Leo Kouwenhoven is a professor in Applied Physics specialized in the field of Quantum NanoScience. His team at the QU Tech Lab designs experiments to place electrons in superpositions.
Here is the full text of Leo’s talk titled “Can we make quantum technology work?” at TEDxAmsterdam.
Leo Kouwenhoven – TEDx Talk Transcript
Good morning, everybody.
Does anybody recognize the picture behind me? What is it? That computer — it was in fact, a mechanical computer, the computer that was used by Alan Turing, which helped to end the Second World War.
What is interesting is that here the wheels are turned, and by registering clicks and no clicks, they could break encrypted messages. So it’s a mechanical motion that actually can do some calculations for us.
Now we use of course electronic computers, where electrical signals encode for our bits zeros and ones. But clicks or no clicks, or zeros and ones, actually is the same principle for encoding information, no difference between the Turing machine and our computers.
What is amazing is that — and what happened in the last 50-60 years is that this machine now fits around your wrist, and it sells as a SmartWatch – the same computing power. Anybody else who has a SmartWatch here? I see a few. Aha! So you fell for the same commercial as I did, I guess.
But it’s smaller, and you see that the revolution in making things smaller is very visible, but what is maybe even more amazing is that the underlying principles of the Turing machine of clicks and no clicks, and in my SmartWatch, it’s still the same; that has not changed – clicks or no clicks, zeros or ones.
Some time ago, a very interesting new idea started to become popular. And there’s not really a single inventor, and in fact, it is not really an invention, it’s rather a change of perspective and the idea is that nature also calculates.
For instance, when light hits a green leaf, it induces all kinds of chemistry, which at the end of the chemistry reaction, it produces oxygen among other things. And the process in between the input light and the output oxygen can be viewed as a calculation.
Why this is an interesting change of perspective is because nature actually calculates a lot faster and a lot smarter than our computers do. The key ingredient that nature uses is, in fact, quantum mechanics.
And the beautiful thing of quantum mechanics is that you don’t have to be a zero or a one, you can be zero and one at the same time, and this option of being zero and one at the same time is used in nature. Zeroes and ones at the same time that they sound acceptable, but clicks and no clicks at the same time that is absurd.
Nevertheless, people like me started to use these principles of quantum mechanics to build a new very powerful computer, it’s called a quantum computer.
How does the quantum computer work?
Why is it actually good? That’s going to be my story today.
Let me first tell you a little bit about myself. I grew up in a small town in the Netherlands on a farm. I was actually doing okay in high school, and I was allowed to go to university. But the only people in my town that I knew who had a university degree was our town veterinarian and our priest.
Now, becoming a priest was not an option for me, so I was going to become a vet. However and unfortunately, my university had no entrance exam, but instead a lottery, which I lost, so I ended up at my second choice which was physics. But my roots remain.
As a farmer son, I keep this pragmatic approach. And nowadays as a professor, if I hear some of my colleagues make very profound theoretical predictions, I think, “Wow, that sounds profound.” Can we actually do something useful with it? And that is also my attitude towards quantum mechanics.
And quantum mechanics is certainly among the most profound scientific ideas that we have around. People like Bohr, and Einstein, discovered the deeper principles of quantum mechanics about 100 years ago. And those principles of quantum mechanics are absurd for us, humans, but not for small particles like electrons.
So what a small particle like an electron can do is for instance not be confined to one single point in space, it can actually occupy different points in space at the same time. How that is possible is actually very impossible to explain in words in our language.
The best thing we can do is just accept that what is absurd for us, is OK for small particles like electrons. So think about this for a moment, because it’s actually very essential for my story that a single particle, the single object can actually be at different locations at the same time. We call this superposition.
When you accept superposition, then you can actually also understand things like chemistry. For instance, a very simple example – the oxygen molecule, and we draw it like two oxygen atoms that are held together by these horizontal lines in the oxygen molecule.
What do these horizontal lines actually stand for?
Well, they share an electron, but this electron does not sit still in the middle between these two atoms. No, it actually divides itself up, goes into a superposition, and occupies the space around each of these two oxygen atoms.
Since these two parts of the same single-electron doesn’t want to be too separated, it actually keeps the oxygen atoms together, so it’s actually a superposition that binds atoms in molecules. Since actually our body consists of molecules, so without superposition, our body would fall apart; and without superposition, all our molecules would fall apart in loose atoms.
So superposition is a good thing. You should like it on Facebook.
Since Einstein, and Bohr, and also other genius scientists developed the principles of quantum mechanics, we’ve been using mostly formulas — formulas to describe nature, as it is given to us.
But now 100 years later, it’s actually time for something new. We now view nature as an information processor, and instead of formulas, we use this symbol to describe that there is an information flow in nature.
And we no longer study nature as it is given to us. We actually have started to design and construct actual subprogram — the machines that we make ourselves; and study how our own designed machines can actually solve quantum problems.
So my job has become to make qubits. Instead of a farmer, instead of a vet, I have become a qubit maker, or a superposition maker. And I want to illustrate that with electrons in boxes.
What you see here in the upper row are two boxes and one electron. In our world, that electron has to choose it can sit on the left or in the right box. In an information description of the same thing, what we say is that in the upper case, when the electron is in the left box, I call it a bit zero; or if it’s in the right box, I call it a bit one. This is actually how we encode information in our normal computers – that’s a bit zero and a bit one.
The special thing that we do in our lab in Delft, is that we can do also superposition. So we can take a single electron and put it in both boxes at the same time, kind of similar the oxygen at a molecule. But now we’ve boxes that we have made ourselves, and if we can also control and program.
When the electron is in both boxes at the same time, in an information description, we say the system is in a qubit state, and the qubit is in a superposition of a bit zero and a bit one at the same time. So it encodes for that information at the same time.
If we have these qubits, we made actually a little animation to illustrate how it can be used to speed up calculations. You see here a labyrinth, and if we put a classical electron into this labyrinth, then the way electrons actually solve classically this problem to find the exit of the labyrinth is what we would do. We try path by path. Every time, we find it’s not the solution, we try again.
Sequentially, we go through the system until we found the exit, but when we find the exit we know we have the right solution. The quantum electron would split itself up, in parallel, in a superposition, take all the paths at the same time; and also reaching the exit, but now a lot faster. And that is the magic of a quantum computer.
All these actions, all these different possibilities, can be checked in a message parallel calculation, and find the answer in a single step, that speeds up the calculation.
What would we do if we have a quantum computer?
What kind of — do we actually have good problems to feed, such a super powerful computer? To answer this question, let’s zoom out a little bit. And ask ourselves, what are actually the big challenges that we face on Earth? The big problems, there’re many big problems. But let’s focus on our natural resources.
Here we are: spoiling energy; we’re wasting materials. Our climate is changing too fast, and many people on Earth don’t get the right medicine. These are very big problems that somehow we have to solve. We have to solve it rather soon at least within the next few decades.
To solve those problems, we need radically new tools. And no one doubts that help from a supercomputer will be of vital importance to actually solve these problems. That’s where the use of the quantum computer can come in.
And universities — I’m working at a university — have started to develop the fundamentals of quantum computers since two decades or so. And in the last few years, also, some of the bigger global IT companies have joined this effort.
When companies join and invest money, they actually have some specific ideas for the purpose, they want to use the quantum computer for, and have made a little list of what they say will be the applications of a quantum computer.
So it’s a list starting with: Electrical cables with zero loss of energy; drug development by solving quantum chemistry problems; predicting material properties for electronics and energy storage; machine learning; optimization problems for robotics; handling big data for sequencing genomes; and airplane design.
But this list is of course, by far not complete. These are just a few examples. It is impossible to predict what you can do with a new technology.
So we started a new institute in Delft, actually, work in a focused way on developing this quantum computer, it’s called Q-Tech.
In this Institute, we make the hardware. By using nanotechnology, and cleanroom fabrication, we make electronic chips with a whole bunch of qubits, that we can program. By programming these chips, we can learn how quantum systems solve problems.
We do that together with electrical engineers, and we make these chips that you see here in the last shot. This is an electronic chip that has a whole bunch of qubits on it.
Nowadays, we can make between five and ten cubits on the chip, and program and control it. We think that we need another ten years or so to make circuits that are big enough to really solve the relevant problems, and have an illustration how it will go, how it will develop from there.
Suppose that we take for instance, the biggest supercomputer that we have nowadays in the world, it’s in U.S. It’s called the Titan. This computer is actually as big as this concert hall. If I want to make this computer two times faster, what I have to do is I have to make it two times bigger, two concert hall.
If this were a quantum computer, it’s the same starting point. Then in order to make it two times faster, I only have to add one cubit to the supercomputer, and one cubit is very tiny, you can’t even see by the naked eye.
So every time, you’re adding a qubit to the quantum computer, it becomes two times faster. That is because the computational power of a quantum computer scales as an exponential. It’s two to the power of the number of qubits.
Exponentials are difficult to understand, but let me try to illustrate it with a number example. I have to go offstage for this example, actually. If I — I’m a linear machine, if I start to give handshakes to everybody in the audience, then you know it’s a linear process and I need ten steps in fact, to give ten handshakes, and I need two, four more. I do this linear, I’m a classical machine.
Suppose I could go into a superposition, then in ten steps, I actually can shake two to the power ten, that is 1024 hands. So in the same sequence, I would have given everybody here in the concert hall a handshake. That is the difference between a classical computer – a small section, and a quantum computer – everybody in the concert hall.
So we’re not talking here about some incremental improvement of computing speed. This is really a revolution in information technology. It’s a game-changer.
Funny enough and lucky enough, it’s nature that actually helps us to develop this quantum computer, and that brings me back to my leaf.
In the photosynthesis reaction, that happens in between the light input where it actually releases electrons in the leaf, that have to find a way from their starting point to some specific end point, where they can release the oxygen.
And in between, there are many different paths for the electron to take to go to the site where it can release the oxygen. Luckily, the electron can go into a superposition, and as in the labyrinth problem can find the end state very quickly, and produce you know, the oxygen for us very efficiently, so our bodies are kept together. We breathe oxygen, all because of superposition, and that brings me to my take-home message: Nature uses quantum mechanics to compute.
And we quantum engineers, have started to begin to make programmable quantum computers, and help solve some of our Earth problems. So I hope that with a quantum computer that we will live better lives. and be more careful with our resources on Earth.
Thank you very much.