OH LIF Research Articles

Characterization of a high-pressure flame facility using high-speed chemiluminescence and OH LIF imaging

Characterization of a high-pressure flame facility using high-speed chemiluminescence and OH LIF imaging

Thermocouple-based thermometry for laminar sooting flames: Implementation of a fast and simple methodology

This work reports the implementation and the validation of the extrapolation method for thermocouple temperature measurements corrected from the radiation losses in sooting flames. This simple method relies on the use of thermocouples having different size diameters and enables a fast and direct determination of the flame temperature by extrapolation to zero diameter. We propose here a detailed study of the possibilities and limitations offered by this method based on experimental measurements and comparison with well-established methods carried out in a laminar diffusion sooting flame. In details, a specific fast insertion setup using four different sized thermocouples has been implemented to record temperature values at different heights in the flame. From these data, we highlight that a linear calibration curve correlates the raw measured temperatures to the flame temperatures corrected of the radiation losses can be easily and rapidly obtained. The impact of soot deposition on the thermocouple on the temperature measurement is also discussed. To assess the reported thermocouple methodology, a direct comparison is made between the temperature profile determined along the vertical central axis of the flame by the extrapolation method with OH and NO LIF thermometry measurements as well as numerical simulation. Finally, we also report the comparison of experimental and simulated radial temperature profiles highlighting the adequate dynamic of the method for temperature profile determination in high temperature gradient conditions (500 K/mm). This work demonstrates that the extrapolation method is an efficient and fast method to determine accurate temperature profiles in flames, even in presence of soot particles up to a few hundred ppb, which can be useful for the development of fast and cheap sensors for either laboratory or larger-scale applications.

A study of the spatial and temporal evolution of auto-ignition kernels using time-resolved tomographic OH-LIF

In this paper, the spatial and temporal evolution of auto-ignition kernels from methane jets propagating into a NOx-vitiated, high-turbulence, hot air co-flow was studied by means of time-resolved tomographic laser-induced fluorescence of OH (Tomo OH-LIF). Measurements were performed using a burst dye laser system at 10 kHz for volumetric laser illumination and a multi-camera arrangement (8-views) for detection of the fluorescence signal. Auto-ignition kernels were detected three-dimensionally and tracked using a robust algorithm based on the intensity gradient of the volumetrically reconstructed signals. The size and location of the detected kernels were evaluated for operating conditions with different Reynolds numbers of the fuel jet. Results showed that auto-ignition randomly occurred with high probability in a well defined fairly axisymmetric radial region with strong fluctuations in the main direction of the flow. The increase of the Reynolds number of the fuel jet resulted in a radial spread of the location of auto-ignition events. The statistical evaluation of the orientation and growth of auto-ignition kernels with respect to the mean flow field showed that the kernels were oriented tangentially to the flow and temporally evolve towards this preferential direction as the ignition events progressed.

A study on the interaction between local flow and flame structure for mixing-controlled Diesel sprays

A detailed study on the spray local flow and flame structure has been performed by means of PIV and laser-sheet LIF techniques under Diesel spray conditions. Operating conditions were based on Engine Combustion Network recommendations. A consistent comparison of inert and reacting axial velocity fields has produced quantitative information on the effect of heat release on the local flow. Local axial velocity has been shown to increase 50–60% compared to the inert case, while the combustion-induced radial expansion of the spray has been quantified in terms of a 0.9–2.1mm radius increase. As a result, the drop in entrainment rate has been quantified around 25% compared to the inert case. Streamline analysis also hints at a reduced entrainment under reacting conditions. A 1D spray model under reacting condition has been used, which confirms the modifications obtained in the main flow metrics when moving from inert to reacting conditions. When comparing the flow evolution with the flame structure, little effect of chemical activity on the spray flow upstream the lift-off length has been evidenced, in spite of the presence of formaldehyde in such regions. Only downstream of the lift-off length, as defined by OH LIF, has a strong change in flow pattern been observed as a result of combustion-induced heat release.

Effect of the mixing fields on the stability and structure of turbulent partially premixed flames in a concentric flow conical nozzle burner

The mixing field is known to be one of the key parameters that affect the stability and structure of partially premixed flames. Data in these flames are now available covering the effects of turbulence, combustion system geometry, level of partially premixing and fuel type. However, quantitative analyses of the flame structure based on the mixing field are not yet available. The aim of this work is to present a comprehensive study of the effects of the mixing fields on the structure and stability of partially premixed methane flames. The mixing field in a concentric flow conical nozzle (CFCN) burner with well-controlled mechanism of the mixing is investigated using Rayleigh scattering technique. The flame stability, structure and flow field of some selected cases are presented using LIF of OH and PIV. The experimental data of the mixing field cover wide ranges of Reynolds number, equivalence ratio and mixing length.The data show that the mixing field is significantly affected by the mixing length and the ratio of the air-to-fuel velocities. The Reynolds number has a minimum effect on the mixing field in high turbulent flow regime and the stability is significantly affected by the turbulence level. The temporal fluctuations of the range of mixture fraction within the mixing field correlate with the flame stability. The highest point of stability occurs at recess distances where fluid mixtures near the jet exit plane are mostly within the flammability limits. This paper provides some correlations between the stability range in mixture fraction space and the turbulence level for different equivalence ratios.

Scalar dissipation rate and scales in swirling turbulent premixed flames

Simultaneous Rayleigh scattering and OH LIF imaging measurements of temperature and OH were used to investigate the properties of turbulent premixed flames, including the nature of the 2D thermal structures and scalar dissipation rate in the Cambridge/Sandia swirling bluff body stabilized flames, with and without the effect of swirl. Swirl creates enhanced turbulence as well as outer flow entrainment, and disrupts the pre-flame zone significantly, whilst the high temperature reaction zone as marked by OH remains relatively intact. In particular, the temperature at the location of maximum OH gradient shows very low variance across the flame region.The 2D image analysis of OH and temperature shows that the corresponding 2D gradients are aligned up to a distance of half the laminar flame thickness away from the flame front, deviating significantly in the case of swirling flames beyond that region. As in previous investigations in diffusion flames, the mean width of the observed thermal structures increases from 300 to 600 µm near the flame, with a main mode around the laminar flame thermal width in the unswirled case. The correlation between 2D thermal dissipation and variance of the reaction progress variable extracted from the images shows a direct proportionality, with a slope which agrees well with theory in the region of high turbulence away from the base. At the base of the flame where turbulence is low, the local scalar dissipation becomes a function of the local temperature via the thermal diffusivity.

Impact of heat release on strain rate field in turbulent premixed Bunsen flames

The effects of combustion on the strain rate field are investigated in turbulent premixed CH4/air Bunsen flames using simultaneous tomographic PIV and OH LIF measurements. Tomographic PIV provides three-dimensional velocity measurements, from which the complete strain rate tensor is determined. The OH LIF measurements are used to determine the position of the flame surface and the flame-normal orientation within the imaging plane. This combination of diagnostic techniques enables quantification of divergence as well as flame-normal and tangential strain rates, which are otherwise biased using only planar measurements. Measurements are compared in three lean-to-stoichiometric flames that have different amounts of heat release and Damköhler numbers greater than unity. The effects of heat release on the principal strain rates and their alignment relative to the local flame normal are analyzed. The extensive strain rate preferentially aligns with the flame normal in the reaction zone, which has been indicated by previous studies. The strength of this alignment increases with increasing heat release and, as a result, the flame-normal strain rate becomes highly extensive. These effects are associated with the gas expansion normal to the flame surface, which is largest for the stoichiometric flame. In the preheat zone, the compressive strain rate has a tendency to align with the flame normal. Away from the flame front, the flame – strain rate alignment is arbitrary in both the reactants and products. The flame-tangential strain rate is on average positive across the flame front, and therefore the turbulent strain rate field contributes to the enhancement of scalar gradients as in passive scalar turbulence. Although increases in heat release result in larger positive values of the divergence as well as flame-normal and tangential strain rates, the tangential strain rate has a weaker dependence on heat release than the flame-normal strain rate and the divergence.

Heat release rate variations in a globally stoichiometric, stratified iso-octane/air turbulent V-flame

Combustion intensity variations along a globally stoichiometric, stratified iso-octane/air turbulent V-flame were measured in the presence of four different equivalence ratio gradients and compared to a reference fully-premixed case. Instantaneous heat release rate (HRR) images were obtained from the simultaneous acquisition of OH and CH2O Planar Laser Induced Fluorescence (PLIF) images, providing the first direct estimates of local heat release rate in a stratified, iso-octane/air turbulent flame. Detailed numerical simulations were conducted to validate the use of (OH LIF)(CH2O LIF) for measurements of HRR in iso-octane/air flames over a broad range of equivalence ratios. Flame images were analyzed for specific mean equivalence ratio gradient effects for near stoichiometric flame regions within a fixed mean equivalence ratio range of 0.95⩽ϕ⩽1.05. The location of the interrogation window, or region of interest (ROI), was obtained through 3-pentanone tracer PLIF of the reactants and ensured that any observed variation could be attributed specifically to mean equivalence ratio gradient effects, rather than simple mixture strength effects. Results showed small but measurable, decreases in the mean HRR for near stoichiometric flame regions within the ROI as the gradient was varied, and differences of up to 4.4% relative to the reference premixed case were directly attributable to the imposed mean ϕ gradient, separate from potential effects of local mixture strength. By contrast, general stratification effects were also considered for flames of comparable global equivalence ratios but with much broader ranges of mixture strength, and results showed offsetting effects of increased 2D flame length (up to ∼17%) and decreased HRR per flame length (up to ∼22%), resulting in small variations in the overall HRR of ∼9%. The present analysis suggests that gradient effects on local flame properties may differ significantly from general stratification effects in which different ranges of mixture strengths are present, and separating these aspects of partially premixed combustion is critical to the proper interpretation of data, where both may be relevant in different applications.

Time-resolved absolute OH density of a nanosecond pulsed discharge in atmospheric pressure He–H2O: absolute calibration, collisional quenching and the importance of charged species in OH production

The time-resolved OH density in a nanosecond pulsed filamentary discharge in an atmospheric pressure He–H2O(0.05%) mixture is measured using laser induced fluorescence. The lifetime of the excited OH(A) state is found to be strongly time dependent during the plasma pulse, with the shortest decay time occurring at the moment the plasma is switched off. The measured LIF intensity is corrected for this time-resolved quenching and calibrated using Rayleigh scattering. Time-resolved electron density , He (3S1) metastable density , gas temperature and optical emission are presented and used in the interpretation of the observed time dependence of the OH density. Based on these experimental data, it is shown that for the present discharge conditions, OH is mainly produced by charge transfer reactions to water followed by dissociative recombination of the water ion. In addition, two often encountered issues with the calibration of OH LIF are highlighted with an example.

Influence of air diffusion on the OH radicals and atomic O distribution in an atmospheric Ar (bio)plasma jet

Treatment of samples with plasmas in biomedical applications often occurs in ambient air. Admixing air into the discharge region may severely affect the formation and destruction of the generated oxidative species. Little is known about the effects of air diffusion on the spatial distribution of OH radicals and O atoms in the afterglow of atmospheric-pressure plasma jets. In our work, these effects are investigated by performing and comparing measurements in ambient air with measurements in a controlled argon atmosphere without the admixture of air, for an argon plasma jet. The spatial distribution of OH is detected by means of laser-induced fluorescence diagnostics (LIF), whereas two-photon laser-induced fluorescence (TALIF) is used for the detection of atomic O. The spatially resolved OH LIF and O TALIF show that, due to the air admixture effects, the reactive species are only concentrated in the vicinity of the central streamline of the afterglow of the jet, with a characteristic discharge diameter of ∼1.5 mm. It is shown that air diffusion has a key role in the recombination loss mechanisms of OH radicals and atomic O especially in the far afterglow region, starting up to ∼4 mm from the nozzle outlet at a low water/oxygen concentration. Furthermore, air diffusion enhances OH and O production in the core of the plasma. The higher density of active species in the discharge in ambient air is likely due to a higher electron density and a more effective electron impact dissociation of H2O and O2 caused by the increasing electrical field, when the discharge is operated in ambient air.

OH radical and temperature measurements during ignition of H2-air mixtures excited by a repetitively pulsed nanosecond discharge

Ignition delay time, time-resolved temperature, and time-resolved absolute OH concentration are measured in mildly preheated H2–air mixtures (T=473–500K), excited by a repetitive nanosecond pulse discharge in a plasma flow reactor. The measurements are done in decaying plasma after the discharge pulse burst (pulse repetition rate 10–40kHz). Ignition delay increases steeply as the number of pulses in the burst is reduced, exhibiting threshold-like behavior. Temperature and OH concentration during the ignition process are measured by two-line OH LIF, as functions of time delay after the discharge burst. Absolute calibration is done using a near-adiabatic flame. The results show that the temperature at the end of the discharge burst is 700–750K, rising to 1500–1600K within a few ms after the burst, indicative of ignition. During ignition, OH concentration increases by a factor of 20–50, from (0.5–1.1)×1014cm−3 at the end of the burst to peak value of (2.3–2.6)×1015cm−3. Kinetic modeling calculations show good agreement with the experimental results, demonstrating quantitative insight into kinetics of plasma-assisted ignition. Modeling calculations demonstrate that ignition is induced primarily by accumulation of H atoms and gradual temperature rise in the discharge, at temperatures significantly lower than autoignition temperature, by up to 200K.

OH density measurement by time-resolved broad band absorption spectroscopy in an Ar–H2O dielectric barrier discharge

We report results of a novel time-resolved broad-band absorption spectroscopy experiment for OH density measurement applied to a pulsed dielectric barrier discharge in Ar/H2O mixtures. The measurement is aimed at the calibration of our previous OH LIF measurements in the same discharge. The apparatus is simple and cheap, being based on a UV LED light source and a non-intensified, non-cooled, gateable linear CCD array as a detector. The set-up is capable of ruling out both medium/long-term drifts of the UV source and of the discharge, and discharge emission from the measurement. Performances of the set-up are discussed, together with possible improvements for its use as a standalone technique.

Time and spatially resolved LIF of OH in a plasma filament in atmospheric pressure He–H2O

The production of OH in a nanosecond pulsed filamentary discharge generated in pin–pin geometry in a He–H2O mixture is studied by time and spatially resolved laser-induced fluorescence. Apart from the OH density the gas temperature and the electron density are also measured. Depending on the applied voltage the discharge is in a different mode. The maximum electron densities in the low- (1.3 kV) and high-density (5 kV) modes are 2 × 1021 m−3 and 7 × 1022 m−3, respectively. The gas temperature in both modes does not exceed 600 K. In the low-density mode the maximum OH density is at the centre of the discharge filament, while in the high-density mode the largest OH density is observed on the edge of the discharge. A chemical model is used to obtain an estimate of the absolute OH density. The chemical model also shows that charge exchange and dissociative recombination can explain the production of OH in the case of the high-density mode.

C212 高速PLIF・ステレオPIV同時計測による火炎ダイナミクスに関する研究(OS-5: これからの燃焼研究(5))

In this study, a turbulent jet flame is studied using simultaneous high speed laser diagnostics. Information from CH and OH PLIF, as well as stereoscopic PIV was obtained on a single plane at a frequensy of 10 kHz. Analysis of the OH PLIF and velocity data using Extended POD, as well as OH LIF data using DMD, showed that in this particular flame the average of the signal is by far the main component of the signal, with no outstanding strong feature. Such diagnostics and post-processing methods open new perspectives on the experimental study of flow-flame interactions and flame dynamics.

Flame chemiluminescence and OH LIF imaging in a hydrogen-fuelled spark-ignition engine

Research into novel internal combustion engines requires consideration of the diversity in future fuels in an attempt to reduce drastically CO2 emissions from vehicles and promote energy sustainability. Hydrogen has been proposed as a possible fuel for future internal combustion engines. Hydrogen’s wide flammability range allows higher engine efficiency with much leaner operation than conventional fuels, for both reduced toxic emissions and no CO2 gases. This paper presents results from an optical study of combustion in a spark-ignition research engine running with direct injection and port injection of hydrogen. Crank-angle resolved flame chemiluminescence images were acquired and post-processed for a series of consecutive cycles in order to calculate in-cylinder rates of flame growth. Laser induced fluorescence of OH was also applied on an in-cylinder plane below the spark plug to record detailed features of the flame front for a series of engine cycles. The tests were performed at various air-to-fuel ratios, typically in a range of φ = 0.50–0.83 at 1000 RPM with 0.5 bar intake pressure. The engine was also run with gasoline in direct-injection and port-injection modes to compare with the operation on hydrogen. The observed combustion characteristics were analysed with respect to laminar and turbulent burning velocities, as well as flame stretch. An attempt was also made to review relevant hydrogen work from the limited literature on the subject and make comparisons were appropriate.

Experimental and computational study of the transition to the flame ball regime at normal gravity

Ultra lean flames propagating in a tube of 12.2mm internal diameter in mixtures of (40mol% H2+60mol% CH4) fuel gas with air have been studied experimentally, using OH Planar Lased Induced Fluorescence (PLIF), and by numerical simulations, for equivalence ratios between 0.30 and 0.50. Hypothetical OH LIF signals were reconstructed from the calculated data and compared with measurements. The simulated OH LIF intensity distributions are found to be similar to the measured ones in the experimental flames for equivalence ratios 0.31⩽φ⩽0.44. Within this range, when the equivalence ratio was decreased down to the lean limit value, the shapes of the simulated and experimental flame fronts changed from relatively large egg-shaped to small, about 3mm diameter, nearly spherical flame balls. Simulations showed the presence of a recirculating flow in near-limit flames. The front of the lean limit flame balls resides entirely within the volume of this recirculating flow. The flame front was nearly uniform in lean limit flame balls and became increasingly non-uniform with the departure from the limit, with an eventually vanishing bottom front. The experimental and simulated flammability limits lie far below the theoretical limit for a planar flame, and simulated flame temperatures exceed adiabatic temperatures, indicating strong preferential diffusion (Lewis number) effects in these flames. These effects are qualitatively analyzed based on the behavior of measured and simulated flame parameters. It is suggested that preferential diffusion in non-uniform flames, well-removed from the lean flammability limit, is primarily controlled by flame stretch, while in near-uniform limit flame balls it is dominated by a curvature effect similar to that in microgravity flame balls.

Flame front tracking in turbulent lean premixed flames using stereo PIV and time-sequenced planar LIF of OH

This paper describes the simultaneous application of time-sequenced laser-induced fluorescence imaging of OH radicals and stereoscopic particle image velocimetry for measurements of the flame front dynamics in lean and premixed LP turbulent flames. The studied flames could be acoustically driven, to simulate phenomena important in LP combustion technologies. In combination with novel image post processing techniques we show how the data obtained can be used to track the flame front contour in a plane defined by the illuminating laser sheets. We consider effects of chemistry and convective fluid motion on the dynamics of the observed displacements and analyse the influence of turbulence and acoustic forcing on the observed contour velocity, a quantity we term as sd2D. We show that this quantity is a valuable and sensitive indicator of flame turbulence interactions, as (a) it is measurable with existing experimental methodologies, and (b) because computational data, e.g. from large eddy simulations, can be post processed in an identical fashion. sd2D is related (to a two-dimensional projection) of the three-dimensional flame displacement speed sd, but artifacts due to out of plane convective motion of the flame surface and the uncertainty in the angle of the flame surface normal have to be carefully considered. Monte Carlo simulations were performed to estimate such effects for several distributions of flame front angle distributions, and it is shown conclusively that sd2D is a sensitive indicator of a quantity related to sd in the flames we study. sd2D was shown to increase linearly both with turbulent intensity and with the amplitude of acousting forcing for the range of conditions studied.

Development of a two-line OH-laser-induced fluorescence thermometry diagnostics strategy for gas-phase temperature measurements in engines

This study aims at optimizing two-line OH thermometry strategies for in-cylinder measurement in internal combustion engines. Various aspects are investigated experimentally, such as the selection of suitable OH lines and the possibility of using a single calibration coefficient for variable mixture composition, temperature, and pressure conditions. Two kinds of experimental systems have been investigated. First, a laminar methane-air burner flame at atmospheric pressure, whose stability allowed the determination of OH-laser-induced fluorescence (LIF) intensity ratios from nonsimultaneous imaging. The temperature distribution in the flame is presented for OH-transition pairs with various temperature sensitivities. The burner flame was studied for equivalence ratios from phi=0.93 to 1.30 in order to check for the stability of calibration over various flame conditions. Additionally, OH LIF images were acquired in an optical engine for the chosen OH transitions yielding data about the effect of pressure on OH LIF signals under realistic experimental conditions.

Computational and experimental study of steady axisymmetric non-premixed methane counterflow flames

We investigated computationally and experimentally the structure of steady axisymmetric, laminar methane/enriched-air diffusion flames. Experimentally, we imaged simultaneously single-photon OH LIF and two-photon CO LIF, which also yielded the forward reaction rate (RR) of the reaction CO + OH → CO2+ H. In addition, particle image velocimetry (PIV) was used to measure the velocity in the proximity of the fuel and oxidizer nozzles, providing detailed boundary conditions for the simulations. Computationally, we solved implicitly the steady state equations in a modified vorticity–velocity formulation on a non-staggered, non-uniform grid. We compared the results along the axis of symmetry from the two-dimensional simulations with those from the one-dimensional model, and showed consistency between them. The comparison between the experimental and computational data yielded excellent agreement for all measured quantities. The field of these two-dimensional flames can be roughly partitioned into two regions: the region between the two reactant nozzles, in which viscous and diffusive effects are confined to the mixing layer and to the nozzle walls, where separation occurs; and a radial development region, which is initially confined by recirculation zones near the burner flanges. Buoyancy is virtually irrelevant in the first region at all but the smallest, and practically irrelevant, strain rates. Buoyancy, on the other hand, does play a role in the growth of the recirculation zones, and in determining the flame location in the outermost region.

Effect of velocity gradient on propagation speed of tribrachial flames in laminar coflow jets

The effect of velocity gradient on the propagation speed of tribrachial flame edge has been investigated experimentally in laminar coflow jets for propane fuel. It was observed that the propagation speed of tribrachial flame showed appreciable deviations at various jet velocities in high mixture fraction gradient regime. From the similarity solutions, it was demonstrated that the velocity gradient varied significantly during the flame propagation. To examine the effect of velocity gradient, detail structures of tribrachial flames were investigated from OH LIF images and Abel transformed images of flame luminosity. It was revealed that the tribrachial point was located on the slanted surface of the premixed wing, and this slanted angle was correlated with the velocity gradient along the stoichiometric contour. The temperature field was visualized qualitatively by the Rayleigh scattering image. The propagation speed of tribrachial flame was corrected by considering the direction of flame propagation with the slanted angle and effective heat conduction to upstream. The corrected propagation speed of tribrachial flame was correlated well. Thus, the mixture fraction gradient together with the velocity gradient affected the propagation speed.