Table of contents

A method for constructing semianalytical strongly correlated wavefunctions for single and molecular quantum dots (QDs) is presented. It employs a two-step approach of symmetry breaking at the Hartree–Fock level and of subsequent restoration of total spin and angular momentum symmetries via projection techniques. Illustrative applications are presented for the case of a two-electron helium-like single QD and a hydrogen-like QD molecule.

This paper reviews the recent developments in first-principles total energy studies of the phenomenological equilibrium 'doping limit rule' that governs the maximum electrical conductivity of semiconductors via extrinsic or intrinsic doping. The rule relates the maximum equilibrium carrier concentrations (electrons or holes) of a wide range of materials to their respective band alignments. The microscopic origin of the mysterious 'doping limit rule' is the spontaneous formation of intrinsic defects: e.g., in n-type semiconductors, the formation of cation vacancies. Recent developments in overcoming the equilibrium doping limits are also discussed: it appears that a common route to significantly increase carrier concentrations is to expand the physically accessible range of the dopant atomic chemical potential by non-equilibrium doping processes, which not only suppresses the formation of the intrinsic defects but also lowers the formation energy of the impurities, thereby significantly increasing their solubility.

Nanomechanics features three-dimensional nanostructuring, which allows full exploitation of the mechanical degree of freedom on the nanometre scale. In this work a number of exemplifying experiments on nano-electromechanical systems realized in silicon materials will be presented. First an introduction to the underlying mechanics will be given and finite element methods required for simulations will be discussed. Further topics presented include measurement methods for probing the mechanical properties of free standing nanowires, sensor applications and nonlinear properties of nanomechanical resonators. Other applications such as parametric frequency tuning are demonstrated and the major sources of dissipation are discussed. Finally, an outlook over the fundamental limits of nanoresonators is given.

A bond-order–bond-length–bond-strength (bond-OLS) correlation mechanism is presented for consistent insight into the origin of the shape-and-size dependence of a nanosolid, aiming to provide guidelines for designing nanomaterials with desired functions. It is proposed that the coordination number imperfection of an atom at a surface causes the remaining bonds of the lower-coordinated surface atom to relax spontaneously; as such, the bond energy rises (in absolute value). The bond energy rise contributes not only to the cohesive energy (E_Coh) of the surface atom but also to the energy density in the relaxed region. E_Coh relates to thermodynamic properties such as self-assembly, phase transition and thermal stability of a nanosolid. The binding energy density rise is responsible for the changes of the system Hamiltonian and related properties, such as the bandgap, core-level shift, phonon frequency and the dielectrics of a nanosolid of which the surface curvature and the portion of surface atoms vary with particle size. The bond-OLS premise, involving no assumptions or freely adjustable parameters, has led to consistency between predictions and experimental observations of a number of outstanding properties of nanosolids.

We report an atomistic investigation of static and dynamic properties of the (0001) surface of α-alumina at different temperatures. Lattice dynamics calculations are performed within the quasiharmonic approximation using a periodic slab geometry and with two parametrizations of the shell model. One of our aims is to prepare the ground for a more precise first-principles treatment, by checking the convergence of dynamical properties with respect to slab thickness, vacuum size and k-point sampling. The surface systematically undergoes a large relaxation, and we study the role of surface vibrational modes in the apparent position of the surface plane. An analysis of the mean square displacement of the atoms shows that the amplitudes of vibration of the atoms are 1.5 time larger at the surface than in the bulk. We find that these amplitudes converge slowly with slab thickness, and that the surface modes involve more than just the surface Al atoms. Another aim is to study the surface free energy, which we find has two favourable properties from the point of view of ease of computation. Firstly, we have calculated it for various coarse and fine samplings of phonon wavevectors in the Brillouin zone and find that it can already be well approximated with a sampling of just two k-points. Secondly, the vibrational contribution to the surface free energy converges much faster with respect to slab thickness than the mean square amplitude of vibration. The values of the surface free energy obtained with the two shell models differ by 20% but have similar temperature dependence and bracket the experimental value.

We consider the equilibrium statistical properties of interfaces submitted to competing interactions: a long-range repulsive Coulomb interaction inherent to the charged interface and a short-range, anisotropic, attractive one due to either elasticity or confinement. We focus on one-dimensional interfaces such as strings. Model systems considered for applications are mainly aggregates of solitons in polyacetylene and other charge-density wave systems, domain lines in uniaxial ferroelectrics and the stripe phase of oxides. At zero temperature, we find a shape instability which leads, via phase transitions, to tilted phases. Depending on the regime, elastic or confinement, the order of the zero-temperature transition changes. Thermal fluctuations lead to a pure Coulomb roughening of the string, in addition to the usual one, and to the presence of angular kinks. We suggest that such instabilities might explain the tilting of stripes in cuprate oxides. The three-dimensional problem of the charged wall is also analysed. The latter experiences instabilities towards various tilted phases separated by a tricritical point in the elastic regime. In the confinement regime, the increase of dimensionality favours either the melting of the wall into a Wigner crystal of its constituent charges or a strongly inclined wall which might have been observed in nickelate oxides.

The common-anion II–VI semiconductor superlattices (SLs) are characterized by a vanishing or a small valence-band offset (VBO). In the case of the lattice-mismatched SLs, the biaxial strain can drastically affect the splitting of the valence-band top states, and therefore be explored in designing type-I character SLs. In the present work, we used the sp³s^* tight-binding method, with the inclusion of strain and spin–orbit coupling effects, to investigate the electronic band structures of the strained CdTe/ZnTe(001) SLs versus the biaxial strain, layer thicknesses and VBO. Our results show that the electron is always confined within the CdTe slabs, whereas the hole behaviour controls the whole SL character. Our theoretical results are compared to the photoluminescence experiments and shown to be consistent with the strain morphology along the SL growth direction as well as the optical and structural qualities of the experimental samples.

The electronic structures of Pd and (Cu + Pd) overlayers on Ru(0001) have been studied. Photoelectron emission spectroscopy based on synchrotron radiation and radiation from a Mg Kα x-ray source has been used to study the valence band of the surfaces and the Ru 3d, Pd 3d and Cu 2p core levels. The interaction of Pd with Ru(0001) is studied in the coverage range of submonolayer to several monolayers. The core level spectra of the Pd 3d_5/2 level show a new component for Pd coverages above 1 ML. In the submonolayer range the binding energy (BE) of the Pd 3d_5/2 level is constant at 335.08 eV. A state characterized as an interface state between the Pd adlayer and the Ru substrate is observed at a BE of 1.22 eV. This state, which dominates the valence band for submonolayer Pd coverages, has not been reported earlier and it is seen neither for clean Ru(0001) nor for multilayers of Pd. It is a characteristic that the overlayer d bands are narrower in the low-coverage regime. Deposition of Cu, on a 0.4 ML Pd/Ru(0001) surface up to 1 ML total coverage of Pd and Cu, results in a mixed two-dimensional overlayer.

A systematic ferromagnetic resonance (FMR) study shows that an in-plane magnetic anisotropy in the patterned submicron rectangular permalloy elements is mainly determined by the element geometry. As the aspect ratio increases, the in-plane shape anisotropy increases, but the contribution from the non-uniform demagnetizing effect decreases. An expression is proposed to represent the in-plane shape anisotropy due to the non-uniform demagnetizing effect. When the magnetic field is applied near the film normal direction, a set of evenly spaced peaks appears on the low-field side of the main FMR peak. The multiple peaks are considered as spin-wave resonance spectra of non-uniformly magnetized elements.

This study explores perpendicular transport through macroscopically inhomogeneous three-dimensional disordered conductors using mesoscopic methods (the real-space Green function technique in a two-probe measuring geometry). The nanoscale samples (containing ∼ 1000 atoms) are modelled by a tight-binding Hamiltonian on a simple cubic lattice where disorder is introduced in the on-site potential energy. I compute the transport properties of: disordered metallic junctions formed by concatenating two homogeneous samples with different kinds of microscopic disorder, a single strongly disordered interface, and multilayers composed of such interfaces and homogeneous layers characterized by different strengths of the same type of microscopic disorder. This allows us to: contrast the resistor model (semiclassical) approach with a fully quantum description of dirty mesoscopic multilayers; study the transmission properties of dirty interfaces (where the Schep–Bauer distribution of transmission eigenvalues is confirmed for a single interface, as well as for a stack of such interfaces that is thinner than the localization length); and elucidate the effect of coupling to ideal leads ('measuring apparatus') on the conductance of both bulk conductors and dirty interfaces. When a multilayer contains a ballistic layer in between two interfaces, its disorder-averaged conductance oscillates as a function of the Fermi energy. I also address some fundamental issues in quantum transport theory—the relationship between the Kubo formula in the exact state representation and the 'mesoscopic Kubo formula' (which gives the exact zero-temperature conductance of a finite-size sample attached to two semi-infinite ideal leads) is thoroughly re-examined by comparing their outcomes for both the junctions and homogeneous samples.

Previous models of the surface structure of ferroelastic domain walls have neglected one potentially important consideration. This is the effect of a volume strain coupled to the order parameter. Such a coupling affects the structure of the domain wall both in the bulk and at the surface. In the bulk only certain components of the strain tensor can relax in response to the stresses generated within the domain wall by this coupling without disrupting the continuity of the lattice. However, at the surface a more general relaxation is possible. This has an asymmetric effect on the domain wall surface structure, causing it to widen at one surface and narrow at the opposite surface.

We report the percolation behaviour of the conductivity of titanium oxynitride films grown by low-pressure metal–organic chemical vapour deposition, composed of TiN_xO_y mixed with TiO₂. The usual DC parameters (t, s and Φ_c), obtained from the effective media theory equations, are compared to the universal values (s = s_un while t < t_un because of the film anisotropy). This is the first example of an electrical continuum percolation applied to columnar films with chemically similar conducting and insulating units (non-heterogeneous percolation) whose mixing is based upon the growth temperature during the film growth.

The catalytic reduction of NO with H₂ is of considerable interest due to the pollution effects of NO in air. Previously, computer simulation studies of this reaction system have been made on the basis of a Langmuir–Hinshelwood (thermal) mechanism. We have studied a model through Monte Carlo simulation, which assumes that the reaction can proceed via a non-thermal (precursor) mechanism. When the motion of the precursor is taken into the first nearest neighbourhoods and NO is always adsorbed in a dissociated form, the model predicts a very small steady reactive window, which is limited by two irreversible phase transitions. However, diffusion of N atoms widens the reactive window considerably. When NO is adsorbed as a molecule (leading to dissociated adsorption if a vacancy pair is available), the motion of the precursor in the first neighbourhood does not provide any reactive region. Remarkably, diffusion of N atoms cannot change the situation in this case. However, desorption of NO provides some interesting results.

We show that electron transport in disordered quantum wires can be described by a modified Cooperon equation, which coincides in form with the Dirac equation for the massive fermions in a (1 + 1)-dimensional system. In this new formalism, we calculate the direct electric current induced by electromagnetic (EM) fields in quasi-one-dimensional rings. This current changes sign, from diamagnetic to paramagnetic, depending on the amplitude and frequency of the time-dependent external EM field.

The decomposition theory of the U(1) gauge potential has been studied. A rigorous proof of the London assumption is given. By making use of this gauge potential decomposition and the ϕ-mapping topological current theory, the precise expression for × is obtained, and the topology in quantum mechanics is discussed.

X-ray diffraction (XRD) and differential scanning calorimetry were used to investigate the crystallization process of amorphous Al₉₀TM_xCe_10−x alloys. Ageing effects were also examined by means of XRD. The structure corresponding to the prepeak for amorphous Al₉₀Fe₅Ce₅ alloys is more stable than the amorphous matrix. However, amorphous Al₉₀Ni₅Ce₅ alloys are not stable during the first crystallization stage, and even decompose at room temperature. The crystallization onset temperature for amorphous Al–Fe–Ce alloys is much higher than that for amorphous Al–Ni–Ce alloys. This is probably caused by the different stability of the structure corresponding to the prepeak. The crystallization onset temperature increases as the Ce/Ni ratio increases in amorphous Al₉₀Ni_xCe_10−x alloys. However, the crystallization onset temperature decreases as the Ce/Fe ratio increases in amorphous Al₉₀Fe_xCe_10−x alloys. A better atomic packing results as the Ce content increases due to size mismatch in Al–Ni–Ce systems and as the Fe content increases due to the increasing size of structural units in Al–Fe–Ce systems.

A high-resolution synchrotron diffraction study of the temperature dependence of the structures of the two layered Bi oxides PbBi₂Ta₂O₉ (PBT) and PbBi₂Nb₂O₉ (PBN) is reported. The transition from the low-temperature orthorhombic ferroelectric structure to the high-temperature tetragonal paraelectric structure in PBN involves an intermediate orthorhombic paraelectric phase. We conclude that this has the same Amam structure as observed in SrBi₂Ta₂O₉ and its Bi-rich analogue. Identifying the sequence of phase transitions in PBT is more challenging since above 200 °C the structure is metrically tetragonal within the resolution of the synchrotron diffraction measurements. By contrast, neutron diffraction measurements suggest that PBT remains orthorhombic to above the Curie temperature ≈430°C. The possible identity of the intermediate space group is discussed.

Raman scattering and cathodoluminescence experiments have been performed to investigate the effect of dislocations on the spatial distribution of point defects and on the free electron concentration in n-type GaAs:Si. An experimentally extended increase of the free electron and (Si_GaV_Ga)²⁻ complex concentrations from the matrix to the dislocation is explained as resulting from the formation of arsenic precipitates around the dislocation by means of computer simulations based on a diffusion–aggregation model.

Recently reported temperature-dependent synchrotron x-ray thermal diffuse scattering data (Holt M, Zschack P, Hong H, Chou M Y and Chiang T C 2001 Phys. Rev. Lett.86 3799) are analysed within an anharmonic electron–phonon interaction model, where the observed zone boundary phonon softening is related to an incipient antiferroelectric instability. The parameter-free model reproduces the temperature dependence of the soft phonon mode in excellent agreement with the experimental data. Predictions for the charge-density-wave-induced gap are made and consequences for the electronic structure changes are discussed.

Electronic and structural properties of copper monoxide (CuO) sintered as a common ceramic and nanoceramic are studied by positron annihilation spectroscopy. A CuO nanoceramic with crystallite size ranging from 15 to 90 nm was prepared from a common one by shock-wave loading. It is found that the momentum distribution of valence electrons in CuO is shifted, as compared with metallic copper, towards higher momentum values. This result is related to the effect of the Cu 3d–O 2p hybridization in the Cu–O ionic covalent bond formation. It is found that open volumes, identified mainly as small agglomerates of oxygen vacancies, appear at the nanoceramic crystallite interfaces. The degree of the Cu–O bond covalency decreases locally at the crystallite interfaces because of an oxygen deficit. The nanocrystalline state in CuO is shown to be thermally stable up to 700 K.

We report on investigations of type I clathrate Si and Ge compounds with Ba partially substituted by rare earth atoms. Novel compounds from framework-deficient solid solutions Ba₈ Al_x Si_42−3/4x □_4−1/4x and Ba₈ Ga_x Si_42−3/4x □_4−1/4x (x = 8, 12, 16; □, open square... lattice defect) have been prepared and characterized. All x-ray intensity data are consistent with the standardized clathrate I-Ba₈Al₁₆Ge₃₀ type structure (space group Pmn). In rare earth substituted clathrates, Eu₂Ba₆M_xSi_46−x (M = Cu, Al, Ga), rare earth atoms completely occupy the 2a position and thus form a new quaternary ordered version of the Ba₈Al₁₆Ge₃₀ structure type. From a geometrical analysis of clathrate crystal structures, a systematic scheme for all known clathrate compounds is proposed. All clathrates studied are metals with low electrical conductivity. The highest Seebeck coefficient in the present series is deduced for Ba₈In₁₆Ge₃₀, S = −75μV K⁻¹, indicating transport processes dominated by electrons as carriers. The Eu-based clathrates investigated exhibit long-range magnetic order as high as 32 K for Eu₂Ba₆Al₈Si₃₆ of presumably ferromagnetic type. Magnetic susceptibility indicates in all cases a 2+ ground state for the Eu ions, in fine agreement with L_III absorption edge spectra.

Using improved Wigner–Brillouin perturbation theory we study resonant electron–phonon interaction in a semiconductor quantum dot. We predict pinning of the excited energy levels to the ground state level plus one optical phonon as a function of the strength of the confinement potential. This effect should be observable through optical spectroscopic measurements.

The in-plane electrical transport and optical properties of the incommensurate intergrowth compounds (SbS)_1.15(TiS₂)_n withn = 1, 2 have been investigated by means of measurements of the electrical resistivity, Hall coefficient and thermopower in the temperature range from 4.2 to 350 K, and by optical spectroscopy in the frequency range 1000–20 000 cm⁻¹ at room temperature. Both the Hall effect and the thermopower indicate transport by electrons. The Hall coefficients show electron donation of about 0.57 and 0.34 of an electron per Ti atom for (SbS)_1.15(TiS₂)_n withn = 1 and 2, respectively. The in-plane resistivity ρ_ab (T) exhibits a non-linear dependence on temperature, which can be described with the formula ρ_ab (T) = ρ₀ + A_ee (T/T_F)² ln (T_F /T) as for a two-dimensional Fermi liquid. Fits according to the Drude model to the room temperature optical reflectivity show that the relaxation rate has a quadratic variation with frequency 1/τ(ω) ∼ ω², also indicating Fermi-liquid behaviour with interelectronic collisions of quasiparticles. Only a small anisotropy in theab-plane is observed in the optical spectra for the electrical field polarized parallel and perpendicular to the incommensurate direction.

We studied the energy spectrum of a spin lattice formed by mXY spin-1/2 chains with Ising inter-chain coupling. For strong ferromagnetic coupling we found that the lowest-energy states for non-zero numbers of inverted spins on each XY chain have n-magnon bound character if n = ml − 1, where l = 2, 3,.... For a spin tube formed by three XY chains with strong ferromagnetic inter-chain coupling, the lowest-energy states are described by the XXZ spin-1/2 chain model with antiferromagnetic coupling.

With the help of the bond operator representation for three S = 1 2 spins, we study the effective Hamiltonian and the phase diagram of a generalized spin-1 2 distorted diamond chain. In the weak-intertriangle-coupling limit, the magnetism of the effective Hamiltonian is studied with second-order perturbation theory and mean-field decoupling. Various phases such as the spin-fluid phase, the dimerized phase and the ferrimagnetic phase are shown to compete. For larger intertriangle interactions, the spin-fluid phase and the dimerized phase can be also described by the effective spin–orbital model and the region of the dimerized phase enlarges around the symmetric point of J₁ = J₂ = J₃ (J₁, J₂ and J₃ are the intratriangle interactions). The magnetization plateaus at 1 3M_s and 2 3M_s in the magnetic field are also studied.

The resistivity, ρ, of ceramic La_1−xCa_xMn_1−yFe_yO₃ with x = 0.3 and y = 0.0–0.09 is found to obey, between a temperature T_v ≈ 310–330 K and the ferromagnetic-to-paramagnetic transition temperature, T_C = 259–119 K (decreasing with y), the Shklovskii–Efros-type variable-range hopping conductivity law, ρ(T) = ρ₀ (T) exp [(T₀ /T)^1/2 ]. This behaviour is governed by generation of a soft Coulomb gap Δ ≈ 0.42 eV in the density of localized states and a rigid gap δ(T) ≈ δ(T_v)(T/T_v)^1/2 with δ(T_v) ≈ 0.16, 0.13 and 0.12 eV at y = 0.03, 0.07 and 0.09, respectively. Deviations from the square root dependence of δ(T), decreasing when y is increased, are observed as T → T_C. The prefactor of the resistivity follows the law ρ₀ (T) ∼ T^m, where m changes from 9/2 at y = 0 to 5/2 in the investigated samples with y = 0.03, 0.07 and 0.09, which is connected to introduction of an additional fluctuating short-range potential by doping with Fe.

Magnetic measurements performed on the LaNi_5−xAl_x system show that the magnetic susceptibilities increase up to a temperature T_max, characteristic for each composition. In the temperature range T ≤ 10 K, a T²-dependence of the magnetic susceptibilities was evidenced for compositions x = 0 and 1. In the high-temperature range the magnetic susceptibilities were analysed as a superposition of a Pauli-paramagnetic term on a Curie–Weiss-type contribution. The effective nickel moments decrease with increasing Al content. X-ray photoelectron spectroscopy measurements and band structure calculations show that the Fermi level is gradually shifted to a region with lower density of states when Al is substituted for Ni. The magnetic behaviour of nickel was analysed in correlation with the sample composition.

The low-energy structures of the quantum ferrimagnetic Heisenberg chain consisting of (5/2, 1/2, 1/2) trimers are investigated theoretically. The results of the linear spin-wave theory are compared with those from the numerical exact diagonalization calculation and the matrix product method for the lowest optical mode. The temperature behaviour of thermodynamical properties such as magnetic susceptibility, specific heat and entropy are analysed in the framework of the modified spin-wave theory. The results of the calculations are used to explain the experimental data obtained for the molecule-based heterospin magnets [Mn(hfac)₂BNO_R ] (R = H, F, Cl, Br) with one-dimensional chain structure.

In the present paper we analyse the contributions of nitro group movements in 2-chloro-nitrobenzene to the nuclear quadrupole resonance (NQR) parameters of the chlorine nucleus in the molecule. We found two contributions to the spin–lattice relaxation time (T₁) and the NQR frequency (ν_Q) due to the onset of nitro group movements in the molecule. One of these contributions is the well-known semirotation of the nitro group around the N–C axis. The other one is attributed to some tilting or tipping of the nitro plane away from the benzene ring introducing some dynamic orientational disorder of this group in the crystal only observed as a contribution from the temperature dependence of T₁ and ν_Q. Its activation energy is similar to that of the nitro group reorientation (21.9 and 23.6 kJ mol⁻¹ for the two processes) and may arise from competing crystalline and steric chlorine nucleus effects. The present investigation shows that in chloronitrobenzenes the NO₂ group dynamic orientational disorder can produce modulation effects on the chlorine T₁ which are large enough to be observed by means of the NQR.

Recent divergence in analysing the magnetization processes in isolated particles between analytical micromagnetics and numerical micromagnetics has focused on whether it is necessary to use nucleation theory in the analysis. Complete saturation is the necessary condition for using nucleation theory. A ferromagnetic elliptical particle can be uniformly magnetized in a large field. As the field decreases, there exists a nucleation field at which the magnetization deviates from uniform magnetization. On the contrary, a ferromagnetic cube can never be saturatedly magnetized in any finite homogeneous field. It is difficult to apply the theory of a nucleation field of an elliptical particle to a cubic particle. One practical way to discuss the 'nucleation' in a cubic particle is to supervise the magnetization changes from a positive quasisaturation state to a negative quasisaturation state, and find what kind of reversal modes appear. In this paper, a three-dimensional micromagnetics model is implemented to analyse the magnetization reversal processes in cubic particles at a field (1.1 × 10⁶ Oe) where the quasisaturation is well developed in a cubic particle. The sizes of particles vary from 400–1000 Å. A fine mesh with 10 × 10 × 10 and a small decreasing step of applied field 10 Oe are used in the calculations. The 'nucleation' in a cubic particle starts from the quasiquantization state (flower state). For a particle whose size is smaller than 1000 Å, the equilibrium magnetization states during magnetization reversal processes are a flower state and an anti-flower state, and a coherent rotation happens when the magnetization state changes from a flower state to an anti-flower state. For a larger particle with a size of 1000 Å, there exist rather complicated equilibrium magnetization states i.e. a flower state, anticlockwise vortex state, intermediate state, clockwise vortex state, and anti-flower state all appear during the reversal processes.

Magnetic properties of samples of the system Ga_xFe_{1 −x}[NiCr]O₄ (x = 0.0, 0.2, 0.4, 0.6, 0.8) are investigated. The assumption is made that two magnetic phase transitions take place for the ferrites with x ≥ 0.2: at T_C the transition from the paramagnetic state to a spin-glass state; and at T_t a second transition from a spin-glass state to a frustrated magnetic phase. This assumption is in good agreement with theoretical models. It is established that long-range magnetic ordering occurs at the temperature T_t.

It is suggested that a frustrated magnetic structure in one sublattice is necessary for the occurrence of anomalous dependences of the spontaneous magnetization on temperature σ_s(T) of N, P or L type. It is established that if the frustrated magnetic structure is present in both sublattices of the ferrite, an anomalous σ_s(T) curve of another type is formed.

Our conclusion is that the reason for the frustration of the magnetic connections in the samples of the system Ga_xFe_{1 −x}[NiCr]O₄ with x ≤ 0.2 is the presence of strong indirect interlattice interaction Fe_A³⁺ – O²⁻ –Cr _B³⁺ of positive sign.

We have performed a Mössbauer spectral analysis of nanocrystalline metastable P 6/mmm SmTi(Fe_1−xCo_x)₉, correlated with structural transformation towards its equilibrium derivative I4/mmm SmTi(Fe_1−xCo_x)₁₁. The Rietveld analysis shows that the 3g site is fully occupied, while the 6ℓ occupation is limited to hexagons surrounding the Fe–Fe dumb-bell pairs 2e. A specific programme for the Wigner–Seitz cell (WSC) calculation of the metastable disordered structure was used. The hyperfine parameter assignment based on the isomer shift correlation with the WSC volumes sequence leads to Co 3g preferential occupation, with Ti location in 6ℓ sites. The mean hyperfine field increases with Co content in connection with the enhancement of the negative core electron polarization term upon additional Co electron filling. The same trend is observed for each individual site leading to the sequence H_HF {2e }≥H_HF {6ℓ}≥H_HF {3g}.

Electron paramagnetic resonance (EPR) of SeO₃⁻ radicals in K₃Na(SeO₄)₂ single crystal is reported. Two SeO₃⁻ radicals of the same orientation but different hyperfine parameters were used as paramagnetic probes to study the ferroelastic phase transition. A gradual change of orientation of the Se–O(1) bond in the bc-plane is observed below 329 K. At room temperature, the Se–O(1) bond is deflected by 3° from the c-axis in the bc-plane. Narrowing of EPR lines above 329 K is associated with an ordering process of the Se–O(1) bonds. The temperature hysteresis indicates the first-order phase transition at 329 K.

The changes in Madelung and non-electrostatic energies of Pb(Zr_xTi_1-x)O₃ (PZT) solid solutions in the high-symmetry (Pm3m) phase have been calculated using the heats of formation from the oxides. The non-electrostatic contribution (ΔE_N) decreases with decreasing x and becomes negative for compositions x ≤ 0.35, corresponding to perovskite tolerance factors t ≥ 1. Correlation of the strong increase in tetragonal distortion in the ferroelectric phase with more exothermic values of ΔE_N suggests a softening of short-range repulsions for Ti-rich compositions. The influence of complex solid solution behaviour on the character of Pm3m ↔ P 4mm transition is investigated by high-temperature specific heat and unit-cell parameter measurements. The implications of the thermochemical results with respect to mean-field theoretical descriptions of lattice instabilities and phase boundaries in the PZT system are briefly discussed.

A theory of free-carrier absorption is given for quasi-one-dimensional ternary semiconducting structures when the carriers are scattered by alloy disorder and the radiation field is polarized along the length of the wire. The free-carrier absorption coefficient is found to be an oscillatory function of the photon frequency and of the area of the cross-section of the wire. It is found that the absorption coefficient increases with decreasing transverse dimension of the quantum wire. The results obtained are compared with those from the quantum theory of free-carrier absorption in quasi-two-dimensional structures. In addition, it was found that in quantum wire the electron–alloy-disorder interaction gives a greater contribution to the absorption than the electron–acoustic phonon interaction.