The study of the behavior of palladium, and noble metals in general, in respect of fundamental gas-metal interactions and reactive chemistry, has attracted a great deal of attention over a great many years. [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15] The intrinsic importance of a range of commercial applications of palladium-not least of which being automotive exhaust catalysis-coupled to a fundamental desire to understand the basic physics and chemistry of such interactions, are the prime drivers for this interest. In the exhaust catalysis application, the catalyst has to operate in conditions wherein the redox potential of the feedstock can change in a highly dynamic manner. Further, the interplay between oxidized and reduced phases of palladium is of central importance to palladium activity in methane oxidation catalysis. [8][9][10] As such, a comprehension of the dynamic redox behavior of Pd and other noble metals is of considerable importance.Recently, we demonstrated the presence of a highly dynamic size/shape variation occurring during CO/NO cycling over 1 wt % Pd/10 % (Ce,Zr)O x /Al 2 O 3 catalysts. [16] Part of this phenomenon comprises a non-oxidative, palladium-redispersion mechanism. Herein we contrast the behavior of chloride-free 2 wt % Pd/Al 2 O 3 catalysts at 673 K during redox cycling using CO as the reductant, and in the absence (NO only) and presence (5 O 2 :1 NO) of O 2 during the oxidizing part of the cycle. We show that, even in the presence of a relatively large amount of oxygen, oxidation cannot compete kinetically with the redispersion events. On the contrary, significant levels of O 2 in the oxidizing feed greatly enhance the degree of palladium redispersion possible before any formal oxidation of Pd 0 to Pd 2+ can be observed. Figure 1 shows the variation in observed Pd-Pd coordination number (CN) derived from analysis of energy-dispersive EXAFS (EDE) data (EXAFS = extended X-ray absorption fine structure spectroscopy; see also Supporting Information) in the two cases studied: the first (Figure 1, open circles) is obtained during cycling of a 2 wt % Pd/Al 2 O 3 sample between flows of 5 % CO/He and 5 % NO/He at 673 K. The second (Figure 1, filled circles) shows results obtained when the total flow in the oxidizing cycle now comprises a mix of 5 O 2 :1 NO. Figure 2 shows the result of translating this raw CN data into an average number of Pd atoms per particle. This translation is done within the framework proposed by Jentys [17] and, as such, a constant particle morphology and 0 % Pd oxidation is assumed.The overall behavior during CO/NO cycling over a 2 wt % Pd/Al 2 O 3 sample is as observed previously.[16] The presence of the CeZr phase is therefore not a prerequisite and these transformations appear intrinsic to Pd nanoparticles. It further appears that, even though there are considerable error