Thanks for the good thoughts and input brother Squiggly!
You're right about condenser capacity, and I have yet to pull it off with vacuum distillation, using all three of our lab condensers and double cold traps in series, with cooling from pumped ice water.
I collect about 20% more solvent with my simple pot still, so it looks like I will have to build an adequate condenser. Any design thoughts that you have on that, would be appreciated. My current thoughts are an upper horizontal header, with "N" down tubes between it and a lower horizontal header.
I envision it sitting in a barrel of brine, with a single tube passing through the bottom of the barrel via a bulkhead fitting, and bleeding into a catch container, kept under vacuum.
How would you design your condenser, from a clean sheet of paper?
If I had the means--I'd design it as a coiled coil (tiny coil wrapped into a big coil) its ALLLL about surface area in terms of design shape. You can also pump through colder coolant as a way to increase capacity (much in the same way you can increase your heat conductance to hit the enthalpy of vaporization you're looking for--you can decrease the temp to hit it from the other direction). Sometimes it's as easy as using a salt/ice/water slurry to get the job done.
Alternatively I'd do some kind of a hybrid between an alihn and graham coil.
Generally a coil is going to be your best bet to increase capacity, just think surface area and your ingenuity will come up with something. Remember that equilibrium plays a large role here--two very important notes on that:
1. Condensation creates heat--so your thermal conductance is highly important. As each unit of heat is gained, the next molecule is that much less likely to condense--the faster we remove the heat, the higher the capacity will be. Keep in mind that molecules work at incredibly high speeds so every little increase in efficiency counts for SO MUCH in this respect.
2. As a result of 1 it becomes a concern how fast solvent drips from your condenser--you don't want it to have anywhere the liquid can settle (another reason a coil works very well).
As for bumping, how much bumping are you seeing with a thin film under vacuum? I see it in our simple pot stills, without boiling chips, but haven't noticed it doing thin film vacuum purging in a Petri dish.
Beware, bumping is an all too common occurrence in thin film evaporation. A good rule of thumb is to never fill the vessel more than half way--but it will all depend on the angle of inclination of the neck as I'm sure you might imagine, because in the end bumping is a liquid transfer that results from this sort of snowball effect.
Usually we think of it as superheating and lack of nucleation causing a boil over when nucleation begins and a chain reaction of nucleation results--boiling chips are thus the answer in stills (as they provide nucleation sites to prevent this).
When you're actually reducing the pressure (you're not talking about a vacuum still are you?)--boiling chips won't get you too far. In the case of boiling under atmospheric pressure you're spending essentially the entire heat up time trying to overcome 1atm of pressure.
With this process you are removing that impedance to boiling in a matter of seconds rather than minutes. Not only should you expect vigorous boiling--but it'd be weird if you didn't get any. Instead of seeking to control nucleation (because often rotovaps are meant to retrieve a pristine sample--unadulterated by anything else, including boil ezers) we just make sure the vessel is big enough to handle the size of bubbles we're going to make (BIG), and also to not overfill the vessel.
A final safety feature is the bump trap, and I highly recommend you install one. If you custom source a glass vessel--it should be no problem to simply purchase a bump trap (or have on custom made). The last thing you want to do is have a high surface area condenser shitted up with oil.
If it is an issue, a labyrinth intake inside the drum, should keep bursting bubbles out of the intake stream.
you'd know better than I on this one--but be aware that the shit is going to bubble--and if you don't plan accordingly you're going to have thin film on every inch of the vessel, rather than in the bottom quarter where you want it.
Good idea on the custom glass drum. I will check with our scientific glass shop for a price!
Borosilicate is a better conductor than 304 stainless, but it is a linear formula and we can always increase the delta T to get the same rates of heat transfer.
I suspected as much with motor speed, and the incremental cost for a variable speed motor isn't that much, so I probably ought to bite the bullet. It would certainly make process development and refinement easier. As noted, the test sled most certainly will have one.
Few final notes:
I'd really recommend making the vessel RBF shaped--this aids in the dispersion of the sample into a thin film. You will end up trapping solvent near the edge of your drum. The idea is for the film to be uniform--which pretty much means a round vessel. (Angle of inclination also plays a role--and this is why the more expensive models allow you to adjust this).
If you're able to source LN2 (I don't know how much it costs)--a cold trap might work, but you must be very careful as there is a chance for an explosion if you pass too much air too quickly over the cold trap when bleeding your manifold.
You might also try a cold finger condenser. As I said earlier you can greatly increase capacity by bringing temps way down. If you construct a cold finger apparatus that will accept a dry ice bath (ethanol -72C, acetone -78C, or LN2 baths can take you all the way to -192) it should be able to overcome whatever you throw at it (with a reasonable size, of course).
