Collisionless fragmentation of nonrotating model n-atom metal clusters (n=12, 13, and 14) is studied using isoergic molecular-dynamics simulations. Minimum-energy paths for fragmentation are mapped out as functions of the distance between the centers of mass of the fragments. These paths provide information on the fragmentation energies for the different fragmentation channels. Fragmentation patterns (distributions of the fragmentation channel probabilities) and global and channel-specific fragmentation rate constants are computed and analyzed as functions of the internal energy and of the size of the clusters. The trends derived from the dynamics are compared with those obtained using the RRK and TST statistical approaches. The dynamics of the fragmentation process is analyzed in terms of characteristic quantities such as the distance between the centers of mass of the fragments, their relative translational energy, and their interaction energy, all considered as functions of time.

The electronic structure of atomic clusters is studied as a function of the geometrical parameters and number of atoms N in the clusters. Self-consistent electron wave functions and energies are calculated for the clusters with closed and open shells for N ranging from 4 to 40 and for the various radii of positive cluster cores. These calculations are done both for the homogeneous and inhomogeneous jellium models of the cluster core. The equilibrium values of jellium core radius have been obtained using ab initio Hartree-Fock calculations for the valence electrons. The importance of the role of nonlocal exchange interaction in the creation of the equilibrium cluster structure is also shown. An adiabatic type of collective vibration mode is predicted. Vibrational freuqencies and amplitudes are calculated for the ‘‘magic’’ number clusters. The influence of the geometrical parameters on the physical properties of clusters is discussed.

Low-frequency approximations for both spontaneous and stimulated bremsstrahlung processes are derived that are applicable to cases in which an electron is scattered from an atom at an energy near the threshold for excitation of the target. Existing approximation procedures are modified to properly account for the strong energy dependence of the cross section, in the form of cusps or rounded steps, caused by the opening up of a new channel. These modifications preserve an essential feature of standard versions of the low-frequency approximation in that the physical, field-free scattering amplitude appears as input to the calculation. An estimate of the error in the Feshbach-Yennie approximation [Nucl. Phys. 37, 150 (1962)] for spontaneous bremsstrahlung is obtained through an evaluation of that contribution to the matrix element corresponding to radiation by the projectile following a virtual excitation of the target. To illustrate the effect of the anomalous energy dependence in a domain containing the threshold, as well as the significance of the above-mentioned correction to the Feshbach-Yennie approximation, calculations are described for a model system; the model is defined by a parametrization of the field-free scattering amplitude that reproduces the correct threshold behavior. A closely analogous treatment is given of laser-assisted scattering near a reaction threshold. A sum rule is derived for this external-field process which, in its simplest form, reproduces a recently conjectured approximation and provides higher-order corrections to it.

We experimentally investigate the transient dynamics of two-level atoms driven by a bichromatic optical excitation field when the bichromatic field is configured to contain a strong resonant component and a weaker nonresonant perturber component. We concentrate on the experimental situation in which the two-level atoms are prepared in an eigenstate (dressed state) of the coupled atom-resonant field system prior to exposure to the perturber field. We experimentally demonstrate perturber-field-induced, optical nutation of the dressed-state populations. We show how nutation of the dressed-state populations manifests itself in the fluorescence signal of the transiently driven atoms. We observe that the detuning of the perturber field from transitions between the dressed states of the atom-resonant field system plays the same role in determining the character of the dressed-state optical nutation as the atom-field detuning in the case of the ordinary optical nutation of atomic populations. Our experiments demonstrate the utility of the dressed-atom picture in understanding the complex dynamics of bichromatically driven atoms in a variety of circumstances.

We have calculated the long-range part of the momentum diffusion coefficient of two atoms in a radiation field using an S-matrix expansion to first order in the dipole-dipole interaction between the atoms. This perturbative treatment is limited to the low-saturation regime and large detunings (‖δ‖≫γ). The physical processes arise from the interaction with the coherent and incoherent parts of the scattering spectrum. We find that in the regime of a constant density the extra diffusion term when averaged over the trap volume depends on the total number of trapped atoms to the third power and is independent of the atom laser detuning.

We present a theoretical treatment of the problem of a Rydberg atom near-resonantly coupled to a mode in a microwave cavity. We analyze the properties of cavity-induced spontaneous emission in the case of the cavity being driven by one or two near-resonant microwave fields. We find that the spontaneous emission of the atom can be turned off by one strong field (dynamical suppression) and that the addition of a second strong field can break this dynamical suppression and allow the atom to fluoresce in the presence of both fields. The analysis is cast both in a full master-equation treatment and in a Floquet treatment of an atom driven bichromatically in vacuum, the latter allowing for an interpretation of the full calculation. The calculations are performed in the intermediately strong coupling limit, in which both atomic and cavity dynamics must be considered.

We present an investigation of an alternative mechanism for sub-Doppler cooling of atoms, based on a coherent population transfer in three-level Λ systems. The mechanism considered is that of a Λ atom interacting with two standing light waves with a mutual spatial phase shift cphi≠0. The spatial dependence of the level populations of the Λ atom for different values of cphi is presented. For cphi≠0, this clearly demonstrates coherent population transfer in an atom with transverse motion along the space-shifted nodes and antinodes of the two standing waves. We show that this allows translational temperatures well below the Doppler limit ${\mathit{T}}_{\mathit{D}}$=${\mathrm{ħ}}_{\gamma }$/${\mathit{k}}_{\mathit{B}}$ to be achieved.

We give a full quantum treatment of the radiative heating mechanism which arises in atomic clouds at low temperatures due to the collisions in the presence of a light field. The time scale of these collisions is comparable to the time scale of spontaneous decay, hence one needs to combine excitation and decay processes with the quantum motion of the colliding atoms into a single model, which we have done. Our study produces a statistical approximate to the full density matrix by combining wave-packet dynamics with the Monte Carlo state-vector method. We also present a quantitative semiclassical treatment based on the Landau-Zener model, and show that it agrees with the quantal results in certain parameter regimes. However, for low temperatures and strong laser fields the full quantum treatment is needed. We describe our numerical procedures and the statistical aspects of the Monte Carlo state-vector method in detail. Our main result is that the low-energy collisions can produce appreciable heating of the cloud.

We present analytical predictions based on a semiclassical theory concerning a single-mode cw laser interaction with longitudinally Zeeman tuned J=1–J=0 atomic systems. In the theoretical model, the laser frequency detuning from the inhomogeneously broadened absorber line center is considered. Equation for the interaction signal are derived for all rotations of a polarizating analyzer transmisson axis with respect to an original laser beam linear polarization. Results are presented for the forward-scattering case when the analyzer transmission axis is orthogonal to the beam polarization. Calculated results on bright and dark ring-shaped light structures are compared with experimental ones originating from a Gaussian beam interaction with a neon 1${\mathit{s}}_4$(J=1, g=1.464)–2${\mathit{p}}_3$(J=0) transition. Calculated results on forward-scattering spectra are presented, i.e., the forward-scattering light intensity as a function of the longitudinal magnetic tuning when the laser frequency is fixed to selected detunings from the absorber line center. Spectra are presented both for the case when the inhomogeneously broadened line originates from a single absorbing isotope and for the case when it originates from overlapping contributions of two isotopes. The results show that laser frequency detunings of tens of megahertz from the absorber line center(s) show up in the forward-scattering spectra. The conclusion is that reliable data on atomic parameters of a nonlinearly absorbing sample can be extracted from experimental spectra only when the sample is monoisotopic and when the laser frequency detuning from the absorber line center is known.

High-order harmonic emission by one electron in a laser field bound to a zero-range potential is extensively discussed. The model yields an expression for the emission rates in the form of a one-dimensional integral that has to be calculated numerically. The solution is based on the quasienergy wave function of the ground state. The approach is very significantly facilitated by suppressing the harmonic components of the wave function at the position of the zero-range potential. This approximation is found to be very accurate except for the third harmonic. In spite of the simplicity of the model, the harmonic spectrum exhibits a very involved structure, occasional harmonics being strongly suppressed, with cusps and spikes for certain evenly spaced intensities. The latter are due to channel closings for the same intensities in above-threshold ionization. The harmonics near and beyond the cutoff of the plateau are amenable to a completely analytical approximation. This approximation shows how the classical model of Krause, Schafer, and Kulander [Phys. Rev. Lett. 68, 3535 (1992)] is embedded in a fully-quantum-mechanical description. Results are also given for the harmonic production rates in an elliptically polarized laser field; they display fair agreement with recent measurements. The model should adequately describe harmonic emission by negative ions with just one bound s state. Moreover, it also gives a fair description of harmonic emission by an atom, particularly if the ground-state energy of the zero-range potential is adjusted not to the binding energy of the atom, but rather to the energy difference between the ground state and the first excited state. The reason why this is appropriate is found in lowest-order perturbation theory, which sheds some light on the physical origin of the plateau.

We describe a numerical method for solving the problem of the interaction of a monoenergetic atomic beam represented by a plane wave with a classical light field at glancing incidence. In the approximation where spontaneous emission is simply treated as a loss, this leads to a vector Schrödinger equation which is solved using a matrix technique. This allows the analysis of arbitrary optical intensity profiles rather than the rectangular profile previously considered by Zhang and Walls [Phys. Rev. Lett. 68, 3287 (1992)]. Selected numerical results show that the method describes a variety of phenomena including state-selective reflection and velocity-tuned (doppleron) resonances.

We investigate the separated-oscillating-fields method when there are nonuniform conditions during the initial and final periods. We obtain an analytical expression for the particle transition probability between two states as a function of the initial and final perturbation frequencies and also of the Larmor precession frequencies. A sharp asymmetry of the transition probability versus the initial and final oscillating-field frequencies is shown. These results may be helpful for the magnetic-resonance experiments with separated oscillating fields.

We provide a rigorous lower bound for the ionization probability of a Rydberg atom under the perturbation by a time-dependent electric field in the form of an ultrashort pulse idealized as a Dirac δ function. This estimate is of the form 1-O(${\mathit{F}}_0^{\mathrm{-}4}$) for large values of the electric field ${\mathit{F}}_0$. For the hydrogen atom we also prove the scaling behavior ${\mathit{F}}_0$∼${\mathit{n}}^{\mathrm{-}1}$, predicted recently by Reinhold et al. [Phys. Rev. Lett. 70, 4026 (1993); J. Phys. B 26, L659 (1993)], for the electric field required to ionize a given percentage of states with principal quantum number n.

We consider radiation trapping in a plane-parallel slab with specularly reflecting boundaries. The reflection coefficient of the walls can be different at the two sides of the slab, and may depend on the angle and frequency of the radiation. We derive a closed-form Green’s function, and give an efficient method for its evaluation. We show that, in most cases, higher-order modes have practically no influence on the temporal behavior of the emergent radiation when the reflection coefficients at both walls are high, but that they are important when only one wall is highly reflecting. We also investigate the difference between diffuse and specular reflection. For center-of-line opacities ${\mathit{k}}_0$L exceeding 0.5, the difference in the lowest-order trapping factor is quite small (less than 3% for Doppler line shapes and less than 6% for Lorentzian shapes). Trapping calculations can therefore be simplified by using the formulas for specular instead of diffuse reflection.

We study internal selective reflection from a dielectric-vapor interface, where three levels of the vapor atoms in a Λ configuration are connected by two light frequencies. The reflectivity contains then both a sub-Doppler structure due to spatial dispersion in the boundary layer and the dark resonance arising from coherent population trapping. We derive analytical expressions for the reflectivity of either one of the two light beams. The contributions from atoms approaching or leaving the interface are strongly different, whereas they are identical for the case of two-state atoms.

Multiphoton single and double ionization of magnesium was studied by measuring electron energy spectra and ion mass spectra using 1-ps laser pulses in the 580–595-nm wavelength and ${10}^{12}$–${10}^{13}$-W/${\mathrm{cm}}^2$ intensity range. In single ionization the (3p${)}^2$ ${}_{}{}^1{}_{}{}^{}$S doubly excited autoionizing state, resonant at the four-photon level, is found to play an important role. Single ionization leaving the ${\mathrm{Mg}}^{+}$ ion in the 3p excited state is strongly enhanced when resonant with the (3p${)}^2$ ${}_{}{}^1{}_{}{}^{}$S state. The amount of above threshold ionization observation also varies strongly in the wavelength range under study. At an intensity of 2×${10}^{13}$ W/${\mathrm{cm}}^2$, sequential double ionization is found to occur dominantly via the excited ionic state ${\mathrm{Mg}}^{+}$(3p). The first ionization step populates this state less than the ion ground state ${\mathrm{Mg}}^{+}$(3s), but this is compensated by the lower order of the second ionization step.

When exposed to intense light of ∼580 nm, the ground state of K shifts up in energy, passing through two photon resonances with Rydberg states, and finally crossing the two-photon ionization limit. We have used laser pulses of varying duration to study the nature of the population transfer from the ground state to the excited state due to the intensity-tuned resonances encountered during the rising edge of the pulse. A dynamic Floquet approach in which the resonances are treated as avoided crossings of the Floquet energy levels is used to model the population transfer and gives excellent agreement with the data. The model is extended into the strong-coupling regime where the ground state interacts with many excited states simultaneously, and we show that this model can be used to describe multiphoton ionization as a series of avoided crossings with the continuum.

We have performed an experimental study of the atomic system ${\mathrm{Rb}}^{\mathrm{-}}$, for which the ground state 5${\mathit{s}}^2$ is coupled to the autoionizing state 5${\mathit{p}}_{1/2}$6s by means of a strong resonant laser field. The investigations are based on irradiation of ${\mathrm{Rb}}^{\mathrm{-}}$ with intense (I≤1.5×${10}^{10}$ W/${\mathrm{cm}}^2$) laser pulses of duration ∼1 ns. Single-photon detachment of ${\mathrm{Rb}}^{\mathrm{-}}$ is suppressed due to the Fano interference between transitions directly to the continuum and transitions via the autoionizing state. In contrast, two-photon absorption is at the same time resonantly enhanced since no such destructive interference is present in this process. These observations constitute experimental evidence of general features predicted from theoretical considerations of autoionizing states in strong laser fields. Furthermore, we discuss the role of autoionizing states in multiphoton double-ionization processes. Electron spectroscopy has been used to show that double ionization of ${\mathrm{Rb}}^{\mathrm{-}}$ primarily is the result of a sequential removal of the two electrons via an excited Rb state. The perspectives for direct two-electron ionization in negative ions are briefly discussed.

We consider a degenerate optical parametric oscillator driven by a stationary field of frequency 2Ω and by a modulated field of frequency Ω with a small modulation (signal) of frequency ω’≪Ω. By using the eigenvectors of an appropriate matrix we identify the conditions to obtain maximum amplification when a single quadrature component is modulated; the amplification coefficient is calculated analytically as a function of the frequency ω’ and of the parameters. Using a method of quantum Langevin equations, we express in analytic form the noise spectrum for a generic quadrature component of the field of frequency Ω. By combining the results concerning amplification and noise we compare the signal-to-noise ratio in the output with that in the input, and determine the range of parameters in which the performance of the system is better than that of the phase-insensitive linear amplifier. In particular we identify the cases in which noiseless amplification is possible. The approach developed in this paper can be applied to a generic cavity-based phase-sensitive amplifier.

A correlation experiment together with previous experimental information supports the existence of a very disordered state in a high-Fresnel-number dye laser that was studied. All the experimental evidence points out that the disordered state can be indentified with weak turbulence, allowing for the description of real turbulence (not only chaos) in a real laser device. The experimental results are analyzed on the basis of a discrete model derived from the Maxwell-Bloch laser equations. The model explains the parameter-independent frequency locking, the power-law dependence of the averaged energy fluctuations on the averaging region, the local decorrelation, and the chaos found in the experiments. Other characteristics of the model equations are also reported.