Run-by-run process control of metal sputter deposition: Combining time series and extended Kalman filter

Juhn Horng Chen*, Tzu Wei Kuo, An-Chen Lee

*Corresponding author for this work

Research output: Contribution to journalArticle

16 Scopus citations

Abstract

By the time series model, this paper constructed the disturbance model for the aluminum sputter deposition process and derived the extending Kalman filter (EKF) controller based on this new disturbance model. Experimental results reveal that ARI(3,1) model appropriately characterizes the dynamic behavior of the disturbance for the processes. The EKF controller which includes information of process noise and measurement noise is able to regulate the model coefficients automatically as the target is replaced or degrades. In this paper, the d-EWMA controller, time-varying d-EWMA controller, age-based d-EWMA controller, and EKF controller have been applied to aluminum sputter deposition processes for predicting deposition rates and comparing their performances. The application of the EKF controller here is proven to improve the estimating accuracy of the aluminum sputter deposition process significantly, regardless of whether the deposition rates are measured at each run or not.

Original languageEnglish
Pages (from-to)278-285
Number of pages8
JournalIEEE Transactions on Semiconductor Manufacturing
Volume20
Issue number3
DOIs
StatePublished - 1 Aug 2007

Keywords

  • D-EWMA controller
  • Deposition rate
  • Extended Kalman filter
  • Time series model
  • Time-varying d-EWMA controller

Fingerprint Dive into the research topics of 'Run-by-run process control of metal sputter deposition: Combining time series and extended Kalman filter'. Together they form a unique fingerprint.

  • Cite this