Annals of Occupational Hygiene Advance Access originally published online on May 17, 2004
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Ann. occup. Hyg., Vol. 48, No. 5, pp. 425-437, 2004
© 2004 British Occupational Hygiene Society
Published by Oxford University Press
Occupational Exposure to Dioxins at UK Worksites
1 Environmental Science Department, Lancaster University, Lancaster LA1 4YQ, UK; 2 Occupational Hygiene Section, Health and Safety Laboratory, Sheffield S3 7HQ, UK; 3 Health and Safety Executive, 1 Hagley Road, Birmingham B16 8HS, UK; 4 Health and Safety Executive, Magdalen House, Stanley Precinct, Bootle, Merseyside L20 3QZ, UK
Received 23 June 2003; in final form 18 November 2003; published online on 7 July 2003
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONSELECTION OF SITES FOR...MATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Following a request from a Governmental Interdepartmental Group, the Health and Safety Executive undertook a polychlorinated dibenzo-p-dioxin (PCDD) and polychorinated dibenzofuran (PCDF) sampling exercise at several work sites in the UK. An initial survey suggested potential PCDD/F production at metal recycling sites, during cement manufacture, at municipal waste incinerators and landfill sites and during the use of thermal oxygen lances. PCDD/F sampling, using static and personal air samplers, revealed that the highest PCDD/F exposures were found at metal recycling sites, particularly aluminium recycling sites. The reasons for these results and the possible consequential intakes are discussed.
Keywords: dioxins; metal processing; occupational exposure; PCDD/F
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONSELECTION OF SITES FOR...MATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Polychlorinated dibenzo-p-dioxins (PCDDs) and polychorinated dibenzofurans (PCDFs) are two large families of chemicals that can be formed as a result of incomplete combustion of organic material. There are a possible 75 PCDD and 135 PCDF congeners (dibenzodioxins and dibenzofurans with up to eight chlorine atoms), but of these, only 17 are considered to be biologically active. These are the 17 congeners chlorinated at the 2, 3, 7 and 8 positions. Considerable concern has been expressed about possible exposure of the UK population to these chemicals. Figure 1 shows the structure of two congeners with most activity shown by the dioxin congener chlorinated at the 2, 3, 7 and 8 positions.
|
The health effects of the 17 biologically active PCDD and PCDF congeners have been described in a statement published by the UK Committee on Toxicity of Chemicals in Food, Consumer Products and the Environment (COT) (COT, 2001). Each of these 17 congeners produces a similar spectrum of health effects, but their potency varies widely. Most, but not all, of the biological effects are due to binding at the aryl or Ah-hydrocarbon receptor.
In addtion to the variations in potency between congeners, there is a wide variation in susceptibility to individual congeners between different species and strains of animals. A more than 8000-fold difference in LD50 values for 2,3,7,8-tetrachlordibenzodioxin (TCDD) exists between the most susceptible and the least susceptible species/strain (EPA, 2000). No single site of toxicity has been identified as the cause of lethality, but a wasting syndrome and hepatotoxicity are commonly observed in animals. Humans seem to be less susceptible; the effect in humans most clearly linked with high dioxin exposure is a skin condition called chloracne. Long-term exposures have produced a range of effects in experimental animals, including immunosuppression and endocrine disruption, but evidence that these occur in humans is uncertain. In relation to carcinogenicity, the UK Committee on Carcinogenicity (COC) concluded in 2001 that ‘TCDD should be regarded as a probable human carcinogen’ (COC, 2001). It was also noted that ‘a threshold approach to risk assessment was likely to be appropriate’, meaning that there may be an exposure level at which there is no increased risk of cancer above background rates. Although COC could not state what this exposure might be, it noted that ‘the excess cancer mortality reported in the heavily-exposed industrial cohorts was small’ and that ‘any increased risk of cancer at background levels of exposure is likely to be extremely small and not detectable by current epidemiological methods’.
There are also concerns in relation to the potential reproductive effects of PCDD/Fs. Studies have been conducted to examine effects in men and women exposed as a result of the Seveso incident, but effects in this population occurred as a result of a peak exposure incident and it is not clear if they are relevant to individuals who receive steady exposure to background levels. Other studies conducted on cohorts in Rotterdam, Gronigen and Amsterdam have suggested an effect of prenatal dioxin exposure background levels on cognitive development. However, additional follow-up work is needed before any clear conclusons can be drawn. There is also evidence, from studies in animals, for adverse reproductive effects as a result of exposure to TCDD in utero. The most sensitive effects were on a number of sperm parameters in rats exposed in utero. These changes reflected functional adverse reproductive effects that have been seen at higher dose levels in a long-term multigeneration reproductive toxicity study in rats. COT considered that these findings were of concern in relation to human health. The recommended UK tolerable daily intake (TDI) of 2 pg WHO toxic equivalent (TEQ)/kg/day is based on this data.
Owing to the considerable variation in the toxicological potency between congeners, toxic equivalency factors (TEFs) that express the potency of each congener relative to the most biologically active congener (TCDD) have been assigned to the 17 biologically active congeners. This enables the overall toxicological potency of any mixture to be expressed in terms of an equivalent dose of TCDD (TEQ).
For the UK COT recommends use of the TEF scheme proposed by the World Health Organisation (WHO) in 1997 (WHO TEFs). In addition to the biologically active dioxin congeners, 12 polychlorinated biphenyls (PCBs) have been identifed that exhibit ‘dioxin-like activity’. These have also been assigned WHO TEFs to rank their biological activity alongside TCDD (Van den Berg et al., 1998). ‘Dioxin-like’ PCBs have a much lower TEF than the active PCDD/Fs.
As part of a UK Government position paper on PCDD/Fs and PCBs that exhibit ‘dioxin-like’ activity, an Interdepartmental Group was asked to provide information on the nature and extent of exposures and intakes of the UK population to these chemicals. This paper was to summarize the current status of PCDD/Fs, to act as a vehicle for a continuing strategy and to recommend future actions to continue the decline in exposure to these substances.
There is a lot of information on PCDD/F intake from non-work activities, with dietary sources assumed to contribute about 95% of the average intake, but there is very little information on exposure to PCDD/Fs from work activities. In order to provide current information on these occupational exposures, the Health and Safety Executive (HSE) commissioned a sampling exercise in certain industry sectors in order to discover whether there was the potential for extra intake of PCDD/Fs in certain industrial sectors that might be significant. High volume static samplers have been used for environmental studies of PCDD/Fs and these were available for use on suitable sites. Recently, workers in Germany have adapted personal samplers to measure PCDD/F exposure from industrial processes (Menzel et al., 1998), so a combination of both samplers was arranged where possible.
The sampling exercise was targeted at sites where the potential for PCDD/F formation was considered highest. The number of samples in this survey was low owing to the complicated and lengthy sampling and analysis procedures for PCDD/Fs, but employee numbers on each site were also generally low and this allowed sampling of most ‘at risk’ workers. Consequentially, the results concentrate on those employees deemed to have the highest potential for PCDD/F exposure and do not represent mean or median exposures for the site as a whole. All personal results represent a variety of tasks, but sampling times usually exceeded 4 h and were designed to represent the full shift. Static samples were taken from areas of peak potential PCDD/F production and represent worst case background exposures.
Only the dioxin and furan congeners were analysed as these are the overwhelming contributors to the toxicity profile. We report below the results of this sampling exercise.
|
SELECTION OF SITES FOR SAMPLING |
|---|
|
TOPABSTRACTINTRODUCTIONSELECTION OF SITES FOR...MATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Industry sectors were chosen for sampling which had the most suitable conditions for PCDD/F formation on the basis of current knowledge.
Suitable conditions for dioxin formation
PCDDs and PCDFs may be produced when organic material is heated in the presence of chlorine, a process catalysed by metals. Numerous experiments have demonstrated that temperatures above 250°C are required for de novo synthesis of PCDD/Fs from organic material (Dickson and Karasek, 1987; Stieglitz and Vogg, 1987; Cains et al., 1997). Peak temperatures for PCDD/F formation in the presence of finely divided ash are around 325°C (Milligan and Altwicker, 1995). PCDD/Fs are formed below 700°C, but destruction starts at 550°C and is complete at temperatures above 1000°C (Menad et al., 1998). However, it is the cooling processes that are a source of much of the PCDD/F in flue gases (Buekens et al., 1998; Chagger et al., 2000), as PCDD/Fs can reform if the precursors are still present. Organic material is essential for PCDD/F formation and carbon remaining in the ash from combustion provides a platform for PCDD/F formation (Groschwitz and Sommer, 2000).
Chlorine is also essential for PCDD/F formation; salts such as copper chloride are efficient chlorine donors (Gullet et al., 2000; Addink and Altwicker, 2001) where organic chlorine is low (Addink et al., 1998). Any associated heavy metals may catalyse the reaction (Stieglitz and Vogg, 1987; Olie et al., 1998), so sites where metals are melted in the presence of organic material were included in the sampling exercise.
Many combustion processes have some or all of the characteristics required for PCDD/F formation and all industry sectors where organic material is burnt at temperatures above 250°C were considered for sampling. The following sectors were included in the sampling exercise.
Metal recycling
Recycling of aluminium, iron, magnesium and zinc involves melting the scrap and casting to form new ingots. Varying quantities of organic material are incorporated into the heating process and the added fluxes may provide a chlorine source.
Metal reclamation plants are known dioxin emitters and the presence of PVC and other contaminants provides a source of organic material and chlorine (Aittola et al., 1993; Menad et al., 1998). Aluminium, copper, zinc and magnesium recycling is performed at lower temperatures than steel production and although copper is no longer recycled in the UK, various copper alloys are used as feedstock. Sampling was therefore undertaken at a range of metal recycling sites.
Aluminium recyclers
Recycling of aluminium can be divided into two sorts depending on the scrap used as feedstock. New scrap generated during production and fabrication is usually totally recycled and this is termed remelting. Old scrap, recovered from articles at the end of their useful life, such as beverage cans, scrap from motor vehicles and aluminium windows, is recycled where it is economically profitable, and this is termed refining (Aluminium Federation, 1998). Old scrap needs more pretreatment and is more likely to be contaminated with organic material. Most scrap is sorted and may be shredded before being baled or charged directly into a melting furnace. Preliminary cleaning should remove oils and organic coatings by heating and drying in rotary driers at temperatures high enough to vaporize or carbonize organic materials but below the melting point of aluminium (660°C) or its alloys.
Melting technology varies and depends on the scrap and the product required. Remelters use more reverbatory furnaces while refiners use reverbatory, rotary, induction and sloping hearth furnaces. Furnaces operate at temperatures between 700 and 750°C. Rotary furnaces are generally used for lower grades of scrap and a salt flux is added. The heat transfer is by conduction as the charge maintains contact with the rotating refractory shell. Impurities are trapped in the upper layer of flux and molten aluminium forms a liquid pool beneath. In the standard reverbatory melting furnace, heat from ignited fuel is reflected back from the curved roof and into the melted charge which circulates inside the furnace. Sloping hearth furnaces are used for the most contaminated scrap and these furnaces operate at temperatures above the melting point of aluminium but below the melting point of contaminants such as iron. The molten aluminium trickles down the hearth into a holding bath. Induction furnaces generate heat by electromagnetic induction where a primary alternating current flows through a coil. They are usually used for clean scrap with a low surface area such as foil and turnings.
Aluminium oxidizes readily and the aluminium oxide floats to the top to form a dross layer that has entrapped metallic aluminium within it plus constituents from the fluxes. This dross is readily skimmed off and is cooled quickly (to minimize further oxidation) and may be pressed in a dross press. Milling of the cold dross produces a powder/globule mixture than can be sieved to recover the metallic part and this can be remelted (usually in a rotary furnace with a salt flux).
Ferrous metal production
This was one industry sector where PCDD/F exposure data were available.
There are two main routes to steel production that are used worldwide: the integrated process route involving the linked processes of sintering, coke making, blast furnace and basic oxygen steel making and electric arc steel making (EAF).
In the integrated route, the only noteworthy source of dioxins is iron ore sintering (Fisher et al., 1998; Alcock et al., 1999; Anderson and Fisher, 2002). Data from the industry indicate that dioxin concentrations in workplace air were typically 0.5 pg/m3 (D.R.Anderson, personal communication, 2001), similar to UK ambient air concentrations in urban areas. The recycling of organic-coated steel is likely to increase in the future, but recent studies (Anderson and Fisher, 2002) indicate that recycling of organic-coated steel in basic oxygen steelmaking had no discernable effect on dioxin emissions.
For EAF, fugitive emission measurements from the roof vents of the melting shop showed higher values (up to 8 pg/m3; D.R.Anderson, personal communication, 2001) but may represent maximum potential concentrations. An EAF site was chosen for the sampling exercise to confirm these results.
Zinc smelting
Hot coke and lump sinter (made from burning raw sulphides and oxides of zinc, lead and cadmium) are added to the top of a furnace and burnt. Hot air and zinc oxides are added at the tuyere injection into the bottom. The zinc floats off and is further refined. The lead bullion is tapped off at the bottom. The scrap zinc oxide dusts injected at the bottom of the furnace have been shown to contain relatively high levels of PCDD/Fs (between 4000 and 5200 ng/kg from two recent measurements made by the company). Although the high temperatures of the furnace (over 1000°C) should destroy the PCDD/Fs, reformation may be possible later.
Magnesium recycling
The processes in the magnesium recycling industry mirror those in the aluminium recyling industry, but the scrap used is usually considerably cleaner. Temperatures in the furnace are around 675°C, low enough for PCDD/F formation, and a suitable site was chosen for sampling.
Cement manufacture
Cement manufacture involves the heating of materials to high temperatures in a kiln. Limestone/shale (dry process) or chalk/clay (wet process) are the starting materials and these are crushed and milled (dry process) or filtered (wet process) before being heated in a kiln at temperatures in the range 900–1600°C to form a clinker. Cement is produced by grinding the clinker with suitable additives, but the important aspect of the process is that there is no ‘bottom waste’ and all of the material is used. Cement kilns are not major polluters of the environment according to inventory figures (Alcock et al., 1999). However, burning hazardous waste such as secondary liquid fuels is becoming more common and some sites are experimenting with old chipped tyres as part of the fuel supply. Metal supports in the tyres will mean that metal is available during the heating and cooling processes and for that reason a site using this fuel was chosen for sampling purposes.
Thermal oxygen cutting
Thermal oxygen cutting (or oxygen arc cutting) is a high temperature (above 2500°C) burning operation in which metal rods are consumed by burning in an oxygen flame. An arc is struck between the workpiece and a tubular electrode covered with flux. Oxygen is fed down the electrode and a metal, which is heated to high temperatures by the arc, burns away to create a metal oxide fume. The 3 m lance is used to remove or to cut through concrete, rock or metal and is used for difficult work in the scrap industry or construction and demolition. By the nature of the work, temperatures are high, metal is involved and organic contamination is likely. German workers (Menzel et al., 1996, 1998) report that during thermal oxygen cutting of scrap metal at demolition sites unpredictable and high air levels of dioxins were found.
A PCDD/F sampling exercise was therefore carried out at a railway tunnel in the UK, where a thermal oxygen lance was used to cut away old concrete and badly corroded metal in order to enlarge the width of the tunnel.
Municipal waste incinerators
Following implementation of the Waste Incineration Directive in 1996 (E89/369/EEC and 89/429/EEC), older UK municipal waste incinerators (MWIs) closed, leaving only 11 with improved air pollution abatement equipment. Higher burn temperatures and rapid quenching of the smoke has reduced the potential for PCDD/F formation, but sampling was undertaken at one modern site to evaluate current occupational exposures.
Power stations
No sampling was undertaken on site because initial assessment suggested limited potential for PCDD/F formation, but as a check, specimens of bottom ash and pulverized fuel ash from one site were taken for analysis.
Landfill sites
Incinerator bottom ash (IBA) represents 80% of the combustion ash from incinerators and is used for road construction and in building blocks. It is collected together with the grate siftings and boiler ash. Air pollution control (APC) ash is technically that part of the ash that is captured by the pollution abatement system (usually bag filters) and includes unspent lime and activated carbon, but is collected together with ‘fly ash’, the part that is removed prior to abatement systems. APC ash is landfilled as special waste because of the lime and heavy metal content and it usually has a PCDD/F concentration much higher than that of IBA. For this reason, both static and personal sampling was undertaken at one UK landfill site while a dedicated ash handling plant was in operation.
Building block manufacture
Not all IBA is landfilled. Some is still used for building block production and if the IBA contains high concentrations of PCDD/Fs the manufacturing of the block, which may contain up to 30% IBA, is a potential source of PCDD/F exposure. The site selected used IBA to produce blocks on the day sampling took place.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONSELECTION OF SITES FOR...MATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Personal and static sampling was undertaken at a range of sites during the period May 2000 to August 2002. Owing to the relatively low number of employees exposed at each site, it was usually possible to obtain personal samples from most of them.
Personal sampling
These were taken onto 32 mm quartz filters (Supleco part no. 21038), backed up by ORBO-1000 polyurethane foam plugs (Supelco part no. 20557). Supelco equipment is available from Sigma Aldrich Co. (Poole, UK). All personal sampling media were soxhlet extracted for 16 h in dichloromethane (DCM) prior to sampling. During sampling the filters were held in custom designed cartridges (Supelco part no. 20037), which allow a gas-tight seal between the filter and ORBO tube during sampling.
It should be noted that these samplers are not designed to sample any specific fraction of airborne dust. This sort of sampling train is widely used for environmental sampling of PCDD/Fs, although this was one of the few times that they have been used for industrial monitoring. The polyurethane plug is intended to capture vapour phase PCDD/Fs. As PCDD/Fs exist in both the gas and particulate phases (though predominantly in the particulate phase), both phases are needed, especially at elevated process temperatures.
Personal air samples were taken at 3 l/min, using GilAir 5 sampling pumps. Air flows were measured before and after sampling using calibrated rotameters. Samples were taken over as long a time period as possible and unless otherwise stated, this was for at least 3 h. There were two reasons for this. First, at sites where personal exposures are particularly low, a larger sample volume maximizes the chances of collecting measurable levels of dioxin. Secondly, such samples are more representative of full-shift exposure. The samplers were mounted in the workers breathing zone, high on the lapel. The glass ORBO tubes were wrapped in foam in order to protect them from breakage during sampling.
Upon completion of sampling, the personal samples were individually wrapped in foil and kept cool (refrigerated where possible) for transportation to the analytical laboratory. At least one field blank was submitted with each set of samples and handled in the same way as the sample modules. It was taken to the sampling location and exposed for a short time, intended to replicate the time it takes to load a sample module.
Static air sampling
The static Hi-Vol samplers were run for 24 h and sample volumes were usually in excess of 100 m3.
Details of the static air sampling methodology can be found in Lee et al. (1999) and the reader should refer to that journal for full details. Briefly, the sampling protocol was based on USEPA method TO4 using General Metal Works high volume air samplers (model PS-1), calibrated prior to use. A diagram of the filter arrangement can be seen in Figure 2.
|
Each sample consisted of a glassfibre filter (11 cm diameter) backed up by two PUF plugs (5 cm x 3 cm radius). The filters were cleaned by baking at 450°C whilst the PUF plugs were pre-extracted by soxhlet, with toluene and then DCM. The volume of air sampled was calculated using the mean flow rate over the sampling period multiplied by the sampling time. Air samples were typically 100 m3 for this study, with a flow rate of 5 m3/h. Additional information pertaining to the sample, such as maximum and minimum temperature and relative humidity, were also recorded.
Analytical methodology
For extraction and clean-up, samples were analysed in batches of 6 or 12, including a field blank and a reference soil. All samples were extracted by soxhlet with toluene for 16 h after addition of 17 [13C12]2,3,7,8-substituted congener recovery spikes (100 pg/sample, spiked directly onto the filter or PUF plug immediately prior to extraction). After extraction, 1 ml of n-nonane was added to the extract as a keeper solvent and the toluene completely removed under reduced pressure on a rotary evaporator before continuing to the two-step clean-up stage. The extract was cleaned by elution through an adsorption chromatography column containing acid-modified silica gel, base-modified silica gel and activated neutral silica gel. The eluate was again concentrated to n-nonane on a rotary evaporator. The extract in n-nonane was applied to a column of 4.5 g aluminium which had been pre-rinsed with 30 ml of n-hexane. The column was first eluted with 20 ml of 7% DCM/n-hexane to separate other halogenated organic compounds that would otherwise interfere with analysis. The column was finally eluted with 20 ml of 1:1 DCM/n-hexane to give a fraction containing the PCDD/Fs. The eluate containing the PCDD/Fs was reduced to 
0.5 ml at 40°C under a stream of dry nitrogen and then quantitatively transferred to a tapered 1 ml gas chromatography vial. An aliquot of 15 µl of a solution containing the injection spike was added and the extract blown down to 15 µl under the same conditions. Recovery efficiencies ranged from 70 to 110%.
Quantitation of PCDD/Fs was carried out by HRGC-HRMS using a HP6890 GC fitted with a HP 6890 series autosampler connected to a Micromass Autospec Ultima high resolution mass spectrometer run in SIR mode with a resolving power of 10 000. Each PCDD/F sample was analysed on two different capillary gas chromatography columns to allow quantitation on a total homologue (i.e. chlorination group) and on an individual 2,3,7,8-substituted congener basis. The columns used were 30 m DB5-ms (0.25 mm i.d., 0.1 µm film thickness) and 60 m SP2331 (0.25 mm i.d., 0.2 µm film thickness). Peak identification was based on retention time as well as ion ratio of the two masses monitored for each homologue. The quantitation software automatically discounts peaks lying outside the set acceptance criteria. The integration of every chromatogram was checked manually before it was accepted. Quantitation of analytes was made by isotope dilution relative to the internal standard using the peak areas of the specific 13C12 surrogate for each 2,3,7,8-PCDD/F analyte. Recoveries of each surrogate were calculated relative to the injection standard. Limits of detection were calculated using a combination of laboratory and field blanks and instrumental sensitivity. Further details of the analytical methodology can be found in Lohmann et al. (2000).
The gas chromatography conditions were as follows: run time 55.50 min; temperature 1, 11.0°C; time 1, 2.00 min; rate 1, 20°C/min; temperature 2, 14.0°C; time 2, 0.00 min; rate 2, 40°C/min; temperature 3, 20.0°C; time 3, 10.00 min; rate 3, 4°C/min; temperature 4, 30.0°C; time 4, 2.00 min; injector temperature, 280°C; inlet line temperature 300°C.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONSELECTION OF SITES FOR...MATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
In all cases, the majority of the PCDD/F was collected on the filters, suggesting that most airborne PCDD/F was particulate; future investigations could concentrate on sampling for the inhalable fraction of the particulates by using a sampler such as the IOM inhalable sampler.
Aluminium recyclers
Most aluminium recycling sites worked a three shift, 24 h cycle with some operating two 12 h shifts. Personal samples were taken for between 4 and 6 h from employees (usually one or two) working around the furnaces and engaged in general furnace work, such as charging, drossing and tapping/casting. These periods of sampling were representative of the full shift. Static samples were taken for 24 h from areas as near to the furnace as possible. Where other areas are included in the sampling, they refer to separate parts of the site.
Tables 1–6 summarize the results from sampling at a total of five sites where aluminium was recycled.
|
|
The average cycle time for rotary furnaces is 6 h, of which 50% of the time is spent in melting, 25% of the time is spent in charging materials and 20% of the time is spent in tapping the furnace. Respiratory protective equipment (RPE) usage was not widespread and tended to be for short periods only and for specified tasks. Site 4 showed the highest air concentrations of PCDD/Fs for both methods of sampling and may reflect the more contaminated scrap that is recycled here.
PCDD/F in air concentrations around reverbatory furnaces were generally lower than those around rotary furnaces and this may reflect the different feedstock.
The personal sample taken from the employee working around a coreless induction furnace from site 1 (Table 3) produced a surprisingly high personal sample yet the personal sample from an adjacent employee was more than an order of magnitude lower.
|
The higher sample came from the employee involved in brass production who charged the furnace with scrap twice during the 5 h sampling period and oversaw a 40 min ingot casting process.
The lower sample came from the employee involved in aluminium bronze production who had a similar schedule of work during the 5 h sampling period.
Extraction at both furnaces appeared similarly efficient close to the capture hood but it was impossible to quantify this because of the heat generation.
The high static PCDD/F in air concentration around the dross processing area in site 4 is comparable with the static samples taken round the rotary furnace at the same site (Table 1) and the personal sample in the dross processing area at site 4 is also in line with the personal samples taken from employees working round the rotary furnaces at this site (Table 1). Examination of the individual congener patterns for samples from this site and examination of the ratios of congeners (not reported) suggested that the dross house air contained a different ratio of congeners to air from the furnace area but that the total ‘dioxin’ equivalence was similar. In other words, the similarity in toxicity equivalence wasn’t due to circulation of background air around the site but was due to a differing but toxicogically similar emission pattern from the cooling dross.
The high PCDD/F concentration for the bag plant waste in site 1 may reflect the high dioxin in air levels found around the coreless induction furnace at this site as the bag plant collects fume and smoke from around the induction furnaces.
Other metal recyclers
The personal samples at the steel production site were taken from furnacemen working around the electric arc furnace. The furnacemen spent about 20% of their time around the furnace and the rest of the time in the control room nearby. The highest static sample (4.7 pg WHO-TEQ/m3) was obtained from near the furnace hopper, the next highest static sample (4.0 pg WHO-TEQ/m3) was obtained from between the furnaces and the lowest static sample (2.4 pg WHO-TEQ/m3) came from the furnace control room.
The magnesium refining site chosen used predominantly high grade, clean magnesium scrap from die casting operations that must be refined before being reintroduced to the casting process. However, it also used lower grade scrap such as swarf and dross and also some aluminium-based material. The scrap was melted on a batch or semi-continuous basis using flux to protect the molten metal from oxidation. The oxide impurities/flux layer is heavier and sinks to the bottom to be removed. Only static samples were taken on site and the two lowest static samples came from the abatement shed where the dust is collected from filter bags, while the two higher static samples came from the foundry area. All values were relatively low, which probably reflects the clean scrap used in these processes.
The highest static sample at the zinc refiner was found at the plant injectors, where the scrap zinc oxides containing potentially high concentrations of PCDD/Fs were added. Only static samples were taken at this site and this area of the site was expected to yield the highest values. The lower static samples were found on the condenser floor and in the bag discharge area.
Cement manufacture
Table 7 summarizes the sampling results from a cement manufacturing site.
|
All air samples had PCDD/F concentrations below the limit of detection (0.6 pg WHO-TEQ/m3 for these sampling periods) and the PCDD/F concentrations in the cement were also very low. Dust collected on site showed a higher concentration of PCDD/Fs, but if the higher value (97.8 ng/kg) was representative of all airborne dust, then the airborne dust levels on site would have to average >6 mg/m3 for the dust PCDD/F concentraion to be above the limit of detection. Visual observation suggested that total dust levels were considerably lower.
Thermal oxygen cutting.
Table 8 shows the results from the sampling exercise at which a thermal oxygen cutter was used to widen the tunnel diameter by melting old and corroded metal struts that braced the tunnel walls.
|
The thermal oxygen cutting took place at least 100 m inside the tunnel and air movement was consequently slow. The highest static sample was obtained by placing the sampler in the path of the visible dark smoke from the cutting procedure, and the north end of the tunnel represented the exit route for the smoke. The sampling period lasted only for the duration of the cutting task (
90 min) and is thus task based; it was designed to detect any significant elevation in PCDD/F levels during this time interval, a typical period for the use of this instrument.
Municipal waste incinerators
The incinerator uses the energy from burning waste to drive a steam turbine to generate electricity. The waste is delivered directly to the plant by collection vehicles and tipped into holding bunkers where it is mixed. The waste is fed by hopper into the incinerator and the bottom ash generated is quenched and automatically transferred via enclosed conveyors to the residue hall. The acidic combustion gases are treated with lime and divided carbon and the ensuing particulates removed by high efficiency bag filters and this APC is landfilled. Two 12 h shifts are worked by employees and their weekly equivalent is 37 h each. Tables 9 and 10 summarize the results of the sampling.
|
|
The personal sample from the tipping hall (Table 6) refers to the employee who performed cleaning duties in the tipping hall area, predominantly using a shovel loader. The personal sample from the residue hall (Table 6) refers to the crane operator who loaded the lorries with bottom ash but who also performed general tasks in this area.
Power station ash
Samples were taken of pulverized fuel ash (PFA), boiler bottom ash and electrostatic precipitator ash from a coal-fired power station. The coal was predominantly from South American sources. The results are shown in Table 11 and demonstrate very low levels of PCDD/Fs in both bottom ash and PFA.
|
Landfill sites
The site takes in general waste and also APC ash by road tanker from incinerators. This latter ash was the reason for the sampling visit. Typically, three to four tanker loads are delivered daily and transfer to a storage silo is by closed system. Ash processing consists of mixing and hydration and the wetted ash is tipped into a truck before being spread on the landfill site. The wet ash is left for about 3 days before being flattened and buried. Work shifts were typically 9 h daily.
Table 12 summarizes the PCDD/F in air concentrations. The plant manager performed various tasks around the ash plant, including monitoring dust levels at various points. Sampling was for 160 min and is deemed to be representative of a shift on and around the ash plant during tanker unloading and truck filling. Another personal sample on the truck driver showed evidence of tampering and is not reported. Table 13 summarizes the PCDD/F concentrations in dust samples from around the on-site ash processing plant. As at the MWI, the on-site dust PCDD/F concentration is commensurate with an urban soil.
|
|
Building block manufacture
As well as being landfilled, IBA may also be used for the production of building blocks. At the site visited, IBA or power station flue ash, limestone and cement are formed into blocks. On the day of the survey, IBA was mixed and pressed to produce ‘green’ blocks that were left to cure.
The raw materials (lacking water and cement) were weighed and conveyed to a mixer where cement and water were added. At the press enclosure, the damp mixture was formed into blocks before being stacked ready for curing in the ovens.
Operators worked a 9 h day loading and working the presses with 1 h general cleaning at the end. Air movement throughout the site was good but dust was controlled by process enclosure and by water addition before pressing. Prior to this visit, it was thought that dioxin exposures were likely to be low, possibly not detectable at this site. Hence, personal sampling was performed for total inhalable particulate, in accordance with MDHS 14, rather than specifically sampling for dioxins. Analysis of bulk samples of dust from the site then allowed an estimation of personal dioxin exposures to be made.
Table 14 summarizes the total inhalable and respirable dust concentrations. The personal samples results are task based and were collected for between 3 and 4 h of routine work. No sampling was undertaken for the final hour of the shift, when cleaning operations take place. Samples of IBA were analysed for PCDD/F and gave values of 10 and 12 ng WHO-TEQ/kg. Assuming that the dust inhaled on site contained 10 ng WHO-TEQ/kg of PCDD/F, then this represents a potential PCDD/F exposure of 0.022, 0.022 and 0.014 pg/m3, respectively, for the loader driver, press operator and cuber operator.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONSELECTION OF SITES FOR...MATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
The highest PCDD/F exposures were found at metal recycling sites, specifically sites that recycled aluminium, and there are obvious reasons why this should be so. One factor is melting temperature; aluminium melts at 660°C and aluminium furnaces generally operate at temperatures well below 800°C, the point at which PCDD/F breakdown becomes rapid. Even if the dioxins and furans were completely broken down in the furnace, escaping fumes and smoke will still contain all the necessary ingredients for reformation of PCDD/Fs in the cooling temperatures because the furnace emits smoke containing organic material, chlorine and metal. Periodic charging with cool material will present a further opportunity for release of particles from the furnace and these particles provide a large surface area for PCDD/F reformation.
Canopy hoods or close capture hoods (canopy hoods with side walls) were the main controls on these furnaces, but visible fumes could often be seen escaping the extraction. Static samplers were arranged to capture the greatest concentration of fugitive emissions during charging and routine operating of the furnace and provide indications of the PCDD/F average air concentrations that were present at these points. Although static samples are useful to show background exposures and can help to pinpoint localized high concentrations, they do not represent true exposures where employees move around or engage in a range of tasks. Around the furnace, convection currents sweep much of the particulate upwards to descend slowly in a cooling cloud towards the floor and continually add to the ambient concentrations. Personal samples collected from employees working around the furnaces showed lower air concentrations of PCDD/Fs than did the static samples, but they were always well within the same order of magnitude. This association was also seen with the iron, zinc and magesium recyclers, but at a lower level.
Rotary furnaces tend to be charged with more contaminated scrap than reverbatory furnaces and the higher PCDD/F levels may reflect this. Another factor is that dust exposure is highest with rotary furnaces (Healy et al., 2001) and the particulate fraction contains the majority of the PCDD/F.
The high PCDD/F exposure from personal and static samples in a separate dross processing area of an aluminium recycler is difficult to explain. Comparison of congener patterns suggested that this exposure was different to the pattern seen in the corresponding rotary furnace area even if the overall TEQ was similar. It would suggest that the fumes and smoke from cooling dross have the potential to form new PCDD/Fs despite there being low levels of organic material available.
The highest value found in personal samples taken from the aluminium recycling sector (54.9 pg WHO-TEQ/m3) (Table 3) was more than twice the next highest, and may represent an unusual sequence of working. Although this employee worked at an induction furnace reycling relatively clean material (copper and zinc) to produce brass, he was engaged in charging, overseeing casting and manually scraping off dross, and this required close proximity to the work. Copper is an excellent catalyst for dioxin formation when conditions are right (Buekens et al., 1998; Olie et al., 1998; Gullet et al., 2000). His colleague, engaged at an adjacent induction furnace producing aluminium bronze from copper, iron, nickel and aluminium, showed a measured exposure of only 3.5 pg WHO-TEQ/m3 and we believe that the difference reflected personal work practices rather than a significantly different PCDD/F production potential. The bag plant dust from this site was collected from around the induction furnaces and contained very high concentrations of PCDD/Fs, suggesting that the fumes/smoke from the process was high in these materials. A slight alteration in posture during working (or slightly less efficient extraction) could therefore produce a much increased PCDD/F exposure.
Since the sampling visit to this site, major alterations have been made to extraction around the induction furnaces and advice has been given on how to reduce exposure from handling dust in the bag plant.
Lower air levels of PCDD/Fs were found at the other metal recycling sites.
Industry data suggested that electric arc furnaces produce most on-site exposure (D.R.Anderson, personal communication, 2001) and our results show similar air concentrations to the industry data, with static samples between 2 and 5 pg WHO-TEQ/m3 and personal samples between up to 9 pg WHO-TEQ/m3.
PCDD/F in air measurements made by static samplers at the zinc/lead refiner varied, with the highest value coming from the area where zinc oxide fines were injected into the process. These powders were known to contain relatively high concentrations of PCDD/Fs, but since the sampling exercise, new sources of zinc oxide wastes have been used.
Magnesium refining uses generally cleaner scrap and despite the relatively low temperature of melting, PCDD/F in air concentrations, measured by static samplers, were low.
On-site static sampling from the cement manufacturer showed no detectable levels of PCDD/Fs in air, the PCDD/F concentrations in the cement were also low and PCDD/F concentrations in the on-site dust were in the high range for urban garden soil (Cox and Creaser, 1995). This site was experimenting with burning small quantities of chipped tyres as part of the fuel for the precalciner (which leads to the kiln) and this method of disposing of unwanted materials is increasing. Recent work on the burning of hazardous waste in cement manufacture showed that dioxin output to the environment seemed to be unaffected by this change (Eduljee, 1999) and our set of sampling results agree with this conclusion.
Our results from the sampling exercise during the use of a thermal oxygen cutter inside a tunnel suggest that even in poorly ventilated areas PCDD/F exposures are relatively low. The use of this instrument is generally sporadic and of short duration, so the short sampling period is representative of normal work. The temperature of the flame is well above the point at which PCDD/Fs would be rapidly destroyed, but there was concern that adjacent areas, especially if contaminated with oils or dirt, would be cooler and susceptible to PCDD/F formation. There was also the possibility of reformation of the toxic materials during the cooling phase. We could find no evidence for this during our short-term sampling on this one occasion but there are many combinations of temperature and contamination possible. Caution should therefore be exercised whenever heavily contaminated, metallic structures are burnt in poorly ventilated areas.
Waste incinerators and landfill sites have been a target of local groups for being dioxin polluters, but this view may be out of date. Air samples taken from areas before and after the burning processes revealed low concentrations of PCDD/Fs. On-site dust PCDD/F concentrations were also low, comparable with urban soil levels (Cox and Creaser, 1995). Recent work from Japan (Kumagai et al., 2002) showed similar on-site dioxin dust levels (0.9, 11 and 33 ng WHO-TEQ/kg) at intermittently burning MWIs, sites that should produce more dioxin than continuously burning incinerators such as the one sampled by ourselves.
PCDD/F concentrations in the air of the MWI, measured by static samplers, were low and personal samplers showed values between 0.16 and 0.25 pg WHO-TEQ/m3. A survey of published data shows that ambient air levels of dioxin vary with the location and season but range from below 0.01 to 0.4 pg /m3 (Lohmann and Jones, 1998).
At the landfill site, the PCDD/F concentration in dust around the APC silos was commensurate with an urban soil, but this is likely to be higher than the surrounding rural soils nearby. The site manager’s PCDD/F personal sample of 1.65 pg WHO-TEQ/m3 was higher than would be expected for a remote rural site (Lohmann and Jones, 1998; Lohmann et al., 1999) and may represent inhalation of PCDD/F from spilled and dried APC ash.
PFA from power stations is the ash collected from electrostatic precipitators. The furnace bottom ash there is similar to IBA but the temperature of the burn is higher in power stations than that in MWIs (1250°C compared with 900°C). PCDD/F concentrations were very low in all samples of power station ash and are similar to results from tests made by the industry on FBA and PFA, where PCDD/F concentrations ranged from 0.05 to 2.4 ng I TEQ/kg (United Kingdom Quality Ash Association, 2002).
All our results represent PCDD/F exposure not PCDD/F intake and this distinction needs to be addressed. Work-related dioxin exposure is significant only if it adds an increment to overall dioxin intake. To model intake from exposure data, we must make certain assumptions, and consider three routes of intake.
Ingestion is a possible route of exposure, whether by hand to mouth transmission or by consumption of food at work. COT assumed 50% bioavailability of ingested dioxin in its derivation of the UK TDI (COT, 2001). Therefore, PCDD/Fs are assumed to be readily absorbed from the gastrointestinal tract. It is difficult to measure exposures of this kind since there is a high dependence on the personal hygiene behaviour of individual workers. On this basis, the contribution of ingestion to the total body burden has not been considered further.
Inhalation is likely to be the main route of intake at UK worksites. For risk assessment purposes, HSE assumes a worker will inhale 10 m3 of air over an 8 h shift; for continuous heavy work this may be an underestimate. RPE usage in all industry sectors surveyed was sporadic and no assumptions are made about any reduction in exposure from its use. We assume that all of the PCDD/Fs adsorbed onto the inhaled dust is taken into the body; this conservative estimate is based on extrapolation from the results of intra-tracheal dosing studies in which animals were dosed with dioxins in an oil/water emulsion or adsorbed onto particulate material (Nessel et al., 1990; Diliberto et al., 1993, 1996). The value of 100% bioavailability is also used by the US EPA (EPA, 2000).
Few studies have examined percentage uptake from dermal exposure, but the limited data suggests that it is reasonable to assume 1% bioavailability across the skin. Dioxins are poorly absorbed across intact skin (Weber et al., 1991). The EPA assumes 1% bioavailability for skin contact with contaminated soils (EPA, 2000). If the skin is damaged, or if protective clothing is not worn or worn incorrectly, dermal absorption may be enhanced. The protection offered by gloves is often overestimated and recent work published in this journal (Garrod et al., 2001) shows that donning and doffing contaminated reusable gloves (a common practice) may be as important a determinant of protection as the glove material. However, in relation to occupational exposure to dioxins, dermal exposures are not considered to make a significant contribution to total body burden.
The values assumed for the routes of exposure are consistent with those used in the draft risk assessment of TCDD and related compounds prepared by the US EPA (EPA, 2000). However, PCDD/Fs have long half-lives in the body and so once absorbed, they will be retained and accumulate on repeated exposure.
Assuming 100% bioavalibility of inhaled dioxins, the highest dioxin exposure (54 pg/m3), from an operator at an induction furnace producing brass, equates to a possible 8 h intake of 540 pg. For a 70 kg employee, this represents between 7 and 8 pg/kg/day, nearly four times the current TDI of 2 pg/kg/day. Observation of this individual suggested that his close proximity to dust and fumes could be reduced considerably by better working practices. Dust from this operation is extracted and collected by filter bags and this dust contained 25 000 ng WHO-TEQ/kg dioxin, a level much higher than any other dusts tested. Back calculations suggest that if this dust is totally responsible for the dioxin intake (and 93% of the dioxin was captured on the filter), then an average dust in air concentration of 2 mg/m3 would produce this personal dioxin value. Following the HSE sampling visit, the company have introduced a canopy over the casting area and made other improvements to the extraction system.
The next highest dioxin exposures were found in operators working near rotary furnaces at aluminium recyclers. The highest personal value measured (25 pg/m3) equates to a possible 8 h intake of 250 pg, and represents nearly twice the current TDI for a 70 kg operator.
However, dioxin exposures at the MWI and the landfill site were considerably lower. The highest value was 1.65 pg WHO-TEQ/m3, and equates to a possible 8 h intake of 16.5 pg, representing just over 0.2 pg/kg for a 70 kg operator, around one-tenth of the TDI.
Although recent reviews have confirmed that there has been a several-fold reduction in exposures and body burdens of dioxins in the general population over the three decades from 1970 to 2000 (ENDS, 2003; Hays and Aylward, 2003), there have been few attempts to quantify the dioxin exposure from work-related activities.
Our results demonstrate that there is a potential for dioxin exposure at some UK workplaces, in particular the metal recycling industry, and this risk must be taken into account when risk assessments are prepared. In the UK, the Control of Substances Hazardous to Health (COSHH) Regulations, 2002, require that employers should assess the health risks associated with dioxin exposure and either prevent exposure or, when this is not reasonably practicable, adequately control exposure. Options that metal recyclers should explore in order to comply with the regulations include eliminating or minimizing the amount of PVC associated with scrap, ensuring that local exhaust ventilation systems are performing satisfactorily, informing employees of the health risks associated with PCDD/F exposure and encouraging good personal hygiene. RPE should only be considered as a last resort.
It should be borne in mind that the number of samples taken was small and that sampling was concentrated in areas where the potential PCDD/F exposure appeared to be greatest. Also, values represent intakes arising from only one day’s activities, and they may not be reproduced every day. Since dioxins accumulate in the body over time, it is an individual’s long-term exposure that will have most impact on their total body burden, and the impact of higher exposures for short periods of time will be balanced by periods of lower exposures. For this reason, some authorites have established tolerable weekly or monthly intakes rather than TDIs as a marker against which to assess human exposures to dioxins. However, if these results represent some current peak exposures, suitable precautions should be taken to reduce this exposure.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONSELECTION OF SITES FOR...MATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
The sampling exercise has revealed the potential for PCDD/F exposure of employees at a range of worksites in the UK. Highest exposures were associated with recycling of metal at elevated temperatures and may contribute a significant proportion to the overall body burden. Partly as a result of this sampling exercise, the HSE has produced guidance on how to reduce exposure to dioxins in the aluminium recycling industry (HSE, 2003).
|
|
|
|
FOOTNOTES |
|---|
* Author to whom correspondence should be addressed. E-mail: colin.davy@hse.gsi.gov.uk
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONSELECTION OF SITES FOR...MATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Addink R, Altwicker ER. (2001) Formation of polychlorinated dibenzo-p-dioxins/dibenzofurans from residual carbon on municipal solid waste incinerator fly ash using NaCl. Chemosphere; 44: 1361–7.[Medline]
Addink R, Espourteille F, Altwicker ER. (1998) Role of inorganic chlorine in the formation of polychlorinated dibenzo-p-dioxins/dibenzofurans from residual carbon on incinerator fly ash. Environ Sci Technol; 32: 3356–9.[CrossRef]
Anderson DR, Fisher R. (2002) Sources of dioxins in the United Kingdom: the steel industry and other sources. Chemosphere; 46: 371–81.[Medline]
Aittola J-P, Paasivirta J, Vattulainen A. (1993) Measurements of organochloro compounds at a metal reclamation plant. Chemosphere; 27: 65–72.
Alcock RE, Gemmill R, Jones KC. (1999) Improvements to the UK PCDD/F and PCB atmospheric emission inventory following an emission measurement programme. Chemosphere; 38: 759–70.[Medline]
Buekens A, Stieglitz L, Huang H, Cornelis E. (1998) Formation of dioxin in industrial combustors and pyrometallurgical plants. Environ Eng Sci; 15: 29–36.
Cains PW, McCausland LJ, Ferrnandez AR, Dyke P. (1997) Polychlorinated dibenzo-p-dioxins and dibenzofuran formation in incineration: effects of fly ash and carbon source. Environ Sci Technol; 31: 776–85.[CrossRef]
Chagger HK, Jones JM, Pourkashanian M, Williams A. (2000) The formation of VOC, PAH and dioxins duirng incineration. Trans Inst Chem Eng; 78B: 53–9.
COC. (2001) Carcinogenicity of 2,3,7,8-tetrachlorodibenzo-p-dioxin, COC/01/S2, July 2001. Available at http:www.doh.gov.uk/tetrachl.htm.
COT. (2001) Statement on the tolerable daily intake for dioxins and dioxin-like polychlorinated biphenyls, COT/2001/07, October 2001. Available at http://www.foodstandards.gov.uk/multimedia/pdfs/cot-diox-full.
Cox EA, Creaser CS. (1995) Determination of polychlorinated biphenyls, polychlorinated dibenzo-p-dioxins and polychlorinated dibenzofurans in UK soils. HMIP 2nd technical report. A7089, DPP Services.
Dickson LC, Karasek FW. (1987) Mechanism of formation of polychlorinated bibenzo-p-dioxins produced on municipal incinerator fly ash from reactions of chlorinated phenols. J Chromatogr; 389: 127–37.[Medline]
Diliberto JJ, Kedderis LB, Jackson JA, Birnbaum LS. (1993) Effects of dose and routes of exposure on the disposition of 2,3,7,8-[3H]tetrabromobibenzo-p-dioxin (TBDD) in the rat. Toxicol Appl Pharmacol; 120: 315–26.[CrossRef][Web of Science][Medline]
Diliberto JJ, Jackson JA, Birnbaum LS. (1996) Comparison of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) disposition following pulmonary, oral, dermal and parenteral exposure to rats. Toxicol Appl Pharmacol; 138: 158–68.[CrossRef][Web of Science][Medline]
Eduljee G. (1999) Waste disposal in cement kilns: a review of dioxin formation and control. Environ Waste Manage; 2: 45–54.
ENDS. (2003) Dioxin and dioxin-like PCBs in the UK diet: 2001 diet study samples. Available on Document Link at endsreport.com.
EPA. (2000) Draft exposure and human health reassessment of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) and related compounds. Available at http:www.epa.gov/ncea.
Fisher R, Fray TAT, Anderson DR. (1998) Investigation of the formation of dioxins in the sintering process. ICSTI/Ironmaking Conference Proceedings, pp. 1183–93.
Garrod ANI, Phillips AM, Pemberton JA. (2001) Potential exposure of hands inside protective gloves—a summary of data from non-agricultural pesticide surveys. Ann Occup Hyg; 45: 55–60.
Groschwitz R, Sommer E. (2000) Dioxine und furane im kremationsprozess und ihr katalytischer abbau. Gefahrstoffe Reinhaltung Luft; 60: 171–7.
Gullet BK, Sarofim AF, Smith KA, Procaccini C. (2000) The role of chlorine in dioxin formation. Trans Inst Chem Eng; 78B: 47–52.
Hays SM, Aylward LL. (2003) Dioxin risks in perspective: past, present, and future. Regulat Toxicol Pharmacol; 37: 202–17.[Medline]
Healy J, Bradley SD, Northage C, Scobbie E. (2001) Inhalation exposure in secondary aluminium smelting. Ann Occup Hyg; 45: 217–25.
HSE. (2003) How to reduce exposure to dioxins in aluminium recycling, INDG 377. Available free from HSE Books.
Kumagai S, Koda S, Miyakita T, Ueno M. (2002) Polychlorinated dibenzo-p-dioxin and dibenzofuran concentrations in serum samples of workers at intermittently burning municipal waste incinerators in Japan. Occup Environ Med; 59: 362–8.
Lee RGM, Green NJ, Lohmann R, Jones KC. (1999) Seasonal, air mass and meteorological influences on atmospheric concentrations of polychlorinated dibenzo-p-dioxins and dibenzofurans (PCDD/Fs): evidence for the importance of diffuse combustion sources. Environ Sci Technol; 33: 2864–71.[CrossRef]
Lohmann R, Jones KC. (1998) Dioxins and furans in air and deposition: a review of levels, behaviour and processes. Sci Total Environ: 219: 53–81.
Lohmann R, Green NJL, Jones KC. (1999) Detailed studies of the factors controlling atmospheric PCDD/F concentrations. Environ Sci Technol; 33: 4440–7.[CrossRef]
Lohmann R, Lee RGM, Green NJ, Jones KC. (2000) Gas-particle partitioning of PCDD/Fs in daily air samples. Atmos Environ; 34: 2529–37.[CrossRef]
Menad N, Bjorkman B, Zhang S, Forssberg E. (1998) Thermodynamic conditions for the formation of dioxin during the recycling of non-ferrous metals from electric and electronic scrap. In B Mishra (ed.), EPD Congress. The Minerals, Metals and Materials Society, pp. 657–73.
Menzel HM, Turcer E, Bienfait HG, Papke O, Ball M, Bolm-Audorff U. (1996) Exposure to PCDDs and PCDFs during welding, cutting and burning of metals. Organohalogen Compounds; 30: 70–5.
Menzel HM, Bolm-Audorff U, Turcer E et al. (1998) Occupational exposure to dioxins by thermal oxygen cutting, welding, and soldering of metals. Environ Health Perspect; 106(suppl 2): 715–21.
Milligan MS, Altwicker ER. (1995) Mechanistic aspects of the de novo synthesis of polychlorinated dibenzo-p-dioxins and furans in fly ash using isotopically labelled reagents. Environ Sci Technol; 29: 1353–8.[CrossRef]
Nessel CS, Amoruso MA, Umbreit TH, Gallo MA. (1990) Hepatic aryl hydrocarbon hydoxylase and cytochrome P450 induction following transpulmonary absorption of TCDD from intratracheally instilled particles. Fundam Appl Toxicol; 15: 500–9.[CrossRef][Web of Science][Medline]
Nessel CS, Amoruso MA, Umbreit TH. (1992) Pulmonary bioavailability and fine particle enrichment of 2,3,7,8-tetrachlorodibenzo-p-dioxin in respirable soil particles. Fundam Appl Toxicol; 19: 279–85.[CrossRef][Web of Science][Medline]
Olie K, Addink R, Schooneboom M. (1998) Metals as catalysts during the formation and decomposition of chlorinated dioxins and furans in incineration processes. J Air Waste Manage Assoc; 48: 101–5
Stieglitz L, Vogg H. (1987). On formation conditions of PCDD/F in fly ash from municipal waste incinerators. Chemosphere; 16: 1917–22.[CrossRef]
Van den Berg M, Birnbaum L, Bosveld ATC et al. (1998) Toxic equivalency factors (TEFs) for PCBs, PCDDs, PCDFs for humans and wildlife. Environ Health Perspect; 106: 775–92.[Web of Science][Medline]
United Kingdom Quality Ash Association. (2002) Statement on dioxin levels in furnace bottom ash and pulverised fuel ash from coal burning power stations, 11 Februaury. Available at www.ukqaa.org.uk.
Weber LWD, Zesch A, Rozman K. (1991) Penetration, distribution and kinetics of 2,3,7,8-tetrachlorodibenzo-p-dioxin in human skin in vitro. Arch Toxicol; 65: 421–8.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  E. Aries, D. R. Anderson, and R. Fisher Exposure Assessment of Workers to Airborne PCDD/Fs, PCBs and PAHs at an Electric Arc Furnace Steelmaking Plant in the UK Ann. Hyg., June 1, 2008; 52(4): 213 - 225. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||