Case Studies in Thermal Engineering 52 (2023) 103669

Contents lists available at ScienceDirect

Case Studies in Thermal Engineering

journal homepage: www.elsevier.com/locate/csite

Study on combustion characteristics of fully premixed

water-cooled biogas burner
Jian Jiao 1, Wang Yungang *, 1, Liu Yufei, Zhang Xingbang, Zhao Qinxin
Key Laboratory of Thermo-Fluid Science and Engineering, Ministry of Education (Xi’an Jiao Tong University), Xi’an, 710049, China

A R T I C L E I N F O A B S T R A C T

Keywords: Biogas is a renewable gas produced by the fermentation of biomass and is an environmentally
Biogas friendly alternative to traditional fossil fuels. To overcome the problems of traditional biogas
Fully premixed combustion burner, such as unstable combustion, easy flameout and backfire, and low thermal intensity, a
Water cooled burner new water-cooled biogas burner based on fully premixed combustion technology was designed
Combustion characteristics and developed. Then, numerical simulations and experimental investigations were conducted to
explore the effects of CO2 content, excess air coefficient, heat load, and cooling water temperature
on the combustion characteristics of the burner. The results revealed that the designed burner
exhibited good adaptability to simulated biogas with methane concentrations ranging from 55 %
to 80 %. Increasing CO2 content resulted in decreased combustion temperatures and reduced NOx
and CO emissions concentration at the burner outlet. In addition, reducing the excess air coef­
ficient enhanced the combustion heat intensity and reduced combustion heat loss. Nevertheless,
excessively low excess air coefficients yielded elevated levels of NOx and CO in the flue gas. The
results showed that the designed burner demonstrated stable operation at a minimum heat load of
35 %, exhibited a favorable load regulation ratio, and had low emissions. It is a promising so­
lution for utilizing biogas as a clean and sustainable energy source.

1. Introduction
Biogas, as a clean and environmental-friendly energy source, can be used similarly to natural gas (NG) for electricity generation,
heating, or as a transportation fuel [1–3]. It holds immense potential due to its diverse sources and minimal carbon emissions during
generation and utilization [4,5]. Due to the different sources and processes, biogas is mainly composed of methane (55–70 % CH4),
carbon dioxide (30–45 % CO2),some traces of nitrogen (0–3% N2), hydrogen (0–1% H2), hydrogen sulfide (0–1% H2S) and volatile
organic compounds, and the composition widely depends on the treated materials and treatment process [6,7]. Due to the dilution of
CO2 to CH4, the calorific value of biogas is not high as that of natural gas, and the stability of the flame is reduced [8]. Therefore, the
current traditional gas burners, if directly used for biogas combustion, will correspondingly encounter issues such as unstable com­
bustion, easy flameout and backfire, and low thermal intensity due to the unique characteristics of biogas. Scholars have increasingly
focused on the stability of biogas composition in burners, seeking solutions to enhance the efficiency and reliability of biogas com­
bustion. For example, Zeng et al. [9] indicated that increasing CO2 content in premixed biogas/air mixtures increases the mixture
ignition delay time and reduces the flame speed. Sahin et al. [10] believed that with the decrease of oxygen concentration, the thermal

* Corresponding author.
E-mail address: [email protected] (W. Yungang).
1
These authors contributed to the work equally and should be regarded as co-first authors.

https://doi.org/10.1016/j.csite.2023.103669
Received 2 August 2023; Received in revised form 13 October 2023; Accepted 23 October 2023
Available online 24 October 2023
2214-157X/© 2023 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license
(http://creativecommons.org/licenses/by/4.0/).
J. Jiao et al. Case Studies in Thermal Engineering 52 (2023) 103669

field in the combustion chamber becomes more uniform. Anggono et al. [11] studied the combustion situation of biogas composed of
66.4 % methane, 30.6 % carbon dioxide and 3 % nitrogen, obtained the laminar combustion velocity of biogas-air mixture, and
explained the basic flame propagation characteristics of biogas. Greco [12] studied the effect of fuel composition on biogas combustion
in laminar one-dimensional premixed flame through numerical simulation, and indicated that the CH4 concentration has a significant
impact on the combustion process. However, when the CH4 volume concentration was fixed at 50 %, there was no major difference
observed in the fuel mixtures, even with changes in CO2 and other gas components.
In order to overcome the combustion problems caused by low calorific value methane, many efforts have been made to make biogas
burn better. According to the primary air coefficient, biogas burners can be divided into diffusion burners, partially premixed burners
and fully premixed burners. For diffusion burners, biogas like other low calorific value fuels has a very narrow stable region owing to
their low burning velocity in conjunction with the unburned flow high velocity. Thus, Saediamiri et al. [13] tried to improve the flame
stability limit by changing the burner geometry, and found that the swirl effect occurs only at relatively moderate to high co-flow
velocity, while the vortex has no obvious effect on the weak co-flow. Ilbas et al. [14] simulated that the changes in turbulator an­
gles highly affect the temperature and emission profiles of the biogas throughout the combustion chamber. However, although they
tried to achieve stable combustion by changing the burner structure, the diffusion combustion itself has a narrow stable region, so it is
not suitable for the use of biogas, which may have fluctuating flow. Besides, Zhen et al. [15] indicated that comparing to the diffusion
flames, the premixed flame yields have higher heat transfer, due to higher flame temperature and larger volume of hot gas in the
premixed flame. Thus, premixed combustion is more suitable for the utilization of biogas. Abdallah et al. [16] designed a slit partially
premixed burner, which can sustain a stable biogas flame up to 30%CO2. However, increasing the CO2 to 40 % eventually led to a weak
and unstable flame regardless of the premixing ratio. Effuggi et al. [17]designed laboratory-scale burners for fully premixed com­
bustion of methane and biogas and found that fully premixed combustion not only significantly reduce NOx emissions, but also prevent
carbon deposition under fuel-rich conditions. Davi et al. [18] developed a 5–10 kW fully premixed biogas burner, which not only has
high combustion efficiency, but also reduces the CO and NOx values by 95 % and 85 %, respectively. Besides, some other scholars have
also studied the combustion characteristics and burner structure of biogas under different combustion modes to different degrees
[19–24]. However, these burners are generally small in power, mostly laboratory experiments or small burners (less than 15 kW), and
have limited capacity for biogas utilization, so that more powerful biogas burners need to be developed.
Based on the above background, it can be concluded that fully premixed combustion is compatible with biogas. However, due to the
instability of biogas composition and source, biogas burners should consider both the reduction of calorific value caused by CO2
dilution and the fluctuation of heat load during operation. With the continuous strict environmental protection policies and increasing
annual output of biogas, biogas burners need to be able to stably burn and control pollutant emissions under a low load state and a wide
load range. High power and industrialization have become the inevitable demand for biogas utilization. In order to further expand the
application range of current biogas burners, it is necessary to design and explore the possibility of high-power fully premixed biogas
combustion and its influencing factors. Based on the design concept of the fully premixed burner, the stable combustion is achieved by
controlling the flow rate to prevent flame-out and tempering, even at a low heat load. In addition, in order to control NOx in com­
bustion products, water-cooled combustion technology is used on the basis of fully premixed combustion. This technology reduces the
temperature of the combustion zone by cooling the root of the flame, thereby controlling the production of thermal NOx, and realizing
efficient and clean combustion of biogas. Based on the above design concept, a new type of fully premixed water-cooled biogas burner
with a rated power of 350 kW was designed in this paper, and comprehensive experimental and numerical investigations were con­
ducted to understand its combustion characteristics and relevant influencing factors.

2. Experiment setup and numerical methods

2.1. Design of the biogas burner
Fully premixed combustion has several advantages, including high combustion heat intensity, rapid flame propagation, and a
compact combustion space requirement. These characteristics effectively address the limitations of biogas, such as its low calorific
value and slow burning speed, making fully premixed combustion the ideal burner form for biogas utilization. Furthermore, the water-
cooled combustion technology effectively reduces thermal NOx, because it can lower the flame root temperature by cooling water.
Thus, this biogas burner is a new design combining fully premixed combustion and water-cooled premixing.
The key to designing a fully premixed water-cooled burner is ensuring that no backfire occurs in the burner. Therefore, the pa­
rameters should be determined based on burner heat load, ultimate propagation speed, and size of the water-cooled tubes. This ensures

Table 1
Biogas burner related design parameters.

No. Parameters Notation Value Unit

1 Center distance between finned tube and burner head D 160 mm

2 Finned tube length L 699 mm
3 Fin height h 1.5 mm
4 Maximum burning area S0 212942.0 mm2
5 Fin gap flow velocity v 2.8 m⋅s− 1
6 Total number of finned tubes N 28 /
7 Ultimate backfire speed v0 0.12 m⋅s− 1
8 Required burning area S1 215627.2 mm2

2
J. Jiao et al. Case Studies in Thermal Engineering 52 (2023) 103669

that combustion occurs without backfire while meeting the load requirements. The finalized parameters of the burner are provided in
Table 1, and the final structure is shown in Fig. 1.
During burner operation, the premixed gas enters the top of the burner head, passes through a double orifice plate in the radial
direction, and then burns through the gaps in the water-cooled tube bundle. The top of the water-cooled tube bundle is arranged with
an upper annular collector. The upper part of the water-cooled tube bundle is equipped with an upper annular collector A partition in
its middle position divides the upper annular collector into two parts. Besides, the water-cooled tube bundle takes the form of a ring
casing. The cooling water first enters into the inner tube of the water-cooled casing pipes from the upper part of the upper annular
collector to flow downward. Subsequently, the water flows upward along the outer tubes under pressure after reaching the lower part
of the upper annular collector. Finally, the cooling water reaches the lower annular collector. Heat exchange is achieved between the
water-cooled tube bundle and the high-temperature zone in the combustion chamber, so the combustion chamber temperature is
reduced, resulting in a low thermodynamic NOx generation.

2.2. Experimental system

Fig. 2 shows the experimental system utilized in this study. First of all, the fan mixes the outside air and the ejecting natural gas with
CO2 in the premix to generate fully premixed simulated biogas gas. Subsequently, the mixed gas enters the biogas burner, passes
through the gap between water-cooled tube bundles, and then is ignited. Finally, the flue gas after combustion is cooled by heat
exchangers and the waste heat is recovered and discharged.
For the measure and control system, the controller takes charge of regulating the fan’s speed to adjust the burner load and manages
the ignition action and the magnetic valve. To control the CO2 flow into the premix, a precision trimmer valve is utilized. Additionally,
the flow of CH4 and CO2 through the premix is measured using a turbine flowmeter and a vortex flowmeter, respectively. To monitor
the combustion process, a K-type thermocouple is strategically positioned at a furnace depth of 460 mm to measure the flame tem­
perature accurately; The flue gas analyzer is used to measure the temperature of flue gas and the content of O2, CO2, CO and NOx in flue
gas. The measuring equipment and corresponding accuracy ranges are shown in Table 2. Besides, the specific experimental conditions
are shown in Table 3.
The composition of biogas is simulated by the ratio between CH4 and CO2, and the heating value of mixed gas fuel is calculated
based by Eq. (1).
∑
n
H1 = Hi · di (1)
i=1

Where Hi is the low heating value of each component, and di is the volume fraction of each component.
The actual amount of air required per cubic meter of gas combustion under standard condition is as follows:
Vt α
Va = (2)
ε

Fig. 1. Schematic diagram of fully premixed water-cooled biogas burner.

3
J. Jiao et al. Case Studies in Thermal Engineering 52 (2023) 103669

Fig. 2. Diagram of fully premixed water-cooled biogas combustion experimental system.

Table 2
Measuring equipment.

Program Device Type Measuring range Accuracy

Flue gas temperature Digital thermometer DTM-180A (-50) − 200 ◦ C ±0.1 ◦ C

Cooling water temperature Digital thermometer DTM-180A (-50) − 200 ◦ C ±0.1 ◦ C
Cooling water flow Electromagnetic flowmeter CKLDG-D15 0.06–6.5 m3⋅h− 1 ±0.5 %
flame temperature K-type thermocouple SMPW-GG-K − 200-1350 ◦ C ±0.004|t|
CO2 volume flow Vortex flowmeter LUGB-DN20 6–30 m3⋅h− 1 ±1.5 %
Nature gas flow Roots flowmeter ZKTD-JLQ-60 0.07–60 m3⋅h− 1 1.0 %~1.5 %
O2 volume fraction Flue gas analyzer Testo 330 0-21 vol% ±0.2 vol%
CO volume fraction Flue gas analyzer Testo 330 0–4000 ppm ±20 ppm
NO volume fraction Flue gas analyzer Testo 330 0–3000 ppm ±5 ppm

Table 3
Experimental conditions and cases during the test.

Case No. φ(CH4)/φ(CO2) α Heat Load m(CH4) m(CO2) m(air) Tw

3 − 1 3 − 1 3 − 1
kW m •h m •h m •h K

1 55 %/45 % 1.3 320 31.4 25.7 388.6 330

2 55 %/45 % 1.25 320 31.4 25.7 373.7 330
3 55 %/45 % 1.2 320 31.4 25.7 358.7 330
4 55 %/45 % 1.15 320 31.4 25.7 343.8 330
5 60 %/40 % 1.3 350 34.3 22.8 424.5 330
6 60 %/40 % 1.25 350 34.3 22.8 408.17 330
7 60 %/40 % 1.2 350 34.3 22.8 391.8 330
8 60 %/40 % 1.15 350 34.3 22.8 375.5 330
9 65 %/35 % 1.3 380 37.1 20 459.2 330
10 65 %/35 % 1.25 380 37.1 20 441.5 330
11 65 %/35 % 1.2 380 37.1 20 423.8 330
12 65 %/35 % 1.15 380 37.1 20 406.2 330
13 70 %/30 % 1.3 410 39.9 17.1 493.8 330
14 70 %/30 % 1.25 410 39.9 17.1 474.8 330
15 70 %/30 % 1.2 410 39.9 17.1 455.8 330
16 70 %/30 % 1.15 410 39.9 17.1 436.8 330
17 75 %/25 % 1.3 440 42.8 14.3 529.7 330
18 75 %/25 % 1.25 440 42.8 14.3 509.3 330
19 75 %/25 % 1.2 440 42.8 14.3 489 330
20 75 %/25 % 1.15 440 42.8 14.3 468.6 330
21 70 %/30 % 1.3 205 20 8.6 274.5 330
22 70 %/30 % 1.3 145 14 6 173.3 330
23 70 %/30 % 1.3 410 39.9 17.1 493.8 300
24 70 %/30 % 1.3 410 39.9 17.1 493.8 315

Where α is excess air coefficient and ε is dry air volume ratio. Vt is theoretical air volume which can be calculated by Eq. (3).
1 [ ∑( n) ]
Vt = 0.5PH2 + 0.5PCO + m + PCm Hn + 1.5PH2 S − PO2 (3)
21 4
Since the composition of biogas is simulated through the distribution of CO2 and CH4 in the experiment, the theoretical air volume
can be simplified by Eq. (4).
2PCH4
Vt = (4)
21

4
J. Jiao et al. Case Studies in Thermal Engineering 52 (2023) 103669

To compare the volume fractions of NOx and CO at the burner outlet under different excess air coefficients, it is essential to convert
them to the same oxygen content. Thus, the specific calculation formula is shown as:
( ) [NO](ppm) 46 21 − [O2 ]
[NOx ] mg · m− 3 = × × [ ] (5)
0.95 22.4 21 − O′2

( ) 28 21 − [O2 ]
[CO] mg · m− 3 = [CO](ppm) × × [ ] (6)
22.4 21 − O′2

Where [O2] is the converted oxygen content of the flue gas, which is uniformly 3.5 %, and [O′2] is the actual oxygen content of the flue
gas.

2.3. Numerical methods

In order to further determine the overall situation of methane combustion in the combustion head, we carried out numerical
calculation of the water-cooled burner involved in the experiment based on ANSYS Fluent 20.0. Firstly, the designed burner, the water-
cooled tubes and the furnace combustion space were modeled in a 1:1 scale. Considering that the designed burner is composed of 28
helical finned tubes, and the flow and combustion results should present a law of central symmetry, a 1/28 portion of the original
model was truncated to create the physical model and calculation grid. Besides, since the numerical calculation does not involve the
fluid-structure coupling, only the fluid domain is considered, and the solid regions such as uniform flow orifice, water-cooled tubes and
other steel walls are removed. Thus, a computational model which can effectively characterize the flow and combustion of premixed
gas between two finned tubes is obtained, as illustrated in Fig. 3.
As shown in Fig. 3, the premixed gas entered from the inlet of the burner head, flowed axially up to the upper wall of the annular
collector below, and then flowed along the radial direction through the uniform flow orifice plate to reach the rear side of the water-
cooled fin bundle. The combustion reaction was carried out through the gap of the water-cooled fin bundle. The outermost side of the
radial direction of the model was the water-cooled wall of the water-cooled jacket. Then, the flue gas flowed out axially through the
exit of the burner after the heat exchange with the jacket. Because the simulation does not consider the temperature change of cooling
water during combustion. As a result, both the water-cooled tube bundle and the annular collector were modeled as cavities with
single-layer walls without thickness.
To accurately simulate the complex combustion process, the turbulence model employed was the realizable k-ε model, which
provides reliable predictions of turbulent flow characteristics. Furthermore, to account for the detailed chemical reaction mechanism
during combustion, the Eddy Dissipation Concept (EDC) model was utilized. The numerical simulation of NOx is obtained by solving
the transport equation of NO.
( )
∂ρmNO ∂ ( ) ∂ ∂mNO
+ ρuj mNO = Γs + RNO (7)
∂t ∂Xj ∂ Xj ∂Xj

Where RNO is the source term and represents the sum of thermal type and fast type NOx.
In terms of the selection of methane combustion reaction mechanism, the widely used methane multi-step reaction model is the
GRI-Mech 3.0 mechanism, which contains 325 elementary reactions and 53 species [25,26]. This mechanism has a wide range of

Fig. 3. Computational model and grids.

5
J. Jiao et al. Case Studies in Thermal Engineering 52 (2023) 103669

applicable conditions, high accuracy and good reaction kinetics. However, due to the complex structures such as uniform flow orifice
and finned tubes in the numerical calculation model, the direct use of the GRI-Mech 3.0 mechanism will lead to excessive simulation
calculation. It is necessary to simplify the GRI-Mech 3.0 mechanism by direct relational graph method (DRG) and principal component
analysis method (PCA) to retain the reaction that has the most significant impact on combustion, so that it has a high consistency with
the detailed mechanism while reducing the calculation amount [27].
Therefore, the 2-step reaction mechanism of Fluent, the 10-step reaction and 43-step reaction mechanism simplified by GRImech
3.0 were selected to simulate a simplified 350 kW burner model. The calculated results were compared with the 325-step reaction
based on GRImech 3.0, as shown in Table 4. It can be seen that 43 elementary reactions are in good agreement with the detailed
mechanism.
The boundary conditions employed in the numerical simulations were as follows: at the burner inlet, a velocity inlet condition was
specified to represent the incoming premixed gas. The burner outlet was set as an outflow outlet, accounting for the exiting flue gas.
The side surfaces of the model were designated as periodic boundary conditions to account for cyclic behavior. Specifically, the fin
bundle and water-cooled wall surfaces were set as the first boundary conditions, while the remaining surfaces were defined as
adiabatic boundary conditions, assuming no heat transfer at those locations.
To ensure accurate and reliable results, grid independence verification was conducted. Five different meshes with mesh numbers of
0.7 million, 1.52 million, 2.47 million, 3.61 million, and 6.08 million were simulated for the physical model, as shown in Fig. 4(a). It is
decided to use 3.61million grid for further calculation after comparison. Fig. 4(b) illustrates that with the grid of 3.61 million, the
difference between numerical and experimental value of CO and NOx outlet concentration is less than 0.8 mg m− 3 and 3.2 mg m− 3
under different effect of excess air coefficient, respectively.

3. Results and discussions

3.1. Effect of excess air coefficient on combustion characteristics
Excess air coefficient (α) plays an important role in burner performance, affecting exhaust gas heat loss, thermal efficiency of gas-
fired boilers, and NOx emissions. Experimental research was conducted on simulated biogas with excess air coefficient from 1.15 to
1.30, and the results are shown in Fig. 5 and Fig. 6. Fig. 5 indicates that as the excess air coefficient increases, excess air absorbs the
heat generated during the combustion reaction, leading to a gradual decrease in combustion temperature. The variation range of the
excess air coefficient on combustion temperature is consistent across simulated biogas with CO2 volume fractions ranging from 25 % to
45 %. The difference between the maximum and minimum combustion temperatures for these compositions is 40.3K, 42.1K, 39.2K,
35.5K, and 39.0K, respectively.
The CO content at the burner outlet is an important parameter to characterize the burner burnout rate. Fig. 6(a) shows that the CO
content at the burner outlet is generally low across all experimental conditions, with the maximum value being less than 10 mg m− 3. In
addition, as the excess air coefficient increases from 1.15 to 1.3, the CO content steadily decreases. This indicates that the excess
oxygen promotes the CO oxidation reaction in the flue gas, leading to increased burnout of simulated biogas.
The NOx content at the outlet is mainly affected by the maximum temperature of the burner and the size of the high-temperature
area, thus exhibiting a consistent trend with the combustion temperature. With varying CO2 volume fractions, the nitrogen oxides at
the outlet decrease by 38.1 %, 36.6 %, 31.7 %, 26.4 % and 25.9 % respectively, as the increase of excess air coefficient. The results
show that the increase of excess air coefficient can effectively control NOx emissions. When α is 1.2, the simulated biogas combustion
with 25 % CO2 volume fraction produces 34.1 mg•m− 3 NOx, while when α is 1.15, the simulated biogas combustion with 25 % and 30
% CO2 volume fraction produces 37.5 mg•m− 3 and 32.5 mg•m− 3 NOx, respectively. Therefore, in order to achieve the NOx outlet
content less than 30 mg•m− 3 emission target while improving the efficiency of the gas boiler, an excess air coefficient greater than 1.2
is recommended.

3.2. Effect of CO2 volume fraction on combustion characteristics

The combustion characteristics of the burner are significantly influenced by the varying CO2 volume fraction in actual biogas,
owing to the dynamic nature of the fermentation environment and other factors.
Fig. 7 illustrates the impact of CO2 volume fraction on combustion characteristics. It is evident that, under the same excess air
coefficient, both combustion temperature and outlet NOx content decrease with CO2 volume fraction. Their similar trends suggest that
the NOx generated during biogas combustion is primarily thermal NOx. Conversely, the outlet CO content increased with CO2 volume
fraction, reaching its maximum value at a CO2 volume fraction of 45 %. This increase is attributed to the higher conversion of CO from
the total gas mixture. However, at higher CO2 volume fractions, the combustion temperature decreases, leading to an inhibitory effect
on the endothermic reaction that converts CO2 into CO. As a result, the maximum outlet CO content remains relatively low, measuring

Table 4
Simulation results of different reaction mechanisms.
3 3
Mechanism reaction Outlet CO concentration/mg•m− Outlet NOx concentration/mg•m−

2-step reaction 1140.4 35.9

10-step reaction 264.1 25.5
43-step reaction 57.7 41.5
325-step reaction 52.2 44.7

6
J. Jiao et al. Case Studies in Thermal Engineering 52 (2023) 103669

Fig. 4. Grid independence verification.

Fig. 5. Influence of excess air coefficient on maximum temperature of burner.

Fig. 6. Influence of excess air coefficient on emission.

only 9.0 mg m− 3.
Besides, the emissions for the new biogas burner were also compared to the results by Valera et al. [28] and Liu et al. [29]. Since the
above reference results all use 15 % oxygen content as the conversion standard, the experimental data are converted and presented
together in Fig. 8. It can be seen from Fig. 8 (a) that NOx emissions from all burners decrease with the increase of CO2. This is because
an increase in CO2 will correspondingly reduce the content of CH4 in effective combustion, which will also reduce the content of N2 and
O2 in the flame zone, resulting in a reduction of NOx. The NOx emission of the new designed burner is lower. Besides, the designed
burner also has a low CO emission (less than 3 ppm), as shown in Fig. 8(b). It indicates the new biogas burner has a good burnout rate,

7
J. Jiao et al. Case Studies in Thermal Engineering 52 (2023) 103669

Fig. 7. Influence of CO2 volume fraction on combustion characteristics of burner.

Fig. 8. NOx and CO outlet concentration for different CO2 fractions.

enabling the fuel and O2 to be fully mixed and burned.

Fig. 9 shows the variation of NOx generation rate during simulated biogas combustion at different CO2 volume fractions. It can be
seen that when the premixed gas swiftly exits the slit between the finned tubes at a high speed, the NOx generation rate in the middle
region is low, indicating a lack of combustion reaction in this zone. Conversely, the NOx generation rate is higher on the left and right
sides of the slit outlet. This is attributed to the location not only at the suction side of the finned tube but also due to the reflux vortex of
high-speed premixed gas in this region. This phenomenon extends the residence time of flue gas, ensuring sufficient combustion re­
actions and resulting in an increased regional temperature. As the CO2 volume fraction increases from 20 % to 45 %, the maximum rate
and range of NOx generation are decreased. The maximum formation rate of NOx decreased from 4.28×10− 7 to 4.95×10− 8 kgmol m− 3
s− 1, indicating that the formation rate of NOx during combustion is closely related to the maximum combustion temperature.

3.3. Effect of combustion heat load on combustion characteristics

For practical burner operation, the heat load varies according to specific requirements, necessitating a burner with a large load
adjustment ratio. The ideal burner should be capable of meeting combustion intensity and pollutant emission requirements under high
load conditions, while also ensuring stable operation without backfiring at low load conditions.
Fig. 10 shows the combustion conditions of simulated biogas at 35 %, 50 %, and 100 % of the heat load, when the excess air
coefficient is 1.3, CO2 volume fraction is 30 %. It shows that the burner can at least maintain stable combustion at 35 % of the full load.
As the combustion heat load increases from 35 % to 100 %, the combustion temperature rises from 948.3K to 1070.1K. The NOx
content at the burner outlet increases from 8.2 mg•m− 3 to 26.1 mg•m− 3, while the CO content decreases from 6.4 mg•m− 3 to zero. The
observed trends can be attributed to the reduction in simulated biogas entering the burner at lower load conditions. Consequently, the

8
J. Jiao et al. Case Studies in Thermal Engineering 52 (2023) 103669

Fig. 9. Effect of CO2 volume fraction on NOx formation rate.

Fig. 10. Effect of different heat load on combustion parameters.

combustion heat intensity weakens, leading to lower maximum combustion temperatures and reduced NOx generation. Conversely, at
higher load conditions, more simulated biogas is supplied to the burner, resulting in enhanced combustion heat intensity, higher
maximum combustion temperatures, and increased NOx generation.
The numerical calculations were further conducted to explore the combustion conditions under different loads, and the results are
presented in Fig. 11. It shows that the burner can operate stably under both high load and low load conditions, with a stable flame and
no backfire observed during ignition at low load. These results underscore the reasonability and reliability of the burner’s design,
ensuring safe and consistent operation. In addition, the temperature contour plot reveals that the heat load significantly influences the
combustion temperature and flame shape. At lower loads, the flame temperature decreases, leading to a smaller high-temperature area
in the combustion chamber. This effect is due to the reduced gas and air flow, resulting in lower combustion intensity and slower flow
through the finned tube gap.
Since the combustion intensity represents the fuel completely burned off per unit volume per unit time, the large combustion
intensity indicates that the combustion speed is fast, or the flame propagation speed is large, and the flame itself is short with a more
concentrated high temperature area. In order to evaluate the combustion intensity, the OH mass fraction is used, as the criterion to
judge the flame front and combustion. Considering the results of literature [30], the OH radical is described as a “weakening of the
flame front” and can show a reduction in the reaction rate. The higher the OH mass fraction, the faster the flame front weakens, that is,
the faster the burning speed and the higher the combustion intensity.

9
J. Jiao et al. Case Studies in Thermal Engineering 52 (2023) 103669

Fig. 11. Temperature contours of burner under different heat loads.

Fig. 12 shows the OH mass fraction contour. With the increase of heat load, the maximum mass fraction of OH increases from
8.35×10− 4 at 35 % load to 1.43×10− 3 at 100 % load. According to the sensitivity analysis of CO [24], the main reaction in the
production of CO species is shown in Eq. (8).
OH + CO⇌H + CO2 (8)

When the combustion heat load increases, the high temperature range expends, enhancing the reaction of active free radicals like OH
and H. Consequently, the branch chain reaction of these free radicals combining with CO to produce CO2 is enhanced. Therefore, the
CO content increases with decreasing load, consistent with the observations in Fig. 10.

3.4. Effect of cooling water temperature on combustion characteristics

Compared with other fully premixed burners, the most distinctive feature of the fully premixed water-cooled burner is that the
combustion head is constantly cooled by cooling water. Fig. 13 shows the effect of the cooling water temperature on the combustion

Fig. 12. OH contours of burner under different heat loads.

10
J. Jiao et al. Case Studies in Thermal Engineering 52 (2023) 103669

characteristics. It can be seen that the higher the cooling water temperature, the higher the combustion temperature. When the cooling
water temperature rises from 300 K to 330 K, the maximum temperature in the combustion chamber increases by 15.7 K, and the outlet
NOx content also increases from 14.3 mg•m− 3 to 26.1 mg•m− 3 accordingly. This positive correlation between combustion heat in­
tensity and nitrogen oxides reaffirms the relationship observed in previous analyses.
Fig. 14 illustrates the variation of OH mass fraction with cooling water temperature. The contour image shows that when the
cooling water temperature increases from 300K to 330K, the maximum OH mass fraction gradually increases. This indicates that a
higher OH mass fraction leads to more rapid combustion reactions, allowing for quicker consumption of biogas and achieving more
complete combustion. As a result, complete combustion corresponds to higher combustion temperatures, and this is reflected in the
increasing NOx generation rate depicted in Fig. 15. These results demonstrate that reducing the cooling water temperature is a feasible
and effective measure to meet nitrogen oxides emission standards when using a fully premixed water-cooled burner. However, the
change trend of CO content at the outlet indicates that decreasing the cooling water temperature leads to a reduction in combustion
heat intensity and a loss of burner combustion efficiency. Therefore, there is a trade-off between NOx emissions and combustion ef­
ficiency in the context of cooling water temperature. The burner’s cooling water temperature must be carefully adjusted to strike the
right balance between reduced NOx emissions and optimal combustion performance.

4. Conclusions
In this paper, a new type of fully premixed water-cooled biogas burner was designed independently, then experimental and nu­
merical research was conducted on the combustion performance of the simulated biogas burner. The main conclusions are as follows:
(1) The biogas burner designed in this paper can achieve stable combustion in the range of excess air coefficient from 1.15 to 1.3.
With the increase of excess air coefficient, the combustion temperature, outlet NOx content and CO content all show a
downward trend according to the experiment. Increasing the excess air coefficient can control the combustion temperature,
reduce the emissions of nitrogen oxides at the outlet, and improve the burner burnout rate.
(2) The designed burner exhibits strong adaptability to gas composition, and the simulated biogas has a good combustion effect
when the volume fraction of CO2 changes from 45 % to 25 %. A decrease in CO2 volume fraction leads to an increase in
maximum combustion temperature and outlet NOx content, while the CO content decreases. This suggests that reducing the CO2
volume fraction enhances the flammability of simulated biogas, increasing combustion heat intensity. Moreover, the higher
combustion temperature facilitates the further oxidation of CO. Conversely, higher CO2 content in simulated biogas hampers
burner burnout rate.
(3) The burner has good adaptability to heat load, achieving stable combustion across 35%–100 % load conditions. With the
reduction of heat load, the combustion temperature and outlet NOx content decrease, while the outlet CO content shows an
increasing trend. The burner meets the nitrogen oxide emission standard under the highest load, and remains stable under low
load, with no risk of backfiring.
(4) Cooling water temperature has an effect on the combustion reaction. Reducing cooling water temperature is conducive to
reducing the concentration of OH in the combustion reaction, reducing the high temperature region and the generation rate of
NOx, so as to maintain the concentration of NOx at a low level.

Authorship contributions
Jiao Jian—Writing-Original Draft, Software, Data Curation, Investigation. Wang Yungang—Conceptualization, Formal analysis,
Data Curation, Supervision. Liu Yufei—Methodology. Zhang Xingbang—Writing – review & editing. Qinxin Zhao—Investigation,
Supervision.

Fig. 13. Effect of cooling water temperature on combustion characteristics.

11
J. Jiao et al. Case Studies in Thermal Engineering 52 (2023) 103669

Fig. 14. OH contours of burner under different cooling water temperature.

Fig. 15. Contour of burner NOx generation rate at different cooling water temperature.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to
influence the work reported in this paper.

Data availability

Data will be made available on request.

Acknowledgement
This research was supported by the National Natural Science Foundation of China (NO.52276085) and K.C. Wong Education
Foundation.

12
J. Jiao et al. Case Studies in Thermal Engineering 52 (2023) 103669

References
[1] C. Varis, S.O. Ozkilic, A biogas power plant from waste heat power generation system using Organic Rankine Cycle and multi-criteria optimization, Case Stud.
Therm. Eng. 44 (2023).
[2] D.G. Gunjo, V.K. Yadav, D.K. Sinha, I.E. Elseesy, G.M.S. Ahmed, M.A.H. Abdelmohimen, Development and performance evaluation of solar heating system for
biogas production process, Case Stud. Therm. Eng. 39 (2022).
[3] B. Ouhammou, A. Mohammed, S. Sliman, A. Jamil, B. Mohammed, F. Karouach, H. El Bari, T. Kousksou, Experimental conception and thermo-energetic analysis
of a solar biogas production system, Case Stud. Therm. Eng. 30 (2022).
[4] P. Weiland, Biogas production: current state and perspectives, Appl. Microbiol. Biotechnol. 85 (2010) 849–860.
[5] N. Scarlat, J.F. Dallemand, F. Fahl, Biogas: developments and perspectives in Europe, Renew. Energy 129 (2018) 457–472.
[6] M. Balat, H. Balat, Biogas as a renewable energy source A review, Energy Sources, Part A Recovery, Util. Environ. Eff. 31 (2009) 1280–1293.
[7] O.W. Awe, Y.Q. Zhao, A. Nzihou, D.P. Minh, N. Lyczko, A review of biogas utilisation, purification and upgrading technologies, Waste And Biomass Valorization
8 (2017) 267–283.
[8] A. Abdulnaim, A. Elkholy, M. Elmously, H. Moneib, W.L. Roberts, A.M. Elbaz, On the Stability and Characteristics of Biogas/Methane/Air Flames Fired by a
Double Swirl Burner, Flow Turbulence And Combustion, 2023.
[9] W. Zeng, H.A. Ma, Y.T. Liang, E.J. Hu, Experimental and modeling study on effects of N-2 and CO2 on ignition characteristics of methane/air mixture, J. Adv.
Res. 6 (2015) 189–201.
[10] M. Sahin, Combustion characteristics of various biogas flames under reduced oxygen concentration conditions, Energy Sources, Part A Recovery, Util. Environ.
Eff. 41 (2019) 2415–2427.
[11] W. Anggono, I.N.G. Wardana, M. Lawes, K.J. Hughes, S. Wahyudi, N. Hamidi, A. Hayakawa, Iop, biogas laminar burning velocity and flammability
characteristics in spark ignited premix combustion, in: 2013 International Conference on Science & Engineering in Mathematics, Chemistry and Physics, vol.
423, Scietech, 2013, 2013.
[12] A. Greco, D. Mira, X. Jiang, Effects of fuel composition on biogas combustion in premixed laminar flames, in: 8th International Conference on Applied Energy
(Icae2016), vol. 105, 2017, pp. 1058–1062.
[13] M. Saediamiri, M. Birouk, J.A. Kozinski, On the stability of a turbulent non-premixed biogas flame: effect of low swirl strength, Combust. Flame 161 (2014)
1326–1336.
[14] M. Ilbas, M. Sahin, Effects of turbulator angle and hydrogen addition on a biogas turbulent diffusion flame, Int. J. Hydrogen Energy 42 (2017) 25735–25743.
[15] H.S. Zhen, C.W. Leung, C.S. Cheung, A comparison of the heat transfer behaviors of biogas-H-2 diffusion and premixed flames, Int. J. Hydrogen Energy 39
(2014) 1137–1144.
[16] M.S. Abdallah, M.S. Mansour, N.K. Allam, Mapping the stability of free-jet biogas fl ames under partially premixed combustion, Energy (2021) 220.
[17] A. Effuggi, D. Gelosa, M. Derudi, R. Rota, Mild combustion of methane-derived fuel mixtures: natural gas and biogas, Combust. Sci. Technol. 180 (2008)
481–493.
[18] S. Devi, N. Sahoo, P. Muthukumar, Combustion of biogas in porous radiant burner: low emission combustion, Innovative Solutions For Energy Transitions 158
(2019) 1116–1121.
[19] I. Sivri, H. Yilmaz, O. Cam, I. Yilmaz, Combustion and emission characteristics of premixed biogas mixtures: an experimental study, Int. J. Hydrogen Energy 47
(2022) 12377–12392.
[20] Z.L. Wei, H.S. Zhen, C.W. Leung, C.S. Cheung, Z.H. Huang, Formations and emissions of CO/NO2/NOx in the laminar premixed biogas-hydrogen flame
undergoing the flame-wall interaction: effects of the variable CO2 proportion, Fuel (2020) 276.
[21] C.J. Mordaunt, W.C. Pierce, Design and preliminary results of an atmospheric-pressure model gas turbine combustor utilizing varying CO2 doping concentration
in CH4 to emulate biogas combustion, Fuel 124 (2014) 258–268.
[22] J. Li, H. Huang, Huhetaoli, Y. Osaka, Y. Bai, N. Kobayashi, Y. Chen, Combustion and heat release characteristics of biogas under hydrogen- and oxygen-enriched
condition, Energies 10 (2017).
[23] H.B. Gao, Z.G. Qu, W.Q. Tao, Y.L. He, J. Zhou, Experimental study of biogas combustion in a two-layer packed bed burner, Energy Fuel. 25 (2011) 2887–2895.
[24] E.E. Fordoei, K. Mazaheri, A. Mohammadpour, Numerical study on the heat transfer characteristics, flame structure, and pollutants emission in the MILD
methane-air, oxygen-enriched and oxy-methane combustion, Energy (2021) 218.
[25] L. Pillier, M. Idir, J. Molet, A. Matynia, S. de Persis, Experimental study and modelling of NOx formation in high pressure counter-flow premixed CH/air flames,
Fuel 150 (2015) 394–407.
[26] Gregory P. Smith, David M. Golden, Michael Frenklach, GRImech3.0 Mechanism, 1999.
[27] J.R. Reisel, Modeling of nitric oxide formation in high-pressure premixed laminar ethane flames, Combust. Flame 120 (2000) 233–241.
[28] A. Valera-Medina, A. Giles, D. Pugh, S. Morris, M. Pohl, A. Ortwein, Investigation of combustion of Emulated biogas in a gas turbine test rig, J. Therm. Sci. 27
(2018) 331–340.
[29] A.G. Liu, Y.D. Yang, L. Chen, W. Zeng, C.J. Wang, Experimental study of biogas combustion and emissions for a micro gas turbine, Fuel (2020) 267.
[30] P.R. Medwell, P.A.M. Kalt, B.B. Dally, Reaction zone weakening effects under hot and diluted oxidant stream conditions, Combust. Sci. Technol. 181 (2009)
937–953.

13