CBDA tincture — 90-day room-temp stability check
Pulled three samples from the September batch and compared COAs at day 0, 45, and 90. CBDA-to-CBD conversion sits at 8% over 90 days at 22°C / amber glass / no light. That's the number we needed before talking about a CBDA-specific product line.
Why we ran the test
Most clinical research on cannabidiolic acid uses fresh extract — produced and tested in the same week. The pharmacy literature on CBDA stability is thin. If we're going to put a CBDA-specific tincture on a shelf for 6-12 months, we need to know: how much of the CBDA decarboxylates to CBD over time?
We've been running CBD tinctures for years. Stable shelf life. But CBDA is more thermally labile, and it's the form the plant actually makes — so for a "raw hemp" or CBDA-forward product, conversion rate at room temperature is the gating data.
We pulled three samples from the September 2025 supercritical CO2 run and stored them at 22°C, in 30 mL amber glass, dark cabinet, 35-45% RH (Phoenix interior conditions). Sampled at day 0, day 45, day 90. Same accredited lab, same Megazyme-equivalent assay, blind labels.
The numbers
Day 0 (2026-01-22): CBDA 47.2 mg/mL · CBD 4.1 mg/mL · ratio 92:8.
Day 45 (2026-03-09): CBDA 45.1 mg/mL · CBD 6.0 mg/mL · ratio 88:12.
Day 90 (2026-04-22): CBDA 43.4 mg/mL · CBD 7.6 mg/mL · ratio 85:15.
Total CBDA loss over 90 days: 8.0%. Total CBD gain: 85.4% (from 4.1 to 7.6 mg/mL). The math closes: the CBDA loss accounts for the CBD gain almost exactly, suggesting clean decarboxylation rather than degradation to inactive products. No detectable CBN, no detectable degradation peaks on the chromatogram.
What this means for product design
8% conversion over 90 days at room temperature is workable for a 6-month shelf-life product if we accept that the COA-stated CBDA value is the day-of-bottling number, and we communicate the conversion to consumers.
Two paths forward:
- Date-stamped COA per batch. Every bottle ships with the day-of-bottling COA + an estimated conversion curve. Consumers know what they're getting, when. The CBDA-favoring use case still works because most users finish a 30 mL bottle in 30-60 days.
- Refrigerated supply chain. Cold storage essentially halts decarboxylation (literature and our own pilot data confirm). More expensive, more logistical complexity, but extends practical shelf life to 12+ months. Probably overkill for a category that hasn't proven consumer demand yet.
Going with path 1 for the soft launch. Adding the conversion-curve language to the product page draft. If the line shows traction, we revisit cold-chain.
Open questions
What does the curve look like at month 6? Month 12? Linear extrapolation says ~16% loss at 6 months, but decarb is generally first-order — so it should slow, not accelerate. We're holding two more samples for the 180-day and 365-day points.
What does sublingual absorption look like for a partially-converted product? The pharmacology is different for CBD vs CBDA (different oral bioavailability, different receptor affinities). A user opening a bottle 4 months out is getting a measurably different ratio than day 0. Whether that matters functionally — open question.
The CBDA-specific product line is provisionally green-lit for late-2026 launch contingent on the 6-month data point coming in clean. Will write that up when we have it.
Goes deeper in
Long-form articles on the science
More from lab notes
Apr 10
Reishi dual extract — three method comparison on the same starting material
Same lot of Ganoderma lucidum fruit body, three extraction methods side-by-side. Beta-glucan and triterpene yields differed enough to change which method we recommend for which product type. Ratios and method specs below.
Mar 28
Q1 2026 heavy-metal screen — six suppliers, two failures
Routine quarterly screen across our six current mushroom-extract suppliers. Two failed the cadmium spec. Both are out of the supply chain pending re-test. Why this matters even when the numbers look small.
Read next
Most recent dispatches across both streams.
April blue oyster bag run — 14% contam, three lessons
Ran 21 supplemented hardwood bags across two flushes. 3 contaminated, 18 fruited. Below the typical 8-12% target but instructive — and the contamination pattern points at one specific moment in the workflow.
Lion's mane fruiting chamber — rebuilt around the CO2 problem
Lion's mane has a real fruiting-stage CO2 sensitivity that doesn't show up in oyster or king trumpet runs. Tore out the old chamber, rebuilt with a switched intake fan + CO2 monitor + scheduled exchange. First post-rebuild flush coming in next week.
Reishi dual extract — three method comparison on the same starting material
Same lot of Ganoderma lucidum fruit body, three extraction methods side-by-side. Beta-glucan and triterpene yields differed enough to change which method we recommend for which product type. Ratios and method specs below.
Shiitake 3782 — getting cold-shock timing right
The 3782 strain wants cold water, but how cold and for how long is the kind of question every guide handwaves. Logged five flushes on a single block under different shock protocols. Sweet spot is narrower than I expected.