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