It seems like something out of science fiction: By exploiting the bizarre world of quantum mechanics, or the physics of the very small (think atoms, electrons, and photons), so-called quantum computers will soon be able to do things that today’s computers cannot. Associate professor of physics Steven Olmschenk, who is creating the building blocks of these remarkable machines, explains how quantum computers work, what makes them special—and how soon we might get our hands on them.
What’s the main difference between quantum computers and conventional, or classical, computers?
In classical computing, information is encoded in binary form as bits or zeros, and ones. You take whatever information you want—words or pictures or whatever—and encode it as zeros and ones.
Whereas if you store some information in an atom or some other quantum mechanical thing, it can be not just a zero or a one, but a superposition of them—which means that you can have both at the same time.
So the main distinction is that quantum bits, or qubits, can be both zero and one at the same time. And when you have many qubits, you can encode lots of different possible configurations simultaneously.
What does that allow quantum computers to do that classical computers can’t?
From a physics standpoint, one of the most interesting things about a quantum computer is that you could use it to simulate other less well-understood quantum systems like molecules and superconductors. It’s really hard to model quantum systems like those on a classical computer because you run out of computational power really fast. If we could mimic such systems with quantum computers, we might learn how to build better materials and medicines.
But one of the most well-publicized examples is decryption, or codebreaking.
When you enter your credit card information online, it gets encrypted so other people can’t steal it. And it’s encoded by an algorithm that relies on the fact that it’s hard to factor large numbers.
A small number like 15 can easily be factored into its prime factors: three and five. But that gets really hard to do when you get numbers that are hundreds of digits long. And the fact that that’s hard to do is the basis for how most of our information is encrypted today.
With a classical computer, factoring such large numbers involves a lot of trial and error. You plug in some numbers and say, “Well, are these the right ones?” And you keep trying different numbers over and over again. And when the number you are factoring is sufficiently large, that trial and error takes a really long time. But with a quantum computer you can perform those operations on all possible inputs at the same time. So quantum computers will be able to solve certain problems much, much faster.
How do you create these magical qubits?
In my lab, I’m using laser-cooled trapped atomic ions.
An ion is just an atom that has lost or gained one or more of its electrons. We kick one or more of the electrons off an atom, so it has a net electric charge. This allows us to use electric fields to trap the resulting ion, which oscillates back and forth in the trap like a ball rolling back and forth at the bottom of a bowl.
The ion will only absorb light at a particular wavelength and frequency. We tune a laser so that the ion can absorb a little bit of its light—but only as it is moving toward the laser. And every time it absorbs a bit of the light, it gets a kick in the opposite direction. So it continually slows down, which is to say it gets colder. Using this kind of laser cooling, we can cool an ion down to a fraction of a degree above absolute zero, which is less than negative 459 degrees Fahrenheit.
Why so cold? The main reason is that if the ions are not cooled, they’ll move all over the place. And we can’t write information into them, or read information out of them, if we don’t know exactly where they are.
How do you store information in an ion?
Let’s say there are two energy levels available in the ion. Call one of them “zero” and the other “one.”
If you can control the energy level of the ion using lasers or microwave radiation, then you can write information into the atom—including information in a superposition of states.
Each individual ion becomes a qubit. And when you have more than one ion, you can have multiple qubits, and then you can make them become entangled and do all sorts of interesting things.
Are there working examples of quantum computers right now?
People have started running small algorithms on 10 or 20 qubits. But the question is, how far do you have to get before you’re doing something that can’t be done on a classical computer? The hip term for that is “quantum supremacy.”
If you’re at 10 or 20 qubits, you’re still working at the level of a classical computer. And when you think about your laptop, it has billions of bits in it. So you might ask, do we have to get billions of qubits?
Fortunately, the answer is no. But with a hundred qubits, you’re going to be able to do things that you cannot do with classical computers. So we’re not so far away; we don’t have to go from 20 qubits to a billion qubits.
So when might the Age of Quantum Supremacy be upon us?
I think that within the next couple of years, we will have quantum computers that can solve specific problems that are beyond the reach of classical computers, like modeling particular molecules.
For a very general system, where you can program in anything you want, I think we’re more like a decade or more away. But given all the potential of quantum computers, it’ll be worth the wait.