Research On Gray-Scale Mask Fabrication And Application Technique Of Micro Optics Element

Design, fabrication and application of micro-optics element (MOE) are put a high value on nowadays with the wide applications of MOE in modern communication, military, space technology, superfinishing, information processing, biomedicine, entertainment, and so on. At present, fabrication limit is a bottle-neck of the MOE development. Therefore, the emphasis of this dissertation is put on the research of MOE fabrication. Moreover, design and application of MOE are also briefly discussed. The main contents and contributions of this dissertation are as follows:1. On the basis of overall analysis of scalar and vector diffraction theory, method of optical -path-difference integration (OPDI) is given for the first time. OPDI is a simple and universal method which can be used to design and analyze MOE with complex structure in scalar domain.(1) The validity of OPDI was proved by comparing the diffraction efficiency computed with OPDI, angular spectrum and rigorous coupled-wave algorithm respectively.(2) With OPDI, a new type of isosceles blazed grating with compensated phase on the obverse side is designed, which avoids the difficulty of inverse alignment and secondary diffraction. The correctness of the new design is validated by the finite difference time domain method.2. A method of gray-scale subdivision expanding by color-gray equivalent is put forward for the first time.(1) The necessity of grayscale-subdivision expand is demonstrated with the analysis of the relation between the depth and intensity of exposure.(2) Two methods of color selection for color equivalent are given. One is testing selection. The other is resolve computation.(3) Color analog mask made with film is easily affected by exterior circumstance and its reproducibility is poor. So the concept of color digital mask is first brought forward. Combination with three color liquid crystal displays (LCD) is given as an example. The formulas of grayscale-subdivision expand is derived.3. Digital Micromirror Device (DMD) is selected to found a new digital gray-scale-mask system for the first time. With digital real-time mask and parallel direct-writing technique, speed of mask fabrication is improved and good experimental results were obtained.4. Based on the digital gray-scale-mask system using DMD, a series of new techniques for digital mask exposure have been put forward as follows:(1) Digital-mask-move technique. This technique can be used to fabricate cylinder lens, sine gratings, micro-lens array with bigger numerical aperture, etc. A model of aperiodic movement for single move-mask and a model of edge-effect for move-mask array were founded respectively. Results of simulation and experiments are given.(2) Digital-mask-rotation technique. This technique can be used to fabricate micro-lens with bigger numerical aperture, beam shaping element with circular symmetry structure and axicon prism, etc. Mathematical models of rotating mask for several common MOEshave been built. Results of simulation and experiments are given.(3) Digital-mask-fractal technique. Mask fractal can be used to solve the problem of energy loss in the edge of reduction lens with finite aperture. Many fractal methods have been put forward, such as period magnifying fractal, step dividing fractal and blend fractal, which can be applied to different type of MOE masks. With mask fractal, the transverse resolution of digital mask exposure is improved. Taking example for binary grating, the energy loss caused by diffraction limit was computed. The result showed that the lost efficiency caused by lens-aperture limit decreases greatly by use of mask fractal. Moreover, the energy on the middle and high orders diffraction has been increased.(4) Digital-mask-coding technique. Coding mask can be used to fabricate special MOE such as beam shaping element, beam splitting element. The coding mode of digital mask is analyzed. The requirement of digital-mask-coding system is discussed.(5) Method of gray-scale-subdivision expanding with two DMDs combination. The exposed depth of digital mask can be finely controlled by subdivided gray-scales. Combining two DMDs and changing their incident intensity ratio, equivalent gray-scales can be expanded more than five times and the longitudinal resolution is improved. Depth error of mask exposure can be decreased to less than 2%. 2DMDs secondary modulating or superimposed modulating simplifies the method of color-gray equivalent. Accuracy and flexibility of digital mask are increased. With 2DMDs combination, digital mask fabricating system can be widely used to make MOE with complex relief structure.All of the new technique for digital mask introduced above also can be applied to the gray mask fabricating system using other electrically addressed spatial light modulators, such as LCD and liquid crystal on silicon (LCOS).5. Error factors of digital gray-scale-mask system using DMD are analyzed systematically. Methods of correction and compensation are given.6. MOE application in precision measurement is discussed. Two new ways of MOE application are put forward for the first time.(1) Application of beam shaping in edge location. Method of substituting energy compress for size compress of focus spot is put forward, which can be used to improve the sensitivity of edge location. Simulation and experiment have been done. The results show that new technique will bring higher sensitivity of intensity change and better linearity of measurement than half-focus technique.(2) Grating filtering method for small-angle diffractive noise. The output of the blazed grating is sensitive to the angle of the incident light. Using this characteristic, a new filtering method for small-angle diffractive noise is found. With blazed grating, the angle between the valid signal and diffractive noise can be magnified. Simulated results show that when the angle between the valid signal and diffractive noise is less than 5°, the angle difference can be enlarged three times if the grating period and blazed angle are optimized. This result can be used to reduce the difficulty of lowpass filter and avoid energy attenuation of valid signal.