(126b) Exploration of the Redox Nature of Cu in Cu/SSZ-13 Catalysts for Selective Catalytic Reduction of NOx by NH3

We further investigate the proposed Cu redox cycle from our recent publication¹, which correlates Cu oxidation state and coordination environment in X-ray absorption spectroscopy (XAS) to density functional theory (DFT) which shows Cu may exist in a mixed 2-coordinate (Cu(I)) and 4-coordinate (Cu(II)) stateunder steady state selective catalytic reduction (SCR) conditions.  To probe this redox cycle, a series of experiments utilizing operando XAS were designed to isolate the oxidizing or reducing SCR half-reactions on Cu/SSZ-13 at 200˚C.  When oxygen was cut off from the standard SCR mixture, forcing the catalyst to perform only the reducing side of the reaction, the catalyst reduced to nearly 75% Cu(I).  The time for Cu to reach this reduced state occurred within 5 minutes of the standard SCR rate diminishing to zero. Similar experiments were performed by cutting off NO from standard SCR and NO₂ from fast SCR.  In the NO₂ cutoff from fast SCR, the catalyst reduced to mixed Cu oxidation states with NO, NH₃, and O₂ still present, corresponding to standard SCR as the oxidizing power of NO₂ was removed.    During NO cutoff, the catalyst was driven to a fully oxidized state because of the oxidizing ability of O₂ in the absence of NO.  Each cutoff included a reverse experiment to bring the Cu oxidation state and the reaction rate back to initial values.  The times for Cu to reach a steady oxidation state and the gas concentrations to become stable were similar except for the NO cutoff and addition, where Cu took longer to reach a steady oxidation state. This was attributed to the mildly reducing nature of NH₃and the initial state of the catalyst upon reintroduction of NO to the system.

We also investigated the effect of NO₂ on the oxidation state of Cu by varying the NO₂ to NO_x ratio from 1:10 up to 1:1.  We observed the amount of Cu(I) present to disappear completely at a 2:5 ratio, while at a 1:10 ratio, about half the Cu(I) of standard SCR was present.  Increasing the amount of NO₂ in the gas caused the reaction conditions to become non-differential, at which point, a small square-shaped x-ray beam was used to collect XAS at the top and the bottom of the bed.  The spectra showed a substantial difference in the dynamic balance of redox sites.  At the top of the bed, NO₂ was more abundant, so the Cu remained mostly oxidized.  At the bottom of the bed, 40-80% of the NO₂ had been reacted, so Cu was seen to be approaching a state similar to that resembling standard SCR, where both Cu(I) and Cu(II) were present.  This is an example of why it is so important to operate under differential conditions when performing operando experiments.

Thus, we confirmed the redox nature of the Cu site in these experiments by eliminating the capability of the redox cycle to close and driving Cu oxidation to a reduced or oxidized state.  The effect of NO₂ on Cu oxidation state also highlighted the sensitivity of this system to initial gas conditions.

1.  J.-S. McEwen, T. Anggara, W.F. Schneider, V.F. Kispersky, J.T. Miller, W.N. Delgass, F.H. Ribeiro, Catalysis Today  (2012), 184, 129– 144, 2012.