Non-volatile programmable directional couplers for high-integration and low-loss photonic computing

The rapid growth of artificial intelligence creates a pressing need for high-speed, low-cost photonic computing technologies, which remain hindered by integration density, optical loss, and computational precision. Here, we demonstrate a non-volatile programmable directional coupler based on low-loss Sb 2 Se 3 phase-change material, featuring an insertion loss of 0.3 dB. The device achieved 197 distinct non-volatile optical states within a coupling length of only 28 μm. This high-precision control was enabled by a novel programming method leveraging a dual-band fiber array. Functioning as a matrix multiplication unit, this device facilitated the photonic neural network to attain an accuracy of 92.27% on the Fashion-MNIST dataset, paving the way for high-density, high-precision, and energy-efficient photonic computing.

Authors

Publication Details

Journal
Photonics Research
Published
2026-01-02
DOI
https://doi.org/10.1364/prj.582627
Primary Topic
Neural Networks and Reservoir Computing
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Non-volatile programmable directional couplers for high-integration and low-loss photonic computing

Qingjiang Li, Cen Liao, Hengyu Zhang, Xinmeng Hao et al.
Photonics Research
Neural Networks and Reservoir Computing
article

Non-volatile programmable directional couplers for high-integration and low-loss photonic computing

Qingjiang Li, Cen Liao, Hengyu Zhang, Xinmeng Hao, Yuan Tian, Hui Xu, Bing Song, Li Xuan, Li Feixiang, Ganlin Xiong
article en

Abstract

The rapid growth of artificial intelligence creates a pressing need for high-speed, low-cost photonic computing technologies, which remain hindered by integration density, optical loss, and computational precision. Here, we demonstrate a non-volatile programmable directional coupler based on low-loss Sb 2 Se 3 phase-change material, featuring an insertion loss of 0.3 dB. The device achieved 197 distinct non-volatile optical states within a coupling length of only 28 μm. This high-precision control was enabled by a novel programming method leveraging a dual-band fiber array. Functioning as a matrix multiplication unit, this device facilitated the photonic neural network to attain an accuracy of 92.27% on the Fashion-MNIST dataset, paving the way for high-density, high-precision, and energy-efficient photonic computing.

Photonics ResearchVol. 14(4)
Affordable and clean energy
Openalex Percentile: Top 97%
Neural Networks and Reservoir Computing
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Non-volatile programmable directional couplers for high-integration and low-loss photonic computing — Qingjiang Li, Cen Liao, et al. · Photonics Research (2026) | TGRS Research Map | TGRS