Lithium possesses the lowest reduction potential and the smallest atomic weight in all metals, has been considered as the most prominent anode material1. Efforts that pursue lithium metal as anode for rechargeable batteries have been started since the 1950s 2; however, lithium dendrite growth and continuous parasitic reactions during the electrodeposition process had soon been recognized and became to be long-standing challenges that persist until today3,4. A turning point is the emergence of lithiated graphite in the 1990s; this intercalation-type approach is a huge compromise in energy density (decrease the specific capacity from 3860 mAh g-1 to 339 mAh g-1 (for LiC6)), however, with very much improved cycling performance 3. The latter advantage makes the lithium-ion batteries (LIBs) soon replaced lithium-metal batteries (LMBs) in academic research and the industrial field. However, after 30 years of the rapid development of LIBs, the continuous demand for higher energy density drive researchers to revisit the lithium metal anode, facing again these two challenges that arose 50 years before. Although varying progress was achieved to prevent dendrite and dead lithium metal, no attempt was made to directly resolve the parasitic reaction due to its extreme reactivity with the electrolytes. Here we designed a nano-structured photoresist membrane separator, which could suppress the continuous parasitic reaction between lithium metal electrode and liquid electrolytes.
Figure 1 (a) Fabrication of the Zr-MOCN@PP membrane via photopolymerization. SEM images of (b) The PP separator; (d) Zr-MOCN@PP. Insets are photos of the membrane. AFM images of (c) The PP separator; (e) Zr-MOCN@PP.
Two years ago, we prepared an interesting metal-organic cluster Zr-MOC, consisting of a Zr6O4(OH)4 core with 12 methacrylic acid ligands, which could be polymerized under UV light to form a crosslinked network membrane. Such a crosslinked membrane has a high specific surface area and a pore size of 1.4-2.7 nm; we were very excited about founding. Because in general, porous materials with high surface area (such as metal-organic frameworks, MOFs), are in the form of powder because their porous feature requires rigid and highly crosslinking. Therefore, they cannot melt or dissolve in any solvent to get a film. In contrast, membranes are usually obtained by solution casting or melting processing of polymers. The fabricable feature of membranes requires flexible structures, which are incompatible for forming nanoscale pores (and accompanied by high surface area).
When we got this uniform nano-porous separator (Figure 1), the first thing that came to my mind was to use it as a separator for lithium metal batteries, which is one major research topic in our lab. To our big surprise, or I should say, we were very much shocked when we saw that the lithium metal still remained metallic lustre after long cycling when using this unique nano-porous separator. Because, the cycled lithium metal anode (with standard commercial PP separator) should always have a thick black surface layer due to the parasitic reactions between lithium metal and electrolyte.
We further carried out a fascinating experiment based on the above-mentioned findings. We prepared “Li” and “THU” shape patterned separator through lithographic technique (Figure 2a, 2b, Figure 3a). Not surprisingly, we found that the parasitic reactions at the patterned area were selectively eliminated, and we even got “Li” and “THU” shape patterned lithium foil after cycling (Figure 2c, Figure 3b). After that, we used in-situ optical microscopy to observe the lithium deposition behaviour directly. Other measurements, such as time-of-flight secondary ion mass spectrometry (ToF-SIMS) and X-ray photoelectron spectra (XPS) were also used to analyze the chemical/electrochemical products on the cycled lithium metal electrodes, which further confirmed this nano-porous separator could suppress both lithium dendrite and continuous parasitic reactions.
Figure 2. (a) Schematic of the preparation process of the patterned separator. (b) The patterned separator of “Li” shape. (c) Voltage versus time profile of symmetric lithium cell with a “Li” shape pattern separator at a fixed current density of 1.0 mA cm-2, each half-cycle lasts 1 h. Insets showed the lithium metal electrodes after 24 h and 96 h stripping/plating cycles, respectively.
Figure 3. (a) The patterned separator of “THU” shape. (b) The cycled lithium metal electrode with a “THU” shape pattern separator at a fixed current density of 1.0 mA cm-2, each half-cycle lasts 1 h.
The effect of this separator in deactivating lithium metal-electrolyte reactivity is so strong that push us to explore the underneath mechanism. In the liquid electrolyte, lithium-ion is strongly bonded with solvents because of its small cation radius and the highly polar solvent, such as ethylene carbonate (EC) — an indispensable component for the commonly used liquid electrolyte. However, the over-stable Li+-solvates (Li(EC)4+) lead to high reduction energy, crossing over the side reactions’ energy barrier. In our work, we found that the desolvation process of Li(EC)4+ would be promoted inside the nano-porous materials. Meanwhile, electro-reduction of the partially dissociated species was energy favourable, and resulted in suppression of side reactions.
For further information, please read our paper “Suppressing Electrolyte-Lithium Metal Reactivity via Li+-Desolvation in Uniform Nano-Porous Separator” published in Nature Communications (DOI : 10.1038/s41467-021-27841-0)
1. Liu, Y., Zhou, G., Liu, K. & Cui, Y. Design of complex nanomaterials for energy storage: past success and future opportunity. Acc. Chem. Res. 50, 2895-2905 (2017).
2. Albertus, P., Babinec, S., Litzelman, S. & Newman, A. Status and challenges in enabling the lithium metal electrode for high-energy and low-cost rechargeable batteries. Nat. Energy 3, 16-21 (2017).
3. Winter, M., Barnett, B. & Xu, K. Before Li ion batteries. Chem. Rev. 118, 11433-11456 (2018).
4. Kim, H. et al. Metallic anodes for next generation secondary batteries. Chem. Soc. Rev. 42, 9011-9034 (2013).