Investigating Gradient Monolayers for Neuromorphic Computing Base on Hybrid Oscillatory Devices
Supervisors: Dr. Saurabh Soni, Prof. dr. Jurriaan Huskens, Prof. dr. Christian A. Nijhuis
Biological neural architecture (such as our brains) provides inspiration towards building faster and more power efficient electronic devices. Soft materials, such as small organic molecules, can be incorporated inside solid-state devices to mimic the functionality of a neuron. To expand the tool-box of neuromorphic computing, new molecular hardware is needed. The goal of this project is to develop such a molecular device to realize long-sought after oscillatory neural networks.1 These devices could consist of two-dimensional assemblies of small molecules contacted with prefabricated metal electrodes. In this project, we aim towards fabrication of gradient monolayers,2,3 which could respond to electric fields providing the feedback to create an oscillatory network.
The student will aim towards finding a suitable system, such as responsive supported lipid bilayers (SLBs)4 functionalized with redox-active moieties, for forming surface gradients (Figure a). Further, the SLBs will be encapsulated in soft devices with PDMS (a transparent rubber) microfluidic channels for forming electrical contacts (when filled with a liquid metal alloy, EGaIn5) and also chemical inputs to affect the molecular properties (an initially prototype design is sketched in Figure b). The dynamic electronic properties of these devices will be measured by applying voltage across the SLBs and measuring their capacitance or current output, modulated by external inputs.
During the project, the student will be exposed to some or all of the following techniques and/or instrumentations:
· Formation of SLBs via self-assembly and their surface characterization
· Fabricating PDMS-based soft microfluidic devices
· Learning impedance spectroscopy and current-voltage measurements
· Techniques such as template-stripping for ultraflat metal electrodes, plasma cleaning, etc.
References:
1. Soman et al., Front. Comput. Neurosci. 12:52. doi: 10.3389/fncom.2018.00052
2. Nicosia et al., J. Mater. Chem. B, 2013, 1, 5417
3. Krabbenborg et al., Chem. Eur. J. 2015, 21, 9638 – 9644
4. N. J. Overeem et al., ChemistrySelect 2020, 5, 9799.
5. Chen et al., ACS Appl. Mater. Interfaces 2019, 11 (23), 21018-21029