Computer chips are now so densely packed with components that they generate heat faster than current cooling systems can remove it. This project tackles a fundamental physics problem: we do not fully understand how heat moves across the tiny interfaces where a nanocoating meets a cooling liquid, or where that liquid turns into vapour. Without that understanding, promising new cooling techniques—such as boiling liquid on nanotextured surfaces or drawing it through nanoporous membranes—cannot be reliably designed or scaled up. The research combines molecular dynamics simulations, machine learning, and new theoretical models to predict heat transfer at these interfaces. The team will then build a first-of-its-kind simulation toolkit and validate it against real experiments. If successful, the work could unlock ultra-efficient cooling for everything from data centres and 5G base stations to electric vehicle power electronics—systems that currently consume vast amounts of energy and water for cooling. This is fundamental engineering science; there is no immediate product. But a reliable simulation-driven design method would replace today’s trial-and-error approach, potentially cutting the environmental cost of our digital infrastructure while allowing electronics to keep shrinking.
View original technical description
Our global technological infrastructure depends on producing smaller, faster, and energy-efficient electronic devices. However, the miniaturisation needed to increase power density leads to larger heat fluxes, representing a bottleneck for next-generation electronics. Current thermal management systems (TMS) are both inadequate and inefficient, requiring vast natural resources. Novel, energy-efficient, and ultrahigh performance TMS are urgently needed in order to sustain economic growth while combating climate change. Nanomaterial-enhanced two-phase cooling has recently shown immense promise, but limitations in our understanding of nanoscale interfacial heat transfer - where the nanocoating or nanostructure meets the coolant liquid and where the liquid meets its vapour - in experiments has prevented this promise from being realised. The proposed work is of fundamental engineering science which will radically improving our understanding of interfacial heat transfer. This will be achieved by: i) combining molecular dynamics (MD) simulations and interfacial phonon analysis to develop new theoretical models for heat transfer across simple solid/liquid interfaces; ii) developing machine-learning-powered MD to extend this approach to realistic nanomaterials and coolant liquids; and iii) building a first-of-its-kind multiphysics simulation toolkit for TMS design, and validating against novel experiments in two case studies involving emerging nanomaterial-enhanced TMS: a) pool boiling using nanocoated surfaces; and b) evaporative cooling within nanoporous membranes. NANO-COOL will fill critical knowledge gaps and open an entirely new research field: "Simulation-driven design of nanomaterial-enhanced two-phase cooling for electronics". Results of NANO-COOL will be published as open-source models, open-access articles, and open data repositories. The long-term ambition of NANO-COOL is to become the enabling framework for novel two-phase TMS design.
Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.
Is something wrong? Let us know