CA1302491C

CA1302491C - Lithium-lithium nitride anode

Info

Publication number: CA1302491C
Application number: CA000561735A
Authority: CA
Inventors: C. David Desjardins; Hossein Sharifian; Gregory K. Maclean
Original assignee: New Brunswick Telephone Co Ltd
Current assignee: New Brunswick Telephone Co Ltd
Priority date: 1987-03-04
Filing date: 1988-03-17
Publication date: 1992-06-02
Anticipated expiration: 2009-06-02
Also published as: EP0281352B1; DE3852188T2; EP0281352A3; DE3852188D1; EP0281352A2; JPS63294666A; AU1276688A; ATE114872T1

Abstract

ABSTRACT OF THE DISCLOSURE An anode material comprising a solid solution of lithium metal and lithium nitride can be used in lithium-based electrochemical cells. The anode material can incorporate lithium alloys, carbon and polymeric materials.

Description

LITHIUM - LITHIUM NITRIDE ANODE

Background of the Invention The invention relates to the use of a mixture of lithium and nitrogen as anode material in rechargeable and non-rechargeable high energy density batteries.
The prohlems associated with the use of pure lithium metal anodes in electrochemical systems, and, in particular, in rechargeable cells have been well. documented in the scientiEic and technical literature (see, for example, U.S. Patents 4,550,064, 4,499,161, 4,118,550, 4,071,664, 4,086,403). The fundamental problem is the thermodynamic and kinetic instability of lithium towards cell materials and the electrolyte especially at the anode/electrolyte interEace. In electrochemical cells, this results in corrosion/passivation of the lithium electrode. Thus, attempts to recycle lithium result in dendritic growth, formation of poorly adherent high surface area deposits of lithium metal on chaxging and slow chemical decomposition of the electrolyte. Ultimately, cell failure occurs due to cell shorting, depletion and/or isolation of the anode and anode passivation Erom electrolyte breakdown products. Short term solutions to the above problems have been ~3~
to use excess lithium metal and/or conductive/protective films.
Another approach has been to use composite electrodes like lithium-aluminum alloy. In such cases, an immobile host material promotes a uniform distribution of lithium having reduced chemical activity. Lithium aluminum alloys have been extensively investigated in nonaqueous electrolyte systems.
E~owever, the electrode experiences large volume changes during deposition/stripping associated with phase transformations.
Such electrodes, on extensive recycling, exhibit severe mechanical instability. Other disadvantages associated with the use of alloys include large voltage and capacity losses relative to pure lithium anodes.

Summary of the Invention It has been found that the above-mentioned problems with anodes made of lithium or lithium-aluminum alloy can be reduced or avoided by the use of anodes made of a mixture of lithium and nitrogen. This invention is therefore primarily directed to lithium-based electrochemical cells having a new anode material which comprises a solid state mixture of lithium metal and nitrogen wherein the nitrogen is in the form of lithium nitride. Anodes comprising the solid mixture or solution formed from this mixture, (said mixture or solution being sometimes referred to in this specification as "Li/Li3N") are intended to be used to replace the lithium anodes in lithium-based electrochemical cells. The terms "mixture" and "solution" are used interchangeably in this specification; the characterization of the solid anode material as a "mixture" or a "solution"
is immaterial. The remaining elemen-ts of lithium-based electrochemical cells, such as -the cathode, separator, and electrolyte, can be used wi~hou-t modification in electrochemical cells made according to this invention, having anodes of Li/Li3N. Such elements have been extensively described in the lithium battery literature.
Mixtures of lithium and lithium nitride have been identified in the scientific literature (e.g. U.S. Patent 4,447,379 to Wagner) as intermediates in the preparation of a solid electrolyte and lithium nitride has been used as a solid ion conductor in solid state cells as reviewed by A. Rabenau (Solid State Ionics, 6 (1982) 277). Surface films of Li3N on lithium foil anodes have also been investigated by Thevenin et al. (Lawrence Berkeley Lab., CA (USA), Jan 1986, Report No.
LBL 20659). However, it has not hitherto been known to use Li/Li3N as an anode material.
The new anode material has reduced chemical/
electrochemical activity towards electrolyte especially at the anode/electrolyte interface, can recycle lithium at very high efficiencies and exhibits no cell voltage losses. Li/Li3N
significantly reduces the rate of interaction between elemental lithium and electrolyte during charge/discharge cycling and in storage. Therefore, an improved lithium-based anode material formed of lithium and lithium nitride, employable with a suitable electrolyte, separator and cathode for use in an electrochemical cell, and, in particular, a lithium rechargeable cell, is contemplated.
Lithium nitride is known to have the highest Li+ ion conductivity of any inorganic lithium salt and has been extensively studied as a solid electrolyte in solid state cells. However, the relatively low decomposition potential of lithium nitride (Ed=0.44V) has limited its commercial use in solid state cells. By contrast, lithium nitride is thermodynamically stable to lithium metal and, in combination with lithium metal, forms a very stable anode material.

Brief Description of the Drawings Figure 1 is the polarization curve of Li/Li3N
(11 mole%N) in lM LiAsF6/2-MeTHF electrolyte at room temperature.
Figures 2 and 3 illustrate the activity of Li/Li3N
(11 mole%N) and pure lithium electrodes during storage time in lM LiAsF6/2-MeTHF based electrolyte and lM LiAsF6/THF
electrolyte respectively at room temperature.
Figure 4 illustrates the exchange current of Li/Li3N (11 mole%N) and pure lithium electrodes in LiAlC14/thionyl chloride electrolyte at room temperature.
Figure 5 illustrates the extent of passivation of pure lithium and Li/Li3N (11 mole%N) during anodic step in a lM LiAsF6 dioxolane-based electrolyte at -20C.

~3~

Figure 6 illustrates the cycling per-formance of lithium and Li/L,i3N electrodes (11 mole~N) in a lM
LiAsF6/2-MeTHF electrolyte at room temperature.

Detailed Description of the Preferred Embodi.ments -Anode material according to the invention can be prepared by melting lithium metal foil in a crucible in an argon-filled dry box, adding nitrogen in the form of lithium nitride powder, nitrogen gas or o-ther suitable nitro~en-containing material to form a homogeneous material, and coolingthe material to room temperature~ The anode material so formed comprises a solid solution oE lithium metal and lithium nitride. Whether nitrogen is added to liquid lithium in the form of lithium nitride, nitrogen gas, or any other nitrogen-containing substance that will release its nitrogen in molten lithium, it has been found that the nitrogen in the solution or mixture is in the form of lithium nitride. The exact composition of the Li/Li3N material is dependent on the process temperature and coolirg rate. The temperature dependency is partially defined by the material's phase diagram (see P.F. Adams et al. J. of the Less-Common Metals, 42 (1975) -1-11, 325-334). Even small amounts of Li/Li3N in the anode are considered an improvement over a pure lithium anode, and electrochemical cells employing such anodes are within the scope of this invention.
In the specific case of the 11 mole%N Li/Li3N
mixture used for the data represented herein, approximately 2g Li3N powder was slowly added to 2g of molten lithium metal in a nickel crucible at approximately 300C. The material was stirred for several minutes until homogeneous and was then cooled to room temperature. The resulting anode material was compressed at about 9000 lb/cm2 into the desired shape.
The anode material can be rolled or extruded into a foil, which is considered a preferred form for use in electrochemical cells.
A 1.12 mole % nitrogen solid solution anode material was prepared by slowly dissolving 6.34g of Li3N in 10~.21g of pure molten lithium at a temperature of approximately 445C.
The molten solution was then poured in-to a mold having a large heat sink to permit rapid cooling. The material solidified almost immediately and the mold was transferred to an extruder for further processing.
In the anode material according to the present invention, the proportion of Li3N in the mixture can be at any effective level. At levels at which the proportion of N in the mixture varies from about 0.06 to 20 mole percent, Li and Li3N form a solid solution at elevated -temperatures, yielding a satisfactory, homogeneous anode material. Within this broad range, a narrower range of about 0.1 to 10 mole percent N is considered preferable.
An electrochemical system that utilizes the Li/Li3N
anode includes a separator, an electrolyte, and a cathode.
Suitable separator materials include those commonly employed in lithium cells such as porous polypropylene and ~3~

glass microfibre materials that are sufficiently porous and wettable by the appropriate electrolyte.
Suitable nonaqueous solvent systems for the electrolyte include all solvents normally employable in lithium cells with particular emphasis on aprotic ether or sulphur analogue systems, preferably dioxolanes, furans, glymes, methoxy methanes, glycol sulfites, sulfolanes, propylene carbonate and combinations thereof. Suitable lithium salts are selected based on their high solubility (e.g.=0.5M) and conductivity in the chosen solvent system but would include, more specifically, anions such as the perchlorates, sulfonates, acetates, borates, closoboranes, hexafluoro or tetrafluoro metallates, halides, aluminates and derivatives thereof. It should be emphasized that normal lithium foil anode dependencies and/or constraints on the choice of an electrolyte (e.g. solute/solvent combinations) do not apply for systems employing the Li/Li3~ anode material. That is, a much wider choice of electrolyte(s) may be used. Some of these electrolytes show considerably improved performance characteristics relative to standard lithium anode materials.
Solid electrolytes such as lithium nitride, lithium iodide and lithium aluminum nitride, as well as polymeric ionic conductors, or combinations thereo-f, may also be used in the present invention.
The use of electrolytes based on inorganic solvents such as sulfur dioxide, thionyl chloride or sulfuryl chloride, ~3~

with an appropriate conductive lithium salt, is also contemplated.
A sultable cathode includes any of those described in the technical and patent literature for lithium-based electrochemical systems. For example cathodes comprising a halogen or halide, a metal oxide, sulphides, selenides, oxyhalides, sulfur dioxide and carbon can be used. Cathodes having an active material comprising a chalcogen or chalcogenide compound of a transition metal are also suitable.
In the case of rechargeable cells, the use of intercalation-type materials and the transition metal sulphides (e.g. ~iS2 or MS2) or oxides (e.g. V6O13), is preferred.
Anodes made aceording to the invention may be comprised solely or substantially solely of Li/Li3N, or they may be eomprised of composite materials, of which Li/Li3N is one component. For example, anode compositions comprising plastic or elastomeric macromoleeular material with ionie conduction, lithium alloys and carbon compounds are described in U.S. Patent 4,517,265 to Belanger et al. Similar anode eompositions may be made aceording to the present invention whieh would inelude Li/Li3N as a component.
Similarly, it is known to use alloys of lithium and aluminum as an anode material (e.g. U.S. Patent 4,002,492 to Rao). Anodes may be made according to the present invention using mixtures of Li/Li3~ and lithium-aluminum alloys, or mixtures of lithium nitride and lithium-aluminum alloys.

~3~Z4~

The use of other lithium alloys as anode materials ls also known in the art, and -the use of mixtures of lithium nitride and one or more lithium alloys such as Li-Si, Li-Sn, Li-Fe, Li-Sb, Li-Bi, Li-B, Li-Pb, Li-As, etc. and other combinations thereof as anode materials is contemplated according to the present invention.
As shown in Figure 1 of -the drawings, the polarization curve of Li/Li3N at room temperature demonstrates that upon the anodic step the current rises to high values. Despite the relatively low decomposition potential of Li3N (0.40V), the cell utilizing Li/Li3N can be discharged up to 7.0 mA/cm2 ln a majority of aprotic solvents commonly used in lithium rechargeable cells. The polarization curve also indicates -that Li/Li3N anodes cycled at low current densities will show only minor overpotentials.
In these cases, the overpotentials are less than the decomposition potential of Li3N.
Figure 2 demonstrates the relative activity of lithium metal and Li/Li3N during storage time at room temperature. In general, the formation of insulating layers on the electrode during storage periods causes an associated reduction of its active surface. The evolution of the corresponding activity loss has been followed by measuring the ratio it/io (io is the exchange current at initial time, it is the exchange current after a specified time later) by means of AC impedance spectroscopy at the end o-f initial treatment (io) and after a standing time (t) on open circuit ~3~

(i.t). The curves in Fiyure 2 show that -the chemical stability of the lithium versus the electrolyte decreases severely with time, while Li/Li3N shows an excellent stability versus solvent under similar experimental conditions.
A similar experiment was carried out employing a relatively more reactive solvent such as tetrahydro-furan ("THF"). It is well known that THF is too reactive for use in secondary cells. Even very pure THF reacts rapidly with lithium and formation of a gel-like film. It is readily apparent from Figure 3 that the surEace activity of Li/Li3N
decreases moderately during a 50 hour storage time at room temperature, while the surface of a pure lithium electrode became entirely inactive during the same period.
These physical phenomenon support the conclusion that Li/Li3N offers a significantly longer shelf life, in which an immobile host material such as Li3N promotes a uniform lithium distribution of reduced activity. This is particularly true for the lithium electrodes in primary cells where one constantly faces problems resulting from the presence of passivating layers at the surface of the lithium metal which limit the transport of the active materials to the interface.
Moreover, excessive passivation of the pure lithium anode entails a long delay time for the attainment of a steady state cell voltage on initial discharge. All these problems can be alleviated markedly by employing a Li/Li3N anode, where substantially reduced chemical activity is observed.

~L3~
Moreover, cells employing Li/Li3N anodes could be extremely important in present day primary cell technology.
One of the most promising of these is the lithium thionyl chloride cell, in whictl LiAlC14 is added to the liquid thionyl chloride in order to increase the conductivity and facilitate Li~ ion transport. The high stability of lithium in LiAlC14/SOCl~ solution is due to a protecting surface film of LiCl which is formed as the electrode makes contact with the electrolyte. While the film makes it possible to construct a battery from a thermodynamically unstable combination such as Li in SOC12, an attendent voltage delay must be overcome in order to have a practical system.
In this connection, 1,i/Li3N electrodes have been studied in thionyl chloride and sulfur dioxide solutions.
Figure 4 depicts the percentage of the exchange current of the cell employing Li and Li/Li3N (11 mole%N) in thionyl chloride/LiAlC14 (1.6M) during storage at room temperature.
The magnitude of exchange current is indicative of the surface activity of the electrode which is in turn directly related to the thickness of the passive film formed on the electrode surface. It is readily apparent in Figure 4 that the exchange current of the lithium electrode decreases with increasing storage time, while the cell utilizing a Li/Li3N electrode exhibits a greater exchange current under similar experimental conditions.
The LiCl film which grows continuously on lithium causes an initial voltage delay, as has been reported previously. It is further believed that the initial voltage delay of the Li/SOC12 cell depends on the effective LiCl film thickness growth rate. Thus, the initial voltage delay of a cell utilizing a Li/Li3N anode would be significantly smaller than the cell utilizing a Li metal anode. This is particularly true for longer storage times, where the Li cell suffers significantly from the continuous growth of LiCl film on the lithium with a corresponding gradual increase in the cell's initial voltage delay.
Figure 5 illus-trates the activity of lithium metal and Li/Li3N during anodic stripping at 1.0 mA/cm2 at -20C.
The formation of insulating layers at the lithium metal surface is readily apparent in Figure 5. These layers have been attributed to solvent polymerization initiated by LiAsE'6 decomposition produc-ts. Deterioration o-f the electrochemical properties of the metal/electrolyte interface due to the formation of passivating layers could be ascribed to a significant ohmic drop overpotential resulted from lower conductivity as well as higher viscosity of the electrolyte at lower temperatures. Consequentlyl this limits the amount of lithium metal in each cycle. It is further evident from Figure 5 that the extent of passivation of Li/Li3N is not as severe as that for lithium electrodes at -20C during an anodic stripping at 1.0 mA/cm2, despite the greater ohmic drop of the cell utilizing Li/Li3N.
Figure 6 shows the discharge capacity of lithium and Li/Li3N as a function of cycle number at room temperature.

The cathode was made from an intercalation material, namely TiS2. The cell was completed with a glass fiber separator soaked with l.OM LiAsF6 in 2-methyltetrahydrofuran. The results of the cycling experiment demonstrate the higher theoretical capacity of the Li/Li3N electrode relative to pure lithium anodes after cycling. In further testing, the cell utiliæing a pure lithium anode completed only 50 cycles at 30~ of its theoretical capacity, while the cell utilizing a Li/Li3N anode completed 125 cycles at 40% of its theoretical capacity.
From the above data, it may be appreciated that the present invention provides a viable solution to numerous problems associated with the use o-f pure lithium metal anodes in lithium galvanic cells and, in particular, li-thium rechargeab,e cells. A suggested explanation for the superior properties of Li/Li3N as an anode material relates to the formation of micropores at the surface of the electrode during discharge processing. The porous Li/Li3N electrode formed on discharge may act as a three dimensional electrode, thereby reducing significantly the local current density and eliminating the mass redistribution of lithium on extending cycling. Furthermore, morphological studies performed on Li/Li3~ by means of a scanning electron microscope have shown that anodic polarization effectively cleans the anode surface.
The microporous structure o~ the substrate was clearly revealed after the electrode surface was anodically polarized at about +50 mV overpotential and 10 coulombs/cm2 charge density.

During the cathodic step, the lithium began to grow on a clean surface. It is suggested that the improvement in the electrode properties is directly linked to the presence of microcavities generated at the electrode surface during the initial anodic step. Moreover, the incorporation of lithium into the Li/Li3N anode during subsequent cathodic steps would take place through these microcavities. However, the dimension of the cavities only ensures the passage of small molecules such as lithium while bulkier solvent molecules are not able to penetrate into the cavities. Based on this model, the first anodic polarization plays a determinant role in the quality of the electrode's properties, and generation of cavities which permit the passage of lithium ions alone and not the solvent or solute molecules. As part of this model, it is further contemplated that lithium nitride in addition to acting as a host species may also promote lithium ion conductivity within the anode~ In addition, preliminary lithium NMR studies indicate complete delocalization of lithium electrons which averages out and lowers the total lithium reactivity. It is believed that a combination of all these effects results in improved cycle and longer shelf life as observed in the studies of the Li/Li3N electrode.
While the invention has been described in detail herein in accord with certain preferred embodiments thereof, many modifications and changes therein may be effected by those skilled in the art. Accordingly, it is intended by the ; - 14 ~

~3~Z~

appended claims to cover all such modi-fications and changes as fall within the true spirit and scope of the invention.

Claims

WHAT IS CLAIMED IS:

1. An electrochemical cell comprising:
(a) an anode comprising a solid solution of lithium metal and lithium nitride, wherein the proportion of nitrogen in said solution is in the range of about 0.1 to 20 mole percent;
(b) a cathode;
(c) a non-aqueous electrolyte comprising a solvent and a lithium salt dissolved therein; and (d) a porous separator.

2. An electrochemical cell in accordance with claim 1 wherein the proportion of nitrogen in the solution is in the range of about 1.12 to 20 mole percent.

3. An electrochemical cell comprising:
(a) an anode comprising a solid solution of lithium nitride and an alloy selected from the group consisting of Li-Al, Li-Si, Li-Sn, Li-Fe, Li-Sb, Li-Bi, Li-B and Li-Pb, wherein the proportion of nitrogen in the said solution is in the range of about 0.1 to 20 mole percent;
(b) a cathode;
(c) a non-aqueous electrolyte comprising a solvent and a lithium salt dissolved therein; and (d) a porous separator.

4. An electrochemical cell in accordance with claim 1 wherein the proportion of nitrogen in the solution is in the range of about 1.12 to 20 mole percent.

5. An electrochemical cell according to claim 1, 2, 3 or 4 wherein said porous separator comprises polypropylene.

6. An electrochemical cell according to claim 1, 2, 3 or 4 wherein said porous separator comprises glass microfiber material.

7. An electrochemical cell according to claim 1, 2, 3 or 4 wherein said lithium salt has an anion selected from the group consisting of halide ions, hexafluorometallate ions, tetrafluorometallate ions, perchlorate ions, sulfonate ions, borate ions, thiocyanate ions, aluminate ions, closoborane ions and acetate ions and derivatives thereof.

8. An electrochemical cell according to claim 1, 2, 3 or 4 wherein said solvent is an organic solvent selected from the group consisting of ethers, dioxolanes, furans, glymes, glycol sulfites, sulfolanes and carbonates.

9. An electrochemical cell according to claim 1, 2, 3 or 4 wherein the cathode active material comprises a chalcogen or chalcogenide compound of a transition metal.

10. An electrochemical cell according to claim 1, 2, 3 or 4 wherein said cathode comprises a substance selected from the group comprising a hologen, a halide, a metal oxide, a sulphide, a selenide, an oxyhalide, sulfur dioxide and carbon.

11. An electrochemical cell according to claim 1, 2, 3 or 4 wherein said electrolyte comprises a conductive, lithium salt and an inorganic solvent selected from the group consisting of sulfur dioxide, thionyl chloride and sulfuryl chloride.

12. An improved lithium-based electrochemical cell of the type having an anode, a cathode, an electrolyte and a separator, wherein the improvement comprises an anode comprising lithium and lithium nitride, said anode being formed by adding nitrogen to molten lithium and cooling and shaping the resulting material, said material having a proportion of nitrogen in the range of about 0.1 to 20 mole percent.

13. An anode for use in a lithium-based electrochemical cell comprising a solid solution of lithium metal and lithium nitride, where the proportion of nitrogen in said solution is in the range of about 0.1 to 20 mole percent, and wherein said anode is in the form of a foil.

14. An anode according to claim 19 wherein the proportion of nitrogen in the solution is in the range of about 1.12 to 20 mole percent.