By Vivek Bakshi, EUV Litho, Inc.

EUV Sources remain the key enabler to move EUVL into manufacturing, and we look forward to the upcoming 2015 Source Workshop (November 9-11, 2015, Dublin, Ireland) for the latest developments and status of EUV Source technology. Both high-volume manufacturing (HVM) level and metrology EUV sources are needed for chip manufacturing using EUVL. For HVM sources, power level and availability are needed to generate cost effective throughput.

We expect to hear from the user (Intel) as well as source makers (ASML and Gigaphoton) on the latest performance of these sources. At last summer’s 2015 EUVL Workshop we learned that 80 W sources are in the field, but availability needs to improve. So we look forward to finding out the latest performance results from users and suppliers. We will also have a session of FEL-based EUV sources in this workshop. FEL is currently being explored as a technology option for 1000 W and higher EUV sources.

Metrology sources are equally important as they support mask defect metrology tools. Our agenda includes all current EUV metrology source suppliers as well as most research organizations that are working to develop these types of sources. For metrology sources, brightness, stability and source availability are the key matrix of performance.

2015 EUVL Workshop Update

I usually take a week or two to summarize the workshops that I organize or attend, as I am not a reporter but more of an analyst. However, I did not get to report on the highlights from the EUVL Workshop, as I was out on paternity leave. However, proceedings were made available after the workshop, as always.

Although much interest has been shown over the last year in the high absorbing EUV resists – which can reduce the source power requirements for HVM scanner – we learned in the 2015 EUVL Workshop that they are not ready (and may not be ready for many years to come) for commercial use. Hence, the pressure remains for source power scaling, as it will continue to be primary enabler of increased throughput in the near future.

Earlier this year, KLA-Tencor put the patterned mask defect metrology development on the back burner. We need this mask defect metrology tool for patterned wafers for manufacturing via EUVL. Hopefully, their planned merger with LAM will revitalize this program. In any case, the performance (throughput) of this and other mask defect metrology tools still depends on the metrology sources. We expect to hear about the status of these sources in the 2015 Source Workshop.