⇦Chinese Physics Letters, 2022, Vol. 39, No. 2, Article code 028202 Electrochemical Role of Transition Metals in Sn–Fe Alloy Revealed by Operando Magnetometry Le-Qing Zhang (张乐清)1, Qing-Tao Xia (夏清涛)1, Zhao-Hui Li (李召辉)1, Yuan-Yuan Han (韩媛媛)1, Xi-Xiang Xu (徐熙祥)1, Xin-Long Zhao (赵新龙)1, Xia Wang (王霞)1, Yuan-Yuan Pan (潘圆圆)1, Hong-Sen Li (李洪森)1, and Qiang Li (李强)1,2* Affiliations 1College of Physics, Qingdao University, Qingdao 266071, China 2Weihai Innovation Institute, Qingdao University, Weihai 264200, China Received 5 November 2021; accepted 23 December 2021; published online 29 January 2022 *Corresponding author. Email: liqiang@qdu.edu.cn Citation Text: Zhang L Q, Xia Q T, Li Z H et al. 2022 Chin. Phys. Lett. 39 028202    Abstract As promising materials, alloy-type anode materials have been intensively investigated in both academia and industry. To release huge volume expansion during alloying/dealloying process, they are usually doped with transition metals. However, the electrochemical role of transition metals has not been fully understood. Here, pure Sn$_{3}$Fe films were deposited by sputtering, and the electrochemical mechanism was systematically investigated by operando magnetometry. We confirmed that Fe particles liberated by Li insertion recombine partially with Sn during the delithiation, while the stepwise increase in magnetization with the cycles demonstrates growth of Fe nanoparticles. In addition, we also found an unconventional increase of magnetization in the charging process, which can be attributed to the space charge storage at the interface of Fe/Li$_{x}$Sn. These critical findings pave the way for the mechanism understanding and development of high-performance Sn based alloy electrode materials.
cpl-39-2-028202-fig1.png
cpl-39-2-028202-fig2.png
cpl-39-2-028202-fig3.png
cpl-39-2-028202-fig4.png
 DOI:10.1088/0256-307X/39/2/028202 © 2022 Chinese Physics Society Article Text As the great supports for smart grids, electric vehicles, and micro/nanodevices, lithium-ion batteries (LIBs) have become cornerstone of energy storage technology.[1–7] The ever-growing technological demand for higher-energy systems keeps catalyzing the exploration of higher-capacity LIB electrode materials.[8–11] Reacting with Li via alloying reaction, tin has long been considered as a promising alternative to graphite anodes due to its high specific capacities.[11,12] However, substantial volumetric expansion of 257% on the complete Li insertion in Sn causes fragmentation/loss of the particles, thus leading to a rapid deterioration in the cycling.[11,13–17] In order to mitigate this issue, transition metals (Fe, Co, Ni, etc.) have been usually introduced as a buffer matrix to relieve the volume expansion.[18–22] Early successful commercialization of Sn–Co alloy electrodes was released in the SONY Nexelion battery in 2004.[23] So far, despite the great progress made in Sn-based alloys for LIBs, the role of transition metals in electrochemical reaction upon lithiation/delithiation is still unclear and controversial.[24–26] Among the various Sn-based anodes, being earth-abundant, cost-effective and eco-friendly, Sn–Fe alloys have aroused intense interest and have been extensively investigated in recent years.[27–29] However, it is still a matter of debate whether Fe, as an inactive center, can be alloyed reversibly with Sn during cycling. By using in situ x-ray diffraction and in situ Mössbauer spectroscopy, Dahn et al. found that the iron particles formed during discharge were not completely inactive and re-alloyed with Sn following charged.[30] By contrast, Nwokeke et al. proposed that the morphology of iron nanoparticles produced while embedded lithium was maintained in the subsequent cycles being completely inert.[31] Nevertheless, Whittingham et al. reported that some Fe particles remained after the first cycle of charging without recombining with Sn revealed by a combination of x-ray diffraction and x-ray absorption spectroscopy.[32] Therefore, it is crucial to unravel the electrochemical mechanism during alloying reaction, especially the role of transition metals.[33–35] Due to the high sensitivity to magnetic materials, magnetometry provides a powerful tool to gain insights into the electrochemical mechanism from the perspective of the charge and magnetic interaction.[36–38] To data, magneto-electrochemistry is now understood being common, suggesting promising new concepts of uncovering electrochemical mechanism through magnetic theory and techniques.[39] In this work, we prepared high-pure Sn–Fe alloys without binders and conductive additives via magnetron sputtering deposition, guaranteeing all of the magnetic variations are mainly due to iron element.[8,40] A magnetometry coupled with electrochemistry was performed to monitor, in situ, the time evolution of voltage and magnetization in Sn–Fe/Li cells. The significantly higher magnetization at 2 V compared to the original samples indicates that the Fe particles are partially re-alloyed with Sn as the removal of Li, while the rest of Fe remains metallic elemental state during cycling. Further magnetometry proved that the stepwise increase in magnetization with the cycles is caused by the increase in the size of iron particles. Excitingly, abnormal magnetic response during charging process reveals the phenomenon of space charge storage in alloy materials, which provides extra capacity for batteries. These critical findings help open the way for better understanding the role of transition metals in electrochemical alloying/dealloying processes and the rational design of high-performance electrode materials for LIBs.
cpl-39-2-028202-fig1.png
Fig. 1. Structural and compositional characterization of the initial sample, and the basic electrochemical properties for the Sn–Fe alloy anode in LIBs. The EDX element mapping diagram of the Sn element (a) and Fe element (b), marked blue and yellow, respectively. The XPS spectrum of the Sn $3d$ (c) and Fe $2p$ (d) in Sn–Fe alloy film. (e) XRD pattern of the Sn–Fe and Cu substrate (SAED image of Sn–Fe in the inset). (f) Galvanostatic charge-discharge curves of the initial three cycles at 100 mA$\cdot$g$^{-1}$.
Pure Sn–Fe alloys were prepared by magnetron co-evaporation from tin and iron targets (purity 99.99%) with an Ar gas pressure of 22.5 mTorr at room temperature. The deposition powers applied to the Fe and Sn targets were 10 W and 10 W, respectively. During the sputtering process, the substrate (copper foil) was rotated to ensure uniformity of Sn–Fe alloys. The mass was obtained by measuring the difference of substrate before and after deposition via electrobalance (METTLER TOLEDO). The thickness of the nanocomposite film is approximately as determined by atomic force microscope (AFM, Park systems XE7). Predictably, TEM energy dispersive x-ray spectroscopy (EDX) mapping images [Figs. 1(a) and 1(b)] provided clear evidence of high homogeneity of Sn and Fe elements in the films. Furthermore, the EDX spectrum in Fig. S1 of the Supplementary Material shows the Sn and Fe atomic ratios of 77% and 23%, respectively, thus the electrodes are approximated as Sn$_{3}$Fe. Figures 1(c) and 1(d) show the high-resolution XPS spectra of Sn $3d$ and Fe $2p$ collected from the as-growing samples. In Fig. 1(c), the binding energy difference between Sn $3d_{5/2}$ (484.9 eV) and Sn $3d_{3/2}$ (493.9 eV) is 8.5 eV,[41] which excludes the presence of tin oxides. The two characteristic peaks centered at 706.7 and 719.8 eV correspond well with the $2p_{3/2}$ and $2p_{1/2}$ peaks of metallic Fe [Fig. 1(d)].[42,43] The XRD patterns in Fig. 1(e) show that there is no diffraction peak other than those of the Cu substrate, which indicates that the thin film is in the nanocrystalline or amorphous state. TEM selective area electron diffraction (SAED) diagram shown in the inset exhibits a diffuse ring, which also shows that the sample is in the amorphous state. According to previous reports,[44] amorphous phase is favorable to release the volume expansion of the tin-based anode, which ensures excellent electrochemical performance in the cycle life. Sn–Fe alloys were assembled into LIBs to serve as working electrodes, with LiPF$_{6}$ (1 M, in $1\!:\!1$ w/w ethylene carbonate/dimethyl carbonate) utilized as the electrolyte. Then, electrochemical properties of Sn$_{3}$Fe electrode were evaluated through charge/discharge curves at constant current density of 100 mA$\cdot$g$^{-1}$ as shown in Fig. 1(f). The second and third charge/discharge curves almost overlap, yielding a reversible capacity of 628 mA$\cdot$h$\cdot$g$^{-1}$. Figure S2 in the Supplementary Material shows that a capacity of 605 mA$\cdot$h$\cdot$g$^{-1}$ was delivered by Sn$_{3}$Fe LIBs in the tenth cycle, indicating the good cycling stability. Distinctly, Sn–Fe anodes not only have a better cycle stability through reducing the volume expansion, but also have a higher capacity performance than pure Sn anodes. To understand the electrochemical mechanism, coupling operando magnetometry with cyclic voltammetry (CV) of the Sn$_{3}$Fe LIBs measurement was performed in Fig. 2. As shown in Fig. 2(a), the cathodic peaks from 0.68 to 0.18 V correspond to the progressive formation of Li$_{x}$Sn ($0 \le x \le 4.4$) alloys, which also means the dealloying process of Fe from Sn$_{3}$Fe. Accordingly, the anodic peaks from 0.4 to 0.9 V are related to the stepwise decomposition of Li$_{x}$Sn with the lithium extraction process, which is consistent with the previous reports.[14] The irreversible capacity loss in the first 1–1.5 V discharge results from the generation of the solid electrolyte interface.[22,32] The temporal evolutions in the magnetization of Sn$_{3}$Fe follow the periodicity of the voltage under the condition of constant magnetic field 1 T and 300 K [Fig. 2(b)]. The magnetization values given in emu$\cdot$g$^{-1}$ are defined per unit weight of active materials (Sn$_{3}$Fe). As revealed by Fig. S3 in the Supplementary Material, continuously sampling through magnetization on a cell assembled with or without active materials, the tiny constant value of the background signal during cycling makes no effect in variation of the magnetization. This means that no obstacle existed in further exploring the electrochemical role of transition metals in alloys. The linear magnetic background signals from the other cell assembly components, such as the Li metal, electrolyte, separator and copper foil, is deducted from the total magnetic moment in Fig. 2(b). The initial low magnetization indicates that the prepared sample is mainly in the form of alloy. Before reaching 0.31 V upon lithiation, the magnetization remains stationary. Until the discharge voltage is below 0.31 V, there is a sharp increase in magnetization, which is related to the dealloying of Fe from Sn$_{3}$Fe. In the subsequent charging process, the voltage range of 0.67–1.01 V corresponds to a significant decrease in magnetization, indicating the re-formation of Sn–Fe alloy. However, the higher magnetization at 2 V compared to that of the initial film proves that a part of Fe remains as metallic element state. The fact that Fe formed by the first lithiation is partially reversible upon delithiation is revealed, that is, only part of Fe is reversibly alloyed with Sn during first cycling, and the rest of Fe is inactive. Concluded from the smaller magnitude of the magnetic change, the active metallic Fe decreased gradually.[45]
cpl-39-2-028202-fig2.png
Fig. 2. Reaction mechanism of the Sn$_{3}$Fe anode in lithium-ion batteries. (a) Cyclic voltammetry (CV) test of first three cycles with a scan rate of 0.5 mV$\cdot$s$^{-1}$. (b) Operando magnetic response of LIBs under an applied magnetic field strength of 1 T conducted during the CV test.
Although the magnetic response of the following two cycles keeps the same trend, it is noteworthy that the overall magnetization is stepwise increase with cycles progress. According to previous reports,[45] this may be due to the fact that Fe nanoparticles produced by lithiation are superparamagnetic and increase in size during the cycling. The Fe nanoparticles produced by the insertion of Li are so small as to show superparamagnetic behavior with magnetization dependent strongly on temperature.[46,47] The formation of Fe nanoparticles was examined by high resolution TEM (HRTEM) of the Sn$_{3}$Fe anode discharged to 0 V as visualized in Fig. 3(a). It shows that Fe is approximately a few nanometers in size, and the clear lattice stripe with a spacing of 0.202 nm corresponds to the (110) plane.[48] The rings of selected area electron diffraction (SAED) pattern in Fig. 3(b) can be assigned to the (310), (200) and (110) planes of the Fe phase, which verifies the composition of the nanometer-scale Fe.[49] However, due to the limitation of the detected region, HRTEM can only characterize the local region, and its penetration depth is generally less than 200 nm, thus the information of the entire electrode cannot be obtained.
cpl-39-2-028202-fig3.png
Fig. 3. (a) HRTEM and (b) SAED spectra of Sn$_{3}$Fe LIBs discharging to 0.01 V. (c) Hysteresis loop after the first discharge at different temperatures. (d) ZFC/FC curves of Sn–Fe LIBs under different cycles.
In order to further determine this hypothesis that the magnetization increases step by step as the increment in superparamagnetic Fe grains size, it deserves a further exploration of the magnetization as a function of temperature and magnetic field. The magnetic behavior of Sn$_{3}$Fe is explored within the initial discharge process to 0 V. In the hysteresis loops of Fig. 3(c), Sn$_{3}$Fe LIBs display the expected superparamagnetic characteristics at 300 K, with a magnetization exposition of unsaturation and no coercivity even at a high field of 7 T. Figure 3(c) shows that the fitted curves coincide the experimental curves at 300 K and 5 K, and the diameter of superparamagnetic Fe nanoparticles calculated by Langevin fitting at 300 K is 4.36 nm, which is in great alignment with 4.28 nm calculated at 5 K (Information I in the Supplementary Material). Subsequently, we characterized the magnetization of Sn$_{3}$Fe LIBs that discharged to 0 V as a function of temperature (FC and ZFC) in Fig. 3(d). The electrode demonstrates the typical signature of superparamagnetic particles with a maximum in the ZFC around 65 K and 200 K (blocking temperature, $T_{\rm B}$) corresponding to the first and fifth cycles. In accordance with the relationship between the particle diameter $d$ and $T_{\rm B}$ (Information II in the Supplementary Material), the average diameter of Fe nanoparticles can be estimated to be 4.4 nm in the first cycle. A significantly higher $T_{\rm B}$ in the fifth cycle implies the superparamagnetic nanoparticles with larger size about 6.5 nm, which may be related with the volume expansion during the alloying process.[45] Therefore, through the hysteresis loops and FC/ZFC measurement, we proved that the overall magnetization gradually increased during the cycling was caused by the increasing size of superparamagnetic Fe nanoparticles. Although the inactive Fe produced by the first lithiation process does not participate in the electrochemical reaction, it can improve the electrochemical performance of the batteries as buffer matrix and good electronic conductor.[30] However, excessive aggregated superparamagnetic Fe may form an impenetrable skin on the surface of grains, preventing the adequate transport of Li$^{+}$ in intermetallic and resulted in less than excellent performance.[50–52]
cpl-39-2-028202-fig4.png
Fig. 4. (a) Interface storage model diagram. (b) Schematic of spin-polarized density of states at the surface of ferromagnetic metal grains before and after electronic filling.
Surprisingly, as shown in the green dotted frame marked in Fig. 2(b), the magnetic responses exhibit an abnormal rise during the charging process from 0 to 0.67 V, which is contrary to the decline of magnetization due to the decrease of iron content. This abnormal magnetic response suggests a space charge storage phenomenon that was recently been magnetically proved.[39] The spin-polarized charges automatically store at the interface of the Fe/Li$_{x}$Sn heterostructure, that is to say, Li$^{+}$ ions are adsorbed on one side of Li$_{x}$Sn while electrons are injected into adjacent Fe grains [Fig. 4(a)]. The charges accumulated at the Fe nanograins interface will cause the magnetization changes, due to the spin splitting of iron's 3$d$ energy levels [Fig. 4(b)].[53] On account of the net magnetization, closely related to the effective magnetic moment of spin-polarized electrons, the change in the saturation magnetization should be a monotonic reduction during electron filling. Inversely, magnetization increases as electrons dissipate from Fe at the interface, corresponding to the magnetic response in the voltage range of 0–0.67 V during charging. The abnormal magnetic response demonstrated the existence of space charge phenomenon in Sn–Fe alloy, which provides extra capacity.[39,54–56] However, the magnetization decrease caused by the accumulation of electrons on Fe upon lithiation is not observed. This can be attributed to that the signal of a decrease in magnetization caused by space charge phenomenon is offset by the generation of metallic Fe nanoparticles. We excellently proved the existence of space charge effect in Sn–Fe alloy, which provides a path for design of electrode materials with high performance. In conclusion, high purity Sn–Fe alloy was prepared by magnetron sputtering and its electrochemical mechanism was researched by operando magnetometry. A relatively higher magnetic response after charging than the initial discharge was observed, indicating a partially reversible alloying/dealloying process. Further magnetic measurement showed that the distinctive stepwise rising in magnetization was caused by the larger Fe particles as the number of cycle increase. In addition, we found the mechanism of space charge storage in alloy materials, which provides extra capacity for batteries. These crucial results reveal the role of transition metals in the electrochemical process, and provide a theoretical cornerstone for design of high-performance electrode materials. Acknowledgements. This work was supported by the National Natural Science Foundation of China (Grant Nos. 22179066 and 11504192), and the Natural Science Foundation of Shandong Province (Grant No. ZR2020MA073). References Building better batteriesStructural Reorganization–Based Nanomaterials as Anodes for Lithium‐Ion Batteries: Design, Preparation, and PerformanceEmbedding and Publishing Interactive, 3-Dimensional, Scientific Figures in Portable Document Format (PDF) FilesRecent Developments on and Prospects for Electrode Materials with Hierarchical Structures for Lithium-Ion BatteriesFeasible engineering of cathode electrolyte interphase enables the profoundly improved electrochemical properties in dual-ion batteryTemperature-Dependent Dynamic Properties of Li x Mn 2 O 4 in Monte Carlo SimulationsFe, N co-doped amorphous carbon as efficient electrode materials for fast and stable Na/K-storageAdvanced Matrixes for Binder‐Free Nanostructured Electrodes in Lithium‐Ion BatteriesNanomaterials for Rechargeable Lithium BatteriesChallenges for Rechargeable Li BatteriesLi-alloy based anode materials for Li secondary batteriesLi-Ions Transport Promoting and Highly Stable Solid–Electrolyte Interface on Si in Multilayer Si/C through Thickness ControlAnodes for lithium batteries: tin revisitedEnhanced Electrochemical Performance of Fe 0.74 Sn 5 @Reduced Graphene Oxide Nanocomposite Anodes for Both Li-Ion and Na-Ion BatteriesMetallic Sn-Based Anode Materials: Application in High-Performance Lithium-Ion and Sodium-Ion BatteriesNovel insight toward engineering of arrayed Cu@Sn nanoelectrodes: Rational microstructure refinement and its remarkable “harvesting effect” on lithium storage capabilityEncapsulating Sn Nanoparticles in Amorphous Carbon Nanotubes for Enhanced Lithium Storage PropertiesHigh-performing Sn–Co nanowire electrodes as anodes for lithium-ion batteriesAntimonides (FeSb2, CrSb2) with orthorhombic structure and their nanocomposites for rechargeable Li-ion batteriesA Chronicle Review of Nonsilicon (Sn, Sb, Ge)‐Based Lithium/Sodium‐Ion Battery Alloying AnodesNi-Sn-based hybrid composite anodes for high-performance lithium-ion batteriesDouble-Network Nanostructured Hydrogel-Derived Ultrafine Sn–Fe Alloy in Three-Dimensional Carbon Framework for Enhanced Lithium StorageExploring Anomalous Charge Storage in Anode Materials for Next-Generation Li Rechargeable BatteriesMagnetic Field–Suppressed Lithium Dendrite Growth for Stable Lithium‐Metal BatteriesStable lithium electrodeposition in liquid and nanoporous solid electrolytesQuantitative Analysis of the Initial Restructuring Step of Nanostructured FeSn 2 -Based Anodes for Li-Ion BatteriesElectrochemical impedance characterization of FeSn2 electrodes for Li-ion batteriesThe Anode Challenge for Lithium‐Ion Batteries: A Mechanochemically Synthesized Sn–Fe–C Composite Anode Surpasses Graphitic CarbonMechanically Alloyed Sn‐Fe(‐C) Powders as Anode Materials for Li‐Ion Batteries: I. The Sn2Fe ‐ C SystemElectron Paramagnetic Resonance, X-ray Diffraction, Mössbauer Spectroscopy, and Electrochemical Studies on Nanocrystalline FeSn 2 Obtained by Reduction of Salts in Tetraethylene GlycolReaction Mechanism of the Sn 2 Fe Anode in Lithium-Ion BatteriesVisualizing Battery Reactions and Processes by Using In Situ and In Operando MicroscopiesReal-time monitoring of stress development during electrochemical cycling of electrode materials for Li-ion batteries: overview and perspectivesIn Situ Electrochemistry of Rechargeable Battery Materials: Status Report and PerspectivesIn Situ Flow MAS NMR Spectroscopy and Synchrotron PDF Analyses of the Local Response of the Brønsted Acidic Site in SAPO-34 during Hydration at Elevated TemperaturesOperando magnetometry on Li x CoO2 during charging/dischargingCharging of lithium cobalt oxide battery cathodes studied by means of magnetometryExtra storage capacity in transition metal oxide lithium-ion batteries revealed by in situ magnetometryPVD amorphous carbon coated 3D NiCo2O4 on carbon cloth as flexible electrode for both sodium and lithium storageXPS and ToF-SIMS study of Sn–Co alloy thin films as anode for lithium ion batteryA novel stabilized carbon-coated nZVI as heterogeneous persulfate catalyst for enhanced degradation of 4-chlorophenolRobust longitudinal spin-Seebeck effect in Bi-YIG thin filmsAmorphous CoSnO 3 @C nanoboxes with superior lithium storage capabilityOperando electron magnetic measurements of Li-ion batteriesMagnetic Properties of Nanostructured MaterialsSynthesis, Properties, and Applications of Iron NanoparticlesConversion Reaction Mechanisms in Lithium Ion Batteries: Study of the Binary Metal Fluoride ElectrodesRemarkably Improved Electrode Performance of Bulk MnS by Forming a Solid Solution with FeS - Understanding the Li Storage MechanismMechanically Alloyed Sn‐Fe(‐C) Powders as Anode Materials for Li‐Ion Batteries: II. The Sn‐Fe SystemXPS and ToF-SIMS Study of Electrode Processes on Sn−Ni Alloy Anodes for Li-Ion BatteriesIn situ 119 Sn Mössbauer Effect Study of Li−CoSn 2 Electrochemical SystemFirst Principles Study on Na x Li 1− x FePO 4 As Cathode Material for Rechargeable Lithium BatteriesLi-ionic control of magnetism through spin capacitance and conversionOperando Magnetometry Probing the Charge Storage Mechanism of CoO Lithium‐Ion BatteriesReacquainting the Electrochemical Conversion Mechanism of FeS 2 Sodium-Ion Batteries by Operando Magnetometry
[1] Armand M and Tarascon J M 2008 Nature 451 652
[2] Han Y et al. 2020 Small 16 1902841
[3] Barnes D G et al. 2013 PLOS ONE 8 e69446
[4] Zhou L M et al. 2018 Adv. Energy Mater. 8 1701415
[5] Li W H et al. 2020 J. Energy Chem. 50 416
[6] Ouyang C et al. 2005 Chin. Phys. Lett. 22 489
[7] Liu Y et al. 2021 Electrochim. Acta 396 139265
[8] Zhang L et al. 2020 Adv. Mater. 32 1908445
[9] Bruce P G et al. 2008 Angew. Chem. Int. Ed. Engl. 47 2930
[10] Goodenough J B and Kim Y 2010 Chem. Mater. 22 587
[11] Park C M et al. 2010 Chem. Soc. Rev. 39 3115
[12] Zhao Y et al. 2019 ACS Nano 13 5602
[13] Yang S et al. 2003 Electrochem. Commun. 5 587
[14] Xin F X et al. 2015 ACS Appl. Mater. & Interfaces 7 7912
[15] Ying H and Han W Q 2017 Adv. Sci. 4 1700298
[16] Wu X et al. 2016 J. Power Sources 307 753
[17] Zhou X et al. 2016 Adv. Energy Mater. 6 1601177
[18] Ferrara G et al. 2012 J. Power Sources 211 103
[19] Park C and Sohn H 2010 Electrochim. Acta 17 4987
[20] Liang S et al. 2020 Small Methods 4 2000218
[21] Nguyen T L et al. 2018 Electrochim. Acta 278 25
[22] Shi H et al. 2018 Nano Lett. 18 3193
[23]Sony's new nexelion hybrid lithium ion bateries to have thirty-percent more capacity than conventional offering 2005 https://www.sony.com/en/SonyInfo/News/Press/200502/05-006E/
[24] Kim H et al. 2020 Chem. Rev. 120 6934
[25] Shen K et al. 2019 Adv. Energy Mater. 9 1900260
[26] Lu Y et al. 2014 Nat. Mater. 13 961
[27] Chamas M et al. 2013 Chem. Mater. 25 2410
[28] Chamas M et al. 2011 Electrochim. Acta 56 6732
[29] Dong Z et al. 2016 Adv. Sci. 3 1500229
[30] Mao O et al. 2019 J. Electrochem. Soc. 146 405
[31] Nwokeke U G et al. 2010 Chem. Mater. 22 2268
[32] Dong Z et al. 2019 ACS Omega 4 22345
[33] Wu Y and Liu N 2018 Chem 4 438
[34] Jangid M K and Mukhopadhyay A 2019 J. Mater. Chem. A 7 23679
[35] Yang Y et al. 2017 Adv. Mater. 29 1606922
[36] Kalantzopoulos G N et al. 2018 ChemPhysChem 19 519
[37] Topolovec S et al. 2016 J. Solid State Electrochem. 20 1491
[38] Klinser G et al. 2016 Solid State Ionics 293 64
[39] Li Q et al. 2021 Nat. Mater. 20 76
[40] Wang K et al. 2017 Carbon 125 375
[41] Li J T et al. 2010 J. Power Sources 195 8251
[42] Li S et al. 2020 Environ. Int. 138 105639
[43] Siegel G et al. 2015 Sci. Rep. 4 4429
[44] Wang Z et al. 2013 Energy & Environ. Sci. 6 87
[45] Gershinsky G et al. 2014 Energy & Environ. Sci. 7 2012
[46] Diandra L L and Rieke R D 1996 Chem. Mater. 8 1770
[47] Huber D L 2005 Small 1 482
[48] Wang F et al. 2011 J. Am. Chem. Soc. 133 18828
[49] Zhao L et al. 2014 Adv. Funct. Mater. 24 5557
[50] Mao O and Dahna J 1999 J. Electrochem. Soc. 146 414
[51] Li J T et al. 2011 J. Phys. Chem. C 115 7012
[52] Ionica-Bousquet C M et al. 2006 Chem. Mater. 18 6442
[53] Ouyang C Y et al. 2006 Chin. Phys. Lett. 23 61
[54] Zhang F et al. 2021 Matter 4 3605
[55] Li H et al. 2021 Adv. Mater. 33 2006629
[56] Li Z et al. 2021 J. Am. Chem. Soc. 143 12800