Researchers at Stockholm University have used ultra-short X-ray laser pulses at facilities in South Korea to identify a long-hypothesized critical point in supercooled water, where two distinct liquid forms converge. The discovery, reported in Science, caps more than a decade of increasingly precise experiments that have chased water’s strangest behavior into a temperature zone once considered impossible to study. If confirmed by independent teams, the finding could reshape how scientists model everything from cloud formation to the survival of biological cells in extreme cold.
Two Liquids Hidden Inside One
Water already has a well-known critical point at high temperature and pressure, where liquid and vapor become indistinguishable. The new result concerns a second, far more elusive critical point deep in the supercooled regime, where water remains liquid well below its normal freezing temperature. At this liquid-liquid critical point, high-density and low-density forms of liquid water merge, producing dramatic swings in compressibility and density fluctuations.
Theoretical simulations have long predicted that deeply supercooled water should split into these two liquid phases under pressure. Computational work using realistic molecular models mapped the likely location of this critical point in the phase diagram, giving experimentalists a concrete target. The challenge was reaching the right temperatures without the water freezing first, a problem that kept the critical point locked inside what physicists call “no man’s land,” where conventional experiments fail because ice forms faster than measurements can be made.
Catching Liquid Water Before It Freezes
The experimental breakthrough rests on X-ray free-electron lasers, or XFELs, which fire femtosecond pulses fast enough to capture a snapshot of liquid structure before ice crystals can form. A foundational step came from work on supercooled microdroplets, which demonstrated that a “diffract-before-destroy” approach could record liquid water’s molecular arrangement below the temperature where ice normally nucleates spontaneously. In that scheme, micrometer-sized droplets are injected into a vacuum, evaporative cooling drives them to extreme supercooling, and an ultrafast X-ray pulse scatters from the liquid before crystallization has time to propagate.
Building on that method, a separate team applied an infrared-pump and X-ray-probe technique to amorphous ice under pressure, tracking structural changes on nanosecond-to-microsecond timescales. That experiment reported clear signatures of a transition between high- and low-density liquid-like structures, offering the first direct experimental glimpse of the phase boundary that theory had predicted. Although the sample started as amorphous ice rather than bulk liquid, the observed rearrangements pointed to an underlying liquid–liquid transition line buried in the supercooled region.
Maxima at 227 K and the Widom Line
A key piece of evidence came from XFEL scattering measurements on evaporatively cooled water droplets that reached approximately 227 kelvin. At that temperature, both isothermal compressibility and correlation length (two quantities that describe how strongly density fluctuations are amplified and how far they extend through a liquid) exhibited pronounced maxima. Those peaks are consistent with what physicists call a Widom line, a thermodynamic echo that extends outward from a critical point into the one-phase region above it. Crossing this line, the liquid’s response functions change sharply but continuously, as if skirting the edge of a hidden phase separation.
A separate infrared-pump and X-ray-probe experiment on evaporatively cooled micrometre-sized droplets added another crucial coordinate. That study reported a dynamic crossover near 233 kelvin in structural relaxation times, supported by molecular dynamics simulations. The crossover marks the temperature at which the liquid’s internal rearrangements slow down abruptly, shifting from relatively simple, high-temperature dynamics to a more sluggish, cooperative motion characteristic of deeply supercooled liquids. Such a change is expected when the system passes near a critical region, where fluctuations in local structure become large and long-lived.
Additional analysis of the same crossover, accessed through an institutional portal at Springer Nature, reinforced the interpretation that the 233 kelvin feature is tied to underlying critical fluctuations rather than to crystallization or experimental artifacts. Together, the 227 kelvin compressibility maximum and the 233 kelvin dynamic crossover bracket a narrow temperature window that matches theoretical predictions with striking precision, suggesting that both experiments are sensing the same buried critical landscape from different angles.
From Discontinuous to Continuous
The Stockholm University team’s latest result pushes the evidence further. By probing supercooled water at pressures that straddle the predicted critical point, they observed a crossover from a discontinuous to a continuous transition, with broad, slow structural variations at the crossover itself. In more accessible terms, at pressures below the critical point, the two liquid forms behave like distinct phases that can switch abruptly, akin to water freezing into ice. Above it, the distinction gradually fades, and the liquid’s structure shifts smoothly from one character to another without a sharp boundary.
That pattern is the textbook signature of a critical point in any fluid, where the line between two phases terminates and gives way to a single, continuously varying state. Observing it in water closes a gap that simulations alone could not fill and ties together years of scattered experimental hints into a coherent phase diagram. The work also shows how sensitive the transition is to both pressure and temperature, underscoring why earlier measurements that averaged over broad conditions struggled to see a clear signal.
Parallel XFEL measurements on supercooled water–glycerol microdroplets have helped refine the experimental protocol and test its robustness. That research characterized density fluctuations and structural heterogeneity in a related supercooled liquid mixture, providing methodological checks on droplet alignment, beam diagnostics, and how adding a solute shifts the apparent critical region. The glycerol experiments do not prove the pure-water critical point on their own, but they validate the droplet-stream techniques, data analysis tools, and temperature estimates that the Stockholm group relies on for its pure-water measurements.
Why a Hidden Critical Point Matters
Most coverage of this result has focused on the intellectual puzzle: does water really have two liquid forms? That framing, while accurate, understates the practical stakes. Supercooled water is not just a laboratory curiosity. It exists in vast quantities inside high-altitude clouds, where droplets can remain liquid down to temperatures below -30 °C before freezing. The way those droplets respond to tiny perturbations in temperature and pressure (how easily they compress, how quickly they rearrange internally, and how prone they are to crystallize) feeds directly into cloud lifetime, precipitation patterns, and the radiative balance of Earth’s atmosphere.
If a liquid–liquid critical point lurks near the conditions relevant to mixed-phase clouds, then small environmental changes could trigger disproportionately large shifts in droplet properties. That, in turn, could influence how efficiently clouds seed ice crystals, how they scatter sunlight, and how they interact with aerosols. Climate models typically treat supercooled water with simplified parameterizations; incorporating a critical region with strongly varying response functions could refine those treatments and improve predictions of cloud feedbacks.
Biological systems offer another arena where the new findings may matter. Many organisms (from overwintering insects to freeze-tolerant plants) rely on supercooled water inside their tissues or extracellular spaces. Near a critical point, fluctuations in local density and structure become intense, potentially affecting how solutes cluster, how membranes interact with their surroundings, and how easily ice nuclei form. A better grasp of water’s hidden phase behavior could inform cryopreservation strategies, where controlling ice formation and glassy states is central to keeping cells and tissues viable at low temperatures.
On the theoretical side, confirming a liquid–liquid critical point would cement water’s status as a paradigmatic “anomalous” liquid whose behavior cannot be captured by simple models. It would provide a unifying explanation for a host of oddities, such as the sharp rise in heat capacity and compressibility upon cooling, that have long resisted tidy interpretation. At the same time, the new data will challenge modelers to reconcile subtle experimental signatures with the predictions of competing molecular potentials, potentially driving the development of more accurate descriptions of hydrogen bonding and local order.
The work is not the final word. Independent experiments at other XFEL facilities, using alternative sample geometries and analysis methods, will be essential to test how robust the reported critical signatures are. But by threading a path through “no man’s land” with ultrafast X-rays, the Stockholm group and their collaborators have transformed a decades-old conjecture into a concrete, testable picture: inside the familiar liquid we drink and depend on, two distinct liquid states appear to meet at a hidden critical point, shaping water’s behavior in some of the coldest corners of nature.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
