Deterministic Design of Chemistry and Mesostructure in Li-Ion Battery Electrodes Paul V. Braun*,†,‡ and John B. Cook*,‡ †
Department of Materials Science and Engineering, Frederick Seitz Materials Research Laboratory, and Beckman Institute for Advanced Science and Technology, University of Illinois at UrbanaChampaign, Urbana, Illinois 61801, United States ‡ Xerion Advanced Battery Corporation, 3100 Research Boulevard St. 320, Kettering, Ohio 45420, United States ABSTRACT: All battery electrodes have complex internal three-dimensional architectures that have traditionally been formed through the random packing of the electrode components. What is now emerging is a new concept in battery electrode design, where the important electronic and ionic pathways, as well as the chemical interactions between the components of the electrode, are deterministically designed. Deterministic design enables far better properties than are possible through random packing, including dramatic improvements in both power and energy. Such a design approach is particularly attractive for emerging high-energy-density materials, which require available free volume as they swell during cycling. In addition to controlled structure, another important facet of the design of such systems is the stable chemical linkages between the active material and the conductive network that survive the lithiation and delithiation processes. In this Perspective, we discuss and provide our views on deterministically designed battery electrodes.
O
to overcome the fundamental limitations while taking into account practical realities, including safety. As background, today’s high-energy Li-ion battery electrodes are formed by casting a homogenized mixture of active chargestorage material, conductive additive, and polymer binder onto a foil current collector (typically Cu for the anode and Al for the cathode); after various processing steps, a ∼50−100 μm laminate with ∼25−40% porosity is formed. During cycling, current is generated in the active material and transferred outside the cell by electron conduction, first through neighboring active material particles or through a contiguous conductive additive pathway and then through the foil current collector. Particularly in the case of cathode materials (e.g., LiCoO2, LiMn2O4, LiFePO4), the electrical conductivity of the active material is typically quite low, in a range of ∼10−4−10−9 S/cm,1 and so a few percent by mass conductive additive is necessary in the cathode. The easiest approach to overcome the limitations of emerging active materials is unfortunately the least practical. The performance of most active materials can be improved significantly simply by making the material quite thin or by forming it into an electrode with a very low active material loading. Both of these approaches are practical dead ends, however, and thus have limited value unless they enable the elucidation of fundamental properties that can later be exploited. For a classical transition metal oxide cathode (e.g., LiCoO2) or intercalation anode (e.g., graphite) where the volume of the host does not change significantly with cycling, any approach that does not provide high active material loading
ver the past decade, there has been an explosion in research on new materials and structures for electrochemical energy storage, primarily lithium based, driven by the nearly universal goal of increasing energy density (energy per unit volume or energy per unit mass) and motivated by the needs of applications ranging from portable electronics to electric vehicles. Although incremental gains in energy density continue to be made through reductions in the thicknesses of cell components (separators, packaging, and foils) and increases in the volume fractions of active materials in the electrodes, at some point, these gains will yield unacceptable reductions in the margin of safety. Providing high energy density, fast charging, and safe usage is mutually exclusive with current transition metal oxide cathodes (e.g., LiCoO2, LiNiCoMnO2, LiNiCoAlO2, LiMn2O4) and anodes (e.g. graphite). If one is open to moving to new materials, increasing energy density is rather straightforward, and there are numerous examples in the academic literature of materials with energy densities far exceeding commercial materials. However, in most cases, the high capacity provided by these new materials requires unacceptable trade-offs. The trade-offs can be economic, which perhaps creative engineering or economies of scale could overcome, or fundamental, including low ion/electron conductivities, large voltage hystereses upon cycling, low (cathode) or high (anode) operating voltages, and large volume changes with cycling. These fundamental problems cannot be solved simply by small tweaks to the electrode or electrolyte. Solving these problems requires both a deep understanding of the physical mechanism of energy storage in these materials and structures and creative solutions © XXXX American Chemical Society
A
DOI: 10.1021/acsnano.8b01885 ACS Nano XXXX, XXX, XXX−XXX
Perspective
www.acsnano.org
Cite This: ACS Nano XXXX, XXX, XXX−XXX
ACS Nano
Perspective
unreactive at the potentials required to store lithium in those active materials (typically below 1 V versus Li/Li+). Carbonaceous scaffolds, such as fibers and nanotubes, have relatively high electronic conductivities, and their density (and, therefore, the inactive mass contributed to the electrode) is low compared to that of metals. However, below 1 V versus Li/Li+, we have concerns that solid-electrolyte interphase (SEI) formation, solvent co-intercalation, Li-ion intercalation, graphitic exfoliation, and deleterious volume expansion may begin to detract from the advantages of carbon. Examples of potentially acceptable carbon-based scaffold materials include systems where the carbon is in the form of graphene or single- to fewlayer nanotubes, which undergo limited intercalation and/or support a stable SEI.12−14 Another potential advantage of carbon and its derivatives is that it can form covalent bonds with many materials (e.g., sulfur, silicon, and Fe3O4), which may confer strong mechanical adhesion and low interfacial resistance between the active material and the scaffold.
is not acceptable as it will reduce the volumetric energy density of the resulting battery. In the emerging alloying and conversion systems, where the volume of the active material changes significantly during cycling, a new opportunity emerges. In these systems, it is imperative that the active material is not packed densely in its low-volume state (nonlithiated/uncharged state), as otherwise the electrode will pulverize when the active material expands and contracts during cycling. Examples of active materials that undergo large volume changes include silicon and tin, which expand 300−400% upon lithiation,2−5 and Fe3O4, which expands nearly 200% upon conversion to Fe + Li2O.6 Making transformative and concurrent gains in energy, kinetics, and safety may indeed be possible if one is willing to consider combining fundamental modifications to the electrode architecture with new high-energy materials such as these, which, in fact, require such modifications.
In this issue of ACS Nano, Marschilok, Reichmanis, and co-workers report the addition of a carboxylated conjugated polymer to provide chemical bonding, using an electrically conductive polymer, between the active material and a three-dimensional carbon nanotube web.
In addition to providing the necessary free volume, the scaffold should also be electrically conductive, which limits the choice to carbons, metals, conductive polymers, and conductive ceramics. Metal-based scaffolds such as nickel and copper, although heavier than carbon, are highly conductive, do not alloy with Li and Na, and are mechanically stable, enabling their use as porous current collectors for high-energy materials such as tin and silicon, which undergo large volume changes with cycling. The fact that metals are both mechanically robust and highly conductive is probably important for obtaining long cycle lifetimes with good kinetics.7,11,15,16 As Braun and co-workers showed, metallic foams with precisely controlled morphology, tortuosity, surface area, mechanical properties, and metallic volume content can be formed with pores below 1 μm in diameter.8 For active materials such as silicon and tin, which must be thin to prevent pulverization during cycling, it is such foams that are most interesting (if the pores are larger, the active material loading is too low). Figure 2A,B shows a StructurePore foam conformally electrodeposited with ∼200 nm film of tin, resulting in an areal loading of ∼10 mAh/cm2 (75 μm electrode thickness), nearly 3-fold greater than the areal loading of current commercial anodes.17 Even when highly loaded with tin, the unique StructurePore morphology and appropriate surface area supports ∼70% capacity retention after 100 cycles. One major drawback of metal scaffolds is that they are considerably denser than carbon, which can add mass to an electrode compared to carbon, but this drawback can be offset in part through the intrinsic strong mechanical properties of Ni and Cu, which may enable the metal foam to be directly used as a current collector. For example, Xerion’s Advanced Battery Corp 200 μm thick StructurePore foam has the same mass as a 10 μm thick copper foil current collector (10 μm is a typical copper foil thickness used for double-sided graphite anodes). Thus, if such a scaffold were to replace the traditional copper foil current collector, the net contribution to the mass of a battery would be minimal. An attractive method used to generate even higher surface area metallic scaffolds other than templating is free-corrosion dealloying. This method starts from a multiphase alloy, which
Electrode Architecture. For electrode materials that undergo large volume changes with cycling, it is critical that the structure retain ion and electrically conductive networks during cycling. It is well-known that if alloying and conversion materials are simply grown as solid films on a substrate that they fail within a few cycles due to pulverization and concurrent loss of electrical connectivity. A way to retain the interconnectivity of the structure during the large volume changes is required. Many groups have explored options to accomplish this goal, including direct deposition of the active material on three-dimensionally structured metal foams (Figure 1A−G),7,8 encapsulation of the active material inside of carbon spheres,9 block-co-polymer templated porous silicon,10 or porous tin synthesized from a dealloyed tin−magnesium alloy.11 In this issue of ACS Nano, Marschilok, Reichmanis, and co-workers report the addition of a carboxylated conjugated polymer to provide chemical bonding, using an electrically conductive polymer, between the active material and a three-dimensional (3D) carbon nanotube web (Figure 1H− J).12 What is particularly notable about this work is how the authors effectively illustrate the importance of considering both the electrical and mechanical interactions between the many different components of a Li-ion battery electrode. Regardless of the type of the specific charge storage chemistry, intimate mechanical and electrical communication of the active material with the current collector is imperative for energy, rate performance, and safety. Scaffold Requirements. In addition to providing the necessary free volume, the scaffold should also be electrically conductive, which limits the choice to carbons, metals, conductive polymers, and conductive ceramics. Considering the desire to commercialize high-capacity anode materials like silicon and tin, the scaffold should be electrochemically B
DOI: 10.1021/acsnano.8b01885 ACS Nano XXXX, XXX, XXX−XXX
ACS Nano
Perspective
Figure 1. (A) Schematic of a battery containing a bicontinuous cathode. (B) Illustration of the four primary resistances in a battery electrode. (C) Bicontinuous electrode fabrication process. The electrolytically active phase is yellow, and the porous metal current collector is green. The electrolyte fills the remaining pores. Images A−C adapted with permission from ref 8. Copyright 2011 Macmillan Publishers Limited. Scanning electron microscope images of (D) a Ni inverse opal after template removal (plane view), (E) an electropolished Ni inverse opal (plane view), (F) a cross section of the electropolished Ni scaffold, and (G) the structure after silicon deposition via chemical vapor deposition (plane view). Images D−G adapted from ref 7. Copyright 2012 American Chemical Society. Fabrication of few-walled carbon nanotube (FWNT) web electrode. (H) Schematic representation of the overall fabrication procedure for FWNT web electrode, composed of PEG-sFe3O4/CB/FWNT/poly[3-(potassium-4-butanoate) thiophene]. (I) Cycling performance (capacity retention as a function of cycle number) collected for the current density of 0.5 C (∼450 mA g−1) between 0.01 and 3 V (open circle: Coulombic efficiency of FWNT web electrode). (J) Schematics of the formation of Fe-carboxylate complex. Images H−J adapted from ref 12. Copyright 2018 American Chemical Society.
includes a sacrificial component that can be removed either chemically or electrochemically.18 The surface areas of these materials range between 5−200 m2/g (pores of 50 nm diameter or less),19−21 enabling even greater loadings of high-capacity active materials. Figure 2C−F shows a nanoporous gold scaffold produced by dealloying silver from a gold−silver alloy. Tin was deposited via electroless plating, yielding good cycle lifetimes and fast kinetics as a result of the decreased ion diffusion path lengths. For obvious reasons gold, cannot be used because of its cost, but this work highlights a strategy to produce nanostructured tin electrodes with a good cycle life and fast kinetic properties. As a more practical solution, copper, nickel, and numerous bimetallic scaffolds can be formed using this dealloying method.22,23 A potential drawback of these highsurface-area foams is their relatively high metal volume content,
which can add inactive mass to the battery, thereby diminishing the returns of using a high-capacity active material. Furthermore, the first cycle Coulombic efficiency of these high-surface-area materials is low (e.g., ∼70% or less), which, in practice, will add the requirement of a prelithiation step.24,25 A scaffold-free dealloying method to create high-surface-area (19 m2/g) tin with pores between 100 and 300 nm is illustrated in Figure 2G,H. Specifically, the porous tin is synthesized by dealloying magnesium from a tin−magnesium alloy. A trade-off of this approach is that although a conductive scaffold is not required to create the necessary porosity, because the product of the dealloying is a powder, a polymer binder with conductive additives is required. When tested as a slurry electrode, the material shows a good cycle life and fast kinetics. We suggest C
DOI: 10.1021/acsnano.8b01885 ACS Nano XXXX, XXX, XXX−XXX
ACS Nano
Perspective
Figure 3. (A) Scanning electron microscope (SEM) images of planar LiCoO2 films (∼20% porosity) electroplated on both sides of an Al foil. (B) Higher-magnification view of the LiCoO2 coating. (C) Optical images of LiCoO2 electroplated on the Al foil and this electrode rolled into a 5 mm diameter tube (inset). Scanning electron microscope images of (D) the open-cell carbon foam and (E) the LiCoO2/carbon foam electrode. (F) Lower-magnification view of a ∼0.5 mm thick LiCoO2/carbon foam electrode, with LiCoO2 plated uniformly throughout the foam. Scanning electron microscope images of (G) the three-dimensional cellulose nanofibril scaffold and the LiCoO2 electrodes electroplated on this scaffold with (H) ∼1 mA·h cm−2 loading and (I) ∼3 mA·h cm−2 loading. Adapted with permission from ref 26. Copyright 2017 Zhang et al.
Figure 2. (A,B) Low- and high-magnification scanning electron microscope (SEM) images of the StructurePore Ni scaffold. Scale bar in inset of (B) is 500 nm.17 (C,D) Synthetic scheme for forming nanoporous Au-supported nanocrystalline tin. The dealloying process in (C) generates a three-dimensional (3D) nanoporous Au substrate, the pores of which are traps for the nanocrystalline silicon in (D). Corresponding SEM images in (E,F) reveal the microstructure of the 3D nanoporous Au-supported nanocrystalline silicon thin foil for the stages shown in (C,D), respectively. Images C−F adapted with permission from ref 19. Copyright 2011 WileyVCH Verlag GmbH & Co. (G,H) Scanning electron microscopy images of nanoporous tin (NP-Sn) at different magnifications. Lowmagnification image of as-synthesized NP-Sn powder indicates that it consists of randomly shaped NP-Sn grains with sizes in the sub10 μm range. The higher-magnification images show the porous nanostructure, which consists of 100−300 nm ligaments and pores. Adapted from ref 11. Copyright 2017 American Chemical Society.
undergo tremendous volume changes during cycling, good adhesion to the current collector will be critical for achieving high power and long cycle lifetimes. Practical Considerations: Tabbing, Uniformity, and Manufacturability. The practical use of porous 3D scaffolds is limited to the extent that reliable electrical connections can be made to the outside world. In commercial cells, metal tabs are usually attached to the electrodes via direct resistance welding or ultrasonic welding. Although we have found that tabs can be attached to metal foams, we suspect tabbing to carbon will be quite challenging. One option is to deposit the carbon scaffold on a conventional current collector; however, the addition of a conventional current collector decreases the advantage of using a scaffold. Another practical consideration is the uniformity of the active material loading. Modern manufacturing methods control the active material loading across the electrode within ∼2%, which is the equivalent of just a micrometer or two in total electrode thickness. Controlling the active material loading with 2% in a scaffolded system is likely to be challenging. Along with these two issues, there are numerous other manufacturing issues to consider. Is the scaffold formation process scalable? Are the energy inputs to scaffold formation and active material deposition on the scaffold acceptable? The margins associated with Li-ion batteries are slim, and if expensive manufacturing steps are required to form a scaffolded electrode, the performance advantages need to be considerable and not incremental.
that porous metals, like tin, provide an interesting opportunity for creating scaffold-free porous monolithic electrodes. Active Material Deposition. The requirements for an electrically conductive scaffold add significant constraints to the synthesis of the active material to be incorporated within the scaffold. If the active material is synthesized independent of the scaffold, there must be a method either to add the active material (most likely in nanoparticle form) to a preformed scaffold or to form the scaffold in the presence of the active material. The other option is to synthesize the active material within the scaffold, with the challenge that the commonly required active material synthesis conditions (e.g., temperatures exceeding 700 °C in an oxidizing environment) will deteriorate almost any scaffold material. These constraints have motivated synthetic methods, primarily electrochemical and hydrothermal, that are amenable to direct formation of active material in a high-performance form on the scaffold (Figure 3).13,26 Synthetic methods leverage the advantage of a direct connection of the active material to the scaffold and current collector, which not only decreases the interfacial resistance but also improves the mechanical adhesion. In particular, for systems such as tin, silicon, and conversion oxides, which
CONCLUSIONS AND PROSPECTS All battery electrodes have 3D architectures, and considerable effort has been made to ensure that these architectures provide D
DOI: 10.1021/acsnano.8b01885 ACS Nano XXXX, XXX, XXX−XXX
ACS Nano
Perspective
(12) Kwon, Y. H.; Park, J. J.; Housel, L. M.; Minnici, K.; Zhang, G.; Lee, S. R.; Lee, S. W.; Chen, Z.; Noda, S.; Takeuchi, E. S.; Takeuchi, K. J.; Marschilok, A. C.; Reichmanis, E. Carbon Nanotube Web with Carboxylated Polythiophene “Assist” for High-Performance Battery Electrodes. ACS Nano 2018, DOI: 10.1021/acsnano.7b08918. (13) Kim, H.-S.; Cook, J. B.; Tolbert, S. H.; Dunn, B. The Development of Pseudocapacitive Properties in Nanosized-MoO2. J. Electrochem. Soc. 2015, 162, A5083−A5090. (14) Ko, M.; Chae, S.; Jeong, S.; Oh, P.; Cho, J. Elastic a-Silicon Nanoparticle Backboned Graphene Hybrid as a Self-Compacting Anode for High-Rate Lithium Ion Batteries. ACS Nano 2014, 8, 8591− 8599. (15) Zhang, H.; Shi, T.; Wetzel, D. J.; Nuzzo, R. G.; Braun, P. V. 3D Scaffolded Nickel−Tin Li-Ion Anodes with Enhanced Cyclability. Adv. Mater. 2016, 28, 742−747. (16) Ning, H.; Pikul, J. H.; Zhang, R.; Li, X.; Xu, S.; Wang, J.; Rogers, J. A.; King, W. P.; Braun, P. V. Holographic Patterning of HighPerformance On-Chip 3D Lithium-Ion Microbatteries. Proc. Natl. Acad. Sci. U. S. A. 2015, 112, 6573−6578. (17) Sun, P.; Davis, J.; Cao, L.; Jiang, Z.; Cook, J. B.; Ning, H.; Liu, J.; Kim, S.; Fan, F.; Nuzzo, R. G.; Braun, P. V. High Capacity 3D Structured Tin-Based Electroplated Li-ion Battery Anodes (submitted for publication). (18) Erlebacher, J.; Aziz, M. J.; Karma, A.; Dimitrov, N.; Sieradzki, K. Evolution of Nanoporosity in Dealloying. Nature 2001, 410, 450−453. (19) Yu, Y.; Gu, L.; Lang, X.; Zhu, C.; Fujita, T.; Chen, M.; Maier, J. Li Storage in 3D Nanoporous Au-Supported Nanocrystalline Tin. Adv. Mater. 2011, 23, 2443−2447. (20) Detsi, E.; Cook, J. B.; Lesel, B. K.; Turner, C. L.; Liang, Y.-L.; Robbennolt, S.; Tolbert, S. H. Mesoporous Ni60Fe30Mn10-Alloy Based Metal/Metal Oxide Composite Thick Films as Highly Active and Robust Oxygen Evolution Catalysts. Energy Environ. Sci. 2016, 9, 540−549. (21) Detsi, E.; Vukovic, Z.; Punzhin, S.; Bronsveld, P. M.; Onck, P. R.; Hosson, J. T. M. D. Fine-Tuning the Feature Size of Nanoporous Silver. CrystEngComm 2012, 14, 5402−5406. (22) Fujita, T.; Kanoko, Y.; Ito, Y.; Chen, L.; Hirata, A.; Kashani, H.; Iwatsu, O.; Chen, M. Nanoporous Metal Papers for Scalable Hierarchical Electrode. Adv. Sci. 2015, 2, 1500086. (23) Kunduraci, M. Dealloying Technique in the Synthesis of Lithium-Ion Battery Anode Materials. J. Solid State Electrochem. 2016, 20, 2105−2111. (24) Forney, M. W.; Ganter, M. J.; Staub, J. W.; Ridgley, R. D.; Landi, B. J. Prelithiation of Silicon−Carbon Nanotube Anodes for Lithium Ion Batteries by Stabilized Lithium Metal Powder (SLMP). Nano Lett. 2013, 13, 4158−4163. (25) Kim, H. J.; Choi, S.; Lee, S. J.; Seo, M. W.; Lee, J. G.; Deniz, E.; Lee, Y. J.; Kim, E. K.; Choi, J. W. Controlled Prelithiation of Silicon Monoxide for High Performance Lithium-Ion Rechargeable Full Cells. Nano Lett. 2016, 16, 282−288. (26) Zhang, H.; Ning, H.; Busbee, J.; Shen, Z.; Kiggins, C.; Hua, Y.; Eaves, J.; Davis, J.; Shi, T.; Shao, Y.-T.; Zuo, J.-M.; Hong, X.; Chan, Y.; Wang, S.; Wang, P.; Sun, P.; Xu, S.; Liu, J.; Braun, P. V. Electroplating Lithium Transition Metal Oxides. Sci. Adv. 2017, 3, e1602427.
electronic and ionic connectivity and minimize undesirable chemical reactions. We are seeing now that the important elements of a battery electrode can be deterministically designed, including the electron conduction pathways, the ion conduction pathways, the free volume required to provide space for active materials to swell into during cycling and, now as Marschilok, Reichmanis, and co-workers report, the chemical linkages between the active material and the conductive network. In particular, now that the community is starting to take advantage of high-energy-density anode materials where free volume in the delithiated state is acceptable and, in fact, preferable, deterministic design concepts have real opportunities for impact, and, not surprisingly, the number of reports in this space is increasing rapidly.
AUTHOR INFORMATION Corresponding Authors
*E-mail:
[email protected]. *E-mail:
[email protected]. ORCID
Paul V. Braun: 0000-0003-4079-8160 Notes
The authors declare no competing financial interest.
REFERENCES (1) Park, M.; Zhang, X.; Chung, M.; Less, G. B.; Sastry, A. M. A Review of Conduction Phenomena in Li-Ion Batteries. J. Power Sources 2010, 195, 7904−7929. (2) Cook, J. B.; Lin, T. C.; Detsi, E.; Weker, J. N.; Tolbert, S. H. Using X-Ray Microscopy To Understand How Nanoporous Materials Can Be Used To Reduce the Large Volume Change in Alloy Anodes. Nano Lett. 2017, 17, 870−877. (3) Huang, J. Y.; Zhong, L.; Wang, C. M.; Sullivan, J. P.; Xu, W.; Zhang, L. Q.; Mao, S. X.; Hudak, N. S.; Liu, X. H.; Subramanian, A.; Fan, H.; Qi, L.; Kushima, A.; Li, J. In Situ Observation of the Electrochemical Lithiation of a Single SnO2 Nanowire Electrode. Science 2010, 330, 1515−1520. (4) McDowell, M. T.; Lee, S. W.; Nix, W. D.; Cui, Y. 25th Anniversary Article: Understanding the Lithiation of Silicon and Other Alloying Anodes for Lithium-Ion Batteries. Adv. Mater. 2013, 25, 4966−4985. (5) Nitta, N.; Yushin, G. High-Capacity Anode Materials for LithiumIon Batteries: Choice of Elements and Structures for Active Particles. Part. Part. Syst. Charact. 2014, 31, 317−336. (6) Taberna, P. L.; Mitra, S.; Poizot, P.; Simon, P.; Tarascon, J. M. High Rate Capabilities Fe3O4-Based Cu Nano-Architectured Electrodes for Lithium-Ion Battery Applications. Nat. Mater. 2006, 5, 567− 573. (7) Zhang, H.; Braun, P. V. Three-Dimensional Metal Scaffold Supported Bicontinuous Silicon Battery Anodes. Nano Lett. 2012, 12, 2778−2783. (8) Zhang, H.; Yu, X.; Braun, P. V. Three-Dimensional Bicontinuous Ultrafast-Charge and -Discharge Bulk Battery Electrodes. Nat. Nanotechnol. 2011, 6, 277−281. (9) Liu, N.; Lu, Z.; Zhao, J.; McDowell, M. T.; Lee, H.-W.; Zhao, W.; Cui, Y. A Pomegranate-Inspired Nanoscale Design for Large-VolumeChange Lithium Battery Anodes. Nat. Nanotechnol. 2014, 9, 187−192. (10) Cook, J. B.; Kim, H.-S.; Lin, T. C.; Robbennolt, S.; Detsi, E.; Dunn, B. S.; Tolbert, S. H. Tuning Porosity and Surface Area in Mesoporous Silicon for Application in Li-Ion Battery Electrodes. ACS Appl. Mater. Interfaces 2017, 9, 19063−19073. (11) Cook, J. B.; Detsi, E.; Liu, Y.; Liang, Y.-L.; Kim, H.-S.; Petrissans, X.; Dunn, B.; Tolbert, S. H. Nanoporous Tin with a Granular Hierarchical Ligament Morphology as a Highly Stable Li-Ion Battery Anode. ACS Appl. Mater. Interfaces 2017, 9, 293−303. E
DOI: 10.1021/acsnano.8b01885 ACS Nano XXXX, XXX, XXX−XXX
