| With the development of the semiconductor industry,element integration is higher and higher,the size of the key unit has reached tens of nanometers.In the production process,the measurement feature size is often difficult to break through the focus,in the case of last resort,even need to destroy the sample to measure.The current technical approach has been difficult to meet the current component size requirements.Atomic force microscopy(AFM)is a common tool in nanoscience and technology that can not damage the sample during scanning.AFM scanning imaging before the need for manual or semi-automatic operation steps to achieve the probe and sample approach process,low degree of automation,cumbersome operation,the probe vulnerable to damage.And most of the current research AFM can only achieve two-dimensional or 2.5-dimensional scanning imaging of the sample process.The above problems have seriously affected the level of automation in the production process and the technical methods for testing the samples.Therefore,AFM in industrial production is often faced with a lot of technical improvements.The main contents of the thesis are as follows:(1)Through reviewing the relevant technical literature at home and abroad,the research object of the paper is determined,and the basic research status of the atomic force microscope system and the development trend of the related technology are summarized,and the automatic combination of coarse and fine Positioning method for automatic focusing,on the basis of this proposed probe-sample automatic approximation process.Finally,a three-dimensional measurement of the sample is proposed.(2)The auto focus method is used to improve the auto-focusing algorithm in computer vision.A method of the mean edge Laplacian operator is proposed,which has strong anti-noise performance and improved focus Accuracy and algorithm of universality,can adapt to AFM long stroke,different texture samples automatically focus.(3)The paper proposed a combination of coarse and fine segmented automatic positioning method.In the coarse positioning process,a mean algorithm based on the strong edge Laplace operator is proposed for autofocus.This algorithm has high noise endurance,can adapt the long travel and different texture samples.In the fine positioning process,with the precise force feedback control method,when the force between sample and probe exceeds the set point,the probe can be automatically backed up by the Z axis nanopositioning piezo stage,so that the probe tip is effectively protected.Through the effective integration of the two methods,the automatic feeding process can be realized.Therefore,it can improve the ease of use and efficiency of AFM.(4)The paper proposed a two-dimensional feedback control method for measuring the three-dimensional structure.The control system can adjust the feedback motion angle of the probe according to the different side wall direction,so that the probe tip is able to contact the sidewall effectively,and reach the purpose of detecting the sidewall morphology.In this paper,the 3D measurement method is verified experimentally on the self-made 3D-AFM platform,and the sidewall profile information and 3D topography of the grid structure are obtained.The experimental results show that this method can effectively realize the 3D shape measurement of the nano structure samples... |
PDF Full Text Request