From the 1960s into the early 1990s, roof structures, beams and timber cladding were chemically treated on a vast scale, usually with two active ingredients in combination: pentachlorophenol (PCP) against fungi and lindane (the γ-isomer of hexachlorocyclohexane) against insects, often in a ratio of about ten to one. They were sold under brand names such as Xylamon or Xyladecor. In buildings from that period, treated timber is the rule rather than the exception. Both substances are semi-volatile: they off-gas slowly from the wood over years and accumulate in house dust, long after the last treatment. The contamination arises from that, not from the wood itself.
Why the contamination persists, and whom it affects
PCP and lindane are semi-volatile organic compounds. They evaporate over years from the treated wood, deposit on surfaces and collect in house dust, which thus becomes the actual long-term reservoir (UBA 2008). This pathway explains why the substances are still measurable in room air and dust decades after application, and it also explains why the contamination is not confined to the attic: dust is displaced, and the substances turn up in the living spaces below.
That is decisive for the question of who in the household is most exposed. Not the person who occasionally enters the attic, but the small child who spends most of its time indoors, plays on the floor and regularly takes up house dust via hand-to-mouth contact (UBA 2008). The relevant intake runs through the dust, not through brief stays under the roof. A rarely used but treated roof structure is therefore not a closed matter as long as people live below it.
Where treated timber sits
- Roof structures: the classic location. Rafters, purlins and collar beams were coated over the surface or treated in a dip bath.
- Timber-beam ceilings and half-timbering: in old buildings with visible wood.
- Cellar beams: moisture-exposed wood was treated especially intensively.
- Cladding and panelling: wall panelling and ceiling panels.
A sign can be a brownish, greenish or oily discolouration of the wood, occasionally also a chemical smell, especially on warm days or in poorly ventilated rooms. The visual impression is not reliable; only an analysis gives certainty.
Why PCP and lindane are dangerous
PCP and lindane act through two entirely different mechanisms. PCP attacks the cell's energy metabolism and damages above all the liver (ATSDR 2022). A second, historically important problem is an impurity: technical PCP was only about 86 to 90 percent pure and contained dioxins and furans as manufacturing by-products (ATSDR 2022). It is these accompanying substances, not the PCP itself, that account for the chloracne and part of the immune effects associated with technical PCP (ATSDR 2022). Lindane, by contrast, is a neurotoxin: it dampens the inhibitory signalling of the central nervous system and can thereby trigger overexcitation up to convulsions (IPCS 1996); it also affects the liver, is hormonally active, and, because it is fat-soluble, distributes into fatty tissue.
Both substances are classified as carcinogenic to humans (Group 1), each on the basis of non-Hodgkin lymphoma: PCP since the reassessment by the International Agency for Research on Cancer, IARC (Monograph 117, IARC 2019), lindane since the 2015 evaluation (Monograph 113, IARC 2015). This classification is serious, but it needs to be read correctly. Group 1 describes how well the carcinogenic effect in humans is established, not the level of risk in a particular situation. It was established in occupationally heavily exposed workers, whose exposure lies orders of magnitude above what a treated roof structure releases into room air. It therefore does not follow from the classification that such a roof structure causes cancer, but that the substances are to be taken seriously and exposure avoided as far as possible. How high the contamination is in a specific case, only measurement can tell.
For experts, or those who want to become one: How PCP and lindane act in the body, and why the chloracne does not come from PCP
PCP is an uncoupler of oxidative phosphorylation: in the mitochondria it breaks the coupling between the respiratory chain and ATP formation, so that the absorbed energy is lost as heat instead of being stored as ATP. At high acute intake this manifests as overheating, sweating and a racing heart; fatal poisonings among heavily exposed workers are documented (ATSDR 2022). In the chronic, low-dose range, liver damage is to the fore.
The chloracne and the immunological effects once attributed wholesale to PCP are with high probability effects of the impurities. Technical PCP was produced by chlorinating phenol at high temperatures, which formed polychlorinated dibenzodioxins and furans (PCDD/PCDF); pure PCP did not cause chloracne in animal experiments, technical PCP did (ATSDR 2022). Chloracne is the classic signature of dioxin exposure. Important for interpreting the cancer data: IARC attributes the carcinogenic effect of PCP expressly to the PCP itself and not to the dioxin impurity (IARC 2019); the 2,3,7,8-TCDD is separately classified as carcinogenic (Group 1, IARC 2012).
Lindane is a non-competitive antagonist at the GABA-A-gated chloride channel (binding at the picrotoxin site). By blocking the inhibitory chloride influx, it disinhibits the central nervous system, which shows as tremor, ataxia and convulsions (IPCS 1996). A dataset of its own comes from the former use of lindane as a medicine against scabies and lice, where seizures and deaths were described (ATSDR 2024). With an octanol-water partition coefficient (log Kow) around 3.7, lindane is lipophilic and distributes into fatty tissue; lindane itself, however, is metabolised and excreted comparatively quickly, whereas the especially long-lived β-isomer of HCH remains in fatty tissue for years (ATSDR 2024).
What the studies show
The cancer classification does not rest on a single result but on a converging picture from several study types. For PCP, two independent cohorts of workers in PCP manufacture show elevated rates of non-Hodgkin lymphoma, a third cohort of sawmill workers links rising PCP exposure to rising lymphoma risk, and case-control studies confirm the association. For lindane, a meta-analysis and a prospective cohort of farmers deliver the same signal. That cohorts, case-control studies and meta-analyses point in the same direction is why IARC speaks of an established association. This evidence too comes from the occupational sphere and says nothing about the level of a residential exposure; it grounds the case for clarifying and reducing the contamination, not for extrapolating it into a personal cancer risk. The details, with figures, are in the box.
For experts, or those who want to become one: The epidemiological evidence in detail, and the dispute over the "wood-preservative syndrome"
PCP and non-Hodgkin lymphoma. IARC (Monograph 117) bases the classification on several worker studies. In the cohort of US PCP manufacture (Dow/Michigan), mortality from non-Hodgkin lymphoma was elevated in the PCP-exposed group (standardised mortality ratio (SMR) 2.4; 95% confidence interval 1.0 to 4.8; Collins 2009); the authors are Dow-affiliated, and the conflict of interest is disclosed in the paper. An independent cohort of the US occupational-safety institute NIOSH likewise found elevated lymphoma mortality (SMR around 1.8; Ruder & Yiin 2011). Particularly informative, because closer to real wood-preservative use, is a large cohort of Canadian sawmill workers: here the frequency of non-Hodgkin lymphoma rose with cumulative PCP exposure, while the co-used tetrachlorophenol showed no such trend (Demers 2006). Case-control studies support the picture: a pooled Swedish analysis found an elevated lymphoma risk for chlorophenols (Hardell 2002), and a nested case-control study within the international IARC register separated the contributions of phenoxy herbicides, chlorophenols and dioxins (Kogevinas 1995).
Lindane and non-Hodgkin lymphoma. A systematic review with meta-analysis pools the observational studies and finds for lindane a summary relative risk of about 1.6 (Schinasi & Leon 2014). The large prospective US cohort of the Agricultural Health Study additionally shows a dose-response relationship: lymphoma risk rose with increasing lindane exposure (Purdue 2007). Prospective cohort and meta-analysis together were the basis of the IARC classification in 2015 (IARC 2015).
Putting the level in perspective. These findings come from the occupational sphere, with exposures far above those of a dwelling. They establish that the substances are carcinogenic, not that a residential contamination carries a particular cancer risk. From the combination of long-lived contamination, a sensitive population group (small children) and the precautionarily assumed absence of a threshold follows the precautionary principle: clarify and reduce the contamination, rather than assume a "safe" concentration.
The "wood-preservative trial" and the "syndrome". In Germany, the Frankfurt Regional Court in 1993 convicted two managing directors of the manufacturer Desowag of negligent bodily harm to residents of treated homes. The Federal Court of Justice overturned the judgment in 1995, however on a procedural ground and not because causation had been scientifically refuted (BGH 1995). The individual toxic effects (liver damage, acute uncoupling poisoning, convulsions from lindane, dioxin-related chloracne, the elevated lymphoma risk) are established. A unified, clearly delimited "wood-preservative syndrome" of diverse non-specific complaints at low residential exposure, by contrast, is scientifically contested as an independent clinical entity and not established.
What the bans achieved. The effect of the bans can be read off in humans. In the German Environmental Specimen Bank, the PCP concentration in blood plasma fell markedly after the ban, from over 30 µg/l in the mid-1980s to a few µg/l around the turn of the millennium and further thereafter (UBA-ESB). At the population level the bans thus worked; the remaining problem is local indoor reservoirs in old buildings.
The chemistry behind it
Why do these substances persist so long? Because they are chemically built to be stable. The many chlorine-carbon bonds make them hard to break down, their low water solubility and their fat solubility let them linger in materials and organisms. For PCP the dioxin impurity from manufacture is added; for lindane a legacy of its own magnitude, since for every tonne of the active isomer, historically eight to twelve tonnes of unusable sister isomers were produced (Vijgen 2011). The specifics are in the box.
For experts, or those who want to become one: Synthesis and dioxin formation, the HCH-isomer problem, and why the substances are persistent
PCP and the dioxins. Technical pentachlorophenol was obtained by chlorinating phenol (or by hydrolysing hexachlorobenzene) at high temperatures. If the alkali salts of PCP are heated above about 300 °C, two molecules condense into polychlorinated dibenzodioxins; because the precursor is pentachlorinated, the highly chlorinated congeners dominate, above all octachlorodibenzodioxin (OCDD) and the heptachloro compounds, while the especially toxic 2,3,7,8-TCDD occurs only in traces (WHO 1987; IARC 2019). This explains why technical PCP carries a dioxin burden whose congener pattern differs from that of other dioxin sources.
The HCH-isomer problem. Hexachlorocyclohexane forms during the photochlorination of benzene as a mixture of several isomers (α, β, γ, δ and others). Only the γ-isomer is practically insecticidal, and lindane is material purified to over 99 percent γ-HCH. Technical HCH, by contrast, contains only about 10 to 15 percent γ-HCH; the purification correspondingly leaves large quantities of the other isomers. On balance, for every tonne of lindane about eight to twelve tonnes of waste isomers arose, a global legacy of millions of tonnes (Vijgen 2011). The β-isomer is the most persistent: its chlorine atoms all stand equatorial, which geometrically blocks the elimination of hydrogen chloride needed for breakdown; β-HCH therefore reacts orders of magnitude more slowly than the other isomers and dominates in the environment and in human fatty tissue (ATSDR 2024).
Persistence. PCP is a weak acid (acid constant pKa around 4.7) and at indoor and environmental pH is present predominantly as the phenolate anion, which co-determines its distribution and mobility; its octanol-water coefficient (log Kow) is around 5. Lindane, with log Kow around 3.7, is lipophilic and only sparingly soluble in water. Both are semi-volatile enough to off-gas from the wood over years, and both are listed as persistent organic pollutants (POP); the HCH isomers in particular are strongly bioaccumulative and are carried through the atmosphere into remote regions. Persistence, bioaccumulation and long-range transport are the criteria by which both were taken into the Stockholm Convention and the POP Regulation (Vijgen 2011).
Guide values for indoor air
There is no binding limit for PCP or lindane in indoor air in Austria; German guide values are used for assessment. For PCP the German PCP-Richtlinie names two levels: 0.1 µg per cubic metre of room air as a precautionary and remediation-target value, and 1 µg per cubic metre as the value above which remediation is required (each as an annual mean). These are precautionary and action values, not toxicological thresholds below which nothing happens. They give a measurement its yardstick: only the measured concentration in room air and house dust shows whether and how urgently to act.
How the contamination is determined
It begins with a sample: wood shavings from the suspect component, supplemented by a house-dust sample from the occupied rooms and, where appropriate, a room-air sample, examined in an accredited laboratory for PCP, lindane and related substances. In all, four sample types come into consideration; they answer different questions and are not interchangeable: the wood says whether and with what it was treated; the house dust shows the exposure-relevant reservoir; room air measures the current inhalation exposure; a blood or urine sample captures the amount actually taken up. The dust sample from the living area is especially informative because it comes closest to real intake, precisely where children live. Self-sampling is inadvisable, because improper handling tends to spread the contamination rather than clarify it; sampling, like the later remediation, should be carried out professionally. How the laboratory measures and what the limits are is in the box.
For experts, or those who want to become one: Sample types, measurement methods, detection limits and the limits of the analytics
What each sample answers. A material sample of the wood (result in mg per kg) clarifies whether a component was treated; it cannot be converted into a statement about room air or health (UBA-HBM 1997). House dust is the time-integrating, exposure-relevant reservoir and the most informative sample for small children; for it the AGÖF gives statistical orientation values (not health values) (AGÖF 2007). Room air is the snapshot the PCP-Richtlinie is written against. Biomonitoring in blood or urine captures the internal dose across all uptake routes; air concentration and internal dose, however, correlate only weakly, which is why the four levels complement rather than replace one another (UBA-HBM 1997).
Methods. After extraction and clean-up, the substances are separated by gas chromatography and determined by mass spectrometry (GC/MS) or the electron-capture detector (GC/ECD), which is very sensitive to organochlorine compounds (VDI 4301 Blatt 2). PCP must be derivatised for this, usually by acetylation, because the free phenol is too polar for gas chromatography; lindane, as a neutral organochlorine compound, is measured directly. Room air is sampled actively, pumped through a collector of glass-fibre filter and polyurethane foam to capture the gas and particle phases (DGUV/IFA). The limit of quantification for PCP in room air is about 0.05 µg/m³, for house dust in the region of 0.1 mg/kg.
Regulatory basis. The measurement strategy for PCP and lindane indoors is governed by VDI 4300 Blatt 4, the measurement method by VDI 4301 Blatt 2; the assessment follows the PCP-Richtlinie (annual mean). For the internal dose, Germany has health-based HBM values (for instance HBM-I 40 µg/l and HBM-II 70 µg/l PCP in serum) and statistical reference values for the background burden; these value types must not be confused. An accredited laboratory (to ISO/IEC 17025) is decisive for reliable results, since interlaboratory trials show considerable scatter.
Limits. A wood result cannot be converted into an air exposure; air and internal dose correlate only weakly; the air concentration varies with temperature (higher on warm days), which is why the guide values refer to the annual mean; a single air sample cannot rule out contamination. House dust is more reproducible than a single air measurement, which makes it the robust starting point.
The legal framework
PCP has been banned in Austria since 1991 (Ordinance BGBl. No. 58/1991, based on the Chemicals Act); lindane was abandoned as a wood-preservative active ingredient and is no longer authorised today. Both are regulated internationally as persistent organic pollutants (POP) under the Stockholm Convention, implemented in the EU through the POP Regulation (EU) 2019/1021, Annex I (POP Regulation 2019). PCP was additionally subject to a REACH restriction (Annex XVII), which was repealed in early 2021 upon its transfer into the stricter POP regime (Regulation (EU) 2020/2096). Treated wood that is removed is, depending on the level of contamination, hazardous waste; in the European waste catalogue it carries the code 17 02 04* (glass, plastic and wood containing dangerous substances), in Austria the corresponding key number of the Waste Catalogue Ordinance (AVV 2020).
What to do
- Suspicion, but no measurement: the most important step is clarification, and with an eye to the occupied rooms, not just the attic. Where small children live in the household, the house-dust sample in the living area takes priority.
- Loft conversion planned: have it measured before the conversion. Contamination changes the approach and the costs considerably, and the conversion itself releases bound substances.
- Contamination confirmed: the most reliable solution is to remove the treated wood. Where that is not possible, planing off the surface or sealing reduces the off-gassing without ending it entirely, accompanied by good ventilation and the removal of the contaminated dust. Which measure is effective depends on the findings.
Wood preservatives are one of several invisible legacies in the attic and old building. In buildings of the same period it is also worth looking at asbestos and old mineral wool insulation.
Sources
- ATSDR (2022): Toxicological Profile for Pentachlorophenol. Agency for Toxic Substances and Disease Registry, U.S. Department of Health and Human Services. Purity of technical PCP (86 to 90 %) and PCDD/PCDF impurity; uncoupling of oxidative phosphorylation; liver damage; chloracne and immune effects as effects of the impurities. ncbi.nlm.nih.gov/books/NBK590407
- ATSDR (2024): Toxicological Profile for Hexachlorocyclohexanes (HCH). Agency for Toxic Substances and Disease Registry. β-HCH as the most persistent isomer; accumulation in fatty tissue; lindane neurotoxicity from medicinal use. atsdr.cdc.gov
- AGÖF (2007): Vorläufige AGÖF-Orientierungswerte für mittel- und schwerflüchtige organische Verbindungen im Hausstaub (provisional AGÖF orientation values for semi-volatile organic compounds in house dust), as of autumn 2007 (with one amendment in 2022). Arbeitsgemeinschaft ökologischer Forschungsinstitute. Statistically derived percentile values from the analysis data of AGÖF institutes (not health-based values); includes PCP and lindane. agoef.de
- AVV (2020): Waste Catalogue Ordinance (Abfallverzeichnisverordnung), BGBl. II No. 409/2020. Treated wood containing dangerous substances; EU waste code 17 02 04*. ris.bka.gv.at
- BGH (1995): Federal Court of Justice, judgment of 2 August 1995 (2 StR 221/94), wood-preservative proceedings; the Frankfurt judgment overturned on a procedural ground. Official reports BGHSt 41, 206. bundesgerichtshof.de
- Collins et al. (2009): Mortality Rates Among Workers Exposed to Dioxins in the Manufacture of Pentachlorophenol. Journal of Occupational & Environmental Medicine 51(10):1212–1219. Elevated non-Hodgkin lymphoma mortality in the PCP-manufacture cohort (Dow; conflict of interest disclosed). doi.org/10.1097/JOM.0b013e3181badd4e
- Demers et al. (2006): Cancer and Occupational Exposure to Pentachlorophenol and Tetrachlorophenol (Canada). Cancer Causes & Control 17(6):749–758. Non-Hodgkin lymphoma trend with cumulative PCP, not TCP, exposure in a sawmill cohort. doi.org/10.1007/s10552-006-0007-9
- DGUV/IFA: Institute for Occupational Safety and Health of the German Social Accident Insurance, method for determining PCP and lindane (active sampling on glass-fibre filter/polyurethane foam; staged approach: wood, dust, air, biomonitoring). dguv.de/ifa
- Hardell et al. (2002): Exposure to Pesticides as Risk Factor for Non-Hodgkin's Lymphoma and Hairy Cell Leukemia: Pooled Analysis of Two Swedish Case-control Studies. Leukemia & Lymphoma 43(5):1043–1049. Elevated lymphoma risk with chlorophenol exposure. doi.org/10.1080/10428190290021560
- IARC (2012): Chemical Agents and Related Occupations, Vol. 100F; 2,3,7,8-tetrachlorodibenzo-para-dioxin as carcinogenic to humans (Group 1). ncbi.nlm.nih.gov/books/NBK304398
- IARC (2015): DDT, Lindane, and 2,4-D. IARC Monographs, Vol. 113 (evaluation 2015). Lindane as carcinogenic to humans (Group 1), basis non-Hodgkin lymphoma. publications.iarc.fr
- IARC (2019): Pentachlorophenol and Some Related Compounds. IARC Monographs, Vol. 117. Pentachlorophenol Group 1, basis non-Hodgkin lymphoma; carcinogenic effect attributed to the PCP itself. ncbi.nlm.nih.gov/books/NBK543343
- IPCS (1996): Lindane (gamma-HCH). Poisons Information Monograph PIM 859, International Programme on Chemical Safety, WHO. GABA-A action; neurotoxicity; lipophilicity and accumulation in fatty tissue. inchem.org
- Kogevinas et al. (1995): Soft Tissue Sarcoma and Non-Hodgkin's Lymphoma in Workers Exposed to Phenoxy Herbicides, Chlorophenols, and Dioxins. Epidemiology 6(4):396–402. Nested case-control study within the international IARC register. doi.org/10.1097/00001648-199507000-00012
- PCP-Richtlinie: Guideline for the assessment and remediation of pentachlorophenol-contaminated building materials and components in buildings, German Bauministerkonferenz (ARGEBAU), 1996 version with 1997 correction. Indoor guide values 0.1 µg/m³ (precaution/remediation target) and 1 µg/m³ (remediation required), annual mean.
- POP Regulation (2019): Regulation (EU) 2019/1021 on persistent organic pollutants, Annex I (PCP and lindane/HCH listed), implementing the Stockholm Convention. eur-lex.europa.eu
- Purdue et al. (2007): Occupational exposure to organochlorine insecticides and cancer incidence in the Agricultural Health Study. International Journal of Cancer 120(3):642–649. Dose-response relationship of lindane and non-Hodgkin lymphoma in a prospective cohort. doi.org/10.1002/ijc.22258
- Ruder & Yiin (2011): Mortality of US pentachlorophenol production workers through 2005. Chemosphere 83(6):851–861. Elevated non-Hodgkin lymphoma mortality in the NIOSH cohort. doi.org/10.1016/j.chemosphere.2011.02.064
- Schinasi & Leon (2014): Non-Hodgkin Lymphoma and Occupational Exposure to Agricultural Pesticide Chemical Groups and Active Ingredients: A Systematic Review and Meta-Analysis. International Journal of Environmental Research and Public Health 11(4):4449–4527. Summary relative risk for lindane about 1.6. doi.org/10.3390/ijerph110404449
- UBA (2008): German Environmental Survey for Children 2003/06 (KUS), base report "Hausstaub" (house dust): substance levels in house dust from households with children in Germany (Müssig-Zufika et al., WaBoLu-Hefte 02/08, German Environment Agency / Umweltbundesamt), together with the accompanying UBA press release No. 12/2008 of 22 February 2008. House dust as indicator and reservoir of semi- and non-volatile substances, PCP quantifiable in 83 percent of house-dust samples, uptake by children via typical hand-to-mouth behaviour. umweltbundesamt.de
- UBA-ESB: German Environmental Specimen Bank (Umweltprobenbank des Bundes), time series of pentachlorophenol in blood plasma; marked decline after the ban. umweltprobenbank.de
- UBA-HBM (1997): Human Biomonitoring Commission of the German Umweltbundesamt: Stoffmonographie Pentachlorphenol, Referenz- und Human-Biomonitoring-Werte (HBM). Bundesgesundheitsblatt 40(6):212–222. Wood material values cannot be converted into room-air or health statements; no good correlation between air concentration and internal dose. doi.org/10.1007/BF03042913
- VDI 4300 Blatt 4 / VDI 4301 Blatt 2: Association of German Engineers, measurement of indoor-air pollutants: measurement strategy and measurement method for PCP and lindane (GC/MS and GC/ECD). vdi.de
- Vijgen et al. (2011): Hexachlorocyclohexane (HCH) as new Stockholm Convention POPs: a global perspective on the management of Lindane and its waste isomers. Environmental Science and Pollution Research 18(2):152–162. Isomer ratio, 8 to 12 t waste isomers per t lindane, β-HCH persistence. doi.org/10.1007/s11356-010-0417-9
- WHO (1987): Pentachlorophenol. Environmental Health Criteria 71, International Programme on Chemical Safety, World Health Organization, Geneva. Formation of PCDD/PCDF from PCP salts on heating. inchem.org
Questions on the topic? → Contact