Navigation and service

How does a quantum computer work?

Mehrere Aufgaben können gleichzeitig gerechnet werden

Quantum computers do not work like conventional computers. The latter use bits, in other words, zeros and ones. In order for a component to serve as a bit – that is, as the smallest possible information unit of a computer – it must have one property above all: it must be able to assume two clearly distinguishable states, which can be interpreted as zero and one. Conventional computer chips calculate with billions of microscopically small semiconductor transistors that function like tiny flip switches for the electrical current. A control voltage can switch them between “on” and “off”, between zero and one.

The basic units of a quantum computer, on the other hand, are quantum bits, so-called qubits. They cannot only assume the value one or zero – but both at the same time. In fact, the qubit – as long as it is not called up – is in a state of superposition of different states, which is described in quantum mechanics most vividly using the image of “Schrödinger’s cat”. It is both zero and one and also every possible value in between; a world that eludes our everyday experience and partly contradicts it.

This is due to the information carriers that are used, such as atoms, electrons or photons, which are not yet subject to the classical laws of the macroscopic world. For them, the rules of quantum mechanics apply, which – unlike in classical microelectronics – can be used to create and use bizarre-looking states. One of these rules is the possible overlaying of states. Physicists speak of superposition. It allows quantum objects to assume an ambiguous state or several states simultaneously. A second phenomenon is the so-called quantum entanglement: In an arrangement of qubits, the states of the individual particles can be linked (physically: entangled). They are then connected to each other as if by magic. Because of these two phenomena, quantum computers are capable of performing many operations simultaneously with each switching operation.

In order to perform a calculation with many qubits, they must therefore be in entangled superpositions of states – referred to as a quantum coherent state. It is then that the change of one qubit affects all others at the same time. This is not easy, however. Qubits and their quantum states are very sensitive to external influences such as heat or radiation. These interference factors dissolve the entangled states again after a few microseconds. The more qubits are connected, the more fragile their common quantum state becomes. Therefore, the lifetime of a piece of quantum information also depends on how well the researchers manage to shield the computer from the environment – or to correct the arising errors.

Twenty years ago, David DiVincenzo, pioneer in quantum information and director at the Jülich Peter Grünberg Institute, summarised five criteria that a universal quantum computer must meet:

1. The system consists of a scalable system of well characterised, that is, understood qubits

2. It must be possible to bring the qubits to a defined initial state

3. A universal set of elementary quantum logic gates, that is, arithmetic operations, can be performed

4. Individual qubits (at least one) can be measured

5. The coherence time of the system is much longer than the operation time of a gate, or rather, of an arithmetic operation

It is still very uncertain which physical systems are ultimately best suited to realise a universal quantum computer. Even though the most advanced technologies currently rely on superconducting qubits, there are reasons why other approaches based on semiconductor qubits or exotic materials such as topological insulators may prove superior in the long run.