We report the observation of the anisotropic linear polarization of porous Si photoluminescence measured in two excitation geometries. In the normal excitation geometry (exciting beam normal to the sample (100) surface) linear luminescence polarization of as much as 20% is seen parallel to the excitation polarization. In the edge excitation geometry (exciting light incident on a cleaved edge of the sample) the luminescence polarization is aligned mainly in the [100] direction (normal to the surface). The effect is described within the framework of a dielectric model in which porous Si is considered as an aggregate of slightly deformed, elongated and flattened, dielectric elliptical Si nanocrystals with preferred orientation in the [100] direction.

We observe a quadratic rise of the absorption coefBcient with excitation energy in photoluminescence excitation spectra of porous silicon. Extrapolation to a, = 0 yields an average band gap of microporous silicon about 0.2 eV above the luminescence line. Good agreement is obtained with an estimate of the band gap from the position of the second luminescence line of porous silicon in the infrared spectral region. Further analysis of the line shape using difFerent luminescence detection energies shows that, in addition to the size distribution of crystallites, there exists a second contribution to the linewidth.During the past years there has been an increased interest in the luminescence properties of porous Si (PS). This material has efBcient luminescence in the visible regime even at room temperature. Although many of studies have been devoted to various aspects of PS, it is not yet clear what is the origin of the light emission; this question is still under intense debate.There are numerous models that have been suggested in the literature. Generally they can be classified into four major categories: Radiative recombination is assumed to occur via quantum confined excitons, localized electronic states on the surface of the crystallites, '4 defects in the oxidic coverage of the crystallites, and even within certain Si-based chemical compounds, like siloxene.Canham was the first to ascribe the visible luminescence to quantum confinement of the excited electronhole pair inside the small Si structures, resulting in a luminescence energy well above the bulk Si band gap.It was further argued that the nonradiative recombination is much reduced due to the good surface passivation and the fact that the carriers are confined insid. e the crystallites and cannot diffuse far away to reach a nonradiative center. Calcott et al. 2 have shown that the luminescence at low temperature has a phonon structure; they argued that this proves that both the absorption of the light and the reemission occurs within the Si crystallites. However, there is a variety of experiments that cannot be simply explained by this model. For example, it was shown that the chemical environment affects the luminescence properties.This result suggests that the enlarged surface area of the PS plays a role in the light emission. The existence of localized states on the surface of the Si network has been shown using various measurements. 3 Furthermore, it was demonstrated that the in&ared emission observed &om PS is related to a radiative recombination process, which involves a dangling bond. state at the surface of a crystallite~o '~T he reason for the diKculty in determining the radia-tive processes in PS is related to the highly nonhomogeneous structure of the material. Crystallites having different sizes and shapes, will result in a broad distribution of confinement energies. Therefore, it is not easy to identify where the absorption as well as the related emission takes place. Several attempts were done to translate the PL spectral shape into a size di...

No abstract

No abstract

No abstract