The far-infrared fine-structure line [C ii] at 1900.5 GHz is known to be one of the brightest cooling lines in local galaxies, and therefore it has been suggested to be an efficient tracer for star formation in very high redshift galaxies. However, recent results for galaxies at z > 6 have yielded numerous non-detections in star-forming galaxies, except for quasars and submillimetre galaxies. We report the results of ALMA observations of two lensed, star-forming galaxies at z = 6.029 and z = 6.703. The galaxy A383-5.1 (star formation rate [SFR] of 3.2 M⊙ yr−1 and magnification of μ = 11.4 ± 1.9) shows a line detection with $L_{\rm [C\,\small {II}]} = 8.9\times 10^{6}$ L⊙, making it the lowest $L_{\rm [C\,\small {II}]}$ detection at z > 6. For MS0451-H (SFR = 0.4 M⊙ yr−1 and μ = 100 ± 20) we provide an upper limit of $L_{\rm [C\,\small {II}]} < 3\times 10^{5}$ L⊙, which is 1 dex below the local SFR–$L_{\rm [C\,\small {II}]}$ relations. The results are consistent with predictions for low-metallicity galaxies at z > 6; however, other effects could also play a role in terms of decreasing L[CII]. The detection of A383-5.1 is encouraging and suggests that detections are possible, but much fainter than initially predicted.

We present a new algorithm for stacking radio interferometric data in the uv-domain. The performance of uv-stacking is compared to the stacking of fully imaged data using simulated Atacama Large Millimeter/sub-millimeter Array (ALMA) and the Karl G. Jansky Very Large Array (VLA) deep extragalactic surveys. We find that image-and uv-stacking produce similar results, however, uv-stacking is typically the more robust method. An advantage of the uv-stacking algorithm is the availability of uv-data post stacking, which makes it possible to identify and remove problematic baselines. For deep VLA surveys uv-stacking yields a signal-to-noise ratio that is up to 20 per cent higher than image-stacking. Furthermore, we have investigated stacking of resolved sources with a simulated VLA data set where 1.5 ′′ (10-12 kpc at z ∼ 1 − 4 ) sources are stacked. We find that uv-stacking, where a model is fitted directly to the visibilities, significantly improves the accuracy and robustness of the size estimates. While scientific motivation for this work is studying faint, high-z galaxies, the algorithm analysed here would also be applicable in other fields of astronomy. Stacking of radio interferometric data is also expected to play a big role for future surveys with telescopes such as LOFAR and Square Kilometre Array (SKA). 1 Using a standard cosmology with H 0 = 67.3 km s −1 Mpc −1 , Ω Λ = 0.685 and Ωm = 0.315 (Planck Collaboration et al. 2014).

We study the sub-mm properties of color-selected galaxies via a stacking analysis applied for the first time to interferometric data at sub-mm wavelengths. We base our study on 344 GHz ALMA continuum observations of ∼ 20 ′′ -wide fields centered on 86 sub-mm sources detected in the LABOCA Extended Chandra Deep Field South Sub-mm Survey (LESS). We select various classes of galaxies (K-selected, star-forming sBzK galaxies, extremely red objects and distant red galaxies) according to their optical/NIR fluxes. We find clear, > 10-σ detections in the stacked images of all these galaxy classes. We include in our stacking analysis Herschel/SPIRE data to constrain the dust SED of these galaxies. We find that their dust emission is well described by a modified black body with T dust ≈ 30 K and β = 1.6 and IR luminosities of (5 − 11) × 10 11 L ⊙ , or implied star formation rates of 75-140 M ⊙ yr −1 . We compare our results with those of previous studies based on single-dish observations at 870 µm and find that our flux densities are a factor 2-3 higher than previous estimates. The discrepancy is observed also after removing sources individually detected in ALESS maps. We report a similar discrepancy by repeating our analysis on 1.4 GHz observations of the whole ECDFS. Hence we find tentative evidence that galaxies that are associated in projected and redshift space with submm bright sources are brighter than the average population. Finally, we put our findings in the context of the cosmic star formation rate density as a function of redshift.

Studying molecular gas properties in merging galaxies gives us important clues to the onset and evolution of interaction-triggered starbursts. NGC 4194 (the Medusa merger) is particularly interesting to study, since its FIR-to-CO luminosity ratio rivals that of ultraluminous galaxies (ULIRGs), despite its lower luminosity compared to ULIRGs, which indicates a high star formation efficiency (SFE) that is relative to even most spirals and ULIRGs. We study the molecular medium at an angular resolution of 0.65 × 0.52 (∼120 × 98 pc) through our observations of 12 CO 2−1 emission using the Submillimeter Array (SMA). We compare our 12 CO 2−1 maps with the optical Hubble Space Telescope and high angular resolution radio continuum images to study the relationship between molecular gas and the other components of the starburst region. The molecular gas is tracing the complicated dust lane structure of NGC 4194 with the brightest emission being located in an off-nuclear ring-like structure with ∼320 pc radius, the Eye of the Medusa. The bulk CO emission of the ring is found south of the kinematical center of NGC 4194. The northern tip of the ring is associated with the galaxy nucleus, where the radio continuum has its peak. Large velocity widths associated with the radio nucleus support the notion of NGC 4194 hosting an active galactic nucleus. A prominent, secondary emission maximum in the radio continuum is located inside the molecular ring. This suggests that the morphology of the ring is partially influenced by massive supernova explosions. From the combined evidence, we propose that the Eye of the Medusa contains a shell of swept up material where we identify a number of giant molecular associations. We propose that the Eye may be the site of an efficient starburst of 5−7 M yr −1 , but it would still constitute only a fraction of the 30−50 M yr −1 star formation rate of the Medusa. Furthermore, we find that ∼50% of the molecular mass of NGC 4194 is found in extended filamentary-like structures tracing the minor and major axis dust lanes. We suggest that molecular gas is transported along these lanes, providing the central starburst region with fuel. Interestingly, a comparison with locations of super star clusters (SSCs) reveal that the molecular gas and the SSCs are not co-spatial.