Constraints on the common-envelope evolution process from wide triple systems

TL;DR

This paper introduces a novel method to constrain the common-envelope mass-loss timescale in binary systems by analyzing the stability of wide triple systems, providing insights into the CE evolution process.

Contribution

The study proposes using wide triple systems to measure CE mass-loss timescales, offering a new observational constraint on CE models.

Findings

Mass-loss timescales of 10^3-10^5 years are most consistent with observed systems.

Longer timescales challenge traditional dynamical CE models.

Results suggest dust-driven winds may explain longer CE ejection timescales.

Abstract

Common envelope (CE) is an important phase in the evolution of interacting evolved binary systems. The interaction of the binary components during the CE evolution (CEE) stage gives rise to orbital inspiral and the formation of a short-period binary or a merger, on the expense of extending and/or ejecting the envelope. CEE is not well understood, as hydrodynamical simulations show that only a fraction of the CE-mass is ejected during the dynamical inspiral, in contrast with observations of post-CE binaries. Different CE models suggest different timescales are involved in the CE-ejection, and hence a measurement of the CE-ejection timescale could provide direct constraints on the CEE-process. Here we propose a novel method for constraining the mass-loss timescale of the CE, using post-CE binaries which are part of wide-orbit triple systems. The orbit/existence of a third companion constrains the CE mass-loss timescale, since rapid CE mass-loss may disrupt the triple system, while slower CE mass-loss may change the orbit of the third companion without disrupting it. As first test-cases we examine two observed post-CE binaries in wide triples, Wolf-1130 and GD-319. We follow their evolution due to mass-loss using analytic and numerical tools, and consider different mass-loss functions. We calculate a wide grid of binary parameters and mass-loss timescales in order to determine the most probable mass-loss timescale leading to the observed properties of the systems. We find that mass-loss timescales of the order of $10^{3}-10^{5}{\rm yr}$ are the most likely to explain these systems. Such long timescales are in tension with most of the CE mass-loss models, which predict shorter, dynamical timescales, but are potentially consistent with the longer timescales expected from the dust-driven winds model for CE ejection.