When you lift a parquet board in an old building, there is sometimes a black, viscous, tar-like layer beneath it. Two such layers can look completely identical to the eye and still differ by orders of magnitude in their content of carcinogenic compounds. The reason lies not in the appearance but in the origin: coal tar is a high-temperature product of coal coking, bitumen a residue of crude-oil distillation, and these two origins leave an opposite molecular fingerprint.
This article follows three threads the usual account leaves out: why the origin explains the order-of-magnitude difference and how to tell tar from bitumen beyond a bare lab value; why, of all things, the most dangerous compounds betray themselves not by smell or outgassing but travel invisibly with the dust; and why a high material value is a reason for calm clarification, not for panic.
Tar is cooked, bitumen is distilled
Polycyclic aromatic hydrocarbons (PAH) are a group of fused carbon rings. They form when organic material burns incompletely or is decomposed under heat, and they are concentrated in coal tar. This tar is a by-product of the coking of coal: the coal is heated under exclusion of oxygen to up to about 1,300 °C (IARC 2012, Monograph 100F), driving off tars and oils that are then fractionally distilled, from light oil through carbolic and creosote oil to anthracene oil, while the residue, the pitch, remains: roughly half of the tar, and a veritable PAH concentrate.
Bitumen has a completely different biography. It is the non-volatile residue of the vacuum distillation of crude oil, and it is never brought to the temperatures at which new ring systems form; the processes of asphalt production stay below about 400 °C (Machado et al. 1993). It is precisely this temperature axis that separates the two materials. Why the heat produces PAH at all, and preferentially the heavy, unsubstituted ring systems, is the actual reason the fingerprint works later on.
For experts (or the expert-curious): how heat builds PAH
At combustion and pyrolysis temperatures the first aromatic ring assembles from small radicals, for instance through the recombination of two propargyl radicals (C3H3), and then grows repetitively through alternating hydrogen abstraction and acetylene addition. This mechanism carries the name HACA (hydrogen abstraction, C2H2 addition; Frenklach 2002).
Decisive for the fingerprint is why high temperature preferentially builds unsubstituted parent PAH: the acetylene addition is reversible and only becomes irreversible when the system locks onto particularly stable, compact, pericondensed structures, the so-called stabilomers (descriptor after Stein and Fahr 1985); the dominant growth path follows the smallest thermodynamic resistance (Frenklach 2002). The signature of a hot process is therefore a pattern of bare, more-highly-condensed parent PAH, not of alkylated side chains. That is the molecular reason coal tar looks so different from any petroleum product.
The molecular fingerprint: tar or bitumen?
Optically the two cannot be told apart; chemically they are opposites. Coal tar (pyrogenic, from high-temperature pyrolysis) is dominated by unsubstituted parent PAH and carries a great deal of benzo[a]pyrene; bitumen (petrogenic) is marked by alkylated homologs and carries very little. How large the difference becomes was measured by Machado et al. (1993) on the same analyses: benzo[a]pyrene stood at 18,100 mg/kg in coal-tar pitch but below 6 mg/kg in the whole asphalt (the two asphalt samples measured 3.2 and 5.3 mg/kg), more than three orders of magnitude lower; the sum of the 16 EPA PAH was 294,300 in the pitch versus 50.4 and 122.3 mg/kg in the asphalt (Machado et al. 1993, Table 2). The fumes of the heated materials too contained two to three orders of magnitude more benzo[a]pyrene and were about a hundred times more mutagenic.
A laboratory reads this difference in origin not only off the pure benzo[a]pyrene value but off the ratios of certain isomers to one another, the actual fingerprint. Intellectual honesty is required here: these ratios are indications, not proof. Weathering and photo-oxidation shift the more sensitive ones, and of all things the canonical source for the method bears the subtitle "a critical appraisal" (Yunker et al. 2002).
For experts (or the expert-curious): the diagnostic isomer ratios
Yunker et al. (2002) established four ratios of same-mass isomer pairs with named thresholds:
| Ratio (mass) | petrogenic | transition | combustion / pyrogenic |
|---|---|---|---|
| Anthracene/(An+Phenanthrene) [178] | < 0.10 | > 0.10 | |
| Fluoranthene/(Fl+Pyrene) [202] | < 0.40 | 0.40 to 0.50 | > 0.50 (wood/coal, i.e. tar) |
| Benz[a]anthracene/(BaA+Chrysene) [228] | < 0.20 | 0.20 to 0.35 | > 0.35 |
| IcdP/(IcdP+Benzo[ghi]perylene) [276] | < 0.20 | 0.20 to 0.50 | > 0.50 |
Why the fluoranthene/pyrene ratio works is a fine piece of chemistry: fluoranthene and pyrene are isomers of the same formula C16H10, but pyrene is more stable by about 66 kJ/mol. The rearrangement of fluoranthene to pyrene is exothermic yet sits kinetically frozen behind a high barrier; a combustion quenches the mixture far from equilibrium and thereby preserves the actually less stable fluoranthene (Khiri et al. 2019). A high fluoranthene share is thus the imprint of a hot, rapidly cooled process.
Not all ratios are equally robust. Resistant to weathering are fluoranthene/(Fl+pyrene) and IcdP/(IcdP+BghiP); sensitive (photolabile) are anthracene/(An+Phe) and benz[a]anthracene/(BaA+chrysene), because the linear anthracene decays faster than phenanthrene and can thereby make a genuinely pyrogenic sample appear "petrogenic" (Tobiszewski and Namieśnik 2012). A second, more robust tool is the distribution of the alkyl homologs: petrogenic bell-shaped with a maximum at the singly to doubly substituted compounds, pyrogenic dominated by the unsubstituted parent compound (Youngblood and Blumer 1975). One subtlety often reproduced wrongly: pyrogenic PAH by no means "lack" the alkyl homologs; combustion samples contain whole series of them; what distinguishes them is the shape of the distribution, not its presence or absence (Youngblood and Blumer 1975).
Where the tar sits, and where the most is
Coal tar is intrinsically about 20 percent PAH (16 EPA) by mass, a figure that can be computed directly from its composition (naphthalene about 10 percent, phenanthrene about 4.5, fluoranthene about 3, pyrene about 2; Franck and Stadelhofer 1987). From this follows a rule of thumb that orders the whole material story: the purer the tar product, the higher the PAH load.
The famous black parquet adhesive is tar-based, but installed as a solvent-thinned thin film. Its benzo[a]pyrene content scatters widely: across about 2,500 samples examined, the median and 95th percentile lay at roughly 800 and 8,000 mg benzo[a]pyrene per kilogram of adhesive respectively (DIBt PAH guidance 2000). Until the end of the 1960s, parquet was frequently glued with coal-tar pitch; tar-containing adhesives have not been produced in Germany since the mid-1970s (DIBt PAH guidance 2000).
The real blind spot, however, is not the adhesive. Tar-containing sealants and damp-proofing, that is tar coatings on basement masonry, tar roofing felt, tar cork and horizontal barriers, are not present as a thin, diluted film but almost in pure form as the binder. At about 20 percent PAH in pure tar, that means contents in the percent range, far above the solvent-thinned thin-film adhesive. Precisely for this reason the plastered, inconspicuous damp-proofing in the wall often carries the highest load in the entire building, while it is left standing during partial renovations and only exposed after deconstruction.
The most important qualification remains the same as for the adhesive: not every black building material is tar-containing. Pure bitumen contains only about 2 mg benzo[a]pyrene per kilogram (DIBt PAH guidance 2000), roughly four orders of magnitude less than tar pitch. A tar-like or mothball-like smell (naphthalene is the most volatile PAH) can be a hint but is no proof; and an ordinary bitumen mastic asphalt is not to be confused with a tar-containing one. Only analysis gives certainty.
Why it is dangerous, and what is often misunderstood about it
The sentence "PAH outgas from the material" is simply wrong for the dangerous members. The partitioning between air and dust follows the ring number: the light two- to three-ring PAH are in the vapour phase, from four rings it tips over, and the five- to six-ring carcinogens such as benzo[a]pyrene are practically non-volatile (vapour pressure on the order of 10⁻⁹ mmHg) and cling to the dust (ATSDR; WHO 2010).
From this follows an inversion of intuition. What you can smell, the naphthalene with its mothball note, is the most volatile and at the same time by far the weakest carcinogen of the group (toxicity equivalency factor 0.001; IARC Group 2B). The actual carcinogen benzo[a]pyrene (factor 1, IARC Group 1) is odourless and does not outgas; it travels bound to dust as soon as the material is worked or destroyed. "No tar smell" therefore does not mean "little benzo[a]pyrene". Naphthalene itself, incidentally, is an indoor concern in its own right, with a WHO guideline of 10 µg/m³ (WHO 2010).
The path from material to person thus decides the risk: an intact, sealed floor releases little; it becomes critical when sanding, milling, drilling or tearing out generate fine, PAH-laden dust that spreads through the flat and settles permanently in the house dust. PAH are then taken up by three routes: inhalation of the dust, oral uptake via hand-to-mouth contact (disproportionately in small children), and through the skin, which tar PAH penetrate well.
An honest differentiation is needed here, and it cuts both ways. A high material value is a reason for calm clarification, not for panic, because the material content is not the absorbed dose. When the Frankfurt public health office examined the PAH metabolites in the urine of 347 children under six from such flats, it found no statistically significant additional burden between the children from heavily benzo[a]pyrene-contaminated and those from tar-free flats (Heudorf and Angerer 2000). The reason is precisely not that the tar is harmless, but that a person's PAH burden comes from many sources: more than 90 percent of the daily total intake comes from food, indoor air usually lies well below 1 ng benzo[a]pyrene per cubic metre, and for the PAH metabolites in children's urine a connection with passive smoking rather than with the adhesive was suggested (Stahl et al. 2004). The tar adhesive is thus one real source among several real sources. That is exactly why a robust assessment looks at the whole exposure picture and at the dust and exposure data, not at the single largest material figure alone.
Honesty also requires noting that this reassuring reading did not go unchallenged: the official risk assessment of the contaminated flats at the time was criticised as methodologically flawed (Obenland 2003). Regardless of the outcome of that debate, the practical consequence remained precautionary: above defined house-dust values, reduction measures are recommended. And a qualification that is often drawn too broadly: the cancer sites well documented for tar and soot are skin and lung (IARC 2012, Monograph 100F); the sufficient evidence for bladder cancer belongs to aluminium production, not to parquet adhesive across the board.
For experts (or the expert-curious): from benzo[a]pyrene to mutation
Benzo[a]pyrene is a pro-carcinogen; only its metabolite becomes dangerous. Via the enzymes CYP1A1/1A2/1B1 and epoxide hydrolase, (+)-anti-benzo[a]pyrene-7,8-diol-9,10-epoxide (BPDE) forms, a bay-region diol epoxide (IARC 2012, Monograph 100F). It binds covalently to the exocyclic nitrogen N2 of guanine.
How direct the bridge from this molecule to human cancer is was shown by Denissenko et al. (1996): they mapped the BPDE adducts along the tumour-suppressor gene P53 and found strong, selective adduct formation at the guanines of codons 157, 248 and 273, that is exactly at the mutational hotspots of human lung cancer; the resulting G-to-T transversions concentrate to more than 90 percent on the poorly repaired, non-transcribed strand. This is a direct molecular link between a defined chemical carcinogen and the mutational pattern of a human cancer. The diol-epoxide pathway itself goes back to Sims et al. (1974).
The regulatory frame in Austria
Tar-containing construction and demolition waste counts as hazardous in Austria. In the waste catalogue, the Waste Catalogue Ordinance 2020 (Abfallverzeichnisverordnung) lists it under the Austrian waste key number (Schlüsselnummer) 54913 ("tar residues"); this is not to be confused with the EU waste-list code 17 03 03*, which denotes the same waste in the European system. A finding is best served by naming both. The decisive hazard property is HP 7 (carcinogenic), which applies from 0.1 percent by mass of a substance classified as Carc 1A/1B, for benzo[a]pyrene therefore from 1,000 mg/kg, which is not to be confused with the German 50 mg/kg material value (Waste Catalogue Ordinance 2020).
Worker protection applies to removal. Contrary to what is often stated, Austria does have a binding benzo[a]pyrene value at the workplace: the Limit Values Ordinance 2025 (Grenzwerteverordnung) sets a technical guide concentration of 0.001 mg/m³ as a daily mean (0.005 mg/m³ in pitch-rod production and the oven area of coking plants) and lists coal tars, pitches and oils expressly as "clearly carcinogenic". Ambient air too has a binding limit, 1 ng benzo[a]pyrene per cubic metre as an annual mean (Ambient Air Quality Act, since 2013). What is missing is, of all compartments, the indoor one: for benzo[a]pyrene or the PAH sum there is no binding indoor-air value in Austria; the only PAH-related indoor value is a non-binding guideline for naphthalene of about 10 µg/m³ (BMK 2024).
German rules are drawn upon in Austria as the acknowledged state of the art. TRGS 551 ("Tar and other pyrolysis products from organic material") classifies a material as carcinogenic from 50 mg/kg benzo[a]pyrene. It is important that this threshold is a material classification and not a remediation trigger: the actual remediation recommendations of the DIBt attach to house-dust values, to 100 mg benzo[a]pyrene per kilogram of fresh dust in living spaces and 10 mg/kg where infants and small children are affected (DIBt PAH guidance 2000; Stahl et al. 2004). Conversely, a content below 10 mg benzo[a]pyrene per kilogram in the adhesive practically rules out a relevant tar fraction.
What to do
From all this follows a sober approach. An intact, sealed floor without odour is no acute reason to act; the situation should be documented and taken into account at the next renovation. Before a floor renovation in a building erected before about 1980, it is worth clarifying beforehand what is stuck under the parquet and what sits as damp-proofing in the plinth and basement wall. If tar is found, its removal does not belong in one's own hands: no sanding, milling or tearing out, but controlled deconstruction with containment, negative pressure, extraction and protective equipment by a specialist firm, with disposal as hazardous waste (SN 54913, HP 7). Sampling on one's own is inadvisable, because even scratching at suspect material generates dust and risks skin contact.
Why a single benzo[a]pyrene value underestimates the toxicity of a tar mixture, and how the toxicity equivalent remedies this, is covered in the article One value is not enough: the toxicity equivalent for PAH. Which pollutants are to be expected in which building era is set out in Pollutants by construction year; asbestos in floor coverings and adhesives is covered in How to identify asbestos, and lead in paints and pipes in Lead in old buildings.
A look back
That of all things coal tar stands at the beginning of cancer research is no accident. As early as 1775 the London surgeon Percivall Pott described scrotal cancer in young chimney sweeps and traced it to the soot, the first clearly named occupational chemical carcinogen (documented in IARC 2012, Monograph 100F). In 1915 Yamagiwa and Ichikawa produced carcinomas experimentally with tar, and in 1933 Cook, Hewett and Hieger isolated from coal tar the cancer-producing hydrocarbon, benzo[a]pyrene, the first chemically pure carcinogen (Cook et al. 1933). It is the same index substance on which the assessment of tar-containing materials still orients itself today.
Sources
- IARC/WHO (2012), Monograph 100F: benzo[a]pyrene (Group 1), coal tar and pitch, soot and chimney sweeps (Pott), coking temperatures, bay-region activation. Free: publications.iarc.who.int.
- Machado, M.L. et al. (1993): Evaluation of the Relationship between PAH Content and Mutagenic Activity of Fumes from Roofing and Paving Asphalts and Coal Tar Pitch. Fundamental and Applied Toxicology 21(4): 492–499.
- Yunker, M.B. et al. (2002): PAHs in the Fraser River basin: a critical appraisal of PAH ratios as indicators of PAH source and composition. Organic Geochemistry 33(4): 489–515.
- Tobiszewski, M. & Namieśnik, J. (2012): PAH diagnostic ratios for the identification of pollution emission sources. Environmental Pollution 162: 110–119.
- Khiri, D. et al. (2019): Theoretical Investigation of the Reaction of Pyrene Formation from Fluoranthene. Journal of Physical Chemistry A 123(34): 7491–7498.
- Youngblood, W.W. & Blumer, M. (1975): Polycyclic aromatic hydrocarbons in the environment: homologous series in soils and recent marine sediments. Geochimica et Cosmochimica Acta 39: 1303–1314.
- Frenklach, M. (2002): Reaction mechanism of soot formation in flames. Physical Chemistry Chemical Physics 4: 2028–2037. Stabilomer descriptor: Stein, S.E. & Fahr, A. (1985), J. Phys. Chem. 89: 3714.
- Denissenko, M.F. et al. (1996): Preferential Formation of Benzo[a]pyrene Adducts at Lung Cancer Mutational Hotspots in P53. Science 274(5286): 430–432. Diol-epoxide pathway: Sims, P. et al. (1974), Nature 252: 326–328.
- Heudorf, U. & Angerer, J. (2000): Humanbiomonitoring auf PAK-Metaboliten im Urin von Kindern aus Wohnungen mit PAK-haltigem Parkettkleber. Umweltmedizin in Forschung und Praxis 5(4): 218–226.
- Stahl, T., Heudorf, U., Moriske, H.-J. & Mersch-Sundermann, V. (2004): Polyzyklische aromatische Kohlenwasserstoffe (PAH) im Innenraum. Bundesgesundheitsblatt 47: 868–881.
- Obenland, H. (ARGUK-Umweltlabor, 2003): US-Housings, five years on. A review of a chapter of official risk assessment.
- Cook, J.W., Hewett, C.L. & Hieger, I. (1933): The Isolation of a Cancer-producing Hydrocarbon from Coal Tar. Journal of the Chemical Society 1933: 395–405.
- WHO Regional Office for Europe (2010): WHO Guidelines for Indoor Air Quality: Selected Pollutants (naphthalene guideline, PAH, benzo[a]pyrene as indicator).
- ATSDR: Toxicological Profiles for Coal Tar/Coal Tar Pitch, Wood Creosote and Polycyclic Aromatic Hydrocarbons.
- Franck, H.-G. & Stadelhofer, J.W. (1987): Industrielle Aromatenchemie. Springer (coal-tar composition).
- Waste Catalogue Ordinance 2020 (Abfallverzeichnisverordnung, BGBl. II No. 409/2020), SN 54913 / EU code 17 03 03*, HP 7, RIS: ris.bka.gv.at.
- Limit Values Ordinance 2025 (Grenzwerteverordnung, BGBl. II No. 339/2025), benzo[a]pyrene TRK; Ambient Air Quality Act (BaP 1 ng/m³); BMK (2024): guideline for naphthalene in indoor air.
- TRGS 551 "Tar and other pyrolysis products from organic material" (BAuA); DIBt, PAH guidance (DIBt-Mitteilungen 2000), house-dust remediation values.
Questions on the topic? Free Initial Consultation.