What's causing this spontaneous decarbbing?

I oversee operations in two CO2 (waters extractors) extraction labs and ~thought~ our process was identical. We’ve been seeing a very confusing situation: in one facility for the past few months, our finished oil has been spontaneously decarboxylating slowly and interfering with our cartridge filling.

What happens is that once it’s warmed to about 45-50C and perturbed (stirred, swirled, pushed through a syringe tip), it will decarb, which forms bubbles and if it’s bad will cause the oil to foam out the top of the cartridge before we cap them. I have confirmed it is CO2 gas forming by bubbling it through CaO solution. We don’t fully decarb our oil, as it’s full-spectrum and there is that pesky mass loss we’d like to avoid if possible. Here is a link to a video (the audio is irrelevant). The oil is warm, which explains its low viscosity–I stir it to show the bubbles form, wait for some time to let them rise to the top and then repeat to show more perturbation will cause more bubbles to form.

Other things I have eliminated:

  • The material itself–this occurs with bud we’ve grown ourselves (organic/no pesticides) and with outside growers’ material (also tested clean for pesticides)
    *the extraction methods and settings should be the same between instruments
  • The solvents have all been switched out for new containers and batches, and this still occurs.
    *the average ratio of THC:THCA is similar across both facilities and doesn’t seem to matter, if there’s any THCA left, it will slowly decarb at these relatively low temperatures (~40-50C) and atmospheric pressure conditions.

Has anyone ever see this before and know how I can prevent it from happening to stabilize the oil?! I’ve been struggling with this for awhile and am starting to run out of variable to eliminate…I unfortunately cannot transfer pre-processed or in-process material between facilities as they’re in different states and the boss says it’s too risky.

you’ve nailed down that it’s CO2.

Have you done the potency analysis to prove it’s decarb?

I’ve seen that at my facility before we changed our decarb/terpene processes. I’m mostly surprised it’s only happening at one facility. We always just decarbed more to get rid of the issue. Now we decarb fully before reintroducing terps.

I’m inclined to think that without some type of wax content, decarb doesn’t require quite as much energy, especially after being exposed to rapid pressure change. I don’t have any real science associated with this concept. I have noticed that exposed terpenes are vastly more volatile than when suspended in waxy dabs. So, perhaps cannabinoids can be more reactive as they’re more exposed.

1 Like

We made the same modification at Care By Design, more thorough decarb before flavoring and filling.

1 Like

I have looked at potency over the course of warming and stirring/perturbing. After a couple hours , the THCA% change is barely / not detectable. After 5-6 hours of warming/perturbing, the THCA% reduction is only a couple percentage points.

Do you feel that points to decarb? Enough to cause foam?

is there ZERO THCA% reduction in the non-fizzy facility after six hrs of poking at it?

I’d be more inclined to blame something in the resin holding on to your extraction solvent until you warm it sufficiently. Although that’s pretty hard to argue with two folks telling you they “fixed it” by decarbing pre-extraction.

Unless you posit that it’s the THCA that’s keeping the CO2 in solution…

Wouldn’t she be ridding the solution of residual extraction CO2 in the dewaxing/rotovaping steps?

Derive the terpenes from cannabis. Or buy cannabis derived ones. I personally believe that the plants know what terpenes go best with what ratios of cannabinoids. I have also been referred to as a “typical folk medicine retard” though. Not sure what’s accurate.

Sure, but are those a given? or an assumption?

My understanding is that not everybody uses second solvent dewaxing on their oil.

We certainly had a picture up here for a bit recently of carts being filled with what was purported to be unwinterized CO2 extract.

I’m also playing devils advocate here, because exploring only one side of this might leave us ignorant.

Edit: yeah, you know you winterize. crap…

Oh, I was under the impression no one was selling unwinterized for at least a few years. I suppose I could be wrong, I would like to think people know better.

yeah, that was apparently the way the poster of said image felt too.

I just figured if the hydrocarbon and ethanol aficionados can bypass initial pickup, it should be possible with CO2 as well. I have zero experience (with CO2) to back that up.

I don’t believe the foaming is due to any specific contaminant bubbling out–I have proven it’s CO2 gas by capturing it and bubbling through an aqueous solution of CaO, which turns the solution cloudy (general chemistry experiment).
The reduction of potency is difficult to measure with high accuracy without multiple replicate data points, which I’m unable to obtain at this time. I basically have one data point at t=0, t=2 or 3 hrs, and t=6 or 7 hrs for 3 different samples. I do see a small reduction over many hours of periodically stirring and allowing the bubbles to escape, but after only a couple hours, the is no change within the accuracy of the method.

There should not be any residual solvent that’s escaping at atmospheric pressure that didn’t already escape in the hours of rotovapping at reduced pressure.
The craziest thing to me (as someone else mentioned) is that we never see this at one facility and we see it with essentially every single batch in our other lab. But it wasn’t always this way with that other lab…it used to be normal/well-behaved.

And to answer the Stonejh77 (not very related question) we only use 100% cannabis derived compounds (keeping each batch separate and strain specific) in our final formulation. If you read all the comments above, I mention we don’t fully decarb our oil before finishing–this is the problem. It is only partially decarbbed after the CO2 extraction and is not decarbbed any further in our post-extraction process. We may ultiimately need to decarb to prevent this if I can’t pinpoint the problem and solve it, but I’m hoping to avoid the 5-6% mass loss afterwards.

Ok, so you do winterize, so while not given, it was a safe assumption. so much for that idea.

The fact that it is only at one facility is fascinating.

I can’t tell from the above response if you are seeing an equivalent drop in THCA over 5-6 hrs at the non-foaming facility. lack of said drop would certainly point at decarb as the primary CO2 source.

Yes, we do winterize, however, I do know of at least one other company that fills carts with unwinterized CO2 oil.

This bubbling is resolved by decarbbing the oil completely near the end of the process. So ultimately, I can solve the problem, but I worked pretty hard to eke out a few % gains in extraction efficiency this past year, don’t want to throw all that away with a 6% mass loss in the last step. And also just fundamentally as a scientist I want to know why one location and not the other! :laughing:

Another science-related discussion topics related to this issue: When we do decarb fully, the finished oil is a couple shades darker–even when I try to do it as “gently” / low temperature as possible. What’s the easiest way to minimize/prevent this? Using nitrogen blanket during decarb process?

1 Like

Use inert gas, the color change can come from plant sugars, particulate, and anything else in there that will turn brown from heat. Think about fry oil, it is pretty light colored when you start. Near then end when all the little particles turn brown the oil appears brown. After the oil is used so many times it is filtered. The resulting oil is lighter, but not as good as new fry oil. This is because some of the dark colors are soluble in the oil.

I’m just going to venture a guess here, but iconsidering that the situation is different at different facilities, you should look for differences that are catalytic in nature.
This ioil when it leaves the system is put in a series of containers, I would literally walk through the process from there and see who touches the containers and how.
Look at differences in the water at the facilities. old pipes, etc.
The only other thing i can thinkof is that the acidic environments in the extractors themselves might be different and residul acidity (if that’s even possible) could be helping the decarb along in one place or another.
I’m sure when you find it, the answer will be highly revealing.
good luck. -Bone

1 Like

Maybe your temp readings are wrong? Try switching the thermocouple or IR, or use both, just to rule that out. Maybe some weird operator error?

As a co2 extractor who decarbs almost fully prior to extraction I would like to see more on this and learn more about this. There are a lot of factors which contribute to final coloration of an extract and understanding more about every facet of that interests me greatly. Should we maybe start a separate post/thread?

Another contributing factor to darker color is your settings on the CO2 extractor. We found that just backing off the time per pound by 30 minutes from 2 hrs to 1.5 hrs greatly lightened up the crude coming out of the machine. Also far fewer waxes too. A good lesson in diminishing returns. Takes a while to dial in your machine but it pays in the end.

That being said, full decarb before extraction will darken your crude.