Apsidal motion of the eclipsing binary AS Camelopardalis: discrepancy resolved

Abstract

We present a spectroscopic study of the eclipsing binary system AS Camelopardalis, the first such study based on phase-resolved CCD echelle spectra. Via a spectral disentangling analysis we measure the minimum masses of the stars to be M_A sin^3 i = 3.213 +/- 0.007 M_sun and M_B sin^3 i = 2.323 +/- 0.006 M_sun, their effective temperatures to be T_eff(A) = 12840 +/- 120 K and T_eff(B) = 10580 +/- 240 K, and their projected rotational velocities to be v_A sin i_A = 14.5 +/- 0.1 km/s and v_B sin i_B = 4.6 +/- 0.1 km/s. These projected rotational velocities appear to be much lower than the synchronous values. We show that measurements of the apsidal motion of the system suffer from a degeneracy between orbital eccentricity and apsidal motion rate. We use our spectroscopically-measured e = 0.164 +/- 0.001 to break this degeneracy and measure (d omega_obs / dt) = 0.133 +/- 0.010 deg/yr. Subtracting the relativistic contribution of (d omega_GR / dt) = 0.0963 +/- 0.0002 deg/yr yields the contribution due to tidal torques: (d omega_cl / dt) = 0.037 +/- 0.010 deg/yr. This value is much smaller than the rate predicted by stellar theory, 0.40--0.87 deg/yr. We interpret this as a misalignment between the orbital axis of the close binary and the rotational axes of its component stars, which also explains their apparently low rotational velocities. The observed and predicted apsidal motion rates could be brought into agreement if the stars were rotating three times faster than synchronous about axes perpendicular to the orbital axis. Measurement of the Rossiter-McLaughlin effect can be used to confirm this interpretation.