Li 2 S is a high-capacity cathode material for lithium metal-free rechargeable batteries. It has a theoretical capacity of 1166 mAh/g, which is nearly 1 order of magnitude higher than traditional metal oxides/phosphates cathodes. However, Li 2 S is usually considered to be electrochemically inactive due to its high electronic resistivity and low lithiumion diffusivity. In this paper, we discover that a large potential barrier (∼1 V) exists at the beginning of charging for Li 2 S. By applying a higher voltage cutoff, this barrier can be overcome and Li 2 S becomes active. Moreover, this barrier does not appear again in the following cycling. Subsequent cycling shows that the material behaves similar to common sulfur cathodes with high energy efficiency. The initial discharge capacity is greater than 800 mAh/g for even 10 μm Li 2 S particles. Moreover, after 10 cycles, the capacity is stabilized around 500−550 mAh/g with a capacity decay rate of only ∼0.25% per cycle. The origin of the initial barrier is found to be the phase nucleation of polysulfides, but the amplitude of barrier is mainly due to two factors: (a) charge transfer directly between Li 2 S and electrolyte without polysulfide and (b) lithium-ion diffusion in Li 2 S. These results demonstrate a simple and scalable approach to utilizing Li 2 S as the cathode material for rechargeable lithium-ion batteries with high specific energy. ■ INTRODUCTIONRechargeable lithium-ion batteries have been widely used in portable electronics and are promising for applications in electric vehicles and smart grids. 1−4 However, due to limited capacity in both electrodes, the specific energy of Li-ion batteries needs to be improved significantly to fulfill the requirements in these applications. 5,6 Significant improvement has been achieved in the development of high-capacity materials to replace carbon-based anodes, such as silicon 7−12 and tin. 13 However, state-of-the-art cathode materials have a capacity less than one-half of the carbon anode. Accordingly, breakthroughs in cathodes are urgently needed to increase the specific energy of lithium-ion batteries. Current metal oxide and phosphate cathodes possess an intrinsic capacity limit of ∼300 mAh/g, with a potential of maximum 130% increase in the specific energy if all the capacity can be used. 14,15 In contrast, Li 2 S has a specific capacity of 1166 mAh/g, four times that of the limit in oxide/phosphate cathodes. 15,16 Considering pairing with Si anodes with 2000 mAh/g capacity, the specific energy of a Li 2 S-based lithium-ion battery could be 60% higher than the theoretical limit of metal oxide/phosphate counterparts ( Figure 1A, see Supporting Information for details) and three times that of the current LiCoO 2 /graphite system. Moreover, Li 2 S could be paired with a lithium-free anode, preventing safety concerns and low Coulomb efficiency of lithium metal in Li/S batteries. 17,18 The main hindrance for utilizing Li 2 S is that it is both electronically and ionically insulating. Therefore, Li 2 S was...

Rechargeable lithium-sulfur (Li-S) batteries hold great potential for high-performance energy storage systems because they have a high theoretical specific energy, low cost, and are eco-friendly. However, the structural and morphological changes during electrochemical reactions are still not well understood. In this Article, these changes in Li-S batteries are studied in operando by X-ray diffraction and transmission X-ray microscopy. We show recrystallization of sulfur by the end of the charge cycle is dependent on the preparation technique of the sulfur cathode. On the other hand, it was found that crystalline Li(2)S does not form at the end of discharge for all sulfur cathodes studied. Furthermore, during cycling the bulk of soluble polysulfides remains trapped within the cathode matrix. Our results differ from previous ex situ results. This highlights the importance of in operando studies and suggests possible strategies to improve cycle life.

Results are presented of single crystal structural, thermodynamic, and reflectivity measurements of the double-perovskite Ba2NaOsO6. These characterize the material as a 5d1 ferromagnetic Mott insulator with an ordered moment of ∼ 0.2 µB per formula unit and TC = 6.8(3) K. The magnetic entropy associated with this phase transition is close to Rln2, indicating that the quartet groundstate anticipated from consideration of the crystal structure is split, consistent with a scenario in which the ferromagnetism is associated with orbital ordering.PACS numbers: 75.50. Dd, 75.30.Cr, 71.70.Ej The interplay between spin, orbital and charge degrees of freedom in 3d transition metal oxides has proven to be a rich area of research in recent years. Despite the wide array of interesting physics found in these materials, much less is known about whether similar behavior can be found in related 4d and 5d systems, for which both the extent of the d-orbitals and larger spin-orbit coupling cause a different balance between the relevant energy scales. In this respect, oxides of osmium are of particular interest because the element can take formal valences from 4+ to 7+, corresponding to electron configurations 5d4 to 5d 1 . In this instance, we examine the simplest case of a 5d1 osmate for which the magnetic properties indicate that orbital ordering may indeed play a significant role.Simple oxides of osmium are typically Pauli paramagnets due to the large extent of the 5d orbitals. Examples include the binary oxide OsO 2 [1, 2] and the simple perovskites AOsO 3 (A= Sr, Ba) [3]. However, more complex oxides, including the double and triple perovskites La 2 NaOsO 6 [4], Ba 2 AOsO 6 (A = Li, Na) [5,6] and Ba 3 AOs 2 O 9 (A = Li, Na) [7], appear to exhibit local moment behavior. Presumably the large separation of Os ions in these more complex structures leads to a Mott insulating state, and indeed these and related materials are most often found to be antiferromagnetic. Of the above materials and their near relations containing no other magnetic ions, Ba 2 NaOsO 6 distinguishes itself as the only osmate with a substantial ferromagnetic moment (∼0.2 µ B ) in the ordered state [5].Weak ferromagnetism has been previously observed in other 5d transition metal oxides containing iridium. BaIrO 3 exhibits a saturated moment of 0.03 µ B , which has been attributed to small exchange splitting associated with charge density wave formation [8]. Sr 2 IrO 4 and Sr 3 Ir 2 O 7 exhibit similarly small saturated moments, attributed variously to either spin canting in an antiferromagnet due to the low crystal symmetry [9] or to a borderline metallic Stoner scenario [10,11]. The ferromagnetic moment in Ba 2 NaOsO 6 is substantially larger than in these materials. Furthermore, at room temperature the material has an undistorted double-perovskite structure, space group Fm3m (inset to Fig. 1) [5], in which OsO 6 octahedra are neither distorted nor rotated with respect to each other or the underlying lattice [12]. Such a high crystal symmetry, if preserve...