Date: Fri, 3 JUN 94 20:24:3 EST
Subject: Re: XEDS;SEM;Standardizing spectrum
From: WALCKSD@ml.wpafb.af.mil
Status: RO

I assume that you have a good specimen current meter. The poor man's
way of making a Faraday cup is the following. Take a small block of
aluminum (any metal) and drill a hole about 2mm diameter (size of
drill???) relatively deep. Then take another drill the same diameter as
the outside diameter of a Pt or Mo 100 um aperture and drill to a very
small distance (just deep enough to accept the aperture). Silver paint
the aperture in place. With the large L/D ratio of this block, when the
electron beam is inside of the hole, essentially all of the secondary
and backscattered electrons that are generated will be captured. The
current that you measure will be the beam current. In the SEM, the hole
should look black. Tilting this whole relative to the sample normal by
a degree or two insures the capture of most of the electrons because of
the strong forward scattering of the backscattered electrons at large
tilt angles.

Another acceptable way to do this is to drill a small wire size hole in
the Al block relatively deeply. Tilt the hole as above. As long as the
hole appears dark in the SEM, you have a reasonable approximation to the
beam current.

Now that you have the beam current, you can put this parameter into your
software package. Record the beam current for your knowns and unknowns.
The software will make a calculation to adjust for differences in
spectra recorded at different beam currents. You might consider
standardizing your measurements with a particular beam current.

You can use this setup to measure the stability of your microscope.
Measure the beam current before and after collecting a typical spectrum;
if it is changing significantly the amount of time that you are
collecting data, you should probably call the service engineer. Your
microscope should be reasonably stable over the 3 to 5 minutes that you
might collect a long spectra. If it changes slightly, use the average
of the beam currents for your value. The problem with this setup, is
that you must move your sample when you bring the block over. If you
have an electrometer available, Fullam used to sell a reasonably priced
Faraday cup arrangement that can be brought into the beam without moving
your sample. It is mounted on a gimbal on a flange that can replace one
of your blank flanges on the microscope. This unit worked fairly well.

Hope this helps. Have fun.

Scott D. Walck
Materials Directorate
WL/MLBT, Bldg. 654
2941 P St. Ste. 1
Wright-Patterson AFB, OH 45433-7750

walcksd@ml.wpafb.af.mil
PHONE: (513) 255-5791
FAX: (513) 255-9019



---------------------- Replied Message Body ----------------------
Date: 6-2-94 2:34pm
From: {henne@sfu.ca}:ddn:wpafb
To: WALCKSD:ml:wpafb, SCHELTFJ:ml:wpafb
Subj: XEDS;SEM;Standardizing spectrum
Orig-Author: {Daniel Henne }:ddn:wpafb
-----------------------------------------------------------
Hello, humble Masters student here.
My question is regarding the standardization of one samples x-ray
spectrum to another. If my SEM was nice and stable I suppose I could
just take a spectrum for a certain period of time. Or if I had some
nice low noise amplifiers I could look at the sample current and
integrate that. However, my SEM isn't very stable and I'm not sure I
want to mess around with the sample current. Has anyone heard of any
other methods or can point to some articles regarding this subject.

Thanks in advance.
Dan Henne
Simon Fraser University
Vancouver, Canada.
email: henne@sfu.ca