Isomerization to Delta 8 THC


I work in a medical cannabis lab and recently we have been discussing ways to isomerize either CBD or d9 THC into d8 THC. I’ve seen several processes that claim to yeild this result but before I start dumping H2SO4 into prefectly good oil I wanted to discuss the best approach. I am also working without a hood and under some strict local guidelines so I wanted to discuss alternative solvents in some cases.

From CBD:

Dissolve 1g of CBD in 10ml 0.005molar H2SO4 in glacial acetic acid.
Let solution stand at room temperature.
After 3 hours CBD has been converted into 52% delta-9-THC and 2% delta-8-THC.
After 3 days a mixture of 15% delta-9-THC, 54% delta-8-THC, 10% delta-8-iso-THC and 10% CBD has formed.
Work up:
Pour into water (100ml or more) and add cautiously with stirring (foaming!) enough sodium bicarbonate (15g should do) to raise the pH above 7. Extract with petrol ether, wash with water, evaporate solvent to get mixture of starting CBD and predominantly delta-9- or delta-8-THC with some other byproducts depending on the reaction time.

Question 1: is the chemistry in this SOP sound?

Question 2: could I substitute heptane for the petrol ether in this SOP?

Question 3: in the “wash with water” step would it be better to use straight up RO water or to use a saturated saline?

I’ve also read and heard from several sources that the use of activated carbon or a strong acid in a short path distillation can also isomerize d9 to d8. I have not come across any SOPs that sound legit, however. Does anyone have thoughts on a direct isomerization from d9 to d8? In particular as part of a short path process?

I appreciate everyone’s input! We have several patients who could really benefit from d8. Your help helps them!


do you have the ability to tell in-house if you’ve made delta 8?
Or will you be paying a stranger? (3rd party analytics).


It will be 3rd party. We are mandated to use a particular lab and we do not yet have in house analytics. Our lab is able to differentiate between the isomers though.


Dialling in your process will take more than a couple of tests. Getting in-house analytics early rather than late would be my advice. even TLC, which can be had for the price of a couple of 3rd party tests, would presumably give you a leg up on this.

I say presumably because I’ve not seen delta-8 vs delta-9 on a TLC plate.

I have seen fairly high purity delta-8 on my SRI GC/FID.


I don’t however know how this sample was made. @Photon_noir does.

I blame the peak shapes and rather poor resolution on the abuse the column has taken recently (only some of it my fault). I believe a new one has been ordered. The numbers are fairly solid, although all calibration was done with CBD isolate (so it won’t be perfect).

Edit: these guys used 2-D TLC, and claim resolution of delta8 from delta9. they list different migration for delta-8 and delta-9 in both dimensions, as well as different color staining with fast blue. so the cheap TLC kit should get you close enough to do your initial work ups.


Hmm. Well it looks like going around the hand to get to the thumb, and the thumb is gloved in CBD and Δ9-THC and (Δ8-iso-THC?) you do not want, which causes excessive need for purification. Why not just CBD in solvent to Δ8 via acid with a non-polar affinity? Use the alkaline beam test to determine completion, then analyze. With that, I have gotten 90% Δ8-THC and 5% Δ9-THC at just over room temp in less than an hour… and that was using synthetic CBD which tested at 96% pure, but which had unknown reagent left in it. I was not aware of the nature of the CBD impurity at the time, but it showed up as 1% THCa on single wavelength UV detection of chromatography, which made no sense. Upon observing the DAD of that peak, it was shown to be something else entirely. Only after cross analysis with nIR and modeled algorithms was I able to ascertain that it was either a cutting agent or leftover reagent of some sort. Still unsure what it is, but using pure CBD next time will yield better results, I’m sure.


Thanks! That’s great info. What acid would you recommend? Would the H2SO4 in glacial acetic acid, as discussed above, create the correct environment?

1 Like

Sulfuric acid alone is better than a mixture with acetic acid, but I noticed even “pure” (98% in water) sulfuric forms more byproducts than the acid I prefer to use. PM me to apply for a small consultation if you like, or simply read more of the literature for acids commonly used in the process. :wink:


Photon_Noir, I am interested in contracting you for assistance with this subject.


Okay. You can PM me here or on Instagram. :blush:


Use a lewis acid bf3 etherate?


According to wikipedia… BF3 can bond to even weak lewis acids.

Googling “Can double bonds act as lewis bases” described a reaction involving BF3. The ether part is also aprotic. CBD to THC?

1 Like

Yeah maybe nvm just looked at the MSDS haha…

1 Like

@slooty @QGA BF3 is one type. There are many. Most of them are pyrophoric or otherwise quite hazardous and difficult to work with. Even Aluminum Chloride works.


General question, is the acid you are prefering to use a lewis acid?

1 Like

Yea cbd to delta 8 and delta 9 are both patented processes you can look up and from my memory they both use lewis acids.


Yes. As I said above, Lewis acids that are not also Brönstead acids work for CBD to Δ9, while Brönstead acids work for CBD to Δ8.


I’ve got a jar of that(ALCL) laying around from a terp based polymer project…

1 Like

Oh? How old is it? Has it been opened in moist air?

1 Like

Less than 4 mo old and only opened once in CO…so normally very arid air.

1 Like

Nice! I’ll take it off your hands!

1 Like