Adaptive Segmentationmicro-segmentation March 19, 2015

A New Day for Soldering Semiconductors

Jimmy Ray Purser,

While knocking back an seemingly endless stream of coffee and pancakes at breakfast today, I read an article in NASA Tech Briefs about a new form of soldering that will absolutely change how fast we deploy semiconductors in manufacturing.  

If you REALLY wanna geek it up, here’s the actual published work.

Jimmy Ray's Tech Log

Soldering 101

Soldering has been around for a really long time. Due to its low melting point, it’s used to bond wires, plumbing pipes, and other electronics. What doesn’t soldering work for? Joining semiconductors together.

Semiconductors are highly sensitive to impurities in the solder itself. Technically speaking, the actual problem is at the sub-atomic level. Metal-to-metal semiconductors form what’s called a Schottky barrier. Now, this is all groovy if we are talking about diodes, but for transistors, it’s a real pain in the tail. It’s kinda like wearing too much flair. Too much Schottky and you just turned your transistors into diodes. Too little, and now you have an ohmic contact that’s basically a straight piece of wire.

Both outcomes suck. And due to impurities, both are difficult to predict. So a best practice has been not to mess around with soldering semis and to just move on. Until recently, when some folks at the University of Chicago thought, “Hmmm. Football season is over. Let’s work on this whole soldering thingy while we wait for the Cubs to start playing.”

How do you solve a problem like Tin Whiskers?

Looks like they may have found an answer. They grabbed a Tupperware, mixed up some lead and cadmium, tossed in a pinch of bismuth, and then broke out the astral runes and caste a spell from the lunar spellbook. KA-ZAP! A new chalcogenidometallates solder was born! A chalcogenidometallate is not a dude from Hellriser. It’s a compound of some chemistry thingy dealing with the permutations in multiple levels of crystal formation that, to be honest, I just do not understand that well. I suck at chemistry.

After reading the paper, I wonder how they were able to solve the “tin whisker” problem. Besides being a totally groovy name for a Southern Rock band, tin whiskers are basically electrically conductive, crystalline structures of tin that actually grow from surfaces where tin based-solder is used. That’s right! Like a beard! These can grow up to 10mm and cause all kind of problems, like short circuits.

Tin whiskers aside, the results have been very impressive, indeed! The Chi-Town Fermi Level Nihilists (that’s their bowling league team name) have soldered semis together with absolutely zero change in electrical properties. Considering that with this new form of solder, we will be able to three-dimensionally print and mold electro, opto, photovoltaic electronics on an incredibly small scale, this is a massive, Nobel-worthy achievement.

—Jimmy Ray


Adaptive Segmentationmicro-segmentation
Share this post: