248奈米深紫外光微影成像製程寬容度最適化

Abstract

實驗設計應用於微影成像技術有多項優點，最大的優點為在不增加額外製程成本下，以改變現有參數與製程設定，達成增加製程寬容度的目的。運用二階段實驗設計與多重品質特性能有效率分析可控因子對微影製程阻劑關鍵尺寸值的影響，且可自結果中決定一組最適合的製程參數設定。 本論文以田口玄一(Taguchi Genichi)實驗設計法，針對0.25微米 深紫外光微影製程，利用L8、L9、L18直交表研究相關阻劑參數與光學參數對微影製程的影響。以Hoechst AZ DX-1300p技術資料為比較標準，其聚焦深度為1.22微米；曝光寬容度為5.86%。模擬發現對0.25微米密集線/隙圖案而言，阻劑參數最適化設定為預烤溫度110 ℃、預烤時間70分鐘、曝後烤時間90秒、曝後烤溫度115℃、顯影時間40秒；光學參數最適化設定為內環相擾度當量值0.35、外環相擾度當量值0.7、內外環相擾度當量值比1/2、數值孔徑0.6、減光式相移圖罩透光率9%，再搭配圖案偏差式光學鄰近效應修正。最適化結果使聚焦深度自1.22微米增為1.66微米，增加0.44微米(增加率36%)；曝光寬容度自5.86%增為13.3%，增加7.44%(增加率127%)。 面對未來細線化製程的要求，變因亦隨之增加。製程相關參數的最適化，將是決 定微影技術成敗的關鍵。運用電腦模擬搭配田口實驗設計法，可有效率且整合性的分析與研究眾多製程參數。

Design of experiment (DOE) has several merits when applied to microlithography. The most important merit is that it could increase the process latitude by changing current parameters and process settings without increasing extra cost. Using two level DOE and multi-quality characteristic could efficiently analyze the influence of the controllable factors on the critical dimension of resist in microlithographic process, and obtain one set of the most suitable process parameters. In this thesis, the Taguchi Genichi methodology of experimental design has been applied by using L8, L9 and L18 orthogonal arrays to study the effects of resist parameters and optical parameters on the 248 nm deep-UV lithography processes for the 0.25 um linewidth patterns. Using Hoechst AZ DX-1300p resist technical data as a standard for comparison, in which the depth of focus is 1.22 um and the exposure latitude is 5.86%. Simulation of resist parameter optimization found out that prebake temperature 110℃, prebake time 70 minutes, post exposure bake time 90 seconds, post exposure bake temperature 115℃ and development time 40 seconds are the optimal parameters for 0.25 um dense line/space. Simulation of optical parameter optimization found out that inner ring sigma equivalent 0.35, outer ring sigma equivalent 0.7, ratio of inner ring sigma equivalent over outer ring sigma equivalent 1/2, numerical aperture 0.6, transmittance of attenuated phase shift mask 9% and using mask bias for optical proximity correction are the optimal parameters for 0.25 um dense line/space. The optimized result s indicated that depth of focus increased from 1.22 um to 1.66 um (increasing 36%); exposure latitude increased from 5.86% to 13.3% (increasing 127%). Face on the demanding of narrower linewidth processing in the future, the variances increase accordingly. The optimization of related process parameters is the key point for the success of microlithography technology. Using computer simulation combined with Taguchi DOE method could analyze and study various process parameters efficiently and integratedly.