Dynamic neural control for a plasma etch process

Abstract

This paper presents results and commentary on using a cascade neural network and a policy-iteration optimization routine to provide suggested process setpoints for recovery from long-term machine drift in a LAM 4520 6-in dielectric etcher. Traditional plasma etch variables such as pressures, gas flows, temperatures, RF power, etc, are combined with a generalized representation of the time dependent effects of maintenance events to predict film etch rates, uniformity, and selectivity, A cascade neural-network model is developed using 15 months of data divided into train, test, and validation sets. The neural model both fits the validation data well and captures the nonuniformity in the in-control region of the machine. Two control algorithms use this model in a predictive configuration to identify input state changes, including maintenance events, to bring an out-of-control situation back into control. The overall goal of the optimization is to reduce equipment downtime and decrease cost of ownership of the tool by speeding up response time and extending the lifetime of consumable parts. The optimization routines were tested on 11 out-of-control situations and successfully suggested reasonable low-cost solutions to each for bringing the system back into control.

Estimation and Control in Semiconductor Etch: Practice and Possibilities

“…This is not a substantial issue in high-volume manufacturing, such as semiconductor etch, since most production plants are equipped with extensive databases for the recording of large numbers of process variables. Such data have been used in [82], [123], and [131]- [133] to train a neural network for the mapping of etch process input/ output relationships. A neural network model, predicting etch rate, is combined with a real-time optimizer in [131] to provide process setpoints to alleviate long-term process drift and sensitivity to PM interventions.…”

“…Such data have been used in [82], [123], and [131]- [133] to train a neural network for the mapping of etch process input/ output relationships. A neural network model, predicting etch rate, is combined with a real-time optimizer in [131] to provide process setpoints to alleviate long-term process drift and sensitivity to PM interventions. The ANN model is trained with extensive historical process data prior to use.…”

An IC yield enhancement approach by ARMA modeling and dynamic process control

“…Some works utilize neural networks to deal with the complexities [16][17][18]. In this paper, we ascribe the run to run variation of a process step to only one factor, the time.…”

“…Typically an IC process involves hundreds of steps. APC methods are used to improve qualities of those key process steps, such as implantation [15], photolithography [16], and plasma etching [17], etc. Some IC process steps are so complicated that they are dealt with by neural networks in recent works [17][18][19].…”

Advanced analysis of dynamic neural control advisories for process optimization and parts maintenance

“…Finally, by including a riskbased objective function in all optimization analyses, the DNC is able to quantify the degree to which a corrective adjustment is needed and the degree to which the overall risk of operation is reduced by performing the optimized recommendations. The DNC returns on a wafer-to-wafer basis a series of risk-reduction ranked corrective actions that may include process setpoint adjustments specific to recipe and maintenance corrective actions [9]. Within the same model, the DNC discriminates between several repair actions based on their impact on wafer quality and can predict wafer quality for all quality variables after one or more simulated repair actions .…”