Dear Steve,

I agree with Uwe re the use of Louwers-Hapert (L-H) stopcocks. We have used
the traditional 'Max Coleman' pear shaped reaction vessels for off-line
carbonate analysis. These are fitted with L-H 9mm stopcocks. We have had no
problems with any residual excess pressure after a 12 hour reaction period.
The valves are ultra-reliable and in my experience have an undetectable
leak rate.

In contrast whenever I have used valves with teflon stems there always
seems to be a problem or two. I'm not sure if this is due to gas leaking
past the o-rings, or due to porosity/permeability problems with the teflon.
I think Carolyn's reply shows that it is possibly a problem with the
teflon. However, I have also noticed another problem with the teflon valve
stems of Young's valves. This occured when initially setting up for
hydrogen isotope analysis using Zn reduction. After heating the reaction
tube and allowing then allowing to cool there seemed to be a degree of
'relaxation' in the stem as though it was taking on a permanent
deformation. This resulted in the seal opening somewhat and leaking. The
upshot of this is a tendency to 'overtighten' the valve initially so that
it remained sealed. See my comments in the next para re: the sealing of the
L-H valve.

There are other design advantages to the L-H valves. Unlike many valves of
the Ace or Young's design in which the seal is made by compressing the
o-ring against a valve seat, the L-H valve seals by compressing the o-ring
between the inner stem and an outer precision glass bore. It is not
possible to overtighten the valve, in the conventional sense, and damage
the o-ring, or even worse break the glass valve! The L-H valves are also
readily automated using either vacuum or compressed air actuators. Whilst
in manual form they cost a little more than equivalent ace/young style
valves in manual form, as an automated valve they are a fraction of the
price of say a nupro equivalent.

Best wishes,

Paul