This paper describes a methodology based on the combined utilization of both a multisensor system and an optimized artificial neural network that has been applied to equipment utilized for the production of doped silicon dioxide films. The model exhibits an average relative error around 1% in predicting the concentrations of dopants and the thickness of the oxide layer. One of the major benefits of such a predictor is the ability of providing an on-line estimate of the process yield, thus avoiding off-line testing and gaining a significant reduction of risks of wafer loss. The neural model here described is currently utilized as a control tool at the Texas Instruments Avezzano, Italy, plant
Modeling of APCVD doped Silicon Dioxide deposition process by a modular neural network
Corrado Di Natale;Emanuela Proietti;Arnaldo d'Amico
1999
Abstract
This paper describes a methodology based on the combined utilization of both a multisensor system and an optimized artificial neural network that has been applied to equipment utilized for the production of doped silicon dioxide films. The model exhibits an average relative error around 1% in predicting the concentrations of dopants and the thickness of the oxide layer. One of the major benefits of such a predictor is the ability of providing an on-line estimate of the process yield, thus avoiding off-line testing and gaining a significant reduction of risks of wafer loss. The neural model here described is currently utilized as a control tool at the Texas Instruments Avezzano, Italy, plantI documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.