Kim Sang-uk (left), a professor at KAIST, and researcher Yang Geon-guk, who led the study. [KAIST]
Kim Sang-uk (left), a professor at KAIST, and researcher Yang Geon-guk, who led the study. [KAIST]

South Korean researchers have developed a technology that can instantly verify whether an object is counterfeit simply by shining light on it.

KAIST announced Wednesday that a research team led by Professor Kim Sang-uk of the Department of Materials Science and Engineering, working jointly with a team led by Professor Kwon Seok-jun at Sungkyunkwan University, has developed a new foundational security technology. The technology uses randomly generated "colloidal nano patterns" — unique microscopic designs formed by clusters of particles too small to see with the naked eye — to authenticate objects using nothing more than ordinary light.

As AI advances make cyberattacks increasingly sophisticated, and as future quantum computers threaten to crack conventional encryption, technologies that use the unique physical characteristics of products and devices as a security measure are drawing growing attention.

The leading example of this approach is Physical Unclonable Function, or PUF.

Just as no two human fingerprints are alike, the arrangement of tiny particles that form spontaneously differs every time. Even when the same materials and methods are used, it is extremely difficult to reproduce the exact position and orientation of each particle. The researchers exploited precisely these differences as an "artificial fingerprint" to verify the authenticity of products and devices.

Existing high-security PUF systems, however, have often required expensive microscopes or specialized equipment capable of precisely measuring the properties of light in order to read such minute differences.

To overcome this limitation, the team focused on making the fingerprint hard to copy but easy to read.

Spherical particles measuring hundreds of nanometers — billionths of a meter — were allowed to self-assemble on a water surface. Because the position and orientation of the particles vary each time, a distinct pattern forms naturally with every iteration.

Reading this "fingerprint" is remarkably simple.

A schematic diagram of the research on hard-to-replicate 'artificial fingerprints' (AI-generated image). [KAIST]
A schematic diagram of the research on hard-to-replicate 'artificial fingerprints' (AI-generated image). [KAIST]

When ordinary light — such as a smartphone flashlight — is shone on the surface, the arrangement of the nanoparticles produces a unique combination of colors and reflection patterns. Shining a laser pointer causes the light to scatter in multiple directions as it interacts with the fine particle structure, generating yet another distinct light pattern.

The team pre-registered the two light patterns produced by each product and verified authenticity by comparing them against the patterns that appear on the actual item — in effect, reading a single "nano fingerprint" using two different light sources: a flashlight and a laser.

To forge such a label, a counterfeiter would need to replicate not only the nanoparticle structure itself but also the precise color and reflection pattern produced by a flashlight and the light pattern produced by a laser, making duplication extremely difficult.

The team also succeeded in transferring the nano structure onto a variety of surfaces, including flexible plastic, metal, transparent film and hydrogel, a soft gel-like material capable of absorbing large amounts of water.

"The key achievement of this research is that we created a random structure that is hard to copy while also making it easy to verify authenticity with simple tools like a flashlight or laser pointer," Kim said. "We expect this to develop into a next-generation security technology that can be readily used in everyday settings, from electronic device authentication to anti-counterfeiting labels."

The findings were published in the international journal Nature Communications.


nbgkoo@heraldcorp.com