- Materials nanostructural design
- Materials evolutionary design
- Strain engineering
- Materials engineering
- Active photonics
- Phase Change Memory
Apply for PhD position here
PhD Project by Li Lu
Tuneable hyperbolic metamaterials
Hyperbolic metamaterials (HMMs) exhibit a hyperbolic dispersion relation, which is rarely observed in natural materials. Therefore, HMMs are promising for many photonics applications such as sub-wavelength imaging, biosensing, and spontaneous emission enhancement etc. They can be formed by layering dielectric and metallic nano films. The aim of this project is to make tuneable HMMs.
Tuneable nanoantenna arrays metasurfaces
Metasurfaces made of dielectric materials can have similar properties to plasmonic structures and can be used manipulate light, with the advantage of having lower loss. Here we applied PCM to design all-dielectric Huygens’ metasurfaces. Exploitation the refractive index change of PCM, we can dynamically control the light in the visible and N-IR spectrum range. Essentially, the phase of light through each nanoresonator can be tuned by the structural phase transition of PCM. This can allow us to make beam steering devices.
Reconfigurable InP waveguide components
Photonics integrated circuits (PIC) based on the InP platform are technologically important because they can manipulate, filter, generate, and amplify light. It is, therefore, an attractive platform for telecommunication. Reconfiguring the path of light in PIC is important because it opens the possibility to bespoke reprogrammable optical chips. PCMs can have large non-volatile optical contrast between amorphous and crystalline states, and therefore we applied PCM into on-chip waveguide components. By switching the structural state of PCM, we can change the effective index of the waveguide mode, and further achieve optical switching in different waveguide devices such as directional couplers, and Mach-Zehnder interferometer.
Project done by Li Tian
Antimony Trisuphide (Sb2S3) Hybrid Multi-Bit Memory Device
Non-volatile memory is getting more important in the development of the IC industry. The ability to retain data when power is turned off become a huge advantage when creating applications for computer and communication fields. As technology getting advanced, the demand for NVM has grown significantly as well. The need to store huge amounts of information and a high-density memory is increasing, this leads to development of new-generation information storage technology. Sb2S3 is a phase change material that can behave as a ReRAM. Ionic diffusion through Sb2S3 occurs when an electric voltage is applied and when it is heated above its crystallisation temperature, it can switch from amorphous state to crystalline state. To enable the ionic diffusion, an asymmetric device is fabricated. By using two different materials as top and bottom electrode, there will be difference in electrical potential that will create a field in between them. Thus, the charge particles will move in response to it. Combining this effect with phase change, it could enable a 2-bit memory cell. Therefore, by using a hybrid phase change and resistive memory concept, this is useful for high-density data storage applications. We believe that these multi-bit non-volatile memory devices will also be useful in solid-state artificial neural networks.
Projects done by Ning Jing:
Highly efficient switching phase change materials
This work aims to improve the switching properties of phase change materials. GST225 is the most applied phase change materials due to its rapid crystallization and good electrical and optical contrast between crystalline and amorphous states. The idea is to design an iPCMs superlattices with low switching energy. This project will involve both theoretical simulations and experiments
Superlattice growth of chalcogenides
The properties depend on the crystal structures. This work aims to improve the crystal quality of phase change materials, such as Sb2Te3 – GeTe superlattice. The design is to understand the mechanism and optimize the growth process using design of experiment, screen out the significant factor in growth and fabricate the well-textured layers for further application. This project will involve both theoretical simulations and experiments.
Project by Ting Yu
Hardware Neural Networks
With the advent of Artificial Intelligence and the Internet of Things, neural network accelerator chips are gaining traction to efficiently perform data-centric tasks. However current devices are made of CMOS transistor technologies, which have limited performance as we approach transistor scaling limits. Beyond-CMOS technologies are currently investigated to develop new hardware neural network (HNN) components. This includes two-terminal electrical devices and reprogrammable photonic devices. Our research work focus on the study and development of novel reconfigurable Chalcogenide materials for these beyond-CMOS devices. The materials developed will be mainly used in emerging non-volatile and volatile memory devices, and reconfigurable photonics.