Highly Stable Lithium Metal Anode Constructed by Three-Dimensional Lithiophilic Materials

: Although lithium metal anode has irreplaceable advantages, such as ultra-high specific energy density and ultra-low redox potential, a variety of issues, i.e., short cycle life, low Coulomb efficiency, and tendency to cause fire explosions caused by lithium dendrite growth and high reactivity to the electrolyte, seriously hinder the practical progress of lithium metal anode. This perspective summarizes how 3D lithiophilic materials have stabilized lithium metal anodes in recent years by improving the uneven deposition of lithium metal, alleviating the volume expansion of lithium metal anodes, and limiting dendrite growth. Simultaneously, the issues of the 3D composite lithium anodes in practical application are concluded and the research direction of 3D composite lithium anode is prospected.


Introduction
High energy density and excellent cycling performance make lithium-ion batteries one of the first choices for electric vehicles and large-scale renewable energy storage systems [1,2].However, at present, the specific energy of LIBs is approaching the limit, and it is urgent that we find a novel battery system with a high specific energy density.Lithium metal anode is well known as the "Holy Grail" in the battery field due to its super-high specific capacity (3860 mAh g −1 ) and the most negative potential (−3.040 V vs. SHE) [1,2].Especially when matched with the high specific energy sulfur anode, the theoretical energy density can reach as high as 2567 Wh kg −1 , which is much higher than that of commercial LIBs [3][4][5][6].However, the electrochemical stripping and deposition reaction of the lithium metal anode is not uniform, which will generate a large number of lithium dendrites that can easily pierce the separator and cause short circuits [7][8][9].After many cycles, it even leads to the deactivation of electrode powder [1].At the same time, as infinite volume change occurs during the lithium metal deposition/stripping process, it is difficult to form a stable solid electrolyte interface film, resulting in reaction reversibility deterioration, poor cycling stability of lithium metal battery cycle, as well as low Coulomb efficiency (CE) [10,11].Therefore, how to construct a stable electrochemical interface and inhibit the uncontrollable formation of lithium dendrites has always been a key scientific problem that needs to be solved urgently in the research of lithium anode.
From the perspective of thermodynamics, the Gibbs free energy during the deposition of lithium in the high-dimensional and low-dimensional deposition phases differs only very slightly [12].Lithium metal is hence more likely to develop dendrites.According to the theory of Kinetics, Li + can easily accumulate at one location to form dendrites because there is insufficient solid-phase diffusion of lithium metal.
Researchers have proposed a variety of effective modification strategies as shown in Figure 1, including liquid electrolyte modification [10,13], construction of solid and quasi-solid electrolytes [2, 10,14,15], design of interface protective film, construction of three-dimensional (3D) conductive skeleton [16][17][18][19][20], etc., to inhibit the lithium dendrite growth, promoting the cycling performance of lithium metal anode.Compared with other modification methods, the 3D conductive skeleton can be used as a carrier for efficient lithium deposition and stripping, which has obvious advantages in solving the dendrite and powder of lithium metal and greatly improving the cycle life of lithium anode [16,17].However, after several cycles of the 3D conductive skeleton, lithium metal will inevitably accumulate on the electrode surface, which will affect the long cycle life of the electrode.In order to solve this problem, researchers proposed a modified strategy of gradient lithophilia [21,22].The so-called gradient lithophilia is to induce metal lithium to preferentially deposit inside or at the bottom of the skeleton by taking advantage of the differences in the binding between different components of the skeleton material and lithium, thus effectively inhibiting the growth of dendrites.
Researchers have proposed a variety of effective modification strategies as shown in Figure 1, including liquid electrolyte modification [10,13], construction of solid and quasisolid electrolytes [2, 10,14,15], design of interface protective film, construction of three-dimensional (3D) conductive skeleton [16][17][18][19][20], etc., to inhibit the lithium dendrite growth, promoting the cycling performance of lithium metal anode.Compared with other modification methods, the 3D conductive skeleton can be used as a carrier for efficient lithium deposition and stripping, which has obvious advantages in solving the dendrite and powder of lithium metal and greatly improving the cycle life of lithium anode [16,17].However, after several cycles of the 3D conductive skeleton, lithium metal will inevitably accumulate on the electrode surface, which will affect the long cycle life of the electrode.In order to solve this problem, researchers proposed a modified strategy of gradient lithophilia [21,22].The so-called gradient lithophilia is to induce metal lithium to preferentially deposit inside or at the bottom of the skeleton by taking advantage of the differences in the binding between different components of the skeleton material and lithium, thus effectively inhibiting the growth of dendrites.This perspective is focusing on the latest progress of 3D lithiophilic materials in stabilizing lithium anodes, such as 3D metal-based materials, 3D carbon materials, and 3D polymers.After analyzing the mechanism of how these 3D lithiophilic materials stabilize the lithium anode, finally, we conclude that to realize the industrialization of lithium metal batteries, the dendrite and volume expansion of lithium should be solved first of all and should combine different modification methods to be studied and explored.The future research direction of 3D composite lithium anode is also prospected.

3D Metal-Based Materials to Stabilize Lithium Anode
According to the theory put forward by Wang et al. [23], if two substances could react with each other, there will be an affinity between them.Lithium metal is active, so a considerable number of conductive or non-conductive element metals and metal compounds have an affinity with lithium metal.3D metal matrix framework materials, including conductive or non-conductive metals [24][25][26], metal alloys, and metal compounds [27][28][29], can induce lithium metal to deposit uniformly in a variety of ways, such as through the formation of high lithium-ion conductive layers [30], the formation of lithiophilic compounds [31], the induction of active nucleation seeds [32], the formation nanopore struc- This perspective is focusing on the latest progress of 3D lithiophilic materials in stabilizing lithium anodes, such as 3D metal-based materials, 3D carbon materials, and 3D polymers.After analyzing the mechanism of how these 3D lithiophilic materials stabilize the lithium anode, finally, we conclude that to realize the industrialization of lithium metal batteries, the dendrite and volume expansion of lithium should be solved first of all and should combine different modification methods to be studied and explored.The future research direction of 3D composite lithium anode is also prospected.

3D Metal-Based Materials to Stabilize Lithium Anode
According to the theory put forward by Wang et al. [23], if two substances could react with each other, there will be an affinity between them.Lithium metal is active, so a considerable number of conductive or non-conductive element metals and metal compounds have an affinity with lithium metal.3D metal matrix framework materials, including conductive or non-conductive metals [24][25][26], metal alloys, and metal compounds [27][28][29], can induce lithium metal to deposit uniformly in a variety of ways, such as through the formation of high lithium-ion conductive layers [30], the formation of lithiophilic compounds [31], the induction of active nucleation seeds [32], the formation nanopore structures [33], and effective surface area expansion [34].

3D Lithiophilic Metal Collector
Metal-based materials, such as copper and nickel, were first used as lithium anode collectors due to their excellent electrical conductivity and chemical stability.Compared with the two-dimensional (2D) current collector, the 3D current collector has a larger specific surface area, which can effectively reduce the current density and slow down the generation of lithium dendrites [16].The 3D porous structure can also restrict the free growth of lithium metal and effectively slow down the infinite volume expansion of the lithium anode [16].Therefore, in recent years, it has received extensive attention from researchers.However, due to the poor compressibility of metal-based materials, researchers often use template etching and electrodeposition to obtain foam-like and network-like 3D metal current collectors [35,36], and also use wet chemical synthesis to obtain 3D nanowire/nanotube structures [37].
For example, Guo's research group prepared a 3D reticulated Cu current collector with a high aspect ratio by dehydrating Cu(OH) 2 to CuO on copper foil and then reducing it as shown in Figure 2a [34].Compared with an ordinary commercial 2D Cu foil current collector, using this 3D Cu current collector, the growth of lithium anode dendrite was effectively inhibited.It can continuously operate safely for 600 h under 0.2 mA cm −2 with low voltage hysteresis as shown in Figure 2b.Subsequently, they compared the lithium wettability on different current collectors as shown in Figure 2c [23]: as can be seen, molten Li shows a poor wettability on planar Cu and the contact angle of planar Cu is as high as 140 • .While the surface wetting of molten Li onto various 3D substrates, i.e., Cu foam, Ni foam, Fe foam, etc., is improved, which suggested the 3D structure is beneficial for enhancing the current collector's lithiophilicity.Yu's group synthesized Cu nanowires by the wet chemical synthesis method [37], then dispersed them in ethanol hydrazine hydrate solvent, and obtained a self-supporting Cu nanowire membrane as the current collector through the solvent self-volatilization assembly.Using this 3D Cu nanowire current collector, up to 7.5 mAh cm −2 lithium can be electrodeposited, and the Li cathode could stably cycle 200 cycles with an average CE as high as 98.6%, much higher than using the common 2D Cu current collector.In addition, Fokko M. Mulder et al. prepared a honeycomb-like porous 3D Ni@Cu current collector by electrodeposition of a hydrogen bubble dynamic template (HBDT) [36].The collector prepared by this method has a highly stable cycle performance of lithium plating/stripping (more than 300 cycles under 0.5 mA cm −2 and more than 200 cycles under 1.0 mA cm −2 ).Such superior stability is attributed to their ability to effectively accommodate lithium deposition in porous networks, and to make charge distribution delocalized, while Luo's group reported a light 3D titanium current collector to alleviate the incompatibility between electrolyte additives and the collector [38].When the electrolyte contains polysulfide additives, the collector has good corrosion resistance.Li having been deposited at 5 mAh cm −2 , after 10 cycles, no lithium dendrite was observed, and the CE was as high as 99%.
Therefore, the stable cycle of lithium metal anode, on the condition that there is high current density and large amount deposition, can be achieved by designing a 3D porous structured metal current collector.However, the 3D metal-based current collector without lithium affinity easily causes lithium metal to deposit on the top of the 3D porous structure on the condition that there is high current and a mass of deposition, rather than inside the 3D structure [16].Therefore, how to build a lithiophilic 3D porous collector is a research hotspot now.
Lithiophilic 3D metal collectors are mainly obtained from modifying the surface and even the interior of 3D Cu or Ni metal collectors-for example, alloying or metal compound modification.Cui's group first proposed the notion of lithium affinity [39].They investigated the nucleation overpotential of lithium on a substrate of 11 elements (Cu, Al, Zn, Pt, Au, Ag, Sn, Mg, Si, Ni, and C), and found the Cu, Ni, C, Sn, and Si substrates showed obvious nucleation overpotentials.In terms of Ni and Cu, in particular, the great nucleation overpotential resulted from a large thermodynamic mismatch between the substrate and lithium.However, Zn, Mg, Ag, and Au substrates show zero nucleation overpotential after complete lithiation, which effectively eliminates the nucleation barrier.Thus, the above metals can serve as lithiophilic centers for preferential lithium nucleation to regulate uniform nucleation and even subsequent growth.
Batteries 2023, 9, 30 4 of 18 overpotential after complete lithiation, which effectively eliminates the nucleation barrier.Thus, the above metals can serve as lithiophilic centers for preferential lithium nucleation to regulate uniform nucleation and even subsequent growth.Cu foil as a current collector [34].Copyrights @ 2015 Macmillan Publishers Limited.All rights reserved.(c) Wettability of molten Li onto various substrates [23].Copyrights @ 2019 Macmillan Publishers Limited.All rights reserved.
For example, Lai's research group used electroplating technology to deposit Ag particles on copper foil [40].When the deposition time was 10 s, the size of silver particles ranged from 0.5 μm to 2.0 μm, as shown in Figure 3a.The Ag particles are uniformly covered on the copper foam collector, which can control the nucleation of Li preferentially and also reduce the nucleation overpotential.Therefore, the silver particle-modified copper foam collector (Ag@Cu foam) illustrated excellent fast charge-discharge performance and long cycle life, as shown in Figure 3b.At 2.0 mA cm −2 , the 3D Ag@Cu foam collector exhibited a high CE of 97.5% in 400 cycles, while the foam Cu collector only exhibited a finite cycle life of ~110 cycles.Huang's group reported an Au-modified Cu foil as the current collector [41].The Au-modification transformed into LixAu alloy and solid solution, successfully reducing the nuclear forming barrier of Li deposition (Figure 3c,d) and inducing the dendrite-free Li morphology (Figure 3e,f).Therefore, the Li2S||Au/Cu full cell demonstrated remarkably improved stability and initial Coulombic efficiency, and achieved a high energy density of up to 626 Wh kg −1 , while Zhang et al. reported a ZnOmodified brass alloy current collector [42].Based on the low layered fault energy of Cu-Zn alloy, they proposed a new surface atomic diffusion strategy, which utilized the negative Gibbs free energy generated by surface oxidation and succedent displacement reaction during air heat treatment to achieve uniform coverage of ZnO nanoparticles on the brass net.The lithium deposition nucleation overpotential was decreased by using the brass mesh modified with ZnO as a lithiophilic current collector, and the 3D structure can For example, Lai's research group used electroplating technology to deposit Ag particles on copper foil [40].When the deposition time was 10 s, the size of silver particles ranged from 0.5 µm to 2.0 µm, as shown in Figure 3a.The Ag particles are uniformly covered on the copper foam collector, which can control the nucleation of Li preferentially and also reduce the nucleation overpotential.Therefore, the silver particle-modified copper foam collector (Ag@Cu foam) illustrated excellent fast charge-discharge performance and long cycle life, as shown in Figure 3b.At 2.0 mA cm −2 , the 3D Ag@Cu foam collector exhibited a high CE of 97.5% in 400 cycles, while the foam Cu collector only exhibited a finite cycle life of ~110 cycles.Huang's group reported an Au-modified Cu foil as the current collector [41].The Au-modification transformed into Li x Au alloy and solid solution, successfully reducing the nuclear forming barrier of Li deposition (Figure 3c,d) and inducing the dendrite-free Li morphology (Figure 3e,f).Therefore, the Li 2 S||Au/Cu full cell demonstrated remarkably improved stability and initial Coulombic efficiency, and achieved a high energy density of up to 626 Wh kg −1 , while Zhang et al. reported a ZnO-modified brass alloy current collector [42].Based on the low layered fault energy of Cu-Zn alloy, they proposed a new surface atomic diffusion strategy, which utilized the negative Gibbs free energy generated by surface oxidation and succedent displacement reaction during air heat treatment to achieve uniform coverage of ZnO nanoparticles on the brass net.The lithium deposition nucleation overpotential was decreased by using the brass mesh modified with ZnO as a lithiophilic current collector, and the 3D structure can reduce the surface current density, make the lithium-ion distribution uniform, and effectively inhibit the lithium dendrite growth.Using this ZnO-modified lithiophilic 3D collector, the symmetric cell achieved long-term cycle stability at 2.0 mA cm −2 for 500 cycles (Figure 3g).reduce the surface current density, make the lithium-ion distribution uniform, and effectively inhibit the lithium dendrite growth.Using this ZnO-modified lithiophilic 3D collector, the symmetric cell achieved long-term cycle stability at 2.0 mA cm −2 for 500 cycles (Figure 3g).

3D Lithium Alloy and 3D Metal/Lithium Complex Anode
The unsupported lithium metal anode is often accompanied by unstable SEI film, uncontrollable dendrite growth, low Coulombic efficiency, short cycle life, and nearly infinite volume expansion in practical application [11,22].Therefore, researchers developed various kinds of strategies to modify the lithium anode in order to improve its performance, and one of the breakthrough studies is the construction of a lithium alloy anode [43][44][45][46][47][48][49][50][51].Through alloying, the uniform deposition of lithium metal is induced, which can greatly alleviate and improve the problems of lithium metal anode, thus improving the electrochemical performance and practicability of the battery.

3D Lithium Alloy and 3D Metal/Lithium Complex Anode
The unsupported lithium metal anode is often accompanied by unstable SEI film, uncontrollable dendrite growth, low Coulombic efficiency, short cycle life, and nearly infinite volume expansion in practical application [11,22].Therefore, researchers developed various kinds of strategies to modify the lithium anode in order to improve its performance, and one of the breakthrough studies is the construction of a lithium alloy anode [43][44][45][46][47][48][49][50][51].Through alloying, the uniform deposition of lithium metal is induced, which can greatly alleviate and improve the problems of lithium metal anode, thus improving the electrochemical performance and practicability of the battery.
Professor Mai and his collaborators developed a lithiophilic 3D Li-B-Mg alloy anode [45].According to the calculation of density functional theory (DFT), adding an Mg element to the three-dimensional LiB skeleton can fundamentally improve the adsorption energy of lithium as shown in Figure 4a.At the same time, Li-B-Mg composite skeleton can also provide less interfacial resistance, maintain structural stability in the long-time cycle process, and limit the infinite volume expansion of Li.Therefore, the Mg-doped Li-B skeleton has excellent lithiophilicity and conductivity, which helped to reduce local current density and homogenize lithium ion distribution, thus inhibiting dendrite generation as shown in Figure 4b.The soft-clad all-cell assembled with LiCoO 2 positive electrode and Li-B-Mg alloy negative electrode has good electrochemical performance (Figure 4c), which proved that it can be potentially applied to large-scale commercial production.While pure lithium metal is preferable for setups that permit stack pressures in the MPa range, magnesium alloying can boost lithium consumption when no external pressure is applied [52].Contact loss can also be effectively avoided by adding a suitable proportion of Mg, i.e., 10%, to the Li metal anode [52].This shows the wide solid solution range, the robustness of the magnesium framework, and the comparatively high Li diffusion coefficient in magnesium (∼10 −7 cm 2 •s −1 for the vapor-deposited Li-Mg alloy) [52][53][54][55][56].
Since then, more and more lithiophilic 3D lithium alloy anodes, such as Li-Bi [48], Li-Ga [43], Li-Sn [46], Li-Ag [50], Li-In [46], Li-Mg [51], etc., have been reported, and they verify the advantages of the lithium-philic alloying 3D anode.For the negative electrode of lithium metal alloy, we need to note that only the appropriate proportion of metal added can play a positive role.
Professor Mai and his collaborators developed a lithiophilic 3D Li-B-Mg alloy anode [45].According to the calculation of density functional theory (DFT), adding an Mg element to the three-dimensional LiB skeleton can fundamentally improve the adsorption energy of lithium as shown in Figure 4a.At the same time, Li-B-Mg composite skeleton can also provide less interfacial resistance, maintain structural stability in the long-time cycle process, and limit the infinite volume expansion of Li.Therefore, the Mg-doped Li-B skeleton has excellent lithiophilicity and conductivity, which helped to reduce local current density and homogenize lithium ion distribution, thus inhibiting dendrite generation as shown in Figure 4b.The soft-clad all-cell assembled with LiCoO2 positive electrode and Li-B-Mg alloy negative electrode has good electrochemical performance (Figure 4c), which proved that it can be potentially applied to large-scale commercial production.While pure lithium metal is preferable for setups that permit stack pressures in the MPa range, magnesium alloying can boost lithium consumption when no external pressure is applied [52].Contact loss can also be effectively avoided by adding a suitable proportion of Mg, i.e., 10%, to the Li metal anode [52].This shows the wide solid solution range, the robustness of the magnesium framework, and the comparatively high Li diffusion coefficient in magnesium (∼10 −7 cm 2 •s −1 for the vapor-deposited Li-Mg alloy) [52][53][54][55][56].
Since then, more and more lithiophilic 3D lithium alloy anodes, such as Li-Bi [48], Li-Ga [43], Li-Sn [46], Li-Ag [50], Li-In [46], Li-Mg [51], etc., have been reported, and they verify the advantages of the lithium-philic alloying 3D anode.For the negative electrode of lithium metal alloy, we need to note that only the appropriate proportion of metal added can play a positive role.In addition to the alloying of lithium metal anode, another common lithium anode protection strategy is to build a lithiophilic 3D complex lithium anode [42,[57][58][59][60][61][62][63][64][65][66][67][68][69][70].On the one hand, the unique 3D structure and good electrical conductivity can build a complete and uniform conductive path in the internal lithium metal anode, reducing the local current density, and thus, delaying or inhibiting dendrite.On the other hand, by introducing a lithiophilic substrate to support the lithium anode, lithium-ion flux is homogenized and uniform lithium deposition is obtained.In addition to the alloying of lithium metal anode, another common lithium anode protection strategy is to build a lithiophilic 3D complex lithium anode [42,[57][58][59][60][61][62][63][64][65][66][67][68][69][70].On the one hand, the unique 3D structure and good electrical conductivity can build a complete and uniform conductive path in the internal lithium metal anode, reducing the local current density, and thus, delaying or inhibiting dendrite.On the other hand, by introducing a lithiophilic substrate to support the lithium anode, lithium-ion flux is homogenized and uniform lithium deposition is obtained.
For instance, Manthiram et al. developed a 3D hybrid structure of Mo 2 N-modified carbon fiber (CNF) as a lithium anode host material by a simple in situ reduction-nitride strategy [31].The Mo 2 N@CNF hybrid structure has several advantages as a lithium anode host.Firstly, Mo 2 N with strong lithophilia and uniform distribution can be used as a pre-implanted nucleation seed to guide lithium nucleation and deposition homogenously along the 3D skeleton.Secondly, the CNF skeleton with high conductivity and high surface area can be used as a 3D fluid collector to promote rapid electron transfer and uniform electric field distribution, effectively avoiding charge accumulation and strengthening the inhibition of lithium dendrite growth.Thirdly, the framework with abundant pores and excellent mechanical strength supplies abundant lithium housing sites, which greatly alleviates the problem of the infinite volume expansion of lithium and ensures electrode integrity during prolonged cycling.Therefore, using this Mo 2 N-modified carbon fiber as a host material, dendrite-free lithium stripping/plating and ultra-low voltage overpotential can be achieved, and the long life and Coulomb efficiency of symmetric cells can be greatly improved as shown in Figure 5a,b.Professor Yang and Professor Kang jointly reported a light 3D Cu nanowire fluid collector with gradient phosphatization as shown in Figure 5c [32].The gradient phosphide with good lipophilicity and high ionic conductivity enabled lithium to compact nucleate and stably deposit in the porous structure as shown in Figure 5d.Li deposition with a loading capacity of about 44 wt% and a capacity of 3 mAh cm −2 was achieved, and the average CE is up to 97.3%.At 2 mA cm −2 , the symmetric cell has a lifetime of 300 h as shown in Figure 5e.
Batteries 2023, 9, 30 7 of 18 For instance, Manthiram et al. developed a 3D hybrid structure of Mo2N-modified carbon fiber (CNF) as a lithium anode host material by a simple in situ reduction-nitride strategy [31].The Mo2N@CNF hybrid structure has several advantages as a lithium anode host.Firstly, Mo2N with strong lithophilia and uniform distribution can be used as a preimplanted nucleation seed to guide lithium nucleation and deposition homogenously along the 3D skeleton.Secondly, the CNF skeleton with high conductivity and high surface area can be used as a 3D fluid collector to promote rapid electron transfer and uniform electric field distribution, effectively avoiding charge accumulation and strengthening the inhibition of lithium dendrite growth.Thirdly, the framework with abundant pores and excellent mechanical strength supplies abundant lithium housing sites, which greatly alleviates the problem of the infinite volume expansion of lithium and ensures electrode integrity during prolonged cycling.Therefore, using this Mo2N-modified carbon fiber as a host material, dendrite-free lithium stripping/plating and ultra-low voltage overpotential can be achieved, and the long life and Coulomb efficiency of symmetric cells can be greatly improved as shown in Figure 5a,b.Professor Yang and Professor Kang jointly reported a light 3D Cu nanowire fluid collector with gradient phosphatization as shown in Figure 5c [32].The gradient phosphide with good lipophilicity and high ionic conductivity enabled lithium to compact nucleate and stably deposit in the porous structure as shown in Figure 5d.Li deposition with a loading capacity of about 44 wt% and a capacity of 3 mAh cm −2 was achieved, and the average CE is up to 97.3%.At 2 mA cm −2 , the symmetric cell has a lifetime of 300 h as shown in Figure 5e.Recently, researchers, based on lithiophilic materials, have developed a variety of lithiophilic 3D metal matrix skeletons to serve as the host for lithium metal anode.The design of a 3D metal skeleton can not only play a supporting role, but also induce uniform deposition of lithium metal in different ways, so as to greatly alleviate the infinite volume expansion of lithium anode and severe dendrite growth.

3D Lithiophilic Carbon Materials
3D carbon-based materials are employed to stabilize lithium metal anode, mainly based on graphene, carbon nanotubes/fibers, hollow carbon spheres, and other carbon nanocomposites with lightweight, porous structures and stable electrochemical properties [71][72][73][74][75][76][77].These advantages ensure that they will not seriously affect the high specific capacity of lithium metal and its conductivity.More importantly, the pore structure and lithiophilicity of nano-carbon skeleton materials can be regulated according to the actual requirements.
For example, Zhang's group developed an unstacked graphene framework via CH 4 chemical vapor deposition on magnesium aluminum hydroxide and then removed the template of magnesium aluminum hydroxide [78].The prepared 3D unstacked graphene was then coated on the copper foil for Li deposition.As a result, the CE could be improved to 93% at 0.5 mA cm −2 and a cycle capacity of 0.5 mAh cm −2 , while the CE of ordinary copper foil was only 65% to 85% (Figure 6a).In addition, under the subsequent 2 mA cm −2 lithium stripping/plating test experiments, the 3D graphene-based lithium anode has a stable voltage distribution within 800 cycles (Figure 6b).In the same year, a layered reduced graphene oxide with nanoscale interlayer space was also employed as a 3D host for lithium anodes (Figure 6c) [79].The 3D reduced graphene oxide-lithium composite anode owned a specific capacity of 3390 mAh g −1 with low volume expansion (20%) and good mechanical flexibility.The composite 3D electrode showed a relatively low overpotential (80 mV) at 3 mA cm −2 , and when matched with the cathode of LiCoO 2 , the full cell also showed a good rate performance.Hu's group prepared a kind of carbonized wood with a vertical pore structure of 73% porosity and used it as a 3D conductive skeleton as shown in Figure 6d-f [80].The vertical channel structure limits the growth direction of lithium metal, increases the space utilization of lithium metal, and avoids huge volume changes.When the lithium anode encapsulating this skeleton material cycled at the current of 3 mA cm −2 , the cycle time could reach 150 h and the polarization voltage is only 90 mV.
In order to add the lithiophilicity, functional group modification or heteroatom doping of various kinds of nanocarbon materials is a general, simple, and facile strategy.For instance [73,76,[81][82][83][84][85][86]], Liu's group reported mesoporous carbon nanofibers that were modified with amino groups for a self-smoothing Li-C anode as shown in Figure 7a [87].Not only does such a lithium-carbon anode own the conductive skeleton and porous structure to limit the dendrite growth, but also, the surface chemistry of the carbon host is significantly altered by amine functionalization in terms of its interaction with lithium (Figure 7b,c).Computer simulations showed that Li tended to nucleate and grow into twodimensional Li clusters around the -NH sites on the carbon surface due to more favorable energies (Figure 7d,e), which guides the deposition of Li along the carbon fiber surface, as shown in Figure 7f.As a result, when this composite anode matched with the NMC622 (LiNi 0.6 Mn 0.2 Co 0.2 O 2 ) cathode, after accounting for all the active and inactive components, the completed full cell may achieve an energy density of 350-380 Wh kg −1 and a steady cycling capacity of up to 200 cycles.Zhang's group annealed the graphene oxide in an NH 3 atmosphere to obtain the N-doped graphene and then employed it as the Li plating matrix [83].They found that the N-doped graphene with doping pyrrolic and pyridinic N are lithiophilic, which can guide the Li nucleation and introduce the uniform Li + distribution on the lithium anode surface.Accordingly, the N-doped graphene-modified Li anode demonstrated a dendrite-free morphology as well as an excellent electrochemical performance, i.e., stably cycling 200 cycles at 1 mA cm −2 with a high CE of 98%.Xie et al. reported a fluorinated graphene (FG)-modified Li negative electrode (LFG) for high-performance lithium−oxygen (Li−O 2 ) cells [88].As shown in the in situ optical microscope in Figure 7g, it was evident that, during lithium deposition, a large number of Li dendrites develop on the exposed Li surface, which will cause the SEI layer to break, the Li electrode to deteriorate, and even the cell to short circuit, while only slight surface variation could be seen on the LFG electrode, indicating a consistent Li deposition and restricted Li dendrite growth.The in situ observation showed that merely a 3 wt% FG introduction increases the rate capability and cycling life of Li electrodes significantly.The cells with LFG exhibit significantly more stable voltage profiles than the half cells with bare Li, even at high areal capacities of up to 5 mA h cm −2 or high current densities of up to 5 mA cm −2 .Additionally, compared to cells with pure lithium anodes, Li-O 2 cells with LFG anodes exhibit a longer cycle life.In order to add the lithiophilicity, functional group modification or heteroatom doping of various kinds of nanocarbon materials is a general, simple, and facile strategy.For instance [73,76,[81][82][83][84][85][86]], Liu's group reported mesoporous carbon nanofibers that were modified with amino groups for a self-smoothing Li-C anode as shown in Figure 7a [87].Not only does such a lithium-carbon anode own the conductive skeleton and porous growth.The in situ observation showed that merely a 3 wt% FG introduction increases the rate capability and cycling life of Li electrodes significantly.The cells with LFG exhibit significantly more stable voltage profiles than the half cells with bare Li, even at high areal capacities of up to 5 mA h cm −2 or high current densities of up to 5 mA cm −2 .Additionally, compared to cells with pure lithium anodes, Li-O2 cells with LFG anodes exhibit a longer cycle life.[87].Copyrights © 2022, Springer Nature.(g) After 0, 15, and 30 min, in situ optical images of the plating processes of bare Li and LFG at a current density of 1 mA cm −2 were taken [88].Copyrights © 2022 American Chemical Society.

3D Organic Polymer Materials
As mentioned above, the deposition of lithium metal on the surface of conductive material as the skeleton may still lead to the generation of lithium dendrites.To solve this problem, through the design of polymer surface groups, researchers can make the polymer skeleton have a variety of functions, such as regulating the concentration of lithium ions on the electrode surface [89][90][91][92], facilitating uniform nucleation [92][93][94], participating in the formation reaction of SEI [95][96][97][98], and promoting the migration of lithium ions [90,91,99].
For example, Lu's group designed a 3D porous poly-melamine-formaldehyde (PMF) polymer framework (Figure 8a) containing a variety of polar functional groups (such as amine and triazine) on the surface [89], which effectively homogenizes the flux of lithium ions, and the large pores inside the framework can accommodate a large amount of lithium metal, alleviating the volume expansion of lithium metal.As a result, the 3D Li/PMF composite electrode shows enhanced electrochemical performances with smooth voltage plateaus and low hysteresis, and the CE could still reach as high as 94.7% after 50 cycles at an ultrahigh current density of 10 mA cm −2 (Figure 8b).A polyethyleneimine porous sponge (PSS) polymer framework is designed by Wang's group [99], which, in order to facilitate mass transfer in lithium metal batteries, can cause a range of electrokinetic phenomena, such as electrokinetic surface conduction, electroosmosis, and electrophoresis (Figure 8c).PPS has a strong affinity for lithium ions, which could concentrate lithium ions in the pore structure of the skeleton, resulting in a local concentration of lithium ions higher than the bulk solution as shown in Figure 8c.The lithium-ion automatic aggregation and electric pump feature of this 3D PPS can simultaneously overcome the diffusion limit and decrease the concentration polarization, so as to adjust the lithium-ion concentration difference to solve the problem of uncontrollable lithium deposition (such as lithium dendrite growth).Recently, Zhang's group employed the interaction between the 3D polymer skeleton and solvent molecules to promote the formation of stable SEI in the composite lithium anode and achieve the synergistic effect of electrolyte and 3D skeleton (Figure 8d) [95].The electrospun polyacrylonitrile (PAN) with strong polarity was used as the framework material, and fluoroethylene carbonate (FEC) and dimethyl carbonate (DMC) were used as the solvent of electrolyte.It is proved by theoretical calculation that the interaction between PAN and FEC molecules is stronger than that between PAN and DMC molecules.Therefore, FEC aggregates on the surface of lithium metal under the induction of the skeleton to form a LiF-rich SEI layer, thus enhancing the uniformity of Li deposition.Therefore, a composite Li anode with a PAN polar host could deliver 145 cycles, while the Li anode with a no-polar host only delivers 90 cycles.
Previous research has proven that the 3D polymer framework played an important role in slowing down the volume expansion and inducing lithium deposition during lithium metal cycling.However, the manufacturing cost, mechanical properties, density, processability, and electrochemical stability of polymers should be considered comprehensively when designing 3D polymer skeleton materials.Previous research has proven that the 3D polymer framework played an important role in slowing down the volume expansion and inducing lithium deposition during lithium metal cycling.However, the manufacturing cost, mechanical properties, density, processability, and electrochemical stability of polymers should be considered comprehensively when designing 3D polymer skeleton materials.

Conclusions
Although based on the lithiophilic 3D conductive skeleton construction, the cycle life, rate capability, and Coulombic efficiency of lithium metal anode have been improved, and even though some of the performance is better than the existing commercial anode materials, it should be pointed out that the commercialization progress of lithium battery

Conclusions
Although based on the lithiophilic 3D conductive skeleton construction, the cycle life, rate capability, and Coulombic efficiency of lithium metal anode have been improved, and even though some of the performance is better than the existing commercial anode materials, it should be pointed out that the commercialization progress of lithium battery based on 3D metal lithium anode is slow.That is because: First, many fundamental issues are still confusing-for example, the alloys, doping atoms, and polar functional groups all exhibit the stated lithiophilicity.Are the interactions between essences that cause lithiophilicity the same?Is the lithiophilicity specific to the Li atom or Li ion?The definition and boundary of lithiophilicity need further investigation from our researchers.
Second, uncontrolled growth of dendrites and excessive side reactions need to be further solved.Even though utilizing the 3D structure and lithiophilic property could realize the precise regulation of lithium deposition, the precise regulation of lithium deposition depends on the precise control of the 3D skeleton microstructure, which makes the preparation process too complicated and difficult to ensure the consistency of materials in the process of large-scale production, and will greatly increase the production cost [100,101].
Third, some problems have not been paid attention to in the laboratory battery test, i.e., the thickness of the electrode and the expansion of the electrode [102]: commercial softpack batteries have requirements on the thickness and expansion of the electrode, wherein the volume expansion of the lithium-based composite anode should be less than 15% and the thickness of the 3D electrode should be controlled below 50 µm (commonly, the 3D skeleton is almost hundreds of µm) after being filled with lithium [100,101,103], before it can be applied to the commercial soft-pack battery.How to achieve a thinner 3D conductive skeleton with a low expansion rate (<15%) is a great challenge from the material design.In the face of the above challenges, first of all, the 3D skeleton material should take into account the rigidity and toughness, reduce the micropores and increase the densely packed pores, induce the dense deposition of lithium metal, and accommodate the volume change during the deposition and stripping of lithium metal [17,100,104].Secondly, it is still necessary to accurately control the deposition site of lithium metal to avoid deposition between the electrode and diaphragm interface, and improve cycle stability [7,21,105].Finally, the synthesis method should easily realize the structure and thickness control of 3D skeleton electrode materials, and easily scale production, improving the consistency of materials [16,17].
Additionally, at present, there are vast publications that reported that 3D lithiophilic materials are beneficial for enhancing the performances of a lithium anode, but there is little systematic research on the effects of doping atoms and functional groups, electrical conductivity, and pore structure (in the case of host materials) on the deposition behavior of lithiophilic materials [106].Consequently, in order to provide a more thorough understanding of Li's actions in lithiophilic materials, we must consider advanced in situ characterizations, i.e., in situ TEM [107], SEM [108], AFM [109], etc.The creation of electrodes that can support active lithiophilic sites during the full plating/stripping processes is anticipated to be guided by theoretical models developed using finite element analysis, density functional theory, and phase-field theory, as well as those models.
All in all, there are many problems and challenges in the lithium metal composite anode, and it is difficult to effectively solve all the problems by relying on a single modification method, i.e., 3D construction.In contrast, the synergistic effect of various strategies can better solve the problems faced by the lithium anode [110].For example, matching a 3D lithium metal composite anode with a modified electrolyte is a more effective technical path to improving comprehensive performance.Electrolyte modification can effectively construct stable solid electrolyte interface film, which has obvious advantages in reducing side reactions and improving the CE of electrodes.However, we also need to point out that the electrolyte amount has to decrease for practically high-energy-density batteries from >30 to <3 g Ah −1 for Li batteries coupled with nickel cobalt manganese oxide/lithium cobalt oxide (NCM/LCO) cathode, and from electrolyte/sulfur ratio >10 to <3 for Li-S batteries.
At the same time, in the process of the commercial application of lithium metal anode, safety cannot be ignored, which is also a factor to be considered in the design of 3D electrodes with electrolyte modification.The use of a 3D alloying anode with a modified electrolyte could more effectively reduce the anode activity and improve the safety of the battery.

Figure 1 .
Figure 1.Various strategies to inhibit lithium dendrite growth.

Figure 1 .
Figure 1.Various strategies to inhibit lithium dendrite growth.

Figure 2 .
Figure 2. (a) Schematic presentation of the procedures to prepare a 3D porous Cu foil from a planar Cu foil.(b) Li metal plating/stripping at 0.2 mA cm −2 in symmetric Li|Li@Cu cells with planar or 3D

Figure 2 .
Figure 2. (a) Schematic presentation of the procedures to prepare a 3D porous Cu foil from a planar Cu foil.(b) Li metal plating/stripping at 0.2 mA cm −2 in symmetric Li|Li@Cu cells with planar or 3D Cu foil as a current collector [34].Copyrights @ 2015 Macmillan Publishers Limited.All rights reserved.(c) Wettability of molten Li onto various substrates [23].Copyrights @ 2019 Macmillan Publishers Limited.All rights reserved.

Figure 3 .
Figure 3. (a) SEM images of Ag@Cu foam with plated silver for 10.0 s; (b) Coulombic efficiencies of Ag@Cu and Cu foam at 2.0 mA cm −2 [40].Copyrights © 2022 Published by Elsevier.Discharge voltage curves of (c) Li||Cu, and (d) Li||Au/Cu batteries at 10 μA cm −2 , 0.1 mAh cm −2 ; the SEM images of Cu foil surface in the Li2S||Cu cell, and (e) Cu/Au foil surface in the Li2S||Au/Cu cell (f) for the first cycle at 0.05 C [41].Copyrights © 2022 Published by Elsevier.(g) Voltage profiles of the ZnO−CuZn mesh and Cu foil in symmetrical batteries at 2 mA cm −2 and 1 mA h cm −2 [42].Copyrights © 2022 American Chemical Society.

Figure 3 .
Figure 3. (a) SEM images of Ag@Cu foam with plated silver for 10.0 s; (b) Coulombic efficiencies of Ag@Cu and Cu foam at 2.0 mA cm −2 [40].Copyrights © 2022 Published by Elsevier.Discharge voltage curves of (c) Li||Cu, and (d) Li||Au/Cu batteries at 10 µA cm −2 , 0.1 mAh cm −2 ; the SEM images of Cu foil surface in the Li 2 S||Cu cell, and (e) Cu/Au foil surface in the Li 2 S||Au/Cu cell (f) for the first cycle at 0.05 C [41].Copyrights © 2022 Published by Elsevier.(g) Voltage profiles of the ZnO−CuZn mesh and Cu foil in symmetrical batteries at 2 mA cm −2 and 1 mA h cm −2 [42].Copyrights © 2022 American Chemical Society.

Figure 4 .
Figure 4. (a) Adsorption energies of one lithium atom on the surface of LiB, Mg, and LiB/Mg, respectively; (b) The stripping/plating process of lithium on LiB skeleton; (c) Pouch-type batteries using Li-B-Mg anode and LiCoO2 cathode in the carbonate-based electrolyte [45].Copyrights © 2022, WILEY-VCH.

Figure 4 .
Figure 4. (a) Adsorption energies of one lithium atom on the surface of LiB, Mg, and LiB/Mg, respectively; (b) The stripping/plating process of lithium on LiB skeleton; (c) Pouch-type batteries using Li-B-Mg anode and LiCoO 2 cathode in the carbonate-based electrolyte [45].Copyrights © 2022, WILEY-VCH.

Figure 6 .
Figure 6.(a) Average CE and its variance for Cu foil and graphene-based anode with various cycling capacities at 0.5 mA cm −2 ; (b) Voltage profiles for graphene-and Cu foil-based anode in 800 cycles at 2.0 mA cm −2 , fixing capacity at 0.1 mA h cm −2 [78].Copyrights © 2022, WILEY-VCH.(c) Synthesis process from GO film (left) to a sparked rGO film (middle) to a layered Li-rGO composite film (right) with the corresponding digital photos [79].Copyrights © 2022, Macmillan Publishers Limited.(d) The synthesis process of Li/C-wood composite; (e) Top-view SEM images of ZnO-coated C-wood; (f) Cross-sectional SEM images of Li infusion in the channel void space [80].Copyrights © 2022, Macmillan Publishers Limited.

Figure 6 .
Figure 6.(a) Average CE and its variance for Cu foil and graphene-based anode with various cycling capacities at 0.5 mA cm −2 ; (b) Voltage profiles for graphene-and Cu foil-based anode in 800 cycles at 2.0 mA cm −2 , fixing capacity at 0.1 mA h cm −2 [78].Copyrights © 2022, WILEY-VCH.(c) Synthesis process from GO film (left) to a sparked rGO film (middle) to a layered Li-rGO composite film (right) with the corresponding digital photos [79].Copyrights © 2022, Macmillan Publishers Limited.(d) The synthesis process of Li/C-wood composite; (e) Top-view SEM images of ZnO-coated C-wood; (f) Cross-sectional SEM images of Li infusion in the channel void space [80].Copyrights © 2022, Macmillan Publishers Limited.

Figure 7 .
Figure 7. (a) -NH functional groups modified the surface of mesoporous CNFs; (b) Spontaneous infiltration of Li into CNFs; (c) Schematic of rough Li-C surface after electrochemical deposition and stripping process: grey for C; blue for Li; orange for the electrolyte.DFT calculation of lithium growth on (d) ammonia-modified, and (e) unmodified carbon surfaces: grey for C; blue for N; white for H; purple for Li; (f) The SEM image of Li deposition on an untreated carbon film[87].Copyrights © 2022, Springer Nature.(g) After 0, 15, and 30 min, in situ optical images of the plating processes of bare Li and LFG at a current density of 1 mA cm −2 were taken[88].Copyrights © 2022 American Chemical Society.
and Key Disciplines of Chemical Engineering and Technology in Chongqing Colleges and Universities during the 13th Five Year Plan provided the financial support.Institutional Review Board Statement: Not applicable.