Measurements of pressure broadening of N2 in the anisotropic tensor component of spontaneous Raman spectra

Abstract

Spontaneous Raman scattering from nitrogen is well-suited for temperature measurements in combustion experiments, especially at high pressure which increases the Raman signal because of the higher density in the measurement volume. In this work we investigate high density effects on the anisotropic tensor component of ro-vibrational spontaneous Raman scattering, which must be understood to obtain accurate thermometry in high pressure gases using high-fidelity Raman simulations. We measure the collision broadening for the anisotropic component of spontaneous Raman scattering from room temperature nitrogen over the pressure range 10–70 atm for three gas compositions: pure nitrogen, air, and nitrogen in argon. Line broadening coefficients inferred from these measurements were found to be 14 ± 5% larger than the corresponding broadening coefficients of the isotropic Q branch. Air broadening coefficients were found to be very similar and about 2.5% smaller than nitrogen self-broadening coefficients. Argon broadening coefficients were 25% smaller at rotational quantum number 7 and about 50% smaller at rotational quantum number 21. Additionally, we found that our unmixed line model for the O and S branches gave good fits for all pressures, which indicates that line mixing effects are not significant in the O and S branches over this range of pressures. Using indirect experimental evidence we infer that line mixing effects in the anisotropic component of the Q branch were below the threshold set by our experimental spectral resolution at pressures up to 70 atm at room temperature. Assuming that the anisotropic Q branch lines mix like the isotropic lines results in a small systematic error in the inferred temperature at flame temperatures, with the error increasing slowly with pressure. The bias can be easily removed by modeling the anisotropic spectrum separately from the isotropic spectrum. Line mixing effects should be included in the model of the isotropic part of the Raman spectrum, but can probably be neglected in the anisotropic part of the spontaneous Raman spectrum of N 2 for pressures below 400 atm at flame temperatures.