Decoding Quantum Computing

IRA FLATOW: This is Science Friday. I’m Ira Flatow. Lately, I’ve been thinking a lot about disruptive technologies. And by that, I mean new ways of doing things that, if they pan out, could dramatically change the way we work and live. And one of those things is quantum computing, a way of doing calculations not with the ones and zeros of electrons flowing through transistors and electronic chips, but by harnessing the strange physics of the quantum world.

I’ve heard a lot about quantum computing, as I’m sure you have. But how does it work? Why should I care? I want to know. So I’ve asked an expert who lectures on this subject to come on the show.

Dr. Doug Stone is deputy director of the Yale Quantum Institute and Carl A. Morse professor of applied physics at Yale University. He’s also author of Einstein and the Quantum, one of my favorite books. Welcome back to Science Friday, Doug.

DOUG STONE: It’s great to be back, Ira.

IRA FLATOW: Nice to have you. As I say, you have lectured a lot about why should we care about quantum computing. In a nutshell, can you tell us why we should care?

DOUG STONE: Well, this is, to me, the most surprising thing in modern physics in 50 years. When I went into graduate school, there was no such thing. And all these other things like Higgs boson and gravitational waves, people were already thinking about, but this idea of a quantum computer was completely out of left field. It promises to give us solutions to problems that are at the forefront of technology and health and understanding the environment and so on. It gives us access to a computing technology that has never existed and could be revolutionary if it all pans out.

IRA FLATOW: OK, let’s talk about some of the details. First, let’s define terms. What exactly is quantum computing?

DOUG STONE: So the important thing about quantum computing is that the information is not stored in the forms of ones and zeros, which are electrical signals in a standard computer. But instead, it’s stored in terms of some quantum object. It could be an atom or molecule. Or often now we have artificial atoms, nanostructures that behave like atoms and behave in a quantum way. And the most important thing is the information is stored not as a one or a zero, but as what we call a superposition of a one and a zero, something that could, depending on each measurement, be a zero or a one, but actually also has additional information to specify what its state is.

IRA FLATOW: Sounds a little bit like Schrodinger’s cat.

DOUG STONE: Yes, and I would say the most exciting and surprising thing in this whole area is that having things be uncertain, you would think would be just bad for a computer. Why do you want to start with bits that are actually unstable and might be zero or one? It’s hard enough without that, you would think. And the surprising discovery of mathematicians, theoretical physicists, is that actually, that can be harnessed to do essentially enormously parallel processing. And that allows you to do things that conventional computers will never be able to do.

IRA FLATOW: I mentioned your book about Einstein in my introduction. And I know that you have written a lot about Einstein and the ideas of entanglement. How key is that to understanding quantum computing?

DOUG STONE: It’s very key to understanding it, which is somewhat unfortunate, because it’s a very subtle and difficult concept. But it has been proven that a quantum computer will not have an advantage over a conventional computer if you don’t use entanglement, which is a property of quantum states. And it’s a subtle property, but let me try to explain it.

So imagine I have two coins, heads and tails. We’ll call them magic coins. And they have the property that each one independently is a fair coin, half heads, half tails, randomly when you toss it. But when you toss them both, if one is heads, the other is also heads, or if one is tails, the other is also tails. They’re perfectly correlated.

And so you might think that’s not very valuable, OK, because I still have these random coins. I don’t know if they’re going to be heads or tails. But it turns out you do know something. What you know is you’ll never have heads, tails or tails, heads. And therefore there is information stored in the entanglement, which can then be exploited and processed.

IRA FLATOW: Interesting. If you were to point me to a box and say, over there is my quantum computer, and I opened up the box, what would I see inside that box?

DOUG STONE: Well, that’s a great question. And stepping back a little bit, nobody in 1990, almost nobody, thought that a quantum computer would be useful in any disruptive way. Maybe just as a curiosity. And therefore, no one had seriously said, what’s the best way to do this?

And so once it became clear from the mathematical analysis that it could be enormously useful, then people started saying, what’s the best way to build it? It was completely open. So there are designs where the quantum bits are molecules or atoms or ions. And there are designs where the quantum bits are photons. And then there are designs where the quantum bits are sort of electronic excitations, more like a conventional computer.

And the latter is the one maybe I’ll tell you about, because it’s the one we do here at Yale and we’ve pioneered. And that’s where you have microwave circuits fed in with cables. So the cables would look almost like your cable TV. But then they go into a big can, which is incredibly insulated, and then cooled down almost to absolute zero so that the circuits will behave like quantum objects and not like your light switch.

And you can actually use them to store quantum states. In your conventional computer, your one and your zero is a voltage. It’s not fluctuating much. It’s a definite value.

Your quantum computer, your qubit, whatever it is, whether it’s a microwave circuit, a superconducting circuit, or an ion, is in this uncertain state. And any interaction with the environment will tend to measure it and make it fall into a definite state. And then you lose the whole quantumness that allows you to do these unprecedented computations.

IRA FLATOW: What are the properties of the quantum state and the qubits that make the computer, a quantum computer, so much more powerful? You talk about how much more powerful a quantum computer is than, say, an Intel-based processor.

DOUG STONE: It’s a tricky thing to quantify because it’s not that it runs fast. It just runs on different principles. It might run slow or at same speed as an Intel processor, but it’s doing things in a quantum parallel way that no Intel processor can do.

So the example I’ll use is to think about storing ones and zeros in a memory, in a register in a memory. Let’s just choose 300 bits, not the billions in your computer. OK, 300 bits is 2 to the 300 or 10 to the 91. That’s a trillion trillion trillion trillion trillion trillion trillion, maybe, if I said that the right number of times, states even with 300 bits. So if you go through and try to create all the different states of one and zero in your memory, it would take you longer than the age of the universe to even store all the different states or process through them. Not that you’d want to do that, necessarily, but it’s an indication of the limitation.

Now, with a quantum computer, I just need 300 steps, if there’s 300 bits, to create all the different possible states. I won’t go into the details of how. That’s 300 steps that I can do in a microsecond. So in a millisecond, I can do something that you could never do on a conventional computer in the age of the universe.

IRA FLATOW: Wow. That really is fast. Do you see quantum computers in general use, or would one be purposely built for a specific problem?

DOUG STONE: Definitely the latter, for two reasons. First, quantum computers are not actually great for handling big data. It’s really for computing things that are too difficult to compute with conventional computers, maybe couldn’t be computed in the age of the universe, as I was just saying.

So where you have a big tough compute problem, you would have a special-purpose cloud quantum computer, and it would do that for you. And it would solve that part of the problem. But the big data storage or just the everyday types of things you can already do well with conventional technology will not be replaced.

IRA FLATOW: So what kind of problem are you talking about, then, a typical problem?

DOUG STONE: So one kind of problem is the routing problem. If you have to go to 100 different locations in some order, and you have to figure out the most efficient way of doing that, that turns out to be a very, very hard problem. It’s called the traveling salesman problem. And it’s very hard to solve on a conventional computer. It’s exactly the kind of problem you would want to solve for routing on a network, and that may be why Google is so interested in quantum computing.

IRA FLATOW: Interesting. Some technologists are saying that if and when we get quantum computers, that will be the end of e-commerce over the internet, as your encrypted passwords could all be broken by someone listening in and possessing a quantum computer. Is that a valid concern?

DOUG STONE: Well, I would say not, because I think people are developing quantum-secure technologies. And we have no general proof that quantum computers can solve all the different encoding schemes that work pretty well on the internet. But it’s interesting you mention that, because the thing that kind of just turned everybody sideways and created this mad rush to quantum computing was the mathematical proof that one of the main forms of encryption, the RSA encryption, which is used for securely accessing your credit card on the internet– that could be broken by a high-end, well-functioning quantum computer, which we do not have yet. But that was how the sort of gold rush started in 1994, when that was proven.

And there’s this other interesting kind of duality to it, which is that we can prove that with quantum channels, you can have unbreakable codes. So if you’re willing to switch to all quantum channels, there actually– you would be completely quantum secure.

IRA FLATOW: Do we have to redo the whole internet for that, or could those–

DOUG STONE: That would be very hard, I agree. You’d have to redo a lot. But if you had specific channels that you wanted to make hyper-secure, we already have technology to do that. It’s not released who’s doing it, but some banks have expressed interest. The government certainly has expressed interest. So there may already be a quantum-secure channel from the Pentagon to the White House we just don’t know about.

IRA FLATOW: So is that where the money is coming from in this field? I mean, who’s spending money on this besides Yale?

DOUG STONE: Well, Yale is not spending nearly as much as we would like. No, they’re spending enough. But yeah, so certainly DOD. I mean, if you tell a general that you can break the other people’s codes, they get excited, right? So the DOD is very excited.

But now I think people are realizing that the big compute problems can be incredibly important for the information technology sector, and also just for any kind of difficult problem in materials. In the end, all materials problems are doing a computation of a quantum problem with a classical computer. And it’s been shown that for many problems, a quantum computer can be much faster for that. So now, just to do chemistry or materials science or medical imaging or something, this might eventually be useful.

A great example that I learned about is this so-called fixing of nitrogen. So ammonia is an incredibly valuable chemical, I read. It’s a precursor in 45% of the world’s food, between fertilizer and other ways in which ammonia comes into food. And then, of course, there’s also explosives and other things you can make with ammonia.

So we have a process that’s over 100 years old for doing it, called the Haber-Bosch process, which takes a lot of energy. It takes like 2% of the world’s energy to use to make all this ammonia that we need. And it emits 3% of our carbon, all our carbon. Amazing.

IRA FLATOW: Wow.

DOUG STONE: Well, it turns out that there’s a nitrogenase, which is a biomolecule enzyme that can just interact with air and produce ammonia for plants. It’s called FeMoco. I don’t know that much more about it.

But we have no idea how it works. It’s a big molecule. It’s a complicated thing. It’s beyond our current scientific understanding.

That’s the kind of problem that you could imagine a quantum computer could simulate. We could go in and pull it apart with a quantum computer, understand. And the point is that this biomolecule is doing it at room temperature, whereas all this energy comes from having to heat up the ammonia in the Haber-Bosch process to 450 degrees centigrade, put it at high pressure, et cetera. That’s why it takes so much energy. If we could just do it at room temperature and pressure, we would save these huge quantities of energy, carbon emission, et cetera.

IRA FLATOW: This is Science Friday from WNYC Studios. You know, you’ve said we don’t have a functioning quantum computer yet. What’s limiting this from being a transformative thing? Is it the technology, the understanding, the cost? What will limit the impact here? And when do you think we’ll have a functioning quantum computer?

DOUG STONE: So it’s certainly gotten much more realistic in the last five years that we’ll have something. I mean, right now, six or seven companies, including IBM and Microsoft, they’ll let you play on their quantum cloud even though they’re not doing anything really important with it yet because the computers aren’t good enough. But it’s really gotten much closer to being realistic.

So what’s making it quite difficult is what I said before, that making such a fragile system robust and being able to make it do this very specific computational task that you want– it’s never been attempted before in the history of man. I mean, there’s nothing comparable in any of our physics that’s as delicate as making, say, a 1,000-qubit quantum computer. So everything is being pioneered. We don’t even know which technology is best. And people are making progress. But I think it’s somewhere in between physics and engineering that we’re stuck.

IRA FLATOW: Do you think this is a technology that we’ll all have equal access to, or will the world be divided into quantum haves and quantum have-nots?

DOUG STONE: Well, I do believe that all of these companies will provide quantum cloud services if everything pans out. So if you have money, even if you’re not in America, you’ll have access. Now, initially, I wouldn’t be surprised if defense departments were trying to keep some things classified and so on. And we know that 2/3 of the money being spent on quantum computing is actually being sent by China. So China is really interested in this, and I don’t know what they’re going to do if they get it.

But anyway, I think it’s going to be expensive. I don’t know if it will ever come down so that you could use that just with your cell phone, going to ask something from the quantum cloud. I do not know if that will ever happen.

IRA FLATOW: Doug, we have run out of time. So much to talk about, so little time. Thank you for being with us today.

DOUG STONE: Well, thank you.

IRA FLATOW: Very interesting stuff. Dr. Doug Stone is deputy director of the Yale Quantum Institute and Carl A. Morse professor of applied physics at Yale University. You might want to check out his YouTube lecture on why we should care about quantum computing. It’s got a lot of good extra stuff in there.

Copyright © 2021 Science Friday Initiative. All rights reserved. Science Friday transcripts are produced on a tight deadline by 3Play Media. Fidelity to the original aired/published audio or video file might vary, and text might be updated or amended in the future. For the authoritative record of Science Friday’s programming, please visit the original aired/published recording. For terms of use and more information, visit our policies pages at http://www.sciencefriday.com/about/policies/