Peristalsis works, but there are other ways.  In any event, tight seals
(low leakages) are essential.  You should read two review articles on
micro-pumps.  One is by Juan Santiago, the other by Peter Woias.

You should also look at the technology employed by Tandem Diabetes to
move small amounts of insulin around.  Do a patent search on both Tandem
Diabetes (assignee name), as well as its predecessor company, Phluid
Corp.  One of the inventors' names is Paul DiPerna.

Finally, you should look into the latest MFC technology available
commercially.  A quick google search on 'ultra-low flow mass flow
controller' will do the trick.  I have no vested interest in any of
these guys, but for instance its specs indicate the Sierra Micro-Trak
MFC gives control down to 0.1 sccm of N2, which is a factor of 50 below
your present minimum.

Al Henning

====================
Ned Flanders wrote:

OK, I know, this is off-topic, and I will be lucky if my post survives
the wrath of the list moderators. But still, since I know that there are
some really smart heads around here, I decided to ask, maybe someone
would know:

I am trying to achieve low flow rates of some (mostly inert) gases.
Our flow controllers can achieve the lowest flow of 5 ml/min, which is
too much for my purpose. We have a bunch of peristatic pumps, that can
achieve down to a few microliter/min, but they are used, as far as I
know, with liquids only, because of the compressibility of gases. My
question is, if I forego precision of flow rate control, could I use a
peristatic pump to pump a gas? Are there any thermodynamic obstacles,
and with which gases?

Many friendly thanks in advance

m