The interstellar clouds are not observable in depth with optical telescopes as they are full of dust which absorbs all the visible light. However, these clouds are interesting to study because they are the birth places of the future stars and future planets around these baby-stars. These clouds are made of two components, the dust grains discussed here and the gas, mostly hydrogen (as a molecule, H2) but also numerous other molecules, such as the carbon monoxyde (CO), water (H2O), ammonia (NH3), etc. The astronomers have developped a whole range of instruments to be able to study the inner parts of these clouds to understand the birth of stars and of future planetary systems. This instrumentation covers all the infra-red and radio domains. The infra-red being blocked by the terrestrial atmosphere (the famous greenhouse effect is directly related to it), observatories have to be put into space. It is there that the latest of the infra-red satellites, Spitzer, scrutinize with the best present sensitivity the interstellar clouds in the mid- (3 to 8 µm) and far- (24 to 160 µm) infra-red.
Traditionally, to study the dust clouds, one observes either their far-infra-red emission or their near- or mid-infra-red absorption of the star light (the effect is clearly seen at 8 µm - see Fig. 1 - where the absorption is very strong due to the silicates which are widespread in interstellar grains). In the case of L183, a cloud which the team knows quite well, a diffuse emission, clearly visible in the images at 3.6 and 4.5 µm (Fig. 1) has been identified for the first time. This emission which comes from the inside of the cloud (compare the difference with the image in I band, in the very near-infra-red, Fig. 2), clearly follows the densest part of the cloud (the one seen as a dark patch at 8 µm or in emission when observing the N2H+ ion, unknown on Earth, which we presented here].
This emission cannot originate neither from large aromatic molecules (PAHs or Polycyclic Aromatic Hydrocarbons) famous for their emission in the mid-infra-red nor from hot grains (at least 1000 K !). Indeed, PAHs have a clear spectral signature which has components in the 3.6 µm band but are not supposed to emit in the 4.5 µm band and on the opposite should be very bright in the 5.8 and 8 µm bands. This is very different from present observations. The emitting region is also embedded deeply enough to be situated in the part of the cloud for which we have already shown that it is extremely cold (only 7 K ! ) and this would be totally incompatible with 1000 K grains. The only remaining possibility is that the ambient light due to the nearby stars and the galaxy as a whole is diffracted by the dust grains. However, ordinary interstellar dust grains (between 0.02 and 0.1 µm) are too small to scatter the mid-infra-red light and only bigger grains (between 0.5 and 1 µm) can explain what is seen here.
To model the cloud, a first set of 3D models is computed which reproduces the extinction of the star light. Among this set, a few representative cases are picked up for which the light propagation through the 3D structures is traced, evaluating at each step the influence due to the grains in terms of absorption and scattering as a function of their size. At the end, the models looking closely to the observations are selected. The team has found that the best model is the one for which the grain size increases from the border of the cloud to the center. Indeed, the bigger the grains, the stronger the light scattering. This explains why the effect is strong in the densest central regions and not at the surface of the cloud (Fig. 3 and 4, the differences between observations and model are mostly due to the fact that our input dust map used to build the 3D structure is old and of limited quality). 3D models are particularly difficult to construct because the studied "objects" can not be handled upside down to visualize, understand and measure their shape. Without being completely original, this new 3D model should be better constrained thanks to these supplementary informations and represents a significant step forward to better understand the cloud shapes.
The advantage of this method is that it is very sensitive to the properties of the grains which we measure directly and therefore it allows to strongly constrain the properties of the cloud in combination with the classical measurements of dust absorption and emission. One will be able to extract convincing 3D models from these data which will allow the researchers to better understand the star and planetary formation inside these clouds. Spitzer is the first infra-red satellite to have the appropriate sensitivity and wavelength coverage which have allowed this discovery. The JWST, future large space telescope to which Europe contributes, with a much better sensitivity will allow us to use this tool routinely and to expand this investigation to a large number of clouds throughout the Galaxy.