Why you can't use water to make cryoporometric measurements of the pore size distributions in meteorites - or in high iron content clays, rocks or concrete.

Many porous materials have high susceptibility magnetic gradients in the pores, due to

the presence of iron or other magnetic materials. Thus if probe liquids are placed in the

pores they exhibit fast decaying signals with a short T2*. Usually the actual T2 of the

liquids is also reduced, due the presence of paramagnetic ions in the pore walls. The

usual solution in NMR is to measure an echo (or echo train) at short times. However,

recent work [J. Phys.: Condens. Matter 19, 415117, 2007.] has shown that water/ice

systems near a pore wall form rotator phase plastic ice, with T2 relaxation times in the

region of 100 to 200 ms. Thus if a NMR cryoporometric measurement is attempted

with a measurement time significantly less than 1 or 2 milli-seconds, the result is to

make a measurement based on the phase properties of the brittle to plastic ice phase

transition, not that of the brittle ice to water phase transition. This gives rise to

artefacts of small pore sizes that may not actually be present. This work successfully

uses a-polar liquids instead.