Based on a previously developed polarization component method, a phenomenological model is constructed for the generation of polarized radiation in single-mode semiconductor lasers and applied to polarization switching. This model gives a good description of the experimentally observed features of the polarization process. At the same time, a number of effects related to the polarization switching rate, hysteresis characteristics, and polarization switching during optical injection are interpreted anew in terms of the formation dynamics of the laser radiation.Introduction. Sudden switching in the polarization state of the output of semiconductor injection lasers has been known for quite some time [1,2] and is usually associated with the attainment of conditions such that the differences in the gain and loss coefficients for two orthogonally polarized modes are equal [3]. In end-on laser diodes these conditions are reached either through use of an external pressure [4] or through a temperature change [3]. In many cases the polarization switching effect is bistable [5,6], so it has come into widespread use in the development of various kinds of devices for optoelectronic systems [7].Further interest in this effect was stimulated by the discovery of a "spontaneous" polarization switching effect in vertical-cavity surface-emitting lasers (VCSEL) [8], which placed significant limitations on the use of these systems in polarization sensitive optoelectronic devices. Earlier approaches to the interpretation of the polarization switching effect (dispersion in the gain coefficient, as well as the anisotropy in the active layer induced by internal stresses) [9] were not adequate for describing the various manifestations of this effect and other mechanisms had to be taken into account (e.g., the thermal lens effect [10]). For this reason, the spin-flip model [11,12], which is based on a generalization of the theory of gas lasers, has come into widespread use; it relates the polarization switching effects to a change in the difference in the populations of the sublevels of the conduction band and heavy holes in the region of a quantum well owing to electron spin relaxation. This model gives a completely natural description of the two types of polarization switching (with increasing and decreasing output frequency) in VCSELs and makes it possible to take into account a whole series of specific mechanisms (for example, the influence of the relation between intrinsic and induced anisotropies [13]).The existence, in general, of two different approaches to interpreting polarization switching in VCSELs has made it necessary to search for some generalizations at the level of a microscopic description (the change in the band structure owing to microscopic stresses, elimination of the degeneracy in the energy levels, variations in effective mass, etc.) [14]; however, the numerical calculations for these models are extremely cumbersome [15,16] and the models themselves are unsuitable for practical applications. On the other hand, a sim...

UDC 621.373.8 Various dynamic effects during polarization switching in single-mode semiconductor lasers are examined on the basis of a previously developed approach. This model is found to provide a good description of a wide range of experimentally observed phenomena. At the same time, it offers a simple and intuitive interpretation of the nature and character of polarization switching effects that is related to the formation dynamics of the laser radiation. Introduction.Although the sudden switching of the polarization state of the output from semiconductor injection lasers is an effect that has been known for quite a long time [1,2], only recently have the dynamical aspects of this phenomenon been studied in the case of vertical-cavity surface-emitting lasers (VCSEL) [3][4][5][6]. This interest originates in the possible use of VCSEL in optoelectronic systems, while "spontaneous" polarization switching is undesirable in polarization sensitive systems.Dynamic effects in VCSEL are usually studied in terms of a spin-flip model (SFM) [7,8] which contains two coupled spin subsystems that generate circularly polarized orthogonal components. Linearly polarized modes are formed by matching the phases of these components as a consequence of stability conditions [7]. However, the fluctuations in the output radiation increase rapidly in the polarization switching region, and the phase can no longer be regarded as an adequately defined quantity. Thus, in the polarization switching region, partially polarized dynamic states [9] (an incoherent mix of different polarizations) are observed. These things cast doubt on the validity of the SFM for the dynamics of polarization switching in semiconductor injection lasers.In an earlier paper [10] we studied polarization switching effects in a cw single-mode semiconductor injection laser. The polarization component method (PCM) [11] made it possible to relate all the basic features of polarization switching to the formation dynamics for polarized radiation in the active layer of a laser diode. This paper is an analysis of polarization switching dynamics in single-mode semiconductor lasers based on a model using the polarization component method.Theoretical Model. The basis of the theoretical model is the PCM, in terms of which a plane wave field E(z, t) propagating along the z axis is represented in the form of a superposition of polarization components:

Problem statement: Correct description of the anomalous phenomena determined by self-point defects in implanted silicon desires knowledge of their properties. Interstitial Si atoms themselves display anomalies in their behavior and firstly in existence of two very different values of the diffusion coefficient. Approach: We analyzed experimental results and proposed the model of diffusion of interstitial Si atoms in silicon in two shapes. Results: At low saturation Si atoms diffuse as isolated atoms with a low diffusion coefficient (~10^-12 cm² sec^-1 at 900°C). At high supersaturation interstitial atoms diffused as Si-Si pairs, which had lower activation energy of migration and higher diffusion coefficients (~10^-7 cm2 sec^-1). Conclusion: The high diffusivity pairs were formed when two Si atoms hit in the same interstice. The atoms were not bound to one another by covalent bond. In a pair atoms were retained by a potential relief of the crystal

No abstract

No abstract