Thin Is In, But Not Too Thin!

efficiencies.

Recently there has been a great deal of interest in CdTe and CIGS based photovoltaics with CdTe being the current leader in volume manufacture. However several issues confront these materials if truly large scale deployment is contemplated. This includes toxicity of these materials although many companies claim to have this under control. Another much more difficult issue with these materials is their availability. As Martin Green states, “In success CdTe and CIGS technologies will ultimately guarantee their eventual failure. This will be by pushing Te and In prices beyond the threshold for profitability, as recently with polysilicon prices.”[1]

In view of the above, although “thin is in” a case can be made for not being too thin! This case is based on the fact that silicon wafer thicknesses in today’s high-volume manufacturing technology are much thicker than needed for achieving high energy conversion with the thickness being driven by the difficulty of handling and processing wafers of thicknesses much below about 160 µm. Figure 1[2] shows that the minimum wafer thickness required to achieve the theoretical maximum efficiency from silicon solar cells is ~40 to 50 microns, not the current ~180 microns.

However, in addition to the basic problem of slicing silicon ingots into wafers of these thicknesses and the inevitability of kerf loss, handling and processing such thin silicon wafers are basically impossible tasks. The technology limitation of current wafer-based technology from the perspective of wafer thickness can be seen in Figure 2.

In loose analogy with the continuing reduction in the critical dimensions for integrated circuits, the continuing reduction in wafer thickness has been called the Moore’s Law for PV (Figure 2). In further analogy with Moore’s Law for ICs, where continuing reduction in the CD is slowing down due to fundamental limitations, wafer thickness reductions for PV are also slowing down, due to fundamental limitations. Both technologies need radical departures from conventional scaling.

Crystal Solar is developing such a radical departure from conventional practice for manufacturing solar cells with an absorber (wafer) thickness that is, as Goldilocks would say it, it is neither too thick nor too thin, but just right! This technology is based on fabricating single crystal wafers by depositing silicon from the vapor phase on to appropriately prepared substrates using an epitaxial deposition process. The positioning of this technology is depicted in Figure 3.

This approach enables the fabrication, handling, processing and packaging of very thin (< 50 microns thick) single crystal silicon wafers and solar cells. This will substantially reduce the amount of silicon utilized with a major impact on the overall materials usage in PV manufacturing. When this technology is transitioned into manufacturing, we project direct manufacturing costs well under $1/Wp to be achievable with high-efficiency PV modules with a direct impact on lowering systems costs. This technology will have the disruptive potential to dramatically reduce manufacturing costs as follows:

• Silicon utilization is reduced to about 20 percent of that utilized in current wafer-based technology – ~300 µm (slice + kerf) for current technology versus < 50 µm using epitaxial tech-nology.
• The traditional supply chain – polysilicon production in Siemens reactors, crystal growth or casting, ingot cropping, squaring and wafering – are eliminated by the direct gas (tri-chlorosilane)-to-wafer process involving epitaxy (Figure 4). This reduces process complexity and is substantially more capital-efficient as compared to traditional technology.

Although high-quality, very thin silicon wafers can be produced by this approach, novel and innovative approaches have to be developed for handling, processing and packaging.

At Crystal Solar, such processes have been developed for fabricating very thin solar cells and packaging them for the completed PV module. Figure 5 shows an example of mini modules ~1 ft. X 1 ft. using ~50 micron thick solar cells.

When transitioned into volume manufacturing, this technology is expected to enable the lowest manufacturing costs of all PV technologies with the cost advantages of thin film technologies and the efficiency, reliability and non-toxicity of earth-abundant silicon PV.

References
1. “Price and Supply Constraints on Te and In Photovoltaics,” Martin A. Green, IEEE PV SEC, 2011
2. “Limiting efficiency of Silicon Solar Cells,” Tiedle et al. IEEE Trans. Electron Devices, vol. ED-31, No. 5, May 1984.

About the Author
K.V. Ravi is the chief technology officer of Crystal Solar, where he is responsible for the development of technology for the manufacturing of low-cost silicon wafers, solar cells and modules. His other affiliations have been Applied Materials, Intel, Motorola, Texas Instruments and Mobil Solar Energy Corporation. He has a Ph.D. in materials science from

FPV4_N_WP_NTM_Ravi.pdf