Polarized Luminescence of Defects in CuGaSe 2
- PDF / 786,380 Bytes
- 8 Pages / 612 x 792 pts (letter) Page_size
- 2 Downloads / 215 Views
1012-Y13-01
Polarized Luminescence of Defects in CuGaSe2 Susanne Siebentritt1,2, Sven Augustin2, Niklas Papathanasiou2, Damon Hebert3, Angus Rockett3, J¸rgen Bl‰sing4, and Martha Ch Lux-Steiner2 1 Faculty of Science, University of Luxembourg, 162a, avenue de la FaÔencerie, Luxembourg, L1511, Luxembourg 2 Hahn-Meitner-Institute, Glienicker Str. 100, Berlin, 14109, Germany 3 Dept. of Mat. Science and Engineering, University of Illinois, Urbana, IL, 61801 4 Otto-von-Guericke Universit‰t Magdeburg, Magdeburg, 39016, Germany ABSTRACT The linear polarisation of luminescence light allows conclusions on the symmetry of defects in semiconductors with non-cubic symmetry, like chalcopyrites, for which three shallow acceptors have been identified by photoluminescence. The polarisation dependent photoluminescence allows determination of the symmetry of the defects relative to the c-axis of the crystal. A simple geometrical model implies that chalcogen sites show a predominant direction perpendicular to the c-axis, while metal sites show a predominant direction parallel to the c-axis. Since all three shallow acceptors show polarization parallel to the c-axis, it can be concluded that they are situated on a metal site.
INTRODUCTION Polarisation dependent photoluminescence has been applied in the past for the analysis of the symmetry in molecules or colour centres. [1, 2] In semiconductors it has been used for the analysis of the band structure (see e.g. [3-5]), but hardly for the analysis of point defects. Other methods have been developed to analyse the physico-chemical nature of defects, but they are not necessarily applicable to all semiconductors or defects. In particular native defects are often difficult to control, and methods like extended X-ray absorption fine structure (EXAFS) or perturbed angular correlation spectroscopy (PACS) are not directly applicable. Electron spin resonance (ESR) has been successful in identifying native defects in a number of semiconductors (see e.g. [6] and references therein). But for the whole group of I-III-VI2 chalcopyrite semiconductors no reliable identification of native defects by ESR has been possible so far (see e.g. [7, 8] [9]). But these crystals are accessible by polarisation dependent photoluminescence. The tetragonal structure defines the c-axis as the predominant direction. Polarisation is expected to be parallel or perpendicular to this axis.[1] Previous photoluminescence studies [10] in combination with Hall [11] and cathodoluminescence measurements [12] have shown that the shallow defects in CuGaSe2 consist of three acceptors and one donor. [13] The same defect structure has been found in CuInSe2. [14] In the effort of relating the energy positions of the defects to defect structures one can resort to ab initio calculations.[15-17] However it is not possible to make a direct connection between the calculated defect energies and the measured ones, because for CuGaSe2 as well as for CuInSe2 most of the calculated defect ionisation energies are much higher than observed experim
Data Loading...
