Hydrodynamics with gas–grain chemistry and radiative transfer: comparing dynamical and static models

Abstract

Context. We study the evolution of chemical-abundance gradients using dynamical and static models of starless cores. Aims. We aim to quantify if the chemical abundance gradients given by a dynamical model of core collapse, which includes time-dependent changes in density and temperature, differ greatly from abundances derived from static models where the density and temperature structures of the core are kept fixed as the chemistry evolves. Methods. We developed a new one-dimensional spherically symmetric hydrodynamics code that couples the hydrodynamics equations with a comprehensive time-dependent gas–grain chemical model, including deuterium and spin-state chemistry, and radiative transfer calculations to derive self-consistent time-dependent chemical-abundance gradients. We apply the code to model the collapse of a starless core up to the point when the infall flow becomes supersonic. Results. The abundances predicted by the dynamical and static models are almost identical at early times during the quiescent phase of core evolution. After the onset of core collapse, the results from the two models begin to diverge: at late times the static model generally underestimates abundances in the high-density regions near the core center, and overestimates them in the outer parts of the core. Deuterated species are clearly overproduced by the static model near the center of the model core. On the other hand, simulated lines of NH3 and N2H+ are brighter in the dynamical model because they originate in the central part of the core where the dynamical model predicts higher abundances than the static model. The reason for these differences is that the static model ignores the history of the density and temperature profiles which has a large impact on the abundances, and therefore on the molecular lines. Our results also indicate that the use of a very limited chemical network in hydrodynamical simulations may lead to an overestimate of the collapse timescale, and in some cases may prevent the collapse altogether. Limiting the set of molecular coolants has a similar effect. In our model, most of the line cooling near the center of the core is due to HCN, CO, and NO. Conclusions. Our results show that the use of a static physical model is not a reliable method of simulating chemical abundances in starless cores after the onset of gravitational collapse. The abundance differences between the dynamical and static models translate to large differences in line emission profiles, showing that the difference between the models is at the observable level. The adoption of complex chemistry and a comprehensive set of cooling molecules is necessary to model the collapse adequately.