[labnetwork] Mystery White Solid in ICP-CL

[Ryan Rivers]

I've seen that in our lam 9600 for years. My suspicion is that it's boron
oxide or nitride as a byproduct of the BCl3 reactions. A wet clean should
suffice, but SF6/O2 should also mitigate it. (BF3 is volatile, and F can
slowly break the B-O and B-N bonds.)

I would check your chamber and gas panel leak up rates if you are getting a
lot of formation rapidly, because an atmosphere leak will generate this
stuff much faster. But it's going to form even if your system is sealed
tight, especially if you start etching metal oxides on purpose.

Theoretically adding a diluent gas should help but I've never needed to try
that hard to mitigate it as long as we keep up with cleaning.

You should assume the substance has unreacted B-Cl in its matrix and
generally proceed with the appropriate caution for handling a material that
can volatilize chlorine in very small quantities. We stick an exhaust
snorkel over the chamber and let it absorb water from the air for quite a
while before working on it. It'll pull moisture from the air and offgas HCl.

On Tue, Jun 9, 2026, 5:04 PM Emma Anquillare <eanquillare at gc.cuny.edu>
wrote:

> Dear LabNetwork Hive Mind,
>
> I went to manually scrub out our ICP-Cl Etcher with IPA as I usually do
> every 1-2 weeks, and opened up the chamber to find large quantities of
> white powder caked along the upper portion of the metal rim near where it
> meets the ceramic liner.  Large amounts of it seemed to have fallen onto
> the table as well. (See Photos). Has anyone ever encountered this? Any idea
> what it might be?
>
> Here are my suspicions and more details:
>
>    1. One of my colleagues commented that it looked like the Aluminum/a
>    she gets in her deposition tools, and I was wondering if something is
>    leeching out of the liner and it needs to be replaced? Might explain why
>    it's concentrated at the border, and appears in such visibly large
>    quantities for a nano-scale etch tool
>    2. The immediate suspect of course is also a solid etch product,  but
>    we have not had any new or unusual materials in the chamber since the last
>    cleaning two weeks ago (or at least, as far as users have told us). We have
>    recently changed our clean recipes so that instead of just a blanket O2
>    clean for everything, there are now customized cleans based on what people
>    are etching, but people seem to be following the rules on this pretty well
>    (see attached). I have also attached a summary of what has been etched in
>    the chamber since the last manual clean and using what gasses.  (We are
>    aware LiNbO3 is not volatile, thus the regular manual cleanings). One thing
>    that also occurred to me was the possibility of LiF formation with
>    increasing amounts of CF4 being run from the new cleans and reacting with
>    sputtered LiNbO3 in the chamber (but again, this seems like an uncannily
>    large quantity of solid for that).
>
>
> I collected some with kapton tape for XPS analysis, but curious to see if
> anyone else has encountered this, especially if it's an issue with the tool
> liner itself.
>
> Thanks!
> Emma
>
>
>
> *_____________________________________________*
> *Emma Anquillare, PhD*
> Research Scientist
> ASRC Nanofabrication Facility
> City University of New York
>
> *Catalyzing Change, Celebrating Gains: *
> *A decade of visionary science for the public good.*
> _______________________________________________
> labnetwork mailing list
> labnetwork at mtl.mit.edu
> https://mtl.mit.edu/mailman/listinfo.cgi/labnetwork
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mtl.mit.edu/pipermail/labnetwork/attachments/20260609/36b68dcf/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Outlook-rv1hrm15.jpg
Type: image/jpeg
Size: 141606 bytes
Desc: not available
URL: <https://mtl.mit.edu/pipermail/labnetwork/attachments/20260609/36b68dcf/attachment.jpg>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Outlook-rv1hrm15.jpg
Type: image/jpeg
Size: 141606 bytes
Desc: not available
URL: <https://mtl.mit.edu/pipermail/labnetwork/attachments/20260609/36b68dcf/attachment-0001.jpg>