Chip-to-Chip Quantum Teleportation: What Bristol Physicists Did

Facebook ↗WhatsApp ↗

Originally published 2020-02-16.

Quantum Teleportation Between Photonic Chips

In December 2019, a team from the University of Bristol and the Technical University of Denmark (DTU) reported the first quantum teleportation between two separate silicon chips. The quantum state of a single photon on one chip was transferred to a photon on a second chip, with a fidelity of 91%. The results were published in Nature Physics.

That sentence packs in a lot of jargon. Below, we unpack what was actually moved between the chips, why the word "teleportation" is both accurate and misleading, and why engineers care about doing this on tiny, mass-producible chips rather than on a laboratory bench.

What the experiment achieved

The researchers built programmable circuits for light on silicon, the same material used in ordinary computer processors. Each chip carried nanoscale waveguides that guide and steer single photons, along with sources that create those photons in the first place.

Their circuit included four photon sources that were almost perfectly identical to one another. That uniformity matters: quantum effects between photons only work well when the particles cannot be told apart. With them, the team created entangled states of up to four photons and processed up to four qubits in a reprogrammable circuit.

The headline result was teleportation from one chip to another. A quantum state carried by a photon on the first chip was reproduced on a photon on the second chip, with 91% fidelity. Fidelity is a measure of how closely the received state matches the original; 100% would mean a perfect copy of the state, while values well above the classical limit show that genuinely quantum effects were at work.

The team also showed entanglement swapping, a related technique in which two particles that have never met become entangled through measurements on their partners. That is one of the basic building blocks of a future quantum internet.

What quantum teleportation is, and what it is not

The name invites images from science fiction, so it helps to be precise. Quantum teleportation does not move matter or energy. Nothing physically jumps from one chip to the other. What moves is information: the exact quantum state of a particle.

The idea was proposed in 1993 by Charles Bennett and five colleagues, and the first laboratory demonstrations came in 1997 from groups in Rome and Innsbruck. The recipe goes like this:

  1. Two particles are prepared in an entangled state and shared between the sender and the receiver.
  2. The sender lets the particle whose state should be transferred interact with their half of the entangled pair, and measures the two together.
  3. The sender then tells the receiver the result through an ordinary channel, such as a cable or radio signal: two classical bits for every qubit.
  4. Using those two bits, the receiver applies a simple correction to their particle, which then carries the original state.

Two consequences follow. First, the original state is destroyed in the process, so nothing is cloned; physics forbids making a perfect copy of an unknown quantum state. Second, because step three needs a normal signal, teleportation can never send information faster than light.

Entanglement in plain words

Entanglement is the resource that makes this possible. When two particles are entangled, their properties are linked so that measuring one immediately tells you something about the other, however far apart they are. Einstein was uneasy about this and famously called it "spooky action at a distance."

A common myth is that entanglement lets you signal instantly across the universe. It does not. Each individual measurement result looks random on its own; the correlation only becomes visible when the two sides compare notes, and that comparison travels at or below light speed. Entanglement is better thought of as a shared resource that makes certain tasks possible, not as a hidden telephone line.

Why doing it on silicon chips matters

Earlier teleportation records were set over impressive distances. In 2017 a Chinese team sent quantum states from a ground station in Tibet up to the Micius satellite, up to about 1,400 km away. Those experiments needed large optical tables, telescopes and carefully aligned equipment.

The Bristol and DTU work points in a different direction: miniaturisation. The chips were made with CMOS fabrication, the standard process used across the semiconductor industry. Yunhong Ding of DTU credited the low-loss silicon photonics made there for making the experiment possible. If quantum light circuits can be produced in the same factories as normal chips, they could one day be made cheaply and in large numbers, much as classical electronics were.

Lead author Jianwei Wang, by then at Peking University, suggested that combining quantum photonic devices with classical electronic controls on a single chip could lead to fully chip-based quantum networks.

How close is this to everyday technology?

Not very, at least not yet. The experiment ran in a controlled laboratory, and the chips still relied on external lasers, detectors and cooling equipment that will not fit inside a laptop. A 91% fidelity is excellent for a first chip-to-chip demonstration, but practical networks need even higher figures and much longer distances.

The realistic near-term uses lie elsewhere:

  • Linking quantum processors. A large quantum computer may be built from many smaller modules. Teleportation is a way to pass quantum information between them without disturbing it.
  • Quantum repeaters. Entanglement swapping can extend secure links beyond the distance that light in optical fibre can travel without being lost.
  • Secure communication. Quantum key distribution, already tested commercially, benefits from compact and reliable photon sources like the ones shown here.

Common misunderstandings to avoid

  • "Scientists teleported an object." No. They transferred the state of a single photon.
  • "Information was sent without any connection." The quantum state was not carried by an electrical link, but the protocol still needs a classical signal between the two sides.
  • "This is faster than light." It is not, for the reason explained above.
  • "Quantum computers will arrive next year because of this." The work is one useful step among many. Building large, error-corrected quantum machines remains a long-term engineering effort.

What makes the result notable is not a dramatic headline but its practicality: a key quantum operation performed between two chips built with the industry's everyday manufacturing tools.

Sources: Llewellyn et al., Chip-to-chip quantum teleportation and multi-photon entanglement in silicon, Nature Physics 16, 148–153 (2020); University of Bristol press release, 23 December 2019; Wikipedia: Quantum teleportation.

COMMENTS