close

Вход

Забыли?

вход по аккаунту

?

Effect of Char Properties on N2O Formation and Reduction during Circulating Fluidized Bed Coal Combustion.

код для вставкиСкачать
Dev. Chem Eng. Mineral Process., 9(3/4),pp.343-3SS. 2001.
Effect of Char Properties on NZO Formation
and Reduction during Circulating Fluidized
Bed Coal Combustion
Hao Liu, Toshinori Kojima", Bo Feng, Dechang Liu
and JidongLu
National Laboratory of Coal Combustion, Huazhong University of
Science and Technology, Wuhan, Hubei, 430074, P. R. China
*Department of Industrial Chemistry, Seikei University, 3-3-1 Kichijoji
Kita-machi, Musashino-shi, Tokyo 180-8633, Japan
The mechanism and effect of char on N20 formation and reduction during circulating
fluidized bed coal combustion were examined in a bench-scale circulating fluidized
bed It was found that char derivedfiom coal with high volatile matter content has a
higher conversion ratiofrom char-N into N 2 0 during char combustion. Furthermore,
char with greater pore volume and specific suqace area not only shows a higher
conversion ratiofiom char-N into N20, but also has stronger N20reduction ability.
Big pores play the main part in N20 diffusion inside char particles whereas small
pores play the main part of chemical reaction. Molecular d i m i o n and Knudsen
d i f i i o n are both important for N20 d i m i o n in char. The reactivity of char
concerning N20 formation and reduction is closely related to its pore characteristics,
which is an important factor to account for the diverse N 2 0 emissions
porn
combustion of various coals in a circulatina fluidized bed.
Introduction
Fluidized bed combustion technology has gained extensive application owing to its
excellent fuel flexibility, high combustion efficiency and environmentally favorable
* Authorfor correspondence.
343
H a 0 Liu, Toshinori Kojima, Bo Feng, et al.
performance. It has been recognized as a clean coal combustion technology. However,
it has been discovered that the nitrous oxide (N20)emission from fluidized-bed coal
combustion is much higher than that from pulverized coal combustion. N2O emission
from circulating fluidized-bed coal combustion is even higher than that fkom bubbling
fluidized-bed coal combustion.
N20 formation in bubbling fluidized-bed coal combustion has been widely studied.
In recent years, many investigations of N 2 0 formation in circulating fluidized-bed
coal combustion have also been conducted. On the other hand, char plays an
important role in N20 formation and reduction in circulating fluidized-bed coal
combustion owing to the high particle concentration, particularly the much higher
particle concentration in the upper part of a circulating fluidized-bed combustor than
in the freeboard of a bubbling fluidized-bed combustor. The fact that N 2 0 formed
during combustion can be heterogeneously reduced by carbonaceous residues
produced in-situ is known [1-41. De Soete [4] has studied the conversion of char-N in
a fixed bed reactor in the temperature range of 750-1050K.He found that the fraction
of char-nitrogen transformed into NO and N20 during combustion is roughly
proportional to the degree of carbon burnout. He proposed that the mechanism of N20
formation from char combustion is the reaction of (-CN) and (-CNO) sites. Mochizuki
et al. [ 5 ] measured N20 concentrations fiom the combustion of three types of char at
1073K. They detected no N 2 0 in the absence of either NO or O2 as reactant. The
conversion ratio of char-N to N 2 0 was 1-9 %, dependent on coal type and reaction
conditions. They suggested that the pathway of NzO formation is the reaction of NO
with char-N. Suuki et al. [6] studied the effect of devolatilization temperature,
reactor type and combustion temperature on N 2 0 and NO formation from char
combustion. They found that the volatile matter in char is very important for hel-N
conversion into N20. N20 decreased when char was burned in a reactor with a shorter
residence time compared to a reactor with a longer gas residence time. They argued
the possibility of N20 formation from the heterogeneous reaction of NO and char.
Tullin et al. [7] also studied the mechanism of N20 formation from char combustion
in a small-scale fluidized bed reactor. They proposed that the principal path for N20
formation is fiom the reaction of NO with char nitrogen in the presence of oxygen.
344
Effect of Char Properties on N,O during CFB Coal Combustion
The importance of oxygen is in consuming the carbon to provide a mechanism for the
release of nitrogen which is h l y bound in coal mostly in heterocyclic structures.
Moreover, Jones et al. [8] studied NO, and N20 release during coal char combustion
using carbon-13 materials as models and concluded that char having a high
heteroatom content gave N2 as the major product rather than NzO. The relative
amounts of N2, NtO and NO formed during combustion appear to have a complex
dependence upon temperature of combustion, rate of combustion, oxygen content of
the char, nitrogen content and functionality in the char. Hayhurst et al. [9] studied
NO, and N20 formation in a fluidized bed combustor during the burning of coal
volatiles and char. They found that coals of higher rank gave more N20 and also more
NO, ffom both stages of combustion, i.e. the burning of the volatiles and also of char.
On the other hand, lower rank coals give a greater fraction of the NO, and N20during
devolatilization, rather than the subsequent stage of char combustion. Matos et al. [lo]
developed a method of calculating internal surface areas of particles from kinetic rate
data and obtained the rate constants for the NO/char reaction. Krammer and Sarofim
El13 examined reaction of char nitrogen during fluidized bed coal combustion and
found that NzO was formed in amounts that increased with increasing NO
concentration, showing the importance of NO for N20 formation. Teng et al. [12]
studied the global kinetics of carbon/N20 reaction with a TGA and argued that the
reaction order with respect to N20 partial pressure is not constant, which was
considered to be due to the influence of pore resistance variation with the carbon
burn-off level. They argued that since the global activation energy for N20-carbon
reaction varies with the bum-off level of the carbon, it is suspected that the diffusion
of N20 through the porous structure of the carbon may play an important role in this
reaction. Rodriguez-Mirasol et al. [13] studied N20-char reaction in a fixed bed and
obtained the reaction kinetics. They also concluded that the kinetic parameters for
N 2 0 decomposition on a char surface depended on char structure. Furthermore, in our
previous work, we studied the mechanisms of N20 formation ffom char combustion
in a bubbling fluidized-bed and nitrogen oxide emission from a circulating fluidized
bed combustor, with possible pathways of N20 formation from char combustion in a
proposed bubbling fluidized bed [14- 151.
345
Hao Liu, Toshinori Kojim, Bo Feng, et al.
Despite the above-mentioned research, studies concerning the effect of char
property on N20 formation and reduction are limited and there are still many
unknowns to be clarified. Particularly in circulating fluidized bed coal combustion,
such as the effect of microscopic structure of char on its NzO reduction ability, the
influence of property of primary coal on the characteristics of char as well as the
reaction mechanism inside the char particles and so on. Besides that, research under
conditions close to that of practical circulating fluidized bed coal combustion is also
needed because experimental conditions, such as heating rate, etc., may influence the
results.
The main objective of this paper is to clarify, through experiments under
conditions close to a practical circulating fluidized bed, the mechanism and effect of
char property on N20 formation and reduction during circulating fluidized bed coal
combustion. In particular the relationship with pore characteristics of char, in order to
understand the mechanism of N2O formation from coal combustion in a circulating
fluidized bed.
Experimental Details
The experiments were conducted in a bench-scale circulating fluidized-bed as shown
in Figure 1. The reactor was made of a quartz tube with an inner diameter of 0.02 m
and height of 0.6 m. The reactor was heated by an electric furnace and the
temperature was measured with a thermocouple. To avoid the potential catalytic
effect of the thermocouple on the reaction system, the thermocouple was located
outside the tube near the wall. The measurements showed that the temperature
difference between the inside and outside was less than 10K. The reactant gases were
well mixed through a mixer and entered the reactor from the bottom. The composition
of the reactant gases was (at the inlet): 21% O2for N20 formation experiments and
1020 ppm N 2 0 for NzO reduction experiments, and with argon as a balance in all
experiments. A cyclone was used to separate solids from flue gas and the separated
particles were fed back to the reactor through a L-valve. Silicon sand, with size of
1.8-2.3~10~
mywas used as bed material. Char particles of the same size were fed
into the reactor with a very small fluidized-bed feeder. The total gas flow rate was
346
Efect of Char Properties OR N,O during CFB Coal Cornburtion
6.3x 10” Nm‘/s. In all experiments, argon was set as a balance of gas concentration.
The flue gases were cooled and moisture and particulates removed. Then the
concentrations of NO,, 02,
CO and C 0 2 were measured on-line. NO, was analyzed
with a chemiluminescent analyzer, C 0 2 and CO with an NDIR analyzer, and O2 with
a paramagnetic analyzer. The concentration of N20 in the exhausted gas was
measured through a gas chromatograph (GC) equipped with an ECD detector. A gas
sample was obtained and stored in a glass container for measurements after passing
through an ice bath and an NaOH solution. The concentration of N 2 0 in each gas
sample was measured three times, and the average value was taken. Experiments
showed that the differences among the measurements were within 10 %. The coal
chars used in the experiments were prepared through devolatilizing the coals in a
nitrogen stream at a temperature of 1173K for 15 minuets. Table 1 records their
properties.
Table I . Ultimate andproximate analyses of coals and chars.
Ultimate analyses
Fuel
Coal A
CharA
CoalB
CharB
Coal C
CharC
Coal D
CharD
Coal E
CharE
Proximate analyses
(daf, wt %)
C
79.5
95.4
83.7
90.0
89.4
91.6
88.9
91.7
89.8
91.9
H
4.9
1.8
5.4
2.1
4.1
1.8
4.2
2.2
4.7
2.8
(wt %)
N
1.0
0.7
1.3
1.5
1.2
1.3
1.4
1.1
1.1
1.3
FC
62.9
VM
30.1
47.5
16.3
31.3
73.7
13.3
12.2
71.2
10.9
16.9
71.5
17.5
10.5
Ash
4.0
In order to gain a better insight into the mechanism of N 2 0 formation and
reduction during char combustion in a circulating fluidized bed, pore characterization
was conducted with gas adsorption (N2 at 77K) analyses. The measurement
instrument was made by Micrometrics, Co. Ltd. and N2 adsorption was conducted
under various pressures. The pore characteristics were obtained from the adsorption
curve of N2 and calculation by the BET method.
347
H a o Liu, Toshinori Kojima, Bo Feng, er al.
Experimental Results
NzO Formation
First, we studied N20 formation from char combustion. Coal char was bumed in the
reactor at 973K and N2O formation from fuel-N in coal char was investigated by
measuring N20 concentration at the outlet of the reactor. Inlet 0 2 concentration was
21% and oxygen-fuel stoichiometric ratio was 1.2. In order to compare the N20
formation characteristics of these chars, N20 concentrations were converted into
conversion ratio from char-N into N20, because the N contents of the five chars were
different. In order to reveal the effect of char property on N20 formation from char
combustion, pore structure characterization of the char was conducted with gas
adsorption (N2 at 77K)analyses. N2 adsorption was conducted under various pressures.
The pore characteristics were obtained from the adsorption curve of N2 and
calculation by BET method. Table 2 summarizes the specific surface area and pore
volume data of various chars. Figure 2 shows the results of conversion ratio from
char-N into N20 versus specific surface area of the char. It can be seen eom Figure 2
that char combustion also produces a significant amount of N20 and the conversion
ratio from char-N into N 2 0 increases with the specific surface area of char. The
relationship between conversion ratio from char-N into NzO and pore volume of char
is shown in Figure 3. Obviously, conversion ratio from char-N into N 2 0 also increases
with pore volume.
Table 2. Microscopic characteristics of char.
Type of char
CharA CharB CharC CharD CharE
Specific surface area (lo"m'kg)-. 10.94 9.29
1.61
4.94
2.04
Pore volume (10" rnjkg)
13.19 13.29 6.19
6.99
5.52
348
Effect of Char Properties on N,O during CFB Coal Combustion
I-
I
AT
(for N,O)
Figure I . Schematic of the bench-scale circulatingjluidized-bed reactor.
35L
'
'
'
'
I
'
'
'
'
I
30 25
'
'
,
'
4
a
-
10 -
20
15
a
'a
5-
u
0
0
5
10
15
Specific surface area (lo3rn2/kg)
Figure 2. Conversion ratio from char-N into N 2 0 versus speclfic surface area of
char, inlet O2 concentration = 21 % and oxygen-fuel stoichiometric ratio = 1.2.
349
H a 0 Liu, Toshinon Kojim, Bo Feng,et aL
35c
-
I
30 -
25
-
20
-
0
0
*'*
1510 -
-
5-
-
0
5
10
15
20
pore volume (1o m3/kg)
Figure 3. Conversion ratio from char-N into N 2 0 versus pore volume of char, inlet
0 2
concentration = 21 % and oxygen-&el stoichiometric ratio = 1.2.
40
6o
i
I
01
950
I
I
.
I
J
1000 1050 1100 1150 1200 1250
I
1
I
.
1
1
.
Figure 4. N 2 0 reduction ratio by char, inlet N 2 0 = 1020ppm, argon as balance.
350
Effect of Char Properties on N 2 0 during CFB Coal Combustion
Reduction of N20 by Char
N2O reduction by char was examined with three types of char prepared fiom three
coals. An inlet N 2 0 (1 020 ppm) was fed into the reactor fiom the bottom, with Ar as
balance, so as to derive the pure effect of N 2 0 reduction by char. Experiments were
conducted at various temperatures and the results are shown in Figure 4. Here N 2 0
reduction ratio was defined as the ratio of the reduced N20 amount to the inlet N 2 0
amount. It was revealed that different chars have different N 2 0 reduction ability, but
the N 2 0 reduction ratio for the three chars all increase with temperature. The different
ability of N 2 0 reduction demonstrated by different chars suggested that N 2 0 reduction
ability is dependent on a property of the char itself. The mechanism of N 2 0 reduction
by char is probably the effect of active sites, such as (-C),(-CO), etc.. The possible
heterogeneous reactions [4]include:
NtO + (-C)+ N2 + (-CO)
(1)
NzO + (-CO)+ N2 + (-COz)
(2)
Rodriguez et al. [ 131 also found diverse kinetics of N 2 0 reduction by different chars.
In this work, char A demonstrated the strongest N20 reduction ability. Char A was
made fiom a coal of high volatile-matter content, which may account for its strong
ability for N20 reduction. Further research is needed to clarify the mechanism.
Effect of Microscopic Characteristics of Char on N20 Reduction
In order to gain a better insight into the mechanism of N 2 0 reduction by char, the pore
structure characterization of char was conducted with gas adsorption (N2 at 77K)
analyses. The pore diameter distribution characteristics were obtained from the
adsorption curve of N2 and calculation with BET method. Figure 5 shows the specific
surface area distribution versus pore diameter for various chars. Figure 6 shows the
pore volume distribution versus pore diameter for various chars.
351
Ha0 Liu, Toshinori Kojimu, Bo Feng, et al.
---Char
A
W h a rB
-C-CharC
-+Char
--
D
-
%
D
o
~ ~ ' " ' " ~ " " ~ " ' ~ ' ~ " ' " ' ' ' ' '
0
20
40
60 80 100 120 140
Pore diameter (1O-99m)
Figure 5. Cumulativespecific surface area versus pore diameterfor various chars.
Pore diameter (lo-' m)
Figure 6. Cumulativepore volume versus pore diameterfor various chars.
From Tables 1 and 2, and Figures 2 and 3, it can be seen that those chars made
from coals with high volatile-matter content have greater pore volume and specific
surface area, and they also demonstrated a higher conversion ratio from char-N into
N20. Figure 4 showed that those chars with greater pore volume and specific surface
352
Effect of Char Propenies on N 2 0 during CFB Coal Combustion
area also have stronger NzO reduction ability. Therefore, we may attribute the
different N20 formation and reduction characteristics of various chars to the
difference in their internal pore structures developed by devolatilization of different
coals.
We also investigated the diffision regime of NzO inside the pores of char particles,
to gain a better understanding of the effect of char property on N 2 0 formation and
reduction. In general, the diffusion regime inside the pores of a particle can be
estimated through a dimensionless number, i.e. Knudsen number:
K , =Aid
(3)
where d is pore diameter (m); and ;1is the mean free path of a molecule (m) where:
where k = c,,lcv, ratio of specific heat capacities; v = kinematic viscosity (m’/s); a =
speed of sound (ds).
Table 3. Meanfree path of N20 molecule and critical pore diametersfor molecular
dimion and Knudren difision.
Temperature (K)
Mean fiee path of N 2 0 molecule (1 0” m)
Critical pore diameter for Knudsen diffusion (lo-’ m)
Critical pore diameter for molecular diffusion
m)
973
349.0
34.9
3490.0
1073
366.0
36.6
3660.0
1173
383.0
38.3
3830.0
Table 3 lists the mean free path of an N20 molecule and critical pore diameters for
molecular diffusion and Knudsen difhsion inside a particle. By comparing the
microscopic characteristics shown in Figure 6 and Table 3, it can be seen that the
diffusion regime of N 2 0 in char belongs to a transition regime, i.e. molecular
difision and Knudsen diffusion coexist and both are important. In addition, Figure 6
reveals that pores with a diameter greater than 1x 10’ m account for over 60% of the
353
Ha0 Liu, Toshinori Kojima, Bo Feng, et al.
total pore volume, which means it is the big pores that play the main part in diffusion,
because diffusion is directly related to pore volume. Alternatively, Figure 5 reveals
that pores with a diameter smaller than 1x 1Om*m account for over 75% of the total
pore surface area, which suggests it is the small pores that play the main part in
chemical reaction, because reaction is directly related to the surface area of pores.
Conclusions
The mechanism and effect of char on N20 formation and reduction during circulating
fluidized bed coal combustion were examined in a bench-scale circulating fluidizedbed. The following conclusions were reached for circulating fluidized bed coal
combustion:
1. Char derived from coal with high volatile-matter content has a higher conversion
ratio from char-N into N 2 0 during char combustion, owing to the greater pore
volume and specific surface area of char developed during devolatilization of coal
with high volatile-matter content.
2. Char with greater pore volume and specific surface area has not only a higher
conversion ratio of char-N into N20, but also stronger N20 reduction ability.
.
I
3.
Big pores play the main part in N 2 0 diffusion inside char particles, whereas small
pores play the main role in chemical reaction. Molecular diffusion and Knudsen
diffusion are both important for N20diffusion in char particles.
4. This work revealed that the reactivity of char concerning N20 formation and
reduction is closely related to its pore characteristics. This is an important factor
to account for the diverse N 2 0 emissions from combustion of various coals in a
circulating fluidized bed.
Acknowledgments
The authors are gratefbl to the National Natural Science Foundation of China for
financial support of this work.
354
Effect of Char Properzies on N,O during CFB Coal Combustion
References
1.
2.
3.
Wojtowics M. A.; Pels, J. R. and Moulijn, J. A. 1993. Combustion of coal as a source of N 2 0
emission. Fuel Process. Technol., 34, 1-71.
Gulyurtlu, I.; Esparteiro, H. and Cabrita, I. 1994. NzO formation during fluidized bed combustion of
chars. Fuel. 73. 1098- 1102.
h a n d , L.-E. &d Lecher, B. 1994. Reduction of NzO in a circulating fluidized-bed combustor. Fuel,
73,1389- 1397.
h - s o e t e , G.G. 1990. Heterogeneous N20 and NO formarion from bound nitrogen atoms during coal
char combustion. Symp. (Int.) Combust., [Proc.], 23 1257-1264.
5. Mochizuki, M; Koike, J. and Horio, M. 1992. The mechanisms of nitrous oxide formation from
fluidized bed char combustion. 5th International Workshop on Nitrous Oxide Emissions, NIRE,
Tsukubq Japan.
6. Suzuki, Y.; Moritomi, H.; Kido, N.; Ikeda, M.; Suzuki, K. and Torikai, K. 1992. Nitrous oxide
formation from char and heterogeneous reactions with char and CaO. 5th International Workshop on
Nitrous Oxide Emissions, NIRE, Tsukuba, Japan.
7. Tullin, C. L.; Sarofim, A. F. and Beer, J. M. Proc. 1993. Formation of NO and NzO from coal
combustion: the relative importance of volatile and char nitrogen. 12th Int. Conf. Fluidized Bed
Combust, ASME 599-609.
8, Jones, J. M. and Thomas, K. M. 1995. Carbon-13 materials as models for NO, and NzO release during
coal char combustion. Carbon, 33, 1129-1139.
9. Hayhunt, A. N. and Lawrence, A. D. 1996. The amounts of NO, and NzO formed in a fluidized bed
combustor during the burning of coal volatiles and also of char. Combustion and Flame, 341-357.
10. Matos, M. A,; Pereira, F. J. M. A. and Ventura. M. P. 1991. Internal area determination of reactive
particles using kinetic measurements. Fuel, 70,3843.
1 I. Krammer, G. F. and Sarofim, A. F. 1994. Reaction of char nitrogen during fluidized bed coal
combustion-influence of nitric oxide and oxygen on nitrous oxide. Combustion and Flame, 97. 1184.
124.
12. Teng, H.; Lin, H. C. and Hsieh, Y. S. 1997. Thermogravimetric studies on the global kinetics of carbon
gasification in nitrous oxide. Ind. Eng. Chem. Res., 36,523-529.
13. Rodriguez-Mirasol, J.; Ooms, A. C.; Pels, J. R.; Kapteijn, F. and Moulijin, J. A. 1994. NO and N20
decomposition over coal char at fluidized-bed combustion conditions. Combust. Flame, 99,499-507.
14. Feng, B.; Liu, H.; Yuan, J. W.; Lin, Z. J. and Liu, D. C. 1996. Mechanisms of N20 formation from
char combustion. Energy and Fuels, 10,203-208.
15. Feng, B.; Liu, H.; Yuan, J. W.; Lin, Z. 1. and Liu, D. C. 1996. Nitrogen oxides emission from a
circulating fluidized bed combustor. Int. J. Energy Research, 20, 1015-1025.
Received: 30 October 2000; Accepted a f tr revision: 14 March 2001.
355
Документ
Категория
Без категории
Просмотров
0
Размер файла
529 Кб
Теги
effect, properties, coal, formation, reduction, fluidized, bed, combustion, chars, circulating, n2o
1/--страниц
Пожаловаться на содержимое документа