19. Dioxins and Furans in the Chemical Industry

by Ms. Veronique Garny

Dioxins is the general collective term for chlorinated aromatic compounds, consisting of a group called polychloro dibenzo-p-dioxins and polychloro dibenzofurans. The first group (PCDDs) consists of 75 and the second (PCDFs) of 135 congeners. Characteristic of all the compounds is a high melting point, a low vapour pressure, a low water solubility and a reasonable solubility in and affinity for apolar media. These differences in affinity make it possible to predict where dioxins will occur and how to remove these compounds.

Physical properties:

Melting point > 593° K

Vapour pressure at 298° K < 10^-6 Pa

Partition coefficients organic media/water > 10⁶

As a consequence of this, accumulation or adsorption takes place in organic matter in sediments, suspended solids, fly ash, soot and fatty tissues of organisms. Dioxins will not be found at significant levels in water except adsorbed onto solid particles in suspension.

These properties are important to predict where PCDDs/PCDFs will be found, if produced during manufacturing processes, and to define the removal methods.

Dioxin concentrations in the environment are usually very low, measured in fractions of a gram. These are typically expressed per kilogram of soil, per cubic metre of air, and per litre of water. To avoid confusion, concentrations are expressed in the units that minimise the number of zeros and so the following units are used widely in this report.

1 milligram (mg) = 0.001 g

1 microgram (µg) = 0.000001 g

1 nanogram (ng) = 0.000000001 g

1 picogram (pg) = 0.000000000001 g

1 femtogram (fg) = 0.000000000000001 g

Dioxins occur in the environment in complex mixtures of the 210 congeners. Most of the 210 dioxin congeners are thought to pose no risk to human health, and only 17 congeners with chlorine are reported to have potential health effects. Since the congeners have different toxicities, it is difficult to determine the overall toxicity of any mixture. An international system has therefore been developed which assigns toxicities to each congener relative to the most toxic form (namely 2,3,7,8-TCDD). Seventeen dioxin congeners have been identified as having significant toxicity and have been assigned the Toxic Equivalent Factors (TEF) in Table 2.

Dioxins Factor Furans Factor

2,3,7,-TCDD 1 2,3,7,8-TCDF 0.1
1,2,3,7,8-PeCDD 0.5 2,3,4,7,S-PeCDF 0.5
1,2,3 ,7,8-PeCDF 0.05

1,2.3,4,7,8-HxCDD) 1 ,2,3,4,7,8-MxCDF
1,2,3,6,7,8-HxCDD) 0.1 1,2,3,7,8,9-HxCDF
1,2,3,7,8,9-HxCDD) 1 ,2,3,6,7,8-HxCDF 0.1
2,3,4,6,7,8-MxCDF

l,234678-HpCDD 0.01
1,2,3,4,6,7~8-1tpCDF) 0.01
1 ,2,3,4,7,8,9-MpCDF)
OCDD 0.001
OCDF 0.001

The TEF is effectively a weighting factor which, if multiplied by the known concentration of a congener, gives a Toxic Equivalent (TEQ). The toxicity of any mixture is given by the sum of the TEQs. For example, Table 3 shows a soil sample containing three dioxin congeners at different concentrations. If these concentrations were simply summed together, then the sample would be reported as containing 60 ng/kg of dioxin.

However, this ignores the fact that 1,2,3,4,7,8-HxCDF is ten times less toxic than 2,3,7,8-TCDD. By applying the toxicity factors to each congener and summing the results it can be seen that the Toxic Equivalent (TEQ) is 23 ng/kg. The TEQ figure therefore takes account of the relative strengths of dioxins and enables comparison with other results.

(ng/kg) (TEF) (TEQ)

2,3,7,8-TCDD 10 1 10

1,2,3,7,8-PeCDD 20 0.5 10

1,2,3,4,7,8-HxCDF 30 0.1 3

Total concentration = 60 ng/kg Total TEQ = 23 ng/kg

Dioxins are not produced intentionally and have no known use, but they occur in the environment in air, soil and water. Dioxins occur as a result of natural processes (forest fires, volcanoes) although these sources are relatively unimportant when compared to the amounts released from human activities. Interest in dioxins has increased over the last 20 years due to an awareness of the contribution of man-made dioxins, and also to improved techniques for the measurement and detection of dioxins at extremely low levels.

Major public interest in dioxins originated with the 1976 incident at Seveso, Italy when a phenol/dioxin mixture was released after an explosion at a trichlorophenol manufacturing plant. Dioxins also gained notoriety as a contaminant in "Agent Orange" which was used as a leaf defoliant in Vietnam.

The chemical industry is not a major source of PCDDs/PCDFs according to national emission inventories in European countries such as Germany, the UK, the Netherlands and Sweden. Combustion processes are considered to be an important primary source of PCDDs/PCDFs (ref. Wormgoor, Björndal, Quass). Most thermal processes which involve burning organic and inorganic compounds containing chlorine atoms e.g. sodium chloride, result in the formation of PCDDs/PCDFs. Of special importance is the incineration of various types of municipal, hospital and hazardous wastes. Production of steel and non-ferrous metals (copper, magnesium, nickel…) are other notable sources.

It is commonly accepted today that formation of dioxin-like compounds in flue gases happens in a temperature range of 200 to 400°C, especially in poor combustion conditions leading to incomplete conversion and the formation of carbon in the fly ashes. Levels of dioxins in flue gases of incinerators depends on the well known 3 T rule:

• Temperature above 850°C
• Time (residence) typically 2 seconds
• Turbulence optimised by furnace geometry and secondary air.

As PCDDs/PCDFs are mainly fixed onto fly ashes, the effectiveness of the dust separation is of paramount importance.

Recently the UK Authorities published a review of dioxin emissions into air (1995). The dominant source was the incineration of municipal solid waste representing 70% of the total industrial sources. Other major emissions, representing 23% of the total industrial emissions were:

• steel mills
• combustion of coal (energy, fuel)
• iron and steel plants
• non-ferrous metals operations
• hospital wastes

Each of these 5 sources is of approximately equal importance. Emissions from the chemical industry are very much smaller than the above sources. Similar date was reported by Quass et al in this report prepared on behalf of the European Commission DG XI.

In a number of European countries, new and/or existing incineration plants have to comply with the emission limit of 0.1 ng TEQ/m³. These new regulations will drastically reduce the dioxin emissions from the exhaust gases from waste incinerators.

A national inventory was published by the RIVM for Holland in 1993 (Emmissies van Dioxinen in Nederland) concerning the estimated dioxin emissions to air in 1991. From a total of 484 g I.TEQ per year, the chemical production processes accounted for of an emission level of 0.5 grams per year.

In the next sections of this paper we will assess emissions from the chemical industry processes and from end-product uses.

Dioxins can be formed in chemical processes~ where the element chlorine is involved. The following processes in particular have been identified as sources of dioxins formation.

- Pentachloronhenol (PCP)

- the order of magnitude of furans formed during oxychlorination

- the distribution of furans in consecutive process steps

- the potential outlets and actual discharge into the environment.

The actual amount of furans formed with the EDC production is less relevant, as only a small fraction, if any, of the fliran compounds produced is discharged into the environment.

5.2 The main portion of furans formed originates during crude EDC production by oxychlorination. After distillation, these furans are all concentrated in the heavy ends residues fractions. In line with BAT guidelines, these heavy ends should be decomposed by thermal treatment with recovery of energy (steam) and/or hydrogen chloride. In these installations, furans are fully destroyed. Formation of new dioxins is effectively limited by controlled design and operation conditions:

- Temperature, turbulence, residence time and oxygen concentration are the keys to a complete destruction and the final dioxin level concentration in the off-gases. Typical temperatures of 1,1000C are used in combination with a residence time of 2 seconds. However, good turbulence in the reactive zone or higher temperatures are compatible with lower residence times. Specific burners have been developed to limit or even suppress the need for additional fuel even with these wastes of low caloric value: 8400 to 10450 kJ/kg.

In order to avoid any further chemical transformation in the off-gases after the reactive zone and the eventual heat recovery boiler, a quench column is provided to quickly cool the exhaust gases and allow the further treatments.

- The environmental performance of these units is good regulated by a Directive of the European Union.

5.3 The other portion of furans is contained in the solids fraction of the by product water, where the furans are adsorbed. The reaction effluent is processed before being discharged outside the plant boundaries. Treatment facilities can include stripping, flocculation, settling, filtration and biotreatment. Combinations of these units are in operation at VCM production sites. With biological treatment, an additional removal of dioxin related components may take place by adsorption on active sludge depending on the efficiency of the primary physico-chemical treatment. Disposal of this biosludge therefore takes care of this potential contamination.

The residues of the treatment processes are either incinerated as chemical waste or disposed of in controlled deposits.

For emission into water, the ECVM BAT is restricting to a level of less than 1 microgram TEQ per tonne of ethylene dichioride capacity.

5.4 Waste disposal. The main quantity of waste e.g. the heavy fractions are combusted as indicated in section 5.2. Other contaminated wastes are: used oxychlorination catalysts, metal sludges from the physico-chemical treatment, and biosludges from the biotreatment.

- The contamination of the latter can be avoided with a good efficiency of removal of the primary treatment e.g. filtration.

- These wastes are either incinerated as chemical wastes or disposed of in controlled deposits.

5.5 Analysis of the oxychlorination off gases have revealed that there are no dioxin related compounds in these gases. After treatment for solvent recovery these are combusted before they are emitted to the atmosphere. The presence of chlorinated compounds in the gas entering the combustion step is however compatible with an emission standard of less than 0.1ng/Nm³ of TEQ in the off-gases.

We would like to conclude this chapter 5 by making an observation. This chapter provides fact-based information on the best practices to control the emissions of dioxins from the chlorine industry and PVC in particular. The expansion of the chlorine and PVC industry based on public demand for its products has been going on for more than 40 years. According to numerous international studies, dioxin levels have steadily declined and significantly declined since 1970. This dioxin reduction occurred at the same time PVC production has almost tripled worldwide. These long-term trends are demonstrated in the scientific publication of Allcock and Jones and illustrated in fig 1. of this paper.

Extensive analysis of dioxin related compounds in the marketed products have been performed in the last decade. Programmes have been performed in Europe and in the USA.

In the Dioxin 96 meeting in Amsterdam, Hans Wagenaar et al presented a paper entitled: "Analysis of PCDDs and PCDFs in virgin suspension PVC resin". They concluded that: the results demonstrated that virgin suspension PVC resin from 11 production sites in Europe does not contain PCDDs and PCDFs at concentrations above the limit of quantification, which is less than 2 parts per trillion (ppt) (TEQ).

In the same meeting, W. F. Caroll et al presented results obtained in a similar study in the USA, and including EDC. For the PVC pipe resins, results ranged from not detected to 1.4 ppt TEQ (for 12 samples). For bottles and packaging resins, on 8 samples, the results ranged from 0 to 0.7 ppt TEQ. No 2,3,7,8-TCDD, which is believed to be the most toxic dioxin congener, was detected in any sample. For EDC, which is distilled to high purity, one could expect that non-volatile contaminants like dioxin related compounds should not be present. This was confirmed by the analytical results of 0.0 to 0.2 ppt TEQ.

In Germany, a regulation stipulates that any marketed product should have less than 2 ppb of 2,3,7,8-TCDD and less than 5 ppb for all the sum of other isomers as listed in the "Verordnung" (ref. Basler, A.). As a result, the German VCI (Verband der Chemischen Industrie) investigated more than 50 products including those produced in processes where dioxin-like compounds formation are known, for example:

- chlorobenzenes

- chlorophenols

- dyes from chloroaniline

- chlorotoluenes

- chloronitrobenzenes

All the products investigated complied with the limits indicated above.

Waste incinerators burning hazardous or municipal wastes which contain chlorine can exhibit high dioxin concentrations. This observation has led to some regulators or public interest groups calling for limits on the input of chlorinated compounds into waste incinerators. In this section, we examine the available data to determine if the hypothesis of relationship between chlorine input and dioxins output is supported by field experience on waste incineration.

According to J. Vehlow (Forschungszentrum Karlsruhe Technik und Umwelt Institut für Technische Chemie), who investigated the behaviour of a large range of chlorinated and brominated products in the feed of municipal solid waste incinerators. His conclusion was:

"The increased Cl and Br levels (in the feed) caused no significant increase of the concentration of PCDDs or PCDFs in the raw gas".

These findings are in line with many studies concerning the possible impact of PVC materials in the feed of municipal waste incinerators.

The most extensive study was performed in the United States, where the impact of the waste feed chlorine content on PCDDs and PCDFs emissions was analysed on 155 facilities (The study has been published in 1995 by the American Society of Mechanical Engineers, H.G. Rigo et al). The conclusion was:

Similar conclusions on the lack of relationship between the chlorine content of the feed and the PCDDs and PCDFs emissions have been made by many other studies. In particular by: Pr. Ballschmiter University of Ulm (Germany, 1990), R. de Fré, Centre d’Etudes Nucleaires de Mol (Belgium, 1986), Joseph Visalli, New York State Energy Research and Development Authority (USA, 1987), Guigliano and Cernuschi, Politecnico di Milano (Italy), J. Kanters and Robert Louw, University of Leiden (1992), Karasek et Guiochon (1983), University of Waterloo (Canada) and Ecole Polytechnique (France), and E.W.B de Leeren et al TNO Report (Nederland, August 1990).

All these studies concluded that there is no significant relationship between chlorine in incinerator feeds and dioxin emissions from waste incinerators. The way forward is to apply Best Available Techniques - BAT on process design, abatement techniques and achievable release levels of waste incinerators as is practised in most European countries.

Chemical processes where PCDD/PCDF can be formed have either been stopped (e.g. PCBs) or adequate best techniques (BAT) are installed to drastically reduce their emissions into the environment.

A good illustration of this is the self-regulatory Charter adopted by the European PVC industry which includes dioxin emission guidelines based on Best Available Techniques.

The dioxin content of PVC is below detectable limit.

The chemical industry, including the chlorine industry, does not produce significant emissions of dioxin related compounds into the environment.

Inadequately equipped waste incinerators appear to be the main source of dioxins in Europe’s atmosphere. There is no significant relationship between chlorine content in incinerator feeds and dioxin emissions. PCDD/PCDF emissions into the environment can be effectively controlled by applying the Best Available Techniques (process control, abatement techniques).

• Allcock and Jones, Environ. Sci. Tech. 30, 3133, (1996)
• Ballschmiter, K., University of Ulm, Dioxins and Furans from Waste, Chemische Industrie 7/90, (1990)
• Basler, A., Dioxins and Related Compounds, Environ. Sci. & Pollut. Res., 2(2) 117-121, (1995)
• Björndal, H., UNEP supported Expert Meeting to review the Outcome of a Project on Alternatives to POPs, Stockholm, May 20-22, (1996)
• Carroll, W.F., et al, Characterisation of emissions of dioxins and furans from ethylene dichloride (EDC) and vinyl chloride (VCM) and polyvinylchloride (PVC) in the United States, Dioxin 96’ meeting, April 14-16, (1996)
• ECVM, On the environmental impact of the manufacture of polyvinylchloride, a description of Best Available Techniques, (1994)
• de Fré, R., Centre d’Etudes Nucleaires de Mol, Belgium, Dioxin level in the emissions of Belgian municipal incinerators, Chemosphere, Vol. 15 N° 9-12, p. 1255-1260, (1986)
• Giugliano, M. and Cernuschi, S, Politecnico di Milano, The emission of dioxins and related compounds from the incineration of municipal solid wastes with high content of organic chlorine
• HMIP Her Majesty Inspectorate for Pollution, A review of dioxin emissions in the UK, (1995)
• Kanters, J. and Louw, R., Center of chemistry and the environment, Leiden University, Netherlands
• Karasek, F.W.,Viau, A.C., and Guiochon, G., University of Waterloo Ontario, Ecole Polytechnique France, Journal of Chromatography 270, p. 227-234, (1983)
• de Leeren, E.W.B. et al., Environmental Aspects of Wastes Incineration, TNO Report, (August 1990)
• Pr., Louw, University of Leiden, Nederland, PVC waste incineration and "dioxin", Report on a project carried out from Nov. 90 to May 92, (1992)
• Marple et al, Env. Sci. Technology, 20, 18, (1986)
• Marple et al, In "learning from dioxins" American Chemical Society Solving Hazardous Waste Problems, p. 105-113, (1987)
• Mill, T., in "Dioxins in the Environment" (Kamrin and Rodgers) Hemisphere Publ. Corp. Washington, DC, p. 173-194, (1985)
• OSPARCOM, Best Available Techniques for the vinyl chloride industry, (1996)
• Papp, R., Organochlorine Waste Management, Pure and Applied Chemistry, Vol 68, N° 9, p. 1801-1808, (1996)
• Quass et al, Identification of Relevant Industrial Sources of Dioxins and Furans in Europe,North Rhine-Westphalia State Environment Agency (report prepared on behalf of the European Commission DG XI.
• Rappe, C., Sources and environmental concentrations of dioxins and related compounds, Pure and Applied Chemistry, Vol. 68 N° p. 1781-1789, (1996)
• Rigo et al., The relationship between chlorine in waste streams and dioxin emissions from combustions, The American Society of Mechanical Engineers, (1995)
• RIVM, Emissies van dioxinen in Nederland, Report N° 770 501 003, (April 1993)
• Rodorf B., Chemosphere 15 1325, (1986), Proceedings of the 3rd Symposium European Conference on Thermal Analysis and Calorimetry, Termochimica Acta, (1984)
• Schroy et al, Physical properties of 2,3,7,8-tetrachlorodibenzodioxin, Technical Publication 891 American Society for Testing and Materials, Philadelphia, USA, (1985)
• Sheffield, A., Sources and releases of PCDDs/PCDFs to the Canadian environment Chemosphere, Vol 14, N°6/7, p. 811-814, (1985)
• Umweltbundesamt Report 58/95
• Vehlow, J. et al, Co-combustion of electric and electronic wastes in the Karlsruhe test incinerator TAMARA. Paper presented at Recycle 97, Geneva, (February 1997)
• Viselli, J.R., Midwest Research Institute New York State Energy Research and Development Authority, Report 87-16, (June 1987)
• Wagenaar et al, Analysis of PCDDs and PCDFs in virgin suspension PVC resins, Dioxin 96’ meeting, April 14-16, (1996)
• Wormgoor, J.W., Sources of Dioxin Emissions into the Air in Western Europe, TNO Institute of Environmental and Energy Technology, reference number 94-405, (1994)