Langmuir 2007, 23, 10831-10836
10831
Electrocatalytic Oxidation of Formic Acid and Methanol on Pt Deposits on Au(111) Jandee Kim,† Changhoon Jung,† Choong K. Rhee,*,† and Tae-hoon Lim‡ Department of Chemistry, Chungnam National UniVersity, Daejeon 305-764, Korea, and Center for Fuel Cell Research, Korea Institute of Science & Technology, Seoul 136-791, Korea ReceiVed May 12, 2007. In Final Form: July 22, 2007 This work presents characteristics of Pt deposits on Au(111) obtained by the use of spontaneous deposition and investigated by electrochemical scanning tunneling microscopy (EC-STM). On such prepared and STM characterized Au(111)/Pt surfaces, we studied electrocatalytic oxidation of formic acid and methanol. We show that the first monatomic layer of Pt displays a (x3 × x3)R30° surface structure, while the second layer is (1 × 1). After prolonged deposition, multilayer Pt deposits are formed selectively on Au(111) surface steps and are 1-20 nm wide and one to five layers thick. On the optimized Au(111)/Pt surface, formic acid oxidation rates are enhanced by a factor of 20 compared to those of pure Pt(111). The (x3 × x3)R30°-Pt yields very low methanol oxidation rates, but the rates increase significantly with further Pt growth.
1. Introduction Platinum is an important catalyst in electrocatalytic reactions, such as methanol oxidation toward direct methanol fuel cell.1,2 Practical platinum catalysts are in the shape of a nanoparticle on carbon supports. The sizes of such Pt particles are a few nanometers in diameter,3-7 mainly to increase the electrochemically effective surface area. Because of the nanoscale size, however, the electrochemical properties of Pt nanoelectrodes are different from those of Pt electrodes of conventional sizes. On commercially available Pt nanoelectrodes on Vulcan, for example, oxidation of formic acid increases as the size of the Pt nanoelectrode decreases, while oxidation of methanol varies in a reverse way.8 This example implies that special attention should be paid in understanding the electrochemistry of Pt nanoelectrodes. However, it is extremely difficult to experimentally approach practical Pt nanoelectrodes, especially at the molecular or atomic level, due to their extremely small sizes and the complicated structures of the carbon supports. Another way to experimentally access Pt nanoelectrodes would be via Pt deposits on catalytically inert and flat supports, that is, mimicking Pt nanoelectrodes on carbon supports. In such a direction, Pt deposits on HOPG9-11 and Au substrates12-21 have been investigated. The numerous investigations revealed various aspects of electrocatalytic reactions on Pt nanoelectrodes: for * To whom correspondence should be addressed. E-mail: ckrhee@ cnu.ac.kr. Fax: 82-42-821-8896. Telephone: 82-42-821-5483. † Chungnam National University. ‡ Korea Institute of Science & Technology. (1) Chrzanowski, W.; Wieckowski, A. Interfacial Electrochemistry; Marcel Dekker: New York, 1999; p 937. (2) Iwasita, T. Methanol and CO electrooxidation. In Handbook of Fuel Cells: Fundamentals, Technology and Applications; Vielstich, W., Lamm, A., Gasteiger, H. A., Eds.; John Wiley & Sons: Chichester, 2003; Vol. 2, p 603. (3) Arico, A. S.; Baglio, V.; Modica, E.; Blasi, A. D.; Antonucci, V. Electrochem. Commun. 2004, 6, 164. (4) Boxall, D. L.; Deluga, G. A.; Kenik, E. A.; King, W. D.; Lukehart, C. M. Chem. Mater. 2001, 13, 891. (5) King, W. D.; Corn, J. D.; Murphy, O. J.; Boxall, D. L.; Kenik, E. A.; Kwiatkowski, K. C.; Stock, S. R.; Lukehart, C. M. J. Phys. Chem. B 2003, 107, 5467. (6) Li, W.; Wang, X.; Chen, Z.; Waje, M.; Yan, Y. J. Phys. Chem. B 2006, 110, 15353. (7) Su, F.; Zeng, J.; Bao, X.; Yu, Y.; Lee, J. Y.; Zhao, X. S. Chem. Mater. 2005, 17, 3960. (8) Park, S.; Xie, Y.; Weaver, M. J. Langmuir 2002, 18, 5792.
example, CO oxidation on colloidal Pt nanoelectrodes on polycrystalline Au depends on the sizes of the Pt particles.13 Specifically, the transient of CO oxidation on 3 nm Pt particles is significantly different from that on polycrystalline Pt electrodes of conventional sizes, while those on Pt particles larger than 16 nm are similar. The aim of this work is to understand oxidation of formic acid and methanol on Pt deposits on Au(111) produced by the use of spontaneous deposition. The atomically flat Au(111) substrate makes it possible to characterize the Pt deposits with electrochemical scanning microscopy (EC-STM). In parallel, oxidation of formic acid and methanol is studied on the surfaces of the Pt deposits. Thus, we correlate the structural information of Pt deposits with their catalytic activities in the oxidation of formic acid and methanol. Specifically, formic acid oxidation favors small and thin Pt deposits, while methanol oxidation prefers to wide and thick ones. 2. Experimental Section The single crystal electrodes of Au and Pt used in this work were prepared with the well-known bead method. In voltammetric works, a single crystal bead at the end of a wire of Au (99.999%, 0.5 mm diameter, Aldrich) or Pt (99.99%, 0.5 mm diameter, Aldrich) was cut and polished to mirror-finish. In EC-STM works, one of the (9) Lee, I.; Chan, K.-Y.; Phillips, D. L. Appl. Surf. Sci. 1998, 136, 321. (10) Lee, I.; Chan, K.-Y.; Phillips, D. L. Ultramicroscopy 1998, 75, 69. (11) Zoval, J. V.; Lee, J.; Gorer, S.; Penner, R. M. J. Phys. Chem. B 1998, 102, 1166. (12) Brankovic, S. R.; Wang, J. X.; Adzic, R. R. Surf. Sci. 2001, 474, L173. (13) Friedrich, K. A.; Henglein, F.; Stimming, U.; Unkauf, W. Electrochim. Acta 2000, 45, 3283. (14) Friedrich, K. A.; Marmann, A.; Stimming, U.; Unkauf, W.; Vogel, R. Fresenius’ J. Anal. Chem. 1997, 358, 163. (15) Guilln-Villafuerte, O.; Garcia, G.; Anula, B.; Pastor, E.; Blanco, M. C.; Lopez-Quintela, M. A.; Hernandez-Creus, A.; Planes, G. A. Angew. Chem., Int. Ed. 2006, 45, 4266. (16) Kokkinidis, G.; Stoychev, D.; Lazarov, V.; Papoutsis, A.; Milchev, A. J. Electroanal. Chem. 2001, 511, 20. (17) Kongkanand, A.; Kuwabata, S. J. Phys. Chem. B 2005, 109, 23190. (18) Nagahara, Y.; Hara, M.; Yoshimoto, S.; Inukai, J.; Yau, S.-L.; Itaya, K. J. Phys. Chem. B 2004, 108, 3224. (19) Pedersen, M. Ø.; Helveg, S.; Ruban, A.; Stensgaard, I.; Lægsgaard, E.; Nørskov, J. K.; Besenbacher, F. Surf. Sci. 1999, 426, 395. (20) Uosaki, K.; Ye, S.; Naohara, H.; Oda, Y.; Haba, T.; Kondo, T. J. Phys. Chem. B 1997, 101, 7566. (21) Waibel, H.-F.; Kleinert, M.; Kibler, L. A.; Kolb, D. M. Electrochim. Acta 2002, 47, 1461.
10.1021/la701377n CCC: $37.00 © 2007 American Chemical Society Published on Web 09/06/2007
10832 Langmuir, Vol. 23, No. 21, 2007
Kim et al. was rinsed promptly with water, and the electrode was immediately subjected to a cathodic voltammetric scan to reduce the deposited layer of Pt. In EC-STM experiments, a Au single crystal bead, covered with Pt deposits and protected with water, was assembled into a homemade cell, and the cell was immediately filled with 0.05 M H2SO4 solution. The cell was then inserted into STM instrument (Nanoscope III), and the electrode potential was held at 0 V to reduce the Pt deposits. To initiate an EC-STM experiment, a W tip (99.9%, 0.5 mm, Aldrich), electrochemically etched and insulated with molten polyethylene, was engaged instantly onto one of the (111) facets of a Au single crystal bead. Evaluation of catalytic activities of Pt deposits on Au(111) electrodes was carried out using 0.1 M methanol (99.9%, HPLC grade, Johnson Matthey) and 0.1 M formic acid (99%, Suprapur, Merck) in 0.05 M H2SO4 solution. In voltammetric and EC-STM measurements, a conventional three electrode system, consisting of a Pt counter electrode and a Ag/ AgCl reference electrode, was employed. The potentials reported in this work were against the reference electrode.
3. Results and Discussion
Figure 1. Typical cyclic voltammograms of Pt deposits on Au(111) formed by the use of spontaneous deposition of PtCl62- in 0.05 M H2SO4 solution. The respective deposition times are (a) 0 s (i.e., clean Au(111)), (b) 10 s, (c) 60 s, (d) 150 s, (e) 600 s, and (f) 1500 s. Scan rate: 50 mV/sec.
(111) facets on the bead surface was used as prepared. In both works, the clean and ordered surfaces were obtained after annealing in a hydrogen flame and subsequent quenching in hydrogen-saturated water. Spontaneous deposition of Pt on Au(111) electrodes was exclusively used in this work to obtain Pt deposits. After preparation of a Au(111) electrode by flaming and quenching, the electrode was dipped into an aqueous solution of 40 µM H2PtCl6 (99.9%, Wako, Japan) and 0.05 M H2SO4 (Suprapur, Merck) for a proper amount of time. The residual Pt-containing solution on the electrode surface
3.1. Electrochemical Behavior of Pt Adsorbates on Au(111). Figure 1 shows typical cyclic voltammograms of Pt deposits on Au(111) obtained by the use of spontaneous deposition. The spontaneous deposition of Pt was carried out by dipping a clean Au(111) electrode (Figure 1a; see Figure S1 of the Supporting Information for details) into 40 µM H2PtCl6 + 0.05 M H2SO4 solution for a proper amount of time.18 After removing the remnant of the Pt-containing solution on the electrode surface with water, the open circuit potential was measured to be roughly 0.6 V in 0.05 M H2SO4 solution. A reductive current peak of spontaneously deposited Pt was normally observed at 0.35 V in the following initial cathodic scan. As shown in Figure 1b-f, the steady-state voltammograms are significantly different from that of bare Au(111). In the voltammograms, characteristics of Pt are clearly seen, that is, hydrogen adsorption region (-0.25-0.03 V) and surface oxidation/reduction redox couple (0.6-0.4 V), with a concomitant appearance of the Au(111) characteristic at 0.25 V. As the deposition time increased, the Au(111) deposit became smaller. It is notable that the shapes of the hydrogen regions are
Figure 2. Typical STM images of Pt deposits on Au(111) observed at 0 V in 0.05 M H2SO4 solution. The respective deposition times are (a) 10 s, (b) 30 s, (c) 60 s, (d) 150 s, (e) 600 s, and (f) 1500 s. Image size: 200 nm × 200 nm.
Electrocatalytic Oxidation on Au(III)/Pt Surfaces
Langmuir, Vol. 23, No. 21, 2007 10833
Figure 3. Population of Pt deposits of various widths. The respective deposition times are (9) 10 s, (2) 150 s, (b) 600 s, and (1) 1500 s.
not exactly same as the conventional one, probably due to the nanoscale size of the Pt deposits. 3.2. Electrochemical STM Images of Pt Deposits on Au(111). Figure 2 displays typical STM images of Pt deposits on Au(111) observed at 0 V. The images show clearly that Pt deposits locate selectively at the steps between wide terraces and the edges of Au islands, which is consistent with the previous reports.14,18,21 Such a site selective formation of nanosized deposits was also demonstrated on Au(111)/Ru.22 When the deposition time is less than 60 s, Pt deposits locate at the rims of the Au islands as shown in the upper panel of Figure 2. As the deposition time becomes longer than 150 s, Pt deposits at the centers of large Au islands (>15 nm) are frequently observed (compare the circles in Figure 2c and d). A further increase in the deposition time yields higher and larger Pt deposits on the step edges at the expense of small deposits as shown in Figure 2e and f. Apparent decreases in the areas covered by Pt deposits will be discussed in detail in conjunction with Pt coverage below. The formation of such large deposits by clustering small deposits is a result of the high surface energy of Pt.23 On the other hand, it is notable that Pt deposits are absent on some steps and edges as indicated with the arrows in Figure 2, which is not understood yet. Figure 3 shows the population of Pt deposits as a function of the width of the Pt deposit. The population of Pt deposits is defined as the ratio of the counted number of Pt deposits of a certain width to the total number of Pt deposits. When the deposition time is 10 s, most of the Pt deposits are 2-5 nm wide. As the deposition time increases, the distribution in population becomes quite broader and the width of significantly populated Pt deposits spans from 10 to 20 nm. Pt deposits on Au(111) are multilayered. Figure 4a is a constant current mode image of a Au(111) surface after contacting with the Pt-containing solution for 150 s. A line profile crossing the deposits of different contrasts is shown in Figure 4b. In Figure 4b, site a is a wide terrace of Au(111), and site b is the surface of a monatomic Au island (roughly 20 nm wide) on the terrace. Considering that Pt deposits of a particular width are rarely found in Figure 3 after 150 s of deposition, the assignment of site b to a monatomic Au island could be justified. On the specific Au island, there are two Pt deposits of different heights. The Pt deposit, designated as c, is ∼0.56 nm high, while the other Pt deposit, indicated as d, is ∼0.28 nm high. These specific height values indicate that Pt deposits could be not only one layer thick but also two layers thick. In Figure 4c, the pasted regions are two-layer-thick Pt deposits in an STM image after deposition for 150 s. The number of stacked layers in the Pt deposits generally (22) Strbac, S.; Johnston, C. M.; Lu, G. Q.; Crown, A.; Wieckowski, A. Surf. Sci. 2004, 573, 80. (23) Bauer, E.; van der Merwe, J. H. Phys. ReV. B 1986, 33, 3657.
Figure 4. Pt deposits of multilayers on Au(111): (a) a constant current mode image (25 nm × 25 nm) of a Au(111) surface after deposition for 150 s, (b) a profile along the line in (a), and (c) an image (40 nm × 40 nm) in which two-layer-thick deposits are pasted.
Figure 5. Population of different height Pt deposits as a function of deposition time. The numbers of Pt layers are (9) one, (2) two, (b) three, and (1) more than four.
increases along with deposition time and spans from one to five within the deposition time scale in this work (less than 1500 s). Figure 5 shows a population distribution of Pt deposits of different heights as a function of deposition time. When the deposition time is 10 s, Pt deposits are exclusively monatomic. As the deposition time increases up to 100 s, the population of two-layer-thick Pt deposits increases, but not significantly. A deposition time longer than 100 s yields an increase in the populations of two- and three-layer-thick deposits with an abrupt decrease in the population of monatomic deposits, indicating
10834 Langmuir, Vol. 23, No. 21, 2007
Kim et al.
Figure 6. Atomic arrangements inside Pt deposits on Au(111). The upper panel presents (a) a constant height mode image of a monatomic layer of Pt (15 nm × 15 nm), (b) a line profile crossing the deposit in (a), and (c) its atomic scale image of (x3 × x3)R30°-Pt (2.5 nm × 2.5 nm). The lower panel presents (d) a constant height mode image of a two-layer-thick deposit of Pt (30 nm × 30 nm), (e) a line profile crossing the deposit in (d), and (f) its atomic scale image of (1 × 1)-Pt (2.5 nm × 2.5 nm).
that multilayer Pt deposits start appearing on the tops of monatomic Pt deposits. After 600 s of deposition time, however, the population of two-layer-thick Pt deposits decreases, while that of Pt deposits with more than three layers continuously increases. The results presented so far would provide a brief description of the growth process of Pt deposits on Au(111). In the initial step of spontaneous deposition, monatomic deposits are exclusively produced. In the following stage, the Pt species in solution continuously deposit on top of the monatomic deposits to form the second and third layers. As the contact time with the Ptcontaining solution becomes longer, Pt deposits grow to much wider and higher deposits by depositing more Pt species from the solution and clustering small deposits on the surface. The overall growth process of Pt deposits observed in this work is at least qualitatively in agreement with that of the works of Kolb21 and Itaya,18 but contrasts that of deposits formed from PtBr42-.18 3.3. Atomic Arrangements of Pt Atoms in Pt Deposits. Figure 6 shows arrangements of Pt atoms in Pt deposits on Au(111). The upper panel of Figure 6 concerns a monatomic Pt deposit. The line profile crossing the deposit in Figure 6a and b indicates that the imaged deposit is monatomic. In an atomic scale image of the specific deposit (Figure 6c), the observed spots are arranged in a hexagonal symmetry, and the distance between spots is 0.51 ( 0.02 nm. Considering that the distance between the Au atoms is 0.288 nm and that the atomic rows in Figure 6c are rotated by 30° against the lattice vectors of the Au(111) substrate, the atomic arrangement in the monatomic Pt deposit is assigned to (x3 × x3)R30°-Pt. The bottom panel of Figure 6 deals with a two-layer-thick Pt deposit. Again, Figure 6d and e indicate that the specific Pt deposit of interest is a two-layer-thick deposit on a step edge. The spot periodicity in an image of the two-layer-thick deposit (Figure 6f) is 0.28 ( 0.03 nm. In addition, the lattice vectors of the atomic image in Figure 6f are parallel to those of the Au(111) substrate, so that the atomic arrangement is concluded to be (1 × 1)-Pt. Uosaki
Figure 7. Variations of Pt coverage on Au(111) estimated with two quantities: (b) hydrogen charge of Pt in voltammograms and (2) projected area of Pt deposits in STM images. See the text for details.
and co-workers have observed a bulk phase of Pt(111)-(1 × 1) on Au(111) after electrochemical deposition of Pt.20 Contrastingly, the (1 × 1)-Pt structure is observed on top of the first Pt layer of (x3 × x3)R30° on Au(111). Imaging the top layers of Pt deposits whose height was more than three layers was not successful, certainly due to the high local roughness around such Pt deposits. 3.4. Coverage of Pt Deposits on Au(111). Figure 7 shows the variations of Pt coverage as estimated with two independent quantities: hydrogen adsorption/desorption charge in voltammograms and projected area of Pt deposits in STM images. The coverage of Pt, defined as the ratio of the number of electrochemically active Pt atoms to that of surface Au atoms, was estimated by integrating the current from -0.27 to 0.03 V after subtraction of a current background similar to that of the voltammogram of Au(111) in the potential region. Here, the numbers of Pt and Au atoms were calculated from the integrated charge and the geometrical area of the used Au electrodes, respectively. On the other hand, Pt coverage, defined as the ratio of the sum of projected areas of Pt deposits in an STM image
Electrocatalytic Oxidation on Au(III)/Pt Surfaces
Figure 8. Cyclic voltammograms of formic acid oxidation on Pt deposits in 0.1 M HCOOH + 0.05 M H2SO4 solution: (a) Pt(111), and Pt deposits formed on Au(111) after deposition for (b) 10 s, (c) 60 s, (d) 150 s, (e) 600 s, and (f) 1500 s. Scan rate: 50 mV/sec.
to the scanned area, was estimated by measuring geometrical areas of Pt deposits without considering their heights. Therefore, the coverage values estimated with two different quantities reflect different but coverage-related properties: the coulometric coverage reflects the number of electrochemically active Pt atoms, while the geometrical coverage represents the area of Au(111) covered by Pt deposits. When deposition time is less than 100 s (region I), the coverage values estimated in the different ways are identical within experimental uncertainty. This consistency confirms that, in the specific deposition time range, the Pt deposits are monatomic. In a narrow time interval between 100 and 150 s (region II), the coulometric coverage increases rapidly, while the geometric coverage remains fairly constant. This particular behavior indicates that the direction of growth is not lateral but vertical to form compact layers of (1 × 1)-Pt on top of the monatomic Pt deposits of (x3 × x3)R30°. Thus, the number of electrochemically active Pt atoms increases. With a further increase in deposition time (region III), the Pt deposits grow vertically to be thicker. Despite continuous deposition, however, the number of electrochemically active Pt atoms decreases because of vertical growth. When the deposition time is more than a few hundreds of seconds (region IV), interestingly, the geometrical coverage and coulometric coverage decrease simultaneously. This particular decrease in both coulometric and geometric coverages results surely from severe clustering among the Pt deposits as seen in Figure 2f and from continuous vertical growth. 3.5. Oxidation of Formic Acid and Methanol On Pt Deposits. Figure 8 shows cyclic voltammograms of formic acid oxidation on Pt deposits formed on Au(111). Because the pure Au(111) surface is not active in formic acid oxidation, the observed current in Figure 8 is absolutely ascribable to formic acid oxidation on Pt deposits. Comparison of a formic acid oxidation voltammogram on pure Pt(111) (Figure 8a)24-31 with (24) Batista, E. A.; Iwasita, T. Langmuir 2006, 22, 7912. (25) Chen, Y.-X.; Heinen, M.; Jusys, Z.; Behm, R. J. Langmuir 2006, 22, 10399. (26) Clavilier, J.; Parsons, R. J. Electroanal. Chem. 1981, 124, 321.
Langmuir, Vol. 23, No. 21, 2007 10835
Figure 9. Cyclic voltammograms of methanol oxidation on Pt deposits in 0.1 M CH3OH + 0.05 M H2SO4 solution: (a) Pt(111), and Pt deposits formed on Au(111) after deposition for (b) 10 s, (c) 60 s, (d) 150 s, (e) 600 s, and (f) 1500 s. Voltammograms represented by the red dashed line were obtained in 0.05 M H2SO4 solution. Scan rate: 50 mV/sec.
that on Pt deposits (Figure 8b-f) reveals a few differences. The catalytic activity of Pt deposits toward formic acid oxidation is enhanced as indicated by the increased maximum oxidation current in the cathodic scan. For example, the maximum current observed on the Pt deposits formed after 150 s of deposition is larger by a factor of 10 than that on Pt(111), although the Pt coverage on Au(111) is much less than 1. In addition, the potential of the current maximum shifts in the cathodic direction from 0.34 to 0.2 V. Furthermore, the current curves of the anodic and cathodic scans are close enough, especially when the deposition time is 60-150 s (Figure 8c and d), indicating that the amounts of catalytic poison formed on the Pt deposits are not as much as those on Pt(111). Thus, the catalytic enhancement of Pt deposits in formic acid oxidation may result from the suppression of catalytic poison formation during the adsorption of formic acid on Pt surfaces. Figure 9 displays cyclic voltammograms of methanol oxidation on Pt deposits formed on Au(111). On Pt(111), the maximum oxidation current is observed at 0.4 V as shown in Figure 9a. On the Pt deposits, however, the methanol oxidation peak at 0.51 V is barely visible after 150 s of deposition (Figure 9d), even when the coulometric coverage of Pt is very high (Figure 7). Indeed, methanol oxidation is negligible when the deposition time is less than 60 s (Figure 9b and c). With prolonged deposition times longer than 600 s, the oxidation current becomes clearly distinguishable, but it is much less than that on Pt(111). The oxidation behavior of methanol on Pt deposits, contrasting that of formic acid, discloses that Pt deposits on Au(111) are not effective in methanol oxidation. (27) Kita, H.; Lei, H.-W. J. Electroanal. Chem. 1995, 388, 167. (28) Leiva, E.; Iwasita, T.; Herrero, E.; Feliu, J. M. Langmuir 1997, 13, 6287. (29) Macia, M. D.; Herrero, E.; Feliu, J. M. J. Electroanal. Chem. 2003, 554555, 25. (30) Samjeske, G.; Miki, A.; Ye, S.; Osawa, M. J. Phys. Chem. B 2006, 110, 16559. (31) Schmidt, T. J.; Behm, R. J. Langmuir 2000, 16, 8159.
10836 Langmuir, Vol. 23, No. 21, 2007
Figure 10. Variations of oxidation turnover frequencies of (b) formic acid and (2) methanol on Pt deposits as a function of deposition time.
Figure 10 presents plots of turnover frequencies of Pt deposits on Au(111) during the oxidation of formic acid and methanol along with deposition time. The turnover frequency is defined as the number of oxidized organic molecules per Pt atom within 1 s, and it is estimated with the maximum oxidation current in the cyclic voltammogram, the hydrogen charge of the electrode, and the number of electrons transferred in the oxidation of formic acid or methanol. Because the coulometric coverage of Pt, that is, the number of electrochemically active Pt atoms, varies with deposition time, the estimated turnover frequency is equivalent to the normalized catalytic activity to deliver semiquantitative information. The turnover frequency of formic acid varies remarkably depending on the deposition time. When the deposition time is less than 100 s (i.e., region I in Figure 7), the turnover frequency in formic acid oxidation increases slowly. In this particular deposition time range (10-100 s), most of the Pt deposits are monolayers of (x3 × x3)R30° atomic arrangement, and the population of multilayer Pt deposits increases slowly as shown in Figure 4. When the deposition time is 120-150 s, the turnover frequency of formic acid increases abruptly (in region II in Figure 7) and then declines gradually as the deposition time becomes longer (in region III in Figure 7). Specifically, the turnover frequency changes from 17 s-1 for Pt(111) to 327 s-1 for the Pt deposits after 150 s of deposition; the enhancement factor is approximately 20. Such an abrupt increase is relevant to the increase in the populations of two-layer-thick and/or three-layerthick Pt deposits (Figure 4), where the formation of poison is minimal (Figure 8). As the Pt deposits become wider and thicker (region IV in Figure 7), however, the turnover frequency decreases
Kim et al.
gradually, probably due to the initiation of poison formation as demonstrated in Figure 8e and f. Considering that the top layer of Pt deposits is (1 × 1), the turnover frequency of formic acid oxidation on Pt deposits may converge to that of Pt(111), that is, a reference of an infinitely wide (1 × 1)-Pt surface with infinite thickness. The variation of turnover frequency of methanol oxidation is quite contrasting to that of formic acid oxidation. When deposition time is less than 100 s, the turnover frequency is negligible. As the population of multilayer Pt deposits is more than 40% (i.e., deposition time longer than 150 s), the activity of methanol oxidation becomes notable and increases continuously. Analogous to the case of formic acid, it would be reasonable to extrapolate the turnover frequency of methanol oxidation on Pt deposits to that on Pt(111).
4. Summary We have presented catalytic activity variations of formic acid and methanol on Pt deposits formed by the use of spontaneous deposition on Au(111). The major findings in this work are summarized below: (i) The Pt deposits are stacked layers of (1 × 1)-Pt on top of the first monatomic Pt layer of (x3 × x3)R30°, and their width and thickness are dependent on deposition time. (ii) On the Pt deposits of (x3 × x3)R30°, formic acid oxidation takes place, while methanol oxidation does not occur. (iii) On the two- and three-layer-thick Pt deposits of (1 × 1), formic acid oxidation is remarkably enhanced, while methanol oxidation is barely observable. (iv) Small and thin Pt deposits are advantageous to formic acid oxidation but disadvantageous to methanol oxidation. Acknowledgment. The authors appreciate the permission of the Central Research Facility, Chungnam National University, Korea, to use the STM instrument. This work was supported by the Program of Development of Fundamental Technology for Direct Liquid Fuel Cell (DLFC). Supporting Information Available: Cylic voltammograms of Au(111) in H2SO4 solutions of different concentrations (Figure S1). This material is available free of charge via the Internet at http://pubs.acs. org. LA701377N
