| Forside | | Indhold | | Forrige | | Næste |

Sundhedsmæssig vurdering af PCB-holdige bygningsfuger

Summary and conclusions

• Perspectives

ABSTRACT

PCB has been used as a plasticizer in building sealants in the period from around 1950, and until it was prohibited in Denmark in 1976.  A questionnaire survey, including 100 buildings, and a detailed chemical analysis of PCB in old sealant materials, indoor surface dust and indoor air in 10 buildings were performed in order to estimate the remaining mass of PCB in old sealants and to estimate the resulting human PCB exposure in buildings and its potential health effects. The total remaining mass of PCB in sealants in Danish buildings was estimated to be in the range of 6-21 ton. Concentrations in indoor air ranged from below 30 to just above 1000 ng/m³. Concentrations in indoor surface dust ranged from below 30 to just above 2000 ng/g. The toxicity of PCBs varies significantly between congeners. Detailed analysis of congener mixture and congener toxicity is required to evaluate health impact of exposures in buildings. The PCBs measured in indoor air were mainly the lower chlorinated, non-dioxin-like congeners. The highest concentration measured was estimated to result in a daily exposure of about 70 times below the no observed adverse effect level (NOAEL) in experimental animals. The congener composition in indoor dust resembled more commercial mixtures. The highest measured concentration was estimated to result in an exposure that was 3500 times below the lowest observed adverse effect level (LOAEL) from animal tests.

INTRODUCTION

PCB is a group of organochlorine compounds with two interconnected benzene rings that can contain as many as 10 chlorine atoms. The 209 PCB-congeners have different toxicological and physical-chemical properties depending on the number of chlorine atoms and the substitution pattern. The mixture of the various congeners differed in the different commercial PCB-mixtures.

Most PCB-congeners with low chlorine content are metabolized and excreted rather quickly in animals and humans, while the congeners with high chlorine content are very persistent, lipophilic and bioaccumulates in fat tissues and biomagnifies through natural food chains. The congener composition of the PCB exposure from food intake and from indoor climate differs considerably.

The toxicity of PCBs differs also significantly between congeners. The most toxic are the so-called dioxin-like PCBs (van den Berg et al. 1998). But the health effects of other bio-accumulating congeners should also be considered. Health effects of PCB include effects on skin, liver, thyroid gland, reproductive organs, central nervous system and immune system. Furthermore, PCB may cause cancer and reduce fertility. The main source of PCB exposure in the general population is food intake, especially fish with high fat content from certain polluted waters. The PCB content is particularly high in breast milk, and consequently it is of principal interest to reduce other intake routes for infants (WHO 2003).

PCB possesses attractive physical properties, such as good electrical insulation, high thermal stability, high viscosity and fire resistance. Furthermore, PCB has been used to make sealants stable, soft and flexible. Until the middle of the seventies PCB was frequently used in a number of building construction products, including sealant and glue in double-glazed units used in energy saving windows as sealant between window frames and walls, and as sealant between light and heavy elements of walls.

All uses of PCB have been banned in Denmark since 1976. Caulking and sealants are suspected of being the biggest remaining man-made deposits of PCB in. Since the 1970s replacements in connection with renovation works have reduced the size of this deposit. Emissions during the past 30 years are expected to have further reduced the size of the deposits and to have polluted the structures and soils around the old joints filled with PCB-containing compounds.

The purpose of the present study was

• To quantify the amount of PCB still contained in joints in buildings,
• To assess the contamination of indoor air and dusts in buildings with PCB-containing materials,
• To evaluate potential health risks associated with this contamination.

METHOD

The data gathering comprised

1. A questionnaire survey including approximately 100 houses, and
2. Chemical sampling and analysis of old sealant materials, indoor air and indoor surface dust in 10 of these houses.

Based on address lists supplied by local authorities in the City of Copenhagen, a questionnaire was mailed to building administrators for offices, schools, childcare institutions and apartment blocs. Furthermore, the questionnaire was mailed to some owners of single-family houses. Approximately 600 questionnaires were sent out and 100 were filled in and returned.

The questionnaire contained information about construction year of the building, area size of the building, length of inner joints, length of outer joints and appearance of joint fillers and sealants. The selection of addresses was not made with an attempt to assure representativeness, and the response rate of 17% was low but systematic bias was not expected, since the type of sealant was not expected to influence inclusion criteria or response rate.

Ten buildings were selected for more detailed assessments.  Based on information found in the questionnaire replies, all buildings were suspected of having PCB-containing sealants.

A cylindrical sealant sample sized 5mm x 10mm was taken by driving a sharp edged tube with a diameter of 5mm into the sealant. If possible, this was done both at outer and inner joints. Immediately after sampling the tube with sealant was wrapped in aluminum foil and placed in sealed plastic bags until chemical analysis could be made.

A specially made filter cassette was mounted on a traditional vacuum cleaner (1600W) for dust sampling. Dust was collected on circular glass fiber filters with a diameter of 80mm. The suction opening was 10mm x 60mm, and by small wheels the opening was kept at a distance of 5mm from the sampling surface. A floor area of 2m² was sampled during 2 minutes near the place, where the joint sample was taken. The filters were immediately wrapped in aluminum foil and placed in sealed plastic bags until chemical analysis could be made.

Duplicate indoor air samples were taken by passing an air flow of 1 L/min through XAD-2 absorption tubes. The sampling time was approximately 17 hours, giving a sampling volume around 1m³. The tubes were capped and placed in sealed plastic bags until chemical analysis could be made.

Approximately 3 months after the first sampling, when all samples had been taken, they were analyzed in one batch. Liquid extraction and gas chromatography were used to quantify the following 22 PCB-congeners 28, 31, 44, 49, 52, 99, 101, 105, 110, 118, 128, 138, 149, 151, 153, 156, 170, 180, 187, 188, 194 and 209.

RESULTS

Based on analyses of questionnaire replies, the main results from chemical analysis and some statistical information from the building registration it was possible to estimate the amount of PCB still contained in old sealants. In the period from around 1950, when PCB-containing sealants were introduced, and until 1976, when PCB-containing sealants were prohibited, 37% of the building stocks in Denmark were constructed. This building mass equals a total floor area of 143 mio. m². The remaining PCB can be calculated by multiplying the relevant building area by area-specific length of sealed joints with an estimate of the length-specific weight of sealant. The total weight of sealants should be reduced only including the fractions without major joint filler replacement, with rubber-like sealants and actually containing PCB of the rubber-like sealants from the relevant period. Finally this reduced mass of sealants should be multiplied by the average total concentration of PCB obtained from the chemical analysis to give a very rough estimate of total remaining mass. The main results of the questionnaire survey are summarized in Table E1 together with a general summary of results from the chemical analysis.

Table E1. Main results of the questionnaire survey based on 100 replies and summary of average results from chemical analysis of 7 rubber-like sealants from the 4 buildings with PCB-containing sealants.

The very uncertain estimate of remaining sum of the analyzed PCB congeners in Danish buildings becomes 5600 kg. An alternative to this figure may be found by multiplying the sum of the 7 most commonly analyzed congeners by a factor 5 to account for the many congeners not included in the analysis. This would result in an estimate of 13000 kg. Still a larger estimate may be found by applying the highest factor found in the literature equal to 8. This high correction factor results in a total remaining mass of PCB in Danish sealants of 21000 kg.

Details about PCB content in joint sealants are given in Table E2, contamination of indoor surface dust are given in Table E3 and the concentrations of PCB in indoor air are given in Table E4.

Table E2. PCB content (µg/g) in indoor joint sealants ( I ), outdoor joint sealants ( O ) and sealant in double-glazed units in windows ( W ). ∑₇ is the sum of congeners number 28, 52, 101, 118, 138, 153 and 180. ∑_n is the sum of all identified congeners of the 22 that were analysed for. The counter n is the number of congeners that were measured above the reporting limit of 0.5µg/g. ND means that all congeners were below reporting limit.

Results in Table E2 indicate that inner sealants in Buildings 6, 7, 9 and 10 contain PCB, while inner sealant in Building 5 only contains traces of PCB. Inner rubber-like sealants were not found in the remaining buildings. The PCB content is much lower than in the original commercial mixtures with the PCB content ranging from 5 to 30%

Table E3. PCB contamination (ng/g) of indoor surface dust in the different buildings ( D ). Reporting limit was 30ng/g. Other denominations are like Table 2.

PCB contents in indoor dust shown in Table E3 show some PCB in all buildings except Building 5. Highest concentrations were found in Buildings 6, 7, 9 and 10; that were also the buildings with significant content of PCB in inner sealants.

Table E4. The two measured concentrations of PCB (ng/m³) in indoor air in the different buildings ( A ). Reporting limit was 30ng/ m³. Other denominations are like Table 2.

The values for PCB in indoor air shown in Table E4 also give the highest concentrations in the buildings with significant content of PCB in inner sealants. The two measurements of PCB in indoor air are in good agreement with each other except in Building 6 where a minor discrepancy was seen. No explanation for this has been found.

Figure E1. Relation between total PCB in indoor air and inner sealant.

Figure E. Relation between total PCB in dust and inner sealant.

To show these relations visually, Figure E1 presents PCB concentration in indoor air in relation to the concentration in inner sealant. Figure E2 presents PCB in dust in relation to PCB in inner sealant. Finally Figure E3 presents the relation between PCB in dust in relation to PCB in indoor air.

Figure E3. Relation between total PCB in dust and indoor air.

All three figures seem to indicate clear interrelation between PCB content in the different compartments.

DISCUSSION

There are a number of other possible sources for PCB contamination of indoor environments. However the clear relations between PCB content in inner sealants, indoor air and indoor surface dust indicate that the inner sealant is a major source of PCB content in these compartments.

The detailed analysis of indoor air showed as expected that the PCB-congeners determined were mainly non-dioxin like congeners. If a person with a body mass of 60 kg daily inhales 15 m³/day of air with a PCB concentration of 1 µg/m³ (approximately equal to the maximal concentration found in this study), then the resulting exposure is 15 µg/person or 250ng/kg body mass/day. This is approximately 100 times below the no observed adverse effect level (NOAEL) of 30000-40000ng/kg body weight/day for effects on liver and thyroid gland found in 90-days oral toxicity  tests for three non-dioxin like PCB with rats (ATSDR 2000, Bilag B).

Intake of PCB with indoor dust was assessed assuming a daily intake of 50mg dust and the highest concentration found in this study of 2 µg/g dust. The resulting exposure is 0.1 µg/person or 2 ng/kg body mass/day. The detailed analysis showed congener composition like the original commercial mixtures. For these mixtures animal tests have showed a lowest observed adverse effect (LOAEL) level of 5 µg/kg body mass/day (ATSDR 2000). The LOAEL is more than 3500 times higher than the estimated highest daily intake of indoor dust.

The selected buildings showed PCB concentrations in indoor air in the range from below 30 to just above 1000 ng/m³. This large difference makes it very probable that buildings with even higher indoor air concentrations may be found. A combination of low air change rates and large amounts of sealant with high PCB content may increase the contamination.

CONCLUSIONS

Based on the data from this limited investigation, the total remaining mass of PCB in joint fillers in Danish buildings can be estimated with great uncertainty to be between 6 and 21 ton.

The investigation shows relations between concentrations of PCB in sealants, indoor air and surface dust. Furthermore, PCB concentrations in the soil around the buildings seem to be somewhat related to the concentrations in outer sealants.

In and around those buildings that still have PCB-containing sealants the sealants will still be a major source of contamination of indoor air, surface dust and the soil around the buildings.

PCB in sealants may only have a small contribution to the building users’ exposure to the most toxic dioxin-like PCB-congeners. That exposure is primary due to intake of contaminated food.

Results from this limited investigation indicate that the most important contribution to the exposure to the most volatile non-dioxin-like PCB-congeners will be PCB in indoor air of buildings with PCB-containing sealants. The levels in indoor air in these buildings may results in an unwanted reduction of the margin of safety.

Data also point to a risk that there in Denmark may be buildings with a far higher PCB-content in the sealants, and exposure to indoor air during several years in these buildings may pose a risk of serious health effects.

Perspectives

This investigation shows that PCB may be contained in joints between window frames and surrounding structures, and that PCB may be emitted to the surroundings. However, this limited study does, not give support to more comprehensive conclusions regarding possible health risks because of the remaining mass of PCB in Danish buildings.

If the aim is a more certain assessment of PCB in Danish buildings then a more comprehensive investigation is required. This may comprise of sealants samples from each of some hundred Danish buildings from the relevant period. Most of the samples in this investigation were taken from window joints. If a new investigation is performed, it is recommended to include glue from double glazing and other construction materials with possible PCB-contamination.

It is rather expensive to analyse samples for PCB-content. Therefore there is a need for more economic and faster methods to identify PCB-containing materials.

Occupational exposure in relation to the renovation of PCB containing materials is not assessed in this report. There is, however, a need for knowledge about required means to reduce exposure during renovation works.

Based on information from the Danish handlers of dangerous waste it may be assumed that much Danish construction wastes containing PCB are not handled according to legislation. There is, therefore, a need for dissemination of knowledge concerning requirements, problems and options in relation to identification and handling of waste with a possible PCB-content among the directly involved.

REFERENCES (for this summary – more in the report in Danish)

Van den Berg M, Birnbaum L, Bosveld ATC, Brunström B, Cook P, Feeley M, Giesy JP, Hanberg A, Hasegawa R, Kennedy SW, Kubiak T, Larsen JC, van Leeuwen FXR, Liem AKD, Nolt C, Peterson RE, Poellinger L, Safe S, Schrenck D, Tillitt D, Tysklind M, Younes M, Wærn F and Zacharewski T (1998). Toxic Equivalency Factors (TEFs) for PCBs, PCDDs, PCDFs for Humans and for Wildlife. Environ Health Perspect 106, 775-792.

WHO (2003). Polychlorinated biphenyls: human health aspects. Concise International Chemical Assessment Document 55. http://www.inchem.org/documents/cicads/cicads/cicad55.htm

ATSDR (2000). Toxicological Profile for Polychlorinated Biphenyls (Update). U.S. Department of Health & Human Services, Public Health Service, Agency for Toxic Substances and Disease Registry.

| Forside | | Indhold | | Forrige | | Næste | | Top |

Version 1.0 Marts 2009, © Miljøstyrelsen.