Impact of contamination on hydrogenated amorphous silicon thin films & solar cells

This thesis deals with atmospheric contamination and cross contamination of boron (single chamber process) of the intrinsic absorber layer (i-layer) of p.i-n thin-film solar cells based on hydrogenated amorphous silicon. The atmospheric contaminations were introduced by means of intentional leaks. Hereby, the focus is on the influence of contamination species(oxygen & nitrogen), quantity of contamination (leak flow), source of contamination (leaks at chamber wall or in the process gas pipe), and plasma power on the properties of solar cells. Thereby, the minimum requirements for the purity of vacuum and process gas as well as leak conditions of the recipient and gas pipe system have been determined. Additionally, deposition regimes were developed, where the incorporation of impurities is significantly suppressed. For standard processes critical levels of nitrogen and oxygen contamination are determined to be ~ 4×10$^{18}$ cm$^{-3}$ and ~ 2×10$^{19}$ cm$^{-3}$, respectively, for a leak situated at the chamber wall. Above these concentrations the solar cell efficiency deteriorates. In literature,incorporation of oxygen and nitrogen in doping configuration is assumed to be the reason for the cell deterioration. This assumption is supported by additional material studies of contaminated absorber layers done in this work. The difference in critical concentration is due to the higher doping efficiency of nitrogen compared to that for oxygen. Nevertheless, applying an air leak the critical concentrations of O and N are reached almost simultaneously since the incorporation probability of oxygen is about one order of magnitude higher compared to that for nitrogen. Applying a leak in the process gas pipe the critical oxygen contamination level increases to ~ 2×10$^{20}$ cm $^{-}$3 whereas the critical nitrogen level remains unchanged compared to a chamber wall leak. Applying a deposition regime with a very high flow of process gases for oxygen the higher critical contamination level is found as well for a chamber wall leak. Possible reasons for the dependence of the critical oxygen contamination on the deposition regime and the oxygen source may be the following. Before entering the plasma zone the oxygen can react with hydrogen atoms, silane (very high reaction probability), or silane radicals. Mainly, three reaction products may result, namely SiO$_{2}$ clusters, siloxane (SiH$_{3}$-O-SiH$_{3}$), and OH / H$_{2}$O molecules.The SiO$_{2}$ is assumed to drop to the bottom of the reactor or to be pumped out. If oxygen is incorporated from siloxane, it is assumed to be electrically inactive. In contrast, oxygen incorporated from hydrogen oxygen molecules leads to oxygen incorporation in an electrically active configuration. For a source gas pipe leak, the density of silane in the gas pipe is high compared to that for the chamber wall leak. For the chamber wall leak the hydrogen density is assumed to be much higher compared to that for the process gas pipe leak. Thus, the critical oxygen concentration is suggested to increases due to the resulting higher siloxane to OH / H$_{2}$O ratio for the gas pipe leak compared to that in the vicinity of the chamber wall leak. The different critical levels [...