Universal calibration of quantitative OH-LIPF measurements in hydrocarbon flames at elevated pressures

Quantitative measurements of radical concentrations are important in studies of the structure of flames. In this research, a quantitative analysis was performed using laser-induced predissociative fluorescence of OH radicals (OH-LIPF) in high-pressure ($1$–$5$ atm) premixed methane-air and propane-air flat flames $\phi=0.7-1.3$. OH was excited ($A^2\Sigma^+$, $\nu'= 3-X^2\Pi$, $\nu''=0$, and $\mathrm{P}_28$) using a $\mathrm{KrF}$ excimer laser and $(3,2)$ band fluorescence was observed. OH fluorescence intensities were calibrated against the OH concentrations calculated from flame simulation results in the postflame zone using the CHEMKIN premix flame code in conjunction with the GRI-Mech $3.0$ reaction mechanism. The accuracy and spatial resolution of temperature measurements are important factors for the correctness of the corresponding flame simulations, especially in the reaction zone near the burner surface. In this work, a carefully constructed thermocouple ($\mathrm{R}$-type, $50\mu$m) positioning system was used to identify the temperatures above the burner surface. With careful evaluations of quenching rates, Voigt profiles, and normalization against room-air $\mathrm{N}_2$ Raman scattering intensity, a universal calibration constant [$C_T=(1.076 \pm 0.174)\cdot 10^{16}$ molecules/cm$^3$] was determined. The OH concentrations obtained by flame simulations showed good agreement with the quantitative OH-LIPF measurements in all methane-air flames and fuel lean $(\phi=0.7-0.8)$ propane-air flames. However, a $2$-fold to $5$-fold discrepancy was obtained in propane flames at $\phi>0.9$. This may be caused by the lack of $\mathrm{C}_3$ reaction paths in the GRI mechanism and/or the inaccuracy of the thermochemical data for large molecules.