Sciencemadness Discussion Board

Solvent Pot Problems

Mumbles - 11-8-2011 at 21:59

I've been put in charge of the solvent stills in the lab I work in. About the only one we really run anymore is a sodium/benzophenone THF still. It has been behaving extremely oddly. No one I've talked to locally seem to have any ideas or suggestions, not even the PI.

It is a 3L still that we fill with unstabilized Fischer Optima grade THF. A fresh still with freshly cut sodium (~10g cut into less than 1/4" cubes) and benzophenone turns the charactaristic blue color after a few days of refluxing under N2. I can distill it and collect it and all is well in the world. The problem comes when I try to refill it. Typically I top it off with fresh THF. I then attach a septa, bubble it with N2 for 20 minutes or so, re-attach the glass stopper, and heat it to reflux. The problem is that every time I do this the whole thing turns orange. I currently have what looks like a boiling pot of orange juice at the moment. Fresh bottle, partly used bottle it makes no difference. I'm currently going to let this one reflux for a while to see if it improves. Typically the post doc dumps it out, and we have to quench the sodium, clean everything, and start from scratch. This is impossibly inefficient, and make getting anything requiring dry THF a chore.

Does anyone see anything wrong with my methods, or something I am over looking? I've never heard of grease on the joints, or teflon tape or anything doing this. Should I maybe transfer in the THF through a septa with positive nitrogen pressure instead of opening it briefly, even with a strong N2 flow?

About the only reference to this problem I've found is on UC Davis' website, and all it says is to regenerate it. It really should not be required to regenerate every time we need to refill the still.

smuv - 12-8-2011 at 00:11

Add more sodium, if that doesn't work add more benzophenone. It shouldn't take you more than a couple hours reflux to get the purple/blue color. Might be worth while to determine how much water is in your THF. There are better ways of doing it, but i would just NMR a CDCl3 blank and then 5% THF/CDCl3 and compare the integral of the water peak vs residual chloroform (to scale by 0.95 in THF sample).

Also, if you just leave the still be, under inert gas, will it stay blue? or will it go green (or orange?).

Fleaker - 12-8-2011 at 08:13

More sodium is what's needed. Greasing the joints with silicone grease is a bad idea; they make this wonderful grease called Krytox. Works a treat for when you don't want to see the Corning silicone in your NMR spectra.

You guys might consider just installing a solvent purification system (it's basically sieve + basic alumina). They work well enough that LAH going into the freshly purified THF is quiet.

Mumbles - 12-8-2011 at 09:03

We have a Grubbs type solvent purification system for everything except THF. You know, everything except the solvent we easily use the most. I've just been going off of how I was shown, but that clearly isn't working so I'm looking for something new. Any silicone grease that washes in seems to stay in the pot after distillation of what we need, but I'll give it a shot.

The solvent does not appear to be extremely dry to begin with. There is a bit of noticeable bubbling upon adding fresh sodium to the pot when making it fresh. Someone else I was talking to about this recently mentioned that they pre-dry their THF with calcium hydride, perhaps that is worth a shot as well. I realize that it should only take a few hours to at the most overnight, but right now I just need to make it work reliably.

Yes, if I leave the still to it's own devices after it's functioning properly, the solvent stays dark purple/blue essentially indefinitely. There have been some that easily stayed fresh for 6-8weeks while we were at a low in use. That is about as long as we can go on 2L or so.

fledarmus - 12-8-2011 at 12:31

If a significant amount of water gets into the still all at once, it can coat your sodium with NaOH and deactivate it. Usually adding some fresh sodium or cutting up some of the old sodium with a spatula will start it back up, but if this happens every time you add fresh THF, I'm guessing your THF is much too wet. We always used commercial anhydrous THF to refill the still, and filled it by cannula through a septum. That way we never had to break the nitrogen atmosphere, either. If you're using normal solvent grade THF, definitely pre-dry it before adding it to your still.

ziqquratu - 12-8-2011 at 21:33

We have this problem a bit, too. Fledarmus is correct, though - too much water results in coating the Na with NaOH and reduced benzophenone scum, but if you can remove that coating, it should be all good again. You can do so with a spatula or a rod, and while it's fiddly, it works. For some stills - depending on size and particularly if you use reasonably fine Na wire - you can also throw in some chunks of broken glass and shake the crap out of it to scrape some of that coating away. (By the way, wire is better than chunks - larger surface area = harder to inactivate the surface; in a pinch, you're better to smash your chunks with a hammer [under an inert atmosphere! A nitrogen-flushed ziplock bag works well] into thin sheets and use those).

Adding more Na to the still is rather pointless (most of the time, there's plenty of active metal still in there, it's just coated). It can also make life more dangerous when it ultimately comes time to quench the still - because you're only using a small amount of the available Na each time you refill, there's always a HUGE amount of unreacted metal remaining, but coated with crap... leading to a greater chance of a fire when you quench it.

For preparing the THF before adding, the best thing is to dry it over molecular sieves for a day or two to remove the bulk of the water, then to sparge with dry N2 for 20-30 minutes BEFORE adding it to the still (from memory, water decomposes the ketyl stoichiometrically, but oxygen does it catalytically, so in fact it's far more important to get rid of oxygen from your solvent in order to avoid decolouration of the still!). Ultimately, try to have the THF you're adding as close to dry and degassed as possible, so you don't keep re-coating the Na.

Also, we've found that, once the benzophenone concentration reaches a certain level, we get the ketyl distilling out with the solvent, contaminating the collected stuff (but, on the plus side, at least we can see it's contaminated!). At that point, there's no choice but to make a new still. To avoid this, keep the benzophenone to a minimum, and only add more to a decoloured still as a last resort (ie. after trying the scraping and refluxing at least a few hours).

The other tip would be to make your still as large as possible (ours is 3L, would be 5 if I could find a suitable flask!) so you minimise the frequency of refills.

Oh, one more thing - when making a fresh still, reflux it a couple hours to thoroughly dry the solvent with the Na BEFORE you add the benzophenone. That way, the amount of scum deposited on the Na is minimised from the start (and you'll get more refills before needing to remake it). Also, the still should then turn blue almost immediately on heating it again!

[Edited on 13-8-2011 by ziqquratu]

Fleaker - 16-8-2011 at 20:50

Sodium wire is what I've been wont to add, not Na chunks. Higher surface area relative to volume. Of course, Na shot also works but it sounds like your very wet THF would be a bad idea to introduce. I meant the fluorogrease comment more as for the stopcocks/ground glass on the output.

Like mentioned above, pre-drying it wouldn't be a bad idea, even if it's just running through a column of CaCl2. It is oxygen that destroys the ketyl radical. Great advice from ziqqaratu.

I'm not a fan of stills, especially big stills. Can go very bad very quick (think fire) with idiots at the helm, and it's just more metal to quench with IPA.

l-PEI - 9-3-2014 at 02:24

The orange colour you see is the polymeric dianion of benzophenone. This is good, as if the still goes completely anhydrous the excess sodium will react again. The problem is it is no longer an indicator. Either increase the quantity of benzophenone added or decrease the quantity of sodium.

Stirring is more effective than refluxing, and the best system is the use of potassium vice sodium (it melts), and stirring the THF pot at reflux. The benzophenone should go blue in less than a minute under those conditions. Your stir bar will go black due to attack on the fluoroether, but this isn't a problem.