Yeah, whenever I have a CoA that shows anything abnormal, I ask for the chromatogram. It won’t always help figure out your problem, but it does help keep labs honest.
One testing lab refused to show the chromatogram because it was “proprietary”. I wasn’t asking for their method or anything so I called bullshit. I ended up getting a chromatogram and it was a total mess. Baseline all over, bad separation, etc.
Anyways, when you pay 50-100 dollars a sample for only getting cannabinoids tested, they should show you the damn graph your results are based on.
@vortal based on the COA you posted, it looks like you’re using ProVerde. You’re right, they are very professional and experienced. Seeing Chris talk back in 2014 motivated me to start KCA with my partners.
@kcalabs I have a hypothetical especially seeing as you guys haven’t seen compliant d8 yet… once you finally see compliant d8, give it your approval(here’s the hypothetical.) What if the dea says they say they see d9 in the product, how good of a defense will one have with results from your facility?
@globtek33 we do not “give approval” based on our testing. We simply report the numbers.
In terms of a defense of our results, it will come down to the validation of the method and how it compares to what the DEA does. From my understanding, the liability falls on the possessor of the material, which is why many labs can issue ND results with no repercussion.
I think further discussion can include the employed method, and comparison of qualiltatif vs quantification results. There is for instance forensic litterature suporting how gc is supetior to lc, and fid or ms superior to uv/conductimetry…
These cannabinoid testing labs are testing absolute concentration, so 72wt% imo is not that concerning since the analyst diluted a neat sample be fore it goes into the hplc. Now, of course they would dilute with an analytical balance and with eppendorfs, but this is something you need to do your due diligence on. How are they prepping these exactly? A small dilution error can easily lead to results like this being off from the actual. The next thing to look at is the relative abundance, which this sample shows a purity of less than 95% relative abundance of Delta 8 to measured cannabinoids. But hey, maybe the people you buy it from measure relative abundance another way, you need to figure that out. Lastly, the reason all these cannabinoid labs suck is cause they are medical labs designed to quantify absolute values. This is different from an analytical lab in significant ways.
@YumYam, sample preparation is critical to the analysis and arguably the most important part. We use an internal standard as a quality control measure to ensure samples are diluted correctly. If the sample is not diluted correctly the internal standard will reflect that and the sample will have to be re-prepped.
I’m not saying they did it wrong. An internal standard should always be used imo, and a calibrator run every week or so depending on usage. I think the problem here is that there are unknowns in this sample, which need to be identified with priority as unknowns can be more detrimental than a simple non compliance for d9, especially if the manufacturer doesn’t know what they’re doing other than making something “compliant”
I don’t even really care about the compliance at this point, even though for D8 to catch on we will need complaince. I want to know what these mystery compounds are. Mostly for my own curiosity.
I’m sure we can predict quite a few of them and measure them by analytical hplc. Think about what the reaction actually is and what possibilities exist for side products. It’s dependent on what you have present at the time of conversion but I’m sure I could predict several of them
Also ask for recovery values from method validation accuracy studies. Any lab with a validated method would be happy to share this. Unless the recovery is terrible, which would indicate a poor method.
I highly doubt this lab has a poor method… I should be receiving a chromatogram at some point today or tommorow, I was offered a total data package including results for QC’s, calibration curves, etc. for a fee of course…
Go all out, get a Waters UPLC. We have one in the lab I work in and these peaks look like tooth picks. Sharp and well defined. Love the thing to death but it can be a expensive to upkeep. PM and warranties recommended.
The reason why they didn`t want to give you the chromatogram is because it was a mess, although the “proprietory info” argument is valid as well. It takes thousands of dollars and months of work to develop those methods. If you are someone who has a solid HPLC/GC experience, then you can deduce a lot of info from the chromatogram and significantly reduce your cost of entry into the testing industry.
As for the baseline all over - it happens - not too much of an issue, software can deal with it. If using an HPLC, it can be caused by various things, such as gas bubbles in the mobile phase - shouldn`t be happening, but can; if in 2 analyses in a row, then it is obviously a problem. As for separation - most labs now use fast methods to increase the number of tests per hour. It causes an issue in separation - here, there is a lab that charges $210/potency test. They have 6 or 7 peaks coming at once within a minute. Separating d8-THC from d9-THC is not easy, many labs have issues with doing it as there is very little difference in molecular structure.
There are absolutely times when proprietary makes sense. I agree 100%.
I just don’t think that in most cases there is any top secret info to even give away. Most labs just run the method the HPLC company hucks their way. And most of the remaining methods are the same with just minor changes.
Things that make sense as properiety (at least at this point in time) is isomer separation or something else specialty.
I second that sentiment @BR-Instrument. But also as someone who spent months and thousands of dollars on an HPLC cannabinoids method development, I understand the proprietary argument - there typically is a way how to remove the proprietary info in the HPLC software, but if you have something unique then I guess people are defensive. Not all methods are the same - for instance, we developed a small HPLC at a fraction of the cost compared to the large companies. This came with lots of research that went into the method development and validation. Actually, to develop the method is not that difficult, but making it work properly is an issue - many threads on this forum about issues with cannabinoids HPLC methods and testing. Anyway, hope it wont happen again.