[Paper Review] The VLT-FLAMES Tarantula Survey XXII. Multiplicity properties of the B-type stars

P. R. Dunstall, P. L. Dufton+8 more|UvA-DARE (University of Amsterdam)|May 26, 2015

Stellar, planetary, and galactic studiesPhysics and Astronomy35 references73 citations

TL;DR

This study analyzes multi-epoch spectroscopy of 408 B-type stars in the 30 Doradus region using the VLT-FLAMES Tarantula Survey to determine their spectroscopic binary properties. It finds an observed binary fraction of 25% ± 2%, with intrinsic multiplicity of 58% ± 11% after correcting for observational biases, indicating that B-type stars in 30 Dor have multiplicity properties consistent with O-type stars, implying widespread binary interaction among massive star progenitors.

ABSTRACT

We investigate the multiplicity properties of 408 B-type stars observed in the 30 Doradus region of the Large Magellanic Cloud with multi-epoch spectroscopy from the VLT-FLAMES Tarantula Survey (VFTS). We use a cross-correlation method to estimate relative radial velocities from the helium and metal absorption lines for each of our targets. Objects with significant radial-velocity variations (and with an amplitude larger than 16 km/s) are classified as spectroscopic binaries. We find an observed spectroscopic binary fraction (defined by periods of <10^3.5 d and mass ratios >0.1) for the B-type stars, f_B(obs) = 0.25 +/- 0.02, which appears constant across the field of view, except for the two older clusters (Hodge 301 and SL 639). These two clusters have significantly lower fractions of 0.08 +/- 0.08 and 0.10 +/- 0.09, respectively. Using synthetic populations and a model of our observed epochs and their potential biases, we constrain the intrinsic multiplicity properties of the dwarf and giant (i.e. relatively unevolved) B-type stars in 30 Dor. We obtain a present-day binary fraction f_B(true) = 0.58 +/- 0.11, with a flat period distribution. Within the uncertainties, the multiplicity properties of the B-type stars agree with those for the O stars in 30 Dor from the VFTS.

Motivation & Objective

• To determine the observed spectroscopic binary fraction of B-type stars in the 30 Doradus region using multi-epoch spectroscopy.
• To correct for observational biases in detecting radial velocity variations to estimate the intrinsic binary fraction of unevolved B-type stars.
• To compare the multiplicity properties of B-type stars with those of O-type stars in the same region to assess the role of binarity in massive star evolution.
• To constrain the period and mass-ratio distributions of B-type binaries in 30 Doradus using synthetic population modeling.
• To assess the implications of binary interaction for the evolutionary pathways and supernova progenitor populations of massive stars.

Proposed method

• Cross-correlation of helium and metal absorption lines in multi-epoch spectra to measure relative radial velocities for each B-type star.
• Identification of spectroscopic binaries based on radial velocity variations exceeding 16 km s⁻¹, consistent with orbital motion.
• Application of a Monte Carlo synthetic population modeling approach to simulate detection probabilities and correct for observational biases.
• Use of a global merit function to jointly constrain the intrinsic binary fraction, period distribution (parameterized by π), and mass-ratio distribution (parameterized by κ).
• Estimation of detection probability for B-type binaries in the VFTS survey as 40% ± 10% based on model comparisons.
• Comparison of observed and simulated cumulative distributions of RV amplitudes and variability timescales to validate model parameters.

Experimental results

Research questions

• RQ1 What is the observed spectroscopic binary fraction of B-type stars in the 30 Doradus region based on multi-epoch VLT-FLAMES data?
• RQ2 What is the intrinsic binary fraction of unevolved B-type stars in 30 Doradus after correcting for observational biases?
• RQ3 How do the period and mass-ratio distributions of B-type binaries in 30 Doradus compare to those of O-type stars in the same region?
• RQ4 To what extent do the multiplicity properties of B-type stars in 30 Doradus align with those of O-type stars, and what does this imply for massive star evolution?
• RQ5 What fraction of stars currently classified as single are likely to be unresolved binaries due to detection limitations?

Key findings

• The observed spectroscopic binary fraction for B-type stars in 30 Doradus is 25% ± 2%, with a significant drop in the two older clusters Hodge 301 and SL 639 (0.08 ± 0.08 and 0.10 ± 0.09, respectively).
• After correcting for observational biases, the intrinsic binary fraction of unevolved B-type stars is estimated to be 58% ± 11%, indicating a high prevalence of binary systems at birth.
• The period distribution of B-type binaries is consistent with a flat distribution, with no strong evidence for a preferred period range.
• The mass-ratio distribution is poorly constrained, with a best-fit exponent κ ≈ -0.5 dex, but uncertainties are large and systematics may bias the result.
• The detection probability for B-type binaries in the VFTS survey is estimated at 40% ± 10%, implying that nearly half of the binaries may remain undetected.
• The multiplicity properties of B-type stars are consistent with those of O-type stars in 30 Doradus, suggesting that binary interaction affects a large fraction of massive star progenitors.