Within the next 6 years, a new generation of more sensitive gravitational wave detectors is expected to start searching for astrophysical sources of gravitational waves. It is widely expected that these instruments will detect gravitational waves within weeks of becoming operational, beginning a new era of astronomy. The research I propose aims to increase the sensitivity these 'second generation' detectors and is critical for the success of planned 'third generation' detectors, which may operate at cryogenic temperatures. Gravitational waves are fluctuations in the curvature of space-time, produced by the asymmetric acceleration of mass and predicted to be emitted by astrophysical objects such as inspiralling binary neutron stars, interacting black holes and supernovae. Current long-baseline gravitational wave detectors use laser interferometry to monitor the relative displacements of test-masses, which are coated to form mirrors that are highly reflective at 1064 nm. These are suspended as pendulums up to several kilometres apart. Gravitational waves are predicted to induce displacements of less than ~10^-19m in the length of the arms. This displacement is so small that that thermally induced vibrations of the mirrors and their suspensions form an important limit to detector sensitivity. In particular, the mechanical dissipation associated with the ion-beam sputtered mirror coatings has been identified as an important noise source which will limit the sensitivity of future detectors. I propose to develop methods of reducing the mechanical loss, and thus the thermal noise contribution, of the reflective coatings which are crucial to the operation of these detectors. This will increase the sensitivity of future detectors, enabling them to search a larger volume of the Universe for sources of gravitational waves. In particular, gravity wave signals from coalescing compact binary systems, such as two orbiting neutron stars, are predicted to be emitted in the frequency band at which coating thermal noise is expected to limit the sensitivity of future detectors. The coatings used in current detectors consist of alternating layers of tantalum pentoxide (tantala) and silica. Experiments have shown that the mechanical dissipation arises primarily from the tantala layers and that the mechanical loss of tantala can be reduced by doping the material with titanium dioxide (titania). My recent research has shown the presence of a low-temperature dissipation peak in a tantala coating doped with titania, which is characteristic of a particular dissipation mechanism, possibly associated with the tantalum-oxygen bond oscillating between two stable energy states. I would undertake a series of experiments to gain an understanding of the fundamental physics associated with mechanical loss in the coating materials. As noted above, cryogenic loss measurements can be a powerful method of understanding the loss mechanisms in a material. In particular, I plan to investigate whether the shape, height and temperature of the low temperature dissipation peak can be altered by varying the doping concentration, or by heat treatment of the coating, with the aim of developing a detailed model of dissipation initially in tantala. I propose to participate in research planned by the LIGO Scientific Collaboration into possible new coating materials and develop models to understand and reduce their mechanical dissipation. The results of these studies should enable me to develop coating designs with lower mechanical loss, and hence thermal noise, than currently possible, with corresponding enhancement to the sensitivity of advanced gravitational wave detectors. This proposed is also critical for the success of future cryogenically cooled gravitational wave detectors such as the 'Einstein Telescope' which is the subject of a current EC FP7 design study.