Ko-Optimasi Source dan Pola Mask Berdasarkan Algoritma Multi-Objektif Particle Swarm Optimization

Arthananda, Zendhiastara (2017) Ko-Optimasi Source dan Pola Mask Berdasarkan Algoritma Multi-Objektif Particle Swarm Optimization. Masters thesis, Institut Teknologi Sepuluh Nopember.

[img] Text
1214201040-Master_Thesis.pdf
Restricted to Repository staff only

Download (3MB) | Request a copy

Abstract

Studi ini mengintegrasikan algoritma multi-objektif particle swarm optimization (MOPSO) kedalam proses ko-optimasi source dan mask (SMO) untuk meningkatkan performa lithografi pada sinar ekstrim ultraviolet (EUV). Sebuah metode proses secara simultan dari source dan pola reticle dikembangkan pada riset ini. Untuk konstruksi source berbentuk bebas (freeform) , sebuah optimasi berbasis pixel digunakan pada platform PC. Algoritma MOPSO digunakan untuk menghasilkan source berbentuk bebas (Source Freform). Model berbasis pendekatan koreksi optik (Optical Proximity Correction or OPC) digunakan untuk mengoreksi pola dari mask layout. Dengan mempertimbangkan karakteristik dari sistem lithografi EUV, metode SMO dikembangkan dengan algoritma MOPSO menggunakan dua fungsi tujuan: error (EPE) dan bias horizontal/vertikal. Sebuah pola satu-dimensi line/space (L/S) digunakan sebagai informasi dasar untuk menguji Pareto dari algoritma SMO. Kemudian, pola 2D dengan half-pitch 22-nm diuji menggunakan algoritma yang sama. Algoritma MOPSO berhasil untuk mengkonstruksi solusi non-dominan (non-dominated) dari source Freeform dan Pareto. Indikator performa menunjukkan kondisi process windows (PW) seperti aerial image, exposure latitutde (EL), depth of focus (DOF) dan bias. Algoritma menunjukan bahwa PW meningkat untuk EL namun DOF menunjukkan penurunan. EL meningkat sebesar 5.26% dan DOF menurun sebesar 11.34% untuk 1D L/S. EL dan DOF meningkat 43.6% dan 18.11% untuk pola 2D. ============================================================================================ This thesis integrates multi - objective particle swarm optimization (MOPSO) algorithm into the source and mask co - optimization (SMO) process to enhance the extreme ultraviolet (EUV) lithography imaging performance. A simultaneous source and reticle pattern process method is developed in this research. For the freeform source construction, a pixelated - based optimiz ation process was performed on PC platform. The MOPSO algorithm was applied to generate freeform source. Model - based optical proximity correction (OPC) was applied to correct the mask layout patterns. Considering the characteristics of the EUV lithography system, the developed SMO with the MOPSO algorithm is constrained by two cost functions: the edge placement error (EPE) and horizontal/vertical bias. A one - dimensional line/space (L/S) pattern is used as the baseline information to test the Pareto behavior of the developed SMO algorithm. Then, the 2D pattern with half - pitch 22 - nm was assessed using the developed algorithm. The proposed MOPSO algorithm succeeded to construct non - dominated solutions of freeform sources and Pareto front which four of those sol utions are presented. The performance indicators include process windows (PW) condition such as the aerial image contrast, exposure latitude (EL), depth of focus (DOF), and bias errors. The proposed algorithm shows that the common PW conditions improved on EL while the DOF is slightly suffering. The EL increased for 5.26% and DOF suffers for 11.34% in 1D L/S and both EL and DOF increased for 43.6% and 18.11%, respectively for the 2D pattern.

Item Type: Thesis (Masters)
Uncontrolled Keywords: lithografi ekstrim ultraviolet (EUV), multi-objektif particle swarm optimization (MOPSO), process window (PW), optimasi source dan mask (SMO), extreme ultraviolet (EUV) lithography, source mask optimization
Subjects: Q Science > QA Mathematics
Q Science > QA Mathematics > QA76.9 Computer algorithms. Virtual Reality. Computer simulation.
Divisions: Faculty of Mathematics and Science > Mathematics > (S2) Master Theses
Depositing User: Zendhiastara Arthananda
Date Deposited: 22 Feb 2018 01:46
Last Modified: 22 Feb 2018 01:46
URI: http://repository.its.ac.id/id/eprint/50891

Actions (login required)

View Item View Item