Boukhelef, M., et al.: Numerical Study of the Injection Conditions Effect ...

THERMAL SCIENCE: Year 2022, Vol. 26, No. 5A, pp. 3741-3750 3741

NUMERICAL STUDY OF THE INJECTION CONDITIONS EFFECT

ON THE BEHAVIOR OF HYDROGEN-AIR DIFFUSION FLAME
by

Mohamed BOUKHELEF a, Mounir ALLICHE b*,

Mohammed SENOUCI a,c, and Habib MEROUANE a
a
Laboratory of Physic Quantic and Mathematical Modelization of Matter (LPQ3M),
University of Mascara, Mascara, Algeria
b
Renewable Energy and Materials Laboratory (LERM), Faculty of Technology,
University of Medea, Medea, Algeria
c
Higher School in Electrical and Energy Engineering of Oran, USTO, Oran, Algeria
Original scientific paper
https://doi.org/10.2298/TSCI210530314B

In order to respond to the increased demand for clean energy without harming
the atmosphere through polluting emissions, energy production from the hydrogen
combustion become largely used. This work presents a numerical study of the in-
jection conditions effect on the structure of the H2-air diffusion flame. The aim is to
reproduce a practical case of non-polluting combustion and resulting in very high
temperatures. The configuration is composed of two axisymmetric coaxial jets, as
can be found in the diffusion burners. A presumed probability density function
approach is used to describe the chemistry-turbulence interaction. The k-ε model
of turbulence is used. Particular attention is given to phenomena anchoring or
blowout of the flame.
Key words: diffusion flame, H2-air, PDF, turbulence, k-ε model, CFD

Introduction
The development of energy systems requires substantial support in terms of fossil re-
sources. The activities inherent to domestic and industrial needs require the supply of fuels, the
majority of which are hydrocarbons. In addition availability and cost issues, products derived
from petroleum and natural gas provide great energy, but release large amounts of pollutants
[1, 2]. As a result, the international community is turning to alternative fuels, offering good
ecological performance at attractive production costs [3-6].
In this context, hydrogen appears to satisfy these criteria. In fact, it offers a low energy
level at ignition and has a wide range of flammability, but still has the disadvantage of having a
low density and to be very diffuse, which leads to technological problems, respectively related
to sealing and storage [7, 8]. Although premixed flames are less polluting and more used in
energy installations as diffusion flames, premixed hydrogen-air flame presents great risks of
transition detonation and return of flames (flashback) [9-11].
As a fuel, hydrogen is involved in a several applications from internal combustion
engines where it is mixed with natural gas, through gas turbines (mixed with methane) to
(H2-air) burners (low NOx), and rocket engines (H2-O2) [12-15].

*
Corresponding author, e-mails: alliche.mounir@univ-medea.dz; alliche_m@yahoo.fr
 Boukhelef, M., et al.: Numerical Study of the Injection Conditions Effect ...
3742 THERMAL SCIENCE: Year 2022, Vol. 26, No. 5A, pp. 3741-3750

The aim of this investigation is to use CFD approach as an effective computer based
simulationol to study and develop the hydrogen combustion processes [16]. Several con-
straints can get involved to make an experiments of combustion not recommended where,
these ones may cause a harmful danger to the entourage [2, 5]. For the hydrogen combustion
case is even worse, due to the characteristics of the hydrogen substance itself. This is the
reason why it is preferred to use a high fidelity computation systems [16-18]. The CFD and
modelling techniques are becoming increasingly important tools to assess the impact of hy-
drogen combustion in the operating burners [2, 15]. Developments in the field of computer
simulation would be extremely useful in predicting flames and combustion process, without
the need for expensive and time-consuming field tests. However, the complete description
of particle trajectories, chemistry reactions and energy production is needed. This numerical
investigation is about making a numerical approach to comprehend how injection conditions
can affect the hydrogen flame behavior which is dependent to the same geometric configura-
tion [5, 16].
In the present study, the impacts of tube diameter and fuel velocity on combustion
efficiency were examined. The present work mainly concerns the numerical simulation of the
phenomenon of turbulent combustion, in a non-premixed flame of the H2-air type. The aim is
to produce a numerical investigation of non-polluting combustion and giving rise to very high
temperatures (of the order of 2000 K) [19]. The studied configuration is composed of two co-
axial axis-symmetric jets, such as can be found in diffusion burners. In this simulation, the k-ε
turbulence model was used with a reduced chemical mechanism. We examine the effect of in-
jection conditions on the structure of the diffusion flame, more specifically, speed, temperature
and static pressure. We are particularly interested in the phenomena of attachment or blowing
of the flame.
Operating conditions
The studied configuration is a set of
two turbulent jets in an axis-symmetric burn-
er, composed of a coaxial injector, (r1 and r2,
of length, l, opening on a cylindrical pipe of
length, L, and radius, R , fig. 1. An air stream
Figure 1. Geometrical configuration of standard composition (O2% = 21.008%,
N2% = 78.992%) is sent via the axial injector at a temperature Tair = 300 K, with a turbulence
intensity of the order of 15% which corresponds to a fluctuating speed given:
15
′ =
Vair V (1)
100
The combustible stream (H2% = 100%) is blown peripherally by the annular injector
with an average speed VH2=200 m/s and a temperature TH2 = 300 K. The blowing is character-
ized by an intensity of turbulence equal to 15%:
The injection Reynolds number related to the fuel flow is given:
ρ H 2 VH 2
= R H2 = r1 8934 (2)
µH2
Mathematical modelling
In this study, a CFD approach is applied. The prediction of the aerothermochemical
variables of fluid-flow is available by solving the RANS equations, energy and transport of
Boukhelef, M., et al.: Numerical Study of the Injection Conditions Effect ...
THERMAL SCIENCE: Year 2022, Vol. 26, No. 5A, pp. 3741-3750 3743

chemical species [20-22]. Turbulence is described by k-ɛ model [16-25]. This is useful in order
to correct an overestimation due to radial expansion of the round jet coming from the ring in-
jector [26, 27].
Thus, the governing equations include the continuity eq. (3), momentum eq. (4), and
additional conservation equations, such as energy eq. (6) and species concentrations eq. (7). In
turbulence circumstance, two others equations are added eqs. (10) and (11).
We can write these equations as following [20-28]:
–– Continuity equation
∂ρ ∂
+ ( ρ ui ) =
0 (3)
∂t ∂xi
–– Momentum equations

∂   ∂ui ∂u j 2 ∂ul   ∂
∂
∂t
( ρ ui ) +
∂
(
∂xi
ρ ui u j =
−)∂p
+ µ  + − δ ij
∂xi ∂x j   ∂x j ∂xi 3
 +
∂xl   ∂x j
− ρ ui′u ′j ( ) (4)
 
For combustion cases, all fluids are assumed as Newtonian and the viscous stress
tensor:
 ∂ui ∂u j  2  ∂uk 
τ ij = µ  +  − µδ ij   (5)
 ∂x j ∂xi  3  ∂xk 
where µ is the molecular viscosity which depends on the fluid. The Kronecker delta is δ = 1, if
i = j, 0 otherwise.
–– Energy equation

∂     
∂t
( ρ E ) + ∇ V ( ρ E + p ) = ∇  keff ∇T −
 ∑ ( )
h j J j + τ eff V  + Sh

(6)
 j 
where keff is the effective conductivity, Jj – the diffusion flux of species j, and Sh – the combina-
tion of heat chemical reactant and every other heat resource which are defined.
–– Species equation
∂ ( ρYα ) (
∂ ρ u j Yα )= ∂J αj
α= (7)
+ − +w α, 1, 2,3..., n
∂t ∂x j ∂x j
where n is the number of species, J αj – the molecular diffusivity flux of the species α in the jth
co-ordinate direction, wα – the mass reaction rate of this species per unit volume, and Yα – the
mass fraction of species α.
The diffusive flux, J αj , can be approximated:
µ ∂Yα ∂Y
J αj =− =− ρ Dα α (8)
Scα ∂x j ∂x j
where Scα is the Schmidt number of the species α, defined:
µ
Scα = (9)
ρ Dα
where D is the molecular diffusivity of the species α relative to the other species.
 Boukhelef, M., et al.: Numerical Study of the Injection Conditions Effect ...
3744 THERMAL SCIENCE: Year 2022, Vol. 26, No. 5A, pp. 3741-3750

Turbulence-combustion modelling
The RANS are used to calculate the transport of the averaged flow quantities, with the
entire range of turbulent time and length scales being modeled. The RANS approach greatly
reduces the required computational effort and resources compared to LES and DNS. This is
why RANS method is widely adopted for practical engineering applications.
In this work, the RNG k-ε turbulence model is used for determining the turbulence
behavior which defined by the transport equations [5, 28, 29]:
∂ ∂ ∂  ∂k 
( ρ k )=
+ ( ρ kui ) α k µeff  + Gk + Gb − ρε − YM + Sk (10)
∂t ∂xi ∂x j  ∂x j 

∂ ∂ ∂  ∂ε  ε ε2
( ρε )=
+ ( ρε ui ) αε µeff  + C1ε ( Gk + C3ε Gb ) + C2ε − Rε + Sε (11)
∂t ∂xi ∂x j  ∂x j  k k

k2
=µt ρ=
Cµ Cµ ρϑ (12)
ε
Cµ ρη 3 (1 − ηη0 ) ε 2
Rε = (13)
1 + βη 3 k
where ϑ the velocity is scale and  – the length scale [30].
All the coefficients of eqs. (10)-(13) are illustrated in tab. 1.
Table 1. Constants values used for turbulence models [29]
Cμ σk σε C1ε C2ε
RNG k-ε 0.0845 1.3900 1.3900 1.4200 1.6800

where η ≡ Skε, η0 = 4.38, and β = 0.012, [29].

The effective viscosity is given by the next equation:
µeff= µ + µt (14)
where μ is the eddy-viscosity and μt – the turbulence viscosity.
The hydrogen combustion phenomena is modeled by using PDF/equilibrium mixture
fraction technic [18, 22, 30]. The PDF combustion model is based on the mixture fraction ap-
proach with an assumption of fast chemistry [4, 16]. This, offers some advantages over the EDS
or EDS-finite-rate models. In addition, it allows intermediate species prediction, more thorough
turbulence-chemistry coupling [6, 15, 30].
In order to determine how the injection conditions influenced the hydrogen flame
behavior, this study is divided in two parts; in the first, one air (23% O2, 79% N2) is used as
oxidizer; the chemical kinetics reactions used is reduced mechanism of twelve steps reactions
and nine species. The second one pure O2 (100% O2, 0% N2) with twelve steps reactions and
eight species. The two cases are based on Boivin’s mechanism [31].
To solve the RANS equation, a turbulence model is used for determining the tur-
bulence behavior in the momentum equations and, a combustion model is used to obtain the
time-averaged reaction rate for the both energy and space continuity.
In this work, the turbulence is modeled using the RNG k-ε turbulence model. The
hydrogen combination is treated using the PDF-equilibrium mixture fraction model. Also, the
heat transfer is treated via the discrete-beam method [22, 30].
Boukhelef, M., et al.: Numerical Study of the Injection Conditions Effect ...
THERMAL SCIENCE: Year 2022, Vol. 26, No. 5A, pp. 3741-3750 3745

Numerical method
In this investigation, we have used the CFD package Ansys FLUENT 12.0 in order
to model combustion and heat transfer phenomenon in a diffusion hydrogen-air (or hydro-
gen-di-oxygen) flame. All the governing equations including momentum, energy, species trans-
port, turbulence and chemical reactions are solved using the finite volume approach coupled
to turbulence and PDF models. The velocity and pressure fields are linked via SIMPLEC algo-
rithm. The grid nodes are interpolated via the second-order schemes [22].
The numerical simulations are carried out on 2-D, axis-symmetric, quadrilateral and
non-uniform grid-mesh of co-ordinates 14442 cells. The grid spacing decreases far from the
reaction zones, the mesh form is illustrated on fig. 2. A grid independence test was performed
which it leads to use 14442 cells grid, fig. 3.

Figure 2. Grid form type used in this study Figure 3. Grid independence test: temperature
distribution along the axial direction

The residual of the energy equation should be below 1–6 and for all other variables
it is set at 10–4 to ensure the convergence of the solution. The swirl velocity of components is
eliminated in the steady-state CFD study. To investigate various aspects of combustion, the
simulation is done in stoichiometric equivalence ratios (Φ = 1).
Second-order discretization scheme is applied to solve all governing equations. The
residual of the energy equation should be below 10–6 and for all other variables it is set at 10–4
to ensure the convergence of the solution. The swirl velocity of components is eliminated in the
steady-state CFD study. To investigate various aspects of combustion, the simulation is done in
stoichiometric equivalence ratios (Φ = 1).
The instantaneous thermochemical state is described by means of the mixing fraction
which represents a single passive scalar when the diffusion coefficients are equal for all chem-
ical species. The mixing fraction is defined [32]:
zi − zi ,ox
f = (15)
zi , f − zi ,ox

Results and discussion

Further analysis is conducted on the behavior of the diffusion flame with respect to
the composition of the oxidizing current, which we assume to be pure oxygen (100% O2, 0%
N2). The analysis involves varying the injection rate of the hydrogen flow and the temperature
of the oxygen.
 Boukhelef, M., et al.: Numerical Study of the Injection Conditions Effect ...
3746 THERMAL SCIENCE: Year 2022, Vol. 26, No. 5A, pp. 3741-3750

Influence of the air blowing speed

To study the behavior of the temperature in function of different air velocities in-
jection are performed. Hence, the temperature distribution analysis shows that whenever the
air blowing speed increases, the flame front es-
capes even more from the injection plane, but
the flame no longer takes off from the central
injector, fig. 4. It can be seen that the maximum
temperature is proportional to the air blowing
speed. In this context, a maximum temperature
of 1370 K is recorded for Vair = 5 m/s. The tem-
perature is rising to 1630K for Vair = 40 m/s.
This means that the consumption of hydrogen
increases, but the maximum temperature shifts
backwards, unlike the flame front which es-
capes forward, forming a kind of flame enve-
Figure 4. Effect of the air blowing speed lope by enclosing the jet air. This can consider-
on the behavior of the flame ably increase the air-hydrogen contact surface,
which promotes better combustion.
Study of the H2-O2 flame
We will deal here with the effect of two parameters, namely:
–– The hydrogen injection speed.
–– The oxygen inlet temperature.
Effect of hydrogen injection rate
In this case, the rate of introduction of
oxygen is 1 m/s and a temperature of 85 K. For
hydrogen, the inlet temperature is 287 K, inject-
ed at the following respective speeds; 90 m/s,
180 m/s, and 236 m/s. Figure 5 shows that the
maximum temperature increases significantly
depending on the rate of H2 injection. For an
axial station located at x = 480 mm, the max-
imum values recorded are 2848 K when VH2 =
Figure 5. Effect of hydrogen injection speed on 236 m/s and 2839 K for VH2 = 90 m/s. Which
the static temperature fields represents a speed difference about ΔVH2 = 146
m/s. This may indicate that the speed of hydro-
gen injection does not have a great effect on
the value of the maximum temperature to be
reach. On the other hand, we can clearly see its
influence on the shape of the flame.
Figure 6 illustrates the temperature vari-
ations along the direction of the central axis of
the combustion chamber. The curve for VH2 =
236 m/s is located above that of VH2 = 90 m/s,
this explains why the flame magnitude for
Figure 6. Effect of hydrogen injection VH2 = 236 m/s is slightly large compared to that
speed on the temperature
Boukhelef, M., et al.: Numerical Study of the Injection Conditions Effect ...
THERMAL SCIENCE: Year 2022, Vol. 26, No. 5A, pp. 3741-3750 3747

of VH2 = 90 m/s. The flame core keeps the same shape and illustrates a flame hanging from the
nozzle of the injector. The two graphs approximate each other up to x = 250 mm which rep-
resents a point of intersection.
On the other hand, fig. 7 shows that the
slopes of the mass fractions are important, which
reveals a rapid disappearance of hydrogen. In-
deed, hydrogen has become almost non-existent
beyond x = 600 mm. We also note that the graph
of VH2 = 90 m/s is relatively sharp compared to
the others, this is explained by the fact that the
flame hangs in a place closer to the injectors,
than in the case of VH2 = 236 m/s where we can
note a blowing of the flame which begins to take
place. Figure 7. Effect of hydrogen injection
speed on the H2 mass fraction
Effect of oxygen inlet temperature
In this case, the parameters relating to the H2 flow are maintained (VH2 = 236 m/s
and TH2 = 287 K). The oxygen inlet temperatures are changed as TO2 = 85 K, TO2 = 120 K, and
TO2 = 300 K. Figure 8 shows the effect of the oxygen inlet temperature does not seem to be very
large on the axial distribution of the static temperature. The visualization of the combustion
temperature fields, fig. 9, shows the shapes of the temperatures are the same. However, at the
output of the burner, the output temperature for TO2 = 85 K is higher than that of TO2= 300 K.

Figure 8. Static temperature distribution on Figure 9. Static temperature fields for

the central axis different O2 injection temperatures

Figure 10. Mass fraction of H2 for Figure 11. The H2 reaction rate for
different O2 inlet temperatures different O2 inlet temperatures
 Boukhelef, M., et al.: Numerical Study of the Injection Conditions Effect ...
3748 THERMAL SCIENCE: Year 2022, Vol. 26, No. 5A, pp. 3741-3750

In addition, fig. 10 shows that the slope of hydrogen consumption in the case of com-
bustion with oxygen is greater than the case with air, the oxidation reactions are therefore, fast-
er, although the oxygen injection temperature only slightly influences the consumption of H2.
This is justified by the variation in the reaction rate plotted as a function of the inlet temperature
of the oxygen, fig. 11.
Comparison between hydrogen flames
for the case of oxygen and the case of air
In comparison of the case of pure oxygen taken as oxidant (100% O2, 0% N2), fig. 12
shows that the thermal energy released during combustion is greater than for air case. Indeed,
the temperature variation on the axis shows the maximum temperature for oxygen (2850 K) is
greater than that for air (2130 K) as shown in fig. 13 with a relative rise by 33.80%.
On the other hand, the slope of the plot of oxygen is greater than that of air; which
means the reactions in the case of oxygen are faster. This is the reason why the position of the
peak of case A is upstream compared to the B case, fig. 13, also, this one requires more time to
reach its maximum value, although the maximum temperature for B case is very lower than in
A case.

Figure 12. Temperature field for hydrogen Figure 13. Temperature variation for hydrogen
combustion difference for air and oxygen case combustion for air and oxygen case

Conclusions
This paper concerns a numerical investigation for the effect of the injection conditions
on the structure of the diffusion flame, more specifically, speed and temperature was examined.
We are particularly interested in the flame blowing phenomena.
An important remark can be concluded: although nitrogen has an important mass
fraction in the air constitutions (78.99% N2) but hydrogen-air leads to generate non-significant
traces of NOx. On the other hand, it has been observed that increasing the hydrogen injection
rate causes the flame to blow away from the injection region, also, the areas with a high tem-
perature gradient are pushed towards the outlet of the combustion chamber away from the
injector, which promotes cooling of the nozzle region. The high fuel injection speeds decrease
the maximum temperature at the central zone of the tube while they promote an increase in tem-
peratures near the walls of the chamber. The maximum temperature decreases with increasing
fuel velocity.
Therefore, the high temperature inlet leads to flame clings to the injection nozzle.
Which can represent a dangerous situation by melting the nozzle if the residual time exceeds
a certain limit and gives rise to a possibility of flashback if the injection diameter exceeds the
Boukhelef, M., et al.: Numerical Study of the Injection Conditions Effect ...
THERMAL SCIENCE: Year 2022, Vol. 26, No. 5A, pp. 3741-3750 3749

quenching distance of the flame. Also, the flame length decreases when the air inlet temperature
and/or the air inlet velocity are increasing so the flame clinging to the injection nozzle. The
temperature increases continuously as it advances towards the end of the tube for a low inlet air
velocity. That means, for high air speeds, the temperature increases and the length of the flame
decreases.
On the other hand, the comparison between the combustion of hydrogen with oxygen
with that with air has shown the energy released by the combustion with oxygen is greater for
that of the combustion with air. Indeed, by varying the speed of air injection, it has been ob-
served that the flame front moves away from the outlet of the axial injector without leaving it.
However, the increasing in the hydrogen injection rate pushes the flame front away from the
injection nozzle. The areas of high temperature gradient are pushed out of the tube away from
the injector, which helps to cool the nozzle region. The high fuel injection speeds decrease the
maximum temperature at the central zone of the tube while promoting an increase in the fuel
speed. In addition, it noticed in the two cases previously discussed for combustion with oxygen,
the hydrogen appears to be completely consumed upstream of the station x = 600 mm.
Nomenclature
C1ε, C2ε, C3ε – constants in k-ε model V – overall velocity vector, [ms–1]
Di – diffusion coefficient of species i Vair, Vgas – air, gas velocity, [ms–1]
E, hi – enthalpy of species i, [J] x, y – radial, axial co-ordinates, [mm]
f – mixture fraction YM – fluctuating dilatation in compressible
Gk – generation of turbulence kinetic energy due turbulence to the overall dissipation rate
to mean velocity gradients
Gb – generation of turbulence kinetic energy due Greek symbols
to buoyancy ε – dissipation rate of turbulent kinetic
Ji – diffusive flux [molm–2s–1] energy, [m2s–3]
k – turbulence kinetic energy [m2s–2] μ – molecular viscosity
nj – stochiometric coefficient for product j μt – dynamic turbulence viscosity, [kgm–1s–1]
p – absolute pressure, [Pa] ρ – mass density
Reg – Reynolds number for gas σk, σε – turbulent Prandtl number for k and ε,
Sk, Sε – user-defined source term respectively
T – temperature, [K] ϑ – velocity scale
ui – density-weightedcartesian velocity ω~n, wi – density-weighted net rate of production of
component species n, i [molm3s–1]
References
[1] Phebe, A. O., Samuel A. S., A Review of Renewable Energy Sources, Sustainability Issues and Climate
Change Mitigation, Cogent Engineering, 3 (2016), 1, 1167990
[2] Koten, H., Hydrogen Effects on the Diesel Engine Performance and Emissions, International Journal of
Hydrogen Energy, 43 (2018), 22, pp. 10511-10519
[3] Jose Luis, A., Juan Carlos, B., The Energy Transitionwards Hydrogen Utilization for Green Life and
Sustainable Human Development in Patagonia, International Journal of Hydrogen Energy, 45 (2020),
47, pp. 25627-25645
[4] Bakic, V., et al., Technical Analysis of Photovoltaic/Wind Systems with Hydrogen Storage, Thermal Sci-
ence, 16 (2012), 3, pp. 865-875
[5] Riahi, Z., et al., Numerical Study of Turbulent Normal Diffusion Flame CH4-air Stabilized by Coaxial
Burner, Thermal Science, 17 (2013), 4, pp. 1207-1219
[6] Alrbai, M., et al., Influence of Hydrogen as a Fuel Additive on Combustion and Emissions Characteristics
of a Free Piston Engine, Thermal Science, 24 (2019), 1, pp. 71-71
[7] Gupta Ram B., Hydrogen Fuel: Production, Transport, and Storage, CRC Press Taylor & Francis Group,
Boca Raton, Fla., USA, 2009
[8] Eichlseder H., et al., The Potential of Hydrogen Internal Combustion Engines in a Future Mobility Sce-
nario, SAE Technical paper, 2003-01-2267, 2003
 Boukhelef, M., et al.: Numerical Study of the Injection Conditions Effect ...
3750 THERMAL SCIENCE: Year 2022, Vol. 26, No. 5A, pp. 3741-3750

[9] Lemmon, E. W., Thermophysical Properties of fluids, CRC Handbook of Chemistry and Physics, 90th ed.
CRC Press, Boca Raton, Fla., USA, Section 6: Fluid Properties, 6.21-6.31, 2009
[10] Law, C. K., Combustion Physics, Cambridge University Press, New York, USA, 2006
[11] Alliche, M., et al., Extinction Conditions of a Premixed Flame in a Channel, Combustion and Flame, 157
(2010), 6, pp.1060-1070
[12] Syred N., et al., Effect of Inlet and Outlet Configurations on Blow-off and Flashback with Premixed
Combustion for Methane and a High Hydrogen Content Fuel in a Generic Swirl Burner, Applied Energy,
116 (2014), 1, pp. 288-296
[13] Karagoz, Y., Koten, H., Effect of Different Levels of Hydrogen + LPG Addition on Emissions and Per-
formance of a Compression Ignition Engine, Journal of Thermal Engineering, 5 (2019), 2, Special Issue
9, pp. 58-69
[14] Karagoz Y., Effect of Hydrogen Addition at Different Levels on Emissions and Performance of a Diesel
Engine, Journal of Thermal Engineering, 4 (2018), 2, Special Issue 7, pp. 1780-1790
[15] Alliche, M., et al., Effect of Bluff-Body Shape on Stability of Hydrogen-Air Flame in Narrow Channel,
in: Progress in Renewable Hydrogen and other Sustainable Energy Carriers, (ed. K. Abdallah), Springer
Proceedings in Energy, New York, USA, Chapter 30, 2021, pp. 231-238
[16] Umyshev, D., et al., Experimental Investigation of Distance between v-Gutters on Flame Stabilization and
NOx Emissions, Thermal Science, 23 (2019), 5-B, pp. 2971-2981
[17] Yang, W., Blasiak, W., Numerical Study of Fuel Temperature Influence on Single Gas Jet Combustion in
Highly Preheated and Oxygen Deficient Air, Energy, 30 (2005), 2, pp. 385-398
[18] Alliche, M., Chikh, S., Study of Non-Premixed Turbulent Flame of Hydrogen/Air Downstream Co-Cur-
rent Injector, International Journal of Hydrogen Energy, 43 (2018), 6, pp. 3577-3585
[19] Wang, C. J., et al., Predicting Radiative Characteristics of Hydrogen and Hydrogen/Methane Jet Fires
Using FireFOAM, International Journal of Hydrogen Energy, 39 (2014), 35, pp. 20560-20569
[20] Dakka, S., Numerical Analysis of Flame Characteristics and Stability for Conical Nozzle Burner, Journal
of Thermal Engineering, 5 (2019), 5, pp. 422-445
[21] Shi, L., et al., A Model of Steam Reforming of Iso-Octane: The Effect of Thermal Boundary Conditions
on Hydrogen Production and Reactor Temperature, International Journal of Hydrogen Energy, 33 (2008),
17, pp. 4577-4585
[22] Hua, Jingsong, et al., Numerical Simulation of the Combustion of Hydrogen Air Mixture in Micro-Scaled
Chambers – Part II: CFD Analysis for a Micro-Combustor, Chemical Engineering Science, 60 (2005), 13,
pp. 3507-3515
[23] ***, FLUENT Ansys, Theory Guide. Ansys Inc., 2009
[24] Khaladi, F. Z., et al., Numerical Simulation of CH4-H2-Air Non-Premixed Flame Stabilized by a Bluff
Body, Energy Procedia, 139 (2017), Dec., pp. 530-536
[25] Pope, S. B., The PDF Methods for Turbulent Reactive Flows, Progress in Energy and Combustion Sci-
ence, 11 (1985), 2, pp. 119-192
[26] Obieglo, A., et al., Comparative Study of Modelling a Hydrogen Non-premixed Turbulent Flame, Com-
bustion and Flame, 122 (2000), 1-2, pp. 176-194
[27] Nanduri, J. R., et al., Assessment of RANS-Based Turbulent Combustion Models for Prediction of Emis-
sions from Lean Premixed Combustion of Methane, Combustion Science and Technology, 182 (2010), 7,
pp. 794-821
[28] Bilger, R. W., Turbulent Flows with Non-Premixed Reactants, in: Turbulent Reacting Flows, Topics in
Applied Physics, (Eds.. Libby P. A., Williams F. A.), Springer, Berlin, Heidelberg, 44 (1980), pp. 65-113
[29] Yakhot, V., Orszag, S. A., Renormalization Group Analysis of Turbulence, I. Basic Theory, Journal Sci.
Comput., 1 (1986), Mar., pp. 3-51
[30] Raman, V., et al., Eulerian Transported Probability Density Function sub-Filter Model for Large-Eddy
Simulations of Turbulent Combustion, Combustion Theory and Modelling, 10 (2006), 3, pp. 439-458
[31] ***, https://www.cerfacs.fr/cantera/mechanisms/hydro.php#boiv
[32] Turns, S. R., An Introduction Combustion: Concepts and Applications, 3rd ed., Mc Graw Hill, New York,
USA, 2012

Paper submitted: May 30, 2021 © 2022 Society of Thermal Engineers of Serbia
Paper revised: August 21, 2021 Published by the Vinča Institute of Nuclear Sciences, Belgrade, Serbia.
Paper accepted: September 21, 2021 This is an open access article distributed under the CC BY-NC-ND 4.0 terms and conditions