Lithium-ion batteries (LIBs) operating at a low temperature are highly wanted in the cold seasons or locations for different applications such as electric vehicles, submarines, and airplanes. At low temperatures, lithium ion (Li⁺) migration rate decreases and reaction rate slows down, resulting in increased internal resistance of the battery, decreased reversible capacity, reduced range of electric vehicles, and may even induce lithium dendrite growth, increasing safety risks. Compared with graphite anode, lithium metal anode has higher energy density (3860 mAh g^-1), which is an ideal anode material for LIBs. In-depth understanding of the microstructure and the variation rule of performance with temperature of lithium metal is the key to break through the bottleneck of low-temperature reaction kinetics of LIBs and improve their low-temperature performance.

Fig.1 Schematic diagram of ion diffusion and charge transfer during Li plating at room/low temperature.

Recently, Xuefeng Wang, Associate Researcher, Institute of Physics/Beijing National Research Center for Condensed Matter Physics, Chinese Academy of Sciences (CAS), and Zhaoshang Wang (co-corresponding author), et al. have utilized a variety of testing and analytical methods, such as cryo-high-resolution transmission electron microscopy (cryo-HRTEM), electron energy-loss spectroscopy (EELS), X-ray photoelectron spectroscopy (XPS), and electrochemical impedance spectroscopy (EIS), to study the the transport behavior and interfacial phase evolution of Li⁺ in lithium-metal batteries under different temperature conditions were investigated, and the constitutive link between them and the electrochemical performance was revealed.

Fig.2 a–c Schematic illustrations of solvation structure in three electrolytes. D Raman spectra of three electrolytes at 25 °C. e Molecular structure, HOMO, and LUMO energy levels for LiPF₆, LiFSI, EC, DMC, FEC, and MTFA. Energy levels for LiPF₆, LiFSI, EC, DMC, and FEC are referenced from Wang et al.60 while that of MTFA was calculated by the same method. f–h Coulombic efficiencies of Li||Cu cells in three electrolytes under a current density of 0.5 mA cm⁻²for 1.0 mAh cm⁻²at different temperatures.

The results show that, kinetically, lowering the temperature increases the kinetic energy barrier of the reaction during lithium deposition, which slows down the Li⁺ transport through the electrolyte and interfacial phase (SEI), as well as the rate of charge transfer, including the desolvation, electrolyte decomposition, and lithium deposition processes. This leads to increased cell polarization and lithium dendrite growth at low temperatures. In addition, thermodynamically, lowering the temperature alters the decomposition reaction paths of lithium salts and solvents in the electrolyte, leading to incomplete decomposition/reaction of lithium salts and solvents and the formation of interfacial phases enriched with unstable organic intermediates, which are unfavorable for Li⁺ transport therein. Compared to the charge transfer impedance (R_ct) associated with the desolvation process, the impedance of Li⁺ transport through the interfacial phase (R_SEI) is the main step limiting the reaction rate at low temperatures. By modulating the solvation structure of Li⁺ in the electrolyte, e.g., by using electrolyte solvents with lower lowest unoccupied orbital (LUMO) energy levels and polar groups, inorganic-rich interfacial phases are generated to improve their temperature tolerance (meaning that the SEI components and structure are less affected by temperature changes). These findings contribute to an in-depth understanding of the Li⁺ behavior and interfacial phase evolution during temperature-regulated lithium deposition/dissolution, deepen the understanding of the bottleneck of reaction kinetics inside the battery, and provide a theoretical basis for the design and performance improvement of low-temperature batteries.

Fig.3 Top views and cross-section (insets) views of the Li deposits (1.0 mAh cm⁻²) in the Li||Cu cell using different electrolytes under a current density of 0.5 mA cm⁻² at 25 (a–c), 0 (d–f) and −20 °C (g–i).

Fig.4 a Ionic conductivities for the three electrolytes at different temperatures. b the setup of the three-electrode cell for the EIS test. c Equivalent circuit. d The fitting results of Rct in LTO (Li₄Ti₅O₁₂ electrode partially lithiated to 85 mAh g⁻¹) ||LTO (Li₄Ti₅O₁₂ electrode partially lithiated to 65 mAh g⁻¹) cell using the equivalent circuit shown in (c), the inset of (d) is the Arrhenius behavior of the resistance corresponding to Li+ desolvation1. e,f The fitting results of Rb and Rinterface in Li||Cu cell after deposition (0.5 mA cm⁻², 1.0 mAh cm⁻²), the inset of (f) is the RSEI of Li||Cu cell.

Fig.5 Cryo-HRTEM images (a–i) and statistical analysis (j–l) of deposited Li metal using different electrolytes at 25 (a–c, and j), 0 (d–f, and k), and −20 °C (g–i, and l). The thickness (m) and content of SEI (n) in different electrolytes and temperatures. The error bars in (m,n) represent the standard deviation of three independent measurements. (Larger images of (a–i) are shown in Supplementary Figs.12–14).

Fig.6 Cryo-TEM images (a,b), EELS mapping (c), EDS results (d), and EELS spectra of Li K-edge (e) of indirect SEI in LiFSI–MTFA/FEC electrolyte. The inset of (b) is the corresponding fast Fourier transform pattern. (Larger image of (b) is shown in Supplementary Fig. 20).

Fig.7 a, b C 1 s (a) and F 1 s (b) XPS spectra of the SEI layer on the deposited Li metal in different electrolytes and temperatures. c Bond breaking modes at different temperatures for Li salt and solvents in the three electrolytes.

The results were published in Nature Communications titled "Temperature-dependent interphase formation and Li⁺ transport in lithium metal batteries". This work was supported by the National Key Research and Development Program of China, the National Natural Science Foundation of China, and the Natural Science Foundation of Beijing Municipality.