Can Immersion Gold PCBs Oxidize?

Phil
Welcome to Board Talk. This is Jim Hall and Phil Zarrow, the Assembly Brothers, Pick and Place here to discuss surface mount processes, equipment, materials, philosophy, whatever.  So what is whatever today, Jim?

Jim
We have a question from J.T. "I have immersion gold PCBs that have developed oxidation. Is there a specific material/process to recover these boards?"

Phil
Somehow there's a disconnect here, one does not normally associate oxidation with gold finishes.

Jim
I think it's impossible, isn't it?

Phil
It'd be really, really hard to do.

Jim
I think it takes a nuclear accelerator to get oxygen to combine with gold. So there's obviously something else going on.

Phil
There is something amiss here. I would say among other things get thee to a failure analysis lab henceforth because I don't really know what's going on.

Jim
I would think of historical sources first. Have these been exposed? Obviously, the thing that comes to mind is some sort of contamination. Something has grown under the surface, from handling, from an atmosphere, from some other process, from something that was or wasn't done in the initial board fabrication process where they put on the gold finish.

If you can track that down historically, it may be a little less painful than going to an analysis lab, but then that being not possible, the only way you're gonna really know what it is to have it chemically analyzed.

Phil
Or you may be a victim of the dreaded plague of black pad.

Jim
Phil, it doesn't say that it's ENIG. It says immersion gold. Immersion gold.  So that could be gold over copper.  In which case it might be something else  

Phil
But it could be something very, very likely something in the planning process.

Jim
Let's be honest. Gold is fabulously expensive today so any manufacturer who's putting gold on is going to try to get away with absolutely the minimum amount of gold that - perhaps pushing the low end of a thickness specification.

So if you have a porous gold - either over nickel or over copper - now you can get oxygen fusing through a porosity and causing some bad stuff to happen, which may change color.

Phil
And again, a failure analysis lab would be a great place to turn to determine exactly what you are seeing there.

Jim
They could measure the thickness of the coating and so forth.  But again, going back to your supplier and getting their specifications, checking your certificate of compliance - which hopefully you have.

Phil
And hopefully it hasn't oxidized away, but yep, absolutely. Should be going through the specs.

Jim
Unless I really have forgotten that much of my chemistry, I don't think that you're chemically mixing oxygen and gold.

Phil
There's something amiss here and most likely at your board shop. But again, send it to a lab and find out what's really going on.  And as far as what's going on here, this is Phil Zarrow and Jim Hall, the Assembly Brothers.

Thank you for tuning in. And whatever the surface finish you're trying to solder to, whatever tarnish you get

Jim
Don't solder like my brother.

Phil
And don't solder, or oxidize, like my brother.

Immersion Gold over Ni (ENIG) is by nature a porous and extremely thin finish. Unfortunately ENIG is the most common surface finish in the world (for no good technical reason) - leading to an incredible amount of reliability problems.

ENIG is no good for solder joint reliability in portable electronics because of formation of the brittle Ni-Sn InterMetallic Compound. It therefore have been banned by most of the serious mobile phone brands for BGA/CSP footprints the last 5 - 10 years. OSP is significantly cheaper and gives much better solder joint strength.

ENIG is furthermore a very unreliable surface finish for electromechanical contacts or keypads: the porosity and low scratch resistance of the typical only 25-60 nm thin gold leads to exposed Ni that easily corrodes - resulting in contact reliability issues.

If you want to have a PCB contact pad you can rely on it have to be thick electrolytic gold > 0.8um or Carbon surface finish. Take a look at my technical paper: New trends for PWB surface finishes in mobile phone applications. You can download it from here: http://cwngroup.dk/about/technical-papers/

Our experience shows that ENIG with gold plating thickness in the low 2- 2.5 micron range can cause some nickel oxidation, especially if the gold is on the porous side. Typically ENIG boards require only a mild flux but to solder these boards just use a stronger flux. No other rework can really be done.

Even non-porous layers will get problems if there is no barrier layer between the gold and copper. There should be a well sealed nickel layer (Electroless Nickel-Gold) between the gold and copper. If the nickel is porous, copper atoms migrate through the gold to the surface where they oxidize to copper oxide.

You are correct gold does not oxidize, but immersion gold is porous no matter how much you deposit. A thinner layer of gold does not necessarily mean less protection of the underlying metal, but it does mean that the underlying metal experienced less corrosion during the immersion process.

Pure gold does not oxidize. Co-deposited metals can. However, many problems are caused by the porosity of the gold coating, allowing the substrate metals (Ni or Cu) to oxidize and causing solderability problems.

Immersion gold on copper is usually thin (approximately 5 microinches) and has a degree of porosity which will lead to oxidation. In addition, if the coating is just gold over copper, diffusion will continue until all the gold has gone. Even the current ENIG (electroless nickel/immersion gold) process has a limited life for solderability because the immersion gold is thin.

I agree with Edward. We have been using ENIG boards for 5 years and have never had an oxidation problem. Boards are not stored more than a year or so and are in a humidity controlled environment. If the boards are ENIG cleaning and reflow of the boards may drive off the oxidation.

Any gold immersion coated board should be placed in a metallized ESD bag and sealed well. Then store in a nitrogen purged dry box. It will still only last up to 6 months before the gold becomes duller due to the diffusion of the substrate. The more expensive and much thicker auto-catalytic electroless gold process will be a better approach.