Two real-world tests of quantum memories bring a quantum internet closer to reality

In the quest to build a quantum internet, scientists are putting their memories to the test. Quantum memories, that is.

Quantum memories are devices that store fragile information in the realm of the very small. Theyre an essential component for scientists vision of quantum networks that could allow new types of communication, from ultra-secure messaging to linking up far-flung quantum computers (SN: 6/28/23). Such memories would help scientists establish quantum connections, or entanglement, throughout a network (SN: 2/12/20).

Now, two teams of scientists have entangled quantum memories in networks nestled into cities, where the hustle and bustle of urban life can pose challenges to quantum communications.

These two impressive studies are pushing out of the lab and into real-world implementations, says physicist Benjamin Sussman of the University of Ottawa, who was not involved with the research. These are not just toy systems, but are really the first steps toward what future networks will look like.

In a network of two quantum memories connected by a telecommunications fiber link that traversed a 35-kilometer loop through Boston and Cambridge, Mass., scientists maintained entanglement for about a second, physicist Can Knaut and colleagues report in the May 16 Nature. That doesnt sound like a lot for us, but in the domain of quantum, where everything is more fleeting, one second is actually a really long time, says Knaut, of Harvard University. 

The researchers used quantum memories built from a tiny hunk of diamond in which two of the diamonds normal carbon atoms are replaced by one atom of silicon. That substitution creates a defect that serves as a quantum bit, or qubit. In fact, the defect serves as two qubits one thats short-lived, and another long-lived qubit that acts as the memory. Scientists prodded the short-lived qubit with a photon, or particle of light. The researchers used that qubit as a go-between in order to entangle the long-lived qubit with the photon. Then the scientists sent the photon through the fiber and repeated the process to entangle the long-lived qubits in each memory.

Meanwhile, in Hefei, China, entanglement was achieved in a network with three quantum memories separated by fiber links of about 20 kilometers, researchers report in the same issue of Nature. 

This teams quantum memory was based on a large ensemble of rubidium atoms about 1 millimeter in diameter. When hit with a laser, the ensemble of atoms can emit a photon. Rather than shuttling the photon directly to another quantum memory, the photon was sent to a centrally located station, where it was measured along with a photon sent from another memory. That generated entanglement between the two distant parts of the network.

Meeting up in the middle meant the photons didnt have to travel all the way to the other side of the network, an added bonus. This scheme is rather efficient, but its experimental realization is rather challenging, says experimental physicist Xiao-Hui Bao of the University of Science & Technology of China in Hefei. The technique required the team to find methods to correct for changes in the length of the fibers due to temperature shifts and other effects that could cause problems. This painstaking effort is called phase stabilization. This is the main technology advance we made in this paper, Bao says.

In contrast, the Boston network had no central station and didnt require phase stabilization. But both teams achieved whats called heralded entanglement. That means that a signal is sent to confirm that the entanglement was established, which demands that the entanglement persists long enough for information to make its way across the network. That confirmation is important for using such networks for practical applications, says physicist Wolfgang Tittel, who was not involved with either study.

If you compare how these two different groups have achieved [heralded entanglement], you see that there are more differences than similarities, and I find that great, says Tittel, of the University of Geneva. There are different approaches which are all still very, very promising.