DFT Study of Interaction of Azoles with Cu(111) and Al(111) Surfaces

ARTICLE pubs.acs.org/JPCC

DFT Study of Interaction of Azoles with Cu(111) and Al(111) Surfaces: Role of Azole Nitrogen Atoms and DipoleDipole Interactions Natasa Kovacevic and Anton Kokalj* Department of Physical and Organic Chemistry, Jozef Stefan Institute, Jamova 39, SI-1000 Ljubljana, Slovenia

bS Supporting Information ABSTRACT: Azoles and their derivatives are among the often used organic corrosion inhibitors for copper. For this reason, the adsorption of four azole molecules—imidazole, 1,2,3-triazole, tetrazole, and pentazole—on Cu(111) and Al(111) surfaces has been studied and characterized using density functional theory (DFT) calculations. We find that the molecules weakly adsorb in an upright geometry onto the top site of Cu(111) via single nitrogen atom; except for triazole the bonding with two nitrogen atoms to a bridge site becomes slightly preferred at very low coverage. Molecular electronic structure is only weakly perturbed upon adsorption and the moleculesurface interaction involves the hybridization between molecular σ orbitals and metal states, yet the main contribution to bonding comes from the electrostatic dipole interactions due to a large dipole moment of azole molecules. Also, the lateral intermolecular repulsion can be significant and very long ranged. With increasing the number of nitrogen atoms in azole ring the molecular electronegativity and chemical hardness linearly increase. The harder the molecule the more difficult the hybridization with metal states, which can explain why with the increasing number of nitrogen atoms in azole ring the moleculeCu(111) bond strength decreases linearly, being 0.69, 0.55, 0.43, and 0.22 eV for imidazole, 1,2,3triazole, tetrazole, and pentazole, respectively. The same bonding trend with very similar adsorption energies is also found on Al(111).

1. INTRODUCTION Some organic molecules have the ability to remarkably slow down the corrosion of metals and alloys; they are called corrosion inhibitors. In particular, heterocyclic organic compounds consisting of π-system or N, O, P, or S heteroatoms are known to inhibit efficiently corrosion of metals. Azoles (five-membered heterocyclic molecules consisting of one or more nitrogen atoms) are among the popular organic corrosion inhibitors.13 Imidazole, triazole, and tetrazole derivatives are well-known as efficient corrosion inhibitors for various metals and alloys.3 The interaction of azole molecules with copper surfaces is usually explained as being due to hybridization of molecular π-system or nitrogen lone pair orbitals with metal states.2 Despite the fact that the interaction between the inhibitor molecules and metal surfaces is considered to be important in achieving the inhibition of corrosion, only a few state-of-the-art density functional theory (DFT) studies were devoted to explicit characterization of the inhibitorsurface bonding.48 Instead many computational studies rely on straightforward correlation between the electronic properties of isolated inhibitor molecules and their inhibition characteristics, thus surpassing the explicit moleculesurface characterization (e.g., see ref 9 and the references therein). In this Article, we consider the adsorption of four azole molecules—imidazole, 1,2,3-triazole, tetrazole, and pentazole— on Cu(111) and Al(111) surfaces by means of atomistic computer simulations based on density-functional theory and r 2011 American Chemical Society

plane-wave pseudopotential method. Whereas imidazole, triazole, and tetrazole molecules are known as efficient corrosion inhibitors,3 the pentazole molecule is considered for the sake of definiteness. Lewis structures of these molecules and the numbering of N atoms are illustrated in Figure 1. For brevity reasons, the 1,2,3-triazole will be denotated as triazole in the following. The aim of this study is to characterize in detail the bonding of neutral azole molecules onto copper and aluminum surfaces (deprotonated (dehydrogenated) azoles will be considered in subsequent publication) and to determine how the number of nitrogen atoms in azole ring affects the moleculemetal interaction. The main focus is devoted to Cu(111), whereas the Al(111) is used to complement the results and provide firmer understanding of the molecule surface bonding. (Corresponding results are provided in the Supporting Information.) Moreover, we also evaluate the applicability of the molecular reactivity indicators (those emerging from the theoretical formalization10 of the HSAB (hard and soft acids and bases) principle11) in the case of a molecule surface interaction. Although such indicators found a widespread application in the quantum chemical studies of corrosion inhibitors,1221 the subject is less consolidated in the field of Received: July 25, 2011 Revised: October 31, 2011 Published: October 31, 2011 24189

dx.doi.org/10.1021/jp207076w | J. Phys. Chem. C 2011, 115, 24189–24197

The Journal of Physical Chemistry C

ARTICLE

2.2. Evaluation of Adsorption Energies. The adsorption energy at a given coverage is calculated as

Eads ¼ Emol=surf ðEsurf þ Emol Þ Figure 1. Lewis structures of imidazole, 1,2,3-triazole, tetrazole, and pentazole. Numbering of N atoms is also indicated.

surface reactivity, and only a few studies rigorously consider their applicability in this context.2224 We further point out that because of a large dipole moment of azole molecules the lateral dipoledipole interactions can be very long-ranged and may extend beyond the nearest-neighbor distance of a few tens of bohrs. A proper evaluation of the adsorption energies calculated by first-principle computations employing periodic boundary conditions is therefore required. For this reason, the adsorption energies are extrapolated to the zero coverage limit by a scheme recently developed by one of the authors.25

2. TECHNICAL DETAILS AND DEFINITIONS 2.1. Computational. Calculations were performed within the framework of DFT using a generalized gradient approximation of PerdewBurkeErnzerhof (PBE).26 The molecule/surface systems were modeled by the ultrasoft pseudopotential method27,28 using the PWscf code from the Quantum ESPRESSO distribution.29 Orbitals were expanded in a plane-wave basis set up to a kinetic energy cutoff of 30 Ry (240 Ry for the charge density cutoff). Surfaces were modeled by periodic multislab model consisting of four (111) layers, with the bottom layer constrained to the bulk positions and the in-plane lattice spacing fixed to the calculated equilibrium bulk lattice parameter: 3.67 Å for Cu30 and 4.06 Å for Al.31 All other degrees of freedom were relaxed. Azole molecules were adsorbed on one side of the slab and the thickness of the vacuum region (the distance between the top of the ad-molecule and the adjacent slab) was set to ∼20 Å. Dipole correction of Bengtsson32 was applied to cancel an artificial electric field that develops along the direction normal to the surface due to periodic boundary conditions imposed on the electrostatic potential. Molecular adsorption on Cu(111) was modeled at various coverages, Θ, ranging from 1/9 to 1/36 ML (Θ is defined as the inverse of the number of surface metal atoms per molecule) using the (3 3), (4 4), (5 5), and (6 6) supercells, whereas adsorption on Al(111) was modeled solely by (5 5) supercell. Brillouin-zone (BZ) integrations were performed with Gaussin-smearing33 special-point technique34 using a smearing parameter of 0.03 Ry and a 3 3 1 k-point mesh for (3 3) supercell and 2 2 1 k-point mesh for larger supercells. The calculations of isolated molecules were performed, for convenience reasons, with the local Gaussian-type-orbital basis set and GAUSSIAN0935 program because the evaluation of vertical ionization potentials and electron affinities involve calculations of charged species. The 6-311++G(d,p) basis set was used because it gives, according to convergency tests, results comparable to those obtained from plane-wave calculations; the differences in energies and bond lengths between the 6-311+ +G(d,p) and plane-wave basis sets are ∼0.01 eV and triazole > tetrazole > pentazole. The optimal zero-coverage adsorption energies are 0.69, 0.55, 0.43, and 0.22 eV for imidazole, triazole, tetrazole, and pentazole, respectively. 24192

dx.doi.org/10.1021/jp207076w |J. Phys. Chem. C 2011, 115, 24189–24197


ARTICLE

Figure 3. Optimized structures of chemisorbed imidazole, triazole, tetrazole, and pentazole on Cu(111). The stablest top and the stablest bridgebonded structures per molecule are shown in top and bottom panels, respectively. The NCu bond lengths are also stated.

Figure 4. PBE adsorption energy of (a) imidazole, (b) triazole, (c) tetrazole, and (d) pentazole on Cu(111) as a function of lateral Rnn distance (defined graphically in the inset) for top and bridge-bonded molecules. Fitted curves are calculated according to eqs 3 and 13, whereas the thin dash-dotted horizontal lines indicate the corresponding extrapolated zero-coverage (Rnn = ∞) adsorption energies, E∞ ads. Calculations were performed with (3 3), (4 4), (5 5), and (6 6) supercells corresponding to 1/9, 1/16, 1/25, and 1/36 ML coverage, respectively.

Table 1. Extrapolated Zero Coverage Adsorption Energies, E∞ ads, of Identified Adsorption Modes of Imidazole, Triazole, Tetrazole, and Pentazole on Cu(111)a E∞ ads (eV) N2

N3

N4

N2+N3

N3+N4

0.69 ( 0.02

imidazole triazole

0.50 ( 0.04

0.52 ( 0.03

tetrazole pentazole

0.39 ( 0.08 0.22 ( 0.06

0.36 ( 0.04 0.17 ( 0.06b

0.55 ( 0.03 0.43 ( 0.02 0.17 ( 0.06b

0.32 ( 0.05 0.09 ( 0.02

0.42 ( 0.08 0.14 ( 0.06

a b The most exothermic E∞ adsof each molecule is accentuated. The estimated extrapolation errors are also indicated. For pentazole, the N3 and N4 bonding modes are symmetry equivalent.

To explain the trend of the adsorption energy, we perform electronic structure analysis and also utilize the molecular reactivity indicators, defined in Section 2.3. Moreover, to yield easier comparison, we choose to analyze the top sites for all molecules, although for triazole the bridge site is slightly more stable than the top site at very low coverages (by 0.03 eV; this tiny difference does not change the adsorption energy trend). Figure 5a plots the PBE calculated vertical ionization potential and electron affinity of the four molecules, whereas the resulting

electronegativities and absolute hardnesses are plotted in Figure 5b. It is seen that ionization potential increases monotonously as passing from imidazole to pentazole (I ∈ [9.0,11.8] eV), whereas electron affinity is similar for all four molecules (i.e., A ∈ [0.7, 0.3] eV). Because of almost constant A the χmol and ηmol display similar trends; that is, both increase monotonously as passing from imidazole to pentazole. Whereas imidazole is less electronegative (4.32 eV), pentazole is more electronegative (5.62 eV) than the Cu(111), for which the calculated work 24193



Figure 5. (a) Vertical ionization potential and electron affinity of the four considered molecules. (b) Corresponding Mulliken electronegativity and absolute hardness calculated as χ = (I + A)/2 and η = (I A)/2.

Figure 6. (a) Correlation between the molecular net charge, calculated by Bader analysis, and the HSAB estimated moleculetometal charge transfer ΔN. (Solid points stand for the stablest top adsorption site; blue crosses mark the slightly less stable second-stablest top adsorption sites for triazole and tetrazole.) (b) Correlation between the adsorption energy and ΔN; E0 versus ΔN (dashed-dotted line) and E∞ ads versus ΔN (dashed line). The estimated accuracies of E∞ ads and E0 are indicated by error bars.

function is Φ = 4.83 eV.24 This indicates that the molecule-tometal charge transfer should also decrease as passing from imidazole to pentazole; indeed the direction of charge transfer should be the opposite for the latter, that is, from metal-tomolecule. This picture is confirmed by Bader charge analysis; although we find that the calculated Bader charge of adsorbed molecule depends on the adsorption site, these calculations confirm that the molecule-to-metal charge transfer follows the order: imidazole > triazole > tetrazole > pentazole. The proportionality between the Bader molecular net charge of the stablest top adsorption site and the ΔN estimate, defined by eq 18a, is plotted in Figure 6a, and it is seen that a correlation between the two is acceptable. Already Pearson stated that the ΔN values should be a measure of relative bond strengths provided that a series of similar acids and bases are compared. 50 It has been recently shown that this concept applies also for adsorbates on metal surfaces.2224 This is also found currently, and the proportionality between the current adsorption energies (i.e., E∞ ads and E 0 , defined by eq 14) and the ΔN estimate is plotted in Figure 6b. The correlation between the E ∞ ads and ΔN is rather excellent and much better than the correlation between the E 0 and ΔN, although the E 0 is a more appropriate measure of covalent interaction. Note, however, that the E0 values suffer from large uncertainties, whereas the E∞ ads estimation is much more reliable and accurate. (See the last paragraph of Section 2.2.) Despite the good correlation between the adsorption energy and the ΔN, an intriguing issue nevertheless emerges.

ARTICLE

Namely, the strength of the formed moleculemetal bond should be proportional to the magnitude of ΔN irrespective of the direction of charge transfer; note that the ΔE of eq 18b is proportional to (Φ χmol)2 , whereas the ΔN Φ χmol and can be therefore either positive (Φ > χmol, molecule donates charge to metal) or negative (Φ < χmol, metal donates charge to molecule). However, the magnitude of charge transfer is not the smallest for pentazole, which bonds the weakest to Cu(111). This issue should be, therefore, clarified. A possible explanation could be that (i) the relative amount of the charge donation and back-donation is different for different molecules, (ii) the very small current charge transfers are not decisive, or (iii) the calculated charges suffer from systematic error. To shed some light onto this question, an electronic structure analysis of the moleculesurface bonding is presented below. Figure 7 plots the charge density difference, ΔF(r) = Fmol/surf(r) Fsurf(r) Fmol(r), and the projected density of states (PDOS) for all four molecules on Cu(111). For triazole and tetrazole, the two most stable top sites per molecule are considered. In the ΔF(r) plots, red (blue) color represents electron charge accumulation (deficit) regions. The formation of direct NCu bonds is clearly seen by the charge redistribution and the charge flows from the involved N and Cu atoms toward the center of NCu bonds (red ellipsoidal shapes). The NCu charge redistribution is similar for all four molecules. It can be further observed from the charge density difference plots that in addition to NCu bonds the molecules interact with the surface also with the bottommost H atoms, as seen by charge accumulation located in between them and the surface. This implies the formation of a XH 3 3 3 Cu(X = N or C) hydrogen bond, thus making the molecule more strongly bound to the surface. Such type of H bonds is known to form in metal complexes51,52 and between adsorbed molecules and metal surfaces.7 Imidazole molecule forms two CH 3 3 3 Cu hydrogen bonds, whereas pentazole forms one NH 3 3 3 Cu hydrogen bond. Triazole and tetrazole form either one CH 3 3 3 Cu hydrogen bond (top row panels of Figure 7) or one NH 3 3 3 Cu hydrogen bond (bottom row panels). The latter hydrogen bond is stronger, as seen by more pronounced charge accumulation in between the H atom and the surface. Note that on the basis of maps of electrostatic potential (Figure 2) the N3 or N4 atoms should be more reactive than the N2 atom. However, the molecule bonded via N3 or N4 to the surface cannot form the NH 3 3 3 Cu hydrogen bond, which is why the bonding via the N2 atom is competitive to that of N3 or N4. The PDOS plots shown beneath the ΔF(r) plots in Figure 7 reveal a spiky molecular PDOS for all four molecules, indicating a weak moleculesurface interaction. The HOMO peaks are located 3 to 4 eV below the Fermi energy, whereas LUMO peaks are located 1 to 2 eV above the Fermi energy. Because of spikiness of molecular states and a vanishing molecular PDOS in between the HOMO and LUMO peaks, it can be therefore concluded that the HOMO and LUMO peaks are well below and above the Fermi energy, respectively. Hence molecular donation and back-donation are negligible, compatible with the very small net molecular calculated Bader charges. Finer details of the weak moleculesurface interaction are provided by Figure 8, which displays the integrated local density of states (ILDOS) analysis of imidazole molecule before and after the adsorption on Cu(111); imidazole was chosen because it bonds the strongest to the surface. The integration of the 24194



ARTICLE

Figure 7. Charge density difference, ΔF(r) = Fmol/surf(r) Fsurf(r) Fmol(r), and density of states projected to the molecule (blue) and the Cu atom (brown) beneath it for the four considered molecules on Cu(111). Top panels show adsorption structures where the CH 3 3 3 Cu bonds develop, whereas bottom panels show adsorption structures with the NH 3 3 3 Cu bonds. The ΔF(r) plots are drawn with seven contours in linear scale from 0.006 to +0.006e/a30; the blue (red) color represents the electron deficit (excess) regions; that is, charge flows from blue to red regions.

individual PDOS peaks and visualization of the corresponding ILDOSes clearly reveal that molecular states are only weakly affected by adsorption because the ILDOS plots (after adsorption) are very similar to the square modulus of molecular orbitals of isolated molecule. This is a clear evidence of the weak moleculesurface interaction. It can be further noticed that only the 15σ (HOMO1) orbital forms a bonding state with the Cu. For this reason, it is considerably downshifted in energy upon adsorption, and its peak is located below the peak of 2π (HOMO2) state. Interestingly, the ILDOS analysis reveals that although either the HOMO or the LUMO is a π-type orbital (see Figure S1 in the Supporting Information, which shows the high-lying molecular orbitals of the four molecules), the moleculemetal interaction mainly involves the σ-type molecular orbitals. Nevertheless, the hybridization between the metal and molecular states has little effect on net bonding because the Cu has the d band completely below the Fermi energy. Instead, it enables the molecule to come into a close contact with the surface to gain in the electrostatic dipole interaction due to smaller dipoletoimage-dipole distance, dim. Indeed, as seen from Figure 6b, the main contribution to the adsorption energy comes from dipole interactions (given

by the E∞ ads E0 difference, cf. eq 3). In this respect, the molecular chemical hardness should also play an important role. Namely, the softer the molecule, the easier the hybridization with the metal states and the more facile the close contact between the molecule and the surface. The E0 values, which can be taken as a measure of covalent interaction, should therefore show proportionality to molecular hardnesses, which is indeed the case, as shown in Figure 9. To support further the hypothesis that chemical hardness is a more decisive molecular parameter than the electronegativity for determining the moleculesurface bond strength in the current case, we also considered the molecular adsorption on Al(111), and the corresponding results are presented in the Supporting Information. The Al(111) was chosen because it is less electronegative than all considered molecules. (The calculated work function of Al(111) is 4.17 eV.)24 On the other hand, the Cu(111) is more electronegative than imidazole and less electronegative than pentazole. Moreover, the bonding of adsorbates to Al(111) is not very different from that of Cu(111), provided that there is no charge back-donation from metal d states into the empty molecular orbitals,53,54 because both metals have only delocalized sp-type states around the Fermi energy. 24195



ARTICLE

Figure 8. Integrated local density of states (ILDOS) analysis of imidazole on Cu(111). Top panels show the PDOS and ILDOSes of the molecule 6 Å above the surface (before interaction), whereas bottom panels show the PDOS and ILDOSes of the adsorbed molecule.

Figure 9. Correlation between the adsorption energy and molecular absolute hardness; E0 versus ηmol (solid line) and E∞ ads versus ηmol (dashed line). The estimated accuracies of E∞ ads and E0 are indicated by error bars.

We indeed find that the bonding of the four azole molecules is similar on Cu(111) and Al(111): not only is the bonding trend the same in both cases, that is, imidazole > triazole > tetrazole > pentazole, but also the adsorption energies are very similar on the two metals. (See Table S1 in the Supporting Information.) The fact that the bonding trend is the same on Al(111) and Cu(111) and compatible with the trend of molecular chemical hardness despite the difference in their electronegativities (compared with molecular electronegativities) proves the above assertion about the role of chemical hardness.

4. CONCLUSIONS We have characterized the bonding of azole molecules onto Cu(111) and Al(111) surfaces and explained the role the azole nitrogen atoms play in the moleculesurface bonding. Their effect is two-fold; first their presence is the origin of the large molecular dipole moment, which provides the main contribution to molecule surface bonding, that is, self-induced dipole to image-dipole interaction. Such type of interaction is not important solely for azoles but can be expected for many highly polar molecules. This finding brings a new point of view to the interaction of azole molecules with metal surfaces, such as copper, which has been traditionally explained as being due to hybridization of molecular π-system or nitrogen lone pair orbitals with metal states. Second, with increasing the number of nitrogen atoms in azole ring the molecules become more electronegative and chemically harder, the latter resulting in diminished moleculesurface bond strength. In this respect, we have shown that the molecular reactivity indicators, emerging from the theoretical formalization of the HSAB principle, are valuable also in the context of molecular adsorption and can be utilized to explain the adsorption energy trends. ’ ASSOCIATED CONTENT

bS

Supporting Information. Description of the dependence of lateral dipoledipole interactions on the orientation of

24196


The Journal of Physical Chemistry C molecular dipole. A figure showing the square modulus of molecular orbitals of imidazole, triazole, tetrazole, and pentazole. Results about the adsorption of these azole molecules on Al(111). This material is available free of charge via the Internet at http://pubs.acs.org.

’ AUTHOR INFORMATION Corresponding Author

*Tel: +386-1-477-35-23. Fax: +386-1-477-38-22. E-mail: tone. [email protected]. URL: http://www-k3.ijs.si/kokalj/.

’ ACKNOWLEDGMENT This work has been supported by the Slovenian Research Agency (grant nos. J1-2240 and P2-0148). ’ REFERENCES (1) Antonijevic, M. M.; Milic, S. M.; Petrovic, M. B. Corros. Sci. 2009, 51, 1228–1237. (2) Antonijevic, M. M.; Petrovic, M. B. Int. J. Electrochem. Sci. 2008, 3, 1–28. (3) Kuznetsov, Y. I.; Kazansky, L. P. Russ. Chem. Rev. 2008, 77, 219–232. (4) Jiang, Y.; Adams, J. B. Surf. Sci. 2003, 529, 428–442. (5) Jiang, Y.; Adams, J. B.; Sun, D. J. Phys. Chem. B 2004, 108, 12851–12857. (6) Blajiev, O.; Hubin, A. Electrochim. Acta 2004, 49, 2761–2770. (7) Kokalj, A.; Peljhan, S. Langmuir 2010, 26, 14582–14593. (8) Kokalj, A.; Peljhan, S.; Finsgar, M.; Milosev, I. J. Am. Chem. Soc. 2010, 132, 16657–16668. (9) Gece, G. Corros. Sci. 2008, 50, 2981–2992. (10) Parr, R. G.; Pearson, R. G. J. Am. Chem. Soc. 1983, 105, 7512–7516. (11) Pearson, R. G. J. Am. Chem. Soc. 1963, 85, 3533–3539. (12) Sastri, V.; Perumareddi, J. Corrosion 1997, 53, 617–622. (13) Lukovits, I.; Kalman, E.; Zucchi, F. Corrosion 2001, 57, 3–8. (14) Martinez, S. Mater. Chem. Phys. 2003, 77, 97–102. (15) Martinez, S.; Stagljar, I. J. Mol. Struct. (THEOCHEM) 2003, 640, 167–174. (16) Lashkari, M.; Arshadi, M. R. Chem. Phys. 2004, 299, 131–137. (17) Khaled, K. F. Appl. Surf. Sci. 2008, 255, 1811–1818. (18) Finsgar, M.; Lesar, A.; Kokalj, A.; Milosev, I. Electrochim. Acta 2008, 53, 8287–8297. (19) Lesar, A.; Milosev, I. Chem. Phys. Lett. 2009, 483, 198–203. (20) Kokalj, A. Electrochim. Acta 2010, 56, 745–755. (21) Kovacevic, N.; Kokalj, A. Corros. Sci. 2011, 53, 909–921. (22) Crawford, P.; Hu, P. J. Phys. Chem. B 2006, 110, 4157–4161. (23) Crawford, P.; Hu, P. J. Phys. Chem. B 2006, 110, 24929–24935. (24) Kokalj, A. Chem. Phys. 2011, in press. DOI: 10.1016/j. chemphys.2011.10.021. (25) Kokalj, A. Phys. Rev. B 2011, 84, 045418. (26) Perdew, J. P.; Burke, K.; Ernzerhof, M. Phys. Rev. Lett. 1996, 77, 3865–3868. (27) Vanderbilt, D. Phys. Rev. B 1990, 41, 7892–7895. (28) Ultrasoft pseudopotentials (US PPs) for H, C, N, and Cu were taken from the Quantum Espresso PseudoPotential Download Page: http://www.quantum-espresso.org/pseudo.php (Files: H.pbe-rrkjus. UPF, C.pbe-rrkjus.UPF, N.pbe-rrkjus.UPF, and Cu.pbe-d-rrkjus.UPF. For Al, the norm-conserving pseudopotential was used, file: Al.pbe-rrkj. UPF). (29) Giannozzi, P.; Baroni, S.; Bonini, N.; Calandra, M.; Car, R.; Cavazzoni, C.; Ceresoli, D.; Chiarotti, G. L.; Cococcioni, M.; Dabo, I.; Corso, A. D.; de Gironcoli, S.; Fabris, S.; Fratesi, G.; Gebauer, R.; Gerstmann, U.; Gougoussis, C.; Kokalj, A.; Lazzeri, M.; Martin-Samos, L.; Marzari, N.; Mauri, F.; Mazzarello, R.; Paolini, S.; Pasquarello, A.;

ARTICLE

Paulatto, L.; Sbraccia, C.; Scandolo, S.; Sclauzero, G.; Seitsonen, A. P.; Smogunov, A.; Umari, P.; Wentzcovitch, R. M. J. Phys.: Condens. Matter 2009, 21, 395502. (30) Peljhan, S.; Kokalj, A. J. Phys. Chem. C 2009, 113, 14363–14376. (31) Kokalj, A.; Dal Corso, A.; de Gironcoli, S.; Baroni, S. Surf. Sci. 2002, 507510, 62–68. (32) Bengtsson, L. Phys. Rev. B 1999, 59, 12301–12304. (33) Methfessel, M.; Paxton, A. T. Phys. Rev. B 1989, 40, 3616–3621. (34) Monkhorst, H. J.; Pack, J. D. Phys. Rev. B 1976, 13, 5188–5192. (35) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.; Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian 09, revision A.1; Gaussian, Inc.: Wallingford, CT, 2009. (36) Kokalj, A. Comput. Mater. Sci. 2003, 28, 155168. Code available from: http://www.xcrysden.org/. (37) Dipoles oriented parallel to the surface and forming a perfect infinite lattice will experience no torque due to a self-generated electric field provided the lattice displays the mirror plane symmetry and the dipoles point along the mirror-plane symmetry direction. (38) Maschhoff, B. L.; Cowin, J. P. J. Chem. Phys. 1994, 101, 8138–8151. (39) Kokalj, A.; Kovacevic, N. Chem. Phys. Lett. 2011, 507, 181–184. (40) Perdew, J. P.; Parr, R. G.; Levy, M.; Balduz, J. L. Phys. Rev. Lett. 1982, 49, 1691–1694. (41) Cohen, M. H.; Wasserman, A. J. Phys. Chem. A 2007, 111, 2229–2242. (42) Ayers, P. W. J. Math. Chem. 2008, 43, 285–303. (43) Parr, R. G.; Yang, W. Density-Functional Theory of Atoms and Molecules; Oxford University Press: New York, 1989. (44) Parr, R. G.; Donnelly, R. A.; Levy, M.; Palke, W. E. J. Chem. Phys. 1978, 68, 3801–3807. (45) Iczkowski, R. P.; Margrave, J. L. J. Am. Chem. Soc. 1961, 83, 3547–3551. (46) Sanderson, R. T. Science 1951, 114, 670–672. (47) Although the electronegativity of metal surface is given by its work function, some other estimates, such as atomic Mulliken electronegativity, are frequently used instead.24 It has been, however, shown that for molecular adsorbates the work function is appropriate for usage in eqs 18a and 8b.24 As for the absolute hardness of bulk metals, it is related to the inverse of their density of states at Fermi energy,55 which is an extremely small number. (48) Bondi, A. J. Phys. Chem. 1964, 68, 441–451. (49) Peljhan, S.; Kokalj, A. Phys. Chem. Chem. Phys. 2011, 13, 20408–20417. (50) Pearson, R. G. Inorg. Chem. 1984, 23, 4675–4679. (51) Crabtree, R. H.; Eisenstein, O.; Sini, G.; Peris, E. J. Organomet. Chem. 1998, 567, 7–11. (52) Yao, W.; Eisenstein, O.; Crabtree, R. H. Inorg. Chim. Acta 1997, 254, 105–111. (53) Hammer, B.; Nørskov, J. K. Surf. Sci. 1995, 343, 211–220. (54) Hammer, B.; Nørskov, J. K. Adv. Catal. 2000, 45, 71–129. (55) Yang, W.; Parr, R. G. Proc. Natl. Acad. Sci. U.S.A. 1985, 82, 6723–6726.

24197


DFT Study of Interaction of Azoles with Cu(111) and Al(111) Surfaces

Recommend Documents