Studies of nano-structured liquids in confined geometry and at surfaces.

This is a progress report on elucidating the behaviour of liquids,

in particular water, in confined geometry on the nano- to mesoscale,

and at interfaces. There are important measurements still

to make, conclusions still to be drawn, and above all leaps of

understanding still to be made. However, a number of important

features in the behaviour of these systems have recently become

clearer.

Nano-structuring of liquids and their crystals changes their

Gibbs free energy, and hence their dynamics. This may most readily

be probed by monitoring the alteration of phase changes as a

function of temperature, together with changes in other parameters,

particularly the confinement diameter. Such studies may be

performed by monitoring the change in the pressure (at constant

temperature) of the liquid in its own vapour (Kelvin equation), or

by monitoring the change in the freezing/melting temperature

(at constant pressure) of a crystal in its own liquid (Gibbs–Thomson

equation).

In the latter case the melting and freezing temperatures of liquids

are modified by the changes in the volumetric Gibbs free energy

due to nanostructuring; this is related to the surface energy of

the curved interface between the crystal and its own liquid. This is

thus dependent on the geometry of the interface between the crystal

and its liquid. There is still discussion on this point as to the exact

geometric constants and functional forms that are applicable

for different confining geometries. Experimental evidence is presented

for the cases of cylindrical pores (SBA-15), and for pores

that on average are spherical (sol–gel). However, reconciling this

comparative data with melting/freezing temperatures in each of

these systems still pose a number of questions.

It is well known that bulk brittle ice has a hexagonal structure,

while brittle ice that forms in pores may be cubic in structure [1,2],

Figs. 10 and 11. Adjacent surfaces appear to further alter the

dynamics and structure of confined liquids and their crystals, leading

in the case of a water/ice system to a state of enhanced rotational

motion (plastic ice) just below the confined freezing/

melting transitions. This plastic ice layer appears to form at both

the ice–silica interface and the ice–vapour surface, and reversibly

transforms to brittle ice at lower temperatures. There is good evidence

to suggest that the plastic ice at a silica interface transforms

to cubic ice, while the plastic ice at vapour surfaces transforms to

hexagonal ice. That this plastic ice may correspond to a layer at the

crystal surface is suggested by the presence of only amorphous ice

in confined systems with small dimensions (<3 nm diameter),

whereas systems with larger dimensions (10 nm) contain brittle

cubic ice and also some hexagonal ice (if a vapour interface is present);

even larger systems (>30 nm) contain predominately hexagonal

ice. It is conjectured that this layer of plastic ice at vapour

surfaces may be present at the myriad of such interfaces in macroscopic

systems, such as snow-packs, glaciers and icebergs, and may

be an explanation for the need for plastic terms in the macroscopic

dynamical models of these systems [3].

These results also point the way forward for a wide-range of

cryoporometric metrology studies of systems that are ‘difficult’

for NMR, such as high iron content clays and rocks, as well as aged

concrete. Results are presented for cryoporometric measurements

on meteorite samples with a significant metallic content, exhibiting

T2 relaxation times down to 2.5 us.