著者
Ohara Koji Mitsui Akio Mori Masahiro Onodera Yohei Shiotani Shinya Koyama Yukinori Orikasa Yuki Murakami Miwa Shimoda Keiji Mori Kazuhiro Fukunaga Toshiharu Arai Hajime Uchimoto Yoshiharu Ogumi Zempachi
出版者
Nature Publishing Group
雑誌
Scientific reports (ISSN:20452322)
巻号頁・発行日
vol.6, 2016-02-19
被引用文献数
107

次世代硫化物ガラス電解質の構造解明に成功 -複雑なガラス構造中のリチウムイオン伝導制御に期待-. 京都大学プレスリリース. 2016-02-22.The atomic and electronic structures of binary Li2S-P2S5 glasses used as solid electrolytes are modeled by a combination of density functional theory (DFT) and reverse Monte Carlo (RMC) simulation using synchrotron X-ray diffraction, neutron diffraction, and Raman spectroscopy data. The ratio of PSx polyhedral anions based on the Raman spectroscopic results is reflected in the glassy structures of the 67Li2S-33P2S5, 70Li2S-30P2S5, and 75Li2S-25P2S5 glasses, and the plausible structures represent the lithium ion distributions around them. It is found that the edge sharing between PSx and LiSy polyhedra increases at a high Li2S content, and the free volume around PSx polyhedra decreases. It is conjectured that Li(+) ions around the face of PSx polyhedra are clearly affected by the polarization of anions. The electronic structure of the DFT/RMC model suggests that the electron transfer between the P ion and the bridging sulfur (BS) ion weakens the positive charge of the P ion in the P2S7 anions. The P2S7 anions of the weak electrostatic repulsion would causes it to more strongly attract Li(+) ions than the PS4 and P2S6 anions, and suppress the lithium ionic conduction. Thus, the control of the edge sharing between PSx and LiSy polyhedra without the electron transfer between the P ion and the BS ion is expected to facilitate lithium ionic conduction in the above solid electrolytes.
著者
INUI Takao KAJITANI Hisashi NARITA Hideaki MORI Kazuhiro
出版者
公益社団法人日本船舶海洋工学会
雑誌
Selected papers from the journal of the Society of Naval Architects of Japan
巻号頁・発行日
vol.9, pp.49-64, 1972-03

A set of three 2m models (M-8, 9, 10) are wave-analyzed, where importance is placed in the bow wave analysis (Part I) rather than in the resultant wave analysis (Part. II). The models are generated from the central vertical plane distribution of sources m(ξ,ζ), whose draftwise distribution is varied as U, V and Λ types under the restriction of ∫^1_0m(ξ) =m(ξ,ζ)dζ=fixed. The results obtained are as follows: Part I-Bow Wave Analysis: (1) The "measured" bow wave patterns show not only the parallel shift (Δx, Δy) but also the "local" distortion in co-ordination to the "calculated" wave patterns where Δy is roughly estimated as B/2. (2) The "measured" bow wave amplitude A_F(θ) is found smaller than the "calculated" amplitude throughout the whole range ofθ=0〜π/2. (3) The comparison among the three tested models shows that the linearized theory does not work so well as far as the frameline effect is concerned. This may suggest a need of some kind of empirical correction factor α_2(ζ) to be applied for the "wave-making" strength of sources, such as 0 < α_2(ζ)&ltlarr;- 1, for 0 &ltlarr;-ζ&ltlarr;- 1 withα_2(0)<1 ,α_2(1)= 1 , whereζ=f/T (f = immersion of source, T = draft of distribution plane). Part II-Resultant Wave Analysis: (1) The resultant of bow and stern waves of Model M-8, the proto-type, is wave-analyzed. (2) Newman's truncation formula is not valid except y=0. (3) The effect of finite transverse separation (y) is of less importance than the truncation error. (4) A remarkable discrepancy is observed between "measured" and "calculated" wave amplitudes, particularly in the smaller range of θor the transverse wave component. (5) Beside viscosity effect, sheltering effect is supposed to be the cause for this discrepancy.