A major goal of the PV industry is to develop high-efficiency, low-cost solar cells and devices. Recently, several approaches have been proposed for approaching the thermodynamic limit of solar energy conversion. It has been shown that the practical efficiency of different approaches can be substantially different, regardless of the fact that the theoretical efficiencies may be similar. A broad approach, combining select components and techniques from different technologies introduces the valuable advantage of expanding the range of possible materials and processes. A consortium with around 20 contributors, led by DuPont and the University of Delaware, with substantial funding from DARPA, aims to create devices that operate at 50% efficiency in production for $1,000 m2. This aggressive goal is realisable because of recent rapid advances in materials processing, non-imaging optics, and solar cell architectures. ANU has recently joined the team, to work on silicon cells for tandem stacks. An innovative optical design, integrated with the cell design, offers the possibility of optimising the optics, the cells, and the circuit topology to produce an ultra- high efficiency device.

High efficiency approaches and limitations

One method for approaching thermodynamic limiting efficiencies, ranging from 68% to 87% depending on the solar spectrum and the degree of concentration used, uses multiple junction tandem devices under concentration. The single largest loss mechanism for any PV technology is spectral mismatch. Photons below the bandgap are not absorbed; and the excess energy of absorbed photons over and above the bandgap is also lost.

The assumptions made for detailed balance analysis include (i) a standard solar spectrum; (ii) a single photon generates a single electron-hole pair; (iii) a single bandgap in the cell; and (iv) constant temperature across the cell. Any ultra-high efficiency approach must circumvent at least one of these four assumptions.

Consideration of the above shows there are five basic approaches to ultra-high efficiency devices, Honsberg (2005):

1. Multiple junction solar cells;

2. Multiple spectrum solar cells, (where the solarspectrum is changed into a different spectrum, within the bounds of conservation of energy);

3. Multiple absorption path solar cells (for example with two photons being absorbed to produce a single electron-hole pair, or two electron-hole pairs produced by the absorption of a single photon);

4. Multiple energy level solar cells, (more than one quasi-Fermi level separation);

5. Multiple temperature solar cells (involving the extraction of energy from variations in lattice or carrier temperature); and

6. AC solar cells, which have the potential for high efficiencies, but require Terahertz devices functioning as rectenna to be developed.

The theoretical efficiency limit for each of the ultra-high efficiency approaches is similar, but the physical mechanisms and practical efficiency limits for each are substantially different. For example, the efficiency of a multi-junction device can be increased by increasing the number of junctions. However, this pre-supposes that high quality materials with the correct band gap, and well-understood technologies, are available. This is not the case, so a practical approach to ultra-high efficiency devices must avoid the requirement for a large number of ideal materials and processes.