Eclipsing time variations in close binaries produced by azimuthal dynamo waves

TL;DR

This paper proposes that azimuthal dynamo waves in stars can explain eclipsing time variations in close binaries, offering a self-consistent alternative to planetary or other mechanisms, with simulations matching observed O-C diagram features.

Contribution

It introduces a model where non-axisymmetric magnetic fields cause ETVs, demonstrating their ability to reproduce observed O-C diagram shapes and amplitudes.

Findings

ADWs produce characteristic O-C diagram shapes similar to observations.

ETV amplitudes range from tens to hundreds of seconds.

ADWs are easily excited in rapidly rotating stars, reducing energetic constraints.

Abstract

The nature of eclipsing time variations (ETVs) in post-common-envelope binaries (PCEBs) is still unknown. Circumbinary planets routinely fail the test of time and the Applegate mechanism has energetic constraints and problems in reproducing observations. Based on recent analytic models of magnetically-induced ETVs and stellar dynamo simulations, we aim at explaining ETVs via non-axisymmetric magnetic fields that drift in the azimuthal direction of the star, know as azimuthal dynamo waves (ADWs). We implement a time-varying non-axisymmetric quadrupole moment ($Q$) in a binary system. We solve for the dynamics of the system, compute the resulting eclipsing times, and construct O-C, diagrams. We perform several simulations with different amplitudes of $Q$, periods, stellar masses and binary separations. ADWs naturally give rise to characteristic shapes in the O-C diagram that resemble observations. Depending on how fast $Q$ changes, the solutions can have a sharp decrease in O-C producing amplitudes such as the one obtained in QS Vir, or sinusoidal-like shapes such as in V471 Tau or NN Ser. We also find that the amplitude of the eclipsing times varies from tens to hundreds of seconds. ADWs offer a self-consistent explanation for ETVs as they are expected in dynamo theory. They can explain a variety of features in the observed O-C diagrams. As suggested by dynamo simulations, ADWs are easily excited in rapidly rotating stars, alleviating energetic constrains required in the context of the Applegate mechanism. They produce non-axis $Q$ that in turn produce ETVs that can account for the long-term variation of the O-C, diagrams. We expect in this case that the resulting O-C diagrams are not strictly periodic, unlike explanations based on a third body that would imply a strict periodicity unless additional mechanisms are being invoked.