2005 Articles

Lead-free soldering was probably the first metal joining practiced. Is its end near? And what of solderless assembly?

What’s at the top of your worry list? RoHS and lead-free, right? But progress is being made and freedom from lead – or at least lead reduction – is in sight. How long did it take to get the lead out? Just a few years? Let’s take a brief look at the history of solder assembly.

Metallurgy is one of the ancient arts. In previous articles^1,2 it was estimated that solder was first used about 7,000 years ago; Mesopotamians, Egyptians and others in the region melted metals with blow-pipe-assisted flames hot enough to melt gold. Gold and silver were the metals of choice and naturally occurring alloys of gold-silver are found in these locations. Almost certainly the natural alloy was melted and used to solder pieces of gold together. Gold-silver is probably the original lead-free solder developed in the name of art.

Thousands of years later, the practical Romans were faced with the challenge of joining two pieces of metal together and solved it with solder. Since the metal to be joined was lead sheet, the solder contained lead. Lead sheeting used to line the aqueducts was readily sealed by heating the joints with strips of tin to form a tin-lead eutectic joint, giving birth to SnPb solder. And we are moving back. Ironically, the first solders contained silver just like modern solders; we’ve replaced gold with tin and copper to come full circle.

Electronics has made tremendous progress during the past century starting with vacuum tubes, moving to solid state and now heading toward nanoelectronic devices. The semiconductor industry can now fabricate hundreds of millions of transistors on a single stamp-sized chip with nanoscale dimensions. Perhaps in 10 years, that industry will shift to nanotechnology devices that may not even contain silicon. The electronic device field certainly represents the epitome of stellar technical progress. Circuitry, as well, is advancing: microvias are laser drilled, micron-dimensioned conductors are formed by additive processes and direct-write ink-jetting is being developed. But what about circuit assembly?

About 20 years ago, we moved the solder joint from the bottom of the circuit board to the surface – the surface mount “revolution.” Today, we are at the threshold of lead-free, a 7,000-year-old idea whose time has come round again. But the basic assembly process is still the same. We join metal A to metal B with solder C. Granted, we are changing the solder alloy, but there is no fundamental difference. What if we had applied the time and money of lead-free to modernizing the process to invent solderless assembly?

In fact, we have solderless assembly today, but it is quite limited. There are plug-in packages like the pin grid array (PGA), assembly materials like conductive adhesives and even metal fusion bonding, wire bonding and ultrasonic flip chip. But why not a universal solderless assembly? How about “Metal Velcro,” or a Lego-like interconnect? These ideas have been discussed for years with limited work done, but not enough. Perhaps MEMS 3-D fabrication techniques can now produce a connectable surface on chips or packages that readily mates to PCBs with complimentary bond pads. Or maybe something will emerge from the nanoparticle field. But is there really any incentive after just about breaking the bank with lead-free? Will we still be soldering in 7,000 years, or even in 70 years? Probably not!

Two important fields will bring fundamental change to components: optoelectronics and nanoelectronics. A time will come, sooner than many realize, when devices will no longer be made with silicon-based transistors. Photons will handle some of the signal transport tasks. Devices of the future could be based on carbon nanotube (CNT) materials, or something yet undiscovered, and many signals will be transmitted over transparent, nonmetallic conduits. These ideas are already on technology roadmaps and are planned events, not speculation. It is probable that metal conductors will not be needed as organic nanostructures are implemented with better electrical, thermal and mechanical properties; these CNT properties are extraordinary. When the device paradigm shift occurs, solder could well become a relic. Before you dismiss the idea of organic devices and interconnects, consider that you are reading and comprehending at this very moment using the world’s greatest data processing and logic system and it does not contain a single solder joint.

References

1. K. Gilleo, “The First 7,000 Years of Soldering, Part I” Circuits Assembly, October 1994, pp. 30-34.
2. K. Gilleo, “The First 7000 Years of Soldering, Part II,” Circuits Assembly, November 1994, pp. 44-45.

Dr. Ken Gilleo is with ET-Trends LLC; email: et-trends@cox.net.