The ATP project with Lucent Technologies (formerly AT&T) Bell Laboratories significantly improved the accuracy of precision reflective optics - complex multilayer-coated mirrors - that are critical for extreme ultraviolet (EUV) lithography. EUV, or soft x-ray, technology is one of several possible approaches to advanced lithography for manufacturing chips.

The goal of the project was to discover whether it is possible to create ultrahigh-precision aspherical mirrors that properly reflect EUV wavelengths for use in lithography. This was a high-risk, technically challenging project. ATP cost sharing enabled Lucent to move ahead with a project that otherwise would have been difficult to justify, particularly because so much of the funding would go to collaborators outside the company. Ultimately, the ATP project showed that the technical obstacles were surmountable and that the optics can be manufactured, measured and aligned.

Characterizing the complex shapes of these mirror surfaces with the high level of precision required for EUV lithography was well beyond the state of the art when the ATP project began. Working with Lucent, Tropel developed a specialized interferometer to measure aspheric surface characteristics, a device that it now uses in other applications. Lucent, in collaboration with Brookhaven and Sandia National Laboratories and the University of Wisconsin, developed other techniques required to characterize aspheric mirrors. The project also generated increased understanding of multilayer-coated aspherical optics and optics surface finishing, advanced techniques for multilayer coating of mirrors, improved methods for mirror alignment, and new test equipment.

To see whether this new technology would work, Lucent and its collaborators conducted a two-stage, round-robin test. In the first stage, four subcontractors fabricated prototype mirrors using the knowledge created in the project. Then each subcontractor tested mirrors fabricated by each of the four. The mirrors made by Tinsley Laboratories proved to be dramatically better than any of this type ever seen before.

Commercialization Status

When this project began, it was uncertain whether aspheric mirrors with the high level of accuracy required for EUV lithography could be made. And even if they could, it was not clear whether they could be measured with sufficient accuracy to verify that they met the extreme precision demanded by the specifications. Thus, this high-risk project aimed to find out whether the EUV approach to lithography deserved further consideration or whether the mirrors constituted a "show stopper" technical barrier that could not be surmounted. The project demonstrated that the mirror technical barrier could, indeed, be overcome.

Progress on all the advanced-lithography candidate technologies developed in parallel at industry and government laboratories during the early 1990s. As data accumulated, Lucent decided in 1995-1996 (well after the ATP project ended) to reduce its effort in EUV lithography and focus its attention on another option - scattering with angular limitation projection electron-beam lithography (SCALPEL) - which it deemed more promising. Lucent still monitors developments in all areas of advanced lithography, and substantial work on EUV lithography continues elsewhere, particularly at Lawrence Livermore and Sandia National Laboratories.

In 1996 Intel, AMD and Motorola formed the Extreme Ultraviolet Limited Liability Company to pursue EUV lithography. In September 1997, this consortium and the Virtual National Laboratory (a collaboration of Lawrence Berkeley, Lawrence Livermore and Sandia National Laboratories) agreed to collaborate on the development of EUV lithography. EUV systems would draw on the optics work from the ATP project and related technology developed at the national laboratories. The three chip makers intend to invest about $250 million over three years in the collaboration to determine whether the technology is commercially viable and, if it is, to pursue commercialization via lithography equipment manufacturers.

It is too early to tell whether the EUV or one of the other approaches to lithography will ultimately win in the marketplace. But it is clear that the ATP project has helped the industry understand the technical barriers to one major candidate technology and how to overcome them. The ATP project results are important to this effort because the kind of aspheric mirrors that Tinsley learned to make under contract to Lucent will be a critical component of the EUV lithography equipment.

ATP-Project Benefits Could Be Huge

Benefits have already started accruing to Tinsley, which produced the best aspheric mirrors, and to its customers who use the mirrors. Tinsley attributes much of its recent success to the ATP project, because the company was able to apply the improved manufacturing processes - developed to supply aspheric optics for the project - to all its products. Tinsley's sales have approximately doubled since the ATP project. Furthermore, in just 27 months the value of Tinsley's stock increased 600 percent, indicating the value the market places on the company's enhanced capabilities. Tropel and its customers are also continuing to reap benefits from the interferometer.

If EUV lithography equipment incorporating the new aspheric mirror technology becomes the technology of choice for the next generation of chip-making equipment, the benefits of the ATP project would be far broader. The new technology would have a huge economic impact on the semiconductor industry and generate spillover benefits to companies that use the improved computer chips in a wide variety of products, as well as to consumers who use these products. Even if another lithography approach becomes the technology of choice, benefits to companies like Tinsley and Tropel and to their customers will continue to accrue.

This project illustrates the important fact that a lack of immediate commercialization after an ATP project ends does not mean that the new technology will not eventually be commercialized and yield large benefits.

Information gathered in this project helped Lucent better understand the technical issues related to EUV lithography. Publication of numerous technical papers resulting from the project has advanced the state of the art for everyone in this technical community. And although Lucent later decided to pursue an alternative lithography approach, other companies have incorporated the ATP-funded technology into research and development work that could lead to systems that are commercialized in the future.

Go to other sections of Chapter 5: ELECTRONICS
	Expanding the Number of Light Signals in an Optical Fiber
	Manufacturing Technology for High-Performance Optoelectronic Devices
	Processes for Growing Large, Single Silicon Carbide Crystals
	Harnessing Cheap Diode Lasers to Power a Low-Cost Surgical Laser
	Lowering the Cost and Improving the Quality of Computer Chips
	A Gas Method to "Dry" Clean Computer-Chip Wafers
	Low-Cost Night-Vision Technology
	Large-Scale Diode-Array Laser Technology for X-Ray Lithography
	Using High-Temperature Superconductivity to Improve Cellular Phone Transmission
	Exploiting Alexandrite's Unique Properties for a Less-Expensive, More-Reliable Tunable Laser
	Precision Mirrors for Advanced Lithography
	Joining Several Chips Into One Complex Integrated Circuit
	Computer RAM Chips That Hold Memory When Power Is Off
	A Feedback-Controlled, Metallo-Organic Chemical Vapor Deposition Reactor
	Flat Fluorescent Lamps for Displays