[NewCandle] Anodizable metals and essential conditions for glow discharge?

[Horace Heffner]

On Sep 15, 2008, at 12:08 PM, Nick Reiter wrote:
>
> Could one take say a sputtered conformal coating of
> Al2O3, or SiO2, or ZnO, make it 1000A thick on an
> aluminum or other metal substrate, and get an AGD from
> it?  Does the oxide coating have to be constantly
> re-generating, or native, or just there as dielectric
> layer for tunneling?
>
> It may be a couple of weeks until I can try this out,
> but I do have the ability to sputter some SiO2 as a
> dielectric barrier.

My experience with this, again if memory serves, is that operating in  
pure DC mode, in some conditions, tends to build up a barrier until  
the width reduces the current. You then have to push voltage to  
continue the same glow.  This can eventually lead to operation in  
electrospark mode.   Running in AC tends to delay or avoid this.   
Running in a reduced concentration electrolyte can avoid an excessive  
anodizing rate.

Electron tunneling rate is *very* sensitive to barrier width d.   
Width d is in an exp(d*k) term.  Tunneling is limited to layers a few  
atoms thick, I think.  Impurities in the the barrier layer permit  
tunneling between islands of potential, and thus a much thicker  
barrier. This kind of stage-wise tunneling may also be glow related.

The big old electrolytic cells used by phone companies as voltage  
regulators exhibited the glow, and they ran for years without any  
problems, so that kind of stability can be achieved, but I think it  
may be achieved using very large anions, like those in acetic acid.

It also may be significant that the conditioned layers act not only  
as a tunneling barrier, but as a diode.  The capacitance effect, the  
i vs V "eye" occurs when AC is used, thus when the two electrode  
surface diodes are opposed.

Ordinary semiconductor diodes, when pushed very near failure voltage,  
I think also can glow. They also breakdown conduct sporadically,  
mimicking the electrospark mode, until final failure occurs. There is  
no electrolyte handy there to heal them.

Best regards,

Horace Heffner
http://www.mtaonline.net/~hheffner/