My first job out of grad school was on the Extreme Ultra-Violet Lithography (EUVL) project at Sandia in Livermore California in 2001…
We were developing the technology that would eventually allow chip-makers to make chips with feature sizes as small as is possible to do with silicon… features so small that they are 10′s of atoms across…
The project was funded at an immense level by a large consortium of chip making companies…and was given high priority at Sandia and Lawrence Livermore national labs. It was creating new technology and solving problems that seemed insurmountable.
I was the laser engineer, in charge of integrating the 1500 Watt YAG laser into the lithography machine. Previously the most powerful laser I had any experience with was 30 Watts.
Our team was working on the Xenon path to creating the EUVL light at 13.6 nm wavelength, and the tin path as well. Before I left the project to work at Keck, I repeatedly argued for the two laser solution, using a pre-pulse laser to prepare the target for the High power laser that generated the EUV light… eventually, as shown in the video below, that is the path they went down.
What I worked on most was the Xenon path- we had a cryogenic nozzle that fired a jet of liquid xenon into a high vacuum where it froze into xenon Ice droplets that we then hit with a 1,500 Watt YAG laser beam. This plasma emitted light at 13.6 nm that was collected with first of their kind large multilayer mirrors made up of alternating layers of Molybdenum and silicon- these layers were only 10′s of atoms thick and had to have their thickness controlled to a few atoms.
It was an exciting, fast paced project, with incredible resources… funded at 50 million dollars a year for 5 years.
It took a LONG time to resolve all the technical challenges, but finally, in ~2018, the technology was put into use, making the worlds most sophisticated chips… the machine is one of the most complicated in the world. Firing tin droplets at a rate of 50,000 droplets a second at 80m/s and hitting each droplet precisely with a pre-pulse laser and then main high power laser in mid flight.
This technology was intended to reach the end of Moores Law for silicon chips- if you make feature sizes smaller than what is ultimately possible with this technology, quantum effects start to dominate and the chips will not function as intended… quantum tunnelling between features becomes an issue.
There is now work being done on photonic (light-based) technologies to push computing speed higher- optical computers.
I learned a lot on the project, but I joined it towards the end when they were transitioning from a 40Watt laser to the 1500Watt laser… 9-11 happened, and it became clear that once the EUVL project wrapped up, I would likely be put on anti-terrorism projects- like using laser spectroscopy to identify anthrax in the air, or something along those lines- I was not enthusiastic about these possibilities, so- on a whim- I applied to be the Laser engineer on the Keck Laser Guide Star- and got the job.