ORR is one of the most studied electrochemical reactions due to its tremendous fundamental and practical importance. Oxygen is a common, readily accessible oxidizing agent and, therefore, the Pt ORR cathode is part of many energy conversion devices, e.g. fuel cells. Unfortunately, slow kinetics of the ORR negatively affects the performance and it is currently one of the main bottleneck in large scale fuel cells commercialization. It is partly caused by the presence of surface Pt oxides, which slow the reaction rate and trap reaction intermediates on the surface. The oxide formation and reduction is also known to cause dissolution of the Pt catalyst, which further degrades the performance.

Even though the electrochemical formation of surface oxides on platinum surface has been extensively studied in the past, there are still many questions unanswered, mainly about the detailed structure of the oxide and its growth mechanism [1 and references therein]. Most of the studies were performed in the absence of O₂, the fuel cell oxidant, and therefore they are less relevant to the fuel cell operation, as gaseous O₂ can modify the oxidation potentials and mechanism. Given the above, further fundamental understanding of the Pt oxidation mechanism and its atomistic picture is clearly needed in order to determine the role of surface oxides in ORR and its effect on the fuel cell performance.

Here we show the results of in-situ study of electrochemical oxide formation on Pt(111) and how it is influenced by presence of O₂ during ORR. Furthermore, we show that surface reorganization processes taking place during the oxidation/reduction cycle are governed by the ad-atom surface diffusion dynamics and closely resemble the dynamics found under the vacuum conditions.

The place exchange process associated with the initial stages of oxidation is followed dynamically during cyclic voltammetry (CV) and potential step experiments in the presence and absence of oxygen. Detailed analysis at two potentials shows that the reconstruction is consistent with a place exchange process between Pt and O atoms, in which the exchanged Pt atoms are directly above their original positions in the Pt(111) lattice. The reconstruction initiates with the CV peak at 1.06 V vs RHE, even though repeated cycling to 1.15 V leads to no changes in the CV. Adding O₂ to the electrolyte does not have any significant effect on the oxidation behavior, in contrast to some literature reports, and the O₂ accelerated Pt dissolution is not caused by the negative shift in the oxidation potential. Furthermore, the ORR current decreases before oxidation, implying that the presence of the surface oxide is not the limiting factor in the ORR and the high ORR overpotential is solely due to the slow ORR mechanism on an unreconstructed surface.

The surface restructuring upon electrochemical oxidation/reduction shows a characteristic ripening behavior where Pt islands grow and become more prominent and homogeneous in size with increasing number of cycles. Their characteristic lateral dimensions primarily depend on the upper potential limit of the cycle and only slightly increase with cycle number. The structural evolution of the Pt surface morphology strongly resembles that found in studies of Pt(111) homoepitaxial growth and ion erosion in ultrahigh vacuum. This finding shows that the electrolyte does not need to be included in the ab-initio investigations of the Pt surface structure change during ORR.