In 2016, the European Union announced an ambitious research agenda for advancing developments in the newly emerged field of Quantum Technologies.
Titled The European Quantum Technologies Flagship, the initiative focuses on four main areas, or pillars:
- Quantum computing: the ability to compute otherwise insoluble problems, as well as processing vast amounts of data, faster than classical computing, recognising patterns and training artificial intelligence systems, to solve world problems: e.g. to help diagnose and treat diseases or optimise energy use in smart cities.
- Quantum simulation: the ability to build functioning complex systems to design new drugs, fertilisers and materials for real world applications before widespread roll out.
- Quantum communication: the ability to create 100% secure, unhackable systems to converse as well as to protect sensitive financial and medical data sets. Impossible to intercept without detection.
- Quantum metrology and sensing: the ability to create highly accurate sensors with applications in the consumer and medical sectors. Ideal for high-precision navigation and the Internet of Things.
The €1bn QT Flagship initiative, will span 10 years (started 2018) will support over 20 cutting edge projects and will bring together the collective knowledge of 5,000+ scientists across 17 EU member states. Together, the UK and their European neighbours will make a whole spectrum of quantum technology better, faster, more secure, scalable and saleable.
What is quantum supremacy and why is it important?
‘Quantum supremacy means that you can run a problem quicker on a quantum computer than on a classical computer. In classical computing, if you want to make a code safe, you make it longer; when it becomes longer, it gets more and more difficult to crack.
‘However, quantum computing is a game changer. It is now exponentially faster than a classical computer at breaking code. So, if one were to have quantum supremacy in code breaking, they could potentially read all our messages on the internet, see bank transfers - in fact everything that has thus far been secured using computational complexity, which is vulnerable to attacks from quantum computers.
‘In short, no classical system is secure against a quantum computer attack. ‘This is why the quantum race has started; it has huge military implications as well - knowing what is going on in other states by being able to read all their secure information. As a scientist, you can always find usages where you think, “okay I don't want this getting into the wrong hands…”’
This is the second quantum revolution. What was the first quantum revolution?
‘It started as a theoretical revolution. If you go back to [Max] Planck [1858-1947], he introduced the quanta in order to explain black body radiation. If you look at radiation coming from the sun, the distribution of the colours follows a certain curve - at the end of the 19th century, nobody could explain why. Planck introduced us to small quanta, that mathematically solved the problem, but it was just a theoretical tool.
‘Then came [Werner] Heisenberg [1901-1976], [Erwin] Schrödinger [1887-1961] and so on; they really took quantum mechanics further and developed theory of nature out of it, which could explain all of our measurements and results when dealing with small objects.
‘When you look at atoms and how electrons can actually orbit atoms, this could not be explained in a classical way. Using quantum mechanics, you had a theory which could explain those things. That was in the 1920s and 30s.’
So how did theory move into practice?
‘People said, "Okay, can we make use of this?" The answer is yes, because the way we are communicating now [England to Austria], over the internet, in front of a computer which has a microprocessor in it and through the internet connecting us using optical fibres.
'The microprocessors are made out of transistors - so that's a semiconductor material - and fibre optics use lasers and light particles. Those inventions, the laser, the transistor, the microchip, are all heavily dependent on understanding nature at a quantum level. It started off with a theoretical idea but the whole world is determined by quantum effects.’
What is different from the first quantum revolution to the second?
‘The first quantum revolution was understanding nature, explaining how quantum systems and particles like atoms and protons behave in groups of millions and billions - and experimenting to make the first applications out of it.
'Now in the second quantum revolution, for the first time in history, it's really possible to manipulate not many atoms or many electrons in a solid - but single ones. So, we can describe the state and evolution of single electrons and single photons - and we can actually manipulate it. So, we can create, manipulate and detect those single quantum states.
‘The power now for this second quantum revolution; quantum technology and quantum computers, all comes from the fact that we can really play and manipulate those single quanta.’
Why was isolating “one” so important?
‘So, the single is important because one of the main concepts in quantum mechanics is super-position. That means if you have two possible quantum states, the system can be in either of the two but it can also be in a “mixture” of the two. This is not possible classically. Classically, it would only be in one state or the other but not in both together. And if you have many atoms or quanta, they can be between states, but as a mass, will average out.
‘However, If you have a single electron, which is in those two possible states and also in between, you can really harness this power of the super-position. ‘That's mainly done in quantum computing using the quantum bit, or qubit.
Qubits can be in 0 and 1 at the same time. If you feed in your calculation and your computer calculates with zeroes and ones (so, with all possible combinations) you only have to run through the computing program once. If you have a classical computer, you will have to do it first with the zero until it comes to the result and then with the one. So, [quantum computing] speeds things up.
‘The challenge for quantum computing now is to make a large device with hundreds or thousands of these single quanta connected together individually.
‘Quantum states can also be used as precise sensors (quantum sensing). Quantum states are very easily disturbed by the outside environment and therefore they make very sensitive sensors as they can measure how much an electron has moved from the zero state, for example, to the one state. That enables this large increase in sensitivity for making measurements.’
How will you make this a commercial proposition?
‘I'm focusing mainly on quantum communication with quantum key distribution (QKD), working with particles of light and the quantum states of photons rather than electrons. Photons travel very easily in air or over optical fibres; QKD allows information to be sent absolutely secretly as information is inscribed on single photons. Because they are single photons, they behave differently than a normal communication link where you have a strong laser pulse containing many photons. Single photons cannot be copied as explained in the quantum non-cloning theorem.
‘The Heisenberg uncertainty principle, states that if you make measurements on a quantum system, you will disturb it. That means if somebody tries to gain information on the line, or they eavesdrop, it will change this very fragile single photon and that can be detected.
‘I think in our world, this secure communication element is extremely sought after. If you’re dealing with online shopping, you need protection for your payment. If you’re banking, it’s important to make sure no one can see your financial dealings. Higher up, we also need to think about how this can protect top level secrets and government information.
‘We know it works in the lab. There are one or two companies who are starting to sell products but they're still very bulky, very expensive - we're talking about hundreds of thousands of euros and the UNIQORN project (supported by the QT flagship) is about getting the cost down and making this mainstream.’
So, it’s 2020. The costs are still largely prohibitive. What’s the first step to the mass-market?
‘So, in the first instance, there will be quantum computer centres. This is because quantum computers, in the beginning, will be very costly and very big - so there will probably be just a few in each country at central points.
‘Using quantum communication, it will be possible to link to the computers, run any programme you want to run and get the results back without having the need for anybody third parties; you will have the results - not the quantum computer. This is called blind computing.’
So, will there only be five computers in the world? Or will the few turn into the many, like our own digital revolution?
‘Once we have made devices more affordable, then it will be up to the user base. When the first iPhone was launched, the CEO of Microsoft, Steve Ballmer, said something along the lines of, "It's complete nonsense. Nobody will buy it."
They don't know yet what to use it for but once everybody has it and suddenly has Facebook and social media then people are empowered by the tech and something creative is happening. It is kind of a feedback loop. Once you have it, you want more of it. So, we are hoping to show what's possible, bring it to a wider base, then let the creativity happen.’
Will quantum be an add-on? Or is it something completely different?
‘In the case of quantum communications using photonics, we can use quite a lot of the standard equipment already used in phones, communication networks and in routers. The idea for us would be add-on. We want to keep the current communication infrastructure as much as possible but just enhance it with quantum technology - maybe with a small transmitter that can switch between classical communication and quantum communication.
‘Mixing the technologies, putting a very weak quantum signal next to a very large pulse which has billions of photons, risks losing the “quantumness” again. We are working on ‘shielding’ the quantum channel from the other ones - so you can still use the same optical fibres and run your normal internet traffic over it - but on the side you have your quantum channel. I think it would be very costly to develop a completely discrete second network.’
So it’s evolution rather than revolution?
‘The protocol is, in principle, absolutely secure. But to truly trust your device, you have to trust that the manufacturer built the device to specifications. This is something completely normal to classical IT and manufacturers - components are made to certain standards; they are certified and they are verified. However, quantum standards are a completely new field to our physicists. This is why we're seeing this monumental shift from basic science to engineering.
‘Moving to a quantum communication network will probably start with an already trusted and government contracted big military supplier who will begin to build those systems.’
In your pan-European project, you work with 17 partner institutions. Will the UK still participate, post Brexit?
‘We are in the transition period until the end of the year. Then after that, I really hope that the UK is like Switzerland, subscribing to the research funding, so we can at least collaborate and exchange ideas. In research, the UK (specifically the University of Bristol) has been a good partner.
‘UNIQORN isn’t just about testing in a lab, it’s about the process of fabrication to integration, packaging and then demonstrating the application - we have to have the value chain and to have that you need the correct partners at each stage. To find that in a single country is impossible, so you find the resources you need from partners across Europe. The UK got in early (Quantum Hubs) and we are all benefitting from the UK’s expertise and findings.’
When correctly programmed, could AI and quantum computing solve the world’s problems?
'AI looks for patterns in huge amounts of data. So, it's really good at replicating what a human would do: we have this huge network of neurons and we try to find connections. I'm not an expert on quantum computers, nor on [biology] but I would say that what a quantum computer would be good at is drug development - on simulating how protein folding takes place.
‘A protein, although it has many tens of thousands of atoms ,should still be treated as a quantum object. At the moment, we can run a simulation with classical computers which is just an approximation - and if it becomes too big, this approximation fails and the results are not valid anymore. So, if you can really simulate those larger molecules with a quantum computer and the simulation is true to life, we can get a better understanding of what is happening on the biological level and can think about new drugs, maybe new therapies and similar. '
The quantum skills pipeline
For something this new, how do you find or train the workforce?
‘We need a lot of engineers on the optic side, on cryogenics, on software and electronic engineering to cut across all of those disciplines. Right from the start, we need to teach our children about quantum as part of the curriculum.
‘Software engineers should also learn how to program a quantum computer. Train them on the differences, how a classical bit compares to a qubit. For electrical engineers, if they make systems for classical communication, they need to know how the system is adapted for quantum transmission. It’s too late to start training people after they’ve finished their studies, people have to really grow up with it.
‘Back in the day, if you wanted to start a career in computing, you’d start scripting or programming, then you’d start trying things out by soldering some circuits to make a little robot or whatever. These are things we had and they have started a digital revolution. We don’t have this (the ability to practise) on the quantum level, yet.
‘At the moment, it’s still too costly and too cumbersome to really roll it out on a large scale, so we have to wait a few more years until the cost comes down and children will grow up with it.’
Do you think there will be quantum Raspberry Pi's in the future?
‘We won't have them next year but I think in five years’ time that's a possibility. For example, you can already make some small quantum computing programs via IBM on remote access and then run it on their big machines.
‘The flagship programme is working towards making small scale quantum computers for smaller companies. Then when costs come down, it will become more mass-producible, perhaps like the home computing revolution of the 1980s.
‘I think to really see what is possible and to fully understand it, many more people have to get involved. At the moment it's just a small group of physicists, maybe a couple hundred groups in the world altogether. It's still not a lot.’