Ammonia is a promising alternative fuel for CO2 emission mitigation. The use of ammonia blends allows for more flexibility compared to pure ammonia fuel and is often considered for immediate CO2 emission reduction in existing facilities running on natural gas. However, fundamental studies on these fuel mixtures remain scarce. This study thus focuses on ammonia/methane blend fuels. The effect of ammonia on methane jet flame stabilization is investigated using a non-premixed jet flame configuration to observe the flame stabilization mechanisms, the flame-burner interactions and how ammonia addition affects the attached flame stabilization up to liftoff. The flame tip position was observed using CH* chemiluminescence. Heat transfer to the burner was monitored by temperature measurement at the burner lip and inside the burner to observe the impact of ammonia addition on thermal interactions. The main stabilization regimes described for the methane non-premixed jet flame are still observed in the case of ammonia addition. However, the transition between those regimes appeared to be shifted toward larger velocities relative to the methane case due to ammonia addition. Those changes could be related to the change in the mixture combustion properties which affects both flame position, heat transfer to the burner and in turn the transition between the identified stabilization regimes. The dynamic leading to liftoff was further analyzed to highlight how ammonia addition perturbated the stabilization balance up to liftoff.

Ammonia combustion gas turbines have drawn much attention for their potential to power hydrogen-carrier applications. In 2014, 21-kW of power generation was achieved with kerosene-ammonia co-combustion using a 50-kW-class micro gas turbine. In 2015, methane-ammonia co-combustion and 100% ammonia gas combustion were separately employed to generate 42-kW of power in both cases. However, this microgas turbine still requires kerosene to start. In gas turbines, the suitability of the air flow for ignition depends on the fuel. Low- NOx combustors were developed using the staged combustion concept to achieve rich-lean combustion. These combustors can burn kerosene at start-up but not at full load. We have also developed a micro gas turbine system that utilizes 100% ammonia gas in combustion during full-load operation. However, start-up with ammonia gas is difficult without hot air. The ignition of ammonia gas is so difficult that during the startup process, a more easily ignitable fuel is also used. In this study, we developed a new 50-kW-class micro gas turbine that can be started with gaseous fuel. Start-up with methane gas was achieved using a newly designed low-NOx combustor. In addition, because hydrogen can be easily obtained from ammonia decomposition, the addition of hydrogen gas to ammonia gas has garnered attention for ammonia gas turbine applications. The application of hydrogen to initiate combustion in a micro gas turbine was also investigated.

<p>In this study, the flame propagation experiments of ammonia/methane/air were conducted using a fan-stirred constant volume vessel in order to clarify the effect of turbulence on the propagation limits of the ammonia/methane/air. Results show that the ammonia/methane/air mixture with a 0.9 equivalence ratio can propagate at the highest turbulence intensity and flames can propagate in larger turbulence Karlovitz number at smaller Markstein number. These tendencies are considered to be due to the diffusional-thermal instability of the flame surface.</p>

Methane-ammonia mixtures have potentials as low-carbon fuels for gas turbines, however significantly high fuel NOx production in their flames present challenges to their application. This study aims to provide deep insight into the physical and chemical processes involved in the formation and control of emissions from the combustion of CH4-NH3-air with up to 30% ammonia by heat fraction in gas turbine combustors. Hence, laser diagnostics techniques such as Particle Image Velocimetry (PIV), and Planar Laser Induced Fluorescence (PLIF) imaging, in addition to Fourier Transform Infrared (FTIR) gas analysis were employed to study the flow field, flame structure and emissions characteristics of a micro gas turbine swirl combustor fuelled with CH4-NH3-air mixtures. The control of emissions from the flames was further studied using Large Eddy Simulation (LES) of a model swirl combustor. The results show that NOx emissions from premixed CH4-NH3-air in single-stage combustion were more than 5000 ppmv at equivalence ratios, Phi = 0.8-1.1, which is about twice more than the values already reported for NH3-air. Trends in NOx emissions correspond with the trends in OH radicals concentration in the combustor owing to the relevance of OH radicals in fuel NOx production. Emissions control leading to significantly low emissions such as 49 ppmv of NOx, 2 ppmv of CO and approximately zero N2O, HCN and NH3 emissions with a 99.8% combustion efficiency was achieved using rich-lean combustion. An optimum Phi of the primary combustion zone for low NOx emission was identified, which varied from 1.30 to 1.35 depending on the ammonia fraction. For Phi richer (leaner) than the optimum Phi, NOx emission increased due to an increase in NOx production in the secondary (primary) combustion zone. Rich-lean combustion of CH4-NH3-air emitted less NOx than that of NH3-air because the higher flame speed of CH4-NH3-air mixtures ensured lower NOx production in the secondary combustion zone. (C) 2019 The Combustion Institute. Published by Elsevier Inc. All rights reserved.

This study is dedicated to understanding the combustion and emission characteristics of turbulent non-premixed ammonia (NH3)/air and methane (CH4)/air swirl flames in a rich-lean gas turbine-like combustor at high pressure under various wall thermal boundary conditions. In this study, the emission characteristics of both flames were obtained through numerical simulations using large-eddy simulations with the finite-rate chemistry technique. In addition, for NH3/air flames, simultaneous NO and OH planar laser-induced fluorescence (PLIF) images were acquired in order to qualitatively verify the numerical results. The results show that the minimum NO emission could be obtained when the primary zone global equivalence ratio (phi(global/pri)) was 1.1, irrespective of the wall thermal condition, using a rich-lean cornbustor in NH3/air flames, whereas in CH4/air flames, the maximum NO emission was obtained with a phi(global/pri) value of 1.0. Moreover, in NH3/air flames, the local NO concentration is largely dependent on the local OH concentration, whereas in CH4/air flames, the local NO concentration is largely dependent on the local temperature. The NO-OH correlation in NH3/air flames was experimentally verified using simultaneous NO and OH PLIF images. Additionally, it was found that, in NH3/air flames, the wall heat losses due to the combustor wall cooling greatly affected the NH3 oxidation and led to significant emissions of unburnt NH3, although lower NO emission resulted from the combustor wall cooling. This was primarily because of the lower OH concentration level in the flame region owing to the wall heat losses. (C) 2019 The Combustion Institute. Published by Elsevier Inc. All rights reserved.

<p>This study focuses on the application of OH planar laser-induced fluorescence (OH-PLIF) in high-pressure rocket combustion conditions, up to 7.0 MPa. The signal to noise ratio of PLIF degrades in high-pressure combustion owing to effects such as line broadening and interference from intense chemiluminescence. The OH(2,0) band excitation method was applied to obtain the OH(2,1) fluorescence emitted near 290 nm and filter out the intense OH(0,0) band chemiluminescence emitted near 308 nm. The gaseous H₂/O₂ (GH₂/GO₂) jet diffusion flame was formed using a recessed coaxial shear injector. The GH₂/GO₂ injection Reynolds number, <i>Re</i> (<i>Re</i>_H2/<i>Re</i>_O2 ≈ 2320/22800–4660/45600), was varied to examine the variation of the flame structure and reaction zone thickness under each pressure condition <i>P_c</i>, and <i>Re</i> injection condition. In addition, the variation of the experimentally derived full width at half maximum (FWHM) of the radial OH distribution, <i>δ</i>_OH, with the Damköehler number, <i>Da</i>, was compared with that of the simulated FWHM of the OH mole fraction, <i>δ</i>_OH-SIM. The OH distribution was clearly observed in the instantaneous PLIF image while eliminating the intense OH chemiluminescence even in the highest pressure condition of 7.0 MPa, which is a pressure higher than any of the previous OH-PLIF studies conducted on rocket combustion. The flame structure showed the typical characteristics of a turbulent jet diffusion flame and depended on <i>Re</i> rather than on the chamber pressure <i>P_c</i>. The variation of <i>δ</i>_OH with <i>Da</i> corresponded qualitatively with <i>δ</i>_OH-SIM and showed the characteristics of flame stretch in the vicinity of the injector.</p>

Low-NOx NH3-air combustion power generation technology was developed by using a 50-kWe class micro gas-turbine system at the National Institute of Advanced Industrial Science and Technology (AIST), Japan, for the first time. Based on the global demand for carbon-free power generation as well as recent advances involving gas-turbine technologies, such as heat-regenerative cycles, rapid fuel mixing using strong swirling flows, and two-stage combustion with equivalence ratio control, we developed a low-NOx NH3-air non-premixed combustor for the gas-turbine system. Considering a previously performed numerical analysis, which proved that the NO reduction level depends on the equivalence ratio of the primary combustion zone in a NH3-air swirl burner, an experimental study using a combustor test rig was carried out. Results showed that eliminating air flow through primary dilution holes moves the point of the lowest NO emissions to the lesser fuel flow rate. Based on findings derived by using a test rig, a rich-lean low NOx combustor was newly manufactured for actual gas-turbine operations. As a result, the NH3 single fueled low-NOx combustion gas-turbine power generation using the rich-lean combustion concept succeeded over a wide range of power and rotational speeds, i.e., below 10-40 kWe and 75,000-80,000 rpm, respectively. The NO emissions were reduced to 337 ppm (16% O-2), which was about one-third of that of the base system. Simultaneously, unburnt NH3 was reduced significantly, especially at the low electrical power output, which was indicative of the wider operating range with high combustion efficiency. In addition, N2O emissions, which have a large Global Warming Potential (GWP) of 298, were reduced significantly, thus demonstrating the potential of NH3 gas-turbine power generation with low environmental impacts. (C) 2018 The Combustion Institute. Published by Elsevier Inc. All rights reserved.

In this study, NO emission characteristics of turbulent non-premixed CH4/NH3/air flames (energy fraction of NH3 (ENH3) of 20%) in a rich-lean gas turbine like combustor were numerically studied using a large eddy simulation technique and detailed chemistry, and compared with CH4/air (in which thermal NO is dominant) and NH3/air (in which Fuel NO is dominant) flames. The study found that NO production in CH4/NH3/air flames is primarily dependent on OH concentration, like NH3/air flames, which confirmed that the NO is mainly fuel origin. Consequently, the study found that low NO emission in the order of 200 ppm can be achieved using a rich-lean combustor at primary zone global equivalence ratio of 1.4 owing to low OH concentration. However in lean and stoichiometric conditions, NO emissions from CH4/NH3/air flames were two times higher than those of NH3/air flames and ten times higher than those of CH4/air flames due to high OH concentration. In addition, CH4/NH3 mixing enhances the burning characteristics of NH3, and even until global equivalence ratio of 1.4, unburnt NH3 not get into the secondary stage from the primary stage.

Recently ammonia has attracted interest as a hydrogen energy carrier as well as a carbon-free fuel. Fundamental combustion characteristics of ammonia mixtures, such as the laminar burning velocity and ignition characteristics, have been clarified. However, product gas characteristics which are important for the use of ammonia in combustors have not been clarified because of difficulty of experimentation. In this study, product gas characteristics of strain stabilized ammonia/air premixed laminar flames were experimentally investigated. An FTIR gas analyzer were employed for gas analysis. Although required amount of sampling gas for FTIR gas analyzer is relatively large, it could be reduced using the dilution sampling method. As a result, product gas characteristics from lean to rich ammonia/air flames were evaluated. Then, analysis of the chemical kinetics of ammonia/air flames was performed using the detailed reaction mechanism developed by Okafor et al. The results showed that some important reactions may need to be updated for accurate modelling of the product gas characteristics.

© Asia-Pacific Conference on Combustion, ASPACC 2019.All right reserved. With the renewed interest in ammonia as a fuel, the potentials of methane-ammonia mixtures as low-carbon fuels for gas turbines are being evaluated. The significantly high NOx production in the flames of ammonia-containing mixtures presents an important challenge to their application. In this study, PIV, OH-PLIF, and FTIR exhaust gas analysis were employed to study the characteristics of a micro gas turbine combustor fueled with CH4-NH3-air mixtures at 0.10 MPa, and methods for controlling emissions from the flames are proposed. The concentration of ammonia in the fuel which was expressed in terms of the heat fraction, NH3, varied from 0 to 0.3. The results show that NO emissions from the ammonia-containing flames can be more than 5000 ppmv (parts per million volume) in single-stage combustion. Emissions control was achieved with two-stage rich-lean combustion and an optimum equivalence ratio upstream of the combustor for low NOx emission was found to be around 1.30 for NH3= 0.20 and 0.30, and 1.35 for NH3= 0.10. At these optimum upstream equivalence ratios, the NO emission from the combustor was in the order of 40 ppmv at 16% O2 concentration with complete fuel consumption over a wide range of overall equivalence ratios.

High temperature flames, which can be produced via oxygen-enriched combustion, potentially have improved the combustion characteristics over air-only combustion flames due to their substantially higher flame temperatures. These conditions necessitate the use of non-intrusive optical measuring methods to measure the temperature and the chemical species in the flame. To develop an optical measurement calibration burner that can be used at high pressure and temperature conditions, a new calibration burner which employed water-cooled multi-hole nozzle was developed in this study. Premixed CH4/O2/N2 oxygen-enriched conditions were selected to investigate both the heat-resisting properties of the developed burner nozzle and the burner’s flame characteristics. OH-Planar Laser-Induced Fluorescence (OH-PLIF) measurements were conducted on the flames to observe the OH distributions. The flame temperature at 0.10 MPa was derived using a Boltzmann-plot for the OH fluorescence excitations. To verify the variation of molecular concentration with equivalence ratio for the experimental flames qualitatively, the experimentally acquired OH and CH chemiluminescence intensities were compared with the simulated partial pressure of OH* and CH*, respectively. Experimental results showed that the CH4/O2/N2 flames were stabilized on the burner nozzle in a wide range of oxygen-enrichment ratio, from 0.40 to 1.0 at atmospheric pressure. At an oxygen-enrichment ratio of 0.45, the flames were also stabilized in pressure conditions up to 0.49 MPa, while the inner nozzle temperature was lower than 400 K. The OH-PLIF images showed that the OH was distributed almost uniformly along the axial direction of the burner, and demonstrated similar characteristics to that of a flat flame. The derived maximum flame temperature at atmospheric pressure was approximately 2650 K at an oxygen-enrichment ratio of 0.80. The variation of the OH and CH chemiluminescence intensities with change of equivalence ratios corresponded roughly with the simulated partial pressures of OH* and CH* at each pressure condition.

With the renewed interest in ammonia as a carbon-neutral fuel, mixtures of ammonia and methane are also being considered as fuel. In order to develop gas turbine combustors for the fuels, development of reaction mechanisms that accurately model the burning velocity and emissions from the flames is important. In this study, the laminar burning velocity of premixed methane–ammonia–air mixtures were studied experimentally and numerically over a wide range of equivalence ratios and ammonia concentrations. Ammonia concentration in the fuel, expressed in terms of the heat fraction of NH3 in the fuel, was varied from 0 to 0.3 while the equivalence ratio was varied from 0.8 to 1.3. The experiments were conducted using a constant volume chamber, at 298 K and 0.10 MPa. The burning velocity decreased with an increase in ammonia concentration. The numerical results showed that the kinetic mechanism by Tian et al. largely underestimates the unstretched laminar burning velocity owing mainly to the dominance of HCO (+H, OH, O2) = CO (+H2, H2O, HO2) over HCO = CO + H in the conversion of HCO to CO. GRI Mech 3.0 predicts the burning velocity of the mixture closely however some reactions relevant to the burning velocity and NO reduction in methane–ammonia flames are missing in the mechanism. A detailed reaction mechanism was developed based on GRI Mech 3.0 and the mechanism by Tian et al. and validated with the experimental results. The temperature and species profiles computed with the present model agree with that of GRI Mech 3.0 for methane–air flames. On the other hand, the NO profile computed with the present model agrees with Tian et al.’s mechanism for methane–ammonia flames with high ammonia concentration. Furthermore, the burned gas Markstein length was measured and was found to increase with equivalence ratio and ammonia concentration.

Ammonia is a possible candidate for use as a hydrogen energy carrier as well as a carbon free fuel. In this study, flame stability and emission characteristics of swirl stabilized ammonia/air premixed flames were experimentally investigated. Results showed that ammonia/air premixed flame could be stabilized for various equivalence ratios and inlet flow velocity conditions in a swirl burner without any additives to enhance the reaction of ammonia even though the laminar burning velocity of ammonia is very slow. The lean and rich blowoff limits were found to be close to the flammability limits of the ammonia flame. In addition, emission characteristics were investigated using an FTIR gas analyzer. The NO concentration decreased and ammonia concentration increased under rich conditions. Moreover, it was found that there is an equivalence ratio in rich condition in which NO and ammonia emission are in the same order. (C) 2017 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved.

A demonstration test with the aim to show the potential of ammonia-fired power plant is planned using a micro gas turbine. 50kW class turbine system firing kerosene is selected as a base model. Over 40kW of power generation was achieved by firing ammonia gas only. Over 40kW of power generation was also achieved by firing mixture of ammonia and methane. However ammonia gas supply increases NOx in the exhaust gas dramatically. NOx concentration in the exhaust gas of gas turbine reached at over 600ppm. In the case of the gas turbine operation firing kerosene-ammonia with 31kW of power generation at 75,000rpm of rotating speed, the LHV (Lower Heating Value) ratio of ammonia to the total supplied fuel was changed from 0% to 100% in detail. NO emission increases rapidly to around 400ppm with ammonia at 7% of LHV ratio of ammonia. Then NO emission increases gradually to 600ppm with ammonia at 27% of LHV ratio of ammonia. NO emission has the peak around 60% of LHV ratio of ammonia. NO emission decreases below 500ppm at 100% of LHV ratio of ammonia. The gas turbine operation firing methane-ammonia with 31kW of power generation at 75,000rpm of rotating speed was also tried. NO emission increases rapidly to around 470ppm with ammonia at 7% of LHV ratio of ammonia. Then NO emission increases gradually to 600ppm with ammonia around 30% of LHV ratio of ammonia. NO emission has the peak at 65% of LHV ratio of ammonia. NO emission decreases below 500ppm at 100% of LHV ratio of ammonia. Since the ammonia flame in the prototype combustor seems to be inhomogeneous, ammonia combustion in the prototype combustor may have high NOx region and low NOx region. Therefore there is a possibility of low-NOx combustion. Flame observation was planned to know combustion state for improvement toward the low NOx combustor. Flame observation from the combustor exit was available by extending the combustor exit with the adaptor of the bent coaxial tubes and the quartz window. Swirling flames of ammonia, methane and methane-ammonia were observed near the center axis of the combustor. Flame observation at 39.1kW of power generation was succeeded. In the case of the flame observation, fuel consumption increased due to increase of the heat loss from the combustor. The emissions of NO and NH3 clearly depend on the combustion inlet temperature at 75,000rpm of rotating speed. The emissions of NO and NH3 in the case of the flame observation setting corresponds to the emission in the case of the normal setting at the condition that the power output is 11.2kW lower.

To protect against global warming, a massive influx of renewable energy is expected. Although hydrogen is a renewable media, its storage and transportation in large quantity is difficult. Ammonia, however, is a hydrogen energy carrier and carbon-free fuel, and its storage and transportation technology is already established. Although ammonia combustion was studied in the 1960s in the USA, the development of an ammonia combustion gas turbine had been abandoned because combustion efficiency was unacceptably low. Since that time, in the combustion field, ammonia has been thought of as a fuel N additive in the study of NOx formation. Recent demand for hydrogen carrier revives the usage of ammonia combustion, but no one has attempted an actual design setup for ammonia combustion gas turbine power generation. The National Institute of Advanced Industrial Science and Technology (AIST) in Japan successfully performed ammonia-kerosene co-fired gas turbine power generation in 2014, and ammonia-fired gas turbine power generation in 2015. In the facilities, a regenerator-heated, diffusion-combustion micro-gas turbine is used, and its high combustor inlet temperature enables high thermal efficiency of ammonia combustion compared with that of methane combustion. Adoption of the regenerator increased combustor inlet temperature and enhanced flame stability in ammonia-air combustion. Although NOx emission from a gas turbine combustor is high, a Selective Catalytic Reduction (SCR) after gas turbine combustor reduces NOx emission to less than 10 ppm. This means that the ammonia combustion gas turbine design, abandoned in the 1960s for its unacceptably low combustion efficiency, has performed successfully with regenerator and SCR technology. However, the weakness of these facilities was that they required large-size SCR equipment in order to suppress a high concentration of NOx. Although NOx reduction in the combustion process is desirable, low NOx combustion technology is difficult because ammonia had been thought of as a source of fuel-NO. In the case of premixed ammonia-air flame, there exists a low emission window of NOx and NH3 in a certain equivalence ratio, but combustion intensity is very low because the laminar burning velocity of NH3-air is one-fifth that of CH4-air. This means that, when utilizing the window of premixed ammonia-air flame, scale-up of the combustion chamber or fuel additives for enhancement of flame stability is necessary. This study shows that the addition of H-2 is effective for low NOx combustion with high combustion efficiency. In addition, H-2 can be easily made from NH3 decomposition. The other option is diffusion combustion. Further research on low NOx combustion is needed.

© 2017 AIChE. All rights reserved. A massive influx of renewable energy is required in order to mitigate global warming. Although H2 is a renewable medium, its storage and transportation in large quantities has some problems. NH3 fuel, however, is an H2 energy carrier and carbon-free fuel, and its storage and transportation technology is already established. Although NH3 fuel combustion was studied in the 1960s in the USA, the development of an NH3 fuel gas turbine had been abandoned because the combustion efficiency was unacceptably low. The recent demand for H2 energy carriers has revived the interest of NH3 fuel; however an actual design setup for NH3 fuel gas turbine power generation has not been attempted. The National Institute of Advanced Industrial Science and Technology (AIST) in Japan, in collaboration with Tohoku University successfully achieved NH3-kerosene gas turbine power generation in 2014, and NH3 fuel gas turbine power generation in 2015. The facility consists of an NH3 fuel-supply system, NH3 gas compressor, 50 kWe-class micro gas turbine, Selective Catalytic Reduction (SCR) NOx-reduction apparatus, and loading equipment. The micro gas turbine utilizes a diffusion combustor to stabilize the flame and a regenerative heat exchanger to improve the thermal efficiency of the gas turbine cycle. Considering that the laminar burning velocity of NH3/air is one-fifth that of CH4/air, the combustion intensity of NH3/air is very low. In addition, NH3 has been thought of as a fuel N additive in the study of NOx formation. Although NOx reduction is desirable in the combustion process, low NOx combustion technology is difficult because NH3 has been thought of as the source of fuel-NO. Hence, in order to evaluate the performance of NH3 fuel gas turbine power generation, it is important to focus on the combustion stability and combustion emissions. Earlier, we have reported the combustion emission characteristics of NH3 fuel gas turbines; in the case of NH3/air, concentrations of NO and unburnt NH3 strongly depend on the combustor inlet temperature, and in the case of NH3/CH4/air, the concentration of NO at a constant electric power output depends on the NH3 ratio in the mixture. This report additionally shows that other than NO, the combustion emissions of NO2 and N2O also decrease considerably at high electric power output. These results arise from the restriction of the eigen balance of fuel, air, and heat, because the compressor and turbine are connected by a single shaft in the gas turbine. In the next step, the element test was carried out using a combustor test rig to develop a low NOx combustor. It is difficult to characterize combustion emissions with the former parameters in the case of a combustor test rig, because there is no restriction of the quantity of fuel, air, and combustor inlet temperature. Thus, this paper reports the combustion emissions of NH3 fuel gas turbine re-characterized with respect to the other parameters, such as fuel flow rate, overall equivalence ratio, combustor pressure, and combustion temperature.

© 2018 Combustion Institute. All Rights Reserved. In order to efficiently utilize ammonia as a gas turbine fuel, it is relevant to understand the combustion characteristics of ammonia in a gas turbine combustor. In this study stabilization and emission characteristics of non-premixed ammonia flames were investigated experimentally using a micro gas turbine swirl burner. The effects of ammonia injection angle, global equivalence ratio of the non-premixed flame, and ambient pressure on flame stabilization and emissions of unburned NH3 and NO were investigated at gas inlet temperature of 298 K. One dimensional numerical calculations using ANSYS Chemkin PRO were performed to extend understanding of the experimental results. Ammonia flames were stabilized over a range of equivalence ratios. An increase in the ammonia injection angle to the burner axis resulted in increased flame stability and more efficient combustion. Ammonia emission increased with an increase in global equivalence ratio. On the other hand, NO emission decreased with global equivalence ratio owing to the increased reduction of NO by amine radicals. With an increase in pressure, NO emissions decreased due to a decrease in H and OH radicals concentration.

The effects of unburned-gas temperature on the characteristics of cellular premixed flames generated by hydrodynamic and diffusive-thermal instabilities were numerically investigated. Two-dimensional reactive flow was calculated in large space, based on the compressible Navier-Stokes equations including a one-step irreversible chemical reaction. The dynamic behavior of cellular premixed flames, i.e. the coalescence and division of cells, appeared in large space owing to intrinsic instability. The behavior of flame fronts became more unstable with a decrease in unburned-gas temperature, even though the burning velocity of a planar flame reduced. This was due to the strength of thermal-expansion effects and to the enlargement of Zeldovich numbers. We found that the average burning velocity of a cellular flame normalized by that of a planar flame increased as the unburned-gas temperature became lower and the space size became larger. To elucidate the increase of burning velocity, we proposed the new model and showed that the normalized increment factor of burning velocity became larger under low unburned-gas temperature. In addition, we performed fractal analysis to consider the fractal dimension for three-dimensional flow. The obtained fractal dimension corresponding to laminar flames was nearly identical to the experimental and numerical results of turbulent flames.

The purpose of this study was to develop an advanced method to measure the properties of rocket combustion using the OH(2,0) band-excited Planer Laser-induced Fluorescence (OH-PLIF) method. This diagnostic method was applied to capture images of the high-pressure H-2/O-2 jet diffusion flames found in typical liquefied bi-propellant rocket combustion. In addition, axisymmetric numerical simulations of H-2/O-2 jet flames modeling the experimental conditions were conducted to evaluate the consistency of the OH-PLIF imaging results and to predict the OH chemiluminescence intensity and flame temperature at high pressure. Experimental results show that it is possible to detect the OH(2,1) band fluorescence effectively by eliminating the interference of OH(0,0)-band chemiluminescence under high-pressure conditions of up to 2.0 MPa. The OH fluorescence signal distributed near the injector face almost corresponded to the OH molar concentration distributions simulated by numerical simulations. Moreover, the simulated pressure dependence of the local OH+ peak mole concentration reasonably corresponded to that of the local peak chemiluminescence intensity of the experimental chemiluminescence images.

For the first time, NH3-air combustion power generation has been successfully realized using a 50 kW class micro gas turbine system at the National Institute of Advanced Industrial Science and Technology (AIST), Japan. Based on the global demand for carbon-free power generation as well as recent advances involving gas-turbine technologies, such as heat-regenerative cycles, rapid fuel mixing using strong swirling flows, and NOx reduction using selective catalytic reduction (SCR), allow us to realize NH3-air combustion gas-turbine system, which was abandoned in the 1960's. In the present system, the combustor adopted gaseous NH3 fuel and diffusion combustion to enhance flame stability. The NH3 pre-cracking apparatus for combustion enhancement using generated H-2 was not employed. The NH3-air combustion gas-turbine power generation system can be operated over a wide range of power and rotational speeds, i.e., 18.4 kW to 44.4 kW and 70,000 rpm to 80,000 rpm, respectively. The combustion efficiency of the NH3 -air gas turbine ranged from 89% to 96% at 80,000 rpm. The emission of NO and unburnt NH3 depends on the combustor inlet temperature. Emission data indicates that there are NH3 fuel-rich and fuel-lean regions in the primary combustion zone. It is presumed that unburnt NH3 is released from the fuel-rich region, while NO is released from the fuel-lean region. When diluted air enters the secondary combustion zone, unburnt NH3 is expected to react with NO through selective non-catalytic reduction (SNCR). NH3-CH4-air combustion operation tests were also performed and the results show that the increase of the NH3 fuel ratio significantly increases the NO emission, whereas it decreases the NO conversion ratio. To achieve low NOx combustion in NH3-air combustion gas turbines, it is suggested to burn large quantities of NH3 fuel and produce both rich and lean fuel mixtures in the primary combustion zone. (C) 2016 by The Combustion Institute. Published by Elsevier Inc.

The objective of this study is to explore the effects of incident shock waves on combustion downstream of a cavity flameholder. The images of flame chemiluminescence showed that the flame looks extinguished downstream of an incident shock wave, as well as OH-PLIF measurements indicated higher OH concentration upstream of the incident shock wave, while lower OH concentration downstream of the incident shock wave. The results of two-dimensional numerical simulations for reacting flows showed that the incident shock wave generates a strong recirculation zone and reverse flows, and the main flow curved around the zone, corresponding well to the flame structure seen by experiments. In addition, it was found that about 90% of the injected H2 was converted to H2O at the rear edge of the calculated wall domain when an incident shock wave was introduced, which means that cavity-stabilized flames is not extinguished by the incident shock wave but burnt out there, thus the fuel was almost fully consumed. That is to say, the incident shock wave enhances chemical reactions because the generated recirculation zone and low speed region increases residence time and increased turbulent kinetic energy enhances the mixing.

Unstable behaviors of cellular premixed flames caused by hydrodynamic and diffusive-thermal instabilities under high- and low-temperature environment were studied numerically. Unsteady reactive flow was calculated in large space, based on the compressible Navier-Stokes equation including chemical reaction. As the unburned-gas temperature became higher (lower), the growth rate increased (decreased) and the unstable range widened (narrowed), which was due to the enlargement (reduction) of the burning velocity of a planar flame. On the other hand, the normalized growth rate decreased (increased) and the normalized unstable range narrowed (widened) under the high (low) temperature conditions. This was due to the weakness (strength) of thermal-expansion effects and to the reduction (enlargement) of Zeldovich numbers. Furthermore, unstable behaviors of cellular flames, i.e. the coalescence and divide of cells, appeared owing to hydrodynamic and diffusive-thermal instabilities. We found that the burning velocity of a cellular flame changed drastically with time, and the average burning velocity of a cellular flame normalized by that of a planar flame decreased (increased) as the unburned-gas temperature became higher (lower). In addition, the burning velocity of a cellular flame depended strongly on the space size. As the space size became larger, the burning velocity increased monotonously. This was because that the long-wavelength components of disturbances played a significant role in the dynamics of cellular premixed flames.

<p>The stability limits and emission characteristics of ammonia-air diffusion flames stabilized by swirl were investigated using a 50 kW micro gas turbine burner. Two liners were used, a cylindrical glass liner and a tapered gas turbine (GT) model liner. The flames were stabilized at equivalence ratios of 0.7 to 1.4 at inlet gas temperature of 298 K. The GT model liner resulted to a slightly wider stability range and lower unburnt ammonia emission. For both liners, ammonia emission increased while NOx emission decreased with equivalence ratio. The sum of NOx and ammonia emissions were found to be minimal at stoichiometric condition.</p>

A demonstration test with the aim to show the potential of ammonia-fired power plant is planned using a micro gas turbine. 50kW class turbine system firing kerosene is selected as a base model. A standard combustor is replaced by a prototype combustor which enables a bi fuel supply of kerosene and ammonia gas. Diffusion combustion is employed in the prototype combustor due to its flame stability. Demonstration test firing ammonia gas was achieved using a new facility of large amount of ammonia supply. The gas turbine started firing kerosene and increased its electric power output. After achievement of stable power output, ammonia gas was started to be supplied and its flow rate increased gradually. 41.8kW power output was achieved by firing ammonia gas only. Ammonia gas supply increases NOx in the exhaust gas dramatically. However post-combustion clean-up of the exhaust gas via Selective Catalytic Reduction can reduce NOx successfully.

Experiments and numerical simulations were conducted to explore the interaction between an incident shockwave and a flame-holding region downstream of a ramp injector. It was found that when an incident shockwave was introduced downstream of the ramp injector, a flame-holding region with red chemiluminescence accompanying high-temperature water vapor formed, while no apparent flame-holding region was observed without it. OH-PLIF measurements were also performed. The concentration of OH radical increased downstream of the incident shockwave, indicating that combustion was enhanced in the downstream region. Results of wall pressure measurements and numerical simulations indicate that the pressure increase and enhanced mixing contribute to flame-holding downstream of the incident shockwave. The mainstream was compressed downstream of the shockwave, resulting in an increased reaction rate between the hydrogen and mainstream air, as well as enhanced mixing due to baroclinic torque generated by the pressure gradient.

Ammonia has promising features as a carbon-free fuel without greenhouse gas emission. However, due to its low combustion intensity, the combustion characteristics of ammonia have been rarely investigated. The design of ammonia based industrial applications requires the development of effective turbulent ammonia combustion models. Thus, a study of the flame stretch effect in ammonia/air premixed combustion is necessary. The objective of this research was to study the ammonia flame extinction stretch rate both experimentally and numerically and to investigate the effects of pressure on its extinction characteristics. Experiments were conducted using a counterflow flame burner. Numerical simulations were run on CHEMKIN-PRO, using the PREMIX opposed flow model, for various detailed chemistry mechanisms. The effects of pressure on the extinction stretch rate of ammonia/air premixed flame were compared with that of methane/air flame, and the effects of pressure on the detailed reaction paths were clarified. It was found that the extinction stretch rate of ammonia/air flame is low compared with that of methane/air flame, as could be expected from its low laminar burning velocity. However, the increase of extinction stretch rate with pressure was found to be greater in the case of ammonia/air flame. From detailed chemistry analysis, it was found that the different dependence on pressure of the reaction path of the two fuels could explain this difference. Indeed, the heat release process and flame strength are affected by a greater dependence on pressure of the reactions contributing the most to heat released in the case of methane/air flame. For methane/air flame, CH3 is consumed in the main heat releasing reactions, and they experience competition by the third body recombination, 2CH(3)+MC2H6+M at high pressure. In the case of ammonia/air flame, the heat release process is mainly related to NH3+OH reversible arrow NH2+H2O, NH2+OH reversible arrow NH+H2O and NH2+NO reversible arrow N-2+H2O, which remain preponderant when pressure increases. Thus, the decrease of characteristic reaction time with pressure was found to be greater in the case of ammonia/air flame, explaining a larger increase in extinction stretch rate when pressure increases.

Statistical flame front structure of turbulent premixed flames at high pressure up to 1.0 MPa was measured on a nozzle-type Bunsen burner with OH-PLIF technique. Turbulent burning velocity, flame surface density and flame brush thickness, as well as the local curvature and radius of curvature were derived from the experimental OH-PLIF images. Turbulence-flame interaction was analyzed based on the geometric parameters combined with laminar flame properties and turbulence length scales. Results show that the flame wrinkles at high pressure are dominated by small scale cusps superimposed with large scale flame branches which is a general characteristic of the turbulent premixed flames at high pressure. S-T/S-L increases remarkably with u'/S-L, and the influence of elevated pressure on S-T/S-L is significant. This is mainly due to the increase of flame front area caused by the turbulence wrinkling. Flame surface density significantly increases with the increase of pressure indicating that there is a large amount of fine cusps and small wrinkles in the flame front at high pressure. This would be due to the enhancement of the flame instability represented by effective Lewis number Le(eff) and flame intrinsic instability scale l(i). With the increase of turbulence intensity, the Sigma at high pressure increases while slightly decreases at normal pressure. The most frequent length scale of the flame front moves to smaller value and the possibility increases with the increase of u'/S-L for all pressures. The effect of flame intrinsic instability on finer flame front at high pressure is mainly on the formation of a large number of convex structures which enlarge the effective contact surface between flame front and unburned reactants, resulting in the increase of S-T/S-L. (C) 2015 Elsevier Inc. All rights reserved.

Ammonia is expected to be useful not only as a hydrogen-energy carrier but also as a carbon-free fuel. In order to design an ammonia fueled combustor, fundamental flame characteristics of ammonia must be understood. However, knowledge of the characteristics of ammonia/air flames, especially at the high pressures, has been insufficient. In this study, the unstretched laminar burning velocity and the Markstein length of ammonia/air premixed flames at various pressures up to 0.5 MPa were experimentally clarified for the first time. Spherically propagating premixed flames, which propagate in a constant volume combustion chamber, were observed using high-speed schlieren photography. Results indicate that the maximum value of unstretched laminar burning velocities is less than 7 cm/s within the examined conditions and is lower than those of hydrocarbon flames. The unstretched laminar burning velocity decreases with the increase in the initial mixture pressure, tendency being the same as that of hydrocarbon flames. The burned gas Markstein length increases with the increase in the equivalence ratio, the tendency being the same as that of hydrogen/air flames and methane/air flames. The burned gas Markstein lengths at 0.1 MPa are higher than those at 0.3 MPa and 0.5 MPa. However, the values of burned gas Markstein length at 0.3 MPa and 0.5 MPa are almost the same. In addition, numerical simulations using CHEMKIN-PRO with five detailed reaction mechanisms which are presently applicable for the ammonia/ air combustion were also conducted. However, qualitative predictions of unstretched laminar burning velocity using those reaction mechanisms are inaccurate. Thus, further improvements of reaction mechanisms are essential for application of ammonia/air premixed flames. (C) 2015 Elsevier Ltd. All rights reserved.

Ammonia shows promise not only as a hydrogen-energy carrier but also as a carbon-free fuel. However, combustion intensity of ammonia must be improved to enable its application to practical combustors. In order to achieve this, hydrogen-added ammonia/air flames were experimentally and numerically investigated at elevated pressures up to 0.5 MPa. The hydrogen ratio, which is defined as the hydrogen concentration in the fuel mixture, was varied from 0 to 1.0. The unstretched laminar burning velocity and Markstein length of spherically propagating laminar flames were experimentally evaluated. The results showed that, unstretched laminar burning velocity increases non-linearly with an increase in the hydrogen ratio. The Markstein length varies non-monotonically with an increase in the hydrogen ratio. The unstretched laminar burning velocity, and the Markstein length decrease with an increase in the initial mixture pressure. Although the decrease in the Markstein length is larger when the initial mixture pressure increases from 0.1 to 0.3 MPa, the values of Markstein lengths at 0.5 MPa are almost the same as those at 0.3 MPa. Copyright (C) 2015, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.

In premixed turbulent combustion, flame surface density (FSD) is a key parameter and can be used to estimate the system reaction rates. Even though laser diagnostic technics (Mie scattering or OH/CH-PLIF) on flame front provided very useful information of flame front wrinkles, the measurement is limited to a plane on which wrinkles in the third direction is unavailable. In this study, the estimation of 3D FSD (Sigma) and the global fuel consumption rate (W) from planar measurements of 2D FSD (Sigma(2D)) was conducted on a Bunsen-type burner fueled with methane/air mixture at the equivalence ratio of 0.9. Assuming symmetry of the mean flow, five different models designated as Method 1 to Method 5 (M1 to M5) based on different additional assumptions were utilized. M1 connected the 2D to 3D FSD with a typical value of 0.69. M2 and M3 were based on isotropic flame front normal vector distribution and identical phi and theta distribution which designate the direction angle of front normal in 3D space and the measurement plane. M4 and M5 assumed that normal vector fluctuation intensity of transverse direction was similar with x or y direction on the measurement plane, respectively. W was also obtained by integrating the 1: within the flame domain and the flame stretch factor, I-0 was evaluated based on fractal analysis of the 2D measurements. For all methods, the results are satisfying. Results of M1 indicate that a typical direction cosine value of 0.69 is valid for the turbulent Bunsen flame in this study and the satisfied W estimation under higher turbulent intensities is provided. Results of M2 are relatively rough for overestimating W by about 40% under most conditions because of its intrinsic deficiency of the 1/&lt; cos phi &gt;(s), evaluation. M3 based on the assumed identical cosine value of mean direction angle of 3D and 2D flame front presented by 4) and 0 gave rather good estimation as M4. M4 and M5 provide the best evaluation of W, absolute error within 17% except low turbulence conditions (u'/S-L approximate to 0.2 and 0.4) of M4, by the normal vector fluctuation analysis. 2D data, as expected, underestimates W. Better W can be obtained considering the flame stretch factor, I-0. (C) 2015 The Combustion Institute. Published by Elsevier Inc. All rights reserved.

Ammonia is a carbon-free fuel and its application to internal combustion engines is expected. However, few studies on ammonia flames, especially at high pressures, have been carried out because ammonia has not been considered to be a fuel owing to its lower combustion intensity. Most NOx, which is formed by ammonia combustion, is considered to be the fuel NOx. The objectives of this study were to investigate the fundamental characteristics of NOx experimentally, such as NO emission and chemiluminescence of ammonia/air flames not only at the atmospheric pressure but also under high pressures and to explore NO formation/reduction mechanisms using numerical simulation. Experiments were carried out using a nozzle-type burner. NH₂ ammonia α band spectra were observed, and it was clarified that the color of ammonia flame is mainly determined by the NH₂ ammonia α band and H₂O spectra. Burned gas was sampled from ammonia flame stabilized at the burner. The mole fraction of NO decreased with the increase in equivalence ratio at atmospheric pressure. Reaction flow analysis was performed, and it was clarified that the decrease in the mole fraction of NO for rich mixtures was caused by NH_<i>i</i> (<i>i</i> = 2, 1, 0). High pressure experiments were performed using a high pressure combustion facility for stoichiometric ammonia flame. Consequently, the decrease in the mole fraction of NO was experimentally observed and its tendency was found to qualitatively agree with the results of the numerical simulation. It was clarified that the third body reaction of OH + H + M ⇔ H₂O + M plays an important role in the reduction of the mole fraction of NO at the high pressure.

The intrinsic instability of three-dimensional (3D) premixed flames under low-and high-temperature conditions was numerically treated to study the effects of unburned-gas temperature on hydrodynamic and diffusive-thermal instabilities. Superimposing a sinusoidal disturbance with sufficiently small amplitude on a planar flame, we obtained the relation between the growth rate and wave number, i.e., the dispersion relation. As the unburned-gas temperature became lower, the growth rate decreased and the unstable range narrowed due to the decrease of the burning velocity of a planar flame. At sufficiently small wave numbers, the obtained numerical results were consistent with the theoretical solutions. When the Lewis number was small, we obtained a large growth rate and wide unstable range due to diffusive-thermal effects. The growth rate and wave number were normalized by the burning velocity of a planar flame and preheat zone thickness. The normalized growth rate increased and the normalized unstable range widened with a decrease of unburned-gas temperature. This was because thermal-expansion effects became stronger owing to the increase of the temperature ratio of burned and unburned gases. To elucidate the characteristics of cellular flames generated by intrinsic instability, we superimposed a disturbance with the critical wave number corresponding to the maximum growth rate, i.e., the linearly most unstable wave number. The superimposed disturbance evolved, and a hexagonal cellular flame formed. The behavior of cellular flames became stronger as the unburned-gas temperature became lower, even though the growth rate decreased. The burning velocity of a cellular flame normalized by that of a planar flame increased due to the strength of thermal-expansion effects.

In order to investigate oxyfuel combustion characteristics of typical composition of coal gasification syngas connected to CCS systems. Instantaneous flame front structure of turbulent premixed flames of CO/H-2/O-2/CO2 mixtures which represent syngas oxyfuel combustion was quantitatively studied comparing with CH4/air and syngas/air flames by using a nozzle-type Bunsen burner. Hot-wire anemometer and OH-FLIP were used to measure the turbulent flow and detect the instantaneous flame front structure, respectively. Image processing and statistical analyzing were performed using the Matlab Software. Flame surface density, mean progress variable, local curvature radius, mean flame volume, and flame thickness, were obtained. Results show that turbulent premixed flames of syngas possess wrinkled flame front structure which is a general feature of turbulent premixed flames. Flame surface density for the CO/H-2/O-2/CO2 flame is much larger than that of CO/H-2/O-2/air and CH4/air flames. This is mainly caused by the smaller flame intrinsic instability scale, which would lead to smaller scales and less flame passivity response to turbulence presented by Markstain length, which reduce the local flame stretch against turbulence vortex. Peak value of Possibility Density Function (PDF) distribution of local curvature radius, R, for CO/H-2/ O-2/CO2 flames is larger than those of CO/H-2/O-2/air and CHI/air flames at both positive and negative side and the corresponding R of absolute peak PDF is the smallest. This demonstrates that the most frequent scale is the smallest for CO/H-2/O-2/CO2 flames. Mean flame volume of CO/H-2/O-2/CO2 flame is smaller than that of CHilair flame even smaller than that of CO/H-2/O-2/air flame. This would be due to the lower flame height and smaller flame wrinkles. Copyright (c) 2014, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.

Instantaneous flame front structure of turbulent premixed CH4/H-2/air flames (hydrogen fraction of 0%, 5%, 10% and 20% by mole fraction) was investigated quantitatively using a nozzle-type Bunsen burner. Hot wire anemometer and OH-PLIF were used to measure the turbulent flow and detect the instantaneous flame front structure, respectively. Turbulent burning velocity, S-T, flame surface density, Sigma, and mean flame volume, V-f, were calculated by processing the OH-PLIF images. Results show that the flame front structures of the turbulent premixed flames are the wrinkled flame front and it becomes much finer with the increase of turbulence intensity as well as hydrogen fraction. The value of S-T/S-L significantly increases with the increase of u'/S-L and it slightly increases with the increase of hydrogen fraction. Flame surface density profile are symmetric and gives its maximum value at about &lt; c &gt; = 0.5. Hydrogen addition slightly enhances the I and the tendency is more obvious under higher turbulence intensity. The decrease of Sigma with the increase of turbulence intensity is mainly due to the effect of flame volume. The mean flame volume of flame region obviously increases with the increase of turbulence intensity within the experimental range due to the increase in depth of the large scale flame wrinkles and flame height. Hydrogen addition is not a predominant factor within the hydrogen fraction range in this study. (C) 2013 Elsevier Inc. All rights reserved.

Instantaneous flame front structure of turbulent premixed CH4/H-2/air flames (hydrogen fraction of 0%, 5%, 10% and 20% by mole fraction) was investigated quantitatively using a nozzle-type Bunsen burner. Hot wire anemometer and OH-PLIF were used to measure the turbulent flow and detect the instantaneous flame front structure, respectively. Turbulent burning velocity, S-T, flame surface density, Sigma, and mean flame volume, V-f, were calculated by processing the OH-PLIF images. Results show that the flame front structures of the turbulent premixed flames are the wrinkled flame front and it becomes much finer with the increase of turbulence intensity as well as hydrogen fraction. The value of S-T/S-L significantly increases with the increase of u'/S-L and it slightly increases with the increase of hydrogen fraction. Flame surface density profile are symmetric and gives its maximum value at about &lt; c &gt; = 0.5. Hydrogen addition slightly enhances the I and the tendency is more obvious under higher turbulence intensity. The decrease of Sigma with the increase of turbulence intensity is mainly due to the effect of flame volume. The mean flame volume of flame region obviously increases with the increase of turbulence intensity within the experimental range due to the increase in depth of the large scale flame wrinkles and flame height. Hydrogen addition is not a predominant factor within the hydrogen fraction range in this study. (C) 2013 Elsevier Inc. All rights reserved.

Instantaneous flame front structure of syngas turbulent premixed flames including the local radius of curvature, the characteristic radius of curvature, the fractal inner cutoff scale and the local flame angle were derived from the experimental OH-PLIF images. The CO/H-2/CO2/air flames as a model of syngas/air combustion were investigated at pressure of 0.5 MPa and compared to that of CH4/air flames. The convex and concave structures of the flame front were detected and statistical analysis including the PDF and ADF of the local radius of curvature and local flame angle were conducted. Results show that the flame front of turbulent premixed flames at high pressure is a wrinkled flame front with small scale convex and concave structures superimposed with large scale flame branches. The convex structures are much more frequent than the concave ones on flame front which reflects a general characteristic of the turbulent premixed flames at high pressure. The syngas flames possess much wrinkled flame front with much smaller fine cusps structure compared to that of CH4/air flames and the main difference is on the convex structure. The effect of turbulence on the general wrinkled scale of flame front is much weaker than that of the smallest wrinkled scale. The general wrinkled scale is mainly dominated by the turbulence vortex scale, while, the smallest wrinkled scale is strongly affected by the flame intrinsic instability. The effect of flame intrinsic instability on flame front of turbulent premixed flame is mainly on the formation of a large number of convex structure propagating to the unburned reactants and enlarge the effective contact surface between flame front and unburned reactants. (C) 2013 The Combustion Institute. Published by Elsevier Inc. All rights reserved.

Instantaneous flame front structures of the turbulent premixed flames of syngas/air and CH4/air mixtures were investigated using OH-PLIF technique at high pressure up to 1.0MPa, through which the turbulent burning velocities were derived and correlated with the turbulence intensity. Results show that both syngas/air and CH4/air mixtures, ST/SL increases remarkably with the increase of u'/SL particularly in the weak turbulence region. For the syngas/air mixture, the intensity of flame front wrinkle is promoted with the increase of hydrogen fraction in the syngas due to the increased preferential diffusive-thermal instability. Compared to CH4/air mixture, the syngas flames possess much wrinkled flame front with much smaller fine cusps structure, and with increasing u'/SL, the rate of the increase of ST/SL for the syngas/air mixtures is more significant than that of CH4/air mixtures. This demonstrates that the increase of flame front area due to turbulence wrinkling is promoted by flame intrinsic instability for syngas/air mixtures. The values of ST/SL for all mixtures increase with the increase of pressure because of the decrease of flame thickness which promotes the hydrodynamic instability. A general correlation of turbulent burning velocity for the syngas/air and CH4/air mixtures was obtained in the form of ST/SL∝a[(P/P0)(u'/SL)]n. © 2013 Elsevier Inc.

We studied numerically the intrinsic instability of high-temperature premixed flames, where the burned-gas temperature was constant, under the conditions of constant density and constant pressure in the unburned gas. A sinusoidal disturbance with sufficiently small amplitude was superimposed on a planar flame to obtain the relation between the growth rate and wave number, i.e. the dispersion relation. As the unburned-gas temperature became higher, the growth rate increased and the unstable range widened, which was due to the increase of the burning velocity of a planar flame. In sufficiently small wave-number range, the obtained numerical results were consistent with the theoretical solutions. When the growth rate and wave number were normalized, the same dispersion relations were found under the conditions both of constant density and constant pressure in the unburned gas. The normalized growth rate decreased with an increase of the unburned-gas temperature, and the normalized unstable range narrowed. This was because that the thermal-expansion effects became weaker owing to the decrease of the difference in temperature between the burned and unburned gases. To clarify the characteristics of cellular flames induced by intrinsic instability, we superimposed a disturbance with the critical wavelength. The superimposed disturbance evolved, and a cellular-shaped flame front formed. The behavior of cellular flames became milder as the unburned-gas temperature became higher, even though the growth rate increased. The burning velocity of a cellular flame normalized by that of a planar flame decreased, which was due to the weakness of the thermal-expansion effects and diffusive-thermal effects. Moreover, the burning velocity of a cellular flame increased monotonously as the length of computational domain became larger, and the dependence of burning velocity on domain length became weaker with an increase of the unburned-gas temperature.

Flame front structure of turbulent premixed CH4/H 2/air flames at various hydrogen fractions was investigated with OH-PLIF technique. A nozzle-type burner was used to achieve the stabilized turbulent premixed flames. Hot-wire anemometer measurement and OH-PLIF observation were performed to measure the turbulent flow and detect the instantaneous flame front structure, respectively. The hydrogen fractions of 0%, 5%, 10% and 20% were studied. Results show that the flame front structures of the turbulent premixed flames are wrinkled flame front with small scale convex and concave structures compared to that of the laminar-flame front. The wrinkle intensity of flame front is promoted with the increase of turbulence intensity as well as hydrogen fraction. Hydrogen addition promotes the flame intrinsic instability which leads to the active response of laminar flame to turbulence and results in the much more wrinkled flame front structure. The value of S T/SL increases monotonically with the increase of u′/SL and hydrogen fraction. The increase of S T/SL with the increase of hydrogen fraction is mainly attributed to the diffusive-thermal instability effects represented by the effective Lewis number, Leeff. A general correlation between S T/SL and u′/SL is provided from the experimental data fitting in the form of ST/SL ∞ a(u′/SL)n, and the exponent, n, gives the constant value of 0.35 for all conditions and at various hydrogen fractions. Copyright © 2013, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.

Flame front structure characteristics of turbulent premixed flames including the local radius of curvature, fractal inner cutoff scale and local flame angle were calculated from the experimental OH-PLIF images of CH4/air flames and those diluted with CO2 and superheated H2O at 0.5 MPa and 573 K. The convex and concave structures of the flame front were detected and statistical analysis including PDF and ADF of the local radius of curvature and the local flame angle was conducted. Results showed that the flame front of turbulent premixed flames at high pressure and high temperature is a wrinkled flame front with small scale convex and concave cusps superimposed on large scale branches, while the whole flame front tends to be convex to the unburned mixture. The effect of EGR gas dilution on the flame front structure of turbulent premixed flames is dominated by CO2 rather than H2O dilution. The quantitative flame front characteristics reveal the complicated effect of CO2 dilution on turbulent premixed flames characteristics observed in our previous research. In the case of CO2 dilution, some local wrinkled structures become sharp and propagate deep into the burned mixture, resulting in a decrease of the fractal inner cutoff scale, the formation of a thick flame brush and the enlargement of the mean flame volume, while the overall wrinkled scale increases significantly due to the suppression of the concave structure with CO2 dilution, leading to a decrease of the turbulent flame front area, and subsequently to a decrease of ST/SL. (C) 2012 The Combustion Institute. Published by Elsevier Inc. All rights reserved.

Characteristics of turbulent premixed flames of a CO/H-2/O-2 mixture highly diluted with CO2 (CO/H-2/CO2/O-2 flame) at high pressures up to 1.0 MPa were experimentally investigated. The CO/H-2 ratio, equivalence ratio and CO2 mole fraction were determined considering the typical composition of coal gasification syngas, laminar burning velocity, adiabatic flame temperature and stoichiometry for IGCC gas-turbine combustors connected to CCS systems. OH-PLIF and flame radiation measurement were performed for Bunsen-type flames stabilized in a high-pressure chamber. Using OH-PLIF images, flame surface density, mean volume of turbulent flame regions and turbulent burning velocity were calculated and compared with those for CH4/air flames and model coal gasification syngas flames burnt with air (CO/H-2/CO2/air flame). The flame surface density for the CO/H-2/CO2/O-2 flames was much greater than that for the CH4/air flames, even greater than that of the CO/H-2/CO2/air flames, presumably due to less flame passivity against turbulent vortex motion caused by smaller Markstein length and smaller scales of flame wrinkles at high pressure. The mean volume of the turbulent flame region for the CO/H-2/CO2/O-2 flames was close to that of CO/H-2/CO2/air flames, while much smaller than that of the CH4/air flames, which was also explicable based on the Markstein length effects on turbulent flames at high pressure. ST/SL of the model syngas flames was larger than that of the CH4/air flames and it was noted that the difference in turbulence Reynolds number caused by smaller kinematic viscosity of the CO/H-2/CO2/O-2 mixture should be considered to understand the ST/SL characteristics. Total radiation intensity of the CO/H-2/CO2/O-2 flame was about 1.6 times stronger than that of CH4/air flames due to the very high CO2 concentration, CO2 being a highly radiative species, indicating very high heat-load for gas-turbine combustors that should be considered for combustor design. (C) 2012 The Combustion Institute. Published by Elsevier Inc. All rights reserved.

Flame front structure characteristics of turbulent premixed flames including the local radius of curvature, fractal inner cutoff scale and local flame angle were calculated from the experimental OH-PLIF images of CH4/air flames and those diluted with CO2 and superheated H2O at 0.5 MPa and 573 K. The convex and concave structures of the flame front were detected and statistical analysis including PDF and ADF of the local radius of curvature and the local flame angle was conducted. Results showed that the flame front of turbulent premixed flames at high pressure and high temperature is a wrinkled flame front with small scale convex and concave cusps superimposed on large scale branches, while the whole flame front tends to be convex to the unburned mixture. The effect of EGR gas dilution on the flame front structure of turbulent premixed flames is dominated by CO2 rather than H2O dilution. The quantitative flame front characteristics reveal the complicated effect of CO2 dilution on turbulent premixed flames characteristics observed in our previous research. In the case of CO2 dilution, some local wrinkled structures become sharp and propagate deep into the burned mixture, resulting in a decrease of the fractal inner cutoff scale, the formation of a thick flame brush and the enlargement of the mean flame volume, while the overall wrinkled scale increases significantly due to the suppression of the concave structure with CO2 dilution, leading to a decrease of the turbulent flame front area, and subsequently to a decrease of ST/SL. (C) 2012 The Combustion Institute. Published by Elsevier Inc. All rights reserved.

Laminar burning velocities of CO-H-2-CO2-O-2 flames were measured by using the outwardly spherical propagating flame method. The effect of large fraction of hydrogen and CO2 on flame radiation, chemical reaction, and intrinsic flame instability were investigated. Results show that the laminar burning velocities of CO-H-2-CO2-O-2 mixtures increase with the increase of hydrogen fraction and decrease with the increase of CO2 fraction. The effect of hydrogen fraction on laminar burning velocity is weakened with the increase of CO2 fraction. The Davis et al. syngas mechanism can be used to calculate the syngas oxyfuel combustion at low hydrogen and CO2 fraction but needs to be revised and validated by additional experimental data for the high hydrogen and CO2 fraction. The radiation of syngas oxyfuel flame is much stronger than that of syngas air and hydrocarbons air flame due to the existence of large amount of CO2 in the flame. The CO2 acts as an inhibitor in the reaction process of syngas oxyfuel combustion due to the competition of the reactions of H + O-2 = O + OH, CO + OH = CO2 + H and H + O-2(+M) = HO2(+M) on H radical. Flame cellular structure is promoted with the increase of hydrogen fraction and is suppressed with the increase of CO2 fraction due to the combination effect of hydrodynamic and thermal-diffusive instability. Copyright (C) 2012, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.

JFE Engineering has developed the advanced stoker-type incineration system "JFE Hyper 21 Stoker System". It is equipped with some new technologies to meet requirement to a municipal solid waste incineration system. Main technology of this system is High-temperature Air Combustion Technology (HiCOT), which realizes low excess air combustion. This paper describes the numerical simulation of combustion behavior in JFE Hyper 21 Stoker System. Firstly, a few kinds of kinetic mechanisms of combustion reaction were examined to select suitable one for a large-scaled numerical simulation of HiCOT. After we selected Yetter mechanism, two kinds of three-dimensional models, HiCOT zone model and entire incinerator model, were made and numerical simulation was performed by using these models. In this paper, we show the effectivity of these models that clarify the combustion behavior in JFE Hyper 21 Stoker System using HiCOT. © 2012 The Japan Society of Mechanical Engineers.

Combustion experiments were performed for polyethylene (PE) in stagnation-point flow to investigate the fundamental characteristics of PE in high-temperature air combustion (HiTAC). Air diluted by nitrogen, water vapor, and carbon dioxide was tested as the oxidizer, and the oxidizer temperature was varied from 300 K to 773 K in order to investigate the effect of the dilution and temperature on the regression rate, the extinction limit, and the sooting limit of PE. The kinetic parameters of PE pyrolysis under the combustion conditions were estimated from the regression rate experimental data using a new method that combines experiments and numerical simulations. It was found that the previously reported kinetic parameters obtained with thermal gravimetric analysis (TGA) were much smaller than those obtained in this study. It was also found that the kinetic parameters varied when a different dilution gas was used, indicating that PE pyrolysis is affected by the composition of oxidizer.

The effects of the unburned-gas temperature and Lewis number on the intrinsic instability of high-temperature premixed flames under the constant-enthalpy conditions were investigated by two-dimensional unsteady calculations of reactive flows. A sinusoidal disturbance with sufficiently small amplitude was superimposed on a planar flame to obtain the relation between the growth rate and wave number, i.e. the dispersion relation. As the unburned-gas temperature became higher, the growth rate increased and the unstable range widened. This was due to the increase of the burning velocity of a planar flame. In addition, the obtained numerical results were consistent with the theoretical solutions in small wave-number region. As the Lewis number became smaller (larger), the growth rate increased (decreased) and the unstable range widened (narrowed), which was due to diffusive-thermal effects. The dispersion relation yielded the linearly most unstable wave number, i.e. the critical wave number. The critical wave number increased as the unburned-gas temperature became higher. Thus, the critical wavelength shrank, and then the cell size shrank. To clarify the characteristics of cellular flames induced by intrinsic instability, a finite disturbance with the critical wavelength was superimposed. The superimposed disturbance evolved, and a cellular-shaped front formed. In all Lewis numbers, the behavior of cellular flames became milder as the unburned-gas temperature became higher, even though the growth rate increased. The normalized burning velocities of cellular flames decreased monotonously. This was because that the thermal-expansion effects became weaker owing to the decrease of the difference in temperature between the burned and unburned gases, which was generated by the conditions of constant enthalpy, i.e. constant burned-gas temperature.

Combustion experiments were performed for polyethylene (PE) in stagnation-point flow to investigate the fundamental characteristics of PE in high-temperature air combustion (HiTAC). Air diluted by nitrogen, water vapor, and carbon dioxide was tested as the oxidizer, and the oxidizer temperature was varied from 300 to 773 K, to investigate the effect of the dilution and temperature on the regression rate, the extinction limit, and the sooting limit of PE. The kinetic parameters of PE pyrolysis under the combustion conditions were estimated from the regression rate experimental data using a new method combining experiments and numerical simulations. It was found that the previously reported kinetic parameters obtained with thermal gravimetric analysis (TGA) were much smaller than those obtained in this study, and that the kinetic parameters of pyrolysis were affected by dilution.

The interaction between an incident shock wave and a transverse jet flow for mixing and combustion in a supersonic airstream was investigated experimentally and numerically. NO planar laser induced fluorescence (NO-PLIF) and particle imaging velocimetry (PIV) for non-reactive flows and three-dimensional reactive/non-reactive numerical simulations were conducted to examine the effect of the incident shock wave on the three-dimensional flow structure and mixing mechanism between the airstream and the injected gas downstream of the injection slot. Results of NO-PLIF measurement and numerical simulation showed that, in the case without the incident shock wave, injected gas is seldom seen in the recirculation zone just downstream of the injection slot, while the injected gas with higher concentration is almost uniformly distributed in the recirculation zone when the incident shock wave is introduced downstream of the injection slot. Moreover, it was shown by the numerical simulations that the profiles of the local equivalence ratio is in the combustible range due to the enhanced entrainment of the airstream when the incident shock wave is introduced downstream of the injection slot. A large-scale recirculation in the direction parallel to the wall is generated by the three-dimensional flow effects, which enhances the mixing and extends the residence time in the recirculation zone in the case with incident shock wave downstream of the injection slot, the recirculation flow being confirmed successfully by PIV measurements as well. The results of three-dimensional reactive numerical simulations were in good agreement with the experimental flame-holding characteristics at a lower total temperature, which showed that flame-holding can be attained only when the incident shock wave was introduced downstream of the injection slot, confirming that the formation of three-dimensional and large-scale recirculation flow downstream of the injection slot enlarges the recirculation zone and enhances the mixing to produce the conditions for robust flame-holding. (C) 2010 The Combustion Institute. Published by Elsevier Inc. All rights reserved.

Experimental study on turbulent premixed flames for a CO/H-2/CO2/air mixture as a model of coal gasification syngas in a high pressure environment was performed up to 1.0 MPa. CH4/air flames with laminar burning velocity and adiabatic flame temperature identical to those of CO/H-2/CO2/air mixture were also employed, and then the flame structure and turbulent burning velocity analyzed using OH-PLIF as well as flame radiation characteristics were compared. Results showed that in the case of the CO/H-2/CO2/air flame, very fine cusps were seen, these small cusps being generated on a large scale wrinkled flame front even for low u'/S-L. Flame surface density of the CO/H-2/CO2/air flame was higher than that of the CH4/air flame, and the mean volume of the turbulent flame region of CO/H-2/CO2/air flame was smaller than that of the CH4/air flames at high pressure. Bending of S-T/S-L with u'/S-L for the CO/H-2/CO2/air flame was not observed in this experiment, while such bending was clearly seen for the CH4/air flame. The difference in the bending characteristics can be explained based on the scale relation between the smallest scale of flame wrinkles represented by the fractal inner cutoff and characteristic flame instability scale. The spectrum of flame radiation for the CO/H-2/CO2/air flame at 0.5 MPa showed continuous emission in the visible range of the wavelength, while that for CH4/air flames for the equivalence ratio of unity showed very intense H2O emission in the IR range. When the flames with identical adiabatic flame temperature for the CO/H-2/CO2/air flame and the CH4/air flame were compared, the total radiation intensity of the former mixture was observed to be strong. The intense emission from the CO/H-2/CO2/air flame is thought to be one of the mechanisms which generate very fine flame cusps as enhanced cellar instability for smaller Lewis number. (C) 2010 The Combustion Institute. Published by Elsevier Inc. All rights reserved.

Two-dimensional unsteady calculations of reactive flow on the basis of the compressible Navier-Stokes equation were performed to elucidate the effects of radiation on the dynamic behavior of cellular premixed flames generated by intrinsic instability. The disturbance superimposed on a planar flame evolved owing to hydrodynamic and diffusive-thermal effects, and then the cellular flame front formed. The dynamic behavior of cellular flames appeared at low Lewis numbers, and it became stronger as radiative heat loss increased. The island of the unburned gas surrounded by the burned gas was observed in non-adiabatic flames with low Lewis numbers. The average cell size of the non-adiabatic flame was slightly small compared with the adiabatic flame, even though the critical wavelength of the former flame was larger than that of the latter flame. This indicates that the radiation has a pronounced influence on the dynamics of premixed flames with low Lewis numbers. Owing to the dynamic behavior, the burning velocity of cellular flames changed drastically with time, which was due principally to the combination and division of cells. The average burning velocity of the non-adiabatic cellular flame was somewhat small compared with the adiabatic cellular flame. This is because that the burning velocity of planar flames decreases owing to radiation and that the dynamic behavior of cellular flames becomes stronger. In addition, the average burning velocity became larger monotonically as the length of computational domain increased. The reason is that the long wavelength components of disturbances play a significant role in the front shape of cellular flames. (C) 2010 The Combustion Institute. Published by Elsevier Inc. All rights reserved.

An experimental study on CH4/air propagating flames in a packed pebble bed was performed over a wide range of ambient pressures and flow velocities to clarify the combustion characteristics in a packed bed at high pressure. It focused on the turbulent flame propagation in cases that the flame thickness was sufficiently smaller than the void scale of the packed bed at high pressure. The flame speed was successfully measured at pressures of 0.1-1.0 MPa and flow velocities of 5-130 cm/s using a high-speed video camera, and then the relationships between the flame speed and pressure, flow velocity, pebble diameter and pebble Reynolds number, Re-p, were examined. Turbulence measurements at high pressure were also conducted using a 2-D pseudo packed pebble bed to investigate the turbulence intensities in the flow channels and the estimated turbulent flame structure. Results of flame propagation experiments showed that the flame speed in a packed bed had a minimum value at Re-p congruent to 150 and rapidly increased with the increase of the flow velocity and pressure. Results obtained in the turbulence measurements indicated that this change of the flame propagation mode was caused by the transition from laminar to turbulent flow when the pressure increased. A significant correlation was seen between the flame speed in a packed bed, S-r, at high pressure and the turbulent burning velocity, S-T, estimated from the results of turbulence measurements and previous studies on turbulent premixed Bunsen flames by Kobayashi et al., indicating the similarity between the turbulent premixed combustion with and without packed pebbles at high pressure. Flame extinction during propagation at high pressure was also seen as well as at ordinary pressure, the extinction mechanism being explained based on the competition between the heat release in turbulent flames and turbulent heat transfer to the pebbles. (C) 2010 The Combustion Institute. Published by Elsevier Inc. All rights reserved.

The enhancement and hysteresis behavior of the burning rate of single droplet combustion in the presence of airstream oscillation observed in previously performed microgravity experiments at elevated pressure up to 1.0 MPa were numerically investigated. Excellent agreement with the experimental results was obtained and the mechanisms of these phenomena were examined based on precise numerical data on instantaneous droplet diameter variations corresponding to the unsteady airstream velocity, flow fields around the droplet, and flame movement during combustion. Results show that, depending on the oscillation Reynolds number, which is a function of pressure, flow amplitude and droplet diameter, there are three mechanisms involved in the enhancement of burning rate. In the cases of low oscillation Reynolds numbers, a diffusion-time-delay has a significant effect on the flame front movement and thus, on heat from the flame to the droplet. In the cases of high oscillation Reynolds numbers, a vortex generated outside the droplet flame promotes the motion of the flame, especially in the wake region, and thus enhancing the droplet burning rate. In addition to these two mechanisms, the forced convection during the acceleration period of the flow oscillation causes overshooting of the droplet burning rate due to instantaneous imbalances of the airflow momentum with the Stefan flow. These three mechanisms explain the predominant role of the highest velocity of the oscillatory airstream in determination of the mean burning rate constant and droplet lifetime. Results also show that the hysteresis behavior of the burning rate is a consequence of the different responses of the flame to the deceleration period compared with the responses to the acceleration period under the existence of those three mechanisms. (C) 2009 The Combustion Institute. Published by Elsevier Inc. All rights reserved.

Numerical studies on the behaviors of combustion of 1-butanol fuel droplet at presence of upstream velocity oscillation are performed. Fuel droplet has an initial diameter of 1.25 mm and ambiance pressure and temperature are 0.4 MPa and 300 K, respectively. These conditions are those in which the microgravity experiments in literature conducted. In the excellent agreement with the experimental data, numerical results show a significant enhancement of the burning rate of droplet compared to what is predicted by quasi-steady film theory models. The mechanism of the enhancement of burning rate is clarified then by observation of a new mechanism that is named thermal-drag, TD. It is shown, depending on the amplitude and frequency of the upstream velocity oscillation, the flame in wake region of droplet can move toward the droplet surface by the force of vortex flow motions produced by the TD mechanism. It is verified that such movement of the flame is responsible for the enhancement of the burning rate and deviation of the response of the evaporation process form the predictions of the quasi-steady model. Frequency analysis of the burning rate reveals that at low frequency and amplitude the FFT diagram of the burning rate contains of only one main peak synchronies with the frequency of upstream velocity oscillation, which implies a quasi-steady response. However; at high frequency and amplitude the diagram includes of wide range of frequencies beside of the main peak that readily shows deviation from the quasi-steady conditions. In the latter, the study on the response of the combustion to the upstream velocity fluctuations in which the fluctuations contains of three wave numbers shows the amplification of the effects of low frequency fluctuations rather than that of damping of the effects of high frequency fluctuations on the evaporation processes.

Numerical analysis of CH4/O-2/H2O laminar premixed flame under various conditions of pressure, equivalence ratio and steam concentration was performed using GRI-Mech 3.0 and the mechanism proposed by Davis and Law, which consists of C1 to C6 hydrocarbons in addition to GRI-Mech 3.0. The pressure dependence of laminar burning velocity and flame structure under fuel-rich conditions was focused on. Effects of the formation of higher hydrocarbons under fuel-rich conditions were also clarified using the mechanism proposed by Davis and Law. Results showed that for extremely fuel-rich conditions, laminar burning velocity increases as pressure increases for both mechanisms. The increase of laminar burning velocity is caused by the shift of the oxidation pathway of CH3 radical from the C2 Route to the C1 Route. The formation of C3-C6 hydrocarbons has only a small effect on laminar burning velocity. Under fuel-rich conditions, super-adiabatic flame temperature (SAFT) occurs and its pressure dependency was clarified.

A 20-step reduced kinetic mechanism of ethanol, a potential sustainable energy source as a biofuel, was developed based on the detailed reaction mechanism proposed by Saxena and Williams using the Computational Singular Perturbation (CSP) method based on the Quasi-steady State Assumption (QSSA). Feasibility evaluation of the reduced kinetic mechanism for multi-dimensional flame analysis, i.e., the difference in numerical results and convergence time between the detailed reaction mechanism and the reduced kinetic mechanism, was also performed to investigate the applicability of the ethanol reduced kinetic mechanism to the development of practical combustors. To consider further industrial applications, the reduced kinetic mechanism was incorporated into the commercial computational fluid dynamics (CFD) code FLUENT 6.3.26 using the User Defined Function (UDF) code developed in the present study. Numerical results calculated with the detailed reaction mechanism and the reduced kinetic mechanism, i.e., temperature profiles, chemical species profiles and laminar burning velocities, were in good agreement for both two-dimensional premixed and non-premixed flame calculations. Convergence time using the reduced kinetic mechanism was considerably reduced compared to that using the detailed reaction mechanism, indicating the applicability and advantage of a reduced kinetic mechanism based on QSSA for multi-dimensional flame analysis. An additional reduction of the computational time was achieved by using both the reduced kinetic mechanism and In Situ Adaptive Tabulation (ISAT) solver by Pope et al.

A numerical study of the heat and mass transfer from an evaporating fuel droplet in oscillatory flow was performed. The flow was assumed to be laminar and axisymmetric, and the droplet was assumed to maintain its spherical shape during its lifetime. Based on these assumptions, the conservation equations in a general curvilinear coordinate were solved numerically. The behaviors of droplet evaporation in the oscillatory flow were investigated by analyzing the effects of flow oscillation on the evaporation process of a n-heptane fuel droplet at high pressure. The response of the time history of the square of droplet diameter and space-averaged Nusselt numbers to the main flow oscillation were investigated in frequency band of 1-75 Hz with various oscillation amplitudes. Results showed that, depending on the frequency and amplitude of the oscillation, there are different modes of response of the evaporation process to the flow oscillation. One response mode is synchronous with the main flow oscillation, and thus the quasi-steady condition is attained. Another mode is asynchronous with the flow oscillation and is highly unsteady. As for the evaporation rate, however, in all conditions is more greatly enhanced in oscillatory flow than in quiescent air. To quantify the conditions of the transition from quasi-steady to unsteady, the response of the boundary layer around the droplet surface to the flow oscillation was investigated. The results led to including the oscillation Strouhal number as a criteria for the transition. The numerical results showed that at a low Strouhal number, a quasi-steady boundary layer is formed in response to the flow oscillation, whereas by increasing the oscillation Strouhal number, the phenomena become unsteady. (C) 2009 Elsevier Inc. All rights reserved.

An experimental study was performed to investigate the flame-holding of a hydrogen jet injected into a supersonic cross-flow interacting with an incident shock wave. Pre-burned rich hydrogen or air was injected into the supersonic airflow at Mach 2.5. The injection pressure was 1.2 MPa or 1.6 MPa. The deflection angle of the shock generator was 10 degrees. Velocity profiles near the wall and of the recirculation zone around the injection slot could successfully be measured using the PTV method newly-devised in this study. The velocity profiles showed that the reattachment point of the recirculation zone downstream of the injection slot moved further downstream when the incident shock wave was introduced downstream of the injection slot. The reattachment point with the incident shock wave at the injection pressure of 1.6 MPa was not very different from that of 1.2 MPa. However, the extinction limit at the injection pressure of 1.6 MPa extended further than that of 1.2 MPa.

The dynamic behavior of premixed flames propagating in non-uniform velocity fields was investigated to assess the significance of intrinsic instability in turbulent combustion. Two-dimensional unsteady calculations of reactive flows based on the compressible Navier-Stokes equations including a one-step irreversible chemical reaction were performed. A sinusoidal disturbance was superimposed on the velocity field of the unburned gas, and its wavelength was set equal to the linearly most unstable wavelength, i.e. the critical wavelength, at the Lewis number Le = 1.0. To investigate the effects of intrinsic instability on the dynamic behavior of premixed flames, we treated two basic types of intrinsic instability, i.e. hydrodynamic instability and diffusive-thermal instability. The dynamic behavior of cellular flames generated both by intrinsic instability and by velocity disturbances was numerically simulated. When Le = 1.0, at the initial evolution, we observed a cellular flame whose cell size was equal to the wavelength of velocity disturbances. After that, cells combined together, and one large cell appeared. When Le = 0.5, on the other hand, several cells smaller than the wavelength of velocity disturbances were found, and the combination and division of cells were observed. This is because that the size of cells caused only by intrinsic instability is smaller than the wavelength of velocity disturbances. Thus, the dynamic behavior of premixed flames is drastically affected not only by velocity disturbances but also by intrinsic instability. The burning velocity of cellular flames propagating in non-uniform velocity fields was larger than that of planar flames, since cellular flames had larger surface area. The burning velocity became larger as the intensity of velocity disturbances became higher, and the dependence of the burning velocity on the intensity was strongly affected by the Lewis number. The relative significance of intrinsic instability on the evolution of turbulent premixed flames was identified by comparing the growth rate and production rate of flame fronts due to intrinsic instability and turbulence. In the region where the growth rate due to intrinsic instability was larger than that due to turbulence, the dynamic behavior was strongly affected by intrinsic instability. This region includes the combustion conditions of many industrial combustors, and then intrinsic instability plays a significant role in the characteristics of turbulent combustion.

This paper treats two-dimensional unsteady calculations of reactive flow on the basis of the compressible Navier-Stokes equation to elucidate the effects of radiation on the dynamic behavior of premixed flames. The disturbance superimposed on premixed flames evolved owing to intrinsic instability, and then the cellular-flame front formed. The dynamic behavior of cellular flames appeared at low Lewis numbers, and it became stronger as the radiative-loss parameter increased. Owing to the dynamic behavior, the burning velocity of cellular flames changed drastically with time, which was due mainly to the combination and division of cells. At low Lewis numbers, the burning velocity of the cellular flame with radiative loss was somewhat small compared with the adiabatic cellular flame. This was due to the decrease of the burning velocity of a planar flame and to the increase of surface area of a cellular flame. In addition, the average burning velocity became larger monotonically as the computational-domain length increased.

Combustion experiments on polypropylene (PP) and polyethylene terephethalate (PET) in stagnation-point flow were performed to investigate the fundamental characteristics of these polymers in high-temperature air combustion (HiTAC). Numerical study was also performed to estimate the regression rate and to determine the kinetic parameter of pyrolysis. In the experiments on PP combustion, extinction limits and sooting limits were found to be extended when highly preheated air was used. In the case of H2O and CO2 dilutions, the dilution enhanced regression rates at low stretch rates. In the case of PET combustion, results indicated that regression rates depended on the production rate of char. The kinetic parameters of PP pyrolysis under the combustion conditions were estimated using a new method in which experiments and numerical simulations were combined. The regression rates calculated using the kinetic parameters obtained in the present study were in good agreement with those of the experiments in contrast with the numerical results using the kinetic parameters obtained using the previously reported thermal gravimetric analysis (TGA).

A study on droplet combustion in unsteady force convection at high pressure under microgravity conditions was performed. The hysteresis loop of the instantaneous burning rate of a single suspended 1-butanol droplet was observed for the first time. Results showed that the classical quasi-steady film model cannot describe droplet combustion in an unsteady flow. Based on precise experimental observation and by utilizing dimensional analysis of the energy conservation equation, a new criterion is herein proposed for the condition in which the quasi-steady assumption is valid and for that in which it is not, The dimensional analysis led to formulation of a new time scale. Based on the time scale which we call the response-time-scale, a new Damkohler number, termed the response-Damkohler-number was formulated. Using the definition of the new time scale and that of the Damkohler number, unsteady behaviors of droplet combustion under conditions of various pressures and varying force convection were examined. Finally, using the response-Damkohler-number and the deviation factor between the actual instantaneous burning rate and the burning rate predicted by the quasi-steady theory, droplet combustion was categorized into four specific regimes. This study is also of fundamental interest in terms of the effects Of turbulence on droplet evaporation and combustion in spray flames. (c) 2008 The Combustion Institute. Published by Elsevier Inc. All rights reserved.

Characteristics of stretched methane-air flames stabilized in a forward stagnation region in front of all inert hot wall with constant temperatures were examined computationally and experimentally. Attention was paid to the bifurcations of stretched flames. To minimize potential experimental errors caused by natural convection, all the experiments were conducted under 10-second microgravity conditions at the JAMIC drop tower in Hokkaido, Japan. Computations for methane-air flames were performed with detailed chemistry and variable properties. Optically thin radiation model was employed using the Planck mean absorption coefficients for CO(2), H(2)O, CO and CH(4). Statistical narrow-band model was also used. Flame bifurcations, that is, the coexistence of multi-temperature flames at the identical conditions of the stretch rate, equivalence ratio and wall temperature were observed. Combustion was completed in the high-temperature flames which stand far from the hot wall, however, incomplete combustion was found in the low-temperature flames which locates close to the hot wall. Hence, levels of the CO concentration in the low-temperature flames were relatively high at the wall and those in high-temperature flames were negligibly low. Flame locations were experimentally measured at the conditions of the equivalence ratio around 0.6 and the wall temperature of 973 K by changing the stretch rate from 6 to 10 s(-1). Two flame locations were successfully confirmed in the identical conditions of the stretch rate, equivalence ratio and wall temperature. Computational and experimental results were compared with each other and discussed in terms of the flame bifurcations. (C) 2009 The Combustion Institute. Published by Elsevier Inc. All rights reserved.

Combustion experiments on polypropylene (PP) and polyethylene terephethalate (PET) in stagnation-point flow were performed to investigate the fundamental characteristics of these polymers in high-temperature air combustion (HiTAC). Numerical study was also performed to estimate the regression rate and to determine the kinetic parameter of pyrolysis. In the experiments on PP combustion, extinction limits and sooting limits were found to be extended when highly preheated air was used. In the case of H2O and CO2 dilutions, the dilution enhanced regression rates at low stretch rates. In the case of PET combustion, results indicated that regression rates depended on the production rate of char. The kinetic parameters of PP pyrolysis under the combustion conditions were estimated using a new method in which experiments and numerical simulations were combined. The regression rates calculated using the kinetic parameters obtained in the present study were in good agreement with those of the experiments in contrast with the numerical results using the kinetic parameters obtained using the previously reported thermal gravimetric analysis (TGA).

Methane/air turbulent premixed flames diluted with superheated water vapor at high-pressure and high-temperature were experimentally investigated to explore the effects of recycled water vapor on turbulent flame characteristics from the viewpoint or applying high-temperature air combustion (HiTAC) to high-load combustors as well as elucidating those effects of exhaust gas recirculation (EGR) in IC engine. A newly devised water evaporator was installed in a high-pressure chamber and superheated water vapor was successfully supplied to air up to 1.0 MPa and 573 K. The maximum dilution ratio defined as the ratio of the molar fraction of H(2)O to those of air and H(2)O was 0.1. Turbulent burning velocity, mean volume and the structure of the turbulent flame region were compared with those of flames diluted with CO(2), reported previously, which is another major species in recycled burnt gas. Results showed that the effects of superheated water vapor dilution on turbulent burning velocity, S(T), normalized by laminar burning velocity, S(L), was much weaker than that of CO(2) dilution. The mean volume of the turbulent flame region defined as the region between &lt; c &gt; = 0.1 and (c) = 0.9, was scarcely changed either. This means that the effects of recycled burnt gas on the structure of turbulent premixed flames at high pressure and high-temperature is predominated by CO(2) but not by superheated water vapor, indicating that Suppression of combustion oscillation in premixed-type gas-turbine combustors by extension of the Volume of heat release region is due to recycled CO(2). The emission indices of CO and NO(x) were also measured at high pressure, and it was proved that water vapor dilution is effective to restrain CO emission, which is a possible defect of HiTAC when it is applied to gas-turbine combustors. (c) 2009 The Combustion Institute. Published by Elsevier Inc. All rights reserved.

An experimental Study for 1-butanol single droplet flames ill constant and oscillatory flow fields was conducted under microgravity conditions at elevated pressure. In the constant flow experiments, flow velocities from 0 to 40 cm/s were tested. Using obtained data of d(2), the burning rate constants were evaluated. The burning rate constant in the quiescent condition was also calculated Successfully at high pressure by the extrapolation method based oil the Frossling relation. In the oscillatory flow experiments, the flow velocities were varied from 0 to 40 cm/s at the frequencies of 2-40 Hz. Results showed that the burning rate constant during the droplet lifetime varied following the quasi-steady relation at 0.1 M Pa: however. in the conditions with higher frequencies at 0.4 MPa, the average burning velocity became larger than that for the constant flow case with the velocity equivalent to the maximum velocity in the oscillatory flow. Under the condition where the burning rate constant increased, it wits observed that the flame did not sufficiently move back upstream, leading to enhancement of the heat transfer from the flame to the droplet surface. Therefore, the instantaneous burning rate constant increased. To investigate the mechanism of Such flame behavior, the ratio of two characteristic times tau(f)/tau(D) (tau(f): flow oscillation characteristic time tau(D): diffusion characteristic time), were compared. As the flow Oscillatory frequency increased, tau(f)/tau(D) becomes smaller, tau(f)/tau(D) also became smaller at high pressure. If tau(f)/tau(D) is small due to the small mass diffusion rate, the droplet flame Could not move back to the appropriate position for the minimum velocity in steady flow, leading to an increase of the burning rate constant, especially in the case of higher frequency at high pressure. (C) 2009 The Combustion Institute. Published by Elsevier Inc. All rights reserved.

A study on droplet combustion in unsteady force convection at high pressure under microgravity conditions was performed. The hysteresis loop of the instantaneous burning rate of a single suspended 1-butanol droplet was observed for the first time. Results showed that the classical quasi-steady film model cannot describe droplet combustion in an unsteady flow. Based on precise experimental observation and by utilizing dimensional analysis of the energy conservation equation, a new criterion is herein proposed for the condition in which the quasi-steady assumption is valid and for that in which it is not, The dimensional analysis led to formulation of a new time scale. Based on the time scale which we call the response-time-scale, a new Damkohler number, termed the response-Damkohler-number was formulated. Using the definition of the new time scale and that of the Damkohler number, unsteady behaviors of droplet combustion under conditions of various pressures and varying force convection were examined. Finally, using the response-Damkohler-number and the deviation factor between the actual instantaneous burning rate and the burning rate predicted by the quasi-steady theory, droplet combustion was categorized into four specific regimes. This study is also of fundamental interest in terms of the effects Of turbulence on droplet evaporation and combustion in spray flames. (c) 2008 The Combustion Institute. Published by Elsevier Inc. All rights reserved.

Experimental and numerical studies on the interaction between combustion of a hydrogen jet and an incident shock wave were performed. When the incident shock wave was introduced upstream of the injection slot, the boundary-layer separation region was expanded extensively. The penetration height, defined as the height of the Mach disk, with the incident shock wave was less than that without the incident shock wave. Combustion was confirmed when the incident shock wave was introduced downstream of the fuel injection slot. However, combustion was not confirmed with the incident shock wave upstream of fuel injection slot. The mechanism of these phenomena was examined based on the numerical simulation results in terms of the characteristic residence time ill the separation region near the fuel injection slot.

Combustion of polypropylene (PP) in a high-temperature, low-oxygen oxidizer enriched with H2O and CO2 in stagnation-point flow was studied numerically to explore fundamental characteristics of polymer incineration under the condition of high-temperature air combustion (HiTAC). The detailed chemistry of propene as a decomposition gas was used for the calculation. Two typical gas radiation models, i.e., the optically thin model (OTM) and the statistical narrow-band (SNB) model, were employed to clarify the effect of gas radiation on PP combustion as well as the validity of the use of these models under HiTAC conditions because the H2O and CO2 included in the burnt gas recirculated in HiTAC furnaces are highly radiative species. Results showed that, under HiTAC conditions, calculations using OTM overpredicts regression rates compared with those using SNB, indicating that OTM is not suitable for use with polymer combustion under HiTAC conditions, while the differences of these gas radiation models were slight when ordinary air at high temperature was used as an oxidizer. It was also shown that when SNB was used, CO2 enrichment in the oxidizer hardly enhanced the regression because CO2 near the PP surface behaves as a barrier against the radiative heat flux. The most effective condition to increase the regression rate is to maintain a higher concentration of H2O in the high-temperature oxidizer so as to enhance the gas radiation to the PP surface.

An experimental study was performed to investigate the flame-holding of hydrogen jet injected into supersonic cross-flow interacting with an incident shock wave. Pre-burned rich hydrogen or air was injected in supersonic airflow at Mach 2.5. The injection pressure was 1.2MPa or 1.6MPa. The deflection angle of the shock generator was 10 deg. Velocity profiles near the wall and those of recirculation zones around the injection slot could successfully be measured using newly-devised PTV method in this study. The velocity profiles showed that re-attachment point of separation downstream of the injection slot moves downstream when the incident shock wave is introduced to downstream of the injection slot. The re-attachment point with the incident shock wave in case of the injection pressure of 1.6MPa was not so different from that of 1.2MPa. However, extinction limit in case of the injection pressure of 1.6MPa extended than that of 1.2MPa.

Experimental and numerical studies of the interaction between combustion of a hydrogen jet and an incident shock wave were performed. In the case that an incident shock wave was introduced upstream of the injection slot, the boundary-layer separation region was extensively expanded. The penetration height of the Mach disk with an incident shock wave was less than that without an incident shock wave. Combustion was confirmed when the incident shock wave was introduced downstream of the fuel injection slot, while, with the incident shock wave upstream of fuel injection slot, combustion was not confirmed. The mechanism of these phenomena was discussed based on the results of numerical simulation in terms of the residence time in the separation region near the fuel injection slot.

Turbulent premixed flames of mixtures of CH4 and air diluted with CO2 at high pressure and high temperature were experimentally studied to clarify the effects of exhaust gas recirculation (EGR), especially the effects of CO2 in EGR gases, on turbulent flame characteristics. The maximum CO2 dilution ratio, defined as the ratio of the molar fraction of CO2 to those of air and CO2, was 0.1. The mixture was preheated up to 573 K and the maximum pressure was 1.0 MPa. Bunsen-type turbulent premixed flames of the mixtures were stabilized in a high-pressure chamber. OH-PLIF visualizations of the flames were performed. By analyzing the OH-PLIF images, turbulent burning velocity, mean volume of turbulent flame region, and mean fuel consumption rate were calculated. Results showed that the turbulent burning velocity, S-T, normalized using laminar burning velocity, S-L, became smaller when the mixture was diluted with CO2. When the turbulent flame region was defined as the region between &lt; c &gt; = 0.1 and &lt; c &gt; = 0.9, the mean volume of the flame region increased in the case of CO2 dilution. Moreover, the mean fuel consumption rate in the flame region decreased with increasing CO2 dilution ratio. This effect was stronger than the decrease in mass fraction of fuel due to CO2 dilution. The decrease in the smallest wrinkling scale of the flame front with increasing turbulence Reynolds number in the case Of CO2 dilution was more significant than that in the case of no CO2 dilution, corresponding well to the scale relation due to turbulence and intrinsic flame instability proposed previously. These results, as well as the previously reported effects of the profiles of the heat-release region on combustion oscillation, imply that exhaust gas recirculation for high-pressure, high-temperature turbulent premixed flames is effective for restraining combustion oscillation of premixed-type gas-turbine combustors. (C) 2006 The Combustion Institute. Published by Elsevier Inc. All rights reserved.

Two high-pressure combustion phenomena recently observed by the author's group are reviewed. The first one is the flame spread of a droplet array in the supercritical pressure range of the fuel in microgravity. Microgravity experiments are essential for research on droplet combustion, especially at high pressure, because of the large Grashof number in normal gravity. The flame spread rate for an n-decane droplet array was measured at high pressure, and a fuel-vapor jet was found to be generated due to an imbalance of surface tension of the droplet surface, leading to a higher flame spread rate. The second phenomenon is turbulent premixed combustion at high pressure and high temperature, environmental conditions of which are very close to those in SI engines and premixed-type gas turbine combustors. Information on the flame characteristics in such conditions has been very limited. A high-pressure combustion test facility with a large high-pressure combustion chamber enabled us to stabilize turbulent premixed flames with a high turbulence Reynolds number and to perform flame observations and measurements for extended period using lasers. Turbulent burning velocity was successfully measured and significant effects of intrinsic flame instability on flame structure and turbulent burning velocity at high pressure were revealed. Effects of CO, dilution on high-pressure and high-temperature premixed flames were also investigated to evaluate the fundamental effects of exhaust gas recirculation (EGR) in practical high-load high-pressure combustors.

A new microscopic model of the interaction between droplet flames and fine vortex tubes which compose a coherent structure of turbulence was developed. Three non-dimensional numbers were introduced to extend the length scale and time scale so as to be suitable for microgravity experiments using droplets of combustion of about I mm, in diameter. An experimental apparatus for combustion of a single droplet and that of an array of two droplets in varying airflow was developed, and experiments were performed in microgravity and normal gravity at pressures up to 2.0 MPa for n-nonane and ethanol as fuels. Variations of the instantaneous burning rate constant, K-i, in response to the varying flow velocity was successfully observed. At high pressure, the effects of droplet Reynolds number Re on K-i was clearly seen, while the effects of natural convection, which increases K-i with Re, was seen in normal gravity even in the forced airflows. As for the experiments on combustion of an array of two droplets, K-i reduction of the downstream droplet became weak when the flow direction was varied. However, the K-i reduction of the downstream droplet for flow direction variations was clearly seen for n-nonane droplets but almost not for ethanol droplets. The interaction mechanism between upstream and downstream droplets is considered to result from the elimination of oxidizer supply to the downstream droplet, indicating strong interaction effects of n-nonane droplets for a stoichiometric oxygen-fuel ratio of n-nonane (i.e., 14.0) greater than that of ethanol (i.e., 3.0). (C) 2005 Elsevier Inc. All rights reserved.

The effects of injection modeling on the combustion of hydrogen in a supersonic airflow are considered here using both numerical and experimental approaches. Two types of injection, single and double from the base of a strut, are analyzed in depth. The solutions to the governing equations of the numerical simulation including Navier-Stokes (N-S) equations, turbulence model, and finite full chemistry are brought in using a high-order numerical scheme. In particular, the laser-induced-fluorescence (LIF) method is effectively used for the OH radical recording. The flame structures of two different injection models are simulated separately using both numerical and experimental techniques for comparison. As a result of changes in the flowfield as well as mixing phenomenon, the flame structure of the double-slit injection model is significantly different from the one for the single injection. By increasing the slit interval of the double-slit injection, the model can face the blowout when d &GT; 8 mm. Furthermore, an increase in the slit width can also affect the flowfield of the wake significantly so that the combustion zone will expand downstream.

Turbulent burning velocities for methane/air mixtures at pressures ranging from atmospheric pressure up to 1.0 MPa and mixture temperatures of 300 and 573 K were measured, which covers the typical operating conditions of premixed-type gas-turbine combustors. A bunsen-type flame stabilized in a high-pressure chamber was used, and OH-PLIF visualization was performed with the pressure and mixture temperature being kept constant. In addition to a burner with an outlet diameter of 20 mm for the high-pressure experiments, a large-scale burner with an outlet diameter of 60 mm was used at atmospheric pressure to extend the turbulence Reynolds number based on the Taylor microscale, Rλ, as a common parameter to compare the pressure and temperature effects. It was confirmed that Rλ over 100 could be attained and that u′/SL could be extended even at atmospheric pressure. Based on the contours of the mean progress variable 〈c〉= 0.1 determined using OH-PLIF images, turbulent burning velocity was measured. ST/SL was also found to be greatly affected by pressure for preheated mixtures at 573 K. The bending tendency of the ST/SL curves with u′/SL was seen regardless of pressure and mixture temperature and the Rk region where the bending occurs corresponded well to the region where the smallest scale of flame wrinkling measured as a fractal inner-cutoff approaches the characteristic flame instability scale and becomes almost constant. A power law of ST/SL with (P/P0)(u′/SL) was clearly seen when ST was determined using 〈c〉= 0.1 contours, and the exponent was close to 0.4, indicating agreement with the previous results using the mean flame cone method and the significant pressure effects on turbulent burning velocity.

The unstable behavior of cellular premixed flames induced by intrinsic instability is studied by two-dimensional unsteady calculations of reactive flows. In the present numerical simulation, the compressible Navier-Stokes equation including a one-step irreversible chemical reaction is employed. We consider two basic types of phenomena to account for the intrinsic instability of premixed flames, i.e., hydrodynamic and diffusive-thermal effects. The hydrodynamic effect is caused by the thermal expansion through the flame front the diffusive-thermal effect is caused by the preferential diffusion of mass versus heat. A disturbance with several wavelength components is superimposed on a planar flame, and the formation of a cellular flame induced by hydrodynamic and diffusive-thermal effects is numerically simulated. After the cellular-flame formation, the combination and division of cells are observed. The behavior of cellularflame fronts becomes more unstable when the Lewis number is lower than unity, since the diffusive-thermal effect has a great influence on the unstable behavior. The cell size changes with time, and its average is greater than the critical wavelength and becomes smaller by decreasing the Lewis number. The flame velocity of cellular flames depends strongly on the length of computational domain in the direction tangential to the flame front. As the length of computational domain increases, the flame velocity becomes larger. This is because the long-wavelength components of disturbances play an important role in the shape of cellular flames, i.e., in the flame-surface area.

An oblique shock wave in a supersonic flow field was measured and visualized using Particle Tracking Velocimetry (PTV). The beginning of the velocity drop at the shock wave was sharper than that detected by general Particle Image Velocimetry (PIV). The location of the shock wave in the obtained velocity vector was in good agreement with that in Schlieren images because PTV has the advantage of high spatial resolution. The statistical values of turbulence in the airstream were successfully estimated. Another advantage of PTV is low calculation load. Real-time PTV was developed in this study. The image capture and the tracking calculation were performed in real time and the obtained vector data were sent to another computer using a network socket. It is considered that the transdisciplinary approach between experiment and numerical simulation can be performed using velocity vectors which will be sent to a supercomputer in real time. It is conformed that the obtained vector data and the real-time PTV are capable of realizing a transdisciplinary measurement system in future.

The unstable behavior of cellular premixed flames induced by intrinsic instability is studied by two-dimensional unsteady calculations of reactive flows. In the present numerical simulation, the compressible Navier-Stokes equation including a one-step irreversible chemical reaction is employed. We consider two basic types of phenomena to account for the intrinsic instability of premixed flames, i.e., hydrodynamic and diffusive-thermal effects. The hydrodynamic effect is caused by the thermal expansion through the flame front; the diffusive-thermal effect is caused by the preferential diffusion of mass versus heat. A disturbance with several wavelength components is superimposed on a planar flame, and the formation of a cellular flame induced by hydrodynamic and diffusive-thermal effects is numerically simulated. After the cellular-flame formation, the combination and division of cells are observed. The behavior of cellular-flame fronts becomes more unstable when the Lewis number is lower than unity, since the diffusive-thermal effect has a great influence on the unstable behavior. The cell size changes with time, and its average is greater than the critical wavelength and becomes smaller by decreasing the Lewis number. The flame velocity of cellular flames depends strongly on the length of computational domain in the direction tangential to the flame front. As the length of computational domain increases, the flame velocity becomes larger. This is because the long-wavelength components of disturbances play an important role in the shape of cellular flames, i.e., in the flame-surface area. (c) 2004 The Combustion Institute. Published by Elsevier Inc. All rights reserved.

Turbulent burning velocities for methane/air mixtures at pressures ranging from atmospheric pressure up to 1.0 MPa and mixture temperatures of 300 and 573 K were measured, which covers the typical operating conditions of premixed-type gas-turbine combustors. A bunsen-type flame stabilized in a high-pressure chamber was used, and OH-PLIF visualization was performed with the pressure and mixture temperature being kept constant. In addition to a burner with an outlet diameter of 20 mm for the high-pressure experiments, a large-scale burner with an outlet diameter of 60 mm was used at atmospheric pressure to extend the turbulence Reynolds number based on the Taylor microscale, R-gimel, as a common parameter to compare the pressure and temperature effects. It was confirmed that R-gimel over 100 could be attained and that u'/SL could be extended even at atmospheric pressure. Based on the contours of the mean progress variable (c) = 0.1 determined using OH-PLIF images, turbulent burning velocity was measured. S-T/S-L was also found to be greatly affected by pressure for preheated mixtures at 573 K. The bending tendency of the S-T/S-L curves with U'/S-L was seen regardless of pressure and mixture temperature and the R-gimel region where the bending occurs corresponded well to the region where the smallest scale of flame wrinkling measured as a fractal inner-cutoff approaches the characteristic flame instability scale and becomes almost constant. A power law Of S-T/S-L with (P/P-0)(u'/S-L) was clearly seen when S-T was determined using (c) = 0.1 contours, and the exponent was close to 0.4, indicating agreement with the previous results using the mean flame cone method and the significant pressure effects on turbulent burning velocity. (c) 2004 The Combustion Institute. Published by Elsevier Inc. All rights reserved.

This paper describes the effect of flame position and its spatial variation on prediction accuracy of combustion oscillation in a dry low emission (DLE) combustor. A one-dimensional linear model has been developed. The flame is usually treated as fixed and located at a single axial plane in conventional analysis. However, in practice, the flame position varies during operation. A new flame model considering this variation by a spatial distribution function has been developed. Variation of flame position is empirically measured by using UV images of OH radicals in the oscillating flame. A triangular distribution function is introduced into the flame model because it is similar to the experimentally obtained distribution function. A 'top-hat' distribution is also considered to test the influence of distribution shape on the result. Numerical results are compared with experimental data. The triangular flame model shows better prediction of the stability boundaries of combustion oscillation compared with the simple flame sheet model. The results of the top-hat flame model differ from those of the experiment. It is found that consideration of the spatial distribution yields good results for the DLE combustor. (c) 2004 The Combustion Institute. Published by Elsevier Inc. All rights reserved.

Experimental and numerical studies on laminar burning velocities of a stoiehiometric CH4/air mixture were performed at elevated pressure and elevated temperature. A stoichiometric CH4/air mixture was diluted by helium to restrain the intrinsic flame instabilities occurring at elevated pressure. Measurements of laminar burning velocity were conducted by a technique which measures the instantaneous local burning velocity using a particle tracking velocimetry (PTV) and planar laser induced fluorescence for OH radical ( OH-PLIF) simultaneously for burner-stabilized flames. Laminar burning velocities were determined by the average values of local burning velocities in the region where the Karlovitz number are sufficiently small, meaning that the effect of flame stretch and curvature can be neglected. Numerical simulations were also conducted using the one-dimensional premixed flame code. Detailed reaction mechanisms and a reduced mechanism were examined, and their results were compared with experimental results to investigate the feasibility of the mechanisms for predicting the flame characteristics at elevated pressure and elevated temperature.

The extinction limits of stretched diffusion flames of CO, H-2, CH4 and their mixtures in conjunction with room-temperature air and high-temperature oxidizers up to 773 K consisting of O-2 + N-2 and O-2 + H2O, were studied experimentally and numerically, in order to investigate the combustion characteristics of the mixed fuels produced from combustible wastes in incinerators. The cylindrical porous Tsuji-type burner was used to obtain a stretched diffusion flame in the present experiments. The numerical simulations were performed on a one-dimensional counterflow diffusion flame with detailed chemical reactions and a radiative heat loss term as optically thin limit in the energy conservation equation. The stretch rate, the fuel and the oxygen compositions, and the oxidizer temperature were the experimental and the numerical parameters. The numerical predictions were generally consistent with the experimental results. The stretch rate at extinction increased with the oxidizer temperature and the oxygen concentration. For instance, when the oxidizer temperature increased from 300 K to 773 K for the CH4/air flames, the stretch rate at extinction increased by approximately one order. For the CO/air flames, the stretch rate at extinction increased with H-2 concentration in fuel because the CO flame cannot be self-sustained in dry air and the mixing of a small amount of H-2 was needed. When H2O vapor was used for the dilution gas in the high- temperature oxidizer instead of N-2, the stretch rate at extinction decreased due to the large heat capacity. flame CO flames mixed with a small amount of H-2 or hydrocarbon gases have a thick visible flame and show a continuous emission spectrum having a maximum at 430 nm.

The mechanism determining the scale of the smallest flame wrinkles of high-pressure, high-temperature turbulent premised flames was investigated. The fractal inner cutoff of OH planar laser-induced fluorescence images, which is the smallest scale of flame wrinkles, was analyzed for burner-stabilized flames in a high-pressure chamber. Precise measurement of the energy spectrum of turbulence was also performed and the relationship between the intrinsic flame instability and flow turbulence was examined. Experiments were performed for CH4/air mixtures of 300 and 573 K at 0.1, 0.5, and 1.0 MPa. The experimental results clearly showed the Kolmogorov's similarity law for non-dimensional energy spectra of flow turbulence at high pressure and high temperature. A characteristic scale equivalent to the average vortex-tube diameter, l(v), which is about 10 times larger than the Kolmogorov scale revealed by recent direct numerical simulation, was used as a scale corresponding to the largest wave number of initial flow disturbances in an unburned mixture. At high pressure, the fractal inner cutoff, epsilon(i), and l(v) decrease with turbulence Reynolds number based on Taylor microscale, R-lambda, regardless of mixture temperature. The magnitude of epsilon(i) is close to l(v), when l(v) is larger than the characteristic instability scale corresponding to the maximum growth rate of flame instability, li. When R-lambda increases further and l(v) becomes smaller than l(v), epsilon(i) becomes almost constant. At atmospheric pressure, the relationship between l(v), l(v), and epsilon(i) was not obvious, but the correlation of epsilon(i) with the integral scale, l(g), was rather significant. These characteristics of epsilon(i) variation for high-pressure, high-temperature turbulent premixed flames can be explained using the scale-relation model based on l(g), l(v), and l(v) for R-lambda variation proposed in this study.

Flame propagation experiments on n-decane spray were performed in microgravity to investigate the flame propagation mechanism of a spray for less volatile fuels. After the spray was dispersed into an acrylic propagation tube with an inner diameter of 62 mm and a length of 535 mm, about 5 s were required for a quiescent spray to form under conditions of microgravity The spray was ignited by an electrically heated nichrome wire just after the Sauter mean diameter (SMD) and the concentration of spray were measured by a laser droplet analyzer. The variation of flame propagation speed with the SMD was determined. Results showed that the change of the flame propagation speed with the SMD had a maximum for a constant equivalence ratio. This can be reasonably explained by analogy with the flame spread from a droplet to a neighboring droplet in a droplet array, that is, the flame spread of a droplet array and the flame propagation speed of spray reach a maximum when the spacing between a droplet and a neighboring droplet is approximately equal to the flame radius of the droplet. Therefore, all experimental data could be arranged on one curve when the lateral axis was taken as the spacing between droplets divided by the flame diameter. Even when methane of 1 vol %, which is outside the flammability limit, was added to spray with a constant overall equivalence ratio, the flame propagation speeds fell on the integrated curve.

The flame spread phenomena of an n-decane droplet array in the supercritical pressure range were experimentally investigated in microgravity. Experiments were conducted at pressures up to 5.0 MPa, which is over the critical pressure of n-decane. Observations of the flame-spread phenomenon were conducted using OH-radical emission, Schlieren, and back-lit images recorded by a high-speed video camera. The flame-spread rates were calculated on the basis of the time history of the OH-emission images. In microgravity, the flame-spread rate decreased with increasing pressure, had a minimum at a pressure around half of the critical pressure, and then increased again. It had a maximum at the pressure near the critical pressure and then decreased gradually with pressure. In normal gravity; the flame-spread rate monotonously decreased and there was a pressure limit beyond which flame spread did not occur. Around the critical pressure, a jet-like flow of fuel vapor from an unburned droplet heated by the flame of a burning droplet was observed. The fuel-vapor jet from the side opposite the heating region of the unburned droplet reached another adjoining unburned droplet and then flame propagated along the jet, leading to heating prior to ignition of the unburned droplet. The mechanism of the fuel-vapor jet was examined based on the shear flow near the droplet surface induced by Marangoni convection of the unburned droplet heated non-uniformly by the burning droplet. The droplet internal flow rate was measured in normal gravity and confirmed the existence of Marangoni convection. The internal flow rate increased with pressure and had a maximum near the critical pressure. It was expected that the mechanism responsible for the maximum flame-spread rate near the critical pressure was the enhanced heat and mass transfer caused by the fuel-vapor jet and flame propagation along that jet.

The structures and stability of lifted combustion zones have been simulated with detailed chemistry and transport properties in an axisymmetric laminar fuel (CH,) jet and outer co-flow of the (O-2 + N-2) oxidizer whose initial temperature is 300 K, 700 K and 1200 K. A set of numerical simulations was executed by increasing the N-2 dilution ratio, Z (mole fraction of N-2 in the oxidizer). The results showed that at 300 K, the lifted combustion zone had a triple flame structure where the rich premixed wing is smaller than the lean one and the trailing diffusion flame immediately inclined to the fuel side from the triple point as well as the leading edge of the triple flame was shifted away from the jet axis as Z increased. As the initial temperature increased, the combustion zones were lifted at larger Z values than the one at 300 K. Especially, for 1200 K, it was found that the lifted combustion zones, when expressed in terms of the heat release rate, have become so weak that a flameless triple combustion zone was formed due to the high dilution ratio and high preheat temperature. The numerical simulations on the response of the lifted triple combustion zone to the initial fuel velocity were also carried out, and the results showed that the lifted combustion zone using a high preheated temperature was very stable in the near field.

The objectives of the present study are to measure NOx emission of counterflow diffusion flame, to compare the findings with numerical results, and finally to demonstrate efficacious effect of high-temperature air with low concentration of oxygen on NOx emission. Recently, high-temperature air with low concentration of oxygen is used for various industrial furnaces, resulting high efficiency and low emission of pollutants. Since high-temperature air increases NOx emission and air with low concentration of oxygen decreases it, these effects are competitive. Measurement and computation were conducted to clarify these two effects by use of counterflow diffusion flame. Since it is difficult to employ very high temperature over 1100 K in a laboratory-scale apparatus, a quantitative agreement between experimental and numerical results was confirmed first, and then a numerical approach was used to obtain a larger effect of low oxygen to reduce NOx emission. In the experiments, the methane concentration is changed from 10 to 30 vol% diluted by nitrogen, oxygen from 10 to 21 vol%, and air temperature from room temperature to 1100 K. The total amount of NOx sufficiently agreed between experimental and numerical results, although NO and NO2 could not be separated. By the numerical method, it was found that NOx emission from the counterflow diffusion flame of high-temperature low-oxygen air of 1500 K and 5% oxygen is comparable with that of room-temperature air of 21% oxygen. (C) 2002 Editions scientifiques et medicales Elsevier SAS. All rights reserved.

Flame propagation experiments on n-decane spray were performed in microgravity to investigate the flame propagation mechanism of a spray for less volatile fuels. After the spray was dispersed into an acrylic propagation tube with an inner diameter of 62 mm and a length of 535 mm, about 5 s were required for a quiescent spray to form under conditions of microgravity. The spray was ignited by an electrically heated nichrome wire just after the Sauter mean diameter (SMD) and the concentration of spray were measured by a laser droplet analyzer. The variation of flame propagation speed with the SMD was determined. Results showed that the change of the flame propagation speed with the SMD had a maximum for a constant equivalence ratio. This can be reasonably explained by analogy with the flame spread from a droplet to a neighboring droplet in a droplet array, that is, the flame spread of a droplet array and the flame propagation speed of spray reach a maximum when the spacing between a droplet and a neighboring droplet is approximately equal to the flame radius of the droplet. Therefore, all experimental data could be arranged on one curve when the lateral axis was taken as the spacing between droplets divided by the flame diameter. Even when methane of 1 vol %, which is outside the flammability limit, was added to spray with a constant overall equivalence ratio, the flame propagation speeds fell on the integrated curve.

The combined effects of non-gray radiation and pressure on the propagation and flammability limits of premixed CH4/O-2/CO2 flames were investigated numerically at an ambient temperature of 300 K over the pressure range of 0.04-0.50 MPa and the CO2 dilution range of 0.40-0.79 molar fraction of CO2 in the oxidant. A statistical narrow-band (SNB) model and an optically thin model (OTM) were used to describe flame radiation heat transfer in conjunction with a detailed chemical mechanism and transport properties. Results show that the effects of non-gray radiation and pressure on the burning velocity and flammability limits of the flame become significant as the amount of CO2 added is increased. Moreover, the flammability limits based on the OTM can be significantly extended by the spectral reabsorption mechanism described by the SNB model. Comparisons of the calculated results based on the SNB model with the experimental results in microgravity by Abbud-Madrid and Ronney show reasonable agreement. (C) 2001 by The Combustion Institute.

The flame surface density for turbulent premixed combustion on a nozzle-type burner is measured by planar laser-induced fluorescence (PLIF) and image processing techniques. The maximum flame surface density tends to show linear dependence on the K-factor given as a function of the integral length scale and u(0)'/S-L. The flame surface density shows an asymmetric profile in (c) over bar space with the peak location correlated in terms of the dimensionless parameter, N-B, which represents the degree of gradient or countergradient diffusion by turbulence. At values of N-B close to unity the peak occurs at a value of (c) over bar of about 0.7. As N-B increases above unity, the peak moves to a lower value in (c) over bar space, approaching a symmetric profile. The thickness of a turbulent flame brush nondimensionalized by the integral length scale tends to show linear dependence on the H-factor which is obtained by integrating the first moment equation of the reaction progress variable. The flame surface density increases at a higher ambient pressure due to decrease in the laminar burning velocity and the length scales of flame wrinkling. (C) 2000 by The Combustion Institute.

Experimental and numerical studies on the laminar burning velocity of hydrogen-air mixtures were performed. Measurements of laminar burning velocities were conducted using a new technique based on particle tracking velocimetry (PTV) and image processing for burner-stabilized flames in a high-pressure chamber. Equivalence ratios of the mixtures were varied from 0.6 to 3.0 for the pressure range from 0.1 to 0.5 MPa. A numerical simulation was conducted considering detailed reaction mechanisms and transport properties At high pressures, the experimental and numerical results agreed reasonably well with each other for mixtures of equivalence ratios of 1,0 and 2.0, while discrepancies were seen for the equivalence ratio 3.0. These discrepancies can be diminished effectively by modifying the rate-coefficient expressions of recombination reactions. (C) 2000 Elsevier Science Inc. All rights reserved.

Pulsating flame propagation in mixtures consisting of gas and particles is calculated numerically using a simplified model of particle-cloud combustion. The key heat transfer processes are based on the radiation heat transfer between particles and the heat transfer between gas and particles. Therefore, the model in which the flame propagates into a gas mixture seeded with inert particles is used to simplify the problem. The basic mechanisms of pulsating flame propagation can be explained as follows. When the burning velocity is rapid, there is insufficient time for the particles to be heated by radiation from the burned region. Therefore, the burning velocity slows down because the heat is absorbed by cold particles in and behind the flame. On the other hand, when the burning velocity is slow, the particles in front of the flame are sufficiently heated by radiation from the hot particles behind the flame so that the flame gathers speed. After passing through this period, the flame meets the cold particles, and its speed is again decreased. Thus, the flame oscillates in propagation. Pulsating flame propagation appears under certain conditions of the equivalence ratio, the weight concentration of particles, and the particle diameter. In the case of low weight concentration of particles and large particles, radiation preheating is not sufficient so that pulsation does not occur. Also, in the case of small particles, the particle temperature rises quite fast and the temperature in the system equilibrates rapidly so that the pulsation disappears. Tendencies of calculaton results agree with those of experimental results obtained from the PMMA particle cloud in microgravity.

The objectives of this study were to observe ignition events and to measure ignition times of a droplet array of n-decane placed in a high-temperature low-speed airflow under microgravity field. Due to the difficulty of making droplets of the same size within a short period of time in a drop capsule, imitation droplets made of porous ceramic balls soaked with n-decane were used. Experimental conditions were a droplet diameter of 1 mm, droplet spacing within the range of 0 to 6 mm, airflow velocity of 0 to 10 cm/s, and an airflow temperature of 925 K. According to OH emission images taken by a high speed camera with an OH band-path filter, ignition occurred around the droplets simultaneously at zero airflow velocity. At higher airflow velocities of more than several centimeters per second, however, ignition was initiated in the wake flow of the droplets and the flame spreads to the forward region of droplets. A range of droplet spacing existed in which ignition times of droplet arrays were less than those of a single droplet and had a minimum ignition time at a certain spacing. The spacing of this minimum ignition time increased with an increase of airflow velocity.

Pulsating flame propagation in mixtures consisting of gas and particles is calculated numerically using a simplified model of particle-cloud combustion. The key heat transfer processes are based on the radiation heat transfer between particles and the heat transfer between gas and particles Therefore, the model in which the flame propagates into a gas mixture seeded with inert particles is used to simplify the problem. The basic mechanisms of pulsating flame propagation can be explained as follows. When the burning velocity is rapid, there is insufficient time for the particles to be heated by radiation from the burned region. Therefore. the burning velocity slows down because the heat is absorbed by cold particles in and behind the flame. On the other hand, when the burning velocity is slow, the particles in front of the flame are sufficiently heated by radiation from the hot particles behind the flame so that the flame gathers speed. After passing through this period, the flame meets the cold particles, and its speed is again decreased. Thus, the flame oscillates in propagation. Pulsating flame propagation appears under certain conditions of the equivalence ratio, the weight concentration of particles, and the particle diameter. In the case of low weight concentration of particles and large particles, radiation preheating is not sufficient so that pulsation does not occur. Also, in the case of small particles, the particle temperature rises quite fast and the temperature in the system equilibrates rapidly so that the pulsation disappears. Tendencies of calculaton results agree with those of experimental results obtained from the PMMA particle cloud in microgravity.

To explore the mechanism determining the smallest scale of flame wrinkles and turbulent burning velocity in a high-pressure environment, OH planar laser-induced fluorescence (PLIF) images of turbulent and non-turbulent premixed flames stabilized in a high-pressure chamber were analyzed for CH4/air and C3H8/air mixtures. Fractal analysis was performed to investigate the characteristics of the scale and complexity of the flame wrinkles. It was found that fractal dimension increased with increasing u ' /S-L for the whole pressure range in the experiments. The increase in the dimension was rapid at higher pressure. The fractal inner cutoff decreased with u ' /S-L and pressure. However, at high pressure, the variation of the fractal inner cutoff with u ' /S-L was very small, showing that the inner cutoff is almost constant over the wide range of u ' /S-L Comparison between the inner cutoff and various characteristic scales of turbulent flames was made. It was proved that, at high pressure, a significant correlation exists between the inner cutoff and the characteristic scale of flame instability, that is, Darrieus-Landau instability combined with diffusive thermal effects, where the characteristic instability scale is defined based on the wavenumber at the maximum growth rate of flame disturbances. Flame instability of the high-pressure flame without flow turbulence also was observed. The nominal burning velocity enlarged by the flame-area increase due to the flame instability and its variation with pressure was measured using a mean angle method for OH-PLIF images at pressure up to 3.0 MPa, and the pressure exponent was found to be 0.4. Based on these results, a concept to explain the pressure effects that appeared in the general correlation of the turbulent burning velocity obtained by Kobayashi el al. was proposed. That is, flame instability which produces small-scale wrinkles is significant in a high-pressure environment and overlaps with flame-area increase due to the turbulence, causing larger S-T/S-L.

Experimental investigation of the effect of blended fuel on flame spread along droplet array has been conducted. Flame spread rate is measured using high-speed chemiluminescence images of an OH radical. The flame spread is observed with the initial droplet diameter, droplet spacing, and the mixing ratio of n-heptane and n-hexadecane. The mode of flame spread is categorized into two types: a continuous mode and an intermittent one. It is seen that flame spread rate is sensitively dependent on the relative flame position to droplet spacing. For a large droplet, the flame spread time is governed by a volatile fuel (heptane), but for a small droplet, it is controlled by a less volatile fuel (hexadecane). Copyright (C) 1999 John Wiley & Sons, Ltd.

Lean flammability limits of heterogeneous mixtures of combustible solid particles and air were measured in a microgravity field. Polymethylmethacrylate (PMMA) spherical particles with mass median diameters of 5.0, 8.4, 13.5, 30.4, and 48.5 mu m were instantaneously and uniformly dispersed into a cylindrical closed vessel using an air jet dispersion device. The mixture became quiescent without sedimentation of the PMMA particles at 6 seconds after the dispersion in microgravity. The mixture was ignited at the center of the vessel by a hot wire, and then a spherically propagating flame was formed. Flame behavior in the vessel near the lean flammability limit was observed with pressure histories, CCD camera images, and flame speeds measured by ionization probes. The experiments were performed for the initial conditions at room temperature and atmospheric pressure in microgravity. Results showed that the equivalence ratios at the lean flammability limits for the particles whose mean diameters were 8.4, 13.5, 30.4, and 48.5 mu m were 0.68, 0.76, 0.86, and 1.05, respectively, indicating that the equivalence ratio at the limit increased linearly as the particle diameters were increased. (C) 1999 by The Combustion Institute.

Experimental investigations on flame spread along a droplet array have been conducted at elevated pressures up to supercritical pressures of the fuel droplet under normal gravity and microgravity. The flame spread rate is measured using high-speed chemiluminescence images of OH radicals and direct visualization is employed to observe the images of the vaporizing fuel around the unburnt droplet. The mode of flame spread is categorized into two: a continuous mode and an intermittent one. There exist a limit droplet spacing and a limit ambient pressure in normal gravity, above which flame spread does not occur. It is seen that flame spread rate is dependent upon the relative position of flame to droplet spacing, In microgravity, the limit droplet spacing of flame spread and the droplet spacing of maximum flame spread rate are larger than those in normal gravity. In microgravity, the flame spread rate with ambient pressure decreases initially, shows a minimum, and then decreases again after taking a maximum, Flame spread time is determined by competing effects between the increased transfer time of the thermal boundary layer due to reduced dame diameter and the decreased ignition delay time in terms of the increase of ambient pressure. In normal gravity, the flame spread rate with ambient pressure decreases monotonically and there exists a limit ambient pressure, except at small droplet spacing, under which flame spread extends to the range of supercritical pressures of fuel, This is because natural convection induces the upward flow of hot gases into a plume above the burning droplets and limits the lateral transfer of thermal boundary layer. Consequently, it is found that flame spread behaviour under microgravity is considerably different from that under normal gravity due to the absence of natural convection, Copyright (C) 1999 John Wiley & Sons, Ltd.

Near-infrared (IR) and ultraviolet (UV) emission profiles of flame balls at microgravity conditions in H-2-O-2-diluent mixtures were measured in the JAMIC 10 s drop-tower and compared to numerical simulations and supplemental KC135 aircraft mu g experiments. Measured flame ball radii based on images obtained in the JAMIC, KC135, and recent space experiments (IR only) were quite consistent, indicating that radius is a rather robust property of flame balls. The predicted IR radii were always smaller than UV radii, whereas the experiments always showed the opposite behavior. Agreement between measured and predicted flame ball properties was closer for UV radii than IR radii in H-2-air mixtures but closer for IR radii in H-2-O-2-CO2 mixtures. The large experimental IR radii in H-2-air tests is particularly difficult to interpret even when uncertainties in chemical and radiation models are considered. Experimental radii would be consistent with a chemiluminescence reaction of the form HO2 + HO2 --&gt; H2O2 + O-2 producing an excited state of H2O2, since HO2 is consumed at large radii through this reaction and its exothermicity is sufficient to create excited states that could emit at the observed wavelengths, however, no appropriate transition of H-2-H-2* could be identified. (C) 1998 by The Combustion Institute.

Experimental and numerical investigations were performed for the Iaminar burning velocity and the flame structure of laminar premixed CH_4/O_2/CO_2 flames. Measurements of the laminar burning velocity were conducted by using a flame cone angle method for a circular nozzle burner. Numerical simulation was performed using one-dimensional plane flame code including radiation heat loss with an optically thin model. It was shown that the laminar burning velocity decrease with CO_2 addition even though the adiabatic flame temperature is the same as that for CH_4/Air flames. The radiation heat loss is significant for the CH_4/O_2/CO_2 flames, and the flame temperature and laminar burning velocity decreases when the radiation heat loss is considered. Effects of thermal properties, radiation, and chemical reaction on the determination of the laminar burning velocity of CH_4/O_2/CO_2 flames were discussed.

Flame spread phenomena in a suspended fuel droplet array were experimentally investigated for n-decane and n-hexadecane in a high-pressure ambience. Seven droplets of the same size were arranged at equal horizontal spacings. Flame spread rates were measured based on OH emission histories detected by a high-speed video camera with an image intensifier for droplet diameters of 0.50, 0.75, and 1.0 mm at ambient pressures from 0.1 to 2.0 MPa. Results show that, as droplet spacing becomes smaller, flame spread rate increases and attains a maximum value at a specific spacing. A further decrease in droplet spacing causes the spread rate to decrease due to the large latent heat of vaporization. Experiments were also conducted in a microgravity field to determine ii these characteristics of flame spread are affected by natural convection.

Turbulent burning velocities for lean C2H4-air and C3H8-air mixtures were measured for Bunsen-type turbulent premixed flames stabilized in a high-pressure chamber keeping the chamber pressure constant. The experiments were performed for a pressure range up to 1.0 MPa and for u'/S-L range up to 25. The effects of equivalence ratio, Lewis number, and pressure exponent of laminar burning velocity were examined. Results show that for both C2H4-air and C3H8-air flames, S-T/S-L increases rapidly with increasing u'/S-L, particularly for weak turbulence (u'/S-L &lt; 1). However, the rate of increase for the C3H8-air flames was smaller than that for C2H4-air flames, indicating that flame area increase due to hydrodynamic instability at high pressure was restrained by diffusive-thermal effects because of the larger Lewis number of C3H8-air mixtures. The data for a wider range of u'/S-L showed that the u'/S-L dependence of S-T/S-L for C2H4-air flames is almost the same as that for CH4-air flames reported by Kobayashi et al. within the scattering of the data. On the other hand, S-T/S-L for C3H8-air flames showed a smaller S-T/S-L than in the cases of C2H4-air and CH,-air flames. The verifications of existing correlations between S-T/S-L, u'/S-L, and turbulence Reynolds number, R-1, were made to elucidate effects of elevated pressure on turbulent burning velocity and the relationship between these parameters. Results show that the data for our experiments had a similar tendency to the general correlation proposed by Abdel-Gayed et al. on turbulent burning velocity but had a larger S-T/S-L than the correlation. A 1/4 power law of R-1 was seen for data in u'/S-L &gt; 1.0, consistent with the fractal flamelet model of turbulent burning velocity proposed by Gouldin. The best correlation between these parameters derived from the experimental data was S-T/S-L proportional to [(P/P-0)(u'/S-L)](n) for the whole range of u'/S-L and the exponent, n, is close to 0.4.

Although a large number of studies have been made on heterogeneous combustion of solid particle clouds, the combustion mechanisms of these systems are not well understood due to the complexity of the combustion processes and difficulties in conducting experiments. The purpose of this paper is to show a particular phenomenon found in flame propagation of PMMA particle clouds. The experiments were performed in microgravity to pl event sedimentation of the particles and the effect of buoyancy on the flame using a closed-vessel method and spherical polymethylmethacrylate (PMMA) particles as fuel. Observation of the flame using a CCD video camera and the time history of pressure and ionization current showed that the flame propagation with alternating fast amid slow modes occurred and that the flame was characterized by oscillation. This phenomenon is called "pulsating flame" in this paper. The pulsating flame appeals only near the lean flammability limit. As fuel concentration increases from the lean flammability limit, the frequency of time pulsating flame increases first and then decreases; beyond a certain equivalence ratio, the pulsating flame does not occur. Time frequency also decreases with increasing mean diameter of the particles. Explanation of the mechanism of this pulsating flame phenomenon was attempted based on die concept of heat absorption of die particles in burned gas and radiative heat transfer to the particles in a fresh mixture.

In order to explore the characteristics of turbulence and turbulent premised flames in a high-pressure environment, a nozzle-type burner with a turbulence generator was installed in a high-pressure chamber. Turbulence measurements and combustion experiments were conducted with the chamber pressure up to 3.0 MPa. Methane-air mixtures were used for the combustion experiments and confirmed that the turbulent premixed flames were successfully stabilized. Flame observations were made using instantaneous Schlieren photographs and high-speed laser tomography. Turbulence measurements were conducted using a hot-wire anemometer installed in the high-pressure chamber. It was found that the scales of turbulence generated by perforated plates at elevated pressure are smaller than those at atmospheric pressure. From flame observations, the following features of the flames at elevated pressure were found: (1) wrinkled structures of the flames become very fine and complex, and the cusps become sharp as pressure rises; (2) the flamelet breaks at many points of the flames and the scales of broken flamelets become small; (3) small-scale parts of the flame front convex to the unburned mixture frequently occur and move quickly to the unburned side. The effects of ambient pressure on turbulence characteristics and possible mechanisms which produce the wrinkled structure of the fine scales and generate flame front disturbances in the high-pressure environment are discussed. Copy-right (C) 1997 by The Combustion Institute

To explore the effects of ambient pressures on the turbulent burning velocity in a high-pressure environment, turbulent premixed flames of lean methane-air mixtures stabilized with a nozzle-type burner in a high-pressure chamber were investigated experimentally. Continuous combustion was investigated up to pressures of 3.0 MPa. Measurements of turbulent burning velocity were made using a mean-angle method based on a technique involving laser tomography and image processing. Results show that the effects of elevated pressure on turbulent burning velocity are significant and that the ratio of turbulent to laminar burning velocities, S-T/S-L, increases with both turbulence intensity u' and pressure, reaching a value of 30 at 3.0 MPa under the present experimental conditions. The increases in S-T/S-L with increasing u'/S-L are rapid at high pressure, particularly for small u', that is, in the region of weak turbulence. An interesting similarity of S-T/S-L variations was observed between the effect of pressure found in this experiment and the effect of a density jump as analyzed by Cambray and Joulin. Flame front instability theory based on Sivashinsky's formulation was applied to flames in high-pressure environments; it was found that the region of wave numbers where the flame front is unstable extends to larger wave numbers with increasing pressure because the diffusive-thermal effect, which stabilizes the hydrodynamic instability, weakens. This suggests that hydrodynamic instability, which enlarges the total flame area, plays an important role in the rapid increase of S-T/S-L with pressure in high-pressure environments.

Flame propagation experiments were conducted for a PMMA particle cloud in a microgravity field provided by the JAMIC (Japan Microgravity Center), whose microgravity level was 10(-4)-10(-5) g and duration was 10 s. PMMA-polymethyl methacrylate sphere particles with various mass-median diameters from 5.0 to 30.4 mu m were dispersed in a flame propagation tube with an inner diameter of 38 mm, and then the cloud was ignited 6 s after the release of the drop capsule, because vortex-like motions generated at the time of dispersion stopped after several seconds. The results showed that the maximum flame propagation speed decreased and the equivalence ratio at this point moved to the rich side when the particle size increased. When a very small amount of methane was added to the air, the plots of the flame propagation speed versus the equivalence ratio changed greatly for particles with a mass-median diameter of 8.4 mu m, which is different from 5.0-mm particles. It was concluded that for 8.4-mm particles, the flame behavior shifted from a nongaseous type of flame to a gaseous type of flame when a very small amount of methane is mixed with the air Also, the new arrangement of data, based an the nondimensional distance between particles, Verified the change of the flame propagation speed for large-particle clouds. This implies that the flame of a large particle cloud propagates basically according to flame spread between particles.

Ignition experiments on a suspended droplet matrix quickly immersed in a high-temperature ambience were conducted for hexadecane and decane in a microgravity field(10(-5) g) with a test duration of 4.5 s in the 100-meter drop-shaft of the Microgravity Laboratory of Japan (MGLAB) in Gifu, Japan. Small porous ceramic balls soaked with liquid fuel were employed a model droplets. The effect of the initial diameter and the spacing of the droplets on ignition delay time were measured. Also, by using a high-speed video camera with an image intensifier, OH emission images from the droplet matrix were observed. With a decrease in droplet spacing, in general, thermal interaction between droplets becomes significant and ignition time increases monotonically. However, in the case of small droplets, ignition time has a minimum when spacing is small, being shorter than that of a single droplet. The dependence of the vaporization time and the reaction time on the initial droplet diameter for a single droplet ignition basically elucidated the ignition time behaviour of a droplet matrix.

In order to obtain the ignition behavior at supercritical pressures, ignition times of a single fuel dropletwere measured in high-pressure high-temperature ambient. A suspended droplet of n-hexadecane or n-heptane with a diameter of 0.35-1.4 mm was quickly immersed in an electric furnace with a temperature up to 950 K. Attachment of the droplet, movement of the furnace, and ignition measurement were carried out in an air vessel with pressures up to 3 MPa. At low pressures, ignition times of both fuels decreased with the initial droplet diameter and thenincreased. Therefore, the ignition time variation with the initial droplet diameter has a minimum. This phenomenon, however, disappeared at high pressures. Also, the ignitable limit of droplet diameter, below which the droplet vaporized completely before ignition, decreased as pressure increased. In the case of a droplet burning at high pressures, the preceding experiment showed that the burning rate constant increased and had a maximum around the critical pressure of fuel. This is significantly caused by variable properties around the critical point such as thermal conductivity and diffusion coefficient and therefore, the present ignition time was expected to show similar characteristics due to the same reason. Ignition time, however, decreased monotonously with pressure, and even at supercritical pressures, the ignition time behavior did not change much. Being different from the case of combustion, it is suggested that drastic changes of properties did not take place in ignition processes. © 1994 Combustion Institute.

The flame propagation experiments on clouds of purely spherical PMMA particles in a microgravity environment were conducted by using the Japan Microgravity Center (JAMIC) drop shaft, where a microgravity condition of 10-4 g for 10 s is available. The exact measurement of the burning velocity of the particle cloud was impossible due to the particle sedimentation in normal gravity up to now. The particle cloud was created using a fluidized-bed-type device and suspended in the flame propagation tube. The cloud was ignited at the open end of the tube, and the flame speed was measured by charge coupled device (CCD) video camera images. The flame speed in normal gravity was also measured, and the two groups of results were compared. The results showed that the flame speed in normal gravity was considerably larger than for ordinary gaseous flames, since turbulent combustion occurred due to the residual turbulence of the flow and the turbulence generated by the particle sedimentation. On the other hand, in the microgravity environment, when the cloud was ignited 6 s after the release of the capsule, the particles were quiescent and dispersed with sufficient uniformity, indicating the effectiveness of the long duration micorgravity environment on the decay of turbulence. The flame speed decreased drastically in comparison with normal gravity cases, but the dependence of the flame speed on the particle concentration was similar to that in normal gravity. © 1994 Combustion Institute.

The actual flame stretch rates of the stretched cylindrical premixed flame and counterflow twin flames were measured with LDV for the methane-air and propane-air mixtures over a wide range of equivalence ratios. The stretch rate at the flame center at the point of flame extinction was considered to be the maximum stretch rate for the existence of cylindrical and plane flamelets. It was shown that the actual flame stretch rate at the point of flame extinction for the stretched cylindrical flame was smaller than that for the counterflow twin flames except for the case of very rich propane-air mixtures. The conditions for the existence of the cylindrical flamelet in turbulent premixed flames, which has a similarity to the fine structure of turbulent premixed flames proposed by Chomiak (1970, 1972, 1977), were discussed using the stretch rate data at the extinction point and the turbulence parameters. It was found that the cylindrical flamelet could not exist in highly turbulent flames where the distributed reaction zones were realized because of the limitation of resistibility to the extinction due to local strain. Furthermore, the cylindrical flamelet should be limited to mixtures such as rich propane-air and lean methane-air in which the minimum flame diameter at the extinction point is comparable to the Kolmogorov microscale.

The flow fields of a counterflow nozzle burner and counterflow twin flames for propane-air mixture stabilized with that burner were measured using LDV. The effects of the burner dimension, i.e., the outlet diameter of two opposed contraction nozzles and the distance between them, on the velocity profiles and extinction limits of the twin flames were examined. Results show that the effects of burner dimension on the extinction limit do not appear when the velocity gradient upstream of the front edge of preheating zone is used as a flow parameter and that the ratio of nozzle distance to nozzle outlet diameter is the main factor determining the dimensional effects on the flow characteristics. It is also shown that the flame stretch rate of the twin flames defined at the stagnation plane is about two times as large as that defined at the cold boundary of the flame.

Inside a coaxial flow diffusion flame of the ordinary type (fuel ejection into an air atmosphere), another diffusion flame of the reverse type (air ejection into a fuel atmosphere) was formed. The sooting limits of this double diffusion flame were measured for propane with three different types of burners, and the effect of several factors, such as the amount of inner air, flame-to-flame distance, thermal condition between the two flames, and flow stretch, were investigated. On the basis of the results, the effectiveness of making use of the double flame in the suppression of soot emission from diffusion flames of the coaxial flow type was discussed. The main findings were : (1) The double flame condition suppressed soot emission more effectively. (2) The sooting limit was controlled by the outer flame temperature, and therefore, adjusting the factors so as to increase this temperature may lead to much stronger soot suppression.

For the basic investigation of a premixed flame subjected to the effects of both flame stretch and flame curvature, we took up the stretched cylindrical flame formed inside the porous cylinder when the mixture was ejected uniformly from the inner wall surface of it. With methane/air and propane/air mixtures, flame behavior observations and measurements of extinction limit and flame diameter were made for these flames, and the effects of flame stretch and flame curvature were discussed, including those of heat loss to the burner wall, buoyancy, and Lewis numbers of reactants. Theoretical analysis was also made and extinction limits of the cylindrical flame and counter-flow twin flames, which have no flame curvature effects, were calculated and compared. The results showed that the curvature of the cylindrical flame propagating outward weakness resistibility to the flame extinction due to stretch.