Looks like I am based on their website.
Not a one trick pony like the other devices I’ve used.
Guessing it won’t be trivial to achieve your goal, because training it requires feeding it a set of knowns, and we’ve seen how problematic that can be with converted D8
Please share what Perkin has to say on the subject.
So I can learn something new… 
Of course! I’ve gotta run and grab some dry ice for the short path then I’ll make the call.
I have used FTIR for several years from positive/negative identification vs a standard to using it to troubleshoot the presence of an active ingredient in a formulation or extraction solution. Feel free to shoot me a PM and I’ll be happy to help provide any insight and help that I can.
I had some fun at the Spannabis, where some people where showcasing the Sage stuff, and also the Gemmacert… I could fool those device with a 7% CBD flower. They would say this is a 15%THC + 15%CBD, or things like that…
The various form of IR spectroscopy is not on point at all for such complex matrices.
It could work well for pure d9/d8 mixes… or I believe could be a major asset for just identifying roughly the cannabinoid profiles of natural samples (I was thinking Raman would be even better).
The basic thing with such device, is to have an access to the raw data output.
That’s the starting point… and then it needs thorough work of background determination, data de-convolution and modeling, and eventually calibration for specific matrices…
A reason why the cannabis devoted device that are sold to people are pretty useless…
The only thing I’m really trying to achieve with this thing is wether or not my reaction produced pure D8 or not. If it shows that the sample contains only D8 then I’ll send that sample out to someone who actually knows what they are doing. It’s a cost savings thing. I was hoping this would be a simple quick way to say ok this looks promising let’s send it out or this has a bunch of unidentifiable stuff in it so the process didn’t work for what I’m trying to achieve.
TLC can tell you alot about side product formation. It’s an art to some extent, but once you get it down a great inexpensive tool.
What do you mean by pure d8? Under 0.3% d9?
Or under 0.1% d9? Etc. There will always be at least a few molecules of d9, nothing is completely pure. So what you are asking is if the ir you have is capable of detecting x amount of d9. The answer is ir isn’t great at quantification of molecules. Is good at qualitatively looking at them. Yes some advances have been made but you’ll have to calibrate and that’s a major pain. Better off using gc or hplc. Notice none of the 3rd party testing labs use ir for the tests they sell
Yes I do and I had an Agilent 1200 But I sold it months ago unfortunately. This IR actually came with the HPLC on a trade deal I did. If its not going to do what I need I’ll probably just sell this thing. I’m having a hard time letting it go though because I learned my lesson on selling the 1200.
Let me know if you decide to sell it! Does it come with software?
It seems to. It has a thumb drive plugged into the USB on the tower. I powered it up one day and from what I can tell everything is there including the software.
dm sent, not sure how to delete posts…
Ok so Perkin said that as long as I have a standard to compare against I could pretty much detect what ever I want. The gentleman told me that this is going to just spit out a bunch of peaks. Like the Pics from @bigbone. He also said we could build a quntum model to asses percentages.
To start this investigation, you need a set of characterized samples with contrasting d8/d9 compositions. Ideally, you should have pure D9 (d8 <0.01%), and pure D8 (d9 < 0.01%)…
Overall, since we are dealing with d9/d8 dominated matrix but still with varying other contributors, I hardly think you could get below 1% detection threshold. And that would already be an achievement.
Yeah there’s a good chance any identifying peaks will be obscured by noise at such low concentration.
Has anyone done this? Never tried FT-IR on cannabis flower, but I really doubt you can distinguish between D8 and D9 as there is only 1 adjacent double bond difference - I doubt youd be able to see difference; I cant see how the technology would allow for it.
Sounds right @JedClampet. That`s all that should be needed - buy standards, run those and then run samples to interpret - it may not be easy or possible due to the technique being an FT-IR, which seems to be interpreted only within the major functional groups.
I haven`t run FT-IR on flower - always wondered why a light-based technique is done on a solid material/flower - why not extract first? Sounds like a whole world of trouble introduced when doing light-based techniques on a flower.
Basically you would analyze pure d8-THC and save the spectrum and then analyze pure d9-THC and save the spectrum. If you overlay the 2, or perform a spectral subtraction in the software you might find that there is a peak that is unique to d9-THC that’s not seen in d8-THC. Then you would take the presence of that peak in a spectrum of your product to mean the presence of d9-THC in your product.
