Far-infrared emission of massive stars

TL;DR

This study analyzes the far-infrared excess emission in a sample of massive stars, revealing that half exhibit excess likely due to ionized gas or circumstellar dust, with detailed spectral and imaging data supporting these findings.

Contribution

It provides a comprehensive multi-wavelength analysis of far-IR excess in massive stars, identifying its possible origins as ionized gas or circumstellar dust, and rules out scattered-light disks in some cases.

Findings

Half of the studied stars show far-IR excess.

Far-IR excess correlates with stronger Pfund and Humphrey lines.

Excess is explained by free-free emission or large dust shells.

Abstract

We present results of the analysis of a sample of 22 stars of spectral types from O7 to B5 and luminosity classes I-V for which Spitzer/IRS spectra are available. The IRS spectra of these stars are examined for signs of excess infrared (IR) emission by comparison with stellar atmospheric spectra. We find that the spectra of half of the studied stars are dominated by excess emission in the far-IR, including all six super- and bright giants. In order to examine the origin of the far-IR excess, we supplement the Spitzer data with optical high-resolution echelle spectroscopy ($\lambda/\Delta \lambda \sim 10^5$), near-IR high-contrast coronagraphic imaging taken with the SPHERE instrument at VLT with a spatial resolution of 0.05", and WISE and Herschel photometry. In the optical region, we detect various absorption and emission lines (H$\alpha$, CIII, and NIII) irrespective of the far-IR excess. Pfund($\alpha$) and Humphrey($\alpha$) lines are observed at the same time as the far-IR excess. These lines are stronger in stars with far-IR excess than in stars without excess. A scattered-light disk in the central r < 2.5" region of the far-IR excess stars HD149404, HD151804, and HD154368 can be excluded from H band imaging down to a 1$\sigma$ contrast of $F(r)/F_{*} \sim 10^{-6}$. The far-IR excess is fit either by a free-free component from ionized gas as for the winds of hot stars or a large (1pc) circumstellar dust shell. The putative dust envelopes required to explain the excess have a visual extinction as low as a few hundred $\mu$-mag.