DIY CO2 Extraction?
Supercritical carbon dioxide
Standard carbon dioxide pressure-temperature phase diagram from Wiki, plus another showing the droopy nose normally not shown on CO@ SCFE charts, but which can be exploited.
Supercritical carbon dioxide is a fluid state of carbon dioxide where it is held at or above its critical temperature and critical pressure.
Carbon dioxide usually behaves as a gas in air at standard temperature and pressure (STP), or as a solid called dry ice when frozen. If the temperature and pressure are both increased from STP to be at or above the critical point for carbon dioxide, it can adopt properties midway between a gas and a liquid. More specifically, it behaves as a supercritical fluid above its critical temperature (31.1 °C) and critical pressure (72.9 atm/7.39 MPa), expanding to fill its container like a gas but with a density like that of a liquid.
Supercritical CO2 is becoming an important commercial and industrial solvent due to its role in chemical extraction in addition to its low toxicity and environmental impact. The relatively low temperature of the process and the stability of CO2 also allows most compounds to be extracted with little damage or denaturing. In addition, the solubility of many extracted compounds in CO2 vary with pressure,[1]permitting selective extractions.
Soooo, in us’n layman’s terms, we use plain old CO2, that we have gotten hotter than 31.1C/87.98F, while under at least 1086 pounds per square inch pressure.
In that state, the CO2 becomes a super solvent, which has higher penetration power, and can be fine tuned to focus in on specific constitutes, by varying the pressure, temperature, and by the use of co-solvents like ethanol or hexane.
Fair enough, and easy enough to accomplish several ways! Traditionally a high pressure intensifier pump is used to achieve the required pressures.
Cryo-pumping also works. You start with liquid CO2 or dry ice and heat it in an enclosed space.
My first conceptual of a system that was semi affordable, was a simple minded system using liquid CO2 and heat to run it at about 1500 PSI. That eliminated the pumps normally used and by using a liquid Dewar with a dip tube, with both a gas head and a liquid tap, the need for any liquid pumps is eliminated as well.
It is designed to use readily available Schedule 180 austenitic stainless pipe and its largest component is small enough to fit in my 7″ X 36″ lathe. My original plan was to use an existing hydraulic cylinder for my vessel, but alas, carbon steel embrittles excessively at cryogenic temperatures.
My design calls for achieving pressure by simply heating the pipe with available pipe band heaters.
Pressure was controlled using a conventional, though expensive back pressure regulator. Please note the attached conceptual, which reflects my thinking four to five years ago. I put the project on the back burner after reading the SFE CO2 patents for extracting cannabis, and noting that CO2 isn’t a very aggressive solvent and the patented processes used significant fluid exchange rates, as well as relatively long spans of time.
To compound that, reports from those testing SFE CO2 extractions reported poor taste and effects, so I put CO2 extraction on the back burner to just watch for awhile.
Within the last couple of years several folks have done some interesting work starting with dry ice. The first that I noted was able to maintain high enough pressure in a test tube to keep dry ice from sublimating as it warmed, so as to extract some Limonene from an orange peel.
That led me to question whether super critical pressures were required to extract the essential oils from cannabis that we typically extract with other solvents, and further experiments by others suggests that it is not. Some even report better results at subcritical pressures, especially with the use of co-solvents.
Sub critical extraction presents an interesting twist from a simplicity standpoint, so of course my simple mind immediately envisioned replacing the test tube with a pressure vessel containing dry ice in the bottom, and a basket of material suspended above it. Screw down lid and bring it up to temperature with a pipe clamp heater for pressure.
Flooding is achieved by simply turning it upside down to soak the material and right side up to drain. That step could be repeated back and forth, to soak the material and keep the boundary layers diminished. On setting it upright the final time, we could bleed off the liquid into an expansion chamber and then to atmosphere, leaving behind the oil.
I have decided to build a 2″ X 24″ prototype, and have acquired two resources on this project that emboldens me. One is the offer from a dear old friend, to allow me the use of his aerospace pressure and vacuum equipment manufacturing facilities for giggles and old times sake.
The other is an agreement with another old friend, who owns a cryogenic equipment manufacturing plant back east and is familiar with SFE.
We have a mutual engineering support agreement in effect, in exchange for my support on one of their projects with a former aerospace customer of mine, who is now their customer as well.
For my support interfacing their equipment with equipment of my own design and manufacture before retirement, they will support me in developing a simple minded DIY SFE system, that hopefully more of us’ns can afford to build, or have built.
So here is my first simple minded conceptual using liquid:
Here is my simple minded conceptual starting with dry ice:
OOPs, solly, watch this space!
After working all day on the drawing, my 2002 32 bit Auto Cad program is not able to access my printer on my 64 bit processor, and I’m not smart enough to figure it out, so I have to find a computer hero or heroine first.
The way my conceptual works, is the bottom of the cylinder is filled with dry ice , on top of which sits a wire basket of plant material. Clamp on band heaters turn the dry ice to liquid, which is circulated and soaked through the material by simply turning the cylinder upside down.
After flipping it about a few times, over an extended soak, the vessel is set upright and attached to the lower expansion chamber via a hydraulic quick disconnect. The liquid is bled off, decompressed, and the oil collected from the decompression vessel via the bottom drain valve, as well as via the detachable bottom.
Lu lu lu lu lu lu lu lu lu lu lu lu lu lu lu!!!!!!!!!!!!!!!!!!!!!!!! Progress!
Well thank the Great Spirit for ex-brother in laws, who can sort out computer problems, so that I can print from virtual mode. Hee, hee, hee, kept him at tossed the sister……….
Here is my simple minded conceptual of how such a dry ice system might work. More detail prints once I have exercised some resources.
After further discussion with another experimenter, I have decided to build the system to operate at 5000 psi and below. Instead of using schedule 160 stainless pipe, I will gun grill a billet of stainless, but will keep the bore at one inch.
Hi ya’ll!
Been out talking to folks running commercial CO2 SFE units and would like to share some selected excerpts from their letters, to keep their identities confidential:
Hi Graywolf, here is some pics of the unit i have been running for 8+ years. Our whole thing with this unit was to have maximum fractionation capabilities.
The extraction vessel is rated to 10,000psi, and made of stainless!!
We use a basket to hold the material and I’m having a natural cloth bag made that will go into the vessel and hopefully allow me to extract 2x current volume.
There are some great graphs around the net that show the combinations of pressure and temps and the resulting density of co2. In general high pressure (we go up to 10,000psi) works well in combination with high temps (around 100c) and lower pressure around 5000 psi goes well with about 60c.
I tried high pressure and while i need to try again to confirm, there seemed to be some degradation to the overall quality and definitely the fragrance. the yield was higher and its much more efficient requiring less co2.
I already noticed on the bit of extract that i dissolved in ethanol and then winterized and filtered, the flavor is no where near what it was before going in the ethanol. I plan to try and get the waxes out with only co2.
I cant see any advantage to using either hexane or ethanol or a mixture as a co solvent?? the compounds are easily soluble in co2 and usually those types of things will help to extract more polar compounds such as chlorophyll. It all depends on what your goals are? maybe he is using that mixture to help fractionate out the waxes??
I can do a small run next week using those co- solvents and report back if i see anything magic or some beautiful separation. maybe he is just some chemical engineer using a method that was used for another project and he is trying to apply it there?
It seems really counterproductive to use co solvents and contaminate a great extract… maybe he can only do really low pressure and he requires the co solvent to help the efficiency??
I found material with a moisture content of around 10% to be fine. I don’t know what the implications of a slight amount of carbonic acid could mean?
I have heard it can make oils rancid, but have run vegetable oils at 10+% moisture and had them test very low in peroxide values in respect to oxidation and rancidity. I do not see any problems with around 10% moisture, the water sometimes comes out with the extract and is sitting on top and the few mls there can be poured off.
And from yet another hero!
I. Extraction was performed by sub and supercritical CO2:
Subcritical (liquid): 10-12 ºC @ 65-70 bar for approximately 6 hrs.
Supercritical (liqid/gas): 40-45 ºC @ 230 bar for approximately 4 hrs.
II. Separation of fats and sugars, by winterization, chromatography and high speed centrifugation.
This part was very critical in the purification schema and required analysis of both the anaylate and precipitate.
III. Purification, at this point one has two options, Flash Chromatography or Wiped Film Evaporation (Short-path)
I performed both and in both techniques, one can obtain purities in the range of 95-99% total THC (∆9, ∆8 and ∆6) with the purity of ∆9 ranging anywhere from 78-98% depending on the parameters set.
One last comment, probably the most important….. absolutely no petroleum derived hydrocarbon solvent was used. Except for the small amounts used for LC/MS analysis. This was my driving force to see if it could actually be pulled off.
12-21-12
We are moving this project from research to active experimentation in our FY 2013, now that the shack and Terpenators projects are a fait accompli. What I did in FY 2012, was simply research and develop resources.
It was good that I did, as further research put things in better perspective.
While I determined that my simple minded CO2 SCFE system would work, it raises a the bigger germane question of why is this trip necessary, given how effective BHO and some of the other extraction methods are. Why spend the money for this more expensive fish trap, if we already have the fish?
The answer lies in that what CO2 SFE extraction shines at, is selective extraction. By fine tuning the pressure and temperature, as well as using co-solvents, specific components may be extracted, while leaving behind everything else.
That is of less value when discussing cannabis essential oil extraction, in that it has been determined that it is the entourage effects of all the terpenes present, including the diterpene cannabinoids, that are responsible for its medicinal properties. That however doesn’t mean that we can’t specifically focus on that list of terpenes.
The other thing that my research made clear, and that is using one vessel to process the material and raise the pressure in, would result in extractions along the entire temperature/pressure curve. Woops, no selectivity there, but easily solved by bringing the liquid CO2 or dry ice up to a higher temperature and pressure in a separate vessel, and then equalizing it with the one containing the plant material! Easy, but the system complexity grows.
One of the interesting things that my research turned up during tete a tetes with brothers owning and operating commercial CO2 SCFE’s, and bumblers like myself whomping up DYI ghetto extraction equipment, is the wide range of processes that different operators have achieved a modicum of success at.
One operator simply took the valve out of a CO2 bottle and filled it with plant material and dry ice, before replacing the valve and heating the tank to SCFE ranges. He then inverted the tank and vented the CO2 into a catch container, where it left cannabis essential oils behind.
He claimed his best success using co-solvents at ~5500 psi and that his product was superior to all competing, but alas, I was never able to sample it and his spamming got him banned from the forums where we shared information.
Commercial SCFE operators report successful sub critical products as low as 350psi, and another at 850psi, but I haven’t been able to sample their successes, to confirm my own opinion of product quality. There in lies the rub of course, as there is no common measurement and each creator loves his creation.
Fortune has ostensibly smiled on us however, in that two of the operators that I have been conversing with, have agreed to work more closely with us and to supply confirmation samples. Hee, hee, hee……………………………
Even simpler DIY SCFE C02 6-2-13
Hee, hee, hee, you have to admire simplicity. Here is the system by
jyndustriez which I mentioned, that simply puts the plant material in an
off the shelf CO2 tank, and floods it with liquid C02 from a second off
the shelf tank, before heating to achieve super critical state.
More information on how he accomplishes that at http://jyndustriez.blogspot.com/, but here are some pictures he has shared:
8-25-2013
Progress! We got to play with a donated SCFE CO2
extraction sample and were able to heat it under vacuum and observe the
effects.
As received, the sample was in a “wax” hydrate form, and was not aromatic. The flavor was mild, and the effects were good.
As it was a small sample, I was unable to winterize
it to remove the waxes, but I was able to heat it to 115F, under
vacuum, and observe what happened.
The sample out gassed significant CO2, even though
at 115F decarboxylation was minimal, and never quite turned clear, due
to the high wax content, but that isn’t unique, as BHO extracts are
improved by wax removal as well.
On the surface however, the wax content seemed
high, compared to our typical BHO extraction, so I look forward to a
large enough sample to winterize and determine just how much.
Here is a starting and ending photo, demonstrating the ending wax content.
To add to our excitement, last week we got a guided tour of Eden Labs during our recent business trip to Seattle! What fun!!!
Patrick, one of Eden Lab’s bright new sales
consultants, attended one of our BHO classes and subsequently invited us
for a tour, so how could we resist, hee, hee, hee, snicker, snark,
snort………………?
Always a pleasure to see how the pros from Dover do things, and I do likes the way they detail a package.
We lucked out in finding the original creator
available and got our guided tour from Fritz himself. Besides his
latest creation, we got to see not only his original test sled, but some
of the R&D that may make great strides someday, when Fritz can find
enough time to finish them, between current product demands……….
Their latest all stainless creation was in final
assembly and testing, and was capable of extracting not only with
subcritical and Super critical C02, but also with refrigerants like
butane and propane.
It was appointed well, and used pneumatic intensifiers both for the hydrocarbon pump, as well as the SCFE CO2 pump.
It also sported a unique closure system, which
simply spins close and that greatly simplifies loading, as compared to a
12 bolt flange with a torque wrench.
The vessel itself uses a Schedule 160 stainless
core, but adds water jackets, so as to permit the use of a commercial
heater/chiller for temperature control.
The unit that we looked at, had a $60K base price,
plus the price of the hydrocarbon extraction capabilities. Not cheap,
and more complicated that most folks should attempt at home, but if you
need a machine capable of a quality extract, in commercial quantities,
you will need to step up to a system of this caliber, which includes
pumps to cycle the C02, and C02 recycle to recover it afterwards.
Fritz did note that one of the things on his list
of thangs he is juggling, is a smaller lab sized unit, which fills the
market window for a smaller unit that doesn’t recycle. Looking forward
to see what comes up there!
From a process perspective, Fritz shared that the
aromatic extractions are best done at low pressures and as the pressure
goes up, so does the rate of chlorophyll extraction. Based on his and
other input, I will focus on the range from 800 psi to 3500 psi. While
I’ve heard operating parameters up to 9500 psi, the lower ranges sound
more believable.
As unbelievable as it may sound, ah think sometimes
operators shuck me, cause they are trying to protect process secrets
and lead me astray.
10-27-13
Lu lu lu lu lu lu lu lu lu lu lu lu lu lu
lu!!!!!!!!!!!!!!!!!!!!!!!1 The Great Spirit smiled and JYNdustriez
loaned us a CO2 extraction setup and will show us how to use it, when he
visits to watch Dr. Fischedick’s lecture on 11-2.
I’ve also located a local brother with an Eden Lab
machine, who is amenable to experimentation, so things are heating up in
the area of CO2 extraction at Skunk Pharm Research. We will have
machines from both ends of the price scale to experiment with.
Sound CO2 has also extended us an invitation to
visit, and they run Apeks equipment, soooo even though our CO2 project
was slow getting in motion this year, it looks like it is finishing with
a bang.
