Quasimolecular x-ray transitions into the 1sσ orbital of the ${\mathrm{Cl}}^{16+}$-Ar collision system were measured in coincidence with scattered particles for collision energies of 20, 10, 5, and 2.5 MeV. The acceleration-deceleration method was used to produce K-shell vacancy bearing projectiles at low velocities. Because K-shell vacancies were brought into the collisions, the excited states of the transient quasimolecules could decay by x-ray emission either on the incoming or on the outgoing parts of the collision trajectory. The resulting interference structures in the quasimolecular x-ray spectra are clearly observed for impact parameters between 600 and 3000 fm. The analysis of these spectra has produced the first experimental determinations of 2pπ-1sσ transition energies as a function of internuclear distance. Details of the experimental procedure, results, and analysis are presented; the velocity and impact-parameter dependencies of spectral decay widths are investigated; and the effects of collision broadening for quasimolecular x rays are discussed.

Recent experiments have shown that it is possible to excite a specific character of an autoionizing state with high selectivity, even far from that character’s resonance. When a state is excited in this manner, the branching ratio with which it decays into various continua may acquire an energy dependence. We have observed dramatic variations in the branching ratio for the decay of 6pns states of barium as a function of energy when they are excited far off-resonance. Using a formalism based on the multichannel-quantum-defect theory, we have derived simple expressions to explain these variations. Essentially, we show that the decay of a state is determined by its couplings to the continua only near its resonance. Far from its resonance, its branching ratio is determined primarily by the couplings of the continua to other near-resonant states.

We derive an infinite-order correction to the adiabatic approximation for the polarization induced in a two-level system by nearly resonant laser irradiation in the low-Rabi-frequency limit for two general classes of field envelopes. We find that Rabi oscillations at the resonant sideband frequency are a general occurrence and study the influence of the pulse shape on the form of the asymptotically decreasing amplitude of the Rabi oscillations as a function of the detuning and time constant. We go beyond the low-Rabi-frequency limit by comparing the analytic solution with numerical solutions of Schrödinger’s time-dependent equation. For symmetric laser-pulse envelopes, the numerical solutions predict eigenvalues of the pulse area at which the amplitude of the Rabi oscillations is zero. The phase of the temporal oscillations changes by π at these eigenvalues. For the special case of a hyperbolic-secant envelope, these eigenvalues correspond to the 2nπ-area pulses of self-induced transparency. For large-area pulses, the central region of the polarization as a function of time contains additional oscillations, the number of oscillations being determined by the number of pulse-area eigenvalues. For a propagating pulse, these oscillations are impressed on the field and amplified, thereby initiating pulse breakup (nonresonant self-induced transparency).

We formulate a quantum-statistical theory of the generation of the fields in a two-photon medium in which several competing nonlinear processes such as four-wave mixing, two-photon absorption, and amplified spontaneous emission are taking place. We derive the equation for the density matrix of the generated fields. We give numerical results for the time-dependent solutions, which enable us to understand the growth and stabilization of the fields in such a medium. We show the generation of new types of the coherent states of the radiation field. Our analysis shows that the generated fields, both in transient and the steady-state domain, have very striking quantum properties leading to squeezing, sub-Poissonian statistics, and violations of Cauchy-Schwarz inequalities. Wherever possible we compare our results with the experiment of Malcuit et al. and the semiclassical theory of Boyd et al.

The density-matrix approach, in the quasistatic or adiabatic limit, is applied to the case of resonant four-wave interactions in four-level systems resulting in explicit analytical expressions for the nonlinear optical parametric and nonparametric polarizations, including ac Stark effects. The results are applicable to any closed-linkage diagram of this type but are applied only to the specific case of resonant frequency tripling. The results revealed the presence of up to six Stark-shifted nonparametric resonances and the presence of saturation and interference effects on the parametric susceptibility. In particular, the third-order susceptibility is shown to saturate to a very small value under high-field conditions which restricts the range of useful pump intensities. Under near-resonant to resonant conditions, these imply that because of either nonparametric loss present at the pump or harmonic frequencies or power-dependent dispersive and saturation effects, conversion efficiencies higher than 25% may be difficult to realize in practice.

The nonlinear evolution of a free-electron laser (FEL) amplifier is investigated for a configuration in which an electron beam propagates through an overmoded rectangular waveguide in the presence of a planar wiggler with parabolically tapered pole pieces. The analysis is fully three dimensional and describes the evolution of an arbitrary number of resonant TE and/or TM modes of the rectangular guide as well as the trajectories of an ensemble of electrons. Numerical simulations are conducted for parameters consistent with the 35-GHz amplifier experiment performed by Orzechowski and co-workers [Phys. Rev. Lett. 54, 889 (1985); 57, 2172 (1986)], in which the ${\mathrm{TE}}_{01}$, ${\mathrm{TE}}_{21}$, and ${\mathrm{TM}}_{21}$ modes were observed. The theory is found to be in good agreement with the experiment. Surprisingly, comparison with a single-mode analysis shows that the enhancement of the efficiency of the ${\mathrm{TE}}_{01}$ mode obtained by means of a tapered wiggler is significantly greater (as well as being in substantial agreement with the experiment) when the ${\mathrm{TE}}_{21}$ and ${\mathrm{TM}}_{21}$ modes are included in the simulation.

A strong pump-power dependence of the Rayleigh-type optical-mixing signal for a water solution of malachite green is reported. Experimental evidence of an exponential pump-power dependence is given. The essential characteristics of the pump-power dependence and the frequency spectrum of the nonlinear signal are qualitatively reproduced by a theoretical model where coupling between pump, probe, and signal fields and propagation effects are considered.

A molecular-dynamics simulation of the stress autocorrelation function η(t) is reported for a dense-liquid-rubidium model. The data indicate a rapid decay of the correlation function for t≲0.3 ps and a somewhat slower time dependence thereafter. Mode-coupling theory, and an alternative dynamical approach involving a conditional probability distribution function, are discussed and then applied in an attempt to explain these data. The former is shown to give a very good account of our results at intermediate and long times, and a simple means of extrapolating the theory to small times is suggested. In our case excellent overall agreement with the computer data is achieved. The dependence of η(t) on the intermediate-scattering function F(q,t), as predicted by the mode-coupling result, confirms the importance of collective atomic rearrangements in the dynamics of shear-stress relaxation.

We have used variational wave functions for the ground state of the tdμ molecular ion to find the probability that after the t-d fusion the system will be in a state with the muon bound to the α particle. This probability, known as the sticking fraction, was found to be 0.8860×${10}^{\mathrm{-}2}$.

Cross sections are calculated for charge transfer into excited states of ${\mathrm{C}}^{5+}$ and ${\mathrm{O}}^{7+}$ from collisions of ${\mathrm{C}}^{6+}$ and ${\mathrm{O}}^{8+}$ with hydrogen atoms in the n=2 and 3 levels. Effective emission cross sections that account for cascading are then derived for spectral lines emitted when the product ions decay. These are compared to the emission cross sections calculated for excitation by ground-state hydrogen. It is shown that in some circumstances the contribution from excited hydrogen atoms must be taken into account when visible lines are used to determine the oxygen or carbon density from charge-exchange spectroscopy in plasmas. Influences on Lyman-series spectra in the x-ray region are also discussed.

The direct [φ(r)→S(k)] and inverse [S(k)→φ(r)] problems of liquid pair structure can be solved by a hypernetted-chain (HNC) integral equation provided the bridge functions B(r) are known. The asymptotic high-density properties of the HNC equation are mapped on the Onsager lower bound to the potential energy, which features ‘‘atoms’’ and ‘‘molecules’’ as mathematical constructs (e.g., the confined-atom Thomas-Fermi model for dense bulk matter). This asymptotic HNC Onsager ‘‘state’’ provides a starting point for analyzing the structure of dense fluids—like the ideal-gas state for dilute fluids. Using only the asymptotic properties of the HNC equation and the single assumption that B(r) is nonsingular, I present the first direct calculation of the bridge functions for a highly correlated fluid and derive their universal characteristics.

This experimental paper studies the time-dependent convective flows in Hele-Shaw slots which are heated from below. The evolution of oscillatory phenomena is traced through the time-dependent regime which is bounded on either side by regions of steady flow. The history of thermal events is recorded by thermocouples which protrude into the test fluid (silicone oil, $N_{\mathrm{Pr}=37\mathrm{}}$) at the four corners of the cavity. Flow patterns are visualized by holographic interferometry and temperature signals are analyzed using Fourier transforms and correlation functions. The results of this investigation suggest that the richness of the dynamic behavior of time-dependent convective flows in high-Prandtl-number fluids is determined by (i) the symmetry of the flow pattern and (ii) by the interaction of individual, spatially distributed instability events.

The motion of the freezing front between a needle-shaped crystal and a supercooled liquid is analyzed for situations where there is forced convection aligned with the crystal axis. It is shown that in the absence of capillary effects the shape of the crystal-melt interface is a paraboloid of revolution, similar to that found in situations where diffusion is the sole heat-transfer mechanism. A relation between the supercooling, the product of tip velocity and tip radius, and the strength of the flow is derived which reduces to the well-known Ivantsov theory in the absence of convection.

In this paper we present molecular-dynamics simulations for a simple liquid-liquid-amphiphile model system. We have determined the density profiles and the influence of the amphiphilic particles on the interfacial tension of the liquid-liquid interface. The amphiphilic particles distribute preferentially to the interface. The addition of a small amount of these surface-active particles gives rise to a substantial decrease of the interfacial tension.

A new type of electrostatic drift waves, which can exist in a plasma whose electrons can be grouped into distinct hot and cold components, is discussed. These waves involve electron dynamics only and can be in a higher-frequency range than that of the ordinary drift waves for which ion motion is crucial. Nonlinear states associated with these drift waves are also investigated.

The quantum limitations on the maximum communication rate (capacity) possible through a single noiseless channel with signals of finite duration are investigated. They may be summarized in the characteristic information function defined here. In the absence of dispersion this description is a Lorentz-invariant one, and applies also in the presence of exterior gravitational fields. For long duration the characteristic information function corresponds to the standard quantum-channel-capacity formula of Gordon, of Lebedev and Levitin, and of Pendry. For finite signal duration it proves useful to distinguish between heralded and self-heralding signals, according to whether their arrival is anticipated or not. The two types have different characteristic functions. These are calculated here for occupation-number signal states in channels where the carrier quantum field may be represented by independent modes. The Gordon-Lebedev-Levitin-Pendry formula provides an upper bound to the exact results for both types of signals. The linear bound on communication rate of Bremermann and the present author bounds the capacity only for self-heralding signals. However, it is the better estimator of capacity for self-heralding signals with modest information content.

This paper investigates the quantum dynamics of a particle on a fermionic surface, e.g., an ion on a metallic surface. The question addressed is whether mapping from a fermion bath to a boson bath can be achieved in this case, as has been shown for several other cases. Using a previously established method, we demonstrate explicitly that in a certain limit, the influence of the coupling between the particle and the bulk fermions is the same as if the particle interacts linearly with an ohmic boson bath, regardless of the coupling strength. The obtained viscosity η is expressed explicitly in terms of the scattering phase shift at the Fermi surface; in particular, the s-wave scattering has no effects on the dynamics.

A density-functional theory is developed in which the local density of a ionic fluid near an interface can be calculated. To find this local fluid structure, the Helmholz free energy is approximated using a perturbation expansion around an optimized reference state. The density of this reference state follows directly from the theory once an approximation for the direct correlation function in the homogeneous reference state is given; i.e., no coarse-graining procedure has to be imposed beforehand. Using the mean-spherical-approximation to the direct correlation function, the theory is applied to three different physical situations. In the restricted primitive electrolyte model near a charged wall we find layering of the counterions, and on adding a neutral third component we find spontaneous charge inversion; i.e., a negatively charged wall develops a positive potential. In the molten-salt regime the model shows very strong oscillations in the potential as a function of the distance from the wall, due to the fluctuation corrections.

The first high-resolution specific-heat measurement on octylphenyl-2-chloro-4-(p-cyano benzoyloxy) benzoate (${\mathrm{DB}}_8$ClCN) is reported. At 390.2 K, the material undergoes a second-order phase transition from a bilayer smectic-A phase (Sm-$A_2$) to a bilayer smectic-C phase (Sm-$C_2$). On cooling, the specific heat of ${\mathrm{DB}}_8$ClCN is independent of temperature in the Sm-$A_2$ phase and increases at $T_c$ in a steplike fashion. In the Sm-$C_2$ phase, the specific heat again becomes temperature independent. This is in striking contrast to previously reported Sm-A–Sm-C transition data, all of which conform to a Landau model with an unusually large sixth-order term.

A new theory of linear and nonlinear optics of liquid crystals is presented. It is based on the ab initio macroscopic Lagrangian theory of Lax and Nelson [Phys. Rev. B 4, 3694 (1971)] for anisotropic crystals and the dissipation function theory of Lam and Lax [Phys. Fluids 21, 9 (1978)] for liquid crystals. The theory is completely deductive and the effects of internal excitations can be included in principle. In this paper the formalism is presented with application to nematic liquid crystals. Dielectric constants with frequency dependence are given. Within the electric dipole approximation an effective equation of motion for the director is derived. It agrees with that of Zel’dovich, Yabiryan, and Chilingaryan (Zh. Eksp. Teor. Fiz. 81, 72 (1981) [Sov. Phys.—JETP 54, 32 (1981)]). This equation describes the optically induced reorientation of the molecules. Flexoelectric constants depending on the optical frequency are presented. The flexoelectric effect in our case arises from the induced dipoles due to director gradients and should exist even in molecules without permanent dipoles or quadrupole moments. This physical mechanism differs from that proposed by Meyer [Phys. Rev. Lett. 22, 918 (1969)] and by Prost and Marcerou [J. Phys. (Paris) 38, 315 (1977)]. Dissipative effects are also discussed with the use of a dissipation function.