Abstract
Cepheid masses are a fundamental parameter for studying the pulsation and evolution of intermediate-mass stars, however the determination of their values is a long-standing problem since decades. Most of Cepheid masses are derived using stellar evolution or pulsation models, but they differ by 10-20 %. Binary Cepheids offer the unique opportunity to make progress in resolving this mass discrepancy, but so far, only a few Cepheids have a dynamical measurement of their mass. The first problem in studying binary Cepheids is that the pulsating star outshines its hot main-sequence companion at wavelengths longer than 0.5 {μm}. In addition, because of the close orbit of the companions (< 40 mas), the system cannot be spatially resolved with a single-dish 10 m-class telescope. We are using long-baseline interferometry as a powerful tool to reach a high dynamic range (∼ 1 % in H) and high spatial resolution. The astrometric position of the Cepheid companions can be measured at several orbital epochs, and then later combined with spectroscopic measurements to derive all the orbital elements, the mass and the distance of the Cepheids.
