Numerical simulation and verification of gas transport during an atomic layer deposition process

Atomic Layer Deposition (ALD) is a process used to deposit nanometer scale films for use in semiconductor electronics. The reactor consists of a warm wall horizontal flow tube, a substrate mounted on a disk downstream from the inlet, and cyclic flow between a reactant gas, a purging gas and a gas that preps the surface of the substrate. The objective is to achieve a uniform coating on the substrate layer by layer in minimal time. It is possible to use in situ monitoring of the gas phase and deposition to modify layer formation. Process improvement is currently accomplished experimentally by monitoring the precursor delivery and the growth of the film and adjusting the parameters: flow rates, temperature, pressure, concentrations, etc. Accurate simulation and optimization can decrease processing time and cost and increase control during product development. In addition, increased accuracy of gas transport simulation can be used to analyze reaction and diffusion rates, reaction mechanisms and other physical properties. In this paper we introduce the first comprehensive numerical solution of the Dusty-Gas Model including the complete binary diffusion term. We derive a concentration dependent Damkohler number relevant to the purge step of the process. The simulation matched the experimental data at a specific Damkohler number and further variation of the parameter confirmed existing experimentally observed phenomena.