Assessment of evolutionary status of eclipsing binaries using light-curve parameters and spectral classification

TL;DR

This paper presents a new, efficient method for classifying eclipsing binary stars based on light-curve data and spectral types, enabling large-scale analysis of stellar evolution.

Contribution

The authors developed and validated a fast, flexible classification procedure applicable to large surveys, identifying rare and unclassified binary systems.

Findings

Successfully classified over 4700 binaries with unknown types.

Detected rare evolutionary classes and systems with unusual parameters.

Identified evolutionary classes for 50 previously unclassified cluster binaries.

Abstract

We have developed a procedure for the classification of eclipsing binaries from their light-curve parameters and spectral type. The procedure was tested on more than 1000 systems with known classification, and its efficiency was estimated for every evolutionary status we use. The procedure was applied to about 4700 binaries with no classification, and the vast majority of them was classified successfully. Systems of relatively rare evolutionary classes were detected in that process, as well as systems with unusual and/or contradictory parameters. Also, for 50 previously unclassified cluster binaries evolutionary classes were identified. These stars can serve as tracers for age and distance estimation of their parent stellar systems. The procedure proved itself as fast, flexible and effective enough to be applied to large ground based and space born surveys, containing tens of thousands of eclipsing binaries.