• No results found

The Swedish charge on nitrogen oxides

N/A
N/A
Protected

Academic year: 2021

Share "The Swedish charge on nitrogen oxides"

Copied!
12
0
0

Loading.... (view fulltext now)

Full text

(1)

The Swedish charge on

nitrogen oxides – Cost-effective

emission reduction

UNDER INFORMA-TION FACTS THE SWEDISH EPA PRESENTS FACTS ABOUT DIFFERENT ISSUES

INF

O

RMA

TION

FACTS

Emissions of nitrogen oxides (NOX) contribute to several environmental problems. Together with sulphur dioxide (SO2), emissions of nitrogen oxides are the main causes of acidification, which results in widespread damage to vegetation in forests and lakes. In addition, nitrogen oxides contribute to the formation of ground-level ozone, which has adverse effects on vegetation and human health, and they are a main cause of eutrophication in forest soils and on sea beds.

Emissions of nitrogen oxides are connected to several of the 15 national environmental quality objectives, such as Natural acidification only and Zero eutrophication. The acidification objective has an interim target, stating that emissions of nitrogen oxides are to be reduced to 148 000 tonnes per year by 2010. This is in line with the EC Air Quality Directive1 and would mean a 56 per cent reduction compared to 1990 levels. Emissions are also regulated in the UNECE-CLRTAP Protocol2 to abate acidification, eutrophication and ground-level ozone.

In the 1980s acidification was a major problem in Sweden. In 1985, the Swedish Parliament decided that airborne emissions of nitrogen oxides should be reduced by 30 per cent by 1995, compared to 1980 levels. In line with proposals put forward by the Swedish Environmental Charges Commission in the late 1980s, a charge on emissions of nitrogen oxides from energy generation at combustion plants was introduced on 1 January 1992. The intention was to achieve a more rapid reduction in emissions of nitrogen oxides than was considered possible by relying on the

administrative guidelines in place at that time. There was also a will to provide an incentive for cost-effective emission reductions in excess of these administrative guidelines.

In 1980, total emissions of NOX in Sweden amounted to about 450 000 tonnes3.

Total emissions of NOX in Sweden from 1990 to 2003 are shown in Figure 1.

_____________________________________

1 Also referred to as the NEC (National Emissions Ceiling) directive.

2 United Nations Economic Commission For Europe – Convention on Long-Range Transboundary Air Pollution. The protocol is also referred to as the Gothenburg Protocol. 3Methods for calculating emissions of NO

X have changed. This figure is an indication of the size

(2)

2

Figure 1. Total emissions of nitrogen oxides in Sweden 1990-2003, Source: Swedish EPA.

Emission charge and refund

According to the NOX Act the charge is to be paid for emissions of nitrogen oxides from boilers, stationary combustion engines and gas turbines with a useful energy production of at least 25 gigawatt hours (GWh) per year. Almost all production units targeted by the NOX Act are boilers. In the following text the word boiler is used for simplicity. The NOX charge is based on actual recorded emissions. It is imposed irrespective of the fuel used and is levied at a rate of SEK 40 per kg of emitted NOX4 (equal to about € 4.3).

To avoid distorting the pattern of competition between those plants which are subject to the NOX charge and those that are not, the system is designed so that all revenue except the cost of administration is returned to the participating plants, in proportion to their production of useful energy. Boilers with high emissions relative to their energy output are net payers to the system, and sources with low emissions relative to energy output are net recipients. This feature of the system encourages the targeted plants to reduce their emissions of nitrogen oxides per unit of energy to the lowest possible level.5 Because of the refund mechanism, the targeted polluters have been relatively content with the charge. This mechanism also made it possible to levy the charge at a level as high as 40 SEK per kg. In 2004 the refund was 8.94 SEK per MWh useful energy (about € 0.97 per MWh).

The decision to set the unit charge at 40 SEK per kg NOX was based on engineering data on expected effectiveness and costs of abatement investments at electricity power stations and district heating plants. The abatement cost was found to range between 3 and 84 SEK per kg reduced NOX. A charge of 40 SEK per kg was therefore considered reasonable. The charge has remained constant in nominal terms since introduction.

4 The sum of NO and NO

2 converted into NO2.

5 For calculation examples, see Appendix A

0 50 100 150 200 250 300 350 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 100 0 tonnes

(3)

3

Assessment of emissions

The charge is applied to measured emissions, or to presumptive emissions levels of 250 milligrams per megajoule (mg per MJ) for boilers and 600 mg per MJ for gas turbines. Plant operators may choose to pay the charge on the basis of presumptive emissions levels or by installing measuring equipment. In most cases the

presumptive emissions levels are substantially higher than the actual emissions, so measurement is generally preferred. The presumptive levels are also applied when the measuring equipment has been out of order, or does not comply with the specifications required by the Swedish Environmental Protection Agency (Swedish EPA). To allow time for maintenance and calibration of the measuring equipment, operators may estimate emissions for a maximum of 37 hours each month, on the basis of emissions under similar operating conditions. For another 60 days (1440 hours) the emissions can be calculated as the emissions from similar operating conditions multiplied by 1.5.

Extension of the charge system

Monitoring emissions requires relatively heavy investment in monitoring equipment. In order not to be overly cumbersome, the charge was therefore initially confined to about 124 combustion plants (182 boilers) producing at least 50 GWh of useful energy per boiler. Because of its effectiveness in emission reduction and falling monitoring costs, the charge system was later extended. As from 1996, all boilers producing at least 40 GWh per year were included. In 1997, the system was further expanded to include all boilers producing at least 25 GWh of useful energy per year. Today, about 260 plants (400 boilers) are subject to the charge.

Figure 2 shows a breakdown of NOX emissions in Sweden in 2003 among various sectors. NOX emissions from stationary combustion account for 42 per cent of total emissions. Out of this figure, boilers subject to the NOX charge are only responsible for 8 per cent.

Fuel used by the boilers subject to the charge in 2004 is shown in Figure 3. Half of the energy produced is based on biofuel.

(4)

4

Figure 2. Emissions of nitrogen oxides in Sweden 2003, sector by sector. Source: Swedish EPA.

Figure 3. Fuel used by boilers subject to the charge in 2004 based on energy

production. Source: NOX charge database.

Administration

The Swedish EPA is the taxation authority for the NOX charge. Charge payers must register with the Swedish EPA by submitting a return for boilers subject to the charge no later than 25 January each year. The return includes the amount of NOX emitted and the energy produced. The Swedish EPA checks these returns and a net payment/refund is calculated for each company6. Polluters that are net payers are invoiced on 1 September and companies that are to receive money are paid by 1 December.

The equipment for continuous monitoring of NOX permanently installed on a boiler must be checked by an independently accredited inspector at least once a year to ensure that the system meets the quality requirements stipulated by the Swedish EPA. The installed monitoring equipment is checked against an independent monitoring system in a parallel measurement. When submitting their returns, companies must also enclose the inspection report.

The Swedish Board for Technical Accreditation (SWEDAC), a government agency, accredits the inspectors. At the beginning of 1992, there were only a few accredited inspectors. Today there are 15.

6 The net payment/refund calculations for two different plants can be found in Appendix A.

Gas 13% Oil 14% Coal 6% Peat 4% Biofuel 50% Waste 13% Transport 51% Other stationary combustion

34%

Plants subject to the NOx charge

8%

Industrial processes 7%

(5)

5

In addition to scrutinising returns and administering incoming and outgoing payments, the Swedish EPA audits plants subject to the charge independently of the inspections made by the accredited companies.

The following elements are examined in the audit:

• Calculations and measurement of emissions and flue gas flow • Calibration procedures

• The daily report system • Maintenance procedures

• Calculations and measurements of useful energy

• The measurements carried out by the accredited inspector • Examination of the return (if appropriate)

By the end of 2005, most of the plants now subject to the charge had been audited. The administrative work carried out by the Swedish EPA in 2005 represented 5 man-years at a cost of SEK 4.6 million (approximately € 495,000). This is

equivalent to approximately 0.7 per cent of the total charge amount.

Monitoring and calculations of emissions

The Swedish EPA has published requirements for the performance properties of the automatic monitoring systems used for determination of the NOX emissions. The measured NOX concentrations combined with the accurate flue gas flow make up the basic data for the calculation of the NOX emission. For 87 per cent of the boilers, the flue gas flow is calculated from fuel data (elementary composition, specific heat), fuel consumption and oxygen or carbon dioxide concentration in the stack. For 13 per cent of the boilers, the flue gas flow is determined by direct measurements. Automatic measurements of the NOX concentration can be performed in two fundamentally different ways: extractive methods and in situ methods. In the extractive method, a gas sample is sucked from the stack and transported to an external measuring device through a sampling line. In situ measurements are performed inside the stack. 85 per cent of the boilers use extractive methods. Of these, 98 per cent are based on photometric determination.

Every hour, the mass flow of NOX (kg/h) is calculated as NO2 by multiplying the measured NOX concentration and the flue gas flow. The average hourly values are then added together to give the total annual emission, which is to be reported to the Swedish EPA.

For calculation of revenue, it is also necessary to monitor the amount of useful energy produced. Monitoring equipment for this normally already exists for other purposes.

(6)

6

Emission reduction

Since the Swedish Parliament passed legislation introducing the NOX charge in June 1990 the specific emissions have dropped from an average of about 160 milligrams of NOX per megajoule (mg/MJ) of energy input to about 55 mg/MJ7, equivalent to 65 per cent. (See Figure 4). It was concluded in an earlier evaluation of the NOX charge that 50 per cent of the emission reduction that was achieved between 1990 and 1992 (when the NOX charge came into effect) was due to the introduction of the charge. With the 1996 and 1997 expansions, which brought 100 new boilers into the system, total and specific emissions increased.

Figure 4. Total and specific NOX emissions from boilers subject to the NOX charge

between 1992 and 2004 and estimated emissions from these boilers in 1990.8

7 Emissions of 160 and 60 milligrams per megajoule (mg/MJ) of energy input would correspond, for plants burning coal or biofuels, to about 430 and 160 mg/m3. For plants fired by gas or oil, it

would be about 590 and 220 mg/m3.

8 Total emissions in 1990 are based on average specific emissions that year and energy output in 1992. 0 5000 10000 15000 20000 25000 30000 -90 -91 -92 -93 -94 -95 -96 -97 -98 -99 -00 -01 -02 -03 -04 0,000 0,050 0,100 0,150 0,200 0,250 0,300 0,350 0,400 0,450 0,500 0,550 0,600 total NOX-emissions Specific NOX-emissions

(7)

7

Figure 5. Mean NOX emissions in kg NOX per MWh useful energy produced by boilers

targeted by the NOX charge in 1992-2004.

0,000 0,050 0,100 0,150 0,200 0,250 0,300 0,350 0,400 0,450 -92 -93 -94 -95 -96 -97 -98 -99 -00 -01 -02 -03 -04 kg/MWh

Boilers targeted from 1992 Boilers targeted from 1996/97 All boilers

Figure 6. Emission reductions for different types of fuel at boilers targeted by the

NOX charge in 1992-2004. 0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 kg /M W h WasteBiofuel Coal Oil Gas

For the boilers that entered the system in 1992 and that have remained within the system ever since, specific emissions have fallen by approximately 42 per cent, from about 0.40 kg NOX/ MWh useful energy to 0.23 kg NOX/MWh (See Figure 5). The smaller plants, which entered the system later, have only reduced their emissions by approximately 26 per cent.

All sectors have substantially reduced their emissions since the charge was introduced, see figure 7.

(8)

8

Figure 7. Specific NOX-emissions for different sectors 1992-2004. Each bar

represents one year, starting with 1992 to the far left for each sector, and the years increasing toward the right.

0,000 0,100 0,200 0,300 0,400 0,500 0,600

Avfallsförbränning Kemiindustri Kraft- och värmeverk Livsmedelsindustri Massa- och

pappersindustri Metallindustri Träindustri kg/ M W h

Different industries have had varying degrees of success in their efforts to reduce their NOX emissions in a cost-effective way. Heat and cogeneration plants as a group have always been net recipients of payments, i.e. “winners” in the system, since the charge came into effect. One reason is that the energy sector started to invest in NOX reducing measures at an earlier stage and at a higher annual cost than the pulp and paper and chemical sectors. For boilers where the energy output varies a lot, it is more difficult to optimize operation with regards to NOX emissions. These boilers tend to be more common within industry, whereas energy sector boilers tend to be operated at a more even level. For net payers and receivers by sector, see figure 8.

Figure 8. Net payers and net receivers sector by sector 1992-2004. Each bar represents one year, starting with 1992 to the far left for each sector, and the years increasing toward the right.

Com bine d he at a nd p ow er Ch em ical indu stry Foo d in dust ry Pul p an d pa per indu stry Met al in dust ry Woo d in dust ry Was te in cine ratio n -50 -40 -30 -20 -10 0 10 20 30 40 50 Net p a ye rs [ M illion SEK] N e t rec e iv ers

(9)

9

The reduction in emissions of NOX cannot solely be ascribed to the effect of the NOX charge system. A large number of boilers are also subject to specific

regulations stipulated in the plant’s operating permit under the Environmental Code.

NOX-abatement measures

There are two principal methods of reducing NOX emissions: combustion measures (primary measures) and flue gas cleaning (secondary measures). The costs and use of these differ widely. For the boilers subject to the charge in 2004, 58 per cent had taken some kind of abatement measure.

Combustion measures

Combustion measures to reduce emissions include low-NOX burners, flue gas recirculation, air staging, over or rotating fire air, reburning and “fine tuning” of the combustion system. Changes in operating procedures for a given plant have, in several cases, had a substantial impact on the level of NOX emissions.

An inexpensive way of reducing the formation of NOX during combustion is to “trim” the combustion process, i.e. to choose temperature, oxygen rate and other combustion parameters in order to optimise combustion efficiency. According to a 1996 study9, about 70 per cent of the boilers had either been “trimming” their boilers or had installed different kinds of combustion measures. More than half of the trimming measures were reported to have been implemented at zero cost.

Combustion measures are sometimes combined with flue gas cleaning.

Flue gas cleaning

There are two different methods of flue gas cleaning: selective non-catalytic reduction (SNCR) and selective catalytic reduction (SCR). These methods can also be combined. SNCR is based on the principle of feeding urea or ammonia (NH4) into the combustion chamber. The urea/ammonia reacts with the NOX in the chamber and in the flue gas to form, nitrogen (N2) and water (H2O). The emission reduction rate varies between 20 - 70 per cent but is normally around 50 per cent. Before the introduction of the NOX charge, SNCR had not been used at all in Swedish plants.

SCR uses a catalytic ceramic to convert the NOX to N2 and H2O. This method is the most effective but is also a more costly way to reduce NOX emissions, especially given the relatively small size, by international standards, of the plants concerned. The emission reduction rate is often as high as 90 per cent. Eleven boilers (3 per cent) have so far had SCR installed.

9 L. Höglund, (1999) ”Essays on environmental regulation with application to Sweden”, School of Economic and Commercial Law at Gothenburg University, Department of Economics.

(10)

10

Table 1. Type of NOX reducing measures for boilers subject to the charge in 2004

implemented by sector10.

Energy Pulp and

paper Waste combustion Wood industry All sectors

Flue gas treatment SNCR/SCR 28 % 22 % 77 % 0 % 29 % Flue gas recirculation 22 % 22 % 45 % 38 % 25 % Other combustion measures 20 % 35 % 12 % 5 % 21 % No measure 44 % 38 % 16 % 58 % 42 %

The NOX charge represents the greatest reason why NOX abatement measures were implemented during the 1990s for boilers that are subject to the charge. For half of the measures, it was stated that the NOX charge was the deciding factor in

implementing NOX abatement measure11. Comparatively large reductions have been possible at what has been reported as zero or very low cost. One reason for

substantial reductions being possible at very low cost may be that the

implementation of NOX abatement measures itself has intensified the spotlight on boilers. Implementation has often been preceded by thorough scrutiny of the boilers and their functions. During this process, other efficiency-improving and cost-saving measures have been discovered and subsequently implemented. Knowledge of abatement measures for NOX has also improved due to the existence of the NOX charge.

According to a 1996 study, the average costs of measures to reduce emissions as a result of the NOX charge was SEK 7.5 (about € 0.8) per kilogram of reduced NOX. Reduced emissions at zero cost or even at a profit were reported for about 30 per cent of boilers.

Side-effects

There is a risk of other emissions increasing when measures are taken to reduce NOX emissions.

Carbon monoxide emissions

Modification of combustion techniques with lower temperature and/or oxygen rate in the combustion chamber, known as “trimming”, may lead to less complete combustion with reduced efficiency and often increased emissions of carbon monoxide (CO) and other non-combusted compounds. Carbon monoxide is a poisonous gas. The concentrations that reach humans from stationary combustion sources are generally too low to create any serious health problems. The concern over emissions of carbon monoxide from stationary sources relates rather to carbon monoxide as an indicator of less complete combustion and increased emissions of more harmful pollutants such as volatile organic compounds (VOC) and polycyclic aromatic hydrocarbons (PAH).

10 This table gives a rough indication of the actual state of affairs. In reality, a combination of measures is often implemented. Source: NOX charge database.

11 L. Höglund, (1999) ”Essays on environmental regulation with application to Sweden”, School of Economic and Commercial Law at Gothenburg University, Department of Economics.

(11)

11

Emissions of nitrous oxide

Flue gas treatment using urea and certain combustion conditions may result in increased emissions of nitrous oxide (N2O). Effects on emissions of nitrous oxide from NOX abatement arise mainly when non-catalytic flue gas treatment (SNCR) with urea is used to reduce NOX in the flue gases. An increase in emissions of nitrous oxide is undesirable since it is a very stable greenhouse gas. Nitrous oxide and the decay products, such as nitrogen oxide (NO), are also involved in the atmospheric chemistry of stratospheric ozone. Hence, emissions of nitrous oxide contribute to depletion of the ozone layer.

Ammonia emissions

When using non-catalytic (SNCR) or catalytic (SCR) flue gas treatment, ammonia or urea is used to reduce emissions of NOX. Not all the ammonia reacts chemically; some will be released with the flue gases, collected in condensate or fixed in fly ash. Ammonia emissions contribute to acidification and eutrophication of land and waters.

Effects on emissions

Emissions of ammonia from boilers subject to the NOX charge with SNCR/SCR were calculated in 2000 as 210-300 tonnes. For nitrous oxides the emissions were 530-620 tonnes. This is equivalent to 0.5 per cent of total emissions in Sweden for ammonia and 3 per cent for nitrous oxide.12

Future development

The introduction of the NOX charge has been a crucial factor in the considerable reduction of emissions of nitrogen oxides that has taken place at large-scale combustion plants in Sweden since 1990.

One possible change is to extend the NOX charge to include the majority of NOX emissions from industrial processes. However, this will require the system to be divided into different categories of plants with similar competition conditions.

Another possible development of the system is to increase the charge per kg of emitted NOX from the current level of SEK 40 (about € 4.3). An increase to 50 SEK per kg would lead to a decrease of NOX emissions of more than 5 000 tonnes, according to a 2004 investigation13. This is equivalent to a decrease of about 30 per cent in emissions from the boilers subject to the charge.

The following reports have summaries in English and can be downloaded from the Swedish EPA online bookshop.

Report 5335, Reducing NOX-emission – An evaluation of the Nitrogen Oxide Charge, (2003)

Report 5356, Suggestions for cost effective reduction of nitrogen oxides emissions, (2004)

Report 5525, Expansion of the NOX charge system, (2005)

12 Utsläpp av ammoniak och lustgas från förbränningsanläggningar med SNCR/SCR, Swedish Environmental Protection Agency, 2002.

13 Förslag för kostnadseffektiv minskning av kväveoxidutsläpp, Swedish Environmental Protection Agency, Report 5356 (2004).

(12)

Naturvårdsverket, SE-106 48 Stockholm. Visit Blekholmsterrassen 36. Phone: +46-8-698 10 00, Fax: +46-8-20 29 25, E-mail: natur@naturvardsverket.se Internet: www.naturvardsverket.se Customer Service: Phone: +46-8-505 933 40, Fax: +46-8-505 933 99, E-mail: natur@cm.se Postal adress: CM-Gruppen, Box 110 93, SE-161 11 Bromma. Internet: www.naturvardsverket.se/bokhandeln

Appendix A Calculation examples

Two examples of plants that are covered by the NOX charge system are given below.

First, the base credit amount for a specific year (2003) is calculated. The base credit amount is equal to the total amount of money collected (minus costs) divided by the total number of megawatt hours of useful energy produced.

All financial amounts in SEK (SEK 1 = € 0.11).

Calculation of the base credit amount for 2003

Total amount of NOX from the boilers: 15 835 659 kg

Total income from charges in 2003: 15 835 659 * 40 = SEK 633 426 360 Amount brought forward from 2002: SEK 10 751 606

Swedish EPA administration costs: SEK 4 155 000 Amount excluded from distribution14: SEK 15 000 000

Amount to be distributed: 663 426 360 + 10 751 606 - 4 155 000 - 15,000,000 = SEK 625 022 966

Useful energy produced at all plants in 2003: 66 136 158 MWh Base credit amount: SEK 625 022 966 / 66 136 158 = 9,45 SEK/MWh

Plant A

Measured NOX emissions: 14 601 kg

NOX charge: 40 SEK/kg *14 601 kg = SEK 584 040 Useful energy produced: 37 495 MWh

Energy credit balance for plant A: 37 495 * 9.45 SEK/MWh = SEK 354 328 Net payment = Environmental charge - Energy credit balance = 584 040 – 354 328 = SEK 229 712

Plant A has relatively high specific NOX emissions and has to pay the net sum of SEK 229 712 (about € 24 700) for 2003.

Plant B

Measured NOX emissions: 110 577 kg

NOX charge: 40 SEK/kg *110 577 = SEK 4 423 080 Useful energy produced: 548 374 MWh

Energy credit balance for plant B: 548 374 MWh * 9,45 SEK/MWh = SEK 5 182 134

Net payment = Environmental charge - Energy credit balance = 4 423 080 - 5 182 134 = SEK -759 054.

Plant B has low specific emissions and is a “financial winner” in the system. It receives a net payment of SEK 759 054 (about € 81 600) for 2003.

______________________________

14Amount that is reserved each year to be used for any necessary adjustments to the

Figure

Figure 1. Total emissions of nitrogen oxides in Sweden 1990-2003, Source: Swedish  EPA
Figure 2. Emissions of nitrogen oxides in Sweden 2003, sector by sector. Source:  Swedish EPA
Figure 4. Total and specific NO X  emissions from boilers subject to the NO X  charge
Figure 6. Emission reductions for different types of fuel at boilers targeted by the
+3

References

Related documents

46 Konkreta exempel skulle kunna vara främjandeinsatser för affärsänglar/affärsängelnätverk, skapa arenor där aktörer från utbuds- och efterfrågesidan kan mötas eller

The increasing availability of data and attention to services has increased the understanding of the contribution of services to innovation and productivity in

The weld cap appearance is best when the oxides from laser cutting remains (Samples 1-2), ie. no underfill, near net-shape. For Sample 4 when no oxygen is inside the

According to Jonsson (2005), companies that apply a carrying charge in inventory management ought to investigate how it is determined in order achieve a reasonable size. Closely

In the fermentation with 30 g/L glucose, the cells used the remaining glucose after cell growth and L -arginine production stopped, likely due to utilization of L -arginine as

The following assumptions are used, if not otherwise stated: Swedish electricity mix; typical values for process efficiency (energy) and transport distances; emissions from

CEN protocols for the test of diffusive samplers for measurements in ambient air were published in 2002/2003 85-87 .These protocols describe the test procedure to be used in

The Swedish House of Finance at the Stockholm School of Economics is Sweden’s national research center for financial economics. It hosts internationally distinguished researchers,