6/26/96
questions, and was wondering if anyone out there had any ideas about
something our 200 kV TEM has been doing. We have a tungsten filament in
our microscope, and I've noticed that there are 2 intensity maxima when
I slowly saturate the filament. If I go to crossover and slowly
desaturate the filament, the tip(s) can be imaged at both maxima and the
images are different, but both look filament tip-like. The beam current
also seems really high. The two maxima occurred with a previous
filament and the beam current was also really high (at the second
intensity maximum, the first maximum is at the current I'd expect from
prior experience). Our technician doesn't seem to find this unusual, so
perhaps it's not a problem, but we go through many filaments in a
typical year. Anyone know if this is a problem? Anyone know how to fix
it ?
Regina Valluzzi
rv@squeaky.pse.umass.edu
tungsten filament's false peak?
Mark Darus
General Electric Co.
darus@cle.dnet.ge.com
the gun. If that doesn't work contact your service man.
Bob J.
bob@befvax.uchicago.edu
actually happening in the electron gun is that the emission current through
the filament is increased. This current flows through a bias resistor, and
this causes the bias voltage between the filament and the grid cap to
increase to such a level that the equipotential surfaces around the filament
tip attain sufficient curvature to cause the electron beam to be focused to
aspot somewhere in front of the grid cap. This process was first fully
understood by Haine & Epstein in 1952, and is nicely described in Haine's
book, "The Electron Microscope" which was published by Interscience in the
mid 1950's (Chs. VI and VII). I do not know the exact characteristics of
your electron gun, which depend on the size and shape of the hole in the grid
cap and the position of the filament relative to that hole, but is sounds as
though you have a situation where you are able to take the process through
two successive focusing conditions. It is quite possible to do this in
electron lenses - we once had a microprobe that had such a strong power
supply for the condenser lens that we could run the beam through two
successive focus conditions if we ran the power to the lens through the whole
range provided by the power supply. While this was an electromagnetic lens,
and you are dealing with a very primative electrostatic lens, it nonetheless
sounds like what is happening in your gun.
In any event, filament life is exponentially dependent on filament
temperature. So, to increase filament life you should certainly operate at
the the crossover or saturation condition corresponding to the lowest
filament heating current.
Wil Bigelow
Wil_Bigelow@mse.engin.umich.edu
filament heating process is the same. You just accelerate the electrons
through a bigger potential after they are emitted.
I don't undertand thermoelectric emission so I can't tell you why, but the
two peaks are normal. The second one should be a bit higher and is the
normal working position for an EM. It might be run at the first peak, but
slight changes in filament heating current either way would lead to a drop
in the emission current.
The saturation current will drop with time as the filament thins.
Therefore, the first peak on a new filament may be at the control setting
for the second peak on an old filament. Again this is normal. A thinner
piece of tungsten doesn't need as much current to generate the same heat for
electron emission.
Regarding lifetimes, be sure not to overheat the filament. The extra heat
goes into vaporiizng tungsten atoms. We heat our filament to the second peak
and then back off the heating current ever so slightly so that we just begin
to see a drop in the emission current. (A linescan profile helps give us the
sensitivity to see this change. Screen brightness or emission current is
less sensitive to the changes.) With careful saturation, we have been able
to routinely get lifetimes of 50-100 hours per filament. I think that is
pretty much the standard. We can drop to below 30 hours if we have
over-saturation or if the column vacuum is poor.
Does this answer your questions? What kind of lifetimes are you experiencing?
----------------------------------------------------
Warren E. Straszheim
270 Metals Development, Ames Lab/ISU, Ames IA, 50011
Phone: 515-294-8187 FAX: 515-294-3091
E-Mail: wes@ameslab.gov (or: wesaia@iastate.edu)
http://www.public.iastate.edu/~iprt_info/cfce/ (re: coal)
http://www.public.iastate.edu/~wesaia/marl/ (re: SEM)
called false), and second or full saturation. Most labs work at full satu-tion
I believe, but some work at the first (usually to extend filament life, with some
sacrifice of brightness. The gun bias also affects saturation. By limiting the increase in emission current to 8-10 microamps above the dark current, your
filament life should be normal. Poor vacuum in the gun will also shorten filament
life. You could also check that the filament is set back properly from the wehnelt.
John Hunt
hunt@msc.cornell.edu