Hi there fellow spinlanders,
Hope this message finds you well. This is a question for those of you
with a recovery system. I've been meaning to write to the group that
here at UVM we *finally* got our recovery system installed at the end of
July. I think you probably saw my message via Greg Wylie at TAMU who
posted a synopsis to his query. The whole process, getting to the
installation part was quite extended, since our faculty member who was
awarded the NIGMS supplementary grant for the system, received this
funding TWO years ago! Even though our new building had all the
recovery pipelines already installed and a dedicated room in the
basement, there were still extensive building modifications needed for
the system to be installed.
We ended up purchasing a medium pressure system from CryoMech that has
the capability of producing 15L per day (LHeP15) and uses a bag to
collect the gas. There were some funky things that happened at
installation, but we ended up only adding 1 cylinder of helium to the
system, plus whatever boil off came off of a Bruker Ascend magnet fill
when the system was installed. So for a few months we were only in
'liquid production' mode, since we only have two NMR magnets and the
faculty's MCD (has a supercon that needs to be ramped up) is still not
set up for recovery. We also have an EPR, which will be used with a
cryostat and LHe, so all in all not many systems that need the liquid.
Prior to installation of the recovery system we had to build manifolds
for the two NMR magnets. I had considered using 1 manifold for both
magnets, but the way my lab is set up, that wasn't possible. I have
many of you to thank for your assistance in figuring out how these
manifolds work and all the components that they require. Many of you
had sent me photos and detailed explanations and I would like to thank
first and foremost Greg Wylie at TAMU for lots of photos and videos, a
supplies list, and back and forth discussions, and now an excellent web
page on helium recovery. I would also like to thank Johannes Leisen at
Georgia Tech, Jurgen Graf in Heidelberg, Allan Kershaw at USC, Tara
Sprules and Robin Stein at McGill and last but not least Eric Paulson at
Yale for the detailed drawing with part numbers of a manifold and one of
the first web sites on recovery. Without your folks' advice, I'm not
sure how my manifolds would have gotten to materialize!
As I said, this e-mail has been a long time coming, and now I have an
issue that possibly some of you could give some advice on. We finally
started drawing liquid from our system on Oct. 18th (last week), but the
liquid transfer from the production dewar into the transfer dewar has
been very inefficient and it's been taking us hours to transfer. During
our first try, the transfer dewar was warm, so it took a while for
liquid to collect in the dewar. We spent 2h on the 18th and then an
additional 2h two days later and ended up with only about 45L in the
dewar. The 2nd transfer was on Monday this week, fortunately the dewar
was still cold and I'm unsure of how many hours the transfer took (2-3h
I believe), since I couldn't be there, but we ended up with less than
40L in the dewar, and this was the liquid to fill my Oxford 500 magnet
and alas, the magnet only got to 72% full. Then I attempted
transferring yesterday afternoon and again this morning, I think that
the dewar was cold, but this morning it took me 3.5h of filling and I
had only emptied out 10L from the production dewar (!). And after
thumping the transfer dewar, possibly there was 1" of liquid, but the
fluctuation frequency almost immediately indicated the liquid to gas
state. At this point I don't have a good way of telling when my
transfer dewar is full - we do have a differential pressure gauge on the
transfer dewar, but it was set up incorrectly, so we have to wait until
the dewar is warm to fix this. Here are some of my questions:
When the transfer dewar is warm, what is the best way to cool it off?
With so few liquid He needs, I'm afraid we will have a warm dewar quite
often. The faculty member wants to use LN2 to cool it off, since they
do that with their MCD all the time. It seems that would be quite time
consuming!
When we are in the middle of the transfer, a lot of gas is being
produced, so the bag fills up and the compressor to compress it into
cylinders goes on. Is it OK to continue to transfer through this
process? We're concerned that the compressor being on and the dewar
being connected to the bag creates vacuum in the dewar and the He gas is
being 'sucked out'.
The pressure gauge of the plant typically ranges between 2psi and 10psi
(big gauge at the top of the production dewar) and often during our
fill, this drops to 2psi and perhaps even lower. So perhaps at this
point there's not enough 'push gas' to get the liquid over (?). But the
purifier is connected the whole time though likely liquid isn't being
produced during the fill. Should I only start the transfer when the
pressure is closer to 10 psi?
The amount of time, LN2 and electrical power to recover our own helium,
makes me wonder if it's all worth it. I know it's more worth it in
facilities with many more magnets, but in a small facility this is a
good question. Perhaps in a year when we're used to all the maintenance
activities and "the cycle" of producing helium, this might not seem as
time consuming, but at this point it makes me wonder. Although a big
benefit is not having to rely on an outside supplier AND not being
succumbed to the skyrocketing prices of helium! So... at this point I'm
celebrating my first magnet fill with our own produced liquid! Yay!
Thanks in advance for your suggestions to my concerns above. I will
post a synopsis if there's enough of us who could benefit from this
information!
Happy Halloween!
Cheers,
Monika
--
Monika Ivancic, PhD
NMR Facility Manager (W103A)
Department of Chemistry
82 University Pl.
University of Vermont
Burlington, VT 05405
(802) 656-0285 office
Monika.Ivancic_at_uvm.edu
Received on Thu Oct 28 2021 - 10:17:51 MST