Quantum memory array brings us closer to a quantum RAM

Internet, social media and digital technologies have completely transformed how we establish commercial, personal and professional relations. Basically, this company is based on the exchange of information expressed in terms of bits. This basic information unit can be either a 0 or a 1, and it is generally represented in electrical circuits, for example, in two tension levels (one representing the bit in state 0 and the other representing state 1).

The possibility of storing and manipulating bits effectively throws the database of digital electronics and allows modern devices to perform a variety of tasks, ranging from email and reading music to digital simulations. These processes are only possible thanks to key material components such as random access memory (RAM), which offer temporary storage and recovery at data demand.

In parallel, the progress of quantum physics have led to a new type of information unit: QUBIT. Unlike conventional bits, which are strictly 0 or 1, qubits can exist in a superposition of the two states at a time. This opens up new possibilities for processing and storing information, although its practical implications are always in development.

Future quantum computers and a quantum internet will also require quantum memories (especially quantum memories with random access) to store and recover qubits. Despite several existing approaches to code qubits and to implement quantum memories, no “stallion” has yet emerged.

From now on, ICFO researchers, Dr. Markus Teller, Susana Plascencia, Cristina Sastre Jachimska, Dr. Samuele Grandi, led by Professor Icrea Hugues de Riedmatten, have reached an important step in the development of quantum memories in the solid state – one of the most promising platforms for the storage of quantum information.

In a recent Physical review X Article, they use a table of ten individually controllable memories to store qubits in arbitrary combinations of memory cells and recover them on request. These results are based on a previous Quantum information NPJ Publication, where they first introduced the table.

Their work focuses on two quedations of qubit widely used in quantum photonic technologies: the coding of the path, where the qubit is defined by which memory the photon enters and the coding of the bine in time, where the qubit is coded in the arrival time of the photon (at an anterior or subsequent time interval). For the latter, the team used a unique characteristic of its approach: the possibility of storing photons in several time locations in each memory cell.

Ten cells, a crystal: advanced quantum communications

In the Quantum information NPJ Paper, the team created a range of ten quantum memories using a praseodymium doped crystal cooled at 3 Kelvin inside a cryostat. In this crystal, they allocated 250 “locations” of storage or spatio -temporal modes, each potentially storing a photon – the current world record for a solid device with recovery on demand. Such an achievement is really remarkable, because the abilities on demand are technically very difficult to implement, and yet they are essential to synchronize quantum networks.

The team then used a similar configuration – ten cells of memory addresses individually but with fewer modes available – to store several qubits and recover them on demand, which finally led to Physical review X article. To do this, acousto-optical deflectors have led laser pulses to write and read qubits in arbitrary combination of memory cells. The posterior analysis of the recovered photons showed that the quantum memory network retained the quantum states of origin with reasonable fidelity.

To present the potential of their configuration, the team stored two time bar qubits and recalled the two at the same time. These capacities bring us closer to a quantum memory in the solid state in the solid state with random access, with applications in quantum IT and communications.

“We plan to combine this platform with a source of photonic cluster states for quantum computer-based computer”, shares Dr. Markus Teller, the first co-author of the study. “In this scenario, the quantum memory picture would store more and more photons until a large quantum tangle is formed. Then, quantum operations could begin.”

The system could also advance quantum repeaters, the backbone of the future quantum internet. These devices aim to extend quantum communication over large distances by distributing the quantum resource of the tangle between successive segments.

“Previous experiences with solids had to stop after only a few dozen entanglement attempts, waiting for a success signal,” said Susana Plascencia, ICFO researcher and co-author of the study. “With our table, we no longer need to wait for the success signal (at least, up to a certain distance). Instead, we can move on to another memory cell and continue to try.”

The filling of this inactivity time with new attempts could increase the rate to which entanglement – and therefore quantum information – is transferred to long distances.

To fully exploit the potential of quantum memory networks multiplexed over time, the next challenge will be to increase performance (for example, in terms of efficiency and storage time), to increase the number of memory cells and be able to store tangles.

Overall, this study represents a significant step towards the quantum equivalent of RAM, whose implications in quantum communications and IT remain open.