Toward Zero Carbon Emissions: Investigating the Combustion Performance of Shaped Microcombustors Using H₂/Air and NH₃/Air Mixtures

Abstract

The research effort in the microcombustor field has recently increased due to the demand for high-performance systems in microelectromechanical and micro power generation devices. To address rising concerns about pollutants from fossil sources, zero-carbon fuels such as hydrogen (H₂) and ammonia (NH₃) have been considered as an alternative in microcombustion processes. In a microcombustor, the surface area-to-volume ratio is much higher compared to conventional combustion systems, resulting in faster heat transfer rates and more intense combustion reactions. However, achieving efficient mixing of fuel and an oxidizer in a microcombustor can be challenging due to its small size, particularly for highly reactive fuels like H₂. For NH₃, challenges in microcombustion involve a low reactive, high ignition temperature (923 K vs. 793 K of H₂) and high concentration of NO_x combustion products. Therefore, studying the performance of these fuels in microcombustors is important for developing clean energy technologies. In this paper, to explore features of non-premixed NH₃/air and H₂/air combustion in micro-scale combustors, an Ansys Fluent numerical investigation was conducted on a Y-shaped microcombustor. Results show that for combustion with H₂, stationary flames can be achieved even at lower equivalence ratios. Additionally, the pollutants generated from H₂ in the flame are generally twice those of NH₃. The overall efficiency of the microcombustor is two times greater for NH₃ conditions than for H₂ conditions.

1. Introduction

1.1. Background

The 17 Sustainable Development Goals (SDGs) outlined for 2030 emphasize the importance of improving the performances of microthermal energy conversion devices [1]. Combustion-based microdevices have revolutionized the field of energy conversion in the past 30 years due to their high energy density, low weight, compact size, and long lifespan [2,3,4,5,6,7,8,9]. These devices not only offer energy benefits but also have significant economic and social implications. Microcombustors are essential components in compact energy systems, such as micro-thermophotovoltaic [9,10,11] and micro-thermoelectric systems [12,13]. In these applications, the main challenge is to achieve the highest possible temperature difference between the inlet and outlet to reach wall temperatures between 1300 and 1800 K [14]. The microscale power generation length scale is typically defined as being below 1 mm [15]. The microcombustion technology revolution has led to systems that are lighter and smaller and have higher energy density [1,16]. It has been widely adopted in various engineering power systems such as micro rotor engines and turbines [17,18]. The high area-to-volume ratio and short mixture residence time make maintaining a stable flame in micro burner systems a challenging task [19,20,21,22].

The study of flame stability is crucial in numerous fields including energy production, propulsion systems, and safety engineering. The stability of flames is influenced by various factors including heat and flow recirculation, which have been shown by multiple researchers to have a significant impact. Addressing flame stability challenges requires innovative approaches, including exploring new mixing configurations and implementing advanced thermal management strategies [23,24].

Conducting experimental microcombustion research is complex, time-consuming, and expensive, which is why significant computational effort is invested in numerical modeling to reduce the need for experimental validation and final design phases [25]. Micro-Electro-Mechanic Systems (MEMSs) are a promising alternative to a combustion device, but their effectiveness is hindered by various challenges such as batteries’ short life, recharging time, and energy density [26]. Currently, the best battery performances achieve 0.5 kJ/g [27], which is 300 times less than hydrogen and leads to reduced resources and cost [28]. Short battery life and long recharging times are other persistent issues with conventional batteries.

1.2. Literature Review

The impact of geometry on microcombustion applications is a highly relevant topic today, as evidenced by numerous studies [29]. Variations in geometry can greatly influence thermophysical properties such as heat conduction and residence time in the combustion chamber, affecting the device performances. For instance, Peng et al. [30] conducted numerical simulations to explore the effects of reducing wall thickness in hydrogen–air combustion to improve the flame area and wall temperature. They simulated a premixed H₂-air combustion in the microcombustors with different backward-facing steps and wall thicknesses.

Another crucial aspect that has recently been investigated is the impact of increased thermal conductivity on the combustion area, as discussed in [31]. The researchers studied three catalytic microcombustors made of various materials. They compared their experimental results with those obtained from CFD simulations and observed that the performance of each material varied.

Improved flame stability can be achieved through better control of thermal energy loss, resulting in a higher inlet mass flow rate, increased maximum temperature, and improved combustion performance. Effective flame recirculation and thermal management in microcombustors can also enhance residence time and stabilize the flame [32,33]. Recently, Resende et al. [34] analyzed the flame dynamics of a H₂/air mixture in a wavy micro-channel. Their results showed that at low inlet velocity (4 m/s), the flame became stable, and, at higher inlet velocities, the flame showed pulsatory burst dynamics.

Despite hydrogen’s advantages, there are still challenges associated with its storage, distribution, and infrastructure. NH₃ potential as an alternative fuel stems from its H₂-rich composition and the absence of carbon, which makes it an attractive option in the pursuit of reducing greenhouse gas emissions (GHGs) [35,36,37]. Microcombustion has serious heat loss due to the large ratio of surface area to volume. This leads to some unwanted characteristics such as a narrow blow-out limit. NH₃ has relatively low flame propagation speed and chemical activity compared to other hydrocarbon fuels. Therefore, the combustion condition for NH₃ is demanding especially in a microcombustor. This work compares the flame behavior of a H₂-air and NH₃/air mixture for combustion in a Y-shaped microcombustor.

The effects of using the mixture of NH₃/air are discussed in terms of emissions and combustion efficiency compared to the H₂ condition.

Within the present work, authors start from their previous studies [38] to develop a numerical model is developed in a Chemkin environment to study the combustion evolution for the mixture in a non-premixed microcombustor. The Chemkin model is used to validate the laminar burning velocity (LBV). Secondly, the developed scheme of chemical equations developed in Chemkin that considers all the species provided from the combustion process is exported to steady-state CFD simulations that use a continuum Navier–Stokes approach. The validity of the model is confirmed by comparing simulation results with experimental data in the case of H₂. The simulations are performed using Ansys Fluent 2023 [39].

There are only a few studies on microcombustion fueled with NH₃ and several studies using methane (CH₄) or H₂, also in a blend, as fuel. Due to the diverse conversion pathways of NH₃ combustion, different oxides will be generated depending on the fuel. This leads to changes in energy conversion efficiency. The main goal of the study proposed here is to underline the effects of different zero-carbon fuels in a microcombustor on the improvement in thermal performances and on environmental impact of these kinds of fuels.

Utilizing a numerical CFD approach, the paper carries out a step beyond the present knowledge of the thermo-chemical processes distinguishing a microcombustor based on new technology, as it investigates fundamental features that are strongly related to the inefficiency of the heat thermal exchange in the energy conversion process.

2. Materials and Method

2.1. Geometric Features of the Microcombustor

The present study investigates a specific model geometry, featuring a Y-shaped combustion chamber of 200 mm in length with a 90-degree angle and two inlet channels for the considered fuel (hydrogen and ammonia) and dry air, as previously described by Xiang et al., 2020 [40], Xang et al., 2021 [41]. The combustion chamber walls are made of 1-mm-thick quartz glass, and the pressure and temperature are set to 1 atm and 300 K, respectively.

2.2. Modelling Combustion and Kinetic Mechanism Details

The CFD software Fluent 2023 [39] is employed to address the core equations representing conservation of mass, momentum, energy, and chemical species within the fluid, leveraging a second-order upwind scheme. This methodology is reputed for its enhanced accuracy and stability when simulating fluid flows exhibiting strong gradients and discontinuities. A coupled algorithm, recognized as a pressure–velocity coupling technique, is utilized in the simulation. This technique embodies a linkage between velocity and pressure corrections to uphold mass conservation and deduce the pressure field, accompanied by a second-order pressure spatial discretization. Gradients are ascertained utilizing the least squares cell-based scheme. Convergence in the simulation is inferred when the residuals for all parameters fall below 10⁻⁴ and the average temperature and the mass fractions of main combustion products are constant. A three-dimensional (3D) steady model is adopted, and the impact of gravity is disregarded. The fundamental governing equations for heat and mass transfer are presented. A segregated solution solver coupled with a sub-relaxation method is employed.

$$\frac{\partial \rho }{\partial t}+\frac{\partial (\rho u_i)}{\partial x_i}=0$$

(1)

$$\frac{\partial (\rho u_j)}{\partial t}+\frac{\partial (\rho u_i{u}_j-{\tau }_{ij})}{\partial x_i}=\frac{\partial p}{{\partial x}_j}$$

(2)

$$\frac{\partial (\rho h)}{\partial t}+\frac{\partial (\rho u_{i}h)}{\partial x_i}=\frac{\partial (\lambda \partial T_f)}{\partial t}-\sum j\frac{\partial (h_j{J}_j)}{\partial x_i}+\sum j{h}_j{R}_j$$

(3)

$$\frac{\partial ({\rho Y}_i)}{\partial t}+\frac{\rho u_i{Y}_i}{\partial x_i}=\frac{\partial J_i}{\partial x_i}-R_i$$

(4)

where $\rho$ is the gas density, p is the pressure, u is the velocity, ${\tau }_{ij}$ is the stress tensor, h is total enthalpy, Ji is the diffusion flux of species i, Y_i is the mass fraction of species, T is the temperature, Rj is the net rate of production of species j through chemical reaction i, $\lambda$ is the thermal conductivity. The effects of convection and radiation (P1 model) on the heat losses are computed considering Equation (5). An ideal gas flow model is adopted with a mixing law, c_p.

$${\dot{Q}}_{loss}=h_{0}A\left(T_w-T_0\right)+\epsilon \sigma A(T_w^4-T_0^4)$$

(5)

A numerical model is proposed in this study to analyze combustion chemical effects. The combustion mechanism proposed by Mei et al. [42] was modified and validated. This mechanism incorporates the base H₂ mechanism from Hashemi et al. [43] and the NH₃ sub-mechanism from Shrestha et al. [44] and includes reactions of excited species, such as O₂(a1Δg) and O(1D), introduced by Konnov [45].

2.3. Validation of Laminar Burning Velocity Using CHEMKIN

To investigate the flame speed of NH₃/air and H₂/air mixtures, numerical simulations with different mechanisms were performed using CHEMKIN [46]. A premixed laminar flame speed reactor (PLFSR) calculation module was used in this work to predict laminar burning velocities (LBVs). All the simulations were converged to a grid-independent solution.

The experimental results reported by Mei et al. [42], Ronney [47], and Lhuillier et al. [48], as well as the simulation results reported by Han et al. [49], San Diego et al. [50], Mathieu and Petersen [51], and Li et al. [52], were utilized to validate the updated version of the NH₃/air combustion mechanism by correlating the LBVs of NH₃/air flame at various equivalence ratios as shown in Figure 1. This figure illustrates that, under lean, stoichiometric, and slightly rich (<1.2) conditions, the current mechanism demonstrated good agreement with the experimental results of Mei [42] and Ronney [47]. However, there are slight deviations on the rich side because of the low LBV values of NH₃/air mixtures.

The validation of the H₂/air combustion mechanism was achieved by establishing a correlation between the LBV experimental data reported by Burke et al. [53], Pareja et al. [54], and Taylor [55] as well as simulation results reported by Alekseev et al. [56] and Konnov et al. [45] at a temperature of 298 K and a pressure of 1 atm across a range of different equivalence ratios, as depicted in Figure 2. In this case, the predicted values using the chosen mechanism are in good agreement with experimental data as well.

The findings of these studies demonstrate that the ammonia/air mixture’s LBV peaks at Φ = 1.1–1.15, whereas the hydrogen/air mixture’s LBV peak value is approximately Φ = 1.50–1.70, 40 times greater than the ammonia/air mixture.

2.4. Computational Domain and Boundary Conditions

Two separate channels are used for the fueling of fuel and air. One channel is for pure H₂ or NH₃ and the other channel is for air. Both channels represent the inlet conditions with constant flow rates and an inlet temperature of 300 K. A pressure–outlet boundary condition of 1 atm is set at the outlet. The wall is modeled as a no-slip wall with heat transfer to the surroundings accounted for by adopting mixed thermal conditions that include a heat transfer coefficient of 20 W/m²K. The specific quartz glass specific heat capacity and thermal conductivity are 750 J/kgK and 2 W/mK [57]. A direct comparison between H₂ and NH₃ as fuels was conducted, setting a mixture velocity of 6 m/s for H₂ and different Φ from lean blow-out to 1. For NH_3, mass flow rates were computed based on a different Lower Heating Value (LHV), ensuring an equivalent heat of the reaction source for each fuel type at the same Φ.

Table 1. Velocity boundary conditions for H₂.

Heat of Reaction Source (W)	Φ	VH2 (m/s)	Vair (m/s)
15.2	0.5	1	5
17.6	0.6	1.16	4.83
19.9	0.7	1.31	4.69
22.1	0.8	1.46	4.54
24.1	0.9	1.59	4.41
26.4	1	1.72	4.28

and

Table 2. Inlet velocity magnitude boundary conditions for NH₃.

Heat of Reaction Source (W)	Φ	VNH3 (m/s)	Vair (m/s)
17.6	0.6	0.873	6.06
19.9	0.7	0.98	5.81
22.1	0.8	1.09	5.67
24.1	0.9	1.17	5.49
26.4	1	1.29	5.33

report the inlet velocity value for every considered case.

Mesh is realized using hexahedral cells, except for the confluence zone, where prismatic cells are used (Figure 3).

To ensure the grid independence, the simulation results were compared at v = 6 m/s, the H₂ case condition with three different grids. To reduce computational times without having any substantial effect on results, the inlet length has been reduced by 100 mm. Three different mesh resolutions, 140,000, 175,000, and 237,000 cells, were compared. The Grid Convergence Index (GCI) for three meshes has been computed [58]. This process allowed us to ascertain the discretization error associated with the numerical solution, improving the overall accuracy and reliability of the models. Refinement was applied in the mixing chamber with r = 1.5. It is known that the relative error for a function f is equal to

$$\mathsf{\epsilon }=\left|\frac{{\mathrm{f}}_2-{\mathrm{f}}_1 }{{\mathrm{f}}_2}\right|$$

(6)

$$\mathrm{P}=\ln (\frac{{\mathrm{f}}_3-{\mathrm{f}}_2}{{\mathrm{f}}_2-{\mathrm{f}}_1})/\ln (\mathrm{r})$$

(7)

$$GCI= \frac{f_s \epsilon }{r^P-1}$$

(8)

$$f^{*}=\frac{ r^P{f}_3-{ f}_2 }{r^P-1}$$

(9)

The safety factor was equal to 1.25. Temperature values are displayed in

Table 3. Temperature values at x = 9 mm, y = 0, and z = 0.785.

	Temperature (K)
Refined Mesh	2359.8
Middle Mesh	2349.9
Coarse Mesh	2328.4

at x = 9 mm, y = 0, and z = 0.785 for H₂-air combustion, velocity being equal to 6 m/s and Φ equal to 1.

GCI for 1/2 mesh was 0.2% while for 2/3 mesh, it was 0.5%. Moreover, the extrapolated f* value was computed with Equation (9) and the predicted value is equal to 2362 K. Thus, the middle mesh has only a 0.5% error, inside the 5% confidence. Figure 4 presents the temperature profiles along the centerline, emphasizing the consistency of our results across different mesh sizes and underscoring the accuracy of our grid independence verification.

2.5. Validation of 3D Model

The simulation results were compared to experimental data for an unburned mixture velocity of 6 m/s and various values for Φ ranging from 0.5 to 1. However, for Φ values lower than 0.6, the simulations did not exhibit a stable flame, and this lack of stability is consistent with the findings in the experimental data. There is no specific numerical parameter available to directly define the chemical flame length and facilitate a direct comparison with the experimental visible flame length. However, correlations between the flame and the mole fraction of OH distribution have been established. Nevertheless, in this study, a promising correlation is observed between the distribution of the heat of reactions and the flame length. By using a threshold of 1 W/m³, meaningful insights can be gained from the relationship between the heat of reactions and flame length. This correlation proves to be valuable in understanding the combustion process within the microcombustor. Figure 5 shows the trend of the heat of the reaction correlated with the flame length identified from the experimental data [41]. Despite some discrepancies, such as the data point at 0.9, the overall trend of the function can be considered in good agreement with the experimental data for the simulation.

3. Results

Considering that variations in Φ can lead to changes in inlet velocity, it is expected that these changes will significantly impact the circulation flow field and subsequently reduce temperature, ultimately affecting the formation of NO. Therefore, we assessed the impact of several critical parameters on NO formation by varying the fuel-oxidizer Φ within a range of 0.5 to 1.0 with an equivalent provided power volumetric flow rate. A direct comparison between H₂ and NH₃ as fuels was conducted following the boundary conditions displayed in Table 1 and Table 2. As anticipated, experimental results indicate that for H₂ fuel, a flame is present when the Φ is greater than 0.5. On the other hand, for NH₃ fuel, the minimum Φ required for flame ignition is 0.7. The curves for ammonia are shifted downstream in comparison to those for hydrogen (see Figure 6).

This is due to the lower laminar burning velocity of ammonia in comparison to H₂ and HC fuels, which makes it more difficult to sustain. This is due to the lower laminar burning velocity of ammonia in comparison to hydrogen and hydrocarbon fuels, which makes it more difficult to sustain a flame at lower equivalence ratios. Due to different Reynolds and fuel chemistry, ammonia flames move toward the outlet, becoming a flame at lower equivalence ratios. Due to different Reynolds and fuel chemistry, ammonia flames move toward the outlet. Figure 7 shows temperature centerlines for different equivalence ratio values at the same source of power of 26 W and 19.9. At Φ = 1, the maximum temperature of ammonia/air flames is about 200 K lower than that of H₂/air flames at the same Φ. The flame temperature is mostly affected by the specific heat of combustion and the flame speed. As the flame temperature and specific heat of combustion are quite higher for H₂, the flame temperature is higher. The variation in the Φ demonstrates distinct behaviors between H₂ and NH₃ in combustion scenarios. As the Φ varies, the peak temperature of H₂ exhibits a significant increase, while the peak temperature of NH₃ remains relatively stable. The effect of Φ on NH₃ is more pronounced in the extension of the flame region at temperatures exceeding 1800 K, as evident from Figure 7.

In the simulations of NH₃ combustion conducted, it was observed that higher equivalence ratios resulted in lower levels of nitrogen oxides (NO_x) (Figure 8).

This can be attributed to several factors inherent in the combustion dynamics of NH₃. Enhanced combustion efficiency in ammonia combustion processes at elevated equivalence ratios plays a pivotal role in mitigating NO_x emissions. This is primarily because higher equivalence ratios facilitate a more complete combustion of the fuel, predominantly yielding water and nitrogen. Consequently, this reduces the incidence of partial combustion, a process that typically leads to the formation of NO_x. Therefore, the modulation of equivalence ratios in ammonia combustion emerges as a critical parameter in curtailing the generation of nitrogen oxides, aligning with environmental directives and sustainability goals. On the other hand, one important NO production route from the combustion of nitrogen-free fuels in air, such as hydrogen, is the thermal-NO mechanism, usually referred to as the extended Zel’dovich mechanism [59], which is favored at high temperatures. So, the mole fraction of the NO value is two times that of the ammonia/air combustion (Figure 8b). The symmetry plane contours of the different species produced at ϕ = 0.9 are presented for the symmetry plane predicted in the case of the NH₃/air mixture. In Figure 9, the temperature symmetry plane field is provided.

NH₃ burning decreases both flame temperature (Figure 9a,b) and the concentration of the light radicals, as OH, relative to H_2, (Figure 10a,b) which negatively impacts the NH₃ flame reactivity, and thus flame burning velocity. Hence, the NH₃ flame is shifted forward compared to that of H₂. This information is reflected in both the NO concentration and the concentration of water vapor produced through combustion in both cases (see Figure 11a,b). A high concentration of NO is found to be associated with excess oxygen (O₂) atoms and OH radicals. This is the reason why NO concentrations are lower for NH₃ at higher equivalence ratios. OH primarily reacts with NH₃ through H abstraction. Other secondary consumption steps include reactions with H and O, with NH₂ being the common product. Oxidation of NHi (i = 0, 1, 2) may primarily lead to NO formation through an HNO intermediate or to NO reduction through NHi + NO reactions, depending on the concentration of O/H radicals. The abundance of O/H radicals leads to the conversion of NHi + O and may inhibit the reduction of NO by NHi radicals. Furthermore, the HNO intermediate channel is also a main NO production path in NH₃/air flames. HNO is mainly due to the reaction of NH₂ with O atoms. HNO is converted to NO mostly reacting with H, OH, and O₂, and through thermal dissociation.

Microcombustor Efficiency Evaluation

The efficiency of the system is calculated as the ratio of the enthalpy increase between the inlet and outlet to the total input thermal power, which encompasses the output chemical power from the complete combustion of fuel considered (see Equation (4)). Equation (11) is used to calculate the fuel conversion efficiency, denoted as ${\eta }_f$, which serves as an indicator of whether complete combustion has been achieved in the chamber wall.

$${\eta }_c=\frac{{\dot{m}}_{exit} {c_p T}_{exit}-{{\dot{m}}_{inlet} c_p T}_{inlet}}{{\dot{m}}_f{ LHV}_f}$$

(10)

$${\eta }_f=1-\frac{{ }_{{\dot{m}}_{f outlet} }}{{\dot{m}}_{f inlet}}$$

(11)

Table 4. Microcombustor predicted efficiencies and heat losses through the walls.

Φ	${\mathit{\eta }}_{\mathbf{f}\mathbf{N}{\mathit{H}}_3}$	${\mathit{\eta }}_{\mathbf{f}{\mathit{H}}_2}$	${\mathit{\eta }}_{\mathbf{c}\mathbf{N}{\mathit{H}}_3}$	${\mathit{\eta }}_{\mathbf{c}{\mathit{H}}_2}$	${\dot{\mathit{Q}}}_{\mathit{N}{\mathit{H}}_3}(\mathit{W})$	${\dot{\mathit{Q}}}_{{\mathit{H}}_2}(\mathbf{W})$
0.7	99.5%	99.9%	19.4%	9.6%	−16.5	−17.4
0.8	99.9%	99.9%	18.1%	9.3%	−18.5	−19.9
0.9	99.9%	99.9%	16.9%	8.8%	−20.1	−20.6
1	99.8%	99.5%	16.5%	8.4%	−23.2	−24.0

illustrates the combustion efficiency, revealing that the overall efficiency of the microcombustor is higher under NH₃ fueling conditions compared to hydrogen conditions. In all the configurations examined, whether using NH₃ or H₂ as fuels, we consistently observe nearly complete combustion. This is evident in the fuel conversion efficiency consistently exceeding 99%. A crucial factor contributing to this high efficiency is the system’s geometry.

This design promotes more effective mixing of the fuel with the oxidizer and ensures an ample residence time for the combustion process to reach completion.

NH₃ has a lower LHV, specifically 18.6 MJ for NH₃ compared to 190 MJ for H₂. So, the combustion efficiency is greater mainly because of the forward shift of the flame, which results in lower heat losses.

It is important to emphasize that in this study, the comparison was made at equal equivalence ratios and supplied power, but the mixture mass flow rates are different and significantly higher for NH₃, thus making the efficiency comparison useful but in need of a further in-depth analysis.

Higher equivalence ratios correspond to lower efficiencies due to significant chamber wall losses: in fact, even though the thermal power input increases, the wall losses grow more significantly, resulting in a decrease in combustion efficiency. Equation (5) results, as presented in Table 4, depict the heat losses through the walls for each case. The notably higher diffusivity of H₂ significantly amplifies the mixture’s thermal diffusivity, leading to a more rapid spread of heat and enhanced heat transfer to the combustion chamber walls. This phenomenon is clearly illustrated in Figure 7, where the temperature of H₂ decreases.

Additionally, an essential factor influencing the heat transfer process is the role of radiation. In the combustion of hydrogen, a larger amount of water vapor is produced compared to NH₃ combustion. This increase in water vapor results in heightened radiation, contributing to the observed greater wall losses in H₂ combustion.

4. Conclusions

A Y-shaped microcombustor designed for eco-friendly fuels is analyzed in this research.

Due to their distinct attributes, microcombustors exhibit different combustion dynamics for H₂ and NH₃ when compared to larger-scale combustion apparatuses.

Numerical simulations developed in an Ansys Fluent environment investigate the NO_x emission behaviors considering the impact of the flow field.

A 3D CFD computational framework incorporating intricate chemical–kinetic mechanisms is crafted and authenticated through juxtaposition with existing experimental findings from the literature.

The outcomes demonstrate the influence of the different fuels on the generated species, emissions, and the microcombustor efficiency.

Thermal performance alongside NO_x generation processes is assessed under scenarios of NH₂/air and H₂/air mixtures, with a steady flow velocity of 6 m/s and varying equivalence ratios. The NO_x formation in the microcombustor follows the mechanism of NO_x following the temperature distribution in agreement with the Zeldovich theory.

Specifically, with H₂ combustion, stable flames are attainable even at reduced equivalence ratios, leading to elevated flame temperatures and pollutant emissions.

The microcombustor-predicted efficiencies and heat losses through the walls are evaluated and discussed.

This research could represent an important tool in the microcombustor field, and thus in MEMS devices, which are a current research topic deeply investigated in these past few years.