[Paper Review] The CORALIE survey for southern extra-solar planets IX. A 1.3-day period brown dwarf disguised as a planet

N. C. Santos, M. Mayor+9 more|ArXiv.org|Jun 13, 2002

Stellar, planetary, and galactic studiesPhysics and Astronomy27 references182 citations

TL;DR

This paper identifies a 1.3-day radial-velocity signal in the HD 41004 AB binary system, initially interpreted as a low-mass planet around the K0V star HD 41004 A. However, detailed analysis of the bisector of the cross-correlation function (CCF) and spectral blending reveals the signal arises from a brown dwarf companion orbiting the M-dwarf component HD 41004 B, making it the first known brown dwarf in a 1.3-day orbit around an M dwarf. The study highlights the critical importance of bisector analysis to avoid false planet detections in binary systems.

ABSTRACT

In this article we present the case of HD 41004 AB, a system composed of a K0V star and a 3.7-magnitude fainter M-dwarf companion separated by only 0.5 arcsec. An analysis of CORALIE radial-velocity measurements has revealed a variation with an amplitude of about 50m/s and a periodicity of 1.3days. This radial-velocity signal is consistent with the expected variation induced by the presence a very low mass giant planetary companion to HD 41004 A, whose light dominates the spectra. The radial-velocity measurements were then complemented with a photometric campaign and with the analysis of the bisector of the CORALIE Cross-Correlation Function (CCF). While the former revealed no significant variations within the observational precision of 0.003-0.004 mag (except for an observed flare event), the bisector analysis showed that the line profiles are varying in phase with the radial-velocity. This latter result, complemented with a series of simulations, has shown that we can explain the observations by considering that HD 41004 B has a brown-dwarf companion orbiting with the observed 1.3-day period. If confirmed, this detection represents the first discovery of a brown dwarf in a very short period (1.3-day) orbit around an M dwarf. Finally, this case should be taken as a serious warning about the importance of analyzing the bisector when looking for planets using radial-velocity techniques.

Motivation & Objective

• To investigate the origin of a 1.3-day radial-velocity signal detected in the HD 41004 AB binary system, initially interpreted as a low-mass planet around the primary K0V star.
• To determine whether the signal arises from a planetary companion or from stellar activity or instrumental effects.
• To assess the reliability of radial-velocity techniques in binary systems where spectral blending occurs.
• To evaluate the role of the bisector of the cross-correlation function (CCF) in detecting false positive signals.

Proposed method

• Obtained 86 radial-velocity measurements using the CORALIE spectrograph on the 1.2-m Euler telescope at La Silla Observatory.
• Conducted a photometric campaign using the Strömgren Automatic Telescope (SAT) to search for transit-like variations.
• Analyzed the bisector of the CORALIE cross-correlation function (CCF) to detect line profile distortions indicative of spectral blending.
• Performed simulations to model the Doppler shift of the M-dwarf component relative to the K-dwarf, showing how relative line shifts can mimic radial velocity variations.
• Used stellar parameters and radial velocity amplitudes to constrain the mass and orbital period of the unseen companion.
• Calibrated the CCF surface (W_fit) to derive [Fe/H] metallicities, enabling consistent abundance analysis across the survey.

Experimental results

Research questions

• RQ1 Can the 1.3-day radial-velocity signal in HD 41004 AB be explained by a planetary companion around the K0V primary star?
• RQ2 Is the observed radial-velocity variation due to intrinsic stellar activity or spectral blending in a binary system?
• RQ3 To what extent do variations in the bisector of the CCF indicate spectral line profile distortions caused by a moving companion in a binary system?
• RQ4 Can the observed radial velocity and bisector variations be explained by the Doppler shift of the M-dwarf companion relative to the primary?
• RQ5 What are the implications of this system for the reliability of radial-velocity planet searches in binary stars?

Key findings

• The 1.3-day radial-velocity signal with an amplitude of ~50 m s⁻¹ is not caused by a planet around HD 41004 A, but by the Doppler motion of the M-dwarf companion HD 41004 B.
• The bisector of the CCF varies in phase with the radial velocity, indicating line profile distortions due to spectral blending from the moving M-dwarf.
• Photometric observations showed no significant variations within ~0.003–0.004 mag, ruling out transits or large photometric variability.
• Simulations confirm that the relative motion of the M-dwarf (with radial velocity amplitude of a few km s⁻¹) can produce the observed radial velocity and bisector variations.
• The system hosts a brown dwarf with a mass likely above the deuterium-burning limit, orbiting the M-dwarf in a 1.3-day period, marking the first such discovery.
• The study demonstrates that failure to analyze the bisector can lead to false planet detections in binary systems, especially when one component is much fainter.