12230
J. Phys. Chem. C 2010, 114, 12230–12233
Site Switching from Di-σ Ethylene to π-Bonded Ethylene in the Presence of Coadsorbed Nitrogen on Pt(111) Jun Yin,† Michael Trenary,‡ and Randall Meyer*,† Departments of Chemical Engineering and Chemistry, UniVersity of Illinois at Chicago, Chicago, Illinois 60607 ReceiVed: April 7, 2010; ReVised Manuscript ReceiVed: May 30, 2010
The formation of π-bonded ethylene on a nitrogen-covered Pt(111) surface was studied by reflection absorption infrared spectroscopy (RAIRS) and temperature-programmed desorption (TPD). The observation with RAIRS of a peak at 975 cm-1 due to the dCH2 wagging (out of plane) mode is a clear indication of the presence of π-bonded ethylene on the surface. This form of ethylene desorbs from the surface at 216 K, a temperature much lower than ethylene desorption from the clean surface. Density functional theory calculations confirm that π-bonded C2H4 becomes thermoneutral compared with di-σ C2H4 when 0.25 ML of N atoms are coadsorbed on the Pt(111) surface. Introduction The adsorption and hydrogenation of ethylene on Pt surfaces have been extensively studied as model systems for alkene hydrogenation.1-5 Under ultrahigh vacuum (UHV) conditions below 240 K, ethylene is known to adsorb on Pt(111) via two σ bonds to the metal surface. The di-σ bonded ethylene then dehydrogenates to the ethylidyne (CCH3) species upon annealing. Although ethylidene (CHCH3) and surface ethyl species have been argued as relevant hydrogenation reaction intermediates in RAIRS and kinetic studies, it is now generally appreciated that π-bonded ethylene is the critical species for the ethylene hydrogenation reaction.4-7 Cremer and coworkers studied ethylene hydrogenation on the Pt(111) surface using sum frequency generation (SFG) under reaction conditions and determined that π-bonded ethylene is the dominant reaction intermediate that leads to the formation of ethane.7 Whereas varying the concentration of di-σ bonded ethylene did not affect the hydrogenation rate at 295 K, the coverage of ethylidyne was directly proportional to the presence of di-σ bonded ethylene, indicating that both ethylidyne and di-σ bonded ethylene are spectator species.8-10 In contrast, the π-bonded ethylene peak is unchanged by the coverage of ethylidyne. However, in Cremer’s experiment, the features ascribed to π-bonded ethylene are weak, indicating that it is present as a minority species on the surface. Further study has revealed that two types of π-bonded ethylene may potentially exist on the Pt(111) surface: physisorbed and chemisorbed. Physisorbed π-bonded ethylene can be observed on the clean Pt(111) surface at low temperature, but above 52 K, it begins to convert to di-σ bonded ethylene.11 Chemisorbed π-bonded ethylene has been reported to form on Pt(111) surfaces modified with such coadsorbates as oxygen12 and potassium or cesium.11,13,14 These studies have typically relied on indirect methods for determining the presence of π-bonded ethylene. In addition, Kubota et al. have used reflection absorption infrared spectroscopy (RAIRS) to detect π-bonded ethylene on Pt(111) at 112 K in the presence of 10-3 Torr of ethylene.15 The π-bonded ethylene disappeared as soon * Corresponding author. E-mail:
[email protected]. † Department of Chemical Engineering. ‡ Department of Chemistry.
as the gas-phase ethylene was pumped away and UHV was restored. So far, there is no RAIRS evidence of chemisorbed π-bonded ethylene on Pt(111) under UHV conditions. Here we demonstrate that N atoms on the Pt(111) surface can stabilize π-bonded ethylene relative to di-σ bonded ethylene, a result supported by density functional theory (DFT) calculations. In promoting the stability of the form of ethylene that is reactive toward hydrogenation on Pt(111), nitrogen is revealed to have the desirable ability to alter the chemical properties of a catalyst surface subtly while not being a reactant itself. Experimental Section The experiments were performed in two separate UHV chambers using two different Pt(111) crystals. The TPD results were obtained in a chamber (chamber 1) with a base pressure of ∼1 × 10-10 Torr. Detailed description of the system can been found elsewhere.16 In brief, the system is equipped with low-energy electron diffraction (LEED), an XPS system, and a quadrupole mass spectrometer (QMS) (UTI 100C) for TPD. The heating rate in the TPD experiment was 2 K/s. The data were smoothed using the Savitzky-Golay method. The RAIRS experiments were performed in a second chamber (chamber 2) with a base pressure of ∼2 × 10-10 Torr. A detailed description of this system can be found elsewhere.17 In brief, this UHV chamber is equipped for AES, LEED, and RAIRS using a commercial Fourier transform infrared (FTIR) spectrometer (Bruker IFS 66v/S). The IR beam enters and exits the UHV chamber through differentially pumped O-ring-sealed KBr windows and passes through a polarizer before reaching the infrared detector. An MCT (HgCdTe) detector and a SiC IR source were used. In cases where the sample was annealed to a temperature > 90 K, the sample was cooled back to 90 K before a spectrum was acquired. The background reference spectrum was also taken at 90 K. The resolution for the spectra is 4 cm-1. The Pt(111) surfaces were cleaned and judged free of impurities by a standard procedure previously described.18 Ammonia (99.9992%), oxygen (99.998%), and ethylene (99.8%) were purchased from Matheson Trigas and used without further purification. The preparation of a well-ordered p(2 × 2)-N/Pt(111) surface has been discussed previously in detail.19 In brief, the 0.25 monolayer (ML) N-Pt(111) surface was created by the oxida-
10.1021/jp103116m 2010 American Chemical Society Published on Web 06/24/2010
Site Switching from Di-σ C2H4 to π-Bonded C2H4
Figure 1. Temperature-programmed desorption of 0.8 L of ethylene from clean Pt(111) and from 0.25 ML of nitrogen on Pt(111). Both N2 and C2H4 contribute to the m/z ) 28 signal, but only C2H4 contributes to the m/z ) 27 signal, which is due to the C2H3 fragment of ethylene.
tive dehydrogenation of ammonia. Ammonia and oxygen are coadsorbed at 85 K, followed by annealing the sample to 400 K, generating water (which desorbs) and leaving behind a p(2 × 2)-N/Pt(111) surface with a nitrogen coverage of 0.25 ML. Computational Methodology The DFT calculations in this work were performed using the Vienna ab initio simulation package (VASP).20,21 A plane-wave basis set with a cutoff energy of 400 eV and ultrasoft Vanderbilt pseudopotentials (US-PP) were employed.22 Calculations were performed using the Perdew-Wang (PW-91) form of the exchange-correlation functional23 for thermodynamics of C2H4 adsorption. The (2 × 2) unit cell representing the system is four layers thick, in which the two uppermost metal layers were allowed to relax and approximately five layers of vacuum were used to separate the slabs. The Brillouin zone is sampled with a uniform 7 × 7 × 1 k-point grid (Monkhorst-Pack), as determined by convergence tests.24 In general, the structures’ geometries were optimized within a convergence tolerance of 10-3 eV. Results and Discussion Figure 1 shows the TPD results for m/e ) 28 and 27 after exposing 0.8 L of ethylene to both clean and nitrogen-covered Pt(111). On clean Pt(111), the saturation coverage of ethylene gives only a single peak around 300 K in agreement with previous work. Simultaneous to ethylene desorption, some di-σ bonded ethylene is converted to ethylidyne. When C2H4 is exposed to a surface precovered by 0.25 ML of nitrogen atoms, two peaks are seen at 216 and 300 K. Additional features for m/e ) 28 are observed above 400 K (but not for m/e ) 27), which are ascribed to N2 desorption (to be further discussed in a forthcoming publication). Figure 2 shows RAIRS spectra following exposure of 0.8 L of C2H4 to the p(2 × 2)-N/Pt(111) surface at 90 K. The most characteristic and intense peak due to π-bonded ethylene15,25,26
J. Phys. Chem. C, Vol. 114, No. 28, 2010 12231
Figure 2. RAIRS spectra following exposure of 0.8 L of ethylene to 0.25 ML of nitrogen on Pt(111).
is the dCH2 wagging (out-of-plane) mode, which appears at 975 cm-1 in Figure 2. The peak at 2920 cm-1 is assigned to the CH2 symmetric stretching mode of di-σ bonded ethylene. The peak at 2964 cm-1 follows the same changes with coverage and temperature as the 975 cm-1 peak and is therefore assigned to a CH stretching mode of π-bonded ethylene. The high frequency of the CH stretch peak at 3083 cm-1 indicates that it is also due to a form of π-bonded ethylene. The fact that this peak disappears with annealing to only 130 K, whereas the 975 and 2964 cm-1 peaks persists to 210 K, indicates that the 3083 cm-1 is due to a form of π-bonded ethylene that is less stabilized by the presence of N atoms on the surface. The π-bonded ethylene species disappears upon annealing to 210 K, whereas di-σ bonded ethylene remains on the surface before ultimately converting to ethylidyne at 310 K. Therefore, we assign the ethylene desorption peak at 216 K in TPD to the π-bonded ethylene species because the dCH2 wagging mode of π-bonded ethylene is diminished at the same temperature. Previous DFT calculations suggest that π-bonded ethylene occupies an atop site on Pt(111),27 in analogous fashion to its bonding in organometallic clusters.28 Therefore, one possible explanation for the observation of π-bonded ethylene at low ethylene coverages on the p(2 × 2)-N/Pt(111) surface is that N atoms block the preferred adsorption sites for di-σ bonded ethylene. However, our DFT calculations indicate that nitrogen atoms occupy fcc three-fold hollow sites29 whereas di-σ bonded ethylene bonds atop/atop to the Pt(111) surface in agreement with earlier experiments,30,31 allowing coadsorption of 0.25 ML of C2H4 on the N-Pt(111) surface. This implies that a simple site blocking mechanism is not likely to control the adsorption of C2H4. The calculations also confirm that whereas adsorption of di-σ C2H4 on clean Pt(111) in an atop/atop arrangement is favored by 0.44 eV over π-C2H4 at a C2H4 coverage of 0.25 ML, the situation changes dramatically in the presence of coadsorbed nitrogen atoms. On the N-Pt(111) surface, the favored site for di-σ bonded ethylene switches from atop/atop to atop/bridge where the molecule is centered over the fcc site.
12232
J. Phys. Chem. C, Vol. 114, No. 28, 2010
Figure 3. PDOS for clean Pt(111) and for Pt(111) covered with 0.25 ML of N.
In addition, π-C2H4 and di-σ C2H4 are essentially thermoneutral (di-σ C2H4 is favored by only 0.01 eV), indicating that both species should exist on the surface depending upon variations in local coverages. As depicted in Figure 3, density of states calculations of the nitrogen-covered surface reveal a shift in the d-band center of the metal away from the Fermi edge (from -1.94 to -2.52 eV on the 0.25 ML of N surface). Following the d-band model of Norskov and Hammer,32 this shift in the
Yin et al. d-band center readily explains a reduction in the reactivity of the surface toward adsorbates when N atoms are coadsorbed. Indeed, the adsorption energy of di-σ C2H4 reduces from -1.13 eV on Pt(111) to -0.48 eV on the nitrogen-covered surface. As indicated above, the extent to which the bonding of π-C2H4 is weakened is much less than that of di-σ C2H4. This can be explained by the lower coordination of π-bonded ethylene to the surface. In an effort to better understand the results, further experiments varying the coverage of C2H4 were performed. As shown in Figure 4, the presence of π-bonded C2H4 is maximized at an exposure of 0.8 L of C2H4 to the nitrogen-precovered surface. TPD data mirror that of the IR data, although it should be noted that the peak ascribed to π-bonded ethylene shrinks and narrows as the coverage of C2H4 increases. At higher exposure, more di-σ bonded ethylene is formed, which subsequently converts to ethylidyne when the surface is heated to higher temperatures. The reduced formation of π-bonded ethylene at high ethylene coverages suggests that the favored bonding mode for C2H4 on the N-Pt(111) surface may switch again back to di-σ atop/ atop in an analogous way to which the favored adsorption site switches on the Pt(111) surface from di-σ to π-bonded as coverage increases. DFT calculations could not, however, locate favorable adsorption configurations for higher C2H4 coverages. It is well known that halides and alkali metals may serve as promoters in metallic heterogeneous catalysts for many impor-
Figure 4. RAIRS spectra of different ethylene exposures to the p(2 × 2)-N/Pt(111) surface at 90 K, followed by annealing to different temperatures. (a) 0, (b) 0.4, (c) 0.8, and (d) 1.6 L.
Site Switching from Di-σ C2H4 to π-Bonded C2H4 tant reactions such as ethylene epoxidation33 and Fischer-Tropsch synthesis.34 Our work suggests that nitrogen atoms may be an effective promoter for the selective hydrogenation of alkenes as the bonding of alkenes to metal surfaces shifts from a spectator state to one that can easily be hydrogenated. By altering the nitrogen coverage, one may tune the electronic structure of the metal and potentially shift the selectivity dramatically. Furthermore, nitrogen atoms do not react with the molecular species and therefore present an intriguing option for promotion and selectivity control in other systems. Conclusions The formation of π-bonded ethylene on p(2 × 2)-N/Pt(111) has been observed by a combination of RAIRS and TPD. DFT calculations find that π-bonded C2H4 is thermoneutral compared with di-σ C2H4 when 0.25 ML of C2H4 is adsorbed on p(2 × 2)-N/Pt(111). However, changing the coverage of ethylene changes the relative portions of π-bonded C2H4 and di-σ C2H4 because of the repulsive interactions between ethylene and N atoms. Our results suggest that N may serve as a promoter for hydrogenation of alkenes. Acknowledgment. This work was supported by a grant from the U.S. National Science Foundation, CHE-0714562. References and Notes (1) Zaera, F. J. Am. Chem. Soc. 1989, 111, 4240. (2) Zaera, F.; Janssens, T. V. W.; Ofner, H. Surf. Sci. 1996, 368, 371. (3) Cremer, P. S.; Somorjai, G. A. J. Chem. Soc., Faraday Trans. 1995, 91, 3671. (4) Cremer, P.; Stanners, C.; Niemantsverdriet, J. W.; Shen, Y. R.; Somorjai, G. Surf. Sci. 1995, 328, 111. (5) Deng, R. P.; Herceg, E.; Trenary, M. Surf. Sci. 2004, 560, L195. (6) Wasylenko, W.; Frei, H. J. Phys. Chem. B 2005, 109, 16873. (7) Cremer, P. S.; Su, X. C.; Shen, Y. R.; Somorjai, G. A. J. Am. Chem. Soc. 1996, 118, 2942.
J. Phys. Chem. C, Vol. 114, No. 28, 2010 12233 (8) Beebe, T. P.; Yates, J. T. J. Am. Chem. Soc. 1986, 108, 663. (9) Davis, S. M.; Zaera, F.; Gordon, B. E.; Somorjai, G. A. J. Catal. 1985, 92, 240. (10) Rekoske, J. E.; Cortright, R. D.; Goddard, S. A.; Sharma, S. B.; Dumesic, J. A. J. Phys. Chem. 1992, 96, 1880. (11) Cassuto, A.; Mane, M.; Jupille, J.; Tourillon, G.; Parent, P. J. Phys. Chem. 1992, 96, 5987. (12) Steininger, H.; Ibach, H.; Lehwald, S. Surf. Sci. 1982, 117, 685. (13) Cassuto, A.; Touffaire, M.; Hugenschmidt, M.; Dolle, P.; Jupille, J. Vacuum 1990, 41, 161. (14) Cassuto, A.; Kiss, J.; White, J. M. Surf. Sci. 1991, 255, 289. (15) Kubota, J.; Ichihara, S.; Kondo, J. N.; Domen, K.; Hirose, C. Surf. Sci. 1996, 357, 634. (16) Kang, D. H.; Trenary, M. Surf. Sci. 2000, 470, L13. (17) Brubaker, M. E.; Trenary, M. J. Chem. Phys. 1986, 85, 6100. (18) Jentz, D.; Celio, H.; Mills, P.; Trenary, M. Surf. Sci. 1995, 341, 1. (19) Herceg, E.; Jones, J.; Mudiyanselage, K.; Trenary, M. Surf. Sci. 2006, 600, 4563. (20) Kresse, G.; Furthmuller, J. Phys. ReV. B 1996, 54, 11169. (21) Kresse, G.; Furthmuller, J. Comput. Mater. Sci. 1996, 6, 15. (22) Vanderbilt, D. Phys. ReV. B 1990, 41, 7892. (23) Perdew, J. P.; Wang, Y. Phys. ReV. B 1992, 45, 13244. (24) Monkhorst, H. J.; Pack, J. D. Phys. ReV. B 1976, 13, 5188. (25) Kubota, J.; Ohtani, T.; Kondo, J. N.; Hirose, C.; Domen, K. Appl. Surf. Sci. 1997, 121, 548. (26) Kubota, J.; Ichihara, S.; Kondo, J. N.; Domen, K.; Hirose, C. Langmuir 1996, 12, 1926. (27) Jacob, T.; Goddard, W. A. J. Phys. Chem. B 2005, 109, 297. (28) Grogan, M. J.; Nakamoto, K. J. Am. Chem. Soc. 1966, 88, 5454. (29) Mudiyanselage, K.; Trenary, M.; Meyer, R. J. J. Phys. Chem. C 2007, 111, 7127. (30) Kesmodel, L. L.; Dubois, L. H.; Somorjai, G. A. J. Chem. Phys. 1979, 70, 2180. (31) Starke, U.; Barbieri, A.; Materer, N.; Vanhove, M. A.; Somorjai, G. A. Surf. Sci. 1993, 286, 1. (32) Hammer, B.; Norskov, J. K. AdV. Catal. 2000, 45, 71. (33) Serafin, J. G.; Liu, A. C.; Seyedmonir, S. R. J. Mol. Catal. A: Chem. 1998, 131, 157. (34) Morales, F.; Weckhuysen, B. M. Promotion Effects in Co-Based Fischer-Tropsch Catalysis. In SPR: Catalysis; Spivey, J. J., Ed.; Royal Society of Chemistry: London, 2006; Vol. 19, p 1.
JP103116M
