Nanocrystalline silicon is an attractive material for solar cells. It has very small grains, about 20 nm, and yet its electronic properties are very similar to those of crystalline silicon. The material exhibits smaller mobilities than crystalline Silicon, but the minority carrier lifetimes are reasonable. It is known that the properties of the material depend critically upon deposition parameters, in particular ,the degree of grain boundary passivation achieved during growth and grain size. Previous work has shown that as the material grows, the grains tend to agglomerate into a cluster, and the development of this cluster leads to poorer electronic properties. The traditional method for overcoming such clustering has been to change the hydrogen to silane dilution ratio as the material grows, keeping the material near its crystalline to amorphous transition zone. However, this method is dependent upon the precise growth chemistry and is not suitable for mass production.

In this project, we develop a new device design, a superlattice comprising alternating layers of amorphous and nanocrystalline silicon, which allows one to precisely control the agglomeration of grains without having to resort to hydrogen profiling techniques. We study structural properties such as grain size and the degree of crystallinity, and electronic properties such as carrier diffusion lengths and defect densities. We show that an appropriate design of the superlattice allows one to minimize defect densities and maximize carrier diffusion lengths. We also study how to reduce series resistance in solar cells, and show that an appropriate combination of superlattice and contacts can lead to devices with high fill factors and good solar cell efficiencies.

We also report on a new discovery, namely that the optical absorption itself depends critically upon grain size. Larger grain sizes, up to 50 nm, lead to increased optical absorption, a totally unexpected and very useful discovery for devices, since higher absorption translates into larger current densities. We show that such grain sizes can be achieved using deposition at higher temperatures. We develop a new technique, post-deposition annealing, to help passivate the grain boundaries in devices prepared at higher temperatures. Without such annealing, the device properties for devices grown at higher temperatures with larger grains are very poor. We show that when the devices are grown with larger grains, and then passivated using this new annealing technique, the current and efficiencies increase by 40% compared to similar devices prepared at lower temperatures.

We also propose the concept of a "Multiphase cell" by alloying germanium with silicon in the back layer of the device. Since germanium has a stronger absorption co-efficient than silicon, we show that we can enhance the infrared spectral response and also achieve good IV curves by optimally grading the germane flow.

Finally, we implement these concepts on textured backreflectors to further enhance the efficiencies by light trapping.