Pressure, not confinement, drives water reactivity
For a decade, scientists watched the same experiment yield opposite results — water squeezed into nanoscale spaces sometimes grew more reactive, sometimes did not, and no one could explain why. A team led by Xavier Advincula has now traced the contradiction to its source: it was never confinement itself that changed water's behavior, but the immense internal pressures — reaching several gigapascals — that confinement quietly generates. Published in Science Advances, the finding restores coherence to a fractured field and opens a deliberate path toward engineering the chemistry of water at its smallest scales.
- A decade of contradictory experimental results had left researchers unable to trust their own data on how water behaves when squeezed into spaces billionths of a meter across.
- Machine learning simulations revealed that van der Waals forces between atom-thin sheets generate gigapascal-scale pressures internally — pressures comparable to deep Earth — without any external force applied.
- When confined water was compared directly with bulk water under identical pressure, the reactivity gap nearly disappeared, exposing pressure — not confinement — as the true driver.
- Surface chemistry proved to be a secondary but real lever: hexagonal boron nitride chemically bonded with hydroxide ions at droplet edges, stabilizing them and lowering the energy cost of water splitting in ways that inert graphene could not.
- The field is now converging on a practical design principle — choose confining materials and control internal pressures deliberately to tune water reactivity for hydrogen fuel cells, batteries, and catalytic membranes.
For a decade, a stubborn contradiction haunted nanoscale water research. Experiments showed that water squeezed into spaces billionths of a meter across sometimes became dramatically more reactive — and sometimes did not. The disagreement resisted explanation, leaving researchers unable to reconcile their own findings.
A team led by Xavier Advincula has now resolved the puzzle. Using machine learning simulations capable of quantum mechanical accuracy across a far wider range of conditions than traditional methods allow, the researchers studied water trapped between sheets of graphene and hexagonal boron nitride — materials that are structurally similar but chemically opposite. What they found reframed the entire question: the water droplets were experiencing internal pressures of several gigapascals, generated not by any external apparatus but by van der Waals attraction pulling the surrounding sheets together. These are pressures found deep inside the Earth, arising spontaneously from the geometry of confinement.
Those pressures, it turned out, explained the reactivity boost almost entirely. When confined water was compared with ordinary bulk water held at the same pressure, both split into hydronium and hydroxide ions at essentially the same rate. Confinement alone was not the cause — pressure was. As Cambridge's Angelos Michaelides observed, once pressure and chemical potential were properly accounted for, much of the apparent mystery dissolved into thermodynamics.
The surrounding material still played a meaningful role. Around hexagonal boron nitride, hydroxide ions bonded chemically with the surface at droplet edges, stabilizing the reaction products and lowering the energy required for water to split further. Graphene, chemically inert, offered no such participation. The identity of the confining material could actively shape what happened inside.
The practical consequences extend across technologies that depend on water moving through tight spaces — hydrogen fuel cells, batteries, ion-selective membranes, and catalytic systems. Engineers can now approach these systems with a clearer design logic: select confining materials whose surfaces interact with water's dissociation products, and control the pressures those materials generate. The researchers plan to extend the work into more realistic environments, including the defects and edges common in real materials, and to screen broad families of two-dimensional surfaces for combinations that can enhance or suppress reactivity on demand. A decade of conflicting data has found its explanation, and with it, a new foundation for nanoscale chemical engineering.
For years, scientists have puzzled over a stubborn contradiction in their labs. Water squeezed into spaces billionths of a meter across seemed to behave differently depending on who was measuring it and how. Some experiments showed it became more reactive. Others did not. The disagreement persisted for a decade, frustrating researchers who could not figure out why the same phenomenon kept producing opposite results.
Now a team led by Xavier Advincula has cracked the mystery, and the answer rewrites what we thought we knew about water at the nanoscale. The reactivity boost that researchers observed was not actually caused by confinement itself—the act of squeezing water into tiny spaces. It was caused by pressure. And once you account for that pressure correctly, the contradictions vanish. The findings, published in Science Advances, matter because water trapped in nanoscale pores, membranes, and biological channels is everywhere—in nature and in the technologies we build.
Water has a fundamental chemical property: it splits into two charged particles, the hydronium ion and the hydroxide ion. This splitting determines pH, the measure of acidity or alkalinity, and it drives everything from the enzymes that keep cells alive to the reactions inside batteries. Scientists wanted to know whether confining water to spaces just billionths of a meter across changed how readily this splitting happened. The question seemed simple. The answer turned out to be layered.
The researchers used machine learning simulations that could reproduce quantum mechanical accuracy while exploring a much wider range of conditions than traditional computational methods allowed. They examined water trapped between sheets of graphene and hexagonal boron nitride, two materials that are each just one atom thick and structurally similar, but chemically very different. What the simulations revealed was striking: water droplets confined between these materials experience internal pressures of several gigapascals—pressures comparable to those found deep inside the Earth. No external force was applied. The pressure developed naturally because of van der Waals attraction, the weak force between individual atoms that becomes remarkably strong across the large surface area of two-dimensional materials, pulling the sheets together and compressing the water trapped between them.
These intense pressures greatly increased the splitting of water molecules. But when the researchers compared confined water with ordinary bulk water exposed to the same pressure, both behaved essentially the same way. The increased reactivity came primarily from pressure itself, not from confinement alone. As Angelos Michaelides of the University of Cambridge noted, what surprised the team most was how much of the apparent confinement effect could be explained by thermodynamics. Once pressure and chemical potential were properly accounted for, much of the complexity simply resolved.
Yet the surrounding material still mattered. In water droplets confined by hexagonal boron nitride, hydroxide ions that formed around the edges bonded chemically with the surrounding material. This stabilized the ions, lowered the energy required for water to split, and increased the amount of dissociation. The same effect did not occur with graphene because its chemically inert surface did not participate in the reaction. The material surrounding confined water could actively shape its chemical behavior.
The implications reach into technologies that depend on confined water: hydrogen fuel cells, batteries, ion selective membranes, and catalytic systems. Rather than focusing solely on pore size or channel dimensions, engineers can now tailor water reactivity by choosing a confining material whose surfaces interact with the products of water dissociation and by controlling the pressures generated within confined spaces. The researchers plan to study more realistic environments that include the defects and edges commonly found in practical materials, and to screen large families of two-dimensional materials and surface chemistries to identify combinations that can either enhance or suppress water reactivity for specific applications. A decade of conflicting studies has finally found its resolution, and in that resolution lies a practical design principle for the next generation of nanoscale chemical engineering.
Notable Quotes
When we compared systems under equivalent thermodynamic conditions, the effect of confinement largely disappeared. The contradictions in the literature were largely because scientists were comparing systems at different effective pressures or densities without realizing it.— Xavier R. Advincula, lead author
Rather than focusing solely on the size of pores or channels, we can tailor water reactivity by choosing a confining material whose surfaces interact with the products of water dissociation and by controlling the pressures generated within confined spaces.— Dr. Christoph Schran, Cavendish Laboratory
The Hearth Conversation Another angle on the story
So for ten years, scientists got different answers to the same question. What was actually going wrong?
They were comparing systems at different pressures without realizing it. When you squeeze water into a tiny space, the walls naturally press inward with enormous force—several gigapascals. Some experiments accounted for that pressure, others didn't. So they were measuring different things and calling them the same thing.
But the pressure isn't applied from outside. Where does it come from?
Van der Waals attraction. It's a weak force between atoms, but when you have two sheets of material just one atom thick with a huge surface area, that weak force adds up to something enormous. The sheets pull together and compress whatever is trapped between them.
So confinement itself doesn't make water more reactive?
Not inherently, no. The reactivity increase comes from the pressure. If you take ordinary water and expose it to the same pressure, it behaves the same way as confined water.
Then why does the material surrounding the water matter?
Because the surface can interact with the products of water splitting. Hexagonal boron nitride can chemically bond with hydroxide ions that form at the edges. That stabilizes them and makes the splitting easier. Graphene can't do that because it's chemically inert.
So you could engineer this? Choose materials strategically?
Exactly. You're not just picking based on pore size anymore. You pick based on whether the surface will help or hinder the reaction you want, and you control the internal pressure. It's a design principle.
What happens next?
They want to test this against real laboratory measurements and study materials with actual defects and imperfections. Then screen different combinations of two-dimensional materials to find the best ones for specific applications—fuel cells, batteries, that sort of thing.