CO-promoted N2 adsorption on copper atoms

PAPER

www.rsc.org/pccp | Physical Chemistry Chemical Physics

CO-promoted N2 adsorption on copper atoms Zhang-Hui Luab and Qiang Xu*ab Received 7th January 2010, Accepted 24th March 2010 First published as an Advance Article on the web 21st May 2010 DOI: 10.1039/c000398k Reactions of laser-ablated copper atoms with carbon monoxide and dinitrogen in excess argon have been investigated by matrix-isolation infrared spectroscopy. The NNCuCO and (NN)2CuCO complexes, as well as copper carbonyls, are formed as reaction products during sample deposition and further annealing whereas no copper dinitrogen complexes were observed. These carbonylcopper dinitrogen complexes are characterized on the basis of the results of the isotopic substitution and the reagent concentration change. Density functional theory calculations of the geometry structures, vibrational frequencies, relative absorption intensities, and isotopic shifts strongly support the experimental assignments. The experimental results reveal promoted adsorption of N2 on Cu atoms by preadsorbed CO. The theoretical results for reaction characteristics between the coadsorbed CO and N2 agree with the experimental findings. The joint investigations provide insights with regard to CO and N2 cooperative adsorption effects and consequent reaction mechanisms.

Introduction The interaction of copper centers with small molecules (i.e., CO, N2, O2, CO2, H2, CH4, etc.) is of considerable interest in the widely different fields of catalysis, synthesis, atmospheric chemistry, and biology.1 Among these small molecules, carbon monoxide is one of the most important in transition-metal chemistry from an academic or an industrial viewpoint.1,2 Many industrial processes employ CO as reagent and metal compounds as heterogeneous catalysts and involve the intermediates of metal carbonyls. For instance, copper polycarbonyl cations have been used as catalysts for carbonylation reactions of olefins, alcohols, and saturated hydrocarbons.3,4 On a more fundamental level, transition metal carbonyls are important models for complex systems such as chemisorption on metal surfaces5 or binding of small molecules at the active sites of metallo-proteins. The long-standing goal of elucidating mechanisms of the reactions involving CO has motivated numerous experimental and theoretical investigations of the interactions between CO and Cu atoms,6–16 clusters,17–21 and surfaces.22–27 Cu(CO)1–4+, Cu(CO)1–3, and Cu(CO)1–3 have been synthesized using the matrix isolation technique, and their infrared spectra in neon and argon matrices have been reported.12 Dinitrogen fixation and activation are one of the most challenging and important subjects in chemistry.28,29 N2 is isoelectronic with CO, whereas the reactivity of N2 toward Cu is quite different from that of CO. The co-condensation of Cu vapor in neat N2 matrices produced only the Cu2(N2) and Cu2(N2)2 complexes, suggesting that Cu dimers are active toward N2 whereas naked Cu(0) atoms are inert toward N2.30 It was interesting to find that the reactivity of copper a

National Institute of Advanced Industrial Science and Technology (AIST), Ikeda, Osaka 563-8577, Japan. E-mail: [email protected] b Graduate School of Engineering, Kobe University, Nada Ku, Kobe, Hyogo 657-8501, Japan

This journal is

c

the Owner Societies 2010

was greatly affected by the presence of CO in the N2 matrix, whereas the formation of Cux(CO)y(N2)z complexes was suggested while these complexes were unidentified.30–32 This prompted us to undertake the present matrix isolation study of the reactivity on Cu atoms toward CO and N2. Recent studies have shown that, with the aid of isotopic substitution, matrix isolation infrared spectroscopy, combined with density functional theory (DFT) calculation is very powerful in investigating the spectrum, structure, and bonding of novel species and the related reaction mechanisms.33–41 In this paper, we present experiments and density functional theory calculations on the reactions of Cu atoms with CO and N2, aiming at elucidation of the role of CO in the adsorption of N2 and the bonding nature of carbonylcopper dinitrogen complexes. IR spectroscopy coupled with theoretical calculation provides evidence for the formation of the new carbonylcopper dinitrogen complexes NNCuCO and (NN)2CuCO. The joint investigations show that, although Cu atoms are inert toward N2,30 the preadsorption of CO cooperatively enables the adsorption of N2 on Cu atoms.

Methods The experiments for laser ablation and matrix isolation infrared spectroscopy are similar to those previously reported.42,43 Briefly, the Nd:YAG laser fundamental (1064 nm, 10 Hz repetition rate with 10 ns pulse width) was focused on the rotating Cu target. The laser energy was varied from 5 to 30 mJ pulse 1. The laser-ablated Cu atoms were co-deposited with CO and N2 mixtures in excess argon onto a CsI window cooled normally to 7 K by means of a closed-cycle helium refrigerator. CO (99.95%, Japan Fine Products), 13C16O (99%, 18O o1%, ICON), 12C18O (99%, ICON), N2 (99.95%, SUZUKI SHOKAN Co., Ltd.), 15N2 (99.8%, SHOKO Co., Ltd.), and mixed isotopic samples were used to prepare the CO/N2/ Ar mixtures. In general, matrix samples were deposited for Phys. Chem. Chem. Phys., 2010, 12, 7077–7082 | 7077

30–60 min with a typical rate of 2–4 mmol h 1. After sample deposition, IR spectra were recorded on a BIO-RAD FTS-6000e spectrometer at 0.5 cm 1 resolution using a liquid nitrogen cooled HgCdTe (MCT) detector for the spectral range of 5000–400 cm 1. Samples were annealed at different temperatures and subjected to broad-band irradiation (l > 250 nm) using a high-pressure mercury arc lamp (Ushio, 100 W). Density functional theory calculations were performed to predict the structures and vibrational frequencies of the observed reaction products using the Gaussian 03 program.44 The B3LYP and BP86 density functional method were utilized.45–47 The 6-311+G(d) basis set was used for C, O, and N atoms,48,49 and the Wachters-Hay all-electron basis set was used for Cu atoms.50,51 All geometrical parameters were fully optimized and the harmonic vibrational frequencies were calculated with analytical second derivatives. Trial calculations and recent investigations have shown that such computational methods can provide reliable information for metal complexes, such as infrared frequencies, relative absorption intensities, and isotopic shifts.33–38

Fig. 1 Infrared spectra in the 2120–1820 cm 1 region from co-deposition of laser-ablated Cu atoms with CO/Ar or CO/N2/Ar samples. (a) 0.1% CO, 40 min sample deposition, (b) after annealing to 25 K; (c) 0.05% CO + 0.1% N2, 1 h sample deposition, (d) after annealing to 25 K, (e) after 10 min of broad-band irradiation, and (f) after annealing to 30 K.

Results and discussion Experiments have been done for laser-ablated Cu atom reactions with CO and N2 in excess argon with different CO and N2 concentrations. Typical infrared spectra for the products in the selected regions are illustrated in Fig. 1 and 2, and the absorption bands in different isotopic experiments are listed in Table 1. As shown in Fig. 1 (traces a and b) and Table 1, absorptions common to these experiments such as Cu(CO), Cu(CO)2, Cu(CO)3, Cu2(CO)2, and Cux(CO)y have been reported previously12,52 and are not discussed here. Similar experiments with N2 in excess argon (N2/Ar o 10%) did not produce copper dinitrogen complexes, in accord with previous experiments.30 However, new absorptions at 2088.0, 2066.4, 1929.3, and 1912.2 cm 1 appeared during sample deposition when CO/N2 mixture was used as reactant gas (Fig. 1, trace c). These bands are observed in lower laser power experiments and are not favored with higher laser energy and low CO/N2 mixture concentration, suggesting that one copper atom is involved. Doping with CCl4 has no effect on these bands, suggesting that the reaction products are neutral. The stepwise annealing and photolysis behavior of the product absorptions is also shown in the figures and will be discussed below. Similar matrix experiments with Ag and Au produce only silver and gold carbonyls as reported previously.53,54 Density functional theory calculations have been carried out for the possible isomers and electronic states of the potential product molecules. Fig. 3 shows the optimized structures, electronic ground states, and point groups calculated at the B3LYP/6-311+G(d) and BP86/6-311+G(d) levels. The results calculated with the B3LYP functional are in agreement with those from BP86 calculations. Hereafter, mainly B3LYP results are presented for discussion. Table 2 reports a comparison of the observed and calculated isotopic IR frequency (unscaled) and isotopic ratios for the N–N and C–O stretching modes of the new products. The energetic analysis for possible reactions are given in Table 3. Molecular orbital depictions of 7078 | Phys. Chem. Chem. Phys., 2010, 12, 7077–7082

Fig. 2 Infrared spectra in the 2120–1800 cm 1 region for laserablated Cu atoms co-deposited with isotopic CO/N2 mixtures in Ar after annealing to 25 K. (a) 0.05% CO + 0.1% N2, (b) 0.05% 13CO + 0.1% N2, (c) 0.05% 12CO + 0.05% 13CO + 0.15% N2, (d) 0.05% C18O + 0.1% N2, (e) 0.05% C16O + 0.05% C18O + 0.15% N2, (f) 0.05% CO + 0.15% 15N2, and (g) 0.05% CO + 0.15% 14 N2 + 0.15% 15N2.

the highest occupied molecular orbitals (HOMOs) and HOMO-1s of the new products are illustrated in Fig. 4. NNCuCO New absorptions at 2066.4 and 1912.2 cm 1 (Table 1 and Fig. 1) appear together during sample deposition. They exhibit the same behavior through deposition, annealing and broad-band irradiation, suggesting that they are due to different vibrational modes of the same molecule. The lower band at 1912.2 cm 1 shifts to 1871.7 cm 1 with 13C16O and to 1875.1 cm 1 with 12C18O, exhibiting isotopic frequency ratios (12C16O/13C16O: 1.0216; 12C16O/12C18O: 1.0198) characteristic of C–O stretching vibrations. Only the sum of pure isotopic bands is observed in the mixed 12C16O+13C16O + N2 This journal is

c


Table 1

IR Absorptions (in cm 1) observed from reaction of laser-ablated Cu atoms with CO and N2 mixtures in excess argon

CO + N2

13

C18O + N2

CO +

2088.0 2066.4 2034.1 2028.4 2010.2 1986.0 1975.9 1936.4 1929.3 1921.4 1912.2 1891.2 1876.5

2086.9 2065.4 1989.2 1984.1 1966.2 1941.1 1931.3 1892.1 1887.8 1877.4 1871.7 1847.5 1832.7

2087.7 2065.6 1991.4 1986.4 1962.7 1941.0 1931.1 1893.5 1890.1 1879.0 1875.1 1850.4 1836.7

2020.8 2000.3 2034.1 2028.4 2010.2 1986.0 1975.9 1936.4 1921.7 1921.4 1903.3 1891.2 1876.5

CO + N2

15

N2

(Fig. 2, trace c) and 12C16O + 12C18O + N2 (Fig. 2, trace e) experiments, implying that only one CO unit is involved in this mode.55 This band is about 98 cm 1 lower than the band for CuCO (Cs) and about 21 cm 1 higher than the antisymmetric C–O stretching vibration frequency for Cu(CO)2 (DNh) in an argon matrix.12,52 Furthermore, the 1912.2 cm 1 band also shows a small shift with 15N2 (8.9 cm 1) (Table 1 and Fig. 2), suggesting that the N2 unit is involved in this complex. The upper band at 2066.4 cm 1 shows small shifts (1.0 and 0.9 cm 1) with 13C16O and 13C18O, respectively, but a large shift (66.1 cm 1) with 15N2. The 12C16O/13C16O isotopic ratio of 1.0005, 12C16O/12C18O isotopic ratio of 1.0004 and 14N/15N isotopic ratio of 1.0331 suggest that this band is mainly due to a terminal N–N stretching vibration with small coupling to CO. This band is about 261 cm 1 lower compared to free N256 and about 218 cm 1 higher than the band for Cu2(N2).30 The mixed CO + 14N2 + 15N2 isotopic spectra (Fig. 2, trace g) only provides the sum of pure isotopic bands, indicating that only one N2 unit is involved in this complex.55 Accordingly, the absorptions at 2066.4 and 1912.2 cm 1 are assigned to the N–N and C–O stretching modes of NNCuCO, respectively. The assignment of the above bands to NNCuCO is in excellent agreement with DFT calculations. The NNCuCO

R(12CO/13CO)

R(C16O/C18O)

R(14N/15N)

Assignment

1.0005 1.0005 1.0226 1.0223 1.0224 1.0231 1.0231 1.0234 1.0220 1.0234 1.0216 1.0237 1.0239

1.0001 1.0004 1.0214 1.0212 1.0242 1.0232 1.0232 1.0227 1.0207 1.0226 1.0198 1.0221 1.0217

1.0333 1.0331 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0040 1.0000 1.0047 1.0000 1.0000

(NN)2CuCO NNCuCO Cux(CO)y Cux(CO)y CuCO Cu(CO)3 Cu(CO)3 site Cu2(CO)2 (NN)2CuCO Cu2(CO)2 NNCuCO Cu(CO)2 Cu(CO)2 site

Fig. 4 Molecular orbital depictions of the highest occupied molecular orbitals (HOMOs) and HOMO 1s of the carbonylcopper dinitrogen complexes calculated at the B3LYP/6-311+G(d) level.

complex is predicted to have a linear geometry with a 2P ground state (Fig. 3), consistent with previous theoretical calculations.30 A structural isomer with side-bonded N2 is 8.3 kcal mol 1 higher in energy than the linear doublet (Fig. 3). The N–N and C–O stretching vibrational frequencies of the linear NNCuCO complex are calculated to be 2165.3 and 2000.0 cm 1 (Table 2), respectively, it requires a 0.955 scale factor to fit the corresponding experimental values 2066.4

Fig. 3 Optimized structures (bond lengths in angstrom, bond angles in degree), electronic ground states, point groups of the possible isomers calculated at the B3LYP/6-311+G(d) and BP86/6-311+G(d) (in parentheses) levels.

This journal is

c


Phys. Chem. Chem. Phys., 2010, 12, 7077–7082 | 7079

Table 2

Comparison of observed and calculated IR frequency (unscaled) and isotopic ratios for the new products Freq./cm

Species NNCuCO

Obsd 2066.4 1912.2

(NN)2CuCO

2088.0 1929.3

a

1

R(12CO/13CO)

R(C16O/C18O)

R(14N/15N)

Calcd

Mode

Obsd

Calcd

Obsd

Calcd

Obsd

Calcd

B3LYP BP86 B3LYP BP86 B3LYP BP86 B3LYP BP86

nN–Na

1.0005

1.0004

1.0216

nN–N

1.0005

nC–Oa

1.0220

1.0047 1.0031 1.0176 1.0189 1.0000 1.0000 1.0215 1.0214

1.0331

nC–Oa

1.0050 1.0034 1.0193 1.0212 1.0000 1.0001 1.0224 1.0230

1.0224 1.0239 1.0123 1.0109 1.0350 1.0350 1.0036 1.0034

2165.3 2084.7 2000.0 1967.8 2146.2 2078.1 2033.6 1986.4

1.0198 1.0001 1.0207

1.0047 1.0333 1.0040

In these modes, the CO str and NN str motions are strongly coupled.

Table 3 Energetics for possible reactions computed by the B3LYP/ 6-311+G(d) method Reaction (1) (2) (3) (4)

Cu + CO - CuCO Cu + N2 - CuNN CuCO + N2 - NNCuCO NNCuCO + N2 - (NN)2CuCO

DEa 8.0 0.002 7.2 4.0

a DE is the reaction energy (kcal mol 1) and a negative value of energy denotes that the reaction is exothermic.

and 1912.2 cm 1, which is in line with other B3LYP scale factors.57,58 For the C–O stretching mode of the linear NNCuCO complex, the calculated 12C16O/13C16O, 12C16O/12C18O, and 14 N/15N isotopic frequency ratios of 1.0193, 1.0176, and 1.0123 (Table 2) are consistent with the experimental observations, 1.0216, 1.0198, and 1.0047, respectively. The calculated 12 16 C O/13C16O, 12C16O/12C18O, and 14N/15N isotopic frequency ratios of the N–N stretching mode of the linear NNCuCO complex also agree with the experimental values (Table 2). (NN)2CuCO New bands at 2088.0 and 1929.3 cm 1 appear together during sample deposition, increase upon sample annealing to 25 K, disappear after broad-band irradiation, and almost recover after further annealing to 30 K (Table 1 and Fig. 1). The lower band at 1929.3 cm 1 shifts to 1887.8 cm 1 with 13C16O and to 1890.1 cm 1 with 12C18O, exhibiting isotopic frequency ratios (12C16O/13C16O: 1.0220; 12C16O/12C18O: 1.0207) characteristic of C–O stretching vibrations. The mixed 12C16O+13C16O + N2 (Fig. 2, trace c) and 12C16O + 12C18O + N2 (Fig. 2, trace e) isotopic spectra only provide the sum of pure isotopic bands, implying that one CO unit is involved in this mode.55 This band is about 81 cm 1 lower than the C–O vibrational frequency of CuCO (Cs),12,52 and about 17 cm 1 higher than the band for NNCuCO (CNV) in the present experiments, respectively. Furthermore, the 1929.3 cm 1 band also shows a small shift 7.6 cm 1 with 15N2 (Table 1 and Fig. 2), suggesting that the N2 unit is involved in this complex. The upper band at 2088.0 cm 1 shows a large shift with 15N2 (14N2/15N2: 1.0333), indicating that this mode is mainly due to a N–N stretching vibration. This band is about 240 cm 1 lower as compared to free N256 and about 192 cm 1 higher than the band for Cu2(N2)2.30 A triplet isotopic pattern is observed in the mixed CO + 14N2 + 15N2 isotopic spectra (Fig. 2, trace g), implying 7080 | Phys. Chem. Chem. Phys., 2010, 12, 7077–7082

that two N2 units are involved in the complex.55 Accordingly, the absorptions at 2088.0 and 1929.3 cm 1 are assigned to the asymmetric N–N and C–O stretching modes of (NN)2CuCO, respectively. B3LYP calculations predicted a planar (C2v) equilibrium structures with linear CuCO and CuNN units for (NN)2CuCO.30 Our B3LYP and BP86 calculations give results consistent with the previous calculations.30 The (NN)2CuCO complex is predicted to have a 2B1 ground state with a C2v symmetry (Fig. 5). The symmetric and asymmetric N–N and C–O stretching frequencies of the (NN)2CuCO species are calculated to be 2267.8, 2146.2, and 2033.6 cm 1, respectively. The intensity of the symmetric N–N stretching vibration 2267.8 cm 1 in (NN)2CuCO is predicted to be about 1/7 that of the asymmetric N–N stretching vibration 2146.2 cm 1, which is not readily observed and is consistent with the absence of the symmetric N–N stretching vibration of the (NN)2CuCO complex from the present experiments. The calculated 12 16 C O/13C16O, 12C16O/12C18O, and 14N/15N isotopic frequency ratios of the asymmetric N–N and C–O stretching vibrations are consistent with the experimental values (Table 2), respectively, which support the identification of the (NN)2CuCO complex. Reaction mechanisms Under the present experimental conditions, laser-ablated Cu atoms react with CO and N2 mixtures in the excess argon matrices to produce carbonylcopper dinitrogen species as well as copper carbonyls. Binary copper carbonyls Cu(CO)1–3 have been prepared by reactions of Cu atoms with CO, and identified using infrared spectroscopy.12,52 However, the isoelectronic Cu(NN)1–3 were found to be unbound with respect to Cu + N2.30 Unlike for Cu(NN)1–3, the N2 molecule is clearly bound in NNCuCO and (NN)2CuCO from our present experiments. The present experiments clearly show promoted adsorption of N2 on Cu atoms by preadsorbed CO. The assignment is supported by DFT calculations of energy changes for possible reactions. Our experimental observations suggest that the NNCuCO species is formed by reactions of CuCO and N2 whereas the (NN)2CuCO species is formed via the N2 addition (Table 3, reactions (3) and (4)). The CuCO intermediate is produced from reaction (1) during deposition and after annealing. As can be seen in Table 3, reactions (1), (3), and (4) are exothermic by 8.0, 7.2 and 4.0 kcal mol 1 at the B3LYP/6-311+G(d) level, respectively, but the energy of This journal is

c


reaction (2) is practically zero, and hence, the thermodynamic driving force to give CuNN is limited. Reactions (1) and (2) are exothermic by 8.0 and 0.002 kcal mol 1, respectively, implying that the Cu atoms react with CO but not with N2, in agreement with the experimental observations. However, the reaction of CuCO with N2 is calculated to be exothermic by 7.2 kcal mol 1, suggesting that the reactivity of Cu is greatly affected by the presence of CO and the CuCO species are active toward N2, consistent with the experimental results. As N2 does not react with the Cu atoms, CO preadsorption is responsible for the activation of the Cu atoms to enable cooperative N2 coadsorption. CO chemisorption-induced activation of metal centers was also observed for Au clusters.59–62 CO preadsorption enables O2 coadsorption to the Au clusters.61,62 The preadsorption of one ligand (CO in our case) can effectively change the electronic structure of the metal atoms or clusters that might enable the subsequent reaction with further reactant molecules that would otherwise not be possible. Similar cooperative effects have been recently observed in the Aux+ clusters with O2 and H2 system, where the coordination of O2 to Aux+ clusters was promoted by H2.63 As illustrated in Fig. 4, the highest occupied molecular orbitals (HOMOs) in the NNCuCO and (NN)2CuCO complexes are of p-type whereas the HOMO-1s are of s-type. Similar to the transition metal carbonyl complexes, the bonding in carbonylcopper dinitrogen complexes is dominated by the s-type donation of electrons from CO and N2 to the empty s orbital of Cu and the p-type back-donation from the Cu orbitals into the antibonding orbitals of CO and N2, in accord with the Dewar-Chatt-Duncanson (DCD) complexation model.64,65 Such mechanism weakens the CO and N2 bonds, lowering the C–O and N–N stretching vibrational frequency, consistent with the experimental observations.

Conclusions New carbonylcopper dinitrogen complexes NNCuCO and (NN)2CuCO have been prepared by the reactions of laserablated Cu atoms with N2 in excess argon. The absorptions at 2066.4 (2088.0) and 1912.2 (1929.3) cm 1 are assigned to the N–N and C–O stretching vibrations of the NNCuCO ((NN)2CuCO) molecules, respectively, on the basis of the isotopic shifts and mixed isotopic splitting patterns. DFT calculations have been performed, which lend support to the experimental assignments of the matrix infrared spectra. The present experimental and theoretical results show that, although Cu atoms are known to be inactive toward N2, the preadsorption of CO cooperatively enables the adsorption of N2 on Cu atoms.

Acknowledgements The authors thank the reviewers for valuable suggestions and comments. This work was supported by AIST and a Grant-inAid for Scientific Research (B) (Grant No. 17350012) from the Ministry of Education, Culture, Sports, Science and Technology (MEXT) of Japan. Z.-H.L. acknowledges support from Marubun Research Promotion Foundation (MRPF). This journal is

c


References 1 F. A. Cotton, G. Wilkinson, C. A. Murillo and M. Bochmann, Advanced Inorganic Chemistry, Wiley, New York, 6th edn, 1999. 2 E. L. Muetterties and J. Stein, Chem. Rev., 1979, 79, 479. 3 Y. Souma and H. Sano, J. Org. Chem., 1973, 38, 2016. 4 Q. Xu, Coord. Chem. Rev., 2002, 231, 83 and references therein. 5 G. A. Somorjai, Introduction to Surface Chemistry and Catalysis, Wiley, New York, 1994. 6 F. Meyer, Y. M. Chen and P. B. Armentrout, J. Am. Chem. Soc., 1995, 117, 4071. 7 H. Huber, E. P. Kunding, M. Moskovits and G. A. Ozin, J. Am. Chem. Soc., 1975, 97, 2097. 8 P. H. Kasai and P. M. Hones, J. Am. Chem. Soc., 1985, 107, 813. 9 J. H. B. Chenier, C. A. Hampson, J. A. Howard and B. Mile, J. Phys. Chem., 1989, 93, 114. 10 M. A. Blitz, S. A. Mitchell and P. A. Hackett, J. Phys. Chem., 1991, 95, 8719. 11 D. Post and E. J. Baerends, J. Chem. Phys., 1983, 78, 5663. 12 M. F. Zhou and L. Andrews, J. Chem. Phys., 1999, 111, 4548. 13 G. H. Jeung, J. Am. Chem. Soc., 1992, 114, 3211. 14 C. W. Bauschlicher, J. Chem. Phys., 1994, 100, 1215. 15 C. Adamo and F. Lelj, J. Chem. Phys., 1995, 103, 10605. 16 V. Barone, J. Phys. Chem., 1995, 99, 11659. 17 M. Moskovits and J. E. Hulse, J. Phys. Chem., 1977, 81, 2004. 18 Z. X. Cao, Y. J. Wang, J. Zhu, W. Wu and Q. N. Zhang, J. Phys. Chem. B, 2002, 106, 9649. 19 A. Poater, M. Duran, P. Jaque, A. Toro-Labbe and M. Sola, J. Phys. Chem. B, 2006, 110, 6526. 20 W. Liu, Y. H. Zhao, E. J. Lavernia and Q. Jiang, J. Phys. Chem. C, 2008, 112, 7672. 21 L. Jiang and Q. Xu, J. Chem. Phys., 2008, 128, 124317. 22 W. Eberhardt and E. W. Plummer, Phys. Rev. B: Condens. Matter, 1983, 28, 3605. 23 E. Borguet and H. L. Dai, J. Chem. Phys., 1994, 101, 9080. 24 A. P. Graham, F. Hofmann, J. P. Toennies, G. P. Williams, C. J. Hirschmugl and J. Ellis, J. Chem. Phys., 1998, 108, 7825. 25 O. Koga, S. Teruya, K. Matsuda, M. Minami, N. Hoshi and Y. Hori, Electrochim. Acta, 2005, 50, 2475. 26 E. Borguet and H. L. Dai, J. Phys. Chem. B, 2005, 109, 8509. 27 C. Liu and R. G. Tobin, J. Chem. Phys., 2007, 126, 124705. 28 J. B. Howard and D. C. Rees, Chem. Rev., 1996, 96, 2965. 29 H. J. Himmel and M. Reiher, Angew. Chem., Int. Ed., 2006, 45, 6264. 30 F. Elustondo, J. Mascetti and I. Papai, J. Phys. Chem. A, 2000, 104, 3572. 31 S. N. Ceasro and S. Dobos, Mikrokim. Acta (Suppl.), 1997, 14, 387. 32 S. Dobos and S. N. Ceasro, High Temp. Mater. Sci., 1997, 37, 81. 33 M. F. Zhou, L. Andrews and C. W. Bauschlicher, Chem. Rev., 2001, 101, 1931. 34 Y. Gong, M. F. Zhou and L. Andrews, Chem. Rev., 2009, 109, 6765. 35 G. J. Wang and M. F. Zhou, Int. Rev. Phys. Chem., 2008, 27, 1. 36 L. Jiang and Q. Xu, J. Am. Chem. Soc., 2005, 127, 42. 37 Q. Xu, L. Jiang and N. Tsumori, Angew. Chem., Int. Ed., 2005, 44, 4338. 38 H. G. Cho, X. F. Wang and L. Andrews, J. Am. Chem. Soc., 2005, 127, 465. 39 G. J. Wang, Y. Gong, M. H. Chen and M. F. Zhou, J. Am. Chem. Soc., 2006, 128, 5974. 40 M. F. Zhou, X. Jin, Y. Gong and J. Li, Angew. Chem., Int. Ed., 2007, 46, 2911. 41 X. F. Wang and L. Andrews, J. Am. Chem. Soc., 2008, 130, 6766. 42 T. R. Burkohlder and L. Andrews, J. Chem. Phys., 1991, 95, 8697. 43 Z. H. Lu, L. Jiang and Q. Xu, J. Chem. Phys., 2009, 131, 034512. 44 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, J. A. Montgomery Jr, T. Vreven, K. N. Kudin, J. C. Burant, J. M. Millam, S. S. Iyengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli,

Phys. Chem. Chem. Phys., 2010, 12, 7077–7082 | 7081

45 46 47 48 49 50 51 52

J. Ochterski, P. Y. Ayala, K. Morokuma, G. A. Voth, P. Salvador, J. J. Dannenberg, V. G. Zakrzewski, S. Dapprich, A. D. Daniels, M. C. Strain, O. Farkas, D. K. Malick, A. D. Rabuck, K. Raghavachari, J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul, S. Clifford, J. Cioslowski, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara, M. Challacombe, P. M. W. Gill, B. G. Johnson, W. Chen, M. W. Wong, C. Gonzalez and J. A. Pople, GAUSSIAN 03 (Revision B.04), Gaussian, Inc., Pittsburgh, PA, 2003, and references therein. J. P. Perdew, Phys. Rev. B: Condens. Matter, 1986, 33, 8822. A. D. Becke, J. Chem. Phys., 1993, 98, 5648. C. Lee, E. Yang and R. G. Parr, Phys. Rev. B: Condens. Matter, 1988, 37, 785. A. D. McLean and G. S. Chandler, J. Chem. Phys., 1980, 72, 5639. R. Krishnan, J. S. Binkley, R. Seeger and J. A. Pople, J. Chem. Phys., 1980, 72, 650. A. J. H. Wachter, J. Chem. Phys., 1970, 52, 1033. P. J. Hay, J. Chem. Phys., 1977, 66, 4377. L. Jiang and Q. Xu, J. Phys. Chem. A, 2007, 111, 2690.

7082 | Phys. Chem. Chem. Phys., 2010, 12, 7077–7082

53 B. Liang and L. Andrews, J. Phys. Chem. A, 2000, 104, 9156. 54 L. Jiang and Q. Xu, J. Phys. Chem. A, 2005, 109, 1026. 55 J. H. Darling and J. S. Ogden, J. Chem. Soc., Dalton Trans., 1972, 2496. 56 M. F. Zhou and L. Andrews, J. Phys. Chem. A, 1998, 102, 9061. 57 A. P. Scott and L. Radom, J. Phys. Chem., 1996, 100, 16502. 58 I. Bytheway and M. W. Wong, Chem. Phys. Lett., 1998, 282, 219. 59 H. J Zhai, B. Kiran, B. Dai, J. Li and L. S. Wang, J. Am. Chem. Soc., 2005, 127, 12098. 60 H. J. Zhai and L. S. Wang, J. Chem. Phys., 2005, 122, 051101. 61 W. T. Wallace and R. L. Whetten, J. Am. Chem. Soc., 2002, 124, 7499. 62 J. Hagen, L. D. Socaciu, M. Elijazyfer, U. Heiz, T. M. Bernhardt and L. Woste, Phys. Chem. Chem. Phys., 2002, 4, 1707. 63 S. M. Lang, T. M. Bernhardt, R. N. Barnett, B. Yoon and U. Landman, J. Am. Chem. Soc., 2009, 131, 8939. 64 J. Chatt, G. A. Rowe and A. A. Williams, Proc. Chem. Soc., London, 1957, 208. 65 J. Chatt, L. A. Duncanson and R. G. Guy, J. Chem. Soc., 1961, 827.

This journal is

c
