Optical Proximity Correction Solutions Research Articles

Inverse lithography physics-informed deep neural level set for mask optimization.

As the feature size of integrated circuits continues to decrease, optical proximity correction (OPC) has emerged as a crucial resolution enhancement technology for ensuring high printability in the lithography process. Recently, level set-based inverse lithography technology (ILT) has drawn considerable attention as a promising OPC solution, showcasing its powerful pattern fidelity, especially in advanced processing. However, the massive computational time consumption of ILT limits its applicability to mainly correcting partial layers and hotspot regions. Deep learning (DL) methods have shown great potential in accelerating ILT. However, the lack of domain knowledge of inverse lithography limits the ability of DL-based algorithms in process window (PW) enhancement, etc. In this paper, we propose an inverse lithography physics-informed deep neural level set (ILDLS) approach for mask optimization. This approach utilizes level set-based ILT as a layer within the DL framework and iteratively conducts mask prediction and correction to significantly enhance printability and PW in comparison with results from pure DL and ILT. With this approach, the computational efficiency is significantly improved compared with ILT. By gearing up DL with the knowledge of inverse lithography physics, ILDLS provides a new and efficient mask optimization solution.

Can optical proximity correction solution be learned? The learning limit and a general learning framework

BackgroundOptical proximity correction (OPC) is an indispensable technology that has been propelling the advancement of computational lithography technology. To tightly control edge placement error (EPE) and maintain lithography process window, the demands on OPC computational resources and OPC turnaround time are growing rapidly with alarming acceleration. To tame the trend, machine learning technologies have been explored; however, an in-depth discussion on OPC solution learning limit is still lacking.AimWe aim to present an in-depth discussion on OPC solution learning limit and propose a general machine learning OPC framework that can be extended to curvilinear mask OPC technology.ApproachIn this study, we first investigate the machine learning OPC learning limit by examining noninverse lithography technology (non-ILT) OPC solution space characteristics inherited from edge segmentation and control point setting rules and then propose a general machine learning OPC framework that can take full advantage of deep convolution neural network (DCNN) learning capability while being able to preserve mask data high resolution.ResultsWith this machine learning OPC framework, we have achieved models with average absolute model error <1 nm for 14-nm metal layer. With single GPU, the average time for machine learning OPC models to produce results of 3840 nm × 3840 nm area is 8.74 ms for single channel input model and 12.65 ms for six channels input model.ConclusionsFor non-ILT OPC solution, there is an intrinsic learning limit inherited from edge segmentation rules. Machine learning OPC models should be content with learning low order OPC solutions. This intrinsic learning limit of non-ILT OPC solution may diminish for ILT OPC solution when the constraint on degrees of freedom of OPC solution is lifted. The machine learning OPC framework we proposed is general and extendable to curvilinear OPC technology.

DAMO: Deep Agile Mask Optimization for Full-Chip Scale

Continuous scaling of the very-large-scale integration system leaves a significant challenge on manufacturing; thus optical proximity correction (OPC) is widely applied in conventional design flow for manufacturability optimization. Traditional techniques conduct OPC by leveraging a lithography model but may suffer from prohibitive computational overhead. In addition, most of them focus on optimizing a single and local clip instead of addressing how to tackle the full-chip scale. In this article, we present DAMO, a high-performance and scalable deep-learning-enabled OPC system for full-chip scale. It is an end-to-end mask optimization paradigm that contains a deep lithography simulator (DLS) for lithography modeling and a deep mask generator (DMG) for mask pattern generation. Moreover, a novel layout splitting algorithm customized for DAMO is proposed, composed of DBSCAN clustering and KMeans++ clustering, to handle the full-chip OPC problem. Further, graph-based computation and parallelism techniques are proposed to deploy our GPU algorithms to accelerate computations. Extensive experiments show that DAMO outperforms state-of-the-art OPC solutions in both academia and industrial commercial toolkit.

Neural-ILT 2.0: Migrating ILT to Domain-Specific and Multitask-Enabled Neural Network

Optical proximity correction (OPC) in modern design closures has become extremely expensive and challenging. Conventional model-based OPC encounters performance degradation and large process variation, while aggressive approach, such as inverse lithography technology (ILT), suffers from large computational overhead for both mask optimization and mask writing processes. In this article, we developed Neural-ILT, an end-to-end learning-based OPC framework, which literally conducts mask prediction and ILT correction for a given layout in a single neural network, with the objectives of: 1) mask printability enhancement; 2) mask complexity optimization; and 3) flow acceleration. A domain-specific model pretraining recipe, which introduces the domain knowledge of lithography system, is proposed to help Neural-ILT achieving faster and better convergence. Quantitative results show that compared to the state-of-the-art (SOTA) learning-based OPC solutions and conventional OPC flows, Neural-ILT can achieve <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$15\times $ </tex-math></inline-formula> to <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$30\times $ </tex-math></inline-formula> turnaround time (TAT) speedup and the best mask printability with relatively lower mask complexity. Based on the developed infrastructure, we further investigated the feasibility of handling multiple mask optimization tasks for different datasets within a common Neural-ILT platform. We believe this work could bridge well-developed deep learning toolkits to GPU-based high-performance lithographic computations to achieve groundbreaking performance boosting on various computational lithography-related tasks.

Information theoretical approaches in computational lithography.

Computational lithography is nowadays playing an indispensible role in improving the imaging performance of optical lithography systems. This paper develops a new and powerful approach to computational lithography by introducing an information theoretical channel modeling in partially coherent lithography systems. A statistical model is built up based on the lithography imaging model to characterize the information transfer between the mask and print images. Then, this paper calculates the optimal information transfer (OIT) in partially coherent lithography systems, and derives the theoretical limit of image fidelity for optical proximity correction (OPC), which is used extensively in computational lithography. Finally, the proposed information theoretical approaches are applied to improve the OPC solutions obtained by the gradient-based algorithm. A set of simulations are provided to verify the proposed information theoretical model and approaches.

Computational algorithms for optimizing mask layouts in proximity printing

We use numerical simulation and optimization algorithms to apply optical proximity correction (OPC) to different mask structures for exposure on a Mask Aligner in proximity mode. The range of OPC solutions reaches from subtle changes in the mask layout to masks that have lost almost all resemblance to the structure to be printed. An extreme case is the printing of periodic structures where we make use of the image reversal caused by the Talbot effect. We also show the improvement that can be obtained when co-optimizing the angular spectrum of the illumination, i.e. when performing a source-mask optimization (SMO).