Figure 8

Experimental curve rescaling. DW-NMR signals of water diffusion coming from two different experiments are rescaled according to Fig. 5 (see “Methods”). Panel (a) diffusion of water inside water. The rescaled curve obtained at different values of the pulse field gradient strengths g, makes it possible to use the Stejskal–Tanner formula (22). The fitted values of D are \(D=2.1\times 10^{-9} \; {\text {m}}^2/{\text {s}}\) for \(g=0.064\) T/m, \(D=2.0\times 10^{-9} {\text {m}}^2/{\text {s}}\) for \(g=0.089\) T/m and \(D=2.1\times 10^{-9} {\text {m}}^2/{\text {s}}\) for \(g=0.127\) T/m with an error of about \(10^{-12}\) in all cases. The \(\zeta\) values are of order of \(10^5\), meaning that the NPG approximation holds in this case \(\delta \zeta \gg 1\). Panel (b) DW-NMR attenuation signals of water molecules diffusing in samples filled up with polystyrene microbeads mixture with nominal average diameters of 6, 10, 40 \(\upmu\)m^37,39. Symbols refer to the values of g in Panel (a). The lack of collapse between the rescaled curves does not allow the use of the anomalous expression (24) for fitting. Inset: the same DW-NMR attenuation signals of panel (b) are plotted versus b value, showing a no solid rescaling.