WO2015027147A1 - Hydrolysis deposition - Google Patents

Abstract

Materials comprised of a substrate and layer materials, and a method of making the same are disclosed. The substrate can be porous, and the layer material can be selectively located in the pores. The method exposes a substrate to an amount of water or water vapor selected to substantially locate only where deposition is required such as in the pores of a porous substrate. The water then hydrolyses a precursor material. The resulting product deposits selectively where the water had located. Controlling the amount of water controls the proportion of the surface that is covered by the layer and the depth of that layer. Successive cycles can result in multiple layers to either increase the thickness of the layer or to deposit layers with different compositions. Metal oxide layers thus deposited can be subsequently converted to other compositions such as metal nitrides, carbides, phosphides and oxynitrides.

Description

HYDROLYSIS DEPOSITION

CROSS REFERENCE TO RELATED APPLICATION

This application claims the benefit of the earlier filing date of U.S. provisional patent application No. 61/869,003, filed August 22, 2013, which is incorporated herein by reference in its entirety.

FIELD

This invention concerns a method for depositing materials onto a substrate using hydrolysis deposition, with certain embodiments selecting the location and amount of deposition by controlling the amount of water, and embodiments of products made by the method.

BACKGROUND

There are several known deposition methods. Despite these known methods, it has been a significant challenge for both academia and industry to form a thin coating on materials, especially nanoporous materials, on a large scale. Most deposition techniques, such as physical vapor deposition, plasma-assisted/ion-beam-assisted techniques, pulsed laser deposition and magnetron sputtering, are designed preferentially for coating planar substrates. These techniques typically cannot form a coating that extends into the pores.

There are other disadvantages associated with known techniques. For example, chemical vapor deposition (CVD) relies heavily on the reactivity between the precursors and the substrates. Electroless deposition requires sensitizer and catalyst nanoparticles to be deposited first, which leads to relatively thick coatings. Electrodeposition uses plating baths that require unique combinations of additives designed to be effective for a specific metal. Underpotential deposition (UPD) is a method on an atomic scale, but it is restricted by limited

substrate/deposition combinations that exhibit high bonding tendencies. Methods such as dip coating and spin coating involve solvent evaporation, which often suffers from unfavorable surface energy effects and a lack of fine control. In addition, supercritical drying is needed.

Currently, the state-of-the-art deposition method is atomic layer deposition (ALD). However, it has not yet been demonstrated that this method is effective for coating nanoporous materials, particularly mesoporous (2 nm < pore size < 50 nm) and microporous materials (pore size < 2 nm). ALD relies on a sequential self-limiting surface chemistry process employing gaseous precursors. Recently development on fluidized reaction bed and rotatory reactor has improved its scalability. Yet the self-limiting reaction nature of ALD still ultimately limits its scalability. Plus, the high cost of ALD associated to rare precursors and complicated equipment inhibits the commercial material processing applications.

Hydrolysis deposition methods are under-developed, particularly for applications for nanoporous materials. Controlled hydrolysis and condensation techniques, also called sol-gel processes, have been investigated for coating a variety of structures including microspheres, fibers and carbon nanotubes. Precursor and water concentrations are critical for these processes, such as the well-known Stober process, to control the hydrolysis rates and scale. Infiltration of mesoporous silica and carbon by sol-gel processes has been well established. Infiltration methods typically cannot achieve fine pore loading control and generally result in pores fully filled with material. Moreover, precursor materials, water and porous substrates are all mixed together. Substrates only provide a nucleation surface for hydrolysis reactants and water that are freely dispersed in the reaction system. Thus, with these techniques, it is difficult to control both where the hydrolysis occurs, and the quantity of the material deposited.

SUMMARY

In view of the above, a new method is needed for the controlled deposition of one phase or a mixture of phases as a material layer onto a variety of substrates or matrices, including porous substrates or matrices. Disclosed embodiments of the present method address that need and provide a method for selectively depositing a material onto a substrate.

For certain disclosed embodiments, a substrate is first exposed to an amount of water. The water can be in the form of water vapor. The substrate is then exposed to a dry, nonaqueous solution containing a hydrolysable compound. The hydrolysable compound is hydrolyzed by water adsorbed by the substrate, thereby forming a layer of hydrolysis product on the substrate. This layer only forms where the water was present. By controlling where the water locates, the location of the deposition can be controlled, allowing for selective deposition on portions of the substrate surface. In some embodiments, the substrate is a porous substrate, having a surface that defines a plurality of pores, each pore being defined by a pore surface that is a portion of the substrate surface. Exposing the substrate to an amount of water may comprise exposing the surface of the substrate to the amount of water vapor sufficient to locate the water vapor in the pores. In some embodiments the water locates substantially in the pores of the porous substrate, and in certain embodiments, the water vapor locates only in the pores. In other embodiments, however, the water locates substantially on the substrate surface and not in the pores. In addition, by controlling the amount of water to which the substrate is exposed, the amount of hydrolysis product formed can also be controlled. In some examples, the substrate surface has an interior portion that defines the pores and an exterior portion that does not define the pores, and at least 50% of the area of the exterior portion is not covered by the layer of hydrolysis product. In certain embodiments, the layer of hydrolysis product in located only in the pores.

Multiple layers can be made by repeating the process. Each added layer can comprise a different hydrolysis product; or multiple layers of the same product can be added to form a thicker composite layer; or some layers can be made from the same material and others from one or more different materials.

The method can be used to apply a material to a wide variety of substrates. Exemplary substrates include, but are not limited to, carbon, silica, alumina, titanium oxide, zeolites, metal organic frameworks, or polymeric structures, such as structures of cellulose. In certain embodiments, these substrates can be porous and the amount of water can be selected to ensure that it substantially adsorbs onto the surface of the pores. In particular embodiments, the substrate was CMK-3 or activated carbon, which are both examples of a porous carbon substrate, or porous silica.

Metal oxides are examples of hydrolysis products that can be deposited onto the substrate. These metal oxides can include, but are not limited to, S1O2, AI2O3, SC2O3, T1O2, V2O3, V2O5, Cr₂0₃, Mn0₂, Ga₂0₃, FeO, Fe₂0₃, Fe₃0₄, CoO, Co₃0₄, NiO, N1O2, CuO, Cu₂0, ZnO, SrO, Y2O3, Zr0₂, CdO, Ag₂0, Rh₂0₃, Nb₂0₃, Nb₂0₅, W2O3, W0₂, WO3, M0O2, M0O3, RuC-2, Re0₃, Re₂0₇, Ir0₂, PdO, PtO, Sn0₂, Sb₂0₃, Te0₂, Ge0₂, PbO, Pb0₂, Ce₂0₅, Ce0₂, EU2O3, or combinations thereof. In one particular embodiment the metal oxide was T1O2, and in another the metal oxide was Sn0₂, and in yet another particular embodiment the metal oxide

These metal oxide layers can be converted into other compositions. For example, in some embodiments substrates comprising a metal oxide layer deposited according to disclosed embodiments were treated with ammonia to form metal nitride layers. In one particular embodiment the metal nitride was TiN. Other metal compounds that can be formed include metal carbides, metal phosphides and metal oxynitrides.

For hydrophobic substrates the surface can be activated prior to the exposure to water. In certain embodiments carbon substrates were treated with an oxidizing agent to form oxygen- containing functional groups, such as carboxyl groups, on the substrate surface, which facilitates the water's adsorption onto the surface of the substrate. In some particular embodiments the oxidizing agent was ammonium persulfate, and in another embodiment it was nitric acid.

Also disclosed are materials, devices and products made by various embodiments of the disclosed method. These materials comprise a substrate comprising one or more material layers deposited according to disclosed embodiments of the present invention. In some embodiments the substrates were selected from porous carbon, including activated carbon, CMK-3 and porous graphene, porous silica, porous alumina, porous titanium oxide, porous zeolites, porous metal organic frameworks, porous structures of cellulose or other porous polymeric structures. The layers can comprise metal oxides, metal nitrides, metal carbides, metal phosphides or metal oxynitrides, or combinations thereof.

In certain particular embodiments the substrate was CMK-3 and a layer of either tin oxide, titanium oxide or titanium nitride was substantially located in the pores. In another embodiment the material was activated carbon with titanium oxide substantially deposited in the pores. In yet another embodiment the material was porous silica with a layer of tin oxide substantially located in the pores.

The foregoing and other objects, features, and advantages of the disclosed embodiments will become more apparent from the following detailed description, which proceeds with reference to the accompanying FIG.s. BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a schematic drawing illustrating features of certain disclosed embodiments of the present invention.

FIG. 2 is a schematic drawing illustrating features associated with conventional hydrolysis deposition methods.

FIG. 3 is a schematic drawing illustrating one embodiment of a precise water-loading device scalable for commercial production.

FIG. 4 schematically illustrates hydrolysis-based, sub-nanometer deposition on porous carbons according to certain disclosed embodiments of the present invention: A) porous carbon substrate; B) functionalized carbon substrate; C) water-loaded carbon substrate; and D) carbon substrate with an oxide deposition.

FIG. 5 is an SEM image of a Sn0₂/CMK-3 composite.

FIG. 6 is a Tin EDX mapping a Sn0₂/CMK-3 composite.

FIG. 7 is a schematic diagram of an electrical double layer capacitor. FIG. 8 is a schematic diagram of a battery.

FIG. 9 is a schematic diagram of a hydrogen fuel cell.

FIG. 10 is a schematic diagram of a gas sensor.

FIG. 11 is a graph of mass versus temperature illustrating the TGA curves of water- loaded samples formed by exposing C-APS to measured volumes of water vapor.

FIG. 12 is a schematic drawing illustrating a laboratory- scale, precise water loading device used for certain working embodiments of the present invention.

FIG. 13 is a plot of mass versus temperature illustrating the thermogravimetric analysis (TGA) of different samples.

FIG. 14 is a graph of volume of nitrogen adsorbed versus relative pressure to illustrate the N₂ sorption isotherms of samples of CMK-3, C-APS and C-TiO₂-100.

FIG. 15 is a graph of volume of nitrogen adsorbed versus relative pressure to compare the N₂ sorption isotherms of different samples made according to disclosed embodiments of the present invention.

FIG. 16 is a graph of pore size distributions of different samples made according to disclosed embodiments of the present invention.

FIG. 17 is an XRD pattern of Ti0₂ deposited on CMK-3.

FIG. 18 is an XRD pattern of a sample of bulk Ti0₂ powder.

FIG. 19 is a representative SEM image of a Ti0₂/CMK-3 composite according to disclosed embodiments of the present invention.

FIG. 20 is a TEM image of a Ti0₂/CMK-3 composite.

FIG. 21 is an EDX mapping image for carbon in a Ti0₂/CMK-3 composite.

FIG. 22 is an EDX mapping image for titanium in a Ti0₂/CMK-3 composite.

FIG. 23 is a graph of volume of nitrogen adsorbed versus relative pressure to illustrate the N₂ sorption isotherms of the C0₂ activated carbon.

FIG. 24 is a graph of the pore size distribution of the C0₂ activated carbon.

FIG. 25 are SEM and EDX images of Ti0₂ deposited on activated carbon.

FIG. 26 is a wide angle XRD pattern of TiN deposited on CMK-3.

FIG. 27 is X-ray photon spectra of CMK-3 with a layer of TiN and titanium 2P₃/₂ signals, illustrating the contributions of Ti0₂, TiON and TiN.

FIG. 28 is a TEM image of a TiN/CMK-3 composite.

FIG. 29 is an EDX mapping image of carbon in a TiN/CMK-3 composite.

FIG. 30 is an EDX mapping image of nitrogen in a TiN/CMK-3 composite. FIG. 31 is an EDX mapping image of titanium in a TiN/CMK-3 composite.

FIG. 32 is a graph of volume of nitrogen adsorbed versus relative pressure to illustrate the N₂ sorption isotherms of samples of C-Ti-N and C-TiO₂-100.

FIG. 33 is a graph of capacitance versus voltage illustrating the cyclic voltammograms of a TiN/CMK-3 composite at different scanning rates.

FIG. 34 is a graph of capacitance versus voltage illustrating the cyclic voltammograms of CMK-3 at different scanning rates.

FIG. 35 is a graph plots of capacitance versus voltage illustrating the cyclic

voltammograms at different scanning rates of a sample of C-APS that has been subjected to nitridation conditions to investigate possible N-doping effects.

FIG. 36 is a graph of voltage versus time for CMK-3 and a TiN/CMK-3 composite, illustrating the galvanostatic charge/discharge profiles at different current rates.

FIG. 37 is a graph that provides Electrochemical Impedance Spectroscopy (EIS) Nyquist plots obtained at the frequency range from 200 kHz to 10 mHz.

FIG. 38 is a graph that provides wide-angle XRD patterns of Sn0₂/CMK composites after one, two and three hydrolysis deposition cycles.

FIG. 39 is a graph of volume adsorbed versus relative pressure illustrating nitrogen adsorption/desorption isotherms of CMK-3, 1-Sn-CMK, 2-Sn-CMK and 3-Sn-CMK.

FIG. 40 is a graph of pore volume versus pore diameter illustrating the pore size distribution profiles of CMK-3 and the Sn0₂-CMK composites after one, two and three hydrolysis deposition cycles.

FIG. 41 is a graph that provides small angle XRD patterns of CMK-3 and C-APS.

FIG. 42 is a graph of weight loss versus temperature illustrating the TGA profiles of CMK-3 and the Sn0₂-CMK composites after one, two and three hydrolysis deposition cycles.

FIG. 43 is an SEM image of a 2-Sn-CMK composite.

FIG. 44 is a higher magnification image of the composite of FIG. 43.

FIG. 45 is a TEM image of a 2-SN-CMK composite.

FIG. 46 is an HRTEM image of a 2-SN-CMK composite.

FIG. 47 is an HRTEM image of a 2-SN-CMK composite showing Sn0₂ crystallites. FIG. 48 is a TEM image of a 2-Sn-CMK composite after 300 galvanostatic cycles.

FIG. 49 is a photograph that provides TEM images of 1-Sn-CMK (top) and 3-Sn-CK (bottom). FIG. 50 is a photograph that provides an HAADF-STEM image of 2-Sn-CMK and the corresponding elemental mapping images of carbon, tin and oxygen.

FIG. 51 is an EDX spectrum of 2-Sn-CMK, with the insets showing the SEM image of the part analyzed and the corresponding elemental composition.

FIG. 52 is a graph of potential versus specific capacity, illustrating the first-cycle galvanostatic discharge/charge profiles of Sn0₂/CMK compositions after one, two and three hydrolysis deposition cycles at a current density of 200 mAg^"1.

FIG. 53 is a graph of potential versus specific capacity illustrating the first-cycle galvanostatic discharge/charge profiles of C-APS at a current density of 200 mAg^"1 between 0.01 V and 2 V.

FIG. 54 is a graph of specific capacity versus cycle number illustrating the charge capacity cycling data for 1-Sn-CMK, 2-Sn-CMK and 3-Sn-CMK at a current density of 200 mAg^"1.

FIG. 55 is a graph of capacitance versus voltage illustrating the cyclic voltammograms of the initial four cycles of 2-Sn-CMK at a scan rate of 0.5 mVs^"1.

FIG. 56 is a graph of voltage versus specific capacity illustrating the galvanostatic discharge/charge profiles of 2-Sn-CMK for various cycles.

FIG. 57 is a graph of specific capacitance and coulombic efficiency versus cycle number illustrating the long term galvanostatic cycling performance and the corresponding coulombic efficiency for 2-Sn-CMK at a current density of 200 mAg^"1.

FIG. 58 is a photograph that provides a HAADF-STEM image of 2-Sn-CMK after 300 charge/discharge cycles and the corresponding elemental mapping images of carbon, tin and oxygen.

FIG. 59 is a photograph that provides TEM images of 3-Sn-CMK before and after 200 charge/discharge cycles at a current density of 500 mAg^"1.

FIG. 60 is a graph of potential versus specific capacity illustrating the charge/discharge profiles of 2-Sn-CMK cycled at various current densities.

FIG. 61 is a graph of weight loss versus temperature illustrating the TGA profiles of V₂0₅-RFC nanocomposites after the first five hydrolysis deposition cycles.

FIG. 62 is a graph of weight loss versus temperature for pure RFC.

FIG. 63 is a graph that provides the XRD patterns of 37-V₂0₅-RFC, 50-V₂O₅-RFC and 70-V₂O₅-RFC. FIG. 64 is a graph that provides the XRD pattern for bulk vanadium oxide formed by hydrolysis.

FIG. 65 is the XPS spectra for 55-V₂0₅-RFC (top) and 70-V₂O₅-RFC (bottom).

FIG. 66 is an SEM image of RFC.

FIG. 67 is an SEM image of 55-V₂0₅-RFC.

FIG. 68 is an expansion of the boxed area from FIG. 67.

FIG. 69 is an SEM image of 70-V₂O₅-RFC.

FIG. 70 is a TEM image of 55-V₂0₅-RFC.

FIG. 71 is a HAADF-STEM image of 55-V₂0₅-RFC.

FIG. 72 is a photograph that provides the vanadium EDX mapping image corresponding to the image of FIG. 71.

FIG. 73 is a photograph that provides the carbon EDX mapping image corresponding to the image of FIG. 71.

FIG. 74 is a graph of potential versus cycle number illustrating the galvanostatic charge/discharge profiles of RFC, bulk V₂0₅ and various composites.

FIG. 75 is a graph of specific capacitance versus cycle number illustrating the initial cycling performance of RFC, 55-V₂0₅-RFC, and 70-V₂O₅-RFC.

FIG. 76 is a graph of potential versus specific capacity illustrating the galvanostatic charge/discharge profiles for the third cycle of 55-V₂0s-RFC.

FIG. 77 is a graph of potential versus specific capacity illustrating the charge/discharge profiles for bulk V₂0₅.

FIG. 78 is a graph of current versus potential illustrating the cyclic voltammogram profiles for the first three cycles of 55-V₂0s-RFC.

FIG. 79 is a graph that provides a comparison of the CV profiles for 55-V₂0₅-RFC and 70-V₂O₅-RFC.

FIG. 80 is a graph of potential versus specific capacity illustrating the galvanostatic charge/discharge profiles for 55-V₂0₅-RFC at various current densities.

FIG. 81 is a graph of specific capacitance versus cycle number illustrating the rate and cycling performance of 55-V₂0₅-RFC at different current densities.

FIG. 82 is a graph that provides a linear plot of ih versus scanning rate (v^1/2).

FIG. 83 is a graph of current versus potential illustrating the CV profiles of 55-V₂0₅- RFC at different scanning rates. FIG. 84 is a graph of current versus potential illustrating the CV profile of 55-V2O5-RFC at a sweep rate of 5 mV/s.

FIG. 85 is a graph of current versus potential illustrating the CV profile of 55-V2O5-RFC at a sweep rate of 0.5 mV/s.

FIG. 86 is a graph of current versus potential comparing the CV profiles of 55-V2O5-

RFC and 1.0 M TEABF₄ in PC at 5 mV/s.

FIG. 87 is a graph of normalized current versus potential comparing the CV profiles of 55-V2O5-RFC and V2O5 at 0.5 mV/s.

FIG. 88 is a graph that provides Nyquist plots of RFC, 55-V₂0₅-RFC and bulk V2O5. FIG. 89 is a graph that provides the fitting for the Nyquist plot of 55-V2O5-RFC and the equivalent circuit.

FIG. 90 is a graph that provides the Nyquist plot and fitting data for bulk V2O5.

DETAILED DESCRIPTION

I. Definitions

As used herein, the singular terms "a," "an," and "the" include plural referents unless context clearly indicates otherwise. Also, as used herein, the term "comprises" means "includes.

Layer material refers to a material that forms at least a partial layer on at least a portion of the surface of the substrate. The layer may be substantially continuous or it may be discontinuous. The layer material may comprise a material that is deposited onto the substrate, according to disclosed embodiments of the present invention; a material that has been chemically modified after deposition by a post-deposition treatment, according to disclosed embodiments of the present method; or combinations thereof. Examples of layer materials include, but are not limited to, metal oxides, metal nitrides, metal carbides, metal phosphides, metal oxynitrides, or combinations thereof.

Composite refers to a substrate where at least a portion of the surface is covered by at least one layer material.

Mesoporous refers to a pore size between from about 50 nm to about 2 nm.

Metal carbide refers to a compound comprising a metal or metalloid and carbon.

Examples of metal carbides include, but are not limited to, SiC, TiC, VC, NbC, WC, or combinations thereof. Metal nitride refers to a compound comprising a metal or metalloid and nitrogen.

Examples of metal nitrides include, but are not limited to, TiN, Sr₃N₂, ScN, W₂N, WN, WN₂, VN, NbN, or combinations thereof.

Metal oxide refers to a compound comprising a metal or metalloid and oxygen.

Examples of metal oxides include, but are not limited to, Si0₂, A1₂0₃, Sc₂0₃, Ti0₂, V₂0₃, V2O5, Cr₂0₃, Mn0₂, Ga₂0₃, FeO, Fe₂0₃, Fe₃0₄, CoO, Co₃0₄, NiO, N1O2, CuO, Cu₂0, ZnO, SrO, Y₂0₃, Zr0₂, CdO, Ag₂0, Rh₂0₃, Nb₂0₃, Nb₂0₅, W₂0₃, W0₂, W0₃, M0O2, Mo0₃, Ru0₂, Re0₃, Re₂0₇, Ir0₂, PdO, PtO, Sn0₂, Sb₂0₃, Te0₂, Ge0₂, PbO, Pb0₂, Ce₂0₅, Ce0₂, Eu₂0₃, or combinations thereof.

Metal oxynitride refers to a compound comprising a metal or metalloid, oxygen and nitrogen, with a chemical formula M_xO_yN_z, where M refers to one or more metal or metalloid atoms, and x, y and z are each independently greater than zero. Examples of metal oxynitrides include, but are not limited to, Si₂N₂0, aluminum oxynitride (AION) and titanium oxynitride.

Metal phosphide refers to a compound comprising a metal or metalloid and phosphorus. Examples of metal phosphides include, but are not limited to, aluminum phosphide, copper phosphide, titanium phosphide or combinations thereof.

Microporous refers to a pore size of less than or equal to about 2 nm.

Nanoporous refers to a pore size of less than or equal to about 100 nm.

Oxidizing agent refers to a compound that removes one or more electrons from another compound in an oxidation-reduction (redox) reaction. Examples of oxidizing agents include, but are not limited to, nitric acid, ammonium persulfate, oxygen/ozone, potassium

permanganate, potassium dichromate, potassium chlorochromate (PCC), hydrogen peroxide, or combinations thereof.

Pore size is the average diameter of the pores for a particular substrate. The pore size is determined by a commonly used N₂ sorption technique coupled with a calculation method called Barrett-Joyner-Halenda method.

Pore volume is the sum of the volumes of all the pores. The pore volume is determined by the N₂ sorption method.

Porous material or porous substrate refers to a material containing or defining pores, which may or may not extend completely through the material. The skeletal portion of the porous material may be referred to as the matrix, and a pore is a space in the material that is not occupied by the material. Porous materials exhibit a lower density than non-porous materials of the same composition. Example of porous materials include, but are not limited to, porous carbon, including activated carbon, CMK-3, and porous graphene, porous metal oxides, such as porous S1O2, porous T1O2, porous AI2O3, porous zeolites, porous metal organic frameworks, porous polymeric structures, including porous structures of cellulose, and combinations thereof. II. Composites

A. Overview

Presently disclosed embodiments concern a hydrolysis-based deposition method for depositing materials onto a substrate. In certain embodiments that deposition has a

subnanometer thickness. Suitable substrates include non-porous and porous materials, including nanoporous (mesoporous or microporous) materials. In some embodiments deposition was performed at atmospheric pressure. Deposition proceeds by reacting a hydrolysable compound with water that is pre-adsorbed by the substrate, as schematically shown in FIG. 1. This is in contrast to a typical sol-gel process, where water and precursor are mixed together (FIG. 2). A person of ordinary skill in the art will recognize that hydrolysis using embodiments of the present invention can only occur where water exists and stops when the water is consumed. In some embodiments only a portion of the substrate surface is exposed to the water, thereby limiting the deposition of the hydrolysable material to that portion. Thus, a person of ordinary skill in the art will appreciate embodiments of the disclosed method can be used to achieve selective deposition on only a portion of a surface, such as selectively coating a flat surface, or depositing a material in a pattern onto the surface.

Porous materials adsorb water to different levels depending on the environmental humidity, and some embodiments of the disclosed method utilized this property to control the hydrolysis. The deposition thickness can be adjusted by regulating how much water is available to be adsorbed prior to hydrolysis, and also by how long the water-loaded substrate is in contact with the solution of hydrolysable compound. Thicker layers can also be made by successive hydrolysis deposition cycles. For certain working embodiments, a single deposition cycle at atmospheric pressure deposited a layer with a subnanometer thickness. Additionally, in some embodiments a carbon substrate was pre-treated with an oxidizing agent to produce an oxidized carbon substrate to facilitate water adsorption to the substrate. The resulting metal oxide layer was later converted into a metal nitride layer. Other deposited layers can be converted into a variety of other compositions such as metal oxynitrides, metal carbides or metal phosphides. B. Substrates

Embodiments of the disclosed method deposit one or more layers onto a substrate. A person of ordinary skill in the art will appreciate that the substrate material is substantially unlimited. Suitable substrates can include, but are not limited to, carbon, silica, titanium oxide, alumina, zeolites, metal organic frameworks or polymeric structures. The substrates can be of almost any size and shape, such as particles, wafers, sheets, bars, bricks, rods or complex shapes. A person of ordinary skill in the art will appreciate that the shape of the substrate is not limiting as long as the portion of the surface onto which the deposition is to occur can be accessed by the water and the solution of hydroly sable material. The surface of the substrate may be substantially smooth, or substantially textured, or a combination thereof. A portion of the substrate surface maybe substantially flat, or curved or a combination thereof. The substrate surface may also be a non-contiguous surface.

In certain embodiments the substrate materials are porous. Porous materials can have a pore size of about 1 micron or less, preferably 100 nm or less, more preferably 50 nm or less. A minimum pore size for hydrolytic deposition into substrate pores is the diameter of a water molecule, which is about 0.3 nm. In some embodiments, the porous substrate has a substrate surface where each pore has a pore surface defined by a portion of the surface, and an exterior surface portion that is not located in the volume of the pores. In some embodiments, more than 50% of the surface of the pores cannot be reached by a ray of light from any angle.

Substrate surfaces may be pre-treated prior to exposure to water. In some embodiments substrates with a hydrophobic surface were pre-treated to facilitate water adsorption. In other embodiments substrate surfaces may be treated to hinder water adsorption. The pre-treatment may substantially cover the substrate surface, or it may be a partial coverage, for example, to facilitate deposition in a pattern on the substrate, or to selectively deposit on a portion of a flat surface.

In some particular embodiments the substrate was a carbon material, preferably a nanoporous carbon material. In some particular embodiments the substrate was CMK-3 nanoparticles, and in others the substrate was activated carbon. In another embodiment the substrate was porous silica. Other examples of porous substrates include, but are not limited to, porous alumina, porous titanium oxide, porous graphene, porous zeolites, porous metal organic frameworks, or porous polymeric structures such as porous structures of cellulose. C. Precursor compounds

Embodiments of the disclosed method use the hydrolysis of a hydrolysable precursor compound to deposit material onto the substrate. Therefore suitable precursor compounds are those that will hydrolyze. When the precursor compound hydrolyzes the resulting hydrolysis product is deposited onto the substrate.

In certain embodiments the precursor material is a metal alkoxide. Metal alkoxides are compounds in which at least one metal or metalloid is bonded to one or more alkyl groups through an intermediate oxygen. The alkyl groups can be straight chains, branched, cyclized or a combination thereof. Examples of alkyl groups include, but are not limited to, alkyl groups having 1 to at least 10 carbon atoms, such as methyl, ethyl, n-propyl, isopropyl, cyclopropyl, n- butyl, isobutyl, sec-butyl, tert-butyl, cyclobutyl, n-pentyl, isopentyl, neo-pentyl, cyclopentyl, n- hexyl and cyclohexyl. The alkyl chain maybe include heteroatoms, such as oxygen, nitrogen and sulfur. Examples of alkoxides comprising such heteroalkyl chains include

methoxyethoxide, and methoxypropoxide. An alkoxide group may also be substituted. For example, alkoxide groups may be substituted with one or more heteroatoms, such as with haloethoxides, including trifluoroethoxide as an example. The metal alkoxide compound may comprise other groups or atoms in addition to the alkoxide group, such as alkyl groups, chelating groups such as diketones, for example 2,4-pentanedione (acac), amines, thiols and halides. Examples of metal alkoxides include, but are not limited to, titanium tetraisopropoxide, titanium ethoxide, zirconium ethoxide, aluminum isopropoxide, vanadium oxytriisopropoxide and titanium tert-butoxide. Exemplar metal alkoxides comprising additional groups or atoms include Sn(OEt)₂(acac)2, diethylaluminum ethoxide and chlorotriisopropoxytitanium.

In other embodiments the precursor material is a metal phenoxide. Metal phenoxides are compounds in which an aryl group is bonded to one or more metal through an intermediate oxygen. The aryl group may be substituted, for example with halides, alkyl groups, alkoxy group, amines and alkylamines.

In further embodiments the precursor material is a metal halide. Metal halides are compounds in which at least one metal or metalloid is bonded to one or more halides, such as fluorides, chlorides, bromides or iodides or combinations thereof. Examples of metal halides include, but are not limited to, Agl, Bil₃, CaF₂, Cal₂, Cdl, CuBr, Cul, Mgl₂, Nil₃, Pbl₂, Sbl ,

Snl₂, Snl₄, Til₄, Znl₂, Fel₂, Gel₄, FeCl₂, FeCl₃, SnCl₂, SnCl₄, InCl₃, SbCl₃, SbCls, GeCl₄, SiCl₄, ZnCl₂, TiCl₂, TiCl₃, TiCl₄, AgCl, CuAgI₂, CuCdI₂, CuBiI₄, CuPbI₃, CuSnls and SbSI. D. Layer materials

The layer material forms at least a partial layer on a substrate surface, particularly at a selected location, such as on the portion of the substrate surface that defines the pores, which portion of the surface is referred to herein as the interior surface portion. The portion of the substrate surface that does not define the pores is referred to herein as the exterior surface portion. The layer can be substantially continuous or discontinuous. In some embodiments the layer substantially coats the entire interior surface portion. In some embodiments, at least 50 % of the substrate exterior surface portion is not covered by the layer material, such as at least 75%, at least 90%, at least 95% or at least 99%. In some embodiments, none of the exterior surface portion is covered by the layer material, to the extent to which a person of ordinary skill in the art can reasonably determine using standard techniques such as scanning electron microscopy (SEM), transmission electron microscopy (TEM), high-angle angular dark-field scanning TEM (HAADF-STEM), energy-dispersive X-ray spectroscopy (EDX) elemental mapping and combinations thereof. In particular embodiments, SEM is used to determine if the deposited phase is on the exterior surface. In certain embodiments there are multiple layers. These layers can each comprise a different layer material, or at least two layers may comprise the same layer material. In some embodiments the layer material is the material deposited by the hydrolysis of the precursor material. In other embodiments the layer material is the material resulting from post-deposition processing of the material deposited by the hydrolysis of the precursor material.

In some embodiments the layer material comprised a metal oxide. Particular examples of metal oxide layer materials included vanadium oxide, titanium oxide and tin oxide. In another embodiment the layer material comprised a metal nitride. A particular example of a metal nitride layer material was titanium nitride.

Other examples of layer materials include, but are not limited to, metal carbides, metal oxynitrides or metal phosphides. Metal carbides can be formed in a variety of ways, for example, by magnesiothermic reactions on the surface of porous carbon that is coated by transition metal oxides, or by treatment with a suitable carbiding agent, such as carbon compounds having 10 or fewer carbon atoms, such as CH₄, C2H6, C2H_4j CCl₄, C3H8, C6H6, CH3CI or combinations thereof. Metal oxynitrides can be formed by treating or exposing nitrides deposited on the surface of substrates to one or more oxidizing agents, such as air or oxygen. Metal phosphides can be formed by heating oxide coated substrates under an atmosphere comprising a phosphorus-containing gas. It will be readily apparent to one of ordinary skill in the art that contact between the layer material and the atmosphere may result in a reaction at the surface of the layer material, producing at least a partial layer of impurities such as oxides. Thus, solely by way of an example, a layer material comprising titanium nitride may have, on its surface, oxides of titanium nitride, TiO_xN_y, and/or titanium oxide.

III. General Embodiments of a Method for Forming Composites

A general method for forming the substrate/layer material composite comprises:

1) exposing the substrate to an amount of water; and

2) exposing the substrate to a non-aqueous solution of precursor compound thereby only depositing the hydrolysis product where water is present on the substrate.

In certain embodiments of the disclosed method the substrate was exposed to water vapor. FIG. 3 illustrates a device 300 that could be used to provide a selected amount of water vapor to a substrate. The substrate 310 is placed in the sample chamber 320 of an oven 330. An agitator 340 is connected to a motor 350. Water is heated in a container 360 by a heater 370 to produce a water vapor/air mixture. A selected amount of this water vapor/air mixture is introduced into the sample chamber 320 through an inlet 380. The oven is maintained at a temperature sufficient to ensure that the water vapor does not condense to water droplets. After the substrate has been exposed to the water vapor for a sufficient time for the required adsorption to take place, the sample chamber is flushed with nitrogen gas from a nitrogen source 390 to remove excess water vapor.

The amount of water vapor is calculated using the tabulated vapor pressures for water. According to Dalton's law, each component gas in the mixture of gases exerts a partial pressure proportional to its molar percentage. For example, at 80 °C, the vapor pressure of water is 355.1 mmHg. If the atmospheric pressure is 760 mmHg, the molar percentage of water in the water/air mixture is 355.1/760 = 46.7%. The total moles of gas mixture can be calculated by the ideal gas law: PV = nRT, where n is the molar number, R is the gas constant, 0.082

(L-atm)/(mol- K). The total moles, n, can be calculated since P, V, T are all known, and the portion of water can be estimated as well. Using this, or other techniques, a known amount of water vapor can be selected sufficient to produce a desired amount of hydrolysis product, and/or to locate in the pores. In some embodiments, the water vapor locates substantially in the pores of the substrate. The substrate is then exposed to the hydrolysable precursor compound. In some embodiments the precursor compound is maintained in a non-aqueous solution. In certain embodiments the substrate is added to a solution of the precursor compound. In other embodiments the solution is added to the substrate, or to a portion of the substrate. The precursor compound may be present in a molar excess compared to the adsorbed water, or there may be about an equal amount of precursor compound and water, or there may be less precursor compound than water. The substrate may be agitated while in the solution, such as by stirring, to ensure efficient mixing of the substrate and precursor compound. The substrate is kept in contact with the precursor material for a time period sufficient to allow the adsorbed water to hydrolyze the precursor compound, thereby depositing the layer material. The time period can be from about 1 minute to greater than 24 hours, preferably from about 15 minutes to about 10 hours, more preferably from about 45 minutes to about 2 hours. In some embodiments the time period is selected to achieve a particular quantity of deposition. For example, longer time periods, such as longer than about 1 hour, may be selected to achieve greater amounts of deposition. After hydrolysis, the substrate is removed from contact with the solution. In some embodiments this was achieved by filtration. In some embodiments, the substrate is

subsequently heated at a temperature of from greater that about 25 °C to greater than 400 °C, such as from about 100 °C to about 400 °C, or from about 200 °C to about 300 °C. The heating is performed for a time period of from about 1 minute to greater than 24 hours, preferably from about 30 minutes to about 10 hours, more preferably from about 90 minutes to about 2.5 hours. In some embodiments, the substrate is heated in air.

Suitable non-aqueous solvents for the precursor compound solution are any which dissolve the precursor compound and do not interfere with the deposition process according to the present invention, such as by dissolving the layer material. Polar and non-polar aprotic solvents, such as acetonitrile, DMSO, dimethylformamide, toluene, benzene, hexanes, cyclohexane, tetrahydrofuran, ethers, such as diethyl ether, methyl tert-butyl ether and diphenyl ether, cyclic ethers, such as 1,3-dioxolane and dioxane, and pyridine can be used. Protic solvents including alcohols can also be used. Examples of suitable alcohols include methanol, ethanol, n-propanol, isopropanol, n-butanol, iso-butanol, sec-butanol, tert-butanol and phenol. In some embodiments the alcohol is selected to match the alkoxide present in the precursor material. For example, isopropanol for a metal isopropoxide. In certain embodiments the solvent is selected to be miscible with water. In some embodiments these process steps may be repeated to increase the thickness of the layer material or to achieve a desired loading, such as a desired weight percent, of the layer material. In certain other disclosed embodiments at least one additional step may be included, such as changing the composition of the layer material from, for example, a metal oxide to a metal nitride, carbide or phosphide. For some embodiments involving a hydrophobic substrate, such as a carbon-based substrate, the substrate may be treated prior to exposure to the water, to facilitate water adsorption. In some working embodiments this treatment comprised contacting the substrate with an oxidizing agent. In some embodiments the surface of the substrate may be functionalized by the treatment, such as forming carboxyl groups on the surface. Other functional groups that may result from the treatment include, but are not limited to, aldehydes, ketones, hydroxyls, halides, amines, thiols, ethers, nitros, sulfonic acids, sulfonates, phosphonic acids and phosphates.

FIG. 4 schematically illustrates one disclosed embodiment for one cycle of hydrolysis deposition on a porous carbon substrate. In FIG. 4B the surface of the porous carbon was exposed to an oxidizing agent. The substrate was then exposed to a selected amount of water, which located substantially in the pores (FIG. 4C). Water-loaded carbon samples were then soaked in a non-aqueous solution of a hydrolysable compound (FIG. 4D). The hydrolysable compound was selected such that, upon hydrolysis, the desired layer material was deposited as the hydrolysis product.

In an exemplary embodiment, CMK-3, a mesoporous carbon, was selected as the carbon backbone for T1O2 or TiN composites. The preparation of CMK-3 can be found herein, and in detail from the literature, for example in Jun, S. et al. J. Am. Chem. Soc. (2000), 122, 10712- 10713. Briefly, mesoporous silica was prepared from Pluronic P123 and tetraethylorthosilicate in hydrochloric acid solution. The mesoporous silica was then added to a solution of sucrose in acidic water. The solution was heated and sonicated. After carbonization the silica was removed by treatment with a HF solution leaving the mesoporous carbon CMK-3.

CMK-3 was functionalized according to disclosed embodiments of the present invention by oxidation with ammonium persulfate, thereby forming carboxyl groups on the surface.

Without being bound to any particular theory, in addition to increasing the hydrophobicity of the surface, acids can also catalyze hydrolysis of metal alkoxide precursors. For example, in the formation of Ti0₂ a first step of Ti(0'Pr)₄ hydrolysis may be the following:

Ti(OⁱPr)₄ + [H₃0]⁺ -> Ti(0'Pr)₃(OH₂ ⁺) + j-PrOH. The functionalized CMK-3, hereafter referred to as C-APS, was then loaded with an amount of water before hydrolysis. In other embodiments porous carbon materials were not oxidized and were directly used for water loading and hydrolysis. In order to know the limits of T1O2 loading by this strategy, the minimum and maximum hydrolysis inside C-APS was investigated. To determine the minimum hydrolysis C-APS was dried and soaked for one hour in a pre-dried solution of titanium tetraisopropoxide (TTIP) under a nitrogen atmosphere. To determine the maximum hydrolysis the C-APS was saturated with water vapor prior to being exposed to the TTIP solution. The resulting samples were referred to as C-T1O2-O and C-T1O2- 100 respectively.

For loadings in between the minimum and maximum levels, the porous substrate was exposed to specific amounts of water vapor. By limiting the amount of water vapor C-APS was exposed to, samples with about 33% and 67% loading were prepared.

Metal oxides can be converted into metal nitrides by nitridation. In some embodiments a post-deposition treatment with ammonia was used to form metal nitrides from deposited metal oxides. In one particular embodiment a deposition of T1O2 on CMK-3 was converted to TiN by treatment with ammonia gas at a temperature from about 500 °C to about 1000 °C, preferably from about 700 °C to about 900 °C. The CMK-3 was exposed to the ammonia from about 3 hours to about 9 hours, preferably from about 5 hours to about 7 hours. The TiN layer improved the conductivity of the CMK-3 substrate. The composite takes advantage of both light carbon to provide a framework with a high surface area and TiN nanoparticles for a better electronic conductivity and power performance.

A person of ordinary skill in the art will appreciate that the disclosed embodiments are not limited to T1O2. For example, in certain working embodiments SnC was deposited onto CMK-3 using tin(IV) tert-butoxide as the precursor material. FIG. 5 shows the SEM image of Sn02/CMK-3 composite with the surface free from SnC , and FIG. 6 shows the corresponding tin elemental EDX mapping. In another embodiment SnC was loaded onto porous silica.

IV. General Applications for the Composites.

A person of ordinary skill in the art will appreciate that disclosed embodiments of the present invention can be used to make composites and devices for a variety of applications. In some embodiments the composites can be used to make electronic components. Examples of electronic components include, but are not limited to, capacitors and supercapacitors, such as electrical double layer capacitors (EDLCs). FIG. 7 schematically shows the components and configuration of one embodiment of an EDLC device 700. EDLCs operate on an electro- adsorption mechanism where electron and electron holes reside very near to the surface of two polarized carbon electrodes 710 and 720. The electrodes are made from TiN/CMK-3 composites made according to disclosed embodiments of the present invention. The electrodes are placed on either side of a separator 730, which is soaked with an electrolyte. Solvated ions in electrolyte are electrostatically adsorbed to the electrodes, forming two EDLs connected in series by the electrolyte. The EDLC is charged from a voltage source 740, and the stored charge can be accessed through the current collectors 750 and 760, which are metallic substrates that provide electrical contact between electrode materials and external circuit.

In some other embodiments the composites can be used to make batteries. In particular embodiments, porous carbon/Ti0₂ composites or porous carbon/Sn0₂ composites, made according to disclosed embodiments of the present invention, can be used as an electrode material for Li-ion or Na-ion batteries. FIG. 8 schematically shows the components and configuration of one embodiment of a battery device 800. Electrodes 810 and 820 comprise a porous carbon/metal oxide composite that can be made according to disclosed embodiments in which the metal oxide is substantially located within the pores. The electrodes are separated by a separator 830 soaked with a lithium or sodium electrolyte. Electrical contacts 840 and 850, attached to electrodes 810 and 820 respectively, provide electrical energy to an external component to be powered by the battery.

In another embodiment porous carbon/Ti0₂, made according to disclosed embodiments of the present invention, can be used as a support for fuel cell catalysts. FIG. 9 schematically shows the components and configuration of one embodiment of a fuel cell device 900.

Hydrogen gas enters through inlet 910 and oxygen enters from inlet 950. The gases bubble through electrodes comprising a porous carbon/Ti0₂ composite 920 and 940 into the electrolyte 930, which is typically a concentrated potassium hydroxide or concentrated sodium hydroxide solution. Electrical contacts 960 and 970 provide electrical energy to an external devise to be powered by the fuel cell. Unused hydrogen gas leaves the cell through an outlet 980 and a mixture of unreacted oxygen and water, the product of the reaction between hydrogen and oxygen, leaves the cell through outlet 990.

In other embodiments activated carbon/Ti0₂ composites, made according to the present invention, can be used as a hydrogen storage material. T1O2 catalyzes dissociation of molecular hydrogen and promotes chemisorption of hydrogen by the activated carbon/Ti0₂ composite. The composites can be used as catalysts. In a particular embodiment, activated carbon/Ti0₂ composites, made according to the present invention, can be used as a photo catalyst for water treatment. Activated carbon adsorbs contaminating organic molecules in the water. Ti0₂ then acts as a photocatalyst, catalyzing the degradation of the organic contaminants by ultraviolet radiation, such as from the sun. In other embodiments porous alumina can be used as the substrate for catalysts for organic synthesis.

The composites can be used in sensors. In a particular embodiment porous silica/SnC composites, made according to the present invention, can be used as a functioning material for gas sensors. FIG. 10 schematically shows the components and configuration of one embodiment of a sensor 1000. A porous silica substrate 1010 has Sn0₂ substantially coating the inside of the pores 1020. When the metal oxide reacts with a toxic gas, such as carbon monoxide, electrical signals are transmitted along the lead 1030 to a detector 1040.

One exemplary application is in EDLCs, which play an important role for energy storage technologies. Compared to batteries, EDLCs are more competitive in many applications that need high power, long cycle life and high round- trip efficiency. The capacitance of a device is linearly proportional to the surface area of the electrode materials. In commercial devices, activated carbons with very large surface areas ranging from about 2,000 to about 3,000 m²/g are used as electrodes due to the scalable manufacturing, reasonable cost, and high capacitance. Significant advancement in capacitance has been made very recently on activated carbons.

However, the amorphous characteristics of activated carbon limit its conductivity and hence the power performance of EDLCs. A metallic coating, a few nanometers thick, onto the surfaces of the porous carbon electrodes would further improve the power performance of EDLCs, making them comparable to electrolytic capacitors. Cost-effective metals are often not

electrochemically compatible with EDLCs, which prohibits their usage. A coating made of refractory conducting ceramics, particularly TiN, is favorable because of its properties, including high hardness, corrosion resistance, and high conductivity (5xl0⁴ S/cm), two orders more conductive than graphene. TiN nanostructures including nanoporous structures, nanoparticles, CNT-doped TiN nanocrystals and nanowires grown on carbon-fiber paper have been investigated as electrode materials for supercapacitors. Yet, the high density of TiN of 5.4 g/cc makes it too heavy as an electrode by itself for practical supercapacitors. Some particular embodiments of the disclosed method produced a layer of TiN on porous carbon substrates. The carbon framework provided a light backbone for the highly conductive TiN layer which helped enhance the kinetics in EDLs. CMK-TiN samples and CMK-3 were investigated as electrode materials for supercapacitors. Results from cyclic voltammetry clearly suggested that the TiN/CMK-3 composite helped improve the kinetic response of the electrode.

Another exemplary application is in electrodes for batteries such as lithium-ion and sodium-ion batteries. Intense efforts have been devoted to exploring new electrode materials for Li-ion batteries (LIBs) in pursuit of higher energy density and better cycling stability.

Compared to conventional graphitic anodes, alloying anodes have attracted much attention due to their high theoretical capacities. The primary challenge for alloying anodes is the very large volumetric changes during the Li alloying/dealloying processes, which cause electrode pulverization and rapid capacity reduction. Among the alloying anodes, Sn0₂ is an attractive option due to its high theoretical capacity. During discharge, Sn0₂ anodes operate though the following reactions [Eqs. (1) and (2)].

SnO₂ + 4 Li⁺ + 4 e^" »~ Sn + 2 Li₂O (1)

Sn + x Li⁺ + x e^{" Li}xSn (x <= 4.4) (2)

The theoretical capacity is 782 mAh g^"1 under the assumption that only the

alloying/dealloying reactions are reversible. The value is 1494 mAh g^"1 if the conversion between Sn and Sn0₂ is electrochemically reversible. The hurdle for Sn0₂ as an anode has been its rapid capacity fading associated with the large volumetric expansion of Sn (up to 259%) during Li⁺ insertion. To address this problem one possible approach is to down-size the Sn0₂ crystallites to minimize the strain during the alloying/dealloying reactions. Various Sn0₂ nanostructures, such as nanotubes, nanosheets, hollow spheres, microboxes, and mesoporous Sn0₂ have been investigated. Studies on the correlation between Sn0₂ crystallite size and cycling stability revealed that Sn0₂ nanoparticles of about 3 nm in diameter can exhibit a high capacity with improved cycling. However, maintaining a stable capacity for these single- component Sn0₂ nanostructures has been very challenging.

CMK-Sn0₂ samples were investigated as electrode materials for LIBs. Results from galvanostatic cycling clearly suggested that an anode comprising a CMK-Sn0₂ composite exhibited superior stable cycling performance compared to other anode materials, and retained a high capacity after 300 cycles.

Sodium-ion batteries (SIBs) represent one of the most promising alternatives to LIBs for energy storage, due to sodium's low cost, abundance, and distribution all over the world. As an emerging technology, SIBs are limited by a lack of high-performance electrodes. Not surprisingly, many cathode candidates for SIBs are borrowed from LIBs, such as layered transition metal oxides, metal phosphates, and Prussian blue analogues. Among them, V2O5, as a promising material, has exhibited great performance in LIBs. Early studies showed that V2O5 aerogels can perform reversible sodium insertion/deinsertion. Recently, it has been reported that bilayered V2O5 materials exhibit a high capacity, above 170 mAh/g, as well as stable cycling performance in SIBs. Compared to the bilayered V2O5, bulk orthorhombic V2O5 exhibits a much lower capacity, 60 mAh/g. For orthorhombic V2O5, the major hurdle is the slow sodium- ion diffusion in the compact crystal structure.

Porous nanocarbon-orthorhombic V2O5 composites were investigated as electrode materials for SIBs. Results clearly suggested the hydrolysis deposition method disclosed herein allowed control of the loading levels of V2O5 in nanoporous carbon, and revealed the

electrochemical properties of orthorhombic V2O5 nanoparticles as a cathode in SIBs.

V. Examples

The following examples are provided to illustrate certain features of working

embodiments. A person of ordinary skill in the art will appreciate that the scope of the disclosed embodiments is not limited to the features exemplified by these working embodiments.

Preparation of CMK-3 particles

CMK-3 has a long-range ordered P6mm hexagonal structure with linear channels going throughout the particles, which facilitates an investigation of the extremely thin deposition by microscopy techniques. CMK-3 also has a well-explored uniform pore size, large specific pore volume and high surface area. This was helpful for characterizing the level of T1O2 loading.

CMK-3 was prepared by a hard template method using mesoporous silica, SBA-15. For the synthesis of SBA-15 with a controlled morphology 2 grams of Pluronic P123

(HO(CH2CH20)2o(CH2CH(CH3)0)7o(CH2CH₂0)2oH) was dissolved in 60 milliliters of 2 M HC1 at 38 °C. Tetraethylorthosilicate (TEOS, 4.2 grams) was added into the above solution with vigorous stirring. The mixture was stirred for just 6 minutes then remained quiescent for 24 hours at 38 °C. The mixture was subsequently heated at 100 °C for another 24 hours in an autoclave. The as-synthesized SBA-15 with short-rod morphology was collected by filtration, dried and calcined in air at 550 °C. A nanocasting method was utilized to fabricate CMK-3 using SBA-15 as a hard template. Sucrose (1.25 grams) was dissolved in 5.0 milliliters of water containing 0.14 grams of H2SO4. Surfactant-free SBA-15 (1.0 grams) was then dispersed into the above solution and the mixture was sonicated for 1 hour, heated at 100 °C for 12 hours and then at 160 °C for another 12 hours. The impregnation process was repeated once with another 5.0 milliliters of an aqueous solution containing 0.8 grams of sucrose and 0.09 grams of H2SO4. The composite was completely carbonized at 900 °C for 5 hours in an argon atmosphere. To remove the SBA-15 silica template, the composite was stirred in a 5 % HF solution at room temperature for 4 hours. Alternatively, sodium hydroxide solutions can be used to dissolve the silica.

Preparation of activated carbon

To prepare activated carbon, about 2 grams of commercially available carbon fibers (Osaka Gas company, Ltd.) were placed into a horizontal cylindrical furnace. The furnace was purged with CO2 for 30 minutes at room temperature, and then the temperature was increased at 5 °C/minute up to the desired activation temperature of 910 °C. Once the activation temperature was reached the material was held at that temperature for a desired period of activation, typically from about 10 hours to about 30 hours, more preferably from about 15 hours to about 20 hours, under a flow of CO2 with a flow rate of about 100 milliliter s/minute. At the end of the activation period, the sample was cooled down to room temperature under CO2.

Preparation of C-APS by ammonium persulfate treatment

0.3 grams of CMK-3 was added to 30 milliliters of a freshly-prepared aqueous solution of ammonium persulfate ((NH₄)2S208) (1.0 M) and H₂S0₄ (2.0 M). The mixture was stirred at 60 °C for 6 hours. Then the oxidized CMK-3 (C-APS) was filtered, rinsed with deionized water and dried overnight in an oven at 80 °C. The number of carboxylic groups on C-APS was measured by Boehm titration to be 1.96 mmol/g. Boehm titration procedure

The amount of surface functional groups was determined using Boehm titration. In Boehm titration, the following assumptions were made to distinguish between the carbon- oxygen functionalities based on their acidity: NaOH is the strongest base and it neutralizes all Bronsted acids, including phenols, lactonic and carboxylic groups, while NaC0₃ neutralizes carboxylic and lactonic groups and NaHC0₃ neutralizes only carboxylic acid groups. Briefly, 0.2 grams of C-APS was dispersed in 20 milliliters of NaHC0₃ (0.05 M) solution and the mixture was stirred for 48 hours. The solution was then allowed to remain quiescent for 24 hours. A 5 milliliter aliquot was neutralized with 0.05 M cone. HC1 and then back titrated against a standardized NaOH solution using phenolphthalein as indicator.

Calculation of C- APS water-loading

To generate a saturated reference, C-APS was exposed to excessive water vapor for one hour and the adsorbed water mass in C-APS was measured by thermo gravimetric analysis (TGA) (-16 wt%, see FIG. 11). FIG. 12 illustrates a possible device 1200 for exposing the substrate to a specific amount of water vapor in a laboratory setting. A certain amount of degassed C-APS 1240 was placed into a dry plastic syringe 1210. A certain volume of water vapor/air mixture was removed from a bottle 1220 containing water 1230 at 80 °C into the syringe containing C-APS. The volume was calculated according to Dalton's law using the tabulated vapor pressures of water. The syringe was agitated for 10 minutes at 80 °C in order to facilitate adsorption of water onto C-APS. By exposing C-APS to volumes of water vapor corresponding to 133%, 100%, 67%, and 33% of the saturated reference, 7.5 wt%, 7 wt%, 6 wt%, and 4 wt% respectively were loaded, as revealed by the TGA curves in FIG. 11. A TGA curve of the degassed C-APS served as the baseline for water loading. Note that the water adsorption does not increase proportionally with the volumes of water vapor available for the carbon to adsorb. These results clearly demonstrate a good degree of controllability of water loading in C-APS, which provides for controlled hydrolysis of T1O2 in C-APS.

Preparation of T1O2/CMK-3 composites

C-APS was degassed under nitrogen for 12 hours at 250 °C, to eliminate any adsorbed water, and then soaked for one hour in a 1,3-dioxolane solution of TTIP (10 % volume) under a nitrogen atmosphere in a glovebox. The composite was filtered before thermal gravimetric analysis (TGA) was conducted in air to measure the loading of T1O2. CMK-3 is nearly ashless, as shown in its TGA in air (FIG. 13: CMK-3). When dry CMK-3-APS was soaked in the TTIP solution, forming composite C-T1O2-O, the TGA indicated about 6.2 wt% of T1O2 formed (FIG. 13: C-T1O2-O). T1O2 formation was probably due to residual water from 1,3-dioxolane and TTIP in spite of both liquids being dried with molecular sieve (pore size: 4A) before use. b) C-Ti0₂-33 and C-Ti0₂-67

Samples of C-APS were prepared with about 33 % and 67 % of water loading as described above. The resulting Ti0₂-composite samples were referred to as C-Ti0₂-33 and C- Ti0₂-67. There were 9.1 wt and 11.3 wt of T1O2 loaded onto CMK-3, respectively, as determined by TGA (FIG. 13: C-Ti0₂-33 and C-Ti0₂-67). The increase of T1O2 percentage in the composites confirmed the precision of water loadings into the structure.

A sample of C-APS was taken in a small vial and kept in a holder. This holder was then kept in a big bottle of water heated at 80 °C. The sample was allowed to equilibrate in a water vapor/air mixture for 1 hour. This saturated sample was then soaked for one hour in a 1,3- dioxolane solution of TTIP (10 % volume) under a nitrogen atmosphere in a glovebox. The composite was filtered before TGA was conducted in air to measure the loading of T1O2. T1O2 loading reaches to 12.6 wt with the maximum loading of water, referred to as C-T1O2-IOO

FIG. 14 shows the N₂ adsorption/desorption isotherms of CMK-3, C-APS and C-T1O2- 100. The isotherms of C-T1O2-O, C-Ti0₂-33, and C-Ti0₂-67 were similar to that of C-T1O2-IOO (FIG. 15: isotherms of C-Ti0₂-67, C-Ti0₂-33 and C-T1O2-O are moved upward by 100, 200, and 300 cm³/g STP, respectively, for a better comparison of the curves). Table 1 summarizes the physical characteristics of the above samples. It was evident that the oxidation treatment slightly damaged the structure of CMK-3, as revealed by lowered specific surface area and pore volume of C-APS.

The specific surface area and pore volume of samples were further decreased in C-T1O2 samples. Most importantly, the pore volume and surface area of the final composites were inversely proportional to water loading, which clearly showed the controllability of the method. The mesoporous pore size did not change, which indicated that the hydrolysis mainly occurred in micropores. This was evident by the decrease in microporous surface area, as shown in Table 1. The pore size distributions (PSD) of the samples are shown in FIG. 16. The PSD peaks of the samples were very close despite different specific surface areas and pore sizes and varying amounts of deposition. Table 1

Physical Characteristics of Samples

The wide-angle X-ray diffraction (XRD) pattern of C-T1O2-IOO did not exhibit any crystalline features (FIG. 17). Hydrolysis-formed bulk T1O2 was also prepared by exposing TIPP to moisture in a Petri dish. The bulk T1O2 powder exhibited well-resolved peaks in an XRD pattern (FIG. 18). Comparing the above, both XRD patterns suggested that the lack of T1O2 crystalline features in the XRD pattern of C-T1O2-IOO may be due to the extremely small coherence lengths of T1O2 clusters.

Scanning electron microscopy (SEM) of C-T1O2-IOO showed the typical rod-like morphology of CMK-3 (FIG. 19). The clean surface morphology of carbon was free of any T1O2 phase, indicating a complete encapsulation of T1O2 inside the CMK-3 framework. The uniformity of the T1O2 deposition was revealed by high-angle angular dark field scanning transmission electron microscopy (HAADF-STEM) (FIG. 20) equipped with an energy dispersive X-ray (EDX) attachment for elemental mapping (FIG. 21 and 22). It is well-known that HAADF-STEM is sensitive to the contrast of electron densities from different elements, which helps estimate the particle size of deposited T1O2 in the carbon matrix. The absence of T1O2 nanoparticles demonstrates that the deposited T1O2 particles are, in fact, less than 1.0 nm, the resolution limit of the instrument in its STEM mode. As shown in FIG. 22, the titanium distribution was very uniform in a composite of C-TiC"2-67. This confirmed that Ti0₂ can be loaded into the pores of porous carbon by the disclosed method. Preparation of T1O2I activated carbon composites

A sample of activated carbon was taken in a small vial and kept in a holder. This holder was then kept in a big bottle of water heated at 80 °C. The sample was allowed to equilibrate in water vapor/air for 1 hour. The sample was then soaked for one hour in a 1,3-dioxolane solution of TTIP (5 % volume) under a nitrogen atmosphere in a glovebox. The composite was filtered in the glovebox and dried in an oven at 80 °C. This carbon exhibits a high surface area of 2715 m²/g and an average pore size around 2.3 nm (FIGS. 23 and 24). A representative SEM image with the corresponding Ti EDX mapping reveals the homogeneous deposition of Ti0₂ inside the activated carbon as well (FIG. 25). To the best of the applicants' knowledge, this is the first time that this type carbon can be coated by an oxide with such a high degree of uniformity.

Preparation of TiN/CMK-3 composites

For the preparation of TiN/CMK-3 composites, the previously prepared Ti0₂-composite with porous carbon was nitrided in an NH3 atmosphere with a flow rate of 300 cc/minute at 850 °C for 6 hours. The composition of the titanium compounds in the modified layer material after nitridation was identified using X-ray diffraction spectroscopy (XRD), X-ray photon

spectroscopy (XPS) and TEM. The XRD pattern of CMK-TiN corresponded to a single cubic phase of TiN (JCPDS 38-1420), as shown in FIG. 26. The XRD pattern indicated the absence of crystalline Ti0₂ phases. The broadness of the peaks revealed the tiny sizes of TiN

nanocrystallites inside the carbon framework. The domain size of TiN was estimated by the Scherrer Equation to be 4.5 nm, slightly larger than the 3.5 nm pore size of CMK-3. A certain degree of structure damage of CMK-3 due to the growth of TiN was expected. XPS was used to further investigate the oxidation states of titanium, as shown in FIG. 27. The peak of Ti 2p₃/₂ from CMK-TiN was de-convoluted into three components, at 456.3 eV (18%) for TiN, 457.6X eV (25%) for TiO_xN_y and 458.8 eV (57%) for T1O2. The existence of T1O2 component in the sample was expected due to the inevitable exposure of the TiN nanocrystallites to air, which are more reactive than their bulk counterparts. A strong T1O2 portion in XPS and an absence of XRD peaks of T1O2 phases indicated that it was the amorphous T1O2 that was formed on the surface of TiN or TiO_xN_y during the sample handling and transfer. This is a known issue and can be resolved, if needed, by excluding oxygen. The TEM image (FIG. 28) combined with the corresponding EDX mapping (FIGS. 29-31) confirmed the well-maintained distribution homogeneity of titanium phases after the nitridation. The N₂ sorption isotherm of C-Ti-N is shown in FIG. 32 and its characteristics are listed in Table 1. All the characteristics of C-Ti-N were larger than its C-T1O2-IOO counterpart. The pore size grew from 3.9 nm to 4.3 nm. This was most likely caused by the formation process of TiN particles from Ti0₂ particles. By forming larger particles, the micropores originally occupied by Ti0₂ are emptied after nitridation.

Preparation of SnO^porous silica composite

SBA-15 with a high surface area (830 m²/g) and uniform mesopores was used as the mesoporous silica. A sample of SBA-15 was taken in a small vial and kept in a holder. This holder was then kept in a big bottle of water heated at 80 °C. The sample was allowed to equilibrate in water vapor/air for 1 hour. The water-loaded silica was dispersed in a solution of Tin (IV) tert-butoxide in 1,3 dioxane (10 vol ) for 1 hour in a glovebox. The solution was then filtered in the glovebox and dried in an oven at 80 °C.

Preparation of Sn02/CMK-3 composite

A sample of C-APS was taken in a small vial and kept in a holder. This holder was then kept in a big bottle of water heated at 80 °C. The sample was allowed to equilibrate in a water vapor/air mixture for 1 hour. The water-loaded carbon C-APS was dispersed in a solution of tin(IV) tert-butoxide in 1,3 dioxane (10 vol ) for 1 hour in a glovebox. It was then filtered in the glovebox and dried in an oven at 80 °C. The organic groups were expelled by heating the sample in a muffle furnace at 300 °C. This procedure was repeated from the water-loading step for a second and third loading of Sn0₂/CMK-3 composite.

Electrochemical measurements

A two-electrode cell configuration, as illustrated by device 700 (FIG. 7), was used to measure the electrochemical performance of samples. Electrodes 710 and 720 were composed of 90 wt active mass and 10 wt poly(vinylidene fluoride) binder. The materials were slurry- cast from a cyclopentanone suspension onto a carbon-fiber paper current collector (Model:

2050A). The electrodes were dried at 120 °C under vacuum for 12 hours and then cut into 10 mm disks. The active mass loadings were ~1 mg/cm². Then, two identical (by weight and size) electrodes were assembled in coin-type cells that used polypropylene films as the separator 730 and a CH3CN solution of 1.0 M ammonium tetrafluoroborate (NH4BF4) as the electrolyte.

Cyclic voltammetry (CV), galvanostatic charge/discharge, and Electrochemical Impedance Spectroscopy (EIS) were carried out on a VMP-3 multi-channel workstation at room temperature.

Cyclic voltammetry profiles were collected in a voltage window from 0 to 2.5 volts and at different scanning rates of 50, 100, 200, and 500 mV/s. As FIG. 33 shows, CV curves of C- Ti-N maintained the rectangular shape very well even at 500 mV/s and area enclosed in CV did not decrease much upon higher scanning rates, which indicates high surface conductivity. In sharp contrast, the CV curves of CMK-3 were flattened dramatically upon higher scanning rates (FIG. 34), and similar CV results were observed for a reference sample of C-APS subjected to nitridation conditions to investigate possible N-doping effects (FIG. 35). The CV results suggest that the TiN coating significantly helped improve the kinetic response of the electrode although some T1O2 and TiON phases were formed on the surface of TiN nanocrystallites, and the rate improvement was certainly not from the N-doping effect.

Galvanostatic charge/discharge profiles were collected in a voltage window from 0 to 2.5 Volts. FIG. 36 shows the galvanostatic charge and discharge profiles of CMK-3 and C-Ti-N at different current rates. The two materials showed similar capacitance values but very much different equivalent series resistance (ESR). For example, at a current rate of 0.5 A/g, the ESR was 21.9 Ω and 118 Ω, respectively, for C-Ti-N and CMK-3.

EIS was carried out with the potential amplitude of 10 mV at the frequency range of 200 kHz to 10 mHz. The voltage drop at the beginning of discharge {Vdrop) was used to estimate the equivalent series resistance, RESR, at a constant current of / with the formula of RESR = Vdmp (2I). FIG. 37 shows the Nyquist plots obtained at the frequency range from 200 kHz to 10 mHz. C- Ti-N exhibited a vastly smaller semicircle than CMK-3 did (1 Ω vs. 95 Ω), which indicated a much lower resistance at the interface between the electrode and electrolyte for C-Ti-N than CMK-3. The enlarged view at the high-frequency region for the C-Ti-N is shown in the inset.

Investigation of CMK-Sn02 as an anode material for lithium-ion batteries

I. Preparation

Step I: carbon surface functionalization: CMK-3 was treated with a mild oxidant, namely, an acidic aqueous solution of 1.0 M (NH₄)₂S₂08 (APS) and 2 M H₂S0₄ to form C-APS. 0.3 grams CMK-3 was added to 30 milliliters of a freshly prepared acidic 1.0 M solution of APS. The mixture was stirred and heated at 60 °C for 6 hours. Then the solids were collected by filtration, washed with deionized water, and dried overnight in an oven at 80 °C. Step II: water loading: Degassed C-APS (50 milligrams) was allowed to equilibrate in an ambient water vapor/air mixture with 100% relative pressure at 80 °C for 1 hour. The sample was poured into a Petri dish and kept under a non-moving atmosphere for a further 1 hour.

Step III: hydrolytic deposition of Sn0₂: Water-loaded C-APS was dispersed and soaked for 1 hour in a dilute solution of tin(rV) butoxide in 1,3-dioxolane (10 vol%). The solid product was collected by filtration in a glovebox. Samples were then heated at 300 °C under air for 5 hours to form Sn0₂. The material collected was designated 1-Sn-CMK. To increase the Sn0₂ loading, the sequential water-loading and hydrolysis process was repeated, and the resulting products were denoted 2-Sn-CMK and 3-Sn-CMK.

FIG. 38 shows the X-ray diffraction patterns of materials prepared in one, two, and three hydrolysis deposition cycles. They exhibited broad peaks at 25.7, 33.2, 36.9, 50.8, and 62.3°, which can be indexed to (110), (101), (200), (211), and (301) planes of tetragonal Sn0₂ (JCPDS 41-1445), respectively. These broad peaks suggested the existence of nanosized crystal domains of Sn0₂. With increasing number of deposition cycles, the increased intensity of the XRD peaks and slightly sharper peaks indicated a larger Sn0₂ due to layer-by-layer growth. The domain size estimated by means of the Scherrer equation increased from about 2 to about 4 nm from the first to the third cycle, and the corresponding full width at half-maximum of the (110) peak in the XRD patterns decreased from 7.3 to 3.7°. Table 2 summarizes the results of surface-area and porosity measurements. The nitrogen sorption isotherms and pore size distributions of the composites are presented in FIGS. 39 and 40, respectively. Treatment of CMK-3 with APS led to a slight decrease in specific surface area and pore volume. As observed in the low-angle XRD patterns (FIG. 41), the long-range order was well maintained. After one deposition cycle, the surface area and pore volume decreased from 1238 m²g^_1 and 1.37 cm³g^_1 for C-APS to 968 m²g^_1 and 0.962 cm³g^_1, respectively. These parameters further decreased after the second and third cycles.

Table 2.

Physical characteristics of samples

The mass loading of the deposited Sn0₂ in the composites was measured by

thermogravimetric analysis (TGA) in air (FIG. 42). The small weight loss below 100 °C was attributed to water evaporation. Almost no weight loss was observed between 100 and 400 °C, demonstrating the thermal stability of the composites in air up to 400 °C. The weight loss from 400 °C to 750 °C was due to carbon combustion. With increasing number of hydrolysis deposition cycles, the percentage of Sn0₂ in the composites increased from 21 to 66 wt . When normalized for per gram of carbon, 0.26, 0.43, and 1.30 g of Sn0₂ was deposited after the first, second, and third deposition cycles, respectively. The dramatic increase in Sn0₂ deposition in later cycles was attributed to the enhanced hydrophilic properties of the composite surface after formation of an Sn0₂ layer.

The morphology of the Sn0₂-CMK composites was probed by SEM and TEM (FIGS. 43-49). In the SEM images, the absence of Sn0₂ nanoparticles indicated successful loading of Sn0₂ into the porous structure of CMK-3 (FIGS. 43 and 44). In the TEM image, fine dispersion of Sn0₂ nanoparticles inside the channels of CMK-3 was evident (FIG. 45). High-resolution TEM analysis of 2-Sn-CMK showed the presence of crystalline Sn0₂ particles embedded in the carbon matrix (FIGS. 46 and 47). FIG. 50 shows the high-angle annular dark- field (HAADF) scanning TEM (HAADF-STEM) image and the corresponding carbon, oxygen, and tin energy- dispersive X-ray spectroscopy (EDX) mapping images. The HAADF-STEM and elemental- mapping data revealed that the Sn0₂ nanoparticles in the composite were of uniform size and were homogenously dispersed within the mesoporous channels. EDX also indicated 43 wt Sn0₂ in the composite, which corroborated the TGA measurements (FIG. 51). //. Assembly and testing of lithium-ion batteries

The electrochemical characteristics of the Sn0₂/CMK composites were investigated in coin-style half-cells with lithium-metal foil as the counter electrode and 1.0 M LiPF₆ in ethylene carbonate/dimethyl carbonate (1/1 w/w) as electrolyte. Slurries were prepared by mixing 80 wt% active material (Sn0₂/CMK composite), 10 wt% carbon black (Super-P), and 10 wt% poly(vinylidene fluoride). The mixture was suspended in N-methyl-2-pyrrolidinone before casting onto a copper foil current collector by the doctor-blade method. The typical active-mass loading was around 1-1.3 mg cm^"2. Lithium-foil anodes were polished in an argon environment before use as the counter/reference electrode. The electrodes were assembled into a coin cell in a glovebox. Glass-fiber membrane was used as the separator, and 1 M LiPF₆ in ethylene carbonate/dimethyl carbonate was used as the electrolyte. Galvanostatic cycling was conducted on an Arbin BT2000 system, and CVs were collected on a VMP-3 multichannel workstation at a scanning rate of 0.5 mVs^"1 at room temperature.

The specific capacity was calculated on the basis of the total mass of a composite electrode. FIG. 52 shows the first galvanostatic discharge/charge profiles of the Sn0₂/CMK composites cycled between 0.01 and 2.0 V at a current density of 200 mAg^"1. For 2-Sn-CMK, the first discharge (lithiation) delivered a capacity of 2675 mAhg^"1 and a charge (delithiation) capacity of 1054 mAhg^"1 with 39% reversible capacity, while 1-Sn-CMK and 3-Sn-CMK had 33 and 46% reversible capacity, respectively. Comparison with the first discharge/charge profiles of C-APS with a 22% reversible capacity revealed that a major portion of the irreversible capacity in the Sn0₂/carbon composites was contributed by C-APS forming a solid electrolyte interphase (SEI) layer (FIG. 53). The reversibility increased with increasing Sn0₂ loading. An SEI layer can be formed with the Sn0₂ active mass as well, which would contribute to the irreversible capacity.

The first delithiation capacity of 2-Sn-CMK of 1054 mAh g^"1 was much higher than those of 630 mAhg^"1 for 1-Sn-CMK and 800 mAh g^"1 for 3-Sn-CMK. It is generally accepted that for Sn0₂ electrodes only the Li-Sn alloying process, which involves 4.4 Li⁺ per Sn atom, is reversible, and the theoretical capacity of Sn0₂ electrodes is 782 mAhg^"1. Based on the initial delithiation capacity of C-APS (603 mAh g^"1, FIG. 53), the capacity contribution of Sn0₂ to the first delithiation capacity of 2-Sn-CMK was 698 mAh/gram of composite. The specific capacity of Sn0₂ was calculated to be 1703 mAh which was even higher than the theoretical capacity of Sn0₂ if 8.4 Li⁺ is reversibly inserted/deinserted (1494 mAh g^"1). Without being bound to a particular theory, this surprising result may be due to the synergetic effect between porous carbon and Sn0₂ nanoparticles, which might alter and enhance the electrochemical properties of the carbon matrix after Sn0₂ deposition. The actual contribution from carbon in the composite may be underestimated. On the other hand, this result indicated that the conversion reaction of Sn0₂ to Sn [see Eq. (1), above] in 2-Sn-CMK is reversible to a large extent. The nanocrystalline nature of Sn0₂ and the uniform dispersion of Sn0₂ phases in a high-surface-area matrix should greatly enhance the electrochemical reactivity of Sn0₂. Similar observations have been reported for approximately 2-6 nm Sn0₂ nanoparticles embedded in a graphene matrix, whereby the contribution from Sn0₂ in the graphene composite reaches 94% of its theoretical capacity if both alloying and conversion reactions are considered to be reversible.

The first 50 charge/discharge cycles were tested at a current density of 200 mA g^"1 for

Sn0₂/C composites after one, two, and three hydrolysis deposition cycles, and the corresponding delithiation capacities are shown in FIG. 54. The 2-Sn-CMK composite showed a higher capacity of 684 mAh g^"1 after 50 cycles. Surprisingly, 3-Sn-CMK showed a lower capacity of 455 mAh and lower capacity retention (57%) after 50 cycles, compared to 2-Sn-CMK (65%). Without being bound to a particular theory, this might be due, in part, to the loading of Sn0₂ and the available pore volume in the composites for volumetric expansion of Sn. 2-Sn-CMK contains 0.695 g of Sn0₂ per gram of carbon. According to Equation (1), Sn0₂ undergoes a conversion reaction to form Sn and Li₂0, and Sn further alloys with Li⁺ to form Li₄.4Sn. The volume occupied by Li₄.₄Sn and Li₂0 was estimated by assuming that the nanophases of Sn and Li₂0 had the same density as their bulk counterparts. Thus, in 2-Sn-CMK, the Sn0₂-derived mass occupied a volume of 0.402 cm³ per gram of carbon after the conversion and alloying reactions (assuming 259% volumetric expansion of Sn). Considering that C-APS has a specific pore volume of 1.37 cm³g^_1, there was sufficient pore volume in 2-Sn-CMK after lithiation of Sn. Similarly, in 3-Sn-CMK, the Sn0₂-derived mass after complete conversion and lithiation reactions occupies a volume of 1.17 cm³ per gram of carbon. This expansion theoretically filled 85% of the available pore volume in C-APS, which may have caused a serious strain on alloying and limit electrolyte imbibement. The large capacity difference for composites with different active mass loadings clearly showed the impact of the empty space in the composites on the electrochemical performance of the encapsulated Sn0₂. These results, for the first time, demonstrated the importance of porosity manipulation in nanoporous composites.

Due to its excellent performance, further electrochemical investigations were carried out for 2-Sn-CMK. FIG. 55 shows cyclic voltammograms (CVs) for 2-Sn-CMK at a scanning rate of 0.5 mV/s in the voltage window from 0.01 to 2.0 V. In the cathodic scan, a large irreversible reduction peak was observed at 0.7 V, which is attributed to decomposition of Sn0₂ into Sn and Li₂0, and also to the formation of an SEI on Sn phases and carbon. In the following cycles, a cathodic peak at 0.8 V was still present, indicative of a certain degree of reversible conversion of Sn0₂. A pair of reversible redox peaks at 0.14 and 0.55 V were assigned to lithium alloying and dealloying with tin. These observations were in accordance with previous reports on Sn0₂- based anodes. The largely unchanged peak current intensity, except for the first cycle, indicated good cycling stability of the composite electrode.

Long-term galvanostatic cycling was conducted for 2-Sn-CMK between 0.01 and 2.0 V at a current density of 200 mAg^"1. The charge/discharge profiles of Sn0₂/CMK composite in the 1st, 5th, 25th, and 300th cycles and the cycling profile are shown in FIGS. 56 and 57. The coulombic efficiency increased from 85% in the second cycle to 99.6% in the 300th cycle.

These results demonstrated the promising cycling stability of Sn0₂ supported on mesoporous carbon, and were attributed to the well-managed micro structure of the composites. To further analyze the structural stability, a TEM study was carried out on 2-Sn-CMK after 300

charge/discharge cycles (FIG. 48). From the TEM image, it was evident that the ordered mesoporous structure of CMK-3 was well maintained and there was no severe aggregation of Sn0₂ particles. The corresponding HAADF-STEM and EDX mapping of carbon, tin, and oxygen revealed uniform distribution of Sn0₂ nanoparticles in the carbon matrix after 300 cycles (FIG. 58). The microscopy results corroborated the high capacity, high rate capability, and cycling stability of 2-Sn-CMK. In sharp contrast, after 200 charge/discharge cycles, 3-Sn- CMK showed significant aggregation of Sn0₂ particles (FIG. 59). The results demonstrated the importance of the manipulation of empty pore volume for accommodating the large volumetric expansion/contraction of the tin nanoparticles during the Li insertion/extraction process.

The 2-Sn-CMK composite also exhibited excellent rate performance. The rate capability was investigated at current densities of 100, 200, 500, 1000, and 2000 mAg^"1, and the charge/discharge profiles are presented in FIG. 60. At a current density of 2000 mAg^"1, a high capacity of 320 mAg^"1 was retained. This was higher than the reported rate performance of Sn0₂-CNT, carbon-coated Sn0₂ platelets, and Sn0₂ embedded in mesoporous carbon.

Moreover, even at high current densities, a high coulombic efficiency of 99% was maintained. The superior rate performance was attributed to the high porosity of the composite and the conductive carbon framework. Investigation of orthorhombic V2O5 nanoparticles generated in nanoporous carbon as a cathode material for sodium-ion batteries

I. Preparation

Porous carbon (RFC) was prepared according to the procedure reported previously in the literature using resorcinol and formaldehyde as precursors. Briefly, resorcinol was dissolved and mixed in an aqueous solution of formaldehyde. Then, silica colloidal suspension was added to the mixture under vigorous stirring. After a heating treatment in air at 100 °C followed by pyrolysis under Ar at 900 °C, the product was soaked in an HF aqueous solution to remove the silica. The final product was collected by filtration and dried.

Hydrolysis deposition:

Step I: Water loading. Degassed RFC (50 mg) was allowed to equilibrate in an ambient water vapor/air mixture with 100% relative pressure at 80 °C for one hour. The sample was then poured in to a Petridish and kept in a non-moving atmosphere for one additional hour.

Step II: Hydrolysis deposition of V₂Os. Water-loaded RFC was dispersed and soaked for one hour in a dilute solution of vanadium triisopropoxide solvated in cyclohexane (10 vol%). The solid product was collected by filtration in an inert atmosphere in a glovebox. Samples were then heated at 225 °C in air for 2 hours to eliminate the organic moieties and form the oxide. The above steps were repeated until a desired loading of V2O5 was achieved in the nanoparticles. The obtained composites were denoted as X-V2O5-RFC, where x represents the weight percentage of V2O5 in the composites.

The V2O5 loading was measured by thermogravimetric analysis (TGA) in air. FIG. 61 shows the TGA data for V2O5-RFC composites obtained after the first to fifth hydrolysis deposition cycles. For pure RFC, weight loss due to carbon oxidation occurred from about 500 °C (FIG. 62). For the V2O5-RFC samples, the oxidation onset temperature dramatically dropped from about 500 to about 300 °C, which implied that the coated vanadium oxide catalyzed carbon oxidation. After one hydrolysis deposition cycle, the TGA plot displayed two well-defined weight losses from 300 to 375 °C and from 400 to 500 °C, respectively. Without being bound to a particular theory, the lower-temperature weight loss was attributed to the carbon oxidation catalyzed by vanadium oxide, while the higher temperature weight loss was attributed to the noncatalyzed oxidation of carbon in air. Two-carbon oxidation events indicated that the first hydrolysis deposition cycle did not construct a continuous V2O5 coating on the RFC. However, after the second deposition cycle with 37 wt % of V2O5 loaded in the composite, only one sloping weight loss region was observed. After the third and the fifth cycles, the weight loss at the higher-temperature regions further diminished, which indicated that repeated deposition cycles improved V2O5 coverage on the carbon surface.

//. Characterization

The structure of V2O5 in the nano-composites was characterized by XRD (FIG. 63). For comparison, bulk vanadium oxide formed in a Petri dish by hydrolysis of the same precursor, vanadium triisopropoxide, was also characterized (FIG. 64). As expected, all nanocomposites exhibited much broader peaks due to the small particle sizes, which could still be assigned to the orthorhombic phase of V2O5. The major XRD peak from the V2O5/RFC nanocomposite at about 26° matched the peak (110) displayed by the bulk V2O5 formed by the same hydrolysis method. By comparing this XRD pattern from the nanocomposite to the XRD patterns of bilayered V2O5 in literature, it was clear that the nanoparticles in the composites were not of a bilayered structure. These nanopartides formed by the hydrolysis deposition method appeared

orthorhombic in structure. With more deposition cycles, the domain size of the deposited V2O5 was larger. Estimated by the Scherrer Equation, the domain size increased from 5 to 7 nm from 55-V2O5-RFC (after three deposition cycles) to 7O-V2O5-RFC (after five cycles).

Surface area and porosity characteristics of the composites were analyzed by nitrogen sorption measurements, and the results are summarized in Table 3.

Table 3.

Physical characteristics of RFC and V2O5-RFC samples

As can be seen in Table 3, as V2O5 loading increased, the specific surface area and pore volume of the composite dropped, indicating an effective pore filling inside RFC framework. By loading 55 and 70 wt % of V2O5 in the RF carbon, 6.8 and 13.1% of the available pore volume was filled by V2O5 in the composite.

The chemical compositions of the vanadium oxide coated on carbon was further identified by X-ray photoelectron spectroscopy (XPS) where the V 2p2/3 peak could be de- convoluted into two binding energies: 515.5 and 517.3 eV, corresponding to V and V , respectively (FIG. 65). The V⁴⁺/V⁵⁺ ratio was about 16% for both 55-V₂0₅-RFC and 70-V₂O₅- RFC. The results suggested the presence of oxygen vacancies in the V2O5 lattice. Previous studies on V2O5 for LIBs implied that such vacancies may enhance the electrochemical performance of V2O5 samples.

Additionally, the samples were assessed to determine whether the V2O5 had been completely encapsulated into the pores of nanoporous carbon. The surface morphology of RFC and 55-V2O5-RFC were scanned by scanning electron microscopy (SEM). However, no differences could be identified (FIGS. 66-68). With the higher loading of 70 wt % V2O5, however, needle-like particles of V2O5 appeared on the surface of carbon (FIG. 69), indicating that there may be a practical limit for exclusively depositing V2O5 inside the pores, even though the specific pore volume of RFC theoretically allowed 94 wt % V2O5 to be loaded.

The distribution of V2O5 was then investigated in 55-V₂0s-RFC by transmission electron microscopy (TEM). As FIG. 70 shows, 55-V₂0s-RFC exhibited a highly porous structure.

Furthermore, based on the high-angle angular dark-field scanning TEM (HAADF-STEM) and the corresponding energy-dispersive X-ray spectroscopy (EDX) elemental mapping, it was evident that the hydrolysis deposition of vanadium oxide was indeed uniformly dispersed inside the porous carbon framework (FIGS. 71-73). Electrochemical performance

Electrode slurries were prepared by mixing 70 wt% active material (V2O5/RFC composite), 20 wt% carbon black (Super-P) and 10 wt% polyvinylidene fluoride (PVDF). The mixture was suspended in N-methyl-2-pyrrolidinone (NMP) before casting onto an aluminium foil current collector by the doctor blade method. The typical active-mass loading was around 1-1.3 mg/cm². Sodium foil anodes were polished in an argon environment before being used as the counter/reference electrode. The electrodes were assembled in a coin cell in a glovebox. Glass-fiber membrane was used as the separator, and 1.0 M NaC10₄ in propylene carbonate (PC) was used as the electrolyte. Galvanostatic cycling was conducted on an Arbin BT2000 system, and cyclic voltammograms (CVs) were collected on a VMP-3 multi-channel workstation at a scanning rate of 0.5 mVs^"1 at room temperature.

Initial electrochemical analyses were conducted at a current density of 40 mA/g in a potential window from 1.5 to 3.8 V versus Na⁺/Na. FIG. 74 shows the charge/discharge profiles for RFC, bulk V2O5, 35-V₂0₅-RFC, 55-V₂0₅-RFC, and 70-V₂O₅-RFC composites. Nearly linear voltage profiles were observed as a function of inserted sodium ions in V2O5-RFC composites. Composites of 35-V2O5-RFC and 55-V2O5-RFC exhibited a discharge capacity of 123 and 183 mAh/g, respectively, while RFC also shows a capacity of 48 mAh/g due to the electrical double layer capacitance. Note that the specific capacity is calculated based on the total mass of the composites. FIG. 75 provides the initial cycling performance of RFC, 55- V2O5-RFC, and 7O-V2O5-RFC. By assuming that RFC in the composites contributed the same capacity as in its pure form to the capacity of the composites, the specific capacity of V2O5 in the 55-V2O5-RFC was estimated to be 276 mAh/g. This was one of the highest capacity values reported for orthorhombic V2O5 in SIBs. Because the cutoff potential of 1.5 V was chosen arbitrarily, it was in fact very difficult to predict the exact oxidation state of vanadium at this potential. If this was a two-electron transfer and all the vanadium was in a 5⁺ oxidation state, the theoretical capacity would be 294 mAh/g. In this case, however, about 16% of the vanadium was in 4⁺ oxidation state, as revealed by the XPS results, resulting in a theoretical capacity for vanadium oxide in the nanocomposite of 247 mAh/g. This was significantly lower than the experimental observation. Therefore, at 1.5 V versus Na⁺/Na the oxidation state of the vanadium was assumed to be a mixture of 4⁺ and 3⁺.

The charge/discharge profile for the third cycle of 55-V2O5-RFC at a current density of 40 mA/g in a potential window between 1.5 V to 3.8 V, is shown in FIG. 76. The 55-V2O5-RFC composite exhibited a columbic efficiency of 93-95%. Surprisingly, 7O-V2O5-RFC, with more V2O5 loading, exhibited a similar capacity to that of 55-V2O5-RFC. This implied that the sodium-ion storage in V2O5 may partially originate from a surface pseudocapacitance and more V2O5 loading did not necessarily increase the surface area of active mass. For comparison, bulk V2O5 electrode was also analyzed under the same conditions, and exhibited a much lower specific capacity of 15 mAh/g (FIG. 77). The poor performance of the bulk orthorhombic V2O5 was consistent with previous reports.

55-V2O5-RFC was selected for further investigation to understand the electrochemical performance of the hydrolysis deposition-enabled composites. Cyclic voltammetry (CV) scans were carried out at a sweeping rate of 0.5 mV/s in the voltage window from 1.5 to 3.8 V versus Na⁺/Na. As FIG. 78 shows, during the first cathodic scan, a broad reduction peak occurred from 3 to 2.25 V, followed by another one from 2.0 to 1.5 V. As revealed by XPS (FIG. 65), vanadium in the composites was mainly in a 5⁺ oxidation state. Thus, the two sequential cathodic peaks may be ascribed to the reduction processes from V⁵⁺ to V⁴⁺ and from V⁴⁺ to V³⁺, respectively. In the first anodic scan, the broad 'hill' was attributed to the desodiation process. Interestingly, in the second and third anodic scans, the oxidation current evolved into two better resolved peaks: one centered at about 2.3 V and the other at about 3.0 V, suggested the sequential extraction of sodium ions from V2O5. 55-V2O5-RFC also exhibited better kinetics than 7O-V2O5-RFC, which was evident by comparing the redox peak separation in CV curves (FIG. 79). This correlated with the smaller particle size of V2O5 in 55-V2O5-RFC.

The rate performance of 55-V2O5-RFC was characterized at different galvanostatic current densities (FIG. 80). Impressively, the composite retained 52% of its capacity when the rate was increased from 40 to 640 mA/g. This was comparable to the high rate performance from the bilayered materials. Interestingly, there was a plateau centered on 2.7 V during the charging process, which became more pronounced at higher current rates. Without being bound to a particular theory, this suggested that the desodiation may occur through different pathways than the sodiation process. FIG. 81 provides the cycling performance of 55-V2O5-RFC over 70 cycles at different current densities, indicating that after 70 cycles a specific capacity of 140 mAh/g was retained at 40 mA/g.

Due to the small particle sizes of V2O5, the composite may exhibit significant pseudocapacitance. V2O5 nanosheets, V20₅/carbon nanotubes, and porous V2O5/C composites have previously been studied for pseudocapacitor applications. To study pseudocapacitance in the V2O5-RFC composites, it was necessary to deconvolute the total current based on two different charge- storage mechanisms: the diffusion-controlled Faradaic reaction and the capacitive behavior, including the surface pseudocapacitance, and the non-Faradaic electrical double layer capacitance. Accordingly, the current response (i) at a fixed potential (V) can be represented by the following equation:

j(V) = kiv + k₂v^1/2

where kiv represents the total capacitive contribution, and k2V^1/2 represents the contribution of diffusion-controlled Faradaic intercalation process. A linear plot of ih as a function of v^1/2 was used to determine the slope (ki) and intercept (k2) (FIG. 82). By

determining the value of k, and k2, the capacitive charge storage at a certain voltage was estimated.

Results showed that the capacitive charge storage was a significant contributor to the total capacity, particularly at high sweeping rates. CV curves of 55-V2O5-RFC at sweeping rates from 0.5 to 20 mV/s between 1.5 to 3.8 V versus Na⁺/Na were used to quantify the capacitive contribution (FIG. 83). FIGS. 84 and 85 show the CV profiles for the calculated capacitive current and the total measured current for 55-V2O5-RFC at a sweep rate of 5 and 0.5 mV/s, respectively, with the estimated capacitive contribution to the total current shown in the shaded regions. At 0.5 mV/s, the capacitive process accounted for only about 20% of the total charge storage, estimated by the enclosed area, while at 5 mV/s, this value increased to about 40%. Furthermore, it was evident that the current from diffusion-controlled redox reactions dominated from 2.5 to 3 V. At the potentials lower or higher, the current response comprised a significant capacitive portion.

In order to further differentiate pseudocapacitive behavior from the double layer capacitance, an electrolyte was used with very large ions that cannot be inserted into V2O5, namely, 1.0 M tetraethylammonium (TEA⁺) tetrafluoroborate (BF₄ ^~) in propylene carbonate. The obtained capacitance was less than 5% of the total charge storage. This meant that the majority of the capacitive charge storage was due to the pseudocapacitance of V2O5 (FIG. 86). The light grey shaded area in FIG. 86 indicates the total capacitive contribution to the total current and the dark grey shaded area shows the contribution from the double-layer capacitance. The high pseudocapacitance was due to the high surface area of coated V2O5 in porous carbon, as evident by the vast difference between CV profiles of 55-V2O5-RFC and bulk V2O5 (FIG. 87).

In order to further understand the electrochemical properties of the materials,

electrochemical impedance spectroscopy (EIS) measurements were carried out on bulk V2O5, RFC, and 55V2O5-RFC at a frequency range of 200 kHz to 10 mHz with a signal amplitude of 10 mV. As FIG. 88 shows, all the Nyquist plots featured two well-defined regions, a depressed semicircle followed by an inclined straight line at the low-frequency region. FIG. 89 inset shows the equivalent circuit that gave rise to a good fit for the EIS spectra of 55-V2O5-RFC and bulk V2O5 (FIG. 90) with the fitting parameters presented in Table 4. In the equivalent circuit, R_s represented the equivalent series resistance (ESR) that includes all Ohmic resistance due to the electrolyte and other parts of the cell. Cdi and Cf were the constant phase elements (CPE), revealing the nonideal capacitance due to the surface roughness, while R_ct and Rf stood for the charge transfer resistance through the electrode/electrolyte interface and the contacts in between V2O5 particles, respectively. The Warburg element, W₀, reflected the solid-state diffusion of Na⁺ ions inside the V2O5 particles. 55-V2O5-RFC exhibited a much smaller Rf (1.5 Ω) than that of bulk V2O5 (24.7 Ω), which was attributed to the smaller particle sizes of V2O5. Table 4.

Electrode resistance obtained from the equivalent circuit fitting of EIS results

In conclusion, for the first time superior electrochemical performance of orthorhombic V2O5 encapsulated in nanoporous carbon was observed. In the V2O5/RFC composites, the loading levels of V2O5 were controlled by adjusting the number of hydrolysis deposition cycles. As a promising cathode material for SIBs, V2O5-RFC composites exhibited a reversible capacity of over 170 mAh/g at 40 mA/g and 92 mAh/g at a very high current density of 640 mA/g. In 55-V2O5-RFC, V2O5 alone exhibited an impressive specific capacity of 276 mAh/g. Modeling revealed the pseudocapacitance behavior of orthorhombic V2O5 nanoparticles that accounted for a significant portion of the total capacity. The excellent electrochemical properties of V2O5- RFC composites were attributed to the small particle sizes, fine dispersion, and controlled loading of V2O5 in nanoporous carbon.

In view of the many possible embodiments to which the principles of the disclosed invention may be applied, it should be recognized that the illustrated embodiments are only preferred examples of the invention and should not be taken as limiting the scope of the invention. Rather, the scope of the invention is defined by the following claims. We therefore claim as our invention all that comes within the scope and spirit of these claims.

Claims

We claim:

1. A method, comprising:

exposing a porous substrate to an amount of water vapor, the substrate having a surface that defines a plurality of pores, each pore being defined by a pore surface that is a portion of the substrate surface, the exposing comprising exposing the surface of the substrate to the amount of water vapor sufficient to locate the water vapor in the pores; and

forming a layer comprising a hydrolysis product that at least partially covers pore surfaces that define at least some of the pores by contacting the surface of the substrate with a non-aqueous solution of at least one hydrolysable compound.

2. The method of claim 1, wherein the amount of water vapor is an amount sufficient to locate the water vapor only in the pores.

3. The method of claim 1, wherein the substrate is exposed to a volume of water that is less than or equal to the pore volume.

4. The method of claim 1, further comprising selecting the amount of water vapor to provide a selected amount of the hydrolysis product.

5. The method of claim 1, further comprising serially applying multiple layers.

6. The method of claim 5, wherein applying multiple layers comprises:

applying a first layer having a first composition; and

applying a second layer having a second composition that is different from the first composition.

7. The method of claim 5, wherein applying multiple layers comprises:

applying a first layer having a first composition; and

applying a second layer having a second composition that is the same as the first composition.

8. The method of claim 1, wherein the substrate surface has an interior portion that defines the pores and an exterior portion that does not define the pores, and at least 50% of the area of the exterior portion is not covered by the layer.

9. The method of claim 1, wherein the layer is located only in the pore surfaces.

10. The method of claim 1, further comprising contacting the substrate with an oxidizing agent prior to exposing the substrate to the water vapor.

11. The method of claim 1, wherein the substrate is selected from porous carbon, porous silica, porous alumina, porous titanium oxide, porous zeolites, porous metal organic frameworks or porous polymeric structures.

12. The method of claim 11, wherein the substrate is porous carbon.

13. The method of claim 11, wherein the substrate is porous silica.

14. The method of claim 12, wherein the porous carbon is CMK-3.

15. The method of claim 12, wherein the porous carbon is activated carbon.

16. The method of one of claims 1-15, wherein the hydrolysable compound is selected to produce a hydrolysis product comprising a metal oxide.

17. The method of claim 16, wherein the metal oxide is selected from T1O2, Si0₂,

AI2O3, Sc₂0₃, V2O3, V2O5, Cr₂0₃, Mn0₂, Ga₂0₃, FeO, Fe₂0₃, Fe₃0₄, CoO, Co₃0₄, NiO, N1O2, CuO, Cu₂0, ZnO, SrO, Y2O3, Zr0₂, CdO, Ag₂0, Rh₂0₃, Nb₂0₃, Nb₂0₅, W2O3, W0₂, WO3, M0O2, M0O3, Ru0₂, Re0₃, Re₂0₇, Ir0₂, PdO, PtO, Sn0₂, Sb₂0₃, Te0₂, Ge0₂, PbO, Pb0₂, Ce₂0₅, Ce0₂ or Eu₂0₃.

18. The method of claim 16, wherein the metal oxide is T1O2.

19. The method of claim 16, wherein the metal oxide is Sn0₂.

20. The method of claim 16, wherein the metal oxide is V2O5.

21. The method of one of claims 1-16, further comprising contacting the layer with ammonia, to form a metal nitride.

22. The method of claim 21, wherein the hydrolysable compound is selected to such that the metal nitride is titanium nitride.

23. The method of one of claims 1-15, further comprising converting the layer comprising a hydrolysis product to a layer comprising a metal carbide.

24. The method of one of claims 1-15, further comprising converting the layer comprising a hydrolysis product to a layer comprising a metal phosphide.

25. The method of one of claims 1-15, further comprising converting the layer comprising a hydrolysis product to a layer comprising a metal oxynitride.

26. The method of claim 10, wherein the oxidizing agent is nitric acid.

27. The method of claim 10, wherein the oxidizing agent is ammonium persulfate.

28. The method of claim 1, wherein the hydrolysable compound comprises a metal alkoxide.

29. The method of claim 1, wherein the hydrolysable compound comprises a metal halide.

30. A method, comprising:

providing a porous substrate, the substrate having a surface that defines a plurality of pores, each pore being defined by a pore surface that is a portion of the substrate surface, the substrate surface also having a portion that does not define the pores;

locating water vapor into the pores; and forming a layer comprising a hydrolysis product that at least partially covers the pore surfaces and leaves at least 50% of the portion that does not define the pores uncovered by contacting the substrate with a non-aqueous solution comprising at least one hydrolysable compound.

31. A method, comprising:

contacting a carbon substrate with an oxidizing agent, the substrate having a surface that defines a plurality of pores, each pore being defined by a pore surface that is a portion of the surface, and an exterior portion of the surface that does not define the pores;

locating water vapor into the pores; and

forming a layer comprising a hydrolysis product that at least partially covers the pore surfaces and leaves at least 50% of the portion that does not define the pores uncovered by contacting the substrate with a non-aqueous solution comprising at least one hydrolysable compound.

32. The method of one of claims 30-31, further comprising contacting the layer with ammonia.

33. A material, comprising:

a substrate, comprising a surface, a plurality of pores, each pore being defined by a pore surface that is a portion of the surface, and an exterior portion that does not define the pores; and one or more layers, each layer independently comprising a layer material, and each layer located such that the pore surface is at least partially covered by one or more layers and at least 50% of the exterior portion is not covered by a layer.

34. The material of claim 33, wherein the exterior portion is not covered by the layer.

35. The material of claim 33, wherein the layer material has a composition different from that of the substrate.

36. The material of claim 33, wherein the substrate is selected from porous carbon, porous silica, porous alumina, porous titanium oxide, porous zeolites, porous metal organic frameworks, or porous polymeric structures.

37. The material of claim 36, wherein the substrate is porous carbon.

38. The material of claim 37, wherein the substrate is CMK-3.

39. The material of claim 37, wherein the substrate is activated carbon.

40. The material of claim 36, wherein the substrate is porous silica.

41. The material of one of claims 33-40, wherein the layer material comprises a metal oxide.

42. The material of claim 41, wherein the metal oxide is selected from T1O2, Si0₂, AI2O3, Sc₂0₃, V2O3, V2O5, Cr₂0₃, Mn0₂, Ga₂0₃, FeO, Fe₂0₃, Fe₃0₄, CoO, Co₃0₄, NiO, N1O2, CuO, Cu₂0, ZnO, SrO, Y2O3, Zr0₂, CdO, Ag₂0, Rh₂0₃, Nb₂0₃, Nb₂0₅, W2O3, W0₂, WO3, M0O2, M0O3, Ru0₂, Re0₃, Re₂0₇, Ir0₂, PdO, PtO, Sn0₂, Sb₂0₃, Te0₂, Ge0₂, PbO, Pb0₂, Ce₂0₅, Ce0₂ or Eu₂0₃.

43. The material of claim 42, wherein the metal oxide is T1O2.

44. The material of claim 42, wherein the metal oxide is Sn0₂.

45. The material of claim 42, wherein the metal oxide is V2O5.

46. The material of one of claims 33-40, wherein the layer material comprises a metal nitride.

47. The material of one of claims 33-40, wherein the layer material comprises a metal carbide.

48. The material of one of claims 33-40, wherein the layer material comprises a metal phosphide.

49. The material of one of claims 33-40, wherein the layer material comprises a metal oxynitride.

50. The material of claim 46, wherein the metal nitride is TiN.

51. A material, comprising:

a substrate selected from porous carbon, porous silica, porous alumina, porous titanium oxide, porous zeolites, or porous metal organic frameworks, comprising a surface, a plurality of pores, each pore being defined by a pore surface that is a portion of the surface, and an exterior portion that does not define the pores; and

one or more layers, each independently comprising a layer material selected from metal oxide, metal nitride, metal carbide, metal phosphide or metal oxynitride, and each layer located such that the pore surface is at least partially covered by one or more layers and at least 50% of the exterior portion is not covered by a layer.

52. The material of claim 51, wherein the metal oxide is selected from T1O2, Si0₂, AI2O3, Sc₂0₃, V2O3, V2O5, Cr₂0₃, Mn0₂, Ga₂0₃, FeO, Fe₂0₃, Fe₃0₄, CoO, C0O2, NiO, N1O2, CuO, Cu₂0, ZnO, SrO, Y2O3, Zr0₂, CdO, Ag₂0, Rh₂0₃, Nb₂0₃, Nb₂0₅, W2O3, W0₂, WO3, M0O2, M0O3, RuC-2, Re0₃, Ir0₂, PdO, PtO, Sn0₂, Sb₂0₃, Te0₂, Ge0₂, PbO, Pb0₂, Ce₂0₅, Ce0₂

53. A device, comprising the material of any one of claims 33-52.

54. The device of claim 53, wherein the device is selected from a capacitor, a supercapacitor, a battery, a gas sensor, a hydrogen storage device, a water treatment device, fuel cell or a catalyst.

55. The device of claim 54, wherein the device is an electric double layer capacitor.

56. The device of claim 54, wherein the device is a battery.

57. The device of claim 56, wherein the battery is a sodium ion battery or a lithium ion battery.

58. The device of claim 56, wherein the battery comprises an electrode comprising the material.

59. The device of claim 58, wherein the material comprises T1O2.

60. The device of claim 58, wherein the material comprises V2O5.

61. A battery comprising the material of any one of claims 33-52.