Coadsorption of NO and NH 3 on Cu(111): The Formation of the Stabilized (2 × 2) Coadlayer Tsuyoshi Sueyoshi, Takehiko Sasaki, and Yasuhiro Iwasawa* Department of Chemistry, Graduate School of Science, The UniVersity of Tokyo, Hongo, Bunkyo-ku, Tokyo 113, Japan ReceiVed: February 29, 1996; In Final Form: April 30, 1996 X The coadsorption of NO and NH 3 on Cu(111) was investigated by means of LEED, HREELS, and TPD. A new ordered (2 × 2) structure was found after annealing to 180 K NO-NH 3 coadlayers prepared from various relative exposures of NO and NH 3 at 100 K. A N 2 O desorption peak from the coadlayer appeared at 220 K, which was 70 K higher than the desorption peak from a pure NO layer. A new NH 3 desorption peak also appeared at 220 K, while there was only a broad peak centered at 150 K for a pure ammonia adlayer. These results suggest that the coadlayer is more stabilized than the pure layers owing to the attractive interaction between NO(a) and NH 3 (a). While NO in pure NO layers adsorbs on atop, bridge, and 3-fold hollow sites, it was found that NO in the (2 × 2) structure occupied only 3-fold hollow sites in a linear configuration and the frequency of ν(N-O) shifted downward from that in pure NO layers. This change in adsorption state of NO correlates with enhanced occupation of the antibonding 2π* orbital of NO due to charge transfer from NH 3 through the substrate, leading to more attractive interaction. A real space model for the (2 × 2) structure was proposed on the basis of the results of TPD and HREELS. 1. Introduction The coadsorption of different kinds of molecules on metal surfaces has received much attention for further understanding chemical reactions on surfaces covered with different adsorbates, like catalytic reactions. 1 Coadsorption is possible to induce ordering of the structures of adsorbed molecules; for example, ordered structures consisting of CO and benzene have been reported on many metal surfaces such as Rh(111), 2-7 Pd(111), 8 Pt(111), 6, 9 Ru(001), 10, 11 and Ni(111). 12 Ordered structures consisting of NO and benzene have also been examined on Ru(001) 10, 13 and Ni(111). 14 We have investigated the effect of coadsorption on surface chemical processes on Ru(001). The (2 × 2) structures were formed in CO-NH 3 and CO-methylamine coadlayers on Ru(001), and the dissociation paths of NH 3 and methylamine were influenced by coadsorbed CO. 15-18 Decomposition of C 2 H 2 was affected by CO in the coadlayer consisting of CO and C 2 H 2 . 19 NH 3 is often used as a reductant in selective catalytic reduction (SCR) of NO. NO reduction is indispensable for processing combustion products. Recently, high reactivity of Cu-based catalysts in NO reduction gained considerable inter- est. 20,21 Accumulated knowledge about the behavior of adsorbed NO on Cu with and without coadsorbates is needed for molecular-level understanding and rational control of surface NO reduction. The purpose of the present study is to investigate the structure and reactivity of the coadlayer of NO and NH 3 on Cu(111). The NO-NH 3 coadlayer on Pt(111) was investigated by Gland et al. 22 and Burgess et al. 23 In these studies, upward shift of the desorption peaks for NO and NH 3 in TPD was reported. Adsorption states and reactivity of NO in the pure layers on Cu single-crystal surfaces were examined by various methods such as XPS and HREELS. 24-26 For NO adsorption on Cu(111), So et al. reported by means of HREELS that bridge sites were preferred at low NO coverage and atop and 3-fold hollow sites were also occupied at high NO coverage. 26 As to ammonia adsorption on Cu(111), Hertel et al. investigated photodesorption and thermal desorption of ammonia on Cu(111). 27 In the present study a new (2 × 2) structure was found for the NO-NH 3 coadlayer on Cu(111) prepared by exposing Cu(111) to NO and NH 3 at 100 K and subsequently annealing to 180 K. In the TPD spectra of N 2 O and NH 3 from the (2 × 2) coadlayer, desorption peaks shifted to higher temperatures than those from each pure layer. These results are indicative of an attractive interaction between NO(a) and NH 3 (a) on Cu(111). 2. Experimental Section All experiments were carried out in an UHV chamber equipped with rearview LEED/AES optics, a quadrupole mass spectrometer, and a home built HREELS spectrometer, as described elsewhere. 28 The mass spectrometer was enclosed in a glass cap with a hole of 6 mm i. The contribution except that from a sample surface could be suppressed when the sample was placed in front of the hole at a distance of 2 cm. 28,29 A Cu(111) sample with a purity of 5N was purchased from MaTerial-Technologie and Kristalle GmbH (Ju ¨ lich), which was used in our previous study. 28 The sample was suspended by a U-shaped Ta wire threaded through two in-plane holes. 30 The sample temperature was measured by a thermocouple comprised of W-Re (5%) and W-Re (26%) spot-welded to the Ta wire in the vicinity of the sample. The sample was cooled down to 100 K by a liquid nitrogen reservoir installed in a sample holder and heated resistively to 1000 K at 2 K/s using a PID controller. TPD measurements were performed at a heating rate of 2 K/s. The sample was cleaned by cycles of Ar + sputtering (0.5 keV, 5 µA/cm 2 , 50 min) at 300 K and by annealing at 730 K for 20 min. The sample cleanliness was checked by AES and LEED. Gases used in this study were introduced into the chamber by back-filling through variable leak valves independently. 15 N 18 O (MSD, 15 N 99.7 atom %, 18 O 98.8 atom %) was used * Corresponding author. Fax: +81-3-3814-2627. E-mail: iwasawa@chem. s.u-tokyo.ac.jp. X Abstract published in AdVance ACS Abstracts, July 1, 1996. 13646 J. Phys. Chem. 1996, 100, 13646-13654 S0022-3654(96)00626-0 CCC: $12.00 © 1996 American Chemical Society