Since their inception in 2007, nanotube-silicon heterojunction solar cells have experienced rapid improvement due to the diligent work of several research groups. These devices have quickly reached a point where they might begin to possibly compete with current well-established silicon solar technologies; however the development of industrial-scale nanotube synthesis and purification capabilities remains problematic. Although there has been significant recent progress in improving performance, the precise classification of nanotube-silicon heterojunctions has remained ambiguous. In this thesis, I use type, chirality and length purified single-wall carbon nanotubes to clarify the nature of this particular class of solar cell. The junctions that I assembled were made from freestanding nanotube sheets that showed remarkable stability in response to repeated crumpling and folding during fluid processing, which suggests that the films could be well suited to flexible device platforms. Despite modest ideality factors, the best diodes created in this study met or exceeded state-of-the-art device characteristics, but with a surprising lack of any significant dependence on sample type. The data further suggest that these devices might be simultaneously categorized as either Schottky or p-n junctions. More importantly, the results of this study demonstrate the manner in which band-gap engineering can optimize these devices while emphasizing the important role of the junction morphology.