Radiative properties of free excitons in a single GaAs quantum well are studied under resonant excitation. Enhanced radiative recombination of the excitons, caused by the breakdown of the translational symmetry of the system, is evidenced by the very short lifetime as well as by the strong intensity of the signal. Dephasing mechanisms, by transferring the excitons to nonradiative states, increase the observed lifetime. We deduce a radiative lifetime of 10±4 ps in the absence of dephasing mechanisms.

Electronic states in quantum wires are studied using computer simulations of the growth to characterize the compositional disorder and a tight-binding Hamiltonian to determine the electronic densities of states. Calculations for monolayer slices of wires generated from the statistics of the simulated wires reveal the degrding effect of interface profile fluctuations and islands on the ideal quasi-one-dimensional electronic characteristics. Fixed-width fluctuations of the wire profiles, on the other hand have had a relatively minor effect on the first few subbands. Calculations of the localization length are used to characterize the mobility within the quantum wires.

The magnetic exchange splitting of epitaxial Fe films is determined using inverse photoemission. fcc Fe grown on Cu(100) is found in two phases, one obtained by deposition at 300 K with a splitting of 1.2 eV, and the second with a splitting of 0.9 eV obtained after annealing above 450 K. bcc Fe on Ag(100) exhibits a splitting of 1.8 eV, similar to that of bulk Fe. The magnetic splitting of the 3d states is shown to be correlated linearly with the local magnetic moment, with a slope of 1 eV/${\mathrm{\mu }}_{\mathit{B}}$. This relation holds not only for ferromagnets, but also for antiferromagnets, spin glasses, and the free atoms.

We present a numerical study of a topologically disordered, overdamped array of Josephson junctions subjected to dc and ac currents. We find that vortices are nucleated by the ac current, apart from the ones produced at defects. After a complex transient, these vortices settle into a parity-broken, time-periodic, axisymmetric coherent vortex state characterized by rows of vortices lying along a tilted axis. This locked-in state leads to giant half-integer pseudo-steps in the I-V characteristics that are hysteretic in nature. Possible connections of these results to recent experiments are discussed.

We point out that the London model for the reversible magnetization M of high-κ type-II superconductors in the intermediate-field region is quantitatively incorrect, and we discuss the implications on our understanding of the mixed state of type-II superconductors. We also show that the apparently linear dependence of M vs lnH in the intermediate-field region, a behavior that has been observed experimentally, can be obtained from a previously proposed variational model.

At strong magnetic fields, Landau quantization invalidates the semiclassical approximations which underly the Ginzburg-Landau (GL) theory of the mixed states of type-II superconductors. We have solved the microscopic mean-field equations for the case of a two-dimensional electron system in the strong magnetic-field limit. For delta-function attractive interactions there exist n+1 pairing channels in the nth Landau level. For n>0, two channels share the maximum ${\mathit{T}}_{\mathit{c}}$, and the order parameter differs markedly from expectations based on GL theory.

We show that superconductive pairing is possible in the presence of an arbitrarily strong short-range repulsion between electrons if the gap is an odd function of k-${\mathit{k}}_{\mathit{F}}$. We discuss how the behavior of Δ(k-${\mathit{k}}_{\mathit{F}}$) close to the Fermi surface is related to (i) the low-energy tunneling density of states, and (ii) the attractive part of the potential. We show that some features of the tunneling data in high-${\mathit{T}}_{\mathit{c}}$ cuprates, including the fact that the maximum hardly moves when T approaches ${\mathit{T}}_{\mathit{c}}$, are natural consequences of this type of pairing.

Magnetization measurements in field on a ${\mathrm{Bi}}_2$${\mathrm{Sr}}_2$${\mathrm{CaCu}}_2$${\mathrm{O}}_8$ single crystal do not reveal any sharp phase transition between the normal and superconducting states. The M(H) curves for temperatures about ${\mathit{T}}_{\mathit{c}}$ in fields up to 5 T all display diamagnetic behavior. Above the mean-field transition temperature ${\mathit{T}}_{\mathit{c}}^{\mathrm{MF}}$, quasi-two-dimensional diamagnetic fluctuations are responsible. Below ${\mathit{T}}_{\mathit{c}}^{\mathrm{MF}}$ strong deviations from the conventional mean-field (Abrikosov) behavior lead to an apparent ${\mathit{T}}_{\mathit{c}}$(B) inadequacy of the mean-field theory to describe the vortex state below the crossover line ${\mathit{H}}_{\mathit{c}2}$(T).

We present the results of an x-ray photoelectron spectroscopy investigation in the real-time regime of the kinetics of high-temperature oxidation of the Si(100) surface. The dependence of the net concentration of silicon atoms in all oxidation states (i.e., ${\mathrm{Si}}^{1+}$ ${\mathrm{Si}}^{2+}$, ${\mathrm{Si}}^{3+}$, ${\mathrm{Si}}^{4+}$) on oxygen exposure is found to exhibit a ‘‘step’’-like behavior, each ‘‘step’’ corresponding to one oxide layer. The results obtained suggest a mechanism of layer-by-layer growth of initial oxide layers, with oxide-phase formation taking place at the Si-${\mathrm{SiO}}_2$ interface.

We develop a perurbation theory for the asymptotic survival probability of an interacting particle system which can become trapped in an absorbing state. The method is applied to a simple model describing the poisoning of a catalytic surface. Analyais of the resulting series (in powers of the annihilation rate) leads to very precise estimates for the location of the critical point and the order-parameter exponent β. Results for similar models and other expansions are presented briefly.

Observation of negative ions desorbing from an adsorbate-covered metal surface following UV laser irradiation gives direct evidence that the mechanism responsible is charge-transfer photodissociation (CT-PDIS). We have studied photoemission of ions for ${\mathrm{CCl}}_4$/Ag(111)+hν→Cl-(g)+${\mathrm{CCl}}_3$/Ag(111). The CT-PDIS mechanism is supported by comparison of the yield of desorbing ${\mathrm{Cl}}^{\mathrm{-}}$ and photoemitted electrons with the variation in UV photon energy and surface coverage. Charge transfer leading to negative-ion emission is also shown to occur in the absence of photoelectron emission (i.e., hν<φ) for ${\mathrm{CCl}}_4$/Ag(111).

On the basis of recent experimental results obtained using a model biological system, we argue that well-known physical phenomena like wetting and percolation may play a significant role in morphogenetic changes which occur during development. We also present results of scanning microscopy studies on the structure of the system, which support our theoretical conclusions.

We consider wandering of a nonrelativistic particle in a time-dependent random potential in d spatial dimensions. Its root-mean-square displacement from the initial position increases superdiffusively with time t as ${\mathit{t}}^{9/8}$ for d>1, and as ${\mathit{t}}^{6/5}$ in d=1. Its kinetic energy increases as ${\mathit{t}}^{1/2}$ for d>1, and as ${\mathit{t}}^{2/5}$ in d=1. These scaling behaviors hold for both the classical and the corresponding quantum-mechanical problem in continuous space-time and differ from those of lattice models.

We propose a pair of new quasilocal constructions for the four-momentum and mass of the gravitational and matter fields within generic two-surfaces. We show that the momenta are future pointing when the dominant energy condition is satisfied on a spanning three-surface and the two-surface is suitably convex. The new definition gives zero in flat space and the correct results in linearized theory, at spacelike infinity and for round spheres. At null infinity at least one definition gives the Bondi mass. These definitions can be embedded in definitions of angular momentum.

We show that theories which couple the antisymmetric part of a nonsymmetric-tensor gravitational field to the electromagnetic field predict a depolarization of the Zeeman components of spectral lines emitted by extended, magnetically active regions near the limb of the Sun. This new effect is a consequence of the violation of the Einstein equivalence principle by these nonsymmetric alternatives to general relativity. We show that existing solar-physics data limit the extent of such depolarization and imply that the Sun’s antisymmetric charge ${\mathit{l}}_{\mathrm{\odot }}^2$ must be less than (535 km${)}^2$ in Moffat’s theory, the prototypical nonsymmetric theory of gravity.

We compare two previously introduced theoretical definitions of the ${\mathit{Z}}^0$ mass and show that one of them, frequently used, is gauge dependent in O(${\mathrm{\alpha }}^3$) in a wide class of gauges and in O(${\mathrm{\alpha }}^2$) in the complementary class. We then discuss a slightly modified version that circumvents these problems, is endowed with desirable theoretical properties, and can be identified with the mass measured at the CERN ${\mathit{e}}^{+}$${\mathit{e}}^{\mathrm{-}}$ collider LEP. The simple formulation that emerges is applied to illustrate properties of the physical amplitude, related to gauge invariance, and the role played by the two definitions of mass.

Absolute photoionization cross sections for the 6s6p ${}_{}{}^1{}_{}{}^{}$P excited state of Ba have been measured in the threshold region and found to agree substantially with theory, thus resolving previous discrepancies. Use was made of the ‘‘magic’’ angle applied to the relative orientation of the linear polarizations of the exciting and ionizing lasers; at this orientation angle the measured cross section is the same as if the excited state were populated isotropically. This phenomenon, which has general application to electric dipole photoabsorption processes, allowed us to make the measurements in two different ways, with excellent agreement between them.

Specific molecular mechanisms are proposed for associative ionization collisions of ultracold sodium atoms in a hybrid optical trap. When an intense, strongly detuned optical trap laser is on, the ionization rate is modulated by molecular bound-state resonances which are strongly affected by field dressing. When the weak, slightly detuned optical molasses lasers are on to provide cooling, an excitation mechanism which produces the two excited atoms at large internuclear separation in different hyperfine states accounts for the lower observed ionization rate.

Julienne recently predicted a dramatic laser-intensity-dependent modification of the associative ionization (AI) rate in ultracold collisions. We observe such a modification, but with a behavior inconsistent with the originally proposed mechanism. Furthermore, we find resonant structure in the spectrum of AI rate versus laser frequency, showing the importance of molecular bound states in the AI process. These observations are explained in a new theoretical treatment by Julienne and Heather (preceding Letter).

The ion kinetic structure of a planar collisional shock front in a fully ionized plasma is investigated using a new Vlasov-Fokker-Planck code. The effects of ionic viscosity and ionic thermal conduction are found to be much larger than assumed in usual hydrodynamic plasma simulations with classical transport coefficients. This might have consequences on the numerical modeling of inertial-confinement fusion targets.