There is a notion that radiation may be limiting qubit lifetime. Even if you use low-radiation materials and shielding, you still have to worry about muons generated by cosmic rays. They are very penetrating – some reach kilometers down into the rock.
If so, we may need to build quantum computers several klicks underground. The deeper, the better: another factor of ten attenuation of the muon flux could materially extend qubit lifetime and drastically improve performance.
We could perhaps use the old Homestake mine in South Dakota, now the location for DUSEL (Deep Underground Science and Engineering Laboratory ). But it’s only 2.4 km down.
The best mines ( for our purpose) are more than 3 kilometers deep, and a few approach 4 km. The problem is, of all the ten deepest mines in the world, only one is in the US.
The United States faces a mine shaft gap.
West Hunter 
Or build the qubits “rad hard”, so plot the probability of that being achieved against the cost of building a whole facility down the deepest US mine, and compare with the cost of lead bricks.
Iron is better and cheaper, which has the best combination of primary shielding and minimal secondary neutron production. Lead is very bad poison.
Alternatively go to South Africa which has lots of deep mineshafts.
The crooks in SA will never let us near there.
Here is idea: the Moon. The maria-heavy near side used to be tectonically active. It has lavatubes which already get us part of the way down under there.
Also, whatever minerals the South African precambrian craton has, so does the Moon.
A mine on the Moon should get us the needed 4 km down.
When TV villains ask our hero whether he wants to do it it the easy way, or the hard way, they can quote you to show just how hard it can be.
This post feels to me like it is building up to a pun. But I have no idea what “mine shaft gap” could refer to. Too many “muons” on the other hand is a common problem …
I think it refers to the ending of Doctor Strangelove. The russian ambassador reveals they have a doomsday device that will automatically retaliate, irradiating the entire surface of the world. The war room realize they will have to live underground for centuries, and start talking about the “mine shaft gap”, as a parallel to “missile gap”.
I didn’t even think about Dr Strangelove. A mine shaft is literally a gap.
A reference to the notorious “Missile Gap” of the 1960 election.
Rather, the excellent Kubrick film Dr. Strangelove, which itself referenced the missile gap: https://www.youtube.com/watch?v=ybSzoLCCX-Y
Of course. Dr Strangelove. Nine attractive girls for each man.
10, not nine.
OMG I forgot about that – and I’ve seen it about a dozen times.
“Mineshaft gap” is basic, Science Fiction Fandom 101 reference. Even many mundanes get it.
To understand this blog, you need to grok SF Fandom 201, 301, 401 stuff.
This is non-negotiable.
Either you are trolling or you are mundane who really does not belong here 😦
So, where does being an insufferable doofus fall on the list of requirements for blog participation?
I have actually seen “Dr. Strangelove” and remember it quite well. Just not in English. Inconceivable, I know. On the other hand, I don’t really care whether I belong here or not.
“Mineshaft gap” is basic, Science Fiction Fandom 101 reference. Even many mundanes get it.
To understand this blog, you need to grok SF Fandom 201, 301, 401 stuff.
This is non-negotiable.
Either you are trolling or you are mundane who really does not belong here 😦
I only come here for the Firesign Theater references, myself.
Not mentioned here but valuable in that genre was The Credibility Gap, especially Woodshtick: 49 baggy-pants comedians locked in a hotel room in upstate New York.
The group eventually included lots of big names before they were big: Michael McKean, Harry Shearer, David Lander, Albert Brooks, and Christopher Ross.
I only get half of the SF references because I was a Society For Creative Anachronism nerd myself. A college roommate of mine who barely graduated still leads pirate raids every year in North Carolina.
Regrettably yes, but you know it is a sacrifice required for the future of the human race.
Aren’t those South African mines not useful for that purpose because of the South Atlantic Anomaly thing, which could enable cosmic rays to go even deeper? Or it has nothing to do with that?
Good one, Greg!
Don’t forget about neutrinos interacting to produce electrons, or muons, or taus. See https://icecube.wisc.edu.
I was in a pizza joint not long ago and ran into a dude with a t-shirt that said something about living (or getting mail?) at the south pole, and it turned out he worked at that facility.
Small world…
“The problem is, of all the ten deepest mines in the world, none are in the US.” Oh never mind, some of the potholes on your roads will soon enter that category.
Am I wrong, or is it true that quantum algorithms never hangs?
Sure, Kubrick. But do mind the Gemeinschaft–Gesellschaft gap.
The Simulation could only have been built with quantum technology. But what happens when the sims themselves develop quantum computing – the fabric of space-time could literally be turned inside out. These muons seem all too convenient.
No. 7: Empire mine, 3.4 km, California?
Taking this literally, rather than as a science fiction reference, the limiting factor on mine depth isn’t cost or mining tech. It’s the temperature of the rock. Assuming you have a few hundred million dollars you can readily sink a 10,000 foot shaft anywhere in theory. But most parts of the Earth it will be too hot for humans to work that deep, even with an ice slurry plant consuming enough power to run a small city. Only certain parts of cratons, the ancient cores of continents, have a low enough geothermal gradient. Like Kaapvaal craton with the world’s deepest mines around Johannesburg or the Baltic Shield where the Soviets drilled the record 40,000 foot borehole.
Presumably on the principle that Marx was a 40,000 foot bore?
So we need to dig ourselves into an even deeper hole?
A muon is a bit like a very fast moving hand grenade. It has a fuse of about one microsecond in round numbers and it doesn’t have to hit the target dead on, it just needs to get close. You’d think that would be good enough. Light only travels about 300 meters in a microsecond. You’re 3000 meters deep. No problem. But, we don’t live in a Newtonian world. We have to consider special relativity. Because the muon’s fuse is ticking in the muon’s reference frame, not the target quantum computer’s reference frame, and the close you get to the speed of light, the lower time moves in the muon’s reference frame relative to our reference frame.
A muon has a mass of about 105 MeV/c^2 (I’ll omit the c^2 going forward adopting the convention that c=1). For the fuse to blow on the muon safely before it hits our quantum computer 3000 meters down, we need to get time ticking in the muon’s reference frame as of the time it hits the surface of the Earth up to the rate its kinetic energy is down to about 90 MeV. And, a muon, is a negatively charged particle with the same charge as an electron (although we have to worry about antimuons which are positively charged as well) so it will lose momentum if it encounters an electric field, for a regular muon, either a negatively charged electric field below it, or a positively charged electric field above it. the trouble is that some muons come in with lots of kinetic energy. The peak of the distribution is only about 300 MeV which is manageable to slow down to what we need it to be, but the most powerful one every detected has about a trillion times more kinetic energy. Shielding is just a really crude version of trying to slow down the incoming muons and it helps with the ordinary ones, but it has its limits. And, if we can get the barrier that reduces the kinetic energy of incoming muons higher in elevation (e.g. at the top of a very tall tower or nearby mountain peak or in a weather balloon), the muon has more distance to cover which leaves us more distance for the fuse to run, which gives us more of a margin of error allowing us to get the kinetic energy down to merely 120 MeV perhaps. We might not be able to have an EM field shield at that elevation 24-7 to stop every stray muons, but sometimes cosmic rays come in bursts that can tip us off to amp up the protection, and we aren’t necessarily operating our quantum computer 24-7 either.
But, even if we can’t slow down the muon enough to get it to blow its fuse before it hits our target, the effort is still worth it, because the more we slow it down, the more effective the next phase of our defenses can be.
The more sensible option is to take advantage of the fact that the part of our quantum computer that we want to protect is small (0.1 meters or less), and if we can keep the muon even say, 0.4 meters away from the target, we’re probably fine because the electromagnetic charge of an isolated fast moving muon isn’t very big and decays with distance squared. So, basically, all we have to do is keep the high energy muons that the shielding doesn’t stop out of a space the size of a beach ball.
To do that we also want an electromagnetic field, in classical electromagnetism terms, one with a hyperbolic shape above our target, to pull incoming muons on a course towards our target slightly off course so it misses our target. All we need to do is nudge the course of the incoming high energy muons bound for our target by about 0.5 meters over the 3000 meters it travels from the surface to our target, which is much easier than slowing it down enough for it to blow its fuse.
Alas, unlike classical electromagnetism which can erect a perfect force field wall that way, we have to deal with quantum electrodynamics to protect our quantum computer, which means that the incoming muon can tunnel through our force field and that the more powerful it is, the easier it is for it to do so. So, every once in a while a muon is going to slip through and get too close to our computer and f- things up and that’s just a cost of doing business with quantum computers. But, are classically hyperbolic shaped electromagnetic deflector field should be a lot more efficient than the brute force shielding from digging deeper mines which faces inherent limits because the Earth’s core it hot and emits its own radioactivity that the portion of the cruse below our quantum computer is also shielding it from.
But, really we’re looking at a three part solution. First, an electric field shield for selected times when due to critical quantum computer operations or a tipoff of incoming cosmic rays, we are on high alert, second, the brute force shielding of putting our computer deep in a mine shaft the stops the routine easy stuff, and third a deflector oriented electric field hyperbolic in geometry to nudge incoming muons that the first two levels don’t stop enough off course that it will usually miss the target.
Should be obvious but at the end of the first paragraph “the closeR you get to the speed of light, the Slower time moves” and in the next paragraph “we need to get time ticking in the muon’s reference frame (as of the time it hits the surface of the Earth) up to the rate IT HAS WHEN its kinetic energy is down to about 90 MeV.” Later in the same paragraph “the most powerful one EVER detected”. Probably a few other minor typos too.
A question for the readers to ponder: are quantum computers actually viable? Analog computing has been shown to be difficult enough, I’d think.
Since there are many approaches, almost certainly possible. Analog comparison – not appropriate.
You have this point of view: https://www.quantamagazine.org/gil-kalais-argument-against-quantum-computers-20180207/
This kind of theories, of course, depends on simplified models. So in practice it could turn out false.
If the Mohole hasn’t been filled up, stick ’em down there.