Research On Digital Lithography Imaging Algorithm And Mask Optimization Methods

As a mask generator, DMD not only has the characteristics of flexibility, low cost, fast speed, but also can display any shape of the "mask" graphics and can be modified in real time. Therefore, DMD –based digital lithography technique has aroused great attention. At present, the technology in the world is still at the exploratory research stage. The research of digital lithography imaging theory and the fast and accurate digital graphics optimization algorithm have important theoretical and practical significance for the DMD-based digital lithography technique. In order to explore fast imaging algorithm and improve the imaging quality for DMD-based digital lithography technique, we carry out a systemic study on the DMD-based digital lithography technique. The main contents and contributions of this dissertation are as follows:1. Study with partial coherent imaging theory, a rapid computation method for DMD-based digital lithography partially coherent imaging algorithm is proposed combined DMD mirror characteristics with matrix multiplication approach. At the same time, in order to improve the performance of the proposed algorithm, two kinds of acceleration methods are put forward. The first one can enhance the efficiency of the algorithm with the reduction of the number of feature vector matrix by SVD, and the second one can enhance the compute speed with the reduction of the sampling number of the light source with the proposed radius sampling mechanism. Simulation results show that the algorithm has high calculation accuracy with less time consumption, and the two acceleration methods can effectively improve the efficiency of the algorithm.2. With analysis of the characteristics of the DMD as a digital mask, the system response function of the single mirror is set up, and a pixel-based imaging algorithm is also proposed. In this method, DMD-based digital lithography system is regarded as a linear invariant system, the whole DMD digital mask imaging is realized with the superposition of single mirror imaging, and it provides an effective method to analyze the influence of the DMD mirror state on the image quality of lithography imaging. The simulation results validate the effectiveness of the proposed algorithm.3. In order to improve the resolution of DMD digital lithography system, a mask optimization method based on inverse theory is proposed. In the method, DMD gray scale mask is expressed with binary mask by using of fine sampling method. In the method, the optimization gray scale of DMD mirrors can obtained by gradients algorithm. In additional, in order to overcome the crosstalk of DMD mirrors, the concepts of the key pixel and the bound pixel are proposed. The method overcomes the shortage of the high compute complexity of using directly with pulse width modulation method for DMD gray scale imaging. Simulation results show that the proposed method improve the imaging quality effectively, and has the advantages of simple calculation, good optimization effect.4. As low quality of the 3D structure constructed by traditional DMD gray tone image method, a method to improve the quality of 3D structure constructed with DMD based gray tone lithography is explored. In the method, an optimization model has been set up, in which the exposure time of DMD mirrors are act as optimization variables, the difference of expected height and reconstruction height of the constructed structure are act as performance indicator. In order to reduce the computational complexity, the relationship table between the DMD mirrors and the pixels of the resin surface is built, and the conjugate gradients algorithm(CGA) is adopted to resolve the optimization problem because of the high scale variable of performance indicator. Simulation results show that the proposed algorithm can effectively improve the quality of three dimension structure, and has good prospect in the engineering application.