What Are Brown Dwarfs? The Failed Stars Between Planets and Suns

Photo: Marco Milanesi / Pexels
A brown dwarf is an object that formed the way a star does but never gathered enough mass to burn ordinary hydrogen for the long haul. Astronomers define the class mainly by mass, roughly 13 to 80 times that of Jupiter, and by what happens inside: a brief spell of deuterium fusion, then a slow fade to lower and lower temperatures. Here is where those mass boundaries come from, how brown dwarfs sort into L, T, and Y types, what the 2024 split of Gliese 229B revealed, and why the line between a brown dwarf and a planet stays blurry.
What makes an object a brown dwarf?
Brown dwarfs are defined by mass, and the two edges of the definition are nuclear thresholds. At the low edge, near 13 Jupiter masses (written MJ), the core becomes warm enough to fuse deuterium, the heavy isotope of hydrogen whose nucleus holds one proton and one neutron. At the high edge, near 80 MJ (about 0.08 times the mass of the Sun), the core reaches the temperature needed to fuse ordinary hydrogen steadily, and the object is a low-mass star, an M dwarf. Everything between those two lines is a brown dwarf. (Wikipedia: Brown dwarf)
Think of a gas stove with the dial turned only part of the way. Below a certain setting there is no flame at all, and above it the flame stays lit indefinitely. Brown dwarfs sit in the awkward middle: enough heat to flicker for a while, not enough to keep burning. A planet is the stove that never ignites, and a red dwarf is the one that burns for trillions of years.
The edges are rules of thumb, not sharp physical cliffs. The hydrogen-burning limit shifts with an object's composition: theory has put it at 0.07 to 0.08 solar masses for stars with the Sun's makeup and higher, near 0.09, for stars poor in heavy elements. The 13 MJ deuterium line is similarly approximate, and it is often described as a convenient convention rather than a quantity with deep physical meaning.
| Object | Approximate mass | What happens in the core |
|---|---|---|
| Planet (IAU working definition) | Below about 13 MJ | No deuterium fusion |
| Lower-mass brown dwarf | About 13 to 63 MJ | Deuterium fuses for a time; lithium survives |
| Higher-mass brown dwarf | About 63 to 80 MJ | Deuterium and lithium both fuse; hydrogen fusion never becomes self-sustaining |
| Red dwarf star | Above about 80 MJ | Steady hydrogen fusion |
Source: mass thresholds as summarized in Wikipedia's Brown dwarf article; lithium-burning limit of about 63 to 65 MJ.
The lithium test
Fusion thresholds are hard to measure directly, so astronomers use a chemical shortcut. Lithium-7 is destroyed at a lower core temperature than hydrogen. A true star, which gets hotter inside, burns its lithium away. Most brown dwarfs never get hot enough to do so, and objects above roughly 65 MJ burn theirs within about half a billion years. A cool, faint object whose spectrum still shows lithium is therefore very likely substellar.
The test was decisive in the mid-1990s. Teide 1, a faint object in the Pleiades star cluster, was found in images from January 1994, and its lithium was confirmed in November 1995 using the Keck telescope. Its mass is estimated at 55 plus or minus 15 MJ. (Wikipedia: Brown dwarf)
The surprise: brown dwarfs are roughly the size of Jupiter whether they weigh 15 times as much or 75 times as much. Adding mass to a brown dwarf makes it denser, not bigger.
Why a brown dwarf cannot hold a hydrogen fire
A star stays up against its own gravity because its core is hot. Fusion releases energy, the hot gas pushes outward, and the push balances the pull. A brown dwarf contracts under gravity just as a forming star does, heating as it shrinks, but it hits a different wall before the core gets hot enough for hydrogen to burn steadily.
That wall is electron degeneracy pressure. The Pauli exclusion principle says that two electrons cannot occupy the same quantum state. Squeeze matter dense enough and the electrons run out of low-energy states, so they are forced into higher-energy ones, which creates an outward pressure that does not depend on temperature. Compare it with a crowded stadium: once every seat is taken, more people cannot fit however hard the crowd pushes, and the building does not need to be hot to resist. The same effect holds up white dwarfs, the dense remnants of Sun-like stars (White Dwarfs: The Fate of 97% of All Stars).
Once degeneracy pressure takes over, the brown dwarf stops shrinking and begins to cool. Its radius settles close to that of Jupiter: across the whole brown dwarf mass range, radii vary by only about 10 to 15 percent. That produces the counterintuitive pattern in the box above. A heavier brown dwarf is more tightly packed rather than larger, so a 70 MJ object has about the same width as one with a fifth of the mass.
What fuels the early glow
Early in life, a brown dwarf is hot from the energy released by contraction, and objects above the deuterium threshold get an additional, temporary boost from fusing the deuterium they inherited. Deuterium is scarce, only a few nuclei for every 100,000 hydrogen nuclei, so the fuel does not last. Once it is used up, the extra heat source disappears and the brown dwarf continues to dim for the rest of its existence. (Wikipedia: Brown dwarf)
Fusion in the Sun is described in How the Sun Works: Nuclear Fusion, and the broader life cycle that brown dwarfs fall outside is covered in How Stars Are Born and Die.
From prediction to discovery: a short history
The idea came before the evidence. Astrophysicist Shiv Kumar worked out in the early 1960s that objects too light to burn hydrogen should exist, and the first name for them was "black dwarfs." Jill Tarter proposed the name brown dwarf in 1975, in her PhD thesis at the University of California, Berkeley. For two decades the class had no confirmed members.
An early hint came in 1988 with GD 165B, a cool companion to a white dwarf that did not fit any known category. It is now the prototype of the L dwarfs. The decisive year was 1995. Teide 1 was confirmed in the Pleiades with the lithium test, and Gliese 229B was found next to a nearby red dwarf by a Caltech team at Palomar Observatory. Gliese 229B showed methane absorption, a signature of cool gas-giant atmospheres that stars do not have. Which of the two counts as "first" depends on whether you prioritize the lithium test or the methane signature. (Caltech)
After 1995 the numbers grew steadily. By 2013, catalogs contained more than 900 L dwarfs and 355 T dwarfs, and the total of identified brown dwarfs has since passed 1,800. The next class, the very cold Y dwarfs, was announced in 2011 using the Wide-field Infrared Survey Explorer (WISE) spacecraft. (Wikipedia: Brown dwarf; Cushing et al. 2011)
Gliese 229B turned out to be two objects
For nearly 30 years, Gliese 229B was the textbook example of a brown dwarf, and it also carried an embarrassing problem. Its measured mass, about 71 MJ from the way it tugs on its host star, was too high for how dim it was. A brown dwarf that heavy should be 2 to 6 times brighter than Gliese 229B actually is. Either the evolutionary models were wrong or the mass was misleading. (Xuan et al. 2024)
In October 2024, a team led by Jerry Xuan used two instruments on the European Southern Observatory's Very Large Telescope in Chile to resolve the puzzle. The GRAVITY interferometer, which combines light from several telescopes to see fine detail, split the object into two separate points. The CRIRES+ spectrograph independently saw the Doppler shifts of two bodies, one moving toward Earth while the other moved away, switching back and forth as they orbited. Each instrument alone resolves the pair. (Caltech)
| Property | Value (Xuan et al. 2024) |
|---|---|
| Component masses | 38.1 plus or minus 1.0 MJ and 34.4 plus or minus 1.5 MJ |
| Orbital period | 12.1 days |
| Semimajor axis | 0.042 astronomical units, about 16 times the Earth-Moon distance |
| Brightness ratio at 2 micrometers | 0.47 plus or minus 0.03 |
| Orbit around host M dwarf | About every 250 years (as a pair) |
Sources: Xuan et al., arXiv:2410.11953; Caltech news.
Two dimmer objects of about 38 and 34 MJ together outweigh and out-shine one of 71 MJ in exactly the way the observations required. As Xuan put it, "we were wrong all along about the nature of the object. It's not one but two." A follow-up orbit analysis in 2025 refined the masses slightly, to about 37.7 and 33.4 MJ. The case is a useful reminder that a mass measured from orbital motion describes everything that moves together, whether that is one body or several. (Caltech; Thompson et al. 2025)
Spectral classes: L, T, and Y
A brown dwarf starts out hot enough to resemble a very cool star, then moves down the spectral sequence as it cools. The sequence for substellar objects runs M, L, T, Y, and an object's class depends on both its mass and its age. A 15 MJ object and a 70 MJ object can share a temperature at different times in their lives. For that reason, no single temperature or luminosity can be assigned to each class without caveats. (Wikipedia: Stellar classification)
| Class | Approximate temperature | Defining spectral features | Notes |
|---|---|---|---|
| M (late) | 2,100 to 3,500 K | Metal oxides such as titanium oxide | Includes the lowest-mass stars |
| L | 1,300 to 2,100 K | Metal hydrides, strong alkali-metal lines | Most are brown dwarfs; some are stars; very dark red |
| T | About 600 to 1,300 K (some sources say 550 K) | Methane, water | Gliese 229B was the first found |
| Y | Below about 600 K | Ammonia likely; methane and water | Identified with WISE in 2011 |
Sources: Wikipedia: Brown dwarf, Wikipedia: Stellar classification, Cushing et al. 2011. Boundaries vary slightly between sources.
L dwarfs: cloudy and dark red
L dwarfs are cooler than M stars, are very dark red, and shine brightest in the infrared. Their atmospheres are cool enough that metal hydrides and alkali metals are prominent in the spectrum. Most L dwarfs are brown dwarfs, but the youngest and heaviest are low-mass stars, so a spectral type alone does not prove an object is substellar. That is one reason the lithium test still matters. Reading these features relies on the techniques in What Is Spectroscopy?.
T dwarfs: the methane class
T dwarfs are cooler still, around 600 to 1,300 K, and their emission peaks in the infrared. Methane is prominent, which is why Gliese 229B's spectrum looked so different from any star's in 1995: a gas that appears in the atmospheres of Jupiter and Saturn was dominating the light of an object outside the solar system. Between the L and T classes, the silicate and iron clouds that hang in L dwarf atmospheres dissipate, a change astronomers call the L/T transition.
Y dwarfs: close to room temperature
Y dwarfs are the coldest brown dwarfs and are best studied at mid-infrared wavelengths, where they emit most of their energy, which is why a space telescope such as WISE was needed to find them (How Space Telescopes See in Infrared). In 2011, a team led by Michael Cushing reported seven ultracool brown dwarfs from WISE data. Six were classified as Y dwarfs, with temperatures of roughly 300 to 500 K. One of them, WISEP J1828+2650, became the archetype of the class. Ammonia is the suspected defining feature, though it is hard to separate from water and methane absorption in the same wavelength range. (Cushing et al. 2011)
Weather on a failed star
Brown dwarf atmospheres are cool enough to hold clouds and warm enough to churn. The cloud types change with temperature. In L dwarfs, grains of silicates and iron condense into clouds. As an object cools into the T class, those clouds clear from view, and models by Caroline Morley and colleagues show that other condensates, such as sodium sulfide (Na2S), manganese sulfide, zinc sulfide, and potassium chloride, can form in the cooler T dwarf atmospheres and make their spectra look "cloudy" in a different way. (Morley et al. 2012)
The clouds are not static. Hubble and Spitzer observations of the brown dwarf 2MASS J22282889-4310262 have tracked shifting cloud patterns on its surface. A 2020 study measured wind speeds of about 650 meters per second, with an uncertainty of 310, on 2MASS J10475385+2124234. In 2024, the James Webb Space Telescope detected clouds of silicate grains, at temperatures of about 875 to 1,026 degrees Celsius, in the atmosphere of the Luhman 16 pair. Jupiter's storms are the nearest analogy, and The Great Red Spot: Science Behind Jupiter's 350-Year Storm shows how long a gas-giant weather system can last. (Wikipedia: Brown dwarf)
Rotation drives much of this. Most brown dwarfs spin quickly, and the fastest known, 2MASS J03480772-6022270, has a rotation period of just 1.080 hours. Rapid spin helps generate strong magnetic fields, and roughly 5 to 10 percent of brown dwarfs appear to produce radio emission. TVLM 513-46546 showed what may be the first brown dwarf aurora, and its equator rotates faster than its poles, much as differential rotation shapes the Sun and Jupiter. (Wikipedia: Brown dwarf)
The coldest known: WISE 0855-0714
The record holder for coolest brown dwarf is WISE J085510.83-071442.5, usually called WISE 0855. It was identified in April 2014 by Kevin Luhman in repeated WISE observations, and it stood out for its motion: 8.1 arcseconds per year across the sky, the third highest of any known star or brown dwarf. That fast motion is a sign of proximity, and a parallax measurement placed it about 2.2 parsecs, or 7.2 light-years, from the Sun. (Luhman 2014)
Luhman's comparison with evolutionary models gave an effective temperature of 225 to 260 K, which is about minus 48 to minus 13 degrees Celsius, the range of a cold winter on Earth, and a mass of 3 to 10 MJ for an assumed age of 1 to 10 billion years. Those figures come from comparing the object's brightness with evolutionary models, and an object this cold is hard to model, so later estimates may differ. (Luhman 2014)
At such low temperatures, models predict that water may condense into ice clouds in the atmosphere, though how far that has been confirmed by observation is still debated. At 3 to 10 MJ, WISE 0855 may sit below the deuterium-burning line. If so, it is a planetary-mass object by the usual mass cut, even though it may have formed like a star. That tension is the subject of a later section.
How brown dwarfs form
The leading idea is that most brown dwarfs form the same way stars do: a dense knot of gas in a molecular cloud collapses, and the collapse simply ends with too little mass to start steady hydrogen fusion. Many young brown dwarfs are surrounded by disks of gas and dust, as young stars are, which is consistent with a star-like origin. As of 2024, about 67 proto-brown-dwarf and 26 pre-brown-dwarf candidates were known, and the smallest brown dwarf found so far, Cha 110913-773444, has about 8 MJ. (Wikipedia: Brown dwarf)
Related reading on the earliest stages is in What Is a Protoplanetary Disk and How Do Planets Form? and in the discussion of the Orion Molecular Cloud, a nearby star factory.
The brown dwarf desert
One pattern is hard to explain: brown dwarfs are rare close to ordinary stars. Fewer than 1 percent of Sun-like stars have a brown dwarf companion within about 3 to 5 astronomical units. Astronomers call this gap the brown dwarf desert. It suggests that making a companion in the brown dwarf mass range close to a Sun-like star is difficult. Systems that do have such a companion, like Gliese 229, tend to have it far out. (Wikipedia: Brown dwarf)
Pairs and the ejection idea
Brown dwarfs also come in pairs. About 24 percent of L dwarfs and about 8 percent of late-T and early-Y dwarfs are binaries. The Y+Y binary WISE J0336-0143, confirmed with the James Webb Space Telescope in 2023, has a mass ratio of 0.62 plus or minus 0.05 and a separation of 0.97 astronomical units. An older proposal held that brown dwarfs are stellar embryos ejected from multiple-star systems before they could gain mass, but the wide binary 2M1101AB, whose members are more than 20 astronomical units apart, is hard to reconcile with ejection. A pair that wide would be hard to keep intact through an ejection. (Wikipedia: Brown dwarf)
Binary stars and their orbital mechanics are covered in What Is a Binary Star System?.
Where the line to planets blurs
The cleanest rule says an object under about 13 MJ is a planet and one above is a brown dwarf. In practice the definition is messier. The International Astronomical Union's 2018 working definition counts an object below 13 MJ, with a mass ratio to its host of about 1 to 25 or less, as a planet. The NASA Exoplanet Archive, meanwhile, includes objects up to 30 MJ. Two catalogs, two cutoffs. (Wikipedia: Brown dwarf)
The conflict is sharpest for free-floating objects with no host. A 5 MJ body in open space might have formed like a planet in a disk and been thrown out, or like a star from a collapsing cloud. Mass alone cannot tell the two apart, so many astronomers sort objects by formation history, which is rarely observable. Free-floating planetary-mass objects are the subject of What Are Rogue Planets?, and the wider planet catalog is explained in What Are Exoplanets and Could Any of Them Host Life?.
The JuMBO controversy
The blur became a public argument in 2023, when astronomers Samuel Pearson and Mark McCaughrean reported, from James Webb Space Telescope images of the Orion Nebula, hundreds of planetary-mass objects, including about 40 apparent pairs, nicknamed JuMBOs (Jupiter-mass binary objects). The reported masses ran from 0.7 to 13 MJ and the separations from tens to a few hundred astronomical units. Such wide pairs of such small objects were hard to explain with either planet or star formation. (Wikipedia: JuMBO)
Later work has cut into the claim. Luhman reanalyzed the data in 2024 and found most JuMBOs were not in his substellar sample, with JuMBO 24 turning out to be a low-mass star. In 2025, spectroscopy showed that at least one component of seven of the pairs is a background source rather than a member of the cluster. Separate theoretical work argued that most JuMBOs could not survive the dense environment they sit in. Whether any real JuMBOs remain is an open question. (Wikipedia: JuMBO)
How many brown dwarfs are there?
Brown dwarfs are faint and cool, so they are easy to miss. Estimates for the solar neighborhood suggest there may be as many as six stars for every brown dwarf. A 2017 estimate based on the star cluster RCW 38 put the Milky Way's total at 25 to 100 billion brown dwarfs, compared with 100 to 400 billion stars. (Wikipedia: Brown dwarf)
Those numbers are large, but brown dwarfs weigh little. If a typical one has a few hundredths of a solar mass, 25 to 100 billion of them add up to a few billion solar masses, a small share of the galaxy's stellar mass and a tiny fraction of the dark matter halo around it. That is why astronomers do not consider brown dwarfs a solution to the dark matter problem. The evidence for dark matter and the candidates for it are laid out in Dark Matter and Dark Energy Explained.
Why astronomers care about them
Brown dwarfs are test beds for exoplanet science. A planet orbiting a star is lost in the star's glare, but a free-floating brown dwarf can be studied alone, which makes its atmospheric chemistry, clouds, and winds easier to measure. The same physics applies to young giant planets that are imaged directly, so brown dwarf spectra serve as templates for reading planet spectra. The James Webb Space Telescope's results on these cool objects fit into the survey in What James Webb Telescope Has Actually Found.
They also change the picture of where planets might exist. Planets orbiting brown dwarfs are expected to be carbon-rich and water-poor, and the habitable zone around a T dwarf is narrow and close in, about 0.005 astronomical units, shrinking as the dwarf cools. A planet that close would probably be locked with one face toward its host, a process described in What Is Tidal Locking?. (Wikipedia: Brown dwarf)
Finally, they mark the lower end of the stellar sequence. The classification scheme that sorts stars from O to M extends into L, T, and Y for substellar objects, as described in Types of Stars.
Common confusions
| Claim | What the evidence says |
|---|---|
| Brown dwarfs are brown | Not in any simple sense. L dwarfs are very dark red, and the cooler types emit mostly infrared, so their visible color is a poor guide to what they look like |
| They are just big planets | They are classified by mass and, ideally, by how they formed. Many probably formed like stars, from collapsing clouds |
| They never fuse anything | Objects above about 13 MJ fuse deuterium for a time, and those above about 63 to 65 MJ can burn lithium |
| A heavier brown dwarf is bigger | Radii vary by only 10 to 15 percent across the range, because degenerate matter gets denser rather than wider |
| Brown dwarfs explain dark matter | Their combined mass is far too small, even with tens of billions of them |
| A brown dwarf's spectral type reveals its mass | Not on its own, because a cooling brown dwarf moves through L, T, and Y types over time |
Frequently asked questions
Q: What is the difference between a brown dwarf and a star? A: A star sustains hydrogen fusion in its core for billions or trillions of years. A brown dwarf is too light, below about 80 Jupiter masses, so it can fuse deuterium for a while and in some cases lithium, then fades as it cools. The boundary varies a little with composition.
Q: What is the difference between a brown dwarf and a planet? A: The common rule is mass: roughly 13 Jupiter masses and below is a planet, and above that is a brown dwarf. The International Astronomical Union's 2018 working definition adds a mass ratio to the host, while the NASA Exoplanet Archive lists objects up to 30 Jupiter masses. Formation history matters but is hard to observe.
Q: How hot is a brown dwarf? A: It depends on mass and age. Young ones start as hot as late M-type stars, which run from about 2,100 to 3,500 K. L dwarfs are about 1,300 to 2,100 K, T dwarfs about 600 to 1,300 K, and Y dwarfs below roughly 600 K. The coldest known, WISE 0855-0714, was estimated at 225 to 260 K in its discovery paper.
Q: What was special about Gliese 229B? A: It was one of the first confirmed brown dwarfs, found in 1995, and the first T dwarf, with methane absorption in its spectrum. In 2024, two instruments on the Very Large Telescope resolved it into two brown dwarfs of about 38 and 34 Jupiter masses that orbit each other every 12.1 days.
Q: How are brown dwarfs found? A: They are faint and mostly glow in the infrared, so they are found in infrared sky surveys such as WISE and followed up with spectra. A spectral signature such as methane, or lithium in a cool object, points to a substellar nature.
Q: Can brown dwarfs have planets? A: Planets have been reported around some brown dwarfs, including strong evidence, reported in 2025, for a circumbinary planet around 2M1510. Models suggest such planets would be carbon-rich and water-poor, and any habitable zone would be very close to the dwarf and shrink with time.
The takeaway
A brown dwarf is best understood as a failure that turned out to be informative. It collapses like a star and then stalls, held up by quantum pressure instead of fusion, and it spends billions of years cooling from the glow of a dim star down to the temperature of a cold winter day. In between, it hosts clouds of iron, silicate, and salts, storms, auroras, and spin periods as short as an hour. The 2024 split of Gliese 229B shows how much of what we thought we knew rested on assuming a single body, and the disputes over JuMBOs and over WISE 0855's mass show that the line between planet and brown dwarf is still being drawn.
Sources
- Xuan, J. W. et al. (2024), The cool brown dwarf Gliese 229 B is a close binary
- Caltech (2024), It's Twins! Mystery of Famed Brown Dwarf Solved
- Thompson, W. et al. (2025), On the Orbit of the Binary Brown Dwarf Companion GL229 Ba and Bb
- Luhman, K. L. (2014), Discovery of a ~250 K Brown Dwarf at 2 pc from the Sun
- Cushing, M. C. et al. (2011), The Discovery of Y Dwarfs Using Data from the Wide-field Infrared Survey Explorer (WISE)
- Morley, C. V. et al. (2012), Neglected Clouds in T and Y Dwarf Atmospheres
- Wikipedia: Brown dwarf, Stellar classification, and Jupiter-mass binary object, used for summary figures and cross-checked against the primary papers above where possible
Related Reading
Follow new articles
Add SciHub101 to your feed reader to see new science and technology stories.
About the Author
SciHub101 is an independent science and technology publication founded and edited by Buğra Sözeri, a software engineer and app developer based in Türkiye. AI tools assist with drafting and research. Source documentation in the existing archive is uneven, and articles are being reviewed and corrected. The Editorial Team byline identifies editorial responsibility; it does not certify specialist or independent human review. Where we compare products, anything built by our own team is labelled as such — see the editorial guidelines and About page for how corrections and disclosures work.
Related Articles

What Is Tidal Locking? How Gravity Synchronizes Worlds
Tidal locking happens when a body's rotation period equals its orbital period, so one face always points at its partner. The Moon is locked to Earth, Mercury is caught in a 3:2 resonance instead, and many planets around red dwarf stars are probably locked to their stars.

The History of Space Exploration: From Sputnik to Artemis
Space exploration began on October 4, 1957, when the Soviet Union launched Sputnik 1. In the decades since, humans have walked on the Moon, robots have visited every planet, a space station has been continuously crewed since 2000, and Artemis II sent four astronauts around the Moon in April 2026.
Kepler's Three Laws: Why Planets Speed Up Near the Sun
Kepler's three laws explained with NASA data: why orbits are ellipses, why planets speed up near the Sun, and how P² = a³ weighs planets and finds exoplanets.
Mercury Explained: Iron Core, Polar Ice, Odd Orbit
Mercury has an oversized iron core, ice at its poles, and a day longer than its year. What MESSENGER found and what BepiColombo is due to do in November 2026.
What Are the Van Allen Belts? Earth's Trapped Radiation
The Van Allen belts are zones where Earth's magnetic field traps radiation. How Explorer 1 found them, how they work, and how Apollo and Artemis crossed them.
Why Is the Night Sky Dark? Olbers' Paradox Explained
The night sky is dark because the universe has a finite age, not because stars are few. How Olbers' paradox works, who solved it, and what it reveals.