In article , Harold and Susan Vordos says...

You actually can see the distortion from that.

That's an interesting observation. If you don't know, the reason the
nickel is plated to begin with is to prevent migration of the gold to the
base metal. In my years of refining I discovered it was pretty much common
practice to find nickel under gold plating when stripping. Sir T.K. Rose
concluded long ago that gold and silver have an affinity for certain metals
and will migrate to them when in close proximity, and they need not be
molten or even hot. Ingots of pure silver and pure gold end up
transferring to one another when in contact, for example. Nickel prevents
the migration. If what you say is correct, unless gold is plated quite
thick (I have no idea what that may mean) , there is likely no benefit in
using it at all, because, over time, and it's not long, it simply vanishes.
One problem gets replaced with another. Comments?

Transition metals are commonly used as both diffusion barriers and
as adhesion layers. (Ti, Ni, Mo, W, etc)

The first prevents the plated, or deposited, layer from diffussing into
the the substrate. This is a real issue in semiconductor processing,
where it's important to keep layers of metal (wiring) on chips from
being contaminated by the layers adjacent. This is especially important
because some of the processing is carried out at elevated temperatures.
A layer of, say, Ti, that is about 200 atomic layers thick will
suffice to keep the layers of the cake from smearing out.

Adhesion layers are another issue. This is why nickel strikes are
used under gold plating as a rule.

Adhesion is a peculiar thing - what makes one thing 'stick' to another?
At atomic levels this question is a) of considerable practical interest
and b) not entirely well understood. There's a lot of research being
done on the fundamentals.

I can recall desperately trying to figure out some way to deposit
copper on sapphire stubstrates - but it would *not* stick. The
common trick is to put down a Ni adhesion layer, but nickel is magnetic
and for our work that was unacceptable.

That company I mentioned (trompeter) makes rf connectors for low
level, high freqency applications. All of the other connectors
I investigated were magnetic because of the ni strike in the plating
process. Because they build for the small-signal rf market - mostly
satallite communications gear - they understand that distortion of
the rf signal can occur from any magnetic materials in the connector,
so they use something they call a "pulse plating" process to
eliminate the Ni strike beforehand. This is specifically to
replace an *adhesion* layer, not a diffusion barrier.

Does the gold diffuse through into the brass metal underneath?

I honestly don't know. They may use some other kind of diffusion
barrier, but I've never seen them degrade over the 20 or so years
I've been using them. I will look up in their catalog at work
and find their spec for plating thickness.

Jim


--
==================================================
please reply to:
JRR(zero) at pkmfgvm4 (dot) vnet (dot) ibm (dot) com
==================================================