The Galaxy's Chemical Clock: How Yttrium and Magnesium Reveal Stellar Ages
Based on a sample of 736 field stars (new atmospheric parameters and Mg and Y abundances, accounting for non-LTE effects, for 528; asteroseismic ages for 307; isochrone ages for 221), spatial variations of the [Y/Mg]–age relation in the Galactic disk are explored. Two new triple and nine double spectroscopic systems are identified. Systematic differences in the [Y/Mg]–age dependence are found, reflecting an inhomogeneous history of star formation and enrichment. Overall, [Y/Mg] rises with metallicity, but at supersolar values the trend weakens and becomes flatter than for solar-metallicity stars, where young ones show lower [Y/Mg] and old ones elevated [Y/Mg].
Context
Age is the most important and most elusive parameter of a star. Without it, we cannot reconstruct the history of the Galaxy. Cecilia Payne-Gaposchkin showed that stars are mostly hydrogen and helium, and Fred Hoyle and Margaret Burbidge explained how nuclear reactions inside them forge elements. The abundance ratios of certain elements, such as yttrium and magnesium, change predictably due to their different sources and injection timescales into the interstellar medium. This led to the idea of 'chemical clocks.' However, previous studies have shown that these clocks tick differently in various parts of the Galaxy. This work expands the sample and, for the first time, extensively uses precise asteroseismic ages for calibration, while correcting systematics arising from departures from thermodynamic equilibrium (non-LTE).
Methods
For 528 field stars from the TESS mission, spectra were obtained using high-resolution spectroscopy with the VUES spectrograph (resolution ~68,000) at the Molėtai Observatory. Spectral synthesis was used to measure Mg and Y abundances with non-LTE corrections. Ages were determined in two ways: for 307 stars—asteroseismically (by analyzing solar-like oscillations: measuring the frequency of maximum power and the large separation), and for the rest—via the isochrone method (comparison with evolutionary tracks). Asteroseismic ages were verified with the SPInS code and showed good agreement. Classification into thin and thick disk was done rigorously: based on kinematics (U, V, W velocities using the Doppler effect), membership probabilities, and alpha-element enrichment (Mg). All parameters are compiled in a final catalog.
Results
The [Y/Mg]-age relation shows clear spatial variations. In the outer disk region (R > 8.5 kpc), the slope is steepest: log Age = –7.092 [Y/Mg] + 9.206 with a Pearson correlation coefficient of –0.60. Toward the Sun (7.5–8.5 kpc), the slope decreases, and in the inner disk (R < 7.5 kpc) the trend is nearly flat (PCC = –0.22). The thick disk shows almost no correlation (PCC = –0.20), indicating rapid formation before significant AGB star contribution. Splitting by metallicity reveals that stars with [Fe/H] > –0.2 have, on average, higher [Y/Mg] across all ages, but for stars with [Fe/H] > 0.1, the trend flattens: at young ages, [Y/Mg] is lower than for solar-metallicity stars. This agrees with theoretical expectations: at high metallicity, there are fewer neutrons per iron seed, and the s-process produces more light elements, including yttrium, which saturates the clock.
Implications
The results mean that Galactic chemical clocks are not universal: their calibration must account for a star's birthplace. The transition from a flat relation in the thick disk to a steep one in the outer thin disk reflects a radial gradient in star-formation efficiency. The inner regions formed quickly and became saturated with elements, while the outer regions evolved more slowly, allowing the s-process to accumulate yttrium. This is direct observational evidence that Galactic evolution proceeded differently at different radii. For chemical evolution theory, this requires a revision of models that assume homogeneous enrichment.
Future development
In the future, machine learning methods will enable multidimensional calibrations that simultaneously account for age, metallicity, position, and kinematics. Similar studies are planned for other pairs (Ba/Mg, Ce/Mg), although for barium, the 'barium puzzle'—the difficulty of analyzing its lines—poses a challenge. An important step will be integration with Gaia data and future missions like PLATO, which will provide asteroseismic ages for thousands of stars.
Impact
The results will impact stellar astrophysics, Galactic archaeology, age determination methods, and chemical evolution models.
Next steps
The next step is to apply the method to open clusters with precisely known ages and to red clump stars to refine the spatial trends. It is also necessary to include more stars from the inner disk to confirm the flattening at super-solar metallicities.
Key open problems
The research connects to open questions: the origin of s- and r-process elements, the formation of galactic disk structure, and the age of the Universe problem (local clocks must agree with cosmological ones). It also touches on the issue of convective mixing in AGB stars and nucleosynthesis beyond iron.
🎯 Yttrium, used as a chemical clock, is named after the Swedish village of Ytterby, where a mineral containing this element was found. Magnesium—one of the most common metals in Earth's crust—is born mostly in Type II supernovae. So the clock ticks thanks to explosions and slow burning in stars.
Key numbers
- total sample size: 736 stars
- stars with asteroseismic ages: 307
- spectral resolution: 68,000
- typical [Fe/H] error: ±0.09 dex
- minimum sample age: 0.17 Gyr