Hi AMMRL Colleagues.
Thanks to all who responded to my request for help rejuvenating loose spinners!
As expected, the seemingly simple question of improving spinner grip touched
on many issues, including:
* O-ring materials and sizing
* NMR tube quality and institutional expectations
* Sample hygiene
* Spinner cleaning practices and frequency
SHORT ANSWER - REPLACE THE O-RINGS WITH SMALLER ID O-RINGS:
Grip rejuvenation was accomplished by replacing all the spinner O-rings with
smaller ones, which tighten spinner grip by flexing the spinner prongs farther
into the spinner bore. (original size appears to be 3 mm thick x 8.5 mm ID)
The new O-rings are made of either Viton fluoroelastomer or fluorosilicone,
which are both rated for greater chemical resistance than Buna-N orregular
silicone rubber.
Blue POM spinners were first equipped with 3x8mm O-rings, evaluated for grip
(see procedure below), and classified as either "normal" or "loose". The loose
spinners were tightened further by having their top O-ring replaced with a
thicker one bearing an even smaller inner diameter (I.D.), size "dash 203"
(3.53x7.52mm).
* for normal spinners: 3x8mm, black Viton, temp range -26 to +204 °C,
McMaster-Carr #1295N313<
https://www.mcmaster.com/1295N313/>
* for top of loose spinners: Dash size 203, brown Viton, temp range -26
to +204 °C, McMaster-Carr #5267T166<
https://www.mcmaster.com/5267T166/>
Gray/frosted Kel-F spinners, used in many VT experiments (rated for -4 to +120 °C),
were first provided with “dash size 109" (2.62x7.59mm) blue low-temperature
fluorosilicone O-rings (temp range -62 to +204 °C), evaluated for grip, and
classified as either "normal" or "loose". The loose spinners had their top O-rings
replaced with "dash size 203" O-rings of the same blue material. O-rings made of
this material are more expensive than others, and are not a good choice for
supporting large numbers of spinners in an autosampler on a RT-only instrument.
* for normal spinners: dash size 109, blue low-temp fluorosilicone, McMaster-Carr #8333T168<https://www.mcmaster.com/8333T168/>
* for top of loose spinners: dash size 203, blue low-temp fluorosilicone, McMaster-Carr #8333T24<https://www.mcmaster.com/8333T24/>
OPTICAL BANDS:
Several spinners exhibited damage to the reflective stickers essential for
proper optical sensing. These stickers were replaced with new ones ordered
from Bruker Labscape, #Z42852
<
https://bruker-labscape.store/collections/all-mr-items/products/label-standardbore-spinners-for-5-mm-2-5-mm>.
DISCUSSION:
This project was motivated by an incident in which a loose spinner/tube interaction
nearly damaged our SampleCase autosampler-equipped instrument. The problem presented
itself as a tube stuck at the top of the magnet, jammed between the magnet bore and
the SampleCase transport tube. When originally loaded into the carousel, the spinner
barely held the tube, but it was inserted into the magnet normally. It appears that
at the final "click" drop into the probe, the sample dropped in the spinner. When it
ejected, the tube stuck out below the transport tube and got jammed - luckily not
breaking in the magnet bore. The student who reported the problem removed the
problematic sample and took this troubling video demonstrating the fit between the
tube and spinner (it's just 2-sec long, so I suggest you right-click on the YouTube
Play button and get it to loop): https://youtu.be/TtdYh-A4Y9w
Life would be simpler if we could affordably ensure all our spinners and tubes were
identical, but that's not practical in our open-access environment.
Spinner Variability:
Spinner grip definitely depended on the individual unit. Our spinner supply
(~180 blue POM and ~15 Kel-F, plus a few ceramic) includes items obtained
via instrument purchases over the years, and it appears that subtle differences
in manufacture can make the difference between one gripping a tube tightly and
another letting the same tube slip through. Our blue POM spinners include some
that have "curved" prongs, and the rest have "straight" ones. The curved prongs
have a narrower neck and visibly respond more when supplied with a new O-ring.
No curved-prong spinners were classified as loose with the first round of O-rings;
they may be more fragile, but they perform better, in my opinion. Of the
straight-prong blue POM spinners, about half were still loose even after being
equipped with 3x8mm O-rings, and they needed extra tightening.
Tube supply and quality:
Tube quality in our facility is variable, but most are high-throughput/economy/disposable
tubes. Most of the experiments here support organic small molecule synthesis, and
the chemists appear to be getting the answers they need from data acquired with the
cheapest tubes they can find. Training here includes discussion of the importance
of tube quality, but acknowledges the reality that the PIs can't/won't support
widespread use of $15 tubes when $1 tubes suffice. That said, coaching trainees on
purchasing tips has little impact. Most labs have one person in charge of ordering
supplies, and most new users inherit tubes from someone senior. Different labs get
tubes from different suppliers, and we don't have a handy stockroom that could
help establish an institutional norm. My next step is to reach out to faculty directly.
Training:
To safeguard the instruments, I encourage our researchers to have a low threshold
for disposing problematic tubes. I now include haptic training for entry-level
certification, in which new users feel the tightness of a normal tube/spinner
interaction, then feel one that is too loose. I tell them if they encounter a
loose interaction they must find a different spinner, and if they try three and
all fail they must go back to lab and throw out that tube. I'm also encouraging
them to throw out tubes that are chipped and scratched. I do advise them to get
a couple of 500 MHz precision tubes to use on special occasions like final
characterization and VT work, and they should keep them in a disguised velvet-lined
box at the back of a private drawer so no one else uses them.
O-ring material, an unexpected problem, and why use different sizes sometimes:
My selection of material changed over the course of this project, as I learned
more about them. One person suggested I use dash size 203 O-rings made of red
silicone, and they worked in preliminary tests, so I outfitted all 60 spinners
in our SampleCasePlus with them. Grip was good, though a little too tight for
some tubes. Over the course of two weeks, however, on three separate occasions
I found one of these O-rings on the floor next to the autosampler, it having
popped off the bottom of the spinners. I can't be certain why, but I think is
has to do with the dynamic interaction of the spinner and white tube housing
of the SampleCase carousel. The silicone material is tangibly tacky, and the
dash 203 size has a larger *outer* diameter than others. I speculate that casual
users pulling spinners out of the carousel sometimes tilt the spinner in a way
the rubs the O-ring on the inner surface of the white tube, where it may stick
and roll off the spinner. The large O.D. of the dash 203 O-rings confers less
clearance between O-ring and white support tube than smaller O-rings. Further,
the larger thickness (3.53 mm) of the dash 203 may be larger than can be
accommodated firmly in the spinner's groove.
Proper function appears to have been achieved by using Viton rubber instead
of silicone, and by using the thicker O-rings only on the top prongs of the
spinner, which do not interact with the white plastic tube in the carousel.
Special thanks go to Dean Olson for pointing me to “dash size" options,
and to Martha Morton for suggesting use of the blue chemical-resistant
low-temperature Viton flurosilicone. Further thanks goes to Raphael Stoll,
who is using (perhaps developed) an accessory “spinner clamp" that prevents
tubes from slipping: https://nmrtools.com/nmr-produkte . These look good, but
I think I'd have a hard time convincing people to use them in my open-access
facility.
Additional information about the process of assessment and other insights are
provided below, along with a compendium of helpful responses.
Many thanks to respondents: Dean Olson, Sameer Al-Abud-Wahid, Thomas Eubanks,
Charlie Fry, Dee-Hua Huang, Martha Morton, Ta-Chung Ong, Heather Schenck,
Raphael Stoll, George Sukenick, Vanessa Timmerman, Paul White, and Letitia Yao.
I hope to see you all as soon as circumstances allow.
* Josh
ADDITIONAL DETAILS: Testing and replacement procedure; O-ring notes, compendium of responses
TESTING & REPLACEMENT PROCEDURE
* Two representative unused tubes were identified for testing: a Wilmad
528-PP ("large" diam., caliper<
https://www.mcmaster.com/8619A67/>-measured
at 4.96 mm), and a Norell S55-00GS-050-1780 ("narrow" diam, measured at
4.92 mm diam).
* Each spinner had its existing O-rings removed and was cleaned:
* Exterior gently scrubbed in lab glassware detergent solution with
a soft brush and sponge
* Bore was gently scrubbed in detergent with a hobby-grade pipe cleaner
* Batches of spinners were rinsed/drained multiple times in tap water,
then DI water, then 70% isopropanol and left to dry in a colander in a hood
* Each spinner was supplied with a pair of "default" O-rings (3x8 mm for
blue POM, dash 109 for Kel-F) and manually tested for grip tightness with the
narrow tube. Spinners that did not grip tightly were designated as "loose"
and set aside. Each spinner designated as "normal" with the narrow tube was
loaded with the large tube to check that it was not excessively tight.
* Each spinner exhibiting acceptable tightness for both narrow and large
tubes were marked on both the top O-ring and the inner surface of the top portion
of the spinner, then either brough into use or placed in a case for the future.
* For each spinner designated as loose, the top O-ring was replaced with
a dash 203 size O-ring and retested. A very small number of spinners failed
to grip sufficiently even with that O-ring arrangement and they were discarded.
The ones that passed were marked as described above and brought into use or stored.
O-RINGS - Misc notes:
* Simple replacement of old with new O-rings of the same size has no perceptible
effect on grip strength in my hands. (3mm "thick" x 8.5mm I.D. Buna-N rubber).
* A non-exhaustive exploration of the effects of O-rings with different widths
and I.D. indicated that grip strength is highly dependent on I.D., but not so much
on width. For example, 1.6x8.5 and 3x8.5mm O-rings provided similar strength,
whereas 3x7.9 was much tighter than 3x8.5.
* Dash size 203 red silicone O-rings appear to be made to English units
1/8" x 16" "fractional" width x I.D. (0.139"x 0.296" "actuals"), which
would be approximately 3.2x7.9mm "fractional" dimensions.The most important
dimension is the smaller inner diameter (7.9 vs 8.5 mm), which pulls the spinner
"tongs" closer together.
RESPONSES:
Outside of restricting the quality of tube or vendor, replacing toroids on
spinners is the key. you can find them at
https://nmrtubes.com/spinner-turbine-toroids.php<
https://urldefense.com/v3/__https:/nmrtubes.com/spinner-turbine-toroids.php__;!!BpyFHLRN4TMTrA!r_LKbxZqu56ZkR5F6NXMuaeHOJ_iqgL79FFuDhMO0N_sXoe0m1HEgVqIxiJx0boL8Ss$>
Thank you for such interesting insight! I've had similar issues too and mainly
I have been replacing O-rings and scrubbing the inside bore with pipe cleaner
and 70% iPrOH like yourself. If you receive a lot of responses, please do send
them out as I would love to hear what everyone else's solutions are too.
We use replacement o-rings from McMaster-Carr. We have not worn out any blue
spinners in 6 years of use on the SampleXpress automation system. We never have
problems with spinners, but we clean them with ethanol once in a while.
For us we have to replace a few spinners every year because they chip when
people drop them on the floor or they simply disappear from the lab. Bruker
offers promotion/discount occasionally and I wait till then to put in the order.
Do you have chargeback? Then make the cheap tubes expensive (which they actually are)
What if “these are the tubes that I recommend. When a problem crops up due to
use of other tube types or defective tubes, then charges for the repair as
well as for the full downtime (and maybe also tech hourly rates for the repairs?)
will be applied
Or is that a dream? :-)
That being said - I find that the PEEK sample holders from Wilmad work pretty well.
I have to take off one of the two O rings for good tubes. When I clean the holders
I use a cotton swab, kimwipe, and some methanol or IPA.
The only thing that I am really fussy about with my users is the type and treatment
of NMR tubes. A good NMR tube is $13-$20 each and according to my observations is
reused it seems for a lifetime (it seems like the gold ring in LOTR Smeagol's
"my precious" :-) ). Limited volume MS vials are a dollar or two each and are tossed
after each use. Cost of NMR tubes pretty much pay for themselves compared to MS vials.
When I have new users, regardless of experience, I require that they have a full day
training - which includes 1 hour about NMR tubes.
It's not as horrible as it sounds (LOL) - most researchers are interested -
part of it is showing a shim stack, inside of a spare probe, and making points about
shimming, spectral quality, etc - Why a bad quality tube gives bad spectral results,
wastes time shimming, even why it's a bad idea to use less than 0.5 ml in a 5mm tube.
I also make the important point that other people are doing the same research as them
in other institutions, and if the one who publishes first gets the citations, the patents,
etc and the second place guy gets maybe published and a pat on the back - so time wasted
getting spectra well you know the rest.
They even know what the lock signal actually is and what they are looking at on the display.
It's a lot of work for me to spend the day but I'd rather this than cleaning up messes.
For a while I was having some issues with users - I found a way to give new users a pack
of good quality NMR tubes.
That helped get over the hump. I don't do that anymore - practically everyone is in compliance.
In 28 years we've had literally less than 10 broken NMR tubes in the NMR instruments without
sample changers.
Most of my users are chemists - so lots and lots and lots of samples - reaction monitoring.
With sample changer - a few dozen, less than 50 :-) not great but still not
too bad - occasionally it is the changer's fault (a story in itself) and when policy was
changed to "use cap only once", it is rare.
For the higher field 600 cryo with SampleJet, they have 2.5 hrs more training which includes
chemist's 2-D experiments. The SampleJet is very picky for tube as well as cap.
The machine picks up the sample by the cap. Caps can be used (pretty much) only once.
If tube is broken at the top - sample is lost.
If tube is wrong length - fails or loses sample.
If sample is lost down the core - cryo needs to be cycled, instrument is down for the day (or longer).
There's some peer pressure; people hate it when the cryo is down.
When I had some problems (DNP people using junk tubes from their desktop NMRs), I told everyone
if I see something doesn't meet policy, sample is removed run or not and they have to come to me
if they want it - because we need to keep the instrument up and running for everyone.
I try to move move forward and not lay blame - but also be truthful.
When something happens, everyone knows why it happened
Either a message on the machine or a short mass email to the users and PIs.
Not necessarily the who - that is destructive when people point fingers - usually only I know
the "who" and we talk.
If someone points fingers, I point out our policy is to be professional and
move forward.
I know the pain of some spinners fitting and some not, although not to the
extent that none of the spinners would grip.
Until recently I worked in a University and although we didn't have the throughput
in numbers you have per month (maybe 500-1000 at max), some of the spinners that were
used for the oldest instrument certainly were much older than 10 years. They never
bought new spinners for the instrument they had since 2007, so I2m fairly sure
that over the time we would have seen similar wear.
I also don't know of any university having the money to buy completely new spinners
on a regular basis, even with bulk discount.
On of the culprits my colleagues from Bristol suspected recently where a (Jeol) spinner
just straight slid off the sample was the moisturiser used by the students since they
continuously sanitise their hands. Maybe there is some degree of this involved. Granted,
the Jeol spinners are different, but additional "lubricant" could have an additional effect here.
From what I've heard from colleagues, most just replace the O-rings and I would be
tempted to go down in diameter a bit from the ones you mentioned. Maybe to something
like 8mm ID x 14mm OD x 3mm cross section.
Given that all spinners have two O-rings, you could see if you have some of
the "good" spinners (where all but the odd tube fit well in the spinner) and
then swap one of the O-rings onto another spinner where only some samples fit.
If you have a "good" O-ring from spinners that grips well, take it off, leave
it for a few days and then get it measured by mechanical workshops or similar.
That may give you the dimensions "required". You could also swap them out like
a mix-and-match and see if it improves changing either the bottom O-ring or the top O-ring.
The thing I did differently for the cleaning, is I usually took both O-rings off,
then wiping them well from the outside and inside with both water an (neat)
isopropanol. For the inside I'd usually use a cotton bud and for the o-rinds in
most cases it's just been fibres and fluff, so a good rub with a slightly damp
wipe would do. But I definitely would take care to minimise exposure to alcohol
for the O-rings. Once all spinners are clean, I usually put the O-rings back on
and they are ready to go.
In regard to the tubes, I've noticed that the high-throughput and economy tubes
from Wilmad tend to be thinner and have a greater variation (4.9467 - 0.0191 rather
than 4.9635 - 0.0065 of the 528-PP) so they may be worth using when testing.
I hope any of this is of use and you get your issues resolved soon!
I inherited a very tight operation from my predecessor we only allow Wilmad
Precision or Shigemi tubes, and students are trained to wipe down tubes with
isopropanol and to handle spinners with Kimwipes to avoid transferring skin
oils to the magnet bore. We're also a smaller operation than you, so I have
an easier time of things!
We are not picky about the MHz rating of the tubes, only that they are precision
grade. We routinely use Wilmad 200 MHz tubes (506-PP-7) for 600 MHz metabolomics
and have DSS linewidths between 0.6-0.8 Hz. They cost 9 CAD each. Whenever I have
a person "complain" about our tube requirements, I point out that a broken tube
in our Prodigy probe would result in a few weeks/months of lost usage and a repair
bill in the thousands.
I have noticed that one of our spinners is loose and occasionally a sample
will slip. About half the time that happens, the tube ends up being an Economy
grade tube that someone got as a freebie, etc., and tried to use.
For cleaning spinners, we use isopropanol/water and drag a wetted Kimwipe through
the bore to clean it. Haven't needed anything more than that.
We have also had good luck with our blue spinners, many being in service since
2012. Occasionally we lose one, but that has been rare (my guess is 2-3 per year).
We do not work real hard to keep them clean; they seem to be mostly fine, with
pretty stringent words to students to not touch the spinning surfaces. We have
used replacement o-rings that I believe are the same as D's (red silicone dash 203).
These work fine. I prefer the white kel-f spinners, but only when we're doing
VT since they're so expensive. They work fine over the primary ranges we run
at: -120 to +120C. Haven't used a ceramic spinner in forever.
Our spinners don't see quite as much work; maybe 70 samples a day or so. We
have two SampleJets that handle the higher volume use.
It seems the extra cleaning you're doing might be counterproductive. That's
too bad if so. Such energy should be rewarded. But one of the reasons I
haven't worked on keeping ours cleaner is a worry that the plastic would swell
and/or warp with any type of solvent being introduced too often.
We do clean those Bruker's blue spinner every few weeks by removing the o-rings
from the spinners, then put both the spinners and o-rings in a ultrasonic cleaner
with DI water (sometime bottle drinking water) for 10 or 15 minutes. We use
cotton applicator with methanol or ethanol to clean the bore of the spinners
after taking them out from the ultrasonic cleaner. We set them aside for them
to dry, as well as let the o-rings "relax". We will put the o-rings back to
the spinner right before putting the spinners back to service. This seems to
help the spinner's performance. We've had our spinners since 2015 and I have
a small bucket of broken ones. Bruker used to make them more robustly, but
changed the design around 2013.
We clean with iPrOH, kimwipes for the outside, q-tips for the inside.
I do ask that people not use the economy/high-throughput tubes since they
are more likely to be loose (or too tight).
Yes, they do still get used, but periodically I will ask students to
gather them all up and I will replace them with abandoned tubes.
we have replaced the o-rings with NMR Spinnerclamps:
https://nmrtools.com/nmr-produkte
This has completely solved the problem. As the the NMR Spinnerclamps
(there are different types for each diameter, i.e. 3, 4 or 5 mm nmr
tubes) adjust to all nmr tubes for a tight fit, even economy grade tubes
will be handled without a problem.
Josh Kurutz, PhD
NMR Facility Manager, Chemistry Dept.
https://voices.uchicago.edu/chemnmr/
jkurutz_at_uchicago.edu<mailto:jkurutz_at_uchicago.edu>
Received on Fri Aug 13 2021 - 06:51:13 MST