Strain Analysis in PbTe/PbMnTe Superlattices by X-Ray Diffraction

  • PDF / 296,085 Bytes
  • 6 Pages / 420.48 x 639 pts Page_size
  • 104 Downloads / 251 Views

DOWNLOAD

REPORT


STRAIN ANALYSIS IN PbTe/PbMnTe SUPERLATTICES BY X-RAY DIFFRACTION

E.P. KOPPENSTEINER, H. KRENN, N. FRANK AND G. BAUER fdr Halbleiterphysik, A-4040 Linz, Austria Institut

ABSTRACT The strain tensor components of the PbTe and PbMnTe constituent layers of three superlattices grown on (111) oriented BaF 2 substrates are determined from an X-ray analysis. Symmetric (222) and asymmetric (426), (224), (244) Bragg reflections were analysed and a generalization of the Hornstra-Bartels model is used, assuming that the SL (superlattice) can be simulated accordingly. Since in the multilayers also the satellites of the symmetric and asymmetric reflections have been considered the strain tensor components were determined with an accuracy of 1.5x10-4. Consequently the Mn-content of the PbMnTe layers could be determined with an. accuracy of ± 0.03% INTRODUCTION IV-VI compound dilute magnetic semiconductor quantum wells are of interest because of their potential use for mid-infrared laser applications. Their band gaps can be tuned by the incorporation of either manganese or europium (1]. External magnetic fields influence the gaps as well as the band offsets

[2].

In PbMnTe [3], which crystallizes for small Mn contents in the NaCl structure, the decreases of the lattice constant with increasing Mn content is given by a(Pb1 -xMnxTe) = a(PbTe) 0.00491 x(%) [A]. In this paper we report on X-ray diffraction using CuKa 1 radiation in 0-20 scans. The samples investigated were grown by MBE on cleaved (111) BaF 2 substrates, depositing first a PbTe buffer layer followed by 18 to 30 bilayers of certain periods. Buffer thicknesses, layer thicknesses and the number of periods are listed in Table I for the samples SL142, SL144 and SL161, respectively. The growth conditions were such that for sample SL142 a Mn-content of 5%, for sample SL144 a content of 3% and for sample SL161 a content of 2% were expected within the PbMnTe layers. However, due to the lattice mismatch between the layers constituteing the SL and the mismatch with respect to the substrate, for the determination of the real Mn-content a complete strain analysis is necessary, which also gives information about the state of relaxation within the buffer layer and so about the remaining influence coming from the substrate [4,8,9]. Using symmetric and asymmetric reflections the elements of the strain tensor of both the PbTe and PbMnTe layers can be determined following a generalization of the original Hornstra-Bartels [5] model by B.Ortner [6] under the following assumptions: (i) All PbTe layers including the buffer have the same strain

Mat. Res. Soc. Symp. Proc. Vol. 239. @1992 Materials Research Society

522

tensor components. (ii) All the PbMnTe layers have the same Mn content and the same strain tensor components. (iii) The elastic constants of PbTe and of PbMnTe are assumed reasonable •or x = 2-5 %) and to be the same (which is 6 N/mm 2 , c12=0.77x10 following values5 were used: c 1 1=10.8x10 N/mmz, c 4 4=1.3x10 N/mmz [7]. assumed to be (iv) The BaF 2 substrate (0.5mm thick