An electroactive spinel solid solution, LiFexMn2−xO4 false(0⩽x⩽0.5false), was prepared and investigated. X-ray Rietveld refinements showed a small amount of transition metal ions in tetrahedral 8a sites in Fe-substituted samples. The 57normalFe Mössbauer spectrum of LiFe0.5Mn1.5O4 at 300 K is composed of two doublets and attributed to trivalent Fe. In the voltage range 3.0 and 5.3 V vs. normalLi/Li+, two reversible plateaus appeared at 4.9 and 3.9 V during discharge. Combined data of X-ray diffraction, Mn K-edge X-ray absorption near-edge structure (XANES), and 57normalFe Mössbauer spectroscopy obtained in situ during charge and discharge demonstrate that the plateau around 4 V on charging is associated with the oxidation of Mn3+ to Mn4+ and the plateau around 5 V with the partial oxidation of Fe3+ to Fe4+. © 2001 The Electrochemical Society. All rights reserved.

LiFeO2­Li2MnO3 solid solution was synthesized using solid-state reaction and hydrothermal-postannealing methods and characterized as a positive electrode material for rechargeable lithium batteries. Although the maximum Fe content false[normalFe/false(normalFe+normalMnfalse)false] was limited up to 30% by solid-state reaction, the content can extend up to 75% by the hydrothermal-postannealing method. Neutron and X-ray Rietveld analysis reveal that the basic structure of the sample is a layered rock-salt structure isostructural with LiCoO2 false(R3¯mfalse) in which Fe ions exist on both Li (3a) and Co (3b) sites. Elemental analysis and 57normalFe Mössbauer spectra show Fe ions exist as 3+/4+ mixed-valence state after the samples were postannealed above 650°C. The initial charge capacity of Li/sample cells was above 100 mAh/g when the upper voltage limit was 4.3 V. The plateau around 4 V was observed for all Li/sample cells on first discharge. The maximum of initial discharge capacity was about 100 mAh/g down to 2.5 V for the Li/(50% Fe-substituted sample) cell, when the positive electrode was obtained by postannealing at 650°C in air. The capacity fading of the 4 V plateau could be suppressed by adjusting the Fe content to less than 50%, postannealing temperature between 600 and 700°C, and by 10% Ni substitution. © 2002 The Electrochemical Society. All rights reserved.

Lithium cobalt oxide (LiCoO 2 ) with ␣-NaFeO 2 -type (layered rock-salt) structure is now extensively used as a positive electrode material in commercial lithium-ion batteries. In this structure, LiO 6 and CoO 6 octahedra share their corners and stack alternatively along the c axis direction, which allows two-dimensional diffusion of Li ions during electrochemical deintercalation and intercalation. Because the Co in LiCoO 2 is a relatively rare metal, dopants such as Ni, 1 Cr, 2 Mn, 3 B, 4 Al 5 have been used to reduce material costs. Although Fe doping 6,7 is among the effective methods to accomplish this aim, the incorporation of Fe has been limited to 10% per formula unit (LiCo 0.9 Fe 0.1 O 2 ) by solid-state reaction at high temperature (800ЊC 6,7 ). Although Alcantara et al. 7 claimed that the solubility limit of Fe in LiCoO 2 was 20%, a small nonindexable peak near the 101 peak (2 ϭ 36Њ) was observed for LiCo 0.8 Fe 0.2 O 2 . This limitation originates from sample contamination by the formation of cubic ␣-LiFeO 2 with a cation-disordered structure, as for the case of Fe-doped LiNiO 2 , 8 because LiFeO 2 , isostructural with LiCoO 2 (layered-LiFeO 2 ), is a metastable form and can be obtained by a softchemical synthetic route such as ion-exchange and hydrothermal reactions. A soft-chemistry method is necessary to form rhombohedral Fe-doped LiCoO 2 with high Fe content.Charge-discharge voltages and capacities of Li/LiCoO 2 cells are sensitive to the starting materials and LiCoO 2 preparation conditions. 9-12 Some structural models for Li-extracted LiCoO 2 have been reported. 9,10 Concerned about Fe doping effects into LiNiO 2 , Reimers et al. 8 showed that a single-phase layered structure could be obtained up to LiNi 0.8 Fe 0.2 O 2 and that the charge and discharge capacities decreased with increasing Fe content. Recently, Prado et al. 13 reported that a homogeneous solid solution could be obtained up to LiNi 0.7 Fe 0.3 O 2 , a rhombohedral symmetry was maintained up to Li 0.4 Ni 0.9 Fe 0.1 O 2 during electrochemical oxidation, and some of the trivalent iron was oxidized to the tetravalent state after electrochemical charging of a Li/LiNi 0.9 Fe 0.1 O 2 cell. Although Alcantata et al. examined the cation distribution and electrochemical properties of Fe-doped LiCoO 2 , 7 no systematic study of changes of crystal structure and the valence state of 3d metals after charge-discharge cycle tests has been reported for Li/LiCo 1Ϫx Fe x O 2 cells.We have successfully obtained metastable layered LiFeO 2 14 and layered LiMnO 2 15 with an ␣-NaMnO 2 structure similar to ␣-NaFeO 2 using a mixed-alkaline hydrothermal reaction below 300ЊC. Up to 25% Fe-doped LiCoO 2 could be prepared by hydrothermal reaction at 220ЊC. 16 Structural changes and valency of Co and Fe during charge-discharge tests of Li/LiCo 1Ϫx Fe x O 2 cells were studied to understand the Fe doping effect on the LiCoO 2 positive electrode. ExperimentalIron-doped LiCoO 2 samples (LiCo 1Ϫx Fe x O 2 nominal x ϭ 0.00 (sample A), 0.05, 0.10, 0.15, and 0.25) were ...

Li 4/3 Ti 2/3 O 2 -LiFeO 2 solid solution, Li (4 2 x)/3 Ti (2 2 2x)/3 Fe x O 2 (0.18 ¡ x ¡ 0.67), which has the cubic rock-salt structure (Fm3 ¯m, average particle size less than 100 nm), was synthesized from Fe-Ti co-precipitates by hydrothermal reaction with excess LiOH and KClO 3 at 220 uC. Calcination of the products with lithium hydroxide in an oxidative atmosphere leads to the oxidation of trivalent iron to a 31/41 mixed valence state. Hydrothermally-obtained Li 1.2 Ti 0.4 Fe 0.4 O 2 gave maximum initial charge (266 mA h g 21 ) and discharge capacities (153 mA h g 21 around 3 V) between 2.5 and 4.8 V. Calcination enabled us not only to improve the crystallinity, but also suppress the discharge capacity fading with cycle number. Two plateaus around 3 and 4 V were observed on discharging by decreasing the amount of Li extraction (0.4 Li per chemical formula). Although the cubic rock-salt structure was retained during both charge and discharge processes, a partial 3d-cation displacement from octahedral 4a to tetrahedral 8c sites and some oxygen loss were observed after electrochemical delithiation. In-situ 57 Fe Mo ¨ssbauer spectroscopy showed evidence of the Fe 31 /Fe 41 redox only around the 4 V region.

A 5 V class ceramic/polymer composite all-solid-state lithium battery was prepared. The cell configuration was ͓Li 3 PO 4 coated LiNi 0.5 Mn 1.5 O 4 ͉solid polymer electrolyte͉ Li͔. The total cell impedance was 4 k⍀ at 333 K and the discharge capacity was 100 mAh g Ϫ1 with a discharge voltage plateau in both 4.7 and 4.1 V regions. X-ray absorption near-edge structure results indicated that both transition metal ions, Ni and Mn, involved in the oxidation/reduction processes. The cell without Li 3 PO 4 showed a lower discharge voltage plateau ͑Ͻ3.5 V͒ than the composite one. Although the Li 3 PO 4 film was so thin that it could be nearly removed with only 2 min of Ar etching in X-ray photoelectron spectroscopy, Li 3 PO 4 is thought to have a function as a solid electrolyte interface between LiNi 0.5 Mn 1.5 O 4 and SPE to prevent the degradation of solid polymer electrolyte.Lithium-ion cells, which consist of lithium transition-metaloxide positive electrodes and carbon negative electrodes, are widely used in portable electric devices. 1 However, the issue of the genuine safety of the cell must be resolved in scaling up of the cell for use in load leveling ͑LL͒ and zero-emission vehicles ͑ZEV͒. 2 Namely, the combination of a flammable organic electrolyte with a highly oxidized positive electrode can explode easily if sufficient care is not taken. Some safety systems have already been applied in commercialized cells, which have worked well in small-scale cells. However, not all of these safety systems may work well in the large-scale cell. Therefore, a solvent-free all-solid-state battery, which is the most suitable for a large-scale battery system, is needed. There are two types of lithium conductive solid electrolyte, the polymer type and the inorganic type. The polymer type was mainly developed for large-scale batteries because it easily tolerates volume change of the electrode materials during charge and discharge. However, combinating a high voltage positive electrode with the polymer electrolyte is difficult because of the poor oxidation resistivity of the ethylene oxide units, and therefore a 3 V class positive electrode, such as vanadium oxide, V 2 O 5 , was applied to the polymer-type system. 3 On the other hand, an inorganic electrolyte has high oxidation resistivity; for example, a lithium phosphorous oxynitride electrolyte ͑Li-PON͒ showed resistivity at over 5 V. 4,5 However, the ceramic electrolyte cannot tolerate the volume change of electrodes, particularly in large scale batteries, during charge and discharge. Thus, we proposed the concept of a composite in which an inorganic electrolyte is placed at the positive electrode surface and a polymer electrolyte is placed at the negative electrode surface. In this setup, a high voltage positive material can be utilized with a lithium metal negative electrode in an all-solid-state battery. Furthermore, the composite system has the potential to permit scaling of the cell due to the contribution of the polymer flexibility. We previously demonstrated the combina...