Subscriber access provided by University of South Dakota
Viewpoint
Prospects of Platinum-Based Nanostructures for The Electrocatalytic Reduction of Oxygen Lei Wang, Adam Holewinski, and Chao Wang ACS Catal., Just Accepted Manuscript • Publication Date (Web): 28 Aug 2018 Downloaded from http://pubs.acs.org on August 28, 2018
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 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 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.
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 32 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
ACS Catalysis
Prospects of Platinum-Based Nanostructures for The Electrocatalytic Reduction of Oxygen
Lei Wang1, Adam Holewinski*2,3, and Chao Wang*1 1
Department of Chemical and Biomolecular Engineering, Johns Hopkins University, Baltimore, Maryland
21218 2
Department of Chemical and Biological Engineering and 3Renewable and Sustainable Energy Institute,
University of Colorado, Boulder, CO 80309 *Email:
[email protected];
[email protected] 1 ACS Paragon Plus Environment
ACS Catalysis 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
Page 2 of 32
Abstract Platinum-based electrocatalysts for fuel cells have been the subject of a vast collection of research over the past several decades.
While the intrinsic area-specific activity of these
materials has arguably been optimized, the mass-specific activity and long-term durability still have room for improvement, and achieving a co-optimization of activity and stability requires consideration of a number of tradeoffs. We provide here a viewpoint on prospects for the continued development of ORR electrocatalysts, focusing on recent advances and guidelines for maximizing the efficient use of platinum. Our discussion starts with a brief review of Pt-based alloy electrocatalysts, illustrating what we perceive as a necessary progression toward core/shell nanostructures—particularly those comprised of a non-precious core and a Pt shell. Shell thickness is discussed in terms of tradeoffs between stabilization of the nonprecious metal core and permission of strain and/or ligand effects on the surface—in other words how effectively the core and surface can “communicate”. Recent examples are then heighted that demonstrate the growth of Pt shells on non-precious core materials and note promising gains in durability without significant sacrifice of the mass activity or cost. Finally, remaining questions to be addressed are summarized, and attention is called to possible directions for further development of core/shell structured electrocatalysts.
Keywords: oxygen reduction reaction, fuel cell, alloy nanoparticles, core shell nanostructures, platinum electrocatalysts
2 ACS Paragon Plus Environment
Page 3 of 32 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
ACS Catalysis
Introduction Hydrogen/oxygen (H2/O2) fuel cells have been intensively pursued as "green" devices for electrical-chemical energy conversion.1 Electrical vehicles powered by proton-exchange membrane fuel cells (PEMFCs) have been commercialized, with deployment growing drastically as hydrogen fueling stations begin to proliferate.2 For PEMFCs, Pt is the most active singlemetal catalyst for both the oxygen reduction reaction (ORR) at the cathode and the hydrogen oxidation reaction (HOR) at the anode. While the state-of-the-art electrocatalysts have been successfully formulated (e.g., by alloying) to reduce the content of precious metals, Pt remains a major constituent in all viable materials.3 Given the scarcity and cost of Pt, further reduction of its usage―particularly for the kinetically more sluggish ORR―thus represents a primary challenge for the wide implementation of fuel cell technologies. Many efforts have been devoted to enhancing the ORR activity by employing bimetallic or multimetallic alloys of Pt with other transition metals (Co, Ni, Cu, Ga, Y, etc.4-12). Alloying represents a common approach to tailor the performance of heterogeneous catalysts, as the heteroatomic structure and interactions can induce modifications to the surface property and reactivity through ligand,13, 14 strain15 and/or ensemble14 effects. For the ORR, it has been found that Pt-alloy catalysts that bind to oxygenated intermediates (e.g., *OH, where * denotes an adsorption site on the catalyst surface) slightly weaker than pure Pt exhibit improved kinetics,11, 16
although there may be some ambiguity in the precise optimum.17 It may be noted that several
Pt-based alloy catalysts (such as the nanostructured thin film (NSTF) catalysts developed by 3M18, 19) have met the 2020 DOE target for mass activity (0.44 A/gPt) in membrane assembly electrodes (MEAs). Despite the progress on catalytic activity improvement, Pt-based alloy catalysts are challenged by the lack of long-term stability under ORR-relevant electrochemical 3 ACS Paragon Plus Environment
ACS Catalysis 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
Page 4 of 32
conditions, and the corresponding DOE target for catalytic durability (10,000 potential cycles,27, 65 and many of these durability tests have not been taken to the full benchmark standard of 30,000 cycles. It may be argued that the most significant activity losses are seen very early in cycling, but one may also notice that real PEMFCs often involve even harsher conditions than the AST protocols (potential cycling between 0.6 and 1.0 V). For example, excursions to very high potentials (up to 1.5 V66, 67) during startup have been reported for the oxygen electrode. Moreover, none of the Pt-alloy electrocatalysts with high mass activities (e.g., >1 A/mgPt) have been shown to be able to stabilize more than 30 atom% of NPM content after extensive potential cycling (10,000 cycles or above). This indicates that the nanostructure generated from dealloying is not stable against continual leaching. Further modification of the surface and crystal structures by applying thermal treatments,22, 51, 55, 68, 69 chemical aging prior to the electrocatalysis7 or introducing heteroatomic dopants10 has been shown to improve the catalytic activity and durability to a degree, but preservation of both high activity and high NPM content after extensive potential cycles has not been demonstrated so far.
Prospects of Core/Shell Nanostructures as ORR Electrocatalysts Core/shell nanostructures are promising alternatives to alloys for the development of advanced ORR electrocatalysts.8, 25, 27, 29, 56, 70-75 As mentioned in the Introduction, a core/shell nanostructure comprising a nonprecious, 3d metal core and a Pt shell could substantially reduce the content of precious metals in the electrocatalyst, while still preserving the interactions between the two metals toward catalytic activity enhancements (Figure 1). In an NPM/Pt 8 ACS Paragon Plus Environment
Page 9 of 32 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
ACS Catalysis
core/shell nanoparticle, a thicker shell may provide better protection of the core metal from leaching out under electrochemical reaction reactions, whereas a thinner Pt layer would be more desirable for taking advantage of the ligand and/or strain effects to enhance the catalytic activity.76 Regarding tradeoffs with shell thickness, we note again that the general guideline for optimizing the specific activity of Pt-based ORR electrocatalysts is to perturb the Pt sites so that they bind to oxygenated intermediates slightly weaker than pure Pt. This can be achieved by altering the Pt-Pt separation (strain effect) or replacing near neighbors of Pt with atoms of different identity (ligand effect). Compressive strain can roughly be considered to increase the bonding interaction between the neighboring undercoordinated surface atoms, making them less apt to bond with approaching adsorbates. At a more fundamental level, this phenomenon is often interpreted in terms of shifts in the d-electron band center of Pt,77-79 which broadens due to increased d-d overlap during compression. The d-band must shift down to preserve constant filling (i.e. nine d-electrons per Pt) as it broadens. This downshift causes adsorbate anti-bonding states to populate to a greater degree when the adsorbate hybridizes with Pt, weakening the bond. The ligand effect can also roughly be interpreted in terms of changes to d-orbital overlap when a host (here Pt) atom is swapped with another element while holding the lattice constant constant—i.e. a major component of the interaction can be considered in terms of whether the ligand atom will overlap with Pt orbitals more or less than another Pt would. However, care must be taken in this interpretation as additional complexities can also play a role. For example, ligands with large electronegativity differences can yield charge transfer (between sp-states; dstates tends to retain their atomic filling80-82), which can change the adsorbate-surface bond and attenuate, or even reverse, the simpler picture83, 84. In comparison to strain, ligand effects are 9 ACS Paragon Plus Environment
ACS Catalysis 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
very short range. Kitchin et al.80,
81
Page 10 of 32
performed a careful delineation of ligand vs. strain
contributions to adsorption using DFT calculations on bimetallic skin-alloy systems (e.g. replacing a single subsurface layer with a new metal while retaining the original lattice constant). Adsorption trends were found to correlate well with interatomic matrix elements (effectively a measure of orbital overlap with the neighbor85), which fall off with distance at approximately d-5. This means that direct influence of a ligand atom is minimal from even the second nearestneighbor position. Explicit calculations of small molecule adsorption on pseudomorphic overlayer systems (Pd on Au86; Pt on Ru87) of varying thickness (mimicking a core/shell arrangement for extended surfaces) confirm that adsorption energies converge to an essentially constant value after as few as two atomic layers are deposited on a substrate86,87, perhaps three for more disparate materials (e.g. metal on oxide substrate).88 Compared to the ligand effect, the prospect of surface strain engineering is more promising in core/shell nanostructures owing to its effectiveness over a larger distance. Previous work on dealloyed Pt-bimetallic catalysts has demonstrated relaxed, but residual, strain in the formed Pt shells that are multiple atomic-layers thick. For example, anomalous XRD measurements by Strasser et al.12 showed that the Pt lattice in the shell (~0.6 nm thick) of surface-depleted PtCu3 alloy nanoparticles (ca. 4-5 nm in diameter) is compressed by as much as 4% (versus bulk Pt crystal).12 Gan et al. used aberration-corrected STEM imaging to map the strain distribution in dealloyed PtFe nanoparticles (~4.5 nm in diameter) and found a gradient compressive strain in the Pt shell, varying from 4% in the third subsurface layer to ~2% in the top surface layer.89 In another study by the Strasser’s group, STEM and EELS-based elemental mapping were combined to characterize dealloyed Pt-Ni alloy nanoparticles, which indicated strain effects influencing activities at distances as large as 2.6 nm (or ~10 atomic layers of Pt).90
10 ACS Paragon Plus Environment
Page 11 of 32 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
ACS Catalysis
From these studies of dealloyed Pt-bimetallic electrocatalysts, it is conceivable that, in NPM/PM core/shell nanoparticles, the surface strain also decreases as the shell thickness increases. Thereby an optimal shell thickness can be postulated to exist that balances the catalytic activity enhancements (desiring thinner shell) and stabilization of the nanostructure (thicker shell preferred). In light of present experimental evidences, we believe there remains an open synthetic question as to whether appreciable lattice strains can be engineered to propagate over larger shell thicknesses that also enhances the nanostructure stability. On the other hand, practical mass activity requirements necessitate that active Pt shells are not unduly thick. To aid in assessing the likelihood of simultaneously meeting these competing criteria, we present in Figure 4a set of benchmarking curves illustrating relationships between the mass activity, specific activity, particle core size, and shell thickness, assuming the core is made completely of nonprecious materials and the shell is pure Pt. As these characteristics obviously have a number of complex interdependencies (particularly the specific activity varying with shell thickness and core composition), we simply intend for the plots to represent the constraints that must be met to reach DOE mass activity targets provided that certain properties are exhibited. Particle size effect on the specific activity, known for the ORR on Pt-based nanoparticulated catalysts,35, 91-94 is not included for simplicity, and one may expect to see the relations discussed here at any given particle sizes. In Figures 4a and b we show the variation of mass activity as a function of particle core size for a series of shell thicknesses, under the assumption of specific activity established at 0.2 mA/cm2 (typical commercial Pt/C level) and 2mA/cm2 (state-of-the-art core/shell catalyst). Figures 4c and d present the inverse, where, for a given specific activity, one can see the maximum allowable shell thickness vs. core size under constraints of particular mass activities.
11 ACS Paragon Plus Environment
ACS Catalysis 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
Page 12 of 32
These plots select a common case of 0.2 A/g and the optimistic value of 0.5 A/g, which has been achieved (stability not thoroughly established though) and slightly exceeds the DOE 2020 target of 0.44 A/g. A final criterion to note is that while well-coordinated Pt shells are considered stable, Pt can corrode much more quickly from small nanoparticles, where the shell may still contain many undercoordinated sites. We thus have placed an additional line marking a “stable zone” for Pt shells, where the overall particle (core plus shell) diameter is at minimum 5 nm. Under this constraint, it can be seen that there is very little room to manipulate core-shell mass activity by the overall particle size—in the limit of large particles the mass activities and maximum allowable shell thicknesses converge to constant values as the radius of curvature approaches the planar limit. On the other hand, this means that controlling the particle size of NPM cores may not be a necessary focal area if the NPM is sufficiently low cost.
Recent Advances in NPM/PM Core/Shell Electrocatalysts As an early example of exploiting core/shell nanostructures as ORR electrocatalysts, Adzic et al. developed a series of so-called “Pt monolayer” electrocatalysts via galvanic displacement of underpotentially deposited Cu (Cu-UPD) on different substrates (e.g., Pd,25 AuNi0.5Fe95 and PdAu96). Although significant improvement in both specific and mass activities have been observed on these so-called “monolayer Pt” catalysts, the need for noble-metal substrates has limited the reduction of catalyst costs, and long-term (>10,000 potential cycles) electrochemical stability has not been demonstrated by this approach. More robust control in core/shell nanostructures can be achieved by seed-mediated growth in colloidal phases. Wang et al. used Au nanoparticles as seeds to grow Au/Pt3Fe core/shell nanoparticles and demonstrated superior catalytic durability (
