Evolution of soot size distribution in premixed ethylene/air and ethylene/benzene/air flames: Experimental and modeling study

The effect of benzene concentration in the initial fuel on the evolution of soot size distribution in ethylene/air and ethylene/benzene/air flat flames was characterized by experimental measurements and model predictions of size and number concentration within the flames. Experimentally, a scanning mobility particle sizer was used to allow spatially resolved and online measurements of particle concentration and sizes in the nanometer-size range. The model couples a detailed kinetic scheme with a discrete-sectional approach to follow the transition from gas-phase to nascent particles and their coagulation to larger soot particles. The evolution of soot size distribution (experimental and modeled) in pure ethylene and ethylene flames doped with benzene showed a typical nucleation-sized (since particles do not actually nucleate in the classical sense particle inception is often used in place of nucleation) mode close to the burner surface, and a bimodal behavior at greater height above burner (HAB). However, major features were distinguished between the data sets. The growth of nucleation and agglomeration-sized particles was faster for ethylene/benzene/air flames, evidenced by the earlier presence of bimodality in these flames. The most significant changes in size distribution were attributed to an increase in benzene concentration in the initial fuel. However, these changes were more evident for high temperature flames. In agreement with the experimental data, the model also predicted the decrease of nucleation-sized particles in the postflame region for ethylene flames doped with benzene. This behavior was associated with the decrease of soot precursors after the main oxidation zone of the flames.     

A model of particle nucleation in premixed ethylene flames

A detailed model of particle inception is proposed to delve into the physical structure and chemistry of combustion-formed particles. A sectional method is used, from a previously developed kinetic mechanism of particle formation with a double discretization of the particle phase in terms of C and H atom number. The present model also distinguishes between different particle structures based on their state of aggregation; single high molecular mass molecules, cluster of molecules and aggregates of clusters. The model predicts the mass of particles, hydrogen content and internal structure. It represents a first approach in following the chemical evolution and internal structure of the particles formed in flames, coupled with the main pyrolysis and oxidation of the fuel.The model is tested in atmospheric premixed flat flames of ethylene and the effect of fuel equivalence ratio on particle morphology is analyzed. Molecular weight growth of aromatic compounds and the inception of particles are predicted. The morphology of the particles and the number of molecules in the clusters at particle inception are also indicated.     

A stochastic approach to calculate the particle size distribution function of soot particles in laminar premixed flames

We introduce an efficient stochastic approach to solve the population balance equation that describes the formation and oxidation of soot particles in a laminar premixed flame. The approach is based on a stochastic particle system representing the ensemble of soot particles. The processes contributing to the formation and oxidation of soot particles are treated in a probabilistic manner. The stochastic algorithm, which makes use of an efficient majorant kernel and the method of fictitious jumps, resolves the entire soot particle distribution (PSDF) without introducing additional closure assumptions. A fuel-rich laminar premixed acetylene flame is computed using a detailed kinetic soot model. Solutions are obtained for both, the stochastic approach and the method of moments combined with a modified version of the Premix, CHEMKIN code. In this manner, the accuracy of the method of moments in a laminar premixed flame simulation is investigated. It is found that the accuracy for the first moment is excellent (5% error), and mean error for rest of the moments is within 25%. Also the effect of the oxidation of the smallest particles (burnout) has been quantified but was found not to be important in the flame investigated. The time evolution of computed size distributions and their integral properties are compared to experimental measurements and the agreement was found to be satisfactory. Finally, the efficiency of the stochastic method is studied.     

Quantitative measurement of soot particle size distribution in premixed flames - The burner-stabilized stagnation flame approach

A burner-stabilized, stagnation flame technique is introduced. In this technique, a previously developed sampling probe is combined with a water-cooled circular plate such that the combination simultaneously acts as a flow stagnation surface and soot sample probe for mobility particle sizing. The technique allows for a rigorous definition of the boundary conditions of the flame with probe intrusion and enables less ambiguous comparison between experiment and model. Tests on a 16.3% ethylene-23.7% oxygen-argon flame at atmospheric pressure show that with the boundary temperatures of the burner and stagnation surfaces accurately determined, the entire temperature field may be reproduced by pseudo one-dimensional stagnation reacting flow simulation using these temperature values as the input boundary conditions. Soot particle size distribution functions were determined for the burner-stabilized, stagnation flame at several burner-to-stagnation surface separations. It was found that the tubular probe developed earlier perturbs the flow and flame temperature in a way which is better described by a one-dimensional stagnation reacting flow than by a burner-stabilized flame free of probe intrusion.     

Multimodal ultrafine particles from pulverized coal combustion in a laboratory scale reactor

Particle size distribution functions have been measured in a ethanol fueled flame reactor fed with a low amount of pulverized coal particles. The reactor is operated in low (5.0 vol.%) and high (76.5 vol.%) oxygen concentrations using two high volatile bituminous Colombian and Indonesian coals. A carbon black powder is also oxidized in the same conditions. Generated particles are sampled using rapid-dilution probes and the size distribution functions are measured on-line by a high resolution Differential Mobility Analyzer. Results clearly show that ultrafine particles, those with sizes lower than 100 nm, have a multimodal size distribution function. These particles have huge number concentrations in both investigated conditions whereas their formation is enhanced in the oxygen enriched condition. Ultrafine particles are almost totally dominated in number by the fraction having sizes below 30 nm. Nanoparticles also account for a significant fraction of total particle mass and slowly coagulate in the reactor. The shape of the size distribution functions is not affected by the coal type, at least for the two investigated coals. Results suggest that ultrafine particles form through the vaporization-nucleation-growth pathway involving inorganic ashes. Moreover the contribution of carbonaceous particles seems particularly important for size smaller than 5 nm.     

On evolution of particle size distribution functions of incipient soot in premixed ethylene-oxygen-argon flames

The evolution of the soot particle size distribution function (PSDF) and particle morphology are studied for premixed ethylene-oxygen-argon flat flames at equivalence ratio ?=2.07 over the maximum flame temperature range of 1600-1900 K. Experiments were carried out using an in-situ probe sampling method in tandem with a scanning mobility particle sizer (SMPS), yielding the PSDF for various distances from the burner surface. The morphology of the particles was examined by transmission electron microscopy (TEM) and atomic force microscopy (AFM). Within the particle size range that can be detected, the PSDF transitions from an apparent unimodal PSDF for high temperature flames (Tf>∼1800 K) to a bimodal PSDF at lower temperatures (Tf<∼1800 K). The bimodal PSDF has a noticeable trough that separates the nucleation tail and log-normal mode. This mode-transition trough had been previously thought to occur at a fixed particle size, but these results show a continuous shift of the trough location towards smaller sizes with increasing flame temperature. TEM images show that the particles are spherical, even when the PSDF is bimodal, suggesting that the bimodality occurs as the primary particles undergoes mass and size growth, and is not a result of particle aggregation. Atomic force microscopy of substrate deposited particles shows that particles spread and form hill like structures upon impact with the substrate surface, indicating that the particles are liquid-like at the time of impact.     

Particulate Matter size distribution in the exhaust gas of a modern diesel Engine fuelled with a biodiesel blend

The characteristics of particulate mater size distribution in the exhaust gas of an automotive diesel engine have been studied for a biodiesel blend of 30% rapeseed methyl ester (RME) and 70% ultra low sulphur diesel (ULSD) by volume (B30). The engine, a twin-turbo charged V6 equipped with a common rail fuel injection system, was operated on 16 steady-state points extracted from a corresponding New European Driving Cycle test with no engine system modification and a fast differential mobility spectrometer was used to determine the particulate number concentration and distribution. It is shown that the number-size distribution is dependent on engine operating conditions including the rate of exhaust gas recirculation (EGR). Compared with ULSD, B30 leads to a 41% smaller average size of the particles with EGR but gives rise to a higher number concentration under certain engine operating conditions, with the differences varying between nucleation and accumulation mode. The calculated particle total mass for B30 combustion aerosol is lower than the value with ULSD for all the engine operating conditions tested. The average B30 aerosol was 28% smaller in size on mass basis, compared to ULSD aerosol. For both fuels, the relationship between the particle total number and total mass has been found to be directly correlated and both the number and the mass of particles increase when the mean diameters of particles increase.     

Characterization of the effects of hydrogen addition in premixed methane/air flames

The use of hydrogenated fuels shows considerable promise for applications in gas turbines and internal combustion engines. In the present work, the effects of hydrogen addition in methane/air flames are investigated using both a laminar flame propagation facility and a high-pressure turbulent flame facility. The aim of this research is to contribute to the characterization of lean methane/hydrogen/air premixed turbulent flames at high pressures, by studying the flame front geometry, the flame surface density and the instantaneous flame front thermal thickness distributions. The experiments and analyses show that a small amount of hydrogen addition in turbulent premixed methane–air flames introduces changes in both instantaneous and average flame characteristics.     

Charge fraction distribution of nucleation mode particles: New insight on the particle formation mechanism

We measured particle size distributions of total and singly charged nanoparticles in premixed flames with different flame stoichiometry and temperature to investigate particle inception.Particle charging in flames occurs by diffusion charging involving ions formed by chemi-ionization reactions in the flame front. It can be described by a Boltzmann charge fraction distribution evaluated at the local flame temperature where the particles interact with the chemi-ions. As the particles coagulate in the post flame zone, their charge fraction is reduced. The charge distribution of the coagulated aerosol again results in a Boltzmann curve, this time evaluated at the local post flame gas temperature where the particles had their last coagulation event. Particle nucleation in the post flame zone, where chemi-ions are drastically reduced, produces uncharged particles.Considering the above charging processes, the charge fraction of the nucleation mode contains information on the location within the flame these particles were formed. The results show that in flames near the particle inception threshold, particles are charged close to the flame front and remain charged even late in the post flame zone. Furthermore, smaller particles undergo less charge neutralization by coagulation as they travel through the post flame zone than larger particles. A different scenario is observed in richer flames; the smaller particles eventually become uncharged, indicating that significant amounts of freshly nucleated particles in these flames are formed in the post flame zone. Whether nucleation preferentially occurs close to the flame front or persists into the post flame zone also depends on flame temperature.     

Experimental and kinetic modeling study on methylcyclohexane pyrolysis and combustion

Methylcyclohexane is the simplest alkylated cyclohexane, and has been broadly used as the representative cycloalkane component in fuel surrogates. Understanding its combustion chemistry is crucial for developing kinetic models of larger cycloalkanes and practical fuels. In this work, the synchrotron vacuum ultraviolet photoionization mass spectrometry combined with molecular-beam sampling was used to investigate the species formed during the pyrolysis of methylcyclohexane and in premixed flame of methylcyclohexane. A number of pyrolysis and flame intermediates were identified and quantified, especially including radicals (e.g. CH₃, C₃H₃, C₃H₅ and C₅H₅) and cyclic C6- and C7-intermediates (benzene, 1,3-cyclohexadiene, cyclohexene, toluene, C₇H₁₀ and C₇H₁₂, etc.). In particular, the observation of cyclic C6- and C7-intermediates provides important experimental evidence to clarify the special formation channels of toluene and benzene which were observed with high concentrations in both pyrolysis and flame of methylcyclohexane. Furthermore, the rate constants of H-abstraction of methylcyclohexane via H attack, and the isomerization and decomposition of the formed cyclic C₇H₁₃ radicals were calculated in this work. A kinetic model of methylcyclohexane combustion with 249 species and 1570 reactions was developed including a new sub-mechanism of MCH. The rate of production and sensitivity analysis were carried out to elucidate methylcyclohexane consumption, and toluene and benzene formation under various pyrolytic and flame conditions. Furthermore, the present kinetic model was also validated by experimental data from literatures on speciation in premixed flames, ignition delays and laminar flame speeds.     

Particulate emissions from laser ignited and spark ignited hydrogen fueled engines

Exponentially increasing energy demand and stricter emission legislations have motivated researchers to explore alternative fuels and advanced engine technologies, which are more efficient and environment friendly. In last two decades, hydrogen has emerged as promising alternative fuel for internal combustion (IC) engines and vehicles. For gaseous fuels, laser ignition (LI) has emerged as a novel ignition technique due to its superior characteristics, leading to improved combustion, engine performance and emission characteristics. Numerous advantages of LI system such as flexibility of plasma location, lower NO_x emissions and capability of igniting ultra-lean fuel–air mixture makes LI system superior compared to conventional spark ignition (SI) system. This study experimentally compares particulate emissions from hydrogen fueled engine ignited by LI and SI systems. Experiments were performed in a constant speed engine prototype, which was suitably modified to operate on gaseous fuels using both LI as well as SI systems. Particulate were characterized using engine exhaust particle sizer (EEPS) spectrometer. Results showed that LI engine resulted in relatively higher particulate number concentration as well as particulate mass compared to SI engine. In both ignition systems, particulate emissions increased with increasing engine load however rate of increase was relatively higher in LI system. Relatively larger count mean diameter (CMD) of particulate emitted from SI engine compared to LI engine was another important observation. This showed emission of relatively smaller particles in larger numbers from LI engine, compared to baseline SI engine.     

Exploratory studies of modeling approaches for hydrogen triple flames

A review of triple flame modeling is first presented, which demonstrates the need for additional work in this area. Building on previous methods described in the literature, a hybrid model that uses a weighted average of one-dimensional premixed and diffusion flamelet reaction rates has been proposed and evaluated for a hydrogen triple flame. Results indicated that some type of progress variable is needed for application of the diffusion flamelet contribution. Weighting the premixed flamelet reaction rate contribution at 100%, it is shown that peak temperatures between the model and a case employing detailed chemistry vary 7.5%, while heat release rate, flame speed, and mass fraction contours agree well.A second model, based on a library of reaction rates built from numerical studies which directly resolve the propagating triple flame has also been tested. Computational time for the baseline case is shown to be reduced by a factor of 3 ½ in comparison to use of detailed chemistry. The role of scalar dissipation rate as a necessary independent variable to the library has also been investigated using simulations with variable mixing layer thicknesses. Overall, it is found that large changes in local mixture fraction gradient cause rather small changes in propagation speed and total heat release rate of the hydrogen triple flame. This implies that such a model may be useful for CFD simulations that do not employ spatial resolution capable of resolving the triple flame itself.     

The effect of biodiesel and bioethanol blended diesel fuel on nanoparticles and exhaust emissions from CRDI diesel engine

Biofuel (biodiesel, bioethanol) is considered one of the most promising alternative fuels to petrol fuels. The objective of the work is to study the characteristics of the particle size distribution, the reaction characteristics of nanoparticles on the catalyst, and the exhaust emission characteristics when a common rail direct injection (CRDI) diesel engine is run on biofuel-blended diesel fuels. In this study, the engine performance, emission characteristics, and particle size distribution of a CRDI diesel engine that was equipped with a warm-up catalytic converters (WCC) or a catalyzed particulate filter (CPF) were examined in an ECE (Economic Commission Europe) R49 test and a European stationary cycle (ESC) test. The engine performance under a biofuel-blended diesel fuel was similar to that under D100 fuel, and the high fuel consumption was due to the lowered calorific value that ensued from mixing with biofuels. The use of a biodiesel–diesel blend fuel reduced the total hydrocarbon (THC) and carbon monoxide (CO) emissions but increased nitrogen oxide (NO_x) emissions due to the increased oxygen content in the fuel. The smoke emission was reduced by 50% with the use of the bioethanol–diesel blend. Emission conversion efficiencies in the WCC and CPF under biofuel-blended diesel fuels were similar to those under D100 fuel. The use of biofuel-blended diesel fuel reduced the total number of particles emitted from the engine; however, the use of biodiesel–diesel blends resulted in more emissions of particles that were smaller than 50 nm, when compared with the use of D100. The use of a mixed fuel of biodiesel and bioethanol (BD15E5) was much more effective for the reduction of the particle number and particle mass, when compared to the use of BD20 fuel.     

Comprehensive analysis on the effect of lube oil on particle emissions through gas exhaust measurement and chemical characterization of condensed exhaust from a DI SI engine fueled with hydrogen

This research study aims to investigate the causes of the particles emitted by a spark ignition engine fueled with hydrogen. The experiments were carried out on a single cylinder 250 cm³ direct injection spark ignition engine. Two operating conditions at 2000 rpm both full and low load, representative of typical urban conditions, were investigated. A physical characterization of the particles, size and number, was performed through an Engine Exhaust Particle Sizer coupled to a single diluter. Chemical characterization was carried out on the condensed exhaust. The simultaneous analysis of the physical properties and their chemical characterization allows to point out not only the role of the oil on the particle emissions but also to give an important information on its state/composition, if it was unburned and oxidized. Particles were detected with conventional spectrometer at low load while at high load the noise signal ratio is too high to distinguish the presence of particles. More detailed chemical techniques highlighted the presence of PAH, alkyl-PAHs, oxy-PAHs and unburned hydrocarbon in the exhaust due to the mineral oil.     

Effects of ellipsoidal and regular hexahedral particles on the performance of the waste heat recovery equipment in a methanol reforming hydrogen production system

Hydrogen production from methanol has attracted attention due to its wide range of raw material sources and mature technology. Using waste heat of industrial high temperature solid particles like blast slag and steel slag etc. To provide vaporization heat and reaction heat for the reaction between methanol and water is an emerging technology for hydrogen production from methanol, which can save additional thermal energy resources. Herein, the performances of equipment that uses the waste heat of ellipsoidal and regular hexahedral particles to provide a heat source for methanol to hydrogen were explored by the DEM-CFD method. Compared with spherical particles of the same equivalent diameter, ellipsoidal and regular hexahedral particles have poor fluidity in the stagnant area, and the empty area is enlarged and irregular in shape. The average velocity peaks of the ellipsoidal and regular hexahedron particles are larger than those of spherical particles, and the overall mean velocity fluctuation of ellipsoidal particles is similar to that of spherical particles while the regular hexahedron particles' is larger. The average temperature drop rate of the ellipsoidal and regular hexahedral particles is slower than that of spherical particles, the uniformity of temperature distribution is worse than that of spherical particles. The ellipsoidal and regular hexahedral particles’ average effective heat transfer coefficient is smaller than that of spherical particles, and the heat transfer effect is weaker than that of spherical particles. The effective heat transfer coefficient of ellipsoidal particles is 2.95 W/(m⁻²∙K⁻¹) lower than that of spherical particles and the effective heat transfer coefficient of hexahedral particles is 6.09 W/(m⁻²∙K⁻¹) lower than that of spherical particles. Therefore, compared with the spherical particles of the same equivalent diameter, ellipsoidal and regular hexahedral particles produce less hydrogen.     

Impact of waste cooking oil in biodiesel blends on particle size distributions from a city-car engine

Using biodiesel as a blending component in diesel engine has demonstrated to reduce hydrocarbon and particulate matter emissions. Literature showed that biodiesel type, engine architecture and test conditions deeply affect performance and emission characteristics. Among suitable biodiesel fuels, waste cooking oil (WCO) is considered very attractive due to the reduced environmental impact without sacrificing engine performance.This paper aims at investigating how mixing ratio of biodiesel from WCO and mineral diesel affects the particle size distributions of a current state of art small displacement diesel engine.Experimental tests have been performed on an up-to date light common rail diesel engine. Its complete operative field has been investigated. The results obtained show that the use of biodiesel blends from WCO reduces the total number of particles emitted from the engine with respect to the diesel fuel; the reduction is more evident as the percentage of biodiesel in the blend increases. The number of particles in WCO biodiesel soot with diameter smaller than 10 nm is reduced as compared to diesel fuel; the same trend is observed for diameters larger than 200 nm; comparable particle numbers were obtained in the ultrafine range (Dp < 100 nm).     

Effect of hydrogen addition on NOx formation in high-pressure counter-flow premixed CH4/air flames

A laboratory-scale laminar counterflow burner was used to investigate NO formation in high pressure premixed CH₄/H₂/air flames. New experimental results on NO measurements by LIF were obtained at high pressure in CH₄/H₂/air flames with H₂ content fixed at 20% in the fuel at pressures ranging from 0.1 to 0.7 MPa and an equivalence ratio progressively decreased from 0.74 to 0.6. The effects of hydrogen addition, equivalence ratio and pressure are discussed. These results are satisfactorily compared to the simulations using two detailed mechanisms: GDFkin®3.0_NOmecha2.0 and the mechanism from Klippenstein et al., which are the most recent high-pressure NOx formation mechanisms available in the literature. A kinetic analysis based on Rate of Production/Rate of Consumption and sensitivity analyses of NO is then presented to identify the main pathways that lead to the formation and consumption of NO. In addition, the effect of hydrogen addition on NO formation pathways is described and analysed.     

Studies on formation and control of combustion particulate matter in China: A review

Airborne particulate matter (PM) now exceeds sulfur dioxide and nitrogen oxides to become principal urban pollutant in most major cities of China. This paper gives an overview of fundamental studies on the formation and control of combustion PM from many research groups in China. About 62.8% major cities in China have lowered their annual mean PM₁₀ concentrations to less than 100 μg/m³ as of year 2006. The coal combustion source contributes 15–20% to fine particulates in Beijing because of the coal-dominant energy consumption structure. Overall, in mainland China the PM emission from coal-fired power plants totals 3.81 million tonnes per year, accounting for 44.6% of the total PM. Then, the characteristics of PM₁₀ from both pulverized coal plants and circulated-fluidized bed coal plants are discussed. Finally, the R&D of emission control technologies of PM₁₀ including combustion modification, electrically enhanced fabric filtration and novel agglomeration approaches are reviewed in detail.     

Reprint of: Studies on formation and control of combustion particulate matter in China: A review

Airborne particulate matter (PM) now exceeds sulfur dioxide and nitrogen oxides to become principal urban pollutant in most major cities of China. This paper gives an overview of fundamental studies on the formation and control of combustion PM from many research groups in China. About 62.8% major cities in China have lowered their annual mean PM₁₀ concentrations to less than 100 μg/m³ as of year 2006. The coal combustion source contributes 15–20% to fine particulates in Beijing because of the coal-dominant energy consumption structure. Overall, in mainland China the PM emission from coal-fired power plants totals 3.81 million tonnes per year, accounting for 44.6% of the total PM. Then, the characteristics of PM₁₀ from both pulverized coal plants and circulated-fluidized bed coal plants are discussed. Finally, the R&D of emission control technologies of PM₁₀ including combustion modification, electrically enhanced fabric filtration and novel agglomeration approaches are reviewed in detail.     

Lewis number effect on the propagation of premixed flames in narrow adiabatic channels: Symmetric and non-symmetric flames and their linear stability analysis

The propagation of premixed flames with different Lewis numbers in a planar channel subject to a Poiseuille flow is considered within the diffusive-thermal model for steady and time-dependent cases. It was found that, depending on the Lewis number and the flow rate, symmetric and non-symmetric flames are possible. The existence of multiple steady solutions in cases of the low Lewis number is demonstrated. The time-dependent simulations carried out for high Lewis number flames also showed the symmetric and non-symmetric oscillatory solutions.Linear stability analysis of two-dimensional steady-states was performed using a practical method developed in the paper and applied to calculate the main eigenvalue. It was shown that for symmetric flames with a low Lewis number the increase in the flow rate leads to a loss of stability with subsequent formation of non-symmetric solutions. For flames with a high Lewis number the Poiseuille flow produces a stabilization effect. The results of the stability analysis were successfully compared with the results of direct numerical simulations.