Background and scope:

Power consumption and heat dissipation are the most urgent challenges in electronics ranging from mobile devices to large data centres and becomes especially relevant for smart edge devices. Advanced chip designs are lowering energy consumption of microelectronic components, devices and systems, while increasing performance such as speed, capacity, reliability and security. Applications include artificial intelligence, communications, computing and sensing.

Various strategies are and have been tested, but still there is much room to improve energy consumption towards near-fundamental limits, through the co-design of geometry, materials, circuits, and integration in a holistic approach.

The overall goal of this challenge is to explore novel materials and beyond CMOS devices, non-von Neumann architectures and alternative information processing paradigms to drastically reduce energy consumption in order to meet application-specific needs of smart edge devices and circuits.

Specific objectives:

The overall objective of this challenge is to explore solutions (starting at TRL 1/2) that will have a drastic impact on decreasing the power consumption of any smart edge device, but specially for Edge Processing and memories, Edge Sensing and Imaging, Edge Communication and Edge Power Management. The proposed solutions should start at TRL 1-2 and reach TRL 3-4.

The projects, supported under this Challenge are expected to address one or more of the following aspects:

• fundamental issues like heat dissipation at nanoscale that has turned out to be the most critical bottleneck in information processing covering the design of a device from the understanding of the physics and the nanoscale thermal transport at component level to circumvent the “heat valley”, selecting the materials and process solutions.
• demonstration of the potential of the developed technologies for energy savings and contained environmental footprint towards responsible smart edge devices.

The proposed developments may cover (amongst others):

At Design level.

• Computer modelling based on the fundamental understanding of heat transport across layers and interfaces, harvesting fluctuations instead of fighting them for computing or the use of different state variables, e.g., spins, photons, phonons or mechanical switches, instead of charge.
• Analysis of the dissipation mechanisms in signal transmission and conversion, heat removal from hot spots in components and circuits, potential for energy conversion at the nano-scale, etc.

At Materials/Process levels:

• Novel or unique electrical, mechanical and optical interconnections or other switching mechanisms
• Efficient heat dissipation new materials for in-chip heat dissipation, e.g., 2D materials
• Embedding energy harvesters in the final devices and/or circuits
• Effective 3D multi-die heterogeneous integration including advanced packaging, heterogeneous integration, and modular design of components (such as chiplets)

At Device/Architecture levels:

• Molecular electronic circuits
• Beyond CMOS. Non mainstream semiconductor transistors including a plausible circuit concept, e.g., single electron transistors.
• Novel non-von Neumman architectures and alternative processing approaches

Expected outcomes and impacts:

The portfolio of projects selected under this Challenge is expected to collectively:

• derive fundamental bounds for energy consumption and designing practical and basic scenarios to minimize the energy costs of the different processes
• harness energetic efficiencies as optimization tools to operate smart technological choices to build smart edge devices.

The expected impact from this Challenge is to open an unprecedented way for the reduction of power consumption in information processing, transmitting, etc. by developing new fundamental technology solutions going from advanced materials to advanced devices and circuits, that holistically will allow a drastic reduction of energy consumption of smart edge solutions.

Specific conditions

The applicants must describe in their proposal energy-based metrics for the technologies and methodologies to measure them and establish benchmarks.

Stable, abundant and non-toxic materials which withstand device and circuit processing steps should be used.