Device dependent control of chemical-mechanical polishing of dielectric films

Patel, Neil; Miller, Gerald A.; Guinn, C.; Sánchez, Altamirano; Jenkins, S.T.

doi:10.1109/66.857944

Cited by 29 publications

(17 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Normally, the calculation of weights of the feedforward model at th run is obtained by iteratively (run-to-run) minimizing the estimation error function as (3) where is the estimate error evaluated at at th run and is a vector with elements , and . Using the first order Taylor expansion of around the initial values , we obtain (4) for a small , where is the Jacobian matrix . Thus can be represented as where…”

Section: B Neural Network Trainingmentioning

confidence: 99%

“…We randomly pick up 16 runs of data among the 32 DOE data sets as the training data set. Specifically we pick runs 1,2,4,6,8,10,13,15,16,17,18,21,23,25, 26, and 28. By trying different neural network structures, we finally used one hidden layer neural network with 12 hidden neurons for the feedforward neural network process model, i.e.,…”

Section: B Performance Of the Neural Network Process Modelmentioning

confidence: 99%

“…The controller developed by [3] extended the predictor-corrector control (PCC) scheme in [2] and was applied to CMP processes. In [4], a concept of "sheet film equivalent" (SFE) was used to design a R2R CMP control scheme for dielectric applications, and the experimental results from Texas Instruments fab were compared with the EWMA control scheme. Statistical surface response methods (SRM) are widely used to model the relationship between process parameters and responses in semiconductor manufacturing.…”

Section: Hemical-mechanical Planarization (Cmp)mentioning

confidence: 99%

See 2 more Smart Citations

Neural network based uniformity profile control of linear chemical-mechanical planarization

2003

IEEE Trans. Semicond. Manufact.

View full text Add to dashboard Cite

In this paper, a neural network based uniformity controller is developed for the linear chemical-mechanical planarization (CMP) process. The control law utilizes the metrology measurements of the wafer uniformity profile and tunes the pressures of different air-bearing zones on Lam linear CMP polishers. A feedforward neural network is used to self-learn the CMP process model and a direct inverse control with neural network is utilized to regulate the process to the target. Simulation and experimental results are presented to illustrate the control system performance. Compared with the results by using statistical surface response methods (SRM), the proposed control system can give more accurate uniformity profiles and more flexibility.Index Terms-Chemical-mechanical planarization (CMP), neural networks, uniformity, run-to-run control.

show abstract

Section: B Neural Network Trainingmentioning

confidence: 99%

Section: B Performance Of the Neural Network Process Modelmentioning

confidence: 99%

Section: Hemical-mechanical Planarization (Cmp)mentioning

confidence: 99%

See 1 more Smart Citation

Neural network based uniformity profile control of linear chemical-mechanical planarization

2003

IEEE Trans. Semicond. Manufact.

View full text Add to dashboard Cite

show abstract

“…[1]. Patel et al suggest the volume of the wafer surface step as Sheet Film Equivalent (SFE) be used to express the product-by-product polishing characteristics relating to polishing volume.…”

Section: Introductionmentioning

confidence: 98%

CMP process run-to-run control to adjust margin in control limit

Morisawa

Abe

Tachikawa

et al. 2011

Precision Engineering

View full text Add to dashboard Cite

“…Many research has been focused on tool-induced variation, 1,4,5,8 and the product-induced variation has also been addressed recently. 6,9 In this work, we focus on two published removal rate models, namely CMP model based on topography factor 10 (simplified as TopFac model) and CMP model based on sheet film equivalent 6 (simplified as SFE model), and their application in CMP control. In Section 2, a brief review of these two models is given, and the equivalency between them is derived; In Section 3, a new control method is proposed based on the sheet film equivalent model.…”

Section: Introductionmentioning

confidence: 99%

A new run-to-run control method for oxide CMP process

Wang

2005

SPIE Proceedings

View full text Add to dashboard Cite

The main objective of the CMP run-to-run controller is to reduce the lot-to-lot variation in the post-polish oxide film thickness. Besides tool-induced variation, product-induced variation is also a significant source of variation to the CMP process. But the need to compensate for device pattern dependencies has not been addressed until recently. In this work, two removal rate models (topography factor model and sheet film equivalent model) are compared and the equivalency between them is derived. A new control method is proposed based on the sheet film equivalent model, which shows significant improvement compared to the traditional control method based on topography factor model. The performance of the proposed control algorithm is demonstrated using both simulated and industrial examples.

show abstract

Device dependent control of chemical-mechanical polishing of dielectric films

Cited by 29 publications

References 7 publications

Neural network based uniformity profile control of linear chemical-mechanical planarization

Neural network based uniformity profile control of linear chemical-mechanical planarization

CMP process run-to-run control to adjust margin in control limit

A new run-to-run control method for oxide CMP process

Contact Info

Product

Resources

About