Recently, there has been worldwide enthusiasm in applying lithium metal as an anode material to batteries. Lithium metal battery could triple the specific energy of today's typical electric vehicle (EV) batteries from 170–200 Wh/kg to 500 Wh/kg. In fact, lithium metal has been intensively studied as an anode material since the 1970s due to its high theoretical capacity (3860 mAh/g) and lowest redox potential (–3.040 V vs. standard hydrogen electrode) . However, uncontrollable dendritic Li growth and limited coulombic efficiency during repeated Li deposition/stripping have prevented practical applications. Incidents of fire followed by manufacture recalls diminished the industrial efforts in the early 1990s . With the urgent need for next-generation high-energy-density battery technology, such as Li–S and Li–air batteries, lithium metal research has been revitalized and attracted regenerated interest in recent years . Key issues relating to Li metal are the following: deposition of Li metal tends to occur anywhere with unpredictable shape; deposition of Li metal tends to be highly dendritic; the irreversible reaction between Li metal and electrolyte leads to low coulombic efficiency and severe capacity decay . Recent progress in improving the morphology and coulombic efficiency of Li metal with novel electrolyte additive, solid electrolyte, nanoscale interface design, etc. has been reviewed by Xu et al. . However, there is no effective approach to regulate Li deposition with control over both position and morphology.
In a recent report, Yan et al. showed an elegant approach to selective deposition and encapsulation of Li metal in hollow carbon shells through seeded growth, which eliminated dendrite formation and enabled improved cycling over 300 cycles . After investigating 11 metal substrates and their Li-metal binary phase diagrams, the authors discovered a substrate-dependent Li-metal nucleation phenomenon: no nucleation barrier is present for metals exhibiting a finite solubility in lithium, whereas appreciable nucleation barrier exists for materials with negligible solubility. Such a discovery immediately led to a demonstration of spatially controlled Li deposition on a patterned substrate. Au strips were patterned on copper substrates as seeds for Li nucleation where the LixAu alloy served as a buffer layer with no nucleation barrier. A Li-metal anode made of a nanocapsule assembly consisting of hollow carbon spheres with embedded Au nanoparticle seeds was designed and tested. Amorphous carbon shells exhibit a slightly higher nucleation barrier than that of Au nanoparticles, therefore promoting Li nucleation and growth inside the carbon spheres (Fig. 1a). This hypothesis was visualized and confirmed by in situ TEM characterization (Fig. 1b and c). Stable cycling over 300 cycles with 98% coulombic efficiency was demonstrated in alkyl carbonate electrolytes. To be practical, Li-metal anodes with coulombic efficiency of at least 99.9% are necessary because irreversible reaction between Li and electrolyte will lead to the accumulation of solid electrolyte interphase and capacity decay.
This work presents an unexploited approach to precisely control Li-metal nucleation during electrochemical plating, which is not only timely, but also thought-provoking. Although there are still many obstacles to overcome, the advancements in nanomaterial and nanostructure synthesis, operando characterization techniques, as well as modeling tools, which did not exist in early 1990s, make us optimistic that the ‘holy-grail’ lithium-metal challenges could be successfully tackled in the near future.