Advances in physics are making possible radically new kinds of computers. An interview with Cecilia López, from the Massachusetts Institute of Technology.
What is quantum computing?
Generally, “quantum information processors” are devices that operate in a way in which the particular effects of quantum physics become preponderant. Quantum computers are specifically those devices meant to “compute” things. Other types of quantum devices are for example those related to communication and cryptography, or quantum devices focused in the measurement of properties (for example, the strength of a magnetic field) with a precision that can’t be achieved by other means (quantum metrology).
In these devices, information is stored in quantum bits (qubits). In a classic processor, information is codified in electrical impulses: lack of current is 0, and the presence of current is 1. In quantum devices, the codification is done in atoms, nuclei, photons, or similar systems that can be in at least two states (which we call 0 and 1) but which by their nature can be in a superposition of 0 and 1: with a certain probability in 0, and with a certain probability in 1.
This opens tremendous potentialities, because those operations that we would classically run over the 0 bit and the 1 bit, in quantum processing can be run over 0 and 1 at the same time (the so called quantum parallelism), and thus with many qubits at the same time. This also generates a very particular phenomenon called entanglement, which happens when we have a device with two or more qubits: observing the state of one of them induces modifications in the state of the other, that can for example be far away. This property is key for some operations in quantum communication like teleportation and transmission with superdense codes.
There are of course other relevant properties of quantum information processing, but those two are probably the most representative.
Will quantum computers be competitors (or inheritors) of current computing technologies?
Quantum processors aren’t equivalent to current classical computers. They don’t attempt to compete, but rather to solve problems which classic computers cannot and will never be able to solve efficiently.
The most famous example is Shor’s factorization algorithm. The best known classical algorithm requires a number of operations that grows too fast (exponentially) with the size of the number to factorize. Shor’s algorithm, using quantum computing, makes it possible with a number of operations that grows more slowly (polynomially). To illustrate the issue of efficiency, consider the following: if with the best computers factorizing a number of two or three bits takes a few seconds with either method, to factorize a number of a hundred bits would take classically at least five months, while Shor’s algorithm would require less than a day.
Factorizing numbers is not a problem that interests the average computer user: nobody will buy a quantum computer with this purpose. However, the fact that factorial decomposition is so hard — numbers of thousands of bits are used — allows the encryption of information (for example a credit card number during an online purchase) knowing that it would take anyone a lifetime to break the code.
There are many computational problems that classically lack an efficient solution: they take too much time or too many resources as the “size of the problem” grows.
An example of this that is very important for scientists is the simulation of quantum systems composed of many particles (which is related to the number of qubits): nothing better than a quantum computer to solve a quantum physics problem!
It’s there where a quantum computer can make a difference. More than a competition, I’d say that quantum processors seek to fill a vacuum that nobody else can fill. It’s not possible for a classical computer to solve these problems efficiently; it’s a basic issue, not a problem of optimization or technological competence.
How advanced would you say is quantum computing research in terms of possible commercial applications?
Quantum computers as such are quite far from being usable commercially. Today you can find computers with 4-5 qubits that operate with a low margin of error during a relatively short but arbitrary series of operations. There are also many devices that operate with up to 12-13 qubits, but which can only perform very specific tasks, and aren’t yet universal computers. In general it’s relatively simple to control a few qubits, but attempting to expand the capacity of the quantum processor runs into many problems, which today are being studied over the possible physical systems that would constitute these quantum computers: ion traps, cold atoms, photons, superconducting circuits, carbon nanotubes, etc.
However, thanks to the search of a quantum computing, many technologies have been developed that are much closer to commercialization. A few companies are already working in the sale of products related to quantum communication. The first bank transference using quantum cryptography (with entangled photons) was realized in Austria in 2004. Without going any farther, in Buenos Aires the Laboratorio de Óptica Cuántica is being developed under the Universidad de Buenos Aires (UBA) and the Instituto de Investigaciones Científicas y Técnicas para la Defensa (CITEDEF). They have already built a quantum random numbers generator, and are preparing to implement a quantum key distribution system. This is the basis of commercial quantum cryptography devices already sold in other places.
Quantum metrology is also an area that has been developed studying the phenomenon of quantum interference, which is related to the superposition of states that I mentioned before. For example, commercial magnetometers are already manufactured based on superconducting elements with quantum interference, which at the same time is a technology used in the development of quantum computers.
Atomic clocks (which can attain an unprecedented precision) are also on the way to reach commercialization. The same path can be seen for transistors based on quantum effects, which can control the flow of electric current down to the level of a single electron.
The important conclusion is that scientific development isn’t always lineally directed to technological and commercial development, but however ends generating it. The field of quantum information is where the quantum world, made of many small quantum manipulable units of information, is discovered and sought to control. And this has already proven to be an incredibly fertile field for the development of technologies and devices more powerful or efficient than the current ones.