Woven Heat Sinks Are Next?

Seems this fiber has greater heat conductivity than copper. The only common materials that are better are silver & diamond.
http://www.physorg.com/news/2012-03-spider-silk-metals.html

This ought to lead to some interesting developments indeed.

I thought I had read a while back that they could duplicate the actual silk, just not spin it.

Maybe some use for that "failed" chemical?

now we just need to genetically engineer spiders that make superconducting silk.

The last I heard of this was probably 10+ years ago when they were looking at it as structural material for a possible space elevator.

They had succeeded then in engineering goats to produce it in their milk. Spider goat, spider goat, does whatever a spider goat does.

But as Geek said, after filtering they couldn't get it into very useful fibers.

http://www.bbc.co.uk/news/science-environment-17232058
Spider silk spun into violin strings

Though, it's not made of spider silk, they do have flexible heat conducting sheet using graphite that is better than copper or silver, and approaches diamond.
Pyrolytic Graphite Sheet

Digikey:
http://search.digikey.com/us/en/products/E...11438-ND/429229
Datasheet:
http://industrial.panasonic.com/www-data/p.../AYA0000CE2.pdf

Amusing...
Features:
Excellent thermal conductivity
...
Low thermal resistance

Apparently they don't have much else to say about it?

Data shows high thermal conductivity IN PLANE ONLY. This is lateral conductivity, which is nice for spreading heat, say over a circuit board or plastic housing, but doesn't help as a filler pad. Modulus or hardness is not specified so it's impossible to tell how it would do at filling gaps. Most likely, it's not as soft as a silicone pad (not that silicone pads have much thermal conductivity anyway). It's of no value for thermally conductive insulators, because it doesn't conduct.

Since it's laterally conductive, it would be of the most value in increasing the heat dissipation capacity of a circuit board. Most boards are made with 35um or so copper (i.e., 1 oz), which spreads heat efficiently for about half an inch distance (i.e., the diffusion length), meaning you have diminishing returns on heatsinking value for copper pours larger than an inch across. This particular sheet is 100um, which if it were copper, would increase the diffusion length to about 1.7"; because it's about twice the conductivity, this material extends that further, to a region about 2.4" across. That's about 6 times the area, so the dissipation will be that much higher -- about 6W versus 1W without using any more copper on your circuit board. Curiously, because the conductivity varies so much with thickness, you might find a thinner sheet has nearly the same properties.

Because it's another layer, the sheet does allow you to fill the area on your circuit board with more components, which would've otherwise been wasted on heat spreading area. It doesn't allow you more heat dissipation on a row of power transistors -- area is area. For that, you need more cooling power (fan, heat sink, heat pipe, etc.).

Since the thermal conductivity is double, the equivalent thickness of copper is 0.2mm (7.8 mil, ~5.8 oz) foil. The next closest McMaster product is 7 mil, 2" x 50ft = 0.77 m^2, and costs 80.99 USD. The above product is only 0.0225 m^2 and costs 83.71 USD at Digikey pricing. By area, you get 34 times more in copper than PGS.

Copper has higher electrical conductivity, giving some useful magnetic shielding at most frequencies, is probably about as formable (but does fatigue after repeated plastic deformation), can be stamped, drawn, rolled, etc., and can be soldered. It still must be insulated where insulation is required, and various tapes and coatings serve the same purpose equally.

I must say, aside from the flexibility, I really don't see the point to the stuff.

Tim

Easier to work with. Easy to cut in whatever form required.Used, per example, to line the casing of notebooks. Where .1mm matters.

Yeah, silicone pads don't conduct heat well, do they. Neither do most thermal compounds. I was quite surprised when I compared relative thermal conductivities.
It was after learning this that I started including a flat lapping procedure in my projects involving hi power heatsinking. I've found heat conductivity goes up substantially if both surfaces are lapped flat within .003mm or better. It's amazing just how un-flat extruded heatsinks & device bottoms are. I've seen salvaged units with more than .2 mm worth of uneveness.
This is also the reason I strongly prefer mica insulators, because mica can be split down to incredible thinness. I'm not sure what the limit is, but I've gotten TO-220 sized pieces down to under .013mm.