Molten additive binary systems having a common anion yield much more heuristic information for a better understanding of the mechanism of ionic conduction than pure melts do. In most of such systems the Chemla effect occurs; the mobility isotherms of two cations have a crossing point called the Chemla crossing point. In binary multivalent cation systems, the larger cation is more mobile than the smaller one. An internal mobility u of a monovalent cation is generally expressed by an empirical equation: u = [A/(V_m − V₀)]exp(−E/RT), where V_m is the molar volume; A, V₀ and E are constants nearly independent of the kind of the second cation. Deviation from this equation could be interpreted in terms of the agitation effect (positive deviation), and the tranquilization effect or the free space effect (negative deviation). The dynamic dissociation model well accounts for the above-mentioned findings; in other words, the internal mobility is strongly related to the dissociative motion of neighboring cation and anion. The effect of highly polarizable cations on internal mobilities is also discussed.

The structures of potassium borate glasses and melts with 10 and 30 mol % K₂O have been studied by the use of pulsed neutron total scattering measurement. With an increase in K₂O content, the first peak of the experimental radial distribution function corresponding to nearest neighbor B-O correlation spread asymmetrically to the higher bond distance r side, and the coordination number of O atoms around a reference B atom, Nb–o, increased from 3.0 to 3.4 reflecting a structural change from BO₃ triangle units to BO₄ tetrahedral units. It can be concluded from the neutron results that some BO₃ units present are converted to BO₄ units on the addition of up to 30 mol % K₂O to the B₂O₃ melt in the same way as in K₂O-B₂O₃ glasses. XAFS studies of network modifying cations in alkali borate glasses have made an important contribution to our understanding of the structure of glass. The local structure in K₂O-B₂O₃ glasses has been determined in order to the elucidate the environment of network modifying K⁺ ions. The average coordination number, N_K-O, of oxygen around K⁺ cations and the mean distance, r_K-O, are found to be approximately 6 and 2.83∼2.86 Å, respectively. The XAFS results are in good agreement with X-ray diffraction and MD simulation results.

Electrochemical behavior of trivalent vanadium species was investigated in a Lewis basic 1-ethyl-3-methylimidazolium chloride (EMIC)-aluminum trichloride (AlCl₃) molten salt at room temperature. The reduction of V(III) was observed around −1.2 V vs. Al(III)/Al in the cyclic voltammogram on platinum electrode. The reduction process of V(III) seems electrochemically irreversible as observed in aqueous solution.

The relative difference in the internal cation mobilities in the molten binary system NdCl₃-KCl was measured by countercurrent electromigration method at 1073 K. Isotherm of the internal cation mobilities was estimated with available densities at this temperature. The internal cation mobility of Nd³⁺ becomes drastically small with decrease of its concentration, which is explained by stable complex formation of NdCl₆³⁻.

The short range structure of Na₃ErCl₆ melt at 973 K and K₃ErCl₆ melt at 1053 K was investigated by X-ray diffraction. The nearest neighbor distance of Er³⁺-Cl⁻ and the coordination number of Cl⁻ around Er³⁺ in Na₃ErCl₆ melt, for instance, were estimated to be 2.61 Å and 5.7, respectively, and the Er³⁺-Er³⁺ and Cl⁻-Cl⁻ distances 5.15 Å and 3.90 Å, respectively. The ratio of the anion-anion pair distance to the cation-anion pair distance (r₋₋/r₊₋) was, therefore, 1.49, being close to 1.41, characteristic of octahedral geometry. These results indicate that the stable ErCl₆³⁻ octahedra exist in Na₃ErCl₆ and K₃ErCl₆ melts. Probably, complex ions such as Er₂Cl₁₀⁴⁻ and Er₂Cl₁₁⁵⁻ expected to be in the pure ErCl₃ melt are little formed at these compositions according to the geometrical calculations of bond lengths and bond angles.

Raman spectra were measured for the mixture of alkali metal propionates and water in the composition range of hydrate melt to highly concentrated solution. Band positions and band widths for the C₁-C₂ stretching, C-O stretching in the propionate anion, and O-H stretching bands in water molecule were investigated for various compositions and temperatures. The dependence of Raman bands on composition, cation, and temperature is discussed in view of ion-ion, ion-water, and water-water interactions, suggesting that the direct interaction between cation and anion increases with decreasing water content.

Surface tensions of basic NaCl-AlCl₃ and LiCl-AlCl₃ binary melts were measured by using capillary-rise method up to 1200 K precisely. It was found that the adding very small amount of AlCl₃ into NaCl or LiCl decreases the surface tension very steeply to less than half value of NaCl or LiCl. The surface enthalpy and surface entropy also decreases steeply. These phenomena are similar to the action by surface active agent although the phenomena are rare in molten salt systems. This is due to the formation of the compound of NaAlCl₄ and LiAlCl₄. The compound is considered to be concentrated in surface phase due to their lower coulombic energy, and resulted the decrease of surface tension and surface energy, and furthermore the increase of ordering on surface phase. Based on present results and the thermodynamic consideration, the concentrations of NaAlCl₄ and LiAlCl₄ in the surface phase were calculated.

Density, viscosity and surface tension of molten aluminum chloride have been measured precisely to understand its unique physicochemical properties by using a float method, a capillary viscometry and a capillary rise method, respectively. The apparatuses used are all made of quartz or Pyrex glass and the melt was sealed in them completely. Density, viscosity and surface tension obtained are considerably lower than those of other conventional molten salts. The temperature corresponding to critical temperature could be estimated from the surface tension. The hypothetical melting temperature by assuming that the molecular solid which consists of Al₂Cl₆ was estimated to be almost 150 K lower than melting temperature of real ionic crystal of aluminum chloride. These results suggest that the molten aluminum chloride is different in the condition of the study from other typical molten salts, namely it is considerably superheated liquid due to the structural change on melting from ionic crystal to molecular liquid. The lower values measured in this work are attributed to such a superheated state.

The thermodynamic model has been discussed for the surface tension of molten salt mixtures in the common ion alkali-halide, carbonate, sulfate and nitrate systems. In addition, the extension of the above model has been investigated to alkaline-earth halide systems and reciprocal alkali-halide systems. Based on the present approach, the thermodynamic databases, which are usually used to the calculation of phase diagrams of salt mixtures, can be applied to the evaluation of the surface tension of molten salt mixtures.

We have investigated the calculation conditions necessary for precise and efficient evaluation of the thermal conductivity of a molten salt based upon the equilibrium Green-Kubo formula by molecular dynamics simulation. The suitable parameters are also presented for the Ewald sums in the calculations of the interionic force and the energy current. The statistical error was estimated at ±10% under the calculation conditions. The dependence of the calculated results was examined on various calculation conditions. The number of time steps of at least 50000, which corresponds to 200 ps, was required in order to obtain a fully converged value of the time integral of the correlation functions. No significant dependence was found on the number of particles used in the simulated system compared with the statistical error. The dependence on the time step, the interval between time origins when the correlation functions were sampled, and the upper integral limit for the correlation functions was negligible compared with the statistical error. The addition of a small velocity vector to the ions, however, significantly influenced the calculated value of the thermal conductivity.

The ε values in the internal cation mobilities of (Li, Na, K)(CO₃)_0.5 of the eutectic composition (43.5-31.5-25.0 mol %) in the temperature range from 773 K to 1023 K have been measured by a countercurrent electromigration method, that is the Klemm method, where ε is defined as (u_i − u_j)/u_a; u_a is the average internal cation mobility and subscripts i and j refer to any two of three cations. From the present data and the densities and electric conductivities, the internal cation mobilities of u_Li, u_Na and u_K have been calculated. The internal mobility of Li⁺, u_Li, was compared to the empirical equation expressed for binary alkali carbonates. The orders of the internal cation mobilities were u_K < u_Li < u_Na at 773-873 K, u_Li < u_K < u_Na at 923 K, and u_Li < u_Na < u_K above 973 K, although these were broadly equal to one another. These orders could be interpreted in terms of the dynamic dissociation model previously presented.

The surface tension and the density of molten Li₂CO₃-X₂CO₃ (X : Na, K, Rb, and Cs) and Na₂CO₃-Y₂CO₃ (Y : K, Rb, and Cs) binary carbonate systems were measured using maximum bubble pressure technique. The addition of Rb₂CO₃ or Cs₂CO₃ to Li₂CO₃ or Na₂CO₃ effectively decreases the surface tension of mixtures. At constant temperature, the relationship between surface tension and composition in all binary carbonate systems was well correlated by the Guggenheim’s Equation for regular solution. The molar volume of Li₂CO₃-Rb₂CO₃ and Li₂CO₃-Cs₂CO₃ systems shows positive deviation from additivity.

The solubility of prelithiated NiO and prelithiated cobaltite as cathode materials for MCFC in 62 Li-38 K eutectic carbonate melts was studied with immersion test method. It was found that the solubility of either prelithiated NiO or prelithiated cobaltite was affected by its lithium (Li) content. The higher Li content was, the lower solubility could be in the range of 0∼0.25 mole fraction of Li. With the prelithiated NiO at Li content of 0.2 mole fraction, Ni solubility was reduced to 3∼4 × 10⁻⁶ mole fraction, which was only one-third of that of the normal NiO cathode in 30% CO₂-70% air at 1 atm. It was obtained that the solubility of prelithiated cobaltite and even LiCoO₂ were higher than the prelithiated NiO in the low O₂ concentration gas. This suggested that even LiCoO₂ was not stable in the O₂ depleting condition, and more soluble than NiO.

The electrode kinetics and mechanism of oxygen reduction at gold electrodes in (53 + 47) mol % (Li + Na)CO₃, (62 + 38) mol% (Li + K)CO₃ and (43.5 + 31.5 + 25) mol % (Li + Na + K)CO₃ melts have been investigated under pressurized Air/CO₂ oxidant gas conditions in a temperature range of 823-973 K. The total pressure was changed from 1.0 to 5.0 atm with a constant ratio of Air/CO₂ = 70/30, 80/20, 90/10 or 95/5. It was observed that the oxygen reduction current in cyclic voltammograms increased with the total pressure, indicating that the concentration of reactive species increased. The analyses of the gas partial pressure dependence on Warburg coefficients, which were obtained by ac impedance measurements, revealed that the reaction process in the tested conditions was dominated by the mixed diffusion of superoxide ion O₂⁻ and dissolved CO₂.

The formations of Li-doped NiO and LiNiO₂ in molten carbonates have been studied in O₂ atmospheres from 873 K to 1173 K. NiO was reacted in Li containing molten carbonates; Li₂CO₃, (Li_0.62K_0.38)₂CO₃ and (Li_0.52Na_0.48)₂CO₃ melt for 20 h. The rate of lithiation reaction of NiO was measured by TGA and the products were analyzed by XRD. LiNiO₂ was formed in O₂ atmosphere at higher temperatures. The formation rate of LiNiO₂ was the fastest in Li₂CO₃ and the temperature of LiNiO₂ formation in Li/K carbonate and Li/Na carbonate was higher than that in Li₂CO₃. The Li content in molten carbonate and the oxygen in the atmosphere would have important roles for the lithiation of NiO and the formation of LiNiO₂ as well as the reaction temperature.

The stabilities of lithium zirconate (Li₂ZrO₃) and zirconia in molten carbonates were tested. There are some disputes about the stable zirconium compound in molten carbonates. We considered a possible reversible change between lithium zirconate and zirconia by means of the thermodynamic calculation and experiments. All results suggest that the change not only depends on temperature and CO₂ partial pressure but also on the presence of molten carbonates, e.g., alkali carbonate in molar ratio of Li : K = 62 : 38. In the presence of the molten carbonates, the change proceeds reversibly.

A new way to solve the nickel deposition problem in the matrix of MCFC has been proposed, i.e. by using the matrix of LiAlO₂ powder coated with cobalt oxide to decrease the concentration of reducing agents (such as H₂) in the matrix through the reaction with in situ oxide agents. The possibility of the co-existence of H₂ and O₂ at a trace amount in the presence of normal γ-LiAlO₂ matrix powder around 650 °C was verified by the gas chromatography analysis. It was found that H₂ amount was sharply reduced in the presence of γ-LiAlO₂ matrix powder coated with cobalt oxide, that means the cobalt oxide have the catalytic function for the reaction of H₂ and O₂ even while both H₂ and O₂ are at a trace amount. Electrochemical measurement showed that the redox potential (Eredox) in the normal matrix was almost the same in the tested distance ratios, and the value was less than −1000 mV. However, the Eredox value in the matrix coated with cobalt oxide increased linearly with the distance ratio from the anode side to the cathode side. According to the nickel deposition potential, it was deduced that Ni deposite zone might be limited within the region near the anode side. Therefore, it was anticipated that the short-circuit time of a cell would be postponed by use of the matrix with cobalt oxide.

The effect of nickel oxide such as NiO_1+x and LiNiO₂ in the Ni based composite anode on the physico-chemical properties of oxidized layer formed on the composite anode prepared by sintering or dispersion coating method was studied. The galvanostatic and potentiostatic polarization behaviors of the Ni, the Ni-NiO, and the Ni-LiNiO₂ composite anodes were investigated to elucidate the anodic processes. The electrical conductivity of the oxidized layer formed on the Ni-NiO and the Ni-LiNiO₂ composite anodes was large in a well dehydrated melt of NH₄F·2HF and its resistance to corrosion during polarization at potentials higher than 4.6 V vs. H₂ was high compared with those on the Ni anode, because of the higher content of nickel oxide in the oxidized layer on the anode.

A novel SiH₄ production method using molten salt electrochemical process has been proposed. In order to examine the possibility of the method, a potential-pH⁻ diagram was constructed for the Si-H system in a LiCl-KCl eutectic melt. The electrode behavior of a single crystal Si electrode was investigated by cyclic voltammetry in molten LiCl-KCl-LiH(5 mol %) systems at 673 K. The result showed that a passivation film was formed on the electrode by anodic polarization and that further polarization leads to breakdown of the passivation film. From the IR absorption spectroscopy of the evolved gas, SiH₄ production was confirmed. Fluoride ion addition to the melt was found to be effective for promoting the SiH₄ production. The current efficiency for the SiH₄ evolution was improved to about 20% by using a lower LiH(0.8 mol %) system, which suggests that SiH₄ evolution mainly proceeds by the indirect reaction of chlorosilanes.

The electrochemical behavior of hydride ion on a nickel electrode in a LiF-NaF-KF eutectic melt at 773 K has been studied by means of gas chromatography, cyclic voltammetry, chronopotentiometry and chronoamperometry. It was found that hydride ion is anodically oxidized at around 0.45 V vs. M⁺/M (M = alkali metal) and pure hydrogen gas is generated according to the following reaction:

H⁻ → l/2 H₂ + e⁻

This anodic process is a reversible, one-electron reaction governed by diffusion transport. The diffusion coefficient of hydride ion, , was estimated to be around 1 × 10⁻⁵ cm² s⁻¹ at 773 K.

Electrochemical study on redox reactions of polybismuth cations has been carried out in the acidic 1-metyl-3-benzyl imidazolium chloride (MeBzImCl) + 67 mol % AlCl₃ melt, where MeBzImCl was synthesized from 1-methyl imidazole and benzyl chloride. The reduction of Bi₄⁴⁺ to Bi₈²⁺ (Bi₄⁴⁺ + 3e = 1 / 2 Bi₈²⁺) occurs at the cathodic peak around 0.6 V for BiI dissolved into this melt. The number of electron which took part into reduction at the cathodic peak around 0.6 V was determined as 2.8 ± 0.3 by polarography for BiI. For Bi₆Cl₇ dissolved into this melt, the reductions of Bi³⁺ to Bi⁺, Bi₅³⁺ to Bi₈²⁺ and Bi₈²⁺ to Bi metal occur at the cathodic peaks around 0.9, 0.7 and 0.2 V, respectively. The oxidations of Bi₅³⁺ to Bi³⁺ and Bi₈²⁺ to Bi³⁺ occur at the anodic peaks around 1.45 and 1.67 V for both BiI and Bi₆Cl₇.

A Magnesium-Nickel hydrogen absorption alloy, Mg₂Ni, was produced as a film by molten salts electrolysis and diffusion process. Magnesium was electrodeposited on a nickel substrate from a 25 mol % NaCl - 5 mol % KC1 -70 mol % CaCl₂ mixture as a supporting electrolyte, and additionally 10 mol % MgCl₂ with respect to the supporting electrolyte as magnesium-ion source. The electrodeposition was conducted at 873∼1073 K and with cathodic current densities between 0.02∼0.10 A cm⁻² for 3.6∼18 ks. After the electrolysis, the specimens were subjected to X-ray diffraction and EPMA line analysis to identify the phase of compounds and to measure the thickness of the film. As a result, only hydrogen absorption alloy films, Mg₂Ni, were obtained with 20 µm and 27 µm thickness from electrodeposition with 0.10 A cm⁻² for 3.6 ks at 1023 K and with 0.02 A cm⁻² for 18 ks at 873 K, respectively. At 1073 K, the intermetallic compound MgNi₂ was also detected between the Mg₂Ni layer and the nickel substrate. These specimens were subjected to hydrogen absorption tests in 30 mass % KOH solution at 298 K. It was confirmed that these specimens absorbed hydrogen.

Direct electrowinning of liquid titanium metal has been investigated by using a DC-ESR (Direct Current Electro Slag Remelting) apparatus. A small scale DC-ESR unit was operated in a DC reverse polarity mode, where a graphite rod was used as anode and a steel or a copper base-plate was used as cathode. The used salt was CaO-CaF₂-TiO₂ mixture. The distance between the cathode and the anode was essential to the electrolysis. A metal pool was formed on the base-plate and the current efficiency for the reduction was up to 18% with the proper distance between the electrodes. Some amount of titanium was electrodeposited on the base-plate though its state changed with the electrolytic condition. Pure titanium metal pieces were obtained in the solidified salt after the run with the bigger distance. It is concluded that the electrowinning of liquid titanium metal by the present process is possible if sufficient heat to form a metal pool can be supplied at the bigger distance between the electrodes.

The effect of oxide ion was investigated in the electrodeposition of tungsten in an acidic ZnCl₂-NaCl (60-40 mol %) melt at 450 °C. Cyclic voltammetry was performed in melts which contained various tungsten compounds and different amounts of ZnO. In constant potential electrolysis, metallic tungsten was obtained from K₂WO₄, WO₃, WOCl₄, and WCl₆, but not from KWCl₆, K₂WCl₆, WCl₄, K₃W₂Cl₉ and WCl₂. When ZnO was added, metallic tungsten deposited from KWCl₆, K₂WCl₆, and WCl₄. It seems that the oxide ion or an oxochlorotungsten species participates in the electrodeposition process.

This investigation was carried out to find a molten salt which enables a smooth electrodeposition of tungsten. The KF-B₂O₃-WO₃ molten salt was selected on the basis of the acid-base cooperative reaction theory. A composition of KF(67 mol %)-B₂O₃(26 mol %)-WO₃(7 mol %) was selected as the preferred bath composition. Using a tungsten anode, smooth electrodeposits were obtained at 1123 K and at a current density ranging from 250 Am⁻² to 550 Am⁻² on nickel, copper, stainless steel, and mild steel substrates. When a graphite anode was used, smooth electrodeposits accompanied by dendrites were obtained. The Vickers microhardness number of the electrodeposit was 422 (load 0.49 N). The smooth electrodeposit had tendencies to form the 111 fiber texture or a mixed texture of the 111, 211 and 321 fiber textures.

The electrodeposition of the Nb-Sn alloy from the SnCl₂–NbCl₅–EMIC melt was investigated. From the results of the physical properties and cyclic voltammetry, the possibility of electrodeposition of the Nb-Sn alloy was suggested, which was subsequently obtained from the melt by electrodeposition. However, the Nb content in the deposits was very low. The low Nb content was related to the low limiting current density of Nb. From the results of bath temperature, melt composition and pulse electrolysis for increasing the Nb content in the deposits, the Nb-Sn alloy containing 69.1 wt % Nb was obtained from the 2.8 mol % SnCl₂–68.6 mol % NbCl₅–28.6 mol % EMIC melt at 200 mA cm⁻² with a pulse period of 10 ms, duty ratio of 0.2 at 160 °C.

The vanadium bronze films were synthesized onto the ITO electrode by electrolysis in molten dimethylsulphone (DMSO₂)-V₂O₅ and DMSO₂-V₂O₅-LiNO₃ systems at the temperature about 150 °C. The obtained film was analyzed by x-ray diffraction. Moreover, the electrochromic (EC) reactions of the vanadium bronze film in room temperature solution of propylene carbonate (PC) or high temperature solution of DMSO₂ were investigated. The lithium vanadium bronze film was discussed on the basis for high temperature electrochromic materials.

To decrease the reaction polarization of the cathodic reaction of Cl₂ + 2e⁻ → 2Cl⁻ in Al-Cl₂ electrochemical cell in NaCl-KCl melt at 1023 K, we developed a cell system with a vertically rotating electrode composed of sliced graphite disks which are partially immersed in the melt. The apparent reaction resistance of the cathodic reaction was evaluated from transient potential and current changes of the cell. The reaction resistance decreases with an increase of the number of the graphite disks. The resistance is a function of the rotation rate of the electrode and exhibits the minimum value around a rotation rate of 20 rpm. The dependence on the rotation rate may be explained from Cl₂ concentration in a thin melt layer formed on the graphite electrode during exposure to Cl₂ gas phase. The rotation may introduce a continuous supply of the thin melt layer with a high concentration of Cl₂ to a reaction place at a meniscus part on the graphite electrode. However, if the rotation rate is too high, the exposure time of the thin melt layer to Cl₂ gas is not enough to establish an equilibrium concentration of Cl₂ in the layer.

The reaction distributions in not only the circumferential but also the longitudinal directions of the positive electrode of a Na/SeCl₄ cell utilizing the basic AlCl₃-NaCl melt are discussed. The two-dimensional potential distributions during charge, discharge, and open circuit are measured using tungsten microelectrodes inserted in a dual-mat graphite electrode and the local potentials are compared with the emf of the Na/SeCl₄, cell to calculate the radial and axial distributions of the formal oxidation states of selenium. The radial reaction distribution during discharge is uniform independently of current densities, while the axial one is almost homogeneous except at deeper stages of discharge at a high current density. The radial and axial reaction distributions during charge are also uniform until the occurrence of the melt acidification which is confirmed from the result that the local potential becomes nobler than the emf of the cell in the fully charged state. The uniformity of the reaction distribution in the dual-mat graphite electrode, especially in the radial direction, causes a uniform deposition of solid NaCl on the graphite electrode, consequently inducing the enhanced depth of discharge and the improved utilization of tetravalent selenium.

An AlCl₃-1-ethyl-3-methylimidazolium chloride (EMIC) melt has a wide electrochemical window, and it is nonflammable with no ignition. We have developed the rocking chair type lithium secondary battery with an AlCl₃-EMIC melt containing excess LiCl with a saturated solubility as the nonflammable electrolyte. In this paper, when LiCl was added in the acidic melt (AlCl₃ > 50 mol %), the solubility of the LiCl and the changes in the physical properties were measured. Moreover, SOCl₂ or Li metal was treated to completely remove any small amounts of Al₂Cl₇⁻, which was the ionic species for deposited Al, that remained in the melt. As a result, it was confirmed that Al₂Cl₇⁻ was completely removed based on a cyclic voltammogram (C.V.) using a W electrode. The reversible deposition and dissolution behavior of lithium was indicated from the C.V. for a Li electrode in the SOCl₂ containing melt, and it was found that a Li metal would perform as the negative electrode active material.

Room Temperature Molten Salt (RTMS) based on 1-methyl-3-ethylimidazolium chloride/AlCl₃/LiAlCl₄ is recently shown to provide a promising medium for lithium battery due to its high current capability and inertness towards active material. In the present work, the electrochemical properties of a Li_xMn₂O₄ electrode in RTMS, the most commonly used positive electrode material, were investigated by cyclic voltammetry, coulometric titration and constant current cycling. From cyclic voltammetric studies, the Li_xMn₂O₄ electrode in RTMS was found to exhibit the same electrochemical behavior as in other non-aqueous electrolytes. However, a new and very large irreversible anodic peak was found due to the insertion of AlCl₄⁻ into the carbon current collector. During coulometric studies, coulombic efficiencies greater than 96% were obtained at composition close to x = 1 in Li_xMn₂O₄. However, for Range I (0 < x < l), a rapid decrease in coulombic efficiency was observed at x less than 0.6. For Range II (1 < x < 2), close to 86% of the electrode material could be used. Thus, Range II was selected for battery application. For cycling at Range I, greater than 95% cycling efficiencies were obtained up to insertion/extraction capacities of 60 mAh/g, whereas at Range II, 98% cycling efficiencies at the first 20 cycles were obtained up to 120 mAh/g. The difference was attributed to the irreversible insertion of AlCl₄⁻ anions into the carbon current collector at high anodic potential and hence less lithium was extracted from the Li_xMn₂O₄ electrode in 0 < x < 1. The electrochemical performance of the LiMn₂O₄ electrode as positive electrode material for secondary lithium battery at different lithium insertion in RTMS was discussed in the light of the results obtained.

The effects of current density and temperature on the current efficiency for uranium metal deposition were investigated by electrolysis at the cathodic current densities of 1.34∼22 A/cm² in the fluoride salts in the temperature range of 1150∼1200 °C. The electrolyte consisted of 74 wt. % BaF₂, 11 wt. % LiF, and 15 wt. % UF₄. On the basis of the theoretical assumption, the factors affecting the current efficiency were discussed. The current efficiency for uranium deposition increased with increasing current density. A neutral hearth type of electrode in a small cell or a horizontal type of electrode with cathode studs in a large cell enabled to conduct electrolysis at a high cathodic current density. From these results, it is found that the current efficiencies for uranium deposition in both cells were high compared with those reported previously.

A neutron diffraction measurement has been performed for molten TlCl, a monovalent transition metal chloride, at 800 K. The radial distribution function showed very broad peak owing to the large polarizability of the ions. However, the ratio of the r value of the second peak to the first 1.43 seems to show NaCl type local structure of the melt in spite of the CsCl type crystal structure. This finding has been supported by molecular dynamics calculation.

The solubility of lead monoxide was measured in an AlCl₃-NaCl molten mixture at 160-320 °C in a Pyrex glass ampoule. The solubility was minimal at 50 mol % AlCl₃ and increased with the temperature and the concentration of AlCl₃ in the melt, but had a maximum at 70 mol % AlCl₃.