







































MN. Rahman et al. /Future Energy                                                                                          February 2026| Volume 05 | Issue 01| Pages 41-52 

41 

 

 

 

Article 

Ammonia as a hydrogen carrier: LES of ammonia-solid 

fuel firing at varying air staging ratios  

Mohammad Nurizat Rahman1*, Muhamad Shazarizul Haziq Mohd Samsuri2, Suzana Yusup3, 

Ismail Shariff 2   

1Energy Markets and Strategy, Energy Systems, DNV, 118227 Singapore 
2Generation Unit, Generation and Environment, TNB Research, Kajang, Selangor, 43000, Malaysia 
3Faculty of Engineering and Technology, Sunway University, Selangor, 47500, Malaysia 

A R T I C L E   I N F O 
 

Article history: 
Received 02 October 2025  
Received in revised form 
17 November 2025 
Accepted 15 December 2025 
 
Keywords: 
Large Eddy Simulation (LES), Solid fuel, Coal, 
Ammonia, Power Generation, 
Air-staging combustion 
 
*Corresponding author 
Email address:  
mohammadnurizatrahman@gmail.com 
 
DOI: 10.55670/fpll.fuen.5.1.5 

A B S T R A C T 
 

Hydrogen carriers, such as Ammonia (NH3), is anticipated to be used as a 
carbon-free alternative for solid fuels, such as coal. Hence, the effect of air 
staging ratio (ASR) on emissions from NH3 co-firing with sub-bituminous coal 
was numerically investigated in a small-scale coal combustor via a Large Eddy 
Simulation (LES) method. The validation with experimental data demonstrated 
a difference in nitrogen oxides (NOx) and temperature profiles of less than 10 
%. Carbon dioxide (CO2) and sulphur dioxide (SO2) levels are decreasing as the 
NH3 percentage rises, but ASR has minimal influence. Increasing the ASR from 
20 to 60 % resulted in NOx reduction, except for 60 calorific (cal.) % NH3, where 
NOx began to grow at ASR 60 %. In the said case, peak temperature was 
recorded in the over-fire Air (OFA) zone due to considerable unburned carbon 
(UC) oxidation, resulting in an increase in thermal NOx. Due to oxygen 
deficiency, coal volatiles and NH3 are thought to burn in the firing zone due to 
dominant devolatilization, resulting in significant UC/char oxidation in the OFA 
zone. Overall, with proper ASR tuning, NH3 co-firing can produce low CO2, SO2, 
and NOx, and existing coal-fired utility ASR technology can be used. 

1. Introduction  

Nowadays, the electricity sector accounts for a 

significant share of worldwide carbon dioxide (CO2) 

emissions [1-2], primarily because coal-fired thermal power 

plants supply a substantial share of global electricity demand, 

owing to their large reserves and affordability [2-6]. It is a 

known fact that coal-fired thermal power plants emit more 

CO2 than other generation systems [7], since coal has a 

relatively higher carbon content than almost all other fossil 

fuels [8]. And it serves as one of the main anthropogenic CO2 

emission sources [9], accounting for 41 % of worldwide CO2 

emissions in 2023 [10]. A reduction in greenhouse gas (GHG) 

emissions is critical as a key control measure for climate 

change issues, and it is now becoming a worldwide accord [9-

13]. The attainment of a sustainable society is extensively 

spoken about, and the pressure on coal-fired utilities to 

decarbonise is heightening [5]. At the UN's Climate Change 

Conference (COP26), more than 40 nations made pledges to 

abandon coal. Despite the excitement surrounding the net-

zero carbon transition, the global energy crisis in recent years 

has led to a rush for coal demand once more, demonstrating 

that the rapid switch to renewable energy is, in fact, more 

difficult than anticipated. One of the solutions to this issue is 

to gradually phase out coal, so that there is an adequate 

amount of time for low- and/or zero-carbon technologies to 

reach economies of scale [2]. Therefore, while waiting for 

these technologies and their associated supply chains to fully 

mature, decreasing negative emissions from existing coal-

fired thermal power plants is critical for reducing the carbon 

footprint and eventually establishing the targeted net-zero 

society [9-10]. Therefore, various techniques are being 

established to reduce CO2 emissions from the said plants, 

including integrated gasification combined cycle (IGCC), 

ultra-supercritical technology [10], oxy-fuel combustion, 

carbon capture and storage (CCS) [7], double reheat 

technology [10], and the use of low-/zero-carbon fuels [11]. 

When it comes to employing these types of fuels, biofuels 

(such as biogas and biomass) are appealing fuels for co-firing 

applications. Yet, variability in harvesting periods indicates 

fluctuations in feedstock supply and presents significant 

hurdles for both operations and market sentiment. Carbon-

free fuels, such as hydrogen, are another option, expected to 

play an increasingly important role in creating a net-zero 

society, especially in hard-to-abate sectors [11-15]. However, 

owing to its unique properties [11], the storage and 

transportation of hydrogen remain relatively complex 

 

 

Future Energy 

Open Access Journal 

https://doi.org/10.55670/fpll.fuen.5.1.5 

 

 

 

 

 

 

 

 

 

February 2026| Volume 05 | Issue 01 | Pages 41-52 

Journal homepage: https://fupubco.com/fuen 

 
ISSN 2832-0328 

mailto:mohammadnurizatrahman@gmail.com
https://doi.org/10.55670/fpll.fuen.5.1.5
https://fupubco.com/fuen


MN. Rahman et al. /Future Energy                                                                                          February 2026| Volume 05 | Issue 01| Pages 41-52 

42 

 

[13,16]. Hydrogen carriers, such as ammonia (NH3), are 

effective alternatives to pure hydrogen, as demonstrated in a 

report by Rahman [13], owing to their relatively high 

hydrogen density compared with other hydrogen carriers, 

such as organic hydrides [16-17]. Additionally, NH3 is a 

desirable vector for hydrogen because it retains roughly 90% 

of the energy from the hydrogen feedstock and is much easier 

to store due to its ease of liquefaction [18]. Having said that, 

NH3 is an enabler of the hydrogen economy, not a competitor 

[19]. Furthermore, as a fuel, NH3 can be employed in existing 

plant-related combustion systems without the need for a 

procedure to extract its hydrogen content [18], albeit with 

tuning works to be done where necessary. NH3 has been 

regularly and widely employed as a denitrification material in 

boilers and plants. The latest movement in decarbonising the 

shipping industry with NH3 as a marine fuel is an obvious 

indication of some of its benefits that can be potentially 

adapted to power generation systems, such as production 

scalability, a comparatively adequate energy density with 

relatively simple storage needs as opposed to pure H2, and 

secured usage in existing industrial processes. NH3 is already 

an essential commodity chemical worldwide, with a mature 

storage and distribution supply chain that is becoming 

increasingly crucial to vital aspects of global society, such as 

the fertiliser industry [17]. As such, infrastructures for NH3 

storage and delivery are well developed [3, 12], and its initial 

and operational costs are expected to be relatively lower than 

those of other low-carbon/carbon-free fuels and hydrogen 

carriers [7], with several techno-economic studies indicating 

that it has the potential to be the lowest cost zero-carbon 

option. Hence, NH3 could be a preferable candidate for 

widespread use in the future because it is both a hydrogen 

carrier and a potential replacement for conventional fossil 

fuels. Hence, it is clear that interest in NH3 as an electricity-

generated component is growing, especially with its prospect 

as a crucial fuel/feedstock for decarbonising power 

generation [2]. 

Co-firing of coal and NH3 has recently been regarded as a 

promising method for minimising CO2 emissions from coal-

fired thermal power plants [17]. For the production of NH3 as 

a fuel, the "blue NH3" is expected to be created using CCS to 

capture carbon emissions [10]. The "green NH3" could be 

produced from hydrogen electrolysis from water, utilising the 

excess renewable electricity (e.g., wind and solar power) [9]. 

Hence, “green NH3” is a derivative renewable fuel that 

functions as a transporter and a storage option for renewable 

energy [17]. At the same time, it is important to note that the 

reduction in emissions from NH3 co-firing is also highly 

affected by the NH3 production method [18]. Yet, NH3 co-

firing remains one of the most realistic and appealing options 

compared to most strategies for fuel blending scenarios [19]. 

Co-firing methods aim to maximise the utilisation of facilities 

in existing coal-fired power plants, potentially reducing 

resource waste and financial/opportunity losses due to 

power plant early retirement [20]. Moreover, reducing the 

carbon content of the main fuel stream (in this case, coal) 

could help cut CO2 emissions. However, given the limited 

capacity of NH3 production facilities worldwide, it is unlikely 

that NH3 will fully replace coal in the short to medium term 

[3]. Possibly in the future, especially in the scenario of NH3 

yield sees an exponential increase due to its use as a hydrogen 

carrier (combined with government fuel incentives), the 

associated cost of NH3 procurement has the potential to fall 

even further, and it can thus be used as one of the main energy 

vectors for power generation systems [7]. By utilising green 

NH3 as a carbon-free fuel, the integration of existing coal-fired 

power plants and low-carbon renewable options could be 

realised, providing an essential engineering pathway for a 

potentially cleaner coal-fired power generation [10]. Thus, at 

the moment, co-firing NH3 with pulverised coal in a boiler is 

seen as a quick and viable means of effectively lowering CO2 

emissions from these systems, as the world waits for the 

eventual full phase-out of coal-fired power plants [17]. 

Despite this, co-firing NH3 in coal-fired boiler systems 

could result in elevated nitrogen oxide (NOx) emissions, 

owing to its much higher fuel-nitrogen content [9, 12]. These 

impacts must be investigated in advance, including their 

emissions and combustion aspects, if NH3 co-firing is to be 

widely used in the future. Various organisations worldwide 

have researched the combustion properties of NH3 [7]. In the 

context of combustion assessments, comprehensive data on 

NH3 flame propagation across varying operational scenarios 

aid combustion dynamics assessment and the development of 

comprehensive reaction mechanisms [12, 16].  

Recent advances in research on NH3 co-firing technology 

with pulverised coal have been published [3,20]. IHI 

Corporation, for instance, successfully tested NH3 co-firing in 

a pulverised coal combustor system (10 MW size), with NH3 

accounting for 20% of the co-firing ratio. Moreover, the 

results from these experimental works have demonstrated 

that for the case of NH3 co-firing, unburned carbon (UC) in fly 

ash and NOx emission could be comparable to those in the 

scenario of full coal firing, if a proper NH3 injection technology 

is adopted. In separate experimental research, the 

characteristics of pulverised coal+NH3 co-firing were 

explored in detail via a horizontally shaped single burner with 

the feeding rate of 100 kg/hr (coal) [3]. Their results have 

shown that when the NH3 was injected from the burner’s 

centre with NH3 of 20 cal.%, NOx concentration in the flue gas 

has been seen to be elevated by roughly 20 % from the pure 

coal firing case, and UC in fly ash increased in a moderate 

manner. These findings suggest that controlling/tuning NH3-

coal co-firing while preserving both flame stability and NOx 

emissions is a significant challenge that requires further 

research to achieve acceptable combustion characteristics. 

Tuning the air staging ratio (ASR) is another potential 

approach for decreasing NOx emissions when co-firing coal 

and NH3. This is the same strategy employed in existing coal-

fired power plants, where airflow is distributed via the 

combustion air and then separated into the burner zone and 

the over-fire air (OFA) zone above the burner regions [21-22]. 

The existing air staging system in most coal-fired utility 

furnaces is a technical advantage. As a result, its start-up and 

operating costs are expected to be relatively low, as it 

requires essentially no alteration to the existing coal-fired 

power system. The only significant change in co-firing NH3 is 

at the burner, where a novel NH3-co-fired burner is required 

to inject NH3 with coal [20]. While several studies have 

examined the co-firing of coal and NH3 to evaluate its 

combustion and emission properties, there have been very 

few investigations into the effects of ASR on these properties. 

Moreover, a number of important works on NH3 co-firing 



MN. Rahman et al. /Future Energy                                                                                          February 2026| Volume 05 | Issue 01| Pages 41-52 

43 

 

mostly assumed an ideal reactor network model [4, 5, 7, 9, 

12]. This assumption is known to neglect the 3D effects of 

fluid dynamics. Weng [5], for instance, examined the presence 

of sulphur and alkali species in NH3 conversion processes in a 

post-flame environment using ideal reactor networks 

(Chemkin PRO software), with a focus on the characteristics 

of NO emissions and the slip of NH3 in flue gases. 

Despite the fact that NH3 and low-rank coals have 

comparable energy density [19], it is clear that important NH3 

properties, for example its relatively low laminar burning 

velocities and energy [16, 23], as well as elevated ignition 

energies and high auto-ignition temperatures [19], make it 

more difficult to combust NH3 efficiently and achieve the 

operational and combustion performances required by 

existing coal-fired power plants. In fact, more assessment is 

necessary before NH3 is used in existing coal-fired power 

plants to better understand its combustion and emission 

properties, mainly through further research on the kinetics 

and fluid dynamics of NH3 co-firing/full-firing, to aid flow 

tuning, such as ASR control. 

One viable option is computational fluid dynamics (CFD), 

which can be a useful tool for detailed investigations of the 

impact of ASR on NH3 co-firing. Subsequently, it can aid in the 

tuning of the ASR for actual coal-NH3 firing for NOx reduction 

in power plants. CFD has been frequently utilised to examine 

heat transmission and combustion dynamics in pulverised 

coal-fired utility furnaces. Zhang [3], for instance, used CFD 

approaches to examine the impact of NH3 ratio on coal+NH3 

co-firing in a pulverised coal combustor facility. The 

modelling findings were compared with experimental data on 

NOx, CO2, and UC levels. However, when compared to NOx 

readings from experimental data, the modelling approaches 

that use the Reynolds-averaged Navier-Stokes (RANS) model 

to resolve turbulence flow produced a 35 % and 47 % 

difference in NOx and UC levels, respectively.  

Cardoso [2] has also used one of the RANS-based models, 

the k- realizable model, to investigate NH3 and biomass co-

firing in a pilot-scale fluidised bed reactor system. The 

predicted CO2 and NO emissions from coal+biomass co-firing 

were compared to the actual CO2 and NO emissions from 

coal+biomass co-firing experiments. While the validations 

show reasonable agreement in terms of trend, there is no 

direct validation with real coal-NH3 co-firing. Therefore, it is 

still not entirely safe to assume that it can reliably simulate 

coal-NH3 co-firing when the validation basis is for different 

fuel blends (coal-biomass). Furthermore, post-processing is 

the most commonly used approach for NOx modelling, in 

which the main gas compositions, temperature, and velocity 

distributions are first obtained from combustion numerical 

calculations; then NOx-related reactions are added [3]. As a 

result, it is advised to use the combustion simulation 

parameter to evaluate a reasonable validation. Furthermore, 

actual NH3 co-firing test data should be used as a validation 

benchmark. 

In terms of resolving turbulence flow, while the use of the 

RANS model is beneficial to account for the accuracy and 

efficiency of the computational processes of NH3 co-firing [2], 

there is still more research required to fully comprehend the 

coal-NH3 flame dynamics and its associated emissions, which 

can only be achieved through the use of Large Eddy 

Simulation (LES). The key advantage of LES over RANS 

approaches is that it treats turbulence-chemistry interactions 

(TCI) more realistically [24]. As a result, higher fidelity to 

simulation results can be achieved. Through the low-pass 

filtering method, the LES reduces the computational burden 

by ignoring small length scales that require significant 

computational effort. Instead, the effect of these small length 

scales will be modelled using sub-grid scale models [25]. 

Therefore, LES is a powerful algorithm with a good trade-

off between reliable combustion dynamics predictions and 

computational cost. While LES has been commonly used for 

NH3 combustion simulations, the majority of these studies 

have been focused on co-firing with gaseous fuel, primarily 

natural gas and methane (CH4) [26-28]. The use of LES for 

coal+NH3 co-firing research is still lacking. Furthermore, the 

effect of varying ASR on NH3 co-firing must be studied further 

for future implementation in coal-fired power plants. Hence, 

reliable prediction of flame structures, temperature 

dynamics, and NOx levels is an important goal in the 

numerical modelling of coal+NH3 co-firing via LES in order to 

provide a holistic risk assessment for coal+NH3 co-firing at 

varying ASRs. As a result, in this study, CFD assessments using 

a detailed LES were performed to examine the effects of NH3 

co-firing on emissions in a small-scale coal combustor facility 

at various ASRs. Among the studied pollutants are CO2, SO2, 

and NOx. Sub-bituminous coal rank was used since it is the 

most commonly used coal rank in Malaysia’s coal-fired power 

plants [29-30]. The prediction accuracy of the CFD approach 

was first evaluated by comparing it to actual testing data from 

the pulverised coal combustor testing facility. 

2. Experimental setup  

The coal+NH3 co-firing test was carried out in TNB 
Research's pulverised coal combustor facility, as shown in 
Figure 1. The thermal input of 150 kW is employed for the co-
firing assessment. NH3 was inserted radially and 
concentrically from the centre of the coal injector burner, 
allowing direct mixing with the entering coal and air. This is 
also to lower the flow velocity of NH3, allowing for a longer 
residential mixing time with the coal+air mixture. The coal 
combustor facility has been built with a flue gas analyser, K-
type thermocouples at combustor sections, and a single swirl 
burner. The air and fuel inlets, the main combustion section, 
the heat exchanger, and the cyclone are the main parts of the 
combustor.  

The facility also includes a high-temperature glass 

window for viewing the flame profile and temperature 

measurements. Within the facility, the embedded combustor 

is configured in an L-shape to simulate a typical coal-fired 

boiler layout. This allows for the distinction between 

radiation, where the high-temperature combustor/flame 

zone occurs, and convection zones. The coal sample was 

transported using primary air (PA). Another flow of air, 

known as secondary air (SA), acts as an oxidising component 

for the primary combustion process. Throughout the test 

program, the PA flow rate was held constant at 9 Nm3/hr, 

while the coal, SA, and NH3 volume flow rates were calculated 

on a calorific/thermal input basis and maintained at 150 kW. 

The combustor main body, or the longer section of the L-

shape, is made up of four tube sections that measure 3.3 m in 

length and have an internal diameter of 0.6 m each. Each tube 

section contains one K-type thermocouple for temperature 

measurement. Section 1 is upstream, and Sections 2, 3, and 4 



MN. Rahman et al. /Future Energy                                                                                          February 2026| Volume 05 | Issue 01| Pages 41-52 

44 

 

come after it. The downstream part of the combustor (the 

shorter length) is about 0.91 m long and has an internal 

diameter of 0.3 m. 

 
Figure 1. TNB Research's pulverised coal combustor facility 

The gas temperatures and compositions at the 

combustor outlet, comprising CO2, SO2, and NOx, were 

constantly measured. One of the key aspects of ensuring the 

success of the testing is to ensure that the temperature from 

the combustion process within the combustor reaches a 

sufficiently high temperature condition and then maintains 

that temperature condition throughout the testing process so 

that it can mimic the actual temperature condition within the 

combustion zone of the actual boiler temperature condition. 

Hence, to achieve this flame temperature, liquefied petroleum 

gas (LPG) was also fed at a certain range of mass flow rate, 

roughly 2 kg/hr, with manual flow rate tuning required 

intermittently to maintain the desired temperature. The 

presence of NH3 was also assessed in the flue gas via a 

detector tube to ensure that no slip of NH3 occurred. 

Discharge of NH3 into the surrounding area is prohibited due 

to its toxicity [20]. For modelling validation purposes, out of 

all the testing cases that were done, one test case of pure coal 

firing and one test case of NH3 co-firing with a 60 cal.% 

proportion were selected. The ASR (fraction of OFA) in both 

of these cases was around 20 %. The properties of the fuels 

utilised in experiments and numerical studies are shown in 

Table 1. 

3. Numerical setup 

CFD techniques were used to simulate the combustion 

dynamics of pure coal firing and NH3 co-firing at various ASRs 

and NH3 co-firing ratios. ANSYS FLUENT 19.0 was used, and 

the majority of the default CFD solvers and models were 

already embedded.  

 

 

 

 

 

 

 

 

 

 

 

 

 

Yet, a new model establishment was needed to account 

for coal kinetics. The coal-firing simulation accounts for three 

key stages of the coal combustion process: devolatilization of 

coal, the subsequent conversion/reaction of char, and volatile 

reactions (from volatiles released during early 

devolatilization). The coal network model and the coal 

database from TNB Research’s analytical fuel laboratory were 

used to determine the composition of volatiles and the 

corresponding rate constants for coal devolatilization. Our 

prior studies [29-30] show the chemical reactions and coal 

combustion models employed in the current CFD 

assessments. The compressible and reacting Navier-Stokes 

(NS) equations were used in the numerical model. The 

pressure-based solver was used to solve the governing 

equations. Turbulence was solved using the Large Eddy 

Simulation (LES) model [31]. Wan [32] and Sun [33] provide 

detailed information on the formulations utilized in the NS 

equations and LES model for coal combustion. All of the 

equations were discretised using second-order upwind 

methods. The coal particle trajectories were traced via a 

Lagrangian approach, which took into account turbulent 

dispersion factors in the coal trajectories, an important 

consideration due to the turbulence mixing of the coal that 

occurs in the combustor facility. Another consideration is 

radiation, which was simulated using the discrete ordinate 

(DO) approach. The radiation model setup includes 

discretisation, which includes an angular direction of 5 

divisions, as well as polar and azimuthal orientations of 3 

pixels each. The weighted-sum-of-gray-gases model 

(WSGGM) [3] was used to predict gas emissivity. 

Within the entire CFD frameworks, a different method 

known as a post-processing technique was employed to 

predict NOx emissions from the simulated combustion 

processes of these fuels. The technique first allowed all 

combustion iterations to occur, during which all key 

modelling results were obtained, including temperature, 

major gas composition, and velocity distributions. Then, 

reactions of hydrogen cyanide (HCN), NH3, thermal NOx, and 

the eventual NOx reduction by released/residual char were 

incorporated after these key iterations in the combustion 

computation had finished. Having said that, it occurred during 

post-processing, not during the main processing. Accordingly, 

NOx-related species such as NO, NH3, HCN, O, hydroxide (OH), 

and N were computed at this stage. Turbulence, flow, other 

major gas compositions such as CO2, oxygen, hydrogen, and 

carbon monoxide (CO), as well as energy and radiation 

equations, were already solved during the main processing 

phase. Figure 2 summarizes the calculation approaches for 

the kinetics of coal-NH3 combustion and the corresponding 

NOx emissions.  

 

 

 

 

 

 

 

 

 

 

 

Table 1. Fuel properties 

Proximate analysis, wt. %, ad., coal 
(TM-Total moisture, VM-Volatile matter, 

FC-Fixed carbon, AC-Ash content) 

Ultimate analysis, wt. %, ad., coal 
(C-Carbon, H-Hydrogen, N-Nitrogen, O-Oxygen, S-

Sulphur) 

GCV- Gross Calorific Value, 
ad., coal (kcal/kg) 

TM VM FC AC C H N O S Coal A NH3 LPG 

24 41 39 2 68.7 4.4 0.9 23.7 0.2 6,449 5,374 11,775 

 



MN. Rahman et al. /Future Energy                                                                                          February 2026| Volume 05 | Issue 01| Pages 41-52 

45 

 

 
Figure 2. Calculation approaches for the coal-NH3 combustion 
kinetics and NOx 

Table 2 depicts the main conditions of simulated cases. 

This study included 16 cases. The same heat input utilised in 

the experiment was employed in all numerical cases. As 

previously stated, the co-firing ratio was calculated on a 

calorific/thermal input basis and set at 150 kW. The coal, NH3, 

and LPG flow rates are shown in Table 2. All cases have an 

equivalence ratio (Ø) of 1, indicating a stoichiometric 

condition. The combustion air (PA + SA) flow rate was 

determined using the coal/NH3/LPG ratio at a stoichiometric 

condition. The coal-to-NH3 co-firing ratio was varied, and NH3 

co-firing was increased from 0 to 60 cal.% at various ASRs. 

While the maximum ASR for a coal-fired air-staged 

combustion system is typically no more than 40 % due to fuel 

burnout issues, the ASR was set to 60 % for the current 

assessment to gain combustion insights for future ASR tuning 

of coal-NH3 firing in actual coal-fired power plants. The 

overall ratio of coal+NH3 to LPG was kept constant to ensure 

a constant LPG flow rate, with the LPG ratio kept at 27 cal.% 

and the rest being coal+NH3 ratio (83 cal.%).  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

The exact drawing of the combustor facility was used to 

generate the 3D model of its computational domain, which 

was later discretised mainly with hexahedral meshes using 

the Assembly Meshing method. A mesh-independent test was 

performed to ensure that the modelling results were not 

affected by the number of meshes. Table 3 presents the 

properties of the meshes used in the computational domain. 

Since the quality of meshes affects the spatial discretisation 

error, meshes were generated with both skewness and 

orthogonality taken into account to reflect overall mesh 

quality. Orthogonality indicates how closely the angles 

between consecutive mesh faces approach the ideal mesh 

angle. The scale of orthogonal quality ranges from 0 to 1, with 

1 indicating the highest quality [34]. Skewness reflects how 

close the mesh is to an optimal equiangular mesh. 

Heavily skewed meshes and their corresponding faces 

are unsuitable, as the governing equations are solved under 

the assumption that meshes are roughly equiangular. 

Skewness also ranges from 0 to 1, where values close to zero 

have the slightest deviation from a normalised equiangular 

angle [34]. All meshes were optimised to achieve high quality 

across both mesh quality metrics, as shown in Table 3. 

The predicted velocity and NOx (case C12) at the coal 

combustor outlet as the mesh count changes are shown in 

Figure 3. When the mesh number is increased from 2.412 

million to 4.171 million, NOx and velocity change little, with 

variations of less than 1 %. Hence, 2.412 million meshes were 

selected for the coal combustor domain.  

Figure 4 illustrates the mesh-independent model of the 

coal combustor’s computational (fluid) domain, as well as the 

boundary inlet details. Since the upstream region (inlet) 

entails intricate reactions and mixing of NH3, coal, air, and 

LPG, a finer mesh was generated in that zone. 

 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Table 2. Primary conditions of simulated cases 

Cases ASR (%) Coal + NH3 = 82 cal.% LPG (cal.%) Coal (kg/hr) NH3 (kg/hr) LPG (kg/hr) Ø 

Coal (cal.%) NH3 (cal.%) 

C10 0 100 0 27 10.00 0.00 2.00 1 

C12 20 80 20 8.00 2.40 

C13 40 60 40 6.00 4.79 

C14 60 40 60 4.00 7.19 

N20 0 100 0 10.00 0.00 

N22 20 80 20 8.00 2.40 

N23 40 60 40 6.00 4.79 

N24 60 40 60 4.00 7.19 

N30 0 100 0 10.00 0.00 

N32 20 80 20 8.00 2.40 

N33 40 60 40 6.00 4.79 

N34 60 40 60 4.00 7.19 

N40 0 100 0 10.00 0.00 

N42 20 80 20 8.00 2.40 

N43 40 60 40 4.00 4.79 

N44 60 40 60 4.00 7.19 

 



MN. Rahman et al. /Future Energy                                                                                          February 2026| Volume 05 | Issue 01| Pages 41-52 

46 

 

Table 3. Mesh characteristics 

 

 

Figure 3. Predicted velocity and NOx at varying mesh counts 
(location: combustor outlet) 

 

  

(a) (b) 

 
Figure 4. Coal combustor domain with (a) mesh model and (b) 
boundary inlets in plane AA view 

4. Results and discussion 

Figures 5 and Figure 6 display the validation results for 
NOx levels and temperature profiles from experiments and 
numerical simulations. The numerical results from the 
simulated cases with varied ASRs were thoroughly evaluated 
in subsequent sections based on predicted CO2, SO2, and NOx. 
For validation, NOx and temperature profile data from 
experimental coal firing and NH3 co-firing were compared to 
the numerical results from cases C12 and N42, respectively. 

4.1 Validation I: NOx concentrations 
According to Figure 5, NH3 co-firing results in higher NOx 

concentrations at the combustor exit in both the simulation 
and the experiment than in coal combustion. Based on the 

validation cases, the NOx concentrations in the coal-firing and 
NH3-cofiring cases are 265 and 1,573 ppm, respectively 
(according to the experimental results). Therefore, as the 
fraction of NH3 climbs to 60 cal.%, the NOx concentration 
increases by around 494 %, roughly five times more than that 
produced by the pure coal combustion scenario. It is well 
known that decreasing prompt and thermal NOx is always 
challenging in combustion involving carbon-dominant fuels. 
Yet, it appears that adding NH3 enhances the likelihood of 
fuel-bound nitrogen to interact with oxygen in the air, as seen 
in the experimental results in Figure 5. The synergistic effect 
of the fuel-bound nitrogen results in a considerable increase 
in the total NOx emissions, which consists of fuel, prompt, and 
thermal NOx. 

According to the numerical results, the established 
model appears to perform reasonably well when comparing 
the NOx results from both the model and the testing, with 
discrepancies in NOx concentrations between the two (for 
both coal firing and NH3 co-firing) being below 10 %. Despite 
the model's ability to simulate a broadly similar increasing 
trend in NOx emissions when NH3 co-firing is employed 
compared to the 100% coal firing case, there remains a slight 
variation between NOx values simulated by the established 
model and the ones obtained from the actual tests. One 
explanation for the minor discrepancies is that the techniques 
for NOx modelling used in this study, as previously mentioned, 
are based on a method that relies heavily on a set of 
mechanism that is semi-empirical in basis [3]. Of course, in 
the actual kinetics process from these fuels’ combustion, the 
reaction pathways for the NOx production are substantially 
more intricate, necessitating a greater computing cost to 
resolve the key chemical kinetics that are involved [12]. 
Therefore, semi-empirical mechanisms, while excellent for 
parametric and/or scaling assessments, may not be fully 
sufficient to provide absolutely correct NOx emissions. 
Nonetheless, when compared to the testing data, the 
simulation results shown in Figure 5 still showed a NOx 
difference with testing results below 10 %. Furthermore, 
quantitatively, it can be observed that the trend in the testing 
data matches the trend in NOx emissions simulated by the 
established numerical model. Hence, adequate validation can 
be reasonably claimed as the model can model NOx emissions 
with reliable accuracy for both pure coal firing and NH3 co-
firing. Safety-wise, it is important to note that the slip of NH3 
was not detected at the outlet of the combustor facility 
throughout the tests, which can be safely assumed to mean 
that there are no safety issues related to toxic NH3 slip being 
detected.  

Figure 5. NOx results from numerical and experimental works (ASR 
20 %) 

Average 
element 

size (mm) 

Meshes, 
106 

Orthogonal 
quality 

Skewness 

300 0.069 0.916 0.094 

200 0.081 0.936 0.079 

100 0.120 0.971 0.051 

50 0.162 0.990 0.024 

30 0.273 0.991 0.041 

20 0.530 0.993 0.037 

15 0.958 0.992 0.025 

13 1.425 0.990 0.016 

10 2.412 0.995 0.014 

8 4.171 0.996 0.019 



MN. Rahman et al. /Future Energy                                                                                          February 2026| Volume 05 | Issue 01| Pages 41-52 

47 

 

It can be reasonably postulated that this is mainly 
because of the direction in which the NH3 was injected into 
the combustor, where it is injected radially and concentrically 
from the coal injector’s centre at the burner inlet. This allows 
for a higher residence time for mixing with incoming 
combustion air and coal, hence reducing the jet velocity of 
NH3 entering the combustor. With the combined effects of 
these higher mixing rates, the potential for NH3 to escape 
from the recirculation zone is lower as well (as supported by 
a previous study [7]). This recirculation zone is important for 
flame stability in the combustor. Hence, the directional 
injection factor of NH3 is potentially the key reason for the 
inexistence of NH3 slip. 

4.2 Validation II: Temperature profiles 
As shown in Figure 6, the one-point temperature in each 

tube section of the coal combustor testing facility was 
measured using thermocouples. The testing data showed that 
the temperature difference (average) between the NH3 co-
firing and 100% coal firing cases varied between 0.9 and 6.1 
%. The average difference in numerical data between the two 
cases ranged from 0.4% to 4.9%. Due to the constant thermal 
input, differences in temperature were almost negligible in 
both cases: 100% coal firing and NH3 co-firing. Of the four 
tube sections of the combustor, Section 2 had the highest 
temperature (testing data), and it also had the highest 
temperature in the modelling assessments. The OFA 
enrichment between the two middle sections, Sections 2 and 
3, which serves as a thermal NOx mitigation method and also 
stops the flame front from elongating to reach between 
Sections 3 and 4, resulted in a significant temperature drop 
after Section 2. From the experimental data, the maximum 
temperatures are 1,149 °C for 100% coal firing and 1,121 °C 
for NH3 co-firing, indicating a measured peak temperature 
difference of 2.4%. The numerical model predicts maximum 
temperatures of 1,222 and 1,184 °C for the 100% coal-firing 
and NH3 co-firing cases, respectively. Therefore, there is a 3% 
difference in peak temperature between these two fuel 
scenarios, according to modelling results. Peak temperature 
differences between testing and numerical results are 5.9 and 
5.3 % for 100% coal and NH3 co-firing cases, respectively. 

A possible cause of these minor differences is the 
calculation of the turbulence-chemistry model in the current 
CFD frameworks. It is important to note that the turbulence-
combustion model used in the current study provides one of 
the best possible balances between the efficiency and 
accuracy of reacting flow CFD modelling. However, it remains 
a known fact that the LES algorithm filters the comparatively 
small-sized eddies to be modelled purely via the sub-grid 
formulation [31]. As a result, the current CFD framework does 
not fully capture the overall range of length scales. This could 
result in minor regional differences in flame front dynamics 
and corresponding temperatures.  

Still, the peak temperature difference between 
experimental and numerical results is less than 6 %.  
Qualitatively, the numerical predictions for the temperature 
profile positioning and trend nearly match those measured 
from test data. Therefore, the validation can be considered 
satisfactory, and the model can reliably predict temperature 
behaviour in both NH3 co-firing and 100% coal-firing cases. 

4.3 Impacts on CO2 emission 
CO2 compositions in flue gas changed dramatically when 

NH3 was co-fired in the coal combustor, as illustrated in 
Figure 7. In NH3 co-firing situations, the fraction of CO2 in flue 
gas was much lower than in the coal-fired base case. As 
anticipated, the CO2 fraction decreased as the NH3 co-firing 

ratio increased. When the NH3 co-firing ratio reached 60 
cal.%, the CO2 emission dropped to 7.8 % (for 0 % ASR), equal 
to a 41 % reduction in CO2 emissions as compared to the coal-
fired base scenario. CO2 emissions fell to 7.9, 8.0, and 8.1 % at 
ASRs of 20, 40, and 60 %, respectively (co-firing ratio 60 
cal.%). This equates to a reduction in CO2 emissions of 41 to 
42 % when compared to the coal-fired base scenario. 

The numerical results show that the CO2 emission ranges 
for coal firing, 20, 40, and 60 cal.% NH3 cases are 13.2 to 14.0 
%, 11.2 to 12.1 %, 9.7 to 10.5 %, and 7.8 to 8.2 %, respectively. 
As a result, increasing the percentage of NH3 in the fuel to 20, 
40, and 60 cal.% reduces CO2 emissions by roughly 13.1, 25.7, 
and 41.7 %, respectively, when compared to the coal-fired 
case. Since NH3 is a carbon-free fuel, replacing coal with NH3 
reduces CO2 emissions directly, with the decline ratio roughly 
equalling the fuel replacement ratio on a heating value basis. 

The percentage difference between the lowest and 
highest CO2 emissions when the ASRs were varied is less than 
10% in all cases shown in Figure 7. Hence, the difference in 
ASR has no substantial effect on CO2 emissions. While greater 
ASRs will result in a considerable increase in UC from the 
main combustion zone due to the lack of oxygen caused by 
reduced airflow, the remaining airflow was subsequently 
injected into the OFA zone to complete the oxidation of UC to 
generate CO2. Since the combustion temperature has a 
relatively lesser effect on CO2 production than thermal NOx 
[35], the considerably lower temperature of injected OFA 
could still complete the oxidation of UC to generate CO2. 
Hence, despite the obvious CO2 decrement as NH3 increased, 
the main finding from this assessment is that the staging air 
element has literally no impact on CO2 emissions during coal-
NH3 combustion. 

 
Figure 6. Temperature results from numerical and experimental 
works (cases C12 and N42) 

 

 
Figure 7. CO2 data at varying NH3 ratios and ASRs 



MN. Rahman et al. /Future Energy                                                                                          February 2026| Volume 05 | Issue 01| Pages 41-52 

48 

 

4.4 Impacts on SO2 emission 
As shown in Figure 8, the SO2 concentrations in flue gas 

altered significantly when NH3 was co-fired in the coal 
combustor. The concentration of SO2 in flue gas was 
substantially lower in NH3 co-firing scenarios than in the coal-
fired base case. As the NH3 co-firing ratio increased, the SO2 
concentration decreased. When the NH3 co-firing ratio 
reached 60 cal.%, SO2 emissions plummeted to 12 ppm (for 0 
% ASR), representing a 75 % reduction in SO2 emissions as 
opposed to the coal-fired base case.  

When the NH3 co-firing ratio reached 60 cal.%, SO2 
emissions dropped to 13.0, 13.1, and 13.2 ppm at ASRs of 20, 
40, and 60 %, respectively. When compared to the coal-fired 
base scenario, this translates to a 71 to 72 % reduction in SO2 
emissions. The numerical results show that the SO2 emission 
ranges for coal firing, 20, 40, and 60 cal.% NH3 cases are 45.5 
to 47.0 ppm, 34.1 to 36.7 ppm, 23.0 to 25.1 ppm, and 12.0 to 
13.2 ppm, respectively. Hence, raising the amount of NH3 in 
the fuel to 20, 40, and 60 cal.% reduces SO2 emissions by 
approximately 24.1, 47.9, and 72.5 %, respectively, as 
compared to the coal-fired case.  

Because this assessment was conducted at a pilot coal 
combustor testing facility, the injected fuel flow rates are 
significantly lower than the actual fuel flow rates in coal-fired 
power plants. As a result, the predicted SO2 concentration 
values in Figure 8 are very low in comparison to the actual SO2 
emissions from industrial coal combustion systems such as 
coal-fired power plants. Moreover, it is crucial to note that the 
sulphur concentration of the sub-bituminous coal utilised in 
the current assessment (Table 1) is lower than the sulphur 
content of common sub-bituminous coals. Hence, the SO2 
concentration in the coal-fired base scenario is already 
substantially lower. Nonetheless, for parametric analysis, the 
percentage reduction of SO2 when NH3 was co-fired is one of 
the most relevant research outcomes. NH3 is not only a 
carbon-free fuel, but it also contains no sulphur. As a result, 
increasing the amount of NH3 has resulted in a significant 
decrease in SO2 concentrations due to a lack of sulphur as a 
portion of coal is replaced by NH3. 

The percentage disparity in SO2 emissions between the 
lowest and highest ASRs is likewise less than 10 % in all cases 
shown in Figure 8. As a result, the variation in ASR has no 
discernible influence on SO2 emissions. However, notable 
discoveries can be found in the coal-fired base case, where no 
ASR was used. It was predicted that a slightly higher SO2 
concentration would occur in this case. The rise in SO2 
emissions might be explained by the leaner fuel condition 
(main combustion zone) in the coal firing case, where no ASR 
was used. With greater oxygen available in the main 
combustion zone, it reduces the likelihood of other 
sulphurous substances such as hydrogen sulphide (H2S), 
carbonyl sulphide (COS), and carbon disulphide (CS2) being 
created, as most of the available oxygen oxidises to become 
SO2. This is also corroborated by prior findings that a fuel-rich 
region will lead to a reduction in SO2 emissions from coal 
combustion [35]. As a result, for the coal firing situations 
depicted in Figure 8, the adoption of an ASR is predicted to 
result in lower SO2 emissions than the no staging ratio case 
due to the fuel-rich condition in the primary combustion zone 
as oxygen availability decreases. 

Furthermore, during coal devolatilization, a portion of 
sulphur is liberated from the coal as one of the volatiles, and 
the remainder of sulphur remains within the char (residual of 
coal after devolatilization) [35]. As a result, sulphur retention 
in char/UC is suggested to be another explanation for slightly 
reduced SO2 in the ASR cases. This is because, in the ASR 

cases, the fuel-rich zone occurs within the main combustion 
zone, so the devolatilization intensity decreases, lowering the 
char combustion intensity, which is mostly affected by 
volatiles combustion. Hence, in a fuel-rich environment of 
coal firing, the sulphur in char/UC has the tendency to release 
a lesser amount of sulphur than in a fuel-lean environment. 

 

 

Figure 8. SO2 data at varying NH3 ratios and ASRs 

4.5 Impacts on NOx emission 
The NOx concentrations in flue gas changed greatly when 

NH3 was co-fired in the coal combustor, as shown in Figure 9. 
NOx concentrations in flue gas were notably higher in NH3 co-
firing scenarios than in the coal-fired baseline case. The NOx 
concentration grew as the NH3 co-firing ratio increased. When 
the NH3 co-firing ratio reached 60 cal.%, NOx emissions 
soared to 1,918 ppm (for 0 % ASR), reflecting a staggering 
632.06 % increase in NOx emissions, approximately six times 
that of the coal-fired base case.   

The numerical findings show that the NOx emission 
ranges for coal firing, 20, 40, and 60 cal.% NH3 cases are 250 
to 282 ppm, 492 to 895 ppm, 685 to 1,460 ppm, and 1221 to 
1918 ppm, respectively. Therefore, compared to the coal-
fired scenario, increasing the amount of NH3 in the fuel to 20, 
40, and 60 cal. % increases NOx emissions by an average of 
146.5, 311.1, and 480.2 %, respectively. 

However, as the NH3 co-firing ratio reached 60 cal.%, NOx 
emissions reduced to 1,670 and 1,221 ppm for ASRs of 20 and 
40 %, respectively. Yet, as soon as the staging ratio for the 
aforesaid NH3 co-firing ratio hit 60 %, the NOx emission began 
to increase. The NOx reduction trend is also visible in 20 and 
40 cal.% of NH3 co-firing cases. NOx emissions were reduced 
to 672, 532, and 492 ppm in NH3 20 cal.% co-firing cases with 
ASR increases of 20, 40, and 60 %, respectively. With an 
increase in ASR of 20, 40, and 60 % for NH3 40 cal.% co-firing, 
NOx emissions were lowered to 1,383, 793, and 685 ppm, 
respectively. 

Furthermore, with an ASR of 60 %, the NOx emission 
from the 20 and 40 cal.% NH3 co-firing approaches the NOx 
emission from the coal firing base case. Using a 60 % ASR, the 
percentage difference between the coal firing base case and 
NH3 co-firing cases was reduced to 74.5 (20 cal.% NH3) and 
142.9 % (40 cal.% NH3). These percentage differences are 
much lower than those obtained when no air staging was 
used, which are 241.6 (40 cal.% NH3), 457.3 (40 cal.% NH3), 
and 632.1 % (60 cal.% NH3). Therefore, it was expected that 
the use of air staging would greatly aid in reducing NOx when 
NH3 was co-fired. 

 

 

 



MN. Rahman et al. /Future Energy                                                                                          February 2026| Volume 05 | Issue 01| Pages 41-52 

49 

 

 

Figure 9. NOx data at varying NH3 ratios and ASRs 

Adopting a 60 % ASR severely curtailed the airflow for 
the main combustion zone, resulting in an extremely rich in 
fuel state with a large deficit in oxygen concentration. As a 
result, it significantly reduces the oxidation intensity with the 
nitrogen in fuel, resulting in a reduction in fuel NOx. 
Furthermore, the very fuel-rich main combustion zone 
reduces temperature, which reduces the creation of thermal 
NOx [36]. The remaining injected airflow (60 %) is supplied 
by the OFA technology above the main combustion zone via a 
separate injector situated above the said zone. The firing 
process is effectively terminated in this OFA zone. 
Consequently, the relatively low temperature in the OFA 
injection zone helps to further reduce the production of 
thermal NOx. Furthermore, the injected SA surrounded the 
core combustion zone. As a result, a relatively low 
temperature in the oxygen-enriched afterburning zone 
provides a further reduction in the generation of thermal NOx. 
This is a common approach used in coal-fired boiler Low NOx 
burners [36]. Previous kinetic studies have also supported the 
aforementioned findings, revealing that for NH3 co-firing with 
coal, thermal NOx and fuel NOx are the dominant NOx types 
produced [4, 12]. 

Thus, it can be observed that limiting oxygen availability 
during the crucial stage of coal devolatilization is the most 
efficient way to reduce NOx formation from the NH3 co-firing. 
The coal devolatilization is hypothesised to occur primarily in 
the primary combustion zone due to its relatively short 
processes as opposed to the char oxidation [30, 37] since the 
combustion zone in the current coal combustor is relatively 
smaller than that in the actual coal-fired boiler. Later on in the 
operation, more air (oxygen) can be introduced by the OFA 
technology to finish char reactions, lower UC, and maintain 
high combustion efficiency [22]. 

However, as the ASR surpasses 40 %, the NOx reduction 
gradient begins to decrease for 20 and 40 cal.% NH3 but 
begins to climb for 60 cal.% NH3. Previous research has also 
indicated that when the ASR crosses a particular quantity, NOx 
levels begin to rise [38]. This is because when the ASR exceeds 
a particular threshold value, a large proportion of NOx 
production is repressed at the primary combustion zone, 
where the air ratio is small. Yet, a significant amount of UC 
stays in the said position and is burned at the OFA zone to 
complete its oxidation. Since a substantial quantity of UC is 
burned in the OFA zone, the combustion intensity increases, 
causing the nitrogen content in the fuel to oxidize and 
produce NOx in the OFA zone. In other words, while the main 
combustion zone generates the highest temperature along the 
coal combustor, the high UC content at the OFA zone 
generates another peak combustion temperature, commonly 
known as delayed combustion, resulting in nitrogen oxidation 
in fuel at a high temperature in the OFA zone.  

Figure 10 depicts temperature contours within the coal 
combustor model (excluding the convection zone – shorter 

length region) at NH3 60 cal.% at various ASRs. There is a 
completely visible second temperature peak at the OFA zone, 
especially at 60 % ASR. Hence, it promotes the development 
of thermal NOx at the OFA zone. The temperature distribution 
near the outlet, where the temperature is higher as opposed 
to lower/no ASRs, can also indicate delayed combustion at 
high ASRs. Figure 10 also shows that the flame temperature is 
higher in the absence of ASR than in its presence, indicating 
that more oxygen is available to achieve a higher combustion 
temperature. The flame temperature has been observed to 
decrease as ASR increases due to a reduction in available 
oxygen in the main firing area. 
 

Figure 10. Temperature contours (NH3 60 cal. %) at (a) 0 %, (b) 20 
%, (c) 40 %, and (d) 60 % ASRs 

In Figure 9, the degree of ASR has a smaller effect on NOx 
emissions in the coal-firing base scenario than in the NH3 co-
firing scenarios. The predicted NOx levels decreased slightly, 
from 262 to 250 ppm, as the ASR increased from 0 to 20 % in 
the coal-fired base scenario. However, when the staging ratio 
reaches 40% and 60 %, there is a slight increase in NOx in the 
coal-firing base scenario. The increase, however, is not as 
significant as in the NH3 60 cal.% case. As previously stated, 
devolatilization occurred primarily in the main combustion 
zone. Because the char reaction has a faster ignition rate than 
coal, the VM of coal and NH3 are assumed to burn in the 
primary firing zone [32], resulting in a slight increase when 
ASR 40 and 60 % were implemented in the coal firing base 
scenario. The amount of these components in NH3 co-firing, 
however, is greater than in coal firing. Therefore, it is 
postulated that when NH3 co-firing reaches 60 cal.%, along 
with a high ASR (60 %), the air-to-VM ratio in the main 
combustion zone is lower than in the coal-only case. Hence, a 
substantially larger amount of char/UC remains, which is 
burned at the OFA position, causing thermal NOx to be much 
higher than in the coal firing case. 



MN. Rahman et al. /Future Energy                                                                                          February 2026| Volume 05 | Issue 01| Pages 41-52 

50 

 

Furthermore, these findings were supported by a 
comprehensive kinetics reactor network 0D modelling by 
Ishihara [7]. Despite the fact that their analysis ignored the 
impacts of fluid dynamics and 3D characteristics, their 
kinetics pathway showed that for a relatively high NH3 co-
firing ratio, complete NH3 reactions cannot be achieved in the 
primary firing area. As a result, it demonstrates the prospect 
of nitrogen-related radicals being combusted further 
upstream in the coal combustor region, such as the OFA zone 
with significant oxygen availability, as predicted by current 
numerical assessments. The devolatilization, volatiles 
reaction, and NH3 reaction are more prominent in the 
combustion zone than the char/UC reactions due to the 
significant amount of VM and NH3 in 60 cal.% NH3 co-firing. 
The aforesaid hypothesis has also been supported by prior 
research, which indicated that NH3 co-firing, both in fuel-rich 
and fuel-lean conditions, promoted coal devolatilization and 
volatile release [8]. Furthermore, with the use of sub-
bituminous coal in the current study, which is known to have 
a higher VM than common bituminous coals [25], the 
devolatilization and volatiles reactions will dominate even 
more. Therefore, the current findings support prior 0D kinetic 
modelling studies that revealed that NH3 reactions inhibit 
char oxidation [7]. 

5. Conclusion 

In the context of reducing CO2 emissions from pulverised 

coal-fired boilers, the paper studied the potential of NH3 to be 

utilised as a carbon-free substitute for coal, at least partially, 

to aid in the progressive phase-out of coal. In this research, 

the co-firing of NH3 with sub-bituminous coal was 

numerically studied via a detailed LES assessment at various 

ASRs in a small-scale coal-fired testing combustor. In-depth 

insight into predicted emissions, including CO2, SO2, and NOx, 

was successfully obtained, and an appropriate ASR for a 

number of NH3 co-firing percentages was discovered to 

enable the reduction of such emissions. The prediction 

accuracy of the model’s results was first evaluated by 

validating it with actual testing data from TNB Research’s coal 

combustor testing facility, which revealed a below 10 % 

difference in NOx emissions and temperature results for both 

100% coal firing and NH3 co-firing cases. As a result, the 

validation can be deemed satisfactory, and the model can 

forecast the expected emissions with reliable accuracy for 

both these cases. All in all, the research has demonstrated that 

coal+NH3 co-firing can produce low CO2, SO2, and NOx 

emissions as opposed to pure coal-firing with proper ASR 

tuning. As the variation of ASRs was studied, important 

findings regarding emission characteristics were discovered, 

where the synergistic effect of NH3 co-firing with coal plays a 

significant part in the selection of ASR to be used at a specific 

NH3 co-firing case. The increase in NOx when the staging ratio 

reached 60 % revealed that devolatilization occurred 

primarily in the main combustion zone due to the smaller 

firing space compared to actual coal-fired power plants, as 

well as the shorter devolatilization duration compared to char 

reaction. As a result, reaction, VM of coal, and NH3 are 

assumed to be what burns at the primary firing zone, with the 

amount of these components in NH3 co-firing being greater 

than in a pure coal firing scenario. This causes considerable 

UC oxidation in the OFA zone, which has high oxygen 

availability, and results in a relatively higher combustion 

temperature in the OFA zone. These findings contribute to 

vital information for combustion tuning in actual coal-fired 

power plants to achieve NH3+coal co-co-firing and reduce CO2 

emissions. The numerical results also confirmed that NH3 can 

be used as an alternative fuel to reduce CO2 emissions in 

actual coal-fired power boilers. The air staging method has 

been shown to provide reduced CO2, SO2, and NOx emissions 

when the ASR is properly tuned to ensure acceptable staging 

combustion within the furnace. As a result, existing coal-fired 

power plant air staging technology can be utilised to provide 

proper NH3 co-firing while attaining low NOx emissions. This 

is critical information for retrofit setups because it is possible 

that the only adjustment required is the addition of a new NH3 

burner. However, in the current work, the combustion zone is 

simplified, and the effects of multiple burners (which actual 

coal-fired boilers have), such as flame interplay, are not fully 

examined. Plus, the size of the coal combustor employed is 

smaller than that of a typical coal-fired utility boiler. When 

several burners are used, as in commercial boilers, it is 

expected that estimating the amount of emissions will 

become more complex due to the presence of additional flame 

interaction and/or the stronger influence of mixing residence 

time. Further research can be conducted to investigate the 

impact of multiple burners and different NH3 burner designs 

on emission and combustion characteristics. It is also 

expected that a higher NH3 co-firing ratio will be required in 

the future to accomplish further CO2 emission reductions. 

Hence, future research can include an increase in the NH3 co-

firing ratio. 

Ethical issue 
The authors are aware of and comply with best practices in 
publication ethics, specifically concerning authorship 
(avoidance of guest authorship), dual submission, 
manipulation of figures, competing interests, and compliance 
with policies on research ethics. The authors adhere to 
publication requirements that the submitted work is original 
and has not been published elsewhere in any language. 

Data availability statement 
The manuscript contains all the data. However, more data will 

be available upon request from the corresponding author. 

Conflict of interest 

This work was funded by the TNB Research Seeding Fund 

(Grant No. TNBR/SF 429/2022). All research activities and 

analysis associated with this paper were completed by the 

authors during their respective tenures at TNB Research, 

regardless of their current affiliations at the time of 

submission. 

References 

[1] Cesaro Z, Ives M, Nayak-Luke R, Mason M, Banares-
Alcantara R. Ammonia to power: Forecasting the 
levelized cost of electricity from green ammonia in 
large-scale power plants. Appl Energy. 2021;  
https://doi.org/10.1016/j.apenergy.2020.116009 

[2] Cardoso JS, Silva V, Chavando JAM, Eusebio D, Hall MJ. 
Numerical modelling of the coal phase-out through 
ammonia and biomass co-firing in a pilot-scale 
fluidized bed reactor. JFUECO. 2022;  
https://doi.org/10.1016/j.jfueco.2022.100055 

[3] Zhang J, Ito T, Ishii H, Ishihara S, Fujimori T. 
Numerical investigation on ammonia co-firing in a 
pulverized coal combustion facility: Effect of 



MN. Rahman et al. /Future Energy                                                                                          February 2026| Volume 05 | Issue 01| Pages 41-52 

51 

 

ammonia co-firing ratio. Fuel. 2020; 
https://doi.org/10.1016/j.fuel.2020.117166 

[4] Tsukada N, Kinoshita N, Kabuki Y, Taguchi Y, 
Takashima Y, Tsumura T, Taniguchi M. Role of OH 
Radical in Fuel-NOx Formation during Cocombustion 
of Ammonia with Hydrogen, Methane, Coal, and 
Biomass. Energy Fuels. 2020; 
https://dx.doi.org/10.1021/acs.energyfuels.0c00356 

[5] Weng W, Li Zhongshan, Marshall P, Glarborg P. 
Participation of alkali and sulfur in ammonia 
combustion chemistry: Investigation for 
ammonia/solid fuel co-firing applications. Combust 
Flame. 2022;  
https://doi.org/10.1016/j.combustflame.2022.11223
6 

[6] Chen P, Jiang B, Wang H, Gu M, Fang Y, Wang P. 
Experimental and theoretical calculations study on 
heterogeneous reduction of NO by char/NH3 in the 
reduction zone of ammonia co-firing with pulverized 
coal: Influence of mineral Fe. Fuel. 2022; 
https://doi.org/10.1016/j.fuel.2021.122374 

[7] Ishihara S, Zhang J, Ito T. Numerical calculation with 
detailed chemistry on ammonia co-firing in a coal-
fired boiler: Effect of ammonia co-firing ratio on NO 
emissions. Fuel. 2020; 
https://doi.org/10.1016/j.fuel.2020.117742 

[8] Zhu J, Liu X, Xu Y, Xu J, Wang H, Zhang K, Cheng X, Yu 
D. Probing into Volatile Combustion Flame and 
Particulate Formation Behavior During the Coal and 
Ammonia Co-firing Process. Energy Fuels. 2021; 
https://doi.org/10.1021/acs.energyfuels.2c01450 

[9] Wang X, Fan W, Chen J, Feng G, Zhang X. Experimental 
study and kinetic analysis of the impact of ammonia 
co-firing ratio on products formation characteristics 
in ammonia/coal co-firing process. Fuel. 2022; 
https://doi.org/10.1016/j.fuel.2022.125496 

[10] Yuan T, Xu Y, Wang C, Gao P, Zhao P, Zhao L, Jin C, Li 
C, He B. Calculation of CO2 emissions from coal-fired 
power plants based on OCO-2/3 satellite 
observations and divergence model: Energy. 2025; 
https://doi.org/10.1016/j.energy.2024.134303  

[11] Chen P, Fang Y, Wang P, Gu M, Luo K, Fan J. The effect 
of ammonia co-firing on NO heterogeneous reduction 
in the high-temperature reduction zone of coal air-
staging combustion: Experimental and quantum 
chemistry study. Combust Flame. 2022; 
https://doi.org/10.1016/j.combustflame.2021.11185
7 

[12] Ishihara S, Zhang J, Ito T. Numerical calculation with 
detailed chemistry of effect of ammonia co-firing on 
NO emissions in a coal-fired boiler. Fuel. 2020; 
https://doi.org/10.1016/j.fuel.2019.116924 

[13] Rahman MN, Wahid MA. Renewable-based zero-
carbon fuels for the use of power generation: A case 
study in Malaysia supported by updated 
developments worldwide. Energy Rep. 2021; 
https://doi.org/10.1016/j.egyr.2021.04.005 

[14] Rahman MN, Ujir MH, Wahid MA, Yasin MFM. A 
single‑step chemistry mechanism for biogas 
supersonic combustion velocity with nitrogen 
dilution. J Therm Anal Calorim. 2022; 
https://doi.org/10.1007/s10973-022-11356-x 

[15] Rahman MN, Shahril N, Yusup S. Hydrogen-Enriched 
Natural Gas Swirling Flame Characteristics: A 
Numerical Analysis. CFD Lett. 2022; 
https://doi.org/10.37934/cfdl.14.7.100112 

[16] Chen P, Wang H, Jiang B, Wang Y, Gu M, Chen G, 
Huang X. An experimental and theoretical study of NO 
heterogeneous reduction in the reduction zone of 
ammonia co-firing in a coal-fired boiler: Influence of 
CO. Fuel Process Technol. 2022; 
https://doi.org/10.1016/j.fuproc.2022.107184 

[17] Cardoso JS, Silva V, Eusebio D, Tarelho LAC, Hall MJ, 
Dana AG. Numerical modelling of ammonia-coal co-
firing in a pilot-scale fluidized bed reactor: Influence 
of ammonia addition for emissions control. Energy 
Convers Manag. 2022; 
https://doi.org/10.1016/j.enconman.2022.115226 

[18] Stocks M, Fazeli R, Hughes L, Beck FJ. Global 
emissions implications from co-combusting ammonia 
in coal fired power stations: An analysis of the Japan-
Australia supply chain. J Clean Prod. 2022; 
https://doi.org/10.1016/j.jclepro.2021.130092 

[19] Valera-Medina A, Xiao H, Owen-Jones M, David WIF, 
Bowen PJ. Ammonia for power. Prog Energy Combust 
Sci. 2018; 
https://doi.org/10.1016/j.pecs.2018.07.001 

[20] Tamura M, Gotou T, Ishii H, Riechelmann D. 
Experimental investigation of ammonia combustion 
in a bench scale 1.2 MW-thermal pulverised coal 
firing furnace. Appl Energy. 2020; 
https://doi.org/10.1016/j.apenergy.2020.115580 

[21] Ouyang Z, Liu W, Man C, Zhu J, Liu J. Experimental 
study on combustion, flame and NOX emission of 
pulverized coal preheated by a preheating burner. 
Fuel Process Technol. 2018; 
https://doi.org/10.1016/j.fuproc.2018.07.008 

[22] Kim W, Lee D, Park S. Experimental study on 
optimization of over-fire air in modified combustion 
condition with selective catalytic reduction. J Mech 
Sci Technol. 2011; https://doi.org/10.1007/s12206-
011-0208-3 

[23] Lee H, Lee M. Recent Advances in Ammonia 
Combustion Technology in Thermal Power 
Generation System for Carbon Emission Reduction. 
Energies. 2021; 
https://doi.org/10.3390/en14185604 

[24] Pestheruwe N, Huangwei Z, Lin M, Ruoyang Y. 
Numerical simulations on the combustion and 
emission characteristics of a non-premixed NH3/CH4 
swirling flame using LES-FGM method. AIAA SCITECH 
Forum. 2023. https://doi.org/10.2514/6.2023-1722 

[25] Dounia O, Vermorel O, Misdariis A, Poinsot T. 
Influence of kinetics on DDT simulations. Combustion 
and Flame. 2019. 
https://doi.org/10.1016/j.combustflame.2018.11.00
9 

[26] Taamallah S, Vogiatzaki K, Alzahrani FM, Mokheimer 
EM, Habib MA, Ghoniem AF. Fuel flexibility, stability 
and emissions in premixed hydrogen-rich gas turbine 
combustion: Technology, fundamentals, and 
numerical simulations. Applied energy. 2015. 
https://doi.org/10.1016/j.apenergy.2015.04.044 

[27] Meloni R, Nassini PC, Andreini A. Model development 
for the simulation of the hydrogen addition effect 
onto the NOx emission of an industrial combustor. 
Fuel. 2022. 
https://doi.org/10.1016/j.fuel.2022.125278 

[28] Zheng K, Yu M, Liang Y, Zheng L, Wen X. Large eddy 
simulation of premixed hydrogen/methane/air flame 
propagation in a closed duct. International Journal of 



MN. Rahman et al. /Future Energy                                                                                          February 2026| Volume 05 | Issue 01| Pages 41-52 

52 

 

Hydrogen Energy. 2018. 
https://doi.org/10.1016/j.ijhydene.2018.01.045 

[29] Rahman MN, Othman NF. A Numerical Model for Ash 
Deposition Based on Actual Operating Conditions of a 
700 MW Coal-Fired Power Plant: Validation Feedback 
Loop via Structural Similarity Indexes (SSIMs). CFD 
Lett. 2022; https://doi.org/10.37934/cfdl.14.1.99111 

[30] Rahman MN. Optimisation of Solid Fuel In-furnace 
Blending for an Opposed-firing Utility Boiler: A 
Numerical Analysis. CFD Lett. 2022; 
https://doi.org/10.37934/cfdl.14.9.89107 

[31] Sun W, Zhong W, Echekki T. Large eddy simulation of 
non-premixed pulverized coal combustion in corner-
fired furnace for various excess air ratios. Appl Math 
Model. 2019; 
https://doi.org/10.1016/j.apm.2019.05.017 

[32] Wan K, Xia J, Wang Z, Pourkashanian M, Cen K. Large-
eddy Simulation of Pilot-assisted Pulverized-coal 
Combustion in a Weakly Turbulent Jet. Flow 
Turbulence Combust. 2017; 
https://doi.org/10.1007/s10494-017-9817-y 

[33] Sun W, Zhong W, Zhang J, Echekki T. Large Eddy 
Simulation on the Effects of Coal Particles Size on 
Turbulent Combustion Characteristics and NOx 
Formation Inside a Corner-Fired Furnace. J Energy 
Resour Technol. 2021; 
https://doi.org/10.1115/1.4048864 

[34] Silva VB, João C. Computational fluid dynamics 
applied to waste to energy processes: a hands-on 
approach. Butterworth-Heinemann; 2020. 

[35] Hu Y, Naito S, Kobayashi N, Hasatani M. CO2, NOx and 
SO2 emissions from the combustion of coal with high 
oxygen concentration gases. Fuel. 2000; 
https://doi.org/10.1016/S0016-2361(00)00047-8 

[36] Askarova AS, Messerle VE, Ustimenko AB, Bolegenova 
SA, Maximov VY, Yergalieva AB. Reduction of noxious 
substance emissions at the pulverized fuel 
combustion in the combustor of the BKZ-160 boiler of 
the Almaty heat electropower station using the 
“Overfire Air” technology*. Thermophys 
Aeromechanics. 2016; 
https://doi.org/10.1134/S0869864316010133 

[37] Speight JG. Coal-fired power generation handbook. 
John Wiley & Sons; 2021. ISBN: 978-1-119-51013-0 

[38] Nagatani G, Ishii H, Ito K, Ohno E, Okuma Y. 
Development of Co-Firing Method of Pulverized Coal 
and Ammonia to Reduce Greenhouse Gas Emissions. 
IHI Engineering Review. 2020. 
https://www.ihi.co.jp/en/technology/sdgs/topic01/
pdf/Vol53No1_F.pdf 

 
This article is an open-access article distributed under the 

terms and conditions of the Creative Commons Attribution 

(CC BY) license 

(https://creativecommons.org/licenses/by/4.0/). 

Abbreviations 

CFD Computational Fluid Dynamics 
COP26 UN's Climate Change Conference 
IGCC Integrated Gasification Combined Cycle 
CCS Carbon Capture and Storage 
UC Unburned carbon 
ASR Air Staging Ratio 
TM Total moisture 
FC Fixed carbon 
GCV Gross Calorific Value 
H Hydrogen element 
O Oxygen element 
OFA Over-fire Air 
PA Primary Air 
DO Discrete Ordinate 
HCN Hydrogen Cyanide 
CO  Carbon Monoxide 
H2S Hydrogen Sulphide 
CS2 Carbon Disulphide 
NOx Nitrogen Oxides 
NH3 Ammonia 
SO2 Sulphur Dioxide 
CO2 Carbon Dioxide  
SA Secondary Air 
LPG Liquefied Petroleum Gas 
VM Volatile matter 
AC Ash content 
C Carbon element 
N Nitrogen element 
S Sulphur element 
NS Navier–Stokes 
LES Large Eddy Simulation 
WSGGM Weighted-sum-of-gray-gases Model 
OH Hydroxide 
ad Air-dried basis 
COS Carbonyl Sulphide 

https://creativecommons.org/licenses/by/4.0/

