Chilling bulk ETOH with LN2

We’re putting together a cryo etoh extraction system and I’m still trying to wrap my head around the best way to chill the bulk solvent using Ln2 (storing enough to run a full day). My initial designs called for using a basic jacketed fermentor outfitted with an internal coil wrapping around the inside of the tank through which we could run our Ln2. An overhead stirrer would provide agitation.

Unanswered Questions: Would the part of the internal coil that becomes exposed as the level of the etoh drops with consecutive runs become frosted over? Would this affect the rest of the solvent in the tank? How long would it take to chill the solvent to say -70c? Is it possible to even control Ln2 temps in this way? Better to go with a large Flat Plate Heat Exchanger and chill inline to reaction vessel or on recirculation to bulk tank?

Any insights would be amazing.

1 Like

I would strongly suggest using a counterflow heat exchanger instead.

These come in several varieties, the most popular being a “tube in tube” counter flow coil and a “shell and tube” column.

These have two separate flow chambers that share large amounts of surface area, allowing much quicker cooling with no condensation contamination.


If you can pump the ethanol around, a plate style heat exchanger will have much more surface area for heat transfer than most tube/shell style heat exchangers (though they are a close runner-up). Plate style heat exchange is not self-draining so it can be harder to clean, but I have yet to need to clean mine beyond the ethanol that already circulates through it.

I use a centrifugal, magnetic drive pump by Finish Thompson.

Circulate the ethanol through the heat exchanger until desired temperature is achieved. I need to go grab me a liquid dewar and see how much LN2 it’ll take to really chill 250L ethanol. The Touch Science -80C/30L can hardly hang at -35C on a full load. It reaches -40°C on 200L, and -50°C on 150L. Cooling takes about 5-6 hours with this method. Mind you that the room is way too hot and needs more A/C


I wonder if a combination of the two would be beneficial. Fphe Ln2 recirculating etoh to bring the bulk tank down to temp/ touch science to maintain. Regardless, coming up with a good way of insulating everything would seem a good idea. I’ve been looking at biotech cryo-storage vessels wondering if there’s something to be done there.

vacuum is a great insulator. the jacketed fermenter I’m using for a still has a second jacket over the first that has been evacuated for insulation (a fairly common practice I believe). Even driving the jacket at 95C for 12hrs produces no appreciable change in external temp (on the jacket portions of the tank).

When I did the math on chilling EtOH with liq N2, it seemed like it was going to take most of a 200l Dewar to chill my 100gal of solvent from 20C to -25C. Yeah, I could have a had half dozen dewars delivered every week, but I opted for a mechanical chiller. If I need to go colder for specific extractions, then I’ll bring liq N2 into play.

I’d redo the maths here, but I can’t locate the handy chart my gas supplier provided.

1 Like

With your cryo insulated vessel do you run your chiller 100% of the time? Or is it sufficent to turn off the chiller for downtime? What is the most economical option you’ve found for maintaining your solvent temp?

I’ll have to get back to you on that.

Right now I’m still putting kegs of solvent in freezers, 24hrs before I need them.

Both my craigslist -80’s are dead now.
-40C still going strong. holds 3 kegs.
I have been putting solvent in -20 just to pre-chill it before the -40.

Were I to stay at this scale, I’d probably go with liq N2. lifting kegs full of 190 in and out of freezers is getting old.

The best thing about -40C is I can pull a keg out without worrying about personal protective gear. pulling them from a -80C required gloves and a lab coat.


Yikes! That sounds heavy.

What method of using Ln2 for chilling were your calculations based on (ie counterflow/ recirculation; jacketed vessel; immersion exchanger; etc)?

Someone (Air Gas?) was trying to sell us a system. They left me a flyer with the cooling capacity of Liq N2. I’d already done the maths to figure out what I needed to remove to go from +20 to -20 so I plugged their numbers in. As it was a piece of paper, it’s probably been recycled. I’ve asked google to find a copy for me a couple of times without joy.

I suspect I was looking at direct injection of Liq N2 into the solvent. just to get a ball park on how much it would take. I told the boss that there was no way a 200l Dewar would bring 100gal down to -40C before looking at the flyer he had. I either got lucky or cheated. (which is why I’d like to find that flyer and redo the math for peer review)

Here’s what I used to size my chiller.

Ethanol = 46g/mol

has a heat capacity of 112J/mol.K

and a density of 0.7893g/cm3

1gal = 3785.41cm3 (mls)

=> density = 2990g/gal

=> 100gal = 299000g

=> 299000g/46g/mol = 6500mol.

to lower by 6500mol by 40 degrees C (or Kelvin) would be 40 * 6500 * 112 = 29,120,000J = 29,120kJ

you can then pick a time frame in which you want to remove that heat, and size your chiller (convert to BtU). Chiller guy said he got the same answer via a different route.

As I was sick and tired of minions (ok, me too) going to the freezer to get a keg of cold solvent, only to find there wasn’t one. I spec’ed dropping 100gal in an hr (sans losses, and not looking at heat transfer efficiency).

The argument was that one could walk in of a morning and suddenly realize “shit, I need to extract 100lbs of cannabis today…I’d better get some solvent chilled!”

Who the hell forgets they need to process 100lbs of cannabis? So how come they keep forgetting they need to process 15lb every day? (they don’t quite yet. which is the problem)

What that really means is I can scale to 1000gal without going back for more cold** which was the eventual Hemp target I was given when I started looking at this.

When looking at plate and frame heat exchangers, it was pointed out that a double wall separating the glycol from my food grade solvent would be a good idea, as the glycol was going to be at higher pressure than the ethanol, so any leaks would be from glycol to ethanol without that double wall.

Originally this was a three vessel setup destined for an outdoor Hemp farm.100gal each: reactor, Still, receiver, with the reactor used to cool the solvent before adding cannabis. Condenser cooling was to be provided by well water, with water then being used on plants (after cooling of course).

No CUP was involved. Filtering was an unknown. The project manager of the Hemp grow has a PhD and deep experience in filtration/particle distribution, so I figured she’d dig me out of that hole if I set the rig up on her door step and looked pathetic :wink:

As it happens, I would have stumbled across’s solution just in the nick of time… Industrial Filtration Equipment Q&A

Scaling from 100lb => 1000lb was either bigger vessels or in parallel, depending on results from 100gal.

At some point, recovery from a standard still is going to become less than desirable. As recovery times increase, dwell time in the still increases. eventually you start destroying enough of your input cannabinoids that its not worth it.

But… rules changed, so plans changed. Oregon says that we can process Hemp in a licensed Cannabis facility…so lets do that instead…after some of the parts have been ordered.

Current plan is to use the Still condenser as the primary heat exchanger, and supplement with an in tank coil in the receiver. Doing the solvent “storage” and chilling in the receiver, and drawing as needed for processing in the CUP. As the receiver empties, the boiler is filled (by the CUP). done with the CUP? turn the still on. Come back in the morning to extract and cold solvent. rinse & repeat.

plans call for “trim” heat exchanger right before the CUP. using liq N2 here to chill by the CUPfull would allow solvent temp to be selected on a per run basis, not relevant for Hemp, but appropriate for boutique cannabis.

**given that I’m now using the chiller to cool the still condenser, I’ve got more heat to remove. I’m running at a vacuum and aiming at a boil temp between 25 & 35C, so there’s not a whole lot more.

Can’t wait to play with this thing!!


they were heavy when I started. We have an engine hoist that the minions use.

I’m mostly too lazy to go get the hoist, and just lift them out.

it becomes easier the more you do it :slight_smile:

restaurant style upright freezers are nice. a walkin would be better.

You remind me of my business partner, he’s always lifting stuff too heavy for him. I keep telling him to stop and get someone to help him lift. You see the tank link I sent you? Is that big enough for your endevor?

yeah, saw the link, didn’t see the utility till you mentioned a reservoir for the struggle chiller.

I’ve absolutely got use for a big ol’ vat full of cold. Have suggested the carcasses of my two dead -80’s for that role a couple of times.

Why not? You’ve got them, put them back to use!

1 Like


bah ha!

1 Like

I’ve been talking with a heat exchanger company who recently made some recommendations for a BFPE to chill ethanol from 20 to -80 in-line using Ln2. I gave them a specific solvent transfer rate through the exchanger and the presumed temp of Ln2. I figure I can transfer the solvent faster or slower to control the temp. I’m still confused about how the Ln2 is best moved to the exchanger. Gas or Liquid flood? Using some sort of thermally controlled cryo-valve sounds like a good idea. Seems a rapid thermal expansion of Ln2 in the exchanger would be dangerous.

The exchanger seems relatively small. Not having a firm grasp of thermal dynamics, I’m looking at this exchanger saying, hmm is this really going to do the job? Looking at surface area and cooling capacity of Ln2… How do I calculate this?

Fill the exchanger from the bottom with liquid, let gas come off of the top into an overflow tube. If you see liquid you’ve used too much. I know there’s gotta be a better way to do it automatically.


Incoming criticism my friend… If a hoist is needed for your fellow humans then it is needed for your own safety.

I agree with previous posts about cooler capacity. My father owned a Tavern Bar and this was our livlihood. A tavern MUST have a walk in cooler. Ours looked built by hand and a good job too. Make it out of wood. Use 8" studs and ¾ plywood. Then instead of foam or fiberglass obtain clean sawdust and fill the spaces between the studs to the top with course dry sawdust. Then this is placed inside a shed but away from any heat or direct sunlight. Our walk in was plumbed up with a outdoor compressor which ran mostly when the door opened and closed often.

The door was a bit of a bank vault style so just as thick and huge hinges but without a doubt if cooling solvent is the goal this would be the way to do it. Then you have ramps built in and hand trucks or you can simply plumb a line (like a beer line) from the cooler to point of use. I have seen this “cooler” design also used by my grandpa with no cold compressor. He used it to store year around vegetables from the garden and it stayed cool and even in temp even on hot days.