Separated At Birth: Comparing Chemical Compositions of Binaries

Title: Identical or fraternal twins? : The chemical homogeneity of wide binaries from Gaia DR2

Authors: Keith Hawkins, Madeline Lucey, Yuan-Sen Ting, Alexander Ji, Dustin Katzberg, Megan Thompson, Kareem El-Badry, Joanna Teske, Tyler Nelson, Andreia Carrillo

First author’s institute: Department of Astronomy, The University of Texas at Austin, 2515 Speedway Boulevard, Austin, TX 78712, USA

Status: Accepted for publication in MNRAS [open access on arXiv]

To understand our place in the Milky Way, we need to understand its history. Answering questions about the dynamical evolution of our galaxy — how its different components evolved and interacted — is the driving force behind the field of Galactic Archaeology. Using information about stellar movements, ages and chemical compositions, scientists can construct a picture of how the layout of the Milky Way has changed over the course of history.

One of the most important techniques in Galactic Archaeology is chemical tagging. As the universe ages, stars will create heavier elements and disperse these throughout their interstellar neighbourhood. As time goes on, different regions of the universe will have different chemical abundances unique to that time period. The technique of chemical tagging, then, rests on the assumption that stars with similar chemical compositions were born together. If this holds true, chemical tagging could be used to identify stellar clusters that have since dispersed, or accreted material.

However, in order for this technique to work, it is important to test these assumptions. There are two core principles underlying the use of chemical tagging. Stars that form in groups are:

  • Chemically homogenous (they have the same composition of elements), and
  • Unique in those chemical compositions from other groups.

The best way to test these assumptions is through wide binaries. These stars are useful for this purpose because they formed together, but have not interacted during their lifetime, thereby changing any elemental abundances.

Todays authors identify 50 stars in 25 wide binary pairs to test these two principles. With precise distances from the Gaia mission, and spectroscopy from the McDonald Observatory, they were for the first time able to place strong constraints on whether chemical tagging can be applied to Milky Way stars.

Are wide binaries chemically homogenous?

In order to determine whether stars born in groups have the same chemical composition, todays authors performed thorough spectroscopic measurements on all 50 stars in their 25 wide binary pairs (see Figure 1). They measured 23 different elements, but focused primarily on the typically well constrained iron-to-hydrogen ratio, [Fe/H], expressed in logarithmic units of dex.

For 20 out of the 25 stars they studied, the difference in [Fe/H] within the binaries was within ~0.02 dex, small enough to call them homogenous. For the remaining 5 stars the difference was roughly 0.10 dex which, while still small, was enough to arouse suspicion. For ratios of other elements measured, any differences within the binaries were found to be within the measurement uncertainty of 0.08 dex.

Figure 1: A subsection of the observed spectra for 4 of the 25 binary systems. One component is shown as a dotted line, and the other with a solid line. In the lower three cases, the differences are extremely small, and the difference in their chemical abundances are low. For the top example, there are some more noticeable differences in the spectra, and the difference in their chemical abundances is higher.

Are binaries unique in chemical composition?

In order to check whether the agreement in chemical abundance within the binary pairs is distinctly different from other stars, the authors compared the differences in chemical abundance ratios between random pairings of the stars in their sample. The results of this can be seen in Figure 2.

If stars had similar chemical compositions regardless of their formation history, these random pairings would show similar differences in chemical abundance to the binary pairs. Instead, the data shows that the differences between binary stars are much smaller, making them far more homogenous than when compared to random field stars.

Figure 2: Violin diagrams showing the distribution of differences between binaries in their pairs (orange, left) and with random stars from the sample (green, right). The top shows the ratio between each element and hydrogen (H), and the bottom shows the ratio between each element and iron (Fe). As expected, the differences within the binary pairs are much smaller than when compared to random stars. The solid lines represent a 0.05 dex spread on both plots.

What’s next?

These results are incredibly encouraging for the use of chemical tagging in galactic archaeology, finding that in 80% of cases, binaries have near-identical chemical compositions. In future, the authors hope to increase their stellar sample, in order to see if the 20% of stars that are not homogenous are an over- or underestimate. Understanding the reasons behind why these stars do not share the same chemical compositions is the next big step in this field, and will fully unlock the potential of chemical tagging.

About Oliver Hall

Oliver just started a postdoc at ESA ESTEC in the Netherlands, after completing a PhD at the University of Birmingham, UK. His research focuses on asteroseismology, the study of stellar pulsations, and what it can tell us about stellar populations. When not doing research he enjoys playing piano, walking, and not moving from the sofa all weekend with a good book, show, or game.

Leave a Reply