Data sampling in MR relaxation

Four time spacing methods for the sampling of magnetic resonance relaxation data are investigated: linear, logarithmic, geometric with time offset, and pure geometric spacing methods. They are compared with the respective normalized root-mean-square errors of the four parameters of biexponential inversion recoveries. Linear spacing was found to be inappropriate for NMR. Geometric spacing may be the method of choice to detect an unexpected exponent when sampling is performed from 1/28.7 of the shortest time constant up to five times the longest one.