Critical Current Density in Solid-State Lithium Metal Batteries

16163028, 2021, 18, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/adfm.202009925 by Institute Of Chemical Technology, Mumbai, Wiley Online Library on [30/12/2022].
See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Review
www.afm-journal.de
Critical Current Density in Solid-State Lithium Metal

Batteries: Mechanism, Influences, and Strategies
Yang Lu, Chen-Zi Zhao, Hong Yuan, Xin-Bing Cheng, Jia-Qi Huang, and Qiang Zhang*
electrolytes (lithium phosphorus oxyni-

Solid-state lithium (Li) metal batteries (SSLMBs) have become a research tride, LiPON[25]), lithium borohydrides,[26]
hotspot in the energy storage field due to the much-enhanced safety and and solid-state polymers[27–30] are believed
high energy density. However, the SSLMBs suffer from failures including to enable the practical applications of Li
dendrite-induced short circuits and contact-loss-induced high impedance, metal anodes.[31,32] The high shear mod-
ulus of solid-state electrolytes can hinder
which are highly related to the Li plating/stripping kinetics and hinder the
free dendrite extension.[33] Moreover, the
practical application of SSLMBs. The maximum endurable current density of high thermostability of solid-state electro-
lithium battery cycling without cell failure in SSLMB is generally defined as lyte also enables a wide operation tempera-
critical current density (CCD). Therefore, CCD is an important parameter for ture, avoiding the issue of a thermal run
the application of SSLMBs, which can help to determine the rate-determining away.[34] Therefore, solid-state Li metal bat-
steps of Li kinetics in solid-state batteries. Herein, the theoretical and prac- teries (SSLMBs) are promising to realize an
improved safety and energy density.[35–37]
tical meanings for CCD from the fundamental thermodynamic and kinetic
The maximum available current density
principles, failure mechanisms, CCD identifications, and influence factors for of a solid-state battery is defined as critical
improving CCD performances are systematically reviewed. Based on these current density (CCD), which is related
fundamental understandings, a series of strategies and outlooks for future to the power density of batteries and is of
researches on SSLMB are presented, endeavoring on increasing CCD for vital importance to solve the range anxiety
of current electric vehicles (EV).[38–40] CCD
practical SSLMBs.
works as an important index to compare
the performances of Li metal anodes and
indicates many intrinsic behaviors at high
current densities in solid-state batteries. Comprehensive under-
1. Introduction
standings of CCD will help to unravel multiple hidden factors
Nowadays, the energy density of conventional lithium-ion bat- in Li-SSEs interfaces.[41]
teries is gradually impending their theoretical values. Li metal In fact, there are still large gaps of available current density
anode is an important precondition to realize a higher energy in SSLMBs comparing with practical requirements (Figure 1).
density, as it possesses a high specific capacity of 3860 mAh g−1 The industry standards for power batteries are shown in Table 1,
and the negative potential of −3.040 V versus H+/H2.[1–4] How- including IEC (International Electrotechnical Commission,
ever, the formation of lithium dendrites will shorten the cycle International), ISO (International Organization for Standardi-
lifespan and possibly trigger safety accidents in traditional bat- zation, International), QC/T (Recommended Chinese Automo-
teries with organic liquid electrolytes.[5–7] bile Standards, China), GB/T (Recommended Chinese National
The emerging solid-state electrolytes (SSEs) with high ionic Standards, China), NB/T (Recommended Standards of Chinese
conductivity, such as oxide garnets,[8,9] sodium superionic National Energy Administration, China), SAE (Society of Auto-
conductor (NASICON),[10,11] sulfide electrolytes,[12–24] thin-film motive Engineers, USA), and VDA (Verband Der Automobilin-
dustrie, Germany), where a high working current (>4 mA cm−2
or even 10–40 mA cm−2) is usually required.[42–44] However,
Dr. Y. Lu, Dr. C.-Z. Zhao, Dr. X.-B. Cheng, Prof. Q. Zhang lab-level SSLMBs are usually cycled at a current density lower
Beijing Key Laboratory of Green Chemical Reaction than 0.5 mA cm−2. Much attention has been paid on pursuing
Engineering and Technology the breakthrough on energy density, while the available cur-
Department of Chemical Engineering
Tsinghua University
rent densities is also essential in the design towards practical Li
Beijing 100084, P.R. China metal pouch cells.[45–47] As shown in Figure 1, the conventional
E-mail: zhang-qiang@mails.tsinghua.edu.cn pouch cells using Li metal anode with a satisfied energy density
Dr. H. Yuan, Prof. J.-Q. Huang (>300 Wh kg−1) and high single-cell capacity (>1 Ah) can deliver
Advanced Research Institute of Multidisciplinary Science a current density of 0.5–1.5 mA cm−2 with liquid electrolyte.[48,49]
Beijing Institute of Technology As reported by Samsung, the energy-dense solid-state pouch
Beijing 100081, P.R. China
cells (900 Wh L−1) can deliver the available current density of
The ORCID identification number(s) for the author(s) of this article
can be found under https://doi.org/10.1002/adfm.202009925.
3.4 mA cm−2 at 60 °C.[50] Although currently the results meas-
ured with SSEs (e.g., 1–2 mA cm−2 for Li-garnet and Li-sulfide
DOI: 10.1002/adfm.202009925 systems) still fall behind that measured with liquid electrolytes,
Adv. Funct. Mater. 2021, 31, 2009925 2009925 (1 of 33) © 2021 Wiley-VCH GmbH
16163028, 2021, 18, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/adfm.202009925 by Institute Of Chemical Technology, Mumbai, Wiley Online Library on [30/12/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.afm-journal.de
Figure 1. The available current densities of state-of-the-art batteries at commercial level and lab level, considering the real capacity and cycle lifespan.
a) The radar figure and performance comparisons among the commercial industrial standards, Li metal pouch cells with liquid electrolytes, and solid-
state pouch cells. Data sources: Li metal pouch cell RT-1,[48] Li metal pouch cell RT-2,[199] solid-state pouch cell 60 °C.[50] b) The performances of lab-level
batteries. Data sources: Li metal-LE,[200] Li metal garnet,[51,126] Li metal-sulfide,[166,177] and Li metal-hybrid.[201]
these results prove the potential of SSLMBs to achieve a higher elusive. Hence, deeper and comprehensive theoretical under-
power density. standings are essential to realize a quantum leap on CCD.
Nowadays, the acquired CCD results cannot provide enough In this review, we underline the importance of the com-
information due to the non-standard evaluations. Some meas- prehensive understandings on CCD. The flow path on deter-
urements lack practical rationality (e.g., measured with a low mining the CCD is shown in Figure 2. Underlying the Li
areal capacity) and consistency (e.g., various evaluation proto- anode/solid-electrolyte interfaces in SSBs, we systematically
cols). Different measurement details can significantly influence summarize the fundamental understanding on electrochemical
the results. For example, cycling a battery under temperature mechanism of Li metal dynamics, the failure mechanisms of Li
higher than 100 °C can help to realize a current density of metal solid-state batteries and their critical influencing factors,
>20 mA cm−2 in garnet solid-state electrolyte systems.[51] Practi- trying to integrate the CCD with failure and limitation mecha-
cally, the electrode/electrolyte interfaces in SSBs dominate the nisms to the emerging strategies. On the basis of these fun-
CCD, while the understandings on the interfaces, including damental understandings, a series of solutions and previews
the: 1) mass transport of Li ions; 2) intrinsic interfacial phys- for improving CCD performances are also put forward on the
ical and chemical properties; 3) the Li stripping and plating development for high-energy SSLMBs, promoting practical
kinetics; and 4) interfacial evolution regularity[52] still remain evaluation in future research.[2]
Table 1. Standards for current densities towards a power battery for EVs.
Country Standard code Measurement standard (cycle) Current density [mA cm−2]
International IEC-62660:2-2010 1/3 C for BEV 1.34
1 C for HEV 4
ISO-12405-4:2018 1 C, 2 C (standard), 10 C (if applicable) for a high-power pack 4, 8, 40 (if applicable)
1/3 C (standard), 1 C, 2 C (if applicable) for a high-energy pack 1.34, 4, 8 (if applicable)
China QC/T 743-2006 1/3 C (basic unit for measurement) 1.34
1.5 C for energy type 6
4 C for power type 16
GB/T-31467.2-2015 1 C for room temperature 4
1/3 C for low temperature 1.34
NB/T-33024-2016 High-energy type: 0.5 C 2
High-power type: 1 C 4
USA SAE J2929-2011 1 C for HEV/PHEV 4
C/3 for BEV 1.34
Germany VDA 2007 1 C, 10 C, 20 C 4, 40, 80
Notes: Current density is calculated by areal capacity in a commercial Li ion battery (4 mAh cm−2); HEV: Hybrid Electric Vehicle; BEV: Battery Electric Vehicle; PHEV: Plug-in
Hybrid Electric Vehicle; IEC: International Electrotechnical Commission; ISO: International Organization for Standardization; QC/T: Recommended Chinese Automobile
Standards; NB/T: Recommended Standards of Chinese National Energy Administration; GB/T: Recommended Chinese National Standards; SAE: Society of Automotive
Engineers; VDA: Verband Der Automobilindustrie.
Figure 2. Schematic diagram of key issues on determining the critical current density.
2. The CCD Theory Underlying Solid Interface etc.) are also simplified as LLZO) system,[51] exceeding which
battery failure will occur. In solid-state Li metal batteries, Li
2.1. The Development of CCD dendrite penetration constitutes the main causes for short cir-
cuit. Recently, improving CCD has been treated as an important
The development of CCD can be traced back to the application target for SSLMB research. Naturally, the CCD can be involved
of solid-state sodium batteries. The timeline of development into a series of topics underlying metallic lithium/solid-electro-
history on CCD in SSLMBs is shown in Figure 3. In 1980s, lyte interfaces, including the current density dependent interfa-
the available maximum current density is measured in Na/β″- cial kinetics (e.g., ionic diffusion, Li metal nucleation/growth,
Al2O3/Na symmetric cells to indicate the interfacial wetta- and void production/annihilation), thermodynamics (e.g., para-
bility.[53] With the emergence of many new types of SSEs in sitic reactions), and intrinsic quality of SSEs (e.g., degradation
lithium batteries, the concept of CCD is first put forward by of the interfaces and the electrolytes). Deepened understand-
Sakamoto et al. in an Li/Li7La3Zr2O12 (the doped ones (e.g., ings will promote the development of CCD and contribute to
Li7−xLa3Zr2−xMxO12 M = Ta, Nb; Li7−3xNxLa3Zr2O12 N = Al, Ga, standardizing future research.
Figure 3. Timeline of the development of critical current densities (CCDs) in solid-state batteries.[8,13,15,17,41,53,71,80,126,144,158] Reproduced with permission.[41]
Copyright 2017, American Chemical Society. Reproduced with permission.[80] Copyright 2017, Elsevier. Reproduced with permission.[88] Copyright 2018,
the Royal Society of Chemistry.
Figure 4. The lithium nucleation mechanism at solid-state interfaces. a) The surface energy works as barriers for nucleation, and the
overpotential provides the driven force for Li embryo growth. b) The inner pressure derived from the mechanical strength of current collectors deter-
mines the nuclei size. c) The relationship between the size of the lithium nuclei and the imposed current density. Cycling at a higher current density
induces larger overpotentials, which leads to the small nuclei. Cycling at a lower current density induces lower overpotentials, the nuclei inclines to
grow up.
2.2. Interfacial Lithium Chemistry −1

 3 
ε EM  3 ( ri + t ) ν  (1)
ηs = θ −
Theoretically, CCD reflects the maximum available ionic trans- ρ (1 − ν )  2 ( r + t ) 3 − r 3  1 − ν 
portation and charge transfer at the unit area and the time   i i  
without fatal system damage. The unit ionic flux passing across
the interfaces and interfacial charge transfer rate co-determine where ηs represents overpotential, εθ represents strain at
the CCD value. This is actually related with the interfacial θ direction, E represents Young’s modulus, ν represents
lithium chemistry and is the reflection of the essence of solid Poisson ratio, ri represents the nuclei radius, M and ρ rep-
interfaces. The related lithium chemistry processes can be cate- resent the molar mass and density of lithium metal, respec-
gorized into: a) interfacial lithium kinetics including the forma- tively. These results prove that the increased Young’s modulus
tion/consumption of Li ions and the elimination/formation of is related to increased overpotential, resulting in smaller Li
Li atoms at working interfaces; b) interface morphological evo- nuclei.
lution including the void formation and atom diffusions; and The current densities also influence the initial morpholo-
c) interface degradation and failure. gies of the plated lithium. Iriyama et al. systematically probed
the role of current densities on the morphology of deposited
lithium based on LiPON electrolyte.[60] The nucleation overpo-
2.2.1. Basic Lithium Kinetics tentials are dependent on the current densities in a cell. The
nucleation overpotentials can be predicted by the deviation of
There are a series of models for lithium deposition based on peak potential ηNC. According to the Nernst equation, the Li
liquid electrolytes, such as heterogeneous nucleation model, plating process is described as:
space charge model, and solid-electrolyte interphase induced +
Li
dendrite growth model.[5,54–56] The lithium deposition in solid- RT aLiPON
E = E0 + ln 0 (2)
state systems share some similarities, while the unique chem- F aad
ical/mechanical state of various solid electrolytes will lead to
0
novel lithium kinetics. where aad represents the activity of polarized and adsorbed Li
Li+
The lithium nucleation process is analyzed based on sur- atom, aLiPON represents the activity of Li+ in bulk LiPON.
face energy and chemical potential,[57] where the overpotential The overpotential is marked as ηk. The activity of aad at dif-
contributes to overcoming the surface energy and triggers Li ferent current densities is calculated as:
embryo formation (Figure 4a). The lithium nucleation in solid-
state systems is first observed in LiPON thin-film batteries.[58]  Fη 
Amorphous LiPON delivers good stability and high mechanical aad = aad
0
exp  k  (3)
 RT 
strength, avoiding the grain boundary to form a uniform local
ionic distribution. Motoyama et al. took Cu as current collec-
∆µ = Fηk . (4)
tors to probe the plating behaviors of Li nuclei (Figure 4b). The
nucleation overpotential and nuclei sizes are directly related to The increased chemical potential Δμ influences the critical
the Young’s modulus of the current collectors.[59] The rigid cur- radius and the Gibbs free energy of nucleation. As calculated,
rent collector provides sustained high inner pressure, resulting subsequent formulas are given as follows
in the larger size of nuclei with fewer nuclei sites at the same
γV
current density. The relationship between the overpotential and rc = (5)
the nuclei size is as follows: Fηk
π hγ 2 mainly contributes the Rint, because of the extremely weak RSEI.

∆G = (6) 1
Fηk
Hence considering the R ∝ A , A representing the real active
where rc represents critical radius, γ represents unit Gibbs free contact interfacial area, the continuously increasing Rint indi-
energy of the nuclei, V represents atomic volume, and h repre- cates the constant loss of contacts. In contrast, the reduced
sents the height of deposited nuclei. These calculations are also lithium atoms will refill in the vacancies at the interfaces, recov-
accomplished by Iriyama et al.[60] (Figure 4c). ering the interfacial contacts during the deposition process.
These calculations prove that a higher overpotential (ηk) reduces Therefore, the evolution of Rint is reversible during lithium
the critical radius (nuclei size). The nuclei sizes and morphologies plating and stripping at a low current density of 100 µA cm−2.
are also detected. The large nuclei with whisker growth are formed The kinetics process is accurately displayed by galvanostatic
with low current density (5 µA cm−2) and the small nuclei of about electrochemical impedance spectroscopy.[67] The time–polariza-
200 nm are formed with a current density of 500 µA cm−2. There- tion and time–impedance profiles clearly represent the inter-
fore, the higher current density can render tiny lithium nuclei, facial evolution tendency. Koshikawa et al. also measured the
which is in accordance with the experimental observations. These evolution of Rint during lithium plating and stripping utilizing
nucleation estimations are based on the amorphous LiPON elec- a three-electrode system.[68] Actually, the interfacial kinetics of
trolytes. However, the utilization of polycrystalline oxide or sulfide lithium is much more complicated due to the inhomogeneity
electrolytes introduces more interfacial defects. of the surface microstructures including flaws, crystal orien-
tations, grains, grain boundaries, and contaminates,[69,70] and
these defects will amplify the differences under high current
2.2.2. Interface Evolution and Degradation densities. In order to unravel the surficial asperities on influ-
encing the lithium deposition, the Li-free deposition in half
Nowadays, a series of theoretical models have been proposed cells are observed.
to predict the lithium plating/stripping behaviors. The den- The solid-state half cells can clearly distinguish the lithium
sity-driven model put forward by Ahmad et al. adopted shear kinetics. Most of measurements towards anodic evaluation in
modulus and molar volume ratio as critical parameters.[61] solid-state batteries (SSBs) are conducted in symmetric cells.
Garnet-type Li7La3Zr2O12 (LLZO) solid-state electrolyte exhibits The pre-existing lithium can easily be deformed and con-
high stability against Li metal, serving as an optimal model for tacted with SSEs well concealing the actual lithium kinetics.[67]
the research on interfacial kinetics to exclude the contribution In solid-state half cells, the kinetics of lithium growth can be
of parasitic reactions.[8,62,63] Krauskopf et al employed Al-doped recognized. Initially, the Li-LLZO-Cu half batteries can clearly
Li0.65Al0.25La3Zr2O12 as a model system to probe the fundamental capture the continuous impedance evolution that indicates
phenomena of the Li metal.[52] During the stripping process, the lithium appearance and vanishing. The interfacial lithium con-
Li atom (Li0) will become Li+ and be transported into the SSE, tent can be reflected in the discharge and charge profiles in half
leaving an electron and lithium vacancy (VLi). Repeated stripping cells. The plating lithium results in the reduction of interfacial
processes result in gradually-accumulated vacancies forming impedance. The depletion on stripped lithium is consistent
continuously a void layer without external pressure. The Li+ com- with the drastically emerging Rint (Figure 5b). The homog-
bines with an electron and fills into the VLi during the plating enous LiPON film is appropriate to eliminate the influence of
process. These processes are represented as following formula: the interfacial asperity and conduct optical observations. The
growth of lithium whiskers is monitored at LiPON-Cu inter-
Li  Li + + e − + VLi (7) faces. After the nucleation, the lithium whiskers can rapidly
penetrate the current collectors and sprout out above the cur-
The relationship between interfacial impedances and the rent collectors.[59] The rapid discovery of lithium whisker on
voltage profiles are measured during the dissolution and depo- probing can prove the rapid charge transfer (Figure 5c).[64] The
sition, unraveling the formation and annihilation of the void uncontrollable lithium deposition can exhibit variable mor-
layer at electrode–electrolyte interfaces (Figure 5a).[52] The phologies. Krauskopf et al. and Hu et al. both observed special
interfacial voids dominate the interfacial impedance (or charge deposition morphologies of lithium in half cells. The Cu films
transfer impedance).[64] The essences of interfacial impedance or carbon nanotube foams are pressed on solid-state electrolyte
are derived from the interfacial ionic conductive resistance at to assemble the Li-LLZO half cells.[67,71] Dendrite-like lithium
SEI layers (RSEI) and charge transfer processes (Rct). nucleates and grows underneath the Cu film. The free lithium
also displays whisker growth on LLZO surfaces (Figure 5d).[72]
R int = RSEI + Rct (8) The diameters of the whiskers range from hundred nano
meters to several micrometers, which are determined by the
RT
Rct = (9) pinhole cavities on electrolyte surfaces. The whiskers with
Fi0, short length and thick roots are achieved at a low current den-
As displayed, Rct is charge transfer impedance, and i0 is the sity, while whiskers with thin and long length morphology are
exchange current density. Actually, the charge transfer of Li+/Li delivered at a high current density. These phenomena reflect
at the solid-state interfaces is extremely rapid. It is measured that that the lithium nucleation and deposition behaviors are fully
the exchange current density can reach as high as >50 mA cm−2 controlled by the surficial morphology, current densities, and
in the Li/LLZO and Li/Li2S-P2S5 system, proving the intrinsic the surficial crystal nucleation kinetics.[67] Wang et al. utilized
rapid charge-transfer.[64–66] As measured with LLZO, the Rct carbon nanotube foam as current collectors to investigate
Figure 5. The lithium kinetics during plating and stripping in SSBs. a) The dynamic processes during lithium plating and stripping in symmetric cells.
In continuous cycling, the interfacial voids of lithium metal continuously produce and annihilate. Reproduced with permission.[52] Copyright 2019,
American Chemical Society. b) The lithium plating and stripping in solid-state Li-Cu batteries. The interfacial impedance exhibits obvious changes
and the bulk almost remains at a constant. Reproduced with permission.[67] Copyright 2019, Elsevier. c) The rapid charge transfer induces the dis-
covery of the lithium whiskers.[64] d) The lithium whiskers of different sizes at LLZO interface. The different morphologies of lithium whiskers are also
determined by the current densities. Reproduced with permission.[67] Copyright 2019, Elsevier. e) The lithium “mushroom” at the CNT-LLZO interface
proves the importance of the existence of a current collector for stable lithium plating. Reproduced with permission.[71] Copyright 2017, American
Chemical Society. f) The tendency of lithium growth with different current collectors. The free plating of lithium metal exhibits an obvious tendency
for dendrite propagation, and the alloy interlayers or the pre-existed lithium metal can guide the uniform plating. Reproduced with permission.[67]
Copyright 2019, Elsevier.
lithium deposition. They discovered the “lithium mushroom” the failure of alloy layers. Generally, the alloy can accelerate the
in the carbon networks, which derived from the special interfa- lithium diffusion and contribute to suppressing Li accumula-
cial contacts (Figure 5e).[71] These results confirm that the initial tion (Figure 5f).[52,67,77]
lithium kinetics also obey the nucleation and growth process. The polymer electrolyte is the special case in SSEs. The
The nucleation is determined by the surficial morphology and polymer electrolytes exhibit a low mechanical strength, which
current density, and the growth is highly relied on the formed can be easily penetrated by lithium filaments. The ionic trans-
nucleation sites. portation is fulfilled by the coordinated Li ions at the polymer
The kinetics of Li deposition will be regulated through the chains. The dendrite propagation is similar to that in liquid
alternation of substrates. The formation of alloyed lithium solid electrolytes. Monroe et al. simulated the dendrite formation
solutions render very limited nucleation overpotential and thus process in polymer SSEs.[78]
guide the lithium nucleation in an alloy anode.[73] Lithium ions Briefly, both surface homogeneity and current density are
insert into the metal (e.g., Au,[74] In[75,76]) to form an alloy layer the most important factors on interfacial kinetics. Introducing
and get deposited. The introduction of Li-Au alloy layer is dem- a lithium alloy contributes to reducing Li deposition overpoten-
onstrated to retard the dendrite penetration and reduce the tial and inhibiting Li dendrite penetration. Guiding Li growth at
void generation during the Li stripping process. While, exces- a high current density with an imperfect surface is essential to
sive lithium leads to the dispersion of alloy elements, causing achieve a stable cycling at high rates.
Figure 6. Typical Li metal failures’ modes with different types of SSEs. a) The dendrites intend to grow along the grain boundaries theoretically. Repro-
duced with permission.[80] Copyright 2017, Elsevier. b) The scale up SEM images of lithium filaments at the grain boundary. Reproduced with permis-
sion.[80] Copyright 2017, Elsevier. The lithium filaments in c) sulfide SSEs. Reproduced with permission.[122] Copyright 2017, Wiley-VCH. d) Polymer
SSEs. Reproduced with permission.[81] Copyright 2020, Wiley-VCH. e) The voids accumulation induced open circuit in polymer SSEs.[82] f) The Li voids
accumulation in LLZO based SSEs.[83]
2.2.3. Interface Failure (2) Partly short circuit
The anodic failure including the short circuit and open circuit Partly short circuit is the primary stage of the complete short
determines the CCD value. Lithium dendrite is widely observed circuit, but the short circuit pauses or terminates subsequently.
in different type of SSEs.[79] The dendrites tend to extend The lithium filaments just penetrate the interfaces, not the whole
along the grain boundaries (Figure 6a). The grain bounda- bulk SSEs. The polarization voltage drops slightly, which repre-
ries in garnet SSEs can be fully penetrated by Li filaments sents the partial reduction of the bulk impedance.[87] Because
(Figure 6b).[80] In contrast, due to the poorer adhesion between of the existence of bulk electrolyte impedance, the partly short
sulfide particles, the resistance for dendrite growth is lower circuit cannot be regarded as a complete failure of a solid-state
than that in oxide SSEs, exhibiting clear dendrites (Figure 6c) battery. As displayed by Lu et al., the polarization will increase fol-
in sulfide SSEs. The tree-like and needle-like lithium whiskers lowing the Ohm’s law with the increase of external currents.[88,89]
can be observed in polymer SSEs (Figure 6d).[81] Except for
dendrite-induced failures, the voids-induced open circuit is also (3) Soft short circuit
an important type for battery failures. The interfacial void accu-
mulation can also be detected in polymer based[82] (Figure 6e) Soft short circuit refers to the slight voltage drop during bat-
SSLMBs and the inorganic SSLMBs (Figure 6f).[83] Different tery operations, embodied in the severe fluctuation of charging
types of failure mechanisms in solid-state batteries are intro- curves. After regulating the external factors (e.g., current den-
duced below. sity, pressure, or temperature), the soft short circuit can be
recovered.[90] The soft short circuit usually occurs in polymer
(1) Hard short circuit electrolytes or solid–liquid hybrid electrolytes. Distinguished
from the stiff inorganics, the partial fluidity enables polymer
Hard short circuit in SSLMBs derives from the rapid den- materials to form uniform interfacial contacts. The soft short
drite penetration. The dendrites directly connect the anode and circuit is resulted from the formation and dissolution of tiny
cathode, which is unable to be recovered. The dendrites can dendrites. The high viscosity of the polymer/partial liquid
be valued as parallel leakage resistance in the fitting circuit of restrains the size of dendrites. These tiny dendrites can be
impedance spectra. The electrochemical phenomena and the easily fused, providing chances for battery recovery.[90] However,
corresponding fitting circuits are summarized in Figure 7. In the soft short circuit will induce a severe overcharge pheno
galvanostatic measurements, the voltage or polarization exhibit menon, leading to a waste of energy.
a sudden and drastic drop.[84] The hard short circuits are usu-
ally difficult to forecast.[85,86] Under some circumstances, this (4) Contact loss
hard short circuits are accompanied with severe polarizations.
This kind of short circuit generally occurs in stiff inorganic The electrode–electrolyte contact loss is mainly attributed to
SSEs. Especially, the low strength areas (e.g., grain bounda- the vacancy accumulation at interface and the volume expansion/
ries and defects) exhibit lower shear modulus than that of bulk shrinkage in the continuous stripping of Li metal.[91] The escape
SSEs, which are preferred areas for these completely dendrite of Li atoms will generate vacancy sites during the lithium strip-
propagations.[86] ping. Vacancy accumulation leads to the formation of pores and
Figure 7. Three types of dendrite formation and their corresponding charge profiles and fitting circuits. A rapid growth of thick dendrite induces
the short circuit in a sudden-death style. Finite penetration of dendrite penetration leads to the partly short circuit. Reproduced with permission.[88]
Copyright 2018, the Royal Society of Chemistry. The continuous production and annihilation of tiny dendrites result in the soft short circuit.[85,86]
defects at the working interfaces. These defects gradually reduce progressing of solid-state batteries, various measurement pro-
the effective interfacial area, inducing the partially concentrated tocols for CCD are developed and lead to the different CCD
Li-ion flux. The concentrated Li-ion flux will further exacerbate results.[92] The schematic illustrations of internal evolutions of
the contact loss. Kazyak et al. observed the inhomogeneous Li different measurement protocols are illustrated in Figure 9a,b.
depletion and void accumulation at high current densities and The protocols of evaluating CCDs can be categorized as follows:
high areal capacities (Figure 8).[83] It is proved that the interfacial
capacitances are mainly derived from the interfacial voids. The (1) Time control
microelectrodes are employed to observe the void accumulation A time-control test is that the time is set in same particular
(Figure 8a). Obvious lithium depletions can be observed after length (10 min, 30 min, 1 h, or even longer) in every cycle,
lithium stripping with the capacity of 1.07 and 1.47 mAh cm−2 while the current density is increased in a fixed step. As
(Figure 8b,c). The depleted regions indicate the contact loss areas, displayed in Figure 9a, the capacity is gradually increasing
resulting in the increased polarization (Figure 8d) and inter- with each current step. Figure 9c exhibits an example of
facial impedance (Figure 8e). During lithium plating, the huge time-control measurement, where the polarization increases
volume expansion will also lead to more cracks and consequently continuously until a short-circuit failure in time–voltage
accelerate lithium pulverization.[84] Besides, volume changes of profiles. This method helps to assess the endurance with
cathode materials will also lead to contact loss in the cathodic simultaneously increased current density and capacity.
structure. Contact loss is a dynamic behavior and is accumulated
(2) Capacity control
over cycles. Therefore, maintaining a conformal interface during
repeated cycling is essential to extend the life span of SSLMBs. The capacity control method is based on equal areal capacity
in each cycle (e.g., 0.2, 0.5, 1 mAh cm−2, or more). The oper-
ation time gradually shrinks with the extended cycle because
2.3. Rational Identifications for CCD the capacity is constant (Figure 9b,d). This protocol is effec-
tive to evaluate the influence of capacity during cycling.
2.3.1. Basic Measurement Protocols
(3) Endurable time control
The detection towards battery failure with specific current The measurement only set as one cycle in each current
density is the identification for CCD. Generally, the CCD is density in the time-control test. The measurement remains
measured by stepwise increasing currents. However, with the for several cycles in each current density at this protocol
Figure 8. Contact loss at anodic view of SSLMBs. a) Microelectrodes design for experiments. Depletion regions of b) 1.07 mAh cm−2 stripping and
c) 1.47 mAh cm−2. d) The polarization increasing after contact loss. e) Impedance evolution with continuously stripping.[83]
(Figure 9e). This method can effectively evaluate the role on cells. However, a stable polarization increase is still accompa-
current density and cycling number simultaneously. nied with the elevated current densities. It is a typical “fake
stable phenomena”, which exhibits a continuous increased
(4) Long cycling
polarization without a voltage drop. The EIS at room tempera-
This is the commonly-used protocol for measuring the sta- ture before cycling is 70–80 Ohm (Figure 10b), while the much-
bility of lithium/SSE interface in symmetric cells, which decreased impedance (6 Ohm) at room temperature proves that
aims at evaluating the interfacial accumulation effects the short circuits have occurred (Figure 10b, insert). After the
during long-time operation. However, the operation current short circuits, this symmetric cell exhibits a flat polarization
is relatively low in most of reports. without fluctuation when cycled at a high current density of
10 mA cm−2 (Figure 10c). A flat polarization with a high current
density or capacity is also a common characteristic of the “fake
2.3.2. “Fake Stable” Phenomena stable” phenomenon.
Two methods can help to identify the “fake stable” pheno
The erroneous identification for CCD should be avoided. Most menon when measuring CCD. The most direct method is to
CCDs are identified by the voltage polarization curves. How- compare the initial impedance and the cycled impedance under
ever, the impedance spectra before and after cycling indicate room temperature, where a sudden impedance decrease indi-
that short circuits occur before the voltage profile drops to zero. cates short circuits. Moreover, it renders a stable cycling if the
As aforementioned in Section 2.2.2, the total impedance for calculated DC resistance by Ohm’s law is at the same magni-
symmetric cells is composed of bulk impedance of SSEs, Rct tude with the result from impedance spectra, which is effec-
and RSEI. The polarizations are also derived from the interfa- tive for room-temperature measurements. However, a high
cial overpotential and the Ohm impedance from the bulk SSEs. temperature operation will shrink the polarization which is
According to the DC Ohm’s law, the polarization voltage (Upol) not convenient for identification, where the polarization pro-
U files can also indicate the “fake stable” phenomena. A standard
can basically obey the R pol = pol . The erroneous identifica-
I measure polarization profile with a high current density or high capacity
tion on CCD usually happens at extremely high current den- will exhibit a polarization increase due to the void accumulation
sity or high temperature. Although a short circuit occurs due to at the interfaces (Figure 10d). In comparison, the “fake stable”
dendrite formation, weak electronic impedance will still remain will show an unusual flat and stable polarization where no void
and contribute to the polarization increase with elevated cur- accumulation phenomena can be observed. Once an extremely
rent densities. high “CCD” with a flat polarization has occurred, it can prob-
Herein, our experimental results are provided to unravel ably be attributed to “fake stable” phenomena unless a rapid
the “fake stable” phenomena. This result is measured at 60 °C, Li0 diffusion is determined to eliminate the influences of void
with a cycling capacity of 2 mAh cm−2. The current density is accumulations such as in a Li alloy system. Consequently, an
ranged from 0.5 to 10 mA cm−2. As displayed in Figure 10a, EIS evaluation before and after cycling is essential to validate
the dashed square marks the real failure points in symmetric the results of CCD.
Figure 9. CCD measurement methods. Schematic illustration of CCD measured by with a) time control and b) capacity control methods. Typical charge
and discharge profiles of CCD measured by c) time control. Reproduced with permission.[202] Copyright 2018, American Chemical Society. d) Capacity
control. Reproduced with permission.[42] Copyright 2019, Elsevier. e) Endurable time control. Reproduced with permission.[119] Copyright 2019, Elsevier.
2.4. Rate-Determining Steps for CCD the effective area of for charge transfer is one of the main
rate-determining processes. The mass transport at the inter-
The major target for CCD is used to evaluate the anodic face is also an important factor. The ionic diffusion in inor-
dynamic stability with a series of current density and capacity. ganic SSEs (10−9–10−7 cm2 S−1, room temperature)[18,93] are
The CCD aims at evaluating the rate-determining steps for the basically much faster than that of the Li atom self-diffusion
Li metal and SSE interfaces. The solid–solid Li kinetics are (10−12–10−11 cm2 S−1).[77]
determined by three main processes: ionic transport, charge Another main rate-determining process for mass trans-
transfer reaction, and the Li self-diffusion (Li adatom diffusion port can be attributed to the Li atom transport for inorganic
and voids production and annihilation), which are displayed in systems. If a polymer electrolyte is employed, the mass trans-
Figure 11. port determining factor should be reconsidered due to the low
The Li kinetics is highly relied on the quality of solid– transference number (t+ 0.2–0.3) and Li+ diffusion coefficient
solid interfaces. The exchange current density of intrinsic Li (10−8 cm2 S−1 at 60 °C).[94] These models are based on a per-
charge transfer reactions is as high as 50 mA cm−2.[66] Hence fect SSE without external influence factors. Virtually, other
Figure 10. The “fake stable” phenomena measured in Li/Li7P3S11/Li symmetric cells at 60 °C. a) The cycling profiles for CCD with constant capacity of
2 mAh cm−2 and current density range from 0.5–10 mA cm−2. b) The impedance before cycling (outside) and after cycling (inside). c) The time–voltage
profiles of typical “fake stable” phenomena. d) The demonstration for normal cycling and “fake stable” cycling.
influence factors can also alter the rate-determining process of CCD in solid batteries is more sensitive to external factors than
Li deposition. And the CCD will display special changes under those with liquid electrolytes. The major issues are summa-
different factors. rized in Figure 12 and are discussed as follows.
3. The Influence Factors of CCD 3.1. Interfacial Contact
The CCD can be affected by many internal and external fac- The interfacial contacts are directly related to the interfacial
tors. These factors are usually coupled with each other. Because physical contacts and chemical stability. The physical contact is
of the poor interfacial adjustments in a solid-state system, the one of the most fundamental prerequisites for SSBs.[95] Without
Figure 11. Theoretical model of CCD and the corresponding rate-determining steps. Date sources: The diffusion coefficient of inorganic SSEs[18,93] and
polymer SSEs.[94] The exchange current density[65,66] and charge transfer impedance.[64]
Figure 12. Schematic for key influence factors of CCD. Basically, interfacial contacts including physical contact and interfacial reactions. Intrinsic defects
in solid-state electrolyte are also major influence factors on homogenous Li+ flux, and defects provide spaces for dendrite propagation. Dynamic evolu-
tion such as the propagation of voids, mismatches of diffusion balances among Li atom, vacancies, and Li+ are core mechanisms on failure at high
current densities. Pressure can convert the point-to-point contacts into face-to-face contacts, and also reduce the defects in SSEs. Temperature can
improve the diffusion in electrode and electrolyte, delaying the mismatch of diffusion.
the fluidities to connect the electrolyte and the electrode, the voids will remain at the rough places of the interfaces. If there
transport pathway of Li ions can only be built by solid–solid are contaminates at the boundaries or when operated under
contacts.[96] low pressure, the electrode/electrolyte interface will render a
The SSEs and electrodes (e.g., Li metal) possess unique point-to-point contact (Figure 13b, left) or uneven face-to-face
mechanical properties such as hardness, flexibility, and creep contact (Figure 13b, right). The limited physical contact contrib-
deformation. The theoretical model and practical analysis are utes to high local current densities, consequently accelerating
shown in Figure 13. Persson and co-workers[97] reviewed the Li dendrite growth.[52,98]
basic contact mechanics theories, pointing out that the contact The interfacial reactions between electrolytes and Li metal
regions are determined by the elastoplasticity, the squeezing anodes induce volume changes, generate additional products,
force, and the roughness. The solid–solid contacts are usu- and therefore determine the interfacial contacts.[99] Virtually,
ally formed between the soft matrices and comparatively hard many kinds of SSEs exhibit poor chemical stability against
matrices. The soft substances are squeezed into the interfacial the Li metal. The decomposition of some SSEs with high-
gaps to form a face-to-face contact (Figure 13a). A perfect face- valence metal element will continue until battery failure if a
to-face contact is hard to accomplish. At microscopic view, tiny mixed conducting layer is formed. This kind of interphase is
Figure 13. The interfacial contacts: from the theoretical description to the experimental phenomena. a) Theoretical mode. Reproduced with permis-
sion.[97] Copyright 2006, Elsevier. b) Interfacial contacts in LLZO.[52,84] Reproduced with permission.[52] Copyright 2019, American Chemical Society.
Reproduced with permission.[84] Copyright 2018, Elsevier. c) Chemical degradation of Li/SSE.[102,110,172] Reproduced with permission.[102] Copyright 2019,
American Chemical Society. Reproduced with permission.[110] Copyright 2018, Elsevier. Reproduced with permission.[172] Copyright 2018, the Royal
Society of Chemistry. d) Void induced short circuit. Reproduced with permission.[114] Copyright 2019, Elsevier. e) Lithium pulverization in Li/LLZO.
Reproduced with permission.[84] Copyright 2018, Elsevier.
defined as mixed conductive interphase (MCI). For instance, conduction ability, facilitating the continuous decomposi-
Ge4+ in Li1.5Al0.5Ge1.5(PO4)3 will be reduced into Li-Ge. The tion.[62,100–103] Meanwhile, the heterogeneous morphologies
reduced high-valence metal element will retain the electronic lead to uneven interfacial chemical reactions, which is a
Figure 14. Detects in SSEs influencing CCD. a) The SSEs with abnormal grain growth smash after contacting with molten lithium. Reproduced with
permission.[119] Copyright 2019, Elsevier. b) The relationship between internal defects and CCD value. The low relative density of SSEs results in an
uneven ionic flux and a large space for dendrite extension. Reproduced with permission.[88] Copyright 2018, the Royal Society of Chemistry. c) The den-
drite propagation induced by surficial flaws in SSE single crystals[121] and d) the lithium deposition induced the crack propagation in SSEs. Reproduced
with permission.[122] Copyright 2017, Wiley-VCH. e) The schematic illustration of dendrite-induced high inner stresses on degrading the interfaces.
Reproduced with permission.[122] Copyright 2017, Wiley-VCH.
potential inducement for dendrite propagation (Figure 13c). 3.2. Intrinsic Properties of Electrolyte
Polyethylene oxide (PEO) exhibits the thermodynamic ten-
dency to be decomposed into Li2O, C2H4, H2, and LixCy, The intrinsic properties of the SSEs include the relative den-
etc.[104,105] The Li2S-P2S5(-LiX X = Cl, Br, or I) sulfide solid-state sity, compacted density, chemical components, and mechanical
electrolyte will be decomposed into Li2S and Li3P.[106–109] These strength.[115] Most SSEs are polycrystalline, which are composed
decomposition products are electronically insulated, which are of multiple single crystals, grain boundaries, isolated SSE par-
similar to the solid-electrolyte interphase (SEI) layer derived ticles, impurities, and internal flaws. These internal factors
from the direct reactions between lithium and organic electro- determine the internal ionic flux and guide the nucleation or
lytes in liquid-based battery systems.[55] However, the artificial extension of lithium filaments. Different chemical components
interfaces might be degraded by the volume expansion during (e.g., dopants or functional additives) will also determine the
cycling.[107,110,111] intrinsic properties of SSEs.[116]
A perfect interface guarantees face-to-face contacts, pro- The relative density of a solid-state electrolyte deteriorates the
viding homogenous Li-ion flux. The LLZO electrolytes possess critical behavior of SSLMBs. For instance, the LLZO solid-state
high chemical stability in a wide temperature range, which electrolytes are inclined to display a statement of “abnormal
permits the lithium melting treatment on the LLZO electro- grain growth” during the sintering of an LLZO electrolyte.[89]
lyte at a high temperature (>200 °C).[112,113] After the melting The abnormal grains will destroy the internal homogenous sit-
treatment on LLZO, the molten Li fills into the surficial fluc- uation of SSE, including the ionic conduction and the mechan-
tuations forming a static face-to-face contact. Nevertheless, the ical strength.[70,117,118] After the molten lithium treatment, there
stable status cannot be hold in the continuous dynamic opera- are fragmentations of LLZO SSEs with abnormal grain growth,
tions. The high current density results in a rapid injection of due to the reduced mechanical strength and inhomogeneous
vacancies, and the voids will accumulate at the areas with high internal thermal conduction (Figure 14a).[119]
impedance, reflecting a partial contact loss (Figure 13d).[114] The The ionic conductivities at grain is generally higher than that
stepwise generated contact loss is related to the continuous at the grain boundary, hence the abnormal grain growth will
concentrating on Li-ion flux, finally causing the dendrite pene- concentrate the Li-ion flux. As calculated by Srinivasan et al.,
tration. Besides, the volume expansion and subsequent pulver- a phase map between applied current density and grain size is
ization are also discovered in prolonged cycling (Figure 13e). predicted. Similarly, the current distributes at the grain bound-
This self-activated anodic deterioration stimulates the dendrite aries with the rise of grain size. The dendrite filaments are
formation. inclined to extend through these areas along the grain boundary
Figure 15. The capacity influences the interfacial voids. a) The schematic illustrates the gaps between lithium metal and SSEs after stripping lithium.
Reproduced with permission.[126] Copyright 2019, Wiley-VCH. b) After continuous stripping at a high capacity and a high current density, the interfacial
voids will accumulate.[83]
in these circumstances. The grain size is less than 100 nm; the The CCD is not only dependent on the densities of
current concentration also decreases gradually, reflecting as a SSEs influence, but also the dopants in solid-state electro-
high CCD.[85] lytes.[123,124] Persci et al. demonstrated the relation of dopants
The dendrite penetration can be clearly induced by internal with CCD in an LLZO electrolyte. It is measured that the
defects.[88] Although the Li-Al alloy can help to increase the grain boundary capacity, the bulk capacity, and total ionic
CCD, LLZO with defects also cannot endure a high current conductivity exhibit an obvious discrepancy with different
density. It is demonstrated that the relative density (LLZO) dopants, which are determined as the derivation on the dif-
of >97% can achieve a CCD of >0.7 mA cm−2. The decreased ferent results towards CCD.[125] Zhu et al. also discovered
relative density of 92–93% leads to the drop of a CCD value the Nb doped LLZO exhibited a lower CCD than the Al and
of 0.4 mA cm−2. A LLZO with an extremely low relative den- Ta doped ones.[123] Different dopants significantly affect the
sity (<90%) can be penetrated at a low current density of stability against the Li metal and subsequently influence the
0.2 mA cm−2 (Figure 14b).[88] Nevertheless, a good or even per- CCD performance.
fect interfacial contact is realized, internal flaws will also drasti-
cally weaken the CCD behaviors. Shen et al. also analyzed the
pore connectivity in LLZO on influencing the CCD value. A low 3.3. Capacity on Lithium Plating and Stripping
porosity of LLZO reflects on higher CCD.[120] In subsequent
studies, lithium filament propagation can be induced from The raising areal capacity during Li stripping results in
surficial to internal crystals. Chiang et al. systematically evalu- severe contact loss in all-solid-state systems. As reported,
ated the dendrite propagation in a garnet single crystal by con- 1 mAh cm−2 corresponds to a lithium thickness of 4.9 µm
trolling the surface flaws.[121,122] The lithium tips can induce the (Figure 15a).[126] Nowadays, although some reported solid
crack propagation in single crystal SSEs. Even a single crystal batteries exhibit a long lifespan, the cycling capacities are
LLZTO can be split by the Li dendrites (Figure 14c). The micro- far from practical batteries’. In commercial lithium ion bat-
electrodes are utilized to monitor these processes. As shown in teries, the cathodic areal capacity should exceed 4 mAh cm−2.
Figure 14d, cracks will be rapidly developed during continuous After being fully discharged, the thickness of the void layer
Li plating, which will finally lead to Li penetration. The lithium will reach nearly 20 µm. The stripped void layer represents
metal preferentially gathers in the flaws, and then the lithium the contact loss, and a higher current density will accelerate
ion flux concentrates at the flaw area leading to overpotential. the void accumulation (Figure 15b).[83] In addition, massive
The overpotential is the driven factor of the stress, which is lithium cycling will lead to a high volume expansion, accel-
determined by the size of flaws. The subsequent deposition of erating the contact loss. The loss of active plating/stripping
lithium will impose an additional stress including shear force, sites is difficult to recover due to the sluggish and uncontrol-
triggering the crack of the SSEs (Figure 14e). This research is lable surficial Li adatom diffusion according to the terrace-
also validated by a finite element modeling to assess the current ledge-kink model. It is discussed in Section 2.2.2. that the
density in SSEs.[122] lithium growth exhibits a complexity which is highly related
Figure 16. The diffusion issues of Li0/VLi at the interfaces. a) The evolution tendency of the impedances of bulk, grain boundaries, and interfaces at
different temperatures. Reproduced with permission.[42] Copyright 2019, Elsevier. b) The extremely improved CCD performances at 160 °C. Reproduced
with permission.[51] Copyright 2016, Elsevier. c) The Li0 accumulation due to the insufficient voids at the interfaces. Reproduced with permission.[42]
Copyright 2019, Elsevier. d) The diffusion dominated model for the flux balance between Li vacancies in the Li metal electrode and the Li+ in LLZO
SSEs. Reproduced with permission.[52] Copyright 2019, American Chemical Society.
with initial morphology and the current density. In a high embrittles at a low temperature. The deformation of softened
capacity cycling it is difficult to fulfill a reversible morpholog- lithium will contribute to improving the interfacial contacts
ical evolution, and high current density also deteriorates the under a same stack pressure. Hence the interfacial impedances
morphological reversibility, which results in the poor CCD are reduced with the increasing temperature. The reduced
performance in a working cell. impedance still remains after recovering to room temperature.
It is concluded that the increased temperature is beneficial to
construct conformal interfaces. According to the practical meas-
3.4. Temperature urements by Sakamoto et al., the interfacial impedance can be
reduced to 2.7 Ω cm2 at 175 °C with the ascending temperature.
The batteries are expected to be stably-operated in a wide tem- The CCD can also reach a high value of 20 mA cm−2 at 160 °C
perature range in practical applications under harsh condi- (Figure 16b). The reduced interfacial impedance is directly
tions. The working temperature significantly affects the value related to the improved CCD. The interfacial impedance can
of CCD. The corresponding mechanism is divided into two also achieve a low value of <2 Ω cm2 through some other inter-
aspects, including the intrinsic physical properties of lithium facial modification strategies[41,88] at room temperature. How-
and the electrochemical properties (e.g., ionic conduction in ever, their corresponding CCDs don’t fulfill a similar value of
lithium SSEs and the interface and the diffusion coefficients over 10 mA cm−2. The improved temperature exerts enormous
in an electrode–electrolyte system). As measured by Krauskopf functions on other aspects in the SSBs. The low interfacial
et al., the interfacial activation energy is similar to that of the impedance is only a fundamental precondition to permit the
bulk LLZO electrolyte (Figure 16a). Both the interfacial conduc- high Li+ flux, but it is not a determinant.
tivity and bulk conductivity obey the Arrhenius formula. Hence Another reason on the high CCD at high temperature can be
the interfacial impedance can theoretically achieve a low value. ascribed to the improved diffusion of Li0/VLi and Li+. These fac-
In addition, the temperature determines the local strain rate tors are also temperature-dependent. A high CCD indicates a
of lithium. The lithium softens at a higher temperature and rapid diffusion of lithium ions and lithium atoms at interfaces,
a solid-state electrolyte, and bulk lithium metal. The rapid rise taking an LLZO electrolyte as example, the interfacial elemen-
of these diffusion factors accelerates the lithium diffusion in tary process is denoted as the following formula:[52]
electrode, electrolyte, and the interfaces, which can delay the
existence of the diffusive mismatch at ascending current den- Li  Li Li× + − V ′ Li + ( LLZO)  + e ′ ( Li ) + VLi× ( Li ) (12)

sities. Hence the diffusion balance of electrode–electrolyte
can be realized at a higher current density. This can be treated During lithium stripping, at the surface, a lithium vacant
as a rational interpretation for the high CCD value at high site VLi× (Li) and electron e′(Li) are left in the bulk Li metal, and
temperature. the Li ion transports into the available Li ion channel in the
LLZO, which is the vacant site V ′Li+ (LLZO) or interstitial site of
LLZO at the interface (Figure 16d). These leftover Li vacancies
3.5. Interfacial Diffusion Balance can migrate into the bulk metal. It can be speculated that the
combined effect on the migration of Li vacancy in the electrode
As mentioned in Section 3.4, a higher temperature can signifi- and the mobility of Li ion in the SSE determine the interfacial
cantly improve the CCD in SSBs because of the increasing dif- dynamics. As aforementioned, the mobility of Li+ in SSE is much
fusive capability of Li+. In liquid electrolytes, a diffusion-limited faster than that of Li0 in bulk metal. It is concluded that not only
model is adopted to interpret the dendrite formation at a high the Li0 adatom diffusion determines the interfacial phenomena,
current density. The interfacial mass transportation plays vital but the vacancy diffusion dominates the interfacial dynamics.
roles on constraining current densities. The analogous conjec- According to the vacancy mode, a detailed description on these
ture is proposed in a solid state system. As claimed by Sharafi dynamics processes is put forward as follows: the internal diffu-
et al., a higher CCD represents a larger Li ion flux at the unit sion of vacancy is fast enough to deal with the Li solution at low
area on the interfaces.[42] The ionic transference number is con- current densities. The vacancy sites can migrate into the bulk,
sidered as 1 in a solid-state electrolyte. All the operation cur- which helps to retain a stationary vacancy concentration at the
rent can be ascribed to the Li+ transport. Hence the flux of Li+ interface. The vacancy pile-up occurs at the interface once the
migration (Jm) is displayed as: production of vacancies surpasses their diffusion speed, which
results in the morphological instability. Janek and Schmalzried
J et al. also put forward the defect relaxation model. They took Ag
Jm = (10)
e and Ag+ SSE as examples, which is similar to Li+ in LLZO.
There is a reversible transition between Li0 (Li atom) and Li+  c V (ξ = 0 )  o DV

at the interfaces in an Li metal electrode. The stack pressure jV =  1 −  c V τ (13)
 c Vo V

induces the creep deformation of Li metal. The mass transport
on the stripping of Li0 in Li metal is denoted as Li° flux (J0):
icrit = z·F· j V (14)
J 0 = J diffusion + J creep (11)

V is the vacancy in Li metal, τV is the relaxation time, DV is
There is a discrepancy of the diffusive capability between the diffusion coefficient of Li metal vacancies DV = DLi·xV−1, jV,cri
Li0 in the Li metal and Li+ in the solid-state electrolyte. The is the limiting flux of vacancy, cV is the vacancy concentration
diffusion coefficient of Li+ in the LLZO is measured as c Vo ≈ 8 × 10−9 mol cm−3, and τV is the relaxation time to estab-
2.15 × 10−9 cm2 s−1. Comparatively, the self-diffusion coefficient lish the vacancy equilibrium in the electrodes. The limiting flux
of the Li metal is reported as 5.6 × 10−11 cm2 s−1. If the current of the vacancy is defined as jV. When the concentration of the
density remains in a low value (Jm << J0), the diffusion of Li0 vacancy reaches the maximum value at the interface cV(ξ = 0),
is sufficient to supply the Li loss at the interfaces and leave the vacancy supersaturation status is defined as jV,crit. Once the
homogenous vacancy sites for a new lithium deposition. The current density exceeds the critical vacancy diffusion, more
plated Li° flux can also replenish the vacancy sites with suffi- vacancy will remain at the interfaces, resulting in the accumu-
cient speed by diffusion and creep during the plating process. lation of voids near the interface. This should be seen as the
Therefore, the planar morphology can remain at low current beginning of a morphological collapse. If there is no vacancy
densities. In contrast, the Li0 is not rapid enough to migrate diffusion process in SSBs, contact loss will rapidly happen for a
into the bulk lithium upon the plating of fresh Li0 at high cur- planar electrode in a short time at any current density because
rent densities (Jm > J0). The new plated Li° can only compul- of the rapid formation of the vacancy. However, SSBs can ful-
sively accumulate at the interfaces, resulting in the “non-planar” fill a high areal capacity at a low working current density virtu-
character (Figure 16c). The accumulated Li metal works as new ally. It validates the rationality of this theory. In addition, these
nuclei sites, which induces the morphological instability, even parameters including τV, DV, and c Vo are also exponential tem-
the dendrite penetration. The mass transport limitation occurs perature dependent. Therefore, the fulfilled critical current can
in the lithium electrode (adatom diffusion), which is in contrast be estimated theoretically under different temperatures. In con-
to that in a liquid electrolyte. In a liquid electrolyte, Li+ diffu- clusion, the vacancy diffusion model obeys the practical pheno
sion is the major limitation on mass transport. Moreover, the mena, which provides a powerful mechanism understanding
concentration and diffusion of vacancy are not negligible. In on interfacial dynamics in Li metal SSBs.
EIS spectra, the increase of Rint directly reflects the phenomena The weakness of diffusion is derived from the metal elec-
of contact loss. As described in the Kroger–Vink notation, trode in inorganic electrolyte systems. The vacancy diffusion
and atom rearrangement dominate the charge transfer process. pre-pressed lithium can also avoid the subsequent creep induced
In contrast, due to the crystalline and high solvation towards short circuits.[130] Li metal is softer than most inorganic solid-
Li+ by polymer chains, the Li+ transport in polymer electro- state electrolytes in Li metal based SSBs. The deformation of Li
lyte is sluggish between the available solvation sites.[127] The metal dominates the deformation-dependent impedance reduc-
ionic conductivities are generally lower than that of the inor- tion. Taking the resistive layer into consideration, the interfacial
ganic ones (e.g., oxide and sulfide SSEs). Besides, both cations impedance is summarized in following formula:
and anions diffuse in polymer matrices simultaneously in a
polymer electrolyte. Taking lithium bis(fluorosulfonyl)imide ρ π·H 1 π·HLi (15)
R int = · = ·
(LiTFSI) as example, diffusion coefficient DTFSI− is higher 2 F 2·σ Li+ F
than the cation one DLi+.[128] The transference number of Li+
(t+) can only reach a low value of ≈0.3–0.4 in routine polymer Rint is the interfacial impedance, σ Li+ is the ionic conductivity
electrolytes.[129] A balance is still difficult to negotiate between of SSE, HLi is the Vickers hardness of Li metal, and F is the
an Li0/vacancy diffusion in electrode and the Li+ transport in external pressure. It is concluded that the Rint is proportional
electrolyte at working interfaces. The poor ionic conduction to the F−1/2. According to Equation (12), an extremely high
dominates the interfacial failure. The poor mass transport of a pressure can significantly eliminate the interfacial impedance
polymer electrolyte induces a concentration gradient at a high (Figure 17a). The Equation (12) is based on the ideal planar
current density, which can induce the tendency on dendrite for- model with circular point contacts, ignoring the microscale
mation. Improving the Li+ diffusion in polymer electrolyte is surficial asperities of SSEs. The pressure-dependent deforma-
a feasible effective route based on the theoretical mechanisms. tion of Li metal will overcome the influence of the roughness
Because of the rapid transport of Li+, the accumulation of on the electrolyte (Figure 17b). After practical experimentation,
0
Li results in the dendrite formation in inorganic SSEs. The Krauskopf et al. achieved an areal impedance of 0 Ω cm2 after
accumulation of vacancies leads to contact loss, which will operating the pressure for hundreds of megapascals.[52] The
further concentrate the Li+/vacancy fluxes. On the contrary, pressure can also improve the surface adhesion between an Li
the ionic diffusion in SSEs dominates the interfacial stability metal and SSE. It is concluded that the interfacial strength is
in polymer SSEs. Intrinsically, the mismatch among diffusion relative to the interfacial impedance. This is a new visual angle
fluxes of Li+, Li0, and corresponding vacancies simultaneously on the advantage of the high external pressure.[131] Not only
determine the CCD value and the interfacial morphological sta- that, the pressure induced Li creep can also play vital roles. As
bility. In essence, a higher CCD represents on realizing the Li+- displayed in Section 3.5, the creep induced Li atom flux (Jcreep)
vacancy-Li0 balance with high interfacial fluxes. Ameliorating can compensate the insufficiency of the ionic flux at a high cur-
the diffusion issues in electrode or electrolyte is an important rent density, which can improve the vacancy diffusion.[42,132]
breakthrough point. Hence the Rint remains stable after the Li stripping (Figure 17c).
Wang et al. put forward a concept of “critical pressure”. This
research is also based on an LLZO system.[133] The critical pres-
3.6. Pressure sure means that the pressure induced Jcreep is insufficient to
avoid the void formation below the critical pressure. The voids
Pressure is also considered as an important external factor for at the interface can be fully eliminated after conducting the
SSBs. The physical and mechanical properties include the elastic pressure over critical pressure.[133] A critical pressure σcrit is an
plastic and creep mechanical features. The good ductility of Li emerging standard for stable Li kinetics (Figure 17d). In sulfide
metal enables their deformations by an external force. Espe- SSE system, the critical stripping current also exhibits a pres-
cially, there is obvious creep applied to a wide range of applied sure-dependent character. The creep behavior of lithium metal
stress at a high operation temperature. Predicting the influ- is emphasized as a dominated factor.[132]
ences derived from the deformation of Li metal is essential This conclusion is based on the rigid inorganic SSEs (e.g.,
to realize the implementation of the Li metal anode in SSBs. ceramics). At once considering the yield strength of SSEs, the
Masias et al. systematically measured the Young’s modulus, influence of the external stress should be reconsidered. The
shear modulus, time-dependent deformation (creep), and stress deformation of elastic polymer SSEs will import new effects.[134]
dependent deformation.[130] The elastic and plastic properties The sulfide SSEs prepared by cold pressing also exhibit poor
are directly related to the solid–solid contacts. The deformed Li yield stress. The SSE particles only adhere with each other. The
gradually flows to the side boundary of batteries without a hydro- Li creeps at a high stack pressure can be passively emerged
static pressure, which finally induces the short circuit. The pre- into the SSE and subsequently short the batteries. As eluci-
stressed system constrains the undesirable flow of Li metal. The dated by Doux et al., a stack pressure of 5 MPa delivered a pro-
deformation of Li metal also imports microstructural evolutions, longed cycling of 1000 h (Figure 17e).[135] Hence, considering
which significantly reduce the interfacial troubles and enhance the mechanical endurance capability of SSEs, only a moderate
the CCD values. Because of the pressure-dependent deforma- external stack pressure can balance the replenishing towards
tion of Li metal, the ascending pressure helps to increase the interfacial fluctuations, voids elimination, and the cycling sta-
interfacial contacts. Owing to the interfacial asperities, the load bility at a high current density. The performance of pouch cells
bearing contact area is much smaller than the actual area of can be regulated more obviously by the altered external pres-
electrodes in most routine situations. The contact spots will gen- sure.[136] Virtually, a specific pressure can only impose a specific
erally transfer into contact areas after implementing the hydro- creep regulation at anodic interfaces. There are competitions
static pressure, which is induced by the plastic deformation. The among Li plating, stripping, and creep. If the morphological
Figure 17. The pressure factors on the influence towards the anodic performance. a) The external pressure contributes to reducing the interfacial
impedance. Reproduced with permission.[52] Copyright 2019, American Chemical Society. b) The evolution of interfacial impedance during continuous
lithium plating and stripping. Reproduced with permission.[52] Copyright 2019, American Chemical Society. c) The schematic illustration exhibits the
pressure on reducing the pore formation at the interfaces. Reproduced with permission.[52] Copyright 2019, American Chemical Society. d) The role
of critical stresses on the interfacial morphological stability. Reproduced with permission.[133] Copyright 2019, Elsevier. e) The stack pressure induced
short circuits in sulfide all-solid-state batteries. Reproduced with permission.[135] Copyright 2019, Wiley-VCH. f) The lithium plating induces the internal
pressure in solid-state batteries.[138]
evolution can’t be annihilated, continuous cycling at high cur- accumulate at these areas at the cracks or the tips of the Li
rent density can also destroy the interfaces. filament, which will split off the robust SSEs. This proves the
The external pressure displays more positive effects. How- internal stress is inevitable. The parasitic reactions with lithium
ever, the internal stress will occur and import a hidden danger metal will induce the volume expansion in SSE systems, which
in the dynamic processes of the battery operation. The internal produces the excessive internal stress. This phenomenon can
stress can emerge with following processes. The electrochem- be treated as another reason for the mechanical damages. As
ical process can also render additional internal pressure for an displayed by McDowell and co-workers, the internal cracks
Li metal anode.[137] It has been demonstrated that in Section 3.2 are gradually discovered and interconnected after continuous
that the surficial defects will lead to SSE degradations. Virtu- plating and stripping, leading the chemomechanical failure.[139]
ally, most SSB cells are assembled with an external stack pres- A higher current density and high real capacity can also result
sure. The Li metal anode is restrained in rigid boundaries by in the internal stress, accelerating the degradation of a working
solid-state electrolyte. The plated Li can deposit freely without cell.
boundary constraint. The plated Li metal is compressed with
rigid boundaries, inducing the creep effects of the original Li.
After plating the Li metal, the unit time dt is corresponding to 3.7. Disagreement for CCD Results
the plated Li with the height of dh. The strain rate is signifi-
cantly related to the current density. As calculated, the power- Although many experiments have been conducted with
law creep dominates the deformation of Li metal at high cur- similar or same SSE compositions and Li metal anodes, the
rent densities (Figure 17f).[138] An extremely high stress will acquired CCD results are not in agreement. The CCD is not an
evaluation for a single material, but a comprehensive evaluation asperities, which reduce the active interfacial areas.[140,141]
for battery systems, including the interfaces, electrolytes, elec- Constructing a lithiophilic interface is strongly requested
trodes, external factors (e.g., pressure, temperature etc.), and to improve the interfacial contacts.[142,143] Many lithiophilic
cycling protocols. The influence factors are listed as follows: materials (e.g., Au, Al, Si, graphite, etc.) and eliminating lithi-
1) the interfacial discrepancies in different experiments. The ophobic contaminates such as Li2CO3 has been proved as effec-
CCD results are highly related to the interfacial status including tive choices for reducing the interfacial impedance in garnet
the surface homogeneities. Removing the surficial contami- electrolytes.[144–147] The Li2CO3 can be removed by carbon
nations (e.g., Li2CO3 on LLZO) contributes to improving the reduction at high temperature (Li2CO3 + C = Li2O + 2CO),
performances. In comparison, the intrinsic rough surface and acid pre-treatment (2H+ + Li2CO3 = 2Li+ + CO2 + H2O), and
contaminated surface of same materials cannot realize a same mechanical polishing.[147–149] The lithiophilic transitions are
performance; 2) the discrepancies on SSEs qualities. The inter- common strategies in LLZO based SSLMBs, and the improved
face and internal defects will exhibit discrepancies. The internal CCD value can reach from 0.3–1.5 mA cm−2. In addition, the
defects in SSEs will not only influence the internal Li+ flux but lithium composites such as LiC, LiBN, LiC3N4, etc. can
also provide spaces for dendrite penetration, which is carefully also realize lithiophilic properties delivering the CCD of about
discussed in Section 3.2; 3) discrepancies on external factors. 1.5 mA cm−2.[150–152]
Many solid-state batteries are measured in different battery
(2) External pressure
types, such as coin cells, mold type Swagelok cells, or tube cells.
The adopted pressures and the intrinsic discrepancies for dif- Considering the chemical instability of Li metal against
ferent battery types will obviously influence the CCD results; sulfide SSEs, cold pressing or even isostatic pressing are the
and 4) measurement protocols. The precise evaluation for CCD most adopted methods to fulfill the anodic interfacial con-
is a systematic work, which requires standardized measure- tacts.[153] As described in Section 3.6, external-pressure-induced
ment protocols and accurately-controlled experimental details. lithium creep can improve the interfacial contacts. However,
the initial good contact cannot guarantee a stable cycling
because of the contact loss during high volume changes at a
4. Strategies to Improve CCD
high capacity. Hence more strategies are also in urgent need for
The strategies towards improving CCD are derived from solving practical applications.
the intrinsic kinetic problems and regulating the key factors.
Considering the possible main rate-determining steps for CCD,
the present strategies are still mainly concentrated on solving 4.1.2. Enhancing Interfacial Stability
some specific issues such as active areas or stabilizing the
interfacial chemistry. The strategies include constructing a lithi- As demonstrated in Section 3.1, the interfacial instabilities are
ophilic interface, optimizing solid-state electrolytes, and lean mainly caused by physical contact loss and chemical parasitic
liquid therapy. Some emerging strategies concentrated on ame- reactions. Hence physical structural design for lithium hosts
liorating interfacial diffusion are gradually getting attention. and interfacial chemical regulations are important guidance.
Herein, we summarized the reported strategies with acquired The lithium hosts can be classified as the artificial hosts and
CCD values in Table 2. Key measurement parameters including natural hosts. The chemical regulation is concentrated on SEI
real areal capacity, temperature, and duration time are also regulation.
listed for comparison. The mind map on guiding the strate-
(1) Artificial 3D anodic hosts
gies is shown in Figure 18. The strategies are mainly classified
by charge transfer, ionic transport, and dynamic stability. The Considering the volume expansion and high local current
active areas determine the effective areas for charge transfer. densities, constructing an ionic conductive host provides new
Then the mass transport includes the interfacial ionic trans- opportunities to settle the anodic trouble. The 3D structure of
port and Li atom transport. The ionic transport and Li atom SSEs can improve the active interfacial areas, reduce the local
transport contributes to the dynamic interfacial stability. The area current density, and work as supporting skeletons for
interfacial contact can be improved by lithiophilic modification, lithium metal.[154] Hu et al. designed a 3D structural garnet
utilizing a 3D host, and adopting moderate external factors. framework by tape casting slurry with mixed LLZO powder,
The ionic transport can be enhanced by constructing a stable pore former (e.g., polymethyl methacrylate (PMMA)), and a
interface to eliminate the electronically conductive MCI. The binder (Figure 19a).[155] Porous structures are acquired by de-
dynamic stability can be improved by enhancing chemical dif- binding and sintering the casted precursor tapes. The porous
fusion or introducing liquid-phase therapy in Li metal anodes. LLZO can be prepared as a sandwich structure as a porous-
dense-porous one (Figure 19b). The porous structures con-
4.1. Increasing Active Areas tribute to ameliorating the volume expansion of Li metal and
dense structures possessing high ionic conductivity.[156] Contin-
4.1.1. Improving Interfacial Contact uous plating and stripping behaviors can be achieved in these
ionic conductive hosts. The lithium can be completely depos-
(1) Constructing lithiophilic interfaces ited into the 3D structures (Figure 19c).[157] Although the porous
structures of mixed conductive hosts are similar with the ones
Due to the interfacial lithiophobic contaminants, molten of pure ionic conductive 3D hosts, the mixed conductors guide
lithium metal cannot be directly filled into the surficial the lithium deposition in the host directly due to their intrinsic
Table 2. The state-of-the-art CCD values with cycling protocols.
Method-electrolyte CCD value [mA cm−2] Areal capacity [mAh cm−2] Duration time (per cycle) Temperature [°C] Ref
Oxide electrolyte and
interfacial treatment
Au sputtered LLZO 0.8 0.4 30 min 50 [144]
Li2CO3 free LLZO 0.9 0.45 30 min 60 [147]
Li-Mg alloy LLZO 2 2 1h 25 [126]
Si CVD LLZO 0.2 0.0167 5 min 25 [145]
Pencil drawing interface LLZO 0.3 0.3 1h 25 [203]
Li-Al alloy LLZO 1.2 0.2 10 min 25 [88]
2.3 0.4 10 min 60
ALD Al2O3 LLZO 0.2 0.1 30 min 25 [204]
ALD ZnO LLZO 0.1 <0.01 – 25 [205]
Thermal treatment LLZO 0.3 0.15 30 min 25 [41]
0.2 0.2 1h
Li-C3N4-LLZO 1.5 0.25 10 min 25 [152]
Li-ITO-LLZO 1.05 ≈0.8 ≈45 min 30 [206]
Evaporated Li on LLZO 7 0.2 – 100 [42]
0.93 – 25
H3PO4 treatment on LLZO 0.8 0.4 30 min 25 [149]
0.5 0.25 30 min
Li-Ag coated LLZO 0.75 1.5 2h 25 [207]
0.5 0.25 30 min
Li LLZO 13.3 0.36 – 25 [208]
Sputtered Li on LLZO 20 10 30 min 160 [51]
3.5 1.75 130
Heat treating/Au coated LLZO 6 – – 60 [209]
1 3 3h
Li3OCl coating LLZO 0.05 0.05 1h 25 [210]
Li-Na-LLZO 1.5 0.05 – 25 [211]
TVD-carbon-LLZO 1.2 0.6 30 min 25 [36]
COF coating LLZO 3 – – 25 [212]
Modified electrolyte
No mother powder densified 1.15 0.58 30 min 25 [202]
LLZO
MgO densified LLZO 0.95 0.498 30 min 25 [118]
LiAlO2 modified LLZO 0.75 0.375 30 min 25 [175]
Li3PO4 densified LLZO 0.6 0.3 30 min 25 [174]
3D LLZO host 10 2.5 15 min 25 [157]
7.5 45 min
3D C-LLZO host 1 3 3h 25 [158]
Sulfide electrolyte
Li-LiH2PO4 LGPS 0.1 0.1 1h 25 [171]
LiI incorporation Li2S-P2S5 3.9 3.9 1h 100 [167]
Li-LiF-Li3PS4 >2 – 1h 25 [166]
LiF/LiI-Li7P3S11 0.1 0.1 1h 25 [168]
Li6PS5−xOxBr 0.45 0.225 30 min 25 [213]
0.4 0.2 30 min
Li7P2S8Br0.5I0.5 0.5 0.5 1h 25 [22]
Li-SiS2-Li2S-P2S5 1.5 1.5 1h 25 [214]
Table 2. Continued.
Method-electrolyte CCD value [mA cm−2] Areal capacity [mAh cm−2] Duration time (per cycle) Temperature [°C] Ref
Oxide electrolyte and
Au coated Li3PS4 1.3 6.5 5h 100 [215]
Li3PS4-LiI 1.7 8.5 5h 100 [110]
1.25 5 4h
Li1.27In β-Li3PS4 1 1 1h 25 [216]
Li-Gr-LGPS 10 0.25 – – [217]
Li-Li6PS5Cl1−xFx 6.37 5 47 min [177]
Polymer and liquid therapy
Polymer in ceramic 0.5 0.5 1h 55 [218]
RGO PEO-LLZO 1 1 1h 60 [219]
Liquid with Li metal LLZO 1 1 1h 25 [190]
PEC-LiMNT-FEC 0.5 0.5 1h 25 [220]
Gel-ceramic multilayer 0.8 1 1.25 h 25 [192]
Double layer polymer 0.1 0.1 1h 65 [29]
(PMA-PEO)
PEO-Li3PS4 0.1 0.1 1h 60 [221]
MOF-PEO-IL 0.2 0.2 1h 60 [222]
Nano LLZO-PEO 3 3 1h 60 [223]
Nanowire LLZO-PEO 0.5 0.5 1h 25 [224]
LLZO-PEO 0.1 – – 60 [179]
CNPE 4.8 1.2 15 min 25 [201]
2.4 3.6 1.5 h 25
Phosphorene PEO 1 1 1h 25 [225]
Sodium SSBs
Na-β″-Al2O3 12 0.25 – 25 [226]
5.5 3
Na β″-Al2O3 2.88 0.48 ≈10 min 70 [227]
PMA: Poly(N-methyl-malonic amide); FEC: Fluoroethylene carbonate; CNPE: Cross-linked nanocomposite polymer electrolyte; MNT: Montmorillonite; PEC: Poly(ethylene
carbonate).
electric conductivity (Figure 19d). Hence batteries with porous

SSEs can realize a much higher CCD (>1.0 mA cm−2 even
10 mA cm2) with a high areal capacity (>1.0 mAh cm−2) than
that with conventional planar SSE and electrodes.[157,158] A 3D
electrode can also help to ameliorate volume expansion during
lithium plating. Li et al. analyzed the lithium plating behavior
in a 3D Ti electrode, and concluded the superiority of 3D struc-
tures on improving the lithium deposition in SSLMBs.[159] Nev-
ertheless, building 3D SSE structures will not only increase the
complexity on battery manufacture, the fragilities of 3D porous
frameworks will also hinder their practical usage.
(2) Natural 3D host
Li-rich alloy can be seen as a natural mixed conductive host.
After high-capacity lithium stripping, the porous alloy skeleton
remains at the interfaces, which guarantees the anodic con-
tacts at the Li-lean stage. Hence, an alloy anode can realize a
high CCD with a high operational real capacity with a planar
electrode and electrolyte. The alloy anodes avoid complicated
Figure 18. Strategies towards improving CCD in SSLMBs. design and material preparation, and the rational design for
Figure 19. Artificial 3D structures as lithium hosts. a) The precursor of 3D porous garnet SSEs prepared by tape casting. Reproduced with permis-
sion.[157] Copyright 2019, Elsevier. b) The microscopic images of the 3D garnet frameworks. Reproduced with permission.[157] Copyright 2019, Elsevier.
c) The lithium deposition and solution in 3D pure ion conductive structures. Reproduced with permission.[157] Copyright 2019, Elsevier. d) Schematic
illustration exhibits of 3D mixed conductive structures and the corresponding lithium kinetics. Reproduced with permission.[158] Copyright 2018,
American Chemical Society.
alloy components is promising for full batteries. The alloy also validate the improved chemical stability can help to ameliorate
possesses an improved Li diffusion coefficient, which will be electrochemical performances.
discussed subsequently.
(3) Artificial SEI
4.1.3. Electrolyte Optimization
The parasitic reactions between Li metal and SSEs can
induce morphological breakdown and the uneven ionic con- (1) Ultra-fine surface and densified SSEs
ductive interfaces, which are unfavorable for uniform lithium The surficial fluctuation should be eliminated to reduce the
plating. Similar to the modification towards Li metal anodes interfacial differences for uniform ionic conduction. These
in routine liquid electrolytes, an artificial SEI is constructed to ultra-fine surface achieved by careful polishing can deliver
reduce the side reactions and works as robust shields to resist improved performances.[172] Grain boundary strength is useful
the dendrite penetration. to reduce the tendency of the dendrite penetration along the
Typical sulfide SSE and many oxide SSEs possess an obvious grain boundaries. Virtually, rational densification of the oxide
potential gap against Li metal according to the theoretical cal- SSEs can reduce the defects, which help to enhance CCD. Den-
culation.[160] The relative chemical potential of the interface sified SSEs realize the improved ionic conductivity, providing
(μint) should be the intermediate value between that of the Li sufficient ionic flux. La2Zr2O7 densified LLZO SSEs can achieve
metal (μLi) and the SSEs (μSSE), which can suppress the subse- an ionic conductivity to 10−3 S cm−1.[88,173] Densified grain bound-
quent decomposition of the solid-state electrolyte and the elec- aries also improve the CCD performance. Yamada et al. inves-
trode materials.[161–163] Besides, an ideal artificial layer should tigated the influence of LiOH and Li2CO3 to manipulate the
possess very small electronic conductivity and high ionic con- grain boundaries, which realize a CCD of 0.6 mA cm−2.[174] The
ductivity. The side reactions are perfectly suppressed, and the grain boundary modified by MgO[118] and LiAlO2,[175] fulfilled
high ionic conductivity can avoid the interfacial polarization an increase of a range from 0.75–0.95 mA cm−2 in CCD values.
and lithium accumulation. Examples such as boron nitride The high intrinsic quality of SSEs realizes a slight improvement
for Li-Li1+xAlxGe2−x(PO4)3 x = 0.3–0.5 (LATP) interface,[164] for CCD. If more chemical decoration is involved, the interface
and Ge for Li-Li1+xGe2−x(PO4)3 x = 0.3–0.5 (LAGP) SSEs,[165] and bulk properties can be ameliorated simultaneously.
both following the aforementioned design principles. Au,[74]
(2) Additives for SSEs
LiF,[166] and LiI[110,167,168] have been proved as effective SEI com-
ponents for sulfide SSE systems. The LiF modification can Adding beneficial additives in SSEs aims at dealing with
realize the CCD of 2.0 mA cm−2.[166] The organic polymer and interfacial instability or digesting the formed dendrite by
lithium salts construct protective layer between Li metal and chemical reactions, which help to mitigate the dendrite issues.
Li10GeP2S12(LGPS) SSEs, enabling a stable operation for over Both Li3PO4 in LLZO powder before sintering can react with
3000 h.[169] The Li3PO4 has been selected as a robust SEI com- the penetrated lithium dendrites.[176] In sulfide SSEs, Xu et al.
ponent to suppress dendrite growth and parasitic reactions,[170] added methoxyperfluorobutane (HFE) in sulfide SSEs to con-
which can also play similar roles in solid-state system such vert the internal active dendrites into inert LiF.[168] The additives
as garnet SSE[149] or sulfide SSE.[171] These results powerfully also help to stabilize the interfaces. Tatsumisago et al. added
Figure 20. The mechanisms on the interfacial remedy by an Li-rich alloy. a) Conformal interfacial contacts at the deficient state of Li. Reproduced with
permission.[126] Copyright 2019, Wiley-VCH. b) The improvement of the diffusion of Li0/VLi.[77]
LiI in Li3PS4 SSEs to reduce the interfacial parasitic reac- SSEs, vacancy and Li0 diffusion are main issues to realize the
tion and volume expansion, realizing a high current den- interfacial balances. The Li atom and the diffusion on Li voids
sity of 1.5 mA cm−2 with the real capacity of 7.5 mAh cm−2 at are ensured as the rate determining steps in inorganic systems.
100 °C.[110] Han et al. also conducted similar strategy. After LiI Virtually, the LiIn alloy have been widely applied in sulfide
incorporation, the CCD value can be improved to 3.9 mA cm−2 bases SSLMBs, which can easily realize stable cycling with
at 100 °C.[167] Zhao et al. designed doped Li6PS5Cl1−xFx electro- the current density of >1 mA cm−2. However, the real mecha-
lyte, which can form LiF-based interface, realizing an amazing nisms have not been proved. The understanding on the specific
current density of 6.37 mA cm−2, with an areal capacity of properties of Li alloys in SSLMBs has drawn attention previ-
5.0 mAh cm−2.[177] ously. Basically, Wang et al. proved that the Li-alloy skeleton
remains at the interfaces after high-capacity lithium stripping,
reducing the influence of contact loss (Figure 20a). The lithium
4.1.4. Limitations for Improving Active Areas alloy can also promote fast Li diffusion. Krauskopf et al. used
LLZO system to probe the influence of Li alloy on improving
Basic solutions for improving active areas dominate the Li the anodic performances (Figure 20b).[77] The diffusion co-effi-
metal researches. These strategies are very effective at the ini- ciencies in an Li-Mg alloy can be increased from 0.8 × 10−11 to
tial stage. However, these solutions are gradually approaching 2.3 × 10−11 cm2 s−1. This can be attributed to the altering on dif-
the intrinsic bottlenecks. These methods can guarantee the ini- fusion mode in Li alloy. The diffusion of Li0 in an alloy obeys the
tial active areas but cannot overcome the voids accumulation 1D chemical diffusion model by the Fick’s second law, which
during the continuous dynamic processes such as Li stripping. breaks the limitation on Li diffusion. As expected, the alloy
The common planar structure of SSLMBs will be significantly anode can realize a high CCD value over 2 mA cm−2.[88,126] The
restrained by the interfacial voids and they are hard to recover depletion of active Li can also be significantly delayed due to
due to the uncontrollable Li surficial adatom diffusion. The pre- the improved diffusion capability and the elimination of inter-
sent strategies can mostly realize an interfacial impedance of facial voids. Although the temperature and external pressure
<10 Ω cm2. Hence these rate determining steps of the initial can both improve the diffusion of Li0 or vacancies improving
interfacial contact have been partly overcome. The voids accu- the intrinsic diffusion can reduce the dependence on external
mulations induced by high current density and high capacity factors. However, the corresponding research for the diffusion
can hardly be solved. It seems that although an artificial host balance is extremely insufficient and should be underlined.
can address the voids dilemmas, the complicated fabrication
processes may weaken the practical value. Improving the active
area is insufficient to break the bottlenecks. The mass transport 4.2.2. Increasing the Transference Number of Li+ in Polymer SSEs
issue must be considered to realize a breakthrough.
Li+ diffusion in polymer SSEs dominates the rate determining
steps on dendrite propagation. Therefore, increasing the ionic
4.2. Mass Transport Balance View transference number is employed to ameliorate the diffusion
issues in polymer matrices. The anion immobilization is con-
4.2.1. Improving the Vacancy and Li0 Diffusion in Anodes sidered as an important thought to improve Li-ion migration
in polymer SSEs. Importing strong Lewis acid to solvate the
As mentioned in Section 3.5, the vacancy accumulation results anions, adding filler to restrain the mobility of anions and
in contact loss and Li0 accumulation at interfaces, leading to regulating the chain components have all effectively improved
the dendrite growth. Due to the rapid diffusive speed of Li+ in the transference number.[178–181] Increasing the transference
number of polymer SSE can reduce the tendency of dendrite regulate the interfacial ionic conductions.[188] The Li+ flux is
propagation. However, the relationship between CCD and the not highly reliable to the solid–solid contact. The produced
polymer transference numbers deserves to be further devel- voids can also be emerged by liquid. The liquid therapy
oped. In addition, the CCDs of pure polymer are overall rela- is mainly applied to ameliorate the cathodic troubles.[189]
tively poor than that in inorganic SSEs. The polymer systems This strategy can also significantly improve the anodic per-
are still valuable deserving to be fully developed. formance. As measured by Lu et al., the liquid enables the
stable operation of hybrid symmetric cells of high current
density (>1.0 mA cm−2) (Figure 21).[84] The liquid between
4.2.3. Interfacial Therapy by Lean Liquid or Polymer Li-SSE alters the Li deposition style (Figure 21a1–a3). The
liquid provides homogeneous Li flux which helps to reduce
Stable lithium plating and stripping at a high current den- the inner dendrite in SSEs. The dendrite can be blocked by
sity in an all-solid-state mode is very difficult to achieve. SSEs with improved current density from 0.1 to 1.5 mA cm−2
The liquid or polymer wetting are the compromised strate- (Figure 21b–d). As expected, the stable cycling in asymmetric
gies for practical batteries.[182–187] The liquid or polymer cells with liquid electrolyte at single side also remained. Li
electrolytes can immerse into the interfacial fluctuations to et al. designed a semi-liquid Li metal coupling with the LLZO
Figure 21. The liquid phase therapy on influencing the CCD. The symmetric cells operated in an all-solid-state mode, hybrid mode, and solid–liquid
asymmetric mode. Reproduced with permission.[84] Copyright 2018, Elsevier. (a1) Schematic illustration of a symmetric cell measured in the all-solid-
state mode. b) The dendrite will penetrate at a low current density (0.1 mA cm−2). (a2) Schematic illustration of symmetric cells at a hybrid mode,
where liquid phase therapy is conducted at both sides). c) The symmetric cell can realize stable cycling at a high current density (1.5 mA cm−2). (a3)
Schematic show of asymmetric cells of a semi-solid-state mode, where liquid therapy is conducted at the single side. d) No short circuit occurs at a
high available current density of 1.5 mA cm−2.
electrolyte. The operation current density can easily reach to 5. Conclusions and Outlooks
1.0 mA cm−2.[190] The hybrid electrolyte is usually conducted
in a sandwiched structure or mixture styles.[183,191–193] The An energy-dense solid-state Li metal anode suffers from various
lean liquid can significantly improve the ionic conduc- interfacial concerns, while the batteries operated at a high cur-
tion and interfacial contacts in practical batteries.[194] The rent density for high-power situations will exacerbate the inter-
lithium plating and stripping deliver the characters of that facial degradations. The high utilization of an Li metal anode is
in liquid electrolyte.[195] In hybrid or quasi solid-state elec- also one of the most prerequisite issues to fully utilize the energy
trolyte, many conventional strategies can be transplanted provided by solid-state Li chemistry. Consequently, cycling under
to widen the thoughts. The SSEs plays as physical barriers a high current density (>3 mA cm−2) and high areal capacity (>3
for dendrite extension, and fire retardant to ameliorate the mAh cm−2) is essential for the development of SSLMBs.[43]
safety.[196–198] After cooperating with the liquid or gel polymer, There is still a large gap between the targets and current
the level of performance can approach that in conventional SSLMB performances. CCD is not only an important criterion
batteries, which is the shortcut for the industrial developing to evaluate the SSLMBs, but also provide chances to compre-
direction.[32] hensively understand the interfacial behaviors, failure mecha-
nisms, and other factors influencing electrochemical perfor-
mances. The relationship among basic behaviors, influence,
4.2.4. Limitations of Addressing the Mass Transport Balance and CCD strategies are integrated in Figure 22. The key conclu-
sions are summarized as follows.
The mass transport issues still need to be fully addressed
(1) The significance of CCD
to break the limitation on the intrinsic poor self-diffusion
Li metal. There is still a lack of effective strategies on The major target for CCD is to evaluate the Li dynamic sta-
improving the Li atom diffusion. Li alloy is an effective bility with a series of current densities and capacities. It helps
method to realize improved diffusion. The Li alloy will suffer to ensure the endurance of an Li interface with a specific cur-
from problems such as low Li contents, volume expan- rent density and capacity, which can provide references to con-
sions, and the loss on energy density. More new designs to struct SSLMBs with high power density.
improve Li diffusion or voids diffusion are in urgent need.
(2) Basic Li kinetics for CCD
In addition, breaking mass transport limitation requires
high-quality SSEs and sufficient active areas. The poor The nucleation and growth kinetics of Li metal are influ-
intrinsic ionic conduction and low Li+ transference number enced by the current density. The high current densities tend to
of polymer SSEs still fall behind those of inorganic ones. induce small nuclei with thin and prolonged whiskers, consti-
Consequently, the applications of polymer electrolytes with tuting the origin of morphological instabilities. A high current
Li metal anode will confront more challenges in high-power density and high capacity will lead to rapid Li void accumula-
situations. tions, which will deteriorate the interfacial stability.
Figure 22. Summary of the critical current density, including its mechanism, influence factors, and strategies.
(3) Rate-determining steps for CCD (3) Developing non-destructive characterization techniques
Non-destructive, high-resolution and in situ characterization
The rate-determining steps of CCD help to unravel the methods will provide more fundamental understandings about
essence of the limitation for the instability of Li-SSE inter- the buried interfaces in solid-state systems.
faces. The basic rate-determining steps are Li ionic transport,
(4) Establishing computational simulations
charge transfer, and the Li adatom diffusion and voids diffu-
sion. The charge transfer is controlled by active contact area. The theoretical computational study towards Li-SSE inter-
A different SSEs system will be controlled by different sys- faces especially adopting the critical situations can pro-
tems. The polymer system is limited by the mass transport vide important references for experiments and mechanism
of Li ion and the inorganic system is controlled by Li atom interpretation. A rational model validated by experimental
and Li voids diffusion. The rate-determining steps are derived results can provide profound understandings on solid-state
from the ideal model for solid-state Li kinetics, which will be electrochemistry.
significantly influenced by many practical influence factors.
(5) Establishing the standard for evaluating the CCD
(4) Rational identification for CCD
CCD values are measured in varies of methods presently.
The identification of CCD is based on the standardized Some test protocols are far from practical conditions, and many
protocols with time control, capacity control, and long cycling external conditions (e.g., temperature, pressure) have not been
methods. Those measurement protocols offer various informa- accurately controlled. Hence, a clear standard for CCD evalua-
tion under different practical situations. Rational identification tion should be built. The CCD value should be clearly related
on CCD should avoid the “fake stable” phenomena. with the areal capacity, temperature, and pressure. Also, CCD
should be tested under practical areal capacities and full cells,
(5) Key influence factors for CCD
as employing a high areal capacity are essential to construct
The CCD will be influenced by many internal and external fac- energy-dense devices.
tors. These factors are interfacial contact, intrinsic quality of SSEs,
(6) Conducting CCD evaluation in scale-up SSLMBs
mass transport (diffusion) balance, pressure, and temperature.
Most of the CCD results are measured in coin cells or mold
(6) The strategies to improve CCD
cells, which can provide precisely-controlled measurement con-
The strategies towards improving CCD must concentrate on ditions (e.g., pressure) with less interfacial inhomogeneity. The
the rate-determining steps and their influence factors. The pre- enlarged batteries fabricated by industry techniques with com-
sent strategies have paid much attention on the active area fac- plex interfaces will expose more problems under the critical
tors and achieved many progresses. The mass transport issues evaluation at practical levels.
are gradually emphasized, which are major points to realize a Generally, realizing high current density in SSLMBs is an
performance breakthrough. important issue for their practical applications. The synergism
Elevating CCD is essential for the development of SSLMBs from materials, chemistry, and mechanics will inspire more pro-
with brilliant prospects. More insights into the intrinsic science gresses and in-depth understandings, shedding lights in devel-
and engineering are still of vital importance in the following areas: oping next-generation high-energy and high-power batteries.
(1) Enhancing fundamental understandings on solid-state elec-
trochemistry
Acknowledgements
The solid–solid Li kinetics are involved in multiple solid-state
This work was supported by National Natural Science Foundation
electrochemistry phenomena. Many Li kinetics behaviors have of China (22075029, 21805161, 21808124, 21825501, and U1801257),
not been fully unraveled such as the voids evolution rules, the National Key Research and Development Program (2016YFA0202500),
influences of specific morphology on Li plating/stripping, and Beijing Municipal Natural Science Foundation (Z20J00043), China
their accumulation effects. More basic theories on solid-state Postdoctoral Science Foundation (2019M660659, BX20190168), the
electrochemistry should be demonstrated in details. Some elec- “Shuimu Tsinghua Scholar Program,” and Tsinghua University Initiative
trochemical theories in liquid electrolytes should be carefully Scientific Research Program.
transferred and validated in solid-state systems. All of them
contribute to ensuring the accurate rate-determining factors for
solid-state Li kinetics. Conflict of Interest
(2) Emphasizing the diffusion balance issues The authors declare no conflict of interest.
The diffusion issues usually act as the rate-determining steps

for Li kinetics. Enhancing the diffusion ability of Li0, VLi, or Li+ Keywords
contributes to constructing the interfacial balance of Li0/VLi and
Li+ at a higher current density, postponing the local accumula- battery failure, critical current density, dendrites, Li kinetics, Li metal
anodes, solid-state batteries, standards
tion of voids, Li atoms, or concentration gradients. These issues
contribute to the dynamic interfacial evolutions. The presented Received: November 19, 2020
effective strategies on solving the diffusion issues are still Revised: January 5, 2021
insufficient. Published online: February 22, 2021
[1] B. D. McCloskey, J. Phys. Chem. Lett. 2015, 6, 4581. [32] R. Chen, Q. Li, X. Yu, L. Chen, H. Li, Chem. Rev. 2020, 120, 6820.
[2] H. Li, Joule 2019, 3, 911. [33] C. Monroe, J. Newman, J. Electrochem. Soc. 2005, 152, A396.
[3] G. Yasin, M. Arif, T. Mehtab, X. Lu, D. Yu, N. Muhammad, [34] J. C. Bachman, S. Muy, A. Grimaud, H. H. Chang, N. Pour,
M. T. Nazir, H. Song, Energy Storage Mater. 2020, 25, 644. S. F. Lux, O. Paschos, F. Maglia, S. Lupart, P. Lamp, L. Giordano,
[4] R. Xu, X.-B. Cheng, C. Yan, X.-Q. Zhang, Y. Xiao, C.-Z. Zhao, Y. Shao-Horn, Chem. Rev. 2016, 116, 140.
J.-Q. Huang, Q. Zhang, Matter 2019, 1, 317. [35] Z. Gao, H. Sun, L. Fu, F. Ye, Y. Zhang, W. Luo, Y. Huang, Adv.
[5] X. B. Cheng, R. Zhang, C. Z. Zhao, Q. Zhang, Chem. Rev. 2017, 117, Mater. 2018, 30, 1705702.
10403. [36] W. Feng, X. Dong, X. Zhang, Z. Lai, P. Li, C. Wang, Y. Wang, Y. Xia,
[6] H. Liu, X. B. Cheng, R. Xu, X. Q. Zhang, C. Yan, J. Q. Huang, Angew. Chem., Int. Ed. 2020, 59, 5346.
Q. Zhang, Adv. Energy Mater. 2019, 9, 1902254. [37] L.-P. Hou, H. Yuan, C.-Z. Zhao, L. Xu, G.-L. Zhu, H.-X. Nan,
[7] Y. Liang, C. Z. Zhao, H. Yuan, Y. Chen, W. Zhang, J. Q. Huang, X.-B. Cheng, Q.-B. Liu, C.-X. He, J.-Q. Huang, Q. Zhang, Energy
D. Yu, Y. Liu, M. M. Titirici, Y. L. Chueh, H. Yu, Q. Zhang, InfoMat Storage Mater. 2020, 25, 436.
2019, 1, 6. [38] X. Feng, M. Ouyang, X. Liu, L. Lu, Y. Xia, X. He, Energy Storage
[8] R. Murugan, V. Thangadurai, W. Weppner, Angew. Chem., Int. Ed. Mater. 2018, 10, 246.
2007, 46, 7778. [39] Y. Liu, Y. Zhu, Y. Cui, Nat. Energy 2019, 4, 540.
[9] C. Wang, K. Fu, S. P. Kammampata, D. W. McOwen, A. J. Samson, [40] A. Tomaszewska, Z. Chu, X. Feng, S. O’Kane, X. Liu, J. Chen, C. Ji,
L. Zhang, G. T. Hitz, A. M. Nolan, E. D. Wachsman, Y. Mo, E. Endler, R. Li, L. Liu, Y. Li, S. Zheng, S. Vetterlein, M. Gao, J. Du,
V. Thangadurai, L. Hu, Chem. Rev. 2020, 120, 4257. M. Parkes, M. Ouyang, M. Marinescu, G. Offer, B. Wu, eTranspor-
[10] P. Knauth, Solid State Ionics 2009, 180, 911. tation 2019, 1, 100011.
[11] W. Xiao, J. Wang, L. Fan, J. Zhang, X. Li, Energy Storage Mater. [41] A. Sharafi, E. Kazyak, A. L. Davis, S. Yu, T. Thompson, D. J. Siegel,
2019, 19, 379. N. P. Dasgupta, J. Sakamoto, Chem. Mater. 2017, 29, 7961.
[12] M. Tachez, J. P. Malugani, R. Mercier, G. Robert, Solid State Ionics [42] M. Wang, J. B. Wolfenstine, J. Sakamoto, Electrochim. Acta 2019,
1984, 14, 181. 296, 842.
[13] H. Yamane, M. Shibata, Y. Shimane, T. Junke, Y. Seino, S. Adams, [43] P. Albertus, S. Babinec, S. Litzelman, A. Newman, Nat. Energy
K. Minami, A. Hayashi, M. Tatsumisago, Solid State Ionics 2007, 2017, 3, 16.
178, 1163. [44] Z. Lin, T. Liu, X. Ai, C. Liang, Nat. Commun. 2018, 9, 5262.
[14] Ö. U. Kudu, T. Famprikis, B. Fleutot, M.-D. Braida, T. Le Mercier, [45] S. R. Chen, C. J. Niu, H. Lee, Q. Y. Li, L. Yu, W. Xu, J. G. Zhang,
M. S. Islam, C. Masquelier, J. Power Sources 2018, 407, 31. E. J. Dufek, M. S. Whittingham, S. Meng, J. Xiao, J. Liu, Joule 2019,
[15] N. Kamaya, K. Homma, Y. Yamakawa, M. Hirayama, R. Kanno, 3, 1094.
M. Yonemura, T. Kamiyama, Y. Kato, S. Hama, K. Kawamoto, [46] S. Jiao, J. Zheng, Q. Li, X. Li, M. H. Engelhard, R. Cao, J.-G. Zhang,
A. Mitsui, Nat. Mater. 2011, 10, 682. W. Xu, Joule 2018, 2, 110.
[16] R. Kanno, M. Maruyama, J. Electrochem. Soc. 2001, 148, A742. [47] H. Yuan, H. X. Nan, C. Z. Zhao, G. L. Zhu, Y. Lu, X. B. Cheng,
[17] H. J. Deiseroth, S. T. Kong, H. Eckert, J. Vannahme, C. Reiner, Q. B. Liu, C. X. He, J. Q. Huang, Q. Zhang, Batteries Supercaps
T. Zaiss, M. Schlosser, Angew. Chem., Int. Ed. 2008, 47, 755. 2020, 3, 596.
[18] P. Adeli, J. D. Bazak, K. H. Park, I. Kochetkov, A. Huq, [48] C. Niu, H. Lee, S. Chen, Q. Li, J. Du, W. Xu, J.-G. Zhang,
G. R. Goward, L. F. Nazar, Angew. Chem., Int. Ed. 2019, 58, 8681. M. S. Whittingham, J. Xiao, J. Liu, Nat. Energy 2019, 4, 551.
[19] M. A. Kraft, S. Ohno, T. Zinkevich, R. Koerver, S. P. Culver, [49] X. Q. Zhang, T. Li, B. Q. Li, R. Zhang, P. Shi, C. Yan, J. Q. Huang,
T. Fuchs, A. Senyshyn, S. Indris, B. J. Morgan, W. G. Zeier, J. Am. Q. Zhang, Angew. Chem., Int. Ed. 2020, 59, 3252.
Chem. Soc. 2018, 140, 16330. [50] Y.-G. Lee, S. Fujiki, C. Jung, N. Suzuki, N. Yashiro, R. Omoda,
[20] M. A. Kraft, S. P. Culver, M. Calderon, F. Böcher, T. Krauskopf, D.-S. Ko, T. Shiratsuchi, T. Sugimoto, S. Ryu, J. H. Ku, T. Watanabe,
A. Senyshyn, C. Dietrich, A. Zevalkink, J. Janek, W. G. Zeier, J. Am. Y. Park, Y. Aihara, D. Im, I. T. Han, Nat. Energy 2020, 5, 299.
Chem. Soc. 2017, 139, 10909. [51] A. Sharafi, H. M. Meyer, J. Nanda, J. Wolfenstine, J. Sakamoto,
[21] E. Rangasamy, Z. Liu, M. Gobet, K. Pilar, G. Sahu, W. Zhou, J. Power Sources 2016, 302, 135.
H. Wu, S. Greenbaum, C. Liang, J. Am. Chem. Soc. 2015, 137, 1384. [52] T. Krauskopf, H. Hartmann, W. G. Zeier, J. Janek, ACS Appl. Mater.
[22] Y. X. Wang, D. P. Lu, J. Xiao, Y. He, G. J. Harvey, C. M. Wang, Interfaces 2019, 11, 14463.
J. G. Zhang, J. Liu, Energy Storage Mater. 2019, 19, 80. [53] L. Viswanathan, A. V. Virkar, J. Mater. Sci. 1982, 17, 753.
[23] S. J. Sedlmaier, S. Indris, C. Dietrich, M. Yavuz, C. Dräger, [54] A. Pei, G. Zheng, F. Shi, Y. Li, Y. Cui, Nano Lett. 2017, 17, 1132.
F. von Seggern, H. Sommer, J. Janek, Chem. Mater. 2017, 29, [55] X. B. Cheng, R. Zhang, C. Z. Zhao, F. Wei, J. G. Zhang, Q. Zhang,
1830. Adv. Sci. 2016, 3, 1500213.
[24] Y. Lu, S. Gu, X. Hong, K. Rui, X. Huang, J. Jin, C. Chen, J. Yang, [56] X. Shen, R. Zhang, X. Chen, X. B. Cheng, X. Li, Q. Zhang, Adv.
Z. Wen, Energy Storage Mater. 2018, 11, 16. Energy Mater. 2020, 10, 1903645.
[25] X. H. Yu, J. B. Bates, G. E. Jellison, F. X. Hart, J. Electrochem. Soc. [57] S. S. Shishvan, N. A. Fleck, R. M. McMeeking, V. S. Deshpande,
1997, 144, 524. J. Power Sources 2020, 456, 227989.
[26] K. Kisu, S. Kim, R. Yoshida, H. Oguchi, N. Toyama, S.-i. Orimo, [58] R. Raj, J. Wolfenstine, J. Power Sources 2017, 343, 119.
J. Energy Chem. 2020, 50, 424. [59] M. Motoyama, M. Ejiri, Y. Iriyama, J. Electrochem. Soc. 2015, 162,
[27] C.-Y. Chiang, Y. J. Shen, M. J. Reddy, P. P. Chu, J. Power Sources A7067.
2003, 123, 222. [60] F. Sagane, K.-i. Ikeda, K. Okita, H. Sano, H. Sakaebe, Y. Iriyama,
[28] H. Zhang, J. Zhang, J. Ma, G. Xu, T. Dong, G. Cui, Electrochem. J. Power Sources 2013, 233, 34.
Energy Rev. 2019, 2, 128. [61] Z. Ahmad, V. Viswanathan, Phys. Rev. Lett. 2017, 119, 056003.
[29] W. Zhou, Z. Wang, Y. Pu, Y. Li, S. Xin, X. Li, J. Chen, [62] F. Han, Y. Zhu, X. He, Y. Mo, C. Wang, Adv. Energy Mater. 2016, 6,
J. B. Goodenough, Adv. Mater. 2019, 31, 1805574. 1501590.
[30] S. Wang, P. Xiong, J. Zhang, G. Wang, Energy Storage Mater. 2020, [63] C. Ma, Y. Cheng, K. Yin, J. Luo, A. Sharafi, J. Sakamoto, J. Li,
29, 310. K. L. More, N. J. Dudney, M. Chi, Nano Lett. 2016, 16, 7030.
[31] S. Ramakumar, C. Deviannapoorani, L. Dhivya, L. S. Shankar, [64] T. Krauskopf, B. Mogwitz, H. Hartmann, D. K. Singh, W. G. Zeier,
R. Murugan, Prog. Mater. Sci. 2017, 88, 325. J. Janek, Adv. Energy Mater. 2020, 10, 2000945.
[65] M. Chiku, W. Tsujiwaki, E. Higuchi, H. Inoue, J. Power Sources [97] B. N. J. Persson, Surf. Sci. Rep. 2006, 61, 201.
2013, 244, 675. [98] H.-K. Tian, Y. Qi, J. Electrochem. Soc. 2017, 164, E3512.
[66] M. Chiku, W. Tsujiwaki, E. Higuchi, H. Inoue, Electrochemistry [99] H. Tang, Z. Deng, Z. Lin, Z. Wang, I.-H. Chu, C. Chen, Z. Zhu,
2012, 80, 740. C. Zheng, S. P. Ong, Chem. Mater. 2017, 30, 163.
[67] T. Krauskopf, R. Dippel, H. Hartmann, K. Peppler, B. Mogwitz, [100] H. Chung, B. Kang, Chem. Mater. 2017, 29, 8611.
F. H. Richter, W. G. Zeier, J. Janek, Joule 2019, 3, 2030. [101] S. Wenzel, S. Randau, T. Leichtweiß, D. A. Weber, J. Sann,
[68] H. Koshikawa, S. Matsuda, K. Kamiya, M. Miyayama, Y. Kubo, W. G. Zeier, J. Janek, Chem. Mater. 2016, 28, 2400.
K. Uosaki, K. Hashimoto, S. Nakanishi, J. Power Sources 2018, 376, [102] J. A. Lewis, F. J. Q. Cortes, M. G. Boebinger, J. Tippens,
147. T. S. Marchese, N. Kondekar, X. Liu, M. Chi, M. T. McDowell, ACS
[69] K. Kerman, A. Luntz, V. Viswanathan, Y.-M. Chiang, Z. Chen, Energy Lett. 2019, 4, 591.
J. Electrochem. Soc. 2017, 164, A1731. [103] L. E. Camacho-Forero, P. B. Balbuena, J. Power Sources 2018, 396,
[70] L. Cheng, W. Chen, M. Kunz, K. Persson, N. Tamura, G. Chen, 782.
M. Doeff, ACS Appl. Mater. Interfaces 2015, 7, 2073. [104] A. Mirsakiyeva, M. Ebadi, C. M. Araujo, D. Brandell, P. Broqvist,
[71] C. Wang, Y. Gong, J. Dai, L. Zhang, H. Xie, G. Pastel, B. Liu, J. Kullgren, J. Phys. Chem. C 2019, 123, 22851.
E. Wachsman, H. Wang, L. Hu, J. Am. Chem. Soc. 2017, 139, 14257. [105] M. Golozar, P. Hovington, A. Paolella, S. Bessette, M. Lagace,
[72] M. Motoyama, Y. Tanaka, T. Yamamoto, N. Tsuchimine, P. Bouchard, H. Demers, R. Gauvin, K. Zaghib, Nano Lett. 2018,
S. Kobayashi, Y. Iriyama, ACS Appl. Energy Mater. 2019, 2, 6720. 18, 7583.
[73] K. Yan, Z. Lu, H.-W. Lee, F. Xiong, P.-C. Hsu, Y. Li, J. Zhao, S. Chu, [106] S. Wenzel, T. Leichtweiss, D. Krüger, J. Sann, J. Janek, Solid State
Y. Cui, Nat. Energy 2016, 1, 16010. Ionics 2015, 278, 98.
[74] A. Kato, H. Kowada, M. Deguchi, C. Hotehama, A. Hayashi, [107] S. Wenzel, S. J. Sedlmaier, C. Dietrich, W. G. Zeier, J. Janek, Solid
M. Tatsumisago, Solid State Ionics 2018, 322, 1. State Ionics 2018, 318, 102.
[75] H. Nagata, Y. Chikusa, J. Power Sources 2014, 264, 206. [108] L. Sang, R. T. Haasch, A. A. Gewirth, R. G. Nuzzo, Chem. Mater.
[76] F. Sun, K. Dong, M. Osenberg, A. Hilger, S. Risse, Y. Lu, 2017, 29, 3029.
P. H. Kamm, M. Klaus, H. Markötter, F. García-Moreno, T. Arlt, [109] D. H. S. Tan, E. A. Wu, H. Nguyen, Z. Chen, M. A. T. Marple,
I. Manke, J. Mater. Chem. A 2018, 6, 22489. J.-M. Doux, X. Wang, H. Yang, A. Banerjee, Y. S. Meng, ACS Energy
[77] T. Krauskopf, B. Mogwitz, C. Rosenbach, W. G. Zeier, J. Janek, Adv. Lett. 2019, 4, 2418.
Energy Mater. 2019, 9, 1902568. [110] M. Suyama, A. Kato, A. Sakuda, A. Hayashi, M. Tatsumisago, Elec-
[78] C. Monroe, J. Newman, J. Electrochem. Soc. 2003, 150, A1377. trochim. Acta 2018, 286, 158.
[79] C.-Z. Zhao, H. Duan, J.-Q. Huang, J. Zhang, Q. Zhang, Y.-G. Guo, [111] S. Wenzel, D. A. Weber, T. Leichtweiss, M. R. Busche, J. Sann,
L.-J. Wan, Sci. China: Chem. 2019, 62, 1286. J. Janek, Solid State Ionics 2016, 286, 24.
[80] E. J. Cheng, A. Sharafi, J. Sakamoto, Electrochim. Acta 2017, 223, [112] J. Wolfenstine, J. L. Allen, J. Read, J. Sakamoto, J. Mater. Sci. 2013,
85. 48, 5846.
[81] Y. Shi, J. Wan, G. X. Liu, T. T. Zuo, Y. X. Song, B. Liu, Y. G. Guo, [113] Y. Jin, K. Liu, J. Lang, D. Zhuo, Z. Huang, C.-a. Wang, H. Wu,
R. Wen, L. J. Wan, Angew. Chem., Int. Ed. 2020, 59, 18120. Y. Cui, Nat. Energy 2018, 3, 732.
[82] D. Devaux, K. J. Harry, D. Y. Parkinson, R. Yuan, D. T. Hallinan, [114] W. Manalastas, J. Rikarte, R. J. Chater, R. Brugge, A. Aguadero,
A. A. MacDowell, N. P. Balsara, J. Electrochem. Soc. 2015, 162, L. Buannic, A. Llordés, F. Aguesse, J. Kilner, J. Power Sources 2019,
A1301. 412, 287.
[83] E. Kazyak, R. Garcia-Mendez, W. S. LePage, A. Sharafi, A. L. Davis, [115] C. Wang, W. Ping, Q. Bai, H. Cui, R. Hensleigh, R. Wang,
A. J. Sanchez, K. H. Chen, C. Haslam, J. Sakamoto, N. P. Dasgupta, A. H. Brozena, Z. Xu, J. Dai, Y. Pei, C. Zheng, G. Pastel, J. Gao,
Matter 2020, 2, 1025. X. Wang, H. Wang, J. C. Zhao, B. Yang, X. R. Zheng, J. Luo, Y. Mo,
[84] Y. Lu, X. Huang, Z. Song, K. Rui, Q. Wang, S. Gu, J. Yang, T. Xiu, B. Dunn, L. Hu, Science 2020, 368, 521.
M. E. Badding, Z. Wen, Energy Storage Mater. 2018, 15, 282. [116] Y. Zhu, X. He, Y. Mo, J. Mater. Chem. A 2016, 4, 3253.
[85] P. Barai, K. Higa, A. T. Ngo, L. A. Curtiss, V. Srinivasan, J. Electro- [117] X. Huang, C. Liu, Y. Lu, T. Xiu, J. Jin, M. E. Badding, Z. Wen,
chem. Soc. 2019, 166, A1752. J. Power Sources 2018, 382, 190.
[86] S. Yu, D. J. Siegel, ACS Appl. Mater. Interfaces 2018, 10, 38151. [118] X. Huang, Y. Lu, Z. Song, T. Xiu, M. E. Badding, Z. Wen, J. Energy
[87] X. Huang, L. Wu, Z. Huang, J. Lin, X. Xu, Energy Storage Sci. Chem. 2019, 39, 8.
Technol. 2020, 9, 479. [119] J. Su, X. Huang, Z. Song, T. Xiu, M. E. Badding, J. Jin, Z. Wen,
[88] Y. Lu, X. Huang, Y. Ruan, Q. Wang, R. Kun, J. Yang, Z. Wen, Ceram. Int. 2019, 45, 14991.
J. Mater. Chem. A 2018, 6, 18853. [120] F. Shen, M. B. Dixit, X. Xiao, K. B. Hatzell, ACS Energy Lett. 2018,
[89] X. Huang, Y. Lu, Z. Song, K. Rui, Q. Wang, T. Xiu, M. E. Badding, 3, 1056.
Z. Wen, Energy Storage Mater. 2019, 22, 207. [121] T. Swamy, R. Park, B. W. Sheldon, D. Rettenwander, L. Porz,
[90] W. Ping, C. Wang, Z. Lin, E. Hitz, C. Yang, H. Wang, L. Hu, Adv. S. Berendts, R. Uecker, W. C. Carter, Y.-M. Chiang, J. Electrochem.
Energy Mater. 2020, 10, 2000702. Soc. 2018, 165, A3648.
[91] P. Wang, W. Qu, W. L. Song, H. Chen, R. Chen, D. Fang, Adv. Funct. [122] L. Porz, T. Swamy, B. W. Sheldon, D. Rettenwander, T. Frömling,
Mater. 2019, 29, 1900950. H. L. Thaman, S. Berendts, R. Uecker, W. C. Carter, Y.-M. Chiang,
[92] Y. Ruan, Y. Lu, Y. Li, C. Zheng, J. Su, J. Jin, T. Xiu, Z. Song, Adv. Energy Mater. 2017, 7, 1701003.
M. E. Badding, Z. Wen, Adv. Funct. Mater. 2020, 31, 2007815. [123] Y. Zhu, J. G. Connell, S. Tepavcevic, P. Zapol, R. Garcia-Mendez,
[93] K. Hayamizu, Y. Terada, K. Kataoka, J. Akimoto, J. Chem. Phys. N. J. Taylor, J. Sakamoto, B. J. Ingram, L. A. Curtiss, J. W. Freeland,
2019, 150, 194502. D. D. Fong, N. M. Markovic, Adv. Energy Mater. 2019, 9,
[94] Q. Zheng, D. M. Pesko, B. M. Savoie, K. Timachova, A. L. Hasan, 1803440.
M. C. Smith, T. F. Miller, G. W. Coates, N. P. Balsara, Macromol- [124] D. Rettenwander, R. Wagner, A. Reyer, M. Bonta, L. Cheng,
ecules 2018, 51, 2847. M. M. Doeff, A. Limbeck, M. Wilkening, G. Amthauer, J. Phys.
[95] A. Banerjee, X. Wang, C. Fang, E. A. Wu, Y. S. Meng, Chem. Rev. Chem. C 2018, 122, 3780.
2020, 120, 6878. [125] Federico M. Pesci, R. H. Brugge, A. K. O. Hekselman, A. Cavallaro,
[96] R. Usiskin, J. Maier, Adv. Energy Mater. 2021, 11, 2001455. R. J. Chater, A. Aguadero, J. Mater. Chem. A 2018, 6, 19817.
[126] C. Yang, H. Xie, W. Ping, K. Fu, B. Liu, J. Rao, J. Dai, C. Wang, [156] C. Yang, L. Zhang, B. Liu, S. Xu, T. Hamann, D. McOwen, J. Dai,
G. Pastel, L. Hu, Adv. Mater. 2019, 31, 1804815. W. Luo, Y. Gong, E. D. Wachsman, L. Hu, Proc. Natl. Acad. Sci.
[127] Z. Xue, D. He, X. Xie, J. Mater. Chem. A 2015, 3, 19218. USA 2018, 115, 3770.
[128] K. Timachova, H. Watanabe, N. P. Balsara, Macromolecules 2015, [157] G. T. Hitz, D. W. McOwen, L. Zhang, Z. Ma, Z. Fu, Y. Wen,
48, 7882. Y. Gong, J. Dai, T. R. Hamann, L. Hu, E. D. Wachsman, Mater.
[129] L. O. Valo̸en, J. N. Reimers, J. Electrochem. Soc. 2005, 152, A882. Today 2019, 22, 50.
[130] A. Masias, N. Felten, R. Garcia-Mendez, J. Wolfenstine, [158] S. Xu, D. W. McOwen, C. Wang, L. Zhang, W. Luo, C. Chen, Y. Li,
J. Sakamoto, J. Mater. Sci. 2018, 54, 2585. Y. Gong, J. Dai, Y. Kuang, C. Yang, T. R. Hamann, E. D. Wachsman,
[131] M. Wang, J. Sakamoto, J. Power Sources 2018, 377, 7. L. Hu, Nano Lett. 2018, 18, 3926.
[132] J. Kasemchainan, S. Zekoll, D. Spencer Jolly, Z. Ning, G. O. Hartley, [159] Q. Li, T. Yi, X. Wang, H. Pan, B. Quan, T. Liang, X. Guo, X. Yu,
J. Marrow, P. G. Bruce, Nat. Mater. 2019, 18, 1105. H. Wang, X. Huang, L. Chen, H. Li, Nano Energy 2019, 63, 103895.
[133] M. J. Wang, R. Choudhury, J. Sakamoto, Joule 2019, 3, 2165. [160] Y. Zhu, X. He, Y. Mo, Adv. Sci. 2017, 4, 1600517.
[134] P. Barai, K. Higa, V. Srinivasan, Phys. Chem. Chem. Phys. 2017, 19, [161] L. Sang, K. L. Bassett, F. C. Castro, M. J. Young, L. Chen,
20493. R. T. Haasch, J. W. Elam, V. P. Dravid, R. G. Nuzzo, A. A. Gewirth,
[135] J. M. Doux, H. Nguyen, D. H. S. Tan, A. Banerjee, X. Wang, Chem. Mater. 2018, 30, 8747.
E. A. Wu, C. Jo, H. Yang, Y. S. Meng, Adv. Energy Mater. 2019, 10, [162] T. Nakamura, K. Amezawa, J. Kulisch, W. G. Zeier, J. Janek, ACS
1903253. Appl. Mater. Interfaces 2019, 11, 19968.
[136] A. J. Louli, J. Li, S. Trussler, C. R. Fell, J. R. Dahn, J. Electrochem. [163] A. M. Nolan, Y. Zhu, X. He, Q. Bai, Y. Mo, Joule 2018, 2, 2016.
Soc. 2017, 164, A2689. [164] Q. Cheng, A. Li, N. Li, S. Li, A. Zangiabadi, T.-D. Li, W. Huang,
[137] G. Bucci, J. Christensen, J. Power Sources 2019, 441, 227186. A. C. Li, T. Jin, Q. Song, W. Xu, N. Ni, H. Zhai, M. Dontigny,
[138] W. S. LePage, Y. Chen, E. Kazyak, K.-H. Chen, A. J. Sanchez, A. Poli, K. Zaghib, X. Chuan, D. Su, K. Yan, Y. Yang, Joule 2019, 3, 1510.
E. M. Arruda, M. D. Thouless, N. P. Dasgupta, J. Electrochem. Soc. [165] Y. Liu, C. Li, B. Li, H. Song, Z. Cheng, M. Chen, P. He, H. Zhou,
2019, 166, A89. Adv. Energy Mater. 2018, 8, 1702374.
[139] J. Tippens, J. C. Miers, A. Afshar, J. A. Lewis, F. J. Q. Cortes, [166] X. Fan, X. Ji, F. Han, J. Yue, J. Chen, L. Chen, T. Deng, J. Jiang,
H. Qiao, T. S. Marchese, C. V. Di Leo, C. Saldana, M. T. McDowell, C. Wang, Sci. Adv. 2018, 4, eaau9245.
ACS Energy Lett. 2019, 4, 1475. [167] F. Han, J. Yue, X. Zhu, C. Wang, Adv. Energy Mater. 2018, 8,
[140] C. Liu, K. Rui, C. Shen, M. E. Badding, G. Zhang, Z. Wen, J. Power 1703644.
Sources 2015, 282, 286. [168] R. Xu, F. Han, X. Ji, X. Fan, J. Tu, C. Wang, Nano Energy 2018, 53,
[141] R. H. Brugge, A. K. O. Hekselman, A. Cavallaro, F. M. Pesci, 958.
R. J. Chater, J. A. Kilner, A. Aguadero, Chem. Mater. 2018, 30, 3704. [169] Y. Gao, D. Wang, Y. C. Li, Z. Yu, T. E. Mallouk, D. Wang, Angew.
[142] J. Wang, H. Wang, J. Xie, A. Yang, A. Pei, C.-L. Wu, F. Shi, Y. Liu, Chem., Int. Ed. 2018, 57, 13608.
D. Lin, Y. Gong, Y. Cui, Energy Storage Mater. 2018, 14, 345. [170] N. W. Li, Y. X. Yin, C. P. Yang, Y. G. Guo, Adv. Mater. 2016, 28, 1853.
[143] C. T. Zhou, A. J. Samson, K. Hofstetter, V. Thangadurai, Sustain- [171] Z. Zhang, S. Chen, J. Yang, J. Wang, L. Yao, X. Yao, P. Cui, X. Xu,
able Energy Fuels 2018, 2, 2165. ACS Appl. Mater. Interfaces 2018, 10, 2556.
[144] C. L. Tsai, V. Roddatis, C. V. Chandran, Q. Ma, S. Uhlenbruck, [172] S. Wang, J. Wang, J. Liu, H. Song, Y. Liu, P. Wang, P. He, J. Xu,
M. Bram, P. Heitjans, O. Guillon, ACS Appl. Mater. Interfaces 2016, H. Zhou, J. Mater. Chem. A 2018, 6, 21248.
8, 10617. [173] X. Huang, C. Shen, K. Rui, J. Jin, M. Wu, X. Wu, Z. Wen, JOM 2016,
[145] W. Luo, Y. Gong, Y. Zhu, K. K. Fu, J. Dai, S. D. Lacey, C. Wang, 68, 2593.
B. Liu, X. Han, Y. Mo, E. D. Wachsman, L. Hu, J. Am. Chem. Soc. [174] R. Hongahally Basappa, T. Ito, T. Morimura, R. Bekarevich,
2016, 138, 12258. K. Mitsuishi, H. Yamada, J. Power Sources 2017, 363, 145.
[146] K. K. Fu, Y. Gong, B. Liu, Y. Zhu, S. Xu, Y. Yao, W. Luo, C. Wang, [175] Y. Song, L. Yang, W. Zhao, Z. Wang, Y. Zhao, Z. Wang, Q. Zhao,
S. D. Lacey, J. Dai, Y. Chen, Y. Mo, E. Wachsman, L. Hu, Sci. Adv. H. Liu, F. Pan, Adv. Energy Mater. 2019, 9, 1900671.
2017, 3, e1601659. [176] B. Xu, W. Li, H. Duan, H. Wang, Y. Guo, H. Li, H. Liu, J. Power
[147] J. F. Wu, B. W. Pu, D. Wang, S. Q. Shi, N. Zhao, X. Guo, X. Guo, Sources 2017, 354, 68.
ACS Appl. Mater. Interfaces 2019, 11, 898. [177] F. Zhao, Q. Sun, C. Yu, S. Zhang, K. Adair, S. Wang, Y. Liu, Y. Zhao,
[148] Y. Li, X. Chen, A. Dolocan, Z. Cui, S. Xin, L. Xue, H. Xu, K. Park, J. Liang, C. Wang, X. Li, X. Li, W. Xia, R. Li, H. Huang, L. Zhang,
J. B. Goodenough, J. Am. Chem. Soc. 2018, 140, 6448. S. Zhao, S. Lu, X. Sun, ACS Energy Lett. 2020, 5, 1035.
[149] Y. Ruan, Y. Lu, X. Huang, J. Su, C. Sun, J. Jin, Z. Wen, J. Mater. [178] B. M. Savoie, M. A. Webb, T. F. Miller, 3rd, J. Phys. Chem. Lett.
Chem. A 2019, 7, 14565. 2017, 8, 641.
[150] J. Duan, W. Wu, A. M. Nolan, T. Wang, J. Wen, C. Hu, Y. Mo, [179] C. Z. Zhao, X. Q. Zhang, X. B. Cheng, R. Zhang, R. Xu, P. Y. Chen,
W. Luo, Y. Huang, Adv. Mater. 2019, 31, 1807243. H. J. Peng, J. Q. Huang, Q. Zhang, Proc. Natl. Acad. Sci. U. S. A.
[151] J. Wen, Y. Huang, J. Duan, Y. Wu, W. Luo, L. Zhou, C. Hu, 2017, 114, 11069.
L. Huang, X. Zheng, W. Yang, Z. Wen, Y. Huang, ACS Nano 2019, [180] G. Jo, H. Jeon, M. J. Park, ACS Macro Lett. 2015, 4, 225.
13, 14549. [181] K. M. Diederichsen, E. J. McShane, B. D. McCloskey, ACS Energy
[152] Y. Huang, B. Chen, J. Duan, F. Yang, T. Wang, Z. Wang, Lett. 2017, 2, 2563.
W. Yang, C. Hu, W. Luo, Y. Huang, Angew. Chem., Int. Ed. 2020, 59, [182] W. Zhou, S. Wang, Y. Li, S. Xin, A. Manthiram, J. B. Goodenough,
3699. J. Am. Chem. Soc. 2016, 138, 9385.
[153] Z. Lin, Z. Liu, W. Fu, N. J. Dudney, C. Liang, Angew. Chem., Int. Ed. [183] Q. Wang, J. Jin, X. Wu, G. Ma, J. Yang, Z. Wen, Phys. Chem. Chem.
2013, 52, 7460. Phys. 2014, 16, 21225.
[154] S.-S. Chi, Y. Liu, N. Zhao, X. Guo, C.-W. Nan, L.-Z. Fan, Energy [184] X. Zhang, T. Liu, S. Zhang, X. Huang, B. Xu, Y. Lin, B. Xu, L. Li,
Storage Mater. 2019, 17, 309. C. W. Nan, Y. Shen, J. Am. Chem. Soc. 2017, 139, 13779.
[155] K. Fu, Y. H. Gong, G. T. Hitz, D. W. McOwen, Y. J. Li, S. M. Xu, [185] Y. Li, B. Xu, H. Xu, H. Duan, X. Lü, S. Xin, W. Zhou, L. Xue, G. Fu,
Y. Wen, L. Zhang, C. W. Wang, G. Pastel, J. Q. Dai, B. Y. Liu, H. Xie, A. Manthiram, J. B. Goodenough, Angew. Chem., Int. Ed. 2017, 56,
Y. G. Yao, E. D. Wachsman, L. B. Hu, Energy Environ. Sci. 2017, 10, 753.
1568. [186] K. Yoshima, Y. Harada, N. Takami, J. Power Sources 2016, 302, 283.
[187] C.-Z. Zhao, B.-C. Zhao, C. Yan, X.-Q. Zhang, J.-Q. Huang, Y. Mo, [209] N. J. Taylor, S. Stangeland-Molo, C. G. Haslam, A. Sharafi,
X. Xu, H. Li, Q. Zhang, Energy Storage Mater. 2020, 24, 75. T. Thompson, M. Wang, R. Garcia-Mendez, J. Sakamoto, J. Power
[188] J. Y. Liang, X. X. Zeng, X. D. Zhang, T. T. Zuo, M. Yan, Y. X. Yin, Sources 2018, 396, 314.
J. L. Shi, X. W. Wu, Y. G. Guo, L. J. Wan, J. Am. Chem. Soc. 2019, [210] Y. Tian, F. Ding, H. Zhong, C. Liu, Y.-B. He, J. Liu, X. Liu, Q. Xu,
141, 9165. Energy Storage Mater. 2018, 14, 49.
[189] D. Y. Oh, Y. J. Nam, K. H. Park, S. H. Jung, K. T. Kim, A. R. Ha, [211] X. Fu, T. Wang, W. Shen, M. Jiang, Y. Wang, Q. Dai, D. Wang,
Y. S. Jung, Adv. Energy Mater. 2019, 9, 1802927. Z. Qiu, Y. Zhang, K. Deng, Q. Zeng, N. Zhao, X. Guo, Z. Liu, J. Liu,
[190] S. Li, H. Wang, J. Cuthbert, T. Liu, J. F. Whitacre, K. Matyjaszewski, Z. Peng, Adv. Mater. 2020, 32, 2000575.
Joule 2019, 3, 1637. [212] Z. Cheng, M. Xie, Y. Mao, J. Ou, S. Zhang, Z. Zhao, J. Li, F. Fu,
[191] X. Liu, X. Li, H. Li, H. B. Wu, Chem. - Eur. J. 2018, 24, 18293. J. Wu, Y. Shen, D. Lu, H. Chen, Adv. Energy Mater. 2020, 10,
[192] Q. S. Wang, Z. Y. Wen, J. Jin, J. Guo, X. Huang, J. H. Yang, 1904230.
C. H. Chen, Chem. Commun. 2016, 52, 1637. [213] Z. Zhang, L. Zhang, X. Yan, H. Wang, Y. Liu, C. Yu, X. Cao,
[193] M. Keller, A. Varzi, S. Passerini, J. Power Sources 2018, 392, 206. L. van Eijck, B. Wen, J. Power Sources 2019, 410–411, 162.
[194] Z. Zhang, L. Zhang, Y. Liu, H. Wang, C. Yu, H. Zeng, L. M. Wang, [214] R. Zhao, S. Kmiec, G. Hu, S. W. Martin, ACS Appl. Mater. Interfaces
B. Xu, ChemSusChem 2018, 11, 3774. 2020, 12, 2327.
[195] C. Yang, K. Fu, Y. Zhang, E. Hitz, L. Hu, Adv. Mater. 2017, 29, [215] A. Kato, M. Suyama, C. Hotehama, H. Kowada, A. Sakuda,
1701169. A. Hayashi, M. Tatsumisago, J. Electrochem. Soc. 2018, 165,
[196] C. Z. Zhao, P. Y. Chen, R. Zhang, X. Chen, B. Q. Li, X. Q. Zhang, A1950.
X. B. Cheng, Q. Zhang, Sci. Adv. 2018, 4, eaat3446. [216] A. L. Santhosha, L. Medenbach, J. R. Buchheim, P. Adelhelm, Bat-
[197] R. Xu, Y. Xiao, R. Zhang, X. B. Cheng, C. Z. Zhao, X. Q. Zhang, teries Supercaps 2019, 2, 524.
C. Yan, Q. Zhang, J. Q. Huang, Adv. Mater. 2019, 31, 1808392. [217] Y. Su, L. Ye, W. Fitzhugh, Y. Wang, E. Gil-González, I. Kim, X. Li,
[198] D. Xu, J. Su, J. Jin, C. Sun, Y. Ruan, C. Chen, Z. Wen, Adv. Energy Energy Environ. Sci. 2020, 13, 908.
Mater. 2019, 9, 1900611. [218] L. Chen, Y. Li, S.-P. Li, L.-Z. Fan, C.-W. Nan, J. B. Goodenough,
[199] X. Q. Zhang, X. Chen, X. B. Cheng, B. Q. Li, X. Shen, C. Yan, Nano Energy 2018, 46, 176.
J. Q. Huang, Q. Zhang, Angew. Chem., Int. Ed. 2018, 57, 5301. [219] Y. Liu, D. Lin, Y. Jin, K. Liu, X. Tao, Q. Zhang, X. Zhang, Y. Cui, Sci.
[200] Y. J. Gong, J. W. Heo, H. Lee, H. Kim, J. Cho, S. Pyo, H. Yun, H. Kim, Adv. 2017, 3, eaao0713.
S. Y. Park, J. Yoo, Y. S. Kim, Adv. Energy Mater. 2020, 10, 2001479. [220] L. Chen, W. Li, L. Z. Fan, C. W. Nan, Q. Zhang, Adv. Funct. Mater.
[201] S. Tang, Q. Lan, L. Xu, J. Liang, P. Lou, C. Liu, L. Mai, Y.-C. Cao, 2019, 29, 1901047.
S. Cheng, Nano Energy 2020, 71, 104600. [221] S. Chen, J. Wang, Z. Zhang, L. Wu, L. Yao, Z. Wei, Y. Deng, D. Xie,
[202] X. Huang, Y. Lu, H. Guo, Z. Song, T. Xiu, M. E. Badding, Z. Wen, X. Yao, X. Xu, J. Power Sources 2018, 387, 72.
ACS Appl. Energy Mater. 2018, 1, 5355. [222] J. F. Wu, X. Guo, Small 2019, 15, 1804413.
[203] Y. Shao, H. Wang, Z. Gong, D. Wang, B. Zheng, J. Zhu, Y. Lu, [223] J. Zhang, N. Zhao, M. Zhang, Y. Li, P. K. Chu, X. Guo, Z. Di,
Y.-S. Hu, X. Guo, H. Li, X. Huang, Y. Yang, C.-W. Nan, L. Chen, X. Wang, H. Li, Nano Energy 2016, 28, 447.
ACS Energy Lett. 2018, 3, 1212. [224] K. K. Fu, Y. Gong, J. Dai, A. Gong, X. Han, Y. Yao, C. Wang,
[204] X. Han, Y. Gong, K. K. Fu, X. He, G. T. Hitz, J. Dai, A. Pearse, B. Liu, Y. Wang, Y. Chen, C. Yan, Y. Li, E. D. Wachsman, L. Hu, Proc. Natl.
H. Wang, G. Rubloff, Y. Mo, V. Thangadurai, E. D. Wachsman, Acad. Sci. USA 2016, 113, 7094.
L. Hu, Nat. Mater. 2017, 16, 572. [225] R. Rojaee, S. Cavallo, S. Mogurampelly, B. K. Wheatle, V. Yurkiv,
[205] C. Wang, Y. Gong, B. Liu, K. Fu, Y. Yao, E. Hitz, Y. Li, J. Dai, S. Xu, R. Deivanayagam, T. Foroozan, M. G. Rasul, S. Sharifi-Asl,
W. Luo, E. D. Wachsman, L. Hu, Nano Lett. 2017, 17, 565. A. H. Phakatkar, M. Cheng, S. B. Son, Y. Pan, F. Mashayek,
[206] J. Lou, G. Wang, Y. Xia, C. Liang, H. Huang, Y. Gan, X. Tao, V. Ganesan, R. Shahbazian-Yassar, Adv. Funct. Mater. 2020, 30,
J. Zhang, W. Zhang, J. Power Sources 2020, 448, 227440. 1910749.
[207] M. Cai, Y. Lu, J. Su, Y. Ruan, C. Chen, B. V. R. Chowdari, Z. Wen, [226] M. C. Bay, M. Wang, R. Grissa, M. V. F. Heinz, J. Sakamoto,
ACS Appl. Mater. Interfaces 2019, 11, 35030. C. Battaglia, Adv. Energy Mater. 2019, 10, 1902899.
[208] H. Zheng, S. Wu, R. Tian, Z. Xu, H. Zhu, H. Duan, H. Liu, Adv. [227] L. Liu, X. Qi, Q. Ma, X. Rong, Y. S. Hu, Z. Zhou, H. Li, X. Huang,
Funct. Mater. 2019, 30, 1906189. L. Chen, ACS Appl. Mater. Interfaces 2016, 8, 32631.
Yang Lu completed his Bachelor’s degree in Tianjin University, and received Ph. D. degree in
Shanghai Institute of Ceramics, Chinese Academy of Science. He joined Prof. Zhang’s group
as Postdoctoral Researcher at Department of Chemical Engineering in Tsinghua University.
His current research focuses on solid-state battery, solid-state electrolyte, and their interfacial
modifications.
Chen-Zi Zhao obtained her Ph.D. from Department of Chemical Engineering (2020), and B. Eng.
from School of Materials Science and Engineering, Tsinghua University (2015). Her main areas of
research address energy-storage materials, including electrolytes and lithium metal anodes.
Hong Yuan received his Ph.D. degree from Beijing Institute of Technology in 2017. Now he is
an Associate Professor in Advanced Research Institute of Multidisciplinary Science (ARIMS)
in Beijing Institute of Technology. His research interests are focused on a fundamental under-
standing of energy chemistry in advanced Li−S batteries, Li-ion batteries, and solid-state batteries.
Xin-Bing Cheng received his B. Eng. from Tianjin University in 2012 and Ph.D. from Tsinghua
University in 2017 under the guidance of Prof. Qiang Zhang. He is currently a Research Assistant
Professor at Tsinghua University. His current research interests are mainly focused on energy
storage materials and devices.
Jia-Qi Huang is a Professor in the Advanced Research Institute for Multidisciplinary Science
at Beijing Institute of Technology. He received his Bachelor and Ph.D. degree in Chemical
Engineering from Tsinghua University. His research focuses on interface electrochemistry and
advanced energy materials design in high-energy-density rechargeable batteries, especially for
lithium-sulfur batteries and lithium metal anode.
Qiang Zhang is a Full Professor at Tsinghua University. He received his Bachelor and Ph.D.
degree from Tsinghua University, China in 2004 and 2009, respectively. He subsequently held
Research Associate/Postdoc Research Fellow positions in the Case Western Reserve University,
USA, and Fritz Haber Institute of the Max Planck Society, Germany. His current research inter-
ests are advanced energy materials, including dendrite-free lithium metal anode, lithium sulfur
batteries, and electrocatalysis, especially the structure design and full demonstration of advanced
energy materials in working devices.

Critical Current Density in Solid-State Lithium Metal Batteries

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Critical Current Density in Solid-State Lithium Metal Batteries

Uploaded by

Copyright:

Available Formats

16163028, 2021, 18, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/adfm.202009925 by Institute Of Chemical Technology, Mumbai, Wiley Online Library on [30/12/2022].

Critical Current Density in Solid-State Lithium Metal

electrolytes (lithium phosphorus oxyni-

2.2. Interfacial Lithium Chemistry −1

π hγ 2 mainly contributes the Rint, because of the extremely weak RSEI.

2.2.3. Interface Failure (2) Partly short circuit

3. The Influence Factors of CCD 3.1. Interfacial Contact

There is a reversible transition between Li0 (Li atom) and Li+  c V (ξ = 0 )  o DV

Table 2. The state-of-the-art CCD values with cycling protocols.

electric conductivity (Figure 19d). Hence batteries with porous

The diffusion issues usually act as the rate-determining steps

You might also like