Dear fellow researchers,
I am currently trying to build thin film tantalum pentoxide integrated capacitors. I use 8''/200mm silicon wafers with a 100nm thick silicon oxide layer. I sputtered 200nm of TiW and on top of that 1 µm of copper. This I want to use as the bottom electrode. I then sputtered 250nm of tantalum and anodized it in a mixture of distilled water, tartaric acid and ethylene glycol at 120V and 80mA.
Now to my problem: I encountered a lot of defects on the tantalum pentoxide layer which I am not sure why (pictures are attached). My first guess was that maybe copper diffuses into the oxide layer during anodization, and causes these defects. This is why I then tried a slightly different setup. Everything was the same except that I now sputtered another 200nm TiW layer between the copper and tantalum layers. I now don't see these defects any more.
My question is if anyone else encountered this problem because I can't find any papers about copper diffusion during anodization. Also, I am not sure if it was copper diffusion or maybe a breakdown of the oxide layer or any other reason. I would find the diffusion of copper quite unintuitive because I read that tantalum is being used as a diffusion barrier for copper.
I would appreciate any ideas or confirmation of my theory.