General information

The rapidly expanding field of artificial intelligence (AI) and machine learning exposes the limitations of conventional Von Neumann architecture. Neuromorphic computing, inspired by the highly energy efficient functionning of the brain, has emerged as a solution for efficient unsupervised learning, particularly relevant qith the advent of edge computing and visual computing applications. Solar cell-inspired materials, offering persistent photoconductivity allowing to simulate synaptic plasticity, have the potential to revolutionise neuromorphic visual computing. The SOLIS project forms an international consortium of experts from photovoltaics and materials science to explore inorganic thin film materials' potential as artificial visual synapses. These materials, tuneable and stable, promise reliable optically controlled MEMRISTORS suitable for diverse light intensities and wavelenghts. The collaboration, emphasising staff exchanges and transparent sharing of data, methods and persons, aims to reinforce our understanding, innovate with materials like 2D MXenes, and establish a shared framework for optoelectronic characterisation of visual synapses. The project aligns with Europe's objective to catch up in the field of AI and possibly become a leader in hardware-level machine learning, offering opportunities for scientists from Third Countries and EU countries alike. SOLIS will be an important milestone for EU research and the PV field as a whole, unlocking new applications for inorganic thin film materials and offering a paradigm shift toward visual computing and AI free from the constraints of current materials and architectures.

Sun sensors have many applications including, for instance, the orientation of satellites. These sensors determine the direction of light based on geometrical principles under a ray approximation of light. This usually involve some sort of light projection or non-planar structures, which can limit accuracy and reliability, and require precise alignment. In this project we aim to develop a new concept of light angle detection based on the wave nature of light. In particular, the sensor will be based on the angle-dependence of the transmittance spectrum through a layer designed to interfere with light (e.g. rugate filter, grating or a 2D metasurface, among others). In order to obtain a compact sensor design, the transmission spectra will be sampled at different spectral regions through a small matrix of photodetectors and absorptive filters (see the figure as an example of one ‘pixel’ detector). The angle of light will be determined from the relative intensity detected by the individual pixels of the matrix. Notice that the structure would be somewhat similar to a CCD image sensor but with a just few large pixels and the inclusion of the interference layer, which might also differ between "pixels". Also notice that, here, pixels do not play an imaging role. And some implementations might require as few as two "pixels". We envisage that a sensor based in this principle would present several advantages over the standard approaches: (1) The sensor will work on the full hemisphere, (2) superior sensitivity with peak sensitivity outside the normal incidence direction, (3) better performance in the prescience of diffuse or stray light, and (4) fullyintegrated MOS-compatible on-chip solution. However, further investigation, both theoretically and experimentally, needs to be performed in order to demonstrate whether all these potential advantages can be realized.

The MENTOR research initiative will provide a comprehensive and versatile technical platform for the development of next-generation indoor photovoltaics (IPVs) that efficiently re-use energy from artificial illumination to power electronics, eventually contributing to an energy- and climate-neutral future. MENTOR aims to unlock the full potential of IPVs taking into consideration growing concerns about sustainability, through the establishment of the first international network of 8 universities, 7 industrial partners, and 5 research centers. The consortium will cover all the key aspects and technologies related to IPVs, including sustainable design, organic and inorganic materials synthesis, photovoltaics manufacturing and characterization, device physics and modelling, theoretical and machine learning-driven approaches, photovoltaics recycling, and industrial processing. MENTOR will establish an interdisciplinary, intersectoral, and global program of doctoral training and research that propels the development of new leaders capable of directing academic and industrial R&D on renewable energy, electronics, and sustainability through the successful implementation of 16 doctoral candidate (DC) individual projects. This research initiative will amplify the recently recognized importance of IPVs for the sustainable powering of the IoT by advancing novel material designs, processing methods, device architectures, theoretical models, and characterization standards across disciplines and sectors.