Modification of Au Nanoparticles Dispersed on Carbon Support Using ...

pubs.acs.org/Langmuir © 2010 American Chemical Society

Modification of Au Nanoparticles Dispersed on Carbon Support Using Spontaneous Deposition of Pt toward Formic Acid Oxidation Sechul Kim,† Changhoon Jung,† Jandee Kim,† Choong Kyun Rhee,*,†,‡ Sung-Min Choi,§ and Tae-Hoon Lim^ †

Department of Chemistry, Chungnam National University, Daejeon, 305-704, South Korea, ‡Graduate School of Analytical Science and Technology, Chungnam National University, Daejeon, 305-704, South Korea, § Central Laboratory, Hanbat National University, Daejeon, 305-719, South Korea, and ^ Center for Fuel Cell Research, Korea Institute of Science and Technology, Seoul, 136-791, South Korea Received September 9, 2009. Revised Manuscript Received January 10, 2010

This work presents formic acid oxidation on Pt deposits on Au nanoparticles dispersed on Vulcan XC-72R. The Pt deposits were produced using spontaneous deposition method contacting the Au nanoparticles with solutions containing Pt complex ions in various concentrations. The Pt deposits were characterized using CO stripping coulometry, X-ray photoelectron spectroscopy, and inductively coupled plasma atomic emission spectroscopy. When the Pt concentration is 10-5-10-4 M, the Pt deposits are nanoislands of monatomic height. In the concentration range of 10-4-10-3 M, the Pt deposits are most likely two-layer-thick nanofeatures. As Pt concentration increases further, the deposits become wider and thicker. Voltammetric behavior of Pt deposits reveals that on Pt deposits, dehydrogenation path is activated at the expense of poison-forming dehydration path. Furthermore, chronoamperometric measurement of the catalytic activity of Pt deposits supports that the two-layer-thick Pt deposits are most efficient in formic acid oxidation among the studied Pt deposits on Au nanoparticles. The enhancement factor of the particular Pt deposits is 2 in terms of turnover frequency, compared with a commercial Pt catalyst. Details are discussed in conjunction with Pt deposits on Au(111).

1. Introduction Platinum is a major key element of electrocatalysts for oxidation of small organic molecules such as methanol and formic acid.1-5 Because of the high price of the precious metal, in particular, Pt catalysts need to be cost-effective to be utilized in commercial applications. One approach toward practical usage in fuel cells is to disperse Pt nanoparticles on carbon supports to increase their electrochemically effective surface areas, so that Pt atoms in nanoparticles are exposed to fuel molecules as much as possible. As the size of Pt nanoparticles becomes smaller, however, their catalytic behavior normally alters (“particle size effect”).6,7 Concerning the size effect, it is generally accepted that variation in *Corresponding author. Telephone: þ82 42 821 5483. Fax: þ82 42 821 8896. E-mail: [email protected]. (1) Waszczuk, P.; Crown, A.; Mitrovski, S.; Wieckowski, A. In Handbook of Fuel Cells; Vielstich, W., Yokokawa, H., Gasteiger, H. A., Eds.; John Wiley & Sons Ltd.: New York, 2003; Vol. 2, p 635. (2) Lamy, C. Electrochim. Acta 1984, 29, 1581–1588. (3) Hamnett, A. Catal. Today 1997, 38, 445–457. (4) Rice, C.; Ha, S.; Masel, R. I.; Waszczuk, P.; Wieckowski, A.; Barnard, T. J. Power Sources 2002, 111, 83–89. (5) Iwasita, T.; Hoster, H.; John-Anacker, A.; Lin, W. F.; Vielstich, W. Langmuir 2000, 16, 522–529. (6) Park, S.; Xie, Y.; Weaver, M. J. Langmuir 2002, 18, 5792–5798. (7) Kinoshita, K. Electrochemical Oxygen Technology; John Wiley & Sons, Inc: New York, 1992; p 43. (8) Kabbabi, A.; Gloaguen, F.; Andolfatto, F.; Durand, R. J. Electroanal. Chem. 1994, 373, 251–254. (9) Mayrhofer, K. J. J.; Blizanac, B. B.; Arenz, M.; Stamenkovic, V. R.; Ross, P. N.; Markovic, N. M. J. Phys. Chem. B 2005, 109, 14433–14440. (10) Frelink, T.; Visscher, W.; van Veen, J. A. R. J. Electroanal. Chem. 1995, 382, 65–72. (11) Mukerjee, S.; McBreen, J. J. Electroanal. Chem. 1998, 448, 163–171. (12) Han, B. C.; Miranda, C. R.; Ceder, G. Phys. Rev. B 2008, 77, 075410–9. (13) Maillard, F.; Eikerling, M.; Cherstiouk, O. V.; Schreier, S.; Savinova, E.; Stimming, U. Faraday Discuss. 2004, 125, 357–377.

Langmuir 2010, 26(6), 4497–4505

specific catalytic activity (current per unit surface area) comes from geometrical and electronic factors.8-17 Another approach to increase the efficiency of Pt as catalysts for fuel cell applications is to modify their surfaces with various foreign metals. For instance, Ru and Bi would be the most studied surface modifiers for oxidation of methanol1,5 and formic acid,18-20 respectively. In addition, the methods to modify the Pt surfaces include alloying,3,21-29 (14) Godoi, D. R. M.; Perez, J.; Villullas, H. M. J. Electrochem. Soc. 2007, 154, B474–B479. (15) Bergamaski, K.; Pinheiro, A. L. N.; Teixeira-Neto, E.; Nart, F. C. J. Phys. Chem. B 2006, 110, 19271–19279. (16) Markovic, N. M.; Schmidt, T. J.; Stamenkovi, V.; Ross, P. N. Fuel Cells 2001, 1, 105–116. (17) Chrzanowski, W.; Kim, H.; Wieckowski, A. Catal. Lett. 1998, 50, 69–75. (18) Clavilier, J.; Fernandez-Vega, A.; Feliu, J. M.; Aldaz, A. J. Electroanal. Chem. 1989, 258, 89–100. (19) Clavilier, J.; Femandez-Vega, A.; Feliu, J. M.; Aldaz, A. J. Electroanal. Chem. 1989, 261, 113–125. (20) Kim, B.-J.; Kwon, K.; Rhee, C. K.; Han, J.; Lim, T.-H. Electrochim. Acta 2008, 53, 7744–7750. (21) Reddington, E.; Sapienza, A.; Gurau, B.; Viswanathan, R.; Sarangapani, S.; Smotkin, E. S.; Mallouk, T. E. Science 1998, 280, 1735–1737. (22) Iwasita, T.; Nart, F. C.; Vielstich, W. Ber. Bunsen-Ges. Phys. Chem. 1990, 94, 1030. (23) Gasteiger, H. A.; Markovic, N.; Ross, J. P. N.; Cairns, E. J. J. Electrochem. Soc. 1994, 141, 1795–1803. (24) Markovic, N. M.; Gasteiger, H. A.; Ross, P. N.; Jiang, X.; Villegas, I.; Weaver, M. J. Electrochim. Acta 1995, 40, 91–98. (25) Iwasita, T.; Hoster, H.; John-Anacker, A.; Lin, W. F.; Vielstich, W. Langmuir 1999, 16, 522–529. (26) Hoster, H.; Iwasita, T.; Baumgartner, H.; Vielstich, W. J. Electrochem. Soc. 2001, 148, A496–A501. (27) Hoster, H.; Iwasita, T.; Baumgartner, H.; Vielstich, W. Phys. Chem. Chem. Phys. 2001, 3, 337–346. (28) Gurau, B.; Viswanathan, R.; Liu, R.; Lafrenz, T. J.; Ley, K. L.; Smotkin, E. S.; Reddington, E.; Sapienza, A.; Chan, B. C.; Mallouk, T. E.; Sarangapani, S. J. Phys. Chem. B 1998, 102, 9997–10003. (29) Kabbabi, A.; Faure, R.; Durand, R.; Beden, B.; Hahn, F.; Leger, J. M.; Lamy, C. J. Electroanal. Chem. 1998, 444, 41–53.

Published on Web 01/21/2010

DOI: 10.1021/la903357c

4497

Article

electrochemical deposition,30-32 vacuum evaporation,25 and spontaneous adsorption.17,33-37 Understanding of electrocatalysis on Pt nanoparticles is limited. The reason for such a limitation is due to difficulties in experimental approaches at the molecular level, which results from the extremely small sizes of nanoparticles and the complicated structure of carbon supports. One of the research approaches concerning the electrocatalysis on Pt surfaces is examination of disk-type Pt surfaces of various crystallographic orientations. Formic acid oxidation on Pt single crystal electrode surfaces would be an excellent example. Specifically, Pt(100) and Pt(110) are known to be very active, despite significant catalytic poison formation, while the Pt(111) surface is less active, albeit with a smaller amount of catalytic poison than Pt(100) and Pt(110).38,39 Furthermore, the monatomic steps on Pt(111) surface promote poison formation to decrease the catalytic performance of Pt(111), while the monatomic steps on Pt(100) and Pt(110) reduce poison formation to enhance the activities of Pt(100) and Pt(110).40 However, Pt nanoparticles show unique catalytic behavior different from those of disk-type Pt electrodes of polycrystalline, (111) and (100) surfaces as exemplified by us.41 An alternative to the approach to practical nanoparticle electrodes would be experimental mimicking of nanoparticles on flat substrates.42,43 One example for such efforts would be colloidal Pt nanoelectrodes on polycrystalline Au disk electrode, revealing that the transient of CO oxidation on Pt nanoparticles of 3 nm differs from those on conventional Pt electrodes, although that on Pt nanoelectrodes larger than 16 nm is similar. To understand the oxidation of formic acid and methanol on Pt nanoparticles, our group has carried out a mimicking experiment on Pt deposits in nanometer sizes on Au(111).44 Comparing the oxidative behavior of formic acid and methanol on Pt(111), Pt deposits on Au(111) were found to enhance formic acid oxidation significantly but to prohibit methanol oxidation. It was clearly demonstrated that the enhancement of formic acid oxidation was due to suppression of poison-forming dehydration (HCOOH f H2O þ poison, and poison þ H2O f CO2 þ 2 Hþ þ 2e) and simultaneous promotion of dehydrogenation (HCOOH f CO2 þ 2 Hþ þ 2e), while the negligible oxidation of methanol was ascribed to lack of adsorption of methanol. A strong correlation was found between the catalytic activities and the structural atomic-level information obtained using electrochemical scanning tunneling microscopy (EC-STM). Specifically, the maximum current of formic acid oxidation was obtained on (30) Frelink, T.; Visscher, W.; van Veen, J. A. R. Surf. Sci. 1995, 335, 353–360. (31) Frelink, T.; Visscher, W.; van Veen, J. A. R. Langmuir 1996, 12, 3702–3708. (32) Abd El Meguid, E. A.; Berenz, P.; Baltruschat, H. J. Electroanal. Chem. 1999, 467, 50–59. (33) Kim, H.; Rabelo de Moraes, I.; Tremiliosi-Filho, G.; Haasch, R.; Wieckowski, A. Surf. Sci. 2001, 474, L203–L212. (34) Babu, P. K.; Tong, Y. Y.; Kim, H. S.; Wieckowski, A. J. Electroanal. Chem. 2002, 524-525, 157–167. (35) Tong; Kim, H. S.; Babu, P. K.; Waszczuk, P.; Wieckowski, A.; Oldfield, E. J. Am. Chem. Soc. 2001, 124, 468–473. (36) Crown, A.; Wieckowski, A. Phys. Chem. Chem. Phys. 2001, 3, 3290–3296. (37) Strbac, S.; Johnston, C. M.; Lu, G. Q.; Crown, A.; Wieckowski, A. Surf. Sci. 2004, 573, 80–99. (38) Iwasita, T.; Xia, X. H.; Liess, H. D.; Vielstich, W. J. Phys. Chem. B 1997, 101, 7542–7547. (39) Iwasita, T.; Xia, X.; Herrero, E.; Liess, H.-D. Langmuir 1996, 12, 4260– 4265. (40) Maci, M. D.; Herrero, E.; Feliu, J. M. Electrochim. Acta 2002, 47, 3653– 3661. (41) Rhee, C. K.; Kim, B.-J.; Ham, C.; Kim, Y.-J.; Song, K.; Kwon, K. Langmuir 2009, 25, 7140–7147. (42) Friedrich, K. A.; Henglein, F.; Stimming, U.; Unkauf, W. Electrochim. Acta 2000, 45, 3283–3293. (43) Friedrich, K. A.; Marmann, A.; Stimming, U.; Unkauf, W.; Vogel, R. Fresenius J. Anal. Chem. 1997, 358, 163–165. (44) Kim, J.; Jung, C.; Rhee, C. K.; Lim, T.-h. Langmuir 2007, 23, 10831–10836.

4498 DOI: 10.1021/la903357c

Kim et al.

two-layer-thick and 5 nm wide Pt deposits. As the Pt deposits on Au(111) become wider and thicker, however, their activities of oxidation of methanol and formic acid approached to those of Pt(111), in terms of turnover frequency. Namely, small and thin Pt deposits are more effective in formic acid oxidation, while wide and thick ones are efficient in methanol oxidation. The aim of this work is to examine the electrocatalytic activity of formic acid oxidation on Pt deposits on Au nanoparticles toward more practical applications than our previous work on Au(111).44 Pt deposits on Au nanoparticles were produced using spontaneous deposition of Pt onto the surfaces of Au nanoparticles. The Pt deposits were characterized by the CO stripping method, X-ray photoelectron spectroscopy (XPS), and inductively coupled plasma atomic emission spectroscopy (ICP-AES). The electrochemical behavior of the Pt deposits in formic acid oxidation was scrutinized using cyclic voltammetry and chronoamperometry. It was found that in terms of turnover frequency, Pt deposits on Au nanoparticles are more efficient that a commercial Pt nanoparticle catalyst in the oxidation of formic acid. A comparison of the Pt deposits on Au nanoparticles with those on Au(111) is fully discussed.

2. Experimental Section Au nanoparticles dispersed on a carbon support were prepared using citrate method.45 The specific preparation procedure is as follows: (a) 800 mL of 1 wt % sodium citrate solution (98%, Sigma-Aldrich) was mixed with 800 mL of a solution containing 0.0666 g of HAuCl4 3 H2O (ACS reagent, Sigma-Aldrich), (b) 10 mL of ice-cold 0.0035 wt % NaBH4 (reagent grade, g98.5%, Sigma-Aldrich) solution was slowly added into the solution mixture being stirred to produce Au nanoparticles, (c) the solution of Au nanoparticles was mixed with a solution dispersing carbon supports, Vulcan XC-72R (Cabot International), so that the Au nanoparticles settle on the supports during stirring for 24 h, and (d) the carbon supports with Au nanoparticles were filtered, washed, and dried at 70 C under vacuum for more than 3 h. The Au nanoparticles on the carbon supports were characterized with a transmission microscope (TEM, JEM-2010F, JEOL). The loaded amount of Au was determined gravimetrically with a thermal gravimeter (TA Instruments, SDT 2960 Simultaneous DTA-TGA) to be 10 ( 1 wt %. The surfaces of the Au nanoparticles on the carbon support were modified using spontaneous deposition of Pt. The spontaneous deposition of Pt was achieved by contacting the Au nanoparticles for 3 h with 0.5 M H2SO4 (95-97%, Merck) containing Pt complex ions as prepared with H2PtCl6 3 6H2O (98%, Wako). After stirring, the Au nanoparticles modified with Pt were thoroughly washed with 0.5 M H2SO4 solution and electrochemically reduced at 0.1 V (versus a Ag/AgCl reference electrode) in 0.5 M H2SO4 solution for 30 min to stabilize the Pt adsorbates for the following treatments such as filtering, drying and grinding. The electrochemical reduction of the adsorbed Pt species was carried out by placing the catalyst powder in a Au bowl, immersing the whole bowl into 0.5 M H2SO4 solution, and applying electrochemical potential in a conventional way. The Pt concentrations of the studied solutions containing Pt ions were in the range 10-5-10-1 M. In electrochemical measurements, a conventional three electrode system was employed. The working electrodes employed in this work were prepared by spreading a slurry mixture of the Ptmodified Au nanoparticles, 5% Nafion solution (Wako) and water, on a Au disk electrode, as detailed in refs 20 and 41. The counter electrode was a Pt gauze, and the reference electrode was a homemade Ag/AgCl electrode in 1.0 M NaCl. The potential (45) Hayat, M. A. Colloidal Gold: Principles, Methods, and Applications; Academic Press: San Diego, CA, 1989.

Langmuir 2010, 26(6), 4497–4505

Kim et al.

Article

reported in this work is against the reference electrode. Cyclic voltammetry and chronoamperometry were employed for diagnostic purpose and catalytic activity measurements, respectively. In chronoamperometric measurements, a rotating disk electrode configuration with rotating speed of 1000 rpm was utilized to remove CO2 bubbles efficiently. The concentration of formic acid (99%, Wako) in 0.5 M H2SO4 solution was 2.0 M. Before electrochemical measurements, the working solutions were purged with N2 at least for 30 min. Also, a commercially available Pt nanoparticle catalyst (HiSPEC2000, Johnson Matthey) was investigated as a reference. The deposited amounts of Pt on Au nanoparticles were measured with three independent methods: the coulometry of adsorbed CO, XPS, and ICP-AES. For measurements of CO stripping charges, CO adsorption was carried out by immersing the working electrodes into 0.5 M H2SO4 solution saturated with a CO gas (99.99%, Air Liquid Korea) at 0 V for 15 min. Stripping of CO was performed in CO-free 0.5 M H2SO4 solution after rinsing the electrochemical cell with a copious amount of 0.5 M H2SO4 solution under potential control.46 XPS measurements were carried out using an instrument (Thermo Electron Co., MultiLab 2000, USA) utilizing a monochromatic Mg KR X-ray beam (1253.6 eV) as an excitation source and a multichannel hemispherical electron energy analyzer operated at a constant pass energy of 15 eV. The observed spectra were curve-fitted using a mixed Gaussian-Lorentzian line shape and Shirley baselines. In ICP-AES measurements, the sample powders of 0.05 g were completely dissolved in a mixture of H2SO4 (2 mL), HNO3 (3 mL) and HCl (9 mL) at 180 C for 2 h. After dilution to 25 mL, the solutions were fed into an instrument (iCAP 6500, Thermo), and the concentrations of Pt and Au were determined using proper calibration curves.

Figure 1. Typical TEM micrographs of Au nanoparticles (a) before and (b) after loading on Vulcan XC-72R. Size distributions of Au nanoparticles (c) before and (d) after loading.

3. Results and Discussion 3.1. Au Nanoparticles on Vulcan XC-72R. Figure 1 shows micrographs of Au nanoparticles before and after loading on Vulcan XC-72R. A typical picture of Au nanoparticles before loading on the carbon support, Figure 1a, clearly shows that the distribution of Au particles is homogeneous in size. After loading, the Au nanoparticles are dispersed evenly on the carbon support without any spatial preference (Figure 1b). However, the size distributions of the Au nanoparticles before and after loading are not identical to each other as demonstrated in Figure 1, parts c and d. Specifically, the size distribution before loading (4.8 ( 1.1 nm) is very narrow, while the distribution after loading (6.1 ( 3.0 nm) is relatively broad. In particular, larger Au particles not notable before loading are observable frequently as indicated with the arrows in Figure 1b. Thus, the difference in the size distributions before and after loading Au nanoparticles on the carbon supports is likely ascribable to the aggregation of Au nanoparticles during the loading procedure or to the presence of an excess NaBH4 which, during the loading step, modifies the distribution of the initial small Au nanoparticles. 3.2. Characterization of Pt Spontaneously Deposited on Au Nanoparticles. Figure 2 shows the cyclic voltammograms of Au nanoparticles on Vulcan XC-72R, modified using spontaneous deposition of Pt. Figure 2a is a cyclic voltammogram of plain Au nanoparticles, showing surface redox peaks (1.2 V for oxidation and 0.9 V for reduction) which are characteristics of polycrystalline Au electrodes. Compared with the cyclic voltammogram of a conventional Au disk electrode (Figure S1b in the Supporting Information), the current observed in the double layer region (-0.28 to þ0.8 V) is fairly large, which may be due to electrochemical double layer charging of the carbon support (46) Jung, C.; Kim, J.; Rhee, C. K. Langmuir 2007, 23, 9495–9500.

Langmuir 2010, 26(6), 4497–4505

Figure 2. Cyclic voltammograms of Pt deposits on Au nanoparticles in 0.5 M H2SO4. The Pt concentrations for spontaneous deposition of Pt are (a) 0 M, (b) 10-5 M, (c) 10-4 M, (d) 10-3 M, (e) 10-2 M, and (f ) 10-1 M. The cyclic voltammograms of plain Au nanoparticles (dashed lines) are juxtaposed for convenience. The current scale is normalized to unit weight of Au. Scan rate: 10 mV/s.

(Figure S1c in the Supporting Information).47 Spontaneous deposition of Pt on Au nanoparticles induces a modification in the cyclic voltammogram of plain Au nanoparticles, depending on Pt concentration. When Pt concentration is lower than 10-4 M, the modification is not significant (Figure 2, parts b and c). However, Pt concentrations higher than 10-4 M change the voltammograms substantially as shown in Figure 2d-f. As Pt concentration increases, specifically, a broad anodic current starting at 0.6 V becomes notable, while a cathodic peak at 0.4 V appears simultaneously. Because a Pt disk electrode shows an oxidation current starting at 0.6 V and a reduction peak at 0.47 V in the same solution (Figure S1, parts d and f in the Supporting Information), it is obvious that the observed voltammetric features result from the redox behavior of the Pt deposits on (47) Kim, T.; Ham, C.; Rhee, C. K.; Yoon, S.-H.; Tsuji, M.; Mochida, I. Electrochim. Acta 2008, 53, 5789–5795.

DOI: 10.1021/la903357c

4499

Article

Kim et al.

Figure 3. Plots of electrochemical coverage of Pt (0) and spectroscopic atomic ratio of Pt to Au measured using XPS (4) and ICP-AES (O) as a function of Pt concentration. The charge density is normalized to unit weight of Au.

Au nanoparticles. Furthermore, the current peak of surface Au oxide reduction at 0.9 V is suppressed as Pt concentration increases, indicating that the surfaces of Au nanoparticles are covered with Pt deposits. Contrasting to the clearly notable Pt surface redox behavior, the charges related to hydrogen adsorption are not quite well developing, although they are conceivable. The particular poor definition of the hydrogen region may be related to the difference in the sizes of Pt objects: Pt deposits in a nanometer scale on Au substrate and conventional Pt electrodes in a scale larger than micrometer. Such a poorly defined hydrogen region on Pt object in a nanometer regime has been reported by us44 and independently by others.48 On the other hand, a possibility of adsorption of Pt ions on the surfaces of the carbon supports would be discarded as fully discussed with experimental evidence in Supporting Information. The amounts of Pt spontaneously deposited on Au nanoparticles were determined using three independent methods: stripping coulometry of adsorbed CO, XPS, and ICP-AES. The coulometry of CO stripping provide directly the numbers of electrochemically active Pt atoms (i.e., exposed to electrolyte), while XPS results offer at least semiquantitative Pt/Au atomic ratios near the surfaces (i.e., within escape depths of photoelectrons). The ICP-AES measurements provide the atomic ratios of Pt to the whole Au, including their absolute concentrations. The deposited amounts of Pt, as measured using the three methods, are displayed in Figure 3 and discussed in detail as below. Figure 4 shows typical stripping voltammograms of CO adsorbed on Pt spontaneously deposited on Au nanoparticles. On plain Au nanoelectrodes, there is no current relevant to CO oxidation (Figure 4a). After spontaneous deposition of Pt, a CO stripping peak with a long tail becomes discernible at 0.55 V, whose charge depends upon the Pt concentration of the Ptcontaining solution (Figure 4b-f ). In Figure 3, the absolute stripping charge of CO per unit mass of Au and the coverage of Pt are plotted simultaneously as a function of Pt concentration. Here, the Pt coverages (θ) on Au nanoparticles (defined as the number of electrochemically active Pt atoms to the number of surface Au atoms) were calculated according to the following equation: θPt ¼ ðQCO =420 μC=cm2 Þ=ðQAu =400 μC=cm2 Þ (48) Kristian, N.; Yu, Y.; Gunawan, P.; Xu, R.; Deng, W.; Liu, X.; Wang, X. Electrochim. Acta 2009, 54, 4916–4924.

4500 DOI: 10.1021/la903357c

Figure 4. Stripping voltammograms of CO adsorbed on Pt deposits on Au nanoparticles in 0.5 M H2SO4 solution. The Pt concentrations for spontaneous deposition of Pt are (a) 0 M, (b) 10-5 M, (c) 10-4 M, (d) 10-3 M, (e) 10-2 M, and (f ) 10-1 M. The cyclic voltammograms of Pt deposits on Au nanoparticles after stripping of CO (faint solid lines) are juxtaposed for convenience. The current scale is normalized to unit weight of Au. Scan rate: 10 mV/s.

Figure 5. X-ray photoelectron spectra of Pt 4f region obtained from Pt deposits on Au nanoparticles. The Pt concentrations for spontaneous deposition of Pt are (a) 10-1 M, (b) 10-3 M, (c) 10-4 M, and (d) 10-5 M. The curve-fitting indicates that the observed spectra consist of the Pt 4f photoelectrons from metallic Pt, PtO, and PtO2, and the Au 5p photoelectrons.

where QCO and QAu stand for the CO stripping charge on Pt deposits and the surface oxide reduction charge of plain Au nanoparticles, respectively. The specific numbers, 420 and 400 μC/cm2, are the CO stripping charge per unit surface area on Pt(111)13,49 and the surface oxide reduction charge of polycrystalline Au,50 respectively. Figure 5 shows typical XPS spectra of Pt spontaneously deposited on Au nanoparticles. It is clear that as the Pt concentration of the Pt-containing solution increases, the amount of deposited Pt increases as revealed by the increase in the peak areas. For further analysis, the obtained peaks were curve-fitted, based on an assumption that the peaks from Pt, PtO, PtO2 and Au were involved. In particular, it is worthy to address that because (49) Feliu, J. M.; Orts, J. M.; Femandez-Vega, A.; Aldaz, A.; Clavilier, J. J. Electroanal. Chem. 1990, 296, 191–201. (50) Trasatti, S.; Petrii, O. A. Pure Appl. Chem. 1991, 63, 711–734.

Langmuir 2010, 26(6), 4497–4505

Kim et al.

of small surface concentration of Pt on Au nanoparticles, Au 5p1/2 peak (the sensitivity factors for Au 5p1/2 and Pt 4f7/2 are 0.353 and 8.890, respectively51) was included as one of the possible components in the curve-fitting procedure. Indeed, six peaks of the three Pt species and one peak related to Au appear as significant components of the observed peaks. Qualitatively, the Pt 4f7/2 peaks of metallic Pt, PtO, and PtO2 appear at 71.0, 73.5, and 75.6 eV, respectively. Quantitatively, the peak areas of the Pt species indicate that the Pt adlayers on the Au nanoparticles crudely consist of 35% metallic Pt, 50% PtO, and 15% PtO2. The presence of the Pt oxides, although the spontaneously deposited Pt was reduced electrochemically, is trivial due to air exposure during further handing in air. The atomic ratio of the deposited Pt to the Au substrate were semiquantitatively calculated with the ratio of the Pt 4f7/2 peak areas of the three Pt species to the peak area of Au 4f7/2 (not shown here), on the assumption that the sensitivity factors of Pt 4f7/2 and Au 4f7/2 are 8.890 and 9.790, respectively,51 and are plotted in Figure 3. It is worthy to address a detail concerning measurement error in ICP-AES experiments. The observed relative errors in determination of Au concentration are in the range of 10-15%. Because the loaded amount of Au is 10 ( 1 wt %, the observed errors should be originated from ICP-AES measurements, most likely from the sampling preparation procedure, not from ICP-AES itself. Indeed, when the Pt concentration is higher than 10-3 M (or, the dissolved Pt amount is significant), the values of Pt contents scatter with an error range of 10-15%. Another thing to clarify is that when the Pt concentration is 10-5 M (equivalently, the dissolved Pt concentration is extremely small.), the values of Pt deviate by a factor of 2-3, indicating a serious preparation error. Thus, the atomic ratios of the concentrations observed using ICP-AES, except the one obtained when the Pt concentration is 10-5 M, are presented in Figure 3. Figure 3 shows the variations of the surface amount of Pt spontaneously deposited on Au nanoparticles as measured with the three independent electrochemical and spectroscopic means. It is clearly revealed that the spontaneous deposition of Pt on Au nanoparticle surfaces becomes significant when Pt concentration is higher than 10-4 M. Furthermore, it is notable that when Pt concentration is more than 10-3 M, there are some differences in the variation of the measured surface amount of Pt. The electrochemical coverages (i.e., surface atomic ratios) are larger than the spectroscopic atomic ratios measured using XPS and ICP-AES. Specifically, the electrochemical atomic ratios are larger than the XPS and ICP-AES ones roughly by a factor of 2 and 4, respectively. It should be recalled that XPS technique can measure a few Au layers below the surface with in a sampling depth, while the electrochemical method can measure only the surface atoms directly exposed to the electrolyte. In addition, it should be emphasized that the ICP-AES method measures the whole Au and Pt on the carbon supports. Thus, the atomic ratios estimated with XPS and ICP-AES techniques are inherently smaller than those measured with the electrochemical method. On the other hand, the electrochemically measured coverage declines certainly in the Pt concentration region of 10-3-10-1 M, while the spectroscopic atomic ratios incline slightly. Despite a minute increase in the surface amount of Pt on Au nanoparticle surfaces, the decline in the number of electrochemically active Pt atoms may imply that as the concentration of Pt ion increases in solution phase, the number of layers in islands of the Pt ion increases, so that the number of Pt atoms exposed to an electrochemical environment decreases after electrochemical reduction. Our (51) Scofield, J. H. J. Electron Spectrosc. Relat. Phenom. 1976, 8, 129–137.

Langmuir 2010, 26(6), 4497–4505

Article

Figure 6. Plot of specific surface area of Pt deposits on Au nanoparticles as a function of Pt concentration. The surface areas and the mass were obtained using CO stripping charges and ICP-AES, respectively. The specific area of commercial Pt nanoparticles is displayed with a red solid line. The hypothetical surface areas of 1 g of Pt are illustrated in cyan and blue lines, when the Pt atoms are spread into monolayers of (100) and (111) arrangements, respectively.

previous work on spontaneous deposition of Pt on Au(111)44 has demonstrated that the coverage of electrochemically active Pt decreases after reaching to a maximum value (0.25) as the Pt concentration increases. Such a decline in the number of electrochemically active Pt atoms has been ascribed, based on EC-STM measurements, to the formation of multilayer nanoislands due to the high surface energy of Pt.52 As the number of layers of the deposited Pt increases, two phenomena acting in opposite ways in XPS measurements takes place. One is the attenuation of Au signal by thicker and wider Pt deposits, and the other one is the attenuation of Pt signals due to forming multiple-layer deposits. As a result, the XPS atomic ratio may increase slightly. On the other hand, the ICP-AES atomic ratio would show a rather small increase because an increment in Pt concentration is divided by a relatively high Au concentration. Figure 6 provides another aspect concerning the morphology of Pt deposits on Au nanoparticles. The specific surface area of Pt deposits (m2/gPt) is plotted against the concentration of Pt. The surface area and amount of Pt were measured using CO stripping and ICP-AES. It should be noted that because the amount of Pt obtained using ICP-AES after contacting with Pt solution of 10-5 M was very small and scattered by a factor of 2-3, the specific value are not included in Figure 6. Considering the significant error propagation especially in division by a extreme small number, the exclusion of the particular value would be justified. When the Pt concentration is 10-4 M, the surface area of Pt deposits is roughly 160 m2/gPt. A simple calculation tells that if 1 g of Pt is spread to a monatomic layer, the specific surface area would be 204 and 237 m2/gPt in (111) and (100) crystallographic geometries, respectively. Also, the horizontal line indicates the specific value of the commercial Pt nanoparticles (around 110 m2/gPt) whose diameter is around 3 nm. Then, the specific value of 160 m2/gPt indicate that all of the Pt deposits produced with 10-4 M Pt solution are not certainly deposits of monatomic layer; in other words, there are Pt deposits of multiple layers although the entire deposits are not multilayer deposits. When the Pt concentration is more than 10-3 M, on the other hand, the specific surface area becomes lower than 100 m2/gPt, indicating that the Pt deposits becomes thicker. If Pt deposits are distinctive particles such as cuboctahedrons or icosahedrons, then (52) Bauer, E.; van der Merwe, J. H. Phys. Rev. B 1986, 33, 3657.

DOI: 10.1021/la903357c

4501

Article

Kim et al.

Figure 7. Schematics of Pt deposits on Au nanoparticles. The Pt concentrations for spontaneous deposition of Pt are (a) below 10-4 M (region I), (b) in the range of 10-4-10-3 M (region II), and (c) above 10-3 M (region III).

the size of Pt deposits should be larger than 3 nm to have smaller specific surface area than the commercial Pt nanoparticles. Not observed at all was an increase in the number of particle after contacting Au nanoparticles with a Pt solution whose concentration is higher than 10-3 M. Figure 7 illustrates schematics of Pt deposits on Au nanoparticles, depending on Pt coverage. The variation of electrochemical Pt coverage is closely similar to that observed on Pt deposits on Au(111), so that the studied concentration range of Pt in this work could be classified into three regimes as we have done in ref 44. Regime I is the one where the electrochemical Pt coverage is relatively low (