Search

Amongst the more than 18 different forms of water ice, only the common hexagonal phase and the cubic phase are present in nature on Earth. Nonetheless, it is now widely recognized that all samples of ‘cubic ice’ discovered so far do not have a fully cubic crystal structure but instead are stacking-disordered forms of ice I (namely, ice Isd), which contain both hexagonal and cubic stacking sequences of hydrogen-bonded water molecules. Here, we describe a method to obtain large quantities of cubic ice Ic with high structural purity. Cubic ice Ic is formed by heating a powder of D2O ice XVII obtained from annealing of pristine C0 hydrate samples under dynamic vacuum. Neutron diffraction experiments performed on two different instruments and Raman spectroscopy measurements confirm the structural purity of the cubic ice, Ic. These findings contribute to a better understanding of ice I polymorphism and the existence of the two natural ice forms. Structurally pure cubic ice Ic is obtained from the controlled heating of D2O ice XVII.

The reduction of Fe 2 O 3 and CoO oxides supported on carbon xerogels was studied to elucidate the effect of the nanoconfinement of the catalyst in carbon matrices. Resorcinol formaldehyde xerogels were synthesized, impregnated with iron and cobalt nitrates, and subsequently heated to obtain the oxides. The mechanism of oxide reduction to a metal with hydrogen was investigated by in-situ synchrotron X-ray diffraction in dynamic, non-isothermal conditions. Kinetic profiles of the reactions are obtained by plotting the diffraction intensities of selected Bragg peaks vs. temperature. The extracted Temperature-Programmed-Reduction (TPR) diagrams were best fitted by the Avrami-Erofeev and the n -order kinetic models for the iron and the cobalt oxide reduction reactions, respectively. The activation energies for the two-step reduction of Fe 2 O 3 to FeO and then to Fe are 80 and 15.1 kJ mol −1 . Such results may contribute to developing efficient and inexpensive catalysts based on non-noble metals like Fe, Co, via deposition of metal complexes on mesoporous supports.

Deuterated ice XVII, a metastable solid water polymorph, was filled with Ne and O2 at p ≈ 100 kPa and studied by in situ neutron diffraction (ILL, France). Powder patterns were collected in the ranges of 20-50 K (Ne) and 4.6-90 K (O2). Rietveld refinement and difference Fourier techniques showed that the gas molecules were located inside the hexagonal channels of the host ice. Both Ne atoms and O2 molecules are arranged in a spiral-like configuration off the channel axis, preserving the P6122 symmetry of the host in the case of Ne, but reducing it to P61 in O2. A larger Ne absorption compared to Ne-filled ice II is observed, which is consistent with longer host-guest contacts producing smaller hydrophobic repulsion. In O2-filled ice XVII, instead, short O-D distances (2.37 Å) have attractive character and stabilize the structure.

Experiments of deuterium absorption/desorption were performed on a foil of Pd 77 Ag 23 (wt%) alloy in the 78–196 °C temperature and 1–4 bar pressure ranges under the neutron beam at ILL (Grenoble). Powder diffraction patterns were collected on the D1B diffractometer ( λ = 1.2871 A) on the Pd 0.772 Ag 0.228 D ν deuterated alloy, and the face-centred-cubic structure was Rietveld-refined locating the D atom in the octahedral site with a variable occupancy ν (=D/M ratio). The results of ν ( T ) curves at different p (D 2 ) pressures are discussed in comparison with literature data on H and D absorption into Pd Ag alloys from thermodynamic measurements. A negative deviation from predictions of Sieverts' law is shown by the dependence of D occupancy on pressure. This effect relates the behaviour of the activity coefficient γ D to D–D and D–M interactions in the solid solution phase.

The deuteration reaction of a Pd 0.772 Ag 0.228 foil was studied by neutron powder diffraction measurements vs. time in different T and p(D 2 ) conditions (D20 and D2B high-resolution diffractometers at ILL, France). The system has great importance as permeable membrane for hydrogen separation. At p = 1 bar and T < 200 °C Bragg peak splitting indicates the presence of two cubic phases, to be identified with the α and β modifications of palladium hydride. Rietveld refinements provided the corresponding a α and a β lattice constants, the ν α and ν β occupancies of D, and the β phase volume fraction vs. time during deuteration progress. A continuous α → β transformation takes place till reaction end. At T = 200 °C or p > 1 bar a single phase was always observed. The phase diagram of Pd 0.772 Ag 0.228 D ν is thus shown to present a α+β two-phase domain less extended but similar to that of PdH ν , at variance with most thermodynamic measurements suggesting full single-phase behavior. The ν(t) kinetic data of deuteration at 80 °C/1 bar can be interpreted by a mechanism based on a first-order nucleation step followed by a zero-order step of surface adsorption and dissociation of the D 2 molecule.

Deuterated ice XVII, a metastable solid water polymorph, was filled with Ne and O-2 at p approximate to 100 kPa and studied by in situ neutron diffraction (ILL, France). Powder patterns were collected in the ranges of 20-50 K (Ne) and 4.6-90 K (O-2). Rietveld refinement and difference Fourier techniques showed that the gas molecules were located inside the hexagonal channels of the host ice. Both Ne atoms and O-2 molecules are arranged in a spiral-like configuration off the channel axis, preserving the P6(1)22 symmetry of the host in the case of Ne, but reducing it to P6(1) in O-2. A larger Ne absorption compared to Ne-filled ice II is observed, which is consistent with longer host-guest contacts producing smaller hydrophobic repulsion. In O-2-filled ice XVII, instead, short O-D distances (2.37 angstrom) have attractive character and stabilize the structure.

TiNb{sub 2}O{sub 7}, a good candidate as anode in lithium batteries, was treated with n-butyllithium to synthesize Li{sub x}TiNb{sub 2}O{sub 7} phases similar to those formed during electrochemical reactions. The Li{sub 2.67}TiNb{sub 2}O{sub 7} and Li{sub 3.33}TiNb{sub 2}O{sub 7} compounds, monoclinic C2/m, were studied by time-of-flight powder neutron diffraction. Their crystal structures, containing 3×3 blocks of (Ti,Nb)O{sub 6} octahedra, were Rietveld refined including Li positions. The Li atoms are distributed in similar amounts over sites with (Li–O) Coordination Numbers 5 and 4, although CN=5 should be preferred for having lower energy. Quantum-mechanical calculations were also performed, determining the average charge–discharge voltages to be 1.415 and 1.571 V for Li{sub 3.33}TiNb{sub 2}O{sub 7} and Li{sub 2.67}TiNb{sub 2}O{sub 7}, respectively, in good agreement with experimental results. An analysis of the theoretical charge distribution shows that, on lithiation, the chemical reduction of Ti and Nb atoms concentrates in the more condensed peripheral octahedra of the 3×3 block. This corresponds to electrons moving into a partly spin-polarized small band which gives rise to semi-metallic conductivity. - Graphical abstract: Structure of monoclinic Li{sub 3.33}TiNb{sub 2}O{sub 7}, and corresponding electron density of states. - Highlights: • Lithium was chemically inserted into the battery anode material TiNb{submore » 2}O{sub 7}. • Li atoms are distributed over four- and fivefold oxygen coordinated sites. • On lithiation Nb and Ti atoms are reduced according to a peculiar structural pattern. • Theoretically predicted charge–discharge voltages fit experiments very well.« less

Lithium was found to be inserted into FeNb11O29 by reaction with n-butyllithium, producing Li11FeNb11O 29 samples with good crystallinity, and the Amma crystallographic- shear structures of both compounds (a = 28.7093(6)/28.4036(4), b = 3.8256(1)/4.08447(9), c = 20.6241(4)/20.7067(2) Å, respectively) were Rietveld-refined by high-resolution neutron powder diffraction. Lithium atoms of Li11FeNb11O29 were located both in 4-fold- and in 5-fold-coordinated positions, lying respectively inside and between the 4 × 3 perovskite blocks of Nb(Fe)O6 octahedra. Electrochemical measurements on a FeNb11O29/ LixFeNb 11O29 electrode vs metallic Li showed that lithium can be intercalated/deintercalated reversibly in the 1.1-2.5 V range with a stable capacity of 185 mAh/g, which roughly corresponds to 11 Li atoms per f.u. as obtained by chemical lithiation. The rather uniform structural distribution found for Li atoms is consistent with the good electrode reversibility, which makes this material promising as anode in rechargeable lithium batteries. © 2014 American Chemical Society.

The rates of the irreversibile LiBH4 +Mg2FeH 6 → LiH + 2MgH2 + FeB + 5/2H2 and reversible (with significant sorption/desorption hysteresis) LiBH4 + 1/2Mg ↠LiH + 1/2MgB2 + 3/2H2 reactions were measured by isothermal-isobaric experiments in a Sievert-type apparatus. Measurements were done at several temperature T and overpressure Δp/p values, deriving the rate constants k(T, Δp/p) by Avrami's fitting of reaction advancement vs time. The results could be rationalized on the basis of the k = A exp(-Ea/RT) = A0 exp[-Ea0/RT + a(Δp/p)] kinetic formula, which couples the standard Arrhenius approach for thermal effects with an exponential dependence of the rate constant on overpressure. The empirical a coefficient varies with temperature in a way that requires the activation energy and entropy to depend linearly on Δp/p. For the first of the above reactions, Ea = -151(Δp/p) + 118 kJ mol-1 and ln(A/min-1) = -34(Δp/p) + 16; similar values are obtained for the second one. Relations of this kinetic model with the thermodynamic driving force ΔG and with equations of electrochemical kinetics, where overpressure is replaced by overvoltage, are discussed. © 2013 American Chemical Society.

First-principles quantum-mechanical calculations (CRYSTAL09 code, B3LYP functional) were performed on alkali titanates A2TinO 2n+1 with layered structure (n=3,4,6). Monoclinic structural types with unshifted (P21/m) and with shifted (C2/m) layers were considered. Crystal energies and full structural details were obtained for all Li, Na, and K phases. Neutron diffraction data were collected on powder samples of P21/m-Li2Ti3O7 (a=9.3146(3), b=3.7522(1), c=7.5447(3) Å, β=97.611(4)°) and C2/m-K 2Ti4O9 (a=18.2578(8), b = 3.79160(9), c=12.0242(4) Å, β=106.459(4)°) and their structures were Rietveld-refined. Computed energies show the P21/m arrangement as favoured over the C2/m one for n=3, and the opposite holds for n=6. In the n=4 case the P21/m configuration is predicted to be more stable for Li and Na, and the C2/m one for K titanates. Analysis of Li-O and K-O crystal-chemical environments from experiment and theory shows that the alkali atom bonding is stabilized/destabilized in the different phases consistently with the energy trend. © 2013 Elsevier Inc.

The dehydrogenation reaction of the 0.6LiBH4-0.4Mg(BH 4)2 eutectic system was investigated by Temperature-Programmed-Desorption and Pressure-Composition-Isotherm methods, in the range of 25-540 °C and 0.1-150 bar of p(H2). A sequence of four decomposition steps was found by TPD measurements; they occur at 235, 315, 365 and 460 °C for p(H2) = 3 bar, with a clear T decrease with respect to pure LiBH4 and Mg(BH4)2. In the PCI experiments, the first two steps could not be resolved but appeared merged in a single process. The amounts of H2 release at each step and the ΔrH and ΔrS values derived from van't Hoff plots were analyzed and compared with known results for relevant possible reactions. A scheme of interpretation was then proposed for all four processes. In particular, a fraction of LiBH4 and Mg(BH4)2 would react together in the range of 300-350 °C according to 2LiBH 4 + Mg(BH4)2 → 2B + 2LiH + MgB2 + 7H2, thus explaining the quite large H2 yield therein observed. The first and fourth steps correspond to decompositions of pure remaining Mg(BH4)2 and LiBH4, respectively, and the third one to dehydrogenation of MgH2 produced in the first step. © 2011, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.

The LiBH4?Mg(BH4)2 system has been investigated as a possible hydrogen storage material. Several composites were synthesized by ball milling, namely, xLiBH4?(1?x)Mg(BH 4)2 with x = 0, 0.10, 0.25, 0.33, 0.40, 0.50, 0.60, 0.66, 0.75, 0.80, 0.90, 1. The physical mixture was investigated by using X- ray powder diffraction and thermal analysis. Interestingly, already a small amount of LiBH4 makes the α to β transition of Mg(BH 4)2 reversible, which has not been reported before. The eutectic composition was found to exist at 0.50 < x < 0.60 exhibiting a eutectic melting at 180 °C. A phase diagram was built based on the data obtained in this study. Furthermore, the decomposition of the material begins right after the melting; thus, the decomposition temperature of the composite is much lower than those of the pure borohydrides. At 270 °C the x = 0.50 composite releases about 7.0 wt % of hydrogen. © 2011 American Chemical Society.

The Li-rich phases of superionic LLTO lithium conductor were modelled by a 2 a p × 2 a p × 4 a p perovskite supercell hosting one Li 5 La 9 ⎕ 2 (TiO 3 ) 16 formula unit. The disordered La/Li/⎕ distribution in the A-type perovskite cages was resolved into several ordered configurations, characterized by a short-range order parameter σ related to the number of nearest-neighbour La-vacant A sites. Quantum-mechanical ab initio calculations of the total energy were performed by the periodic code CRYSTAL09, on the basis of a B3LYP Hamiltonian and Gaussian-type localized basis sets. The structural configurations were least-energy optimized within the Pm monoclinic symmetry, finding a progressive stabilization for larger σ values, i.e. with clustering of La-vacancies and then with increasing connectivity of LaO 12 coordination polyhedra. By the frozen ion technique, one- and two-dimensional mechanisms of lithium diffusion within (001) layers of thickness a p were investigated. The lowest energy barrier of 0.31 eV, in good agreement with the experimental activation energy of 0.35 eV, was determined for the 1D pathways. The result holds both for one and for two mobile Li atoms per 2 a p × 2 a p layer unit-cell, according to a peculiar correlation between the motions of neighbouring ions.

Samples with MgmAlNin composition (m, n ≤ 3) were synthesized by ball milling in form of crystalline nanoparticles, and were found to be a single phase with partially disordered CsCl-type cubic structure. For compositions with large m and small n, minor quantities of Mg and Mg2Ni are segregated. XRD Rietveld refinements indicate that the two P m over(3, ̄) m sublattices basically have (Mg, Al) and (Mg, Al, Ni) occupations, respectively. Hydrogenation experiments on the Mg2AlNi2 sample by the PCI technique showed a reversible absorption/desorption of 1.4 wt% H2 with formation/decomposition of the MgH2 and Mg2NiH4 hydrides. The van't Hoff plot gives ΔrH = 67.7 kJ mol-1 (enthalpy of dehydrogenation), and T0 = 524 K (temperature of dehydrogenation at p = 0.1 MPa). A kinetic study of the gas release at 535 K showed a satisfactory behaviour, with a first fast step followed by a slower one in both the absorption and desorption processes. © 2008 Elsevier B.V. All rights reserved.

LiTi 2 NbO 7 was synthesized from CsTi 2 NbO 7 by direct Cs + /Li + ion exchange and subsequent thermal decomposition of the hydrated form. Neutron powder diffraction data were collected at high-resolution (ILL, France) and analyzed by Rietveld refinements and Fourier difference techniques, revealing a layer-like crystal structure (orthorhombic Pbnm , a = 9.2476(6), b = 16.955(2), c = 3.7542(2) A) partly similar to that of monoclinic LiTi 3 O 7 . Lithium is tetrahedrically coordinated and bridges adjacent layers of (Ti,Nb)O 6 octahedra. Nb atoms are strongly ordered in one of the three independent sites available for Ti/Nb, thus compensating for the unbalance of negative charge from the surrounding O atoms. Electrochemical measurements were performed on a LiTi 2 NbO 7 electrode vs. Li/Li + couple. Overlapping Ti 4 + /Ti 3 + and Nb 5 + /Nb 4 + redox processes occur around 1.4 V, with a specific charge of 245 mAh/g (about 2.8 electrons per f.u.) in the 2.30 to 1.15 V range. Charge-discharge cycling results show a reversible and stable specific capacity of 220 mAh/g at low current density, indicating that this material is a promising alternative to Li 4 Ti 5 O 12 spinel for reversible anode applications in lithium batteries.

Elemental Ge nanophases are obtained in silica starting from a sol-gel synthesis. The process comprises the preparation of silica xerogel doped by Ge(IV), substituting for Si(IV) in the oxide network, followed by reduction and phase separation during the sintering process by reaction with H-2 or CO. X-ray diffraction, Raman spectroscopy and high resolution transmission electron microscopy show that reactions with H-2 give rise in the porous silica network to nanometer-sized crystallization of elemental cubic germanium. (c) 2007 Elsevier Ltd. All rights reserved.

The vibrational spectrum of Mg2SiO4 olivine was calculated at the Γ point by using the periodic ab initio CRYSTAL program. An all electron localized Gaussian-type basis set and the B3LYP Hamiltonian were employed. The full set of frequencies (35 IR active, 36 Raman active, 10 "silent" modes) was computed and compared to experimental data from different sources (four for IR and four for Raman). A generally good agreement is observed with experiment (the mean absolute difference ranging from 7 to 10 cm-1 for the various sets), when some of the experimental frequencies, whose attribution is uncertain or appears to be affected by large errors, are not taken into account. A small number of observed peaks are not consistent with calculated frequencies, and a few theoretical peaks do not correspond to measured values. The implications are discussed in detail. The full set of modes are characterized using different tools, namely isotopic substitution, direct inspection of the eigenvectors and graphical representation, so as to obtain a consistent mode assignment. © Springer-Verlag 2006.

The Li+ ion conductor Li0.3La0.567TiO 3 was synthesized as a metastable phase quenched in liquid N 2 from 1300°C, and it was studied by time-of-flight powder neutron diffraction at high-resolution (HRPD instrument, ISIS Facility, U.K.). At room temperature a tetragonal perovskite superstructure was found, with a √2ap × √2ap × 2ap unit cell (a = 5.48027(1), c = 7.75591(2) Å, and Z = 4), I4/mcm space group, and a0a0c- tilt system of the TiO6 octahedra. Superlattice hkl reflections with odd l were systematically weaker and broader. By Fourier difference maps, the Li atom was located in a disordered position inside the La vacancy hollow, close to the bottleneck site at the center of the Ti4O4 square window. The Rietveld refinement converged to Rp = 6.56%, RBragg(even l) = 3.26%, and RBragg(odd l) = 11.6%. A two-dimensional pathway of Li+ ion mobility inside the structure, based on the coupled ordering of lithium and lanthanum, is proposed. At about 600°C the unit cell transforms into a p × ap × ap, with a disappearance of superlattice reflections. A tetragonal pseudo-cubic P4/mmm structure (a = 3.89644(1), c = 3.89748(2) Å, and Z = 1) was found (T = 650°C) and Rietveld refined to Rp = 6.10% and RB = 4.86%. Lithium was located on the layer of Ti atoms rather than on that of La as in the room-temperature structure. © 2006 American Chemical Society.

The five piezoelectric stress coefficients eik of orthorhombic Im 2m NaNO2 were calculated by ab initio quantum mechanical methods, employing the Berry phase theory, with an all-electron basis set of localized Gaussian-like functions and either a DFT-GGA or a Hartree-Fock Hamiltonian. DFT results are larger by about 30% than HF ones. The purely ionic and electronic contributions to piezoelectricity were evaluated by computing the internal and external (clamped-ion), respectively, components of each coefficient. In the e34 case the clamped-ion component is dominant, indicating that the electronic contribution prevails over the ionic one of opposite sign. In all other cases the overall piezoelectric effect is ruled by the ionic contribution. The full set of elastic constants was also computed, by means of which the piezoelectric strain coefficients dik could be derived. These are discussed and compared to experimental data from the literature. © 2005 IOP Publishing Ltd.

The monoclinic P21/m and orthorhombic Pmmn (Watanabe et al's-type) mechanisms of the high-pressure phase transition of NaCl between the B1 (rocksalt, Fm3̄m) and B2 (CsCl-like, Pm3̄m) cubic phases were investigated by ab initio DFT techniques with all-electron localized basis sets. Enthalpy profiles versus the order parameter were computed at constant pressures of 15, 26.3 (equilibrium) and 35 GPa for each pathway. The monoclinic path shows a lower activation enthalpy at the equilibrium pressure, but at different p values (hysteresis effects) the other mechanism becomes competitive. In the P21/m case, sharp jumps of structural parameters are observed along the transformation coordinate, which can be explained by a mechanism based on discontinuous sliding of alternating pairs of (100) atomic layers. This accounts also for the predicted formation of a metastable intermediate Cmcm phase with TlI-like structure, similar to that observed experimentally at high pressure in AgCl, and the relations with the KOH structure are discussed, too. On the other hand, along the Pmmn pathway the structural parameters vary quite smoothly, indicating a continuous motion of neighbouring atomic planes within the constraint of the additional mirror symmetry.

Powder neutron diffraction data at high resolution (HRPD instrument, ISIS Facility, U.K.) were collected on the lithium ion conductor α-LiZr2(PO4)3 at 673 K (a = 8.84448(3) A, c = 22.2875(1) A), and 873 K (a = 8.83667(4) A, c = 22.4177(2) A). Previous data measured at 423 K were included in the analysis. By Fourier maps and Rietveld refinement of the rhombohedral R3c crystal structure, it was shown that most lithium occupies the 6-fold disordered primary Li1 site, located off the symmetry center, at all temperatures. However, a 3-fold disordered secondary Li2 site was discovered, between two adjacent ZrO6 octahedra along the z axis, where a minor part of lithium is located. The fraction of lithium on Li2 increases with temperature from 10% (423 K) to 22% (873 K), and this may affect the mechanism of Li+ ion mobility. Li2 was not taken into account in the previous analysis of the 423 K data. Small but significant configurational changes of the Li1 and Li2 coordination environments occur between 423 and 87...

High-resolution powder neutron diffraction data (HRPD, ISIS Facility, U.K.) were collected on (H3O)Zr2(PO4)3 between 4 and 300 K. In the range 175-180 K, a ferroelastic first-order phase transition from the rhombohedral R3̄c NASICON structure to a monoclinic distorted one was discovered. Rietveld structure refinements were performed at 300 K (a = 8.74848(3), c = 23.7598(1) Å) and 4.5 K (space group C2/c or Cc, a = 15.0663(1), b = 8.7878(1), c = 9.3611(1) Å, β = 122.498(1)°). At RT, the hydronium ion H3O+ is pyramidal disordered over two inversion-related configurations. At LT, H3O+ becomes nearly planar, with two H atoms ordered and hydrogen-bonded to neighboring oxygens, and the third H atom disordered over two close positions with 2/3 and 1/3 statistical occupancies. Relations with the structural features of hydronium in other structures and the nature of order-disorder are discussed. An atomistic mechanism for proton conductivity is proposed.

The lithium conductors Li{sub 2}FeTi(PO{sub 4}){sub 3} and Li{sub 2}FeZr(PO{sub 4}){sub 3}, synthesized by solid-state reaction and characterized by X-ray powder diffractometry, were studied structurally at room temperature by neutron powder diffraction at high resolution (HRPD, ISIS Facility, U.K.). By trial-and-error and Rietveld refinements (R{sub p}=0.111, R(F{sup 2})=0.112), the first compound (orthorhombic Pbca, Z=8; a=8.5515(1), b=8.6229(1), c=23.9116(3) {angstrom}) was shown to have a complex superstructure sharing features of both the {alpha} and {beta} NASICON-type phases of LiZr{sub 2}(PO{sub 4}){sub 3}. Four (001) layers of PO{sub 4} and (Fe, Ti)O{sub 6} polyhedra are present per unit-cell, and they are related both by -1 inversion centers ({alpha} structure) and by a glide planes ({beta} structure). Ti{sup 4+} and Fe{sup 3+} order in the two interlayer regions, respectively. Owing to the structure complexity, only half of the lithium atoms could be refined in tetrahedral coordination with =1.99 {angstrom}. Li{sub 2}FeZr(PO{sub 4}){sub 3} (orthorhombic Pbna, Z=4; a=8.70559(8), b=8.78572(9), c=12.2202(1) {angstrom}) proved to be similar to {beta}-LiZr2(PO4)3; however, by Fourier synthesis and Rietveld refinement (R{sub p}=0.0618, R(F{sup 2})=0.0574) Li was located in a fully ordered tetrahedral configuration with =2.01 {angstrom}, instead of being disordered as in the phase of LiZr{submore » 2}(PO{sub 4}){sub 3}.« less

In order to develop new electrolytes for all-solid-state rocking chair lithium batteries, the NASICON-type compounds Li2FeZr(PO 4)3 and Li2FeTi(PO4)3 were investigated by powder X-ray diffraction technique and impedance spectroscopy. Li2FeZr(PO4)3 is orthorhombic Pbna (a=8.706(3), b=8.786(2), c= 12.220(5) Å) and Li 2FeTi(PO4)3) is orthorhombic Pbca (a=8.557(3), b=8.624(3), c=23.919(6) Å). They show no phase transitions from RT to 800°C. In the same temperature range logσT vs. 1/T show no slope variations. The activation energies for the ionic conductivity were 0.62 and 0.64 eV for Li2FeTi(PO4), and Li2FeTi(PO 4)3, respectively. In order to better evaluate the present results they were compared with those of α and ß-LiZr 2(PO4)3 phases, which were also prepared and characterised. A change of activation energy from 0.47 eV to 1.03 eV was observed in the case of ß phase, at about 300 °C; attributed to the ß (orthorhombic) ↔ ß' (monoclinic) phase transition. In the α phase the activation energy 0.47 eV in the temperature range 150-850 °C. The Li2FeZr(PO4)3 and Li 2FeTi(PO4)3 compounds can be interesting for applications as solid electrolytes in high temperature (> 300 °C) lithium batteries.

High-resolution powder neutron diffraction data have been collected T=423 K on the rhombohedral alpha phase of LiZr2(PO4)(3) (HRPD, ISIS Facility, UK). By Rietveld refinement analysis (space group R (3) over barc; Z=6; n=8.85493(3), c = 22.1440(1) Angstrom) and Fourier difference maps, lithium was found to be disordered over six positions related by (3) over bar symmetry in the large M' hole of the NASICON structure with centre at 0 0 0. The final agreement indexes are R-p=0.0352, wR(p) = 0.0436, R(F-2)= 0.1289 for 46 refined parameters. Li shows a very distorted tetrahedral coordination, with = 2.27 Angstrom; this result, together with the large number of empty sites available for hopping, due to disorder, accounts for the high Li+ mobility in the alpha phase. The order-disorder nature of the alpha-alpha' phase transition of LiZr2(PO4)(3) is discussed, and structural relations between NASICON and corundum alpha -Al2O3 are pointed out. (C) 2000 Elsevier Science B.V. All rights reserved

Neutron diffraction profiles at high resolution (HRPD, ISIS Facility, U.K.) were collected on powder samples of LiZr2(PO4)3 at 20°C (β′phase, monoclinic P21/n, Z=4; a=8.81277(4), b=8.94520(5), c=12.37540(6) A, β=90.801(1)°) and at 350°C (β phase, orthorhombic Pbna, Z=4; a=8.84303(5), b=8.94120(6 ), c=12.41301(8) A). All Li sites were located by difference Fourier maps in tetrahedral coordination, and both structures were Rietveld-refined to wRp=0.0353 (β′) and 0.0429 (β). The β′ structure is a distortion of β, with a [100] pseudo-twofold axis x, 1/2−y,−z relating all atoms but lithium, which is fully ordered (〈Li–O〉=2.02 A). In the β phase, Li is disordered over four sites, of which two (Li1 and Li2) are symmetry-independent with occupancies 0.34(1) and 0.16(1), respectively, and 〈Li–O〉=2.17 and 2.16 A. The disorder fully explains the higher Li+ mobility in the β with respect to the β′ phase observed from ionic conductivity data. Mechanisms of ion transport are proposed, and the relationship to the NASICON-type α′/α phase of LiZr2(PO4)3 is analyzed in detail.

Band structure and ground-state total energy calculations were performed on layered ${\mathrm{LiCoO}}_{2}$ (rhombohedral $R3\ifmmode\bar\else\textasciimacron\fi{}m)$ and ${\mathrm{Li}}_{0.5}{\mathrm{CoO}}_{2}$ (monoclinic $P2/m)$ by periodic Hartree-Fock methods, using effective core pseudopotential basis sets. Insulating and conducting behaviors are correctly predicted for the former and latter compound, respectively, on the basis of nonmagnetic closed-shell wave functions. Lithium diffusion was simulated in ${\mathrm{Li}}_{0.5}{\mathrm{CoO}}_{2}$ along the [100] and [110] directions, by relaxing all atomic positions in Pm and $P1\ifmmode\bar\else\textasciimacron\fi{}$ space groups and obtaining the corresponding energy profiles as a function of the ${\mathrm{Li}}^{+}$ position. An a posteriori density-functional theory-based correction for the correlation energy was included. The two-step diffusion mechanism $(\mathrm{octahedra}\stackrel{\ensuremath{\rightarrow}}{l}\mathrm{tetrahedra}\stackrel{\ensuremath{\rightarrow}}{l}\mathrm{octahedral}\mathrm{ }\mathrm{site})$ is predicted to be favoured with respect to the direct jump between octahedral sites, with an activation energy of 0.13 against 0.27 eV. A significant cobalt-oxygen charge transfer is observed during the diffusion process, and its meaning is analyzed.

High-resolution powder neutron diffraction data have been collected at room temperature on the triclinic alpha' phase of LiZr2(PO4)(3) (HRPB, ISIS Facility, U.K.). Rietveld refinement analysis (space group C (1) over bar; Z = 4; a = 15.0718(2), b = 8.8556(1), c = 9.1233(1) Angstrom, alpha = 89.660(1), beta = 123.912(1), gamma = 90.429(1)degrees) and Fourier difference maps have revealed the presence of monoclinic pseudo-symmetry -x, y, 1/2 - z for all atoms but lithium, which is disordered over two sites Li1 and Li2 with occupancies 0.71(3) and 0.29(3), respectively. The final agreement indexes are R-p = 0.0672, wR(p) = 0.0919, R(F-2) = 0.0603 for 92 refined parameters. Both Li1 and Li2 are in distorted tetrahedral coordinations, with (Li-O) = 2.09 and 2.18 Angstrom, respectively. Pairs of centrosymmetrical Li1 sites are located within half of the available M' holes of the NASICON structure, which are split into two independent sets by triclinic breaking of symmetry. This accounts for the lower conductivity of the triclinic phase. (C) 1999 Elsevier Science B.V. All rights reserved

The high-pressure equation of state of the normal spinels MgCr2O4 (picrochromite), MnCr2O4 and ZnCr2O4, and their reaction of decomposition into Cr2O3 (eskolaite) and MO (rocksalt-type) component oxides, were investigated by periodic unrestricted Hartree-Fock calculations. All-electron basis sets, and an a posteriori correction for the electron correlation energy, based on Density-Functional-Theory, were employed. Interpolation of results by the P-V Murnaghan equation of state yielded the equilibrium volume and energy, and the bulk modulus and its pressure derivative, for each of the seven phases (three spinels, three rocksalt oxides and eskolaite) considered. The simulated behaviour of interatomic distances vs pressure shows similar compressibilities of M-O bonds in both octahedral and tetrahedral coordinations. Binding energies and formation enthalpies of spinels from oxides are also computed and compared to available experimental data. The predicted decomposition pressures of Mg, Mn and Zn chromium spinels are 19, 23 and 34 GPa, respectively. The greater stability of ZnCr2O4 is related to Zn2+ being better suited to tetrahedral coordination than the other M2+ cations. Such results are strongly supported by the excellent agreement previously obtained between simulated (11 GPa) and experimental (13 GPa) pressures of the decomposition of MgAl2O4 spinel into corundum and periclase.

International audience; Monoclinic α-Nb2O5 was chemically lithiated by reaction with n-butyllithium, mimicking the product of electrochemical discharge of a niobium oxide cathode vs. a Li anode. The compound was investigated by neutron powder diffraction (D2B equipment at ILL, France) and its structure was Rietveld refined in space group P2 to wRp = 0.045, locating the Li atoms inserted in the α-Nb2O5 framework. The ensuing chemical formula is Li12/7Nb2O5. Some Li atoms are more strongly bonded (five coordinated O atoms), some are less strongly bonded (coordination number = 4). Starting from the experimental structure, first-principles periodic DFT calculations based on the hybrid B3LYP functional were performed. The electrochemical voltage of Li insertion was computed to be 1.67 V, fully consistent with the experimental 1.60 V plateau vs. capacity. The analysis of the electron band structure shows that lithiation changes the insulating oxide into a semi-metal; some of the extra electrons inserted with lithium become spin-polarized and give the material weak ferromagnetic properties.

Periodic bond chain (PBC) analysis is performed using a potential function fitted to elastic, structural and vibrational properties of aragonite. The resulting equilibrium form of the crystal, bounded by six forms having F character, is dominated by {0 1 1}, {1 1 0} and {1 0 2}, other forms being {1 1 1}, {0 1 0} and {0 0 1}, the corresponding specific surface energy γ has a mean value of ≈940erg/cm 2 . The theoretical growth shape is defined by five F-forms: {1 1 0}, {0 1 0}, {0 1 1}, {0 0 1}, {1 0 2} and a stepped one, {0 2 1}. Its habit depends on what value of the attachment energy of {0 1 1} is considered. The agreement between the theoretical growth morphology and that of natural crystals is not fully satisfactory: as a matter of fact three out of the eight forms bounding the averaged natural morphology (i.e., {1 1 1}, {0 1 2} and {1 2 1}) are missing in the theoretical growth morphology. From the observation of the natural gypsum/aragonite epitaxy, a simple model is proposed to explain the striking morphological difference between the pseudo-symmetrical zones [0 1 0] and [1 1 0] in natural aragonite crystals. The variation on the attachment energy of {0 1 0}, induced by water adsorption, improves the fit between theory and observation around these zones, especially when the distribution of their vicinal faces is considered. The same reasoning on water adsorption is extended to the specific surface energy of the crystal: a theoretical value of 330 erg/cm 2 is calculated for γ erystal/water of the face (0 1 0).

The ground-state electronic and magnetic properties of rhombohedral ${\mathrm{Ti}}_{2}$${\mathrm{O}}_{3}$ and ${\mathrm{V}}_{2}$${\mathrm{O}}_{3}$ have been investigated by ab initio all-electron periodic Hartree-Fock calculations. Both unrestricted open-shell and restricted closed-shell methods have been employed, with basis sets of atomic orbitals represented by contracted Gaussian-type functions. The ${\mathrm{t}}_{2\mathrm{g}}$ degeneracy of d electrons is removed by the rhombohedral field, giving rise to orbital ordering between ${\mathrm{a}}_{1\mathrm{g}}$ and ${\mathrm{e}}_{\mathrm{g}}^{\mathrm{\ensuremath{\pi}}}$ levels. The self-consistent-field variational solutions are spin-polarized insulating states with single (${\mathrm{a}}_{1\mathrm{g}}$) and double (${\mathrm{e}}_{\mathrm{g}}^{\mathrm{\ensuremath{\pi}}}$) electron occupations for ${\mathrm{Ti}}_{2}$${\mathrm{O}}_{3}$ and ${\mathrm{V}}_{2}$${\mathrm{O}}_{3}$, respectively. Conducting or semiconducting states, with different relative energies of ${\mathrm{a}}_{1\mathrm{g}}$ and ${\mathrm{e}}_{\mathrm{g}}^{\mathrm{\ensuremath{\pi}}}$ bands, have also been obtained by changing the c/a ratio of the hexagonal unit cell. The charge transfer into ${\mathrm{e}}_{\mathrm{g}}^{\mathrm{\ensuremath{\sigma}}}$ levels is discussed and compared to the ${\mathrm{Cr}}_{2}$${\mathrm{O}}_{3}$ and ${\mathrm{Fe}}_{2}$${\mathrm{O}}_{3}$ behavior.

Periodic unrestricted Hartree–Fock methods, with all-electron basis sets and aposteriori estimate of correlation energy, have been used to study the ground-state total energy of the title compounds with rhombohedral Rc corundum-type structure. Co2O3 and Ni2O3 are shown to be stable oxides with quite regular binding energies at the athermal limit. The relative stabilities of the AF1 (R3c), AF2 (R) and AF3 (R32) antiferromagnetic structures have been investigated, showing that Cr2O3 adopts the AF1 configuration, while Fe2O3, Co2O3 and Ni2O3 prefer AF2; this is accounted for in terms of σ- and π-type superexchange interactions. The spin polarization of electron transfer from O2- to Me3+ has been analysed, showing that only minority-spin electrons are transferred for Me=Fe, Co and Ni, while the total transfer is doubled for the other oxides by the majority-spin contribution.

The formation energies of the title alkali layered titanates (n = 3, 4, 6) are determined by DFT calculations for different monoclinic structure configurations.

The periodic unrestricted Hartree-Fock method has been adopted to calculate the ground-state spin-polarized wave function and total energy of rhombohedral Cr2O3 (corundum-type structure). Contracted Gaussian-type functions represent the atomic orbitals of Cr and O atoms. The least-energy structural configuration as a function of pressure has been derived, as well as elastic properties and equationof-state parameters. Results agree satisfactorily with experimental data. The relative stability of the antiferromagnetic and ferromagnetic spin arrangements is investigated. Density of electronic states, Mulliken population data and charge and spin density maps locate Cr2O3 at the borderline between Mott-Hubbard and charge-transfer insulators. Important differences appear in the covalent character of the chemical bond and in the spin polarization of the anions, when compared to Fe2O3.

Antiferromagnetic rhombohedral \ensuremath{\alpha}-${\mathrm{Fe}}_{2}$${\mathrm{O}}_{3}$ has been studied by calculations of the ground-state spin-polarized wave function and total energy, using the ab initio periodic unrestricted Hartree-Fock approach. All-electron basis sets of contracted Gaussian-type functions are employed to represent the O and Fe atoms (18 and 27 orbitals, respectively); Fe is alternatively described by a large-core pseudopotential plus 18 valence-shell orbitals. Computations have been performed for both the antiferromagnetic (AF) and ferromagnetic (FM) structures; the correct relative stability is reproduced, with \ensuremath{\Delta}E(AF-FM)=-0.0027 hartree per formula unit (including a correction for correlation energy). The dependence of \ensuremath{\Delta}E(AF-FM) on variations of the Fe-O bond lengths and Fe-O-Fe' angles involved in superexchange is analyzed, finding that for some configurations the FM structure becomes more stable than the AF one. The athermal equation of state, equilibrium crystal structure, elastic bulk modulus, and binding energy have been computed and compared to experimental quantities. An analysis of the density of electronic states show that the band gap is of p-d rather than d-d type, confirming the charge-transfer-insulator nature of hematite as inferred from photoelectron spectra. The overall shape of the valence band is also fully consistent with spectroscopic results. Mulliken electron population data indicate a charge back transfer of 0.29\ensuremath{\Vert}e\ensuremath{\Vert} from ${\mathrm{O}}^{2\mathrm{\ensuremath{-}}}$ to the d shell of ${\mathrm{Fe}}^{3+}$, causing a partial spin pairing with a magnetic moment of 4.7${\mathrm{\ensuremath{\mu}}}_{\mathit{B}}$.

Joint decomposition of hydrides may be energetically favored, if stable mixed compounds are formed. This "hydride destabilization" improves the energetics of H2 release from hydrogen storage materials. The sequence of dehydrogenation reactions of the 2LiBH4-Mg 2FeH6 composite was studied by PCI (pressure-composition- isotherm) and TPD (temperature-programmed-desorption) techniques in a Sievert apparatus. Produced phases were identified by ex-situ X-ray diffraction and FTIR spectroscopy. Three distinct plateaus are detected on each isotherm: A, B, and C on decreasing pressure. The A reaction, involving formation of FeB, MgH 2, and LiH, occurs at higher pressure/lower temperature than dehydrogenation of either pure hydrides; these are then effectively destabilized thermodynamically. The B process is plain decomposition of MgH2, and in C, the magnesium produced reacts with left LiBH4 forming MgB 2 and LiH. The B + C sequence is fully reversible, and it corresponds to two-step dehydrogenation of the LiBH4/MgH2 system. Reaction enthalpies and entropies were obtained through van't Hoff plots of all processes, thus providing a full thermodynamic characterization of the system. Interestingly, it ensues that destabilization of pure hydrides with reaction A is due primarily to an entropic rather than enthalpic effect. © 2012 American Chemical Society.

In the present study we report neutron spectroscopic measurements on polycrystalline lithium imide, namely, incoherent inelastic neutron scattering at 20 K, and neutron Compton scattering from 10 K up to room temperature. From the former technique the H-projected density of phonon states up to 100 meV is derived, while the latter works out the spherically averaged single-particle (i.e., H, Li, and N) momentum distributions and, from this, the mean kinetic energies. Only for H at the lowest investigated temperature, non-Gaussian components of its momentum distribution are detected. However, these components do not seem directly connected to the system anharmonicity, being fully compatible with the simple N-H bond anisotropy. Neutron data are also complemented by ab initio lattice dynamics simulations, both harmonic and, at room temperature, carried out in the framework of the so-called quantum colored noise thermostat method. The single-particle mean kinetic energies in lithium imide as a function of temperature show a quite peculiar behavior at the moment not reproduced by ab initio lattice dynamics methods, at least as far as H and Li are concerned. As matter of fact, neither their low temperature values nor their temperature trends can be precisely explained in terms of standard phonon calculations. © 2012 American Institute of Physics.

The ground-state crystal energies of cubic ${\mathrm{MgAl}}_{2}$${\mathrm{O}}_{4}$ (spinel) and MgO (periclase) and of rhombohedral \ensuremath{\alpha}-${\mathrm{Al}}_{2}$${\mathrm{O}}_{3}$ (corundum) have been calculated at different volumes, relaxing the corresponding structures, by all-electron periodic Hartree-Fock methods (crystal program). Basis sets of contracted Gaussian-type functions are employed for the 18 atomic (including d) orbitals representing each of the Mg, Al, and O atoms. Mulliken net atomic charges ${\mathit{z}}_{\mathrm{Mg}}$=1.86\ensuremath{\Vert}e\ensuremath{\Vert} (MgO), ${\mathit{z}}_{\mathrm{Al}}$=2.30\ensuremath{\Vert}e\ensuremath{\Vert} (\ensuremath{\alpha}-${\mathrm{Al}}_{2}$${\mathrm{O}}_{3}$), ${\mathit{z}}_{\mathrm{Mg}}$=1.74\ensuremath{\Vert}e\ensuremath{\Vert}, and ${\mathit{z}}_{\mathrm{Al}}$=2.24\ensuremath{\Vert}e\ensuremath{\Vert} (spinel) are obtained. The elastic bulk modulus, the Murnaghan equation of state p(V) at the athermal limit, the Mg-O and Al-O bond compressibilities, and the binding energy have been derived for each phase (and the elastic constants ${\mathit{C}}_{11}$ and ${\mathit{C}}_{12}$ for spinel only). Comparison with existing experimental data is discussed. The enthalpy change for spinel decomposition into the simple oxides has been computed as a function of pressure, including a correction for the electron correlation energy based on local-density-functional theory. A decomposition pressure of 11 GPa at T=0 K is predicted, against values of 8 and 13 GPa derived from experimental thermodynamic data and from direct compression experiments, respectively.

A powder sample of 2M 1 muscovite, (K 0.90 Na 0.07 )(Al 1.63 Fe 0.23 Mg 0.16 Ti 0.03 )(Si 3.20 Al 0.80 )O 10 (OH) 2 , was studied by time-of-flight neutron diffraction at 1 bar and at 2 (± 0.1) GPa. Rietveld refinements of the crystal structure were performed, varying all atomic coordinates in the room-pressure case (profile R w = 0.013, Bragg intensities R w = 0.040) and only the z coordinates of basal O atoms plus xyz of O and H of the OH group at 2 GPa (profile R w = 0.035, Bragg intensities R w = 0.181). In both cases the lattice constants were refined: a = 5.2108(4), b = 9.0399(8), c = 20.021(2) A, β = 95.76(1)° (1 bar), a = 5.187(2), b = 8.995(4), c = 19.502(4) A, β = 95.78(2)° (2 GPa). The unit-cell compression shows deviations from linear elastic behaviour. Inter-layer K-O bonds are compressed along z more than the c lattice constant. The O-H group forms an angle of 85(1)° with c* and establishes three very loose contacts (H..O from 2.62 to 2.67 A) with neighbouring O atoms. This environment does not change appreciably at 2 GPa

La scienza dei materiali e nata nella seconda meta del Novecento, con la caratteristica fondamentale di essere interdisciplinare fra la fsica e la chimica della materia solida. Il grande sviluppo autonomo di questo settore scientifico, e il fatto che in esso la distinzione tra aspetti fsici e aspetti chimici perda spesso significato, ha giustifcato la creazione di una nuova disciplina indipendente. L’altro connotato centrale della scienza dei materiali e la sua stretta relazione con gli aspetti tecnologici dei fenomeni che sono suo oggetto di studio, come il concetto stesso di “materiale”, intrinsecamente legato alle proprieta applicative, chiaramente testimonia. Bastera pensare ai semiconduttori, materiale base dei dispositivi elettronici e informatici, ai polimeri (materie plastiche), alle leghe metalliche speciali, ai ceramici avanzati, ai vetri per la fotonica, agli ossidi intercalati per le batterie ricaricabili al litio, e a tanti altri esempi consimili.