INTRODUCTION
Lithium metal is a promising anode candidate for developing high-energy-density Li-based batteries due to its ultrahigh specific capacity (3860 mA h g-1) and low redox potential (-3.040 V versus the standard hydrogen electrode).[1-3] Unfortunately, uncontrollable lithium dendrite growth during the electrochemical process and the unstable electrode/electrolyte interface result in poor cycle life, low Coulombic efficiency (CE), and even explosion. [4-6]
Among various strategies attempted to tackle the aforementioned problems, [7-12] constructing a three-dimensional (3D) current collector is an efficient method to mitigate dendrite growth as the 3D configuration with a high specific surface area and the porous structure can reduce the local current density and accommodate the volume change upon repeated stripping/plating.[13,14] A number of successful demonstrations of uniform Li deposition by adopting the 3D strategy can be found in the literature, such as porous metal-based foams and 3D carbon-based frameworks.[15-18] However, the high specific surface area of a 3D host is double-bladed, as it also leads to an increased interfacial reaction between the Li metal and electrolyte, making the already unstable interface (between Li and existing electrolytes) worse, and serving as a major contributor to the low Coulombic efficiency commonly associated with the 3D configuration.[19,20] To improve the CE, constructing artificial solid-electrolyte interphase (SEI) to reduce the irreversible reaction between electrolyte and lithium metal is a commonly adopted method.[21,22] To date, most of the artificial SEI layers are constructed either on lithium metal or planar current collectors,[23-25] although interfacial stability is a particularly critical issue in the 3D configuration and building an efficient artificial SEI layer on the 3D host may represent a most promising solution to simultaneously achieving long cycle stability and high CE. Recently, a couple of attempts have been made to build SEI on 3D electrode hosts. For example, Zhai et al. designed a 3D g-C3N4/G/g-C3N4architecture, and a high average CE (99.1%) was achieved for 500 cycles at 1 mA cm-2/1 mA h cm-2.[26] Liu et al. reported a LiBr-LiF-rich SEI on the surface of 3D sponge nickel, and the assembled cell exhibited a good cycle life (over 800 cycles at 1 mA cm-2/1 mA h cm-2). [27] Carbon cloth is an important 3D substrate in constructing composite lithium metal anode because of its good chemical stability, excellent mechanical performance, and electrical conductivity. Its 3D architecture has been found to effectively reduce the local current density and accommodate lithium. However, the commercial carbon cloth is lithiophobic, and the existing modification strategies mainly focused on functional groups grafting (-C=O, -COOH), heteroatom doping (N/S/P co-doping, O doping), and lithiophilic material decorating. [28-31] Most recently, Cao et al. reported CoSe2 anchored carbon fiber cloth to enhance lithiophilicity by forming Co/Li2Se, and the system exhibited excellent cycle performance.[32]
Herein, we demonstrated uniform surface decoration of carbon cloth by ZnS/ZnO nanotubes (ZSONT/CC), serving as a multifunctional skeleton to host lithium metal for anode applications. The formed nanotube uniformly covers the entire surface of the individual carbon fiber. Most importantly, during the electrochemical pre-lithiation process, ZnS and ZnO serve as precursors for in situ generations of SEI containing Li2S, Li2O, and LiZn, which materials are most desirable for SEI functionality. More specifically, it is known that Li-Zn alloy enhances the affinity of the 3D host to lithium and thus guides the uniform lithium deposition.[33]Li2S possesses a low Li diffusion energy barrier (0.28 eV) and a high ionic conductivity (10-5 S cm-1), which enable fast Li+transport kinetics.[34] Li2O has an affinity for lithium and can also reduce the elastic strain energy between the plated lithium and the substrate because of the lower crystalline mismatch.[35] As a result, the Li/ZSONT/CC half-cell can last for 700 hours at 4 mA cm-2 / 4 mA h cm-2. A stable CE of 99.2% is achieved over 400 cycles at 1 mA cm-2/ 1 mA h cm-2. When assembled with LiFePO4cathode, the full cell exhibits an excellent cycle performance (with 71% capacity retention after 4000 cycles) and improved rate performance. The present work provides an effective strategy in forming SEI on 3D electrode host with possible control on the SEI content.