DX-centers in CdTe and ZnTe Observed by Locally Sensitive Probe Atoms
- PDF / 148,052 Bytes
- 6 Pages / 612 x 792 pts (letter) Page_size
- 48 Downloads / 200 Views
B1.3.1
DX-centers in CdTe and ZnTe observed by locally sensitive probe atoms S. Lany, H. Wolf and Th. Wichert Technische Physik, Universität des Saarlandes, 66041 Saarbrücken, Germany ABSTRACT Heavily In-doped CdTe and ZnTe crystals are investigated using perturbed γγ-angular correlation experiments. By means of characteristic electric field gradients, it is shown that in the Cd (Zn) poor regime, the compensation is governed by A-center formation. After annealing in Cd (Zn) rich atmosphere, the formation of DX-centers is observed and identified as the prevailing compensation mechanism in the Cd (Zn) rich regime. INTRODUCTION In the II-VI semiconductors CdTe and ZnTe, n-type doping is subject to the well-known phenomenon of self-compensation: While electron concentrations in the 1018 cm−3 range have been reported in CdTe [1], it is difficult to achieve any n-type conductivity in ZnTe [2]. The present work focuses on the microscopic origin of the self-compensation for CdTe and ZnTe doped with Indium donors. The formation of two different compensating centers has been proposed in literature in order to explain the limited doping efficiency. The pair formation of the donor with an acceptor-like cation vacancy (A-center) has been proposed originally in Ref. [3]. In CdTe, for instance, such InCd-VCd pairs have been observed using various experimental techniques [4, 5, 6]. An alternative explanation for self-compensation is given on the basis of theoretical studies which predict the transition of a shallow donor into a DX-center, i.e. a deep, doubly occupied state induced by a large lattice relaxation [7, 8]. For In-doped CdTe, persistent photoconductivity (PPC) and metastable behavior observed in high-pressure experiments have been related in literature to the DX-center model [9, 10]. In order to identify the local structure of the In dopants, we make use of the fact that the electric field gradient (EFG) is very sensitive to the local environment about the site of a suitable probe atom. The EFG is conveniently characterized by Vzz, the strongest component of the traceless EFG tensor (|Vxx| ≤ | Vyy| ≤ |Vzz|), and the asymmetry parameter η = (Vxx−Vyy) / Vzz. Because of the condition Vxx + Vyy + Vzz = 0 for a traceless tensor, the EFG vanishes at lattice sites with cubic symmetry. In the cubic zincblende structure of CdTe and ZnTe, any non-zero EFG indicates a local disturbance of the semiconductor lattice, e.g. caused by a defect. The EFG is not only sensitive to impurity atoms and native defects close to the probe atom, but also to the actual strength of the atomic relaxation around the probe [11]. Therefore, an identification of both the A-center and the DX-center is possible by means of characteristic EFG. In order to measure the EFG, we employ the perturbed γγ-angular correlation (PAC) technique, which has proven to be very useful to investigate defects in semiconductors [12]. Thus, radioactive 111In probe atoms are introduced into the semiconductor crystals. By the time compensation occurs, the 111In atoms are incorp
Data Loading...
