A new measurement of the expansion history of the Universe at z=1.26 with cosmic chronometers in VANDELS

We derive a new constraint on the expansion history of the Universe by applying the cosmic chronometers method, studying the age evolution of high-redshift galaxies with a full-spectral-fitting approach. We select a sample of 39 massive ($log(M/M_\odot)>10.8$) and passive ($log(sSFR/yr^{-1})<-11$) galaxies from the data release 4 of the VANDELS survey at $1<z<1.5$, combining different selection criteria to minimize the potential contamination by star-forming outliers. We perform full-spectral-fitting jointly on spectra and photometry of our sources with the code BAGPIPES, without any cosmological assumption on the age of the population. The derived physical properties of the selected galaxies are characteristic of a passive population, with short star formation timescales ($<\tau>=0.28\pm0.02$ Gyr), low dust extinction ($<A_{V,dust}>=0.43\pm0.02$ mag), and sub-solar metallicities ($<Z/Z_{\odot}>=0.44\pm0.01$). The ages show a decreasing trend with redshift compatible with a standard cosmological model, even if no cosmological constraint is assumed in the fit, and a clear mass-downsizing pattern. Testing the impact of the star formation history on the results, we find only a maximum 2\% fluctuation in age and metallicity. By fitting the median age-redshift relation with a flat $\Lambda$CDM model and assuming a Gaussian prior on $\Omega_{M,0}= 0.3\pm0.02$ from late-Universe probes, we obtain $H_0=67_{-15}^{+14}\:km\:s^{-1}\:Mpc^{-1}$. In the end, we derive a new estimate of the Hubble parameter with the cosmic chronometers method, $H(z=1.26)=135\pm65\:km\:s^{-1}\:Mpc^{-1}$ including statistical and systematic errors. While the error budget is currently dominated by the scarcity of the sample, this work proves the potential strength of the cosmic chronometers approach up to $z>1$, especially in view of incoming large spectroscopic surveys like Euclid. (abridged)