Water self-diffusion studies in complex materials with fast-relaxing components: static and pulsed field methods revisited

Field gradient NMR techniques have proved very useful for measurements of molecular self-diffusion coefficients in many materials ranging from simple liquids to more complex materials such as polymer melts, porous media and biological tissues. In such multicomponent systems, spin relaxation phenomena may not only lead to a reduction of the signal intensity available for the field gradient NMR experiment but also to systematical errors in the measured diffusion coefficient due to unwanted relaxation time weighting. While pulsed field gradient (PFG) techniques allow better separation of relaxation and diffusion effects, they require at the same time longer echo times. This in turn aggravates unwanted weighting by transverse relaxation. For materials containing fast-relaxing components such as clayey sediments or partially hydrated cement the use of static field gradient (SFG) techniques suggests substantial benefits due to shorter echo times (i.e. less transverse relaxation time weighting). However, the separation of relaxation and diffusion effects is more difficult in SFG. In this contribution, we present the echo ratio approach to separate both effects and discuss its limitations. As we will show on the basis of experimental and calculated results, the echo ratio technique works well for homogeneous samples and complex materials with sufficiently long longitudinal relaxation times. In materials with components for which T1 is comparable or shorter than the diffusion time, however, this approach also fails and no meaningful interpretation of the SFG data on the basis of simple fit functions is possible so far