C Ordered Intermetallic

Jan 20, 2015 - ... and Chemical Biology, Cornell University, Ithaca, New York 14853, ..... Kie Yong Cho , Yong Sik Yeom , Heun Young Seo , Pradip Kuma...
1 downloads 0 Views 420KB Size
Download PDF
Letter pubs.acs.org/NanoLett

Morphology and Activity Tuning of Cu3Pt/C Ordered Intermetallic Nanoparticles by Selective Electrochemical Dealloying Deli Wang,*,† Yingchao Yu,‡ Jing Zhu,† Sufen Liu,† David A. Muller,§ and Héctor D. Abruña*,‡ †

Key Laboratory for Large-Format Battery Materials and System, Ministry of Education, School of Chemistry and Chemical Engineering, Huazhong University of Science and Technology, 430074, Wuhan, People’s Republic of China ‡ Department of Chemistry and Chemical Biology, Cornell University, Ithaca, New York 14853, United States § Department of Applied and Engineering Physics and Kavli Institute for Nanoscale Science, Cornell University, Ithaca, New York 14853, United States S Supporting Information *

ABSTRACT: Improving the catalytic activity of Pt-based bimetallic nanoparticles is a key challenge in the application of proton-exchange membrane fuel cells. Electrochemical dealloying represents a powerful approach for tuning the surface structure and morphology of these catalyst nanoparticles. We present a comprehensive study of using electrochemical dealloying methods to control the morphology of ordered Cu3Pt/C intermetallic nanoparticles, which could dramatically affect their electrocatalytic activity for the oxygen reduction reaction (ORR). Depending on the electrochemical dealloying conditions, the nanoparticles with Pt-rich core−shell or porous structures were formed. We further demonstrate that the core−shell and porous morphologies can be combined to achieve the highest ORR activity. This strategy provides new guidelines for optimizing nanoparticles synthesis and improving electrocatalytic activity. KEYWORDS: Fuel cells, oxygen reduction reaction, electrocatalyst, ordered intermetallic, dealloying

C

In recent years, dealloying of Pt-based alloys has been widely used for improving electrocatalytic activity for the ORR in fuel cells.17−22 Core−shell or nanoporous structured materials have been formed depending on the dealloying conditions. An atmosphere-controlled chemical dealloying method was also used by the Strasser group to control the formation of nanoporous Pt−Ni bimetallic nanoparticles.23 Performing the dealloying under an air atmosphere produced a nanoporous structure, while a core−shell structure was formed under a N2 atmosphere.23 By electrochemically dealloying a Pt−Cu bimetallic alloy, Pt-rich shell, core−shell Pt−Cu nanoparticles were formed, which exhibited enhanced catalytic activity for the ORR.22,24 The nanoparticles could exhibit a single core−shell, multiple cores−shell, or surface pits and nanoscale pore structures depending on the particle size.25 A recent work, using an electrochemical method, has revealed that the dealloying of PtCo3 nanoparticles could result in a complex spongy multicore/shell structure.26 Our recent study has indicated that by forming an ordered intermetallic core, the ORR activity of Pt−Co/C nanoparticles

ontrolling nanoparticle morphology is a crucial issue in numerous applications, including optical,1,2 electrical and magnetic materials,3 chemical or biosensing,4,5 and especially in catalysis due to the improvement of catalytic activity.6 A number of strategies have been developed to synthesize these morphology-controlled nanoparticles. For example, hollowstructured metal oxides, chalcogenides, and phosphide nanocrystals have been successfully synthesized through the nanoscale Kirkendall effect.7−9 By using suitable organometallic precursor and surfactants, the shape and size of colloidal nanocrystals can be controlled, as reported by the Alivisatos group.10 Tetrahexahedral platinum nanocrystals have been prepared by the Sun group via an electrochemical treatment of Pt nanospheres employing a square-wave potential used for small organic molecule electro-oxidation.11 Xia et al. synthesized Pd−Pt bimetallic nanodendrites consisting of a dense array of Pt branches on a Pd core, that exhibited high activity for the ORR in fuel cells.12 Other strategies, such as dealloying methods, have also been used to form nanoporous structure in noble metal alloys.13−15 The nanopores were formed due to the depletion of the less noble metal component. This hollowing effect did not just occur at the surface of the particles but also through the surface and into the particles to support the continuing dissolution.13,16 © XXXX American Chemical Society

Received: November 30, 2014 Revised: January 13, 2015

A

DOI: 10.1021/nl504597j Nano Lett. XXXX, XXX, XXX−XXX

Letter

Nano Letters

Figure 1. (a) ADF-STEM images and EELS composite mapping of the as-prepared Cu3Pt/C intermetallic nanoparticles with Cu L2,3 and Pt N3 edges in red and green, respectively. (b) ADF-STEM images and EELS composite mapping of the particles after 5000 potential cycles from +0.05 to +1.0 V at a scan rate of 50 mV/s in N2-purged 0.1 M HClO4 solution at room temperature.

increased significantly when compared to a disordered alloy.27 Moreover, the morphology and the activity of ordered intermetallic Cu3Pt/C nanoparticles could be tuned to form hollowed structures through chemical dealloying or core−shell structures via electrochemical dealloying.21 Dealloying of lessnoble elements has become a key strategy for creating Pt-based catalysts in all design categories as articulated by Debe.28 Although great efforts have been focused on controlling dealloying conditions, a more systematic study using electrochemical methods is necessary in this field. We present a comprehensive study that demonstrates that by varying the electrochemical dealloying conditions, such as scan rate, upper potential limit, scan time, and the number of cycles, the morphology and ORR activity of Cu3Pt/C ordered intermetallic nanoparticles can be tuned dramatically. Ordered intermetallic Cu3Pt/C nanoparticles were chosen for several reasons. First, the Pt−Cu system has been employed in dealloying studies of fuel cell catalysts.20,22,24 However, to date there is no established unifying principle or paradigm showing the tuning of nanoscale morphology via electrochemical methods. Second, ordered intermetallic nanoparticles provide well-defined composition and structure, representing an ideal model material for electrochemical dealloying studies.21 A “one pot, two step” strategy was used for the preparation of Cu3Pt/C ordered intermetallic nanoparticles as reported in our previous work.21 In brief, an impregnation-reduction method was adapted to form a Pt−Cu alloy, followed by a post heat-treatment to form an ordered intermetallic phase. The as-prepared Cu3Pt/C ordered intermetallic nanoparticles exhibited a uniform elemental distribution of Pt and Cu from electron energy loss spectroscopic maps (EELS) (Figure 1a, refer to our previous publication21 for more details on analyzing EELS data). A 0.6−1.0 nm thickness Pt-rich shell formed on the surface of the nanoparticles after electrochemically dealloying at a scan rate of 50 mV/s for 5000 cycles between +0.05 and +1.0 V (versus RHE) in a N2-saturated 0.1 M HClO4 (Figure 1b). By carefully analyzing the reports of dealloyed nanoparticles, we have found that the morphology and activity vary broadly across research groups.25,26 Even when they all used electrochemical dealloying strategies, the detailed operating conditions might be different. Thus, we think it is important to systematically study dealloying conditions so as to provide useful guidelines for future studies. Cyclic voltammograms (CVs) of the as-prepared sample deposited on a glassy carbon (GC) electrode were performed continuously employing different upper limit potentials at a scan rate of 50 mV/s for three cycles (Supporting Information

Figure S1). The current in the hydrogen region gradually increased with increasing upper potential limit. A bulk Cu dissolution current was observed in the anodic scan beyond +0.35 V versus RHE, while Cu redeposition in the cathodic scan took place when the upper potential limit was under +0.07 V versus RHE. There was almost no bulk Cu dissolution and redeposition when the upper potential limit was over +1.0 V versus RHE. When the upper potential limit reached +1.2 V versus RHE, the cyclic voltammograms presented features typical of a Pt surface with clear hydrogen adsorption/ desorption and Pt oxide formation and reduction peaks. Two types of electrochemical dealloying methods were employed on the Cu3Pt/C ordered intermetallic nanoparticles: potential cycling and potentiostatic method. For the potential cycling method, parameters such as upper limit potential, scan rate, and the number of cycles were varied. To ensure clear correlations, we chose to change only one parameter at a time. For example, if the upper potential limit was fixed at +1.0 V versus RHE and the cycle number was fixed at 5000 we only changed the scan rate from 50 mV/s to 1 V/s and observed some unexpected results. If the scan rate was slow (50 mV/s), a core−shell structure was formed (Figure 1b). When the scan rate was increased to 1 V/s but the number of cycles was kept at 5000, the morphology of the nanoparticles remained virtually unchanged. A pair of overview scanning transmission electron microscopy (STEM) images in Figure 2a,b show that these noncore−shell nanoparticles were well dispersed on the carbon support with virtually no agglomeration. The energy dispersive X-ray (EDX) line profile (Figure 2d) across the particle (Figure 2c) shows that the distribution of Pt and Cu was uniform with no obvious core−shell structure within the resolution of EDX mapping. Because the total cycling time at a scan rate of 1 V/s for 5000 cycles was only about 2.6 h, it seems that the time was not long enough to leach out all the Cu from the particle. The CV curves (Supporting Information Figure S2) clearly show that even after 5000 potential cycles, the Pt features were still not evident on the surface, which is consistent with the previous observations. When the nanoparticles were cycled at 1 V/s for 100 000 cycles, the total cycling time was about 53 h, which would be the same as for the data obtained at a scan rate of 50 mV/s for 5000 cycles. The original spherical nanoparticles, especially the larger ones, turned into cubic nanoparticles. As shown in Figure 3, the morphology of the particles was significantly changed (compare Figure 3 with Figure 2b), with an evident Pt-rich shell on the surface (Figure 3d). The new morphology was formed because of the rearrangement of the surface during fast-scan potential cycling. B

DOI: 10.1021/nl504597j Nano Lett. XXXX, XXX, XXX−XXX

Letter

Nano Letters

CVs of commercial Pt/C with Cu3Pt/C nanoparticles after 100 000 cycles (Supporting Information Figure S3), we can see clearly for Pt/C, there are two peaks at 0.16 and 0.22 V, corresponding to hydrogen desorption on Pt(110) and (100) plane, respectively. While only one peak at 0.28 V originates from the hydrogen desorption on Pt(100) plane for Cu3Pt/C nanoparticles after 100 000 cycles, indicating that the (100) facets dominate in the Cu3Pt/C nanoparticles after fast CV treatment. The morphology change of Cu3Pt/C nanoparticles was mainly caused by the reconfiguration of Pt atoms during cycling. When the scan rate is low, most of the dissolved Pt ions diffused in the bulk solution, while during fast-scan potential cycling, the dissolved Pt ions on the surface have no time to diffuse and then redeposited on the electrode surface, forming Pt(100) facets dominated cubic structure. In the CO stripping curves (Figure 4b), two prepeaks and a main peak appeared after 100 000 cycles, which is in good agreement with results reported for cubic Pt nanoparticles.31 These CVs and CO stripping experiments provide evidence, albeit indirect, that the surface changed with faster scan rates. The ORR polarization curves in Figure 4c exhibited a positive shift in the half-wave potential (E1/2) of 20 mV after 5000 potential cycles as compared with Pt/C, while the E1/2 continued to shift positive by about 40 mV after 100 000 cycles, indicating an enhancement in activity for the ORR. The activity comparison, including mass activity (MA) and specific activity (SA) is presented in Figure 4d. At +0.9 V, the nanoparticles, after 5000 cycles and 100 000 potential cycles, exhibited a MA of 0.26 and 0.46 A/mgPt respectively, which is 2.5 to 5 times that of Pt/C, while the SA was enhanced by a factor of 6 to 7. While maintaining the same sweep rate of 50 mV/s and number of cycles of 5000, we increased the upper limiting potential to +1.2 V versus RHE (Supporting Information Figure S4). The first three cycles show rapid Cu dissolution with no obvious Pt voltammetric characteristics. After 5000 cycles, two well-defined hydrogen adsorption/ desorption peaks at +0.13 and +0.28 V versus RHE appeared, along with Pt oxide formation and reduction peaks at +0.9 and +0.8 V versus RHE respectively. The morphology of the particles turned out to be hollow as shown in Figure 5. Interestingly, the shape of the nanoparticles also changed. The shape change might be caused by the redeposition of the electrochemically dissolved Pt ions during cycling. These hollow structures greatly improved the surface-to-volume ratio and thus could potentially promote the ORR mass activity. As shown in Supporting Information Figure S5, the ORR polarization curve on the electrode after 1000 potential cycles exhibited a positive shift in E1/2 of about 50 mV, although with an activity decay after 5000 potential cycles (∼10 mV negative shift in E1/2). We further changed the dealloying conditions by using a potentiostatic method (Supporting Information Figure S6). Figures 6 and 7 show the STEM images of the nanoparticles after applying a constant potential of +0.8 and +1.0 V versus RHE for 3 h, respectively. When the potential was controlled at +0.8 V, only a small fraction of the particles, especially the relatively larger particles, formed porous structures. A similar phenomenon has also been reported by Oezaslan et al.25 The porous multiple-core/shell structures could be formed during electrochemical dealloying of PtCo3/C and PtCu3/C particles when particle sizes were larger than 30 nm.25 When the potential was held at +1.0 V versus RHE, the majority of particles presented dark spots in the ADF-STEM image (Figure

Figure 2. (a,b) BF and ADF-STEM overview images after 5000 potential cycles from +0.05 to +1.0 V at a scan rate of 1 V/s in N2purged 0.1 M HClO4 solution at room temperature. (c,d) ADF-STEM image of one representative particle and the EDX line profile across the particle.

Figure 3. ADF-STEM overview images of Cu3Pt/C intermetallic nanoparticles after 100 000 potential cycles from +0.05 to +1.0 V at a scan rate of 1 V/s in N2-purged 0.1 M HClO4 solution at room temperature (a) and EELS maps of Cu (b), Pt (c) and the composite of Pt and Cu (d).

Similar phenomena have also been observed on polycrystalline Pt wires to form Pt(100) single crystal surfaces.29 The electrochemical properties of the nanoparticles after 5000 and 100 000 cycles at a scan rate of 1 V/s were evaluated at a rotating disk electrode (RDE). Figure 4a,b shows the CVs and CO stripping of the two samples at a scan rate of 50 mV/s. It can be observed that in the CV curves (Figure 4a), the hydrogen under-potential-deposition (UPD) region shifted to positive potentials by about 30 mV. The peak between +0.2 to +0.4 V versus RHE after 100 000 cycles resembled (100) domains as described in the literature.30,31 When comparing the C

DOI: 10.1021/nl504597j Nano Lett. XXXX, XXX, XXX−XXX

Letter

Nano Letters

Figure 4. (a) CV curves in N2-purged 0.1 M HClO4 solution at a scan rate of 50 mV/s. (b) CO stripping in N2-purged 0.1 M HClO4 solution at a scan rate of 50 mV/s. (c) ORR polarization curves in O2-saturated 0.1 M HClO4 solution at rotation rate of 1600 rpm, and a scan rate of 5 mV/s. (d) Mass and specific activity for ORR comparison of Cu3Pt/C intermetallic nanoparticles after 5000 (5K) and 100 000 (100 K) potential cycles from +0.05 to +1.0 V in N2-purged 0.1 M HClO4 solution at a scan rate of 1 V s−1 at room temperature.

Figure 5. ADF-STEM overview images of Cu3Pt/C intermetallic nanoparticles after 5000 potential cycles from +0.05 to +1.2 V at a scan rate of 50 mV/s in N2-purged 0.1 M HClO4 solution at room temperature (a), three typical particles (b−d), and EELS maps Pt (e) Cu (f), and the composite of Pt and Cu (g) for the particle c.

Figure 6. BF (a,c) and ADF-STEM (b,d) overview images of Cu3Pt/C intermetallic nanoparticles after holding the potential at +0.8 V for 3 h in N2-purged 0.1 M HClO4 solution at room temperature. (e−g) Three particles with porous structures.

7b), corresponding to internal voids or empty space within the particle. From the line profiles in Figure 7c,d, these dark spot regions lacked both Pt and Cu. The ORR electrocatalytic activity of the particles after holding the potential at +0.8 V versus RHE was lower than that of pure Pt (Figure 8a), likely because the surface still had a small fraction of Cu remaining, and the particles were not fully hollow. However, after applying a constant potential of +1.0 V for 3 h the ORR activity was significantly improved. As shown in Figure 8a, the half-wave potential for such a material was about 70 mV positive of Pt/C. Subsequent CV scans enhanced the mass activity further. The improved ORR activity may be attributed at least in part to a larger electrochemical surface area (ESA), as is evident in Figure 8b with the areas for both hydrogen adsorption− desorption and oxide formation−reduction regions increasing significantly after the application of higher potentials. After holding the potential at +1.0 V, the majority of the particles

formed nanoporous structures, providing a larger ESA. Additionally, there was only one broad peak without wellresolved weakly- and strongly- adsorbed peaks in the hydrogen region, which is similar to our previous report on hollowed Pt− Cu nanoparticles formed by a chemical leaching method.21 The current for the hydrogen adsorption−desorption peaks decreased after 100 potential cycles, while in the oxide formation-reduction region both the onset potential of surface oxides formation and reduction shifted positively, suggesting a weakening adsorption strength of the oxygenated species on the particle surface.32,33 The decrease in the oxygen-containing species adsorption strength can also be verified by the different onset potentials for CO stripping on different electrodes (Supporting Information Figure S7). The enhanced ORR D

DOI: 10.1021/nl504597j Nano Lett. XXXX, XXX, XXX−XXX

Letter

Nano Letters

nanoporous structure gradually coarsened due to coalescence,34 forming larger pores; on the other hand, the Pt dissolution and redeposition also occurred on the particles surface, gradually forming a Pt-rich shell. After 1000 potential cycles, the Pt shell was too thick (>1 nm) resulting in a relaxation of the lattice strain and a corresponding decay in activity for the ORR (Supporting Information Figure S8). In conclusion, we have presented a comprehensive study of the effects of nanoscale electrochemical dealloying. We have successfully synthesized a group of Pt−Cu electrocatalysts, which exhibited various morphologies depending on the dealloying parameters and various activities toward the ORR. Our findings suggest that a critical upper limit voltage exists for creating spongy electrocatalysts, although other factors such as scan rate and total cycle numbers also appear to play important roles. By carefully tuning the electrochemical cycling parameters, the shape, size, preferred facet and surface structure of these electrocatalyts can be tuned at the nanoscale. We optimized these conditions to enable a Pt rich shell with the largest specific surface area in addition to the Pt−Cu lattice strain effect in the core. Beyond this study, we have enabled such a database for similar platinum binary materials under electrochemical dealloying conditions. This can serve as a guide in the search for better performing electrocatalysts capable of meeting the goals of high activity, longer stability, and low cost.

Figure 7. BF (a) and ADF-STEM (b) overview images of Cu3Pt/C intermetallic nanoparticles after holding the potential at +1.0 V for 3 h. (c,d) ADF-STEM image of one particle and the EDX line profile across the particle. The line in c indicates the profile where the EDX in d is taken.

■

ASSOCIATED CONTENT

S Supporting Information *

Experimental section and additional electrochemical testing results. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Authors

*E-mail: (D.W.) [email protected]. *E-mail: (H.D.A.) [email protected]. Notes

The authors declare no competing financial interest.

■ Figure 8. (a) ORR polarization curves in O2-saturated 0.1 M HClO4 solution at rotation rate of 1600 rpm, and a scan rate of 5 mV/s and (b) CV curves comparison in N2-purged 0.1 M HClO4 solution of Cu3Pt/C intermetallic nanoparticles after holding the potential at +0.8 and +1.0 V for 3 h, respectively, in N2-purged 0.1 M HClO4 solution at room temperature and after holding the potential at +1.0 V and then performing CV cycling from +0.05 to +1.0 V at a scan rate of 50 mV/s for 100 cycles. (c) EELS maps of two typical particles after holding the potential at +1.0 V, and then performing CV cycling from +0.05 to +1.0 V at a scan rate of 50 mV/s for 1000 cycles.

ACKNOWLEDGMENTS This work was supported by the National Science Foundation of China (21306060), the Program for New Century Excellent Talents in Universities of China (NCET-13-0237), the Doctoral Fund of Ministry of Education of China (20130142120039), the Fundamental Research Funds for the Central University (2013TS136, 2014YQ009) and the Department of Energy through Grant DE-FG02-87ER45298, by the Energy Materials Center at Cornell, an Energy Frontier Research Center funded by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences under Award Number DE-SC0001086. This work made use of TEM facilities of the Cornell Center for Materials Research (CCMR). We thank the Analytical and Testing Center of Huazhong University of Science and Technology for allowing us to use its facilities.

activity after 100 potential cycles could also be attributed to a surface lattice strain.22,23 Figure 8c shows the EELS mappings of two nanoparticles after 1000 potential cycles. Although the particles maintained a nanoporous structure after cycling, the morphology was different from the as-prepared particles (Figure 7). On one hand, during the potential cycling the

(1) Hsueh, H. Y.; Chen, H. Y.; Hung, Y. C.; Ling, Y. C.; Gwo, S.; Ho, R. M. Adv. Mater. 2013, 25, 1780−1786. (2) Ma, Y. Y.; Li, W. Y.; Cho, E. C.; Li, Z. Y.; Yu, T. K.; Zeng, J.; Xie, Z. X.; Xia, Y. N. ACS Nano 2010, 4, 6725−6734. (3) Sun, S. H.; Murray, C. B.; Weller, D.; Folks, L.; Moser, A. Science 2000, 287, 1989−1992.

■ E

REFERENCES

DOI: 10.1021/nl504597j Nano Lett. XXXX, XXX, XXX−XXX

Letter

Nano Letters (4) Xiong, B.; Zhou, R.; Hao, J. R.; Jia, Y. H.; He, Y.; Yeung, E. S. Nat. Commun. 2013, 4, 1708−1716. (5) Anker, J. N.; Hall, W. P.; Lyandres, O.; Shah, N. C.; Zhao, J.; Van Duyne, R. P. Nat. Mater. 2008, 7, 442−453. (6) Lee, I.; Delbecq, F.; Morales, R.; Albiter, M. A.; Zaera, F. Nat. Mater. 2009, 8, 132−138. (7) Moreau, L. M.; Ha, D. H.; Zhang, H. T.; Hoyden, R.; Muller, D. A.; Robinson, R. D. Chem. Mater. 2013, 25, 2394−2403. (8) Wang, W. S.; Dahl, M.; Yin, Y. D. Chem. Mater. 2013, 25, 1179− 1189. (9) Yin, Y. D.; Rioux, R. M.; Erdonmez, C. K.; Hughes, S.; Somorjai, G. A.; Alivisatos, A. P. Science 2004, 304, 711−714. (10) Puntes, V. F.; Krishnan, K. M.; Alivisatos, A. P. Science 2001, 291, 2115−2117. (11) Tian, N.; Zhou, Z. Y.; Sun, S. G.; Ding, Y.; Wang, Z. L. Science 2007, 316, 732−735. (12) Lim, B.; Jiang, M. J.; Camargo, P. H. C.; Cho, E. C.; Tao, J.; Lu, X. M.; Zhu, Y. M.; Xia, Y. N. Science 2009, 324, 1302−1305. (13) Forty, A. J. Nature 1979, 282, 597−598. (14) Erlebacher, J.; Aziz, M. J.; Karma, A.; Dimitrov, N.; Sieradzki, K. Nature 2001, 410, 450−453. (15) Ding, Y.; Erlebacher, J. J. Am. Chem. Soc. 2003, 125, 7772−7773. (16) Stratmann, M.; Rohwerder, M. Nature 2001, 410, 420−423. (17) Shui, J. I.; Chen, C.; Li, J. C. M. Adv. Funct. Mater. 2011, 21, 3357−3362. (18) Mani, P.; Srivastava, R.; Strasser, P. J. Power Sources 2011, 196, 666−673. (19) Srivastava, R.; Mani, P.; Hahn, N.; Strasser, P. Angew. Chem., Int. Ed. 2007, 46, 8988−8991. (20) Yang, R.; Leisch, J.; Strasser, P.; Toney, M. F. Chem. Mater. 2010, 22, 4712−4720. (21) Wang, D. L.; Yu, Y. C.; Xin, H. L. L.; Hovden, R.; Ercius, P.; Mundy, J. A.; Chen, H.; Richard, J. H.; Muller, D. A.; DiSalvo, F. J.; Abruna, H. D. Nano Lett. 2012, 12, 5230−5238. (22) Strasser, P.; Koh, S.; Anniyev, T.; Greeley, J.; More, K.; Yu, C.; Liu, Z.; Kaya, S.; Nordlund, D.; Ogasawara, H.; Toney, M. F.; Nilsson, A. Nat. Chem. 2010, 2, 454−460. (23) Gan, L.; Heggen, M.; O’Malley, R.; Theobald, B.; Strasser, P. Nano Lett. 2013, 13, 1131−1138. (24) Koh, S.; Strasser, P. J. Am. Chem. Soc. 2007, 129, 12624−12625. (25) Oezaslan, M.; Heggen, M.; Strasser, P. J. Am. Chem. Soc. 2012, 134, 514−524. (26) Liu, Z. Y.; Xin, H. L.; Yu, Z. Q.; Zhu, Y.; Zhang, J. L.; Mundy, J. A.; Muller, D. A.; Wagner, F. T. J. Electrochem. Soc. 2012, 159, F554− F559. (27) Wang, D. L.; Xin, H. L. L.; Hovden, R.; Wang, H. S.; Yu, Y. C.; Muller, D. A.; DiSalvo, F. J.; Abruna, H. D. Nat. Mater. 2013, 12, 81− 87. (28) Debe, M. K. Nature 2012, 486, 43−51. (29) Cervino, R. M.; Arvia, A. J.; Vielstich, W. Surf. Sci. 1985, 154, 623−628. (30) Vidal-Iglesias, F. J.; Aran-Ais, R. M.; Solla-Gullon, J.; Herrero, E.; Feliu, J. M. ACS Catal. 2012, 2, 901−910. (31) Nakamura, M.; Hanioka, Y.; Ouchida, W.; Yamada, M.; Hoshi, N. ChemPhysChem 2009, 10, 2719−2724. (32) Stamenkovic, V.; Schmidt, T. J.; Ross, P. N.; Markovic, N. M. J. Phys. Chem. B 2002, 106, 11970−11979. (33) Wang, X.; Tian, W.; Zhai, T. Y.; Zhi, C. Y.; Bando, Y.; Golberg, D. J. Mater. Chem. 2012, 22, 23310−23326. (34) Yu, Y. C.; Xin, H. L. L.; Hovden, R.; Wang, D. L.; Rus, E. D.; Mundy, J. A.; Muller, D. A.; Abruna, H. D. Nano Lett. 2012, 12, 4417− 4423.

F

DOI: 10.1021/nl504597j Nano Lett. XXXX, XXX, XXX−XXX