Evidence for a bloated massive protostar in IRAS20126+4104

TL;DR

This study confirms periodic mid-infrared emission in the massive protostar IRAS20126+4104, indicating a bloated star with a long rotation period, based on 19 years of multi-epoch data analysis.

Contribution

It provides the first detailed analysis of long-term mid-IR variability in a high-mass protostar, supporting the rotation model of periodic emission.

Findings

IRAS20126+4104 exhibits a ~6.8-year periodic IR emission.

Both lobes show anticorrelated emission with a phase difference of ~2.5 years.

The protostar is likely bloated with a radius of ~200 R_sun.

Abstract

Variability is a well known phenomenon in low-mass young stellar objects, but in recent years the monitoring of methanol masers and infrared continuum emission has permitted the detection of both burst-like episodes and periodic variations also in high-mass (proto)stars. Multi-epoch studies on large samples of these objects have become possible thanks to the NEOWISE database, which surveyed the sky in the mid-IR for about a decade. Our goal is to analyse the mid-IR emission from the well studied massive protostar IRAS20126+4104 and confirm the hypothesis that such emission is periodic, as proposed in previous studies. We take advantage of the NEOWISE, ALLWISE, and Spitzer databases to obtain 24 images of the 3.4 $\mu$m emission from IRAS20126+4104 spanning 19 years, with $\sim$6 months sampling over a decade. With these data we create a light curve for each lobe of the bipolar nebulosity/outflow associated with the protostar. Our results confirm that the IR emission from IRAS20126+4104 varies regularly with a period of $\sim$6.8 yr. The period is the same for both lobes, but their emissions are anticorrelated with a phase difference of $\sim$2.5 yr. The variation is consistent with that found in previous studies for the 6 GHz CH$_3$OH masers and the near-IR emission from the lobes. After discussing four possible ``clocks'' that could determine the observed periodicity, we rule out all but a model involving rotation of the star with a spot obscuring $\sim$20% of the stellar surface. The long rotation period implies that the 12 $M_\odot$ protostar is bloated, with a radius of $\sim$200 $R_\odot$.