Date: Sat, 23 Jul 1994 14:26:13 +0800
Subject: Si contamination in SX-50 and overlap corrns
From: B.Robinson@dem.csiro.au (Bruce Robinson)
X-Mailer:

John,

I have a query which I would be grateful if you would forward to SX users on
your list. I am not sure that I have the most current one. I hope you do
not mind being mailperson.

Si contamination in the SX-50.

One of our users noticed a year ago that our metal standards all seemed to
be contaminated with about 100 ppm of Silicon. This seemed proportional to
the time the standards spent in the SX. The most commonly used standards
seemed to have more silicon. I tried to postulate that the Si came from
other sources. However, yesterday I looked at the contamination marks
around the reference "Cross" on the normal Cameca 25 mm smaple holder. We
use this for standards, and we have used the engraved "Cross" to check stage
positions etc. It has quite a level of beam contamimation build up from
years of work (although we normally do not use this holder for routine
work). To my surprise, I found a lot of Si in the EDS spectrum from near
the cross, and much less in areas normally out of reach of the beam. The
EDS levels I guessed at about 1% Si or more, which is a lot!.

1. Would other users mind seeing if there is any silicon near the "Cross"
on their Cameca sample holders.

2. Does anyone have any idea of where the Si comes from.

Our SX is over 7 years old, and it still has its original diffusion pump oil
in it. We have changed rotary pump oils regularly, using straight mineral
oils. Our SX was of the vintage where there was concern about contamination
levels, and I recall that Cameca was using a brand of "diffusion-pump" oil
in the rotary pumps in an attempt to minimise contamination levels from
rotary pump backstreaming.

3. Does anyone have any other hints on how the Si is getting on the
surfaces, even when the beam has not been there previously. The Si level
does not seem to build up rapidly over time with beam usage. I left the
beam on one spot for a while, and the levels did not seem to be rising,
although the precision was not tremendous. I may do some more precise work
at higher currents.


Overlap to Ti on V.
Our major element analysis package for the PDP11 SX (WANUSX) offers overlap
correction if required. We are currently using 4.7% apparent V from 100% Ti
as the factor on PET. The overlap should be much less if LiF is used. I
have plotted the spectra out for two of our PET xls and one of our LiFs
using our Specscan routine. I can fax spectra to anyone interested, or can
"attach" Excel-4 graphs and spreadsheets for those who can decode attached
files in Mime, Binhex or Uuencode.

Regards to all users from distant Oz.

--------------------------------------------------------------------
| Bruce Robinson |e-mail: B.Robinson@DEM.CSIRO.AU |
| Electron Beam Laboratory | ,-_|\ |
| CSIRO, Exploration and Mining |fax 61 9 387 6046 /CSIRO\ |
| P.O. Wembley WA 6011 |phone 61 9 387 0378 \_,-._/ |
| AUSTRALIA |home 61 9 384 7409 v |
--------------------------------------------------------------------