Diffusion measurements at long observation times in the presence of spatially variable internal magnetic field gradients

Diffusion measurements in heterogeneous media may contain a significant source of error, the influence of the coupling between the applied and internal magnetic field gradients on the attenuation of the NMR signal. The application of bipolar magnetic field gradients has been introduced to suppress this error. The basic assumption for the successful removal of the coupling is that the diffusing molecules are experiencing a constant internal gradient during the experiment. We will provide theoretical and experimental evidence that the application of bipolar magnetic field gradients may fail to suppress the effect from all the cross terms between internal and applied gradients effectively at long observation times. It will be shown experimentally that a successful suppression of the cross terms is strongly dependent on the observation time, and on the tau value in the bipolar pulsed field gradient stimulated echo experiment. (C) 2000 Academic Press.