Constraining the overcontact phase in massive binary evolution: III. Period stability of known B+B and O+B overcontact systems

Context. Binary systems play a crucial role in massive star evolution. Systems composed of B-type and O-type stars are of particular interest due to their potential to lead to very energetic phenomena or the merging of exotic compact objects. Aims. We aim to determine the orbital period variations of a sample of B+B and O+B massive overcontact binaries, with the primary objectives of characterizing the evolutionary timescales of these systems and addressing the existing discrepancy between observational data and theoretical predictions derived from population synthesis models. Methods. We used PERIOD04 to analyze archival photometric data going back a century for a sample of seven binary systems to measure their orbital periods. We then determine the period variations using a linear fit. Results. We find that the period variation timescales of five truly overcontact binary systems align with the nuclear timescale, in agreement with previous findings for more massive overcontact binaries. Additionally, we noticed a clear distinction between the five systems that had been unambiguously classified as overcontact systems and both SV Cen and VFTS 066, which seem to be evolving on thermal timescales and might be misclassified as overcontact systems. Conclusions. In the case of the five overcontact binaries, our results indicate a noticeable mismatch between the observational data and the theoretical predictions derived from population synthesis models. Furthermore, our results suggest that additional physical mechanisms must be investigated to compare the observed variations more thoroughly with theoretical predictions.