Structural studies of ice: single crystals at 165 GPa

Ice has the richest pressure-temperature phase diagram known and its physics revolves around the hydrogen-bond which loosely and elusively binds hydrogen atoms to neighbouring molecules. Understanding this «H-bond» is central to all structural studies of this material, and more particularly its eventual 'symmetrisation' under pressure, that is when the H-atoms become equally strongly bonded to the neighbouring oxygen atoms. This new phase of ice, called ice X, has never been observed.

Above 2 GPa(1 GPa = 10,000 bar) we find ice VIII, a low temperature proton ordered polymorph, and ice VII, the ambient temperature proton disordered polymorph which has a cubic structure (see Figure 15). Ice VII has been studied up to 103 GPa [1] as a powder using vertically focused monochromatic radiation and image plates. A sufficiently large number of ice VII reflections were followed up to the highest pressure to characterise the stress distribution within the sample and correct the measured pressure accordingly.


The corrected compressibility data are shown in Figure 16.

However, the diffracted intensity was not sufficient to follow the evolution of the hydrogen-only '111' reflection above 18 GPa. This important reflection is forbidden in the body centered cubic oxygen lattice, but makes fleeting appearances as the hydrogen atoms tend to preferentially localise on the body diagonal. When visible, it should thus give information on the state of proton (that is hydrogen) ordering.

To bring out this very weak reflection, single crystals of ice VII were grown and taken to a new record pressure in single crystal diffraction: data were collected up to 165 GPa using the technique developed for hydrogen (see last year's Highlights). These data are also shown in Figure 16. As expected the '111' reflection was indeed followed over the whole pressure range and it revealed different regimes of proton ordering. We now have some evidence of proton localisation leading to ice X at 165 GPa, and if confirmed this should end a fascinating quest which started several decades ago.




[1] E. Wolanin (a), Ph. Pruzan (a), J.C. Chervin (a), B. Canny (a), M. Gauthier (a), D. Häusermann (b) and M. Hanfland (b)

[2] P. Loubeyre (a), R. LeToullec (a), M. Hanfland (b) and D. Häusermann (b)

(a) Physique des milieux condensés, Univ. Pierre et Marie Curie, Paris VI (France)

(b) ESRF



Towards the centre of the earth



The pressure at the centre of the earth is expected to be of the order of 350 GPa, and the temperature 5500 Kelvin. A little closer to our habitat, and more important for our daily life, it is now becoming possible to relate interactions between the earth's mantle and core at a depth of 2900 km with intense volcanic activities and changes in the earth's magnetic field for example. Down at the core-mantle boundary, pressures and temperatures are in the region of 135 GPa and 3500 K, and it is thus an exceptional materials laboratory where unusual chemical reactions and new structures are found. Consequently large efforts are being made to reproduce these extreme conditions using various experimental techniques and to combine these conditions with in situ X-ray characterisation.

Such studies require the development of special tools, in particular improved diamond-anvil cells able to compress larger samples to the highest possible pressures. Reaching pressures of several megabar (several hundreds of GPa) is very difficult, especially in X-ray diffraction measurements as mechanical support has to be sacrificed to achieve sufficient geometrical access. The limitations are the ultimate strength of the diamond anvils and other cell components.

To better understand these problems, a high intensity microbeam was used to study the elastic deformation, material strength and pressure distribution of diamond-anvil cell components (anvils, gasket and sample) up to 340 GPa (3.4 megabars). Pressures in the 300 GPa range were obtained in samples of rhenium, tantalum, tungsten, molybdenum and iron using bevelled diamond anvils with tip diameters of 10 to 25 µm.

Figure 17 shows how the strength of the cell components limits the maximum pressure obtained: the pressure increases quickly with load up to about 230 GPa, then most of the applied force goes into plastic deformation of the cell components. The deformation of the diamond anvil tip was monitored from X-ray transmission measurements and the pressure distribution across the sample and gasket from diffraction measurements using the same microbeam.

Figure 18 shows the deformation of the diamond anvils and rhenium gasket: at 300 GPa only a few µm2 of very thin sample (maximum X-ray transmission) are left on the tip of the diamonds. The main finding was that the change in pressure versus load behaviour is related to the deformation of the diamond anvils. These measurements of anvil deformation and associated pressure distribution provide crucial information on the optimisation of tip sizes and bevel angles of anvils which are important for improving the design of diamond anvil cells for ultrahigh pressure studies.

All these measurements were made using the diffraction and transmission signals from a 5 x 5 µm2 X-ray beam. It was the first time that X-ray diffraction under pressure was carried out with such a small beam.




H.K. Mao (a), G. Shen (a), R.J. Hemley (a), J. Badro (b),

Ph. Gillet (b), M. Hanfland (c) and D. Häusermann (c)

(a) Geophysical Laboratory and Center for High-Pressure Research, Carnegie Institution of Washington (USA)

(b) Laboratoire de Sciences de la Terre, Ecole Normale Supérieure des Sciences, Lyon (France)

(c) ESRF