Like all other laser interferometer, LISA depends on the stability of the optical path length between optical components. Changes in the mechanical distances between certain components on the optical bench and changes in the index of refraction of transmissive optical components can reduce the final sensitivity of the LISA interferometer. In most experiments, fluctuations in the temperature are the main reason for these changes. However, thermal expansion coefficients of most materials are well known and experienced thermal engineers will design each spacecraft such that thermally induced fluctuations can be neglected. But LISA's requirements are unusually demanding and many materials undergo additional temperature-independent dimensional changes. These changes are thought to be correlated with annealing or stress relaxation processes in the material. They often seem to depend not only on the material itself but also on the history of the particular piece, how it was produced, machined, and long it was left alone to relax. The details of these processes are neither well understood nor well quantified, but it can be expected that they will be in many cases well above the LISA requirements. Just recall that 10pm is only a small fraction of the diameter of an atom.

Experimental studies of the stability of LISA components require an environment with a temperature stability such that the temperature related expansions or contractions are smaller than the requirements. One way to achieve such a stable environment is to use several l ayers of gold-coated stainless steel sheets as radiation shields inside a vacuum chamber. The temperature inside the enclosed volume can have uK stability which should be sufficient for most interesting materials.

We have build such a vacuum chamber and started measuring the stability of different materials. The material under test will define the distance between two mirrors, these mirrors face each other and form an optical cavity or a Fabry-Perot. The frequency of a laser will be stabilized to a resonance frequency of the cavity. A second laser will be stabilized to a reference cavity. Changes in the length of the test cavity will change the difference frequency between the lasers. The difference frequency will be monitored and later analyzed.

Materials which we plan to test include ULE, Zerodur, Silicon Carbide, Super Invar, and Carbon Fiber Reinforced Polymers (CFRP). These materials are the most likely materials to be used in interferometric space missions. The reference cavity for LISA's laser frequency stabilization will most likely be made from ULE or Zerodur. ULE is tested at Goddard Space Flight Center and showed a stability of better than 30fm. Our group focuses on Zerodur and hopes to achieve similar results in the near future.

The separation between the primary and the secondary mirror in the LISA telescope is one of the most critical distances in terms of mechanical stability. The telescope is the most complex mechanical apparatus within the optical train and it has to face the cold open space. Therefore, it is not as well insulated as the rest of the interferometer.

The dimensional stability of the telescope has to be better than 10pm to ensure that dimensional changes in the separation between primary and secondary mirror do not limit the sensitivity. Different materials and designs are currently discussed for the LISA telescope. Possible materials include ULE, Zerodur, Silicon Carbide, and CFRP. ULE and Zerodur have the lowest coefficient of thermal expansion (CTE); this is the reason why they are used for reference cavities. However, ULE and Zerodur are brittle and the mass penalty for an all-ULE or all-Zerodur telescope would be unnecessarily large. Remember that the requirements on the reference cavities are about 3 orders of magnitude more stringent than on the telescope.

Silicon Carbide and CFRP are two materials with modest CTE. However, both materials are very strong and can be used to build complex light-weight structures. We plan to measure the dimensional stability of a Silicon Carbide support structure and of different types of CFRP and also study different bonding techniques to attach multiple pieces of the same or of different materials together.