CO-Ar collisions: ab initio model matches experimental spectra at a sub percent level over a wide pressure range

Pure rotational R(0) and R(1) lines of CO in Ar baths: experimental broadening, shifting and mixing parameters in a wide pressure range versus ab initio calculations

“…To reduce the number of variable parameters we neglected the speed-dependence of the line mixing parameter Y , because its impact was found to be much less than the experimental noise level for the R(0) line. 30 Additional theoretical calculations for both lines confirm this fact (Section 4). For fitting the experimental spectra, the general profile should be multiplied by the integrated intensity I of the line and complemented by an additional “pedestal” function F c ( ν ) responsible for the continuum absorption in pure CO or CO–Ar mixtures.…”

“…The value of ν diff = 0.926 MHz torr −1 for CO in Ar at 299 K was reported in the earlier work. 44 Significantly smaller values of the velocity changing collision rates, namely 0.301 MHz torr −1 for the R(0) line in Paper I 30 and 0.334 MHz torr −1 for the R(1) line in the present work, were calculated ab initio for 297.5 K as the real part of the complex Dicke parameter ν opt (see eqn (A.4) from Paper I for details). We used these ab initio values as fixed parameters of the qSDRP model at the first stage of the experimental data treatment within the line-by-line fitting procedure.…”

“…The speed-dependent shifting parameter Δ 2 is very small (as follows from our ab initio calculations, Fig. 3 from Paper I 30 ) for the lines under study and could not be reliably retrieved at the current sensitivity level of the spectrometer. Therefore, we set Δ 2 = 0 in the model and only Δ 0 was varied.…”

“…The number density was calculated using an ideal gas approximation. It was verified in Paper I 30 that accounting for non-ideality of gas leads to number density correction, which does not exceed 0.1% for the experimental conditions of the resonator spectrometer and thus can be safely neglected.…”

“…aCybulski potential, 36 see Serov et al 30 for details.bFrom Seleznev et al 27 cFrom Luo et al 52 dHITRAN stated uncertainties: less than 1%.eStatistical uncertainty of fit, see Section 4 for details.fUncertainty due to the intermolecular potential, see text for details.gHITRAN stated uncertainties: 2–5%.hUncertainty is roughly estimated as about 10%.iHITRAN stated uncertainties: 5–10%.jHITRAN stated uncertainties: more than 20%.…”