POS4-1098
Defect-Engineered Chiral Plasmonic Nanoparticle Assemblies for Multichannel Optical Physical Unclonable Functions
Topic
S4. Colloids, Interfaces, and Molecular Assemblies for Functional Soft Materials
When and Where
Oct 1, 2026
08:30 - 09:30
Room 301 (Grand Ballroom)
Session Chairs
Jinkee HONG
Ki Su KIM
Presenter(s)
Siwon Lee (Sungkyunkwan university)
Co-Author(s)
Abstract
With the explosive growth of IoT devices, authenticating the unique and unclonable identity of hardware has become a key challenge. However, conventional optical physical unclonable functions (PUFs) remain limited because the structural origin of their randomness is unverified, defect statistics are rarely quantified, and challenge–response diversity is constrained by single-channel readout. Here, the two-dimensional integration of Helicoid III (H3) chiral nanoparticles (NPs) was induced through controlled process conditions to generate differences in the integration tendency, exploiting the resulting defects not as targets for elimination but as a source of quantitatively controllable physical randomness. The defect fraction was quantified through a statistical analysis of the FE-SEM images, verifying the reproducibility within a narrow margin of error. Depending on the defect fraction, the chiral optical properties at 550 and 650 nm exhibited wavelength-dependent opposite trends, the origin of which was elucidated by cross-analysis integrating Transmission Electron Microscopy (TEM) coupled with Electron Energy-Loss Spectroscopy (EELS) measurements and Finite Element Method (FEM) simulations. Using a self-constructed Circular Dichroism (CD) optical setup, the wavelength-dependent polarization response was converted into pixel-wise images for PUF analysis. At both wavelengths, the inter-Hamming distance (HD) and intra-HD approached their ideal values (~0.499 and ~0.02), and the CRP-HD reached ~0.499, verifying both the enhanced encoding capacity and high PUF performance robust against AI-based modeling attacks. By quantitatively defining disorder, establishing reproducibility through sample statistics, and deriving evaluation criteria from intrinsic material properties, this study opens a path toward scalable, low-power, miniaturized physical-layer authentication materials.
