Article: Testing Old Cannabis and Why We Need Stability Testing


In Canada if you take a look at the label on the package of the dried cannabis flower you bought you might notice that an expiration date isn’t noted and says “No Expiry Date Determined”. In June of 2019 I had posted about two products that were eight months old and Thomas Fraleigh, President and Founder of Vivariant Laboratories reached out and was willing to test this aged cannabis as my concern was on potency loss and not getting the full value from my purchase.

Before we get into this further I’d like to caution that it’s hard to say a drop in THC is 100% due to age. As Thomas advised you’re always going to see a difference when you test a licensed producer (LP) bud and compare it to the label value. Below he speaks of the sampling challenges and stability testing, and from there I take the results and give my final thoughts.

Sampling Challenges

Cannabinoids are not distributed evenly across a cannabis plant. A bud sampled from the top of a plant and the bottom of the same plant can vary in concentration, same goes for different clones in a grow room. Even on a single bud the trichomes have a distribution I often liken to “marble cake”. 

When a customer provides a sample for potency we typically ask for a minimum of 1 gram but suggest 5 grams. We then grind the sample to a homogenous (even) powder from which we take the material we actually analyze. The grinding is an important step because it takes the variable product and creates an average however that average is only as good as the input material that the customer provides. If the customer only submits one bud, the average result we return will be reflective of that one bud. If they take six buds from various parts of the plant, we grind those to a powder and that average is then more reflective of the whole plant. 

But how many plants, and what parts of those plants do you pull from in order to sample a whole grow room at an LP? Unfortunately there is no clear guidance on that matter right now. Health Canada (HC) has very carefully removed themselves from responsibility for sampling stating instead that it is up to the LP to make sure that they have a “good protocol for sampling”. Health Canada has also not placed any formal size definitions on what constitutes a batch or lot, so you could have someone sending in 1 gram sample for a 10 KG batch or a 1 gram sample for a 100 KG batch. 

The only time that HC has ever formalized any cannabis sampling plans was the now-obsolete 2004 Industrial Hemp Technical Manual. During this era each and every hemp farmer had to have their field sampled by a third party lab to check that THC levels were below 0.3%. The lab would send a technician who would walk in a very specific manner through the field and collect 60 flowering inflorescences (think of a branch off the main stem with the leaves and flowers), divide those into two groups of 30, 30 stay with the farmer as a reference and 30 go back to the lab for analysis. The THC determination would be based on the 30 inflorescences sent to the lab. This document is now obsolete and these guidelines no longer apply. 

Today the only hemp farmers that have to do THC testing are those breeding viable seed and clear sampling protocols for that testing have not been re-issued yet. I call attention to this old document though because it’s the only time that HC has given clear sampling guidance for cannabis/hemp because they wanted to keep an eye on the “evil THC” in hemp. They don’t seem as motivated these days to issue guidance on sampling for determination of THC and CBD at the medical and adult use cannabis producers. While mandating 30 branches sample size from a cannabis crop to determine potency would cause the LPs to riot it would be nice to have some sort of sampling guidelines from HC on the topic. 

It’s important to note that even with a huge sample input amount, which is going to give you a great representative average, if you were to re-test buds from that batch against that value you will still have outliers. There were some articles going around in USA from when state markets and testing labs were coming online with headlines like “omg I sent my buds to two different labs and got two different results” and that is to be expected for all the reasons we discussed above. I had a friend ask once why we don’t present values in ranges which I thought was a good question. My thoughts are that people in general don’t like ranges and the labels are also confusing enough as it is today. If you were going to go with a range you would need to conduct multiple repeat tests so the cost would go up a lot as well.

Personally I think that the single value is sufficient, we do need more consensus on sampling practices. Note that for products like oil solutions sampling is less of an issue because as liquid solutions these products are more homogenous to begin with. The problem re-surfaces again in edibles and topicals though. Cannabinoids are highly viscous and difficult to blend. If your mixing process does not evenly mix your cannabinoid you’re back at marble cake again, perhaps literally! 

Stability Testing and ‘old ass weed’

When we were first introduced on the topic of some old cannabis you received from an LP. We discussed how testing cannabis for cannabinoids isn’t necessarily a clear indicator of stability. There seems to be a consensus that THC breaks down into CBN but the kinetics of that are poorly understood. Stability studies are used to determine expiration dates and shelf life. They involve taking many products in its final container and putting them in a room of controlled temperature and humidity (“stability chamber”) and periodically pulling some out for analysis. 

When you test for stability you need to have stability markers, for example, what do I test for to show that this material is still stable? One thing you could do is repeat the tests on the Certificate of Analysis (CoA) until it fails to meet spec. For cannabis there is no pass/fail limit for cannabinoid content and the variability in cannabinoid content in between buds (as mentioned above) is going to complicate this. 

So what else? Microbes are one option. Keep in mind that microbial cells don’t grow at low water concentrations so very dry cannabis won’t show much microbial growth over time. Pesticides are unlikely to magically appear unless they leach in from packaging components. Metals may be leached from the container over time, especially vape pen products. Terpene content may diminish over time. The thing is, I think you can have old cannabis that still passes quality control long after it has lost it’s bag appeal. Again, there’s no consensus on how stability studies are performed especially on dry flower. If you meet LPs who are engaged in GMP production this is an interesting question to ask them as all GMP products must bear an expiration date and that expiration date needs to be determined by stability studies. 

The lab results

First in the lab Jean Guy was tested from Aphria which I ordered from the Cannabis by Shoppers Drug Mart (SDM) medical program. The label was stated this was packaged October 9th 2018 and was tested on September 18 2019, the potential THC amount was noted at 17.24% on the bottle.

According to this report the Total THC amount was 16.9%, a slight loss compared to the advertised potential of 17.24%. 

The second lab tested product didn’t fair as well and was Rockstar from Tilray which was also ordered from the SDM medical program. The package date was October 17 2018 and as tested September 18 2019, the potential THC amount was noted at 24% THC on the bottle.

According to the lab report the Total THC amount was 18.2%, that’s a large difference and not the potential 24% THC I paid for.

Final thoughts

One product had a tiny change and one significantly had a lower value. The takeaway here is the word “potential” THC on the label and the proper need of stability testing to determine when quality begins to degrade and when does it go bad. The cannabis I had wasn’t expired but the quality wasn’t where it should’ve been.

Vivariant Laboratories is one of the authorized laboratories in Canada to conduct analytical testing of cannabis under the Cannabis Act, and you can find more about them at for more details on potency testing that starts at $125 each and you can reach Thomas Fraleigh directly on Twitter under the handle @tfraleigh.

Thanks for reading!

Posted in Articles and tagged .