Scientists at the National Institute of Standards and Technology (NIST) have developed a new device that measures the motion of super-tiny particles traversing distances almost unimaginably small—shorter than the diameter of a hydrogen atom, or less than one-millionth the width of a human hair. Not only can the handheld device sense the atomic-scale motion of its tiny parts with unprecedented precision, but the researchers have devised a method to mass produce the highly sensitive measuring tool.
It’s relatively easy to measure small movements of large objects but much more difficult when the moving parts are on the scale of nanometers, or billionths of a meter. The ability to accurately measure tiny displacements of microscopic bodies has applications in sensing trace amounts of hazardous biological or chemical agents, perfecting the movement of miniature robots, accurately deploying airbags and detecting extremely weak sound waves traveling through thin films.
The researchers measured subatomic-scale motion in a gold nanoparticle. They did this by engineering a small air gap, about 15 nanometers in width, between the gold nanoparticle and a gold sheet. This gap is so small that laser light cannot penetrate it.
However, the light energized surface plasmons—the collective, wave-like motion of groups of electrons confined to travel along the boundary between the gold surface and the air.
The researchers exploited the light’s wavelength, the distance between successive peaks of the light wave. With the right choice of wavelength, or equivalently, its frequency, the laser light causes plasmons of a particular frequency to oscillate back and forth, or resonate, along the gap, like the reverberations of a plucked guitar string. Meanwhile, as the nanoparticle moves, it changes the width of the gap and, like tuning a guitar string, changes the frequency at which the plasmons resonate.
Schematic shows laser light interacting with a plasmonic gap resonator, a miniature device designed at NIST to measure with unprecedented precision the nanoscale motions of nanoparticles. An incident laser beam (pink beam at left) strikes the resonator, which consists of two layers of gold separated by an air gap. The top gold layer is embedded in an array of tiny cantilevers (violet)—vibrating devices resembling a miniature diving board. When a cantilever moves, it changes the width of the air gap, which, in turn, changes the intensity of the laser light reflected from the resonator. The modulation of the light reveals the displacement of the tiny cantilever. Credit: Brian Roxworthy, NIST/CNST
These optical micrographs provide a top-down view of several plasmonic gap resonators and zoom in on a single device. Bottom right shows schematic of a single device. Credit: Brian Roxworthy, NIST/CNST
The interaction between the laser light and the plasmons is critical for sensing tiny displacements from nanoscale particles, notes Aksyuk. Light can’t easily detect the location or motion of an object smaller than the wavelength of the laser, but converting the light to plasmons overcomes this limitation. Because the plasmons are confined to the tiny gap, they are more sensitive than light is for sensing the motion of small objects like the gold nanoparticle.
The amount of laser light reflected back from the plasmon device reveals the width of the gap and the motion of the nanoparticle. Suppose, for example, that the gap changes—due to the motion of the nanoparticle—in such a way that the natural frequency, or resonance, of the plasmons more closely matches the frequency of the laser light. In that case, the plasmons are able to absorb more energy from the laser light, and less light is reflected.
To use this motion-sensing technique in a practical device, Aksyuk and Roxworthy embedded the gold nanoparticle in a microscopic-scale mechanical structure—a vibrating cantilever, sort of a miniature diving board—that was a few micrometers long, made of silicon nitride. Even when they’re not set in motion, such devices never sit perfectly still, but vibrate at high frequency, jostled by the random motion of their molecules at room temperature. Even though the amplitude of the vibration was tiny—moving subatomic distances—it was easy to detect with the new plasmonic technique. Similar, though typically larger, mechanical structures are commonly used for both scientific measurements and practical sensors; for example, detecting motion and orientation in cars and smartphones. The NIST scientists hope their new way of measuring motion at the nanoscale will help to further miniaturize and improve performance of many such micromechanical systems.
“This architecture paves the way for advances in nanomechanical sensing,” the researchers write. “We can detect tiny motion more locally and precisely with these plasmonic resonators than any other way of doing it,” said Aksyuk.
The team’s fabrication approach allows production of some 25,000 of the devices on a computer chip, with each device tailored to detect motion according to the needs of the manufacturer.
The researchers have introduced a new class of plasmonic resonators embedded into nanomechanical devices, and demonstrated them by measuring motion of sub-picogram cantilevers with an unprecedented combination of high sensitivity and small footprint. The low levels of input-referred mechanical motion noise in our motion measurements result from the unique features of our LGP resonators, namely the combination of extremely large optomechanical coupling strength, high plasmonic quality factor, and the large optical cross section of the LGP resonators, which increase the modulation of the reflected signal by small motion. The broadband nature of the LGP mode (full-width half maximum of 35 nm for both devices) can benefit chip-scale motion detection applications. Specifically, motion can be transduced from multiple devices, including device arrays, simultaneously using a single laser wavelength, despite small fabrication variations that produce device-to-device variation in λLGP. Moreover, inexpensive incoherent illumination sources, such as light-emitting diodes with spectral bandwidth filtered down to 10 nm or more, can potentially be used for these measurements in addition to lasers.
Beyond motion measurement, the dynamic LGP modes are attractive for future fundamental studies of nonlinear optomechanical coupling. The pNEMS architecture admits arbitrary planar resonator and mechanical structure shapes, while precisely defining the large-aspect-ratio, nanoscale gap via a thin sacrificial layer. Therefore, the smallest lateral features (165 nm in this work) can be straightforwardly patterned using 193 nm photolithography for wafer-scale batch fabrication. This approach to creating plasmomechanical systems is backend-compatible with complementary metal oxide semiconductor (CMOS) processing. enabling future monolithic integration with high-speed multi-channel control circuits. Moreover, owing to large and gOM, only about 10 nm of electrostatic or thermal actuation is sufficient to shift the plasmonic resonance by one linewidth. Thus, pNEMS may enable not only large nanomechanical sensing assays, but also rapidly reconfigurable photonic devices, providing a path to realizing large-scale, randomly accessible photonic metamaterials.
Plasmonic structures couple oscillating electromagnetic fields to conduction electrons in noble metals and thereby can confine optical-frequency excitations at nanometre scales. This confinement both facilitates miniaturization of nanophotonic devices and makes their response highly sensitive to mechanical motion. Mechanically coupled plasmonic devices thus hold great promise as building blocks for next-generation reconfigurable optics and metasurfaces. However, a flexible approach for accurately batch-fabricating high-performance plasmomechanical devices is currently lacking. Here we introduce an architecture integrating individual plasmonic structures with precise, nanometre features into tunable mechanical resonators. The localized gap plasmon resonators strongly couple light and mechanical motion within a three-dimensional, sub-diffraction volume, yielding large quality factors and record optomechanical coupling strength of 2 THz·nm−1. Utilizing these features, we demonstrate sensitive and spatially localized optical transduction of mechanical motion with a noise floor of 6 fm·Hz− 1/ 2, representing a 1.5 orders of magnitude improvement over existing localized plasmomechanical systems.