The conventionally accepted reason that platinum alloys are more active for the O2 reduction reaction than pure platinum is that reaction products such as adsorbed oxygen atoms (Oad) and hydroxyls (OHad) are less strongly adsorbed on the alloy surface, allowing more O2 to adsorb [1]. This is a reasonable view, but experimental results obtained in this laboratory suggest a much different picture, one in which higher coverages of reaction products such as Oad and OHad on alloy surfaces are correlated with higher O2 reduction activity [2-4]. Specifically, with X-ray photoelectron spectroscopic measurements combined with electrochemical measurements on the same surface (EC-XPS), we find a buildup of Oad, with increasing amounts at lower potentials [2,3]. This result is just the opposite of that expected from the conventional model. Similarly, with quartz crystal microbalance/electrochemical measurements (EQCM), we find that there are elevated levels of oxygenated species on the alloy surface during active O2 reduction, even though the mass resolution is not sufficiently high to distinguish between Oad, OHad and H2Oad. The Brønsted-Evans-Polanyi (BEP) principle, which states that the activation energies of a set of similar reactions are proportional to their enthalpies [5], has become well accepted, even though it is an empirical relationship. This principle implies that the adsorption energy of two Oad atoms should be large compared to that of the undissociated O2 molecule, specifically, in an electrochemical potential region in which the overall reaction is energetically favored. Furthermore, it implies that a platinum skin/platinum alloy surface such as Pt skin/Pt3Co(111) should have a larger overall negative change in energy for the O2dissociation step than that for pure Pt(111). This simplest of all models seems consistent with our experimental results. However, most theoretical calculations have found a weakening of the adsorption strengths of all of the oxygenated species, i.e., O2, Oad and OHadon Pt skin/Pt alloy surfaces. Thus, this situation presents a highly intriguing challenge for theoreticians. Of course, it can be conceived that reaction products might be so tightly bound to the surface that the desorption process might become rate-limiting. However, our experimental results are inconsistent with this idea. Finally, we note that both our EC-XPS and EQCM results clearly show that the coverages of oxygenated species such as Oad and OHad are much different for the oxygenated vs. deoxygenated solutions in the potential region in which O2 reduction is proceeding at significant rates. This is especially true for the Pt skin/Pt3Co alloy surface. This is because a high rate of O2reduction inevitably implies a high yield of dissociated products. In our presentation, based on recent DFT calculations, we will suggest a reasonable solution to the dilemma described above. We will also comment on the need to take into account orbital energies, as well as the overall Fermi energy, in order to fully understand the energetics of key reaction such as O2 dissociation. Acknowledgement: The authors gratefully acknowledge support from the “Research on Nanotechnology for High Performance Fuel Cells” project of NEDO, Japan. References Y. Xu, A. V. Ruban, and M. Mavrikakis, J. Amer. Chem. Soc., 126, 4717-4725 (2004).M. Wakisaka, M. Watanabe, and H. Uchida, Mechanism of an Enhanced Oxygen Reduction Reaction at Platinum-Based Electrocatalysts: Identification and Quantification of Oxygen Species Adsorbed on Electrodes by X-Ray Photoelectron Spectroscopy. In Fuel Cell Science: Theory, Fundamentals, and Biocatalysis, Editors, A. Wieckowski, J. K. Nørskov, pp. 147-168, John Wiley & Sons, Inc., Hoboken, NJ, USA (2010).M. Watanabe, D. A. Tryk, M. Wakisaka, H. Yano, H. Uchida, Electrochim. Acta, 84, 187-201 (2012).J. Omura, M. Wakisaka, D. A. Tryk, H. Uchida, and M. Watanabe, submitted.T. Bligaard, J. K. Nørskov, S. Dahl, J. Matthiesen, C. H. Christensen, J. Sehested, J. Catal. 224, 206-217 (2004).