Global Climate & Energy Project
Solar EnergyDesign and Fabrication of the First All-Carbon-Based Solar Cell
Start Date: September 2011
Zhenan Bao, Department of Chemical Engineering, Stanford University
The goal of this research is to develop the first all-carbon-based solar cell. Carbon is an attractive material for photovoltaics because of its low cost and abundance. However, pure semiconducting carbon is difficult to obtain, and there are technical drawbacks with carbon conductors and cathodes. This work will address those limiting issues. The impact of this research and the platform technologies to be developed will have applications beyond solar cells and will contribute to the field of flexible, low-cost electronics as a whole.
BackgroundCarbon nanotubes (CNT) are folded tubes made from a single layer of graphite. Depending on the folding direction relative to the hexagonally arranged carbon atoms, a variety of CNTs with different chiralities is possible (Figure 1). Single-walled carbon nanotubes (SWNTs) are ideal components for electronic devices, because of their superior electronic and mechanical properties and environmental stability. A large percentage of SWNTs are semiconducting, while others exhibit metallic behavior. SWNTs can be readily dispersed in water or organic solvents with the assistance of surfactants or polymers, making them compatible with low-cost, large-area roll-to-roll coating processes.
ApproachDespite their great electronic properties, carbon-based materials have only been tested as individual components of solar cells. For example, carbon nanotubes and graphene are being pursued as potential candidates for the replacement of indium tin oxide (ITO) transparent anodes. ITO is becoming more expensive due its high demand in the display and solar industries, and the low abundance of indium. CNTs and graphene have been incorporated into anodes and organic solar cells. In some cases, improved power conversion efficiency has been reported. However, the exact role of CNTs in these devices is unclear as a mixture of metallic and semiconducting CNTs were often used. Buckyballs, on the other hand, are commonly used as electron acceptors for organic solar cells. To date, no other organic acceptor material has achieved power conversion efficiencies anywhere near that of buckyballs.
The proposed architectures for all-carbon-based solar cells are shown in Figure 2. In the vertical structure (Fig. 2a), a CNT or graphene transparent electrode will be used as the anode. Sunlight will be absorbed by the semiconducting carbon nanotubes (sc-SWNTs) to generate excitons (electron-hole pairs). These excitons will be split at the interface between the CNTs and the carbon 60 (C60) or carbon 70 (C70) acceptor to generate holes and electrons. Some excitons can also form in the acceptor layer, especially with more light-absorbing C70 derivatives. The generated electrons will be transferred from CNTs to the buckyballs, while holes remain in the CNTs. The electrons will flow towards the highly n-doped carbon cathode. Horizontal devices will be fabricated to take advantage of the high-charge transport properties of aligned sc-SWNTs and/or buckyball (C60) microribbons. Similar horizontal architecture has been reported for a single-nanowire solar cell (Briseno et al.).
The research team will develop carbon-based semiconductors, which will also have applications for other semiconductor devices, such as transistors, sensors and photodetectors. Design rules for p- and n-type dopants will also be developed. These dopants are of great importance for both carbon-based and organic solar cells, as well as organic light-emitting diodes and organic electronics. A previously developed sorting method will be used to achieve a higher electronic quality. Confocal Raman Spectroscopy will be used to characterize both the chirality and the electronic type (metallic vs. semiconducting) of the dispersed CNTs. Ultra-violet, visible and near-infra red spectrographs will be used to characterize the absorption of the dispersed SWNTs. Simple transistors will be constructed using the sorted SWNTs to determine if the SWNTS are semiconducting or metallic.