AmericanChemicalSociety.com

The discovery of the exceptional NO decomposition activity of Cu-ZSM-5 in 1986 by Iwamoto set the stage for the continued research into the deNOx capabilities of Cu-Zeolites. Cu-ZSM-5 has shown superior conversion and selectivity in direct NO decomposition, hydrocarbon assisted Selective Catalytic Reduction (SCR), and NH₃-SCR. Considerable advancements have been made in the determination of how this system is superior to other copper-exchanged zeolites, in the understanding of the active sites in Cu-ZSM-5  and the mechanism of NO decomposition. The main drawback of Cu-ZSM-5, however, is its lack of hydrothermal stability.  This is due to dealumination of the zeolite framework, where at these elevated temperatures the aluminum in the framework becomes unstable and detaches from the tetrahedral framework positions. Copper migration also occurs resulting in the formation of copper oxide and copper aluminate clusters.  Along with these problems, ZSM-5 is susceptible to adsorbing hydrocarbons at low temperatures that can generate heat as the temperature is raised, damaging to the zeolite structure. These results obviously have a negative impact on the further development of Cu-ZSM-5 for automobile exhaust abatement, since water is ever present in internal combustion engine exhaust streams and catalytic converters are exposed to high temperatures during the exhaust cycle. 

Recently, other copper zeolite systems have been discovered as SCR catalysts with improved activity and selectivity in the decomposition of NO. Others have reported data showing the enhanced performance of metal exchanged small-pored zeolites and Cu-Chabazite (Cu-CHA) in the NO decomposition of exhaust gas streams.  These catalysts have higher SCR activity at low temperatures and improved hydrothermal stability over existing Cu-zeolites.  Reaction experiments with Cu-CHA show NO conversion to be 90-100% over a temperature range of 250-450°C, and even after hydrothermally aging the zeolite at 800°C, conversion is still in excess of 80%. We used Rietveld refinements to show the increased thermal stability of copper exchanged small-pore zeolites, and to establish the coordination of copper ions within the frameworks of Cu-SSZ-13 (CHA) and Cu-SSZ-16 (AFX).

In this ACS-PRF grant we first investigated the selective catalytic reduction (SCR) activity along with the hydrothermal stability of these copper exchanged small-pore zeolites. The NH₃-SCR activity and hydrothermal stability of copper-exchanged small-pore zeolites has been investigated for SSZ-13, SSZ-16, SAPO-34, Sigma-1, and Nu-3.  The results show the excellent performance of Cu-SSZ-13 and Cu-SSZ-16 in the decomposition of nitrogen oxides.  These materials were also shown to be more hydrothermally stable than the medium-pore Cu-ZSM-5 and maintain their high SCR activity even after being steamed for periods of up to 15 hours.  The degree of copper-exchange is an important factor in achieving high hydrothermal stability, as we showed that samples of Cu-SSZ-13 and Cu-SSZ-16 with higher copper loadings were much less stable than the samples with a “intermediate” copper loading.  Decreasing the copper loading and dimensionality of the framework reduces the SCR activity of the zeolite, by limiting the number of and accessibility to active sites within the zeolite.  Heteroatom substitution of Ga for Al in SSZ-13 decreases the stability of the zeolite and leads to the formation of extra-framework gallium and loss of Bronsted acid sites.  Isostructural to SSZ-13, Cu-SAPO-34 was also investigated as a catalyst for NH₃-SCR.  The activity of Cu-SAPO-34 was similar to that of Cu-SSZ-13, and its high SCR activity was maintained even after steaming.  Within the limits of these experiments, the performances of Cu-SSZ-13 and Cu-SAPO-34 are seen to be superior to those of the other zeolites studied for the NH₃-SCR of nitric oxide.  A manuscript summarizing these results has been submitted to Applied Catalysis B: Environmental and we expect it will be published shortly.

In the second part of this investigation, we focused on determining the structure of the active site in Cu-SSZ-13 catalysts using a combination of catalytic, in-situ and ex-situ UV/vis spectroscopy, X-ray absorption spectroscopy (EXAFS and XANES) and diffraction analysis. These studies were carried out in collaboration with Prof. Andrew Beale at Utrech University. The advantage of this zeolite structure is that it has only one site for coordination of the Cu ions in contrast to most of the other zeolite frameworks where there are a variety of locations where ions can bind, each with a distinct coordination environment.  Consequently, EXAFS and XANES can be analyzed with the confidence that the spectra represent the structure of the Cu site that is relevant in enhancing catalytic activity.  The analysis of the EXAFS data find that Cu is coordinated to three oxygen atoms above the plane of the 6-membered ring of the structure of chabazite (SSZ-13).  This site is quite similar to the site observed by X-ray diffraction on dehydrated samples of SSZ-13 conducted by our group prior to this ACS-PRF grant.  UV/vis spectra of the catalysts before and after reaction (ex-situ) and also obtained in-situ at reaction conditions (200°C in a stream containing ammonia, NO and O₂) show that the oxidation state of the Cu ions does not change during the reaction.  This study demonstrates, for the first time, the structure and electronic properties of the catalytic site responsible for this important reaction.  Catalytic investigattions of NO decomposition (in the absence of ammonia) showed a striking difference between the SSZ-13 zeolite and the better known Cu-ZSM-5 catalyst.  Cu-SSZ-13 is essentially inactive for NO decomposition, in contrast to Cu-ZSM-5 which is quite active.  Importantly, it has been shown prior to our research that Cu-ZSM-5 zeolites have µ-Cu-O-Cu dimers in the zeolite pores, in addition to isolated Cu(II) ions in other parts of the zeolite framework.  Our results clearly show that the isolate Cu ions are active in the ammonia SCR reaction, but are inactive for NO decomposition.  By default, we conclude that the µ-Cu-O-Cu dimers (bis-µ-oxo-dicopper units) are in fact the sites responsible for the NO decomposition activity. A report summarizing these results has been accepted for publication in Chemical Communications.  We expect this report to be published in the near future.