This contribution presents the results of an X-ray photoelectron spectroscopy (XPS) study aimed at gaining further insight into the interactions between Si, Ti, TiN, and Pt. Our approach involves the sequential deposition of Ti, TiN, and Pt onto a clean, well-characterized Si(111) surface, a substrate known to react with Pt . These depositions were carried out in situ under ultra-high vacuum (UHV) in a chamber equipped to perform angle-resolved XPS (ARXPS) and various other types of characterization. It will be shown that layers of TiN only ca. 1 nm thick are sufficient to prevent interdiffusion between Pt and Si.

The electrochemical properties of carbonaceous films prepared by catalytic chemical vapor deposition on clean Ni(111) using an equimolar CO:CH4 mixture at a total pressure ca. 2 x 10(-5) kPa (0.133 kPa = 1 Torr) as carbon source have been examined under ultrahigh vacuum (UHV) conditions (6.7 x 10(-11) kPa) in pre-electrolyzed LiClO4-poly(ethylene) oxide, (PEO), solutions (1 Li+:36 ethylene oxide units) at ca. 55degreesC. All steps involved were carried out in the same UHV chamber under highly controlled conditions without exposure to the ambient atmosphere. Despite these precautions, the first scan for the supported carbon film, C/Ni(111), in the negative direction initiated at similar to2.0 V vs. Li/Li+ extending down to 0.05 V, yielded a total negative charge exceeding by about one-third that observed in the subsequent scan in the positive direction. This irreversible capacity loss was much reduced during the second and subsequent cycles. Nevertheless, well-defined cyclic voltammetric peaks could be clearly discerned over the featureless background observed for clean Ni(111) in the potential region 1.2 < E < 0.05 V vs. Li/Li+. Based on coulometric data and the total amount of carbon in the film as determined from AES analysis the Li+/C stoichiometric ratio was estimated to be 1:3. (C) 2004 The Electrochemical Society.

Adsorption of small amounts of D(2)O (ca. 0.01 L exposure) on CO(sat)/Pt(100) surfaces in ultrahigh vacuum at 105 K was found to split the asymmetric peak at 2100 cm(-1) in the infrared reflection absorption spectra attributed to the stretching of CO adsorbed on atop sites into two clearly defined features: an intense component, which shifted toward lower energies due to surface hydration of adsorbed CO (originally at 2100 cm(-1), peak a), and a smaller peak centered at 2094 cm(-1) (peak b), which remained fixed in position even after closing the D(2)O dosing valve. The energies of peaks a and b, as determined by statistical analyses, correlated very well with those reported in the literature for CO adsorbed at high coverages on Pt(100) originally in the (5 x 20) or (hex) reconstruction, and on the unreconstructed Pt(100)-(1 x 1), respectively, at 90 K. On these bases, the asymmetry of the peak observed for CO(sat)/Pt(100) (no D(2)O dosing) is ascribed to the presence of CO linearly adsorbed on these two different sites on the surface, for which the rate of hydration is larger for the (5 x 20) compared to the (1 x 1) phases.

The vibrational spectra of CO adsorbed on Ru-modified Pt(100) surfaces prepared by chemical vapor deposition (condensation of Ru(3)(CO)(12) at 105 K followed by X-ray irradiation and thermal decomposition at 650 K in ultrahigh vacuum, UHV) was investigated by time-resolved infrared reflection absorption spectroscopy (IRAS) in UHV. Spectra were recorded while Ru/Pt(100) bimetallic surfaces (theta(Ru) = 0.24 and 0.52 by X-ray photoelectron spectroscopy, XPS) were dosed with gas-phase CO. Analysis of the data revealed that for a wide range of calibrated CO exposures, the linear CO-stretching region displays two features: a higher energy peak (2085-2100 cm(-1)), attributed to CO adsorbed on pristine Pt(100) sites, and a lower energy peak (2066-2092 cm(-1)), ascribed to adsorption of CO on sites on the surface induced by the presence of Ru. Similar experiments were performed on bimetallic specimens annealed repeatedly in UHV to 650 K to promote partial Ru dissolution into the lattice and thus render surfaces gradually enriched in Pt. For all surfaces and CO exposures examined, the total integrated area under the two CO spectral features remained fairly constant and equal in value to the corresponding areas found for bare Pt(100). If it is assumed that a fixed exposure leads to a fixed coverage on both bare and Ru-modified Pt(100)surfaces, and the thermal treatment leads to an exchange of Ru by Pt sites without altering significantly the total number of metal sites on the surface, the absorption cross sections for both of these peaks are virtually the same.

Changes in the properties of CO adsorbed at saturation coverages on Pt(100) induced by subsequent coadsorption of fixed amounts of D2O at 105 K in ultrahigh vacuum (UHV) were monitored by time-resolved infrared reflection absorption spectroscopy (tr-IRAS). The linear- and bridge-bonded CO stretching features were found to change in intensity and shift toward lower energies as a function of time at fixed CO and D2O coverages. Also observed was the development of multiple features in both CO spectral regions depending on the amount of D2O on the surface. These findings indicate that, under the conditions of these experiments, the interfacial dynamics are relatively slow, on the order of minutes, involving a gradual rearrangement of adsorbed CO and D2O on the surface to yield surface solvated CO, as has been suggested in the literature (Kizhakevariam et al. J. Chem. Phys. 1994, 100, 6750). This factor should be considered when comparing, quantitatively, shifts induced by water coadsorption with CO on Pt single crystals in UHV with CO adsorption on the same surfaces in electrochemical environments.

The voltammetric characteristics of polycrystalline Au and W electrodes cleaned (thermal annealing at 1100 K) and characterized (Auger electron spectroscopy) in ultrahigh vacuum (UHV) have been examined in ultrapure AlCl3/1-ethyl-3-methylimidazolium chloride (EtMeImCl) melts in UHV. These experiments were performed using a custom-designed transfer system that allows for the all-Al electrochemical cell to be filled with EtMeImCl in an auxiliary UHV chamber and later transferred under UHV to the main UHV chamber that houses the Auger electron spectrometer. The results obtained for the underpotential (UPD) and bulk deposition of Al on Au were found to be very similar to those reported in the literature for measurements carried out under 1 atm of an inert gas in a glovebox. For the far more reactive W surfaces, voltammetric features ascribed to the stripping of underpotential-deposited Al could be observed following a single scan from 1.0 V vs Al3+/Al to a potential negative enough for bulk deposition of Al to ensue. This behavior is unlike that reported in the literature for experiments performed in a glovebox, which required either extensive potential cycling in the Al bulk deposition and stripping region or excursions to potentials positive enough for chlorine evolution to ensue for Al UPD features to be clearly discerned. These observations open new prospects for fundamental electrochemical studies of well-characterized, highly reactive metals, including single crystals, in a variety of low vapor pressure ionic liquids.

The mode of bonding of CO to Ru-modified Pt(100) surfaces prepared by chemical vapor deposition was investigated using infrared reflection absorption spectroscopy (IRAS) in ultrahigh vacuum (UHV). Spectra recorded in the CO stretching region for freshly prepared Ru(theta(Ru) ca. 0.52)/Pt(100) exposed to saturation coverage of CO displayed, in addition to a rather weak peak ascribed to CO bound to bridging sites, a single asymmetric peak centered at ca. 2092 cm(-1). Statistical analyses of IRAS spectra recorded in experiments in which a Ru(theta(Ru) ca. 0.52)/Pt(100) surface was annealed sequentially to 650 K for ca. 2.5 h prior to a saturation exposure of CO yielded two peaks at 2100 cm(-1) and 2092 cm(-1) with a clearly identifiable isosbestic point, characteristic of a quantitative conversion of one type of species into the other. This behavior has been attributed to a thermally induced surface incorporation of Ru into Pt, as has been shown with low-energy ion-scattering spectroscopy (LEIS) for other Ru-modified low-index single-crystal Pt surfaces (Davies et al. Electrochim. Acta 1998, 44, 1181), which effectively enriches the surface with Pt while reducing the Ru coverage.

The nature of the reaction products generated by exposure of a model cyclic ester, gamma-butyrolactone (gamma-BL), two cyclic ethers, dioxalane (DIOX) and tetrahydrofuran (THF), and a cyclic alkyl carbonate, propylene carbonate (PC), toward metallic Li at room temperature was examined in ultrahigh vacuum by infrared reflection absorption, Auger electron (AES), and X-ray photoelectron (XPS) spectroscopies. The features observed in the spectra were consistent with Li butyrate and Li beta-keto ester in the case of gamma-BL, and a dialkoxide derivative of Li with small amounts of Li oxide for experiments involving PC as the main reaction products. The behavior of both these solvents differs greatly from that found of the cyclic ethers DIOX and THF, for which the XPS spectra yielded small amounts of Li oxide and Li carbide as the only reaction products. These two species are most likely not derived from reactions involving DIOX or THF but rather from reactions between metallic Li and CO desorbed from the walls of the chamber during prolonged dosings. The extent of the reaction involving gamma-BL, as evidenced by the loss of the AES peak at 52 eV characteristic of metallic Li, far exceeded that observed with PC and the linear alkyl carbonates reported previously in this laboratory, for which the aforementioned feature could be discerned clearly even after extensive exposure. All these findings agree with those reported by Aurbach et al.(1) for experiments involving Li metal exposure to the same solvents in liquid phase, which suggests that conclusions drawn from ex situ experiments of the type originally devised by our research group may also be applicable to reactions in condensed phase, including electrochemical environments. (C) 2003 The Electrochemical Society.

The vibrational properties of vapor-deposited lithium films exposed to propylene carbonate (PC) vapors at room temperature have been examined by infrared reflection-absorption spectroscopy (IRAS) in ultrahigh vacuum (UHV) using a specially designed chamber. The sharply defined spectral features observed were found to be incompatible with those of either, lithium propoxide, lithium alkyl carbonate, lithium carbonate, or mixtures thereof. Instead, the IRAS spectrum of PC/Li/Ni was found to be identical to that obtained by exposing Li to 1,2-propanediol vapors under otherwise identical conditions and may thus be attributed to the corresponding alkoxide derivative.

The infrared reflection absorption spectrum of clean lithium films supported on a polycrystalline Ni surface exposed to propylene carbonate (PC) vapors at room temperature has been recorded using a specially designed ultrahigh vacuum (UHV) chamber. The sharply defined spectral features observed were found to be incompatible with those of either, lithium propoxide, lithium alkyl carbonate, lithium carbonate or mixtures thereof. This observation is at variance with conclusions drawn by other workers based on X-ray photoelectron spectroscopy (XPS) of PC films condensed at cryogenic temperatures onto Li films supported on metal substrates, which had been heated and characterized by XPS in stages to ca. 300 K. Instead, the few infrared peaks in the 1000 -1600 cm(-1) region appear to be associated with more complex alkoxide species. Furthermore, the O(1s) and C(1s) XPS spectra of Li/Ni(poly) exposed to 1-propanol at ca. 300 K, to generate lithium propoxide, were virtually identical to those observed for the same substrate exposed to PC under the same conditions, indicating that XPS lacks sufficient specificity to discriminate between various Li alkoxide species.

X-ray photoelectron spectroscopy (XPS) was used to examine the reactivity of metallic Li toward gas-phase symmetric and asymmetric alkyl linear carbonates of the form ROCO 2 R' (R and R' = -CH 3 , -CH 2 CH 3 ) at room temperature in ultrahigh vacuum (UHV). Comparison of the C (1s) and O (1s) XPS features obtained in virtually identical experiments involving methanol and ethanol allowed LiOCH 3 and LiOCH 2 CH 3 to be clearly identified as the major products of the reaction between Li and dimethyl (DMC) and diethyl (DEC) carbonates, respectively, and a mixture of the two alkoxides for ethylmethyl (EMC) carbonate. These findings are in agreement with UHV Fourier transform infrared studies of the same systems reported earlier in our laboratory. Also found in the XPS spectra for ROCO 2 R'/Li interfaces was an O (Is) peak at 529.6 eV attributed to Li oxide. In the case of DEC and EMC, evidence was obtained for the presence of an additional O (Is) feature at 534.76 eV attributed to Li ethyl carbonate. This behavior was unlike that found for DMC for which no signals due to Li methyl carbonate could be discerned. The conclusions emerging from this study further support the general mechanism for Li reactivity toward ROCO 2 R' suggested by Aurbach and co-workers.

X-ray photoelectron spectroscopy (XPS) was used to examine the reactivity of metallic Li toward gas-phase symmetric and asymmetric alkyl linear carbonates of the form ROCO2R' (R and R' = -CH3, -CH2CH3) at room temperature in ultrahigh vacuum (UHV). Comparison of the C (1s) and O (1s) XPS features obtained in virtually identical experiments involving methanol and ethanol allowed LiOCH3 and LiOCH2CH3 to be clearly identified as the major products of the reaction between Li and dimethyl (DMC) and diethyl (DEC) carbonates, respectively, and a mixture of the two alkoxides for ethylmethyl (EMC) carbonate. These findings are in agreement with UHV Fourier transform infrared studies of the same systems reported earlier in our laboratory. Also found in the XPS spectra for ROCO2R'/Li interfaces was an O (1s) peak at 529.6 eV attributed to Li oxide. In the case of DEC and EMC, evidence was obtained for the presence of an additional O (1s) feature at 534.76 eV attributed to Li ethyl carbonate. This behavior was unlike that found for DMC for which no signals due to Li methyl carbonate could be discerned. The conclusions emerging from this study further support the general mechanism for Li reactivity toward ROCO2R' suggested by Aurbach and co-workers. (C) 2002 The Electrochemical Society.

The reactivity of copper toward liquid anhydrous hydrofluoric acid (AHF) has been examined using ex situ X-ray photoelectron spectroscopy (XPS). Exposure of either oxide-free or air-oxidized Cu surfaces to AHF yielded XPS spectra consistent with the presence of a CuF2 layer devoid of oxygen. Cyclic voltammetric experiments involving oxide-free Cu electrodes in hexafluorobutanol (HFB)+AHF displayed a prominent oxidation peak during the first few cycles. XPS analysis of Cu electrodes emersed at potentials positive to this feature provided evidence for the presence of a CuF2 layer much thicker than that found by simple immersion in AHF. Furthermore, the onset potential for both oxidation and subsequent reduction of the film was 0.0 V versus a Cu electrode in the same solution, which suggests that the Cu/CuF2 redox couple controls the potential of the reference electrode.

The influence of substrate defect sites on the morphology of Pt vapor deposited on the basal plane of highly oriented pyrolytic graphite (HOPG(bp)) has been investigated in situ (in ultrahigh vacuum (UHV)) by X-ray photoelectron spectroscopy (XPS), and ex situ by high-resolution scanning electron microscopy (HRSEM), and scanning tunneling microscopy (STM). Defects were introduced on selected sections of HOPG(bp) surfaces via Ar+ bombardment by employing specially designed masks. The effect of such defects on the electronic and structural nature of the Pt clusters formed was examined by evaporating the metal simultaneously on both damaged and nominally pristine sections of HOPG(bp). HRSEM and STM images of Pt deposits formed at 300 K on pristine HOPG(bp) were found to display three-dimensional fractal-type aggregates. Image analysis of such Pt clusters gave a fractal dimension DL=1.7 for Pt on Pt, using area (A)/perimeter (P) relations, and DL=1.5 for Pt on HOPG(bp), based on the count box method. Possible origins of this difference include higher cluster edge atom diffusion rates for Pt in the first monolayer (ML) and/or small cluster mobilities as in the model investigated by Meakin [Phys. Rev. Lett. 51 (1983) 1119.]. This overall behavior was unlike that observed for Pt vapor deposited on sections of Ar+-bombarded HOPG(bp) surfaces, which yielded instead clusters roughly circular in shape, ca. 1 nm in diameter, and one to two atomic layers high. Such disparate morphologies, however, are in agreement with models proposed in the literature which predict formation of highly ramified structures for diffusion limited aggregation (DLA) on substrates with low density of defect sites and small round clusters when the defect density is increased. Also studied in this work were the effects of post-deposition thermal annealing on the nature of the deposits.

Optically smooth Zn films supported on Cu-coated glass and quartz substrates have been obtained by physical vapor deposition of Zn in metallic form. The method employed involves resistive heating of a Mo boat tilled with high purity Zn shot in an Ar atmosphere at pressures of about 3-5 mTorr. Atomic force microscopy images revealed that the resulting Zn deposits consist of smooth features (rms roughness ca. 0.3 nm) with dimensions on the order of 200 nm. Preliminary results indicate that the electrochemical behavior of these films in strongly alkaline solutions is somewhat different than that observed for Zn in bulk commercial form. (C) 2001 The Electrochemical Society. [DOI: 10.1149/1.1376120] All rights reserved.

Low-energy electron diffraction (LEED), X-ray photoelectron spectroscopy (XPS), and GO-temperature; programmed desorption (TPD) were used to characterize ruthenium modified Pt(100) surfaces of very high purity and controlled stoichiometry prepared in ultrahigh vacuum (UHV) by irradiating Ru-3(CO)(12) films condensed on cold Pt substrates at 150 K with X-rays, and subsequent annealing at ca. 620 K. The presence of Ru an Pt(100) lifted the (5 x 20) reconstruction characteristic of the bare clean substrate; however, the reconstruction reappeared as-the bimetallic surfaces were briefly annealed to ca. 900 K. Exposure of nonannealed Ru(theta (Ru) greater than or equal to 0.22)/Pt(100), where theta (Ru) represents the Ru coverage in monolayers, to large exposures of CO at ca. 200 K yielded smaller theta (CO), as well as TPD peaks with onset desorption temperatures, T-des(CO), ca. 50 K lower than those observed for bare Pt(100). More strikingly, however, the CO-TPD spectra of CO-saturated Ru(theta (Ru) = 0.42)/Pt(100), which had been briefly annealed to 900 K, displayed T-des(CO) as low as 250 K, very similar to desorption temperatures reported for Pt-modified Ru(0001) by de Mongeot et al.

Certain aspects of the electrochemical fluorination (ECF) of hexafluorobutanol (HFB) in anhydrous HF (AHF) on Ni electrodes have been examined by mass spectrometry (MS) and X-ray photoelectron spectroscopy (XPS). Measurements were performed using a portable ultrahigh vacuum compatible chamber that allows for the transfer and characterization of specimens without exposure to the ambient atmosphere. Quasi on line MS analysis of the gases released during ECF of HFB/AHF solutions revealed features attributable to perfluorobutyryl fluoride [PBF = CF3CF2CF2C(O)F] with no evidence of the presence of other reaction products. Ex situ XPS analysis of Ni electrodes following ECF yielded spectra consistent with the formation of a rather thick, irregular layer of NiF2 of average thickness of about 150-200 monolayers, with small contributions due to O (6% ) and C (7%) localized primarily in the surface region. (C) 2001 The Electrochemical Society.

The reactivity of nominally clean polycrystalline Ni toward gas-phase and liquid anhydrous hydrofluoric acid (AHF) has been examined by X-ray photoelectron spectroscopy (XPS). To avoid contamination with atmospheric components, experiments were carried out using a portable ultrahigh vacuum (UHV)-compatible chamber to transfer Ni specimens between a UHV system that houses the XPS spectrometer and an additional UHV-compatible chamber where the actual exposures were performed.For exposures of ca. 10 minutes, the extent of surface oxidation of clean Ni to form NiF2, as calculated from Ni 2p and F 1s XPS spectra, was found to be significantly higher for gas-, compared to liquid-phase AHF. Thin (ca. 100 nm), mirror-like Ni films sputtered onto the surface of a sapphire disk yielded upon exposure to gas phase AHF (after contact with the laboratory atmosphere) scanning electron microscope images displaying a coherent, patchy structure with raised junctures. This behavior was ascribed to internal compressive stresses induced by the formation of a nickel (II) fluoride layer on the metal surface.

The electrochemical properties of clean and oxygen-contaminated polycrystalline Ag surfaces have been examined in LiClO4/polyethylene oxide solutions in ultrahigh vacuum (UHV) environments at temperatures in the range 323-333 K. Unlike the behavior observed for Au and Ni under the same experimental conditions, no clearly defined voltammetric peaks were found during the first and subsequent cycles in the range 2.20-0.25 V vs. Li/Li+ initiated at the open-circuit potential, 1.75 V vs. Li/Li+. instead, the scans in the negative direction were characterized by two adjoining regions in which the current increased linearly with potential, albeit at different rates, and the subsequent scans in the positive direction yielded comparatively much smaller currents largely independent of the applied potential. Integration of the voltammetric curves over the potential range 0.25 < E < 2.20 V vs. Li/Li+ revealed a pronounced imbalance between the charges obtained in the scans in the negative (Q(-)) and positive (Q(+)) directions. This phenomenon was attributed, by and large, to the high rates of Li dissolution into Ag at these temperatures, consistent with the presence of a low-temperature eutectic in the Li-Ag phase diagram. Additional support for this view was obtained from UHV nonelectrochemical measurements involving vapor-deposited Li onto Ag, for which the amount of Li on the surface, as monitored by Auger electron spectroscopy, decreased markedly upon increasing the temperature from ca. 300 to 350 K. The voltammetry of oxygen-contaminated Ag surfaces was characterized by a well-defined peak in the scan in the positive direction centered at ca. 1.3 V, which persisted upon continuous cycling. Although the process responsible for this feature has not yet been identified, it provides a marker for detecting oxygen impurities on Ag in this electrolyte. (C) 1999 The Electrochemical Society. S0013-4651(98)10-067-8. All rights reserved.

The underpotential deposition (UPD) of lithium on polycrystalline Ni and Ni(111) from LiClO4/poly(ethylene)oxide (PEO) and LiI/PEO electrolytes was examined by cyclic voltammetry in, ultrahigh vacuum (UHV) at temperatures, T, in the range 330-340 K. At least two well-defined UPD peaks (A and B), and their corresponding stripping counterparts (A' and B'), were identified in the region 0.25-2.0 V vs. Li[C/R]. Their combined charge, Q(A + B) (or Q(A' + B')), estimated from the smoother Ni(111) specimen, was about 40 mu C/cm(2), i.e. equivalent to a Li coverage (theta(Li)) of ca, 0.15, assuming Li+ undergoes full discharge. The presence of more than a single Li UPD voltammetric feature is consistent with low energy electron diffraction (LEED) studies of K, Cs and Li adsorbed on Ni(111), which revealed different surface superstructures as a function of the alkali metal coverage (theta(alk)) for 300 < T < 350 K. Furthermore, the small values of theta(Li) found just prior to bulk Li electrodeposition, are in harmony with (i) additional LEED information, which indicates that a second alkali metal layer begins to form for theta(alk)less than or equal to 0.5 and (ii) the rapid decrease in the work function of Ni, Phi(Ni) (and other high work function metals) as a function of theta(alk) to values lower than theta(alk) for theta(alk) < 0.3. Electrodeposition of bulk Li on Ni displayed a nucleation/growth loop and a sharp stripping peak with no evidence for alloy formation. Marked changes in the voltammetric features could be observed after dosing polycrystalline Ni surfaces with carbon, and especially oxygen, supporting the view that peaks A and B (and A' and B') can indeed be ascribed to Li UPD land stripping) and not to effects associated with superficial impurities. (C) 1998 Elsevier Science Ltd. All rights reserved.