Effects of SO2 and NOx control on energy-efficiency power generation

ARTICLE IN PRESS

Energy Policy 35 (2007) 3898–3908 www.elsevier.com/locate/enpol

Effects of SO2 and NOx control on energy-efficiency power generation W.H.J. Graus, E. Worrell Ecofys, Kanaalweg 16-G, 3526 KL, Utrecht, The Netherlands Received 6 December 2006; accepted 10 January 2007 Available online 21 March 2007

Abstract The aim of this paper is to give an overview of the effects of SO2 and NOx pollution control on the energy efficiency of fossil-fired power generation for the following countries: Australia, China, France, Germany, India, Japan, Nordic countries (Denmark, Finland, Sweden, and Norway aggregated), South Korea, UK and Ireland, and United States. Together these countries generate 65% of fossilfired power generation worldwide. The level of SO2 and NOx control seems to vary widely for the included countries. The highest level of desulphurisation and denitrification is present in Japan, Germany, Nordic countries and South Korea. These countries also have the lowest NOx and SO2 emissions per unit of power generated. Limited pollution control is implemented in India and China, resulting in high specific NOx and SO2 emissions. The effect of NOx and SO2 control on net energy efficiency is estimated to be around 2% for coal-fired power plants and 1% for natural gas-fired power plants. The average power use for NOx and SO2 control for fossil-fired power generation is estimated to be between 1.2% and 1.5% of power generation output for countries with high levels of pollution control; leading to an effect on net efficiency of fossil-fired power generation of 0.5–0.6% points. r 2007 Elsevier Ltd. All rights reserved. Keywords: Pollution control; Power generation efficiency; Emission control

1. Introduction International comparisons of the energy efficiency of power generation can be used to compare a country’s performance to that of other countries. The results can be used to determine potential energy savings and greenhouse gas emission reduction. Factors influencing energy efficiency of power generation are, e.g. the type and age of power plants and the fuel types and technologies used. The net efficiency of a power plant is influenced by the amount of auxiliary power consumption for internal use. Strict pollution limits increase auxiliary power consumption of power plants and thereby reduce overall energy efficiency. Few studies have been done regarding the effects of SO2 and NOx abatement technologies on the energy efficiency of power generation. Although estimates for the energy consumption of most SO2 and NOx abatement technolo-

gies are available, the loss of energy efficiency on a country level has not been studied. In this paper we explore the differences in applied desulphurisation and denitrification technologies and the effect on power generation efficiency for a selection of countries. These countries are Australia, China, France, Germany, India, Japan, Nordic countries (Denmark, Finland, Sweden and Norway aggregated), South Korea, UK and Ireland, and United States. Together these countries generate 65% of fossil-fired power generation worldwide. This paper is structured as follows. Section 2 describes the approach used in this analysis. Section 3 gives an overview of data input. Section 4 gives the results of the analysis and Section 5 discusses uncertainties. Conclusions are given in Section 6.

2. Approach Corresponding author. Tel.: +31 302808324; fax: +31 302808301.

E-mail address: [email protected] (W.H.J. Graus). 0301-4215/$ - see front matter r 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.enpol.2007.01.011

The approach used in this analysis can be divided into three steps.

ARTICLE IN PRESS 2

W.H.J. Graus, E. Worrell / Energy Policy 35 (2007) 3898–3908

In the first step, we aim to give an overview of pollution control technologies applied on a country level. For coalfired power plants, we use the CleanPower5 database from the IEA Clean Coal Centre from IEACR (2006). For natural gas and oil-fired power plants, we use the World Electric Power Plant (WEPP) Database from Platts (2006). In the second step, we give an overview of the effects of different desulphurisation and denitrification technologies on energy efficiency. For the most frequently applied technologies, a literature research is done regarding energy consumption and efficiency losses resulting from the technologies. In the third step, we determine the effects of SO2 and NOx control on the energy efficiency of fossil-fired power generation for the countries involved. The results from the first step, in terms of the level of applied NOx and SO2 control technologies, are combined with the results from the second step. This gives an indication of the energy efficiency loss due to NOx and SO2 emission control for fossil-fired power generation of the observed countries. The energy efficiency of fossil-fired power generation for the included countries is taken from Graus and Worrell (2006). This study includes public power and CHP plants and is based on IEA Energy Balances, 2005 and National Statistics. Formula (1) gives the energy efficiency of power generation used in this study: E ¼ ðP þ HsÞ=I

(1)

where E is the energy efficiency of power generation, P the power production from public power plants and public CHP plants, H the heat output from public CHP plants, s the correction factor between heat and electricity, defined as the reduction in electricity production per unit of heat extracted, and I the fuel input for public power plants and public CHP plants. Heat extraction causes the energy efficiency of electricity generation to decrease although the overall efficiency for heat and electricity production is higher than when the two are generated separately. Therefore, a correction for heat extraction is applied. This correction reflects the amount of electricity production lost per unit of heat extracted from the electricity plant(s). For district heating systems, the substitution factors vary between 0.15 and 0.2. Graus and Worrell (2006) use a correction factor of 0.175. 3. Data input 3.1. SO2 and NOx control technologies SO2 and NOx emissions resulting from fossil fuel combustion depend on a number of factors like fuel type (e.g. thermal content, sulphur content and sulphur retained in the ash) and combustion conditions. Unabated SO2 emissions are particularly high for coal combustion, typically around 230 (anthracite) to 390 (bituminous coal) g/GJNCV, and highest for lignite; 525 g g/GJNCV. For oil,

SO2 emissions are somewhat lower typically around 70 (gas oil) to 350 (fuel oil) g/GJNCV. NOx emissions from coal range from around 90 (fluidised bed) to 540 (boiler) g/GJNCV. For oil typical NOx emissions are 200 g/GJNCV and for natural gas, between 30 (combined cycle) and 130 (boiler) g/GJNCV (Barrett, 2004). 3.1.1. SO2 control Besides a switch to fuels with a low sulphur content, there are currently two main types of desulphurisation technologies: wet scrubbers and dry scrubbers. Other SO2 emission control technologies are sorbent injection and combined NOx and SO2 control. Wet scrubbers are used where SO2 control requirements are most stringent and are the most widely used flue gas desulphurisation (FGD) technology. Wet scrubbers can achieve removal efficiencies as high as 99%. Spray dryer scrubbers are used for applications with less demanding SO2 requirements. Dry scrubbers are often superior to other designs in terms of cost efficiency, thanks to their simplicity and lower capital and maintenance costs. Combined SO2/NOx removal processes are fairly complex and costly. However, emerging technologies have the potential to reduce SO2 and NOx emissions for less than the combined cost of conventional FGD for SO2 control and selective catalytic reduction (SCR) for NOx control. Most processes are in the development stage, although some processes are commercially used in low- to mediumsulphur coal-fired plants [IEA Clean Coal Centre (2006) and ICAC (2006)]. 3.1.2. NOx control There are a few basic types of NOx emission controls: advanced combustor design to suppress NOx formation, wet controls using water or steam injection to reduce combustion temperatures, and post-combustion controls (flue gas treatment). The emission reduction from combustion modification can be 20–70%, depending on the technologies applied (IPPC, 2005). Combustion modification options are, e.g. separating the combustion process in stages. This staging partially delays the combustion process and results in a cooler flame that suppresses thermal NOx formation. Alternatively, air ports can be installed in addition to existing burners (overfire air operation). Burners can then be operated with low excess air, which inhibits NOx formation, while the overfire air ensures complete burnout. Another option is to recirculate part of the flue gases (usually less than 30%) into the combustion air. Flue gas recirculation results in a reduction of available oxygen in the combustion zone and leads to a decrease of the flame temperature and reduces both fuel-bound nitrogen conversion and thermal NOx formation. For gas turbines, dry low-NOx combustors (DLNC) can minimise combustion temperatures by providing a lean

ARTICLE IN PRESS W.H.J. Graus, E. Worrell / Energy Policy 35 (2007) 3898–3908

premixed air/fuel mixture, where air and fuel are mixed before entering the combustor. This minimises fuel-rich pockets. DLNC can achieve emission reductions of about 94% (California Air Resources Board, 2000). Water or steam injection can be used to reduce NOx emissions from gas turbines and involves the injection of a small amount of water or steam via a nozzle into the immediate vicinity of the combustor burner flame. Water injection typically results in a NOx reduction efficiency of about 70%, while steam injection can achieve approximately 80% control (California Air Resources Board, 2000). Post-combustion control reduces NOx emissions already formed. They can be implemented independently or in combination with primary measures. Main post-combustion technologies are SCR and selective non-catalytic reduction (SNCR). SCR systems selectively reduce NOx by injecting ammonia (NH3) into the exhaust gas stream upstream of a catalyst. NOx, ammonia, and oxygen react Table 1 Abbreviations Abbreviations

Full name

DLNC DNOX DESOX EMULS FGD

(Dry) low NOx combustors Unspecified NOx removal equipment Unspecified type of SO2 control equipment Emulsified fuel oil Flue gas desulphurisation Unspecified type of FGD scrubber Flue gas recirculation Low-NOx burners Overfire air Oxidation catalyst Selective catalytic reduction Selective non-catalytic reduction Combined SO2 and NOx control Steam injection Two-stage combustion Water injection

FGR or FR LNB OFA OXI SCR SNCR SNOX STM IJ TSC WAT IJ or WI

3

on the surface of the catalyst to form molecular nitrogen (N2) and water. SCR systems typically reduce NOx emissions by 80–95%. A SNCR system is operated without a catalyst and generally reduces NOx emissions by 30–50%. [IPPC, 2005] 3.2. Applied emission control technologies on a country level In this subsection, we give an overview of NOx and SO2 control technologies that are applied in the countries observed. Table 1 shows abbreviations used in the figures for the different control technologies. 3.2.1. Coal-fired power plants Fig. 1 shows the level of SO2 control for coal-fired power plants. As can be seen in Fig. 1, wet scrubbing is the most commonly used technology for SO2 control in coal-fired power plants, followed by dry scrubbing. SO2 control is particularly high for South Korea, Japan, Germany and the Nordic countries, and low for China, Australia, France and India. Fig. 2 shows combustion modification applied for reducing thermal NOx emissions in coal-fired power plants. Low NOx burners are widely applied for coal-fired power plants. UK and Ireland, and the Nordic countries have the highest share of low NOx burners. For Japan the figure shows a high share of overfire air and flue gas recirculation. Fig. 3 shows the level of flue gas treatment applied for removing NOx emissions from flue gas. SCR is the most often applied technology for flue gas treatment, followed by SNCR which is applied in the United States and Nordic countries. Japan and Germany have the highest share of installed SCR systems, 65% and 53%, respectively. The differences between countries in terms of SO2 and NOx control applied in coal-fired power plants are large,

100% 90% 80% 70%

Planned FGD

60%

SNOX Furnace sorbent injection

50%

Control in fluidised bed

40%

Dry scrubber

30%

Wet scrubber

20% 10% In di a

an y co un tri U K es + Ire U la ni nd te d St at es C hi na Au st ra lia Fr an ce

m

pa n

er N

or d

ic

G

Ja

Ko

re

a

0%

Fig. 1. SO2 control for coal-fired power plants in 2005 (IEACR, 2006). Note: Based on units 4100 MW in operation, including industrial power plants. The share refers to percentage of units equipped with pollution control device.

ARTICLE IN PRESS W.H.J. Graus, E. Worrell / Energy Policy 35 (2007) 3898–3908

4

100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0%

ranging from 0% to 90% for some technologies. The specific emissions of SO2 and NOx/kWh power generated are therefore expected to differ widely as well.

s St at es Au st ra lia C hi na Fr an ce In di a

an

un

m

te d

N

or

U

di

ni

c

co

G er

Ire +

K U

3.2.2. Gas-fired power plants Fig. 4 shows the share of gas-based capacity that is equipped with NOx emission control. In general, DLNC is the most applied NOx abatement technology for gas-fired power plants. Japan has a high share of SCR and two-stage combustion/flue-gas recirculation.

tri e

y

nd

a

la

re Ko

Ja

pa

n

LNB OFA FGR

3.2.3. Oil-fired power plants Fig. 5 shows the share of oil-based capacity that is equipped with NOx emission control. Note that oil-fired power generation is very small (o5 TWh) for Australia, Germany, UK and Ireland, France and the Nordic countries. Fig. 5 shows that SCR is the most applied NOx abatement technology for oil-fired power plants. The level of SCR is high for South Korea and Japan. Japan has additionally a high share of two-staged combustion in combination with flue gas recirculation. Fig. 6 shows the share of oil-based capacity with SO2 control. Fig. 6 shows that wet scrubbers are applied on a limited scale in Germany, Japan and Nordic countries. In general, SO2 control for oil-based capacity seems limited.

Fig. 2. NOx combustion modification of coal-based power plants in 2005 (IEACR, 2006). Note: Sorted on sum of three combustion control technologies. Note that the technologies can be applied simultaneously.

70% SNCR SCR

60% 50% 40% 30% 20% 10%

lia

a

c

K

+

Au

st

ra

di

nd

In

C

Ire

hi

la

na

ce an Fr

co

Ko

re

a

es un

at St

d te

or

U

di

ni

3.3. Specific SO2 and NOx emissions

N

U

tri

es

y an m

er G

Ja

pa

n

0%

Fig. 7 shows SO2 emissions per kWh for coal and oilfired power generation.

Fig. 3. Flue gas treatment for NOx emissions of coal-based power plants in 2005 (IEACR, 2006).

80% 70%

N/A Other DNOX SCR LNB/SCR DLNC/SCR WAT IJ STM IJ TSC/FR LNB/WI OFA/FGR LNB/OFA LNB DLNC

60% 50% 40% 30% 20% 10%

ra lia

Au st

In di a

an y m

er

d ni te U

G

St

at e

s

an ce Fr

hi na C

re a

So

ut h

Ko

es un tri

nd di c

co

Ire la N or

U K

+

Ja pa n

0%

Fig. 4. Share of gas-based capacity with NOx control in 2005 (Platts, 2006). Note: ‘‘Not applicable (N/A)’’ refers to internal combustion plants or older steam-electric power plants with no pollution control. In case data fields are empty this means information on NOx and SO2 control systems or processes is absent.

ARTICLE IN PRESS W.H.J. Graus, E. Worrell / Energy Policy 35 (2007) 3898–3908

5

70% 60% N/A DNOX SNCR SCR WAT IJ EMULS TSC/FR TSC LNB/OFA LNB DLNC

50% 40% 30% 20% 10%

K+

Ire

la

nd

lia st ra U

Au

C

hi

na

e

ia

nc Fr a

N

or

U

di

ni

c

te d

co

In d

s tri e un

m

St at es

y an

a re

G er

Ko

So

ut h

Ja

pa

n

0%

Fig. 5. Share of oil-based capacity with NOx control in 2005 (Platts, 2006). Note: ‘‘Not applicable (N/A)’’ refers to internal combustion plants or older steam-electric power plants with no pollution control. In case data fields are empty this means information on NOx and SO2 control systems or processes is absent.

100%

N/A Influidised bed control FGD Dry scrubber Wetscrubber

90% 80% 70% 60% 50% 40% 30% 20% 10%

nd

lia

a

Ire

U

K

+

Au

st

la

ra

di In

p co an un tri So es ut h Ko U re ni a te d St at es C hi na Fr an ce

Ja

N or di c

G er

m

an

y

0%

Fig. 6. Share of oil-based capacity with SO2 control in 2005 (Platts, 2006).

Fig. 7 shows a strong decrease in SO2 emissions per kWh for many countries, in some cases specific emissions reduce nearly tenfold. The differences between countries in terms of specific SO2 emissions are large ranging from 0.5 to 5.5 g/kWh in 2002 (factor 11). Japan has the lowest specific SO2 emissions, followed by Germany, South Korea (data for 1999) and Nordic countries, respectively 0.5, 0.9, 1.3 and 1.3 g/kWh. The highest level of specific SO2 emissions is 5.5 for the UK and Ireland and 4.4 for the United States in 2002. For the UK and Ireland, additional wet scrubbers are planned to be applied to around 30% of coal capacity. Furthermore a number of existing wet scrubbers became operational after 2002. SO2 emissions for the UK and Ireland are therefore expected to decrease further.

According to OSC (2006), SO2 emissions from thermal power plants in India are 7.4 g/kWh in 1998. For China, average SO2 emissions from coal-fired power plants are 7.3 g/kWh in 2000, according to Jintian et al. (2002). Based on LBNL (2004)1 SO2 emissions from coal- and oil-fired power plants in China are 6.5 g/kWh in 2000 and 5.7 g/kWh in 2002. Fig. 8 shows NOx emissions per kWh for fossil-fired power generation. Also NOx emissions have decreased for many countries. The differences between countries in terms of specific NOx emissions are again large, ranging from 0.4 to 3.6 in 2002 (factor 9). Japan has the lowest specific NOx emissions, followed by Germany and South Korea (data for 1999), 0.4, 0.7 and 0.9 g/kWh, respectively. The highest level of specific NOx emissions is 3.6 for France and 2.9 for Australia. According to OSC (2006), NOx emissions from thermal power plants in India are 7.0 g/kWh in 1998. For China, average NOx emissions from coal-fired power plants are 3.7 g/kWh in 2002, based on Tonooka et al. (2006). Factors influencing specific NOx and SO2 emissions for fossil-fired power generation are besides the level of pollution control and the energy efficiency of power generation, the fuel type that is used (i.e. share of coal in the fuel mix and its sulphur content). Fig. 9 shows the share of gas-fired, coal-fired and oilfired power generation in total fossil power generation.

1 SO2af0 emissions for electric utilites are divided by by oil- and coalfired electricity generation in public power plants (IEA, 2005).

ARTICLE IN PRESS W.H.J. Graus, E. Worrell / Energy Policy 35 (2007) 3898–3908

6 14

SO2 emissions (g/kWh)

12

10

Australia France

8

Germany Japan Korea

6

Nordic countries UK+ Ireland

4

United States

2

0 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002

Fig. 7. SO2 emissions per kWh for coal- and oil-fired power generation (IEA (2005) and OECD (2004)). Note: SO2 emissions from electric power stations taken from OECD (2004) are divided by oil- and coal-fired electricity generation in public power plants (IEA, 2005).

6

5

NOx emissions (g/kWh)

Australia

4

France Germany Japan

3

Korea Nordic countries

2

UK+ Ireland United States

1

0 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002

Fig. 8. NOx emissions per kWh for fossil-fired power generation (IEA (2005) and OECD (2004)).

3.4. Emission limits Table 2 gives an overview of SO2 and NOx emission limits for power plants in the different countries. In case no emission limits are available a short description is provided of relevant policies in the field of SO2 and NOx emissions of power plants. 3.5. Effect of SO2 and NOx control on power generation efficiency In the previous subsections we saw that the level of pollution control and specific SO2 and NOx emissions

differ widely per country. In this subsection we will look at the effects of SO2 and NOx control on power generation efficiency. Table 3 gives an overview of the impact of the SO2 and NOx control technologies on the energy efficiency of power generation. Table 4 shows that the effect of NOx and SO2 control on power generation efficiency consist mainly of an increase of auxiliary power consumption for fans and pumps to drive the pollution control devices, etc. The effect on net energy efficiency is around 2% for coal-fired power plants (wet scrubber, SCR and combustion modification) and 1% for gas-fired power plants (SCR and combustion

ARTICLE IN PRESS W.H.J. Graus, E. Worrell / Energy Policy 35 (2007) 3898–3908

Coal

Oil

Gas

100% 90% 80% 70% 60% 50% 40% 30% 20%

7

modification). In some cases there is also an effect on gross efficiency of at most 1% due to loss of boiler efficiency or additional fuel consumption. Fig. 10 shows the total auxiliary power consumption in electricity, CHP and heat plants, including auto-producers, as share of total gross power generation in 2003. The difference in auxiliary power consumption between countries can be explained by, e.g. (1) the fuel mix for power generation (e.g. auxiliary power consumption for natural gas plants is generally lower than for coal plants), (2) the level of pollution control, (3) efficiency of pumps, compressors, etc. and operation and maintenance and (4) the energy efficiency of power generation.

10% 0%

te d

ni U

c di N

or

U

a St at es G er m an y In di a Au st ra lia C hi na Fr an ce

re

s un

co

Ko

tri e

n pa

Ja

K

+

Ire

la

nd

4. Results

Fig. 9. Share gas, oil and coal in total fossil power generation in 2003 (IEA, 2005).

Table 4 shows the level of applied NOx and SO2 control technologies for the observed countries by fuel type. In the last row, an estimate is given of the power consumption of the pollution control technology as a share of total gross power output of a power plant.

Table 2 NOx and SO2 emission limits for power plants or related policies SO2 (mg/Nm3)a European Union [Entec (2005), Barrett (2004)] EU Large Combustion Plant Existing plants4500 MW Directive (LCPD) (2001/80/EC) 400 for solid and liquid fuels 35 for gaseous fuels

NOx (mg/Nm3)b

Existing plants4500 MW 500 for solid fuels 400 for liquid fuels 200 for gaseous fuels

To be applied from 1 January 2008

Plants built after 20034300 MW 200 for solid and liquid fuelsc 35 for gaseous fuels

Plants built after 20034300 MW 200 for solid and liquid fuels 100 for gaseous fuels

Germany

300 for solid fuels. In addition plants have to comply with a sulphur removal rate of at least 85%.

200 for solid fuels 135–150 for gaseous fuels

Finland

—

200 for plants 41000 MWth

United States [Entec (2005), Rubin et al. (2002), Schwarz (2003)] US acid rain programme (capPhase two (started in 2000): 1475 and-trade programme started in 1995)

490–1060 (depending on boiler type)

Federal New Source Performance Standard (NSPS)

1475 for solid fuels 985 for gas and liquid fuels

615–985 for solid fuels 370–615 for liquid fuels 245–615 for gaseous fuels (depending on fuel characteristics) Plants modified after 1997 185

North Carolina Clean Smokestacks Actb

200 for coal

100–120 for coal

Japan [World Bank (2001), Ministry of Environment Japan (2006), Tanaka (2000)] Air Pollution Control Law The emission limits for sulphur dioxide range from 170–3000d and vary according to geographical location and height of exhaust stack. Sulphur content limits for fuels are o0.5–1.2%. The average sulphur content of the heavy oil used in Japan was 2.5% in 1967, but by 1992 had dropped to 1.05%.

Large-scale boilers 410–513 for coale 267–308 for oil 123–205 for gas (depending on size)

ARTICLE IN PRESS 8

W.H.J. Graus, E. Worrell / Energy Policy 35 (2007) 3898–3908

Table 2 (continued ) SO2 (mg/Nm3)a

NOx (mg/Nm3)b

South Korea [Kim and Kim, 1999] Air Quality Control Law (1999) Existing power plants4500 MW 429–772 for solid fuels 429–515 for liquid fuels New power plants4500 MW 343 for solid and liquid fuels Sulphur content limits for fuels are o0.1–1.0%, depending on region and fuel type. China (Battelle Memorial Institute, 1998; Jintian et al., 2002)

India [EIA, 2004b]

Australia [EIA, 2002]

China is attempting to control sulphur emissions associated with power production by implementing flue gas desulphurisation (FGD) facilities. Government agencies in the power sector have also made efforts to improve the quality of coal supplies to reduce pollution. The average sulphur content in coal supplied to power plants decreased from 1.2% in 1988 to 1.05% in 1999

Power plants4500 MW 718 for solid fuels 513 and 1950 for liquid fuels 820 and 1025 for gas (upper value is for internal combustion engines)

The Chinese government has yet to develop a specific standard for NOx emissions for power production, although the former Ministry of Electric Power Industry had requested that new thermal power plants with a unit capacity over 300 megawatts use low NOx emission technologies.

The Central Pollution Control Board has been slow to set sulphur dioxide (SO2) emission limits for coal-fired power plants, mainly because most of the coal mined in India is low in sulphur content. Coal-fired power plants do not face any nitrogen oxide (NOx) emissions limits either Australia does not have specific emission limits for power plants. The reason is that sulphur dioxide, for example, is not a major pollutant, both because of the low-sulphur nature of Australia’s coal and because most power plants are located at significant distance from urban areas.

Note that emission standards are mostly set on a fuel input or flue gas output basis. More efficient plants will have lower emissions on a per kWh basis. Input-based emission standards however provide no incentive to reduce emissions by improving the energy-efficiency of power plants. In contrast outputbased standards would provide an incentive to do so. a Emission limits are given in mg/Nm3. Nm3 refers to the volume (m3) of exhaust gas in normal conditions (2731 K, 101.3 kPa) after correction for the water vapour content. b The actual numbers will be different at every facility depending upon facility type, boiler technology and coal type (Schwarz, 2003). c In case emission limits cannot be met due to fuel characteristics, desulphurisation must be at least 95% and maximum emissions 400 mg/Nm3. d Converted with 1 ppm ¼ 2.86 mg/Nm3 for SO2. e Converted with 1 ppm ¼ 2.05 mg/Nm3 for NO2.

Table 5 shows the gross efficiencies calculated in the study Graus and Worrell (2006) and give estimates for net efficiency, based on Fig. 10. In the two right-hand columns, the estimated power use for NOx and SO2 control is given as a share of gross power generation and the net efficiency excluding NOx and SO2 control is calculated. The net efficiency refers to net electricity output divided by fuel input, while the gross efficiency refers to gross electricity output divided by fuel input. The difference between gross and net electricity output is the power consumption of a power plant for its own use, including power consumption for pollution control. Please note that the uncertainties in the net efficiencies and in the power use for NOx and SO2 pollution control are significant. The share of own power consumption in gross power generation (see Fig. 10) refers to all heat and power plants in a country and may not be representative for the fossil-fired power generation share. These results therefore merely aim to show the sensitivity of net efficiencies to NOx and SO2 pollution control. The results in Table 5 show that power use for NOx and SO2 control is estimated to be between 1.2% and 1.5% for

countries with high levels of pollution control; Japan, Nordic countries, South Korea and Germany. The effect on net efficiency of NOx and SO2 abatement is 0–0.6% points (e.g. 40.6% instead of 40.0%) for the included countries, depending on the abatement level. Fig. 11 shows the results from Table 5 in graph form. 5. Discussion of uncertainties There are two main sources of uncertainty in this analysis: 1. the level of pollution control for the countries, 2. the effect of pollution control on energy efficiency. The first source of uncertainty is present in the estimation of the level of pollution control for the included countries. The uncertainty is estimated to be low for coal-fired power plants. The IEACR database is quite complete with a high detail level of information. The uncertainty for natural gas and oil-fired power plants is higher. A significant share of

ARTICLE IN PRESS W.H.J. Graus, E. Worrell / Energy Policy 35 (2007) 3898–3908

9

Table 3 Impact SO2 and NOx control technology on efficiency power generation (ICAC (2006), IPPC (2005), World Bank (2006), Rubin et al. (2002), COEN (2006), Stork (2005), Liere et al. (2001), Carnegie Mellon University (2006))

SO2 wet scrubbers SO2 dry scrubbers SO2 sorbent injection NOx SCR NOx SCNR NOx combustion modification

Power consumption as share of unit’s generating capacity

Other effects

1–2% 0.5–1% 0.01%–2% 0.5% 0.1–0.3% Depending on technology

In some cases 1% gross efficiency loss for reheating flue gases — 2% drop boiler efficiency for furnace sorbent injection is possible — — Depending on technology

 Fan power for flue gas recirculation (o0.5%)

 Flue gas recirculation: 0.2–0.8%a reduced boiler efficiency is possible  Steam injection: fuel consumption for steam production  Swirl flash system: swirl flash may increase power output while increase fuel input, leading to a net energy-efficiency improvement of 0–2%

 At some plants, the use of low-NOx burners can increase the share of unburned carbon caused by changes in furnace firing conditions, leading to a slight energy-efficiency drop a

Boiler efficiency loss is 0.0325 times share of FGR (COEN, 2006). Typical amounts of flue gas recirculation are 5–25% (IPPC, 2005).

Table 4 Level of applied NOx and SO2 pollution control (excluding planned pollution control) Coal

Australia China France Germany India Japan Korea Nordic countries UK+Ireland United States Power consumption

Gas

Oil

SO2 control (%)

NOx combustion control (%)

SCR (%)

NOx control (%)

SO2 control (%)

NOx control (%)

1 8 0 78 0 78 94 75 21 30

23 4 0 49 0 63 69 75 90 48

0 1 15 53 0 65 17 25 0 29

15 38 32 24 15 70 42 44 51 32

0 0 0 16 0 14 1 7 0 0

2 7 8 18 10 63 36 16 0 13

1.5

0.5

0.5

1

1.5

1

10% 9% 8% 7% 6% 5% 4% 3% 2% 1% 0% Nordic Japan countries

France

Korea

UK+ Ireland

United GermanyAustralia States

India

China

Fig. 10. Auxiliary power consumption in power and heat plants as share in total gross power generation in 2003 (IEA extended energy balances 2005).

ARTICLE IN PRESS W.H.J. Graus, E. Worrell / Energy Policy 35 (2007) 3898–3908

10

generation output together with national pollution control limits are in line with the found results. A second source of uncertainty is the effect of pollution control on energy efficiency of power generation. Although a reasonable amount of literature sources is available regarding power consumption of different pollution control technologies, the effects in terms of steam consumption and other possible effects on efficiency are limited. Especially for combustion modification for NOx control, the effects are complicated and depend on of a lot factors. Actual measurements of the effects at power plants were not available for this study. Although both sources of uncertainty limit the use of the detailed results of this study, the observed trends and relative magnitude of the found impacts for the analysed countries is unlikely to differ much.

data fields in the Platts database is empty, indicating that information on pollution abatement is not available. The overall uncertainty in the level of pollution control in a country is however estimated to be limited since the specific NOx and SO2 emissions on a per unit power Table 5 Gross and net fossil efficiencies with and without pollution control Gross efficiency 2003 (%)

Net efficiencya 2003 (%)

Power use for NOx and SO2 controlb (%)

Net efficiency excl. NOx and SO2 controlc (%)

Japan UK+Ireland Nordic countries Korea Germany United States France Australia China India

43.1 43.3 41.9 40.1 39.3 36.7 34.9 35.0 33.3 32.0

41.6 41.3 40.8 38.3 36.9 34.9 33.4 32.6 30.4 29.7

1.2 0.6 1.2 1.4 1.5 0.7 0.1 0.1 0.1 0.0

42.1 41.5 41.3 38.8 37.5 35.1 33.4 32.7 30.5 29.7

Source

Graus and Worrell (2006)

Estimate based on Fig. 10

Estimate based on Table 4 and Fig. 9

Result

6. Conclusion Pollution abatement technologies affect the net efficiency of power generation, as they often lead to increased internal consumption of power (and steam for specific technologies). NOx and SO2 emissions and the level of desulphurisation and denitrification seem to vary widely for the included countries. The highest level of desulphurisation and denitrification is present in the countries Japan, Germany, Nordic countries and South Korea. These countries also have the lowest NOx and SO2 emissions per kWh power generated. Limited pollution control is implemented in India and China, resulting in high specific NOx and SO2 emissions for these countries. The effect of NOx and SO2 control technologies on generation efficiency mainly consists of power consumption by the installations. The effect on net energy efficiency is estimated to be around 2% for coal-fired power plants (wet scrubber, SCR and combustion modification) and 1%

Net efficiency (%) ¼ gross efficiency (%)(1–own power consumption (%)). b Power use for NOx and SO2 control (as % of gross fossil power generation) ¼ SUMPRODUCT (power consumption for control technology (%) Table 4 last rowapplied level for country (%) Table 4share of coal, oil or gas in fossil-fired power generation (%), depending on control technology, Fig. 9). c Net efficiency (%) in case there would be no NOx and SO2 control ¼ net efficiency (%)/(1–power consumption for NOx and SO2 control (%)). a

45% 43%

Gross efficiency

41%

Net efficiency excl. NOx and SO2 control

39%

Net efficiency

37% 35% 33% 31% 29% 27%

a In di

a hi n C

ra lia Au st

an ce Fr

St

at

es

an y ni te d U

er m G

Ko re a

es tri co un

or di c N

U

K+

Ire

Ja

la nd

pa n

25%

Fig. 11. Gross efficiency and net fossil efficiency (with and without NOx and SO2 control).

ARTICLE IN PRESS W.H.J. Graus, E. Worrell / Energy Policy 35 (2007) 3898–3908

for gas-fired power plants (SCR and combustion modification). The effect on gross efficiency of SO2 and NOx control is at most 1% due to loss of boiler efficiency or additional fuel consumption, but probably lower. Power use for NOx and SO2 control is estimated to be between 1.2% and 1.5% of gross power output for countries with high levels of pollution control (Japan, Nordic countries, South Korea and Germany), leading to an effect on net efficiency of NOx and SO2 abatement of 0.5–0.6% points. Acknowledgements This analysis was funded by the Japanese Central Research Institute of Electric Power Industry (CRIEPI). We would like to thank CRIEPI for reviewing an earlier draft of this paper. Despite all their efforts, any remaining errors are the responsibility of the authors. The views expressed in this paper do not necessarily reflect those of CRIEPI. References Barrett, M., 2004. Atmospheric Emissions from Large Point Sources in Europe SENCO, UK. Battelle Memorial Institute, 1998. China’s Electric Power Options: An Analysis of Economic and Environmental Costs, Advanced International Studies Unit, Pacific Northwest National Laboratory, Washington, USA /http://www.pnl.gov/aisu/pubs/chinapwr.pdfS. California Air Resources Board, 2000. Air Resources Board (ARB) Guidance for Power Plant Siting and Best Available Control Technology, US /http://www.arb.ca.gov/energy/powerpl/appcfin. pdfS. Carnegie Mellon University, 2006. IECM Version 5.11. Integrated Environmental Control Model /http://www.netl.doe.gov/technologies/ coalpower/ewr/pubs/cmu-iecm.htmlS. COEN, 2006. Steam injection versus FGR for the reduction of thermal NOx. Technical Bulletin #20–103, US /http://www.coen.com/mrktlit/ brochures/pdf/tb20103.pdfS. EIA, 2002. Australia: Environmental Issues. Energy Information Administration (EIA), US Department of Energy (DOE), Washington, DC, US /http://www.eia.doe.gov/emeu/cabs/ausenv.htmlS. EIA, 2004. India: Environmental Issues. Energy Information Administration (EIA), US Department of Energy (DOE), Washington, DC, US /http://www.eia.doe.gov/emeu/cabs/indiaenv.htmlS. Entec, 2005. Preparation of the review relating to the large combustion plant directive. A Report for European Commission, Environment Directorate General, UK /http://www.eu.nl/environment/air/pdf/ final_report_05225.pdfS. Graus, W.H.J., Worrell, E., 2006. Comparison of efficiency fossil power generation. Ecofys, Utrecht, The Netherlands. ICAC, 2006. Acid Gas/SO2 Control Technologies, Institute of Clean Air Companies. Washington, US /http://www.icac.com/i4a/pages/index. cfm?pageid=3401S. IEA, 2005. Energy balances of OECD countries 1960–2003. Energy balances of non-OECD Countries 1971–2003. International Energy Agency (IEA), Paris, France.

11

IEA Clean Coal Centre, 2006. Wet Scrubbers for SO2 Control /http:// www.iea-coal.org.uk/content/default.asp?PageId=87S. IEACR, 2006. CleanPower5 Database, IEA Clean Coal Centre, UK /http://www.iea-al.org.uk/content/default.asp?PageId=60&LanguageId =0S. IPPC, 2005. Reference Document on Best Available Technology for Large Combustion Plants, Integrated Pollution Prevention and Control (IPPC), Seville, Spain. /http://eippcb.jrc.es/pages/FAbout.htmS. Jintian, Y., Dong, C.A.O., Chazhong G.E., Shuting, G.A.O., 2002. Air Pollution Control Strategy for China’s Power Sector, Chinese Academy for Environmental Planning, Beijing, China /http:// www.cse.polyu.edu.hk/activi/BAQ2002/BAQ2002_files/Powerpoint/ Subworkshop3/sw3a-3Yang_paper.docS. Kim, E.D., Kim, J.S., 1999. Air Pollutant Control Strategies for Thermal Power Plants in KEPCO, Korea Electric Power Corporation, South Korea /http://www.netl.doe.gov/publications/proceedings/99/99korea/ edkim.pdfS. LBNL, 2004. China Energy Databook. Version 6.0. June 2004. China Energy Group, Lawrence Berkeley National Laboratory (LBNL), Berkeley, US. Liere, J. van, Laagland, G.H.M., Meijer, C.G., 2001. Retrofit of Gas Turbines by SwirlFlash Over-spray, Alpha Power Systems and KEMA, Arnhem, The Netherlands /http://www.alphapowersystems. nl/NewOrleans.pdfS. Ministry of Environment Japan, 2006. Emission Standards for Soot and Dust, and NOx, Government of Japan /http://www.env.go.jp/en/air/ aq/air/air4.htmlS. OECD, 2004. OECD Environmental Data Compendium 2004. Organisation for Economic Co-operation and Development, Paris, France. OSC, 2006. Anthropogenic Emissions from Energy Activities in India: Generation and Source Characterization. Emissions from Coal Fired Thermal Power Plants in India, US /http://www.osc.edu/research/ pcrm/emissions/thermalemissions.shtmlS. Platts, 2006. World Electric Power Plant Database (WEPP), US /http:// www.platts.com/infostore/product_info.php?products_id=85S. Rubin, E.S., Berkenpas, M.B., Farrell, A. (Center for Energy and Environmental Studies), Gibbon, G.A., Smith, D.N. (National Energy Technology Laboratory, US DOE), 2002. Multi-pollutant Emission Control of Electric Power Plants, Pitsburgh, US. Schwarz, P.M., 2003. Multiple Pollutant Efficiency Standards for Electricity Production, Global Institute of Energy and Environmental Systems, UNC Charlotte and Senior Economist, Research Triangle Institute, US /http://enviro.colorado.edu/Ws2003/Ws2003/Schwarz. docS. Stork, 2005. SwirlFlash—The power of water. Gas Turbine Power Augmentation and NOx Reduction, The Netherlands /http://www. stork.com/thermeq/Downloads/TechnicalServices/STQ/SwirlFlash_ folder_E_2005_LR2.pdfS. Tanaka, N., 2000. Air Pollution in Urban Area in Japan, Air Quality Management Division. Environmental Management Bureau, Ministry of the Environment, Japanese Government /http://www.asiairnet.org/ publications/pubs/tanakadoc.pdfS. Tonooka, Y., Kannari, A., Murano, K., 2005. Emission inventory comparison: China case in 2002. In: MICS Asia Workshop, IIASA, 14–15 February 2005 /http://www.iiasa.ac.at/rains/meetings/ 7thMICS-Asia/papers/tonooka.pdfS. World Bank, 2001. Thermal Power—Guidelines for New Plants /http:// www.worldbank.or.jp/02event/01seminar/pdf_bs/bs17_material.pdfS. World Bank, 2006. Wet Flue Gas Desulfurization (FGD) /http:// www.worldbank.org/html/fpd/em/power/EA/mitigatn/aqsowet.stmS.