Experimental and Kinetic Study on Laminar Burning Velocities of High Ratio Hydrogen Addition to CH₄+O₂+N₂ and NG+O₂+N₂ Flames

Abstract

In 2020, energy-related CO₂ emissions reached 31.5 Gt, leading to an unprecedented atmospheric CO₂ level of 412.5 ppm. Hydrogen blending in natural gas (NG) is a solution for maximizing clean energy utilization and enabling long-distance H₂ transport through pipelines. However, insufficient comprehension concerning the combustion characteristics of NG, specifically when blended with a high proportion of hydrogen up to 80%, particularly with minority species, persists. Utilizing the heat flux method at room temperature and 1 atm, this experiment investigated the laminar burning velocities of CH₄/NG/H₂/air/He flames incorporating minority species, specifically C₂H₆ and C₃H_8, within NG. The results point out the regularity of S_L enhancement, reaching its maximum at an equivalence ratio of 1.4. Furthermore, the propensity for the enhancement of laminar burning velocity aligned with the observed thermoacoustic oscillation instability during fuel-rich regimes. The experimental findings were contrasted with kinetic simulations, utilizing the GRI 3.0 and San Diego mechanisms to facilitate analysis. The inclusion of H₂ augments the chemical reactions within the preheating zone, while the thermal effect from temperature is negligible. Both experimental and simulated results revealed that CH₄ and NG with a large proportion of H₂ had no difference, no matter whether from a laminar burning velocity or a kinetic analysis aspect.

1. Introduction

A significant energy demand has prompted profound apprehension due to the substantial surge in carbon dioxide (CO₂) emissions. In the year 2020, global CO₂ emissions from energy sources reached 31.5 Gt, resulting in an unprecedented level of atmospheric CO₂ of 412.5 ppm—exhibiting approximately a 50% increase compared to the levels observed at the onset of the Industrial Revolution [1]. Hydrogen energy, described as a clean, efficient, and multi-application energy carrier, could be a connection between traditional fossil and renewable energy. However, the challenge of hydrogen storage and transportation remains a big obstacle. The mixing of hydrogen with natural gas (NG) will be an important solution to transport H₂ in existing long-distance NG pipelines. Moreover, as an engine fuel, the synergy between natural gas and hydrogen is complementary. The incorporation of hydrogen amplifies the efficiency of natural gas combustion, diminishes carbon output, and expands its ignition threshold. Natural gas mitigates the challenges posed by hydrogen, including a low ignition energy, rapid combustion rate, and susceptibility to flash ignition and deflagration in realistic applications [2,3]. Therefore, it is crucial to explore the fundamental combustion properties of hydrogen/natural gas blends under more realistic conditions.

Laminar burning velocity, S_L, is a fundamental parameter of laminar premixed combustion and serves as the basis for studying turbulent combustion [4]. The investigation into laminar burning velocity aids in enhancing the comprehension of diverse combustion processes, including quenching, blow-off, and flashback [5]. Moreover, laminar burning velocity serves as a valuable tool for validating chemical kinetic mechanisms.

Methane is the dominant component of NG. The S_L of CH₄ with added H₂ has already been thoroughly investigated in diverse publications [6,7,8,9,10,11]. De Goey et al. [7] expanded upon the classical theoretical framework proposed by Peters and Williams for CH₄/air flames to examine the structural characteristics of CH₄/H₂/air flames at ϕ = 1. According to the theoretical analysis, with hydrogen addition a decreasing inner-layer temperature was observed, and it dominated the increasing burning velocity. Halter et al. [8] investigated the impact of H₂ addition (up to a 20% mixing ratio) and pressure on the properties of methane/air laminar flames across diverse equivalence ratios. The findings revealed that the increase in hydrogen resulted in an elevation of the laminar burning velocity and a decline in the flame’s sensitivity to stretch. Hermanns et al. [9] studied the impact of adding H₂ to CH₄ flames, with H₂ volumetric proportions reaching up to 40%, employing the heat flux method. Several studies on CH₄+H₂+air S_L under increasing conditions (300–650 K) have been complemented by Berwal et al. [6]. A power-law correlation was established to create a description of the temperature dependency, while at ϕ = 1.1 the temperature exponent exhibited a parabolic trend with a minimum value. Hu et al. [10,11] investigated the S_L of methane/hydrogen/air mixtures, utilizing a constant volume combustion chamber method. These experimental outcomes indicated two typical characteristics of CH₄+H₂+air combustion, where in ${\chi }_{H_2}$= 0.1–0.5 and ${\chi }_{H_2}$ = 0.9–1.0 conditions, in H₂ combustion zones, there exists a linear increment, and in transition regimes with H₂ mixed (${\chi }_{H_2}$= 0.6–0.8) there exists a logarithmic amplification in S_L.

Typically, besides methane, there always exist minority species in natural gas, like ethane and propane. The mixture of these components with hydrogen has not been investigated well. Nilsson et al. [4] examined the influence of H₂ incorporation on the S_L of methane/ethane/propane blends. It was found that, compared to pure methane, the heavier hydrocarbon exhibited a reduced augmentation in S_L with hydrogen addition. Huang et al. [12] analyzed the S_L of natural gas with hydrogen volumetric fractions between 0 and 100% and ϕ ranging from 0.6 to 1.4. With increasing hydrogen content, S_L increased exponentially, while the flame instability intensified.

However, research about the effects of H₂ on real composition NG, especially with large proportions of H₂ (up to 0.8), has not been investigated and understood well. Furthermore, the responses of different parts of the components of natural gas to H₂ addition have rarely been investigated, as well. Therefore, it is necessary to investigate the effect of real composition NG mixed with high proportions of H₂.

Nevertheless, as mentioned above, high S_L arising from a large proportion of H₂ addition to NG will generate combustion instability. It is acknowledged that the high reactivity of hydrogen will raise multiple safety concerns when blended with ethane, methane, and propane. These effects also lead to significant changes in the ignition and propagation characteristic parameters of NG/H₂/air flames: a reduction in minimum ignition energies and self-ignition delay times, alongside the growth in adiabatic flame temperatures and laminar burning velocity [13,14,15,16]. Several studies have revealed the effects of H₂ on natural gas components, including the broadening of the flammability limits of CH₄/air flames [13,17,18] and increases in S_L [19,20,21,22]. Meanwhile, research has extensively focused on hydrogen-enriched combustion instability, not only from acoustically coupled pulsating motion aspects [23], but also from intense interactions between hydrodynamic and acoustic modes [24]. Following these theories, combustion instability has also been observed and taken into consideration in this study.

Based on the above-mentioned motivations, this research aims to fill in gaps regarding a large proportion of hydrogen, which can reach up to 0.8, mixed with a complex composition of real natural gas. The aims of this study are as follows: (1) to measure the precise laminar burning velocities for H₂ mixed with Chinese typical Natural Gas, as

Table 1. Composition of studied natural gas (NG).

Types of Natural Gas	Methane (%)	Ethane (%)	Propane (%)
NGOne	98	1.53	0.47
NGTwo	98.47	1.53	0
NGThree	99.53	0	0.47

lists (#97 volume fraction: 96.68% CH₄ + 1.53% C₂H₆ + 0.47% C₃H₈) [25], with varying volumetric proportions at 1 atm and room temperature, and to analyze thermoacoustic oscillation between different experimental working conditions; and (2) to validate and understand the profoundly kinetic mechanism influencing the S_L of blended fuels with H₂ added.

2. Materials and Methods

2.1. Experimental Setup

The S_L of H₂/CH₄/NG/He/air flames were experimentally measured through the heat flux method under 1 atm and room temperature conditions. Figure 1 displays the current laboratory configuration schematic representation in this study. The heat flux method, including its principles and associated uncertainties, has been comprehensively described in our prior publications [26,27,28,29]. The setup was mainly composed of a heat flux burner and a gas supply system. Gas flow rates were regulated using 5 mass flow controllers (MFC, Alicat) to generate unburnt gas mixtures based on experimental parameters. The gas mixture was directed through the burner chamber and ignited above the burner plate, resulting in the formation of a planar laminar flame. The burner plate featured a 30 mm diameter perforated hexagonal pattern with 0.3 mm diameter apertures and a 0.4 mm spacing [30].

Table 1 lists the three types of natural gas adopted in the present investigation, named as NGOne, NGTwo, and NGThree. NGOne is a typical Natural Gas in China, and the others are artificial ones, for exploring the effects of different hydrocarbon fractions on S_L and the combustion process.

Table 2. Experimental conditions adopted in this work.

Case	Mole Fraction of Fuel Components		Helium	Equivalence $\mathbf{Ratio} \mathit{\varphi }$	Unburnt Gas Temperature Tu (K)
	H2	Natural Gas *
a1/b1/c1/d1	0	1	No	0.6–1.4	298
a2/b2/c2/d2	0.2	0.8	No	0.6–1.4	298
a3/b3/c3/d3	0.4	0.6	No	0.6–1.4	298
a4/b4/c4/d4	0.6	0.4	No	0.6–1.4	298
a5/b5/c5/d5	0.7	0.3	Yes	0.6–1.25	298
a6/b6/c6/d6	0.8	0.2	Yes	0.6–1.3	298

* Note: Natural gas indicates pure methane (a1–6), NGOne (b1–6), NGTwo (c1–6), or NGThree (d1–6).

outlines the experimental conditions employed. Notably, in conditions involving high hydrogen mixing ratios (up to 70% and 80%), the unburned gas mixture was diluted with helium to maintain laminar burning velocities within the heat flux S_L tolerance range of approximately 70 cm/s. The volumetric fraction of helium, ${\chi }_{He}$, in the total gas mixture was 30%, as described in Equation (1). The mole fractions of hydrogen, methane, or natural gas within the fuel composition, ${\chi }_M$, are described in Equation (2) with the subscript [M] denoting H₂, CH₄, NGOne, NGTwo, or NGThree.

$${\chi }_{He}=V_{He}/\left(V_{fuel}+V_{air}\right)$$

(1)

$${\chi }_M=V_M/\left(V_{H_2}+V_{{C_{x}H}_y}\right)$$

(2)

The equivalence ratio ϕ is defined as follows, in Equation (3):

$$\varphi =\frac{\left(V_{H_2}+V_{C_x{H}_y}\right)/V_{air}}{{\left[\left(V_{H_2}+V_{C_x{H}_y}\right)/V_{air}\right]}_{Stoichiometric}}$$

(3)

2.2. Modeling Details

For modeling, the PREMIX module of Chemkin Pro was utilized with curvature and gradient parameters set at 0.02 for grid independence. Several kinetic mechanisms, the GRI 3.0 mechanism [31], San Diego mechanism [32], the USC mech Version II [33], Aramco 2.0 [34], and the CRECK mechanism [35], have been developed to describe hydrogen and hydrocarbon chemical reaction processes. The most mature and complete mechanism is GRI 3.0, where the validation of the model for C1–C4 hydrocarbons has demonstrated its accurate prediction of laminar burning velocities for methane, ethane, propane, and methane/hydrogen blends. In the present work, GRI 3.0 was therefore selected for the following kinetic analysis, and the San Diego mechanism was adopted to investigate the cases with helium addition.

3. Results and Discussion

3.1. Experimental Results of Methane and Hydrocarbon Mixtures at 298 K

The laminar burning velocity of methane/air at 298 K and 1 atm has been studied completely. Through extensive experimental procedures and statistical analysis, the scatter in S_L values for methane/air across different setups and environments was within ±2 cm/s. To assess the reliability of the present laboratory setup, Figure 2a presents the measured S_L of methane/air flames, considering varying equivalence ratios at room temperature and 1 atm from the current work and the ones by Han et al. [28], Wang et al. [29], and Nilsson et al. [4], who also used the heat flux method. The measured results of this study align with published data, affirming the experimental setup’s validity. Figure 2c,d show the measured S_L of natural gas/air flames, which were simulated using the five selected mechanisms. Generally, there was good agreement between the simulations and experimental findings, except for a slight overprediction at ϕ < 0.8 and ϕ > 1.3. At NGOne, for the fuel-lean and fuel-rich side, simulations seemed to overpredict a little compared with other fuels. Moreover, the GRI 3.0 matched well with NGOne experimental results around stoichiometric conditions, where the laminar burning velocities were 37.99 cm/s, 39.45 cm/s, and 40.7 cm/s, respectively, for NGOne, NGTwo, and NGThree. All of the experimental data are provided in Table S1 of the Supplementary Material.

3.1.1. Effect of H₂ Mole Fraction on Laminar Burning Velocity

Figure 3a illustrates the experimental and modeled S_L of CH₄/H₂/air/He mixtures at room temperature and 1 atm, graphed as a function of the equivalence ratio, with ${\chi }_{H_2}$ spanning 0 to 0.6 without helium and 0.7 to 0.8 with helium dilution. The laminar burning velocities of NGOne, NGTwo, and NGThree mixed with H₂ are displayed in Figure 3b–d. Detailed experimental data can be found in Table S1 of the Supplementary Material.

In Figure 3, the literature data were obtained from previous works of Han et al. [28], Wang et al. [29], Halter et al. [8], Hu et al. [10] and Huang et al. [12]. When comparing experimental and simulated results, the GRI 3.0 mechanism demonstrated better accuracy in predicting the S_L of methane/ethane/propane/hydrogen blends compared to methane/hydrogen blends. Furthermore, it is crucial to emphasize that as ${\chi }_{H_2}$ increased, the discrepancy between experimental results and simulations decreased, no matter whether it was with helium dilution or not. As for methane, NGOne, NGTwo, and NGThree blended with hydrogen, the equivalence ratios of peak values gradually rose from ϕ = 1.05 to around ϕ = 1.1 under the 60% H₂ mole fraction, even up to about ϕ = 1.15 under the 80% H₂ mole fraction. The corresponding S_L for methane/hydrogen blends were 36.56 cm/s, 74.28 cm/s, and 79.53 cm/s, respectively. Even based on helium addition, the tendency of increasing ϕ with H₂ blends remained, and this phenomenon may be derived from the peak S_L of pure hydrogen occurring at around ϕ = 1.7, so that its addition to C_xH_y affects the equivalence ratio, shifting the peak velocity towards the fuel-rich side. Helium, in this case, acts as a diluent in the fuel and oxidizer mixture.

The hydrogen addition to hydrocarbon blends led to non-monotonic impacts on the S_L at diverse equivalence ratios. To assess the magnitude of the S_L enhancement resulting from hydrogen addition to CH₄- or CH₄/C₂H₆/C₃H₈-blended flames, a parameter $\xi$ was introduced to quantify the normalized effect of enhancement; in Equation (4), taking $S_L\left(0.4,\varphi \right)$ as an example, $\xi \left(0.4-0.2,\varphi \right)$ can be denoted as [27]:

$$\xi \left(0.4-0.2,\varphi \right)=\frac{S_L\left(0.4,\varphi \right)-S_L\left(0.2,\varphi \right)}{S_L\left(0.2,\varphi \right)}\times 100\%$$

(4)

where $S_L\left(0.4,\varphi \right)$ indicates the laminar burning velocity of H₂/CH₄/NGOne/NGTwo/NGThree mixed with a H₂ ratio of 0.4, and $\xi \left(0.4-0.2,\varphi \right)$ represents the enhancement between $S_L\left(0.4,\varphi \right)$ and $S_L\left(0.2,\varphi \right)$. The potential systematic errors associated with the heat flux method can be considered negligible, and the consistent experimental setup across different conditions further reduces potential uncertainties. According to Equation (5), the quantification of $\xi$ uncertainty can be determined by [30]:

$$△\xi =\left|\frac{d\xi }{d{S}_L\left({\chi }_{H_2},\varphi \right)}\right|·△S_L\left({\chi }_{H_2},\varphi \right)$$

(5)

Figure 4 displays the enhancement parameter $\xi$ for CH₄/air, NGOne/air, NGTwo/air and NGThree/air flames with 20–80% H₂ added at room temperature and 1 atm with respect to the equivalence ratio. Apart from extremely fragmented data points, it can be pointed out that the results from both experimental and simulation analyses revealed that the most prominent enhancements of S_L occurred at fuel-rich sides for all hydrocarbon blends at different ${\chi }_{H_2}$. Hydrogen addition significantly amplified the sensitivity of off-stoichiometric flames, owing to the reduced reactivity and chemical heat release observed in lean and rich premixed flames, as compared to stoichiometric conditions. In addition, the fuel-rich side contained a large amount of hydrogen, so peak values occurred at these conditions. The equivalence ratio of peak S_L enhancement achieved around ϕ = 1.4, and its peak values continually increased with hydrogen addition, where the $\xi$ value was attained from 37.99% to 114.71%. With hydrogen added, the maximum values of S_L enhancement tended to the fuel-rich side. Helium addition did not affect this trend, indicating that it mainly contributes to the physical characteristics of S_L rather than chemical effects. Notably, the largest S_L enhancement was observed for the case with 80% hydrogen addition.

3.1.2. Effect of CH₄/C₂H₆/C₃H₈ Mole Fraction on Laminar Burning Velocity

Figure 3 shows different levels of fitness between experimental S_L and predicted ones. From these figures, these five mechanisms are both consistent with CH₄, NGOne, NGTwo, and NGThree blended with H₂. The trends of equivalence ratios corresponding to peak S_L increase with H₂ addition did not change between the four studied fuels. This means that the influence of H₂ is barely irreversible under different saturated hydrocarbon conditions.

Figure 4 shows the enhancement parameter $\xi$ of four experimental gasses. The simulation and measured data revealed the non-monotonic rule of S_L enhancement with H₂ increasing. The best agreement between the experiments and modeling of the NGTwo condition is notable compared with the other three types of fuel. Meanwhile, the laminar flames generated constant and regular beeps under fuel-rich conditions. At the same time, the combustion stability was weakened, manifested as jumping flames as well as irregular flame shapes, which indicates that the current S_L is undesirable. The corresponding beeps phenomenon is detailed in

Table 3. The beeps phenomenon occurs in this work.

Mole Fraction of H2 (%)	Natural Gas Type	Beeps Occurrence Conditions
0.6	NGThree	$\varphi$ = 1.4
0.7	CH4, NGThree	$\varphi$ = 1.3
0.7	NGOne, NGTwo	$\varphi$ = 1.25
0.8	CH4, NGOne, NGTwo, NGThree	$\varphi$ = 1.35

.

The constant and regular beeps during combustion on heat flux could be explained by the thermoacoustic oscillation instability introduced before by Matsuyama et al. [23] and Hemchandra et al. [24] According to the discussion stated above, the condition of 70% hydrogen addition was more unstable, and under the condition of NGThree mixed with H₂, which mainly consisted of CH₄ and C₃H₈, beeps occurred ahead at ${\chi }_{H_2}$= 0.6. This phenomenon provides valuable guidance for determining the appropriate volumetric fraction of H₂ in hydrocarbon blends to avoid unstable and hazardous conditions in future industrial hydrogen-doped combustion applications.

3.2. Kinetic Analysis

To obtain the underlying patterns of key intermediate species during combustion, the behaviors of species, H, O, OH, NO, and CH₃ were presented. The profiles obtained from H₂/CH₄/NGOne/air flames at ϕ = 1.4 (where S_L enhancement attained its maximum value) are displayed in Figure 5, where the solid line represents the ${\chi }_{H_2}$ = 0 condition, the dot-dash line represents the ${\chi }_{H_2}$ = 0.4 condition, and the dotted line represents the ${\chi }_{H_2}$ = 0.6 condition. Figure 5 illustrates the significantly elevated mole fraction of the H radical, and the OH radical, CH₃ radical, and O radical appeared subsequently. With the H₂ addition increasing, the analysis revealed that the maximum mole fractions of these species predominantly concentrated in the upstream region of the flame, displaying an increasing trend. This suggests that these radicals substantially impacted combustion within the preheating zone (0.05–0.1 cm). Such regularity can interpret the thermoacoustic oscillations mentioned above and the areas where thermoacoustic oscillation instability occurs in the preheating zone (0.05–0.1 cm), according to previous analysis.

Meanwhile, the NO component had a trend of rising first and then falling from ${\chi }_{H_2}$ = 0 to ${\chi }_{H_2}$ = 0.6, thus meaning that although H₂ addition may result in higher NO emission initially, due to high temperature, NO emission will decline under the condition of a large proportion of hydrogen doping due to relatively complete combustion, which can even be lower than the 0% H₂ condition in the downstream region of the flame. Additionally, when comparing H₂/CH₄/air flames, the radicals in H₂/NGOne/air flames reached their peak values further upstream in the flame, and the reaction areas of CH₃ radicals became narrower. This indicates that the reaction of H₂/NGOne/air occurs earlier and intensifies, explaining the higher instability observed in NGOne blended with H₂, as discussed in Table 3. H₂/NGTwo/air and H₂/NGThree/air flames had the same regularity compared to H₂/CH₄/air flame.

Figure 6 gives a clear display of H₂/CH₄/air/He and H₂/NGOne/air/He flames, which revealed the same trend consistent with these radicals in H₂/CH₄/NGOne/air flames without He. In comparison to the H₂/CH₄/air/He flame, the H radical in the H₂/NGOne/air/He flame was slightly higher at the 70% H₂ condition and lower at the 80% H₂ condition.

Figure 7 depicts the peak mole fractions of multiple crucial radicals at various positions within the flame, which represent the stoichiometric and H₂/CH₄/air conditions. In Figure 7, Figure 8 and Figure 9, the flame temperature profile is also depicted to illustrate the thermal effect arising from H₂ addition. Corresponding to the discussion mentioned above, the growth of the H₂ ratio in H₂/CH₄/air would prompt the maximum mole fraction of H, O, and OH radicals but reduce peak mole fraction of the CH₃ radical, which could potentially arise from the limited availability of C_xH_y radicals in the hydrogen-rich flame, for stoichiometric H₂/CH₄/air flames. The slopes of O, OH radicals’ growth and CH₃ radical decline were much gentler and flatter than the slope of H radical growth. The violent slope of the H radical growth matched well with the increment of S_L as the H₂ proportions grew, which indicates that the S_L enhancement exhibited a robust positive correlation with it. The slopes of H, O, OH, and CH₃ radicals in H₂/NGOne/NGTwo/NGThree/air/He exhibited the same trends as H₂/CH₄/air/He.

Figure 8 represents the maximum mole fraction, adiabatic temperature, and laminar burning velocity of studied radicals at ${\chi }_{H_2}$ = 0.2, 0.8 conditions of H₂/CH₄/air/He flames at diverse equivalence ratios. To aid in visualization, the mole fraction of CH₃ radical was multiplied by 10, and denoted as “CH₃×10”. H₂/NGOne/air/He, H₂/NGTwo/air/He, and H₂/NGThree/air/He had the same regularity as H₂/CH₄/air/He. According to Figure 8, which illustrates the equivalence-ratio-dependent variations in radical concentration, adiabatic temperature and S_L, the chemical effect, represented by H radical concentration, exerted a more pronounced influence on laminar burning velocity than adiabatic temperature as H₂ proportions ranged from 0.2 to 0.8. With the H₂ proportion growing, the equivalence ratios of the peak temperature barely changed, while the equivalence ratios corresponding to the maximum H radical concentrations tended to be higher. Note that the mole fractions of the CH₃ radical declined violently because of the decrease of C_xH_y in mixtures. Meanwhile, the mole fraction of the species under investigation initially rose and subsequently decreased as the equivalence ratio varied, and equivalence ratios of their peak values had a deviation to the fuel-rich sides, consistent with the regularity of S_L presented above.

Figure 9 portrays the concentration of NOx with H₂ ratio from 0 to 0.8 in H₂/CH₄/air/He flames. EINOx, as a NOx emission index, is reported as grams of NO₂-equivalent formed per kilogram of fuel consumed. In Figure 9a, with the H₂ volumetric ratio rising from 0% to 60%, EINOx tended to generate downstream while pure CH₄ EINOx reached its local peak value upstream. All generation origin zones were far from the preheating zone (0.05–0.1 cm). In Figure 9b, the EINOx of the stoichiometric condition increased with the H₂ ratio, consistent with T_ad growth. Furthermore, with the H₂ fraction ranging from 0% to 30%, EINOx grew and then, with the H₂ fraction ranging from 30% to 60%, EINOx declined at ϕ = 1.4. Under helium dilution, the extreme point of EINOx occurred at ${\chi }_{H_2}$ = 0.72. This phenomenon could be attributed to incomplete combustion on the fuel-rich side. It is interesting to point out that in Figure 9c, although the H₂ ratio attains 0.7 and 0.8, EINOx is even smaller than other experimental conditions because of the helium dilution added, resulting in a low temperature and mole fraction of total fuel in the reactant.

Figure 10 illustrates the fuel-rich side’s rate of production (ROP) of H radicals at ϕ = 1.4, providing insights into the mechanism underlying the observed behavior of various radicals in the upstream flame zone with H₂ and He added. The top 10 vital elementary reactions involving active radical production or consumption are represented by different colors. Generally, reactions R1 and R2 have a prevailing influence on H radical formation at different conditions, which were associated with CO/H₂ chemistry. With hydrogen addition increasing without helium, the H radical main reaction areas kept moving upstream of the flame and the reaction zones corresponding to peak values narrowed, with peak values growing accompanied. Under 30% He blended conditions, the H radical main reaction areas moved downstream and widened slightly, but were still angled towards upstream with the H₂ volumetric fraction rising. Associated with discussions of mole fraction mentioned above, H₂/NGOne/air/He, H₂/NGTwo/air/He, and H₂/NGThree/air/He had the same regularity of H radical reactions, which indicates that the H radicals in H₂/NGOne/NGTwo/NGThree/air/He flames attained the peak further upstream. To sum up, the trends of H radicals, particularly under 30% He blending conditions, provide further insights into the thermoacoustic oscillation observed in large-scale hydrogen doping experiments.

$$\mathrm{O}\mathrm{H}+{\mathrm{H}}_2=\mathrm{H}+{\mathrm{H}}_2\mathrm{O}\hspace{1em}\hspace{1em}\hspace{1em}(\mathrm{R}1)$$

$$\mathrm{O}+{\mathrm{H}}_2=\mathrm{H}+\mathrm{O}\mathrm{H}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}(\mathrm{R}2)$$

To further elucidate the mechanisms underlying the augmentation of laminar burning velocity with H₂ blends with CH₄ and natural gases, detailed kinetic analyses have been carried out through GRI 3.0 without He blends and the San Diego mechanism with He blends. Figure 11 presents the A-factor sensitivities of S_L for the 10 most influential reactions in H₂/CH₄/air flames under the specified conditions ($\varphi$ = 0.6, 1.0, 1.3, and 1.4) for ${\chi }_{H_2}$ = 0 and 0.4 individually. Figure 11 reveals that the reaction R3, which has been considered the pivotal chain-branching reaction in hydrocarbon flame [36], plays a dominant role in positive sensitivity under various circumstances. This phenomenon may arise from the significance of the reaction R3 in H₂/CO chemistries, which could analogize kinetic processes between H₂/CO and H₂/C_xH_y. The chain termination reaction R4 exhibits the most significant negative sensitivity, indicating the notable impact of C-containing species like CH₃ radical on laminar burning velocity.

$$\mathrm{H}+{\mathrm{O}}_2=\mathrm{O}+\mathrm{O}\mathrm{H}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}(\mathrm{R}3)$$

$${\mathrm{C}\mathrm{H}}_3+\mathrm{H}\left(+\mathrm{M}\right)={\mathrm{C}\mathrm{H}}_4\left(+\mathrm{M}\right)\hspace{1em}\hspace{1em}\hspace{0.17em}(\mathrm{R}4)$$

$$\mathrm{H}{\mathrm{O}}_2+\mathrm{C}{\mathrm{H}}_3=\mathrm{O}\mathrm{H}+\mathrm{C}{\mathrm{H}}_3\mathrm{O}\hspace{1em}\hspace{1em}\hspace{1em}(\mathrm{R}5)$$

$$\mathrm{O}+\mathrm{C}{\mathrm{H}}_3=>\mathrm{H}+{\mathrm{H}}_2+\mathrm{C}\mathrm{O}\hspace{1em}\hspace{1em}\hspace{1em}(\mathrm{R}6)$$

With the hydrogen addition increasing from 0 to 0.4, the positive sensitivity factors of the reaction R5 decreased significantly, while the positive sensitivity factors of the reaction R6 increased substantially. This indicates that the reactions R5 and R6 have a competing relationship to consume CH₃ radical, and under large proportions of hydrogen doping circumstances CH₃ radical has been transferred to generate the H radical instead of OH radical. Of particular note is the growth rate of the positive sensitivity factors of the reaction R6, which is particularly much larger at ϕ = 1.4, with the H₂ mole fraction increasing. This tendency may be ascribed to the high temperature and reactivity of highly active free radicals under large proportions of hydrogen doping reactions at fuel-rich conditions.

As to the instability of thermoacoustic oscillation mentioned above at a 70% and 80% H₂ volumetric fraction with helium blended, Figure 12 points out that the reaction R7 makes a replacement for the reaction R6, which means that the chain-branching reaction R6 was no more vital. The disappearance of such a chain-branching reaction could track back the temperature drop due to He addition, corresponding to the analysis of H ROP. According to Figure 12, it can also be found that the reaction R8 has the second most negative sensitivity factors accelerating kinetic processes through this H-atom abstraction reaction. Other sensitivity analyses on three types of H₂/NG/air/He flames have similar trends compared with Figure 11 and Figure 12.

$$\mathrm{O}+\mathrm{C}{\mathrm{H}}_3=\mathrm{H}+\mathrm{C}{\mathrm{H}}_2\mathrm{O}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}(\mathrm{R}7)$$

$$\mathrm{H}+\mathrm{O}\mathrm{H}+\mathrm{M}={\mathrm{H}}_2\mathrm{O}+\mathrm{M}\hspace{1em}\hspace{1em}\hspace{1em}(\mathrm{R}8)$$

4. Conclusions

This study investigated the augmentation of H₂ addition into CH₄/NGOne/NGTwo/NGThree/air with and without He at room temperature and 1 atm. The experimental measurements adopted the heat flux method and kinetic modeling using GRI 3.0 and the San Diego mechanism to compare the consistency between laboratory results and simulations. In summary, the primary findings are summarized as follows.

• The S_L measured by the heat flux method fits well with GRI 3.0 and San Diego mechanism simulations at various conditions. Furthermore, S_L has much more consistency with simulated lines in the case of large proportions of hydrogen doping. At the same time, combustion stability weakens the manifestation in jumping flames as well as irregular flame shapes for the conditions of large H₂ volumetric fraction. The constant and regular beeps accompanying combustion on heat flux could be explained by thermoacoustic oscillation instability.
• The significant factor, S_L enhancement, has been observed to reach peak values when fuel-rich in contrast to lean and stoichiometric fuel blends. The equivalence ratio of peak S_L enhancement achieved around ϕ = 1.4 and continued to increase with hydrogen addition, where the $\mathsf{\xi }$ could reach from 37.99% to 114.71%, respectively.
• With H₂ addition growing, the analysis revealed that the maximum mole fractions of these species predominantly concentrate in the upstream region of the flame, presenting an increasing trend. This suggests that these radicals substantially impact combustion within the preheating zone. (0.05–0.1 cm). On the other hand, in hydrogen blended mixtures, the chemical effect outweighs the thermal effect arising from increased temperature.
• When ${\chi }_{H_2}$ varied from 0.3 to 0.6, the peak S_L enhancement achieved around ϕ = 1.4 and, while at this condition, EINOx descended. Blending with a H₂ ratio beyond 0.7 is not recommended because of the intense thermoacoustic oscillation instability, especially for natural gas mainly composed of CH₄ and C₃H₈. These phenomena could inform the amount of H₂ volumetric fraction (recommended range is 0.3–0.6) used in hydrocarbon blends to achieve high efficiency combustion and low emission and avoid unstable and dangerous conditions.