Anal. Chem. 2000, 72, 2969-2975
Electrochemical and XPS Characterization of Composite Modified Electrodes Obtained by Nickel Deposition on Noble Metals Innocenzo G. Casella* and Maria Gatta
Dipartimento di Chimica, Universita´ degli Studi della Basilicata, Via N. Sauro 85, 85100 Potenza, Italy
A chemically modified electrode composed of nickel oxyhydroxide film deposited on noble metals (i.e., gold or platinum) was characterized in an alkaline medium by cyclic voltammetry and XPS (X-ray photoelectron spectroscopy) techniques. The nickel was deposited on the gold substrate in an alkaline medium by various strategies: cycling the potential between -0.1 and 0.65 V vs SCE, in potentiostatic conditions at potentials comprised between 0.0 and 0.55 V and by simple immersion of the electrode in non-deaerated 0.2 M NaOH solutions containing 3 mM K2Ni(CN)4. The effects of several experimental parameters such as applied potentials, pH, tetracyanonickelate concentration, electrode substrate, etc., on the nickel film formation and growth were evaluated. The electroactivity of the resulting composite gold-nickel electrode was investigated in an alkaline medium toward the oxidation of carbohydrates using arabinose as a model compound. Oxide/hydroxide films of transition metals constitute a wide class of materials which are widely used in many areas, such as electrocatalysis or electrosynthesis,1,2 battery and material science,3,4 electrochromic devices,5,6 interfacial charge, and electron transfer.7,8 To increase the surface area of the active phase, highly dispersed metal oxide/hydroxide particles are frequently prepared by various deposition techniques such as chemisorption of metal ions on the traditional electrode substrates or polymer matrixes, electrochemical deposition of metals or their electroactive complexes, and sputtering of metal particles in vacuum systems. The dispersed microparticles are physically separated from each other, and a highly active electrode suitable for efficient electrocatalysis can be obtained. Thus, highly dispersed metal catalysts on inert electrode surfaces and methods for their preparation are of great practical as well as theoretical interest. Electrochemical techniques (1) O’Sullivan, E. J. M.; Calvo, E. J. In Electrode Kinetics: Reactions; Compton, R. G., Ed.; Elsevier: Amsterdam, 1987; Vol. 27, Chapter 3. (2) Trasatti, S. The electrochemistry of novel materials. In Transition metal oxides: versatile materials for electrocatalysis; Lipkowski, J., Ross, P. N., Eds.; VCH Publishers: New York, 1994; p 207. (3) Lambert, C. M.; Nazri, G.; Yu, P. C. Sol. Energy Mater. 1987, 16, 1. (4) Honda, K.; Hayashi, H. J. Electrochem. Soc. 1987, 134, 1330. (5) Passerini, S.; Scrosati, B.; Gorenstein, A.; Andersson, A. M.; Granqvist, C. G. J. Electrochem. Soc. 1989, 136, 3394. (6) Chigane, M.; Ishikawa, M. Electrochim. Acta 1997, 42, 1515. (7) Kubota, L. T.; Gushikem, Y. J. Electroanal. Chem. 1993, 362, 219. (8) Tanaka, K.; Tamamushi, R. J. Electroanal. Chem. 1995, 380, 279. 10.1021/ac9913863 CCC: $19.00 Published on Web 06/30/2000
© 2000 American Chemical Society
are frequently employed for the metal deposition because very thin and uniform films with a high degree of reproducibility can be obtained. In electroanalysis, modified electrodes prepared by electrodeposition of cyanometalate complexes,9-14 cathodic or anodic deposition of metal ions on an inert surface or entrapped into organic polymer matrixes15-21 are widely characterized and proposed as electrocatalytic systems toward the electrooxidation of several organic compounds. An alternative strategy to prepare CMEs (chemically modified electrodes) with increased catalytic properties is to employ underpotential deposition of heavy metal ad-atoms or to use co-deposition of transition metals on the inert electrode substrate.22-24 The enhancement of the electrocatalytic activity by co-deposition of transition metals may be interpreted in terms of decreased electrode poisoning. Nevetheless, one of the problems in the use of CMEs is the gradual change in the mechanical integrity of the catalytic film for extended periods of time under applied potentials. Thus, the modified electrodes, prepared by attaching a metal catalyst to the inert electrode surface by covalent bonds, polymer film coatings, or adherent insoluble metal hydroxide, very often show a continuous diminution of the metal loading with subsequent loss of the catalytic activity. The partial solubilization of the catalyst and the mechanical erosion of the metal particles, particularly during operation in flowing solutions, are the main factors causing loss of the electrochemical activity. To overcome these problems, modification strategies of the CMEs with which both electro(9) Itaya, K.; Uchida, I.; Neff, V. D. Acc. Chem. Res. 1986, 19, 162. (10) Kulesza, P. J. J. Electroanal. Chem. 1987, 220, 295. (11) Cox, J. A.; Jaworski, R. K.; Kulesza, P. J. Electroanalysis 1991, 3, 869. (12) Joseph, J.; Gomathi, H.; Prabhakara Rao, G. Electrochim. Acta 1991, 36, 1537. (13) Cataldi, T. R. I.; Campa, C.; Centonze, D. Anal. Chem. 1995, 67, 3740. (14) Narayanan, S. S.; Scholz, F. Electroanalysis 1999, 11, 465. (15) Kost, K. M.; Bartak, D. E.; Kazee, B.; Kuwana, T. Anal. Chem. 1988, 60, 2379. (16) Chigane, M.; Ishikawa, M. Electrochim. Acta 1997, 42, 1515. (17) Yet, J.-H. Fedkiw, P. S. Electrochim. Acta 1996, 41, 221. (18) Casella, I. G.; Guascito, M. R.; Sannazzaro, M. G. J. Electroanal. Chem. 1999, 462, 202. (19) Casella, I. G. Electrochim. Acta 1999, 44, 3353. (20) Yang, H.; Lu, T.; Xue, K.; Sun, S.; Lu, G.; Chen, S. J. Electrochem. Soc. 1997, 144, 2302. (21) Kulesza, P. J.; Matczak, M.; Wolkiewicz, A.; Grzybowska, B.; Galkowski, M.; Malik, M. A.; Wieckowski, A. Electrochim. Acta 1999, 44, 2131. (22) Raj, I. A.; Vasu, K. I. J. Appl. Electrochem. 1990, 20, 32. (23) Yang, Y.; Zhou, Y.; Cha, C.; Carroll, W. M. J. Electroanal. Chem. 1992, 338, 251. (24) Schrebler, R.; del Valle, M. A.; Gomez, H.; Veas, C.; Cordova, R. J. Electroanal. Chem. 1995, 380, 219.
Analytical Chemistry, Vol. 72, No. 13, July 1, 2000 2969
chemical activity of the redox mediator and its film deposition on the electrode surface can be performed in the same experimental conditions (i.e., pH, background electrolyte, and applied potentials) are required to preserve the catalytic performance. The purpose of the present study is focused on the deposition of nickel particles on the noble metals in the same experimental conditions where the nickel catalyst shows the maximum electrochemical activity. Particular attention is devoted to its preparation by adsorption of tetracyanonickelate ions and subsequent nickel hydroxide precipitation on the gold electrode surface. The resulting composite gold-nickel (Au-Ni) and platinum-nickel (Pt-Ni) electrodes are characterized by cyclic voltammetry (CV) and X-ray photoelectron spectroscopy (XPS) techniques in high pH solutions. EXPERIMENTAL SECTION Reagents. Potassium tetracyanonickelate(II), sodium hydroxide and arabinose are purchased from Aldrich Chemical Co. and were used as received. Other chemicals employed were of reagent grade and were used without further purification. All solutions were prepared just prior to use with deionized and doubly distilled water. Unless otherwise specified, experiments were performed using 0.2 M NaOH as background electrolyte. Apparatus. Cyclic voltammetry was performed by an EG & G Princeton Applied Research (PAR) model 273 potentiostat/ galvanostat. Data acquisition and potentiostat control were accomplished with a 486/50 MHz IBM-compatible computer running the M270 electrochemical research software (EG&G). All experiments were carried out at room temperature in a standard threeelectrode glass cell using the Au-Ni as a working electrode, a SCE as a reference electrode, and a platinum foil as a counter electrode. The gold, platinum, and glassy carbon substrate electrode used in CV (geometric area, 0.125 cm2) were purchased from PAR. All current densities are quoted in terms of milliamperes per square centimeter of apparent geometric area of the gold electrode. X-ray photoelectron spectra were collected using a Leybold LH ×1 spectrometer using unmonochromatized Al KR radiation (1486.6 eV). The source was operated at 13 kV and 20 mA. The binding energy (BE) scale was calibrated with respect to the Cu2p3/2 (932.7 eV, with a full width at half maximum (fwhm) of 1.75 eV) and Au4f7/2 (84.0 eV with a fwhm of 1.20 eV) signals. Spectra were recorded only after the wide scan showed that no features arose from the copper tape and from the sample rod. Wide and detailed spectra were collected in fixed analyzer transmission (FAT) mode with a pass energy of 50 eV and a channel width of 1.0 and 0.1 eV, respectively. The vacuum in the analysis chamber was always stronger than 5 × 10-9 mbar. The kinetic energy axis origin in all spectra was not corrected for surface charging, but peak positions (BE) in the text are corrected by referring to the C1s peak after setting its BE to 285.0 eV. Data acquisition and spectra analysis were accomplished with a data processing program.25 The background was subtracted before curve-fitting analysis, using a nonlinear baseline correction function of the type suggested by Shirley,26 which assumes that the background on which the photoelectron peaks are sitting is (25) Desimoni, E.; Biader Ceipidor, U. J. Electron Spectrosc. Relat. Phenom. 1991, 56, 189. (26) Shirley, D. A. Phys. Rev. B: Condens. Matter 1972, 5, 4709.
2970 Analytical Chemistry, Vol. 72, No. 13, July 1, 2000
Figure 1. Cyclic voltammograms of nickel film growth on a gold electrode in a 0.2 M NaOH aqueous solution containing 3 mM K2Ni(CN)4. The potential was continuously cycled at 50 mV s-1 between -0.1 and 0.7 V vs SCE.
determined by the intensity of the photopeaks themselves. Thus, the magnitude of the background correction at any point is proportional to the integrated spectral intensity on the low-bindingenergy side of the point. Elemental surface stoichiometries were obtained from peak area ratios corrected by sensitivity factors derived by Wagner.27 Gaussian/Lorenzian sum functions were used to fit the Au4f7/2, Ni2p3/2, and O1s peaks’ line shapes. Parameters to fit the line shape of Au4f7/2, Ni2p3/2, and Pt4f7/2 were obtained by analyzing the relevant metals after 30 min of sputtering with 4 kV argon ions. Electrode Preparation. Prior to each electrode modification, traces of nickel species were removed from the gold (or platinum) surface by soaking the electrode in concentrated hydrochloric acid (37% w/w) for few minutes. The electrode was then polished with 0.05-µm R-alumina powder on a polishing microcloth and washed with doubly distilled water. Films of nickel oxyhydroxide on the gold surface were obtained by voltage cycling (50 mV s-1) between -0.1 and 0.7 V vs SCE or by simple immersion of the gold electrode in non-deareated 0.2 M NaOH plus 3 mM K2Ni(CN)4. The deposition of nickel film on the platinum surface is performed by cycling the potential between -0.7 and 0.65 V vs SCE in 0.2 M NaOH solution containing 3 mM of tetracyanonickelate (II) ions. The surface concentration of nickel sites (ΓNi) was evaluated by cycling the potential between -0.1 and 0.65 V vs SCE in 0.2 M NaOH solution and then determining the charge under the cathodic wave centered at about +0.35 V, assuming that all the nickel redox sites are electroactive on the voltammetric time scale. RESULTS AND DISCUSSION Nickel Film Deposition in an Alkaline Medium. A typical example of cyclic voltammograms obtained during continuous potential cycling for nickel film growth on a gold surface in 0.2 M NaOH plus 3 mM K2Ni(CN)4 is shown in Figure 1. As is characteristic of conducting-film formation, the relevant deposit remains electroactive during electrodeposition so that the redox waves IIa/IIc grow with the number of scans. In addition, the currents of the wave, Ic, relevant to the gold reduction process, are virtually identical to those observed for the bare gold electrode (see Figure 1). The absence of any screening effects on the gold activity is due to the good conductivity and ion permeability of (27) Wagner, C. D. Anal. Chem. 1979, 51, 466.
Table 1. Dependence of the Lower and Higher Potential Limits on the Nickel Surface Concentration (ΓNi)a
Figure 2. Dependence of the rate of nickel film deposition on the tetracyanonickelate (II) concentration. The gold electrode was continuously cycled (100 cycles) at 50 mV s-1 in 0.2 M NaOH solution containing various concentrations of K2Ni(CN)4. The rates of nickel deposition were evaluated by integration of the cathodic peak IIc (referred at the 5th cycle) obtained by cycling the potential at 50 mV s-1 between -0.1 and 0.65 V vs SCE in 0.2 M NaOH solution. The net charge was then related against the relevant cycle number.
the nickel film. In agreement with the literature,18,28-30 the wave Ic is related to gold oxide reduction, whereas the redox waves IIa/IIc are associated with the NiII/NiIII redox transition. The growth of the nickel film was monitored as a function of potential cycle number by measurement of the surface concentration of nickel sites (ΓNi) obtained for each growth cycle. The relevant plot, surface concentration of nickel vs cycle number, calculated in the range 0-200 cycle numbers, gives a straight line with a correlation coefficient of 0.9996. Thus, nickel deposits of about 110 nmol cm-2 (∼26-28 nm in thickness), corresponding to about 200 cycles, can be easily obtained in an alkaline medium. The good linearity between the surface nickel concentration and the cycle number confirms the conducting character of the nickel deposit. The rate of nickel deposition was monitored as a function of the tetracyanonickelate(II) concentration. As can be seen in Figure 2, the rate of deposition increases with the tetracyanonickelate(II) bulk concentration up to 3 mM, while it remains practically constant for higher concentrations. The behavior is characteristic of an electrochemical process in which adsorption of reactants or their subsequent reactions on the electrode surface are involved in the rate-determining step. To elucidate some aspect of the nickel deposition on the gold substrate, the effect of changing the lower and higher potential limits on cyclic voltammograms is investigated. Table 1 summarizes the relevant results. As can be seen, a sensible increase in the nickel surface concentration was observed when the higher potential limit was increased, while a diminution of the deposited nickel was obtained when the lower potential limit was increased. Indeed, a greater extension of the potential window induces an increase of the rate of deposition of the nickel on the gold substrate. Perhaps the formation of nickel(III) species and/or the reduction of gold oxide caused by the large excursion of the potentials, leads to an increased rate of the nickel deposition. (28) Bruckenstein, S.; Shay, M. J. Electroanal. Chem. 1985, 188, 131. (29) Fleischmann, M.; Korinex, K.; Pletcher, D. J. Electroanal. Chem. 1971, 31, 39. (30) Casella, I. G.; Desimoni, E. Electroanalysis 1996, 8, 447.
potential window (V vs SCE)
ΓNi (nmol cm-2)
-0.1-0.65 0.0-0.65 0.2-0.65 0.3-0.65 -0.1-0.75 -0.1-0.85 -0.1-0.95
4.1 3.0 0.4 0.1 6.7 12.7 38.5
a The nickel depositions on the gold substrate were performed by voltage cycling in non-deaerated 0.2 M NaOH solution containing 3 mM K2Ni(CN)4. The potential was scanned 10 times at 50 mV s-1. The nickel surface concentration (ΓNi) of the composite Au-Ni electrode was estimated by integration of the cathodic peak IIc (referred at the 5th cycle) obtained by cycling the potential at 50 mV s-1 between -0.1 and 0.65 V vs SCE in 0.2 M NaOH solution.
Figure 3. Dependence of the rate of nickel deposition on the applied potential. The gold electrode was cycled (10 cycles) between -0.1 and 0.65 V in 0.2 M NaOH solution containing 3 mM K2Ni(CN)4 and polarized at various applied potentials for 300 s.
It is interesting to observe that, after the initial nickel deposition, by cycling the potential (i.e., 10 cycles, ∼0.7 nmol cm-2 of deposited film), further nickel deposition is obtained in potentiostatic conditions. Figure 3 shows the rate of deposition of the nickel on the gold surface obtained after 5 min of deposition time at various applied potentials in 0.2 M NaOH solution plus 3 mM K2Ni(CN)4. As can be seen, the rate of deposition of the nickel on the gold electrode substrate is practically independent of the specific applied potentials. Moreover, deposition of nickel on the gold surface is obtained also by simple immersion of the electrode in the electrolytic medium (0.2 M NaOH solutions plus 3 mM K2Ni(CN)4). Thus, after 300 min of immersion of the gold electrode in the electrolytic solution a deposit of nickel of about 28 nmol cm-2 is obtained. Nevertheless, it is important to underline that, without a preliminary film deposition of nickel by the voltage cycling technique, a negligible deposition of nickel by the chemisorption process is observed also after prolonged treatment. In fact, after ∼30 h of the electrode immersion in the electrolytic solution, a deposit of ∼10 nmol cm-2 is obtained. This clearly demonstrates that the first monolayers of nickel precipitated on the gold surface act as an efficient catalyst for further nickel deposition. The reason for this catalytic effect is not clear at the present time. The previous observations support the suggestion that gold complexation with the cyano group and a simultaneous precipitaAnalytical Chemistry, Vol. 72, No. 13, July 1, 2000
2971
Table 2. Effect of pH on the Nickel Surface Concentration (ΓNi) cycling voltage,a -0.1-0.65 V pH 12.7 13.0 13.3 13.5 13.6 13.7
ΓNi
c (nmol
cm-2)
9.6 10.1 14.8 15.9 16.7 18.9
electrode immersionb pH
ΓNi (nmol cm-2)
7.7 9.0 10 11 12 14
0.5 1.3 2.5 4.5 5.0 6.8
a The nickel deposition on the gold substrate were performed by voltage cycling in non-deaerated 0.2 M NaOH solution containing 3 mM K2Ni(CN)4. b The nickel film was deposited by immersion of the gold electrode in 0.2 M NaOH solution containing 12 mM K2Ni(CN)4 for 3 h. c The nickel surface concentration (ΓNi) of the composite AuNi electrodes was estimated as described in Table 1.
tion of nickel hydroxide on the electrode surface occur in alkaline solutions. In particular, considering the results reported in Table 1 where the concomitant formation of NiIII at high potential and AuI at low potential is postulated, a probable overall reaction of nickel precipitation can be schematically represented as follows
Ni(CN)42- + 2Au + 2OH- S Ni(OH)2 + 2Au(CN)2- + 2 e- (1) Vv NiOOH + H+ + e-
(2)
Upon cycling the nickel deposit between Ni(OH)2 and NiOOH in an alkaline medium, both redox waves IIa/IIc increase as additional nickel is precipitated on the gold surface. To verify the hydroxide effect on the nickel film deposition, experiments were performed both by voltage cycling between -0.1 and 0.65 V vs SCE and by direct chemisorption of nickel on the gold substrate by immersion of the electrode in solutions at various pHs containing tetracyanonickelate (II). Table 2 summarizes the relevant results. Most significant is the observation that, either under potentiodynamic conditions or by a direct chemisorption process, the pH induces the same effect on the nickel deposition. The above observations are consistent with the conclusion that nickel film deposition on the gold substrate proceeds following the same reaction mechanism independent of the experimental conditions of deposition (i.e., potendiodynamic, potentiostatic, or by simple electrode immersion in tetreacyanonickelate solutions). In addition, independent of the preparation procedures, the nickel film appears to be uniform, smooth, and with a good degree of adhesion on the gold electrode surface. To better understand the nickel deposition on the gold surface and to extend the deposition procedures to other inert electrode surfaces, nickel deposition on glassy carbon and platinum electrodes was also investigated. As expected, nickel deposition on the glassy carbon was not observed after continuous potential cycling between -1.0 and 1.0 V vs SCE in 0.2 M NaOH solution containing 3 mM K2Ni(CN)4. In addition, no nickel deposition was obtained also after about 90 h of electrode immersion in tetracyanonichelate (II) solution. This result confirms the previous hypothesis (see reaction 1 and 2) that nickel precipitation on the 2972 Analytical Chemistry, Vol. 72, No. 13, July 1, 2000
Figure 4. Cyclic voltammograms of nickel film growth on a platinum electrode in a 0.2 M NaOH aqueous solution containing 3 mM K2Ni(CN)4. The potential was continuously cycled at 50 mV s-1 between -0.7 and 0.65 V vs SCE. (a) 10th cycle, (b) 30th cycle, (c) 50th cycle, (d) 70th cycle, (e) 90th cycle. Table 3. Nickel Deposition on the Platinum Electrode Substratea potential window (V vs SCE)
ΓNib (nmol cm-2)
-0.7-0.65 -0.7-0.5 -0.7-0.4 -0.7-0.3 -0.6-0.65 -0.5-0.65 -0.4-0.65
2.8 2.0 1.5 1.1 2.1 1.4 0.2
a The nickel depositions on the platinum electrode substrate were performed by voltage cycling in non-deaerated 0.2 M NaOH solution containing 3 mM K2Ni(CN)4. The potential was scanned 10 times at 50 mV s-1. The nickel surface concentration (ΓNi) of the composite Pt-Ni electrode was estimated by integration of the cathodic peak IIc (referred at the 5th cycle), obtained by cycling the potential at 50 mV s-1 between -0.7 and 0.65 V vs SCE in 0.2 M NaOH solution. b Dependence of the lower and higher potential limits on the nickel surface concentration (ΓNi).
electrode surface is accompanied by a simultaneous complexation of gold species (i.e., AuI) with cyano groups. In sharp contrast to the behavior on glassy carbon, nickel deposition is performed on the platinum substrate by cyclic voltammetry. Figure 4 shows the growth of the nickel film on the platinum electrode (indicated as Pt-Ni) during successive potential cycling between -0.7 and 0.65 V vs SCE in a 0.2 M NaOH solution containing 3 mM K2Ni(CN)4. The redox waves relevant at the NiII/NiIII couple recognizable at about 0.4 V show a voltammetric behavior similar to the nickel film deposited on the gold electrode (see Figure 1). Nevertheless, the rate of nickel deposition is very low if compared with that observed on the gold substrate. In this case, the deposition rate is about 0.02 nmol cycle-1. In view of the results shown above for nickel deposition on gold substrate, the effect of changing the lower and higher potential limits on cyclic voltammograms is also investigated. Table 3 summarizes the relevant results. As can be seen, an extension of the potential window induces an increase in the efficiency of the nickel deposition. In particular, the rate of nickel deposition is strongly influenced by lower potential limits. This result further supports the hypothesis that nickel deposition involves the simultaneous complexation of the platinum species with cyano groups. Nevertheless, in contrast with the gold electrode substrate, no deposition of nickel film was observed either by potentiostatic
Figure 5. Cyclic voltammetric response (5th cycle) at the Au-Ni electrode in 0.2 M NaOH solution containing 9 mM arabinose (solid curve). Scan rate, 50 mV s-1.
conditions in the range of potentials comprised between 0.0 and 0.5 V vs SCE or by simple electrode immersion in alkaline solution containing tetracyanonickelate ions. Perhaps, the high difference of the potential values between the redox couples NiIII/NiII centered at about 0.4 V and Pt II/Pt0 at - 0.4 V (see wave Ic in Figure 4) is responsible for the absence of any deposit of nickel film under potentiostatic conditions or by a simple chemisorption process. Electrocatalytic Activity of the Composite Au-Ni Electrode in Alkaline Media. Considered the greatest asset of the nickel deposition on the gold substrates by potentiodynamic or potentiostatic conditions and/or by simple electrode immersion in alkaline solutions containing tetracyanonichelate ions, the electrocatalytic activity of the composite Au-Ni electrode toward organic compound oxidation is explored. It is well-known that in an alkaline medium both gold and nickel materials show important electroactivity for the oxidation of several classes of organic compounds.31-36 Thus, on the basis of their specific electroactivity, the electrochemical behavior of the composite Au-Ni electrode was tested in an alkaline medium upon addition of a representative carbohydrate (i.e., arabinose). Figure 5 shows the relevant cyclic voltammogram for 9.0 mM arabinose (solid curve) in 0.2 M NaOH. As can be seen, upon addition of arabinose there are two marked increases in current in the region of the Au0 and NiIII (i.e., below 0.25 V and above 0.35 V, respectively). Thus, the concomitant presence on the electrode surface of gold and nickel leads to a greater extension of the potential window in which there are significant oxidation currents. A similar behavior was observed for other polyhydric compounds investigated. XPS Measurements. In this study a comparative XPS analysis using gold or platinum substrates after various treatments in 0.2 M NaOH solution containing 3 mM K2Ni(CN)4 was carried out considering only the changes in the relative intensities of the (31) Beden, B.; Cetin, I.; Kahyaoglu, A.; Takky, D.; Lamy, C. J. Catal. 1987, 104, 37. (32) LaCourse, W. R.; Mead, D. A., Jr.; Johnson, D. C. Anal. Chem. 1990, 62, 220. (33) Johnson, D. C.; Dopperpuhl, D.; Roberts, R.; Vandeberg, P. J. Chromatogr. 1993, 640, 79. (34) Robertson, P. M. J. Electroanal. Chem. 1980, 111, 97. (35) Reim, R. E.; Van Effen, R. M. Anal. Chem. 1986, 58, 3203. (36) Casella, I. G.; Cataldi, T. R. I.; Salvi, A. M.; Desimoni, E. Anal. Chem. 1993, 65, 314.
Figure 6. XPS detailed spectra of (A) the Ni2p3/2 region, (B) the O1s region, and (C) the N1s region for the Au-Ni electrode. The electrode was prepared by cycling the potential between -0.1 and 0.65 V vs SCE (50 mV s-1, 100 cycles) in 0.2 M NaOH solution containing 3 mM K2Ni(CN)4.
detailed Au4f7/2, Pt4f7/2, Ni2p3/2, N1s, and O1s regions. Figure 6 depicts the high-resolution spectrum of the Ni2p3/2, O1s, and N1s regions of the gold substrate electrode cycled between -0.1 and 0.65 V vs SCE in 0.2 M NaOH solution containing 3 mM K2Ni(CN)4. As can be seen, the Ni2p3/2 spectrum shows a complex structure with intense satellite signals of high binding energies adjacent to the main peaks, which are ascribed to a multielectron excitation (shake-up peaks). The main signal Analytical Chemistry, Vol. 72, No. 13, July 1, 2000
2973
Table 4. XPS Data of Composite Au-Ni and Pt-Ni Electrodesa Ni2p3/2
O1s
electrode
treatment
1
2
RNiII/NiIII
Au-Ni Au-Ni Au-Ni Au-Ni Au-Ni Pt-Ni
potentiost (0.0 V)b potentiost (0.3 V)b potentiost (0.5 V)b voltage cyclingc immersiond voltage cyclinge
856.4 856.3 856.2 856.4 856.6 856.3
858.0 858.2 857.9 858.2 858.5 858.0
3.2 3.0 1.7 3.0 4.9 3.9
1 530.8 529.7
N1s
2
3
1
2
532.2 532.2 531.6 532.1 532.1 532.1
533.6 533.7 533.4 533.2 533.2 533.2
398.2 398.1 398.1 397.9 398.0 398.1
400.6 400.2 400.2 399.9 400.1 400.2
3
RN(1)/Ni
402.2 402.0
0.3 0.07 0.07 0.075 0.2 0.3
402.2
a The electrodes were treated in 0.2 M NaOH plus 3 mM K Ni(CN) solution. b The electrode was prepared by cycling the potential between 2 4 -0.1 and 0.65 V vs SCE (50 mV s-1, 50 cycles) and successively polarized at the indicated potential for 4500 s. c The electrode was prepared by -1 cycling the potential between -0.1 and 0.65 V vs SCE (50 mV s , 100 cycles). d The electrode was prepared by simple immersion of the gold substrate in alkaline solution for 43 h. e The electrode was prepared by cycling the potential between -0.7 V and 0.65 V vs SCE (50 mV s-1, 100 cycles).
of the Ni2p3/2 peak gives two contributions of 856.4 ( 0.2 and 858.2 ( 0.3 eV assigned to Ni(OH)2 and NiOOH, respectively.18,37 The relevant O1s signal reported in Figure 6B shows two different peaks at 532.2 ( 0.4 (1) and 533.5 eV (2) assigned to OH- and adsorbed oxygen or water, respectively.18,38,39 Occasionally a third contribution at 530.2 ( 0.6 eV (3) and assigned to O2- is also observed.18 The low content of the O2- contribution on the O1s signal (generally the percent contribution of the O2- peak on the O1s signal is lower than 15%) suggests the substantial absence of the NiO in the nickel-deposited film. Moreover, the shape of the Ni2p3/2 signal confirms the previous hypothesis. Nevertheless, the specific assignment of the O1s peaks is more complex and rather ambiguous owing to uncertainty in the BE values of the various oxygen species and the degree of disorder and roughness of the specific surface considered.18,36-42 Thus, the large peak centered at 532.1 eV (see Figure 6B), assigned to the hydroxyl group (OH-), probably also includes the O2- species. In fact, a broadened O1s band for other similar oxyhydroxide species (i.e., FeOOH) under lower resolution was previously observed.43-45 The lattice oxygen (O2-) line in FeOOH is shifted significantly higher than that for the structurally similar Fe2O3. This suggests that the proton on the hydroxyl oxygen has some interaction with the O2- oxygen. Thus, a partial overlap between the peaks of the OHand O2- species can occur in this experimental context. However, the exact assignment of these features is not essential since the presence on the electrode surface of Ni(OH)2 and NiOOH species is confirmed by the peaks at 856.2 and 858.2 eV of BE observed in the Ni2p3/2 region. The curve fitting of the N1s signal gives three different contributions with binding energies of 398.0 ( 0.2 (1), 400.2 ( 0.3 (2), and 401.9 ( 0.3 eV (3), which, on the basis of their BEs, were assigned to the CN group, the imine group, and amide/amine group, respectively.40,41 Independently of the (37) Visscher, W.; Barendrecht, E. J. Electroanal. Chem. 1983, 154, 69. (38) Kim, K. S.; Winograd, N. Surf. Sci. 1974, 43, 625. (39) Dikinson, T.; Povey, A. F.; Sherwood, P. M. A. J. Chem. Soc., Faraday Trans. 1975, 1(71) 298. (40) Pireaux, J. J.; Liehr, M.; Thiry, P. A.; Delrue, J. P.; Gaudano, R. Surf. Sci. 1984, 141, 221. (41) Briggs, D.; Seah, M. P. Practical Surface Analysis, Vol. 1, 2nd ed.; Wiley & Sons: New York, 1990. (42) Desimoni, E.; Salvi, A. M.; Biader Ceipidor, U.; Casella, I. G. J. Electron Spectrosc. Relat. Phenom. 1994, 70, 1. (43) Kim, K. S.; Vinograd, N.; Davis, R. E. J. Am. Chem. Soc. 1971, 93, 6296. (44) Allen, G. C.; Curtis, M. T.; Hooper, A. J.; Tucker, P. M. J. Chem. Soc., Dalton Trans. 1974, 14, 1525. (45) McIntyre, N. S.; Zetaruk, D. G. Anal. Chem. 1977, 49, 1521.
2974 Analytical Chemistry, Vol. 72, No. 13, July 1, 2000
electrochemical or chemical treatments in alkaline media containing tetracyanonickelate ions, the shape of the Au4f signals is identical to those observed for a pure gold foil obtained after sputtering with 4 kV argon ions. The Au4f region always shows a single spin-orbit split doublet, with a well-resolved Au4f7/2 asymmetric peak at a binding energy of 83.9 ( 0.1 eV, which was assigned to Au0.18,40 In contrast, the Pt4f region relevant at the platinum electrodes, which cycled between -0.7 and 0.65 V vs SCE in the electrolytic medium, shows generally two well-resolved Pt4f7/2 peaks (not shown here) at 71.0 ( 0.3 and 72.3 ( 0.3 eV, which can be ascribed to Pt0 with adsorbed oxygen and/or amide/ imide groups and PtII species, respectively.42,43 The binding-energy values (Ni2p3/2, O1s, and N1s regions) and the relevant surface atomic ratio, obtained from gold and platinum electrode surfaces after treatment in alkaline solution containing tetracyanonickelate ions, are reported in Table 4. While no conclusions regarding the direct reaction between noble metals and cyano groups can be drawn from the XPS study, the atomic surface ratios revealed some interesting features. As can be seen in Table 4, the atomic surface ratio RN1s(1)/Ni2p3/2 is very low and appeared generally between 0.07 and 0.3, while the surface atomic ratio between the total area of the N1s signal and the Ni2P3/2 signal is comprised between 0.7 and 1.1. In contrast, as expected for the powder pellet of K2Ni(CN)4, a ratio of 4 between the area of the N1s and Ni2p3/2 signals was obtained. The substantial absence of cyano groups on the electrode surface, in agreement with the previous results obtained in CV, suggests that in alkaline media an adsorption/ dissociation of the Ni(CN)42- complex on the noble metal substrate occurs before the nickel hydroxide/oxyhydroxide precipitation process. Moreover, the presence on the electrode surface of only Ni(OH)2 and NiOOH species confirms this hypothesis. In addition, it is interesting to underline that the surface composition of the modified electrodes is quite independent of the specific procedure of the electrode modification (i.e., potentiostatic, potentiodynamic, or simple electrode immersion in alkaline solution). Thus, the voltage cycling procedure induces only a catalytic effect on the nickel hydroxide/oxyhydroxide deposition. CONCLUSIONS A new modification procedure of noble metals (i.e., Au, Pt) with nickel oxyhydroxide film obtained by tetracyanonickelate (II) ions in an alkaline medium is described. The nickel film was deposited by voltage cycling, by applying potentiostatic conditions,
or by simple immersion of a gold electrode in 0.2 M NaOH solution plus 3 mM K2Ni(CN)4. The effects on the rate of deposition and growth of the nickel film of various experimental parameters, such as cycle number, potential window, electrode substrate, pH, etc., were evaluated. The nickel film is deposited on noble metals by direct exchange of cyano groups with gold or platinum species and simultaneous precipitation of oxyhydroxide film on the electrode surface. The composite Au-Ni electrode shows interesting catalytic activity toward the electrooxidation of
carbohydrates in an alkaline medium. ACKNOWLEDGMENT This work was supported by Ministero dell’Universita` e della Ricerca Scientifica e Tecnologica (MURST). Received for review December 5, 1999. Accepted March 24, 2000. AC9913863
Analytical Chemistry, Vol. 72, No. 13, July 1, 2000
2975
