On the formation and abundance of CO in AGB envelopes

TL;DR

This paper uses kinetic modeling to quantify the residual gaseous carbon in carbon-rich AGB star envelopes, revealing it is much higher than traditional estimates and depends on reaction rates, impacting cosmic carbon grain formation.

Contribution

It introduces a kinetic approach to accurately determine residual gaseous carbon in AGB envelopes, challenging the conventional rule of thumb and emphasizing the importance of reaction rates.

Findings

Residual gaseous carbon ranges from 55 to 144 atoms per million H atoms.

The formation rate of CH significantly influences CO formation and residual carbon.

Higher reaction rates of C+H→CH are necessary to match observed CO abundances.

Abstract

It is generally considered, as a rule of thumb, that carbon monoxide forms very early in envelopes of AGB stars, and that it consumes most of the carbon, or most of the oxygen, depending on whether the photosphere is oxygen-rich or carbon-rich, respectively. \rm This work focuses on the latter case, with the purpose of quantifying the remaining fraction of gaseous carbon which is then available for forming carbonaceous grains. Since AGB stars are (probably) the main providers of cosmic carbon grains, this residual fraction is essential in establishing the validity of current grain models. Here, we use a kinetic treatment to follow the chemical evolution of circumstellar shells towards steady state. It is shown that the residual fraction depends essentially on the atomic ratio of pristine gaseous carbon and oxygen, and on the cross-section for CH formation by collision of C and H atoms. It lies between 55 and 144 C atoms per million H atoms, depending on the values adopted for unknown reaction rates and cosmic C abundance. This is much larger than predicted by the rule of thumb recalled above. The present results depend strongly on the rate of the reaction C+H-->CH: far from thermodynamic equilibrium (which is the case here), CO cannot be formed if this rate is as low as generally assumed. We have, therefore, estimated this rate by chemically modelling the reaction and found it indeed much higher, and high enough to yield CO abundances compatible with observations. An accurate experimental rate determination is highly desirable.