Michael W. Renfro

Experimental study of temperature and CH radical location in partially premixed CH4/air coflow flames

Abstract As part of an ongoing investigation of an exhaust NOx emission index minimum measured for partially premixed flames, radial temperature profiles and CH radical locations were measured in atmospheric-pressure, partially premixed, coflow, methane/air flames with fuel-side equivalence ratios of 1.6, 2.0, and 3.5, at three axial heights above the burner. The work was undertaken because of the importance of temperature and CH radical behavior in NO formation chemistry. Thin-filament pyrometry was found to be more appropriate than thermocouple thermometry for temperature measurements in partially premixed flames. Results demonstrated that the 1.6-equivalence-ratio flame exhibited classical double-flame structure, the 2.0-equivalence-ratio flame was a merged flame, and the 3.5-equivalence-ratio flame exhibited diffusion-flame structure. Signals from CH∗ chemiluminescence and CH laser-induced fluorescence provide evidence that, for the present measurement locations, double flames exhibit single CH peaks which can be associated with their premixed component flames. Double CH radical peaks, which were predicted to occur in low-strain-rate flames, were not found for the limited number of flame conditions and locations studied. In the near-burner region, the premixed and nonpremixed component flames of the ΦB = 1.6 double flame diverge radially with increasing downstream distance and merge together for larger values of ΦB.

Scalar time-series measurements in turbulent CH4/H2/N2 nonpremixed flames: CH

Time-series measurements of CH concentrations [CH] are reported in a series of methane/hydrogen/nitrogen diffusion flames. Power spectral densities (PSDs) and autocorrelation functions are computed from the time series, permitting a detailed investigation of [CH] fluctuation time scales. The effects of fluorescence lifetime fluctuations are found to be negligible for quantitative determination of the PSDs. A similar study of hydroxyl concentrations in nonpremixed hydrogen/argon flames recently demonstrated that OH PSDs collapse to a single curve when normalized by the integral time scale, in agreement with mixture fraction statistics. However, for the present measurements, [CH] data on the fuel-side of the peak [CH] location exhibit low-frequency differences in their PSDs with respect to both the [OH] and air-side [CH] PSDs. These differences could be associated with the closer proximity of the CH radical to the shear layer and thus to enhanced fluctuations in the local mixture fraction. This possibility can be examined via measurements of other scalars.

Quantitative hydroxyl concentration time-series measurements in turbulent nonpremixed flames.

Quantitative hydroxyl concentration time-series measurements have been obtained by picosecond time-resolved laser-induced fluorescence in a series of methane-air and hydrogen-argon-air nonpremixed flames. The recovery of a quantitative time series is complicated by the need to account for fluctuations in the fluorescence lifetime. We have recently developed instrumentation that enables the simultaneous measurement of fluorescence signal and lifetime. The present research represents the first application of this technique to turbulent flames. The correction for hydroxyl lifetime fluctuations is shown to be significant for mean concentrations and thus probability density functions but negligible for power spectral densities (PSDs). The hydroxyl PSDs were found to vary slightly with radial and axial location in the flames and to vary significantly with Reynolds number. However, the PSDs in the H(2)-Ar-air flames are nearly identical to those in the CH(4)-air flames.

Hydroxyl time-series measurements in laminar and moderately turbulent methane/air diffusion flames

Picosecond time-resolved laser-induced fluorescence (PITLIF) is a developing technique used to probe minor-species concentrations in flames at rates sufficient for the study of turbulent fluctuations. This method has previously been applied to the measurement of CH signals in laminar and low-Reynolds number turbulent diffusion flames. In the present work, Laser-induced Fluorescence (LIF) measurements of OH are obtained in the same flames. The peak concentrations of the hydroxyl radical provide a signal-to-noise ratio sufficient to extend the measurements to a variety of axial and radial locations relative to the flame front. The time-series measured are used to compute power spectral densities (PSDs) which provide frequency information not available from more typical probability density functions. The PSDs are compared to those from the previous CH work and to expectations for scalars in moderately turbulent flames from the literature. These measurements provide the first known PSDs of the hydroxyl radical in a low-Reynolds number turbulent nonpremixed flame.

Time-series measurements of CH concentration in turbulent CH 4 /air flames by use of picosecond time-resolved laser-induced fluorescence

We report a developing technique capable of making continuous time-series measurements of naturally occurring minor-species concentrations. The high repetition rate of the mode-locked laser used in this technique allows for the study of transient combustion events, such as turbulence, and their effect on minor-species concentrations. The technique is applied to make CH fluorescence time-series measurements and to calculate power spectral densities in a turbulent nonpremixed flame. To our knowledge, the reported time series represents the first such measurement for a naturally occurring minor species in a turbulent flame.

Scalar time-series simulations using flamelet state relationships for turbulent non-premixed flames

Simulations of scalar (OH, CH, and number density) time series and comparisons to experimental data are presented using a laminar flamelet model. Realistic time series for mixture fraction (Z) were constructed by employing measured Z mean and rms values in conjunction with realistic power spectral densities (PSDs), probability density functions (PDFs), and integral time scales. A unique procedure was implemented to permit simultaneous specification of both the PSD and PDF shapes for Z. These Z time series were mapped to other scalar time series by using flamelet state relationships from a strained laminar flame code. The simulated statistics are compared to recent data in hydrogen/methane/nitrogen flames. The predictions of OH and CH time scales directly depend on the time scale for Z; however, they are not identical because of the narrow state relationships for reactive scalars. The model successfully captures complicated features in the radial distribution of OH integral time scales. In a separate inverse calculation, the simulation is used to estimate Z time scales from each of the measured OH, CH, and number density data. In each case the Z time scale is found to be ∼0.75 ms on the jet centerline. In contrast to time scales from non-reacting jet studies, this Z time scale is nearly invariant with axial height and at low axial heights varies only slowly with radial location, implying that convective scaling (jet width/local velocity) may be insufficient for the accurate description of mixing time scales in jets with heat release.

SEMI-QUANTITATIVE MEASUREMENTS OF CH CONCENTRATION IN ATMOSPHERIC-PRESSURE COUNTERFLOW DIFFUSION FLAMES USING PICOSECOND LASER-INDUCED FLUORESCENCE

Measurements of relative CH concentrations and absolute fluorescence lifetimes are reported in atmospheric pressure, laminar, counterflow diffusion flames using picosecond laser-induced fluorescence combined with a novel photon-counting technique. Three fuels consisting of nitrogen-diluted methane, ethane, or acetylene were used with global strain rates varying from 20 s-1 to 60 s-1. The concentrations were normalized by a measurement in one of the methane flames. The resulting (CH) ratios are corrected for quenching effects and are quantitative with uncertainties of typically ±7% (95% confidence interval). The results were compared to predictions from OPPDIF using two chemical mechanisms, GRI-2·11 and GRI-3·0. Both mechanisms correctly predict the relative CH concentrations between the ethane and methane flames but overpredict those between the acetylene and methane flame. Predictions of fluorescence lifetimes were computed from the OPPDIF species and temperature profiles using species-dependent quenching cross sections from the literature. The predicted fluorescence lifetimes are larger than the measurements by 10—15% in the methane and acetylene flames and by 50% in the ethane flames.

Photon-counting technique for rapid fluorescence-decay measurement

We report on a novel laser-induced fluorescence triple-integration method (LIFTIME) that is capable of making rapid, continuous fluorescence lifetime measurements by a unique photon-counting technique. The LIFTIME has been convolved with picosecond time-resolved laser-induced fluorescence, which employs a high-repetition-rate mode-locked laser, permitting the eventual monitoring of instantaneous species concentrations in turbulent flames. We verify the technique by application of the LIFTIME to two known fluorescence media, diphenyloxazole (PPO) and quinine sulfate monohydrate (QSM). PPO has a fluorescence lifetime of 1.28 ns, whereas QSM has a fluorescence lifetime that can be varied from 1.0 to 3.0 ns. From these liquid samples we demonstrate that fluorescence lifetime can currently be monitored at a sampling rate of up to 500 Hz with less than 10% uncertainty (1 sigma) .

Cross sections for quenching of CH A 2δ, ν′=0, by N2 and H2O from 1740 to 2160 K

Measurements of time-resolved CH fluorescence are reported for 77 methane-based counterflow diffusionflames at atmospheric pressure. Seven oxidizer mixtures, with variations in O 2 /N 2 /Ar mole fractions, are used with each of 11 CH 4 /N 2 /Ar fuel mixtures to vary the resulting nitrogen mole fraction without altering the flame temperature. A wide range of fluorescence lifetimes is observed at the locations of peak CH concentration. A procedure for extracting single-collider thermally averaged cross sections for nitrogen in these flames is developed despite the complicated collisional environments. Because of linear relationships between nitrogen and other collider mole fractions in the 7 oxidizer mixtures, the effects of quenching from species other than nitrogen can be subtracted without knowledge of their quenching rate coefficients. The 11 fuel mixtures are used to parametrically assess the N 2 cross-section dependence on temperature from 1740 to 2160 K. The results show that N 2 quenching of CH A 2 Δ, ν′=0, at flame temperatures is less efficient by a factor of 3 than the correlation of Tamura et al., which relies on extrapolation from a single measurement at 1300 K. Using the improved nitrogen cross-section correlation of σ Q,N 2 =1.53×10 −4 T 1.23 exp(−522.1/ T ), water is found to be an equally important quencher of CH in flames. The collision environment for the 77 diffusion flames is predicted using OPPDIF with GRI-Mech 3.0 kinetics, and the resulting data are used to suggest improvements to the H 2 O cross section. The present data are consistent with a constant H 2 O thermally averaged cross section versus temperature as opposed to a constant quenching rate coefficient.

Laser-induced Fluorescence Triple-integration Method Applied to Hydroxyl Concentration and Fluorescence Lifetime Measurements

We report quantitative hydroxyl concentrations obtained by using a new laser-induced fluorescence triple-integration method (LIFTIME), which is capable of rapid and continuous fluorescence lifetime measurements via a unique photon-counting technique. LIFTIME has been convolved with picosecond time-resolved laser-induced fluorescence to permit the rapid monitoring of instantaneous species concentrations in flames. Here, LIFTIME is used to measure hydroxyl concentrations and fluorescence lifetimes at a sampling rate of 1 Hz in eight premixed laminar flat flames and in one laminar opposed flow diffusion flame. Fluorescence lifetimes as a function of axial position are generally obtained with less than 5% uncertainty, while concentrations at the same locations are obtained with less than 10% uncertainty (95% confidence interval). The hydroxyl concentration measurements are shown to agree well with modeling predictions and with previous laser-saturated fluorescence measurements. The measurements are also compa...