¶¡ÏãÔ°AV

 

Thesis Defense

Investigation of complex impurity centers in zinc oxide

Abbas Nakhlband, ¶¡ÏãÔ°AV Physics
Location: Online

Thursday, 08 April 2021 02:30PM PDT
Facebook
Twitter
LinkedIn
Reddit
SMS
Email
Copy

Synopsis

Zinc oxide is a potentially useful material in optoelectronics. It has a 3.3 eV direct bandgap which makes it a comparatively inexpensive candidate for replacing gallium nitride as a UV light emitter. Production of stable p-type ZnO is the biggest challenge for the introduction of this material to industry. Many attempts to produce stable p-type ZnO by doping single atom point defects have failed. The remaining dopant candidates are complex defects including two or more components. In this thesis, we mostly focus on the study of bound excitons corresponding to Li related defects. We show that the line at 3353.4 meV in the UV spectra of ZnO, previously assigned to tin, is a complex donor impurity including Sn and Li atoms. We performed annealing experiments, diffusing the Li content out of the lattice. The intensity of the 3353.4 meV line decreased as a result of annealing, which suggests the involvement of Li in this impurity. This process was reversed by the introduction of Li into the sample and the original intensity was restored. This was strong evidence of the involvement of Li in this impurity. We found a 0.4 eV ± 0.2eV activation energy for the ejection of Li from the complex using the annealing results. Density functional analysis of a complex consisting of Li and Sn in the nearest neighboring Zn sites results in much bigger activation energies for different mechanisms of ejection of Li. This evidence suggests that the impurity is more complicated than Li and Sn in nearest neighboring Zn sites. We tried to further investigate this impurity by introducing a different isotope of Li. Substitution of natural Li (96% 7Li) with 6Li showed a change of -0.022±0.008 in the energy position of the recombination energy of the D0X attributed to the line at 3353.4 eV. Our theoretical prediction of the shift suggested a lower limit of -0.031 for the expected shift. In addition, we observed new Li-related lines and showed they are due to bound excitons. We also investigated the effect of the environment in which Li doping is done on the emergence of these new lines.