Dear NMR Colleagues:
Following is a summary of the responses that I received and the
eventual solution of the problem.
I thank Martha Morton, Joseph Dumais, Johannes Leisen, Nancy Washton, Jerry
Hu, Banghao Chen, Bert Heise, Johanna Baldus, Jari Sinkkonen, Bob Berno,
Nader Amadeu for their great suggestions. Learned a lot.
Original question:
We have a Bruker CP/MAS VTN 4mm probe which has a chronic arcing
problem (spikes
on FID when detecting X with 1H decoupling). I sent it to Bruker for repair
every time it occurred and the probe would come back working fine for a
short while then the problem would reappear. Have you had any similar issues?
How do you have it fixed, and would you recommend some experts for these
problems?
Responses:
-----------
Our probe had the same problem. It was from the stator moving back and
forth. A grad student fixed it mostly with copper mesh.
-------------
I would suggest looking carefully at the flexible leads below the rotor
housing. That tilting rotor house is convenient for loading samples but it
seems that the flexible leads are sometimes the source of arcing.
-----------
Usually the Bruker probe people know what they are doing and they should be
able to permanently repair the problem. I would look what they did: clean
coil and stator? exchange any of those two?
Sometimes "arcing channels" might have burned into the stator and a good
cleaning might inactivate them for a time. The next time your probe
is arcing I would just try a very good cleaning of the stator , before
sending the probe back to bruker.
Another issue you should look at is at the experiments and samples you are
running. There are samples which easily cause arcing.
I would also be mindful about the experiments. Many CP-MAS experiments work
well with a 50 kHz decoupling and not more than 1K sampling points for the
FID.
--------------------
We had a similar issue with an HX probe, and we spend a lot of time
diagnosing it until we finally found that the arcing is actually not arcing
but an interference from a radio station (400 MHz for 1H ~ 100 MHZ for 13C
~ FM radio). You can simply check this by setting all pulse powers to 0W
and see if you still observe your “arcing” signal.
--------------
you could ask the guys of Phoenixnmr.com <
http://phoenixnmr.com/>, maybe
they can help.
-----------
could this be caused by moisture in the air used for the MAS? What is the
dew point of the compressed air you are using? I have worked in a lab, were
reoccurring problems during high power decoupling were related to a the
compressed air being much wetter than it was supposed to be. It is worth
checking. (Of course if you use N2 boil-off for spinning, you should be
fine.)
-----------
Our probe (4 mm CP-MAS VTN) has been twice in factory, but the problem
continues to exist. We had also another loaner probe and it had the same
problem. However, we had also a loaner DVT design 3.2 mm probe and it
didn't have any issues.
my feeling is still that the problem may be caused by not dry enough
compressed air. VTN and DVT probes have different design and different
setup in VTU unit (external and standard, respectively). We should get
during this Autumn a 4 mm iProbe, which should be DVT type. I hope
arcing problems
end then.
-------------
Open up the probe and inspect the entire RF circuit for both channels. It
is almost certain that the problem you are seeing is with the 1H-channel,
but if you have the probe open it doesn't hurt to look at the broadband
channel as well.
----------
For me, I observed some visual signs of where the arcing was happening,
very near the tuning capacitor.
If you see some charring, clean that area with isopropyl alcohol as best
you can. When it arcs once and leaves that deposit, then that is one
reason for the probe to arc again and again in the same spot. Cleaning it
will help stop that from happening.
Then, you look for a possible reason why it might have arced in the first
place. Check the solder connections especially. Are they clean and flat,
or are there bumps and possibly weak connections.
If they replaced the capacitor when you sent it in for repair, then it
shouldn't fail again so soon, but maybe they bought a bad batch of
capacitors.
If you would like to talk to someone other than Bruker to have a look at
possibly repairing the probe, you can try DOTY scientific, Revolution NMR
and Phoenix NMR.
David Doty has many years of experience with RF circuits for NMR and might
be able to help you out.
And it doesn't hurt to talk to the folks at Revolution or Phoenix as well.
-------------
1. Pulse high power overnight to maybe oxidize the contact surfaces;
2. Pulse high power with open circuit in a dark room to identify
sparks;
3. If you spot a position, isolate the parts with some material (eg
teflon) and pull them apart, if possible.
4. Arching can continue even after step 3. Arching can go around
isolating materials.
5. Arching happens mostly on decoupling. Confirm that with an
oscilloscope. Change your decoupling sequence. Spinal64 is better the
tppm and so forth.
6. Optimize your decoupling power and go for the lowest value you can
live with.
7. The company nmr-service in Erfurt, Germany can also polish the parts
and try to change some parts, but no guarantee as well.
8. Pray!
-----------------
Removing the coil from the stator is a better option but depending on the
particular probe it can be quite tricky to put the coil back since the coil
can easily be misaligned. Before that you might want to try to turn the
probe upside down, find a way to hold it and immerse the entire top of the
probe into Methanol. Keep it immersed in the MeOH for a few minutes and
then exchange the MeOH and immerse it again. Do this several times. Some
people also use an ultrasonic bath. Bruker recommends that you use Methanol
for all cleaning procedures.
-----------
Things that I tried:
1. Wipe clean the coil with a Q-tip (the wood end wrapped with Kimwipe
and soaked with ethanol).
2. Put the probe upside down and blow bearing and drive gas (1000 mbar
each) for 30 minutes. These two steps did not solve the problem.
3. Come to the lab at night, shutting all the lights, pulsing with
cover removed, and observe sparks. I found that there are so many light
sources in the lab e.g. numerous LED's and lights from adjacent labs and
buildings that are impossible to shut off or cover. Gave up with this
diagnosis.
4. The moisture in our system does not seem to be a problem.
5. Regarding the practice of soaking the probe head in methanol, Joe
Grifoni of Bruker cautioned that this could dissolve some paint and
generate a background signal, so I did not do it.
6. While playing with these activities, I noticed that one of the
flexible copper braid leads was partially broken. Joe mentioned that they
always check this when they repair MAS probes as the braids would fatigue
by constant bending back and forth. He kindly sent me a piece of the braid
and after it was soldered into place, the arcing problem disappeared. (Two
suggestions that I received, as listed above, pointed to this cause, but I
failed to notice it the first time I examined. One really needs to take a
close look to see it.)
7. Phoenix and Revolution currently don't repair Bruker probes.
In summary, there are many potential causes of the arcing problem. In the
past years, our probe has developed arcing problems multiple times
(repaired by Bruker in the past and in house this time), but upon a closer
look, it appears like every time the cause was somewhat different. We never
fail to learn in the field of NMR!
--
Weiguo Hu
Director of NMR Labs/Chemistry & Materials
Conte Polymer Research Center
120 Governors Dr
University of Massachusetts
Amherst, MA 01003
Received on Thu Sep 22 2022 - 21:14:56 MST