CO investigation of

We report on the results of an IRAM-30 m search for CO emission lines in three galaxies at intermediate redshifts. The idea was to investigate the molecular content of galaxies that are bright in the infrared at $z=0.4{-}1.5$, a redshift desert for molecular line studies because poorly investigated as yet. We integrated 8-10 h per source and did not succeed in detecting any of the sources. From our upper limits, we were able to constrain the molecular gas content in these systems to less than 4 to 8 $\times$ $10^9~M_\odot$, assuming a CO-to-H2 conversion factor ($\alpha=0.8~M_\odot$/(K km s-1 pc2)). We stress the current difficulty of selecting sources with a detectable molecular content, a problem that will be faced by the ALMA First Science projects