Structured Silicon Devices as a Potential Means to Cost-Effective Solar Cells
Introduction | Macroporous Silicon | Rod Arrays
Introduction
Concerns about anthropogenic global climate change and national energy security are currently driving intense research into alternatives to fossil fuels. Prof. Nathan Lewis gives an excellent talk covering the current energy situation, the prognosis for global CO2 levels, and possible sources of carbon-free energy. One of the key conclusions of the talk is that solar energy is uniquely suited to provide enough energy to power the entire planet. For example, more energy from the sun strikes the Earth in an hour than all the energy consumed by humans in an entire year. However, harvesting this energy in a cost-effective manner is challenging.
Current planar photovoltaics suffer from an intrinsic coupling of the the purity of the material to its light absorption properties. This is because any efficient solar cell must be thick enough to absorb most of the light, but its base material must also be pure enough so that all of the excited charge carriers can escape to a junction. One possible way to decouple these two properties is to imagine using structured devices, for example, arrays of silicon wires.
My research has focused on both understanding the possible challenges associated with building a rod array photovoltaic device and the initial testing of candidate rod array structures. In both of these tasks, I have made extensive use of photoelectrochemistry. By using an electrolyte containing both dimethylferrocene (Me2Fc) and dimethylferrocenium (Me2Fc+), high barrier-height contact can be made conformally to the rod arrays without the need for diffused p-n junctions. Planar or structured silicon samples are used as the working electrode, and the cell is referenced to a Luggin capillary. This allows me to rapidly test different electrodes to determine their properties as solar cells.
Macroporous Silicon back to top
As a model system for a rod array solar cell, I initially fabricated macroporous silicon and tested it in the photoelectrochemical cell shown above. Because macroporous silicon is still highly structured, there is the possibility of minority carriers being collected at the pore sidewalls. Macropores are fabricated from pure, single crystalline silicon by anodization in HF-containing solutions and compared to equivalent planar samples, so that the only variable is the geometry change. Scanning Electron Microscopy (SEM) of macroporous samples yields images similar to these (left is a top down view and right is a cross-section):
Running these samples in the photoelectrochemical cell gave two very important results for the concept of structured silicon solar cells. First, it was found that the degradation of cell behavior was minimal even with a large enhancement in surface area. This implies that, with proper control of the interface and materials properties, it is possible to build an efficient cell with a structured geometry. Second, we found that some loss of open circuit voltage (Voc) was found that could not be accounted for by surface recombination or other surface effects. After reconsidering the device physics of these structured cells, it became clear that, even with perfect materials and interfaces, a loss of Voc is expected due to the decreased flux of charge carriers through the interface. Although the results with macroporous silicon do not unequivically prove this assertion, they do support the idea. For more information, see this paper:
Maiolo, J. R.; Atwater, H. A.; Lewis, N. S. Macroporous Silicon as a Model for Silicon Wire Array Solar Cells, Journal of Physical Chemistry C 2008, 112, 6194-6201
Rod Arrays back to top
In concert with my work on macroporous silicon, others in my group and the group of Harry Atwater (also at Caltech) were working to produce silicon rod arrays grown by the chemical vapor deposition (CVD) method. They have developed an excellent technique that can produce high-fidelity arrays of vertically aligned, single-crystalline rods starting from gas phase precursors (a—top view, b—side view):
Initial tests on these rod arrays were performed using the same photoelectrochemical cell as was used for the macroporous silicon samples. In these tests, highly doped substrate wafers were used to prevent photocurrent and photovoltage contributions from the single crystalline substrate. The initial results showing photo-response from the rod arrays were published in this paper:
Maiolo, J. R.; Kayes, B. M.; Filler, M. A.; Putnam, M. C.; Kelzenberg, M. D.; Atwater, H. A.; Lewis, N. S. High Aspect Ratio Silicon Wire Array Photoelectrochemical Cells, Journal of the American Chemical Society 2007, 129(41), 12346-12347
Work on producing an efficient device from these rod arrays, and on understanding the materials properties of the produced rods is currently in progress. In particular, studies of rods produced using different catalysts and efforts to deconvolute the behavior of the rods from that of the substrate are currently underway.