Silver Solution Ink Makes Faster Flexible Circuits

judgecorp writes "Silver-based compounds dissolved in ammonia could make finer and more flexible circuits, according to researchers at the University of Illinois. Existing inkjet based circuit printing systems use particles which are less predictable. The silver-based ink remains dissolved until the ammonia evaporates, and can be delivered through 100nm nozzles. In all senses, it's a better solution."

Comments

"Solution"... get it?

[jpwilliams]

Puns FTW!

Re:"Solution"... get it?

[Cyberax]

To nitpick, Russian space vehicles were called "Salyut" (which is Russian for "Salute").

Corrosion?

[Kenja]

There's a reason copper or gold is used in circuit boards despite silver being a far better conductor, how does this new solution avoid corrosion?

Re:Corrosion?

[Sponge Bath]

This suggests corrosion is not a problem. Highlights: "It is quite resistant to corrosion and does not oxidize easily, although it readily forms a surface tarnish of silver sulfide." and "Silver plating is widely used on contacts and other conductive parts in electrical apparatus such as switchgear and motor control centers because of the superior conductivity and longevity."

Re:Corrosion?

[introcept]

It doesn't need to, it should be fine under a layer solder mask/conformal coat/lacquer with plated contacts if necessary.
In a flexible circuit you'd be sandwiching it between two layer of polymer anyway

Re:Corrosion?

[davester666]

That's why everybody uses gold contacts. It's more expensive [and people see the word 'gold' and automatically overvalue it compared with most anything else, particularly if it uses the word 'silver'], and is also wears out faster through use, hence also needs to be replaced more often, while providing no noticeable difference in actual performance.

Re:Corrosion?

[LordLimecat]

Copper corrodes far worse than silver does.

Re:Corrosion?

[wierd_w]

Assuming this is metalic silver being deposited (this process seems eerily similar to traditional silver mirroring processes with ammonia, dextrose, and silver nitrate) then it can be rapidly plated with gold using a solution bath. (Gold dissolved in nitric and hydrochloric acid mixture)

An interesting idea would be to print the two solutions (dextrose + ammonia) + (silver nitrate) with a dual head printer, then print dissolved gold after it has had time to deposit.

This would chemically replace/plate (depending on thickness) the silver traces with gold ones. A simple distilled water wash afterwards would clean up the piece.

Makes me wonder if I could repurpose a color print cartridge for this process. I strongly suspect that the nitrate and aqua regia solutions would be very very bad for the printheads though. (Dextrose + ammonia solution would probably just clog and nothing more.)

Re:Corrosion?

[Anonymous Coward]

There's a reason copper or gold is used in circuit boards despite silver being a far better conductor.

Silver isn't a better electrical conductor than copper, it's a better thermal conductor.

Silver is tremendously better at conducting heat than aluminum, and slightly better than copper and gold. In the past, I've melted blobs of silver solder over flat aluminum heatsinks, mostly to absorb temperature spikes (to steady temperatures).

How does this new solution avoid corrosion?

It isn't really necessary to avoid the corrosion of silver in most cases (in breathable air). Unlike copper oxide (a semiconductor/insulator), or aluminum oxide (insulator), silver oxide remains a decent electrical conductor. Conducting electricity despite oxidation is why it's great for electrical connectors, because all bare metal corrodes in the presence of oxygen (except gold; most of them form protective oxide layers on the surface like stainless steel, but most oxides have high electrical resistance).

Regardless of the metal being used, you can protect the circuit traces against abrasion, shorts and corrosion with a protective coating. Tinning the traces would be an obvious choice, if the heat from soldering doesn't melt or burn the PCB material. Smearing silicone sealant across it might work, but any general purpose PCB coatings will just flake off something flexible.

Corrosion isn't always oxidation, though. Another poster mentioned silver sulfide, which is different (aka silver tarnish). Oxides and sulfides of metals have different physical properties, silver oxide conducts but other reactions may or may not. Keep in mind you'd only get silver sulfide from hydrogen sulfide (air pollution or other biological sources). So, unless you're eating off the circuit board, or expose it to smog, oxidation is the primary type of corrosion.

Re:Corrosion?

[tmosley]

Uhh, silver is the best metallic conductor of electricity their is. Get your facts straight.

Every fucking month

[atari2600a]

We get a Silver Ink story, as if it hadn't been invented years ago available at any Frys electronics ready to mend any severed circuit.

Re:Every fucking month

[tobt4josh]

The fact that this new ink reaches the bulk conductivity of pure silver upon annealing is nothing short of amazing. My company has been working on some printed electronics application, and most of the conductive inks (including silver based) have conductivities that are too low to conduct any useful amount of current.

Re:Every fucking month

[atari2600a]

Depends on the scale I guess-- I have a cousin that once used a silver ink pen to repair a garage door opener. Not sure where it was on the circuit but something tells me if the circuit fractured to begin with it must have been one of the 5HP move-y bits

Re:Every fucking month

[LuxuryYacht]

There are a few silver nanoparticle inkjet inks on the market with very low resistance. ~0.2 ohm per square for water based and even lower for solvent based inks.

Re:Every fucking month

[Hank the Lion]

There are a few silver nanoparticle inkjet inks on the market with very low resistance. ~0.2 ohm per square for water based and even lower for solvent based inks.

I wouldn't call 0.2 ohm per square low resistance.
If I would print a trace of 10 mils width with a length of just one inch (1000 mils) with the ink you mention, it would have a resistance of 20 ohms.
This system must have far lower resistance to be useful.

Re:Every fucking month

[LuxuryYacht]

How much current do you need to pass through your circuits?

http://onelabs.com/prntelec0000.htm has systems for even lower conductivity fluids.

Re:dpi?

[SmlFreshwaterBuffalo]

I'm sure it also has to do with the resolution of the nozzle positioning system as well as the spread of the droplets once they hit the surface. To get true 10 million dpi, I think you would need to be able to create 2 dots 100 nm apart without them shorting together.

Of course, it sounds like the resolution is still much better than existing technology, just not 10 million dpi.

Link to the research group doing this:

[Hartree]

http://colloids.matse.illinois.edu/

Jennifer Lewis' research group here at the University of Illinois did this work.

They've got a link on that page to a youtube video that shows how to make and use this conducting ink, but it goes through Boing Boing and is down at the moment due to the SOPA protest.

Here's a direct link: http://www.youtube.com/watch?v=dfNByi-rrO4

Seriously cool work.

Not this sense

[Prime Mover]

> In all senses, it's a better solution
Ammonia? Not in my sense of smell.

Tell me when they have a silver/Mendocino Oatmeal Stout solution.

Seem to need a big shipping crane + tracks

[Joe_Dragon]

Maybe this will work in a open area building stuff in a line but trying to fit that in to a area with other stuff in the way? Maybe to build / puttogether parts of the crane on site.

What hills and and places with uneven ground?

Re:Seem to need a big shipping crane + tracks

[similar_name]

I believe you're looking for this.

Forget faster circuits

[Trogre]

I want to see this tech used in cables, particularly headphone and low-voltage power cables.

This is assuming it would fare better than braided copper.

Re:Forget faster circuits

[aaronb1138]

I have a feeling unless the printing was done braided, the skin effect would still be an issue in the analog world. On the other hand, you might be able to get Litz windings incredibly cheap compared to the current methods. Too bad the marketing folks will hear Litz and increase the cost 10x anyways.

Re:Cage match: Techies vs Silver Hoarders

[tmosley]

Considering that this particular incarnation would likely use a highly processed and EXPENSIVE form of silver (last time I checked, nanosilver cost about $10,000 a kilo, about ten times the current price, I doubt you would have any problem getting them to sell it to you.

And you should also note that silver is dramatically underpriced as it is, and as a result, we have eaten through literally thousands of years of stockpiles. World stockpiles are the lowest they have been since the time of the Spartans, yet as this article demonstrates, more and more new uses for the remaining silver keep appearing.

How to Fill a Computer printer cartridge

[chrisphotonic]

Now we just need an order form for this, and a good how to, on filling up a used printer cartridge with this ink.

I've always wanted to build a small CNC machine to make custom PCB circuit boards, but with this, I might not have to.

The right solution

[gnarlin]

I myself recommend a seven percent solution for the best effect.

Re:dpi?

[RockDoctor]

does that mean 10 million dpi, or does it not work like that

No.

The 100nm dimension would be precisely what it says : the diameter of the nozzles. But the nozzles are one end of a chamber in which the droplet accumulates before being fired - and the chamber is of larger diameter, otherwise it wouldn't be a "nozzle", would it?

The chambers also have mechanisms in their walls, principally the heaters that vaporize the ink, to propel the droplet out of the nozzle. Those heaters aren't going to be "small". They may have other mechanisms, such as attitional drainage pipes for when the ink head is being "cleaned".

Then surrounding that will be the mechanical supports to hold the nozzle in place, and to maintain the correct relationship from nozzle to nozzle.

So, your actual ink-jet mechanism is going to be a lot larger than your nozzle size.

This being Slashdot, I suppose a car analogy is in order. A car is generally a way of getting one person from one place to another. A person is 0.5m in diameter (most countries ; 1m in America), so you can fit 20 efficient cars side-by side on an 10m-wide roadway (5/6th of a dozen cars in America, in non-metric terms).