5 Conclusion

We have described the concept of the Achromatic Interfero Coronagraph, based on destructive interferences and we have derived analytically its theoretical capabilities relating to the detection of a faint companion close to a star. The two main advantages of this device are its inherent achromaticity (which allows choosing freely both a working wavelength and a spectral bandwidth) and its ability to sense very closely the neighbouring of a star (or other unresolved source), namely at angular distances of a fraction of the first radius of the Airy pattern. This latter feature of our coronagraph is not available in other coronagraphs using Lyot masks (Mouillet et al. [1997]) for which sensing is not possible closer than several Airy radii, while achromaticity is not provided by both Lyot Coronagraphs and the Phase Mask Coronagraph (Roddier & Roddier [1997]).

Beside these advantages, some specific technical constraints in terms of internal adjustment have to be considered, namely optical quality and accurate nulling of optical path difference. As for every coronagraph, observations from the ground require adaptive optics. The goal of adaptive optics here is to minimize unwanted residual light in the image plane, caused by departure from centro-symmetry of the complex amplitudes of interfering fields.

In such conditions and even when adaptive optics correction is not fully achieved, detection capabilities for ground-based observations makes our coronagraph a powerful tool for the detection of faint sources located in the neighbouring of a star, as close as a fraction of the theoretical angular resolution set by diffraction. Theoretical estimates given in this paper show that at angular distances comparable to the ones explored by other coronagraphs, our coronagraph has comparable detection capabilities. At angular distances, unreachable by other coronagraphs, clear detection of companion 6 magnitudes fainter than a star with K-magnitude of 5 is a truly reasonable goal for a 4-m class telescope equipped with adaptive optics. This magnitude difference can be significantly increased by an appropriate observing strategy (separate and adjacent spectral bands) by longer integration time and by appropriate data processing. Owing to expected progress on adaptive optics capabilities, detection of a close companion, at an angular distance smaller than the diffraction limit, with $\Delta m$ up to 12 (and beyond when increasing integration time) seems to be within reach from raw data with our coronagraph.