Monday, June 8, 2009

nice set up

I'm pushing a bunch of material through a known reaction that, before my boss and I improved it, sucked. It worked very well on small scale and the yield and quality of the product dropped off precipitously as it was scaled up. Now the reaction can be performed on most any scale and yields of purified product are typically 90% or above.

Now there is only one small problem, the solvent for the reaction is triglyme. I run the reaction quite concentrated, but I still have hundreds of milliliters of solvent to deal with that has 50-100 g of crude product in it. Extraction on this scale is doable, but vacuum distillation is not only easier and faster in this case, it leads to significantly more pure product.

Herein lies the issue, triglyme is a bitch to distill. It has a relatively high boiling point, typically 80-100 centigrade with the vacuum my pump generates, and is prone to bumping. By prone to bumping I mean that the shit bumps almost continuously when contaminated with a product mixture, and treated in the fashion I usually treat short path distillations.

The first time I ran this reaction at 50 g scale in 500 mL of triglyme, I distilled a significant portion of the solvent off starting with the 1 L flask the reaction took place in, rinsed into a 250 mL flask, and removed the remaining solvent, rinsed into a 100 mL flask and distilled the product. There was some bumping over of the product, but not too much, and the reaction-purification sequence took all day. All in all not too bad for a 50 g reaction, but I still needed to put a few hundred more grams through it in the coming days.

I remembered a set up my boss had mentioned to me in another context once that he used for distillation of a heat sensitive compound. Many people are aware of the falling film distillation apparatus, and they are available from places like Aldrich for the low low price of $800 for the lovely 24/40 model.

an easily broken paper weight that cost almost as much as an ounce of gold

When my boss needed to do some sensitive distillation a long time ago, he made up an exceedingly useful contraption to effect very similar conditions using common lab glassware. My target compounds are plenty stable to heat, but I thought the apparatus would be usefull to deal with the bumping issue. Here is my practice run with the apparatus.


It's a 3-neck 250 mL rbf fitted with a stopper, a pressure equalizing addition funnel, and a short path. Notice that the addition funnel has a glass stopcock, teflon stopcocks do not work well at all for vacuum applications. The oil bath is heated to ~150 centigrade and the mixture is added to the hot flask a little at a time. It distills rapidly, and there is never a significant amount at any one time so the bumping is inconsequential, additionally I spin the shit out of it (1,000-1,200 rpm) which helps quite a bit. This pic is a practice run, and contains the solvent from the 50 g reaction containing a trace of product, hence the slight color. I was able to distill off 500 mL water white solvent leaving the crude product in 30-40 minutes this way.

When I ran the reaction on 100 g scale I was able to distill off all the solvent, and then distill the product in less than half the time the first 50 g reaction distillation took.


Jim said...

So if you avioded the teflon stopcock why use a teflon thermometer adapter? Do they hold a vacuum?

scientist 1 said...

Unfortunately it was all I had. My preferred short path, with a small vigreux section and joint for a ground glass thermometer, is out for repair.

The teflon thermometer holders do hold vacuum well when the o-rings are in good shape, but I don't think I would trust one well enough to distill something pyrophoric.

Dr. Miller said...

That is cool. While I hope that I never have to distill 1L of diglyme, at least now I know a good way. That is a super easy solution that is not at all obvious. Nice.

james said...

Hmm... I wonder if that works as well with water

scientist 1 said...

It should, although the hi-vac is probably a bit much. Perhaps aspirator vacuum or adjustable pump might be more appropriate.

I've been having good luck with water on my rotovap lately. I try to make sure the flask is less than half full, and pull the vacuum down slowly in many small increments.

james said...

the best luck i've had with water is having a flask that looks way too large, and pulling down slowly till i can get it on the high vac, as long as it isn't bumping *too* much

scientist 1 said...

I neglected to mention that I run my bath hotter for higher boiling solvents too. Usually 65 centigrade is enough for solvents such as 2-propanol or water, but I' have gone higher.

I spin the flask pretty fast, but not so fast that the liquid is thrashing around inside. I like to see the surface of the liquid significantly perturbed, but if it is splashing or sheeting too far up the sides of the flask those points seem to flash more readily.

I haven't stripped more than a few hundred milliliters of water in quite a while, but the rotovap has worked well for it every time. I don't need to go to the high vac until the end to remove the last vestiges of solvent.

Dr. Miller said...

Jim tells me that Steve got a lyophilizer. That is great way to get rid of large amounts of water without bumping. Freeze dry that reaction.

Mike said...

Running reactions in water is overrated. Green chemistry is an oxymoron.