ABSTRACT
Lithography simulation is one of the key steps in physical verification, enabled by the substantial optical and resist models. A resist model bridges the aerial image simulation to printed patterns. While the effectiveness of learning-based solutions for resist modeling has been demonstrated, they are considerably data-demanding. Meanwhile, a set of manufactured data for a specific lithography configuration is only valid for the training of one single model, indicating low data efficiency. Due to the complexity of the manufacturing process, obtaining enough data for acceptable accuracy becomes very expensive in terms of both time and cost, especially during the evolution of technology generations when the design space is intensively explored. In this work, we propose a new resist modeling framework for contact layers that utilizes existing data from old technology nodes to reduce the amount of data required from a target lithography configuration. Our framework based on residual neural networks and transfer learning techniques is effective within a competitive range of accuracy, i.e., 2-10X reduction on the amount of training data with comparable accuracy to the state-of-the-art learning approach.
- Martín Abadi, Ashish Agarwal, Paul Barham, Eugene Brevdo, Zhifeng Chen, et al. 2015. TensorFlow: Large-Scale Machine Learning on Heterogeneous Systems. (2015). https://www.tensorflow.orgGoogle Scholar
- Yoshua Bengio, Patrice Simard, and Paolo Frasconi. 1994. Learning long-term dependencies with gradient descent is difficult. IEEE transactions on neural networks 5, 2 (1994), 157--166. Google ScholarDigital Library
- Xavier Glorot and Yoshua Bengio. 2010. Understanding the difficulty of training deep feedforward neural networks. In Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics. 249--256.Google Scholar
- Ian Goodfellow, Yoshua Bengio, and Aaron Courville. 2016. Deep learning. MIT press. Google ScholarDigital Library
- Allan Gu and Avideh Zakhor. 2008. Optical proximity correction with linear regression. IEEE Transactions on Semiconductor Manufacturing (TSM) 21, 2 (2008), 263--271.Google ScholarCross Ref
- Josiah P Hanna and Peter Stone. 2017. Grounded Action Transformation for Robot Learning in Simulation. In AAAI. 3834--3840.Google Scholar
- Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. 2016. Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition. 770--778.Google ScholarCross Ref
- Ningning Jia and Edmund Y Lam. 2010. Machine learning for inverse lithography: using stochastic gradient descent for robust photomask synthesis. Journal of Optics 12, 4 (2010), 045601.Google ScholarCross Ref
- Diederik Kingma and Jimmy Ba. 2014. Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980 (2014).Google Scholar
- Lars Liebmann, Albert Chu, and Paul Gutwin. 2015. The daunting complexity of scaling to 7nm without EUV: Pushing DTCO to the extreme. In Proceedings of SPIE, Vol. 9427.Google Scholar
- Lars Liebmann, Jia Zeng, Xuelian Zhu, Lei Yuan, Guillaume Bouche, and Jongwook Kye. 2016. Overcoming scaling barriers through design technology CoOptimization. In VLSI Technology, 2016 IEEE Symposium on. IEEE, 1--2.Google ScholarCross Ref
- Lars W Liebmann, Scott M Mansfield, Alfred K Wong, Mark A Lavin, William C Leipold, and Timothy G Dunham. 2001. TCAD development for lithography resolution enhancement. IBM Journal of Research and Development 45, 5 (2001), 651--665. Google ScholarDigital Library
- Yibo Lin, Xiaoqing Xu, Jiaojiao Ou, and David Z Pan. 2017. Machine learning for mask/wafer hotspot detection and mask synthesis. In Photomask Technology, Vol. 10451. International Society for Optics and Photonics, 104510A.Google Scholar
- Rui Luo. 2013. Optical proximity correction using a multilayer perceptron neural network. Journal of Optics 15, 7 (2013), 075708.Google ScholarCross Ref
- Xu Ma, Xuejiao Zhao, Zhiqiang Wang, Yanqiu Li, Shengjie Zhao, and Lu Zhang. 2017. Fast lithography aerial image calculation method based on machine learning. Applied Optics 56, 23 (2017), 6485--6495.Google ScholarCross Ref
- Tetsuake Matsunawa, Shigeke Nojima, and Toshiya Kotani. 2016. Automatic Layout Feature Extraction for Lithography Hotspot Detection Based on Deep Neural Network. In Proceedings of SPIE.Google Scholar
- Tetsuaki Matsunawa, Bei Yu, and David Z Pan. 2016. Optical proximity correction with hierarchical Bayes model. Journal of Micro/Nanolithography, MEMS, and MOEMS 15, 2 (2016), 021009--021009.Google ScholarCross Ref
- Mentor Graphics. 2008. Calibre Verification User's Manual. (2008).Google Scholar
- Sinno Jialin Pan and Qiang Yang. 2010. A survey on transfer learning. IEEE Transactions on knowledge and data engineering 22, 10 (2010), 1345--1359. Google ScholarDigital Library
- Andrei A Rusu, Neil C Rabinowitz, Guillaume Desjardins, Hubert Soyer, James Kirkpatrick, Koray Kavukcuoglu, Razvan Pascanu, and Raia Hadsell. 2016. Progressive neural networks. arXiv preprint arXiv:1606.04671 (2016).Google Scholar
- Seongbo Shim, Suhyeong Choi, and Youngsoo Shin. {n. d.}. Machine Learning-Based Resist 3D Model. In Proc. of SPIE Vol, Vol. 10147. 101471D--1.Google Scholar
- Moojoon Shin and Jee-Hyong Lee. 2016. Accurate Lithography Hotspot Detection Using Deep Convolutional Neural Networks. In Journal of Micro/Nanolithography, MEMS, and MOEMS (JM3).Google Scholar
- Synopsys. 2016. Sentaurus Lithography. https://www.synopsys.com/silicon/mask-synthesis/sentaurus-lithography.html. (2016).Google Scholar
- Chin Boon Tan, Kar Kit Koh, Dongqing Zhang, and Yee Mei Foong. 2015. Sub-resolution assist feature (SRAF) printing prediction using logistic regression. In Proceedings of SPIE. 94261Y--94261Y.Google Scholar
- Yuki Watanabe, Taiki Kimura, Tetsuaki Matsunawa, and Shigeki Nojima. 2017. Accurate lithography simulation model based on convolutional neural networks. In SPIE Advanced Lithography. International Society for Optics and Photonics, 101470K--101470K.Google Scholar
- Jen-Yi Wuu, Fedor G Pikus, and Malgorzata Marek-Sadowska. 2011. Efficient approach to early detection of lithographic hotspots using machine learning systems and pattern matching. In SPIE Advanced Lithography. International Society for Optics and Photonics, 79740U--79740U.Google Scholar
- Xiaoqing Xu, Yibo Lin, Meng Li, Tetsuaki Matsunawa, Shigeki Nojima, Chikaaki Kodama, Toshiya Kotani, and David Z. Pan. 2017. Sub-Resolution Assist Feature Generation with Supervised Data Learning. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems (TCAD) PP, 99 (2017).Google Scholar
- Haoyu Yang, Yajun Lin, Bei Yu, and F.Y. Evangeline Young. 2017. Lithography Hotspot Detection: From Shallow to Deep Learning. In IEEE International System-on-Chip Conference (SOCC).Google Scholar
- Haoyu Yang, Jing Su, Yi Zou, Bei Yu, and F.Y. Evangeline Young. 2017. Layout Hotspot Detection with Feature Tensor Generation and Deep Biased Learning. In ACM/IEEE Design Automation Conference (DAC). Google ScholarDigital Library
Index Terms
- Data Efficient Lithography Modeling with Residual Neural Networks and Transfer Learning
Recommendations
Fine pattern transfer by nanocasting lithography
A novel pattern transfer technique has been demonstrated by simple procedure based on a nanoimprint lithography. A polymer is coated on a mold by a spin coater and cast into fine grooves on the mold. After evaporation of a solvent, an adhesive resin is ...
Fabrication of hp 32nm resist patterns using near-field lithography
Near-field lithography (NFL) has no fundamental limit such as the diffraction limit of light. However, in order to fabricate resist patterns with hp 32nm, thorough optimization of various processes are indispensable. Previously, we reported on the use ...
A non-destructive method for the removal of residual resist in imprinted patterns
In this paper we describe a bi-layer resist imprinting procedure in which the residual resist remaining after the imprint step was removed by selective dissolution. This eliminates the risk of dry etch damage to the substrate or pattern and variations ...
Comments