At MWC SemiAccurate met Xtalic, a company making a better gold for contacts through nanotechnology. Actually they are making a better material than gold for contacts and a cheaper one too.
The word nanotech has been vastly overused but Xtalic is doing it right for the right reasons. If you are wondering how one can do better than gold, a very corrosion resistant metal with great electrical conductivity, there are actually a number of ways. The main one is cost and even with high production costs, Xtalic should be able to beat the price of gold. More importantly is corrosion resistance.
1, 3, 10, and 180 minutes in the tub
The four samples you see above are pins for a connector, USB if memory serves. Three are gold plate and one is coated with Xtalic’s LUNA(R). Hint: It is the non-gold colored one. All four were placed in a bath for corrosion testing, the number of minutes of immersion are written on the sample, 1, 3, 10, and 180 minutes. As you can see the gold-plated materials showed visible corrosion at 3 minutes and by 10 minutes were falling apart. The Xtalic coated samples were just fine after 180 minutes.
Most people who took chemistry are probably pretty confused by this, gold doesn’t corrode without a very nasty cocktail called aqua regia which is not used in the test bath. Why do the gold coated contacts corrode then? Porosity, basically gold coating are porous which lets the corrosive fluids through. This then chews up the metals underneath and it mechanically degrades. This is a fancy way of saying it falls apart, a few microns gold is not a good structural material.
The way Xtalic makes its coating is of course nanotechnology, more specifically controlling the grain size of of the coating through electrodeposition. The ones we are fixated on are the non-porous varieties though, and as you can see they work pretty well.
Why is this important in the real world? There is the obvious problem of gold being really expensive, insert a joke about it being the gold standard for coatings here. Needless to say Xtalic says they are 50-80% the cost of a gold coating with the same electrical properties and less porosity. Believe it or not the bit about porosity is probably the most important of the three.
Porosity matters because in some places the air is really nasty. If you have ever been to Asian manufacturing cities on a sunny day you now what I am talking about. In general pollutants in the air form quite acidic compounds when they get wet, you know from rain or humidity. Or lungs but we will ignore that corner case for this article. Electronics put outside in such environments will begin to show corrosive effects after three or so months so gold plating is a minimum requirement. As you can see from the above pictures however it is not a panacea.
So that is where Xtalic comes in, non-porous so airborne corrosive compounds don’t get in, plus it is cheaper to produce than gold. More interestingly it said to have the strength of steel, not that it matters much when it is a few microns thick. Since it is made via electrodeposition it is a plating process, not a forming one, so it should be easier to make complex and delicate traces with. See?
Update 4/25/16 @ 11:50am: The ultra-strong alloy is not the gold LUNA(R), it is their Aluminum alloy that has the added strength.
Traces for USB/SD or similar contacts
Overall Xtalic has a very interesting technology. It is cheaper, better, and about the same as the incumbent depending on which aspect you are looking at. It uses some common metals and controls the nanocrystaline structure during deposition. Xtalic calls this chemistry, process, and technique, and from our hazy college chemical engineering days, we think that is quite appropriate. It will be interesting to see how widely this material spreads in the market.S|A
Update 4/25/16 @ 11:50am: Several changes made to the materials names and properties after an update from Xtalic. The versions above should be more accurate.
Latest posts by Charlie Demerjian (see all)
- Intel kills off the 10nm process - Oct 22, 2018
- HyperX ships it’s 60 millionth enthusiast memory module - Oct 15, 2018
- Bittware/Nallatech water cools 300W of Xilinx FPGA - Oct 12, 2018
- More on Intel’s 10nm process problems - Sep 17, 2018
- Intel puts out another 14nm 2020 server platform - Sep 11, 2018