Matalized film capacitors question. (involving sputtering) / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

Matalized film capacitors question. (involving sputtering)

Move Thread

LAN_403

Ash Small

Tue Jun 19 2012, 01:13AM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

I've been reading a few threads here regarding matalized film caps, I assume these are manufactured by sputtering aluminium onto the plastic film (please correct me if my assumtion is incorrect).

It occured to me that HV caps could be produced by sputtering copper (which is very easy to sputter) onto a suitable plastic film (polythene for example).

My question is this:

Sputtering is performed inside a vacuum chamber. Plastics (like polythene) tend to outgas in a vacuum. What pressure should such a system operate at for best performance? Is copper suitable for this application, or should I be considering metals like aluminium, which are much more difficult to sputter, but which can be deposited more easily using other vapour deposition techniques?

HV capacitors produced in this manner would not suffer from the drawbacks of the 'usual' homebrew HV capacitors, such as corona breakdown, etc.

plazmatron

Tue Jun 19 2012, 01:49AM

Registered Member #1134 Joined: Tue Nov 20 2007, 04:39PM
Location: Bonnie Scotland
Posts: 351

Ash Small wrote ...

Sputtering is performed inside a vacuum chamber. Plastics (like polythene) tend to outgas in a vacuum. What pressure should such a system operate at for best performance? Is copper suitable for this application, or should I be considering metals like aluminium, which are much more difficult to sputter, but which can be deposited more easily using other vapour deposition techniques?

Outgassing, on a continuously pumped system is no problem at all, so long as the gasses aren't evolved at a rate faster than your pumps can handle. Even the smallest diffusion pumps, will pump upwards of 50L/sec. Only in a sealed system does outgassing become a serious problem.

There is at least one person, that metallised some paper here:

I see no reason why copper would be unsuitable, it does sputter easily, so easily in fact, that I try not to use it in any glow discharge experiments anymore, since I got fed up with cleaning my glasswork!

Les

Mattski

Tue Jun 19 2012, 03:13AM

Registered Member #1792 Joined: Fri Oct 31 2008, 08:12PM
Location: University of California
Posts: 527

I wonder how a vacuum deposited metal would compare to plated, using a system like

with subsequent electroplating to thicken the metal?

Ash Small

Tue Jun 19 2012, 12:28PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

Mattski wrote ...

I wonder how a vacuum deposited metal would compare to plated, using a system like

with subsequent electroplating to thicken the metal?

I've been looking into this a bit, Mattski, and, from what I've read, the substrate (polythene, in this case), after cleaning, must be made hydrophyllic (I've not yet found any details on this part), then needs to be 'activated' using silver nitrate (or the salt of another noble metal), prior to the electroless plating stage.

While one could purchase a proprietary kit such as the one you linked to, I'm tempted to try the sputtering of copper, as copper sputters so easily (as Les points out, it's difficult to stop it sputtering).

I have a rotary vacuum feedthrough, which I could use to turn a roller inside the chamber, to coat a whole roll of polythene, I'd need to mask the edges, and not sputter the ends of the roll, for insulation purposes. I imagine adhesive tape of some sort could be used for this.

The main advantage of 'dry chemistry' compared to 'wet chemistry', in this case, is that it would avoid having to overcome the hydrophobic nature of the plastic film.

Does anyone know what metals are used in commercially produced matallized film capacitors? I assume it's usually aluminium, but I'm only guessing here.

plazmatron

Tue Jun 19 2012, 01:38PM

Registered Member #1134 Joined: Tue Nov 20 2007, 04:39PM
Location: Bonnie Scotland
Posts: 351

Ash Small wrote ...

Does anyone know what metals are used in commercially produced matallized film capacitors? I assume it's usually aluminium, but I'm only guessing here.

It is in fact Aluminium. I don't know the real reasons for this, but here are a few guesses:

The low price and melting point of aluminium saves money.
Aluminium forms a passivation layer in air, preventing further oxidation (I would imagine a thin layer of copper or other reactive metals, would oxidise quite rapidly (certainly in terms of component lifetime), rendering it useless)
Also, I am not sure how other metals 'stick' to the substrate.

Although I said in the beginning, I see no reason why one can't use copper, there are always good reasons why manufacturers use what they use. After really thinking about it, I would be inclined to solve whatever issues you are having with aluminium deposition, but having said that, there is nothing at all to stop you trying it with copper.

Les

Pinky's Brain

Tue Jun 19 2012, 02:10PM

Registered Member #2901 Joined: Thu Jun 03 2010, 01:25PM
Location:
Posts: 837

Ash Small wrote ...

It occured to me that HV caps could be produced by sputtering copper (which is very easy to sputter) onto a suitable plastic film (polythene for example).

I've said it before and I'll say it again ... the only plastics which make sense are BoPET and BoPP, nothing else gets close.

wrote ...
HV capacitors produced in this manner would not suffer from the drawbacks of the 'usual' homebrew HV capacitors, such as corona breakdown, etc.

A single layer in oil won't generate much corona, but as soon as you start stacking layers or rolling them up it becomes very likely you will catch some air bubbles.

Mattski

Tue Jun 19 2012, 05:04PM

Registered Member #1792 Joined: Fri Oct 31 2008, 08:12PM
Location: University of California
Posts: 527

I think the reactivity of copper, in particular that it reacts with and damages silicon and gate insulators, is why aluminum was first more popular for deposition in the semiconductor industry. To use copper they first deposit a barrier layer of another metal like tungsten.

I found one source which suggests that copper might not have good adhesion to polyethylene, though I don't have access to the full paper unfortunately:

Ash Small

Tue Jun 19 2012, 09:48PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

Pinky's Brain wrote ...

I've said it before and I'll say it again ... the only plastics which make sense are BoPET and BoPP, nothing else gets close.

Yes, I do realise this (and you've said it many times before), but I was planning on trying this with PE first, mainly because I can get it easily.

Pinky's Brain wrote ...

A single layer in oil won't generate much corona, but as soon as you start stacking layers or rolling them up it becomes very likely you will catch some air bubbles.

Not if I roll them inside the vacuum chamber. I do have a rotary feedthrough.

Mattski wrote ...

I think the reactivity of copper, in particular that it reacts with and damages silicon and gate insulators, is why aluminum was first more popular for deposition in the semiconductor industry. To use copper they first deposit a barrier layer of another metal like tungsten.

I found one source which suggests that copper might not have good adhesion to polyethylene, though I don't have access to the full paper unfortunately:

I understand sputtering will stick to anything. I do plan on trying aluminium, but I'm still at the early stages here. There are a number of sites describing how amateurs have coated telescope lenses with aluminium, using vapour deposition techniques. It's just that Cu is much easier to sputter than Al. I realise that I may have oxidation problems, but these should be avoided with vacuum impregnation and/or oil submersion, so it 'may' actually work.

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob