Global Research PartnershipsCreate free account
Research Collaboration Index

Broad Area Award · 2026

Physical Sciences & Engineering: the ten finalists

The ten finalist universities are the ten highest-scoring across all 10 subject indices in physical sciences & engineering (Chemical Engineering, Chemistry, Computer Science, Earth & Planetary Sciences, Energy, Engineering, Environmental Science, Materials Science, Mathematics, Physics & Astronomy), from 756 universities ranked in every one of them. 10 finalists are represented by a verified academic and project from the subject awards.

Finalists, not winners. The ten finalists in each broad area are the ten highest-scoring universities across every subject index in that area (mean subject score, complete coverage required). Exact positions publish with the full index on 15 September 2026 at 09:00 UK. Winners are decided by editorial review and announced on 15 September 2026. The Awards are exclusive to partner universities.

The finalists

Ten universities leading physical sciences & engineering.

Etienne Fluet‐Chouinard

🇨🇭 ETH Zurich · Environmental Science

Extensive global wetland loss over the past three centuries

How much of the world's wetland has disappeared, and over what timescale? Examining change across the past three centuries, the work assembles a long historical account of global wetland loss, extending the record well beyond the recent decades that most assessments are able to cover.

Nature · 2023 · 914 citations

Iain Staffell

🇬🇧 Imperial College London · Energy

Realistic roles for hydrogen in the future energy transition

Where does hydrogen genuinely belong in a decarbonising energy system? Addressing that question, the work sets out realistic roles for hydrogen in the future energy transition, examining where the fuel can contribute as energy systems change and, by implication, where its promise has limits.

Nature Reviews Clean Technology · 2025 · 406 citations

Hanry Yu

🇸🇬 National University of Singapore · Engineering

Organoids

Organoids are tissue-engineered models grown in the laboratory that reproduce much of the structure and function of real human tissue. Surveying the materials and methods used to culture, grow, differentiate and mature them, the work explains how these systems support fundamental studies of development, regeneration and repair, alongside diagnostics, disease modelling, drug discovery and personalised medicine.

Nature Reviews Methods Primers · 2022 · 976 citations

María Pérez‐Ortiz

🇬🇧 University College London · Earth & Planetary Sciences

Seasonal Arctic sea ice forecasting with probabilistic deep learning

Arctic sea ice is shrinking year-round, with far-reaching consequences for local communities, polar ecosystems and global climate. The team built IceNet, a probabilistic deep learning system trained on climate simulations and observations that forecasts sea ice concentration six months ahead, outperforming a state-of-the-art dynamical model for summer sea ice and extreme ice events.

Nature Communications · 2021 · 291 citations

Honghui He

🇺🇸 University of California, Berkeley · Engineering

Polarisation optics for biomedical and clinical applications: a review

Polarised light carries vectorial information that biological tissue transforms in revealing ways, and such techniques have been harnessed in biological and clinical research for decades. The work draws together methodologies and applications in tissue polarimetry, with emphasis on the Stokes-Mueller formalism, and surveys recent breakthroughs, development trends and potential multimodal uses alongside other techniques.

Light Science & Applications · 2021 · 598 citations

Xiao Hua

🇬🇧 University of Cambridge · Chemistry

Revisiting metal fluorides as lithium-ion battery cathodes

Metal fluorides have long been considered as cathode materials for lithium-ion batteries, and this work revisits the case for them. Appearing in the materials literature as a reassessment rather than a single experiment, it reconsiders what these compounds offer as cathodes and which questions about using them remain open.

Nature Materials · 2021 · 222 citations

Vivian Wing‐Wah Yam

🇭🇰 University of Hong Kong · Chemistry

Toward the Design and Construction of Supramolecular Functional Molecular Materials Based on Metal-Metal Interactions

Supramolecular materials assemble themselves through noncovalent forces such as hydrogen bonding, and metal-metal interactions have emerged as an unconventional addition to that toolkit. Exploring how these interactions can guide the design and construction of self-assembled metal-based materials with rich spectroscopic functionalities, the work aims to stimulate new research directions despite the challenge of controlling such hierarchical architectures.

Journal of the American Chemical Society · 2022 · 134 citations

David Hunter

🇬🇧 University of Oxford · Environmental Science

Pollution and health: a progress update

Drawing on the Global Burden of Diseases study for 2019, this work updates the earlier Lancet Commission estimate that pollution caused 9 million premature deaths a year, and finds the toll broadly unchanged at roughly one death in six worldwide. Deaths linked to household air and water pollution have fallen, but ambient air pollution and toxic chemicals such as lead have taken their place.

The Lancet Planetary Health · 2022 · 2,094 citations

Anna Heath

🇨🇦 University of Toronto · Mathematics

Economic Evaluation of Cost and Time Required for a Platform Trial vs Conventional Trials

Platform trials let new treatments join a study already under way, but little is known about what they cost to run. Surveying 146 international experts and drawing on the real entry dates of ten interventions into a long-running prostate cancer trial, this evaluation compares the money and time required for a platform trial against a series of conventional ones.

JAMA Network Open · 2022 · 61 citations

Xiuping Jia

🇦🇺 UNSW Sydney · Engineering

SpectralGPT: Spectral Remote Sensing Foundation Model

Most artificial intelligence models for imagery are built for ordinary photographs, not the rich spectral data captured from space. The team created SpectralGPT, the first universal remote sensing foundation model purpose-built for spectral imagery, using a three-dimensional generative pretrained transformer that accommodates images of varying sizes, resolutions, time series and regions.

IEEE Transactions on Pattern Analysis and Machine Intelligence · 2024 · 810 citations

Exclusive to partners

The Awards are open to partner universities.

Finalists are identified from open data; winners are decided by our editorial team and announced on 15 September 2026.

Explore partnership →

← All awards