Skip to main content

How Not to Do It

How Not to Do It: NMR Magnets

Here’s an NMR imaging blog with details of a recent problem in an Indian facility. Two people ended up stuck to the machine, pinned by an oxygen cylinder (!) that one of them brought into the room. Both sustained injuries.
There are two questions here: one is how anyone is allowed to wheel a ferromagnetic metal cylinder anywhere near an NMR magnet, and the other is how it took so long to quench the magnet once the accident had occurred. That latter point is addressed by the blog link above – the hospital is saying that the emergency quench circuit malfunctioned, and that it took four hours for a GE technician to arrive and get things shut down. I’m no NMR hardware expert, but I wonder about that one myself. As that blog post concludes:

Whether or not GE was really at fault in Mumbai we shall learn eventually, I hope. (I have heard rumors that some sites like to bypass their quench circuit in order to avoid having the cost of recharging the magnet should the quench button get activated. Insert your own exclamations of disbelief here because I’m incredulous.) In the mean time, this sorry saga is an opportunity for all of us to review our own procedures and take the extra moments to ensure that we’ve done everything humanely possible to eliminate risks. There really is no excuse.

34 comments on “How Not to Do It: NMR Magnets”

  1. Anonymous says:

    It’s an MRI magnet…

  2. Teddy Z says:

    I would never say this never happens here. I remember oncce in grad school the guys in the stockroom had placed the liquid nitrogen cylinder on a metal rollie thingee, which just looked the non-ferromagnetic one. A labmate rolled it in to the room to do the LN2 fill on a 500 and it started heading to the magnet. She tried to stop it, her arm got caught, and she broke it.

  3. Rhenium says:

    MRI is just NMR, but without the N…
    Oh and you don’t have to suspend your person in CDCl3.

  4. Anon says:

    A couple of jobs ago, someone on the overnight cleaning crew moved a freestanding metal coat rack from the hallway into the NMR room. No injuries, but the coat rack hit the magnet hard enough to cause a quench.

  5. Anonymous says:

    You sometimes have to bring ferromagnetic helium gas cylinders into the magnet room to pressurize the liquid helium dewar during helium fills. You keep it as far as possible from the magnet, especially if the magnet is not shielded, and use a long gas line. You must secure the cylinder to the wall as well- you can’t let it rest on the cart.

  6. oldnuke says:

    @3 Although a little chloroform might give the people pinned to the magnet a little relief! 🙂

  7. bad wolf says:

    @6: 2001 is “not too long ago”?

  8. Algirdas says:

    Oldnuke & Rhenium,
    I was told some of patients that are claustrophobic or simply can’t take the racket from the gradient coils do get MRI scans while sedated. Chloroform very apropos.

  9. pgwu says:

    When I was doing my postdoc in Baltimore I heard a story about a professor in medical school trying to change a liquid tank by himself Friday afternoon and was stuck between an NMR instrument and the tank for the whole weekend.

  10. luysii says:

    A less benign use of MRI (NMR to you) was the fatal rupture of a clipped intracranial aneurysm in a patient with a ferromagnetic aneurysm clip [ Radiol. vol. 187 pp. 612 – 614, 855 – 856 ’93 ].
    Why does medicine use the term Magnetic Resonance Imaging (MRI) rather than NMR (Nuclear Magnetic Resonance)? Because we’d never get anyone to stick their heads inside anything with nuclear in its name. Seriously, this is the actual reason.

  11. Magrinho says:

    I saw this happen on Gilligan’s Island. Gotta be careful around magnets.

  12. DrSnowboard says:

    There was an incident at GW just after the merger. New shiny Stevenage facility, new NMRs, metal ladder……..Fortunately only a damaged career

  13. Anonymous says:

    I wonder what would happen if you got pins and needles in your legs while waiting for an MRI scan? Would they abort the scan in case the pins and needles reacted to the magnet?

  14. xfin31 says:

    And on the opening day of the GSK 800 MHz facility, I think a photographer’s tripod got sucked in, quenched the magnet, which had to be rebuilt …

  15. Boot says:

    Did my PhD in solid-state NMR.
    I asked specifically if the shifter spanners that we used to attach the LN2 fill lines were ferromagnetic, I was told they weren’t.
    One Friday morning (we usually topped up the magnets on Friday) I was talking to a mate who was working on another machine while I was filling our magnet. In the process I gestured with the spanner. The next thing I knew was my arm being pulled towards the magnet then a very loud metallic clang.
    Turns out my supervisor was a much better NMR guy than metallurgist! My mate still thinks the look on my face was one of the funniest things he’s ever seen…

    1. Sal says:

      Re: Boot, “spanners”.
      I heard this safety rule in some other context (forget which) where you never take someone’s word on whether something’s safe, but always make sure by testing. Seems like an eminently sensible rule in situations where “unsafe” treated as “safe” can lead to disaster, if testing is feasible.

      I think I’ve even heard a version where, if posed such a question, you weren’t allowed to answer “safe”, it was either “unsafe” or “I don’t know”.

  16. dearieme says:

    I know an MRI lab that got planning permission to be built only after it changed its description from an NMR lab. The residents of the nearby Clerk Maxwell Road had objected to it under its original name.
    Am I alone in finding that a toe-tickler?

  17. Anonymous says:

    Probably a stupid question, but do most NMRs have this strong of a magnetic field? I always used to take out my wallet before running a sample to avoid wiping credit cards, but I haven’t done that for years and I’ve never had a problem. I’m not sure I’d stop myself and think about it if I was about to bring some big metal thing into the room (of course, I can’t think of circumstances where I would be doing that).

  18. Secondaire says:

    I had a friend who worked for an NMR apps group in grad school. Can’t remember where this happened, but there was an NMR facility where things kept going wonky with the magnetic field of one instrument during overnight acquisitions. The staff couldn’t figure out what was going on *at all* and tried every kind of troubleshooting known to mankind. Finally, someone suggested they put a camera in the NMR room, and lo and behold, every night, the resident custodian would use a metal (ferromagnetic) folding chair to “wrestle” with the magnetic field…

  19. Algirdas says:

    @18 Anonymous re “do most NMRs have this strong of a magnetic field”
    Yes. Majority of NMR spectrometer magnets have field strengths (300 MHz proton = 7 Tesla) stronger than majority of MRI machines.
    Keep in mind though, that both Bruker and Varian (R.I.P.) worked steadily on shielding technologies for magnets. Modern “ultrashielded” magnets have the 5 gauss line within magnet legs.
    Having said that, over the years in our NMR lab, which up to recently exclusively featured old Oxford magnets with their very wide 5 gauss cicles, I have forgotten my wallet on me couple times while putting the sample into the magnet. No ill effects on any cards observed. YMMV.

  20. Anon says:

    @18 I once wanted to run an NMR at -78 C to monitor a reaction and brought the tube to the instrument in a dry ice/acetone bath all held together in an ordinary clamp stand. So as not to let it warm up too much in between, I began passing the whole clamp stand to the technician atop a ladder. It started being pulled towards it with great force – new underpants moment! It was only thanks to my own super-human strength that I could wrestle it away.

  21. Anonymous says:

    The non-sparking tools made of berylco (Be-Cu alloy) are also suited for use in strong magnetic fields.

  22. Darren B says:

    There is no emergency quench button on most (if not all) high-field NMR systems. The “emergency quench circuit” is just the conduit to discharge the electricity safely. This only occurs when the magnet feels like quenching, like during the initial charging. So if your O2 tank/other large metal object doesn’t sufficiently perturb the system, the magnet is not going to quench on its own.
    We do not let anyone into the room untrained, or with anything large and metallic in their hands. Lab tours stay outside; janitors too. Service people are escorted by trained personnel.
    Transfer lines can be bought to connect anything large and metallic (i.e. gas cylinders) from a safe distance. I would rather use rubber tubing to connect to a distant gas line, than for a tourniquet.

  23. Anonymous says:

    I had a USB stick get trashed by an unshielded 500 MHz magnet. I was doing a helium fill and was standing next to it for awhile. I think it takes awhile for the magnets to wipe the cards. I’ve walked up to this same magnet and walked away with my wallet and everything survived.

  24. Jon says:

    I used to wear a combination analog/digital watch in grad school. If I didn’t take it off before going to the ancient 400 MHz NMR (it was something like 19 years old by then), the magnetic field was strong enough to stop the analog part of the watch from moving. I could tell by synchronizing the analog and digital displays before going near the magnet. The analog watch would lose a few seconds every time.

  25. Canageek says:

    While I was doing my undergrad someone took a steel dewer near our least-shielded NMR (an old 500 MHz). At night, without signing the instrument out. Bad, but didn’t damage the instrument. What DID damage it a fair bit was when they tried to physically pull the dewer off the magnet to cover it up. One of our NMR techs told me that the only reason it didn’t quench was that they’d done a helium fill 3 days before.

  26. CarlW says:

    I just had an idea to potentially improve the safety around MRI machines. In the hallway outside the MRI room, have a moderately powerful permanent magnet chained to the wall, and a sign saying to test whether metal objects are magnetic before bringing them into the room.
    Would this be helpful?

  27. Algirdas says:

    @27 CarlW
    I don’t think this would be very helpful. Dangers of ferromagnetic items near NMR magnets are well known, and people tend to be quite vigilant about that.
    Many thousands of MRI scans and NMR spectra are recorded around the world every single day of the year, and more than 99% are uneventful. When some sad accident like the one discussed in this post happens it is a newsworthy event – precisely because it is so rare. Hence I’d argue that the current level of awareness and diligence is good enough. We should be as vigilant as ever and not get complacent, but at the same time no additional measures are required.

  28. NC says:

    #25 has reminded me. Do you think my smartwatch will get ruined by the magnet? I take it off every time I take an NMR, and the NMR director and I were debating it.

  29. Martin says:

    Our old Gemini 300 got hit by a rotary floor polisher many years ago and quenched. Irony was that the cleaning supervisor had been warned about the magnets strength and so took it upon himself to step in and take over from the regular cleaner to clean around the magnet. Got a bit too close and bang, sucked in underneath and wedged it between the legs hard enough that even after it quenched it took a bit of pulling to get it out. Luckily little probe end damage just bent the tuning knobs a bit. Still have the photos somewhere of people clowning around for posed shots afterwards

  30. sepisp says:

    Why isn’t there a simple metal detector in the room door?

  31. Calli Arcale says:

    Ooooh, MRI. Amazing technology; my brother works with it as a physicist. But it can go pretty spectacularly wrong. One thing I’ve kind of wondered — how does it work for those mobile MRI machines? It would seem rather hazardous to me for them to be just rolling down the street with a powerful magnetic field like that.
    Also, I am reminded of an MRI quench story that came up on The Daily WTF, a blog about IT horrors and coding catastrophes. The story is likely heavily dramatized. Short version: to dispose of two odious engineers who were bogging the team for this MRI monitoring software down, an engineer submitted a false report of a failure that could not be reproduced in simulation. It would need to be tested in the lab, which was offsite but had a working MRI machine. Unfortunately, the scheme backfired, as the odious engineers decided to instead use a fielded machine at a hospital instead. (Maybe it was a shorter drive?) They rolled out their “fix” to the nonexistent bug, without telling anybody, and some time subsequently the monitoring system failed to activate the quenching system and it exploded.

  32. Mike A says:

    Medical MRI gear does indeed have field strengths that will do that sort of thing. Chemical MRI gear? I’m not sure, as my knowledge dates to 50 years back, when Rice had a Shiny! New! NMR rig and an equally Shiny! New! Ruby Laser! I suspect the field strength increases as a function of either the diameter of the imaging aperture or its volume. Certainly someone else “here” will know.

Comments are closed.