針對多電子束直寫系統之鄰近效應修正

Abstract

隨著製程演進，超大型積體電路之最小關鍵尺寸已到達物理極限，傳統光學曝光使用的光源因解析度已到達極限而逐漸無法滿足需求，有許多下一代製程的曝光系統被提出，電子束曝光是其中具有潛力者。電子束因為其高解析度及精準性，可以精準地將電路資訊傳至曝光系統。許多在資料傳輸上遇到的問題，例如其產出的限制，可藉由大量平行的電子束系統來克服，資料壓縮的技術也能解決電子束曝光系統在傳輸方面的限制。然而，曝光過程中電子的各種效應仍需要解決，本篇論文主要針對鄰近效應，鄰近效應是電子的散射所造成的現象，會造成額外不需要的曝光，而使得曝光後的電路與實際電路有所 差距，這個差別可能會讓電路效能不彰，是個需要解決的議題。本篇論文將要探討的問題是，如何在不變更大量平行之電子束系統中的各項參數設定之下，改變原先的電路佈局，使其在正常曝光程序之後能與實際電路差距不遠，用以修正鄰近效應造成的影響，這個方法稱作為電子束鄰近效應修正。本篇論文提出的修正方法，大幅改善鄰近效應的影響。然而，此方法因故使原先電路佈局變得更加複雜，而造成壓縮效能降低。因此，修正的好壞程度必須與壓縮效能取得平衡點，本篇論文亦提出考慮壓縮效能的修正方法，得以證明兩者結合的演算法具有發展的潛力與研究價值。

The feature size of Integrated Circuits (IC) is getting smaller and smaller due to the desired continuation of Moore’s Law, but the resolution of today’s techniques is not enough. Hence, various candidates of next-generation lithography are presented. Electron beam lithography is a potential next-generation lithography because of its high resolution and accuracy. The drawback of e-beam lithography is its low throughp ut, which can be solved by applying a massive parallel writing system. In addition, compression algorithms are proposed to compensate the low speed of e-beam systems. However, some process effects are needed to be improved. Among them, proximity effect causes the original layout circuit differ from the exposed one, which may fail the whole circuit and is intolerable. In this thesis, we proposed rule-base electron beam proximity correction (EPC) methods to modify the original input circuits before the exposure process. The purpose is not to modify the parameters of the e-beam system while compensate the influence of proximity effect after exposure. There is a trade-off between correction and compression, and we conclude that these two techniques can be combined to find a balance between them.