---
title: 'HD 143811(AB)b: Directly Imaged Circumbinary Companion'
url: https://www.emergentmind.com/topics/hd-143811-ab-b-abc34709-b8f3-4a02-b4c9-4cf7feefc7b8
type: topic
---

# HD 143811(AB)b: Directly Imaged Circumbinary Companion

HD 143811(AB)b, also written HD 143811 AB b, is a directly imaged substellar companion orbiting the young spectroscopic binary HD 143811 AB in the Scorpius–Centaurus association. The system lies at a Gaia DR3 distance of \(136.85 \pm 0.38\) pc from a parallax of \(7.3065 \pm 0.0204\) mas, and the age adopted for planet characterization is \(13 \pm 4\) Myr. The companion was confirmed through multi-epoch high-contrast imaging and common proper motion analysis, and current analyses place it on a wide circumbinary orbit with a semi-major axis near \(64\) au and a mass of order \(5.6\)–\(6.1~M_\mathrm{Jup}\), depending on the adopted atmospheric and evolutionary framework [2509.06729].

## 1. Host system and stellar architecture

HD 143811 AB is a double-lined spectroscopic binary (SB2) identified with HIP 78663 and associated with Sco–Cen. Probabilistic membership from BANYAN \(\Sigma\) favors Upper Scorpius at \(72.1\%\), with a \(27.8\%\) chance of Upper Centaurus–Lupus and \(0.2\%\) field. The space motion is \((U, V, W) = (-2.68 \pm 0.04, -18.10 \pm 0.05, -4.68 \pm 0.02)\) km s\(^{-1}\). The adopted age used in the planet analyses is \(13 \pm 4\) Myr, synthesized from typical subgroup ages, although a stellar SED fit yielded a model-dependent age of \(20.9^{+1.8}_{-1.7}\) Myr [2509.06727].

The inner binary is unusually compact among hosts of directly imaged planets. Its spectroscopic orbit has period \(18.59098 \pm 0.00007\) days, eccentricity \(e = 0.4935 \pm 0.0013\), systemic velocity \(\gamma = 0.48 \pm 0.02\) km s\(^{-1}\), and mass ratio \(q = 0.885 \pm 0.003\). The radial-velocity semi-amplitudes are \(K_1 = 22.76 \pm 0.05\) km s\(^{-1}\) and \(K_2 = 25.71 \pm 0.06\) km s\(^{-1}\), and the argument of periastron for the primary is \(\omega_A = 155.91 \pm 0.18\) deg. Combining the spectroscopic constraints with stellar evolutionary modeling gives component masses \(M_A = 1.30^{+0.03}_{-0.05}~M_\odot\) and \(M_B = 1.15^{+0.03}_{-0.04}~M_\odot\), with total mass \(M_\mathrm{tot} \approx 2.46 \pm 0.07~M_\odot\). Kepler’s law then implies a binary semi-major axis \(a = 0.1854^{+0.0014}_{-0.0024}\) au.

The stellar properties are consistent with a near-equal-mass F+F pair. PHOENIX-based spectral fitting yielded \([\mathrm{M/H}] = -0.18 \pm 0.10\), \(T_\mathrm{eff}(A) = 6439 \pm 132\) K, \(T_\mathrm{eff}(B) = 5900 \pm 157\) K, \(v \sin i(A) = 6.8 \pm 0.6\) km s\(^{-1}\), and \(v \sin i(B) = 2.8 \pm 1.4\) km s\(^{-1}\). A separate MIST+ATLAS9 SED fit to blended Gaia and 2MASS photometry gave \([\mathrm{M/H}] = -0.15^{+0.07}_{-0.08}\), \(A_V = 0.16 \pm 0.01\) mag, and somewhat hotter temperatures, \(T_\mathrm{eff}(A) = 6751^{+207}_{-148}\) K and \(T_\mathrm{eff}(B) = 6349^{+122}_{-133}\) K. The authors note that the SED-based temperatures are systematically higher than the spectral-fitting values and attribute this to extinction–temperature covariance and model systematics; the masses are described as more robust than \(T_\mathrm{eff}\).

The planet therefore resides in a hierarchical circumbinary configuration: an \(\sim 18.6\)-day eccentric SB2 at \(\sim 0.19\) au, orbited by a giant planetary-mass companion at \(\sim 60\) au. This architecture is rare among directly imaged systems and provides the dynamical context for all subsequent interpretation.

## 2. Detection history and confirmation as a bound companion

The companion was initially detected in Gemini Planet Imager observations and later confirmed with Keck/NIRC2 and, in an independent analysis, with SPHERE. The primary discovery paper reports GPI integral-field spectroscopy in \(H\) band \((1.50\text{–}1.80~\mu\mathrm{m}; \Delta\lambda/\lambda \approx 40)\) on 2016 Apr 30 and 2019 Aug 11, together with Keck/NIRC2 \(L'\)-band imaging \((3.426\text{–}4.126~\mu\mathrm{m})\) on 2022 Jun 10. The GPI epochs comprised \(36 \times 60\) s and \(45 \times 60\) s integrations, while the NIRC2 data consisted of 175 frames of 60 coadds \(\times\) 0.5 s, for a total of 87.5 min [2509.06729].

| Epoch | Instrument / band | Relative astrometry |
|---|---|---|
| 2016 | GPI \(H\) | \(\rho = 429 \pm 3\) mas, \(\theta = 11.9 \pm 1.0\) deg |
| 2019 | GPI \(H\) | \(\rho = 429 \pm 10\) mas, \(\theta = 15 \pm 3\) deg |
| 2022 | Keck/NIRC2 \(L'\) | \(\rho = 421 \pm 9\) mas, \(\theta = 18.7 \pm 1.2\) deg |

The 2016 GPI detection had signal-to-noise ratio \(7.7\) using a forward model matched filter with an L-type template. Its measured \(H\)-band flux ratio was \((1.4 \pm 0.3) \times 10^{-5}\), corresponding to apparent \(H = 19.82 \pm 0.19\), absolute \(M_H = 14.14 \pm 0.19\), and integrated \(H\)-band flux \((1.37 \pm 0.24) \times 10^{-17}\) W m\(^{-2}\) \(\mu\)m\(^{-1}\). The 2019 GPI epoch had SNR \(3.5\), flux ratio \(0.78^{+0.87}_{-0.46} \times 10^{-5}\), and integrated flux \(0.77^{+0.80}_{-0.48} \times 10^{-17}\) W m\(^{-2}\) \(\mu\)m\(^{-1}\). The 2022 Keck/NIRC2 data gave an \(L'\)-band flux ratio \((1.8 \pm 0.8) \times 10^{-4}\), apparent \(L' = 16.56 \pm 0.32\), absolute \(M_{L'} = 10.87 \pm 0.32\), and flux \((1.26 \pm 0.36) \times 10^{-17}\) W m\(^{-2}\) \(\mu\)m\(^{-1}\).

Common proper motion was established by comparison with the Gaia DR3 proper motion and parallax of the host. The expected trajectory of a stationary background source is strongly inconsistent with the measured astrometry, and the observed separations disagree with the background track at \(\gtrsim 20\sigma\). Bound-companion orbital tracks generated by OFTI from the first epoch are consistent with the later epochs, confirming that HD 143811 AB b is gravitationally bound.

An independent SPHERE-based study reached the same conclusion with a longer temporal baseline. That analysis used 2016 and 2019 GPI data plus SPHERE/IRDIS and IFS observations obtained in 2025, reporting astrometry of \(\rho = 427.8 \pm 3.0\) mas, \(\theta = 11.79 \pm 0.46^\circ\) in 2016; \(\rho = 427.3 \pm 3.3\) mas, \(\theta = 16.29 \pm 0.38^\circ\) in 2019; and \(\rho = 431.1 \pm 0.63\) mas, \(\theta = 21.71 \pm 0.09^\circ\) in 2025. That work derived a relative proper motion \(\Delta \mu = 9.10 \pm 0.32\) mas yr\(^{-1}\), equivalent to \(\Delta v = 5.9 \pm 0.2\) km s\(^{-1}\) at projected separation \(\ge 60\) au, below the escape speed \(8.5 \pm 0.1\) km s\(^{-1}\), and reported a background false-alarm probability of \(3 \times 10^{-4}\) [2509.06009].

## 3. Orbital solution and circumbinary geometry

The planet’s orbit has been modeled with Orbits For The Impatient (OFTI). In the discovery analysis, the priors were uniform in \(\log(a)\), \(e\), \(\cos(i)\), \(\omega\), \(\Omega\), and \(T_0\), with parallax \(7.31 \pm 0.02\) mas and total mass \(2.46 \pm 0.07~M_\odot\). Using three relative astrometry epochs, the fitted posterior summary yielded semi-major axis \(a = 64^{+32}_{-14}\) au, eccentricity \(e = 0.23^{+0.24}_{-0.16}\), inclination \(i = 38 \pm 16^\circ\), and period \(P = 330^{+280}_{-100}\) yr. The posterior medians were \(a = 63.881\) au, \(e = 0.231\), \(i = 37.543^\circ\), \(P = 326.347\) yr, \(\Omega = 180.230^\circ\), \(\omega = 179.950^\circ\), and \(T_0 = 2112.574\), with broad credible intervals for the angular elements [2509.06729].

The period estimate is consistent with Kepler’s third law. In astronomical units and solar masses, the relation is \(P^2 = a^3/M_\mathrm{tot}\). Using \(a \approx 64\) au and \(M_\mathrm{tot} \approx 2.46~M_\odot\) gives \(P \approx \sqrt{64^3/2.46} \approx 326\) yr. The broad uncertainty in \(a\), rather than the small uncertainty in \(M_\mathrm{tot}\), dominates the reported period range.

Projected physical separations derived from \(s = \rho \times d\) are \(58.7 \pm 0.5\) au in 2016, \(58.7 \pm 1.4\) au in 2019, and \(57.6 \pm 1.3\) au in 2022, fully consistent with a fitted semi-major axis near \(64\) au and modest eccentricity. A parallel orbit analysis using GPI plus SPHERE astrometry found closely similar values: \(a = 63^{+30}_{-12}\) au, \(e = 0.19 \pm 0.12\), \(i = 37^{+11}_{-12}\) deg, and \(P = 319^{+248}_{-86}\) yr. The agreement between the two studies indicates that, with present astrometric coverage, the outer orbit is already constrained to be wide, low-to-moderately eccentric, and mostly face-on, while \(\Omega\), \(\omega\), and \(T_0\) remain weakly constrained [2509.06009].

The relation between the inner and outer orbital planes remains unresolved. The host-binary characterization paper inferred a binary inclination \(i_\mathrm{AB} = 22.9^{+0.3}_{-0.2}\) deg, with a spectroscopically degenerate retrograde alternative at \(156.1^{+0.2}_{-0.3}\) deg. Because spectroscopy does not constrain the inner node \(\Omega\), and the planetary \(\Omega\) posterior remains broad, the mutual inclination cannot yet be computed. The current estimates overlap within roughly \(1\sigma\), so coplanarity cannot be ruled out, but neither can it be established [2509.06727].

## 4. Photometry, spectroscopy, and atmospheric modeling

The planet’s spectro-photometric characterization currently rests on a GPI \(H\)-band spectrum and thermal-infrared photometry. The 2016 GPI \(H\)-band spectrum was extracted with KLIP-FM forward modeling and Bayesian PSF fitting, calibrated using satellite spots and the stellar SED, and was described as exhibiting strongly correlated spectral noise typical for GPI but robustly measured flux. The 2022 NIRC2 \(L'\) photometry was calibrated to stellar \(L'\) through the system SED model and was reported as consistent with a cool, substellar companion [2509.06729].

In color–magnitude space, the combined measurements yield \(H_{GPI} - L'_{NIRC2} = 3.3 \pm 0.4\) mag and \(M_{L'} = 10.87 \pm 0.32\) mag. On a color–magnitude diagram, the object lies near late-L/early-T field objects and close to the directly imaged companions HD 206893 B and 2M 1207 B. The discovery paper states that this positioning suggests a very red, dusty atmosphere may be plausible. This is explicitly framed as an interpretation of the photometric placement rather than a direct measurement.

Atmospheric model comparison was carried out against 842 PHOENIX synthetic stellar spectra and 9575 Exo-REM exoplanet atmosphere models, including clouds and disequilibrium chemistry, across \(T_\mathrm{eff} = 400\text{–}13{,}500\) K. The initial goodness-of-fit statistic was
\[
\chi^2 = \sum_i \frac{(f_{\mathrm{obs},i} - f_{\mathrm{model},i})^2}{\sigma_i^2}.
\]
For the GPI spectrum alone, PHOENIX and Exo-REM both matched moderately well, with PHOENIX reaching a maximum \(\chi^2 \approx 3.4\), and Exo-REM slightly preferred. For the joint GPI+NIRC2 fit, Exo-REM was strongly preferred, with best-fit \(\chi^2 = 3.31\), while the PHOENIX templates were rejected. A representative best-fit Exo-REM model had \(T_\mathrm{eff} \approx 1050\) K, \(\log g = 4.5\), \([\mathrm{Fe/H}] = +1.0\) dex, and \(\mathrm{C/O} = 0.35\), but the paper explicitly states that \(\log g\), metallicity, and C/O are not constrained by the data.

A more complete inference used nested sampling with `dynesty` and a Gaussian-process noise model for the correlated GPI spectral noise. That analysis yielded \(T_\mathrm{eff} = 1042^{+178}_{-132}\) K, radius \(R = 1.7^{+0.7}_{-0.4}~R_\mathrm{Jup}\), and luminosity \(L = 3.3^{+0.8}_{-0.6} \times 10^{-5}~L_\odot\). Surface gravity, metallicity, and C/O ratio remained unconstrained across the Exo-REM grid. Under hot-start evolutionary models of Baraffe et al. (2003), and adopting age \(13 \pm 4\) Myr, the derived luminosity corresponds to a mass \(M = 5.6 \pm 1.1~M_\mathrm{Jup}\), with an evolutionary radius \(1.41 \pm 0.03~R_\mathrm{Jup}\), consistent with the atmospheric-fit radius within uncertainties.

A separate GPI+SPHERE analysis used the HADES coupled atmosphere–interior grid with an affine-invariant MCMC, including SPHERE/IRDIS \(H2/H3\) photometry and SPHERE/IFS \(YJ\) upper limits. That work reported \(H2 = 14.31 \pm 0.10\) mag, \(H3 = 13.85 \pm 0.10\) mag, \(H2-H3 = 0.44 \pm 0.14\) mag, and a \(5\sigma\) non-detection limit \(J > 21.34\) mag, implying \(J-H2 > 1.4\) mag. It derived intrinsic temperature \(T_\mathrm{int} = 1000 \pm 30\) K, mass \(M_p = 6.1^{+0.7}_{-0.9}~M_\mathrm{Jup}\), radius \(R_p = 1.4 \pm 0.1~R_\mathrm{Jup}\), atmospheric metallicity \([\mathrm{Fe/H}] = 0.1^{+0.8}_{-0.4}\) dex, cloud sedimentation efficiency \(f_\mathrm{sed} = 2.8^{+1.1}_{-0.8}\), and vigorous vertical mixing \(\log K_{zz} = 9.0^{+1.2}_{-1.5}\). It interpreted the H-band shape and red \(YJ-H\) color as consistent with strong \(\mathrm{H_2O}\) absorption and muted \(\mathrm{CH_4}\) near the L/T transition in a young, cloudy, chemically disequilibrium atmosphere [2509.06009].

## 5. Dynamical regime and formation significance

HD 143811(AB)b belongs to a small class of directly imaged planets orbiting binaries rather than single stars. The discovery paper explicitly places it among few such systems and compares it with HD 106906 b, b Cen b, and WISPIT 1bc, while also noting borderline planetary-mass companions around binaries such as Ross 458C, SR 12 C, 2MASS J01033563–5515561 (AB)b, and ROXs 42B b. Within that group, HD 143811 AB b is distinguished by a moderate separation of about \(60\) au, modest eccentricity near \(0.23\), and youth at \(13 \pm 4\) Myr [2509.06729].

The inner binary is sufficiently tight that the planet resides deep in the stable circumbinary regime. The host-binary study invokes the Holman & Wiegert (1999) critical semimajor axis framework, writing \(a_\mathrm{crit} = a_\mathrm{bin} f(e,\mu)\) with \(\mu = M_B/(M_A+M_B)\). Using \(a_\mathrm{bin} \approx 0.185\) au, \(e \approx 0.494\), and \(\mu \approx 0.469\), it concludes that \(a_\mathrm{crit}\) remains orders of magnitude smaller than the planet’s \(\sim 59\) au orbit, so the system is comfortably within the stable circumbinary regime. A separate analysis, adopting a compact binary with \(a_\mathrm{bin} \lesssim 3\) au, similarly states that the planet’s orbit at \(a \approx 63\) au comfortably exceeds any plausible critical stability radius [2509.06727].

The formation interpretation remains open. The discovery paper states that the planet’s moderate separation, modest eccentricity, and youthful age provide a valuable data point for understanding planet formation and dynamical evolution in circumbinary environments where core accretion is challenged but not excluded. The host-binary study adds that the tight inner binary likely truncated and structured the primordial circumbinary disk, setting initial conditions for planet formation and migration, and that the large binary–planet separation ratio of \(\sim 300\) is consistent with long-term stability. It further states that the observed configuration suggests formation in a circumbinary disk at tens of au, or post-formation outward migration or scattering, but does not distinguish among these scenarios.

The system may also include circum-system debris. Spitzer detections indicate a cold debris belt with fractional infrared luminosity between \(3.3 \times 10^{-5}\) and \(2.4 \times 10^{-4}\), inferred temperatures of \(\sim 68\text{–}140\) K, and blackbody radii of \(\sim 8.5\text{–}90\) au, although the host-binary paper emphasizes that grain properties and the then-unknown binary nature can shift radii outward and that a 70 \(\mu\)m non-detection was reported. This suggests that the architecture may eventually permit joint analysis of binary, planet, and disk dynamics, but current disk parameters remain too uncertain for strong conclusions.

## 6. Open questions and future characterization

Several central parameters remain weakly constrained. For the planet, the current three-epoch orbit fits already limit \(a\), \(e\), \(i\), and \(P\), but \(\Omega\), \(\omega\), and \(T_0\) remain broad. For the inner binary, spectroscopy fixes the period, eccentricity, and mass ratio with high precision but cannot determine the longitude of the ascending node, and radial velocities alone leave a mirror degeneracy between prograde and retrograde orbital inclinations. Consequently, the full three-dimensional architecture and mutual inclination are not yet known [2509.06727].

Continued high-precision relative astrometry is expected to refine the planetary orbital parameters \(a\), \(e\), \(i\), \(\Omega\), \(\omega\), and \(P\), and may eventually permit dynamical mass constraints when combined with absolute astrometry and a well-determined binary orbit. The discovery paper specifically notes that Hipparcos–Gaia accelerations could contribute once the host-binary solution is improved. The host-binary paper recommends VLTI/GRAVITY, since the binary apastron corresponds to approximately \(2\) mas at \(136.9\) pc, potentially allowing recovery of the inner \(\Omega\) and true \(i\), and cites Gaia DR4 individual-scan astrometry as another route to constraining the \(\sim 0.1\) mas photocenter wobble.

Additional spectroscopy across broader wavelength coverage is a major priority. The discovery analysis recommends further observations in \(J\), \(K\), and \(M\) bands and medium-resolution spectroscopy to constrain \(\mathrm{H_2O}\), \(\mathrm{CH_4}\), and CO, as well as cloud properties, metallicity, and C/O, while reducing the impact of correlated-noise systematics. It explicitly recommends follow-up with upgraded high-contrast imagers such as Gemini Planet Imager 2.0, along with complementary Keck observations. The GPI+SPHERE study further identifies JWST NIRSpec/G395M and MIRI/LRS–MRS as promising for resolving molecular and cloud signatures across \(1\text{–}12~\mu\)m, and also points to variability, polarimetry, and ELT-class high-resolution spectroscopy as routes to probing cloud patchiness, particle size, winds, rotation, and disequilibrium chemistry [2509.06009].

The scientific importance of these follow-up programs is twofold. First, they can test whether the apparent consistency of the binary and planetary inclinations reflects genuine coplanarity expected from formation in a common circumbinary disk. Second, they can determine whether HD 143811(AB)b is best interpreted as a planet formed in situ in a circumbinary disk, a product of migration, or an object whose present orbit encodes later dynamical evolution. Present data suggest several of these possibilities, but do not yet isolate a unique formation pathway.

Source: https://www.emergentmind.com/topics/hd-143811-ab-b-abc34709-b8f3-4a02-b4c9-4cf7feefc7b8