Table of contents

The dewetting of thin polymer films and the resulting pattern formation are reviewed and previewed. Being of crucial importance for various applications, the destabilization of an initially homogeneous layer has been investigated intensively. Recently, the growing complexity of the systems under investigation introduced new questions. In this highly active research area, different complementary methods such as experiments, computer simulations and theory have yielded new insights. Particular emphasis is placed on experiments. Future perspectives are discussed in addition.

We outline the current state of experimental study and basic ideas for describing phase transitions in topologically disordered condensed matter, such as liquids and amorphous solids. Reviewing briefly the study of molten elementary substances under pressure, we pay primary attention to the results for liquid Se, S, and P and also to those substances that have not been represented in previous publications, mainly the liquid oxides B₂O₃ and GeO₂. The experimental data reveal the possibility of rather sharp transformations in relatively simple liquids that are smoothed at high temperatures. Comparing the transitions in amorphous solids and in liquids, one should emphasize the metastable and non-ergodic nature of amorphous substances and the existence of static local atomic stresses fluctuating in thermally frozen amorphous networks. In particular, the kinetic study of amorphous–amorphous transformations (AATs) in SiO₂ and GeO₂ glasses and amorphous H₂O ice under pressure highlights a number of anomalous features that distinguish the AATs from ordinary first-order transitions and from transformations in liquids. The recent in situ study of the volume changes in glassy silica a-SiO₂ upon compression at high temperatures provides a new conclusion as regards the existence of two pressure-induced AATs in a-SiO₂ with different microscopic mechanisms of structural rearrangements. We also perform the analysis of two possible kinetic scenarios for AATs, including sharp and diffuse transitions. The key relation determining the transformation scenario is the relationship between the radius of structural correlations in amorphous solid and the size of the critical nucleus of the growing disordered modification. The comparative analysis emphasizes the main difference between the transformations in liquids and amorphous solids that consists in the fact that the transitions in liquids are mainly determined by thermodynamic relationships, whereas the transitions in amorphous solids take place far away from equilibrium and are governed by the corresponding kinetics.

X-ray structural studies on several elemental liquids under high pressure are reviewed. Combination of synchrotron radiation sources and large-volume presses enables us to carry out in situ structural measurements on liquids at high pressures up to several gigapascals. The measurements have revealed that compressions of liquid alkali metals are almost uniform, whereas those of liquids that have covalent components in bonding are mostly anisotropic. For covalent liquids, the volume dependence of the nearest-neighbour distance deviates from (V/V₀)^1/3 behaviour (V being the molar volume and V₀ being the molar volume at zero pressure) and changes in coordination play important roles. In some elements, different types of volume dependence of the nearest-neighbour distances are observed in different pressure ranges. This behaviour suggests that the liquid phase can be divided into regions. Although most of the observed structural changes are continuous, a discovery of an abrupt structural change in liquid phosphorus, which is completed over a pressure range of less than 0.05 GPa around 1 GPa and 1050°C, supports the existence of a first-order liquid–liquid phase transition.

Supercooled liquids in the Y₂O₃–Al₂O₃ system undergo a liquid–liquid phase transition between a high-temperature, high-density amorphous polymorph (HDA form), and one with lower density that is stable at lower temperature (LDA form). The two amorphous polymorphs have the same chemical composition, but they differ in their density (∼4% density difference), and in their heat content (enthalpy) and entropy determined by calorimetry. Here we present new results of structural studies using neutron and high-energy x-ray diffraction to study the structural differences between high-density amorphous (HDA) and low-density amorphous (LDA) polyamorphs. The combined data sets show no large differences in the average nearest-neighbour Al–O or Y–O bond lengths or coordination numbers between the low-and high-density liquids. However, the data indicate that changes occur among the packing geometries and clustering of the Al–O and Y–O coordination polyhedra, i.e., within the second-nearest-neighbour shell defined by the metal–metal (i.e., Y–Y, Y–Al and Al–Al) interactions. Polarizable ion molecular dynamics simulations of Y₂O₃–Al₂O₃ liquids are used to help interpret the pair-correlation functions obtained from x-ray and neutron scattering data. Unexpectedly large density fluctuations are observed to occur during the simulation, that are interpreted as due to dynamic sampling of high-and low-density configurations within the single-phase liquid at temperatures above the critical point or phase transition line. Calculated partial radial distribution functions indicate that the primary differences between HDA and LDA configurations occur among the Y–Y correlations.

The metastable phase equilibria of substances in the disordered, liquid or amorphous, state are considered in T–P coordinates. Application of the thermodynamic model of pseudo-binary solutions is illustrated by means of calculated metastable T–P phase diagrams of water, carbon, and semiconducting compounds, GaSb and Ga₃₈Sb₃₈Ge₂₄. All calculated diagrams include the equilibrium lines between the disordered phases terminating at the critical points and the stability boundaries of these phases, the spinodals. Depending on the mutual disposition of the elements of the stable and metastable T–P diagrams, two additional lines of metastable equilibria between the crystalline high-pressure phase and each of the metastable disordered phases can appear. These lines terminate at the points of intersection with spinodals. These special points are denoted as pseudo-critical points. A combined analysis of the stable and metastable diagrams gives a reasonable interpretation of the experimental data on the phase transformations and on the nature of the anomalous properties of the substances in question as well as allowing prediction of some new effects.

Structures of (FeO)_x(P₂O₅)_1−x glasses with 0.2 ≤ x ≤ 0.5 are studied by x-ray diffraction using high energy photons from a synchrotron. Scattering intensities are obtained up to Q_max of 250 nm⁻¹. P–O, Fe–O and O–O first-neighbour peaks are well resolved in the pair distribution functions. Constant P–O coordination numbers of 4.0 ± 0.1 with distances of 0.155 ± 0.001 nm are found as expected for PO₄ tetrahedra. Mean Fe(II)–O coordination numbers of ∼5 with distances of ∼0.210 nm are extracted from fitting the Fe–O peaks where the oxygen coordination number of the minor fraction of 5–18% Fe(III) was set to six with Fe–O distances of 0.200 nm. The existence of FeO₅ or strongly distorted FeO₆ polyhedra instead of densely packed FeO₆ octahedra for the Fe(II) sites is attributed to the mixed oxide effect. Fe³⁺ cations in rigid FeO₆ octahedra hinder the Fe²⁺ cations in forming well defined octahedral environments.

Thermodynamic model calculations of the surface phase transitions in Ga–Bi alloys are performed, which account in particular for the finite thickness of the adsorption films. The excess chemical potential of bismuth in the films is calculated by employing a screened Coulomb potential. The calculated dependence of the liquid film thickness predicts adsorption and complete wetting transitions, which are in qualitative agreement with the experimental data. On the basis of surface tension calculations the surface freezing–melting transition is modelled and the results coincide with experimental observations via second harmonic generation and surface light scattering. Using an appropriate description of the nucleation mechanism in liquid films the line tension between the liquid and solid films is determined by a fit of experimental data. A complete diagram of the surface liquid–solid phase transitions on the vapour/liquid Ga–Bi interface is derived.

The dynamical behaviour of an entangled network, a melt made of long polymer chains, with an oscillating nanotip, is investigated to provide evidence for a close relationship between the spatial scale and dynamics of the network, from the molecular to the mesoscopic scale. The experimental results show how a dynamic force microscope is able to discriminate between slow and fast relaxation processes related to molecular motion at different scales. In particular, it is shown that within a very small variation of the tip indentation depth, a few ångströms, the relaxation times involved in the process of dissipation vary by more than three orders of magnitude.

The thermal radiation intensity from the mercury/sapphire interface shows an anomalous decrease in the prewetting supercritical region (Ohmasa et al 2000 J. Phys.: Condens. Matter12 A375). This anomaly is a clear indication of the optical diffuse scattering caused by the critical fluctuations in the mercury wetting film formed on the sapphire substrate. In the present work, we measured the thermal radiation spectrum in the wavelength range 500 ≲ λ < 1650  nm, and deduced the diffuse scattering intensity as a function of λ. In this measurement, a new collimator system was employed to eliminate the contribution from the 'backing materials' behind the mercury/sapphire interface. Assuming an Ornstein–Zernike type permittivity correlation function, we have estimated the lateral correlation length and the interface width of the critical fluctuations to be several hundred nanometres and a few nanometres, respectively.

As a result of a pseudosymmetry search, the double perovskite compounds Sr₂MWO₆ (M = Ni, Zn, Co, Cu) were identified as good candidates for presenting Landau-type phase transitions at high temperatures. In the present work, the phases observed at different temperatures are reported and the phase transitions are discussed. The Sr₂MWO₆ (M = Ni, Zn, Co) compounds present I4/m–Fmm phase transitions, while Sr₂CuWO₆ shows an I4/m–I4/mmm phase transition. All these transitions are second order and are consistent with the pseudosymmetry concept. Sr₂ZnWO₆ and Sr₂CoWO₆ also show P2₁/n–I4/m phase transitions and Sr₂CuWO₆ presents a phase transition of the type I4/mmm–Fmm. These transitions are first order. Alongside the experimental results a discussion of some aspects of the study of phase transitions in double perovskite materials is presented.

Near-band-edge photoluminescence properties of as-grown undoped CuGaSe₂ single crystals have been investigated in the range between 10 K and room temperature. The temperature dependence of the exciton gap energy E_g was studied by means of a three-parameter thermodynamic model, the Einstein model, Varshni's model and the Pässler model. The values of the band gap energy at T = 0 K, the effective phonon energy and the Einstein temperature, the cut-off phonon energy of the phonon spectrum as well as a dimensionless constant related to the electron–phonon coupling were estimated. Comparing all applied models points out that a minimum set of four parameters is required for a complete analytical description of the experimentally determined temperature dependence of the exciton gap in CuGaSe₂.

The magnetic and electronic structure of strontium hexaferrite SrFe₁₂O₁₉ has been investigated using density functional theory within the local spin density approximation. The calculations show that in this ferrite all the Fe³⁺ ions are fully spin polarized with S = 5/2. The electronic structure of this ferrite is strongly influenced by the exchange interactions between the iron ions. The most stable form of the ferrite is a ferrimagnet with the Fe³⁺ ions at the 4f sites having their spin polarization anti-parallel to the rest of the Fe³⁺ ions, in agreement with Gorter's prediction. The ferrite SrFe₁₂O₁₉ is calculated to be a semiconductor with both the top of the valence band and the bottom of the conduction band being dominated by Fe 3d states. A strong anisotropy was found for the conductive charge carriers. The calculated results are in good agreement with the available experimental data.

The path integral formalism is applied to derive the full partition function of a generalized Su–Schrieffer–Heeger Hamiltonian describing particle motion in a bath of oscillators. The electronic correlations are computed versus temperature for some choices of oscillator energies. We study the perturbing effect of a time-averaged particle path on the phonon subsystem, deriving the relevant temperature-dependent cumulant corrections to the harmonic partition function and free energy. The method has been applied to compute the total heat capacity up to room temperature: a low temperature upturn in the heat capacity over temperature ratio points to a glassy-like behaviour ascribable to a time-dependent electronic hopping with variable range in the linear chain.

The levels of the 4fⁿ⁻¹5d configuration of Ce³⁺(n = 1), Pr³⁺(n = 2), and Tb³⁺(n = 8) in compounds are compared with each other. A model is presented that, by means of energy shift operations performed on the five 5d levels of Ce³⁺, reproduces the energies of those for Pr³⁺ and Tb³⁺. Using Tb³⁺ data, two main sources of deviations from the shift model are identified. One is related to the size difference between Tb³⁺ and Ce³⁺ which affects the lattice relaxation and the crystal field splitting of the 5d configuration. The other is related to the isotropic exchange interaction between the 5d electron spin and the total spin of the 4f⁷ electrons in Tb³⁺. The exchange splitting is about 1 eV in fluorides, sulfates, and phosphates. In oxides with less strongly bonded oxygen 2p electrons, the exchange splitting decreases to 0.6 eV. The effects of the two deviations on the predictability of the 4fⁿ⁻¹5d energy levels of Tb³⁺ and other lanthanides are discussed.

DyIn₃ orders at T_N = 20 K and undergoes a second spontaneous magnetic transition at 16.5 K. From bulk magnetization measurements, performed on a single crystal along the three main axes of the cubic AuCu₃-type structure, the magnetic phase diagrams have been established. The crystalline electric field (CEF) scheme, in the paramagnetic phase, and the magnetic structures of the spontaneous and low field-induced phases have been probed by neutron techniques. All the magnetic phases studied are found to be multiple q with q belonging to the ⟨1/2, 1/2, 0⟩ star. In the low temperature phase (T < 16.5 K) the structure is double q with moments along twofold axes, whereas above 16.5 K it becomes triple q with moments along threefold axes. The analysis of the experimental results within the periodic molecular field model leads to a coherent interpretation of the spontaneous magnetic transitions, mainly driven by bilinear exchange and CEF interactions. Though the existence of quadrupolar interactions is definitively proved by the stabilization of multiple q magnetic structures, quadrupolar coefficients are found to be one order of magnitude smaller than those previously reported for NdIn₃ and TbIn₃.

We study the interplay between antiferromagnetism and superconductivity in a generalized infinite-U Anderson lattice, where both superconductivity and antiferromagnetic order are introduced phenomenologically in mean field theory. In a certain regime, a quantum phase transition is found which is characterized by an abrupt expulsion of magnetic order by d-wave superconductivity, as externally applied pressure increases. This transition takes place when the d-wave superconducting critical temperature, T_c, intercepts the magnetic critical temperature, T_m, under increasing pressure. Calculations of the quasiparticle bands and density of states in the ordered phases are presented. We calculate the optical conductivity σ(ω) in the clean limit. It is shown that, when the temperature drops below T_m, a double peak structure develops in σ(ω).

Effects of quantum fluctuation and four-spin interaction on the polarization and pyroelectric coefficient of temperature graded ferroelectric film are investigated, based on the transverse Ising model by taking into account the four-spin interaction. It is found that the magnitude and the sign of the polarization gradient are dependent on the imposed temperature gradient, and both the quantum fluctuation and four-spin interaction play an important role in the thermodynamic properties of the temperature graded film. With increasing strength of the quantum fluctuations the sharp peak of the pyroelectric coefficient shifts to lower temperatures. Also the four-spin interaction increases the mean polarization and makes the smooth pyroelectric peak at low temperature more pronounced. Furthermore, the graded ferroelectric film shows the characteristic feature of a first-order phase transition when the four-spin interaction is larger than a critical value, which is dependent on the temperature gradient and the quantum fluctuation.