Improving Carbon Nanotube-Silicon Solar Cells by Solution Processable Metal Chlorides

Carbon nanotube-silicon (CNT-Si) solar cells have the potential of being an alternative to the expensive crystal Si solar cells; however, there are critical limitations, such as the relatively low efficiency, instability, and lack of systematic improvement techniques. Herein, by investigating 17 solution processable, nonprecious metal chlorides, it is shown that appropriate chlorides can significantly improve the behavior of CNT-Si solar cells individually or as mixtures. The results reveal a close relationship between the standard electrode potential of redox couples and the open-circuit voltage (V-OC) of solar cells, in which high V-OC values are usually generated from chlorides with larger potentials. Detailed studies on SnCl2 and a mixture of FeCl3/ZrCl4 reveal mechanisms including doping of CNTs (increase in work function) and acting as antireflection layers, resulting in a substantial increase of both V-OC and short-circuit current density (J(SC)). As a result, stable power conversion efficiencies of 14.8% and 16.2% in CNT-Si solar cells are achieved by simply spin-coating SnCl2 or ZrCl4/FeCl3, respectively. The family of metal chlorides provide many opportunities for overcoming limitations and developing high-performance CNT-Si solar cells toward practical applications.