|dc.description.abstract||To create doped titanium dioxide TiO2 thin films for use in intermediate band solar cells, it is important to be able to deposit thin films of high quality. To achieve this, substrates with surfaces of high quality is needed. Often, preparation is needed to get the correct termination on the substrate surface. HF etching is a widely used method for preparation of strontium titanate substrates, but it is both expensive and challenging, so we have tried using a different approach.
STO substrates have been prepared for deposition of TiO2 thin films using three different methods. One used acetone and isopropanol to wipe the substrates, another used heat annealing at 1000 degrees Celsius for an hour before being washed in an ultrasonic deionized water bath for 30 seconds. The third method used the same steps as the second, but repeated them twice (referred to as waterleaching), resulting in well-defined atomic planes. Substrates prepared by waterleaching have been studied before the treatment, straight after the treatment and one, two, three and six weeks after the treatment using atomic force microscopy (AFM). Untreated, newly treated and nine weeks old treated STO substrates were studied using electron microscopy (SEM), energy dispersive X-ray spectroscopy (EDS) and X-ray photoelectron spectroscopy (XPS).
AFM gave STO substrate surface roughness measurements ranging from 61 to 123.5 pm, with a tendency of increasing roughness as time passed. Line scans over the atomic planes showed that the height of each plane was approximately in integer or half integer values of the STO lattice constant, meaning that the substrates were not completely TiO2 terminated even after waterleaching. Still, XPS measurements showed that waterleacing did in fact increase the TiO2 termination at the substrate surface to a limited extent.
TiO2 thin films were deposited on untreated, newly treated and nine weeks old treated STO substrates by both pulsed laser deposition (PLD) and electron beam physical vapour deposition (EBPVD) at room temperature, due to a malfunctioning substrate heater in the PLD. The deposition rates of the PLD deposited thin films were calculated from profilometer measurements of the thin film thickness and were found to range from 4.58 to 6.04pm/pulse. EDS scans of all six thin films showed that the oxygen content in the PLD deposited thin films was a lot higher than in the EBPVD deposited thin films.
As expected, X-ray diffraction showed the TiO2 thin films to have no, or low, crystallinity. XPS measurements of TiO2 thin films deposited on by PLD and EBPVD on newly treated STO substrates and by PLD on a silicon substrate, showed that the thin film composition was more dependent on the deposition method used than on what type of substrate was used. This was further confirmed by EDS-measurements.||