An Investigation into Metallic Impurities in Silicon for Solar Cells

UoM administered thesis: Phd

  • Authors:
  • Simon Leonard


Photovoltaics is an exciting area of research with the potential to completely change the world's energy landscape. Silicon still dominates the photovoltaics market and shows no sign of being overtaken by other materials systems for large scale manufacture. Huge strides have been made in recent years to reduce the cost of solar modules, mainly through the introduction of mass production solar panel plants. However producing very pure single crystalline silicon is still a relatively expensive, energy intensive process. If cheaper less pure silicon could be cast into multi-crystalline ingots, without significant losses to the conversion efficiency this could be a game changer in the photovoltaics industry. For this to happen we need to have greater knowledge and understanding of the role of metallic impurities in solar silicon. If we can find ways to passivate or getter these impurities in cost effective processes that lend themselves to mass production then this would be the key to cost effective solar energy. In the work in this thesis I have investigated some of the most common and most harmful metallic impurities in silicon solar cells using a combination of Deep Level Transient Spectroscopy (DLTS), Capacitance Voltage (CV) measurements, Secondary Ion Mass Spectroscopy and Tunnelling Electron Microscopy (TEM). The specific transition metals I studied were iron, as it is one of the most common impurities and also titanium and molybdenum, because they are very harmful, have slow diffusivities and hard to get rid of with traditional gettering techniques. I have then looked at using hydrogen to electrically passivate these defects, and show evidence that hydrogen passivation is possible for interstitially incorporated titanium in silicon, but is unlikely to happen for interstitially incorporated iron. Another important part of this thesis was the observation and characterisation of molybdenum nano-precipitates in silicon. We have observed the nano-precipitates both electrically in DLTS, and physically in TEM. The precipitates have very interesting electrical properties, and appear to be very strong minority carrier recombination centres, which would have a very negative effect on solar cell performance. It is possible that these nano-precipitates could form from any of the slow diffusing transition metals, and could be a key reason to explain the efficiency gap between low purity cast silicon and high purity single crystal silicon.


Original languageEnglish
Awarding Institution
  • Bruce Hamilton (Supervisor)
  • Philip Dawson (Supervisor)
Award date31 Dec 2015