Ralph Weber once recommended the following steps to clean cavities.
It got rid of some, but not all of my background:
If your cavity is contaminated, it is easiest if you know what the
contaminant is. Of course, no one will ever admit breaking a sample tube
in the cavity. (Unless you're there when it happens.) I recommend the following
procedure.
1. Put the solid collet (plug) in the bottom stack of the cavity.
2. Place the cover over the irradiation grid. The cavity should now be
fairly leak-tight.
3. Wash with toluene. Pour the solvent into the top stack of the cavity.
Swish the solvent around for a few minutes. Be careful not to get the
solvent on the paint on the outside of the waveguide as most of the
solvents mentioned in this procedure are effective paint strippers.
Pour the solvent out and dispose of it according to the local
environmental laws.
4. Wash the cavity with acetone. Follow the same procedure as in step 3.
5. Wash the cavity with methanol. Follow the same procedure as in step 3.
6. Wash the cavity with distilled and deionized water. Follow the same
procedure as in step 3.
7. Dry the cavity with a stream of clean and dry nitrogen gas.
8. The first solvent is non-polar and each washing is increasing polar.
It is hoped that one of the solvents will dissolved the contamination.
If you know that the contamination is a metal, you can try washing the cavity
with 0.1 M disodium EDTA followed by a thorough washing with distilled and
deionized water.
I see a message from Graeme Hanson has just arrived. If you use cotton swabs
to clean the cavity, be careful. Cotton gives a background signal. We do not
use Brasso (or toothpaste) to clean cavities, so we cannot guarantee the
results. We would appreciate any testimonials as to its efficacy. (Thanks for
the recommendation Graeme.)
The recommendation from Graeme on how to re-attach the front plate is essent-
ially what is done here in Billerica. Take care when crimping the gaskets in
order to avoid tears and holes in the thin gaskets. It is IMPERATIVE to use a
non-magnetic screwdriver. (BeCu is the most common.) It is also very important
to remove your analog watch (if you are wearing one) as it will be permanently
magnetized and will no longer keep correct time. If you have any questions or
require any advice regarding this procedure, please call one of our service
engineers at 508-663-7406 and they will be happy to help you.
Beware of some advice in the literature. In one of Poole's books was the
recommendation to wash cavities with aqua regia. This is a recipe for disaster!
The thin walls of the cavity (required to allow the 100 kHz modulation in) will
be quickly eaten away.
If all else fails, the cavity must be replated. Art Heiss has told me:
It costs 70% of the price of a new cavity to have a cavity replated.
Any cavities which have been stripped and replated by another firm will
cost 100% of the price of a new cavity.
The side walls of the cavity are particularly difficult to replate to a
proper thickness. A too thin or too thick layer will result in poor
performance of the cavity.
Good luck with the cleaning!
Ralph Bruker Applications
rtw@domino.bruker.com (Ralph Weber)
Hope this helps,
dave scott
iowa state university
515-294-4057