Comment on ‘‘Density functional calculation of nuclear magnetic resonance chemical shifts’’ [J. Chem. Phys. 102, 2806 (1995)].

van Wüllen has correctly commented on the lack of gauge invariance of our sum-over-states density functional perturbation theory (SOS-DFPT) method for NMR chemical shifts. It is shown that, for practical basis sets, the improvement of the SOS-DFPT method over an uncoupled approach is greater than the difference in gauge dependence between these two methods. © 1996 American Institute of Physics. [ABSTRACT FROM AUTHOR]