🔬 materials science

Observation of Hexagonal Close-Packed Water Ice at Conditions in Ice Giant Planetary Interiors

Using synchrotron x-ray diffraction in laser-heated diamond anvil cells, researchers discovered that hexagonal close-packed (hcp) water ice, rather than the previously assumed face-centered cubic (fcc) structure, becomes the dominant phase under the high-pressure and high-temperature conditions found in the interiors of Uranus and Neptune, a finding that may significantly impact planetary dynamo models and ice mantle dynamics.

Original authors: Alexis Forestier, Gunnar Weck, Sandra Ninet, Gaston Garbarino, Mohamed Mezouar, Frédéric Datchi, Paul Loubeyre

Published 2026-02-13

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Alexis Forestier, Gunnar Weck, Sandra Ninet, Gaston Garbarino, Mohamed Mezouar, Frédéric Datchi, Paul Loubeyre

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine the inside of a planet like Uranus or Neptune as a giant, multi-layered cake. For decades, scientists have been trying to figure out exactly what the "frosting" in the middle of this cake is made of. We know it's mostly water, but under the crushing weight of the planet's gravity, that water isn't liquid or normal ice. It's a strange, super-dense material called superionic ice.

Until now, scientists thought this superionic ice was arranged in a specific, perfect pattern, like a stack of oranges in a grocery store (called a cubic or fcc structure). They believed this was the only way the ice could exist in the deep, hot, high-pressure heart of these planets.

The Big Discovery
This paper is like finding a secret ingredient in that planetary cake. The researchers, using a machine that acts like a microscopic, super-powerful magnifying glass (a synchrotron), discovered that the ice doesn't just stick to that one pattern. Under certain conditions, it rearranges itself into a hexagonal pattern (like a honeycomb).

Here is the story of how they found it, explained simply:

1. The Pressure Cooker (The Experiment)

To recreate the conditions inside a planet, the scientists used a Diamond Anvil Cell. Imagine two tiny, perfect diamonds pressing down on a speck of water ice smaller than a grain of sand. They squeezed it with a force equivalent to millions of atmospheres (over 200 gigapascals!) and heated it with lasers to temperatures hotter than the surface of the sun (over 2,000°C).

2. The Shape-Shifting Ice

Normally, when you heat ice, it melts. But under this extreme pressure, the ice doesn't melt into a liquid. Instead, it enters a "superionic" state.

The Analogy: Think of the oxygen atoms as the bricks of a wall, and the hydrogen atoms (protons) as bees buzzing around inside the wall.
In this state, the brick wall stays solid and rigid, but the bees (hydrogen) are free to zip around wildly, conducting electricity like a metal wire.

3. The "Stacking" Mystery

The researchers were watching how these "bricks" (oxygen atoms) were stacked.

The Old Theory: They thought the bricks were always stacked in a perfect, repeating cube pattern (like a pyramid of cannonballs).
The New Finding: They saw that as they cycled the temperature up and down, the bricks started to get confused. The perfect cube pattern began to wobble and shift. Some layers of bricks started stacking in a hexagonal (honeycomb) pattern instead.

It's like a game of musical chairs where the players (the atoms) are so crowded they start switching between two different dance formations. At lower pressures, the "cube dance" was dominant. But as they pushed the pressure higher (above 200 GPa) and kept it hot, the "honeycomb dance" (hexagonal) took over completely.

4. Why This Matters for Planets

This isn't just a fun fact about ice; it changes how we understand the "weather" inside Uranus and Neptune.

The Magnetic Field: Planets have magnetic fields generated by moving electrically conductive fluids (like a giant dynamo). Scientists thought the superionic ice layer was a solid, conductive shell that stopped the magnetic field from being simple.
The Twist: The new hexagonal ice might conduct electricity differently than the cubic ice. It's like the difference between a smooth highway and a bumpy, one-way street. The hexagonal ice might be less conductive in certain directions.
The Result: This could explain why Uranus and Neptune have such weird, tilted, and messy magnetic fields. If the inner core is made of this new hexagonal ice, it changes how the "dynamo" works, creating the strange magnetic storms we see from Earth.

The Takeaway

For a long time, we thought the deep interiors of ice giants were made of one specific type of super-ice. This paper tells us the reality is more complex: the ice is a shape-shifter. It can switch between a cubic structure and a hexagonal structure depending on how hot and how heavy the pressure is.

This discovery is like realizing that the foundation of a skyscraper isn't just one type of concrete, but a mix that changes its properties as you go deeper. It forces scientists to rewrite the blueprints for how these mysterious planets work, from their magnetic fields to their internal heat.

1. Problem Statement

Understanding the phase behavior of dense water ice is critical for modeling the geophysical properties of Uranus and Neptune. Current stratified interior models for these ice giants propose a thick layer of "ice" beneath a convective ionic fluid layer.

Current Consensus: It has been widely assumed that water ice in this deep interior adopts a face-centered cubic (fcc) superionic structure (Ice XVIII), recently confirmed experimentally.
The Gap: Theoretical calculations suggest that beyond the body-centered cubic (bcc) Ice X phase, close-packed oxygen sublattices (fcc or hexagonal close-packed/hcp) should exist. However, the stability field of the hcp phase under planetary conditions (high pressure and temperature) had not been clearly observed or distinguished from the fcc phase, despite predictions that their free energies are nearly identical.

2. Methodology

The researchers utilized synchrotron X-ray diffraction (XRD) within laser-heated diamond anvil cells (DAC) to simulate the extreme conditions of ice giant interiors.

Experimental Setup:
- Sample Geometry: An advanced "encapsulated" sample assembly was used, where the water ice was confined between two cup-shaped boron-doped diamond (BDD) laser absorbers. This geometry minimizes thermal gradients via two-sided indirect laser heating (1070 nm YLF lasers).
- Anvil Preparation: Diamond anvils were machined with focused ion beams (FIB) to create pits for the BDD absorbers. Thermal insulation was provided by a ~2.5 µm Al₂O₃ coating.
- Facility: Experiments were conducted at the ESRF-EBS (European Synchrotron Radiation Facility) on the ID27 beamline, using a sub-micrometer X-ray beam (0.3738 Å wavelength) to probe the small sample volume (~12 µm diameter).
Procedure:
- Run 1: Samples were heated to ~2000 K at ~80 GPa to form single-crystal fcc ice, then cooled to observe phase evolution.
- Run 2: Samples were compressed to >200 GPa, followed by thermal annealing and heating cycles up to 2630 K to map the high-pressure, high-temperature phase diagram.
- Pressure/Temperature Measurement: Pressure was determined via the equation of state of the BDD absorbers; temperature was measured using a two-sided achromatic pyrometric system (550–930 nm range).

3. Key Contributions

First Direct Observation: The study provides the first clear experimental observation of a hexagonal close-packed (hcp) oxygen sublattice in water ice under conditions relevant to ice giant interiors.
Phase Coexistence Mechanism: The authors demonstrate that the coexistence of fcc and hcp phases arises from stacking disorder within the oxygen lattice, triggered by temperature cycling and mechanical plasticity.
Superionic Transition Characterization: The paper identifies an anisotropic thermal expansion in the hcp phase, serving as a signature for a superionic transition with direction-dependent proton diffusion.

4. Key Results

Phase Coexistence (80–200 GPa):
- In Run 1 (82 GPa), cooling fcc ice from 2000 K resulted in the deformation of fcc diffraction peaks and the emergence of diffuse scattering. Upon further cooling, distinct hcp Bragg reflections (100, 002, 101) appeared alongside fcc peaks.
- The $d$ -spacings of the fcc (111) and hcp (002) reflections nearly coincided, confirming that the structures differ primarily in the stacking sequence of dense oxygen planes.
- This coexistence was attributed to stacking faults and mechanical plasticity, consistent with recent machine-learning potential simulations showing nearly identical free energies for fcc and hcp arrangements.
Dominance of hcp Phase (>200 GPa):
- In Run 2, as pressure increased beyond 200 GPa and temperatures reached 2250–2630 K, the relative intensity of hcp reflections increased while fcc peaks diminished.
- At 219 GPa and 2630 K, the fcc peaks vanished entirely, leaving the hcp phase as the sole observable ice form. This indicates that the hcp phase becomes thermodynamically dominant over fcc at these extreme conditions.
Anisotropic Superionic Behavior:
- Lattice parameter analysis of the hcp phase at ~197 GPa revealed that thermal expansion is driven almost entirely by elongation along the c-axis.
- This "S-shaped" thermal expansion curve mirrors the crossover observed in fcc superionic ice but suggests an anisotropic proton diffusion mechanism (proton hopping preferentially along the c-axis), similar to observations in superionic ammonia.

5. Significance and Implications

Planetary Dynamo Models: The existence of a stable, deep hcp superionic ice layer has profound implications for the magnetic fields of Uranus and Neptune.
- Current models propose a thin, dynamo-active outer shell surrounding a stable, non-convecting inner shell.
- The observed anisotropy in hcp ice implies lower electrical conductivity compared to the isotropic fcc phase. This makes hcp ice a stronger candidate for the "less conductive" solid inner core required by dynamo models to explain the non-dipolar, non-axisymmetric magnetic fields of these planets.
Mantle Dynamics: The mechanical properties (plasticity) and distinct transport properties of hcp ice could alter the dynamics of the ice mantle, influencing heat transport and convection patterns within the planet.
Future Directions: The findings necessitate further theoretical work on the mechanical plasticity and electrical conductivity of hcp ice, as well as continued experimental efforts to fully map the pressure-temperature stability domains of these competing phases.