NEXAFS Spectroscopy of Homopolypeptides at All Relevant

Jul 31, 2007 - Carefully calibrated high-resolution low-noise near-edge X-ray absorption fine structure spectra of three homopolypetides, viz., polyis...
0 downloads 0 Views 92KB Size
J. Phys. Chem. B 2007, 111, 9803-9807

9803

NEXAFS Spectroscopy of Homopolypeptides at All Relevant Absorption Edges: Polyisoleucine, Polytyrosine, and Polyhistidine Yan Zubavichus,*,†,‡ Andrey Shaporenko,† Michael Grunze,† and Michael Zharnikov*,† Angewandte Physikalische Chemie, UniVersity of Heidelberg, INF 253, 69120 Heidelberg, Germany, and Institute of Organoelement Compounds, Russian Academy of Sciences, 28 VaViloVa st., 119991 Moscow, Russia ReceiVed: May 21, 2007; In Final Form: June 22, 2007

Carefully calibrated high-resolution low-noise near-edge X-ray absorption fine structure spectra of three homopolypetides, viz., polyisoleucine, polytyrosine, and polyhistidine at the C, N, and O K-edges, are compared with the respective spectra of parent amino acids and glycine-derived cyclic dipeptide, 2,5-diketopiperazine. An assignment of the spectral features related to the nitrogen and oxygen atoms constituting the peptide bond is suggested on the basis of a comparative analysis of the experimental spectra as well as theoretical calculations for 2,5-diketopiperazine within the real-space multiple-scattering formalism. A splitting of the π*-feature in the N K-edge spectra is identified, which is probably sensitive to the dominant conformation type of the peptide molecule (i.e., R-helix vs β-sheet).

1. Introduction Near-edge X-ray absorption fine structure (NEXAFS) spectroscopy is a widely recognized powerful tool to probe the electronic structure of organic substances in various environments.1,2 Despite the fact that the measurements are typically accomplished on vacuum-compatible solid samples, recent advancements in instrumentation enabled studies of liquids and solutions either utilizing ultrasonic liquid jets3-5 or relying on the design of a special cell with a thin membrane6,7 semitransparent for soft X-rays and separating a liquid phase from the ultrahigh vacuum. One of the current challenges in this field is the retrieval of chemical and molecular structure information on bioorganic objects, such as functional proteins, from the NEXAFS spectra. The interpretation of the respective experimental data is difficult for various reasons, including the complexity of the molecular composition of the target objects and their intrinsic radiation sensitivity. Nevertheless, numerous reports on successful applications of NEXAFS spectroscopy to studies of amino acids,8-17 small peptides,18-22 proteins,23-26 and even ordered bacterial layers27,28 have appeared during last several years. Since experimental data collection on individual proteins is often extremely challenging, it seems very promising to predict spectra of very complex proteins from those of their simpler constituents.8 Recently, we have reported a complete compendium of the NEXAFS spectra for 22 most common amino acids at all relevant absorption edges, viz., C, N, and O K-edges.17 A next logical step in the quest for the application of NEXAFS spectroscopy to functional proteins should be homopolypeptides. A homopolypetide is a relatively short sequence of one specific sort of amino acid residues joined into a unique molecule via the peptide bonds. Due to this architecture, these molecules are * To whom correspondence should be addressed. E-mails: [email protected] (M. Zharnikov) and yan. [email protected] (Y. Zubavichus). † University of Heidelberg. ‡ Institute of Organoelement Compounds.

only slightly more complex than their parent amino acids, but are much simpler than typical physiologically important peptides and proteins containing many different amino acids. However, the individual building blocks in homopolypetides are already linked with each other in a way typical of proteins. Therefore, homopolypeptides are ideal model objects for the optimization and verification of different experimental approaches and theoretical schemes aimed at the interpretation of fine chemical effects in the spectra of proteins. In this paper, we present the C, N, and O K-edge NEXAFS spectra of three homopolypeptides, viz., polyisoleucine, polytyrosine, and polyhystidine, and analyze spectral manifestations of the peptide bond formation by comparing the spectra of these molecules with those of their parent amino acids as well as with the spectra of 2,5-diketopiperazine, which is the cyclic glycinederived dipeptide and, thus, the smallest molecule containing only “peptide” nitrogen and oxygen atoms. The molecular structures of the amino acids and peptides under study are shown in Figure 1. We note that the target molecules represent distinct chemical subclasses of amino acids, viz., amino acids with aliphatic, aromatic, and heterocyclic side chains, respectively. 2. Experimental Section The as-purchased powders of the respective homopolypeptides (Sigma-Aldrich Chemie GmbH, stated purity >98%, molecular weights >10 kDa, optically pure L-form) were pressed into clean In foil and thinned by a brush to suppress charging effects.17,19 The measurements were performed at the bending magnet beamline HE-SGM of the synchrotron radiation facility BESSY II in Berlin. The energy resolution of the setup was about 0.3 eV; the incident beam was kept at the “magic angle”; i.e., the polar angle of incidence in the p-polarization geometry was kept at 55° with respect to the sample surface. Due to a special procedure for the absolute energy calibration of the experimental spectra against the distinct π*-resonance of highly oriented pyrolytic graphite (285.38 eV),29 the accuracy of the determined positions of the spectral features is estimated at (0.05 eV. The spectra were fully reproducible over several independent runs. No spectral changes due to radiation damage

10.1021/jp073922y CCC: $37.00 © 2007 American Chemical Society Published on Web 07/31/2007

9804 J. Phys. Chem. B, Vol. 111, No. 33, 2007

Zubavichus et al.

Figure 2. Experimental C K-edge NEXAFS spectra of polyisoleucine (a), polytyrosine (a), polyhistidine (c), and the respective pristine amino acids (gray lines). The difference between the spectrum of the amino acid and that of the respective peptide is depicted by dashed line.

Figure 1. Molecular formulas of the homopolypeptides and pristine amino acids used in this study.

within the data acquisition time were detected with the current setup (the photon flux on the order of 1011 photons/s at the spot size about 1.2 × 0.5 mm2). Raw NEXAFS spectra were normalized to the incident photon flux by division through the spectra of clean, freshly sputtered gold sample. This procedure corrected for the energy dependence of the photon flux and removed the artefacts arising by the absorption due to the contaminants containing carbon, nitrogen and oxygen on the optics of the beamline. Further, the resulting spectra were reduced to the standard form by subtracting linear preedge background and normalizing to the unity edge jump (determined by a nearly horizontal plateau 40-50 eV above the respective absorption edges). Theoretical NEXAFS spectra of “peptide” N and O atoms in 2,5-diketopiperazine were calculated using the cluster real-space multiple-scattering (RSMS) approach implemented in the FEFF8 code.30-34 Crystallographic data for 2,5-diketopiperazine reported in the literature35 were used to generate the input files. The scattering potential was calculated self-consistently for a sphere of 5.0 Å around the respective photoionized atom taking into account the presence of a dynamically screened 1s core hole. The full multiple-scattering (FMS) summation was applied to a sphere of 6.5 Å around the photoionized atom, and thus,

the respective probe clusters contained about 150 atoms. Atomic thermal vibrations were partly taken into account within the correlated Debye-Einstein model by assigning a Debye-Waller factor of 0.01 Å2 to all atoms in the probe cluster. FEFF outputs were consistently shifted by a few electronvolts to match the experimental energy scales. Note that it is known that the theoretical approach implemented in FEFF8 fails to precisely reproduce relative intensities and widths of low-lying π*resonances in the NEXAFS spectra of organic molecules, although their energy positions are predicted quite reliably. Nevertheless, we decided to use this code since it produces realistic whole-range simulated spectra that do not require any additional mathematical processing (e.g., manual broadening, summation, convolution, etc.) from the end-user to be compared with the experimental data. 3. Results and Discussion 3.1. C K-Edge Spectra. The C K-edge spectra of polyisoleucine, polytyrosine, and polyhistidine as well as of their parent amino acids and the respective difference spectra are shown in Figure 2. The spectrum of pristine isoleucine in Figure 2a shows three dominant features, viz., a narrow resonance at around 288.6 eV attributable9,17 to the π*(COO) transition and broader σ*resonances at ca. 293.0 and 301.0 eV. The former dominantly comprises σ* C-C and C-N components, whereas the latter corresponds to transitions to σ*-states associated with the carboxylate group. The spectrum of the respective homopolypeptide, polyisoleucine, is similar in general, but demonstrates some distinctions. First of all, the first narrow π*-resonance is slightly (by 20-25%) attenuated in peak intensity and shifted to lower photon energies by 0.4 eV. This clear manifestation of the peptide bond formation has been noted and discussed by several authors,8,18-20 and even a numerical procedure for its correction

NEXAFS Spectroscopy of Homopolypeptides for the simulation of spectra of peptides based on linear combinations of the spectra of the constituent amino acids has been elaborated recently.36 Nevertheless, the spectra in Figure 2a reveal that changes occur not only in the region of narrow π*-resonances, but rather broader σ*-resonances located at higher energies are also affected by the peptide bond formation. Primarily, the feature at 293.0 eV tends to shift to lower energies, whereas the feature at 301.0 eV nearly disappears due to smearing. Note that only one out of six carbon atoms in the isoleucine molecule is directly affected by the polycondensation to the homopolypeptide, whereas the chemical environment of five other carbon atoms remains unchanged. Therefore, we suggest that these rather strong spectral changes are mostly due to the coexistence of different carbon atoms, which are equivalent chemically but slightly different in terms of local geometrical parameters in the conformationally labile peptide molecule. The C K-edge spectrum of pristine tyrosine in Figure 2b reveals a very rich near-edge fine structure, which has been described in earlier papers.9,15,17 Essentially, the first two π*peaks at 285.1 eV (composed of two partly resolved components) and 287 eV arise due to the π*-antibonding states associated with the benzene ring (the latter feature is chemically shifted due to the OH substituent in the benzene ring). Another narrow feature at 288.6 eV, similar to the aforementioned spectrum of isoleucine, is attributed to the carboxylate π*resonance. The σ*-resonances in the spectrum of tyrosine are represented by a rather narrow peak at 290.6 eV (C-OH) and very broad peaks at 293.9 eV (C-C, C-N) and 302.3 eV (mainly CdO). Upon a transition to polytyrosine, the spectrum changes in a way very similar to the above case of isoleucine and polyisoleucine. The π*-peaks at 285.1 and 288.6 eV are attenuated in intensity by 25-30% and the latter shifts by 0.4 eV to lower energies. In the case of tyrosine, one out of nine carbon atoms is involved into the peptide bond formation. The C K-edge spectrum of histidine in Figure 2c is dominated by two π*-resonances at 286.9 and 288.6 eV associated with the imidazole and carboxylate π-systems, respectively. Both features are attenuated by ca. 20% in the spectrum of polyhistidine. Furthermore, the same changes as discussed above are observed at higher photon energies. The difference spectra visualizing the spectral changes upon the peptide bond formation are shown in Figure 2 as dashed lines. For the three homopolypeptide studied here, they are characterized by similar features, Viz., two minima at 288.1 and 291.2 eV and two maxima at 288.7 and 303 eV. 3.2. N K-Edge Spectra. The N K-edge spectra of polyisoleucine, polytyrosine, and polyhistidine are compared to those of their parent amino acids in Figure 3. For the two former peptides, the spectral changes due to the peptide bond formation are very prominent, since all nitrogen atoms present in the molecules are involved and thus change their chemical environment. In the case of polyhistidine, the changes associated with the peptide bond formation affect only one of the three nitrogen atoms, since the imidazole heterocycle remains intact. Essentially, in the case of isoleucine and tyrosine, the N K-edge spectra (panels a and b, respectively, of Figure 3) are represented by broad peaks centered at 406.5 eV, which can straightforwardly be attributed to the σ*(N-C) shape resonance. In the respective polypeptides, this feature is diminished in intensity nearly 2-fold, whereas two new narrow peaks at 401.1 and 402.7 eV and a broader peak at 412.5 eV arise. Such a spectral change can easily be rationalized in view of the chemical changes occurring with the nitrogen atom upon the peptide bond

J. Phys. Chem. B, Vol. 111, No. 33, 2007 9805

Figure 3. Experimental N K-edge NEXAFS spectra of polyisoleucine (a), polytyrosine (b), polyhistidine (c), and the respective pristine amino acids. The difference between the spectrum of the amino acid and that of the respective peptide is depicted by dashed line.

formation. When a former amino nitrogen atom is involved into a peptide bridge, a new C-N bond is formed, which is characterized by a partial double character due to the delocalization of the π-system over the carbamide group. Thus, the first narrow peak corresponds to the electron transition from the N1s core state to the common π*-orbital of the carbamide group, whereas the broad peak at high photon energies is attributed to the σ*-shape resonance associated with the partly double shortened N-C bond. The appearance of the second narrow π*-like resonance at a relatively low energy is rather surprising and will be discussed in somewhat more detail below. In the case of histidine, the N K-edge NEXAFS spectrum (Figure 3c) is dominated by two π*-resonances at 399.7 and 401.3 eV associated with the imidazole π-system. Spectral changes observed upon a transition to polyhistidine are not as prominent as for the two other peptides and can be described as an intensity redistribution of already existing spectral features. Nevertheless, the difference spectrum “his-polyhis” very closely resembles the respective difference spectra for isoleucine- and tyrosine-based peptides: in all three cases, the curves reveal minima at about 401, 402.5, and 414-415.0 eV and a maximum at ∼406 eV. 3.3. O K-Edge Spectra. The O K-edge NEXAFS spectra of polyisoleucine, polytyrosine, and polyhistidine are shown in Figure 4, together with the spectra of the respective amino acids. The spectra of pristine amino acids reveal a narrow π*resonance at ca. 532.5 eV associated with the carboxylate group and a broad irregularly shaped peak spanning from 535 to 550 eV, which is apparently split in the case of isoleucine (Figure 4a) into two components at 539 and 544 eV. A few characteristic changes are observed in the spectra of polypeptides. The dominant π*-resonance is shifted to lower energies by ca. 0.3 eV in a way similar to the C K-edge spectra considered above. Furthermore, the shape of the broader σ*-resonance changes

9806 J. Phys. Chem. B, Vol. 111, No. 33, 2007

Figure 4. Experimental O K-edge NEXAFS spectra of polyisoleucine (a), polytyrosine (b), polyhistidine (c), and the respective pristine amino acids. The difference between the spectrum of the amino acid and that of the respective peptide is depicted by dashed line.

Figure 5. N (left panel) and O (right panel) K-edge NEXAFS spectra of the nitrogen and oxygen atoms constituting the peptide bond.

so that its maximum shifts from 544 to 539 eV. Despite some differences in the spectra of both pristine amino acids and homopolypeptides, the difference spectra are rather similar for all three samples exhibiting a minimum at ∼531.6 eV and two maxima at 532.5 and 543.6 eV. 3.4. NEXAFS Spectral Features of the Peptide Bond. The nitrogen and oxygen K-edge NEXAFS spectra of molecules comprising only “peptide” N and O atoms (i.e., polyisoleucine, polytyrosine, and 2,5-diketopiperazine for the N K-edge spectra and polyisoleucine, polyhistidine, and 2,5-diketopiperazine for the O K-edge spectra) are shown in Figure 5. Nominally, the spectra of these samples have to be very similar since they are originated from chemically equivalent nitrogen and oxygen atoms constituting the peptide bond. The experimental spectra are indeed rather similar, but not identical. First of all, the spectral features of 2,5-diketopiperazine are substantially narrower and more intense than their counterparts in the spectra of homopolypeptides. As has been suggested above, this can be due to the “entropic” factor, i.e., due to the coexistence of a

Zubavichus et al. large number of chemically equivalent atoms within an individual molecule, which, however, experience slightly different intramolecular (as defined by the exact molecular conformation) and intermolecular (H-bonding, etc.) environments. This factor effectively removes the equivalency of the nitrogen and oxygen atoms, which gives rise to intensity losses and broadening of the spectral features. In addition to the experimental spectra, Figure 5 also shows the results of theoretical simulation of the N and O K-edge NEXAFS spectra for 2,5-diketopiperazine using the FEFF8 code together with an assignment of the major features. The theoretical simulation reproduces all essential details of the experimental spectra reasonably well. Noteworthy, the N and O K-edge spectra are rather similar to each other in terms of the number, relative positions and intensities of features, despite their local atomic environments are distinctly different. Indeed, nitrogen is bonded with two carbon atoms, forming a formally single and a partly double bonds, and with a hydrogen atom. Consistently, the three major spectral features are attributed to the π*(N-CO), σ*(N-CH2), and σ*(N-CO) electronic transitions. Since the theoretical approach employed does not use explicit molecular orbital representation, this assignment was made using an original approach described in detail elsewhere.16 It was based on a series of simulation of polarization-dependent spectra for several representative directions of the polarization vector within the internal coordinate system of the model cluster. In particular, the π*(N-CO) component is maximized in intensity when the polarization vector is perpendicular to the plane of the carbamide group. The σ*(N-CH2) and σ*(N-CO) components are maximized when the polarization is parallel to the N-CH2 and N-CO bonds, respectively. Meanwhile, the oxygen atom is bound to only one carbon atom via the double bond delocalized over the carbamide group. It is surprising that such a simple structure gives rise to this rather complex spectral pattern revealing at least three distinct components with some even finer structure. It suggests that the σ-electron density is also delocalized over the carbamide group in a way similar to the π-electron density. The entire region of σ*-resonances is formally attributed to the σ*(CdO) contribution, since both apparent components are maximized when the polarization is directed parallel to this bond. The most significant difference between the N K-edge spectra of the homopolypeptides (i.e., polyisoleucine and polytyrosine) on one side and the experimental and simulated spectra of 2,5diketopiperazine on the other consists of the appearance of a narrow π*-like resonance at 402.7 eV for the former pair. Although its origin is not completely understood for the moment, we suggest that this difference can reflect different molecular conformations of the molecules. The homopolypeptides are expected to adopt the R-helical conformation, which is typical of the majority of proteins, in contrast to the ideally planar geometry of the 2,5-diketopiperazine ring (which, in turn, is analogous to the protein secondary structure elements, such as β-sheets and turns). The exact geometry, in particular, deviations of the carbamide group from the exact planarity strongly affects the extent of π-delocalization. 4. Summary Summarizing all the experimental evidence given above, we conclude that the peptide bond formation upon a transition from an amino acid to the respective homopolypeptide gives rise to several distinct changes in the NEXAFS spectra. In the case of the carbon K-edge spectra, although the peptide contribution is masked by the signal coming from carbon atoms in the side

NEXAFS Spectroscopy of Homopolypeptides chain, the dominant carboxylate π*-resonance shifts by ca. 0.4 eV to lower energies and the intensities of all spectral features are diminished by 20-30%. The nitrogen K-edge spectra change dramatically in terms of both strong intensity redistribution and the emergence of new narrow π*-features. The oxygen K-edge spectra change in way similar to the C K-edge spectra (i.e., a shift by 0.3 eV plus intensity attenuation) and the resultant peptide spectrum becomes similar to the N K-edge spectrum. Note that changes occurring upon the formation of peptides from individual amino acids take place not only in the region of narrow π*-resonances, but rather broader σ*-resonances located at higher energies are also affected by the peptide bond formation. We believe that these results will help to improve on the procedures of empirical simulation of protein spectra within the building-block approach. Acknowledgment. We are grateful to Ch. Wo¨ll (Universita¨t Bochum) for providing us with the experimental equipment at BESSY II and the BESSY II staff for their technical support during beamtime. This work was supported by the German BMBF (projects No. 05 KS4VHA/4 and 05 KS4WWA/6), the Russian Foundation for Basic Research (grant 05-03-32871), Fonds der Chemischen Industrie (M.G.), and the Office of Naval Research (Y.Z.). References and Notes (1) Sto¨hr, J. NEXAFS Spectroscopy; Springer-Verlag: Berlin, 1992. (2) Hahner, G. Chem. Soc. ReV. 2006, 35, 1244. (3) Messer, B. M.; Cappa, C. D.; Smith, J. D.; Drisdell, W. S.; Schwartz, C. P; Cohen R. C; Saykally R. J. J. Phys. Chem. B 2005, 109, 21640. (4) Messer, B. M.; Cappa, C. D.; Smith, J. D.; Wilson, K. R.; Gilles, M. K.; Cohen, R. C.; Saykally, R. J. J. Phys. Chem B 2005, 109, 5375. (5) Messer, B. M.; Cappa, C. D.; Smith, J. D.; Drisdell, W. S.; Schwartz, C. P.; Cohen, R. C.; Saykally, R. J. J. Phys. Chem. B 2005 109 21640. (6) Guo, J.-H.; Luo, Y.; Augustsson, A.; Rubensson, J.-E.; Såthe, C.; Ågren, H.; Siegbahn, H.; Nirdgren, J. Phys. ReV. Lett. 2002, 89, 1374021. (7) Fuchs, O.; Zharnikov, M.; Weinhardt, L.; Blum, M.; Weigant, M.; Zubavichus, Y.; Ba¨r, M.; Maier, F.; Denlinger, J. D.; Heske, C.; Grunze, M.; Umbach, E. Phys. ReV. Lett., submitted for publication. (8) Boese, J.; Osanna, A.; Jacobsen, C.; Kirz, J. J. Electron Spectrosc. 1997, 85, 9. (9) Kaznacheyev, K.; Osanna, A.; Jacobsen, C.; Plashkevych, O.; Vahtras, O.; Ågren, H.; Carravetta, V.; Hitchcock, A. P. J. Phys. Chem. A 2002, 106, 3153.

J. Phys. Chem. B, Vol. 111, No. 33, 2007 9807 (10) Hasselstro¨m, J.; Karis, O.; Weinelt, M.; Wassdahl, N.; Nilsson, A.; Nyberg, M.; Pettersson, L. G. M.; Samant, M. G.; Sto¨hr, J. Surf. Sci. 1998, 407, 221. (11) Nyberg, M.; Hasselstro¨m, J.; Karis, O.; Wassdahl, N.; Weinelt, M.; Nilsson, A.; Pettersson, L. G. M. J. Chem. Phys. 2000, 112, 5420. (12) Tanaka, M.; Nakagawa, K.; Koketsu, T.; Agui, A.; Yaokoya, A. J. Synchrotron. Radiat. 2001, 8, 1009. (13) Tzvetkov, G.; Koller, G.; Zubavichus, Y.; Fuchs, O.; Casu, M. B.; Heske, C.; Umbach, E.; Grunze, M.; Ramsey, M. G.; Netzer, F. P. Langmuir 2004, 20, 10551. (14) Zubavichus, Y.; Fuchs, O.; Weinhardt, L.; Heske, C.; Umbach, E.; Denlinger, J. D.; Grunze, M. Rad. Res. 2004, 161, 346. (15) Zubavichus, Y.; Zharnikov, M.; Shaporenko, A.; Fuchs, O.; Weinhardt, L.; Heske, C.; Umbach, E.; Denlinger, J. D.; Grunze, M. J. Phys. Chem. A 2004, 108, 4557. (16) Zubavichus, Y.; Shaporenko, A.; Grunze, M.; Zharnikov, M. J. Phys. Chem. B 2006, 110, 3420. (17) Zubavichus, Y.; Shaporenko, A.; Grunze, M.; Zharnikov, M. J. Phys. Chem. A 2005, 109, 6998. (18) Gordon, M. L.; Cooper, G.; Morin, C.; Araki, T.; Turci, C. C.; Kaznatcheev, K.; Hitchcock, A. P. J. Phys. Chem. A 2003, 107, 6144. (19) Zubavichus, Y.; Zharnikov, M.; Schaporenko, A.; Grunze, M. J. Electron Spectrosc. 2004, 134, 25. (20) Cooper, G.; Gordon, M.; Tulumello, D.; Turci, C.; Kaznatcheev, K.; Hitchcock, A. P. J. Electron Spectrosc. 2004, 137-140, 795. (21) Wilks, R. G.; MacNaughton, J. B.; Kraatz, H.-B.; Regier, T.; Moewes, A. J. Phys. Chem. B 2006, 110, 5955. (22) Polzonetti, G.; Battocchio, C.; Lucci, G.; Dettin, M.; Gambaretto, R.; Di Bello, C.; Carravetta, V. Mater. Sci. Eng. C 2006, 26, 929. (23) Liu, X.; Jang, C.-H.; Zheng, F.; Ju¨rgensen, A.; Denlinger, J. D.; Dickson, K. A.; Raines, R. T.; Abbott, N. L.; Himpsel, F. J. Langmuir 2006, 22, 7719. (24) Cruz, D. H.; Rousseau, M.-E.; West, M. M.; Pe´zolet, M.; Hitchcock, A. P. Biomacromolecules 2006, 7, 836. (25) Li, L.; Hitchcock, A. P.; Robar, N.; Cornelius, R.; Brash, J. L.; Scholl, A.; Doran, A. J. Phys. Chem. B 2006, 110, 16763. (26) Azioune, A.; Siroti, F.; Tanguy, J.; Jouini, M.; Chehimi, M. M.; Miksa, B.; Slomkowski, S. Electrochim. Acta 2005, 50, 1661. (27) Vyalikh, D. V.; Danzenba¨cher, S.; Mertig, M.; Kirchner, A.; Pompe, W.; Dedkov, Yu. S.; Molodtsov, S. L. Phys. ReV. Lett. 2004, 93, 238103. (28) Vyalikh, D. V.; Kirchner, A.; Danzenba¨cher, S.; Dedkov, Yu. S.; Kade, A.; Mertig, M.; Molodtsov, S. L. J. Phys. Chem. B 2005, 109, 18620. (29) Batson, P. E. Phys. ReV. B 1993, 48, 2608. (30) Ankudinov, A. L.; Ravel, B.; Rehr, J. J.; Conradson, S. D. Phys. ReV. B 1998, 58, 7565. (31) Ankudinov, A. L.; Rehr, J. J. Phys. ReV. B 2000, 62, 2437. (32) Ankudinov, A. L.; Bouldin, C. E.; Rehr, J. J.; Sims, J.; Hung, H. Phys. ReV. B 2002, 65, 104107. (33) Rehr, J. J.; Ankudinov, A. L. Int. J. Quantum Chem. 2003, 95, 487. (34) Rehr, J. J.; Ankudinov, A. L. Coord. Chem. ReV. 2005, 249, 131. (35) Degeilh, R.; Marsh, R. E. Acta Crystallogr. 1959, 12, 1007. (36) Stewart-Ornstein, J.; Hitchcock, A. P.; Herna`ndez-Cruz, D.; Henklein, P.; Overhage, J.; Hilpert, K.; Hale, J. D.; Hancock, R. E. W. J. Phys. Chem. B 2007, 111, 7691.