The Hunt for Free Floating Planets in Our Galaxy

By Dante Minniti, Universidad Andrés Bello

Extrasolar planets is the name given to all the planetary-mass bodies that orbit stars other than the Sun. So far, many extrasolar planets have been discovered, among them are stars that harbour multiple planets. The confirmed record so far is the Sun with 8 well known planets, but we suspect that there may be many more planetary systems like ours in the Milky Way, and there is evidence that the star HD10180 has 9 planets orbiting it.

But even if the planets form around a star, they can be expelled from their parent system, and so there could be isolated planets out there, wandering the vacuum unattached to a parent star. We call these bodies, free floating planets, the worlds that roam free through our Galaxy.

Why do we suspect that there should be free floating planets out there?  Because of the way that planetary systems likely form and evolve. Many protoplanets form in a protoplanetary disk and then they go through a phase of collisions and rearrangement until the final planetary system is stable. While some coalesce to form giant planets and others fall into the star and are destroyed, we expect that many are expelled from the system completely. In fact, the number of free floating planets in our Galaxy might surpass the number of planets associated with stars. We simply do not know…

It is easier to form planets in the outer parts of a protoplanetary disk, where the objects are more loosely attached to the star and prone to be freed from the gravitational pull of the mother star by close encounters with other planets in the forming disk, or even by close passages with neighbouring stars (not collisions, which should be very rare among stars).

ffloatingearthArtists impression of an Earth-sized free-floating planet roaming through our galaxy.  Credit: J. B. Pullen.

It is also important scientifically to know how many of these objects are out there, in order to test our ideas (theories) of planetary formation. Is it the same number of free floating planets as the number contained in planetary systems? Or are they very rare, say 10 times less? Or are they much more numerous, 10 times more numerous for instance? And also, no less interesting, are small rocky free floating planets more numerous than giant gaseous free floating planets?  The last possibility is sobering, as some theories of Solar System formation predict that many more planets were formed in the early Solar System, the majority of which were expelled through collisions.

We know that there should be free floating planets because a few are being detected in young nearby associations. Very young, recently formed planets, emit light just as stars do for a brief period of time (astronomically speaking a few million years), before they cool and become too faint to detect from our vantage point with current technology. These bodies are very young giant planets, and although faint, they can be detected in the infrared before they are older and their surfaces cool down.

But we think that there should be much more free floating planets out there waiting to be discovered. The main problem to find this putative free floating planetary population is that older planets like Earth are small and dark, they do not emit light for themselves. They are so faint that is is impossible to detect them using images in the optical and near-infrared taken with the largest available telescopes (VLTs, Kecks, Geminis, etc), or even with future telescopes (JWST, GMT, TMT, EELT). Some of these planets may emit in radio wavelengths, but again they would be so faint that they are beyond detection even with our most powerful radio telescopes like Arecibo or ALMA.

We are stuck, we cannot detect these faint free floating planets directly. However it is important to know if there are free floating planets out there, and how numerous they are.

But don´t give up hope yet, astronomers sometimes can find clever solutions to seemingly impossible problems. There is a technique that is indirect, because it cannot image the planets, but can detect the gravitational effect cause by their mass. The father of this idea was Einstein, who else. He predicted that the light from a distant object would be bent when it passes close to a massive object. This is because the massive objects produce a deformation of space and time.  This is called the gravitational microlensing technique.  It measures the brightening of a distant source due to the light bending caused by a massive lens that passes in front of it.
The microlensing measurement simply consists of detecting the change of brightness of a distant object (called the source) by the near-perfect alignment of a massive object (called the lens) along the line of sight of the observer (us, with our telescopes).

vvv_microlensing1An example of a microlensing event that was witnessed by the VVV survey, where the star’s brightness is shown with time.  The event is seen towards the right end of the plot where the quick rise and fall in brightness of the observed background star is due to another object passing by between the star and the observer on Earth.

However, it is not that simple. The timescales of microlensing due to free floating planets should be very short, of the order of hours, as opposed to weeks-months for typical stellar mass objects. This timescale is the critical measurement, but if we are able to measure these short-lasting microlensing brightenings, we could detect planets like Earth, or less massive like Mars or even smaller. This is a very difficult measurement because it essentially requires continuous imaging of tens of millions of stars. You can imagine that these microlensing observations create gigantic databases and technical challenges.

A few groups have been pioneering in this field, discovering thousands of microlensing events so far (OGLE, MOA, MACHO, etc.). However, the short timescale microlensing signals due to our putative free floating planets are lacking. Is it because they do not exist, of because the current experiments are not designed to detect such short timescales? We believe the latter. Even though the evidence for free floating planets is slim at present, we believe that they should be very numerous, and we need to refine out hunting tools to discover them.

Very soon, the K2 mission gives us hope to discover these free floating planets towards the central regions of our Galaxy, the bulge. This extended phase of the Kepler space telescope will observe fields towards the Galactic bulge, where the number of stars is high, with a very frequent and continuous cadence, in order to be able to detect microlensing due to free floating planets.

On the other hand, we have been running a survey of the Milky Way´s bulge over the past few years, called the VVV survey (, using the VISTA near-infrared telescope located at ESO Paranal Observatory. It would be wonderful to do simultaneous near-infrared imaging of the K2 bulge fields in order to characterise the sources. The combination of optical and infrared data is very powerful to measure stellar properties. Knowing the stars parameters (luminosities, distances, reddenings, etc.) would allow us to characterise the rare microlensing event due to a free floating planet, and ultimately to measure the mass of the planet. If many of them are discovered, we could potentially refine the statistics in order to know if the free floating Earths are more numerous than the giant planets, among other things. The K2 bulge observations will be carried out in April-May this year, so stay tuned for the results on free floating planets!

vvv_galaxy2The region of our Milky Way galaxy that has been explored by the VVV survey, including coverage of the galactic center.

Finally, let´s go crazy and discuss life in free floating planets. This is pure speculation but I will contend that these objects are some of the most promising life vessels. As they travel across the Milky Way they can spread the seeds of life throughout the Galaxy. The main difficulty for a free floating planet to sustain life as we know it would be the absence of the main energy source: its parent star. However, when Earth-like planets form they are very hot, and then the igneous rock cools on the surface. The cooling process takes billions of years, depending on the original mass of the planet. For example, our Earth is 4.5 billion years old and it has not cooled down completely yet. Evidence of this are the volcanos throughout our planet´s crust, the plate tectonics, and the fact that seismology indicates that our planet nucleus is still igneous.

This volcanic activity potentially provides a continuous alternative source of energy for life in a free floating planet. This should not be surprising, as geothermal vents deep in the oceans sustain a variety of living organisms. In fact, if Earth would lose the Sun for some reason, the oceans would freeze but only on the surface, deep down they would remain liquid for many billions of years.

This is different for asteroids or minor planets, which can also be roaming through the Galaxy, but which are cold objects. Seeds essentially there have to survive the cold vacuum and hibernate for billions of years in order to be a viable source for spreading life, which is more difficult to sustain. Free floating planets constitute better living vessels, and they could be out there by the millions (or billions!).

For those who love to let the imagination run wild, consider the following speculation. A sufficiently advanced intelligent technological civilisation may chose to free its own planet from the parent star on purpose. This would allow them to cruise free through the Galaxy. In fact, in several billion years more our Sun will start to inflate in order to become a red giant star, engulfing the innermost planets of the Solar System in the process. In order to preserve our planet, we may want to liberate it from our mother star before it becomes a red giant. That is if we have the adequate technology in the future (the energy required to do this is enormous), and if we decide not to lose our Earth, where all Humanity´s initial history has developed. We could then decide to leave our mother Sun, becoming a free floating living Earth. It would be either this, or let Earth roast completely, because the Sun will inexorably become a red giant. To travel through our Milky Way in our own Earth is a dream future for an astronomer. And I don´t know about you, but I would like to preserve our heritage in this way…

Mail Attachment

About the Author. Dante Minniti is a Full Professor at the Universidad Andrés Bello in Chile, and Adjunct Scholar for the Vatican Observatory, Italy. He did his undergraduate studies in Astronomy at the Universidad Nacional de Córdoba (Argentina), and obtained his PhD in 1993 at the University of Arizona (USA).  He was a Postdoctoral Fellow of the European Southern Observatory (Germany) in 1993-1996, and a Lawrence Livermore National Laboratory Postdoctoral Fellow (USA) in 1996-1998.  Dante has been a member of the MACHO Team since 1996, and leader of the VVV Science Team since 2006. He was awarded the John Simon Guggenheim Fellowship Prize in 2005, and the Scopus Prize in Physics and Astronomy in 2008. In 2012 he was appointed a Member of the National Academy of Sciences of Argentina.  His broad research interests are: Extrasolar Planets, Astrobiology, Globular Clusters, Stellar Populations, Stellar Evolution, Gravitational Microlensing, Galaxy Formation, and Galactic Structure.  He has authored more than 300 refereed publications, that have accumulated more than 12000 citations in the scientific literature to date, whilst also authoring the recent books “Mundos Lejanos”, “Vistas de la Galaxia”, and “Nuevos Mundos”.  He is the deputy-PI of the “Millennium Institute of Astrophysics (MAS)”.

3 thoughts on “The Hunt for Free Floating Planets in Our Galaxy”

  1. There are several suggestions on how you might have biospheres on nomadic planets. Of course, such planets could have internal “insulated” oceans, such as the one we think exists under the ice of Europa. Abbot and Switzer call these “Steppenwolf” planets (Ap J Letters, 735:L27, 2011), they could be powered by natural radioactivity, or by tidal effects (if the planet has a moon, or is a moon of a larger planet). Intriguingly, Stevenson* points out that a sufficiently large “super Earth” with a predominately hydrogen atmosphere could trap enough heat to support liquid water oceans on its surface. Because there should be nomadic planets closer to Earth than the nearest stars, I predict that the closest biospheres to Earth will actually reside in nomadic planets, not orbiting a star.**



  2. “A few groups have been pioneering in this field, discovering thousands of microlensing events so far (OGLE, MOA, MACHO, etc.). However, the short timescale microlensing signals due to our putative free floating planets are lacking.”

    To clarify, lacking in the surveys, but not entirely lacking!? I haven’t kept up with the results from this field beyond early frequency estimates which were very useful, But I seem to remember early estimates from lucky finds of ~ 1 nomad/star, and Asteroid Initiatives last reference confirms both microlensing detection and at least the same frequency (of Jupiter massed nomads).