Shanghai Optical Machinery Institute has made important progress in computational lithography technology research

According to an article published on the official website of the Chinese Academy of Sciences, Shanghai Optical Machinery Institute has made important progress in computational lithography technology research.

Recently, the Information Optics and Optoelectronic Technology Laboratory of Shanghai Institute of Optics and Fine Mechanics, Chinese Academy of Sciences proposed a fast optical proximity effect correction based on virtual edge and mask pixelation with two-phase sampling. Technology (Optical proximity correction, OPC). The simulation results show that the technology has higher correction efficiency.

Lithography is one of the key technologies in the manufacture of very large-scale integrated circuits. The resolution of lithography determines the feature size of integrated circuits. As the feature size of integrated circuit patterns continues to decrease, the diffraction-limited properties of lithography systems lead to obvious optical proximity effects, which reduce the quality of lithography imaging.

Under the condition that the software and hardware of the lithography machine remain unchanged, the mathematical model and software algorithm is used to optimize the lighting mode, mask pattern, and process parameters, which can effectively improve the resolution of lithography and increase the process window. Such technologies are calculations. Lithography (Computational Lithography) is considered to be a new driving force for the continued development of integrated circuit chips in accordance with Moore's Law.

OPC technology corrects the optical proximity effect by adjusting the transmittance distribution of the mask pattern, thereby improving the imaging quality. Model-based OPC technology is one of the key computational lithography technologies to realize integrated circuit manufacturing at technology nodes of 90nm and below.

The rapid optical proximity effect correction technology based on virtual edges and dual-sampling rate pixelated mask patterns proposed by the researchers of Shanghai Institute of Optics and Mechanics can attribute different types of imaging distortions to two types of imaging abnormalities, namely shrinkage abnormalities. And the external expansion is abnormal.

Using different imaging anomaly detection templates, perform local imaging anomaly detection at the contour offset judgment positions such as the edges and corners of the mask pattern in turn to determine the type of abnormality and the range of the abnormal area.

According to the abnormal detection position and the abnormal area range, the virtual edge is adaptively generated. The local transmittance distribution of the mask is adjusted by moving the virtual edge to correct local imaging abnormalities. With the help of correction strategies and correction constraints, efficient local correction and global contour fidelity control are realized.

In addition, the dual sampling rate pixelated mask makes full use of the diffraction-limited properties of the imaging system performs imaging calculations and abnormal detection on the coarse sampling grid and performs mask correction on the fine sampling grid, which takes into account the efficiency of imaging calculations and Masks correction resolution.

Using a variety of mask graphics to verify, the simulation results show that the correction efficiency of the OPC technology is better than the commonly used OPC technology based on heuristic algorithms.