Simultaneous measurement of localized heat release with OH/CH2O-LIF imaging and spatially integrated OH∗ chemiluminescence in turbulent swirl flames

The in-situ and localized observation of heat release in turbulent flames is important for the validation of computational modeling of turbulent flows with combustion. In the present work we obtain localized information on heat release rate (HRR) by the commonly accepted technique of the simultaneous and single-shot planar imaging of OH and CH₂O concentrations by laser-induced fluorescence (LIF). Additionally, we combine this with the simultaneous line-of-sight and temporally resolved chemiluminescence detection of OH^?, spatially integrated within the flame volume, interrogated by the laser sheets used for the HRR imaging technique. The combined diagnostic methods are demonstrated for a swirl-stabilized, premixed turbulent methane/air flame of 30-kW thermal power, and they show the existence of correlations between both HRR-sensitive diagnostic techniques.     

Heat release rate as represented by [OH] × [CH2O] and its role in autoignition

Data from a recent instantaneous, simultaneous, high-resolution imaging experiment of Rayleigh temperature and laser induced fluorescence (LIF) of OH and CH₂O at the base of a turbulent lifted methane flame issuing into a hot vitiated coflow are analysed and contrasted to reference flames to further investigate the stabilization mechanisms involved. The use of the product of the quantified OH and semi-quantified CH₂O images as a marker for heat release rate is validated for transient autoigniting laminar flames. This is combined with temperature gradient information to investigate the flame structure. Super-equilibrium OH, the nature of the profiles of heat release rate with respect to OH mole fraction, and comparatively high peak heat release rates at low temperature gradients is found in the kernel structures at the flame base, and found to be indicative of autoignition stabilization.     

Quantitative measurement of hydroxyl radical(OH) concentration in premixed flat flame by combining laser-induced fluorescence and direct absorption spectroscopy

An accurate and reasonable technique combining direct absorption spectroscopy and laser-induced fluorescence(LIF)methods is developed to quantitatively measure the concentrations of hydroxyl in CH_4/air flat laminar flame. In our approach, particular attention is paid to the linear laser-induced fluorescence and absorption processes, and experimental details as well. Through measuring the temperature, LIF signal distribution and integrated absorption, spatially absolute OH concentrations profiles are successfully resolved. These experimental results are then compared with the numerical simulation. It is proved that the good quality of the results implies that this method is suitable for calibrating the OH-PLIF measurement in a practical combustor.     

Absolute radical concentration measurements in low-pressure H2/O2 flames during the combustion of graphite

Absolute CN and CH radical concentrations were determined in situ during the combustion of a graphite substrate in premixed, laminar, low-pressure, H₂/O₂ flames for two different equivalence ratios, = 1.0 and = 1.5. For CN measurements, a small amount of NO (1.8%) was added. The concentration of CN was measured by cavity ring-down spectroscopy (CRDS) probing the absorption of the P_1,2 (13) in the B–X (0, 0) band at 388.1 nm, and the concentration of CH was measured by linear unsaturated laser-induced fluorescence (LIF) exciting the fluorescence of the R₁ (4) in the B–X (0, 0) band at 387.4 nm. Temperature measurements were done based on LIF excitation spectra of OH in the A–X (0, 0) band. It was found that the graphite substrate reduces the flame temperature in the vicinity of its surface. The CN concentrations were found to be three times higher for the rich flame than for the stoichiometric flame. CH concentrations were slightly higher for the stoichiometric flame than for the rich flame. The observed CH/CN concentration ratio is substantially lower compared to NO-doped low-pressure CH₄/O₂ flames. The obtained quantitative information can serve as a first calibration point for detailed numerical simulations of the burning solid graphite, which are based on the concept of surface elementary reactions.     

An experimental and numerical study on the adequacy of CH as a flame marker in premixed methane flames

The CH radical is frequently used as a flame marker because it is relatively short-lived and is present over a narrow region in flames. Discontinuities in the CH field are thus often interpreted as localized extinction of the flame. Recently, however, the adequacy of CH laser-induced fluorescence (LIF) as a flame marker was questioned by an experimental study of flame–vortex interactions in highly N₂-diluted premixed methane flames. We demonstrate both experimentally and numerically that anomalies in the transient response of CH in this earlier study were due to reactant composition variations in the vortex. In addition, we evaluate the adequacy of CH LIF as a flame marker over a much broader range of conditions. Previous numerical studies showed that heat release rate correlates reasonably well with peak [HCO] and the concentration product [OH][CH₂O], but poorly with [CH], in highly N₂-diluted premixed methane flames. Here, the correlation between heat release rate and CH is investigated both experimentally, by performing simultaneous measurements of CH, OH, and CH₂O LIF, and numerically. We consider undiluted and N₂-diluted premixed methane flames over a range of strain rates and stoichiometries. Results are reported for flames subjected to unsteady stretch and reactant composition variations. For all N₂-dilution levels considered, the peak CH LIF signal correlates poorly with heat release rate when the stoichiometry of the reactant mixture changes from rich to lean. However, when flames are subjected to stretch, the correlation between CH and heat release rate improves as the N₂-dilution level decreases. The correlation is reasonably good for undiluted flames with equivalence ratios of 0.8 < Φ < 1.2. This result is particularly encouraging, given the relevance of undiluted flames to practical applications, and it motivates further investigation of the parameter space for which difficulties may exist in using CH as a flame marker.     

Effect of laser intensity and of lower-state rotational energy transfer upon temperature measurements made with laser-induced fluorescence

4 –air flame, with OH at 2000 K. We calculate the ratio of LIF intensities that would be induced by doubled dye-laser light near 283 nm, by means of the A←X, 1←0, P₁(7), and Q₂(11) transitions in OH. Here we show that the ratio of LIF signals from those two transitions, and thus the deduced temperature, is sensitive to laser intensity. That is caused mainly by the competition between laser-pumping of molecules out of the lower rotational state and of rotational energy transfer (RET) collisions into that state. A-state collisional effects are normally important, but are minimized here by assuming that they are the same for both transitions. The laser spectral intensity dependence of the fluorescence ratio depends heavily upon the value of the RET coefficients within the X-state. While RET reduces the sensitivity of the observed signal to the laser spectral intensity, the conversion of a measured fluorescence ratio to temperature is particularly difficult. That is because RET rates, and quenching rates, can be a function of local conditions and of the rotational state being populated. Two different models are used to demonstrate these effects, and both predict large effects upon temperature. Received: 19 February 1998/Revised version: 16 June 1998     

Effects of signal corrections on measurements of temperature and OH concentrations using laser-induced fluorescence

Temperature and OH concentrations derived from OH laser-induced fluorescence (LIF) are known to be susceptible to effects such as collisional quenching, laser absorption, and fluorescence trapping. In this paper, a set of analytical and easy-to-implement methods is presented for treating these effects. The significance of these signal corrections on inferred temperature and absolute OH concentration is demonstrated in an atmospheric-pressure, near-stoichiometric CH₄-air flame stabilized on a Hencken burner, for laser excitation of both the A²Σ⁺←X²Π (0,0) and (1,0) bands. It is found that the combined effect of laser attenuation and fluorescence trapping can cause considerable error in the OH number density and temperature if not accounted for, even with A–X(1,0) excitation. The validity of the assumptions used in signal correction (that the excited-state distribution is either thermalized or frozen) is examined using time-dependent modeling of the ro-vibronic states during and after laser excitation. These assumptions are shown to provide good bounding approximations for treating transition-dependent issues in OH LIF, especially for an unknown collisional environment, and it is noted that the proposed methods are generally applicable to LIF-based measurements.     

Single-shot laser-induced fluorescence imaging of formaldehyde with XeF excimer excitation

Single-shot formaldehyde laser-induced fluorescence (LIF) imaging measurements in a technical scale turbulent flame have been obtained using XeF excimer laser excitation in the ?¹A₂-˜X¹A₁ transition at 353.2 nm. Measurements have been carried out in a 150 kW natural gas swirl burner where formaldehyde distribution fields have the potential, in combination with OH concentration fields, to visualize the heat release distribution and therefore give an optimal visualization of flame-front positions. The extended areas where formaldehyde was detected in the swirl flame indicates the presence of low temperature chemistry in preheated gas pockets before ignition. Received: 31 January 2000 / Revised version: 2 March 2000 / Published online: 5 April 2000     

Influence of ferrocene addition to a laminar premixed propene flame: Laser diagnostics, mass spectrometry and numerical simulations

The detailed influence of ferrocene in a low-pressure, fuel-rich, laminar, premixed propene/oxygen/argon flat flame was investigated experimentally using molecular beam sampling mass spectrometry (MBMS), laser-induced fluorescence (LIF), and compared to numerical simulations. MBMS was applied to analyze the species profiles of important intermediates in the flames with and without ferrocene doping. The concentration profile of iron atoms was measured with absorption sensitive LIF, which provides absolute number densities without additional calibrations. The flame temperature was obtained by two-line OH LIF measurements. One dimensional numerical simulations of the flames using detailed models from the literature were performed and the modeling results are compared with the experimental measurements. The iron measurements show reasonable agreement with the numerical simulation, while some discrepancies were found at larger heights. The MBMS measurements show a decrease in flame velocity when ferrocene was added, which was not provided by the model.     

REMPI as a tool for studies of OH radicals in catalytic reactions

2 +? O₂→H₂O on polycrystalline Pt foils has been studied by detection of desorbing OH radicals using the Resonance Enhanced Multiphoton Ionization, REMPI, technique. The measurements were performed at catalyst temperatures of 1000–1400 K and a total pressure below 10^-4 mbar. The studies of OH desorption by REMPI were achieved using a two-photon excitation D²Σ^-–X²Π (1–0), followed by one-photon ionization. The ions were detected in a time-of-flight mass spectrometer, TOF-MS, in order to avoid interference from non-resonantly ionized molecules. By applying TOF-MS, a simultaneous non-resonant ionization and detection of H₂, O₂ and H₂O was achieved. Recorded REMPI spectra were compared with spectra simulated using known molecular constants. The kinetics of the reaction derived from the measurements were compared with what was obtained in earlier LIF detection of OH, performed at higher total reactant pressure using the A–X transition. REMPI TOF-MS is shown to be a complement to LIF for reaction studies below 5×10^-4 mbar total pressure, where LIF is too insensitive for quantification. The reaction kinetics was found to be in agreement with a theoretic model and previous LIF studies. Received: 8 March 1996/Revised version: 4 October 1996     

Laser imaging system for determination of three-dimensional scalar gradients in turbulent flames

An imaging system for the measurement of three-dimensional (3D) scalar gradients in turbulent hydrocarbon flames is described. Combined line imaging of Raman scattering, Rayleigh scattering, and CO laser-induced fluorescence (LIF) allows for simultaneous single-shot line measurements of major species, temperature, mixture fraction, and a one-dimensional surrogate of scalar dissipation rate in hydrocarbon flames, while simultaneous use of two crossed, planar LIF measurements of OH allows for determination of instantaneous flame orientation. In this manner the full 3D scalar dissipation can be estimated in some regions of a turbulent flame on a single-shot basis.     

Application of advanced laser diagnostics for the investigation of the ionization sensor signal in a combustion bomb

The ionization sensor is an electrical probe for diagnostics in internal combustion engines. Laser-induced fluorescence (LIF) imaging of fuel, hydroxyl (OH), and nitric oxide (NO) distributions has been employed to extend our knowledge about the governing processes leading to its signal. By monitoring the flame propagation in quiescent and turbulent mixtures, the cycle-to-cycle variations in the early sensor signal was attributed to the stochastic contact between flame front and electrodes. An analysis of the relationship between gas temperature and sensor current in the post-flame gas suggests a dominant role of alkali traces in the ionization process at the conditions under study. Significant cooling of the burned gas in the vicinity of the electrodes was observed in quiescent mixtures. Imaging of the post-flame gas in turbulent combustion revealed moving structures with varying NO and OH concentrations, which were identified as sources of variation in the sensor current. PACS 51.50.+v; 42.62.Fi; 47.27.-i     

Experimental study of signal trapping of OH laser induced fluorescence and chemiluminescence in flames

The absorption of OH^∗ chemiluminescence and laser-induced fluorescence (LIF) in the exhaust gas of confined premixed laminar CH₄/air flames at atmospheric pressure was investigated. One flame was used as source and a second as absorber. OH LIF was excited in the ν″=0→ν′=1 band of the A–X electronic system around ≈283 nm and spectrally resolved detected in the (0,0) and (1,1) vibrational bands around 305–320 nm. For OH^∗ chemiluminescence, spectrally resolved detection was performed in the wavelength range 280–340 nm. For an absorption path of 54 mm and at T≈2000 K, signal trapping on the order of 10–40% was observed. Signal trapping was most pronounced in the (0,0) band, as expected from the thermal population distribution of OH in the electronic ground state. The spectral distribution of the signals and the wavelength dependence of the signal trapping are addressed in this paper. Implications from the results with respect to detection strategies and chemiluminescence-based equivalence ratio measurements are discussed.     

Flame turbulences recorded at 1 kHz using planar laser induced fluorescence upon hot band excitation of OH radicals

Laser-induced fluorescence (LIF) in flames is excited by a diode-pumped all-solid-state disk laser system which operates at a pulse repetition rate of 1 kHz and a tunable wavelength around 1030 nm. The laser fundamental is converted to 343 nm and used to excite into the hot band transition of OH radicals in H₂/O₂ diffusion flames of an industrial burner and in the premixed flame of a microburner. The OH radical emission around 308 nm is resolved spectrally and spatially using a light-sheet technique. Imaging of the planar LIF (PLIF) by a gated camera visualizes the turbulent flame behavior on the millisecond time scale without averaging. To our knowledge this is the first time that an all-solid-state laser providing at the same time a kHz repetition rate as well as pulse energies of up to 5.5 mJ is available for PLIF observation of OH radicals. PACS 33.50.-j; 42.55.Xi; 82.33.Vx     

Experimental analysis of local flame extinction in a turbulent jet diffusion flame by high repetition 2-D laser techniques and multi-scalar measurements

In this paper, we present a detailed experimental study of turbulence chemistry interactions in the “DLR_B” turbulent jet diffusion flame. The flame operates on mixtures of CH₄, H₂, and N₂ in the fuel stream at Re = 22,800 and is a target flame within the TNF workshop. Extinction and re-ignition events can be tracked in real time and related to the underlying flow field phenomena and temperature fields. Time resolved measurements of OH radical concentration fields are performed in combination with temperature and velocity field measurements. For this purpose, we combined high repetition rate (33 kHz) PLIF imaging with stereoscopic PIV and double pulse Rayleigh imaging techniques. Comparisons are made with results from multi-scalar Raman/Rayleigh/LIF point measurements that reveal the thermochemical state of the flame. The large deviations from equilibrium observed on resulting OH/temperature joint pdfs could be related to strain rate and Damköhler number variations caused by turbulent flow structures leading to frequent extinctions. The 2D measurement series uniquely reveal the underlying mechanism that can lead to such events. Finally, comparisons are made to strained laminar flame calculations, which are generally found to be in good agreement with the measured data.     

Evaluation of heat release indicators in lean premixed H2/Air cellular tubular flames

Heat release rate in combustion systems must be understood in order to control thermoacoustic instabilities, flame extinction, and heat losses. Traditionally OH chemiluminescence (OH*) is used to trace heat release rate (HRR) in H₂/air flames, but its accuracy as a tracer has not been assessed. Lean premixed H₂/air cellular tubular flames are a good test case to evaluate HRR tracers due to the presence of highly reactive flame cells surrounded by regions of near extinction. Comparing the calculated heat release rate to OH* concentration, one finds that [OH*] profiles correlate with the regions of high reactivity (flame cells) but the correlation fails in the low reactivity regions where the HRR is much higher than the [OH*] value indicates. Alternate HRR tracers including [H] and pixel-by-pixel products of [O₂]x[H], [OH]x[H₂], and [O]x[H₂] are analyzed with detailed numerical simulations. The chosen products derive from the main chain reaction steps that contribute to overall HRR in lean, premixed H₂/air flames. Findings suggest that [H] is an accurate yet simple way of tracking HRR. Planar measurements of HRR are possible if LIF measurements of [H] are improved.     

Energy transfer in the A2Σ+ state of OH following v’ = 1 excitation in a low pressure CH4/O2-flame

2 Σ⁺(v’=1) level of OH. Measurements were performed in a laminar premixed flame at 10 Torr total pressure. The low pressure allowed the spatial variation of the effective quenching rate to be determined through the flame front. In addition, the dependence of the quenching rate on rotational quantum number was measured by exciting a series of rotational lines in the range N’=0–16. The results show that the total quenching rate decreases only 17% through the flame front, in the region where OH can be detected. Nevertheless, the absolute value of the quenching rate Q is required if absolute concentrations are to be determined from LIF-signals. The variation both of Q and of the rotational relaxation rate with excited rotational quantum state must be known for quantification of LIF-temperature measurements via the Boltzmann relation. Finally, the rotational and vibrational energy transfer (RET, VET), was investigated by recording the spectrally and temporally resolved fluorescence. For all excited rotational lines, efficient RET to neighbouring rotational states was observed, but only very little VET. Total RET rates were determined from the difference between the time-resolved broadband (total fluorescence) and narrowband (fluorescence from the laser excited level) curves. The experimental results were compared with simulations using a dynamic model, which describes the energy transfer for flame conditions. With the available input data (temperature, major species concentrations and collision-partner specific RET cross sections), good agreement was obtained. Received: 3 February 1997/Revised version: 3 September 1997     

The stabilization mechanism and structure of turbulent hydrocarbon lifted flames

The structure and stabilization mechanism of turbulent lifted non-premixed hydrocarbon flames have been investigated using combined laser imaging techniques. The techniques include Rayleigh scattering, laser induced predissociation fluorescence of OH, LIF of PAH, LIF of CH₂O, and planar imaging velocimetry. The geometrical structure of multi-reaction zones and flow field at the stabilization region have been simultaneously measured in 16 hydrocarbon flames. The data reveal the existence of triple flame structure at the stabilization region of turbulent lifted flames. Increasing the jet velocity leads to an increase of the lift-off height and to a broadening of the lift-off region. Further analysis of the stabilization criterion at the lift-off height based on the premixed nature of triple-flame propagation and flow field data has been presented and discussed.     

A pulse-pileup correction procedure for rapid measurements of hydroxyl concentrations using picosecond time-resolved laser-induced fluorescence

The measurement of fluorescence lifetimes is important for determining minor-species concentrations in flames when using linear laser-induced fluorescence (LIF). Applications of LIF to turbulent flames require that the quenching rate coefficient be determined in less than ∼100 μs. Moreover, the measurement technique must be insensitive to the existence of relatively large backgrounds, such as occur from flame emission. To meet these goals, we have recently developed a rapid, gated photon-counting technique, termed LIFTIME. However, for ultimate application to turbulent time-series measurements, LIFTIME must be extended to photon count rates which unfortunately result in nonlinear discriminator operation. In this paper, a correction technique is derived to permit quantitative measurements of fluorescence lifetimes and concentrations at sampling rates up to 4 kHz. The technique was tested against liquid samples having a known lifetime and is further shown to reproduce previous hydroxyl concentration measurements in a series of laminar flames with total photon count rates of up to ∼35 million detected photoelectrons per second. The fluorescence lifetimes and hydroxyl concentrations are shown to be measured with ∼10% accuracy (68% confidence interval) for sampling times as low as 250 μs. Received: 9 October 1998 / Revised version: 30 December 1998 / Published online: 28 April 1999     

Instantaneous three-dimensional visualization of concentration distributions in turbulent flows with crossed-plane laser-induced fluorescence imaging

A laser-based technique for measuring instantaneous three-dimensional species concentration distributions in turbulent flows is presented. The laser beam from a single laser is formed into two crossed light sheets that illuminate the area of interest. The laser-induced fluorescence (LIF) signal emitted from excited species within both planes is detected with a single camera via a mirror arrangement. Image processing enables the reconstruction of the three-dimensional data set in close proximity to the cutting line of the two light sheets. Three-dimensional intensity gradients are computed and compared to the two-dimensional projections obtained from the two directly observed planes. Volume visualization by digital image processing gives unique insight into the three-dimensional structures within the turbulent processes. We apply this technique to measurements of toluene-LIF in a turbulent, non-reactive mixing process of toluene and air and to hydroxyl (OH) LIF in a turbulent methane-air flame upon excitation at 248 nm with a tunable KrF excimer laser. PACS 42.30.Va; 32.50.+d; 42.62.Cf