Confronting the Challenges in Lithium Anodes for Lithium Metal Batteries

Abstract

With the low redox potential of -3.04 V (vs SHE) and ultrahigh theoretical capacity of 3862 mAh g^-1 , lithium metal has been considered as promising anode material. However, lithium metal battery has ever suffered a trough in the past few decades due to its safety issues. Over the years, the limited energy density of the lithium-ion battery cannot meet the growing demands of the advanced energy storage devices. Therefore, lithium metal anodes receive renewed attention, which have the potential to achieve high-energy batteries. In this review, the history of the lithium anode is reviewed first. Then the failure mechanism of the lithium anode is analyzed, including dendrite, dead lithium, corrosion, and volume expansion of the lithium anode. Further, the strategies to alleviate the lithium anode issues in recent years are discussed emphatically. Eventually, remaining challenges of these strategies and possible research directions of lithium-anode modification are presented to inspire innovation of lithium anode.

Keywords: coulombic efficiency; cyclic performance; high energy density; lithium anodes, lithium metal batteries; practical application.

Scheme 1

Schematic diagram of the development history of lithium metal batteries.

Figure 1

a–e) Schematic illustration of lithium anode's failure process. f) Schematic illustration of native surface layer on lithium and g) schematic illustration of uneven current distribution on the surface of lithium anode. f,g) Reproduced with permission.^{[ 27 ]} Copyright 2009, Elsevier. h) Typical behavior of lithium anode in Li‐Li_xMnO₂ AA batteries at different charging rates. The average size of lithium is associated with the charging rates. Corresponding SEM micrographs of lithium grains at different charging rates are also shown. Reproduced with permission.^{[ 55 ]} Copyright 2002, Elsevier. i) The correlations among anode issues. j) Cyclic performance of lithium metal batteries at various N/P ratios. Reproduced with permission.^{[ 56 ]} Copyright 2021, Wiley‐VCH.

Figure 2

a) Cycling performance of Li||Li symmetric batteries with different electrolytes. The battery with tri‐salt electrolyte shows enhanced lifespan compared with the dual‐salt and mono‐salt cases. Reproduced with permission.^{[ 68 ]} Copyright 2019, American Chemical Society. b) Schematic illustration of the flexible polymer and inorganic superionic conductor network. Reproduced with permission.^{[ 88 ]} Copyright 2018, Wiley‐VCH. c,d) The SEM images of ex situ SEI prepared by immersing lithium in Mn(NO₃)₂‐containing carbonate electrolyte. Reproduced with permission.^{[ 90 ]} Copyright 2018, The Royal Society of Chemistry.

Figure 3

a) Schematic illustration of the layer‐by‐layer design where the polymer membrane acts as an adhesive layer between lithium anode and garnet electrolyte. b) Comparison of Nyquist plots for symmetric batteries with garnet‐polymer layer‐by‐layer and garnet‐only structures, respectively. c) Electrochemical lithium deposition/dissolution cyclic performance of the garnet‐polymer layer‐by‐layer symmetrical batteries at 0.1 mA cm⁻². a–c) Reproduced with permission.^{[ 106 ]} Copyright 2019, Wiley‐VCH. d) Schematic illustration of Li⁺ transport paths in IL@MOF electrolyte. Reproduced with permission.^{[ 110 ]} Copyright 2019, The Royal Society of Chemistry. e) Schematic of “brick‐and‐mortar” microstructure, consisting of LAGP (“brick”) and PEO (“mortar”). f) Cyclic performance of Li||Li symmetric batteries with pure PEO and LAGP‐PEO NCPE. The LAGP‐PEO NCPE improved the lifespan of the symmetric battery apparently. e,f) Reproduced with permission.^{[ 20 ]} Copyright 2020, Wiley‐VCH.

Figure 4

a–d) Top‐view and e–h) side‐view SEM images of the IMF matrixes removed from the batteries after lithium deposition with capacities of a,e) 2.0 and b,f) 10 mAh cm⁻² and after lithium dissolution with capacities of c,g) 6.0 and d,h) 9.8 mAh cm⁻². a‐h) Reproduced with permission.^{[ 120 ]} Copyright 2017, American Chemical Society. i) The photograph of upright lithium anode. j) The schematic illustration of the lithium plating process on upright lithium at the low current density and high current density. i,j) Reproduced with permission.^{[ 121 ]} Copyright 2019, Wiley‐VCH. k) In the case of insulating IMF matrix, lithium tends to fulfill the internal space of the host. Reproduced with permission.^{[ 120 ]} Copyright 2017, American Chemical Society.

Figure 5

a) Dynamical changes of DICu current collector in lithium deposition/dissolution. b) The working mechanism of the DICu at low/high loading of lithium. When the loading is relatively low, the association among Cu MP (microparticles) depends on the interaction between Cu MP and polymer chain, while lithium becomes “binder” instead of PVDF when the loading of lithium increases. c–f) SEM images of DICu current collectors after plating lithium with different deposition capacities. a–f) Reproduced with permission.^{[ 129 ]} Copyright 2020, American Chemical Society. g) Schematic structure of the CG (conductivity gradient) host. h–l) The cross‐sectional side‐view ex situ SEM images of CG electrode at different DOC (depth of charge) indicated the fully utilize of internal space during a whole plating/stripping process. (DOC 0%, 42%, 100%, 71% represented that the lithium loading in the CG host were 0 mAh cm⁻², 3 mAh cm⁻², 7 mAh cm⁻² (fully lithiated/delithiated state), and 2 mAh cm⁻². g–l) Reproduced with permission.^{[ 22 ]} Copyright 2020, Wiley‐VCH. m) Schematic illustration of electron‐transport pathway and Li plating in the SEI wrapped TCA electrode. Reproduced with permission.^{[ 132 ]} Copyright 2019, The Royal Society of Chemistry.

Figure 6

a) A comparison of the stability of CCOF‐Li and bare lithium anode in symmetrical batteries. The side‐view SEM images of b) the bare lithium anode and c) the CCOF‐Li electrode after 100 cycles. The top‐view SEM images of d) the bare lithium anode and e) the CCOF‐Li electrode after 100 cycles. The bare lithium anode has a markedly thicker white dead‐lithium layer. The original thickness of bare‐lithium electrode and CCOF‐Li electrode are 380 and 750 µm, respectively, indicating a larger volume expansion of bare‐lithium anode. a–e) Reproduced with permission.^{[ 140 ]} Copyright 2019, Elsevier. f) Photograph and g) SEM images of CNSSM‐Li composite electrode. h) Comparison of cyclic performance for symmetrical Li||Li batteries using pristine Li foil, stainless steel mesh (SSM)‐Li, carbon nanofibers (CF)‐Li, or carbon‐nitrogen modified stainless steel mesh (CNSSM)‐Li electrodes at the current density of 0.5 mA cm⁻². f–h) Reproduced with permission.^{[ 125 ]} Copyright 2020, Elsevier. i) Illustration of the lithium nucleation and growth process on the backside of the host, in which the growth direction of lithium can be guided. Reproduced with permission.^{[ 135 ]} Copyright 2019, Elsevier. j) The schematic illustration of mixed electron/Li⁺ conductive electrode composing of Li₂₂Sn₅ and lithium. k) The comparison of stability between Li||Li symmetric battery (red) and Li/Li₂₂Sn₅||Li/Li₂₂Sn₅ symmetric battery (blue). j,k) Reproduced with permission.^{[ 142 ]} Copyright 2020, Springer Nature.

Figure 7

a) The fabrication of DICu current collectors with granular piling structure by solution‐casting commercial Cu micro‐particles (MPs) onto Cu foils. Reproduced with permission.^{[ 129 ]} Copyright 2020, American Chemical Society. b) The fabrication of the CG host via vacuum‐assisted infiltration process. Reproduced with permission.^{[ 22 ]} Copyright 2020, Wiley‐VCH. c) The preparation of the CLi by coiling glass fiber film and Li foil into the vortex‐shape structure, like a Swiss roll. Reproduced with permission.^{[ 121 ]} Copyright 2019, Wiley‐VCH. d) The two‐step method to fabricate the CCOF‐Li electrode. Reproduced with permission.^{[ 140 ]} Copyright 2019, Elsevier. e) The preparation of the Li/Li₂₂Sn₅ nanocomposite foil realized by a facile calendaring and folding route. Reproduced with permission.^{[ 142 ]} Copyright 2020, Springer Nature. f) The fabrication process of the CNSSM, final carbon‐nitrogen modified stainless steel mesh, and lithium composite electrode (CNSSM‐Li). The SSM was first carbonized and covered with a dense functional layer, then pressed onto lithium foil via mechanical pressing. Reproduced with permission.^{[ 125 ]} Copyright 2020, Elsevier.

Figure 8

Simulation of Li⁺ distribution in the electrolyte for the battery a) without and b) with porous membrane. a,b) Reproduced with permission.^{[ 145 ]} Copyright 2018, American Chemical Society. SEM of the pore structure for c) the PP membrane and d) the 3DOM PI membrane. c,d) Reproduced with permission.^{[ 146 ]} Copyright 2019, American Chemical Society. Postmortem SEM images of the lithium symmetric battery electrodes obtained after cycling with e) CPC and f) PE separators. e,f) Reproduced with permission.^{[ 147 ]} Copyright 2018, Wiley‐VCH. g) Schematic illustration of fabricating P(V‐B) solid electrolyte and the effect of immobilizing anions. Reproduced with permission.^{[ 151 ]} Copyright 2019, The Royal Society of Chemistry. h) Schematic diagram of the ferroelectric effect of BiFeO₃ promoting the transmission of lithium ions. Reproduced with permission.^{[ 152 ]} Copyright 2021, Elsevier. i) Self‐healing electrostatic shielding effect. K⁺ additive gathered around initial growth tip of lithium metal dendrite. The electrostatic force between K⁺ and Li⁺ causes lithium to deposit in the pits. j) The cyclic performance of the Cu//NMC batteries with the electrolytes when operated at 0.2 mA cm⁻². (EYL1 is the baseline electrolyte consisting of ethylene carbonate (EC) and diethyl carbonate. EYL2 is the electrolyte introducing KPF₆ additive in the EYL1. EYL3 is the electrolyte introducing TMSP (tris (trimethylsilyl) phosphite, the function of TMSP is protecting cathode, which isn't the key point we focus on) additive in the EYL1. EYL4 is the electrolyte introducing KPF₆ additive in the EYL3. i,j) Reproduced with permission.^{[ 153 ]} Copyright 2019, Elsevier.

Figure 9

a) Schematic illustrations of the lithium metal plating on a Cu collector in dilute electrolyte (up) and HCE (down). Reducing the amount of free solvent (TEP, triethylphosphate) is in favor of reducing the decomposition of solvent molecules on the surface of lithium anode. Reproduced with permission.^{[ 158 ]} Copyright 2019, American Chemical Society. b–e) Top and f–i) side‐view SEM images of lithium after cycling using b,f) E‐control (the traditional electrolyte consisting of 1.0 m LiPF₆ in EC/EMC with VC as additive) and three LHCEs: c,g) LiFSI in DMC‐BTFE (bis(2,2,2‐trifluoroethyl) ether), d,h) LiFSI in EC/EMC‐BTFE, and e,i) LiFSI in EC/EMC‐BTFE with lithium difluoro(oxalate)borate (LiDFOB) as an additive. From the results, LHCEs are effective to attenuate the thickness of the corrosion layer on lithium surface. b‐i) Reproduced with permission.^{[ 163 ]} Copyright 2018, American Chemical Society. j) The CEs with cycling in different electrolytes measured in Li||Cu batteries. k,l) Voltage profiles of Li plating and stripping processes at selected cycles with k) HCE (LiFSI‐TMS (tetramethylene sulfone)), and l) LHCE (LiFSI‐TMS‐TTE) as electrolytes. j–l) Reproduced with permission.^{[ 160 ]} Copyright 2018, Elsevier.

Figure 10

a) Schematic of tri‐layer separator suppressing Li dendrites in LMO/GO/Li battery. The middle‐layer GOF etches dendrites to hinder their propagation. Reproduced with permission.^{[ 164 ]} Copyright 2018, Elsevier. Cross‐sectional images of lithium metal electrode in the Li||Li symmetric battery that operated at b) ≈1 mA cm⁻² and c) ≈12 mA cm⁻². b,c) Reproduced with permission.^{[ 165 ]} Copyright 2019, Elsevier.

Figure 11

a) The simulation of dendrites’ morphology formed in 3D conductive current collector with channel length of 100 µm and different channel widths. Reproduced with permission.^{[ 144 ]} Copyright 2019, Elsevier. b) Lithium deposits on the incoherent interface and coherent interface, respectively. When lithium is in contact with the solid state electrolyte, the pores formed on the interface can be seen in the side view (upper) and the bottom view (lower). Reproduced with permission.^{[ 25 ]} Copyright 2021, Wiley‐VCH. The analysis of solvation sheath via the MD simulations. c) The radial distribution function of the Pyr₁₄ ⁺ COR and DCA−COM or TCM−COM. d) The radial distribution function of lithium ion and DCA−COM or TCM−COM. e) Refer to the spatial distribution functions of the center Pyr₁₄ ⁺, anions (red), cations (blue), and Li⁺ (silver). The left image and the right image refer to the system including DCA⁻ and TCM⁻, respectively. f) Refer to the spatial distribution function of the central anion. The color scheme of this image is the same as panel e). c‐f) Reproduced with permission.^{[ 26 ]} Copyright 2021, Wiley‐VCH.