Synthesis and Characterization of Pt–Ag Alloy Nanocages with

Publication Date (Web): September 23, 2016 .... 3D freestanding porous PtAg hollow chain-like networks as efficient electrocatalyst for oxygen reducti...
0 downloads 0 Views 2MB Size
Subscriber access provided by UNIV OF CALIFORNIA SAN DIEGO LIBRARIES

Communication

Synthesis and Characterization of Pt-Ag Alloy Nanocages with Enhanced Activity and Durability towards Oxygen Reduction Xuan Yang, Luke T Roling, Madeline Vara, Ahmed O. Elnabawy, Ming Zhao, Zachary D. Hood, Shixiong Bao, Manos Mavrikakis, and Younan Xia Nano Lett., Just Accepted Manuscript • DOI: 10.1021/acs.nanolett.6b03395 • Publication Date (Web): 23 Sep 2016 Downloaded from http://pubs.acs.org on September 25, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Nano Letters is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 18

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Synthesis and Characterization of Pt−Ag Alloy Nanocages with Enhanced Activity and Durability towards Oxygen Reduction

Xuan Yang,† Luke T. Roling,¶ Madeline Vara,§ Ahmed O. Elnabawy,¶ Ming Zhao,§ Zachary D. Hood,§,‡ Shixiong Bao,† Manos Mavrikakis,*,¶ and Younan Xia*,†,§



The Wallace H. Coulter Department of Biomedical Engineering, Georgia Institute of Technology and Emory University, Atlanta, Georgia 30332, United States



Department of Chemical and Biological Engineering, University of Wisconsin−Madison, Madison, Wisconsin 53706, United States §

School of Chemistry and Biochemistry, School of Chemical and Biomolecular Engineering, Georgia Institute of Technology, Atlanta, Georgia 30332, United States



Center for Nanophase Materials Sciences, Oak Ridge National Laboratory, Oak Ridge, Tennessee 37831, United States

*Corresponding authors: [email protected] (for computation) and [email protected] (for synthesis and electrochemical characterization)

1

ACS Paragon Plus Environment

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Abstract: Engineering the elemental composition of metal nanocrystals offers an effective strategy for the development of catalysts or electrocatalysts with greatly enhanced activity. Herein we report the synthesis of Pt−Ag alloy nanocages with an outer edge length of 18 nm and a wall thickness of about 3 nm. Such nanocages with a composition of Pt19Ag81 could be readily prepared in one step through the galvanic replacement reaction between Ag nanocubes and a Pt(II) precursor. After 10,000 cycles of potential cycling in the range of 0.60−1.0 V as in an accelerated durability test, the composition of the nanocages changed to Pt56Ag44, together with a specific activity of 1.23 mA cm−2 towards oxygen reduction, which was 3.3 times greater than that of a state-of-the-art commercial Pt/C catalyst (0.37 mA cm−2) prior to durability testing. Density functional theory calculations attributed the increased activity to the stabilization of the transition state for breaking the O−O bond in molecular oxygen. Even after 30,000 cycles of potential cycling, the mass activity of the nanocages only dropped from 0.64 to 0.33 A mg−1Pt, which was still about two times as great as that of the pristine Pt/C catalyst (0.19 A mg−1Pt).

KEYWORDS: Platinum-based catalyst, nanocage, Pt−Ag alloy, oxygen reduction reaction, density functional theory

2

ACS Paragon Plus Environment

Page 2 of 18

Page 3 of 18

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Proton exchange membrane fuel cells (PEMFCs) are efficient in converting chemical energy to electricity at high power density and zero emission.1 As such, the PEMFC technology is attractive for future transportation. Although this technology has been around for many decades, its commercialization is still hindered by several challenges, including the low activity and poor durability associated with the Pt/C catalysts currently used to accelerate the oxygen reduction reaction (ORR) on the cathode of a PEMFC.2 To reduce the cost of a PEMFC without compromising the performance, the catalyst need to be greatly enhanced in both mass activity and durability by engineering the shape, size, structure, and composition of the Pt-based nanoparticles.3−7 As a unique class of catalytic materials, Pt-based nanocages have recently been demonstrated with significantly enhanced catalytic activity and durability towards ORR.8−11 Because of their hollow interiors and porous walls, the Pt atoms on the inner surface can also participate in the catalytic reaction, leading to increase in specific surface area and thus the mass activity. Currently, there are two different strategies for the synthesis of Pt nanocages, including selective removal of Pd cores from Pd@PtnL (n = 1−6) core−shell nanocrystals and galvanic replacement between a Pt precursor and sacrificial templates.8−14 In general, two steps are necessary for the synthesis of Pt nanocages through the selective removal of Pd cores, including the deposition of Pt shells on the surfaces of Pd nanocrystals and a subsequent etching process to selectively remove the Pd cores.8−10 Using this approach, Pt-based nanocages with a number of different surface structures have been synthesized. The wall thickness of such Pt nanocages can be precisely controlled up to six atomic layers. Because of their ultrathin walls and thus high utilization efficiency of atoms, these nanocages have shown great catalytic performance towards ORR. The synthesis is, however, time-consuming due to the involvement of multiple steps. In contrast, only one step is needed for the synthesis of Pt nanocages through the galvanic replacement reaction between a Pt precursor and a sacrificial template made of Ag or Pd nanocrystals.11−14 Despite the simplicity of this synthetic process, the Pt nanocages reported in literature have been plagued by relatively large dimensions and thick walls, which are not beneficial to the enhancement of ORR mass activity. In addition, the surfaces of such Pt nanocages are typically very rough, making it difficult to control the surface structure. 3

ACS Paragon Plus Environment

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Here we demonstrate the synthesis of 18-nm Pt−Ag alloy nanocages with smooth surfaces through galvanic replacement. Starting from Ag nanocubes of 16 nm in edge length,15 we have obtained Pt−Ag alloy nanocages mainly enclosed by {100} facets and systematically evaluated the conditions for galvanic replacement to ensure high quality and good reproducibility. The ORR activity of the Pt−Ag nanocages was found to be highly dependent on their composition, which changed from Ag-rich to Pt-rich upon repeated electrochemical cycling. After 10,000 rounds of potential cycling in the range of 0.60−1.0 V, the nanocages showed the highest specific activity of 1.23 mA cm−2. Even after 30,000 rounds of potential cycling, the nanocages still showed a mass activity of 0.33 A mg−1Pt, which was almost two times as great as that of the pristine Pt/C (0.19 A mg−1Pt). Self-consistent density functional theory (DFT) calculations attribute these activity trends to differences in the transition state stabilization for O−O bond breaking events. The first step in the synthesis of Pt−Ag nanocages involves the preparation of small Ag nanocubes by following a protocol recently reported by our group.15 The Ag nanocubes used in this work had an average edge length of 15.8±0.4 nm (Figure S1A). They had a localized surface plasmon resonance (LSPR) peak at 411 nm, together with a shoulder at 357 nm (Figure S1), confirming the small size of the nanocubes. The half bandwidth of the LSPR peak was only 37 nm, indicating a narrow size distribution for the Ag nanocubes. As illustrated in Figure 1, the Ag nanocubes can serve as sacrificial templates for the synthesis of Pt−Ag nanocages through the galvanic replacement reaction with a Pt(II) precursor. As the galvanic replacement reaction proceeds, Pt atoms are deposited on the surface of the Ag nanocubes while Ag atoms are dissolved into the solution. According to previous reports, the as-prepared nanocages could be dealloyed in the presence of an etchant for elemental Ag such as H2O2.16,17 Figure 2, A and B, shows TEM images of the Pt−Ag nanocages before and after etching with 2% aqueous H2O2 for 24 h. The cubic hollow structure was well preserved, demonstrating the stability of the Pt−Ag nanocages in the presence of H2O2. The high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM) image and energy-dispersive X-ray spectroscopy (EDX) elemental mapping demonstrate that the nanocage was indeed composed of a Pt−Ag alloy. 4

ACS Paragon Plus Environment

Page 4 of 18

Page 5 of 18

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Similar to the synthesis of Au nanocages,16,17 the amount of Pt(II) precursor, reaction temperature, and injection method for the Pt(II) precursor are all critical to the formation of nanocages with a smooth surface. Figure S2 shows TEM images of Pt−Ag nanocages obtained after introducing different amounts of the Pt precursor. As the amount of Pt precursor injected into the suspension of Ag nanocubes was increased, the nanocages became more porous. This result demonstrates that more Ag atoms were leached out due to the galvanic replacement reaction between the Ag and Pt(II) precursor. With the introduction of 0.4 mL of Pt(II) precursor (Figure 2), the wall thickness of the nanocages was about 3.1 nm. According to the ICP-MS data, the amount of Pt in the nanocages (with a composition of Pt19Ag81) was only equivalent to two atomic layers of Pt if they would have been deposited in a layer-by-layer fashion on the Ag nanocubes. Note that the Pt−Ag nanocages obtained with the introduction of 0.4 mL of the Pt(II) precursor were used for all electrochemical characterizations. The reaction temperature could significantly affect the structure of the Pt−Ag nanocages (Figure S3). When the reaction temperature was increased to 60 °C, the surface of the nanocages became very rough. The surface of the nanocages became more corrugated with the increase of reaction temperature because the deposition rate was increased more significantly than the diffusion rate of the Pt adatoms. The method used for the injection of the Pt precursor also affected the product quality. When the precursor was injected in a dropwise fashion, only nanocages were obtained (Figure S4). However, when the precursor was injected in one shot, there were also some tiny nanoparticles (with a size of around 2 nm) in the final products due to the involvement of homogeneous nucleation. As the injection rate was increased, more Pt atoms were formed in the solution at the early stage of synthesis, causing the formation of small Pt nanoparticles through homogeneous nucleation. Taken together, dropwise addition of the Pt(II) precursor into a suspension of Ag nanocubes is instrumental to the synthesis of uniform, pure Pt−Ag nanocages. After washing three times with DI water, most PVP on the surface of Pt−Ag nanocages could be removed (Figures S5 and S6). When subjected to repeated potential cycling in the range of 0.60−1.0 V (as in an accelerated durability test), the Ag will be selectively leached out from the walls to generate 5

ACS Paragon Plus Environment

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Pt−Ag alloy nanocages with different compositions. Specifically, the nanocages changed from Pt19Ag81 to Pt46Ag54, Pt56Ag44, Pt65Ag35, and Pt79Ag21 at 5000, 10000, 20000, and 30000 rounds of repeated potential cycling, respectively. By using the rotating disk electrode (RDE) technique, we compared the ORR performance of these Pt−Ag cubic nanocages with that of a commercial Pt/C catalyst. Figure 3A shows the cyclic voltammograms (CVs) of the Pt56Ag44 nanocages supported on carbon and the commercial Pt/C catalyst. We derived the electrochemical active surface areas (ECSAs) of the two catalysts from the charges associated with both the desorption and adsorption of hydrogen and then normalized the values against the amounts of Pt in the samples. Although there was a substantial difference between their particle sizes (18.5±1.2 nm for nanocages vs. 3.2 nm for Pt/C), the specific ECSA of the nanocages was comparable to that of the Pt/C (50.2 m2 g−1Pt vs. 52.3 m2 g−1Pt). Since the Pt−Ag nanocages contained a very small amount of Pt atoms on their surface, the exposed fraction of Pt atoms should be higher than that of the commercial Pt/C catalyst. Additionally, since the nanocages were porous, atoms on both the outer and internal surfaces could participate in ORR. The positive-going ORR polarization curves of the catalysts are shown in Figure 3B. To achieve a better understanding of mass and surface effects, we calculated the kinetic currents of a polarization curve based on the Koutecky−Levich equation and then normalized them against the Pt mass and ECSA to obtain the mass and specific activities (jk,mass and jk,specific), respectively (Figure 3, C and D). Both the mass and specific activities of the Pt56Ag44 nanocages were greatly enhanced relative to the commercial Pt/C catalyst in the potential region of 0.86−0.94 V referenced to the reversible hydrogen electrode (VRHE). At 0.9 VRHE, the mass activity of the Pt56Ag44 nanocages was 0.64 A mg−1Pt, almost 3.4 times as high as the initial mass activity of the Pt/C catalyst (0.19 A mg−1Pt). The specific activity (1.23 mA cm−2) of the nanocages at 0.9 VRHE was enhanced by 3.3 times relative to the initial specific activity of the Pt/C catalyst (0.37 mA cm−2). It is worth noting that both the specific and mass activities of the Pt−Ag nanocages were highly dependent on their composition (Figure 4). The as-prepared Pt19Ag81 nanocages showed a specific activity of 0.49 mA cm−2 and specific ECSA of 52.04 m2 g−1Pt. After 5,000 cycles of accelerated durability test, the composition of the Pt−Ag nanocages changed to Pt46Ag54 due to the selective dissolution of Ag atoms during the potential cycling, as 6

ACS Paragon Plus Environment

Page 6 of 18

Page 7 of 18

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

illustrated in Figure 1. Compared to the as-prepared Pt19Ag81 nanocages, the specific activity of the Pt46Ag54 nanocages increased by 97%, leading to great enhancement in mass activity. After 10,000 cycles of potential cycling, the molar ratio of Pt to Ag in Pt−Ag nanocages increased to 56:44. At the same time, the specific activity of the Pt56Ag44 nanocages increased to 1.23 mA cm−2 (Figure 3). After 20,000 and 30,000 cycles of potential cycling, the compositions of the Pt−Ag nanocages changed to Pt65Ag35 and Pt79Ag21, respectively. Consequently, the specific activities of the Pt−Ag nanocages decreased to 0.81 and 0.64 mA cm−2, respectively. The specific ECSA varied very little even after 30,000 cycles of potential cycling, indicating the good stability of such Pt−Ag nanocages during ORR. The stability of these Pt−Ag nanocages was also confirmed by TEM imaging, as shown in Figure 5. The cubic shape and hollow structure of these nanocages were largely preserved during the accelerated durability tests. Furthermore, the Pt−Ag nanocages still showed a mass activity of 0.33 A mg−1Pt after 30,000 cycles, which was about two times as great as that of the pristine Pt/C catalyst. To better understand the composition-dependent activities of the Pt−Ag nanocages, we performed self-consistent, periodic DFT calculations (GGA-PW91, see the SI for details) to elucidate the ORR mechanism on (100) surfaces of model catalyst surfaces.18−20 In the course of these analyses, we also performed detailed calculations of surface segregation phenomena in the context of ORR, which revealed that the surface composition under ORR conditions tended to differ from the Pt:Ag ratio in the bulk. In general, the surface was enriched in Ag relative to the bulk (see SI for details). As shown in Figure 4B, the most favorable pathway for O−O dissociation on Ag(100) proceeds through the formation of surface OOH (endergonic by 0.35 eV), which dissociates to form O and OH with a barrier of 0.33 eV. Similarly, PtAg3 has an energy barrier of 0.36 eV for the formation of OOH, but its dissociation is more favorable (with a barrier of 0.25 eV), explaining its enhanced activity relative to pure Ag. In contrast, PtAg and Pt3Ag have preferred surface terminations with a Pt:Ag atomic ratio of 1:1 in the exposed layer in the presence of adsorbed O and OH. Adjacent Pt atoms provide bridge sites for the adsorption of O2, which can be dissociated far more easily (with an activation energy less than 0.3 eV) than on the Ag-rich surfaces (which dissociate O2 with an activation energy greater than 1.0 eV) due to the stronger binding 7

ACS Paragon Plus Environment

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

nature of the Pt atoms. The O−O bond in O2 is broken directly on these surfaces and on pure Pt(100) to form adsorbed atomic O. The experimentally observed activity correlates with the transition state energy of this O−O bond-breaking event (Figure 4B): PtAg is the most active and has the most stable O2 transition state (G = −0.67 eV relative to gas-phase O2), followed by Pt3Ag (−0.52 eV) and Pt (−0.47 eV). Our calculations therefore show that a Pt−Ag alloy with a nearly 1:1 ratio of Pt:Ag should yield the highest ORR activity, which qualitatively reproduces the experimental trend. While it would be prohibitively expensive to distinguish between, for example, Pt46Ag54 and Pt56Ag44 (with the latter having the highest experimental activity) through computations with higher composition resolution, our calculations suggest that surface Pt−Pt bridge sites are the active site for ORR (rather than Ag-containing sites) on the surface of the alloys with roughly 1:1 Pt:Ag composition. While the differences in ORR energetics on these Pt−Pt bridge sites is not expected to be large for these very similar alloy compositions (Pt46Ag54 and Pt56Ag44), surface Pt atoms on Pt56Ag44 surface have a higher likelihood of being located adjacent to another Pt atom than in the Pt46Ag54 case and therefore a higher fraction of Pt−Pt bridge sites should be found on the Pt56Ag44 surface. This increased number of active Pt−Pt bridge sites may explain why Pt56Ag44 was most active, in particular when compared with Pt46Ag54, which had a very similar elemental composition. This stabilization trend of the O2 transition state could be attributed to strain and/or ligand effects.21−24 Following the d-band model of Nørskov and coworkers,24,25 the relative position of the d-band center can be used as a simple descriptor for surface activity. To decouple strain and ligand effects and determine their relative contributions towards the observed activity trend, we calculated the d-band center of the surface Pt atoms of strained Pt(100) stretched to the lattice constant of PtAg. These calculations show that the Pt d-band center is raised by only 0.02 eV as a result of the tensile strain; the Pt d-band center is instead raised by 0.2 eV in PtAg relative to pure Pt, which can be primarily attributed to ligand effects from alloying. Similarly, the d-band center of surface Pt in Pt3Ag (which has less Ag) rises by 0.1 eV relative to pure Pt, while the d-band rises by just 0.01 eV due to strain alone (without alloying). This suggests that the rise in d-band center of surface Pt atoms due to ligand effects from the electronegative Ag atoms, rather than lattice strain effects, is the 8

ACS Paragon Plus Environment

Page 8 of 18

Page 9 of 18

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

primary reason for the stabilization of O2 transition state and subsequent improved activity for PtAg and Pt3Ag relative to pure Pt. The experimental activity does not appear to correlate with the calculated energetics of removing adsorbed OH, as shown to be important in our previous work.26,27 Our calculations gave roughly similar energies of OH hydrogenation on all surfaces (within 0.1 eV), with the highest required energy found on PtAg, which has the highest experimental activity. We note that OH stabilization by H2O could be responsible for these variations in OH hydrogenation energy: we uniformly apply a 0.1 eV stabilization to the adsorbed OH (calculated by Greeley et al. for Pt(100)28) on all surfaces. Any decrease in this stabilization on the bimetallic surfaces, particularly due to disruption of the water bilayer caused by lattice expansion with increasing Ag content, would cause the OH hydrogenation energetics to become even more similar and more insignificant compared to the differences in transition state stabilization for O−O bond breaking. Therefore, we believe that the ORR activities are governed by O−O bond breaking kinetics, rather than OH hydrogenation thermodynamics. At a higher Ag proportion, Ag sites will be more prevalent on the surface and the ORR activity is determined by the relative ease of OOH dissociation. A mechanistic change occurs at intermediate Pt:Ag ratios, in which adjacent Pt sites can directly dissociate O2 (instead of OOH); the energy of the O2 transition state (which becomes less stable with increasing Pt content) determines the reaction rate for Pt-rich alloys. More definitive conclusions on reaction rates and rate-determining steps would require additional microkinetic modeling. In conclusion, we have demonstrated the synthesis and catalytic evaluation of a new class of ORR catalysts based on Pt−Ag cubic nanocages. The nanocages were produced via a simple galvanic replacement reaction between Ag nanocubes and a Pt(II) precursor. The mass and specific activities of the Pt−Ag nanocages were highly dependent on their composition. After 10,000 cycles in the accelerated durability test, the mass activity of Pt56Ag44 cubic nanocages showed 3.3-fold enhancement relative to that of a commercial Pt/C before durability testing. Even after 30,000 cycles of an accelerated durability test, the mass activity of Pt−Ag nanocages with a composition of Pt79Ag21 was 1.7 times of that of the commercial Pt/C before durability testing. DFT calculations of the ORR mechanism suggest that the trend in O−O bond breaking energetics is consistent with the experimentally observed activity as a 9

ACS Paragon Plus Environment

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

function of the bimetallic composition. Taken together, this work offers an attractive strategy for designing future catalysts with enhanced activity and durability towards oxygen reduction.

ASSOCIATED CONTENT Supporting Information Supporting Information available: experimental details; TEM images and UV−vis spectrum of the Ag nanocubes; TEM images of Pt−Ag nanocages obtained under different conditions; ICP-MS data for the Pt−Ag nanocages; TEM images of the Pt−Ag nanocages before and after different cycles of accelerated durability tests; FTIR spectra recorded from PVP, Ag nanocubes, and the Pt−Ag nanocages; XPS spectra of the Pt−Ag nanocages; DFT calculations; surface segregation calculations; and ORR calculations. This material is available free of charge via the Internet at http://pubs.acs.org.

AUTHOR INFORMATION Corresponding Authors *E-mail: [email protected] for computation. *E-mail: [email protected] for synthesis and electrochemical characterization.

Notes The authors declare no competing financial interest.

ACKNOWLEDGEMENTS This work was supported in part by start-up funds from Georgia Tech and a grant from DOE-BES (Office of Chemical Sciences, grant DE-FG02-05ER15731). Work at UW-Madison was additionally supported by DOE-BES (Office of Chemical Sciences, grant DE-FG02-05ER15731). ZDH gratefully acknowledges a Graduate Research Fellowship award from the National Science Foundation (DGE-1148903) and the Georgia Tech-ORNL Fellowship. Calculations were performed at supercomputing centers located at the Environmental Molecular Sciences Laboratory, which is sponsored by the DOE Office of

10

ACS Paragon Plus Environment

Page 10 of 18

Page 11 of 18

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Biological and Environmental Research at Pacific Northwest National Laboratory; the Center for Nanoscale Materials at Argonne National Laboratory, supported by DOE contract DE-AC02-06CH11357; and the National Energy Research Scientific Computing Center (NERSC), a DOE Office of Science User Facility supported by DOE contract DE-AC02-05CH11231; and the UW-Madison Center for High Throughput Computing (CHTC), supported by UW-Madison, the Advanced Computing Initiative, the Wisconsin Alumni Research Foundation, the Wisconsin Institutes for Discovery, and the National Science Foundation.

REFERENCES (1) Wang, Y.-J.; Wilkinson, D. P.; Zhang, J. Chem. Rev. 2011, 111, 7625−7651. (2) Debe, M. K. Nature2012, 486, 43−51. (3) Chen, J.; Lim, B.; Lee, E. P.; Xia, Y. Nano Today 2009, 4, 81−95. (4) Stamenkovic, V. R.; Fowler, B.; Mun, B. S.; Wang, G.; Ross, P. N.; Lucas, C. A.; Markovic, N. M. Science 2007, 315, 493−497. (5) Wang, C.; Daimon, H.; Onodera, T.; Koda, T.; Sun, S. Angew. Chem., Int. Ed. 2008, 47, 3588−3591. (6) Zhou, W.; Wu, J.; Yang. H. Nano Lett. 2013, 13, 2870−2874. (7) Chen, C.; Kang, Y.; Huo, Z.; Zhu, Z.; Huang, W.; Xin, H. L.; Snyder, J. D.; Li, D.; Herron, J. A.; Mavrikakis, M.; Chi, M.; More, K. L.; Li, Y.; Markovic, N. M.; Somorjai, G. A.; Yang, P.; Stamenkovic, V. R. Science 2014, 343, 1339−1343. (8) Zhang, L.; Roling, L. T.; Wang, X.; Vara, M.; Chi, M.; Liu, J.; Choi, S.-I.; Park, J.; Herron, J. A.; Xie, Z.; Mavrikakis, M.; Xia, Y. Science 2015, 349, 412−416. (9) Wang, X.; Figueroa-Cosme. L.; Yang, X.; Luo, M.; Liu, J.; Xie, Z.; Xia, Y. Nano Lett. 2016, 16, 1467−1471. (10) Wang, X.; Vara, M.; Luo, M.; Huang, H.; Ruditskiy, A.; Park, J.; Bao, S.; Liu, J.; Howe, J.; Chi, M.; Xie, Z.; Xia, Y. J. Am. Chem. Soc. 2015, 137, 15036−15042. (11) Peng, Z.; Wu, J.; Yang, H. Chem. Mater. 2010, 22, 1098−1106. (12) Chen. J.; Wiley, B.; McLellan, J.; Xiong, Y.; Li, Z.-Y.; Xia, Y. Nano Lett. 2005, 5, 11

ACS Paragon Plus Environment

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

2058−2062. (13) Mahmoud, M. A.; El-Sayed, M. A. Langmuir 2012, 28, 4051−4059. (14) Popa, A.; Samia, A. C. S. Chem. Commun. 2014, 50, 7295−7298. (15) Ruditskiy, A.; Xia, Y. J. Am. Chem. Soc. 2016, 138, 3161−3167. (16) Yang, X.; Yang, M.; Pang, B.; Vara, M.; Xia, Y. Chem. Rev. 2015, 115, 10410−10488. (17) Zhang, Q.; Cobley, C. M.; Zeng, J.; Wen, L.-P.; Chen, J.; Xia, Y. J. Phys. Chem. C 2010, 114, 6396−6400. (18) Kresse, G.; Furthmüller, J. Phys. Rev. B 1996, 54, 11169−11186. (19) Kresse, G.; Furthmüller, J. Comput. Mater. Sci. 1996, 6, 15−50. (20) Perdew, J. P.; Wang, Y. Phys. Rev. B 1992, 45, 13244−13249. (21) Kitchin, J. R.; Nørskov, J. K.; Barteau, M. A.; Chen, J. G. Phys. Rev. Lett. 2004, 93, 156801. (22) Wang, J. X.; Inada, H.; Wu L.; Zhu, Y.; Choi, Y.; Liu, P.; Zhou, W.-P.; Adzic, R. R. J. Am. Chem. Soc. 2009, 131, 17298−17302. (23) Alayoglu, S.; Nilekar, A. U.; Mavrikakis, M.; Eichhorn, B. Nat. Mater. 2008, 7, 333−338. (24) Mavrikakis, M.; Hammer, B.; Nørskov, J. K. Phys. Rev. Lett. 1998, 81, 2819−2822. (25) Bligaard, T.; Nørskov, J. K. Electrochimica Acta 2007, 52, 5512−5516. (26) Xie, S.; Choi, S.-I.; Lu, N.; Roling, L. T.; Herron, J. A.; Zhang, L.; Park, J.; Wang, J.; Kim. M. J.; Xie, Z.; Mavrikakis, M.; Xia, Y. Nano Lett. 2014, 14, 3570−3576. (27) Park, J.; Zhang, L.; Choi, S.-I.; Roling, L. T.; Lu, N.; Herron, J. A.; Xie, S.; Wang, J.; Kim, M. J.; Mavrikakis, M.; Xia, Y. ACS Nano 2015, 9, 2635−2647. (28) Greeley, J.; Rossmeisl, J.; Hellman, A.; Nørskov, J. K. Z. Phys. Chem. 2007, 221, 1209−1220.

12

ACS Paragon Plus Environment

Page 12 of 18

Page 13 of 18

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Figure 1. Schematic illustration depicting the major steps involved in the formation of Pt−Ag nanocages with high ORR activity: (a, b) alloying of Ag with Pt to generate Pt−Ag nanocages during the galvanic replacement reaction between Ag and a Pt(II) precursor and (c) dealloying of Ag from Pt during the accelerated durability test.

13

ACS Paragon Plus Environment

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 2. Typical TEM images of Pt−Ag nanocages (A) before and (B) after etching with 2% aqueous H2O2 for 24 h. The insets in (A) and (B) show TEM images of the nanocages at a higher magnification (scale bar: 10 nm). (C) HAADF-STEM image of a single Pt−Ag nanocage after H2O2 etching. (D−F) EDX mapping of elemental distributions for (D) Ag, (E) Pt, and (F) Ag and Pt.

14

ACS Paragon Plus Environment

Page 14 of 18

Page 15 of 18

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Figure 3. (A) Cyclic voltammograms recorded from Pt56Ag44/C nanocages and a commercial Pt/C catalyst. (B) Comparison of positive-going ORR polarization curves between Pt56Ag44/C nanocages and a commercial Pt/C catalyst. The currents were normalized to the geometric area of the rotating disk electrode. (C and D) Mass and specific activities given as kinetic current density (jk) normalized to the ECSAs and Pt masses of the catalysts, respectively. (E and F) Mass and specific activities measured at 0.9 V.

15

ACS Paragon Plus Environment

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 4. (A) Comparison of the durability of the Pt−Ag nanocages and the commercial Pt/C catalyst after different rounds of repeated potential cycling. The comparison is based on the ORR specific and mass activities at 0.9 VRHE for the catalysts before and after the accelerated ORR durability tests (ADT). (B) Calculated free energy diagram showing the minimum energy pathway for ORR on modeled Pt−Ag(100) surfaces at 0.90 VRHE. The circled regions highlight the transition state energies responsible for the experimentally observed activity trends: PtAg3 is more active than Ag due to more facile dissociation of OOH (red circle), and PtAg is most active due to the most stable transition state energy for direct O2 dissociation (black circle).

16

ACS Paragon Plus Environment

Page 16 of 18

Page 17 of 18

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Figure 5. Typical TEM images of the Pt−Ag nanocages (A and B) before and (C−F) after different rounds of repeated potential cycling: (C) 5,000, (D) 10,000, (E) 20,000, and (F) 30,000 cycles. The cubic shape and hollow structure of the nanocages were essentially retained during the repeated potential cycling despite compositional changes.

17

ACS Paragon Plus Environment

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

TOC

18

ACS Paragon Plus Environment

Page 18 of 18