The objectives of this project are to elucidate the mechanisms of adsorption, desorption, surface reaction, diffusion, tungsten nucleation and film growth in the heterogeneous chemical vapor deposition of tungsten hexafluoride and tungsten hexachloride and to develop fundamentally based, non-empirical reaction rate expressions based on these mechanisms. An experimental approach based on complementary ultrahigh vacuum (UHV) techniques will be employed so that the various reaction steps can be isolated and examined individually under well-defined conditions. Temperature-programmed desorption and reaction spectroscopy (TPD/TPRS) will be used to separate adsorption from decomposition and desorption and to determine the rate constants for each step. Rate expressions based on mechanisms suggested by the TPD/TPRS studies and scanning kinetic spectroscopy (SKS) experiments will be compared to growth rate curves. Clean silicon and clean tungsten surfaces will be investigated to establish behavior of the silicon reduction and hydrogen reduction reactions, respectively, in the absence of chemical species which may act as poisons or promoters. Impurities, dopants and surface insulating layers will then be introduced systematically and their effects on the deposition reactions will be quantified. Successful completion of this research will lead to a greater fundamental understanding of the tungsten deposition reactions, particularly with respect to the limits of selective deposition, the self- limiting growth behavior exhibited on silicon, and to methods for intentionally promoting or poisoning the surface reactions. The proposed research is motivated by new demands placed on materials as lateral and vertical dimensions in semiconductor devices continue to shrink in the drive for faster, more reliable integrated microcircuits. Tungsten possesses properties which make it an attractive candidate for a number of technologically- important applications in very large scale integration (VLSI) circuits. Some manufacturers are already incorporating tungsten interconnects into their circuits, but existing technology leaves much room for improvement in quality control and production of very fine structures.

Project Start
Project End
Budget Start
1987-07-01
Budget End
1989-12-31
Support Year
Fiscal Year
1987
Total Cost
$70,000
Indirect Cost
Name
Arizona State University
Department
Type
DUNS #
City
Tempe
State
AZ
Country
United States
Zip Code
85281