A.I. Lebedev,
I.A. Sluchinskaya, V.N. Demin
Physics Department, Moscow State University, Moscow, 119899, Russia
I.H. Munro, G. van Dorssen
SERC Daresbury Laboratory, Warrington, WA4 4AD, UK
PbTe, SnTe, GeTe and their solid solutions are narrow-gap semiconductors widely used in infrared optoelectronics and for thermoelectric energy conversion. Being doped with different impurities, they exhibit many interesting phenomena such as superconductivity, phase transitions, strong photoconductivity in the far infrared [1]. Despite great technical importance of the photoconductivity that occurs in In-doped crystals at temperatures below 25 K, the nature of this phenomenon is not clear yet.
The present work is a continuation of recent EXAFS studies performed at the NSLS on GeTe, SnTe and PbTe crystals doped with high concentration of indium (10-35 at.%) [2]. Unfortunately, the data obtained in this study were not very useful to interpret our photoelectric data [3] because the experiment was carried out on the samples having the indium concentration too high to be photoconductive and did not consider the effects of deviation from stoichiometry, which strongly influence the physical properties of In-doped samples [3].
EXAFS data were collected at the In K-edge (27.94 keV) at 77 K in fluorescence mode on the station 9.2 equipped with solid-state detector array. The studied samples were PbTe, SnTe and GeTe doped with 1.5-2% of indium along two sections (InTe and In2Te3) of the phase diagrams. EXAFS data were analyzed using FEFF software.
The results of the preliminary data analysis are summarized in Table 1. For comparison, the data obtained for concentrated samples [2] are also included into the table. In the analysis we allowed only one type of atom to enter each shell and revealed tellurium atoms in the first shell and metal atoms in the second shell for all samples.
| Sample | CN | R1, A | sigma12, A2 | R2, A |
|---|---|---|---|---|
| GeTe(2%InTe) | 2.9(2) | 2.896(4) | 0.0051(5) | 4.206(22) |
| GeTe(1%In2Te3) | 3.1(1) | 2.902(4) | 0.0061(4) | 4.210(11) |
| Ge0.9In0.1Te* | 3.7(1) | 2.934(3) | 0.0064(3) | 4.214(5) |
| SnTe(2%InTe) | 3.9(2) | 3.082(6) | 0.0082(8) | 4.448(15) |
| SnTe(1%In2Te3) | 4.2(3) | 3.055(7) | 0.0085(9) | 4.459(19) |
| Sn0.75In0.25Te* | 4.6(2) | 3.114(3) | 0.0074(4) | 4.417(10) |
| PbTe(1.5%InTe) | 3.7(3) | 3.042(9) | 0.0207(21) | 4.569(30) |
| PbTe(0.75%In2Te3) | 3.8(4) | 3.046(15) | 0.0225(38) | 4.508(37) |
| Pb0.75In0.25Te* | 4.0(2) | 3.126(8) | 0.0150(13) | 4.587(13) |
It is seen that the data for diluted and concentrated samples are consistent. The main changes observed for diluted samples are: shortening of interatomic distance R1 and decrease in coordination number CN in the first shell for all three compounds; increase in Debye-Waller factor sigma12 for SnTe and especially for PbTe. Except for the case of SnTe, we did not reveal any difference in obtained parameters for samples with different deviation from stoichiometry. This can indicate that the most of indium atoms probably do not form impurity complexes.
Interatomic distances in the first and second shells revealed by the analysis favour the model with the indium atom entering the metal site, but the reduced coordination number (2.9-4.6 instead of 6 for metal site in NaCl structure) complicates the interpretation. To reconcile these two facts, we suppose that either the first shell consists of two subshells with close interatomic distances (this would correspond to the physical model [4] of two charge states of indium atom in a lattice), or reduction factor S02 for indium atom is unusually small due to easiness of charge redistribution on the In-Te bond for systems under investigation. The further analysis will show which of these suppositions is more realistic.
1. V.I. Kaidanov, Yu.I. Ravich, Sov. Phys.--Usp. 28, 31 (1985).
2. A.I. Lebedev, I.A. Sluchinskaya, V.N. Demin, Q.T. Islam, (unpublished).
3. A.I. Lebedev, Kh.A. Abdullin, Physica Status Solidi(a) 91, 225
(1985).
4. Yu.A. Andreev, K.I. Geiman et al., Fiz. Tech. Poluprov. 9, 1873
(1975).