Pore size dependence with anneal time (15 nm)
In a continuing study on how a post-silicon deposition Ar sputter affects morphology, I annealed three 15-nm films for 10 s, 60 s, and 300 s. If you recall in my first report, I found that the Ar sputter increases the porosity slightly for a 1000 C RTP and dramatically for a 900 C RTP. My motivation for performing the Ar sputter was to seed crystallization along with pore growth at the Si/SiO2 interface. This is in contrast to a “defect-free” amorphous film where nucleation occurs randomly. My goal was to try to increase porosity by initiating crystallization on one side of the film, allowing for the pore growth front to propagate through the entire 15-nm thickness. Here are the results from the anneal time study:
The control membrane was annealed at 1000 C with no Ar sputter. From these results, we can conclude that pore sizes do increase slightly with longer anneal times. My next experiment will be with 30 nm films in which the pre-sputter should have a more dramatic effect.
Any update on the pinhole story with Ar sputter?
Keep in mind I also deposited these wafers at a time no other users were using it so I could do a thorough clean of the chamber without worrying about contamination. We’ll have to see if the low pinhole #s continue once other users start ramping up again…