Microstructure, roughness, the enthalpy of formation of an oxide compared to the adjacent layer are some of the factors affecting the adhesion strength. Is there a specific variable for metal/nitride interfaces? Please, consider the deposition method to be sputtering and taking place without exposure to air. Examples: TiO2/Au vs TiN/Au.
Thank you in advance,
It is in first place surface energy. You want to have it as high as possible but of course this element dependent. On top of that you have process parameters of the PECVD and/or PVD systems. Underlying material should not outgas so many times prior metal is deposited in PVD systems, an insitu degas step is taking place.
Thank you very much, for the informations.
I would like to extend it a bit further. It is correct that O and N does not help with the adhesion of Au.
Au/Ti forms a better adhesion surface, however they diffuse into each other which can extend up to a few microns depending on the post-treatment conditions.
What would be the most stable adhesion layer between Au and SiO2?
Thank you very much for all the answers.
Danke Schon Prof. Stutzman!
The trick for improving adhesion is to engineer interface between film and substrate such that the bonding environment gradually changes from the substrate to the sputtered film.
For reasons pointed out by Prof. Stutzman, Cr is often used as adhesion layer for metal deposition on oxides.
As another example TiN (resulting from post deposition anneal) is used to improve electrical properties of GaN and Ti interface.
http://link.springer.com/article/10.1007%2Fs11664-998-0075-6
Hello Triratna, The link does not direct to a publication.
Thank you for the answer. I am looking for a more fundamental answer which explains the adhesive forces, the interfacial energy and strength.
Cr, Ti are basically used thats true, but what are the ways to make them more stable in the systems I mentiones in my question. What are the reasons lying under good and bad adhesion of oxide vs metal interfaces.
Thank you for trying to answer but it is not deep enough.
I cannot help with the theoretical treatment beyond what was suggested in the previous comments. Some deposition methods with higher energy of the impinging atoms, such as sputtering, can achieve better adhesion to the surface compared to lower energy deposition methods (thermal, e-beam). Also, cleaning the surface prior to deposition can significantly contribute to the adhesion (depending on how clean it was to start with). Moreover, heating it slightly in the vacuum before deposition (~100C) could desorb residual moisture, and if you use sputtering and you can slightly (reverse) sputter the surface before deposition it will improve the adhesion as well.
As for minimizing the diffusion of the adhesion layer's atoms in Au, you can first try minimizing its thickness. I see people using tens of nm of adhesion layers, where much less is usually sufficient. We were getting good results with
You may use an adhesion layer of Ti or Cr few nm thick.
Also, post deposition anneal will be helpful.
During deposition, an adhesion layer is helpful, such as Cr and Ti (5-10 nm). After deposition, you may also use thermal annealing to improve the adhesion strength. Besides, the match of coefficient of thermal expansion between the two layers is also very important for high-temperature applications.
- Badges
- Science topic
- Similar topics
- Physical Sciences
- Materials Science