Additional Funding Sources
This project was supported by the Idaho State University Department of Chemistry C-SURF and Career Path Internship (CPI).
Abstract
A prototype atomic layer deposition (ALD) vacuum chamber was constructed to form the semiconductor GeS2. In previous experiments, PECVD (plasma enhanced chemical vapor deposition) was used to deposit GeCl4 and H2S to form germanium sulfides. However, the stoichiometric ratio of the germanium sulfides was difficult to control. To gain better control over the stoichiometric ratio in the product, PE-ALD can be used. This process releases reactants one at a time instead of continuously, which should allow single layers (monolayers) of each element to be deposited one at a time. Furthermore, the use of plasma can further allow the chemicals to be deposited sequentially, since the plasma can be programmed to pulse in intervals. The results to date include the construction and testing of the PE-ALD chamber. The sequential pulsing of gases and plasma was controlled with an Arduino control unit programmed with LabVIEW software. Argon was continuously pumped into the deposition chamber to purge excess reactant gases, GeCl4 for the Ge monolayer and H2S for the sulfur monolayers. The argon pressure and flow direction were the variables in the experiment. Other parameters varied were the mass flow settings on the reactant mass flow controllers, and the pulse widths of the 24-volt solenoid valves were used to introduce the reactants into the PE-ALD chamber. During initial tests of the deposition chamber, the chamber was monitored with an SRS RGA 200 mass spectrometer. Initially, it took 50 seconds for an H2S pulse to return to its baseline value, and it took 10 seconds for the GeCl4 to enter the deposition chamber and three minutes to return to its baseline value. Refinements to the pumping and timing of the reactant introductions will be discussed. These refinements allowed both gases to return to their baseline values within 20 seconds and allowed GeCl4 to enter the chamber within 2 seconds. Initial depositions, which used a continuous stream of plasma, formed thin films that were about 90% germanium by weight. Current and future experiments will attempt to increase the presence of sulfur in the thin films either through plasma pulsing or with special substrates such as gold-plated silicon.
Constructing and Testing a Plasma Assisted Atomic Layer Deposition Chamber for the Deposition of GeS2 from GeCl4 and H2S Using LabVIEW Controls
A prototype atomic layer deposition (ALD) vacuum chamber was constructed to form the semiconductor GeS2. In previous experiments, PECVD (plasma enhanced chemical vapor deposition) was used to deposit GeCl4 and H2S to form germanium sulfides. However, the stoichiometric ratio of the germanium sulfides was difficult to control. To gain better control over the stoichiometric ratio in the product, PE-ALD can be used. This process releases reactants one at a time instead of continuously, which should allow single layers (monolayers) of each element to be deposited one at a time. Furthermore, the use of plasma can further allow the chemicals to be deposited sequentially, since the plasma can be programmed to pulse in intervals. The results to date include the construction and testing of the PE-ALD chamber. The sequential pulsing of gases and plasma was controlled with an Arduino control unit programmed with LabVIEW software. Argon was continuously pumped into the deposition chamber to purge excess reactant gases, GeCl4 for the Ge monolayer and H2S for the sulfur monolayers. The argon pressure and flow direction were the variables in the experiment. Other parameters varied were the mass flow settings on the reactant mass flow controllers, and the pulse widths of the 24-volt solenoid valves were used to introduce the reactants into the PE-ALD chamber. During initial tests of the deposition chamber, the chamber was monitored with an SRS RGA 200 mass spectrometer. Initially, it took 50 seconds for an H2S pulse to return to its baseline value, and it took 10 seconds for the GeCl4 to enter the deposition chamber and three minutes to return to its baseline value. Refinements to the pumping and timing of the reactant introductions will be discussed. These refinements allowed both gases to return to their baseline values within 20 seconds and allowed GeCl4 to enter the chamber within 2 seconds. Initial depositions, which used a continuous stream of plasma, formed thin films that were about 90% germanium by weight. Current and future experiments will attempt to increase the presence of sulfur in the thin films either through plasma pulsing or with special substrates such as gold-plated silicon.