THIN-FILM ION CONDUCTING MEMBRANE
FIELD OF THE INVENTION The present invention relates to thin-film ion conducting membranes for use in fuel cells and batteries, and is based on the incorporation of ion conducting materials into the matrix of an amorphous carbon nanocomposite.
BACKGROUND OF THE INVENTION Ion conducting membranes with high ionic conductivity are critical enabling technologies for primary and rechargeable batteries, fuel cells and other electrochemical devices such as electrochromic windows and displays and electrochemical sensors. For rechargeable lithium batteries, the key issue is to stabilize the lithium-electrolyte interface against corrosion. In the absence of a stable lithium metal anode, lithium-ion battery manufacturers have resorted to lithiated carbon as anodes to achieve long cycle life. If lithium metal could be used as the anode, it would provide higher energy density. The only stable metallic lithium- electrolyte interfaces that have been demonstrated to date are solid state: lithium against an inorganic glass/ceramic solid-state electrolyte. If this stable interface can be mimicked by inserting a solid-state membrane between the lithium anode and the electrolyte, it will result in high energy density and long cycle life. A key problem has been to produce thin film membranes that have a sufficiently low density of defects to be incorporated into practical devices.
For fuel cells using methanol or other liquid fuels, it is desirable to operate at elevated temperatures for improved oxidation kinetics and reduced catalyst poisoning. This is not possible with current polymer electrolyte membrane fuel cell technology, where the proton conducting membrane incorporates water, limiting the temperature range of operations. Another problem is drag of solvents through the membrane. A solid-state membrane that has high proton conductivity, is impermeable to liquid fuels and can operate at temperatures in the range of 150°C to 200°C would represent a major technical advance.
A preferred source of hydrogen for fuel cells is natural gas, or methane, which can be converted to hydrogen more easily than liquid hydrocarbons. Systems to produce hydrogen from hydrocarbons would be improved by the use of thin-film thermally stable membranes with high hydrogen permeability. There is a need for ion conducting membranes, in particular for proton and lithium ion-conducting membranes, that can be produced as thin, flexible films with a low density of defects.
SUMMARY OF THE INVENTION It is the obj ect of the instant invention to provide an improved ion conducting solid-state membrane that has high ionic conductivity.
It is a further object of the instant invention to provide a membrane that has high conductivity for proton and lithium ions to be applied in fuel cells and lithium batteries, respectively. It is an additional object of the instant invention to provide ion-conducting membranes that can be produced as thin films with a low density of defects.
It is an additional object of the instant invention to provide ion-conducting membranes that can be produced by coating in vacuum.
It is an additional object of the instant invention to provide ion-conducting membranes that are compatible with silicon microchip processing for the production of miniaturized batteries and fuel cells based on microsystems.
The instant invention is directed to a solid-state ion conducting membrane based on the incorporation of ion conducting materials into an amorphous carbon film, in particular an amorphous diamond-like carbon nanocomposite film, produced by co-deposition of the ion conducting material during the growth of the amorphous carbon film. The concentration of the ion conducting material is sufficiently high to provide a continuous phase through the film, thereby providing a connected path for ion conductivity.
As will be appreciated by those skilled in the art, features of one aspect or embodiment of the invention are also applicable to other aspects or embodiments of the invention.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 is a schematic showing a reactor for depositing conducting diamondlike carbon films.
FIG. 2 is a graph showing the conductivity of conducting amorphous diamond-like carbon nanocomposite films containing tungsten, as a function of the tungsten concentration.
DETAILED DESCRIPTION OF THE INVENTION Amorphous carbon films, and particularly amorphous diamond-like carbon nanocomposite (DLCN) films, which consist of a matrix comprising carbon, hydrogen, silicon and oxygen, can be made electrically conducting by incorporating other elements and compounds into the matrix during the growth of the film. DLCN films are obtained by Plasma Assisted Chemical Vapor Deposition (PACVD), where the feedstock is a liquid organo-silicone (e.g. polyphenylmethylsiloxane, or PPMS) used as a plasma-forming substance. FIG. 1 shows schematically a PACVD reactor consisting of a vacuum chamber 1, a magnetron 2, an electromagnet 3, a tungsten thermocathode 4, a feed mechanism for an organo-silicone compound 5, a carousel with substrates 6, a shutter 7, a high frequency voltage source 8, a carbon-silicon beam 9, and an ion beam of a metal or other elements to be incorporated into the DLCN matrix 10.
A conducting DLCN film is made by co-depositing (using for example sputtering, thermal evaporation or electron-beam) additional elements during the growth of the DLCN film. Essentially any element of the periodic table can be embedded into the DLCN matrix, as well as compound materials.
The resistivity of the film is controlled by controlling the composition during film growth. Typically, there is a percolation threshold to high electrical conductivity at 15-20 at.% concentrations of metallic "dopants". Conductivities as high as 103 S/cm have been measured with films incorporating metals such as tungsten. Undoped films have conductivities of 10"10 - 10"12 S/cm. The films can be deposited with good adhesion on metals, ceramics, semiconductors and plastics. Dopant concentrations can
be as high as 45-50 at.%. At low concentrations, the dopants are in the form of randomly dispersed atoms. FIG. 2 shows the conductivity of DLCN films as a function of the concentration of tungsten (W) incorporated into the films. As can be seen, a percolation threshold to high conductivity occurs at approximately 20 at.% DLCN materials are significantly more stable towards high temperatures than typical diamond-like carbon (DLC) materials. Many of the film properties are similar to those of DLC materials: extreme hardness; high wear resistance; high elastic modulus (100 - 400 GPa); low friction coefficient (0.04 - 0.2); and high chemical stability. Incorporating metal atoms into the matrix releases the internal stress. The amorphous carbon nanocomposite materials coat conformally and produce hermetic seals at very low thickness. Even 5-10 nm thick films on steel show effective corrosion protection. This implies that single layer coatings may produce effectively pinhole-free films. The thickness can range up to several micrometers.
Ion conducting membranes can be produced by co-sputtering (or depositing by other physical vapor deposition means such as evaporation, pulsed cathodic arc, electron beam, or by depositing by chemical vapor deposition) ion conducting materials during the DLCN film growth, incorporating ion conducting materials into the amorphous carbon:silicon matrix in sufficient concentration (above the percolation threshold) that the ion conductivity of the film is determined by a continuous phase of the ion conducting material. The estimated concentration will be in the range of 25-50 at.%. The films can be produced in a range of thicknesses.
The membranes may be deposited on metals, ceramics, semiconductors and plastics. The membranes may be part of a multilayer structure with other ion conducting materials such as polymers to increase strength and flexibility. The membranes may also be deposited on silicon wafer substrates for the fabrication of microsystem batteries and fuel cells. Amorphous DLCN films can be dense and pinhole-free coatings at very low thickness. DLCN film deposition is compatible with microelectronics processing and fabrication of microelectromechanical systems (MEMS), thus allowing the membranes of the present invention to be incorporated into MEMS-based fuel cells and microbatteries.
Materials that may be incorporated to produce proton conducting membranes for fuel cells include, but are not limited to, Pd, Ni, Ag and their alloys, e.g., Pd-Cu, Pd-
Ag and V-Ni-Cr; metal hydride compounds such as AB5 structures, e.g., LaNi5, CaNi5, LaNi4.7Alo.3, and other metal hydrides, e.g., Ti(Fe0.9Mno.ι), Ti(Fe0.8Ni0.2), Mg Ni, Mg and Ti; and ion-conducting glasses and ceramics.
Materials that may be incorporated to produce lithium ion conducting membranes for lithium batteries include, but are not limited to, metals, such as LixSn, LixAl, LixZn, LixAg, LixBi, glasses and ceramics, such as LiAlOx and LiLaxTiyO3, lithium- phosphorous oxynitride, Li2S + B2S3 + P2S5 and Li2S + B2O3 + P O5 with LiBr and Lil as dopants.
Various changes and modifications of the composition of the materials and processes will be apparent to those skilled in the art. Such changes and modifications are to be understood as included in the scope of the present invention defined by the claims as follows.