[labnetwork] Nitrogen Purity for Sputtering

[Robert M. HAMILTON]

Mark Heiden,

The UC Berkeley Marvell NanoLab has looked into house, high-purityN2 versus
UHP cylinder N2 in reactive sputter applications.Using a RGA we see more
contamination with using cylinder N2 than our house-delivered N2. As per
your note, the NanoLab N2 is also derived from a cryogenic LN vessel. The
spec from the LN vendor gives impurities of <1 ppm.

Having said this, N2 is delivery throughout the lab is branched to
countless uses, including things like N2 blowoff guns. There is also
significant use of plastic tubing and retractile hoses, all of which are
subject to permeation. Partial-pressure laws invariably mean the N2
delivery system is subject to back-diffusion. Long-ago we made sure that
anytime a regulator goes into the N2 circuit, especially those cheap types,
it is a non self-relieving type, yet another source for back-diffusion.

As a sidebar to this conversation I question whether Teflon-PFA tubing and
the associated fittings offer any advantage over polyethylene tubing when
it comes to delivering N2 with high process purity.Teflon tubing and
fittings are subject to compression-set given time and like other polymers
PTFE and extrudable variants have diffusion issues.

Given the above we decided that in processes where N2 becomes part of a
critical thin film, we will add an N2 purifier cartridge. These typically
come as a compact SS cartridge with face-seal (VCR) fittings and can be
exposed to air, for the short time required for installation. If you pick
one of these make sure the reactive purifying material is a metal-alloy and
not an organic.

In system, gas sticks typically have a pneumatic isolation valve (Nupro)
before and after a mass flow controller (mfc). The best place to install a
purifier is in this circuit. Consider adding an additional valve to isolate
the purifier should an mfc needs change-out or have a VCR cap handy to
protect it.

Given such placement the purifier cartridge will only see the actual amount
of N2 requires by the process and otherwise remain isolated. Given good
supply N2 purity it should last a long time - perhaps into retirements?

Regards,
Bob Hamilton




Robert Hamilton
University of CA, Berkeley
Marvell NanoLab Equipment Manager
Rm 520 Sutardja Dai Hall, MC 1754
Berkeley, CA 94720
Phone 510-809-8618 (desk - preferred)
Mobile 510-325-7557 (my personal mobile)
E-mail preferred: bob at eecs.berkeley.edu
http://nanolab.berkeley.edu/



On Thu, Jul 30, 2015 at 3:12 PM, Mark Heiden <mheiden at engr.ucr.edu> wrote:

> We are moving a sputtering system from a remote installation into our
> cleanroom and the intent has been to take it “off the bottle” and use the
> building process nitrogen system for process gas as well as venting. The
> bottles on the old install were 99.999% but the building process nitrogen
> is not certified to any exact number. The piping comes out of the 9000 gal.
> tank and through a vaporizer, then it goes through filters then splits into
> “house” nitrogen which is copper piping and “process” nitrogen which is all
> stainless steel with 10ra polished inside.
>
>
>
> From the stainless building process nitrogen line regulator, we ran high
> purity Teflon PFA tubing to the point of use filter on the MFC. Since I
> can’t provide a precise purity for the evaporated nitrogen gas coming from
> the bulk tank and the tubing from the regulator to the filters is not
> stainless, a PI that was using the system is terrified that the nitrogen
> won’t be pure enough. Could I get your opinions on what the purity may be
> expected to be and if it is less than 99.999% would the filters correct
> this anyway?
>
>
>
> I’m sure most operations that spent the money for a “process” nitrogen
> system are not then running everything on bottles anyway so how do you
> ensure that the nitrogen being delivered to the systems is pure enough?
>
>
>
> Thanks in advance,
>
>
>
>
>
> Mark Heiden
>
> NanoFab Cleanroom Manager
>
> Center for Nanoscale Science and Engineering
>
> University of California, Riverside
>
> 951-827-2551
>
>
>
>
>
> _______________________________________________
> labnetwork mailing list
> labnetwork at mtl.mit.edu
> https://www-mtl.mit.edu/mailman/listinfo.cgi/labnetwork
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mtl.mit.edu/pipermail/labnetwork/attachments/20150731/7d9d5b89/attachment.html>