- Email: [email protected]
Measurement of CO2, CO, SO2, and NO emissions from coal-based thermal power plants in India
ARTICLE IN PRESS
Atmospheric Environment 42 (2008) 1073–1082 www.elsevier.com/locate/atmosenv
Measurement of CO2, CO, SO2, and NO emissions from coal-based thermal power plants in India N. Chakrabortya,, I. Mukherjeea, A.K. Santraa, S. Chowdhurya, S. Chakrabortyb, S. Bhattacharyac, A.P. Mitrad, C. Sharmac a
Department of Power Engineering, Jadavpur University, 2nd Campus, Salt Lake, Kolkata 700098, India b Department of Mechanical Engineering, Indian Institute of Technology, Kharagpur 721302, India c Winrock International India, 1 Nav Jeevan Vihar, New Delhi 110057, India d National Physical Laboratory, Dr. KS Krishnan Road, New Delhi 110012, India Received 17 April 2007; received in revised form 30 July 2007; accepted 30 October 2007
Abstract Measurements of CO2 (direct GHG) and CO, SO2, NO (indirect GHGs) were conducted on-line at some of the coalbased thermal power plants in India. The objective of the study was three-fold: to quantify the measured emissions in terms of emission coefficient per kg of coal and per kWh of electricity, to calculate the total possible emission from Indian thermal power plants, and subsequently to compare them with some previous studies. Instrument IMR 2800P Flue Gas Analyzer was used on-line to measure the emission rates of CO2, CO, SO2, and NO at 11 numbers of generating units of different ratings. Certain quality assurance (QA) and quality control (QC) techniques were also adopted to gather the data so as to avoid any ambiguity in subsequent data interpretation. For the betterment of data interpretation, the requisite statistical parameters (standard deviation and arithmetic mean) for the measured emissions have been also calculated. The emission coefficients determined for CO2, CO, SO2, and NO have been compared with their corresponding values as obtained in the studies conducted by other groups. The total emissions of CO2, CO, SO2, and NO calculated on the basis of the emission coefficients for the year 2003–2004 have been found to be 465.667, 1.583, 4.058, and 1.129 Tg, respectively. r 2007 Elsevier Ltd. All rights reserved. Keywords: Thermal power plants; On-line measurement; Direct and indirect GHG; Emission coefficient
1. Introduction In relation to preparation of India’s National Communication (NATCOM) to the United Nations Framework Convention on Climate Change Corresponding author. Tel.: +91 332 416 7854;
fax: +91 332 414 7121. E-mail address: [email protected] (N. Chakraborty). 1352-2310/$ - see front matter r 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.atmosenv.2007.10.074
(UNFCCC) and to measure the amount of the direct (CO2) and indirect (CO, SO2, and NO) GHGs (Good Practice Guidance and Uncertainty Management in National Greenhouse Gas Inventories) from the coal-fed thermal power plants in India, a project work was undertaken. Measurements of these gases were carried out on-line using IMR 2800P Flue Gas Analyzer at 11 numbers of generating units of varying ratings over a period of 2 years during 2003–2004 on different days.
ARTICLE IN PRESS 1074
N. Chakraborty et al. / Atmospheric Environment 42 (2008) 1073–1082
The installed electricity generation capacity of the different coal-fed units were 60, 67.5, 210, and 250 MW, respectively, and most of the units were performing with a plant-load factor of almost equals to unity except for a particular plant where the plant-load factor varied between 0.667 and 1.00. Since the Indian power grids are well connected, power can be sold to any location in India. With this it has been observed that generations of the plants under observation were almost equal to the rated generation capacity, and this has been reflected in plant-load factor of almost equals to unity in most of the cases. The calculations for the emission coefficients have been made on the actual generations during the time of measurement. The age of the generating units varied from 5 to 20 years. Studies related to emission measurement and estimations from thermal power plants conducted by different researchers, scientists, and organizations (Mittal and Sharma, 2003b; Jorge et al., 2002; Ryerson et al., 1998; Gillani et al., 1998; Gurjar et al., 2004; Garg et al., 2001; TERI, 2001a; Varshney and Aggarwal, 1992; Chandra and Chandra, 2003; Modeling Anthropogenic Emissions from Energy Activities in India: Generation and Source Characterization) have confirmed the toxic potential of the measured gases particularly with respect to the increasing trend in temperature or in other words global warming and therein lies the importance of carrying out this project work to determine the amount of emissions of these gases particularly for a very fast developing economy like India. The emission coefficients for different gases have been calculated for different category of generating units by applying statistical methods. The figures have been calculated based on repeatedly measured values following IPCC guidelines (Good Practice Guidance and Uncertainty Management in National Greenhouse Gas Inventories). The consistencies of measured values of the generating units were also checked (see Table 1). Certain quality control measures as well as uncertainty reduction methods were adopted during the measurement process, and also during calculations to find out the emission coefficients. The variations in emission of the different gases for the different units of the thermal power plants have been dealt elaborately with specific and pertinent reasonings in the successive paragraphs of Section 3. The emission co-efficients for different gases were obtained for per kWh of electricity generated and per kg of coal utilized. The emission coefficients have been compared with the values as
Table 1 Consistency check in the measurement process with an interval of about 30 min at a fixed load Serial no.
Time of measurement (h)
Measured emission rate at generator unit S4U2 (installed capacity 67.5 MW) SO2 NO CO2 CO (%) (mg m3) (mg m3) (mg m3)
1 2 3 4 5 6 7 8 9 Arithmetic meana
12:09 12:35 13:06 13:34 14:41 15:08 15:35 16:02 16:37
12.0 12.2 12.6 12.4 12.4 12.4 12.9 12.4 13.1
40 30 40 40 40 40 40 40 40
634 636 657 647 541 552 558 556 628
516 509 488 480 488 486 471 476 449
12.49 40
601
484.78
a All the variations are within a range of 710% of the mean value. Emission figure of CO has been considered as 40 mg m3.
obtained in the previous studies (Gurjar et al., 2004; Modeling Anthropogenic Emissions from Energy Activities in India: Generation and Source Characterization). Further, the total estimated emission for CO2 has been compared with the study conducted by OSC (Modeling Anthropogenic Emissions from Energy Activities in India: Generation and Source Characterization). This is probably the first communication presented systematically by an Indian group on emissions only from thermal power plants which is based on measurements carried out on-line in a plant following standard experimental guidelines.
2. Experimental process The instrument used for measuring the direct and the indirect GHG emissions was the IMR 2800P Flue Gas Analyzer made by IMR Inc., USA (Operation Manual of IMR 2800P Flue Gas Analyzer). The instrument was provided with necessary pump and an interconnecting flexible hose with fixed thermocouple sensing wire. As soon as the instrument was switched on, self-calibration started automatically. Fresh air was drawn in by the in-built pump in the instrument from the normal atmosphere through the probe. It also purged out any gas/air present inside the instrument and finally
ARTICLE IN PRESS N. Chakraborty et al. / Atmospheric Environment 42 (2008) 1073–1082
calibration with respect to the oxygen present in the atmosphere at 20.9% value was done. At the time of measurement, electricity generation records and all other relevant details related to the measurement were noted from the generator control room. Other information on corresponding coal/oil inflow rates were also noted from control room. The instantaneous test results from the instrument indicated the following: (a) ambient atmospheric temperature (range, 20 to 120 1C); (b) temperature of flue gas (range, 20 to 1200 1C); (c) composition of flue gas in respect of carbon dioxide (CO2; range, 0–CO2 max. in percentage of volume); oxygen (O2; range, 0–20.9% of volume); carbon monoxide (CO; range, 0–2500 mg m3); sulfur dioxide (SO2; range, 0–11440 mg m3); nitrogen dioxide (NO2; range, 0–205 mg m3); nitric oxide (NO; range, 0–2680 mg m3); excess heat (range, 0–999%); and excess air (range, 0–999%) (Operation Manual of IMR 2800P Flue Gas Analyzer). 2.1. Quality assurance (QA)/quality control activity (QC) The experimental processes were undertaken following certain quality assurance (QA) and quality control (QC) protocol as formulated in the IPCC Good Practice Guidelines (Good Practice Guidance and Uncertainty Management in National Greenhouse Gas Inventories). In the present study, the following quality control measures were adopted: 1. Calibration of the measuring instrument IMR 2800P was done before measurement at each generator unit. Calibration of the instrument IMR 2800P was carried out at a location at the footsteps of the stack since at such locations, the possibility of mixing of the flue gas (emitting out of the stack at a considerable height) with the surrounding air was absolutely nil. So, it can be assumed that the calibration of the instrument with reference to the atmospheric oxygen was acceptable. 2. Measurements were conducted at accessible locations of the flue gas duct. The positions of the locations were maintained to be at sufficient distance from bends and obstructions in the flue ducts; thus possible disturbances arising out of irregular turbulence (due to bends or obstructions) was avoided as per CPCB guidelines
3.
4.
5.
6.
1075
(Environment Acts Rules and Notifications, 1999). The relevant activity data and measured emission figures of this key source category have been properly recorded as per IPCC guidelines (Good Practice Guidance and Uncertainty Management in National Greenhouse Gas Inventories). Measurements were taken repeatedly at regular time intervals to check consistency in emission values. For each source, data transcription and calculations were checked in respect of errors if any. All the measurement, calculations and conversion factors were checked for the purpose as per IPCC guidelines (Good Practice Guidance and Uncertainty Management in National Greenhouse Gas Inventories). Standard statistical methods were applied for data interpretation as per IPCC guidelines (Good Practice Guidance and Uncertainty Management in National Greenhouse Gas Inventories).
2.2. Statistical methods The estimation of the emission co-efficients of different gases has been categorized into four types depending upon the installed capacity of the plants. The emission coefficients have been obtained by calculating the arithmetic mean of data set as per IPCC guidelines (Good Practice Guidance and Uncertainty Management in National Greenhouse Gas Inventories) for each category of plant. The standard deviation of the data set (the square root of the variance) has also been calculated for each category of power plant. This in turn helped reducing the uncertainty in the measurement process. 2.3. Consistency of emission measurement and uncertainty reduction Consistency of the measured emissions was checked to avoid uncertainty and also to ascertain if there is any variation in emission over the period of measurement or not. It was observed that the emissions remain reasonably consistent in a plant throughout the time period of measurement. To illustrate this, a particular case where measurement was conducted over a period of 412 h in a thermal
ARTICLE IN PRESS N. Chakraborty et al. / Atmospheric Environment 42 (2008) 1073–1082
1076
power plant of generation capacity 67.5 MW and generating electricity at a constant rate of 67.5 MW has been provided (see Table 1). During this measurement process, coal supply rate was constant at 42 ton h1. Airflow rate for two numbers of induced draught (ID) fans was also observed to be more or less constant at 90 m3 s1. Measurement was conducted at an interval of approximately 30 min and the unit of measurement was mg m3 for all gases except for CO2, which was measured in percentage of volume. It was observed that the instantaneous emission rate of all the gases like CO2, CO, SO2, and NO were more or less constant over the time period (see Table 1). However, there were little variations in the measured figures (within 710% of mean value); this variation may be due to minor variations in the quality of corresponding coal getting burnt in the furnace. Apart from quality of coal, other factors affecting combustion such as burner positioning, amount of excess air may also result in variations in the emission rates. But when a plant is operating smoothly, it can be considered that the operating performances of all the plant machineries are more or less smooth and constant, and hence the rate of emission of gases will also remain more or less consistent. This process of checking consistency undoubtedly reduces uncertainty in the measurement process which ensures accuracy in measurement. With this, better interpretations of results are possible for quantification of emissions.
3. Results The average emission coefficients, their standard deviation values and range for each category of plant are given in Table 2. The emission coefficient for different gases has been calculated on the basis of actual measurement data. Measurement was recorded in ‘milligram per cubic meter’ and ‘percentage of volume’ units (as applicable in the measuring instrument) and such figures were converted to total emission per hour of electricity generation for each generator unit. The average emission coefficients for 60, 67.5, 210, and 250 MW category of plants have been calculated, respectively, as shown in Table 2. The standard deviations of such emission coefficients have also been calculated. It should be noted that, though coal quality varied from plant to plant, the combustion technology in all the plants is same which is based on pulverized coal burning. However, the type of furnace technology and design of boiler, forced draught (FD) and ID fans have effect on the combustion efficiency resulting variation in emission coefficient of measured gases. All the generating units where measurements were carried out, run with coal supplied from different collieries of Eastern Coalfields Limited, Bharat Coking Coal Limited, and Mahanadi Coalfields Limited. These collieries are located in India. However, sometimes imported coals with higher calorific value are mixed with Indian coals for better combustion. The coal is fed in pulverized form into the boiler furnace. Additionally,
Table 2 Average emission coefficient of different category of thermal power plant Plant (MW)
Emission coefficient
Emission per kg coal CO2 (kg)
CO (g)
Emission per unit (kWh) electricity SO2 (g)
NO (g)
CO2 (kg)
CO (g)
SO2 (g)
NO (g)
60
Average value S.D. Range
1.550 0.493 10.918 4.235 0.804 2.510 5.670 2.210 0.091 0.446 0.964 0.927 0.020 3.224 0.432 0.559 1.480–1.670 0.147–13.07b 9.530–11.720 3.120–5.320 0.776–0.824 0.080–7.140b 5.210–6.211 1.540–2.910
67.5
Average value Rangea
1.734 0.353 12.14 4.590 1.079 0.220 1.652–1.819 0.088–0.353 9.712–15.77 4.250–5.244 1.028–1.132 0.055–0.220
210
Average value S.D. Range
1.705 1.663 19.23 0.295 1.521 3.572 1.481–2.12 0.283–34.98b 15.29–22.85
3.61 0.595 2.84–4.27
1.197 10.680 13.460 2.525 0.208 11.345 2.500 0.415 1.037–1.49 0.198–24.49b 10.70–15.99 1.99–2.99
250
Average value S.D. Range
1.565 0.272 13.835 0.035 0.182 0.488 1.540–1.590 0.400–0.143 13.49–14.18
3.635 0.926 2.98–4.29
0.911 0.161 0.083 0.116 0.852–0.969 0.079–0.243
a
7.550 2.855 6.043–9.812 2.644–3.263
8.104 2.091 0.932 0.395 7.445–8.763 1.812–2.370
This range was measured at only one thermal power plant. As data on other units are not available, figure for S.D. has not been calculated. b Due to oil support in these thermal power plants, the CO emission is high, which has been reflected in the range of emission for CO.
ARTICLE IN PRESS N. Chakraborty et al. / Atmospheric Environment 42 (2008) 1073–1082
1077
categorized in accordance with the installed capacity of the generating units, and each generating unit has been given a name for identification. Other plant data relevant to the measured values, as available from the respective plant authority, are given in Table 4. The possible reasons behind the varying emission rate of different gases in the different plants are discussed in the following paragraphs. However, it will be prudent here to describe the plants in brief before results are presented. The generating units of 60 MW capacity, naming S1U1, S1U2, S1U3, and S1U4, were installed way back in the year 1982–1983; the oldest amongst the
diesel oil is used as and when required to supplement the combustion system for maintaining a stable combustion process during continuous running of the plant. Oil is also used during the starting process of the boiler. So far, no measurements could be taken during such start up process in any plant. All the generating units have individual electrostatic precipitators (ESP) and ID fans; but there was no scrubber in any unit for suppression of any gas. Information on corresponding coal and oil inflow rates were also noted from control room. The emission rates of different gases are given in Table 3. The various test results from different units have been
Table 3 Emission rates of green house gases from different thermal power generating units Generator unit
Installed in year
Installed capacity (MW)
Electricity generation and corresponding emission Generation (MW)
CO2 (kg h1)
CO (kg h1)
SO2 (kg h1)
NO (kg h1)
Flue gas temperature (1C)
60 60 60 60
60 60 60 60
48456.84 48508.67 46550.96 49420.8
428.53a 4.823 33.99 135.14a
312.51 372.64 327.45 347.07
130.210 174.41 92.36 133.99
127 128 139 154
1995 1997
250 250
250 250
242232.40 213002.82
60.85 19.79
2155.75 1861.95
453.10 592.58
137 127
S3U1 S3U3 S3U5 S3U6
1987–1988 1987–1988 1987–1988 1987–1988
210 210 210 210
175 140 175 200
260099.25 167914.63 181389.27 211836.74
4285.32a 375.97 34.63 3067.13a
2106.71 2238.86 2643.67 2139.92
523.52 369.96 434.72 397.53
133 126 127 143
S4U2
1990
72821.31
14.842
192.71
144
S1U1 S1U2 S1U3 S1U4
1982–1983 1982–1983 1982–1983 1982–1983
S2U1 S2U2
a
67.5
67.5
509.681
Due to oil support in these thermal power plants, the CO emission is high, which has been reflected in the range of emission for CO.
Table 4 Fuel feeding and ID fan delivery rates at the time of GHG measurement at different thermal power generating units Generator unit
Generation (MW)
Coal feed rate (ton h1)
Oil feed rate (l h1)
ID fan airflow rate (m3 s1)
Lambda (%)
Excess QA (%)
33 34 36 39
1.4 1.5 1.5 1.5
5.1 5.0 5.8 6.3
S1U1 S1U2 S1U3 S1U4
60 60 60 60
32.8 32.8 29.6 29.6
750 0 0 300
S2U1 S2U2
250 250
152.0 138.0
0 0
246 240
35 36
1.2 1.4
4.6 4.5
S3U1 S3U3 S3U5 S3U6
175 140 175 200
122.5 98.0 122.5 140.0
950 0 0 800
275 240 280 300
35 34 32 32
1.3 1.7 1.9 1.7
5.3 5.9 6.5 7.5
42.0
0
90
30
1.6
6.5
S4U2
67.5
55.0 58.5 58.9 59.6
Ambient temperature (1C)
ARTICLE IN PRESS 1078
N. Chakraborty et al. / Atmospheric Environment 42 (2008) 1073–1082
plants where measurement has been carried out so far. Being more than 20 year old units, their operational efficiency have definitely decreased in comparison to newer plants. Perhaps this is the main reason behind the occasional requirement of fuel oil along with coal in two of the generating units (see Table 4). It is understood from the plant authority that occasionally oil has to be used for maintaining stability of the combustion system in the boiler of different units; this is required when the volatile matter content of coal is o22%. The plant was designed for operation with coal containing 25% volatile matter. However, at the time of experimental measurement for a day it was observed that only units S1U1 and S1U4 were being fuelled with oil at the rate of 750 and 300 l h1, respectively. In these units, measurements were conducted at a horizontal location of rectangular ducts immediately after the location of ID fans (at a height of approximately 5 m from ground level) for individual units carrying flue gas to common stack. The test results are given in Tables 3 and 4, respectively. The plant-load factor for this plant was almost equals to unity at the time of measurements. Measurement was also conducted on a generator unit naming S4U2 of installed capacity 67.5 MW. In this unit, measurement was conducted at a horizontal location of the gas duct (one of two parallel ducts) before the stack. The height of the location was approximately 2 m from ground level. The plant was operating at its rated generation capacity at the time of the measurements. The plant was commissioned in the year 1990. The generating units of 210 MW capacity, naming S3U1, S3U3, S3U5, and S3U6, were designed for coal having 35% ash content. The designed coal supply requirement was 700 kg MWh1. But depending on the quality of available coal, the feeding rate varied from unit to unit, as and when required. The values of coal feeding rates in these four units (given in Table 4) were available verbally from the plant authority as no measured coal-feeding rate was available. Further, unit S3U5 and S3U6 were of improved version over the design of other two comparatively older units. In unit S3U1 and S3U3, measurement was conducted at a horizontal location of the rectangular flue gas duct (one of the four parallel ducts for each unit), in between ID fan and electrostatic precipitator. In unit S3U5 and S3U6, measurement was conducted at a horizontal location of the gas duct (one of two parallel ducts for each unit) just before the stack. In all the cases, the
height of measurement location was approximately 12 m from ground level. Except for one particular day, the plant-load factor varied in between 0.667 and 1.00. For almost all the other days of measurement, the plant-load factor was almost equals to unity. The 250 MW capacity generating unit were the highest in capacity where measurement were carried out. They were built and commissioned in the year 1995 and 1997; they are youngest of all thermal power plants where the measurements were conducted. So, it may be considered here that the efficiency of these two plants, naming S2U1 and S2U2, have not deteriorated much so far. In these two units, measurement on flue gas was carried out in the vertical stack at a height of 92.5 m from ground level. Naturally, the flue gas temperature must have reduced by few degrees centigrade after traveling a considerable distance. Here also, the plant-load factor was almost equals to unity on almost all the days of measurement. The emissions of the direct (CO2) and the indirect (CO, NO, and SO2) GHGs as measured for the different units of the respective thermal power plants designated as S1, S2, S3, and S4, respectively, have been reported in the following sections. 3.1. CO2 emissions from different plants From Table 3, it is seen that the average CO2 emission for S1U4 was maximum whereas that of unit S1U3 was minimum although the same amount of coal (29.6 ton h1) is being fed into both the units. The possible reason behind such a variation may have been the amount of excess air supplied and the combustion efficiency of the units. The carbon dioxide emissions from S1U1and S1U2 were more or less the same (see Table 3). It is also evident from Table 3 that the average CO2 emission rate from unit S3U1 was highest amongst the four units of that category (S3). The reason may be the higher carbon content of mixed coal being supplied to the unit. It was also noted that sometimes the color of flue gas emitting out of the unit was quite blackish. This may be due to additional oil support that was required to control the flame stability. The average CO2 emission from S3U2 was the second highest, whereas the average CO2 emission from the units S3U3 and S3U5 were 167914.63 and 181389.27 kg h1, respectively. The average CO2 emission from unit S2U1 was more than that of S2U2 because more coal
ARTICLE IN PRESS N. Chakraborty et al. / Atmospheric Environment 42 (2008) 1073–1082
(152 ton h1) was being supplied to S2U1 in comparison to that (138 ton h1) of unit S2U2 to meet same amount of electrical load. Measurement was also conducted on a generator unit (naming S4U2) of installed capacity 67.5 MW. The average emission rate of CO2 was 72821.31 kg h1. Here, no comparison could be made with other units of the same category, as measurement was not possible due to unavoidable circumstances in that plant. 3.2. CO emissions from different plants The average CO emission from unit S1U1 was highest as considerable amount of low speed diesel oil was injected along with coal due to lesser combustion capability of the unit (see Table 3). Complete combustion of coal and oil was not possible resulting in higher carbon monoxide emission from the unit. SIU4 recorded the second highest CO emission, as here also additional oil support was provided but to a lesser extent as that to S1U1. The average CO emission from S1U2 was lowest which can be attributed to better combustion efficiency of the system. The average CO emission from S1U3 was 33.99 kg h1. The average CO emission from unit S3U1 was very high. Due to higher coal flow rate and low air supply, combustion was not complete resulting in emission of higher amount of carbon monoxide. However, additional oil support in this unit may be the main reason for such a high CO emission. Additional oil support was also provided to the unit S3U6 but to a lesser extent as compared to S3U1. It recorded the second highest CO emission value. The average CO emission from unit S3U5 was the lowest and this may be attributed to better combustion in the system. The average CO emission from the unit S3U3 was 375.97 kg h1. The average CO emission from unit S2U1 was comparably higher than S2U2 and this may be attributed to better combustion efficiency of the unit S2U2 compared to S2U1. For the generator unit (naming S4U2) of installed capacity 67.5 MW, the average emission rate of CO was measured to be 14.842 kg h1.
1079
attributed to the varying sulfur content of coal fed into the respective units. The average SO2 emissions from all the four units in the S1 category were more or less constant with the unit S1U1 emitting the minimum. The ranges of the SO2 emission for this category of units lie in the range 312.51–372.64 kg h1 (see Table 3). The average SO2 emission for the four units in the S3 category was more or less equal with the maximum emission being recorded in the unit S3U5. The average SO2 emission rates of S3U1, S3U3, S3U5, and S3U6 were 2106.71, 2238.86, 2643.67, and 2139.92 kg h1, respectively. The average SO2 emission for the unit S2U1 was 2155.75 kg h1 whereas that from unit S2U2 was 1861.95 kg h1. Average SO2 emission measured for the category S4U2 was found to be 509.681 kg h1. 3.4. NO emissions from different plants The basic reason behind the variation of NO emissions from the different categories may be attributed to the varying amount of fixed nitrogen content in coal and the variable performance efficiency of the burners of the different categories. From Table 3, it can be seen that the average NO emission from unit S1U2 was the highest. The average NO emissions as measured for the other units S1U1, S1U3, and S1U4 are 130.210, 92.36, and 133.99 kg h1, respectively. The average NO emission from the unit S3U1 of the S3 category was the highest amongst themselves. The possible reason can be attributed to the lesser performance efficiency of the burner compared to the other units of the same category. The emissions from the other units of this category were more or less at par with each other (see Table 3). The average NO emission from unit S2U1 was less in comparison to that of unit S2U2. The reason may be attributed to better performance of burners in S2U1 compared to that of S2U2, thereby producing lesser amount of NO in S2U1 than in S2U2. Another possible reason may be the lesser amount of fixed nitrogen in coal fed into that unit. Measurement was also conducted on a generator unit (naming S4U2) of installed capacity 67.5 MW. The average emission rate of NO was recorded to be 192.71 kg h1.
3.3. SO2 emissions from different plants
4. Discussion
The main reason behind the variation in SO2 emission for the different categories can mostly be
Thermal power plants are one of the main sources of GHG emission throughout the world. For a very
ARTICLE IN PRESS 1080
N. Chakraborty et al. / Atmospheric Environment 42 (2008) 1073–1082
fast developing economy like India, measurement of GHGs’ from the thermal power plants are very much essential in order to find out their values so that necessary policies of reduction of such gases can be formulated. In order to develop strategies of GHG reduction, a realistic inventory of GHG is very much required. For that, determination of methodology-based measured emission coefficients are very much essential. These emission coefficients then will not only help developing the national inventory but will help comparing the national emission with the emission of the developed countries. The ranges of the values of the emissions in per kWh of electricity generated and in per kg of coal utilized obtained in the present study are provided in Table 5 in comparison to those obtained by OSC (Modeling Anthropogenic Emissions from Energy Activities in India: Generation and Source Characterization) and by Gurjar et al. (2004). The emission coefficients calculated in the present study are more or less at par with the values obtained by Gurjar et al. (2004), except for the two indirect GHGs CO and NO (see Table 5). The main variation is likely due to variable combustion technologies in the respective units. At the same time variation of coal quality with regard to carbon content and fixed nitrogen content may have an indirect effect on the variation of CO and NO formation. Particularly in case of CO, additional oil support in some of the units of the plants (see Table 6) may be the other possible reason for this mismatch in CO values. While the co-efficient values obtained for CO2 and SO2 matches closely with that obtained by OSC (Modeling Anthropogenic Emissions from Energy Activities in India: Generation and Source Characterization), a more or less four times higher coefficient values for NO has been reported by OSC. Obviously, there is no doubt that this needs to be
further investigated. However, roughly the differences can be attributed to the varying composition of coal fed into the machine and the combustion process in the unit and their performance. Furthermore, it is worth mentioning that the present study is totally on-line measurement based whereas the study by OSC is based on theoretical calculations. It has been considered that a total of 466,600 GWh electrical energy was produced from coal-fed thermal power plants for the year 2003–2004 (http:// indiabudget.nic.in) in India. This excludes contribution by the captive and other non-conventional power plants. To calculate the total emission of each of these toxic gases, the total range of emission (per kWh of electricity) from all categories of power plants has been selected and their mean emission coefficient has been calculated first. With that the total emission for the year 2003–2004 has been estimated. From Table 7, it can be observed that the corresponding CO2, CO, SO2, and NO values are Table 6 Increased Emission Coefficients due to additional oil support Plant Generation Gas type designation capacity per unit (MW)
Emission
S1
60
CO
(a) 13.07 g kg1 of coal (b) 7.140 g kWh1 of electricity
S3
210
CO
(a) 34.98 g kg1 of coal (b) 24.49 g kWh1 of electricity
Note: The emission rates of CO in the units SIU1 (428.53 kg h1), SIU4 (135.14 kg h1), S3U1 (4285.32 kg h1), and in S3U6 (3067.13 kg h1) were high compared to the other units of the plant.
Table 5 Comparison of emission coefficients Gaseous type
CO2 SO2 NO CO a
Range of measured emission coefficient
Emission coefficients
Present study for the year 2003–2004
Study by OSC in the year 1997–1998
Gurjar et al. (2004)
Present study
0.776–1.49 (kg kWh1) 5.210–15.99 (g kWh1) 1.540–3.263 (g kWh1) 0.055–24.49b
0.8–1.8 (kg kWh1) 4–18 (g kWh1) 6–13.1 (g kWh1) Not available
1.739 (kg kg1 of coal) 14.767 (g kg1 of coal) 0.824a (g kg1 of coal) 0.253 (g kg1 of coal)
1.639 (kg kg1 of coal) 14.031 (g kg1 of coal) 4.018 (g kg1 of coal) 5.153 (g kg1 of coal)
NOx as NO2 measured by Gurjar as 1.263, converted to NO multiplying with a factor of 0.652. Additional oil support.
b
ARTICLE IN PRESS N. Chakraborty et al. / Atmospheric Environment 42 (2008) 1073–1082
465.667, 1.583, 4.058, and 1.129 Tg, respectively, for the year 2003–2004. Considering CO2 to be the major contributor to the green house effect, the measured emission of 465.667 Mton for the year 2003–2004 has been compared with the calculated value obtained by OSC for the year 1997–1998 which was found to be 395 Mton. The thermal power generation for 1997–1998 was 336,000 GWh (Modeling Anthropogenic Emissions from Energy Activities in India: Generation and Source Characterization) while that for the period 2003–2004 was 466,600 GWh (http:// indiabudget.nic.in). This reflects that there is not much variation exists between the present study and that reported by OSC (Modeling Anthropogenic Emissions from Energy Activities in India: Generation and Source Characterization) at least for carbon dioxide taking into consideration the increased generation of electricity from 1997–1998 to 2003–2004. 5. Conclusion From the foregoing discussions, it can be generalized that the emission rates of the direct and indirect GHGs obtained from the Indian thermal power plants varies on various accounts; such as (a) quality of coal mixture; (b) quality of oil, wherever used; (c) quantity of coal and oil required for per unit generation; (d) age of the plant and its maintenance standard; and (e) amount of excess air fed into the furnace. The quality of coal (from different sources) is variable over a wide range. For a particular power plant, a single quality standard cannot be adhered
1081
to, as the mixing ratio of different quality of coal cannot be specified. At the same time, the mixture remains non-homogenous. Moreover, the quantity of coal and oil requirement varies in accordance with their calorific value to meet a specific electrical demand. Again efficiency of the plant and equipment, particularly the combustion and heat absorption is varying in nature. The efficiency is again dependent on the age of the plant and the standard of regular preventive maintenance. Without regular preventive maintenance of burners and air inlet valves, the combustion efficiency cannot be maintained at a desired standard level. Further, arrangements cannot be made for homogenous mixing of air and fuel at every location of the furnace; resulting in non-homogenous temperature zones in the furnace resulting in the variation in both (direct as well as indirect) GHG production. At the same time, the amount of excess air (supplied to the furnace) plays an important role in the combustion and generation of NO and SO2. The amount of electricity generation during different hours of a day is also a factor controlling the GHG production. Even if a plant runs at no load or little load, some fuel has to be supplied for maintaining stability of the furnace. Also all the plants do not run at full load all the time, nor the generation pattern over a day or month is the same for a particular plant. As the variability of all the above factors in all the generating plants in India remains random in nature, the emission coefficients cannot be quantified easily. So, establishing a relationship between above factors and the emission coefficient is not straightforward. Only, more accuracy in estimating total GHG emission can be achieved if measurement is conducted on almost all the power plants in India. To find out qualitatively better emission coefficients, on-line measurement should
Table 7 Total estimated emission of green house gas from Indian thermal power plants in the year 2003–2004 Emission per unit (kWh) of electricity
Range of emission from all power plants Average emission coefficient Total estimated emission during year 2003–2004a (Tg)
CO2
CO
SO2
NO
0.776–1.49 kg 0.998 kg 465.667
0.055–24.49 g 3.393 g 1.583
5.21–15.99 g 8.696 g 4.058
1.54–3.263 g 2.420 g 1.129
Note: 1 Tg ¼ 1 Mton ¼ 1 million metric ton. a Total generation from coal fired thermal power plants in India, has been considered to be 466,600 GWh for the year 2003–2004 (http:// indiabudget.nic.in).
ARTICLE IN PRESS 1082
N. Chakraborty et al. / Atmospheric Environment 42 (2008) 1073–1082
be conducted at all such thermal power plants in India. This will certainly help to determine the total emission as well as strategies of reduction of the GHGs’ from thermal plants in India. The present study sets the direction to such an endeavor. Acknowledgments We express our gratitude to the Ministry of Environment and Forest, Government of India, for selecting us to execute the project. We are grateful to United Nations Development Programme and Winrock International India for their financial support in executing this project. We also acknowledge the support extended by National Physical Laboratory, New Delhi, Central Fuel Research Institute, Dhanbad, with their valued advice in doing this project. We are indebted to Ministry of Power, Government of India, Ministry of Power and Ministry of Environment and Forest, Government of West Bengal, Central Pollution Control Board (Government of India), West Bengal Pollution Control Board (Government of West Bengal) and authorities of different power plants in West Bengal for their active support in executing this project. Lastly, we express our thanks to the authorities of Jadavpur University, without whose all-out support it would not have been possible to make this endeavor a success. References Chandra, A., Chandra, H., 2003. Environmental management for clean power generation of thermal power plants: an Indian perspective. Indian Journal of Air Pollution Control 3 (1), 22–36.
Environment Acts Rules and Notifications, Department of Environment, Government of West Bengal, March 1999. Garg, A., Bhattacharya, S., Shukla, R.P., Dadhwal, K.V., 2001. Regional and sectoral assessment of greenhouse gas emissions in India. Atmospheric Environment 35, 2679–2695. Gillani, N.V., Meagher, J.F., Valente, R.J., Imhoff, R.E., Tanner, R.L., Luria, M., 1998. Relative production of ozone and nitrates in urban and rural power plant plumes, 1. Composite results based on data from 10 field measurement days. Journal of Geophysical Research 103, 22593–22615. Good Practice Guidance and Uncertainty Management in National Greenhouse Gas Inventories. /http://www. ipcc-nggip.iges.or.jp/public/gp/gpgaum.htmS. Gurjar, R.B., Van Aardenne, A.J., Lelieveld, J., Mohan, M., 2004. Emission estimates and trends (1990–2000) for mega city Delhi and implication. Atmospheric Environment 38, 5663–5681. http://indiabudget.nic.in. Jorge, I., Manzini, F., Martinez, M., 2002. Renewable energies in electricity generation for reduction of greenhouse gases in Mexico 2025. AMBIO 31, 35–39. Mittal, M.L., Sharma, C., 2003b. Anthropogenic emissions from energy activities in India: generation and source characterization. Part I: emissions from thermal power generation in India. URL: /http://www.osc.edu/research/pcrm/emissions/ India.pdfS. Modeling anthropogenic emissions from energy activities in India: generation and source characterization. /www.osc. edu/research/pcrm/emissions/contents.shtmlS. Operation Manual of IMR 2800P Flue Gas Analyzer. /www. imrusa.comS. Ryerson, T.B., Buhr, M.P., Frost, G.J., Goldan, P.D., Holloway, J.S., Hubler, G., Jobson, B.T., Kuster, W.C., McKeen, S.A., Parrish, D.D., Roberts, J.M., Sueper, D.T., Trainer, M., Williams, J., Fehsenfeld, C.F., 1998. Emissions lifetimes and ozone formation in power plant plumes. Journal of Geophysical Research 103, 22569–22583. TERI (The Energy and Resources Institute), 2001a. State of environment report for Delhi 2001. Report No. 2000EE65, Supported by the Department of Environment, Government of National Capital Territory, Delhi. Varshney, K.C., Aggarwal, M., 1992. Ozone pollution in the urban environment of Delhi. Atmospheric Environment 3, 291–294.
Atmospheric Environment 42 (2008) 1073–1082 www.elsevier.com/locate/atmosenv
Measurement of CO2, CO, SO2, and NO emissions from coal-based thermal power plants in India N. Chakrabortya,, I. Mukherjeea, A.K. Santraa, S. Chowdhurya, S. Chakrabortyb, S. Bhattacharyac, A.P. Mitrad, C. Sharmac a
Department of Power Engineering, Jadavpur University, 2nd Campus, Salt Lake, Kolkata 700098, India b Department of Mechanical Engineering, Indian Institute of Technology, Kharagpur 721302, India c Winrock International India, 1 Nav Jeevan Vihar, New Delhi 110057, India d National Physical Laboratory, Dr. KS Krishnan Road, New Delhi 110012, India Received 17 April 2007; received in revised form 30 July 2007; accepted 30 October 2007
Abstract Measurements of CO2 (direct GHG) and CO, SO2, NO (indirect GHGs) were conducted on-line at some of the coalbased thermal power plants in India. The objective of the study was three-fold: to quantify the measured emissions in terms of emission coefficient per kg of coal and per kWh of electricity, to calculate the total possible emission from Indian thermal power plants, and subsequently to compare them with some previous studies. Instrument IMR 2800P Flue Gas Analyzer was used on-line to measure the emission rates of CO2, CO, SO2, and NO at 11 numbers of generating units of different ratings. Certain quality assurance (QA) and quality control (QC) techniques were also adopted to gather the data so as to avoid any ambiguity in subsequent data interpretation. For the betterment of data interpretation, the requisite statistical parameters (standard deviation and arithmetic mean) for the measured emissions have been also calculated. The emission coefficients determined for CO2, CO, SO2, and NO have been compared with their corresponding values as obtained in the studies conducted by other groups. The total emissions of CO2, CO, SO2, and NO calculated on the basis of the emission coefficients for the year 2003–2004 have been found to be 465.667, 1.583, 4.058, and 1.129 Tg, respectively. r 2007 Elsevier Ltd. All rights reserved. Keywords: Thermal power plants; On-line measurement; Direct and indirect GHG; Emission coefficient
1. Introduction In relation to preparation of India’s National Communication (NATCOM) to the United Nations Framework Convention on Climate Change Corresponding author. Tel.: +91 332 416 7854;
fax: +91 332 414 7121. E-mail address: [email protected] (N. Chakraborty). 1352-2310/$ - see front matter r 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.atmosenv.2007.10.074
(UNFCCC) and to measure the amount of the direct (CO2) and indirect (CO, SO2, and NO) GHGs (Good Practice Guidance and Uncertainty Management in National Greenhouse Gas Inventories) from the coal-fed thermal power plants in India, a project work was undertaken. Measurements of these gases were carried out on-line using IMR 2800P Flue Gas Analyzer at 11 numbers of generating units of varying ratings over a period of 2 years during 2003–2004 on different days.
ARTICLE IN PRESS 1074
N. Chakraborty et al. / Atmospheric Environment 42 (2008) 1073–1082
The installed electricity generation capacity of the different coal-fed units were 60, 67.5, 210, and 250 MW, respectively, and most of the units were performing with a plant-load factor of almost equals to unity except for a particular plant where the plant-load factor varied between 0.667 and 1.00. Since the Indian power grids are well connected, power can be sold to any location in India. With this it has been observed that generations of the plants under observation were almost equal to the rated generation capacity, and this has been reflected in plant-load factor of almost equals to unity in most of the cases. The calculations for the emission coefficients have been made on the actual generations during the time of measurement. The age of the generating units varied from 5 to 20 years. Studies related to emission measurement and estimations from thermal power plants conducted by different researchers, scientists, and organizations (Mittal and Sharma, 2003b; Jorge et al., 2002; Ryerson et al., 1998; Gillani et al., 1998; Gurjar et al., 2004; Garg et al., 2001; TERI, 2001a; Varshney and Aggarwal, 1992; Chandra and Chandra, 2003; Modeling Anthropogenic Emissions from Energy Activities in India: Generation and Source Characterization) have confirmed the toxic potential of the measured gases particularly with respect to the increasing trend in temperature or in other words global warming and therein lies the importance of carrying out this project work to determine the amount of emissions of these gases particularly for a very fast developing economy like India. The emission coefficients for different gases have been calculated for different category of generating units by applying statistical methods. The figures have been calculated based on repeatedly measured values following IPCC guidelines (Good Practice Guidance and Uncertainty Management in National Greenhouse Gas Inventories). The consistencies of measured values of the generating units were also checked (see Table 1). Certain quality control measures as well as uncertainty reduction methods were adopted during the measurement process, and also during calculations to find out the emission coefficients. The variations in emission of the different gases for the different units of the thermal power plants have been dealt elaborately with specific and pertinent reasonings in the successive paragraphs of Section 3. The emission co-efficients for different gases were obtained for per kWh of electricity generated and per kg of coal utilized. The emission coefficients have been compared with the values as
Table 1 Consistency check in the measurement process with an interval of about 30 min at a fixed load Serial no.
Time of measurement (h)
Measured emission rate at generator unit S4U2 (installed capacity 67.5 MW) SO2 NO CO2 CO (%) (mg m3) (mg m3) (mg m3)
1 2 3 4 5 6 7 8 9 Arithmetic meana
12:09 12:35 13:06 13:34 14:41 15:08 15:35 16:02 16:37
12.0 12.2 12.6 12.4 12.4 12.4 12.9 12.4 13.1
40 30 40 40 40 40 40 40 40
634 636 657 647 541 552 558 556 628
516 509 488 480 488 486 471 476 449
12.49 40
601
484.78
a All the variations are within a range of 710% of the mean value. Emission figure of CO has been considered as 40 mg m3.
obtained in the previous studies (Gurjar et al., 2004; Modeling Anthropogenic Emissions from Energy Activities in India: Generation and Source Characterization). Further, the total estimated emission for CO2 has been compared with the study conducted by OSC (Modeling Anthropogenic Emissions from Energy Activities in India: Generation and Source Characterization). This is probably the first communication presented systematically by an Indian group on emissions only from thermal power plants which is based on measurements carried out on-line in a plant following standard experimental guidelines.
2. Experimental process The instrument used for measuring the direct and the indirect GHG emissions was the IMR 2800P Flue Gas Analyzer made by IMR Inc., USA (Operation Manual of IMR 2800P Flue Gas Analyzer). The instrument was provided with necessary pump and an interconnecting flexible hose with fixed thermocouple sensing wire. As soon as the instrument was switched on, self-calibration started automatically. Fresh air was drawn in by the in-built pump in the instrument from the normal atmosphere through the probe. It also purged out any gas/air present inside the instrument and finally
ARTICLE IN PRESS N. Chakraborty et al. / Atmospheric Environment 42 (2008) 1073–1082
calibration with respect to the oxygen present in the atmosphere at 20.9% value was done. At the time of measurement, electricity generation records and all other relevant details related to the measurement were noted from the generator control room. Other information on corresponding coal/oil inflow rates were also noted from control room. The instantaneous test results from the instrument indicated the following: (a) ambient atmospheric temperature (range, 20 to 120 1C); (b) temperature of flue gas (range, 20 to 1200 1C); (c) composition of flue gas in respect of carbon dioxide (CO2; range, 0–CO2 max. in percentage of volume); oxygen (O2; range, 0–20.9% of volume); carbon monoxide (CO; range, 0–2500 mg m3); sulfur dioxide (SO2; range, 0–11440 mg m3); nitrogen dioxide (NO2; range, 0–205 mg m3); nitric oxide (NO; range, 0–2680 mg m3); excess heat (range, 0–999%); and excess air (range, 0–999%) (Operation Manual of IMR 2800P Flue Gas Analyzer). 2.1. Quality assurance (QA)/quality control activity (QC) The experimental processes were undertaken following certain quality assurance (QA) and quality control (QC) protocol as formulated in the IPCC Good Practice Guidelines (Good Practice Guidance and Uncertainty Management in National Greenhouse Gas Inventories). In the present study, the following quality control measures were adopted: 1. Calibration of the measuring instrument IMR 2800P was done before measurement at each generator unit. Calibration of the instrument IMR 2800P was carried out at a location at the footsteps of the stack since at such locations, the possibility of mixing of the flue gas (emitting out of the stack at a considerable height) with the surrounding air was absolutely nil. So, it can be assumed that the calibration of the instrument with reference to the atmospheric oxygen was acceptable. 2. Measurements were conducted at accessible locations of the flue gas duct. The positions of the locations were maintained to be at sufficient distance from bends and obstructions in the flue ducts; thus possible disturbances arising out of irregular turbulence (due to bends or obstructions) was avoided as per CPCB guidelines
3.
4.
5.
6.
1075
(Environment Acts Rules and Notifications, 1999). The relevant activity data and measured emission figures of this key source category have been properly recorded as per IPCC guidelines (Good Practice Guidance and Uncertainty Management in National Greenhouse Gas Inventories). Measurements were taken repeatedly at regular time intervals to check consistency in emission values. For each source, data transcription and calculations were checked in respect of errors if any. All the measurement, calculations and conversion factors were checked for the purpose as per IPCC guidelines (Good Practice Guidance and Uncertainty Management in National Greenhouse Gas Inventories). Standard statistical methods were applied for data interpretation as per IPCC guidelines (Good Practice Guidance and Uncertainty Management in National Greenhouse Gas Inventories).
2.2. Statistical methods The estimation of the emission co-efficients of different gases has been categorized into four types depending upon the installed capacity of the plants. The emission coefficients have been obtained by calculating the arithmetic mean of data set as per IPCC guidelines (Good Practice Guidance and Uncertainty Management in National Greenhouse Gas Inventories) for each category of plant. The standard deviation of the data set (the square root of the variance) has also been calculated for each category of power plant. This in turn helped reducing the uncertainty in the measurement process. 2.3. Consistency of emission measurement and uncertainty reduction Consistency of the measured emissions was checked to avoid uncertainty and also to ascertain if there is any variation in emission over the period of measurement or not. It was observed that the emissions remain reasonably consistent in a plant throughout the time period of measurement. To illustrate this, a particular case where measurement was conducted over a period of 412 h in a thermal
ARTICLE IN PRESS N. Chakraborty et al. / Atmospheric Environment 42 (2008) 1073–1082
1076
power plant of generation capacity 67.5 MW and generating electricity at a constant rate of 67.5 MW has been provided (see Table 1). During this measurement process, coal supply rate was constant at 42 ton h1. Airflow rate for two numbers of induced draught (ID) fans was also observed to be more or less constant at 90 m3 s1. Measurement was conducted at an interval of approximately 30 min and the unit of measurement was mg m3 for all gases except for CO2, which was measured in percentage of volume. It was observed that the instantaneous emission rate of all the gases like CO2, CO, SO2, and NO were more or less constant over the time period (see Table 1). However, there were little variations in the measured figures (within 710% of mean value); this variation may be due to minor variations in the quality of corresponding coal getting burnt in the furnace. Apart from quality of coal, other factors affecting combustion such as burner positioning, amount of excess air may also result in variations in the emission rates. But when a plant is operating smoothly, it can be considered that the operating performances of all the plant machineries are more or less smooth and constant, and hence the rate of emission of gases will also remain more or less consistent. This process of checking consistency undoubtedly reduces uncertainty in the measurement process which ensures accuracy in measurement. With this, better interpretations of results are possible for quantification of emissions.
3. Results The average emission coefficients, their standard deviation values and range for each category of plant are given in Table 2. The emission coefficient for different gases has been calculated on the basis of actual measurement data. Measurement was recorded in ‘milligram per cubic meter’ and ‘percentage of volume’ units (as applicable in the measuring instrument) and such figures were converted to total emission per hour of electricity generation for each generator unit. The average emission coefficients for 60, 67.5, 210, and 250 MW category of plants have been calculated, respectively, as shown in Table 2. The standard deviations of such emission coefficients have also been calculated. It should be noted that, though coal quality varied from plant to plant, the combustion technology in all the plants is same which is based on pulverized coal burning. However, the type of furnace technology and design of boiler, forced draught (FD) and ID fans have effect on the combustion efficiency resulting variation in emission coefficient of measured gases. All the generating units where measurements were carried out, run with coal supplied from different collieries of Eastern Coalfields Limited, Bharat Coking Coal Limited, and Mahanadi Coalfields Limited. These collieries are located in India. However, sometimes imported coals with higher calorific value are mixed with Indian coals for better combustion. The coal is fed in pulverized form into the boiler furnace. Additionally,
Table 2 Average emission coefficient of different category of thermal power plant Plant (MW)
Emission coefficient
Emission per kg coal CO2 (kg)
CO (g)
Emission per unit (kWh) electricity SO2 (g)
NO (g)
CO2 (kg)
CO (g)
SO2 (g)
NO (g)
60
Average value S.D. Range
1.550 0.493 10.918 4.235 0.804 2.510 5.670 2.210 0.091 0.446 0.964 0.927 0.020 3.224 0.432 0.559 1.480–1.670 0.147–13.07b 9.530–11.720 3.120–5.320 0.776–0.824 0.080–7.140b 5.210–6.211 1.540–2.910
67.5
Average value Rangea
1.734 0.353 12.14 4.590 1.079 0.220 1.652–1.819 0.088–0.353 9.712–15.77 4.250–5.244 1.028–1.132 0.055–0.220
210
Average value S.D. Range
1.705 1.663 19.23 0.295 1.521 3.572 1.481–2.12 0.283–34.98b 15.29–22.85
3.61 0.595 2.84–4.27
1.197 10.680 13.460 2.525 0.208 11.345 2.500 0.415 1.037–1.49 0.198–24.49b 10.70–15.99 1.99–2.99
250
Average value S.D. Range
1.565 0.272 13.835 0.035 0.182 0.488 1.540–1.590 0.400–0.143 13.49–14.18
3.635 0.926 2.98–4.29
0.911 0.161 0.083 0.116 0.852–0.969 0.079–0.243
a
7.550 2.855 6.043–9.812 2.644–3.263
8.104 2.091 0.932 0.395 7.445–8.763 1.812–2.370
This range was measured at only one thermal power plant. As data on other units are not available, figure for S.D. has not been calculated. b Due to oil support in these thermal power plants, the CO emission is high, which has been reflected in the range of emission for CO.
ARTICLE IN PRESS N. Chakraborty et al. / Atmospheric Environment 42 (2008) 1073–1082
1077
categorized in accordance with the installed capacity of the generating units, and each generating unit has been given a name for identification. Other plant data relevant to the measured values, as available from the respective plant authority, are given in Table 4. The possible reasons behind the varying emission rate of different gases in the different plants are discussed in the following paragraphs. However, it will be prudent here to describe the plants in brief before results are presented. The generating units of 60 MW capacity, naming S1U1, S1U2, S1U3, and S1U4, were installed way back in the year 1982–1983; the oldest amongst the
diesel oil is used as and when required to supplement the combustion system for maintaining a stable combustion process during continuous running of the plant. Oil is also used during the starting process of the boiler. So far, no measurements could be taken during such start up process in any plant. All the generating units have individual electrostatic precipitators (ESP) and ID fans; but there was no scrubber in any unit for suppression of any gas. Information on corresponding coal and oil inflow rates were also noted from control room. The emission rates of different gases are given in Table 3. The various test results from different units have been
Table 3 Emission rates of green house gases from different thermal power generating units Generator unit
Installed in year
Installed capacity (MW)
Electricity generation and corresponding emission Generation (MW)
CO2 (kg h1)
CO (kg h1)
SO2 (kg h1)
NO (kg h1)
Flue gas temperature (1C)
60 60 60 60
60 60 60 60
48456.84 48508.67 46550.96 49420.8
428.53a 4.823 33.99 135.14a
312.51 372.64 327.45 347.07
130.210 174.41 92.36 133.99
127 128 139 154
1995 1997
250 250
250 250
242232.40 213002.82
60.85 19.79
2155.75 1861.95
453.10 592.58
137 127
S3U1 S3U3 S3U5 S3U6
1987–1988 1987–1988 1987–1988 1987–1988
210 210 210 210
175 140 175 200
260099.25 167914.63 181389.27 211836.74
4285.32a 375.97 34.63 3067.13a
2106.71 2238.86 2643.67 2139.92
523.52 369.96 434.72 397.53
133 126 127 143
S4U2
1990
72821.31
14.842
192.71
144
S1U1 S1U2 S1U3 S1U4
1982–1983 1982–1983 1982–1983 1982–1983
S2U1 S2U2
a
67.5
67.5
509.681
Due to oil support in these thermal power plants, the CO emission is high, which has been reflected in the range of emission for CO.
Table 4 Fuel feeding and ID fan delivery rates at the time of GHG measurement at different thermal power generating units Generator unit
Generation (MW)
Coal feed rate (ton h1)
Oil feed rate (l h1)
ID fan airflow rate (m3 s1)
Lambda (%)
Excess QA (%)
33 34 36 39
1.4 1.5 1.5 1.5
5.1 5.0 5.8 6.3
S1U1 S1U2 S1U3 S1U4
60 60 60 60
32.8 32.8 29.6 29.6
750 0 0 300
S2U1 S2U2
250 250
152.0 138.0
0 0
246 240
35 36
1.2 1.4
4.6 4.5
S3U1 S3U3 S3U5 S3U6
175 140 175 200
122.5 98.0 122.5 140.0
950 0 0 800
275 240 280 300
35 34 32 32
1.3 1.7 1.9 1.7
5.3 5.9 6.5 7.5
42.0
0
90
30
1.6
6.5
S4U2
67.5
55.0 58.5 58.9 59.6
Ambient temperature (1C)
ARTICLE IN PRESS 1078
N. Chakraborty et al. / Atmospheric Environment 42 (2008) 1073–1082
plants where measurement has been carried out so far. Being more than 20 year old units, their operational efficiency have definitely decreased in comparison to newer plants. Perhaps this is the main reason behind the occasional requirement of fuel oil along with coal in two of the generating units (see Table 4). It is understood from the plant authority that occasionally oil has to be used for maintaining stability of the combustion system in the boiler of different units; this is required when the volatile matter content of coal is o22%. The plant was designed for operation with coal containing 25% volatile matter. However, at the time of experimental measurement for a day it was observed that only units S1U1 and S1U4 were being fuelled with oil at the rate of 750 and 300 l h1, respectively. In these units, measurements were conducted at a horizontal location of rectangular ducts immediately after the location of ID fans (at a height of approximately 5 m from ground level) for individual units carrying flue gas to common stack. The test results are given in Tables 3 and 4, respectively. The plant-load factor for this plant was almost equals to unity at the time of measurements. Measurement was also conducted on a generator unit naming S4U2 of installed capacity 67.5 MW. In this unit, measurement was conducted at a horizontal location of the gas duct (one of two parallel ducts) before the stack. The height of the location was approximately 2 m from ground level. The plant was operating at its rated generation capacity at the time of the measurements. The plant was commissioned in the year 1990. The generating units of 210 MW capacity, naming S3U1, S3U3, S3U5, and S3U6, were designed for coal having 35% ash content. The designed coal supply requirement was 700 kg MWh1. But depending on the quality of available coal, the feeding rate varied from unit to unit, as and when required. The values of coal feeding rates in these four units (given in Table 4) were available verbally from the plant authority as no measured coal-feeding rate was available. Further, unit S3U5 and S3U6 were of improved version over the design of other two comparatively older units. In unit S3U1 and S3U3, measurement was conducted at a horizontal location of the rectangular flue gas duct (one of the four parallel ducts for each unit), in between ID fan and electrostatic precipitator. In unit S3U5 and S3U6, measurement was conducted at a horizontal location of the gas duct (one of two parallel ducts for each unit) just before the stack. In all the cases, the
height of measurement location was approximately 12 m from ground level. Except for one particular day, the plant-load factor varied in between 0.667 and 1.00. For almost all the other days of measurement, the plant-load factor was almost equals to unity. The 250 MW capacity generating unit were the highest in capacity where measurement were carried out. They were built and commissioned in the year 1995 and 1997; they are youngest of all thermal power plants where the measurements were conducted. So, it may be considered here that the efficiency of these two plants, naming S2U1 and S2U2, have not deteriorated much so far. In these two units, measurement on flue gas was carried out in the vertical stack at a height of 92.5 m from ground level. Naturally, the flue gas temperature must have reduced by few degrees centigrade after traveling a considerable distance. Here also, the plant-load factor was almost equals to unity on almost all the days of measurement. The emissions of the direct (CO2) and the indirect (CO, NO, and SO2) GHGs as measured for the different units of the respective thermal power plants designated as S1, S2, S3, and S4, respectively, have been reported in the following sections. 3.1. CO2 emissions from different plants From Table 3, it is seen that the average CO2 emission for S1U4 was maximum whereas that of unit S1U3 was minimum although the same amount of coal (29.6 ton h1) is being fed into both the units. The possible reason behind such a variation may have been the amount of excess air supplied and the combustion efficiency of the units. The carbon dioxide emissions from S1U1and S1U2 were more or less the same (see Table 3). It is also evident from Table 3 that the average CO2 emission rate from unit S3U1 was highest amongst the four units of that category (S3). The reason may be the higher carbon content of mixed coal being supplied to the unit. It was also noted that sometimes the color of flue gas emitting out of the unit was quite blackish. This may be due to additional oil support that was required to control the flame stability. The average CO2 emission from S3U2 was the second highest, whereas the average CO2 emission from the units S3U3 and S3U5 were 167914.63 and 181389.27 kg h1, respectively. The average CO2 emission from unit S2U1 was more than that of S2U2 because more coal
ARTICLE IN PRESS N. Chakraborty et al. / Atmospheric Environment 42 (2008) 1073–1082
(152 ton h1) was being supplied to S2U1 in comparison to that (138 ton h1) of unit S2U2 to meet same amount of electrical load. Measurement was also conducted on a generator unit (naming S4U2) of installed capacity 67.5 MW. The average emission rate of CO2 was 72821.31 kg h1. Here, no comparison could be made with other units of the same category, as measurement was not possible due to unavoidable circumstances in that plant. 3.2. CO emissions from different plants The average CO emission from unit S1U1 was highest as considerable amount of low speed diesel oil was injected along with coal due to lesser combustion capability of the unit (see Table 3). Complete combustion of coal and oil was not possible resulting in higher carbon monoxide emission from the unit. SIU4 recorded the second highest CO emission, as here also additional oil support was provided but to a lesser extent as that to S1U1. The average CO emission from S1U2 was lowest which can be attributed to better combustion efficiency of the system. The average CO emission from S1U3 was 33.99 kg h1. The average CO emission from unit S3U1 was very high. Due to higher coal flow rate and low air supply, combustion was not complete resulting in emission of higher amount of carbon monoxide. However, additional oil support in this unit may be the main reason for such a high CO emission. Additional oil support was also provided to the unit S3U6 but to a lesser extent as compared to S3U1. It recorded the second highest CO emission value. The average CO emission from unit S3U5 was the lowest and this may be attributed to better combustion in the system. The average CO emission from the unit S3U3 was 375.97 kg h1. The average CO emission from unit S2U1 was comparably higher than S2U2 and this may be attributed to better combustion efficiency of the unit S2U2 compared to S2U1. For the generator unit (naming S4U2) of installed capacity 67.5 MW, the average emission rate of CO was measured to be 14.842 kg h1.
1079
attributed to the varying sulfur content of coal fed into the respective units. The average SO2 emissions from all the four units in the S1 category were more or less constant with the unit S1U1 emitting the minimum. The ranges of the SO2 emission for this category of units lie in the range 312.51–372.64 kg h1 (see Table 3). The average SO2 emission for the four units in the S3 category was more or less equal with the maximum emission being recorded in the unit S3U5. The average SO2 emission rates of S3U1, S3U3, S3U5, and S3U6 were 2106.71, 2238.86, 2643.67, and 2139.92 kg h1, respectively. The average SO2 emission for the unit S2U1 was 2155.75 kg h1 whereas that from unit S2U2 was 1861.95 kg h1. Average SO2 emission measured for the category S4U2 was found to be 509.681 kg h1. 3.4. NO emissions from different plants The basic reason behind the variation of NO emissions from the different categories may be attributed to the varying amount of fixed nitrogen content in coal and the variable performance efficiency of the burners of the different categories. From Table 3, it can be seen that the average NO emission from unit S1U2 was the highest. The average NO emissions as measured for the other units S1U1, S1U3, and S1U4 are 130.210, 92.36, and 133.99 kg h1, respectively. The average NO emission from the unit S3U1 of the S3 category was the highest amongst themselves. The possible reason can be attributed to the lesser performance efficiency of the burner compared to the other units of the same category. The emissions from the other units of this category were more or less at par with each other (see Table 3). The average NO emission from unit S2U1 was less in comparison to that of unit S2U2. The reason may be attributed to better performance of burners in S2U1 compared to that of S2U2, thereby producing lesser amount of NO in S2U1 than in S2U2. Another possible reason may be the lesser amount of fixed nitrogen in coal fed into that unit. Measurement was also conducted on a generator unit (naming S4U2) of installed capacity 67.5 MW. The average emission rate of NO was recorded to be 192.71 kg h1.
3.3. SO2 emissions from different plants
4. Discussion
The main reason behind the variation in SO2 emission for the different categories can mostly be
Thermal power plants are one of the main sources of GHG emission throughout the world. For a very
ARTICLE IN PRESS 1080
N. Chakraborty et al. / Atmospheric Environment 42 (2008) 1073–1082
fast developing economy like India, measurement of GHGs’ from the thermal power plants are very much essential in order to find out their values so that necessary policies of reduction of such gases can be formulated. In order to develop strategies of GHG reduction, a realistic inventory of GHG is very much required. For that, determination of methodology-based measured emission coefficients are very much essential. These emission coefficients then will not only help developing the national inventory but will help comparing the national emission with the emission of the developed countries. The ranges of the values of the emissions in per kWh of electricity generated and in per kg of coal utilized obtained in the present study are provided in Table 5 in comparison to those obtained by OSC (Modeling Anthropogenic Emissions from Energy Activities in India: Generation and Source Characterization) and by Gurjar et al. (2004). The emission coefficients calculated in the present study are more or less at par with the values obtained by Gurjar et al. (2004), except for the two indirect GHGs CO and NO (see Table 5). The main variation is likely due to variable combustion technologies in the respective units. At the same time variation of coal quality with regard to carbon content and fixed nitrogen content may have an indirect effect on the variation of CO and NO formation. Particularly in case of CO, additional oil support in some of the units of the plants (see Table 6) may be the other possible reason for this mismatch in CO values. While the co-efficient values obtained for CO2 and SO2 matches closely with that obtained by OSC (Modeling Anthropogenic Emissions from Energy Activities in India: Generation and Source Characterization), a more or less four times higher coefficient values for NO has been reported by OSC. Obviously, there is no doubt that this needs to be
further investigated. However, roughly the differences can be attributed to the varying composition of coal fed into the machine and the combustion process in the unit and their performance. Furthermore, it is worth mentioning that the present study is totally on-line measurement based whereas the study by OSC is based on theoretical calculations. It has been considered that a total of 466,600 GWh electrical energy was produced from coal-fed thermal power plants for the year 2003–2004 (http:// indiabudget.nic.in) in India. This excludes contribution by the captive and other non-conventional power plants. To calculate the total emission of each of these toxic gases, the total range of emission (per kWh of electricity) from all categories of power plants has been selected and their mean emission coefficient has been calculated first. With that the total emission for the year 2003–2004 has been estimated. From Table 7, it can be observed that the corresponding CO2, CO, SO2, and NO values are Table 6 Increased Emission Coefficients due to additional oil support Plant Generation Gas type designation capacity per unit (MW)
Emission
S1
60
CO
(a) 13.07 g kg1 of coal (b) 7.140 g kWh1 of electricity
S3
210
CO
(a) 34.98 g kg1 of coal (b) 24.49 g kWh1 of electricity
Note: The emission rates of CO in the units SIU1 (428.53 kg h1), SIU4 (135.14 kg h1), S3U1 (4285.32 kg h1), and in S3U6 (3067.13 kg h1) were high compared to the other units of the plant.
Table 5 Comparison of emission coefficients Gaseous type
CO2 SO2 NO CO a
Range of measured emission coefficient
Emission coefficients
Present study for the year 2003–2004
Study by OSC in the year 1997–1998
Gurjar et al. (2004)
Present study
0.776–1.49 (kg kWh1) 5.210–15.99 (g kWh1) 1.540–3.263 (g kWh1) 0.055–24.49b
0.8–1.8 (kg kWh1) 4–18 (g kWh1) 6–13.1 (g kWh1) Not available
1.739 (kg kg1 of coal) 14.767 (g kg1 of coal) 0.824a (g kg1 of coal) 0.253 (g kg1 of coal)
1.639 (kg kg1 of coal) 14.031 (g kg1 of coal) 4.018 (g kg1 of coal) 5.153 (g kg1 of coal)
NOx as NO2 measured by Gurjar as 1.263, converted to NO multiplying with a factor of 0.652. Additional oil support.
b
ARTICLE IN PRESS N. Chakraborty et al. / Atmospheric Environment 42 (2008) 1073–1082
465.667, 1.583, 4.058, and 1.129 Tg, respectively, for the year 2003–2004. Considering CO2 to be the major contributor to the green house effect, the measured emission of 465.667 Mton for the year 2003–2004 has been compared with the calculated value obtained by OSC for the year 1997–1998 which was found to be 395 Mton. The thermal power generation for 1997–1998 was 336,000 GWh (Modeling Anthropogenic Emissions from Energy Activities in India: Generation and Source Characterization) while that for the period 2003–2004 was 466,600 GWh (http:// indiabudget.nic.in). This reflects that there is not much variation exists between the present study and that reported by OSC (Modeling Anthropogenic Emissions from Energy Activities in India: Generation and Source Characterization) at least for carbon dioxide taking into consideration the increased generation of electricity from 1997–1998 to 2003–2004. 5. Conclusion From the foregoing discussions, it can be generalized that the emission rates of the direct and indirect GHGs obtained from the Indian thermal power plants varies on various accounts; such as (a) quality of coal mixture; (b) quality of oil, wherever used; (c) quantity of coal and oil required for per unit generation; (d) age of the plant and its maintenance standard; and (e) amount of excess air fed into the furnace. The quality of coal (from different sources) is variable over a wide range. For a particular power plant, a single quality standard cannot be adhered
1081
to, as the mixing ratio of different quality of coal cannot be specified. At the same time, the mixture remains non-homogenous. Moreover, the quantity of coal and oil requirement varies in accordance with their calorific value to meet a specific electrical demand. Again efficiency of the plant and equipment, particularly the combustion and heat absorption is varying in nature. The efficiency is again dependent on the age of the plant and the standard of regular preventive maintenance. Without regular preventive maintenance of burners and air inlet valves, the combustion efficiency cannot be maintained at a desired standard level. Further, arrangements cannot be made for homogenous mixing of air and fuel at every location of the furnace; resulting in non-homogenous temperature zones in the furnace resulting in the variation in both (direct as well as indirect) GHG production. At the same time, the amount of excess air (supplied to the furnace) plays an important role in the combustion and generation of NO and SO2. The amount of electricity generation during different hours of a day is also a factor controlling the GHG production. Even if a plant runs at no load or little load, some fuel has to be supplied for maintaining stability of the furnace. Also all the plants do not run at full load all the time, nor the generation pattern over a day or month is the same for a particular plant. As the variability of all the above factors in all the generating plants in India remains random in nature, the emission coefficients cannot be quantified easily. So, establishing a relationship between above factors and the emission coefficient is not straightforward. Only, more accuracy in estimating total GHG emission can be achieved if measurement is conducted on almost all the power plants in India. To find out qualitatively better emission coefficients, on-line measurement should
Table 7 Total estimated emission of green house gas from Indian thermal power plants in the year 2003–2004 Emission per unit (kWh) of electricity
Range of emission from all power plants Average emission coefficient Total estimated emission during year 2003–2004a (Tg)
CO2
CO
SO2
NO
0.776–1.49 kg 0.998 kg 465.667
0.055–24.49 g 3.393 g 1.583
5.21–15.99 g 8.696 g 4.058
1.54–3.263 g 2.420 g 1.129
Note: 1 Tg ¼ 1 Mton ¼ 1 million metric ton. a Total generation from coal fired thermal power plants in India, has been considered to be 466,600 GWh for the year 2003–2004 (http:// indiabudget.nic.in).
ARTICLE IN PRESS 1082
N. Chakraborty et al. / Atmospheric Environment 42 (2008) 1073–1082
be conducted at all such thermal power plants in India. This will certainly help to determine the total emission as well as strategies of reduction of the GHGs’ from thermal plants in India. The present study sets the direction to such an endeavor. Acknowledgments We express our gratitude to the Ministry of Environment and Forest, Government of India, for selecting us to execute the project. We are grateful to United Nations Development Programme and Winrock International India for their financial support in executing this project. We also acknowledge the support extended by National Physical Laboratory, New Delhi, Central Fuel Research Institute, Dhanbad, with their valued advice in doing this project. We are indebted to Ministry of Power, Government of India, Ministry of Power and Ministry of Environment and Forest, Government of West Bengal, Central Pollution Control Board (Government of India), West Bengal Pollution Control Board (Government of West Bengal) and authorities of different power plants in West Bengal for their active support in executing this project. Lastly, we express our thanks to the authorities of Jadavpur University, without whose all-out support it would not have been possible to make this endeavor a success. References Chandra, A., Chandra, H., 2003. Environmental management for clean power generation of thermal power plants: an Indian perspective. Indian Journal of Air Pollution Control 3 (1), 22–36.
Environment Acts Rules and Notifications, Department of Environment, Government of West Bengal, March 1999. Garg, A., Bhattacharya, S., Shukla, R.P., Dadhwal, K.V., 2001. Regional and sectoral assessment of greenhouse gas emissions in India. Atmospheric Environment 35, 2679–2695. Gillani, N.V., Meagher, J.F., Valente, R.J., Imhoff, R.E., Tanner, R.L., Luria, M., 1998. Relative production of ozone and nitrates in urban and rural power plant plumes, 1. Composite results based on data from 10 field measurement days. Journal of Geophysical Research 103, 22593–22615. Good Practice Guidance and Uncertainty Management in National Greenhouse Gas Inventories. /http://www. ipcc-nggip.iges.or.jp/public/gp/gpgaum.htmS. Gurjar, R.B., Van Aardenne, A.J., Lelieveld, J., Mohan, M., 2004. Emission estimates and trends (1990–2000) for mega city Delhi and implication. Atmospheric Environment 38, 5663–5681. http://indiabudget.nic.in. Jorge, I., Manzini, F., Martinez, M., 2002. Renewable energies in electricity generation for reduction of greenhouse gases in Mexico 2025. AMBIO 31, 35–39. Mittal, M.L., Sharma, C., 2003b. Anthropogenic emissions from energy activities in India: generation and source characterization. Part I: emissions from thermal power generation in India. URL: /http://www.osc.edu/research/pcrm/emissions/ India.pdfS. Modeling anthropogenic emissions from energy activities in India: generation and source characterization. /www.osc. edu/research/pcrm/emissions/contents.shtmlS. Operation Manual of IMR 2800P Flue Gas Analyzer. /www. imrusa.comS. Ryerson, T.B., Buhr, M.P., Frost, G.J., Goldan, P.D., Holloway, J.S., Hubler, G., Jobson, B.T., Kuster, W.C., McKeen, S.A., Parrish, D.D., Roberts, J.M., Sueper, D.T., Trainer, M., Williams, J., Fehsenfeld, C.F., 1998. Emissions lifetimes and ozone formation in power plant plumes. Journal of Geophysical Research 103, 22569–22583. TERI (The Energy and Resources Institute), 2001a. State of environment report for Delhi 2001. Report No. 2000EE65, Supported by the Department of Environment, Government of National Capital Territory, Delhi. Varshney, K.C., Aggarwal, M., 1992. Ozone pollution in the urban environment of Delhi. Atmospheric Environment 3, 291–294.