AFM Vibration Noise Reduction via Squeeze-Film-Damping-Effect Sensing Method

• Published in: IEEE transactions on instrumentation and measurement Vol. 73; pp. 1 - 11
• Main Authors: Zhai, Shenghang, Liang, Liang, Shi, Jialin, Yu, Peng, Yang, Tie, Yang, Yang, Su, Chanmin, Liu, Lianqing
• Format: Journal Article
• Language: English
• Published: New York IEEE 2024; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Actuators; Atomic force microscopy; Atomic force microscopy (AFM); Error analysis; Error reduction; finite impulse response (FIR) filter; Force; Measurement by laser beam; Noise measurement; Noise reduction; Sensors; squeeze-film-damping effect; Topography; Vibration; Vibration control; Vibration damping; Vibration measurement; Vibrations
• ISSN: 0018-9456; 1557-9662
• DOI: 10.1109/TIM.2024.3440377

Large industrial atomic force microscopy (AFM) plays a pivotal role in the metrology of large samples. However, significant vibrations between the large AFM and the sample can limit the resolution of the AFM. By detecting and subtracting these vibrations from the image signal, image noise can be effectively reduced. However, measuring these vibrations presents a challenge, particularly because most large samples are dielectric and fragile, which makes conventional electrical vibration sensors unsuitable. Furthermore, the frequency responses of the vibration sensor and the AFM differ. Therefore, directly subtracting the vibration signal from the AFM's topography signal can lead to considerable errors owing to the magnitude and phase differences between these signals. In our study, we designed a noncontact vibration sensor leveraging the squeeze-film-damping effect. This sensor can measure dielectric samples and achieve a high resolution (better than 0.1 nm). We processed the measured vibration signal using a magnitude-phase adjuster to compensate for the dynamic differences between the sensor and the AFM. We managed to limit the subtraction error to within 1 nm using this adjuster, achieving a reduction of 63%-88% in subtraction errors. Additionally, we validated our vibration reduction method using topography images of standard grating samples.