[Libre-silicon-devel] Maskless lithography
pavel at noa-labs.com
Wed Mar 11 12:30:17 CET 2020
The benefit of DMD is that it's probably physically possible to make
one for short enough wavelength, and then use whatever plasma light
source with it. On the other hand, DMD will probably need a
dramatically different optics that will necessitate somebody PhD level
to work on it. Even in the years of i-line devices, optics was already
gigantic enough, and likely made a big part of the scanner cost.
On Wed, Mar 11, 2020 at 3:15 PM David Lanzendörfer
<leviathan at libresilicon.com> wrote:
> Hi Pavel
> Do you have a contact or so, of those guys?
> I've just moved to Portugal and enjoy this beautiful place with its very very
> nice view everywhere :-)
> (I'd go into the details now about all those beautiful ladies, but this
> mailing list is public xD)
> Maybe you can get us into touch by sending an email to both of us,
> in case you got a mingpian from the folks or so?
> But considering, that they haven't managed to manufacture a working
> prototype, they might not be able to produce anything better, then
> what we could manufacture at INL (The new lab in Portugal) anyway.
> We maybe have to hack something together with a less expensive
> DMD display operated outside the specified ranges...
> There are some micro mirror devices which materials which can also
> reflect UV light, I believe.
> In addition, it would solve the issue with addressing and homogenity
> of the UV light source.
> Am Tuesday, March 10, 2020, 11:45:09 PM WET schrieb Pavel Nikulin:
> > Following on the custom microled device idea. One company I have heard
> > of recently is Hong Kong Beida Jade Bird Display. They still don't
> > have a selling product, and the last time I saw them on an industry
> > event, they were still pitching people around
> > https://www.jb-display.com/projects
> > Maybe they will be open to the idea of a custom device. 435nm is
> > achievable with GaN with which they already work. 435nm is not great,
> > but still better than i-line
> > On Mon, Feb 24, 2020 at 12:43 PM David Lanzendörfer
> > <leviathan at libresilicon.com> wrote:
> > > Hi
> > > So as Staf pointed out, EUV exposure has to be performed at a very very
> > > low
> > > pressure, which is inconvenient to handle.
> > > So I'd prefer DUV or normal UV light.
> > > However, it would be of course fantastic, if we could reach feature sizes
> > > of 50nm or so.
> > > It occured to me, that we will have an e-beam exposure unit available
> > > anyway, at the lab, and that we can deposit all of the materials I've
> > > come across so far, commonly used to build UV LEDs.
> > > Do we wanna design our own (D)UV microLED matrix, maybe?
> > > We have the manufacturing equipment anyway, and it might be a cool
> > > selling point. And it's probably easier to manufacture than MEMs.
> > >
> > > As kind of a side quest :-)
> > >
> > > -lev
> > >
> > > > You can't produce EUV with anything made of solid matter. Even DUV and
> > > > Fluorine lasers (157nm) get absorbed way too enthusiastically. This is
> > > > why Fluorine litho sank on arrival, and the industry stayed on 193nm.
> > > > Switching to 157nm was requiring a change of material technology
> > > > comparable to EUV, but long term gains were not in 157nm's favour.
> > > >
> > > > Both 157nm and EUV can make 25nm feature sizes, but that's only a
> > > > small increment over 30-35 nm what a typical DUV system can do with
> > > > immersion. The industry still went EUV because tech transition with it
> > > > would ease future transition to Xray litho, which would also rely on
> > > > similar resist chemistry and vacuum.
> > _______________________________________________
> > Libresilicon-developers mailing list
> > Libresilicon-developers at list.libresilicon.com
> > https://list.libresilicon.com/mailman/listinfo/libresilicon-developers
More information about the Libresilicon-developers