Converted Cannabinoid Products: What the Surveys Found

What this page is and is not human data

Δ8-THC, Δ10-THC, tetrahydrocannabinol acetates, hexahydrocannabinol and a growing list of related compounds reach the market as products made by chemically transforming cannabidiol rather than by extracting the target cannabinoid from a plant that contains it in quantity. At the level of structure the transformation is a ring closure under acid catalysis, a reaction documented in the chemical literature from Adams and colleagues in 1940 onward, and it is chemically facile. That structural statement is the entirety of what this library says about how it is done: no reagents, no catalysts, no solvents, no concentrations, no temperatures, no times, no work-ups and no yields appear here or anywhere else in this section. What this page is about is the other half of the question, the half that is directly a consumer and buyer safety matter and is well covered in the published analytical literature: what the finished products have actually been found to contain when independent investigators assayed them.

Sources: Adams R 1940 · Meehan-Atrash J 2022 · Lin K 2026

What the product surveys report contested human data

The findings across the published surveys are consistent enough to summarise as a pattern. Residual reaction solvents and residual acid catalyst have been detected in retail products, which is the expected consequence of a transformation and purification step performed outside a pharmaceutical quality framework. Unreacted starting material is present, so cannabidiol frequently appears in a product sold as Δ8-THC, as does Δ9-THC — which matters legally as well as pharmacologically, because a product marketed on its non-Δ9 identity may contain quantities of Δ9-THC sufficient to produce both effects and positive drug tests. The most consistent and least discussed finding is the presence of multiple additional cannabinoid-like species: isomers, regioisomers and side-products that appear in chromatograms as peaks the laboratory cannot name, sometimes accounting for a substantial share of total cannabinoid chromatographic area in a given sample. Olivetol- and resorcinol-related impurities, traceable to the starting materials and side-reactions, have been reported. Metals attributable to catalysis and to processing equipment have been reported alongside the metals contributed by vaping hardware. And running through all of it is a label-accuracy problem: measured content diverging from label claim in both directions, which is the same finding as the broader cannabinoid-product literature has repeatedly produced for CBD extracts and edibles. Poison-centre data for Δ8-THC, Δ10-THC and THC-O-acetate exposures document that these products are producing clinical presentations at scale, including in children, which is the downstream consequence of an unlabelled and unverified dose reaching a general retail channel.

Contested — caveat. These are findings from a modest number of product surveys on samples purchased at particular times in particular markets, and the proportion of products affected and the magnitude of each finding vary considerably between studies. They establish that these contaminants and unassigned species are present in commercial products and that label accuracy is unreliable; they do not establish a prevalence figure for the market as a whole, and a well-run processor with real analytical control is not described by them.

Sources: Meehan-Atrash J 2022 · Lin K 2026 · Helander A 2022 · Burgess A 2024 · Meehan-Atrash J 2021 · Meehan-Atrash J 2017

An unassigned peak is an unidentified compound being consumed human data

This deserves to be pulled out, because it is the finding with the largest gap between its importance and the attention it gets. When a chromatogram of a cannabinoid product shows peaks that the report does not name, the honest reading is that the product contains compounds nobody has identified, in quantities that are sometimes not small, and that are being inhaled or ingested. The usual reason they stay unassigned is not laziness or negligence on the laboratory's part. Identifying a peak requires an authentic reference standard of the suspected compound to match retention and spectral behaviour against, and for novel isomers and side-products of a conversion no such standard is commercially available; a laboratory can often say a peak is cannabinoid-like from its mass spectrum and ultraviolet absorbance while being genuinely unable to name it. That is a structural gap in the analytical supply chain, the same gap that made the synthetic-cannabinoid era unmanageable, appearing again in a legal market. It follows that the correct thing for a buyer or a processor to ask for is not a certificate with no unassigned peaks — which is often unobtainable and is easy to fake by simply not reporting them — but a certificate that reports them honestly: total cannabinoid mass balance, the percentage of chromatographic area that is unassigned, and a statement of which compounds were sought against standards. A report that accounts for 100 percent of a sample with four named cannabinoids and no discussion of the remainder is either a very clean product or an incomplete report, and the report itself should tell you which.

• A named peak is a measurement. An unassigned peak is an unidentified compound present in the product.
• Peaks usually stay unassigned because no reference standard exists, not because the laboratory failed.
• Ask for mass balance and the unassigned fraction, not for a certificate with nothing unexplained on it.
• A report that names four cannabinoids and is silent about everything else has not told you the composition.

Sources: Meehan-Atrash J 2022 · Castaneto MS 2015 · International Organization for Standardization / International Electrotechnical Commission 2017 · Lin K 2026

What a processor or a buyer can actually ask for human data

The practical content of this page is a request list, and every item on it is something a competent processor either has or can obtain. A batch-matched certificate of analysis from a laboratory accredited to ISO/IEC 17025 with the relevant analytes inside its declared scope, rather than a certificate from an unspecified laboratory or one accredited for a different matrix. A full panel rather than potency alone: cannabinoid profile, residual solvents, heavy metals, pesticides, mycotoxins, and microbiological where the matrix warrants it. For a converted product specifically, an elemental scan broad enough to cover catalyst metals rather than the four-metal agricultural list, and residual-solvent analytes that include the solvents actually used in the conversion and purification rather than only the ones used in extraction. Explicit reporting of the unassigned chromatographic fraction and a total-cannabinoid mass balance. A named compound identity, using the actual chemical name and, where they exist, a CAS number and the specified isomer and stereochemistry — a label reading Δ8 without specifying which compounds are present in what proportion is a marketing term, not an identity. Per-serving and per-container milligram figures, computed and stated rather than left to the buyer. And for any unit-dose product, evidence of dose uniformity rather than a single composite assay: increments sampled from different points in the batch, assayed individually, with the spread reported. A supplier who can provide that list is running a real quality operation; a supplier who treats the request as unreasonable has answered the question.

• Batch-matched COA, ISO/IEC 17025 laboratory, relevant analytes inside the declared scope.
• Full panel, not potency-only, on any converted or novel product.
• Residual-solvent analytes that match the solvents actually used, including in conversion and purification.
• Elemental scan broad enough for catalyst metals, not only lead, arsenic, cadmium and mercury.
• Unassigned-peak reporting and total-cannabinoid mass balance.
• Actual chemical identity — compound, isomer, stereochemistry, CAS where it exists.
• Per-serving and per-container milligram figures, and dose-uniformity evidence for unit-dose products.

Sources: International Organization for Standardization / International Electrotechnical Commission 2017 · United States Pharmacopeia 2023 · United States Pharmacopeia 2023 · Meehan-Atrash J 2022 · Vandrey R 2015

The label-accuracy baseline, for context human data

The label problem is not specific to converted cannabinoids; it is the background condition of the whole cannabinoid product market, and knowing the baseline prevents both complacency and overstatement. Independent assay of edible medical cannabis products found that a minority were accurately labelled, with both substantial under-labelling and substantial over-labelling present. Independent assay of cannabidiol extracts sold online found that most were inaccurately labelled, again in both directions, with some products containing detectable Δ9-THC that the label did not mention. Later work on unregulated cannabidiol products reproduced the finding that measured concentration and label claim diverge, and reviewers have documented regional variability in labelled unit strength for edibles on top of the measurement disagreement. Converted and novel cannabinoid products therefore inherit an already-poor label-accuracy baseline and add to it the specific difficulties of a chemically transformed product: additional species to quantify, some with no available standards, and a smaller pool of laboratories with validated methods for them. The implication for a buyer is that a batch-matched certificate is not a formality but the only actual evidence of content, and the implication for the industry is on the market-consequences page.

Sources: Vandrey R 2015 · Bonn-Miller MO 2017 · Johnson E 2022 · Johnson-Arbor K 2023 · Kruger JS 2022 · US Food 2022*