Solar cells are one of the most promising alternative power sources, because the sun provides plentiful, renewable energy. In fact, one hour of sunlight provides enough energy to power all human endeavors for one year.1 Bulk silicon cells, which convert between 14 and 17% of incident light into electricity, make up 90% of the solar cell market. Silicon is widely used because it is the second most abundant element in the earth's crust, and because the electronics industry has already developed infrastructure to process it. Yet pricey, complicated manufacturing makes these photovoltaic (PV) systems more expensive per kilowatt-hour than conventional energy sources. These limitations have driven efforts to develop inexpensive solar modules with efficiencies equivalent to, or better than, existing devices.
Figure 1. The current density vs. voltage data shows the photovoltaic effect under simulated AM1.5 light in a large-area silicon nanowire array-based solar cell. The plot verifies that the device is capable of generating power from incident light. The cell generated a photocurrent of 1.6mA/cm2 and a voltage of 130mV. (A typical bulk silicon device has a photocurrent and voltage of more than 20mA/cm2 and 600mV, respectively.) The left inset shows a nanowire array. The right inset is a high-resolution transmission electron micrograph (TEM) of one nanowire coated with hydrogenated amorphous silicon (a-Si:H).
Various thin-film technologies, such as cadmium telluride, copper indium gallium selenide (CIGS), and hydrogenated amorphous silicon based devices, offer reduced costs, but they are hindered by lower module efficiencies.2 Nanotechnology provides another approach to obtaining cheap, high-efficiency devices. Specifically, we have been exploring the use of silicon nanowires to produce new solar cell architectures on low-cost substrates.3
Silicon nanowires offer several performance and manufacturing benefits that may impact future PV applications. By fabricating p-n junctions conformally around the nanowire structure, the absorption of light can be decoupled from minority carrier diffusion. Therefore, minority carriers only have to diffuse tens to hundreds of nanometers to the charge-separating junction, rather than the tens to hundreds of microns typical of conventional solar cells. Furthermore, the subwavelength light-trapping effects in silicon nanowire arrays improve optical absorption properties relative to thin films.4 Finally, scalable processing methods such as chemical vapor deposition (CVD) can be used to grow nanowires on cheap substrates over large areas, potentially reducing the manufacturing cost.
At General Electric Global Research, we recently used CVD to synthesize silicon nanowire arrays on stainless steel foil using a suitable diffusion barrier (see Figure 1). We used a unique, scalable method to fabricate a large-area photoactive p-n junction conformally around the nanowires. We deposited amorphous silicon using plasma-enhanced chemical vapor deposition, a process commonly used in the electronics and solar industries. While the efficiency of the devices is low (∼0.1%), we showed a promising photocurrent density (∼1.6mA/cm2) and broadband quantum efficiency response. The devices also exhibited the expected performance enhancement in optical properties compared to planar thin-film devices. While we focused primarily on metal substrates, similar structures can also be applied to affordable glass substrates.
The use of a metal foil substrate is critical for three reasons. First, it allows for the formation of cells that are conformable and thus suitable for rooftop and building-integrated applications. Second, metal foil substrates are less expensive than conventional single-crystal silicon substrates. Finally, metal foils may enable the use of roll-to-roll-type processes, creating a new paradigm in the manufacturing of crystalline silicon-based solar cells.
The results demonstrate that fabricating silicon nanowire arrays on metal substrates may be a viable method for creating solar cells. These nano-engineered devices have the potential to impact future PV technologies by providing cost-competitive modules with power conversion on par with today's bulk silicon technologies. Our next steps will focus on improving their efficiency by reducing contact resistance, minimizing shunts, optimizing geometry, and improving the p-n junction quality.
The authors would like to thank D. Merfeld, E. Butterfield, T. Feist, G. Trant, and M. L. Blohm for their support of this work. We also thank M. Larsen for the TEM data.
Loucas Tsakalakos, Joleyn Balch, Jody Fronheiser, Bastiaan Korevaar
Micro & Nano Structures Technologies
General Electric - Global Research Center
Oleg Sulima, James Rand
GE Energy - Solar Technologies
4. L. Tsakalakos, J. Balch, J. Fronheiser, M.-Y. Shih, S. F. LeBoeuf, M. Pietrzykowski, P. J. Codella, B. A. Korevaar, O. Sulima, J. Rand, A. D. Kumar, U. Rapol, Strong broadband optical absorption in silicon nanowire films, J. Nanophoton. 1, no. 1, pp. 013552, 2007.doi:10.1117/1.2768999