蔡坤諭臺灣大學:電機工程學研究所伍海濤Ng, Hoi-TouHoi-TouNg2010-07-012018-07-062010-07-012018-07-062009U0001-1408200915452900http://ntur.lib.ntu.edu.tw//handle/246246/188113低能電子微影術擁有高解析度，低基底損害，增加光阻敏感度的優勢，被認為實踐22奈米製程的技術之一。為了在高斯電子束柵狀掃描系統下增加製程的產量，電子的曝光量，曝光之距離和曝光點大小等參數必須謹慎選取，以達到產量的最佳化。為此我們以一個蒙地卡羅的模擬方法去模擬出光阻的曝光形狀，根據邊緣粗糙度對ITRS上的標準值的比較而把以上所述的參數的邊界找出。同時把曝光形狀作為電子光學系統的判定條件，對電子光學系統的參數作出進一步的最佳化。Low-energy electron beam lithography (LEEBL) is a promising patterning solution for the 22-nm half-pitch node and beyond due to its high resolution, low substrate damage, and increased resist sensitivities. In order to achieve throughout required for high-volume manufacturing, writing parameters such as probe size, pixel size, electron dosage, proximity correction scheme, and number of beams need to be carefully selected in a Gaussian-beam–raster-scan system. In high-throughput LEEBL, line edge roughness (LER) caused by shot noise becomes a critical issue for both device patterning and device performance variability. To characterize these effects, stochastic MOSFET gate patterning with LEEBL is constructed by overlapping energy distributions from an in-house electron scattering Monte Carlo simulation program with various writing parameters. Parameter optimization can help provide initial guidelines and specifications for design and operation of multiple electron beam direct-write systems.Table of Contentsbstract I要 IItatement of Contributions III謝 IVable of Contents Vist of Figures VIIist of Tables IXhapter 1 Introduction 1.1 Microlithography 1.2 Electron Beam Lithography 2.3 Multiple Electron Beam Lithography 4hapter 2 Monte Carlo simulation 6.1 Monte Carlo method 6.2 Scattering phenomena in solid 6.2.1 Polar angle 7.2.2 Azimuth angle 7.2.3 Distance between two collisions 8.2.4 Energy loss 8.2.5 Collision determination 10.3 Point spread function comparison with CASINO 11.4 The simulation time analysis of the original simulation structure and Parallel simulation structure 13.4.1 Algorism improve 13.4.2 Parallel Algorism 17.4.3 Acceleration Results 18hapter 3 Exposure and latent image Electron beam exposure process 20.1 Simulation model 20.1.1 Electron beam spot distribution 22.1.2 Point spread distribution 22.1.3 Threshold level and latent image 23.1.4 Diffusion model 24.2 Patterns parameters definition 26.2.1 Parameters definition 27.2.1.1 Spot size 27.2.1.2 Grid size 27.2.1.3 Dosage 28.2.2 Patterns’ latent image simulation results 28.3 Direct Monte Carlo method 29.3.1 Different of DMC and pervious method in pattern simulation 30hapter 4 Parameter selection methodology 33.1 Parameters boundary finding methodology 33.2 Process of the methodology 33.3 Simulation with DMC and hybrid method including PAG diffusion 35hapter 5 Lens parameters optimization including the whole e-beam system 38.1 Preliminary EOS for MPML2 38.2 Whole system simulation flow 40.2.1 New random variable generator 41.2.2 Source + Enizel lens as an operator 45.3 Parameters definition 47.4 Optimization 49hapter 6 Future Work 53eference 54IOGRAPHIES 581616311 bytesapplication/pdfen-US低能電子微影術邊緣粗糙度蒙地卡羅方法最佳化low energy electron beam lithographyLine Edge RoughnessMonte Carlo methodoptimizationthesishttp://ntur.lib.ntu.edu.tw/bitstream/246246/188113/1/ntu-98-R96921013-1.pdf