Electrode for rechargeable lithium battery and rechargeable lithium battery

Information

  • Patent Grant
  • 7192673
  • Patent Number
    7,192,673
  • Date Filed
    Friday, October 20, 2000
    24 years ago
  • Date Issued
    Tuesday, March 20, 2007
    17 years ago
Abstract
An electrode for a rechargeable lithium battery which includes a thin film composed of active material that expands and shrinks as it stores and releases lithium, e.g., a microcrystalline or amorphous silicon thin film, deposited on a current collector, characterized in that said current collector exhibits a tensile strength (=tensile strength (N/mm2) per sectional area of the current collector material×thickness (mm) of the current collector) of not less than 3.82 N/mm.
Description
TECHNICAL FIELD

The present invention relates to a novel electrode for a rechargeable lithium battery and also to a rechargeable lithium battery using the same.


BACKGROUND ART

The battery performance of rechargeable lithium batteries recently under intensive research and development, such as charge-discharge voltage, cycle life characteristics or storage characteristics, depends largely upon the types of the electrodes used. This has led to the attempts to better battery performance by improving electrode active materials.


The use of metallic lithium for the negative active material, although possible to construct a battery with high energy density per weight and volume, presents a problem that the lithium deposited on charge grows into dendrite which could cause internal short-circuiting.


Rechargeable lithium batteries are reported (Solid State Ionics, 113–115, p 57 (1998)) which use an electrode consisting of aluminum, silicon, tin or the like that is electrochemically alloyed with lithium on charge. Among these, a silicon electrode provides a particularly high theoretical capacity and is promising as a high-capacity negative electrode. For this reason, various rechargeable batteries using silicon for the negative electrode are proposed (Japanese Patent Laying-Open No. Hei 10-255768). However, such alloying negative electrodes fail to provide sufficient cycle characteristics since alloys, as electrode active materials, are themselves pulverized on charge and discharge to reduce current-collecting capabilities.


DISCLOSURE OF THE INVENTION

The inventors of this application have found that an electrode including a thin film composed of active material, e.g., a microcrystalline or amorphous silicon thin film, deposited on a current collector by a film-forming process such as a sputtering or CVD process exhibits improved charge-discharge cycle characteristics and is suitable for incorporation in rechargeable lithium batteries, as demonstrated in Reference Experiments 1–8 which will be later described.


However, since the thin film of active material expands and shrinks as it stores and releases lithium, a stress is engendered in the current collector during a charge-discharge reaction and occasionally causes formation of wrinkles in such an electrode for a rechargeable lithium battery. The formation of wrinkles not only increases a volume of the electrode but also disturbs uniformity of an electrode reaction, resulting in the reduction of an energy density.


It is an object of the present invention to provide an electrode, for use in a rechargeable lithium battery, which includes a thin film composed of active material that expands and shrinks as its stores and releases lithium, deposited on a current collector and which can restrain formation of wrinkles therein during charge and discharge, and also to provide a rechargeable lithium battery using the electrode.


In accordance with a first aspect of the present invention, an electrode for a rechargeable lithium battery is provided including a thin film composed of active material that expands and shrinks as it stores and releases lithium, deposited on a current collector. Characteristically, the current collector exhibits a tensile strength×thickness of not less than 3.82 N/mm when determined by the following equation:


Tensile strength of the current collector (N/mm)=tensile strength (N/mm2) per sectional area of the current collector material×thickness (mm) of the current collector.


In the equation, the tensile strength per sectional area of the current collector material can be measured in accordance with a method specified in Japan Industrial Standards (JIS), for example.


In the first aspect of the present invention, the tensile strength×thickness of the current collector is preferably not less than 7.44 N/mm.


In accordance with a second aspect of the present invention, an electrode for a rechargeable lithium battery is provided which includes a thin film composed of active material that expands and shrinks as it stores and releases lithium, deposited on a current collector. Characteristically, the current collector exhibits a tensile strength of not less than 1.12 N/mm per 1 μm thickness of the thin film of active material.


The tensile strength per 1 μm thickness of the thin film of active material can be determined by the following equation:


(Tensile strength per 1 μm thickness of the thin film of active material)=(tensile strength of the current collector)÷(thickness of the thin film of active material: μm).


The tensile strength value of the current collector in this equation is previously defined in the first aspect.


In the second aspect of the present invention, the tensile strength per 1 μm thickness of the thin film of active material is preferably not less than 2.18 N/mm, more preferably not less than 4.25 N/mm.


In accordance with a third aspect of the present invention, an electrode for a rechargeable lithium battery is provided which includes a thin film composed of active material that expands and shrinks as it stores and releases lithium, deposited on a current collector. Characteristically, the current collector is related to the thin film of active material by a thickness ratio (thickness of the thin film of active material/thickness of the current collector) of not greater than 0.19.


In the third aspect of the present invention, the ratio in thickness of the thin film of active material to the current collector is preferably not greater than 0.098, more preferably not greater than 0.05.


In accordance with the first, second and third aspects of the present invention, the tendency of the electrode to form wrinkles during charge and discharge can be reduced.


The matters common to the first, second and third aspects of the present invention are described hereinafter as those of the “present invention”.


In the present invention, a surface roughness Ra (roughness average) of the current collector is preferably in the range of 0.01–1 μm, more preferably in the range of 0.05–0.5 μm. The surface roughness Ra of the current collector is preferably comparable in level to the surface roughness Ra of the electrolytic copper foil which will be described later. Accordingly, the surface roughness Ra of the current collector is preferably 0.1 μm or larger, more preferably in the range of 0.1–1 μm. The surface roughness Ra, which is specified in Japanese Industrial Standards (JIS B 0601-1994), can be measured as by a surface roughness meter, for example.


In the present invention, the surface roughness Ra of the current collector preferably satisfies the relationship Ra≦t, where t is a thickness of the thin film of active material. It is also preferred that the surface roughness Ra and a mean spacing of local peaks of profile S of the current collector satisfy the relationship 100Ra≧S. The mean spacing of local peaks of profile S, which is specified in Japanese Industrial Standards (JIS B 0601-1994), can be measured as by a surface roughness meter, for example.


The shape of the projections on the current collector surface is not particularly specified, but may preferably be a conical form, for example.


In the present invention, it is preferred that the current collector constituent is allowed to diffuse into the thin film of active material. The diffusion of the current collector constituent into the thin film of active material not only improves adhesion between the current collector and the thin film of active material, but also effectively prevents the thin film of active material from separating from the current collector. As a result, the charge-discharge characteristics are further improved.


In the case where the thin film is composed of active material capable of alloy formation with lithium and the current collector is composed of material incapable of alloy formation with lithium, the diffusion of the current collector constituent lessens expansion and shrinkage of a thin film portion in the vicinity of the current collector during storage and release of lithium. Thus, the thin film of active material can be kept adhered more effectively to the current collector.


Preferably, the concentration of the current collector constituent in the thin film is higher in the vicinity of the current collector and is lower at a location closer to the thin film surface. Due to the presence of such a concentration gradient, the thin film is subjected to less expansion and shrinkage in the vicinity of the current collector so that the thin film can be kept adhered to the current collector, and is permitted to contain a relatively lager amount of active material in the vicinity of its surface so that a high charge-discharge capacity can be retained.


It is preferred that the current collector constituent diffused in the thin film, forms a solid solution, instead of an intermetallic compound, with a component of the thin film. The intermetallic compound, as used herein, refers to a compound which has a specific crystal structure formed via combination of metals in specific proportions. The formation of solid solution, instead of intermetallic compound, between the thin film component and the current collector constituent improves adhesion between the thin film and the current collector, resulting in obtaining the increased charge-discharge capacity.


In the present invention, the thickness of a thin film region into which the current collector constituent is diffused is not particularly specified, but may preferably be 1 μm or greater.


The current collector for use in the present invention is not particularly specified in type, so long as it satisfies the above-described conditions of the present invention. More specifically, the current collector may be comprised of at least one selected from copper, nickel, stainless steel, molybdenum, tungsten and tantalum, for example.


The current collector is preferably thin and thus preferably in the form of a metal foil. The current collector is preferably comprised of a material which does not form an alloy with lithium, particularly preferably of copper. The current collector is preferably in the form of a copper foil which is preferably roughened at its surface. A typical example of such a copper foil is an electrolytic copper foil. This electrolytic copper foil is obtained, for example, from an electrolyte containing copper ions: A metallic drum is immersed in the electrolyte solution and rotated. The introduction of current results in the electrodeposition of copper on a surface of the drum. The electrolytic copper foil is then obtained by removing the deposited copper from the drum. The electrolytic copper foil may be subjected to surface roughening or other surface treatment at one or both sides thereof.


As the current collector, a rolled copper foil, whose surface is roughened by depositing copper thereon via an electrolytic process, may be used.


Also, an interlay may be provided between the current collector and the thin film of active material. In such a case, the interlayer is preferably comprised of a material which contains a constituent easy to diffuse into the thin film of active material, e.g., a copper layer. Such a copper layer may be superimposed on a surface-roughened nickel foil (e.g., electrolytic nickel foil) to provide the current collector. Alternatively, copper may be deposited on a nickel foil by an electrolytic process during which the nickel foil is roughened at its surface.


The thin film of active material in accordance with the present invention can be formed from one or more materials capable of producing compounds or solid solutions with lithium, e.g. at least one selected from elements from Groups IIB, IIIB, IVB and VB of the periodic table and oxides and sulfides of transition metal elements from Periods 4, 5 and 6 of the periodic table.


In the present invention, examples of elements from Groups IIB, IIIB, IVB and VB of the periodic table that can produce compounds or solid solutions with lithium include carbon, aluminum, silicon, phosphorus, zinc, gallium, germanium, arsenic, cadmium, indium, tin, antimony, mercury, thallium, lead and bismuth. Specific examples of transition metal elements from Periods 4, 5 and 6 of the periodic table include scandium, titanium, vanadium, chromium, manganese, iron, cobalt, nickel, copper, zinc, yttrium, zirconium, niobium, molybdenum, technetium, ruthenium, rhodium, palladium, silver, cadmium, lanthanum series elements, hafnium, tantalum, tungsten, rhenium, osmium, iridium, platinum, gold and mercury.


Preferred among the above-listed elements is at least one selected from carbon, silicon, germanium, tin, lead, aluminum, indium, zinc, cadmium, bismuth and mercury. Silicon and/or germanium is more preferred.


In general, silicon is roughly classified by its crystallinity into amorphous silicon, microcrystalline silicon, polycrystalline silicon and single crystal silicon. The term “noncrystalline silicon”, as used herein, is meant to encompass amorphous silicon and microcrystalline silicon and exclude polycrystalline silicon and single crystal silicon. Silicon is identified as the amorphous silicon when Raman spectroscopy detects substantial absence of a peak around 520 cm−1 which characterizes a crystalline region, and as the microcrystalline silicon when Raman spectroscopy detects the substantial presence of a peak around 520 cm−1 corresponding to the crystalline region and a peak around 480 cm−1 which indicates an amorphous region. Hence, the microcrystalline silicon consists substantially of a crystalline region and an amorphous region. Silicon is identified as the single crystal silicon or polycrystalline silicon when Raman spectroscopy detects substantial absence of a peak around 480 cm−1 corresponding to the amorphous region.


In the present invention, the silicon thin film in the microcrystalline or amorphous form is preferably used for the thin film of active material.


In addition to the above-described silicon thin film, a germanium thin film or a silicon-germanium alloy thin film may also be preferably used for the thin film of active material in the present invention. The germanium thin film in the amorphous or microcrystalline form is preferably used. The preferred silicon-germanium alloy thin film is in the microcrystalline or amorphous form. The above-described procedure applied to the silicon thin film can be similarly utilized to determine the microcrystalline or amorphous nature of the germanium and silicon-germanium alloy thin films. The use of silicon and germanium provides good results as evidenced by Examples which will be described hereinafter. Since silicon and germanium can be mixed with each other in arbitraty proportions to produce solid solutions, similar results are expected for the silicon-germanium alloy.


In the present invention, it is preferred that the thin film of active material is divided into columns by gaps formed therein in a manner to extend in its thickness direction and the columnar portions are at their bottoms adhered to the current collector. It is also preferred that a thickness portion of the active material thin film that occupies at least a half of its thickness is preferably divided into columns by such gaps.


Preferably, the gaps are formed by the expansion and shrinkage of the thin film, which may be caused to occur by a charge-discharge reaction, for example. Accordingly, the gaps may be formed by the charge-discharge reaction either after or before the electrode is assembled into a battery. Illustrating one method of forming such gaps in the thin film before it is subjected to a charge-discharge process, the thin film of the electrode before being assembled into a battery is allowed to store and then release lithium or the like so that the thin film is made to expand and then shrink in volume, thereby forming the gaps. Also, the thin film divided by gaps into columns can be lithographically formed by utilizing a resist film patterned by photolithography.


In the case where the thin film of active material has irregularities on its surface, the aforementioned gaps may be formed therein to extend in its thickness direction from valleys of the irregularities on the thin film surface toward the current collector. The irregularities on the thin film surface may conform in shape to those on the current collector surface. That is, depositing the thin film of active material on the current collector with such irregularities results in the formation of the corresponding irregularities on the surface of the thin film.


The columnar portions of the thin film of active material may have various top shapes, but preferably have a round top shape.


The gaps may be formed in advance in the thin film in a manner to extend in its thickness direction along the low-density regions. Such low-density regions may be connected to each other like a network in a planar direction and extend in the thickness direction toward the current collector, for example.


In the present invention, the thin film of active material can be deposited on the current collector by various methods, including, for example, CVD, sputtering, vapor evaporation, spraying and plating processes. Particularly preferred among such thin-film forming methods are CVD, sputtering and vapor evaporation processes.


The thin film of active material in the present invention may be doped with an impurity. Examples of such impurities include elements of the periodic Groups IIIB, IVB, VB and VIB, such as phosphorus, aluminum, arsenic, antimony, boron, gallium, indium, oxygen and nitrogen.


Also, the thin film of active material in the present invention may be made up of a sequence of superimposed layers. These layers may differ from each other in terms of composition, crystallinity, impurity concentration or the like. Such layers may provide a thin film structure graded in its thickness direction. For example, such layers, if properly arranged, can provide a thin film structure wherein the composition, crystallinity, impurity concentration or the like is varied in its thickness direction.


Preferably, the thin film of active material in the present invention stores lithium via formation of an alloy with lithium.


Lithium may be previously stored or incorporated in the thin film of active material in the present invention. Lithium may be added during deposition of the thin film of active material. That is, lithium may be introduced via formation of a lithium-containing thin film. Alternatively, lithium may be added or stored after formation of the thin film of active material. One method is to use an electrochemical mechanism whereby lithium is added or stored in the thin film of active material.


The thickness of the thin film of active material in the present invention is preferably 1 μm or above, for the purpose of obtaining a high charge-discharge capacity.


In the present invention, an interlayer may be provided between the current collector and the thin film to improve adhesion therebetween, as described above. Preferably, such an interlayer may comprise a material capable of alloy formation with both the current collector material and the active material.


The rechargeable lithium battery (lithium secondary battery) of the present invention is characterized as including a negative electrode comprised of the above-described electrode of the present invention, a positive electrode and a nonaqueous electrolyte.


An electrolyte solvent for use in the rechargeable battery of the present invention is not particularly limited in type but can be illustrated by a mixed solvent which contains cyclic carbonate such as ethylene carbonate, propylene carbonate or butylene carbonate and also contains chain carbonate such as dimethyl carbonate, methyl ethyl carbonate or diethyl carbonate. Also applicable is a mixed solvent of the above-listed cyclic carbonate and an ether solvent such as 1,2-dimethoxyethane or 1,2-diethoxyethane or a chain ester such as y-butyrolactone, sulfolane or methyl acetate. Illustrative electrolyte solutes are LiPF6, LiBF4, LiCF3SO3, LiN(CF3SO2)2, LiN(C2F5SO2)2, LiN (CF3SO2) (C4F9SO2), LiC(CF3SO2)3, LiC(C2F5SO2)3, LiAsF6, LiClO4, Li2B10Cl10, Li2B12C12 and mixtures thereof. Other applicable electrolytes include a gelled polymer electrolyte comprised of an electrolyte solution impregnated into a polymer electrolyte such as polyethylene oxide, polyacrylonitrile or polyvinylidene fluoride and inorganic solid electrolytes such as LiI and Li3N, for example. The electrolyte for the recharageable lithium battery of the present invention can be used without limitation, so long as an Li compound as its solute that imparts an ionic conductivity, as well as its solvent that dissolves and retains the Li compound, remain undecomposed at voltages during charge, discharge and storage of the battery.


Examples of positive active materials for the present invention include lithium-containing transition metal oxides such as LiCoO2, LiNiO2, LiMn2O4, LiMnO2, LiCo0.5Ni0.5O2 and LiNi0.7Co0.2Mn0.1O2; lithium-free metal oxides such as MnO2; and the like. Other substances can also be used, without limitation, if they are capable of electrochemical insersion and release of lithium.





BRIEF DESCRIPTION OF THE DRAWINGS


FIG. 1 is a schematic sectional view of a rechargeable lithium battery fabricated in Examples in accordance with the present invention;



FIG. 2 is a photomicrograph (at a magnification of 2,000×) taken using a scanning electron microscope, showing an electrode in accordance with one embodiment of the present invention in its state before charge and discharge;



FIG. 3 is a photomicrograph (at a magnification of 5,000×) taken using a scanning electron microscope, showing an electrode in accordance with one embodiment of the present invention in its state before charge and discharge;



FIG. 4 is a photomicrograph (at a magnification of 500×) taken using a scanning electron microscope, showing an electrode in accordance with one embodiment of the present invention in its state after charges and discharges;



FIG. 5 is a photomicrograph (at a magnification of 2,500×) taken using a scanning electron microscope, showing an electrode in accordance with one embodiment of the present invention in its state after charges and discharges;



FIG. 6 is a photomicrograph (at a magnification of 1,000×) taken using a scanning electron microscope, showing a silicon thin film of an electrode in accordance with one embodiment of the present invention when viewed from above;



FIG. 7 is a photomicrograph (at a magnification of 5,000×) taken using a scanning electron microscope, showing a silicon thin film of an electrode in accordance with one embodiment of the present invention when viewed from above;



FIG. 8 is a photomicrograph (at a magnification of 1,000×) taken using a scanning electron microscope, showing a silicon thin film of an electrode in accordance with one embodiment of the present invention when viewed from a slight angle;



FIG. 9 is a photomicrograph (at a magnification of 5,000×) taken using a scanning electron microscope, showing a silicon thin film of an electrode in accordance with one embodiment of the present invention when viewed from a slight angle;



FIG. 10 is a schematic sectional view, showing a silicon thin film in the process of being divided by gaps into columns;



FIG. 11 a photomicrograph (at a magnification of 12,500×) taken using a transmission electron microscope, showing a section of a silicon thin film of an electrode a3 in accordance with the present invention;



FIG. 12 a photomicrograph (at a magnification of 25,000×) taken using a transmission electron microscope, showing a section of a silicon thin film of an electrode a6 in accordance with the present invention;



FIG. 13 is a diagrammatic representation of the photomicrograph of FIG. 11;



FIG. 14 is a diagrammatic representation of the photomicrograph of FIG. 12;



FIG. 15 is a photomicrograph (at a magnification of 1,000×) taken using a scanning electron microscope, showing a silicon thin film surface of an electrode a3 in accordance with the present invention when viewed from above;



FIG. 16 is a photomicrograph (at a magnification of 1,000×) taken using a scanning electron microscope, showing a silicon thin film surface of an electrode a6 in accordance with the present invention when viewed from above;



FIG. 17 is a graphical representation illustrating concentration profiles of constituent elements in a silicon thin film of an electrode a6 in accordance with the present invention along the depth of the film;



FIG. 18 is a schematic view, showing a construction of an apparatus which is employed when a thin film is formed by a vacuum vapor evaporation technique in examples of the present invention;



FIG. 19 is a photomicrograph (at a magnification of 2,000×) taken using a scanning electron microscope, showing an electrode a7 in accordance with the present invention in its state before charge and discharge;



FIG. 20 is a photomicrograph (at a magnification of 10,000×) taken using a scanning electron microscope, showing the electrode a7 in accordance with the present invention in its state before charge and discharge;



FIG. 21 is a photomicrograph (at a magnification of 2,000×) taken using a scanning electron microscope, showing an electrode a8 in accordance with the present invention in its state before charge and discharge;



FIG. 22 is a photomicrograph (at a magnification of 10,000×) taken using a scanning electron microscope, showing the electrode a8 in accordance with the present invention in its state before charge and discharge;



FIG. 23 is a photomicrograph (at a magnification of 500×) taken using a scanning electron microscope, showing the electrode a7 in accordance with the present invention in its state after charges and discharges;



FIG. 24 is a photomicrograph (at a magnification of 2,500×) taken using a scanning electron microscope, showing the electrode a7 in accordance with the present invention in its state after charges and discharges;



FIG. 25 is a photomicrograph (at a magnification of 500×) taken using a scanning electron microscope, showing the electrode a8 in accordance with the present invention in its state after charges and discharges;



FIG. 26 is a photomicrograph (at a magnification of 2,500×) taken using a scanning electron microscope, showing the electrode a8 in accordance with the present invention in its state after charges and discharges;



FIG. 27 is a photomicrograph (at a magnification of 1,000×) taken using a scanning electron microscope, showing a germanium thin film of the electrode a7 in accordance with the present invention in its state after charges and discharges, when viewed from above;



FIG. 28 is a photomicrograph (at a magnification of 5,000×) taken using a scanning electron microscope, showing a germanium thin film of the electrode a7 in accordance with the present invention in its state after charges and discharges, when viewed from above;



FIG. 29 is a photomicrograph (at a magnification of 1,000×) taken using a scanning electron microscope, showing a germanium thin film of the electrode a7 in accordance with the present invention in its state after charges and discharges, when viewed from a slight angle;



FIG. 30 is a photomicrograph (at a magnification of 5,000×) taken using a scanning electron microscope, showing a germanium thin film of the electrode a7 in accordance with the present invention in its state after charges and discharges, when viewed from a slight angle;



FIG. 31 is a photomicrograph (at a magnification of 1,000×) taken using a scanning electron microscope, showing a germanium thin film of the electrode a8 in accordance with the present invention in its state after charges and discharges, when viewed from above;



FIG. 32 is a photomicrograph (at a magnification of 5,000×) taken using a scanning electron microscope, showing a germanium thin film of the electrode a8 in accordance with the present invention in its state after charges and discharges, when viewed from above;



FIG. 33 is a photomicrograph (at a magnification of 1,000×) taken using a scanning electron microscope, showing a germanium thin film of the electrode a8 in accordance with the present invention in its state after charges and discharges, when viewed from a slight angle;



FIG. 34 is a photomicrograph (at a magnification of 5,000×) taken using a scanning electron microscope, showing a germanium thin film of the electrode a8 in accordance with the present invention in its state after charges and discharges, when viewed from a slight angle;



FIG. 35 is a photomicrograph (at a magnification of 1,000×) taken using a scanning electron microscope, showing a germanium thin film of the electrode a7 in accordance with the present invention in its state before charge and discharge, when viewed from above;



FIG. 36 is a photomicrograph (at a magnification of 1,000×) taken using a scanning electron microscope, showing a germanium thin film of the electrode a8 in accordance with the present invention in its state before charge and discharge, when viewed from above;



FIG. 37 is a graphical representation illustrating concentration profiles of constituent elements in a germanium thin film of the electrode a7 in accordance with the present invention along the depth of the film;



FIG. 38 is a graphical representation, illustrating concentration profiles of constituent elements in a germanium thin film in the electrode a8 in accordance with the present invention along the depth of the film;



FIG. 39 is a photomicrograph (at a magnification of 2,000×) taken using a scanning electron microscope, showing a section of an electrode a-11 in accordance with the present invention before charge and discharge;



FIG. 40 is a photomicrograph (at a magnification of 10,000×) taken using a scanning electron microscope, showing a section of the electrode a-11 in accordance with the present invention before charge and discharge;



FIG. 41 is a photomicrograph (at a magnification of 1,000×) taken using a scanning electron microscope, showing a silicon thin film of the electrode a-11 in accordance with the present invention before charge and discharge, when viewed from above;



FIG. 42 is a photomicrograph (at a magnification of 1,000×) taken using a scanning electron microscope, showing a silicon thin film of the electrode a-11 in accordance with the present invention after charges and discharges, when viewed from above;



FIG. 43 is a photomicrograph (at a magnification of 500,000×) taken using a transmission electron microscope, showing an interface between a copper foil and a silicon thin film and its vicinities;



FIG. 44 is a photomicrograph (at a magnification of 1,000,000×) taken using a transmission electron microscope, showing an interface between a copper foil and a silicon thin film and its vicinities;



FIG. 45 is a graphical representation, illustrating concentration profiles of copper and hydrogen in a mixed layer in an electrode c1 along the depth of the layer; and



FIG. 46 is a graphical representation, illustrating concentration profiles of copper and hydrogen in a mixed layer in an electrode c3 along the depth of the layer.



FIG. 47 is a perspective view showing a rechargeable lithium battery constructed in a further embodiment of the present invention;



FIG. 48 is a schematic sectional view showing a rechargeable lithium battery constructed in a further embodiment of the present invention;



FIG. 49 is a graph showing the relation of the thickness of the copper foil electrode to the charge-discharge cycle characteristics;



FIG. 50 is a view showing the top surface condition of the negative electrode after charges and discharges;



FIG. 51 is a view showing the top surface condition of the negative electrode after charges and discharges;



FIG. 52 is a view showing the back surface condition of the negative electrode after charges and discharges;



FIG. 53 is a view showing the back surface condition of the negative electrode after charges and discharges;



FIG. 54 is a graph showing the relation of the thickness of the silicon thin film electrode to the charge-discharge cycle characteristics; and



FIG. 55 is a view showing the back surface condition of the negative electrode after charges and discharges.





BEST MODE FOR CARRYING OUT THE INVENTION

The present invention is below described in more detail by way of examples. It will be recognized that the following examples merely illustrate the practice of the present invention but are not intended to be limiting thereof. Suitable changes and modifications can be effected without departing from the scope of the present invention.


The following Reference Experiments 1–8 demonstrate that the microcrystalline silicon thin film, amorphous silicon thin film and amorphous germanium thin film, each formed on a current collector such as a copper foil by a CVD or sputtering process, provide good charge-discharge cycle characteristics when they are used for the electrode for a rechargeable lithium battery.


REFERENCE EXPERIMENT 1

(Fabrication of Negative Electrode)


A microcrystalline silicon thin film was formed on a rolled copper foil (18 μm thick) by a CVD method, using the rolled copper foil as a substrate, silane (SiH4) as a source gas and a hydrogen gas as a carrier gas. Specifically, the copper foil as a substrate was placed on a heater within a reaction chamber. An interior of the reaction chamber was evacuated by a vacuum evacuator to a pressure of 1 Pa or lower. The silane gas as a source gas and the hydrogen (H2) gas as a carrier gas were introduced via a source gas inlet port. The substrate was heated to 180° C. on the heater. A degree of vacuum was adjusted by the vacuum pumping apparatus to a reaction pressure. An RF power supply was operated to excite a radio frequency wave which is introduced via an electrode to induce a glow discharge. Detailed thin-film forming conditions are listed in Table 1. In Table 1, a volumetric unit, sccm, indicates a volumetric flow rate (cm3/minute) of a fluid at 0° C. at 1 atmospheric pressure (101.33 kPa) per minute and is an abbreviation of standard cubic centimeters per minute.












TABLE 1







Conditions
During Film Formation









Source Gas (SiH4) Flow Rate
 10 sccm



Carrier Gas (H2) Flow Rate
200 sccm



Substrate Temperature
180° C.



Reaction Pressure
40 Pa



RF Power
555W










The microcrystalline silicon thin film was deposited under the above-specified conditions to a thickness of about 10 μm. Observation by an electron microscope (at 2,000,000× magnification) ascertained noncrystallinity of the thin film in the way that an amorphous region was arranged to surround a crystalline region consisting of microfine crystal grains. A 17 mm diameter piece was punched out from the resulting sample to provide an electrode a1. A piece identical to the electrode a1 was subjected to heat treatment at 400° C. for 3 hours to provide an electrode a2.


For comparative purposes, 90 parts by weight of commercially available single crystal silicon powder (particle diameter of 10 μm) and 10 parts by weight of polytetrafluoroethylene as a binder were mixed. This mixture was pressed in a 17 mm diameter mold to obtain a pellet-form electrode b1.


(Fabrication of Positive Electrode)


Starting materials, Li2CO3 and CoCO3, were weighed such that the atomic ratio of Li and Co, Li:Co, was brought to 1:1, and then mixed in a mortar. The mixture was pressed in a 17 mm diameter mold and calcined in the air at 800° C. for 24 hours to obtain a calcined product consisting of LiCoO2. This was subsequently ground into particles with a mean particle diameter of 20 μm.


80 parts by weight of the resulting LiCoO2 powder, 10 parts by weight of acetylene black as a conducting material and 10 parts by weight of polytetrafluoroethylene as a binder were mixed. The mixture was pressed in a 17 mm diameter mold to obtain a pellet-form positive electrode.


(Preparation of Electrolyte Solution)


1 mole/liter of LiPF6 was dissolved in a mixed solvent containing equi-volumes of ethylene carbonate and diethyl carbonate to prepare an electrolyte solution for use in the following battery construction.


(Construction of Battery)


A coin type rechargeable lithium battery was constructed using the above-fabricated electrode a1, a2 or b1 for the negative electrode, and the above-fabricated positive electrode and the above-prepared electrolyte solution.



FIG. 1 is a schematic sectional view, illustrating a such-constructed rechargeable lithium battery which includes a positive electrode 1, a negative electrode 2, a separator 3, a positive can 4, a negative can 5, a positive current collector 6, a negative current collector 7 and an insulating gasket 8 made of polypropylene.


The positive electrode 1 and negative electrode 2 are disposed on opposite sides of the separator 3. These are enclosed in a battery case composed of the positive can 4 and negative can 5. The positive electrode 1 is connected to the positive can 4 by the positive current collector 6. The negative electrode 2 is connected to the negative can 5 by the negative current collector 7. Such construction enables charge and discharge as a secondary battery.


As a result, batteries A1, A2 and B1 were constructed using the electrodes a1, a2 and b1 for the negative electrode, respectively.


(Measurement of Charge-Discharge Cycle Life Characteristics)


Each battery, excepting the battery B1, was charged at a current of 100 μA at 25° C. until a negative electrode capacity reached 2,000 mAh/g, and then discharged. This was recorded as a unit charge-discharge cycle. Cycling was effected to measure a 50th-cycle capacity retention rate for each battery. The battery B1, which could not be charged to 2,000 mAh/g, was subjected to a cycle test wherein it was charged to 4.2 V and then discharged. The results are given in Table 2.


In Table 2, a hydrogen concentration obtained from SIMS measurement, a ratio of peak intensities around 480 cm−1 and 520 cm−1 as determined by Raman spectral analysis, and a crystal grain size calculated from an X-ray diffraction spectrum and the Scherrer's equation, all for the negative active material of each battery, are also given. Also, the crystal grain size of the negative active material of the battery B1 is given by the particle diameter of the powder since both are considered to be almost equal in value to each other.













TABLE 2






50th-Cycle






Capacity

Ratio of Peak



Retention
Hydrogen
Intensities
Crystal


Battery
Rate
Content
(480 cm−1/520 cm−1)
Grain Size




















A1
85%
4%
0.1
1
nm


A2
78%
0.01%  
0.1
1
nm


B1
 5%
0%
0
10
μm









As can be clearly seen from the results shown in Table 2, the batteries A1 and A2 in accordance with the present invention both exhibit markedly higher capacity retention rates compared to the comparative battery B1.


As such, the use of the microcrystalline silicon thin film for the negative active material results in the marked improvement of charge-discharge cycle characteristics of the rechargeable lithium battery. This is believed due to the following reason: In the microcrystalline silicon thin film, the moderation of expansion and shrinkage which occurs when lithium is stored and released prevents the negative active material from being pulverized and thereby suppresses the possible reduction of current collecting capability.


REFERENCE EXPERIMENT 2

The procedure used in Reference Experiment 1 to construct the battery A1 was followed, except that an electrolytic copper foil (18 μm thick) was used for the current collector as a substrate. That is, a microcrystalline silicon thin film (about 10 μm thick) was deposited on the electrolytic copper foil to fabricate an electrode a3. Using this electrode, a battery A3 was constructed.


Also, the rolled copper foil used in Reference Experiment 1 was subjected to a one-minute griding treatment with a #400 or #120 emery paper to provide a ground copper foil. The procedure used in Reference Experiment 1 to construct the battery A1 was followed, except that such a ground copper foil was used for the current collector as a substrate. That is, a microcrystalline silicon thin film (about 10 μm thick) was deposited on the copper foil to fabricate an electrode. The electrode fabricated using the copper foil ground with a #400 emery paper was designated as an electrode a4 and the electrode fabricated using the copper foil ground with a #120 emery paper was designated as an electrode a5. These electrodes were used to construct batteries A4 and A5 in the same manner as in Reference Experiment 1.


These batteries A3–A5 and the batteries A1 and B1 constructed in Reference Experiment 1 were subjected to a charge-discharge cycle test under the same conditions used in Reference Experiment 1 to obtain a 10th-cycle capacity retention rate for each. The results are given in Table 3. Also given in Table 3 are a surface roughness Ra and a mean spacing of local peaks S for the copper foil, as a current collector, of each of the batteries A1, B1 and A3–A5.


The surface roughness Ra and the mean spacing of local peaks S for each copper foil were measured using a stylus profiler Dektak ST (available from ULVAC Inc.) with a scanning distance of 2.0 mm. The surface roughness Ra was calculated after correction of a deflection portion. The deflection portion was collected using the correction values with a low pass=200 μm and a high pass=20 μm. The surface roughness Ra was automatically calculated and the mean spacing of local peaks S was read from the chart.












TABLE 3









10th-Cycle




Capacity
Current Collector (Copper Foil)











Retention
Surface Roughness Ra
Mean Spacing S


Battery
Rate
(μm)
(μm)













A1
97%
0.037
14


A3
99%
0.188
11


A4
98%
0.184
9


A5
99%
0.223
8


B1
20%
0.037
14









As can be clearly seen from the results given in Table 3, the batteries A3–A5 using the copper foils with higher values of surface roughness Ra for the current collector exhibit improved 10th-cycle capacity retention rates compared to the battery A1 using the copper foil with the lowest value of surface roughness Ra. This is considered due to the folloiwng reason: The copper foil with a higher value of surface roughness Ra, when used for the current collector, improves adhesion between the current collector and the active material. This adhesion improvement reduces the influence of structural change, such as falling-off of the active material that occurs when it expands or shrinks during storage or release of lithium.


REFERENCE EXPERIMENT 3

The batteries A1 and A3 respectively constructed in Reference Experiments 1 and 2 were further subjected to a charge-discharge cycle test under the same test conditions as used in the Reference Experiment 1 to measure a 30th-cycle capacity retention rate. The results are shown in Table 4.












TABLE 4







Battery
30th-Cycle Capacity Retention Rate









A1
91%



A3
97%










As can be clearly seen from the results given in Table 4, the batteries A1 and A3 exhibit good capacity retention rates even on the 30th-cycle. Particularly, the battery A3 using the copper foil with a higher value of surface roughness Ra for the current collector exhibits good capacity retention rate.


The electrode a3 incorporated in the battery A3 was viewed under an electron microscope to observe a condition of its silicon thin film. First, the electrode a3 in its state prior to being incorporated in the battery, i.e., before charge and discharge, was observed using a scanning electron microscope. FIGS. 2 and 3 are photomicrographs (secondary electron images) taken with a scanning electron microscope, both showing the electrode a3 in its state before charge and discharge. FIGS. 2 and 3 are taken at 2,000× and 5,000× magnifications, respectively.


The electrode was embedded in a resin and then sliced to provide a sample. The layers of the embedding resin are found in upper and lower end portions of FIG. 2 and in an upper end portion of FIG. 3.


In FIGS. 2 and 3, a portion that appears slilghtly light indicates the copper foil. The deposited silicon thin film (about 10 μm thick) is found as a dark portion on the copper foil. As shown in FIGS. 2 and 3, irregularities are formed on a surface of the copper foil. Particularly, projections have a generally conical shape. Similar irregularities are formed on a surface of the silicon thin film deposited on the copper foil. Accordingly, the surface irregularities of the silicon thin film appear to generally conform in shape to those formed on the copper foil surface.


Next, the electrode a3 was removed from the battery A3 after 30 cycles, embedded in a resin, and then subjected to observation under a scanning electron microscope in the same manner as described previously. Here, the electrode a3 was removed after discharge. Thus, the observed electrode a3 was in its state after discharge.



FIGS. 4 and 5 are photomicrographs (secondary electron images) taken with a scanning electron microscope, each showing the electrode a3 after discharge. FIGS. 4 and 5 are taken at 500× and 2,500× magnifications, respectively.


As shown in FIGS. 4 and 5, the silicon thin film has gaps that extend in its thickness direction and divide the silicon thin film into columns. Gaps are barely found to extend in a planar direction. A bottom of each columnar portion is found to adhere well to the copper foil as a current collector. Also, each columnar portion has a round top. It is thus understood that these gaps are formed to originate from the valleys of irregularities that were found on the surface of the silicon thin film in its state before charge and discharge.


Further, the surface of the silicon thin film of the electrode a3 after charges and discharges was observed with a scanning electron microscope. FIGS. 6 and 7 are photomicrographs (secondary electron images) taken with a scanning electron microscope, each showing the surface of the silicon thin film when viewed from above. FIGS. 6 and 7 are taken at 1,000× and 5,000× magnifications, respectively. FIGS. 8 and 9 are photomicrographs (secondary electron images) taken with a scanning electron microscope, each showing the surface of the silicon thin film when viewed at a slight angle. FIGS. 8 and 9 are taken at 1,000× and 5,000× magnifications, respectively.


As shown in FIGS. 6–9, the gaps are formed in such a way to surround the columnar portions of the silicon thin film so that spaces are defined between neighboring columnar portions. When the silicon thin film stores lithium on charge, the columnar portions will expand and increase in volume. This increment in volume, however, is believed to be accommodated by those spaces provided around the columnar portions. On discharge, the columnar portions of the silicon thin film release the lithium and shrink to decrease in volume. This decrement in volume is believed to restore the spaces around the columnar portions. Such a columnar structure of the silicon thin film is effective to relax a stress caused by expansion and shrinkage of the active material on charge and discharge, so that falling-off of the active silicon thin film from the current collector can be prevented.


The formation of the gaps which divide the silicon thin film into columns results in a marked increase in contact area thereof with the electrolyte solution. Also, the columnar portions are almost comparable in size to each other. These are believed to allow efficient occurrence of a charge-discharge reaction accompanying storage and release of lithium in the thin film of active material.


Since the individual columnar portions of the silicon thin film adhere to the current collector, as shown in FIGS. 4 and 5, good electrical connection is provided between the active material and the current collector. This is believed to allow effeicient occurrence of the charge-recharge reaction.


As also shown in FIGS. 6–9, each columnar portion has a round top. This provides an electrode structure which prevents localized current concentration and reduces the occurrence such as of a deposition reaction of a lithium metal.



FIG. 10 is a schematic sectional view, illustrating a process whereby the silicon thin film deposited on a copper foil is divided into columns by the gaps formed therein.


As shown in FIG. 10(a), the copper foil 10 has irregularities on its surface 10a. The copper foil with the increased value for surface roughness Ra has the larger irregularities.



FIG. 10(
b) illustrates a noncrystalline silicon thin layer 11 deposited on a rough surface 10a of the copper foil 10. The surface 11a of the silicon thin film 11 is influenced by the irregularities on the surface 10a of the copper foil 10 to have similar irregularities. Before charge and discharge, the silicon thin film 11 remains undivided, as shown in FIG. 10 (b). When charging is effected, the silicon thin film 11 stores lithium therein and expands in volume. During the charge, the silicon thin film 11 appears to expand in both thickness and planar directions of the thin film, although the detail is not clear. During the subsequent discharge reaction, the silicon thin film 11 releases lithium therefrom and shrinks in volume. At this time, a tensile stress is produced in the silicon thin film 11. Probably, such a stress concentrates at valleys 11b of the irregularities on the surface 11a of the silicon thin film 11 to result in the formation of gaps 12 that originate from the valleys 11b and extend in the thickness direction, as shown in FIG. 10 (c). Conceivably, the gaps 12 such formed relax the stress to allow the silicon thin film 11 to shrink without occurrence of falling-off from the copper foil 10.


In the silicon thin film divided into columns in the fashion as described above, the spaces provided around the columnar portions serve to relax the stress resulting from expansion and shrinkage of the active material during the succeeding charge-discharge cycles. This appears to assure repetitive charge-discharge cycling while preventing falling-off of the active material from the current collector.


Further, the electrode a3 incorporating an about 10 μm thick, microcrystalline silicon thin film deposited on an electrolytic copper foil was observed under a transmission electron microscope to study a mechanism whereby the gaps are formed in the silicon thin film. FIG. 11 is a photomicrograph (at a magnification of 12,500×) taken with a transmission electron microscope, showing a section of the electrode a3 before charge and discharge. The observation sample was prepared by slicing the resin-embedded electrode.



FIG. 13 is a diagrammatic representation of the photomicrograph of FIG. 11. In the photomicrograph of FIG. 11 taken with a transmission electron microscope, the silicon thin film 11 is deposited on the surface 10a of the electrolytic copper foil 10, as diagrammatically shown in FIG. 13. The silicon thin film 11 appears light relative to the copper foil 10 in the photomicrograph taken with a transmission electron microscope. In the silicon thin film shown in FIG. 11, light portions are observed in the regions extending between respective valleys 11b and 10b of the irregularities on the surfaces 11a and 10a of the silicon thin film 11 and copper foil 10. These light portions are indicated by single-dotted chain lines A, B and C in FIG. 13. Particularly, the light portion is observed more clearly in the region indicated by A. These regions are considered to be low in density, i.e., low-density regions of the silicon thin film 11. For the purpose of observing such low-density regions in more detail, an electrode a6 was fabricated by depositing an about 2 μm thick, microcrystalline silicon thin film on an electrolytic copper foil under the same conditions as used in the fabrication of the electrode a3.



FIG. 12 is a photomicrograph taken by a transmission electron microscope, showing the electrode a6 when observed in the same manner as described above. FIG. 12 is taken at a magnification of 25,000×. FIG. 14 is a diagrammatic representation of the photomicrograph of FIG. 12. As clear from FIG. 12, a low-density region is also observed in the region D of the electrode a6 that extends between the respective valleys 11b, 10b of the irregularities on the surfaces 11a, 10a of the silicon thin film 11 and the copper foil 10. A detailed observation of the photomicrograph of FIG. 12 reveals a number of fine lines extending in directions shown by the arrows in FIG. 14. It seems very likely that such lines are formed as the silicon thin film grows. It accordingly appears that the silicon thin film 11 grows generally perpendicularly to the surface 10a of the copper foil 10. It also appears that the silicon thin film layer grows in such a direction to collide at the region D with an adjacent silicon thin film layer being deposited and growing on the adjacent inclined surface of the copper foil. Thus the formation of the low-density region D is very likely to have resulted from such a collision at the region D. It also appears that the collision of the silicon thin film layers with each other is continued till the end of thin film formation, and formation of the low-density region also continues until reaching the surface of the silicon thin film.



FIG. 15 is a photomicrograph (secondary electron image) taken with a scanning electron microscope, showing a surface of a silicon thin film of an electrode a3 when observed from above. The electrode a3 shown in FIG. 15 is in its state before charge and discharge. FIG. 15 is viewed at 1,000× magnification. In FIG. 15, portions appearing lightened indicate projections on a surface of the silicon thin film and the surrounding portions appearing darkened indicate valleys on the surface of the silicon thin film. As shown in FIG. 15, the valleys on the surface of the silicon thin film are connected to each other like a network. It is accordingly found that the low-density regions define a continuous network in a plane of the silicon thin film. As shown in FIGS. 11 and 13, such a reticulated low-density region also extends in a thickness direction toward the current collector. The dark portions in FIG. 15 do not indicate the gaps (spaces). This is apparent from the fact that no gap (space) is observed which extends in the thickness direction of the thin film in the photomicrographs of FIGS. 2 and 3 taken by a scanning electron microscope.



FIG. 16 is a photomicrograph (secondary electron image) taken at a magnification of 1,000 using a scanning electron microscope, showing a surface of a silicon thin film, when observed from above, of an electrode a6 in its state before charge and discharge. As apparent from FIG. 16, the valleys in the electrode a6 are also connected to each other like a network. It is accordingly found that the low-density regions are arranged like a network continuous in a planar direction.



FIG. 17 is a graph showing concentration profiles of constituent elements along the thickness of the silicon thin film in the electrode a6. The concentration profiles of constituent elements were obtained via measurement of concentrations of copper (63Cu+) and silicon (Si2+) by SIMS using O2+ as a sputtering source. In FIG. 17, the abscissa indicates a depth (μm) from a surface of the silicon thin film and the ordinate indicates an intensity (count) of each consituent element.


As apparent from FIG. 17, a constituent element of the current collector, copper (Cu), is found to diffuse in the silicon thin film at locations adjacent to the current collector. The copper (Cu) concentration decreases at a location closer to the surface of the silicon thin film. The copper (Cu) concentration is also found to vary in a continous fashion. This demonstrates that a solid solution of silicon and copper, instead of an intermetallic compound thereof, is formed in the copper(Cu)-diffused region.


In view of the previous discussion, the folloiwng is very likely to explain a machanism whereby the gaps are formed in the silicon thin film to extend in its thickness direction as it expands and shrinks during charge and discharge. That is, a stress caused by expansion or shrinkage in volume of the silicon thin film concentrates at valleys of the irregularities on the silicon thin film surface, as previously explained by referring to FIG. 10. Also, in the silicon thin film, there initially exists low-density regions which are relatively low in mechanical-strength, extending from the valleys toward the current collector. As the result of the above-mentioned situations, the gaps (spaces) are likely to be formed along these low-density regions.


Also, as shown in FIG. 17, the diffusion of copper, a constituent element of the current collector, into the silicon thin film creates a concentration gradient of copper therein, so that the copper concentration is higher at a location closer to the current collector and lower at a location closer to the surface of the silicon thin film. Accordingly, a higher concentration of copper nonreactive with lithium and a lower concentration of silicon reactive with lithium exist at a location closer to the current collector. In the vicinity of the current collector, the silicon thin film is believed to store and release less lithium, undergo less expansion and shrinkage, and thus produce a lower level of stress which leads to the reduced formation of the gaps (spaces) which may occasion separation or removal of the silicon thin film from the current collector. As a result, the bottoms of the columnar portions of the silicon thin film can be kept adherent to the current collector.


The silicon thin film divided by such gaps into columns keeps a strong adhesion to the current collector even during charge-discharge cycles. Also, the spaces provided to surround the columnar portions serve to relax the stress caused by expansion and shrinkage of the thin film that occur with charge-discharge cycling. These are believed to be contributors to excellent charge-discharge cycle characteristics.


REFERENCE EXPERIMENT 4

(Fabrication of Electrode a7)


An electrolytic copper foil similar to that for use in the fabrication of the electrode a3 was used for a current collector as a substrate. An amorphous germanium thin film (about 2 μm thick) was formed on this copper foil by an RF sputtering technique to fabricate an electrode a7.


The thin film was formed using germanium as a target, at a sputtering gas (Ar) flow rate of 100 sccm, an ambient substrate temperature (not heated), a reaction pressure of 0.1 Pa, and 200 W RF power.


The resulting germanium thin film was analyzed by Raman spectroscopy which detected the presence of a peak around 274 cm−1 and the absence of a peak around 300 cm−1. This revealed an amorphous nature of the germanium thin film.


(Fabrication of Electrode a8)


An amorphous germanium thin film (about 2 μm thick) was formed on an electrolytic copper foil, similar in type to the current collector of the electrode a7, by using a vapor evaporation technique to fabricate an electrode a8.


Specifically, the germanium thin film was deposited on the substrate by utilizing an apparatus of the construction shown in FIG. 18. Referring to FIG. 18, an ECR plasma source 21 includes a plasma generation chamber 22 to which a microwave power 25 and an Ar gas 26 are supplied. An Ar plasma is generated when the microwave power 25 is supplied to the plasma generation chamber 22. This Ar plasma 23 is directed to exsit from the plasma generation chamber 22 and bombard a substrate 20. The germanium thin film can be deposited on the substrate 20 by utilizing an electron beam from an electron beam (EB) gun disposed below the substrate 20.


The electrolytic copper foil substrate was pretreated by Ar plasma irradiation before the germanium thin film was deposited thereon. A degree of vacuum within the reaction chamber was adjusted to about 0.05 Pa (about 5×10−4 Torr) The substrate was exposed to the Ar plasma under conditions of an Ar gas flow rate of 40 sccm and a supplied microwave power of 200 W. During the Ar plasma irradiation, an bias voltage of −100 V was applied to the substrate. The pretreatment was accomplished by exposing the substrate to the Ar plasma for 15 minutes.


Next, the germanium thin film was deposited at a deposition rate of 1 nm/sec (10 Å/sec) using an electron beam gun. The substrate temperature was ambient temperature (not heated).


The resulting germanium thin film was analyzed by Raman spectroscopy which revealed an amorphous nature of the germanium thin film, as similar to the electrode a7.


(Fabrication of Electrode b2)


80 parts by weight of germanium powder with a mean particle diameter of 10 μm, 10 parts by weight of acetylene black as an electroconductive material, and 10 parts by weight of polytetrafluoroethylene as a binder were mixed. This mixture was pressed in a 17 mm diameter mold to fabricate a pellet-form electrode b2.


(Construction of Batteries)


The procedure of Experiment was repeated, except that the above-fabricated electrodes a7, a8 and b2 were used for the negative electrode, to construct batteries A7, A8 and B2.


(Evaluation of Charge-Discharge Cycle Characteristics) Each battery was charged at a current of 0.1 mA at 25° C. to 4.2 V, and then discharged to 2.75 V. This standard charge-discharge cycling was repeated to measure a capacity retention rate on the 10th cycle. The measurement results are given in Table 5.












TABLE 5







Battery
10th-Cycle Capacity Retention Rate









A7
96%



A8
93%



B2
39%










As apparent from Table 5, the batteries A7 and A8 using the electrodes in accordance with this invention, i.e., the electrodes incorporating the germanium thin film formed on the current collector, for the negative electrode exhibit markedly improved capacity retention rates compared to the battery B2 using the germanium powder for the negative electrode.


(Observation With Electron Microscope)



FIGS. 19 and 20 are photomicrographs (reflection electron images) taken with a scanning electron microscope, each showing a section of the electrode a7 in its state before being charged and discharged. FIGS. 19 and 20 were taken at magnifications of 2,000× and 10,000X, respectively.


Each electrode was resin embedded and then sliced to provide a sample. The embedding resin is observed as layers located in upper and lower end portions of FIG. 19 and in an upper end portion of FIG. 20.


In FIGS. 19 and 20, the copper foil and the germanium thin film appear lightened relative to the rest. A thin layer overlying the copper foil is the germanium thin film. Irregularities are defined on a surface of the copper foil. Similar irregularities are also found on a surface of the germanium thin film. This suggests that the irregularities on the germanium thin film surface were formed to conform in shape to those defined on the copper foil surface.


In FIG. 20, there is observed a dark portion that is located in a germanium thin film region overlying a leftmost valley of the copper foil and extends in a thickness direction of the thin film. This portion is very likely to indicate a region of low density, i.e., a low-density region of the germanium thin film.



FIGS. 21 and 22 are photomicrographs (reflection electron images) taken with a scanning electron microscope, each showing a section of the electrode a8 in its state before being charged and discharged. FIGS. 21 and 22 are taken at magnifications of 2,000× and 10,000×, respectively. Like the electrode a7 shown in FIGS. 19 and 20, a sample of this electrode is embedded in a resin.


In FIGS. 21 and 22, a lightened portion indicates a copper foil and a slightly darker portion carried thereon is a germanium thin film (about 2 μm thick). Irregularities are defined on both surfaces of the germanium thin film and the copper foil of the electrode a8, as analogous to the electrode a7.



FIGS. 23 and 24 are photomicrographs (reflection electron images) taken with a scanning electron microscope, each showing a section of the electrode a7 removed from the battery A7 after 10 cycles. FIGS. 25 and 26 are photo-micrographs (reflection electron images) taken with a scanning electron microscope, each showing a section of the electrode a8 removed from the battery A8 after 10 cycles. In either case, the electrode was resin embedded and then sliced to provide a sample. FIGS. 23 and 25 are both taken at a magnification of 500×. FIGS. 24 and 26 are both taken at a magnification of 2,500×.


In FIGS. 23–26, a portion which appears white on the surface of the germanium thin film is gold coated thereon before it is embedded in a resin. The coating of gold is provided to prevent any reaction which may occur between the germanium thin film and the resin and also define a clear boundary therebetween.


As can be clearly seen from FIGS. 23–26, charge-discharge cycling causes formation of gaps which extend in a thickness direction of the germanium thin film and divide the thin film into columns, as similar to the case of the silicon thin film. Although a small difference in contrast between the copper foil, as a current collector, and the germanium thin film makes it difficult to distinguish a boundary therebetween, the careful observation reveals the presence of columnar portions of the germanium thin film over projections of the current collector and thus good adhesion of the germanium thin film to the current collector.


Unlike the case of silicon thin film, laterally-extending gaps are also observed in the germanium thin film. It is very likely, however, that such gaps were formed when the germanium thin film was polished before proceeding to sectional observation.


Also, the width of a gap (space) between neighboring columnar portions is found to be larger in the germanium thin film than in the silicon thin film. After charge-discharge cycling, the height of the columnar portions measured about 6 μm, which is about three times the initial film thickness of the germanium thin film, 2 μm, before the charge-discharge cycling. This is considered to indicate that when the thin film shrinks on discharge after it has expaneded due to storage of lithium during charge, the shrinkage occurs mainly in a lateral direction, i.e., in a planar direction. It is accordingly believed that the wide gaps (spaces) between the columnar portions result from a small percent shrinkage of the germanium thin film in its thickness direction.



FIGS. 27 and 28 are photomicrographs (secondary electron images) taken with a scanning electron microscope, each showing a germanium thin film of the electrode a7 in its state after charges and discharges, when observed from above. FIGS. 27 and 28 are taken at magnifications of 1,000× and 5,000×, respectively. FIGS. 29 and 30 are photomicrographs (secondary electron images) taken with a scanning electron microscope, each showing the germanium thin film of the electrode a7 in its state after charges and discharges, when observed at a slight angle. FIGS. 29 and 30 are taken at magnifications of 1,000× and 5,000×, respectively.



FIGS. 31 and 32 are photomicrographs (secondary electron images) taken with a scanning electron microscope, each showing a germanium thin film of the electrode a8 in its state after charges and discharges, when observed from above. FIGS. 31 and 32 are taken at magnifications of 1,000× and 5,000×, respectively. FIGS. 33 and 34 are photomicrographs (secondary electron images) taken with a scanning electron microscope, each showing the germanium thin film of the electrode a8 in its state after charges and discharges, when observed at a slight angle. FIGS. 33 and 34 are taken at magnifications of 1,000× and 5,000×, respectively.


As shown in FIGS. 27–34, gaps (spaces) are formed in such a fashion to surround the columnar portions of the germanium thin film to thereby define spaces between the neighboring columnar portions. It is belived that these spaces serve to relax the stress caused by expansion and shrinkage of the active material during charge and discharge, as also described in the previous case of silicon thin film.



FIGS. 35 is a photomicrograph (secondary electron images) taken with a scanning electron microscope, showing a surface of the germanium thin film of the electrode a7 in its state before charge and discharge, when observed from above. FIGS. 36 is a photomicrograph (secondary electron images) taken with a scanning electron microscope, showing a surface of the germanium thin film of the electrode a8 in its state before charge and discharge, when observed from above. FIGS. 35 and 36 are both taken at a magnification of 1,000×.


As shown in FIGS. 35 and 36, the germanium thin film has irregularities on its surface that follow the profile of those defined on the underlying electrolytic copper foil. The valleys of the germanium thin film are connected to each other like a network. It is understood that the gaps extend along the depth of such valleys to define columnar portions in the germanium thin film.


(SIMS Analysis of Concentration Profile Along Depth)



FIG. 37 is a graphical representation illustrating concentration profiles of constituent elements in the electrode a7 along its depth before it is incorporated in a battery, i.e., before charge and discharge. FIG. 38 is a graphical representation illustrating concentration profiles of constituent elements in the electrode a8 along its depth before charge and discharge. The concentration profiles of constituent elements were obtained by a secondary ion mass spectrometry (SIMS) wherein copper (63Cu) and germanium (73Ge) concentrations were measured along the depth from a surface of the thin film using O2+ as a sputtering source. The abscissa indicates a depth (μm) from a surface of the germanium thin film and the ordinate indicates an intensity (count) of each consituent element.


As can be clearly seen from FIGS. 37 and 38, copper (Cu), as a current collector constituent, diffuses into the germanium thin film in the vicinity of the current collector and shows a lower concentration at a location closer to the surface of the germanium thin film.


As discussed above, the germanium thin film contains a current collector constituent, copper, diffused therein, has a higher copper concentration in the vicinity of the current collector, and has a concentration gradient such that a copper concentration becomes lower at a location closer to its surface. Hence, the germanium thin film in the vicinity of the current collector contains a higher concentration of copper unreactive with lithium and a lower concentration of germanium reactive with lithium. In the vicinity of the current collector, the germanium thin film is thus believed to store and release less lithium, undergo less expansion and shrinkage, and produce a lower level of stress. This probably results in the reduced formation of gaps (spaces) which may cause separation or removal of the gemanium thin film from the current collector, so that the bottoms of the columnar portions of the germanium thin film can be kept adhered to the current collector.


As stated above, the germanium thin film in conditions of being divided in columns maintains strong adhesion to the current collector even during charge-discharge cycling. Also, the gaps formed in a fashion to surround the colomnar portions serve to relax the stress caused by expansion and shrinkage during charge-discharge cycles. The excellent charge-discharge cycle characteristics are thus obtained.


REFERENCE EXPERIMENT 5

(Fabrication of Electrode a9)


An electrolytic copper foil (18 μm thick) was used for a current collector as a substrate. A silicon thin film was formed on this electrolytic copper foil by an RF sputtering technique. The sputtering was effected at a sputtering gas (Ar) flow rate of 100 sccm, an ambient substrate temperature (not heated), a reaction pressure of 0.1 Pa (1.0×10−3 Torr) and a 200 W RF power. The silicon thin film was deposited to a thickness of about 2 μm.


The resulting silicon thin film was analyzed by Raman spectroscopy which detected the presence of a peak around 480 cm−1 and the absence of a peak around 520 cm−1. This reveals an amorphous nature of the silicon thin film.


The electrolytic copper foil after silicon thin film deposition thereon was cut into a 2 cm×2 cm size to prepare an electrode a9.


The surface roughness Ra and the mean spacing S of the electrolytic copper foil used were measured using a stylus profiler Dektat3 ST (available from ULVAC Inc.) with a scanning distance of 2.0 mm. The surface roughness Ra and the mean spacing S were determined to be 0.188 μm and 11 μm, respectively.


(Fabrication of Electrode a10)


An electrolytic copper foil similar to that used in the fabrication of the electrode a9 was used for a current collector as a substrate. A silicon thin film was formed on the electrolytic copper foil under the same conditions as used in the fabrication of the electrode a1 of Reference Experiment 1 to a thickness of about 2 μm. The procudere used to prepare the electrode a9 was followed to prepare an electrode a10.


The resulting silicon thin film was analyzed by Raman spectroscopy which detected the presence of peaks around 480 cm−1 and 520 cm−1. This reveals a microcrystalline nature of the silicon thin film. (Fabrication of Comparative Electrode b3) A rolled copper foil for use in the above Reference Experiment 1 was utilized for a current collector as a substrate. The procedure used to fabricate the electrode a9 was followed to form an amorphous silicon thin film (about 2 μm thick) on the rolled copper foil by an RF sputtering technique.


The resulting amorphous silicon thin film was subjected to one-hour annealing at 650° C. The annealed silicon thin film was then analyzed by Raman spectroscopy which revealed disappearance of a peak around 480 cm−1 and detected a sole peak around 520 cm−1. This demonstrates that the annealing results in the formation of a polycrystalline silicon thin film.


The procedure used to prepare the electrode a9 was followed to prepare an electrode b3 from the polycrystalline silicon thin film formed on the rolled copper foil.


The above-described procedure was utilized to measure the surface roughness Ra and the mean spacing S for the rolled copper foil. The rolled copper foil exhibited the surface roughness Ra of 0.037 μm and the mean spacing S of 14 μm.


(Measurement of Charge-Discharge Characteristics)


Each of the above-fabricated electrodes a9, a10 and b3 was used as a work electrode. Metallic lithium was used for both a counter electrode and a reference electrode. Using these electrodes, experimental cells were constructed. The electrolyte solution was identical to that used in the above Reference Experiment 1. In a single electrode cell, reduction of the work electrode is a charge reaction and oxidation thereof is a discharge reaction.


Each experimental cell was charged at a constant current of 0.5 mA at 25° C. until a potential relative to the reference electrode reached 0 V, and then discharged to 2 V. This was recorded as a unit charge-discharge cycle. Cycling was effected to measure 1st- and 5th-cycle discharge capacities and charge-discharge efficiencies. The results are given in Table 6.













TABLE 6







Electrode a9
Electrode a10
Electrode b3



















Substrate
Electrolytic
Electrolytic
Rolled Copper



Copper Foil
Copper Foil
Foil


Thickness of Silicon Thin Film
2 μm
2 μm
2 μm


Annealing
Absent
Absent
650° C., 1 h


Crystallinity of Silicon Thin Film
Amorphous
Microcrystalline
Polycrystalline












Charge-Discharge
1st Cycle
Discharge Capacity
3980
4020
1978


Characteristics

(mAh/g)




Charge-Discharge
100
96
83




Efficiency (%)



5th Cycle
Discharge Capacity
3990
4020
731




(mAh/g)




Charge-Discharge
100
100
75




Efficiency (%)









As apparent from the results shown in Table 6, the electrode a9 using the amorphous silicon thin film for the electrode active material and the electrode a10 using the microcrystalline silicon thin film for the electrode active material, in accordance with the present invention, exhibit higher discharge capacities and superior charge-discharge efficiencies even on the 5th cycle, relative to the comparative electrode b3.


REFERENCE EXPERIMENT 6
EXAMPLES 1–7 AND COMPARATIVE EXAMPLES 1–2

(Fabrication of Current Collector)


Samples 14 specified in Table 7 were used for a current collector as a substrate. Sample 1 was similar to the rolled copper foil used for the current collector of the electrode b3. Samples 24 were prepared in accordance with the following procedure: A rolled copper foil was ground with a #100, #400 or #1000 emery paper to roughen its surface, washed with a purified water and then dried.













TABLE 7





Sample No.
1
2
3
4



















Copper Foil Thickness (μm)
18
18
18
18


Surface Roughness Ra (μm)
0.037
0.1
0.18
1









Each of the above copper foils was used as a substrate. A silicon thin film was deposited on the substrate under the conditions specified in Tables 8–10 by means of an RF argon sputtering apparatus. In Comparative Example 2, the deposited thin film was subsequently subjected to a heat treatment (annealing). In Examples 1–7 and Comparative Example 1, each substrate was pretreated prior to thin film deposition. The pretreatment was performed by generating an ECR plasma in a separately-installed plasma generator and directing the plasma to bombard the substrate for 10 minutes at a 200 W microwave power and an argon gas partial pressure of 0.06 Pa.


The nature of each silicon thin film was identified via analysis by Raman spectroscopy. The results are shown in Tables 8–10.


(Measurement of Charge-Discharge Characteristics)


The silicon-deposited copper foils obtained in Examples 1–7 and Comparative Examples 1–2 were cut into 2 cm×2 cm pieces and then utilized to construct experimental cells in the same manner as in the above Reference Experiment 5. For each cell, the charge-discharge test was performed in the same manner as in the above Reference Experiment 5 to measure 1st-, 5th- and 20th-cycle discharge capacities and charge-discharge efficiencies. The results are shown in Tables 8–10.














TABLE 8







Ex. 1
Ex. 2
Ex. 3
Ex. 4





















Substrate
Substrate type
Sample 2
Sample 3
Sample 4
Sample 3



Surface Roughness Ra
0.1
0.18
1
0.18



Substrate Thickness
18 μm
18 μm
18 μm
18 μm


Thin Film Forming
Thickness of Silicon Thin Film
 2 μm
 2 μm
 2 μm
 2 μm


Conditions
Film-Forming Process
Sputtering
Sputtering
Sputtering
Sputtering



Sputtering Gas
Argon
Argon
Argon
Argon



Ar Flow Rate
100 sccm
100 sccm
100 sccm
100 sccm



Target
99.999% Si
99.999% Si
99.999% Si
99.999% Si




Single Crystal
Single Crystal
Single Crystal
Single Crystal



Sputtering Atomosphere
0.10 Pa
0.10 Pa
0.10 Pa
0.10 Pa



Sputtering Power
200W
200W
200W
200W



Substrate Temperature
20° C.
20° C.
20° C.
200° C.



Pretreatment
Present
Present
Present
Present



Sputtering Time
2 h
2 h
2 h
2 h


Heat Treatment
Heat Treatment
Absent
Absent
Absent
Absent


Conditions
Heat Treating Time






Identification of
Raman Peak at 480 cm−1
Present
Present
Present
Present


Crystallinity
Raman Peak at 520 cm−1
Absent
Absent
Absent
Absent



Crystallinity
Amorphous
Amorphous
Amorphous
Amorphous


1st Cycle
Discharge Capacity (mAh/g)
3980
3978
3975
3980



Charge-Discharge Efficiency (%)
100
100
100
100


5th Cycle
Discharge Capacity (mAh/g)
3990
3981
3980
3990



Charge-Discharge Efficiency (%)
100
100
100
100


20th Cycle
Discharge Capacity (mAh/g)
3990
3980
3981
3990



Charge-Discharge Efficiency (%)
100
100
100
100




















TABLE 9







Ex. 5
Ex. 6
Ex. 7




















Substrate
Substrate type
Sample 3
Sample 3
Sample 3



Surface Roughness Ra
0.18
0.18
0.18



Substrate Thickness
18 μm
18 μm
18 μm


Thin Film Forming
Thickness of Silicon Thin Film
 2 μm
 2 μm
 2 μm


Conditions
Film-Forming Processing
Sputtering
Sputtering
Sputtering



Sputtering Gas
Argon
Argon
Argon



Ar Flow Rate
100 sccm
100 sccm
100 sccm



Target
99.999% Si
99.999% Si
99.999% Si




Single Crystal
Single Crystal
Single Crystal



Sputtering Atomosphere
0.10 Pa
1.0 Pa
10 Pa



Sputtering Power
200W
200W
200W



Substrate Temperature
50° C.
20° C.
20° C.



Pretreatment
Present
Present
Present



Sputtering Time
2 h
1.5 h
2.5 h


Heat Treatment
Heat Treatment
Absent
Absent
Absent


Conditions
Heat Treating Time





Identification of
Raman Peak at 480 cm−1
Present
Present
Present


of Crystallinity
Raman Peak at 520 cm−1
Absent
Absent
Absent



Crystallinity
Amorphous
Amorphous
Amorphous


1st Cycle
Discharge Capacity (mAh/g)
4060
3585
2500



Charge-Discharge Efficiency (%)
100
100
100


5th Cycle
Discharge Capacity (mAh/g)
4060
3592
2505



Charge-Discharge Efficiency (%)
100
100
100


20th Cycle
Discharge Capacity (mAh/g)
4060
3590
2505



Charge-Discharge Efficiency (%)
100
100
100



















TABLE 10







Comp. Ex. 1
Comp. Ex. 2



















Substrate
Substrate type
Sample 3
Sample 1



Surface Roughness Ra
0.18
0.037



Substrate Thickness
18 μm
18 μm


Thin Film Forming
Thickness of Silicon Thin Film
 2 μm
 2 μm


Conditions
Film-Forming Process
Sputtering
Sputtering



Sputtering Gas
Argon
Argon



Ar Flow Rate
100 sccm
100 sccm



Target
99.999% Si
99.999% Si




Single Crystal
Single Crystal



Sputtering Atomosphere
0.10 Pa
0.10 Pa



Sputtering Power
200W
200W



Substrate Temperature
450° C.
20° C.



Pretreatment
Present
Absent



Sputtering Time
2 h
2 h


Heat Treatment
Heat Treatment
Absent
650° C.


Conditions
Heat Treating Time

1 h


Identification of
Raman Peak at 480 cm−1
Absent
Absent


Crystallinity
Raman Peak at 520 cm−1
Present
present



Crystallinity
Polycrystalline
Polycrystalline


1st Cycle
Discharge Capacity (mAh/g)
1250
1978



Charge-Discharge Efficiency (%)
81
83


5th Cycle
Discharge Capacity (mAh/g)
900
731



Charge-Discharge Efficiency (%)
75
75


20th Cycle
Discharge Capacity (mAh/g)
700
350



Charge-Discharge Efficiency (%)
69
59









As can be clearly seen from the results shown in Tables 8–10, the increased discharge capacities and improved charge-discharge cycle characteristics are obtained by utilizing the electrodes obtained via Examples 1–7 that use the amorphous silicon thin film for the electrode active material in accordance with the present invention, relative to utilizing the electrodes obtained via Comparative Examples 1–2 that use the polycrystalline silicon thin film for the electrode active material.


REFERENCE EXPERIMENT 7

An amorphous silicon thin film (about 3 μm thick) was formed on an electrolytic copper foil (18 μm thick, surface roughness Ra=0.188 μm, mean spacing S=6 μm) by an RF sputtering technique to fabricate an electrode a-11. The thin film was deposited using single-crystal silicon as a target, at a sputtering gas (Ar) flow rate of 100 sccm, an ambient substrate temperature (not heated), a reaction pressure of 0.1 Pa, and 200 W RF power.


The resulting silicon thin film was analyzed by Raman spectroscopy which detected the presence of a peak around 480 cm−1 and the absence of a peak around 520 cm−1. This revealed an amorphous nature of the silicon thin film.


The electrode a-11 thus obtained was used to construct a battery A11 in the same manner as in the above Reference Experiment 1. The battery was subjected to a charge-discharge cycle test under the same conditions as in the above Reference Experiment 1 to measure a capacity retention rate on the 30th-cycle. The result is shown in Table 11. In Table 11, the results for the batteries A1 and A3 are also shown.












TABLE 11







Battery
30th-Cycle Capacity Retention Rate









A1
91%



A3
97%



 A11
97%










As apparent from the results shown in Table 11, the battery A11 using the sputter deposited amorphous silicon thin film for the active material also exhibits a good capacity retention rate comparable to those of the batteries A1 and A3 using the microcrystalline silicon thin film for the active material.


The condition of the silicon thin film in the electrode a-11 was observed using an electron microscope. First, a section of the electrode a-11 in its state before charge and discharge was observed with a scanning electron microscope. FIGS. 39 and 40 are photomicrographs (secondary electron images) taken with a scanning electron microscope, each showing a section of the electrode a-11 before charge and discharge. FIGS. 39 and 40 are taken at magnifications of 2,000× and 10,000×, respectively. A sample was prepared by following the procedure used to prepare the samples shown in FIGS. 2 and 3, i.e., by embedding the electrode in a resin and then slicing the resin-embedded electrode.


In FIGS. 39 and 40, a portion that appears relatively light indicates the electrolytic copper foil. The deposited silicon thin film (about 3 μm thick) is found as a dark portion on the copper foil. As shown in FIGS. 39 and 40, irregularities are defined on a surface of the electrolytic copper foil. Particularly, projections have a generally conical shape. Similar irregularities with such conical projections are also formed on a surface of the silicon thin film deposited on the copper foil. Accordingly, the surface irregularities of the silicon thin film appear to conform in shape to those defined on the copper foil surface.



FIG. 41 is a photomicrograph (secondary electron image) taken with a scanning electron microscope, showing a surface of the silicon thin film in the electrode a-11 when viewed at a magnification of 1,000×. As shown in FIG. 41, a number of projections is formed on the silicon thin film surface. As shown in FIGS. 39 and 40, these projections are formed in such a way to follow those defined on the copper foil surface.



FIG. 42 is a photomicrograph (reflection electron image) taken with a scanning electron microscope, showing a surface of the electrode a-11 removed from the battery A11 after 30 cycles in the charge-discharge test. FIG. 42 is a photograph taken at a magnification of 1,000×.


As shown in FIG. 42, gaps (spaces) are formed in the silicon thin filim to extend in its thickness direction and these gaps (spaces) divide the silicon thin film into columns. In the silicon thin film shown in FIGS. 6–9, the gaps are formed in such a way as to define columnar portions each encompassing a single projection on the thin film. On the other hand, in the silicon thin film shown in FIG. 42, the gaps are formed in such a way as to define columnar portions each encompassing plural projections on the thin film. It is also found that the gaps (spaces) are wider in the silicon thin film shown in FIG. 42 than in the silicon thin film shown in FIGS. 6–9.


The battery A11 exhibits a good capacity retention in a manner similar to the battery A3. This is believed to demonstrate that the spaces provided in a way to surround the columnar portions serve to relax the stress caused by expansion and shrinkage of the active material so that charge-discharge cycling can be repeated without occurrence of separation of the active material from the current collector, even in the case where each columnar portion is defined to encompass plural projections on the thin film surface.


REFERENCE EXPERIMENT 8

An about 2 μm thick, microcrystalline silicon thin film was formed on both a rolled copper foil and an electrolytic copper foil (18 μm thick) under the same thin film-forming conditions as used in the fabrication of electrode a1 in Reference Experiment 1. Then, a 17 mm diameter piece was punched out from each sample to provide an electrode c1 incorporating the silicon thin film formed on the rolled copper foil and an electrode c3 incorporating the silicon thin film formed on the electrolytic copper foil. Pieces identical to the electrodes c1 and c3 were heat treated at 400° C. for 3 hours to provide electrodes c2 and c4, respectively.


The procedure of Reference Experiment 1 was followed, except that the electrodes c1–c4 were used for the negative electrode, to construct rechargeable lithium batteries C1–C4. These batteries were measured for charge-discharge cycle life characteristics in the same manner as in Reference Experiment 1. Also, a hydrogen content, a ratio of Raman peak intensities (480 cm−1/520 cm−1) and a crystal grain size were measured for the silicon thin film of each electrode in the same manner as in Reference Experiment 1. The results are shown in Table 12.













TABLE 12






50th-Cycle






Capacity

Ratio of Peak



Retention
Hydrogen
Intensities
Crystal


Battery
Rate
Content
(480 cm−1/520 cm−1)
Grain Size







C1
90%
  4%
0.1
1 nm


C2
85%
0.01%
0.1
1 nm


C3
91%
  4%
0.1
1 nm


C4
87%
0.01%
0.1
1 nm









As demonstrated by the results shown in Table 12, the markedly high capacity retention rates are also obtained for the batteries C1–C4 with the about 2 μm thick microcrystalline silicon thin film.


The electrode c1 incorporating the microcrystalline silicon thin film formed on the rolled copper foil was sliced in its thickenss direction to provide a sample which was subsequently observed with a transmission electron microscope.



FIGS. 43 and 44 are photomicrographs taken with a transmission electron microscope, showing an interface between the copper foil and the silicon thin film and its vicinities in the electrode c1. FIGS. 43 and 44 are taken at magnifications of 500,000× and 1,000,000×. The copper foil is found in a lower portion and the silicon thin film in an upper portion of each photomicrograph.


In FIGS. 43 and 44, a lightened lower portion appears to be a copper foil portion. A portion located in the vicinity of the interface between the copper foil and silicon thin film appears darkened toward the above. This portion (about 30 nm–about 100 nm) seems to be a part of a mixed layer where diffusion of copper from the copper foil into silicon is particularly significant. In this mixed layer, copper (Cu) is probably alloyed with silicon (Si). Also in FIGS. 43 and 44, a particulate portion is observed in the vicinity of an interface between the seeming mixed layer and the copper foil. This particulate portion is found to define an irregular profile along the interface as a result of the diffusion of copper (Cu) into silicon (Si).


Next, concentration profiles of constituent elements along the depth of the mixed layer were observed. For this purpose, the concentrations of copper (63Cu+) and hydrogen (1H+) were measured by SIMS using O2+as a sputtering source. FIG. 45 shows a concentration profile of each constituent element. The abscissa indicates a depth (μm) and the ordinate indicates an atomic density (number of atoms/cm3)


As shown in FIG. 45, the concentration of copper (Cu) in the mixed layer increases at a deeper location, i.e., at a location closer to the copper foil. If the mixed layer is defined as a layer in the silicon thin film that contains at least 1% (1020 atoms/cm3, if expressed in atomic density) of a current collector material, the mixed layer is found to exist in a thickness region which extends from a depth of about 1.9 μm to a depth of about 2.7 μm.


Similarly, a concentration profile of each constituent element along the depth of the mixed layer were observed using SIMS for the the electrode c3 incorporating the about 2 μm thick microcrystalline silicon thin film formed on the electrolytic copper foil. The results are shown in FIG. 46. As shown in FIG. 46, the atomic density of copper already exceeds 1020 atoms/cm3 at the surface of the silicon thin film in the electrode c3. This clearly indicates that the copper diffused across the the silcon thin film to its surface to render the silicon thin film into the form of mixed layer in its entirety. Also, the battery C3 using this electrode c3 exhibits good charage-discharge cycle characteristics. This demonstrates that the silicon thin film still serves as electrode active material, even if it is rendered into the form of mixed layer in its entirety.


As can be clearly seen from FIGS. 45 and 46, the copper concentration varies continuously across the silicon thin film. This accordingly demonstrates that copper exists in the silicon thin film not in the form of an intermetallic compound with silicon but in the form of a solid solution with silicon.


As discussed above, it is ascertained that the mixed layer where copper in the copper foil is mixed with silcon in the silicon thin film is formed at the interface between the copper foil and silicon thin film. The presence of this mixed layer is believed to improve adhesion of the silicon thin film to the copper foil, prevent separation of the silicon thin film from the copper foil as a substrate even if the silicon thin film is subjected to expansion and shrinkage on charge and discharge, and provide good charge-discharge cycle characteristics.


EXPERIMENT A

In the electrode for a rechargeable lithium battery in accordance with the present invention, a stress is produced in the current collector during a charge-discharge reaction since the thin film of active material expands and shrinks as it stores and releases lithium. Such a stress causes the current collector to undergo an irreversible or plastic defomation, resulting in the formation of wrinkles therein. The formation of such wrinkles leads to a volume increase of the battery and a nonuniform electrode reaction, and gives rise to reduction in energy density. In view thereof, the thickness of the electolytic copper foil, as a negative current collector, was varied to study the relation of the tensile strength of the negative current collector to wrinkles formed in the electrode.


(Fabrication of Negative Electrode)


Four types of electrolytic copper foils, i.e., 12 μm, 18 μm, 35 μm and 70 μm thick electrolytic copper foils were used. The thickness of each electrolytic copper foil, as a negative current collector, was measured using a micrometer. The tensile strength (N/m) of each current collector can be calculated by (tensile strength per sectional area of the current collector material: (N/mm2))×(thickness of the current collector: mm), where the tensile strength per sectional area of copper was given as 212.7 N/mm (21.7 kgf/mm2, “Data Book of Metals, Revised 2nd Ed.”, published by Maruzen Co.).


A silicon thin film was formed on each electrolytic copper foil in an Ar atmosphere by an RF sputtering technique. The thin film was deposited using single-crystal silicon as a target, at 350 W RF power, an Ar gas flow rate of 100 sccm, an in-chamber pressure of 0.1 Pa, and an ambient substrate temperature (not heated).


The silicon thin film deposited under the above-specified film-forming conditions was analyzed by Raman spectroscopy which revealed an amorphous nature of the silicon thin film. In each case, the silicon thin film was 3.4 μm thick. The thickness of the silicon thin film deposited on the electrolytic copper foil was determined by measuring, with a surface roughness meter, a thickness of a silicon thin film deposited on a silicon substrate which was placed, together with the electrolytic copper foil, in the chamber. Specifically, in determining the thickness of the silicon thin film, a surface roughness meter was utilized to measure a difference in level at an edge of the silicon thin film by tracing with the probe.


In the formation of the silicon thin film, a mask was utilized to limit deposition of silicon within a 2.5 cm×2.5 cm area on the copper foil. A negative tab was attached onto a copper foil portion that was left uncoated with silicon to complete a negative electrode.


(Fabrication of Positive Electrode)


The LiCoO2 powder obtained in the course of fabricating the positive electrode in Reference Experiment 1 was used to fabricate a positive electrode. In specific, 90 parts by weight of LiCoO2 powder and 5 parts by weight of artificial graphite powder as a conducting material were mixed in a 5 wt. % aqueous N-methylpyrrolidone solution containing 5 parts by weight of polytetrafluoroethylene as a binder to provide a mix slurry for positive electrode. This slurry was coated on a 2 cm×2 cm area of an aluminum foil (18 μm thick) as a positive current collector by a doctor blade technique and then dried to form a layer of positive active material. The amount of the slurry coated was adjusted so that the positive electrode had a 15.75 mAh capacity. A positive tab was attached onto an aluminum foil region that was left uncoated with the layer of positive active material to complete a positive electrode.


(Preparation of Electrolyte Solution)


1 mole/liter of LiPF6 was dissolved in a mixed solvent containing equi-volumes of ethylene carbonate and dimethyl carbonate to prepare an electrolyte solution for use in the following battery construction.


(Construction of Battery)



FIGS. 47 and 48 are a perspective view and a schematic sectional view, respectively illustrating the rechargeable lithium battery as constructed. As shown in FIG. 48, a casing 30 made of a laminated aluminum film encloses the positive and negative electrodes. Silicon as negative active material is deposited on a negative current collector 31 to form a silicon thin film 32 thereon. A layer 34 of positive active material is provided on a positive current collector 33. The silicon thin film 32 and the layer 34 of positive active material are disposed on opposite sides of a separator 35. The casing 30 stores the above-prepared electrolyte solution 36. The casing 30 is heat sealed at edges to define a sealed region 30a. A negative tab 37 attached to the negative current collector 31 extends through this sealed region 30a to outside. A positive tab 38 attached to the positive current collector 33 also extends through the sealed region 30a to outside, although not shown in FIG. 48.


(Charge-Discharge Cycle Test)


The rechargeable lithium batteries constructed in the manner as described above were subjected to a charge-discharge cycle test. Each battery was charged at a current of 9 mA to a charge end capacity of 9 mAh and then discharged at a current of 9 mA to a discharge end voltage of 2.75 V. This unit charge-discharge cycle was repeated 10 times to measure a charge-discharge efficiency on each cycle. 3 samples (1C-1, 1C-2 and 1C-3) were prepared for each of the batteries incorporating the respective copper foils different in thickness from each other. The measurement results thereof are given in FIG. 49.


As indicated in FIG. 49, the change in thickness of the copper foil does not provide a noticeable effect on the charge-discharge cycle characteristics. Regardless of the thickness of the copper foil used, the battery provides good charge-discharge cycle characteristics.


After 10 cycles, the negative electrode was removed from each battery to observe its condition. FIG. 50 and FIG. 51 show conditions of the top surfaces of negative electrodes where the respective silicon thin films were deposited. FIG. 52 and FIG. 53 show conditions of the back surfaces of negative electrodes where the respective silicon thin films were not deposited. In FIGS. 50–53, “12”, “18”, “35” and “70” each indicates a thickness value of the copper foil.


As apparent from FIGS. 52 and 53, the electrode was observed to be fully wrinkled when it used the 12 μm thick copper foil, slightly wrinkled when it used the 18 μm thick copper foil, and little wrinkled when it used the 35 or 75 μm thick copper foil. These results are summarized in Table 13. In Table 13, evaluation was made according to the following standards.


×: fully wrinkled.


◯: slightly wrinkled.


⊚: little wrinkled.













TABLE 13





Thickness of Current Collector (μm)
12
18
35
70



















Tensile Strength of Current Collector
2.552
3.828
7.443
14.886


(N/mm)


Thickness of Thin Film of Active
0.28
0.19
0.10
0.049


Material/Thickness of Current Collector


Evaluation
x












As can be seen from Table 13, the electrode shows a rapid decrease in number of wrinkles formed as the tensile strength of the current collector reaches and exceeds 3.82 N/mm, and becomes little wrinkled as it reaches and exceeds 7.44 N/mm. Also, the electrode has been found to show a rapid decrease in number of wrinkles formed as a ratio in thickness of the silicon thin film to the current collector falls to 0.19 and below, and becomes little wrinkled as it falls to 0.10 and below. This is probably because the current collector, if its tensile strength reaches and exceeds a specific level, undergoes elastic deformation that almost relaxes the stress caused by expansion and shrinkage of the thin film of active material.


EXPERIMENT B

Next, the thickness of the silicon thin film deposited on the electrolytic copper foil which served as a negative current collector was varied to study the relation of the tensile strength of the negative current collector per 1 μm thickness of the silicon thin film to the wrinkles formed in the electrode.


(Fabrication of Negative Electrode)


An electrolytic copper foil of 18 μm thick was used for the negative current collector. The thickness of the electrolytic copper foil was measured using a micrometer and the tensile strength (N/mm) of the current collector was determined in the same manner as in Experiment A.


An amorphous silicon thin film was deposited on the electrolytic copper foil under the same film-forming conditions as used in Experiment A. The thickness of the silicon thin film was varied to provide three types of thin films having different thicknesses; 0.9 μm, 1.8 μm and 3.6 μm. The thickness of each silicon thin film was determined according to the procedure described in Experiment A.


Like Experiment A, a negative tab was attached onto the copper foil to complete a negative electrode.


(Construction of Battery)


The procedures described in Experiment A were followed to fabricate a positive electrode, prepare an electrolyte solution and construct a rechargeable lithium battery.


(Charge-Discharge Cycle Test)


The above-constructed rechargeable lithium batteries incorporating the negative electrodes with the respective silicon thin films different in thickness from each other were subjected to a charge-discharge cycle test under the same conditions as in Experiment A. For each battery, the previously defined unit charge-discharge cycle was repeated 30 times to measure a charge-discharge efficiency at each cycle. 3 samples were prepared for each of the batteries incorporating the negative electrodes with the respective silicon thin films different in thickness from each other. The measurement results thereof are given in FIG. 54. In FIG. 54, “0.9 μm”, “1.8 μm” and “3.6 μm” each indicates a thickness value of the silicon thin film and the indicated charge-discharge efficiency at each cycle is an arithmetic mean of those obtained from three samples.


As apparent from FIG. 54, the change in thickness of the silicon thin film does not provide a noticeable effect on the charge-discharge cycle characteristics. Regardless of the thickness of the silicon thin film, the battery provides good charge-discharge cycle characteristics.


Each battery was subjected to additional 10 cycles of charge and discharge. After a total of 40 cycles, the negative electrode was removed from each battery to observe its condition. FIG. 55 shows conditions of back surfaces of negative electrodes, i.e., their surfaces on which the respective silicon thin films were not deposited. In FIG. 55, “0.9”, “1.8” and “3.6” each indicates a thickness value of the silicon thin film.


As apparent from FIG. 55, the electrode after charges and discharges was observed to be fully wrinkled when it used the 3.6 μm thick silicon thin film, slightly wrinkled when it used the 1.8 μm thick silicon thin film, and little wrinkled when it used the 0.9 μm thick silicon thin film. These results are summarized in Table 14. In Table 14, evaluation was made according to the same standards as used in Table 13.












TABLE 14







Thickness of Silicon Thin Film (μm)
0.9
1.8
3.6


Thickness of Current Collector (μm)
18
18
18


Thickness of Silicon Thin Film/Thickness of
0.0500
0.1000
0.2000


Current Collector


Tensile Strength of Current Collector
3.828
3.828
3.828


(N/mm)


Tensile Strength of Current Collector per μm
4.253
2.127
1.063


Thickness of Silicon Thin Film (N/mm)


Evaluation


x









These results, as well as the results given in Table 13 for Experiment A, are also listed in Table 15, in the order of decreasing tensile strength of the current collector per μm thickness of the silicon thin film.
















TABLE 15







Thickness of Silicon Thin Film (μm)
3.4
0.9
3.4
1.8
3.4
3.6
3.4


Thickness of Current Collector (μm)
70
18
35
18
18
18
12


Thickness of Silicon Thin Film/Thickness of
0.0486
0.0500
0.0971
0.1000
0.1889
0.2000
0.2833


Current Collector


Tensile Strength of Current Collector
14.886
3.828
7.443
3.828
3.828
3.828
2.552


(N/mm)


Tensile Strength of Current Collector per
4.378
4.253
2.189
2.127
1.126
1.063
0.751


μm Thickness of Silicon Thin Film (N/mm)


Evaluation





x
x









As can be clearly seen from Table 15, the electrode shows a rapid decrease in number of wrinkles formed as the tensile strength of the current collector per μm thickness of the silicon thin film reaches and exceeds 1.12 N/mm, shows a further decrease in number of wrinkles to an unnoticeable level as the tensile strength of the negative current collector per μm thickness of the silicon thin film reaches and exceeds 2.18 N/mm. The electrodes denoted by “35” in FIG. 53 and by “0.9” in FIG. 55 are both evaluated as “⊚”. However, the detailed comparison thereof reveals that the electrode denoted by “0.9” in FIG. 55 is less wrinkled. This demonstrates that the electrode becomes further less wrinkled as the tensile strength of the negative current collector per 1 μm thickness of the silicon thin film reaches and exceeds 4.25 N/mm.


The electrode has been found to show a rapid decrease in number of wrinkles formed as a ratio in thickness of the silicon thin film to the current collector falls to 0.19 and below, becomes less wrinkled as it falls to 0.098 and below, and becomes further less wrinkled as it falls to 0.05 and below.


It is believed from the foregoing that the negative current collector, if its tensile strength per 1 μm thickness of the thin film of active material exceeds a specific level, undergoes elastic deformation that almost relaxes the stress caused by expansion and shrinkage of the thin film of active material, thereby reducing the degee of wrinkle formation. Similarly, it is believed that, if a ratio in thickness of the thin film of active material to the negative current collector falls below a specific level, the current collector undergoes elastic deformation that almost relaxes the stress caused by expansion and shrinkage of the thin film of active material, thereby reducing the degee of wrinkle formation.


UTILITY IN INDUSTRY

In accordance with the present invention, an electrode for a rechargeable lithium battery is provided which has a high charge-discharge capacity, exhibits excellent charge-discharge cycle characteristics, and can prevent formation of wrinkles in the electrode during charge and discharge.

Claims
  • 1. An electrode for a rechargeable lithium battery which includes a thin film composed of active material that expands and shrinks as it stores and releases lithium, deposited on a current collector, characterized in that said current collector exhibits a tensile strength×thickness of not less than 3.82 N/mm, a constituent of said current collector is diffused into said thin film of active material, the thin film of active material is divided into columns by gaps formed therein and extending in a thickness direction of said film from a surface of said film toward said current collector such that said columns extend in the thickness direction of the thin film and are separated by said gaps, and said columnar portions are at their bottoms adhered to the current collector where, the tensile strength×thickness of the current collector is determined in this and subsequent claims by the following equation:tensile strength×thickness of the current collector (N/mm)=tensile strength (N/mm2) per sectional area of the current collector material×thickness (mm) of the current collector.
  • 2. The electrode for a rechargeable lithium battery as recited in claim 1, characterized in that said current collector exhibits a tensile strength×thickness of not less than 7.44 N/mm.
  • 3. The electrode for a rechargeable lithium battery as recited in claim 1, characterized in that said current collector has a surface roughness Ra in the range of 0.01–1 μm.
  • 4. The electrode for a rechargeable lithium battery as recited in claim 1, characterized in that said current collector comprises a metal foil roughened at its surface.
  • 5. The electrode for a rechargeable lithium battery as recited in claim 1, characterized in that said current collector comprises a copper foil.
  • 6. The electrode for a rechargeable lithium battery as recited in claim 5, characterized in that said copper foil is an electrolytic copper foil.
  • 7. The electrode for a rechargeable lithium battery as recited in claim 1, characterized in that said current collector comprises a metal foil on which a copper layer is deposited.
  • 8. The electrode for a rechargeable lithium battery as recited in claim 7, characterized in that said metal foil is a nickel foil.
  • 9. The electrode for a rechargeable lithium battery as recited in claim 8, characterized in that said nickel foil is an electrolytic nickel foil.
  • 10. The electrode for a rechargeable lithium battery as recited in claim 1, characterized in that said thin film of active material contains silicon or germanium.
  • 11. The electrode for a rechargeable lithium battery as recited in claim 1, characterized in that said thin film of active material is a silicon, germanium, or silicon-germanium alloy thin film.
  • 12. The electrode for a rechargeable lithium battery as recited in claim 1, characterized in that said thin film of active material is a microcrystalline or amorphous silicon thin film.
  • 13. The electrode for a rechargeable lithium battery as recited in claim 1, characterized in that said thin film of active material is a microcrystalline or amorphous germanium thin film.
  • 14. The electrode for a rechargeable lithium battery as recited in claim 1, characterized in that said thin film of active material is a microcrystalline or amorphous silicon-germanium alloy thin film.
  • 15. The electrode for a rechargeable lithium battery as recited in claim 1, characterized in that said gaps are formed via a charge-discharge reaction of said thin film of active material.
  • 16. The electrode for a rechargeable lithium battery as recited in claim 1, characterized in that a first concentration of said current collector constituent in the thin film of active material is higher in the vicinity of the current collector and a second concentration of said current collector constituent in the thin film of active material is lower at a location closer to a thin film surface.
  • 17. The electrode for a rechargeable lithium battery as recited in claim 1, characterized in that, in said thin film of active material, the diffused current collector constituent forms a solid solution with a component of the thin film of active material.
  • 18. The electrode for a rechargeable lithium battery as recited in claim 1, characterized in that said thin film of active material is deposited by a CVD, sputtering, vapor evaporation, spraying or plating process.
  • 19. The electrode for a rechargeable lithium battery as recited in claim 1, characterized in that said thin film of active material stores lithium via formation of an alloy with lithium.
  • 20. The electrode for a rechargeable lithium battery as recited in claim 1, characterized in that lithium is previously stored or incorporated in said thin film of active material.
  • 21. A rechargeable lithium battery including a negative electrode composed of the electrode as recited in claim 1, a positive electrode and a nonaqueous electrolyte.
  • 22. The rechargeable lithium battery as recited in claim 21, characterized in that said positive electrode contains, as active material, an oxide capable of lithium storage and release.
  • 23. The rechargeable lithium battery as recited in claim 21, characterized in that said positive electrode contains, as active material, a lithium-containing oxide.
Priority Claims (10)
Number Date Country Kind
11/301646 Oct 1999 JP national
11/357808 Dec 1999 JP national
11/365306 Dec 1999 JP national
11/374512 Dec 1999 JP national
2000/039454 Feb 2000 JP national
2000/047675 Feb 2000 JP national
2000/090583 Mar 2000 JP national
2000/105506 Apr 2000 JP national
2000/142755 May 2000 JP national
2000/207274 Jul 2000 JP national
PCT Information
Filing Document Filing Date Country Kind 371c Date
PCT/JP00/07295 10/20/2000 WO 00 8/1/2002
Publishing Document Publishing Date Country Kind
WO01/31723 5/3/2001 WO A
US Referenced Citations (14)
Number Name Date Kind
4997732 Austin et al. Mar 1991 A
5108852 Tomantschger et al. Apr 1992 A
5431803 DiFranco et al. Jul 1995 A
5494762 Isoyama et al. Feb 1996 A
5631100 Yoshino et al. May 1997 A
5641591 Kawakami et al. Jun 1997 A
5888670 Kawakami Mar 1999 A
6048646 Xing et al. Apr 2000 A
6402795 Chu et al. Jun 2002 B1
6432585 Kawakami et al. Aug 2002 B1
6743369 Jito et al. Jun 2004 B2
6746802 Tamura et al. Jun 2004 B2
20020192564 Ota et al. Dec 2002 A1
20040224231 Fujimoto et al. Nov 2004 A1
Foreign Referenced Citations (55)
Number Date Country
1092208 Sep 1994 CN
0 854 529 Jul 1998 EP
0 855 752 Jul 1998 EP
0 910 131 Apr 1999 EP
60-175375 Sep 1985 JP
63-314762 Dec 1988 JP
5-74479 Mar 1993 JP
5-101830 Apr 1993 JP
5-226004 Sep 1993 JP
6-140045 May 1994 JP
6-168737 Jun 1994 JP
6-187994 Jul 1994 JP
7-302588 Nov 1995 JP
8-50922 Feb 1996 JP
8-255610 Oct 1996 JP
9-17441 Jan 1997 JP
9-115523 May 1997 JP
09115523 May 1997 JP
9-147856 Jun 1997 JP
9-213339 Aug 1997 JP
10-21928 Jan 1998 JP
10-199524 Jul 1998 JP
10-208740 Aug 1998 JP
10-223221 Aug 1998 JP
10-294098 Nov 1998 JP
10-326628 Dec 1998 JP
11-73946 Mar 1999 JP
11-86854 Mar 1999 JP
11-86875 Mar 1999 JP
11-135115 May 1999 JP
11-135130 May 1999 JP
11-154508 Jun 1999 JP
11-167930 Jun 1999 JP
11-185744 Jul 1999 JP
11-233116 Aug 1999 JP
11-242954 Sep 1999 JP
11-250892 Sep 1999 JP
11-273683 Oct 1999 JP
11-283628 Oct 1999 JP
11-339811 Dec 1999 JP
2000-11997 Jan 2000 JP
2000-12088 Jan 2000 JP
2000-12089 Jan 2000 JP
2000-12091 Jan 2000 JP
2000-149928 Jan 2000 JP
2000012088 Jan 2000 JP
2000-100429 Apr 2000 JP
2000-133276 May 2000 JP
2000-182623 Jun 2000 JP
2000-208136 Jul 2000 JP
2000-228187 Aug 2000 JP
2000-294249 Oct 2000 JP
2000-294250 Oct 2000 JP
2000-294251 Oct 2000 JP
2001-76710 Mar 2001 JP