We discuss the quantum-statistical properties of a system of absorbing atoms in a ring cavity under the action of a resonant classical field. At a semiclassical level, this system exhibits a bistable behavior in the sense that the transmitted light $E_T$ is a discontinuous function of the incident light $E_I$ with a hysteresis cycle, which resembles a first-order phase transition. Here we specify this analogy stressing the nonthermodynamic properties of this coherently driven system. The quantized field $E_T$ is described by a Glauber $P$ function, which obeys to a Fokker-Planck equation. This equation is derived from a very general master equation in the case of a good quality cavity by neglecting derivatives of order higher than second. A remarkable feature of this equation is that the diffusion coefficient is a nonlinear function of the field intensity. The steady-state solution of the Fokker-Planck equation has the following features. The $P$ function can be single or double peaked according to the value of $E_I$. In fact, our Fokker-Planck equation is like the one of a Brownian particle moving in a potential which presents one minimum or minima separated by a barrier, depending on the value of $E_I$. The peaks of the $P$ function correspond to the minima of the potential. In the double-peaked case, the peaks have different widths. The maxima of the $p$ function occur at values of $E_T$, which coincide with the classical stationary solutions that are stable according to the linear stability analysis. These semiclassical solutions give a discontinuous two-valued function of $E_T$ vs $E_I$ (hysteresis cycle). On the contrary, the mean value $〈E_T〉$ is a continuous single-valued function of $E_I$, which behaves similarly to the Maxwell construction of a first-order phase transition: $〈E_T〉$ jumps from low to high values in a very sharp transition region of $E_I$, which lies between the discontinuity points of the hysteresis cycle. However, the Maxwell rule that we obtain is quite different from that of equilibrium thermodynamics. The mean-square fluctuations of $E_T$ are always very small except in the transition region where they are quite remarkable. We also present an exact expression of the stationary $P$ function, which takes into account all higher-order derivatives, and we show that the Fokker-Planck approximation is very good for calculating mean values and mean-square fluctuations. Finally, we discuss the connections which link bistable absorption to the laser with injected signal and to the laser with a saturable absorber.

The critical behavior of the depolarized Rayleigh spectrum in the system nitrobenzene-$n$-hexane has not been previously accurately assessed because in usual Fabry-Pérot experiments the multiple-scattering contribution completely overwhelms the spectrum within a few degrees of the critical temperature $T_c$. By using a high-performance double-pass Fabry-Pérot interferometer, we have been able to obtain accurate measurements of both linewidth and intensity, from 50°C to temperatures as close as 0.026°C to $T_c$. Two lines due to nitrobenzene molecules in solution are visible. The first line (1) is intense and narrow, while the second line (2) is weak and broad. The spectral intensity of (2) is about ${10}^{-2\mathrm{}}$ times that of (1). The linewidth of (1) is seen to behave as the inverse shear viscosity, with a critical divergence with exponent $Y\approx 0.04$. The intensity of (1) exhibits a power-law divergence with exponent $\Omega \approx 0.2$. These results are compared with data obtained in pure nitrobenzene and $n$-hexane, and in noncritical mixtures of nitrobenzene.

We find that certain cyano Schiff bases and cyanobiphenyls with alkyl or alkoxy end groups exhibiting a bilayer smectic-$A$ phase, also exhibit a reentrant nematic phase at high pressure. A tentative model of the bilayer smectic-$A$ phase and the mechanism whereby the reentrant transition may occur is presented.

Rayleigh linewidth measurements have been performed on methanol and acetonitrile in the critical region using a light-beating spectrometer. The result of this study was that neither the dipole moments nor the strong associations seem to have any influence on the critical behavior of the liquids since the critical exponent $\nu$ for the correlation length $\xi$ measured for methanol and acetonitrile agrees fairly well with the values published for simple liquids. The experimental setup is described and fundamental problems in locating the critical point are treated extensively.

Starting from the Lennard-Jones interatomic potential, a modified cell theory has been used to describe the solid-liquid phase transition in argon. The cell-size variations may be evaluated by a self-consistent condition. With the inclusion of cell-size variations, the transition temperature, the solid and liquid densities, and the liquid-phase radial-distribution functions have been calculated. These ab initio results are in satisfactory agreement with molecular-dynamics calculations as well as experimental data on argon.

Collective density fluctuations in ${}_{}{}^{36}{}_{}{}^{}\mathrm{Ar}$ gas at 456 atm pressure and 295 K have been studied by thermal neutron scattering using time-of-flight techniques. Well-defined collective modes are observed for wave-vector transfers $\overrightarrow{\mathrm{k}}\lesssim 0.1$ ${\mathrm{\AA }}^{-1\mathrm{}}$ which become strongly damped at larger $k$. The data have been compared in detail with predictions of the generalized Enskog equation. Although extrapolation of the predictions to $k=0\mathrm{}$ gives the correct transport coefficients to better than 5% accuracy, the generalized Enskog equation does not properly predict the wave-vector dependence of the scattering law.

Density and current fluctuation spectra of the one-component plasma with plasma parameter $\Gamma =140\mathrm{}$ are calculated self-consistently within a mode-coupling approximation for longitudinal and transverse memory kernels. The theory uses as input the static two- and three-particle-correlation functions. The resulting spectra are in satisfactory agreement with molecular-dynamics calculations for all wave numbers. In particular, the peak structure of density-fluctuation and shear-mode spectra in the intermediate-wave-number regime as well as their interrelationships are discussed in terms of the mode-decay processes.

We present molecular-dynamics computations of the thermal conductivity, and two viscosities of the classical one-component plasma in a rigid, uniform background. The Kubo formulas are proved for the Coulomb case, and it is shown that some care has to be taken in defining the Kubo currents in order to avoid divergences arising from the long range of the Coulomb interactions.

One of the classical statements of the second law of thermodynamics is derived from an assumption about the behavior of an isolated heat reservoir, together with a further assumption, which seems to be a rather weak one, about initial states. The derivation is otherwise based on the unitary property of time evolution in quantum mechanics. An analogous derivation, within the framework of classical mechanics, is also given; in this case Liouville's theorem substitutes for unitarity.

The first four surface virial coefficients in the virial expansion of the surface tension of a hard-sphere fluid are calculated. The virial expansion for surface tension of a hard-spheres-hard-wall system is compared with the results obtained by means of the Percus-Yevick, hypernetted-chain, and scaled particle theories.

New systematic trend curves for the oscillator strengths of the $4s-4p$ resonance transitions of Cu- and Zn-like ions have been derived from a critical analysis. This analysis takes into account recent theoretical results as well as a simulation study of cascade effects in beam-foil experiments which showed that the currently available experimental data for high ions may be too low by as much as 40-70%. The new analysis yields reliable interpolated data for many ions within the Cu and Zn isoelectronic sequences.

A previously published analysis of the orbital contribution to the electronic stopping cross section ($S_e$) for slow-ion-atom collisions has been extended for atoms and singly charged positive ions up to xenon and zinc, respectively. The importance of excited states to the evaluation of $S_e$ is discussed by comparing the orbital contribution of excited states of Ti, ${\mathrm{Sc}}^{+}$, ${\mathrm{V}}^{+}$, ${\mathrm{Fe}}^{+}$ with their respective ground states. The exact values for Brice's $Z$ parameters are obtained and a comparison is made with Brice's values.

Calculations are presented of cross sections for electron capture from atomic hydrogen to $n$, $l$ excited states of $p^{+}$, ${\alpha }^{++}$, and ${\mathrm{Be}}^{+4\mathrm{}}$ at velocities of about 100 keV/amu using both the Brinkman-Kramers (BK) and Jackson-Schiff (JS) approximations, and to excited states of ${\mathrm{O}}^{+8\mathrm{}}$ in the BK approximation. Capture occurs into levels with increasing $n$ and $l$ as the projectile charge $Z_1$ is increased. While the BK and JS ratios $\frac{{\sigma }_{\mathrm{nl}}}{{\Sigma }_{n,l}{\sigma }_{\mathrm{nl}}}$ are quite similar for $p$+H, these two approximations give rather different distributions in $n$ and $l$ for ${\mathrm{Be}}^{+4\mathrm{}}$ where transfer to excited levels dominates.

A simple way to properly take into account the exchange effects within the independent-atom model of electron-molecule scattering is pointed out and applied to $e^{-}$-${\mathrm{H}}_2$ elastic scattering.

An essential step in the proof of the quantum recurrence theorem is shown to follow from the Poincaré recurrence theorem of classical mechanics.

The first-order Faddeev approximation for ionization of atoms by ion impact is discussed within the framework of the impact-parameter approximation. It is suggested that the approximation might be improved by adding a term to the final-state wave function to make the initial- and final-state wave functions orthogonal at all internuclear separations. Calculations have been performed of the differential cross section with respect to energy and angle for electrons ejected in the forward direction from helium atoms by 300-keV protons. Comparison with experiment indicates that the modification to the first-order Faddeev approximation represents a considerable improvement.

In the reduction of the Glauber scattering amplitude for charged-particle-neutral-helium collisions, a modified form of Lommel functions has been introduced by Thomas and Chan for evaluating some integrals. An alternative numerical method is proposed for the evaluation of these integrals.

The experiment of Liran, Levin, Erez, and Jortner is clearly very interesting. Nevertheless, we feel that the numerical values which has been found for the broadening of the sodium $3S-4D$ two-photon transition by collisions with argon atoms is wrong. Using Doppler-free two-photon spectroscopy we have measured this cross-section to be 1050±110 ${\mathrm{\AA }}^2$. This value is in good agreement with other experiments and with theoretical predictions. We present several possible interpretations of the value obtained by Liran et al.

The two-potential eikonal approximation of Ishihara and Chen is used to calculate the cross section for electron-impact excitation of hydrogen ($1s\to 2s, 2p$) at intermediate energies. It is found that the differential cross section in the large-scattering-angle region shows improvement (though it still underestimates the experimental data), as expected, over the usual Glauber-approximation results. There is, however, no improvement in the integrated cross section.