Ž.
The Science of the Total Environment259200097᎐103
Mercury emissions from a coal-fired power plant in
Japan
Takahisa Yokoyama a,U,Kazuo Asakura a,Hiromitsu Matsuda b,
Shigeo Ito b,Naoki Noda b
a Atmospheric Science Department,Komae Rearch Laboratory,Central Rearch Institute of Electric Power Industry,
2-11-1Iwato-kita,Komae-Shi,Tokyo201-8511,Japan
b Chemical Energy Engineering Department,Yokosuka Rearch Laboratory,Central Rearch Institute of Electri
c Power
Industry,2-6-1Nagasaka,Yokosuka-Shi,Kanagawa340-0196,Japan
Received27September1999;accepted30April2000
Abstract
The emissions study for mercury was conducted at a700MW coal-fired plant for the combustion of three types of
coal with mercury concentrations of0.0063,0.0367and0.065mg r kg.The power plant is equipped with a cold-side electrostatic precipitator and wet typeflue gas desulfurization system.During full load operation of the boilers, samples of the input and output streams such as coal,coal ash,ESP ash and post-ESP particulates andflue gas were
Ž. collected.The Hg concentrations in solid were measured by cold-vapor atomic absorption spectrometry AAS after appropriate preparation and acid digestion.Gaous Hg was collected using a mixed solution of potassium permanganate and sulfuric acid and the Hg concentrations in the samples were measured using cold-vapor AAS.The
Ž.Ž.Ž.
results were ud to examine:1overall mass balances;2relative distribution in the power plant;3equilibrium of杜甫怎么死的
Ž.
Hg species using MALT-2calculation program;and4Hg concentrations in stack emissions.The mass balances estimated in this study were100,138and89%,respectively,for the coals.Total Hg concentrations in stack gas were
1.113,0.422and0.712g3,respectively,for the coals.More than99.5%of the Hg in the stack emissions were in
m N
gaous form and the proportion in particulate form was extremely low.The relative distribution of Hg in ESP,FGD and Stack ranged from8.3to55.2%,13.3to69.2%and12.2%to44.4%,respectively.The results indicated that factors other than the Hg concentration of coals and efficiency of pollution control devices might affect Hg emissions from coal-fired plant.The calculated equilibrium of the distribution of Hg species using the MALT2program suggest that it is necessary to consider condens
ation mechanism to interpret the affect of Hg species on the variations of the removal efficiencies of Hg in the ESP.ᮊ2000Elvier Science B.V.All rights rerved.
Keywords:Mercury emissions;Coal-fired power plant;Mass balance;Japan
U Corresponding author.Tel.:q81-334802111,fax:q81-334801942.
Ž.
E-mail address:takahisa@jp T.Yokoyama.
0048-9697r00r$-e front matterᮊ2000Elvier Science B.V.All rights rerved.
Ž.
PII:S0048-96970000552-0
()
T.Yokoyama et al.r The Science of the Total En¨ironment259200097᎐103 98
1.Introduction
The Air Pollution Control Law was amended in 1996and a new environmental policy addressing hazardous air pollutants was added.Currently234 hazardous air pollutants are listed and approxi-mately20heavy metals such as mercury,arnic, etc.,are included.
Since the start of operation of large-scale coal-fired power plants in Japan in the1980s,the environmental effects of mercury have been a subject of high interest.Currently45million tons of different types of coal are burned annually in
生于忧患死于安乐教案
Ž
coal-fired power plants in Japan Coal Project, .
1999.Coal-fired power plants are anticipated to face more stringent environmental regulations re-lated to siting and operation.
The emissions study,sponsored by the Agency of National Resources and Energy,the Ministry of International Trade and Industry,was con-ducted at a700-MW coal-fired plant for the com-bustion of three types of coal,with mercury con-centration levels of0.0063,0.0367and0.065
Ž.
mg r kg.Bad on the study results,the:1over-
心声歌词
Ž.
all mass balances,2relative distribution in the
Ž.
power plant,3calculated equilibrium of Hg
Ž.
species,and4Hg concentrations in stack emis-sions was examined.
2.Outline of the study
2.1.Plant data
The coal-fired plant has a once-through super-critical type boiler.The plant data are summa-rized in Table1.This plant annually burns ap-proximately1.5million tons of bituminous coal which is mostly imported from Australia,China, the United States and Canada.The configuration diagram of the plant
is illustrated in Fig.1.The power plant is equipped with a low-NO burner,
x
two-staged combustion system,lective catalytic
Ž. reduction with ammonia injection SCR,cold-side electrostatic precipitator and wet typeflue gas desulfurization system in which a watered lime-stone slurry is ud to remove sulfur oxides to produce gypsum as its byproduct.
2.2.Sampling
报本During full load operation of the power plant, samples were collected from all the relevant input
Ž
and output streams:coal,coar ash clinker, cinder,air-heater ash,economizer ash,mechani-
.
cal cyclone ash,ESP ash,FGD gypsum,flue gas, process water.Three types of coal were burned. Sa
mpling was conducted three times for each type of coal per day,and was held over a period of3 days.The same type of coal was burned for5᎐6 days including the pre-sampling and during the sampling to provide a stable environment for sampling.Coal samples were collected hourly from a coal-bunker eight times per day.The grab samples were divided and mixed to prepare a reprentative composite sample for the test day. Coar ash and ESP were taken at the lected sampling ports of sample-collection hoppers twice per day.They were divided and mixed to prepare
Table1
Data of coal-fired plant having a once-through supercritical type boiler
Specification Description
Output700MW Fuel Bituminous coal Particulate control Cold side electrostatic precipitator
Ž.
140ЊC160ЊC
NO control Low-NO burners x x
Flue gas re-circulation
Two-staged combustion
SO control Wet type scrubber with x
limestone slurry as absorbent
()
T.Yokoyama et al.r The Science of the Total En¨ironment259200097᎐103
99
Fig.1.Schematic configuration of the coal-fired power plant equipped with a SCR,cold-side electrostatic precipitator and wet type FGD system and input r output streamflows.
reprentative samples for the test day.Other solid samples such as FGD gypsum and limestone were taken at the relevant facilities twice per day and prepared to form the test-day samples with the same procedure as ud for the ash samples. Process water and wastewater were also sampled to collect additional environmental information.
2.3.Measurement
The sample preparation and analytical method for determination of Hg concentrations in the collected samples in this study are summarized in Table2.The Hg concentrations in solids were measured by cold-vapor atomic absorption spec-trometry after appropriate preparation.
Total gaous Hg was collected under non-iso-kinetic conditions using a mixed solution of potas-
Ž钢琴曲纯音乐
sium permanganate and sulfuric acid Japane
.
Industrial Standard K0222Method in impingers and the Hg concentrations in the samples were measured using cold-vapor atomic absorption spectrometry followed by acid digestion.
3.Results and discussion
3.1.Mercury concentrations in coals ud in Japan Annual steam coal consumption by the electric utilities in Japan is approximately40million tons of bituminous coal,which is imported from Aus-tralia,Indonesia,China,The United States,South
ŽAfrica,Canada,Russia and New Zealand Coal
.
Project,1999.The share of Australian coal in the
Ž. total imported coals is the largest66%,FY97. Data for the mercury concentrations in59coal samples including26different coals from major
Ž
importing countries Australia,China,The United
. States,South Africa and domestic coal are sum-
Table2
Sample preparations and analytical methods for determination of Hg concentrations in the collected samples in this study Sample Sample preparation Analytical method
Coal Acid digestion Cold-vapor atomic
Ž.
Ash,Particulate H SO,HNO,KMnO,K S O absorption spectrometry
2434228
Acid digestion Cold-vapor atomic
Ž.
HCl,H SO,KMnO,K S O absorption spectrometry
244228
Process water,Acid digestion Cold-vapor atomic
Ž.
村晚雷震
wastewater H SO,HNO,KMnO,K S O absorption spectrometry
2434228
Gas Acid digestion
Ž.Ž
collected in impingers NH OHиHCl,H SO,HNO,Cold-vapor atomic
2243
.
KMnO,K S O absorption spectrometry
4228
()
T.Yokoyama et al.r The Science of the Total En¨ironment259200097᎐103
100
Table3
a
Hg concentrations in59samples including26different types of coal from USA,Australia,South Africa,China and Japan
Number of Mean value Standard deviation Minimum value Maximum value Ž.Ž.Ž.Ž. samples mg r kg mg r kg mg r kg mg r kg
590.0660.0530.0120.276
a Coals:US10,Australia7,South Africa4,China3,Japan2.
marized in Table3.Among them,for eight dif-ferent coals,multiple consignment samples were taken at different intervals of shipment for analy-sis.The coal samples reprent a large part of currently imported coal.The overall mean value of Hg concentrations in the59samples was0.066 mg r kg with a standard deviation of0.053mg r kg.
3.2.Coal data in this study
In this study three types of US coal with dif-ferent Hg concentrations were burned at differ-ent times.The coal data including Hg concentra-tions are shown in Table4.The Hg concentra-tions were equal to or less than tho of the mean concentrations in the59samples previously men-tioned.The major differences between three coals are the concentrations of mercury,chlorine,ash, oxygen and organic sulfur.Their concentrations are highest in Coal-3.
3.3.O¨erall mass balances
Overall mass balances of mercury on the basis of coal stream for combustion of three coals were Table4
Hg concentrations and the related fuel characteristics in the combusted coals
Component Coal-1Coal-2Coal-3Ž.
Hg mg r kg0.0370.00630.065Ž.
Cl mg r kg207411593
Ž.
Ash%8.84 5.8413.02
Ž.
C%72.6672.0772.04
Ž.
H% 4.56 4.42 4.74
Ž.
O%12.5516.457.72
Ž.
N% 1.14 1.13 1.37
Ž.
Combustible S%0.260.09 1.07
calculated as shown in Table5.The mass balance was100,138and89%for the coal types,Coal-1, Coal-2and Coal-3,respectively.Uncertainty is inherent to mass balance estimation.Uncertainty includesfluctuations in coal,ash,flue gasflow rates,sampling uncertainty and measurement un-certainty.It is usually difficult to estimate each uncertainty parately.If the mass balance clo-sures are between70and130%,it is assumed that the results are acceptable.The results of Coal-1and Coal-3coals were within acceptable
Table5
Mass balances of Hg between output and input streams in a coal-fired power plant for the combustion of three types of coal with different Hg concentrations
Stream Coal-1Coal-2Coal-3
Input Coal g r h7.450 1.33613.33
养殖泥鳅
Output Ash r particulates r gas g r h7.421 1.84611.813 Coar ash g r h0.0420.0080.219
Clinker g r h0.0180.0080.153
Economizer ash g r h0.0010.0000.002
Mechanical cyclone ash g r h0.0060.0000.017
Air heater ash g r h0.0160.0000.047 ESP ash g r h 4.0950.152 1.984
FGD reactants g r h 1.0090.8668.172
Stackflue gas g r h 2.2750.819 1.438
Ž.
Mass balance Input r output=100%10013889
()
T.Yokoyama et al.r The Science of the Total En¨ironment259200097᎐103
101
Fig.2.Volatility of trace elements during combustion being broadly classified into three groups depending on their behav-ior:non-vaporization,volatilization᎐condensation,volatiliza-tion᎐non-condensation.
mass balance closure results.The result of Coal-3 was slightly above130%.
3.4.Relati¨e distribution of mercury in the power plant
During combustion the pulverized coal parti-cles undergo complex changes including the for-mation of char,agglomeration of melted inclu-sions,and vaporization of volatile elements Ž.
Clarke and Sloss,1992.The trace elements bound in coal are distributed into coar ash,ESP ash,FGD reactants andflue gas.As shown in Fig. 2they are broadly classified into three groups
Ž
depending on their behavior Yokoyama et al.,
亚献. 1991;Clarke and Sloss,1992;Meij,1995.
()
ⅷGroup1non-¨aporization:
The elements which have low volatility at fur-
Ž.
nace temperature1200᎐1600ЊC and have no tendency to concentrate in or on combusted bottom ash andfly ash.
()
ⅷGroup2¨olatilization᎐condensation:
The elements which vaporize but then con-den,tend to concentrate in or on thefine particles within the temperature range of the
Ž
electric precipitator hot-side ESP:350ЊC;
.
cold-side ESP:140ЊC.
()
ⅷGroup3¨olatilization᎐non-condensation: The elements which are vaporized most read-ily,remain in the gas pha even atflue gas
Ž.
cleaning temperature50᎐60ЊC.
Mercury belongs to the Group3category and the relative distribution of Hg for the coals are summarized in Table 6.Coar ash includes clinker,economizer ash,mechanical cyclone ash and air heater ash.The relative distribution of Hg in coar ash,ESP,FGD and Stack ranged from 0.5to1.9%,8.3to55.2%,13.6to69.2%and12.2 to44.4%,respectively.The distribution profile varies by coal type.For combustion of Coal-3
Ž.
more than half of Hg69.2%was distributed or removed through the FGD,while for combustion
Ž.
of Coal-1more than50%of Hg55.2%was distributed or removed through the ESP.The difference of coal types were possibly a major factor which affects the control efficiency of ESP and FGD as the coal-fired power plant including ESP and FGD was operated under the same conditions for the combustion of the coals in this study.
3.5.Factors affecting Hg distribution
The forms of Hg species affects the removal
Ž
efficiencies influe gas control systems Moberg et al.,1982;Hall et al.,1991;Von Gutberlet et al., .
1992.
Table6
Relative distribution of Hg in the coal-fired power plant showing the variation by coal type
Ž.Ž.Ž.Ž. Coal type Coar ash%ESP%FGD%Stack% Coal-10.655.213.630.7 Coal-20.58.346.944.4 Coal-3 1.916.869.212.2
