Towards High-Precision Astrometry: Diferential Delay Lines for PRIMA@VLTI

Abstract

Deriving orbital parameters and masses of extrasolar planets by means of measuring the variation of the host star positions requires an astrometric accuracy of 10 microarcsec. To achieve this goal, a consortium with partners from Germany, the Netherlands, and Switzerland, in agreement with ESO, will enhance the PRIMA facility at the VLTI with Differential Delay Lines (DDLs). We give an overview of the PRIMADDL project, which consists of developing hardware, astrometric operation tools, and data reduction software, and outline the anticipated astrometric planet search program to be carried out with this facility. 1 Narrow-angle Astrometry with the VLTI PRIMA, the instrument for Phase Referenced Imaging and Micro-arcsecond Astrometry is currently being developed at ESO. PRIMA will implement the dual-feed capability at the VLTI for both UTs and ATs to enable simultaneous interferometric observations of two objects that are separated by up to 1 arcmin, without requiring a large continuous field of view. PRIMA will be composed of four major sub-systems: Star Separators, Differential Delay Lines (DDLs), a laser metrology system, and Fringe Sensor Units (FSU). The system is designed to perform high-accuracy (10μas) narrow-angle differential astrometry in K-band with two FSUs and, with one FSU in combination with AMBER or MIDI, phase-referenced aperture synthesis imaging. The purpose of the DDLs in differential astrometry is to increase the astrometric accuracy by separating the large OPD correction terms which are common for the two stars from the small differential terms, and to increase the sensitivity by stabilizing the fringe pattern (in a closed loop with the laser metrology) and thus allow for longer integrations. 2 DDLs and Astrometric Software for PRIMA In order to speed up the full implementation of the 10μas astrometric capability of the VLTI and to carry out a large astrometric planet search program, a consortium lead by the Observatoire de Geneve (Switzerland), the Max Planck Institute for Astronomy in Heidelberg (Germany), and the University of Leiden/NOVA (The Netherlands) agreed with ESO to build and deliver the Differential Delay Lines for PRIMA (see Fig. 1) and to provide all necessary operation 266 Launhardt et al. and software tools to perform narrow-angle astrometry at the 10μas level. This includes developing and building all the DDL hardware, the construction and analysis of an astrometric error budget, the establishment of an operations and calibration strategy, and the development of observation preparation and data reduction software: Fig. 1. The PRIMA-DDL project: breakdown of hard and software components 3 The Astrometric Planet Search Program When completed in 2007, we will use the upgraded PRIMA facility to detect and characterize extra-solar planets through the reflex motions of their host stars in the plane of the sky. Two core programs are planned to be carried out over a duration of at least three years: 1) Observe all stars with known radial-velocity planets that are in reach of the VLTI and have a suitable phase reference star. We will resolve the sin i uncertainty of the planet masses and thus constrain the uncertain upper end of the planetary mass function. For stars with multiple planetary systems we will derive the relative inclination of the orbits. We will follow up long-term radial velocity trends and search for new planets in longer-period orbits for which astrometry is more sensitive than the radial velocity method. 2) Planet search around stars of different mass and evolutionary status without known planets. The search for planets by the radial-velocity technique is restricted to stars with narrow and stable spectral lines, thus excluding A and most F stars with their broad spectral lines as well as pre-main sequence stars. Our astrometric planet search program will explicitly include such stars. For nearby (< 20 pc) late-type (F-M) main sequence stars, the primary new discovery space opened by such an astrometric facility would be Saturn down to Uranus-mass planets with orbital periods of a few years (a = 1–5AU).