Table of contents

We present a theory of the dynamics of monatomic liquids built on two basic ideas: (1) the potential surface of the liquid contains three classes of intersecting nearly harmonic valleys, one of which (the `random' class) dominates the potential surface and consists of valleys which all have the same depth and normal-mode spectrum; and (2) the motion of particles in the liquid can be decomposed into oscillations in a single many-body valley, and nearly instantaneous inter-valley transitions called transits. We review the thermodynamic data which led to the theory, and we discuss the results of molecular dynamics (MD) simulations of sodium and Lennard-Jones argon which support the theory in more detail. Then we apply the theory to problems in equilibrium and nonequilibrium statistical mechanics, and we compare the results to experimental data and MD simulations. We also discuss our work in comparison with the quenched normal-mode and instantaneous normal-mode research programmes and suggest directions for future research.

In this paper, the electrical resistivity of liquid Ga_1-xSb_x has been carefully measured as a function of temperature and concentration. For liquid Ga_1-xSb_x, the electrical resistivity versus temperature is linear for all concentrations for which measurements were made except x = 1 and 0.5, where the temperature coefficient of the resistivity increases with the increase of temperature near each melting point, but is independent of temperature above certain temperatures. It is very interesting that the electronic transport properties of liquid Ga-Sb are very different from those of liquid In-Sb, although they both belong to the liquid III-V system. The results obtained in this work have also been discussed on the basis of a `two-structure' model.

Dynamics of water contained in the pores of alumina gel as studied using a combination of the high and medium resolution quasi-elastic neutron scattering (QENS) technique at room temperature and extending to the supercooled region is reported. In the single particle picture of the dynamics of water molecules in confined geometry (Volino-Dianoux model), two types of water are found to be present in the pores of alumina gel. Some water molecules are attached to the surfaces (localized) and others undergo diffusion within the otherwise available space in the pores. The localization radius and diffusion constant (D_loc) characterizing the local dynamics and also the diffusion constant (D_t) and residence time (τ₀) of the water molecules diffusing inside the alumina gel pores are obtained at different temperatures. Water molecules are found to undergo restricted diffusion in the pores at higher temperatures, which approach the bulk-like behaviour in the supercooled region.

Low-temperature time-resolved photoluminescence (PL) experiments have been performed on a semiconductor planar microcavity, which contains two sets of three In_0.13Ga_0.87As/GaAs quantum wells embedded in a 3λ/2 GaAs cavity. The spontaneous emission dynamics of both lower- and upper-branch polaritons is investigated as a function of exciton–cavity detuning under nonresonant optical excitation. It is found that the PL decay times of both branches are independent of cavity detuning while the PL rising kinetics of the lower- and upper-branch polaritons exhibits a significant difference. The rise time of the upper polarition branch shows a strong dependence on cavity detuning, while the rise time of the lower polarition branch is less sensitive to cavity detuning. Our results can be well understood in the framework of the theoretical prediction of Tassone et al.

The structural transition of the interphase boundary between crystalline Al and amorphous aluminium oxide under electron beam irradiation is investigated. Local amorphization took place on the crystalline Al side near the interphase boundary, when the current beam density was higher than 75 A cm^-2 under electron beam irradiation. This was the result of mixing of oxygen into the crystalline Al, which resulted in formation of Al-O bonds. The amorphous Al₂O₃ irradiated by the electron beam was the main source of oxygen.

We report a detailed study of the magnetic and magnetotransport properties of Sr₂FeMoO₆ ceramics having a controlled concentration of antisite (AS) defects. It is found that a high-field differential susceptibility exists in all samples, which increases with AS. Similarly, a high-field magnetoresistivity develops and mimics the differential susceptibility. These observations suggest that antisite defects promote some magnetic frustration. High-resolution electron microscopy studies have allowed observation of the existence of antiphase boundary defects in the Sr₂FeMoO₆ structure.

Using the vector-wave multiple-scattering method we have calculated the photonic band structures of AB₃ and B₃ crystals composed of perfect metal spheres in air. It is shown that AB₃ photonic crystal contains a large photonic band gap and B₃ photonic crystal has more than one gap when the filling ratio of metal spheres exceeds a threshold. The multiple photonic gaps of the B₃ structure are useful in the design of optical multi-channel perfect mirrors and optical multi-band filters in the microwave regime.

⁷Li nuclear magnetic resonance and heat capacity measurements were performed on the metallic spinel LiV₂O₄ down to 30 mK. The temperature dependencies of the linewidth, the Knight shift and the spin-lattice relaxation rate were investigated in the temperature range 30 mK⩽T⩽280 K and at applied magnetic fields of 4.6, 10, 44 and 83 kOe. The longitudinal nuclear magnetization was found to relax following a stretched-exponential form with a stretching exponent in the range 0.5<β<1. For temperatures T<1 K and at the lowest applied magnetic field of 4.6 kOe, we observe a spin-lattice relaxation rate which slows down exponentially, exhibiting a hindering barrier of the order of 1 K. This excitation energy separates different configurations of a highly degenerate ground state of a completely frustrated magnet.

Electric and dielectric measurements have been used to study the superionic conductivity and the ferroelectricity respectively in Rb_0.8(NH₄)_0.2HSO₄ single crystal. This material presents a high level of conductivity at high temperature. Activation energies were determined from the temperature dependence of the alternating-current conductivity in the regions 440-560 K and beyond 560 K; they are respectively 0.15 and 0.10 eV. The dielectric measurements show an intense anomaly at 285 K indicating a ferroelectric-paraelectric phase transition. This phase transition is accompanied by a significant dispersion of the real part of the dielectric constant in the frequency region investigated, in accordance with the high level of conductivity of this material.

Polarized Cu K-edge spectra of CuO are calculated by different methods and compared to experiment. Theoretical spectra were obtained via an all-electron full-potential band-structure calculation based on a pseudopotential technique and by a real-space multiple-scattering method involving a self-consistent muffin-tin potential. Self-consistency in the scattering potential is necessary to describe the low-energy part of the ∥z partial spectral component. The low-energy shoulder in the ∥x component observed in the experiment cannot be explained within a one-electron picture and thus apparently has a many-body origin. However, its particular mechanism will be different from the shake-down process suggested for explaining a similar feature in copper-chlorine compounds.

We present results from in situ x-ray diffraction studies of hydrogenated icosahedral Ti-Zr-Ni quasicrystals carried out under high pressure. The icosahedral Ti₅₃Zr₂₇Ni₂₀ phase and hydrogenated icosahedral Ti₄₅Zr₃₈Ni₁₇ samples that contain 0.32 and 1.45 hydrogen atoms for each metal atom were studied. The icosahedral quasicrystal structure is retained up to the highest pressures investigated. The six-dimensional lattice parameter was obtained as a function of pressure for the different samples. The zero-pressure bulk modulus and its pressure derivative were determined by fitting a Murnaghan-type equation of state to the relative volume change V/V₀. The zero-pressure bulk modulus is respectively obtained as B₀ = 130±10, 105±10, and 110±20 GPa for H/M equal to 0, 0.32, and 1.45 using a constant value for the first derivative B'₀: 5.5±1. It seems, therefore, that the icosahedral Ti-Zr-Ni phase may be more compressible after hydrogenation, but there is no clear difference for different H/M contents. These results are tentatively related to the local structure of these materials.

Some numerical mistakes in figure 1 in [1] have been pointed out [2] and the corrected figure is shown below. Most of the conclusions in [1] remain the same. The only difference is as regards the shift of the critical Reynolds number Re_c for Kn not equal to 0 cases. Re_c becomes 5282, 2480 for Kn=0.001, 0.01, respectively. For the Kn=0.05 case, Re_c is reduced to 290.32. If the viscosity of the normal He II fluid can be evaluated for a certain temperature, then we can calculate the critical velocity in these cases [3].

Figure 1 Velocity-slip effects (Kn) on the neutral-stability boundary of the plane. Poiseuille flow for the normal fluid. Kn=λ/h. λ is the mean free path of the fluid and h is the channel halfwidth (cf. figure 1 of [1]).

Acknowledgment

The author is partially supported by the China Post-Doctoral Foundation.

References

[1] Chu W K-H 2000 J. Phys.: Condens. Matter12 8065

[2] Jie Q Y 2001 private communication

[3] Chu W K-H 2001 Preprint