Researchers Vladimir Shalaev, Purdue’s Bob and Anne Burnett Distinguished Professor, and Demid Sychev, a postdoctoral researcher in Shalaev’s group, have demonstrated a “photonic transistor” operating at single-photon intensities, published in Nature Nanotechnology. This achievement utilizes the avalanche multiplication process within silicon to create a nonlinear refractive index several orders of magnitude higher than previously known materials. This breakthrough, originating from the Elmore Family School of Electrical and Computer Engineering at Purdue University, addresses a long-standing challenge in photonics and paves the way for the development of practical terahertz-speed photonic computing.
Single-Photon Switch and Optical Nonlinearity
Researchers at Purdue University have demonstrated a “photonic transistor” operating at single-photon intensities, achieving optical nonlinearity several orders of magnitude higher than previously known materials. This breakthrough utilizes the avalanche multiplication process from commercial single-photon detectors, where a single photon can trigger a cascade generating up to 1 million electrons. This allows a single-photon control beam to modulate a much more powerful probe beam, effectively functioning as an optical switch with potential for terahertz-speed computing.
This new approach offers three key advantages over existing methods. It operates at room temperature, unlike many current single-photon nonlinearity techniques. The technology is also compatible with complementary metal-oxide-semiconductor processes, allowing for integration into existing semiconductor manufacturing. Critically, the demonstrated device operates at gigahertz speeds and has the potential to reach hundreds of gigahertz, significantly faster than alternative methods.
The implications of this technology extend beyond quantum computing to classical computing, data centers, and optical communications. Existing photonics approaches offer potential for faster, more energy-efficient computing, but have lacked a suitable switching mechanism. This Purdue innovation could enable terahertz clock rates for CPUs, compared to the 5 gigahertz currently achievable, and represents a potential solution to a long-standing problem in the field.
Avalanche Multiplication for Enhanced Signals
Researchers at Purdue University have demonstrated a “photonic transistor” operating at single-photon intensities, achieving optical nonlinearity several orders of magnitude higher than existing materials. This breakthrough utilizes the avalanche multiplication process—commonly found in single-photon detectors—where a single photon can trigger a cascade generating up to 1 million electrons. This process bridges the microscopic quantum world with measurable macroscopic effects, enabling control of powerful optical beams with single photons, a key step toward practical photonic computing.
The Purdue team’s approach offers advantages over existing methods: it operates at room temperature, is compatible with complementary metal-oxide-semiconductor manufacturing, and potentially reaches hundreds of gigahertz speeds. Traditional methods for single-photon nonlinearity often require sensitive, low-temperature quantum systems. This new method offers a compact, semiconductor-based solution that could significantly increase the speed of computing—potentially reaching terahertz clock rates compared to current 5 gigahertz processors—and increase energy efficiency.
This technology leverages a single-photon avalanche diode (SPAD) to function as an optical switch, modulating a powerful probe beam with a single-photon control beam. While current SPADs are not specifically designed for this application, the team plans to fabricate optimized SPADs for this purpose. This advancement has implications beyond computing, potentially transforming data centers, optical communications, and data transfer systems through faster, more energy-efficient photonics.
Ideally, from photons, you can get terahertz clock rates of CPUs, compared to currently existing 5 gigahertz in the best cases. But the problem is that there are no photonic switches like this.
Purdue’s Technology: Advantages and Compatibility
Purdue University researchers have developed a “photonic transistor” operating at single-photon intensities, demonstrating a nonlinear refractive index several orders of magnitude higher than existing materials. This breakthrough utilizes the avalanche multiplication process—typically used in single-photon detectors—to amplify microscopic quantum effects into measurable, macroscopic ones. This approach addresses a key challenge in photonics: achieving strong interaction between photons without requiring high power levels, potentially paving the way for terahertz-speed computers.
The Purdue technology offers three key advantages over existing methods for single-photon nonlinearity. It operates at room temperature—unlike many current approaches—and is compatible with complementary metal-oxide-semiconductor manufacturing, allowing for seamless integration into existing semiconductor processes. Crucially, the device operates at gigahertz speeds, with the potential to reach hundreds of gigahertz, dramatically exceeding the capabilities of current photonic switches and offering a pathway to faster computing.
This advancement isn’t limited to quantum computing; it also holds promise for classical computing, data centers, and optical communications. The team is currently working to optimize the technology by fabricating single-photon avalanche diodes (SPADs) specifically designed for this application. Researchers anticipate this work will lead to significant advancements in both industry and academia, providing a potential solution to a longstanding problem in the field of photonics.
Potential Applications in Computing and Beyond
This research demonstrates a “photonic transistor” operating at single-photon intensities, a breakthrough potentially enabling terahertz-speed computers. The team achieved a nonlinear refractive index several orders of magnitude higher than existing materials by utilizing the avalanche multiplication process found in commercial single-photon detectors. This allows a single photon to control a much more powerful optical beam, effectively acting as an optical switch – a key component for faster computing and data transfer.
The Purdue team’s approach offers several advantages over current methods. It functions at room temperature, unlike many single-photon nonlinearity techniques that require extreme cooling. Crucially, the technology is compatible with complementary metal-oxide-semiconductor (CMOS) manufacturing, allowing for integration into existing semiconductor processes. Furthermore, the device operates at gigahertz speeds, with potential to reach hundreds of gigahertz—far exceeding the 5 gigahertz speeds of today’s best CPUs.
While offering benefits for quantum computing, the researchers believe this technology could be even more transformative for classical computing. The goal is to create photonic computers that consume less energy and operate at terahertz speeds. The lack of efficient photonic switches has been a major barrier, but this new method, utilizing single photons to control light, offers a potential solution and could impact data centers, optical communications, and data transfer systems.
Four-Year Research Journey and Next Steps
The research team’s four-year journey involved iterative experimentation in completely unknown territory, ultimately leading to a demonstrated “photonic transistor” operating at single-photon intensities. The work was conducted at Purdue’s Birck Nanotechnology Center with guidance from Professors Shalaev and Boltasseva. Initial success relied on commercially available single-photon avalanche diodes (SPADs), but the team acknowledges substantial optimization is needed to move the technology forward, marking this as a potential solution to a long-standing problem.
Currently, the team is focused on fabricating their own SPADs specifically designed for this application, aiming to optimize performance beyond the capabilities of commercially available components. They plan to explore different device geometries and materials to enhance the technology further. This ongoing work builds on a demonstration achieving optical nonlinearity several orders of magnitude higher than best-known materials, potentially paving the way for terahertz-speed computing.
The implications of this research extend beyond quantum computing, with potential applications in classical computing, data centers, optical communications, and data transfer systems. The team anticipates clock rates potentially reaching hundreds of gigahertz—a significant improvement over current 5 gigahertz CPU speeds—if this technology is fully realized. Sychev emphasizes that this work represents a potential solution to a problem hindering the development of faster, more energy-efficient photonic computing.
We demonstrated a way to realize a photonic transistor working at single-photon intensities.
Vladimir Shalaev
