Machine-Learning Methods Enable Exhaustive Searches for Active Bimetallic Facets and Reveal Active Site Motifs for CO₂ Reduction

Abstract

Bimetallic catalysts are promising for the most difficult thermal and electrochemical reactions but modeling the many diverse active sites on polycrystalline samples is an open challenge. We present a general framework for addressing this complexity in a systematic and predictive fashion. Active sites for every stable low-index facet of a 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 sites. The activity of these sites is explored in parallel using a neural-network based surrogate model to share information between the many Density Functional Theory (DFT) relaxations, resulting in activity estimates with an order of magnitude fewer explicit DFT calculations. Sites with interesting activity were found and provide targets for follow-up calculations. This process was applied to the electrochemical reduction of CO 2 on nickel gallium bimetallics and indicated that most facets had similar activity to Ni surfaces, but a few exposed Ni sites with a very favorable on-top CO configuration. This motif emerged naturally from the predictive modeling and represents a class of intermetallic CO 2 reduction catalysts. These sites rationalize recent experimental reports of nickel gallium activity and why previous materials screens missed this exciting material. Most importantly these methods suggest that bimetallic catalysts will be discovered by studying facet reactivity and diversity of active sites more systematically.

Dynamic Changes in the Structure, Chemical State and Catalytic Selectivity of Cu Nanocubes during CO2 Electroreduction: Size and Support Effects

“…[1] Among these parameters,t he catalyst structure and chemical state are of particular importance. [7,8,16,21,[30][31][32][33][34][35][36] However,t he presence of (100) facets is not the only factor responsible for the superior activity and selectivity of cubeshaped Cu catalysts,w ith surface roughness,s ubsurface oxygen and Cu I species or Cu/Cu I interfaces formed and/or stabilized under reaction conditions also playing av ery important role. [20][21][22][23][24][25][26][27][28][29] Previous studies [9][10][11] on Cu single crystals have shown the improved C À Cc oupling performance of (100) facets,w hich was further confirmed by the high selectivity towards ethylene observed on cube-shaped Cu catalysts.…”

Application of Artificial Neural Networks for Catalysis: A Review

“…Their screening results are shown in Figure 7. As a further step for the CO 2 electroreduction catalyst search, Ulissi and Nørskov et al developed a framework that combined ANN potential energy fitting and the binding energy prediction for bimetallic alloy design [84]. They concluded that a machine learning technique can enable an exhaustive screening on both bimetallic facets and active atomic ensembles for a given catalysis.…”

“…With these machine learning-assisted methods, it has been shown that some state-of-the-art data mining techniques (e.g., ANNs) are able to perform ultra-fast and precise predictions of catalytic descriptors based on a sufficiently large DFT-calculated database. The research done by Ulissi and Nørskov et al [84] can be a good milestone that illustrates how DFT can work with machine learning and perform a quick and exhaustive catalysts screening and optimization. workers successfully developed a "volcano activity plot" method to estimate the theoretical activities of monometallic heterogeneous catalysts [71][72][73].…”

“…Based on a DFT-calculated database with proper independent variables as the inputs, ANN is able to "learn" the highly-complicated intrinsic properties via a non-linear fitting process. Several successful studies such as the research done by Ulissi and Nørskov et al [84] have shown that machine learning can be a good choice to reduce the computational cost of theoretical catalysis study, and meanwhile, provides precise and ultra-fast screening for new catalyst discovery. (4) To theoretically study the PES for a catalytic system and perform global optimizations, ANN has shown its capacity for mapping the PES for a specific reaction and/or a specific type of catalyst, combined with the MC methods.…”

Tuning the Products of CO2Electroreduction on a Ni3Ga Catalyst Using Carbon Solid Supports

“…XPS of bulk Ni 3 Ga indicates that the material's surface is comprised of a nearly exact 3:1 ratio of Ni and Ga, which has been reported as a relatively poor CO 2 -reducing stoichiometry. 26,41,42 On the other hand, the surfaces of Ni 3 Ga thin films synthesized on HOPG and glassy carbon are both Ga-rich, despite the fact that the bulk compositions of these films exhibit their namesake 3:1 Ni:Ga stoichiometry ( Figure 1). Considering that heterogeneous CO 2 reduction occurs at the electrode's surface, we suggest that the different surface stoichiometries achieved on bulk versus thin film Ni 3 Ga are instrumental in determining the materials' reactivities toward CO 2 in solution.…”