Julio,
I am not sure what your problem is!
During the initial installation of our Micromass Optima IRMS, the gold
seals and other internal parts of the switching or change-over valves had
to be replaced by the installing field engineer, and the switching block
cleaned with solvents and baked out. We had a leakage problem that the
field engineer did not fully explain to me; however, the system was up and
running after this repair was performed.
A few years ago, I went through a nightmarish series of problems and
after eliminating everthing
else, repeated similar repair work in replacing gold sealing rings and
other internal parts. The instrument
went from inoperable to functional with that last bit of repair work.
According to the company and our top of
the line field engineer, the switching valves should not require this type
of repair. Despite these assertions,
I did the repair and our system started working afterwards; this is proof
to me, twofold proof, that sometimes
these valves have problems despite their rugged toughness in operation. We
operated 7 or 8 years prior
to performing the second valve part replacement. Without the help and
advice of the Micromass field engineer, it would have taken much longer to
sort out and repair the problems we had, including the
valve problem.
Notes: Our Micromass was the second system in operation in the US. The new
knife edge part we ordered during the second repair is different from the
ones on our system. The knife-edges on our system are chamfered or
bevelled and are dull as if they were sand-blasted or similarly treated.
The new part we
ordered had knife-edges that were cut to right angles or squared off. I was
not sure if this new part was
geometrically compatible with our old system and therefore, never used it.
Buena Suerte,
Glen Martin
"Julio A.
Villaescusa" To: [log in to unmask]
<[log in to unmask] cc:
> Subject: Need help with Optima
Sent by: Stable
Isotope
Geochemistry
<ISOGEOCHEM@LIS
T.UVM.EDU>
05/06/02 06:11
PM
Please respond
to Stable
Isotope
Geochemistry
Dear Isotope Geochemists,<?xml:namespace prefix = o ns
= "urn:schemas-microsoft-com:office:office" />
I am writing this message to you with the hope of getting some advice in
solving a problem that we have been having with our Optima Mass
Spectrometer during the last year. Probably some of you may have had
similar problems as this one, and your advice in helping us solve it would
be a HUGE blessing to our lab.
For several years we analyzed carbonate samples in our carbonate carousel
(Isocarb) with a typical external precision (1 Std.Dev.) of 0.02 ‰ for
carbon and 0.05 for oxygen. However, one year ago we started getting bad
precision measurements with smaller-size samples (~70 µg, when running
foram samples): ~0.12 ‰ for both carbon and oxygen, almost always being
carbon slightly worst than oxygen (about 0.14 vs 0.1 ‰), which is inverse
to what should be. Our zero enrichment tests (in the dual-inlet) give us
systematically good results (within acceptable “zero rejected delta”
values), but again carbon being slightly higher than oxygen, when it should
be the other way around.
It is important to mention that when we analyze larger-size samples (from
~300 µg, up to 1 mg), our external precision values are good, as always.
The problem of decreased reproducibility seems to be related to smaller
samples. This made us think that there was probably either some source of
contamination, or a tiny leak, that could only be “detected” when running
the smaller samples, increasing the “noise to signal” ratio. But we looked
after many other things while trying to solve our problem. We made sure the
acid was not the problem, then the chiller (giving stable temperatures).
Later we disassembled the piping, all the way from the carousel up to the
cold finger (from C4 to C1), and did an exhaustive cleaning. We checked for
leaks either using Argon or Nitrogen, using the mass spec as a “leak
detector”, and found no leak.
We also checked every valve in the Dual inlet and carousel section for
leaks. We indeed found several cross-seat leakages at valves SM (B1
increases to 20.5E-13) and RM (8.7E-13). Since these values are larger than
5E-13 (m/z 28 in the axial collector), as specified in the users manual, we
changed the diaphragms and the seats for these valves, as well as replaced
the MP3 gold ring. We, however, were not able to reduce the size of the
leak in these valves. Has any of you had adverse effects in your results as
a consequence of cross-seat leakages of this magnitude?
Finally, other line of evidence of problems is that we get a trend in the
delta values (either in C and O) when analyzing samples of increasing
sample size. For instance, when increasing the sample size from 70-100 ug
samples to samples >200 µg, we get the following: a change from -1.40 to
–1.55 ‰ for ä(superscript: 13)C, and a change from –2.5 to –2.7 ‰ for ä
(superscript: 18)O. Also there is trending in the same carousel batch when
running samples of equal size, the trend is towards more negative values
(in both C and O) as the time progresses. These trends are only evident
when running “small” samples of ~70-300 µg. Running larger samples (1 mg)
does not generate a trend.
In practical terms, to give you an idea of our problem, when we run a batch
run of 44 standard samples of ~70-100 µg, we usually get 2 to 4 bad values
in the batch. If we eliminate these bad values from the statistics, the
external precision is acceptable (below 0.1 ‰ for both carbon and oxygen).
The problem is that we never had to reject a bad measurement from our
batches and kept a good reproducibility. If any of you have any suggestions
and/or are interested in seeing our results with more detail, please let me
know. We think we have run out of ideas on how to solve this problem, but
are hopeful that this highly specialized group of isotope ratio mass
specrometrists may provide us some guidance on how to solve our problem,
based on your experience.
Thanking you in advance for your support,
Best regards,
Julio Villaescusa
<?xml:namespace prefix = st1 ns
= "urn:schemas-microsoft-com:office:smarttags" />University of Baja
California
|
