If a practical quantum computer is going to become an everyday thing, qubits have to remain in two states at one time for much longer than they do now. One of the possible candidates for a longer lasting qubit is a copper ion embedded in a large molecule.
A quick primer on qubits is in order. Certain ions have unpaired electrons whose spins can assume either of two spin states, up or down—or in computer speak, 0 or 1. But when struck with a microwave pulse, the unpaired electron can be coerced into assuming both the 0 and 1 state simultaneously. The two states are said to be in superposition. All the qubits created up to now stay in a superposition state for very short periods because the spin states of neighboring atoms quickly destroy the coherent state, making the life of the qubit too short for it to perform the desired number of quantum computations. But researchers have been looking high and low for suitable qubits and for ways to lengthen the periods over which they remain in superposition.
Now, a group at the University of Stuttgart reports, in the 20 October issue of Nature Communications, that it has developed a way to protect the spin of a copper ion by placing it inside a molecule that has relatively few spin-carrying atoms and keeping it far away from hydrogen atoms that carry spins. The copper ion is moved into a neighborhood where it’s surrounded by sulfur and carbon atoms that have no spin, and by nitrogen atoms that have a small magnetic moment. The team reports a coherence time of 68 microseconds at a temperature of 7 K—an order of magnitude better than what can be achieved with similar qubits now. The molecule can still function as a qubit at room temperature, although the coherence time is 1 microsecond. Qubits in nitrogen vacancies in diamonds have scored much longer coherence times, but “you cannot make a quantum computer with a single qubit,” says Joris van Slageren, the chemical physicist at the University of Stuttgart who led the research.
“The coherence time should always be compared to the time that it takes to do one quantum operation, and in our case this is perhaps 20 to 40 nanoseconds,” says van Slageren. “In principle, within the coherence time, we can carry out about 3000 quantum operations, which equals the ‘qubit figure of merit’ of 3000. The usual benchmark is a qubit figure of merit of 10,000; we are a factor of three away from that, so I think this is a major step forward compared to all other quantum bits," says van Slageren.
There is still room for improvement in the coherence time. It will happen with further engineering of the molecule that contains the copper ion. “There are still nuclear spins that can be removed,” says van Slageren.
As quantum computers will have several hundred to several thousand quantum bits that can communicate, research into how to communicate with individual qubits is now important. Up to now, the researchers investigated the average response of probe microwave pulses that incited quantum bits embedded in a solid mass of compounds meant to protect them.
“There are now ways to talk to each quantum bit individually by entanglement with an quantum sensor,” says van Slageren. To achieve this, the team plans to place individual qubit molecules on substrates. “This can be done either by evaporating these molecules in a high vacuum, which leads to very ordered arrays of molecules on surfaces, or by making a solution of the quantum bits and dropping this onto the surface,” says van Slageren. According to the Stuttgart researcher, there are a number of methods for getting the molecules from the solution onto the surface.