I have been working on piezoelectric(PZT) thin film for a few month, I grew thin film of piezoelectric onto a substrate of silicon by sputtering thecnic, then i was tryin to do capacitor for analizing its ferroeletric property , so i deposited first thin film of TiN onto the substrate of silicon, then the piezoelectric and finally a thin film of copper, but what i found is that copper diffuse through the piezoelectric thin film until it reach the TiN, I would like to know any advice about some special for preventing that diffusion.
I suggest to change copper by any other elements like gold since Cu tend to diifuse more than other elements
Hi Brayan, there are a couple of things that might help, firstly as has already been mentioned, try alternative elements to Cu. Even quite thin coatings of gold beneath your thicker Cu layer might be sufficient to reduce diffusion.
Another approach might be to investigate different sputtering conditions that might alter initial nucleation and resulting crystallography which could also reduce grain boundary diffusion.
Whilst I have not worked with sputtered PZT thin films, I suspect early thin film growth is likely to have a significant effect of PZT performance, I have seem similar effects in thin superconducting films and devices.
Good luck.
The role of top and bottom electrode on a Metal/Ferroelectric/Metal stack has been the subject of hundreds of papers during the last 40 years: so you should make some bibliography!
Typically, oxide electrodes like SrRuO3 or La0.5Sr0.5CoO3 or LaNiO3 have been used to get excellent ferroelectric properties as well as very low fatigue level, but these kinds of electrodes needs elevated deposition temperature (around 700°C). On the other hand, metallic electrodes like Pt and sometimes Ir have been used for industrial development of M/F/M devices. Note that the use of identical material for top and bottom electrode is also necessary to limit internal electric field.
I am sorry but using TiN (or even worse Cu) as electrode seems not to be a good choice to get large ferroelectric properties in a M/F/M stack.
NiCr as top electrode is an economical option for getting capacitor structure. However, Oxide electrodes mentioned by Gilles are better option if you want to study fatigue characteristics. RuO2 and IrO2 electrodes have shown good fatigue characteristics for PZT.
Thank you very much everyone, I do really appreaciate your all sugestions.
The best electrode is the one that works.
I've used silver, nickel, chromium, gold and aluminum. They all have advantages and disadvantages. Copper if not done properly will create copper ions that will unzip many polymers reverting them back to monomers. I'd stay away from the refractory metals (Mo, W, etc.) as they are brittle and tend to crack and open circuit during use.
- Badges
- Science topic