Duke Engineers Trap Light in a 10-Nanometer Gap to Build World's Fastest Thermal Detector — 125 Trillionths of a Second

Engineers at Duke University have built the world's fastest thermal photodetector — a device barely thicker than a soap bubble that can sense light across the entire electromagnetic spectrum and generate an electrical signal in just 125 picoseconds, or 125 trillionths of a second. The breakthrough, which runs at frequencies up to 2.8 gigahertz, is hundreds to thousands of times faster than any previous thermal detector ever built, operates at room temperature without any external power source, and could transform multispectral imaging across medicine, agriculture, national security, and space exploration. The research was published in the journal Advanced Functional Materials and funded by the Air Force Office of Scientific Research and the Gordon and Betty Moore Foundation.

The device was developed in the laboratory of Maiken Mikkelsen, professor of electrical and computer engineering at Duke's Pratt School of Engineering, and relies on a phenomenon called plasmonics that essentially traps light inside an impossibly thin material. The architecture uses precisely arranged silver nanocubes — each tens of nanometers across — positioned just 10 nanometers above a flat gold layer. When light strikes the surface of a nanocube, it excites the silver's electrons into a collective oscillation called a plasmon, which traps the light's energy in the narrow gap between the nanocube and the gold below. This captured energy heats an ultra-thin pyroelectric film sandwiched in the gap, which converts that temperature change into an electrical signal in a fraction of a nanosecond.

The device's record-breaking speed stems directly from its extraordinary thinness. Traditional pyroelectric detectors — the class of sensors that convert temperature changes into electricity — require bulk material to absorb enough light to generate a measurable signal. That bulk creates thermal inertia: the sensor heats and cools slowly, limiting how fast it can respond. By using plasmonics to concentrate light with extraordinary efficiency in a gap just 10 nanometers wide, the Duke team needed only an ultrathin pyroelectric layer, which heats and cools in femtoseconds. PhD student Eunso Shin, who led the experimental work, also developed a new, low-cost measurement technique using distributed feedback lasers to precisely verify the device's speed — a method that could itself accelerate the development of other ultrafast detectors across the field.

The spectrum coverage of the detector is as impressive as its speed. Because the plasmonic response can be tuned by adjusting the size and spacing of the nanocubes, the device can be engineered to absorb light at virtually any wavelength from ultraviolet through infrared — or across the entire spectrum simultaneously. This makes it fundamentally different from semiconductor photodetectors, which are each sensitive only to a narrow range of wavelengths and require different materials for different spectral ranges. A single Duke-style device could replace an entire array of conventional sensors, dramatically simplifying the hardware required for multispectral imaging systems.

The applications span a remarkable range of domains. In medicine, multispectral cameras built with the detector could identify cancerous tissue in real time during surgery by distinguishing the different infrared signatures of healthy and malignant cells. In precision agriculture, drones equipped with the device could map crop health, identify irrigation problems, and detect fungal disease across thousands of acres in a single pass without bulky cooling equipment. In national security, satellites could conduct broad-area thermal surveillance without the complex cryogenic cooling systems that current high-speed infrared detectors require. NASA has identified passive, room-temperature broadband detectors as a priority for future deep-space missions where power budgets are severely constrained. "We essentially broke the fundamental trade-off between sensitivity and speed that has constrained thermal detectors for decades," Mikkelsen said. The team said the device could be made even faster still by optimizing the geometry of the gap between the nanocubes and the gold layer.