Context. The H i halo clouds of the Milky Way, and in particular the intermediate-velocity clouds (IVCs), are thought to be connected to Galactic fountain processes. Observations of fountain clouds are important for understanding the role of matter recycling and accretion onto the Galactic disk and subsequent star formation. Aims. Here, we quantify the amount of molecular gas in the Galactic halo. We focus on the rare class of molecular IVCs (MIVCs) and search for new objects. Methods. The H i-FIR correlation was studied across the entire northern and southern Galactic hemispheres at Galactic latitudes |b| > 20• to determine the amount and distribution of molecular gas in IVCs. We used the most recent large-scale H i and FIR data, the Effelsberg Bonn-H i Survey, the Parkes Galactic All-Sky Survey, and the Planck FIR surveys.Results. We present a catalogue of 239 MIVC candidates on the northern and southern Galactic hemispheres. Among these candidates, all previously known MIVCs are recovered except for one single source. The frequency of candidates differs significantly between the northern and southern Galactic hemispheres and between negative and positive LSR velocities as well.Conclusions. In our approach we analyse the local Galactic environment. Extrapolating our results to the entire Galaxy, the global inflow of atomic and molecular IVC gas onto the Milky Way may account for the major fraction of the gaseous mass that is required to sustain the current Galactic star formation rate.

Context. Towards the high galactic latitude sky, the far-infrared (FIR) intensity is tightly correlated to the total hydrogen column density which is made up of atomic (H ) and molecular hydrogen (H 2 ). Above a certain column density threshold, atomic hydrogen turns molecular. Aims. We analyse gas and dust properties of intermediate-velocity clouds (IVCs) in the lower galactic halo to explore their transition from the atomic to the molecular phase. Driven by observations, we investigate the physical processes that transform a purely atomic IVC into a molecular one. Methods. Data from the Effelsberg-Bonn H -Survey (EBHIS) are correlated to FIR wavebands of the Planck satellite and IRIS. Modified black-body emission spectra are fitted to deduce dust optical depths and grain temperatures. We remove the contribution of atomic hydrogen to the FIR intensity to estimate molecular hydrogen column densities. Results. Two IVCs show different FIR properties, despite their similarity in H , such as narrow spectral lines and large column densities. One FIR bright IVC is associated with H 2 , confirmed by 12 CO (1 → 0) emission; the other IVC is FIR dim and shows no FIR excess, which indicates the absence of molecular hydrogen. Conclusions. We propose that the FIR dim and bright IVCs probe the transition between the atomic and molecular gas phase. Triggered by dynamical processes, this transition happens during the descent of IVCs onto the galactic disk. The most natural driver is ram pressure exerted onto the cloud by the increasing halo density. Because of the enhanced pressure, the formation timescale of H 2 is reduced, allowing the formation of large amounts of H 2 within a few Myr.

Context. Intermediate-velocity clouds (IVCs) are HI halo clouds that are likely related to a Galactic fountain process. In-falling IVCs are candidates for the re-accretion of matter onto the Milky Way. Aims. We study the evolution of IVCs at the disk-halo interface, focussing on the transition from atomic to molecular IVCs. We compare an atomic IVC to a molecular IVC and characterise their structural differences in order to investigate how molecular IVCs form high above the Galactic plane. Methods. With high-resolution HI observations of the Westerbork Synthesis Radio Telescope and 12 CO(1 → 0) and 13 CO(1 → 0) observations with the IRAM 30 m telescope, we analyse the small-scale structures within the two clouds. By correlating HI and far-infrared (FIR) dust continuum emission from the Planck satellite, the distribution of molecular hydrogen (H 2 ) is estimated. We conduct a detailed comparison of the HI, FIR, and CO data and study variations of the X CO conversion factor. Results. The atomic IVC does not disclose detectable CO emission. The atomic small-scale structure, as revealed by the highresolution HI data, shows low peak HI column densities and low HI fluxes as compared to the molecular IVC. The molecular IVC exhibits a rich molecular structure and most of the CO emission is observed at the eastern edge of the cloud. There is observational evidence that the molecular IVC is in a transient and, thus, non-equilibrium phase. The average X CO factor is close to the canonical value of the Milky Way disk. Conclusions. We propose that the two IVCs represent different states in a gradual transition from atomic to molecular clouds. The molecular IVC appears to be more condensed allowing the formation of H 2 and CO in shielded regions all over the cloud. Ram pressure may accumulate gas and thus facilitate the formation of H 2 . We show evidence that the atomic IVC will evolve also into a molecular IVC in a few Myr.