Calibration Tips when Calibrating your TEM
It is more complicated that originally imagined!
The following information is based in large part on a contribution to the
Microscopy Listserver by Malcolm Haswell, EM Unit, University of Sunderland,
January 25, 2001.
Quest for the perfect magnification calibration:
There are a number of things that must be considered when checking
magnification on a TEM, and this is true of any TEM, and is not specific to
that of any one manufacturer:
1] Most manufacturers will aim for a consistency of about +/-5% accuracy
because of the laws of diminishing returns. Also they may have an optimum
voltage.
2] There will be variations up and down the magnification scale because a
balance has been struck between several lenses, and when they are excited
differently, this will be expressed differently.
3] There is a very short working distance between lens pole piece and
specimen (just a few millimeters) so if the specimen height varies
by +/- 0.2 mm this will greatly affect the magnification. For instance
if the grid is slightly bent or kinked, if you don't check magnifications
at a standard Z position (eucentric if that's available) then you will get
inconsistencies on the order of 5% without even hardly trying.
Malcolm Haswell says that when he calibrates at eucentric, he notices that
all of his magnifications at 75kv were near or better than +/- 4% whereas
one or two previous sets were mostly worse than +/- 5%.
4] All of the above will change with time as the electronics changes so
calibrations, should be done perhaps every 6 months or year and certainly
after major repairs by service engineers.
5] Consider how you measure your calibration data; one should always
measure the negative directly and never a print of the negative
(use a calibrated graticule eyepiece or travelling microscope).
It might not be universally appreciated, but prints alone can vary by
1 or 2% easily.
6] The calibration standards for TEM will be at best +/- 2% accurate unless
you have some certified statement or certificate saying otherwise. Also,
the nature of the calibration specimens can differ, that is, measurements
with catalase vs. a diffraction grating will give slightly different results.
7] Do not forget the hysteresis factor. On most microscopes one gets the
best calibration data by calibrating at the highest magnification then
working down the range because of inherent hysteresis in the electromagnetic
lenses. Some manufacturers even, at one time, manufactured TEMs with a
special "button" to overcome the hysteresis effect (for example, the old AEI
801); it was called "standardize magnification".
8] Never quote magnifications to 3 significant figures, it is meaningless.
Words of optimism:
These enumerated concerns and difficulties may sound almost like
insurmountable obstacles to the generation of meaningful magnification
calibration curves. But if one is aware of the potential problem areas,
they can be managed to the extent that they can be either reduced or avoided.
For example, if you have calibrated your TEM and see little change between
calibration sessions, then you can put confidence limits on your accuracy.
If it is critical that a user needs accuracy, then photograph a consecutive
field of a calibration sample under exactly the same conditions.
To Ask a Question or Make a Comment
To Place an Order or Request a Quote
Return to:
Saturday July 05, 2008
© Copyright 2008. By Structure Probe, Inc.
Contacting SPI Supplies and Structure Probe, Inc.
All rights reserved.
All trademarks and trade names are the property of their respective owners.
Privacy Policy
Worldwide Distributors, Representatives, and Agents
