- Email: [email protected]
Synergistic effect of co-firing woody biomass with coal on NOx reduction and burnout during air-staged combustion
Experimental Thermal and Fluid Science 71 (2016) 114–125
Contents lists available at ScienceDirect
Experimental Thermal and Fluid Science journal homepage: www.elsevier.com/locate/etfs
Synergistic effect of co-firing woody biomass with coal on NOx reduction and burnout during air-staged combustion Yonmo Sung a,b, Sangmin Lee b, Changhyun Kim c, Dongheon Jun d, Cheoreon Moon b, Gyungmin Choi b,⇑, Duckjool Kim b a
Department of Mechanical Engineering, Imperial College London, London SW7 2AZ, United Kingdom School of Mechanical Engineering, Pusan National University, Busan 609-735, South Korea Plant Management Team, Korea South-East Power Co., Ltd., Gyeongsangnam-do 660-031, South Korea d Taean Thermal Power Complex Division, Korea Western Power Co., Ltd., Chungcheongnam-do 357-914, South Korea b c
a r t i c l e
i n f o
Article history: Received 18 May 2015 Received in revised form 18 October 2015 Accepted 19 October 2015 Available online 23 October 2015 Keywords: Synergistic effect Woody biomass Co-firing NOx reduction Air-staging
a b s t r a c t Hybrid technologies combining fuel blending and air-staging processes have been applied to a pulverized coal fired furnace to reduce NOx emissions. In this study, an Australian bituminous coal (Whitehaven), an Indonesian sub-bituminous coal (Adaro), and an Indonesian woody biomass were selected as fuel with blending ratio of 10%, 20%, and 30% of the low-rank fuels, and the air-staging levels were set to 235 mm, 390 mm, 585 mm, and 760 mm. The purpose of this study was to investigate the synergistic effect of woody biomass co-firing on the level of NOx emissions and the degree of carbon burnout under air-staged conditions. For single coals, sub-bituminous coal was more favorable than that for bituminous coal to reduce NOx emissions due to low fuel-N composition. This tendency was more dominant with increasing the air staging levels. In the co-firing of woody biomass with coal, as the biomass is highly volatile but has a low carbon content, it could be successfully applied to low-NOx combustion under air-staged conditions. In addition, the degree of carbon burnout and the flame temperature both increased. As a result of this research, we determined that hybrid NOx reduction technologies have the potential to reduce exhaust gas emissions and enhance combustion performance. A dominant synergistic effect on NOx reduction and carbon burnout was observed when woody biomass co-firing with coal was applied to air-staged combustion. Ó 2015 Elsevier Inc. All rights reserved.
1. Introduction Coal-fired thermal power plants in Korea currently co-fire bituminous/sub-bituminous coal because of the difficulty in obtaining high-grade coal. The importance of fuel blending technology is being emphasized as it opens up the possibility of using low-rank fuels such as biomass or waste material. Recently, the utilization of both woody biomass and sewage sludge has become commonplace in existing coal-fired thermal power plants because Korea’s renewable portfolio standard (RPS) obligates electricity providers to produce a specified fraction of their electricity from renewable energy, with the remainder being generated from fossil fuel and nuclear reactors. In Korea, 13 electricity providers have been mandated to generate 2% of their gross power output from renewable energy sources, such as solar, wind, hydro, tidal power, fuel cells, hydrogen, biomass, and waste, as a result of the ⇑ Corresponding author. Tel.: +82 51 510 2476; fax: +82 51 512 5236. E-mail address: [email protected] (G. Choi). http://dx.doi.org/10.1016/j.expthermflusci.2015.10.018 0894-1777/Ó 2015 Elsevier Inc. All rights reserved.
implementation of the RPS regulation in January 2012. This percentage is set to rise to 10% in 2022 [1]. Most coal-fired thermal power plants in Korea determined that the best way for them to satisfy the RPS regulations was to introduce the co-firing of biomass since it is the fourth largest source of energy after coal, petroleum, and natural gas [2]. Environmental issues, especially NOx emissions, continue to be one of the most important challenges facing coal-fired utilities. The regulations limiting NOx emissions are actually very severe [3]. For example, as of 2008, the NOx emission limit imposed on power plants over 500 MWth in the European Union (EU) has been 500 mg/Nm3 at 6% O2. This limit will drop to 200 mg/Nm3 at 6% O2 from 1 January 2016. In this context, emission limits for primary air pollutants such as NOx, SOx, and dust from coal-fired thermal power plants in Korea have been continually reinforced, shown in Table 1 [4,5]. The air quality preservation act by Korean ministry of environment requires that new facilities must reduce NOx emissions as 103 mg/Nm3 at 6% O2 from 1 January 2015.
115
Y. Sung et al. / Experimental Thermal and Fluid Science 71 (2016) 114–125 Table 1 Permissible emission standards for NOx, SOx, and dust from coal-fired thermal power plant in South Korea. Pollutants
Facility division
Limit before 2014.12.31
Limit after 2015.1.1
NOx (mg/Nm @6%O2)
Installed before 1996.6.30 Installed after 1996.7.1 Installed after 2015.1.1
<308 <164
<287 <144 <103
SOx (mg/Nm3@6%O2)
Capacity > 100 MW Installed before 1996.6.30 Installed before 2014.12.31 Installed after 2015.1.1 Capacity < 100 MW Installed before 1996.6.30 Installed before 2014.12.31 Installed after 2015.1.1
<286
Capacity Installed Installed Installed Capacity Installed Installed Installed
>500 MW <28 <19
3
Dust (mg/Nm3@6%O2)
before 2001.6.30 after 2001.7.1 after 2015.1.1 before 2001.6.30 after 2001.7.1 after 2015.1.1
To satisfy this impending tightening of the NOx emission limits, technologies such as in-furnace removal (i.e. air-staging, reburning, the use of low-NOx burners, etc.), flue gas purification (i.e. selective catalytic reduction (SCR), selective non-catalytic reduction (SNCR), etc.), and other combustion methods (i.e. coal-water slurry combustion, circulated fluidized bed combustion etc.) have been widely used for coal-based power generation applications [6]. Among the existing methods, air-staged combustion is one of the most efficient and attractive technologies for reducing NOx emissions, because it does not require expensive new equipment, such as SCR systems [7]. Unfortunately, there are also serious disadvantages such as high operating costs, ammonia slip (the passing of ammonia through the SCR reactor), and the easy deactivation of the catalyst in SCR applications [8]. Ammonia slip, in particular, can lead to a blockage in the SCR catalyst due to an unwanted reaction between un-reacted ammonia and unburned carbon. Generally, biomass co-firing should reduce NOx emissions simply because most forms of biomass contain less fuel-N than the coal they replace [9]. Therefore, it is important to minimize the initial formation of NOx by in-furnace removal methods including airstaging and biomass co-firing to avoid having to implement flue gas purification or minimize its operation. Numerous articles dealing with NOx reduction technologies such as air-staged combustion and low-rank fuel blending have been published [10–18]. Ribeirete and Costa [19] investigated the impact of air-staging on the overall performance of a large-scale laboratory furnace fired by an industrial pulverized coal swirl burner. The influences of the axial position of the staged-air injector, the primary zone stoichiometric ratio, the coal type, and the configuration of the staged-air injectors on the pollutant emissions and particle burnout were quantified [18]. Munir et al. [20] showed that the addition of biomass has a positive effect on NOx reduction and carbon burnout under the optimum conditions that were identified as part of their study. Daood et al. [21] investigated NOx control in coal combustion by combining biomass co-firing, oxygen enrichment, and SNCR in a 20-kW combustion facility. It was found that NOx control by SNCR and the oxygen-enriched co-firing of biomass in power station boilers would result in lower NOx emissions and a higher CO2 concentration that would lead to more efficient scrubbing with better carbon burnouts. When a furnace is co-firing biomass and coal, it is important to consider the slagging, fouling, and tube corrosion that may occur as a result of the presence of unburned carbon. This study focused on the effect of air-staged combustion and woody biomass co-firing
<286 <228 <143 <428 <371 <228 <143
<500 MW <37 <28
>100 MW <23 <19 <9 <100 MW <37 <28 <28
on the overall performance, including NOx emissions and carbon burnout, since there are still doubts about the effect on gas emissions of low-rank fuel blending. For having nearly same coal properties of two sub-bituminous coals, NOx emissions from two subbituminous coals (Wara and Adaro), which were blended with bituminous, exhibited an inverse tendency with increasing subbituminous coal blending ratio although the moisture contents of them were different [22]. The effect of biomass co-firing on NOx emissions was found to be dependent upon the particle size, with smaller particles reducing NOx emissions as the biomass co-firing ratio increased [23]. Air-staged combustion is used to reduce the stoichiometric ratio in the primary combustion zone, and thus the fuel-rich condition causes the NOx to act as a reducing species. This reducing environment can be locally intensified by adding low-rank fuels such as woody biomass. In particular, the nitrogen content in woody biomass is relatively lower compared to straw and other agricultural residues, although the woods and the straws decompose in different ways in terms of the release of the N-components [24,25]. Moreover, air-staged combustion is, in general, desirable for reducing NOx emissions; if, however, there is insufficient air in the primary combustion zone, the flame temperature is typically applied to low-temperature processes [23]. Different components in the biomass, i.e., lignin, and semi-cellulous/ cellulous materials, decompose at different temperatures, and thus the biomass starts to release its volatile components at a relatively low temperature [26]. This behavior could help alleviate the drop in temperature caused by deep air-staging through possible synergistic effects when coal is co-fired with woody biomass. There are lots of studies on low-rank fuel blending with air-staged combustion. However, woody biomass co-firing with air staging is rarer to produce synergy effects on reducing NOx emissions and enhancing carbon burnout performances simultaneously, and even fewer studies have been attempted under the combination of several parameters as oxy-firing, air staging, and agricultural biomass co-firing for the synergy of NOx and burnout [20,21]. In this study, a hybrid technology combining fuel blending and air-staging was applied to reduce the NOx emissions from a pulverized coal fired furnace. The combustion and emission characteristics of different blends of coal (bituminous and subbituminous) and woody biomass were investigated using a pulverized coal fired furnace. To understand the details of the air-staged combustion, in this study we undertook detail measurements of gas species concentration, temperature, and carbon burnout under
116
Y. Sung et al. / Experimental Thermal and Fluid Science 71 (2016) 114–125
a range of operating conditions. The purpose of this study was to investigate the synergistic effect of woody biomass co-firing on NOx reduction and carbon burnout under air-staged conditions. 2. Experimental 2.1. Materials In this study, we used three solid fuel samples, namely, an Australian bituminous coal (WH, Whitehaven) as a high-rank coal, an Indonesian sub-bituminous coal (AD, Adaro) as a low-rank coal, an Indonesian woody biomass (BM, Biomass), and blends thereof. Each sample was milled and sieved to the desired particle size as listed in Table 2. The particle size of the samples was estimated by using a laser diffraction particle size analyzer (LS 13 320, Beckman Coulter). The mean particle size of the WH, AD, and BM was 118 lm, 122 lm, and 286 lm, respectively. For the blends, the blending ratio, as based on the energy fraction (heating value), was 10%, 20%, and 30% of the low-rank fuels. These samples were fed directly into a pulverized coal fired furnace without any moisture-related treatment, and the results of the proximate and ultimate analyses, as well as the calorific values, of each fuel sample are listed in Table 3. 2.2. Experimental apparatus and methods The experimental detail of the test rig has already been discussed in previous published articles [27,28]. Fig. 1 is a schematic diagram of the pulverized coal fired furnace. The cylindrical comTable 2 Particle size distributions of fuel samples obtained by the laser diffraction particle size analyzer. Samples
Bulk density (kg/m3)
d10, <10%
WHa ADb BMc
593 624 418
11 19 33
d25, <25%
d50, <50%
d75, <75%
d90, <90%
dmean, mean
106 112 232
180 181 397
244 241 612
118 122 286
Size (lm)
a b c
38 51 113
WH: Australian bituminous coal; Whitehaven. AD: Indonesian sub-bituminous coal; Adaro. BM: Indonesian woody biomass.
Table 3 Proximate and ultimate analysis of solid fuel samples and LPG compositions. Quantity
WH
AD
BM
Proximate analysis (wt%, air-dry basis) Moisturea Moisture Volatiles Fixed carbon Ash
13.04 5.02 30.52 51.08 13.38
26.83 19.38 38.69 39.02 2.91
6.88 6.77 74.85 16.01 2.37
Ultimate analysis (wt%, dry basis) Carbon Hydrogen Oxygen Nitrogen Sulfur Ash Net calorific value (MJ/kg, as-received basis)
72.1 4.62 7.38 1.55 0.26 14.09 25.16
70.4 4.57 20.04 1.29 0.09 3.61 21.19
41.2 6.02 49.7 0.45 0.1 2.53 18.26
Composition (vol%) C2H6 C3H8 C4H10 Net calorific value (MJ/kg) a
as-received basis.
LPG
1 97 2 46.04
bustion chamber is down-fired to facilitate particulate removal. The combustion chamber is made up of three individual watercooled steel segments, each having an internal diameter of 200 mm while being 500 mm high. The roof section and all of the segments are lined with a layer of refractory material (width: 130 mm). Each segment has four pairs of diametrically opposed 120-mm round ports to allow the observation of the combustion. R-type thermocouples and gas sampling ports for recording the in-furnace temperatures and gas concentration are installed in every segment. Six pairs of staging air ports, oriented tangentially to the main swirl flow of the burner, are mounted in the refractory materials. In addition, a thermocouple lance is located in the cooling water circuit of every segment to record the outlet cooling water temperatures. The axial positions of the air-staging level, relative to the burner port, were at 235 (SL1), 390 (SL2), 585 (SL3), and 760 mm (SL4), while the thermocouple lances were placed at 65 (Z1), 210 (Z2), 355 (Z3), 535 (Z4), and 680 mm (Z5). The burner has a coaxial dual-piping structure with a main burner port (inner diameter: 10 mm) and an annular slit (width: 5.5 mm) installed outside the main burner port. Primary air was used to transport the fuel samples from the fuel feeder into the furnace. Primary air and solid fuel are supplied from the main burner port. The secondary air enters the combustion chamber through an annular slit port and encounters a swirl generator with a vane angle of 60°. There is also a refractory burner quarl with a half-angle of 30° with a pair of annulars (inner diameter: 6 mm) to supply liquefied petroleum gas (LPG) fuel which has positive effects on coal particle ignition and flow formation such as an internal recirculation zone (IRZ) and an outer recirculation zone (ORZ) as shown in Fig. 1. The LPG flow rate is held to the minimum rate needed to stabilize the swirl flames [29]. The air and LPG flow rate were controlled by a mass flow controller (KOFLOC-3660, Kojima Instruments). The concentrations of O2, CO2, and NOx were measured in a dry basis using a gas analyzer (VA-3000, HORIBA), which is a multi-gas analyzer that uses non-dispersive infrared absorptiometry, the chemiluminescence method, and magnetic pressure analysis to capture data. Gas measurements were taken using a stainless steel water-cooled dual-pipe probe [27,30]. The diameter of the sampling hole for the probe was 1 mm. A wet sample was drawn through the probe and into the sampling system by an oil-free pump. Most of the moisture was removed by a silica-gel filter. Then, any residual moisture and particles were removed by a pair of filters such that a constant supply of dry, filtered combustion gas was available to the analytical instrument. The analog outputs of the analyzer were transmitted via high-speed A/D boards (NI-DAQ 9172, National Instruments), at a sampling rate of 2 Hz, to a computer where the signals were processed and then the mean values computed. The concentration of gas species was measured using the gas analyzers for about 10 min, five times, and then the ensemble average value was used. The variations in the gas concentrations are expressed by error bars in the graphs presented in the Results and Discussion. Example statistical analysis data is presented in Table 4. Temperature measurements were obtained using 50-lm diameter fine-wire R-type (Pt-Rh 13%/Pt) thermocouples. Each thermocouple’s bead was surrounded by a 10-mm diameter ceramic tube to prevent the thermocouples from becoming attached to unburned coal particles. The ceramic tube thermocouples were calibrated in turbulent jet pulverized coal flames [31,32], relative to unprotected thermocouples. In order to better understand the temperature change with different combustion environments, in-furnace temperatures were measured at the furnace center in radial directions in interval of 10 mm for every axial position Z, shown in Fig. 1, and then 2-d half contours were obtained from
117
Y. Sung et al. / Experimental Thermal and Fluid Science 71 (2016) 114–125
Fig. 1. Schematic diagram of pulverized coal combustion system.
placed in a glass container, in a filter. The collected wet ash samples were then placed in an oven at 383 K for three hours to dehydrate, and then the de-watered particles were burnt in a thermogravimetric analyzer (TGA) (SDT-Q600, TA Instruments) and the weight loss was calculated by applying the ash tracer method [33]. The ash tracer method is prone to analytical errors when used with fuels with low ash contents, which can be of importance for many biomass fuels, in particular woody materials. Scatter in the results caused by analytical errors is less important for fuels with high ash contents [34,35]. The carbon burnout data was obtained using the following equation [36,37]:
Table 4 Example statistics of data analysis of exhaust gas emissions.
WH@SL0 WH@SL1 WH@SL2 WH@SL3 WH@SL4 AD@SL0 AD@SL1 AD@SL2 AD@SL3 AD@SL4
O2 (%) mean, SD
CO2 (%) mean, SD
NO (ppm) mean, SD
NO (mg/Nm3) mean
3.83, 0.16 4.53, 0.14 4.54, 0.11 4.63, 0.08 4.77, 0.15 2.48, 0.11 2.7, 0.21 2.86, 0.2 2.91, 0.11 2.97, 0.07
15.58, 14.94, 14.67, 14.41, 14.23, 17.59, 17.42, 17.23, 17.05, 16.93,
330.87, 4.55 300.44, 4.05 268.81, 5.22 257.01, 5.71 258.69, 3.71 322.38, 3.54 282.05, 6.2 245.2, 8.22 196.52, 6.94 177.63, 4.43
413 375 335 321 323 402 352 306 245 222
0.2 0.2 0.12 0.11 0.19 0.08 0.28 0.24 0.14 0.14
W¼
Notes: SD is the standard deviation, NOx data is corrected at O2 6%, SL0 is the unstaged condition, SL1 to SL4 are the air staging level 1 to 4, respectively.
55 point data. The empty values among axial positions (Z1–Z5) were filled interpolation values, using Tecplot 360 (Tecplot, USA). Fly ash sampling was performed with the aid of a stainless steel, water-cooled, water-quenched probe normally used for gas species sampling. Ash samples were collected with the probe and then
1 ðwk =wx Þ 1 wk
ð1Þ
where W is the carbon burnout, w is the dry mass fraction, and the subscripts k and x refer to the ash content in the input solid-fuel and char sample, respectively. 2.3. Experimental conditions Table 5 summarizes the furnace operating conditions for the air-staging and fuel-blending parametric trials. For all of the exper-
Table 5 Operating conditions of pulverized coal fired furnace. WH Total thermal input (kW) Overall excess air (%) Primary zone stoichiometric ratio, SR1 Solid fuel feeding rate (kg/h) LPG flow rate (slpm) Primary air flow rate (slpm) Secondary air flow rate (slpm) Secondary air swirl number Staging air flow rate (slpm)
AD
16 = 8 (solid fuel) + 8 (LPG) 15 0.9, 1.15 0.9, 1.15 1.14 1.35 5 10 253, 288 265, 303 1.33 35, 0 38, 0
WB10
WB20
WB30
WA10
WA20
WA30
AB30
0.9 1.19
0.9 1.23
0.9 1.27
0.9 1.17
0.9 1.19
0.9 1.21
0.9 1.42
250
246
241
255
256
257
250
34
33
32
35
36
36
34
Notes: WA is the blend of WH coal + AD coal, WB is the blend of WH coal + Biomass, AB30 is the blend of AD coal 70% + Biomass 30%.
118
Y. Sung et al. / Experimental Thermal and Fluid Science 71 (2016) 114–125
air staging and co-firing, the test of WH single coal firing without any air-staging or co-firing, while keeping the SR0 of 1.15, was taken as baseline, and the reduction (%) in the amount of gas emissions and carbon burnout was defined by the following equation:
Reductionð%Þ ¼
baseline value obtained value baseline value
ð2Þ
3. Results and discussion 3.1. Combustion and emission characteristics for single type of coal with air staging level Fig. 2. NOx emission and burnout performance of two single types of coal for different air staging levels.
imental conditions, the thermal input was constant at 16 kWth, based on the net calorific value. The preheating temperatures for the primary, secondary, and staging air were held constant at 338 K, 573 K, and 303 K, respectively, but this preheat thermal energy was not considered in the thermal input calculations. The overall stoichiometric ratio (SR0) was 1.15. The secondary air swirl number was 1.33 [38]. To enable the comparison of the effects of
Several studies showed that the NOx reduction with air-staged combustion primarily depends upon the location of the air-staging level (SL) and the primary zone stoichiometric ratio (SR1) [6,19,20]. In the case of the above studies, as the value of the SR1 fell, as did the NOx emissions. From the previous published article [27], the SR1 value of 0.9 was found to be the optimum for both NOx reduction and carbon burnout (W). Therefore, the fixed SR1 value of 0.9 was used, and the value of SR0 was maintained at 1.15 along with the location of each SL from level 1 (SL1) to level 4 (SL4).
Fig. 3. In-furnace temperature contours for two single types of coal at SR1 of 0.9 with a different SL. (a) WH coal, and (b) AD coal.
Y. Sung et al. / Experimental Thermal and Fluid Science 71 (2016) 114–125
Fig. 4. O2 and CO2 gas emissions of two single types coal for different air staging levels.
Fig. 2 shows the effect of the SL value on NOx emission and burnout characteristics of the WH and AD coals. It can be seen that the NOx emissions decrease as the distance between the SL and the burner exit increases. In particular, the NOx emissions can be seen to be 330, 300, 269, 257, and 259 ppm, from no staging (SL0) to staging level 4 (SL4) with WH coal, and 322, 282, 245, 197, and 178 ppm from SL0 to SL4 with AD coal. When the SL is placed in axial locations relatively close to the burner exit, as in the case of SL1 or SL2, the coal particle residence time in the primary combustion zone is relatively short because of the short length of the fuel rich zone (reducing environment). Conversely, the increased residence time of the coal particles in the fuel rich zone, as in the case with SL3 or SL4, with a relatively lean O2 concentration, promotes a reduction in the amount of NOx emissions. Previous works by Ribeirete and Coata [19] and Munir et al. [20] produced similar results. From Fig. 2, it can be seen that the effect of SL on the
119
NOx emissions differs depending upon the coal type. The differences can be attributed to the rapid production of gas-phase reductive species such as hydrocarbons for sub-bituminous coal, whereas the massive gas-phase reductive species for bituminous coal are mainly derived from the gasification reactions of char, which generally takes relatively longer to complete [8]. Although air-staged combustion in pulverized coal fired boilers can reduce NOx emissions, it could have a negative effect on the carbon burnout (W) performance of the coal particles. Therefore, it is very important to achieve both low NOx emissions and a high W. The W values for the WH and AD coals were examined and are shown in Fig. 2. The value of W decreases monotonously as SL increases, although the value of W for two different types of coal is quite different. For instance, when SL increases from SL1 to SL4, the W for the WH coal drops from 92.85% to 89.11%, while for the AD coal, the W drops from 99.85% to 99%. Although the NOx reduction based on SL0 for the WH coal was 22.32% at SL3 and 21.82% at SL4, and for the AD coal was 40.61% at SL3 and 46.31% at SL4, the W reduction for the WH coal was 2.95% at SL3 and 4.02% at SL4, and for the AD coal was 0.23% at SL3 and 0.86% at SL4. This means that the NOx reduction is insignificant, however, a significant decrease in the W value when the SL increases from SL3 to SL4. Therefore, for the co-firing tests, the condition of SL3 and SR1 values of 0.9 was fixed. Fig. 3 shows the in-furnace temperature contours for two single types of coal at SR1 of 0.9 with a different SL. The maximum temperature and temperature gradient can be seen in the vicinity of the burner, while the temperature decreases along with the furnace height in the burner zone. In Fig. 3a, for the WH coal, the temperature decreased as SL increased at a point 200 mm from the burner port. In Fig. 3b for the AD coal, however, the tendency was found to be quite different. The temperature showed a similar tendency as SL increased. This was mainly because of the relatively high amount of volatile matter but less fixed carbon in the AD coal,
Fig. 5. Co-firing characteristics for bituminous/sub-bituminous coals and bituminous coal/woody biomass at SL3 and SR1 of 0.9. (a) NOx, (b) carbon burnout, (c) O2, and (d) CO2.
120
Y. Sung et al. / Experimental Thermal and Fluid Science 71 (2016) 114–125
Fig. 6. In-furnace temperatures as a function of low-rank fuel blending ratio at SL3 and SR1 of 0.9. (a) 2-d temperature contour for WA blending combination, and (b) 2-d temperature contour for WB blending combination.
which makes the combustion reactivity of the AD coal better than that of the WH coal. For the SL0 and SL1 conditions, the temperature at around a height of 200 mm for the AD coal was lower than that for the WH coal. However, it was higher than that for the WH coal at SL2 to SL4. In particular, this reversed trend in the temperature distributions can be seen more clearly in the downstream region (at a height in excess of 300 mm). These results can explain why the W for the AD coal was higher than that for the WH coal, as shown in Fig. 2. The distributions of the O2 and CO2 gas emissions at the furnace exit for the WH and AD coals are shown in Fig. 4. The O2 and CO2 are complementary species, being the main reactant and product of the combustion, respectively. Although the O2 concentration is proportional to SL, the CO2 distributions showed a completely
opposite tendency. Also, the O2 concentration for the AD coal was lower than that for the WH coal, and the CO2 concentration for the AD coal was higher than that for the WH coal. In this study, the excess 3% O2 was considered with respect to the 15% excess air at SR0 of 1.15. In Fig. 4, for the unstaged conditions, the O2 concentrations for the WH and AD coals were around 3.8% and 2.5%, respectively. This can be explained by the temperature distribution and the carbon burnout characteristics. Although it depends upon the value of SL, the amount of unburned carbon in the fly ash of the WH coal was around 7%, as shown in Fig. 2. Given the reaction of C + O2 ? CO2, we can deduce that the O2 concentration will be more than 3%, with large amounts of unburned carbon remaining in the fly ash. In contrast, the O2 concentration for the AD coal was less than 3%, and the CO2 was higher than that for the WH coal.
Y. Sung et al. / Experimental Thermal and Fluid Science 71 (2016) 114–125
Fig. 7. Radial temperature profiles as a function of low-rank fuel blending ratio at SL3 and SR1 of 0.9. (a) WA at Z1, (b) WA at Z2, (c) WB at Z1, and (d) WB at Z2.
Fig. 8. NOx reduction mechanism of hybrid technology between air staging and low-rank fuel blending.
121
122
Y. Sung et al. / Experimental Thermal and Fluid Science 71 (2016) 114–125
Fig. 9. NOx reduction rate (base case: WH single coal firing without air staging and fuel blending) at SL3 and SR1 of 0.9 with co-firing.
Fig. 10. Burnout reduction rate (base case: WH single coal firing without air staging and fuel blending) at SL3 and SR1 of 0.9 with co-firing.
Moreover, there was basically no unburned carbon, and the overall temperature in the downstream region in the furnace was higher than that of the WH coal. This would certainly have a major impact on the synergistic effect on the combustion and emission performance under co-firing conditions with high- and low-rank fuels.
primary combustion zone than in the case of AD co-firing because the woody biomass has a higher amount of volatile matter that also decomposes faster than coal. Fig. 5c shows the influence of the co-firing ratio on the value of W. The value of W for the WA and WB blends is higher than that in non-blended WH coal combustion for given operating conditions. There is a tradeoff relationship between the NOx emissions and the carbon burnout. With both blending combinations, the value of W was improved as the low-rank fuel blending ratio increased, and the positive effect on the value of W for the biomass blending was relatively large, compared to the AD coal blending. This result is in agreement with those of previous studies [11,20,39]. This behavior could be linked to the higher volatility and reactivity of the woody biomass, making the unburned carbon in the bituminous single coal particles easier to burn when co-firing with woody biomass. For the blends, the maximum value of W for WA30 and WB30 was 93.35% and 94.32% for a given operating condition. Fig. 6 shows the temperature contours with a low-rank fuel blending ratio at an SL3 and SR1 of 0.9. Under every test condition, the temperature distributions exhibited a similar tendency. The temperature decreases with height in the furnace. The overall distributions with the addition of low-rank fuels in ratios from 10% to 30% were higher than in the case of firing with WH coal alone. This feature was very apparent in the case of woody biomass blending, and also in the results of a study on the co-combustion of low-rank coal with biomass [12]. From the radial temperatures in Fig. 7, the temperature for WA30 and WB30 was higher than that for WH single coal combustion when the thermal input was constant, based on the energy fraction. This is because the large volatile gas reaction arising from the low-rank single fuel in the blends assists with the carbon conversion reaction of the high-rank WH coal. Therefore, the W and CO2 values became higher as the biomass blending ratio increased, and thus the O2 value fell due to the enhanced gas phase volatile reaction from the biomass, as shown in Fig. 5.
3.2. Combustion and emission characteristics of blends of coal and biomass Fig. 5 shows the influence of low-rank fuel blending on the value of W and the emissions of gases such as NOx, O2, and CO2 for a given operating condition (SL3 and SR1 of 0.9) for the WA (WH + AD) and WB (WH + woody biomass) blends. NOx emissions that were very different from those of a composition of low-rank fuels were observed, as shown in Fig. 5a. The addition of subbituminous AD coal based on bituminous WH coal had a positive effect on the NOx reduction. This positive effect on the NOx reduction was more significant with the blends of WH coal and woody biomass. The minimum NOx emission for the co-firing at SL3 and SR1 of 0.9 was 222 ppm and 190 ppm for WA30 and WB30, respectively. This is mainly due to the lower nitrogen content of the biomass (fuel-N of 0.45%), compared to the sub-bituminous coal (fuelN of 1.29%) and the bituminous coal (fuel-N of 1.55%). Moreover, the dominant volatile-N compound in bituminous coal combustion is HCN, while in biomass combustion it is both HCN and NH3 [24]. The woody biomass, especially, gives largely HCN with some NH3, and the agricultural biomass gives mainly NH3 [25]. Essentially, there is more volatile matter in the woody biomass (VM of 74.85%) than in the sub-bituminous coal (VM of 38.69%), as shown in Table 3. For blends with a higher biomass proportion, the predominant combustion consisted of gas phase reactions in a strong local reducing environment under fuel-rich, O2-lean conditions. With regard to the residence time of solid fuels in the fuel-rich zone, as the biomass (FC of 16.01% and fuel-O of 49.7%) contains less fixed carbon and more oxygen relative to sub-bituminous coal (FC of 39.02% and fuel-O of 20.04%), the amount of stoichiometric air required based on the energy fraction is less than that required for coal combustion, and then the particle residence time for the biomass blending condition should increase because of the lower bulk velocity at the burner exit, as shown in Table 5. From Fig. 5b and d, we can see that the O2 concentration declines as the low-rank fuel blending ratio increases, while the CO2 distribution exhibits a completely opposite tendency. Also, the O2 concentration for the biomass blends was lower than that in the AD coal blends. Conversely, the CO2 concentration for the biomass blends was higher than that in the AD coal blends. With the co-firing of woody biomass, O2 is consumed more rapidly in the
3.3. Synergistic effect of hybrid NOx reduction technology on combustion and emission characteristics Fig. 8 is a schematic diagram of the NOx reduction method based on the fuel-rich, oxygen-lean concept. The most important point affecting NOx reduction technologies is to create a fuel-rich environment in the main gas/volatile combustion zone [3,8,20]. The basic principal of air-staged combustion is to reduce the stoichiometry in the primary combustion zone in which devolatilization takes place in an O2-lean atmosphere. Additionally, this reducing environment could be made locally strong by adding
Y. Sung et al. / Experimental Thermal and Fluid Science 71 (2016) 114–125
123
Fig. 11. In-furnace temperature contours for a blending proportion of 30% with three blending combinations such as WA (WH + WA), WB (WH + Biomass), AB (AD + Biomass). (a) SL0, and (b) SL3.
low-rank fuels that contain higher proportions of volatile matter and oxygen. The NOx concentration was found to be lower as SL increased because the length of the fuel-rich primary zone became longer as SL increased. Thus, the particle residence time in an O2lean atmosphere became longer. Fig. 9 shows NOx reduction rate at SL3 with co-firing, comparing the unstaged WH single coal firing condition. It can be seen that the NOx reduction with air staging for the WH coal was 22.32%, and for both air staging and co-firing, it showed a higher reduction rate. The NOx reduction for the blends of 10%, 20%, and 30% of WA and WB was 28.59%, 31.83%, and 32.87% for WA, and 32.26%, 38.83%, and 42.63% for WB. This reduction became particularly predominant as the air staging level and low-rank fuel blending ratio increased. As a result, the synergistic effect of the woody biomass blending is more efficient than that of sub-bituminous coal blending for controlling the NOx emissions of air-staged combustion. Fig. 10 compares the W reduction for two blending combinations such as WA (WH + WA) and WB (WH + Biomass) at SL3. It can clearly be seen that the W reduction for the WH coal with air-staging was 2.95%, meaning that the burnout performance fell as the intensity of the air-staging increased, as shown in Fig. 2. The reaction intensity for the staged condition in the main combustion zone is lower than that of the unstaged condition
due to the deficiency in the required theoretical air, and thus the overall combustion of the coal particles remains incomplete in the primary combustion zone. This could explain why the air staging produced greater NOx reduction and less carbon burnout. The W reduction for the 10%, 20%, and 30% blends of WA and WB was 2.05%, 1.43%, and 0.55%, respectively, for WA, and 1.52%, 0.32%, and 1.59%, respectively, for WB. The negative value points to an improvement of the carbon burnout in comparison to unstaged WH single coal firing. An improvement in the carbon burnout was achieved by increasing the low-rank fuel blending ratio, with this positive effect being more dominant for the woody biomass blending. As mentioned above, the woody biomass has a higher VM content than the coal. The tendency of the VM is to be easily evolved in the primary combustion zone, even at lower temperatures than that in the coal. Therefore, the reaction of the unburned carbon in the burnout zone could be more active as a result of the enhanced ignition processes in the primary combustion zone. This phenomenon means that the woody biomass enhances the ignition of the coal, which is supported by the synergistic behavior of the improved NOx reduction and carbon burnout. To illustrate the above synergistic effect, the temperature distributions for a blending proportion of 30% were compared with the results obtained with three blending conditions, including WA30,
124
Y. Sung et al. / Experimental Thermal and Fluid Science 71 (2016) 114–125
WB30, and AB30, as shown in Fig. 11. The AB30 condition, which represents a blend with sub-bituminous coal (AD) and woody biomass showed the lowest temperature distribution for both SL0 and SL3. For blending combinations of bituminous coal/sub-bituminous coal and bituminous coal/biomass at SL3 as shown in Fig. 11b, the temperature for WA30 and WB30 was higher than that observed for WH single coal combustion when the thermal input was constant, based on the energy fraction. This enhanced temperature for WB30 was more predominant than that for WA30 because a large volatile gas reaction from the biomass content of the blends assists with the combustion of the unburned carbon from the WH coal. This synergistic reaction could occur mainly in the primary combustion zone, as shown in Fig. 11b. This is a synergistic effect of fuel blending on combustion performance with air-staged combustion. This research revealed that hybrid NOx reduction technology that combines air staging with low-rank fuel blending has the potential to simultaneously reduce exhaust gas emissions and enhance combustion performance when the combustion of pulverized coal particles is incomplete. This knowledge will be very useful for application to pollution control in existing thermal power plants using woody biomass blending technology. Moreover it can be concluded that the relative NOx reduction and carbon burnout rates are reproducible in the utility boilers. For this, the burner and OFA operations should be optimized with increasing the biomass blending proportion because the changing co-fired fuels leads to the variation of fuel-rich intensity or O2 availability in the primary combustion zone. Thus, the movements of burner tilt and yaw angles are very important to control the coal particle residence time in both primary combustion zone and burnout zone.
4. Conclusions Co-firing and air-staging were applied in an attempt to reduce NOx emissions in a pulverized coal fired furnace. Influences of air staging level, fuel types, and co-firing ratio of woody biomass on NOx emission and burnout characteristics were evaluated. The main results can be summarized as follows: (1) For single coals, sub-bituminous coal was more favorable to reduce NOx emissions due to low fuel-N composition. Specifically, NOx reductions for bituminous and sub-bituminous coal were 22% and 41%, respectively, at a primary zone air ratio of 0.9 and an air staging level 3. This tendency was more dominant with increasing the air staging levels. (2) For the co-firing, given that biomass has a high volatile and low carbon content, it could be successfully applied to the air-staged combustion. The NOx reduction for the blends of 30% was 33% for WA30, and 43% for WB30. In addition, the carbon burnout and flame temperature increased by cofiring woody biomass. The improvement of carbon burnout performance was 0.6% for WA30, and 1.6% for WB30. (3) Hybrid NOx reduction technology combining air staging and low-rank fuel blending has shown the potential to reduce exhaust gas emissions and enhance combustion performance.
Acknowledgements This work was supported by the Human Resources Development program (No. 20144010200780) of the Korea Institute of Energy Technology Evaluation and Planning (KETEP) grant funded by the Korea Government Ministry of Trade, Industry and Energy.
References [1] J. Shin, J. Woo, S. Huh, J. Lee, G. Jeong, Analyzing public preferences and increasing acceptability for the renewable portfolio standard in Korea, Energy Econ. 42 (2014) 17–26. [2] L. Dong, S. Gao, W. Song, J. Li, G. Wu, NO reduction in decoupling combustion of biomass and biomass-coal blend, Energy Fuels 102 (2009) 224–228. [3] W. Fan, Z. Lin, J. Kuang, Y. Li, Impact of air staging along furnace height on NOx emissions from pulverized coal combustion, Fuel Process. Technol. 91 (2010) 625–634. [4] D. Pudasainee, J.H. Kim, Y.C. Seo, Mercury emission trend influenced by air pollutants regulation for coal-fired power plants in Korea, At. Environ. 43 (2009) 6254–6259. [5] L. Sloss, Legislation, Standards and Methods for Mercury Emissions Control, CCC/195, IEA Clean Coal Centre, London, UK, 2012 (Apr). [6] J. Shen, J. Liu, H. Zhang, X. Jiang, NOx emission characteristics of superfine pulverized anthracite coal in air-staged combustion, Energy Convers. Manage. 74 (2013) 454–461. [7] Z. Li, Y. Liu, Z. Chen, Q. Zhu, J. Jia, J. Li, Z. Wang, Y. Qin, Effect of the air temperature on combustion characteristics and NOx emissions from a 0.5 MW pulverized coal-fired furnace with deep air staging, Energy Fuels 26 (2012) 2068–2074. [8] W. Bai, H. Li, L. Deng, H. Liu, D. Che, Air-staged combustion characteristics of pulverized coal under high temperature and strong reducing atmosphere conditions, Energy Fuels 28 (2014) 1820–1828. [9] S. Niksa, G. Liu, L. Felix, P.V. Bush, D.M. Boylan, Predicting NOx emissions from biomass cofiring, Prepr. Pap.-Am. Chem. Soc. Div. Fuel Chem. 48 (2) (2003) 643–648. [10] W.L. VanDeKamp, D.J. Morgan, The co-firing of pulverized bituminous coals with straw, waste paper and municipal sewage sludge, Combust. Sci. Technol. 121 (1996) 317–332. [11] H. Spliethoff, K.R.G. Hein, Effect of co-combustion of biomass on emissions in pulverized fuel furnaces, Fuel Process. Technol. 54 (1998) 189–205. [12] A. Gani, K. Morishita, K. Nishikawa, I. Naruse, Characteristics of co-combustion of low-rank coal with biomass, Energy Fuels 19 (2005) 1652–1659. [13] A. Kazagic, I. Smajevic, Experimental investigation of ash behavior and emissions during combustion of Bosnian coal and biomass, Energy 32 (2007) 2006–2016. [14] S. Li, T. Xu, P. Sun, Q. Zhou, H. Tan, S. Hui, NOx and SOx emissions of a high sulfur self-retention coal during air-staged combustion, Fuel 87 (2008) 723–731. [15] A. Kazagic, I. Smajevic, Synergy effects of co-firing wooden biomass with Bosnian coal, Energy 34 (2009) 699–707. [16] J.P. Smart, R. Patel, G.S. Riley, Oxy-fuel combustion of coal and biomass, the effect on radiative and convective heat transfer and burnout, Combust. Flame 157 (2010) 2230–2240. [17] J. Bai, C. Yu, L. Li, P. Wu, Z. Luo, M. Ni, Experimental study on the NO and N2O formation characteristics during biomass combustion, Energy Fuels 27 (2013) 515–522. [18] J. Wang, W. Fan, Y. Li, M. Xiao, K. Wang, P. Ren, The effect of air staged combustion on NOx emissions in dried lignite combustion, Energy 37 (2012) 725–736. [19] A. Ribeirete, M. Costa, Impact of the air staging on the performance of a pulverized coal fired furnace, Proc. Combust. Inst. 32 (2009) 2667–2673. [20] S. Munir, W. Nimmo, B.M. Gibbs, The effect of air staged, co-combustion of pulverized coal and biomass blends on NOx emissions and combustion efficiency, Fuel 90 (2011) 126–135. [21] S.S. Daood, M.T. Javed, B.M. Gibbs, W. Nimmo, NOx control in coal combustion by combining biomass co-firing, oxygen enrichment and SNCR, Fuel 105 (2013) 283–292. [22] M. Ikeda, H. Makino, H. Morinaga, K. Higashiyama, Y. Kozai, Emission characteristics of NOx and unburned carbon in fly ash during combustion of blends of bituminous/sub-bituminous coals, Fuel 82 (2003) 1851–1857. [23] S.A. Skeen, B.M. Kumfer, R.L. Axelbaum, Nitric oxide emissions during coal and coal/biomass combustion under air-fired and oxy-fuel conditions, Energy Fuels 24 (2010) 4144–4152. [24] L.I. Darvell, L. Ma, J.M. Jones, M. Pourkashanian, A. Williams, Some aspects of modeling NOx formation arising from the combustion of 100% wood in a pulverized fuel furnace, Combust. Sci. Technol. 186 (2014) 672–683. [25] P. Glarborg, A.D. Jensen, J.E. Johnsson, Fuel nitrogen conversion in solid fuel fired systems, Prog. Energy Combust. Sci. 29 (2003) 89–113. [26] T. Wu, M. Gong, E. Lester, P. Hall, Characteristics and synergistic effects of cofiring of coal and carbonaceous wastes, Fuel 104 (2013) 194–200. [27] C. Moon, Y. Sung, S. Eom, G. Choi, NOx emissions and burnout characteristics of bituminous coal, lignite, and their blends in a pulverized coal-fired furnace, Exp. Therm. Fluid Sci. 62 (2015) 99–108. [28] Y. Sung, G. Choi, Effectiveness between swirl intensity and air staging on NOx emissions and burnout characteristics in a pulverized coal fired furnace, Fuel Process. Technol. 139 (2015) 15–24. [29] A. Ribeirete, M. Costa, Detailed measurements in a pulverized-coal-fired largescale laboratory furnace with air staging, Fuel 88 (2009) 40–45. [30] A. Ribeirete, M. Costa, Co-combustion of biomass in a natural gas-fired furnace, Combust. Sci. Technol. 175 (2003) 1953–1977.
Y. Sung et al. / Experimental Thermal and Fluid Science 71 (2016) 114–125 [31] Y.M. Sung, C.E. Moon, J.R. Kim, S.C. Kim, T.H. Kim, S.I. Seo, G.M. Choi, D.J. Kim, Influence of pulverized coal properties on heat release region in turbulent jet pulverized coal flames, Exp. Therm. Fluid Sci. 35 (2011) 694–699. [32] C. Moon, Y. Sung, S. Ahn, T. Kim, G. Choi, D. Kim, Thermochemical and combustion behaviors of coals of different ranks and their blends for pulverized-coal combustion, Appl. Therm. Eng. 54 (2013) 111–119. [33] S. Xiu, Z. Li, B. Li, W. Yi, X. Bai, Devolatilization characteristics of biomass at flash heating rate, Fuel 85 (2006) 664–670. [34] S. Badzioch, P.G.W. Hawksley, Kinetics of thermal decomposition of pulverized coal particles, Ind. Eng. Chem. Process Des. Dev. 9 (4) (1970) 521–530.
125
[35] T.R. Ballantyne, P.J. Ashman, P.J. Mullinger, A new method for determining the conversion of low-ash coals using synthetic ash as a tracer, Fuel 84 (2005) 1980–1985. [36] C. Casaca, M. Costa, Detailed measurements in a laboratory furnace with reburing, Fuel 90 (2011) 1090–1100. [37] G. Wang, R.B. Silva, J.L.T. Aevedo, S. Martins-Dias, M. Costa, Evaluation of the combustion behaviour and ash characteristics of biomass waste derived fuels, pine and coal in a drop tube furnace, Fuel 117 (2014) 809–824. [38] A.K. Gupta, D.G. Lilley, N. Syred, Swirl Flows, Abacus, Kent, 1984. [39] T.H. Ye, J. Azevedo, M. Costa, V. Semiao, Co-combustion of pulverized coal, pine shells, and textile wastes in a propane-fired furnace, Combust. Sci. Technol. 176 (2004) 2071–2104.
Contents lists available at ScienceDirect
Experimental Thermal and Fluid Science journal homepage: www.elsevier.com/locate/etfs
Synergistic effect of co-firing woody biomass with coal on NOx reduction and burnout during air-staged combustion Yonmo Sung a,b, Sangmin Lee b, Changhyun Kim c, Dongheon Jun d, Cheoreon Moon b, Gyungmin Choi b,⇑, Duckjool Kim b a
Department of Mechanical Engineering, Imperial College London, London SW7 2AZ, United Kingdom School of Mechanical Engineering, Pusan National University, Busan 609-735, South Korea Plant Management Team, Korea South-East Power Co., Ltd., Gyeongsangnam-do 660-031, South Korea d Taean Thermal Power Complex Division, Korea Western Power Co., Ltd., Chungcheongnam-do 357-914, South Korea b c
a r t i c l e
i n f o
Article history: Received 18 May 2015 Received in revised form 18 October 2015 Accepted 19 October 2015 Available online 23 October 2015 Keywords: Synergistic effect Woody biomass Co-firing NOx reduction Air-staging
a b s t r a c t Hybrid technologies combining fuel blending and air-staging processes have been applied to a pulverized coal fired furnace to reduce NOx emissions. In this study, an Australian bituminous coal (Whitehaven), an Indonesian sub-bituminous coal (Adaro), and an Indonesian woody biomass were selected as fuel with blending ratio of 10%, 20%, and 30% of the low-rank fuels, and the air-staging levels were set to 235 mm, 390 mm, 585 mm, and 760 mm. The purpose of this study was to investigate the synergistic effect of woody biomass co-firing on the level of NOx emissions and the degree of carbon burnout under air-staged conditions. For single coals, sub-bituminous coal was more favorable than that for bituminous coal to reduce NOx emissions due to low fuel-N composition. This tendency was more dominant with increasing the air staging levels. In the co-firing of woody biomass with coal, as the biomass is highly volatile but has a low carbon content, it could be successfully applied to low-NOx combustion under air-staged conditions. In addition, the degree of carbon burnout and the flame temperature both increased. As a result of this research, we determined that hybrid NOx reduction technologies have the potential to reduce exhaust gas emissions and enhance combustion performance. A dominant synergistic effect on NOx reduction and carbon burnout was observed when woody biomass co-firing with coal was applied to air-staged combustion. Ó 2015 Elsevier Inc. All rights reserved.
1. Introduction Coal-fired thermal power plants in Korea currently co-fire bituminous/sub-bituminous coal because of the difficulty in obtaining high-grade coal. The importance of fuel blending technology is being emphasized as it opens up the possibility of using low-rank fuels such as biomass or waste material. Recently, the utilization of both woody biomass and sewage sludge has become commonplace in existing coal-fired thermal power plants because Korea’s renewable portfolio standard (RPS) obligates electricity providers to produce a specified fraction of their electricity from renewable energy, with the remainder being generated from fossil fuel and nuclear reactors. In Korea, 13 electricity providers have been mandated to generate 2% of their gross power output from renewable energy sources, such as solar, wind, hydro, tidal power, fuel cells, hydrogen, biomass, and waste, as a result of the ⇑ Corresponding author. Tel.: +82 51 510 2476; fax: +82 51 512 5236. E-mail address: [email protected] (G. Choi). http://dx.doi.org/10.1016/j.expthermflusci.2015.10.018 0894-1777/Ó 2015 Elsevier Inc. All rights reserved.
implementation of the RPS regulation in January 2012. This percentage is set to rise to 10% in 2022 [1]. Most coal-fired thermal power plants in Korea determined that the best way for them to satisfy the RPS regulations was to introduce the co-firing of biomass since it is the fourth largest source of energy after coal, petroleum, and natural gas [2]. Environmental issues, especially NOx emissions, continue to be one of the most important challenges facing coal-fired utilities. The regulations limiting NOx emissions are actually very severe [3]. For example, as of 2008, the NOx emission limit imposed on power plants over 500 MWth in the European Union (EU) has been 500 mg/Nm3 at 6% O2. This limit will drop to 200 mg/Nm3 at 6% O2 from 1 January 2016. In this context, emission limits for primary air pollutants such as NOx, SOx, and dust from coal-fired thermal power plants in Korea have been continually reinforced, shown in Table 1 [4,5]. The air quality preservation act by Korean ministry of environment requires that new facilities must reduce NOx emissions as 103 mg/Nm3 at 6% O2 from 1 January 2015.
115
Y. Sung et al. / Experimental Thermal and Fluid Science 71 (2016) 114–125 Table 1 Permissible emission standards for NOx, SOx, and dust from coal-fired thermal power plant in South Korea. Pollutants
Facility division
Limit before 2014.12.31
Limit after 2015.1.1
NOx (mg/Nm @6%O2)
Installed before 1996.6.30 Installed after 1996.7.1 Installed after 2015.1.1
<308 <164
<287 <144 <103
SOx (mg/Nm3@6%O2)
Capacity > 100 MW Installed before 1996.6.30 Installed before 2014.12.31 Installed after 2015.1.1 Capacity < 100 MW Installed before 1996.6.30 Installed before 2014.12.31 Installed after 2015.1.1
<286
Capacity Installed Installed Installed Capacity Installed Installed Installed
>500 MW <28 <19
3
Dust (mg/Nm3@6%O2)
before 2001.6.30 after 2001.7.1 after 2015.1.1 before 2001.6.30 after 2001.7.1 after 2015.1.1
To satisfy this impending tightening of the NOx emission limits, technologies such as in-furnace removal (i.e. air-staging, reburning, the use of low-NOx burners, etc.), flue gas purification (i.e. selective catalytic reduction (SCR), selective non-catalytic reduction (SNCR), etc.), and other combustion methods (i.e. coal-water slurry combustion, circulated fluidized bed combustion etc.) have been widely used for coal-based power generation applications [6]. Among the existing methods, air-staged combustion is one of the most efficient and attractive technologies for reducing NOx emissions, because it does not require expensive new equipment, such as SCR systems [7]. Unfortunately, there are also serious disadvantages such as high operating costs, ammonia slip (the passing of ammonia through the SCR reactor), and the easy deactivation of the catalyst in SCR applications [8]. Ammonia slip, in particular, can lead to a blockage in the SCR catalyst due to an unwanted reaction between un-reacted ammonia and unburned carbon. Generally, biomass co-firing should reduce NOx emissions simply because most forms of biomass contain less fuel-N than the coal they replace [9]. Therefore, it is important to minimize the initial formation of NOx by in-furnace removal methods including airstaging and biomass co-firing to avoid having to implement flue gas purification or minimize its operation. Numerous articles dealing with NOx reduction technologies such as air-staged combustion and low-rank fuel blending have been published [10–18]. Ribeirete and Costa [19] investigated the impact of air-staging on the overall performance of a large-scale laboratory furnace fired by an industrial pulverized coal swirl burner. The influences of the axial position of the staged-air injector, the primary zone stoichiometric ratio, the coal type, and the configuration of the staged-air injectors on the pollutant emissions and particle burnout were quantified [18]. Munir et al. [20] showed that the addition of biomass has a positive effect on NOx reduction and carbon burnout under the optimum conditions that were identified as part of their study. Daood et al. [21] investigated NOx control in coal combustion by combining biomass co-firing, oxygen enrichment, and SNCR in a 20-kW combustion facility. It was found that NOx control by SNCR and the oxygen-enriched co-firing of biomass in power station boilers would result in lower NOx emissions and a higher CO2 concentration that would lead to more efficient scrubbing with better carbon burnouts. When a furnace is co-firing biomass and coal, it is important to consider the slagging, fouling, and tube corrosion that may occur as a result of the presence of unburned carbon. This study focused on the effect of air-staged combustion and woody biomass co-firing
<286 <228 <143 <428 <371 <228 <143
<500 MW <37 <28
>100 MW <23 <19 <9 <100 MW <37 <28 <28
on the overall performance, including NOx emissions and carbon burnout, since there are still doubts about the effect on gas emissions of low-rank fuel blending. For having nearly same coal properties of two sub-bituminous coals, NOx emissions from two subbituminous coals (Wara and Adaro), which were blended with bituminous, exhibited an inverse tendency with increasing subbituminous coal blending ratio although the moisture contents of them were different [22]. The effect of biomass co-firing on NOx emissions was found to be dependent upon the particle size, with smaller particles reducing NOx emissions as the biomass co-firing ratio increased [23]. Air-staged combustion is used to reduce the stoichiometric ratio in the primary combustion zone, and thus the fuel-rich condition causes the NOx to act as a reducing species. This reducing environment can be locally intensified by adding low-rank fuels such as woody biomass. In particular, the nitrogen content in woody biomass is relatively lower compared to straw and other agricultural residues, although the woods and the straws decompose in different ways in terms of the release of the N-components [24,25]. Moreover, air-staged combustion is, in general, desirable for reducing NOx emissions; if, however, there is insufficient air in the primary combustion zone, the flame temperature is typically applied to low-temperature processes [23]. Different components in the biomass, i.e., lignin, and semi-cellulous/ cellulous materials, decompose at different temperatures, and thus the biomass starts to release its volatile components at a relatively low temperature [26]. This behavior could help alleviate the drop in temperature caused by deep air-staging through possible synergistic effects when coal is co-fired with woody biomass. There are lots of studies on low-rank fuel blending with air-staged combustion. However, woody biomass co-firing with air staging is rarer to produce synergy effects on reducing NOx emissions and enhancing carbon burnout performances simultaneously, and even fewer studies have been attempted under the combination of several parameters as oxy-firing, air staging, and agricultural biomass co-firing for the synergy of NOx and burnout [20,21]. In this study, a hybrid technology combining fuel blending and air-staging was applied to reduce the NOx emissions from a pulverized coal fired furnace. The combustion and emission characteristics of different blends of coal (bituminous and subbituminous) and woody biomass were investigated using a pulverized coal fired furnace. To understand the details of the air-staged combustion, in this study we undertook detail measurements of gas species concentration, temperature, and carbon burnout under
116
Y. Sung et al. / Experimental Thermal and Fluid Science 71 (2016) 114–125
a range of operating conditions. The purpose of this study was to investigate the synergistic effect of woody biomass co-firing on NOx reduction and carbon burnout under air-staged conditions. 2. Experimental 2.1. Materials In this study, we used three solid fuel samples, namely, an Australian bituminous coal (WH, Whitehaven) as a high-rank coal, an Indonesian sub-bituminous coal (AD, Adaro) as a low-rank coal, an Indonesian woody biomass (BM, Biomass), and blends thereof. Each sample was milled and sieved to the desired particle size as listed in Table 2. The particle size of the samples was estimated by using a laser diffraction particle size analyzer (LS 13 320, Beckman Coulter). The mean particle size of the WH, AD, and BM was 118 lm, 122 lm, and 286 lm, respectively. For the blends, the blending ratio, as based on the energy fraction (heating value), was 10%, 20%, and 30% of the low-rank fuels. These samples were fed directly into a pulverized coal fired furnace without any moisture-related treatment, and the results of the proximate and ultimate analyses, as well as the calorific values, of each fuel sample are listed in Table 3. 2.2. Experimental apparatus and methods The experimental detail of the test rig has already been discussed in previous published articles [27,28]. Fig. 1 is a schematic diagram of the pulverized coal fired furnace. The cylindrical comTable 2 Particle size distributions of fuel samples obtained by the laser diffraction particle size analyzer. Samples
Bulk density (kg/m3)
d10, <10%
WHa ADb BMc
593 624 418
11 19 33
d25, <25%
d50, <50%
d75, <75%
d90, <90%
dmean, mean
106 112 232
180 181 397
244 241 612
118 122 286
Size (lm)
a b c
38 51 113
WH: Australian bituminous coal; Whitehaven. AD: Indonesian sub-bituminous coal; Adaro. BM: Indonesian woody biomass.
Table 3 Proximate and ultimate analysis of solid fuel samples and LPG compositions. Quantity
WH
AD
BM
Proximate analysis (wt%, air-dry basis) Moisturea Moisture Volatiles Fixed carbon Ash
13.04 5.02 30.52 51.08 13.38
26.83 19.38 38.69 39.02 2.91
6.88 6.77 74.85 16.01 2.37
Ultimate analysis (wt%, dry basis) Carbon Hydrogen Oxygen Nitrogen Sulfur Ash Net calorific value (MJ/kg, as-received basis)
72.1 4.62 7.38 1.55 0.26 14.09 25.16
70.4 4.57 20.04 1.29 0.09 3.61 21.19
41.2 6.02 49.7 0.45 0.1 2.53 18.26
Composition (vol%) C2H6 C3H8 C4H10 Net calorific value (MJ/kg) a
as-received basis.
LPG
1 97 2 46.04
bustion chamber is down-fired to facilitate particulate removal. The combustion chamber is made up of three individual watercooled steel segments, each having an internal diameter of 200 mm while being 500 mm high. The roof section and all of the segments are lined with a layer of refractory material (width: 130 mm). Each segment has four pairs of diametrically opposed 120-mm round ports to allow the observation of the combustion. R-type thermocouples and gas sampling ports for recording the in-furnace temperatures and gas concentration are installed in every segment. Six pairs of staging air ports, oriented tangentially to the main swirl flow of the burner, are mounted in the refractory materials. In addition, a thermocouple lance is located in the cooling water circuit of every segment to record the outlet cooling water temperatures. The axial positions of the air-staging level, relative to the burner port, were at 235 (SL1), 390 (SL2), 585 (SL3), and 760 mm (SL4), while the thermocouple lances were placed at 65 (Z1), 210 (Z2), 355 (Z3), 535 (Z4), and 680 mm (Z5). The burner has a coaxial dual-piping structure with a main burner port (inner diameter: 10 mm) and an annular slit (width: 5.5 mm) installed outside the main burner port. Primary air was used to transport the fuel samples from the fuel feeder into the furnace. Primary air and solid fuel are supplied from the main burner port. The secondary air enters the combustion chamber through an annular slit port and encounters a swirl generator with a vane angle of 60°. There is also a refractory burner quarl with a half-angle of 30° with a pair of annulars (inner diameter: 6 mm) to supply liquefied petroleum gas (LPG) fuel which has positive effects on coal particle ignition and flow formation such as an internal recirculation zone (IRZ) and an outer recirculation zone (ORZ) as shown in Fig. 1. The LPG flow rate is held to the minimum rate needed to stabilize the swirl flames [29]. The air and LPG flow rate were controlled by a mass flow controller (KOFLOC-3660, Kojima Instruments). The concentrations of O2, CO2, and NOx were measured in a dry basis using a gas analyzer (VA-3000, HORIBA), which is a multi-gas analyzer that uses non-dispersive infrared absorptiometry, the chemiluminescence method, and magnetic pressure analysis to capture data. Gas measurements were taken using a stainless steel water-cooled dual-pipe probe [27,30]. The diameter of the sampling hole for the probe was 1 mm. A wet sample was drawn through the probe and into the sampling system by an oil-free pump. Most of the moisture was removed by a silica-gel filter. Then, any residual moisture and particles were removed by a pair of filters such that a constant supply of dry, filtered combustion gas was available to the analytical instrument. The analog outputs of the analyzer were transmitted via high-speed A/D boards (NI-DAQ 9172, National Instruments), at a sampling rate of 2 Hz, to a computer where the signals were processed and then the mean values computed. The concentration of gas species was measured using the gas analyzers for about 10 min, five times, and then the ensemble average value was used. The variations in the gas concentrations are expressed by error bars in the graphs presented in the Results and Discussion. Example statistical analysis data is presented in Table 4. Temperature measurements were obtained using 50-lm diameter fine-wire R-type (Pt-Rh 13%/Pt) thermocouples. Each thermocouple’s bead was surrounded by a 10-mm diameter ceramic tube to prevent the thermocouples from becoming attached to unburned coal particles. The ceramic tube thermocouples were calibrated in turbulent jet pulverized coal flames [31,32], relative to unprotected thermocouples. In order to better understand the temperature change with different combustion environments, in-furnace temperatures were measured at the furnace center in radial directions in interval of 10 mm for every axial position Z, shown in Fig. 1, and then 2-d half contours were obtained from
117
Y. Sung et al. / Experimental Thermal and Fluid Science 71 (2016) 114–125
Fig. 1. Schematic diagram of pulverized coal combustion system.
placed in a glass container, in a filter. The collected wet ash samples were then placed in an oven at 383 K for three hours to dehydrate, and then the de-watered particles were burnt in a thermogravimetric analyzer (TGA) (SDT-Q600, TA Instruments) and the weight loss was calculated by applying the ash tracer method [33]. The ash tracer method is prone to analytical errors when used with fuels with low ash contents, which can be of importance for many biomass fuels, in particular woody materials. Scatter in the results caused by analytical errors is less important for fuels with high ash contents [34,35]. The carbon burnout data was obtained using the following equation [36,37]:
Table 4 Example statistics of data analysis of exhaust gas emissions.
WH@SL0 WH@SL1 WH@SL2 WH@SL3 WH@SL4 AD@SL0 AD@SL1 AD@SL2 AD@SL3 AD@SL4
O2 (%) mean, SD
CO2 (%) mean, SD
NO (ppm) mean, SD
NO (mg/Nm3) mean
3.83, 0.16 4.53, 0.14 4.54, 0.11 4.63, 0.08 4.77, 0.15 2.48, 0.11 2.7, 0.21 2.86, 0.2 2.91, 0.11 2.97, 0.07
15.58, 14.94, 14.67, 14.41, 14.23, 17.59, 17.42, 17.23, 17.05, 16.93,
330.87, 4.55 300.44, 4.05 268.81, 5.22 257.01, 5.71 258.69, 3.71 322.38, 3.54 282.05, 6.2 245.2, 8.22 196.52, 6.94 177.63, 4.43
413 375 335 321 323 402 352 306 245 222
0.2 0.2 0.12 0.11 0.19 0.08 0.28 0.24 0.14 0.14
W¼
Notes: SD is the standard deviation, NOx data is corrected at O2 6%, SL0 is the unstaged condition, SL1 to SL4 are the air staging level 1 to 4, respectively.
55 point data. The empty values among axial positions (Z1–Z5) were filled interpolation values, using Tecplot 360 (Tecplot, USA). Fly ash sampling was performed with the aid of a stainless steel, water-cooled, water-quenched probe normally used for gas species sampling. Ash samples were collected with the probe and then
1 ðwk =wx Þ 1 wk
ð1Þ
where W is the carbon burnout, w is the dry mass fraction, and the subscripts k and x refer to the ash content in the input solid-fuel and char sample, respectively. 2.3. Experimental conditions Table 5 summarizes the furnace operating conditions for the air-staging and fuel-blending parametric trials. For all of the exper-
Table 5 Operating conditions of pulverized coal fired furnace. WH Total thermal input (kW) Overall excess air (%) Primary zone stoichiometric ratio, SR1 Solid fuel feeding rate (kg/h) LPG flow rate (slpm) Primary air flow rate (slpm) Secondary air flow rate (slpm) Secondary air swirl number Staging air flow rate (slpm)
AD
16 = 8 (solid fuel) + 8 (LPG) 15 0.9, 1.15 0.9, 1.15 1.14 1.35 5 10 253, 288 265, 303 1.33 35, 0 38, 0
WB10
WB20
WB30
WA10
WA20
WA30
AB30
0.9 1.19
0.9 1.23
0.9 1.27
0.9 1.17
0.9 1.19
0.9 1.21
0.9 1.42
250
246
241
255
256
257
250
34
33
32
35
36
36
34
Notes: WA is the blend of WH coal + AD coal, WB is the blend of WH coal + Biomass, AB30 is the blend of AD coal 70% + Biomass 30%.
118
Y. Sung et al. / Experimental Thermal and Fluid Science 71 (2016) 114–125
air staging and co-firing, the test of WH single coal firing without any air-staging or co-firing, while keeping the SR0 of 1.15, was taken as baseline, and the reduction (%) in the amount of gas emissions and carbon burnout was defined by the following equation:
Reductionð%Þ ¼
baseline value obtained value baseline value
ð2Þ
3. Results and discussion 3.1. Combustion and emission characteristics for single type of coal with air staging level Fig. 2. NOx emission and burnout performance of two single types of coal for different air staging levels.
imental conditions, the thermal input was constant at 16 kWth, based on the net calorific value. The preheating temperatures for the primary, secondary, and staging air were held constant at 338 K, 573 K, and 303 K, respectively, but this preheat thermal energy was not considered in the thermal input calculations. The overall stoichiometric ratio (SR0) was 1.15. The secondary air swirl number was 1.33 [38]. To enable the comparison of the effects of
Several studies showed that the NOx reduction with air-staged combustion primarily depends upon the location of the air-staging level (SL) and the primary zone stoichiometric ratio (SR1) [6,19,20]. In the case of the above studies, as the value of the SR1 fell, as did the NOx emissions. From the previous published article [27], the SR1 value of 0.9 was found to be the optimum for both NOx reduction and carbon burnout (W). Therefore, the fixed SR1 value of 0.9 was used, and the value of SR0 was maintained at 1.15 along with the location of each SL from level 1 (SL1) to level 4 (SL4).
Fig. 3. In-furnace temperature contours for two single types of coal at SR1 of 0.9 with a different SL. (a) WH coal, and (b) AD coal.
Y. Sung et al. / Experimental Thermal and Fluid Science 71 (2016) 114–125
Fig. 4. O2 and CO2 gas emissions of two single types coal for different air staging levels.
Fig. 2 shows the effect of the SL value on NOx emission and burnout characteristics of the WH and AD coals. It can be seen that the NOx emissions decrease as the distance between the SL and the burner exit increases. In particular, the NOx emissions can be seen to be 330, 300, 269, 257, and 259 ppm, from no staging (SL0) to staging level 4 (SL4) with WH coal, and 322, 282, 245, 197, and 178 ppm from SL0 to SL4 with AD coal. When the SL is placed in axial locations relatively close to the burner exit, as in the case of SL1 or SL2, the coal particle residence time in the primary combustion zone is relatively short because of the short length of the fuel rich zone (reducing environment). Conversely, the increased residence time of the coal particles in the fuel rich zone, as in the case with SL3 or SL4, with a relatively lean O2 concentration, promotes a reduction in the amount of NOx emissions. Previous works by Ribeirete and Coata [19] and Munir et al. [20] produced similar results. From Fig. 2, it can be seen that the effect of SL on the
119
NOx emissions differs depending upon the coal type. The differences can be attributed to the rapid production of gas-phase reductive species such as hydrocarbons for sub-bituminous coal, whereas the massive gas-phase reductive species for bituminous coal are mainly derived from the gasification reactions of char, which generally takes relatively longer to complete [8]. Although air-staged combustion in pulverized coal fired boilers can reduce NOx emissions, it could have a negative effect on the carbon burnout (W) performance of the coal particles. Therefore, it is very important to achieve both low NOx emissions and a high W. The W values for the WH and AD coals were examined and are shown in Fig. 2. The value of W decreases monotonously as SL increases, although the value of W for two different types of coal is quite different. For instance, when SL increases from SL1 to SL4, the W for the WH coal drops from 92.85% to 89.11%, while for the AD coal, the W drops from 99.85% to 99%. Although the NOx reduction based on SL0 for the WH coal was 22.32% at SL3 and 21.82% at SL4, and for the AD coal was 40.61% at SL3 and 46.31% at SL4, the W reduction for the WH coal was 2.95% at SL3 and 4.02% at SL4, and for the AD coal was 0.23% at SL3 and 0.86% at SL4. This means that the NOx reduction is insignificant, however, a significant decrease in the W value when the SL increases from SL3 to SL4. Therefore, for the co-firing tests, the condition of SL3 and SR1 values of 0.9 was fixed. Fig. 3 shows the in-furnace temperature contours for two single types of coal at SR1 of 0.9 with a different SL. The maximum temperature and temperature gradient can be seen in the vicinity of the burner, while the temperature decreases along with the furnace height in the burner zone. In Fig. 3a, for the WH coal, the temperature decreased as SL increased at a point 200 mm from the burner port. In Fig. 3b for the AD coal, however, the tendency was found to be quite different. The temperature showed a similar tendency as SL increased. This was mainly because of the relatively high amount of volatile matter but less fixed carbon in the AD coal,
Fig. 5. Co-firing characteristics for bituminous/sub-bituminous coals and bituminous coal/woody biomass at SL3 and SR1 of 0.9. (a) NOx, (b) carbon burnout, (c) O2, and (d) CO2.
120
Y. Sung et al. / Experimental Thermal and Fluid Science 71 (2016) 114–125
Fig. 6. In-furnace temperatures as a function of low-rank fuel blending ratio at SL3 and SR1 of 0.9. (a) 2-d temperature contour for WA blending combination, and (b) 2-d temperature contour for WB blending combination.
which makes the combustion reactivity of the AD coal better than that of the WH coal. For the SL0 and SL1 conditions, the temperature at around a height of 200 mm for the AD coal was lower than that for the WH coal. However, it was higher than that for the WH coal at SL2 to SL4. In particular, this reversed trend in the temperature distributions can be seen more clearly in the downstream region (at a height in excess of 300 mm). These results can explain why the W for the AD coal was higher than that for the WH coal, as shown in Fig. 2. The distributions of the O2 and CO2 gas emissions at the furnace exit for the WH and AD coals are shown in Fig. 4. The O2 and CO2 are complementary species, being the main reactant and product of the combustion, respectively. Although the O2 concentration is proportional to SL, the CO2 distributions showed a completely
opposite tendency. Also, the O2 concentration for the AD coal was lower than that for the WH coal, and the CO2 concentration for the AD coal was higher than that for the WH coal. In this study, the excess 3% O2 was considered with respect to the 15% excess air at SR0 of 1.15. In Fig. 4, for the unstaged conditions, the O2 concentrations for the WH and AD coals were around 3.8% and 2.5%, respectively. This can be explained by the temperature distribution and the carbon burnout characteristics. Although it depends upon the value of SL, the amount of unburned carbon in the fly ash of the WH coal was around 7%, as shown in Fig. 2. Given the reaction of C + O2 ? CO2, we can deduce that the O2 concentration will be more than 3%, with large amounts of unburned carbon remaining in the fly ash. In contrast, the O2 concentration for the AD coal was less than 3%, and the CO2 was higher than that for the WH coal.
Y. Sung et al. / Experimental Thermal and Fluid Science 71 (2016) 114–125
Fig. 7. Radial temperature profiles as a function of low-rank fuel blending ratio at SL3 and SR1 of 0.9. (a) WA at Z1, (b) WA at Z2, (c) WB at Z1, and (d) WB at Z2.
Fig. 8. NOx reduction mechanism of hybrid technology between air staging and low-rank fuel blending.
121
122
Y. Sung et al. / Experimental Thermal and Fluid Science 71 (2016) 114–125
Fig. 9. NOx reduction rate (base case: WH single coal firing without air staging and fuel blending) at SL3 and SR1 of 0.9 with co-firing.
Fig. 10. Burnout reduction rate (base case: WH single coal firing without air staging and fuel blending) at SL3 and SR1 of 0.9 with co-firing.
Moreover, there was basically no unburned carbon, and the overall temperature in the downstream region in the furnace was higher than that of the WH coal. This would certainly have a major impact on the synergistic effect on the combustion and emission performance under co-firing conditions with high- and low-rank fuels.
primary combustion zone than in the case of AD co-firing because the woody biomass has a higher amount of volatile matter that also decomposes faster than coal. Fig. 5c shows the influence of the co-firing ratio on the value of W. The value of W for the WA and WB blends is higher than that in non-blended WH coal combustion for given operating conditions. There is a tradeoff relationship between the NOx emissions and the carbon burnout. With both blending combinations, the value of W was improved as the low-rank fuel blending ratio increased, and the positive effect on the value of W for the biomass blending was relatively large, compared to the AD coal blending. This result is in agreement with those of previous studies [11,20,39]. This behavior could be linked to the higher volatility and reactivity of the woody biomass, making the unburned carbon in the bituminous single coal particles easier to burn when co-firing with woody biomass. For the blends, the maximum value of W for WA30 and WB30 was 93.35% and 94.32% for a given operating condition. Fig. 6 shows the temperature contours with a low-rank fuel blending ratio at an SL3 and SR1 of 0.9. Under every test condition, the temperature distributions exhibited a similar tendency. The temperature decreases with height in the furnace. The overall distributions with the addition of low-rank fuels in ratios from 10% to 30% were higher than in the case of firing with WH coal alone. This feature was very apparent in the case of woody biomass blending, and also in the results of a study on the co-combustion of low-rank coal with biomass [12]. From the radial temperatures in Fig. 7, the temperature for WA30 and WB30 was higher than that for WH single coal combustion when the thermal input was constant, based on the energy fraction. This is because the large volatile gas reaction arising from the low-rank single fuel in the blends assists with the carbon conversion reaction of the high-rank WH coal. Therefore, the W and CO2 values became higher as the biomass blending ratio increased, and thus the O2 value fell due to the enhanced gas phase volatile reaction from the biomass, as shown in Fig. 5.
3.2. Combustion and emission characteristics of blends of coal and biomass Fig. 5 shows the influence of low-rank fuel blending on the value of W and the emissions of gases such as NOx, O2, and CO2 for a given operating condition (SL3 and SR1 of 0.9) for the WA (WH + AD) and WB (WH + woody biomass) blends. NOx emissions that were very different from those of a composition of low-rank fuels were observed, as shown in Fig. 5a. The addition of subbituminous AD coal based on bituminous WH coal had a positive effect on the NOx reduction. This positive effect on the NOx reduction was more significant with the blends of WH coal and woody biomass. The minimum NOx emission for the co-firing at SL3 and SR1 of 0.9 was 222 ppm and 190 ppm for WA30 and WB30, respectively. This is mainly due to the lower nitrogen content of the biomass (fuel-N of 0.45%), compared to the sub-bituminous coal (fuelN of 1.29%) and the bituminous coal (fuel-N of 1.55%). Moreover, the dominant volatile-N compound in bituminous coal combustion is HCN, while in biomass combustion it is both HCN and NH3 [24]. The woody biomass, especially, gives largely HCN with some NH3, and the agricultural biomass gives mainly NH3 [25]. Essentially, there is more volatile matter in the woody biomass (VM of 74.85%) than in the sub-bituminous coal (VM of 38.69%), as shown in Table 3. For blends with a higher biomass proportion, the predominant combustion consisted of gas phase reactions in a strong local reducing environment under fuel-rich, O2-lean conditions. With regard to the residence time of solid fuels in the fuel-rich zone, as the biomass (FC of 16.01% and fuel-O of 49.7%) contains less fixed carbon and more oxygen relative to sub-bituminous coal (FC of 39.02% and fuel-O of 20.04%), the amount of stoichiometric air required based on the energy fraction is less than that required for coal combustion, and then the particle residence time for the biomass blending condition should increase because of the lower bulk velocity at the burner exit, as shown in Table 5. From Fig. 5b and d, we can see that the O2 concentration declines as the low-rank fuel blending ratio increases, while the CO2 distribution exhibits a completely opposite tendency. Also, the O2 concentration for the biomass blends was lower than that in the AD coal blends. Conversely, the CO2 concentration for the biomass blends was higher than that in the AD coal blends. With the co-firing of woody biomass, O2 is consumed more rapidly in the
3.3. Synergistic effect of hybrid NOx reduction technology on combustion and emission characteristics Fig. 8 is a schematic diagram of the NOx reduction method based on the fuel-rich, oxygen-lean concept. The most important point affecting NOx reduction technologies is to create a fuel-rich environment in the main gas/volatile combustion zone [3,8,20]. The basic principal of air-staged combustion is to reduce the stoichiometry in the primary combustion zone in which devolatilization takes place in an O2-lean atmosphere. Additionally, this reducing environment could be made locally strong by adding
Y. Sung et al. / Experimental Thermal and Fluid Science 71 (2016) 114–125
123
Fig. 11. In-furnace temperature contours for a blending proportion of 30% with three blending combinations such as WA (WH + WA), WB (WH + Biomass), AB (AD + Biomass). (a) SL0, and (b) SL3.
low-rank fuels that contain higher proportions of volatile matter and oxygen. The NOx concentration was found to be lower as SL increased because the length of the fuel-rich primary zone became longer as SL increased. Thus, the particle residence time in an O2lean atmosphere became longer. Fig. 9 shows NOx reduction rate at SL3 with co-firing, comparing the unstaged WH single coal firing condition. It can be seen that the NOx reduction with air staging for the WH coal was 22.32%, and for both air staging and co-firing, it showed a higher reduction rate. The NOx reduction for the blends of 10%, 20%, and 30% of WA and WB was 28.59%, 31.83%, and 32.87% for WA, and 32.26%, 38.83%, and 42.63% for WB. This reduction became particularly predominant as the air staging level and low-rank fuel blending ratio increased. As a result, the synergistic effect of the woody biomass blending is more efficient than that of sub-bituminous coal blending for controlling the NOx emissions of air-staged combustion. Fig. 10 compares the W reduction for two blending combinations such as WA (WH + WA) and WB (WH + Biomass) at SL3. It can clearly be seen that the W reduction for the WH coal with air-staging was 2.95%, meaning that the burnout performance fell as the intensity of the air-staging increased, as shown in Fig. 2. The reaction intensity for the staged condition in the main combustion zone is lower than that of the unstaged condition
due to the deficiency in the required theoretical air, and thus the overall combustion of the coal particles remains incomplete in the primary combustion zone. This could explain why the air staging produced greater NOx reduction and less carbon burnout. The W reduction for the 10%, 20%, and 30% blends of WA and WB was 2.05%, 1.43%, and 0.55%, respectively, for WA, and 1.52%, 0.32%, and 1.59%, respectively, for WB. The negative value points to an improvement of the carbon burnout in comparison to unstaged WH single coal firing. An improvement in the carbon burnout was achieved by increasing the low-rank fuel blending ratio, with this positive effect being more dominant for the woody biomass blending. As mentioned above, the woody biomass has a higher VM content than the coal. The tendency of the VM is to be easily evolved in the primary combustion zone, even at lower temperatures than that in the coal. Therefore, the reaction of the unburned carbon in the burnout zone could be more active as a result of the enhanced ignition processes in the primary combustion zone. This phenomenon means that the woody biomass enhances the ignition of the coal, which is supported by the synergistic behavior of the improved NOx reduction and carbon burnout. To illustrate the above synergistic effect, the temperature distributions for a blending proportion of 30% were compared with the results obtained with three blending conditions, including WA30,
124
Y. Sung et al. / Experimental Thermal and Fluid Science 71 (2016) 114–125
WB30, and AB30, as shown in Fig. 11. The AB30 condition, which represents a blend with sub-bituminous coal (AD) and woody biomass showed the lowest temperature distribution for both SL0 and SL3. For blending combinations of bituminous coal/sub-bituminous coal and bituminous coal/biomass at SL3 as shown in Fig. 11b, the temperature for WA30 and WB30 was higher than that observed for WH single coal combustion when the thermal input was constant, based on the energy fraction. This enhanced temperature for WB30 was more predominant than that for WA30 because a large volatile gas reaction from the biomass content of the blends assists with the combustion of the unburned carbon from the WH coal. This synergistic reaction could occur mainly in the primary combustion zone, as shown in Fig. 11b. This is a synergistic effect of fuel blending on combustion performance with air-staged combustion. This research revealed that hybrid NOx reduction technology that combines air staging with low-rank fuel blending has the potential to simultaneously reduce exhaust gas emissions and enhance combustion performance when the combustion of pulverized coal particles is incomplete. This knowledge will be very useful for application to pollution control in existing thermal power plants using woody biomass blending technology. Moreover it can be concluded that the relative NOx reduction and carbon burnout rates are reproducible in the utility boilers. For this, the burner and OFA operations should be optimized with increasing the biomass blending proportion because the changing co-fired fuels leads to the variation of fuel-rich intensity or O2 availability in the primary combustion zone. Thus, the movements of burner tilt and yaw angles are very important to control the coal particle residence time in both primary combustion zone and burnout zone.
4. Conclusions Co-firing and air-staging were applied in an attempt to reduce NOx emissions in a pulverized coal fired furnace. Influences of air staging level, fuel types, and co-firing ratio of woody biomass on NOx emission and burnout characteristics were evaluated. The main results can be summarized as follows: (1) For single coals, sub-bituminous coal was more favorable to reduce NOx emissions due to low fuel-N composition. Specifically, NOx reductions for bituminous and sub-bituminous coal were 22% and 41%, respectively, at a primary zone air ratio of 0.9 and an air staging level 3. This tendency was more dominant with increasing the air staging levels. (2) For the co-firing, given that biomass has a high volatile and low carbon content, it could be successfully applied to the air-staged combustion. The NOx reduction for the blends of 30% was 33% for WA30, and 43% for WB30. In addition, the carbon burnout and flame temperature increased by cofiring woody biomass. The improvement of carbon burnout performance was 0.6% for WA30, and 1.6% for WB30. (3) Hybrid NOx reduction technology combining air staging and low-rank fuel blending has shown the potential to reduce exhaust gas emissions and enhance combustion performance.
Acknowledgements This work was supported by the Human Resources Development program (No. 20144010200780) of the Korea Institute of Energy Technology Evaluation and Planning (KETEP) grant funded by the Korea Government Ministry of Trade, Industry and Energy.
References [1] J. Shin, J. Woo, S. Huh, J. Lee, G. Jeong, Analyzing public preferences and increasing acceptability for the renewable portfolio standard in Korea, Energy Econ. 42 (2014) 17–26. [2] L. Dong, S. Gao, W. Song, J. Li, G. Wu, NO reduction in decoupling combustion of biomass and biomass-coal blend, Energy Fuels 102 (2009) 224–228. [3] W. Fan, Z. Lin, J. Kuang, Y. Li, Impact of air staging along furnace height on NOx emissions from pulverized coal combustion, Fuel Process. Technol. 91 (2010) 625–634. [4] D. Pudasainee, J.H. Kim, Y.C. Seo, Mercury emission trend influenced by air pollutants regulation for coal-fired power plants in Korea, At. Environ. 43 (2009) 6254–6259. [5] L. Sloss, Legislation, Standards and Methods for Mercury Emissions Control, CCC/195, IEA Clean Coal Centre, London, UK, 2012 (Apr). [6] J. Shen, J. Liu, H. Zhang, X. Jiang, NOx emission characteristics of superfine pulverized anthracite coal in air-staged combustion, Energy Convers. Manage. 74 (2013) 454–461. [7] Z. Li, Y. Liu, Z. Chen, Q. Zhu, J. Jia, J. Li, Z. Wang, Y. Qin, Effect of the air temperature on combustion characteristics and NOx emissions from a 0.5 MW pulverized coal-fired furnace with deep air staging, Energy Fuels 26 (2012) 2068–2074. [8] W. Bai, H. Li, L. Deng, H. Liu, D. Che, Air-staged combustion characteristics of pulverized coal under high temperature and strong reducing atmosphere conditions, Energy Fuels 28 (2014) 1820–1828. [9] S. Niksa, G. Liu, L. Felix, P.V. Bush, D.M. Boylan, Predicting NOx emissions from biomass cofiring, Prepr. Pap.-Am. Chem. Soc. Div. Fuel Chem. 48 (2) (2003) 643–648. [10] W.L. VanDeKamp, D.J. Morgan, The co-firing of pulverized bituminous coals with straw, waste paper and municipal sewage sludge, Combust. Sci. Technol. 121 (1996) 317–332. [11] H. Spliethoff, K.R.G. Hein, Effect of co-combustion of biomass on emissions in pulverized fuel furnaces, Fuel Process. Technol. 54 (1998) 189–205. [12] A. Gani, K. Morishita, K. Nishikawa, I. Naruse, Characteristics of co-combustion of low-rank coal with biomass, Energy Fuels 19 (2005) 1652–1659. [13] A. Kazagic, I. Smajevic, Experimental investigation of ash behavior and emissions during combustion of Bosnian coal and biomass, Energy 32 (2007) 2006–2016. [14] S. Li, T. Xu, P. Sun, Q. Zhou, H. Tan, S. Hui, NOx and SOx emissions of a high sulfur self-retention coal during air-staged combustion, Fuel 87 (2008) 723–731. [15] A. Kazagic, I. Smajevic, Synergy effects of co-firing wooden biomass with Bosnian coal, Energy 34 (2009) 699–707. [16] J.P. Smart, R. Patel, G.S. Riley, Oxy-fuel combustion of coal and biomass, the effect on radiative and convective heat transfer and burnout, Combust. Flame 157 (2010) 2230–2240. [17] J. Bai, C. Yu, L. Li, P. Wu, Z. Luo, M. Ni, Experimental study on the NO and N2O formation characteristics during biomass combustion, Energy Fuels 27 (2013) 515–522. [18] J. Wang, W. Fan, Y. Li, M. Xiao, K. Wang, P. Ren, The effect of air staged combustion on NOx emissions in dried lignite combustion, Energy 37 (2012) 725–736. [19] A. Ribeirete, M. Costa, Impact of the air staging on the performance of a pulverized coal fired furnace, Proc. Combust. Inst. 32 (2009) 2667–2673. [20] S. Munir, W. Nimmo, B.M. Gibbs, The effect of air staged, co-combustion of pulverized coal and biomass blends on NOx emissions and combustion efficiency, Fuel 90 (2011) 126–135. [21] S.S. Daood, M.T. Javed, B.M. Gibbs, W. Nimmo, NOx control in coal combustion by combining biomass co-firing, oxygen enrichment and SNCR, Fuel 105 (2013) 283–292. [22] M. Ikeda, H. Makino, H. Morinaga, K. Higashiyama, Y. Kozai, Emission characteristics of NOx and unburned carbon in fly ash during combustion of blends of bituminous/sub-bituminous coals, Fuel 82 (2003) 1851–1857. [23] S.A. Skeen, B.M. Kumfer, R.L. Axelbaum, Nitric oxide emissions during coal and coal/biomass combustion under air-fired and oxy-fuel conditions, Energy Fuels 24 (2010) 4144–4152. [24] L.I. Darvell, L. Ma, J.M. Jones, M. Pourkashanian, A. Williams, Some aspects of modeling NOx formation arising from the combustion of 100% wood in a pulverized fuel furnace, Combust. Sci. Technol. 186 (2014) 672–683. [25] P. Glarborg, A.D. Jensen, J.E. Johnsson, Fuel nitrogen conversion in solid fuel fired systems, Prog. Energy Combust. Sci. 29 (2003) 89–113. [26] T. Wu, M. Gong, E. Lester, P. Hall, Characteristics and synergistic effects of cofiring of coal and carbonaceous wastes, Fuel 104 (2013) 194–200. [27] C. Moon, Y. Sung, S. Eom, G. Choi, NOx emissions and burnout characteristics of bituminous coal, lignite, and their blends in a pulverized coal-fired furnace, Exp. Therm. Fluid Sci. 62 (2015) 99–108. [28] Y. Sung, G. Choi, Effectiveness between swirl intensity and air staging on NOx emissions and burnout characteristics in a pulverized coal fired furnace, Fuel Process. Technol. 139 (2015) 15–24. [29] A. Ribeirete, M. Costa, Detailed measurements in a pulverized-coal-fired largescale laboratory furnace with air staging, Fuel 88 (2009) 40–45. [30] A. Ribeirete, M. Costa, Co-combustion of biomass in a natural gas-fired furnace, Combust. Sci. Technol. 175 (2003) 1953–1977.
Y. Sung et al. / Experimental Thermal and Fluid Science 71 (2016) 114–125 [31] Y.M. Sung, C.E. Moon, J.R. Kim, S.C. Kim, T.H. Kim, S.I. Seo, G.M. Choi, D.J. Kim, Influence of pulverized coal properties on heat release region in turbulent jet pulverized coal flames, Exp. Therm. Fluid Sci. 35 (2011) 694–699. [32] C. Moon, Y. Sung, S. Ahn, T. Kim, G. Choi, D. Kim, Thermochemical and combustion behaviors of coals of different ranks and their blends for pulverized-coal combustion, Appl. Therm. Eng. 54 (2013) 111–119. [33] S. Xiu, Z. Li, B. Li, W. Yi, X. Bai, Devolatilization characteristics of biomass at flash heating rate, Fuel 85 (2006) 664–670. [34] S. Badzioch, P.G.W. Hawksley, Kinetics of thermal decomposition of pulverized coal particles, Ind. Eng. Chem. Process Des. Dev. 9 (4) (1970) 521–530.
125
[35] T.R. Ballantyne, P.J. Ashman, P.J. Mullinger, A new method for determining the conversion of low-ash coals using synthetic ash as a tracer, Fuel 84 (2005) 1980–1985. [36] C. Casaca, M. Costa, Detailed measurements in a laboratory furnace with reburing, Fuel 90 (2011) 1090–1100. [37] G. Wang, R.B. Silva, J.L.T. Aevedo, S. Martins-Dias, M. Costa, Evaluation of the combustion behaviour and ash characteristics of biomass waste derived fuels, pine and coal in a drop tube furnace, Fuel 117 (2014) 809–824. [38] A.K. Gupta, D.G. Lilley, N. Syred, Swirl Flows, Abacus, Kent, 1984. [39] T.H. Ye, J. Azevedo, M. Costa, V. Semiao, Co-combustion of pulverized coal, pine shells, and textile wastes in a propane-fired furnace, Combust. Sci. Technol. 176 (2004) 2071–2104.