National Chemistry Instrumentation Center (NCIC)
Ateneo de Manila University phone: (632) 924-4601 loc. 2309
Loyola Heights, Quezon City e-mail: etch@ginto.chem.admu.edu.ph
Philippines
------------------------------------------------------------------------------
>Dear AMMRL'ers,
>
>I'm trying to set up a vacuum line for evacuating NMR tubes prior to sealing
>with a torch. What vacuum pressures are sufficient to remove atmospheric
>oxygen from the sample? Also, I'd like to know what various methods are used
>to remove oxygen from the sample: I've heard of freeze-dry methods where the
>sample in the tube is frozen in liquid nitrogen and the tube is evacuated and
>sealed, and also bubbling inert gas (nitrogen and argon) through the
>liquid sample prior to sealing. Anybody out there willing to share their
>opinions on the efficacy of such methods?
>From FOERSTERLING@novell.chem.utk.edu Sat Jun 29 22:00:45 1996
We usually use a diffusion pump with 10-4 torr or less. We normally
freeze the sample in liquid nitrogen, pump down on the frozen
solution, disconnect from the pump and let it thaw. This cycle is
repeated until the gauge doesn't move significantly above 10-4 torr
when the frozen solution is exposed to the vacuum pump.
We normally degass the solvent in a bulb and vacuum transfer it onto
the solution, thus we normally have only few cycles with the actual
NMR tube, reducing the chance of the tube breaking upon thawing.
Hope this helps.
Frank H. Foersterling
**************************************
* Frank-Holger Foersterling *
* Department of Chemistry *
* University of Tennessee *
* Knoxville, TN37996 *
* home phone : (423)-558-9754 *
* e-mail : holger@utk.edu *
**************************************
>From angel@sigma.chem.uic.edu Sun Jun 30 00:57:17 1996
Bubbling an inert gas into the sample is probably the best way of removing
oxygen. Of course, you can still follow up with freezing at liquid N2 and
pulling a vacuum. This should take care of all the oxygen, as long as the
freezing will not destroy your sample. Otherwise, you should just bubble
with an inert gas.
If you use a roughing forepump (an ordinary vacuum pump) with a
liquid nitrogen trap between the sample tube to be sealed and the pump,
the vacuum should be sufficient (This should be in the order of a
Torr.) But this is already doable. What is important is that the
tubes need to be narrowed down at the seal point prior to the actual
sealing. You should do this before putting any sample into the
tube, make the seal point narrow enough and then anneal the tube inside an
oven or furnace. With this treatment, the tube can withstand 50 atm,
without it, 2 atmm can easily blow the seal.
-angel
Georgetown University
>From akj@sigma.chem.uic.edu Sun Jun 30 01:23:15 1996
Hi Edward. Angel asked me to tell you how I deoxygenate samples. It
depends:
If the sample is just an ordinary solution to be run fairly quickly
(perhaps 1 hour) then simply bubble dry N2 thru the solution for 2-3
minutes. This displaces dissolved gases and purges the overhead of
all oxygen. Capping will limit the oxygen within the tube and ordinary
diffusion through a liquid down to the receiver coil takes a LONG time!!
To do the job right on a permanent basis one can put the sample into a
previously prepared tube with the top 1 cm. fused to a 4 mm piece of
ordinary Pyrex glass tubing. The pyrex tubing can have a constriction
previously made with a torch (for sealing afterward--I like to anneal
this well, but its not really necessary), so that when the sample is
placed into the tube the top has a 4mm od piece of tubing portruding
with a constriction perhaps 1 cm from the top of the original nmr tube.
Then the purging of the previous paragraph will remove dissolved gases.
simply pulling a GENTLE vacuum while sealing off will yield a sample
which is permanent, has an inside pressure below atmospheric, will spin
nicely, and can be reused if desired (break open the top, remove the
sample, clean the tube, dry it, and seal another small section of 4mm
od tubing to the top.
Hope this is what you want.
Keith
>From roy@cc.huji.ac.il Sun Jun 30 14:15:51 1996
Dear Edward,
Regarding deoxygenation of samples. The main purpose of deoxygenating NMR
saples is to remove the paramagnetic broadening of oxygen and to increase
longitudinal relaxation times to make NOE easurements easier.
We use a high vacuum line and the freeze pump thaw method for our samples.
The high vacuum that we use is probably more due to the sensitivity of our
samples to oxygen than any NMR consideration. I have not done any detailed
studies on the efficacy of deoxygenation but I am told that 1) Three
freeze-pump-thaw cycles with a good roughing pump is sufficient for NMR.
2) bubbling argon of oxygen-free nitrogen through the soltion is also good.
However, argon is better as is is heavier than air and displaces it from the
tube better. Organic solvents reqiure about 30 seconds bubling and water
about two minutes. Consideer the above as rough guidelines.
Yours,
Roy.
--------------------------------------------------------------------------------
V
I Dr. Roy Hoffman,
I Manager, NMR Laboratory, Department of Organic Chemistry,
A a X The Hebrew University of Jerusalem, Jerusalem, Israel 91904.
XJ lX Tel. (972) 2 6585327, Fax (972) 2 6585345, Email roy@shum.cc.huji.ac.il
X X Web page http://drx.ch.huji.ac.il:8080/hoffman.html
X
--------------------------------------------------------------------------------
Date: Mon, 1 Jul 1996 08:52:08 -0500
From: cgregory@uiuc.edu (Carl Gregory)
It all depends on how much O2 you can tolerate.
The classic reference on this subject is:
The manipulation of air-sensitive compounds
D.F. Shriver, M.A. Drezdzon.
2nd ed.
New York : Wiley,
326 p. : ill. ; 24 cm.
1986
Carl Gregory
Biomedical Magnetic Resonance Laboratory
College of Medicine
University of Illinois
1307 W Park St
Urbana, IL 61801
(217)-244-2350
Date: Mon, 01 Jul 1996 08:10:54 +0000
From: Jim Frye <jimf@newage.otsuka.com>
"Freeze-pump-thaw" is effective and efficient. Use several cycles.
"Purging" works but often leads to unacceptable loss of solvent. If
you purge water, it is best to use helium gas if you will raise the
temperature. Unlike other gases, helium becomes more soluble in water
at higher temperatures, so bubbles will not form. If you will lower the
temperature (e.g., for biological or other heat-sensitive samples), use
argon or nitrogen.
The most convenient pressure range for sealing tubes with a torch is
0.5-0.7 atm. This provides a gentle inward force to help seal the soft
glass. A harder vacuum generally causes the seal to crumple, and 1 atm
causes the seal to bulge as the gas is heated. Use a small flame and
practice on a large number of cheap tubes and disposable samples first.
It is an art. Asymmetric seals will spin poorly, generate large
spinning sidbands and may scratch the glass lining in your probe.
Consider using one of the torchless reuseable sealing systems offered by
NMR tube vendors such as Wilmad (http://www.wilmad.com). The tubes are
expensive initially, but so is your probe and the time of a skillful
glassblower.
-- Sincerely yours, Jim Frye email: jimf@otsuka.com Web site: http://www.otsuka.comOtsuka Electronics USA Inc. phone: (970)484-0428 2607 Midpoint Dr. free phone: (800)4-OTSUKA Ft. Collins, CO 80525 FAX: (970)484-0487 USA direct line: (970)484-1346 x244
From: lskan@chem.sinica.edu.tw (Lou-sing Kan)
Dear Dr. Chainani:
I assumed that you would like to remove the oxygen dissolved in the sample solution in an nmr tube, not the empty tube itself. The organic solvents could dissolve sizable amount of oxygen that may cause trouble. Yes, you can use freeze and thaw method to get rid of the dissolved oxygen.
The procedures are quite simple: 1. freeze the sample by liquid nitrogen, 2. open the stopcock and pull vacuum, 3. close the stopcock and thaw the sample, Repeat 1 - 3 as you wish. Usually twice are enough, one time gives good result too. 4. freeze the sample and open the stopcock and seal the sample by a torch. You may need some practice for step 4. You don't have to have a high vacuum pump to do it.
Extreme cautions have to bear in mind if you want to get rid of oxygen or other gas in aqueous solution. The nmr tube, especially 5mm or smaller, may crack during the freezing process. By the way, the solubility of oxygen in water is low (ca. 5%) and it can be bubbled out by nitrogen.
Please feel free to contact me if you have more questions. Good luck
Lou-sing Kan, Ph. D. Professor of Chemistry Institute of Chemistry Academia Sinica Tel: 886-2-788-4184 FAX: 886-2-783-1237
From: Robert D Scott <scott@iastate.edu>
Edward,
I don"t think that the pressure is critical when you freeze the sample, you just want to remove most of the gas when it is frozen. It is important to remove the pump (clamp the hose) when melting the sample. I usually do two or three freeze thaw cycles before sealing the tube while it is frozen and have been happy with the results.
hope this helps,
dave scott iowa state university 515-294-4057
>From dmiller@hydrogen.ucsc.edu Wed Jul 10 00:28:19 1996
Hi Ed,
Sorry for the time frame, I've been on vacation. As a glassblower I have had to seal many NMR tubes. In most cases I will use house vacuum 1x10-3 and purge sample with nitrogen. Prior to sealing I then freez sample with liquid nitrogen while pumping to prevent any outgasing of the sample while trying to seal it off. In most cases this has worked well in removeing the oxygen from the samples. Sounds like the people you have been talking to are on the same track as myself. If I can be of any help let me know.
Regards, Don
>>
Date: Fri, 12 Jul 1996 17:42:29 -0500 From: "G. Pearson" <gpearson@blue.weeg.uiowa.edu>
> Date: Sat, 29 Jun 1996 15:21:15 +0800 (GMT+0800) > From: "Edward T. Chainani" <etch@ginto.chem.admu.edu.ph> > To: AMMRL <ammrl@bloch.cchem.berkeley.edu> > Subject: Vacuum line > > Dear AMMRL'ers, > > I'm trying to set up a vacuum line for evacuating NMR tubes prior to > sealing with a torch. What vacuum pressures are sufficient to remove > atmospheric oxygen from the sample? Also, I'd like to know what various > methods are used to remove oxygen from the sample: I've heard of freeze- > dry methods where the sample in the tube is frozen in liquid nitrogen and > the tube is evacuated and sealed, and also bubbling inert gas (nitrogen and > argon) through the liquid sample prior to sealing. Anybody out there > willing to share their opinions on the efficacy of such methods? Hi, Edward. You can removed disolved oxygen from samples in an NMR tube by bubbling either N2 or Ar through a capillary inserted to the bottom of the NMR tube. You need to bubble enough gas through the liquid to evaporate a noticable volume of the solvent. Any TMS in the sample will be long gone. MHDS (hexamethyl disiloxane) is a suitable proton reference, and is far less volatile than TMS. 3 freeze/pump/thaw cycles on a high-vacuum manifold will do a good job. I would not trust a roughing pump when doing this. I would expect that the "vacuum" created by a roughing pump would contain crud which could get into your sample and condense at 77K. You might get away with using only a roughing pump like this: * remove O2 by bubbling an inert gas through the sample, * freeze it down (dry ice or liq. N2), * pump a _partial_ vac. on the tube with a roughing pump, * valve off the connection to the roughing pump, and * seal the sample. When I seal off samples I like to leave a fraction of an atmosphere pressure of an inert gas above the liquid. This prevents refluxing in the NMR tube when the sample is above room temperature. If the sample is at 77 K when you seal it off, you want to make sure that the inert-gas pressure is 1/4 atmosphere or below, so that there is no chance that the NMR tube could explode when you warm it up. This is my favorite recipe: valve M +-------------------------------------------\___ ____ to liq. N2 | high-vac manifold, volume = Vm ___ X____ --> trap & +---+ +-------------+ +---------------+ +---/ diffusion pump valve B |X| valve S |X| valve I |X| | | | | | | | | | | | | +--+ +--+ | | | |______ | | | | NMR |________ <--- inert gas | | | | sample such as N2 or Ar +-------+ | | tube ballast bulb |_| volume = Vb You want Vb = Vm _very_ _approximately_ (within a factor of 4 is OK) valves (O=open, X=closed) B S I M OPERATION ----------- --------------------------------------------- X X X X hook everything up O X X O pump out manifold & ballast bulb. X X X O ----------------------------------------------------+ X X X O freeze sample with liq. N2 } freeze/pump/thaw X O X O pump off air } cycle -- repeat X X X O let sample thaw out; dissolved gas } 3 times bubbles out of the sample } ----------------------------------------------------+ X X X O freeze sample with liq. N2, and keep it frozen during the remaining steps. X O X O pump off any remaining trace of air X X X O O X X O evacuate ballast bulb X X X X X X O X admit about 1 atm. inert gas to manifold X X X X O X X X divide inert gas between ballast bulb & manifold X X X X trap some gas in the ballast bulb X X X O evacuate manifold X X X X O X X X divide inert gas between ballast & manifold O O X X admit low pressure gas to sample; seal off sample Final gas pressure is Pf = Vb Vm / ( Vb + Vm )^2 If Vb = Vm, then Pf = 0.25 atm. But even if Vb and Vm differ by a fairly large factor, Pf not much below 0.25 atm. For example if Vm = 3.0 Vb, then Vf = 0.1875 Hope this helps. -- Gerry --------------------------------------------------------------- Gerald A. Pearson INTERNET: gerald-pearson@uiowa.edu Chem. Dept., Univ. of Iowa OFFICE: 319-335-1336 Iowa City, IA 52242-1219, USA LABS: -1332, -0073 FAX -1270 ---------------------------------------------------------------