3 ms·
> sudden resistivity changes at a critical temperature (bizarrely high though that is in this case) How high? Just wondering if it might eventually be possibl
by cma 3y ago
> sudden resistivity changes at a critical temperature (bizarrely high though that is in this case)
How high? Just wondering if it might eventually be possible to use it in some power/clocking distribution layers of a semiconductor chip without bulky cooling.
- jacquesm 3y ago127C!
- elihu 3y agoThat's a good idea. Honestly, I find it pretty amazing that modern CPUs don't just instantly turn into a puddle of molten metal as soon as you turn them on. The critical temperature is 127 C, as the other commenter mentioned. That seems like it's high enough to be useful in a computer CPU. I suppose there might also be applications for power distribution in a data center. You might use a few large, high-efficiency low voltage DC power supplies to supply the whole site instead of hundreds or thousands of individual power supplies if electrical resistance in cables was less of an issue. (I think this sort of thing is done now, but more at the rack level, since long cable runs cause losses.) You might even use 1V power instead of 5V or 12V. (Granted, superconducting materials don't have unlimited current capacity, and bad things happen if you exceed the limit.)
- axiomaticdoubts 3y agoThe critical temperature is not 127 C. They did not actually measure the critical temperature. They just said it's at least 400 K (~127 C), probably because that was the maximum temperature the equipment they used could measure.
- pjc50 3y ago> Just wondering if it might eventually be possible to use it in some power/clocking distribution layers of a semiconductor chip There's an additional obstacle factor here which is that the material has to (a) be "sputterable" onto the surface, which may rule out things with fancy crystal requirements, and (b) not adversely react with the rest of the chip, which is why coppper is ruled out (!) and chips use aluminium for metal layers.
- cma 3y agoCopper replaced aluminum some time ago (starting ~1997) regardless because of the resistivity, and they just had to find different barrier materials to prevent electromigration and overcome other issues.