The not-so-massive black hole in the microquasar GRS1915+105

We present a new dynamical study of the black hole X-ray transient GRS1915+105 making use of near-infrared spectroscopy obtained with X-shooter at the VLT. We detect a large number of donor star absorption features across a wide range of wavelengths spanning the H and K bands. Our 24 epochs covering a baseline of over 1 year permit us to determine a new binary ephemeris including a refined orbital period of P=33.85 +/- 0.16 d. The donor star radial velocity curves deliver a significantly improved determination of the donor semi-amplitude which is both accurate (K_2=126 +/- 1 km/s) and robust against choice of donor star template and spectral features used. We furthermore constrain the donor stars rotational broadening to vsini=21 +/-4 km/s, delivering a binary mass ratio of q=0.042 +/- 0.024. If we combine these new constraints with distance and inclination estimates derived from modelling the radio emission, a black hole mass of M_BH=10.1 +/- 0.6 M_sun is inferred, paired with an evolved mass donor of M_2=0.47 +/- 0.27 M_sun. Our analysis suggests a more typical black hole mass for GRS1915+105 rather than the unusually high values derived in the pioneering dynamical study by Greiner et al. (2001). Our data demonstrate that high-resolution infrared spectroscopy of obscured accreting binaries can deliver dynamical mass determinations with a precision on par with optical studies.

Dynamical constraints on the component masses of the cataclysmic variable WZ Sge

We present phase-resolved spectroscopy of the short period cataclysmic variable WZ Sge obtained with the Hubble Space Telescope. We were able to resolve the orbital motion of a number of absorption lines that likely probe the environment near the accreting white dwarf. The radial velocities derived from simultaneous fits to 13 absorption lines indicate an orbital velocity semi-amplitude of K_UV = 47 +/- 3 km/s. However, we find that the phase zero is offset from the white dwarf ephemeris by +0.1. Our offset and velocity amplitude are very similar to constraints derived from optical emission lines from the quiescent accretion disk, despite the fact that we are probing material much closer to the primary. If we associate the UV amplitude with K_1, our dynamical constraints together with the K_2 estimates from Steeghs et al. (2001) and the known binary inclination of i=77+/-2 imply 0.88<M_1<1.53 M_sun, 0.078 < M_2 < 0.13 M_sun and 0.075<q=M_2/M_1<0.101. If we interpret the mean velocity of the UV lines (-16+/-4 km/s) as being due to the gravitational red-shift caused in the high-g environment near the white dwarf, we find v_grav=56+/-5 km/s which provides an independent estimate on the mass of the primary of M_1=0.85+/-0.04 M_sun when coupled with a mass-radius relation. Our primary mass estimates are in excellent agreement and are also self-consistent with spectrophotometric fits to the UV fluxes despite the observed phase offset. It is at this point unclear what causes the observed phase-offset in the UV spectra and by how much it distorts the radial velocity signature from the underlying white dwarf.