Michèle Gupta

Electronic structure and stability of hydrides of intermetallic compounds

Abstract We have investigated the electronic structure of several intermetallic compounds of AB and AB 5 type and their hydrides using the ab initio self-consistent linear muffin-tin orbitals method. The energy bands, total densities of states as well as their site and partial wave contributions are used to discuss the metal–hydrogen bonding features, the modifications of the electronic states at the Fermi energy on hydrogen absorption and the factors which control the stability. The specific systems investigated are zirconium-based AB compounds (AZr, BCo, Ni) and their hydrides ABH and ABH 3 . We also discuss the effect of substitution at the Ni site in LaNi 5 by several elements of the 3d series (Mn, Fe, Co). In Zr-based AB hydrides, the Zr–H bonding contribution is crucial to the stability of the system due to the lowering of the energy states below the Fermi energy. The total energies of ZrNiH calculated for different site occupancies by the H atoms are in agreement with experimental trends. For the intermetallic compounds LaNi 4 M (MMn, Fe, Co) the Fermi level is found to lie in a narrow M-3d sub-band above the Ni-d states, and the densities of states are high. The lattice expansion accounts for less than 50% of the decrease in the stability, indicating the importance of the chemical substitution effects, besides the lattice expansion contribution. For the Co-substituted compound, the (3g) sites are found to be more stable than the (2c) sites, in agreement with experimental observations, and the maximum hydrogen content of the hydride LaNi 4 CoH 4 appears to be associated with the filling of the Co-d peak.

The electronic structure of hydrogenated intermetallic compounds: Theory☆

Abstract We discuss the results of a theoretical investigation of the electronic properties of some intermetallic hydrogen storage compounds such as FeTiH, FeTiH 2 and Mg 2 NiH 4 . The metal-hydrogen and H-H interactions, which in all cases result in a low energy structure in the density of states, are analysed by means of a partial wave decomposition within the framework of the augmented plane wave method; this allows us both to gain insight into the preferential interactions of hydrogen with each of the two metal components and to predict the main features of the corresponding X-ray emission spectra. We discuss the change in the general position of the Fermi level in the metal d bands which occurs on hydrogenation of FeTi. The theoretical results are used to interpret successfully some experimental data for the electronic specific heat, the magnetic susceptibility and the Mossbauer isomer shifts. We found that cubic Mg 2 NiH 4 is a semiconductor, unlike the hydrides of FeTi which are metallic. This is due to complete filling of the four low energy metal-hydrogen-induced bands and the narrow nickel d bands. The consequences of these results on the hydrogen absorption capacity of intermetallic compounds of the same family are discussed.

A comparative study of the pressure-induced charge redistribution in YBa2Cu3O7 and YBa2Cu4O8

Abstract While the pressure dependence of T c in the single-chain superconductor YBa 2 Cu 3 O 7 is anomalously low, the double-chain compound YBa 2 Cu 4 O 8 exhibits one of the highest positive pressure dependences. Results of electronic structure calculations are presented in this paper in order to understand this difference in behavior, and to shed some light on the charge redistribution under pressure. Our calculations show that the charge redistributions under pressure are quite subtle, and that all sites participate in this charge redistribution. It is difficult to correctly understand the pressure dependence of T c without taking into account this charge redistribution. Roughly speaking, in the YBa 2 Cu 3 O 7 and YBa 2 Cu 4 O 8 families of superconductors, one can consider three separate structural components: CuO 2 planes, CuO chains, and the ionic elements Y, Ba. There is always an increase of the electronic charge at the ionic sites under pressure, and this tends to increase the hole density in the CuO 2 planes. The difference in behavior under pressure in YBa 2 Cu 3 O 7 and YBa 2 Cu 4 O 8 arises due to the chains. While in YBa 2 Cu 4 O 8 they further accept the electronic charge, and hence further increase the hole density in the CuO 2 planes, their behavior is quite the opposite in YBa 2 Cu 3 O 7 where they transfer back the electronic charge and almost compensate for the accumulation of the electronic charge at the ionic sites. The final result is that there is an extremely small increase in the hole density under pressure in YBa 2 Cu 3 O 7 . This difference in behavior of the chains is of structural origin, namely single chain versus double chains, which affects the densities of states in the vicinity of the Fermi level. At a pressure of 4.65 GPa in YBa 2 Cu 4 O 8 we obtain an increase in the hole density of ∼0.019 CuO 2 . This gives a T c ∼ 100 K, in good agreement with experiment.

An experimental and theoretical investigation of the densities of states of dimagnesium cobalt pentahydride

Abstract Recently synthesized and characterized dimagnesium cobalt pentahydride can be considered as a member of the isoelectronic structural series of Mg 2 FeH 6 , Mg 2 CoH 5 and Mg 2 NiH 4 . In this paper we present an investigation of the electronic structure of Mg 2 CoH 5 . An experimental analysis has been made through the soft X-ray Mg Kβ and Co Lα emission spectra. From these experimental distributions, once they have been corrected for oxide contributions, information is drawn about the densities of occupied Mg 3p and Co 3d states. These results are analysed in the light of an augmented plane wave (APW) study of the energy bands, density of states (DOS) and partial wave analysis of the DOS at the atomic sites of Mg 2 CoH 5 in its cubic phase. We discuss the nature of the metal-hydrogen bonds and the character of the occupied transition metal d states. Mg 2 CoH 5 is found to be non-metallic. The origin of the energy gap is ascribed to structural features already observed in isoelectronic compounds. The investigation of empty states also appears to be important since the presence of antibonding metal-d-hydrogen-s antibonding states above the energy gap is a characteristic of this series of compounds. Thus an attempt has been made to study the unoccupied states by an analysis of Mg K absorption and by reabsorption of the Co Lα emission in the case of cobalt.

Origin of the non-metallic behavior in Li2C60, Na2C60 and Mg2C60

Abstract While K-, Rb- and Cs-doped C 60 are metallic and superconducting, doping by Li, Na or Mg surprisingly results only in non-metallic phases. Results of our tight-binding electronic structure calculations show that the latter are band insulators due to the introduction of additional occupied bands either in the gap or just below the conduction band minimum of C 60 . The position of these bands depends critically on the location of the element in the periodic table. In the case of K, Rb or Cs, these bands are found at much higher energies in the conduction band.

The crystal and electronic structure of CaPd3H

Abstract A new calcium substituted palladium hydride, CaPd 3 H, has been synthesized and the metal structure was found to be of the Ni 3 Ti ( D 0 24 ) type. From a Guinier-Hagg X-ray powder diffraction pattern the unit cell dimensions were determined to be a = 5.8617(2) and c = 9.7793(6) A. Resistance measurements did not reveal any superconducting transition, but the alloy has metallic character. From considerations of bonding distances in the binary hydrides, the hydrogen atoms are restricted to the octahedral site coordinated by six palladium atoms. The experimental results were confirmed by theoretical calculations on the electronic structure, using the LMTO method. The d -electron bands are filled at the CaPd 3 H composition, thus limiting the hydrogen occupancy to only one H atom per formula unit.

Mechanism of hole doping in Hg-based cuprate superconductors

Abstract The new Hg-based cuprate superconductors are considered important materials for technological applications. These compounds can be obtained, for instance, from the double-layer Bi- or Tl-based superconductors by the replacement of a (Bi2O2)2+ or (Tl2O1)2+ layer by a single Hg2+ layer. Results of electronic structure calculations for the hole densities in the three Hg-based cuprate superconductors HgBa2CuO4+δ, HgBa2CaCu2O6+δ, HgBa2Ca2Cu3O8+δ are presented in this paper for two oxygen stoichiometries δ=0 and δ=0.125. We find that for δ=0 these compounds are practically undoped due to the absence of oxygen atoms in the plane of Hg. The non-stoichiometry (δ0) in these superconductors plays a crucial role and provides the only source of hole doping. Our calculations show that for δ=0.125 the total carrier density in the CuO2 planes per formula unit is essentially the same in all three compounds with a value ∼0.32 hole/formula unit. This value is larger than that expected on the basis of purely ionic considerations, due to the lowering of Hg states arising from HgO interactions. We also find that in the Hg 1223 superconductor the hole density is approximately uniformly shared between the three CuO2 planes.

Effect of zinc substitution on the carrier density in YBa2Cu3O7−δ superconductors

Abstract We show that the substitution of Zn at the Cu(2) sites in the CuO2 planes in YBa2Cu3O7 results in the collapse of the antibonding Cu–O band in the vicinity of Zn that causes a depletion in the hole carrier density in its immediate neighborhood and a non-uniform distribution in the CuO2 planes, while leaving the overall hole density practically unchanged. This local disruption of the electronic integrity of the CuO2 planes and the non-homogeneity in the carrier density that it causes are important factors that need to be also considered for an understanding of the mechanism of the depression of Tc due to Zn. Our results also give information concerning the superconductor–insulator transition observed in the underdoped compounds due to Zn, which we find to occur due to a large depletion of the hole density at the CuO2 units nearest neighbor to Zn to a critically low value at which the electron localization sets in.

Pressure dependence of the hole concentration in superconducting La1.85Sr0.15CuO4

Abstract Results of electronic structure calculations for La 1.85 Sr 0.15 CuO 4 are presented in order to understand the effect of pressure on the charge redistribution. Our calculations show that the effect of pressure on the charge redistribution are rather subtle. In particular, we find that, contrary to what has been believed, the planar copper-apical oxygen (Cu-O(2)) bond length does not play an important role in determining the charge redistribution under pressure. The deciding role is in fact exercised by the La-O bond lengths. The apical oxygens essentially serve as a conduit for charge transfer from the CuO 2 planes to the lanthanum sites under pressure. With the crystal structure data of both Pei et al. and Nelmes et al., which exhibit Cu-O(2) bond length variations in completely opposite directions with pressure, an increase in the hole concentration in the CuO 2 plane is obtained. However, assuming a linear relationship between T c and the hole concentration, we recover the pressure dependence of T c observed experimentally with the data of Pei et al. while with the data of Nelmes et al. a somewhat higher dependence is obtained.

Electronic structure investigation of the relationship between oxygen ordering and superconductivity in YBa2Cu3O6.5

Abstract Results of electronic structure calculations for the compound YBa2Cu3O6.5 are presented. Three models for the crystal structure have been considered: (a) an alternate chain (AC) model where each alternate chain is fully intact and the adjacent one fully empty, (b) an identical chain (IC) model where all chains are identical but are broken, and (c) tetragonal. Our calculations show that there is practically no charge transfer from the CuO2 planes towards the plane of the chains in the latter two cases. However, in the first case ∼ 0.18 holes/CuO2 are created. This shows that the ordered chains are absolutely essential for the superconductivity and the disorder in the chains destroys superconductivity.