In a combined effort of the Max-Planck-Institute for Gravitational Physics (AEI), the Institute for Gravitational Physics of the Leibniz Universitaet Hannover, and the QUEST cluster of excellence, a 10m prototype interferometer facility is currently being set up at the AEI in Hannover. The prototype interferometer will be used to test and develop some of the techniques for potential future upgrades of the gravitational-wave detector GEO600. Furthermore, experiments to explore quantum mechanical effects in macroscopic objects will be run in this facility.
In the first round of experiments a 10m Michelson interferometer with a sensitivity that is solely limited by quantum noise at all relevant Fourier frequencies will be set up. The sensitivity reached under these conditions is referred to as the Standard Quantum Limit (SQL). However, even this remarkable sensitivity limit can be overcome by the injection of squeezed states of light.
In order to reach such a sensitivity several innovative approaches and techniques are required.
First of all, the experiments have to be enclosed in a proper vacuum envelope to minimise acoustic coupling and effects from changing of the air's refractive index as well as air damping of mechanical resonances. With 100m³ volume, 3m diameter tanks and 1.5m diameter beam tubes the ultra-high vacuum system is designed in a rather generous way, such that it can hold more than one experiment at a time. The pressure inside the system is below 10-6mbar after 12 hours of pumping, after one week it is below 10-7 mbar.
The optical configuration of the core instrument, a sub-SQL interferometer, will be a 10m Michelson interferometer with Fabry-Perot cavities in the arms. No recycling techniques, such as power recycling or signal recycling, will be used in the initial round of experiments. However, high light power is required to approach the SQL regime and hence, arm cavities of finesse F=675 are planned. In this way the resonantly enhanced light power does not travel through the beamsplitter, minimising the thermal load in the substrate. The mirrors of each interferometer arm will be coated with crystalline AlGaAs coatings, reducing thermal noise by a factor of ca. 3 in ASD.
It is required that all relevant optical components will be isolated from seismic motion of the ground by about ten orders of magnitude. As a first stage of vibration isolation all optical components will be mounted on top of passive isolation platforms inside the ultra-high vacuum system. These platforms are based on geometrical anti-springs and inverted pendulum legs (please see Isolation Tables for details). Even though the tables are in close proximity to each other there will inevitably be some residual differential motion between them. In order to further minimise the differential motion of the tables an interferometric link between them will be established. This so-called Suspension Platform Interferometer (SPI) is based on a set of Mach-Zender Interferometers which are read out by a LISA Pathfinder phasemeter. The goal for this stabilisation is to reach a differential displacement between the tables of less than 100pm/sqrt(Hz) and less than 10nrad/sqrt(Hz) at a Fourier frequency of 10mHz.
The optical components themselves will be suspended from multiple cascaded pendulums, which are mounted on top of the isolation tables (please see suspension systems for more details). Each pendulum stage provides 1/f² isolation above its resonance frequency. Steering mirrors and frequency-reference cavity optics will be suspended in steel-wire loops. The 100g core optics of the Michelson interferometer will be all monolithically suspended by four 20µm diameter silica fibres. Such monolithic suspensions, as used for the GEO600, VIRGO+ and AdvLIGO core optics, minimise suspension thermal noise to the required level. Damping of the pendulum resonances will be carried out as a combination of eddy-current damping and active electronical feedback. Length control for the optical cavities will be applied from seismically well isolated reaction pendulums.
The light source for this experiment will be a LZH eLIGO 35W Nd:YAG laser system. Coupling of the laser light into the vacuum system will be done by use of a photonic crystal fibre. Even though substantial fibre modecleaning can be expected a 40cm ring-cavity premodecleaner will be the first optical element in the path. A small fraction of the light is then tapped off to operate a suspended 12.3m triangular ring cavity of finesse F=7300, which serves as a frequency reference at Fourier frequencies well above the pendulum resonances (please see frequency-reference cavity). At lower Fourier frequencies the laser frequency will be stabilised to a Iodine molecular reference. The laser will be stabilised to better than 5x10-9 in relative intensity noise (RIN).
Data acquisition and the implementation of all relevant servo loops are done via a version of the digital AdvLIGO Control and Data system (CDS). The control loops are programmed with the help of Matlab/Simulink or in C code.
The design of the experiment is done such that the sum of all classical noise sources lies well below the sum of the quantum noise in a region between 50Hz and several 100Hz (please see Design and Sensitivity for more details). Once quantum limited sensitivity is reached, the injection of appropriate squeezed states of light or the application of back-action evasion techniques can further improve the sensitivity to below the SQL.
Further down the road, experiments such as thermal noise interferometry or even entanglement of macroscopic test masses can be investigated.