The MPE and VLTI
The Very Large Telescope (VLT) is a group of four 8.2 m telescopes, but it is more than the sum of four telescopes as it has been designed so that the beams collected by each of these four apertures can be combined together in order to perform optical interferometry. This mode of operation on the VLT is called VLT Interferometry (VLTI).
The goal of interferometry is to improve the spatial resolution of the instrument: with the advent of adaptive optics, the resolution of individual telescopes has become diffraction limited, meaning that the minimum distance between two stars necessary to see each of them individually rather than a blend of both depends mostly on the diameter of the telescope. Now, it is in principle possible to achieve the resolution of a 100 m wide telescope by simply putting two smaller telescopes 100 m apart from each other and combining their light in a coherent way. This principle has been applied successfully for decades in the radio and sub-mm domains for several decades with projects such as the Very Large Array (VLA). However, optical interferometry is much more difficult than radio interferometry, mainly because both the atmosphere behaviour and the detection technologies are different between radio and optical frequencies. Several experiments have already demonstrated the feasibility of optical interferometry, such as that performed by A. Labeyrie in 1974 at the Nice observatory; now, the VLTI aims at overcoming the difficulties of optical interferometry in such a way that it will not be an experimental technology anymore, but an everyday tool for tomorrow's astronomer.
The MPE has been committed in VLTI from an early stage on, with the building of LISA, the near infrared camera for VINCI. The MPE is now willing to contribute to the fringe tracker PRIMA: a strong limitation on optical interferometry is imposed by the atmosphere, that changes the path difference between the light rays collected by the two telescopes of the interferometer in a time-dependent manner. For this reason, the interferometric fringes, that arise on the detector as the beams coming from the two telescopes interfere, are constantly moving, and therefore must be acquired quickly (within about 1 ms) in order to be detected, which means that only fairly bright objects (K<14 with the UTs) can be directly observed. However, PRIMA has been designed to overcome this limitation in a similar way to the principle of adaptive optics: it will allow for using a bright reference star to determine the fast-changing atmospheric term in the path difference and correct for it in real time, thus freezing the fringes of the nearby, fainter science object on the detector. The integration time for the science object can then be significantly increased, and the limit luminosity enhanced by about 5 magnitudes. PRIMA is a complex device, made of several subsystems. Each telescope that has to be used with PRIMA must be equipped with a star separator (STS), the piece of hardware that allows for selecting both the reference star and science object within the filed-of-view of the telescopes. The construction of two of these STS is currently underway, and the MPE has offered to fund a third one, that would allow for retrieving two-dimensional information from the science object within a single observation (actually, the simultaneous use of three telescopes instead of two allows to record three times as much independent information from the source within a single observation).
In addition to these contributions into the hardware of VLTI, the MPE is developing a scientific program to make use of this fantastic new tool. For instance, we would like to further our study of the proper motions of stars in the direct vicinity of the Galactic Centre black hole candidate Sgr A*. During more than a decade, we have observed the central arcsecond of the Milky Way with unprecedented resolution in the near infrared, and been able to observe a group of stars that revolve extremely fast around the central point of the Galaxy, coincident with the point-like radio source Sgr A*, proving that a very massive and compact object lays at this location. Observations of the same region with interferometric technics should allow for discovering more stars (about half a dozen according to our estimates), also orbiting the central black holes at even smaller radii, within one resolution element of each of the VLT telescopes of about 0.05 arcsecond diameter. Their proper motions should quickly, within a few years, reveal perturbations that would allow to weigh not only the black hole itself, but also the cusp of faint, undetectable stars that surrounds it.