A model and a prediction method for soot formation in NH3-hydrocarbon fuel blended flame based on nitridation mechanism

Abstract

The application discloses an NH3-carbon-hydrogen fuel blending flame soot generation model and a prediction method based on a nitridation mechanism, and belongs to the technical field of combustion pollution control and numerical simulation. The application constructs a surface nitridation reaction population balance model containing NH3, NH2 and HCN double reaction paths on the basis of a traditional HACA soot model, determines reaction rate constants through molecular dynamics and quantum chemistry calculation, and clearly defines the core hypothesis that inertization of nitrogen-containing functional groups permanently blocks HACA growth. The model is solved by using a shadow projection method and is coupled with an OPPDIF flame solver to realize accurate prediction of soot generation in an ammonia-doped flame. The application can accurately distinguish between chemical soot suppression and dilution effect of ammonia, significantly improves the prediction accuracy of soot volume fraction and particle size, reveals the double-phase soot suppression mechanism of the gas phase surface, and is suitable for soot emission prediction and low-emission design of ammonia-doped combustion equipment such as gas turbines, internal combustion engines and ship propulsion.

Description

Technical Field

[0001] This invention relates to the field of combustion pollution control and numerical simulation technology, specifically to a soot generation model and prediction method for NH3-hydrocarbon fuel blended flames based on the nitriding mechanism. It is applicable to combustion equipment such as gas turbines, internal combustion engines, and ship propulsion systems that use ammonia-hydrocarbon dual fuels as energy sources, and can accurately predict the soot generation characteristics under ammonia blending conditions. Background Technology

[0002] Reducing carbon and particulate emissions from combustion systems is a priority for the energy and transportation sectors. Fossil fuel combustion remains dominant in land and sea transportation, but the resulting carbon dioxide and soot (fine particulate matter, PM2.5) emissions are significantly higher. 2.5 The main components of carbon dioxide (NH3) have clear adverse effects on climate and human health. To reduce carbon emissions, zero-carbon fuels have received widespread attention. Among them, ammonia (NH3) is an important candidate fuel for gas turbines, internal combustion engines, and marine propulsion systems because it contains no carbon, can be synthesized on a large scale through the Haber-Bosch process, can be liquefied, stored, and transported under medium pressure, and can be used as a hydrogen carrier.

[0003] However, ammonia has significant drawbacks in its combustion characteristics: its laminar combustion rate is an order of magnitude lower than that of typical hydrocarbon fuels, its auto-ignition temperature is high, and its flammability limit range is narrow, resulting in poor flame stability, slow ignition, and high unburned fuel emissions in actual burners. Furthermore, the nitrogen content of ammonia makes it prone to generating nitrogen oxides (NOx) during combustion. x To address these issues, researchers have proposed blending ammonia with more reactive hydrocarbon fuels such as methane, propane, and ethylene. This dual-fuel strategy can improve ignition and flame stability while reducing net carbon emissions.

[0004] Further research confirmed that adding ammonia can effectively suppress soot formation in hydrocarbon fuel flames, and the suppression effect far exceeds that of simple dilution or cooling effects: for example, adding 3-4% (volume ratio) NH3 to a propane flame can drastically reduce the soot mass concentration and average particle size; the soot suppression effect of ammonia in a methane co-flow flame is due to chemical effects rather than thermal or dilution effects; in an ethylene counter-diffusion flame, adding 1% NH3 reduces the soot volume fraction by 4-6%; a reduction in polycyclic aromatic hydrocarbons (PAHs, soot precursors) and soot was also observed in n-heptane and ethylene flames, and the nitrogen content on the surface of soot generated by ammonia-doped flames increased, with the presence of cyano or amino functional groups. These phenomena indicate that ammonia not only changes the flame temperature but also alters the chemical pathways of soot precursors and the surface processes of soot particles.

[0005] From a mechanistic perspective, species such as NH2, NCO, CN, and HCN generated from the pyrolysis of NH3 in the gas phase scavenge key free radicals such as H, CH3, C2H2, and C3H3 required for the formation of benzene (Al) and larger PAHs, leading to a decrease in PAH precursor concentration and a slowdown in the hydrogen abstraction-C2H2-addition (HACA) pathway, thereby reducing soot nucleation. Superficially, nitrogen-containing species can directly react with active sites at the edge of aromatic clusters to form inert nitrogen-containing functional groups, occupying these sites to block the HACA growth pathway and even introducing competitive reaction channels. Abinitio and RRKM–ME analyses in existing technologies further confirm that NH3 and NH2 in the flame are more competitive than HCN in acetylene addition at the aromatic edge; they temporarily block active sites and alter the growth pathway, and compared to cyano functional groups, more amino functional groups are formed at the edges of polycyclic aromatic hydrocarbons.

[0006] However, existing technologies have significant shortcomings: First, the kinetic characterization of the reaction between ammonia and soot surfaces is insufficient, and related surface nitriding reactions are rarely explicitly incorporated into soot models; second, although studies have investigated the chemical, thermal, and dilution effects of ammonia separation (e.g., chemical effects dominate smoke suppression at low NH3 ratios, while dilution effects are more important at high ratios), even considering the gas phase mechanism, numerical predictions still systematically underestimate the degree of soot and PAH suppression, indicating that the role of nitrogen species surface chemistry is being neglected; third, existing soot simulation methods lag behind experimental research: for example, some studies combine the gas-phase mechanism and nitrogen chemical mechanism of C2 to naphthalene with standard soot models, which can only capture dilution and temperature effects, and cannot accurately reproduce the measured distribution of C2H2, benzene, or soot volume fractions; although some studies have attempted to improve the coupling between carbon-nitrogen chemistry and soot generation, they still rely on the traditional Frenklach model. The HACA growth and oxidation scheme only adds surface reaction steps sporadically (such as the single reaction of HCN forming cyano groups with surface sites), does not explicitly consider NH3 and NH2, and does not distinguish between different pathways of HCN and soot surface reactions; fourth, existing models mostly treat soot as pure carbonaceous material and do not explicitly assume that "nitrogen-containing functional groups do not participate in subsequent HACA reactions during the entire formation process of soot", which makes it impossible to accurately quantify the blocking effect of surface nitriding reactions on the HACA pathway.

[0007] In summary, the current lack of a comprehensive soot model that includes complete NH3 / NH2 / HCN surface chemistry (especially the HCN dual reaction pathway) and is based on physics makes it impossible to accurately predict the soot generation characteristics of ammonia-infused flames and to fully elucidate the smoke suppression mechanism of ammonia, thus restricting the research and development and application of ammonia fuel combustion systems. Summary of the Invention

[0008] I. Technical problems to be solved To address the shortcomings of existing technologies, such as the lack of a complete soot model incorporating the surface chemistry of NH3 / NH2 / HCN (including the dual reaction pathway of HCN), low accuracy in soot formation prediction, and insufficient elucidation of the ammonia smoke suppression mechanism, this invention aims to solve the following technical problems: (i) Construct a detailed soot population balance model that includes the reactions of NH3, NH2, HCN (including dual reaction pathways) and soot active sites to make up for the shortcomings of existing models that ignore the surface chemistry of nitrogen species and the differences in HCN pathways; (ii) The rate constant of the surface nitriding reaction (including the HCN dual pathway) was determined by molecular dynamics simulation, clarifying the assumption that "nitrogen-containing functional groups no longer participate in the subsequent HACA reaction", providing a solid physical basis for the model and avoiding the limitations of traditional models that rely on empirical parameters; (iii) The soot model is solved by the moment projection method and coupled with the OPPDIF flame solver to achieve accurate prediction of soot generation in ethylene / NH3 diffusion flame; (iv) To reveal the two-phase (gas phase-surface) inhibition mechanism of ammonia on soot generation, clarify the competitive differences between NH3, NH2 and HCN and the contribution of gas phase and surface interaction, and provide a theoretical basis for the design of ammonia-blended combustion systems.

[0009] II. Technical Solution This invention achieves accurate prediction of soot generation in ammonia-doped flames by developing a surface-reactive population balance model and combining precise kinetic parameters (including HCN dual-path parameters) with advanced numerical solution methods. The specific technical solution is as follows: (I) Model Construction: Soot Population Balance Model Including Surface Nitriding Reaction (HCN Dual Pathway) Based on the traditional soot model using the hydrogen abstraction-acetylene addition (HACA) mechanism, this model extends the model by incorporating detailed surface nitridation reactions (including a dual HCN reaction pathway) of active soot sites and nitrogen-containing species (NH3, NH2, HCN). The model assumes all soot particles are spheres, with the carbon number of each individual particle as the internal coordinate. It neglects the inertial forces of soot particles (because the diameter of soot particles in a flame is typically several hundred nanometers, resulting in a small Stokes number), and assumes they move along streamlines. The soot particle swarm equilibrium equation is as follows:

[0010] In the formula The particle number density of soot. This refers to the number of carbon atoms in a single soot particle. For flame density, For flow velocity, The height of the flame. For hot swimming speed, The diffusion coefficient is... For the source terms of soot particles, including nucleation ( ),collision( ), condensation ( ), surface growth ( ), oxidation ( )process, The formula is as follows:

[0011] The specific definitions and calculation methods for each process in the formula are as follows: (1) Nucleation process ( The process involves converting gaseous PAHs into solid initial soot particles, using pyrene (A4) as a gaseous precursor for the soot particles, resulting in a minimum particle size of [missing value]. The source term expression for carbon soot particles is: ( For particle nucleation rate, (Pyrene concentration); (2) Collision process ( The collision and bonding between soot particles form larger aggregates, which are described by the Smoluchowski equation. The source term expression is: ( For size and (carbon soot particle collision efficiency). (3) Condensation process ( ): Pyrene molecules collide and condense on the surface of soot particles, leading to particle size growth. The source term expression is: , The increase in carbon atoms in soot particles during condensation is due to the use of pyrene as the condensate; a single pyrene molecule contains 16 carbon atoms. ; (4) Surface growth process ( ): It includes traditional HACA mechanism reactions and newly added surface nitriding reactions (including HCN dual pathway), and explicitly assumes that "nitrogen-containing functional groups on the soot surface no longer participate in subsequent HACA reactions during the entire formation process of soot".

[0012] 1. Traditional HACA mechanism response: Active sites: Based on the research of Appel et al., two types of active sites exist on the surface of soot. and .

[0013] ： Hydrogen atom sites (saturation sites) bonded to carbon atoms participate in dehydrogenation reactions during soot surface growth. The dehydrogenation reaction is as follows: (The rate constant is shown in S1 in Table 1). This reaction is caused by H atoms attacking and removing hydrogen molecules from saturated sites to generate active sites. The dehydrogenation reaction with OH and the rate constant are shown in Table 1, S2.

[0014] The dehydrogenation active sites (radical sites) on carbon atoms can participate in addition reactions during the growth process on the soot surface. The addition reaction is (The rate constant is shown in S4 in Table 1). In this reaction, the active site reacts with acetylene, increasing the carbon skeleton (the number of carbon atoms increases by 2) and regenerating a saturation site. The termination reaction equation and rate constant with H are shown in Table 1, S3.

[0015] 2. Surface nitriding reaction (including HCN dual pathway) Based on the traditional HACA mechanism, the surface nitridation reaction of nitrogen-containing species (NH3, NH2, HCN) with soot is introduced, and it is explicitly hypothesized that "nitrogen-containing functional groups do not participate in subsequent HACA reactions during soot formation".

[0016] ①Reaction pathway: active site It can undergo surface nitriding reactions with a variety of nitrogen-containing species, as shown in the following reaction formulas: (a) Reaction with HCN (dual pathway): Path 1: (Generates cyano functional group).

[0017] Path 2: ,

[0018] This represents HCN and model compounds (such as 4- Different reaction pathways were used to determine the rate constants of the carbon soot surface reaction, and their kinetic parameters were used to determine the rate constants of the reaction. (See Table 1)

[0019] (b) Reaction with NH2: (Generates amino functional groups).

[0020] (c) Reaction with NH3: (Generates amino functional groups).

[0021] ②Key assumption: All newly generated nitrogen-containing functional groups (such as...) , The ions are assumed to be inert and no longer participate in subsequent HACA growth, thus permanently occupying the active sites and inhibiting soot growth.

[0022] ③ Kinetic parameters: The rate constants of these surface nitriding reactions (such as S8-S11 in Table 1) were determined by molecular dynamics simulations, referencing the reaction data of model compounds (such as 4-phenanthryl) and their corresponding nitrogen-containing species.

[0023] 3. Expression for surface growth source term The net effect of this process is described by the source term in the population balance equation, expressed as:

[0024] In the formula Let be the surface growth reaction rate constant. For containing The surface area of ​​a carbon soot particle per carbon atom The number of carbon atoms added in a single reaction during surface growth (for acetylene addition, =2; for some surface nitriding reactions, it depends on the specific reaction formula).

[0025] (5) Oxidation process ( In contrast to the surface growth process, the oxidation process removes carbon atoms through the surface reaction of soot particles with gaseous substances.

[0026] 1. Oxidizing agents: The model simultaneously considers carbon soot oxidation initiated by molecular oxygen (O2) and hydroxyl radicals (OH).

[0027] 2. Source Term Expression: The source term of this process is explicitly defined by the following formula:

[0028] In the formula: Let be the rate constant of the oxidation reaction. It contains The surface area of ​​a single carbon atom in a single particle. It is the number density of carbon soot particles with a carbon atom number of . This represents the number of carbon atoms that are removed from the particle during each oxidation reaction.

[0029] 3. Reaction details: When O2 oxidation occurs, two carbon atoms are released from the soot particles in each reaction, i.e. =2; When OH oxidation occurs, one carbon atom is removed from the soot particle with each reaction, i.e. =1. The specific reaction formulas and rate constants corresponding to the oxidation process are shown in S5 and S6 in Table 1.

[0030] (II) Key reactions, rate constants, and their determination methods based on molecular dynamics simulations This invention, building upon the traditional HACA-based soot model, innovatively incorporates detailed surface nitridation reactions between active soot sites and key nitrogen-containing species (NH3, NH2, HCN). Specifically, the surface nitridation reaction between HCN and active sites explicitly includes two pathways. Table 1 lists all reactions in the model and their rate constants (following the Arrhenius equation). The kinetic parameters of the newly added surface nitriding reactions (S8-S11 in Table 1a) are not based on empirical estimates, but are determined through molecular dynamics simulations (MD) or quantum chemical calculations, thus laying a solid physical foundation for the model and avoiding the limitation of poor extrapolation ability of traditional model parameters.

[0031] Table 1a. Reaction rate constants for carbon soot growth reaction

[0032] Note: S8-S11 in the table are newly added surface nitriding reactions, of which S9 and S10 are two different pathways for the reaction between HCN and active sites on the surface of soot, corresponding to different pre-exponential factors (A), temperature exponents (n) and activation energies (E); S1-S7 are traditional HACA mechanisms and PAH condensation-related reactions.

[0033] Table 1b: Surface reaction rate of carbon soot

[0034] The specific basis for determining the newly added reaction kinetic parameters in Table 1a is as follows: Reaction S8 (NH2 reaction): Parameters were determined based on molecular dynamics simulations by Wang et al. Simulation results confirmed that NH2 exhibits stronger reactivity with active sites on the carbon soot surface than HCN.

[0035] Reaction S9 and S10 (HCN dual pathway): The parameters were obtained through quantum chemical calculations, corresponding to two pathways in which HCN reacts with model compounds (such as 4-phenanthryl) to generate cyano-containing functional groups with different structures. Pathway 1 (S9) has a relatively low activation energy, while path 2 (S10) has a higher activation energy, reflecting the differences in the reaction pathways.

[0036] Reaction S11 (NH3 reaction): Its kinetic parameters were also determined based on molecular dynamics simulations, and the results showed that NH3 was more competitive than HCN.

[0037] Key Assumption: This model explicitly assumes that all newly generated nitrogen-containing functional groups (such as -CN, -NH2) through the above reactions are considered inert throughout the entire soot formation process and do not participate in subsequent HACA reactions. This core assumption is supported by molecular simulation results and is key to the model's ability to accurately quantify the "permanent" blocking effect of surface nitriding on the HACA pathway, thus solving the problem of existing models systematically underestimating the degree of soot suppression.

[0038] Table 1b presents the calculation forms of surface reaction rates corresponding to each elementary reaction on the soot surface. This is used to unify the numerical calculation rules of surface reactions in the model. Together with the kinetic parameters in Table 1a, it can achieve accurate quantification of the growth, nitriding, and oxidation processes on the soot surface. Among them: S1-S3 are the transformation reactions of active sites on the surface of soot, which only change the number of active sites and do not contribute to the surface rate of soot, so they are recorded as 0; S4 is the HACA surface growth reaction of acetylene addition, S5 is the molecular oxygen oxidation reaction, and S8 is the surface nitridation reaction of NH2 with active sites of soot. All of them are calculated by multiplying the concentration of active sites, the concentration of gaseous reactants, and the particle surface area; S6 is the hydroxyl radical oxidation reaction and S7 is the PAH condensation reaction, which do not depend on active sites and are modeled as collision processes; S9-S1 are the elementary reaction numbers in Table 1a, and their surface rate calculations are respectively classified into S12-S14, and are recorded as 0 in this table; S12 and S13 correspond to the dual pathways of HCN and active sites of soot, respectively, and S14 corresponds to the surface nitridation reaction of NH3 and active sites of soot. All of them are calculated independently by multiplying the concentration of active sites, the concentration of gaseous reactants, and the particle surface area to distinguish the reaction rate differences of different nitrogen-containing species and reaction pathways, and to provide a unified surface rate input for the numerical solution of the moment projection method.

[0039] (iv) Numerical solution: Coupling of the moment projection method with the OPPDIF flame solver To achieve efficient solutions for the soot particle swarm equilibrium model, this invention employs the Moment Projection Method (MPM) as the core numerical algorithm and couples it with the OPPDIF offset diffusion flame solver to uniformly handle the source term contributions of traditional HACA reactions and surface nitriding reactions. This method avoids directly solving complex integral-differential equations by solving the moments of the soot number density function, significantly improving computational efficiency while maintaining accuracy. The specific implementation steps are as follows: (1) Definition and equation establishment of moment The method of moments (MPM) is used to solve the soot population balance equation. First, the soot number density function is defined. Step moment ( (Lower-order moments have explicit physical meaning:) This represents the total number of carbon soot particles. (For the total mass of soot), the population balance equation is transformed into a transport equation in moment form:

[0040] (2) Equation closure: Weighted particle method To close the nonlinear terms in the transport equations of the moment form described above, MPM introduces the concept of weighted particles. This method uses a finite number of weighted particles to approximate a continuous particle distribution, and its core formula is: , For the weighted particle coordinates, For the corresponding weights, (for weighted particle count), and fixed At the smallest particle size, The carbon soot consumption rate is approximated by the coordinates of the first weighted particle. Fixed at the smallest particle size, its weight It can effectively characterize the particle flux loss caused by consumption mechanisms such as oxidation, thereby achieving equation closure.

[0041] (3) Source term closing expression Based on the weighted particle method, the moment source terms of all physical processes contributing to soot evolution in the model are closed. These processes include, but are not limited to: 1. Nucleation source term (moment form): ; 2. Collision source term (moment form): ; 3. Condensation source term (moment form):

[0042] 4. Surface growth source term (moment form, including contributions from C2H2 and nitrogen species (HCN dual pathway)):

[0043]

[0044]

[0045]

[0046]

[0047] In the above surface growth source term, the surface reaction rate constant Each reaction is assigned an independent value, calculated separately according to the Arrhenius equation for the corresponding elementary reaction in Table 1a. Different reactions correspond to different... Numerical value.

[0048] 5. Oxidation source term (moment form)

[0049]

[0050] All source terms share a unified mathematical form, reflecting the consistency of the weighted particle method in handling different physical processes. Ultimately, the total moment source terms... This is the sum of the source terms of each process.

[0051] (4) Algorithm implementation and verification The closed moment transport equations described above are coupled with the OPPDIF flame solver to form a complete simulation framework. This framework can self-consistently solve the turbulent combustion field and soot evolution, and has been successfully applied to the simulation of ethylene / NH3 counter-diffusion flames. Comparison with experimental data verifies the significant accuracy of this numerical algorithm in predicting soot volume fraction and particle size. Results show that the model can successfully reproduce the experimentally observed trend of "increasing NH3 addition leading to a decrease in soot volume fraction and average particle size," and due to the inclusion of the HCN dual reaction pathway and the highly competitive reaction of NH3 / NH2, the predicted results show significantly better agreement with experimental data than traditional models.

[0052] (III) Beneficial Effects Compared with the prior art, the present invention has the following significant advantages: Significantly improved prediction accuracy: By introducing a complete surface nitriding reaction (including the HCN dual pathway) and the explicit assumption that "nitrogen-containing functional groups do not participate in the subsequent HACA reaction", the model makes up for the shortcomings of existing models that ignore the differences in surface chemistry and pathways of nitrogen species. Under the NH3 mixing condition, the model's predicted soot volume fraction and average particle size are in much better agreement with the experimental data than traditional models, and can accurately reflect the differentiated effects of NH3, NH2, and HCN on soot generation. The physical foundation is solid and reliable: the rate constant of the newly added surface nitriding reaction (including the HCN dual pathway) is determined based on molecular dynamics simulations rather than relying on empirical parameters, and the physical properties of NH3 and NH2 being more competitive than HCN are clearly defined, avoiding the problem of poor extrapolation ability of traditional models. The model is more applicable under different ammonia doping ratios and different hydrocarbon fuel conditions. The mechanism is clearly and comprehensively explained: within a unified model framework, the dual-phase inhibition mechanism of ammonia on soot formation is revealed and quantified—the formation of aromatic hydrocarbon precursors (such as A1 and A4) is reduced in the gas phase, while NH3, NH2, and HCN (dual pathways) compete with soot active sites in the surface phase to block the HACA growth pathway. The competitive differences of different nitrogen-containing species and the contribution of gas phase and surface interaction are clarified, providing a clear theoretical basis for the optimization of ammonia-blended combustion systems. High engineering application value: The model is coupled with the OPPDIF flame solver and fully presents the key reaction parameters (Table 1). It can be directly applied to the carbon soot prediction of ammonia-blended combustion equipment such as gas turbines, internal combustion engines, and ship propulsion systems, providing a precise tool for the design and operation parameter optimization of low-emission combustion systems and promoting the industrial application of ammonia fuels in the energy and transportation fields. Attached Figure Description

[0053] Figure 1 This is a flowchart of the algorithm of the present invention.

[0054] Figure 2 Comparison of experimentally measured temperature values ​​along the flame centerline with predicted values ​​from the model of this invention. Figure 3 Comparison of SVF values ​​between the model predictions and experimental measurements under different ammonia blending ratios. Figure 4 Comparison of SVF values ​​predicted by traditional models and experimental measurements under different ammonia blending ratios. Figure 5 Comparison of peak soot volume fraction between the model of this invention and the traditional model Figure 6 The model of this invention differs from the traditional model in terms of average diameter D. 63 Prediction Detailed Implementation The technical solution of the present invention will be further described in detail below with reference to specific experimental cases and simulation verification. This embodiment is only used to explain the present invention and does not constitute a limitation on the scope of protection of the present invention.

[0055] (1) Setting up experimental and simulation conditions Verification experiments were conducted using a counter-diffusion flame apparatus, which consists of two vertically opposed nozzles, each with an inner diameter of 10 mm and a distance of 8 mm between them. The oxidant was supplied from the upper nozzle, and the fuel was introduced from the lower nozzle. The gas-phase chemical reaction employed the NPAC mechanism, which has been calibrated in a C2H4 / NH3 diffusion flame and can accurately predict the concentration distribution of polycyclic aromatic hydrocarbons (PAHs).

[0056] The experimental and simulation conditions are set as follows: 1. Pure ethylene fuel operating condition: ethylene molar fraction is 1, NH3 molar fraction is 0, N2 molar fraction is 0, fuel nozzle outlet velocity is 17.4 cm / s; oxidant side O2 molar fraction is 0.25, N2 molar fraction is 0.75, oxidant nozzle outlet velocity is 23.4 cm / s.

[0057] 2. Ammonia blending condition: The ethylene molar fraction is set to 0.95, 0.9, 0.85, and 0.8 respectively, and the corresponding NH3 molar fraction is 0.05, 0.1, 0.15, and 0.2 respectively. The N2 molar fraction is 0. The fuel and oxidant nozzle outlet velocities are the same as those for the pure ethylene condition.

[0058] 3. Nitrogen-doped control condition: ethylene molar fraction 0.8, N2 molar fraction 0.2, NH3 molar fraction 0, and other parameters unchanged.

[0059] The soot population balance model with surface nitriding reaction (HCN dual path) constructed in this invention is coupled with the moment projection method (MPM) and the OPPDIF offset diffusion flame solver to build a complete simulation framework and simultaneously solve the turbulent combustion field and soot evolution process.

[0060] (2) Model Solving and Verification Process 1. Based on the soot population balance model of this invention, the rate constants of the traditional HACA reaction and the newly added surface nitriding reaction (S8-S11, including HCN dual pathways S9 and S10) in Table 1a are input, following the Arrhenius equation. Calculate the reaction rate.

[0061] 2. Define the first... Step Moment The population balance equation is transformed into a moment transport equation. The weighted particle method is used to close the nonlinear terms of the equation. The first weighted particle is fixed at the minimum particle size to close the source terms of each process, including nucleation, collision, condensation, surface growth and oxidation.

[0062] 3. Couple the OPPDIF flame solver to perform ethylene / NH3 counter-diffusion flame simulation, outputting flame temperature, soot volume fraction (SVF), and average soot particle size (D). 63 )data.

[0063] 4. Compare the simulation results of this invention with the experimental measurement data of traditional models to verify the prediction accuracy and explanatory ability of the ammonia smoke suppression mechanism of this invention.

[0064] (3) Simulation results and creative embodiment 3.1 Temperature Field Prediction like Figure 2 As shown, the model of this invention can accurately capture the peak position and trend of flame centerline temperature under a 20% nitrogen / ammonia mixing ratio. The temperature curves of the ammonia mixing condition and the pure nitrogen dilution condition are very similar, indicating that the model can reasonably distinguish the dilution effect, chemical effect and thermal effect of ammonia, which is consistent with the experimental conclusions of the prior art, and provides an accurate temperature field basis for predicting soot generation.

[0065] 3.2 Prediction of Soot Volume Fraction (SVF) like Figure 3 and Figure 4 As shown, the following conclusions can be drawn: Experiments show that when 20% NH3 is added, the peak volume fraction of soot is reduced by 87.16%, and the higher the amount of ammonia added, the more significant the soot suppression effect.

[0066] The model of this invention, as shown in Table 2, can accurately reproduce the trend of "increased NH3 addition → decreased carbon soot volume fraction", and the predicted values ​​are in high agreement with the experimental values. The relative error of the prediction is 7.69% when the ammonia ratio is 0%, and the error drops to 1.05% when the ammonia ratio is 20%. The higher the ammonia ratio, the higher the prediction accuracy.

[0067] Table 2 shows the deviation between the experimentally measured peak value and the peak value predicted by the model of this invention.

[0068] Traditional models can only capture the decreasing trend of carbon soot, systematically underestimate the degree of carbon soot suppression, and the predicted values ​​deviate significantly from the experimental values.

[0069] like Figure 5 As shown, this invention incorporates the surface nitridation reactions of NH3, NH2, and HCN (dual pathways) with soot active sites and clarifies the assumption that "nitrogen-containing functional groups do not participate in subsequent HACA reactions," thus overcoming the shortcomings of traditional models that neglect the surface chemistry of nitrogen species and cannot quantify the HACA pathway blocking effect. The prediction accuracy is far superior to existing technologies.

[0070] 3.3 Average particle size of carbon soot (D) 63 )predict like Figure 6 As shown, the following conclusions can be drawn: Experiments show that the average particle size of soot reaches its peak at 2.65 mm from the fuel nozzle, and the peak particle size of the mixture with 20% NH3 is much lower than that of the mixture with 20% N2, indicating that ammonia has a chemical inhibitory effect.

[0071] The model of this invention accurately captures the particle size distribution trend, with the peak value basically consistent with the experimental value, and can distinguish between the chemical inhibition of ammonia and the pure dilution effect of nitrogen.

[0072] Traditional model: Without considering surface nitriding reaction, the predicted particle size is too high with 20% NH3 blending, similar to the predicted level with 20% N2 blending, and cannot reflect the chemical smoke suppression effect of ammonia.

[0073] This invention precisely quantifies the blocking effect of ammonia on the growth of soot on the surface by using the dual reaction pathway of HCN and the highly competitive reaction parameters of NH3 and NH2, and clarifies the gas-phase-surface two-phase smoke suppression mechanism. This is a creative technical effect that traditional soot models do not possess.

[0074] (4) Creative summary This specific embodiment verifies the core inventiveness of the present invention compared to the prior art: Construct a dual pathway including NH3, NH2, and HCN. ， A complete carbon soot population balance model for surface nitriding reactions is proposed to address the deficiency in existing models that lack surface chemistry of nitrogen species.

[0075] The newly added reaction rate constant is determined based on molecular dynamics and quantum chemical calculations, abandoning empirical parameters and providing a more solid physical foundation.

[0076] The core assumption of "inertization of nitrogen-containing functional groups and permanent blockage of HACA growth" was clearly defined, and the carbon soot inhibition effect was accurately quantified, with prediction errors far lower than those of traditional models.

[0077] The coupled moment projection method and OPPDIF solver can efficiently achieve accurate prediction of soot generation in ammonia-doped flames, and can be directly applied to the low-emission design of equipment such as gas turbines and internal combustion engines.

[0078] The above embodiments are merely preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the technical principles and core innovation framework of the present invention should be included within the protection scope of the present invention.

Claims

1. A model and prediction method for soot formation in NH3-hydrogen fuel blending flames based on nitriding mechanism, characterized in that, Includes the following steps: A soot population balance model was constructed, which, based on the traditional HACA-based soot surface growth response mechanism, expanded to include nitrogen-containing species NH3, NH2, and... With active sites on the surface of soot Detailed surface nitriding reaction, in which The reaction with the soot surface involves two distinct pathways; the model explicitly assumes that the nitrogen-containing functional groups generated on the soot surface through the surface nitriding reaction are chemically inert throughout the entire soot formation process and no longer participate in the subsequent HACA reaction, thereby permanently blocking the soot surface growth pathway. The general form of the soot population balance model is described by the particle swarm balance equation: The source item Includes surface growth process source terms contributed by the surface nitriding reaction. Its expression is: The This is the overall rate constant for surface growth reactions, including the surface nitriding reaction. Based on molecular dynamics simulations or quantum chemical calculations, the kinetic parameters of the surface nitriding reaction are determined; the soot population balance model is solved using the moment projection method, specifically including: defining the first... Step Moment The particle swarm equilibrium equation is transformed into a moment-of-moment (MOM) transport equation, and the MOM transport equation is solved in a closed loop using the weighted particle method. The solved MOM transport equation is then coupled with the OPPDIF offset diffusion flame solver to simulate and predict the soot generation characteristics in ammonia-doped flames.

2. The prediction method according to claim 1, characterized in that, The surface nitriding reaction specifically includes the following elementary reactions: Among them, reactions S9 and S10 are Two different pathways for reacting with the surface of carbon soot.

3. The prediction method according to claim 2, characterized in that, The phrase "nitrogen-containing functional groups are chemically inert" specifically refers to the formation of nitrogen-containing functional groups from reactions S8 and S11. Functional groups, and those generated by reactions S9 and S10 The functional group is configured in the model not to participate in any subsequent hydrogen abstraction or acetylene addition reactions.

4. The prediction method according to claim 2, characterized in that, The determination of the kinetic parameters specifically refers to the pre-exponential factors of reactions S8 and S11. Temperature index and activation energy The pre-exponential factors of reactions S9 and S10 were determined based on molecular dynamics simulations. Temperature index and activation energy Determined based on quantum chemical calculations.

5. The prediction method according to claim 2 or 4, characterized in that, The rate constant of the surface nitriding reaction Following the Arrhenius formula Among them, reaction S8 , , ; reaction S9 , , ; Reaction S10 , , ; Reaction S11 , , .

6. The prediction method according to claim 1, characterized in that, The moment projection method is used for solving the problem, and the expression for the surface growth moment source term, which includes the contribution of the surface nitriding reaction, after being closed by the weighted particle method, is as follows: Where x represents different surface growth reactions, including the acetylene addition reaction of conventional HACA and reactions S8, S9, S10, and S11 as described in claim 2. The surface reaction rate constant for the corresponding reaction is... This represents the increment of carbon atoms produced in the first reaction.

7. The prediction method according to claim 1, characterized in that, The source terms of the soot population balance model It also includes nucleation process source terms Collision process source term Condensation process source items and oxidation process source items .

8. A predictive system for soot formation in NH3-hydrocarbon fuel blending flames based on nitriding mechanism, characterized in that, include: A model building module for building a soot population balance model as defined in any one of claims 1–7; The parameter determination module is used to determine the kinetic parameters of the surface nitriding reaction based on molecular dynamics simulations or quantum chemical calculations. The solution and prediction module is used to solve the soot population balance model using the moment projection method and couple it with the OPPDIF offset diffusion flame solver to output the prediction results of soot generation characteristics.

9. The prediction system according to claim 8, characterized in that, The solution and prediction module is specifically configured to calculate and couple the closed moment transport equation as described in claim 6.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the method for predicting soot generation in NH3-hydrocarbon fuel blending flames based on the nitriding mechanism as described in any one of claims 1–7.

Images