Sorry guys, couldn’t post the link this morning. Future4200 was down for a while. Never saw that before. The server was returning a 500 error code.
I will run another session real soon, sooner if I get replies from interested people!
Let me know. Today’s run was in 1 hour 15 minutes. The oil run yielded 3 grams from 2 oz hemp.
Nylon actually has an “excellent” rating with acetone. Definitely do be careful about what plastics you use though, because there are many that aren’t.
56g in, 3g out?
Assuming 10% cannabinoids on your input and 90% potency on your output (generous), that puts you in the 50% extraction efficiency range?
Do you have potency on input or output?
Hey Cyclopath, sorry for the lag. I saw ur reply after eating my nightly gummies, so I waited for clarity before replying. You asked a question head-long into my obsession! 
I believe you are correct in seeing a 50% reduction in total oil in the first run. Fortunately, its easy to recover the losses in a second run.
On this run, I used hemp product purchased from BuyOregonHemp.com and have the C0A. CBD is 17% with this strain. On losses, there are three points where this can happen:
First, I did a quick 30 second wash of the 2oz with 99% ISO at room temperature, poured the plant through a nut milk bag for straining, then let the bag drip-dry into the capture container. What’s left behind can account for upwards of 20% loss, to be recoved in a second wash.
Second place of losses - distilling with distilled water. So, the wash metioned aboves resulted in about 2 cups 99% ISO to be boiled down. I added about 1/2 cup of distilled water to this. This forms two emulsions. First being a weak water/alcohol emulsion that can gets boiled out when the water reaches 212f. The second one is much harder to break, and this is the phospholipid, IE, lecithin-like binding of water and oil. This is what’s left at 212f. The water can look clear, typically translucent, all the way to the worst case which looks like whole milk. The variable I have found is the plant phenol type and soak times. This goes back to @beaker 's louching. More on this below. What I am finding, this emulsion can account for about 8-10% ‘loss’ of oil in the first pass. This ‘loss’ in the waste water can be recovered in a second reduction.
The third loss is simple transfer losses which could add up to .25g on this simple setup.
So to total it all up, a single pass results in very clean oil at about 50% of the potential oil volume. That’s backup up with the lab test results showing improved potency from ~33% up to ~78%.
Good news, these losses are recoverable. This means the process is a crude fractioning of the end product in it’s simplest form. This simple process has tradeoff which means the burden of cleanup is messy in the kitchen. Water makes it very easy to separate the best oil, but makes it messy to reach 100% extraction. I think of this process like onion skin layering. The first layer is easy, the others more work to recover; the 80/20 rule dominates. I am currently obsessing on optimzing this second recovery. Your question is my concern, fortunately there is an answer. I will publish how to recover these ‘losses’ in the ‘RxCE Shop Manual’.
Going back to beaker’s work, as I understand louching, its the ratio of water to alcohol + concentration of dissolved plant compounds. The plant contributes to the emulsion depending on the total bi-polar amphilics (phospholipids being one of them) dissolved into the solution. The longer the soak, the more amphilics, the greater the ‘loss’. Fortunately these emulsion-tied-up oils can be boiled out. The hard emulsion does break as the water steams out. Interesting enough, the water wont boil until this emulsion is broken. Apparently, the oils in the emulsion absorb some of the BTU’s being pumped in until the water/oil ratio shifts.
Here, let me toss this out for discussion. Using the prior process called RSO 2.0 2022, where I introduced ‘Silting’, I ran 3 tests this past January and reported the results to Graywolf. Here’s a .PDF of my email writeup and illustrations to report the yields.
I hope this answers your question?
Blog-RSO 2.0 Test Results 01.pdf (2.2 MB)
17% input & 78% output?
56g and 3g
That’s 9.5g in and 2.34g out.
Which is 24% extraction efficiency.
Seems a smidge more than 50% losses
My scale showed 2.97grams, so 2.34 is short. Here’s the COA. If the COA is correct and the calculates are correct, then losses are even greater. The little emulsion that was in the translucent water could not contain 7+ grams of oil. So either the wash left a lot more oils, or the COA is questionable. I’ll run a long soak on 2oz and boil that straight down without water and see what the total crude oil tops out at.
That was 3g x 78% potency == 2.34g CBD in your extract
Follow the cannabinoids. Raw yields are meaningless without potency.
I believe your calculation, but the COA is based on 23.5% so that might account for my higher #. BTW, I see the second page of that COA looks like from another batch. Dunno how that ended up in that box, probably my mistake. Ignore that second page.
Oh, I forgot to mention the losses in Rapid Winterization. I cooled the boiler down to freezing by placing it in a bowl of ice water and in 10 minutes, waxes formed. I poured that thru a paper napkin filter, and it looked to be about 1.5 to 2 grams of waxes. The color was a light green, not black so it has some oil, but was no where near saturated. This is about the color of the green I saw.
I’ll run a couple tests this weekend taking 2oz of this bud and doing a 10 minute soak with 99% ISO, then 10 minute soak with same bud chopped up and soaked with 91% ISO, then boil down both straight away to get a total #. This way, we see the total grams of crude available in this plant. That should give us a clue to the accuracy of the COA.
BTW, I appreciate your dialog. I believe the RxCE process is good science, just not optimized. Your input is helping me narrow that down.
Hey Cyclopath,
This weekend, I setup three room temp 10 minute extractions and boiled those down to crude resin, aka, RSO. These are the control cases. I now have to go back and repeat the three runs using the RxCE process. Right off the top, this is probably the answer, but won’t know for sure till confirmed by the next tests. On the reported video session oil, I was doing 30 second room temperature washes with 99% ISO, versus these three 10 minute soaks.
Here’s the yields on the control cases:
Grn--------91%---------Chopped------------N------------7.8g
----------------------------flower
Blu---------99%---------1/4" bud-------------N------------7.1g
Red--------99%---------1/4" bud-------------Y------------8.4g
In the next couple couple days, I will run these three through the full RxCE and report those yields and send one of those to the lab for potency. RxCE does two ‘wash’ reductions and winterization. I will measure the losses in the waste water and by-product of winterization to account for those.
Looking at the C0A, total cannabinoids at 24% and total CBD at 17% should result in 14g and 10g respectively. Either that COA is wrong or real world solvent extraction leaves a lot behind. What do you see with closed loop BHO and C02 extractions?
Minor observations - 1) I was surprised the chopped plant with 91% ISO didn’t return a higher bulk yield. The Grn test case was to simulate worst case scenario with plant penetrating solvent. 2) Decarbed plant that was then washed turned in a signifcantly higher yield. Not sure thats good, I expected to see a very slight reduction due to lighter Molecules. It will be interesting to see what RxCE does with this.
More to come…
(Edits: Sorry for the dashes in the table. Is there a way to stop the webserver from screwing with my text formatting? Can I change to courier font for the table?)
So last night after I completed the first 3 control tests, I thought a 4th one soaked for 30 seconds should be added to match the video session. Here’s the table with the 4th soak done this morning.
Label Alcohol 2oz Material Soak time Decarbed Yield
Grn 91% Chopped 10m N 7.8g
flower
Blu 99% 1/4in bud 10m N 7.1g
Red 99% 1/4in bud 10m Y 8.4g
Ylw 99% 1/4in bud 30s N 5.0g
Interesting it yielded 5g after 30 seconds but only yielded 7-8g after 20 times longer.
As mentioned above, I’m going to spend the next couple days doing the 3 RxCE runs to match the 10 minute control test soak times.
Observation: 3/5 is 60% of the 5g total. Add in winterization, waste water emulsion and transfer losses, this makes more sense. Its backed up with the potency rising from 34% to 78%.
I’ll be back with the 3 long soak tests.
(edit: added the pic…)
In our testing of most liquid solvents, 45 seconds was sufficient soak contact time to get about as much extraction into the solvent as you’re going to. Longer soaks didn’t pick up substantially more target compounds, especially at temperatures in the range of -20°C to ambient.
Morning!
Finally done. I ran a total of seven 2oz oil runs, four as controls, three as full RxCE processes.
Here are numbers -
Label Alcohol Plant Soak time Decarbed WW* Waxes Yield
Grn#1 91% Chopped 10m N n/a n/a 7.8g
flower
Blu#1 99% 1/4in bud 10m N n/a n/a 7.1g
Red#1 99% 1/4in bud 10m Y n/a n/a 8.4g
Ylw#1 99% 1/4in bud 30s N n/a n/a 5.0g
---
Grn#2 91% chopped 10m N 1g 1g 5.6g 75% ATC**
Blu#2 99% 1/4in bud 10m N .4g .8g 5.3g 74.6% ATC**
Red#2 99% 1/4in bud 10m Y .2g 1g 5.8g 69% ATC**
*WW = Waste Water. Waxes and waste water were boiled down and measured.
**ATC = Available Total Cannabinoids to solvent extraction.
Observations:
First, let me say, I coined the term ATC to qualify the Total Cannabinoids available to solvent extraction since the control tests show a best case 80% yield to the theoretical maximum TC listed in the COA. As with other things, solvent extraction appears to follow the 80/20 rule where 80% is easy, the last 20% is exponentially harder for every additional percent, just like charging batteries, etc. It’s worse in this case as the COA says TC=24%, CBD=17%. At 24%TC, the control cases show a 50% loss; whereas, the 17% shows a more realistic 30% loss. This leads me to believe the COA might not be accurate. A 10 minute wash then boiled down to oil should have returned a higher yield if 24% were accurate.
On to the observations…
The third control Red#1, and test Red#2 used decarbed flower (classic 40min at 240f in the oven…) with the results being wonky. The yield in Red#1 being significantly higher that the first two. I thought it should be lower, then realized it might indicate the first two had losses due to acidic cannabinoids being somewhat water soluble increasing those losses. But the actual test Red#2 yield fell in line with Grn#2 and Blu#2. So the 8g result is still open to discussion. The wonkiness is the fact Red#2 didn’t have higher yield and it flows like cooking oil, a fluid state at room temperature. So decarbing first then subjecting the wash to a water wash to purge water solubles does have an effect, to be investigated further.
Conclusions:
My definition of ATC defines a real-world maximum yield versus the theoretical maximum based only on COA numbers, but not attainable by solvent based extractions (I’m curious what BHO and C02 extractions can achieve). This number is determined by the 10 minute soak control cases.
Using the ATC as the real world maximum yield, you can see how I come by the 25% loss. Standard winterization results in a 15-18% loss by removing waxes. The water wash removes an additional amount upwards of 8-10% by removing water solubles.
These tests do show my 30 second quick wash resulting in a 3g yield is leaving some cannabinoids behind.
The one aspect of RxCE not yet discussed is ‘Rapid Winterization’. I’d like to explore that further here on Future4200, maybe a different thread? Here’s what it looks like. After the removal of the water solubles, redissolving the oil in ethanol results in room temperature wax formations, much like methanol. Placing this in the freezer for 15 minutes then passing it thru a nylon fabric results in the middle frame. Without a vacuum, I gave it a slight squeeze resulting in the third frame. This is the most significant benefit RxCE brings to the table, a way to winterize without the 12-24hr penalty.
(Edits: corrected Red#2 column width and decarb entry to Y instead of N)
oh, forgot to say, I’ll send Blu#2 and Red#2 to the lab for potency testing, will report back the findings…
Not attainable by single stage soak-only liquid solvent based extractions.
We regularly extract over 93% of cannabinoids from the biomass into our solvent in our liquid phase extractions on the first try, and with some tuning for a given biomass type/batch/grind we have been able to consistently get over 99%.
See here for a general description of the procedure: It washes the cannabis and then it washes the cannabis some more
MiniEmu…
RSO was green black goop.
You seem to be “touching” on the concept or concepts involving advantages of using water based extraction system…but back dooring into the field utilizing water mixtures of protic and aprotic polar solvents.
Moreover you have no designed endpoint of what it is you want to compose your new version of colored goop. RSO as a mixture for efficacy curing “what not” …is/was ill defined to begin with. Quo Vadis? Should we assume your newly defined goop cures “cancer as well”.
The only thing I can ascertain is with all that boiling of “the louches” , you are certainly decarbed eliminating all possibility of cannabinoic acids and volitile terps.
Start with pure water work backwards until you get goop.
You right RSO is green black goop. It is nasty to ingest. I started at zero almost 3 years ago during the covid lockdown to improve the oil. What I’ve identified is a simple ‘hack’ on making RSO, that has huge ramifications on quality.
“You seem to be “touching” on the concept or concepts involving advantages of using water based extraction system…but back dooring into the field utilizing water mixtures of protic and aprotic polar solvents.”
That’s where the 3 years has led me, an optimally ordered set of steps to produce near distillate quality oil in the kitchen. Realize you can do the whole process with Ethanol. Distilling in water removes the polar and semi-polar compounds is what optimizes the Rapid Winterization, not the solvent per se.
“Quo Vadis?”
I assume you jumped in on the end of the thread. Did you see the website and RxCE page?
Remember, this is for the home bound MMJ patient. Think of it as the Poor Man’s Refinement process. You can now take 2oz dried plant and make winterized decarbed oil that is just under distillate quality in under 2 hours. This oil is clean enough to be vaped, something you don’t want to do with RSO/FECO. My lab tests shows the process takes TC from 34% to 79%.
As far as efficacy in application, there are many successful stories of application, the research is catching up. Here is Gray Wolf’s article written back in 2012 on SkunkPharmReseach showing how it helped an individual with an opioid addiction. I have a friend in the UK who is cancer free after receiving a late stage diagnosis of bladder cancer. My focus is to improve the oil, not to prove it’s a cure-all.
Actually, by boiling off the alcohol in water, it acts as a buffer to retain cannabinoic acids and above 212f boiling point terps. It’s the reduction step where most elimination happens, so you can make adjustments to finalize the oil. At that stage, you have winterized oil in ethanol as a tincture ready for consumption, all under 2 hours.
That’s the success of this optimization.
Considering that CB1 and CB2 are GPCRs any extract of cannabis strains is going to “hit a bull eye” of some sort.
I actually like your work, especially simplicity and I understand the “preparation at home concept .“
If you want to be really “pure” about your process why not stick to FDA GRAS materials…in this case Water, Ethanol and EthylAcetate. Water alone can get you crystalline THCA hydrates.
Just a point of interest: FYI, the water is catalytic when it comes to decarbing the cannabinoic acids….so my question would be, have you measured the THCA/THC levels at the beginning and end point of various procedures? What is your target…do you want the the cannabinoics or the neutrals or do you want the mixture….water procedures offer a bit of tuning.
If you want a product you can eat in 15 minutes why not promote Bhang as a home remedy as well?
Three buckets and a Buchner will get you there…but you already know that. In the end point, you want “something”extracted from a cannabis plant in an alcohol tincture.(?)
“Something” is most likely cannabinoids. And as you are pointing out…you don’t need the green black (?).
Keep up the good work…excelllent.
Water systems will transform the cannabis industry.
Yes, I am taking the opportunity to “pitch”.
For a 93% return, what was the solvent temperature and soak time? What other factors are involved, such as centrifuge alcohol recovery?
The details are in the thread I linked, but the short version is that rinsing with virgin solvent - which is the same as multiple extractions/soaks - boosts your efficiency substantially.
Every time we’ve played with alcohol extraction, we’ve got over 99% extraction into the solvent, at temperatures ranging between -30°C and ambient.
There is a centrifuge involved in our method, but you can replicate the bulk of our results by using multiple washes.
