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

Mass Flow Analyses of Mercury 2001

2 Uses in Denmark

2.1 Raw materials and semi-manufactured goods
2.2 Use of mercury as a metal
   2.2.1 Dental fillings
   2.2.2 Light sources
   2.2.3 Electrical switches, contacts and relays
   2.2.4 Thermometers
   2.2.5 Monitoring equipment
   2.2.6 Other applications in metallic form
   2.2.7 Summary - metallic mercury
2.3 Use of mercury in chemical compounds
   2.3.1 Batteries
   2.3.2 Chemicals for laboratory and industrial use
   2.3.3 Other uses of chemical compounds
   2.3.4 Summary - chemical compounds
2.4 Mercury as an impurity in other products
   2.4.1 Coal
   2.4.2 Oil and biological fuels
   2.4.3 Cement
   2.4.4 Agricultural lime and commercial fertiliser
   2.4.5 Foodstuffs
   2.4.6 Miscellaneous consumer goods
   2.4.7 Summary - mercury as an impurity

2.1 Raw materials and semi-manufactured goods

Imports, exports and manufacturing of raw materials and semi-manufactured goods containing metallic mercury or mercury compounds, as recorded by Statistics Denmark for the years 1996-2001, are shown in Table 2.1.

Table 2.1 Supply information from Statistics Denmark for metallic mercury and mercury compounds for the years 1996-2001 (tonnes). *1

Notes:
*1 The quantities shown in the table have been obtained from Statistics Denmark and are given in tonnes. The relationship between the quantities shown in the table and the position numbers of the goods statistics are as follows:
Metallic mercury: position numbers 2805.40.10 and 2805.40.90;
Sulphates of mercury: position number 2833.29.70;
Nitrates of mercury: position number 2834.29.30;
Mercury amalgams: position number 2843.90.10.

Of the recorded imports/exports, the following should be noted:

Manufacturing
In 2001, as was also the case in the previous report of 1996, no manufacturing of the above chemical compounds was recorded.

Metallic mercury
Imports of metallic mercury are used in the applications shown in Table 2.2, which is based on information obtained during this mass flow analysis. The difference between the recorded consumption during 2001, of about 600 kg, and total imports, of 1,300 kg, can be attributed to the re-exportation of part of the imported mercury and to stock fluctuations. Thus, it has not been possible to locate 900 kg of pure mercury destined for export through the recycling industry.

Sulphates and nitrates of mercury
In both cases, the information available from Statistics Denmark covers compounds of mercury and lead. As stated in Section 2.3.2, the consumption of mercury-containing compounds is marginal in Denmark today, i.e., 30-60 kg mercury-containing compounds, which corresponds to 20-40 kg of mercury/year. Thus, the remaining imports must be assumed to apply to lead compounds.

Amalgams
Denmark imports/exports a certain quantity of amalgams. According to Statistics Denmark, 2001 witnessed a net export of 9 tonnes of precious metal amalgams.

It has not been possible to identify the listed quantities of imports/exports of precious metal amalgams. Whether or not capsules of mercury and amalgam powder for dental amalgams are recorded by Statistics Denmark remains unknown. Such capsules are not manufactured in Denmark. Therefore it is not possible to explain the large net export for the year 2001.

Table 2.2 Metallic mercury consumed in manufacturing, etc., in Denmark in 1992, 1993 and 2001.

1) The quantity listed for 2001 covers 50% of dentists' consumption of mercury purchased in the form of pure mercury.

2.2 Use of mercury as a metal

2.2.1 Dental fillings

Description
Dental amalgam is a mixture of mercury, silver, copper and tin, of which mercury constitutes between 44 and 51% by weight, although typically 48-49% (Munksgaard, 2001). An average mercury content of 48% has been assumed in this report. Silver constitutes the greater part of the remainder, whereas copper and tin are only included in smaller quantities.

Amalgam is supplied to dentists, partly in the form of pure mercury which is mixed on the dentists' premises with the amalgam powder that contains the other constituents, and partly as individual capsules containing the mercury and amalgam powder, separated by a thin membrane, which ruptures when shaken in special shaker machines. The use of these capsules is considered to result in less waste in the form of excess amalgam, than the method of weighing at the individual clinic. Capsules account for about half of the Danish consumption of amalgam.

Consumption in Denmark
Consultation with the four Danish dealers in mercury for dental amalgam revealed a total Danish consumption of mercury in amalgam of about 1,200 kg in 2001.

Roughly the same result can be obtained from based on performed on the number of fillings. For dentists in private practices, the annual consumption of mercury in amalgam for dental fillings can be calculated from their reports to the National Health Service, of the numbers of amalgam fillings performed. In addition to this, there is municipal dental care (i.e., the school dental service), which is not within the scope of the reports sent to the National Health Service.

Private dentists
Table 2.3 shows the number of amalgam fillings performed in private dental practices in Denmark and reported to the National Health Service for the years 2000 and 2001. The reported number of fillings for all years between 1994 and 2001 can be found in Appendix 1. The previous mass flow analysis (Maag et al., 1996) has corresponding data for the years 1980-1983.

In the reports to the National Health Service - and in Table 2.3 - the size of the fillings are shown as "a," "b" and "c," depending on the number of tooth surfaces that were covered with the filling material. Based on detailed investigations, the amalgam consumed (including waste) for performing "a"-, "b"- and "c"-type fillings has been estimated as follows (Maag et al., 1996; Munksgaard, 2001) (the numbers correspond to those of the previous mass flow analysis):

Table 2.3 The National Health Service's accounting of the number of amalgam fillings performed, distributed by filling size "a," "b" and "c," for the years 2000 and 2001. This account only covers fillings performed in private dental practices 1).

1) The figures reported to the National Health Service are considered to be reliable and precise. The numbers shown in the table include temporary amalgam fillings (so-called "gradual excavation"), which the National Health Service accounted separately for the years 1999-2001.

Public dental care
In January 2002 the National Health Service estimated that 6,000 mercury amalgam fillings are performed annually on milk teeth (Nielsen, 2002). If an average of 0.4 g mercury/filling is assumed, as in the previous report (Maag et al., 1996), 2.4 kg of mercury were used in fillings in milk teeth.

In addition to this amalgam is used in public dental care for the permanent teeth of youths. Each child receives an average of 0.8 fillings/year. On the assumption that 20% are amalgam fillings (type "a"), the total mercury consumption in municipal dental care - about 1m youths - can be estimated at about 100 kg of mercury/year (Holst, 2002).

Total consumption
The total mercury consumption for dental fillings in Denmark in 2001 can, thus, be estimated at about 1,200 kg/year. It is realistic - based on the authors' assessment - to calculate with an uncertainty of ±100 kg. This relatively low uncertainty is due to the fact that the information given by all dealers in amalgam agrees well with the consumption calculated from the number of fillings.

Development trends
The consumption of amalgam for dental fillings has dropped somewhat over the course of the past 20 years. Thus, the total consumption in 1982-83 was calculated as 3,100 kg of mercury and for 1992-93, as 1,800 kg of mercury. The number of amalgam fillings reported to the National Health Service has dropped from 2.7m in 1982 to 1.8m in 1992 and 1.1m in 2001. This reduced consumption is thought in part to be due to improved dental hygiene in the population, and in part to be due to the increased use of alternative filling materials. One significant factor here is probably the prohibition against using mercury in teeth other than the molars and premolars, where fillings are subject to wear (Statutory Order No. 692 of 22 September, 1998 on prohibition of sale and expert mercury and mercury-containing products).

Subsequent fate of amalgam
This report estimates that about 60% of dental mercury ends in teeth. This corresponds to the estimate of the previous mass flow analysis, and it corresponds to about 1 g amalgam/tooth (Munksgaard, 2002), i.e., that between 660 and 780 kg of mercury/year ends in finished dental fillings.

The rest of this consumption sector is collected as excess amalgam or is taken by saliva ejectors together with the mercury from old, excavated fillings.

Excavation of old fillings
The average life-cycle of amalgam fillings is estimated at between 7 and 20 years (Munksgaard, 2002). Thus fillings removed for replacement today were on the average performed in approximately 1988. In 1992-93, dental mercury consumption was about 1,800 kg/year (Maag et al., 1996), whereas the figure in 1982-83 was about 3,100 kg/year (Hansen, 1985). On the expectation that about 50% of this mercury was filled in teeth, about 1,100-1,300 kg of mercury would have been used for dental fillings performed in 1988. Part of this mercury will be excavated with old fillings, part will accompany extracted or lost teeth, and the remainder will be in the teeth of the deceased, who are buried or cremated.

For 2001, the excavation of old fillings is estimated to account for 660-1,100 kg of mercury. This follows from the following figures (from the previous mass flow analysis):

Due to the great attention to the effects of mercury in teeth, there is reason to believe that some people have the amalgam in their fillings replaced with other materials, before the amalgam fillings are worn out. This means that the excavated quantity could exceed the estimated 940 kg. It is difficult to quantify this precisely, but for the purpose of this report, a maximum of 1,100 kg of excavated mercury has been assumed. For this reason, the excavation of old fillings is considered to represent 680-1,100 kg of mercury/year.

Collection of mercury from dental cuspidors
All dental chairs are equipped with a strainer that captures large particles. The strainer is manually emptied from one to five times a week (Marker, 2002; Lindstrøm, 2002). Its content - or at least the visible fragments of amalgam - are separated as mercury-containing waste. The residue is disposed of as ordinary waste or flushed down the sink where the sorting was carried out.

Amalgam filters
After the strainer, the cuspidor system may be equipped with an amalgam filter, which collects 90-99.9% (Bindslev, 1999; Marker, 2002) of the mercury in the wastewater. Most of Denmark's local councils have issued regulations on the use of amalgam separators that have a maximum discharge, e.g., of 5 g mercury/year. The Århus School of Dentistry has conducted questionnaire surveys in all of Denmark's local councils (Bindslev, 2002). It was apparent from the 239 responses that such regulations had not been issued by 27% of the local councils.

The largest filter manufacturer collects most of its filters itself. The filters are disassembled in Denmark and the content of the filters is sent to The Netherlands for processing. The filters are cleaned and reused. In Denmark the cleansing process occurs in a closed system, with recirculation of the water and with filters in the ventilation exhaust. The efficacy of the air filters has not been measured and they are so new that they have not yet been replaced. Based on our knowledge of the number of amalgam filters processed, the total amount of mercury released to air and water is considered to be between 1 and 5 kg.

Another filter manufacturer also collects used filters and part of the mercury-containing waste from dental clinics. This is sent to Sweden where the filters are cleaned for reuse and the mercury is processed. About 30% of the remaining mercury-containing waste from dental clinics is considered to be collected together with the filters and exported.

The last fraction of the filters are sealed when they are replaced and are left at the clinic. Depending on the local council's regulations, they are disposed of together with the clinic's other mercury-containing waste.

One collector of amalgam filters states that, when collected, each filter typically contains 250 g mercury. Although this figure is lower than the assessment of the previous mass flow analysis, it has been verified by the Swedish purchaser. Consultation with the filter manufacturers reveals that 2,900-3,800 filters are sold annually in Denmark. It must be assumed that a corresponding number is collected, i.e., that 690-950 kg of mercury is collected annually from amalgam filters. Of this, an estimated 480-575 kg is exported to The Netherlands and Sweden where it is processed. The remainder is collected and delivered to Kommunekemi.

Discharges to wastewater
Assuming that amalgam filters are 98% effective, 15-20 kg of mercury is discharged to wastewater from dental clinics equipped with amalgam filters. In addition are the discharges from dental clinics that lack such filters, which can be estimated at about 175-240 kg. Thus, the total discharge from dental clinics to wastewater can be calculated as 190-260 kg of mercury/year. Here, it has been assumed that 80% of Danish dental clinics are equipped with amalgam filters and that the total discharge from the remaining 20% of clinics is in proportion thereto. About 27% of Danish local councils lack regulations on filters. This circumstance is also assumed to apply to dental clinics, but as many dental clinics in these areas nevertheless are considered to have filters, 20% of clinics are thought to lack filters. However, there is significant uncertainty associated with this figure.

For the sake of comparison, an investigation of a limited number of clinics, conducted on a single day in Århus, revealed a discharge of 65-842 mg of mercury/dentist (corresponding to 7-185 g of mercury/year) into the municipal sewers from clinics without filters. At clinics with amalgam filters the discharge was 12-99 mg of mercury/dentist (2.6-22 g of mercury/year) (Arenholt-Bindslev, 1996). If this is assumed to be representative of 20% and 80%, respectively, of all approximately 5,000 dentists in Denmark, it corresponds to a discharge of 20-270 kg of mercury/year. This corresponds very well - considering the paucity of data - to the above estimate.

To summarise, the following assessment is made of discharges to wastewater from dental clinics:

A discharge of 20 kg is based on a limited number of measurements and must be assumed on the basis of other available data to be too low, whereas a discharge of 50 kg of mercury/year is not unrealistic. On the other hand, a possible total discharge of as much as 250 kg/year cannot be dismissed. The discharge to wastewater from dental clinics is therefore assessed as 50-250 kg of mercury/year.

Amalgam in extracted and lost teeth
By far the greater part of the teeth lost by adults are extracted by their dentists, whereas only a small number are actually lost. This report considers the number of teeth lost by adults to be insignificant.

According to National Health Service data for the number of teeth extracted annually, 354,519 were extracted in private practice (i.e., teeth that count as adult teeth, see Appendix 1). It must be considered likely that a significant number - perhaps the greater part - of all these teeth contained amalgam fillings. If, as in the previous mass flow analysis (Maag et al., 1996), it is assumed that the average weight of these fillings is 1 g (amalgam), they account for an annual 100-150 kg of mercury.

The quantity of mercury lost and extracted during the years children are treated under the school dental service must also be added to the above figure. However, as the use of mercury in dental fillings has been prohibited since 1 January 1995, with the exception of fillings in molars and premolars, where they are subject to heavy wear, many clinics have stopped using mercury in milk teeth. As most of the teeth extracted and lost in these age groups are milk teeth (mercury consumption for new fillings in milk teeth is about 2.4 kg/year), and only a limited number is estimated to consist of permanent teeth (mercury consumption for new fillings in permanent teeth is about 100 kg/year), the quantity of mercury from teeth lost or extracted under the school dental service is estimated at 20-30 kg/year. In the previous report, when the use of mercury in milk teeth was more widespread, this was estimated at about 250 kg/year.

Thus, it can be estimated that a total of 120-180 kg of mercury/year is disposed of together with extracted and lost teeth. Of this, it is estimated that 20-30%, i.e., about 20-50 kg of mercury, enters the waste stream (refuse), whereas the remainder, 70-160 kg, will be passed to Danish schools of dentistry (which collect teeth for educational purposes), to Kommunekemi or, together with amalgam filters, to the filter manufacturers.

Amalgam in the teeth of the deceased
The previous mass flow analysis calculated with an average of 2 g of mercury in the teeth of the deceased in the Western world. Compared to earlier times, far more people retain their teeth until they die. According to Marker (2002), it is reasonable to calculate with 4 g mercury/deceased.

About 58,000 people died during the course of 2001, of which about 41,000 were cremated and 17,000 were buried (Hansen, 2002). This means that about 170 kg of mercury is passed to Danish crematoria, while about 70 kg is buried in cemeteries. As the crematoria are not equipped with flue gas cleaning equipment capable of capturing mercury vapour, it is to be expected that all of the mercury passed to the crematoria is emitted to the atmosphere.

Mercury-containing waste
Mercury-containing waste consists of excess amalgam from strainers and, possibly, extracted teeth containing amalgam fillings. Typically, this waste is stored by dentists in a water-filled container, until it is delivered to the local receiving station for chemical waste. Individual collectors of filters also collect mercury-containing waste, although there are no longer any actual collectors of excess amalgam or mercury-containing waste (Due, 2002; Metaligen, 2002; Petersen, 2002). The reason for this is stated as being the small quantities, low price, and the complicated procedures that apply when transporting hazardous waste. These factors make it extremely difficult to estimate quantities. Based on the sold and excavated quantities of dental mercury and the quantity in wastewater and amalgam filters, the total quantity of mercury in this waste fraction is estimated at 120-680 kg/year.

Employees at dental clinic are considered to be aware that mercury-containing waste must not be disposed of in refuse and it is assumed that the main part of mercury-containing waste (calculated as 60-80%) is collected separately, whereas the remainder is assumed to enter the refuse stream. In other words, about 50-140 kg of dental mercury enters the refuse stream, whereas 70-540 kg is collected.

About 30% of collected mercury-containing waste is exported together with collected filters destined for export. This corresponds to 20-160 kg of mercury/year. The remainder is sent to Kommunekemi. The mercury balance for dentists is shown in Fig. 2.1. It should be noted that no attempt has been made to estimate the emission of mercury vapour to air from dental clinics. Neither has any attempt been made to quantify the amount of mercury which, together with cotton pledgets and similar low-contamination waste from dental clinics, can be conceived of as entering the refuse stream. There is no information to give reason to believe that these quantities of mercury are of any significance.

Fig. 2.1 Estimated circulation of mercury within Danish dental care in 2001 (calculated as average, rounded numbers, kg of mercury)

Click here to see Fig. 2.1

2.2.2 Light sources

Most electric light sources of the discharge type contain a small quantity of mercury. Discharge light sources function through the application of a high voltage field across a gas, normally mercury vapour, whereon the gas emits light. For instance, ordinary fluorescent tubes function on this principle. Inside fluorescent tubes, the emitted light strikes the fluorescent powder coating on the inside of the tube. The constituent substances of the fluorescent powder then emit light in the desired colours. Fluorescent powder consists primarily of chemical compounds of the elements known as "rare earths".

Table 2.4 shows the numerically most important light sources that contain mercury. For each type and for various manufacturing years, the table also lists the quantity of mercury/light source

Table 2.4 The numerically most important light sources containing mercury and the mercury content of each type.

1) Source: Floyd et al., 2002. The mercury quantities for the individual years have been interpolated from the information received.

2) The previous mass flow analysis lists 5-15 mg of mercury/tube. This is based on information given by manufacturers, waste collectors, and various reports.

Mercury is also used in the following light sources, which are not included in Table 2.4:

- super-high-pressure lamps for photographic purposes;
- mercury vapour lamps for chemical analyses (atom absorption) - so-called "quartz lamps"
- UV sterilisation lamps
- LCD screens for computers and televisions
- tubes in advertising signs (see below).

As sales of these light sources are small, their mercury content makes only a marginal contribution to the total consumption of mercury.

Manufacturing in Denmark
Fluorescent tubes for advertising are manufactured and repaired in Denmark. The consumption of other types of light source depends solely on imports.

Consumption in Denmark
Table 2.5 shows the consumption (thousands of lamps) during the period 1997-2001. The mercury consumption of the same lamps in Denmark is estimated in Table 2.6.

Table 2.5 Consumption of mercury-containing light sources during the years 1997-2001 (lamps x 1,000). 1)

1) Information on the numbers is based on information from Statistics Denmark. Discharge lamps consist, e.g., of mercury vapour lamps, sodium vapour lamps and metal-halide lamps.

Table 2.6 Consumption of mercury in light sources during the years 1997-2001 (kg of mercury). 1)

1) Based on Tables 2.4 and 2.5.

As can be seen from Table 2.6, the total consumption of the lamp types shown in the table for 2001 is estimated at 105 kg of mercury. For this report, it is realistic to calculate with an uncertainty of ±50 kg, based on the authors' assessment.

Specially formed advertising tubes
According to information provided by a number of manufacturers of advertising tubes, consumption for tubes today can be roughly estimated at 1-5 kg of mercury/year. In the previous report for 1992-93 mercury consumption for advertising tubes was listed as 15 kg/year. This market is declining. The mercury is added manually using a thin glass pipette. One small drop is added to each fluorescent tube. Today, some advertising signs are manufactured with diodes and optical fibre. These technologies are advancing rapidly.

LCD screens for computers and televisions
Flat screens, so-called "LCD screens," for computers - also including screens for portable computers - contain mercury-containing light sources (LCD: liquid crystal display). The same applies to LCD screens for televisions, although they are still only assumed to have very limited penetration in Denmark.

According to Floyd et al., 2002, screens for portable computers typically contain 2-4 lamps/screen, depending on the size of the screen. A 15" screen will typically contain four lamps, whereas larger screens can contain 6, 8 or more lamps. In 2001, the typical mercury content of each lamp was between 2.5 and 3.5 mg/lamp (the present requirements for affixing the EU ecolabel to LCD screens state that the mercury content must not exceed 3 mg/lamp).

Sales of portable computers in Denmark amounted to about 75,000 in 1997 (IDC, 1999).

Sales of flat screens for computers can be estimated from the total use of stationary personal computers, i.e., about 500,000 (figure for 1997, IDC, 1999), the assumption that approximately one screen is sold for each personal computer and an assessment of the distribution of new sales of screens over conventional CRT screens and flat screens. One large retail chain estimated that, in 2001, this distribution was about 75% CRT screens and 25% flat screens, but that the distribution shifted very rapidly during 2002, so that the ratio was estimated at 60/40 at the end of the year. This shift towards flat screens is expected to continue.

Sales of televisions with LCD screens are, as mentioned above, very limited (due to the extremely high prices), and no attempt has been made to quantify the mercury consumption of these screens.

Thus, the consumption of mercury for flat screens can be estimated as follows:

portable computers: 75,000 pieces * (2 to 4 lamps) * 0.000003 kg/lamp: 0.5-0.9 kg of mercury;

flat screens for stationary personal computers: 0.25 * 500.000 pieces * (4 to 6 lamps) * 0.000003 kg/lamp: 1.5-2.3 kg of mercury.

As sales of both types of product can have been a little higher in 2001 than in 1997, and as uncertainty is also associated with the numbers of lamps and the mercury content of the lamps, the total consumption of mercury for flat screens in 2001 is estimated at 2-5 kg.

Light sources in vehicles
A major manufacturer and dealer reports that mercury-containing lamps are seldom used in vehicles. A qualified estimate is 1% of vehicle sales and a mercury content of a few milligrams/lamp. About 115,000 cars were sold in Denmark in 2001 (Statistics Denmark, 2002). The assumption that 1% of the cars sold in 2001 were equipped with mercury discharge lamps yields 1,150 cars.

Each lamp contains between 5 and 10 mg of mercury (Ecology Centre, 2001). If we assume six lamps/vehicle, we obtain a total mercury consumption of between 35 and 70 g. We should also mention that mercury-containing lamps are being used to an increasing extent, even in relatively inexpensive vehicles (Ecology Centre, 2001).

Discharge lamps used for back-lighting displays on the instrument panel also contain mercury (in similarity to LCD screens), and substitution is not possible today (Lohse, 2001). The quantity of mercury in the individual units is not known and is highly variable. An American study states that 1.2 mg is used for a navigation system and less than 40 mg, for a speedometer. The same study also shows the vehicle models of the year 2000 that are equipped with discharge lamps. The list is probably not exhaustive, but it includes a number of vehicle manufacturers that are known on the Danish market. This is, however, a question of luxury vehicles. Bearing in mind this consideration, we shall assume that 1% of the vehicles sold in 2001 were equipped with discharge lamps, and if we also assume that each vehicle contains 40 mg of mercury, the total consumption is 46g.

Total consumption
The total consumption of mercury in light sources is, thus, estimated at 60-170 kg of mercury/year, of which 100 ±50 kg is for the most common types (see Table 2.6), 1-5 kg for specially-formed advertising tubes, 2-5 kg for LCD screens (flat screens) and less than 1 kg for backlit display panels in vehicles.

Development trends
The consumption of mercury for light sources has dropped since the previous mass flow analysis, in which it was assessed as 170 kg in 1992. The drop to about 115 kg (the average of 60-170 kg) for 2001 principally occurred because the mercury content/light source is lower today. The consumption of linear fluorescent tubes dropped by almost 2m from 1999 to 2001. The consumption for 1988-1992 was 5-6m tubes. On the other hand, the consumption of energy saver lamps increased, from 0.6m in 1998, to 5.7m in 2001.

Losses in manufacturing
As has already been mentioned, the consumption of mercury for Danish manufacturing of light sources (advertising tubes) is very small. Manufacturing losses are considered to be insignificant.

Losses in conjunction with disposal
The quantity of mercury disposed of with light sources in 2001 can be estimated at about 140 kg (uncertainty interval: ±40 kg). This estimate makes allowance for the fact that most mercury-containing light sources probably have a life-cycle of 8-10 years, and that mercury consumption for light sources for 1992 was assessed as being about 170 kg (Maag et al., 1996). The previous mass flow analysis calculated with a life-cycle of "3-4 years, or more." In this report, the life-cycle has been estimated at 8-10 years as this is the life-cycle usually applied for fluorescent tubes, energy saver lamps and vapour lamps.

Depending in part on local waste-collection schemes, burned-out energy saver lamps and other small light sources are expected mainly to be collected with refuse, whereas other types of light source are collected separately, normally through schemes for bulky waste.

There is a recovery plant in Denmark that processes straight fluorescent tubes. In this plant the metal ends are cut off and sent to smelting works, the mercury-containing powder is blown out and returned to the original manufacturers of fluorescent tubes for reuse. The glass is cleaned and sent to glassworks, so that it can be reused in new light sources. Thus, more than 98% of these fluorescent tubes are reclaimed. The entire process occurs in a closed system, with activated charcoal filters on the output streams. This plant processes about 2.6m straight fluorescent tubes a year. About 70% of the fluorescent powder from the plant can immediately be reused in manufacturing new tubes. Manufacturing takes place abroad. The remainder (about 30%) is sent to Belgium for processing (distillation). 2.6m fluorescent tubes, each containing about 15 mg of mercury, corresponds to 40 kg of mercury. In addition, there is the exporting of about 500,000 straight tubes from other collectors, as well as broken tubes and other mercury-containing light sources. Based on information on consumption of the different types of light source, the author considers this to correspond to about 10 kg of mercury.

According to information provided by Kommunekemi, 156,258 kg of mercury-containing light sources were received in 2001. Fluorescent tubes weigh 50-300 g each and contain about 15 mg of mercury/tube, energy saver lamps weigh 40-100 g and contain 5-10 mg of mercury, while vapour lamps weigh 30-200 g each and contain 25-39 mg of mercury. Assuming that the light sources weigh an average of 100-200 g and contain 15 mg of mercury each, 10-20 kg of mercury were delivered with light sources to Kommunekemi in 2001. Light sources are deposited at Kommunekemi's site with a view to subsequent processing abroad (Naamansen, 2003). In this report, this quantity has been counted with exports.

Altogether, it is estimated that 60-80 kg of mercury (10-20 kg of which is delivered to Kommunekemi) is collected and exported, whereas 20-120 kg of mercury ends in refuse.

Some of the light sources are broken during handling - especially during the collection and sorting of the tubes. This occurs under normal, unfiltered, ventilation conditions. When the light sources are crushed, mercury is emitted to the air. Although there are no figures for the quantity of broken light sources, it does not seem unreasonable to assume that 1-5 of every 100 light sources are broken during the time from when they enter the country until they are incinerated or re-exported. 1-5% of the potential of 100-180 kg of mercury corresponds to 1-9 kg, i.e., the discharge to air resulting from broken light sources can be roughly estimated at 1-9 kg of mercury/year.

Disposal summary:

2.2.3 Electrical switches, contacts and relays

Mercury-containing electrical switches, contacts, and relays have been used in many different situations over the years. Table 2.7 shows the uses of mercury-containing switches, contacts and relays for which, on the basis of available information, application is found, has been found, or could have been found in Denmark within a reasonable period.

Table 2.7 Use of mercury-containing switches, contacts, and relays

The following review of the main applications is focused on the magnitude of consumption. For a more detailed description of the design and modes of operation of different types of mercury based switches, contacts and relays, see (Rasmussen, 1992). However, a brief description of the predominant types is included here. Conventional reed-relays consist of two electrodes mounted in a sealed glass tube. The electrodes are activated by an external magnetic field. In this type of relay, the mercury - or other metal, such as ruthenium or rhodium - is applied to the surfaces of the electrodes in a thin layer. This ensures the maintenance of a smooth surface because these metals, in contrast to others, do not form uneven surfaces, thereby establishing poor contact or causing arcing. Reed-relays are not to be confused with mercury switches, in which a larger quantity of mercury is free to roll around in a glass tube and, at certain angles, to form the conductor between the electrodes that are fused into the wall of the glass tube.

Domestic appliances
As early as 1994, DEMKO stated that mercury-containing thermal cut-outs were not used in domestic appliances, such as coffee machines and flat irons (Maag et al., 1996). The extent to which mercury has been used in these applications and when such use was discontinued is unknown. The discontinuation of such uses could be related to a German prohibition on the use of mercury in a number of household products in the mid-1980s (Rasmussen, 1992). Waste collectors say that they were still receiving, e.g., coffee machines containing mercury, during 2001.

Submersible pumps
A major manufacturer of submersible pumps states that, due to environmental considerations, it discontinued the use of mercury around 1996, and that this generally applies throughout the industry.

The consumption of mercury-containing level switches for submersible pumps, which have a mercury content of 6.8-13.6 g/pump, corresponding to a mercury consumption of 140-400 kg/year, was estimated at 20,000-30,000/year for 1982-83. Sales of submersible pumps for sewerage applications have dropped slightly and, according to the leading manufacturers in 1994, sales were marginal. However, level switches were manufactured in small quantities, although the manufacturer and quantity were not known (Maag et al., 1996).

The greater part of the sales of level switches is thought to be for use in sewerage applications. It is illegal to use mercury-containing level switches in clean water supplies. One supplier has stated that an average mercury content of 10 g/level switch is to be expected. Thus, we estimate that the annual mercury consumption from 1982 to 1993 varied between 200 and 300 kg of mercury (Maag et al., 1996).

Cars
Historically, cars have contained a number of mercury-containing components: ABS brakes, airbags, switches, lamps and instrument-panel displays. However, the use of mercury in certain components is gradually being phased out.

A major importer and dealer in cars is of the opinion that mercury is no longer used in ABS braking systems and airbags. This opinion is confirmed by a trade organisation (Rasmussen, 2002).

According to a 1993-94 assessment, less than 1% of all cars sold contained mercury-containing components. With respect to airbags, it was already known that mercury-containing sensors were being phased out.

The assumption that mercury is no longer used in ABS braking systems and airbags is reinforced by a study undertaken by the Danish EPA of mercury switches in cars; the study lists which car models use mercury switches and the year of manufacture of these cars. The most recent year mentioned is 1996. It is also seen that a number of makes that are well-known in Denmark are completely free of mercury switches (Grau, 2002; Sharp, 2003).

As of 1July 2003, an EU directive prohibits the sale of new cars containing heavy metals. A number of components listed in Annex II of the directive (European Parliament, 2002) are, however, exempt from this prohibition. As far as mercury is concerned, the relevant devices are discharge lamps and instrument-panel displays; this is apparent from a Commission decision (The Commission, 2002). The directive is implemented in a statutory order that sets requirements for the removal of mercury switches from cars before they are scrapped (The Danish Ministry of Environment, 2002).

Remote control of rail traffic
Mercury wetted relays, which are still permitted under the statutory order, can be used for repairs to older equipment. The mercury consumption in this application is not known (Danish National Railway Agency, 2003), but is only considered marginal. Mercury wetted relays are now substituted in new equipment. Instead, modern PLC equipment is used, in which microprocessors have rendered the relays obsolete.

Television sets
In the previous mass flow analysis of mercury, it was clear that the mercury in televisions was used in the cathode ray tube and in relays. According to a major manufacturer of cathode ray tubes, mercury has never been used in cathode ray tubes, and the use of mercury in relays ended in the 1980s. A processor of electronic waste has stated that the number of scrapped televisions containing mercury in 2002 was only marginal - roughly one/week - out of a total of about 3,000. The mercury content of the relays in televisions was estimated at less than 4 mg/set (Maag et al., 1996). As in the previous mass flow analysis, the mercury consumption for television sets is estimated at less than 2 kg.

Computing equipment
Personal computers are not expected to contain mercury-containing relays in quantities that significantly exceed the quantities found in older television sets. As far as is known, such relays were used at the start of the 1990s, primarily in connection with data transmission, i.e., modems. Today, the conventional reed-relays in modems have probably been replaced by microprocessor technology. At the same time, most types of reed-relay are coated with materials other than mercury. Even if there are, in fact, one or a few mercury-containing reed-relays in every modem, the mercury consumption would scarcely exceed 5 kg/year (according to information collected for the previous mass flow analysis, a mercury content of the order of 0.1-10 mg of mercury/relay; according to information on sales of personal computers (IDC, 1999), sales of modems are roughly estimated at about 500,000/year).

Miscellaneous electronics
The use of mercury in electronics in general is considered to be marginal for the same reasons as were given for computing equipment. The report, Elektronik - Erfaringsopsamling 1999 ("Electronics - Collected Experience 1999", Faber et al., 1999), states a mercury content of 0.0009% in printed circuit boards (based on information from the mid-1990s). Based on this information, the total consumption of mercury in relays used in miscellaneous electronic equipment in Denmark, including imported finished goods, is considered scarcely to exceed 10 kg/year.

Industrial plant
The Danish EPA has granted exemption for the sale of special mercury-containing rotating contacts in industrial plants. Such contacts are used when transmitting electrical power and signals to and from rotating machines. This exemption applies to a limited quantity until the end of 2002. The consumption for 2001 is assessed as being about 2 kg.

Flashing lights for rail traffic
Mercury-containing flashing light systems are manufactured in Denmark for the control of rail traffic and level crossings. Such use is still permitted. In 2001, 19 mercury tubes were consumed, each with a mercury content of 10 g. Thus, the total consumption is 190 g. This consumption varies from year to year, but 2001 was an average year in this respect. The consumption for the next five years is also expected to be around 190 g/year. This is significantly less than consumption in the same application in 1992-93 (50-100 kg of mercury).

Burglar alarms
The use of mercury-containing switches in burglar alarms for motorcycles and video tape recorders is considered to be marginal in Denmark and, as in the previous mass flow analysis, must be calculated as being less than 5 kg/year (Maag et al., 1996).

The consumption of gas chromatographs equipped with mercury-containing thermal cut-outs is only marginal in Denmark. Such equipment is only used in laboratories and has a relatively long life-cycle of 5-10 years. The consumption of mercury in this application is roughly estimated at less than 1 kg/year (as in the previous mass flow analysis; Maag, 1996).

Flashing lights in running shoes
Running shoes equipped with flashing lights controlled by mercury-containing switches were marketed in Denmark at the beginning of the 1990s, although imports stopped in the mid-1990s (Maag et al., 1996). A major importer confirms this and states that only shoes without mercury are imported (Bech, 2002). Similarly, Floyd et al., 2002, mention that substitutes have now been found for the mercury-containing switches in flashing shoes for children.

Total consumption
In summary, the total consumption of mercury in switches, contacts and relays can be assessed as being less than 24 kg/year.

Development trends
All information indicates that the use of moderately large mercury-containing switches and relays has been discontinued - or largely discontinued - with the possible exception of minor consumption for repairs to older equipment (cf. use in the remote control of rail traffic). On the other hand, it is difficult to assess the prevalence of mercury-wetted switches/relays although, as mentioned above, this is only a question of very small quantities of mercury/device. According to information from suppliers, reed-relays with mercury-wetted contacts can still be purchased although many types use other contact materials, and it is presumed that such relays are being replaced by microprocessor technology in many applications. As mentioned above, mercury wetted contacts and relays are still exempt under the Danish statutory order on mercury.

Disposal
The loss of mercury through the disposal of products is difficult to assess with any certainty, as such losses are mainly associated with applications that have been discontinued today. Any assessment of this loss must, therefore, make allowance for the historical magnitude of consumption, as well as for the probability that mercury-containing products are still in use.

Around 1992, the total consumption of mercury for electrical contacts, switches and relays was of the order of 200-400 kg/year, of which 200-300 kg was for submersible pumps. The total consumption does not include telephones, many of which apparently have a life-cycle in excess of 20 years, as used telephones and telephone exchanges are the source of most of the relevant collected mercury in switches, contacts and relays (see below).

Many of the products listed in the table may still be in use or in storage, from where they will gradually be discarded (the "box-room" effect). A long life-cycle is expected, for instance, for the mercury-containing switches in submersible pumps.

According to one waste collector, about half of all the freezers collected are equipped with mercury-containing switches. This collector, who receives about 20% of electronic waste in Denmark, collected about 10,000 freezers in 2001. If an average mercury content of 5 g/switch is assumed, this corresponds to a collected annual quantity of 125 kg of mercury from freezers in Denmark.

According to waste collectors, old telephones and telephone exchanges are one of the largest sources of mercury-containing contacts, switches, and relays in electronic waste. In recent years the telephone company TDC has, itself, collected an annual 800-1,000 kg mercury-containing contacts, switches and relays from telephones and telephone exchanges. All telephone exchanges have now been replaced by digital exchanges that contain no mercury, although some of the old, analogue installations are still to be found "inactive" around the country. On the assumption that the mercury content is 70-90% of the weight of the contacts, switches and relays, an annual 600-900 kg of mercury is collected by TDC in telephones and telephone exchanges. In addition, there is the mercury from telephones processed by collectors of electronic waste. If this is assumed to amount to 10-20% of the quantity collected by TDC, an annual total of 650-1,000 kg of mercury is collected with telephones and telephone exchanges. This is exported.

In addition, there is the mercury exported with telephones and telephone exchanges for reuse abroad. For the past three years, complete telephones have been exported for reuse. Such telephones were scrapped in earlier years. According to the exporter, about 10% of the exported telephones are considered to contain mercury. Calculating with 250 kg containers and assuming an average weight of 0.75-1 kg/telephone and an average content of 4-5 g of mercury/telephone, an annual quantity of 50-90 kg of mercury/year is exported in complete telephones.

Waste collectors state that printed circuit boards account for 9-10% of the total quantity of electronics (about 20,000 tonnes, according to the collectors of electronic waste). Assuming a mercury content of 0.0009% (Faber et al., 1999; Rambøll, Hannemann & Højlund A/S, 1995), this corresponds to 14-16 kg of mercury.

Certain old models of coffee machines contain 10-20 g mercury/machine. Based on information from a single waste collector, we consider that about 100 such coffee machines are collected every year. This corresponds to an annual 1-2 kg of mercury.

In addition, there are the previously mentioned contacts, switches and relays, which are considered to constitute a minor proportion of the collected contacts, switches and relays. The author is of the opinion that it is not unrealistic to calculate with 50-150 kg of mercury/year in other types of contact, switch and relay.

Summary of the annual collection of mercury-containing contacts, switches and relays:

All of this is exported; however, Kommunekemi has a temporary store destined for later export for re-processing (Naamansen, 2003; Hohberg, 2003).

It is probable that some of the mercury-containing contacts, switches, and relays end in refuse. The previous mass flow analysis estimated that <160 kg of mercury ends in refuse. This was a question of mercury in telephones, coffee machines, etc. As can be seen from the foregoing, the quantity of mercury collected with telephones is considerably greater than was previously thought to be the case. The life-cycle of other mercury-containing contacts and switches is also possibly significantly longer than 10 years, 20 years is perhaps more likely. The consumption of mercury in electrical equipment was assessed as being 160-520 kg/year in 1982-83. Less than 100 kg of this was in telephones (Hansen, 1985).

Mercury has also been found in the output of incinerator plants, i.e., mercury is still discarded in refuse. As a rough estimate, the quantity of mercury entering refuse is assessed as being 1030% of the collected quantity. This corresponds to 80-400 kg of mercury.

2.2.4 Thermometers

During the last decade, mercury-containing thermometers were principally sold as clinical thermometers and thermometers for controlling industrial processes, mainly the diesel engines of large ships (Maag, 1996). Today, however, mercury-containing thermometers have largely been substituted by electronic thermometers and thermometers containing non-mercury-containing metal mixtures.

Since 1994, it has been prohibited in pursuance of the first mercury statutory order to sell mercury-containing thermometers other than special-purpose thermometers (Ministry of Environment, 1994). It is also apparent from a letter that the Ministry of Health states in a statutory order on medical equipment that equipment, which has been subjected to EEC type approval, can still be marketed (Rüsander, 1994).

The most recent mercury statutory order, of 1998, specifies that such exemption covers (Ministry of the Environment, 1998):

• thermometers that have been subjected to EEC type approval (in conformity to Directive 76/764/EEC and subsequent amendments) until 30 June 2004 (approved clinical thermometers);
   
• special-purpose thermometers: marine industry, until 31 December 1999; combined heat and power stations, until 31 December 1999; calibration of other thermometers, until further notice, analytical equipment, until further notice.

Based on this information, we assume that thermometers that have been subjected to EEC type approval have been on the market throughout the period.

Clinical thermometers
The use of clinical thermometers containing mercury has dropped dramatically over the past 10 years. A large drop at the beginning of the 1990s was apparent as early as in the previous mass flow analysis. This is attributed to growth in the use of electronic thermometers and, possibly, glass thermometers filled with a liquid metal mixture consisting of gallium, indium, and tin (Maag et al., 1996).

Mercury-containing clinical thermometers are not manufactured in Denmark.

An importer of clinical thermometers states that pharmacists no longer purchase mercury-containing thermometers. The Danish Pharmaceutical Association confirms this, but cannot rule out that a few pharmacists purchase mercury-containing thermometers. The policy of not purchasing mercury-containing thermometers was introduced at the beginning of the 1990s, in anticipation of a ban (Riis, 2003).

The fact that mercury-containing clinical thermometers were largely sold to hospitals and other institutions was apparent from the previous mass flow analysis (Maag et al., 1996). We assume that mercury-containing clinical thermometers are no longer purchased for private use, although many such thermometers are probably still in use.

Importers of mercury-containing clinical thermometers for use in hospitals state that about 4,500 mercury-containing thermometers were sold in Denmark in 2001. Consumption of these items has fallen sharply - the corresponding figure was about 50,000 ten years ago. This information agrees with the account given in the previous mass flow analysis. According to importers, an average mercury content of 0.25 g/thermometer can be expected today. The reason for this drop in the quantity of mercury is that the glass industry is now able to manufacture capillary tubes more precisely. The total mercury consumption can be calculated as more than 1 kg.

Other types of thermometer
Consultation with a number of importers and dealers in thermometers for laboratory use reveals about 1,200 recorded sales of mercury-containing thermometers/year. As it is probable that there were some sales from other sources, it is estimated that a total of 1,000-1,500 mercury-containing laboratory thermometers were sold during 2001. These thermometers are expected to contain about 15 g mercury (based on information from suppliers and Norsk Renholds Verks-forening, 2001). This corresponds to a mercury consumption of 15-23 kg/year.

Mercury-containing thermometers are no longer manufactured in Denmark.

Importers state that many thermometers for use in analytical work contain spirit today, although mercury-containing thermometers are still used to a certain extent.

Disposal of clinical thermometers
In the case of the clinical thermometers used in hospitals and institutions, it is to be expected that practically 100% of the broken and discarded thermometers are collected and disposed of as chemical waste. In cases where it is not possible to collect all of the metallic mercury from crushed thermometers, it must be assumed that the residue is collected, together with the wastewater from washing floors.

As the turnover of thermometers in institutions is considered to be quite fast (Maag et al., 1996), it is to be expected that the quantity destined for disposal more or less corresponds to the present consumption of about 1 kg of mercury, whereas a tiny quantity ends in wastewater and refuse.

In the case of clinical thermometers used in private households, it is logical to expect that only a modest proportion of the mercury is collected as chemical waste. In such case, it is probably a question of whole thermometers that are handed in to pharmacists. Other disposal alternatives include losses to refuse (whole thermometers, broken thermometers and mercury, which have been collected or picked up from floors by vacuum cleaners) and wastewater (from washing floor).

The life-cycle of clinical thermometers used in households can be significant, as many thermometers are only used a few times a year, or even less frequently. It is therefore assumed that the average life-cycle of such thermometers is close to 10 years.

Based on Maag et al., 1996, the consumption of mercury from clinical thermometers in households in 1991 was approximately 70 kg (1990, 100 kg; 1992, 20 kg and; 1993, 0 kg). To make the best estimate on the basis of the information available, it should here be assumed that the quantity disposed of in 2001 is equal to the quantity consumed in 1991, broken down as follows (corresponds to the previous mass flow analysis, in which Kommunekemi investigated received waste more closely):

Disposal of other types of thermometer
In 1992-93, about 100 kg of mercury was also used in industrial thermometers and suchlike (Maag et al., 1996). In the previous mass flow analysis, laboratory thermometers were not accounted for separately, but were included in this total. In the case of industrial and laboratory thermometers, it also seems reasonable to assume an average life-cycle of about 10 years, even if there is probably considerable variation, depending on the application in question.

The disposal of industrial thermometers is probably mainly determined by the type of equipment in which the individual thermometer is used (mainly in motors). Mercury in such installations must primarily be expected to be collected by the recycling industry and secondarily, to be lost in wastewater and the waste stream through the dismantling of motors, etc.

As the best estimate on the basis of available knowledge, it will be assumed in this report that the quantity of mercury disposed of together with industrial and laboratory equipment corresponds to the consumption in 1992-93, broken down as follows (same proportions as in the previous mass flow analysis):

Disposal, total
Thus, we estimate that the following quantities of mercury from thermometers were disposed of in 2001:

NB: it goes without saying that there is considerable uncertainty in these estimates, and they should only be understood as an attempt to estimate the order of magnitude of the disposal of mercury from thermometers in Denmark.

2.2.5 Monitoring equipment

Sphygmomanometers
According to information given by a number of suppliers, an annual total of about 170,400 mercury-containing sphygmomanometers are still sold in Denmark. They have an average mercury content of about 5 ml, i.e., 70 g mercury. This corresponds to a total consumption in new sphygmomanometers of 12-28 kg of mercury/year.

Suppliers also state that by far the majority of these are sold to doctors in private practices. The hospital sector has substituted later models without mercury for mercury-containing sphygmomanometers to a greater extent than doctors in private practices. Sales to individuals with no medical background are considered only to consist of sphygmomanometers not containing mercury.

Older mercury-containing sphygmomanometers are only refilled to a very limited extent during calibration/maintenance. Recycled mercury taken from old sphygmomanometers is exclusively used for this purpose - only about 1 kg of mercury/year is used for this purpose.

Miscellaneous equipment
As was the case in the previous mass flow analysis, there is no certain information on the continued use of mercury in other monitoring equipment. However, the trend in consumption in recent decades has been strongly in favour of substitution by other technologies not containing mercury. For this report, a limited effort has been made to trace the continued use of mercury in these applications, although no sign has been found in connection with the collection of data that any significant use of mercury continues, e.g., in schools and industry. (In the case of industry, small quantities have been accounted for in other sections of this report.) This impression is also supported by the fact that the mercury statutory order and the requirement on permission to purchase toxic substances make the purchasing of mercury inconvenient and difficult. In principle, institutions of higher education could be an exception to this, but no specific information on this is available. The consumption of mercury in monitoring equipment is considered to be minimal and should be roughly estimated at <20 kg of mercury/year.

Disposal
In the previous mass flow analysis (data from 1992-1993), the consumption of mercury for monitoring equipment was assessed as having been reasonably constant for many years, at about 500 kg. Our immediate assessment is that the life-cycles of these applications can be very long, and that a certain level of mercury use is apparently maintained, e.g., the sphygmomanometers of doctors in private practices. The same is possibly true of schools' and other educational institutions' stocks of mercury, e.g., in physics laboratories.

Thus, it appears probable that the disposal quantity reflects earlier patterns of use, rather than present patterns. However, it should be emphasised that there is no certain information on the magnitude of such disposal with respect to monitoring equipment. As a rough estimate, it seems probable that the annual disposal quantities from these sources could amount to 100-500 kg of mercury.

This is a question of applications which, as a rule, are deemed to be covered by effective mercury collection schemes. However, it also appears likely that small quantities of mercury are lost to refuse, air and wastewater in connection with the handling of broken equipment. In view of the lack of information on this, we consider that roughly the following distribution of disposal and loss could be of relevance:

2.2.6 Other applications in metallic form

Lighthouses
The Royal Danish Administration of Navigation and Hydrography operates 12 lighthouses, which were built or rebuilt in the latter half of the 19th Century. The lighthouses are equipped with powerful, rotating systems of lenses. The lenses weigh from between a few hundred kg up to four tonnes each. These lenses float in a bath containing from 6 to 200 kg of pure mercury, which minimises friction.

Minor quantities of mercury evaporate from these baths, which contain a total of about 1,400 kg of pure mercury. Assuming an evaporation of 0.5%, <10 kg of mercury is annually emitted to the atmosphere.

The Royal Danish Administration of Navigation and Hydrography annually delivers about 40 kg of mercury to Kommunekemi. Assuming that this quantity is to be replaced by new mercury, the annual mercury consumption is estimated at about 50 kg. The uncertainty of the above figures is assessed at about 20%.

2.2.7 Summary - metallic mercury

The available assessments and knowledge of the intentional use of metallic mercury in Denmark during 2001 and the disposal of mercury in these contexts are shown in Table 2.8.

Table 2.8 Consumption and disposal of metallic mercury used for intentional purposes in Denmark in 2001 (kg/yr of mercury)

2.3 Use of mercury in chemical compounds

2.3.1 Batteries

Mercury is intentionally used in batteries of the types mercuric oxide, silver oxide, zinc/air and alkaline. An EU report (Floyd et al., 2002) shows the mercury content of the different types of battery. These figures apply to batteries sold in the EU. The figures have been confirmed by the Association of Danish Producers and Importers of Dry Batteries (Madsen, 2002). (In the following, the figures obtained from a Danish dealer are shown in brackets.) The mercury content of mercuric oxide batteries is about 32% (30-32%). The mercury in these batteries functions as the cathode. Silver oxide button cells contain about 0.34% mercury, alkaline button cells contain about 0.45% (0.2-0.5%) and zinc/air button cells, about 1.24% (1.1-1.2%). Mercury is added to these batteries to prevent self-discharge and to prevent gas from evolving during discharge, with resulting leakage or, in the worst case, explosion. All lithium batteries and rechargeable batteries have always been free from mercury (Floyd et al., 2002). About 70% of the European market for button cells consists of lithium button cells (Floyd et al., 2002). However, this does not apply to the Danish market, where zinc/air button cells are by far the most prevalent.

Import and the sale of batteries containing more than 0.0005% mercury by weight have been prohibited in Denmark since 1 January 2000. This prohibition applies to loose retail batteries and batteries included in products. However, button cells can contain up to 2% mercury by weight (Statutory Order No. 1044 of 16 December 1999 on certain batteries and accumulators containing dangerous substances). This statutory order is an implementation of an EU directive, i.e., corresponding prohibitions apply in the other EU countries.

Manufacturing in Denmark
Denmark has a limited production of alkaline batteries (not button cells). Based on information obtained from Statistics Denmark (2002b), and on the maximum permissible mercury content of this type of battery of 0.0005%, the consumption of metallic mercury for the manufacture of batteries in Denmark in 2001 is roughly assessed as a maximum of 9 kg/year.

Mercuric oxide batteries
This type of battery is no longer manufactured as it contains more than the permissible 2% of mercury. As the mercury is an active component of the electro-chemical system of these batteries, reduction of the mercury content is not possible in this type of battery (see The Association of Danish Producers and Importers of Dry Batteries website: www.batteri.dk). There is an annual consumption of about 1,000 mercuric oxide batteries in Denmark, see Table 2.9.

Loose retail batteries
Statistics Denmark gives some especially high figures for the importation of mercuric oxide batteries for the years 2000 and 2001 (Statistics Denmark 2001 and 2002). They have probably been incorrectly recorded.

The EPBA has calculated the consumption of different types of button cell for each country in Europe for the period 1996-2000. The Association of Danish Producers and Importers of Dry Batteries has subsequently estimated the consumption for 2001. The result for Denmark can be seen in Table 2.9.

Table 2.9 Consumption of button cells in Denmark (EPBA, 2001) and their mercury content (Madsen, 2001; Jochumsen, 2003)

It can be seen from the table that the consumption of mercuric oxide button cells continues, and that zinc/air batteries constitute the greater part of the button cells. Zinc/air button cells are almost exclusively used in hearing aids (Jochumsen, 2003). The total number of button cells consumed in Denmark was about 10,000/year, for the years 2000 and 2001.

Button cells weigh from 0.2-4 g each. The most used alkaline button cells weigh about 2 g each, whereas zinc/air batteries are often a little smaller, 0.6-0.9 g each (information from a supplier). Assuming an average weight of 1.5-2 g (but 0.6-0.9 for zinc/air) for button cells, the sales figures correspond to an average consumption of button cells of 10-14 tonnes during the period 1996-2001, whereas the average for 2000 and 2001 was 8-11 tonnes.

Table 2.10 shows the consumption of button cells converted into kg of mercury. No allowance has been made for the fact that the mercury content of the individual types of button cell can have dropped during the period.

Table 2.10 Consumption of mercury in button cells in Denmark

The annual consumption of mercury in loose button cells can thus be set to 58-100 kg. This corresponds to about 40% of the consumption in 1997. The drop since then is primarily due to a drop in the consumption of button cells containing mercuric oxide.

Large loose batteries
Batteries of types other than button cells can contain up to 0.0005% mercury. According to Statistics Denmark, about 3,000 tonnes of batteries other than button cells, lithium batteries and manganese dioxide batteries were consumed in 2001. With a mercury content of 0.0005%, this corresponds to about 15 kg of mercury. As many batteries are completely free of mercury today, 15 kg must be considered to be the maximum consumption.

Batteries supplied in finished goods
Mercuric oxide batteries are known to be used in many types of finished product, including hearing aids, photographic equipment, toys and clocks and watches.. For this reason, an attempt has been made to obtain information on the extent to which such batteries can be conceived of as being imported in finished goods.

Mercuric oxide batteries were previously used in hearing aids, but according to the Association of Danish Producers and Importers of Dry Batteries and Metaligen (Jochumsen, 2003), which supply batteries for all hearing aids in Denmark, today's zinc/air batteries can replace all mercuric oxide batteries for use in hearing aids (www.batteri.dk). Hearing aids have largely been designed to operate with zinc/air batteries for the past 20 years (Maag et al., 1996).

Mercuric oxide batteries used to be widely used as the power supplies of photographic light meters (especially in single lens reflex cameras) in earlier years. These devices are now over 20 years old (Petersen, 2002). It is possible that there are still occasional replacement sales of such batteries. Visits to five photographic dealers revealed that two of them still trade in mercuric oxide batteries, albeit in very small quantities. The batteries were purchased in 2000 or earlier.

There are significant imports of batteries in inexpensive electronic articles (watches, pocket calculators, greeting cards (which play melodies), toys and suchlike, bicycle lamps and bicycle computers). The button cells in the toys supplied by the largest toy dealer in Denmark are mainly alkaline. This dealer imports both the button cells and the toys direct from the Far East and has received guarantees from the manufacturers, stating that mercuric oxide batteries are not used (Top-Toy, Vastrup, 2002).

An inspection of goods supplied complete with button cells has been conducted at a large supermarket owned by one of the national supermarket chains, at a bicycle dealer and at a shop specialising in low-priced goods from the Far East. Most of the goods with batteries contained stick batteries. Only a small number of the goods contained button cells. Based on the type designations printed on the packaging, or on the button cells themselves, the button cells used in toys and bicycle computers are alkaline cells, whereas the button cells in bicycle lamps are of the lithium type.

Thus, as was the case in the previous mass flow analysis, no examples were found of imported finished goods that contained mercuric oxide cells.

Total imports of button cells included in finished goods can be estimated as follows (numbers of items taken from Statistics Denmark, 2001 and 2001b):

Pocket calculators (predominantly alkaline - net imports: about 860,000 items/year, of which the author considers it reasonable to assume that 80% contain 1.5-4 g of button cells/calculator, with a mercury content of 0.45% by weight) = 4.6-12 kg of mercury/year;

Wrist-watches and pocket watches (silver oxide, alkaline and lithium - net imports: about 1.34m, all containing 0.3-1.5 g of button cells/item, with a mercury content of 0.45% by weight) = 1.8-9 kg of mercury/year;

Alarm clocks (typically alkaline - net imports: 189,000, of which about 25% are considered to contain 1.5-4 g of button cells, with a mercury content of 0.45% by weight) = 0.32-0.85 kg of mercury/year;

Toys and greeting cards (alkaline - sales, according to information from suppliers and our own estimate: about 1.1m batteries, with a mercury content of 0.2-0.45% by weight) = 3.3-7.3 kg of mercury/year.

The total quantity of mercury in button cells included with finished goods imported into Denmark can be estimated at 10-30 kg of mercury/year.

With respect to the stick batteries found in imported finished goods, the manganese dioxide and alkaline types of battery (the mercury content of which, as in loose retail batteries, is assessed as being tiny) are considered by far the most prevalent. The quantity of stick batteries in finished goods is roughly estimated to constitute about 30% of loose retail stick batteries, i.e., it corresponds to a mercury consumption of less than 5 kg/year.

Total consumption
As of 2001, the total consumption of mercury in batteries in Denmark can thus be assessed at about:

Development trends
Today's closer scrutiny of mercury, together with the prohibitions on more than 2% of mercury by weight in button cells and 0.0005% of mercury in other batteries, has resulted in a dramatic drop in the consumption of mercury in batteries. This is partly because mercuric oxide cells are very rarely used now, partly because the mercury content of the individual batteries has been reduced and, partly, because some batteries no longer contain mercury (applies to many stick batteries). According to the Association of Danish Producers and Importers of Dry Batteries, the complete removal of mercury from button cells is difficult due to the risks of leakage and possible explosion.

According to Statistics Denmark, the consumption of button cells in 2001 remains at the same level as in 1992, i.e., about 15 tonnes, whereas the consumption of loose batteries other than button cells has more than doubled since 1992, from 1,600 tonnes to about 3,700 tonnes.

Disposal
Certain delays must be expected in the disposal, in relation to the consumption, of the mercury in batteries. In practice, mercury-containing batteries were prohibited in Denmark from 1 January 2000. This means that a proportion could very well still have been in use in 2001, or stored by users with a view to later disposal. Up to 1 January 2000, alkaline batteries other than button cells were also permitted to contain more mercury: 0.025%, in contrast to the present maximum of 0.0005%. Consumption of the mercury in mercuric oxide batteries in 1993 was estimated at 280-430 kg, while the consumption for 1994 can be estimated at 440-600 kg (440 kg was recorded by The Association of Danish Producers and Importers of Dry Batteries, plus an estimated maximum of 160 kg imported through other sources, according to Maag et al., 1996). The total consumption of mercury in batteries in 1992-93 was estimated at 400-860 kg/year (Maag et al., 1996).

Consumption of the mercury in batteries has dropped sharply over the course of the past decade, which is especially true of mercuric oxide batteries. Table 2.11 below contains the known information on the consumption of mercury in mercuric oxide batteries.

Table 2.11 Available information on the consumption of mercury in mercuric oxide batteries

Note:
1) The figures have been rounded to take account of their inherent uncertainty.

There is no certain information on the average life-cycle of mercuric oxide batteries (cf. the previous mass flow analysis), but 4-6 years is not an inconceivable period for the time between purchase and final disposal, bearing in mind the fact that the light meters of older photographic equipment (which have a very low current consumption) have been one of the most important applications for mercuric oxide batteries (Maag et al., 1996). If we accept the consumption figures of Table 2.11, the collection potential for 2001 alone could, thus, very well have been in the interval 160-400 kg/year.

In addition, there is the collection potential of the mercury contained in alkaline stick batteries; a rough estimate is 20-40 kg of mercury, based on the trend in mercury concentrations and consumption between 1993 (about 60 kg, according to Maag et al., 1996) and 2001.

The consumption of mercury in types of button cell other than mercuric oxide cells remained reasonably stable between 1993 (70-160 kg according to Maag et al., 1996) and the approximately 60-150 kg/year shown in Table 2.11 for the years 1997-2000. The collection potential is assessed on this basis as being of the same magnitude, i.e., 60-150 kg/year.

Based on the foregoing information, the total estimated collection potential for mercury from batteries can be summarised as follows:

That a significant quantity of mercuric oxide batteries - as well as other mercury-containing products - was disposed of as late as 2000-01 is supported by the fact that the quantity of mercury in refuse was still relatively high in comparison to the 2001 consumption of mercury (see Section 3.2.2, on combustible waste).

The batteries needed for hearing aids are paid for by the public sector. For this reason, there is only a single supplier of such batteries. However, the used batteries are not collected by any central body. The individual users dispose of their batteries themselves. Thus, it is to be expected that a large proportion of these batteries is disposed of in refuse.

Silver oxide button cells are mainly used in watches and watchmakers generally replace these batteries. These button cells are collected by a single firm, which exports the used batteries to England for recovery of the silver. The quantity of mercury collected in silver oxide button cells is 0.5-2.5 kg.

The report Indsamlingssystemer for batterier - eksisterende erfaringer fra Danmark og udlandet ("Collection systems for batteries - experience from Denmark and abroad") (Hansen and Hansen, 2002) has gathered the experience gained from six different collection systems in Denmark (the Municipality of Vejle, AVV, VEGA, REFA, Amagerforbrænding and R98, AKV and the Municipality of Bramming). Manganese dioxide batteries, and to some extent alkaline batteries, are deposited locally. 44-97% (calculated as an average of 60-80% in this report) of the collected batteries are thus deposited locally, whereas the remaining 20-40% of the collected batteries are sent to Kommunekemi.

The same report offers a provisional estimate for the collection potential of 0.41-0.54 kg/inhabitant. Of this, an average of 0.18 (0.16-0.26) kg/inhabitant is collected (Hansen and Hansen, 2002). This corresponds to a collection efficiency of 30-60%. By far the greater part of the remainder ends in refuse. Manganese dioxide and alkaline batteries constitute 72-97% of the collected quantity (Hansen and Hansen, 2002). This figure for the collection efficiency will be used for batteries other than button cells in the subsequent calculations of this report.

Kommunekemi, Elektromiljø and a private collector of silver oxide batteries received a total of about 2.5 tonnes of button cells in 2001. In comparison to the above average consumption of an estimated 10-14 tonnes during the years up to and including 2001 (a little lower in 2000 and 2001), this represents a collection rate for button cells of about 20-30%. By far the greater part of the remainder is considered to enter the refuse stream.

In 2001, Kommunekemi and Elektromiljø received a total in excess of 77 tonnes of batteries other than button cells. According to foreign trade statistics, the total consumption of other loose retail batteries was 3,700 tonnes. Thus, the quantity of batteries other than button cells received by Kommunekemi and Elektromiljø corresponds to about 2% of the consumption.

Based on the foregoing estimates, Table 2.12 shows the distribution of disposal quantities for mercury in batteries around 2001.

Table 2.12 Distribution of disposal quantities for mercury in batteries around 2001

2.3.2 Chemicals for laboratory and industrial use

Mercury and its compounds have traditionally been used for a wide variety of purposes in analytical laboratories. They have been used partly in the form of chemical reagents in various assays, partly in the form of electrodes used for measuring redox conditions (standard calomel electrodes) and partly in metallic form in a variety of analytical instruments used, e.g., in polarimetry, particle counting with so-called "Coulter counters," etc.

Mercury and its compounds have also been used in the chemical industry, e.g., as catalysts in heterogeneous catalysis in which mercury can, for instance, act as the carrier substance for sodium (which actually gives the catalytic effect) and in homogeneous catalysis, in which specific compounds of mercury behave as catalysts in a solution of the reacting substances. The Department of Chemistry at the Technical University of Denmark (Fehrman, 2002), as well as a number of leading industries that produce and use catalysts, have the impression that, although mercury can in principle be used for catalysis in a number of chemical processes, such use is only marginal today. In this context, mercury is considered to be "old-fashioned" chemistry, and much has been done to avoid the use of mercury for the sake of the environment. One major chemical industry also states that the few grams that they use every year are purchased through the suppliers of general laboratory materials, as the importation of these substances (classified as toxins) is too inconvenient from an administrative standpoint, when considered in relation to that industry's minimal consumption. The same industry is also of the opinion that this is generally the case.

The Department of Data on Chemical Products has provided information on a marginal consumption of mercury compounds in the manufacture of a number of different paint and enamel products; see the discussion of these below.

Consumption
Information has been gathered on the consumption of mercury and its compounds by a number of suppliers, which together are considered to cover up to 95% of the general laboratory market in Denmark (the coverage of mercury compounds could be lower). Such information as it has been possible to obtain is, however, considered to be rather sparse for which reason the uncertainty of the figures used in calculations is a little greater than is suggested by the above-mentioned degree of coverage.

The suppliers in question recorded total annual sales of about 14 kg of metallic mercury and 35-40 kg of mercury compounds in 2001 and 2002. The compounds included, e.g., the sulphate, nitrate, chloride, acetate, and salts of mercury, mercury-sodium amalgam, salts of phenylmercury and mercury-containing benzoic acid derivatives. There has also been a very small consumption of mercury-containing organic biocides in biological reagents intended, e.g., for immunological assays; this is a question of ethyl mercury compounds (thimerosal) of the type that are also used in certain vaccines (see the discussion below). As a spontaneous assessment, mercury sulphate, which is still used in standard COD analyses of wastewater, appears to constitute a significant proportion of the consumption of mercury compounds. Although the available information does not make a precise calculation of the mercury content of compounds possible, it is estimated on the basis of substance-specific accounts of part of this consumption to correspond to about 70% of the quantity of mercury compounds.

Because this is a question of data for a single year, because certain data has been given for 2002, and because there are assumed to have been certain sales from other suppliers, the total consumption in laboratories and in industrial chemical applications in 2001 must in this report be estimated at about 10-30 kg of metallic mercury, and 30-60 kg of compounds with a mercury content of about 20-40 kg.

Development trends
Both the suppliers of laboratory materials and contacts in industry consider that the chemical use of mercury has almost ceased in Denmark. The use of mercury sulphate in COD analyses also seems to be dropping, even if this type of analysis is still in use. For instance, a chemicals supplier has stated that mercury and its compounds are not stocked in Denmark, and that since 1999 there have been no sales of many of the mercury-containing substances listed in its catalogue. The mass flow analysis for 1992-93 recorded a sale of 60 kg of mercury in mercury compounds; on this basis, a total consumption of 60-120 kg of mercury/year is estimated for these applications (Maag et al., 1996).

Disposal
By far the greater part of the (limited) consumption of mercury for chemical purposes appears to be used in chemical analysis. The collection of environmentally hazardous chemicals from laboratories in Denmark is considered to be effective, and it is assumed in this report that consumed mercury is collected and disposed of as hazardous waste, and that the losses to wastewater, etc., are insignificant. It is to be expected that the greater part is sent to Kommunekemi. However, a smaller (but unknown) part, in the form of tubes/kits for COD analysis, is returned to the foreign suppliers, which state that it is recycled.

2.3.3 Other uses of chemical compounds

National Register of Chemical Substances and Products, Denmark
For this mass flow analysis, the National Register of Chemical Substances and Products, Denmark, has been searched for uses of mercury and its compounds in chemical products.

All in all, about 400 products containing mercury or mercury compounds, from 25 companies, have been reported since 1997. This is almost exclusively a question of such polymer-based products as hardeners and resins for plastic materials, plastic flooring materials, jointing compounds, etc. No reported products containing mercury are registered as laboratory chemicals.

According to the information obtained from the National Register of Chemical Substances and Products, Denmark, annual imports of mercury and mercury compounds in reported products constitute about 6 kg, of which by far the greater part is recorded as the compound neodecanoate-O)phenylmercury (CAS No. 26545-49-3). This is used in different forms in hardeners and resins for a number of plastic based materials. Based on the molecular weight of the above substance, a total of about 3 kg of mercury is imported.

A total consumption of about 1 kg of mercury (recorded as metallic mercury), for the manufacture of a number of different paint and enamel products, has been reported. In all, this concerns the manufacturing of over 300 products, with a total production of about 7,000 tonnes, distributed over many firms. Although the information given on the concentrations of mercury in the products is inconsistent, it suggests a level of a few mg/kg. No more detailed investigation has been made of the functions of mercury in these products (possible functions can include, e.g., preservatives and catalysts in hardeners).

Exports have been recorded of 0.7 kg of mercury in products, chiefly paint and enamel products in the same categories as were mentioned in the previous paragraph.

The information taken from the National Register of Chemical Substances and Products, Denmark, is based on reports and updates thereof provided on the initiative of the companies that market the relevant products for which reason, significant uncertainty must be expected to be associated with the reported quantities and information on use. Whether or not all of the reported products are still being used is not stated. If we consider the total numbers of products, i.e., including products reported before 1997, total imports amount to about 80 kg of mercury or mercury compounds, whereas the manufacturing and export figures are the same as those listed above. Of the 80 kg, the greater part, i.e., 48 kg, is recorded as "phenylmercuric oleate." No reports of products containing this compound have been recorded since 1997.

It is difficult to assess the magnitude of similar - but unreported - products that are possibly traded in Denmark, especially when seen in the light of the very small quantities involved. The use of mercury in paint and enamel goods is not permitted in Denmark (with the exception of catalysts until further notice) and there is some degree of regulation in the EU and the USA. Any widespread use of mercury in imported products is therefore considered unlikely.

Consumption and disposal
Based on the above information, the consumption of mercury in products of the types recorded in the National Register of Chemical Substances and Products, Denmark, is loosely estimated at between 5 and 50 kg/year.

It largely appears to be a question of products used in buildings or other types of structure. The mercury input will largely follow the materials to which the products are applied and is therefore counted as landfill. Previously mercury compounds in these types of product were primarily used as a preservative in latex paint and, possibly to a limited extent, in pigments. Both of these applications are considered to have ceased in Denmark. The use of mercury compounds in the Danish manufacturing of latex paint probably ended more than 20 years ago. The available information does not offer a basis for making a realistic estimate of the quantity of mercury that is disposed of with old building materials or structures. Thus, we have only calculated with the disposal of a quantity corresponding to the current estimates of consumption.

Fireworks
Certain sources state that mercury compounds are used, or have been used, in explosives and fireworks. Floyd et al., 2002, state that fulminate of mercury is sometimes used in detonators for these purposes.

However, a Swedish study states that mercury is not used in the modern manufacturing of fireworks, and that it was only possible to find trace levels of mercury in assays of six selected types of firework. The levels found represent a total consumption of mercury in fireworks of about 0.01 kg in 1998 (Göteborgs Miljöförvaltning, 1999).

The quantity and representativeness of the Swedish studies are insufficient to completely rule out the use of mercury compounds in fireworks. However, we assume that the consumption of mercury in this application is insignificant.

Vaccines and eye drops
Mercury is still permitted in vaccines and eye drops.

Mercury has not been an ingredient of children's vaccines since 1992. Mercury is used in certain vaccines against influenza and in certain vaccines for travellers (Japanese encephalitis). Mercury is an ingredient in thimerosal, which is a preservative containing an ethyl mercury compound. About half of the Danish vaccines against influenza contain thimerosal. The quantity of mercury in each vaccine is so small (50 µg of thimerosal/dose) that the total Danish consumption can be assessed as being less than 20 g of mercury/year.

According to the Danish Medicines Agency, mercury is not permitted in new pharmaceutical preparations. It is possible that individual eye-drop preparations contain mercury, however this only applies to a few isolated instances.

The total consumption of mercury in vaccines and eye drops is therefore assessed as being exceedingly small (<1 kg).

2.3.4 Summary - chemical compounds

The information available on the consumption of mercury in the form of chemical compounds, as well as losses in conjunction with the manufacturing and use of products containing mercury compounds, is summarised in Table 2.13.

Table 2.13 Consumption and disposal of mercury in chemical compounds for specific purposes in Denmark, 2001 (kg/yr of mercury)

2.4 Mercury as an impurity in other products

2.4.1 Coal

Fossil fuels such as coal and oil have a small, natural mercury content. When burned, most of the mercury content of these fuels will accompany the flue gas. Thus, emissions of mercury to air will vary according to the mercury content of the fuels and the flue gas cleaning equipment used.

The period since 1990 has witnessed a drop in the consumption of coal and coke, which have been replaced by an increasing consumption of natural gas and renewable energy. 6.7m tonnes of coal were consumed for the production of energy in 2001, as opposed to 12m tonnes in 1994.

Coal's content of heavy metals varies to some extent, depending on the country of origin, although there is considerable variation between the individual mines of each country. In the most recent mass flow analysis, the mercury content of coal was estimated at 0.04-0.18 mg of mercury/kg (Maag et al., 1996). Newer analyses show a mercury content of 0.087-0.15 mg of mercury/kg of dry matter (Hald, 2002), from which the total consumption of mercury in coal can be estimated at 590-980 kg of mercury/year. Data from the industry suggests that the average quantity of mercury in the coal consumed is about 550 kg/year, although no interval of uncertainty is stated. In the following, calculations will be based on an amount corresponding to 600-1,000 kg of mercury/year consumed with coal, although the true figure is probably at the low end of this interval.

About 17% of the coal consumed is burned in plants equipped with semi-dry desulphurisation, 69% in plants equipped with wet desulphurisation, and 14% in plants that lack desulphurisation. The mass balances for heavy metals in Danish coal-fired power stations during 2000-01 are shown in Table 2.14.

Table 2.14 Distribution of mercury in quantity of coal consumed, together with residual products and emissions from Danish power stations, and combined heat and power plants. The data has been derived on the basis of mass balances for all Danish coal-fired combined heat and power plants.

1) No mercury has been found in the slag from coal-fired combined heat and power plants equipped with semi-dry desulphurisation or from plants without desulphurisation.

2) Plants that lack desulphurisation usually burn a type of coal that has a low mercury content (Hald, 2002), but as it has not been possible to distinguish between these coals and other types of coal in the underlying data, no consideration is given to this in our calculations.

Emissions to air
As can be seen from Table 2.14, about a third (about 31%) of the mercury consumed is emitted with the flue gas. There appears to be a clear tendency for plant equipped with desulphurisation to emit relatively lower quantities of mercury than plant that lacks desulphurisation. Plants equipped with desulphurisation emit about 25-30% of the consumed mercury, whereas about 50% is emitted by plants without desulphurisation. The total emissions to air in 2001 are estimated at 190-310 kg of mercury. According to information from coal-fired plants, the true figure is at the low end of this interval. In the most recent mass flow analysis, the emission in 1992 was estimated at 300 kg of mercury, so there has been an overall drop in emissions in comparison to 1992. This drop is primarily because the consumption of coal has been halved.

Residual products
The residual products from coal firing include fly ash, slag/bottom ash and desulphurisation products. About 50% of coals' mercury content ends in fly ash, far less than 1% ends in slag and bottom ash whereas about 20% ends in desulphurisation products.

It has not been possible to retrieve data on the quantities of residual products from coal-fired plants for 2001, but data is available for 2000. As the total quantity of coal consumed in 2001 was the same as that consumed in 2000, it will be assumed that the quantities of residual products were the same in 2000 as in 2001. A national total of 554,000 tonnes of fly ash was produced through the burning of coal in 2000. It has only been possible to retrieve detailed information on the disposal of the fly ash produced within ELSAM's area, where the disposal is distributed as follows (Hald, 2002):

• about 32% used in cement production
• about 31% used in the production of asphalt and concrete
• about 30% deposited in landfills or used as filling (construction work)
• about 3% exported to foreign producers of cement/concrete
• about 4% placed in storage.

The 43,000 tonnes of slag/bottom ash were disposed of as follows:

• about 91% used as filling (construction work)
• about 9% was exported.

The products of desulphurisation are used for a number of purposes: plasterboard, cement, sulphur fertiliser, etc.

On condition that the disposal of the residual products from the ELSAM area are representative of the entire country, the disposal of mercury together with the residual products of coal firing in Denmark in 2001 can be stated as shown in the points below (rounded). This condition for distributing the quantities of mercury in residual products could, however, over- or underestimate individual streams.

For instance, the contribution to cement production - based on ELSAM's distribution - amounts to 97,160 kg mercury/year, despite the fact that the quantity of mercury in desulphurised gypsum and fly ash actually reused in cement production can be estimated at 3,591 kg. Whether or not the desulphurised gypsum and fly ash destined for cement production are exclusively taken from coal-fired plants is unknown. The difference could be due to the fact that cement production is only carried out at one location in Denmark, which means that transport costs also influence the extent to which transportation of the residual products is worthwhile. The difference between the two intervals is 62-70 kg of mercury, which quantity must be assumed to be distributed over one or more forms of disposal, i.e., asphalt, etc., landfill and filling, storage or export.

• Cement production: 40-90 kg of mercury/year.
• Asphalt, concrete, gypsum, etc.: 100-360 kg of mercury/year.
• Landfill and filling: 70-110 kg of mercury/year.
• Storage: 10-20 kg of mercury/year.
• Export: 20-30 kg of mercury/year.

2.4.2 Oil and biological fuels

This section accounts for the losses and circulation of mercury in oil and gas from offshore drilling platforms, losses and circulation at refineries as well as in the final gas and oil products. Drilling fluids and mud, which can also be of relevance with respect to mercury, are used in drilling for oil and gas. Figure 2.2 gives a collective overview of the results of this section.

Extraction of oil and gas
The production of oil and natural gas takes place in several fields in the Danish part of the North Sea. Primary separation of the extracted materials into water, oil, and gas phases is carried out at the drilling platforms. 20m m³ of crude oil and 11,116m Nm³ of natural gas (Danish Energy Agency, 2001). Natural gas and oil are transported onshore for refining and sale, whereas the water phase is discharged into the sea. However, a part of the produced water and natural gas is injected into the seabed to maintain the pressure needed to make the oil and gas flow towards the wells. A small quantity of the natural gas is burned at the platforms, either as fuel or for flaring. An overview of what happens to the production of the Danish North Sea fields can be found in Table 2.15.

Table 2.15 Information on production and the subsequent streams of natural gas, crude oil and water from Danish fields in the North Sea (OSPAR, 2002; Danish Energy Agency, 2001).

1) About 443 tonnes of oil is discharged into the sea with the produced water (OSPAR, 2002).

The crude oil is transported onshore, where refineries refine it into finished products. There are three Danish refineries, which process an annual total of 9.4m tonnes of oil (Olieberetning 2002 ("Oil Report 2002")). This corresponds to about 55% of the Danish production of crude oil being processed by Danish refineries, while it is assumed the remainder is exported. Natural gas is dried and recompressed offshore, after which it is piped onshore (Dalsager, 2002). According to Statistics Denmark, less than half of the natural gas delivered onshore is exported.

Mercury occurs in several different forms in oil, water, and natural gas. The ionic forms of mercury will generally follow the water phase, whereas the elementary and organic forms are retained in the carbon phase (Wilhelm, 2001). There is a general lack of information on the mercury contents of oil and natural gas, as well as on mercury's fate in products from offshore extraction and subsequent distribution onshore. This lack of data on mercury does not only apply to the Danish fields, but is generally characteristic of the other fields around the world, as are the great uncertainties associated with the existing data (Wilhelm, 2001).

Crude oil: with respect to the global market, Wilhelm (2001) reports a mercury content in crude oil of from about 1.5 mg/tonne to 3,000 mg/tonne. Wilhelm (2001) reports values from 2.5-9.3 mg/tonne for four assays of the crude oil extracted from the North Sea that is refined by an American refinery. A Danish operator in the North Sea reports values in the literature of 30-100 mg/tonne of mercury in crude oil. Thus, there is great variation in the mercury content of crude oil. Assuming that the mercury content of crude oil corresponds to the results of the four American assays of North Sea crude oil, we obtain a mercury circulation of 43-160 kg/year from the Danish sector. However, the values stated in the literature indicate a potential consumption of many tonnes/year for which reason, specific analyses of mercury from the Danish fields in the North Sea are necessary in order to assess the true circulation of mercury. The four assays considered to be representative of Danish oil from the North Sea will be used in the following estimates of the fate of the mercury found in crude oil.

Natural gas: mercury in natural gas almost exclusively occurs in its elementary form, at concentrations far below the saturation point, indicating that mercury does not exist in liquid form in the underground reservoirs (Wilhelm, 2001). There are no Danish analyses of the mercury content of the natural gas that is pumped up and processed by the drilling platforms. The concentration of mercury in the natural gas burned offshore is assumed by Wilhelm (2001) to be of the order of 1 µg/m³, as is generally the case under American conditions.

Assuming that this concentration applies to Danish natural gas prior to transport onshore, it corresponds to a total circulation of about 11 kg of mercury/year in natural gas processed in Danish fields. All of the extracted gas is treated offshore; part of this is injected, a small proportion is burned and the remainder is piped onshore as ready-to-sell gas. About 2.9 kg of mercury/year is injected with gas, 0.9 kg of mercury/year is flared, and the remainder is piped onshore in gas. Elementary mercury adsorbs to metallic components and the mercury content of natural gas therefore reduces as the distance from the well increases (Wilhelm, 2001). Mercury precipitates in the pipelines, especially when transporting wet gas, which results in its complete disappearance from the gas when it is piped to onshore refineries (Wilhelm, 2001). The fate of the precipitated mercury is unknown, but Wilhelm (2001) estimates that it precipitates together with wax, sand, and other solid particles in the pipelines (gas sludge). The Danish natural gas system from the platforms to the shore is a dry system for which reason, there is a minor loss of mercury during transportation of the gas. Assays of Danish natural gas in the natural gas onshore network (ready-to-sell gas) have shown that the mercury content is below the detection threshold, where the detection threshold - depending on sample volume - is given as <0.1-0.8 µg/m³ (Gruithuijsen, 2001). Wilhelm (2001) gives the mercury content of natural gas as <0.02-0.2 µg/m³. Based on Gruithuijsen (2001), the circulation of mercury in the Danish consumer network is estimated at below 0.73-5.9 kg/year, and it is assumed that this quantity is emitted with flue gas when the gas is burned onshore. Based on the figures used here, the difference between this quantity and the quantity of mercury that leaves the drilling platforms is about 1.5-6.6 kg/year, and this is assumed to be precipitated in the pipelines - either during transportation onshore or at onshore gas treatment plants.

It is of relevance here to mention a study of the flow of mercury in The Netherlands, which states that the mercury content of Dutch natural gas is unusually high and that The Netherlands is the only country known to remove mercury from gas before it reaches the consumer (Maxson & Vonkeman, 1996). According to Maxson & Vonkeman (1996), German natural gas appears to contain higher concentrations of mercury than Dutch natural gas. The sludge from the extraction of natural gas is one of the two greatest and most concentrated sources of mercury in The Netherlands (Maxson & Vonkeman, 1996). The circulation of mercury from sludge, waste, and emissions related to the production of natural gas amounts to 8.5 tonnes in The Netherlands (Maxson & Vonkeman, 1996). Of this, the greater part is found in gas sludge, which as far as is known is separated at onshore plants in The Netherlands. As mentioned above, no specific information is available on the mercury content of Danish natural gas in connection with production, whereas the gas distributed to consumers is very well-documented (Gruithuijsen, 2001). For the time being, there is no clarity as to how mercury follows material streams (discharges to water and emissions to air from the platforms, sludge or solid residual products that are brought onshore, etc.,) using the technology applied in Denmark. However, provisional information indicates that the mercury is perhaps chiefly separated from the gas at the platforms. The Dutch production of natural gas is about nine times greater than that of Denmark (IEA, 2002). However, this alone cannot explain the large differences.

Produced water: assays of produced water discharged from Danish fields in the North Sea reveal an average mercury content of about 0.3 µg/l (OSPAR, 2000). Based on this information, we estimate the circulation of mercury in produced water to be of the order of 6 kg/year. About 60% (about 4 kg of mercury) of this is discharged into the sea, and the remainder, about 2 kg/year, is injected into the seabed, cf. the distribution given in Table 2.15.

Processing at refineries
In most cases, there is no loss of mercury during the transportation of liquids, especially not of mercury in oil (Wilhelm, 2001). Thus, we estimate that, from offshore extraction to the refineries, there is no significant loss of mercury from crude oil. The total circulation at refineries is therefore considered to correspond to the total circulation of crude oil, which is estimated at 40-160 kg of mercury/year. As mentioned above, 9.4m tonnes of crude oil is refined by Danish refineries (Olieberetning 2002), corresponding to a circulation of about 2,090 kg of mercury/year. The remainder, which is exported with crude oil, corresponds to 20-70 kg of mercury/year. It should be noted that these quantities have been derived on the basis of four American assays of crude oil from the North Sea, but that there is great variation in the literature concerning the mercury content of crude oil. Thus, there is a clear need for mercury assays of Danish crude oil, in order to obtain reliable numbers.

There is only limited knowledge of mercury's fate in the refining process itself. At the refineries, mercury ends either in the finished products, in wastewater, in flue gas or in a solid waste product, e.g., in the form of a flue gas cleaning product or sludge from wastewater treatment. There is a general trend towards lower concentrations of mercury in the products as the distillation temperature rises (Wilhelm, 2001).

Wilhelm (2001) reports a Canadian study that discusses emissions from Canadian refineries. The Canadian study does not allow for a possible content of mercury in wastewater or in solid waste products, and it concludes that the quantity of mercury emitted to the atmosphere by Canadian refineries is the difference between the mercury in the crude oil and the mercury in the refined products. The Canadian study concludes that more than 90% of the mercury in crude oil is emitted to the atmosphere during the refining processes at Canadian refineries (Wilhelm, 2001).

Similarly, Wilhelm (2001) refers to an American study that investigates mercury's route through the refining process. This study found that 66% of the mercury in crude oil ends in the refined products, 18% is emitted to air, 13% ends in solid waste products and 3% ends in wastewater from the refining process (Wilhelm, 2001). However, significant uncertainties attach to the results that comprise the basis of the American study. With its point of departure in the American figures above, Table 2.16 shows the estimated circulation of mercury in crude oil at Danish refineries.

Table 2.16 Estimated circulation in Danish refineries

1) Calculations are based on an annual consumption of 9.4m tonnes of crude oil, which corresponds to the total capacity of Danish refineries (Olieberetning 2002). On the basis of no more than four assays of North Sea oil, a mercury concentration of 2.5-9.3 µg/kg in crude oil was also applied in the calculations (Wilhelm, 2001),.

Consumption
The total Danish consumption of oil products in 2001 amounted to 6.2m tonnes (Olieberetning 2001). No new Danish studies have been found of the mercury content in different oil products, for which reason calculation of the circulation is based on analyses taken from the most recent mass flow analysis, as well as on American analytical data. Table 2.17 shows the circulation of mercury in oil products in 2001.

Table 2.17 Circulation of mercury in oil products in 2001

1) Conversion factors for m³ to tonnes based on (DONG, 2002).

2) For all products, the lowest concentration of mercury is based on the most recent mass flow analysis (Maag et al., 1996), whereas a reference to the highest concentration is given - as a note - for each individual product.

3) The highest concentration in the interval is based on (Wilhelm, 2001).

4) The highest concentration in the interval is based on the most recent mass flow analysis (Maag et al., 1996).

5) Wilhelm (2001) states a maximum content of 50 mg of mercury/tonne. However, the same maximum value (5 mg of mercury/tonne) as can be found in the most recent mass flow analysis (Maag et al., 1996) has been used in calculations.

The total Danish consumption of mercury in oil products amounts to 2-30 kg. As oil products are primarily burned without subsequent flue gas cleaning, the mercury content of oil products is assumed to be emitted to the air.

Fig. 2.2 gives a summary of mercury's route in products, from the North Sea and further to treatment and consumption onshore. It also shows the circulation of oil and water based drilling liquids and mud, which are described in more detail in the passage following the figure.

Fig. 2.2 Roughly estimated quantities of mercury in connection with the extraction of oil and gas in Denmark (based on flimsy data).

Click here to see Fig. 2.2

Drilling liquids and mud
Drilling liquids and a number of auxiliary substances (drilling and cementing chemicals) are used in drilling for oil and gas in the North Sea. Drilling liquids are used in exploratory and production operations. Drilling liquids are divided into water based and non-water based (i.e., oil-based drilling mud, OBM).

The use of drilling liquids is needed to counteract the pressure of oil and gas. Thus, drilling liquids contain materials of high density, e.g., barium sulphate, (barytes, BaSO₄).

The annual consumption of OBMs is about 5-8,000 tonnes, and it has been increasing in recent years in step with the increase in activity in the North Sea (Strandmark et al., 2002). The circulation of mercury in OBMs is a maximum of 33 kg of mercury/year and is based on the mercury content of the mineral, barytes, which constitutes about 80% of the drillings (Strandmark et al., 2002). The mercury content of barytes is given as being <5 mg of mercury/kg (Strandmark et al., 2002).

There is significant consumption of water-based drilling liquids - roughly estimated at 30,000-40,000 tonnes/year - and the mercury contribution from these drilling liquids is primarily considered to come from barytes, as is the case in OBMs. Four assays of water-based drilling liquids from Indonesia show a mercury content of 0.14-2.14 mg of mercury/kg (Wilhelm, 2001). This agrees with the mercury content of the barytes used in OBMs, which is <5 mg of mercury/kg (Strandmark et al., 2002).

Based on the assays of mercury in water-based drilling mud from Indonesia, the circulation of mercury in such mud is estimated here as being 4,386 kg of mercury/year. The circulation of mercury in OBMs is considered to be approximately 0.72-32 kg/year. The use and disposal of drillings from oil-based mud is regulated under the international OSPAR convention, which prohibits discharges to the sea (Strandmark et al., 2002). Drillings from OBMs are therefore disposed of in the Danish sector, either by injection into the seabed in the field or by transportation onshore for further processing, during which the oil in the drilling liquids is recovered in the form of base oil, whereas the solid material is used as landfill (Strandmark et al., 2002). As injection into the seabed is an irreversible process, it is impossible to recover the material thus disposed of (Strandmark et al., 2002). In 2002, all drillings from OBMs were transported onshore (Strandmark et al., 2002). The drilling mud is heated during the recovery process and some of its mercury content is considered to be emitted in flue gas. Assuming that half of the mercury is emitted and half is deposited, 0.15-11 kg of mercury/year is emitted to air and 0.15-11 kg of mercury/year is deposited. The circulation of mercury in drilling mud is outlined together with oil products in Fig. 2.2.

The way in which water-based drilling liquids are disposed of is not known, but it appears most likely that they are either discharged into the sea or injected into the seabed. This quantity is assumed to be disposed of in the sea, together with other offshore chemicals. With respect to the auxiliary substances, about 40% of the Danish consumption in the North Sea, corresponding to 55m tonnes, was discharged into the sea in 2001 (OSPAR, 2002).

Biological fuels
The consumption of straw, chipped wood and other wood products has remained largely unchanged during the period since the previous mass flow analysis, and it amounted to about 2.4m tonnes in 2000 (Danish Energy Agency, 2000). Of this, about 0.9m tonnes was consumed in power stations and district heating power stations, while the remainder was consumed in small farm installations, in industry, in private homes, etc.

Heavy metals are found in varying quantities in biomass, depending on how the biomass was cultivated and which plants it was based on. In general, there is only sparse information on the mercury content of biomass, and no newer data has been found on the mercury content of wood and straw in relation to the most recent mass flow analysis. The mercury content of pure wood is estimated (Christensen et al., 1997) at <0.007 mg of mercury/kg dry matter. A major Danish report from the Danish Technological Institute on mass balances for impurities at straw-fired plants covers detailed analyses of 12 different impurities, but mercury is not included (Evald, 1998). In the most recent mass flow analysis, the mercury content of wood is stated as being 1.0-1.5 µg/MJ, or 0.02-0.03 g of mercury/tonne dry matter (Maag et al., 1996). Assuming that the mercury content of straw corresponds to the mercury content of wood, the mercury circulation in biological fuels is about 18-76 kg/year.

Bottom ash (slag) and fly ash from a multi-cyclone or filter are formed when straw or chipped wood are burned in Danish utilities. Most utilities also have flue gas condensing equipment that yields sludge (Morsing & Westborg, 1996). The residual products are normally mixed in a common ash container (Morsing & Westborg, 1996).

When burning straw, the quantity of ash produced constitutes about 4-7% of the quantity burned (Videncenter for halm- og flisfyring ("Know-how center for straw and chip firing"), 1999) and, when chipped wood is burned, 0.4-3% ash is produced (Videncenter for halm- og flisfyring, 2000). Considered against the background of this information, an estimated 36,000-63,000 tonnes of straw ash, and 6,500-49,000 tonnes of wood ash, measured as dry matter, are annually produced. An annual mercury content of 0.9-0.16 mg of mercury/kg dry matter has been measured in a number of assays of straw ash from Danish utilities (Videncenter for halm- og flisfyring, 1999), whereas ash from chipped wood contains <0.05-1 mg of mercury/kg dry matter (Videncenter for halm- og flisfyring, 2000). There is no overall account of how this ash is disposed of, but roughly 2/3 is probably deposited as landfill, while the remainder is applied to agricultural soil. Under these circumstances, about 2.4-10 kg of mercury/year is deposited, while about 1.2-5 kg of mercury/year is applied to agricultural soil. The remainder, which corresponds to about 14-61 kg of mercury/year, is assumed to be emitted together with flue gas.

2.4.3 Cement

Certain recoverable materials, as well as fuels that contribute the most significant quantities of mercury, are used in the production of grey cement. However, the lime used in the process also makes a contribution. The quantity of mercury that derives from the raw materials of cement production amounts to about 140-324 kg/year, and the greater part the mercury contribution comes from the fuels and fly ash. In 2000, 0.2m tonnes of fly ash was used in cement production, which quantity is the same as in 1992. The mercury content of this varies, from between 0.13 to 0.39 mg/kg.

Aalborg Portland produced 2.6m tonnes of cement in 2000. The production of grey and white cement as well as clinker constitutes 2.0, 0.5 and 0.1m tonnes/year respectively. These products constitute about 99% of total production and their mercury content is below the applied detection threshold of 0.05 mg of mercury/kg. The production of cement has remained stable in recent years. The mercury content of the total production is estimated at 54-130 kg of mercury/year, of which the main part is to be found in grey cement. The most recent mass flow analysis of mercury estimated the circulation of mercury in products from cement production at 60-220 kg/year (Maag et al., 1996).

Based on import and export information, the Danish consumption of grey cement is estimated at 1.3m tonnes/year. The consumption in 1992 was about 1.1m tonnes. Based on the available information, the Danish consumption of mercury in cement during 2000 is estimated at 26-65 kg, while it is also estimated that about 28-65 kg of mercury is exported in products.

The total deposited quantity of the mercury in residual products of cement production is estimated at 4.6-18 kg/year; it consists principally of filter dust.

Mercury evaporates at around 350°C, for which reason most of the mercury found in the raw materials evaporates in the cement production processes (Keating, 1997). The emission in flue gas is estimated by the authors at 70-170 kg of mercury/year.

The life-cycle of concrete is relatively long - typically from 30 to 75 years. Crushed concrete and other inorganic building waste is used as filling. About half of the quantity currently consumed in Denmark is a rough estimate of the quantity disposed of. Thus, we estimate that about 13-32 kg of mercury/year is deposited in landfills or used in construction works.

2.4.4 Agricultural lime and commercial fertiliser

Agricultural lime
According to reports to Statistics Denmark, the consumption of agricultural lime has dropped, from 1m tonnes in 1992-93, to 0.44m tonnes in 2001. The trend over the past 10 to 15 years has been towards a falling consumption of agricultural lime as a result of the falling consumption of nitrogen in agricultural soil. Nitrogen reduces the pH value of the soil, for which reason agricultural lime is added to stabilise the pH value. As the mercury content of agricultural lime is considered to be quite stable, we can estimate solely on the basis of the drop in consumption that the quantity of mercury released from agricultural lime to Danish agricultural soil has been more than halved since 1992-93. New assays of mercury in agricultural lime state that the mercury content is below the detection threshold of 0.01 mg/kg dry matter (Poulsen, 2002). Thus, the release of mercury in agricultural lime to Danish agricultural soil is estimated at <4.4 kg of mercury/year.

Commercial fertiliser
The mercury content of commercial fertiliser is very largely dependent on the fertiliser's phosphorus content. The mercury content of raw phosphate is extremely variable and depends on whether the raw phosphate is of volcanic or sedimentary origin. The introduction of a guaranteed level of purity, which guarantees a maximum mercury content of 0.2 mg/kg of fertiliser, now ensures minimal variation in the mercury content of the fertilisers. However, according to information from the producers, the mercury concentration in fertiliser is rarely this high; in most cases it is <0.05 mg/kg. The consumption trend has been dropping over the past 10 to 15 years. In fertilisation year 2000-01, the total consumption of commercial fertiliser amounted to 1.10m tonnes (Plant Department, 2001), which represents a drop of 0.3m tonnes/year in relation to fertilisation year 1992-93. About half of this fertiliser contained phosphorus. Based on the data collected, the total mercury content of commercial fertiliser is estimated at 11-36 kg/year. The most recent mass flow analysis estimated the quantity of mercury in commercial fertiliser at <50 kg/year.

2.4.5 Foodstuffs

The mercury content of foodstuffs derives from natural sources and from many years of man-made pollution. Mercury concentrates in the liver and kidneys of animals (Furness, 2001). Certain types of fish can contain much mercury, and the content varies widely from species to species, as well as between different waters. In Denmark, the average amount of mercury ingested in fish is 0.005 mg/week, although it is very dependent on the individual's consumption of fish (Danish Veterinary and Food Administration, 1996). The Danish Veterinary and Food Administration (1996) states that the Danish population's intake of mercury in all foodstuffs amounts to an average of 0.035 mg/week and, in the highest case, up to about 0.06 mg/week. Against this background, the total circulation in foodstuffs is estimated at 9.6-17 kg of mercury/year. The mercury that is not absorbed by the body ends mainly in wastewater (a small proportion is exhaled). Absorption by the body is extremely dependent on the form in which mercury is ingested. Thus, almost all organic mercury is absorbed after oral ingestion, whereas only 5-10% of the inorganic mercury compounds are absorbed (Danish EPA, 1987). The primary source of organic mercury in foodstuffs is fish. The half-life of mercury accumulated in the body is 7080 days (Danish EPA, 1987). Overall, the body will therefore excrete the greater part of the mercury it ingests. Here, we expect a loss to wastewater corresponding to a circulation of 9.16-17 kg of mercury/year.

With the exception of mercury in the teeth, the quantity of mercury accumulated in the body will be minimal when death occurs. The distribution of mercury varies throughout the body. Thus, the following concentrations of mercury have been found: 8 µg/l of blood, 2 µg/g of hair, 4 µg/l of urine and 10 µg/kg of placenta (wet weight) (OECD, 1994). Some mercury will also be found in fat. Assuming that the entire body (60 kg) has the same mercury concentration as the hair, which is the highest of the above concentrations, a human body contains about 18 mg of mercury. There are about 60,000 deaths in Denmark every year. Using the above assumption, the bodies of the deceased contain a total of about 1 kg of mercury. This represents only a fraction of the mercury contained in teeth. For this reason, we have decided to ignore the quantity of mercury accumulated in the body at the time of death in places other than the teeth.

2.4.6 Miscellaneous consumer goods

In general, mercury must be expected to occur in greater or lesser quantities in practically all goods and products; this is partly due to the natural occurrence of mercury as a trace element and partly due to diffuse pollution by mercury.

The previous mass flow analysis roughly estimated the consumption of mercury as an impurity in goods at 70-1,400 kg of mercury/year (Maag et al., 1996). This quantity was estimated on the basis of the circulation quantities in solid waste destined for recycling, incineration or landfills, and on the assumption of a mercury content in miscellaneous goods of 0.01-0.02 g/tonne (Maag et al., 1996).

It appears reasonable to compare this value for mercury content to the background concentrations of mercury in coal (0.087-0.15 g/tonne), oil (0.0015-3 g/tonne), cement (<0.05 g/tonne dry matter), fertiliser ((<0.05 g/tonne dry matter) (see the previous section) and soil (0.04-0.12 g/tonne dry matter) (Stubsgaard, 2001).

The total quantity of waste in 2001 amounted to about 9.5m tonnes. Assuming a mercury concentration of 0.01-0.2 g/tonne,, this corresponds to 94-1,900 tonnes of mercury. Assuming that the circulation of mercury in waste corresponds to the consumption of impurities in different goods, the consumption of mercury as an impurity can roughly be estimated at 94-1,900 tonnes/year, of which 55-1,100 kg is contained in goods that are recycled. A more precise estimate of this quantity is not possible due to the lack of relevant information.

This quantity of mercury will probably sooner or later be chiefly disposed of in solid waste, a description of which can be found in Table 3.4.

2.4.7 Summary - mercury as an impurity

The available assessments of the mass streams of mercury as an impurity are summarised in Table 2.18.

Table 2.18 Consumption and disposal of mercury as an impurity in Denmark in 2001 (kg/yr of mercury)

1) In the summation, 26-65 kg of mercury in fly ash was subtracted, as it is used in cement production and thus occurs twice in the table.

2) The column "Deposited & suchlike" includes residual products used for building and construction purposes, as well as mercury in waste that is burned or landfilled (cf. Section 3).

3) Mercury in residual products used in industrial production or exported for building and construction purposes (cf. cement production) and mercury as an impurity in goods that are recycled.

4) 0.3-9.5 kg of this is injected into the seabed in connection with drilling operations in the North Sea.

5) Expected to be disposed of in solid waste destined for incineration or landfills (see Table 3.4).

6) Mercury as an impurity in goods that can be reused, see Section 3.1.

7) The manner in which this quantity is disposed of is not known, but it is considered most likely that it is discharged into the sea or injected into the seabed.

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

Version 1.0 June 2004, © Danish Environmental Protection Agency