Optical nano artifact metrics using silicon random nanostructures

Nano artifact metrics exploit unique physical attributes of nanostructured matter for authentication and clone resistance, which is vitally important in the age of Internet-of-Things where securing identities is critical. However, high-cost and huge experimental apparatuses, such as scanning electron microscopy, have been required in the former studies. Herein, we demonstrate an optical approach to characterise the nanoscale-precision signatures of silicon random structures towards realising low-cost and high-value information security technology. Unique and versatile silicon nanostructures are generated via resist collapse phenomena, which contains dimensions that are well below the diffraction limit of light. We exploit the nanoscale precision ability of confocal laser microscopy in the height dimension, and our experimental results demonstrate that the vertical precision of measurement is essential in satisfying the performances required for artifact metrics. Furthermore, by using state-of-the-art nanostructuring technology, we experimentally fabricate clones from the genuine devices. We demonstrate that the statistical properties of the genuine and clone devices are successfully exploited, showing that the liveness-detection-type approach, which is widely deployed in biometrics, is valid in artificially-constructed solid-state nanostructures. These findings pave the way for reasonable and yet sufficiently secure novel principles for information security based on silicon random nanostructures and optical technologies.

💡 Research Summary

The paper presents a low‑cost, vacuum‑free approach to nano‑artifact metrics by exploiting the three‑dimensional height information of silicon random nanostructures measured with a conventional confocal laser microscope (CLM). The authors first generate unique nanoscale patterns through the intentional collapse of e‑beam lithography resist pillars. Each pillar is 100 nm × 100 nm × 100 nm and arranged on a 200 nm square lattice over a 20 µm × 20 µm area; rinsing induces stochastic collapse, producing structures whose lateral dimensions are far below the diffraction limit. One hundred such devices are fabricated on a single wafer.

For authentication, the CLM (Olympus LEXT OLS400) scans a central 126 × 126‑pixel region (pixel size ≈125 nm) and records the height map Aij. This map becomes the device’s template. During verification, a newly measured map Bij is compared to the stored template using the Pearson correlation coefficient R, after allowing a shift of up to three pixels in each direction; the maximum R across shifts quantifies similarity. By varying the decision threshold, the false‑match rate (FMR) and false‑non‑match rate (FNMR) are evaluated. The results show that with a threshold between 0.1 and 0.5, FMR stays below 0.1 while FNMR remains under 0.2, indicating strong discriminability.

Crucially, the authors demonstrate that the vertical (height) resolution of the measurement determines security performance. When the height data are artificially rounded to 1 nm, 10 nm, and 100 nm precision, the 1 nm and 10 nm cases retain the original FMR/FNMR curves, whereas the 100 nm case suffers a dramatic increase in FNMR. This confirms that nanometer‑scale height variations are essential for reliable authentication.

To assess clone resistance, five counterfeit devices are fabricated based on the genuine height data. The authors binarize the height maps and reproduce them using e‑beam lithography, confirming the binary surface profile with atomic‑force microscopy (AFM). Optical CLM images of the clones, however, display continuous height variations due to diffraction and scattering. Statistical analysis of the height histograms reveals that genuine devices follow an almost Gaussian distribution, while clones exhibit skewed distributions. Computing Pearson correlation between the histograms of genuine–genuine pairs and genuine–clone pairs yields two well‑separated clusters (R < 0.2 for genuine pairs, R > 0.27 for genuine‑clone pairs). Setting the decision threshold between these clusters enables reliable detection of cloned devices, analogous to “liveness detection” used in biometric systems.

Overall, the study validates that a conventional confocal microscope can provide the nanometer‑scale vertical precision required for secure nano‑artifact metrics without the need for expensive SEM equipment. The approach offers a practical path toward high‑volume, low‑cost authentication for IoT devices, smart cards, and other security‑critical applications. Future work suggested includes scaling the fabrication process, integrating real‑time verification pipelines, and extending the method to other material platforms.