Scalable Fracture-free SiOC Glass Coating for Robust Silicon

Publication Date (Web): November 5, 2014 ... Citation data is made available by participants in Crossref's Cited-by Linking service. ... Won Chul Cho ...
0 downloads 0 Views 5MB Size
Download PDF
Letter pubs.acs.org/NanoLett

Scalable Fracture-free SiOC Glass Coating for Robust Silicon Nanoparticle Anodes in Lithium Secondary Batteries Sunghun Choi,†,§ Dae Soo Jung,‡,§ and Jang Wook Choi*,† †

Graduate School of Energy, Environment, Water, and Sustainability (EEWS) and Center for Nature-inspired Technology (CNiT) in KAIST Institute NanoCentury, Korea Advanced Institute of Science and Technology (KAIST), 291 Daehakro, Yuseong-gu, Daejeon 305-701, Republic of Korea ‡ Eco-Composite Materials Team, Korea Institute of Ceramic Engineering & Technology (KICET), Seoul 153-801, Republic of Korea S Supporting Information *

ABSTRACT: A variety of silicon (Si) nanostructures and their complex composites have been lately introduced in the lithium ion battery community to address the large volume changes of Si anodes during their repeated charge−discharge cycles. Nevertheless, for largescale manufacturing it is more desirable to use commercial Si nanoparticles with simple surface coating. Most conductive coating materials, however, do not accommodate the volume expansion of the inner Si active phases and resultantly fracture during cycling. To overcome this chronic limitation, herein, we report silicon oxycarbide (SiOC) glass as a new coating material for Si nanoparticle anodes. The SiOC glass phase can expand to some extent due to its active nature in reacting with Li ions and can therefore accommodate the volume changes of the inner Si nanoparticles without disintegration or fracture. The SiOC glass also grows in the form of nanocluster to bridge Si nanoparticles, thereby contributing to the structural integrity of secondary particles during cycling. On the basis of these combined effects, the SiOC-coated Si nanoparticles reach a high reversible capacity of 2093 mAh g−1 with 92% capacity retention after 200 cycles. Furthermore, the coating and subsequent secondary particle formation were produced by high-speed spray pyrolysis based on a single precursor solution. KEYWORDS: Carbon coating, lithium ion battery, silicon anode, silicon oxycarbide, spray pyrolysis

B

enough but are instead subject to fracture during lithiation (Figure 1A), leaving the original capacity fading issue of Si anodes unsolved in long-term cycling viewpoint. While searching for robust coating materials, we have realized that Si monoxide (SiOx, x ∼ 1), a recent alternative phase to pure Si, offers commercial standard cycling performance by accommodating the volume changes of internal Si nanodomains through silica matrix.23−25 Hence, we hypothesized that Si-containing glass materials could fulfill a similar role by their superior mechanical properties when coated on Si NPs. Among such candidates, in the current study we chose silicon oxycarbide (SiOC) glass because SiOC phases consist of amorphous SiOx domains that function as active phases with decent specific capacities around ∼800 mA g−1.26−28 At an optimal thickness, the SiOC coating layer expands to some extent by its own active character in reaction with Li ions and so accommodates the volume expansion of the inner Si NPs in a more robust fashion (Figure 1B) than other coating materials undergoing lesser degrees of volume expansion. But its expansion is not too serious to initiate the aforementioned capacity fading mechanisms intrinsic to bare Si NPs. Overall, the delicate structural control engaging the SiOC coating layers

ecause of the unparalleled theoretical capacity near 4000 mA g−1, silicon (Si) anodes are expected to play a key role in bringing future lithium ion battery (LIB) applications to a reality. Green sustainable vehicles and advanced mobile electronic devices are representative examples along this direction. Despite the promising feature related to the high specific capacity, Si anodes have suffered from insufficient cycle lives in the commercial standards originating from large volume change up to ∼300% between fully charged and discharged states. The severe volume change causes various capacity fading mechanisms, such as pulverization of active material, electrode film delamination, and unstable solid-electrolyte-interphase (SEI) formation.1−3 A variety of Si nanostructures and composites have been developed4−17 or discovered18−22 to manage the large volume change of Si and have resultantly demonstrated significantly improved cycling performance. While these structural approaches represent remarkable progresses in the Si anode research, resolving the chronic issues by simple surface coating onto nanoparticles (NPs) would be more preferred for largescale manufacturing because the simple coating and the use of NPs as an active Si morphology can simplify the processing and lower the cost. However, achieving sustainable coating during repeated volume change of Si has been nontrivial at all because most conductive coating materials including commonly adopted amorphous carbon are not elastic or expandable © 2014 American Chemical Society

Received: September 20, 2014 Revised: November 3, 2014 Published: November 5, 2014 7120

dx.doi.org/10.1021/nl503620z | Nano Lett. 2014, 14, 7120−7125

Nano Letters

Letter

Figure 1. Graphical illustration of Si NP anodes with different coating materials during lithiation/delithiation. (A) Carbon coating versus (B) SiOC coating. (Lower inset) the internal structure of the SiOC coating material.

Figure 2. Characterization of Si@SiOC composite. (A) Formation mechanism of Si@SiOC composite through vapor deposition and clustering step. SEM images of Si@SiOC in (B) low and (C) high magnifications. HRTEM images of Si@SiOC showing (D) conformal SiOC deposition on Si and (E) interconnected SiOC nanoclusters. (Insets of D and E) Corresponding SADPs showing the crystalline Si and amorphous character of SiOC, respectively. (F) Energy-dispersive X-ray spectra from the points 1 and 2 in (C). (G) XRD pattern of Si@SiOC indicative of coexistence of crystalline Si phase and amorphous SiOC phase. (H) TGA curve measured under air at a heating rate of 10 °C min−1, indicating that the free carbon content in the composite is 6 wt %.

to-one particle conversion,29 PTES can be easily vaporized due to its low boiling point of 235 °C at 1 atm and can be subsequently deposited onto the Si NP surfaces via aerosolassisted chemical vapor deposition (AACVD) mechanism (Figure 2A).30,31 The vaporized PTES deposited on the Si NPs transforms into SiOC glass via a pyrolysis process at 800 °C followed by a heat-treatment at the same temperature for 1 h, forming conformal surface coating layers. Also, during this deposition process the SiOC could grow in the form of nanocluster and serve as bridges between Si@SiOC particles,

allows one to catch two challenging rabbits in Si anode operations: the CEs and long-term capacity retention. For the purpose of scalable synthesis of SiOC-coated Si NPs (denoted as Si@SiOC), high-speed spray pyrolysis was employed (Supporting Information Figure S1A). A single precursor solution was prepared by dispersing 6 g Si NPs (ave. 50 nm, MTI, U.S.A.) in 400 mL of 0.1 M phenyltriethoxysilane (PTES), a precursor of glass coating material. Unlike typical spray pyrolysis where precursor droplets are pyrolyzed into final particles with micrometer dimensions based on one drop7121

dx.doi.org/10.1021/nl503620z | Nano Lett. 2014, 14, 7120−7125

Nano Letters

Letter

Figure 3. (A) The first charge/discharge voltage profiles of Si@SiOC measured at 100 mA g−1. (B) Initial discharge (delithiation) capacities and Coulombic efficiencies of various electrode samples. (C) The cycling performances of the given three samples measured at 1C and (D) the corresponding Coulombic efficiencies. (E,F) The SEM images of Si@SiOC after 50 cycles.

information that the free carbon present in the SiOC phase accounts for 6 wt % of the entire composite. As shown in Supporting Information Figure S4, Si@SiOC exhibited a similar tap density to that of bare Si NPs because the packing of the Si@SiOC powder is dictated mainly by that of the core Si NPs. Prior to electrochemical testing of the Si@SiOC composite, the SiOC glass was first characterized for its structural understanding and electrochemical performance as a LIB anode (Supporting Information Figures S5−S7). In the absence of Si NPs in the precursor solution, the SiOC glass was produced in the form of nanocluster with size around 80 nm (Supporting Information Figure S5B,C) and uniform elemental distributions (Supporting Information Figure S5D). It bears internal carbon structure with 16 wt % and poor crystallinity according to TGA (Supporting Information Figure S6B) and Raman results (Supporting Information Figure S6C), respectively. The SiOC particles also contain a large portion of silica nanodomains, which was indicated by a significantly increased porosity (Supporting Information Figure S6D,E) after hydrofluoric acid (HF) treatment. This increased porosity also implies that most of the silica nanodomains are not bonded to the internal carbon network. Also,29Si MAS NMR (Supporting Information Figure S6F) and FT-IR (Supporting Information Figure S6G) spectra of the HF-treated SiOC indicated that there still exist Si−O−C interdomains, such as SiC2O2 and SiCO3, between the SiO2 domains and the carbon network. These domains must contribute to the mechanical strength26,32 of the SiOC coating layer via its superior Vickers hardness (7.0−8.6 versus 6.0−7.0 of SiO2) and elastic modulus (97.9

leading to formation of micron-sized secondary composite particles. See scanning electron microscope (SEM) images in Figures 2B and Supporting Information Figures S2 and S3. Remarkably, the final composite powder was generated under a high-speed flow condition such that carrier gas flow takes only ∼5 s from one end to the other in a tube furnace (tube length: 60 cm). Thus, the current procedure is suitable for large-scale production. See the Supporting Information for experimental details. The Si@SiOC composite was further characterized by using various analytical tools. From high-magnification SEM image (Figure 2C), it was confirmed that the secondary particles consist of interconnected Si@SiOC nanoparticles and bear void space in their internal structure. The conformal coating (Figure 2D) and the particle-to-particle bridging structure (Figure 2E) of the SiOC layers were additionally verified by transmission electron microscope (TEM) analyses. Also, fast Fourier transform (FFT) patterns obtained from the TEM images indicated amorphous nature of the SiOC coating layers (insets of Figure 2D,E) in conjunction with the presence of crystalline Si from the corresponding FFT pattern (Figure 2D inset). Energy-dispersive X-ray spectra (EDS) (Figure 2F) from the points 1 and 2 of the SEM image in Figure 2C reflected distinct compositions of SiOC between the bridging nanocluster and surface coating, as the point 2 showed a higher Si-to-O ratio. Also, X-ray diffraction (XRD) spectrum exhibited an amorphous characteristic of the SiOC glass combined with crystalline Si indicated by discrete Si diffraction peaks. Finally, thermal gravimetric analysis (TGA) provided compositional 7122

dx.doi.org/10.1021/nl503620z | Nano Lett. 2014, 14, 7120−7125

Nano Letters

Letter

Figure 4. TEM characterization of individual Si NPs for the three different electrode cases: (A) bare Si NPs, (B) Si@C, and (C) Si@SiOC before and after cycling at 100 mA g−1.

to-O ratio.37 In our first cycle data, while bare Si NPs showed the highest ICE of 85%, Si@SiOC, Si@C, SiOC, and amorphous carbon showed ICEs of 72, 69, 43, and 28%, respectively, thus confirming that the inherently low ICE of the SiOC phase can be resolved to a large degree by integration with pristine Si. The amorphous carbon also turned out to suffer from low ICE, similarly due to its internal structure that traps Li ions, and for the same reason the amorphous carbon causes the lower ICE of Si@C when used as a coating material. Si@SiOC exhibited superior performance in the long term cyclability. For this testing, all of the samples were cycled at 1C, which correspond to 2.0, 1.7, and 3.2 A g−1 for Si@SiOC, Si@ C, and bare Si NPs, respectively. All of the samples underwent precycling at 0.05C to form stable SEI, and the cycle number in the current evaluation of the cycling performance starts from the subsequent cycle, not the precycling. Although bare Si NPs began with a high capacity of 2779 mAh g−1, its capacity decayed drastically to 901 mAh g−1 after 200 cycles, leading to a poor capacity retention of 33%. In the case of Si@C, its initial capacity of 1266 mAh g−1 was retained for the first 25 cycles but dropped abruptly thereafter. After 200 cycles, the capacity retention was only 44%. The abrupt capacity drop is ascribed to cracks and disintegration of the carbon coating layers after repeated charge−discharge cycles, and from this point its capacity retention behaves similarly to that of bare Si NPs. By sharp contrast, Si@SiOC retained 92% of the capacity after the same 200 cycles, validating our design approach utilizing the SiOC coating. The initial increase in the capacity for the first 25 cycles might be due to the slow activation of the SiOC phase (Supporting Information Figure S7B,C). Also, the capacity retention behavior can be correlated with the CE. The CEs of bare Si NPs started at a low value around 98.2% in the second cycle and dropped immediately to 97.2% through the next seven cycles. Overall, the CEs of Si NPs stayed in the range of 97.2−98.7% throughout the cycling, again owing to the

GPa versus 70 GPa of SiO2),26,33,34 and thus play a role in accommodating the volume change of the Si active phase. Although the SiOC glass showed a relatively low initial CE (ICE) of 43% (Supporting Information Figure S7A) presumably due to Li trapping in the SiO2 matrix,35 it showed very robust cycling performance in a way that the specific capacity of 720 mAh g−1 was not lost at all for 150 cycles (Supporting Information Figures S7B,C), verifying its structural stability to accommodate the volume change of the internal active phase. From differential capacity curves at the 50th and 100th cycles (Supporting Information Figure S7C inset), two anodic peaks were observed at 0.32 and 0.45 V versus Li/Li+, a signature of amorphous Si domains.36 This phenomenon implies that once amorphous Si domains are formed in the first cycle, the overall electrochemical reaction of SiOC remains similar to amorphous Si in the subsequent cycles. When galvanostatically tested (Figure 3A), the Si@SiOC composite showed characteristic lithiation/delithiation plateaus of Si at 0.1/0.42 V versus Li/Li+ with respective capacities of 2907/2093 mAh g−1. From the weight portion (37.5 wt %) and specific capacity (833 mAh g−1) of SiOC, the observed capacities indicate that Si NPs deliver 2763 mAh g−1, and a large portion (∼92%) of Si NPs participate in the reaction with Li ions. To see the SiOC coating effect more distinctively, widely used amorphous carbon-coated Si NPs (denoted as Si@ C) and bare Si NPs were also tested as control samples. For fair comparison, the carbon-coating content (38 wt %, Supporting Information Figure S8) of Si@C was controlled to be similar to the glass coating content (37.5 wt %) of Si@SiOC. As described above, the main motivation of the SiOC coating is the stable structural integrity of Si NPs over long cycling period by utilizing its capability of buffering the volume change of Si NPs. Meanwhile, it should also be noted that the presence of Si NPs in the Si@SiOC composite is expected to increase the poor ICE of the pristine SiOC phase based on the increased Si7123

dx.doi.org/10.1021/nl503620z | Nano Lett. 2014, 14, 7120−7125

Nano Letters

Letter

Si NPs during repeated volume change of Si and thus realize the excellent cycling performance. Overall, the present composite structure can be a competitive intermediate technology connecting the gap between the existing SiOx in the commercial products and ever-challenging bare Si and can thus be a next reasonable destination in the emerging Si anode technology.

combined aforementioned failure mechanisms of bare Si. As in the cycling data, Si@C showed relatively high CEs of 98.0− 98.6% in the cycle range of 2−15. But the value dropped around the cycle number of 15 because of the aforementioned cracks and disintegration of the carbon coating layers. Although the CE was bounced back up around the 70th cycle but never surpassed 99.0%. In comparison, the CE of Si@SiOC reached 90.9 and 99.4% even in the first and second cycles (after the precycling), and steadily increased to 99.7% at the 200th cycle. The average CE of Si@SiOC in the cycle range of 2−200 was 99.3%. The relatively lower CEs in the beginning of the cycling might be due to amorphous SiO2 domains that trap Li ions. However, once the cycling passes this stage, the pulverizationfree character of Si@SiOC allows the CE to stay at high values. The voltage profiles in the early cycle stage are given in Supporting Information Figure S9. Also, Si@SiOC showed good rate capability (Supporting Information Figure S10), presumably due to the SiOx nanodomains surrounded by the carbon network in SiOC coating layer as well as the internal open space in the micron-sized composite particles. As evidence of the structural stability of Si@SiOC during cycling, SEM images (Figure 3E,F) after 50 cycles confirmed preserved powder morphology where the interconnected nanostructures constitute secondary particles. The void space present in the powder is anticipated to release the stress efficaciously during the volume change of the inner active components (Supporting Information Figure S11). The current SiOC coating is fundamentally distinct from the previous surface coating based on silica38,39 and conducting polymers40−42 in that SiOC undergoes a larger volume expansion to follow the volume changes of the inner Si NPs without disintegration. The previous coating would not be as sustainable in tolerating the volume change of Si. To elucidate the coating effect further, each electrode was characterized at different cycle numbers using TEM (Figure 4). Notably, bare Si NPs were observed to fracture even after five cycles (Figure 4A), and the fracture became more significant after 50 cycles. The serious fracture in Si NPs might originate from spatially inhomogeneous43 volume expansion of Si and also indicate that simple nanostructuring may not address fracture problem completely. The fracture must result in unstable SEI formation and explains the observed low CEs. In the case of Si@C (Figure 4B), although a certain portion of the carbon coating layers appeared to remain intact after the first lithiation, cracks in coating layers were frequently found. After the first delithiation, a good number of pulverized particles were observed (white arrow), and the pulverization became more serious after 50 cycles, consistent with its cycling and CE data that started to show serious decaying in the cycle range of 25− 50. Hence, these TEM images confirm that the widely adopted amorphous carbon coating is not a sustainable solution. In the case of Si@SiOC (Figure 4C), however the SiOC coating was found to be very robust maintaining grapelike core−shell structures at all cycling points, including the first lithiation/ delithiation and 50th cycle. The robust structure of Si@SiOC is once again associated with the volume change capability of SiOC that allows it to follow the volume change of inner Si NPs without disintegration. Thus, the improved SEI stability and QEs are mainly from the volume change capability of SiOC that leads to pulverization-free volume expansion of inner Si NPs, not from physical properties of the SEI layers. In conclusion, we have introduced the SiOC glass coating via a scalable spray pyrolysis to maintain the structural integrity of

■

ASSOCIATED CONTENT

* Supporting Information S

Detailed spray pyrolysis setup, additional characterization and electrochemical data of SiOC, and Si@SiOC composite. This material is available free of charge via the Internet at http:// pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Tel: +82-42-350-1719. Fax: +82-42-350-2248. Author Contributions §

S.C. and D.S.J. contributed equally to this work.

Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS J.W.C. acknowledges the financial support by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MEST) (NRF-2010-C1AAA001-0029031, NRF-2012-R1A2A1A01011970, NRF-2014R1A4A1003712). This work was also supported partly by the framework of Research and Development Program of the Korea Institute of Energy Research (KIER) (B4-2424-05).

■

REFERENCES

(1) Wu, H.; Cui, Y. Nano Today 2012, 7, 414−429. (2) Kasavajjula, U.; Wang, C. S.; Appleby, A. J. J. Power Sources 2007, 163, 1003−1039. (3) Szczech, J. R.; Jin, S. Energy Environ. Sci. 2011, 4, 56−72. (4) Jia, H.; Gao, P.; Yang, J.; Wang, J.; Nuli, Y.; Yang, Z. Adv. Energy Mater. 2011, 1, 1036−1039. (5) Magasinski, A.; Dixon, P.; Hertzberg, B.; Kvit, A.; Ayala, J.; Yushin, G. Nat. Mater. 2010, 9, 353−358. (6) Zhao, Y.; Liu, X.; Li, H.; Zhai, T.; Zhou, H. Chem. Commun. 2012, 48, 5079−5081. (7) Ge, M.; Rong, J.; Fang, X.; Zhou, C. Nano Lett. 2012, 12, 2318− 2323. (8) Liu, N.; Lu, Z.; Zhao, J.; McDowell, M. T.; Lee, H.-W.; Zhao, W.; Cui, Y. Nat. Nanotechnol. 2014, 9, 187−192. (9) Bang, B. M.; Lee, J. I.; Kim, H.; Cho, J.; Park, S. Adv. Energy Mater. 2012, 2, 878−883. (10) Zhang, Z.; Wang, Y.; Ren, W.; Tan, Q.; Chen, Y.; Li, H.; Zhong, Z.; Su, F. Angew. Chem., Int. Ed. 2014, 53, 5165−5169. (11) Zhou, X.; Yin, Y.-X.; Wan, L.-J.; Guo, Y.-G. Chem. Commun. 2012, 48, 2198−2200. (12) Nguyen, B. P. N.; Kumar, N. A.; Gaubicher, J.; Duclairoir, F.; Brousse, T.; Crosnier, O.; Dubois, L.; Bidan, G.; Guyomard, D.; Lestriez, B. Adv. Energy Mater. 2013, 3, 1351−1357. (13) Liu, W.-R.; Wu, N.-L.; Shieh, D.-T.; Wu, H.-C.; Yang, M.-H.; Korepp, C.; Besenhard, J. O.; Winter, M. J. Electrochem. Soc. 2007, 154, A97−A102. (14) Chen, X.; Li, X.; Ding, F.; Xu, W.; Xiao, J.; Cao, Y.; Meduri, P.; Liu, J.; Graff, G. L.; Zhang, J.-G. Nano Lett. 2012, 12, 4124−4130. (15) Wu, H.; Chan, G.; Choi, J. W.; Ryu, I.; Yao, Y.; McDowell, M. T.; Lee, S. W.; Jackson, A.; Yang, Y.; Hu, L.; Cui, Y. Nat. Nanotechnol. 2012, 7, 310−315. 7124

dx.doi.org/10.1021/nl503620z | Nano Lett. 2014, 14, 7120−7125

Nano Letters

Letter

(16) Lee, J. K.; Smith, K. B.; Hayner, C. M.; Kung, H. H. Chem. Commun. 2010, 46, 2025−2027. (17) Yu, Y.; Gu, L.; Zhu, C.; Tsukimoto, S.; van Aken, P. A.; Maier, J. Adv. Mater. 2010, 22, 2247−2250. (18) Jung, D. S.; Ryou, M.-H.; Sung, Y. J.; Park, S. B.; Choi, J. W. Proc. Natl. Acad. Sci. U.S.A. 2013, 110, 12229−12234. (19) Yoo, J. K.; Kim, J.; Choi, M. J.; Park, Y. U.; Hong, J.; Baek, K. M.; Kang, K.; Jung, Y. S. Adv. Energy Mater. 2014, DOI: 10.1002/ aenm.201400622. (20) Shen, L.; Guo, X.; Fang, X.; Wang, Z.; Chen, L. J. Power Sources 2012, 213, 229−232. (21) Yao, H.; Zheng, G.; Li, W.; McDowell, M. T.; Seh, Z.; Liu, N.; Lu, Z.; Cui, Y. Nano Lett. 2013, 13, 3385−3390. (22) Chen, X.; Gerasopoulos, K.; Guo, J.; Brown, A.; Wang, C.; Ghodssi, R.; Culver, J. N. ACS Nano 2010, 4, 5366−5372. (23) Park, C.-M.; Choi, W.; Hwa, Y.; Kim, J.-H.; Jeong, G.; Sohn, H.J. J. Mater. Chem. 2010, 20, 4854−4860. (24) Yang, J.; Takeda, Y.; Imanishi, N.; Capiglia, C.; Xie, J.; Yamamoto, O. Solid State Ionics 2002, 152, 125−129. (25) Miyachi, M.; Yamamoto, H.; Kawai, H.; Ohta, T.; Shirakata, M. J. Electrochem. Soc. 2005, 152, A2089−A2091. (26) Pantano, C. G.; Singh, A. K.; Zhang, H. J. Sol-Gel Sci. Technol. 1999, 14, 7−25. (27) Liu, X.; Zheng, M.-C.; Xie, K. J. Power Sources 2011, 196, 10667−10672. (28) Momma, T.; Aoki, S.; Nara, H.; Yokoshima, T.; Osaka, T. Electrochem. Commun. 2011, 13, 969−972. (29) Wang, W.-N.; Purwanto, A.; Lenggoro, I. W.; Okuyama, K.; Chang, H.; Jang, H. D. Ind. Eng. Chem. Res. 2008, 47, 1650−1659. (30) Hou, X.; Choy, K. L. Chem. Vap. Deposition 2006, 12, 583−596. (31) Gurav, A.; Kodas, T.; Pluym, T.; Xiong, Y. Aerosol Sci. Technol. 1993, 19, 411−452. (32) Sorarù, G. D.; Dallapiccola, E.; D’Andrea, G. J. Am. Ceram. Soc. 1996, 79, 2074−2080. (33) Singh, A. K.; Pantano, C. G. J. Am. Ceram. Soc. 1996, 79, 2696− 2704. (34) Kolár,̌ F.; Machovič, V.; Svıtilová, J.; Borecka, L. Mater. Chem. Phys. 2004, 86, 88−98. (35) Chang, W.-S.; Park, C.-M.; Kim, J.-H.; Kim, Y.-U.; Jeong, G.; Sohn, H.-J. Energy Environ. Sci. 2012, 5, 6895−6899. (36) Takezawa, H.; Iwamoto, K.; Ito, S.; Yoshizawa, H. J. Power Sources 2013, 244, 149−157. (37) Chou, C.-Y.; Hwang, G. S. Chem. Mater. 2013, 25, 3435−3440. (38) McDowell, M. T.; Lee, S. W.; Ryu, I.; Wu, H.; Nix, W. D.; Choi, J. W.; Cui, Y. Nano Lett. 2011, 11, 4018−4025. (39) Sim, S.; Oh, P.; Park, S.; Cho, J. Adv. Mater. 2013, 25, 4498− 4503. (40) Wu, H.; Yu, G.; Pan, L.; Liu, N.; McDowell, M. T.; Bao, Z.; Cui, Y. Nat. Commun. 2013, 4, 1943. (41) Yao, Y.; Liu, N.; McDowell, M. T.; Pasta, M.; Cui, Y. Energy Environ. Sci. 2012, 5, 7927−7930. (42) Liu, B.; Soares, P.; Checkles, C.; Zhao, Y.; Yu, G. Nano Lett. 2013, 13, 3414−3419. (43) Gu, M.; Li, Y.; Li, X.; Hu, S.; Zhang, X.; Xu, W.; Thevuthasan, S.; Baer, D. R.; Zhang, J.-G.; Liu, J. ACS Nano 2012, 6, 8439−8447.

7125

dx.doi.org/10.1021/nl503620z | Nano Lett. 2014, 14, 7120−7125