Nitric oxide, chemisorption

R.A. Marbrow, and R.M. Lambert, Chemisorption and surface reactivity of nitric oxide on clean and sodium-dosed Ag(110), Surf. Sci. 61, 317-328 (1976). 

A Japanese group reacted iron vapor with nitric oxide at 77 K (6). Two different species were observed, with Pno at 1800 and 1720 cm , that were assigned as NO species adsorbed on oxidized and metallic iron, respectively. Although no evidence was presented as to the nu-clearity of the products, the authors considered the species to be models for the chemisorption of NO on iron surfaces. 

The catalytic activity of ln/H-ZSM-5 for the selective reduction of nitric oxide (NO) with methane was improved by the addition of Pt and Ir which catalyzed NO oxidation, even in the presence of water vapor. It was also found that the precious metal, particularly Ir loaded in/H-ZSM-5 gave a low reaction order with respect to NO, and then showed a high catalytic activity for the reduction of NO at low concentrations, if compared with ln/H-ZSM-5. The latter effect of the precious metal is attributed to the enhancement of the chemisorption of NO and also to the increase in the amount of NO2 adsorbed on in sites. 

Similar studies6 were reported for the chemisorption of nitric oxide at Rh(lll) with at 10 8 Torr a (2x2)-3NO structure, previously reported by LEED with a different structure with a (3 x 3) periodicity observed above 10 2 Torr (Figure 7.5). The latter has an apparent height of 0.1 nm above the (2 x 2) structure, suggesting that a distortion has occurred of the top surface layer of... 

Figure 7.5 STM images of the chemisorption of nitric oxide at Rh(l 11) at a pressure of 0.03 Torr at 298 K showing the phase transition between (2 x 2) and a (3 x 3) structure, (a) t = 0 s (b) t — I I0 s the phase boundary has now moved and is half-way across the image. (Reproduced from Ref. 6).

Freund s group at the Fritz Flaber Institute have put much emphasis on linking surface science studies with applied catalysts through replicating the latter with model systems without having to resort to the complexity of the real system. A system they have studied in detail is that of nitric oxide chemisorption at a palladium-alumina model catalyst, where they isolated different... 

Of crucial significance in deciding between various models have been estimates of the number of copper atoms required to transform the surface into a (2 x 3)N phase. This was the approach adopted by Takehiro et al 2 in their study of NO dissociation at Cu(110). They concluded that by determining the stoichiometry of the (2 x 3)N phase that there is good evidence for a pseudo-(100) model, where a Cu(ll0) row penetrates into the surface layer per three [ll0]Cu surface rows. It is the formation of the five-coordinated N atoms that drives the reconstruction. The authors are of the view that their observations are inconsistent with the added-row model. The structure of the (2 x 3)N phase produced by implantation of nitrogen atoms appears to be identical with that formed by the dissociative chemisorption of nitric oxide. 

In agreement with the TPR results, the hydrogen chemisorption/pulse reoxidation data provided in Table 8.3 indicate that, indeed, the extents of reduction for the air calcined samples are -20% higher upon standard reduction at 350°C (compare 02 uptake values). Yet in spite of the higher extent of reduction, the H2 desorption amounts, which probe the active site densities (assume H Co = 1 1), indicate that the activated nitric oxide calcined samples have higher site densities on a per gram of catalyst basis. This is due to the much smaller crystallite that is formed. The estimated diameters of the activated air calcined samples are between 27 and 40 nm, while the H2-reduced nitric oxide calcined catalysts result in clusters between 10 and 20 nm, as measured by chemisorption/pulse reoxidation. 

Results of Hydrogen Chemisorption/Pulse Reoxidation Measurements over Activated Silica-Supported Cobalt Catalysts Calcined at 350°C Using either Flowing Air or 5% Nitric Oxide in Nitrogen... 

The results confirm that the novel metal nitrate conversion method using nitric oxide in place of air advocated by Sietsma et al. in patent applications WO 2008029177 and WO 2007071899 leads to, after activation in H2, catalysts with smaller cobalt crystallites, as measured by EXAFS and hydrogen chemisorption/ pulse reoxidation. In spite of the lower extent of cobalt reduction for H2-activated nitric oxide calcined catalysts, which was recorded by TPR, XANES, EXAFS,... 

Clearly the molecular events with iron were complex even at 80 K and low NO pressure, and in order to unravel details we chose to study NO adsorption on copper (42), a metal known to be considerably less reactive in chemisorption than iron. It was anticipated, by analogy with carbon monoxide, that nitric oxide would be molecularly adsorbed on copper at 80 K. This, however, was shown to be incorrect (43), and by contrast it was established that the molecule not only dissociated at 80 K, but NjO was generated catalytically within the adlayer. On warming the adlayer formed at 80 K to 295 K, the surface consisted entirely of chemisorbed oxygen with no evidence for nitrogen adatoms. It was the absence of nitrogen adatoms [with their characteristic N(ls) value] at both 80 and 295 K that misled us (43) initially to suggest that adsorption was entirely molecular at 80 K. 

The divalent catalyst is highly coordinatively unsaturated and therefore exhibits some unusual chemistry (33-42). It has a light green color but quickly truns blue when exposed to N2, indicating a weak chemisorption. Carbon monoxide is adsorbed to yield a violet color, and of course it poisons the polymerization. Up to two molecules can be adsorbed. Olefins also adsorb in a 2 1 ratio, and acetylene is converted to benzene. Polar compounds like alcohols, ethers, or amines are strongly held. Nitric oxide (NO) attaches in a 2 1 ratio. 

Nitric oxide is rapidly desorbed on evacuation. The number of adsorption sites for this reaction was found to increase from 1.5/100 A2 after pretreatment at 25°C to only 2.3/100 A2 after pretreatment at 800°C for afi-Al203 (Degussa). These values are about an order of magnitude higher than those obtained by pyridine adsorption on the same type of alumina (121) and by NH3 adsorption on a y-Al203 (168). The number of a sites as determined by C02 adsorption by Peri (157) is still lower. The chemisorption of N02, therefore, seems to be less specific than that of pyridine, NH3, and C02. The weak absorptions at wave numbers above 2000 cm-1 were tentatively assigned by Parkyns to N02+ species. These may be comparable to the species found in zeolites by Naccache and Ben Taarit (242). 

TABLE XI. Carbon Monoxide, Di-Nitrogen and Nitric Oxide Chemisorption on Metals... 

The influence of the support is undoubted and spillover was further confirmed by the excess of hydrogen chemisorbed by a mechanical mixture of unsupported alloy and TJ-A1203 above that calculated from the known values for the separate components. It was also observed that the chemisorption was slower on the supported than on the unsupported metal and that the greater part of the adsorbate was held reversibly no comment could be made on the possible mediation by traces of water. On the other hand, spillover from platinum-rhenium onto alumina appears to be inhibited for ratios Re/(Pt Re) > 0.6. In an infrared investigation of isocyanate complexes formed between nitric oxide and carbon monoxide, on the surface of rhodium-titania and rhodium-silica catalysts, it seems that the number of complexes exceeded the number of rhodium surface atoms.The supports have a pronounced effect on the location of the isocyanate bond and on the stability of the complexes, with some suggestion of spillover. 

For all catalysts the reaction is first order with respect to NO. Oxygen strongly inhibits the decomposition reaction. The inhibition by oxygen is ascribed to the chemisorption of oxygen on sites on which nitric oxide chemisorption takes place. 

Several catalyst samples of tungsten carbide and W,Mo mixed carbides with different Mo/W atom ratios, have been prepared to test their ability to remove carbon monoxide, nitric oxide and propane from a synthetic exhaust gas simulating automobile emissions. Surface characterization of the catalysts has been performed by X-ray photoelectron spectroscopy (XPS) and selective chemisorption of hydrogen and carbon monoxide. Tungsten carbide exhibits good activity for CO and NO conversion, compared to a standard three-way catalyst based on Pt and Rh. However, this W carbide is ineffective in the oxidation of propane. The Mo,W mixed carbides are markedly different having only a very low activity. 

Transition metal carbides (mainly of W and Mo) have been shown to be effective catalysts in some chemical reactions that are usually catalyzed by noble metals such as Pt and Pd (ref.1). Their remarkable physical properties added to lower cost and better availability could make them good candidates for substitute materials to noble metals in automobile exhaust catalysis. Hence, for this purpose, we have prepared several catalysts of tungsten carbide and W,Mo mixed carbides supported on y alumina with different Mo/W atom ratios. The surface composition has been studied by XPS while the quantitative determination of catalytic sites has been obtained by selective chemisorption of hydrogen and of carbon monoxide. The catalytic performances of these catalysts have been evaluated in the simultaneous conversion of carbon monoxide, nitric oxide and propane from a synthetic exhaust gas. 

---