Born Together: A New Look at Binary Stars

Peering into the universe’s dark, hazy cosmos to catch the first glimpses of infant stars has long challenged astronomers. The cloud of dense gas and dust that gives rise to these stellar embryos also conceals them. However, a recent study published by Aleksey Generozov, a research fellow at the Oden Institute for Computational Engineering and Sciences at The University of Texas at Austin under Professor Stella Offner, has pierced through the clouds and illuminated the birth of stars, suggesting that stars begin their lives in small families, bound in a gravitational embrace from the start. 

He discovered these results through computer simulations built by the STARFORGE Project, a multi-institutional effort involving Generozov and Offner that seeks to model and visualize star formation. The movies produced by the STARFORGE team are mesmerizing, in which a spherical gas cloud collapses under its own gravity while stars pop up throughout it. However, beyond their visual appeal, the simulations are scientifically unprecedented. Offner comments, “The STARFORGE simulations are an excellent sandbox to test binary formation theories, since they are currently the only models that follow the formation of thousands of stars and include all key physics—turbulence, magnetic fields, radiation, and stellar feedback." 

In the Nature Astronomy study, “Low Mass Stellar Binaries are Bound from Birth,” Generozov mined a suite of these STARFORGE models to observe binary stellar systems. He found that for the vast majority of small stars in binary systems, the stars are already bound before the second star flickers on. He remarks, “This means that binaries are really evolving and accreting together,” reshaping how astronomers previously understood these multiple-star systems. 

Prior to these results, Generozov believed the dominant mechanism for binary formation was through capture: two fully formed stars coming together later in life. In 2023, Generozov published a paper supporting this capture theory, proposing “in a star cluster, capture should be the key physics that happens.” But within just two years, he would overturn his own conclusion that capture was the main method for binary formation.

The turning point came as he analyzed multiple STARFORGE simulations of star-producing clouds covering the span of 10-12 million years, incorporating millions of gas particles and complex physics that feeds back on itself. Because these infant stars are veiled in the gas and dust that births them, astronomers struggle to resolve images from telescopes. On the other hand, the sheer number of variables necessary to model stellar formation makes it incredibly difficult to study with computer simulations. Once an astronomer does have a working model, then they have to hop over another giant hurdle: accounting for the effect of gas. Generozov claims this was the hardest part, since in early stages, “stars are relatively small and surrounded by larger gas structures that dominate the local gravitational potential.” 

These domineering structures, known as gas halos, cradle young stars, and when scientists ignore them, most stellar pairs appear unbound. But when gas halos are included, the number of binary stars dramatically increases. This is illustrated in the figure below, which is a snapshot of the first moment two stars awaken. Each circle represents a young star lodged in a dense gas cloud, falling towards each other (as indicated by the black arrows in the figure below). Generozov identified this pair as a binary—only because he accounted for the gas halo. Without this step, he would have overlooked this, and many other pairs, as a binary system. 

Consequently, the new analysis of gas halos, combined with the physical realism and large statistical samples of STARFORGE enabled the team to address how binary stars formed in one of the first studies of its kind. And these scientific advances may be important for uncovering longstanding mysteries in astronomy. 

Offner elaborates,  "Our nearest stellar neighbor, Alpha Centauri, is actually a system containing three bound stars, but we don't have any information about how it formed. By looking at simulations, we can explore how star systems like Alpha Centauri start their lives."

In addition to investigating the origin of binary stellar systems, Generozov tracked how the binaries evolve after they form. “Approximately 40% of all single stars were at one time part of a multiple stellar system,” he stated. Binary systems are remarkably ubiquitous: even solitary stars may have once had a companion or two. And that early partnership could have lingering effects long after they have broken up. “For example, the presence of a binary companion will truncate proto-planetary disks; even stars that are single could experience this,” added Generozov. 

The implications of their study are far-reaching: “Binaries evolve and accrete together, which will affect their planetary systems and chemical evolution,” Offner and Generozov write. If binary stars grow up together, rather than meeting later in life as adults, their gravitational influence on surrounding matter, such as planets, is present early on. “Most stars have planets,” Generozov notes, “and if you have a binary companion, that affects the stability of the outer planets.” In other words, the presence of a companion star from the start could shape how planetary systems develop, including how planets migrate or settle into stable orbits. If the stars formed independently, planetary formation could have followed a very different path, like planets forming farther out before migrating inward. 

Their conclusions are especially important given that most adult stars, unlike our Sun, are members of systems with two or more stars. Just like humans, it is better to have close friends.

Acknowledgements and additional notes

The authors acknowledge the Texas Advanced Computing Center (TACC) at The University of Texas at Austin for providing computational resources and support from NSF AAG 2407522.

* and a comprehensive treatment of stellar feedback, from protostellar jets, radiation, stellar winds, and core-collapse supernovae. Credit: Dávid Guszejnov, Mike Grudić, in collaboration with Planetarium Mannheim. This work for this particular image was also supported by NSF Career grant 1748571, NSF AAG 2107942, and a Cottrell Fellowships Award 27982 from the Research Corporation for Science Advancement. The simulations were run on TACC supercomputers, using allocations AST-190018 and AST21002.

** (as illustrated by the trajectories in the inset), and is bound from birth due to the gas. However, there is uncertainty in how to account for the gas, represented by the red and blue arrows (opaque for the lower estimate and translucent for the upper estimate). Credit: Aleksey Generozov.