Researchers at Seoul National University and the University of Seoul have developed a groundbreaking programmable photonic integrated circuit that offers unprecedented control over the speed of light, enabling it to be slowed down precisely when and where needed. This innovative technology, detailed in the prestigious journal Advanced Science, has the potential to revolutionize optical computing and address critical bottlenecks in the rapidly expanding fields of artificial intelligence and large-scale data processing.

The pioneering work was spearheaded by Professors Namkyoo Park and Sunkyun Yu from the Department of Electrical and Computer Engineering at Seoul National University, in collaboration with Professor Xianji Piao from the School of Electrical and Computer Engineering at the University of Seoul. Their collective expertise has yielded a device that moves beyond the limitations of fixed optical components, ushering in an era of dynamic and adaptable photonic systems.

The insatiable demand for computing power, driven by the exponential growth of generative AI and massive AI models, is placing immense strain on data centers and servers. Conventional electronic semiconductors, despite their widespread use, are encountering fundamental limitations. Their energy consumption is substantial, and their capacity to transmit data at ever-increasing speeds is nearing its physical limits. This growing disparity has reignited interest in optical computing, a paradigm that leverages light instead of electrical signals for information processing. The allure of optical computing lies in its promise of dramatically higher data speeds and significantly reduced power consumption.

However, light, while incredibly fast, presents its own set of challenges. Its inherent, fixed speed makes it difficult to introduce delays or temporarily store optical signals. These capabilities are crucial for implementing essential functions like buffers and memory within optical computers, akin to how electronic systems manage data flow. Without the ability to precisely control the timing and presence of light signals, the full potential of optical computing remains elusive.

To overcome this hurdle, the research team engineered a programmable photonic circuit designed to meticulously control both the speed and the shape of optical signals. Their novel approach offers a level of flexibility in manipulating "slow light" that surpasses previously conceived methods. The ability to dynamically adjust the speed of light within a chip opens up a vast array of new possibilities for optical system design.

Photonic integrated circuits (PICs) are increasingly recognized as a transformative technology for efficient and rapid information processing using light. In the intricate architectures of data centers, optical communication networks, and the computing systems of the future, the sheer speed of signal transmission is only one piece of the puzzle. Equally important is the precise orchestration of these signals, ensuring they arrive at their destinations at the correct time. In many scenarios, a light signal must be intentionally delayed to maintain synchronization with other information streams traversing the system.

A common technique for achieving such delays is known as coupled-resonator-induced transparency (CRIT). CRIT operates by exploiting interference effects among multiple optical resonators. This phenomenon allows light within a specific frequency range to propagate through a device while simultaneously reducing its effective speed. While CRIT has been a valuable tool, traditional implementations suffer from a significant drawback: their operating characteristics are typically fixed during the manufacturing process. This immutability means that once a CRIT device is fabricated, its behavior is largely set.

For instance, engineers aiming to achieve a longer signal delay or to work with a different frequency range would traditionally be compelled to design and manufacture an entirely new photonic device from scratch. This lack of adaptability introduces considerable complexity and cost into the design and production of optical communication hardware and data center infrastructure. It also prolongs development cycles and escalates expenses whenever new or modified capabilities are required. This issue is particularly acute for AI servers and next-generation data centers, where colossal volumes of data must be processed in real-time. The rigidity of conventional optical components has thus remained a significant impediment to the widespread adoption of more practical optical computing systems.

The research team’s breakthrough lies in their development of a fundamentally different strategy. They reimagined the two primary optical states within CRIT systems – the bright mode and the dark mode – not as independent entities but as a unified degree of freedom. This conceptual shift, combined with the integration of two controllable loop couplers, formed the bedrock of their novel design principle for programmable photonic integrated circuits. This innovation allows resonator arrangements that were previously locked into a single configuration post-fabrication to be dynamically adjusted for a variety of purposes.

Using this new CRIT architecture, the researchers successfully demonstrated the ability to delay and control the movement of light as needed. Crucially, they showed that the interference between the bright and dark modes could be managed as a single, integrated design parameter. This holistic approach dramatically expands the flexibility of photonic resonator circuits, liberating them from the constraints of fixed designs that have long characterized the field.

The implications of this enhanced control are far-reaching. The researchers theoretically proved that the two loop couplers can be employed to fine-tune the bandwidth and shape of the optical passband. More importantly, they can precisely control the duration of signal delays and optimize the efficiency with which these signals traverse the circuit. This means that both the speed and the transmission characteristics of optical signals can be reconfigured dynamically across entire systems comprising multiple resonators, rather than being confined to the limitations of a single resonator.

Furthermore, extensive numerical simulations have corroborated the dynamic adjustability of optical pulse speeds while the circuit is actively operating. The results are highly encouraging, indicating that signal delay times can be altered without compromising processing performance. The system also exhibits the remarkable ability to convert the frequency of light without the need for additional, specialized components, further simplifying optical architectures.

To ascertain the practical feasibility of their design, the researchers conducted comprehensive three-dimensional electromagnetic simulations. These simulations evaluated the potential of building the CRIT device on a silicon nitride (Si₃N₄) photonic integrated circuit platform, a widely used and mature technology. They also meticulously assessed a range of real-world factors that could impact the device’s performance during manufacturing and operation. This included an analysis of material losses, variations in resonator quality, signal backscattering, fluctuations in optical coupling, phase errors within the loop couplers, and thermal crosstalk. The simulation results consistently indicated that the proposed structure could maintain reliable operation even under these realistic, challenging conditions.

The study introduces a programmable photonic platform with the remarkable capacity to control both the timing and frequency properties of light signals in real-time. This design offers a compelling solution to the limitations of conventional optical delay devices, which are typically designed for single, fixed functions. The research also suggests a future where several critical optical functions could be integrated onto a single photonic circuit. These potential integrated capabilities include sophisticated signal synchronization, adjustable delay lines, efficient optical buffers, and versatile frequency conversion.

Beyond the immediate applications, the fundamental design principles explored in this research hold promise for a broader impact. The researchers believe that their approach can be extended to a wide spectrum of photonic circuits based on resonators, potentially laying the groundwork for more adaptable and versatile optical signal processing technologies.

The commercialization of this technology could lead to a paradigm shift in data centers and AI servers. A single programmable optical chip could dynamically perform multiple tasks, including precisely controlling signal speed and seamlessly switching between different operational functions. In essence, such a chip could operate much like a software-defined system, with its behavior adaptable to evolving needs and demands. This inherent flexibility promises to enhance the efficiency of information processing in data centers and AI servers, while concurrently reducing their energy footprint.

The consolidation of multiple signal processing functions onto a single chip also portends a future of smaller, more cost-effective optical communication equipment and sensor systems. Looking further ahead, this technology could become a foundational element for industries that rely on extremely high-speed information processing, such as autonomous driving, the development of next-generation communication networks, and the advancement of quantum technologies.

Professor Namkyoo Park, a co-corresponding author of the study, expressed enthusiasm about the research’s significance: "This research is significant in that it proposes a new design principle that allows the flow of light within photonic integrated circuits to be reconfigured as needed, greatly enhancing design flexibility. We plan to expand this technology toward large-scale programmable photonic integrated circuits based on silicon photonics and photonic AI technologies."

Co-first authors Dr. Seungkyun Park and Ph.D. student Beomjoon Chae, who were instrumental in developing the theoretical framework and conducting the numerical analysis, added: "Through this study, we realized that reinterpreting conventional photonic resonator physics from a different perspective can serve as a starting point for discovering new functionalities in photonic integrated circuits. We plan to further develop this research toward practical device implementation and experimental validation." Dr. Seungkyun Park’s affiliations include the InnoCORE PICORE Center at KAIST and the Photonic Systems Laboratory at Seoul National University, where he researches photonic AI and quantum optics. Beomjoon Chae is currently engaged in research on programmable photonic integrated circuits at the Intelligent Wave Systems Laboratory at SNU.

The research received crucial support from the Ministry of Science and ICT through various programs, including the Innovative Research Center (IRC) program, the Basic Research Laboratory (BRL) program, and the Young Researcher Program. Dr. Seungkyun Park also benefited from support through the InnoCORE program (PICORE Center).