| Front page | | Contents | | Previous | | Next |

Substance Flow Analysis for dioxins in Denmark

3. Formation and turnover by energy production activities

In Denmark energy production is based on a mixture of sources, primarily coal, natural gas, oil and biomass, besides also waste incineration, wind and sun energy. This chapter focuses on fossil fuels and biomass. Emission of dioxin from combustion processes involving such materials is well documented by several studies. As no studies, to the best of knowledge, so far have indicated any natural content of dioxin in these materials, dioxin emission must be assumed entirely to be due to "de novo synthesis" during the combustion process and flue gas treatment operations.

3.1 Coal power plants

Consumption of coal and coke in Denmark in 1998 counted for approx. 235,000 TJ or approx. 9.4 million tonnes /Energistyrelsen 2000/, of which approx. 93% was used for production of electricity and heat by central power plants. The remaining was primarily used for energy supply for manufacturing purposes. Around 1100 TJ or approx. 44,000 tonnes was used for heating purposes by households, farmers or market gardens.

Coal incineration will result in around 13-15% residuals, primarily fly ash and to a lesser extent slag, bottom ash, gypsum and other desulphurization products.

Previous Danish measurements of dioxin formation by coal combustion dates back to before 1990 and did not detect dioxin /Nielsen and Blinksbjerg 1989/. In /Jensen 1997/ dioxin emission to air from coal combustion in Denmark has been estimated at 2 g I-TEQ/year corresponding to an emission factor of 0.2 µg I-TEQ/ton, whereas an estimated 40 g I-TEQ was collected and deposited as production residues. The 40 g collected as residues was estimated based on rather old measurements of total dioxin in fly ash transformed by analogy considerations to N-TEQ.

From a newly investigation in 1999 at a Danish coal power plant an emission to air of 4.7 pg I-TEQ/m³ (n,t at 5.8% oxygen) was reported /Fyns Amt 2000/. This emission corresponds to an emission factor of 33 ng I-TEQ/ton coal. The plant in question (Fynsværket section 7) can with respect to the temperature pattern over the flue gas treatment system and dust removal in general be taken as representative to around 99% of the Danish consumption of coal for energy generation /Elsam-Projekt 2000a/.

The European Dioxin Inventory (section on Germany) gives air emission factors for electricity generation by coal power plants in the range of 1.06 - 7.01 mg I-TEQ/TJ. Assuming a conversion factor of 25 GJ/tonne of coal this equals a range of 0.027 - 0.18 mg I-TEQ/t. For residential heating the Inventory (section on Germany) states air emission factors of (Landesumweltamt Nordrhein-Westfalen 1997):

Measurements from the Netherlands on a coal power plant and an industrial coal combustion plant gave air emission factors as follows /Bremmer et al 1994/:

Based on these figures, it seems reasonable to accept, that

- approx. 8.7 million tonnes were combusted at an emission rate of 0.033 µg I-TEQ/ton

- approx. 0.7 million tonnes were combusted at an emission rate of
0.13 - 2.92 mg I- TEQ/ton.

These assumptions result in a total emission to air of 0.4 – 2.3 g I-TEQ/year that should be regarded as below the previous estimate of 2 g I-TEQ/year.

The amount of residues from coal combustion generated in Denmark in 1998 was approx. 1.5 million tonnes of which around 10% was exported for cement manufacturing abroad, and approx. 200.000 tonnes were used for cement manufacturing in Denmark.

No measurements of dioxins in residues from Denmark exist, and literature figures are scarce. /Dyke et al 1997/ quote figures of 0.02-13.5 ng I-TEQ/kg for grate ash and 0.23-0.87 ng I-TEQ/kg for filter dust from cyclones and bag filters. The residues measured originate from industrial plants and may thus not be representative to residues from large coal power plants.

Flyash from electrostatic filters and bag filters is the dominant residue developed in Denmark. Assuming a figure of 0.2-0.9 ng I-TEQ/kg to be valid to the total amount of residues generated in Denmark, the amount of dioxins collected with these residues may be roughly estimated at 0.3 – 1.4 g I-TEQ/year. It is noted that this estimate is considerably below the previous estimate for Denmark (40 g I-TEQ – see the beginning of this section) which is due to different data sources. In recognition of that neither of the data sources likely are representative to the coal types and operating conditions found at coal power plants in Denmark today, the choice is made here to accept a range of 0.3 – 40 g I-TEQ/year as the best estimate for the dioxin amount collected with residues from coal combustion in Denmark.

Of this quantity 10 % corresponding to 0.03 – 4 g I-TEQ is exported, whereas 0.04 – 5.3 g I-TEQ is used for cement manufacturing in Denmark and the remainder directed to depot in Denmark or used for miscellaneous civil works like road construction etc.

3.2 Other fossil fuels

Other fossil fuels cover natural gas and oil products. The consumption of these energy products in Denmark in 1998 can be summarised as follows /Energistyrelsen 2000/:

Around 70% of the consumption of natural gas was used for industrial processes, power generation and other larger scale uses, whereas the remaining 30% mainly was used for residential heating /Energistyrelsen 2000/.

Dioxin formation and disposal

No measurements of dioxin emission related to combustion of oil and natural gas has been undertaken in Denmark.

The European Dioxin Inventory (section on Germany) gives air emission factors for electricity generation by natural gas in the range of 0.02 - 0.03 mg I-TEQ/TJ. Assuming a conversion factor of 40 GJ/1000 Nm³ this equals a range of 0.0008 - 0.0012 ng I-TEQ/Nm³. For residential heating the Inventory (section on Germany) states air emission factors of /Landesumweltamt Nordrhein-Westfalen 1997/:

Assuming that these data are representative to the qualities and processes used in Denmark, and an average density for oil products of 0.9 kg/l can be applied, the dioxin emission to air can be estimated at:

No knowledge concerning the content of dioxins in soot/ash from combustion of natural gas or oil products seems to exist. The amount of soot/ash generated is, however, small. Soot/ash will be directed to landfills. Emissions to wastewater should be regarded as negligible.

3.3 Biomass

The major biomass fuels are straw and wood. There are following types of wooden fuels: firewood, forest wood chips, wood pellets, wood briquettes and wood waste, including bark.

Straw and woods are used as fuels mainly in private homes, district-heating plants and in central and de-central, combined heat and power plants (CHP).

The total energy production by biomass fuels was estimated at 33,601 TJ in 1998 (see table 3.1).

Table 3.1
Energy production in Denmark 1998 based on biomass / Energistyrelsen 2000/

An ongoing study by the Center of Biomass Technology for the Danish Energy Agency has estimated the number of biomass installations in 1998 as shown in table 3.2:

Table 3.2:
Rounded numbers of biomass installation in Denmark 1998.

The knowledge and assessments related to the different types of installations are presented in the following.

3.3.1 Wood stoves

The number of wood stoves in private homes in Denmark is estimated to be about 400,000 stoves. An investigation from the beginning of the 1990s /Houmøller 1995/ showed that 33% of the woods consumed in these stoves were good qualities of hardwood from forestry. The rest included wood from private gardens, replacement of old hedges, industrial surplus wood etc. Paper, cardboard, milk cartoons, painted and impregnated wood waste (reference is made to section 3.6.1) and perhaps also plastics are known to be used to a certain degree, but there are no available studies and therefore no precise knowledge of these partly illegal customs in Denmark. Attention should e.g. be paid to the fact that the ordinary blue colours used in newspapers, on milk packaging etc. are typically based on copper pigments. It should also be noted (see below), that the typical temperatures present in the stoves as well as the chimney belongs to the interval more or less optimal for dioxin formation.

In the first Danish study /Dyrnum et al. 1990/ the total dioxin emission was estimated at 32 g N-TEQ/year with an uncertainty range of 10-50 g N-TEQ/year based on an annual wood consumption of 222,000 tonnes. The flue gas concentrations were <200 ng total dioxin/Nm³ for hardwood, about 1000 ng total dioxin/Nm³ for waste briquettes and about 65,000 ng total dioxin/Nm³ for PCP-treated wood. It was assumed that burning 1kg wood would generate 8.6 Nm³ flue gas. N-TEQ was assumed to correspond to 1.5% of total dioxin.

In a more recent Danish study /Hansen et al. 1994/ the emission concentration from burning hardwood and softwood under controlled representative conditions in commonly sold Danish wood stoves ranged 5.8-53 ng total dioxin/Nm³ or quite similar to the previous study. The average was 12 ng total dioxin/Nm³ or 0.18 ng N-TEQ/Nm³. The emission factor was 1.9 µg N-TEQ/tonnes wood. The total consumption of wood for stoves was 214,000 ton/year in 1992 and based here upon the total emission was estimated at <0.4 g N-TEQ/year ± 60%. In 1995 the Danish consumption of firewood had increased to 578,231 tonnes, and the dioxin emission correspondingly to 1.1 g N-TEQ/year.

In a new Danish investigation clean birch and dried clean excess wood from manufacturing was fired in a new stove /dk-TEKNIK 2000/. The testing covered for both types of wood ordinary firing as well as night firing. Night firing covers the practice of adding a large amount of wood at one time and adjusting the air supply to a minimum in order to allow the fire to continue the night over. In all cases 6-hours sampling covering lightning as well as operation was performed. Ordinary firing gave a dioxin emission (to air) of 5.1 µg I-TEQ/tonnes wood for birch and 1.7 µg I-TEQ/tonnes wood for excess wood, whereas night firing gave emissions factors of 0.52 µg I-TEQ/tonnes wood for birch and 0.56 µg I-TEQ/tonnes wood for excess wood.

In 1993 the Swedish Environmental Protection Agency reported an emission factor for stoves of 0.13-0.3 mg N-TEQ/tonnes wood burned /Swedish EPA 2000/.

In the Netherlands the emission factors for wood stoves and open wood fire places ranged 1.0-3.3 mg I-TEQ/tonnes dry clean wood and 13-29 m g I-TEQ/tonnes dry clean wood, respectively /Bremmer et al. 1994/. In Switzerland wood stoves were estimated to emit 0.77 (open door)-1.25 (closed door) mg I-TEQ/tonnes clean wood and 3,230 mg I-TEQ/tonnes household waste /Schatowitz et al 1994, quoted by Swedish EPA 2000 and US Dioxin Inventory 1998/.

In the most comprehensive German study /Bröker et al. 1992/ the emission factor for stoves burning clean wood was typically 0.71 mg I-TEQ/tonnes wood and ranged 0.53-0.94 mg I-TEQ/tonnes wood. Burning of wood at open fireplaces resulted in a lower typical value of 0.46 mg I-TEQ/tonnes wood and a range of 0.07-1.25 mg I-TEQ/tonnes wood. In another study with inclusion of 30% paper as fuel the dioxin emission concentrations raised about five times /Launhardt et al. 1996/.

The European Dioxin Inventory has assessed the existing investigations published up to the middle of the nineties (including the investigations described above) and has adopted the following default air emission factors for domestic wood combustion /Landesumweltamt Nordrhein-Westfalen 1997/:

This assessment is here accepted as a reasonable illustration of the variations caused by different types of combustible materials used in wood stoves. Considering that the dominant part of the material burned in Denmark is clean wood, but that other materials to some extent will also be included, it is deemed fair to expect the overall picture to be somewhat between a clean wood situation and a slightly contaminated wood situation. An activity of approx. 430,000 tonnes/year burned and an air emission factor of 1-50 mg I-TEQ/ton equals a total air emission of 0.43 – 22 g I-TEQ/year.

No measurements of dioxin concentrations in ash and soot from wood stoves and chimneys are made in Denmark. /Dumler-Gradl et al 1993 & 1995 quoted by Dyke et al 1997/ gives figures of 75 – 500 ng I-TEQ/kg ash and 500 – 9000 ng I-TEQ/kg soot for a wood based household heating system. They also give soot values of 4 – 42000 ng I-TEQ/kg for a household heating system using a mixture of wood, coal and waste. In the last case the maximum value is related to wood burning only. The mean concentrations of dioxin in soot from various wood stoves and ovens were 1.4-3.5 mg I-TEQ/kg soot /Dumler-Gradl et al. 1995 quoted by US Dioxin Inventory 1998/. In Canada the dioxin content in soot from wood stoves was 211 ng/kg / US Dioxin Inventory 1998/.

Assuming an amount of ash of approx. 4,300 tonnes and a dioxin content of 75 – 500 ng I-TEQ/kg ash, the amount of dioxin to be disposed of with ash can be estimated at 0.32 – 2.2 g I-TEQ/year. This ash will be disposed of with other household waste or spread in gardens.

Soot from chimneys will normally be removed by the chimney sweeper. The amount of dioxin collected and disposed of in this context has not been estimated.

3.3.2 Other plants

A significant amount of other biomass combustion plants is operating in Denmark partly as a result of a Danish policy to develop the utilisation of biomass for energy generation. Generally, the materials combusted in biomass plants will be clean materials. However, it must be assumed that a number of plants will also use materials to some extent contaminated by glue, paint or plastics or perhaps disposable pallets or other types slightly contaminated by PCP. No precise knowledge on this issue is available, and it is not possible to quantify the extent to which the materials combusted are contaminated.

Plant activity

The activity of other plants for energy generation from biomass in Denmark is summarised in table 3.3. It may be noted that the activity of farm boilers has been reduced in the last 10 years, as 580,000 tonnes straw was burned at 11,000 farms in 1989.

Dioxin formation and emission factors

As the raw materials, operation conditions as well as flue gas cleaning varies between the different types of biomass combustion plants, it should be expected that dioxin formation and emission would likely vary also. However, as indicated in the following only few investigations on the different types of plants are available. The data available are presented in the following:

Concerning farm boilers for straw an early investigation by /Nielsen and Blinksbjerg 1989/ reported the very low emission concentration of 0.016 ng Eadon-TEQ/Nm³ and an emission factor of 5 ng Eadon-TEQ/GJ. That will correspond to about 3 ng I-TEQ/GJ. The translation factor from Eadon-TEQ to I-TEQ for this source was about 0.6. In this study the dioxin emission was hundred times greater by burning straw bales than loose straw.

Table 3.3 Se her!
Activity of other plants generating energy from biomass in Denmark

In a newer study /Jensen & Nielsen 1996/ three farm boilers using full bales, sliced bales and grated bales, respectively, were investigated. The emission concentrations from the two first mentioned boilers were below the detection limit (<0.02 ng N-TEQ/Nm³). From the grated bale boiler the air emission concentration was calculated to 16 ng N-TEQ/m³ (n, t) at 10% O₂. The airflow was about 600 m³/h and the load was 41 kg/h, thus the emission flux was about 9.6 mg N-TEQ/h corresponding to an air emission factor of 230 mg N-TEQ/tonnes straw. As an average for all 3 plants investigated, the emission factor can be calculated to 77 mg N-TEQ/tonnes straw.

A new Danish investigation from 2000 on one farm boiler using full bales gave air emission factors of 5.3 - 9.2 mg I-TEQ/tonnes straw /dk-TEKNIK 2000/.

Concerning district heating the study of /Jensen & Nielsen 1996/ also covered three district heating plants using straw. The emission concentrations from two of them were above the detection limit (0.01 ng N-TEQ/m³). The two plants in question can be briefly characterised as follows:

Based on these few data, emission factors of an average 1.7mg N-TEQ/ton and min./max. factors of 0 and 5 mg N-TEQ/ton can be calculated for straw at district heaters with flue gas cleaning.

A new Danish investigation from December 1999 on a straw based district heating plant (6.3 MW) equipped with cyclone and bag filter for flue gas cleaning gave air emission factors of 17 -22 ng I-TEQ/ton straw /dk-TEKNIK 2000/. The temperature of the flue gas over the filter ranged from 110 to120ºC.The investigation was based on 3 samplings, each lasting for 6 hours.

Combustion of wood chips (dry excess wood from furniture manufacturing) and crushed chipboards (inclusive glue, plastic or paper coating and misc. additives) in a district heating plant (6.3 MW) was investigated in summer 2000 /dk-TEKNIK 2000/. The plant was equipped with electrostatic filter for flue gas cleaning, and the temperature over the filter ranged within 110 - 120ºC. For each type of fuel 4 samplings each lasting 6 hours were undertaken. However, 2 out of the total 8 samplings were later assessed as contaminated. For wood chips the air emission factors for the remaining 3 samplings were determined as 15, 18 and 39 ng I-TEQ/ton chips respectively, whereas the factors for crushed chipboards were determined as 17, 21 and 33 ng I-TEQ/ton chipboard /dk-TEKNIK 2000/.

One Danish investigation for small stoker boilers based on wood pellets is available. The investigation was carried out on a new boiler using 6 hours' sampling. The dioxin emission (to air) reported ranged within 0.21- 0.53 mg I-TEQ/ton pellets /dk-TEKNIK 2000/.

Measurements of dioxin emission from 3 Danish central or de-central CHP-plants were undertaken in autumn 1999. From each plant 3 measurements of 2 hours representing normal operation were undertaken. The fuel was mainly straw, but 2 of the plants also used wood chips. Based on energy content wood chips counted for up to 35% of the fuel consumption. The flue gas temperatures ranged between 99 and 129ºC. All plants are undertaking flue gas cleaning by electrostatic filter (1 plant) or bag filter (2 plants). The dust emission of all plants is 10 mg/ Nm³ or below. The dioxin concentrations reported range between 0.4 and 5.3 pg I-TEQ/Nm³ /ELSAMprojekt 2000b/. Assuming 10 Nm³/kg of straw or chips, emission factors of 4 – 53 ng I-TEQ/ton can be calculated.

For stokers burning wood slightly contaminated by glue, PUR and other kinds of plastics and operating cyclones for flue gas cleaning Dutch investigations / Bremmer et al. 1994/ reports air emission factors of 3-8 mg I-TEQ/ton wood with a best estimate of 5 mg I-TEQ/ton wood.

For industrial wood combustion including combustion in boilers, gas turbines and stationary engines the European Dioxin Inventory – section on Germany - reports air emission factors of 1 – 500 mg I-TEQ/TJ for clean wood and 0.75 – 6,200 mg I-TEQ/TJ for contaminated wood /Landesumweltamt Nordrhein-Westfalen 1997/. Assuming a conversion factor of 20 GJ/ton wood these emission factors can be expressed as 0.02 – 10 mg I-TEQ/ton wood for clean wood and 0.015 – 125 mg I-TEQ/ton contaminated wood.

The European Dioxin Inventory /Landesumweltamt Nordrhein-Westfalen 1997/ – section on United Kingdom – reports air emission factors of 17 – 50 mg I-TEQ/ton for straw burning. These factors are partly based on /Nielsen and Blinksbjerg 1989/ referred above. Furthermore is reported air emission factors of 1-2 mg I-TEQ/ton for clean wood burning and 9 – 19 mg I-TEQ/ton for burning of treated wood.

The US air emission factors for various industrial wood-fired boilers were between 0.5-1.3 mg I-TEQ/tonnes. Regarding burning of wood stored in seawater the emission factor raised to 17 mg I-TEQ/tonnes (EPA draft report 1998).

Based on these data the following air emission factors are adopted for the current situation in Denmark:

These factors are deemed appropriate for assessing the total emission in Denmark, but it may well be the case that the emission for some biomass plants will be outside the range stated. The factors are argued as follows:

For straw burning without flue gas cleaning which mainly addresses farm boilers the emission factors reflect the actual Danish experience as described above. For straw burning with flue gas cleaning, one is considering partly CHP plants and partly district heating plants. Again the emission factors reflect the actual Danish experience as described above.

Wood burning without flue gas cleaning deals with small stoker boilers operated by small companies and individuals and fired with pellets, chips and for some boilers also crushed chipboards and larger pieces of wood. The emission factors adopted are partly based on the Danish measurement described above and recognise that the dominant type of fuel will be wood pellets, but do also pay respect to the possibility that part of the chips or wood otherwise used could be contaminated.

Wood burning with flue gas cleaning is relevant to district heating plants and CHP plants, in particular industrial CHP plants. The flue gas cleaning facilities relevant will be cyclones or bag filters and to a lesser extent electrostatic filters, whereas real dioxin filters are not assumed to be used. Generally district heating plants will be designed and operated to maximise heat extraction, and the temperature of the flue gas over the filter will typically be close to 100ºC and will certainly not exceed 200ºC. The same applies to most CHP plants. Furthermore, the plants should be expected to be in control of the materials burned. It has not been investigated to what extent district heating plants have permission to burn contaminated materials. However, as burning of materials classified as waste (e.g. chipboards) is financially less attractive due to the Danish waste fee system, it seems fair to assume that this practise is not widespread for district heating plants. Industrial CHP plants will be designed to burn wood waste from the manufacturing activities that may include chipboards, sawdust, bark etc. but occasionally also other materials like paper depending on the design of the individual plant /dk-TEKNIK 2000/. The emission factors adopted reflects the few Danish measurements described above besides paying respect to the possibility that combustion of contaminated materials could take place to a limited extent.

Combustion of biomass and air emission of dioxin can based on table 4.3 and the emission factors adopted above be summarised as follows:

The total emission of dioxins to air from biomass combustion plants in Denmark can thus be estimated at 0.3 – 19 g I-TEQ/year.

It is noted that the air emission of dioxins from combustion of biomass (other sources than wood stoves) in Denmark in a previous report /Jensen, 1997/ has been estimated at 0.07-6.6 g I-TEQ/year (straw burning) and 0.25 g I-TEQ/year (wood burning).

Residues

No measurements of the content of dioxins in residues from biomass combustion have so far been carried out in Denmark. The international data available are presented in the following:

The only study available on residues from straw combustion is a UK study concerning a whole bale straw combustor from which a concentration of 10 ng I-TEQ/kg grate ash was reported /Dykes et al. 1997/. The concentration was considered very low and caused by a high temperature and long residence time on the grate causing destruction of dioxins. As an estimate for assessing the situation in the UK covering both good and poor combustors was adopted the range of 10 – 500 ng I-TEQ/kg ash /Dykes et al. 1997/.

On wood a few studies are available:

Burning of natural wood in different wood combustion systems concentrations of 0.23 – 1.12 ng I-TEQ/kg in bottom ash and 117 – 272 ng I-TEQ/kg in filter ash has been reported / Oehme & Müller 1995/. The same study reported concentrations of 22 ng TEQ/kg bottom ash and 722-7620 ng TEQ/kg in filter ash after burning a mixture of PCP-treated and untreated wood.

In a Swiss study /Wunderli et al 1996/ of natural wood incineration in installations of from 20 kW to 1.8 MW fly ash collected in cyclones and bottom ash contained only low levels of dioxins (0.6 - 8.5 ng I-TEQ/kg) and lower than bio compost. Fly ashes from waste wood incineration had much higher dioxin content of 700-21,000 ng I-TEQ/kg. If the combustion process has been efficient, bottom ashes were as low in dioxin concentration as ashes from clean wood incineration. Otherwise the concentration could be as high as for fly ashes.

At a German test facility for industrial wood combustion burning contaminated wood the dioxin content in filter dust and bottom ash ranged from 30 to 23,300m g I-TEQ/ton dust and 30-3,300 mg I-TEQ/ton ash, respectively /Landesumweltamt Nordrhein-Westfalen 1997/.

In a UK study of a boiler burning treated wood (however, not PCP-treated) was reported grate ash concentrations of 584-1090 ng TEQ/kg and grit ash levels of 891-1070 ng TEQ/kg /Dykes et al. 1997/.

For a stoker burning wood slightly contaminated by PUR, soot collected from the inside of the stack was reported to contain 0.2 mg I-TEQ/kg /Bremmer et al. 1994/.

Based on these data, that does not allow for distinguishing between straw and wood burning the choice has been made to adopt a concentration range of 1 - 1000 ng I-TEQ/kg as representative to ash produced at Danish plants. Based on the data in table 4.3 the total quantity of ash from Danish biomass combustion plants can be added up to around 36,000 tonnes/year corresponding to a dioxin quantity of 0.04 – 36 g I-TEQ/year. The dominant part of this dioxin should be expected to be directed to landfills. However, part of the production from farm boilers may likely be spread on soil.

As ash from farm boilers counts for around 36% of the total ash production, the dioxin quantity to be spread on soil could be up to 0.01 – 13 I-TEQ/year.

3.4 Summary

The assessments and estimates related to formation and turnover of dioxins by energy production activities in Denmark by the end of the nineties and presented in section 3.1 to 3.3 are summarised in table 3.4.

Table 3.4
Summary of formation and turnover of dioxins by energy production activities in Denmark

| Front page | | Contents | | Previous | | Next | | Top |