Water has been studied for centuries, yet it keeps revealing new tricks under pressure. In experiments published in May 2026, an international team led by researchers at the Korea Research Institute of Standards and Science (KRISS) identified two previously unknown forms of high-pressure ice. One of them, tentatively called ice XXII, contains a repeating structural unit of 304 water molecules, dwarfing every ice phase ever documented. For comparison, ordinary ice has just four molecules in its repeating unit, and the well-known high-pressure form ice VII has only two.
The discoveries push the catalog of known ice phases past 20 and open new questions about how water behaves deep inside icy moons and giant planets, where crushing pressures and rapid temperature shifts are the norm rather than the exception.
Catching ice in the act
The key to finding these phases was speed. The research team used a dynamic diamond anvil cell, a device that squeezes a microscopic water sample between two diamond tips while ramping pressure from ambient to roughly two gigapascals in just milliseconds. That is about 20,000 times atmospheric pressure, applied faster than the blink of an eye.
To watch what happened at the molecular level, the team paired the diamond cell with time-resolved X-ray diffraction at the European XFEL, one of the world’s most powerful X-ray free-electron laser facilities. The combination let them observe water molecules snapping into crystalline arrangements in real time, capturing fleeting structures that would vanish before conventional instruments could record them.
“Rapid compression of supercooled liquid water at room temperature produces crystalline outcomes that slow, equilibrium compression simply does not,” the team reported. Under these conditions, water skipped past the stable phases scientists expected and instead formed metastable arrangements, crystals that are real and ordered but not the lowest-energy configuration for those pressures.
Ice XXI: the peer-reviewed discovery
The first new phase, ice XXI, is described in a peer-reviewed study published in Nature Materials. That paper documents ice XXI’s formation at about two gigapascals and room temperature, details its crystal structure using European XFEL diffraction data, and maps the compression pathways that produce it. An accompanying expert commentary in the same journal places ice XXI within decades of research on ice polymorphism, the phenomenon by which a single substance can adopt many different crystal structures depending on pressure and temperature.
Lawrence Livermore National Laboratory, which contributed to the experimental work, confirmed the methods and emphasized that metastable ice phases matter for building a complete picture of water’s behavior under extreme conditions. Separate research published in Nature Communications has independently shown that compressing water quickly can yield unexpected crystalline forms distinct from the ice VII structure that textbooks predict at high pressures.
Ice XXI is now experimentally established. Its existence demonstrates that the toolkit of ultrafast compression and XFEL imaging can reliably uncover ice phases that slower methods miss entirely.
Ice XXII: the 304-molecule giant awaiting confirmation
The bigger headline belongs to ice XXII, but it comes with an important caveat. The 304-molecule unit cell is reported in a preprint posted on arXiv by Kenji Kobayashi and colleagues. Preprints allow researchers to share findings quickly, but they have not yet undergone formal peer review. Until an independent group reproduces the structure or the manuscript clears journal review, the 304-molecule figure should be understood as a reported measurement, not a confirmed fact.
That said, the preprint is not coming out of nowhere. It builds directly on the validated experimental platform described in the Nature Materials paper, using the same dynamic diamond anvil cell and European XFEL setup. According to the manuscript, both ice XXI and ice XXII crystallized directly from supercooled liquid water under rapid compression, with ice XXII’s enormous repeating unit distinguishing it from anything in the existing ice catalog.
No data on ice XXII’s long-term stability have been published. The preprint documents its formation during rapid compression but does not address whether the phase can be preserved or studied once pressure is released. Lawrence Livermore’s public commentary so far focuses on ice XXI; the lab has not specifically addressed ice XXII or its unusually large unit cell.
Why it matters beyond the laboratory
If ice XXII’s structure holds up under scrutiny, it would suggest that the space of possible ice arrangements is far larger than the roughly 20 phases identified over the past century. That has practical consequences for planetary science.
Jupiter’s moon Europa, for instance, harbors a subsurface ocean beneath a thick ice shell. According to the KRISS news release, the discovery could be relevant to such icy worlds, where high pressures and rapid changes in local conditions might favor metastable ice formation. Understanding which ice structures appear under fast compression could refine models of heat transfer and geological activity on Europa and similar bodies like Saturn’s moon Enceladus.
Separating confirmed results from preliminary claims
The picture that emerges is layered. At its core sits a peer-reviewed, experimentally validated finding: rapid compression of water at room temperature can generate new metastable ice phases, and at least one such phase, ice XXI, is now part of the scientific record. Extending outward, the arXiv preprint proposes a second phase with a strikingly complex 304-molecule repeating unit, a claim that is promising but preliminary.
Competing characterizations of the discovery’s novelty also deserve attention. A KRISS news release describes ice XXI using language to the effect of “the world’s first discovery of a new form of ice.” That framing appears to refer to the first new ice phase identified by KRISS rather than the first new ice phase ever found. Other research groups have continued to identify new and rare ice forms independently, so the “world’s first” label applies narrowly to the institution’s own record.
The broader takeaway does not depend on ice XXII alone. Even the confirmed discovery of ice XXI reinforces a growing realization among researchers: water under extreme conditions is far more structurally versatile than long assumed. Each new phase adds a tile to a mosaic that stretches from diamond-tipped laboratory cells to the hidden oceans of distant worlds, and the mosaic is clearly not finished yet.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
