First authors: Yin Yichen, Yang Jingtian, Luo Jinda, Lu Gongxun
Corresponding authors: Professor Yao Hongbin, Professor Li Zhenyu, Professor Tao Xinyong
Communication units: University of Science and Technology of China, Zhejiang University of Technology
【Research Background】
Studies have shown that all-solid-state lithium metal batteries (ASSLMBs) play an important role in solving the safety and energy density problems of traditional lithium-ion batteries. As an important component of ASSLMBs, solid-state electrolytes (SEs) directly affect battery performance, but until now no single SEs have met all the characteristics required by ASSLMBs, including high ionic conductivity, soft lattice for solid-solid interface contact, and wide electrochemical window. Based on this, previously reported inorganic SEs have poor electrode compatibility problems, such as poor contact between electrode interfaces (oxide SEs), low electrode interface stability (sulfide SEs), and rapid deterioration of the interface with lithium metals (halide/sulfide SEs), thereby limiting their practical application in ASSLMBs. Therefore, it is essential to develop lithium superion conductors in ASSLMBs that are compatible with the desired properties.
【Main content】
Here, Professor Hongbin Yao and Professor Zhenyu Li of the University of Science and Technology of China, and Professor Xinyong Tao of Zhejiang University of Technology report a LixMyLnzCl3 lanthanide halide-based solid electrolyte (Ln = lanthanide metal element, M = non-lanthanide metal element). Among them, the lanthanide metal element has low electronegativity, and the metal chloride has excellent oxidation resistance and deformability, and the lanthanide metal halide-based solid electrolyte constructed can be in direct contact with the lithium metal anode and ternary cathode without any modification, and the constructed all-solid-state battery can cycle stably at room temperature.
Further characterization showed that compared with the Li3MCl6 (M=Y, In, Sc and Ho) electrolyte lattice, UCl3 LaCl3 lattice had a large one-dimensional fast Li+ conduction channel, which was connected by La vacancy by Ta doping to form a three-dimensional Li+ migration network. The optimized Li0.388Ta0.238La0.475Cl3 electrolyte had a Li+ conductivity of 3.02 mS cm-1 and an activation energy of 0.197 eV at 30°C. At the same time, based on the gradient interface passivation layer, the lithium metal anode is stabilized, so that the Li-Li symmetrical battery can be stably cycled for up to 5000 hours under the condition of 1 mAh cm-2. In addition, when the battery is directly assembled with the uncoated LiNi0.5Co0.2Mn0.3O2 cathode and bare lithium metal anode, the Li0.388Ta0.238La0.475Cl3 electrolyte can enable the solid-state battery to cycle more than 100 times at a cut-off voltage of 4.35V and an area capacity of 1 mAh cm-2. Most importantly, the authors also demonstrated lanthanide chloride (LnCl3; Ln=La, Ce, Nd, Sm and Gd), indicating that LnCl3 solid-state electrolyte systems can achieve further development in terms of conductivity and practicality.
A related article was published in Nature under the title "A LaCl3-based lithium superionic conductor compatible with lithium metal".
【Graphic analysis】
Li+ conduction in the LaCl3 lattice
In this paper, a lattice of LaCl3 for Li+ conduction is proposed, which has a non-tightly packed anion lattice and cubic tight packing type different from the traditional hexagonal tightly packed lattice, and has a large number of one-dimensional (1D) channels, surrounded by adjacent edges sharing the six columns of the trigonal triangular prism [LaCl9] with an inner diameter of about 4.6 Å in the LaCl3 lattice, providing rich octahedral sites for Li conduction. However, existing one-dimensional channels in the lattice are susceptible to channel blocking, resulting in low ion diffusivity. Therefore, a three-dimensional (3D)Li+ diffusion lattice is constructed by interconnecting 1D channels through La openings.
At the same time, de novo molecular dynamics (AIMD) simulations were carried out based on a model system with La vacancies. The simulation results show that Li+ can move rapidly along the one-dimensional channel at closely spaced sites. Some irregularly elongated Li+ probability densities, indicating that the position of Li+ is disordered, are considered to be a key feature for fast Li+ conduction. Interestingly, the migration of Li+ in adjacent channels is connected by La vacancies. In the LaCl3 lattice, the good interconnection of the three-dimensional Li+ migration path can be clearly seen by removing all [LaCl9] polyhedra and the remaining Li+ probability density isosurfaces.
Figure 1: Structural model of Li+ superionic conductors based on LaCl3 lattice and corresponding Li+ migration mechanisms.
Ta5+ doping enables fast Li+ conduction
High-valent Ta5+ doping has been introduced into the vacancies of oxide-type SEs, and this paper first doped Ta5+ in the LaCl3 lattice to introduce La vacancies. x+5y+3z=3 in LixTayLazCl3 to ensure the electrical neutrality of the doped lattice. According to the collected impedance spectrum, it was finally determined that the Li0.388Ta0.238La0.475Cl3 electrolyte had the best conductivity at room temperature, and the calculated Ea value range was 0.18~0.25eV, which was lower than the recently reported metal halide SEs. After component screening, the maximum σ obtained by Li0.388Ta0.238La0.475Cl3 SE at 30°C was 3.02 mS cm-1.
In order to determine the precise structure of Li0.388Ta0.238La0.475Cl3, the authors used synchrotron radiation X-ray diffraction and neutron powder diffraction to show that in the P63/m space group, Li0.388Ta0.238La0.475Cl3 was almost pure phase (99.11wt%), and determined the coordinates of two types of Li+ in the lattice, including the coordinates of Li1 and Li2 in the channel. In addition, Li1 in the channel forms a compressed [LiCl6] octahedron with the 6Cl- coordinates and a rectangular cone with the vacant coordinates of 5Cl- adjacent to Li2. Notably, 6h1 is a metastable intermediate Li site that provides good Li+ mobility and fast Li+ exchange between adjacent channels. Then, the bond valence energy (BVSE) method is applied to simulate the Li+ migration path and energy barrier in Li0.388Ta0.238La0.475Cl3 based on the ordered structure of minimum electrostatic energy. By calculating 100 one- and two-dimensional diffusion energy barriers, it is found that these energy barriers only fluctuate in a narrow range, indicating that Ta5+ doping has no obvious repulsion or blocking effect on Li+ migration.
Figure 2: Conductivity of Li+ in LixTayLazCl3 and identification of Li+ chemical environment.
Lithium metal anode compatibility
The corresponding solid-state battery performance shows that lithium-metal symmetrical batteries are capable of stable cycling for more than 5000 hours (Figure 3A). Based on the formation of a gradient passivation layer formed by electrically insulating LiCl interface phase at the interface, consistent with the theoretical study of reduced cations, this interface layer can also effectively alleviate the interfacial strain during lithium deposition/peeling and protect SE from lithium metal. At the same time, the dense nanocrystalline properties of Li0.388Ta0.238La0.475Cl3 are also crucial for improving the stability of Li/SE interface. After 50 h of cycling, Li remains in close contact with SE. In addition, low electronic conductivity (1.74×10-10 S cm-1) also inhibits the growth of dendrites in SE particles. Close interfacial contact, uniform nanoscale grain boundaries, and low electron conductivity further ensure that the SE reported in this paper has a high critical current density of up to 5 mA cm-2. At the same time, at a higher current density of 0.5 mA cm-2, a long cycle of more than 1000h is also achieved.
Figure 3: Interface stability of Li0.388Ta0.238La0.475Cl3 SE to lithium metal anode.
Full battery performance
The authors further evaluated the electrochemical performance of Li0.388Ta0.238La0.475Cl3 in a lithium-metal solid-state whole battery, which, when cycled at a rate of 0.44 C, provided by the cell had a specific capacity of 163 mAh g-1, an average area capacity of 1.16 mAh cm-2, and an initial coulombic efficiency (CE) of 84.96% (Figure 4a). The high peak of the differential capacity (dQ/dV) curve during charging in the first cycle can be attributed to the formation of the positive-electrolyte interface (CEI) layer, and the dQ/dV curves of the second and third cycles have good overlap, indicating stable CEI for subsequent cycles (Figure 4b). At the same time, the Li/Li0.388Ta0.238La0.475Cl3/NCM523 full battery can cycle more than 100 revolutions with a capacity retention rate of 81.6% (Figure 4c). In addition, the linear sweep voltammetry shows that the stable electrochemical window (4.35 V) in which the charge voltage exceeds 4.27 V can be attributed to the formation of a kinetic quasi-stable interface between the SE and the positive electrode.
Figure 4: Full battery performance test.
【Conclusion Outlook】
In summary, this paper reports a LaCl3-based lithium superion conductor with a unique UCl3 structure. The optimized Li0.388Ta0.238La0.475Cl3 has high ionic conductivity and excellent electrode compatibility, so that the Li/Li0.388Ta0.238La0.475Cl3/NCM523 full battery can cycle stably, showing an area capacity of 1.16 mAh cm-2. Therefore, the work in this paper not only demonstrates the electrode compatibility of LaCl3, but also stimulates a new UCl3-type SE system based on LnCl3 lattices (Ln=La, Ce, Nd, Sm and Gd) with a variety of element doping options.
【Literature Information】
Yi-Chen Yin, Jing-Tian Yang, Jin-Da Luo, Gong-Xun Lu, Zhongyuan Huang, Jian-Ping Wang, Pai Li, Feng Li, Ye-Chao Wu, Te Tian, Yu-Feng Meng, Hong-Sheng Mo, Yong-Hui Song, Jun-Nan Yang, Li-Zhe Feng, Tao Ma, Wen Wen, Ke Gong, Lin-Jun Wang, Huan-Xin Ju, Yinguo Xiao, Zhenyu Li✉, Xinyong Tao✉, Hong-Bin Yao✉, A LaCl3-based lithium superionic conductor compatible with lithium metal, Nature, 2023, https://doi.org/10.1038/s41586-023-05899-8