The Simple Analysis of Materials Processing Reactors (SAMPR) computer code is meant for the analysis of plasma and nonplasma processes used in manufacturing semiconductors. The code can also be used to analyze any chemical-processing reactor with gaseous (but not liquid) streams.
The code implements a mathematical model that consists of balance equations for the total mass, mass of individual chemical species, and gas energy. In the case of plasma reactions, a plasma power balance is also included.
The balance equations are volume-averaged; in other words, they represent a zero-dimensional (0-d) analysis. For this analysis to be valid, the reactor contents must be well mixed and not exhibit significant gradients of species concentrations or gas temperatures in any part of the reactor. Such perfect mixing conditions can be found in reactors used in chemical process industries. Such ideal conditions may not exist in reactors used for etching, deposition, annealing, or performing any other functions in the course of manufacturing integrated circuits. Nevertheless, approximate solutions to somewhat idealized reactor conditions may be valuable in estimating overall conversion efficiency of feedstock, effluent concentrations, and energy utilization.
SAMPR provides volume-averaged electron density, electron temperature, and concentrations of radicals and ions as functions of pressure, input power, and flow rates. Obviously, detailed information on the departure from uniformity of the plasma, and on fluxes of radicals and ions near a wafer in the reactor is lost in such a global model, but quantitative behavior of the plasma as a function of system parameters or the so-called "scaling laws" can be obtained very rapidly. Generation of such valuable knowledge with minimal computational resources is the attraction of this simple approach.
Also, the results obtained from a 0-d model can provide guidance for further multidimensional simulations. Usually, in a semiconductor-processing situation, the numbers of chemical species and reaction pathways are large. Multidimensional analysis with a large set of reactions and species is computationally intensive. A 0-d analysis in such a case can be used effectively in a systematic study to generate a "reduced chemistry set" that provides reasonable results.
This work was done by M. Meyyappan of Ames Research Center and T. R. Govindan of Applied Research Laboratory.
Inquiries concerning rights for the commercial use of this invention should be addressed to
the Patent Counsel
Ames Research Center; (650) 604-5104
Refer to ARC-13392.
