(Image source: Internet) In 2009, University of Bristol professor Jeremy O'Brien published a research paper describing how on-chip optics can be reused to manipulate individual optical particles and perform quantum operations. On-chip optical components were originally developed

2025/04/1004:46:40 science 1316

(Image source: Internet) In 2009, University of Bristol professor Jeremy O'Brien published a research paper describing how on-chip optics can be reused to manipulate individual optical particles and perform quantum operations. On-chip optical components were originally developed  - DayDayNews

(Image source: Internet)

In 2009, Jeremy O'Brien, professor of at the University of Bristol, published a research paper describing how to reuse on-chip optical components to manipulate single optical particles and perform quantum operations. On-chip optical components were originally developed by the telecommunications industry.

After early photon research, by 2016, O'Brien and three of his academic colleagues Terry Rudolph, Mark Thompson, and Pete Shadbolt created PsiQuantum.

These founders all believe that it will take too long to build a practical quantum computer with traditional methods. At the beginning of the company's establishment, the PsiQuantum team established the goal of building a million qubits fault-tolerant photon quantum computers. They also believe that the only way to make such a machine is to make it in a semiconductor foundry.

Early progress

PsiQuantum first attracted attention about two years ago when it received $150 million in Series C financing, increasing the company's total investment to $215 million.

This investment rule reflects the great interest in the potential of any quantum device PsiQuantum is building. At that time, PsiQuantum operated in stealth mode, so there was little information available about its research.

Last year, after receiving another $450 million in Series D financing, PsiQuantum disclosed more information about its technology. Just a few weeks ago, a $25 million U.S. government grant was jointly awarded to PsiQuantum and its manufacturing partner GlobalFoundries for processing and further developing its photonic quantum computers. Having GlobalFoundries as a partner is a guarantee of quality. GlobalFoundries is a high-quality first-class wafer fab and one of the three largest first-class foundries in the world.

PsiQuantum is currently valued at $3.15 billion, and the quantum roadmap it follows is mainly paved by its own designed stepping stones, with the unique technologies, components and processes needed to build a universal silicon photonic quantum computer in millions of quantum bits.

Basic technology

Classic computers use digital bits to encode information to represent zero or one. Quantum computers use qubits, which can also represent one or zero, or quantum superposition state of a certain number between zero and one at the same time. There are a variety of qubit technologies. IBM, Google and Rigetti use qubits made of small coils that turn into superconductors when the temperature is low. Quantinuum and IonQ act as quantum bits using ions formed by the removal of outer valence electrons from ytterbium atoms. Atom Computing uses isotopes of strontium to make neutral atom spin qubits.

light is used for various operations in superconductors and atoms quantum computers. PsiQuantum also uses light and turns infinitesimal photons into qubits. There are two types of photon qubits: extruded light and single photon. The technology selected by PsiQuantum is single photon qubits.

Using photons as qubits is a complex process. Among the trillions of photons with different frequencies and energies, it is very complicated to determine the quantum state of of a single photon.

Pete Shadbolt is the co-founder and chief scientist of PsiQuantum. His responsibilities include overseeing the application and implementation of technical and scientific related policies and procedures that are crucial to the success of PsiQuantum. After obtaining his PhD in Experimental Photon Quantum Computing from the University of Bristol in 2014, he was engaged in postdoctoral research at Imperial College of Technology and studied photon quantum computing theory. During Bristol , he showed the first variable component quantum feature parser and the first public API for quantum processors. He was awarded the 2014 EPSRC "Rookie" by the British Research Council, and won the EPSRC Inspirational Scientist and Engineer Award, and the European Physics Society Paper Award.

Shadbolt Dr. explained: Detection of a single photon from a beam is similar to collecting a drop of designated water from the widest part of the Amazon River. This process takes place on a coin-sized chip, the PsiQuantum chip embodies extraordinary engineering and physics, and we are constantly improving the chip's fidelity and single-photon source performance.

has strict requirements on photons used as qubits. Consistency and fidelity are crucial to the performance of photonic quantum computers. Therefore, each photon source must have high purity, proper brightness, and produce consistent photons.

Please partner

A year ago, when PsiQuantum announced its Series D financing, the company revealed that it had established a partnership with GlobalFoundries to pioneer manufacturing processes for photonic quantum chips. This manufacturing process produces 300 mm wafer , which contains thousands of single-photon sources and single-photon detector . The wafer also contains interferometer , separator and phase shifter . To control the photonic chip , the GlobalFoundries factory also manufactures advanced electronic CMOS control chips with approximately 750 million transistors.

(Image source: Internet) In 2009, University of Bristol professor Jeremy O'Brien published a research paper describing how on-chip optics can be reused to manipulate individual optical particles and perform quantum operations. On-chip optical components were originally developed  - DayDayNews

GlobalFoundries

Photon advantages

Each qubit technology has its own advantages and disadvantages. PsiQuantum chose to use photons to build its quantum computer for the following reasons:

1. Photons cannot sense heat, and most photon components work at room temperature;

2. PsiQuantum's superconducting quantum photon detector needs to be cooled, but its operating temperature is 100 higher than that of superconducting quantum bits. :

3. Photon qubits are compatible with fiber network, making it easier to route photons between local devices;

4. Photons are not affected by electromagnetic interference . Another major advantage of

photonic qubits is: It can maintain the quantum state for a longer time. For example: Despite the spread of billions of years, the light emitted by distant stars and galaxies still arrives at Earth intact. The longer a qubit holds its polarized quantum state, the more quantum operations it can perform, making quantum computers more powerful.

Why should I start with one million qubits?

Shadbolt said: "We have cracked the code to build a million-quantum quantum computer, and although it is a huge number, it is not difficult. The process we have to do is similar to: putting billions of transistors into our phones. We are now building quantum chip next to laptops and mobile chip on the GlobalFoundries 300mm platform."

According to Dr. Shadbolt, PsiQuantum's custom production line has made great progress. Surprisingly, building a million-quantum bit quantum machine in a foundry has many of the same non-quantum problems as assembling a classic supercomputer, including chip yield, reliability, high-throughput testing, packaging and cooling. "From our announcement of our collaboration with GlobalFoundries, we have produced a large amount of silicon, made seven streams in total, and hundreds of thousands of silicon wafers come out of our doors. We have invested heavily in packaging, assembly systems, integration and fiber connections to ensure the highest efficiency of optical inflow and out of the chip."

PsiQuantum is conducting a lot of research and continuously improving the performance of photonic components and processes. In addition to high-performance optical components, technologies that support this process are also very important, such as optical switches, fiber-to-chip interconnection and bonding methods."In the last few streams of GlobalFoundries, we greatly improved the efficiency of the photon detector, with fewer and fewer photons lost from the system. In our recent chips, we also reduced the waveguide loss to extremely low levels. This involves a lot of innovation. A single photon light source is a great example of irradiating the laser directly into the chip to run a single photon source. The intensity of the laser is about one trillion times the single photon that needs to be detected, so the light on that chip must be attenuated by about one trillion times."

Shadbolt attributes the success of PsiQuantum's manufacturing success to GlobalFoundries. There are significant differences between second-tier chip foundries compared to first-tier chip foundries like GlobalFoundries, and the build chips required by PsiQuantum can only be built through extremely mature manufacturing processes.

He said: "PsiQuantum has two demanding requirements. We need a large number of components, and these components must consistently meet extremely demanding performance requirements. Few reliable partners in the world can achieve this goal, and working with mature chip foundries like GlobalFoundries will be the key to our strategy."

This cooperation is also beneficial to GlobalFoundries, because it also adds PsiQuantum's silicon optical chip production capacity to the factory and gains more new technology experience.

endpoint in the hope

According to Dr. Shadbolt, the initial question was whether it is possible to manufacture a large number of quantum devices in foundries. Incorporating new equipment directly from university labs into the manufacturing process has been difficult, it is time-consuming and very expensive. A nano single photon detector is an example of this.

PsiQuantum's semiconductor roadmap has only a few projects left to be completed. Since a single chip cannot accommodate one million qubits, quantum computers will need to use optical fiber to connect multiple quantum processor chips, and add ultra-high performance optical switch to ultimately realize the stealth transmission and entanglement of single-photon operations between chips.

adbolt "The remaining problem is the optical switch. You might ask, why do people with photonic quantum computing never build anything on a large scale? Or why don't they show a very large entangled state ? The reason is that a special optical switch is needed. It must have very high performance and be better than any existing state-of-the-art optical switch, such as those used in telecom networks. Although the optical switch is a classic device whose only function is to turn on and off the optical signal between waveguides, in quantum computers it must be done with extremely low losses and very high speeds, so the performance requirements for optical switches are extremely high."

can't buy it, so it makes your own

For PsiQuantum In other words, developing optical switches with suitable performance is a success or failure. Since there is currently no suitable commercial optical switch, PsiQuantum has no choice but to make one. They have been investing heavily in the development of high-performance optical switches over the past few years. "This is one of the most exciting things PsiQuantum is doing. Building an extremely high-performance optical switch is the next big thing in our roadmap. We believe this is the key to opening up the huge prospect of optical quantum computing."

Summary

Summary

PsiQuantum is built on the belief that optical quantum is the correct route to building fault-tolerant quantum computers with millions of qubits and should be made based on semiconductor process manufacturing. At the same time, the founder hopes that it will not become a huge machine as the scale of qubits expands.

Considering the high complexity of the overall process of building a million quantum bit quantum computer and the lack of verification of related tools and processes, the progress made by PsiQuantum since its establishment is amazing.It has partnered with the world's best foundries, completed seven streamers and funded six new processes to build the first wafer manufacturing process that integrates superconducting single-photon detectors into conventional silicon photon detectors.

Today, it addresses another challenge by building light switches to fill in the gap.

It's no surprise that ultra-high performance optical switches are a key part of PsiQuantum's plan to build a scalable million-qubit quantum computer. Other quantum companies are also planning to integrate similar optical switching technologies within a decade to extend the modular quantum processing unit architecture. On the other hand, the high-performance optical switches that PsiQuantum is developing can also be sold as a standalone product in the future, deployed in data centers of millions of qubits, connecting tens of thousands of quantum processing units, becoming a source of additional revenue.

After implementing the optical switch, it needs to be integrated into the GlobalFoundries manufacturing process. This is the last step required to complete the PsiQuantum Foundry Assembly process, which will then produce a photonic quantum computer chip.

However, even with a complete end-to-end manufacturing process, it takes more time to build a mature fault-tolerant quantum computer. PsiQuantum will continue to build complete quantum computers around chips produced by GlobalFoundries. To do this, it will require a well-trained workforce and the location and infrastructure that can assemble, integrate, test and distribute large quantum bit photonic quantum computers.

is based on the post-casting workload, the development and assembly of optical switches, and assuming that no major technical problems or delays occur, it is believed that in ten years, PsiQuantum will only provide a photonic quantum computer of any scale.

Shadbolt said: "Although optical switches are very powerful general-purpose technology that many people are interested in, we are not interested in its universality. We are only interested in one thing - building a quantum computer with better performance than all supercomputers on the planet. This is our only goal."

compiled: Wang Yan

Edited: Mu Yi

Edited: Mu Yi

science Category Latest News