For consumers, it seems like an easy question to answer: How much tetrahydrocannabinol (THC) is in that cannabis-infused chocolate bar? For analytical testers, however, the answers are anything but simple.
According to research published last year by David Dawson, PhD, a researcher with Vertosa, a cannabis technology company based in Oakland, Calif., chocolate contains compounds that interfere with labs’ abilities to determine the dose of cannabinoids infused into the product. Dr. Dawson and his team developed precise solutions of cannabinoids—the active ingredients in cannabis—added these to chocolate, and then attempted testing to prove the chocolate contained the exact volume of cannabinoids they had added. Over and over, they were only able to prove that some of the cannabinoids were present, but not all of them.
“Any time we get less [cannabinoids] than we expected, that’s a sign they’re having interactions with the chocolate,” Dr. Dawson says. “The only thing we changed was the identity of the chocolate, and the quantity.” He and his team found a straightforward trend that correlated with a few factors, one of which was that the more chocolate product in the vial, the lower the recovery of cannabinoids. “Chocolate was absorbing, interacting with, and trapping the cannabinoids, reducing the amount in the actual solution,” he adds.
In his research, Dr. Dawson looked at the four most commercially available cannabinoids: THC, known for the “high” associated with cannabis; cannabidiol (CBD), known for a variety of medical effects but also popularly embraced as a relaxation and wellness tool; and two lesser known cannabinoids, cannabinol (CBN) and cannabigerol (CBG), which have only started to appear in commercial cannabis products, in comparatively small doses. Dr. Dawson tested the four cannabinoids in three types of chocolate: dark, milk, and cocoa powder. Each interfered in some way with cannabinoid measurements.
Dr. Dawson is quick to stress that chocolate only poses a problem for analytical testers wishing to get an exact read on cannabinoid contents. “This doesn’t affect the product as it exists in real life,” he says. “The [same] amount of cannabinoids is there, regardless. It doesn’t affect how high you’ll get from eating the chocolate; it doesn’t affect the amount of cannabis in the chocolate. It’s just how we measure it that gets miscalibrated. This is an issue for analytical testing labs, not chocolate producers, and not consumers.”
That may be encouraging news for consumers and producers, but it does little to solve the mystery of how to accurately measure the dose of cannabinoids in infused chocolates.
Interestingly, Dr. Dawson notes that the four cannabinoids he tested behaved differently from each other: Roughly 10% of THC and CBN remained trapped in chocolate products, while CBD and CBG “had a slight downward trajectory, but recovery was fine by any analytical lab standard.”
He realized that THC and CBN molecules have a single phenolic hydroxyl (OH) group, while CBD and CBG have two. After synthesizing a cannabinoid with no OH groups, Dr. Dawson was able to prove that the more OH groups a product has, the more likely it is to create “signal suppression,” preventing analytical testers from getting a clear measurement. “If you have two OH groups, there’s basically no signal suppression; interaction with chocolate is minimal,” he says. “If you have one OH group, there’s a mild or moderate effect. If you have no OH group, there’s a strong effect.”
Amber Wise, PhD, scientific director for Seattle testing lab Medicine Creek Analytics, notes that the 10% difference Dr. Dawson and his team found is not uncommon. “To put analytical test results in this field into perspective, people need to be able to expect at least a 10% difference in the signal [indicating cannabinoids] between any two labs, or any two experiments in a given lab,” Dr. Wise says. “If the signal is 5, 10% of that is 0.5—it’s not the difference between 5% and 15%. This is not a place where we’re unable to test chocolate for cannabinoids or get the right answer. If you had a bad experience with an edible, it’s probably not because of this issue.”
Cannabinoid Solubility in Fats
Dr. Wise joins Dr. Dawson in lamenting how new the field of cannabis testing actually is, and how this isn’t just about chocolate, but about cannabinoids’ incredible solubility in fats, from which it is very difficult to get the dissolved molecules back out to measure their combined dose.
In addition to dried cannabis flower and infused foods and beverages, clients of Dr. Wise’s lab have also brought her topical products for testing. Sometimes, the contents may be very simple, such as only coconut oil and added cannabinoids. Clients, accordingly, often figure that if the mixture is simple, the science behind it must be as well.
Not so, says Dr. Wise. “It’s actually really hard to get a number out of [cannabinoids dissolved in fats],” she adds. “People don’t understand the fundamental solubility issues occurring here. Most everyday people are only familiar with water-based chemistry.”
Dr. Dawson is familiar with fat solubility, so the outcomes he discovered weren’t a total surprise. After all, he says, “chocolate is a very, very complex food matrix.” Considering cocoa solids alone, milk chocolate contains roughly 50 different types, while dark chocolate contains roughly 70. “Those cocoa solids are 50 or 70 unique, identified organic molecules that contribute to the nuanced chocolate flavor,” he says. “On top of that, there are the fats—cocoa butter, say, or milk fats added to milk chocolate. Fats are chemically distinct from the organic flavor molecules. Additionally, we have the sugars added to chocolate. The organic flavors, the sugar, and the fat are three very broad, wide-ranging chemical classes. It is a very complex matrix.”
Helene Hopfer, PhD, the Rasmussen Career Development Professor in Food Science at Penn State’s Department of Food Science in University Park, Penn., concurs. “Chocolate, if you look at it, about 50% is fat—cocoa butter—[and] the other 50% is cocoa solid: starch, polyphenols, proteins. That changes during processing, during roasting. In order to get chocolate, you need to go through that roasting step to create those flavor compounds. It’s a complex food like a lot of other complex foods: Think of wine, distilled spirits, or tea.”
The chocolate products Dr. Dawson tested were 42% fat by weight, and cannabinoids are fervently lipophilic. “Obviously, the cannabinoids are going to have some kind of desire to remain in the matrix,” he says. “They’re having positive chemical reactions with the fats in the chocolate matrix. That’s something that needs to be overcome [in order to get a clear read on cannabinoid contents].”
Some clients have been adamant with Dr. Wise that if they added 10 mg THC per serving of their product and don’t see test results showing 10 mg, it’s a lab error. She disagrees: “It’s not my math that’s wrong. It could be your mixing, or a thousand other molecules in this mixture that are interfering with our signal as a lab. I’m happy to explore it more deeply, but that costs time and money.”
The bad news is that this issue isn’t just about chocolate but impacts any infused product with a high fat content. Dr. Dawson warns that this reaction could occur easily in any baked good, such as brownies. “Anything that’s thick and rich and creamy could very well display an analogous phenomenon,” he adds.
Dr. Hopfer agrees. “This is not a problem unique to chocolate. It will be similar with butter, probably.”
Where to Go from Here
Dr. Wise says that the issue of fats in foods causing diminishing test numbers isn’t impossible to resolve—the products just require a little more work. She suggests that formulators send their labs a “blank” solution of the same food product without cannabinoids. “We can run the blank mixture on its own to see if there are any interfering peaks that might look like cannabinoids,” she says. “We can also add a known amount of cannabinoids and do our extraction to see if we can get back all the cannabinoids we put in. That’s not perfect. David Dawson has been working on his research for a couple of years now; this isn’t something you can figure out overnight. But, we have ways of helping clients make sure there aren’t compounds interfering directly with their signal, and to be sure we can extract cannabinoids out of that matrix.”
Solvent extractions can work as well, Dr. Wise says, as long as labs are willing to be adventurous and try multiple solvents beyond the standard formulations. She suggests trying five or six different solvent systems to determine which one recovers the most cannabinoids.
But above all, Dr. Wise stresses, this problem isn’t the end of the road for infused-chocolate producers, or producers of infused products with high fat content. The challenges in testing chocolate, she says, are complex but, nonetheless, manageable. “As a producer–processor, there are a few things you can do in working with your lab.”