Water site headerMasthead Island, Great Barrier Reef Print-me keygo to Water Visitor Book contributions
Go to my page Water Structure and Science
5^12 6^4 cavity

Ultra low-density ices

V Other clathrates

V Ice-sixteen (ice XVI)

V Ice-seventeen (ice XVII)
V Other potential low-density ices


If ice is stretched (i.e. put under negative pressure), a it may form (meta-) stable ultra-low-density ices. Such ices may be stable at ambient pressure (positive) in the presence of interstitial molecules that prevent structural collapse. However under negative pressure (metastable conditions with respect to mixed gaseous water/hexagonal ice), the need for these stabilizing molecules may not be required. At present there are three such ices, two (ice XVi and ice XVII) have been experimentally proven whilst the other (empty clathrate s-III), although strongly supported by modeling, has not yet been experimentally proven.

Ice-sixteen (Ice XVI)


Clathrate-II crystal structure

For interactive Figures, see Jmol

Ice XVI may be formed from neon hydrate (a clathrate of structure sII) after five days of continuous vacuum pumping to remove all guest neon atoms [2252]. On removal of the neon, ice XVI expands, with the 512 cavities expanding by 3.9 % and the larger 51264 cavities expanding by 3.3 %, both due to the loss of the attractive van der Waals interactions. It is the least dense of all experimentally-known crystalline water phases (0.81 g.cm-3) and is expected to be the stable low-temperature phase of water at negative pressure but collapses and decomposes at 145 K. It has very slightly larger lattice constants (0.1% increase) than the filled hydrate at low temperatures [2252].


The structure of ice XVI is the same as the cage structure of CS-II clathrate hydrate and shown left (the water hydrogen atoms have been left out). Cubic crystals contain sixteen 512 cavities, eight larger 51264 cavities and 136 H2O molecules per unit cell. The tetrahedral 51264 cavities form an open tetrahedral network, with their centers arranged reminiscent to the cubic ice structure and separated by groups of three 512 cavities.


While the thermal expansivity of the filled CS-II hydrates are positive as usual for solids, ice XVI exhibits negative thermal expansivity below 55 K as with other ices such as hexagonal ice [3029].


The hydrogen bonding is random, following the 'ice rules'.

Ice-seventeen (Ice XVII) ice-XVII, from [2796]

Molecular hydrogen (H2) forms a number of different clathrates as the H2 molecules can fit in most cages with multiple H2 in the larger cages (see elsewhere). At high pressures (~400 MPa, 280 K) a different structure forms (an intercalated filled ice rather than a clathrate) with the composition (H2O)2H2 and consisting of interpenetrating spiral chains of water molecules showing topological similarity to the mineral quartz [2773]. All the hydrogen can be released by warming under vacuum for about one hour at a temperature of 110-120 K to give a new metastable hexagonal ice structure with space group P6122 with cell dimensions 6.3305 Å, 6.3305 Å, 6.0580 (a, b, c; 90°, 90°, 120°, 6 molecules, 75 K). The hydrogen bonding is random, following the 'ice rules'.

ice-XVII, from [2796]


Above 130 K, the sample transforms into ice Ih. It is microporous, due to its hexagonal helical pores (with 6.10 Å. channel diameter from the center of the oxygen atoms leaving a space of ~3 Å minimum diameter) and will re-adsorb molecular H2(~1.8 Å diameter) if exposed to it. The structure of the heavy water version of this ice has been determined by neutron diffraction [2796]. It consists of only sheets of pentamers of hydrogen bonded water molecules made up from (H2O)12 partial dodecahedra (see left, the xz plane down the y-axis) that form hexagonal helices down the z axis (see above right). This xz sheet-like structure is also formed in the [CoII(cyclam)Cl2].3H2O (cyclam =1,4,8,11-
tetraazacyclotetradecane) clathrate [1871].

For an interactive Figure, see Jmol.

Other potential low-density ices

3x3 unit cells of clathrate s-III highlighting the two types of cavity

There are very many potential low-density ices that may be discovered by modeling. For example their structures may be related to the 219 four-coordinated structures in the zeolite database [3039]. It has been found that (models of) some of these are stable near absolute zero under negative pressure and have densities approaching zero. These 'aeroices' consist of long struts of hexagonal cross-section joined at polyhedral vertices and enclosing vast empty spaces. The fact that such structures are found to be stable by modeling does not make them more likely to be found experimentally, so that they are not deserving of numbering as new ices. Some are interesting, however.


A cubic S-III clathrate (made up of two large 4126886 and six smaller 4882 cavities with 48 water molecules per unit cell, see left where one each of these novel cavities are highlighted and the water hydrogen atoms have been left out) has been found theoretically by modeling [2507] with a lower density (0.59 g.cm-3 without interstitial molecules) than ice XVI. It is expected to be more stable than ice XVI at pressures below (more negative) than about -400 MPa at 0 °C.


The large 4126886 cavities were stabilized by the presence of dodecahedrane (C20H20) molecules .

(12)4(6)4(4)18 cage showing the 4 12-water rings


Another such low-density structure (0.506 g cm3) is a potential cubic crystalline phase of ice clathrate, which is composed of eight very large icosihexahedral cavities (12464418) (shown right; all 48 balls shown are water molecules forming four tetrahedrally positioned 12-membered rings), eight intermediate dodecahedral cavities (6646), and sixteen very small octahedral cavities (6246) per unit cell (192 H2O, Space group Fd-3) [2862]. The icosihexahedral cavities (12464418) are both strained, due to the large number of four-membered rings, unstable in that they easily collapse, and highly porous due to the large ring sizes.




[Back to Top to top of page]


a Negative pressures are possible for liquids and solids but not possible for gasses. They are achieved by stretching the material, which is then metastable with respect to the production of a vapor state that occurs via a nucleated process with an activation barrier. Material at negative pressure can remain intact for long periods of time until this activation barrier is overcome. [Back]



Home | Site Index | Ices, introduction | Ice-Ih | Ice-Ic | Ice-Isd | II | III | IV | V | VI | VII | VIII | IX | X | XI | XII | XIII | XIV | Amorphous ice | LSBU | Top


This page was established in 2015 and last updated by Martin Chaplin on 14 September, 2017

Creative Commons License
This work is licensed under a Creative Commons Attribution
-Noncommercial-No Derivative Works 2.0 UK: England & Wales License