In Japan, some micro-batteries using solid electrolytes, such as silver ion, copper ion and lithium ion conductors are now under development for applying them to micro-devices, integrated-circuits and very-large-scaleintegration. For these purposes, inorganic crystalline and glassy electrolytes as well as polymeric electrolytes have been developed. For electric energy storage batteries, sodium ion conductors such as beta- or beta"- alumina have been developed as the electrolyte. A 1 MW sodium-sulfur battery demonstration power plant is to be constructed in Japan by the fall of 1991. To the fuel cell applications, oxide ion conductors or hydrogen ion conductors have chiefly been tried to be used, though solid electrolyte fuel cells are now at an early stage of development in Japan. Further, an AMTEC system for thermal energy to electrical energy conversion using beta•- alumina is under development in Japan. In this review paper, these battery applications of solid electrolytes in Japan are described briefly.

Due to the high operating temperatures (900-1000 °C) the material demands upon Solid Oxide Fuel Cell (SOFC) components are quite stringent. Preferably lower operating temperatures (700-800 °C) are desired so that gas feed lines, heat exchangers, and structure components can be fabricated from relatively cheap stainless steel components.

Typically, the materials used in a SOFC are yttria stabilized zirconia (YSZ) as the solid electrolyte, nickel-zirconia cermet as the anode, strontium doped lanthanum manganite as the cathode, and magnesium doped lanthanum chromite as the interconnection material. The electrolyte and interconnect are difficult to fabricate, because they need to be gas tight, yet thin (30-50 microns) and mechanically stable. Due to the high volatility of CrO3 the densification of LaCrO3 into thin layers is a more demanding challenge than the fabrication of the electrolyte.

Electr°Chemical Vapor Deposition is the key technology for making thin layers of the solid electrolyte as well as the interconnection material LaCrO3. In the simplest case the oxide growth can be modeled with the Wagner oxidation theory for metals. In this paper theory and experiment of the growth of ionically conducting YSZ and electronically conducting LaCrO3 will be discussed.

A thin film solid state microbattery has been used to power an O2 microsensor. The microbattery was fabricated using RF magnetron sputtering for the solid oxide/sulfide electrolyte and the TiS2 cathode. Vacuum evaporation was used for the deposition of LiI and Li anode. rhe microbattery is approximately 10 µm in thickness. The microbattery has an OCV of 2.4-2.5 V and shows close to 100% utilization to a 1.8 V cutoff when discharged between 10 and 135 µA/cm2. The microbattery is capable of supplying 2 second pulses of greater than 2 mA/cm2. The microbattery is rechargeable with over 200 cycles of 70% utilization to a 1.8 V cutoff at current densities up to 135 µA/cm2.

The O2 microsensor consists of a silk-screened Au working and Ag electrode. The Ag electrode was anodized to provide an Ag/AgCl counter electrode. The current produced at the sensor is proportional to the dissolved O2 concentration of an aqueous solution upon application of 0.6 V between the two electrodes.

Interest in the electrochromics research in the 80's was mainly directed towards potential applications for variable light transmission windows. Even though large-area electrochromic cells incorporating liquid electrolytes are being investigated, we focused on solid state devices. Such devices offer fewer fabrication problems in large-area applications. The solid electrolyte layer is a key component in the fabrication of the solid state windows. This layer must be transparent, electrochemically stable and display adequate ionic conductivity. Optical efficiencies of asymmetric solid state devices based on tungsten oxide as the electrochromic material and commercially available ionomers and polyelectrolytes - Naflon, poly(styrene sulfonic acid), and poly(2-acrylamido-2-methylpropanesulfonic acid), as the solid electrolyte layer are presented at room temperature and 90 C. Impedance behavior of this asymmetric system is discussed and compared with the behavior observed in other systems.

Nickel oxide sputtered electrodes can undergo a reversible lithium intercalationdeintercalation process which is accompanied by a net electrochromic effect. The characteristics of this type of electrode and the mechanism of the related electrochromic process are discussed on the basis of results obtained by cyclic voltammetry and optical transmittance. The results show that nickel oxide is electrochromically activated by insertion of lithium in the oxide structure. The lithium intercalation process may be promoted by cyclic voltammetry or, more extensively, by polarizing cathodically the nickel oxide electrode in a non-aqueous electrolyte containing a lithium salt. Activated nickel oxide electrodes are of considerable interest for the realization of new types of efficient solid state electrochromic windows..

Thin films of indium oxide, In2O3 (4000 Å), deposited on commercially available In2O3: Sn (ITO)/glass by rf sputtering, have been examined for potential application as a counter-electrode material in an electrochromic device, based on their chemical, structural, and optical properties. Cyclic voltammetry experiments showed that mobile lithium ions can be inserted (chemical reduction) and removed (chemical oxidation) from the host structure of indium oxide. Coulometric titrations showed that the films exhibited a hysteresis behavior for the injection and removal of lithium ions in LixIn2O3 (x=0-0.23). Structural investigation of the indium oxide films, utilizing electron diffraction techniques, indicated that they were crystalline with a crystallite size of 175 Å, in agreement with x-ray diffraction results. Differences in optical transmission between the lithiated and delithiated thin films were no more than 5% in the visible/near-infrared regions of the spectrum.

The insertion of lithium (lithiation) into tungsten trioxide results in the formation of the tungsten bronze LixWO3. Polycrystalline, rf sputter deposited thin films of LixWO3 were investigated for their application in Smart Window Devices. The optical band gap studies of these films revealed the narrowing of the intrinsic band gap as a consequence of lithium insertion. The results suggest that the rigid band model, which is generally adopted in interpreting the electronic structure of the tungsten bronzes may not be applicable in Lix WO3.

Tungsten oxide films were deposited by reactively sputtering a metallic W target in O2Ar atmospheres. Plasma emission spectra were observed as a function of the oxygen content of the chamber inlet gas using an optical multichannel analyser. The gradual loss of W lines as the O2 content was increased to 10% was accompanied by a drastic reduction in deposition rate. Cyclic voltammetry of the films deposited on ITO-coated substrates reveals that films produced under high oxygen conditions have low charge capacities and retain substantial Li upon initial cycling. The photopic coloration efficiency increased with increasing oxygen in the inlet gas. The electrochemical and electrochromic data exhibit breaks in behavior at the 10% O2 break point defined by optical spectroscopy.

Recent work in our laboratory on polymeric organodisulfides has shown these materials to perform well as positive electrodes in solid-state batteries. The polymeric materials have been named solid redox polymerization electrodes (SRPE's) due to the reversible polymerization/depolymerization reaction that occurs on charge/discharge of the electrode. The cell reaction for SRPE-based cells can be described fora simple case as, 2n M + (SRS)n = n M2SRS, where M is an alkali metal (LiNaK) and R is an organic group. In the broader sense SRPE's can have more than two S groups per monomer R unit, and are reversible to other monovalent and divalent metals. In thefully charged state SRPE's consist of polydisulfide polymers and are depolymerized on discharge by scission of sulfur-sulfur bonds, leading to the formation of dithiolate salts in the fully discharged cell. SRPE's are easy to synthesize, are air stable, and should be very inexpensive in bulk quantities. Depending on the redox potential of the polydisulfide and reaction conditions, manydisulfides can be copolymerized by oxidizing a mixture of dithiols, x HSRSH + y HSR'SH + (x+y) l2 = (SRS)x(SR'S)y + 2(x+y) HI, allowing modification of the physical and/or redox properties of the SRPE's. A series of simple aliphatic dithiols including (HSCH2CH2SH), (HSCH2CH2OCH2CH2 SH), and (HSCH2CH2SCH2CH2SH) have been oxidized to polydisulfides and mixtures of the dithiols have been copolymerized. All of the resulting polymers and copolymers were evaluated in solid-state lithium cells, with some of the new materials demonstrating high levels of performance. The utilization of available capacity in the positive electrode was observed to be independent of electrode thickness for a number of SRPE's at loading levels up to 45% by weight. At 90°C, relatively thick positive electrodes based on (SCH2CH2S)n have been discharged to surface capacities of over 20 coulombs/cm2 at a current density of 0.5 mA/cm2. The discharge profiles for most of the aliphatic polydisulfides are exceedingly flat at slightly over 2 volts versus lithium. Although other polydisulfides such as those derived from the dithiazoles exhibit higher cell voltages, the low equivalent weight of materials such as (SCH2CH2S)n [46 g/equiv] and the low cost of such polymers indicates a potential for commercial application.

Tracer diffusion as well as collective transport properties of ionic conductors are investigated within a lattice-gas model of charged particles. Results obtained from Monte Carlo simulations support the conclusion that the interplay of Coulomb-interactions among diffusing ions and structural disorder in the material provides a general mechanism for non-Debye relaxational effects commonly observed by experiment.

Solid electrolytes with structural disorder generally exhibit characteristic deviations from standard-theory spectra. The effect is known as “universal” dynamic response. In the jump-relaxation model, the phenomena are consistently explained in terms of the non-random hopping resulting from the repulsive Coulomb interaction among the mobile ions. In previous stages of the development of the model, the treatment required either crude approximations or extensive numerical calculations. Now. however, we are able to present, for the first time. simple analytic expressions for the relevant time correlation functions, derived from the rate equations of the model. In particular, the dependence of the ionic conductivity on frequency and temperature is now expressed by a simple equation. Furthermore, we recover the Kohlrausch-Williams- Watts behavior and find the KWW exponent. β. and the mismatch parameter of our model, α. to be identical. The validity of the KWW law is shown to be limited to the dispersive regime on the frequency and time scales.

The dynamic bond percolation model was developed to deal with dynamic disorder, treating ion mobility by a percolation model in which the assignment of any site-to-site jump as allowed or forbidden changes on a timescale related to the local reorganizational dynamics of the polymer segments (the renewal time). Here we discuss the special cases of highfrequency spectra and partially crystalline electrolytes. At high frequencies, the present hopping model yields unphysical behavior (frequencyindependent response); we trace this back to the incorrect treatment of short-time dynamics, and show how it can be corrected. For partially crystalline materials, we show that a rollover feature in the spectrum, in the microwave range, can be expected when ions are trapped in isolated regions of high conductivity, such as amorphous pockets in largely crystalline PEO.

Molecular dynamics simulations were carried out in an attempt to reproduce and explain the differing composition dependencies of the ionic conductivities of Na(I)-Ba(II)- and Na(I)-Sr(II)- ß' '-alumina. Impedance spectroscopy measurements of Na1.67-2xBaxMg0.67Al10.33O17, where x = 0.0 - 0.835, showed a distinct minimum in the conductivity when x = 0.67, or 80% exchange of Ba(II) for Na(I). Evidence for mobile cation ordering at that composition was seen in the simulation results and the general change in conductivity with Ba(II) content was reproduced rather well. The results of a similar experimental study of Na(I)-Sr(II)-ß' '-alumina did not show a minimum in the conductivity, but our simulations incorrectly predicted the Sr(II) system to behave in the same fashion as the Ba(II) system. Possible reasons for this discrepancy are suggested, including problems with transferring potentials between oxides and the influence of absorbed water on the experimental results.

The central force nearest neighbor model for glasses is used to discuss the Raman and infrared vibrational data for the family of lithium doped borate glasses B2O3 - xLi2O. The addition of the dopant is shown to cause local structural changes, including the transformation of three-coordinated borons to four-coordinated ones. An extremely simple structural model for the glass gives good qualitative agreement with experiment. The results of lattice dynamics calculations fall within the allowed frequency band limits predicted by network dynamics. The success of this model illustrates the importance of short range order on the vibrational spectra of covalently bonded solids.

A molecular dynamics simulation has been performed on the crystal lithium iodide, LiI. A rigid ion potential was used with parameters fit to thermal expansion, isothermal compressibility, lattice energy and the frequency of the transverse optical mode at the zone center. The current-current correlation function has been calculated at T = 200K and 400K, and from this the absorption and dispersion have been obtained. Anharmonic broadening is observed at the higher temperature.

The effect of intercalated lithium on the electronic band structure of the γ-polytype of InSe has been investigated using a tight-binding method. The energy bands of the pure polytype were calculated and the results compared with previous work. The modifications of the energy bands produced by the introduction of one lithium atom per unit cell were calculated for the lowest potential energy position of the lithium atom in the Van der Waals gap between layers. The results for the changes in the smallest and next-to-smallest direct band gaps are compared with experimental data. An interpretation of a photoluminescence peak produced by lithium intercalation is given.

On the base of numerous experimental data and theoretical approaches it is shown that ion-ion pairs with the intrapair distance about a sum of the corresponding ionic radii is the feature of charge transfer in the superionic conductors.

Ab initio Hartree-Fock band structure calculations have been performed to study the electronic structure of Li3P in the hexagonal P6/mmm crystal structure. The total energy, band structure, density of states, and charge densities are computed. The band structure is very similar to that calculated for lithium nitride with a small indirect gap between Γ and K of 2 eV. However, the charge distribution in Li 3P is more anisotropic with a greater ionicity in the x-y plane compared to the c direction. This is also supported by a large calculated core level split of 6.5 eV of the Li ls core bands.

The structure of polyethylene oxide doped with lithium perchlorate was studied. The macroscopic structure of the polymer-salt complexes was studied by thermal analysis and xray diffraction. The microscopic structure and polymer dynamics were studied by infrared, Raman and nuclear magnetic resonance spectroscopy. Thermal analysis and x-ray diffraction studies clearly revealed the change of crystalline phase to an amorphous phase by the addition of salt. The vibrational analysis showed a conformational transformation of the polymer chain by the addition of salt. Vibrational analysis also showed clearly the formation of free ion, ion-pair and salt cluster in the polymer as the concentration of the salt increased.

Various types of polymeric materials with enhanced electrical properties have been characterized recently. Many of these conducting polymers are of specific interest in solid state electrochemistry and in solid state ionics since they act as novel electrode and electrolyte materials. Indeed, these materials are currently used as improved polymer electrolytes and polymer electrodes for the development of advanced-design electrochemical devices. However, there are still some problems which prevent the wide utilization of these conductors. In this paper we attempt to identify the nature of these problems and discuss their possible solutions.