Are Giant Stars Stealing Lithium from Their Neighbours?

Title: Are Lithium-Rich Giants Binaries?: A Radial Velocity Variability Analysis of 1,400 Giants

Authors: Matias Castro-Tapia, Claudia Aguilera-Gómez, Julio Chanamé

First Author’s Institution: Pontificia Universidad Católica de Chile, Chile

Status: Submitted to Astronomy and Astrophysics [open access]

Somehow, evolved giant stars that are no longer burning hydrogen in their cores have managed to get their hands on unexplained quantities of lithium, despite the theoretical expectation that they will become depleted of lithium during their lifetimes due to processes such as convection or mixing. Trying to understand where this mystery lithium came from is one of the open questions in stellar astrophysics, motivating the authors of today’s paper to investigate a possible mechanism that may be responsible for enriching evolved massive stars with lithium, namely mass transfer from a binary companion. If binary interaction is an important enrichment channel, we should see lithium-rich stars in binary systems at a higher rate than their non-enriched counterparts. 

To understand the rate at which lithium-rich stars interact with a companion, the authors of today’s paper assembled a sample of low-mass giant stars with well-constrained lithium abundances and searched for evidence of binary interaction. One way to do this is to look for variation in a star’s radial velocity, or its velocity along the line of sight. If a star and its companion are orbiting around their centre of mass, it is likely to experience periods where it is travelling towards the observer (negative radial velocity), away from the observer (positive radial velocity), or perpendicular to the observer (zero radial velocity). Therefore, if a star’s radial velocity is changing with time, it’s a pretty good sign that it is interacting with a companion. If the interacting stars tend to have higher levels of lithium than solitary stars, it’s possible that interaction may account for at least some of the unexplained lithium we are seeing. 

First, the authors assembled their sample consisting of all red giant stars with radial velocity and lithium measurements from the Gaia, RAVE, and GALAH surveys, amounting to 1418 stars in total. Then, they calculated values they refer to as wij and wt for each star in their sample to determine whether it is likely found in a binary. More specifically,  wij describes the variation between radial velocity measurements from two different surveys (i.e. Gaia, RAVE, and GALAH), and wt is the sum of the individual values of  wij,  with higher values of wij and wt indicating a star is more likely in a binary. From their calculated values, they gave a classification to each star, with classes 5a and 5b indicating a star is likely interacting with a binary companion, and classes 1 – 4 indicating that the star does not show strong evidence of interaction. They then looked at what fraction of lithium-rich stars fell into classes 5a and 5b compared to the lithium-normal population. 

Figure 1. The fraction of stars in each variability class for the lithium-normal (A(Li) ≤ 1.5) and lithium-rich populations (A(Li)>1.5), with classes 5a and 5b showing strong evidence of radial-velocity variability. The left panel includes all stars in their sample, while the right panel only includes red giant branch stars. Adapted from Figure 5 in the paper.


Their results are shown in Figure 1, which gives the fraction of all stars (left) and the subsample of red giant stars (right) which show signs of interaction in both their lithium-normal (A(Li) ≤ 1.5) and lithium-rich populations (A(Li) > 1.5). Given that classes 5a and 5b show evidence of interaction in the form of radial velocity variation, a larger fraction of lithium-rich stars classed as 5a or 5b would indicate they are more often found in binaries. Looking at Figure 1, it is clear that when the entire population is considered, there is no significant difference between the lithium-rich and lithium-normal samples in regard to the fraction of stars classed as 5a and 5b. However, when only the red giant branch stars are considered, the fraction of class 5a and 5b lithium-rich stars is greater than the fraction for lithium-normal stars, although the large uncertainty makes the difference between the two populations marginal.

Given that radial-velocity variability is not more common in lithium-rich stars compared to the lithium-normal population when the full sample is considered, the authors suggest that binary interaction is not the primary mechanism responsible for the observed lithium enrichment. However, their analysis does allow for the possibility that binary interaction may be a contributing factor for stars on the red giant branch. They suggest that these red giants may have a different origin to the other lithium-rich giant stars that do not show higher rates of variability, but more radial velocity measurements are required to confirm this. Unfortunately for now, it remains unclear as to where these stellar giants stole their lithium from.

Astrobite edited by Brandon Pries

Featured image credit: NASA

Author

  • Sonja Panjkov

    I’m a second-year PhD student at the University of Melbourne. My research focuses on the high-energy emission from the supernova remnants in the Magellanic Clouds. In my spare time, I enjoy hanging out with my cats and going to see live music.

    View all posts

Submit a Comment

Your email address will not be published. Required fields are marked *