Author: Daniel Oberhaus
Today marked the unveiling of a reprogrammable optical chip that is capable of processing photons in an infinite variety of ways. This development marks a massive step toward the realization of a quantum computer capable of wildly outperforming its most powerful classical counterparts, a technological feat that has been dreamt of for decades.
The new chip was heralded by the team of researchers from the University of Bristol and the Japanese telecom company Nippon Telegraph and Telephone as a “quantum optics lab-on-a-chip” because it brings together a number of pre-existing quantum experiments on a single device, drastically cutting down the amount of time and resources needed to design and run experiments to test theories about quantum computing. The work is detailed in a report in Science.
“A whole field of research has essentially been put onto a single optical chip that is easily controlled. The implications of the work go beyond the huge resource savings,”said Dr. Anthony Laing, a research fellow at the University of Bristol’s Centre for Quantum Photonics. “Now anybody can run their own experiments with photons, much like they operate any other piece of software on a computer.”
Quantum computing was first postulated by renowned physicist Richard Feynman in 1981 in response to the problem of simulating quantum mechanics on a computer. Quantum mechanics, which deals with interactions between individual atoms and particles, involves so many variables that the amount of memory required by a classical supercomputer to model quantum interactions is basically impossible. While Feynman was not the first to recognize this problem, he was one of the first to propose a solution: rather than model quantum mechanics on a classical computer, why not just make a quantum computer?
“We carried out a year’s worth of experiments in a matter of hours.”
So, in the 30-odd years that have elapsed since Feynman’s original query, the race has been on to realize Feynman’s quantum computer. The stakes are large, with the advent of quantum computing promising to revolutionize everything from cryptography to enabling pharmacology, the design of new drugs, quicker database searches and the modeling of quantum interactions which previously required massive, costly IRL undertakings like the Large Hadron Collider to observe.
While quantum technologies already exist, a quantum computer that outperforms the most powerful classical computers is still years away. The slow pace of development in quantum computing is largely due to just how hard it is to build and execute experiments to test theories about quantum computing, due to the volatile and delicate nature of quantum systems.
The Bristol and Tippon team’s optical chip may have just blown open the doors for the development of a quantum computer however, ushering in a new era for quantum information processing insofar as the device is able to perform experiments “in a matter of hours” that would have previously taken months to design and execute, in addition to performing processes that were simply impossible before.
“This chip is universal for linear optics, meaning that it can implement any conceivable linear optical protocol that is useful for quantum information processing with photons,” Laing told me over email. “No one chip has before demonstrated such disparate protocols as this one.”
The Bristol team from left to right: Chris Sparrow, Chris Harrold, Jacques Carolan, Dr Anthony Laing. Image: University of Bristol
Prior to today, a chip would be fabricated for a specific task, such as a particular quantum logic gate. According to Laing, this new chip is so versatile that it is capable of implementing a linear optical quantum logic gate that hasn’t even been dreamt up yet.
The silicon based optical chip is something of an equivalent to a modern computer’s central processing unit, but rather than having its data encoded as digital bits (either 1 or 0), the linear optical processing unit (LPU) works by encoding qubits (represented by 1, 0 or some superposition of these states) in polarized photons. It can process photons in an infinite variety of different ways in up to six modes, which can be thought of as analogs to channels or optical fibers.
Such optical approaches to quantum information processing carry with them a number of benefits, such as the relative ease with which researchers can maintain entanglement (a state of affairs in which the quantum state of multiple qubits depend on one another) with a photon as opposed to a particle. According to Laing there are also more practical benefits, such as the fact that photons exhibit low noise characteristics, are well isolated from the environment and can take advantage of well-established technology from the telecom industry.
What makes the team’s LPU different from other optical chips is the apparent ease with which the chip can be reprogrammed for use with different experiments. To demonstrate the chip’s versatility the team conducted a number of quantum information protocols, successfully demonstrating tasks which were not previously possible.
“This chip has demonstrated logical operations between quantum bits (qubits) that are basic logical gates in a quantum computer,” Laing said, enumerating the experiments demonstrated by the team on this chip. “It has also demonstrated entangling operations between qubits, which is a basic requirement for a different type of quantum computer, and it demonstrated 100 boson sampling experiments, which is a specialised quantum computing protocol designed to rapidly show that photons in linear optics can perform a task that cannot be matched by regular classical computers.”
The University of Bristol’s unique Quantum in the Cloud program, the first such service to make quantum processing available to the public, plans to add more chips such as this to its service in an effort to make quantum computing more publicly available. Such accessibility is due in large part to the increasing overlap between public and private investment in quantum information technology, which helps explain the partnership between Bristol and Tippon.
According to the researchers, such collaboration across the public-private divide is entirely necessary if the goal is to make a quantum computer possible, a goal that is looking increasingly realistic thanks to advances such as this new optical chip. As Laing put it, “NTT are geniuses at waveguide technology and it makes perfect sense for us to work with them to realise our ideas.”
While Laing acknowledged that he and his team had plenty of work to complete in the lab before this chip will ever see less experimental applications, the successes detailed in the Science paper bode well for the future of the device and the quantum computer it will help to make possible.
Suffice it to say, if quantum computing was in its infancy yesterday, today it may have just hit puberty.