Combining computational and in situ spectroscopies joint with molecular modeling for determination of reaction intermediates of deNO x process—CuZSM-5 catalyst case study Piotr Pietrzyk a, * , Barbara Gil a , Zbigniew Sojka a,b, ** a Faculty of Chemistry, Jagiellonian University, ul. Ingardena 3, 30-060 Krakow, Poland b Regional Laboratory of Physicochemical Analyses and Structural Research, Jagiellonian University, ul. Ingardena 3, 30-060 Krakow, Poland Available online 13 November 2006 Abstract Interaction of NO with Cu I ZSM-5 catalysts was spectroscopically investigated in static (IR and EPR) and flow (IR) regimes, complemented by DFT quantum chemical calculations. Particular attention was paid to the elucidation of the N–N bond formation mechanism, which (along with O– O bond making) is one of the key issues of a deNO x reaction. The active sites ({Cu I }ZSM-5, {CuO}ZSM-5), the intermediates ({Cu I NO}ZSM-5}, {Cu I N 2 O 2 }ZSM-5}), and the low-temperature spectators (up to 423 K) ({Cu I (NO) 2 }ZSM-5}) appearing during this process were identified. Their assignment and molecular structure was ascertained by joint use of computational spectroscopy and DFT modeling. A new method of discrimination between the conformers of surface dinitrosyls based on the calculations of the relative IR intensities of the symmetric and antisymmetric vibrations was proposed, and the mechanistic importance of copper dinitrosyl conformation was discussed. The inner-sphere versus outer-sphere mechanistic dichotomy of the N–N bond conception was rationalized in terms of the spin density repartition within the Cu–NO moiety of the mononitrosyl intermediate. # 2006 Elsevier B.V. All rights reserved. Keywords: NO dimer; Dinitrosyl; Hyponitrate; IR intensity; CuZSM-5; deNO x mechanism; IR; EPR; DFT 1. Introduction Despite of being thermodynamically favored, catalytic abatement of NO x still remains a challenging problem from both fundamental and practical point of view [1–4]. A number of important mechanistic issues concerning the nature of the active sites and the key intermediates, as well as the molecular level description of the reaction elementary steps have not been definitely resolved yet. In this context wide scientific interest in the CuZSM-5 zeolite stems from its remarkable activity in direct NO decomposition, and the advantage of exhibiting a relatively simple structure. This makes CuZSM-5 a convenient model system for basic mechanistic studies and quantum chemical calculations [5–8]. DFT modeling of heterogeneous catalytic systems is, however, computationally quite demand- ing. Judicious selection of an appropriate cluster model and an adequate calculation scheme is of great importance for obtaining sensible results. In our method of exploring the mechanism of catalytic reactions we combine molecular modeling with computational spectroscopy. Spectroscopic parameters of postulated active sites, reaction intermediates and spectators are calculated and directly compared with available experimental data (EPR magnetic parameters, IR frequencies and intensities). Such approach provides a quantitative bridge between the spectroscopic fingerprints of the investigated species and their molecular structure and reactivity [9]. The backbone of a deNO x process over CuZSM-5 zeolites can be epitomized in the form of two interconnected cycles associated with the formation of N 2 and O 2 . The principal mechanistic steps include the formation of a N–N bond, where NO reactant is transformed into N 2 O intermediary product, and simultaneously the primary {Cu I }Z active sites are converted www.elsevier.com/locate/cattod Catalysis Today 126 (2007) 103–111 * Corresponding author. Tel.: +48 12 663 22 24; fax: +48 12 634 05 15. ** Corresponding author at: Faculty of Chemistry, Jagiellonian University, ul. Ingardena 3, 30-060 Krakow, Poland. Tel.: +48 12 663 22 95; fax: +48 12 634 05 15. E-mail addresses: pietrzyk@chemia.uj.edu.pl (P. Pietrzyk), sojka@chemia.uj.edu.pl (Z. Sojka). 0920-5861/$ – see front matter # 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.cattod.2006.09.033