Stellar composition
Metal-poor stars
Our bodies, like most things on Earth, are composed largely of elements forged in stars. Despite this, we still have much to learn about the cosmic events that led to the elemental abundances we observe today. These events range from the steady nucleosynthesis within stars like the Sun to the dramatic explosions of core-collapse supernovae. The oldest, most metal-poor stars provide a window into the earliest chapters of this story.
Some 100-200 million years after its birth, the first generation of stars transformed the Universe. No primordial star made up entirely of the H, He and Li produced in the Big Bang has yet been discovered but several second generation stars with exceptionally low iron content, down to at least 1/10,000,000 of the solar value, have been found both in the Milky Way and in nearby dwarf galaxies. Most metal-poor stars in the Milky Way halo would have formed at redshifts z = 5 –10 while the recently discovered extremely metal-poor stars in the bulge may even have formed at z > 15 making them the oldest known objects in the Universe.
The remnant of a supernova is enriched in heavy elements created before and during the explosion. Image: the Crab Nebula, ESO
Researchers at our division collect and analyze high-resolution spectra of these rare metal-poor stars using the world's largest telescopes, such as VLT and Keck, to determine their chemical composition. The patterns of elemental abundances provide clues about the masses of the first stars and how heavy elements were distributed into the interstellar medium. Additionally, these stars help probe Big Bang nucleosynthesis by offering insights into the unresolved cosmological lithium problems.
Data releases from large-scale surveys have greatly enhanced our understanding of metal-poor stars. The Gaia-ESO Survey, a large-scale spectroscopic effort, targeted over 100,000 stars throughout the Milky Way, offering valuable insights into the stellar populations of the disk, bulge, and halo. This survey emphasized homogenizing stellar parameters and chemical abundances using sophisticated Bayesian Inference methods, particularly for the metal-poor end, though challenges persisted due to the limited reference set in this regime.
The Gaia Data Release 3 (Gaia DR3) further advanced this work by providing low-resolution spectrophotometry for approximately 220 million sources, enabling metallicity estimates across a wide range, including the very metal-poor domain. This extensive dataset allows researchers to identify and classify various stellar populations, including those in globular clusters, contributing to our broader understanding of stellar evolution and chemical enrichment in the early Universe.
Additional studies, such as the analysis of the globular cluster NGC 1851, reveal the complexities in the formation histories of these ancient stellar populations. For instance, NGC 1851 has been shown to host multiple stellar populations with slight differences in metallicity and significant variations in certain elemental abundances. These findings suggest that the cluster may have originated from a dwarf spheroidal galaxy that was later accreted by the Milky Way, highlighting the intricate history of metal-poor star formation.
Determining the abundances in metal-poor stars remains particularly challenging. At low metallicity, the assumptions of 1D atmospheric models in hydrostatic and thermodynamic equilibrium are most prone to inaccuracies, leading to systematic errors in abundances that can be as large as an order of magnitude. With expertise in model atmospheres, radiative transfer, and atomic data calculations, astronomers in Uppsala are making significant contributions to both observational and theoretical advances in this important field of research.
Benchmark stars
Aldebaran, Arcturus, Pollux, Procyon, and Alpha Centauri have played a part in the culture and mythology of mankind since they were first identified and used to navigate the Earth. Astronomers used them as a reference point to describe the positions of the moon and planets as they moved through the night sky, appearing in the Babylonian Astronomical Diaries dating back to almost 1000 years BC. Today, these stars once again hold a key role as benchmark stars. Alongside other stars, they are used to classify the millions of stars observed in current and upcoming stellar surveys, serving as reference points for studying the chemical evolution of our Milky Way.
We have established a sample of benchmark stars spanning F, G, and K spectral types at varying metallicities, representing a significant portion of the Galaxy’s stellar populations. These spectral types are the most common in the Galaxy, accounting for more than 80% of all stars brighter than 20th magnitude, according to estimates from standard Galactic models. The benchmark stars are carefully selected with as much available data as possible to determine their effective temperature and surface gravity independently of spectroscopy. By using measured angular diameters, bolometric fluxes, parallaxes, and masses estimated from stellar evolution models, orbital dynamics, or asteroseismology, we calculate effective temperature from the Stefan–Boltzmann relation and surface gravity from Newton's law of gravitation. We also gathered high-quality, high-resolution spectra to build a spectral library. This library is used to determine the metal content and individual abundances of more than 20 chemical elements, employing and combining several analytical methods.
Effective temperature and surface gravity from fundamental relations for the 34 Gaia FGK Benchmark Stars. The semi-axes of the ellipses correspond to the uncertainties. Colour indicates metal content relative to the Sun on a logarithmic scale based on iron abundances measured from high-resolution spectra. Credit for the coloured spectra of Procyon, the Sun, and Arcturus: N.A. Sharp, NOAO/AURA/NSF.
The atmospheric parameters derived from these benchmark stars provide a standard for testing and harmonizing large ground- and space-based stellar surveys, such as the ESA mission Gaia, the European Gaia-ESO Public Spectroscopic Survey, and the Australian Large Observing Program GALAH (GALactic Archaeology with HERMES). Using the spectral library and additional tools for adapting the spectra to the requirements of different surveys, we aim to establish a connection between these major projects. This work is conducted in collaboration with several other research groups, primarily in Santiago (Chile) and Bordeaux (France).
Gaia DR3 has expanded the utility of these benchmark stars by offering uniformly derived stellar astrophysical parameters for hundreds of millions of stars throughout the Milky Way. The derived stellar parameters include detailed chemical abundances, atmospheric properties, and evolutionary traits, which all reinforces the importance of these benchmark stars in validating and refining our understanding of stellar characteristics across the Galaxy.
The initial sample of benchmark stars is being expanded to cover more of the parameter space with additional stars. We are involved in several observing programs using large optical/infrared interferometers to measure stellar diameters, including the ESO Very Large Telescope Interferometer on Cerro Paranal in Chile and the CHARA Array on Mount Wilson in California. Future expansions of the benchmark star sample are essential to close gaps in parameter space, ensuring that these benchmarks continue to provide a solid foundation for understanding the chemical and physical properties of stars across the Milky Way.