Great lecture series! At 3:08 you mention that there is no limit on the line widths but how narrow can they be made with still a reasonable uniformity? With trimming do you mean a isotropic blanket plasma etch?
The smallest linewidth possible is a matter of control, which generally improves over time. Today, we regularly can print 20 nm features on an 80 nm pitch with 193nm immersion lithography. We generally use an oxygen plasma trim, which is essentially an isotropic plasma etch of the photoresist.
Chris, thank you so much. I can’t tell you how helpful these are. I really want to hear your response to the essay question that you ask at 24:44. Is there an issue for quadruple patterning ? Also what is the acronym “APF” in the Applied materials image of SADP?
Quick answer to the essay question: Yes. Quad patterning was used in mass production of FLASH memory chips. It is also used for Intel's 10-nm node devices.
+mark Heynen It is very, very difficult to see 10 years into the future. Historically, we have a fairly clear idea of what we would do for the next two nodes, but after that we can't see a path forward. This is know as Sturtevant's Law: The end of optical lithography is always 6-7 years away. That said, I think the in the next 10 years we will see a dramatic slowdown of Moore's Law, with advances coming less from lithography and more from materials and architectures. There will continue to be innovations, but not the regular, clockwork reductions in features sizes that have driven Moore's Law till now. In 10 years, we will look more like a traditional, mature high-tech industry.
Thanks for posting these free online
Very clear, understandable expose of LDP
Thank you sir. It's very helpful. As I'm fresher it's really fascinating to know how things work at fab level.
Superb Explanation
thank you very much, i learn a lot from your lecture
Thank you, sir. It's a very good lecture
PDF copies of all the slides in this course are available at:
www.lithoguru.com/scientist/CHE323/course.html
Great lecture series! At 3:08 you mention that there is no limit on the line widths but how narrow can they be made with still a reasonable uniformity? With trimming do you mean a isotropic blanket plasma etch?
The smallest linewidth possible is a matter of control, which generally improves over time. Today, we regularly can print 20 nm features on an 80 nm pitch with 193nm immersion lithography. We generally use an oxygen plasma trim, which is essentially an isotropic plasma etch of the photoresist.
Chris,
thank you so much. I can’t tell you how helpful these are. I really want to hear your response to the essay question that you ask at 24:44. Is there an issue for quadruple patterning ?
Also what is the acronym “APF” in the Applied materials image of SADP?
Quick answer to the essay question: Yes. Quad patterning was used in mass production of FLASH memory chips. It is also used for Intel's 10-nm node devices.
Nicely explained, I've one question: what node do you think we'll be at in 10 years from now 2026
+mark Heynen It is very, very difficult to see 10 years into the future. Historically, we have a fairly clear idea of what we would do for the next two nodes, but after that we can't see a path forward. This is know as Sturtevant's Law: The end of optical lithography is always 6-7 years away. That said, I think the in the next 10 years we will see a dramatic slowdown of Moore's Law, with advances coming less from lithography and more from materials and architectures. There will continue to be innovations, but not the regular, clockwork reductions in features sizes that have driven Moore's Law till now. In 10 years, we will look more like a traditional, mature high-tech industry.
can we use OAI together with PSM?
Yes, and we often do, depending on the type of PSM.