Formulation notes

Why edibles take ninety minutes to work

Ask three Oregon brands how long their gummy takes to come on and you will get three ranges, all of them wide. The width of that window is not a formulation failure. It is what the digestive tract does to an oil.


Wyld publishes fifteen to forty-five minutes, and up to ninety. Gron publishes twenty to ninety. Smokiez has written that it can be thirty minutes to two hours. None of those companies is being evasive. They are describing, accurately, what happens when cannabinoid oil is eaten.

An oil has to be emulsified before it can be absorbed

Cannabinoids are lipophilic. They dissolve in fat and not in water, and the contents of your gut are mostly water. Before any of that oil can cross into the bloodstream, the body has to do the emulsification work itself: bile salts from the gallbladder break the oil into droplets small enough for lipase to act on, and only then does absorption begin.

That process is not on a timer. It depends on what else is in the stomach, how much fat came with it, when the person last ate, how quickly their stomach empties that day. Two people eating the same gummy are running two different reactions. So are the same person on two different days.

Then comes the liver. Everything absorbed through the gut passes through it before reaching general circulation, and the liver converts a large share of orally ingested THC into 11-hydroxy-THC. That conversion is why an eaten dose behaves differently from an inhaled one, and it adds its own variable delay.

What a nanoemulsion changes, and what it does not

An emulsion does the first step before the product is ever eaten. Instead of asking the gut to break oil into droplets, the oil arrives already dispersed in water as droplets in the tens to low hundreds of nanometres, stabilised by surfactant.

Two things follow. Surface area per unit of oil rises sharply as droplets get smaller, and absorption happens at surfaces. And the rate-limiting step, waiting on bile to do the emulsifying, is no longer in the path, so the timing depends much less on what the person ate for lunch.

What it does not do is bypass the liver. First-pass metabolism still happens, 11-hydroxy-THC is still formed, and an emulsified edible is still an edible. Anyone promising you inhalation-like onset from something swallowed is selling something.

Why this is a commercial problem, not just a technical one

A ninety-minute window is the single most common reason a consumer takes a second dose too early. They take one, nothing happens, they take another. The product did nothing wrong and the experience is still bad, and the brand on the package wears it.

It is also the part of the category a competitor can take from you. When one brand on the shelf can describe its onset as a specification and the one next to it publishes a range an hour and a quarter wide, that difference does the work of a lot of marketing.

What to ask a supplier

  • Droplet size distribution, not the average. A mean of 100nm hides a lot. Ask for D90 and span. The tail is what separates and what behaves unpredictably.
  • How the emulsion was made. Sonication and high-pressure homogenisation both produce nanoemulsions, and they do not produce the same distributions.
  • Whether the number is per batch. A particle size report from a good batch last year is not a specification.

We make water-soluble emulsion to spec in Hillsboro, and put a particle size report and a COA with every lot. If you are reformulating and want to talk about what is achievable in your matrix, tell us what you are making.