2023-12-021 - IONET - Solid-State Ionics Synaptic Transistors for Neuromorphic Computing
Contacts
MF
Acronim & Gínjol codes
IONET
2023-12-021
Main technology offer
New high-performing cost-effective and environmentally friendly TE material (Thermo Electrical). Nano-enabled easy-to-handle flexible fabrics made of thin-walled silicon microtubes
Ownership
Sant Adrià de Besòs, Spain
2008
Institut de Recerca en Energia de Catalunya (IREC)
Public Partners
Readiness Level
3-4 Experimental PoC /
5 Prototype /
6-7 MVP /
8 Industrialization /
9 Commercialization
MIN
1
2
3
4
5
6
7
8
9
10
MAX
MIN
1
2
3
4
5
6
7
8
9
10
MAX
MIN
1
2
3
4
5
6
7
8
9
10
MAX
Impact: ESG & SDG Goals
Market Data
Problem: There is a need for stable and Si-compatible synaptic transistors in Neuromorphic Computing. Synaptic EGTs already showed the potential for being a leading technology in Neuromorphic Computing, however, they are currently based on electrolytes that are intrinsically unstable and difficult to integrate such as ionic liquids or proton conducting polymers, which are obviously sensitive to humidity and temperature and present a poor compatibility with mainstream microelectronics fabrication processes. Alternatively, more convenient robust oxide-ion conductors forces an EGT operation at non-viable elevated temperatures (above 200ºC). The solution: a robust and reliable artificial neural network based on oxide synaptic transistors that will provide the required stability, low-energy consumption and viable operating temperature.
650000000
32500000
7300000
Fab-less like spin-off formation
There is a need for stable and Si-compatible synaptic transistors in Neuromorphic Computing.
Neuromorphic computing transistors
Funding
Currently, we need to develop a neural network based on our synaptic transistor to demonstrate the capability of the technology in a real environment.
Funds to develop the neural network
Development of a Neural network based on IONET synaptic transistors CMOS compatibility study Scale-up demonstration
Technology Status
IONET transistors are stable under temperature, humidity and time; operate in the whole range of standard temperatures for electronic components; IONET transistors present a very linear and symmetric analogic variation of the channel conductance; and they are compatible with mainstream microelectronics fabrication technology
CMOS plants are hesitant to introduce new materials like oxides into their clean rooms. The AI hardware field is rapidly growing, leading to a shortage of skilled workers. Breakthrough technology may scare some end-users looking for well-established products Breakthrough technology may scare some end-users looking for well-established products
IONET technology is protected with a PCT application (Ref: PCT/EP2022/066614) with a positive ISR evaluation. Entered national phases in EP, US, CN, KO. Communication of intention to grant the European patent (20/02/2025)
PCT application (Ref: PCT/EP2022/066614) with a positive ISR evaluation. Entered national phases in EP, US, CN, KO. Communication of intention to grant the European patent (20/02/2025)
Additional information
A robust and reliable artificial neural network for artificial intelligence applications based on oxide synaptic transistors. Technology can be integrated into the mainstream semiconductor manufacturing processes thanks to its easy scalability and compatibility with CMOS technology.