Letter Cite This: Nano Lett. 2017, 17, 7989−7994
pubs.acs.org/NanoLett
Electrospun Thin-Walled CuCo2O4@C Nanotubes as Bifunctional Oxygen Electrocatalysts for Rechargeable Zn−Air Batteries Xiaojun Wang,† Yang Li,† Ting Jin,† Jing Meng,† Lifang Jiao,*,†,‡ Min Zhu,§ and Jun Chen†,‡ †
Key Laboratory of Advanced Energy Materials Chemistry (Ministry of Education), College of Chemistry, Nankai University, Tianjin 300071, China ‡ Collaborative Innovation Center of Chemical Science and Engineering (Tianjin), Nankai University, Tianjin 300071, China § School of Materials Science and Engineering and Guangdong Provincial Key Laboratory of Advanced Energy Storage Materials, South China University of Technology, Guangzhou 510641, China S Supporting Information *
ABSTRACT: Rational design of optimal bifunctional oxygen electrocatalyst with low cost and high activity is greatly desired for realization of rechargeable Zn−air batteries. Herein, we fabricate mesoporous thin-walled CuCo2O4@C with abundant nitrogen-doped nanotubes via coaxial electrospinning technique. Benefiting from high catalytic activity of ultrasmall CuCo2O4 particles, double active specific surface area of mesoporous nanotubes, and strong coupling with N-doped carbon matrix, the obtained CuCo2O4@C exhibits outstanding oxygen electrocatalytic activity and stability, in terms of a positive onset potential (0.951 V) for oxygen reduction reaction (ORR) and a low overpotential (327 mV at 10 mA cm−2) for oxygen evolution reaction (OER). Significantly, when used as cathode catalyst for Zn-air batteries, CuCo2O4@C also displays a low charge−discharge voltage gap (0.79 V at 10 mA cm−2) and a long cycling life (up to 160 cycles for 80 h). With desirable architecture and excellent electrocatalytic properties, the CuCo2O4@C is considered a promising electrocatalyst for Zn−air batteries. KEYWORDS: Thin-walled CuCo2O4@C nanotubes, coaxial electrospinning, bifunctional oxygen electrocatalyst, Zn−air batteries
I
substrate.15,16 For example, CuCo2O4 nanoparticles supported on reduced graphene oxide exhibited a decrease of 14% in current density over 20 000 s of continuous operation for ORR.13 Although a carbon matrix can certainly improve the conductivity, the low availability of large specific surface area is still a weakness for further improving electrocatalytic properties owing to the poor contact areas between the active materials and the electrolyte. Attempts have been made to expose more active sites to increase the utilization rate of catalytic material.17−20 Recently, Cheng and colleagues successfully prepared spinel CuCo2O4 quantum dots anchored on nitrogendoped carbon nanotubes; the enhanced conductivity and the enlarged specific surface areas (116.18 m2 g−1) of catalysts could contribute to a long lifetime (48 h in liquid state) for a Zn−air battery.21 Even though a lot of inspiring works have made great progress in cycling stability for Zn−air batteries, the catalytic activity is still far away from meeting the application of Zn−air batteries because the structure of narrow oxygen/ hydroxyl transportation channels for catalysts result in sluggish diffusion kinetics. Thus, considerable efforts in the rational design of advanced architecture with broadly diffused pathways are still desired.
ncreasing energy demand inspires the development of alternative energy storage and conversion systems with high energy density.1,2 Currently, rechargeable zinc−air (Zn− air) batteries have received worldwide attention due to their high theoretical energy density (1086 Wh kg−1), abundant resource, and environmental benignity.3,4 The state-of-the-art electro-catalysts for Zn-air batteries remain noble metal-based materials (such as Pt, Ir, and their oxides), which can efficiently accelerate oxygen reduction/evolution reactions (ORR/OER). However, the noble metal-based materials still face these issues such as high cost, limited cycling life and catalytic selectivity (Pt for ORR and Ir for OER).5−7 Therefore, exploring optimal bifunctional electrocatalysts for both ORR and OER with economic viability, efficient activity, and superior durability to the realization of Zn−air batteries is still a challenge. Mixed-valence transition-metal oxides with spinel structure have sparked more attention as bifunctional electro-catalyst because of their low price and high catalytic efficiency.8−12 Among them, CuCo2O4 displays the advantage of high catalytic activity by the virtue of their tailored electron between Cu and Co ions. Importantly, the structural flexibility and variable valence states of the spinel oxides provide great potential to fine-tune their catalytic performance.13,14 However, the main issue with bimetallic oxide materials is their low conductivity, resulting in inferior cycling stability. One effective approach is making the bimetallic oxides with a coupling carbon © 2017 American Chemical Society
Received: October 23, 2017 Published: November 22, 2017 7989
DOI: 10.1021/acs.nanolett.7b04502 Nano Lett. 2017, 17, 7989−7994
Letter
Nano Letters Thin-walled porous-nanotubes with open texture could not only provide large specific areas for catalyst reactions, but offer more channels for oxygen/hydroxyl diffusion.16,22 Compared to the hydrothermal method,23 in situ growth approach,24 and nucleation crystallization strategy,25 electrospinning is a facile technique to directly construct the architecture of thin-walled nanotubes with mesoporous surfaces. Peng’s group successfully prepared Fe-based nitrogen/oxygen codoped carbon nanotubes grown on carbon-nanofiber films (FeNO-CNT-CNFFs) via a facile free-surface electrospinning technique, exhibiting the highest bifunctional oxygen catalytic activities in terms of a positive half-wave potential (0.87 V) for ORR and low overpotential (430 mV@10 mA cm−2) for OER.26 Mai’s group also synthesized CoMn2O4 multilevel nanotubes via the electrospinning method, showing high stability with about 92% current retention after 30 000 s for ORR.27 Significantly, utilizing the different pyrolysis rates of different molecular polymer to synthesize thin-walled mesopores nanotubes with more gas channels that cover CuCo2O4 nanoparticles has not yet been reported. It is of great interest to apply electrospinning to design such nanostructured CuCo2O4@C as the bifunctional electrocatalyst for Zn−air batteries. Herein, we report on the fabrication of N-doped thin-walled CuCo2O4@C nanotubes with mesoporous surface via coaxial electrospinning technique and their further application as bifunctional oxygen electrocatalysts for Zn−air batteries. The as-prepared CuCo2O4@C with a specific surface area of 514 m2 g−1 exhibits a positive half-wave potential of 0.850 V for ORR, and a low overpotential of 327 mV at 10 mA cm−2 for OER. Meaningfully, the as-built Zn−air batteries display a low discharge−charge voltage gap (0.79 V at 10 mA cm−2) and long cycle life (up to 160 cycles for 80 h). The excellent electrocatalytic properties are ascribed to the advanced structure of CuCo2O4@C, which provides a variety of superiorities: numerous mesoporous ensuring more channels for oxygen escape and hydroxyl diffusion, double active specific surface area offering more exposed active site for catalytic reaction, interconnected 1D open-ended nanotubes leading to superior electronic contact for external circuit, and doping N also contributing to oxygen redox catalysis. Scheme 1 illustrates the synthesis procedure of thin-walled CuCo2O4@C (denoted as CCO@C) nanotubes via the coaxial electrospinning technique. It mainly includes an electrospinning fiber membrane, followed by stabilization, carbonization, and oxidation treatment of the obtained film. It is worth mentioning that, after the electrospinning process, low- and high-molecular-weight polyacrylonitrile (PAN) is prone to be distributed into two layers under the coaxial action in the radial direction of nanowires. With the temperature gradually increasing, the inner low-weight PAN first undergoes pyrolysis and moves to the boundary between the low- and high- weight PAN, thus creating the open-ended hollow nanotubes. Significantly, under high-temperature calcining in argon and then low-temperature sintering in air, the PAN and metal salts decompose to generate gases, so the uniform mesoporous on surface of tubes are obtained. Figure 1a presents the Rietveld refined X-ray diffraction (XRD) pattern of CCO@C composite. All diffraction peaks match well with cubic spinel-type CuCo2O4 (space group: Fd3m ̅ , JCPDS card No. 78-2177), and refinement results show that the unit cell structure consists of CuO4 tetrahedra and CoO6 octahedra (Figure 1b). The carbon existing in the CCO@C composites was demonstrated by Raman spectrum,
Scheme 1. Schematic Illustration of the Preparation Process for CuCo2O4@C Nanotubesa
a
(1) After coaxial electrospinning process, the low- and highmolecular-weight PAN tend to be distributed into two layers. (2) As the temperature is slowly increased, the inner low-weight PAN first undergoes pyrolysis and then moves to the boundary of the low-/highmolecular-weight PAN. (3) As the temperature continues to rise, the PAN carbonized and metal salts decomposed, thus creating more pores on the tube surface. (4) After sintering in air, CuCo2O4 nanoparticles supported on thin-walled nanotubes with uniform mesoporous surfaces are obtained.
just as shown in Figure S1 (Supporting Information, SI). The two prominent peaks at 1349 and 1591 cm−1 belong to the defect-induced mode (D bands) and graphitic-induced mode (G bands) of carbon, respectively.28 The weak peaks around 453 and 651 cm−1 are assigned to CuCo2O4. Element analysis (EA) was applied to further determine the carbon content of composite, which is approximately 19.64 wt %. This value is consistent with the thermogravimetric analysis (TGA, Figure S9, SI) result. The morphology and nanostructure of CCO@C were characterized via scanning electron microscopy (SEM) and transmission electron microscopy (TEM) technologies. As shown in Figure 1c,d, the CCO@C nanotubes are continuous and rough with abundant pores on the tube surface, which could improve the diffusion kinetics of hydroxide and favor the transportation of oxygen.29 The TEM image (Figure 1e) shows that all of the CuCo2O4 nanoparticles are surrounded by amorphous carbon. It is worth emphasizing that the nanotube outer diameter is around 130 nm, and the thickness of the wall is about 25 nm. The large interior channels (diameter of ∼80 nm) effectively double the active surface area of 514 m2 g−1 (Figure S2, SI), thus improving the reaction activity between hydroxyl, oxygen, and catalyst. Importantly, the nanoparticle size of CuCo2O4 in CCO@C is around 7 nm (Figure S3, SI). As far as we know, the composite that simultaneously controls the smallest CuCo2O4 nanodots and ensures such large internal channels has never been reported. The clear lattice fringes with d-spacings of 0.47 and 0.28 nm correspond to the (111) and (220) planes of CuCo2O4 with an angle of 35° between this two planes (Figure 1f), which is consistent with the standard value. TEM elemental mapping images (Figure 2g) reveal that Cu, Co, O, C and N are uniformly distributed along the hollow carbon matrix. EA and X-ray photoelectron spectroscopy 7990
DOI: 10.1021/acs.nanolett.7b04502 Nano Lett. 2017, 17, 7989−7994
Letter
Nano Letters
Figure 1. (a) Rietveld refined XRD pattern with experimental data (black line), calculated curves (orange dots), position of allowed Bragg reflections (purple vertical bars), and difference profile (red line). (b) Crystal structure with CuO4 tetrahedra (green) and CoO6 octahedra (blue). (c, d) SEM and (e,f) TEM images of CCO@C. (g) TEM elemental mapping images of CCO@C.
Figure 2. (a) ORR polarization curves of L-CCO@C, CCO@C, H-CCO, and Pt/C samples. (b) ORR polarization curves of the CCO@C nanotubes at different rotation speeds and the corresponding K−L plots at different potentials (inset). (c) ORR polarization curves of CCO@C and Pt/C samples before and after 3000 potential cycles. (d) OER polarization curves of L-CCO@C, CCO@C, H-CCO, IrO2, and the overpotential schematic at 10 mA cm−2 (inset). (e) Tafel plots derived from (d). (f) OER polarization curves of CCO@C and IrO2 during the cycling durability test.
(XPS) also illustrate that CuCo2O4@C contains 5.67 wt % N species (Figure S4 and Figure S9, SI). Significantly, different reaction conditions, mainly including diverse carbonization temperature and diverse oxidation temperature, were attempted to optimize the morphology
and catalytic properties. Figure S5 (SI) presents the SEM images of Cu(NO3)2·3H2O/Co(CH3COO)2·4H2O@PAN (denoted as CuCo@PAN) nanofibers. When the composite carbonizes at different temperature, the nanofibers display diverse architecture (Figure S6 (SI)). We can see that a lower 7991
DOI: 10.1021/acs.nanolett.7b04502 Nano Lett. 2017, 17, 7989−7994
Letter
Nano Letters
Figure 3. (a) The scheme of the as-assembled two-electrode rechargeable Zn−air batteries. (b) Charge and discharge polarization curves of LCCO@C, CCO@C, H-CCO, and Pt/C+IrO2 samples. (c) Galvanostatic discharge−charge cycling curves at 2 mA cm−2 using CCO@C and Pt/C +IrO2 as air electrodes, and the optical images of an LED illuminated using CCO@C as air electrodes (inset). (d) Schematic illustration of the advantages of the CCO@C nanotubes as electrocatalysts.
temperature (700 °C) leads to a smooth surface, while a higher temperature (900 °C) results in a rough surface with a certain degree of aggregation for nanoparticles. The possible mechanism that causes such difference mainly resulted from the pyrolysis degree of PAN, the decomposition of metal salts (Cu(NO3)2·3H2O/Co(CH3COO)2·4H2O@PAN). More importantly, compared to the lower carbonization temperature (700 °C), the external wall is more likely to collapse at the higher carbonization temperature (900 °C) because the inner hollow structure has been preliminarily formed. The morphology carbonized under 800 °C, by contrast, is much more suitable for the next oxidation procedure owing to the ideal porousness. For comparison, we adjusted the carbon content by altering the oxidation temperature (higher and lower oxidation temperature-obtained productions named H-CCO and LCCO@C, respectively). The XRD patterns and Raman spectra of L-CCO@C and H-CCO are displayed in Figure S7 (SI), along with SEM and TEM images of these samples (Figure S8, SI). L-CCO@C exists in a reticular morphology with indistinct hollow structure; it mainly results from the lower oxidation temperature not able to removal the carbon completely. However, the H-CCO tubes are stacked together with a narrow vacancy in the middle of the nanotubes because the higher temperature induces some shrinkage of nanotubes.30,31 Considering the uniform nanotube morphology, appropriate carbon content (Figure S9, SI) and optimal specific surface areas (Figure S10, SI), 350 °C should be the suitable oxidation temperature. The ORR activities of CuCo2O4 samples were measured and compared using a rotating ring-disk electrode (RRDE) in O2saturated 0.1 M KOH solution (Figure 2a). Remarkably, the CCO@C shows a much more positive onset potential (0.951 V) than those of the L-CCO@C (0.927 V) and H-CCO (0.888 V). Significantly, the ORR properties of CCO@C nanotubes
are also close to that of Pt/C (onset potential: 0.961 V, halfwave potential: 0.862 V). The electron transfer number per oxygen molecule (n) of CCO@C composite was determined from linear sweep voltammetry (LSV) curves (Figure 2b), n is calculated to be about 3.9, implying that CCO@C composite is able to straightly reduce the H2O to OH− by an efficient four electron reaction. This results mainly because the advanced morphology of thin-walled CCO@C nanotubes with mesoporous surface could offer more active sites and more transport channels for gas transportation. Although the HO2− yield (6.3%) slightly larger than that of Pt/C (3.55%) electrode (Figure S11, SI), this value is still better than many of the nonnoble metal-based catalysts that have been reported.32 The electrochemical durability in terms of ORR is presented in Figure 2c, where the Pt/C electrode is observed to degrade more severely after 3000 repeated cycles with larger positive potential shift, while the CCO@C electrode is relatively stable even after 3000 CV cycles. During the methanol resistance test (Figure S12, SI), interestingly, the Pt/C catalyst shows obviously degraded current density after addition of methanol; however, CCO@C still exhibits a stable current density without distinct recession, indicating the strong methanol resistance and stability of CCO@C. Figure 2d exhibits the OER polarization curves for L-CCO@ C, CCO@C, H-CCO, and commercial IrO2 catalysts. The CCO@C nanotubes show a lower overpotential (327 mV) compared with L-CCO@C (461 mV) and H-CCO (492 mV) at a current density of 10 mA cm−2. The catalytic kinetics for OER are also evaluated by Tafel slope. As revealed in Figure 2e, the Tafel slope of CCO@C nanotubes is about 74.0 mV dec−1, which is smaller than those of L-CCO@C (91.3 mV dec−1) and H-CCO (105.9 mV dec−1), indicating the outstanding OER kinetics of CCO@C samples. Additionally, the overpotential and Tafel slope of CCO@C nanotubes were comparable to 7992
DOI: 10.1021/acs.nanolett.7b04502 Nano Lett. 2017, 17, 7989−7994
Letter
Nano Letters that of commercial IrO2 (316 mV, 58.1 mV dec−1). To further explore the reasons of outstanding OER performance of CCO@C, we have measured the electrochemically active surface area (ECSA) from the electrochemical double-layer capacitance (Cdl). As shown in Figure S13 (SI), the ECSAs of CCO@C, L-CCO@C, and H-CCO were estimated to be 313.8 cm2, 80.0 cm2, and 61.3 cm2, respectively, which means that CCO@C has the largest electrochemically active surface among the three catalysts. The increased ECSA for CCO@C can render a large specific surface area for its catalytically active sites and excellent gas bubble diffusion ability, thus leading to the superior catalytic performance. The excellent OER kinetics of CCO@C composite was also demonstrated by an electrochemical impedance spectroscopy (EIS) test. As shown in Figure S14 (SI), the semicircle at high frequency corresponds to the formation of an electrical double layer, and the semicircle at low frequency can be ascribed to the Faradaic reaction of oxygen evolution. The CCO@C presents superior performance, which mainly because of the large internal hollow tubular structure and abundant pores, would facilitate the charge transfer kinetics. The stability of CCO@C and noble-metal IrO2 catalyst was assessed through 3000 continuous potential cycles test at 1.45−1.65 V. The polarization curves of CCO@C nanotubes slightly decline. By contrast, the stability of IrO2 is much worse than CCO@C (Figure 2f). As a proof-of-concept application, a primary Zn-air battery is assembled to further evaluate its performance under real battery conditions (Figure 3a). Figure 3b shows typical discharge− charge polarization curves. The open circuit voltage (OCV) of Pt/C-IrO2 is 1.45 V, which is slightly higher than that for CCO@C (1.41 V), whereas the L-CCO@C presents a lower OCV of 1.37 V owing to the narrow channels for oxygen/ hydroxyl diffusion. The H-CCO also exhibits inferior OCV (1.35 V), mainly because the low amount of carbon contained in the composite leads to poor conductivity. The CCO@C electrode also exhibits a lower discharge−charge voltage gap of 0.79 V at 10 mA cm−2. Significantly, when CCO@C repeatedly charge−discharges at 2 mA cm−2 for a total of 160 cycles with a 30 min per cycle period, only a little charge/discharge voltage change is observed, which is superior to L-CCO@C, H-CCO, and Pt/C-IrO2 as cathodes for Zn−air batteries (Figure 3c and Figure S15 (SI)). The cycling stability of CCO@C nanotubes at higher current density was also remarkable (10 mA cm−2, Figure S16); the discharge voltage still remain stable up to 60 cycles. One important proof-of-concept illustration is to power two light-emitting diodes (LEDs) by series-connected Zn−air batteries based on a CCO@C air cathode, and the LED could still emit constantly and steadily (Figure 3c (inset) and Video S1 (SI)). On the basis of the advanced structural and outstanding electrocatalytic properties of CCO@C nanotubes (Figure 3d and Tables S1, S2, S3 (SI)), we think that four critical aspects are responsible for their excellent ORR and OER activity and stability for Zn−air batteries: (1) The large interior hollows (diameter of ∼80 nm) and a mass of mesoporous among carbon nanotubes could double the active surface area, which would effectively offer more channels for oxygen/hydroxyl diffusion and more active sites for catalytic reaction. (2) The ultrasmall CuCo2O4 nanoparticles (∼7 nm) could improve the utilization rate, and thus increase the catalytic activity. (3) The structure of uniform 1D nanotubes interconnected into 3D conductive network would provide a superior electronic contact to the external circuit. (4) The N-
doping, (especially for pyridinic-N and graphitic-N) may also contribute to oxygen redox catalysis. Combining advantages of bimetallic oxides and carbon matrix, we synthesize the mesoporous thin-walled CuCo2O4@ C nanotubes with abundant nitrogen element contained by a facile coaxial electrospinning method. Rational design of surface structure through multiporous auxiliary and strong coupling with N-doped carbon substrate could lead to a high activity and a strong durability for bifunctional catalysis toward ORR (positive onset potential: 0.951 V) and OER (low overpotential: 327 mV at 10 mA cm−2). When used as a cathode for Zn−air batteries, CCO@C electrode exhibits a low discharge− charge voltage gap of 0.79 V at 10 mA cm−2 and a long cycle life up to 160 cycles for 80 h at a current density of 2 mA cm−2. Such promising preparation method as well as remarkable electrocatalytic properties would provide a certain reference to the relevant materials.
■
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.nanolett.7b04502. Experimental Section, additional figures (Figure S1− S16) and tables (Table S1−S3) (PDF) Video demonstrating the powering of two LEDs by series-connected Zn−air batteries based on a CCO@C air cathode (AVI)
■
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. ORCID
Lifang Jiao: 0000-0002-4676-997X Min Zhu: 0000-0001-5018-2525 Jun Chen: 0000-0001-8604-9689 Author Contributions
The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This work was financially supported by the National Natural Science Foundation of China (51622102, 51231003, 51571124), MOST(2016YFB0901502), and the 111 Project (B12015).
■
REFERENCES
(1) Nam, G.; Park, J.; Kim, S. T.; Shin, D. B.; Park, N.; Kim, Y.; Lee, J. S.; Cho, J. Nano Lett. 2014, 14, 1870−1876. (2) Li, C.; Han, X.; Cheng, F.; Hu, Y.; Chen, C.; Chen, J. Nat. Commun. 2015, 6, 7345. (3) Jia, Y.; Zhang, L.; Du, A.; Gao, G.; Chen, J.; Yan, X.; Brown, C. L.; Yao, X. Adv. Mater. 2016, 28, 9532−9538. (4) Li, B.; Quan, J.; Loh, A.; Chai, J.; Chen, Y.; Tan, C.; Ge, X.; Hor, T. S.; Liu, Z.; Zhang, H.; Zong, Y. Nano Lett. 2017, 17, 156−163. (5) Cheng, F.; Chen, J. Chem. Soc. Rev. 2012, 41, 2172−2192. (6) Liu, Z.; Zhao, Z.; Wang, Y.; Dou, S.; Yan, D.; Liu, D.; Xia, Z.; Wang, S. Adv. Mater. 2017, 29, 1606207. (7) Li, Y.; Lu, J. ACS Energy Letters. 2017, 2, 1370−1377.
7993
DOI: 10.1021/acs.nanolett.7b04502 Nano Lett. 2017, 17, 7989−7994
Letter
Nano Letters (8) Su, C.-Y.; Cheng, H.; Li, W.; Liu, Z.-Q.; Li, N.; Hou, Z.; Bai, F.Q.; Zhang, H.-X.; Ma, T.-Y. Adv. Energy Mater. 2017, 7, 1602420. (9) Han, X.; Cheng, F.; Chen, C.; Li, F.; Chen, J. Inorg. Chem. Front. 2016, 3, 866−871. (10) Fu, G.; Chen, Y.; Cui, Z.; Li, Y.; Zhou, W.; Xin, S.; Tang, Y.; Goodenough, J. B. Nano Lett. 2016, 16, 6516−6522. (11) Yin, J.; Li, Y.; Lv, F.; Fan, Q.; Zhao, Y. Q.; Zhang, Q.; Wang, W.; Cheng, F.; Xi, P.; Guo, S. ACS Nano 2017, 11, 2275−2283. (12) Zhao, Q.; Yan, Z.; Chen, C.; Chen, J. Chem. Rev. 2017, 117, 10121−10211. (13) Ning, R.; Tian, J.; Asiri, A. M.; Qusti, A. H.; Al-Youbi, A. O.; Sun, X. Langmuir 2013, 29, 13146−13151. (14) Cheng, F.; Shen, J.; Peng, B.; Pan, Y.; Tao, Z.; Chen, J. Nat. Chem. 2011, 3, 79−84. (15) Li, L.; Peng, S.; Lee, J. K. Y.; Ji, D.; Srinivasan, M.; Ramakrishna, S. Nano Energy 2017, 39, 111−139. (16) Zheng, G.; Yang, Y.; Cha, J. J.; Hong, S. S.; Cui, Y. Nano Lett. 2011, 11, 4462−4467. (17) Liang, Y.; Wang, H.; Zhou, J.; Li, Y.; Wang, J.; Regier, T.; Dai, H. J. Am. Chem. Soc. 2012, 134, 3517−3523. (18) Wu, H.; Geng, J.; Ge, H.; Guo, Z.; Wang, Y.; Zheng, G. Adv. Energy Mater. 2016, 6, 1600794. (19) Liu, Q.; Wang, Y.; Dai, L.; Yao, J. Adv. Mater. 2016, 28, 3000− 3006. (20) Liu, J.; Yu, L.; Wu, C.; Wen, Y.; Yin, K.; Chiang, F. K.; Hu, R.; Liu, J.; Sun, L.; Gu, L.; Maier, J.; Yu, Y.; Zhu, M. Nano Lett. 2017, 17, 2034−2042. (21) Cheng, H.; Li, M.-L.; Su, C.-Y.; Li, N.; Liu, Z.-Q. Adv. Funct. Mater. 2017, 1701833. (22) Chen, S.; Duan, J.; Bian, P.; Tang, Y.; Zheng, R.; Qiao, S.-Z. Adv. Energy Mater. 2015, 5, 1500936. (23) Liu, X.; Park, M.; Kim, M. G.; Gupta, S.; Wang, X.; Wu, G.; Cho, J. Nano Energy 2016, 20, 315−325. (24) Wang, M.; Qian, T.; Liu, S.; Zhou, J.; Yan, C. ACS Appl. Mater. Interfaces 2017, 9, 21216−21224. (25) Han, X.; Wu, X.; Zhong, C.; Deng, Y.; Zhao, N.; Hu, W. Nano Energy 2017, 31, 541−550. (26) Niu, C.; Meng, J.; Wang, X.; Han, C.; Yan, M.; Zhao, K.; Xu, X.; Ren, W.; Zhao, Y.; Xu, L.; Zhang, Q.; Zhao, D.; Mai, L. Nat. Commun. 2015, 6, 7402. (27) Meng, J.; Niu, C.; Liu, X.; Liu, Z.; Chen, H.; Wang, X.; Li, J.; Chen, W.; Guo, X.; Mai, L. Nano Res. 2016, 9, 2445−2457. (28) Meng, F.; Zhong, H.; Bao, D.; Yan, J.; Zhang, X. J. Am. Chem. Soc. 2016, 138, 10226−10231. (29) Li, Q.; Cao, R.; Cho, J.; Wu, G. Adv. Energy Mater. 2014, 4, 1301415. (30) Zhu, Y.; Fan, X.; Suo, L.; Luo, C.; Gao, T.; Wang, C. ACS Nano 2016, 10, 1529−1538. (31) Jayaraman, S.; Aravindan, V.; Suresh Kumar, P.; Ling, W. C.; Ramakrishna, S.; Madhavi, S. Chem. Commun. (Cambridge, U. K.) 2013, 49, 6677−6679. (32) Yan, D.; Li, Y.; Huo, J.; Chen, R.; Dai, L.; Wang, S. Adv. Mater. 2017, 1606459.
7994
DOI: 10.1021/acs.nanolett.7b04502 Nano Lett. 2017, 17, 7989−7994
