WO2020018790A1 - Metal coated structures for use as electrodes for batteries and methods of production thereof - Google Patents

Abstract

The invention discloses new methods of coating a structure in a conductive layer. The methods include coating a structure with a metal coordinating polymer, e.g., polydopamine, layer then coating the layer with a layer of metal, e.g., copper. The method can be used on readily available and low- cost porous materials, such as glass fibers and blended fabrics, using low-cost reagents under mild reaction conditions. The invention further provides batteries using structures of the invention as the current collector. Lithium batteries using the structures of the invention as current collectors exhibit no lithium dendrite growth and excellent battery performance over long cycle periods.

Description

METAL COATED STRUCTURES FOR USE AS ELECTRODES FOR BATTERIES AND METHODS OF PRODUCTION THEREOF

STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH

This invention was made with government support under 1708729 and 1420570 awarded by the National Science Foundation. The government has certain rights in the invention.

BACKGROUND OF THE INVENTION

Developing advanced electrode materials is critical for high-energy-density rechargeable batteries. Lithium metal with its extremely high specific capacity (3,860 mA h g ¹ ) and the lowest redox potential (-3.04 V versus the standard hydrogen electrode) has been long regarded as an ideal anode material. However, the dendrite problem has largely limited its commercialization. hi Uncontrolled dendrite growth can penetrate the separator, causing short circuits and even catastrophic explosions.⁴ Lithium dendrites were thought to originate from the inhomogeneity of the Li⁺ concentration on the traditional anode surfaces and the drastic electrode dimensional change during cycling. Pi

Various strategies have been proposed to suppress the dendrite growth. Many approaches focus on improving the stability and uniformity of the solid electrolyte interphase (SEI) layer by optimizing electrolyte components including adding Cs⁺ and Rb⁺,¹³¹ LiF,¹⁴¹ vinylene carbonate (VC)^[51 or L Ss.¹⁶¹ However, due to the“hostless” nature of Li metal with volume expansion toward the separator,¹⁷¹ cracking SEI layers upon cycling can expose the fresh Li metal upon further reaction. Solid or polymer electrolytes are promising to suppress the lithium dendrite growth if the issues of lithium penetration through the cracks or grain boundaries of the solid electrolyte can be solved, together with some other issues such as the lower ionic conductivity or interface reaction of solid and polymer electrolytes.¹⁸¹ As an alternative, ex situ coated artificial SEI layers such as PDMS,¹⁹¹ boron/graphene,¹¹⁰¹ or silica layer^{11 11} present efficient interfacial protection.

Recently, various novel“host” designs have demonstrated their power in changing the Li plating behavior. Conductive micro/nanostructure frameworks have proved to be an effective method to ensure uniform Li deposition and to accommodate the Li volumetric expansion;¹⁸^ these frameworks include the layered reduced graphene oxides, ^{1¾ 121} nickel (Ni) foam¹¹³¹ and metal-coated sponges.¹¹⁴¹ However, the expensive and complex fabrication process may limit practical applications. Developing a simple, low-cost and versatile technique that can transform materials into 3D conductive frameworks remains a challenge in this field.

Accordingly, a simple, low-cost and versatile technique that can transform materials into 3D conductive frameworks for use in batteries would be helpful.

SUMMARY OF THE INVENTION

The invention features structures useful for electrodes or current collectors in a battery where the structure includes the formation of a layer of a metal coordinating polymer, e.g., polydopamine, on the surface of the structure and a layer of metal, e.g., copper, on the metal coordinating polymer, e.g., polydopamine, layer. Structures of the present invention may be incorporated into a battery as a current collector at an electrode of the battery, e.g., the anode.

In one aspect, the invention provides a method of coating a surface of a structure, the method including the steps of: a) forming a metal coordinating polymer, e.g., polydopamine layer, on the surface of the structure; and b) contacting the product of step a) with metal, e.g., copper, ions under conditions to form a layer of metal, e.g., copper, on the surface.

In some embodiments, the structure includes metal, glass, polymer, paper, fabric, carbon, polyester, cotton or a composite thereof. In some embodiments, the structure includes natural or synthetic fibers. In particular embodiments, the structure includes a composite of polyester and cotton.

In some embodiments, the structure is porous. In certain embodiments, the pores in the structure are randomly oriented. In certain embodiments, the pores in the structure are ordered.

In some embodiments, any of step a) or step b) of the method occurs at room temperature. In some embodiments, any of step a) or step b) of the method occurs over a period of 1 -5 days, e.g., 1 day, 2, days, 3 days, 4 days, or 5 days. In particular embodiments, the source of metal ions, e.g., copper ions, is a metal salt, e.g., copper salt.

In another aspect, the invention includes a battery having a) an anode including a current collector including a structure having a metal, e.g., copper, layer on a metal coordinating polymer, e.g., polydopamine, layer, where the current collector is in contact with lithium, sodium, or potassium ions; and b) a cathode comprising a redox active species.

In some embodiments, the structure includes metal, glass, polymer, paper, fabric, carbon, polyester, cotton or a composite thereof. In some embodiments, the structure includes natural or synthetic fibers. In particular embodiments, the structure includes a composite of polyester and cotton.

In some embodiments, the structure is porous. In certain embodiments, the pores in the structure are randomly oriented. In certain embodiments, the pores in the structure are ordered.

In further embodiments, the battery includes a separator between the anode and the cathode. In particular embodiments, the current collector is in contact with lithium ions. In other embodiments, the battery is pressurized, as described herein. For example, the structure may include a fabric when the battery is pressurized. BRIEF DESCRIPTION OF THE DRAWINGS

Figures 1 A-1 E: Fig. 1 A shows digital-camera photographs of 3D current collectors after different treatments. Fig. 1 B is an SEM image of the pristine 3D current collector. Fig. 1 C is an SEM image of a current collector after 500 °C annealing under N2 atmosphere for 2 hours. Fig 1 D is an SEM image of a current collector after soaking in DEC solution. Fig. 1 E is an SEM image of a current collector after cycling for 20 cycles.

Figures 2A-2D: Schematic of converting a porous structure into a conductive 3D current collector.

Figs 2A and 2B show immersion of the glass fiber (GF) in the dopamine solution with the color change from light (Fig. 2A) to dark (Fig. 2B). Fig. 2C shows the simple dip-coating of the polydopamine (PDA)-coated GF of Fig. 2B turned into a Cu coated 3D conductive framework. The enlarged image of Fig. 2C shown in Fig. 2D illustrates the adherence of PDA and Cu films on the original structure.

Figures 3A-3E: Characterization of dip-coating-treated GF frameworks. Fig. 3A shows digital-camera images of original samples (upper), PDA coated samples (middle) and Cu coated samples (bottom). Six different substrates from left to right are metal coin, polyester, polycarbonate, rice paper, glass fiber and nickel foam. Fig. 3B shows XPS spectra of GF, PDA coated GF and Cu coated GF. Figs. 3C-3E show pore size distributions of pristine GF (Fig 3C), PDA coated GF (Fig. 3D), and Cu coated GF (Fig. 3E) in SEM images.

Figure 4: Pore-size distributions of polycarbonate and its corresponding PDA coated and Cu coated structures.

Figure 5: High magnification SEM image of copper coated glass fiber, showing that copper is well coated on the fiber.

Figure 6: XPS characterization of polydopamine-coated and copper coated surfaces in the polycarbonate frameworks.

Figures 7A-7B: 2D simulations performed on COMSOL Multiphysics using the“Tertiary Current Distribution, Nernst-Planck (tcdee)” module. Fig. 7A shows the geometry of the simulation cell with the length expressed in meters. The nuclei of Li (at the bottom of Fig. 7A) have a width of 1 .5pm and height of 2.75pm for the center one and 1 .75pm for the other four. Circles above the nuclei represent the 2D sections of the 3D copper wire network. For simplicity we assume the section is perpendicular to the wire direction. Fig. 7B shown the cell configuration of the battery, with the top line standing for the cathode in a real battery, and the remaining surfaces in black are the anode. We consider the 3D copper wires are connected to the actual current collector on the anode. According to the SEM figures, the deposition of Li to the 3D copper network has a limited thickness. Thus, for simplicity, in the simulation here the deposition is set to happen on the bottom current collector and the Li metal nuclei, but not on the 3D copper wires. Figures 8A-8F: Significantly reduced Li⁺ current density“hot spots” with 3D copper structure. Figures 8A-8C show predicted current density distributions without 3D copper structure after t = 0 (Fig. 8A),

100 (Fig. 8B), and 200 (Fig. 8C) seconds of simulation. Figs. 8D-8F show predicted current density with 3D copper structure after t = 0 (Fig. 8D), 100 (Fig. 8E), and 200 (Fig. 8F) seconds of simulation

Figures 9A-9F: SEM images of the morphology of Li deposited on 2D and 3D current collectors with current density of 0.5 mA crrr² for a total of 1 mAh crrr² of Li. Fig. 9A shows a top view SEM image of the 20th Li plating, and Fig. 9C shows the 100th Li plating on pristine 2D planar current collector. Fig 9B shows a top view SEM image of the 3D porous current collector at the 20th Li plating and Fig. 9D shows a top view SEM image of the 3D porous current collector at the 100th Li plating. Fig. 9E shows a schematic of lithium plating on 2D copper. Fig. 9F shows a schematic of lithium plating on a 3D framework.

Figures 10A-10B: SEM images of the morphology after the deposited Li stripping out of the composites.

Figures 1 1 A-1 1 C: Comparison of Coulombic efficiency (CE) of Li deposition on 2D planar and 3D porous current collectors after Li deposition/stripping at various current rates of 0.5 (Fig. 1 1 A), 1 .0 (Fig. 1 1 A), and 2.0 (Fig. 1 1 A) mA crrr² with the same area capacities of 1 mA h cm^·2.

Figures 12A-12H: SEM images of the morphology of Li deposited on 2D and 3D current collectors with current density of 1 and 2 mA crrr² for a total of 1 mAh crrr² of Li after 20 cycles. Figs. 12A and 12C show top view SEM images of the 20th Li plating on 2D current collectors at 1 mA crrr². Figs.

12B and 12D show top view SEM images of the 20th Li plating on 3D frameworks at 1 mA crrr². Figures 12E and 12G are top view SEM images of the 20th Li plating on 2D current collectors at 2 mA crrr². Figures 12F and 12H are top view SEM images of the 20th Li plating on 3D frameworks at 2 mA crrr².

Figures 13A-13D: Voltage profiles in symmetric Li|Li@Cu cells with 3D Cu foil (Fig. 13A) or planar Cu (Fig. 13B). Fig. 13C shows average voltage hysteresis of Li metal plating/stripping at 0.25 mA crrr². Fig. 13D shows cycling performances of an Li anode with 2D and 3D current collectors in a full cell with a LiFePCL cathode at 0.5 C.

Figures 14A-14B: Fig. 14A shows an SEM image of the 50th Li stripping of the commercial Ni foam coated with Cu at current density of 0.5 mA crrr² for a total of 1 mAh crrr² of Li. Fig. 14B shows an SEM image of the polycarbonate objects with too small pore size (100 to 200 nm) under the same test condition. Figures 15A-15C: Fig. 15A shows a schematic of 3D Cu-fabric current collector preparation by a two- step dip-coating method. Fig. 15B and Fig. 15C show SEM images of prepared Cu-fabric (Fig. 15B) and an individual fiber of Cu-fabric (Fig. 1 5C).

Figure 16: The compressibility of the Cu-fabric. The extension is measured for ten layers of cloth. The initial thickness of each layer is 291 pm. After each step of compressing, pressure was released and re-applied onto the samples, which are denoted as specimen 1 -4. As shown in Fig. 16, the slope converges to the same value for each specimen. The intrinsic compressibility can be obtained from the slope, and has an estimated compressibility b = ( dV/dp)/V of 1.5 x 1CT⁷ N/m.

Figures 17A-17B: Schematics of lithium deposition during a lithium plating/stripping process on a 3D Cu-fabric (Fig. 17A) and 2D bare copper foil (Fig. 17B).

Figures 18A-18E: Fig. 18A shows the Coulombic efficiency of lithium plating/striping on 3D Cu-fabric under different pressure. Figs. 1 8B-1 8C show SEM images of the morphology of the lithium deposition after 20th plating under 0 degrees of pressure (Fig. 18B) and over 90 degrees of pressure (Fig. 1 8C). Figs. 18D-18E show SEM images of the morphology of the lithium deposition after 10th plating on bare 2D copper (Fig. 1 8D) and 3D Cu-fabric (Fig. 18E). (Current density: 2mA/cm², area capacity: 1 mAh/cm²).

Figure 19: SEM image of the morphology of Celgard 2325 (polypropylene-polyethylene separator) in a Li | |Cu-fabric battery after turning the screws of the battery to 240 degrees.

Figures 20A-20L: Scanning electron microscopy images of the morphology of lithium deposition after 10^th, 50^th and 100^th plating on 3D Cu-fabric and 2D bare Cu. Figs. 20A-20C show lithium deposition after the 10^th, 50^th and 100^th plating on 3D Cu-fabric. Figs. 20D-20F show lithium deposition after the 10^th, 50^th and 100^th plating on 2D bare copper. Figs. 20G-20I show lithium deposition after the 10^th, 50^th and 100^th plating on 3D Cu-fabric. Figs. 20J-20L show lithium deposition after the 10^th, 50^th and 100^th plating on 2D bare copper. In Figs. 20A-20F, the scale bar is 1 0 pm, and in Figs. 20G-20L the scale bar is 100 pm.

Figures 21 A-21 C: SEM images of a cross-section view of the Cu-fabric after 1 0^th (Fig. 21 A), 50^th (Fig. 21 B), and 100^th (Fig. 21 C) plating. All experiments were performed at a current density of 2mA/cm² and an area capacity of 1 mAh/cm².

Figures 22A-22D: Modelling lithium deposition under external pressure. Fig. 22A shows a schematic illustration of the model. p_ext and r_s are external pressure and the surface tension pressure, respectively. / is the total friction on the lithium exerted by each layer of cloth. Fig. 22B shows an illustration of the suppression of Li trunk and more even distribution of Li observed in experiments.

Fig. 22C shows the maximum volume of lithium that can be stored in the fabric as a function of external pressure, with estimated compressibility b 1.5 x 10^_7m/N, average gap size s_{0 ¾} 1 mth and friction / ¾ 300N/m. The predicted optimal external pressure of 4.78 MPa is very close to the experimental value by a factor of only 2-3. Fig. 22D shows a schematic illustration of the performance of structured current collector. Darker shade indicates larger lithium storage capacity inside the 3D Cu-fabric structure that has a high porosity.

Figures 23A-23C: SEM images of Cu-fabric based on polyester (Fig. 23A) and silk (Fig. 23B). Fig.

23C shows the Coulombic efficiency of lithium striping/plating on Cu-fabric based on polyester/silk and 2D bare Cu.

Figures 24A-24D: CE of lithium striping/plating on 2D bare copper and 3D Cu-fabric substrate under different current density and area capacity with the electrolyte being EC/DEC=1 :1 ,1 M LiPFe). Fig.

24A shows the CE under 1 mA/cm² current density and 1 mAh/cm² area capacity. The inset in Fig.

24A shows an SEM image of the morphology after cycling 500 cycles. Fig. 24B shows the CE under 2mA/cm² current density and 1 mAh/cm² area capacity. Fig. 24C shows the CE under 1 mA/cm² current density and 2mAh/cm² area capacity. Fig. 24D shows voltage-time profiles of the lithium stripping/plating process with 2mA/cm² current density and 1 mAh/cm² area capacity in Li||Li(black) and Li@Cu-fabric ||Li@Cu-fabric (gray) symmetric batteries.

Figure 25: Coulombic efficiency of lithium striping/plating on Cu-fabric. The electrolyte is

dimethoxyethane (DME)/ 1 ,3-dioxo!ane (DOL)=1 :1 , 1 M LiTFSI, and 1 % UNO3.

DETAILED DESCRIPTION OF THE INVENTION

The invention provides a structure for use as an electrode or current collector, e.g., an anode, in a battery. The structures of the present invention are resistant to lithium dendrite formation, a common detrimental effect in traditional lithium ion batteries, due to their high porosity that allows lithium to deposit uniformly within the pores of the material, which also allows for volumetric changes in the lithium during electrical cycling. The invention further provides methods of coating a surface of a structure using readily available and inexpensive reagents to form a 3D conductive framework.

Polydopamine (PDA) coatings form on nearly all types of material surfaces, giving an extremely versatile and highly effective platform for secondary metal coatings. A dip-coating method can turn virtually every porous material, ranging from metal to semiconductor to insulator, into an effective 3D current collector. This method includes cost effective compounds and operates under mild reaction conditions, making it highly competitive for scale up. For example, the 3D conductive architecture with large specific surface area can greatly reduce the ion flux density and provide enough sites for homogenous Li nuclei distribution and growth. The porous scaffold also acts as a rigid host to accommodate the volume change of the Li metal. Hence, this 3D current collector maintains an enhanced CE of 94% for 600 h at 0.5 mA crrr² and long-term cycling stability is demonstrated in full batteries (Li @ 3D GF-Cu | LiFePCL). This facile and versatile strategy of metal coating makes a great step towards building an ideal scaffold for Li encapsulation and enormously broadens the choices of suitable 3D porous materials for hosting the Li metal. Batteries of the invention can be cycled for 1000 h or more.

Structures

A structure of the present invention includes a base material, a layer of metal coordinating polymer, e.g., polydopamine, in contact with the base material, and an outer layer of a metal, e.g., copper, in contact with the metal coordinating polymer, e.g., polydopamine, layer.

Suitable materials for the base material include porous conducting, semiconducting, or insulating materials. The pores in the structure may be randomly ordered or may have an ordered pattern. Exemplary pore sizes range from 100 nm to 100 pm, e.g., 100 nm to 10 pm, 400 nm to 100 pm, 400 nm to 10 pm, 400 nm to 1 pm, 750 nm to 10 pm, or 750 nm to 1 pm. For example, useful materials for the base material include, but are not limited to, metals, e.g. a metal foam, glass, e.g., glass fiber, polymers, e.g., polyether or polycarbonate, paper, e.g., rice paper, natural fibers, e.g., cotton or silk, synthetic fibers, e.g., polyester, or composites thereof. Fabrics, e.g., including silk, cotton, or polyester, may also be employed. For example, a base material for a structure of the invention may be a fabric blend, e.g., 80% polyester and 20% cotton. In some cases, the fabric includes fibers, e.g., synthetic fibers, having diameters from 10 nm to 100 pm, e.g., 1 0 nm to 100 nm, 10 nm to 500 nm, 10 nm to 1 pm, 10 nm to 50 pm, 100 nm to 1 pm, 100 nm to 10 pm, 100 nm to 100 pm, 500 nm to 1 pm, 500 nm to 10 pm, 500 nm to 100 pm, 1 pm to 50 pm, or 1 pm to 100 pm.

In the present invention, the metal of the outer layer of structure may be transition metal, e.g., Ni, Co, Fe, Mn, Rh, Pt, Cu, Mo, W, or a combination thereof. An exemplary metal is copper. The metal, e.g., copper, may be uniformly coated on the metal coordinating polymer, e.g., polydopamine, layer.

Suitable metal coordinating polymers include polydopamine. In the present context, polydopamine includes unmodified polydopamine and derivatized polydopamine. For example, dopamine may be derivatized according to formula:

(see, e.g., ACS Appl Mater Interfaces. 201 8 Mar 7;10(9):7523-7540).

Specific derivatives include 3,4-dihydroxy-L-phenylalanine, norepinephrine, 6-nitrodopamine, 2- bromo-N-[2-(3,4-dihydroxyphenyl)ethyl]-2-methyl propenamide, and 5-hydroxydopamine. Other polymers may be formed from nitrogen-free phenols and polyphenols. Examples include hydrocaffeic acid, alkylcatechol, thiol-terminated catechols, gallol, pyrogallol, tannic acid, epigallocatechin gallate (EGCG), epicatechin gallate (ECG), and epigallocatechin (EGC). Metal coordinating polymers also include proteins such as Mfp-3 and -5. Other metal coordinating polymers are known in the art.

Batteries

Structures of the present invention may be used to form an electrode, e.g., an anode, or a portion of an electrode, such as a current collector, of a battery.

An exemplary battery of the invention includes a cathode comprising a redox active species, a separator, an electrolyte, and an anode comprising a current collector. The current collector is in contact with ions in the electrolyte. The separator is between the cathode and the anode. Exemplary materials for separators are nonwoven fibers (cotton, nylon, polyesters, glass), polymer films

(polyethylene, polypropylene, poly (tetrafluoroethylene), polyvinyl chloride), ceramics, and naturally occurring substances (rubber, asbestos, cellulose).

The battery may be under pressure, e.g., when the structure includes a fabric, e.g., 80%

polyester/20% cotton. Without wishing to be bound by any particular theory, the application of a pressure to a softer fabric structure allows for the retention of an increased lithium content while modifying the lithium deposition to allow for the formation of a smooth surface without sharp nuclei to form dendrites that may damage the battery separator. Increasing the lithium content of the structure and smoothness of the surface increases Columbic efficiency and electrical, e.g., cycling, performance. The pressure applied to the structure in a battery may be from 0.1 MPa to 20 MPa, e.g., 0.1 MPa to 15 MPa, 0.1 MPa to 5 MPa, 0.1 MPa to 3 MPa, 0.1 MPa to 1 , MPa, 0.4 MPa to 10 MPa, 0.4 MPa to 5 MPa, 0.4 MPa to 3 MPa, 0.4 MPa to 1 MPa, 0.7 MPa to 5 MPa, 0.7 MPa to 3 MPa, 0.7 MPa to 1 MPa, 1 MPa to 3 MPa, 1 MPa to 5 MPa, 1 MPa to 10 MPa, or 1 MPa to 15 MPa.

Electrodes

The structures of the invention may be used as an anode of a battery.

Cathodes for batteries of the present invention include a redox active species. Examples include LiFeP04, UC0O2, UMn204, L MnOs, LiMn02, LiFe02, LiFesOs, UFes04, and lithium nickel manganese cobalt oxide. Others are known in the art.

In one embodiment of a battery of the invention, the cathode includes LiFeP04, PVDF, and carbon black, and the anode includes a structure of the invention that has been lithiated.

Electrolytes

Suitable electrolytes for batteries include alkali metal, e.g., lithium, sodium, or potassium, salts dissolved in an appropriate solvent. The concentration of the alkali metal salt in the electrolyte may be from about 1 M to about 9M, e.g., from about 1 M to about 4M, from about 2M to about 5M, from about 3M to about 6M, from about 4M to about 7M, from about 5M to about 8M, or from about 6M to about 9M, e.g., about 1 M, about 2M, about 3M, about 4M, about 5M, about 6M, about 7M, about 8M, or about 9M. An exemplary alkali metal salt for use in an electrolyte is LiTFSI (lithium

bis(trifluoromethanesulfonyl)imide), LiPF6, LiAsF6, L1BF4, UCF₃SG₃, or LiCIC .

Suitable solvents include polar aprotic solvents, e.g., carbonates (such as diethyl carbonate, ethylene carbonate, and mixtures thereof), dimethoxyethane, dioxolane, and mixtures thereof. Solvents for the electrolyte may include other solutes, e.g., acids (e.g., HCI) or bases (e.g., NaOH or KOH) salts (e.g., KCI or UNO3), or alcohols (e.g., methyl, ethyl, or propyl) or other co-solvents to increase the solubility of a particular component in the electrolyte or to stabilize the interface on the anode or cathode side.

Methods of Coating

The invention features methods of coating a surface of a structure. The method involves forming a metal coordinating polymer, e.g., polydopamine, layer on the surface of the structure and then contacting the metal coordinating polymer, e.g., polydopamine, layer with metal, e.g., copper, ions under conditions to form a layer of metal, e.g., copper, on the surface.

To form a polydopamine layer on the surface of a structure, dopamine may be dissolved into a solution using a suitable solvent, such as a buffer, e.g., 10 mM Tris-HCI, at an appropriate concentration. Prior to dopamine layer formation, the structure to be coated may be heated using methods known in the art, e.g., in air, under vacuum, or in the presence of an inert gas, e.g., N2, to compact the pores of the structure. The dopamine layer may be formed on the surface of the structure by immersing the structure into the solution of dopamine for an appropriate length of time. For example, the thickness of the polydopamine coating on the structure is dependent on the length of time the structure is in contact with the dopamine solution. In some cases, the formation of a suitable layer of dopamine on the structure may take from 1 to 5 days, e.g., 1 day, 2 days, 3, days, 4 days, or 5 days.

To form a layer of copper on the polydopamine layer of the structure, the polydopamine-coated structure may be placed into contact with, e.g., immersed in, a source of copper ions, such as a solution comprising a copper salt, e.g., CuC . A solution of copper ions may include other components, such as acids, e.g., H3BO3, bases, e.g., NaOH, chelators, e.g.,

ethylenediaminetetraacetic acid, or reducing agents, e.g., dimethylamine-borane (DMAB).

Formation of a copper layer on the polydopamine layer of the structure may occur at conditions suitable to allow this process to occur over a longer time scale, e.g., room temperature and pressure, to ensure the formation of a uniform layer of copper on the polydopamine layer of the structure. In some cases, the formation of a suitable layer of copper on the polydopamine layer of the structure may take from 1 to 5 days, e.g., 1 day, 2 days, 3, days, 4 days, or 5 days. EXAMPLES

Example 1. Preparation of Structures

Polydopamine (PDA) coating: dopamine is a small-molecule with both catechol and amine functionalities, and is a simple mimic of the blue mussel ( Mytilus edulis) foot protein. The protein is rich in 3,4-dihydroxy-L-phenylalanine (DOPA) and lysine amino acids and performs an important role in the mussels’ adhesive ability. Dopamine may participate in reactions of bulk solidification and form strong covalent and noncovalent interactions with substrates.

A dopamine solution was obtained by dissolving dopamine into 10 mM Tris-HCL (pH=8.5). In order to have a compact structure after solution soaking, the glass fiber substrate was heated in air at 500°C for 2 h before use. After dipping the substrates into the aqueous dopamine solution, dark brown color was observed over time. The thickness of the PDA coating is a function of the immersion process and can be reached up to 50nm after 24 hours. ^[15al As a result, the immersion process was continued for 24 hours to ensure the formation of the polydopamine coating on substrates. The as-immersed substrates were washed with distilled water and dried at 60 °C.

Electrode metallization: an aqueous solution of 50 mM CuCL, 0.1 M H3BO3, 50 mM

ethylenediaminetetraacetic acid (EDTA) was prepared, followed by a pH adjusting process with 1 M NaOH until the pH is 7. Before dipping the PDA coated substrates into the solution, 0.1 M

dimethylamine-borane (DMAB) was added. The substrates were kept in the solution for 6 hours at 35°C and then 12 hours at room temperature. Again, the Cu coated substrates were rinsed with distilled water and dried at 60 °C under vacuum. The mechanical robustness of the as-prepared electrodes is shown in Figures 1 A-1 E. To demonstrate the mechanical durability of the prepared 3D Cu samples, the samples were treated by 500 °C annealing for 2h, soaking in DEC for 2h, and cycling for 20 times, respectively. Then we compared the digital-camera images and SEM images among original sample and treated samples. Figure 1 A shows that no geometrical or copper color change was observed after annealing, soaking or cycling, indicating the strong mechanical robustness of the coated copper on the fiber. The SEM images in Figures 1 B-1 E show that the porous inner structure of samples stays the same after treatments.

Electrochemistry: All batteries were carefully prepared in the glove box with a Swagelok cell configuration. 3D copper electrode and 2D copper foil were assembled and tested with lithium metal as the anode and Celgard 2500 (polypropylene) as the separator. Batteries were cycled at different current densities ranging from 0.5 mA/cm² to 1 .5 mA/cm² with a capacity of 1 mAh/cm². Lithium was first deposited on the copper electrode for 1 mAh/cm² and then charged to 0.5 V to delithiate from the copper electrode.

For the symmetric battery test, a pre-lithiation process was first conducted using lithium metal as the anode. The lithium metal was then replaced by a new copper electrode. The batteries were charged and discharged for 1 h at current density of 1 mA/cm². For the full battery test, a LiFePC cathode was prepared by coating an active material (80 wt%), carbon black (10 wt%), and PVDF (10 wt%) solution (in NMP) on an Al foil and then dried under vacuum. The pre-lithiated 3D copper are cycled with LiFePC as a counter electrode at 0.5 C.

EC/DEC (1 M LiPFe) was used as electrolyte in all batteries.

The procedure to obtain the 3D porous Cu foil is schematically presented in Figure 2 and illustrated using different original materials in Figure 3A. The preparation process first involved the immersion of the substrates in an aqueous dopamine solution and then dipping polydopamine coated objects into CuCl2 solutions. After the first immersion step of 24 hours the colors of all the samples, including a metal coin, polyester, polycarbonate, paper, glass fiber and nickel foam, changed toward brown, which clearly demonstrated that an adherent polymer film was deposited on the object surface (Figure 3A). Further evidence for dopamine polymerization was found by X-ray-photoelectron-spectroscopy (XPS) measurements. Taking the glass fiber as an example, the characteristic XPS substrate signals for unmodified GF, such as silicon (-100 eV), were highly suppressed after polydopamine coating. Instead, nitrogen (-399.5 eV) in PDA was clearly observed, as shown in Figure 3B. Scanning electron microscopy (SEM) images showed that before and after coating with PDA, the pore-size distributions were almost identical, as shown in Figures 3C-3E; this suggested that the PDA coating has little influence on the pore-size distributions of the substrates. The preserved porous nature was confirmed by quantitative analysis of the porosity measurement. As shown in Figure 4, the PDA coating does not alter the porosity of the substrates for the specific battery application of interest here.

After the PDA coating, the metal-binding ability of catechols was exploited to form homogenous metal coatings via a single dip-coating step. The distinct color transformation after the Cu coating step for all the samples in Figure 3A demonstrated the high efficiency and generality of our method, which can coat Cu onto any original material, from metal to insulator, and with varying microstructures (Figure 5). Furthermore, the procedure only involves low-cost compounds and operations under mild reaction conditions. Similar to the PDA coating, XPS results (Figure 3B) confirmed the success of the metal film deposition and the porosity measurements (Figure 4) showed no significant reduction of the pore sizes. In addition, similar results in the porosity (Figure 4) and XPS signals (Figure 6) using a polycarbonate substrate confirmed that the effectiveness of this procedure was not limited to glass fiber, but is rather general. Therefore, via this simple and low-cost“dip-coating” method, common porous materials can be easily turned into a conductive 3D matrix ready for further Li deposition with conformal Li entrapment for lithium anode applications.

Example 2. Mechanical Properties of Cu-coated Structures

The stabilization of the electrode dimension is highly beneficial in lithium metal anodes. When deposited on a 2D planar substrate, small dendritic Li grows from bare planar electrodes to promote further dendrite growth. Also, drastic volume expansion during continuous cycling can easily destroy the SEI layer, which in turn results in the inhomogeneous concentration of Li⁺ flux to further accelerate the Li dendrite growth and the rapid consumption of the electrolyte. In contrast, the design of the 3D conductive frameworks with large internal surface area can effectively reduce the current density and provide enough space to accommodate the Li deposition and alleviate the drastic volume change of Li metal during battery cycling. Moreover, the interconnecting conductive inner-structure can effectively regulate the electronic and ionic transportation and distribution properties. As shown in Figures 7A-7B and 8A-8F, distinct“hot spot” regions with high local current densities are observed among the top regions of Li metal nuclei. Figures 7A-7B show 2D simulations performed using COMSOL

Multiphysics with the“Tertiary Current Distribution, Nernst-Planck (tcdee)” module. The data shown here used D = 5x10^-6 cm²/s as the Li⁺ ion diffusivity in the electrolyte, which is close to the

experimental value (on the order of 10^-6 cm²/s). The electrolyte concentration, temperature and average current density are all the corresponding experimental values. According to the SEM pictures in Figs. 3C-3E, the 3D copper network has a diameter distribution around 1 pm. Thus, the diameters of the large and small circles are chosen as 1 pm and 0.6pm, and the positions of these circles are randomly chosen. Note that the observed phenomena remain largely unchanged for different directions and sizes of the 3D copper wire. Further according to the SEM images of Figs. 3C-3E, the deposition of Li to the 3D copper network has a limited thickness. These unconstrained hot spots will eventually grow into the Li dendrites. In contrast, the hot spots are effectively prevented with the introduction of the 3D conductive structure due to the more homogeneous distribution of the electric fields. Hence, a more uniform Li deposition should improve the lifespan of Li-metal anodes.

To confirm this hypothesis, the planar copper and the 3D conductive frameworks based on five different materials are used here as the counter electrodes, respectively, to observe the anode morphologies. In this work, a standard carbonate electrolyte (1 M LiPF6 in ethylene carbonate (ECydiethyl carbonate (DEC)) was used without other additives. Compared with either electrolyte, it shows more intrinsic improvement related to the 3D structure, since the SEI layers formed are dominated by Li alkyl carbonate (ROCO2U) in carbonate electrolyte and are usually too fragile to prevent the Li dendrite formation. i¹⁶l First, 1 .0 mA h/cm² lithium was first deposited on planar copper and 3D conductive frameworks at 0.5 mA/cm², and the stripping process was cut-off at 0.5 V. SEM images clearly demonstrated the lithium morphology after 20 and 100 cycles (Figures 9A-9F). The current density in Figures 9A-9F is 0.5 mA cm ², at such a low current density, small Li particles grow from both bare planar and 3D electrodes. However, upon cycling lithium particles are hosted well by the Cu wire in the 3D structure, while lithium metal chunks were formed on the 2D copper foil. For 2D planar copper anodes, there was a large area of bumpy Li plated on the Cu surface as demonstrated in Figures 9A-9F after 20 cycles. As the cycling time increases, electrons tend to accumulate at the sharp ends and drastically amplify the growth of lithium dendrites, consuming much electrolyte, and further resulting in inhomogeneous Li deposition.!^17] On the contrary, the fully plated Li morphology on the 3D copper current collectors differs significantly from that on the planar Cu. Taking the 3D GF-Cu as an example, the surface after 20 cycles remained flat with no obvious Li dendrites from SEM. Upon further plating, micron-sized Li particles gradually deposited in the porous structure and grew into lumps instead of dendritic or mossy morphology. This effective accommodation of lithium is due to the high specific surface area of the 3D conductive structures, where the active surface area for lithium plating is much higher than ordinary planar Cu. Consequently, the actual local current density is significantly reduced in the 3D anode, consistent with our simulation as shown in Figures 7A, 7B, and 8. In addition, we also investigated the morphology change of the 3D porous current collector after Li stripping. As shown in Figures 10A-10B, the integrity of these porous structures can be preserved well when the deposited Li strips out of the composites, indicating good mechanical properties.

Example 3. Electrical Properties of Cu-coated Structures

Coulombic efficiency (CE) is a critical parameter to evaluate the cycling sustainability of Li. Thus two- electrode cells (3D GF-Cu/Li and planar Cu foil /Li) were assembled to investigate the overall electrochemical performance. As expected, the 3D GF-Cu current collector presented a more stable and longer cycling life. The electrode current density was first set as 0.5 mA crrr² for a constant charge capacity of 1 mA h cm^·2. The CE of 3D GF-Cu could maintain an average value of 94% for more than 200 cycles (Figure 1 1 A), while that of the control Cu foil exhibited a continuous degradation during the cycles, indicating the unique advantages of the 3D GF-Cu with a larger reversible capacity and reusable Li. In the initial several cycles, the relatively low CE resulted from the electrode activation and formation progress of a stable SEI layer. The initial plated Li on the 3D framework has a high surface area and it tends to react with electrolyte more than the plated Li on the 2D Cu foil.

After first several cycles, the CE of planar Cu electrode is lower than 3D GF-Cu, which may be related to the formation of dendritic lithium metal and the cracking of SEI. At a high current density of 1 .0 and 2.0 mA crrr² that usually results in low efficiency and rapid dendrite growth, the differences between two samples are even more obvious (Figure 1 1 B, 1 1 C). SEM images of the Li deposited 2D and 3D current collectors at a high current density are shown in Figures 12A-12H, the extremely

inhomogeneous lithium deposition can be obviously observed on the 2D Cu foil, while the surface of 3D GF-Cu is even. The 3D GF-Cu presented superior cycling performance for more than 100 cycles at 2 mA/cm², while the planar Cu foil current collector exhibited much larger fluctuations in coulombic efficiencies and an obvious drop at 50 h (Figure 1 1 C). Previous studies!¹²¹⁵ _’ ¹⁸i have shown that high current rates not only generate more dendrite growth, but also can cause more severe damage to the SEI layer.!^{1 1}· ^19] The stable cycling of the modified cells at relatively high current density hence indicate that the free-standing 3D GF-Cu membrane is effective in suppressing Li dendrite growth.

The benefits of 3D framework can be directly seen from the high rate lithium plating. A much more even distribution of lithium metal was observed in the 3D current collector than the 2D counterpart, while dendritic Li appeared on the 2D planar after high current density cycling. In the 3D framework, lithium metal was hosted by the copper coated fiber, which prevented the formation of dendritic lithium particles. In contrast, the accumulation of extremely inhomogeneous lithium deposition can be observed from the 2D copper foil after lithium plating.

The voltage profiles of Li-plating-stripping process were also investigated for all the samples. The voltage hysteresis is defined as the difference between the voltages of lithiation and delithiation performances, which is mainly influenced by the current density and interfacial properties. Figure 13A shows the stable deposition/stripping behaviors of 3D GF-Cu with the nearly constant hysteresis of 40 mV. The reduced polarization of Li plating/stripping on the 3D GF-Cu is generally stable without obvious fluctuations for 150 cycles. While for 2D Cu foil-based anode, the hysteresis of Li depositing/stripping was much larger (=70 mV) in Figure 13B. The difference of the voltage hysteresis for the 2D and 3D copper foils is summarized in Figure 13C. The large irreversible deposition may induce thick and unstable SEI layers, consume more Li and electrolytes, cause the capacity loss and further lead to short circuits.¹²⁰¹ As evidenced in Figures 9A-9F, large masses of visible dendrites and dead Li were founded on the surface of bare Cu foil after cycling. In contrast, 3D GF-Cu favors low voltage polarization due to the larger surface area of the porous Cu, which can reduce the practical current density and restrict Li plating within the pore space reserved for Li accommodation.

Example 4. Battery Tests using Cu-coated Glass Fiber Structures

Li|LiFeP04 full battery tests were performed (Figure 13D) using bare Li metal and 3D GF-Cu @ Li at 0.5 C. The slight capacity increase in the first few cycles is due to the activation process of the LiFeP04 materials. The 3D GF-Cu @ Li cells exhibited an improved cycling stability, which retained 91 % of their initial discharge capacity after 200 cycles with the stable CE of 99.5%. In comparison, the bare Li metal showed more evident capacity fading, which only delivered 81 % of initial discharge capacity at 200th cycle.

We also investigated the electrochemical performance of the other current collectors derived from the different matrices shown in Figure 3A. Substrates with different pore structures have a great influence on the CE stability. The CE stability of commercial Ni foam with large pore diameters range from 200 to 400 pm was even worse than that of the 2D Cu, which exhibited a continuous degradation after 50 cycles. SEM images shown in Figure 14A of the 50th Li plating and stripping indicated that although Li metal can fill in the foam pores, much of the inserted Li cannot get out of the Cu coated Ni foam, forming dead Li with large irreversible capacities. In contrast, polyether, polycarbonate objects with pore sizes that are too small (100 to 200 nm) act much like the planar 2D current collectors, since most of the deposited Li metal was found on the surface (Figure 14B). Additionally, excellent mechanical stability is also necessary to construct an ideal 3D current collector. Consequently, rice paper substrates present an average CE over 90% over 100 cycles at 0.5 mA cm ², but drops below 70 % after 50 cycles at higher rates such as 1 and 2 mA cm ². Strong conductive 3D current collectors derived from the matrixes with proper pore structures of glass fiber, PC or PETE with pore size range from 400 nm to 1 pm achieved better cycling performance.

Example 5. Development of a Cu-fabric Structure

Copper coated 3D soft fabric (Cu-fabric) current collectors were developed to show superior battery performance under an optimized mechanical stress for the lithium metal anode application. The role of the external pressure can be evaluated by cycling performance and scanning electron microscopy (SEM) imaging. By modelling the dynamics of the lithium deposition, it was found that there exists an optimal external pressure for the cloth to hold the largest amount of lithium inside the pores. Such design principle is directly applied to and confirmed by the experiments.

The framework of the current collector is collected from lab clothing, which is both low cost and scalable. The composition of the clothing fabric is 80% polyester and 20% cotton. Cu-fabric was prepared by a two-step dip-coating method, shown in Figure 15A t^{30 31}!, that is low cost and has no extreme requirements for chemicals, experimental conditions or facilities. The clothing fabric was soaked into 2 mg/mL dopamine solution (dissolved in pFI=8.5, 0.01 M Tris-FICL buffer) for 48h under room temperature. Then, the polydopamine (PDA) coating clothing fabric (PDA@clothing fabric) was collected by washing three times with distilled water. Secondly, the PDA@clothing fabric was dipped into CuCl2 aqueous solution (0.05M CuCL, 0.05M ethylenediaminetetraacetic acid, 0.1 M FI3BO3, NaOFI, pPI=7), and 0.1 M dimethylamine-borane was added as reducing agent. This step was performed at room temperature for 24h until the solution was colorless, and the fabric was covered by a thin copper layer. The Cu-fabric was rinsed by distilled water for three times and then dried at 90°C. The morphology of Cu-fabric was characterized by SEM. The mechanical properties of the Cu fabric were measured by a materials testing machine (Instron, USA). Combined with the application of the optimal external pressure, the 3D conductive fabric can modify the lithium deposition to a smooth surface without sharp nuclei, and shows a superior cycling performance of 500 cycles (1000 h working hours).

Electrochemical Measurement: A Swagelok cell structure was utilized for all battery assemblies. To optimize the external pressure for the battery test, different levels of external pressure were applied during battery assembly by screwing the Swagelok tightly to different degrees. To test coulombic efficiency, 3D Cu-fabric or 2D bare Cu was used as working electrode, lithium foil as counter electrode and Celgard 2325 as separator. The electrolyte is 1 M hexafluorophosphate (LiPFe) in ethylene carbonate (EC) and diethyl carbonate (DEC), EC/DEC=50/50 (v/v), battery grade. All the batteries were assembled in argon filled glove-box. The battery tests were performed in LAND cell test instrument. After rest for 1 h, 1 mAh/cm² lithium was deposited to 3D/2D current collector, and then the batteries were cycled with 1 mA/cm² or 2mA/cm². Higher area capacity (2mAh/cm²) of deposited lithium was also applied in this system. The morphologies of the lithium deposition on 3D/2D current collector were characterized by scanning electron microscopy. During the symmetric cell test, 6mAh lithium was deposited on the 3D/2D current collector, and then the batteries were discharged/charged at 2mA/cm² for 30min in each cycle. The voltage-time profile was collected.

Example 6. Characterization of Cu-fabric Structures

SEM images of the Cu-fabric and single Cu fiber in the fabric are shown in Figures 15B and 15C.

Each clothing fiber was covered by a thin copper layer, and the interweaved structure was observed. The gray arrows in Figure 15C show the Cu layer and clothing fiber inside separately. The fiber is around 10 pm diameter and the distance between fibers is less than 2pm. The elastic modulus of the interweaved fabric was measured to show an estimated compressibility around 1.5 x 10^-7 N/m as shown in Figure 16. The Cu-fabric framework thus serves as a container and a soft cushion that regulates the lithium deposition. Finally, a 3D Cu-fabric is optimized for use in a battery system. The deposited lithium is mostly pressed into the soft Cu-fabric and shows some flat bulks on the surface (Figure 17). The deposited lithium is distributed uniformly on Cu fabric during the plating/striping process, leaving a wave-shaped surface. On the contrary, the dendrite grows quickly on the 2D Cu foil and penetrates through the Celgard separator. The assembled Li-Cu battery also shows a stable cycling performance at high current density and area capacity. The symmetric battery based on lithium deposited on Cu-fabric (Li@Cu-fabric) shows more stable performance compared to that with Li deposited on 2D Cu (Li@2D bare Cu).

Example 7. Lithium Deposition in Battery Configurations using Cu-fabric Structures

Li||Cu batteries were assembled in a Swagelok cell configuration to evaluate the electrochemical performance of the 2D/3D current collectors. The use of a Swagelok cell contributes to a denser lithium deposition due to the pressure inside the cell. Moreover, the reaction between deposited lithium and electrolyte will be depressed to increase the Coulombic efficiency, which represents the ratio between the amount of stripped and plated lithium every cycle. 3D Cu-fabric is more sensitive to pressure, since it is a soft substrate with special microstructure. In order to choose an optimal pressure for battery test, different external pressure was applied on the battery by screwing the Swagelok cell tightly until 0, 90 and 180 degrees (Figure 18A). After 1 mAh/cm² lithium was deposited to the 3D Cu-fabric current collector, the batteries cycled at 2mA/cm² current density. We focused on the CE of the battery cycling, the battery under 90 degrees and 180 degrees pressure showed higher CE (>90%) than the 0 degree one (<85%). Compared with the SEM images of lithium deposition under 0 degrees and over 90 degrees pressure (Figures 18B and 1 8C), it is clear that the deposited lithium is more porous without enough pressure, which would consume much more electrolyte. The effect of substrate structure was also not manifested after the lithium deposition if external pressure is not large enough. The 90 degrees tightness showed the best cycling performance for over 200 cycles with CE over 90%. When the degrees increased to 1 80, the battery also failed quickly. It is possible that lithium was pressed into the holes of the separator (Celgard 2325), which was subsequently damaged. As Figure 19 shows, the holes of the Celgard separator were filled by lithium particles when too much pressure was applied (>240 degree). As a result, screwing the Swagelok cell to 90 degrees is the optimized condition for better battery performance, equivalent to an external pressure of -0.81 MPa.

To observe if the 3D Cu-fabric can regulate lithium deposition, SEM morphology images of 2D/3D current collector surface after 10^th plating was collected, using Swagelok cell with screwing tight to 90 degrees (Figures 18D and 18E). There is significant difference between the 2D and 3D current collector. On the 2D hard bare Cu surface, the lithium deposited with more sharp nuclei and holes. Some of the deposited nuclei are tree-like, which are more likely to induce large dendrite growth and damage the separator. In contrast, the tree-like lithium deposited onto soft 3D Cu-fabric substrate was squeezed flat (Figure 20E). No sharp protruding area was observed on the surface of 3D Cu current collector. The results indicated that 3D soft Cu-fabric has the potential to modify the shape of the lithium deposition and improve the corresponding long-cycle performance.

To further investigate the lithium deposition process on 2D hard and 3D soft Cu current collector,

Li||Cu batteries were assembled to run for 10/50/100 cycles at 2 mA/cm² current density with

1 mAh/cm² area capacity. After the last plating curve, the batteries were disassembled, and the current collectors were studied with SEM. First, the difference of the surface morphology details was studied during cycling. The scale bars in Figures 20A-20F are 10pm. On 3D Cu-fabric (Figures 20A-20C), the deposited lithium was pressed into separate droplets after 10 cycles. The droplets are around 1 0-20 pm. When the battery kept cycling to 50 cycles, some of the droplets fused together. Flat regions around 40pm were generated, most of which were connected by“bridges” (Figure 20B dashed lines). After 100 cycles, the flat regions composed of lithium droplets grew larger to around hundreds of pm (Figure 20C dashed line area). During cycling, the deposited lithium was pressed into a flat area.

From the cross-section view of the 3D current collector after deposition (Figure 21 ), most of the deposited lithium was inserted into the Cu-fabric. The surface of the current collector was always smooth. On the contrary, the lithium deposited on 2D bare Cu appeared to be much sharper as shown in Figures 20D and 20E. The small nuclei around 2pm existed everywhere after 10 cycles and more smaller nuclei were generated after 50 cycles, and worsened after 100 cycles when the small nuclei stacked together. The comparison between 2D/3D current collectors is more clearly seen in a larger area. The scale bars in Figures 20E-20L are 100pm. With the 3D soft Cu-fabric cushion, the squeezed lithium droplets gradually contacted each other, and the surface of the current collector was covered with lithium uniformly. The wave structure of the fabric remained, and the plated lithium was dense after 100 cycles. However, on 2D hard Cu, the deposited lithium was porous during the cycling. After 10 cycles, the lithium was mossy with many holes. After 50 cycles, several large nuclei appeared and after 100 cycles, a lithium dendrite grew larger than 200pm (where the black arrow points to in Figure 20L). This dendrite was likely to penetrate the separator and short-circuit the battery.

According to the SEM results, the 3D Cu-fabric provided a soft substrate to regulate the shape of the deposited lithium and a large storage space for deposited lithium.

The SEM pictures showing lithium deposition inside the fabric and a gradual disappearance of the geometric features of fabric surface indicates a key balance, as illustrated in Figure 22A. At the beginning of the lithium deposition, bulk lithium is squeezed into the porous region inside the fabric by the external pressure. The two resistance forces are the surface tension pressure r_s = 2 a/s ( a is the surface tension coefficient of Li metal at room temperature) because of the curved surface of the squeezed-in lithium and the friction / between the squeezed-in lithium and the cloth fibers, which is proportional to the number of layers N the lithium has penetrated. This squeezed-in process of lithium will be stopped when the resistance forces grow larger than the external pressure. Since the mobility of Li metal is limited, this process is limited by kinetics. Thus, there is simultaneous plating of Li metal along with squeezing effect (Figure 22A). As is revealed by experiment, the Cu fabric also shows much better ability to suppress a critical step in lithium dendrite formation, the trunk growth of lithium P¹L As shown in Figure 22B, because the modulus of Li (4.9 GPa) is much larger than the modulus of the separator (« 1 GPa), for bare Cu case with a hard Cu substrate, the Li trunk induced pressure Ap cannot reshape the Li trunk. Thus, with the growing of the Li trunk, the deformation of the separator and extra pressure are also accumulating, leading finally to the cracking of the separator. However, in the Cu-fabric case, because of the gaps between the fibers, the extra pressure induced by the Li trunk is sufficiently conducted to the bottom of the bulk Li. This extra pressure drives these Li metal to deform toward the inside the fabric as well as spreading out. For more significant trunk growth with larger Ap, the extra pressure at the bottom, Ar', Ar” and Ap'" are all larger. Thus, the porous nature of the cloth provides a robust negative feedback that prevents the growth of Li trunks.

Quantitatively, the key force balance is expressed as

Pext S = Nf + p_a - S .

Suppose the compressibility b = is a constant under the experiment conditions, we have

where V₀ is the original volume of the pores under no pressure. With cylindrical symmetry about y direction (Figure 22A), V_pore is proportional to s². Also, because the total volume of Li penetration inside the cloth per unit length in y direction is V_{ln cioth} = kN s² (k 1), assuming the friction / is also a constant, we can write the equilibrium formula for V_{ln cloth},

Interestingly, this equation predicts an optimal external pressure that maximizes V_{ln cloth}, which is opt ₌ _2 2ae^1/3

^Pext 3b 3 s₀

as a function of only materials-specific quantities. Based on above calculation and the experimental results shown in the figure, we obtained the maximum amount of Li storable inside the fabric as a function of the external pressure, using the experimentally measured of estimated compressibility b ~ 1.5 x 10^-7m/N, average gap size s_{0 ¾} Impi and friction / 300N/m. (Figure 22C). We identified an optimal pressure that is quite close to the experimental pressure with a difference of only a factor less than 2 or 3.

For example, the porosity of the cloth without pressure is 10% and the thickness and area are 200 pm and 1 cm², under optimal pressure, the porous volume decreases to about 1 /e. Thus, under optimal pressure the total volume inside the cloth is

0.1

100 X— * 0.74mm³

while the total volume of Li in one charge or discharge is 0.48mm³. Thus, for the 200 pm sample used, Li penetration almost to the other side of the fabric is observed (Figure 21 ). Based on our analysis, increasing the porosity of the cloth and decreasing the compressibility are two ways for making a commercially favorable thin Cu-fabric. Our methods in principle can make, for example, a fabric that supports all Li metal to be submerged inside with 25 pm thickness, if the fabric is made by a porous material with 20% porosity and compressibility much smaller than our cloth. This is also shown in Figure 22D, a schematic illustration of the performance of structured current collector. We can see that while smaller compressibility and larger porosity is preferred, while too large porosity leads to reduced structure-Li interaction as in the case of 3D Cu wires.

We tested the electrochemical performance of fabric with different materials composition and thickness after Cu coating (Figure 23). The Cu-fabric based on polyester (784pm) and silk (128 pm) both showed over 300 cycles cycling performance at 1 mA/cm² current density, and 1 mAh/cm² area capacity. This indicates that the fabric structure of the anode made the long stable cycling lithium metal battery possible with a variety of structures, and especially, with thinner fabric. Note that the materials used are all readily available fabrics. It is desirable to make a specifically designed material achieve similar performance with much smaller thickness than 100 pm.

Li||Cu batteries were also applied to test the long-cycle electrochemical performance of 3D Cu-fabric electrode. Lithium was deposited to the 3D/2D current collector, then the battery cycled by charging to a cut-off voltage at 0.5V and discharging at 1 mA/cm² or 2mA/cm². The Coulombic efficiency is the main parameter we considered to evaluate the cycling stability. When CE decreases quickly or shows large fluctuations, it usually means that lithium dendrite have grown large enough to penetrate through the separator. The Coulombic efficiencies of lithium stripping/plating process on 2D/3D current collector were shown in Figures 24A-24C. In Figure 24A, the CE of 3D Cu-fabric electrode maintains over 90% stability for 500 cycles, 10OOh at 1 mA/cm² current density and 1 mAh/cm² area capacity. On the contrary, the CE of 2D bare Cu electrode dropped to be less than 80% quickly after about 60 cycles. It is clear that the 3D Cu-fabric shows a much longer cycle life. The battery after 500 cycles was disassembled and characterized the surface morphology of the 3D Cu-fabric. No obvious nucleation existed on the electrode surface and the shape of the fiber was still maintained. The 3D Cu-fabric regulated the lithium deposition into the fiber and reduced the damage to the separator. At higher current density of 2mA/cm², batteries with 3D Cu-fabric current collector still cycled over 200 cycles while the battery with 2D bare Cu current collector failed after 30 cycles (Figure 24B). At higher area capacity of 2mAh/cm², 3D Cu-fabric electrode battery maintained CE over 90% for 100 cycle, 400h. Under the same test condition, 2D bare Cu electrode battery short circuited after 60 cycles, 240h (Figure 24C). Comparing the CE performance between 2D/3D Cu current collector, the 3D Cu- fabric greatly improved the stability of lithium metal anode. These results are consistent with simulation and SEM images of morphology as shown in Figures 18A-18E, Figures 20A-20L, and Figures 24A-24D. As a result, the soft Cu-fabric structure was proved to be a versatile method, applicable to various materials for improving the safety of lithium metal anode. To get higher coulombic efficiency, an ether-based electrolyte was chosen, and CE comes to over 95% with stable cycling over 300 cycles. Li@Cu||Li@Cu symmetric batteries were also used for testing the stability of electrodes (Figure 24D). The lithium was firstly deposited to the 2D/3D Cu current collector and then collected to assemble symmetric batteries. The batteries were cycled at a high current density of 2mA/cm², and

discharge/charge for 30 min. The polarization of Li@3D Cu-fabric battery (=0.1 V) is much smaller than that of Li@2D bare Cu battery (-0.2V) for the first several cycles. After 50 cycles, the Li@3D Cu- fabric battery was still stable, while the Li@2D bare Cu battery showed large polarization. Based on the comparison of electrochemical performances between 2D and 3D Cu-fabric current collectors, 3D Cu-fabric electrode shows great promise to achieve a more stable lithium deposition on the anode at higher current density (Figure 25).

In Examples 5-7, we developed and demonstrated a soft 3D Cu-fabric current collector which achieved superior long-cycle performance with the commercial carbonate electrolyte. Combining the experimental and simulation results, we suggested an optimal external pressure to the battery assemble. The lithium deposition was regulated and squeezed into the Cu-fabric during cycling. The 3D Cu-fabric surface maintained smoothness without obvious dendrite growth after cycling.

Simulations were also performed to revel the mechanism of suppressing lithium dendrite growth by 3D Cu-fabric. A principle for 3D-structured current collector design was also obtained. The 3D Cu fabric described herein is prepared from readily available clothing fibers using a scalable dip-coating method, which increases applicability for large-scale industrial production due to the low cost and straightforward processing.

REFERENCES

[1 ] a) W. Xu, J. Wang, F. Ding, X. Chen, E. Nasybulin, Y. Zhang, J.-G. Zhang, Energy Environ. Sci. 2014, 7, 513; b) D. Lin, Y. Liu, Y. Cui, Nat. Nanotechnol. 2017, 12, 194; c) M.-H. Ryou, D. J. Lee, J.-N. Lee, Y. M. Lee, J.-K. Park, J. W. Choi, Adv. Eng. Mater. 2012, 2, 645.

[2] a) D. Wang, W. Zhang, W. Zheng, X. Cui, T. Rojo, Q. Zhang, Adv. Sci. 2016, 4, 1 ; b) R. Zhang, X. B. Cheng, C. Z. Zhao, H. J. Peng, J. L. Shi, J. Q. Huang, J. Wang, F. Wei, Q. Zhang, Adv. Mater. 2016, 28, 2155; c) K. Zhang, G.-H. Lee, M. Park, W. Li, Y.-M. Kang, Adv. Eng. Mater. 2016, 6, 160081 1 .

[3] F. Ding, W. Xu, G. L. Graff, J. Zhang, M. L. Sushko, X. Chen, Y. Shao, M. H. Engelhard, Z. Nie, J. Xiao, X. Liu, P. V. Sushko, J. Liu, J.-G. Zhang, J. Am. Chem. Soc. 2013, 135, 4450.

[4] Y. Lu, Z. Tu, L. A. Archer, Nat. Mater. 2014, 13, 961 .

[5] F. Ding, W. Xu, X. Chen, J. Zhang, M. H. Engelhard, Y. Zhang, B. R. Johnson, J. V. Crum, T.

A. Blake, X. Liu, J. G. Zhang, J. Electrochem. Soc. 2013, 160, A1894.

[6] W. Li, H. Yao, K. Yan, G. Zheng, Z. Liang, Y. M. Chiang, Y. Cui, Nat. Commun. 2015, 6,

7436.

[7] a) D. Lin, Y. Liu, Z. Liang, H. W. Lee, J. Sun, H. Wang, K. Yan, J. Xie, Y. Cui, Nat.

Nanotechnol. 2016, 11, 626; b) G. Zheng, S. W. Lee, Z. Liang, H. W. Lee, K. Yan, H. Yao, H. Wang, W. Li, S. Chu, Y. Cui, Nat. Nanotechnol. 2014, 9, 618; c) D. Lin, J. Zhao, J. Sun, H. Yao, Y. Liu, K. Yan, Y. Cui, Proc. Natl. Acad. Sci. U. S. A. 2017, 114, 4613.

[8] a) D. Zhou, R. Liu, Y.-B. He, F. Li, M. Liu, B. Li, Q.-H. Yang, Q. Cai, F. Kang, Adv. Eng.

Mater. 2016, 6, 1502214; b) J.-C. Daigle, A. Vijh, P. Hovington, C. Gagnon, J. Hamel-Paquet, S. Verreault, N. Turcotte, D. Clement, A. Guerfi, K. Zaghib, J. Power Sources 2015, 279, 372; c) J. Zhang, B. Sun, X. Huang, S. Chen, G. Wang, Sci. Rep. 2014, 4, 6007; d) S. Choudhury, R. Mangal,

A. Agrawal, L. A. Archer, Nat. Commun. 2015, 6, 10101 ; e) P. L. Kuo, C. A. Wu, C. Y. Lu, C. H. Tsao, C. H. Hsu, S. S. Hou, ACS Appl. Mater. Interfaces 2014, 6, 31 56; f) Q. Lu, Y. B. He, Q. Yu, B. Li, Y.

V. Kaneti, Y. Yao, F. Kang, Q. H. Yang, Adv. Mater. 2017, 29, 13; g) Q. Lu, Y. B. He, Q. Yu, B. Li, Y. V. Kaneti, Y. Yao, F. Kang, Q. H. Yang, Adv. Mater. 2017, 29.

[9] B. Zhu, Y. Jin, X. Hu, Q. Zheng, S. Zhang, Q. Wang, J. Zhu, Adv. Mater. 2017, 29.

[10] a) H. K. Kang, S. G. Woo, J. H. Kim, J. S. Yu, S. R. Lee, Y. J. Kim, ACS Appl. Mater.

Interfaces 2016, 8, 26895; b) J.-S. Kim, D. W. Kim, H. T. Jung, J. W. Choi, Chem. Mater. 2015, 27, 2780.

[1 1 ] W. Liu, W. Li, D. Zhuo, G. Zheng, Z. Lu, K. Liu, Y. Cui, ACS Cent. Sci. 2017, 3, 135.

[12] a) X. B. Cheng, T. Z. Hou, R. Zhang, H. J. Peng, C. Z. Zhao, J. Q. Huang, Q. Zhang, Adv.

Mater., 2016, 28, 2888; b) R. Zhang, X. R. Chen, X. Chen, X. B. Cheng, X. Q. Zhang, C. Yan, Q. Zhang, Angew. Chem. Int. Ed. Engl. 2017, 56, 7764; c) R. Mukherjee, A. V. Thomas, D. Datta, E. Singh, J. Li, O. Eksik, V. B. Shenoy, N. Koratkar, Nat. Commun. 2014, 5, 3710; d) K. Olszowska, J. Pang, P. S. Wrobel, L. Zhao, H. Q. Ta, Z. Liu, B. Trzebicka, A. Bachmatiuk, M. H. Rummeli, Synth. Met. 2017, 234, 53.

[13] S.-S. Chi, Y. Liu, W.-L. Song, L.-Z. Fan, Q. Zhang, Adv. Funct. Mater. 2017, DOI:

10.1 002/adfm.2017003481 700348.

[14] S. Liang, Y. Li, J. Yang, J. Zhang, C. He, Y. Liu, X. Zhou, Adv. Mater. Techno!. 2016, 1.

[15] a) H. Lee, S. M. Dellatore, W. M. Miller, P. B. Messersmith, Science 2007, 318, 426; b) C. Zhang, Y. Lv, W. Z. Qiu, A. He, Z. K. Xu, ACS Appl. Mater. Interfaces 2017, 9, 14437; c) J. H. WAITE, M. L. TANZER, Science 1981 , 212, 1038.

[16] a) K. Yan, H. W. Lee, T. Gao, G. Zheng, H. Yao, H. Wang, Z. Lu, Y. Zhou, Z. Liang, Z. Liu, S. Chu, Y. Cui, Nano Lett. 2014, 14, 6016; b) J. Qian, W. A. Henderson, W. Xu, P. Bhattacharya, M. Engelhard, O. Borodin, J. G. Zhang, Nat. Commun. 2015, 6, 6362; c) R. Mogi, M. Inaba, S.-K. Jeong, Y. Iriyama, T. Abe, Z. Ogumi, J. Electrochem. Soc. 2002, 149, A1 578; d) H. Ota, Y. Sakata, X. Wang, J. Sasahara, E. Yasukawa, J. Electrochem. Soc. 2004, 151, A437.

[17] a) Z. Liang, D. Lin, J. Zhao, Z. Lu, Y. Liu, C. Liu, Y. Lu, H. Wang, K. Yan, X. Tao, Y. Cui, Proc. Natl. Acad. Sci. 2016, 113, 2862; b) K. Yan, Z. Lu, H.-W. Lee, F. Xiong, P.-C. Hsu, Y. Li, J. Zhao, S. Chu, Y. Cui, Nat. Energy 2016, 1 , 1 6010; c) M. D. Tikekar, S. Choudhury, Z. Tu, L. A. Archer, Nat. Energy 2016, 1, 161 14.

[18] a) Y. Liu, D. Lin, Z. Liang, J. Zhao, K. Yan, Y. Cui, Nat. Commun. 2016, 7, 10992; b) Z. Liang, G. Zheng, C. Liu, N. Liu, W. Li, K. Yan, H. Yao, P. C. Hsu, S. Chu, Y. Cui, Nano Lett. 2015, 15, 2910; c) W. Liu, D. Lin, A. Pei, Y. Cui, J. Am. Chem. Soc. 2016, 138, 15443. [19] S. M. Wood, C. H. Pham, R. Rodriguez, S. S. Nathan, A. D. Dolocan, H. Celio, J. P. de Souza, K. C. Klavetter, A. Heller, C. B. Mullins, ACS Energy Lett. 2016, 1, 414.

[20] a) L. L. Lu, J. Ge, J. N. Yang, S. M. Chen, H. B. Yao, F. Zhou, S. H. Yu, Nano Lett. 2016, 16, 4431 ; b) C. P. Yang, Y. X. Yin, S. F. Zhang, N. W. Li, Y. G. Guo, Nat. Commun. 2015, 6, 8058.

[21 ] R. Zhang, N. W. Li, X. B. Cheng, Y. X. Yin, Q. Zhang, Y. G. Guo, Adv. Sci. 2017, 4, DOI 10.1 002/advs.201600445.

[22] C. P. Yang, Y. X. Yin, S. F. Zhang, N. W. Li, Y. G. Guo, Nat. Commun. 2015, 6, DOI 10.1 038/ncomms9058.

[23] L. Liu, Y. X. Yin, J. Y. Li, N. W. Li, X. X. Zeng, H. Ye, Y. G. Guo, L. J. Wan, Joule 2017, 1, 563.

[24] P. Bai, J. Li, F. R. Brushett, M. Z. Bazant, Energy Environ. Sci. 2016, 9, 3221 .

[25] X. Wang, W. Zeng, L. Hong, W. Xu, H. Yang, F. Wang, H. Duan, M. Tang, H. Jiang, Nat.

Energy 201 8, 3, 227.

[26] T. Hirai, J. Electrochem. Soc. 1994, 141, 61 1 .

[27] C. Monroe, J. Newman, J. Electrochem. Soc. 2005, 152, A396.

[28] C. Monroe, J. Newman, J. Electrochem. Soc. 2003, 150, A1377.

[29] H. J. Chang, N. M. Trease, A. J. Ilott, D. Zeng, L. S. Du, A. Jerschow, C. P. Grey, J. Phys. Chem. C 2015, 119, 16443.

[30] L. Y. Qi, L. Shang, X. Chen, L. Ye, W. Zhang, P. Feng, W. Zou, N. Cao, H. H. Zhou, D. A. Weitz, X. Li, Adv. Mater. Interfaces 2018, 5, DOI 1 0.1 002/admi.201800807.

[31 ] H. Lee, S. M. Dellatore, W. M. Miller, P. B. Messersmith, Science (80). 2007, 318, 426.

Other embodiments are in the claims.

Claims

What is claimed is: CLAIMS

1 . A method for coating a surface of a structure, the method comprising:

a) forming a metal coordinating polymer layer on the surface of the structure; and b) contacting the product of (a) with metal ions under conditions to form a layer of metal on the surface.

2. The method of claim 1 , wherein the structure comprises metal, glass, polymer, paper, fabric, carbon, polyester, cotton or a composite thereof.

3. The method of claim 1 , wherein the structure is porous.

4. The method of claim 3, wherein the pores in the structure are randomly oriented.

5. The method of claim 3, wherein the pores in the structure are ordered.

6. The method of claim 1 , wherein the structure comprises natural or synthetic fibers.

7. The method of claim 2, wherein the structure comprises a composite of polyester and cotton.

8. The method of claim 1 , wherein the metal coordinating polymer is polydopamine.

9. The method of claim 1 , wherein the metal is copper.

10. The method of claim 1 , wherein the metal coordinating polymer is polydopamine and the metal is copper.

11 . A battery comprising:

a) an anode comprising a current collector comprising a structure having a metal layer on a metal coordinating polymer layer, wherein the current collector is in contact with lithium, sodium, or potassium ions; and

b) a cathode comprising a redox active species.

12. The battery of claim 11 , wherein the structure comprises metal, glass, polymer, paper, fabric, carbon, polyester, cotton or a composite thereof.

13. The battery of claim 11 , wherein the structure is porous.

14. The battery of claim 13, wherein the pores in the structure are randomly oriented.

15. The battery of claim 13, wherein the pore in the structure are ordered.

16. The battery of claim 11 , wherein the structure comprises natural or synthetic fibers.

17. The battery of claim 12, wherein the structure comprises a composite of polyester and cotton.

18. The battery of claim 11 , wherein the metal coordinating polymer is polydopamine.

19. The battery of claim 11 , wherein the metal is copper.

20. The battery of claim 11 , wherein the metal coordinating polymer is polydopamine and the metal is copper.