Chinese Chemical Letters  2018, Vol. 29 Issue (12): 1777-1780   PDF    
High performance lithium-sulfur batteries by facilely coating a conductive carbon nanotube or graphene layer
Yuchi Yanga, Chen Chena, Jianhua Hua, Yonghui Dengc,d, Yi Zhangb,*, Dong Yanga,*     
a State Key Laboratory of Molecular Engineering of Polymers, Department of Macromolecular Science, Fudan University, Shanghai 200433, China;
b School of Materials Science and Energy Engineering, Foshan University, Foshan 528000, China;
c Department of Chemistry, State Key Laboratory of Molecular Engineering of Polymers, Shanghai Key Laboratory of Molecular Catalysis and Innovative Materials, iChEM, Fudan University, Shanghai 200433, China;
d State Key Lab of Transducer Technology, Shanghai Institute of Microsystem and Information Technology, Chinese Academy of Sciences, Shanghai 200050, China
Abstract: Lithium sulfur (Li-S) batteries are regarded as promising candidates for next-generation rechargeable batteries. However, the insulation characteristic of sulfur and severe polysulfide dissolution hindered their development. We presented a facile approach to fabricate Li-S batteries by coating commercial carbon nanotube or graphene slurries on normal sulfur cathode electrode to construct a dual-layer cathode electrode. The conductive CNT or graphene layer could not only improve the conductivity of sulfur cathode, but also suppress the polysulfide diffusion. The CNT@S cathode delivered a high reversible capacity of 740 mAh/g over 300 cycles at 1 C and 870 mAh/g over 100 cycles at 0.2 C. Furthermore, this strategy could be realized on the commercial product line of lithium-ion batteries, which made it possible to large-scale produce Li-S batteries.
Keywords: Lithium sulfur batteries     Conductive coatings     Carbon nanotube     Graphene     Electrochemical performance    

As a promising candidate for next-generation rechargeable battery, lithium sulfur (Li-S) battery has attracted increasing attention [1, 2]. Sulfur possesses many outstanding advantages, such as low cost, natural abundance, nontoxicity, and environmental friendliness. Furthermore, sulfur has a high theoretical capacity of 1675 mAh/g and theoretical specific energy of 2600 Wh/kg [3, 4]. However, the commercialization of Li-S batteries was hindered by the insulation characteristic of sulfur and Li2S, and the shuttle effect of soluble polysulfide intermediates [5-10].

Tremendous efforts have been devoted to resolve these obstacles. A common approach was to mix sulfur with conductive carbon-based materials, such as CNT and graphene, to enhance the conductivity of the S cathode [1, 3, 11]. For example, Nazar et al. have prepared reduced graphene oxides and sulfur hybrid composites in one step, which showed an initial discharge capacity of 705 mAh/g at 0.2 C [12]. Xiao et al. have coupled S with etched MWCNTs as cathode to give a high initial capacity of 1382 mAh/g at 0.2 C [13]. Although a simple mixture of sulfur with conductive carbon could effectively improve the initial capacities due to the enhanced conductivity, the polysulfide intermediates still easily diffused into the electrolyte and migrated to the anode side, which would result in a shuttle reaction and loss of the active materials [11]. Thus, porous and hollow carbon materials were developed to host S, which could simultaneously enhance the conductivity of S cathode and compress the diffusion of polysulfide intermediates by the chemical and physical absorption [14-17]. Sun et al. have rationally designed hollow core-shell interlinked carbon spheres for advanced Li-S batteries, which exhibited a high initial capacity of 1100 mAh/g at 0.5 C and excellent cycle stability of 960 mAh/g after 200 cycles at 0.5 C [18]. However, it was time and cost consuming to fabricate the complex carbon hosts. Furthermore, the loading amount of sulfur was seriously confined, mostly below than 70 wt% [19-21]. To maintain the structure of the porous or hollow carbon hosts, the compacted density of the cathode plate was also greatly limited. All these factors were unfavorable for large-scale production of Li-S batteries. Therefore, it was still a challenging work to explore a facile and economical strategy to large-scale produce Li-S batteries with a high initial capacity and good cycling performance.

Herein, we reported a facile and scalable strategy to fabricate sulfur cathode plate. This strategy could be easily operated on the commercial product line of lithium-ion battery. Firstly, a mixed slurry of sulfur powder, carbon black (super-P) and binder (PVDF), was coated on an aluminum foil. Then, CNT or graphene slurry was coated on the sulfur layer to form a dual-layer structure (CNT@S or graphene@S). The smooth and compact CNT or graphene layers could efficiently enhance the conductivity and trap the polysulfide by physical and chemical adsorption [22]. Furthermore, by coating a protective CNT layer on the Al foil current collector, the capacity of Li-S batteries could be further improved. This scalable and facile strategy made it possible to commercialize Li-S batteries.

The morphologies of various cathode electrodes were characterized by SEM. As shown in Fig. 1a, the surface of the normal S cathode electrode was rough, and the S powder was irregularly surrounded with super-P particles. After coating CNT or graphene, the cathode electrodes presented a dense and smooth surface (Figs. 1b and c). The graphene or CNT layer could prevent the direct exposure of the S cathode to the electrolyte. The cross-section SEM images of the CNT@S and graphene@S electrodes were shown in Figs. 1d and e. The average thicknesses of the CNT and graphene layers was 7 μm and 4 μm, respectively. Both CNT and graphene layers were tightly adhered on the S layer.

Download:
Fig. 1. SEM images of (a) normal S cathode, (b) graphene@S, (c) CNT@S, (d) crosssection CNT@S and (e) cross-section graphene@S.

The CNT layer was much thicker than the graphene layer at a similar weight ratio of about 0.6 mg/cm2. It was because the CNT formed a 3D interconnected framework, while the graphene formed a compactly layer-by-layer packed structure. Thus, the CNT layer could accommodate more redundant space, and was more profitable for the electrolyte infiltration and lithium-ion diffusion [18]. The corresponding energy-dispersive X-ray (EDS) mapping was shown in Figs. S1a and b (Supporting information), which also demonstrated that both of the CNT and graphene layers were tightly adhered to the S layer as a protective layer.

To evaluate the effects of the conductive layers on the sulfur cathode electrode, the electrochemical performance of the electrode was investigated by two-electro half-cells and all the measurements were tested at the room temperature. A cyclic voltammetry (CV) of the cathode was tested at a scan rate of 0.1 mV/s from 1.7 V to 2.7 V. Fig. 2a revealed that the CNT@S cathode featured two reduction peaks and two oxidation peaks, representing the redox reactions of the active materials. In the first cycle, the first reduction peak at 2.30 V was attributed to the transformation from S8 to Li2Sx (4 ≤ x ≤ 8), and the second one at 2.02 V was ascribed to the process of liquid-solid phase transitions (reduction from Li2S4 to Li2S2). The two oxidation peaks at 2.36 V and 2.43 V corresponded to the formation of Li2S6 and S8, respectively. The results were consistent with the reported literatures [23-27]. In the following cycles, the redox peaks maintained unchanged, indicating a great capacity retention of CNT@S cathode. The similar consequences were obtained for the graphene@S cathode, as shown in Fig. 2b. As a control, the CV of the nomal sulfur cathode electrode was measured under the same condition. As shown in Fig. 2c, the ill-defined CV curves of the common sulfur cathode electrode illustrated that the sulfur cathode possessed a high polarization and a depressed reaction kinetics due to the poor conductivity [28]. All these results suggested that the conductive CNT or graphene layer could efficiently suppress the polarization (over potential) of the sulfur cathode due to the enhanced conductivity.

Download:
Fig. 2. CV curves of CNT@S (a), graphene@S (b), normal S cathode (c), respectively, at 0.1 mV/s. Gralvanostatic voltage profiles of CNT@S (d), graphene@S (e), S cathode (f), respectively.

The different CV characteristics of the three cathodes were also observed in the galvanostatic charge/discharge curves at diverse current densities from 0.1 C to 3 C (1 C = 1675 mA/g). As depicted in Fig. 2d, the CNT@S cathode presented an admirable voltage plateaus, which was consistent with the redox peaks in the CV measurement, indicating the great electron and ion transport properties. Furthermore, the low overpotential (ΔE = 130 mV), calculated by the charge and discharge plateau at 0.1 C, showed the positive effect of the CNT layer on the polarization of the cathode and the reaction kinetics [27, 29]. The graphene@S cathode exhibited a little larger overpotential (ΔE = 142 mV) in Fig. 2e, attributing to a little weaker redox-reaction kinetics. On the contrary, the normal S cathode exhibited even worse redox reaction kinetics in Fig. 2f. The voltage plateau disappeared at 0.2 C, demonstrating the poor conductivity and high polarization of the S cathode.

Fig. 3 showed the long-term cycling performance of the three cathode electrodes at 0.2 C and 1 C. The CNT@S cathode exhibited a high initial discharge capacity of 1050 mAh/g at 0.2 C (Fig. 3a). After 100 cycles, a high discharge capacity of 870 mAh/g was still maintained, corresponding to an outstanding capacity retention of 80.7%. The graphene@S cathode delivered a little weaker performance and its specific discharge capacity decreased from 960 mAh/g to 750 mAh/g with a capacity retention of 78.1% after 100 cycles. On the contrary, the initial capacity of the normal sulfur cathode was only 413 mAh/g and the capacity retention was 40% after 100 cyclesat 0.2 C. The high capacities and the extraordinary cycle stability of the CNT@S and graphene@S cathodes suggested that the conductive coating layers could improve the conductivity of cathode and suppress the dissolution of polysulfides, which would lead to a higher utilization ratio of the active materials [30]. Moreover, the superior performance of the CNT@S was originated from the 3D interconnected framework of CNT layer, which could effectively depress the shuttle effect of soluble polysulfide intermediates. Additionally, the specific capacities of the three electrodes at 1 C also demonstrated the superiority of the CNT@S cathode, as shown in Fig. 3b. The discharge capacity of the CNT@S changed from 920 mAh/g to 740mAh/g with a capacity decay of 0.06% per cycle over 300 cycles The discharge capacityof graphene@S cathodes delivered from 610 mAh/g to 540 mAh/g, with a capacity decay of 0.04% per cycle over 300 cycles. On the contrary, the normal S cathode, exhibiting an initial capacity of 280 mAh/g, dramatically decreased at the original cycles and finally failed.

Download:
Fig. 3. Long-term cycling performance of CNT@S (the red curve), Graphene@S (the green curve), S cathode (the black curve) at 0.2 C (a),1 C (b). (c) Rate performance of CNT@S (the red curve), Graphene@S (the green curve), and S cathode (the black curve).

To further study the effect of the conductive coatings on the reversibility of the sulfur cathode, the rate capacities of the CNT@S, graphene@S, and normal S cathode electrodes were investigated under various current densities (Fig. 3c). Similar to the long-term cycling results, the CNT@S and graphene@S cathodes, especially the CNT@S cathode, exerted more outstanding performance than the normal S cathode. As it could be seen from Fig. 3c, when the current density increased from 0.1 C, 0.2 C, 0.5 C, 1 C, 2 C to 3 C, the CNT@S cathode delivered discharge capacities of 1060, 980, 880, 800, 770 and 690 mAh/g, respectively. Moreover, when the current density was recovered to 0.1 C and 0.5 C, the reverse capacities of the CNT@S turned back to 990 and 840 mAh/g, indicating that the CNT conductive layer could effectively stabilize the S cathode under high current density, and enhance the reaction kinetics [31]. Similarly, the graphene@S cathode delivered stable capacities of 970, 770, 600, 500, 460 and 400 mAh/g at 0.1 C, 0.2 C, 0.5 C, 1 C, 2 C and 3 C, respectively. When the current density was recovered to 0.1 C and 0.5 C, the reverse capacity of the graphene@S could maintain 830 and 580 mAh/g. In the control experiment, the normal S cathode exhibited a low discharge capacity of 290 mAh/g at 0.1 C, and the discharge capacity sharply decreased and failed at 0.5 C, due to the poor conductivity and reversibility of the normal S cathode.

To verify the absorption effects of CNT and graphene layers to the polysulfide intermediates, an absorption experiment was carried out according to ref. [22]. Li2S6 solution was prepared by dissolving Li2S and S with a molar ratio of 1:5 in a mixed solvent of 1, 3-dioxolane (DOL) and dimethoxyethane (DME) with a volume ratio of 1:1. CNT and graphene layers with the same weight ratio were coated on an Al foil and dried at 50 ℃ in vacuo overnight, followed by sinking into the Li2S6 solution for 24 h. As it can been seen in Fig. S2 (Supporting information), the color of the solutions with CNT or graphene turned into much lighter than the pristine Li2S solution, indicating that the CNT and graphene layers could effectively adsorb the polysulfide.

The superior redox kinetics and higher initial discharge capacity of the CNT@S cathode were ascribed to the better conductivity of the CNT layer. To further investigate it, EIS measurements were performed on the CNT@S, graphene@S, and normal S cathode electrodes before cycling. As it was shown in Fig. S3a (Supporting information), all the Nyquist plots of the CNT@S, graphene@S, and normal S cathode electrodes were composed of one semicircle in high-frequency region and an inclined line in low-frequency region. The semicircle in the highfrequency region was ascribed to the charge-transfer resistance and the inclined line represented diffusion process [32, 33]. The Rct (20.1 V) of the CNT@S was smaller than the Rct (30.9 V) of the graphene@S, demonstrating the higher conductivity of the CNT layer than the graphene layer. In addition, the normal S exhibited a much larger Rct (83.7 V), indicating the insulation characteristic of sulfur. It was apparent that the conductive layers could effectively decrease the electrode resistance, leading to a low overpotential and high initial capacity. Furthermore, the EIS measurements after 20 cycling at 0.2 C were investigated at the low discharge plateau. As shown in Fig. S3b (Supporting information), all of the three cathodes exhibited another more semicircle in medium-frequency regions, corresponding to the formation of solid Li2S2/Li2S films [34, 35]. The normal S cathode exhibited a large semicircle in medium-frequency region, related to the formation of thick nonconductive films on the surface of the cathode. In contrast, the CNT@S and graphene@S cathode exhibited a smaller semicircle in medium-frequency region, indicating that the conductive layers effectively alleviated the formation of solid Li2S2/Li2S films and improved the mass transport. The charge-transfer resistance of CNT@S and graphene@S cathodes after cycling was closed to the value before cycling, which might because of the excellent stability and high conductivity of the conductive layers [32]. At the low frequency, the cycled CNT@S cathode exhibited a higher slope than the graphene@S and normal S cathodes, suggesting a lower diffusion resistance of CNT layer [32]. The interlinked framework of CNT was more favorable for the electronic and ionic conduction, leading to the lower charge-transfer and diffusion resistance. Furthermore, the conductive layer changed the interface between the S cathode and the electrolyte, which would suppress the polysulfides, leading to the formation of thinner nonconductive films. In addition, the four-point probe test was carried out to illustrate the enhanced conductivity. The conductivity of CNT@S layers was 22.13 S/cm, while the conductivity of graphene-S layers was 7.35 S/cm. The results indicated the conductive lays effectively enhance the conductivity of the electrode, consistent with the EIS measurement.

SEM-EDS was employed to investigate the structural evolution of active materials after cycling. Fig. S4a in Supporting information showed the stable morphology of the CNT@S cathode after cycling. The EDS mapping image of the CNT@S cathode after cycling showed that slight sulfur embedded in the CNT layer, owing to the trapping polysulfide intermediates by CNT layer [36-38]. The graphene@S cathode after cycling exhibited the similar behavior (Fig. S4b in Supporting information). It was figured out that the presence of the conductive layers could improve the redox accessibility and alleviate the shuttle effect by physical and chemical absorption, of which results were consistent with the absorption experiment.

It was reported that the common Al foil might reduce the long-cycle stability of Li-S batteries due to its instability [39, 40]. Considering the superior performance of the CNT conductive layer from the previous experiments, we covered the Al foil with a thin CNT layer to further enhance the performance of Li-S batteries. The cross section of the CNT@S cathode and the EDS mapping images were shown in Figs. S5a and b (Supporting information), which displayed a thin carbon layer inserted between the Al foil and the sulfur layer. The CNT@S cathode with the CNT conductive layer on the Al foil exhibited a high initial capacity of 1180 mAh/g and a specific capacity of 950 mAh/g after 300 cycles at 1 C (Fig. S4c in Supporting information), both higher than the normal CNT@S cathode. Moreover, the capacity fade exhibited a little less than 0.06% per cycle, indicating the outstanding cycle stability. It was speculated that the conductive layer could also stabilize the Al current collector as a protective layer and enhance the conductivity of the integrate electrode, resulting in a higher capacity.

Moreover, it was crucial to increase the loading amount of sulfur active material in the electrode for increasing the energy density of Li-S batteries [28, 41-43]. Using CNT as the conductive layer, three CNT@S cathode electrodes with different sulfur loading amount of about 1.8, 2.8 and 3.8 mg/cm2 were prepared by the facile approach. As shown in Figs. S6a and b (Supporting information) (the cell 1 mentioned in Figs. 3a and b), the cell 2 (the sulfur loading amount was about 2.8 mg/cm2) delivered an initial capacity of 1040 mAh/g at 0.2 C and 890 mAh/g at 1 C, and maintained a high capacity of 710 mAh/g after 100 cycles at 0.2 C, and 625 mAh/g after 300 cycles at 1 C. The cell 3 (the sulfur loading amount was about 3.8 mg/cm2) exhibited an initial capacity of 930 mAh/g at 0.2 C, 667 mAh/g at 1 C, and maintained the capacity of 623 mAh/g after 100 cycles at 0.2 C, and 505 mAh/g after 300 cycles at 1 C. The remarkable capacity and cycling performance with high sulfur loading amount presented great potential of the facile approach for the high energy-density Li-S batteries.

In summary, we have demonstrated a facile and effective approach to large-scale fabricate Li-S batteries with a remarkable performance. By coating a conductive layer on normal S cathode electrode, the initial capacity and capacity retention of the Li-S batteries were highly enhanced. The superior CNT@S cathode delivered a high reversible capacity of 740 mAh/g over 300 cycles at 1 C and 870 mAh/g over 100 cycles at 0.2 C, owing to the better conductivity and polysulfides confinement by physical and chemical absorption of CNT conductive layer. Furthermore, by coating a protective CNT layer on the Al foil current collector, the capacity of Li-S batteries could be further elevated to 950 mAh/g over 300 cycles at 1 C. Even at a high sulfur loading amount of 3.8 mg/cm2, the CNT@S batteries still showed a high initial capacity of 930 mAh/g at 0.2 C. The impressive performance was attributed to the interface transformation by conductive coatings, which could enhance the conductivity and suppress the shuttle effect. Importantly, this strategy could be conducted in commercial product line of lithium-ion battery without any modification, which made large-scale production of Li-S batteries possible.

Acknowledgment

The authors thank the financial support from the National Natural Science Foundation of China (Nos. 51573030, 51573028 and 51773042).

Appendix A. Supplementary data

Supplementary material related to this article can befound, in the online version, at doi:https://doi.org/10.1016/j.cclet.2018.08.013.

References
[1]
Y. Yang, G.Y. Zheng, Y. Cui, Chem. Soc. Rev. 42 (2013) 3018-3232. DOI:10.1039/c2cs35256g
[2]
A. Rosenman, E. Markevich, G. Salitra, et al., Adv. Energy Mater. 5 (2015) 1500212. DOI:10.1002/aenm.201500212
[3]
Y.P. Xie, H.W. Cheng, W. Chai, et al., Chin. Chem. Lett. 28 (2017) 738-742. DOI:10.1016/j.cclet.2016.07.030
[4]
L. Chen, L.L. Shaw, J. Power Sources 267 (2014) 770-783. DOI:10.1016/j.jpowsour.2014.05.111
[5]
Y.X. Yin, S. Xin, Y.G. Guo, L.J. Wan, Angew. Chem. Int. Ed. 52 (2013) 13186-13200. DOI:10.1002/anie.201304762
[6]
L. Borchardt, M. Oschatz, S. Kaskel, Chem.-Eur. J. 22 (2016) 7324-7351. DOI:10.1002/chem.201600040
[7]
Z.W. Seh, Y.M. Sun, Q.F. Zhang, et al., Chem. Soc. Rev. 45 (2016) 5605-5634. DOI:10.1039/C5CS00410A
[8]
R. Xu, Y.Z. Sun, Y.F. Wang, Y. Cui, Chin. Chem. Lett. 28 (2017) 2235-2238. DOI:10.1016/j.cclet.2017.09.065
[9]
O.W. Sheng, C.B. Jin, J.M. Luo, et al., J. Mater. Chem. A 5 (2017) 12934-12942. DOI:10.1039/C7TA03699J
[10]
O.W. Sheng, C.B. Jin, J.M. Luo, et al., Nano Lett. 18 (2018) 3104-3112. DOI:10.1021/acs.nanolett.8b00659
[11]
Z. Li, Y.M. Huang, L.X. Yuan, Z.X. Hao, Y.H. Huang, Carbon 92 (2015) 41-63. DOI:10.1016/j.carbon.2015.03.008
[12]
S. Evers, L.F. Nazar, Chem. Commun. 48 (2012) 1233-1235. DOI:10.1039/C2CC16726C
[13]
Z.B. Xiao, Z. Yang, H.G. Nie, et al., J. Mater. Chem. A 2 (2014) 8683-8689. DOI:10.1039/C4TA00630E
[14]
Z. Li, H.B. Wu, X.W. Lou, Energy Environ. Sci. 9 (2016) 3061-3070. DOI:10.1039/C6EE02364A
[15]
M.D. Zhang, C. Yu, J. Yang, et al., J. Mater. Chem. A 5 (2017) 10380-10386. DOI:10.1039/C7TA01512G
[16]
Y.B. An, Q.Z. Zhu, L.F. Hu, et al., J. Mater. Chem. A 4 (2016) 15605-15611. DOI:10.1039/C6TA06088A
[17]
P. Chiochan, N. Phattharasupakun, J. Wutthiprom, et al., Electrochim. Acta 237 (2017) 78-86. DOI:10.1016/j.electacta.2017.03.199
[18]
Q. Sun, B. He, X.Q. Zhang, A.H. Lu, ACS Nano 9 (2015) 8504-8513. DOI:10.1021/acsnano.5b03488
[19]
J.H. Yan, X.B. Liu, M. Yao, et al., Chem. Mater. 27 (2015) 5080-5087. DOI:10.1021/acs.chemmater.5b01780
[20]
W.D. Zhou, B.K. Guo, H.C. Gao, J.B. Goodenough, Adv. Energy Mater. 6 (2016) 1502059. DOI:10.1002/aenm.201502059
[21]
W. Tong, Y.D. Huang, W. Jia, et al., J. Alloy Compd. 731 (2018) 964-970. DOI:10.1016/j.jallcom.2017.10.115
[22]
C.B. Jin, W.K. Zhang, Z.Z. Zhuang, et al., J. Mater. Chem. A 5 (2017) 632-640. DOI:10.1039/C6TA07620C
[23]
S.K. Liu, X.B. Hong, Y.J. Li, et al., Chin. Chem. Lett. 28 (2017) 412-416. DOI:10.1016/j.cclet.2016.10.038
[24]
K. Kumaresan, Y. Mikhaylik, R.E. White, J. Electrochem. Soc. 155 (2008) A576. DOI:10.1149/1.2937304
[25]
X.L. Ji, L.F. Nazar, J. Mater. Chem. 20 (2010) 9821. DOI:10.1039/b925751a
[26]
Y.W. Chen, S.Z. Niu, W. Lv, C. Zhang, Q.H. Yang, Chin. Chem. Lett. (2018), doi: http://dx.doi.org/10.1016/j.cclet.2018.04.019.
[27]
H.J. Yu, H.W. Li, S.Y. Yuan, et al., Nano Res. 10 (2017) 2495-2507. DOI:10.1007/s12274-017-1454-1
[28]
Y.N. Liu, G.L. Feng, X.D. Guo, et al., J. Alloy Compd. 748 (2018) 100-110. DOI:10.1016/j.jallcom.2018.03.110
[29]
J.H. Zheng, G.N. Guo, H.W. Li, et al., ACS Energy Lett. 2 (2017) 1105-1114. DOI:10.1021/acsenergylett.7b00230
[30]
C. Zheng, S.Z. Niu, W. Lv, et al., Nano Energy 33 (2017) 306-312. DOI:10.1016/j.nanoen.2017.01.040
[31]
D.H. Liu, C. Zhang, G.M. Zhou, et al., Adv. Sci. (2017) 1700270.
[32]
M. Yan, Y. Zhang, Y. Li, et al., J. Mater. Chem. A 4 (2016) 9403-9412. DOI:10.1039/C6TA03211G
[33]
N.A. Cañas, K. Hirose, B. Pascucci, et al., Electrochim. Acta 97 (2013) 42-51. DOI:10.1016/j.electacta.2013.02.101
[34]
Z.F. Deng, Z.A. Zhang, Y.Q. Lai, et al., J. Electrochem. Soc. 160 (2013) A553-A558. DOI:10.1149/2.026304jes
[35]
J.H. Yan, X.B. Liu, B.Y. Li, Adv. Sci. 3 (2016) 1600101. DOI:10.1002/advs.v3.12
[36]
S.H. Chung, C.H. Chang, A. Manthiram, Small 12 (2016) 939-950. DOI:10.1002/smll.201503167
[37]
L. Qie, A. Manthiram, Adv. Mater. 27 (2015) 1694-1700. DOI:10.1002/adma.201405689
[38]
S.H. Chung, P. Han, A. Manthiram, ACS Appl. Mater. Inter. 8 (2016) 4709-4717. DOI:10.1021/acsami.5b12012
[39]
H.J. Peng, W.T. Xu, L. Zhu, et al., Adv. Funct. Mater. 26 (2016) 6351-6358. DOI:10.1002/adfm.v26.35
[40]
S.T. Myung, Y. Hitoshi, Y.K. Sun, J. Mater. Chem. 21 (2011) 9891-9911. DOI:10.1039/c0jm04353b
[41]
D.P. Lv, J.M. Zheng, Q.Y. Li, et al., Adv. Energy Mater. 5 (2015) 1402290. DOI:10.1002/aenm.201402290
[42]
Z.J. Guo, B. Zhang, D.J. Li, et al., Electrochim. Acta 230 (2017) 181-188. DOI:10.1016/j.electacta.2017.01.174
[43]
Z. Wu, W. Wang, Y.T. Wang, et al., Electrochim. Acta 224 (2017) 527-533. DOI:10.1016/j.electacta.2016.12.072