Chinese Chemical Letters  2021, Vol. 32 Issue (1): 203-209   PDF    
Ionic/electronic conductivity regulation of n-type polyoxadiazole lithium sulfonate conductive polymer binders for high-performance silicon microparticle anodes
Yuanyuan Yua, Huihui Gaoa, Jiadeng Zhub, Dazhe Lia, Fengxia Wanga, Chunhui Jianga, Tianhaoyue Zhonga, Shuheng Lianga, Mengjin Jianga,*     
a College of Polymer Science and Engineering, Sichuan University, Chengdu 610065, China;
b Chemical Sciences Division, Oak Ridge National Laboratory, Oak Ridge, TN 37831, United States
Abstract: Low-cost silicon microparticles (SiMP), as a substitute for nanostructured silicon, easily suffer from cracks and fractured during the electrochemical cycle. A novel n-type conductive polymer binder with excellent electronic and ionic conductivities as well as good adhesion, has been successfully designed and applied for high-performance SiMP anodes in lithium-ion batteries to address this problem. Its unique features are attributed to the strong electron-withdrawing oxadiazole ring structure with sulfonate polar groups. The combination of rigid and flexible components in the polymer ensures its good mechanical strength and ductility, which is beneficial to suppress the expansion and contraction of SiMP s during the charge/discharge process. By fine-tuning the monomer ratio, the conjugation and sulfonation degrees of the polymer can be precisely controlled to regulate its ionic and electronic conductivities, which has been systematically analyzed with the help of an electrochemical test method, filling in the gap on the conductivity measurement of the polymer in the doping state. The experimental results indicate that the cell with the developed n-type polymer binder and SiMP (~0.5 μm) anodes achieves much better cycling performance than traditional non-conductive binders. It has been considered that the initial capacity of the SiMP anode is controlled by the synergetic effect of ionic and electronic conductivity of the binder, and the capacity retention mainly depends on its electronic conductivity when the ionic conductivity is sufficient. It is worth noting that the fundamental research of this work is also applicable to other battery systems using conductive polymers in order to achieve high energy density, broadening their practical applications.
Keywords: n-Doping    Conductive binder    Electronic conductivity    Ionic conductivity    High-performance silicon microparticle    anodes    

Lithium-ion batteries (LIBs) have attracted tremendous attention because of their excellent characteristics, such as high energy density, excellent electrochemical energy conversion mechanisms, low self-discharge, lightweight, etc. [1-4], which have been widely used in portable electronic devices, electric vehicles, and hybrid power [5-8]. Nevertheless, the negative electrode of commercial LIBs is graphite with a low theoretical specific capacity of only ∼370 mAh/g, which is far from meeting the needs of modern society for high-energy-density batteries [9, 10]. Among various promising anode materials, significant attention is paid to silicon (Si), which has the highest theoretical specific capacity (4200 mAh/g), low Li-ion insertion and removal platform, abundant resources, and good safety [11-15]. However, the volume expansion of Si can be up to 400% during the insertion of Li ions, which could crack electrodes, ultimately cause rapid capacity decay and short cycle life, and severely hinder its practical applications in LIBs [16-19]. In order to solve this challenge, many efforts have been made to modify the structure of Si, including zero-dimensional (0D) Si nanoparticles, 1D Si nanowires, core-shell Si nanorods, etc., which effectively alleviated the mechanical strain generated by the volume change and improved the electrochemical performance of the Si-based electrode. Although nanostructures have been proven to be effective in improving the cycle performance of silicon anode, silicon nanomaterials are still costly and have not been practically used due to their complex process [17, 20]. Comparatively, low-cost SiMP shows more commercial prospects. However, there is a huge challenge associated with using SiMP [21]. Typically, Si particles larger than ∼150 nm and Si nanowires larger than ∼250 nm will be severely pulverized during lithification [22, 23]. Therefore, it is necessary to adopt specific measures to suppress the loss of electrode capacity caused by such fracture. In this regard, using high-performance binders is one of the most straightforward and economic approaches for achieving high-performance and practical SiMP anodes even though only a tiny amount of binders can maintain SiMP electrodes' integrity with good electrical connectivity [24].

The traditional SiMP anode is composed of SiMP particles, conductive additives, and non-conductive binders (Fig. 1a). The non-conductive binders can no longer satisfy the preparation of high-performance SiMP anode since they cannot sustain the good electrical connection between SiMP particles and copper foil/conductive additives after volume expansion, leading to a low capacity and fast capacity decay [25, 26]. In contrast, the conductive binder can achieve the dual functions of electrical connection and bonding (Fig. 1b) [27-29]. However, the currently studied conductive polymer binders, such as polyaniline (PANI) hydrogels [30], poly(phenanthrene quinone) (PPQ) [31], poly(3, 4-ethylene dioxythiophene): poly(styrene sulfonate) (PEDOT: PSS) [32-34], all have exhibited limited performance, which mainly stems from the dedoping of the p-type conductive polymer caused by the SiMP anode working at a strong reduction potential, reducing or even losing the electronic conductivity [2]. Therefore, n-type conductive polymer binders are being studied, which can maintain stable n-doping sate to conduct electrons even under the reducing environment of SiMP anodes. Although many studies have made tremendous breakthroughs and achieved high-performance SiMP anodes, the research on the conductivity of n-type conductive binder is still lacking, and there is no in-depth study on the measurement of its conductivity [24, 35].

Download:
Fig. 1. Schematics of the technical approaches to address the volume change issue in Si materials. (a) The traditional SiMP anode is composed of conductive additive, non-conductive binder, and silicon particle. (b) n-type conductive polymer binder is used to keep both electrical and mechanical integrity of SiMP anode. (c) Synthesis schematic of the novel n-type conductive Li-SPOD, with sulfonate groups for adhesion and ionic conductivity, and oxadiazole ring for reducing LUMO energy level.

To address these issues, an n-type conductive polymer binder that can conduct electrons and ions even under the reducing environment for anodes has been developed in this study. Our previous research has demonstrated that the aromatic polyoxadiazole (POD) has an excellent heat resistance and mechanical properties, which are utilized in many high-performance films and fibers [36, 37]. Its unique oxadiazole structure with strong electron-withdrawing properties enables the polymer to be n-doped [38]. However, due to the rigid structure of POD, it has poor processability since it is only soluble in strong polar solvents such as sulfuric acid. Therefore, we propose to improve its processability via the introduction of 4, 4-biphenyl ether structure, which can be grafted with lithium sulfonate polar groups to enhance the adhesion and ionic conductivity of the polymer [39], obtaining a high-performance SiMP-based anode binder.

In this paper, a series of lithium sulfonated polyoxadiazole (Li-SPOD) binders with different sulfonation degrees by changing the ratio of terephthalic acid and 4, 4-diphenyl ether dicarboxylic acid were successfully synthesized via a one-step process (Fig. 1c). The polymers exhibit excellent ionic conductivity, good electronic conductivity, and mechanical properties, which provide strong adhesion between active material, conductive additives, and current collector. The polymers have a good electrolyte absorption rate to ensure effective Li-ion transmission. The effects of polymer conjugation and sulfonation degree on the cycling performance and rate capability of SiMP anodes were also studied. Moreover, aiming at the current research gap in systematic analysis and verification of electronic conductivity of conductive polymer binders, the electronic and ionic conductivity of the n-doped conductive polymer was disassembled by an electrochemical method and calculated, which is of remarkable significance for the study of conductive polymers.

The structures of the Li-SPODs were analyzed by infrared attenuation total reflection spectroscopy (ATR-FTIR), as shown in Fig. 2a. The characteristic peak at 1614 cm−1 derives from the double bond vibration of the benzene ring on the main chain. The three peaks at 1262, 1090, and 1029 cm−1 correspond to stretching oscillations of S=O on cyclized sulfonate groups [40]. The peaks appearing at 1309, 1240, 1197, and 1150 cm-1 are attributed to the stretching vibration absorption peaks of O=S=O and S-O on lithium sulfonate groups, respectively [36]. The peaks at 1018 cm−1/947 cm−1 and 1538 cm−1 are assigned to the stretching vibration of oxadiazole ring C-O-C [41] and the stretching vibration of C=N in the oxadiazole ring [42, 43]. As the DPE content increases, the corresponding peak intensity of the sulfonate group becomes larger by comparing four curves, indicating the increased number of grafted sulfonate groups. 1H NMR was also performed to identify the structure of Li-SPODs (Fig. S1 in Supporting information). The chemical shifts of some hydrogen atoms coincide because their chemical environments are similar [40]. Fig. 2b shows X-ray diffraction (XRD) results of the prepared Li-SPOD films, which are close to each other in shape. No distinct crystallization peak appears from 5° to 50° with only two broad diffraction peaks, indicating that the SD has no significant effect on the crystallinity of Li-SPOD polymers.

Download:
Fig. 2. Structure and properties of Li-SPODs: (a) ATR spectra, combined with 1H NMR spectroscopy analysis, indicating that the polymers with the expected structure were obtained. (b) XRD patterns, indicating that the SD has no significant effect on the crystallinity of Li-SPOD polymers. (c) Test results, including tensile strength and elongation, in which the 45/55 Li-SPOD shows the highest tensile strength. (d) Good electrolyte uptake.

Due to the different mass ratios of TPA and DPE in Li-SPODs, the content of sulfonate groups and the degree of conjugation are different. The degree of sulfonation directly affects ionic conductivity, and the degree of conjugation determines electronic conductivity [44]. Through element analysis, the carbon (C) and sulfur (S) element contents of Li-SPODs were obtained. The sulfonation degree (SD) of Li-SPOD was calculated in combination with the formula SD = WS/WC×(12 + 21n)/(4 + 8n) (where WS is the mass fraction of S in Li-SPODs, WC is the mass fraction of C in Li-SPODs, n is the molar ratio of DPE to TPA), as shown in Table S1 (Supporting information). It exhibits that the SD decreases while the DPE content reduces. Correspondingly, the decrease of ether structure, which destroys the molecular conjugate, is conducive to increasing the conjugation degree of Li-SPOD.

The mechanical properties of the binder are also critical [45] since the better the mechanical properties of the binder, the better it can tightly bond SiMP with copper foil, enabling high charge and discharge performance of SiMP anodes [46, 47]. As the TPA content increases, the tensile strength of the Li-SPOD films increases (Fig. 2c), resulting from its rigid structure. The strength of Li-SPOD can reach 78.3 MPa with a ratio of TPA/DPE = 45/55, while the strength of 50/50 Li-SPOD decreases due to its poor solubility, inducing more defects in the film-forming process. The elongation at break gradually decreases with the increase of TPA, because the TPA lacks the flexible ether structure.

The electrolyte uptake of the binder is also a key indicator, which directly affects the transmission of ions [48]. The electrolyte uptake (ω) increases with the increase of soaking days, as shown in Fig. 2d. As the DPE proportion increases, the electrolyte uptake of Li-SPODs gradually increases.

Fig. S2 (Supporting information) displays the cyclic voltammograms (CV) of the cells tested under 0∼2.5 V (Li/Li+). CV curves show that all the polymers have two reduction peaks at 1.5 V and 0.5 V (Li/Li+), indicating that they can be electrochemically doped at both potentials. As the DPE decreases, the reduction peak area gradually increases, exhibiting the increase of the doping rate, which may be due to an increase in the degree of conjugation. However, the 50/50 Li-SPOD, which should theoretically have a high doping rate, shows the opposite trend. It might be caused by the poor swell ability of the 50/50 Li-SPOD in the electrolyte, which makes ion transport difficult. Furthermore, the cycles provide good reproducibility with excellent overlapping of all the peaks, suggesting a stable electrochemical process. These results show that the Li-SPODs can undergo a reversible redox reaction.

There has little research on the conductivity of n-type conductive binders in the doping state. The enhanced conductivity after doping has only been proved through theoretical calculations or simple impedance experiments without systematic analysis [2, 49]. Therefore, a strategy has been designed to analyze the ionic and electronic conductivities of the binder in this study. The Li-SPODs were assembled into a semi-blocked coin cell (one side was stainless steel, and the other was Li metal) without any electrolyte to separate and calculate their electronic and ionic conductivities at the doping state, as shown in Fig. 3a. First, the electrochemical impedance spectroscopies of the cells were tested (Fig. 3b), and the total conductivity (σt) can be calculated from the bulk impedance, which is composed of ionic and electronic conductivities, according to the formula σt = L/(RS) (where L is the thickness, R is the measured bulk impedance, and S is the area of the Li-SPOD film) [50]. Then the direct-current polarization curves of cells were recorded (Fig. 3c). When the current tends to be stable, the ions no longer migrate, and the equilibrium current (ie) is only contributed by electronic conduction. Thus, the electronic conductivity (σe) can be calculated according to the equation σe = (Lie)/(US) (where L is the thickness of the film, and S is the film area (2.0096 cm2), and U is the applied voltage) [51]. Finally, the ionic conductivity (σi) can be obtained by subtracting the σe from the σt, as summarized in Table S2 (Supporting information).

Download:
Fig. 3. (a) Configuration of the setup for electronic and ionic conductivity tests, the Li-SPODs were assembled into a semi-blocked coin cell (one side was stainless steel, and the other was Li metal). (b) Impedance spectra for total electronic conductivity of Li-SPOD films. (b) DC polarization curves for electronic conductivity of Li-SPOD films at 0.01 V.

With the decrease in the proportion of DPE, the σi of the Li-SPOD decreases, while the σe slowly increases. The DPE provides more sulfonate groups, but it deteriorates the conjugation structure of the polymer chain, which brings two different effects on the ionic and electronic conductivities of the Li-SPOD. The σe of the 45/55 Li-SPOD film is up to 8.36 × 10−7 S/cm, while that of the 50/50 Li-SPOD is only 4.4 × 10−7 S/cm. The low σe might be caused by the low swelling degree and the doping difficulty of the 50/50 Li-SPOD.

In order to prove the improved conductivity of Li-SPOD, the impedance spectra of SiMP anodes prepared with 45/55 Li-SPOD, PAALi, and CMC binders at different potentials were collected (Fig. 4a). The results show that when the voltage is reduced from 2 V to 0.1 V (vs. Li/Li+), the interface impedance of the SiMP anode prepared with 45/55 Li-SPOD drops sharply, while there is no significant change for either PAALi or CMC. It indicates that the doping processes of Li-SPOD at 1.5 V and 0.5 V (Li/Li+) increase the conductivity of the polymer [24].

Download:
Fig. 4. (a) Impedance spectra at various potentials for demonstrating the increased conductivity, where the interface impedance of the SiMP anode prepared with 45/55 Li-SPOD drops sharply, while either PAALi or CMC change little and (b) CV curves of the cell with 45/55 Li-SPOD, PAALi, CMC binders. (c) 180° peeling test results of SiMP anodes with various binders, and the SiMP anode with 45/55 Li-SPOD still has the best adhesion, consistent with the tensile test results. (d) Cycling performance and (e) rate capability of the cells with different binders, and the cell with 45/55 Li-SPOD shows better cycle stability and capacity retention than that with PAALi or CMC.

Fig. 4b exhibits that the electrode has a broad reduction peak near 0.5 V due to the formation of the solid electrolyte interface (SEI) layer on the surface of the SiMP anodes at the first cycle. In the second cycle, a clear reduction peak appears at 0.25 V, which corresponds to the reaction potential of Li ions inserted into SiMPs to form a SiMP-Li alloy. There are two oxidation peaks at 0.35 V and 0.5 V, which are assigned to the extraction of Li ions from the SiMPs. Moreover, the current density of the redox peak increases significantly with the increased cycles, because as the reaction proceeds, more and more SiMPs participate in the reaction, making the silicon particles steadily activated [2].

Fig. 4c displays a 180° peeling test, which was used to evaluate binders' adhesion. Among the synthetic polymers, the SiMP anode with 45/55 Li-SPOD has the largest adhesion force (1.20 N/mm), while that of the 30/70 Li-SPOD exhibits the smallest adhesion force (0.48 N/mm), which is consistent with the tensile results. Compared with the reference binders, the SiMP anode with 45/55 Li-SPOD still has the best adhesion, followed by PAALi, and CMC is the worst because of the presence of sulfonate and high molecular strength for Li-SPOD.

The cycling performance of the cells with various binders is shown in Fig. 4d. The cells with 30/70 Li-SPOD (the highest σi but lowest σe) and 45/55 Li-SPOD (low σi but the highest σe) have much better initial specific capacities, which are 3000 and 3200 mAh/g, respectively. After 100 cycles, the specific capacity of the former reduces to only 658 mAh/g, and that of the later maintains 1710 mAh/g. It is because that the electronic conductivity of 30/70 Li-SPOD is low. With the increase of the cycles, the SiMP anode with 30/70 Li-SPOD loses the electrical connection, resulting in a severe capacity decay. The 45/55 Li-SPOD exhibits better capacity retention, which is derived from the highest electronic conductivity. Although its ionic conductivity is lower than the 30/70 Li-SPOD, it still can have an acceptable value (near 10−4 S/cm), which is sufficient to provide ion transport for SiMP anode. The cell with 50/50Li-SPOD has the lowest initial capacity among the four Li-SPODs due to its lowest electronic conductivity and ionic conductivity. Compared with the traditionally commercial PAALi binder, the 45/55 Li-SPOD, due to its excellent ionic conductivity, best electronic conductivity, and good mechanical properties, enables SiMP anode with a higher initial capacity and capacity retention. The cell with CMC has the worst cycle performance because CMC is brittle and cannot adapt to the stress caused by the significant volume change of the SiMP anode during the cycles [52]. According to the analysis, it can be concluded that the initial capacity of the SiMP anode is determined by the synergetic effect of the electronic and ionic conductivities of binders. However, the maintenance of the capacity of the SiMP anode is mainly controlled by the electronic conductivity.

The rate capability of the cells with SiMP particles and 45/55Li-SPOD, PAALi, and CMC has also been evaluated (Fig. 4e). The cell containing 45/55Li-SPOD indicates the best rate capacity because it has not only high electrolyte uptake, which allows Li ions to reach the surface of the SiMP particles quickly but also its good adhesion and excellent conductivity, which contribute to tolerating the rapid expansion at high rates and maintaining the electrical connection. Thereby, the polarization phenomenon can be relieved to help the cell retain high capacities at high current densities. Although PAALi SiMP anode presents a high discharge capacity at a low current density of 0.05 C, it drops faster than that of the cell with 45/55Li-SPOD at a high current density because of its stiffness and non-conductive nature [13]. A high capacity of 2400 mAh/g for the cell with the 45/55 Li-SPOD binder can be achieved when the current density is back to 0.1 C, demonstrating its excellent rate capability. In contrast, the capacity of the cell with the CMC binder has almost a zero capacity because the CMC binder is brittle, causing the electrode broken during the rapid expansion of SiMP.

SEM was performed to explore the surface and cross-sectional changes of the electrode made with 45/55 Li-SPOD, PAALi, and CMC binders before and after cycling tests. The surface and cross-sectional SEM images of SiMP anodes are shown in Fig. S3 (Supporting information). Before cycling, all the surface of the electrode with three binders is even and smooth. After 100 cycles, the volume of SiMPs increases significantly with a cracked surface. Among them, the CMC one has the deepest cracks on the surface, and the volume expansion of the SiMPss is the most dramatic, which is mainly because the CMC is relatively brittle, it ruptures with SiMP expansion during the cycle, and cannot bond the SiMPs well, resulting in a rapid capacity decay [53]. The surface of 45/55 Li-SPOD SiMP is flat, and the cracks are shallow due to its high strength and good ductility. From the cross-sectional images, the SiMP anodes contact well with the copper foil before the cycling test, while after 100 cycles, the active material of the electrode and the copper current collector has different degrees of separation. Among them, the separation degree of SiMP anode prepared by CMC is the greatest, and that of PAALi is second, which means that some SiMP could be deactivated due to the loss of electrical connection, significantly affecting the capacity retention. The SiMP anode prepared by 45/55 Li-SPOD is tightly connected to the current collector, ensuring the mechanical integrity and electrical connection of the electrode because of its excellent mechanical strength.

In summary, this paper reports an economical n-type conductive polymer binder Li-SPOD, which can maintain a stable doping state at the operating potential for SiMP anodes. The ionic and electronic conductivities of the polymer can be fine-tuned by optimizing the proportion of DPE structure in the molecule. The ionic and electronic conductivities of Li-SPOD binders have also been disassembled and systematically analyzed with the electrochemical methods, filling the gap in the research on the conductivity of conductive polymers. Among them, the cell prepared by Li-SPOD with the best performance shows a much better discharge capacity and capacity retention rate than traditionally commercial PAALi and CMC binders, due to its excellent ionic conductivity and good electronic conductivity in doping state. The experimental results show that the initial capacity of the SiMP anode is controlled by the combined influence of ionic and electronic conductance, rather than being dominated by one of them alone. On the other hand, the capacity retention rate of the SiMP anode is mainly determined by the electronic conductivity when the ionic conductivity is sufficient.

Declaration of competing interest

The authors declare no competing financial interest.

Acknowledgments

This work was supported by the Fundamental Research Funds for Central Universities of China and the Key Research and Development Projects of Sichuan (No. 2020YFG0127), and the authors gratefully acknowledge the State Key Laboratory of Polymer Materials Engineering, College of Polymer Science and Engineering, Sichuan University and the Analytical & Testing Centre of Sichuan University.

Appendix A. Supplementary data

Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.1016/j.cclet.2020.10.010.

References
[1]
J. Son, T.V.õ Ngọc, S. Cho, et al., J. Power Sources 458 (2020) 228054.
[2]
D. Liu, Y. Zhao, R. Tan, et al., Nano Energy 36 (2017) 206-212. DOI:10.1016/j.nanoen.2017.04.043
[3]
Y. Cai, Y. Li, B. Jin, et al., ACS Appl. Mater. Interfaces 11 (2019) 46800-46807. DOI:10.1021/acsami.9b16387
[4]
L. Zhang, Y. Ding, J. Song, Chin. Chem. Lett. 29 (2018) 1773-1776. DOI:10.1016/j.cclet.2018.03.008
[5]
S. Hu, Z. Cai, T. Huang, H. Zhang, A. Yu, ACS Appl. Mater. Interfaces 11 (2019) 4311-4317. DOI:10.1021/acsami.8b15695
[6]
M.T. Jeena, J.I. Lee, S.H. Kim, et al., ACS Appl. Mater. Interfaces 6 (2014) 18001-18007. DOI:10.1021/am504854x
[7]
S.Y. Lee, Y. Choi, S.H. Kwon, J.S. Bae, E.D. Jeong, J. Ind, Eng. Chem. 74 (2019) 216-222.
[8]
H. Ye, D. Lei, L. Shen, et al., Chin. Chem. Lett. 31 (2020) 570-574. DOI:10.1016/j.cclet.2019.04.047
[9]
N.S. Choi, S.Y. Ha, Y. Lee, et al., J. Electrochem. Sci. Technol. 6 (2015) 35-49. DOI:10.33961/JECST.2015.6.2.35
[10]
Z. Zheng, H.H. Wu, H. Liu, et al., ACS Nano 14 (2020) 9545-9561. DOI:10.1021/acsnano.9b08575
[11]
E. Bulut, E. Güzel, N. Yuca, O.S. Taskin, J. Appl. Polym. Sci. 137 (2020) 48303.
[12]
L. Wei, C. Chen, Z. Hou, H. Wei, Sci. Rep. 6 (2016) 19583.
[13]
C. Chen, F. Chen, L. Liu, J. Zhao, F. Wang, Electrochim. Acta 326 (2019) 134964.
[14]
X. Liu, J. Zai, A. Iqbal, et al., J. Colloid Interface Sci. 565 (2020) 270-277. DOI:10.1016/j.jcis.2020.01.028
[15]
Y. Zeng, Y. Huang, N. Liu, et al., J. Energy Chem. 54 (2021) 727-735. DOI:10.1016/j.jechem.2020.06.022
[16]
R. Guo, S. Zhang, H. Ying, et al., ACS Appl. Mater. Interfaces 11 (2019) 14051-14058. DOI:10.1021/acsami.8b21936
[17]
J. Choi, K. Kim, J. Jeong, et al., ACS Appl. Mater. Interfaces 7 (2015) 14851-14858. DOI:10.1021/acsami.5b03364
[18]
Q. Zhang, H. Chen, L. Luo, et al., Energy Environ. Sci. 11 (2018) 669-681. DOI:10.1039/C8EE00239H
[19]
W. An, B. Gao, S. Mei, et al., Nat. Commun. 10 (2019) 1447.
[20]
X. Zhang, X. Qiu, D. Kong, et al., ACS Nano 11 (2017) 7476-7484. DOI:10.1021/acsnano.7b03942
[21]
Y. Li, K. Yan, H.W. Lee, et al., Nat. Energy 1 (2016) 15029. DOI:10.1038/nenergy.2015.29
[22]
X.H. Liu, L. Zhong, S. Huang, et al., ACS Nano 6 (2012) 1522-1531. DOI:10.1021/nn204476h
[23]
Z. Chen, C. Wang, J. Lopez, et al., Adv. Energy Mater. 5 (2015) 1401826. DOI:10.1002/aenm.201401826
[24]
G. Liu, S. Xun, N. Vukmirovic, et al., Adv. Mater. 23 (2011) 4679-4683. DOI:10.1002/adma.201102421
[25]
O.S. Taskin, N. Yuca, J. Papavasiliou, G. Avgouropoulos, Mater. Lett. 273 (2020) 127918.
[26]
S. Hu, L. Wang, T. Huang, A. Yu, J. Power Sources 449 (2020) 227472.
[27]
R. Tang, L. Ma, Y. Zhang, et al., ChemElectroChem 7 (2020) 1992-2000. DOI:10.1002/celc.201902152
[28]
C. Jiang, H. Li, C. Wang, Sci. Bull. (Beijing) 62 (2017) 1473-1490. DOI:10.1016/j.scib.2017.10.011
[29]
C. Wang, C. Jiang, Y. Xu, et al., Adv. Mater. 28 (2016) 9182-9187. DOI:10.1002/adma.201603240
[30]
H.C. Tao, X.L. Yang, L.L. Zhang, S.B. Ni, J. Solid State Electrochem. 18 (2014) 1989-1994. DOI:10.1007/s10008-014-2439-6
[31]
S.M. Kim, M.H. Kim, S.Y. Choi, et al., Energy Environ. Sci. 8 (2015) 1538-1543. DOI:10.1039/C5EE00472A
[32]
D. Shao, H. Zhong, L. Zhang, ChemElectroChem 1 (2014) 1679-1687. DOI:10.1002/celc.201402210
[33]
X. Liu, A. Iqbal, N. Ali, R. Qi, X. Qian, ACS Appl. Mater. Interfaces 12 (2020) 19431-19438. DOI:10.1021/acsami.0c00755
[34]
X. Liu, J. Zai, A. Iqbal, et al., J. Colloid Interface Sci. 565 (2020) 270-277. DOI:10.1016/j.jcis.2020.01.028
[35]
M. Ling, M. Liu, T. Zheng, T. Zhang, G. Liu, J. Electrochem. Soc. 164 (2017) A545-A548.
[36]
X. Yan, Z. Li, W. Zhou, et al., Intern. Forum Therm. Stud. 126 (2016) 1301-1311.
[37]
X. Yang, E. Jia, G. Ye, J. Xu, Polym. Bull. 72 (2015) 1067-1080. DOI:10.1007/s00289-015-1322-7
[38]
W. Changsheng, J. Gun-Young, S.B. Andrei, R.B. Martin, C.P. Michael, J. Mater. Chem. 12 (2002) 173-180. DOI:10.1039/b106907c
[39]
A. Abdolmaleki, K. Eskandari, M.R. Molavian, Polymer 87 (2016) 181-193. DOI:10.1016/j.polymer.2016.02.011
[40]
S.O. Ilyin, A.Y. Yadykova, V.V. Makarova, V.S. Yashchenko, Y.V. Matveenko, Polym. Intern. 69 (2020) 1243-1255. DOI:10.1002/pi.6068
[41]
C. Hamciuc, E. Hamciuc, M. Bruma, et al., Polymer 42 (2001) 5955-5961. DOI:10.1016/S0032-3861(01)00081-7
[42]
A. Abdolmaleki, M. Zhiani, M. Maleki, S. Borandeh, K. Firouz, Polymer 75 (2015) 17-24. DOI:10.1016/j.polymer.2015.08.021
[43]
P. Mormile, L. Petti, M. Gillo, et al., Mater. Chem. Phys. 77 (2003) 945-951. DOI:10.1016/S0254-0584(02)00187-6
[44]
H. Gao, J. Mao, D. Li, et al., J. Electrochem. Soc. 167 (2020) 070518. DOI:10.1149/1945-7111/ab6e5c
[45]
L. Yu, J. Liu, S. He, et al., J. Phys. Chem. Solid 135 (2019) 109113. DOI:10.1016/j.jpcs.2019.109113
[46]
D. Yao, J. Feng, J. Wang, Y. Deng, C. Wang, J. Power Sources 463 (2020) 228188. DOI:10.1016/j.jpowsour.2020.228188
[47]
X. Zhang, W.L. Song, H.S. Chen, D. Fang, J. Power Sources 465 (2020) 228290. DOI:10.1016/j.jpowsour.2020.228290
[48]
M. Wu, X. Xiao, N. Vukmirovic, et al., J. Am. Chem. Soc. 135 (2013) 12048-12056. DOI:10.1021/ja4054465
[49]
S. Gao, F. Sun, A. Brady, et al., Nano Energy 73 (2020) 104804.
[50]
I.P. Marozau, D. Marrero-López, A.L. Shaula, et al., Electrochim. Acta 49 (2004) 3517-3524. DOI:10.1016/j.electacta.2004.03.022
[51]
R. Merkle, P. Gutbrod, P. Reinold, et al., Polymer 132 (2017) 216-226. DOI:10.1016/j.polymer.2017.10.064
[52]
K. Lee, T.H. Kim, Electrochim. Acta 283 (2018) 260-268. DOI:10.1016/j.electacta.2018.06.175
[53]
D. Bo, H. Xuanning, C. Zhenfei, et al., Inorg. Chem. Commun. 113 (2020) 107771. DOI:10.1016/j.inoche.2020.107771