Materials built from MO6 octahedra linked to XO4 tetrahedra are good candidates for studying the different factors that determine the electrode potential. Among them, olivine-like LiMPO4 (M = transition metal) phosphates are especially interesting. When pressure is applied to LiMPO4 (M = Ni and Fe), a phase transition is induced. However, instead of the well-known olivine ⇔ spinel transformation, a transition to a new phase is observed (β‘). The arrangements of the metal ions (including phosphorus) in the two structures are very similar; thus, the main difference between them is due to the oxygen arrangement in a similar matrix. Raman spectroscopy has confirmed the structural model proposed for the high-pressure phase, in particular the modification in the lithium coordination from 6- to 4-fold upon synthesis under pressure. Among the olivines LiMPO4 (M = Mn, Ni, and Fe), the iron-containing one is only active up to 5.1 V. On the other hand, none of the high-pressure materials is electrochemically active; this can be explained by the change in the electrostatic field at the transition metal position.

Two new double tungstates, Li2Co(WO4)2 and Li2Ni(WO4)2, have been prepared by solid-state reaction and characterized. The structures of these isostructural compounds (triclinic, space group P1̄) have been determined from X-ray and neutron diffraction data and found to be built up of alternating layers of zigzag rows of edge-sharing WO6 octahedra and MO6 octahedra (M = Co2+ or Ni2+) as in wolframite-like phases. However, the MO6 octahedra are arranged in columns but not connected to each other; perpendicular to these columns there are rows of edge-sharing LiO6 octahedra which also connect the different MO6 octahedra. The structure of a previously reported lithium−copper(II) double tungstate with the same stoichiometry, Li2Cu(WO4)2, has also been determined and found to be similar to that proposed for these new double tungstates. The three compounds melt incongruently at temperatures between 750 and 770 °C. Conductivity measurements revealed that these compounds are not good ionic conductors probably because of the full occupancy of lithium positions which hinders the motion of the ions along the (LiO6)∞ columns.

Li 3 Fe͑MoO 4 ͒ 3 undergoes a complex electrochemical reaction with lithium in which the reduction of Fe 3+ and Mo 6+ takes place along the first discharge of the cell at about 2.4 and 1.8 V, respectively. The intercalation process involved is fully reversible for low lithium contents, Li 3+x Fe͑MoO 4 ͒ 3 with 0 Ͻ x Ͻ 1, the inserted compound Li 3+1 Fe 2+ ͑MoO 4 ͒ 3 retaining the Li 3 Fe 3+ ͑MoO 4 ͒ 3 parent framework with only a slight increase of the cell volume ͑0.85%͒. In contrast, the electrochemical reaction of Li 3 Fe͑MoO 4 ͒ 3 with five lithium ions originates an irreversible decomposition of this material into a mixture of a FeO-type compound and an amorphous lithium-molybdate phase. This in situ formed blend is electrochemically active, being able to intercalate and deintercalate three lithium ions at an average voltage of 2 V ͑reversible specific capacity of 150 Ah/kg͒. The full discharge of the cells ͑down to the vicinity of 0 V͒ proceeds through the complete and irreversible reduction of Li 3 Fe͑MoO 4 ͒ 3 with 25 lithium ions, resulting in the breakdown of any existing crystalline framework.

Li 3 Fe͑MoO 4 ͒ 3 reversibly inserts 1 lithium ion per formula down to 2 V vs Li. A preliminary room temperature phase diagram for Li 3+x Fe͑MoO 4 ͒ 3 ͑0 Ͻ x Ͻ 1͒ is constructed combining electrochemical and in situ X-ray diffraction results. Single-phase regions are detected at x = 0, 0.75, and 1. The crystalline structure of the final compound Li 3+1 Fe͑MoO 4 ͒ 3 is derived from that of Li 3 Fe͑MoO 4 ͒ 3 by completely filling the tunnel formed by square pyramidal sites along the a-axis of the structure. The close structural relationship between the host and inserted compounds ensures a good capacity retention of this material over prolonged electrochemical cycling in lithium cells.Since the demonstration of reversible electrochemical lithium insertion-extraction in LiFePO 4 in 1997, 1 compounds built up from polyatomic anions have attracted much attention as an alternative to LiMO 2 ͑M = transition metal͒ layered materials as positive electrode for rechargeable lithium batteries. Materials where the redox activity is due to the Fe +2 /Fe +3 couple seem to be the most attractive in terms of electrochemical performance, low cost, and environmental friendliness. Besides iron phosphates, 2-9 the electrochemistry of iron molybdates Fe 2 ͑MoO 4 ͒ 3 , 10,11 FeMoO 4 , 12 and more recently Li 3 Fe͑MoO 4 ͒ 3 have also been investigated. 13 The compound Li 3 Fe͑MoO 4 ͒ 3 presents a NASICON-related structure 14,15 built up from interconnected octahedra, tetrahedra, and trigonal prisms, displaying good lithium conductivity that appears to be related to the motion of lithium within the prismatic channels along the a axis ͑ 300°C = 6.6 ϫ 0 −7 Scm −1 , 600°C = 1.4 ϫ 10 −3 Scm −1 ͒. 16 In a previous paper 13 we have shown that the high mobility of lithium ions is also evidenced during the insertion of lithium, up to 1 ion per formula unit, that is associated with the reduction of Fe +3 to Fe +2 . Lithium intercalation within Li 3+x Fe͑MoO 4 ͒ 3 in the range 0 ഛ x ഛ 1 proceeds with slight structural modifications, excellent reversibility, and fast kinetics, characteristics that make this compound attractive as electrode material in electrochemical devices. To complete the study of this compound, it is important to fully understand the evolution of the electrode upon lithium insertion. With the aim of analyzing the structural changes induced by lithium insertion into Li 3 Fe͑MoO 4 ͒ 3 , and to identify the different phases eventually formed along the intercalation reaction, in this work we have followed the lithium insertion process by in situ synchrotron X-ray diffraction ͑SXRD͒. As it will be shown, although the framework of Li 3 Fe͑MoO 4 ͒ 3 is retained all along the insertion reaction, at least two other single phases are formed: Li 3+0.8 Fe͑MoO 4 ͒ 3 and Li 3+1 Fe͑MoO 4 ͒ 3 . ExperimentalThe starting material Li 3 Fe͑MoO 4 ͒ 3 was prepared by solid-state reaction from Fe 2 O 3 , LiNO 3 , and Mo 7 O 24 ͑NH 4 ͒ 6 ·4H 2 O. Stoichiometric mixtures of these reactants were uniaxially pressed into 12 mm diameter pellets and fired i...