Dear netters,
Thanks for all the responses regarding the possible
causes of the wavey baseline that I was seeing on our
Waters HPLC. Some of the suggestions that I received are
summarized as follows:
1) Use highly degassed solvents and any tiny bubble in the
pump head should eventually dissolve. Helium sparging
helps as vacuum degassing often leads to new bubbles being
introduced into the line. (Thanks to Jim)
2) Noise may be due to a bubble in the detector which moves
back and forth in front of the lens. A syringe
is useful to push solvent through detector to get rid of
it. There may also be other junk in the detector which
moves back and forth in the flow cell (e.g. mold)
(Thanks to Jayne, Aida and Vick)
3) Noise may be due to a cracked flow cell. Pump cycles
will cause some fluctuations in pressure which then
translate into actual changes in the dimensions of the
cell. Look for possible leaks in flow cell althought they
may not be evident. (Thanks to Himmel)
4) Watch out for the quality of TFA that is used as it may
sometimes be highly absorbing in the low UV range. Thus
any small variations in solvent composition might affect
the absorbance dramatically. A higher wavelength may
help to reduce these fluctuations. (Thanks to John L.)
5) The problem with wavey baseline may be due to the
tubing being changed to narrow bore. I suppose this would
have to do with mixing and pulsations. (Thanks to Andreas)
6) If trouble is due to bubble in the pump try pumping 80%
methanol/water through the system with some backpressure
from an old column since methanol is a very good wetting
agent. If that fails, try dismanteling and
ultrasonicating the pump head and the check vlves in 80%
methanol/water. The small amount of water will apparently
help the methanol to wet the surface of the pump where the
bubble is sticking.
(Thanks to Andrew)
Thanks for all those ideas. I had a few of my own. As I
believed the problem to be in one of the pump heads of pump
B. I actually measured the flow coming out of the pump
heads and found a 1-2% difference in the amount of solvent
coming out of the two pump heads. While this seems like a
small variation I thought it must be the cause of the
fluctuations. Several possibilities exist to explain the
cause of this:
a) 1-2ul Bubble in right pump head causing the observed
1-2% variation in the 100ul stroke.
b) small particle in the right pump head.
c) leakage out of one of the check valves or the pump seal
or the fitting associated with the right pump head.
d) sovent starvation due to blockage of the inlet to the
right pump head.
I finally just called the Waters representative and he said
there was nothing wrong with the system. That sure is bad
news when you are trying to fix a problem that doesn't
exist. He did say that a mixer would help to get the
baseline into shape. So all that trouble and all we needed
was a mixer. If I had only known that!
Would anybody out there be able to recommend a good mixer
to place between the pumps and the injection valve? I
would appreciate any suggestions.
Thanks again for all your help. Sorry if I missed anybody
in the summary.
Regards,
John
_________________________________________________________
John Marcus Marcus at tpp.uq.oz.au (Dr J.Marcus)
Cooperative Research Centre for Tropical Plant Pathology
5th Level John Hines Building
University of Queensland
St. Lucia, QLD 4072
AUSTRALIA
Fax: 61-7-365-4771
Phone: 61-7-365-4764
