In April 2007, T.Corbitt et. al. demonstrated a new style optical cooling and trapping for a gram-scale mirror, utilizing an ultra-small length measurement technique that had been developed to detect gravitational waves (GWs). The general theory of relativity predicted the existence of GWs as ripples of space-time in the scale of 10-23 [1/rHz] strain (3 x 10-19 [m/rHz] in length). In their idea, an optical spring is generated in a detuned-Fabry-Perot (DFP) cavity including a gram-scale mirror. Then, the gram-scale mirror is optically trapped by optical spring restoring force inside the cavity, and also the optical spring mode (~1kHz) displacement is reduced by optical velocity damping force. The detuning means that the mirror position or the laser frequency is set to off-resonate the FP cavity (Figure.1). You can easily imagine that the optical restoring or anti-restoring force will be generated by the radiation pressure inside the cavity according the position of the mirror and the laser frequency. Recently, the velocity damping (accelerating) force was also found to be simultaneously generated in the DFP. In the case that the optical spring force is restoring force, the velocity accelerating force is generated. In contrast, in the case that the optical spring force is anti-restoring force, the velocity damping force is generated. According to this contradiction, the optical trapping of the mirror and the optical cooling of the mode seemed to be impossible because of thermal vibration that is the lasting vibration source. This difficulty was, however, was overcome by use of two laser beams, whose amount of detuning were set oppositely and differently. Because of the property that the only velocity damping (accelerating) force is filtered by the cavity response, the residual optical spring force (Fs) can be kept to be restoring force, even if the residual velocity force (Fv) is set to be damping force (Figure2). Finally, the gram-scale mirror was optically trapped and the optical spring mode was damped at the level of 10-17 [m/rHz]. If the optical spring mode displacement is reduced around 10-18 [m/rHz], which maybe be dominated by thermal noise of the mirrors, the effective temperature will be below 1 [mK], and the quantum number will reach around 1000. If the thermal noise can be reduced by low-vibration noise cryocoolers, and the displacement might be 10-19 [m/rHz] that has been realized in the present GW detectors, micro Kelvin effective temperature and decade quantum number might be realized. We hope resulting quantum phenomena of the gram-scale mirror.

（1）http://www.icrr.u-tokyo.ac.jp/gr/gr.html

（2）http://www.icrr.u-tokyo.ac.jp/gr/GWforPeople.html