Elucidation of the Co⁴⁺ state with strong charge-transfer effects in charged LiCoO₂ by resonant soft X-ray emission spectroscopy at the Co L₃ edge

Abstract

To understand the electronic-structure change of LiCoO₂, a widely used cathode material in Li-ion batteries, during charge–discharge, we performed ex situ soft X-ray absorption spectroscopy (XAS) and resonant soft X-ray emission spectroscopy (RXES) of the Co L₃ edge in combination with charge-transfer multiplet calculations. The RXES profile significantly changed for the charged state at 4.2 V vs. Li/Li⁺, corresponding to the oxidation reaction from a Co³⁺ low-spin state for the initial state, while the XAS profile exhibited small changes. For the 4.2-V charged state, we confirmed that approximately half of the initial Co³⁺ ions were oxidized to Co⁴⁺ ions. The multiplet calculation of the RXES results revealed that the Co⁴⁺ state has a negative charge-transfer energy and the d⁶L̲ state (L̲ is a ligand hole) is the most stable. Therefore, the O 2p hole created by the strong charge-transfer effect plays a major role in the redox reaction of LiCoO₂.

---

1. Introduction

In the past few decades, Li-ion batteries (LIBs) have been extensively developed for various applications. To apply LIBs to electric vehicles and large-scale energy-storage systems, their energy density and safety need to be improved. Increasing the charge–discharge capacity of the cathodes of LIBs is particularly important for increasing the energy density. The charge–discharge capacity is directly connected with the redox reaction. In the case of LiCoO₂, which is a typical cathode material, the formal valence of Co changes according to the equation LiCo³⁺O₂ ⇔ 0.5Li⁺ + 0.5e⁻ + Li_0.5Co⁴⁺_0.5Co³⁺_0.5O₂ with charge (Li extraction)/discharge (Li insertion) within a cutoff voltage range of 3.0–4.2 V (vs. Li/Li⁺), where the capacity is approximately 150 mA h g⁻¹.^1,2 To confirm the Co³⁺/Co⁴⁺ redox reaction, many groups have investigated the electronic structure of LiCoO₂ by various methods.^1–9 X-ray absorption spectroscopy (XAS) is one of the most powerful tools to analyze the electronic structure of cathode materials. In the case of 3d transition-metal (TM) oxides (TMOs) involving LiCoO₂, TM K-edge (mainly the 1s → 4p transition) and L_2,3-edge XAS (2p_1/2,3/2 → 3d transition) have been extensively performed. For example, the Co³⁺ low-spin state has been confirmed to be the initial state of LiCoO₂ by Co K-edge and L_2,3-edge XAS.^8–16 For the charge process of LiCoO₂, Co K-edge XAS studies have shown that the main peak shifts to the high-energy side, and Co L_2,3-edge XAS studies have revealed that the XAS line shape changes, which suggest the oxidation reaction from the Co³⁺ state to the Co⁴⁺ state.^15,16

In this study, we focus on the Co⁴⁺ state in the charge process because the electronic configuration and Co 3d–O 2p orbital hybridization of the Co–O bond have been unclear. Resonant soft X-ray emission spectroscopy (RXES) combined with theoretical charge-transfer (CT) multiplet calculations is a promising method to investigate the Co 3d electronic structure. RXES is the secondary process after XAS, reflecting the element and orbital-selective occupied (valence) state.^17–19 Recently, RXES has attracted considerable interest for studying electrode materials,^20–22 including Li-rich layered oxides with oxygen redox.^23–29 For LiCoO₂, the oxygen redox reaction has been investigated by O K RXES,^30,31 but the reaction occurs at voltage around 4.6 V vs. Li/Li⁺, which is higher than the voltage range in which the Co³⁺/Co⁴⁺ redox occurs in practical applications (below ∼4.2 V vs. Li/Li⁺). Here, we report a Co L₃ (2p–3d–2p resonance) ex situ RXES study of LiCoO₂ in three states (the initial state, the 4.2-V charged state, and the discharged state after 4.2-V charging) to clarify the intrinsic role of the O 2p–Co 3d CT effect in the redox reaction below 4.2 V.

2. Method

LiCoO₂ powder was fabricated by following the procedure below. First, CoCl₂·6H₂O (0.0025 mol) and LiNO₃ (0.020 mol) were dissolved in H₂O (10 ml). The solution was then dried at 140 °C in vacuum for 17 h. The dried reactant was placed in a Al₂O₃ crucible with a lid. The crucible was placed in a muffle furnace and heated at 800 °C in air for 3 h. Finally, the sample was washed with deionized water and dried under vacuum conditions. The LiCoO₂ powder was pasted with super-P Li as a carbon additive to increase the conductivity and polytetrafluoroethylene as a binder. The paste was then pressed onto Al mesh as a current collector to fabricate the working electrode. A three-electrode beaker cell was assembled with the LiCoO₂ working electrode, a Li–metal counter electrode, a Li–metal reference electrode, and a 1 M LiClO₄/ethylene carbonate–diethyl carbonate electrolyte solution. The charge/discharge experiments of the beaker cells were performed by cyclic voltammetry (CV). The scan speed was set to 0.5 mV s⁻¹, and the cutoff voltages were 4.2 V (vs. Li/Li⁺) for charging and 3.0 V for discharging. The charged/discharged LiCoO₂ samples were removed from the cell in a glove box, and they were then transferred to a vacuum chamber using a transfer vessel without exposure to air.

The XAS measurements were performed at BL07LSU of SPring-8³² using the surface-sensitive total-electron yield (TEY) mode. The RXES experiments were performed using the ultrahigh-resolution X-ray emission spectrometer HORNET at BL07LSU of SPring-8.³³ The energy resolution of the incident beam was 100 meV, and the total energy resolution for the RXES measurements was ΔE = 330 meV at 785.0 eV. All of the XAS and RXES measurements were performed at room temperature. We also performed spectral analyses using Quanty software programs based on CT multiplet calculation.^34,35 In the calculation, the Co 3d orbitals with the Co 2p core–hole effect and O 2p orbitals in a CoO₆ octahedron were considered. To describe the ligand-to-metal CT, configuration interactions were used as a linear combination of the basis of each electron configuration with the CT effect (dⁿ, dⁿ⁺¹L̲, and dⁿ⁺²L̲², where L̲ denotes a ligand hole).

3. Results and discussion

The CV curves of LiCoO₂ are shown in Fig. 1. The voltage range was set to 3.0–4.2 V (vs. Li/Li⁺). The open-circuit voltage before cycling was 3.2 V. A broad structure was observed around 4.1 V in the charge process. This indicates the oxidation reaction due to Li extraction from LiCoO₂. For the discharge process, the reduction peak was observed at 3.8 V owing to reinsertion of Li.

Fig. 1 CV curves of LiCoO₂ with a scan speed of 0.5 mV s⁻¹.

To investigate the unoccupied Co 3d states and select the incident photon energy (excitation energy) for RXES, TEY XAS measurements were performed. The Co L₃-edge TEY XAS results for LiCoO₂ are shown in Fig. 2. According to the XAS line profiles in previous reports,^8,9,14 the initial and discharged states are attributed to the Co³⁺ state, suggesting a reversible redox reaction for the Co 3d state. The small multiplet splitting for the initial and discharged states suggests that the e_g orbital is almost unoccupied in the Co³⁺ (d⁶) low-spin (LS) state.¹⁴ The LS Co³⁺ has occupied six t_2g orbitals (which should be further split into and a_1g orbitals in the lower symmetry as reported in ref. 14) and all the e_g orbitals are empty.

Fig. 2 Ex situ Co L₃-edge TEY XAS spectra corresponding to the three states in Fig. 1. The dotted line indicates the excitation energy (779.5 eV) for the RXES measurements.

The main peak position in the 4.2-V charged-state spectrum slightly shifted to the higher energy region and the shoulder structures at 776.8–778.4 and 781.0–782.0 eV were enhanced (indicated by the arrows in Fig. 2). These results are similar to those in previous reports.^8–14 The changes for the 4.2-V charged state indicate the appearance of the Co⁴⁺ state. The slight shift of the main peak suggests oxidation of Co, which should be accompanied with the slight increase of the structure at 781.0–782.0 eV. The increase of the shoulder structure at 776.8–778.4 eV is attributed to t_2g hole (on the a_1g orbital in the lower symmetry) created by charge.¹⁴ Moreover, it is expected that approximately half of the Co³⁺ sites remained (i.e., approximately 0.5Li was extracted) at the upper cutoff voltage of 4.2 V, which is usual value for practical use to maintain the host framework.^1,2

To further investigate the Co³⁺ state in the initial state and determine how it changes by the charge process, the excitation energy for RXES was selected to be 779.5 eV, which is the Co L₃-edge main peak position for the initial and discharged states (indicated by the dashed line in Fig. 2). The same excitation energy was used for the charged sample.

The RXES results at the Co L₃ edge are shown in Fig. 3. The peak at 779.5 eV corresponds to the elastic scattering where the emission energy equals the excitation energy. The sharp peaks from 775.0 to 779.0 eV for the initial state are related to the dd excitations. The broad structure below 775.0 eV is mainly related to the CT excitation between the Co 3d and O 2p orbitals. In addition, some fluorescence components should be involved for the whole range. These features for the initial state are almost the same as those for the discharged state, suggesting a reversible redox reaction of Co, as for the XAS results. The line shape for the 4.2-V charged state is greatly different from those for the initial and discharged states. Large changes in the Co 3d electronic structure by charging were observed for Co L₃ RXES, while Co L₃-edge XAS showed small changes. In particular, the three dd-excitation peaks observed in the initial state greatly changed in the 4.2-V charged state. These changes can be attributed to the partial oxidation of the Co³⁺ state to the Co⁴⁺ state.

Fig. 3 Ex situ RXES results at the Co L₃ edge of LiCoO₂ for the three states shown in Fig. 1.

To analyze the RXES spectra, we performed multiplet calculations including CT effects using the Quanty program. The RXES spectrum with 779.5-eV excitation includes a fluorescence component that was not included in the calculation. For the initial state, we also measured the RXES spectrum with 793.8-eV excitation corresponding to the post-edge region of the Co L₃ edge. The RXES spectrum with 793.8-eV excitation (on-resonance for the L₂ edge) below ∼780 eV in the emission-energy scale is regarded as the off-resonant spectrum for the L₃ edge. We subtracted the off-resonant spectrum from the on-resonant 779.5-eV RXES spectrum to remove the fluorescence. The subtracted spectrum (spectrum A) is compared with the Co³⁺ CT multiplet calculation with a CoO₆ cluster model in Fig. 4(a). The calculated result well reproduces the experiment, except for slight differences of the peak intensities of the dd excitations around 778 eV. In addition, the elastic scattering at 779.5 eV in the experimental spectrum is not reproduced well, because it depends on the experimental settings, including the polarization of the incident and emitted photons. The electronic-structure parameters used in the calculation are summarized in Table 1. For the calculation, O_h symmetry was considered for simplicity, while it has been reported that the CoO₆ octahedron has trigonal D_3d symmetry slightly deforming compared to the O_h symmetry.³⁶ The crystal-field splitting 10 Dq value of 1.4 eV is sufficiently strong to maintain the d⁶ LS state^37,38 considering the Tanabe–Sugano diagram for this configuration. The CT energy (Δ), corresponding to the energy difference of E(d⁷L̲) − E(d⁶) (L̲ is a ligand hole), is 0.6 eV, resulting in a strong CT effect. Indeed, the calculated weight of charge-transferred d⁷L̲ configuration (52.1%) is higher than that of d⁶ configuration (39.0%) as shown in Table 2.

Fig. 4 (a) Subtraction of the off-resonant spectrum from the on resonant spectrum for the initial state and fitting result of the Co³⁺ state. (b) Subtraction of the off-resonant spectrum from the on resonant spectrum for the charged state. (c) Difference analysis for the charged state and fitting result of the Co⁴⁺ state.

Table 1 Electronic-structure parameters used in the calculations (in eV)

	Co^3+ (Oh symmetry)	Co^4+ (Oh symmetry)
10 Dq	1.4	1.6
Δ	0.6	−1.0
U dd	6.5	7.0
Q	7.5	7.5
V(eg)	2.2	3.0
V(t2g)	1.1	1.5

---

Table 2 Calculated weights of each electron configuration and calculated effective 3d electron numbers 〈n_d〉

Co^3+ (Oh symmetry)		Co^4+ (Oh symmetry)
d^6	39.0%	d^5	27.3%
d^7L̲	52.1%	d^6L̲	57.1%
d^8L̲^2	8.9%	d^7L̲^2	15.6%
〈nd〉	6.7	〈nd〉	5.9

---

For the 4.2-V charged state, to extract the Co⁴⁺ component, we performed a difference analysis. Firstly, we subtracted the off-resonant spectrum for the charged state from the on-resonant 779.5-eV RXES spectrum in the same way for the initial state, which is spectrum B in Fig. 4(b). Next, a difference spectrum (spectrum C) was constructed using the formula of [spectrum B (charged)] − 0.5 × [spectrum A (initial)] (Fig. 4(c)). Here, we assumed that the Co³⁺ state was unchanged between the initial and charged states. The coefficient of 0.5 in the formula is an assumption based on the nominal chemical equation of LiCo³⁺O₂ ⇔ 0.5Li⁺ + 0.5e⁻ + Li_0.5Co^3.5+O₂, while it is difficult to accurately estimate the ratio of Co³⁺/Co⁴⁺ from the intensity of RXES spectra.

The difference in line shape between spectrum C and spectrum A suggests the oxidation reaction of the Co³⁺ LS to Co⁴⁺ LS states. The shoulder structure of the 777.9-eV peak in the dd excitation region of spectrum C should correspond to the decrease of the 778.3-eV peak in spectrum A, suggesting the removal of an electron from the fully occupied t_2g band of the Co³⁺ LS state in the initial state as Fig. 5 shows. Spectrum C was successfully fitted by the Co⁴⁺ calculation for the dd and CT excitations (Fig. 4(c)). Indeed, the Co⁴⁺ calculation reproduces the decrease of the 777.9-eV peak. The electronic-structure parameters for the Co⁴⁺ calculation are summarized in Table 1. For Co⁴⁺, Δ is defined as E(d⁶L̲) − E(d⁵). The Δ value was determined to be −1.0 eV. The negative Δ means that the charge-transferred d⁶L̲ configuration is considerably dominant compared to the d⁵ configuration. Negative Δ in Co⁴⁺ configuration has been also found for SrCoO₃.³⁹ Indeed, the calculated weight of d⁵ configuration is as low as 27.3%, while that of charge-transferred d⁶L̲ configuration is calculated to be 57.1% (Table 2). The calculated effective 3d electron number 〈n_d〉 for Co⁴⁺ is 5.9, which is much larger than 5.0 (Table 2). Moreover, the calculated transfer integral between the Co 3d and O 2p orbitals V(e_g) is 3.0 eV, which suggests that the hybridization of the σ bonding is not weak, and V(t_2g) for the π bonding is 1.5 eV. Therefore, the CT effect of the Co⁴⁺ state is further stronger than that of the Co³⁺ state. These results indicate that the O 2p orbital plays an essential role in the redox reaction. The holes created by the charge reaction tend to be located on the p–d hybridized orbital. In that the holes are not localized on the Co 3d orbital, the consideration is partly consistent with previous XAS,⁸ X-ray Compton scattering,⁷ and theoretical studies⁴⁰ which suggest that the hole should be on the O 2p orbital rather than the Co 3d orbital. We conclude that the O 2p–Co 3d hybridized orbital is redox active for LiCoO₂.

Fig. 5 Schematic images of dd excitation for the Co³⁺ and Co⁴⁺ states. For simplicity, O_h symmetry is considered. The yellow and red allows indicate the absorption and emission processes. The 779.5-eV excitation of RXES in Fig. 3 and 4 corresponds to the excitation (absorption) to only the e_g band for the Co⁴⁺ state.

A similar mechanism has been reported in Mn L-edge XAS and RXES studies of LiMn₂O₄ with the Mn³⁺/Mn⁴⁺ redox couple^22,41 and Ni L-edge XAS studies of NaFe_0.5Ni_0.5O₂ with the Ni³⁺/Ni⁴⁺ redox couple for the Na-ion insertion/extraction.⁴² In general, the p–d hybridization in TMOs is particularly important for the physical properties.^43,44 The p–d hybridization in TMOs tends to become stronger for higher oxidation states, such as the tetravalent state.⁴⁵ Thus, the redox reaction in TMO-based cathode materials should be considered for the p–d hybridization in the [TM]O₆ octahedron rather than the localized 3d orbital of the TM. The situation is different from that for polyanion-type LiFePO₄, in which the localized Fe 3d orbital in the FeO₆ octahedron mainly contribute to the redox reaction.²⁰ LiFePO₄ is also a prototypical cathode material, but it should be separately classified from TMO-based cathode materials such as LiCoO₂ and LiMn₂O₄ in terms of the electronic structure and redox reaction.

The calculated spectra for both the Co³⁺ and Co⁴⁺ states agreed the experimental RXES spectra (Fig. 4), even though the symmetry in the calculation was simplified to O_h symmetry. One reason for the agreement is that the degree of deformation of the CoO₆ octahedron from O_h symmetry is not large. Another reason is the strong CT effects and strong p–d hybridization. It is most likely that the interaction of the Co–O bond should be very strong, resulting in the symmetry effect not being significant on the RXES spectra. In the near future, we will perform a similar calculation with lower symmetry to further verify the symmetry effect.

4. Conclusions

We have performed ex situ RXES at the Co L₃ edge of LiCoO₂ in combination with multiplet calculation and succeeded in separately clarifying the valence states of Co³⁺ and Co⁴⁺. Both the Co³⁺ and Co⁴⁺ states show strong CT effects from the O 2p to Co 3d orbitals. The Co⁴⁺ state in the charged state has a negative CT energy, which means that the charge-transferred d⁶L̲ configuration is dominant, resulting in an effective 3d-electron number 〈n_d〉 of 5.9. Similar scenarios have been suggested in previous reports for other typical oxide-based cathode materials like LiMn₂O₄. The higher valence state in those oxides, such as the tetravalent state, commonly have strong CT effects. Thus, present and those results indicate that the O 2p orbital plays an essential role in the redox reaction through the strong p–d hybridization in oxide-based cathode materials. In addition, RXES in combination with difference analysis and multiplet calculation is a very powerful method to distinguish a specific oxidation state from mixed valence states. Extracting the higher valence state in the charged state of cathode materials is particularly important to understand the redox reaction. For example, control of the CT effect by some means, such as element substitution based on such electronic-structure analysis using RXES, will be effective to further improve the electrode performance.

Author contributions

D. A., E. H. and A. Y. conceived and directed the project. D. A. and E. H. synthesized and characterized LiCoO₂. D. A., H. N., H. M. and E. H. constructed the operando cell. T. S., Y. N., R. W. and F. D. G. performed the multiplet calculations. D. A., T. S., H. K., K. Y., J. M., and Y. H. performed the soft X-ray absorption/emission spectroscopy. All of the authors wrote the manuscript.

Data availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was supported by the Ministry of Education, Culture, Sports, Science and Technology (MEXT), Japan; Grant-in-Aid for Specially Promoted Research (No. 15H05701). Part of this work was conducted on the basis of the International Joint Research Program for Innovative Energy Technology by the Ministry of Economy, Trade and Industry (METI), Japan. The XAS and RXES measurements at BL07LSU of SPring-8 were performed by the joint research in SRRO and ISSP, the University of Tokyo (Proposal No. 2017A7531, 2018B7588, 2019B7457, 2020A7477, 2021A7494, 2021A7495, 2021B7429, and 2022A7447).

References

1. K. Mizushima, P. C. Jones, P. J. Wiseman and J. B. Goodenough, Mat. Res. Bull., 1980, 15, 783–789.
2. K. Mizushima, P. C. Jones, P. J. Wiseman and J. B. Goodenough, Solid State Ionics, 1981, 3–4, 171–174.
3. J. N. Reimers and J. R. Dahn, J. Elecrochem. Soc., 1992, 139, 2091–2097.
4. J. van Elp, J. L. Wieland, H. Eskes, P. Kuiper, G. A. Sawatzky, F. M. F. de Groot and T. S. Turner, Phys. Rev. B: Condens. Matter Mater. Phys., 1991, 44, 6090–6103.
5. J. Sugiyama, H. Nozaki, J. H. Brewer, E. J. Ansaldo, G. D. Morris and C. Delmas, Phys. Rev. B: Condens. Matter Mater. Phys., 2005, 72, 144424.
6. M. Okubo, J. Kim, T. Kudo, H. S. Zhou and I. Honma, J. Phys. Chem. C, 2009, 113, 15337–15342.
7. B. Barbiellini, K. Suzuki, Y. Orikasa, S. Kaprzyk, M. Itou, K. Yamamoto, Y. J. Wang, H. Hafiz, R. Yamada, Y. Uchimoto, A. Bansil, Y. Sakurai and H. Sakurai, Appl. Phys. Lett., 2016, 109, 073102.
8. Y. Uchimoto, H. Sawada and T. Yao, J. Synchrotron Radiat., 2001, 8, 872–873.
9. W.-S. Yoon, K.-B. Kim, M.-G. Kim, M.-K. Lee, H.-J. Shin, J.-M. Lee, J.-S. Lee and C.-H. Yo, J. Phys. Chem. B, 2002, 106, 2526–2532.
10. V. R. Galakhov, E. Z. Kurmaev, S. Uhlenbrock, M. Neumann, D. G. Kellerman and V. S. Gorshkov, Solid State Commun., 1996, 99, 221–224.
11. V. R. Galakhov, M. Neumann and D. G. Kellerman, Appl. Phys. A: Mater. Sci. Process., 2009, 94, 497–500.
12. V. R. Galakhov, N. A. Ovechkina, A. S. Shkvarin, S. N. Shamin, E. Z. Kurmaev, K. Kuepper, A. F. Takács, M. Raekers, S. Robin, M. Neumann, G.-N. Gavrila, A. S. Semenova, D. G. Kellerman, T. Käämbre and J. Nordgren, Phys. Rev. B: Condens. Matter Mater. Phys., 2006, 74, 045120.
13. D. Ensling, A. Thissen, S. Laubach, P. C. Schmidt and W. Jaegermann, Phys. Rev. B: Condens. Matter Mater. Phys., 2011, 82, 195431.
14. T. Mizokawa, Y. Wakisaka, T. Sudayama, C. Iwai, K. Miyoshi, J. Takeuchi, H. Wadati, D. G. Hawthorn, T. Z. Regier and G. A. Sawatzky, Phys. Rev. Lett., 2013, 111, 056404.
15. C. J. Partridge, C. T. Love, K. E. Swider-Lyons, M. E. Twigg and D. E. Ramaker, J. Solid State Chem., 2013, 203, 134–144.
16. J. M. Rosolen, P. Ballirano, M. Berrettoni, F. Decker and M. Gregorkiewitz, Ionics, 1997, 3, 345–354.
17. A. Kotani and S. Shin, Rev. Mod. Phys., 2001, 73, 203–246.
18. L. J. Ament, M. van Veenendaal, T. P. Devereaux, J. P. Hill and J. van den Brink, Rev. Mod. Phys., 2011, 83, 705–767.
19. F. de Groot and A. Kotani, Core level spectroscopy of solids, CRC Press, Boca Raton, Florida, 2008.
20. D. Asakura, Y. Nanba, Y. Makinose, H. Matsuda and E. Hosono, ChemPhysChem, 2018, 19, 988–992.
21. D. Asakura, Y. Nanba, Y. Makinose, H. Matsuda and E. Hosono, Phys. Chem. Chem. Phys., 2017, 19, 16507–16511.
22. D. Asakura, Y. Nanba, E. Hosono, M. Okubo, H. Niwa, H. Kiuchi, J. Miyawaki and Y. Harada, Phys. Chem. Chem. Phys., 2019, 21, 18363–18369.
23. T. Sudayama, K. Uehara, T. Mukai, D. Asakura, X.-M. Shi, A. Tsuchimoto, B. Mortemard de Boisse, T. Shimada, E. Watanabe, Y. Harada, M. Nakayama, M. Okubo and A. Yamada, Energy Environ. Sci., 2020, 13, 1492–1500.
24. W. E. Gent, K. Lim, Y. F. Liang, Q. H. Li, T. Barnes, S. J. Ahn, K. H. Stone, M. Mclntire, J. Y. Hong, J. H. Song, Y. Y. Li, A. Mehta, S. Ermon, T. Tyliszczak, D. Kilcoyne, D. Vine, J. H. Park, S. K. Doo, M. F. Toney, W. L. Yang, D. Prendergast and W. C. Chueh, Nat. Commun., 2017, 8, 2091.
25. K. Luo, M. R. Roberts, R. Hao, N. Guerrini, D. M. Pickup, Y.-S. Liu, K. Edström, J.-H. Guo, A. V. Chadwick, L. C. Duda and P. G. Bruce, Nat. Chem., 2016, 8, 684–691.
26. R. A. House, U. Maitra, M. A. Pérez-Osorio, J. G. Lozano, L. Jin, J. W. Somerville, L. C. Duda, A. Nag, A. Walters, K.-J. Zhou, M. R. Roberts and P. G. Bruce, Nature, 2020, 577, 502–508.
27. R. A. House, G. J. Rees, M. A. Pérez-Osorio, J. G. Lozano, J.-J. Marie, E. Boivin, A. W. Robertson, A. Nag, M. Garcia-Fernandez, K.-J. Zhou and P. G. Bruce, Nat. Energy, 2020, 5, 777–785.
28. R. A. House, J.-J. Marie, M. A. Pérez-Osorio, G. J. Rees, E. Boivin and P. G. Bruce, Nat. Energy, 2021, 6, 781–789.
29. L. Massel, K. Hikima, H. Rensmo, K. Suzuki, M. Hirayama, C. Xu, R. Younesi, Y.-S. Liu, J.-H. Guo, R. Kanno, M. Hahlin and L.-C. Duda, J. Phys. Chem. C, 2019, 123, 28519–28526.
30. J.-N. Zhang, Q. Li, C. Ouyang, X. Yu, M. Ge, X. Huang, E. Hu, C. Ma, S. Li, R. Xiao, W. Yang, Y. Chu, Y. Liu, H. Yu, X.-Q. Yang, X. Huang, L. Chen and H. Li, Nat. Energy, 2019, 4, 594–603.
31. E. Hu, Q. Li, X. Wang, F. Meng, J. Liu, J.-N. Zhang, K. Page, W. Xu, L. Gu, R. Xiao, X. Huang, L. Chen, W. Yang, X. Yu and X.-Q. Yang, Joule, 2021, 5, 720–736.
32. S. Yamamoto, Y. Senba, T. Tanaka, H. Ohashi, T. Hirono, H. Kimura, M. Fujisawa, J. Miyawaki, A. Harasawa, T. Seike, S. Takahashi, N. Nariyama, T. Matsushita, M. Takeuchi, T. Ohata, Y. Furukawa, K. Takeshita, S. Goto, Y. Harada, S. Shin, H. Kitamura, A. Kakizaki, M. Oshima and I. Mastuda, J. Synchrotron Radiat., 2014, 21, 352–365.
33. Y. Harada, M. Kobayashi, H. Niwa, Y. Senba, H. Ohashi, T. Tokushima, Y. Horikawa, S. Shin and M. Oshima, Rev. Sci. Instrum., 2012, 83, 013116.
34. M. W. Haverkort, M. Zwierzcki and O. K. Andersen, Phys. Rev. B: Condens. Matter Mater. Phys., 2012, 85, 165113.
35. R.-P. Wang, B. Liu, R. J. Green, M. U. Delgado-Jaime, M. Ghiasi, T. Schmitt, M. M. van Schooneveld and F. M. F. de Groot, J. Phys. Chem. C, 2017, 121, 24919–24928.
36. H.-J. Kim, Y. U. Jeong, J.-H. Lee and J.-J. Kim, J. Power Sources, 2006, 159, 233–236.
37. D. Qian, Y. Hinuma, H. Chen, L.-S. Du, K. J. Carroll, G. Ceder, C. P. Grey and Y. S. Meng, J. Am. Chem. Soc., 2012, 134, 6096–6099.
38. L. A. Montoro, M. Abbate and J. M. Rosolen, Electrochem. Solid State Lett., 2000, 3, 410–412.
39. R. H. Potze, G. A. Sawatzky and M. Abbate, Phys. Rev. B: Condens. Matter Mater. Phys., 1995, 51, 11501–11506.
40. Y. Yamauchi and H. Nakai, J. Electrochem. Soc., 2013, 160, A1364–A1368.
41. D. Asakura, E. Hosono, H. Niwa, H. Kiuchi, J. Miyawaki, Y. Nanba, M. Okubo, H. Matsuda, H. S. Zhou, M. Oshima and Y. Harada, Electrochem. Commun., 2015, 50, 93–96.
42. Y. Nanba, T. Iwao, B. Mortemard de Boisse, W. Zao, E. Hosono, D. Asakura, H. Niwa, H. Kiuchi, J. Miyawaki, Y. Harada, M. Okubo and A. Yamada, Chem. Mater., 2016, 28, 1058–1065.
43. M. Imada, A. Fujimori and Y. Tokura, Rev. Mod. Phys., 1998, 70, 1039–1263.
44. A. Fujimori and F. Minami, Phys. Rev. B: Condens. Matter Mater. Phys., 1984, 30, 957–971.
45. K. Kawai, D. Asakura, S.-I. Nishimura and A. Yamada, Chem. Commun., 2019, 55, 13717–13720.

---

Footnotes

† Present address: Research Initiative for Supra-Materials, Shinshu University, Nagano, Nagano 380-8553, Japan.

‡ Present address: NanoTerasu Center, National Institutes for Quantum Science and Technology, Sendai, Miyagi 980-8579, Japan.

§ Present address: Deutsches Elektronen Synchrotron DESY, Hamburg 22607, Germany.

---