2004A&A...415..971D

Very deep images of the Galactic globular cluster M4 (NGC6121) through the F606W and F814W filters were taken in 2001 with the WFPC2 on board the HST. A first published analysis of this data set (Richer et al., 2002ApJ...574L.151R) produced the result that the age of M4 is 12.7±0.7Gyr (Hansen et al., 2002ApJ...574L.155H), thus setting a robust lower limit to the age of the universe. In view of the great astronomical importance of getting this number right, we have subjected the same data set to the simplest possible photometric analysis that completely avoids uncertain assumptions about the origin of the detected sources. This analysis clearly reveals both a thin main sequence, from which can be deduced the deepest statistically complete mass function yet determined for a globular cluster, and a white dwarf (WD) sequence extending all the way down to the 5σ detection limit at I≃27. The WD sequence is abruptly terminated at exactly this limit as expected by detection statistics. Using our most recent theoretical WD models (Prada Moroni & Straniero, 2002ApJ...581..585P) to obtain the expected WD sequence for different ages in the observed bandpasses, we find that the data so far obtained do not reach the peak of the WD luminosity function, thus only allowing one to set a lower limit to the age of M4 of ∼9Gyr. Thus, the problem of determining the absolute age of a globular cluster and, therefore, the onset of GC formation with cosmologically significant accuracy remains completely open. Only observations several magnitudes deeper than the limit obtained so far would allow one to approach this objective.