Checking It Twice: Using [C/N]-Masses and Asteroseismic Masses as a Diagnostic of Mass Loss and Transfer on the RGB

TL;DR

The paper develops an empirical framework to diagnose RGB mass loss and possible mass-transfer events by pairing birth-mass indicators from [C/N] with current masses from asteroseismology in field stars. By binning stars in [C/N]-[Fe/H] (and separate high-alpha samples by [Fe/H]) and comparing RGB versus RC masses, the authors quantify net RGB mass loss and show it decreases with metallicity and birth mass, contradicting constant-η Reimers' prescriptions. They introduce a metallicity- and mass-dependent calibration for Reimers' η to better match observations and identify 207 outlier stars likely formed through mass transfer, a population that may influence Galactic archaeology studies. The work highlights the value of combining spectroscopy and asteroseismology for disentangling stellar evolution effects in the field and outlines paths for future expansion with TESS and complementary spectroscopic gravities.

Abstract

The surface [C/N] of red giants is correlated with birth mass, but not directly impacted by mass loss. Exploiting this, we compare asteroseismic masses of red giants with the same [C/N] and but different evolutionary states. We find bulk differences between stars at the beginning of the red giant branch and in the subsequent evolutionary phase, the red clump, providing a direct constraint on the strength of net RGB mass loss in field stars. We find that net mass loss decreases with metallicity and mass, matching recent studies for field giants, but contradicting expectations from the widely used Reimers' mass loss formula. We propose a mass- and metallicity-dependent Reimers' $η$ calibration that reproduces the empirical trends that we see. In addition, we identify 207 stars (3.33% of our sample) that are clear outliers from their population in these birth mass bins, which we believe are likely candidates for mass transfer events. These stars do not show any obvious discrepancies in abundances or binary properties from their counterparts. This population should be accounted for in Galactic archaeological studies. Further follow-up is required to quantify their occurrence rate and origin.

Figures (7)

• Figure 1: The RGB and RC stars in our sample, with lines showing the grid of [C/N] and [Fe/H] windows used in our analysis. The bins overlap, so a single window of 0.05 dex is a 3x3 grid of the shown squares. The green, purple, and orange regions show the windows used in the three panels of Figure \ref{['fig:examplewindows']}.
• Figure 2: Mass versus log g in [Fe/H]=0 windows at [C/N]$=-0.1,-0.2$ and $-0.4$. Blue points are LRGB, and red points are RC. The shaded bar at the top of each panel indicates the color of the shaded region in Figure \ref{['fig:Windowbins']} that is being plotted in that panel. The log g dependence of mass on RC is expected because stars have similar radii on the red clump at different masses.
• Figure 3: Net mass loss versus [C/N] and [Fe/H]. In the low [C/N] - high [Fe/H] corner, the trend continues showing negative mass loss due to the earlier stated effects, so we adopt a minimum value of 0 for the color scale.
• Figure 4: RGB (blue), RC (red) masses and mass differences (purple) versus [Fe/H] for the high alpha stars (left-most panels) and various [C/N] values. At lower [C/N] values, the uncertainty increases as the range in masses within the window grows. The trend from Li_2025 is shown in green.
• Figure 5: Same as Figure \ref{['fig:panelcn']}, but now using windows of fixed RGB mass. MIST mass loss values for the given mass are additionally plotted in the bottom panels with both the standard $\eta$ (solid line) and scaled by the variable $\eta$ in equation \ref{['eq:eta']} (dashed line). The primary change of the new $\eta$ is a metallicity-dependence as seen in the field stars.