Analysis of O-2 Adsorption Stability and Strength Over Gold Clusters Using DFT and Logistic Regression

In this work, the stability and strength of O-2 adsorption over Au2-10 clusters were studied. The density functional theory (DFT) computed adsorption data were classified using multiple logistic regression. The effects of user defined descriptors were analyzed and it was found that stable O-2 adsorption requires the presence of an unpaired electron while its strength depends on the size and the charge of the cluster. As the size of the cluster increases, the probability of strong adsorption decreases, and the odds of finding strong adsorption is higher for the anionic clusters compared to neutral and cationic clusters. The effects of the electronic properties were also studied and HOMO-LUMO gap was found to be the most significant property determining the stability of O-2 adsorption; as its value increases, the probability of stable adsorption decreases. The strength of the adsorption, on the other hand, was found to be mostly dependent on the ionization potential, which has a negative effect.