Exoplanet Observations Sharpen Picture of Planetary Formation
As planetary systems go, ours is odd. The Sun is more massive than all but a tiny fraction of the thousands of stars known to host exoplanets. And the planets in our system have a wide range of sizes, whereas the most common planetary system consists of about four similar-sized planets packed in tight orbits. A grand unified theory of planetary formation—one that accounts for the most common systems, the least common, and the ones in between—remains elusive. But two recent exoplanet investigations, the first involving a free-floating planet and the second concerning an uncommonly young host star, may spur progress toward such a theory.
Despite the lack of a comprehensive theory of planetary formation, astronomers have a good scenario of the process. A molecular cloud collapses to form a protostar surrounded by a disk. Planets assemble from disk material, and the type of planet—rocky, gaseous, or icy—is dictated largely by the distance from the star. Interactions with unincorporated disk material engender circular obits and inward migration. Once the disk has dissipated, the system becomes dynamically unstable, meaning gravitational interactions can amplify orbital perturbations (see Viewpoint: Tackling the Puzzle of Our Solar System’s Stability). Close-in planets may suffer collisions and migrate farther inward but remain bound. Outer planets, which tend to be gas giants, may suffer ejections.
Astronomers have clues supporting this planet-formation scenario. Specifically, there is indirect evidence of gas-giant ejection coming from observations of planets with large orbital eccentricities. More direct evidence would come from the detection of ejected or so-called free-floating planets (FFPs). These faint objects have so far avoided detection. However, a group of astronomers may have lucked out with the chance alignment between one of these planets and a background star.
Subo Dong of Peking University and his collaborators used data from ground- and space-based telescopes to observe an object gravitationally lensing a red giant as it passed between the star and Earth [1]. “Microlensing” events like this—in which a star is momentarily brightened by a massive object crossing the line of sight—were first observed by astronomers in the 1990s. Since then, about 20,000 events have been cataloged, almost all of them caused by small-mass stars called red dwarfs
But a dozen or so microlensing events have parameters—corresponding to the lens mass and distance—that place them in the tail of the lensing distribution, below the so-called Einstein desert. These microlenses are presumed to be FFPs, but other objects could mimic an FFP signal. For example, failed stars, such as brown dwarfs, have a small mass (below 80 Jupiters), which precludes them from burning hydrogen and helium like normal stars. These objects are, however, massive enough (more than about 3 Jupiters) that they can form by themselves through gravitational collapse of a cloud—making them distinct from planets that, by definition, form around a host star.
Distinguishing a failed-star lens from an FFP lens is difficult, as there is a degeneracy in microlensing measurements. A massive, distant object can produce the same amount of lensing as a less-massive, nearby object. Dong and his collaborators broke the degeneracy through good fortune. On May 3, 2024, a microlensing event was observed by three separate ground-based telescopes in the Southern Hemisphere. The event’s short duration of two days pointed to an FFP. Fortunately, the Gaia spacecraft was also monitoring the same part of the sky from a vantage 1.6 million km from Earth and was able to record the lensing event two hours later. Combining the observations, Dong and his collaborators disentangled the mass of the microlens from its distance from Earth.
The estimated mass of the microlens, in the sub-Jupiter range, is well below the threshold for gravitational collapse, so the object must have formed as a planet around a star. The planet presumably was booted from its birth system and now roams space alone as an FFP. The only plausible alternative to an FFP is a star-bound planet whose orbit is so wide that its parent star neither participated in the lensing nor appeared within the telescopes’ fields of view. (To an alien astronomer, Neptune and Uranus might look like FFPs.) A new observing campaign that begins in March could settle the question.
The dynamical instabilities that cause outer planets to leave or acquire eccentric orbits also shape the fates of inner planets. In the second investigation, John Livingston of the National Astronomical Observatory of Japan and his collaborators explored a set of inner planets in an extremely young system [2]. The host star, V1298 Tau, has a mass that is about the same as the Sun’s, which makes it rare among nearby stars. But its complement of four close-in, sub-Neptune-mass planets makes it commonplace among exoplanet systems. What makes V1298 Tau especially interesting to astronomers, however, is its relatively young age of 10–30 million years old, offering a rare view of the most common planetary configuration in its youth.
The V1298 Tau system was discovered by the Kepler spacecraft, which detected 2600 planets during its ten-year mission through the transiting method. In this method, a large collection of stars is monitored, looking for dips in brightness caused by passing planets. The planets around V1298 Tau were spotted at the end of the Kepler mission, after the failure of two reaction wheels forced the telescope to hunt for planets in the ecliptic plane rather than its original field near the constellation Cygnus. Using the V1298 Tau data, astronomers inferred the presence of four close-in planets of large radii, but they could not determine the planets’ masses nor the orbital period of the outermost planet [3].
Livingston and his collaborators filled in the gaps. With nine years of space- and ground-based observations, the researchers recorded small orbital changes, so-called transit-timing variations (TTVs), caused by gravitational interactions among the system’s planets. Their analysis of the TTVs revealed that V1298 Tau’s planets have masses of 5, 6, 13, and 15 Earth masses in orbits of 8, 12, 24, and 49 days, respectively. The planets’ radii are unusually large at 5, 7, 9, and 10 Earth radii. The mean densities of V1298 Tau’s planets are in the range 0.08–0.20 g/cm3, far below that of Earth (5.5 g/cm3) and of Neptune (1.6 g/cm3). The implication is that V1298 Tau’s planets are significantly puffed up.
That finding fits in with a proposal made five years ago [4]. In the so-called fluffy planet scenario, young systems like V1298 Tau harbor young planets whose still-significant internal heat inflate their atmospheres. At the same time, the corresponding host star is “active,” meaning it has strong magnetic fields and rapid rotation, which causes it to emit intense radiation at x-ray and ultraviolet wavelengths. This radiation eventually boils the fluffy atmospheres away to leave the types of planets—super-Earths and sub-Neptunes—that are found in abundance around mature stars.
Astronomer Sean Raymond of the University of Bordeaux in France points out the important role of the protoplanetary disk in the evolution of planetary systems. The disk’s dissipation and the dynamical instability that ensues leads both to the inward migration of super-Earths and sub-Neptunes and to the expulsion of gas giants. “So, in my mind, the two new papers show puzzle pieces that nicely fit together into a bigger picture,” Raymond says.
–Charles Day
Charles Day is a Senior Editor for Physics Magazine.
References
- S. Dong et al., “A free-floating-planet microlensing event caused by a Saturn-mass object,” Science 391, 96 (2026).
- J. H. Livingston et al., “A young progenitor for the most common planetary systems in the Galaxy,” Nature 649, 310 (2026).
- T. J. David et al., “Four newborn planets transiting the young solar analog V1298 Tau,” Astrophys. J., Lett. 885, L12 (2019).
- J. E. Owen and Y. Wu, “Atmospheres of low-mass planets: The “boil-off”,” Astrophys. J. 817, 107 (2016).