Crystal Technology Overcomes CaF2 Barriers
Aaron Hand, Managing Editor -- Semiconductor International, 12/1/2001
Hot on the heels of solidifying patent coverage of its core technologies, a startup company has revealed crystal growth and purification techniques that have the potential to wipe away most doubts about the readiness of calcium fluoride (CaF2) optics for 157 nm lithography. The industry has determined that CaF2 is the most suitable optical lens material to deal with 157 nm sources, but many still question whether there will be enough quality material to meet the demand for lithography systems.
The uncertainty that remains isn't surprising when current CaF2 growth techniques provide a yield of about 3%. Despite continued announcements from crystal suppliers about factory expansions and increased capacity, the trifling yields that are common do not instill confidence. And, with such yields and their corresponding capital costs, new entrants to the crystal market don't stand a chance. But Single Crystal Technologies LLC (SCT, Gilbert, Ariz.) has found a way to beat the odds, developing crystal growth and purification techniques that promise to increase CaF2 yield to 90% and offer purity levels of at least 99.99999%.
Today's crystal growers are relying on a technique that has changed little in the past 70 years. The Bridgman-Stockbarger method, introduced in the 1920s, depends on good heat conductance within the growing crystal. Because CaF2 is a heat insulator, it has considerable issues with heat dissipation, which ultimately reduces yields. Also, with current methods, crystal sizes are usually limited to 15 in. diameters. SCT's growth technique manages heat dissipation, and enables crystals of any size.
To grow a crystal plate, SCT's technique first purifies a crystal source material. The key to the purification process is that the reactive gas is released at the heat source, at the base of the crystal material. The gas reacts with impurities, causing the impurities to float to the top of the source material. The gaseous products are then removed by vacuum while the heavy products fall to the bottom of the melt. The melt moves away from the heater and the crystal is formed. The process is repeated, refining the crystal, and a dopant is added in a final heating zone. Ultimately, divided portions of the crystal are heat-treated and annealed.
Even if a crystal grower were to use only SCT's purification process, relying on the Bridgman-Stockbarger method for CaF2 growth, yields could be increased by ~3× current levels. Not only would the increased yield of SCT's technology enable crystal growers to increase CaF2 capacity, the process also takes much less time to achieve — two days vs. about 52 days for current methods, according to Ken Schroeder, SCT's president and COO.
The technique is not only relevant to CaF2. In fact, the original work — involving some 20 years of research by Kiril A. Pandelisev, founder, CEO and chief science officer of SCT — was done on other crystal materials. The purification technique could also purify barium fluoride (BaF2), helping to realize theories of ~30% BaF2 in the 157 nm optics mix, Pandelisev noted. SCT plans to apply its technology to other materials in the future.
SCT is working on a prototype system to produce large volumes of ultrapure CaF2 crystals, with the first test batches expected to be delivered next year. However, the company plans to license its technology to other companies with the capital to produce larger volumes.
For additional information on lithography, go to www.semiconductor.net/lithography.
