Monday, September 01, 2008

Anhydrous Solvents Part 2: The THF Still - Do's and Don'ts

The THF still is one of the most common reasons for serious fires at chemistry departments around the world. As solvent purification systems (SPS) become more and more common this problem will gradually disappear. However for now SPS is still pretty expensive so the THF stills will linger around for some time to come. There are many reasonable ways to construct a THF still. All you need for a basic THF still is:
  • A round bottom flask
  • A still head
  • A reflux condenser
  • An inert gas inlet and gas bubbler
  • A heating mantle
  • Sodium, benzophenone and some THF
The main reason why THF stills blow up is incompetent users. It only takes one sloppy chemists to screw things up in a big way. However, even with competent, careful chemists THF stills occasionally blow up. So what can you do to avoid the THF still disaster?
First and foremost train all users and/or limit the number of people allowed to turn it on and off. Have an on/off sheet where people record their name, time etc.
  • Get a flow-stop switch - This clever little gizmo shuts the power off if the water stops running. This will stop most accidents before they happen. Your electrician will know what to do.
  • Get a "hat" for your heating mantle (see picture) to avoid water dripping in and setting the whole thing on fire.
  • Use Teflon joints or Teflon sleeves to protect joints from fusing (see picture)
  • Don't use excessive amounts of sodium. Ideally use a sodium press to make thin sodium thread. This will give you a large surface area and ensure that you don't end up with a big fat ball of sodium when you eventually have to quench the still.
  • Never turn the inert gas off.
  • Have enough THF in the system. Always ensure that the round bottom flask (RBF) is filled sufficiently with THF to allow the still head to fill up and leave the RBF at least on third full.
  • ONLY fill the THF still with dry THF. Don't just grab any old THF bottle from the solvent cabinet. Have THF bottles specifically for the still only.
  • Don't put a stirring bar in the RBF. This will only result in a big fat ball of sodium down the track. Sodium wire is generally enough for controlled boiling. If you must add something throw some boiling chips in.
  • BE PATIENT! Don't turn the heating mantle to the max because you are in a hurry. It take 30-60 minutes to get a THF still up and running in a controlled fashion.
  • Place your heating mantle on a lab jack. When the shit hits the fan you can lower the heating mantle and stop heat transfer immediately.
I think that is enough for today. There are many things to consider and I have by no means covered everything but this should be a good start. D!


Bunsen Honeydew said...

Great topic for a post. The one thing I have not been able to find is a sodium press. Any suggestions?

Rifleman_82 said...

Hmm... my lab's old THF still was set up similarly, but we had somewhat different procedures. It's been running for years without any mishap, and I feel strongly there are other ways of doing it safely.

We NEVER left the THF still unattended. That is the most important safety control, IMHO. Whenever it was used, the operator was always checking on it every now and again.

Instead of a mantle, we used an oil bath, raised on a lab jack as necessary. When THF was needed, we turned on the heating stepwise, and only sufficient for the mixture to reflux briskly. Since THF boils at 66 C (+ a bit due to boiling point elevation), while sodium melts at 98 C, there should be no problem with using magnetic stirring. The decomposition crap tends to be tarry, which will quickly render your boiling chips ineffective.

N2 gas was turned on only when we were heating. With the no-return bubbler, it was quite a waste of inert gas to bubble when no heating was taking place.

We allowed the THF to reflux for 15 minutes or so, to wash any dissolved O2 or H2O down to the still to be destroyed, then closed the stopcock and started collecting THF. Accidents have happened where the stopcock was closed, and the unattended pot was heated to dryness.

While we didn't specifically reserve dry solvent bottles for the still, we pre-dried a sufficient quantity over sodium wire/pellets overnight before adding them to the still. That should decompose any peroxides.

In any case, if you do purchase dry, high-purity THF packed under N2, you can simply send it into the glovebox and use it as received.

Daniel Sejer said...

Bunsen Honeydew, if you don't have a sodium press then I believe that you will have to cut sodium pieces of with a knife. One place I worked they had a metal rolling pin that they would use to make really thin sodium sheets. Nothing really beats the sodium wire I'm afraid. D!

Rifleman_82 said...

One alternative I've seen, to sodium wire, is to cut chunks of sodium, put it in a clean plastic bag, and mash it into thin flakes with a mallet. If desired, cut them into ribbons using a pair of scissors, and you're done. In comparison, this method is much faster than turning the crank on the sodium press.

Daniel Sejer said...

Rifleman, thanks for taking the time for your extensive contribution. There are many variables. The most critical is the users. Are they responsible or not? Some places there will be countless users on several floors and not leaving it unattended becomes impossible. In fact not forgetting about it completely can become difficult if you haven't been using it yourself for a long time.
- I have had a fair bit of experience with the stirring bar in the still. I seriously don't see the point. What are we trying to achieve by stirring it? I have regenerated more stills than I like to think about and the ones with the stirring bar nearly always contain a potential fireball.
- I agree that boiling chips are more or less pointless but if it makes people relax let them go for it.
- I sincerely believe in never turning the N2 completely off. When someone accidentally has done it in the past the still content will go brown within a few days shortening the stills lifetime.
- I agree that the THF has to gently reflux for a while to ensure that everything is nice and dry before you start collecting
- AR grade THF is more or less dry and contains a stabiliser. Open it pour it into the still close the bottle and write THF still only all over the label. No peroxides to worry about and no pre-drying required.
- However, the bottomline is that I can't wait to have a SPS installed in my lab so I don't have to worry about this stuff anymore. D!

Rifleman_82 said...

I just noticed your picture close up. Nice idea about your receiver - it's now impossible to heat to dryness! We used a modified pressure-equalizing addition funnel, which afforded no such protection.

How do you fill the pear-shaped flask? It seems that the solvent will flood the stopcock, and without a way to relieve the pressure, the flow will quickly stop. Also, how do you dry the pear-shaped flask? Does not seem realistic to connect the entire still to a Schlenk line and vac-refill? Or is this for a less-demanding application?

What is a teflon joint? Or is it a regular standard taper joint with teflon sleeves?

Daniel Sejer said...

Rifleman, I can't really take credit for the still. I only started working here in July. In fact I would have constructed it differently but it works and it is closely monitored so there is no problems. Only three people are allowed to turn the still on and off to prevent problems. Not that I know for sure but I assume that the pearshaped flask is just to keep things fairly dry in there and to avoid any dripping in case of a leak. I'm sure that people attach their own dry flask when they want some. So far I have been getting all my THF out with a dry glass syringe. I'll leave it to the guys that built it to elaborate if they feel so inclined. Teflon sleeves are more than adequate but you can get what is essentially a Teflon tap adaptor. In this case it converts a B29 joint to a B24 with a B24 reflux condenser attached. If you click on the image it will enlarge and you may be able to see what I'm talking about. Personally I would go for the sleeves. Cheaper and more versatile in my opinon. D!

milkshake said...

we have been using Na dispersion in paraffin wax - it is commercially available (from Fluka, or it can be made very easily), it cuts with a knife like a chocolate without fire hazzard and due to huge surface area the benzophenone ketyl color develops promptly, with one hour. The wax does not distill over and the added benefit is that if someone forgets to re-fill the stil there is leftover wax to coat the sodium.

The best way to operate THF stil is to have a dedicated hood for it, and one person who is authosised to take care of it and train the new users.

There was a nasty ether explosion in Prague with ether stil because they were using heating mantle and someone messed with the variac regulators, etheether overboiled and spilled in the hood before the ether vapors ignited, it blew the door to the next lab where some elderly distinguished chemist worked at the bench. Nobody got hurt but the explosion made impression on the old man, and few weeks later when a LARGE lady from administration visited his lab with some paperwork, and sneezed behind his back, the old man immediately took cover, ducking behind his bench like a barman in a saloon

Anonymous said...

Last year we changed our 5 Litre THF still 2 times so each one lasts ~ 6 months if properly maintained.

Most important is pre-drying THF we use to fill it up with.

We use sodium ribbon, our wire press always gets clogged up. Plus in my experince the wire does not last as long

Daniel Sejer said...

That sodium wire doesn't last as long makes no sense. Of course it does. It has to react to stop working. This is why sodium wire is great. It gets consumed so that when you regenerate the still you are less likely to have a fire. With wire you tend to use a lot less than by other methods which is why it appears to last for a shorter time. Just add a bit more after a few months. How a sodium press gets cloged up is beyond me. Obviously you have to clean the die straight after use but provided this is done every time a functional press should not clog up. D!

Anonymous said...

This Na dispersion in paraffin is interesting. Can you use it to make sodamide? Or is commercial grade sodamide perfectly acceptable? I want to take 2-pentynoic acid to 3-pentynoic acid through the dianion. Apparently, when it reprotonates, it's almost exclusively as the 3-regioisomer. Supposedly you do it with 2 equivs NaNH2 in ammonia, but I must say I'm dreading this reaction.

milkshake said...

I would buy a good grad of NaNH2 from Acros, condense some NH3 from a tank to it and there you go.

If you do it ftom Na and liq NH3 and Fe(III) catalyst you dont need any dispersion, Na chunks dssolve in NH3(l) quite fast. I am sure if you search sodium amide in Org Syn you will get a reliable, detailed write up. Dont modify it.

broland said...

here's a nasty mistake i saw. some kook was visiting our lab and decided to use a clamp to close off the hose that connected his N2 line after he was done distilling. i don't know what he was thinking (maybe he wanted to get solvent out of it later???) but the next time he went to distill toulene, he forgot that the line was clamped and funny thing about a closed systems like that is will explode. nasty.

just never clamp a gas line. N2 is cheap, leave it at low flo.

Alexey said...

I think the idea with Na dispersion is great and I will probably switch ot it from now on.
But if you don't have Na dispersion, one guy taught me to add a few pieces of "soft" glass (from Pasteur pipettes) and a stirrer to the still. It scratches sodium and almost does not scratch the flask. So, if I set up the still today and need to get blue ether,THF or benzene tomorrow in the morning, I let sodium foil to stir with glass at rt overnight. (We make a "sodium foil" just by fast rolling with a test tube on a bench). Normally, after this procedure on the next day 1-2 h of reflux is enough to get dark blue color. Although refluxing toluene melts sodium, scratching overnight at rt accelerates getting blue color on the next day as well.

Anonymous said...

I don't know what all the fuss is about...we cool kids use Grubbs-type solvent systems. I think it's way safer, although I suppose you might argue about whether it's the same quality solvent. At my school, we replaced all our alkali metal-based stills with the columns. We still use stills for amine bases and such, but it's typically calcium hydride.

Daniel Sejer said...

Cool, rich kids would be more accurate. Many departments simply don't have the money for a Solvent Purification Systsem (SPS). Personally, I would love to have a SPS and ultimately everyone will be using these systems. If the safety people at Universities around the world would go to the Heads of Departments, Vice Chancellors etc. and compare the cost of SPS to rebuilding a burnt out Chemistry Lab I'm sure SPS would be introduced much faster. D!

Anonymous said...

we have a SPS here but the THF from it is dodgy. THF will also polymerise on the column and break the solvent system, so, in my experience (requiring properly dry, not organic dry, solvents, as I'm an organometallic chemist) still-dried THF is still necessary.

mike said...

I agree that a communal THF still is potentially dangerous if set up wrong and no one decommissions it regularly.
I always use Na dispersion in oil now with good pre-dried THF.
Fisher HPLC grade THF is pretty dry to start with. A 2/3 full 2 L still with approx 5 g Na/oil starts off green/blue and is that lovely purple colour after a couple of hours refluxing and ready to use. The best part about Na/oil is it is so easy and quick to quench the still with EtOH and then water. There is no residual "crusty Na" waiting to catch fire.

Anonymous said...

I have heard that the potassium is much superior to sodium, both in terms of efficancy and handling. Since it melts at the boiling point of THF it always provides a clean active surface. No need for cutting or pressing, just add a lump of metal. Moreover, after cooling it form a single lump that can be fished out prior to clean up of the still, no need for dangerous quenching. Fish it out and add it back when cleaning is finished.

Anonymous said...

we have spare always dried rbf for THF still. Just in case we need to quench still (Na) then we replace it with spare new one.