New predictions for radiation-driven, steady-state mass-loss and wind-momentum from hot, massive stars: II. A grid of O-type stars in the Galaxy and the Magellanic Clouds

Context. Reliable predictions of mass-loss rates are important for massive-star evolution computations. Aims. We aim to provide predictions for mass-loss rates and wind-momentum rates of O-type stars, while carefully studying the behaviour of these winds as functions of stellar parameters, such as luminosity and metallicity. Methods. We used newly developed steady-state models of radiation-driven winds to compute the global properties of a grid of O-stars. The self-consistent models were calculated by means of an iterative solution to the equation of motion using full non-local thermodynamic equilibrium radiative transfer in the co-moving frame to compute the radiative acceleration. In order to study winds in different galactic environments, the grid covers main-sequence stars, giants, and supergiants in the Galaxy and both Magellanic Clouds. Results. We find a strong dependence of mass-loss on both luminosity and metallicity. Mean values across the grid are (M) over dot similar to L-*(2.2) and (M) over dot Z(*)(0.95); however, we also find a somewhat stronger dependence on metallicity for lower luminosities. Similarly, the mass loss-luminosity relation is somewhat steeper for the Small Magellanic Cloud (SMC) than for the Galaxy. In addition, the computed rates are systematically lower (by a factor 2 and more) than those commonly used in stellar-evolution calculations. Overall, our results are in good agreement with observations in the Galaxy that properly account for wind-clumping, with empirical (M) over dot versus Z(*) scaling relations and with observations of O-dwarfs in the SMC. Conclusions. Our results provide simple fit relations for mass-loss rates and wind momenta of massive O-stars stars as functions of luminosity and metallicity, which are valid in the range T-eff = 28 000-45 000 K. Due to the systematically lower values for (M) over dot, our new models suggest that new rates might be needed in evolution simulations of massive stars.