Why SiC6 excitation needed new collisional data
SiC6 was recently identified toward the carbon-rich circumstellar envelope IRC+10216 through rotational transitions with J=26–40, and the observations yielded a column density of roughly 8×10^11 cm^-2 and a rotational temperature of about 19 K. Turning those detections into physical conditions requires state-to-state collisional rate coefficients, because the level populations depend on how efficiently collisions redistribute rotational energy.
Before this work, no state-to-state rotationally inelastic rates for SiC6 with He or H2 had been reported. The paper fills that gap for SiC6–He and uses the resulting dataset to test whether the computationally cheaper coupled-states (CS) approximation is adequate at the high rotational levels that the observations actually probe.
How the SiC6–He potential and scattering calculations were built
The authors constructed a two-dimensional SiC6–He potential energy surface with the RCCSD(T)-F12 method inside the rigid-rotor approximation. Quantum scattering calculations were then run with the MOLSCAT package, using close-coupling (CC) for lower rotational levels and CS for higher levels.
Convergence checks kept the SiC6 collision cross-section error within 5%. The radial propagation range was RMIN = 2.0 Å to RMAX = 20.0 Å, and for collision energies below 50 cm^-1 the integration step was 30 steps per de Broglie wavelength. These numerical choices matter because they set the reliability of the cross sections that are later thermally averaged into rate coefficients.
Where CC and CS rate coefficients actually diverge
CC and CS calculations show similar energy dependence overall, but rate coefficients for some weak transitions differ by up to a factor of 2–3, especially in the low-energy resonance region. For the dominant rate coefficients, the two methods generally agree within a factor of two, consistent with earlier CC/CS comparisons for linear molecules such as HC3N, HC5N, and HC7N.
This distinction matters: the largest discrepancies are concentrated in weak transitions and low-energy resonances, not in the strong transitions that dominate level populations. That is why the authors adopt CS rate coefficients for the radiative transfer simulations rather than treating the CC/CS gap as disqualifying.
What the CS rate coefficients cover
Using the CS method, the authors obtained rate coefficients involving rotational levels up to j=40 over temperatures from 3 to 100 K. The high-j cross sections show distinct resonance structures at low collision energy; as energy increases these resonances vanish and the cross sections decrease smoothly, dropping to approximately 2 Å^2 at 500 cm^-1.
After thermal averaging, the rate coefficients for all transitions rise rapidly with temperature, peak between roughly 25 K and 30 K, and then decline slowly. Adjacent high-j transitions have very similar values, with the rate coefficient slightly lower for higher initial j at any given temperature. The dataset therefore covers exactly the rotational range needed for the observed J=26–40 transitions.
Why CS is enough for RADEX modelling
RADEX radiative transfer simulations for the 20→19 transition at kinetic temperatures of 5, 10, and 15 K produced excitation and brightness temperatures from CC and CS rate coefficients that almost completely overlap. Slight discrepancies appear only at low density, n(H2) < 10^3 cm^-3, and diminish further as column density increases.
In all cases the optical depth τ was much less than 1 at the column densities considered (8×10^10, 8×10^11, and 8×10^12 cm^-2), confirming optically thin lines and making TEX essentially independent of the SiC6 column density. The practical implication is that for relatively high rotational transitions, the CC/CS differences are smoothed out during RADEX calculations, so CS is adequate for non-LTE modelling under the conditions explored.
The 40→39 transition is subthermally excited
For the 40→39 transition, the excitation temperature approaches the cosmic microwave background temperature at low density and increases toward the kinetic temperature as density rises. At the typical low density of IRC+10216, n(H2) ~ 10^4 cm^-3, the 40→39 transition is subthermally excited.
This is the paper's most consequential physical result for interpretation: the observationally derived Trot of about 19 K should be read together with a non-LTE excitation model rather than directly identified with the kinetic temperature. The rotational temperature alone is not evidence for either LTE or non-LTE; it is a constraint to be tested.
What the SiC6–H2 scaling does not yet settle
Because H2 is the dominant collider in astrophysical molecular gas, the authors also estimated SiC6–H2 rate coefficients by reduced-mass scaling from the SiC6–He data. These rates are explicitly provisional, and the paper states that explicit SiC6–H2 scattering calculations will be required to quantify the associated uncertainty.
That limitation sits directly on the H2 numbers, not on the SiC6–He dataset. The SiC6–He rate coefficients up to j=40 and 100 K, together with the CS assessment, remain the paper's first quantitative state-to-state collisional description of SiC6 and the molecular data needed for future excitation and abundance studies.
The priority is high-J non-LTE modelling
The decision-relevant split in this paper is between the strong, high-j transitions that dominate the observed J=26–40 emission and the weak, low-energy-resonance transitions where CC and CS diverge by a factor of 2–3. For the former, CS rate coefficients reproduce CC excitation and brightness temperatures closely enough that the cheaper method is justified; for the latter, the discrepancy is real but contributes minimally to level populations.
The practical threshold to carry forward is density: at n(H2) ~ 10^4 cm^-3 the 40→39 transition is subthermal, so any abundance or kinetic-temperature inference from the IRC+10216 SiC6 lines must be made through a non-LTE model using these state-to-state rates rather than from the rotational temperature alone.
Sources used in this answer
Rotational excitation of SiC 6 by collisions with helium: cross sections, rate coefficients and non-LTE modeling
The paper provides the first state-to-state rotationally inelastic rate coefficients for SiC6–He collisions, covering j ≤ 40 and 3–100 K, and shows that the coupled-states approximation gives nearly identical RADEX excitation and brightness temperatures to close-coupling for the high-J transitions relevant to IRC+10216, where the 40→39 transition is subthermally excited at n(H2) ~ 10^4 cm^-3.