Why Is Alpha Centauri So Hard to Find Planets Around?
<sub>Image courtesy of NASA: [esawebb.org/images/weic2515c/](https://esawebb.org/images/weic2515c/)</sub>
*Written by OC Wanderer, author of Destiny Among the Stars, a litrpg sci-fi series set in the real stellar neighborhood. Published April 7, 2026.*
**Short answer:** Three problems stack on top of each other. Alpha Centauri A and B are a binary pair (two stars orbiting each other), which makes planet formation harder, makes detection noisier, and created the conditions for the most famous false-positive planet detection in history. The only confirmed planet in the system orbits the third star, Proxima Centauri, not either of the bright pair.
The Binary Orbit Is the Root of Every Problem
Alpha Centauri A and B are not a conveniently distant pair. They orbit each other on a highly eccentric path: semi-major axis ~23.4 AU, eccentricity ~0.52. Their separation swings from about 11 AU at closest approach to 36 AU at widest. That geometry does three things at once.
First, it truncates the protoplanetary disk around each star. Any circumstellar disk that extended beyond roughly 3 AU from either star would have been gravitationally shredded by the companion over millions of years. Both disks were left with roughly 30 Earth masses of raw material to work with. That is a slim budget for building terrestrial planets.
Second, it disrupts planet formation where it matters most. Thebault, Marzari, and Scholl (2008) modeled the planetesimal collision stage around Alpha Cen B and found that the binary's gravitational tugs accelerate collision velocities into the erosion regime across most of the habitable zone. Instead of small rocks sticking together, they grind each other down. The conditions that allowed Earth to form simply did not exist at the same orbital distances around Alpha Cen B, unless the binary happened to start with significantly wider separation or lower eccentricity than its current configuration.
Third, once you accept that some planet might have formed anyway, you run into the detection floor: current radial velocity instruments cannot see an Earth-mass world in the habitable zone of either star. Zhao et al. (2017) combined over a decade of data from three separate spectrographs and concluded the minimum detectable mass around Alpha Cen A is roughly 53 Earth masses in the habitable zone. For Alpha Cen B, that number improves to about 8.4 Earth masses. An actual Earth is invisible to all existing instruments.
**Bottom line:** Alpha Centauri's binary nature simultaneously makes planets harder to form, harder to detect, and harder to confirm once a candidate appears.
Are Planets Dynamically Possible at All?
Yes. Planets that somehow formed would be stable. That is worth separating from the formation problem.
In binary star systems, planets can survive in two configurations: an S-type orbit, where the planet circles one star and the second acts as a distant perturber, and a P-type orbit, where the planet circles both stars from far outside the pair. For Alpha Centauri, P-type orbits require semi-major axes beyond roughly 80 AU. There is no habitable zone out there.
S-type orbits are the relevant scenario. Quarles and Lissauer (2016) mapped the stability landscape numerically and found that prograde orbits around Alpha Cen A remain stable out to about 2.56 AU, and around Alpha Cen B to about 3 AU. Both stars' habitable zones sit inside those limits. Andrade-Ines and Michtchenko (2014) confirmed that orbital inclinations below 40 degrees produce stable, regular motion with eccentricity variations comparable to Earth's.
The subtlety is that a planet in the habitable zone is not on a simple circular orbit. The binary companion applies a forced eccentricity, a small but real wobble in the planet's orbital shape that the companion imposes over centuries. Quarles et al. (2018) showed that planets initialized near their forced eccentricity value are actually more stable than those starting on circular orbits, which suggests initial conditions at formation matter a great deal.
Forgan (2012) added a climate wrinkle: even a geologically stable planet inside Alpha Cen B's habitable zone would experience periodic temperature oscillations of a few Kelvin as the binary orbit completes each roughly 80-year cycle. Not catastrophic for most atmospheric compositions, but enough to push a planet near the habitable zone edges in and out of snowball or hot states.
**Bottom line:** Planets in the habitable zones of Alpha Cen A and B would survive indefinitely if they formed. The formation step is the barrier, not long-term stability.
Why Radial Velocity Detection Fails Here
Radial velocity is the dominant method for finding planets around nearby stars. When a planet orbits a star, it tugs the star in a small circle. That circular motion produces a periodic Doppler shift in the star's light spectrum, which a spectrograph can measure as the star's velocity wobbling toward and away from Earth.
For an Earth-mass planet at habitable-zone distances from a Sun-like star, that wobble is about 9 cm per second. Stars produce their own velocity noise at levels that dwarf this:
- **Acoustic oscillations** (p-modes) ring the stellar surface like a bell at amplitudes of 0.5 to 1 meter per second, roughly every 5 to 15 minutes. - **Granulation noise** from convective cells rising and sinking at the stellar surface contributes correlated velocity patterns at about 1 meter per second on timescales of hours to days. - **Starspot and activity-cycle jitter** from rotating magnetic features on the stellar surface generates periodic velocity shifts at the stellar rotation period, and slow drifts of 1 to 9 meters per second through an activity cycle that can span 10 years.
All of this noise is correlated in time, not random. Correlated noise is particularly dangerous because it produces peaks in period-finding analyses that look statistically significant but are not real. Standard significance tests assume noise is uncorrelated. When it is not, false detections become common.
Alpha Centauri A and B are unusually quiet stars magnetically, which should help. But the binary geometry adds another layer: the spectrograph cannot always cleanly isolate one star's light from the other's. Lisogorskyi, Jones, and Feng (2019) showed that contamination from the companion's spectrum bleeds into HARPS measurements, requiring careful filtering of roughly 5 percent of the dataset to achieve meaningful noise reduction.
**Bottom line:** Stellar noise in alpha Centauri's bright pair operates at 1 to 9 meters per second. An Earth-mass planet in the habitable zone would produce a signal at 9 centimeters per second. The two differ by a factor of 10 to 100.
The Alpha Centauri Bb Incident
In October 2012, a team of astronomers led by Xavier Dumusque published a paper in *Nature* announcing the detection of an Earth-mass planet orbiting Alpha Centauri B. The team had four years of HARPS data, 459 spectra, and a 0.51 meters-per-second periodic signal with a 3.236-day period. The planet was named Alpha Centauri Bb.
The announcement made international news. Earth's nearest stellar neighbor had an Earth-sized planet.
It was not real.
The detection had required multiple signal-removal steps: subtract the large Doppler motion from the binary orbit, subtract stellar activity signals using a chromospheric activity indicator, then search the residuals for planetary signals. Each subtraction step was mathematically careful. But the combination of those steps, applied to that particular dataset with that particular time-sampling pattern, amplified a spectral peak at 3.236 days that had nothing to do with a planet.
Artie Hatzes (2013) flagged the problem within months, showing that two different activity-removal methods returned conflicting significance estimates for the same signal. The decisive analysis came from Rajpaul, Aigrain, and Roberts (2015), who named their paper precisely: "Ghost in the Time Series: No Planet for Alpha Cen B." They ran the identical analysis pipeline on synthetic datasets containing pure random noise at the same time stamps as the real observations. Those synthetic datasets, with no planet signal at all, produced the same apparent 3-sigma detection at 3.236 days.
The "planet" was an artifact of when the telescope pointed at the sky, not of what was orbiting the star. The technical term is a window function artifact: a false peak created by the interaction of the time-sampling cadence with the signal-removal procedure.
Dumusque publicly acknowledged the finding: "We are not 100 percent sure, but probably the planet is not there." It is listed as unconfirmed in all subsequent literature and is almost certainly non-existent.
The episode forced a methodological upgrade across radial velocity astronomy. Any new Earth-mass planet candidate around an active star must now pass a higher bar: signal-to-noise above 5 sigma, validation using a Gaussian process noise model that jointly fits RV data and stellar activity indicators, and a synthetic injection-recovery test proving the signal does not appear in activity-only data. The field learned what it did not know how to do from the mistake.
**Bottom line:** The only announced planet detection around Alpha Centauri A or B was a statistical artifact that became a cautionary tale for the entire field of radial velocity astronomy.
What Actually Is Confirmed
As of 2026, no planet has been confirmed around Alpha Centauri A or Alpha Centauri B. But the picture shifted in August 2025.
Using Webb's Mid-Infrared Instrument (MIRI), astronomers detected excess infrared light around Alpha Centauri A consistent with a Saturn-mass gas giant orbiting at roughly 2 AU. The team ruled out dust disks, background objects, and instrument artifacts before settling on a planet candidate as the best explanation. The key distinction from the Bb debacle: this is a direct thermal emission signal, not a radial velocity extraction that required multiple noise-subtraction steps. It is harder to manufacture from statistical artifacts. The researchers are careful to call it a candidate. Follow-up observations in February and April 2025 did not re-detect the object, which the team explained through orbital modeling rather than treating as a retraction. The finding stands as the strongest direct evidence yet for a planet around Alpha Centauri A.
A 2021 candidate from the Wagner et al. team using mid-infrared imaging from the VLT was less certain — the same paper flagged a possible instrumental origin. The 2025 Webb result is independent of that earlier attempt and uses a more capable instrument.
The one firmly confirmed planet in the system orbits Proxima Centauri, the third and faintest member. Proxima b, announced in 2016 by Anglada-Escudé and 30 co-authors, is a minimum 1.27 Earth-mass world on an 11.2-day orbit within Proxima's habitable zone. Proxima is an M-dwarf flare star, so the planet's habitability is a separate and complicated question. But the detection itself is solid.
Proxima is gravitationally bound to the Alpha Centauri AB pair. Kervella, Thévenin, and Lovis (2017) confirmed that Proxima orbits the pair on a 550,000-year period, currently near its farthest point at about 13,000 AU separation. All three stars likely formed from the same molecular cloud, share the same age of 5 to 7 billion years, and share the same bulk composition.
**Bottom line:** Proxima b is the only confirmed planet in the system. A Saturn-mass candidate around Alpha Centauri A detected by Webb in 2025 is the strongest evidence yet that the bright pair hosts planets too.
What TOLIMAN Is Trying to Do
The TOLIMAN mission is a small satellite project led by Peter Tuthill's group at the University of Sydney, with partners including NASA's Jet Propulsion Laboratory and Breakthrough Initiatives. Its launch is targeted for 2027.
TOLIMAN approaches the detection problem differently. It does not use radial velocity. It uses astrometry: instead of measuring Doppler velocity shifts, it measures the tiny wobble in a star's position on the sky caused by an orbiting planet's gravitational pull. For an Earth-mass planet at habitable-zone distances, that wobble is approximately 1 micro-arcsecond per year, roughly ten millionths of a single detector pixel.
The technical innovation is a diffractive pupil mask that spreads starlight from both Alpha Centauri A and B into an overlapping fringe pattern. The relative position of the two stars is encoded in those fringes with precision that is largely immune to the spacecraft's own pointing jitter and optical distortions, the systematic errors that normally defeat high-precision astrometry from space.
This approach sidesteps both the stellar noise problem that defeats radial velocity searches and the transit probability problem that makes space photometry unlikely to work on a nearby, well-characterized system like Alpha Centauri.
If a rocky planet exists in the habitable zone of Alpha Cen A or B, TOLIMAN would be able to detect it. The mission will operate for three years.
**Bottom line:** TOLIMAN is the first instrument purpose-built to find Earth-mass planets around the Alpha Centauri pair, using a detection method that bypasses the noise problems that have defeated every previous attempt.
What Does the Science Actually Expect to Find?
The honest answer is that nobody knows, and the uncertainty runs in both directions.
Cuello and Sucerquia (2024) synthesized the current picture: rocky planets are dynamically possible below 3 AU around each star, but the disk mass budget was thin and planet formation conditions were marginal. The binary's current orbital eccentricity may have been even higher during the epoch of planet formation, which would have made the formation conditions worse. Or the binary may have formed at wider separation, which would have helped.
Worth and Sigurdsson (2016) modeled the three-star system dynamics and found that Proxima's gravitational perturbations on the AB disk were modest and most likely did not prevent terrestrial planet formation around A or B. So the faint third member is probably not the obstacle.
The 2025 Webb detection of a Saturn-mass candidate around Alpha Cen A at 2 AU fits comfortably inside the dynamical stability limit Quarles and Lissauer established. A gas giant at that distance is consistent with the theoretical models — and its presence does not rule out smaller rocky planets closer in. If anything, a confirmed giant planet in the system strengthens the case that planet formation happened here despite the binary's interference.
The simplest summary is no longer "planets are undetected." It is: a gas giant candidate exists, and the rocky planets the habitable zone models permit have not been ruled out.
**Bottom line:** The science went from "probably possible" to "probably present" between 2024 and 2025. Confirmation is still pending, but the direction of evidence has changed.
Reality vs. Fiction
| | Real science | What sci-fi often does | |-|---|---| | Alpha Centauri's planets | No confirmed planets around A or B; a Saturn-mass candidate detected by Webb in 2025 is the strongest evidence yet; Proxima b confirmed around the faint third member | Treats Alpha Centauri as an obvious target with ready-made habitable worlds | | Binary star planet formation | The binary orbit truncates disks and drives erosive collisions in the habitable zone | Binary stars treated as cosmetically interesting but functionally the same as single stars | | The Bb detection | Retracted; a window function artifact from time-sampling the observations | Widely cited for years in popular media as the nearest known exoplanet | | Radial velocity precision | 9 cm/s signal; 1 m/s noise floor; Earth-mass planets currently undetectable | Detection presented as a straightforward measurement problem, not a signal-to-noise crisis |
> [!lore] The crew in *Destiny Among the Stars* finds sixteen planets in this system on arrival. This is why Earth had no idea they were there. [Start Reading Today →](https://ocwanderer.com/storytime/story/destiny-among-the-stars)
How This Shows Up in Destiny Among the Stars
In *Destiny Among the Stars*, the crew arrives at the Alpha Centauri system with one planet on their list: Proxima b. That is the confirmed world, the one with real data behind it, and the mission is built around it. Nobody on the Triumph of Darron is expecting a surprise from the binary pair.
The surprise comes anyway. Once they are actually in the vicinity of the system, their instruments start finding planets around Alpha Cen A and B that no telescope at Earth could have detected. Sixteen in total across the system, including the Proxima b they came for. One of them, the fourth planet of Alpha Centauri A, turns out to be Midnight Veil with a toxic atmosphere. The signal-to-noise problem that defeats radial velocity searches from 4.24 light-years away simply does not exist when you are already there.
That detail carries the real-science weight cleanly. The detection floor described in this post, eight times Earth's mass from Earth's distance, is a distance problem as much as an instrument problem. It captures something the Fermi paradox literature dances around but rarely states directly: absence of evidence from far away is not evidence of absence. The Alpha Centauri system looked empty from a distance. It was not.
The Alpha Centauri Bb incident is in the worldbuilding as historical context, not an active plot element. But the broader lesson it points to runs throughout the story: the gap between what institutions announce and what is actually true, and the specific danger of trusting a confident answer that came from instruments that could not reach the thing they were measuring.
What Luca Thinks About This
*Luca Rossi is the twenty-year-old captain of the Triumph of Darron in* Destiny Among the Stars. *Hand him the research. This is what you get.*
Going into this mission, the briefing was honest about what we knew and what we didn't. Proxima b was the target. Confirmed planet, habitable zone, real data behind it. Alpha Centauri A and B were the bright pair we'd fly past to get there, interesting as navigational landmarks but a question mark for anything else. The instruments on Earth couldn't tell us what was orbiting them. Nobody failed to try. The noise from the stars themselves was louder than any Earth-sized signal could be from that distance.
I remember Danny mentioning the detection floor numbers before departure. Eight point four Earth masses minimum to see anything around Alpha Cen B. Fifty-three around Alpha Cen A. Meaning if there was an actual Earth there, we would not know it until we showed up. That was just the reality of the situation.
What I didn't expect was how fast it changed once we were actually there. We mapped the system on arrival. Sixteen planets. Proxima b was one of them. Sixteen. The binary pair had been sitting on a full planetary system the entire time and Earth had no idea because the instruments couldn't reach it.
That's the number I keep coming back to. Planets existing there was never the surprise. What got me was how confidently we didn't know. There's a difference between "we haven't confirmed it" and "it probably isn't there," and somewhere along the way those two things got treated as the same statement.
Proxima b was still the one that mattered most. That's what we came for. But sixteen planets is a different kind of answer than zero.
> [!lore] If you want a LitRPG where the sci-fi is as hard as the leveling system, check out *Destiny Among the Stars*. [Available now.](https://ocwanderer.com/storytime/story/destiny-among-the-stars)
Related Questions
**Is Proxima b the only planet in the Alpha Centauri system?** It is the only confirmed planet. Proxima b orbits Proxima Centauri, the faint red dwarf bound to the Alpha Centauri AB pair. No planets have been confirmed around Alpha Centauri A or B, and the most famous candidate, Alpha Centauri Bb, was retracted as a false positive in 2015.
**Could Alpha Centauri A or B have planets that we simply cannot detect yet?** Yes. The current radial velocity detection floor around Alpha Cen B is roughly 8.4 Earth masses in the habitable zone, and around Alpha Cen A roughly 53 Earth masses. An Earth-mass planet would produce a signal about 10 to 100 times smaller than current instruments can resolve from stellar noise. TOLIMAN, launching in 2027, uses a different detection method that could reach Earth-mass sensitivity.
**Did the binary nature of Alpha Centauri prevent planets from forming there?** Possibly, but not certainly. The binary's orbital geometry truncated the protoplanetary disk around each star and drove erosive collision velocities among early planetesimals. Thebault et al. (2008) found that planet formation in the habitable zone of Alpha Cen B would have required the binary to be wider or less eccentric than its present configuration. Some planet formation may have occurred, but the conditions were marginal.
**What was Alpha Centauri Bb?** Alpha Centauri Bb was a planet announced in 2012 by Dumusque et al. in *Nature*, based on four years of HARPS radial velocity data. It was not confirmed and is almost certainly not real. Rajpaul et al. (2015) showed the signal was a statistical artifact of the time-sampling pattern of the observations: a "window function ghost." The same apparent signal appeared in synthetic datasets with no planet at all.
**How far apart are Alpha Centauri A and B?** Their separation varies from about 11 AU at closest approach to about 36 AU at widest, completing a full orbit in approximately 80 years. The average separation is about 23.4 AU, roughly the distance from the Sun to Uranus. Their current orbital eccentricity of 0.52 is unusually high for a binary system and is a primary reason why planet formation conditions were disrupted.
**What is TOLIMAN designed to detect?** TOLIMAN is a small satellite designed to measure the astrometric wobble of Alpha Centauri A and B caused by orbiting planets. It uses a diffractive pupil mask to encode the stars' relative positions in a fringe pattern with micro-arcsecond precision, bypassing the stellar activity noise that defeats radial velocity searches. It is targeted to launch in 2027 for a three-year mission.
> [!info] Related Articles > > - [Why Is Alpha Centauri So Hard to Find Planets Around?](https://ocwanderer.com/blog/why-is-alpha-centauri-so-hard-to-find-planets-around)
Sources
1. Cuello, N. and Sucerquia, M. "Alpha Centauri: Disc Dynamics, Planet Stability, Detectability." *Universe*, 2024. [arXiv:2401.16003](https://arxiv.org/abs/2401.16003)
2. Thebault, P., Marzari, F., and Scholl, H. "Planet formation in the habitable zone of alpha Centauri B." *MNRAS Letters*, 2008. [arXiv:0811.0673](https://arxiv.org/abs/0811.0673)
3. Andrade-Ines, E. and Michtchenko, T. A. "Dynamical Stability of Terrestrial Planets in the Binary alpha Centauri System." *MNRAS*, 2014. DOI: [10.1093/mnras/stu1591](https://doi.org/10.1093/mnras/stu1591)
4. Zhao, L. L., Fischer, D. A., Brewer, J. M., Giguere, M., and Rojas-Ayala, B. "Planet Detectability in the Alpha Centauri System." *Astronomical Journal*, 2017. DOI: [10.3847/1538-3881/aa9bea](https://doi.org/10.3847/1538-3881/aa9bea) | [arXiv:1711.06320](https://arxiv.org/abs/1711.06320)
5. Lisogorskyi, M., Jones, H., and Feng, F. "Activity and telluric contamination in HARPS observations of Alpha Centauri B." *MNRAS*, 2019. DOI: [10.1093/mnras/stz694](https://doi.org/10.1093/mnras/stz694)
6. Dumusque, X., Pepe, F., Lovis, C., et al. "An Earth-Mass Planet Orbiting alpha Centauri B." *Nature*, 491, 207-211, 2012. DOI: [10.1038/nature11572](https://doi.org/10.1038/nature11572)
7. Hatzes, A. P. "Radial Velocity Detection of Earth-Mass Planets in the Presence of Activity Noise." *ApJ*, 770, 133, 2013. DOI: [10.1088/0004-637X/770/2/133](https://doi.org/10.1088/0004-637X/770/2/133) | [arXiv:1305.4960](https://arxiv.org/abs/1305.4960)
8. Rajpaul, V., Aigrain, S., and Roberts, S. J. "Ghost in the Time Series: No Planet for Alpha Cen B." *MNRAS Letters*, 456, L6-L10, 2016. DOI: [10.1093/mnrasl/slv164](https://doi.org/10.1093/mnrasl/slv164) | [arXiv:1510.05598](https://arxiv.org/abs/1510.05598)
9. Quarles, B. and Lissauer, J. J. "Long-Term Stability of Planets in the alpha Centauri System." *AJ*, 151, 111, 2016. DOI: [10.3847/0004-6256/151/5/111](https://doi.org/10.3847/0004-6256/151/5/111) | [arXiv:1604.04917](https://arxiv.org/abs/1604.04917)
10. Quarles, B., Lissauer, J. J., and Kaib, N. "Long-Term Stability of Planets in the alpha Centauri System, II: Forced Eccentricities." *AJ*, 155, 2018. DOI: [10.3847/1538-3881/aaa197](https://doi.org/10.3847/1538-3881/aaa197) | [arXiv:1801.06116](https://arxiv.org/abs/1801.06116)
11. Anglada-Escude, G. et al. "A terrestrial planet candidate in a temperate orbit around Proxima Centauri." *Nature*, 536, 437, 2016. DOI: [10.1038/nature19106](https://doi.org/10.1038/nature19106)
12. Kervella, P., Thevenin, F., and Lovis, C. "Proxima's orbit around alpha Centauri." *A&A*, 598, L7, 2017. DOI: [10.1051/0004-6361/201629930](https://doi.org/10.1051/0004-6361/201629930) | [arXiv:1611.03495](https://arxiv.org/abs/1611.03495)
13. Worth, R. and Sigurdsson, S. "Effects of Proxima Centauri on Planet Formation in Alpha Centauri." *ApJ*, 2016. [arXiv:1607.03090](https://arxiv.org/abs/1607.03090)
14. Tuthill, P. et al. "Getting to know the neighbours: Earth analogues in Alpha Centauri with the TOLIMAN space telescope." *SPIE Proceedings*, 2024. [ADS: 2024SPIE13092E..0CT](https://ui.adsabs.harvard.edu/abs/2024SPIE13092E..0CT/abstract)
15. Holman, M. J. and Wiegert, P. A. "Long-Term Stability of Planets in Binary Systems." *AJ*, 117, 621-628, 1999. DOI: [10.1086/300695](https://doi.org/10.1086/300695)
16. TOLIMAN Space Telescope official site. University of Sydney. [https://toliman.space/](https://toliman.space/)
17. ESO Press Release eso1629. "Planet Found in Habitable Zone Around Nearest Star." August 2016. [https://www.eso.org/public/news/eso1629/](https://www.eso.org/public/news/eso1629/)
18. NASA Webb Mission. "NASA's Webb Finds New Evidence for Planet Around Closest Solar Twin." August 7, 2025. [https://science.nasa.gov/missions/webb/nasas-webb-finds-new-evidence-for-planet-around-closest-solar-twin/](https://science.nasa.gov/missions/webb/nasas-webb-finds-new-evidence-for-planet-around-closest-solar-twin/)