Jeonbuk National University (JBNU) Division of Advanced Materials Engineering (Major in Information Materials Engineering) research team led by Professor Park Yushin, together with researchers from the Korea Institute of Science and Technology (KIST), Sogang University, and Dongguk University, has developed a new photonic computing device that performs computation using only light without electricity. This research was conducted with support from the Mid-Career Researcher Support Program (RS-2025-00517064) of the Ministry of Science and ICT and the National Research Foundation of Korea (NRF), and by the Korea Institute of Science and Technology (KIST) institutional research program. The research outcomes were published in the latest issue of Advanced Materials (IF 26.8, JCR Materials Science, Multidisciplinary top 2.1%). The paper is titled “Light-Driven Reconfigurable Logic in a Monolithic Perovskite Device via Nonlinear Photoresponse Switching (http://doi.org/10.1002/adma.202509566).

The core of this study is the precise control of charge motion within a special material (halide perovskite), enabling the direction of current (positive/negative) to change solely based on the intensity and position of incident light. This allows a single device to perform and reconfigure all basic logic operations such as AND, OR, and NOT. A key feature is that it operates even when the voltage is set to 0 V. In other words, the work demonstrates the practical feasibility of light-based computing that consumes almost no power. The paper's first author is PhD candidate Andan-tae (KU–KIST Convergence Graduate School).

Unlike previous approaches that required complex circuits or many transistors, the team's technique implements computation by exploiting the intrinsic properties of the material itself. This suggests potential for developing a low-power, highly integrated photonic computing platform that is particularly important for small devices and battery-powered applications.

● First author: Andan-tae (PhD candidate, KU–KIST Convergence Graduate School). Collaborating institutions: KIST; Sogang University (Prof. Min‑A Seo); Dongguk University (Prof. Soon‑Chul Kwon)

● Operation without electricity: Logic operations possible using only light in the 0 V state

● All operations in a single device: A single device can be reconfigured to implement all basic gates such as AND, OR, NOT, XOR

● Ultra-low power and high integration: Reduced circuit complexity lowers energy consumption and heat generation

● Potential for real-time optical computation: Foundation for systems that directly convert camera or sensor inputs into computation and decisions

In 2022 (Nature Communications) the research team first demonstrated a device that computes using only light without electricity, implementing five basic logic operations in a single device. The present study advances that work by enabling all basic logic operations to be freely reconfigured and executed within one device. In other words, it establishes the basis for a mature photonic computing device that operates solely by light. The technology is well-suited to become compact, nearly power-free, fast-decision sensor–computing devices, with applications in autonomous driving, robotics, smart manufacturing, healthcare, and environmental and safety monitoring. However, to reach commercial products, additional verification of durability, lifetime, packaging, and chip connectivity is required. The research team plans to continue work in the sequence: device arraying → reliability evaluation → system integration.

Expected effects and application areas

● Edge AI: Integrated autonomous sensors for sensing–computation–decision-making in environments with constrained power infrastructure

● Optical neuromorphic computing: Hardware for learning and inference based on direct optical computation of visual information

● Defense, space, and environmental monitoring: Highly reliable, ultra-low-power platforms

● Next-generation optical communications and security: Integration of optical signal processing and computation

Professor Park Yushin said, “This work is significant because it performs computation using only the material’s intrinsic properties without stacking many complex circuits. It can be extended into a new form of intelligent system in which sensors themselves perform computation—i.e., systems that ‘see and immediately decide.’” He added that the technology could contribute to the development of real-world perception and decision systems such as Physical AI, which Jeonbuk Province has shown interest in.

PhD candidate Andan-tae (KU–KIST Convergence Graduate School) stated, “It is very meaningful to have advanced research begun during my internship into this achievement. I am grateful that the work was accepted by a journal of international repute.” Andan-tae is expected to graduate in February 2026 and has attracted significant attention, receiving postdoctoral offers from leading global sensor and semiconductor research groups based on outstanding research results.