Tuesday, October 08, 2002


Summary of Responses to my question posted to AMMRL on October 02, 2002:


Now that cryo-probes are becoming more ubiquitous, I have been asked a specific question, to which I don't know the answer.  Is it practical to have a cryo-probe equipped spectrometer that is used in an "open-access" lab (for use by "routine" users)?

Of course, the sensitivity benefits are well-known for large-molecule work, and for trace analysis of natural products; however, as far as I know most (or all) facilities with these probes only allow the NMR experts to use them.  If my impression is wrong (i.e. if there are labs in which the average user has access to cryo-probe-equipped spectrometers), I'd like to know about it.

Please reply directly to me if you can shed light on this issue.  More and more people are asking me about the practicality of these probes in various settings, and I really don't have any answers.  I will, of course, post a summary of the informative responses that I receive.

Thanks in advance,

-Rich Shoemaker

Responses From AMMRL Members: 

  1. (Neil Jacobsen [neil@u.arizona.edu])        We are considering a cryo-probe on a new 600, primarily for protein work but it will be "open access" to fill all the available time.  I don't think there is any reason that the probe should be limited to experts (although beginners should be banned so we don't get depth gauges in the probe, etc.).  I just did a demo trip in Billerica and Clemens Anklin brags that the probe is "just like any other probe" except for the ensitivity.
Because of the fixed expenses ($12,500 / year for mandatory maintenance contract) we will charge 50% more for time on this instrument.  I think that alone will weed out the people who are looking for time on "any NMR".

  1. (Chris Rithner [cdr@colostate.edu]) Hey Rich...the only information I can give you is with regard to their installation and removal.  Basically, it is not recommended that they be un-installed (or warmed up) once they are installed.  Too many free/thaw cycles are not good on the RTSC, I guess.  And that would affect how you might use one in an open access facility if you don't have a magnet to spare for one, a magnet dedicted to one, and have to change probes to accommodate other uses.  The probes are apparently not "tuneable" in the classic sense (the Varian probes aren't yet, perhaps Bruker's are different) so in my facility we would have trouble dedicating one multi-purpose instrument to their use.  The Varian probes will accommodate either a flow module or standard paddles which is a neat touch.  Big gain, I guess, for organics (not lossy) and reasonable gains even for salty samples.  Good luck with this.  I think that a flow system could be devised that would revolutionize how we do "walk-up" nmr in support of organic labs.  Maybe all the pieces aren't there yet and a new system with cryo-flow would be expensive ($750 to $850K for 500) but perhaps in a year or two.
  1. (Rainer Wechselberger [wechselberger@panda.chem.uu.nl])  We do not really have open access to our cryo-probe system (mainly because we don't have open access to any of our systems), but in general the same users who are using the other instruments are also using the cryo-probe system. Assuming that your 'open-access'-users are using 'standard'-experiments, which are setup by you anyway, I don't see that much of a problem. The main thing you really have to take care of with those systems is that the heating of the sample/probe keeps limited. Dangerous in this respect are TOCSYs (especially heteronuclear) or heteronuclear decoupling where the duration or the power is 'excessive'. We trained our people to check the temperature response of the probe first thing, when they start a new experiment. That worked fine so far. But even if something goes wrong with that, this is only a medium catastrophy, leading to an automatic warm-up cycle of the cryo-system (and to half a day of down-time...). As far as I understood, usually no damage on the system should to be expected. 
  2. (Rainer.Haessner@ch.tum.de ) meanwhile we have two of these cryoprobes, one for protein NMR and one for routine analysis.  Both of them are by no means uncritical in use. For routine use there are three main problems.  a) the Q factor of these probes is very high. Due to this fact the influence of the solvent of the Q factor is much larger than in the case of conventional probes. As a consequence the 90 degree pulse lengths varies strongly. In the case of conventional probes you have a variation of about 20% and in the case of cryoprobes the variations are larger than 100% between CDCl3, Acetone and DMSO. This is valid for the case of optimal tuning and matching.  b) Tuning and matching in routine work is a pain. Even slight changes of the sample detunes the probe. Until now there is no ATM unit available for cryoprobes. This detuning additionally modifies the 90 degree pulse length.  c) Due to the high Q factor the probes are much more sensitive against high power output especially of the proton transmitter. Usually you get arcing starting from 8 W output from the proton transmitter.  The main problem for the protein NMR is the high Q factor again. Water suppression is much more complicated in comparison to conventional probes.  And a last remark: you have to use nitrogen as the flow gas  for the probe. Oherwise forget a high shim quality.  That's comparative easy to achieve.  B u t : at   no   time the gas flow may stop!!! Your cryoprobe is dead after ten minutes without gas flow! We ensure this using two ony way valves one for nitrogen gas with a pressure of 1.6 bar and one for pressurized air of 1.3 bar. If the nitrogen fails, pressurized air ensures the life of the probe, but of course the measurement cannot be used.    Please excuse my bad english.
  3. (Frank Koehn [KOEHNF@wyeth.com])  We have 2 cryoprbes in our facility which are used predominanty for small molecule work. The first of them  a 5mm triple resonance Bruker 500 MHz probe, we have had operational a little over 2 years. The sensitivity aspect of these is a tremendous asset for limited mass samples and allows one to solve problems using experiments  unatainable with conventional probes. However, they are not well suited for wide open access for a number of reasons. 
The first is that they must be precisely tuned and matched to each particular sample in order to fully realize the sensitivity gain and pulsewidth reproducibility, even within the same solvent. In most open acces environments one sees a wide variety of samples and users. Some samples are mass limited but others are many mg and in a variety of solvents. Cryoprobes are more difficult to tune than a conventional probe, and cannot be tuned in automation. (At least not ones we have)  This makes them unsuitable for open access environments where multiple users , usually not spectroscopists and usually in  somewhat of a hurry, walkup and use the spectrometer. It is actually possible to have too much sample, so users must be instructed to limit the amount of material, otherwise one runs into receiver gain issues, etc. Ironically, it is also difficult to work at low sample temperature.

Secondly, there is quite a bit of additional delicate hardware, transfer lines, extra cables, etc.  around the magnet which makes it less convenient to insert and retrieve samples. This might not be much of an issue in a spectroscopy lab but in open access one must remember that any inconvenience, trip hazard etc.  is multiplied many times. 

Third, open access probes need to be robust and rugged. The cost ($200K plus extra infrastructure) , difficulty of repair, and mechanical sensitivity of these probes outweighs their impact in an open access environment. They have high infrastructure requirements.

An exception to this would be limited open access, say a natural products or biological NMR group, where a small number of OA users usually with mass limited samples, could be trained in the cryoprobes nuances and get the full benefit of the hardware. 

  1. (rolf tschudin [tschudin@helix.nih.gov])    We have 3 of these. Used by small groups, 3 to 5 people each, after a little training. With one person responsible for the system. They are fairly routine systems most of the time.
Cheers              Rolf
  1. (Peter Demou [peter.demou.b@bayer.com])      We have a cryoprobe on our DMX-500, but it is NOT used for open access. The reason I caution against this is that the vacuum line and the helium gas line are obstacles that come out the front of the magnet and should not be bumped at any time.  Sure everything is supported and the fittings are tight, but there some unexplained messages that pop up in the log file such as (pressure loss in system).  This one indicated that it lost some helium that it made up from the "bottle".  You also have to expect to warm it up every 6 - 8 or so weeks so it can purge itself.  I hope you know that the annual service contract on the cryo platform is $15400.  One might think that a time and materials deal would be better, but the annual service that is required plus time and materials ( a new cold head for the platform and adsorber unit for the helium compressor) come out to the same, but your probe isn't covered.  A warm up and cool down cycle takes about 5 hours if there are no problems.  The probe is so heavy that it takes two people to remove and install it.  Forget about cleaning all the juice from broken samples easily.  Sure you can get a factor of 3 in S/N but you can end up with a warm up / cool down cycle necessary at the wrong time.  Yes, there is warning in the log that indicates the impending need for this, and you can plan for it, but it can happen by itself unexpectedly. Also, the helium compressor cannot be on a UPS unit.  The lockes rotor current is 110A!  In reality the circuit that runs it is three phase 60A @ 208VAC.  The cryo platfrom can be on a UPS with the console.  So, you can end up in a situationwhere the compressor can shutdown.  The system then needs to eject the sample to keep it from possibly freezing in the probe. If you get one it will keep you on your toes.
Best regards,


  1. (Industrial Lab: anonymous by request, but someone who's opinion I respect!)     For a factor of four in signal to noise is it worth it to spend ten times more money for the initial probe and an equal amount for a backup??? With walkup use, how long before someone breaks a sample in the probe? Or how long until a foreign object falls down into the probe? I'm not accusing the everyday user of being careless, in fact quite the contrary. However, in my experience of managing walkup operations, Murphy?s law applies perfectly. If it can happen, it will.
     Most of the results I have seen using cryoprobes exhibit nice s/n improvements for organic samples where temperature is not important. Otherwise they still seem to be having problems controlling sample temperature. This usually becomes apparent in spectra of sucrose where the water lineshape is very noticeable. Some of the better results I have seen involve 3mm samples in a 5mm tube.

  1. (Stephanie Mabry [mabry@umbc.edu]) I should be able to answer your question this spring.  We have a cryoprobe for our department 500 MHz instrument, but installation has been delayed (yet again).  I am hoping we will be up and running by the end of the year.
Currently this instrument is open for "advanced" users, i.e. gradate students and post-docs who are willing to take the extra time to train on the Bruker software.  These users need higher sensitivity and resolution or are doing 2D experiments.  Note that the majority of our NMR users are happy with the 300 MHz QE/Tecmag instrument since they want a spectrum in 15 minutes or less.
  1. (George Sukenick [g-sukenick@ski.mskcc.org])  OK we don't have one, but I have been thinking about it. You might think that for organic synthesis such a probe would be overkill, but quite often C13 (and other X, and 2D's) are required for small quantities. When the next step of a synthesis is hanging on an NMR result ("as the sample decomposes before my eyes..." :-) time is literally money and it is essential to get people on and off the instrument as  efficiently as possible.
However, the other consideration : $$ of the probe, and it needs to be sent in for repair anytime someone breaks a tube (or too many dirty tubes are inserted...) the people who use that probe should be well selected.
  1. (Sudha Veeraraghavan [Sudha.Veeraraghavan@uth.tmc.edu])  Dear Rich, We have just bought a cryoprobe. Given that it costs $220k  and maintenance is likely to cost another 12k per year anyway, I would  only have it be used for biomolecular NMR applications and only by those grad students and postdocs, of NMR labs, who get direct supervision & training in NMR use.
  1. (Rudi Nunlist [rnunlist@purcell.cchem.berkeley.edu]) Hi Rich, from what I gather, this pretty much requires a dedicated instrument. Given that, if some qualified, properly trained users can help out the others, it  should work. Perhaps a sample changer would be useful? Of course, the only sure way is to get one and find out yourself ;=)


  1. (John Harwood [jharwood@uic.edu] )  FWIW, here at UIC we are planning on the next year or two to try to acquire a 5mm C/H cryoprobe to go in a new 500 MHz Bruker system.   This will be part of our departmental facility and will have the same open access as our other machines.  I may be incredibly naive, but I have not seen or been told anything about croprobes (at least the Bruker setup) that suggests to me they wouldn't be OK to use in an open facility.  What I like about the cryoprobe is that we can get this extra sensitivity while still using standard 5mm samples that are consistent with our other spectrometers.

Best regards