Isolating Cannabis Terpenes, Part II: Actual Process

[WaterFarmFan]

Here is a diagram of my DIY setup that I am going to test. Going to experiment with lots of variables, including different temp ramps and automated exhaust (terpene collection) cycles. Any thoughts???? Still curious how co2 sublimation (from dry ice under vacuum in oven) would affect nitrogen sparge. Later in the process, as temps rise and h2o (from flower soaking) begins to evaporate, wouldn't water collect in cold traps and dilute terpenes?

EDIT - REMOVED PHOTO FOR UPDATED VERSION BELOW

WFF

 

[WaterFarmFan]

Entrainment of terpene(oid) in gas stream is an issue I've experienced, especially given that Anular cold traps are not really designed to "catch the terps". (really meant to minimize pump/process contamination). In vac oven, the unheated window also presents issue, as they like to condense there first. Playing with water content and other variables. Trying it with a heat mat on the window, and a new trap design that should de-entrain my target

Hey Doc! What starting and end temps have you experimented with? How long is your process under vacuum? How often are you exhausting and at what pressures? Brilliant idea about the heating pad with separate PID on the window! Thanks for your time!

WFF

 

[WaterFarmFan]

OK. After some additional reading and thought, I have modified my diagram for better vacuum flow. The automated collection process would be triggered by the short cycle repeat timer, which turns the vacuum pump on. The first delay relay would trigger a closing of the vacuum pump vent. After 10 seconds, the second delay timer relay would trigger a simultaneous opening of main vacuum line and closing of nitrogen back feed. After another 10 seconds, the third relay would trigger the opening of a throttled ball valve at back fill and initiate flow for collection. After a set collection period, third relay would close, followed by 2nd and 1st and then pump. Relays would have to be in sync and changed for each experimental collection time in the beginning. The premise is to create a repeatable process that has minimum inputs to potentially alter collection.

It is pretty clear that I am not an expert, but I love to learn and experiment with things! If you have some thoughts on this setup, please jump in!

WFF

 

[Dr.Terp]

Nice work WFF! I'm reviewing your drawing in detail now, looks proper! Maybe I got enough stuff laying around lab/shop to rig something similar. I have been running manually and rather "blind", just to get a general proof concept and see some terps and feel for they flow.

One issue I have found is the vac oven might not be the best vessel,especially as we scale this. Im working on some designs for material chamber that will give better mass transfer and flow control.
In the works is frustoconical geometry with top. middle, and bottom temp zones. Operated like fluidized bed. As I write this I think of using Jacketed column (maybe with expanded volume section as top section) ProPak in between layers of biomass. (much quicker parts to obtain)

I'll share with this thread the results, as I am taking measurements now, and hopefully be able to add WFFs instrumentation suggestions.

My favorite season of the year is right around the corner!

 

[WaterFarmFan]

Nice work WFF! I'm reviewing your drawing in detail now, looks proper! Maybe I got enough stuff laying around lab/shop to rig something similar. I have been running manually and rather "blind", just to get a general proof concept and see some terps and feel for they flow.

Thanks! Been doing lots of reading and slowly assembling all of the pieces. I have read patents and papers on industrial food processing that have led to many ahh-haa moments (for me). My planned workflow has changed from above in that I will be using liquid nitrogen to pre-process fresh and semi-dried flowers into a hash consistency, adding (a yet to be determined amount of) distilled h20 and then immediately adding more ln2 to freeze material and placing container in vacuum oven. I have a 1 1/2" thick slab of PTFE that a double walled stainless bowl (with pre-processed flower, distilled h20 and ln2) will sit on. Will pull full vacuum and bleed n2 from backfill. Want to experiment with making ln2 solidfy to further shatter trichome walls and release terpenes when distilled h20 (and h20 from plant matter) evaporates.

One issue I have found is the vac oven might not be the best vessel,especially as we scale this. Im working on some designs for material chamber that will give better mass transfer and flow control.
In the works is frustoconical geometry with top. middle, and bottom temp zones. Operated like fluidized bed. As I write this I think of using Jacketed column (maybe with expanded volume section as top section) ProPak in between layers of biomass. (much quicker parts to obtain)

I'll share with this thread the results, as I am taking measurements now, and hopefully be able to add WFFs instrumentation suggestions.

My favorite season of the year is right around the corner!

I am just about to start testing with my vacuum oven, but I too have some ideas for a custom chamber. I am most interested in using ultrasound sonication during evaporation. I think that I will have a custom vacuum chamber lid milled that integrates a ln2 dripper, nitrogen backfill & accurate gauge port. Do you think that in such a chamber design, that it is more efficient to collect terpenes from the bottom, much like the OP did with mason jar on bottom, but with different tech???

WFF

 

[WaterFarmFan]

While part of me loves to take things apart, I would not make such drastic modifications to my vacuum oven. Here is a mock up with some ideas for using a simple vacuum chamber with some mods. The premise would not to be get really hot (less than 120F) but repeated freeze thaw cycles using a smallish amount of ln2 to initiate the refreeze. H20 dripper would allow entire process to be completed without opening lid if more water is needed to aid evaporation. Ultrasonic would be used after thaw while water is in liquid state. Could be automated in quite a few places.

WFF

 

[HashoftheTitans]

I had decent results with a much more crude version of this. I ground kief and dry ice in a ninja blender and spread it out onto parchment on trays in the oven, about a pound per shelf in my 1.9 oven. I pulled a deep vacuum through a series of cold traps and slowly increased temps to 135f. I ended up with 15ml of delicious terps and about 50ml of hydrosol. The kief was two years old so the terp content was fairly low. However, the terps didn't taste the least bit degraded and they were water clear.

 

[WaterFarmFan]

I had decent results with a much more crude version of this. I ground kief and dry ice in a ninja blender and spread it out onto parchment on trays in the oven, about a pound per shelf in my 1.9 oven. I pulled a deep vacuum through a series of cold traps and slowly increased temps to 135f. I ended up with 15ml of delicious terps and about 50ml of hydrosol. The kief was two years old so the terp content was fairly low. However, the terps didn't taste the least bit degraded and they were water clear.

Nice! What temps were you using in your cold traps? Did you make any attempt on post processing the hydrosol to extract water miscible terpenes? From my reading, seems chromatography might be one of the few viable options.

WFF

 

[bloyd]

Curious if anybody is collecting terps from there freeze dryers. Just started playing with one and I am disappointed with the terp loss due to vacuum despite cold temperatures. The water collected from the frozen walls is milky white and terp rich but doesn't seem to separate under normal conditions. Thoughts?

 

[HashoftheTitans]

Nice! What temps were you using in your cold traps? Did you make any attempt on post processing the hydrosol to extract water miscible terpenes? From my reading, seems chromatography might be one of the few viable options.

WFF

2 @ -40 and the last one on dry ice. the first picked up most of it and there was almost none in the last trap. the flavors didn't differ so I combined them all and collected with a sep funnel. I gave the hydrosol to a home brewer in hopes to get some canna IPA. That was a year ago and I'm still waiting.

 

[live resin]

Nice! What temps were you using in your cold traps? Did you make any attempt on post processing the hydrosol to extract water miscible terpenes? From my reading, seems chromatography might be one of the few viable options.

WFF

Have you tried centrifuging the hydrosol? You might not recover all the water miscible terps due to azeotropes but probably enough to be worth the trouble.

Or try to freeze the water out to separate the hydrosol?

 

[WaterFarmFan]

2 @ -40 and the last one on dry ice. the first picked up most of it and there was almost none in the last trap. the flavors didn't differ so I combined them all and collected with a sep funnel. I gave the hydrosol to a home brewer in hopes to get some canna IPA. That was a year ago and I'm still waiting.

Good info. Since I will be adding additional distilled h20, which will create additional hydrosol, I want to play with temps with my two stage traps. My working logic is that the first trap is the warmest, possibly -20c to -10c range to start using glycol/water/dry ice, which easily will condense h20 but let the more volatile terpenes pass through. Second stage is in the -70c to -110c range using 4x 200mm detachable 24/40 cold traps in series with ethanol/iso/ether and dry ice/ln2 combos. It will create fractions, but like you, I will add them all together, as the goal is to add back terpenes to shatter to create a full spectrum extract that has the taste and entourage effect of the premium flowers being used as source material.

 

[WaterFarmFan]

Curious if anybody is collecting terps from there freeze dryers. Just started playing with one and I am disappointed with the terp loss due to vacuum despite cold temperatures. The water collected from the frozen walls is milky white and terp rich but doesn't seem to separate under normal conditions. Thoughts?

Bingo! This is the basic tech that we are replicating, but IMHO using a more suitable vehicle in a vacuum oven, which is designed to ramp temps higher and faster than a freeze dryer (I think????). A freeze dryer kills the oven in its ability to create sustained low temps. Haha - A heater heats and freezer freezes who knew?!?!?!?! I overcome this be using ln2 directly on material and pulling vacuum. Thus, my starting point is very low, much lower than freeze dryer, and the vacuum oven allows an orderly temp ramp. My oven PID starts at 0C, so for safety of oven components, set at this temp to begin process.

I do not have a freeze dryer, but I have read about there use for industrial extraction of terpenes. I think that they need to be modified with additional tech to be used a dedicated terp extractor. Not sure though...

 

[WaterFarmFan]

Have you tried centrifuging the hydrosol? You might not recover all the water miscible terps due to azeotropes but probably enough to be worth the trouble.

Or try to freeze the water out to separate the hydrosol?

I did recently get a centrifuge, but I read (on reddit I believe) that some folks tried and did not have much success with this method on hydrosol. You could obviously short path, but the heat and vacuum required would alter the profile significantly (I read). Freezing (temperature differential) in a chromatography column is on my list of testing. I also plan to safely modify my centrifuge to connect an air pump that blows through a wort chiller at 0c to -20c to chill contents and perhaps aid in better separate.

I did read a paper where they used a separatory funnel, that was pre-loaded with distilled water and a layer of methanol on top, directly connected to cold trap. It showed that the methanol captured the azeotropes before they made it to the water and became miscible (I think?), but methanol could not be used to separate miscible elements after the fact (thus the layer on top I think?). The problem with this approach (to me) is that you now have to effectively purge methanol without damaging or losing specific terpenes. Doable with chromatography...

 

[G.O. Joe]

From my reading, seems chromatography might be one of the few viable options.

There's adding some salt and letting that sit in the freezer, or what I'd do which is shake with ether.

There may be more-soluble non-terpene organics present acting a bit as solvent. These may be a part of or the whole desired essence.

 

[WaterFarmFan]

There's adding some salt and letting that sit in the freezer, or what I'd do which is shake with ether.

Thanks! I have read of "salting". Would the ether form a separation layer or extract non-h20 compounds, and if so how to separate compounds from ether?

There may be more-soluble non-terpene organics present acting a bit as solvent. These may be a part of or the whole desired essence.

My thoughts as well, and it why I have interest in really trying to process the hydrosol. There just might be some compounds that aid in entourage effect. Want to squeeze the last little bit of juice out of the orange...

 

[WaterFarmFan]

I have read multiple citations about using 60 Å, 230-400 mesh, 40-63 μm particle silica gel for the chromatography separation of hydrosol.

https://www.emdmillipore.com/US/en/product/Silica-gel-60-(0.040-0.063-mm),MDA_CHEM-109385

My understanding is that if I mix the silica gel above with my hydrosol and add it to a chromatography column, that the water will be absorbed by the silica leaving only the terpenes behind as I run it through the bottom frit? Does this sound right?

 

[G.O. Joe]

The great things about ether are, a whole lot of organics will dissolve in it, it evaporates very easily, and it has limited solubility in brine especially cold. I'm not sure about adding straight water to flash gel I have some right here and could test it maybe. Organics will bond somewhat to silica until eluted with something that overcomes those forces.

 

[WaterFarmFan]

The great things about ether are, a whole lot of organics will dissolve in it, it evaporates very easily, and it has limited solubility in brine especially cold. I'm not sure about adding straight water to flash gel I have some right here and could test it maybe. Organics will bond somewhat to silica until eluted with something that overcomes those forces.

Would you recommend evaporating ether via vacuum cold boiling or just sit out at room temp and evaporate with a non-heated stirrer? Would you recommend a chilled brine and ether at the same time for hydrosol extraction?

 

[WaterFarmFan]

How many ml of ether would you add per 100ml of hydrosol solution?

I am not a chemist, but if I understand properly, the oil and some water based terpenes would collect at top of separatory funnel and be separated yielding a hydrosol. Thus, all of the compounds that I would like to extract from hydrosol are water-soluable and have a boiling point that is significantly higher than ether. Thus, it should be relatively easy to separate ether at low temperatures in order to not alter organic compounds. I think?