1997A&A...327.1054B

We have performed evolutionary calculations of very-low-mass stars from 0.08 to 0.8M_☉ for different metallicites from [M/H]=-2.0 to -1.0 and we have tabulated the mechanical, thermal and photometric characteristics of these models. The calculations include the most recent interior physics and improved non-grey atmosphere models. The models reproduce the entire main sequences of the globular clusters observed with the Hubble Space Telescope over the afore-mentioned range of metallicity. Comparisons are made in the WFPC2 Flight system including the F555, F606 and F814 filters, and in the standard Johnson-Cousins system. We examine the effects of different physical parameters, mixing-length, α-enriched elements, helium fraction, as well as the accuracy of the photometric transformations of the HST data into standard systems. We derive mass-effective temperature and mass-magnitude relationships and we compare the results with the ones obtained with different grey-like approximations. These latter are shown to yield inaccurate relations, in particular near the hydrogen-burning limit. We derive new hydrogen-burning minimum masses, and the corresponding absolute magnitudes, for the different metallicities. We predict color-magnitude diagrams in the infrared NICMOS filters, to be used for the next generation of the HST observations, providing mass-magnitudes relationships in these colors down to the brown-dwarf limit. We show that the expected signature of the stellar to substellar transition in color-magnitude diagrams is a severe blueshift in the infrared colors, due to the increasing collision-induced absorption of molecular hydrogen with increasing density and decreasing temperature. At last, we apply these calculations to the observed halo field stars, which yields a precise determination of their metallicity, and thus of their galactic origin. We find no evidence for significant differences between the halo field stars and the globular cluster sequences.