We simulate several states of liquid cesium and rubidium with first-principles molecular dynamics. The Hellmann-Feynman atomic forces are obtained from a quantum-mechanical calculation of its electronic structure based on the local-density approximation of the density-functional formalism and using the pseudopotential plane-wave method. We compare our results with experimental data and other theoretical predictions for the structure and dynamics of liquid cesium and rubidium. We find a good agreement between our first-principles results, other theoretical predictions, and experiment.

Incoherent exciton dynamics in one-dimensional perfect lattices with traps at sites arranged according to aperiodic deterministic sequences is studied. We focus our attention on Thue-Morse and Fibonacci systems as canonical examples of self-similar aperiodic systems. Solving numerically the corresponding master equation we evaluate the survival probability and the mean-square displacement of an exciton initially created at a single site. Results are compared to systems of the same size with the same concentration of traps randomly as well as periodically distributed over the whole lattice. Excitons progressively extend over the lattice on increasing time and, in this sense, they act as a probe of the particular arrangements of traps in each system considered. The analysis of the characteristic features of their time decay indicates that exciton dynamics in self-similar aperiodic arrangements of traps is quite close to that observed in periodic ones, but differs significantly from that corresponding to random lattices. We also report on characteristic features of exciton motion suggesting that Fibonacci and Thue-Morse orderings might be clearly observed by appropriate experimental measurements. In the conclusions we comment on the implications of our work on the way towards a unified theory of the ordering of matter.

We calculate the electrical resistivity (ER) ρ of a fractal network from the view of the scattering of extended electronic states with both phonons and fractons and obtain different dependences of ER on the fractal dimensionality ${\mathit{d}}_{\mathit{f}}$, temperature T, fracton dimensionality d${\mathrm{̃}}_{\mathit{f}}$, and characteristic length ${\mathit{l}}_{\mathit{c}}$, for different-order interactions and different Euclidean dimensionalities d. As to the first interaction, ρ is proportional to T at the high-T limit, as known, and ρ∼aT+${\mathit{bT}}_{\mathit{f}}^{3\mathit{d}\mathrm{̃}}$/${\mathit{d}}_{\mathit{f}}$+d${\mathrm{̃}}_{\mathit{f}}$-1 at some low-T ranges. In the second-order case, ρ is a constant at the high-T limit, which is consistent with some recent experiments. In particular, we find that before a special fractal dimensionality ${\mathit{d}}_{\mathit{f}}^0$, there exists a minimum in the ρ-T curve, while after it ρ is a monotonically increasing function of T. The form of the ρ-${\mathit{d}}_{\mathit{f}}$ curve also shows different characteristics when d changes from 2 to 3. Finally, we discuss the percolating network and obtain scalar laws and scalar exponents.

Measurements of the spin-lattice relaxation time, ${\mathit{T}}_1$, of F centers in mixed alkali halide crystals, ${\mathrm{KBr}}_{\mathit{x}}$${\mathrm{Cl}}_{1\mathrm{-}\mathit{x}}$, have been made using magneto-optical techniques: magnetic circular dichroism (MCD) and optical pumping. These techniques consist of monitoring the recovery of the MCD signal to equilibrium, after an intense pulse of circularly polarized light. They were used to obtain the dependence of ${\mathit{T}}_1$ on the magnetic field up to 6.5 T and for several relative halide concentrations x at pumped liquid-He temperature (2 K). A computer-assisted system designed to measure relaxation pulses of the order of 1 ms is described. A model based on phonon modulation of the hyperfine interaction was modified to introduce the relative concentration x. This proves to be the main mechanism for spin-lattice relaxation for the temperature and magnetic-field range considered here, as described for unmixed alkali halides.

We study the electron dynamics in one-dimensional periodic and aperiodic (Thue-Morse) crystals. For the periodic chain the mean-square displacement ${\mathrm{\sigma }}^2$(t) and the integrated autocorrelation function C(t) of a quantum wave packet put on an initial site display unlimited ballistic motion defined by the asymptotic ${\mathrm{\sigma }}^2$(t)∼${\mathit{t}}^2$ and the nonexponential C(t)∼${\mathit{t}}^{\mathrm{-}1}$ laws. In the spatial distribution of the wave packet we find unexpected ‘‘phase chaos’’ features. For the aperiodic Thue-Morse sequence our numerical results show super-diffusion [${\mathrm{\sigma }}^2$(t)∼${\mathit{t}}^{3/2}$] and interesting self-similar oscillations for the ‘‘return to the origin’’ probability. The transmission coefficient of plane waves scattered from the aperiodic Thue-Morse sequence is shown to display self-similar band structure with completely transmissive modes which are related to the corresponding dynamics.

In earlier work unusual excited exciton-phonon states have been found for prototype exciton-phonon Hamiltonians (dimer, trimer). These states have been calculated numerically by diagonalizing the respective Fulton-Gouterman equations. Their physical character is such that the motion of the effective exciton becomes free whereas in the vibrational subspace the dynamics is governed by an effective stiffening of the nearest-neighbor interaction. In the present work it is shown analytically by means of a Fröhlich-type transformation, how the stiffening in the phonon dynamics is generated, such that the phononic part for these nonconventional exciton-phonon states is well described by squeezed oscillatory functional forms.

Of all the first series transition metals, nickel in diamond gives strong optical features characteristic of the charge state of the defect. A multiplet structure for each of these charged states is deduced from results of cluster model calculations and shown to be consistent with experiment. The various defects found to date fall into a pattern centered on the neutrally charged form of the defect. This neutral form has completely filled energy levels and is therefore weakly bonded to the lattice. This could explain why only charged forms of the defect are apparently observed. It is also suggested that while interstitial nickel will have a transition-metal character, a substitutional nickel center will display characteristics not unlike a positively charged vacancy.

Single and multiple-pulse impulsive stimulated Raman-scattering (ISRS) experiments are used to investigate the anomalously high damping rates of the lowest frequency ${\mathit{A}}_1$ polariton in ${\mathrm{LiTaO}}_3$. The wave-vector dependence of the response reveals coupling of the polariton to a weakly Raman-active, low-frequency relaxational mode as well as to two underdamped modes of a different symmetry. Inclusion of the coupled modes yields quantitative agreement between the results and the predictions of a previously developed theory for light scattering from polaritons.

An approach for nuclear quadrupole spin relaxation by anharmonic lattice vibrations of quadrupole nuclei situated at molecular or ion groups is presented. The temperature dependence at constant crystalline configuration is analyzed. This approach takes advantage of the existence of tightly bound molecules or molecular groups. Fluctuations causing relaxation are considered as collective vibrations of the whole lattice. Anharmonicity is explicitly included by involving cubic and quartic terms of the lattice vibrations Hamiltonian. The spectral densities of the spin-lattice Hamiltonian are expressed in terms of the phonon spectral densities and the temperature Green-function perturbation formalism is used for calculating the lower-order contributions. Three kinds of processes are found to contribute mainly to the spin-lattice relaxation time (${\mathit{T}}_1$), namely, anharmonic Raman (AR), first-order Raman (1R), and a third mechanism originated in the self-energy effects on the two-particle spectral density. The AR and 1R processes yield a quadratic temperature dependence of ${\mathit{T}}_1$ while the third mechanism gives a cubic temperature dependence. This last term, being characteristic of molecular crystals, represents a sort of mixing of the AR and 1R mechanisms, and its influence strongly depends on the crystal symmetry.

We report results of resistance measurements on Cu-Cr at several concentrations. We investigate the temperature ${\mathit{T}}_{\mathit{m}}$ of the resistance maximum of this spin glass as a function of sample size and applied magnetic field. From a fit of ${\mathit{T}}_{\mathit{m}}$ vs applied field, we obtain parameters that allow us to compare samples with different widths and thicknesses. We infer that the interimpurity interaction decreases with reduced sample size, and estimate the length scale for modifications of this interaction. We examine the magnetoresistance to gain further information about the effect of size. We also discuss the effect of concentration. The results are discussed within the context of previous observations of the reduction of the Kondo resistance contribution due to finite sample size. Our measurements span the regime where a resistance maximum is observed through the limit where the Kondo slope is the dominant contribution of the magnetic impurities to the resistivity. By adjusting the size of the samples to probe these regimes, we are able to illustrate the interplay between the spin glass and the Kondo resistivity contributions, deriving size effect information about both phenomena.

Low-temperature zero-field (1.4–80 K) and magnetic-field (1.23–7.53 T, 1.4–20 K) heat capacities, magnetization, magnetoresistance (2–25 K), and magnetic susceptibility (1.5–275 K) measurements have been carried out on a polycrystalline sample of ${\mathrm{CePtGe}}_2$. The compound orders ferrimagnetically at ${\mathit{T}}_0$=3.7–3.8 K; all of the studied properties show sharp anomalies around this temperature. Below ${\mathit{T}}_0$ a metamagnetic spin-flip (or -flop) phase transition occurs even at extremely low applied fields (about 0.2 T). A ferromagneticlike downturn in the χ(T${)}^{\mathrm{-}1}$ data was not observed in the low-field M(H) data. This led us to examine the possibility of crystal-field effects on the two crystallographically inequivalent cerium ions giving different sublattice magnetizations which are then coupled with a negative exchange coupling constant. Solving the crystal-field Hamiltonian for each cerium and using mean-field theory to couple the two cerium ions we obtain a good fit to the ${\mathrm{\chi }}^{\mathrm{-}1}$ data in the paramagnetic regime. An analysis of the magnetic dependence of the heat capacity at high temperature (10<T<20 K) indicates that the electronic specific-heat constant, γ, is >257 mJ/mol ${\mathrm{K}}^2$, which is more than twice as large as that found at low temperatures (1.4<T<3 K) in the zero-field heat-capacity data for the ferrimagnetically ordered state, i.e., γ=93(1) mJ/mol ${\mathrm{K}}^2$.

An external magnetic field causes a significant (more than twofold) increase of γ deduced from high-temperature-range data (i.e., the paramagnetic state) showing no saturation even at 7.5 T, and a slight rise of the electronic specific-heat coefficient for the magnetically ordered state up to ∼125 mJ/mol ${\mathrm{K}}^2$ with saturation at this value in a much lower field (2.5 T). The entropy (ΔS) associated with the ferrimagnetic phase transition is estimated to be 2.9 J/mol K, which is only 0.35R or 50% of the theoretically expected R ln2 value. The heat-capacity and resistivity measurements suggest that ${\mathrm{CePtGe}}_2$ is a moderately heavy-fermion system.

Ordered phase and first-order transition are investigated for the infinite-range Potts-glass model. A convergence condition of the free energy for the Thouless, Anderson, and Palmer (TAP) mean-field theory is obtained. Solutions for the mean-field equations in the TAP approach are obtained numerically for p=5 where p is the number of the Potts states. A class of solutions with partially restored permutation symmetry for the Potts index is found. Interpretation of these solutions in the replica approach is given. The first-order transition does not occur by crossing of the paramagnetic and the glass free energies. The free energy of the glass phase is always larger than that of the paramagnetic phase as has been suggested by the replica approach. Confluence of solutions with a rise in temperature for the glass phase seldom occurs in contrast with the Ising case.

We study a harmonic triangular lattice, which relaxes in the presence of an incommensurate short-wavelength potential. Monte Carlo simulations reveal that the elastic lattice exhibits only short-ranged translational correlations, despite the absence of defects in either lattice. Extended orientational order, however, persists in the presence of the background. Translational correlation lengths exhibit approximate power-law dependence upon cooling rate and background strength. Our results may be relevant to Wigner crystals, atomic monolayers on crystals surfaces, and flux-line and magnetic bubble lattices.

In the static replica symmetric approximation, the quantum Heisenberg spin-glass model with the infinite-ranged random Dzyaloshinskii-Moriya interaction in external magnetic fields is investigated in detail. The local susceptibility and the corresponding order parameters are calculated, which are in good agreement with those of thermofield calculations. The dependencies of entropy and specific heat on temperature are studied numerically at various magnetic fields and for different spin numbers. It is shown that the entropy is always positive and the quantum fluctuation has very strong effect on specific heat. Our results show that the typical specific heat of a spin glass for different magnetic fields exhibits crossover behavior, which has been obtained experimentally but not explained theoretically before.

The singular behavior of the magnetic susceptibility as a function of field just above the coexistence region of a ferromagnet is studied taking into acount (in contrast to previous treatments) the demagnetizing effect that spreads the coexistence line into a band. A detailed calculation is presented for the longitudinal susceptibility of a thin, normally magnetized film. It is found that the results of the earlier calculations cease to be valid for sufficiently thin films in a given field, and for films of given thickness they fail at fields below a certain value. In particular, the apparent singularity disappears altogether for sufficiently small fields, i.e., the longitudinal susceptibility reaches a limiting value as the field tends to zero.

Ferromagnetic properties of the molecular alloy complexes TDAE-(${\mathrm{C}}_{60}$${)}_{1\mathrm{-}\mathit{x}}$(${\mathrm{C}}_{70}$${)}_{\mathit{x}}$ (x=0.1–0.9), where TDAE is tetrakis(dimethylamino)ethylene, are studied by means of ESR and static magnetic susceptibility and magnetization experiments. Comparison of the magnetic properties of the present alloy complexes with those of TDAE-${\mathrm{C}}_{60}$ and -${\mathrm{C}}_{70}$ is made. The ESR g values and linewidths of these complexes remarkably increase below the Curie temperature ${\mathit{T}}_{\mathit{c}}$(x) at x<0.5. Moreover, these complexes with x=0.2, 0.4, and 0.5 show the S-shaped magnetization curves indicative of ferromagnetic correlation of spins below the ${\mathit{T}}_{\mathit{c}}$(x). The ${\mathit{T}}_{\mathit{c}}$(x) is suppressed with increasing x in such a way that ${\mathit{T}}_{\mathit{c}}$(0)=17.5 K and ${\mathit{T}}_{\mathit{c}}$(0.5)=6.1 K. This suppression is not so strong as one may expect for a quasi-one-dimensional itinerant ferromagnet. The three-dimensional spin correlations between ${\mathrm{C}}_{60}$ molecules both along the c axis and in the ab plane are important for this interesting magnetism.

Dynamical behavior of vortices is investigated for a two-dimensional classical Heisenberg antiferromagnet with easy-plane anisotropy. Equations of motion for the vortices are derived with the use of the Lagrangian formulation. The effect of a magnetic field applied in the XY plane is considered. The implications of the theory developed here to the calculations of dynamical correlation functions are presented. However, since pair interactions are neglected, these calculations are relevant only for temperatures below the Kosterlitz-Thouless transition temperature.

We report the results of temperature-dependent conductivity ${\mathrm{\sigma }}_{\mathrm{dc}}$(T), thermoelectric power S(T), microwave dielectric constant ${\mathrm{\epsilon }}_{\mathrm{mw}}$(T) (f=6.5 GHz) measurements, and x-ray diffraction studies for the camphor-sulfonic-acid-doped poly(o-toluidine) fibers prepared in m-cresol (POT-CSA). Compared to previously studied hydrochloric-acid-doped poly(o-toluidine) samples, POT-CSA fibers show significantly higher ${\mathrm{\sigma }}_{\mathrm{dc}}$ with a much weaker T dependence, much larger ${\mathrm{\epsilon }}_{\mathrm{mw}}$, and dramatically different temperature dependence of thermoelectric power with S(T)=A+BT, where A<0 and B>0. These results demonstrate the formation of the delocalized electronic state in the poly(o-toluidine) system, despite the asymmetry of the repeat unit.

Photoemission data in which the signals from the first two atomic layers are well distinguished from that of the bulk are used to determine the kinetic-energy dependence of the inelastic electron mean free path in Be metal. Using a layer attenuation model, the electron mean free path is extracted from experimental core-level data recorded from the Be(0001) and Be(101¯0) surfaces. The values determined for the two surfaces show a good overall agreement and the mean free path is found to exhibit a pronounced minimum at an electron kinetic energy of around 25 eV. The experimental results are compared with the mean free path calculated using the approach originally applied by Quinn. The experimental and calculated values are found to agree fairly well at higher energies. At lower energies the experimental values drop significantly below the theoretical values and locate the minimum at a markedly smaller kinetic energy. Possible reasons for this discrepancy are discussed.

The transport properties of excess electrons have been measured by the time-of-flight (TOF) method in high-density gaseous and liquid argon. In addition to the electric drift field a magnetic field perpendicular to it was applied. The drift mobility ${\mathrm{\mu }}_{\mathit{D}}$ and the Hall angle ${\mathrm{\theta }}_{\mathit{H}}$ were measured at electric-field strengths 5<E<200 V/cm and magnetic-field strengths 0<B<5 T. The density of the argon was varied in the range 5.5≤N≤125×${10}^{20}$ ${\mathrm{cm}}^{\mathrm{-}3}$ in the gas phase and 93≤N≤173×${10}^{20}$ ${\mathrm{cm}}^{\mathrm{-}3}$ in the liquid phase. The experimental results in the gas phase are compared with model calculations. For N<70×${10}^{20}$ ${\mathrm{cm}}^{\mathrm{-}3}$ good agreement between measured and calculated values is found, while discrepancies are observed at higher densities. The possibility to adapt similar models based on a kinetic approach to the liquid are discussed. The difference of about 30% between the mobilities measured in the gas and liquid phase at the same density are qualitatively explained by the difference in the static structure function.