What Happens To A Molecule When You Set Fire To It
A chemical's safety file describes the intact molecule. It rarely describes what that molecule becomes at 600°C — and for one widely used class, we now have measurements.
A chemical's safety file describes the intact molecule. Toxicologists feed it to animals, measure what happens, and derive a dose.
What that file almost never describes is what the molecule becomes at 600 degrees.
For most agricultural chemistry this doesn't matter, because most residues are eaten. Combustion changes the question entirely — and for one family of compounds there is now real data on what that change looks like.
The experiment that has been done
Triazole compounds — a large class built around a five-membered ring containing three nitrogen atoms — are among the most widely used substances in horticulture. In 2019 and 2020, researchers at Poland's Central Institute for Labour Protection did something unusual: they burned two of them under controlled, standardised fire conditions and identified what came off.
The apparatus matters, because it is what makes the results credible. Simultaneous thermal analysis from 25 to 900 degrees with online infrared spectroscopy; a steady-state tube furnace built to the ISO/TS 19700 standard; solid-phase microextraction feeding gas chromatography–mass spectrometry; and three separate fire conditions — 350 degrees for oxidative pyrolysis, 650 for a small flaming fire, 825 for post-flashover conditions.
What came off
The products were not the parent compound.
At 350 degrees, triadimenol yielded 4-chlorophenol, chlorinated benzaldehyde and chlorinated ethoxybenzene. At 650 degrees, large quantities of chlorinated benzofurans appeared. At 825 degrees those persisted and were joined by polycyclic aromatic hydrocarbons — naphthalene, acenaphthylene, fluorene, anthracene. Infrared analysis identified hydrogen cyanide coming directly off the triazole ring.
The authors' own summary lists chlorophenols, chlorinated naphthalenes, chlorinated biphenyl and dibenzofuran among the effluents. That is a materially different hazard profile from the substance that went in.
Three things this does not establish
It does not establish dioxin formation. The researchers measured *precursors* — chlorophenols, chlorobenzenes, chlorinated benzofurans. Their statement about polychlorinated dibenzofurans is explicitly conditional: such compounds "may also be presented" in fire effluents. No study has measured a single dioxin or furan congener from any triazole compound. Anyone telling you burning these substances produces dioxins has gone beyond the evidence.
It does not establish quantities. Products were identified by spectral library matching rather than against authentic standards, and quantification was relative peak area normalised to 100 per cent — no absolute yields, no mass balance. The work establishes *that* these compounds form. It cannot tell you *how much*.
And it was done on two specific compounds, not on a whole class. Applying it to a structurally similar molecule is inference, and should be labelled as such.
The counter-evidence, which is real
Honesty requires the other side of this.
The only growth regulator ever actually measured in smoke behaved far less dramatically. A 2012 study of maleic hydrazide on tobacco, using pyrolysis gas chromatography at 750 degrees, found that only 1.4 to 3.7 per cent reached mainstream smoke, that the great majority was destroyed during burning, and that its breakdown products — ammonia, carbon dioxide, butanoic acid — were small and unremarkable. No novel toxicant appeared.
The tobacco literature suggests a mechanism for why. Mainstream smoke residues appear to consist largely of unpyrolysed parent compound carried over by steam distillation, while pyrolysis products go preferentially into sidestream smoke. If that partitioning holds generally, then what reaches a user's lung is dominated by the intact compound rather than exotic combustion products.
And where regulators have actually required pyrolysis studies — for tobacco — the results have not raised alarm. Washington State's agriculture department records that "data from pyrolysis studies have not led to an exceedance in the level of concern".
The precedent nobody followed up
There is an older piece of work worth knowing, because it shows how tractable this question is.
In 1979, forensic scientists published an analysis of a quaternary ammonium herbicide pyrolysed directly on plant material. Using coupled gas chromatography and infrared spectroscopy they identified its breakdown products, established that conversion was complete at 610 degrees, and then detected those same products in genuinely contaminated material seized from the field.
That study is forty-seven years old. It used equipment now considered primitive. It answered exactly the question that remains open today for every plant growth regulator in commercial use: put the residue on the plant, burn it under realistic conditions, identify what comes off.
Nobody has run it since — not for paclobutrazol, not for daminozide, not for chlormequat.
Frequently asked questions
What happens to pesticides when they are burned?
It depends on the compound, and for most it has never been measured. Where it has been studied, triazole compounds burned under standardised fire conditions produced 4-chlorophenol, chlorinated benzofurans, chlorinated naphthalenes, polycyclic aromatic hydrocarbons and hydrogen cyanide. However, a growth regulator studied on tobacco was largely destroyed during burning, with only 1.4–3.7% reaching mainstream smoke and unremarkable breakdown products.
Does burning agricultural chemicals produce dioxins?
No study has measured a dioxin or furan congener from any triazole compound. Research has identified dioxin *precursors* — chlorophenols, chlorobenzenes and chlorinated benzofurans — and the researchers themselves framed the dioxin question conditionally, stating such compounds 'may also be presented' in fire effluents. Claims that burning these substances produces dioxins go beyond what has been measured.
Do combustion products appear on a chemical's safety file?
Rarely, and where they do appear on safety data sheets they are typically inferred from elemental composition rather than measured. Pesticide registration does not require combustion or pyrolysis testing for compounds that are not intended to be burned. Reference doses describe the intact molecule administered orally.
References
- 1.Borucka & Celiński, Chemical Engineering Transactions 77:139–144 (2019)
- 2.Borucka et al., J Thermal Analysis and Calorimetry 139:1493–1506 (2020)
- 3.Zhang et al., The Scientific World Journal 2012:451471 (maleic hydrazide)
- 4.Summoogum et al., Chemosphere 85:143–150 (2011)
- 5.Beutler, Varano & DerMarderosian, J Forensic Sci 24(4):808–813 (1979)
- 6.WSDA Criteria for Pesticides, Publication 398 (2020)
Entries in a similar key
9 min read
What Are Plant Growth Regulators, And Are They Bad?
A plain-English guide to plant growth regulators — what they do, which ones raise genuine safety questions, and what the evidence actually shows.
8 min read
The Study Nobody Has Run
There is a rule in US pesticide regulation that almost nobody outside the field knows about — and the fact that it applies to exactly one crop is the most interesting thing in this subject.
8 min read
Eating It And Breathing It Are Not The Same Exposure
Almost every number in agricultural toxicology describes swallowing. Two things change when a substance is inhaled instead — and the extrapolation between them has never been validated.
Editorial note: Journal is the Golden Ratio Clinics editorial section. Articles are general health-service writing and do not constitute medical advice. All clinical decisions are made by your doctor during a consultation.
All Journal entries