Leveraging Optical Lever Techniques for NEMS Resonators

The precise measurement of mechanical resonances in nanoscale devices (NEMS) is critical for both fundamental research and technological applications. Various read-out techniques exist to characterize micro- and nanoscale resonators, each with distinct advantages and limitations. Among these, interferometric optical methods are widely used due to their high sensitivity in displacement measurements. One of the earliest optical read-out methods is the optical lever technique, which has been successfully implemented in various applications. Its simplicity and robustness make it particularly useful in Atomic Force Microscopy (AFM), yet its adoption in other NEMS implementations remains limited. Recent studies have demonstrated that an optical lever system can mitigate optical back-action effects on membrane-based NEMS, achieving a read-out noise floor two orders of magnitude below the standard quantum limit. The limited adoption and lack of comparative analysis of optical lever read-out techniques for NEMS motivate this study. This work investigates the feasibility of using the optical lever technique for frequency measurements of NEMS resonators. To evaluate the effectiveness of this approach, an experimental setup was designed to enable a direct comparison with two established measurement methods: (i) an electrical read-out technique, the magnetomotive read-out, and (ii) an optical read-out technique, the laser Doppler vibrometer. The impact of each method on the frequency stability of the resonators is assessed under comparable conditions. Additionally, a NEMS application utilizing a read-out scheme is selected to compare the performance of the optical lever technique and the magnetomotive read-out, demonstrating the impact of the optical method relative to the electrical approach.The results confirm that the optical lever technique provides performance comparable to interferometric methods and is well-suited for membrane-based NEMS. The achieved noise performance surpasses that of the electrical read-out. These findings contribute to the ongoing exploration of alternative transduction mechanisms for nanoscale mechanical systems, offering a promising approach for stable and high-precision sensing applications.