New constraints on dust emission and UV attenuation of z=6.5-7.5 galaxies from millimeter observations

We have targeted two recently discovered Lyman break galaxies (LBGs) to search for dust continuum and [CII] 158 micron line emission. The strongly lensed z~6.8 LBG A1703-zD1 behind the galaxy cluster Abell 1703, and the spectroscopically confirmed z=7.508 LBG z8-GND-5296 in the GOODS-N field have been observed with the Plateau de Bure interferometer (PdBI) at 1.2mm. These observations have been combined with those of three z>6.5 Lya emitters (named HCM6A, Himiko, and IOK-1), for which deep measurements were recently obtained with the PdBI and ALMA. [CII] is undetected in both galaxies, providing a deep upper limit for Abell1703-zD1, comparable to recent ALMA non-detections. Dust continuum emission from Abell1703-zD1 and z8-GND-5296 is not detected with an rms of 0.12 and 0.16 mJy/beam. From these non-detections we derive upper limits on their IR luminosity and star formation rate, dust mass, and UV attenuation. Thanks to strong gravitational lensing the limit for Abell1703-zD1 is probing the sub-LIRG regime ($L_{IR} <8.1 \times 10^{10}$ Lsun) and very low dust masses ($M_d<1.6 \times 10^7$ Msun). We find that all five galaxies are compatible with the Calzetti IRX-$\beta$ relation, their UV attenuation is compatible with several indirect estimates from other methods (the UV slope, extrapolation of the attenuation measured from the IR/UV ratio at lower redshift, and SED fits), and the dust-to-stellar mass ratio is not incompatible with that of galaxies from z=0 to 3. For their stellar mass the high-z galaxies studied here have an attenuation below the one expected from the mean relation of low redshift (z<1.5) galaxies. More and deeper (sub)-mm data are clearly needed to directly determine the UV attenuation and dust content of the dominant population of high-z star-forming galaxies and to establish more firmly their dependence on stellar mass, redshift, and other properties.