# Synthesis of a Semiconducting 2D Material for Novel Strong PUFs

## TL;DR

This paper reports on the synthesis of tungsten disulfide, a 2D semiconducting material, for creating novel physical unclonable functions (PUFs) to enhance device security in IoT applications.

## Contribution

It introduces a new approach to fabricating PUFs using chemically vapor-deposited WS2, demonstrating large-scale synthesis and initial patterning for security applications.

## Key findings

- Successful synthesis of WS2 via CVD with optimized parameters
- Formation of a large-scale WS2 surface suitable for PUFs
- Initial patterning and contact formation on WS2 surface

## Abstract

The exponentially growing number of interconnected devices in the Internet of Things poses an increasing amount of challenges to the field of cyber security and encryption. For authenticated use and communication, each device must securely store and apply secret keys without giving a potential attacker the possibility to access them neither by software nor hardware attacks. In 2002, the concept of Physical Unclonable Function emerged to counteract the inherent threats, associated with this trend. Physical disordered systems are excited by external stimuli and can thereby be uniquely identified. 2D materials have profoundly different characteristics when isolated from their bulk form to few- or monolayers. Transition metal dichalcogenides represent a potential PUF candidate since, unlike graphene, they feature a bandgap and are thus semiconducting. Furthermore, a random percolation network can be formed from them, which is expected to possess an electrical response that is uniquely identifiable yet unpredictable. This work has built the foundation for implementing a PUF consisting of a TMD, namely tungsten disulfide. The material has been synthesized using chemical vapor deposition in the close proximity approach. In order to fabricate a large-scale inter-connected surface consisting of WS2, the growth parameters are selected and enclosed by a preliminary trial and error optimization and ultimately optimized by Design of Experiments. This yields a potentially available PUF area of several millimeters, which was then patterned and contacted with initial trials.