×

Modeling and control of a semiconductor manufacturing process with an automata network: An example in plasma etch processing. (English) Zbl 0847.90073

Summary: A neural network model of a plasma gate etch process is described and compared with a statistical model. The neural network model has a correlation of 0.68 while the statistical model has a correlation of 0.45. From our model we deduce that the flow rate of the etching gases and the induced d.c.-bias are the key factors driving the etching and thus the remaining oxide thickness at the end of the etch. An adaptive neural network controller for wafer-to-plasma etch control is also described. It uses real time process signatures and historical data from a relational database for the computation of the overetch time for the current wafer etching within the reactor. For an MOS gate etch the standard deviation of the oxide thickness between the gate and the source (or drain) is in the range of 10Å. This is comparable to open-loop control or timed etch where the operator selects the ideal overetch time. The controller has thus achieved better than human equivalence.

MSC:

90B30 Production models
90B90 Case-oriented studies in operations research
92B20 Neural networks for/in biological studies, artificial life and related topics
PDFBibTeX XMLCite
Full Text: DOI

References:

[1] Jenkins, M. W.; Mocella, M. T.; Allen, K. D.; Swain, H. H., The modeling of plasma etching processes using response surface methodology, Solid State Technol., 175-182 (1986), April
[2] Chapman, B., Glow Discharge Processes: Sputting and Plasma Etching (1980), Wiley: Wiley New York
[3] van Roosmalen, A. J.; Baggerman, J. A.G.; Brader, S. J.H., Dry Etching for VLSI (1991), Plenum Press: Plenum Press New York
[4] Manos, D. M.; Flamm, D. L., Plasma Etching an Introduction (1989), Academic Press: Academic Press New York
[5] Rietman, E. A.; Patel, S. H.; Lory, E. R., Neural network control of a plasma gate etch: early steps in wafer-to-wafer process control, (Fifteenth IEEE/CHMT Int. Electronics Manufacturing Technol. Symp.. Fifteenth IEEE/CHMT Int. Electronics Manufacturing Technol. Symp., Santa Clara, CA (4-6 October 1993)), 454-457
[6] W. E. Willenbock Jr, Personal communication.; W. E. Willenbock Jr, Personal communication.
[7] Curtis, B. J.; Brunner, H. R., The reactive ion etching of tantalum silicide polysilicon bilayers in silicon tetrachloride and hydrogen chloride, J. Electrochem. Soc., 136, 1463-1468 (1989)
[8] Hornik, K.; Stinchcombe, M.; White, H., Universal approximation of an unknown mapping and its derivatives using multilayer feedforward networks, Neural Networks, 3, 551-560 (1990)
[9] Rumelhart, D. E.; Hinton, G. E.; Williams, R. J., Learning internal representations by error propagation, (Rumelhart; McClelland, Parallel Distributed Processing: Explorations in the Microstructure of Cognition, Vol. 1. Foundations (1986), MIT Press: MIT Press Cambridge, MA)
[10] Weiss, S. M.; Kukikowski, C. A., Computer Systems that Learn (1991), Morgan Kaufmann: Morgan Kaufmann San Mateo, CA
[11] McCulloch, W. S.; Pitts, W., A logical calculus of the ideas immanent in nervous activity, Bull. Math. Biophys., 5, 115-133 (1943) · Zbl 0063.03860
[12] Widrow, B.; Hoff, M. E., Adaptive Switching Circuits, ((1960), IRE WESCON, Convention Record IRE: IRE WESCON, Convention Record IRE New York), 96-104
[13] Widrow, B.; Hoff, M. E., Associative storage and retrieval of digital information in networks of adaptive neurons, (Bernard, E. E.; Kane, M. R., Biol. Prototypes Synthe. Syst., Vol. 1 (1962), Plenum Press: Plenum Press New York)
[14] Minsky, M. L.; Papert, S. A., Perceptions: An Introduction to Computational Geometry (1969), MIT Press: MIT Press Cambridge, MA · Zbl 0197.43702
[15] le Cun, Y., Learning process in an asymmetric threshold network, (Bienestock, E.; Soulie, F. Fogelman; Weisbach, G., Disordered Systems and Biological Organization. Disordered Systems and Biological Organization, NATO ASI Series F, Vol. 20 (1986), Springer: Springer Berlin), 233-240 · Zbl 1356.82026
[16] Mocella, M. T.; Bondur, J. A.; Turner, T. R., Etch process characterization using neural network methodology: a case study. Process module metrology, control and clustering, (SPIE Conf. Proc., Vol. 1594 (1991))
[17] C. D. Himmel, B. Kim and G. S. May, A comparison of statistically-based and neural network models of plasma etch behavior. Proc. 1992 Int. Semiconductor Manufac. Sci. Symp.; C. D. Himmel, B. Kim and G. S. May, A comparison of statistically-based and neural network models of plasma etch behavior. Proc. 1992 Int. Semiconductor Manufac. Sci. Symp.
[18] Himmel, C. D.; May, G. S., Advantages of plasma etch modeling using neural networks over statistical techniques, IEEE Trans. Semicond. Manufac., 6, 103-111 (1993)
[19] Kim, B.; May, G. S., An optimal neural network process model for plasma etching, IEEE Trans. Semicond. Manufac., 7, 12-21 (1994)
[20] Huang, Y. L.; Edger, T. F.; Himmelblau, D. M.; Trachtenberg, I., Constructing a reliable neural network model for a plasma etching process using limited experimental data, IEEE Trans. Semicond. Manufac., 7, 333-344 (1994)
[21] Rietman, E. A.; Lory, E. R., Use of neural networks in modeling semiconductor manufacturing processes: an example for plasma etch modeling, IEEE Trans. Semicond. Manufac., 6, 343-347 (1993)
[22] Chambers, J. M.; Hastie, T. J., Statistical Modelling (1992), S. Wadsworth & Brooks: S. Wadsworth & Brooks Pacific Grove, CA
[23] Gallinari, P.; Thiria, S.; Badran, F.; Fogelman-Soulie, F., On the relation between discriminant analysis and multilayer perceptions, Neural Networks, 4, 349-360 (1991)
[24] Klimasauskas, C. C., Neural Nets Tell Why, Dr. Dobb’s J., 16-24 (1991), April
[25] Deif, A., Sensitivity Analysis in Linear Systems (1983), Springer: Springer Berlin
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.