From Molecules to Machines: A Decade-Long Vision Realized

A landmark international collaboration led by Newcastle University has developed the world’s most efficient integrated light-harvesting and storage system for powering autonomous Artificial Intelligence (AI) at the edge of the Internet of Things (IoT).

The technology, described in the journal Energy & Environmental Science (Royal Society of Chemistry), pioneers a battery-free, maintenance-free platform for next-generation smart sensors and devices—heralding a transformative shift toward sustainable, intelligent infrastructure.

At the heart of this breakthrough is an innovative three-terminal photocapacitor—a device that merges a high-efficiency hybrid photovoltaic, a molecularly engineered supercapacitor, and eco-friendly, mushroom-derived chitosan membranes into a seamless system. This compact unit achieves a record photocharging voltage of 0.9 V and overall charging efficiency of 18% under typical indoor lighting, enabling continuous, battery-free operation of IoT networks and edge AI. In real-world tests, the platform powered image recognition tasks with 93% accuracy at just 0.8 mJ per inference, outperforming commercial silicon modules by a factor of 3.5 in throughput.

Professor Marina Freitag, Chair of Energy, Royal Society University Research Fellow, Newcastle University, who co-conceived and led the project, said: “This has been an idea brewing for almost a decade, bringing together everything from fundamental molecular engineering to real-world edge AI applications. I am absolutely delighted to see it finally realised—not just as an academic curiosity, but as a fully integrated, working system. It proves that only through deep, international collaboration can we solve the multi-faceted challenges of sustainable, intelligent technology.”

Why This Matters: A Sustainable Future for Billions of Devices

With over 30 billion IoT devices projected by 2030, the challenge of powering ubiquitous, wireless, smart systems—without toxic batteries or grid connection—is one of the defining issues in technology and sustainability. This work demonstrates a viable, high-performance solution for indoor environments, paving the way for zero-maintenance, energy-autonomous infrastructure in homes, hospitals, factories, and cities. It directly answers the United Nations Sustainable Development Goal 7 for affordable and clean energy and could help reduce the environmental impact of billions of disposable batteries annually.

Broader Implications: Building the Foundation for Smart, Sustainable Societies

The implications stretch far beyond the lab. This technology is a game-changer for smart cities, healthcare, industrial automation, and environmental monitoring—enabling networks of sensors and edge devices that require zero maintenance and have minimal environmental impact. By uniting molecular engineering, biodegradable materials, advanced simulation, and real-world AI integration, the team has set a new benchmark for what is possible when science and collaboration know no borders.

Professor Freitag adds: “Collaboration is the only way to tackle the multi-faceted problems of tomorrow’s technology. Our joint success is not just a scientific breakthrough—it’s a template for how global, cross-disciplinary teams can deliver the innovations society needs.”

International and Interdisciplinary Collaboration at Its Best

This breakthrough was only possible through a powerful, global team effort, uniting expertise in chemistry, materials science, device physics, AI, and systems integration. This milestone was achieved through an exceptional international effort, with the University of Rome “Tor Vergata” at its heart.

The Tor Vergata team, led by Dr Francesca De Rossi and Professor Francesca Brunetti, drove the pioneering integration of advanced supercapacitor technology, seamlessly marrying material innovation with device-level functionality. Their expertise in hybrid electronics and energy storage formed the backbone of the project’s system design, and their leadership in device assembly and performance testing was crucial to the breakthrough.

At Newcastle University, Marie Skłodowska-Curie Fellow Dr Natalie Flores-Diaz spearheaded device engineering and experimental direction, working closely with Timo Keller, Harvey Morritt, who advanced the development of high-performance polyviologen materials. The Spanish team, Dr Zaida Perez-Bassart, Dr Amparo Lopez-Rubio, and Dr Maria Jose Fabra Rovira, brought a sustainability edge with their innovative mushroom-derived chitosan membranes, ensuring the devices were not only efficient but also eco-friendly.

The Technical University of Munich, with Richard Freitag and Professor Alessio Gagliardi, translated these material advances into real-world impact, demonstrating the photocapacitors’ ability to power edge AI and IoT networks in practice.

Meanwhile, theoretical insights from Francesca Fasulo, Professor Ana Belen Muñoz-García, and Professor Michele Pavone at the University of Naples Federico II revealed why the new materials performed so robustly at the molecular level. The EPFL Lausanne team, Sandy Sanchez Alonso and Prof. Michael Grätzel provided gold-standard device and film characterization, validating each stage of progress.

Funders and Support

The work was supported by the EU Horizon 2020 Marie Skłodowska-Curie Actions (grant 101028536 to NFD), EPSRC UKRI (EP/W006340/1, EP/V035819/1), Royal Society University Research Fellowships (URF/R1/191286) to MF, Severo Ochoa Centre of Excellence (CEX2021-001189-S), and the CETPartnership SPOT-IT project.

Read the full article:
Flores-Diaz, N., De Rossi, F., et al. “Unlocking High-Performance Photocapacitors for Edge Computing in Low-Light Environments.” Energy & Environmental Science, 2025. DOI: 10.1039/D5EE01052G

Open data and code:
https://github.com/FreitagTeam/PhotoCap