Free-fall of LISA Test Masses: a new torsion pendulum to test translational acceleration

Scientific community excitement about gravitational waves is growing year on year: availability of next-generation experiments with improved sensitivity and bandwidth, together with the new astronomical observation capabilities gravitational waves could provide, are a promise of new, amazing science. Among the many complementary operative or proposed gravitational wave experiments, LISA [1] stands out as the first high-sensitivity, low-frequency interferometric space-borne gravitational wave observatory. It promises wonderful possibilities in terms of the number and variety of observable sources, many of which will provide a stable signal in the LISA band with a huge Signal-to-Noise Ratio. Moreover, exciting new science is a probable outcome of its observations. However, the strain sensitivity LISA scientific goal (10×10⁻²¹ 1/Hz¹/² at 1 mHz) is pushing the technological limits of almost all its subsystems. Among them, the Gravitational Reference Sensor (GRS) plays a crucial role in fulfilling the 3×10⁻¹⁵ m/(s²Hz¹/²) requirement in terms of purity of geodesic motion of LISA reference free-falling Test Masses (TM), which sets the instrument sensitivity limit below about 1 mHz. The GRS must assure at the same time high readout sensitivity and low disturbances on the measured TM. Given the unprecedented level of performance needed, thorough testing is required. While a dedicated space mission, the LISA Technology Package [2, 3] technological demonstration experiment, is scheduled to be launched in 2010, extensive ground testing is mandatory to exclude the most threatening sources of disturbances and give a preliminary characterization of the noise model on which the mission's concepts and noise budget apportioning are based. Measuring disturbances below 10 fN/Hz¹/² on Earth in the LISA measurement bandwidth from 3×10⁻⁵ to 0.1 Hz is a demanding experimental task and required the design and commissioning of dedicated facilities. The torsion pendulum currently used by the Trento LISA group has demonstrated the possibility to effectively characterize the GRS and place relevant upper limits on the GRS-related force disturbance. Nevertheless, torsion pendulums used so far have some intrinsic limitations due to the model-dependent conversion of the measured torque noise into corresponding force disturbance and risk being insensitive to some force contributions that could be potentially important. The object of the work presented here has been the building and commissioning of a new concept torsion pendulum facility, directly sensitive to forces exerted by the GRS on the TM, thus significantly reducing the possibility of missing important contributions to the force noise and allowing for simpler, more solid estimation of force-disturbance upper limits. The facility is now operating and, although further debugging and improvement are possible, it has already been used for some preliminary measurements. As well as providing a more representative confirmation of previously obtained results, it has significantly improved the force-noise upper limit, reaching the level required by LTP down to about 1 mHz. We give here an overview of this thesis, pointing out the relevant topics and how they are organized among the ten chapters. In Chapter 1 we will briefly introduce gravitational waves, their sources, and the exciting science that can derive from their observation. We will then describe the Laser Interferometer Space Antenna (LISA) mission, its scope, and the technological challenges it poses: the LISA sensitivity goal and the consequent requirements will be presented, leading us to the need for LTP, the technological demonstration mission, and for extensive ground testing. In Chapter 2 we will focus on one of the core parts of the LISA mission, the capacitive Gravitational Reference Sensor (GRS), which has been designed and is currently being tested in Trento. Along with an introduction to the operating principles and a closer look at its design, we will enumerate the known sources of force noise related to the GRS and describe those which are most threatening. From this, the need for a representative ground testing campaign will clearly arise. Chapter 3 will present the torsion pendulum as a suitable GRS ground testing facility. We will briefly describe the "single Test Mass" (1-TM) facility used in Trento for several years and give a general introduction to the techniques employed to characterize the disturbances that the sensor can exert on a LISA Test Mass (TM) suspended inside it. Besides this facility’s successes, we will then point out its limits, being sensitive only to torques acting on the TM. An evolution of the design, directly sensitive to forces acting on the TM and capable of performing more representative measurements, will then be proposed. The achievable results and their interpretation will be discussed. In the following chapter, the detailed design of the core parts of the new "four Test Mass" (4-TM) testing facility will be presented along with its expected performance. Two different implementations of the pendulum inertial member + capacitive sensors assembly (one for preliminary instrument debugging, the other for actual GRS testing) will be introduced, to be further discussed in the following chapters along with experimental results obtained. Facility features and subsystems common to both implementations, including the suspension system, vacuum system, thermal stabilization, and diagnostics, will then be described. In Chapter 5 we will describe the preliminary, simplified implementation of the proposed 4-TM torsion pendulum. While it was not suitable for GRS testing, it served as a test of feasibility as well as a guideline for the setup of the final facility, including subsystems. It shared the base geometry and fundamental aspects of the final project and was also used to probe achievable performance and give a rough estimate of laboratory gravitational noise background. We will present results that show a nearly thermal noise-limited background and measurements proving that gravitational noise would not have represented a limiting factor for the GRS facility’s performance. Chapter 6 will then be dedicated to the 4-TM torsion pendulum configuration for GRS testing. While most of the core and diagnostic subsystems are common with the preliminary facility, we will mention the minor upgrades and provide a detailed description of the two main changes: the inertial member, made up of four LISA-like hollow TM, and the so-called Stiffness Compensator (STC), an additional sensor designed and built in Trento as a support for readout enhancement and stiffness cross-talk minimization. The first-ever upper limit set by directly measuring the force noise exerted by the GRS on a LISA-like TM inside it, in a flight-representative configuration, will be presented in Chapter 7. Data analysis techniques and related issues will be extensively discussed before showing their application to some noise runs, as well as to a statistical average among two different sets of them, corresponding to two functionality statuses of the facility. The measured excess noise with respect to the expected background, although not null, will be shown to be low enough to set upper limits on the force noise at the level of LTP requirements, and a factor of 10 from LISA, at about 1 mHz. In Chapter 8 and Chapter 9 we describe preliminary measurements of two of the GRS-related effects that need on-ground investigation and characterization: GRS-TM spring-like coupling and stray voltage imbalances. These serve both to verify the sensor electrostatic model and to study potentially threatening electrostatic properties. After introducing the measurement techniques, we will present partial characterizations of the relevant parameters, which are found to be in agreement with previous results obtained with the 1-TM facility, but much more representative of the interactions that can produce potentially disturbing effects along the TM X-axis, used in flight for interferometric measurement aimed at detecting gravitational waves. Finally, Chapter 10 will summarize the most interesting and innovative aspects of the work presented throughout this thesis: the building, commissioning, and first valuable investigation campaigns of a new generation torsion pendulum able to set upper limits of unprecedented representativity by directly measuring the force acting on a LISA-like TM inside the GRS. Besides the obtained results, the work presented resulted in the availability of a new instrument that will help in reducing the risk of failure of an innovative, exciting scientific mission such as LISA.