High- and low-density activity series are given for a two-dimensional lattice-gas model in which each anisotropic molecule can take up one of two orientations. The model is ‘‘icelike’’ in that low-energy particle configurations have an open structure, while the close-packed structure has a higher energy. At high temperature there is a second-order transition to a zigzag type of structure at high density, while at low temperatures there is also a phase of intermediate density reflecting an open structure of parallel linear stacks of molecules. At low temperature the open structure must be broken to obtain the close-packed system. For comparison, an analogous one-dimensional model is solved exactly.

We report a computer-simulation study of the phase diagram of a system of hard spherocylinders with aspect ratio L/D between 0 and 5. For L/D≤3 only isotropic liquid and crystalline solid phases can occur. For L/D=3 a mechanically stable smectic phase occurs, but it is thermodynamically unstable. For larger L/D values, stable smectic and nematic phases are present.

Correlation spectroscopy measurements performed on molten and supercooled antimony trichloride with the homodyne technique show correlation functions that have a nonexponential behavior. Two well-defined distributions of correlation times can be observed in different temporal regions. This behavior is discussed in terms of a structural relaxation of clusters dynamically formed by intermolecular and interchain bonds. The Arrhenius plot of these correlation times shows a linear behavior with the same activation energy for both. In contrast, the activation energy of shear viscosity has a different value, showing that the processes determining the temperature behavior of τ and ${\mathrm{\eta }}_{\mathit{s}}$ are different.

The light-induced frequency locking of a nematic liquid crystal in a Fabry-Pérot interferometer with a periodic optical drive has been studied. The frequency-locking phenomena and the quasiperiodic oscillation of the output intensity of the system were observed experimentally. The oscillations over wide-band power spectra were observed between the frequency-locking regions for larger drive amplitudes. A numerical simulation of the system reveals Arnol’d tongues in the parameter plane and Farey ordering for the frequency locking. Chaotic behavior and two routes to chaos are obtained. The effects of noise are analyzed.

The relationship between the one-dimensional kinetic Ising model at zero temperature and diffusion annihilation in one dimension is studied. Explicit asymptotic results for the average domain size, average magnetization squared, and pair-correlation function are derived for the Ising model for arbitrary initial magnetization. These results are compared with known results for diffusion annihilation, and it is shown that there is only partial equivalence between the Ising model and diffusion annihilation. The results of Monte Carlo simulations for the domain-size distribution function for different initial magnetizations are also presented. In contrast to the case of diffusion annihilation, the domain-size distribution scaling function h(x) is found to depend nontrivially on the initial magnetization. The exponent τ characterizing the small-x behavior of h(x) is determined exactly and is shown rigorously to be the same for both the Ising model and diffusion annihilation.

A combined effect of sticking probability and finite viscosity ratio is studied on the pattern formation in Laplacian growth. A renormalization-group theory is developed to study the crossover phenomena between the diffusion-limited aggregation (DLA) and nonfractal structure. A two-stage crossover phenomenon is analyzed by using a three-parameter position-space renormalization-group method. A global flow diagram in three-parameter space is obtained. It is found that there are three nontrivial fixed points, the first Eden point, the DLA point and the second Eden point. The second Eden point corresponding to the dense structure is stable in all directions, while the first Eden point and the DLA point are saddle points. When the sticking probability P is small and the viscosity ratio is finite, the aggregate must cross over from the dense structure, through the DLA fractal, finally to the dense aggregate.

Planar gold layers of thickness 50–200 nm, both free-standing and supported on substrates of 1-μm polypropylene, were irradiated with 3-ns laser pulses of 0.53 μm wavelength at ${10}^{12}$ and ${10}^{13}$ W/${\mathrm{cm}}^2$. The x-ray emission from the rear and front of the target was measured with symmetrically placed x-ray diodes and x-ray transmission grating film spectrometers. The transmitted laser light was also monitored. Under optimized conditions it is possible to obtain up to 30% conversion of the laser energy in the form of soft x rays from the rear surface of the target with less than 2% laser light transmission. The measured overall conversion efficiency is approximately in agreement with hydrodynamic simulations. The spectral and temporal features of the emission are discussed with respect to the opacity of the hot dense target material.

An impurity placed in a plasma modifies the particle-particle correlations in the plasma. This impurity-plasma-plasma correction is relevant to ion-microfield calculations in hot dense plasmas as well as to models of the fractional quantum Hall excitations in terms of impurity-plasma systems. We show how to calculate such impurity-plasma-plasma corrections and apply them to a calculation of the Baranger-Mozer (BM) second-order microfield at He, Li, Be, and B impurities in a hydrogen plasma. Such a calculation for a dense plasma requires a definition of the electric field at the impurity due to an individual plasma ion in the plasma. We show how the traditional uniform jellium background assumption can be transcended via a deconvolution of the electron density obtained from a density-functional calculation for the plasma that incorporates the fully self-consistent nonlinear screening effects. Finally we carry out all-order resummations of the BM series beyond second order using the weighted-chain-sum method and also two models of the adjustable parameter exponential approximation.

Experiments were conducted on planar slab targets of carbon, aluminum, and copper using a 20-J 5-nsec, s- and p-polarized Nd:glass laser for the studies of saturation phenomena associated with two-plasmon-decay instability by monitoring the spectral profiles of 3/2${\mathrm{\omega }}_0$ emissions at different laser intensities. It was observed that the emissions from carbon plasma showed the existence of wavelength-dependent saturation domain of the instability at and beyond a laser intensity of ${10}^{14}$ W/${\mathrm{cm}}^2$, but the emission from copper and aluminum did not show the saturation behavior. The existence or absence of saturation behavior is explained on the basis of excitation of various parametric decay instabilities modulated by damping processes in a plasma. The suitability of temperature diagnostics from the peak shifts of these emissions is also discussed.

A diagnostic method has been devised and used to measure local transport velocities of negative hydrogen ions (${\mathrm{H}}^{\mathrm{-}}$) within a gas-discharge source chamber. The method is based on laser-pulse photodetachment of negative ions, inducing local changes in the electron and ion densities, whose space and time evolution are traced in detail by means of laser pulses and probes. The technique, which has general applicability, can be used in plasmas where conventional probe or spectroscopic methods are not practicable. A new plasma transport process—monopolar drift—in which particles with the same charge counterflow—is identified by means of this technique. This process can be dominant in multispecies plasmas, i.e., those containing negative as well as positive ions. A kinetic analysis of the plasma evolution in space-time, following localized photodetachment of the negative ions, is presented. The results are compared with the first experimental data obtained on ${\mathrm{H}}^{\mathrm{-}}$, demonstrating the principal modes of the diagnostic technique. The consistency between analysis and experiments indicates that the method is generally useful, and principles for its extension are laid out.

The focusing of particles by a thin plasma lens is analyzed with physical, linearized fluid and particle-in-cell computational models. For parameters similar to next-generation linear colliders, the plasma lens strength can exceed 100 MG/cm, and the luminosity can be enhanced by an order of magnitude by passing each beam through an appropriate plasma slab. The plasma electrons affect the focusing by shifting so as to (partially or completely) charge neutralize the beam. Both overdense and underdense plasma lenses are described (plasma density ${\mathit{n}}_0$ greater or less than beam density ${\mathit{n}}_{\mathit{b}}$). The former case applies equally well to ${\mathit{e}}^{+}$ and ${\mathit{e}}^{\mathrm{-}}$ beams, while the latter has distinct advantages for ${\mathit{e}}^{\mathrm{-}}$ beams (including smaller aberrations and background). The effects of spherical and longitudinal aberrations, emittance, plasma boundaries, and non-linear-plasma dynamics on the final spot size are discussed.

The maximum energy exchange of two harmonically coupled nonlinear oscillators is investigated. We calculate the maximum energy exchange close to resonance and show that the corresponding resonance curves have a universal shape and become broader and smaller when the amplitude-frequency coupling becomes large. Since there is a large variety of nonlinear oscillators where the trajectories are nearly homothetic curves in a phase-space representation, we furthermore investigate the special situation where the oscillators are homothetic. We argue that in this case there is a scaling of the maximum energy exchange at resonance. Numerical investigations show that these relations remain valid if the oscillators are slightly damped or perturbed by random noise.

In this work I prove that, under high gravitational acceleration, nonexponential behavior (strong localization) exists in a cord with N random masses and free boundary conditions. This is due to the fact that the product law for the transfer matrix becomes additive, and not multiplicative, in the random parameter associated with each matrix. This delocalization regime is also valid in the weak-frequency approximation.

We develop an argument that a neural network with asynchronous updating dynamics is capable of sequential retrieval of properly embedded spatial patterns. It is not necessary that the patterns be random, but they have to satisfy certain restrictions. The network is a generalization of the Hopfield model with a specific type of asymmetric synaptic connections. No time delay is introduced in signal transmission, in contrast to some of the recently proposed networks.

The behavior of an analog neural network with parallel dynamics is studied analytically and numerically for two associative-memory learning algorithms, the Hebb rule and the pseudoinverse rule. Phase diagrams in the parameter space of analog gain β and storage ratio α are presented. For both learning rules, the networks have large ‘‘recall’’ phases in which retrieval states exist and convergence to a fixed point is guaranteed by a global stability criterion. We also demonstrate numerically that using a reduced analog gain increases the probability of recall starting from a random initial state. This phenomenon is comparable to thermal annealing used to escape local minima but has the advantage of being deterministic, and therefore easily implemented in electronic hardware. Similarities and differences between analog neural networks and networks with two-state neurons at finite temperature are also discussed.

Target relaxation due to defect diffusion is studied for the case in which N defects are initially distributed at random over an M-site domain embedded in a one-dimensional semi-infinite lattice. A general expression is found for the target survival probability. This expression is valid for arbitrary values of the time and the relaxation rate.

We show that the generalized master equation has a compact solution that is exactly the same as the usual iterative solution.

By formulating the problem of a single polymer chain in a steady extensional flow as a zero-component field theory, the scaling exponent for the flow rate is obtained exactly in terms of known static and linear dynamic exponents.

The results of a nonlinear stability numerical analysis for two- and three-dimensional solitons are presented. Soliton dynamics are described by the Kadomtsev-Petviashvili equation and by an equation suggested by one of us [E. Infeld, Postepy Fiz. 34, 215 (1983)]. The bending turns out to be the simplest perturbation of the system. The results of the simulation confirm stability in both cases.

We show that induced nearest-neighbor bond-orientational ordering present in some two-dimensional liquid-crystal systems modifies the usual logarithmic infrared divergence of director fluctuations. We find that the presence of an induced bond ordering introduces a crossover length scale between two different asymptotic regimes with fluctuations increasing logarithmically with the system size in both regimes. We thus suggest that, for liquid-crystal systems where induced bond ordering exists, the interpretation of finite-size-scaling results in terms of a model that considers only fluctuations in the molecular orientation is questionable.