Fuel Nitrogen Conversion in Solid Fuel Fired Systems

Progress in Energy and Combustion Science 29 (2003) 89–113
www.elsevier.com/locate/pecs
Fuel nitrogen conversion in solid fuel fired systems

P. Glarborg*, A.D. Jensen, J.E. Johnsson
Department of Chemical Engineering, Technical University of Denmark, Building 229, 2800 Lyngby, Denmark
Received 24 January 2002; accepted 15 October 2002
Abstract
Understanding of the chemical and physical processes that govern formation and destruction of nitrogen oxides (NOx) in
combustion of solid fuels continues to be a challenge. Even though this area has been the subject of extensive research over the
last three decades, there are still unresolved issues that may limit the potential of primary measures for NOx control.
In most solid fuel fired systems oxidation of fuel-bound nitrogen constitutes the dominating source of nitrogen oxides. The
present paper reviews some fundamental aspects of fuel nitrogen conversion in these systems, emphasizing mostly combustion
of coal since most previous work deal with this fuel. However, also results on biomass combustion is discussed. Homogeneous
and heterogeneous pathways in fuel NO formation and destruction are discussed and the effect of fuel characteristics,
devolatilization conditions and combustion mode on the oxidation selectivity towards NO and N2 is evaluated.
Results indicate that even under idealized conditions, such as a laminar pulverized-fuel flame, the governing mechanisms for
fuel nitrogen conversion are not completely understood. Light gases, tar, char and soot may all be important vehicles for fuel-N
conversion, with their relative importance depending on fuel rank and reaction conditions. Oxygen availability and fuel-
nitrogen level are major parameters determining the oxidation selectivity of fuel-N towards NO and N2, but also the ability of
char and soot to reduce NO is potentially important. The impact of fuel/oxidizer mixing pattern on NO formation appears to be
less important in solid-fuel flames than in homogeneous flames.
q 2003 Elsevier Science Ltd. All rights reserved.
Keywords: Combustion; Coal; Biomass; Nitrogen oxides; Emissions
Contents
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
2. NOx formation mechanisms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
3. Solid fuel characterization. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
4. Devolatilization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
4.1. Amount and partitioning of N-volatiles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
5. Volatile oxidation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
5.1. Tar-N conversion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
5.2. Soot-N conversion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
5.3. Conversion of gas-phase cyclic nitrogen species . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
5.4. Conversion of light N-volatiles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
6. Char oxidation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
6.1. The char nitrogen oxidation mechanism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
6.2. Formation of N2O from char-N. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
6.3. Reduction of NO on char . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
6.4. Effect of reaction conditions on NO yield . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
* Corresponding author. Tel.: þ 45-4525-2840; fax: þ 45-4588-2258.

E-mail address: pgl@olivia.kt.dtu.dk (P. Glarborg).
0360-1285/03/$ - see front matter q 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0360-1285(02)00031-X
90 P. Glarborg et al. / Progress in Energy and Combustion Science 29 (2003) 89–113
7. Fuel-N conversion in laminar flames . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104

8. Concluding remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
1. Introduction istics in various systems. A large part of this work is

discussed in reviews dealing with various aspects of the
Emission of oxides of nitrogen from combustion and nitrogen chemistry in combustion, i.e. gas-phase nitrogen
high temperature industrial processes continues to be a chemistry [1], nitrogen bonding in coal [2], char nitrogen
major environmental concern. Nitrogen oxides, collectively oxidation and reduction of NO [3– 5], NOx formation and
termed NOx, are formed either from fixation of N2 in the destruction in pulverized fuel systems [6 – 10], in FBC [11],
combustion air at high temperatures or from oxidation of and in gasification [12], control methods [13 – 16], formation
nitrogen chemically bound in the fuel. Largely all combus- and destruction of N2O [17], and handling of NOx chemistry
tion processes lead to formation of NOx, emitted mostly as in computational fluid dynamics modeling [18]. Further-
nitric oxide (NO) with smaller amounts of nitrogen dioxide more, general reviews on solid fuel combustion, e.g.
(NO2). Nitric oxide is subsequently oxidized to NO2 in the [19– 21], often discuss aspects of pollutant formation.
atmosphere. Nitrous oxide (N2O) emissions can be signifi- The objective of the present work is to provide an
cant in fluidized bed combustion (FBC), but are negligible overview of the current understanding of fuel nitrogen
in most combustion systems. Nitric oxide and nitrogen conversion in solid fuel fired systems. The review mainly
dioxide are acid rain precursors and participate in the deals with coal combustion, since most previous work focus
generation of photochemical smog, while nitrous oxide is a on this fuel, but also recent results on biomass combustion
greenhouse gas. are discussed. We emphasize the effect of parameters that
The concern about NOx emissions and the need to comply conceivably affect the oxidation selectivity towards NO and
with increasingly stringent regulations have motivated a vast N2. These include fuel characteristics (fuel type, amount and
amount of research. Restrictions that used to apply only to functional form of N in fuel), particle size, devolatilization
large combustion systems like power plants are now a conditions and partitioning of N among pyrolysis products,
concern on ever smaller scales and continue to be a driving fuel/oxidizer mixing patterns, and reaction conditions
force for research and development in many industrial high (temperature, stoichiometry, reaction time, etc.). We focus
temperature processes. Furthermore, in the current develop- mostly on fuel-N conversion under idealized conditions,
ment a wide range of solid fuels can be expected to either in laboratory experiments or in ideal reactor
contribute to the energy supply. In addition to coal, different modeling. Heterogeneously catalyzed nitrogen reactions,
biomass fuels and household and industrial wastes become which can be important in systems such as FBC with high
more important. The use of novel fuels and fuel-mixtures solids concentrations (ash, bed material, limestone), are
introduces new challenges in pollution control. discussed elsewhere [11] and are not treated in the present
Solid fuel combustion technologies include pulverized paper. To utilize the knowledge of nitrogen chemistry to
fuel combustion (PFC), combustion on grates (GC), and improve practical combustion systems, models are required
FBC. A significant expertise has been achieved in reducing that can handle the physical and chemical complexities of
nitrogen oxides emissions from pulverized-coal-burning these systems. Computational fluid dynamic models are
equipment, either by primary means (in particular low-NOx often used to analyze and predict the effect of process
burners or over-fire air injection), by reducing agent conditions on NOx emissions. Discussion of these models
injection (selective non-catalytic reduction of NO or are outside the scope of the present work, but results
reburning) or by catalytic flue gas cleaning. Some of these important for the development of submodels describing the
methods may not be viable for fuels like biomass and waste. various phases in the fuel-N conversion are presented.
These fuels are commonly burned on a grate or in a fluidized
bed; technologies that preclude the use of low-NOx burners.
Furthermore, catalytic flue gas cleaning is often inapplicable 2. NOx formation mechanisms
due to catalyst deactivation caused by trace components
such as alkali metals in the flue gas from biomass or waste. In combustion processes nitrogen oxides can be formed
A better understanding of the nitrogen chemistry in solid from fixation of molecular nitrogen in the combustion air or
fuel fired systems is helpful mainly in improving primary from nitrogen chemically bound in the fuel. Formation of
measures for NOx control. Previous efforts have involved NO from N2 occurs primarily by thermal NO formation. This
fundamental research in the detailed mechanisms of mechanism proceeds through the reaction sequence [1,22]
formation and destruction of NO as well as applied studies
on the effect of process parameters on emission character- O þ N2 O NO þ N ðR1Þ
P. Glarborg et al. / Progress in Energy and Combustion Science 29 (2003) 89–113 91
N þ O2 O NO þ O ðR2Þ Table 1
Typical nitrogen content in selected fuels [3,12,20,21,29]
N þ OH O NO þ H ðR3Þ
Fuel N content (wt%)
The first step (R1) has a high activation energy and
formation of thermal NO is most important at temperatures Straw 0.3–1.5
above 1800 K. Other sources of NO from N2 include prompt Other agricultural residues 0.4–3.5
NO [23], initiated by attack of the CH radical on the N2 triple Wood 0.03– 1.0
bond [1,24,25], Peat 0.5–2.5
Coal 0.5–2.5
CH þ N2 O NCN þ H ðR4Þ Paper 0.1–0.2
and the N2O mechanism [26], initiated by RDF 0.8
Tires 0.3
O þ N2 þ M O N2 O þ M ðR5Þ Household waste 0.5–1.0
Plastic waste 0.0
The reactive nitrogen compounds formed (NCN, N2O) may
Sewage sludge 2.5–6.5
subsequently be oxidized to NO or recycled to N2, dependent
on reaction conditions [1].
With a few exceptions, such as plastic waste, all solid
fuels contain some organically bound nitrogen. Formation contain of the order of 1% nitrogen. However, wood, paper
of NO from oxidation of fuel-N is termed fuel-NO. The and plastic waste typically contain much less nitrogen, while
overall reaction sequence for this source of NO is fairly well sewage sludge may have a very high nitrogen content. The
established. When particle combustion begins, the fuel-N is nitrogen content in trees is about 1%, with most of the
distributed between the volatiles and the solid char matrix. nitrogen present in leaves and needles. Wood bark contains
The distribution of nitrogen between char and volatiles, as typically 0.3– 0.5% and wood matter below 0.1%.
well as the volatile nitrogen composition, depends mainly The relative importance of fuel-NO and thermal NO
on the fuel structure and the temperature. For bituminous depends on the nitrogen content of the fuel, on the
coals the volatile-N consists mostly of tarry compounds that combustion temperature and on the fuel/air mixing pattern.
at high temperatures decay rapidly to hydrogen cyanide High temperature pulverized fuel systems like smelting
(HCN) or soot-nitrogen. For low rank coals and biomass, furnaces or cement kilns may yield significant amounts of
light nitrogen species may be released directly from the thermal NO, but this is an exception. In pulverized coal
solid matrix. Hydrogen cyanide is oxidized to NO via
combustion fuel-N conversion to NO typically accounts for
intermediate amines or, depending on stoichiometry and
more than 80% of the NO emission. Fig. 1 shows results
fuel-N concentration, converted to N2 by the recombination
from Pershing and Wendt [30] for the relative yields of fuel-
of NO with another nitrogen containing species. The
NO and thermal NO in pulverized coal combustion. In
remaining char-N and soot-N undergo heterogeneous
selected experiments the combustion air (O2/N2) were
oxidation to NO and N2 or it may at high temperatures
evolve as light components such as HCN. Nitric oxide, once
formed, may be recycled by hydrocarbon radicals to cyanide
or reduced to N2 by surface reactions on char or soot.
At low temperatures, as in fluidized bed reactors, other
gas-phase nitrogen species are detected in significant
amounts. In addition to HCN, also ammonia (NH3) and
cyanuric acid (HNCO) are released during pyrolysis, in
particular for low rank coals and biomass. Under these
conditions also conversion of fuel-nitrogen to N2O may be
significant [17,27]. The most important source of N2O
appears to be oxidation in the gas-phase of HCN [28], but
char-N oxidation may also contribute [4,11].
In solid fuel systems, oxidation of fuel-N is generally the
major source of NO, with some contribution from thermal
NO. Other NO formation mechanisms are usually negligible
in these systems, even though prompt NO could be of
importance in combustion of high-volatile fuels with low
N-content, and NO formation through N2O conceivably
could play a role in pressurized combustion systems. Table 1
shows typical values for fuel-N content for selected fuels. Fig. 1. Sources of NOx emissions in turbulent diffusion pulverized
Solid fuels such as coal, peat, most agricultural residues, etc. coal flames [30].
replaced by a nitrogen free synthetic air mixture (O2/Ar/ functionality in biomass is expected to vary with type of
CO2), securing that fuel-N conversion was the only source biomass, location in the biomass and with age. However,
of NO. The NO formation from volatiles constitutes most of both pyridinic and pyrrolic structures (in porphyrine) and
the fuel-NO, while char-N oxidation is a minor source in amino structures (in proteins) may be important [49].
these unstaged flames. The high levels of NO formed from It has proven difficult to relate the functional form of
fuel-nitrogen inhibits the thermal NO formation since the fuel-N with the HCN/NH3 release ratio or with the NO/N2O
initiating step (R1) in this mechanism may be reversed formation selectivity [40,41,43,50]. Results indicate that the
NO þ N O O þ N2 ðR1bÞ nitrogen functionality is less important for fuel-N oxidation
than the O/N ratio in the fuel [41,51]. As the oxygen content
Due to this interaction between the two NO formation in the fuel increases compared to nitrogen, both the HCN/
mechanisms, temperatures in excess of 2200 K may be NH3 ratio during pyrolysis and the N2O/NO ratio during
required for a significant contribution from thermal NO in FBC decrease. This trend is similar to what has been
coal-dust flames [31]. observed for model compounds, as discussed below.
Combustion of biomass and waste is mostly carried out
on a grate or in a fluidized bed combustor; systems with
comparatively low combustion temperatures. In these
systems the yields of thermal NO can be considered to be 4. Devolatilization
small or negligible.
Devolatilization is the thermal decomposition of the fuel.
The evolution and burning of volatiles involve a range of
3. Solid fuel characterization complex physical – chemical processes that play a crucial
role in determining flame stability, ignition, volatiles
The fuels of primary interest in this review are coal and composition, and pollutant formation. Because of the
biomass. Coal is formed from biomass in a process that importance of devolatilization to flame properties, it has
requires high pressures and very long times. In this process been studied extensively. Most of the work have focused on
oxygen and hydrogen are eliminated and fuels from coal devolatilization, which is the subject of several reviews
anthracite and coal to peat and biomass are ranked according
[52– 55]; less is known about devolatilization of biomass
to their oxygen and hydrogen content [32]. The fuels are
and alternative fuels [20,21,56– 60].
composed of both organic and inorganic constituent parts.
The products of devolatilization are char and volatiles.
The organic part consists of C, H and O with minor fractions
The volatiles consist of tar and light gaseous species. The
of N and S. The elements Al, Ca, Fe, Hg, K, Mg, Na, Si and
amount and composition of volatiles depend mostly on the
Ti are either organically bound (Ca and Na) or present as
final temperature and fuel type, but other parameters such as
mineral inclusions in the form of mixed oxides (Al, Ca, Hg,
particle size, heating rate, and pressure have also some effect
K, Mg, Na, Si and Ti) or pyrite (Fe) [33]. Coal and biomass
[52]. The general characteristics of the devolatilization
also contain trace amounts of the compounds As, B, Be, Cd,
process are fairly well established, at least for coal [55].
Co, Cr, Ga and Pb, which may be associated with both the
organic and the inorganic phase. Prior to the primary devolatilization, low temperature cross-
The speciation and release rate of nitrogen volatiles linking takes place in coals with a high O content, i.e. low
during devolatilization conceivably relate to the functional rank coals, accompanied by evolution of H2O and CO2.
forms of nitrogen in the fuel. Nitrogen is largely During the primary devolatilization, weak aliphatic bonds in
organically bound in the fuel, even though some inorganic the solid fuel matrix are broken, producing fragments which
nitrogen may be present as ammonium ions in high rank evolve as tar, provided the fragment can vaporize at the
coals [2,34]. Spectroscopic investigations with X-ray given temperature. The tar, which is defined as the volatiles
photoelectron spectroscopy [3,35– 44] and X-ray absorp- released that condense at room temperature, consists of
tion near edge structure [45,46] show that nitrogen in coal hundreds to thousands of organic species. The fragments,
is mainly bound in pyrrolic (five-ring) and pyridinic (six- which do not vaporize, repolymerize and stabilize the
ring) functionalities. Also presence of quaternary nitrogen evolving char matrix. After the breaking of weak bonds, the
has been observed [39,40,42,44,47,48], but the nature of functional groups decompose, releasing CO2, light aliphatic
this species is not yet resolved [4]. Pyrrolic, pyridinic and gases, and H2O. The tar release is most pronounced for
quaternary nitrogen typically account for 50 – 80, 20 – 40, bituminous coals, while lower rank coals due to the low
and 0 – 20%, respectively, of the total nitrogen content. temperature cross-linking release less tar but higher
Small amounts (5– 10%) of aromatic amines have also amounts of light gases.
been reported [45,46], but in general the amine content The secondary devolatilization involves decomposition
appears to be small or negligible. The rank dependence of of the char and tar formed during the primary stage. During
the nitrogen functionality in coal appears to be minor, but the secondary stage additional gas such as CH4, CO, and H2
the issue is still open to discussion. The nitrogen as well as light nitrogen species are released as remaining
aliphatic side chains are released and ring condensation

occurs in the solid matrix.
Characteristics of the devolatilization process are fuel
specific. Bituminous coals yield comparatively large
amounts of tar compared to other coals [61]. Low
temperature cross-linking restrains the formation of mol-
ecular fragments in low rank coals and the initial degree of
cross-linking is too high in anthracites for significant
fragmentation to take place.
The secondary pyrolysis of coal tars occurs at tempera-
tures above 1150 K and is strongly temperature and rank
dependent [62,63]. IR spectra and chemical analysis
indicate that the structure of tar is similar to that of the
parent coal [64,65] and the rank dependency of the tar
decomposition can be attributed to this structural similarity.
The secondary pyrolysis of coal tar results in formation of
soot and light gases such as H2, CO, C2H2, and C2H4 [62,66,
67]. In addition, single ring aromatic molecules have been
observed from thermal decomposition of some tars [63].
The strong propensity of the tar evolved from coal to form Fig. 2. Nitrogen concentration in char versus pyrolysis temperature
soot has been observed in a number of studies [66,68– 76]. for selected fuels and various experimental methods. The data are
shown as the ratio of the nitrogen content in the char and in the
The primary soot particles are formed directly from tar and
parent fuel (on a dry and ash free basis). The particle diameters in
subsequent soot growth occurs through addition of hydro-
the experiments with coal were less than 200 mm and the heating
carbons and/or polyaromatic compounds [68,72,76]. rates were in the range of 200 –20 000 K/s at which rate the
Compared to coal, biomass fuels are characterized by influence of heating rate is modest [80,83]. Open symbols denote
lower char yields and higher yields of both tar and gases coal data [40,64,80,84,85]; closed symbols denote data for peat [89]
[56]. The pyrolysis characteristics of wood and agricultural and straw [88].
residues (wheat straw, olive husks, grape residues, rice
husks) appear to be comparable [58]. (lignites) and biomass than for coals of higher rank [79,86].
Tar yields in the range 50 – 80% have been estimated for However, as the temperature increases biomass appears to
small biomass particles and comparatively low temperatures release the nitrogen more readily than most coals. About
[77]. Larger particles yield smaller amounts of tar due to 80% of the nitrogen in biomass (straw, wood) is released
intraparticle conversion of the tar. Tars from biomass are during pyrolysis at temperatures above 850– 900 K [90].
more reactive and decomposes thermally at lower tempera- The limited data on biomass and peat pyrolysis indicate that
tures compared to coal tar. They would be expected to yield the preferential release of nitrogen occurs at lower
smaller amounts of soot due to the high oxygen content. temperatures for these fuels compared to coal, but more
work is needed to confirm this.
4.1. Amount and partitioning of N-volatiles For coal, the nitrogen volatiles evolve comparatively
late in the primary pyrolysis [66,79,80,91,92]. The time
The split of nitrogen in the fuel into volatiles and char delay is fuel specific, increasing with decreasing rank of
during devolatilization, and the type of the volatiles, is the fuel [66,92]. Essentially all the nitrogen (85 – 100%)
potentially important for NOx formation. The fraction of released during the primary coal pyrolysis does so in the form
nitrogen released with volatiles during devolatilization of aromatic compounds in tar and oils [71,80,87]; the fraction
depends on the fuel type, the temperature, and the residence released as light gases is minor or negligible. For bituminous
time. It increases with the oxygen content in the parent fuel, coals, the amount of nitrogen released as condensible
i.e. with decreasing rank [78]. At low temperature or products during pyrolysis appears to be proportional to the
residence time nitrogen is preferentially retained in the char, tar yield and the nitrogen content in the fuel [61,64,92– 95].
while at high temperatures it is depleted [40,64,66,79 –88]. For low rank coals, only minor amounts of nitrogen is
Fig. 2 shows the nitrogen concentration in char versus released with the tar in the primary pyrolysis and the early
temperature for a range of coals, a peat and a biomass, based light gas species also contain little nitrogen [92]. The major
on various experimental techniques [40,64,80,83 –85,88, release of nitrogen volatiles for low rank coals occurs during
89]. The initial enrichment of nitrogen in the coal char the secondary pyrolysis of the char residue.
occurs mainly in the temperature range 600– 1200 K. It can During secondary pyrolysis of char and tar, the
be attributed to the tight bonding of the nitrogen in pyridinic heteroaromatic rings may decompose, releasing the nitrogen
and pyrrolic structures. The preferential retention of mostly as HCN to the gas-phase. For char this happens on
nitrogen at low temperatures is greater for low rank coals time scales that are considerably longer than the primary
devolatilization and requires high temperatures [65,79– 81, released as a primary product of pyrolysis. Both at low and
96]; at temperatures above 2200 K and several minutes of high heating rates NH3 has been detected prior to HCN in
residence time, only carbon is left in the char [79]. Char pyrolysis of low rank coals [92,109]. The source of NH3 is
decomposition is believed to be responsible for most of the presumably direct cleavage of amino groups and amides.
HCN production during primary to early secondary For coal, the ammonia release has been associated with the
pyrolysis [97]. presence of quaternary nitrogen which is present in larger
As the secondary pyrolysis proceeds, tar may become the quantities in low rank coals [92]. The heating rate
major source of nitrogen release to the gas phase. In addition dependence of the NH3 yield in coal pyrolysis indicate
to release as non-condensible gases, tar-nitrogen may be that, in addition to the direct route, NH3 can be produced
incorporated into a carbonaceous soot matrix or it may from HCN in a secondary reaction [65,110]. Also larger
undergo structural rearrangements among the condensible particle sizes appear to favor formation of NH3 compared to
products [62,71,72,98– 100]. For bituminous coals, as much HCN [109]. It has been proposed [78] that HCN
as 25% of the nitrogen released during pyrolysis may be hydrogenation occurs on the char surface as HCN diffuses
incorporated into soot during secondary pyrolysis [72]. For out of a porous particle. At low heating rates there is enough
other fuels the soot-nitrogen is less significant. The nitrogen gas residence time within the char pores to allow for a
content in the soot appears to be proportional to the nitrogen secondary reaction of HCN and coal hydrogen to produce
fraction in the parent coal [76]. The nitrogen is incorporated NH3, while at high heating rates the residence time is
very early and the total fraction of coal nitrogen incorpor- reduced and HCN conversion does not occur. The HCN
ated into soot remains constant even while soot yields conversion on the char surface may also involve a free
dramatically increase with increases in residence time [71, radical mechanism [109,111,112], but more work is needed
72]. The soot growth occurs largely through addition of to identify this source of NH3. Intra-particle conversion of
hydrocarbons and/or polyaromatic compounds with no or HCN to NH3 would be expected to be most important for
little nitrogen content. large particles and for low heating rates.
Pyrolysis at lower temperatures under fluidized bed Presence of calcium, potassium and iron may catalyze
conditions yields a larger proportion of light nitrogen conversion of fuel-N in gas – solid and/or solid phase
containing gases compared to tar, with the amount reactions. Recent studies show that, in addition to NH3
increasing with decreasing rank [97]. Under these con- and HCN, significant amounts of N2 may be formed during
ditions, the high particle loading may conceivably enhance pyrolysis of low rank coals and lignite [113 –117]. Calcium
the tar decomposition through gas – solid or heterogeneously and potassium can promote formation of NH3 and N2 and
catalyzed reactions. suppress HCN [116,118], while iron may promote N2
The light nitrogen species formed during primary and formation at the expense of N in light gases, tar and char
secondary pyrolysis mostly consist of hydrogen cyanide [119,120].
(HCN) and ammonia (NH3). Minor amounts of cyanuric Despite the physical and chemical complexity of
acid (HNCO) have been measured in fluidized bed pyrolysis devolatilization, significant progress has been made in
[97,101], while it has not been detected in entrained flow modeling this process. Several of these models, i.e.
reactor experiments [92]. It should be noted that detection of FLASHCHAIN [121], FG-DVC [78] and CPD [100], have
the nitrogen speciation based on wet chemical methods been modified to include nitrogen release. In these models
should be interpreted with caution. Methods using electro- first-order kinetics are used to describe the rate of release of
des may exhibit cross-sensitivity between HCN and C2H2 nitrogen, assuming that the nitrogen atoms are randomly
[102], and in solution HNCO may be hydrolyzed to NH3 distributed throughout the aromatic clusters of the solid coal
[103]. It is well established, however, that the nitrogen matrix. The differences of the models are discussed further
volatiles partitioning depends on fuel type, pyrolysis elsewhere [100]. Solid state 13C nuclear magnetic resonance,
temperature and heating rate. Devolatilization of bituminous which is used to characterize the aromatic clusters in the solid
coals produces mainly HCN, while more NH3 evolves from matrix, has provided important input to the network
lower rank coals and biomass [41,56,81,92,99,104– 107]. devolatilization models [122] and has recently lead to the
Higher heating rates appear to increase the HCN/NH3 ratio development of a model for nitrogen evolution during coal
for both coal and biomass fuels [56,78,108]. At high heating pyrolysis based on a free-radical mechanism [95].
rates HCN is the main product observed in bituminous coal
pyrolysis [78,108], while pyrolysis of low rank coals, peat
and bark yields significant amounts of NH3 [41,92,106]. At
low heating rates NH3 is generally the dominant product, 5. Volatile oxidation
both for coals and biomass [56,92].
There are strong indications that NH3 is formed both by Despite the detailed knowledge about hydrocarbon
direct release from the solid matrix and from hydrogenation oxidation chemistry developed over the years, the under-
of HCN on the surface of the solid matrix. In pyrolysis of standing of volatiles oxidation in solid fuel combustion
biomass, and to some extent low rank coals, NH3 is mainly remains limited. This can be attributed both to the complexity
of the volatiles composition, involving a range of com- Thermal cracking of tar at temperatures below 1100 K
ponents from hydrogen to heavy organic compounds and tar, yields only small amounts of light gas-phase nitrogen [97,
and to the complex physical and chemical interactions during 109,125]. The initial reactions of the tar involves loss of side
oxidation. Volatiles conversion may occur on the internal or chains of carbon, hydrogen and oxygen [92]. Higher
external particle surface, in the pore system, in the particle temperatures are required for significant release of gaseous
boundary layer, or in the bulk flow, where the oxidation rate nitrogen species due to the thermal stability of the
may be either kinetically or mixing limited. heterocyclic structures that contain the tar-nitrogen. Fast
The thermal decomposition and oxidation of reactive release of nitrogen, mainly as HCN, requires temperatures
nitrogen components has been studied extensively. To above 1400 K; under these conditions opening of the
avoid the complexity of heterogeneous systems, most aromatic rings are facilitated. At lower temperatures in
studies have been conducted for homogeneous systems, fluidized bed pyrolysis (900 – 1300 K), also NH3 and HNCO
often under idealized conditions. A wide range of nitrogen have been detected [125].
compounds has been used to simulate volatile nitrogen High heating rate experiments show that the nitrogen
behavior in flames. It is generally assumed in modeling release from tars of different coal types is very similar [92].
that nitrogen containing volatiles either consist of or are This indicates that the reactivity of the tar-nitrogen
rapidly converted to HCN and/or NH3. However, the functionalities during secondary pyrolysis is largely rank
nitrogen in the volatiles is partitioned between tar, soot independent. However, this similarity may not extend to
and gas-phase components, as described above, and even biomass fuels. Under slow heating rate conditions, differ-
in pyrolysis at high temperatures not all the nitrogen in ences have been reported for nitrogen reactivity of tars from
the volatiles appears in the gas phase as HCN and NH3. coal, peat and bark [56].
The amount of nitrogen containing volatiles transformed The thermal stability of nitrogen species in the tar
into light gas-phase nitrogen species increases with appears to obey the following order: pyrrolic , pyridinic ,
pyrolysis temperature and with decreasing rank of the cyanoaromatic [39,123]. The fact that the reactivities of
fuel. For coal, reported values range from low numbers up pyrrole type structures are greater than those of pyridine
to nearly 60% [81,92]. type structures are in agreement with model compound
studies, as discussed below. A higher pyrolysis temperature
5.1. Tar-N conversion is required to release nitrogen as HCN from pyridine groups
than from pyrrole groups.
The tar-nitrogen conversion may have a significant effect Fig.3 shows first order rate data for tar consumption, tar-
on the fuel-NO formation. Formation of HCN and NH3 has N release of light nitrogen species (HCN and NH3), and
been shown to follow the production of tar and a large pyrrole and pyridine decay. The overall tar decay occurs
fraction of these light N-containing gases may come from with an activation energy of about 77 kJ/mol [66]. This
cracking reactions of the tar [39]. Unfortunately the value is too low to be consistent with direct ring opening of
knowledge about rate and products of the thermal the aromatic structures, but may be associated with the initial
degradation of the tar is still limited. The problem is
complex, partly because the tar consists of a large number of
different compounds and partly because the characteristics
of the tar may be dependent both on the parent fuel and on
the pyrolysis conditions. Our knowledge of the tar chemistry
comes from structural analysis of different tars and from
experimental studies of tar pyrolysis and oxidation.
Furthermore, useful information is available from model
compound studies.
The nitrogen in the tar is largely incorporated in the same
structures as in the parent coal, i.e. mainly in pyrrolic and
pyridinic structures that are released intact during the
primary devolatilization. However, the nitrogen function-
ality of tars includes a variety of structures (pyrrole,
pyridines, quinolines and indoles as well as cyanoaromatics
like benzonitrile and cyanonaphthalenes [94,123,124]) and
it may change with both heating rate, pyrolysis temperature
and molecular mass [44,124]. The cyanoaromatics are not
present in the parent fuel, but may be formed from
conversion of pyridinic nitrogen [124]. The amount of Fig. 3. First order rate constants for tar decay [66] and for HCN/NH3
quaternary nitrogen in the tar is reduced compared to the formation from tar [92,125]. Rate constants for pyrrole [82,126,127]
parent coal [44,124]. and pyridine decay [82,126] are shown for comparison.
Fig. 4. Selected nitrogen containing cyclic compounds. Pyridinic structures include pyridine, picoline (a pyridine with a methyl group), pyridol
(a pyridine with an OH-group), quinoline (pyridine ring fused to benzene), and acridine (dibenzo-derivative of pyridine). Related structures
include pyrazine, pyrimidine, and pyridazine (six-membered rings with two nitrogen atoms). Pyrrolic compounds include pyrrole,
methylpyrrole, indole (pyrrole ring fused to benzene), and carbazole (dibenzo-derivative of pyrrole).
loss of side chains of carbon, hydrogen and oxygen. The char-N oxidation (discussed below) that soot-N is largely
reported rates for release of HCN and NH3 from the tar are oxidized to NO by reaction with O2.
in good agreement [92,125]. The activation energy for the Soot is efficient in reducing NO, when both components
tar-nitrogen release is higher than for the overall tar are present simultaneously in pulverized fuel flames [66].
decomposition, with reported values ranging from 140 to Soot formed from gas-phase precursors reacts rapidly with
220 kJ/mol [92,125]. While the tar-N decomposition rate is NO at high temperatures [128,129] and soot from coal tar
comparable to that of pyridine, the activation energy for tar- would be expected also to have a high reactivity towards
nitrogen release is significantly lower than that for thermal NO. However, soot formed from gas-phase precursors and
decomposition of pyridine and pyrrole, which is about soot formed from coal tar have quite different characteristics
300 kJ/mol [82,126,127]. The lower activation energy for and could be expected to react differently. Depending on the
HCN release from tar compared to model compounds is heat treatment, coal derived soot may have more aliphatic
conceivably due to the presence of reactive species in the tar attachments and heteroatoms than soot from hydrocarbon
that promote fragmentation. This could occur through a free flames [76]. The high nitrogen content is another difference.
radical mechanism similar to that proposed for coal Also soot formed from nucleation of gas-phase precursors
pyrolysis [95]. may contain some nitrogen, presumably from reaction of
During the secondary pyrolysis of tar at high tempera- soot with NO, but levels are much lower than for coal-
tures, most of the tar-N is incorporated into soot (early derived soot [76].
stages) or released as HCN (late stages). However, during
intermediate stages of the pyrolysis release of a number of 5.3. Conversion of gas-phase cyclic nitrogen species
nitrogen-containing polyaromatic compounds, primarily
two-ring structures, has been reported [94]. These appear Pyrolysis and oxidation of a number of model com-
to be more stable thermally than the corresponding nitrogen- pounds have been studied in order to know more about fuel-
free polyaromatic compounds and may potentially consti- nitrogen transformations in combustion. These include both
tute an emission problem. cyclic nitrogen species (Fig. 4) and aliphatic nitrogen
species (Fig. 5). As nitrogen in most solid fuels is believed
to be incorporated mainly in pyrrolic and pyridinic
5.2. Soot-N conversion structures, most investigators have studied compounds
Soot formed from tar through secondary pyrolysis may

contain significant amounts of bound nitrogen [71,72,76].
Little is known about the fate of the soot-N. The nitrogen in
coal-derived soot has been shown to be thermally very
stable [72,122]. It would be expected from analogy with Fig. 5. Selected nitrogen containing aliphatic compounds.
with these functionalities. However, it is obvious from the The principal nitrogen-containing product is cyanoacetylene
results of Fig. 3 that there are important differences between (HCCCN), which at high temperatures is rapidly converted
the decomposition characteristics of model compounds and to HCN. Other major products are acetylene and hydrogen.
the behavior of tar-nitrogen. Furthermore, the differences in The initiation, which is also believed to be the rate limiting
experimental conditions and in pyrolysis/oxidation path- step [136], is elimination of a hydrogen atom from pyridine
ways make a detailed comparison of the data for different to form a pyridyl radical. The pyridyl radical undergoes ring
model compounds very difficult. Even pseudo first-order opening to form an open-chain nitrile, which subsequently
rate constants are difficult to compare because of the decomposes and leads to the major products C2H2 and
different experimental conditions. However, available data HCCCN.
provide information on decay rates and product yields for In general HCN is the most important product in pyrolysis
the different compounds as a function of reaction conditions. of the model compounds and HCN yields are increasing with
Furthermore they constitute a database for developing and temperature. High temperature shock tube results for
validating detailed kinetic models that may be extrapolated decomposition of pyrimidine and pyrazine show similar
to practical conditions. results for HCN, while HCN yields from the decomposition
Studies on pyridinic structures include pyridine [126, of picoline are significantly lower [141]. This indicates that
130– 139], picoline [140,141], pyridol [142], quinoline HCN is easier formed from components with two N-atoms
[130,133,143] and acridine [142]. In addition, work has in the ring. It could be expected that nitriles which contain the
been reported on pyrazine/pyrimidine/pyridazine [135,136, –CxN group would yield high amounts of HCN, but this is
144,145]. The work on pyrrolic compounds include pyrrole not supported by the available data.
[127,130,133,146– 150], methylpyrrole [151], indole [133, It is noteworthy that the decomposition products of
152] and carbazole [133,152]. pyridinic and pyrrolic model compounds do not include
It is useful when discussing decomposition of gas-phase NH3, independently of the nitrogen functionality. However,
nitrogen compounds to define a short time scale (,1 ms), if the model compound contains oxygen functional groups
representative of the near-particle residence time, and a long (hydroxyl, carboxyl or carbonyl groups), NH3 may be
time scale (,1 s), of the order of the overall gas residence formed during pyrolysis [142]. The product ratio NH3/HCN
time in many combustion systems. The temperature for 50% is influenced by the oxygen functionality and its position
conversion of model compounds at the two time scales is relative to nitrogen in the molecule. The compounds with a
listed in Table 2. hydroxyl group on the carbon next to the nitrogen atom
Rapid thermal decomposition of pyrrole requires tem- show the highest NH3/HCN ratios [142].
peratures above 1300 K. The major products are crotonitrile The presence of O2 opens up alternative consumption
(CH3CHCHCN), allyl cyanide (CH2CHCH2CN), HCN and pathways for the model compounds. In general the
propyne/allene (C3H4) [127,146]. These components are temperature for onset of rapid consumption decreases
largely converted to HCN and C2H2 at higher temperatures compared to inert conditions and the consumption rate
and sufficient reaction times. Detailed chemical kinetic increases. However, it is difficult to identify general trends,
models for the pyrrole decay still rely mostly on estimated as only a few oxidation studies are available in literature.
parameters, but recent theoretical work has allowed the The conversion at FBC temperatures is generally low, but it
development of a more reliable mechanism for the complex increases with temperature and O2 level [130,139,142,153,
pyrolysis chemistry [149]. The rate limiting step is the ring 154]. For pyridine and pyrrole oxidizing conditions shift the
opening, which is believed to involve a hydrogen migration temperature for onset of rapid conversion 200 – 250 K
in pyrrole to form a carbene which subsequently breaks up towards lower values [130].
and form HCN and propyne [149]. While few details are available on the oxidation
Pyridines are more stable than pyrroles [126,130,133]. chemistry of pyrroles, a chemical kinetic model has been
The decomposition of pyridine to form HCN occurs at proposed for pyridine oxidation [139]. Similar to pyrolysis,
considerably higher temperatures and with a fundamentally reaction is believed to be initiated by formation of the
different mechanism to that of pyrrole [127,134]. At short pyridyl radical, which may be formed both by C – H bond
time scales temperatures of more than 1500 K are required. fission of pyridine or by reaction of pyridine with O2.
However, the initiation steps are not rate determining in the
Table 2 presence of oxygen. Reactions of pyridine with mainly O
Temperatures for 50% conversion of fuel nitrogen compounds at atoms and of pyridyl with O2 result in formation of
long and short residence times, based on experimental results [126, oxygenated intermediates which subsequently may lead to
130,133,134,137,146] a range of products and intermediates, including pyrrole.
T50% at 1 s (K) T50% at 1 ms (K)
5.4. Conversion of light N-volatiles
Pyridine 1170–1230 1580–1700
Pyrrole 1110–1180 1300–1450
The light N-containing species that are important in fuel-
N oxidation are mainly HCN and NH3. For this reason most
work in the past on gas-phase fuel nitrogen conversion has

focused on these species. The oxidation chemistry of
hydrogen cyanide has been studied under flame conditions
[1,155], in flow reactors [154,156 – 159], in jet-stirred
reactors [160,161] and in shock tubes [162]. Even more
results are available for the high temperature chemistry of
ammonia, which is the most extensively studied gas-phase
nitrogen species. Results on oxidation of NH3 are available
from flames [163– 171], shock tubes [172 – 180] and flow
reactors [157,159,181]. In addition, a number of modeling
studies and reviews of ammonia chemistry have been
reported [1,182– 184]. Supplementing the oxidation studies,
the extensive work on the kinetics of the thermal DeNOx
process [1,185– 187] has been an important source of
knowledge on ammonia chemistry. In this process NH3 is
used as an additive to reduce NO to N2 in a temperature
window around 1250 K. Even though the chemistry of this
process is not the focus of the present paper, it is quite
important for amine/NO interactions at temperatures below
1400 K. Fig. 6. Conversion of gas-phase fixed nitrogen to NO in premixed
In addition to HCN and NH3, also HNCO and various flames as function of excess air ratio (HCN [205], NH3 [201,205,
nitriles may play some role in fuel-N conversion. Few 206], CH3CN [206], pyridine [207,208]).
results are available on HNCO oxidation [188]; some
additional information comes from studies of the concentration ([209], see Fig. 7). Higher levels of fuel-N
RAPRENOx process [1,188 –191], where HNCO acts as a lead to decreasing fractional conversion to NO, since
reducing agent for NO in a process similar to thermal reactions between two nitrogen-containing species are
DeNOx. Nitriles have not been observed in the coal solid promoted compared to reaction of nitrogen radicals with
matrix, but they may be formed from the decomposition of O2 leading to NO. The distribution of fixed-nitrogen
other nitrogen functionalities [135,144,192] and acetonitrile species leaving the burner in fuel-rich flames is largely
has been observed in coal pyrolysis gases [193]. The independent of the form of the nitrogen in the fuel, both for
chemistry of a number of nitriles has been studied, including laminar flames [207,210,211] and turbulent flames [212].
acetonitrile [194– 197], acrylonitrile [198] and benzonitrile In laminar hydrocarbon flames only HCN and NO obtain
[126,130,199]. significant levels in the primary reaction zone; amines are
The oxidation chemistry of the light N-containing
species in gas-phase was reviewed by Miller and Bowman
[1]. Subsequent studies [158,159,183,184,187,188,200]
have refined the mechanism, but the major reaction
pathways have generally been confirmed. The oxidation
chemistry of hydrogen cyanide and ammonia is quite well
established compared to that of nitriles and nitrogen
containing ring compounds. However, even for the light
nitrogen compounds some uncertainties remain, in particu-
lar in the understanding of their chemistry under very fuel-
rich conditions.
A number of laboratory studies on homogeneous
laminar flames doped with fuel-N have shown that the
speciation of the gas-phase nitrogen compounds (amines,
cyanides, five- or six-membered ring structures) does not
have a significant effect on the NO yield, neither in lean
premixed flames [201– 203] or in diffusion flames [204].
Fig. 6 shows the conversion to NO of various fixed
nitrogen species in laminar premixed flames as a function
of excess air ratio. Under lean conditions the fixed nitrogen Fig. 7. Conversion of NH3 to NO in premixed CH4/NH3/air flames
is oxidized to NO with a high conversion efficiency. The as function of excess air ratio and inlet NH3 concentration [209].
selectivity is not very dependent on the fuel or the flame Symbols: (W) 1.0% NH3, ( £ ) 5.0% NH3, (K) 9.3% NH3, and (þ )
temperature [202], but varies significantly with the fuel-N 23.6% NH3.
Fig. 8. Reaction path diagram illustrating major steps in volatile-N

conversion in flames for different nitrogen species (HCN, NH3,
HNCO, CH3CN and pyridine) at moderate fuel-N concentrations. Fig. 9. Plug flow modeling predictions of the NO formation as a
Solid lines denote elementary reaction pathways, while dashed function of temperature in oxidation of a CH4/CO/H2 mixture doped
arrows denote routes that involve intermediates and reactions not with 1000 ppm of a fixed nitrogen compound (A, HCN; S, NH3; K,
shown. The diagram is based on modeling predictions with a HNCO; L, CH3CN; W, pyridine) under fuel-rich conditions ðl ¼
detailed reaction mechanism drawn from literature ([200] sup- 0:85Þ and a residence time of 0.3 s. Reaction mechanism as in Fig. 8.
plemented with [139,184]).
formed mainly in the post-flame gases [207,210,211]. The FN species other than NO. Fig. 10 shows modeling
spread in data for fuel-rich conditions in Fig. 6 is due to predictions for very lean oxidation of HCN, NH3, HNCO,
differences in fuel-N content [209] and residence time and an HCN/NH3/NO mixture as function of temperature.
effects [201] rather than differences in fuel-N speciation. The calculations are performed for conditions representative
The reason for the small influence of the initial N of the rich-lean transition in staged combustion [213], where
speciation is that at the high temperatures prevailing in the remaining combustibles mainly consist of CO. The
flames, the nitrogen atom in the fuel nitrogen species is temperature for onset of reaction is largely determined by
sequentially stripped of the H and C elements it is bonded to, the moist CO oxidation, but the selectivity for forming NO,
ending up as NH or N. This is shown in the reaction diagram N2O, and N2 is specific for the different FN components.
in Fig. 8. At lower temperatures a number of alternative reaction
Even though the nitrogen species initially follow pathways for NCO and NH2 become competitive and both
different oxidation paths, the steps that determine the
selectivity towards NO and N2 are the same. All species
feed into the amine radical pool (the dashed box in Fig. 8),
where the subsequent reactions of NH and N only depend on
the flame conditions, i.e. stoichiometry, temperature, total
fixed nitrogen level, etc. Hydrogen cyanide, acetonitrile and
pyridine are converted to NH in a sequence of reactions
mainly involving CN and NCO, while ammonia and
cyanuric acid are oxidized through NH2. For this reason
the nitrogen volatiles partitioning is often of little import-
ance for the NO formation.
Unfortunately the similarities between N-oxidation
pathways shown in Fig. 8 break down at lower temperatures
or under very fuel-rich conditions. Here significant
differences may be found in the rate and products of
conversion, dependent on the N speciation. Fig. 9 shows
modeling predictions with a detailed chemical kinetic model
for conversion of various fixed nitrogen compounds to NO
in plug-flow under slightly fuel-rich conditions as a function
of temperature. Similar to the results for the premixed Fig. 10. Plug flow modeling predictions for lean oxidation of moist
flames (Fig. 6) little difference in selectivity is found in most CO (2% CO and 4% O2) with 600 ppm fixed nitrogen added (W,
of the temperature regime. However, as the temperature HCN; K, NH3; L, HNCO; £ , HCN/NH3/NO mixture) and a
decreases larger deviations are predicted, in particular for residence time of 0.2 s. Reaction mechanism as in Fig. 8.
the overall reaction rate and product-N speciation may vary the solid matrix and NO/N2O or gas-phase reactions such
significantly between HCN, NH3 and HNCO. In Fig. 10 this as oxidation of HCN to NO, N2O or N2.
occurs at temperatures below 1150 K, but at other reaction Results indicate that during oxidation of coal char the
conditions significant differences may persist at higher char-N release rate compared to the carbon release rate may
temperatures. The oxidation selectivity to NO and N2 is be larger, similar or lower, dependent on the reaction
different for NCO and NH2, but perhaps more important the conditions and the reactivity of the char. Low temperature,
reaction low heating rate oxidation experiments indicate that
nitrogen is preferentially retained in the char at early
NCO þ NO O N2 O þ CO ðR6Þ burnout stages [218,219], mostly in pyridinic structures
while pyrrolic N is oxidized. At higher heating rates and
is a significant source of N2O, while amine/NO interactions higher temperatures the oxidation of carbon and nitrogen in
only yield small amounts of N2O [1,158,187,213]. For this char appears to be non-selective [220]. At high temperatures
reason species such as HCN or HNCO that are partly the nitrogen release rate may be considerably higher than
oxidized through the NCO radical may yield significant that of carbon at the onset of char combustion due to
amounts of N2O at temperatures typical of FBC [28]. pyrolytic reactions, whereas during the final stages of
oxidation the fractional loss of nitrogen is comparable to
that of carbon [4,86,221].
6. Char oxidation Combustion of small, highly devolatilized char particles
at single particle conditions (1000– 1400 K, 5 – 20% O2) has
In coal combustion, both in pulverized fuel systems and shown that char nitrogen is oxidized to NO with a yield of
in fluidized beds, oxidation of char-N is most likely a major 75– 100% [214 – 216]. This indicates that nitrogen in char is
contributor to formation of NO. In both systems, conversion largely oxidized to NO and that NO yields far below 100%
of volatile nitrogen to N2 is achieved through controlled air are caused by reduction of NO in the pores of the char
addition, using over-fire air (PFC, FBC) or low-NOx-burners particle in which the NO was formed, or by a reaction with
(PFC). In FBC the effect is further enhanced by the low other char particles. This conclusion is supported by recent
combustion temperatures. In biomass combustion a com- results from fixed bed experiments [217,222,223]. The data
paratively larger fraction of the fuel-N is released with the in Fig. 11 show that under single particle conditions the net
volatiles, but the significance of char-NO in these systems is char-N to NO conversion lies in the range 75 – 100%, but the
not well established. conversion efficiency rapidly decreases as the amount of
char increases. Preliminary results indicate that char
6.1. The char nitrogen oxidation mechanism nitrogen in biomass has a similar oxidation behavior [224].
There is now consensus that NO is the major product
Even though it would appear to be fairly straight-forward formed from oxidation of coal char-N by O2. In addition to
to characterize char-N oxidation as a function of fuel-type NO, HCN [219,223,225– 228] and HNCO [223,229] may be
and reaction conditions, both the mechanism of the char-N released during the oxidation process in quantities
oxidation and the selectivity towards NO and N2 under
combustion conditions have long been in dispute. Reported
conversions of char-N to NO range from below 30% [31,79,
96,104] to almost 100% [214 – 217]. Furthermore, the
speciation of the char-N oxidation products has been subject
of discussion; a number of species (NO, N2O, HCN, NH3,
HNCO, N2) have been observed [4,11].
The differences between various experimental results
can to some extent be explained by differences in fuel
type. However, char oxidation experiments are often
difficult to interpret due to a number of physical and
chemical complexities. In order to obtain results relevant
for combustion, the chars should be prepared under
conditions which simulate the heating rate, particle size
and pyrolysis temperature experienced in real systems [3].
However, convenient laboratory methods such as TGA
suffer from lower heating and combustion rates than
encountered in practical systems. Furthermore, experimen-
tal characterization may be complicated by parallel release Fig. 11. Net NO formation selectivity from char combustion in a
of volatile-N, for instance as HCN, or by subsequent fixed bed as a function of initial char mass for a pulverized (10–
reactions of primary products, i.e. reactions between 20 mm) bituminous coal char [217].
corresponding to up to 20% of the char-N. At the low formation of N2O from reaction of char-N with gaseous NO
temperatures typical of FBC, which most investigations in a reaction similar to (S3) can generally be disregarded. To
cover, very little HCN is released from coal char under account for the observed N2O, two mechanisms have been
oxygen-free conditions [230] and it is assumed that HCN proposed. The homogeneous mechanism involves release of
and HNCO result from the heterogeneous oxidative process HCN and HNCO during char oxidation, followed by
itself rather than from slow devolatilization [219,223]. reaction of the NCO radical with NO to form N2O (reaction
Formation of HCN during char oxidation appears to be most (R6)) during the subsequent oxidation. The heterogeneous
pronounced at temperatures below 1100 K [223]. mechanism involves the following global steps [246]
Even though details remain to be clarified, the overall 1
reaction mechanism for char-NO formation and subsequent CðNÞ þ 2 O2 ! CðNCOÞ ðS5Þ
reduction of NO on the char appears to be fairly well
CðNCOÞ þ NO ! Cf þ N2 O þ CO ðS6Þ
established from recent work [4,222,223,231]. The first step
is the formation of NO Here C(NCO) is an oxidized char nitrogen species,
CðNÞ þ O2 ! NO þ CðOÞ ðS1Þ presumably an – NCO or – CNO type species.
The occurrence of the homogeneous mechanism is
Here C(N) and C(O) denote nitrogen and oxygen species, supported by a considerable amount of experimental
respectively, on the char surface. The NO formed may evidence, while the importance of the heterogeneous
subsequently be readsorbed on the char surface by a mechanism is still open to discussion. In support of the
dissociative chemisorption homogeneous mechanism, little or negligible N2O for-
2Cf þ NO ! C0 ðNÞ þ CðOÞ ðS2Þ mation has been detected directly from char oxidation
[223], while increased gas residence time above the particle
0
Cf denotes a free carbon site, while C (N) is a nitrogen bed promotes N2O formation [216,253,254]. Further
surface species, which may be different from the char-N site support is provided by experiments on char oxidation
C(N). Reaction (S2) leads to accumulation of N on the char where a chemical agent is added to perturb the gas-phase
surface, at least at lower temperatures. Reaction between chemistry (but not any heterogeneous reactions). The
C0 (N) and NO from the gas phase may lead to formation of experiments of Winter and coworkers [225,255,256]
free nitrogen showed that in the presence of iodine, which is a strong
C0 ðNÞ þ NO ! N2 þ CðOÞ ðS3Þ inhibitor for gas-phase reactions, HCN levels increased and
N2O levels decreased. In the presence of MgO particles,
There is evidence [222,223] that in the 900– 1200 K range which are known to catalyze oxidation of HCN to NO, both
reaction (S3) is the major step forming N2, while the solid the HCN and N2O levels decreased significantly [247].
state recombination of two C0 (N) species These results indicate that N2O is formed from secondary
reactions of HCN and HNCO. The levels of N2O observed
C0 ðNÞ þ C0 ðNÞ ! N2 þ 2Cf ðS4Þ
during char oxidation are typically in the range 0 – 15% of
is less important. char-N [41,242,248,257], i.e. values consistent with the
The validity of the proposed mechanisms (S1)– (S3) for HCN/HNCO release and the oxidation selectivity towards
char-N oxidation over a wider temperature range needs to be N2O of these species [158,188].
confirmed. However, the scheme appears to be in agreement The heterogeneous mechanism is supported by exper-
with the observations made on char-N oxidation. The NO iments showing that N2O formation proceeds for a while
yield decreases with the bulk gas NO concentration [221], after the O2 supply is shut off, indicating the importance of
with an increasing particle size [215,232], with increasing reactions of surface complexes with NO [4,230]. Modeling
pressure [233– 236], and with increasing char reactivity studies of char oxidation under single particle oxidation
[227,237– 241]. Increased particle size and pressure both [258] and in a circulating fluidized bed [259] also indicate
yield a longer residence time for NO in the pores and some contribution from a heterogeneous mechanism, but
consequently a larger reduction. The instantaneous NO yield direct evidence has not been presented. Conceivably, the
increases during burnout [223,242,243] because the particle relative importance of the homogeneous and heterogeneous
becomes smaller and more porous and thereby allows larger mechanisms depends on char characteristics and reaction
amounts of NO to escape before being reduced. conditions.
6.2. Formation of N2O from char-N 6.3. Reduction of NO on char
In addition to NO, N2 and HCN, considerable amounts of The reduction of NO by char is a complex process, which
N2O may be formed during char oxidation at fluidized bed is not completely understood. The reported reaction order
temperatures [3,11,230,232,244 – 250]. The presence of varies, but most studies find that the reaction is first order in
oxygen is required to form N2O from char [244,251], NO [5]. At temperatures above 950 K there is a change of
except at very high bulk gas NO concentrations [252], and mechanism/activation energy [3,5]. Below this temperature
this range, referred to as the bulk group, were all obtained

with char, which was prepared separately, and represent
steady-state values for the char þ NO reaction. The scatter
in these data can largely be attributed to differences in char
type and surface area. However, two determinations deviate
significantly from the bulk group: a low rate constant for
lignite char [263] and a high rate constant for bituminous
coal char [217]. The data of Jensen [217], which were
obtained with a char directly in continuation of pyrolysis,
indicate a reaction rate that is 10 – 100 times faster than the
Aarna and Suuberg rate constant. This value represents a
transient char reactivity towards NO; Jensen et al. observed
that the rate of the char þ NO reaction decreased by more
than an order of magnitude from the initial value, as steady
state was approached. The Song rate constant is a steady-
state value, obtained with a char prepared at a high
temperature of 1750 K. This value is lower than the bulk
Fig. 12. Rate constant for the char þ NO reaction. (S) bituminous data by 1 –2 orders of magnitude.
coal char [217,244,261,262]; (K) lignite char [263,264]; ( £ ) We attribute the differences between the Jensen rate, the
biomass char [265]. The thick solid line represent the recommen- bulk group, and the Song rate mostly to char deactivation prior
dation of Aarna and Suuberg [5] for reaction of coal char with NO; to reaction, due to high pyrolysis temperatures and/or long
their rate constant based on surface area was converted to mass holding times. The reactivity of a char depends strongly on the
based units assuming a char surface area of 100 m2/g. The thin solid pyrolysis temperature; chars prepared at lower temperatures
lines represent order of magnitude intervals for the Aarna and are more reactive both towards O2 and NO [238,269]. Loss of
Suuberg recommendation.
reactivity through thermal annealing has been shown to be
quite important for the char þ O2 reaction [270– 279] as well
N2 and CO2 are formed, while at higher temperatures also as for char þ NO [5]. However, other factors than thermal
CO becomes a major product with the CO/CO2 ratio deactivation may also contribute to the loss of reactivity. The
increasing with temperature. Presence of CO is known early stage reactivity of transient char-gas reactions is often
promote the reaction rate, because char catalyzes reduction considerably higher than the steady-state value obtained at
of NO by CO [5,260]. long reaction times. The enhanced early reactivity has been
Fig. 12 shows an Arrhenius plot for the reaction of char attributed partly to accumulation of nitrogen oxide on the char
from bituminous coal, lignite and biomass (wheat straw) surface [264,280], probably through the formation of C(N)
with NO. Also the rate constant proposed by Aarna and complexes. In addition to deactivation and accumulation of
Suuberg [5] for the carbon þ NO reaction is shown (thick NO, a large number of very reactive sites may be present
solid line). This rate constant, which was obtained from initially on the carbon surface, due to formation of
averaging a large amount of data, agrees within a factor of intermediate surface complexes, high concentration of
10 to most reported values. The rates shown are based on dangling bonds, presence of disordered carbon, etc. [281].
mass instead of surface area; the Aarna and Suuberg rate Reduction of char reactivity towards NO over time may also
constant was converted from a surface area based value to a be due to formation of surface oxides [260,282].
mass based value assuming a typical coal char surface area In general, char reactivity towards NO would be
of 100 m2/g. Since both the surface area and the effective- expected to increase with the fuel volatility, similar to
ness factor [266] may vary significantly between coals, this what has been observed for other char gasification reactions.
value is only approximate. Total surface area may also vary However, the data shown in Fig. 12 as well as those
with measurement technique and the appropriate surface presented by Aarna and Suuberg [5] do not show a clear
area for normalizing rates is still in dispute [267]. Recent effect of fuel type. For comparable char preparation and
results [268] indicate that the measurements on the char þ reaction conditions, the steady state rate constants for lignite
NO rates correlate better with the surface area associated and biomass appear to be somewhat higher than those of
with macropores and some mesopores (.20 nm by mercury bituminous coals (Fig. 12). This may be attributed to higher
porosimetry) than with BET areas. In addition to surface contents of Ca, Fe and Mg, which may catalyze reaction,
area, the reactivity of char may depend on the accessibility and to a higher BET surface [261]. It has been reported that
and reactivity of specific sites, including catalytic active chars from peat and lignite may be more reactive towards
sites created by inorganic impurities. NO than chars from bituminous coal and wood [283], but
Independent of fuel, most of the data in Fig. 12 are in more work is needed to confirm this observation.
agreement with the Aarna and Suuberg rate constant [5], Little information is available on the potential catalytic
within their reported variation of factor of 10. The data in effect of inorganic species such as Fe or Ca on the char-N
oxidation. For bituminous coals the effect appears to be

small [222], but it may be different for low rank coals and
biomass.
6.4. Effect of reaction conditions on NO yield
Experiments under single particle conditions have shown

that for temperatures typical of FBC reactors, higher O2
concentrations decrease the yield of NO [284,285], while
fixed bed experiments at medium and higher temperatures
indicate that the effect of O2 level is small [217,223]. The
reason for the more pronounced effect of O2 at lower
temperatures is not clear, but it is conceivable that higher O2
concentrations promote formation of a reactive char-N
intermediate (either C(CNO) on the surface [246] or gas-
phase HCN) that reacts rapidly with NO to form N2O. At Fig. 14. Conversion of char-N to N2O as function of temperature in
higher temperatures formation of this N-intermediate is no fluid bed experiments [257]. Symbols: (W) anthracite, (K)
longer thermodynamically favored and this mechanism is bituminous coal, (L) subbituminous coal, and (X) lignite.
less important.
The effect of stoichiometry is more pronounced at higher experiments, both under single particle conditions [247,257]
temperatures and higher particle loadings. Entrained flow and at higher particle loadings [232,245,249,288] and it
reactor experiments with char from pulverized coal [286] corresponds to a decrease in N2O yield of roughly the same
indicate that the NO yield increases with the oxygen magnitude (Fig. 14). The behavior can be explained in terms
availability, from about 30 –40% of the released char-N at of the homogeneous mechanism. In this interpretation the
reducing conditions up to 70 – 80% at fuel-lean conditions. N2O yield from HCN/HNCO oxidation peaks at lower
Fig. 13 shows the fractional conversion of char-N to NO temperatures, while as the temperature increases, the
as function of temperature in coal char oxidation. Despite formation of HCN decreases. Furthermore, the oxidation
the scatter due to differences in coal rank and reaction of HCN/HNCO to NO is favored compared to N2O
conditions, the data indicate that the NO yield increases with formation at higher temperatures, with the transition
temperature up to a certain temperature, above which it temperature depending on the gas-phase composition, in
decreases with temperature. The peak in NO at a certain particular the concentration of combustibles. However, the
temperature in char-N oxidation has been noted previously experimental data may also be explained in terms of the
[232,287], but it is not yet fully explained. heterogeneous N2O mechanism and models based on this
The increase in NO yield with temperature under scheme can reproduce qualitatively the behavior of NO as
fluidized bed conditions has been observed in a number of function of temperature [246,250,289]. In this interpret-
ation, the reactive char-N intermediate C(CNO) removes a
large fraction of the NO formed at lower temperatures,
converting it to N2O. As the temperature increases, the
surface concentration of C(CNO) decreases, leading to an
increase in NO as well as a decrease in N2O, consistent with
observations.
The importance of N2O in the char-N conversion
decreases rapidly with increasing temperature, partly
because N2O formation is less favored and partly due to
homogeneous and heterogeneous decomposition of N2O. A
number of other mechanisms may also play a role in the NO
yield from char oxidation as function of temperature. These
include diffusion phenomena involving NO and O 2,
decreasing char reactivity with increasing temperature,
and catalytic reduction of NO by CO on char.
The fact that NO decreases with temperature above
Fig. 13. Conversion of char-N to NO as function of temperature. (W) 1100–1300 K can be explained in terms of the competition
anthracite (fluid bed [257], fixed bed [217]), (K) bituminous coal between the char þ O2 and the char þ NO reactions. At
(fluid bed [257], fixed bed [11,217]), (L) subbituminous coal (fluid kinetic control the char þ O2 reaction is much faster than
bed [257]), X lignite (fluid bed [257], entrained flow reactor [96]). char þ NO [5]. However, at high temperatures and/or larger
particles the char þ O2 reaction becomes diffusion-limited

and an increase in temperature promotes mainly the
char þ NO reaction.
Modeling the char-N conversion to NO remains a major
challenge. Even though the overall mechanism of NO
formation and subsequent partial reduction of NO on the
char surface to N2 is fairly well understood, a quantitative
prediction of this process is very difficult. The char reactivity
cannot presently be estimated a priori with any confidence,
partly due to the complexity of physical and chemical
processes occurring during devolatilization, such as thermal
annealing, change in surface area, loss of catalytic com-
ponents, etc. that may affect the char reactivity significantly.
In modeling of pulverized fuel systems, which are charac-
terized by solids residence times of a few seconds,
comparatively high rate constant values for char þ NO,
corresponding to the initial rate, may be appropriate unless
the pyrolysis temperature is very high. Single particle
modeling studies of char-N oxidation indicate that a char þ Fig. 15. Effect of flame mode and fuel-N concentration on fuel-N
NO reaction rate constant higher than those generally conversion to NO ðhNO Þ in lean gas and coal-dust flames. The fuel-
reported in literature is necessary to explain experimental N level is shown as the equivalent NO concentration in the flue gas,
results [290]. A fast char þ NO rate constant may also be provided all fuel-N is oxidized to NO. (L) CH4/NH3 laminar
required [217] to explain experimental data on reburning premixed flames [202,209]; (W) CH4/NH3 laminar diffusion flames
with coal char [291]. However, it is worth noting that in a [299]; (W) turbulent ‘premixed’ coal-dust flame [300]; (X) turbulent
axial diffusion coal-dust flame [300]; (þ) swirling diffusion coal-
recent high temperature entrained flow reactor study on char
dust flames covering range of coals [104].
oxidation [286], the char þ NO reaction appeared to be of
minor importance, perhaps due to comparatively severe transformations occurring. Furthermore, this time scale,
conditions in the char preparation. In fluidized bed or grate which is proportional to the square of the particle diameter,
combustion the char reactivity will reach the steady-state varies significantly within a typical pulverized fuel size
value at time scales significantly shorter than the solids distribution. For comparison, char burnout in pulverized
residence in the system and for modeling these applications a
systems generally requires more than 300 ms [30].
lower mean reaction rate would be expected.
Fig. 15 compares experimental results from overall lean
gas flames and coal-dust flames, both for premixed and non-
premixed systems. The results show that in homogeneous
7. Fuel-N conversion in laminar flames systems the fuel-N to NO conversion, here termed hNO ; is
80 – 100%, while values below 30% are observed for
Solid fuel combustion in practical systems involves a diffusion flames with corresponding fuel-N seeding levels.
complex interaction between fluid dynamics, turbulence, Comparing homogeneous and heterogeneous flames, the
heat transfer, and homogeneous and heterogeneous chem- results support the general observation [19] that hNO for
istry; phenomena that all affect fuel-nitrogen conversion. the ‘premixed’ coal-dust flame is significantly lower than
These systems typically require the use of computational the corresponding values for premixed gaseous flames.
fluid dynamics for interpretation. However, with the present While staging is known to be efficient for fuel-NO control in
understanding, prediction of fuel-NO formation in solid fuel pulverized fuel flames [104], the difference in fuel-N
systems is difficult even under conditions with a simple and conversion between lean premixed and non-premixed
well-defined flow field. Few detailed experimental studies coal-dust flames is seen to be significantly smaller than for
have been reported that investigate the flame chemistry of homogeneous flames. The solid-fuel diffusion flames appear
pulverized fuels under such conditions. The reported work to yield higher amounts of fuel-NO than the corresponding
deals with either relatively dense fuel suspension [81,99, homogeneous diffusion flames.
292– 297], representative of pulverized coal suspensions, or Features that distinguish the heterogeneous premixed
close to single particle conditions [66,79,298]. flame compared to the homogeneous flame include local
Unfortunately, results from the primary zone of pulverized mixing phenomena around the devolatilizing particles,
fuel flames are inherently difficult to interpret. Individual presence of reactive solid material (char and soot), and
particles ignite, burn out their volatiles and achieve quasi- finally that part of the fuel-N conversion proceeds through
steady char oxidation in time scales of 10–75 ms [99,298], heterogeneous oxidation of the char-N. Most of the
making it very difficult to monitor in detail the chemical difference in NO emission has been attributed to
the presence of local diffusion flames surrounding single temperatures volatile-N from coal is mostly released with
particles [7,19]. The burning rates of the non-condensible tar, emerging subsequently mostly as HCN, but it may also
volatiles are comparable to that of hydrogen [301] and high be incorporated in soot. For biomass and low rank coals,
enough to support an envelope flame. However, it is likely significant amounts of NH3 may be released directly from
that both mixing processes and heterogeneous processes the solid matrix during devolatilization.
influence the conversion of N-volatiles in the vicinity of the The chemistry of the volatile nitrogen is complicated,
particle [292], with the relative importance dependent on since it may involve oxidation of tar-N, soot-N, cyclic
both fuel type, particle size and particle loading. compounds, and light species such as HCN and NH3.
Single particle experiments [66,298] show conversions Knowledge of conversion of tar-N and soot-N is still
of fuel-N to NO of 50 – 100%, depending on coal type and limited. However, at high temperatures in flames the gas-
particle size, while at higher fuel loadings the NO yield phase volatile nitrogen, independent of speciation, is
decreases significantly [298]. This behavior is similar to sequentially stripped of carbon and hydrogen, and the
that observed for char-N oxidation (Fig. 11). At higher oxidation selectivity towards NO and N2 is largely
particle loadings the devolatilization products may com- determined by the competition between reaction of N
bine to produce a surrounding gaseous environment with atoms with OH/O2 (to form NO) and with NO (to form
high concentrations of nitrogen species and low or zero N2). This competition is mainly governed by the
oxygen concentrations [298]. However, also high concen- stoichiometry (O2 availability) and the total volatile-N
trations of char and soot may serve to limit the NO concentration. At lower temperatures, such as in fluid bed
concentration. It has been observed in pulverized coal combustion, the volatile-N speciation may affect the
flames that, except for low-volatility coals, non-conden- propensity to form NO and N2O, since it becomes
sible fuels, char and soot to some extent burn simul- important whether the oxidation proceeds through the
taneously [99]. The initial reactivity of char towards NO is NH2 or NCO radical.
very high [217] and may have been underestimated in The nitrogen in the char is largely oxidized to NO for
previous analyses. Also soot may be very active; there are small particles and single particle conditions. However, as
indications that coal type effects relate to the propensity to the char loading (or particle size) increases, reaction of char
form soot. The NO yield for bituminous coals under single with NO may serve to efficiently remove NO and reduce the
particle conditions is much lower than for subbituminous char-N to NO conversion efficiency. The reactivity of char
coals, and this difference has been attributed to a rapid towards NO apparently depends both on fuel rank and
reduction of NO by soot [66]. pyrolysis conditions. The transient reactivity of char may be
The char-N conversion to NO decreases with bulk NO an order of magnitude higher than the steady-state value
concentration and may become negative, provided the initial obtained at long reaction times; an effect which may be
NO level is sufficiently high [221]. A net reduction of NO is important in short time-scale systems such as pulverized
thus possible during char burnout and has been observed in fuel firing.
lean premixed pulverized fuel flames [302]. This effect There are important differences between the effect of
would be most important in flames with more reactive chars, flame mode on fuel-N oxidation in homogeneous and
with higher particle loadings, and/or at higher temperatures heterogeneous flames. In gaseous flames the fuel nitrogen
in the char oxidation zone. A higher temperature favors the oxidation selectivity towards NO ranges from 80 – 100%
char þ NO reaction compared to the char þ O2 reaction in premixed flames to 10 – 30% in non-premixed flames
which will typically be diffusion limited. However, a very at typical fuel-N levels. In lean coal-dust flames the NO
high pyrolysis temperature may have an adverse effect on yield is much less dependent on flame mode, with the
both NO and char burnout, as it enhances thermal selectivity for NO ranging from 30 to 50%. Differences
deactivation of the char. between the homogeneous and heterogeneous flames
include mixing phenomena around devolatilizing par-
ticles, oxidation of nitrogen from the solid matrix, and
presence of reactive solid material (char, soot) that may
8. Concluding remarks
reduce NO.
Most of our understanding of fuel nitrogen conversion in

solid fuel fired systems involves coal as a fuel; compara-
tively little is known about other fuels such as biomass. Acknowledgements
However, even for coal there are unresolved issues that may
limit the potential of primary measures for NOx control. The authors would like to thank Adel Sarofim, Tom
The functional form of the nitrogen in the fuel shows no Fletcher and Brian Haynes for helpful comments. This
simple correlation either with rank or with the HCN/NH3 work, which is part of the research program CHEC
ratio in the volatiles. However, it affects the nitrogen release (Combustion and Harmful Emission Control), was funded
rate for both devolatilization and oxidation. At high by Elsam, Energi E2, and the Danish Ministry of Energy.
References [20] Jenkins BM, Baxter LL, Miles Jr. TR. Combustion of
biomass. Fuel Process Technol 1998;54:17 –46.
[21] Werther J, Saenger M, Hartge E-U, Ogada T, Siagi Z.
[1] Miller JA, Bowman CT. Mechanism and modeling of
Combustion of agricultural residues. Prog Energy Combust
nitrogen chemistry in combustion. Prog Energy Combust
Sci 2000;26:1–27.
Sci 1989;15:287–338.
[22] Zeldovich YB. The oxidation of nitrogen in combustion and
[2] Davidson RM. Nitrogen in coal. IEA Coal Research; 1994.
explosions. Acta Physicochem USSR 1946;21:577–628.
IEAPER/08.
[23] Fenimore CP. Proc Combust Inst 1971;13:373 –9.
[3] Thomas KM. The release of nitrogen oxides during char
[24] Glarborg P, Miller JA, Kee RJ. Kinetic modeling and
combustion. Fuel 1997;76:457–73.
sensitivity analysis of nitrogen oxide formation in well-
[4] Molina A, Eddings EG, Pershing DW, Sarofim AF. Char
stirred reactors. Combust Flame 1986;65:177–202.
nitrogen conversion: implications to emissions from coal-
[25] Moskaleva LV, Lin MC. The spin-conserved reaction CH þ
fired utility boilers. Prog Energy Combust Sci 2000;26:
N2 ! H þ NCN : a major pathway to prompt NO studied by
507–31.
quantum/statistical theory calculations and kinetic modeling
[5] Aarna I, Suuberg EM. A review of the kinetics of the nitric
of rate constant. Proc Combust Inst 2000;28:2393– 401.
oxide–carbon reaction. Fuel 1997;76:475–91.
[26] Malte P, Pratt DT. Proc Combust Inst 1975;15:1061–70.
[6] Abbas T, Costa M, Costen P, Godoy S, Lockwood FC, Ou JJ,
[27] Åmand L-E, Leckner B, Andersson S. Formation of N2O in
Romo-Millares C, Zhou J. NOx formation and reduction
circulating fluidized bed boilers. Energy Fuels 1991;5:
mechanisms in pulverized coal flames. Fuel 1994;73:
815 –23.
1423–36.
[28] Kilpinen P, Hupa M. Homogeneous N2O chemistry at
[7] Wendt JOL. Mechanisms governing the formation and fluidized bed conditions: a kinetic modeling study. Combust
destruction of NOx and other nitrogeneous species in low Flame 1991;85:94–104.
NOx coal combustion systems. Combust Sci Technol 1995; [29] Sander B. Properties of Danish biofuels and the requirements
108:323– 44. for power production. Biomass Bioenergy 1997;12:177–83.
[8] Williams A, Pourkashanian M, Jones JM, Rowlands L. A [30] Pershing DW, Wendt JOL. Relative contributions of volatile
review of NOx formation and reduction mechanisms in nitrogen and char nitrogen to NOx emissions from pulverized
combustion systems, with particular reference to coal. J Inst coal flames. Ind Engng Chem Process Des Dev 1979;18:
Energy 1997;70:102 –13. 60 –7.
[9] van der Lans RP, Glarborg P, Dam-Johansen K. Influence [31] Pershing DW, Wendt JOL. Pulverized coal combustion:
of process parameters on nitrogen oxide formation in the influence of flame temperature and coal composition
pulverized coal burners. Prog Energy Combust Sci 1997; on thermal and fuel NOx. Proc Combust Inst 1976;16:
23:349–77. 389 –99.
[10] Mitchell SC. NOx in pulverized coal combustion. London: [32] Speight JG. The chemistry and technology of coal. New
IEA Coal Research; 1998. CCC/05. York: Marcel Dekker; 1983.
[11] Johnsson JE. Formation and reduction of nitrogen oxides in [33] Frandsen FJ, Dam-Johansen K, Rasmussen P. Trace elements
fluidized-bed combustion. Fuel 1994;73:1398–415. from combustion and gasification of coal: an equilibrium
[12] Leppälahti J, Koljonen T. Nitrogen evolution from coal, peat approach. Prog Energy Combust Sci 1994;20:115–38.
and wood during gasification: literature review. Fuel Proc [34] Buckley AN, Kelly MD, Nelson PF, Riley KW. Inorganic
Technol 1995;43:1–45. nitrogen in Australian semi-anthracites; implications for
[13] Morrison GF. Nitrogen oxides from coal combustion— determining organic nitrogen functionality in bituminous
abatement and control. London: IEA Coal Research; 1980. coals by X-ray photoelectron spectroscopy. Fuel Process
ICTIS/TR11. Technol 1995;43:47 –60.
[14] Bowman CT. Control of combustion-generated nitrogen [35] Perry DL, Grint A. Application of XPS to coal characteriz-
oxide emissions: technology driven by regulation. Proc ation. Fuel 1983;62:1024–33.
Combust Inst 1992;24:859–78. [36] Bartle KD, Perry DL, Wallace S. The functionality of
[15] Muzio LJ, Quartucy GC. Implementing NOx control: nitrogen in coal and derived liquids: an XPS study. Fuel
research to application. Prog Energy Combust Sci 1996;23: Process Technol 1987;15:351 –61.
233–66. [37] Burchill P, Welch LS. Variation of nitrogen content and
[16] Bowman CT. Control of combustion-generated nitrogen functionality with rank for some UK bituminous coal. Fuel
oxide emissions: 20th century progress and 21st century 1989;68:100–4.
challenges. Proceedings of the Second Asia– Pacific Con- [38] Wallace S, Bartle KD, Perry DL. Quantification of nitrogen
ference on Combustion; 1999. functional groups in coal and coal derived products. Fuel
[17] Kramlich JC, Linak WP. Nitrous oxide behavior in the 1989;68:1450–5.
atmosphere, and in combustion and industrial systems. Prog [39] Nelson PF, Buckley A, Kelly MD. Functional forms of
Energy Combust Sci 1994;20:149–202. nitrogen in coals and the release of coal nitrogen as NOx
[18] Hill SC, Smoot LD. Modeling of nitrogen oxides formation precursors (HCN and NH3). Proc Combust Inst 1992;24:
and destruction in combustion systems. Prog Energy 1259–67.
Combust Sci 2000;26:417–58. [40] Kambara S, Takarada T, Yamamoto Y, Kato K. Relation
[19] Wendt JOL. Fundamental coal combustion mechanisms and between functional forms of coal nitrogen and formation of
pollutant formation in furnaces. Prog Energy Combust Sci NOx precursors during rapid pyrolysis. Energy Fuels 1993;7:
1980;6:201– 22. 1013–20.
[41] Aho MJ, Hämäläinen JP, Tummavouri JL. Importance of [62] Freihaut JD, Proscia WM, Seery DJ. Energy Fuels 1989;3:
solid fuel properties to nitrogen oxide formation through 692.
HCN and NH3 in small particle combustion. Combust Flame [63] Li C-Z, Nelson PF. Fate of aromatic ring systems during
1993;95:22– 30. thermal cracking of tars in a fluidized-bed reactor. Energy
[42] Kelemen SR, Gorbaty ML, Kwiatek PJ. Quantification of Fuels 1996;10:1083–90.
nitrogen forms in Argonne Premium Coals. Energy Fuels [64] Solomon PR, Colket MB. Evolution of fuel nitrogen in coal
1994;8:896– 906. devolatilization. Fuel 1978;57:749.
[43] Wójtowicz MA, Pels JR, Moulijn JA. The fate of nitrogen [65] Baumann H, Möller P. Pyrolysis of hard coals under fluidised
functionalities in coal during pyrolysis and combustion. Fuel bed combustor conditions. Erdöl Kohle-Erdgas Petrochem
1995;74:507 –16. 1991;44:29–33.
[44] Kelemen SR, Gorbaty ML, Kwiatek PJ, Fletcher TH, Watt [66] Haussmann GJ, Kruger CH. Evolution and reaction of coal
M, Solum MS, Pugmire RJ. Nitrogen transformations in coal fuel nitrogen during rapid oxidative pyrolysis and combus-
during pyrolysis. Energy Fuels 1998;12:159– 73. tion. Proc Combust Inst 1990;23:1265–71.
[45] Mullins OC, Mitra-Kirtley S, Van Elp J, Cramer SP. [67] Lau CW, Niksa S. The impact of soot on the combustion
Molecular structure of nitrogen in coal from XANES characteristics of coal particles of various types. Combust
spectroscopy. Appl Spectrosc 1993;47:1268–75. Flame 1993;95:1–21.
[46] Mitra-Kirtley S, Mullins OC, Van Elp J, Cramer SP. Nitrogen [68] Wornat MJ, Sarofim AF, Longwell JP, Lafleur AL. Energy
chemical structure in petroleum asphaltene and coal by X-ray Fuels 1988;2:775–82.
absorption spectroscopy. Fuel 1993;72:133 –5. [69] McLean WJ, Hardesty DR, Pohl JH. Observations of
[47] Buckley AN. Nitrogen functionality in coals and coal-tar devolatilizing pulverized coal particles in a combustion
pitch determined by X-ray photoelectron spectroscopy. Fuel environment. Proc Combust Inst 1981;18:1239–48.
Process Technol 1994;38:165–79. [70] Seeker WR, Samuelsen GS, Heap MP, Trolinger JD. The
[48] Pels JR. Nitrous oxide in coal combustion. Delft, The thermal decomposition of pulverized coal particles. Proc
Netherlands: University of Technology; 1995. Combust Inst 1981;18:1213– 26.
[49] Hall DO, Overend RP. Biomass regenerable energy. New [71] Chen JC, Niksa S. Suppressed nitrogen evolution from coal-
York: Wiley; 1987. derived soot and low-volatility chars. Proc Combust Inst
[50] Kambara S, Takarada T, Toyoshima M, Kato K. Relation 1992;24:1269–76.
between functional forms of coal nitrogen and NOx [72] Chen JC, Castagnoli C, Niksa S. Coal devolatilization during
emissions from pulverized coal combustion. Fuel 1995; rapid transient heating. 2. Secondary pyrolysis. Energy Fuels
74:1247–53. 1992;6:264–71.
[51] Hämäläinen JP, Aho MJ. Effect of fuel composition on the [73] Ma J, Fletcher TH, Webb BW. Proc Combust Inst 1996;26:
conversion of volatile solid fuel-N to N2O and NO. Fuel 3161–7.
1995;74:1922 –4. [74] Fletcher TH, Ma J, Brown AL, Webb BW. Prog Energy
[52] Howard JB. In: Elliott MA, editor. Fundamentals of coal Combust Sci 1997;23:283 –301.
pyrolysis and hydropyrolysis. The chemistry of coal [75] Brown AL, Fletcher TH. Modeling soot derived from
utilization [chapter 12], New York: Wiley; 1981. pulverized coal. Energy Fuels 1997;17:745 –57.
[53] Gavalas GR, Coal pyrolysis. New York: Elsevier; 1982. [76] Rigby J, Ma J, Webb BW, Fletcher TH. Transformations of
[54] Wall TF. The combustion of coal as pulverized fuel through coal-derived soot at elevated temperature. Energy Fuels
swirl burners. In: Lawn CJ, editor. Combustion treatise. 2001;15:52–9.
Principles of combustion engineering for boilers. New York: [77] Miller RS, Bellan J. Tar yield and collection from the
Academic Press; 1987 [chapter 3]. pyrolysis of large biomass particles. Combust Sci Technol
[55] Solomon PR, Serio MA, Suuberg EM. Coal pyrolysis: 1997;127:97–118.
experiments kinetic rates and mechanisms. Prog Energy [78] Bassilakis R, Zhao Y, Solomon PR, Serio MA. Sulfur and
Combust Sci 1992;18:133–220. nitrogen evolution in the argonne coals: experiment and
[56] Leppalahti J. Formation of NH3 and HCN in slow-heating- modeling. Energy Fuels 1993;7:710–20.
rate inert pyrolysis of peat, coal and bark. Fuel 1995;74: [79] Pohl JH, Sarofim AF. Devolatilization and oxidation of coal
1363–8. combustion. Proc Combust Inst 1976;16:491– 501.
[57] Chen Y, Charpenay S, Jensen A, Wòjtowicz MA, Serio MA. [80] Blair DW, Wendt JOL, Bartok W. Evolution of nitrogen and
Modeling of biomass pyrolysis kinetics under combustion other species during controlled pyrolysis of coal. Proc
conditions. Proc Combust Inst 1998;27:1327–34. Combust Inst 1976;16:475– 89.
[58] Di Blasi C, Signorelli G, Di Russo C, Rea G. Product [81] Phong-Anant D, Wibberley LJ, Wall TF. Nitrogen oxide
distribution from pyrolysis of wood and agricultural residues. formation from Australian coals. Combust Flame 1985;62:
Ind Engng Chem Res 1999;38:2216–24. 21 –30.
[59] Niksa S. Predicting the rapid devolatilization of diverse [82] Bruinsma OSL, Geertsma RS, Oudhuis ABJ, Kapteijn F,
forms of biomass with bio-FLASHCHAIN. Proc Combust Moulijn JA. Measurement of C, H, N-release from coals
Inst 2000;28:2727–33. during pyrolysis. Fuel 1988;67:1190–6.
[60] Bassilakis R, Carangelo RM, Wòjtowicz MA. TG-FTIR [83] Jahns P. Verbleib des Schwefels und Stickstoffs bei der
analysis of biomass pyrolysis. Fuel 2001;80:1765 –86. Kohlenpyrolyse und dessen Bedeutung fur Thermische
[61] Freihaut JD, Zabielsky MF, Seery DJ. A parametric Kohleprozesse. Universitat-Gesamthochschule Essen; 1990.
investigation of tar release in coal devolatilization. Proc [84] Chen Y, Matsuda H, Hasatani M. In: Michaelian KH, editor.
Combust Inst 1982;19:1159–67. Behaviors of coal-N release during coal pyrolysis. Proceed-
ings of the International Conference on Coal Science, [106] Rüdiger H, Greul U, Spliethoff H, Hein KRG. Distribution of
London: Butterworth/Heinemann; 1993. p. 55– 8. fuel nitrogen in pyrolysis products used for reburning. Fuel
[85] Man CK, Pendlebury KJ, Gibbins JR. Laboratory measure- 1997;76:201–5.
ment of N release under combustion conditions and [107] Friebel J, Köpsel RFW. The fate of nitrogen during pyrolysis
comparison with plant NOx formation. Fuel Process Technol of German low rank coals: a parameter study. Fuel 1999;78:
1993;36:117 –22. 923 –32.
[86] Baxter LL, Mitchell RE, Fletcher TH, Hurt RH. Nitrogen [108] Solomon PR, Hamblen DG, Carangelo RM, Krause JL. Coal
release during coal combustion. Energy Fuels 1996;10: thermal decomposition in an entrained flow reactor. Proc
188–96. Combust Inst 1982;19:1139– 49.
[87] Takagi H, Isoda T, Kusakabe K, Morooka S. Effects of coal [109] Wòjtowicz MA, Bassilakis R, Serio MA. Ammonia and
structures on denitrogenation during flash pyrolysis. Energy hydrogen cyanide release during coal pyrolysis at low
Fuels 1999;13:934– 40. heating rates. ACS Div Fuel Chem Prepr 2001;46:133–7.
[88] Stenseng M. Pyrolysis and combustion of biomass. Depart- [110] Ghani UM, Wendt JOL. Proc Combust Inst 1990;23:1281–8.
ment of Chemical Engineering, Technical University of [111] Tan LL, Li C-Z. Fuel 2000;79:1883–9.
Denmark; 2001. [112] Tan LL, Li C-Z. Fuel 2000;79:1891–7.
[89] Scott DS, Piskorz J, Westerberg IB, McKeough P. Flash [113] Staǹczyk K, Boudou JP. Fuel 1994;73:940.
pyrolysis of peat in a fluidized bed. Fuel Process Technol [114] Ohtsuka Y, Mori H, Watanabe T, Asami K. Fuel 1994;73:
1988;18:81– 95. 1093.
[90] Storm C, Rüdiger H, Spliethoff H, Hein KRG. Co-pyrolysis [115] Wu Z, Ohtsuka Y. Remarkable formation of N2 from a
of coal/biomass and coal/sewage sludge mixtures. J Engng Chinese lignite during coal pyrolysis. Energy Fuels 1996;10:
Gas Turbine Power Trans ASME 1999;121:55–63. 1280–1.
[91] Wendt JOL, Pershing DW. Combust Sci Technol 1977;16: [116] Ohtsuka Y, Zhiheng W, Furimsky E. Effect of alkali and
111. alkaline earth metals on nitrogen release during temperature
[92] Zhang H, Fletcher TH. Nitrogen transformations during programmed pyrolysis of coal. Fuel 1997;76:1361–7.
secondary coal pyrolysis. Energy Fuels 2001;15:1512 –22.
[117] Kidena K, Hirose Y, Aibara T, Murata S, Nomura M.
[93] Niksa S. Proc Combust Inst 1994;25:537.
Analysis of nitrogen-containing species during pyrolysis of
[94] Yu LE, Hildemann LM, Niksa S. Characteristics of nitrogen-
coal at two different heating rates. Energy Fuels 2000;14:
containing aromatic compounds in coal tars during secondary
184 –9.
pyrolysis. Fuel 1999;78:377 –85.
[118] Tsubouchi N, Ohshima Y, Xu C, Ohtsuka Y. Enhancement of
[95] Perry ST, Fletcher TH, Solum MS, Pugmire RJ. Modeling
N2 formation from the nitrogen in carbon and coal by
nitrogen evolution during coal pyrolysis based on a global
calcium. Energy Fuels 2001;15:158–62.
free-radical mechanism. Energy Fuels 2000;14:1094–102.
[119] Mori H, Asami K, Ohtsuka Y. Role of iron catalyst in fate of
[96] Song YH, Pohl JH, Beèr JM, Sarofim AF. Nitric oxide
fuel nitrogen during coal pyrolysis. Energy Fuels 1996;10:
formation during pulverized coal combustion. Combust Sci
1022–7.
Technol 1982;28:31.
[120] Wu Z, Ohtsuka Y. Key factors for formation of N2 from low
[97] Li C-Z, Nelson PF, Ledesma EB, Mackie JC. An exper-
rank coals during fixed bed pyrolysis: pyrolysis conditions
imental study of the release of nitrogen from coals pyrolyzed
and inherent minerals. Energy Fuels 1997;11:902–8.
in fluidized-bed reactors. Proc Combust Inst 1996;26:
3205–11. [121] Niksa S. Energy Fuels 1995;9:467.
[98] Wornat MJ, Sarofim AF, Longwell JP. Proc Combust Inst [122] Perry S, Hambly EM, Fletcher TH, Solum MS, Pugmire RJ.
1988;22:135 –43. Solid-State 13C NMR characterization of matched tars and
[99] Niksa S, Cho S. Conversion of fuel-nitrogen in the primary chars from rapid coal devolatilization. Proc Combust Inst
zones of pulverized coal flames. Energy Fuels 1996;10: 2000;28:2313–9.
463–73. [123] Nelson PF, Kelly MD, Wornat MJ. Conversion of fuel
[100] Genetti D, Fletcher TH. Modeling nitrogen release during nitrogen in coal volatiles to NOx precursors under rapid
devolatilization on the basis of chemical structure of coal. heating conditions. Fuel 1991;70:403 –7.
Combust Flame 1998;115:1– 27. [124] Li C-Z, Buckley AN, Nelson PF. Effects of temperature and
[101] Nelson PF, Li C-Z, Ledesma E. Formation of HNCO from molecular mass on the nitrogen functionality of tars produced
the rapid pyrolysis of coals. Energy Fuels 1996;10:264–5. under high heating rate conditions. Fuel 1998;77:157–64.
[102] Lin WC, Wendt JOL, Cole JA, Rumbaugh JE. Acetylene [125] Ledesma EB, Li C-Z, Nelson PF, Mackie JC. Release of
interference with hydrogen cyanide determination by HCN, NH3 and HNCO from the thermal gas-phase cracking
selective ion electrode. Environ Sci Technol 1994;28: of coal pyrolysis tars. Energy Fuels 1998;12:536 –41.
1394–6. [126] Axworthy AE, Dayan VH, Martin GB. Reactions of fuel-
[103] Belson DJ, Strachan AN. Chem Soc Rev 1982;11:41. nitrogen compounds under conditions of inert pyrolysis. Fuel
[104] Chen SL, Heap MP, Pershing DW, Martin GB. Influence of 1978;57:29–35.
coal composition on the fate of volatile and char nitrogen [127] Mackie JC, Colket III MB, Nelson PF, Esler M. Shock tube
during combustion. Proc Combust Inst 1982;19:1271 –80. pyrolysis of pyrrole and kinetic modeling. Int J Chem Kinet
[105] Bose AC, Dannecker KM, Wendt JOL. Composition effects 1991;23:733–60.
on mechanisms governing the destruction of NO and other [128] Cheng MT, Kirsch MJ, Lester TW. Reaction of nitric oxide
nitrogenous species during fuel rich combustion. Energy with bound carbon at flame temperatures. Combust Flame
Fuels 1988;2:301–8. 1989;77:213–7.
[129] von Gersum S, Roth P. Proc Combust Inst 1992;24: [148] Bacskay GB, Martoprawiro M, Mackie JC. The thermal
999–1006. decomposition of pyrrole: an ab initio quantum chemical
[130] Axworthy AE. Chemistry and kinetics of fuel nitrogen study of the potential energy surface associated with the
conversion to nitric oxide. AIChE Symp Ser 71 1975;148: hydrogen cyanide plus propyne channel. Chem Phys Lett
43–50. 1999;300:321–30.
[131] Sugiyama S, Arai N, Hasatani M, Kawamura S. Reducing [149] Martoprawiro M, Bacskay G, Mackie JC. Ab initio quantum
fuel NOx emission: N2 from fuel N. Environ Sci Technol chemical and kinetic modeling study of the pyrolysis kinetics
1978;12:175 –80. of pyrrole. J Phys Chem 1999;103:3923– 34.
[132] Houser TJ, Hull M, Alway RM, Biftu T. Kinetics of [150] Zhai L, Zhou X, Liu R. A theoretical study of pyrolysis
formation of HCN during pyridine pyrolysis. Int J Chem mechanisms of pyrrole. J Phys Chem A 1999;103:3917–22.
Kinet 1980;12:569–74. [151] Lifshitz A, Shweky I, Tamburu C. Thermal decomposition of
[133] Bruinsma OSL, Tromp PJJ, de Sauvage Nolting HJJ, N-methylpyrrole: experimental and modeling study. J Phys
Moulijn JA. Gas phase pyrolysis of coal-related aromatic Chem 1993;97:4442–9.
compounds in a coiled tube flow reactor. 2. Hetero-cyclic [152] Laskin A, Lifshitz A. Isomerization and decomposition of
compounds, their benzo and dibenzo derivatives. Fuel indole. Experimental results and kinetic modeling. J Phys
1988;67:334 –40. Chem A 1997;101:7787–801.
[134] Mackie JC, Colket III MB, Nelson PF. Shock tube pyrolysis [153] Houser TJ, McCarville ME, Houser BD. Kinetics of
of pyridine. J Phys Chem 1990;94:4099 –106. oxidation of pyridine in a flow system. Combust Sci Technol
[135] Doughty A, Mackie JC. Kinetics of thermal decomposition of 1982;27:183–91.
the diazines: shock-tube pyrolysis of pyrimidine. J Chem Soc [154] Houser TJ, McCarville ME, Zhuo-Ying G. Nitric oxide
Faraday 1994;90:541– 8. formation from fuel-nitrogen model compound oxidation.
[136] Jones J, Bacskay GB, Mackie JC, Doughty A. Ab initio Fuel 1988;67:642–50.
studies of the thermal decomposition of azoaromatics: free [155] Miller JA, Branch MC, McLean WJ, Chandler DW, Smooke
radical versus intramolecular mechanism. J Chem Soc MD, Kee RJ. On the conversion of HCN to NO and N2 in
Faraday Trans 1995;91:1587– 92.
H2 –O2 –HCN –Ar flames at low pressure. Proc Combust Inst
[137] Memon HUR, Bartle KD, Taylor JM, Williams A. The shock
1984;20:673–84.
tube pyrolysis of pyridine. Int J Energy Res 2000;24:
[156] Murakami N, Tokunaga K, Sakai M. Formation of NO by the
1141–59.
oxidation of HCN. Nenryo Kyokaishi 1982;61:759 –69.
[138] Ninomiya Y, Dong Z, Suzuki Y, Koketsu J. Theoretical study
[157] Hulgaard T, Dam-Johansen K. Homogeneous nitrous oxide
on the thermal decomposition of pyridine. Fuel 2000;79:
formation and destruction under combustion conditions.
449–57.
AIChE J 1993;39:1342 –54.
[139] Ikeda E, Nicholls P, Mackie JC. A kinetic study of the
[158] Glarborg P, Miller JA. Mechanism and modeling of
oxidation of pyridine. Proc Combust Inst 2000;28:1709– 16.
hydrogen cyanide oxidation in a flow reactor. Combust
[140] Doughty A, Mackie JC. Kinetics of pyrolysis of a coal model
Flame 1994;99:475 –83.
compound 2-picoline, the nitrogen heteroaromatic analogue
[159] Wargadalam VJ, Löffler G, Winter F, Hofbauer H.
of toluene. 2. The 2-picolyl radical and kinetic modeling.
Homogeneous formation of NO and N2O from the oxidation
J Phys Chem 1992;39:10339–48.
of HCN and NH3 at 600–1000 8C. Combust Flame 2000;
[141] Terentis A, Doughty A, Mackie JC. Kinetics of pyrolysis of a
coal model compound 2-picoline, the nitrogen heteroaro- 120:465– 78.
matic analogue of toluene. 1. Product distributions. J Phys [160] Flatness SA, Kramlich JC. Measurement of the branching
Chem 1992;96:10334–9. ratio of NCO þ NO into N2O at 1100 – 1400 K. Proc
[142] Hämäläinen JP, Aho MJ, Tummavuori JL. Formation of Combust Inst 1996;25:567– 73.
nitrogen oxides from fuel-N through HCN and NH3: a model- [161] Dagaut P, Lecomte F, Chevalier S, Cathonnet M. The
compound study. Fuel 1994;73:1894– 8. oxidation of HCN and reactions with nitric oxide: exper-
[143] Laskin A, Lifshitz A. Thermal decomposition of quinoline imental and detailed kinetic modeling. Combust Sci Technol
and isoquinoline. The role of 1-indene imine radical. J Phys 2000;155:105–27.
Chem A 1998;102:928 –46. [162] Higashihara T, Saito K, Murakami I. Oxidation of hydrogen
[144] Doughty A, Mackie JC, Palmer JM. Kinetics of the thermal cyanide in shock waves. Formation of nitrogen monoxide.
decomposition and isomerization of pyrazine (1,4 diazine). J Phys Chem 1983;87:3707–12.
Proc Combust Inst 1994;25:893–900. [163] Fenimore CP, Jones GW. Oxidation of ammonia in flames.
[145] Kiefer JH, Zhang Q, Kern RD, Kato K. Relation between J Phys Chem 1961;65:298–303.
functional forms of coal nitrogen and formation of NOx [164] MacLean DI, Wagner HGg. The structure of the reaction
precursors during rapid pyrolysis. Energy Fuels 1993;7: zones of ammonia–oxygen and hydrazine decomposition
1013–20. flames. Proc Combust Inst 1967;11:871–8.
[146] Lifshitz A, Tamburu C, Suslensky A. J Phys Chem 1989;93: [165] Blint RJ, Dasch CJ. Formation of NO and N2 from NH3 in
5802. flames. ACS Symp Ser 1984;249:87–101.
[147] Bacskay GB, Martoprawiro M, Mackie JC. An ab initio [166] Dasch CJ, Blint RJ. A mechanistic and experimental study of
quantum chemical study of the electronic structure and ammonia in flames. Combust Sci Technol 1984;41:223–44.
stability of the pyrrolyl radical: comparison with the [167] Dean AM, Chou M-S, Stern D. Nitrogen chemistry in flames.
isoelectronic cyclopentadienyl radical. Chem Phys Lett Observations and detailed chemistry. ACS Symp Ser 1984;
1998;290:391 –8. 249:71–86.
[168] Dean AM, Chou M-S, Stern D. Kinetics of rich ammonia [190] Lyon RK, Cole JA. A reexamination of the RapreNOx
flames. Int J Chem Kinet 1984;16:633 –53. process. Combust Flame 1990;82:435–43.
[169] Zabielsky MF, Seery DJ. High temperature measurements of [191] Miller JA, Bowman CT. Kinetic modeling of the reduction of
the rate of the reaction of OH with NH3. Int J Chem Kinet nitric oxide in combustion products by isocyanic acid. Int J
1991;17:1191 –9. Chem Kinet 1991;23:289 –313.
[170] Bian J, Vandooren J, van Tiggelen PJ. Experimental study of [192] Lifshitz A, Bidani M, Agranat A, Suslensky A. Thermal
the structure of an ammonia– oxygen flame. Proc Combust reactions of pyrrolidine at elevated temperatures. Studies
Inst 1986;21:953–63. with a single-pulse shock tube. J Phys Chem 1987;91:
[171] Sausa RC, Singh G, Lemire GW, Anderson WR. Molecular 6043–8.
beam mass spectrometric and modeling studies of neat and [193] Bartle KD, Taylor JM, Williams A. Release of nitrogen
NH3-doped low-pressure H2/N2O/Ar flames: formation and compounds from coal during high temperature pyroprobe gas
consumption of NO. Prog Energy Combust Sci 1992;18: chromatography with selective detection. Fuel 1992;71:
133–220. 714 –5.
[172] Takeyama T, Miyama H. A shock-tube study of the ammonia– [194] Ikeda E, Mackie JC. A kinetic study of the oxidation of
oxygen reaction. Proc Combust Inst 1967;11:845 –52. acetonitrile: a model for NO formation from fuel-bound
[173] Takeyama T, Miyama H. Kinetic studies of ammonia nitrogen. Proc Combust Inst 1996;26:597– 604.
oxidation in shock waves. II. The rate of ammonia [195] Lifshitz A, Tamburu C, Carroll HF. Thermal ignition of
consumption. Bull Chem Soc Jpn 1968;41:1761–5. acetonitrile, experimental results and kinetic modeling. Int J
[174] Bull DC. A shock tube study of the oxidation of ammonia. Chem Kinet 1997;29:839 –49.
Combust Flame 1968;12:603– 10. [196] Lifshitz A, Tamburu C. Thermal decomposition of aceto-
[175] Miyama H. Kinetic studies of ammonia oxidation in shock nitrile, kinetic modeling. Int J Chem Kinet 1998;30:341–7.
waves. IV. Comparison of induction periods for the ignition [197] Sendt K, Ikeda E, Bacskay GB, Mackie JC. Ab initio
of NH3 –O2 –N2 with those for NH3 – O2 –Ar mixtures. Bull quantum chemical and experimental (shock tube) studies of
Chem Soc Jpn 1968;41:1761– 5. the pyrolysis kinetics of acetonitrile. J Phys Chem A 1999;
[176] Miyama H. Ignition of ammonia–oxygen mixtures by shock 103:1054– 72.
waves. J Chem Phys 1968;48:1421–2. [198] Lifshitz A, Bidani M, Suslensky A, Tamburu C. Pyrolysis of
[177] Bradley JN, Butlin RN, Lewis D. Oxidation of ammonia in acrylonitrile at elevated temperatures. Studies with a single-
shock waves. Trans Faraday Soc 1968;64:71 –8. pulse shock tube. J Phys Chem 1989;93:1369 –73.
[178] Drummond LJ. High temperature oxidation of ammonia. [199] Lifshitz A, Cohen Y, Braun-Unkhoff M, Frank P. Thermal
Combust Sci Technol 1972;5:175–82. decomposition of benzonitrile: a combined single pulse
[179] Fujii N, Miyama H, Koshi M, Asaba T. Kinetics of ammonia shock tube—ARAS investigation. Proc Combust Inst 1996;
oxidation in shock waves. Proc Combust Inst 1981;18:873– 83. 26:659–67.
[180] Salimian S, Hanson RK, Kruger CH. Ammonia oxidation in [200] Glarborg P, Alzueta MU, Dam-Johansen K, Miller JA.
shock-heated NH3 – N2O – Ar mixtures. Combust Flame Kinetic modeling of hydrocarbon/nitric oxide interactions in
1984;56:83– 95. a flow reactor. Combust Flame 1998;115:1–27.
[181] Dean AM, Hardy JE, Lyon RK. Kinetics and mechanism of [201] Pfefferle L-D, Churchill SW. NOx production from the
NH3 oxidation. Proc Combust Inst 1982;19:97–105. combustion of ethane doped with ammonia in a thermally
[182] Miller JA, Smooke MD, Green RM, Kee RJ. Kinetic stabilized plug flow burner. Combust Sci Technol 1986;49:
modeling of the oxidation of ammonia in flames. Combust 235 –49.
Flame 1983;34:179. [202] Martin RJ, Brown NJ. Nitrous oxide formation and
[183] Lindstedt RP, Lockwood FC, Selim MA. A detailed kinetic destruction in lean, premixed combustion. Combust Flame
study of ammonia oxidation. Combust Sci Technol 1995; 1990;80:238 –55.
108:231–54. [203] Williams BA, Fleming JW. Radical species profiles in low-
[184] Dean AM, Bozzelli JW. Combustion chemistry of nitrogen. pressure methane flames containing fuel nitrogen com-
In: Gardiner WC, editor. Gas phase combustion chemistry. pounds. Combust Flame 1997;110:1 –13.
New York: Springer; 2000 [chapter 2]. [204] Sarofim AF, Williams GC, Modell M, Slater SM. Conversion
[185] Lyon RK. Kinetics and mechanism of thermal DeNOx: a of fuel nitrogen to nitric oxide in premixed and diffusion
review. 194th Annu ACS Meet; Div Fuel Chem 1987;32:433. flames. AIChE Symp Ser 71 1975;148:51–61.
[186] Glarborg P, Dam-Johansen K, Miller JA, Kee RJ, Coltrin [205] Fenimore CP. Studies of fuel-nitrogen species in rich flame
ME. Modeling the thermal DENOx process in flow reactors. gases. Proc Combust Inst 1979;17:661– 70.
Surface effects and nitrous oxide formation. Int J Chem Kinet [206] Morley C. The mechanism of NO formation from nitrogen
1994;26:421 –36. compounds in hydrogen flames studied by laser fluorescence.
[187] Miller JA, Glarborg P. Modeling the thermal De –NOx Proc Combust Inst 1981;18:23– 32.
process: closing in on a final solution. Int J Chem Kinet 1999; [207] Fenimore CP. Reactions of fuel-nitrogen in rich flame gases.
31:757–65. Combust Flame 1976;26:249–56.
[188] Glarborg P, Kristensen PG, Jensen SH, Dam-Johansen K. A [208] Tang S-K, Churchill SW. The formation of thermal and fuel
flow reactor study of HNCO oxidation chemistry. Combust NOx for radiantly stabilized combustion. Proc Combust Inst
Flame 1994;98:241 –58. 1981;18:73.
[189] Caton JA, Siebers DL. Comparison of nitric oxide removal [209] Wendt JOL, Sternling CV. Effect of ammonia in gaseous
by cyanuric acid and ammonia. Combust Sci Technol 1989; fuels on nitrogen oxide emissions. J Air Pollut Control Assoc
65:277–93. 1974;24:1055–8.
[210] Haynes BS, Iverach D, Kirov NY. The behavior of nitrogen [232] Yue GX, Pereira FJ, Sarofim AF, Beèr JM. Char nitrogen
species in fuel-rich hydrocarbon flames. Proc Combust Inst conversion to NOx in a fluidized bed. Combust Sci Technol
1974;15:1103 –12. 1992;83:245–56.
[211] Haynes BS. Reactions of ammonia and nitric oxide in the [233] Wallman PH, Carlsson RCJ, Leckner B. In: Rubow L,
burnt gases of fuel-rich hydrocarbon –air flames. Combust Commonwealth G, editors. NOx reduction in pressurized
Flame 1977;28:81 –91. fluidized-bed combustion. 12th International Conference on
[212] Foster DE, Keck JC. Factors affecting NOx formation from Fluidized Bed Combustion, New York: ASME; 1993. p.
nitrogen containing fuels. Combust Flame 1980;38: 789 –94.
199–209. [234] Richard JR, Majthoub MA, Aho M, Pirkonen PM. The effect
[213] Kristensen PG, Glarborg P, Dam-Johansen K. Nitrogen of pressure on the formation of nitrogen oxides from coal
chemistry during burnout in fuel-staged combustion. Com- char combustion in a small fixed bed reactor. Fuel 1994;73:
bust Flame 1996;107:211 –22. 1034–8.
[214] Baumann H, Schuler J, Schiller R. Drahtnetzversuche zum [235] Heurtebise C, Richard JR. The effect of pressure on the
Verbleib des Kohlestickstoffs bei der Pyrolyse und Ver- emission of N2O and NO from charcoal. Proceedings of the
brennung. Erdol Erdgas Kohle 1989;105:227–32. Sixth International Workshop on Nitrous Oxide Emissions,
[215] Ninomiya Y, Yokoi K, Arai N, Hasatani M. Characteristics Turku, Finland; 1994. p. 345 –66.
of emission of char NO during the combustion of a single [236] Broden H. Dynamic single particle char combustion and its
particle of coal char. Int Chem Engng 1989;29:512–6. influence on the fate of fuel bound nitrogen. Chalmers
[216] Miettinen H, Paulsson M, Strömberg D. Laboratory study of University of Technology; 1994.
N2O formation from burning char particles at FBC [237] Cahill P, Smith M, Vallender SJ. Characterisation of British
conditions. Energy Fuels 1995;9:10–19. coals for low NOx combustion. Proceedings of the Inter-
[217] Jensen LS, Jannerup HE, Glarborg P, Jensen A, Dam- national Conference on Coal Science, London: Butterworth/
Johansen K. Experimental investigation of NO from Heinemann; 1991. p. 319– 22.
pulverized char combustion. Proc Combust Inst 2000;28: [238] de Andres AIG, Thomas KM. The influence of mineral
2271–8. matter and carbonization conditions on nitrogen release
[218] Lazaro MJ, Ibarra JV, Moliner R, de Andres AG, Thomas during coal combustion. Fuel 1994;73:635 –41.
KM. The release of nitrogen during the combustion of coal [239] Wang W, Brown SD, Thomas KM, Crelling JC. Nitrogen
chars: the role of volatile matter and surface area. Fuel 1996; release from a rank series of coals during temperature
75:1014–24. programmed combustion. Fuel 1994;73:341– 7.
[219] Ashman PJ, Haynes BS, Buckley AN, Nelson PF. The fate of [240] Hindmarsh JC, Wang W, Thomas KM, Crelling JC. The
char-nitrogen in low temperature oxidation. Proc Combust release of nitrogen during the combustion of macerals,
Inst 1998;27:3069–75. microlithotypes and their chars. Fuel 1994;73:1229–34.
[220] Song YH, Beèr JM, Sarofim AF. Oxidation and devolatiliza- [241] Arenillas A, Rubiera F, Pis JJ, Jones JM, Williams A. The
tion of nitrogen in coal char. Combust Sci Technol 1982;28: effect of textural properties of bituminous coal chars on NO
177. emissions. Fuel 1999;78:1779 –85.
[221] Coda B, Kluger F, Fortsch D, Spliethoff H, Hein KRG, [242] Tullin CJ, Goel S, Morihara A, Sarofim AF, Beèr JM. NO and
Tognotti L. Coal-nitrogen release and NOx evolution in air- N2O formation for coal combustion in a fluidized bed: effect
staged combustion. Energy Fuels 1998;12:1322–7. of carbon conversion and bed temperature. Energy Fuels
[222] Aihara T, Matsuoka K, Kyotani T, Tomita A. Mechanism of 1993;7:796–802.
N2 formation during coal char oxidation. Proc Combust Inst [243] Furusawa T, Shimizu T, Kawaguchi H, Kojima T, Chihara Y.
2000;28:2189 –95. Conversion ratio of fuel bond nitrogen of a single coal
[223] Ashman PJ, Haynes BS, Nicholls PM, Nelson PF. Inter- particle to nitric oxide during fluidized bed combustion,
actions of gaseous NO with char during the low temperature experimental and theoretical investigations. Proceedings of
oxidation of coal chars. Proc Combust Inst 2000;28:2171 –9. the International Conference on Coal Science, New York:
[224] Dragheim J, Johnsson JE, Jensen AD. Unpublished results. Elsevier; 1987. p. 853 –6.
[225] Winther F, Wartha C, Löffler G, Hofbauer H. Proc Combust [244] de Soete GG. Heterogeneous N2O and NO formation from
Inst 1996;26:3325–34. bound nitrogen atoms during coal char combustion. Proc
[226] Wang W, Thomas KM. The release of nitrogen species from Combust Inst 1990;23:1257– 64.
carbons during gasification: models for coal char gasification. [245] Wójtowicz MA, Lohuis JAO, Tromp PJJ, Moulijn JA. In:
Fuel 1992;71:871– 7. Anthony EJ, editor. N2O formation in fluidised bed
[227] Wang W, Brown SD, Hindmarsh J, Thomas KM. NOx release combustion of coal. Proceedings of the 11th International
and reactivity of chars from a wide range of coals during Conference on Fluidized Bed Combustion, New York:
combustion. Fuel 1994;73:1381–8. ASME; 1991. p. 1013–20.
[228] Jones JM, Harding AW, Brown SD, Thomas KM. Carbon [246] Goel S, Morihara A, Tullin CJ, Sarofim AF. Effect of NO and
1995;33:833 –43. O2 concentration on N2O formation during coal combustion
[229] Nicholls PM, Nelson PF. Energy Fuels 2000;14:943–4. in a fluidized-bed combustor: modeling results. Proc
[230] Goel S, Zhang B, Sarofim AF. NO and N2O formation during Combust Inst 1994;25:1051– 9.
char combustion: is it HCN or surface attached nitrogen? [247] Klein M, Rotzoll G. Formation of N2O and NO during
Combust Flame 1996;104:213 –7. fluidized bed combustion of a single coal and char particle.
[231] Chambrion P, Kyotani T, Tomita A. Proc Combust Inst 1998; Proceedings of the Sixth International Workshop on Nitrous
27:3053. Oxide Emissions, Turku, Finland; 1994. p. 239–53.
[248] Aho MJ, Paakinen KM, Pirkonen PM, Kilpinen P, Hupa M. [268] Commandrè J-M, Stanmore BR, Salvadore S. The high
The effects of pressure oxygen partial pressure, and temperature reaction of carbon with nitric oxide. Combust
temperature on the formation of N2O, NO, and NO2 from Flame 2002;128:211–6.
pulverized coal. Combust Flame 1995;102:387– 400. [269] Wang W, Thomas KM, Cai HY, Dugwell DR, Kandiyoti R.
[249] Boavida DH, Lobo LS, Gulyurtlu IK. In: Heinschel KJ, NO release and reactivity of chars during combustion: the
editor. N2O formation during the first phase of coal effect of devolatilization temperature and heating rate.
combustion in a fluidised bed. Proceedings of the 13th Energy Fuels 1996;10:409–16.
International Conference on Fluidized Bed Combustion, New [270] Hurt RH, Gibbins JR. Residual carbon from pulverized coal
York: ASME; 1995. p. 881–6. fired boilers. 1. Size distribution and combustion reactivity.
[250] Croiset E, Heurtebise C, Rouan J-P, Richard J-R. Influence of Fuel 1995;74:471–9.
pressure on the heterogeneous formation and destruction of [271] Beeley T, Crelling J, Gibbins J, Hurt RH, Man C, Williamson
nitrogen oxides during char combustion. Combust Flame J, Yang N. Transient high-temperature thermal deactivation
1998;112:33 –44. of monomaceral-rich coal chars. Proc Combust Inst 1996;26:
[251] Krammer G, Sarofim AF. Combust Flame 1994;97:118– 24. 3103–10.
[252] Miettinen H, Abul-Milh M. N2O formation from combustion [272] Senneca O, Russo P, Salatino P, Masi S. The relevance of
of char particles with NO. Energy Fuels 1996;10:421– 4. thermal annealing to the evolution of coal char gasification
[253] Miettinen H. Laboratory study of N2O formation from 15N- reactivity. Carbon 1997;35:141 –51.
isotope-marked NO while burning char particles at FBC [273] Hurt RH, Sun JK, Lunden M. A kinetic model of carbon
conditions. Energy Fuels 1996;10:197–202. burnout in pulverized coal combustion. Combust Flame
[254] Suzuki Y, Moritomi H, Kido N. Proceedings of the Fifth 1998;113:181–97.
International Workshop on Nitrous Oxide Emissions, [274] Senneca O, Salatino P, Masi S. Microstructural changes and
Tsukuba, Japan; 1992 [paper 5-4]. loss of thermal reactivity of chars upon heat treatment. Fuel
[255] Winther F, Wartha C, Hofbauer H. Proceedings of the 14th 1998;77:1483–93.
International Conference on Fluidized Bed Combustion. New [275] Shim H, Hurt RH. Thermal annealing of chars from diverse
York: ASME; 1997. p. 1131. organic precursors under combustion-like conditions. Energy
[256] Winther F, Löffler G, Wartha C, Hofbauer H, Anthony E. Can Fuels 2000;14:340–8.
J Chem Engng 1999;77:275. [276] Zolin A, Jensen AD, Dam-Johansen K. Kinetic analysis of
[257] Pels JR, Wójtowicz MA, Moulijn JA. Rank dependence of char thermal deactivation. Proc Combust Inst 2000;28:
N2O emission in fluidized-bed combustion of coal. Fuel 2181–8.
1993;72:373 –9. [277] Zolin A, Jensen AD, Dam-Johansen K. Coupling thermal
[258] Goel S, Molina A, Sarofim AF. Factors influencing the time- deactivation with oxidation for predicting the combustion of
resolved evolution of NOx HCN and N2O during char a solid fuel. Combust Flame 2001;125:1341–60.
oxidation at fluidized bed conditions. Energy Fuels 2002;16: [278] Zolin A, Jensen AD, Dam-Johansen K. The influence of
823–30. inorganic materials on the thermal deactivation of fuel chars.
[259] Liu H, Kojima T, Feng B, Liu D, Lu J. Effect of Energy Fuels 2001;15:1110–22.
heterogeneous reactions of coal char on nitrous oxide [279] Zolin A, Jensen AD, Dam-Johansen K. Experimental study
formation and reduction in a circulating fluidized bed. of char thermal deactivation. Fuel 2002;81:1065–75.
Energy Fuels 2001;15:696–701. [280] Molina A, Eddings EG, Pershing DW, Sarofim AF.
[260] Aarna I, Suuberg EM. The role of carbon monoxide in the Reduction of nitric oxide on the char surface at pulverized
NO– carbon reaction. Energy Fuels 1999;13:1145– 53. combustion conditions. Proc Combust Inst 2002;29.
[261] Johnsson JE, Dam-Johansen K. In: Anthony EJ, editor. [281] Calo JM, Suuberg EM, Aarna I. The role of surface area in
Formation and reduction of NOx in a fluidized bed the NO– carbon reaction. Energy Fuels 1999;13:761–2.
combustor. Proceedings of the 11th International Conference [282] Aarna I, Suuberg EM. Proc Combust Inst 1998;27:3061 –8.
on Fluidized Bed Combustion, New York: ASME; 1991. p. [283] Zevenhoven R, Hupa M. The reactivity of chars from coal,
1389–96. peat and wood towards NO, with and without CO. Fuel 1998;
[262] Suzuki Y, Moritomi H, Kido N. On the formation mechanism 77:1169–76.
of N2O during circulating fluidized bed combustion. [284] Tullin CJ, Sarofim AF, Beèr JM. Formation of NO and N2O
Proceedings of the Fourth SCEJ Symposium on Circulating in coal combustion: the relative importance of volatile and
Fluidized Bed Combustion; 1991. char nitrogen. J Inst Energy 1973;66:207–15.
[263] Song YH, Beèr JM, Sarofim AF. Reduction of nitric oxide by [285] Lu Y, Jahkola A. Fuel nitrogen behavior during devolatiliza-
coal char at temperatures of 1250–1750 K. Combust Sci tion and char oxidation under pressurized fluidized bed
Technol 1981;25:237–40. combustion conditions. In: Xu X, Zhou L, editors. Proceed-
[264] Guo F, Hecker W. Effect of CaO and burnout on the kinetics ings of the Third International Symposium on Coal
of NO reduction by Beulah Zap char. Proc Combust Inst Combustion, Science and Technology, Turku, Finland;
1996;26:2251. 1995. p. 457 –66.
[265] Sørensen CO, Johnsson JE, Jensen A. Reduction of NO over [286] Nelson PF, Nancarrow PC, Bus J, Prokopiuk A. Fractional
biomass char. Energy Fuels 2001;15:1359– 68. conversion of char N to NO in an entrained flow reactor. Proc
[266] Jensen AD, Johnsson JE. Effective diffusion coefficients in Combust Inst 2002;29.
coal chars. Proc Combust Inst 2000;28:2353–9. [287] Pereira FJ, Beèr JM. A mathematical model of NO
[267] Hurt RH. Structure, properties, and reactivity of solid fuels. formation and destruction in fluidized bed combustion of
Proc Combust Inst 1998;27:2887–904. coal fluidization. In: Davidson JF, Kearns DL, editors.
Proceedings of the Second Engineering Foundation Con- [295] Peck RE, Altenkirch RA, Midkieff KC. Fuel-nitrogen
ference; 1978. p. 401–7. transformations in one-dimensional coal-dust flames. Com-
[288] Furosawa T, Honda T, Takano J, Kunii D. Abatement of bust Flame 1984;55:331.
nitric oxide emission in fluidized bed combustion of coal. [296] Midkieff KC, Altenkirch RA. Particle size effects on the
J Chem Engng Jpn 1978;11:377–84. distribution of fuel nitrogen in one-dimensional coal-dust
[289] Konttinen J, Kilpinen P, Kallio S. Validation of a single flames. Proc Combust Inst 1986;21:1189–98.
particle model for char carbon and char nitrogen oxidation [297] Midkieff KC, Altenkirch RA, Peck RE. Stoichiometry and
under fluidised bed combustion conditions: a single coal-type effects on homogeneous vs heterogeneous combus-
particle modelling. 16th International Conference on tion in combustion environment. Combust Flame 1986;64:253.
Fluidized Bed Combustion, Reno, May 13 – 16, New [298] Kramlich JC, Seeker WR, Samuelsen GS. Observations of
York: ASME; 2001. chemical effects accompanying pulverized coal thermal
[290] Visiona SP, Stanmore BR. Modeling NOx release from a decomposition. Fuel 1988;67:1182–9.
single coal particle. II. Formation NO from char nitrogen. [299] Sullivan NS, Jensen A, Glarborg P, Day MS, Grcar JF, Bell
Combust Flame 1996;106:207 –18. JB, Pope CJ, Kee RJ. NOx formation and ammonia
[291] Østberg M, Glarborg P, Jensen A, Johnsson JE, Pedersen LS, conversion in laminar coflowing nonpremixed methane–air
Dam-Johansen K. A model of the coal reburning process. flames. Combust Flame 2002; 131:285–98.
Proc Combust Inst 1998;27:3027–35. [300] Kremer H, Mechenbier R, Schulz W. In: Lahaye L, Prado G,
[292] Peck RE, Glarborg P, Johnsson JE. Kinetic modeling of editors. Fundamentals of the physical-chemistry of pulver-
nitrogen oxide formation in one-dimensional pulverized coal ized coal combustion. NATO Series E, vol. 137. 1987.
flames. Combust Sci Technol 1991;76:81– 109. p. 304–18.
[293] Altenkirch RA, Mackowski DW, Peck RE, Tong TW. Effect [301] Cho S, Marlowe D, Niksa S. Burning velocities of multi-
of soot on pyrometer measured temperatures in pulverized component organic fuel mixtures derived from various coals.
coal flames. Combust Sci Technol 1984;41:327. Combust Flame 1995;101:399–410.
[294] Peck RE, Midkieff KC, Altenkirch RA. The evolution of [302] Okazaki K, Shishido H, Nishikawa T, Ohtake K. Separation
nitrogen from pulverized subbituminous coal burnt in a one- of the basic factors affecting NO formation in pulverized coal
dimensional flame. Proc Combust Inst 1984;20:1373 –80. combustion. Proc Combust Inst 1984;20:1381– 9.

Fuel Nitrogen Conversion in Solid Fuel Fired Systems

Uploaded by

Copyright:

Available Formats

Fuel Nitrogen Conversion in Solid Fuel Fired Systems

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Fuel Nitrogen Conversion in Solid Fuel Fired Systems

Uploaded by

Copyright:

Available Formats

Progress in Energy and Combustion Science 29 (2003) 89–113

Fuel nitrogen conversion in solid fuel fired systems

* Corresponding author. Tel.: þ 45-4525-2840; fax: þ 45-4588-2258.

7. Fuel-N conversion in laminar flames . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104

1. Introduction istics in various systems. A large part of this work is

aliphatic side chains are released and ring condensation

Soot formed from tar through secondary pyrolysis may

work in the past on gas-phase fuel nitrogen conversion has

Fig. 8. Reaction path diagram illustrating major steps in volatile-N

6.2. Formation of N2O from char-N 6.3. Reduction of NO on char

this range, referred to as the bulk group, were all obtained

oxidation. For bituminous coals the effect appears to be

6.4. Effect of reaction conditions on NO yield

Experiments under single particle conditions have shown

particles the char þ O2 reaction becomes diffusion-limited

Most of our understanding of fuel nitrogen conversion in

You might also like

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.