The Vacuum System
There are three main types of vacuum pump: rough pumps, high
vacuum pumps which need backing, and high vacuum pumps which
don't need backing.
A rough pump can pump down from atmospheric pressure, but
can only reach a rather modest vacuum, at best 0.01 mbar (0.00001
of an atmosphere).
High vacuum pumps can reach good or ultra-high vacuum (UHV),
between log(-6) - log(-10-11) mbar. A high vacuum pump that needs
backing means that it needs another pump (a rough pump) to
work on the 'back end of it' (the high-pressure end), to
take away from it air that it has pumped out of the vacuum
chamber. Examples of such pumps are diffusion pumps and
some types of high-vacuum turbo pumps.
Some UHV pumps, like ion pumps, do not need backing: the
vacuum they deal with has such a low pressure that the pump
can dispose of residual gas atoms simply by accelerating
them into the solid walls of the pumping chamber (this seems
improbable, but it does work).
Other types of pump include
'getters' (at high vacuum) or 'traps' (at rough vacuum),
which simply allow a chemically-reactive surface to adsorb
or absorb stray gas atoms. Cryogenic pumps freeze molecules
to a surface; many microscopes have a 'cold finger' - a
thin, cryogenic pump that is led into the objective pole-
piece to keep contaminants away from the specimen, and is
cooled down from outside by a dewar of liquid nitrogen.
The main vacuum system on a typical electron microscope has
a roughing pump, a diffusion pump and an ion pump.
The roughing pump is a large mechanical contraption which is
easy to hear. It is used to pump the chamber from 'air'
(atmospheric pressure) when necessary, to pump the specimen
transfer chamber (also from atmospheric pressure), and to
back the diffusion pump. Because it is used to do so many
things, it is usually attached to a 'roughing manifold' - a
pipe with lots of other pipes coming off it. By opening and
closing various valves off the roughing manifold, the
roughing pump can be switched from one role to another.
Ask the demonstrator: To point out the roughing pumping on the
schematic diagram on the computer screen. Ask to be shown
it in reality.
How many valves come off the roughing manifold? Where do
they go and what do they do?
The diffusion pump is usually positioned at the bottom of
the column, near the back, and its front end is used to pump
the viewing chamber and the photographic film casement.
There is usually a large empty volume between the back end
of the diffusion pump and the roughing manifold, so that
when the microscope is being used, the roughing pump can be
switched off (it creates a lot of vibration). The empty
volume is very slowly filled with the gas being pumped out
of the chamber by the diffusion pump. From time to time
(once every few hours), it is necessary to empty the backing
line, by pumping it for a short while with the roughing
pump. On modern machines, this all happens automatically,
although the sound of the roughing pump coming on in a
quiet, dark room can be quite surprising.
Ask the demonstrator: To point out the diffusion pump on the
schematic diagram of the vacuum system on the computer
screen, and also in reality.
The gun and top end of the microscope is often pumped by an
ion pump. This section of the microscope is almost entirely
separate from the viewing chamber, except for a small
aperture (called a 'differential pumping aperture') which is
large enough to let the electrons down the column (remember,
they are all confined to a few tens of microns width), but
which is small enough to maintain a differential pressure.
Note that at high vacuum, air does not get sucked through
holes like in a domestic vacuum cleaner because the mean-
free path of air atoms at low pressure can be many metres,
so that they never notice each other. They just ricochet
around the chamber and rarely pass through a small aperture.
In this way, it is possible to have a poor vacuum in the
viewing chamber (say 10-4 mbar) but UHV (10-8 mbar or
better) in the gun chamber.
There is often a valve between the top half of the column
and lower half, so that the bottom section can be let up to
atmospheric in isolation. All this depends on the exact
make of microscope.
Pressures in various parts of the vacuum system are measured
by gauges. There are several types of gauges, but they fall
into two broad categories: low vacuum gauges (usually
Pirani gauges) which monitor areas which go up to
atmospheric pressure; and high vacuum gauges, which are
highly sensitive and are quite often entirely destroyed if
they are exposed to atmospheric pressure.
Ask the demonstrator: To point out the vacuum gauges on the
schematic diagram of the vacuum system on the computer
monitor. To point where their values are displayed.
You don't need to know how the gauges work, but it is worth
bearing in mind that the values shown on the computer screen
are purely notional, and bear no systematic or linear
relationship to the actual pressures in the system. For
example, Pirani gauges mounted on various parts of the
roughing manifold, are virtually insensitive outside of a
range of pressures between about 10-2 mbar and 5mbar: all
that they indicate is that a rough vacuum exists or does not
On a modern machine, the computer controls the whole vacuum
system, and in general these systems are pretty fail safe.
However, as a qualified user of the machine, you must be
aware of what the normal behaviour of the vacuum system
should be, and what the normal values of the vacuum gauges
should read, just in case something goes wrong.
A side-entry stage normally just requires the user to push
in a rod, which is holding the specimen, into the side of
the microscope. As the rod enters, you should hear the
roughing pump come on; a red light may also show. After a
minute or two, the red light extinguishes, and the user must
then rotate the rod, and then very carefully let it be
sucked right the way into the microscope.
Ask the demonstrator: To show you how to load and unload a
specimen, and to supervise you doing it on your own several
Pay particular attention to remember at what angle the
specimen rod goes into the chamber, how long you have wait
before you turn it, and that you are exceedingly careful not
to let the rod get sucked in too quickly under the influence
of atmospheric pressure pushing it into the column.
Also watch what happens on the vacuum screen of the
computer. Do you understand the sequence of valve openings
and closings? Remember, the backing line must never go
above rough vacuum (letting it do so is a good way of
inflicting thousands of pounds of damage to microscope: the
oil in the diffusion pump will back-stream into the column).
What happens to the vacuum gauges and why?
The demonstrator should explain whether you must wait until the
pressure recovers to a certain value, what that value is,
and which gauge monitors it.
Ask the demonstrator: How to load a specimen into the specimen
Note the o-ring, which seals the rod into the column. Never
touch the o-ring, or any part of the specimen holder which
lies deeper into the microscope than the o-ring. A single
fingerprint will torment the next twenty users with terrible
Copyright J M Rodenburg