[labnetwork] SPM spikes, bath life, (Sulphuric Peroxide Mixture)

[Hope Fa-Kaji]

Hi Robert,

This is an anecdotal answer based on my experience in industry.

Bath life and spiking limits will depend greatly on the application for
which you are using SPM. When H2SO4 and H2O2 react to form Caro's acid, the
other liquid byproduct is H2O. As the H2O concentration of the bath
increases, all cleans and etches become less effective.

If you are using the SPM for bulk photoresist removal (i.e. your "dirty"
SPM), I'd think the bath life limitation is driven much more by the
particulate count than the spiking. Also, I assume in this case you have a
much higher H2O2:H2SO4 ratio and are consuming H2SO4 much more quickly.

For other applications like targeted etching or simple cleans, you should
have more room to do more spiking. In particular, if you don't have a
constant flow of wafers to process, I would consider pre-batch spiking
instead of a time based interval. This ensures there is enough H2O2 present
to react and form Caro's acid. Otherwise, the H2O2 could be consumed and/or
evaporate before the wafers enter the bath.

In my previous role (300mm wafer tools with inline filter/heater/pump and
lid covers to minimize evaporation), we ran some studies on H2O2 spiking
and could get 30 batches out of one bath (run at 120C) before the measured
H2SO4 concentration went below 88%. We had previously run to 15 batches but
we increased the limit by increasing the batch limit and running numerous
particle wafer tests to demonstrate no decline in bath quality.

I hope this helps!

On Thu, Mar 6, 2025 at 2:22 PM Macdonald, Robert (GE Aerospace, US) <
robert.macdonald at geaerospace.com> wrote:

> Wondering who is operating SPM baths (Sulphuric Peroxide Mixture).
>
>
>
> We heat Sulphuric to 120C (in a quartz bath), and spike with H2O2, at 1%
> by volume.
>
>
>
> We are doing 10 spikes, or 1 week, then dispose/aspirate (the cooled)
> bath, whichever comes first.
>
>
>
> Does anyone here have opinions or data on the number of spikes they allow?
>
>
>
>
> _______________________________________________
> labnetwork mailing list
> labnetwork at mtl.mit.edu
> https://mtl.mit.edu/mailman/listinfo.cgi/labnetwork
>


-- 
*All the best,*

Hope Fa-Kaji
Semiconductor Equipment Engineer
Microsystems Laboratory
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mtl.mit.edu/pipermail/labnetwork/attachments/20250306/7f5acacf/attachment.html>