Abstract: Atomic arrangements in the mixed-valence oxide U3O7 are refined from high-resolution neutron scattering data. The crystallographic model describes a long-range structural order in a U60O140 primitive cell (space group P42/n) containing distorted cuboctahedral oxygen clusters. By combining experimental data and electronic structure calculations accounting for spin–orbit interactions, we provide robust evidence of an interplay between charge localization and the magnetic moments carried by the uranium atoms. The calculations predict U3O7 to be a semiconducting solid with a band gap of close to 0.32 eV, and a more pronounced charge-transfer insulator behavior as compared to the well-known Mott insulator UO2. Most uranium ions (56 out of 60) occur in 9-fold and 10-fold coordinated environments, surrounding the oxygen clusters, and have a tetravalent (24 out of 60) or pentavalent (32 out of 60) state. The remaining uranium ions (4 out of 60) are not contiguous to the oxygen cuboctahedra and have a very compact, 8-fold coordinated environment with two short (2 × 1.93(3) Å) “oxo-type” bonds. The higher Hirshfeld charge and the diamagnetic character point to a hexavalent state for these four uranium ions. Hence, the valence state distribution corresponds to 24/60 × U(IV) + 32/60 U(V) + 4/60 U(VI). The tetravalent and pentavalent uranium ions are predicted to carry noncollinear magnetic moments (with amplitudes of 1.6 and 0.8 μB, respectively), resulting in canted ferromagnetic order in characteristic layers within the overall fluorite-related structure.

Abstract: Layered Li(M,Li)O2 (where M is a transition metal) ordered rock-salt-type structures are used in advanced metal-ion batteries as one of the best hosts for the reversible intercalation of Li ions. Besides the conventional redox reaction involving oxidation/reduction of the M cation upon Li extraction/insertion, creating oxygen-located holes because of the partial oxygen oxidation increases capacity while maintaining the oxidized oxygen species in the lattice through high covalency of the M–O bonding. Typical degradation mechanism of the Li(M,Li)O2 electrodes involves partially irreversible M cation migration toward the Li positions, resulting in gradual capacity/voltage fade. Here, using LiRhO2 as a model system (isostructural and isoelectronic to LiCoO2), for the first time, we demonstrate an intimate coupling between the oxygen redox and M cation migration. A formation of the oxidized oxygen species upon electrochemical Li extraction coincides with transformation of the layered Li1–xRhO2 structure into the γ-MnO2-type rutile–ramsdellite intergrowth LiyRh3O6 structure with rutile-like [1 × 1] channels along with bigger ramsdellite-like [2 × 1] tunnels through massive and concerted Rh migration toward the empty positions in the Li layers. The oxidized oxygen dimers with the O–O distances as short as 2.26 Å are stabilized in this structure via the local Rh–O configuration reminiscent to that in the μ-peroxo-μ-hydroxo Rh complexes. The LiyRh3O6 structure is remarkably stable upon electrochemical cycling illustrating that proper structural implementation of the oxidized oxygen species can open a pathway toward deliberate employment of the anion redox chemistry in high-capacity/high-voltage positive electrodes for metal-ion batteries. Upon chemical or electrochemical oxidation, layered LiRhO2 shows a unique structural transformation that involves both cation migration and oxidation of oxygen resulting in a stable tunnel-like rutile−ramsdellite intergrowth LiyRh3O6 structure. This structure demonstrates excellent performance with the steady and reversible capacity of ∼200 mAh/g. The stability of LiyRh3O6 is rooted in the accommodation of partially oxidized oxygen species through the formation of short O−O distances that are compatible with the connectivity of RhO6 octahedra.

Abstract: The chemical accessibility of the Celv oxidation state enables redox chemistry to be performed on the naturally coinagemetal -deficient phases CeM1-xSO (M = Cu, Ag). A metastable black compound with the PbFC1 structure type (space group P4/nmm: a = 3.8396(1) angstrom, c = 6.607(4) angstrom, V = 97.40(6) angstrom(3)) and a composition approaching CeSO is obtained by deintercalation of Ag from CeAg0.8SO. High-resolution transmission electron microscopy reveals the presence of large defect-free regions in CeSO, but stacking faults are also evident which can be incorporated into a quantitative model to account for the severe peak anisotropy evident in all the highresolution X-ray and neutron diffractograms of bulk CeSO samples; these suggest that a few percent of residual Ag remains. A strawcolored compound with the filled PbFCI (i.e., ZrSiCuAs- or HfCuSi2type) structure (space group P4/nmm: a = 3.98171(1) angstrom, c = 8.70913(5) angstrom, V = 138.075(1) angstrom 3) and a composition close to LiCeSO, but with small amounts of residual Ag, is obtained by direct reductive lithiation of CeAga8S0 or by insertion of Li into CeSO using chemical or electrochemical means. Computation of the band structure of pure, stoichiometric CeSO predicts it to be a Ce' compound with the 4f-states lying approximately 1 eV above the sulfide-dominated valence band maximum. Accordingly, the effective magnetic moment per Ce ion measured in the CeSO samples is much reduced from the value found for the Ce3+-containing LiCeSO, and the residual paramagnetism corresponds to the Ce3+ ions remaining due to the presence of residual Ag, which presumably reflects the difficulty of stabilizing Ce' in the presence of sulfide (S2-). Comparison of the behavior of CeCu0.8SO with that of CeCu0.8SO reveals much slower reaction kinetics associated with the Cu,_xS layers, and this enables intermediate CeCui LixSO phases to be isolated.

Abstract: CsTlCl3 and CsTlF3 perovskites have been theoretically predicted to be superconductors when properly hole-doped. Both compounds have been previously prepared as pure compounds: CsTlCl3 in a tetragonal (I4/m) and a cubic (Fm3̅m) perovskite polymorph and CsTlF3 as a cubic perovskite (Fm3̅m). In this work, substitution of Tl in CsTlCl3 with Hg is reported, in an attempt to hole-dope the system and induce superconductivity. The whole series CsTl1xHgxCl3 (x = 0.0, 0.1, 0.2, 0.4, 0.6, and 0.8) was prepared. CsTl0.9Hg0.1Cl3 is tetragonal as the more stable phase of CsTlCl3. However, CsTl0.8Hg0.2Cl3 is already cubic with the space group Fm3̅m and with two different positions for Tl+ and Tl3+. For x = 0.4 and 0.5, solid solutions could not be formed. For x ≥ 0.6, the samples are primitive cubic perovskites with one crystallographic position for Tl+, Tl3+, and Hg2+. All of the samples formed are insulating, and there is no signature of superconductivity. X-ray absorption spectroscopy indicates that all of the samples have a mixed-valence state of Tl+ and Tl3+. Raman spectroscopy shows the presence of the active TlClTl stretching mode over the whole series and the intensity of the TlClHg mode increases with increasing Hg content. First-principle calculations confirmed that the phases are insulators in their ground state and that Hg is not a good dopant in the search for superconductivity in this system.