Institute of Fundamental Technological Research

The structural, electronic, and magnetic properties of low-energy rhenium implanted c-Si are examined for the first time. The damage created by rhenium ions and the following partial reconstruction of the silicon host matrix after rapid thermal annealing (RTA) are investigated as a function of the fluence. Rutherford backscattering spectrometry (RBS) results reveal that the implanted ions are located in the near-surface region with the distribution maximum at about 23 nm below the surface. The analysis of rhenium-depth distribution using the McChasy code shows that the implanted Re-ions are located in the interstitial lattice positions. The RTA leads to a partial recovery of the silicon crystal structure. According to the RBS results, the formed inclusions are not coherent with the silicon host matrix causing an increase of the lattice distortion. Analysis of channeled RBS/c spectra carried out by the McChasy code revealed different levels of bent channels in damaged regions suggesting bimodal distribution of inclusions in the silicon. Studies of high-resolution X-ray photoelectron spectroscopy (XPS) conducted after the RTA showed the shift of Re 4f7/2 binding energy (BE) by +0.68 and + 0.85 eV with respect to metallic rhenium for the samples with lower/higher fluencies, respectively. Complex XPS, density functional theory (DFT) simulations, and transmission electron microscopy (TEM) data analysis allowed us to conclude that the near-surface layer of the sample (~10 nm) consists of nanoinclusions with cubic and/or hexagonal ReSi. In the middle area of the samples, much larger nanoinclusions (>10/20 nm for higher/lower fluencies, respectively) containing pure metallic rhenium inside are formed. The RTA increases the magnetic moment of the sample with the lower dose nearly 20-fold, whereas in the sample with the higher dose a 3-fold increment is observed only. The magnetic response of the examined systems after the RTA indicates a presence of magnetic interactions between the nanoinclusions resulting in the system exhibiting super-spin glass or super-ferromagnetism.

rhenium-implanted silicon, RBS, XPS, RTA, TEM, DFT

Epitaxy of semiconductors is a process of tremendous importance in applied science and in the optoelectronics industry. The control of defects introduced during epitaxial growth is a key point in manufacturing devices of high efficiency and durability. In this work, it is demonstrated how useful hybrid reflections are for the study of epitaxial structures with anisotropic strain gradients due to patterned substrates. High accuracy in detecting and distinguishing elastic and plastic relaxations is one of the greatest advantages of measuring this type of reflection, as well as the fact that the method can be exploited in a symmetric reflection geometry on a commercial high-resolution diffractometer.

optoelectronics, Group III-nitride semiconductors, epitaxial growth, X-ray multiple diffraction, interface defects

Spatially resolved X-ray diffraction (SRXRD) is used for micro-imaging of strain in GaAs:Si layers grown by liquid phase epitaxial lateral overgrowth (ELO) on SiO2-masked GaAs substrates. We show that laterally overgrown parts of the layers (wings) are tilted towards the underlying mask. By SRXRD mapping local wing tilt is easily distinguished from macroscopic sample curvature. The direction of the tilt and distribution of tilt magnitude across the width of each layer can also be readily determined. This allows measuring of the shape of the lattice planes in individual ELO stripes. Downward wing tilt disappears completely when the mask is removed by selective etching. Then residual strain in ELO layers is exposed. In particular, upward tilt is found in free-standing ELO wings. Numerical simulations show that this phenomenon is caused by different concentrations of silicon dopant in vertically and laterally grown parts of the layer.

It is well known that classical Stoney's formula for radius of thin film bowing fails when thin film is dislocated. Thereby, a method of determination lattice parameters for each of heterostructure epilayers is needed when inelastic relaxation takes place. In this article, we have developed a method of determination lattice parameters of a heterostructure from radius curvature measurement data. Description of deformation when dislocations are nucleated is carefully analyzed. It is shown that in order to take into account nucleated dislocations in thin film an additional term responsible for increased volume of layers should be included in the analysis. The proposed method is based on the finite deformation elasticity and uses the finite element method and bowing radius measurement by the laser beam reflection method. As an example the nitride alloy heterostructure with GaN as a substrate is analyzed. The verification is performed using X-ray measurement of lattice parameters. A very good correspondence between numerically determined lattice parameters and XRD measurements data are observed.

Zinc oxide has wurtzite structure (wz-ZnO) at ambient conditions. Due to the promising bandgap (4.0-7.8eV) we consider the misfit stress for the growth of rock salt rs-Zn$_x$Mg$_{1-x}$O layers on rock salt MgO. At the ambient conditions, a solid solution of ZnO in MgO is stable only up to 13%. Nevertheless, due to the misfit stress the range of chemical composition of thermodynamically stable layers can be extended. We consider a mechanism of the dislocation network formation at the interface rs-Zn$_x$Mg$_{1-x}$O/MgO. Based on the dislocation theory, many different analytic formulas for critical layer thickness have been derived, cf. Hu (1991), Brown (2002). The formulas concern the critical thickness of the layers which retain thermodynamically stable at atmospheric pressure. On the other hand, for thin layers which lose the stability earlier, before the stress relaxation, we can expect a lower critical thickness. We present a derivation of an analytic formula for the critical thickness of rs-Zn$_x$Mg$_{1-x}$O layers which lose the stability due to the rocksalt-wurtzite phase transition, cf. Lu et al. (2016). In the new formula the dependency of the onset elastic energy $E(sigma, x)$ of the rs$ ightarrow$wz phase transition is taken into account. In the general case this energy depends on the misfit stress and chemical composition.