A combined process and device modeling program for GaAs MESFET digital and microwave integrated-circuit technology is described. GATES (for GaAs transistor engineering models) is designed to meet the needs of process, test, and manufacturing engineers who require answers to questions about how process parameters and tolerances affect device performance and uniformity, but who have limited access to large-computer modeling tools. The program makes extensive use of analytical models. The physical models used for ion implantation, dopant diffusion and other process factors and the methods used to calculate carrier profiles, threshold voltages, layer resistivities. MESFET parasitic resistances, and I-V characteristics are discussed. The I-V curves can be fitted to expressions suitable for use in circuit optimization codes such as SPICE. It is shown how to calibrate some of the physical quantities.>

A computer program SUPREM 3.5, has been developed to simulate processes used to manufacture ion-implanted GaAs integrated circuits. The processes modelled in the present version of the simulator include ion implantation, diffusion, and activation. The simulator includes a routine to calculate the threshold voltage of a MESFET device based on the simulated processing results and on substrate properties. The models used for each dopant and process are based on physical mechanisms when possible, but empirical parameter fitting is sometimes used. Pearson IV distributions are used for the implantation modeling, concentration-independent diffusivities are used for n-type dopants, concentration-dependent diffusivities are used for p-type dopants, and concentration-dependent activation efficiencies are used for all dopants. The simulator models each process individually, as well as integrating the various processes and models together.>

A simulation program that models the gas immersion laser doping (GILD) process is described. This program, which is called LASERMELT, first solves for the silicon melt depth and melt time versus laser energy fluence, and then the impurity dopant profiles using a dopant incorporation and impurity diffusion model. Experimental and simulated dopant profiles and sheet resistance values are given as functions of the laser energy fluence and number of pulses. The determination of liquid phase impurity diffusion coefficients in molten silicon is also described.>