This.
1 Like
Absolutely!!!
Agreed 100%. Even after @StableScott narrowed the outliers, there is still a huge difference between highest and lowest numbers for 8 and 9.
2 Likes
I like how KCAs method VERY clearly doesn’t resolve delta 8 from delta 9 (they’re conjoined peaks right there in the attached COA) and yet everyone be dick riding without proper justification or without criticism of a flawed method… If you’re an analytical lab, you should be able to resolve the two very easily.
I mean, I’ve been using a method for like 5 years that can resolve THC isomers - its pretty standard analytical chemistry/method development. For this reason, it is my experience that KCA shoots low on delta 9 in CBD conversion samples because a bunch of the delta 9 area overlaps with the delta 8 area at the wavelength of integration. Their results should be subject to criticism if they can’t even separate the two compounds they’re allegedly quantifying.
3 Likes
This should be TWO peaks, with some discernable baseline between them. If this is the method being used to quantify delta 9 and delta 9 (no distinct resolution of compounds + using the split peak integration tool), its flaws are glaring.
1 Like
Here is an example of the data used to quantify the d9-THC. It is not this particular sample, but one very similar.
The issue isn’t the resolution between d8-THC and d9-THC. It is the coelution of other isomers.
8 Likes
I agree, that coelution is awful. There are probably six or more isomers in there.
1 Like
This ones a new isomer- D8/9/exo-iso/6a10a-11, you can buy the reference standard at the gas station
7 Likes
Its not a simple matter to confront, no doubt. I’m sure you’ve got a handful of methods up your sleeves, what works best for one analyzing one series of reactions isn’t what works best for another scenario. I’m sure it doesn’t help that you’ve got to attempt to make one extremely robust method whereas if you had the luxury of knowing “this is CBN via iodine material” versus “this is delta 8 via sulfonic acids material” you might prescribe methods that are specialized towards the specific product/impurity profiles. I’ve certainly been involved in tailoring methods for different projects.
2 Likes
I’ve mentioned on here before that we run the same HPLC method on every cannabinoid concentration sample that comes through the door.
These distillates are a whole different world compared to the biomass and isolates running alongside them.
We run a couple dozen analytes on that HPLC, so we sacrifice resolution. LC isn’t cutting it when we’re dealing with several isomers eluting right there.
4 Likes
Have you considered throwing some CBD and p-tsa in a vial on a hot plate and just generating these impurity profiles for your methodology pleasures? Just like…5 grams CBD, 5-10 mL toluene and like 5 mol% p-tsa and burn that at like 100-105C for like 3 days you’ll have the impurity profile you need to do some method development?
5 Likes
Yes, we have a few things we’re currently working on similar to this.
To be frank, we’re competing against labs that charge $25-$35 per sample on both fronts, the farmers’ samples locally and the concentrates, products, and reactions nationally. There are a million things we’d like to do, but you can imagine how priorities get shifted to try and survive in this market.
6 Likes
hell yea, there ya go! good guy KCA warming my cold heart <3
6 Likes
Yrah but that goes for all cannabinoids too- I see up to 20% variance between labs with CBD and CBDa too. Of course its less complex, but these places struggle with the non-linear upper end of their dynamic ranges. Shit gets weird up there without at least an 8-level calibration curve. That becomes a problem for all hemp products when you formulate and you’re too high or too low, especially in states where they have random product testing protocols. Label claims can be based off of results that are wackadoo, placing blame on the manufacturer. The analytical facilities don’t have the product liability if they violate program rules. Gotta test at multiple labs for that reason. Sucks bad to be accused of false label claims- is it the lab who analyzed the product prior to release or the lab who was used to validate the label claim on the other end? Short answer is whomever the program specifies, but they aren’t always right, so yeah its a total nightmare. It’s always the manufacturer is “shady” or “con artists” (still true a lot) but sometimes it’s the lab is “not very good at the one thing we pay them for”… still left with the question of who’s right? .1% in the wrong direction is the difference between a hemp product and scheduled drugs lol. Tough game, folks talk shit but if the mj laws made all the mj processors remove something from their thc thats in the plant already, they’d be bitching lol
5 Likes
I agree with this 100%. Nothing like having to decide between having to bin a whole batch because the compliance test came back way out whack or putting out on the market with a label claim that you know to be BS. Heartbreaking honestly
2 Likes
Yeah and its not like you can be like “i need you to refund me for all these nonsense results that are wrong and worthless” lol. Even running in-house analytics before getting 3rd party analysis is only as good as the 3rd party anyways. Nobody cares about those in-house chromatograms
1 Like
Compliance testing frequently just feels like a tax more than a meaningful service. We’ll probably hit 6 figures in test results we know to be incorrect by more than 15% this year… It’s very very easy to build a validation regime to demonstrate that a lab’s results are not correct but then what do you do? This is where I envy the CBD lab options which aren’t so limited
2 Likes
Wish I could like this 1,000,000 times!
2 Likes
There is a difference, however, the RSD is now 3.37% for d8 which is awesome, that would pass for me. Although the one for d9 is still terrible, and I’d still be removing people, until the data set looked good, AND I’d let them know why AND I’d send the remaining group another set of samples just to make sure. At least we know they can find d8 successfully. I’m going to assume its S/N ratio issues and other automagical integration activities that are causing the issues with d9 - the peak is too small, much of its area is being arbitrarily disregarded, and/or they are adding more area in than they should by tapping it manually.
1 Like
We have good resolution between d8 and d9 on three different platforms.
I’m not sure how many different ways I can say it. The coelution on the HPLC isn’t because d8 and d9 aren’t resolved.
The peak distortion is caused by coelution of other substances that are produced during the synthesis of d8-THC, therefore we use GC-MS to resolve these other substances from d8 and d9.
Happy to have a constructive discussion about this. Maybe we should bring ISO into the discussion.
Do they know there are accredited labs unable to detect the presence of a Schedule 1 compound with their accredited method?
5 Likes