[Libre-silicon-devel] Stepper / Litho system ideas
noah.wickel at rwth-aachen.de
Tue Jan 14 01:20:54 CET 2020
didn't want to spam the mailling list with too much personal introduction right away ;) But ofc you should know a little about me if we might be working on a project together!
>how familiar you are with the silicon chip manufacturing process?
Studying electrical engineering i had the chance to go through the whole manufacturing process for 4µm gate length SOI nmos and pmos devices in a guided weekly lab course. It has been about two years since i was active in the lab, but i have some understanding of the general processes: standard clean, dry and wet etching, lithography with positive and negative processes, growing oxide... the bottom line is, i understand most of it but never worked in industry, or using cutting edge tools. And I am 99% sure RWTH's lab does not have a stepper. We used visible/near UV contact lithography in that course.
If I understand correctly, you plan to use double patterning to make 1µm work on a regular projector's pixel grid and a condenser optics?
I guess that is both a very ambitious but at the same time very tempting project. Looking forward to seeing self-made litho steppers pop up in hackspaces ;D
I guess I could best help with electronics and control programming, since that is an area i have some practical experience (microcontrollers as well as desktop). But not nearly as much as most people you'd meet at CCC -- i just think this is a very interesting project and i have some shallow understanding of almost all the puzzle pieces involved.
Von: Libresilicon-developers <libresilicon-developers-bounces at list.libresilicon.com> im Auftrag von Hagen SANKOWSKI <hsank at posteo.de>
Gesendet: Montag, 13. Januar 2020 02:40
An: libresilicon-developers at list.libresilicon.com
Betreff: Re: [Libre-silicon-devel] Reminder - Next Mumble session on Sunday, 12/01/2020, 21.00 HKT
On 1/12/20 9:08 PM, Wickel, Noah wrote:
> Hello everyone,
> i am Noah, M.Sc. in EE / Nanoelectronics from RWTH Aachen University.
> would be very interested participating in building a stepper scanner.
> What is your idea for tackling the problem? It would be helpful to
> understand what kind of expertise is needed in the team
how familiar you are with the silicon chip manufacturing process?
For our 1 um process we should be able to bring well-shaped structures
at least in this size and with good accuracy on the photo resist.
The wavelength of light is between 750 nm (red) and 380 nm (violett).
Older machines on the market are in the 650 down to 500 nm resolution
range (as we used at HKUST).
But for smaller structures, we need better resolutions. Mostly newer
machines using diffraction (and multi-pattering) therefor.
So I guess, the knowledge someone needs here is a mixture of
- photolithography (how this is done on the wafers)
- physics of Light
- optics for scaling down bigger structures in the mask down to the
small structures we need on the resist
- mechanical (and systems) engineering
- electronics (for controlling the machine)
Our issue here is the following: in HKUST we had a stepper which was
feasible. But now looking around to get into other clean rooms, we often
see that there are *no* steppers. This also means, that a good
photolithography for small clean rooms on Universities is not so quite
common as expected.
We could buy and install one in the clean room we like to go into
(assuming we had a budget for that). But other folks which like to use
our LibreSilicon process in their clean room probably facing the same
problem. Hence, finding a solution which closes this gap for others too
would be nice.
Okay, this sounds a little bit scaring, I know. But probably you can set
your footprint into that field and get your job offer from a equipment
supplier if you like.
support LibreSilicon to get back into the Clean Room
Libresilicon-developers mailing list
Libresilicon-developers at list.libresilicon.com
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Libresilicon-developers