Punktdefekte in Verbindungshalbleitern untersucht mit der gestörten Winkelkorrelation

This work characterizes the doped semiconductors AlN, GaN, their alloys AlGaN and ZnO. They crystallize in the so called wurtzite structure and are used in many devices already (as light emitting diodes or as transparent contacts). For these applications it is important to know how the characteristics of these crystals change with doping. To do this, the local environment of implanted impurities is studied. Important information is the lattice location of these atoms, how many of them can be introduced to the lattice (solubility). Thus it can be determined if these elements are suited as dopants. This work addresses the possible p-doping of AlN and GaN with Cd. Which is very important in the case of AlN, since no efficient method was found top produce p-type AlN, yet. In addition it was shown, that through doping with In a nitrogen vacancy forms a complex with the dopant, which might have a big influence on the properties of the semiconductor. The other question addressed are dilute magnetic semiconductors (DMS), which open up a new field of applications, the so called spintronic. Hereby not only the properties of the charge carriers are used, but also their spins. This allows for faster switching of semiconductors devices by magnetic fields for example. The PhD-thesis determines, if the rather, in this context, unusual elements Rh and Th are suited as dopants for DMS. Since the already studied transition metals like Mn, Cr or V show no ferromagnetic ordering above room temperature. The main prerequisite for DMS-dopants are unpaired d-electrons, mainly 3d, but also 4d (Rh) or 5d (Ta) are possible. This is studied with the method of the perturbed angular correlation (PAC), where the local environment of radioactive isotopes is measured. It is based on the change of the angular correlation over time between two consecutive $\gamma$-rays due to a hyperfine interaction between these probe atoms and the surrounding charge distribution, more precise its electric field gradient (EFG). This method is also sensitive to the interaction between a magnetic field and the magnetic moment of the probe. Thus structural changes of the crystal, the lattice location of the probes, possible defect complexes that form with the probe atom and magnetic fields can be studied. The results of the thesis show, that Cd is a good candidate for p-type doping of AlN. After annealing of the implantation damage, that occurs when the probe atoms are introduced to the crystal ,Cd occupies a substitutional Al lattice site. This configuration is very stable up to high temperatures of about 900 K. The before mentioned nitrogen vacancy forms no complex with the Cd atoms and seems to be an In related effect. These results are backed by $e^-$ -$\gamma$ correlation and RBS/ Channeling measurements. In the case of the possible ferromagnetic ordering of Rh or Ta in AlN, GaN and ZnO only Ta is suited as a possible candidate since it occupies substitutional metal lattice sites, although no magnetic ordering was observed. This is also true for Rh which was recoil implanted as $^{100}$Pd at the 14UD accelerator of the Australian National University in Canberra. In doing so a $^{12}$C beam hits an enriched $^{92}$Zr target with an energy of 80 MeV and via the reaction $^{92}$Zr($^{12}$C,4n) $^{100}$Pd is produced which is directly implanted into the samples and decays to $^{100}$Rh. Although it was possible to anneal the implantation damage with a thermal treatment, only 30 % of the dopants occupied substitutional lattice sites. This show that Pd(Rh) is not suited as dopant for DMS and on the other hand that it is not a good PAC-probe to study AlN, GaN and ZnO. These results are also backed by RBS/ channeling measurements.