Improving the performance of nano-optomechanical systems for use as mass sensors

Abstract: Coupling optical cavities to freely moving mechanical devices results in optomechanical systems. Enabled by advancing fabrication techniques, optomechanical systems are now easily fabricated using silicon-on-insulator chips at the micro- and nanoscale. These nano-optomechanical systems (NOMS) combine nanomechanical devices with photonic integrated circuits for sensitive readout of mechanical displacement. NOMS have many potential applications, including mass detection. In this thesis, we work towards improving the dynamic range of cantilevers coupled to photonic circuits for mass sensing applications. We use an all-optical pump-and-probe approach. First, we derive and demonstrate a simple model for designing optomechanical systems for efficient actuation via the optical gradient force. We achieve large amplitudes of vibration and show the power required to achieve these amplitudes is nearly four times smaller than an unoptimized design. When driving the cantilever to large amplitudes, we observe multiple simultaneous nonlinearities. To understand the source of these nonlinearities, we derive and solve a numerical model that allows us to determine that there are three nonlinearities present in the system; a readout nonlinearity and two distinct optical force nonlinearities. The model is unique in that it incorporates the effect of the pump laser on the optomechanical system. Of all the nonlinearities, the readout nonlinearity is the most detrimental to mass sensing performance as it hinders our ability to determine the amplitude of the mechanical device. To mitigate the readout nonlinearity, we design a photonic integrated circuit to increase the linear range of NOMS readout. This photonic architecture successfully increases the linear range by a factor of four while maintaining the sensitivity of the previous design. These improvements increase the dynamic range, and therefore the limit of detection, by a factor of 2.8. The work presented here will help to improve the viability of NOMS mass sensors in a variety of applications, including portable health care diagnostic tools.