No formation and its reduction through co-flow methane reburn in a pulverised coal combustion proces

Abstract

Keywords: Pulverized Coal; Combustion; Equivalence Ratio; Co-flow Methane; Reburn

Introduction

Coal is the most abundant fuel available on earth. Therefore, it is economical to use in power generation and various metal extraction process. Oxides of nitrogen (NOx), Oxides of sulphur (SOx) and dry and bottom ash

are the major pollutants formed during coal combustion. Out of various NOx like NO, N2O and NO2, NO is the most hazardous and its fraction is higher as compared to other NOx [1]. Presence of nitrogen in coal is a major source of

*Corresponding author. *E-mail address: pghosefme@kiit.ac.in, pghoseju@gmail.com This paper was recommended for publication in revised form by Regional Editor Chandramohan VP Published by Yıldız Technical University Press, İstanbul, Turkey Copyright 2021, Yıldız Technical University. This is an open access article under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

NO. It (Fuel-NO) contributes around 50%-60% of total NO [2] for high grade coal. During combustion of high grade coal like bituminous coal, approximately 60% to 70% of inherent nitrogen is converted to NO [3]. Similarly, atmospheric nitrogen causes Thermal-NO formation by reacting with oxygen or hydroxide at a high temperature [4]. Another minor sources of NO are Prompt-NO and N2O Intermediate NO. It is reported that during bituminous coal combustion up to 30% of total NO may formed due to N2O intermediate NO [5]. However both Prompt-NO and N2O intermediate are relatively less important in industrial furnaces due to smaller quantity[6]. Various mechanism are adopted for NO reduction in industrial furnaces such as; Low NOx burner, Over-fire air, NO reburn, Flue gas recirculation, Flame cooling by water or steam injection etc. In low NO burner; the air-fuel mixture is either burnt in fuel-lean condition or fuel rich condition. In fuel-lean condition, excess air and in fuel-rich condition, excess fuel prevents the flame from overheating. As a result the Thermal-NO formation is reduced. On the other hand, with fuel-rich condition, presence of low oxygen minimizes the NO production. Low NOx burner technology is simple and effective and 40%-60% of NO can be reduced by adapting this methodology [7]. Overall equivalence ratio depends on leanness or richness of the mixture. If the mixture is either too lean or too rich, a low temperature flame exists due to the presence of un-burnt reactants. Hence, the Thermal-NO rate decreases. Moreover, rate of reaction for other NO reaction mechanism such as; Prompt-NO and Fuel-NO strongly depends upon temperature also. Therefore, the effect of overall equivalence ratio on NO formation in low NOx burner is an important parameter. Various research works are performed on low NOx burner for coal combustion to investigate the effect of equivalence ratio on NO formation and other combustion characteristics of flame [8, 9]. Hu et al. [10] intensively studied the impact of equivalence ratio on NO formation. They reported that, NOx emissions increased with increase in equivalence ratio under fuel-lean conditions, then declined dramatically after when equivalence ratio become 0.8. Fuel biasing is another simple method used to reduce NO. In this method either low nitrogen content fuel or nitrogen free fuel is used as secondary fuel. So that the Fuel-NO can be reduced up to a great extent. Biomass cofiring along with coal is used as a fuel biasing method to minimize NOx [11–13]. The nitrogen free gaseous fuel like methane or natural gas can be a good option of biasing fuel. It has many advantages, viz; (1) Methane is a nitrogen free fuel, so Fuel-NO will be reduced by adding this with coal. (2) Total ash will be reduced with the addition of methane. (3) It can be used as re-burn fuel (re-burn fuel should be highly volatile and reactive). NO re-burn is a methodology where generally either hydrocarbon fuel or ammonia are injected at downstream of the flame. Ammonia injection technology called selective catalytic reactor (SCR) is

expensive and required additional fittings. In addition to that ammonia in gaseous form is harmful. Su et al. [14] investigated the NOx destruction through numerical simulation by using pure methane as reburn fuel at downstream. Han et al. [15] used a premixed methane-ammonia blend as reburn fuel for NO reduction, whereas Zarnitz et al. [16] used coal volatile as reburn fuel at downstream of coal burner. Wu et al. [17] investigated the effect of syngas as reburn fuel on NO reduction using both co-flow injection and downstream injection methods. From the investigation it is observed that, 46% NO reduction has been obtained with co-flow reburn fuel injection, whereas by using conventional downstream injection method, only 23% NO reduction has been obtained. From the literature it has been observed that around 30%-40% of NO can be reduced by using reburn fuel at flame downstream. Some of the authors worked on co-flow methane injection in pulverized coal combustion systems experimentally [18] and computationally [19, 20]. But no one investigated the effect of methane on NO reduction rather they used methane as a secondary fuel only. Therefore in this work, a computational simulation is performed to investigate the effect of overall equivalence ratio on total NO formation and reduction. Thermal-NO, Prompt-NO and Fuel-NO mechanism are incorporated for NO formation calculation. For NO reduction, NO-reburn reaction mechanism is used, where co-flow methane is used as reburn fuel. Both Newlands bituminous coal-air and co-flow methane are injected from the same plane. The mass flow rate of coal and methane are calculated in such a way that, 50% of the total heat is contributed by coal and remaining 50% heat is contributed by methane reaction.

Computational Modelling

Coal Properties Newlands bituminous coal is used as base fuel. Various substance obtained from Proximate and Ultimate analysis along with the coal particle size are given in Table 1 [21]. Table 1. Proximate and Ultimate analysis data for Newlands Bituminous coal Proximate analysis [wt%]

Higher calorific value 29.1 MJ/kg Lower calorific value 28.1 MJ/kg Particle maximum diameter 60 µm Particle minimum diameter 5 µm Particle mean diameter 33 µm

The chemical formula for volatile is obtained from Proximate and Ultimate analysis as C4.61 H4.94 O0.46 N0.1219 S0.0155 [22]. Molecular weight of the volatile and high temperature volatile yielding factor is taken as 70 and 2.3 respectively [23]. Physical Model and Governing Equations A two dimensional geometry is prepared to simulate the combustion process through axi-symmetric burner as shown in Figure. 1 Through the central hole, coal is injected along with primary air. An annular hole is provided for coflow methane feeding in to the burner. A wide passage is given beyond the methane inlet for free flow of atmospheric air. Since in this work, solid coal particles are injected into the gas phase domain, the problem is solved under the Eulerian-Lgrangian frame of reference. Favre averaged [24] gas phase governing equations for conservation of mass, momentum and general scalar variable are given in Equation (1), (2) and (3) respectively.

∂ ∂  ∂ϕ  ρu jϕ = Γ φ ,eff + Sϕ , s  ∂x i ∂x i  ∂xi 

∂p ∂ ∂ ’’ ’’ + ρui u j = − τ ij − ρ u i u j + Ss ∂x i ∂x i ∂x i

Phi (φ) represents enthalpy in energy equation and species mass fraction for various chemical species in species transport equations. In energy conservation equation,

µ µ  Γ φ ,eff = Γ φ ,lam + Γ φ ,tur =  + t  where as in species con σ σt 

µ µ  servation equation; Γ φ ,eff = Γ φ ,lam + Γ φ ,tur =  + t  . .  Sc Sct 

To evaluate Reynolds stresses present in momentum equations, Boussinesq hypothesis is used as;

 ∂ui ∂u j  2 ’’ ’’  − ρ u + i u j = µt   − ρkδ ij  ∂x j ∂x i  3

k2 . In order to calculate eddy viscosity ε and Reynold stresses, transport equation for k, ε are solved by using Standard k-ε model [25]. Required boundary conditions are given at different boundaries for initializations of k and ε. The inter-phase source terms for different governing equations are discussed in following sections. Where, µt = ρCµ

Discrete Phase Model A finite number of coal particles having diameter range given in Table 1. are injected through primary air hole. The spread parameter of 4.02 is used to calculate particle diameter distribution by following Rosin-Rammler distribution scheme [26]. The turbulent interaction in between gas and particle phase are simulated by employing Discrete random walk model [27]. A force balance equation is solved to evaluate the drag force exerted on droplet by gas phase and vice versa is expressed as; du pi

2 π ρ p d ( j )  dt 8 ui − u pi ( j ) ui − u pi ( j )  Cdrag

In equation (5) the drag coefficient is calculated by following the spherical drag law [28], expressed as; a a Cdrag = a1 + 2 + 32 , where a1, a2, a3 are model constants Re Re

 ρ ui − u p ( j ) d j  i  is the Reynolds number conand Re  =   µ sidering the flow over spherical particles. A mass balance equation is solved to incorporate the exchange of mass in between two phases because of volatile evolution and char combustion is expressed as;

= hc Ap T − Tp + ε p Apσ θ R4 − Tp4 −  dm ( j )  dmchar ( j ) f h  vol ∆H v + ∆H c  dt  dt 

In the convective part of equation (7), the heat transfer coefficient is evaluated by using Nusselt number correlation suggested by Ranz and Marshal [29]. A large amount of heat is radiated from the flame due to high flame temperature. In addition to that, presence of soot in flame radiated significantly larger amount of heat from the flame. Considering bulk gas as participating media, Discrete Ordinate (DO) [30] radiation modelling method is employed to solve the radiative transport equation. From the solution, the radia . IdΩ   ∫ Ω = 4π  tive temperature θR is calculated as; θ R =  4σ   

A RTE (Radiative Transport Equation) is solved to evaluate radiation intensity. Weighted sum gray gas model (WSGGM) [31] is used to determine the bulk gas absorption coefficient. The absorption coefficient in gas phase includes the effect of absorptivity of soot as well. Scattering and radiative transmission is neglected in gas phase, whereas absorption and scattering both are considered for particle phase. The last term in the equation (7) is the heat transfer from particle surface due to volatile and char combustion. Khan and Greeves soot model [32] is used to include the effect of soot on field temperature. Reaction Modelling The rate of evolution of volatile matter is calculated with the help of single rate kinetic model [33] expressed as;

In equation (8), the kinetic rate coefficient is evaluated as; k = Ae–(E/RT), where pre exponential factor A = 4.747 × 105 and activation energy E = 7.4 × 106. Intrinsic char reaction model [34] is used to predict the char combustion rate. The rate of depletion of char particle mass is expressed as;

are volatile evolution rate dt dt and char combustion rate respectively are expressed mathematically in the following sections. An energy balance equation is solved to include the heat energy exchanges in between gas and particle phases because of convection, radiation and coal burnout. The energy balance equation is expressed as; mp ( j ) c pg

In this model the reaction rate depends on both oxygen diffusion rate towards the core and finite rate of reaction between carbon and oxidant including the effect of coal porosity and porous hole twist as well. In equation (9), D0 and R denoted the oxidant diffusion rate coefficient and chemical reaction rate considering physical nature of the coal respectively and Ap is the particle surface area. Reaction mechanisms with four reactions are considered for either gas phase reaction or particle phase reaction given as following;

C 4.61H 4.94 O0.46 N 0.1219 S0.0155 + 3.32O2 → 4.61CO + 2.47H2 O + 0.0609N2 + 0.0155SO2

For gas phase combustion, Eddy dissipation model proposed by Magnussen and Hjertager [35] is used. The reaction rate in gas phase combustion is evaluated by considering the smallest rate of reaction out of three as;

 YO B1 ∑Yp  ε ω EDM = Bρ min Y f , 2 ,  γ 1 + γ  k 

In this model B and B1 are the model constants and their value are taken as 4 and 0.5 respectively. Ỹf , Ỹo , Ỹp and γ are 2 mass fraction of fuel, oxygen, products and local equivalence ratio. NOx Modelling During combustion of hydrocarbon fuel, a big amount of pollutant NO formation occurs as compared to other NOx such as; N2O and NO2 [1]. Therefore only NO formation has been considered in this work. Moreover the effect of reburn on net NO is also considered by using co-flow methane as reburn fuel.

Thermal-NO, Prompt-NO (originates from atmospheric nitrogen) and Fuel-NO (originated from inherent nitrogen stored in fuel) are considered as the source of NO. Thermal-NO formation is evaluated by following the well known Zeldovich mechanism [36]. By considering the Quasi-steady assumption of nitrogen atom formation, the rate of Thermal-NO is expressed as; d [ NO ] = 2k f ,1 [ XO ]  X N2  dt 2   kr ,1kr ,2 [ X NO ] 1 −    k f ,1  X N  k f ,2  XO    2  2 mol/m 3 − s   kr ,1 [ X NO ] 1 +   k f ,2  XO  + k f ,3 [ XOH ]  2

respectively [37]. The reaction in between hydrocarbon radicals and atmospheric nitrogen within the fuel rich zone under moderate temperature forms Prompt-NO. The rate of Prompt-NO proposed by De Soete [38] is expressed as;

In equation (12), f is a correction factor that depends on carbon atom number of fuel and the equivalence ratio within the reaction zone. kpr is the NO production rate coefficient. a is reaction order for oxygen depends up on the oxygen molar fraction within the reaction zone. During the reaction, fuel nitrogen converted into intermediate species like HCN and NH3. The consequent reaction forms NO by reacting with O2. Also a part of NO reacted with HCN/NH3 again converted to N2. However a significant amount of fuel N also directly converts to NO. The source term for HCN, NH3 and NO (formation of HCN,NH3 and NO from fuel nitrogen conversion) in their corresponding species transport equations are expressed in ref. [36]. The HCN and NH3 depletion rate by reacting with O2 and N2 or formation of NO are expressed as; a MWNO p d [ HCN ] |O2 → NO = k1 [ X HCN ]  XO2  e − E1 / RT (13a) dt RT

MWNO p d [ HCN ] |NO → N2 = −k2 [ X HCN ][ X NO ]e − E2 / RT (13b) dt RT

In this work, the nitrogen contained by volatile and char are taken as 30% and 70% respectively. It is considered that 90% nitrogen convertible to form NO. 80% and 20% of volatile N is allowed to convert HCN and NH3 respectively, whereas direct conversion of volatile N to NO is not considered. 100% of char is converted to direct NO. However NO concentration also decreases due to the reaction of NO with CH radicals within the fuel rich zone. The reaction mechanism of NO with CH radicals is globally expressed as;

CH i + NO → HCN + products ( H 2O, OH , O …etc.) (14) In order to evaluate the NO depletion rate, partial equilibrium approach is used [39], where NO depletion rate and HCN formation rates considering CH4 is reburn fuel are expressed as;

d [ HCN ] = 4 × 10−4 ( ka χ1 + kb χ12 )  XCH 4  [ X NO ] dt

d [ NO ] = −4 × 10−4 ( ka χ1 + kb χ12 )  XCH 4  dt [ X NO ] + (kc χ13 χ2 )  XCH4  [ X NO ]}

that hydrogen radical H is observed as same order with H2 XOH near post flame region in non premixed flame. χ 2 = X H2O , where mole fraction of OH radical is obtained from the reaction; k f 1, kr 1

The rate constants ka, kb, kc, kf1, kr1 in equation (15) and (16) for different re-burn fuel are used from the work Leung et al. [40].

Operating Conditions

Five different overall equivalence ratio conditions have been used in this study as mentioned in Table 2. In all the cases, mass flow rate of coal and methane are kept constant.

By changing the primary air mass flow rate the desired overall equivalence ratio is maintained. However, the coal mass flow and methane mass flow rate are taken in such a way that the heat input by both the fuel are in 50:50 ratio. For the calculations, higher calorific values of coal and methane are used. The higher calorific values of coal and methane are taken as 29MJ/kg and 50MJ/kg respectively. The mass of fuel supplied are calculated as;

Total heating value (5kW ) × heat input percentage from methane

The mass of air supplied is calculated as;  mf   m  a actual Equivaleance ratio =  mf   m  a

where mass flow rate of fuel m f = mcoal + mmethane and stoichiometric fuel-air ratio is calculated as;

In equation (20), ma →CH 4, ma →C and ma →vol are the mass of air required to completely burn a mass of fuel mCH 4, mC and mvol respectively, which are calculated by using the molar coefficient and molecular weight of the fuel and oxidant.

Results And Discussion

Model Validation In order to establish the acceptability of current computational model, few results are compared with an experimental work performed in Japanese Central Research Institute of Electric Power Industry (CRIEPI) [21]. An axisymmetric model is prepared with same dimension as mentioned in the experimental set up. Same inlet conditions are used as mentioned. A parabolic velocity profile for main air is used by keeping given mass flow rate unchanged. Newlands Bituminous coal is used as fuel along with coflow methane pilot fuel. The various computational models has been used in this work are discussed in computational modelling section. The experimental conditions are given in Table 3.

GRID Independent TEST

Grid independent test has been carried out using three different mesh configuration such as; 60000, 40000 and 30000 mesh. Quadrilateral structured elements are used throughout the computational domain. Nitrogen and volatile matter concentration along the axis with different mesh configuration at overall equivalence ration φ = 9.0 are compared as shown in Figure 3. Both N2 and volatile concentration deviates significantly throughout the length with 30000 meshes as compared to 40000 and 60000 meshes. On the other hand, minor deviation in both N2 and volatile concentration has been observed between 40000 and 60000 mesh configured computational domain. However, for the entire simulation 60000 mesh size computational domain is used for simulation. Table 3. Experimental conditions [21] Bulk equivalence ratio

NO Formation And Destruction

The total NO concentration is the sum of Thermal-NO, Prompt-NO and Fuel-NO concentration. Figure 9, illustrates the total NO concentration field. Thermal-NO depends upon flame temperature, availability of atmospheric O2 and N2. From Figures 4a-e, it is also observed that, for lower equivalence ratio cases, plenty of atmospheric nitrogen is available in reaction zone. Therefore, due to very high flame temperature and better availability of O2 and N2, a smaller but intense reaction rate for Thermal-NO has been observed at φ=0.8 and 1.0 as shown in Figure 10.

As overall equivalence ratio increases, due to O2 and N2 deficit and decreased flame temperature, the intensity of Thermal-NO formation rate decreases. However for higher equivalence ratio, a wide reaction zone for Thermal-NO has been observed due to a wider higher temperature region in these cases. Prompt-NO reaction rate is also related to atmospheric N2, where N2 reacts with hydrocarbon radicals to form intermediate species. The consequent reactions of intermediate species with O2 and oxygen radicals form Prompt-NO. Therefore Prompt-NO formation rate become high within the fuel rich zone at moderate temperature. But

Figure 9. Total NO mass concentration for various equivalence ratio condition cases.

Figure 11. Rate of Prompt-NO (kmol/m3-s) contour for various equivalence ratio condition cases.

Figure 10. Rate of Thermal-NO (kmol/m3-s) contour for various equivalence ratio condition cases.

Figure 12. Rate of Fuel-NO (kmol/m3-s) contour for various equivalence ratio condition cases.

Figure 13. a) HCN, NH3 and NO concentration along the axis for a) φ = 0.8 b) φ=1.0 c) φ = 3.0 d) φ = 6.0 e) φ = 9.0.

if local equivalence ratio becomes too high, Prompt-NO decreases due to deficit in O2. Figure 11 illustrates the rate of formation of Prompt-NO at various overall equivalence ratio conditions. For the cases of low equivalence ratio (φ=0.8 and 1.0), despite of lack of rich local equivalence ratio, sufficient O2 and N2 increases the Prompt-NO formation rate in these cases. On the other hand, as the overall equivalence rate increases, despite of availability of richer equivalence ratio, insufficient O2 and oxygen radicals, Prompt-NO decreases. Fuel-NO is the conversion of inherent nitrogen presents in coal to NO directly and/or through intermediate reaction of HCN and NH3. Figure 11 illustrates the Fuel-NO formation and destruction rate followed by the equation 13a–13d. Fuel-NO formation rate is shown with positive sign in contour legend, which majorly depends upon the reaction of HCN/NO with O2 to form NO. Figure 13 shows the intensity of HCN and NH3 along the axis at various equivalence ratio cases. Better conversion of nitrogen to HCN, NH3 and direct NO is obtained in case of higher equivalence ratio as shown in Figure 13. Therefore combustion with a higher equivalence ratio portrays a wider and concentrated NO formation rate as shown in Figure 12. On the other hand, combustion with lower equivalence ratio depicts a narrow

but intense Fuel-NO formation rate due to poor HCN, NH3 and NO concentration and intense flame temperature. Dark blue colour in the contour having negative sign depicts the destruction of Fuel-NO that reacts with HCN and NH3 converted to N2. With the comparison, it is evident that, the Fuel-NO formation rate is higher than other two. Prompt-NO formation rate is much smaller, thus less significant on total NO contribution. From figure 13 it has been observed that, with increase in equivalence ratio, even if there is a better conversion of fuel nitrogen to HCN and NH3 due to rich mixture, but low flame temperature is unable to form much NO. As a result Fuel-NO rate is much higher in case of low equivalence ratio. However a wider range of higher rate of formation of Thermal-NO and Fuel-NO, the concentration of total NO is almost highest in case of φ=3.0 as shown in Figure 9. In the cases of low equivalence ratio (φ=0.8, φ=1.0), even though the formation rate of Thermal, Prompt and Fuel-NO is highest, but limited to a narrow region. Probably higher convection due to very high air velocity causes poor concentration of total NO in the cases of low equivalence ratio (φ=0.8, φ=1.0). Figure 14a illustrates the NO reburn rate by considering CH4 as reburn fuel. In reburn, NO converts majorly to HCN by reacting with hydrocarbon radical. Thus it is a kind of reverse of Prompt-NO formation. From the Figure, it is observed that, as richness of mixture increases, the rate of

Table 4. Maximum and minimum values of temperature and various reaction rates for NO formation and reduction φ = 0.8

Thermal-NO –5.968e-21 0.000207 –9.978e-21 0.000214 –4.4498e-17 3.4689e-5 –1.7643e-10 5.4612e-6 –5.34581e-11 3.3557e-6 rate (kmol/ m3-s) Prompt-NO rate (kmol/ m3-s)

Figure 14. a) Reburn reaction rate (kmol/m3-s) for various equivalence ratio condition cases, b) Total NO contour comparison without reburn and with reburn condition for φ=6.0 and φ=9.0.

NO conversion become higher. Because for this condition, the reburn fuel concentration is higher due to slow primary air entrainment. However in this Figure, negative sign represents the destruction of NO. Other factor such as; higher flame temperature also enhances NO reburn rate. Figure 14b depicts the comparative picture of total NO concentration without considering reburn and with reburn. The comparison for two cases such as; φ=6.0 and φ=9.0 are only given because of significant difference can be visualized easily. Figure 15a and 15b, depicts the mass weighted average of NO concentration across seven different planes along the

axial line with and without reburn respectively for various equivalence ratio conditions. Figure 15a, illustrates the NO reduction efficiency along the axial line. The NO reduction efficiency is calculated as;

NO reduction efficiency = NO without reburn − NOout reburn NO without reburn

NO is not reduced significantly with lower equivalence ratio condition due to lower reburn rate. As a result reduction

Figure 15. a) NO in ppm without reburn, b) NO in ppm with reburn and c) NO reduction efficiency, for various overall equivalence ratio conditions across different axial position.

efficiency decreases throughout along the axis. It is obvious that due to intense reburn rate near inlet region, the reduction efficiency become high at this region. It is decreasing and at outlet plane the reduction efficiency become highest irrespective of equivalence ratio. For φ=9.0, maximum reduction efficiency is around 35% obtained near inlet, whereas at exit plane it is around 15%. But with decrease in overall equivalence ratio, the reduction efficiency suddenly drops. However for φ=0.8, φ=1.0 and φ=3.0 the reduction efficiency is even less than 1% at exit plane. Therefore NO reduction with co-flow reburn method is not much effective for lean mixture condition.

COAL Burnout

Here coal burnout is regarded as, decrease in mass of coal particles due to continuous volatile evolution and char combustion during the movement of particle within the computational domain. Coal burnout percentage across different plane along the axis is calculated as;

Figure 16 illustrates the coal burnout percentage along the length. From the Figure, it is observed that for higher overall equivalence ratio (φ=3.0, 6.0 and 9.0), burnout percentage is around78-85% obtained at exit plane. In these cases, wider temperature field and higher particle residence time due to low air velocity, raise the volatile evolution rate and char combustion as well. On the other hand the as the conditions are reverse for lower equivalence ratio cases, coal burnout percentage decreases in lower equivalence ratio cases. However coal burnout rate increases along the length of the combustor in all the cases. It is obvious that near inlet region, volatile removal and far from inlet the char combustion causes continuous mass losses from coal particle.

Nomenclature

Figure 16. Coal burnout percentage for various overall equivalence ratio conditions across different axial position.

Conclusion

A computational simulation has been performed to study the effect of overall equivalence ratio on combustion performance, NO formation and the reburn effect on NO reduction using CH4 as co-flow fuel. A fixed amount of co-flow methane fuel is supplied through an annular space. Through central hole, primary air is supplied along with fixed amount of coal. The required equivalence ratio is maintained by varying the primary air velocity. Thermal-NO, Prompt-NO and Fuel-NO mechanism is used to predict NO formation, whereas equilibrium approach NO reburn model is used for NO depletion. From the study, it is observed that, Fuel-NO contributes a major fraction to the total NO. A significant contribution of Thermal-NO is also there to the total NO. Prompt-NO formation rate is very low; it has negligible contribution to total the NO. From the study it is observed that, for the given geometric conditions, in the case of low equivalence ratio conditions (φ=0.8 and 1.0), even though the rate of formation of Fuel and Thermal-NO is intense but total NO concentration is very low due to narrow reaction zone near inlet. On the other hand due to mixture richness, at higher equivalence ratio cases (φ=3.0, 6.0 and 9.0), the Total NO is much higher and more concentrated away from inlet region. NO reduction using co-flow methane injection is not an effective method for lower overall equivalence ratio conditions. Because, maximum NO reduction efficiency at φ=0.8,

Data Availability Statement

The authors confirm that the data that supports the findings of this study are available within the article. Raw data that support the finding of this study are available from the corresponding author, upon reasonable request.

Conflict Of Interest

The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Ethics

There are no ethical issues with the publication of this manuscript.

References

1. Edge P, Gharebaghi M, Irons R, Porter R, Porter, Midwest: National Renewable Energy Laboratory Pourkashanian M, et al. Combustion modelling Midwest Research Institute; 2002. [CrossRef] opportunities and challenges for oxy-coal car- [14] Su S, Xiang J, Sun L, Zhang Z, Sun X, Zheng C. bon capture technology. Chem Eng Res Design Numerical simulation of nitric oxide destruction by 2011;89:1470-1493. gaseous fuel reburning in a single-burner furnace.
2. Kuang M, Li Z. Review of gas/particle flow, coal Proceeding of Combustion Institute 2007;31:2795- combustion, and NOx emission characteristics 2803. [CrossRef] within down-fired boilers. Energy 2014;69:144-178. [15] Han X, Wei X, Schnell U, Hein KRG.Detailed mod- [CrossRef] eling of hybrid reburn/SNCR processes for NOx
3. Kambara S, Takarada T, Yamamoto Y, Kato K. reduction in coal-fired furnaces. Combust Flame Relation between functional forms of coal nitrogen 2003;132:374-386. [CrossRef] and formation of nitrogen oxide (NOx) precursors [16] Zarnitz R, Pisupati SV. Evaluation of the use of coal during rapid pyrolysis. Energy Fuels 1993;7:1013- volatiles as reburning fuel for NOx reduction. Fuel 1020. [CrossRef] 2007;86:554-559. [CrossRef]
4. Pershing DW, Wendt J. Pulverized coal combus- tion: The influence of flame temperature and coal BR. Study of syngas co-firing and reburning in a coal composition on thermal and fuel NOx. Symposium fired boiler. Fuel 2004;83:1991-2000. [CrossRef] (International) on Combustion 1977;16:389-399. [18] Choi M, Li X, Kim K, Park Y, Sung Y, Choi G. Analysis [CrossRef] of impact factors for exhaust tube vortex flame gen-
5. Naruse I, Yamamoto Y, Itoh Y, Ohtake K. eration in methane-assisted pulverized coal swirling Fundamental study on N2O formation/decom- flames. Appl Therm Eng 2019;157:113550. [CrossRef] position characteristics by means of low-temper- [19] Yamamoto K, Murota T, Okazaki T, Taniguchi M. ature pulverized coal combustion. Symposium Large eddy simulation of a pulverized coal jet flame (International) on Combustion 1996;26:3213-3221. ignited by a preheated gas flow. Proceeding of [CrossRef] Combustion Institute 2011;33:1771-1778. [CrossRef]
6. Nichols KM, Thompson LM, Empie Jr HL. A review of NOx formation mechanisms in recovery fur- Kempf AM. Large eddy simulation of a pulverised naces. TAPPI J 1991:1-20. 2016 J Ther Eng, Vol. 7, Supp 14, pp. 1–16, December 2021 coal jet flame. Proceedings of Combustion Institute method. Aerospace Research Central US 1988;2:309- 2013;34:2419-2426. [CrossRef] 316. [CrossRef]
7. Hwang SM, Kurose R, Akamatsu F, Tsuji H, Makino H, Katsuki M. Observation of detailed structure of coal combustion with non-gray gas radiation effects. turbulent pulverized-coal flame by optical measure- Combus Sci Technol 2001;166:151-174. [CrossRef] ment. Fluid Therm Eng 2006;49:1316-1327. [CrossRef] [32] Backreedy RI, Fletcher LM, Ma L, Pourkashanian M,
8. Hashimoto N, Kurose R, Hwang SM, Tsuji H, Shirai Williams A. Modelling pulverised coal combustion H. Flame. A numerical simulation of pulverized coal using a detailed coal combustion model.Combust combustion employing a tabulated-devolatiliza- Sci Technol 2006;178:763-787. [CrossRef] tion-process model (TDP model). Combust Flame [33] ANSYS. ANSYS Fluent Theory Guide. 2012;159:353-366. [CrossRef] 2011;15317:724-746.
9. Wen X, Jin H, Stein OT, Fan J, Luo K. Large Eddy Simulation of piloted pulverized coal combustion Pourkashanian M, Williams A. Co-firing pulverised using the velocity-scalar joint filtered density func- coal and biomass: a modeling approach. Proceedings tion model. Fuel 2015;158:494-502. [CrossRef] of Combustion Institute 2005;30:2955-2964. [CrossRef]
10. Versteeg HK, Malalasekera W. An introduction to computational fluid dynamics: the finite volume cal modeling of turbulent combustion with method: London: Pearson Education; 2007. special emphasis on soot formation and combus-
11. Launder BE, Spalding DB. The numerical com- tion. Symposium International on Combustion putation of turbulent flows. Numerical prediction 1977;16:719-729. [CrossRef] of flow, heat transfer, turbulence and combustion. [36] Choi CR, Kim CN. Numerical investigation on the Amsterdam: Elsevier; 1983: 96-116. [CrossRef] flow, combustion and NOx emission characteristics
12. Ghose P, Patra J, Datta A, Mukhopadhyay A. in a 500 MWe tangentially fired pulverized-coal Modelling. Prediction of soot and thermal radiation boiler. Fuel 2009;88:1720-1731. [CrossRef] in a model gas turbine combustor burning kerosene [37] Fluent A. Academic Research, Release 14.7, Help fuel spray at different swirl levels. Combust Theory System, Theory guide. ANSYS; 2014. Model 2016;20:457-485. [CrossRef] [38] De S, SOETE DD G. Overall reaction rates of NO
13. Gorenflo R, Mainardi F, Moretti D, Pagnini G, and N2 formation from fuel nitrogen. Pascal Francis Paradisi P. Discrete random walk models for space– 1974;1093-1102. [CrossRef] time fractional diffusion. Chem Phys 2002;284:521- [39] Su S, Xiang J, Sun X, Zhang Z, Zheng C, Xu M.
14. Morsi SAJ, Alexander AJ. An investigation of par- tion by reburning. Energy Fuel 2006;20:1434-1443. ticle trajectories in two-phase flow systems. J Fluid [CrossRef] Mech 1972;55:193-208. [CrossRef] [40] Leung KM, Lindstedt RP. Detailed kinetic modeling
15. Ranz WE, MarShal WR. Prog. Evaporation fron of C1—C3 alkane diffusion flames. Combust Flame DropS-Part I and II.Chem Eng Prog 1952;48:141-146. 1995;102:129-160. [CrossRef]

Share and Cite