Development of a Photonic Switch via Electro‐Capillarity‐Induced Water Penetration Across a 10‐nm Gap

Background

Narrow, densely packed photonic nanostructures confine near-field resonance, but nanogeometries with sub-100 nm roughness are strongly water repellent, so water-based molecular analysis on photonic surfaces first requires complete wetting of the optical hotspots. How and when water wets these hotspots has been difficult to monitor in situ. Contact-angle measurements can give similar values for different wetting configurations, electron microscopy is difficult to apply to dynamic kinetics, and optical microscopy is limited at the nanoscale.

Design and approach

The FPR is a metal–dielectric–metal tri-layer: a hydrophobic perfluoro-(butenyl vinyl ether) (PBVE) spacer on a 120-nm gold (Au) mirror, capped by a 30-nm Au film that self-assembles into interconnected nanoparticles separated by nanogaps. Oxygen plasma etching through these gaps removes part of the PBVE and leaves nanopillars and air voids, which lowers the effective refractive index of the spacer and blue-shifts the resonance. The Results describe a 150-nm PBVE spacer, whereas the Experimental Section gives 164 nm. When a voltage is applied across a water droplet, charge accumulation in the top Au layer lowers the solid–liquid surface energy and the Cassie-to-Wenzel energy barrier, so water replaces the air in the spacer and the color changes.

How it was evaluated

Four FPRs etched for 0, 20, 40 and 60 s (FPRs I to IV) were characterized by reflectance spectra, simulations based on Bruggeman's effective medium theory, and scanning electron microscopy of 150 gap distances. Electrowetting was measured with 8-µL water droplets under increasing DC voltage, and a surface-energy model was used to calculate the transition voltage for FPR IV. Time-resolved reflectance was recorded every 0.2 s with water sealed against indium-tin-oxide (ITO) glass. The authors then tested optical encryption and decryption, direct writing with a needle electrode held 100 µm above the surface, and colorimetric detection of 0, 10 and 20 wt.% glucose solutions.

Key results

Etching for 0, 20, 40 and 60 s blue-shifted the resonance from 639 nm to 613, 578 and 551 nm, giving light green, green, blue and purple colors. Matching measurements and simulations gave effective refractive indices of 1.34, 1.25, 1.13 and 1.02, corresponding to PBVE nanopillars occupying 100%, 77%, 43% and 8% of the spacer. The gaps in the top Au film followed a gamma distribution with a mean of 11.7 nm (N = 150).

Contact angles decreased with increasing voltage before leveling off, with larger changes on the more heavily etched FPRs. The surface-energy analysis predicted that the Cassie-to-Wenzel transition on FPR IV would begin at 4 V, and droplets on FPR IV changed color above 4 V. With water sealed on FPR IV, 4 V for 1 s turned the purple device green; the green state remained after the voltage was removed and persisted until the water in the spacer evaporated, which the authors attribute to a higher energy barrier for the reverse Wenzel-to-Cassie transition. The resonance shifted by 87 nm (544 to 631 nm) within 0.6 s, which simulations interpreted as an increase in effective index from 1.02 to 1.34 with the air fraction replaced by water (water fraction 0.92).

A 'KIST' pattern etched into the spacer appeared purple on a green background and disappeared after liquid permeation because PBVE and water share a refractive index of 1.34; a patterned ITO electrode instead revealed the message only where liquid entered beneath it (5 V in these experiments). In direct writing, dot diameters grew from 415 µm to 2.47 mm and line widths from 403 µm to 1.31 mm as voltage amplitude or duration increased or stage speed decreased.

With glucose solutions, the resonance shifted by 88 nm (545 to 633 nm) for pure water and by 95 and 101 nm for 10 and 20 wt.% glucose, from which refractive indices of 1.346 and 1.364 were retrieved, compared with literature values of 1.348 and 1.363. The color differences from the pure-water standard were ΔE* = 7.0 and 12.3, above the naked-eye threshold of ΔE* > 1, and the simulation slope gave a minimum visually resolvable concentration difference of 2.5 wt.%.

Open questions

The authors note that the minimum written feature size (about 400 µm) matched the outer diameter of the 27G needle electrode and that a sharper needle would be needed for finer resolution. They propose extending the approach to surface-enhanced Raman spectroscopy, surface-enhanced infrared absorption and terahertz sensing to add chemical selectivity beyond refractive-index sensing.

How it connects to other research

Lubricant-infused slippery surfaces such as SLIPS keep liquids from wetting a textured surface, whereas this FPR uses electrowetting to drive water into a hydrophobic nanostructure on demand and reads the resulting refractive-index change as color. The device builds on an earlier liquid-permeable FPR reported by Yu, Ryu and colleagues (ACS Appl. Mater. Interfaces 2018), whose refractive-index sensitivity of 323.5 nm RIU−1 the authors cite.