Additive regulating Li<sup>+</sup> solvation structure to construct dual LiF−rich electrode electrolyte interphases for sustaining 4.6 V Li||LiCoO<sub>2</sub> batteries

Citation: Xi Tang, Chunlei Zhu, Yulu Yang, Shihan Qi, Mengqiu Cai, Abdullah N. Alodhayb, Jianmin Ma. Additive regulating Li⁺ solvation structure to construct dual LiF−rich electrode electrolyte interphases for sustaining 4.6 V Li||LiCoO₂ batteries[J]. Chinese Chemical Letters, 2024, 35(12): 110014. doi: 10.1016/j.cclet.2024.110014 shu

Additive regulating Li⁺ solvation structure to construct dual LiF−rich electrode electrolyte interphases for sustaining 4.6 V Li||LiCoO₂ batteries

English

Additive regulating Li⁺ solvation structure to construct dual LiF−rich electrode electrolyte interphases for sustaining 4.6 V Li||LiCoO₂ batteries

a.
School of Physics and Electronics, Hunan University, Changsha 410082, China
b.
Department of Physics and Astronomy, College of Science, King Saud University, Riyadh 11451, Saudi Arabia
^* Corresponding author.
E-mail address: nanoelechem@hnu.edu.cn (J. Ma).
Received Date: 29 June 2023
Accepted Date: 14 May 2024
Revised Date: 08 April 2024
Available Online: 15 December 2024

Abstract: The battery energy density can be improved by raising the operating voltage, however, which may lead to rapid capacity decay due to the continuous electrolyte decomposition and the thickening of electrode electrolyte interphases. To address these challenges, we proposed tripropyl phosphate (TPP) as an additive−regulating Li⁺ solvation structure to construct a stable LiF–rich electrode carbonate−based electrolyte interphases for sustaining 4.6 V Li||LiCoO₂ batteries. This optimized interphases could help reduce the resistance and achieve better rate performance and cycling stability. As expected, the Li||LiCoO₂ battery retained 79.4% capacity after 100 cycles at 0.5 C, while the Li||Li symmetric cell also kept a stable plating/stripping process over 450 h at the current density of 1.0 mA/cm² with a deposited amount of 0.5 mAh/cm².

Key words:

As one of typic commercial cathode materials, LiCoO₂ has huge potential as high–capacity cathode for lithium metal batteries (LMBs) [1–3], especially coupled with Li anode [4–6], since only half of the theoretical capacity of LiCoO₂ can be delivered at the cut–off voltage of 4.25 V. Thus, the energy density of Li||LiCoO₂ batteries can be promoted by raising the cut–off voltage [7–10]. Unfortunately, traditional carbonates have strong binding ability with Li⁺ that are difficult to desolvation, which leads to slow reaction kinetics. Moreover, the strong coordination of EC and EMC with Li⁺ lead to serious side reactions with LiCoO₂, resulting in loose porous organic cathode electrolyte interphase (CEI), which is constantly broken/reconstructed during the charge/discharge process, resulting in continuous consumption of electrolyte and rapid capacity decay [11–13]. Moreover, the continuous accumulation of non–conducting and fragile solid electrolyte interphase (SEI) leads to poor interface stability and uncontrolled growth of Li dendrites, which reduce the capacity and coulomb efficiency (CE) [14–19]. Therefore, it is very crucial to promote cycling stability and rate capability by optimizing CEI and SEI in high–voltage Li||LiCoO₂ batteries [20–23].

Electrolyte additive strategy is demonstated to effectively improve CEI/SEI performance [24–28]. Among them, phosphates are well–known co–solvents and additives due to their non–flammability and film–forming properties [29–31]. Moreover, phosphate–based additives are also reported to improve the cycling performance of lithium–ion batteries at elevated temperatures by constructing superior electrode electrolyte interphases [32]. However, the efficient phosphate–based additives are rarely reported in high–voltage Li||LiCoO₂ batteries.

In this work, we proposed an additive regulating Li⁺ solvation structure strategy for constructing dual LiF–rich electrode electrolyte interphases to support 4.6 V Li||LiCoO₂ batteries. For the cathode, the well–formed CEI with inorganic substances restrains the harassing oxidative decomposition of solvents. For the anode, TPP promotes the formation of SEI enriched in fluoride/phosphide components, which facilitates uniform Li deposition and suppresses the formation of Li dendrites [33,34]. This work contributes to a deep comprehensive understanding of the specific working mechanism of phosphate additives.

To investigate the electrochemical stability and solvation effects of the phosphate–based additives, the theoretical prediction based on density functional theory (DFT) and molecular dynamics (MD) simulations were carried out (Fig. 1). Among the molecules, EMC and TPP have the highest HOMO and are easily oxidized while FEC and TPP exhibit the lowest LUMO and are easily reduced (Fig. 1a), which provides a thermodynamic basis for the growth of interphases. The MD simulation results in Figs. 1b-f show that the Li⁺ in the blank electrolyte is closely coordinated with EC and EMC, which are carried to the electrode surface for oxidation to form various loose and unstable organic by–products. As expected, the highly electronegative O atom of TPP coordinates with Li⁺ and occupies the solvated group, reducing the coordination number of anions and solvent molecules due to its large steric hindrance (Fig. S1 in Supporting information). Therefore, in the process of charging, more free PF₆⁻ are preferentially adsorbed on the surface of the positively charged LiCoO₂ under electrostatic action than the Li⁺−solvent complexes, realizing the anion−dominated inner Helmholtz plane (IHP) and the formation of inorganic species−rich CEI layer. At the same time, the solvation degree of weakly coordinated FEC increases with the influence of TPP, which is conducive to the formation of LiF−rich SEI.

Figure 1

Figure 1. Theoretical prediction of phosphate−based additives. (a) LUMO (lowest unoccupied molecular orbital) and HOMO (highest occupied molecular orbital) energy values of carbonate solvent molecules and TPP. Schematic diagrams of the ESP charge distribution in the solvated sheath of (b) blank electrolyte and (c) TPP–contained electrolyte. Molecular dynamics simulation of Li⁺ in different electrolytes, the cumulative coordination number (n(r)) of Li⁺ with (d) PF₆⁻ anion, (e) FEC molecule, and (f) EC molecule.

DownLoad: Full-Size Img PowerPoint

To build a deeper understanding of the phosphate−based additive, the surface chemical compositions of CEI films formed in different electrolytes were characterized by X−ray photoelectron spectroscopy (XPS). As shown in Fig. 2, Figs. S2 and S3 (Supporting information), the depth profile of the LiCoO₂ electrode was detected by Ar sputtering at 2 min and 4 min. The CEI films in all electrolytes contain organic components (C−F, C−O, C=O and C−C/C−H) and inorganic components (Li₂CO₃, LiF and LiPF_xO_y). The C signal peaks in Figs. 2a and b and Fig. S3 show that the organic compounds (C−O, C−C peaks) from the decomposition of carbonate solvents are abundant in the CEI formed in the blank electrolyte, suggesting a large amount of electrolyte consumption during the cycling process. The organic−rich CEI is quite tightly bound to the LiCoO₂ surface and cannot withstand the large volume changes generated by the oxide cathode during the cycling process, resulting in CEI fragmentation and continuous side reactions between the LiCoO₂ and electrolyte during the lithiation/delithiation process [35,36]. In sharp contrast, the anions are well adsorbed in the IHP of LiCoO₂ with the attraction of electrostatic field due to the exisitence of TPP, and then decomposed into the LiF−rich CEI. The bonding energy of the inorganic substances and transition metal oxide is relatively weak, which can accommodate the repeating volume change during cycling and play a positive role in protecting the cathode material [37]. In addition, the decomposition product LiF of PF₆⁻ has a wide electrochemical stability window, which passivates LiCoO₂ surface [38]. The LiF−rich CEI effectively hinders the interface side reactions between the highly oxidized LiCoO₂ and the electrolyte, and reduces solvent molecular consumption. As can be seen from Figs. 2c−g, the strong signals of LiF and LiPF_xO_y in the inner layer of CEI with TPP indicate the formation of anion−derived CEI, which is consistent with the speculation. In addition, due to the highest HOMO, a trace amount of TPP can be also decomposed on LiCoO₂ to provide phosphide components for CEI.

Figure 2

Figure 2. XPS curves of CEI on LCO cathode, formed in different electrolyte systems. C 1s spectra: (a) Blank electrolyte and (b) TPP−contained electrolyte. F 1s spectra: (c) Blank electrolyte and (d) TPP-contained electrolyte. P 2p spectra: (e) Blank electrolyte and (f) TPP-contained electrolyte. (g) Content of LiF of the CEI of blank and TPP−contained electrolytes.

DownLoad: Full-Size Img PowerPoint

To visually assess the enhancement effect of TPP on LiCoO₂ cathode, Li||LiCoO₂ batteries with different electrolytes were measured in a voltage range of 3−4.6 V. As shown in Fig. 3a, the battery with TPP shows a little higher initial discharge capacity of 187.7 mAh/g while the capacity remained at 149.0 mAh/g with a retention rate of 79.4% after 100 cycles. By contrast, the battery without TPP only shows a low retention rate of 67.9% over 100 cycles. The voltage polarization can be demonstrated by the voltage profile curve visually. As shown in Fig. 3b, the capacity of the battery with blank electrolyte decreases and the polarization voltage increases rapidly over 100 cycles, which is attributed to the intense side reactions between the electrolyte and LiCoO₂ as well as the continuous fragmentation/remodeling of CEI. However, the polarization voltage of the battery with TPP has little change (Fig. 3c). In Fig. 3d, the impedance of the battery with TPP is significantly lower than that of the battery with blank electrolyte since the almost same battery impedances before cycling (Fig. S4 in Supporting information). Moreover, the equivalent circuits and fitting results of the EIS were shown in Fig. S5 (Supporting information), which exhibits a much smaller charge transfer impedance (R_ct) of the battery with TPP than that with blank electrolyte after 10 cycles. The EIS results revealed the fast electrochemical reaction kinetics in the electrolyte with TPP. Furthermore, the transmission electron microscopy (TEM) characterizations visually display the explicit shape of CEI (Figs. 3e and f). Due to the extremely low electronic conductivity of LiF, the anion−induced CEI with TPP is thin and compact, while the CEI with blank electrolyte is bumpy. The firm CEI structure with TPP provides the sustainable basis of the repeated lithiation/delithiation. Furthermore, the battery with TPP exhibits enhanced rate capacities of 194.9, 180.3, 168.1, 149.9, 123.3 mAh/g at current densities of 0.2, 0.5, 1, 2, and 5 C, respectively, and returns to 189.2 mAh/cm² at 0.2 C (Fig. 3g), while a capacity of 109.7 mAh/g is delivered at 5 C for battery with blank electrolyte, which indicates the formation of fast Li⁺ transport dynamics in the CEI formed with TPP.

Figure 3

Figure 3. Electrochemical performances of Li||LCO cell assembled with different electrolyte systems: (a) Cycling performances. (b) Charge–discharge profiles of blank electrolyte and (c) TPP contained electrolyte. (d) EIS profiles after 10 cycles. (e, f) TEM images of CEI on the cathode and (g) rate performances, test in a voltage range of 3–4.6 V (vs. Li⁺/Li).

DownLoad: Full-Size Img PowerPoint

Finally, the phosphate−based additive also has an impact on Li plating/stripping. As shown in Figs. 4a and b, the Li||Li symmetric cell with TPP could maintain a stable plating/stripping process over 450 h at the current density of 1.0 mA/cm² with the capacity of 0.5 mAh/cm² while only 285 h is sustained for the cell with blank electrolyte. At lower current density, the cells with/without TPP exhibit the same polarization voltage value. The cell with TPP exhibits the longer plating/stripping process in ~260 h with a lower polarization voltage when the current density increases to 2.0 mA/cm² (Figs. 4c and d). As shown in Figs. 4e and f, Li deposited into a bulb shape in the blank electrolyte. The transport of Li⁺ was hindered by the accumulated layer of dead Li. In contrast, the plated Li anode with TPP exposed a smooth surface without projecting Li dendrites. The bulk Li with dense and overlapping surface maintains structural integrity during the plating/stripping process, avoiding battery safety issues caused by Li dendrites piercing the diaphragm. The surface morphologies of the Li anodes correspond to the performances of the symmetrical cells [39]. The SEI formed on the Li anodes were further characterized by XPS, which are composed of organic/inorganic components including C−C/C−H, C−O/C=O, ROCO₂Li, Li₂CO₃, LiF and LiPF_xO_y. The results are shown in Fig. 4g and Figs. S6−S8 (Supporting information). According to the Li⁺ solvation results, the most easily reducible FEC with low LUMO in the solvation group is enhanced, and the TPP is involved in solvation and decomposed on Li surface, which provides high F and P content for SEI, respectively. Thus, the SEI formed with TPP contains rich LiF produced by the decomposition of FEC, which is a strong electrical insulator for suppressing the continuous consumption of the electrolyte by resisting the electrons that passing through the SEI layer [40,41]. Besides, LiF exhibits lithiophobicity properties and high Young’s modulus, which can promote the lateral growth and the uniform deposition of new Li on the anode surface [42]. Moreover, LiPF_xO_y provides the SEI with high ionic conductivity and reduced impedance, which enables efficient transport of Li⁺ and allows the battery to exhibit better rate performance [43].

Figure 4

Figure 4. Li dendrite inhibition effect of TPP. (a) Li plating/stripping cycling performance and (b) magnified time–voltage profiles during selected cycle time, under 1.0 mA/cm², 0.5 mAh/cm². (c) Li plating/stripping cycling performance and (d) magnified time–voltage profiles during selected cycle time, under 2.0 mA/cm², 1.0 mAh/cm². SEM images of cycled lithium metal anode in different electrolyte systems (after 25 cycles): (e) Blank electrolyte, (f) TPP–contained electrolyte. (g) Content of LiF on anode surface with blank and TPP–contained electrolytes after 10 cycles.

DownLoad: Full-Size Img PowerPoint

In summary, we have successfully explored a phosphate−contained high−voltage electrolyte for 4.6 V Li||LiCoO₂ batteries. TPP could regulate the Li⁺ solvation structure to optimize the structure and components of CEI/SEI. The designed anion−derived CEI with plenty LiF component forms a parclose to protect the LiCoO₂ structure and reduce the persistent side reactions of solvents. At the same time, the TPP−induced SEI with high content of F and P radicals guarantees excellent anode stability. This work demonstrate the effectiveness of electrolyte additives to optimize the structure and components of electrode electrolyte interphases via regulating the Li⁺ solvation structure.

Declaration of competing interest

We herein declare that there is not any conflict in this manuscript.

CRediT authorship contribution statement

Xi Tang: Writing – original draft. Chunlei Zhu: Visualization. Yulu Yang: Formal analysis. Shihan Qi: Formal analysis. Mengqiu Cai: Writing – review & editing. Abdullah N. Alodhayb: Writing – review & editing. Jianmin Ma: Writing – review & editing.

Acknowledgments

The project was supported by the National Natural Science Foundation of China (No. U21A20311) and the Distinguished Scientist Fellowship Program (DSFP) at King Saud University, Riyadh, Saudi Arabia.

Supplementary materials

Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2024.110014.
1. [1]
  J.C. Zhang, Z.D. Liu, C.H. Zeng, et al., Rare Met. 41 (2022) 3946–3956. doi: 10.1007/s12598-022-02070-6
2. [2]
  D. Li, B. Zhang, X. Ou, et al., Chin. Chem. Lett. 32 (2021) 2333–2337. doi: 10.1016/j.cclet.2020.11.074
3. [3]
  Q. Wu, B. Zhang, Y. Lu, J. Energy Chem. 74 (2022) 283–308. doi: 10.1016/j.jechem.2022.07.007
4. [4]
  Y. Wang, Q. Zhang, Z.C. Xue, et al., Adv. Energy Mater. 10 (2020) 2001413. doi: 10.1002/aenm.202001413
5. [5]
  J. Huang, Y. Zhu, Y. Feng, et al., Acta Phys. Chim. Sin. 38 (2022) 2208008.
6. [6]
  X. Chen, Q. Sun, J. Xie, et al., J. Energy Chem. 74 (2022) 91–99. doi: 10.1016/j.jechem.2022.07.006
7. [7]
  G.G. Bizuneh, C. Zhu, J. Huang, et al., Small Methods 7 (2023) 2300079. doi: 10.1002/smtd.202300079
8. [8]
  J. Liu, M. Wu, X. Li, et al., Adv. Energy Mater. 13 (2023) 2300084. doi: 10.1002/aenm.202300084
9. [9]
  K. Guo, C. Zhu, H. Wang, et al., Adv. Energy Mater. 13 (2023) 2204272. doi: 10.1002/aenm.202204272
10. [10]
  F. Xian, J. Li, Z. Hu, et al., Chem. Commun. 56 (2020) 4998–5001. doi: 10.1039/D0CC00049C
11. [11]
  J. Zhang, H. Zhang, S. Weng, et al., Nat. Commun. 14 (2023) 2211. doi: 10.1038/s41467-023-37999-4
12. [12]
  Z. Sun, H. Zhou, X. Luo, et al., J. Power Sources 503 (2021) 230033. doi: 10.1016/j.jpowsour.2021.230033
13. [13]
  Y. Lyu, X. Wu, K. Wang, et al., Adv. Energy Mater. 11 (2021) 2000982. doi: 10.1002/aenm.202000982
14. [14]
  X. Hu, J. Liu, Y. Yang, et al., Chin. Chem. Lett. 34 (2023) 108456. doi: 10.1016/j.cclet.2023.108456
15. [15]
  R. Yu, Z. Li, X. Zhang, X. Guo, Chem. Commun. 58 (2022) 8994–8997. doi: 10.1039/D2CC03096A
16. [16]
  H. Wang, J. He, J. Liu, et al., Adv. Funct. Mater. 31 (2021) 2002578. doi: 10.1002/adfm.202002578
17. [17]
  S. Qi, J. He, J. Liu, et al., Adv. Funct. Mater. 31 (2021) 2009013. doi: 10.1002/adfm.202009013
18. [18]
  B. Roy, P. Cherepanov, C. Nguyen, et al., Adv. Energy Mater. 11 (2021) 2101422. doi: 10.1002/aenm.202101422
19. [19]
  F. Li, J. He, J. Liu, et al., Angew. Chem. Int. Ed. 60 (2021) 6600–6608. doi: 10.1002/anie.202013993
20. [20]
  W. Xue, R. Gao, Z. Shi, et al., Energy Environ. Sci. 14 (2021) 6030–6040. doi: 10.1039/D1EE01265G
21. [21]
  W. Lu, J. Zhang, J. Xu, X. Wu, L. Chen, ACS Appl. Mater. Interfaces 9 (2017) 19313–19318. doi: 10.1021/acsami.7b03024
22. [22]
  K. Zhang, J. Chen, W. Feng, et al., J. Power Sources 553 (2023) 232311. doi: 10.1016/j.jpowsour.2022.232311
23. [23]
  H. Zhang, Z. Zeng, S. Wang, et al., Nano Res. 17 (2024) 2638–2645. doi: 10.1007/s12274-023-6056-5
24. [24]
  S. Tan, Z. Shadike, J. Li, et al., Nat. Energy 7 (2022) 484–494. doi: 10.1038/s41560-022-01020-x
25. [25]
  Z. Guo, H. Yang, Q. Wei, et al., Chin. Chem. Lett. 34 (2023) 108622.
26. [26]
  X. Yang, M. Lin, G. Zheng, et al., Adv. Funct. Mater. 30 (2020) 2004664. doi: 10.1002/adfm.202004664
27. [27]
  Z. Qin, B. Hong, B. Duan, et al., Electrochim. Acta 276 (2018) 412–416. doi: 10.1016/j.electacta.2018.04.193
28. [28]
  Y. Sun, J. Huang, H. Xiang, et al., J. Mater. Sci. Technol. 55 (2020) 198–202. doi: 10.1016/j.jmst.2019.11.002
29. [29]
  P. Shi, H. Zheng, X. Liang, et al., Chem. Commun. 54 (2018) 4453–4456. doi: 10.1039/C8CC00994E
30. [30]
  G. Jiang, J. Liu, Z. Wang, J. Ma, Adv. Funct. Mater. 33 (2023) 2300629. doi: 10.1002/adfm.202300629
31. [31]
  T. Zheng, B. Zhu, J. Xiong, et al., Energy Stor. Mater. 59 (2023) 102782.
32. [32]
  H. Zhao, Y. Qian, S. Hu, et al., ACS Appl. Mater. Interfaces 13 (2021) 29676–29690. doi: 10.1021/acsami.1c06890
33. [33]
  J. Tan, J. Matz, P. Dong, J. Shen, M. Ye, Adv. Energy Mater. 11 (2021) 2100046. doi: 10.1002/aenm.202100046
34. [34]
  D. Wu, J. He, J. Liu, et al., Adv. Energy Mater. 12 (2022) 2200337. doi: 10.1002/aenm.202200337
35. [35]
  C. Yang, X. Liao, X. Zhou, et al., Adv. Mater. 35 (2023) 2210966. doi: 10.1002/adma.202210966
36. [36]
  R. Qi, M. Yang, T. Zheng, et al., Small 20 (2024) 2312087. doi: 10.1002/smll.202312087
37. [37]
  J. Liu, B. Yuan, N. He, et al., Energy Environ. Sci. 16 (2023) 1024–1034. doi: 10.1039/D2EE02411J
38. [38]
  P. Bai, X. Ji, J. Zhang, et al., Angew. Chem. Int. Ed. 61 (2022) e202202731. doi: 10.1002/anie.202202731
39. [39]
  Q.K. Zhang, X.Q. Zhang, J. Wan, et al., Nat. Energy 8 (2023) 725–735. doi: 10.1038/s41560-023-01275-y
40. [40]
  X.Q. Zhang, X.B. Cheng, X. Chen, C. Yan, Q. Zhang, Adv. Funct. Mater. 27 (2017) 1605989. doi: 10.1002/adfm.201605989
41. [41]
  M.Y. Zhou, X.Q. Ding, J.F. Ding, et al., Joule 6 (2022) 2122–2137. doi: 10.1016/j.joule.2022.07.003
42. [42]
  S. Liu, J. Xia, W. Zhang, et al., Angew. Chem. Int. Ed. 61 (2022) e202210522. doi: 10.1002/anie.202210522
43. [43]
  H. Su, Z. Chen, M. Li, et al., Adv. Mater. 35 (2023) 2301171. doi: 10.1002/adma.202301171
Figure 1 Theoretical prediction of phosphate−based additives. (a) LUMO (lowest unoccupied molecular orbital) and HOMO (highest occupied molecular orbital) energy values of carbonate solvent molecules and TPP. Schematic diagrams of the ESP charge distribution in the solvated sheath of (b) blank electrolyte and (c) TPP–contained electrolyte. Molecular dynamics simulation of Li⁺ in different electrolytes, the cumulative coordination number (n(r)) of Li⁺ with (d) PF₆⁻ anion, (e) FEC molecule, and (f) EC molecule.

下载: 全尺寸图片幻灯片

Figure 2 XPS curves of CEI on LCO cathode, formed in different electrolyte systems. C 1s spectra: (a) Blank electrolyte and (b) TPP−contained electrolyte. F 1s spectra: (c) Blank electrolyte and (d) TPP-contained electrolyte. P 2p spectra: (e) Blank electrolyte and (f) TPP-contained electrolyte. (g) Content of LiF of the CEI of blank and TPP−contained electrolytes.

下载: 全尺寸图片幻灯片

Figure 3 Electrochemical performances of Li||LCO cell assembled with different electrolyte systems: (a) Cycling performances. (b) Charge–discharge profiles of blank electrolyte and (c) TPP contained electrolyte. (d) EIS profiles after 10 cycles. (e, f) TEM images of CEI on the cathode and (g) rate performances, test in a voltage range of 3–4.6 V (vs. Li⁺/Li).

下载: 全尺寸图片幻灯片

Figure 4 Li dendrite inhibition effect of TPP. (a) Li plating/stripping cycling performance and (b) magnified time–voltage profiles during selected cycle time, under 1.0 mA/cm², 0.5 mAh/cm². (c) Li plating/stripping cycling performance and (d) magnified time–voltage profiles during selected cycle time, under 2.0 mA/cm², 1.0 mAh/cm². SEM images of cycled lithium metal anode in different electrolyte systems (after 25 cycles): (e) Blank electrolyte, (f) TPP–contained electrolyte. (g) Content of LiF on anode surface with blank and TPP–contained electrolytes after 10 cycles.

下载: 全尺寸图片幻灯片