Kinetic modeling of N incorporation in GaInNAs growth by plasma-assisted molecular-beam epitaxy

We have studied the growth of GaInNAs by a plasma-assisted molecular-beam epitaxy (MBE). It was found that the N-radicals were incorporated into the epitaxial layer like dopant atoms. In the range of 400-500 degrees C, the growth temperature (T-g) mainly affected the crystal quality of GaInNAs rather than the N concentration. The N concentration dropped rapidly when T-g exceeded 500 degrees C. Considering N desorption alone is insufficient to account for the strong falloff of the N concentration with T-g over 500 degrees C, the effect of thermally-activated N surface segregation must be taken into account. The N concentration was independent of the arsenic pressure and the In concentration in GaInNAs layers, but inversely proportional to the growth rate. Based on the experimental results, a kinetic model including N desorption and surface segregation was developed to analyze quantitatively the N incorporation in MBE growth. (C) 2000 American Institute of Physics. [S0003-6951(00)00928-1].

Modeling and optimization of semiconductor manufacturing process with neural networks

A neural network-based process model is proposed to optimize the semiconductor manufacturing process. Being different from some works in several research groups which developed neural network-based models to predict process quality with a set of process variables of only single manufacturing step, we applied this model to wafer fabrication parameters control and wafer lot yield optimization. The original data are collected from a wafer fabrication line, including technological parameters and wafer test results. The wafer lot yield is taken as the optimization target. Learning from historical technological records and wafer test results, the model can predict the wafer yield. To eliminate the "bad" or noisy samples from the sample set, an experimental method was used to determine the number of hidden units so that both good learning ability and prediction capability can be obtained.

Generalized Viscoplastic Modeling of Debris Flow

Various concepts have been proposed or used in the development of rheological models for debris flow. The earliest model developed by Bagnold was based on the concept of the “dispersive” pressure generated by grain collisions. Bagnold’s concept appears to be theoretically sound, but his empirical model has been found to be inconsistent with most theoretical models developed from non-Newtonian fluid mechanics. Although the generality of Bagnold’s model is still at issue, debris-flow modelers in Japan have generally accepted Takahashi’s formulas derived from Bagnold’s model. Some efforts have recently been made by theoreticians in non-Newtonian fluid mechanics to modify or improve Bagnold’s concept or model. A viable rheological model should consist both of a rate-independent part and a rate-dependent part. A generalized viscoplastic fluid (GVF) model that has both parts as well as two major rheological properties (i.e., the normal stress effect and soil yield criterion) is shown to be sufficiently accurate, yet practical, for general use in debris-flow modeling. In fact, Bagnold’s model is found to be only a particular case of the GVF model. Analytical solutions for (steady) uniform debris flows in wide channels are obtained from the GVF model based on Bagnold’s simplified assumption of constant grain concentration.