By Gary Elinoff, contributing writer

No matter how efficient electronic circuitry may be, active electronics must always consume some power. With the advent of the IoT, there will be more instances of sensors residing in remote or dangerous areas where maintenance, including battery charging and replacement, will be difficult or impossible. Supplying power to these sensors, perhaps for years at a time, will present problems. To get past this issue, a new type of sensor based on plasmonic nanostructures has been proposed that doesn’t require actively powered electronics at all. In fact, the presence of the condition that the sensor is on alert for is what, in itself, turns the sensor on.

Plasmonic nanostructures

Plasmonic nanostructures are nano-sized structures that react to a very specific frequency, just like the way a radio tuned to 100 MHz only picks up signals broadcast at the frequency. The sensors were developed for DARPA by a team from Northeastern University headed by Professor Matteo Rinaldi using this principle to react to a distinct frequency — wavelength — of infrared light.

As described by Troy Olsson, manager of DARPA’s N-ZERO program, the sensors will absorb the infrared energy, turn it into heat, and the heat will cause part of the structure to bend. The bending causes that part of the structure to impact another part, effecting an electrical connection. The sensor, which might have lain dormant for months or even years, comes alive, and only then will the circuitry begin to draw power. Rinaldi and his team describe the device as a “plasmonically enhanced micromechanical photoswitch.”

A “plasmonically enhanced micromechanical photoswitch.” Image source: DARPA.

Looking for specific profiles

This frequency-specific behavior means that the sensors can be tuned to identify specific IR sources. Gasses present in vehicle emissions, such as CO, CO₂ , or even water, generate infrared profiles, as do the oxides of nitrogen, NO_X , and sulfur, SO_X . The same is true for the gasses generated by gunshots, fires, and explosions. Olsson explains that a sensor can be constructed to look for any one of those individual gas’s infrared profiles.

But the interesting fact is that each type of event generates its own “spectrum” of gas emissions. As one might imagine, then, the outputs of sensors tuned for different gasses can be “wired” together in the manner of digital logic to react only to a specific type of event. If a fire exhibits two or three gasses, all three will be required to trip the sensor. Similarly, gunshots and explosions, with their own unique gas emission profiles, can be targeted with another set of sensors tuned for each relevant gaseous component. Thus, although originally developed for military purposes, these sensors will have widespread commercial applications, too.