Traditional solar cells with a single band gap utilize less than half the energy in the solar spectrum. Energy is lost because low-energy photons are not absorbed, and high-photons are absorbed inefficiently. These deficiencies – intrinsic to traditional semiconductors – limit the maximum potential solar cell efficiency. One way to circumvent this limitation is with materials containing an energy band that creates electricity by absorbing two low-energy photons – rather than one, as in a traditional solar cell. This absorption process enables more efficient use of high- and low-energy photons, and such absorbers are referred to as intermediate band (IB) materials. IB materials increase the maximum theoretical efficiency of PV devices from 34% to 63% . We are working hard to make such materials a reality.
Adding elements that generate deep electronic levels in a semiconductor absorber material will generally reduce the performance of a solar cell via Shockley-Reed Hall (SRH) recombination. However, adding impurities at concentrations exceeding their equilibrium solubility limit, a process we call hyperdoping, might suppress the detrimental SRH recombination. Our work focuses on unlocking the fundamental physics surrounding the transport and optical properties of these hyperdoped materials.
To understand the fundamental physics of these new material systems, we have developed a new spectroscopy technique to enable study of both the electronic band structure and the carrier dynamics. This technique is particularly suitable for novel IB materials because it is a purely optical technique that requires neither contacts nor a junction, facilitating a rapid synthesis and characterization feedback cycle. We also use established synchrotron-based techniques to understand how deep-level dopants change the band structure of the host material. With x-ray emission spectroscopy (XES), we have identified changes in the band structure upon the formation of a intermediate band . Using extended x-ray absorption fine structure spectroscopy (EXAFS), we have verified that these metastable materials undergo a phase change during low temperature anneals (500K) that de-activates the sub-band gap absorption. However, this meta-stable, highly-absorbing state can be re-activated by annealing at temperatures near the melting point and then rapidly quenching the sample . Synthesizing these findings, we have developed and are testing a framework to select candidate dopants for intermediate bands in silicon.
 Luque and Martí. Phys. Rev. Lett. (1997) vol. 78 (26) pp. 5014-5017
 Sullivan et al. Appl. Phys. Lett. Vol. 99 p142102 (2011)
 B. K. Newman et al. in Proceedings of the 24th European Photovoltaic Solar Energy Conference, Hamburg, Germany 2009, p. 236.
The efficiency limits of solar cells with a single absorber material can also be exceeded by splitting the sunlight into spectral bands, and shining each of these on a different semiconductor. This idea was originally proposed in 1954  but has been limited by the expense of the optics, tracking systems, and materials involved. We are taking a different approach, collaborating with an optics group at Lincoln Laboratory on a spectral splitter that is inexpensive, efficient, and requires only single-axis tracking. Furthermore, we plan to use only material systems that are Earth-abundant and inexpensive. Our initial modeling has indicated that a tandem structure using silicon and cuprous oxide (Cu2O) could be up to 44% efficient versus a maximum efficiency of 34% for silicon alone.
 E.D. Jackson. “Areas for improvement of the semiconductor solar energy converter.” In Transactions of the conference on the use of solar energy. University of Arizona Press: Tucson, Arizona, 1955; 122.
Nature makes extensive use of solar power. In the process of photosynthesis, plants use solar power to split water, combining the products with CO2 to produce a sugar-like fuel that can be stored for use when the sun is not shining. This basic concept is appealing to scientists and engineers, as even the most efficient solar panel cannot provide power when the sun goes down.
Creating an “artificial leaf” that splits water at efficiencies and costs that enable widespread deployment requires two different technologies: a power source (to provide the energy required to split water) and a catalyst (a compound that acts as a sort of chemical lubricant, allowing the water-splitting reaction to proceed more easily). Ideally, we would like to mimic nature with an artificial leaf in which we dunk a solar cell in water, shine light on it, and produce a hydrogen-based fuel. However, the solutions that allow the water-splitting reaction to proceed efficiently tend to be highly corrosive to solar cell materials.
The PV Lab is working to solve this problem for silicon-based solar cells. Silicon solar cells are appealing because they are the most-widely used and best-understood solar technology. Using new catalysts developed by the Nocera group in the Chemistry department at MIT and careful engineering of the silicon-catalyst interfaces, we have successfully demonstrated that silicon can be integrated into an efficient water-splitting process . More generally, we have derived a circuit model for this type of photoelectrochemical cell that allows us to analyze the details of device functionality and ascertain ultimate device efficiency limits. We are currently working to advance the record efficiency for a water-splitting device composed entirely of Earth-abundant materials.
 JJH Pijpers, MT Winkler, Y Surendranatha, T Buonassisi, DG Nocera; “Light-induced water oxidation at silicon electrodes functionalized with a cobalt oxygen-evolving catalyst” Proceedings of the National Academy of Sciences 108 (2011).