Cryptography and carbon: How nanotubes could improve security

Could carbon nanotubes provide intensely secure on-chip encryption for hardware?

Carbon nanotubes are being investigated as a means to create a tough, secure system for computer hardware.

Today's hardware security is made possible by various means, and silicon chips are one of the most popular methods. While silicon chips allow vendors to bring security right down to the stack, they are vulnerable to tampering, information leaks and destruction.

Silicon nanotubes, due to their resilience, conduction properties and size, have been touted as a possible replacement. However, the "purity" level of semiconductor activity can be hampered because of the reactions which result in the nanotubes, and due to their small form factor, arranging them can be difficult -- which, in turn, would undermine the use of silicon nanotubes in security as they could be considered flawed or unstable.

Not all is lost. According to a joint IBM and academic team, a process is available to use these challenges to their advantage to create rough, resilient hardware-based security for our devices.

In a study published online in the publication Nature Nanotechnology, the team says these "inherent imperfections" can be set as the foundation for a low-cost, "unclonable electric random structure" which enhances cryptographic security.

As noted by Ars Technica, the majority of digital cryptography models need the ability to generate a unique set of bits which acts as a key to unlock data. If linked to hardware, this key must be wired onto a chip itself, which again, can be tampered with.

The researchers say that a process which acts to completely randomize the placement of nanotubes could make tampering and cracking hardware-based security extremely difficult.

The process, based on tests with 64-bit hardware, involves carbon nanotubes which are dissolved in water, and a specialized detergent -- negatively charged -- delivers the nanotubes into a positively-charged area of a chip. By alternating the charges, this process can then alter the spacing and placement of nanotubes, which results in their random placement.

In turn, bits on the chip will be randomly conducting or non-conducting. These self-assembled nanotubes (CNTs) will thereby increase the cryptographic potential of a chip, which can be further improved by varying the types of nanotubes used, and whether they are metallic or conducting.

The random nature of the produced chips then enhances the cryptographic potential of the bits, which thereby could improve on-chip cryptographic protection.

"These results show that random bit generation using self-assembled CNTs is a promising approach for low-cost and hard to forge applications," the researchers say. "The addition of CNT random structures into the design toolkit will enable new operating principles for hardware security."

Read on: Top picks

Newsletters

You have been successfully signed up. To sign up for more newsletters or to manage your account, visit the Newsletter Subscription Center.
Subscription failed.
See All
See All