•The clad foil endures the large volume change in Si-based materials.•The clad current collector improves cycling performances of Si-based electrodes.•The developed new clad foil is 2 times stiffer than that of pure Cu foil.•The LFP/SiO cell using clad foil shows stable performance more than 800 cycles.We develop a clad foil current collector with a high tensile strength that endures a large volume change in the active material during the charge and discharge, such as the Si-based materials. The nano-Si negative electrode with the clad current collector retains 76% of the initial capacity after 40 cycles, while the capacity of the nano-Si electrode with a conventional Cu foil drops to less than 70% only after 10 cycles. A full cell with the SiO negative electrode and the LiFePO4 positive electrode retains more than 90% of its capacity at the 10th cycle after 800 cycles. The conventional rolled Cu foil wrinkles during the cycling test. The high-strength clad current foil hardly deforms during the test regardless of the electrode size.

This is the first report about a spinel manganese oxide that serves as a high capacity positive electrode material for the sodium ion battery. By electrochemically extracting Li from a monoclinic layered Li2MnO3, we prepared Li2−xMnO3 (x = 1.6-1.8) of which the Li-extracted domain has a cubic spinel structure. The reversible discharge and charge capacity of Li2−xMnO3 versus the Na negative electrode initially exceeded 200 mA · h · g−1, suggesting that close to one molar equivalent of Na is inserted in and extracted from the formula unit, Li2−xMnO3. The electrode retains the capacity of 160 mA · h · g−1 after 50 cycles. On the other hand, the Li2−xMnO3 electrode versus the Li negative electrode significantly degrades upon cycling.The ex-situ synchrotron X-ray diffraction (SR-XRD) and transmission electron microscopy (TEM) analyses revealed that the spinel domain of Li2−xMnO3 retains its crystallographic structure during the Na insertion and extraction, although the crystal significantly loses its periodicity when Na is inserted. A numerical simulation of the SR-XRD profile suggests that the Na-inserted Li2−xMnO3 has the periodicity of only one to two unit cells while retaining the spinel structure. XAS revealed that the reversible capacity is found to be dominated by the redox between Mn(III) and Mn(IV) and the Mn-Mn distance significantly loses its correlation upon Na-insertion, which is consistent with the broad SR-XRD profile.

The solution density, viscosity and conductivity of MClO4 (M = Li and Na) solutions in propylene carbonate and γ-butyrolactone are measured at the concentrations of <1.5–2.0 mol dm−3. The partial volume of the solute, derived from the density, of NaClO4 is greater than that of LiClO4 as expected. NaClO4 produces less viscous and more conductive solutions than LiClO4 throughout the examined concentration range. Notably, the conductivity of the NaClO4 solutions is 10–20% higher than that of the LiClO4 solutions at T/K = 298. The validity of the empirical cubic root law, Λ(C) = Λ0 − AC1/3, is examined, where Λ and Λ0 are the molar conductivities at the molarity C and at infinite dilution. The meaning of the slope A is interpreted in the theoretical framework of the pseudolattice model.Graphical abstractHighlights► Comparison between transport properties of Li- and Na-based non-aqueous electrolytes. ► Density, viscosity and conductivity of LiClO4 and NaClO4 in PC and γBL were measured. ► Na-based electrolytes have 10–20% higher conductivity than Li-based ones at 0.5–2 M. ► Conductivities were analyzed based on a theory derived from the pseudolattice model.

This study is intended to determine if the capacitive properties are improved when a specific amount of crystalline ruthenium oxide (c-RuO2) is added to an amorphous hydrous ruthenium oxide (a-RuO2) electrode fabricated by the spark plasma sintering technique. For at the cyclic voltammetry scan rates higher than 10 mV s−1, the capacitance of a highly pseudo-capacitive, but less electron-conductive a-RuO2 electrode is augmented by adding 5–20 wt.% of c-RuO2 which is less capacitive, but more electron-conductive than a-RuO2. The capacitance fades when more than 20 wt.% of c-RuO2 is added because the less capacitive nature of c-RuO2 prevails. The proximate cause of this phenomenon is the electronic conductivity, σ, of the composite electrode as we observe a maximum in σ at around a 5–20 wt.% c-RuO2 content. The fact that c-RuO2 is composed of smaller particles than a-RuO2 seems to be related to the maximum σ value for a certain c-RuO2 content of the composite electrode.Highlights► Binderless fabrication of RuO2 pseudocapacitor electrode by spark plasma sintering. ► Addition of crystalline RuO2 to amorphous one improves the capacitive behavior. ► Amorphous:crystalline = 95:5 composite RuO2 electrodes show the best performance. ► High electronic conductivity at this composition results in the above improvement.

In order to understand the final state of the TiCl3 dopant during the dehydrogenation and rehydrogenation cycles of NaAlH4, we determined the reaction stoichiometry between TiCl3 and NaAlH4 by measuring the amount of hydrogen evolution from NaAlH4 with the varying TiCl3 -load. We found that: (i) TiCl3 reacted with 3 M equivalents of NaAlH4 during the doping process of ball-milling, (ii) the Ti dopant continued to react with NaAlH4 during the first dehydrogenation process until total six equivalents of NaAlH4 were consumed, and (iii) Ti fixed Al, not NaH, so that Al became insufficient during the rehydrogenation process. These findings lead to the conclusion that the reaction stoichiometry between Ti and Al is 1:6, which probably yields TiAl6 and plays a catalytic role in the hydrogen storage reactions of Ti-doped NaAlH4.

The spark plasma sintering (SPS) technique was successfully used to mold a hydrous amorphous RuO2electrode without any additives and binders. At the cyclic voltammetry (CV) scan rate of 1 mV s−1, the electrochemical capacitances of the RuO2 electrodes are 600–700 F g−1 for the entire electrode. An increase in the SPS current during the compaction led to the crystallization and dehydration of RuO2, which in turn, resulted in a significant decrease in its capacitance. There is room to improve the rate properties as we observed a steep drop in the capacitance when the CV scan rate was raised.

The difference in capacitive performance between high and low surface area RuO2 electrodes, synthesized with and without a mesoporous silica template, respectively, was investigated in aqueous solutions of sulfuric acid and sulfates by cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS). RuO2 synthesized with the template was crystalline and the formation of the mesoporous structure with a 6.5 nm diameter was confirmed using a transmission electron microscope and the nitrogen adsorption and desorption isotherm. From the CV at the scan rate of 1 mV s−1, the specific capacitance of the high surface area electrode in H2SO4(aq) was determined to be 200 F g−1. The high surface area RuO2 has a three times higher BET specific surface area (140 m2 g−1) than the low surface area sample (39 m2 g−1). Introducing the mesoporous structure was proved effective for increasing the capacitance per mass of the RuO2, though not all the surface functions as a capacitor. Both the CV and EIS suggest that by increasing the charging rate or frequency, the mesoporous structure of the electrode leads to a lower capacitance decrease (higher capacitance retention) than the low surface area electrode. The EIS also indicates that the response time of the capacitor is hardly influenced by the presence of the mesoporous structure.

Hydrogen storage materials containing NaAlH4, LiNH2, Mg(NH2)2, LiH and LiBH4 were subjected to standardized safety tests in order to assess the potential hazards caused by the environmental exposure of these materials. All the materials were judged ‘flammable’, ‘pyrophoric’ and ‘water-reactive’, resulted in being classified as the United Nations Packing Group I, the most stringent category of container regulations in transporting these materials. A small spark energy (1.4 mJ) can trigger an intense dust cloud explosion of the Mg(NH2)2 + LiH system of which the minimum explosive concentration was determined to be 90 mg dm−3. Although this value is lower than those of the hydrogen storage alloys, the minimum explosive concentration of complex hydrides can be comparable to the alloys if expressed in terms of the amount of stored hydrogen in the material. Also examined was the eruption test, a non-standard test, in which the sample powder was pushed out of a container into the atmosphere by pressurized H2. Despite the pyrophoricity, we observed only one explosion of the Ti-doped NaAlH4 in dozens of trials using all the materials. A comparison with other materials points to the inevitability of more cautious measures than metal hydrides when handling these complex hydrides.

The hydrogen desorption and absorption cycles of the LiNH2–LiH system were investigated. We have devised a method to simultaneously measure the hydrogen capacity and the amount of ammonia emitted from the system as a by-product during the hydrogen desorption. The initial hydrogen capacity of ca. 5 mass% exponentially decreased to ca. 2 mass% after 200 cycles at 573 K in which one cycle comprises the hydrogen desorption in vacuo for 3 h and the hydrogen absorption under 3 MPa of pure hydrogen for 2 h. The ammonia concentration in the desorbed hydrogen gradually increased with repeated cycling and abruptly dropped by re-mixing the sample after 100 cycles. By comparing the observed decay rate in hydrogen capacity with the mean ammonia concentration of 0.24 ± 0.05 mol%(NH3/H2), we found that about half the decay in hydrogen capacity can be explained by the loss of the constituent nitrogen due to the ammonia emission.

We investigated at 423 K in vacuo   the hydrogen desorption kinetics of Ti-doped sodium aluminum hydride (NaAlH4)(NaAlH4) prepared with using different Ti precursors and procedures. Initial distinction in kinetics among specimens vanishes through several cycles of dehydrogenation at 473 K in vacuo and rehydrogenation at 423 K under 10 MPa of H2H2. Considering that the temperature of dehydrogenation cycle is above the melting point of NaAlH4NaAlH4, we can assume that the difference in physical nature of the materials, such as particle size, phase distribution and so on, disappears through cycling. The present study tells that the chemical nature of the Ti related species which enhances the reaction kinetics of alanate becomes identical whatever Ti presursor is used and, however, the material is prepared.

It is known that a certain titanium species enhances the dehydrogenation of alanate, MAlH4 (M = Li and Na), in solid state. We carried out the dehydrogenation in solution in order to examine if the titanium species enhances the reaction under homogeneous condition as in solid state. The results show that the titanium species enhances the reaction in solution as well and that the titanium species interacts not only with hydride anions but also with counter cations.

Orimo et al. reported that nanostructured graphite, prepared by using the mechanical grinding under hydrogen atmosphere, contained more than 7 mass% of hydrogen whose thermal desorption spectrum (TDS) showed characteristic two peaks; one is at around 700 K and the other around 1000 K [Appl. Phys. Lett. 75 (1999) 3093; Appl. Phys. A72 (2001) 167; J. Appl. Phys. 90 (2001) 1545]. We confirmed this claim; namely, c.a. 4.5 mass% of hydrogen was detected by TDS in the desorbed gas from graphite powder mechanically ground under hydrogen in a Cr/Ni steel mortar. Yet the mechanism of hydrogenation and the physico-chemical state of adsorbed hydrogen are not known well. We found that the amount of contained hydrogen depends significantly on the grinding mortar. When a Cr steel mortar was used, we obtained 2 mass%; and when an agate mortar was used, only a trace amount of hydrogen was detected. The transmission electron microscopy and the X-ray powder diffractometry indicated that the nanostructured graphite ground in steel mortars contained a large quantity of cementite, Fe3C, to which the iron element was supplied by wearing out of mortar walls during the grinding. We examined the influence of metal particles by intentionally adding iron and nickel powder into graphite during the grinding in the metal-free agate mortar. Although in the agate mortar with metallic additives the hydrogenation did not proceed as much as in the steel mortar, the TDS spectrum showed characteristic features. The presence of catalytic metal particles seems to be a prerequisite for the hydrogenation of graphite under hydrogen by mechanical grinding.