My Little World

Today was a good day. I FINALLY figured out what was causing my erratic leakage problem, it wwas the quality of the silicon wafers I was using. I was using test grade wafers. This week I used virgin prime wafers, and my first wafer had most of it's leakage values reaching the floor of the multimeter. So with the combination of ensuring that the pallet is coated with tantalum, covering the anodization tank, and using expensive virgin prime wafers, I think I finally have a repeatable process worthy of further study. Low leakage at 10 V isn't enough. I have to make sure that the leakage stays low even at 50 V. That is going to be hard since my previous results show that leakage flux increases exponentially with voltage. I've read papers about it being due to an ionic conduction mechanism. Some have mentioned Frenkel-Poole conduction. I've also read that hydrogenating the tantalum metal before anodization increases it's ductility and plasticity. I wonder if hydrogenating the oxide will help fill in those interstitial points and prevent ions from filling in. That would help make the leakage behavior linear, if the leakage theory I've read about is correct. There is also the idea of a pulsed anodization process, and how that affects the structure and stoichiometry of the film being formed. I've started to find papers of people who have modeled this. Turs out that a guy by the name of Lawrence Young has been studying anodized Ta for the past 40 years! He's pretty much the world's expert on this process.

Well I'm going to try and install pkgsrc on my Solaris 9 box. I want to install nethack on this thing.

They talked about me on A.S.U.'s little engineering news magazine. Here is a rough draft of the article they will be publishing soon flexible displays featured on full circle magazine

That giant picture of my mug is being reflected off of a wafer coated with tantalum pentoxide. On top of the film are over 2022 aluminum capacitor dots I deposited with my sputtering machine, King Kong ( formally known at the MRC-603-II ). Behind that is a screen of a wafer map I made using an Electroglas 2001 automated prober. The map shows leakage flux uniformity across the entire wafer area. That's my best film so far, wafer 3 of my first experiments. I have yet to repeat it ^_^. The anodization process is very sensitive to metal contamination, and since the film is made by oxidizing metal it's very hard to reproduce a high quality film. Most of my time is spend trying to figure out where I'm being sloppy inmy process.