(This column, which originally appeared in Global SMT & Packaging magazine 6.5 (May 2006), is also available as a free PDF.)    

I still can’t escape from lead–free! Having spent 5 intense years at Cookson Electronics on the key materials transitions I now find myself coming at it from another direction – how can we use nanomaterials to improve some of the processing issues? The big question in my mind has been ‘What are the pressing real–life processing issues as we approach the RoHS deadline?’

A few weeks ago I had the chance to get back up to speed by attending the Intel Lead Free Symposium sponsored by IPC in Scottsdale, Arizona. This was well attended event with over 120 attendees mainly from North America and Asia – there were, surprisingly, only two attendees from Europe.

Intel chaired the meeting and provided a lot of technical information supported by speakers from Asustek, Hewlett–Packard and IPC as well as a slew of panel presentations from materials vendors.

The goal of the meeting was to encourage "lead free convergence to improve manufacturing robustness," and as you will see later there are some issues of real concern. Nasser Grayeli of Intel set the scene with trends and defined the call to action, which included:

• Gain clarity of global legislation initiatives.
• Industry alignment on reliability assessment.
• Industry collaboration to strengthen the alloy / materials / process knowledge base.
• Industry alignment on materials selection.

There were break–out sessions on key areas – second level interconnect (2LI) flux, 2LI solder joint reliability, high drop performance SAC, and PCB pad cratering (pad pull–off) with strong participation from materials vendors as panelists.

What were my personal takeways?

Divergence – convergence – divergence again.

We thought we had standardized on SAC 305 or 405, but now we see people talking about SAC105 and 205 in order to modify the intermetallics and tweak the properties. A lower % Ag degrades temperature cycling performance but increases shock resistance, so what works in other applications won’t necessarily work for cell phones. Ni and Ge are being introduced in solders, not to mention Al and Zn. And from my own discussions in Japan Bi is being proposed again for the time when the lead works its way out of the assembly arena because of improved processability.

How are contract assembly houses, packaging houses and component suppliers expected to keep these segregated? (Separating lead–free and leaded is bad enough.) Creative ways of segregating some materials are possible, e.g. the triangular lead–free solder bars some manufacturers are producing, which will only fit through the right hole in a solder pot feed, but the more materials, the more difficult it is. This is a really tough issue.

Surface finish shake–out.

There is a strong movement to standardize on silver and OSP as the lead–free finishes on grounds of cost, reliability and performance. Nobody in the meeting supported lead–free HASL. Although the automotive industry which uses this technique widely is exempt for now, they will start bumping up against non–availability of lead–finished components as has happened in aviation electronics and other areas. Tin coatings didn’t see a lot of supporters either, and there were several mentions of the high cost of gold as well as ‘black pad’, both of which eliminate gold’s widespread use.

Although silver and OSP are the apparent winners, they are not without issues. We saw some terrific detective work on subsurface voids – ‘caves’ in silver–finished copper – by Raiyo Aspandiar of Intel. The voids are caused by pinhole corrosion of the copper through the silver and result in planar arrays of micro voids in solder joints. What was nice is that afterwards we had all the major silver finish suppliers on stage showing how this issue could be prevented.

We also heard of the excellent reliability of new varieties of OSP from Helen Holder and Francois Billaut of HP as well as the challenges of testability and of flux compatibility with very different chemistries. Similar to the ‘cave void’ issue we had flux and OSP suppliers talking openly about their products.

The void issue isn’t going away.

We saw some great X–ray videos and a lot of cross–sections, but the consensus is still that even fairly large voids in the center of joints do not seem to affect reliability. Edge voids, small aligned voids and voids near the intermetallic interface can act as crack growth accelerators. ‘Where’ seems much more important than ‘how much’.

What to standardize and how to standardize?

Because of the many proprietary chemistries it is not appropriate for a society such as iNEMI to hold a ‘bake–off’ to decide whose chemistry is best or for a standards organization like IPC to standardize on that composition. That could lead to patent and unfair–competition issues. What is workable is to create a performance standard that products have to pass.

So what was the outcome? Each breakout group gathered interested parties who will follow up and decide the course of action, for example proposing an iNEMI program to gather data for a standard, or going straight to a standards organization like IPC to write standards or guidelines. If that is successful (and we certainly had many of the right people in the room) then we have taken an important first step in getting our hands round the issues.

What really impressed me at this meeting was the openness shown in presenting data about defect mechanisms with real data, warts and all! The issues were being confronted and not danced around.

If you want to get involved in any of the work groups I’d suggest you contact Daryl Sato at Intel daryl.a.sato@intel.com or Tony Hilvers at IPC anthonyhilvers@ipc.org.
 

Keywords : Alan Rae, lead-free

Average user rating

   (0 vote)