Observational mapping of the mass discrepancy in eclipsing binaries

A new self-contained framework for concurrent analysis of photometric and spectroscopic time series^★

¹ Institute of Astronomy, KU Leuven, Celestijnenlaan 200D, B-3001 Leuven, Belgium
² Astrophysics group, Department of Physics, University of Surrey, Guildford GU2 7XH, UK
³ Max Planck Institute for Astrophysics, Karl-Schwarzschild-Straße 1, 85741 Garching, Germany
⁴ Department of Physics, Faculty of Science, University of Zagreb, 10 000 Zagreb, Croatia
⁵ Max Planck Institute for Astronomy, Königstuhl 17, 69117 Heidelberg, Germany
⁶ Department of Astrophysics, IMAPP, Radboud University Nijmegen, PO Box 9010, 6500 GL Nijmegen, The Netherlands

^★★ Corresponding author: nadya.serebriakova@kuleuven.be

Received: 24 December 2024
Accepted: 29 April 2025

Abstract

Context. The mass discrepancy problem, observed in high-mass stars within eclipsing binaries, highlights systematic differences between dynamical and evolutionary mass estimates, challenging the accuracy of stellar evolution models.

Aims. We aim to determine whether analysis methods directly contribute to this discrepancy and to assess how methodological improvements might reduce or clarify it.

Methods. To address this, we developed a new self-contained framework that simultaneously models the photometric and spectroscopic data, minimising biases introduced by traditional iterative approaches and enabling consistent parameter optimisation.

Results. We present this framework alongside validation tests on synthetic data and demonstrate its application to three well-studied observed binaries, including one system known for its pronounced mass discrepancy. The framework recovers multiple viable solutions from distinct local minima, including one that reduces the mass discrepancy.

Conclusions. These results illustrate how methodological biases, rather than evolutionary model assumptions, can contribute to the mass discrepancy problem. We further highlight that external constraints, such as independent distance estimates or evolutionary models, may be necessary to distinguish between degenerate solutions. Expanding this analysis to a larger sample will provide a more complete understanding, with forthcoming results in the next paper in this series.