NLTEmol3D

Was the early Universe carbon-rich? Carbon abundances from 3D non-LTE modelling of molecules

Astronomy is entering a new era of large-scale spectroscopic facilities that will deliver millions of high-quality, wide-band stellar spectra across vast Galactic volumes. This dataset offers a unique opportunity to address scientific questions that depend on identifying rare objects – effectively finding “the needle in a haystack.”

In this DFG-ANR project, we aim to address a fundamental question: Was the early Universe carbon-rich? This question arises from the discovery of carbon-enhanced metal-poor (CEMP) stars In particular, CEMP stars without enhancements in heavy neutron-capture elements are thought to be direct descendants of the first generation of stars. However, their extreme levels of carbon enrichment remain puzzling and lack a robust astrophysical explanation. Current evidence suggests that the earliest phases of the Universe may have been rich in carbonaceous dust, a scenario that is strikingly consistent with recent JWST detections of carbon-enrichment and dust in high-redshift galaxies. If confirmed, this would imply efficient gas fragmentation and the formation of low-mass stars at redshifts z > 10, leading to a substantial population of first-generation stars in the Milky Way.

We aim to provide a decisive test of the CEMP evidence: are metal-poor stars indeed carbon enhanced, as observations suggest? All previous observational studies relied on highly simplified 1D LTE models for carbon diagnostics, but this critical assumption has never been verified from first principles. 

Together with partners at the University of Montpellier and Heidelberg, we will address this question through a highly inter-disciplinary program, combining methods of molecular quantum mechanics, machine learning, stellar models and spectroscopy, and models of Pop III star formation. In particular, we will focus on molecular quantum mechanics calculations for CH photodissociation as well as collisions of CH with atomic hydrogen (reactive and non-reactive process). Those ab initio molecular data will be used to develop a non-local thermodynamic equilibrium model for the CH molecule. The new molecular data will be used to develop a non-local thermodynamic equilibrium model for the CH molecule, which is the only tracer of Carbon in metal-poor stars. Using new observed spectra available to us through 4MOST (the project MPIA is heavily involved with), we will determine carbon abundances of over 50,000 very- and ultra-metal-poor candidates in the Galactic disc and bulge. These measurements will be contrasted with Population III star formation models, yielding new insights into the properties of the first stars in the Milky Way.

This project will establish molecular stellar diagnostics as a powerful new pathway to map stellar nucleosynthesis, trace the chemical enrichment of the Milky Way, and uncover the origin and evolution of the earliest stellar generations.