Research On Fast Image Scanning Method Of Atomic Force Microscopy

Atomic Force Microscopy(AFM)has become one of the most popular scientific instruments for nanoscale characterization and measurement due to its high resolution,non-destructive,flexibility and versatility in various phase environments and different materials imaging.It is widely used in physics,chemistry,materials,biology,medicine,nanoscience and other fields,and serves the semiconductor,aerospace,manufacturing,energy and other industries.Although AFM has wide applicability,the high-resolution imaging of traditional AFM usually takes several minutes,which severely restricts the imaging efficiency of AFM.In view of the limitations of traditional AFM in scanning speed,image quality and comprehensive characterization,this article starting with the establishment of AFM system model,optimizes the design of high-speed Z-positioner,constructs high-speed data acquisition and processing unit,develops fast AFM scanning methods and control strategies,and improves the imaging speed of AFM.First,the dynamic model of probe in typical AFM imaging mode is established,and the simplified model of probe sample interaction force is established by simulation analysis of step response of cantilever to hard substrate in contact mode.Based on the input and output experimental data of the core components in AFM imaging system,the calibration of parameters and the construction of mathematical model are completed.The reliability and accuracy of the model are verified by the tracking test results of typical trajectories.This modeling process provides a theoretical basis for the design of model-based controller.Then,the traditional AFM scanning usually lacks the priori position information of the target of interest,it is difficult to directly carry out high-resolution imaging of the target area,and the image often contains most of the invalid regions,which greatly limits the imaging efficiency of AFM.Regarding the issue above,a fast target boundary tracking and local scanning method is proposed.The boundary tracking algorithm can quickly locate the target and record its boundary points.Based on the perceived target edge information,the local scanning method can achieve fast imaging of only the target coverage area.Experiments on different AFM imaging modes verify the effectiveness of the method.Subsequently,aiming at the problem that the imaging speed of conventional nanopo-sitioning stage is limited in the traditional raster scanning mode,it is proposed to improve the AFM scanning speed and imaging quality by designing high-performance controllers and data post-processing method.The raster scanning mode driven by sine wave avoids the mechanical resonance of the scanner caused by the higher harmonics of the conven-tional triangle wave.An iterative learning controller was developed to effectively sup-press the hysteresis error of the fast axis of the XY piezoelectric scanner;a feedforward-feedback controller was further developed to reduce the response delay and improve the tracking performance.A dual-actuated high-speed Z-nanopositioner is developed,and a H_?controller based on the mixed sensitivity framework was designed to improves the feedback control bandwidth and imaging quality.A data post-processing method based on Kalman filter iterative fusion for optimal estimation of sample morphology is proposed,which further improves the image quality of AFM.A high-throughput data acquisition and real-time digital signal processing system based on a field-programmable gate array platform was constructed,and by combining fast scanning methods and control strategies,the scanning speed of conventional nanopositioning stage was increased from a few hertz to a few tens of hertz,achieving fast and high-quality AFM scan imaging.Finally,in view of the limitation that the existing three-dimensional AFM imag-ing technologies cannot realize the comprehensive fast imaging of the three-dimensional structure,a fast rotation positioning scanning method for revolution structure is proposed.A high-precision rotary transposition micro-platform based on an optical microscopy sys-tem was constructed to compensated the position deviation of the sample during rotation process,and realized fast in-situ positioning and scanning imaging of the sample.A three-dimensional reconstruction method based on image stitching was developed,and a three-dimensional distribution map of sample morphology and nanomechanical proper-ties was completed.