The Origin Of Light Emission From Porous Silicon
- PDF / 358,458 Bytes
- 6 Pages / 414.72 x 648 pts Page_size
- 23 Downloads / 212 Views
ABSTRACT We report on luminescence hole burning experiments, which prove that radiative recombination between quantum confined states is the only viable model for the mechanism of the light emission from porous silicon. We find that more than 90% of the luminescence originates from quantum confined states inside the Si nanocrystals. INTRODUCTION The luminescence mechanism in porous silicon (PS) has been the subject of an intense debate. Several models for the origin of the photoluminescence (PL) have been proposed [1], including the "pure" quantum confinement (QC) model [2] (radiative transitions occur between quantum confined levels inside the Si nanoparticles), the "smart" QC model [3] (radiative recombination via surface states), Si based luminescing compounds (e.g. siloxane [4]) and light emission from certain defects in Si0 2 (e.g. the non-bridging oxygen hole
vacancy) [5].
The phonon steps observed in the resonant PL spectrum have been interpreted as spectroscopic evidence for the "pure" QC model [6]. This interpretation has been challenged by Rosenbauer et al. [7], who found that the luminescence is much less efficient under resonant excitation than under non-resonant one. The rise in the luminescence efficiency with increasing excitation energy is accompanied by a smearing of the phonon spectral structures. They conclude that under the usual (non-resonant) excitation conditions the light emission processes suggested by Calcott et al. are weak. Instead, they argue that most of the light comes from an unspecified, very efficient luminescence source. This proposed source is excited only above ,-,2 eV, and the emission is subject to a strong Stokes, resulting in a PL peak at 1.6 - 2 eV. In this interpretation, the QC levels inside the Si nanocrystals are responsible for only a minor part of the light emission, that can be observed only for excitation below the ,-,2 eV threshold [7]. The argument of Rosenbauer et al. relies on two points: 1. Since phonon steps are not pronounced at high emission energies, emission at this energy range does not involve phonon-assisted processes. 2. Another source of luminescence is postulated. In this paper we address these two assumptions: We show well-pronounced phonon steps at high emission energies, and using luminescence hole burning, we demonstrate that almost all the PL is emitted from quantum-confined states.
311 Mat. Res. Soc. Symp. Proc. Vol. 486 0 1998 Materials Research Society
RESULTS Phonon onsets The PL spectra under resonant conditions at several excitation energies for two different samples is shown in Fig. 1. For sample A (right panel) the peak of the non-resonant PL is at 1.77 eV (see inset) while for sample B the peak is shifted to the blue (2.2 eV). Although both samples are resonantly excited at the same energies, the phonon steps are much more pronounced in the spectra of B. This is due to the different displacements of the excitation energies relative to the respective PL peak energies. The resonant excitation is more selective with regard to nanocrys
Data Loading...
