ARTICLE pubs.acs.org/Langmuir
Photoreduction Kinetics of Sodium Tetrachloroaurate under Synchrotron Soft X-ray Exposure Yuen-Yan Fong,† Bradley R. Visser,† Jason R. Gascooke,‡ Bruce C. C. Cowie,§ Lars Thomsen,§ Gregory F. Metha,† Mark A. Buntine,*,|| and Hugh H. Harris*,† †
Department of Chemistry, The University of Adelaide, Adelaide, SA 5005, Australia School of Chemical and Physical Sciences, The Flinders University of South Australia, GPO Box 2100 Adelaide, SA 5001, Australia § Australian Synchrotron, 800 Blackburn Road, Clayton, VIC 3168, Australia Department of Chemistry, Curtin University, GPO Box U1987 Perth, WA 6845, Australia
)
‡
bS Supporting Information ABSTRACT: We report on the time evolution of the sodium tetrachloroaurate (NaAuCl4) chemical properties as a function of soft X-ray exposure in a dried sample on a silicon surface using X-ray photoelectron spectroscopy (XPS). Our investigations provide mechanistic insight into the photoreduction kinetics from AuIII to AuI and then AuI to Au0. We unambiguously show that XPS photoreduction occurs in stepwise fashion via the AuI state. Both photoreduction steps undergo first-order kinetics.
1. INTRODUCTION Metal nanoparticles have attracted significant interest because of their potential use in applications such as microelectronics, photonics, catalysis, and biochemical sensors.115 Gold nanoparticles can be generated with control over size and shape via a number of techniques. These include chemical reduction from an aqueous tetrachloroaurate solution,16,17 X-ray18,19 and γ-ray irradiation20 of a solid gold salt, chemical synthesis of size-specific clusters (e.g., 55-atom Au),21 direct deposition of size-selected clusters onto a substrate,22 and laser ablation of a metallic gold disk.2328 Previously, we have described the kinetics involved in nanoparticle formation in solution via laser ablation of gold in pure water and solutions of sodium dodecylsulfate.24 We now extend our work to investigate nanoparticle behavior under X-ray irradiation. Herein we determine the reaction pathway and kinetics for the reduction of sodium tetrachloroaurate (NaAuCl4 3 2H2O) under X-ray exposure on a silicon substrate as a precursor to understanding nanoparticle photochemistry. X-ray photoelectron spectroscopy (XPS) is known to cause chemical and structural changes within several investigated molecules.29,30 The technique has been reported to result in the reduction of metal ions by ion, electron, and photon bombardment, referred to as “radiation-induced damage” during the development of the technique. A series of studies have concluded that inorganic salts undergo redox processes during XPS measurements, such as the decomposition of ClO4 to yield r 2011 American Chemical Society
ClO3 and Cl and of MgCl to form Mg and Cl2.29 Complexes of CrVI (e.g., CrO3, K2CrO4) show reduction to CrIII compounds under X-ray exposure.3034 Other studies have reported that X-ray-induced decomposition of the investigated molecules leads to a growth of oxide or halide peaks in the XPS spectrum.18,19,29 Many reports of X-ray exposure leading to the desorption of oxygen or chlorine radicals/anions to generate oxygen or chlorine gas have been published.1719,2931,33,35,36 In 2005, Karadas and co-workers37 published an overview of the tetrachloroauric acid (HAuCl4) photoreduction process from AuIII to Au0 when exposed to 1253.6 eV Mg KR (nonmonochromatized) radiation. They found that the reduction of gold to its metallic state occurs with the production of chlorine gas and the embedment of HCl into the support matrix. Together with further contributions from Ozkaraoglu et al., Karadas and co-workers showed that the disappearance of AuIII and Cl peaks in the XPS spectra occur via first-order kinetics but at differing rates.18,19,37 In this article, we explore the chemical changes and the associated kinetics of sodium tetrachloroaurate (NaAuCl4) exposed to synchrotron soft X-ray radiation. This is achieved by recording high-resolution XPS spectra of the gold 4f region as a Received: February 3, 2011 Revised: May 16, 2011 Published: June 08, 2011 8099
dx.doi.org/10.1021/la200463k | Langmuir 2011, 27, 8099–8104
Langmuir
ARTICLE
function of X-ray exposure time. Our spectra are recorded at a higher resolution than previous studies, providing new insight into the photoreduction process. In particular, we unambiguously show stepwise XPS photoreduction to occur via the AuI state.
2. EXPERIMENTAL METHOD A solution containing 0.02 g of sodium tetrachloroaurate (NaAuCl4 3 2H2O, Aldrich, >99% purity) in 10 mL of water was prepared. One drop of this mixture was deposited on a clean 6 6 mm2 silicon (Si) wafer and subsequently dried under vacuum. The Si wafer was then fixed to double-sided copper tape to ensure the electrical conductivity of the sample. Photoelectron spectra were recorded at the Soft X-ray Beamline at the Australian Synchrotron (AS) using a SPECS Phoibos 150 hemispherical electron analyzer. Samples were continuously irradiated for a period of 900 s. During this irradiation, sequential XPS spectra were recorded by scanning the electron analyzer from 95 to 70 eV in 0.05 eV increments. These scans each took 50 s to complete, followed by an ∼16 s delay before the next scan started. The synchrotron radiation was set to a photon energy of 650 eV38 and adjusted to yield an irradiation spot size of ∼600 μm. The electron analyzer pass energy was set to 10 eV, yielding an instrumental resolution of 295 meV in terms of the resultant electron kinetic energy.38 XPS spectra in the region of the Au 4f peaks were calibrated using the Au 4f7/2 peak (84.0 eV) of gold foil.39 We employed an X-ray photon flux at the AS of approximately 1012 photons mm2 s1, conditions similar to those recently discussed by Hu et al.40 Under these conditions, it is highly unlikely that we are inducing localized heating of NaAuCl4 or the substrate. For example, at a photon energy of 650 eV, the total irradiation power is approximately 100 μW over an area of ∼600 μm2. This is insufficient monochromatic X-ray power to result in appreciable thermal damage. Moreover, under the defocused irradiation conditions, combined with the low detector pass energy of 10 eV, the synchrotron-based XPS studies reported here afford more photoelectrons being produced and recorded at higher resolutions than is possible with laboratory-based instruments. It is these aspects that have enabled our study to provide new insight into the X-ray photoreduction of AuIII. For each recorded XPS spectrum, the data was deconvoluted to obtain the relative population of each gold oxidation state. This was achieved by fitting the spectra to a sum of peaks using nonlinear leastsquares minimization. A Shirley background was applied to remove the electron-scattering background and maintain the intrinsic line shape from the raw data.41,42 A pseudo-Voigt function composed of the sum of Gaussian (30%) and Lorentzian (70%) functions43 was used, with a fullwidth-at-half-maximum of all peaks fixed at 1.28 eV. This functional form for the peaks gave the best fit when comparing all scans. The Au 4f spinorbit (4f7/2 and 4f5/2) energy splitting was fixed at 3.67 eV, but the absolute position was allowed to vary. The ratio of peak areas between the 4f7/2 and 4f 5/2 peaks is expected to appear as 4:3 because of the (2j þ 1) multiplicity of each spinorbit state.44,45 This ratio was either held constant or allowed to vary, depending upon the fitting procedure used. A comprehensive discussion of the fitting procedures employed is given in the Results and Discussion section.
3. RESULTS AND DISCUSSION 3.1. X-ray Photoelectron Spectra. XPS spectra of sodium tetrachloroaurate in the Au 4f region as a function of synchrotron irradiation time are shown in Figure 1. The detailed fitting analysis (see below) indicates that for each oxidation state of gold there are two peaks separated by 3.67 eV, corresponding to the 4f7/2 (lower binding energy) and 4f5/2 (higher binding
Figure 1. XPS spectra of sodium tetrachloroaurate as a function of time under synchrotron X-ray exposure. The dotted lines represent the peak position of the 4f5/2 peaks of each gold species.
energy) spinorbit states. The XPS spectra in Figure 1 are annotated to highlight three sets of doublets, each representing a different oxidation state of gold. Using the known electron binding energies for the various oxidation states of gold,18,19,44 we assign the peaks at 84.0, 85.1, and 87.2 eV as the 4f7/2 peak of Au0, AuI, and AuIII, respectively, and the peaks at 87.6, 88.7, and 91.0 eV as the 4f5/2 peak of Au0, AuI, and AuIII, respectively. The spectra in Figure 1 show that the intensity of each gold oxidation state varies as the sodium tetrachloroaurate sample undergoes continuous irradiation for 900 s. For instance, the intensity of the 4f5/2 AuIII peak decreases quickly as a function of irradiation time, with a concomitant rapid initial growth of the 4f5/2 AuI peak (as seen by comparing the first two spectra). This photoreduction process occurs faster than the timescale of recording each spectrum (50 s). Prior to exposure, we assume that only AuIII is present in the sample. This assumption is consistent with previous reports highlighting the instability of AuI in aqueous solution.17,37,4648 However, the spectrum at the earliest irradiation times (lowest spectrum in Figure 1) shows that during data collection a significant amount of AuIII is converted to AuI prior to the completion of the first scan. Upon further exposure, the 4f5/2 AuIII peak decreases further in intensity and the fully reduced gold peak, 4f5/2 Au0, slowly grows in. Similar trends are apparent for the Au 4f7/2 XPS peaks. Over the 900 s duration of the experiment, all of the peak positions shift by up to 0.4 eV to higher binding energy as shown by the dotted line in Figure 1 for the 4f5/2 peaks. We ascribe this shift in binding energy to changes in the morphology and chemical environment of the gold species as the sample undergoes irradiation. This, however, does not affect the identification of the different oxidation species, which is the salient point of this study. 8100
dx.doi.org/10.1021/la200463k |Langmuir 2011, 27, 8099–8104
Langmuir 3.2. Reinterpretation of Previous Investigations. Our results clearly show that the photoreduction of AuIII in the tetrachloroaurate anion proceeds via reduction to AuI followed by full reduction to metallic Au0. Previous investigations into the photoreduction of AuIII have observed only the disappearance of AuIII and the growth of metallic gold.18,19,37 We are able to observe an AuI intermediate because of the increased resolution in our experiment. In the work of Karadas et al., the 4f7/2 peaks of Au0 and AuI are not resolved, preventing direct observation of the AuI species.37 Likewise, the 4f7/2 peak of AuIII and the 4f5/2 peaks of AuI and Au0 are manifest as a single peak in their spectra. Using our results, we are in a position to understand better the behavior of the previously reported time-dependent XPS spectra. Karadas et al. observed that the 4f7/2 Au0 peak shifted by approximately 0.8 eV to lower binding energy as the X-ray exposure continued37 and attributed this to the nucleation and growth of metallic gold clusters. This reported peak shift is seemingly at odds with our observations, where we detect a shift to higher binding energies. From our results, it is clear that the shift observed by Karadas et al. is actually due to the eventual conversion to Au0 of the AuI intermediate, which forms rapidly from AuIII on initial exposure. The apparent peak positions reported by Karadas are, in fact, determined by the relative populations of AuI and Au0. As such, the peaks appear to shift to lower binding energy with increasing exposure as the relative AuI and Au0 populations change. The splitting of the 4f7/2 Au0 and AuI peaks is approximately 1.1 eV, so the 0.8 eV shift to lower binding energy observed by Karadas et al. is, in fact, an ∼0.3 eV shift of each component to higher binding energy. This is similar to the ∼0.4 eV shift that we observe as shown by the dotted lines in Figure 1; however, the exact cause for this shift to higher energies remains unclear. An equivalent argument also holds for what we now understand to be the combined 4f7/2 AuIII and 4f5/2 AuI and Au0 peak observed by Karadas et al.37 These workers observe this peak to first shift to higher binding energy and then to return to lower binding energy as a function of X-ray exposure. Upon the basis of our observation of AuI, we interpret the shift observed by Karadas et al. as follows: The initial peak position of the 4f7/2 AuIII component was reported to appear at 85.3 eV. As time progresses, the 4f5/2 AuI intermediate peak grows in intensity, causing an apparent shift to higher binding energies. Upon further irradiation, AuI is reduced to metallic gold and consequently the 4f5/2 Au0 component dominates, causing this peak now to shift to lower binding energies. 3.3. Kinetic Modeling of the Photoreduction of Sodium Tetrachloroaurate. The relative population of each of the oxidation states of gold as a function of the irradiation time can be determined by the deconvolution of the XPS spectra. The electron analyzer used in our experiments scans the photoelectron energy over the data collection period. Therefore, during each scan the population of different oxidation states can (and does) vary, and thus the intensity ratio of the two spinorbit components may deviate from the expected value of 4:3 (4f7/2/4f5/2). This is particularly true for the removal of the AuIII species and the production of AuI, which, especially at low exposure times, occur more rapidly than the time between successive scans. Thus, one needs to take into account the time at which each peak was recorded, as determined by the scan start time and the analyzer scan rate. The peaks were fitted with no assumption of the expected 4:3 intensity ratio between the two spinorbit components. For
ARTICLE
Figure 2. Fitted XPS spectra of sodium tetrachloroaurate at select times during synchrotron X-ray irradiation. Superimposed on each experimental spectrum are fits using our kinetic model, the individual fits on the left-hand panel and the global fits on the right-hand panel. (See the text for details.) The spectra are (a) scan 1, (b) scan 2, (c) scan 4, (d) scan 8, and (e) scan 15. The following color scheme is used to display the individual components of the fit: experimental data (orange), Au0 (green), AuI (red), AuIII (blue), and total (black).
selected times, the experimental data and best fit to this data using the above fitting procedure are shown in the left-hand panel of Figure 2. We refer to fitted data using this procedure as “individual fits” because each XPS spectrum is fitted individually. (The right-hand panel of Figure 2 displays fits using a “global” fitting method that has yet to be presented; see below.) Using the peak areas determined from the fitting procedure, populations of each oxidation state are determined as a function of time. This is achieved via normalization of the peak areas by dividing the fitted area by the (2j þ 1) multiplicity (degeneracy) of each spinorbit component, giving a quantity that is proportional to the total population of each oxidation state. For the individual fits described above, we cannot simply add the 8101
dx.doi.org/10.1021/la200463k |Langmuir 2011, 27, 8099–8104
Langmuir
ARTICLE
Table 1. Rate Constants for the Photoreduction of Sodium Tetrachloroauratea fitting procedure individual fit k1 (s1) 1
k1 (s ) 1
k2 (s ) k2 (s1) a
global fit
(6.8 ( 0.3) 102
(6.63 ( 0.05) 102
2
(2.57 ( 0.03) 102
3
(1.88 ( 0.02) 103 (5.2 ( 0.3) 104
(2.27 ( 0.14) 10
(1.91 ( 0.06) 10 (6.2 ( 0.7) 104
See the text for kinetic model details.
Further inspection of the data points in Figure 3 suggests that the photoreduction is proceeding through a stepwise mechanism, where AuIII is first reduced to AuI and then AuI is reduced further to Au0. To model this behavior, we have developed a kinetic model for conversion between the different oxidation states. In this model, we allow interconversion between AuIII and AuI species and between AuI and Au0 species. Each process has a distinct rate of production and is considered reversible. Overall, the simple reaction scheme is described by the series of equilibrium steps in Scheme 1: The kinetic model presented in Scheme 1 contains four unique channels. The resultant rate expressions Figure 3. (a) Population of the different gold species as a function of synchrotron exposure time. (b) The same data recorded within the first 400 s of exposure. The data points represent the relative population of each oxidation state. The lines represent best fits to the data using our model. (See the text for more details.) The error bars represent three standard deviations as determined by the least-squares fitting procedure.
dAuIII ¼ k1 ½AuI k1 ½AuIII dt dAuI ¼ k1 ½AuIII k1 ½AuI k2 ½AuI þ k2 ½Au0 dt and dAu0 ¼ k2 ½AuI k2 ½Au0 dt
Scheme 1. Model Reaction Scheme Highlighting the Reversible Formation of Various Gold Oxidation States
two spinorbit components together because the populations change significantly during each individual scan. Therefore, two unique data points exist for each XPS spectrum because of the two different spinorbit states, with their difference on the time axis resulting from the scan rate of the electron analyzer. Figure 3a shows the population of each species as a function of synchrotron exposure, and Figure 3b presents an expansion of the low exposure time data of Figure 3a. Here, the data points are determined from the fitted peak areas of 4f7/2 and 4f5/2 components of the experimental data, and the solid lines represent fits to a kinetic model yet to be presented. The error bars on the data represent three standard deviations as determined from the fitting process. From these data, it is clear that the population of AuIII decreases dramatically within the first 50 s and then decays further at a slower rate. In addition, the AuI population increases rapidly at low exposure times and subsequently decays slowly at longer exposure times. The Au0 population increases steadily as a function of exposure time. We have used this data to develop a model to describe the X-ray photoreduction kinetics of sodium tetrachloroaurate.
were integrated numerically to determine the relative populations of each oxidation state as a function of time. A nonlinear least-squares fit of each species’ population yields the rate constants reported below. The resulting best fits are presented as lines in Figure 3. The most important aspect of the data-fitting algorithm is that to generate the curves in Figure 3 all four channels are required to fit the data. That is, all four channels (k1, k1, k2, and k2) are required to account for the observed behavior. We refer to the rate constants determined via this fitting method as being derived from individual fits because each XPS spectrum is fitted individually. The values of the rate constants determined via this fitting method are reported in Table 1. The smaller rate constants determined for the conversion of AuI to Au0 result from the experimental observation of the slower, steady production of Au0 as a function of exposure time. The solution involving the channels identified above is indeed unique. From the experimental data, it is obvious that both reverse reactions are necessary because the AuIII and AuI populations do not asymptote to zero. Attempts to allow only the forward reactions to occur (i.e., k1 and k2 rate constants only) or to allow only one of the reverse reactions to occur with both forward reactions simply do not fit the experimental data (Supporting Information). The fit to the data in Figure 3 is very good at longer exposure times, but the agreement at times of less than 100 s (that is, the first two XPS scans) is less satisfactory. To explore the reasons for 8102
dx.doi.org/10.1021/la200463k |Langmuir 2011, 27, 8099–8104
Langmuir this discrepancy, we have also fitted the data using a global fitting procedure such that all spectra are fitted simultaneously while the rate constants are varied. The intensity ratios between all of the peaks in each XPS spectrum were determined by the population of the species at the time that the electron analyzer was recording that particular peak and now assuming that the 4:3 intensity ratio between the 4f7/2 and 4f5/2 peaks holds. All of the XPS spectra were then fitted simultaneously to determine the optimum rate constants. The values of the fitted rate constants for each Aun (n = 0, I, or III) species fitted in this global manner are reported in Table 1 alongside those determined by the individual fitting of the spectra. It can be seen that the resultant rate constants are very similar from each analysis method. The spectra generated via the global fit are shown in the righthand panel of Figure 2. From an inspection of Figure 2, it is seen that the spectra fit well at later times; however, at early times, there is an intensity discrepancy for the AuI 4f5/2 and AuIII 4f7/2 peaks. Even after taking into account the different times at which each of the spinorbit peaks are recorded, the intensity ratios between each pair of spinorbit peaks slightly deviate from the expected 4:3 ratio. Neither fitting the spectra individually nor employing the global fitting process results in a particularly good fit at the shortest exposure times. Recording XPS photoreduction data at faster electron analyzer scanning rates (not currently possible at the AS) is required to explore this issue in greater detail. Over the complete timescale of the experiment, we demonstrate that the relative yields of AuIII, AuI, and Au0 vary in a manner explained by predictions that arise from the kinetic model we propose whereby photoreduction proceeds via the AuI intermediate. Consistent with this observation, very recently Marcum et al. reported on the gas-phase photochemistry of AuIII and noted that the photoreduction process proceeds via “either AuII or AuI depending on the number of neutral ligands lost.”49 Critically, we are unable to explain the experimental observations by any kinetic model that involves the direct photoreduction of AuIII to Au0.
4. CONCLUSIONS In this article, we have reported on the chemical changes and the associated kinetics of sodium tetrachloroaurate (NaAuCl4) exposed to synchrotron soft X-radiation. Our investigation provides new insight into the photoreduction process by recording high-resolution XPS spectra of the gold 4f region as a function of time. Importantly, our studies highlight that the XPS photoreduction occurs via the AuI state. We have developed a kinetic model that involves a stepwise reduction among AuIII, AuI, and Au0 using two different analysis methods: individual fitting and global fitting of the individual gold 4f peaks. Our kinetic analysis accounts for the population of each Au chemical state as a function of irradiation time. ’ ASSOCIATED CONTENT
bS
Supporting Information. Relative population of the different gold species as a function of synchrotron exposure time using different kinetic models. This material is available free of charge via the Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION Corresponding Author
*E-mail:
[email protected];
[email protected].
ARTICLE
’ ACKNOWLEDGMENT We thank Professor Terry Turney from Monash University for supplying the sample of sodium tetrachloroaurate. We also acknowledge Professor Bill Skinner from the University of South Australia for helpful discussions on XPS data analysis. This research was undertaken on the soft X-ray spectroscopy beamline at the Australian Synchrotron, Victoria, Australia. ’ REFERENCES (1) Daniel, M. C.; Astruc, D. Chem. Rev. 2004, 104, 293–346. (2) Hutter, E.; Fendler, J. H. Adv. Mater. 2004, 16, 1685–1706. (3) Henglein, A. J. Phys. Chem. 1993, 97, 5457–5471. (4) Armendariz, V.; Herrera, I.; Peralta-Videa, J. R.; Jose-Yacaman, M.; Troiani, H.; Santiago, P.; Gardea-Torresdey, J. L. J. Nanopart. Res. 2004, 6, 377–382. (5) Curulli, A.; Valentini, F.; Padeletti, G.; Cusma, A.; Ingo, G. M.; Kaciulis, S.; Caschera, D.; Palleschi, G. Sens. Actuators, B 2005, 111, 526–531. (6) Gao, X. Y.; Matsui, H. Adv. Mater. 2005, 17, 2037–2050. (7) Li, C. Z.; Male, K. B.; Hrapovic, S.; Luong, J. H. T. Chem. Commun. 2005, 3924–3926. (8) Schmid, G.; Simon, U. Chem. Commun. 2005, 697–710. (9) Schmid, G.; Corain, B. Eur. J. Inorg. Chem. 2003, 3081–3098. (10) Alvarez-Puebla, R.; Liz-Marzan, L. M.; Javier Garcia de Abajo, F. J. Phys. Chem. Lett. 2010, 1, 2428–2434. (11) Levi, R.; Bar-Sadan, M.; Albu-Yaron, A.; R., P.-B.; Houben, L.; Shahar, C.; Enyashin, A.; Seifert, G.; Prior, Y.; Tenne, R. J. Am. Chem. Soc. 2010, 132, 11214–11222. (12) Liu, J.; He, W. W.; Hu, L. J.; Liu, Z.; Zhou, H. Q.; Wu, X. C.; Sun, L. F. J. Appl. Phys. 2010, 107, 094311. (13) Minati, L.; Torrengo, S.; Speranza, G. Surf. Sci. 2010, 604, 508–512. (14) Rodriguez, J. A.; Stacchiola, D. Phys. Chem. Chem. Phys. 2010, 12, 9557–9565. (15) Santori, C.; Barclay, P. E.; Fu, K.-M. C.; Beausoleil, R. G.; Spillane, S.; Fisch, M. Nanotechnology 2010, 21, 274008. (16) Cohen, L. R.; Pena, L. A.; Seidl, A. J.; Olsen, J. M.; Wekselbaum, J.; Hoggard, P. E. Monatsh. Chem. 2009, 140, 1159–1165. (17) Holt, K. B.; Sabin, G.; Compton, R. G.; Foord, J. S.; Marken, F. Electroanalysis 2002, 14, 797–803. (18) Ozkaraoglu, E.; Tunc, I.; Suzer, S. Surf. Coat. Technol. 2007, 201, 8202–8204. (19) Ozkaraoglu, E.; Tunc, I.; Suzer, S. Polymer 2009, 50, 462–466. (20) Tsuda, T.; Seino, S.; Kuwabata, S. Chem. Commun. 2009, 44, 6792–6794. (21) Turner, M.; Golovko, V. B.; Vaughan, O. P. H.; Abdulkin, P.; Berenguer-Murcia, A.; Tikhov, M. S.; Johnson, B. F. G.; Lambert, R. M. Nature 2008, 454, 981–983. (22) Lim, D. C.; Hwang, C. C.; Gantefor, G.; Kim, Y. D. Phys. Chem. Chem. Phys. 2010, 12, 1517215180. (23) Amendola, V.; Meneghetti, M. Phys. Chem. Chem. Phys. 2009, 11, 3805–3821. (24) Fong, Y. Y.; Gascooke, J. R.; Visser, B. V.; Metha, G. F.; Buntine, M. A. J. Phys. Chem. C 2010, 114, 15931–15940. (25) Hartland, G. V. J. Chem. Phys. 2002, 116, 8048–8055. (26) Inasawa, S.; Sugiyama, M.; Yamaguchi, Y. J. Phys. Chem. B 2005, 109, 3104–3111. (27) Link, S.; Hathcock, D. J.; Nikoobakht, B.; El-Sayed, M. A. Adv. Mater. 2003, 15, 393. (28) Mafune, F. Chem. Phys. Lett. 2004, 397, 133–137. (29) Copperthwaite, R. G. Surf. Interface Anal. 1980, 2, 17–25. (30) Storp, S. Spectrochim. Acta 1985, 40B, 745–756. (31) Fleisch, T. H.; Zajac, G. W.; Schreiner, J. O. Appl. Surf. Sci. 1986, 26, 488–497. (32) Garbassi, F.; Pozzi, L. J. Electron Spectrosc. Relat. Phenom. 1979, 16, 199–203. 8103
dx.doi.org/10.1021/la200463k |Langmuir 2011, 27, 8099–8104
Langmuir
ARTICLE
(33) Sasaki, T.; Williams, R. S.; Wong, J. S.; Shirley, D. A. J. Chem. Phys. 1978, 68, 2718–1714. (34) Wallbank, B.; Johnson, C. E.; Main, I. G. J. Electr. Spectrosc. Relat. Phenom. 1974, 4, 263–269. (35) Imamura, K.; Asai, M.; Tazuke, S.; Okamura, S. Makromol. Chem. 1973, 174, 91–101. (36) Sen, P. K.; Gani, N.; Midya, J. K.; Pal, B. Transition Met. Chem. 2008, 33, 229236. (37) Karadas, F.; Ertas, G.; Ozkaraoglu, E.; Suzer, S. Langmuir 2005, 21, 437–442. (38) Cowie, B. C. C.; Tadich, A.; Thomsen, L. AIP Conf. Proc. 2010, 1234, 307–310. (39) Beattie, D. A.; Kempson, I. M.; Fan, L. J.; Skinner, W. M. Int. J. Miner. Process. 2009, 92, 162–168. (40) Hu, B. B.; Yu, L. F.; Wang, G. J.; Hutton, S.; Huang, H. Z.; Du, Z. L. Measurement 2010, 43, 1299–1304. (41) Shirley, D. A. Phys. Rev. B 1972, 5, 4709–4714. (42) Vegh, J. J. Electron. Spectrosc. Relat. Phenom. 2006, 151, 159–164. (43) Hughes, H. P.; Scarfe, J. A. J. Phys.: Condens. Matter 1996, 8, 1421–1438. (44) Moulder, J. F.; Stickle, W. F.; Sobol, P. E. Handbook of X-ray Photoelectron Spectroscopy; Perkin-Elmer Corporation: Eden Prairie, MN, 1992. (45) St€ohr, J.; Siegmann, H. C. Magnetism: From Fundamentals to Nanoscale Dynamics; Springer: New York, 2006. (46) Greenwood, N. N.; Earnshaw, A. Chemistry of the Elements; Pergamon Press: New York, 1984. (47) Murphy, P. J.; Stevens, G.; LaGrange, M. S. Geochim. Cosmochim. Acta 2000, 64, 479–494. (48) Hanes, G.; Koeber, K.; Kreuzbichler, I.; Neu-Becker, U.; Schwager, B. Swoboda, G.; Nes, G. H.; Hanes, G.; Huisl, W.Gmelin Handbook of Inorganic and Organometallic Chemistry: Au - Gold , 8th ed.; Springer-Verlag: Berlin, 1992; Supplement Vol. B/1. (49) Marcum, J. C.; Kaufman, S. H.; Weber, J. M. J. Phys. Chem. A 2011, 115, 3006–3015.
8104
dx.doi.org/10.1021/la200463k |Langmuir 2011, 27, 8099–8104