fuel-staging and air-staging to reduce nitrogen emission in cfb

Fuel nitrogen is the main cause of nitrogen oxide formation in fluidized bed .... blends. The first blend consisted of 80% bark and 20% coal, and the ...
3 downloads 0 Views 1MB Size
Subscriber access provided by University Library of Southern Denmark

Combustion

FUEL-STAGING AND AIR-STAGING TO REDUCE NITROGEN EMISSION IN CFB COMBUSTION OF BARK AND COAL Heidi Saastamoinen, and Timo J. Leino Energy Fuels, Just Accepted Manuscript • DOI: 10.1021/acs.energyfuels.9b00850 • Publication Date (Web): 16 May 2019 Downloaded from http://pubs.acs.org on May 17, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

1

FUEL-STAGING AND AIR-STAGING TO REDUCE NITROGEN EMISSION

2

IN CFB COMBUSTION OF BARK AND COAL

3

Heidi Saastamoinena,*, Timo Leinoa a

4

VTT Technical Research Centre of Finland, P.O. Box 1603, FI-40101, Jyväskylä, Finland *

5

Tel.: +358 44 568 9860 E-mail address: [email protected]

6 7

Abstract: Nitrogen oxide (NO) and nitrous oxide (N2O) formation in circulating fluidized bed

8

combustion can be controlled by air-staging and fuel-staging. An extensive test campaign was

9

carried out with a pilot scale CFB test rig to observe the possibilities of the methods in spruce

10

bark and bituminous coal combustion as well as in co-combustion. The fuel-staging with liquid

11

petroleum gas (LPG) was done alternately from three locations with three intensities. The air-

12

staging was studied alone as well as during the fuel-staging experiment. The experimental

13

trends for NO and N2O emission formation during the fuel-staging and air-staging are presented

14

in this study. It was observed that air-staging and fuel-staging can have opposing effects on

15

nitrogen oxide emission formation and, thus, when used together, a clear understanding of the

16

fuel behaviour and conditions, as well as NOx chemistry in the combustor, is needed. Under the

17

tested conditions, it was observed that if the air-staging is effective, fuel-staging does not bring

18

further benefits in NO reduction. Instead, LPG feed can increase the emission in the lack of

19

oxygen. However, if it is not possible to carry out air-staging, fuel-staging can be used in

20

generating oxygen-lean reducing zones for NO. N2O concentration was further reduced with

21

LPG also in the tests with effective air-staging.

22

Keywords: CFB, combustion, fuel-staging, air-staging, LPG, coal, bark, biomass

23

24

1

INTRODUCTION

25

NO, NO2, and N2O are the main nitrogen pollutants emitted from combustion [1,2]. Direct formation of NO 2

26

in the fluidized bed combustion is usually insignificant. Instead, NO2 formation is found in the conditions

ACS Paragon Plus Environment

Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 21

27

where rapid cooling takes place [3]. The formed NO is unstable in atmosphere and oxidizes readily to NO2,

28

which causes acidic fallout by reacting with atmospheric moisture. N2O is a greenhouse gas with an

29

increasing atmospheric concentration [4] and global warming potential 265 times that of CO2 [5]. Power

30

plants in operation may have problems meeting the policies [6] implemented to reduce the emissions of

31

nitrogen oxides (NO and NO 2). So far, regulations do not limit N2O emissions.

32 33

The methods for NOx and N2O emission reduction can be divided into primary and secondary techniques, of

34

which the first one is less expensive. The primary techniques are used to reduce emissions already in the

35

furnace by affecting the combustion conditions in order to produce reducing zones. These techniques,

36

unfortunately, usually do not reduce all the harmful nitrogen species. The primary techniques include: (1)

37

low excess air, (2) air-staging, (3) flue gas recirculation, (4) reduced air preheat and (5) fuel-

38

staging/reburning. The secondary techniques are referred to as end-of-pipe techniques, which are used to

39

remove nitrogen oxides from flue gas. These techniques are commonly divided into: (1) selective catalytic

40

reduction (SCR) and (2) selective non-catalytic reduction (SNCR) [20].

41 42

Fuel nitrogen is the main cause of nitrogen oxide formation in fluidized bed combustion. Nitrogen emissions

43

also form from atmospheric nitrogen (prompt-NO and thermal-NO). However, the formation of prompt-NO

44

as result of reactions between N2 and hydrogen radicals from the fuel can be estimated as small [7,8]. The

45

combustion temperature in fluidized bed combustion is far too low for thermal-NO to form (800–1000 °C

46

instead of ≥1300 °C), since thermal dissociation of N2 to N-radicals is needed to start NO formation [9].

47

Lower combustion temperature increases the NO reduction especially when the oxygen concentration is low

48

[10]. Relatively low combustion temperature, however, enhances the formation of N 2O especially in

49

circulating fluidized beds where almost isothermal conditions (typically 850-900 °C with full boiler load)

50

exist in the riser [11].

51 52

Bark with high volatile content releases nitrogen mainly to light gaseous species, promoting the

53

homogeneous mechanism in nitrogen oxides formation [12]. Although, the reaction routes of volatile

54

nitrogen species are complex [11,21], there is conformity in the NO formation. Released nitrogen

ACS Paragon Plus Environment

Page 3 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

55

intermediates (HCN and NH3) react with available oxygen forming NO or in the oxygen lean zone N2;

56

Formed NO can also reduce to N2 [3,14]. Lower O2 content in the combustor generally results in less NOx

57

and N2O [7].

58 59

Gustavsson and Leckner [17,18] state that the N2O formation is highly dependent on the conditions in the

60

furnace. Higher temperature and lower level of excess air decrease emissions. In addition, the solid fuel, fuel

61

characteristics and additives affect the formation of nitrogen emission. Fuels with high content of fixed

62

carbon such as bituminous coal yield relatively high emissions of N2O (100-150 ppm), whereas the N2O

63

emission is lower for lignite and peat and negligible for wood, all with lower content of fixed carbon

64

compared to coal [17,19–21]. The formation is commonly explained by oxidation of char-nitrogen, reaction

65

of NO with char-nitrogen or by catalytic reactions of nitrogen species on the available particle surfaces

66

[22,23]. The high volatile content of fuel also increases the oxygen consumption and produces oxygen-lean

67

zones that decrease the N2O formation and increases its reduction [23]. In a previous study with VTT’s pilot

68

CFB combustor [24], it was found reasonable to assume that the heterogeneous reaction (1) of NO with char

69

nitrogen reduces NO and increases N2O formation when char inventory increases. char  N NO  N 2O

70

(1)

This is in agreement with the literature [25,26] of fluidized bed combustion of coal and char.

71 72

Kinetics of reburning are complex and include various hydrocarbon reaction mechanisms [13,27,28]. During

73

the reburning process, NO can be homogeneously reduced to N2 either with CHi,-radicals (i=0,1,2) and

74

ammonia (NHi, i=0,1,2) radicals (if the reburn fuel contains nitrogen) or with CO. Also, Kilpinen et al. [27]

75

conclude that NO reaction to HCN and with CO and H radicals are effective in reburning with methane. If

76

active fuel-bound carbon sites are available, they can also reduce NO to molecular nitrogen [29]. The main

77

homogeneous NO reduction routes related to reburning in the literature [28–30] can be simplified to:

78

ACS Paragon Plus Environment

Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

NO  

HCNO NCO   HCN     N2 H 2CN NCO  NH

NO  CO  CO2 

Page 4 of 21

(2)

(3)

1 N2 2

79 80

The reburning method has been used for nitrogen emission reduction in pulverized fuel combustion for

81

various fuels [16,17,31–35]. However, studies in fluidized bed combustion conditions are rare. Sirisomboom

82

and Kuprianov [36] studied fuel-staging in biomass-biomass co-combustion and the gas reburning in CFB

83

integrated with gasification has been studied [37]. In this study, the possibilities of the secondary fuel

84

injection to the furnace above the primary zone in circulating fluidized bed combustion and co-combustion of

85

biomass and coal is studied by feeding liquid petroleum gas (LPG) in a pilot-scale combustor.

86

2

EXPERIMENTAL PROCEDURE

87

The effect of air-staging and fuel-staging on nitrogen emission formation in the circulating fluidized bed

88

combustion of coal, bark and mixtures of them was studied with VTT’s 50kW pilot combustor. The

89

schematic diagram of the combustor can be seen in Figure 1. The riser has a cylindrical shape with a

90

diameter of 0.17m and length 8.3m. The walls are constructed with high-temperature and corrosion-resistant

91

steel. Fed air is electrically heated and the furnace zone has been equipped with several electric heating zones

92

for stabilization, and with several ports for solid material sampling.

93 94

Primary air is fed through an air grid. The secondary air is fed 1.3 metres and tertiary air 1 4.7 metres above

95

the grid. Tertiary air 2 is fed just before the cyclone at 8.3 metres above the grid. Primary fuels can be fed

96

through two different lines. Additives can be fed through a separate line. Limestone or other additives were

97

not used during the experiments. The particle size of natural sand used as the bed material was in the range

98

0.1-0.5 mm (mean diameter 0.2 mm). The combustor is equipped with five gas injection lines that can be

99

seen in Figure 1 (LPG1-LPG5). LPG1 is interconnected with the primary air feed in order to mix the

100

additional gas with primary air. LPG2 is located at 3.0, LPG3 at 6.2 and LPG4 at 7.9 metres above the grid.

ACS Paragon Plus Environment

Page 5 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

101

The main flue gas compounds (including O 2, CO2, CO, NO, N2O, NO2, and SO2) were measured with an

102

FTIR gas analyser and with an O2-analyser from 2.0 m above the grid and from the flue gas. The sampling

103

point at 2.0 metres is above the secondary air feed. The temperatures were measured with thermocouples

104

along the CFB riser. All input flow rates are also measured as Nl/min. Laws describing ideal gases are used

105

to correct flow rate according to reactor temperatures. Total flue gas flow rate, which includes combustion

106

gases, is calculated based on measured flue gas mass flow, air feed and O2 content. The residence time of the

107

flue gas is calculated using the flow rates, reactor dimensions and feeding points.

108

109 110

Figure 1. Schematic diagram of VTT’s 50 kW CFB combustor including feeding ports for air and LPG.

111

ACS Paragon Plus Environment

Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 21

112

The combustion tests were conducted with 100% bark, 100% bituminous coal, as well as with two of their

113

blends. The first blend consisted of 80% bark and 20% coal, and the second 27% bark and 73% coal on

114

energy basis (dry fuel). Fuel compositions are shown in

115 116

Table 1. LPG was injected into combustor as a reburn fuel. The used LPG contains ~95 vol-% propane

117

(C3H8) and ~5 vol-% butane (C4H10).

118 119

Table 1. Fuel composition.

Fuel Moisture (wt,%) Ash, dry basis (815 °C) (wt,%) Volatile content (dry basis, wt,%) Heating value, dry basis (kJ/kg) Element analysis, dry basis C-content (wt, %) H-content (wt, %) N-content (wt, %) S-content (wt, %) Cl-content (wt, %)

Finnish spruce bark 12.0 2.6 75.0 18.86

Russian bituminous coal 9.2 16.2 32.8 27.44

50.4 5.9 0.36 0.02 0.012

68.3 4.6 2.22 0.36 0.005

120 121

The total air volume flow (Nl/min) was kept constant, but the feed varied between primary, secondary,

122

tertiary air 1, and tertiary air 2. The air-staging steps can be seen in Table 2. Air-staging was done with and

123

without the LPG feed. The gas injections with 5 kW load from three different locations (LPG2, LPG3, and

124

LPG4) were tested from one location at a time as well as with three different LPG loads (2kW, 4kW and

125

5kW) from the injection point LPG2. The solid fuel feed, 45-50 kW depending on the fuel, was not changed

126

when doing air- or fuel staging to keep the bottom bed conditions comparable.

127 128

Table 2. Air-staging steps. test # 0 1 2

primary air % tot 49 49 49

secondary air % tot 50 43 32

ACS Paragon Plus Environment

tertiary air 1 % tot 0 7 18

tertiary air 2 % tot 1 1 1

Page 7 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

3 4 5 6 7 8

49 49 49 49 49 49

21 16 16 21 32 43

29 34 0 0 0 0

1 1 34 30 19 8

129

130

3

RESULTS AND DISCUSSION

131 132

3.1

133

The further from the grid the combustion air was fed, the longer the flue gas stayed in the combustor due to

134

decreased flue gas velocity. From the initial air feed (49/50/0/1, prim./sec./tert.1/tert.2 vol-% shares

135

respectively) to the staging, in which 33% air was fed into the tertiary air 1 (49/16/33/1), the increase in flue

136

gas residence time in the reactor was 0.5-1 seconds. If part of the combustion air was fed into the tertiary air

137

2 before the cyclone (maximum case 49/16/0/34), the residence time increased up to 1.5 seconds. This means

138

that when air feed to the secondary air is decreased, the flue gas residence time in the oxygen lean zone

139

becomes longer both due to lower oxygen concentration in the combustor and due to decreased flue gas

140

velocity. LPG feed increased the total flow rate less than 0.5 %. Therefore, its effect to residence time can be

141

considered negligible.

Effect of air-staging on the residence time and temperature profile in the combustor

142 143

During the combustion of either 100% bark or mixture containing 80% bark, the increase in air feed

144

increased temperature at the feeding height (Figures 2 and 3). This clearly shows that volatiles burn

145

immediately if oxygen is available. In all tests for bark and mixtures containing bark, a peak in temperature

146

can be seen at the secondary air entry where main part of the volatiles burn. However, during the combustion

147

tests for 100% coal both the secondary and the tertiary air 1 feed cools down the temperature in the

148

combustor locally. The more the air feed was spread, the lower the gradient appeared in temperature at the

149

secondary entry for bark whereas for coal, the decrease in secondary air feed seems to smoothen the

150

temperature differences. It can be observed clearly that combustion at the secondary air entry decreases and

ACS Paragon Plus Environment

Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 21

151

combustion in the upper combustor increases when the share of coal in fuel mixture is increased. When more

152

than 20% of the total air was tertiary air 2, part of the combustion occurred after that level.

153

154 155 156

Figure 2. Temperature profile in the combustor without the gas feed for 100% bark (left) and 100% coal (right).

157 158

Figure 3. Temperature profile in the combustor during the gas feed tests for 100% bark.

159 160

LPG injection increased the mean temperature (25-50 °C) in the combustor since LPG combustion heats up

161

the bed material. It can be seen clearly that LPG burns in oxygen-rich conditions. However, if the combustor

162

is oxygen-lean, combustion occurs closer to or in the cyclone, or in other words due to tertiary air 2. Neither

163

air staging nor LPG feed affect greatly the density profile in the combustor which was close to constant

164

throughout the test set.

ACS Paragon Plus Environment

Page 9 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

165

3.2

166

The blending of fuels affects significantly nitrogen emission formation. The increase of coal share in the fuel

167

decreases the nitrogen conversion to NO and increases its conversion to N2O, as can be seen in Figure 4. If

168

fuel contains mainly or only bark, the nitrogen conversion to N2O is low. For 100% coal, the measured N2O

169

concentration is clearly the highest. In the experiments where 73% share of coal was used, nitrogen

170

conversion to N2O in the stack is in the same range as to NO. When the share of coal increases, the amount

171

of char increases in the combustor and the heterogeneous reaction of NO with char nitrogen reduces NO and

172

increases N2O formation [24–26].

173 174 175

Figure 4. NO and N2O formation with different fuel mixtures (bark/coal) when no LPG is fed and air feed is to primary and secondary air (49/50/0/1).

Nitrogen emission formation during fuel-staging and air-staging

176 177

It was found reasonable to study the test with all bark (100%) and high share of bark (80%) in parallel due to

178

similarities in the results. This is done in chapter 3.2.1. Accordingly, in chapter 3.2.2, also the tests with all

179

coal (100%) and high share of coal (73%) are analysed in parallel.

180 181

3.2.1

Bark as main fuel

182

Carbon monoxide (CO) and char can reduce NO during combustion [27,29,38,39]. In addition, the low

183

excess air causing high CO concentrations reduces NO formation [40]. CO concentration in the stack is low

184

during combustion tests with 100% and 80% bark (under 90 mg/Nm3). The tertiary air reacts with the rest of

ACS Paragon Plus Environment

Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 21

185

the CO if the combustion in the riser is incomplete, and the cyclone is an efficient mixer that aids the

186

combustion of CO. However, the measured CO concentration in the stack is the highest when the NO

187

concentration is the lowest. Considerably high CO concentrations above the secondary air entry are measured

188

when the secondary air feed is small, as can be seen from Figure 5.

189

190 191 192

Figure 5. Mixture containing 80% bark, NO concentration in stack and CO concentration above the secondary air entry versus oxygen concentration after secondary air feed.

193 194

In this study, the largest reduction in NO during the air-staging tests for pure bark (100%) and for the mixture

195

(80% bark, 20% coal) is found when 21% of the total feed is to the secondary air and 49% to primary air and

196

the rest of the air is fed as tertiary airs 1 and 2. Decrease in secondary air feed affects the flue gas velocity in

197

the combustor making the residence time of reacting nitrogen in reducing conditions longer. When the last

198

air feed location was tertiary air 2 instead of tertiary air 1, even more NO is reduced due to extended oxygen-

199

lean zone in the reactor. In the lack of oxygen, nitrogen precursors reduce to atmospheric N2 or NO formed in

200

the dense bed reduces in reaction with CO and/or char-N.

201 202

The N2O formation is insignificant when char inventory is small [41], which was shown in this study as well.

203

Biomass product gas is also found to be an effective N2O reductant in fluidized bed conditions [42]. Lu et al.

ACS Paragon Plus Environment

Page 11 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

204

[43] conducted experiments with natural gas in a circulating fluidized bed combustor. They found, that the

205

gas combustion itself produced a negligible amount of N2O. However, NO formed with the prompt-NO

206

mechanism, which was explained by local hot spots due to uneven burning of gas. In this study, the

207

temperature in the combustor was electrically controlled and such hot spots were not present. Anyhow, the

208

above-mentioned mechanism can affect a commercial-scale combustor.

209 210

The increase in LPG feed reduces NO almost in a linear fashion for bark as can be seen in Figure 6, in which

211

the change in flue gas NO concentration due to LPG feed compared to corresponding air feed without the

212

LPG feed is illustrated. With 5 kW LPG feed to the lowest feeding point (LPG2) above the secondary air

213

entry, 22% less NO forms than without fuel-staging. However, it must be pointed out that there is no

214

reduction with the most reducing air feeds (49/21/29/1). It seems that if the air-staging is effective, fuel-

215

staging is not needed and if used it may increase NO emission. The results are similar with the mixture that

216

contains mainly bark (80%).

217

218 219 220

Figure 6. Bark 100%, the change in flue gas NO concentration due to LPG with three intensities compared to no LPG feed at different air-staging combinations (prim/sec/tert1/tert2).

221 222

In the study by Pels et al. [44] it was observed that during the volatile combustion intermediate NCO, which

223

originates from HCN, reacts to NO at higher temperature whereas lower temperatures favoured N2O

ACS Paragon Plus Environment

Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 21

224

formation. Loeffler et al. [45] also observed that above 800 °C, the conversion to NO increases if radical

225

level in the combustor is high. However, the effect in lower temperatures was opposite, which was explained

226

by reaction (4).

NO  HO2  NO2  OH

(4)

227

Therefore, the temperature increase and high radical level due to LPG feed would expect to lead to higher

228

NO formation during the combustion of bark with high volatile content. However, more NO is reduced when

229

the intensity of LPG feed is increased. Both of the above-mentioned studies were done in laboratory scale

230

equipment in which oxygen concentration can be controlled. In this study, the LPG feed significantly affects

231

the oxygen content above the feeding height. Especially, if part of the combustion air is fed into the tertiary

232

air 2 feed, the lower the location of the LPG feed is in the combustor, the more NO is reduced (Figure 7).

233

Maximum reduction (31%) in NO concentration in the stack is obtained with the lowest air feed (49/32/0/19)

234

and the lowest LPG feed (LPG2). The lower is the increase in reburn fuel in the combustor, the longer is the

235

residence time of nitrogen compounds in oxygen lean zone if tertiary air 1 is not available. Therefore, it is

236

assumed that the decrease in oxygen concentration due to LPG feed plays a greater role in NO reduction than

237

temperature, which, anyhow, accelerates the reactions of the radical pool with NO.

238 239 240

Figure 7. 100% bark, the change in flue gas NO concentration due to LPG feed from three locations compared to no LPG feed at different air-staging combinations (prim/sec/tert1/tert2)

ACS Paragon Plus Environment

Page 13 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

241

Energy & Fuels

3.2.2

Coal as main fuel

242

During bituminous coal combustion, the total nitrogen emission in the stack decreases along with the reduced

243

secondary air feed (see Figure 8). According to an earlier study [24] in which bituminous coal was

244

combusted with VTT’s pilot combustor, NO concentration above dense bed can be correlated with the

245

average bed temperature and oxygen concentration. In this study, the conditions in the dense bed were kept

246

constant to compare the effects of fuel-staging and air-staging. It is assumed that most of the heterogeneous

247

nitrogen reactions on the char surface, such as NO reduction to N2O (1), have already occurred in the dense

248

bed. N2O concentration seems to decrease further in the riser in oxygen-deficient conditions, but NO

249

concentration stays quite stable despite the changes in air feed.

250

251 252 253

Figure 4. 100% coal, NO and N2O concentration in stack and CO concentration above the secondary air entry versus oxygen concentration after secondary air feed.

254

During the coal (100%) combustion tests, LPG feed reduces both NO and N 2O emission, as can be seen in

255

Figure 9. For pure coal, the test matrix included test with all air feed variations only with the LPG feed

256

intensity of 5 kW. NO reduction with LPG is low when tertiary air 1 is not available and the oxygen

257

concentration remains low in the riser. In general, more nitrogen emission is reduced with LPG when there is

258

more oxygen available at the secondary air entry, which is reverse to tests without LPG, in which the

259

decrease in secondary air feed decreases the total nitrogen emission.

ACS Paragon Plus Environment

Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 21

260

261 262 263

Figure 5. 100% Coal. The change in flue gas NO and N2O concentration due to LPG feed compared to no LPG feed at different air-staging combinations (prim/sec/tert1/tert2).

264

The LPG feed reduces NO during the tests for 73% coal except if the secondary air feed is small and part of

265

the combustion air is fed at the tertiary air 2 instead of feeding it at the tertiary air 1 feeding point

266

(49/21/0/30), see Figure 10. Again, it seems that if the air-staging is effective, fuel-staging is not needed.

267

Approximately 34% reduction in NO can be achieved with 5kW LPG feed to the lowest feed point (LPG2)

268

compared to a case in which no LPG is fed. N2O is reduced with LPG in all tests. The reducing effect of LPG

269

is highest in the no-staging test (49/50/0/1) and decreases when the staging is increased. It is also remarkable

270

that LPG reduces more NO and N2O if secondary air feed is high, but if LPG is not available, the reduction

271

increases in the lack of oxygen above the dense bed. Clear conclusions regarding the intensity of LPG feed

272

on NO reduction cannot be made with available measurements.

273 274 275

Figure 10. Mixture containing 73% coal and 27% bark, the change in flue gas NO concentration due to LPG feed with three intensities compared to no LPG feed at different air-staging combinations

ACS Paragon Plus Environment

Page 15 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

276

(prim/sec/tert1/tert2).

277

N2O concentration decreases along with the increase in the intensity of LPG feed, as can be seen in Figure

278

11. High temperature and low air-ratio is found to lower N2O emissions [46,47]. According to the

279

measurements in this study, the increase in LPG feed increases the temperature and lowers the oxygen

280

concentration in the riser. In addition, LPG feed increases N2O reducing radicals in the reactor. Gustavsson

281

and Leckner [17] list the main gas-phase reactions that decompose N2O. They include decomposition via H

282

and OH radicals (5 and 6) and thermal decomposition (7).

N2O  H  N2  OH

(5)

N2O  OH  N2  HO2

(6)

N 2O  N 2  O

(7)

283

They found that the N2O reduction is strongly dependent on temperature and there were no differences

284

between the fuels used for the reduction, meaning that differences between released radicals did not affect

285

the reduction. N2O emission approached zero if the temperature increases above 900 °C. In this study, LPG

286

feed clearly increases temperature throughout the combustor and air feed slightly decreases the temperature.

287 288 289 290 291

Figure 6. Mixture containing 73% coal and 27% bark, the change in flue gas N2O concentration due to LPG feed with three intensities compared to no LPG feed at different air-staging combinations (prim/sec/tert1/test2).

4

CONCLUSIONS

292

The effect of air-staging and fuel-staging on nitrogen emission formation in circulating fluidized bed

293

combustion of coal, bark and their mixtures was studied. LPG was used in fuel-staging. It was observed that

ACS Paragon Plus Environment

Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 21

294

air-staging and fuel-staging can have opposing effects on nitrogen emission formation and, thus, when used

295

together, a clear understanding of the conditions in the combustor is needed. The following conclusions can

296

be made on the study:

297

a) The decrease in secondary air feed decreased oxygen concentration and flue gas velocity in the

298

combustor, which led to lower NO emission due to increased flue gas residence time in the reducing

299

conditions and increased CO concentration in the riser. The largest reduction during the air-staging

300

experiments was obtained when 21% of the total feed was to the secondary air and 49% to primary

301

air and the rest to tertiary air 1 or 2. This applied to all used fuels. If the rest of the combustion air

302

was fed to the tertiary air 2 instead of tertiary air 1, meaning that the flue gas residence time in the

303

reducing zone further increased, the total emission reduction was larger.

304 305

b) The increase in LPG feed reduced NO during bark combustion; a maximum of 31% of NO reduction

306

from flue gas was obtained with a 5kW LPG feed. Such a reduction was not, however, clearly

307

observed with the mixture of 27% bark and 73% coal.

308 309

c) N2O formation was insignificant during bark combustion, but increased with the increase in coal

310

share in the fuel mixture. LPG was found to efficiently reduce N2O when it was formed. The increase

311

in LPG feed increased N2O reduction during the experiments with 27% bark and 73% coal;

312

maximum of 30% N2O reduction from flue gas was obtained with 5kW LPG feed.

313 314

d) If the air-staging was effective, further NO reduction did not occur with fuel-staging. Instead, the

315

LPG feed can increase emissions. However, N2O concentration in the stack was further reduced with

316

LPG, since LPG feed increases the temperature and, thus, speeds up the decomposition of N2O, and

317

increases the N2O reducing radical pool in the combustor.

318 319

e) The decrease in oxygen concentration due to staging with LPG played a bigger role in NO reduction

320

than increase of the temperature due to LPG combustion in pilot CFB combustor, in which the

321

temperature stayed well below Prompt-NO formation temperature.

322

ACS Paragon Plus Environment

Page 17 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

323

f) In the experiments where the fuel-staging reduced the NO in the first place, the lower the LPG feed

324

was located in the combustor, the more NO was reduced. The location of the LPG feed did not

325

clearly affect N2O reduction.

326 327

Unfortunately, the LPG addition did not clearly benefit NO reduction in relation to pure air-staging.

328

However, it is clear that both techniques can have beneficial effects on nitrogen emission reduction. In

329

commercial scale, also lower cost gases such as raw gas from anaerobic digestion, landfill gas or product gas

330

from gasification could be used for fuel-staging. Solid fuels can be used if their combustion behaviour is

331

known so that oxygen-lean NO reducing zone is produced to desired location. To obtain the benefits of the

332

joint use of fuel-staging and air-staging in nitrogen emission reduction during the CFB combustion,

333

validation of sophisticated flow models with nitrogen compound measurements of along the combustor riser

334

would be needed.

335

NOTATION

336

CFB

Circulating fluidized bed

337

LPG

Liquid petroleum gas

338

ACKNOWLEDGEMENTS

339

This work has been done within the project “Biofficiency - Highly efficient biomass CHP plants by handling

340

ash related problems” carried out in European Union’s Horizon 2020 research and innovation programme

341

under grant agreement 727616.

342

REFERENCES

343

[1]

344 345

doi:10.1016/S0082-0784(77)80346-9. [2]

346 347 348

Pohl J, Sarofim A. Devolatilization and oxidation of coal nitrogen. Symp Combust 1977;16:491–501.

Tullin C, Sarofim A, Beer J. Formation of NO and N2O in coal combustion: the relative importance of volatile and char nitrogen. JInstEnergy 1993;66:207.

[3]

Hill SC, Smoot LD. Modeling of nitrogen oxides formation and destruction in combustion systems. Prog Energy Combust Sci 2000;26:417–58. doi:10.1016/S0360-1285(00)00011-3.

ACS Paragon Plus Environment

Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

349

[4]

350 351

Page 18 of 21

IPCC. Climate Change 2013 - The Physical Science Basis. New York, NY, USA: Cambridge University Press; 2013.

[5]

IPCC. Climate Change 2014: Synthesis Report. Geneva, Switzerland: Contribution of Working

352

Groups 1, 2 and 3 to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change

353

[Core Writing Team, R.K. Pachauri and L.A. Mayer (eds.)]; 2014.

354

[6]

European Commission. Directive 2001/80/EC of the European Parliament and of the Council of 23

355

October 2001 on the limitation of emissions of certain pollutants into the air from large combustion

356

plants. Off J Eur Union 2001:1–21. doi:10.1039/ap9842100196.

357

[7]

Hayhurst AN, Lawrence AD. The amounts of NOx and N2O formed in a fluidized bed combustor

358

during the burning of coal volatiles and also of char. Combust Flame 1996;105:341–57.

359

doi:10.1016/0010-2180(95)00215-4.

360

[8]

Fenimore C. Formation of Nitric Oxide in Premixed Hydrocarbon Flames. Pittsburgh, US: 1971.

361

[9]

Zeldovich Y, Sadovikov P, Frank-Kamenetskii D. Oxidation of Nitrogen in Combustion. Moscow-

362 363

Leningrad: Academy of Sciences of USSR, Institute of Chemical Physics; 1947. [10]

Kühnemuth D, Normann F, Andersson K, Johnsson F, Leckner B. Reburning of nitric oxide in oxy-

364

fuel firing-the influence of combustion conditions. Energy and Fuels 2011;25:624–31.

365

doi:10.1021/ef101054t.

366

[11]

367 368

Energy & Fuels 1996;2:970–9. doi:10.1021/ef950253r. [12]

369 370

[13]

[14]

Glarborg P, Jensen AD, Johnsson JE. Fuel nitrogen conversion in solid fuel fired systems. Prog Energy Combust Sci 2003;29:89–113. doi:10.1016/S0360-1285(02)00031-X.

[15]

375 376

Miller J, Bowman C. Mochanism and modelling of nitrogen chemistry in combustion. Prog Energy Combust Sci 1987;15:287–338.

373 374

Abelha P, Gulyurtlu I, Cabrita I. Release of nitrogen precursors from coal and biomass residues in a bubbling fluidized bed. Energy and Fuels 2008;22:363–71. doi:10.1021/ef700430t.

371 372

Johnsson JE. Modeling N 2 O Reduction and Decomposition in a Circulating Fluidized Bed Boiler.

Dagaut P, Glarborg P, Alzueta MU. The oxidation of hydrogen cyanide and related chemistry. Prog Energy Combust Sci 2008;34:1–46. doi:10.1016/j.pecs.2007.02.004.

[16]

Alzueta MU, Glarborg P, Dam-Johansen K. Low temperature interactions between hydrocarbons and

ACS Paragon Plus Environment

Page 19 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

377

nitric oxide: An experimental study. Combust Flame 1997;109:25–36. doi:10.1016/S0010-

378

2180(96)00146-0.

379

[17]

380 381

through Afterburning. Ind Eng Chem Res 1995;34:1419–27. doi:10.1021/ie00043a050. [18]

382 383

Gustavsson L, Leckner B. Abatement of N2O Emissions from Circulating Fluidized Bed Combustion

Gustavsson L. Reduction of the N2O Emission from Fluidized Bed Combustion by Afterburning. Charlmers University of Technology, 1995.

[19]

Chen Z, Lin M, Ignowski J, Kelly B, Linjewile TM, Agarwal PK. Mathematical modeling of fluidized

384

bed combustion. 4: N2O and NOX emissions from the combustion of char. Fuel 2001;80:1259–72.

385

doi:10.1016/S0016-2361(01)00007-2.

386

[20]

387 388

1994;73:1389–97. [21]

389 390

[22]

[23]

[24]

[25]

[26]

[27]

Kilpinen P, Hupa M, Glarborg P. Reburning Chemistry: A Kinetic Modeling Study. Ind Eng Chem Res 1992;31:1477–90. doi:10.1021/ie00006a009.

[28]

403 404

Kilpinen P, Kallio S, Konttinen J, Bariši V. Char-nitrogen oxidation under fluidised bed combustion conditions: Single particle studies. Fuel 2002;81:2349–62. doi:10.1016/S0016-2361(02)00176-X.

401 402

Tullin C, Sarofim A, Beer J. NO and N2O in coal combustion: importance of volatile and char nitrogen. J Inst Energy 1993;66:207–15.

399 400

Tourunen A, Saastamoinen J, Nevalainen H. Experimental trends of NO in circulating fluidized bed combustion. Fuel 2009;88:1333–41. doi:10.1016/j.fuel.2008.12.020.

397 398

Shen BX, Mi T, Liu DC, Feng B, Yao Q, Winter F. N2O emission under fluidized bed combustion condition. Fuel Process Technol 2003;84:13–21. doi:10.1016/S0378-3820(02)00104-2.

395 396

Kilpinen P, Hupa M. Homogeneous N2O chemistry at fluidized bed combustion conditions: A kinetic modeling study. Combust Flame 1991;85:94–104.

393 394

Leckner B. Fluidized bed combustion: Achievements and problems. Symp Combust 1996;26:3231– 41. doi:10.1016/S0082-0784(96)80169-X.

391 392

Åmand L-E, Leckner B. Reduction of N2O in a circulating fluidized-bed combustor. Fuel

Normann F, Andersson K, Johnsson F, Leckner B. Reburning in oxy-fuel combustion: A parametric study of the combustion chemistry. Ind Eng Chem Res 2010;49:9088–94. doi:10.1021/ie101192a.

[29]

Liu H, Hampartsoumian E, Gibbs BM. Evaluation of the optimal fuel characteristics for efficient NO

ACS Paragon Plus Environment

Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

405 406

reduction by coal reburning. Fuel 1997;76:985–93. doi:10.1016/S0016-2361(97)00114-2. [30]

407 408

Magel HC, Schnell U, Hein KRG. 3rd Workshop on Modelling of Chemical Reaction Systems, Heidelberg, 1996 1. 3rd Work Model Chem React Syst Heidelberg, 1996 1 1996:1–10.

[31]

409 410

Smoot LD, Hill SC, Xu H. NOx control through reburning. Prog Energy Combust Sci 1998;24:385– 408. doi:10.1016/S0360-1285(97)00022-1.

[32]

Su S, Xiang J, Sun L, Hu S, Zhang Z, Zhu J. Application of gaseous fuel reburning for controlling

411

nitric

412

doi:10.1016/j.fuproc.2008.10.011.

413

[33]

414 415

[34]

emissions

in

boilers.

Fuel

Process

Technol

2009;90:396–402.

Werle S. Modeling of the reburning process using sewage sludge-derived syngas. Waste Manag

Werle S. Modeling of the reburning process using sewage sludge-derived syngas. Waste Manag 2012;32:753–8. doi:10.1016/j.wasman.2011.10.013.

[35]

418 419

oxide

2012;32:753–8. doi:10.1016/j.wasman.2011.10.013.

416 417

Page 20 of 21

Adams BR, Harding NS. Reburning using biomass for NOx control. Fuel Process Technol 1998;54:249–63. doi:10.1016/S0378-3820(97)00072-6.

[36]

Sirisomboon K, Kuprianov VI. Effects of fuel staging on the NO emission reduction during biomass-

420

biomass co-combustion in a fluidized-bed combustor. Energy and Fuels 2017;31:659–71.

421

doi:10.1021/acs.energyfuels.6b02622.

422

[37]

Hu X, Wang T, Dong Z, Zhang H, Dong C. Research on the gas reburning in a circulating fluidized

423

bed

424

doi:10.3390/en5093167.

425

[38]

426 427

[39]

432

integrated

with

biomass

gasification.

Energies

2012;5:3167–77.

Aarna I, Suuberg EM. Role of carbon monoxide in the NO-carbon reaction. Energy and Fuels

Aarna I, Suuberg EM. A review of the kinetics of the nitric oxide-carbon reaction. Fuel 1997;76:475– 91. doi:10.1016/S0016-2361(96)00212-8.

[40]

430 431

system

1999;13:1145–53. doi:10.1021/ef9900278.

428 429

(CFB)

Lyngfelt A, Amand L, Leckner B. Reversed air staging - a method for reduction of N20 emissions from fluidized bed combustion of coal. Fuel 1998;77:953–9. doi:10.1016/S0016-2361(98)00007-6.

[41]

Goel S, Sarofim A, Kilpinen PIA, Hupa M. Emissions of Nitrogen Oxides From Circulating Fluidized-Bed 1996:3317–24.

ACS Paragon Plus Environment

Page 21 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

433

[42]

434 435

[43]

Lu D, Anthony E. Combustion Characteristics of Natural Gas in a Circulating Fluidized Bed. Jacksonville, Florida, USA: 2003. doi:10.1115/FBC2003-053.

[44]

438 439

Dong C, Hu X, Li Y, Zhang J, Jiang D, Zhang H, et al. Product Gas Combustion in Fluidized Bed for N 2 O Reduction 2009.

436 437

Energy & Fuels

Pels JR, WÓjtowicz MA, Kapteijn F, Moulijn JA. Trade-Off between NOxand N2O in Fluidized-Bed Combustion of Coals. Energy and Fuels 1995;9:743–52. doi:10.1021/ef00053a003.

[45]

Loeffler G, Wartha C, Winter F, Hofbauer H. Study on NO and N2O formation and destruction

440

mechanisms

441

doi:10.1021/ef010228n.

442

[46]

in

a

laboratory-scale

fluidized

bed.

Energy

and

Fuels

2002;16:1024–32.

Lyngfelt A, Åmand L-E, Gustavsson L, Leckner B. Methods for reducing the emission of nitrous

443

oxide from fluidized bed combustion. Energy Convers Manag 1996;37:1297–302. doi:10.1016/0196-

444

8904(95)00336-3.

445

[47]

Li S, Xu M, Jia L, Tan L, Lu Q. Influence of operating parameters on N2O emission in

446

O2/CO2combustion with high oxygen concentration in circulating fluidized bed. Appl Energy

447

2016;173:197–209. doi:10.1016/j.apenergy.2016.02.054.

448

ACS Paragon Plus Environment