In my experiments, I am directly pyrolyzing photoresist (PR) in an inert or reducing gas to form a C film. I have deposited Cu via thermal evaporation (50 or 300 nm), before spin-coating the PR in order to ensure good electrical conductivity for later measurements. I've confirmed that, regardless of modifying pyrolysis temperature (700-900C), heating cycle during pyrolysis, Cu film thickness, or substrate (fused silica or single crystal quartz), the Cu film has de-wet from the substrate to form small spherical droplets after the pyrolysis has finished. How can I prevent the de-wetting of the Cu film? I'd prefer not to increase the Cu film thickness if possible
It is a commonly observed phenomenon that metal films dewet from a silica or quartz substrate if heated above a certain temperature. This is due to the small attractive forces between the substrate and the metal, and the strong interaction between the metal atoms on the other hand. So at elevated temperatures, where the atoms get more and more mobile, the dewetted state is thermodynamically favoured compared to a thin film on the substrate.
What you may try to prevent this is to coat the substrate with a metal of a high melting point such as Cr or Ti, these metals are used to promote the adhesion of thin films on glass or silicon. If you deposit a film of about 5-10 nm, then this underlayer should stuck the substrate, and the Cu adlayer on top will be adherend to this basal layer.
I cannot guarantee, but I think that this may improve the dewetting properties of the Cu, good luck, and keep us updated about the progress!
Dirk
Thank you for your suggestion. As you said, I was also suggested by another researcher to deposit 2-3 nm Cr in between the Cu and the fused silica substrate. Do you imagine there will be much a difference in Cu film stability if I deposit 2 nm vs 10 nm of Cr?
Usually, if you deposit only 2 nm, then this "thin film" will not be a continuous film, but it will already form islands in its native state. So this very thin Cr film will act as a nucleation centre for the Cu film growth in the second step. In contrast, what you need is an adhesion promoter, and thus I guess (but I must say I have no exoperimental evidence for this guess) that a thicker film will be better, because the channels between the Cr clusters will be smaller, and thus the trend of the Cu film to form isolated clusters may be smaller.
Hi, thank you for your suggestion again. The 10 nm Cr layer underneath my Cu layer definitely reduced the degree of dewetting. I was able to get around the problem altogether by also wrapping the sample holding alumina boat in Cu foil and reducing my pyrolysis temperature.
Hi, thank you for your suggestion again. The 10 nm Cr layer underneath my Cu layer definitely reduced the degree of dewetting. I was able to get around the problem altogether by also wrapping the sample holding alumina boat in Cu foil and reducing my pyrolysis temperature.
- Badges
- Science topic