I’m about to set up my GC-MS for unknowns testing. We previously had it set for terpenes and residual solvents but the plan is to set up first for cannabinoids and then move to try and figure out the unknowns. I’d love to offer phytol testing. I think it could be a significant portion of first pass distillate, maybe even a few percentage points. Hopefully we’ll be set up soon!
3 Likes
So you’re going to use a database for the identification of new compounds? When you did terpene analysis did you calibrate for all your terpenes? I have never calibrated a GC only HPLC and UPLC for 11 cannabinoid standards. Im curious to see what you find!
We’re going to try to use a database Shimadzu provided but I expect will only get us so far.
My plan is to submit a sample of hemp mother liquor or crude to a lab which specializes in identifying unknowns. I found a lab that thinks they can ID 98%+ of the compounds in a hemp sample. We’ll take those results and try to apply them to our own system as much as possible.
Calibrating for terpenes was a huge pain, we practically never got a method together we were happy with for all 40 residues. How was running a GC for cannabinoids?
1 Like
I have never used a Shimadzu instrument so I am unfamiliar with their database systems. Agilent uses the NIST program which is great for potential identification. If that lab has standards for 98% of your compounds I would believe them but if not then they are probably just using NIST to determine what the most common structure might be. Either way its always great to have a second set of eyes, instrument, ect. I have never personally done the calibration for terpenes but the other chemist I work with hated it so much. We do not run the GC for cannabinoids, ony terps. I have a HPLC and UPLC for my cannabinoid potency analysis.
1 Like
Thank you for the reply!
To your first point about the potency of the biomass and calculating the expected crude yield based on the weight of input material… I said “we” in my first post, but I probably should have said “my crude supplier” instead. I don’t know exactly how much input/output there was in the batch of crude they gave me. I distilled it and got a TON of cloudy garbage out, which prompted my post here.
“We” does apply to the man running the GC though. He standardizes every day with third party samples, but I don’t believe he uses three different samples to calibrate. Is that what you mean by “triple point”?
Thanks!
We haven’t been testing for THCa and CBDa explicitly since the temp on our GC is set to operate from 200-250 C. Our assumption is there is no need to test for the acid forms explicitly since both would decarb in the GC and contribute to the THC and CBD peaks. Is that a valid assumption?
Also they have been experimenting with different temps in the FFE. Anywhere between 7 C and 75 C. There should be a fair amount of acid form in the crude.
Thanks!
Well then I’m not sure if its a low or high potency. If the crude is only showing 21.88% CBD that leads me to believe either the CBD in your crude is partially decarbed and you have a fair amount of acid form left in there or it really is low potency flower you’re guy is using to extract. If there is a way to get potency on the flower that would help you know what you should be expecting using that formula I provided before. If you’re distilling cloudy that means you have hella lipids and waxes “undesirables” in your crude. You can look up remediation techniques on here and those may help you get more pure crude.
So in order to do a proper calibration for your instrument you need to have a low, medium, and high potency standard with known concentrations of your cannabinoids in each. A triple point calibration just refers to this. You can also do it with 4 or 5 reference points as well to receive higher accuracy. If your chemist is not doing this then I would advise he starts.
I personally think that if you are testing crude then you should be calibrating for CBDa and THCa as well. That way you can do a simple conversion, multiply by .877, to get the total CBD and THC percentages. Just because your GC runs at that high of temps doesn’t necessarily mean that it will fully decarb your extract.
Hope this helps!
2 Likes
It’s definitely a warm ethanol and biomass issue. If you’re going to extract that warm you need more steps to remove impurities. This could be a water wash, filter medias, etc.
Also, leaving ethanol in your crude is messing up your results. Do you have any idea how much ethanol is still in there? If it’s a 1:1 ratio your crude is really at 43%. At least for your analytics you need to remove the ethanol. A full decarb will also partially remove more volatile compounds and terps from the crude increasing potency.
A GC test will decarboxylate any acidic variants so I wouldn’t worry about that.
3 Likes
Thank you, I appreciate the feedback and explanation of how to properly calibrate the GC! Everything you said makes sense to me.
1 Like
There is roughly 20% EtOH in the crude, maybe 25%. I’m going off of how much EtOH collects in my cold traps and how much crude I load into the distillation flask. I’m sure a fair amount will get past the cold trap and leave in the vacuum pump exhaust as well, so also taking that into account.
Why would I need to remove the EtOH for the GC to measure the CBD concentration? My understanding is that when calibrating the GC, a known mass of CBD is used. The mass of CBD in the calibration solution will correspond to an area under the GC’s curve. Then when we run our crude through the GC, the area under the curve tells us the mass of CBD that was in the solution we tested. Knowing the total mass of crude used in the test solution tells us the CBD % in the crude.
My analytical chemistry is a bit rusty so I welcome any feedback/corrections to that. Not trying to dispute you, just trying to understand.
Edit: When I say “crude” I’m talking about the bulk solution, including all remaining ethanol in it. I get that removing the ethanol will increase the concentration of all other materials.
Thanks!
You are totally correct in your understanding except that your determined concentration of cannabinoids in your sample is based off of either an initial mass or volume that is then diluted in a solvent to inject into your GC. If you don’t know how much of that initial mass/volume is from ethanol your sample will just appear more diluted.
For example, when I meassure a sample on my GC I meassure out 100 mg of sample then dilute in 40 mL of solvent. If my 100 mg of sample was half ethanol I would have no way of knowing that, my CBD concentrations would just be much lower.
Technically if your equipment was calibrated right and you had a standard and your solvent isn’t already ethanol you could use a GC to also determine your ethanol concentration and account for that. You can ask your chemist if he has set it up this way. I’ve never heard of anyone doing it this way though.
edit: Just saw your edit to your post. It just depends on how your are defining your crude. Most crude percentages you see on here are post decarb so no ethanol at all. It’s probably a more reliable metric because every run can leave different amounts of ethanol in your crude and give you inconsistent results
2 Likes
Thanks! That all makes perfect sense. I believe the solvent he uses in the GC is acetone, so he could test for ethanol explicitly if he used a wider temp range in the GC. Knowing the ethanol concentration in the “crude” that we receive will help a lot with keeping production consistent.
I get why we would want to define crude as what is left after decarbing and removing all ethanol. I think that’s a better way to compare across samples. Thanks for clearing that up.
Appreciate all the help guys!
Take it to a lab and get a hplc test done.
Did you ever get results back on this distillate-unknowns ID-testing? Interested to see if you’d be willing to share some knowledge just to be kind and helpful!
Deleted
Reposting as a reply:
We didn’t get our in-house system set up for unknown compounds but we outsourced successfully to the lab I mentioned.
Recently we had an ID of 80% purity on some crystalline terpenes separated from distillate. It was a collaboration with another lab who made the crystals. It was the first time I’ve ever seen it identified or highlighted which is sort of cool. It seems to be in almost all d9 distillate
Hoping to release that data soon!
1 Like
Not all compounds have been identified so…