Related Experiment Video
Updated: Mar 27, 2026

09:33
Demonstration of Equal-Intensity Beam Generation by Dielectric Metasurfaces
Published on: June 7, 2019
6.8K
Pixelated electrically reconfigurable metasurfaces for intelligent thermal emission control
Xiu Liu1, Hyeonggyun Kim1, Zexiao Wang1
1Department of Mechanical Engineering, Carnegie Mellon University, Pittsburgh, PA 15213, USA.
Science Advances
|March 25, 2026
Summary
We developed programmable metasurfaces using GeTe materials for dynamic control of thermal-infrared emission. This breakthrough enables faster, more efficient optical data processing for intelligent machines.
Area of Science:
- Optoelectronics
- Materials Science
- Nanotechnology
Background:
- Intelligent machines require advanced hardware for high-density data collection and processing.
- Thermal-infrared emission offers potential for optical data acquisition but faces limitations in speed and control.
- Existing thermal emission technologies are hindered by stochastic nature and slow response times.
Purpose of the Study:
- To demonstrate electrically programmable, pixelated metasurfaces for dynamic control of thermal-infrared emission.
- To overcome the limitations of conventional thermal emission for practical applications in intelligent hardware.
- To establish a versatile platform for reconfigurable photonic systems.
Main Methods:
- Integration of Germanium Telluride (GeTe) phase-change materials into hybrid plasmonic meta-atoms.
- Utilizing strong field confinement for enhanced optical properties.
- Electrically programming pixelated metasurfaces for localized control of thermal emission.
Main Results:
- Achieved fast, nonvolatile switching of thermal-infrared emission with large optical contrast.
- Demonstrated dynamic and localized control over thermal emission.
- Minimized active material usage while maintaining high performance.
Conclusions:
- The developed GeTe-based metasurfaces provide a versatile platform for reconfigurable photonic systems.
- This technology enables high integration density, adaptive functionality, and embedded intelligence for future hardware.
- The approach overcomes key limitations of thermal emission, paving the way for advanced optical data processing.

