Table of contents

We report on the synthesis of a new relaxor compound with chemical formula Pb_0.78Ba_0.22Sc_0.5Ta_0.5O₃ (PBST). Single crystals were obtained by the high-temperature solution growth method. The structure of the new compound is of double-perovskite type with face-centred cubic symmetry at room temperature. The frequency dependence of the dielectric constant of PBST shows a strong dielectric dispersion in a wide temperature range and a dielectric-constant maximum near 200 K at 10 kHz. The local atomic environment was probed by Raman scattering and optical absorption spectroscopy. The results show that the incorporation of Ba deforms the BO₆ octahedra adjoining the BaO₁₂-polyhedra along the direction and shortens the Pb–O bond lengths next to the BaO₁₂-polyhedra within the {111} planes. The random substitution of Ba for Pb leads to a wider distribution in the size and shape of the ferroic species in PBST compared to stoichiometric PbSc_0.5Ta_0.5O₃. The addition of Ba shifts the optical absorption edge to lower energies and gives rise to extra absorption peaks at 460 and 730 nm. The latter peak is related to polar atomic rearrangements in the vicinity of the Ba ions embedded into the PbSc_0.5Ta_0.5O₃ matrix.

The surface of a Cu(311) crystal is formed by a regular distribution of narrow (111) terraces, giving rise to a well-ordered step superlattice. Angle-resolved photoemission experiments demonstrate the existence of a surface resonance related to the L-gap noble-metal surface state. The surface resonance is observed at the edge of the surface Brillouin zone, regularly repeated with the superlattice periodicity, and its wavefunction is referred to the Cu(311) mean surface. The intensity distribution of the surface state in reciprocal space has been probed using different photon energies. The results can be well understood on the basis of a simple model involving the bulk band properties.

The compound Nd₇Rh₃, crystallizing in a Th₇Fe₃-type hexagonal structure, has been shown recently by us to exhibit a signature of a 'magnetic phase-coexistence phenomenon' below 10 K after field cycling. This is uncharacteristic of stoichiometric intermetallic compounds, and bears a relevance to the trends in the field of 'electronic phase-separation'. In order to characterize this compound further, we have carried out dc magnetic susceptibility (χ), electrical resistivity, magnetoresistance and heat-capacity measurements as a function temperature (T = 1.8–300 K). The results reveal that this compound exhibits another unusual finding at the 10 K transition in the sense that the plot of χ(T) shows a sharp increase in the 'field-cooled' cycle, whereas the 'zero-field-cooled' curve shows a downturn below the transition. In addition, the sign of magnetoresistance is negative and the magnitude is large over a wide temperature range in the vicinity of the magnetic ordering temperature, with a sharp variation at 10 K. The results indicate that the transition below 10 K is first-order in its character.

We investigate various ultrafast optical processes in ferromagnetic (III,Mn)V semiconductors induced by femtosecond laser pulses. Two-colour time-resolved magneto-optical spectroscopy has been developed, which allows us to observe a rich array of dynamical phenomena. We isolate several distinct temporal regimes in spin dynamics, interpreting the fast (<1 ps) dynamics as spin heating through sp–d exchange interaction between photo-carriers and Mn ions while the ∼100 ps component is interpreted as a manifestation of spin–lattice relaxation. Charge carrier and phonon dynamics were also carefully studied, showing an ultrashort charge lifetime of photo-injected electrons (∼2 ps) and propagating coherent acoustic phonon wavepackets with a strongly probe energy dependent oscillation period, amplitude and damping.

BaLa₄Ti₄O₁₅ (BLT) is a compound with a Ti-deficient, layered perovskite-related structure. Dense BLT ceramics have been produced and their crystal structures and microstructures are investigated. The average macroscopic structure of BLT powder is considered to be hexagonal with space group , but there is some evidence in high-order x-ray diffraction peaks of asymmetry. Energy-dispersive x-ray analysis and Raman spectroscopy have illustrated that there is a heterogeneous distribution of constituent cations which may induce asymmetry as a result of the presence of two hexagonal phases of slightly different lattice parameter. Despite this inhomogeneity, BLT exhibits a microwave quality factor (Qf) of ∼44 000 GHz, a relative permittivity (ε_r) of ∼45, and a temperature coefficient of resonant frequency (τ_f) of −2 ppm °C⁻¹. In addition, anisotropy in the properties of the ceramic has been observed, with ε_r found to be larger when measured normal to the pressing direction.

Recent experiments have demonstrated the possibility of ultrafast non-thermal control of magnetization in rare-earth orthoferrites and ferrimagnetic garnet films via circularly polarized femtosecond laser pulses. Single and double pump pulses set up ultrafast magnetic fields via the inverse Faraday effect, thereby non-thermally exciting spin dynamics. A theoretical study of coherent control of the magnetization in rare-earth orthoferrites is performed by considering the effect of multiple pulses. The investigation is based on a model for orthoferrites recently proposed for the study of the inverse Faraday effect in the case of a single pump pulse. In the linear regime without damping, interferential effects take place: in-phase pulses induce a coherent enhancement of magnetization. The role of relaxation and nonlinearity is studied in relation to their capability of hampering coherent manipulation of magnetization. After many pulses, the effect of damping induces a stationary behaviour with a periodicity determined by the separation time between successive pulses. Due to nonlinear effects, the magnetization can be characterized by complex beating patterns whose amplitude and periodicity depend on the intensity of exciting pulses.

The magnetic properties of nanoparticles, including their superparamagnetic relaxation and spin orientation, have been found to depend on the aggregation state due to magnetic exchange coupling being established between surface atoms of neighbouring particles. We show that for samples of α-Fe₂O₃ nanoparticles the agglomeration and the interparticle exchange coupling are reversible processes. Grinding or ultrasonic treatment of aggregated particles can disperse the particles and reduce the magnetic coupling, while drying aqueous suspensions of dispersed particles leads to aggregation and re-establishment of magnetic coupling. The establishment of exchange coupling between neighbouring particles gives evidence for overlapping electronic orbitals of surface atoms of neighbouring particles. This is important not only for understanding the magnetic properties but also for understanding other physical properties e.g. the mechanical properties of dried aggregates. The reversibility (and the decoupling of particles) provides information on the strength of the coupling.

An ab initio study of the stability, structural, electronic, vibrational and optical properties of the most stable silicon–carbon binary nanoclusters Si_mC_n (m+n≤5) has been made. A B3LYP-DFT/6-311G(3df) method has been employed to optimize fully the geometries of the nanoclusters. The binding energies (BEs), highest occupied molecular orbital (HOMO)–lowest unoccupied molecular orbital (LUMO) gaps, bond lengths, ionization potentials (IPs), adiabatic electron affinities (EAs), vibrational frequencies, infrared intensities, relative infrared intensities and Raman scattering activities have been computed. In the more stable structures, the carbon atoms are in the majority whereas in the less stable structure the reverse is true. For the clusters containing all the carbon atoms except one silicon atom, the BE increases monotonically with the number of carbon atoms. The ground states of the clusters containing even numbers of the carbon atoms are, in general, lower than those containing odd numbers of carbon atoms. On the other hand, the lowest unoccupied states of the clusters containing even numbers of carbon atoms lie higher than those containing odd numbers of carbon atoms. All the predicted physical quantities are in good agreement with the experimental data wherever available. The growth of these most stable structures should be possible in the experiments.

Room-temperature Co K-edge and La L-edge x-ray absorption studies have been carried out on LaCoO₃. Experimental near-edge structures have been analysed by theoretical LDA+U density of states (DOS) and multiple scattering (MS) calculations. Use of both MS and DOS calculations yields additional information about hybridization of the states of the central atom with neighbouring atoms responsible for producing the near-edge structures. Absorption processes at the Co K-edge, and the La L₁-, L₂-, and L₃-edges have been attributed to electronic transitions from Co  4p, La 6p, La 5d, and La 5d, respectively. All the pre-edge and post-edge features including the shape of the main absorption edge have been generated by taking the convolution of the calculated DOS, indicating that single particle approximation is sufficient to express all experimentally observed major structures. Two pre-edge structures observed in the Co K-edge spectrum are attributed to Co  and quadrupole transitions in contrast to earlier identification of the same to Co  and e_g transitions. The influence of La 6p states on the Co 4p states is such that the inclusion of La atoms in the MS calculations is necessary to generate post-edge structures in the Co K-edge spectrum. The importance of the hybridization of the O 2p state with La 6p and 5d in the L-edge absorption processes is also discussed. A 10% contribution of the quadrupole channel has been estimated in the La L-edges.

The geometric and electronic properties of metal–carbon encaged fullerenes Sc₂C₂@C₈₄ have been studied using the density functional theory at the Becke exchange gradient correction and the Perdew–Wang correlation gradient correction function level with the double numerical atomic orbitals basis sets augmented by polarization functions. The Sc₂C₂ cluster was found to be stable in C₈₄ cage, while the cage expands slightly. The Sc₂C₂ cluster can rotate freely in the cage around the Sc–Sc axis which is coincident with the vertical principal axis of the cage. As the Sc₂C₂ cluster is encaged, the degeneracy of energy splits, and the HOMO–LUMO energy gap becomes smaller than that of the pure C₈₄, which suggests that Sc₂C₂@C₈₄ has higher reactivity than C₈₄. Based on our calculated results, the electronic structure of Sc₂C₂@C₈₄ might be formally described as (Sc₂C₂)⁺¹@(C₈₄)⁻¹ due to the charge transferring from the Sc₂C₂ cluster to C₈₄ cage.

It is argued that the large discrepancy (× 2 in strontium barium niobate SBN) between experimental values from different kinds of experiment for the critical exponent β describing the temperature evolution of the order parameter in relaxors such as SBN arises from the fact that such ferroelectric systems, assumed to be [3D] random field Ising models by Kleemann et al (2002 Europhys. Lett. 57 14), are not in thermal equilibrium. These arguments are illustrated primarily in SBN61—Sr_xBa_1−xNb₂O₆:Ce with x = 0.61 and Ce = ca. 0.7%. An alternative model of Levanyuk and Sigov for defect-dominated dynamics is invoked. The inferred dimensionality of domain walls is also addressed, and the possibility of a d = 5/2 universality class controlled by domain dimensionality is considered as an alternative to defect dynamics; inter alia, four sets of d = 5/2 exponents (with β = 1/2, 1/3, and 1/4) satisfy all known scaling and hyperscaling equalities. The present results support the scepticism about SBN critical exponents emphasized by Chao et al (2005 Phys. Rev. B 72 134105).

The magnetic properties of Nd_0.75Sr_1.25CoO₄ have been investigated by means of dc magnetization, ac susceptibility and electron spin resonance measurements. The specimen presented a deviation from the Curie–Weiss law at T_G∼230 K, ferromagnetic behaviour below T_C∼179 K, and a cusp structure at T_SG∼18 K in dc magnetization. The cusp was confirmed to be due to the existence of spin-glass states, and the deviation was argued to be a Griffiths singularity. Comparing the magnetic properties of the analogous layered cobalt oxides, the anomalous magnetic features of the specimen were attributed to the introduction of rare-earth ions, Nd³⁺.

The general shape of the temperature dependence of the static susceptibility in a biasing field conjugated to the order parameter is analysed with the use of the simplest equation of state compatible with the Widom and Griffiths scaling hypothesis. The corresponding curves are demonstrated to show from two to four inflection points, from which a discontinuous inflection point is found to occur exactly at the critical point whenever the critical exponent of susceptibility differs from one: . The unique inflection point occurring below the temperature of the maximum of the susceptibility in the case of the classical critical exponents, i.e. in the mean field theory, is also shown to be strictly independent of the biasing field. New scaling invariants related to the inflection points are found and their analytical expressions are given for the considered equation of state. The usefulness of the theoretical results to the analysis of experimental data is discussed.

Localization of electronic states in disordered thin layered systems with b layers is studied within the Anderson model of localization using the transfer-matrix method and finite-size scaling of the inverse of the smallest Lyapunov exponent. The results support the one-parameter scaling hypothesis for disorder strengths W studied and b = 1,...,6. The obtained results for the localization length are in quantitative agreement with both the analytical results of the self-consistent theory of localization and the numerical scaling studies of the two-dimensional Anderson model. The localization length near the band centre grows exponentially with b for fixed W but no localization–delocalization transition takes place.

CdTe films were deposited by closed space sublimation from different CdTe:Bi targets (from non-doped up to 0.16 at.%) previously sintered by the Bridgman method. X-ray diffraction measurements demonstrate that CdTe films are formed and Bi is incorporated. Electrical and optical characterizations show that thin films reproduce the bulk material behaviour with a decrease in resistivity and an increase in photoconductivity. Also a limit is found in the increase of photoconductivity properties versus Bi concentration.

First-principles calculations are made on the Ge_1−xMn_xTe diluted magnetic semiconductors with different compositions. The origin of their ferromagnetism is investigated and the carrier-induced ferromagnetism is developed for these diluted magnetic semiconductors. The Mn 3d states localized with divalent character are deep below the Fermi level, and the carriers introduced by defects are supported in our calculation. The p states of Ge also play an important role in the occurrence of ferromagnetism especially at high temperature. It is thought that Ge_1−xMn_xTe with a moderate composition (x = 0.51) of Mn atoms is a good candidate for DMS materials.

The interplay of spin, charge and lattice degrees of freedom in LaSr₂Mn₂O₇ (tetragonal, space group I4/mmm) yields a variety of properties such as a CE-type charge ordering and an A-type spin ordering. We have studied the effect of partial replacement of La by Pr on the magnetic and thermal properties of this system. Compounds of the type La_1−xPr_xSr₂Mn₂O₇ (0≤x≤1) have been prepared and are found to crystallize in the same tetragonal structure. Magnetization, electrical resistivity and specific heat measurements have been carried out on these compounds. Replacing La by Pr in this system affects both the electronic and the magnetic state of the system. It is observed that the charge ordering temperature and the net magnetic moment reduce with the incorporation of Pr in the lattice. In the antiferromagnetic state, the charge carriers in La_1−xPr_xSr₂Mn₂O₇ behave like a weakly scattered two-dimensional gas whose resistivity (ρ) is a logarithmic function of temperature, T (), over a certain temperature interval. This behaviour is enhanced with increasing concentration of Pr in the samples and is proposed to originate from quantum interference. The electronic part of the specific heat shows the presence of the two transitions at T_CO and T_N and a contribution from antiferromagnetic magnons at low temperatures.

We report on the first observation by Raman scattering of charge ordering in thin films of Nd_0.5Ca_0.5MnO₃. We observe the emergence of additional phonon excitations at low temperature that confirms the structural transition accompanying the charge ordering. The properties of our defect-free Nd_0.5Ca_0.5MnO₃ thin films are very similar to those of the single crystals. This behaviour appears to be different from previous Raman reports on Pr_0.5Ca_0.5MnO₃ and La_0.5Ca_0.5MnO₃ thin films. We speculate that the different growth conditions, in particular our higher substrate temperature, are responsible for such a difference.

The electron momentum distributions in icosahedral Al₆₄Cu₂₃Fe₁₃, icosahedral Al₆₃Cu₂₃Ru₁₃ and decagonal Al₆₅Cu₁₅Co₂₀ quasicrystals have been studied using the high-resolution Compton scattering technique. The electron-per-atom ratios (e/a) of the quasicrystals were determined quantitatively for the first time from the Compton profiles. The radii of the Fermi spheres were evaluated from the values of e/a on the basis of the free-electron model. Comparisons between the radius of the Fermi spheres and the size of the quasi-Brillouin zones show that the icosahedral quasicrystals meet the empirical matching condition, while the decagonal quasicrystal does not do this so well. This implies that the Hume-Rothery mechanism works for the formation of the pseudogap near the Fermi level in the icosahedral quasicrystals, although it operates only slightly in the decagonal quasicrystal.

Ferromagnetism with high Curie temperature T_c, well above room temperature, and very small saturation moment has been reported in various carbon and boron systems. It is argued that the magnetization must be very inhomogeneous with only a small fraction of the sample ferromagnetically ordered. It is shown that a possible source of high T_c within the ferromagnetic regions is itinerant electrons occupying a narrow impurity band. Correlation effects do not reduce the effective interaction which enters the Stoner criterion in the same way as in a bulk band. It is also shown how, in the impurity band case, spin wave excitations may not be effective in lowering T_c below its value given by Stoner theory. These ideas are applied to CaB₆ and a thorough review of the experimental situation in this material is given. It is suggested that the intrinsic magnetism of the B₂ and O₂ dimers might be exploited in suitable structures containing these elements.

We have revisited metallic ferromagnetism to examine the effects of band broadening due to electron–electron interaction on the magnetic phase diagram. Based on the generalized single band tight-binding Hamiltonian, we have explored the condition when the magnetization jumps discontinuously, by comparing results obtained by the local minima search and the total energy minimization. We have also investigated how the shape of the density of states (DOS) affects the magnetic phase diagram. We have found that a system with the DOS shape of a concave-type could have the discontinuous magnetization jump as the effective Stoner parameter increases.

First principles calculations, by means of the full-potential linearized augmented plane wave method within the local density approximation, were carried out for the electronic and optical properties of the filled tetrahedral compounds LiMgN, LiMgP and LiMgAs. The bandgap trend in the ternaries is found to be similar to the one encountered in the zinc-blende AlX. The assignment of the structures in the optical spectra and band structure transitions are investigated in detail. The predicted values of the dielectric constants for LiMgN, LiMgP and LiMgAs are close to those of the binary compounds AlN, AlP and AlAs.

In order to clarify the mechanism of the fast localized motion of hydrogen in the cubic (C15-type) Laves phase YMn₂, we have performed quasielastic neutron scattering measurements in YMn₂H_x (x = 0.6 and 1.2) over the temperature range 10–376 K. The behaviour of the elastic incoherent structure factor as a function of momentum transfer (studied up to Q_max≈3 Å⁻¹) is consistent with the two-site motion of H atoms within pairs of closely spaced interstitial g (Y₂Mn₂) sites. Comparison of the data for YMn₂H_0.6 and YMn₂H_1.2 shows that both the hydrogen jump rate and the fraction of H atoms participating in the fast localized motion decrease with increasing H content.

Nanostructured Fe and Fe₅₀Co₅₀ powders were prepared by high-energy ball milling. Microstructural and magnetic properties changes with milling time were followed by x-ray diffraction, differential scanning calorimetry and vibrating sample magnetometry. The nonequilibrium microstructure originates from a grain size reduction to about 12 nm and the introduction of internal strain up to 1.5% (root-mean-square strain). The occurrence of disorder in the ball-milled powders is evidenced by the broad exothermic reaction during the heating of ball-milled samples, the variation of lattice parameters and the increase of the saturation magnetization during the first 3 h of milling for Fe and continuously for the Fe₅₀Co₅₀ powder mixture. According to both the reduction of Fe Curie temperature, T_c, and the increase of the phase transformation , the paramagnetic temperature domain of nanostructured bcc α-Fe is extended by about 50 °C. The Fe₅₀Co₅₀ nanostructured powder behaves as a soft ferromagnet with low values of both the coercive field and the squareness ratio M_r/M_s.

Nanoparticles of NiFe₂O₄ are synthesized using a chemical route involving reactions of nitrates of Ni and Fe, and citric acid. The particles with size as small as 8 nm do not show any superparamagnetic relaxation in room temperature Mössbauer spectroscopy. This suggests that the particular preparation process yields a magnetic structure where the magnetic moments are strong enough to quench superparamagnetism. Over-occupancy in tetrahedral sites is suggested to be a possible explanation.

A method to calculate the quantum states of exciton–phonon complexes in semiconductor nanocrystals is presented. The exciton–phonon complexes are built from a basis set made of products of phonon states and electron–hole pairs, which are coupled through the electron–phonon Fröhlich interaction, and the electron–hole Coulomb and exchange interactions. In CdSe nanocrystals, the conduction band electrons are described by the effective mass equation, while the holes are represented by the spherical 4 × 4 Baldereschi–Lipari Hamiltonian. It is shown that a flexible and complete electron–hole basis, not limited to the 1s–1S_3/2 octet, is essential to obtain converged eigenvalues and the correct polaron shift to the exciton energy. A study of the spectral properties is presented; in particular, the spectral region which involves the lowest exciton–phonon complex eigenstates is analysed in details. Specifically, the non-adiabatic nature of the exciton–phonon dynamics in the nanocrystals examined is clearly shown by the vibron eigenstates that were obtained.

The phonon spectral function of the one-dimensional Holstein model is obtained within weak-coupling and strong-coupling approximations based on analytical self-energy calculations. The characteristic excitations found in the limit of small charge-carrier density are related to the known (electronic) spectral properties of Holstein polarons such as the polaron band dispersion. Particular emphasis is laid on the different physics occurring in the adiabatic and anti-adiabatic regimes, respectively. Comparison is made with a cluster approach exploiting exact numerical results on small systems to yield an approximation for the thermodynamic limit. This method, similar to cluster perturbation theory, confirms the analytical findings, and also yields accurate results in the intermediate-coupling regime.

The tight-binding Hamiltonian and retarded Green function formalism are used to study the local and total density of states of a monatomic quantum wire. An additional atom (adatom) can be coupled with the wire at its side. Also the local and average charges of the perfect and disturbed wire are shown and the analytical formula for the local density of states at each site is obtained. The local density of states at each atom in a wire which consists of an even number of atoms is characterized by minima at the Fermi level, in comparison with a wire disturbed by an adatom, where the density of states is characterized by local peaks. Charge fluctuations (waves) are observed in the wire for the case when the condition for the conductance oscillations is satisfied. For the single electron energies of a wire localized above the Fermi energy the local charge along the wire forms a wave curve which possesses one maximum less than for energies localized below the Fermi level.

We present an implementation of a method to determine different components of the local density of states in the presence of a core hole from inelastic x-ray scattering experiments. The only theoretical input needed beyond experimental data are the embedded atom transition matrix elements. In this work these matrix elements are calculated using a real-space multiple scattering approach. We apply the method to x-ray Raman scattering data for the boron K-edge in MgB₂. The capability of the approach is demonstrated by separating the density of π-, σ- and s-type states at the boron site in MgB₂.

We investigate temperature dependent electronic correlation effects in the conduction bands of gadolinium nitride (GdN) based on the combination of many-body analysis of the multi-band Kondo lattice model and the first principles TB-LMTO band-structure calculations. The physical properties like the quasi-particle density of states (Q-DOS), spectral density (SD) and quasi-particle band structure (Q-BS) are calculated and discussed. The results can be compared with spin and angle resolved inverse photoemission spectroscopy (ARIPS) of the conduction bands of GdN. A redshift of 0.34 eV of the lower band edge () is obtained and found to be in close agreement with earlier theoretical prediction and experimental values reported in the literature.

We calculate spin-Hall conductivities in two dimensional electron systems with Rashba spin–orbit interaction. The salient feature is that, apart from the usual spin-Hall conductivity σ_xy^z, which corresponds to the induction of out-of-plane spin current due to the application of transverse charge current, there is a novel spin-Hall conductivity , which arises due to the induction of transverse spin-polarized current in the transverse direction by the application of in-plane spin-polarized current. This phenomenon, which we call the 'spin–spin' Hall effect, is a spin analogue of the conventional Hall effect, but with no magnetic field. This contribution may be understood through the spin-diffusive equation.

A detailed study of the formation of neutral oxygen vacancies in monoclinic tungsten oxide (WO₃) is performed within the framework of the self-consistent first-principles SIESTA method. This work reveals that the neutral oxygen vacancies are anisotropic with a strong correlation to the structural anisotropy of the WO₃ monoclinic room temperature phase. We show that most of the structural relaxation around the vacancies occurs along a single straight line of W–O–W bonds and that the lowest energy corresponds to the formation of vacancies along the [001] crystallographic direction, where long and short W–O bonds alternate. Moreover, vacancies lead to the partial filling of the conduction band, in which 5d electronic states of the neighbouring W atoms dominate. In term of Mulliken population, the initial charge carried by the removed oxygen atom is almost recovered on the O and W atoms closest to the vacancy.

TlCu_2−xFe_xSe₂ is a p-type metal for x<0.5 which crystallizes in a body-centred tetragonal structure. The metal atoms are situated in ab-planes, ∼7 Å apart, while the metal–metal distance within the plane is ∼2.75 Å. Due to the large difference in cation distances, the solid solutions show magnetic properties of mainly two-dimensional character. The SQUID measurements performed for x = 0.27 give the c-axis as the easy axis of magnetization, but also show clear hysteresis effects at 10 K, indicating a partly ferromagnetic coupling. The magnetic ordering temperature T_c is 55(5) K as found from both SQUID and Mössbauer spectra. At the magnetic hyperfine fields are distributed with a maximum at about 30 T, which are compared to the measured magnetic moment per iron atom, which is 0.97 μ_B/Fe as found from SQUID measurements. The experimental results are compared to results using other methods on isostructural Tl selenides.

We report the results of studies of the effects of annealing conditions on the morphology of ferroelectric nanomesas. The nanomesa patterns were fabricated by self-assembly from continuous ultra-thin Langmuir–Blodgett films of copolymers of vinylidene fluoride and trifluoroethylene. Annealing in the paraelectric phase induced surface reorganization into disc-shaped ferroelectric nanomesas approximately 9 nm thick and 100 nm in diameter. Several factors affect the nanomesa dimensions, such as polymer composition, substrate material, deposition conditions, and annealing temperature. The height and diameter of the nanomesas both increase with increasing annealing temperature. Annealing studies in the ferroelectric–paraelectric coexistence region show that only the paraelectric phase is mobile. From this we conclude that the paraelectric phase supports a kind of plastic crystalline flow connected with dynamic disorder of the polymer conformation.

A first application of the multidimensional interpolation approach to a crystalline system, namely for the determination of the structural parameters of the high-pressure Cmcm phase of InAs, is reported. The method of local three-dimensional (3D) geometry investigation is based on the fitting of the experimental x-ray absorption near edge structure (XANES) data using a multidimensional interpolation of spectra as a function of structural parameters and full multiple scattering (MS) calculations. The procedure is divided into the following steps: the construction of an interpolation polynomial, the calculation of its energy-dependent coefficients using multiple scattering theory, and the minimization of the discrepancy between interpolated and experimental data varying structural parameters. Additionally, the method of XANES calculations has been tested for the reference phase of InAs (with Fm3m symmetry) with well-known structure. To achieve a higher sensitivity of the method on the structural parameters and to exclude the influence of non-structural factors on the results of the fitting, we have analysed the differences between spectra corresponding to the different crystal phases. The best-fit geometry with the cell parameters a = c = 5.366 Å and b = 5.411 Å and the internal cell parameter (which corresponds to the displacement of atoms in the y direction) δ = 0.025, has been found for InAs at 19 GPa.

This paper presents the result of theoretical and experimental studies of the magnetic field dependence of the exciton linewidth in Cd_1−yMg_yTe/Cd_1−xMn_xTe/Cd_1−yMg_yTe quantum wells. The reflection spectra from the structures with various values of x were measured at T = 2 K in magnetic fields up to 3.5 T applied parallel to the structure's growth axis. In the theoretical studies, the fluctuations of both concentration and spin projection of magnetic ions have been considered as the main reasons for the exciton line broadening. It is shown that the experimentally observed broadening of the σ⁻ exciton line and the narrowing of the σ⁺ exciton line with increasing magnetic field are in good agreement with theoretical calculations.

The electronic and magnetic structures of Fe_1−xCu_xCr₂S₄ (0.1≤x≤0.5) spinel sulfides have been investigated systematically by performing photoemission spectroscopy (PES), soft x-ray absorption spectroscopy (XAS), and soft x-ray magnetic circular dichroism (XMCD) measurements using synchrotron radiation. Cr and Cu ions are found to be nearly trivalent (Cr³⁺) and monovalent (Cu⁺), respectively, and their valence states do not change with x. The Fe 2p XAS spectra of Fe_1−xCu_xCr₂S₄ are very similar to that of Fe metal, indicating that the Fe 3d electrons are strongly hybridized to other valence electrons. The Fe and Cr 2p XMCD spectra show that the magnetic moments of Cr ions and Fe ions are aligned antiparallel to each other and that both the Cr and Fe magnetic moments increase with increasing x. The valence-band PES study reveals that the Cr³⁺ () 3d states are located at ∼1.5 eV below E_F. The occupied Fe 3d states consist of the broad states, the states at ∼4 eV below E_F, and the states very close to E_F. The filled Cu 3d¹⁰ states lie at ∼2.5 eV below E_F. This study suggests that the hybridized Fe and S 3p states near E_F play an important role in determining the transport properties of Fe_1−xCu_xCr₂S₄ for x≤0.5.

Electron paramagnetic resonance (EPR) measurements have been made on as-grown single crystals of perovskite-type TlCdF₃ at room temperature in the cubic phase. Signals from two kinds of tetragonal Gd³⁺ centre are observed. One centre observed in the Gd-only-doped crystal is ascribed to a Gd³⁺ ion at a Cd²⁺ site being associated with a vacancy at the nearest Cd²⁺ site. The other centre observed in the Gd, Li co-doped crystal is ascribed to a Gd³⁺ ion at a Cd²⁺ site being associated with a Li⁺ ion at the nearest Cd²⁺ site. The fine-structure terms are discussed by separating them into the uniaxial and cubic terms. Anomalous values of the axial and cubic parameters are obtained for TlCdF₃ in contrast with the parameters for other host crystals ACdF₃ (A = K, Rb, Cs).

The structural, electronic, and magnetic properties of FeSi are investigated by first-principles all-electron density functional calculations. The interplay between non-local exchange and correlations is studied by comparing results obtained with standard density functionals, the hybrid functional B3LYP and the Hartree–Fock ansatz. The calculations are performed using a local basis set. For the standard functionals, previous theoretical results for the non-magnetic semiconducting ground state are well reproduced. With the hybrid functional B3LYP, a magnetic metallic solution is found in addition to the non-magnetic low-energy semiconducting one. The influence of the non-local exchange is also reflected in the increased value of the indirect gap which separates the valence and conduction bands.

We analyse the role of orbital degeneracy in possible magnetic and orbital instabilities by solving exactly a two-site molecule with two orbitals of either e_g or t_2g symmetry at quarter-filling. As a generic feature of both models one finds that the spin and orbital correlations have opposite signs in the low-temperature regime when the orbitals are degenerate, in agreement with the Goodenough–Kanamori rules. While Hund's exchange coupling J_H induces ferromagnetic spin correlations in both models, it is more efficient for t_2g orbitals where the orbital quantum number is conserved along the hopping processes. We show that the ground state and finite temperature properties may change even qualitatively with increasing Coulomb interaction when the crystal field splitting of the two orbitals is finite, and the Goodenough–Kanamori rules may not be followed.

The local Fe structure and corresponding ferromagnetism are different for various concentrations of Fe-doped ZnO (Zn_1−xFe_xO, x = 0–0.07) films, which are prepared on LiNbO₃(104) substrates by reactive magnetron sputtering. X-ray photoelectron spectroscopy and x-ray absorption near-edge structure (XANES) reveal that, when x≤0.04, Fe is in the 2+ state and is incorporated into the wurtzite lattice of ZnO, and as x increases further, a second phase Fe₃O₄ is induced. Furthermore, full multiple-scattering substitution ab initio calculation of Fe K-edge XANES is used to confirm the local structure of Fe in films with different x. The single-phase Fe-doped ZnO films (x≤0.04) exhibit ferromagnetism above room temperature and the mechanism of bound magnetic polarons (BMPs) is proposed to discuss the magnetic properties. The presence of the second phase is responsible for the strong ferromagnetism for higher Fe concentration.

The local electronic structure of Nd₂CuO₄ is determined from ab initio cluster calculations in the framework of density functional theory. Spin-polarized calculations with different multiplicities enable a detailed study of the charge and spin density distributions, using clusters that comprise up to 13 copper atoms in the CuO₂ plane. Electron doping is simulated by two different approaches and the resulting changes in the local charge distribution are studied in detail and compared to the corresponding changes in hole-doped La₂CuO₄. The electric field gradient (EFG) at the copper nucleus is investigated and good agreement is found with experimental values. In particular a careful study of the various contributions to the EFG exhibits that the drastic reduction of the main component of the EFG in the electron-doped material with respect to La₂CuO₄ is due to a reduction of the occupancy of the 3d_3z²−r² atomic orbital. Furthermore, the chemical shieldings at the copper nucleus are determined and are compared to results obtained from nuclear magnetic resonance (NMR) measurements. The magnetic hyperfine coupling constants are derived from the spin density distribution calculated for different spin multiplicities.

Based on a proper definition of the spin current, we investigate the spin-Hall effect of heavy holes in narrow quantum wells in the presence of Rashba spin–orbit coupling by using a spin-density matrix approach. In contrast to previous results obtained on the basis of the conventional definition of the spin current, we arrive at the conclusion that an electric-field-induced steady-state spin-Hall current does not exist in both pure and disordered infinite samples. Only an ac field can induce a spin-Hall effect in such systems. Recently, the same conclusions were derived for the Rashba model, in which the spin splitting of electron states is linear in k.

The full text of this erratum can be found in the PDF file.