We analyze theoretically a one-dimensional model of laser cooling of an atom or ion trapped in a cavity. We assume that the cavity loss rate is much larger than the atom-cavity coupling (bad-cavity limit) and that the atomic excited state is weakly occupied (low saturation limit). After elimination of the cavity mode and the atomic excited state, we derive rate equations for the populations of the trap states. We find that in the Lamb-Dicke limit the atom can be cooled to the ground state of the trap even in the strong confinement limit. This result is interpreted in terms of quantum interferences between different cooling and heating processes involving spontaneous emission in the cavity.

Quantum fluctuations in image irradiance for spatially partially coherent optical imaging are studied as a function of the spatial coherence of the source. The advantage of using incoherent or partially coherent illumination for the quantum noise performance is discussed. The results are compared for illumination by coherent, spatially single-mode squeezed, and spatially multimode squeezed light. It is shown that a signal-to-noise ratio can be improved beyond the shot-noise limit employing the two latter sources.

The probability density function (PDF) of photoelectron counts is investigated theoretically for a fundamental soliton corrupted by the spontaneous-emission noise that is generated by optical amplifiers employed along a fiber-optic transmission line. An analysis based on the orthogonal expansion of the light amplitude makes it possible to find the exact PDF that includes the influence of the soliton width, the hyperbolic-secant pulse shape, and a Lorentzian bandpass optical filter. The derived PDF is applicable to any cascaded chain of amplifiers and, as a practical example, is calculated for a receiver equipped with an optical preamplifier. The pulse width significantly affects photoelectron statistics when the product of the filter bandwidth and the counting interval is small. By comparing our exact PDF with the commonly used approximate Gaussian and Laguerre PDFs, we find that such approximations introduce large errors (by several orders of magnitude) in the tail of the PDF that is relevant for calculation of bit error rates in soliton communication systems.

In this paper we demonstrate that state scarring is a more pervasive and subtle phenomenon than commonly believed. We show that scars can exist in regions where there are no periodic orbits, and that they are due to families of neutrally stable orbits that reside in a neighboring system (‘‘ghosts’’). Furthermore, seemingly unscarred diffuse states can turn out to be supported by a single extensive family of periodic orbits.

Dirac’s equation provides the most rigorous basis known for many calculations in relativistic quantum mechanics. A set of dynamical equations having greater intuitive content can be derived from Dirac’s equation without any approximations. These secondary equations govern the properties of a Dirac particle as functions of time and space and are similar to the corresponding equations governing a classical charged fluid. Several new density functions are implied by these equations and are appropriate for incorporation into the various semiclassical models of physics.

Electron field-emission experiments conclusively eliminate a family of classical stochastic theories that have been proposed as alternatives to quantum mechanics. Modified stochastic theories with certain nonlocal interactions may be postulated and may agree with the experimental data, but these revised theories should not be described as classical stochastic theories.

Valence shell relativistic configuration-interaction calculations for 6p attachments to the Pr ground state yield two bound states for the configuration 4${\mathit{f}}^3$6${\mathit{s}}^2$6p. The two levels (J) and their electron affinities (EA) are (J, EA (eV))=(4, 0.110) and (5, 0.128). We were unable to find any evidence for 5d or 4f attachments. These results are consistent with the relatively modest experimental yields obtained for ${\mathrm{Pr}}^{\mathrm{-}}$.

A measurement of the Ni-like Er 3${\mathit{d}}_{3/2\mathrm{-}}$4${\mathit{f}}_{5/2}$ transition energy is presented. The value differs from a previous prediction that was based on an interpolation of measurements of the transition in elements of nearby atomic numbers. This transition was previously thought to have an energy very near the H-like Al Ly-${\mathrm{\alpha }}_2$ line, making the system a favorable candidate for resonant photopumping. The measured energy difference between the Er 3d-4f and Al Ly-${\mathrm{\alpha }}_2$ transitions is 1.8±0.3 eV or 1040 ppm indicating that the overlap is not as favorable as previously thought. The difficulties with the previous interpolation are discussed. Additional measurements of the 3${\mathit{d}}_{3/2\mathrm{-}}$4${\mathit{f}}_{5/2}$ transition energies in Ni-like Pr, Yb, and Pt are presented that achieve a precision better than 100 ppm. Together with the value for Er, these are used to derive a new interpolation with better predictive capabilities.

Coincidence measurements of the angular distributions of ${\mathrm{N}}^{+}$ dissociated from ${\mathrm{N}}_2^{2+}$ produced by electron impact on ${\mathrm{N}}_2$ have been made. With an incident electron energy of 60 eV, 3.9-, 5.0-, and 7.2-eV ${\mathrm{N}}^{+}$ ions were detected separately using the time-of-flight and coincidence method at selected angles between 40° and 140°. The results show flat angular distributions and the flatness for each case could be ascribed to different origins.

Deuterated ice targets were bombarded by oxygen clusters at incident energies of 115 and 172.5 keV per oxygen atom. The fusion yields were measured by detecting the 3-MeV protons from the D(D,p${)}^3$H channel. Comparisons of the fusion yields from clusters with those from single atoms at the same energy per oxygen atom show no evidence of collective enhancement, in disagreement with a recent report [Bae, Beuhler, Chu, Friedlander, and Friedman, Phys. Rev. A 48, 4461 (1993)].

The maximal population transfer in a two-level system induced by an ultrashort, linearly polarized laser pulse and the optimal pulse shape for maximal population transfer depend critically on the pulse duration and phase. In the case of temporally symmetric pulses, half-cycle pulses are optimal and are capable of inverting the population completely. For antisymmetric pulses, bimodal pulses are optimal but cannot invert the population completely. Deviations from the Rabi formula for the pulse parameters do not occur until the pulses become shorter than a few optical cycles.

We generalize the concept of Schrödinger cat states to macroscopic superpositions of two pairs coherent states. We show that these states can have various nonclassical features such as the violation of the Cauchy-Schwarz inequality, sub-Poissonian photon-number statistics, and squeezing. We propose schemes to generate these states.

The standard quantum limit (SQL) as an error probability can be broken by means of a unitary transformation such as a Kerr medium installed before the conventional homodyne scheme. Such an improvement in the error probability cannot be realized by the conventional unitary transformer like squeezers for fixed coherent state inputs. The physical origin of this effect is nonlinear quantum interference for the components of the output state evolved from the input coherent state. We clarify the conditions for overcoming the SQL by unitary transformation and show several examples of the interaction Hamiltonian which generate unitary processes to break the SQL.

We have derived the normal-ordered form of the SU(1,1) and SU(2) squeeze operators in the multimode bosonic representation and evaluated the coherent-state matrix elements of these operators. The analysis can be applied to any exponential functions of the generators of the Lie algebras su(1,1) and su(2) without difficulty.

Irby et al. [Phys. Rev. A 47, 2957 (1993)] claim that their experimental results, obtained with ${\mathrm{C}}^{+}$, ${\mathrm{C}}^{2+}$, and ${\mathrm{C}}^{3+}$ projectiles on He and Ne targets, strongly support the hypothesis of an independent ‘‘saddle-point’’ ionization mechanism for intermediate ion-atom collision velocities. They state that in previous measurements, in which single-stage electron analyzers were used, this process has been obscured by reflected and secondary electrons. We present experimental evidence that this effect exists, but is insignificant. We suggest that they may have observed a remnant of convoy-cusp electrons, which subsists at an emission angle of 10°.

It is argued that the electron plate-impact contamination reported in the preceding Comment [G. Bernardi and W. Meckbach, Phys. Rev. A 51, 1709 (1995)] cannot be considered as insignificant.

The Vlasov kinetic equation can be viewed as the dynamical equivalent of the static Thomas-Fermi theory. By exploiting this analogy, the closed-orbit theory of the level density, which has been used by various authors for evaluating quantum corrections to the Thomas-Fermi level density, is generalized for calculating quantum corrections to the semiclassical propagator given by the Vlasov method. It is shown that the Vlasov strength function obeys a sum rule analogous to the well-known quantum energy-weighted sum rule. Explicit calculations of quantum corrections are performed for a simplified model of a large metal cluster (electrons in a spherical cavity).

Two general methods are developed to optimally control chaotic dynamics and, for each method, control approximations are presented. The first method consists of seeking a path, defined by an average trajectory, to meet the objective at a target time. The second method has the goal of achieving control by drawing in the irregular trajectories around an imposed regular fiducial trajectory. These two methods are complementary in their approaches to achieving control over chaotic behavior. A Lagrange-multiplier technique and a penalty method are used in the process of finding the optimal solution that minimizes a cost functional. Approximations are introduced for the Lagrange multipliers, based on preserving and controlling only the average trajectory. Numerical results are presented for the two methods. The optimal external perturbation ε(t) influences the chaotic trajectories by redirecting them to a region in phase space where it is possible to maintain a tight bundle of trajectories with a minimum external interaction. Moreover, the results show that the interaction energy is small in comparison with the total energy of the system, indicating an efficient control strategy.