Experimental results on the photoionization spectra of excited rubidium atoms in a well-defined $m_l=1\mathrm{}$ state in the presence of an electric field are reported. Spectra corresponding to a definite polarization of the ionizing light exhibit characteristic resonances which are ascribed either to the non-Coulombic part of the central potential or to the spin-orbit interaction. The different line profiles are interpreted as being due to interferences between discrete states and continua, the observed structures depending directly on the density of oscillator strengths and not simply on the density of continuum states.

In this paper we present the values we obtained for the lifetime of the $5 {}_{}{}^2{}_{}{}^{}P_{\frac{1}{2}}^{}{}_{}{}^{}$ level of ${}_{}{}^{85}{}_{}{}^{}\mathrm{Rb}$ using the Hanle effect. These measurements were carried out using two different methods: through a multichannel signal analyzer by photon counting and through a new method based on the determination of the normalized second-order factorial moment $n^{\mathrm{}\left(2\right)\mathrm{}}\mathrm{}\left(T\right)\mathrm{}$ of the fluorescence obtained. We discuss the results and advantages of the new method.

An exact, closed-form solution is presented for the propagation characteristics of an intense, arbitrarily polarized optical beam in an isotropic medium which is subjected to a dc electric field. The related phenomena of dc-electric-field-induced optical rectification and second-harmonic generation are examined as well as the propagation characteristics of a weak beam in an isotropic medium in the presence of both the intense optical and dc fields. One fundamental and noteworthy aspect of the results presented here is that both the formalism and the resulting solutions for these third-order nonlinear optical phenomena are given explicitly and entirely in terms of the set of real, observable Stokes parameters rather than in terms of the electric field amplitudes and phases.

The steady-state solution of the Langevin equation for a dense exciton-biexciton system is investigated. The corrections to the previous work, the contribution of the field—exciton-biexciton interaction, and the local-field correction are shown to modify the threshold of optical bistability significantly.

It is shown that a simple system without a cavity exhibits optical bistability with no hysteresis but with symmetry-breaking or pitchfork bifurcation. The optical system is composed of a cell containing atoms with spin in the ground state, a $\frac{\lambda }{8}$ plate and a mirror, and an incident light beam linearly polarized and nearly resonant with the atomic absorption line. The system has a positive-feedback loop for the rotation of polarization through competitive optical pumping by ${\sigma }_{\pm }$ circularly polarized components. When the intensity of the incident light exceeds a critical value, symmetry breaking occurs and its polarization is subject to rotation in the clockwise or anticlockwise sense. Correspondingly, atomic spin polarization is produced spontaneously in a direction parallel or antiparallel to the incident beam. This new type of optical bistability is found to be explained well in context with the cusp catastrophe, as well as the ordinary one with hysteresis. It is also found that, in the presence of a transverse magnetic field, self-sustained spin precession occurs, which results in the steady-state modulation of the light polarization at about the Larmor frequency. Experimental evidence of the symmetry breaking and pitchfork bifurcation is obtained by using sodium vapor.

The theory of pump-probe experiments is examined using the nonlinear optical susceptibility ${\chi }^{\mathrm{}\left(3\right)\mathrm{}}$. The results of this calculation are shown to differ from those based on a commonly employed heuristic model of pump-probe experiments which assumes that the pump beam "prepares" the system which the probe beam subsequently monitors. This model consequently accounts only for processes where the interactions with the pump field come first in time. The correct expression for the probe absorption line shape as predicted by $\mathrm{Im}{\chi }^{\mathrm{}\left(3\right)\mathrm{}}$ contains a sum over all the possible time orderings of the various radiative interactions. The interference terms missing from the simplified treatment are shown to account for the spectral hole in homogeneously broadened lines observed recently [L. W. Hillman et al., Opt. Commun. 45, 416 (1983)]. In addition we derive a generalized expression for ${\chi }^{\mathrm{}\left(3\right)\mathrm{}}$ that is valid away from the impact limit and contains a realistic population relaxation ($T_1$) matrix.

A calculation is presented of the unequal-time electric field commutation relation for an electromagnetic field coupled to a general multilevel atom treated as a nonrelativistic system. Use is made of the multipolar Hamiltonian in its dipole approximate form to describe the atom-field interaction. Contributions to the commutator in addition to the usual Pauli-Jordan results for a free field are found, and a physical interpretation of their origin is offered.

For the macroscopic description of thermal electromagnetic fluctuations two methods are frequently used: (i) linear-response theory and the fluctuation-dissipation theorem (FDT of the first kind) and (ii) Langevin theory, i.e., stochastic electrodynamics (FDT of the second kind). The two methods are compared. The identity of both theories in global thermal equilibrium is proven. This identity is expressed by relations between the imaginary and real parts of the Green functions on the one side and volume integrations over products of two Green functions on the other side. The correct treatment of infinitely extended vacuum regions with respect to these integrations is discussed. The local interpretation of the two methods in open systems leads to different results: The Poynting vector calculated by the FDT of the first kind vanishes identically, whereas the FDT of the second kind results in a nonvanishing radiative heat transfer. The basic assumptions which are responsible for the different results are extensively discussed. Simple examples are dealt with to make contact with the phenomenological theory of radiative heat transfer.

The equal-time intensity correlation function and variances of the in-phase and out-of-phase components of the electric field radiated by a system of two interacting atoms in the process of resonance fluorescence are calculated analytically for arbitrary values of the field strength, interatomic interaction, and detuning and are illustrated graphically. It is shown that not only photon antibuching but also squeezing can occur in two-atom resonance fluorescence. The influence of dipole-dipole interaction between the atoms on both nonclassical effects is discussed. It is shown that a considerable amount of photon antibunching and squeezing can be obtained for finite detuning of the laser frequency from atomic resonance, especially when the detuning and dipole-dipole interaction parameters cancel out mutually. The maximum value of squeezing for interacting atoms is shown to be less, however, than that for noninteracting atoms. Very strong dipole-dipole interaction reduces the squeezing effect to zero.

When a turbulent binary fluid mixture is quenched below its consolute point, turbulent mixing competes with an instability toward phase separation. A linear stability analysis in the viscous-convective range of concentration fluctuations shows that spinodal decomposition can be suppressed entirely by sufficiently strong stirring. An instability reappears, however, for deep quenches, when the injection rate of concentration fluctuations exceeds the rate of turbulent mixing. The growing concentration fluctuations which result appear to be confined to the viscous-convective range.

A numerical solution and analytic approximation are obtained for the mean-field continuum equations corresponding to diffusion-controlled aggregation. For $d>\mathrm{}2\mathrm{}$ an asymptotic solution is found with the density of the cluster varying as the inverse radius, which suggests a Hausdorff dimension $D=d-1\mathrm{}$.

We study the effects of thermal noise on the microwave-induced steps of a current-driven Josephson junction using the resistively shunted junction model for the junction. We generalize Stephen's method for calculating the width of the steps to the case of an underdamped junction, and to the case of a subharmonic step, provided that there is no overlapping of different steps. In order to treat the case of overlapping steps we introduce a new approach which enables us to calculate the distribution of fluctuations about as well as the transition rates out of the steps. All our results are usually in the form of explicit analytical expressions that facilitate a comparison with experiments.

We analyze the bifurcation of spatial-temporal dissipative structures beyond a hard-mode instability in open reacting systems operating far from equilibrium when a symmetry group (isomorphic to $Z_2$) acts on the diffusion space. The transversality condition which we study is not the classical Hopf bifurcation situation: Let us denote ${\theta }_1\mathrm{}\left(v\right)\mathrm{}\pm i{\theta }_2\mathrm{}\left(v\right)\mathrm{}$ the eigenvalues of the linearized reaction-diffusion problem that cross the imaginary axis for certain $v=v_c$ ($v$ is the bifurcation parameter) and for certain critical spatial wave number. Then our transversality condition reads ${\theta }_1\left(v_c\right)=0\mathrm{}$, $\frac{d^j{\theta }_1(v=v_c)}{d{v}^j}=0\mathrm{}$, and $\frac{d^l{\theta }_1\mathrm{}(v=vc)\mathrm{}}{d{v}^l}\ne 0$, $2\le j<l$, while in the Hopf picture we had ${\theta }_1\left(v_c\right)=0\mathrm{}$, and $\frac{d{\theta }_1}{dv(v=v_c)}\ne 0$. The system is assumed to have reflection symmetry. If the wave number is odd, the bifurcation diagram presents an additional nonthermodynamic branch which is not present in the Hopf case. The three possible bifurcation diagrams are investigated. Several laws of motion are derived from the restriction that the bifurcation equations should be covariant with respect to the action of the symmetry group.

We present the first quantitative study of the screening potential and the effective cross section for thermonuclear reaction in dense plasmas relevant to the inertially confined fusion experiments and the interior of main-sequence stars. It is shown that the enhancement of the thermonuclear reaction rate due to strong Coulomb correlations has a considerable effect in such a dense plasma.

The formulation of quantum-statistical-mechanical expectation values as long-time averages involving the solution of associated Langevin equations with complex drift terms is developed. As an example, some applications of this method to study the quantized Hall effect are presented.

We have measured the turbidity of a critical mixture of methanol and cyclohexane extremely close to the consolute point. The data span the reduced-temperature range ${10}^{-7\mathrm{}}<t<\mathrm{}{10}^{-3\mathrm{}}$, which is two decades closer to $T_c$ than previous measurements. In this temperature range the turbidity varies approximately as $\ln t$, as expected from the integrated form for Ornstein-Zernike scattering. A thin cell (200-μm optical path) with a very small volume (0.08 ml) was used to avoid multiple scattering. A carefully controlled temperature history was used to mix the sample and to minimize the effects of critical wetting layers. The data are consistent with a correlation-length amplitude ${\xi }_0=3.9\mathrm{}\pm 1.0$ Å, in agreement with the value 3.5 Å calculated from two-scale-factor universality and heat-capacity data from the literature.

The intensity distribution of the critical scattering from sodium di-2-ethylhexylsulfosuccinate $\mathrm{AOT}-{\mathrm{D}}_2\mathrm{O}-n$-alkane water-in-oil (W/O) microemulsions has been measured over an extensive range of droplet volume fractions (3-30 vol%) and temperatures (22 to 43°C) in the critical region. The water/surfactant molar ratio of the microemulsion was kept at a constant value of 40.8, for which previous experiments on the temperature variation have been well documented. A structural model of W/O microemulsions based on well-defined surfactant-coated water droplets is firmly established up to a volume fraction of about 20 vol% for all temperatures studied. Data analysis assumes that the cloud points and subsequent phase separation are caused by concentration fluctuations of polydisperse droplets. The major conclusions drawn from the analysis are as follows. (1) The order parameter of the critical phenomenon can be taken to be the volume fraction of the dispersed droplets. (2) The size and polydispersity of the droplets remain essentially constant in the vicinity of the critical point (for a fixed water/surfactant ratio). (3) The critical phenomenon is driven by an increased attraction between the droplets as the critical point is approached. (4) The critical point can be approached by either raising the temperature at fixed volume fraction or by varying the carbon number of the oil solvent at fixed volume fraction and temperature. (5) The nature of the droplets does not change upon a phase separation into two coexisting microemulsions. The data also gives some evidence that the droplet picture of the microemulsion breaks down at sufficiently high concentrations of water and surfactant.

We report the results of a Monte Carlo simulation of a two-dimensional, one-component plasma with $\frac{1}{r}$ interaction potential. This experiment is made in the grand-canonical ensemble which allows a variation of the number of electrons in the system. We study the influence of the chemical potential on the average number of electrons in the system, as well as the influence of the temperature. We show that the system reacts differently for high and low temperatures. In the former case, the ground state of the plasma is neutrality, whereas in the latter one the system has a natural excess of electrons (of order $\sqrt{N_B}$). From a comparison with the two-dimensional plasma with $\ln \left(\frac{1}{r}\right)$ interaction potential, we try to analyze the causes of that difference and to conjecture what will be the behavior of any plasma with a dynamical space dimension not equal to the electrostatic space dimension.

The motion of an elastic pendulum with two degrees of freedom has been investigated in the vicinity of a separatrix, using the Liouville equation. Even for this simple system, an irreversible kinetic equation of the Fokker-Plank type for the momentum-distribution function has been obtained in the limit of a stiff pendulum. This equation describes a monotonic approach to the "microcanonical equilibrium state" for a given energy surface. The diffusion coefficient for the energy of the unperturbed pendulum in this work is directly related to that obtained by Chirikov's heuristic argument.

Molecular-dynamics simulations of dense gases at 297 K (reduced temperature $T^{*}=\frac{T}{\epsilon }=1.47\mathrm{}$) have been carried out using a Lennard-Jones (LJ) (6-12) pair potential and a truncated LJ potential. Our objective was to study the effects of continuous potentials, with and without attractive forces, on density fluctuations in dense fluids, and to interpret recent neutron scattering measurements on krypton gases which revealed significant deviations from results for hard-sphere fluids. We compare our results with experiments and other studies using molecular dynamics and kinetic theory, but which employed hard-sphere potentials. In particular, predictions of the full width at half maximum for $S(q,w)\mathrm{}$ show improved agreement with experiments for values of $q<\mathrm{}{q}_0$ where $q_0$ is the position of the first diffraction peak in $S\left(q\right)\mathrm{}$, when the full potential is used. This satisfactory agreement verifies the sensitivity of the width of $S(q,w)\mathrm{}$ to potential-dependent effects.