Re: A route to room-temperature superconductivity?

John Schutkeker wrote:
  Quoted Text Here  

Not true. If you could get to RTSC with a diamond-anvil pressure cell, you could put a SQUID magnetometer in there that would be useful for lots of stuff. Maybe even a Josephson junction CPU chip that would run Vista XP Pro at a reasonable speed. What killed IBM's Josephson junction project was the crazy way materials would break down when cycled repeatedly to cryogenic temperatures.

Mark Thorson wrote:
  Quoted Text Here  
I agree that a room-temperature SQUID would be a very useful device, even though it wouldn't be nearly as good as a low-temperature SQUID. Hypres used to make a 100-GHz sampling scope based on SQUIDs, but it needed liquid helium. Pretty amazing device for its time, though.
RE the IBM Josephson project:
They stayed with Pb junctions for too long, and those had the temperature cycling problem in spades, just as you say. They eventually went to niobium edge junctions, which survived very well, but by then it was clear that the speed advantage wasn't enough to justify the requirement for cryogenic temperatures. (I wasn't on the project, I'm about 10 years too young, but I know a bunch of the people who worked on it.)
Cheers,
Phil Hobbs
IBM T. J. Watson Research Center Yorktown Heights NY

  Quoted Text Here  
Bzzzzt, wrong answer! You don't honestly think that they won't break down equally unpleasantly when cycled repeatedly to diamond anvil pressures, do you?
A liquid nitrogen refrigerator is an infinitely cheaper, and much larger, place to build complex widgets, than a diamond anvil. What we really need to bridge the gap between liquid nitrogen temp superconductors and room temp superconductors is a dry ice temperature superconductor (-60F, IIRC), because dry ice refrigerators are even cheaper than liquid nitrogen refrigerators. ?:D
IMO, what makes Josephson Junction computers uninteresting is that they're not a better mousetrap, just a different mousetrap.

John Schutkeker wrote:
  Quoted Text Here  
Larger, sure, but all we're talking about putting in it is a SQUID. You can make those lithographically on one half of the anvil, and then assemble it with a big preload, using some sort of temperature-compensated mount. It ought never to have to be disassembled, so in principle there's no cycling involved. (That of course assumes that the metal seal on the cell didn't ever leak.)
The key thing about cryogenics is that they need constantly power and continual babysitting.
Cheers,
Phil Hobbs

  Quoted Text Here  
I knew you were going to say this, and now we've descended to the nadir of bickering about what's complex and what isn't. According to my definition, anything under the purview of "professional science" is by definition complex. Other categories, like industrial, commercial, hobbyist and amateur science are by definition easier.
Both steps you described above are intricate and difficult to implement, and I shudder to think about the complexity of any steps you've left out. I'm not saying that it can't be done, but just that it's not *obviously* competitive with other ideas, for research dollars. Other ideas can provide any or all of the following benefits - quicker payoff, more important area of research, less effort, cheaper equipment.
  Quoted Text Here  
Sure, but the same was true of the IBM experiment. Theoretically, once you've got your widget working, it's reasonable to require that the user never allow it to return to ambient temp & pressure. In practice, this is yet another headache for the user, and it only takes a couple like that to make them throw up their hands and say "It just ain't worth the hassle." ;(
  Quoted Text Here  
I'm sure that the anvil requires no less power or babysitting than the cryo, and cryo setups are much more familiar to lab people than the anvil. Familiarity = low cost + ease of use.