Hydrodynamically-Modulated Dewetting of Colloidal Solution for Physical Unclonable Functions

Traditional methods of blocking counterfeiting are prone to detection by potential attackers. In response, Physical Unclonable Function (PUF) has emerged as an alternative security measure. PUF is a specialized function that leverages the inherent unpredictability of physical processes to yield unique responses to specific challenges. PUF offers hardware-based security, mitigating vulnerabilities inherent in software-based systems susceptible to hacking and signal jamming. Recent efforts have aimed to develop optical PUF labels using various randomness sources. These PUF's non-deterministic fabrication process ensures a substantial encoding capacity for robust data security. However, excessively disorderly PUF codes disseminated across the information space might exhibit a deficiency in possessing a distinguishing fingerprint key. Using random optical features for security keys mandates statistical image processing for digitization and key generation. Unfortunately, the processing of analog signals to digitized codes may compromise authentication accuracy resulting from an arbitrary threshold during an image analysis.<br/>We tackle this by utilizing the geometric multi-bit patterning technique, manipulating dewetting of colloidal solution for advanced optical PUFs that exhibit both high randomness and determinism. By placing agglomerated dots of colloidal particles at a specific vertex within geometrically confined polygons, these two contrasting features can be provided to the PUF system through the hydrodynamic modulation of the dots as visual cues. Exploiting binary solvent flow unpredictability in confined spaces during dewetting, we fabricate sophisticated PUF labels in periodic pixel arrays. The fundamental principle stems from instability arising from imbalanced solutal-Marangoni flows in evaporating polygonal droplets. Importantly, the dynamic coating procedures have the capability to induce unforeseeable fluctuations in contact angles at every vertex during each coating iteration. These fluctuations perturbed the radial symmetry of the solutal-Marangoni flows within polygonal-shaped droplet, introducing randomness to the final PUF pattern—a phenomenon supported by statistical parameters. This pattern achieves an entropy of 0.977, a uniqueness value of 0.759, and a correlation coefficient of 0.009, all of which are close to the ideal values. It successfully passes the statistical test for random number generators conducted by NIST (National Institute of Standards and Technology), confirming its true randomness, unpredictability, and unclonability. These random PUF labels with quantized signals at deterministic sites enable efficient, accurate, and swift authentication in both encoding and decoding. They also offer reconfigurability, transferability to diverse surfaces, and the potential for enhanced security through fluorescent molecule dyeing.