Probing water motion in heterogeneous systems.

In this Thesis a practical approach is presented to study water mobility in heterogeneous systems by a number of novel NMR sequences. The major part of this Thesis describes how the reliability of diffusion measurements can be improved using some of the novel NMR sequences. The reliability of the data can be further enhanced by combining different NMR characteristics in a single fit routine. In addition, a fast NMR sequence for flow measurements is shown. A wide variety of samples is used to demonstrate how the NMR sequences and the subsequent analysis work.Throughout this Thesis the term heterogeneous system is used whenever a sample contains different physical or chemical environments or compositions, each of which influence the NMR signal in a distinguishable way. One of the effects of such a heterogeneous system is the variation of the magnetic susceptibility within the sample causing so-called in situ magnetic field gradients. These gradients lower the NMR signal amplitude and may cause a substantial deviation of the real diffusion constant from the one measured by NMR using pulsed field gradient spin echo or stimulated echo sequences. In the second half of Chapter 2 an improved version of the PFG multiple spin echo sequence is introduced which minimises the degrading effects of in situ field gradients. While using additional r.f. pulses between the pulsed field gradients to reach the desired reliability no compromises with regard to flexibility were necessary.This flexibility is important to study the change of the apparent diffusion constant with the echo time. This phenomenon, superficially appearing as an artefact, arises because compartments, as for example a vacuole, not only have a characteristic diffusion constant but also a T _{<font size="-1">2</font>} which may significantly differ from other T _{<font size="-1">2</font>} 's in the sample. In Chapter 3 it is shown how a T _{<font size="-1">2</font>} difference can be used to separate diffusion constants (D) even if these have a similar magnitude. Using Diffusion Analysis by Relaxation Time Separation the diffusion constant for water in the vacuole, the cytoplasm and the extra cellular space, respectively, can be distinguished in apple parenchyma tissue. In the second half of Chapter 3 a fast implementation of DARTS PFG NMR is presented which is combined with NMR imaging. Here, spatially localised T _{<font size="-1">2</font>} measurements are preceded by pulsed field gradients for diffusion weighting. In this way the diffusion constant and fractional amplitude of cerebral spinal fluid (CSF) and white and grey matter in cat brain can be measured. The validity of these measurements are supported by Monte Carlo simulations of computer generated 2D data sets.Chapter 4 deals primarily with improving the DARTS technique without imaging (DARTS PFG MSE Carr Purcell Meiboom Gill) and determining the best resolution (for discriminating diffusion constants) which can be obtained, using Monte Carlo simulations on 2D data sets. The use of the 2D fitting routine is extensively studied. Furthermore, existing theoretical models are mutually compared and confronted with the results obtained by the PFG MSE CPMG sequence for different samples. Four heterogeneous samples with different complexity are studied, i.e.:a- A sample consisting of two tubes with different fluids, between which exchange can be ignored.b- Whole Blood where diffusive exchange between the red blood cells (RBC's) and the plasma causes the apparent D of the RBC's to increase with increasing observation time.c-Apple parenchyma tissue where the membrane between the vacuole and the cytoplasm (the tonoplast) is shown to severely restrict, but not prevent, diffusive exchange between these compartments.d- A column with Sephadex beads (porous beads) where flow is introduced. The effect of this flow on the water displacement outside and inside the beads is described. Experiments demonstrated exchange between the flowing fraction and the stationary water in the beads.In all instances it is profitable to combine the diffusion and T _{<font size="-1">2</font>} measurements and analyse the resulting 2D data set as a whole. In this way an intuitive understanding is obtained of diffusion in complex systems measured by NMR. By using the T _{<font size="-1">2</font>} as a label, the resolving power of NMR to distinguish diffusion constants is greatly improved and a difference between the diffusion constants as small as 30 % is demonstrated to be resolvable. None of the presented theories can be used to quantitatively describe the data.In Chapter 5 the subject of flow in heterogeneous systems is studied in further detail. In the first half of this Chapter novel flow measurements in and around the buccal cavity of a Carp are described. These measurements are performed on a standard medical imager without special, fast NMR sequences. The described data can therefore only be used in a qualitative manner. In the second half the line scan flow measurement is introduced. A temporal resolution of 16 ms can be obtained with this sequence allowing accurate, real time flow measurements. The combination of this line scan sequence with displacement imaging yields NMR images which picture the distribution of flow velocities over a line. Demonstrations of displacement imaging are performed in a tube with glass beads and in a pipe with a glass bead filter. The presented data can be used quantitatively.The diffusion and flow measurements described in the Thesis all employ pulsed field gradients to encode for motion. Despite the obvious similarities between the measurements, the optimisation of the NMR sequences results in sequences which can be rather distinct. By careful tuning of the NMR sequences the range of displacements which can thus be measured lies between 5 μm and about 5 cm. This range and the fact that small differences in flow velocities and diffusion constants can be resolved, if necessary using other NMR characteristics, makes NMR a powerful tool to study water mobility in heterogeneous systems