Decarboxylation in the Oven: What It Is and What Temperature to Use

Decarboxylation is a reaction in which acidic cannabinoids lose carbon dioxide under heat and convert into neutral forms, for example CBDA into CBD. What temperature in the oven? The popular “110 degrees for an hour” comes from measurements of extract in a vacuum dryer, not herb. In the closest measurement on ground herb we found, half of CBDA disappeared in a closed laboratory oven after 44.4 minutes at 105°C and after 32.6 minutes at 120°C (Ryu et al., Molecules, 2021). We describe here the chemistry of the reaction, measurements for extract and herb, and home methods from oven to vaporizer. We also show how much material is lost and where the law ends. Parameters apply exclusively to industrial hemp herb, and where measurements are missing, we state it clearly.

KEY INFORMATION
• In ground herb, half of CBDA disappeared after 44.4 min at 105°C (Ryu et al., 2021).
• The recommendation “110°C for an hour” originates from extract measurements, where 8-53% of the sum of acid and neutral form was lost.
• Electric vaporizers at 210°C decarboxylated at least 94.6% of acids.
• Parameters apply only to industrial hemp: total THC and THCA up to 0.3%.

What is decarboxylation and how does it work?

Decarboxylation is the removal of a carboxyl group -COOH from a molecule. The group leaves as carbon dioxide, and a hydrogen atom remains in its place (Urvashi et al., Journal of Cannabis Research, 2024). In cannabis, THCA acid converts to THC, CBDA acid to CBD, and CBGA acid to CBG. The resulting THC weighs 87.7% of the acid’s mass (molar masses from ChEBI database). Heat drives the reaction, and time also slowly contributes.

The plant mainly produces acidic forms, and neutral forms do not occur in significant concentrations. Acids are unstable and degrade under light or heat, for example when smoking or baking (Wang et al., Cannabis and Cannabinoid Research, 2016). This also happens during processing and storage (Nadal et al., British Journal of Pharmacology, 2017). Filer adds that preventing this reaction during harvest and storage is also important (Filer, Cannabis and Cannabinoid Research, 2022).

How slowly does it proceed without heating? Kim and colleagues sampled secretory cavity contents of glandular hairs of Cherry Wine cannabis, with total THC content 0.25%. In the cavity, the CBD to CBDA ratio was 1:99, and in the flower dried for 15 days in darkness at room temperature it was 1:20 (Kim et al., Scientific Reports, 2024). Drying alone, without light and heat, thus converted a measurable part of the acid. This is a result for one greenhouse-grown variety.

The mass balance can be checked without a lab. According to ChEBI, THCA molecule weighs 358.48 g/mol, and THC 314.47 g/mol. The difference, 44.01 g/mol, is the mass of carbon dioxide. For the CBDA and CBD pair, numbers are the same because they have the same molecular formulas. Laboratories have known this reaction for a long time: THC was measured after decarboxylation at 100°C for 15 minutes already in 1967 (cited by Urvashi et al.).

What is decarboxylation outside cannabis?

In chemistry, decarboxylation is any reaction where a molecule loses carbon dioxide, but not every acid does this easily. Organic chemistry textbooks state that by heating alone, only acids with a second carbonyl group two atoms away, i.e. substituted malonic acids and beta-keto acids, lose CO2 (McMurry, Organic Chemistry, OpenStax, 2023). Beta-keto acid then loses CO2 and forms a ketone.

Closest examples outside cannabis are in the body and kitchen. In cells, the pyruvate dehydrogenase complex removes the carboxyl group from pyruvate as CO2 forming acetyl-CoA; this is oxidative decarboxylation (Clark et al., Biology 2e, OpenStax, 2018). In poorly stored fish, e.g. tuna or mackerel, bacterial enzymes remove this group from histidine. The resulting histamine causes scombroid poisoning and does not disappear with cooking because biogenic amines are heat resistant (Saha Turna et al., Heliyon, 2024).

Cannabinoid acids are an unusual case here. They lose the carboxyl group attached to an aromatic ring mainly due to heat alone. Most other decarboxylations are catalyzed (Filer, 2022). The mechanism is described like beta-keto acid, as for salicylic acid, with formic acid from flowers considered a catalyst (cited by Urvashi et al., 2024).

Do acidic forms act differently than neutral ones?

Yes. In binding tests on cells, THCA-A bound weakly to CB1 and CB2 receptors, while THC bound 62 and 125 times stronger respectively (McPartland et al., Cannabis and Cannabinoid Research, 2017). For CB2, acid had no measurable efficacy. Decarboxylation thus changes the compound’s activity profile, not just its name.

This does not mean raw acid is free of THC. In a certified THCA-A standard, authors found 2% THC and state such contamination is almost unavoidable. In the body, THCA-A practically does not decarboxylate: no conversion to THC was detected in rats, and it is detected in human blood and urine (Filer, 2022).

Acidic forms also have their own targets. Cannabinoid acids bound and activated the nuclear PPAR gamma receptor more strongly than their neutral forms, and THCA was neuroprotective in mice (Nadal et al., 2017). Among eleven cannabinoids tested, all acids inhibited DAGL-alpha enzyme, which produces one endocannabinoid (De Petrocellis et al., 2011). These are results from cells and animals, not human studies.

Decarboxylation does not improve the raw material, it just converts one profile into another. Filer describes it as a transition to neutral forms with distinctly different pharmacology, and a process stopped halfway yields a mixture of both. Differences between acids and neutral forms are summarized in the table of acidic cannabinoid forms.

What temperatures and times were measured for extract?

The most cited measurement concerns extract, not herb. Wang et al. heated flower extract in a vacuum dryer at five temperatures from 80 to 145°C. THCA-A approached zero after 30 minutes at 110°C, 9 minutes at 130°C, and 6 minutes at 145°C. Below 100°C, the reaction did not finish within an hour (Wang et al., 2016).

The material was dried extract from the NIDA research program, not industrial hemp herb. Nine cannabinoids were quantified by PDA detector, and mass spectrometry was used for identification. Vacuum drying was chosen to exclude oxygen and light. The reaction followed first-order kinetics from 80°C.

Numbers come from tables in this work, including supplementary table S1. “Zero” after 30 minutes at 110°C is 4.3% THCA-A in the supplement, and the series ends at 50 minutes, not 60. THCA-A is given as a reference point for the whole acid family, not as an instruction. Among tested temperatures, 110°C was the lowest at which THCA-A reacted within an hour.

How long does decarboxylation take for herb, not extract?

Longer than guides state. In a closed, dry laboratory oven, half of CBDA in ground, lyophilized flower disappeared after 44.4 minutes at 105°C, 32.6 minutes at 120°C, and 25.8 minutes at 135°C. At 90°C, half the acid did not react within an hour (Ryu et al., 2021).

Authors describe the material as industrial hemp flower from a Korean genetic resource bank. THCA reacted faster: half disappeared after 39.1 minutes at 105°C. They note that in a closed oven, speed may depend on volume.

We compiled half-conversion times of CBDA from both studies. For extract in vacuum dryer at 110°C, Wang’s rate constant gives about 14 minutes, and for pure CBDA standard at the same temperature about 72 minutes. In ground herb even at higher 120°C, 32.6 minutes was measured. The same acid thus reacts several times faster or slower depending on surroundings and device.

This explains errors in many guides. Wang’s constant extrapolates to about 5% CBDA remaining after an hour at 110°C in extract, but 10.1% was measured after 50 minutes. For herb from Ryu’s work, assuming first-order reaction, about 39% CBDA would remain after an hour at 105°C and 28% at 120°C. This is our calculation, as the paper only gives half-conversion times. Does herb always react slower than extract? Not necessarily, as at 135°C times for extract and flower did not differ significantly. One system’s times cannot be transferred to another.

Why does CBDA react slower than THCA?

No explanation in reviewed literature, though difference was measured many times. At 110°C, rate constant was 1.83 thousandths per second for THCA-A and 0.83 for CBDA. Authors say THCA-A’s constant was about twice that of CBDA and CBGA (Wang et al., 2016).

From temperature dependence of constants, activation energy was calculated: 88 kJ/mol for THCA-A and 112 kJ/mol for CBDA. For CBGA it was 109 kJ/mol. Higher barrier means slower reaction at same temperature, and difference grows in cooler conditions. At 80°C, THCA-A’s constant was 3.6 times higher than CBDA’s.

In ground flower, trend was the same. Ryu et al. found half of THCA disappeared on average 5.1 minutes earlier than half of CBDA at same temperature. Filer calls this difference interesting, described independently by several authors, but no explanation proposed (Filer, 2022).

Practical conclusion: CBDA needs more time at same temperature. Raising temperature shortens process but costs material, as shown below. CBGA, precursor of THCA and CBDA in plant (Walsh et al., 2021), reacts similarly to CBDA (Wang et al., 2016).

What temperature to set in the oven?

Studies do not give a single setting for home ovens. Closest measurement on ground herb covered 90-135°C: at 90°C half of CBDA did not react within an hour, and at 105-135°C half disappeared after 25.8-44.4 minutes (Ryu et al., 2021). The 110°C figure comes from extract measurement where THCA-A disappeared after 30 minutes.

No one measured a home oven. A PubMed query combining cannabis, decarboxylation, and oven on September 19, 2026, yielded two results, none about home ovens. Among ten papers on home oven temperatures, none measured thermostat fluctuations. The dial shows what you set, not the herb’s temperature.

Why not increase temperature? Because faster means costlier. In extract, raising from 110 to 130°C increased CBD sum loss from 18 to 25% (Wang et al., 2016). THC evaporated 4.3 times faster at 120°C than at 100°C, and beta-myrcene 2.1 times faster (Eyal et al., Cannabis and Cannabinoid Research, 2023).

If you stick to the oven, the closest data range is 105-135°C, with time counted in tens of minutes. Lower range means slower reaction, higher means greater losses. An hour in this range does not guarantee full CBDA conversion, and full conversion cannot be checked at home without chromatography.

How to perform decarboxylation in the oven step by step?

To closely replicate measurement conditions. In the herb study, material was ground and heated in a preheated closed oven at 105-135°C, with samples tested every 5 minutes for an hour (Ryu et al., 2021). The steps below follow these conditions. Where measurements are missing, we note it at the step.

Safety: parameters apply exclusively to industrial hemp herb. Label decarboxylated herb and keep out of reach of children and animals. We did not check glass and dish safety in ovens in literature, so follow manufacturer instructions.

1. Check batch test results. Parameters apply to industrial hemp herb with total delta-9-THC and THCA not exceeding 0.3% dry mass.
2. Grind the herb. Flower was ground and sieved in measurement; no one checked how finely to grind at home.
3. Spread thinly on a baking sheet lined with parchment paper. In Moreno et al.’s model, reaction slowed as material mass increased (cited by Urvashi et al.).
4. Preheat oven and check temperature with a separate thermometer at sheet height. Stay within 105-135°C range measured for herb.
5. Cover the sheet or use a dish with lid. In the same model, oxygen accelerated reaction, but in air THC forms CBN, and open dishes lose more vapors.
6. After heating, transfer herb to an airtight container and label with date and process conditions.

How to tell if it worked? Not by weight. Loss of carbon dioxide is 12.3% of reacted acid mass, so for herb with about 8% CBD as in Boffa et al.’s study, this is about one percent of mass. Sample moisture in proficiency testing ranged from 5.8 to 11.4% (Sikora et al., 2025). Weight loss mainly shows evaporated water and volatiles. Degree of conversion is confirmed only by chromatographic analysis.

How many cannabinoids are lost in the reaction itself?

In extract, from about 8 to 53% of the sum of a given cannabinoid, depending on acid. This much the sum of acid and neutral form decreased after reaction in vacuum dryer (Wang et al., 2016). Carbon dioxide loss does not explain this, as it does not change cannabinoid molecule count. Some material simply disappears.

For THCA-A, authors describe conversion as clean, though sum dropped 7.9%. For CBDA and CBGA, some acid disappeared without corresponding neutral form increase. Causes unknown: authors allow evaporation under vacuum or unidentified products. Loss is visible because supercritical chromatography measures both forms in one assay without prior decarboxylation.

Is it worse in a non-vacuum oven? Unknown, as no comparison found. In the same closed oven after 30 minutes at 135°C, neutral forms decomposed less in flower than in extract (Ryu et al., 2021). Losses increase with time. Older measurements of dried THC extracts showed maximum THC after 5-10 minutes at 145°C, then decrease, probably due to evaporation (Veress et al., cited by Wang et al.).

Does high temperature convert THC to CBN?

Not temperature alone. In vacuum dryer without oxygen and light, at 110°C no significant THC oxidation to CBN occurred: concentration stayed around 0.02 mM for 50 minutes (Wang et al., 2016). Same authors write that in presence of oxygen and light, THC easily oxidizes to CBN during decarboxylation.

In reviewed works, popular threshold “above 145°C THC converts to CBN” has no source. Wang’s CBN graph concerns only 110°C, and supplementary table does not list CBN. It is also absent in full texts of Filer, Lanz, Lewis-Bakker, and Eyal.

CBN appeared at higher temperature and longer heating. Pure THCA-A heated 15 minutes fully converted to THC at 160°C, but CBN formed at 160 and 180°C (Dussy et al., cited by Wang et al.). In microwave lab extraction in ethanol at 150°C, CBN was present after 30 minutes in one of three varieties with larger raw material portion, but not after 20 minutes (Lewis-Bakker et al., Cannabis and Cannabinoid Research, 2019).

CBN formation mainly depends on oxygen and time. Air oxidation in darkness produced CBN, and THC loss under light did not increase CBN amount (Fairbairn et al., Journal of Pharmacy and Pharmacology, 1976). CBN is not a stronger THC version but a different compound. In cell tests, it was a strong agonist of rat TRPA1 channel (De Petrocellis et al., 2011).

Which home methods have actually been measured?

Except vaporizer, none. Among kitchen devices and store equipment, only vaporizers were measured. In lab measurement, electric devices at 210°C converted at least 94.6% of acids (Lanz et al., PLoS ONE, 2016). For other methods, no herb measurement found in PubMed or Europe PMC.

Below we summarize what is known about each method. “No one measured” means we found no such study in these databases, not that it definitely does not exist.

Do sous-vide or jar give better results than baking sheet?

Unknown, as no one compared them on herb. Water under normal pressure does not exceed about 100°C, and in extract at 95°C after 50 minutes 46% CBDA remained (Wang et al., 2016). Water bath is thus slow, and slower for CBDA than THCA-A.

We calculated what this means for longer times. From CBDA rate constant at 95°C, half disappears after about 43 minutes. Popular “95°C for 90 minutes” would yield about three quarters CBDA conversion in extract. For 95% conversion, about three hours would be needed. This is extrapolation from extract to a system no one measured.

PubMed query combining cannabis, sous-vide or water bath, and decarboxylation yielded zero results. Only a short flower bath was measured: 5 minutes at 98-100°C gave small decarboxylation and about half loss of main monoterpenes (Romano and Hazekamp, cited by Eyal et al., 2023). Vacuum bag has one advantage based on data: it cuts oxygen, and oxygen in darkness produces CBN from THC (Fairbairn et al., 1976).

With jars it is similar. No herb measurement in jar found, so numbers like “20-25% more monoterpenes” lack known study support. Indirect data concern reactors: in open reactor with hemp seed oil at 120°C, CBD sum dropped 60% (Citti et al., cited by Urvashi et al., 2024). Lewis-Bakker et al. write that in open reactor there is no temperature at which reaction proceeds efficiently without volatile loss. Closing limits vapor escape and air access, but no scale of this gain for herb was given.

Can microwave, air fryer, or decarboxylator replace oven?

Based on studies, this cannot be said. Full decarboxylation was shown only in a lab microwave reactor: ground flower in ethanol, sealed vial, 150°C for 20 minutes (Lewis-Bakker et al., 2019). Such reactor measures and holds sample temperature, while in home microwave you set only power and time.

Microwave oven heats unevenly. Standing wave in cavity creates hot and cold spots depending on shape, position, and material properties, with hotter spots absorbing more energy (Yang et al., Foods, 2021). This review concerns food, not herb. In PubMed, none of 7 papers on cannabis decarboxylation and microwaves concerned home ovens, and in Europe PMC none of 15.

Boffa et al.’s 2024 work cited on this topic measured terpene capture in microwave reactor, not decarboxylation degree. It does not prove microwaves decarboxylate herb well.

For air fryer, Europe PMC returns no cannabis decarboxylation papers, and three results combining it with cannabis concern other topics. For decarboxylators, PubMed gave 7 results and Europe PMC 3. None measured home device. We do not repeat manufacturer claims as they are not measurements.

Does vaporizer perform decarboxylation?

Yes, and it does it by itself. In vitro measurement with five vaporizers given 50 mg herb each. Electric devices set to 210°C converted at least 97.3% THCA-A and 94.6% CBDA, and gas flame device 87.7-93.3% (Lanz et al., 2016). Separate decarboxylation before vaporization is thus unnecessary.

Reaction takes seconds. In Eyal et al.’s 2023 trial, decarboxylation in vaporizer neared completion after 40 seconds at 180°C and was practically full after 20 seconds at 220°C (Eyal et al., 2023). THC evaporated slower: at 180°C 10% after 20 seconds and 20% after 40 seconds. Authors state recommended vaporizer temperature is usually 160-220°C. Temperature selection is discussed in the vaporization temperature guide. Devices with temperature control are in the vaporizers category.

In the body, picture is less clear. In 14 healthy subjects after medical cannabis vaporization, THCA and CBDA were also detected in serum, explained as partial conversion by vaporization (Busardò et al., Pharmaceuticals, 2021). In two smoking simulations, THCA-A converted almost entirely to THC, but only about 30% of substance was recovered as THC (cited by Filer, 2022; Dussy et al., 2005).

What remains in herb after vaporization?

Some cannabinoids. Residues after one session in electric devices set to 210°C contained 1.1 to 15.0% cannabinoids, and in gas device without control 14.4-16.2% (Lanz et al., 2016). From residue assays, authors calculated decarboxylation efficiency at least 94.6% in electric devices, so residue is mainly neutral form.

Devices differed clearly: one left up to 15.0%, another 1.1-2.1%. Lanz et al. tested THC-dominant and CBD-dominant herb, 50 mg each. Residue amount after home session depends on device, temperature, and time, which you cannot measure in kitchen.

Does such herb need decarboxylation again before further processing? No one studied this, and a volunteer study detecting acid forms in blood after vaporization advises caution in generalizing. In PubMed, only residue measurements found are Lanz et al.’s work and a study of microorganisms after heating in vaporizer. Ideas for using such material are described in the post on ABV herb.

How many terpenes escape during heating?

A lot, especially monoterpenes. In lab decarboxylation of flower at 120°C for 30 minutes, vapors were directed into olive oil capturing about 58% of terpenes present in raw material, mainly monoterpenes (Boffa et al., Molecules, 2024). Same authors write about nearly 50% in summary.

Raw material had 8.1% CBD, 0.18% THC, and 91,325 mg terpenes per kg dry mass, including 44.3% beta-myrcene. It was heated in microwave reactor under slight vacuum. Only flower extraction with collected oil at 60 and 100°C raised recovery close to 100% and added sesquiterpenes. This is percent captured from vapors by special installation, not retained in herb. In oven, these compounds escape into kitchen.

Evaporation starts already during drying and intensifies during decarboxylation (Eyal et al., 2023). Older measurement: 30 minutes of flower at 145°C in oven removed almost all sesquiterpenes (Romano and Hazekamp, cited by Eyal). Terpenoids, not cannabinoids, are responsible for cannabis aroma. Beta-caryophyllene often dominates hot decarboxylated extracts and was found in 2008 study to be a selective full CB2 receptor agonist at 100 nM (Russo, British Journal of Pharmacology, 2011).

Terpene-cannabinoid synergy remains a hypothesis Russo describes as needing proof. More about these compounds is in the post on what are cannabis terpenes.

Where does the terpene boiling point table come from?

From McPartland and Russo’s 2001 work, not Russo’s 2011 review, which it is usually attributed to. Figures in that work list THC 157°C, CBD 160-180°C, and beta-caryophyllene 119°C, with note that values were collected from various sources (McPartland and Russo, Journal of Cannabis Therapeutics, 2001).

The 2011 review has no such table. Its two tables concern pharmacology, and text does not mention boiling or Celsius degrees. The source is known but misattributed.

We compared 2001 values with normal boiling points given by Eyal et al.

For four monoterpenes, both columns agree within one or two degrees. Three entries differ: beta-caryophyllene by over 140 degrees, and THC and CBD by over 250. THC and CBD decompose before boiling, so authors estimate boiling from vapor pressure, and under 0.05 torr vacuum THC boils around 155°C (Eyal et al., 2023). The table adds little to decarboxylation, as loss depends on vapor pressure at given temperature. At 130°C beta-pinene evaporated about 900 times faster than alpha-bisabolol. More on this is in the post on vapor pressure, not boiling point, determines terpene loss from herb.

Why does label content not increase after heating?

Because heating does not create new molecules, only converts existing ones. Laboratories convert acid form to neutral using factor 0.877, the same value used by EU law for THC limits in hemp food (Commission Regulation (EU) 2023/915, Annex I).

The “total” content on certificate is neutral form plus acid form multiplied by 0.877, so after full decarboxylation it does not change, and in practice decreases due to losses described above. How to read such numbers is explained in the post on herb label and total THC.

Polish threshold is different, as law sums delta-9-THC and THCA without factor (art. 4 point 5 of the anti-narcotics act). Mass balance shows full decarboxylation reduces this sum’s mass by 12.3% of reacted acid, not increases it. Threshold is a dry mass percentage, and terpene loss also occurs during heating, so calculation alone does not say what percentage remains after process. Heating above threshold does not remove liability under art. 53. Whether heating changes product status is discussed in the law section.

Why can’t home product potency be predicted?

Because no decisive factor is measured at home. In lab heating of flowers of four varieties at 60 and 120°C, decarboxylation rate constants differed between varieties, and new compounds like hydroxyquinones formed in heated samples (Benes et al., Food Research International, 2024).

Similar results in baking. German BfR institute study showed bread baking temperature and time changed delta-9-THC content from -7 to +53%, but raw material variability outweighed process conditions (Lindekamp et al., Food Chemistry, 2025). Even one plant is heterogeneous: in medical variety, top flowers had 12% THC, lower 6% (cited by Filer).

Put together: declaration concerns batch, not your portion. Herb temperature in oven is not measured, and loss in reaction was 8-53% in extract. No study found measuring cannabinoid transfer to home butter in PubMed or Europe PMC. Therefore, we do not calculate portion potency. Components and weak points of such calculation are shown in the post on how to calculate CBD dose in home edibles and butter.

What to do with decarboxylated herb?

If for food, fat is a sensible carrier, as CBD is strongly lipophilic. In 12 healthy volunteers, a high-fat meal raised maximum plasma CBD concentration 4.85 times and area under curve 4.2 times compared to fasting (Taylor et al., CNS Drugs, 2018). This describes CBD in oral solution, not a recommendation.

A review of 24 CBD pharmacokinetic studies found absolute bioavailability measured in humans only for smoking: 31% in one study (Millar et al., Frontiers in Pharmacology, 2018). For oral route, none in that review determined it. EFSA summarized in 2026 that CBD bioavailability is variable and depends on carrier and food (EFSA, 2026). Temperatures and times for melting are described separately in the recipe for cannabis butter.

Can herb be decarboxylated directly in oil? Reaction proceeds slowly: in hemp seed oil at 85°C, half of CBDA disappeared after about 4 days, and with vitamin E added after 17 days (cited by Filer, 2022). At this rate, several hours of melting at this temperature does not replace prior decarboxylation.

Ready oils do not always contain neutral forms, as raw oils with acid form are also sold. Difference is described in the post on raw and decarboxylated oils. If you do not want to process herb, simpler way is ready oil from the oils category, where batch test results should be checked.

How to store decarboxylated herb?

In darkness, cool, and without air access. Classic study showed light was the biggest single factor in cannabinoid loss, and temperature up to 20°C had no effect. Herb kept in darkness at room temperature was relatively stable for 1-2 years (Fairbairn et al., 1976).

Longer observations show role of cool. Over four years at 22°C, in light and darkness, almost all THC in herb and resin samples decomposed with CBN increase, while in fridge and freezer such changes did not occur (Zamengo et al., 2019). THCA-A in herb in glass bottles at room temperature dropped to 80% after 25 months, and at 4°C 94.7% remained (cited by McPartland et al., 2017).

Terpenes behave differently. In a one-year study, terpenoids decreased rapidly in all conditions, most below -20°C and after grinding flowers, and best storage was whole flowers at 4°C (Milay et al., Frontiers in Plant Science, 2020). Cannabinoids in resin and extracts were stable at -20°C for 4 years, according to the cited study.

No one measured how many months home decarboxylated industrial hemp herb lasts. Therefore, we do not give a term. Label container with date to know how long it has stood.

What to watch for when eating cannabinoids?

Primarily delay. With THC edibles, first effect appears after 30-90 minutes, peak after 2-4 hours. Delay and lack of repeatability may lead to eating more than intended (Barrus et al., RTI Press, 2016). For CBD in oral solution, peak concentration was about 4-5 hours (Taylor et al., 2018).

Second risk is children and animals. In USA 2017-2021, 7043 cases of cannabis edible ingestion by children under 6 were reported, 97.7% at home, and 22.7% hospitalized (Tweet et al., Pediatrics, 2023). In a survey of 251 vets, cannabis edibles were most often suspected in animal poisonings, usually eaten unsupervised. Products included cannabis butter and oil (Amissah et al., PLoS ONE, 2022).

Third risk is drugs. EFSA in 2026 concluded CBD safety cannot be established for under 25s, pregnant or breastfeeding women, and people on medications. Human studies showed potential liver damage, especially when combining CBD with drugs (EFSA, 2026). WHO reported adverse effects may result from CBD-drug interactions, though CBD showed no abuse or dependence potential in humans (WHO, CBD critical review, 2018). If you take drugs regularly, are pregnant or breastfeeding, consult a doctor.

Is decarboxylation of hemp herb legal in Poland?

The law does not answer this in one sentence. Industrial hemp, where total delta-9-THC and THCA in flower or fruiting tops does not exceed 0.3% dry mass, is not cannabis herb other than industrial (art. 4 point 5 and 37 of anti-narcotics act, consolidated text Dz.U. 2023 pos. 1939).

Threshold in this wording applies since May 7, 2022 (Dz.U. 2022 pos. 763). It matches EU threshold for crop subsidies from Regulation (EU) 2021/2115 but does not derive from it. Herb above threshold is a narcotic. Giving it a new form is processing, and conversion to another narcotic is transformation (art. 4 points 20 and 19). Both acts against law carry penalty up to 3 years (art. 53 sec. 1).

For industrial hemp, question remains open. Law calls cannabis resin also “other cannabis products containing” THC or THCA without threshold (art. 4 point 38), and narcotics list mentions resin without industrial hemp (consolidated text Dz.U. 2024 pos. 1139). Whether heated herb or product made from it fits this definition is not resolved by law text, so we do not say it is legal. Prescription does not change this: processing outside pharmacy requires permit (art. 35 sec. 1 and 3), and no exemption for patient from art. 53 exists. Selling home product is a separate matter under novel food regulations.

What does this mean in practice?

Decarboxylation is a simple reaction, but numbers come from various materials. “110°C and 30 minutes” is a result for THCA-A in extract (Wang et al., 2016). In ground herb, half of CBDA disappeared after 44.4 minutes at 105°C (Ryu et al., 2021), so an hour in the oven does not guarantee full conversion.

Higher temperature shortens process but increases cannabinoid and terpene loss. Covering limits oxygen access and vapor escape, though no one measured gain for herb. CBN forms mainly with oxygen and time, and 145°C threshold is not confirmed by any reviewed source. Vaporizer does not need separate decarboxylation, as it performs it in seconds. No studies on home oven, jar, or sous-vide were found in PubMed or Europe PMC.

Home process allows reaction but not result calculation. If you need known content, you need batch test from a lab, not home calculator. When eating, remember delay and keep products away from children. All this applies exclusively to industrial hemp herb.

Frequently Asked Questions

What is decarboxylation?

It is a reaction in which a carboxylic acid loses carbon dioxide. In cannabis, heat converts THCA acid into THC, and CBDA acid into CBD. The plant mainly produces acidic forms, and neutral forms do not occur in significant concentrations (Wang et al., 2016). THC constitutes 87.7% of the mass of the acid from which it forms.

What temperature and how long to keep herb in the oven?

No one has measured a home oven. In a closed laboratory oven, half of CBDA in ground herb disappeared after 44.4 minutes at 105°C and after 32.6 minutes at 120°C, but not within an hour at 90°C (Ryu et al., 2021). “110°C for 30 minutes” is a result for THCA-A in extract, not herb.

Can decarboxylation be done in a vaporizer?

The vaporizer does it by itself. In an in vitro measurement, electric devices set to 210°C converted at least 97.3% THCA-A and 94.6% CBDA from 50 mg of herb (Lanz et al., 2016). The recommended operating temperature of vaporizers is usually 160-220°C (Eyal et al., 2023). Separate heating of the herb before vaporization is therefore not necessary.

Does vaporized herb need to be decarboxylated again?

Data indicate that most acids have already reacted in the device: in the trial by Eyal et al. from 2023, decarboxylation at 180°C was nearly complete after 40 seconds. Residues after the session contained 1.1-15.0% cannabinoids (Lanz et al., 2016). How such herb behaves during further processing has not been studied.

Does THC convert to CBN above 145 degrees?

No such threshold was found in any reviewed source. In a vacuum dryer at 110°C, CBN practically did not form, and pure THCA-A produced CBN after 15 minutes at 160 and 180°C (Dussy et al., cited by Wang et al., 2016). CBN formation mainly depends on oxygen and time: oxidation by air in darkness produced CBN (Fairbairn et al., 1976).

Are acidic forms useless before heating?

No, they have a different activity profile. THCA-A bound weakly to CB1 and CB2 receptors, while THC bound 62 and 125 times stronger respectively (McPartland et al., 2017). Cannabinoid acids activated the PPAR gamma receptor more strongly than neutral forms (Nadal et al., 2017). These are results from cells and animals, not human studies.

Is the terpene boiling point table reliable?

Partially. It comes from the work of McPartland and Russo from 2001, not from Russo’s 2011 review. Values for monoterpenes, e.g. myrcene 166-168°C, agree with normal boiling points. Beta-caryophyllene is listed as 119°C instead of about 263°C, and THC as 157°C instead of estimated about 425°C (Eyal et al., 2023).

This article is for informational and educational purposes and does not constitute legal advice. The legal status described applies as of publication date: cannabis regulations may change. Consult a lawyer or current legal acts before making decisions.

Author: Michał Waluk · Published: 2026-05-11 · Updated: 2026-09-20