Total slow-electron (${\mathit{E}}_{\mathit{e}}$≤60 eV) yields as derived from measured electron-emission statistics are presented for impact of various slow (${\mathit{v}}_{\mathit{p}}$≤${10}^5$ m/s) multicharged ions (${\mathrm{N}}^{5+}$,${\mathrm{N}}^{6+}$,${\mathrm{Ne}}^{5+}$–${\mathrm{Ne}}^{10+}$, ${\mathrm{Ar}}^{5+}$–${\mathrm{Ar}}^{16+}$, ${\mathrm{Kr}}^{5+}$–${\mathrm{Kr}}^{10+}$, ${\mathrm{Xe}}^{6+}$,${\mathrm{Xe}}^{8+}$,${\mathrm{Xe}}^{10+}$, and ${\mathrm{I}}^{16+}$,${\mathrm{I}}^{20+}$,${\mathrm{I}}^{23+}$,${\mathrm{I}}^{25+}$) on atomically clean polycrystalline gold. The experimental data can successfully be modeled by classical over-barrier-type calculations, in which way three different electron-emission processes have been identified, i.e., autoionization of the transiently neutralized projectile (‘‘hollow atom’’) on its approach toward the surface, promotion into vacuum of electrons captured by the projectile, and ‘‘peeling off’’ of electrons remaining bound in highly excited states until projectile impact on the surface. The relative importance of these three processes depends strongly both on projectile species and impact velocity.

Emission of slow electrons (${\mathit{E}}_{\mathit{e}}$≤60 eV) induced by impact of slow multicharged ions (impact velocity ${\mathit{v}}_{\mathit{p}}$≤2×${10}^5$ m/s) onto an atomically clean, polycrystalline gold surface has been studied both experimentally and by numerical simulation, based on the resulting electron-emission statistics. The projectile ions (${\mathrm{N}}^{\mathit{q}+}$, q=5,6; ${\mathrm{Ne}}^{\mathit{q}+}$, q=5–10; ${\mathrm{Ar}}^{\mathit{q}+}$, q=5–16; ${\mathrm{Kr}}^{\mathit{q}+}$, q=5–10; ${\mathrm{Xe}}^{\mathit{q}+}$, q=6,8,10; ${\mathrm{I}}^{\mathit{q}+}$, q=16,20,23,25) have been extracted from a recoil ion source pumped by the GSI UNILAC heavy-ion accelerator in Darmstadt, Germany. We discuss the shape of the experimentally obtained electron-emission statistics and, by means of numerical simulation based on the classical over-the-barrier model put forward recently by Burgdörfer, Lerner, and Meyer [Phys. Rev. A 44, 5674 (1991)], identify the various processes contributing to the ‘‘above-surface’’ electron emission, i.e., taking place until projectile impact on the surface. In particular, for impact of slow (E≥50 eV) ${\mathrm{Ar}}^{12+}$ we show that most of the emitted electrons have energies below 50 eV, with the above-surface-produced fast Auger electrons being a small minority of less than 1%.

Kth-power amplitude squeezing (KPAS) is introduced and examined in the nonlinear-harmonic-oscillator model and the Jaynes-Cummings model with an intensity coupling. We discover that, in these two cases, the long-time behavior of KPAS (k≤4) for the strong initial field periodically displays a collapse-revival-valley and squeezed collapse-revival phenomenon, respectively. For a nonclassical state, the relation between KPAS and ordinary squeezing, Hong-Mandel squeezing, and antibunching is investigated. The results show that KPAS is another intrinsic and independent nonclassical effect inequivalent to the last three.

As a generalization of the optical Schrödinger cats, discrete sets of coherent states are considered on a circle in the α plane. It is shown that simple superpositions of Schrödinger cats exhibit amplitude squeezing, similarly to the case of a superposition of several coherent states along a straight line that shows quadrature squeezing. The interference fringes between the coherent states form the annuli of the Fock states in the Wigner-function picture. It is also shown that a continuous superposition of coherent states on a circle can serve as a basis for the representation of any state.

Experiments with erbium-doped fiber lasers demonstrate cw, sinusoidal, and self-pulsing operation. The obtained regimes depend on three control parameters: ion-pair concentration, photon lifetime, and pumping rate. We present a theoretical model which describes the active medium as a mixture of isolated ions and ion pairs. Starting with the adapted laser rate equations we show that the description of the dynamical behavior of this system can be reduced to only four first-order coupled equations. A linear stability analysis demonstrates the existence of self-pulsing for a finite range of pumping rates. At both ends of this range Hopf bifurcations occur: one located near the first laser threshold and the other at a higher pumping ratio, whose position is closely related to the pair concentration. Results of numerical calculations are in good qualitative agreement with our experimental data.

The intensity-noise properties of an injection-locked semiconductor laser were studied experimentally. The constant-current-driven semiconductor laser producing the amplitude-squeezed state whose intensity noise was reduced below the standard quantum limit (SQL) by 0.72 dB was injection-locked by an external master laser. The measured intensity-noise level of the injection-locked semiconductor laser was 0.91 dB below the SQL. This experimental result indicates that a phase-coherent amplitude-squeezed state or squeezed vacuum state together with a reference local oscillator wave can be generated directly by semiconductor laser systems.

Spectral-hole burning (SHB) profoundly affects the modulation and noise properties of laser oscillators and amplifiers at high optical power. The present semiclassical theory of SHB shows that the optical gain should be considered a function of the carrier number and photon rate (rather than photon number) plus a fluctuation at the shot-noise level (for full population inversion). Constant-voltage-driven laser diodes generate amplitude-squeezed light, a result not predicted by previous theories that treat gain compression in a formal way. Amplitude and phase noise of oscillators and amplifiers are considered.

We present a unified model of photonic-band-gap impurities and optical microcavities by using a one-dimensional distributed Bragg reflector. A variable-width center layer creates impurity photon states within the photonic band gap. We show that the best optical features (light localization, electric-field strength, linewidth, and laser threshold gain) are obtained when the layer thickness is half-wave (or a multiple), in which case the photon impurity state is located at the middle of the photonic band gap and the structure is a λ/2 (or higher-order) Fabry-Pérot microcavity.

We study a homogeneously broadened four-level model for a coherently pumped laser with pump and laser fields having crossed linear polarizations. For a parameter range of the type explored in the experiments by Tang et al. [Phys. Rev. A 44, R35 (1991)] the system exhibits a family of type-III-intermittency transitions to chaos in which the onset of intermittency is preceded by period-2, period-3, or period-4 states. We find similarities but also differences between the results of our theory and their experimental results.

Based on the level dependence of collision cross sections, a mechanism of light amplification without inversion in a Doppler-broadened three-level system is further studied. In the rate equations, we use a collision-kernel model and consider all of the decay and pumping rates. The emission and absorption profiles are derived by solving the coupled mass-velocity equations numerically. Then, the possibility of obtaining positive gain with noninversion for an ensemble of particles is shown.

We study the effect of the atomic decay and the pumping statistics in degenerate two-photon oscillators. We obtain an expression for the gain that is valid for all values of the atomic lifetime. Numerical results are presented for regular and Poissonian pumping statistics.

We have examined the ionization of ${\mathrm{Na}}_2$ Rydberg states of principal quantum number n=11–16 converging to rotational-vibrational, v=3–6 and R<10, states of ${\mathrm{Na}}_2^{+}$ by 8-GHz microwave fields. Rydberg states lying below the lowest, v=0, R=0 (v and R are the vibrational and rotational quantum numbers of the molecular ion), ion limit of ${\mathrm{Na}}_2^{+}$ are observed to ionize at approximately E=1/3${\mathit{n}}_0^5$, where ${\mathit{n}}_0$ is the effective quantum number relative to the v=R=0 limit. The agreement with this dependence is not nearly as good as in an atom, and the possible reasons for the deviations are discussed.

An extension of the standard one-photon as well as the two-photon Jaynes-Cummings (JC) model has been done to include transient effects due to a modulation of the atom-field coupling coefficient. An explicit time dependence is given for the case of a linear sweep of the coupling parameter. We have shown the effects of such a sweep on the dynamical evolution of a two-level atom undergoing a one- or two-photon process in a single nondecaying mode of the field in a coherent state. The statistical aspects of the field, such as intensity-intensity correlation and squeezing, are also investigated and are found to be both qualitatively as well as quantitatively different from the standard JC models.

A simple model is introduced to simulate the detection of cavity radiation with the detector placed outside the cavity. Explicit expressions, in terms of initial photon-number distribution, for the time-dependent distributions of photoelectron number and conditional as well as unconditional waiting times are derived. The effect of photodetection on the quantum state of the radiation in the present arrangement is found to be unchanged from that with the detector placed inside the cavity. The well-known Q parameter introduced by Mondel [Opt. Lett. 4, 205 (1979)] is found to be exactly equal to the change in the expectation value of photon number due to the detection of a photon. Photon bunching or antibunching, reflecting classical or nonclassical radiation, is determined by whether conditional waiting time is less or greater than unconditional waiting time. When this criterion is applied to three specific examples, namely, coherent, thermal, and number states, the results are as expected.

In this paper the inherent properties of period-doubling behavior in deeply modulated laser diodes were investigated. The thresholds for the period doubling in terms of the controlling parameters were indicated. With the consideration of Langevin noise in the single-mode rate equations, it was indicated that the noise acted as a virtual Hopf precursor for the event of period doubling. The linewidth and frequency shifting of the noise bump were predicted by Floquet multipliers. Finally, a comparison of the noise bumps obtained from the simulated and analytic methods was also presented.

We report on instabilities in the transmitted polarization of a laser beam crossing a planar Fabry-Pérot resonator filled with sodium vapor, at a null magnetic field, when the input field is linearly polarized and tuned near the center of the D1 line. We have measured bifurcation diagrams in the input-power–cavity-detuning parameter space at various angles of incidence. The system shows monostability, bistability, and limit cycles in the transmitted polarization. Oscillations appear only when the angle of incidence is different from zero and the oscillation phase changes with the position on the transverse profile. The observed dynamics is explained by a simplified model which accounts for Zeeman and hyperfine pumpings, focusing and defocusing effects, and atomic motion. By means of this model, we have determined the codimensionality of the system within a theory of the dynamical systems.

The phase distribution function [after Pegg and Barnett, Phys. Rev. A 39, 1665 (1989)] is calculated for the micromaser and the lowest modal phase diffusion constant is compared to the linewidth calculation of Scully et al. [Phys. Rev. A 44, 5992 (1991)]. Utilizing the Fokker-Planck approach results in a linewidth which does not show ‘‘trapping’’ effects.

A dynamical analysis of nonlinear optical resonators is performed in the framework of a single-mode formalism in the plane-wave regime, taking the finite response time of the nonlinear medium into account. We demonstrate that analytically solvable generic equations can be derived for the time evolution of the output intensity, in the two limiting cases where the medium’s response time is either very large or very small as compared to the cavity buildup time. The stability picture and the characteristic time scales, important for applications in the field of optical information processing, are investigated analytically. This study reveals that a surprising effect may occur where an increase in the material-response time leads to a decrease of the relaxation time of the nonlinear device. This effect is called the speeding-up phenomenon. We also present a numerical study of the full nonlinear dynamics.

We describe ac Stark shift production and measurement in a two-laser mode where a first laser is tuned through a two-photon resonance between the ground state and an excited state in order to record a line shape for the corresponding resonantly enhanced multiphoton ionization. A second laser is tuned to a fixed wavelength which is close to a resonance between the upper state in the two-photon transition and a third state which has dipole-allowed transitions back to the ground state. At low concentrations the second laser leads to very large ac Stark shifts in the two-photon excitation and in the related resonantly enhanced multiphoton ionization. However, at high concentrations a destructive interference occurs involving the three-photon coupling between the third state and the ground state and the one-photon coupling due to the four-wave-mixing field at frequency 2${\mathrm{\omega }}_{\mathit{L}1}$±${\mathrm{\omega }}_{\mathit{L}2}$. This interference leads to a pressure-dependent suppression of the amplitude for the third state. When this destructive interference occurs, the ac Stark shift due to the second laser undergoes a corresponding pressure-dependent suppression. The criteria for the pressure to be high enough to completely suppress the ac Stark shift is that the real part of phase mismatch ‖Δk‖ at the angular frequency 2${\mathrm{\omega }}_{\mathit{L}1}$±${\mathrm{\omega }}_{\mathit{L}2}$ and (or) the absorption coefficient β at the same frequency should be very large, so that with unfocused focused beams ‖Δk‖L>5[(${\mathrm{\Delta }}_{\mathit{s}}^{\mathrm{}\left(\max \right)\mathrm{}}$τ${]}^{1/2}$, where L is the thickness of gas penetrated before the shift is measured, ${\mathrm{\Delta }}_{\mathit{s}}^{\mathrm{}\left(\max \right)\mathrm{}}$ is the maximum ac Stark shift at very low pressure, and τ is the pulse length of the lasers.

At positions and concentrations such that 1<‖Δk‖L≤2[(${\mathrm{\Delta }}_{\mathit{s}}^{\mathrm{}\left(\max \right)\mathrm{}}$τ)${]}^{1/2}$ the ac Stark shift is present but the two-photon excitation is suppressed due to a destructive interference between excitation by two photons from the first laser and by two-photon excitation between a four-wave-mixing photon and a photon from the second laser.

A charged particle in a double well can emit, when driven by an intense laser, intense low-frequency radiation or become strongly localized in one of the wells. By using perturbation theory and an asymptotic analysis, we determine the conditions for which these processes take place for a two-level model.