This paper presents the systematic characterization of the molecular beam epitaxy (MBE) process to quantitatively model the effects of process conditions on film qualities. A fivelayer, undoped AlGaAs and InGaAs single quantum well structure grown on a GaAs substrate is designed and fabricated. Six input factors (time and temperature for oxide removal, substrate temperatures for AlGaAs and InGaAs layer growth, beam equivalent pressure of the As source and quantum well interrupt time) are examined by means of a fractional factorial experiment. Defect density, X-ray diffraction, and photoluminescence are characterized by a static response model developed by training back-propagation neural networks. In addition, two novel approaches for characterizing reflection high-energy electron diffraction (RHEED) signals used in the real-time monitoring of MBE are developed. In the first technique, principal component analysis is used to reduce the dimensionality of the RHEED data set, and the reduced RHEED data set is used to train neural nets to model the process responses. A second technique uses neural nets to model RHEED intensity signals as time series, and matches specific RHEED patterns to ambient process conditions. In each case, the neural process models exhibit good agreement with experimental results.Index Terms-Molecular beam epitaxy (MBE), neural networks, process modeling.

This paper presents a statistically designed experiment for systematic characterization of the molecular beam epitaxy (MBE) process to quantitatively describe the effects of process conditions on the qualities of grown films. This methodology is applied to a five-layer, undoped AlGaAs and InGaAs single quantum well structure grown on a GaAs substrate. Six input factors (time and temperature for oxide removal, substrate temperatures for AlGaAs and InGaAs layer growth, beam equivalent pressure of the As source and quantum well interrupt time) are examined by means of a Resolution IV, 26-2 fractional factorial design requiring sixteen trials. Several responses are characterized, including defect density, x-ray diffraction, and photoluminescence.Results indicate that the manipulation of each of the six factors over the ranges examined are statistically significant and lead to considerable variation in the responses. Following characterization, backpropagation neural networks are trained to model the process responses. The neural process models exhibit very good agreement with experimental results.