6 Conclusion and perspectives

  

Figure 8: Photograph of two prototypes of integrated optics components together with a match to give the spatial scale. Left component features a 3-way beam combiner with photometric calibration channels manufactured with the silicon etching technology (Severi et al. 1999). Right component manufactured with the ion exchange technology and connected with two fibers is a 2-way beam combiner with photometric calibration channels

6.1 Decisive advantages

We have shown the great interest of using integrated optics for astronomical interferometry. Figure 8 shows a three-way beam combiner with photometric calibrating channels made with the silicon etching technology (Severi et al. 1999).

We argue that integrated optics technology which is already industrially mature, presents the following main advantages for astronomical interferometry:

• Small size. A complete instrument can be integrated on a chip typically $5\mbox{ mm}\times20\mbox{ mm}$.
• Stability. The instrument is completely stable while embedded in a substrate.
• Low sensitivity to external constraints: temperature, pressure, mechanical constrains.
• Few opto-mechanical mounts and little alignment required. The only concern is coupling light into the waveguides.
• Simplicity. For a complex instrument, the major efforts are shifted to the design of the mask, not in the construction phase.
• Intrinsic polarization capabilities (Sect.  5.3).
• Low cost. Integrated optics provides very low cost components and instrumentation set-up. Furthermore the price is the same for one or several components since the main cost is in the design and initial realization of the mask.

6.2 Application to interferometry

Integrated optics components are not well-suited for wide wavelength coverage, high spectral dispersion and large field of view. Therefore, we do not think that it will completely replace existing techniques in astronomical interferometry. However with the characteristics presented in this article, we think that integrated optics will be attractive for the two following applications:

• Interferometers with a large number of apertures.
  Whatever the complexity of the instrumental concept, only one component in integrated optics allows combination of several beams and ensures the photometric calibrations at low cost and with limited alignments
• Space-based interferometers.
  Integrated optics components with no internal alignments provide reliable beam combiners for spatial interferometry.

These specific advantages lead us to perform laboratory experiments to validate this approach. An interferometric workbench has been built to completely characterize various components realized by both technologies. First fringes with a white source have been obtained [Berger et al.\ (1999), Paper II] and the integration of an interferometric instrument dedicated to astronomical observations in the H and K atmospheric bands is in progress [Berger et al. (1998)].

Acknowledgements

The authors are grateful to F. Reynaud (IRCOM - Univ. Limoges), P. Pouteau, P. Mottier and M. Séveri (CEA/LETI - Grenoble) for fruitful discussions, and to E. Le Coarer and P. Feautrier for the idea of combining integrated optics and STJ. We would like to thank K. Wallace for carefully reading the manuscript. The works have partially been funded by PNHRA/INSU, CNRS/Ultimatech and DGA/DRET (Contract 971091).