Multistep virtual metrology approaches for semiconductor manufacturing processes

In semiconductor manufacturing, state of the art for wafer quality control relies on product monitoring and feedback control loops; the involved metrology operations are particularly cost-intensive and time-consuming. For this reason, it is a common practice to measure a small subset of a productive lot and devoted to represent the whole lot. Virtual Metrology (VM) methodologies are able to obtain reliable predictions of metrology results at process time; this goal is usually achieved by means of statistical models, linking process data and context information to target measurements. Since production processes involve a high number of sequential operations, it is reasonable to assume that the quality features of a certain wafer (e.g. layer thickness, electrical test results) depend on the whole processing and not only on the last step before measurement. In this paper, we investigate the possibilities to improve the VM quality relying on knowledge collected from previous process steps. We will present two different scheme of multistep VM, along with dataset preparation indications; special consideration will be reserved to regression techniques capable of handling high dimensional input spaces. The proposed multistep approaches will be tested against actual data from semiconductor manufacturing industry.

[1]  C.H. Yu,et al.  Virtual metrology: a solution for wafer to wafer advanced process control , 2005, ISSM 2005, IEEE International Symposium on Semiconductor Manufacturing, 2005..

[2]  Gian Antonio Susto,et al.  A Virtual Metrology system for predicting CVD thickness with equipment variables and qualitative clustering , 2011, ETFA2011.

[3]  Johan A. K. Suykens,et al.  Least Squares Support Vector Machine Classifiers , 1999, Neural Processing Letters.

[4]  Gian Antonio Susto,et al.  A predictive maintenance system based on regularization methods for ion-implantation , 2012, 2012 SEMI Advanced Semiconductor Manufacturing Conference.

[5]  R. Bellman,et al.  History and development of dynamic programming , 1984, IEEE Control Systems Magazine.

[6]  A. E. Hoerl,et al.  Ridge regression: biased estimation for nonorthogonal problems , 2000 .

[7]  Jerome H. Friedman,et al.  On Bias, Variance, 0/1—Loss, and the Curse-of-Dimensionality , 2004, Data Mining and Knowledge Discovery.

[8]  Shinji Okazaki,et al.  Pushing the limits of lithography , 2000, Nature.

[9]  Shane A. Lynn,et al.  Virtual metrology for plasma etch using tool variables , 2009, 2009 IEEE/SEMI Advanced Semiconductor Manufacturing Conference.

[10]  Jürgen Pilz,et al.  Dynamic Maintenance in semiconductor manufacturing using Bayesian networks , 2011, 2011 IEEE International Conference on Automation Science and Engineering.

[11]  Robert Tibshirani,et al.  The Elements of Statistical Learning: Data Mining, Inference, and Prediction, 2nd Edition , 2001, Springer Series in Statistics.

[12]  R. Tibshirani.  Regression Shrinkage and Selection via the Lasso , 1996 .

[13]  Giuseppe De Nicolao,et al.  Multilevel Kernel Methods for Virtual Metrology in Semiconductor Manufacturing , 2011 .

[14]  Giuseppe De Nicolao,et al.  Multilevel Lasso applied to Virtual Metrology in semiconductor manufacturing , 2011, 2011 IEEE International Conference on Automation Science and Engineering.