Low-rank manifold optimization for overlay variations in lithography process

Abstract Overlay variations occur frequently in lithography process, which should be controlled within the tolerance to guarantee the better pattern resolutions. The operational optimization of overlay aims to predict the unknown overlay variations and compensate them into the wafer production. Due to the uncertain yield, the overlay data for learning are usually incomplete, which makes the overlay optimization very challenging. This paper proposes a novel overlay optimization framework called low-rank manifold optimization (LRMO), which provides new insight to address incomplete overlay data via exploiting low-rank property. First, LRMO can use effectively the correlations from incomplete overlay data, which builds a low-rank model for overlay optimization. In addition, LRMO resorts to Riemannian optimization and designs an efficient algorithm for this low-rank model. The proposed LRMO algorithm analyzes the manifold structure of the overlay data and computes accurate overlay variations with a low computational complexity. The experiments validate that LRMO obtains satisfying performance on the operational optimization of overlay variations.

[1]  Thomas F. Edgar,et al.  A new state estimation method for high-mix semiconductor manufacturing processes , 2009 .

[2]  Enrique del Castillo,et al.  Closed-Loop Disturbance Identification and Controller Tuning for Discrete Manufacturing Processes , 2002, Technometrics.

[3]  Sung-Shick Kim,et al.  Neural Networks Fusion to Overlay Control System for Lithography Process , 2006, Australian Conference on Artificial Intelligence.

[4]  An-Chen Lee,et al.  Combined product and tool disturbance estimator for the mix-product process and its application to the removal rate estimation in CMP process , 2012 .

[5]  Shi-Shang Jang,et al.  An Adaptive-Tuning Scheme for G&P EWMA Run-to-Run Control , 2012, IEEE Transactions on Semiconductor Manufacturing.

[6]  N. Patel,et al.  Adaptive optimization of run-to-run controllers: the EWMA example , 2000 .

[7]  Xian-Da Zhang,et al.  Matrix Analysis and Applications , 2017 .

[8]  Zone-Ching Lin,et al.  Multiple linear regression analysis of the overlay accuracy model , 1999, ICMTS 1999.

[9]  Kwang-Hyun Cho,et al.  Run-to-run overlay control of steppers in semiconductor manufacturing systems based on history data analysis and neural network modeling , 2005 .

[10]  Paul M. Thompson,et al.  Multi-source feature learning for joint analysis of incomplete multiple heterogeneous neuroimaging data , 2012, NeuroImage.

[11]  Thomas F. Edgar,et al.  State estimation in high-mix semiconductor manufacturing , 2009 .

[12]  Ying Zheng,et al.  Cycle forecasting EWMA (CF-EWMA) approach for drift and fault in mixed-product run-to-run process , 2010 .

[13]  Charles R. Johnson,et al.  Matrix analysis , 1985, Statistical Inference for Engineers and Data Scientists.

[14]  Jan Mulkens,et al.  Overlay improvements using a real time machine learning algorithm , 2014, Advanced Lithography.

[15]  Shih-Cheng Horng,et al.  Compensating the overlay modeling errors in lithography process of wafer stepper , 2010, 2010 5th IEEE Conference on Industrial Electronics and Applications.

[16]  Fugee Tsung,et al.  A study of variable EWMA controller , 2003 .

[17]  Costas J. Spanos,et al.  One step forward from run-to-run critical dimension control: Across-wafer level critical dimension control through lithography and etch process , 2008 .

[18]  Robert D. Nowak,et al.  Transduction with Matrix Completion: Three Birds with One Stone , 2010, NIPS.

[19]  Kevin Ryan,et al.  Virtual overlay metrology for fault detection supported with integrated metrology and machine learning , 2015, Advanced Lithography.

[20]  Emmanuel J. Candès,et al.  Exact Matrix Completion via Convex Optimization , 2009, Found. Comput. Math..

[21]  Emmanuel J. Candès,et al.  A Singular Value Thresholding Algorithm for Matrix Completion , 2008, SIAM J. Optim..

[22]  Shih-Cheng Horng.  Compensating Modeling Overlay Errors Using the Weighted Least-Squares Estimation , 2014, IEEE Transactions on Semiconductor Manufacturing.

[23]  Richard van Haren,et al.  Pattern recognition and data mining techniques to identify factors in wafer processing and control determining overlay error , 2015, Advanced Lithography.

[24]  Robert E. Mahony,et al.  Optimization Algorithms on Matrix Manifolds , 2007 .

[25]  Shi-Shang Jang,et al.  Stability and performance analysis of mixed product run-to-run control , 2006 .

[26]  Shiqian Ma,et al.  Fixed point and Bregman iterative methods for matrix rank minimization , 2009, Math. Program..

[27]  Hyo-Heon Ko,et al.  Advanced semiconductor fabrication process control using dual filter exponentially weighted moving average , 2012, J. Intell. Manuf..

[28]  Frank Wang,et al.  A 3-tier cooperative control architecture for multi-step semiconductor manufacturing process , 2008 .

[29]  Guy Cohen,et al.  Weighted least squares regression for advanced overlay control , 2012, Advanced Lithography.

[30]  Tianhong Pan,et al.  A G&P EWMA algorithm for high-mix semiconductor manufacturing processes , 2011 .

[31]  Chen-Fu Chien,et al.  MODELING OVERLAY ERRORS AND SAMPLING STRATEGIES TO IMPROVE YIELD , 2001 .