Testing the theory of colliding winds: The periastron passage of 9 Sagittarii

Context.
Colliding winds in massive-star binaries generate non-thermal radio emission and an increased X-ray emission.

Aims.
We study the radio emission of the non-thermal emitter 9 Sgr, which is a long-period and highly eccentric binary.

Methods.
We conducted a monitoring campaign of 9 Sgr around its 2013–2014 periastron passage, obtaining radio, X-ray, and optical spectroscopy data. In this paper, we analyse the radio data.

Results.
The radio data at 3.6, 6, and 20 cm show a maximum near periastron. Other non-thermal radio emitters among O-type binaries have a minimum flux near periastron, because the synchrotron emission is absorbed by the free-free absorption in the two stellar winds. This is not the case for 9 Sgr, because even at periastron the orbital separation is sufficient to maintain the wind-wind collision and its associated synchrotron emission outside the effective free-free radius of both stars. Our standard modelling with a local magnetic field that decreases inversely proportional to the distance from the stars (1/
r
) fails to explain the observed fluxes. Instead, we need to adopt a local magnetic field that decreases faster than 1/
r
.

Conclusions.
While our model provides an acceptably good fit to the data, there is clearly still room for improvement. More advanced modelling should include magnetohydrodynamics and more detailed physics of the injection, acceleration, and cooling of the relativistic electrons. Radio observations at orbital phases away from periastron can put further constraints on the models.