In the quantum realm, at least insofar as we can understand, particles are in a state of superposition, wherein they exist in two or more states simultaneously — a cat that is both dead and alive, so to speak. However, this “coherence” lasts for only a fraction of a second before the whole system decoheres — a phenomenon that marks the transition from the realm of quantum to classical mechanics.
The fact that a coherent quantum state is short-lived is what allows reality as we know it to exist, but, for researchers looking to exploit superposition to create quantum computers, this presents a major roadblock. For these researchers, looking for ways to least delay decoherence — thereby preserving the state of superposition that makes quantum computers so much faster than their conventional counterparts — is a key goal.
Now, in a study published in the latest edition of the journal Quantum Science and Technology, a team of researchers has demonstrated the storage and retrieval of quantum information in a single atom of phosphorus embedded in a silicon crystal. This “quantum memory” was created using the spin of the phosphorous nucleus, which behaved as long-lived qubit — a quantum computing equivalent of a bit.
For the purpose of their experiment, the researchers implanted phosphorus atoms in a 900 nanometer thick layer of enriched silicon. Radio-frequency (RF) pulses were used to transfer spin states (quantum information) to and from the phosphorous nucleus. The electron spin, which is less resilient to electromagnetic stimulation than the nuclear spin, was used as a processing qubit that the scientists used to read and write data.
“We have experimentally demonstrated a nuclear spin quantum memory using a single P atom in silicon. We measured memory coherence times up to 80 ms and process fidelity better than 80 percent,” the researchers wrote in their study. “We have identified a likely cause of the coherence and fidelity limitations, namely a shift in the instantaneous electron resonance frequency after applying RF pulses.”
The next step in the research would be to increase the fidelity, such that the electron’s final state matches its initial state — the one that existed before information was transferred to it from the nucleus — in 99 percent of the cases. If this is achieved, it would mark a significant breakthrough in the efforts to create viable and scalable quantum computers.
“Future work will focus on understanding and eliminating this pulse-induced frequency shift, in order to demonstrate quantum memory fidelities approaching the values of gate fidelity already observed for the operation of the electron and nuclear spin qubits individually,” the researchers wrote in the study.