surfaces or simple metal nanoparticles, their extension to complex particle systems with varying broken-bond strains and/or metal ligands remains elusive due to a formidable computational cost or the lack of an explicit consideration of interatomic interactions.

In this work, we present the effective coordination number (CN_e) as an interaction-aware reactivity descriptor for metal nanocatalysts⁴. The CN_e, quantified by interatomic coupling matrix elements between the site of interest and its all neighbors within a certain cutoff radius, provides a robust description of CO, O₂, and O adsorption energies on Au metal nanoparticles of varying sizes and shapes attributed to its explicit consideration of broken-bond strains (see Fig. 1 for CO as an example). Importantly, the CN_e has a solid physiochemical basis via a direct connection to the moment characteristics (e.g., center) of occupied density of states projected onto valence orbitals of the adsorption site. Furthermore, the CN_e shows promise as a general descriptor for predicting adsorption properties of core-shell and alloyed structures of Au nanoparticles with d¹⁰ metal ligands (e.g., Cu and Ag). The approach can be readily extended to understand and predict reactivity trends of large and more complex metal catalysts with defects, impurities, transition-metal additions, supports, etc, and thus provides a general basis for rational design of metal nanocatalysts.

FIG. 1. Adsorption energies of *CO atop described by CN_e on various surface sites of Au nanoparticles and extended surfaces. The statistics of linear regression (solid line) are also given.

References:

1. Calle-Vallejo, F. et al. Finding optimal surface sites on heterogeneous catalysts by counting nearest neighbors. Science 350, 185¨C189 (2015).

2. Hammer, B. & Nørskov, J. K. Electronic factors determining the reactivity of metal surfaces. Surf. Sci. 343, 211¨C220 (1995).

3. Calle-Vallejo, F. et al. Fast Prediction of Adsorption Properties for Platinum Nanocatalysts with Generalized Coordination Numbers. Angew. Chem. Int. Ed. 53, 8316¨C8319 (2014).

4. Ma, X., Wang, S. & Xin, H. Effective Coordination Number as a Reactivity Descriptor for Coinage Metal Nanocatalysts. Submitted (2016).