From: Daniel Kuespert <0000057d3b6cd9b7-dmarc-request**At_Symbol_Here**LISTS.PRINCETON.EDU>
Subject: Re: [DCHAS-L] Condensation of liquid oxygen in open Dewar flasks of liquid nitrogen
Date: Tue, 16 Mar 2021 07:10:26 -0400
Reply-To: ACS Division of Chemical Health and Safety <DCHAS-L**At_Symbol_Here**Princeton.EDU>
Message-ID: 49E05E64-20DD-4A3D-872C-BE83CC88E0B8**At_Symbol_Here**
In-Reply-To <5786796F-12F6-48A3-BB6E-96FC2DC1A31B**At_Symbol_Here**>

Wow, that's a big dagger to drive in the topic! Thanks to all who responded. I had gotten to wondering if there was anything in the way of data to back the concern, and I think the answer is "no." I've seen this mentioned many times, including in such places as official national lab training documentation and manufacturer SDSs.

It would be interesting to look at how fast oxygen counterdiffuses against the thermal and concentration gradient in the Dewar flask, just to establish whether you can do it at all even if you try. Constructing a finite element model of the situation might make a good undergraduate project; shouldn't be too hard. Since the problem is essentially 1-dimensional, it might even be possible to do it analytically.

Thanks again.


On Mar 15, 2021, at 22:46, ILPI Support <info**At_Symbol_Here**ILPI.COM> wrote:

Bear with me, but I will try to put this one to rest.

The PDF listed immediately below is referring explicitly to the **traps**, which, as previously noted, are a well-recognized hazard and can pose a serious risk if not properly mitigated.  A vacuum trap, by definition, contains a vacuum rather than an inert gas.  When you shut the line down and introduce air into a cold trap, that indeed can condense liquid oxygen and since those traps usually contain organic residues, this poses a serious hazard (especially if one forgets to turn the pump off and continues to draw in additional air) which is what the author is warning about, NOT the Dewar itself.

The question asked in this thread was whether *the Dewars* themselves pose a hazard.  Which in all practical experience, if they do, the risk must be almost negligible or there would have been a documented case of an incident involving a liquid nitrogen Dewar sitting around at atmospheric pressure that condensed enough liquid oxygen to cause an accident. And the resulting lawsuits would have us putting all kinds of "Warning: flammable/explosion hazard" labels on liquid nitrogen Dewars, which we do not. While the evidence of absence is not necessarily absence of evidence, I think the latter is pretty much the case here.

The closest anyone here has come up with to a documented case of a safety hazard/risk was the unverified anecdote in the Scanning article Yaritza found (that resource is a great find, by the way - thanks for contributing that). While pure liquid oxygen will make combustible/flammable materials ignite more easily at 1 atm, the possibility that one could generate a liquid air mixture with sufficient oxygen under typical usage conditions seems rather unlikely.  Given the gross inaccuracies I've heard of even second-hand retelling of lab accident tales, I would have to call this story apocryphal at best.  But, certainly, the advice of a judicious choice of container for liquid nitrogen and reminder of the hazards of liquid oxygen are worthwhile.

So now I will put the dagger into this one:

Yes, liquid oxygen is great at facilitating the ignition of organic materials at room temperature (the liquid oxygen poured onto charcoal being a great example), although ordinary objects already at cryogenic temperature  will not ignite anywhere near as readily.  And we need an ignition source.  So, let's conjecture there was some lint and other organic lab dust in the bottom of the Dewar - I've seen that, so this is a good conjecture. But even if we fully combusted this dust by invoking autoignition, the volume of gases evolved would be minimal - it couldn't explode or cause a mass fire because you're talking about the ignition of a few milligrams of dust.  The fuel would be consumed almost instantly and the minimal heat evolved would boil off some of the liquid air. The magnitude of this would be less than the boiloff we see when filling a warm trap.  No fuel, no fire/explosion. 

So we try harder going back to the original question, so get out a bigger dagger:

We need fuel for this to work.  OK, we're going to pour this putative amount of liquid oxygen-enriched cryogen out on the floor because it's something we've all done in the lab at one point. Let's assume someone managed to condense 50 mL of pure liquid oxygen into our 400 mL Dewar - something I think is basically impossible, but let's run with it. The density of liquid O2 is 1.141 g/mL so we have 57 g of oxygen.  With a molecular weight of 32 g/mol that means we have 1.78 mol of O2. Which when we expand that to room temperature in a flash evaporation and use 1 mol gives 22.4 liters for an ideal gas, we have 40 liters of pure oxygen.  So off on the floor it goes. Some evaporates off immediately while beads of liquid air/oxygen roll around on the gas blanket, so that 40 liters is really maybe 10 or 20 liters of gas being introduced into some volume of laboratory air.

This gets hairy now because the temperature of the gas has a strong gradient, but let's assume we've dumped this out in our lab bay which has a surface are of 9 meters (about 100 sf in round numbers) and we let that gas rise a meter or so to mix, then we have initially put 20 liters of O2 into our nine cubic meters of lower laboratory atmosphere. Nine cubic meters is 9,000 liters, folks.  Air is is 21% oxygen, so this volume of interest initially contained 1,890 liters of oxygen and now contains 1,910 liters  Therefore, we went from 21% oxygen to 21.2%. Yes, there are some gradients, and yes, we still have more cold balls of evaporating liquid outgassing, but we also have our 9,000 cubic meters rapidly exchanging with the rest of the room.  So from a practical perspective, to use a technical term, big whoop.

So the only concern here we can possibly have is some tiny amount of high oxygen concentration somewhere along the floor encountering something that is combustible AND has an ignition source.  The ONLY way we can get there is something that is a flammable vapor already within its flammable limits and some sort of electrical spark.  And for our Dewar dump to contribute to this meaningfully, we had to have been just below the LFL (or we would have ignited already), and pushed the combustion equilibrium to the right with the addition of this putative oxygen.  So yeah, theoretically possible, but so is balancing the federal budget deficit. Actually, I think the latter is more likely.

Finally, I circle back with two observations about the effective of temperature on what are normally very combustible materials.  The same lessons apply to the scenarios discussed above:

When I first learned how to seal NMR tubes under high vacuum, the person performing that used a dry ice acetone bath. I commented on the wisdom of an open flame around acetone and he responded by putting the torch out in the dry ice acetone bath.  A somewhat insane but legitimate demonstration of flash point, flammable limits, kinetics, and heat capacity. 

When I was in grad school, I inherited a 30 mL vial we presumed had diethyl zinc in it because when I picked it up, it jetted out a beautiful blue flame before the tape around the rim crusted it shut again.  We know that dialkyl zincs react with carbon dioxide, so my solution to this lab quandary was to place the vial into a bucket full of dry ice and then smash it.  No flame, no cloud, no explosion because the thermal mass of the system and reduced kinetics kept that under control.

So, while we can speculate all we want, unless someone can produce a documented example of there being a risk let alone a hazard, I consider this one debunked.  And, regardless, the risk infinitesimal, particularly compared to others such as asphyxiation, frostbite, and pressure explosion.

Rob Toreki

Safety Emporium - Lab & Safety Supplies featuring brand names
you know and trust.  Visit us at
esales**At_Symbol_Here**  or toll-free: (866) 326-5412
Fax: (856) 553-6154, PO Box 1003, Blackwood, NJ 08012

On Mar 15, 2021, at 7:27 PM, Yaritza Brinker <YBrinker**At_Symbol_Here**FELE.COM> wrote:

I think you hit the nail on the head... good laboratory practice!

There are videos on the web that demonstrate this issue is real at RT 1atm, but they use an open container. There are also several websites, public SOPs, and LN2 manufacturer documentation that mention this as a potential problem. So, I wonder if the trap itself could be acting as a lid for the Dewar and effectively preventing this issue.

Here's a couple more links that may be of interest. I'll let you google for videos.

"Schlenk Line Design and Safety". The author also recognizes the potential Dewar issue and offers ways to mitigate the risk.

I also liked "Reactivity control using a Schlenk line" by By Tilak Chandra and Jeffrey P. Zebrowski, Journal of Chemical Health & Safety, May/June 2014.

Thank you,

Yaritza Brinker

-----Original Message-----
From: ACS Division of Chemical Health and Safety <DCHAS-L**At_Symbol_Here**Princeton.EDU> On Behalf Of Joseph DiVerdi
Sent: Monday, March 15, 2021 6:22 PM
To: DCHAS-L**At_Symbol_Here**Princeton.EDU
Subject: Re: [DCHAS-L] Condensation of liquid oxygen in open Dewar flasks of liquid nitrogen

** External Email **

In my (pretty long-term) experience working with lots of liquid nitrogen in a variety of situations I, also, have never witnessed accidental or unintentional enrichment of oxygen in the liquid cryogen.

In full disclosure, at the same time I have observed, even participated in, the intentional condensation of liquid oxygen by passing oxygen gas through a collection tube immersed in a bath of liquid nitrogen for various the purpose of thoughtfully and carefully arranged experiments. FWIW, it never spontaneously detonated. And yes, it was sometimes very exciting. :)

Plus, as was mentioned by another list member, there is very good justification why this process is discouraged by the blanket of gaseous nitrogen sustained by the active outflow of boiling nitrogen. (in other words, not only is the observed incidence of the process in question significant; a plausible mechanism explaining the observation exists.)

Yet, with the repetition of an anecdotal report "...We heard of an accident..." the suspicion, the story, the question is enlivened, revivified and sustained.

This thread began with a hypothetical (and doubtful) supposition that oxygen would spontaneously condense in an open-mouthed liquid nitrogen dewar followed by a questionable practice of discarding the dewar contents.

It seems (to this chemist at least) this is a non-issue for standard- and good-laboratory practice. Also, besides this anecdotal report the cited article was quite good and very informative to new experimentalists. I intend to use it with my students.


On Mon, 15 Mar 2021 1:25 PM, Yaritza Brinker wrote:
I worked with SEM/EDS using an LN2 Dewar to cool the EDS detector for about 10yrs. Never seen it happen. However, this older article does anecdotally report it happening and resulting in ignition.

"...We heard of an accident where ignition was due to the condensation of oxygen from room air in liquid nitrogen and subsequent elaboration of an oxygen-rich gas as the nitrogen boiled away. Because of the fire hazard, one should neither allow liquid N2 to evaporate to dryness..."

J. Bastacky, T. L. Hayes, Safety in the Scanning Electron Microscope
Laboratory, SCANNING Vol. 7, 255-272 (1985);data=04

Thank you,

Yaritza Brinker


*From:* ACS Division of Chemical Health and Safety
<DCHAS-L**At_Symbol_Here**Princeton.EDU> *On Behalf Of *Daniel Kuespert
*Sent:* Friday, March 12, 2021 10:25 AM
*To:* DCHAS-L**At_Symbol_Here**Princeton.EDU
*Subject:* [DCHAS-L] Condensation of liquid oxygen in open Dewar
flasks of liquid nitrogen

*** External Email ***

An interesting question occurred to me today: When you leave a Dewar flask of liquid nitrogen open to the atmosphere, such as when you're using it to cool the trap of a Schlenk apparatus, oxygen will condense in it since the boiling point of oxygen is 90 kelvins vs 77 K for LN. How fast this happens will depend on the air-exchange into the flask, so if it's covered loosely, presumably the dynamics will slow down. Eventually, though, the flask contains liquid air, not liquid nitrogen, which could become excessively exciting for someone who empties the flask by dumping it out somewhere near something combustible.

I've not seen any data, though, on how fast this actually happens. Has anyone ever seen data on this? If you have, please let me know. It would be useful to know how long one can really leave an open Dewar sitting around.

Dan Kuespert
Laboratory Safety Advocate
Johns Hopkins University
Daniel Reid Kuespert, PhD, CSP
dankuespert**At_Symbol_Here** <mailto:dankuespert**At_Symbol_Here**>
11101 Wood Elves Way
Columbia, MD 21044
Joseph A. DiVerdi, PhD, MBA
Associate Professor of Chemistry
Colorado State University, Fort Collins, CO, USA
+1.970.980.5868 -

For more information about the DCHAS-L e-mail list, contact the Divisional membership chair at membership**At_Symbol_Here** Follow us on Twitter **At_Symbol_Here**acsdchas

Electronic Transmission Confidentiality Notice
The information contained in this electronic transmission is private, confidential, the property of the sender, and intended for the use of the recipient(s), only. If you are not the addressee, any disclosure, copying, distribution or use of this information for any purpose is strictly prohibited. If you have received this information in error, please notify the sender, YBrinker**At_Symbol_Here**, immediately by e-mail and then delete this message. Thank you.

For more information about the DCHAS-L e-mail list, contact the Divisional membership chair at membership**At_Symbol_Here**
Follow us on Twitter **At_Symbol_Here**acsdchas

--- For more information about the DCHAS-L e-mail list, contact the Divisional membership chair at membership**At_Symbol_Here** Follow us on Twitter **At_Symbol_Here**acsdchas

Previous post   |  Top of Page   |   Next post

The content of this page reflects the personal opinion(s) of the author(s) only, not the American Chemical Society, ILPI, Safety Emporium, or any other party. Use of any information on this page is at the reader's own risk. Unauthorized reproduction of these materials is prohibited. Send questions/comments about the archive to
The maintenance and hosting of the DCHAS-L archive is provided through the generous support of Safety Emporium.