COMPARISON OF METHODS FOR DERIVING ATOMIC CHARGES FROM THE ELECTROSTATIC POTENTIAL AND MOMENTS

Four methods for deriving partial atomic charges from the quantum chem ical electrostatic potential (CHELP, CHELPG, Merz-Kollman, and RESP) h ave been compared and critically evaluated. It is shown the charges st rongly depend on how and where the potential points are selected. Two alternative methods are suggested to avoid the arbitrariness in the po int-selection schemes and van der Waals exclusion radii: CHELP-BOW, wh ich also estimates the charges from the electrostatic potential, but w ith potential points that are Boltzmann-weighted after their occurrenc e in actual simulations using the energy function of the program in wh ich the charges will be used, and CHELMO, which estimates the charges directly from the electrostatic multipole moments. Different criteria for the quality of the charges are discussed. The CHELMO method gives the best multipole moments for small and medium-sized polar systems, w hereas the CHELP-BOW charges reproduce best the total interaction ener gy in actual simulations. Among the standard methods, the Merz-Kollman charges give the best moments and potentials, but they show an apprec iable dependence on the orientation of the molecule. We have also exam ined the recent warning that charges derived by a least-squares fit to the electrostatic potential normally are not statistically valid. It is shown that no rank-deficiency problems are encountered for molecule s with up to 84 atoms if the least-squares fit is performed using pseu doinverses calculated by singular value decomposition and if constrain ts are treated by elimination. (C) 1998 John Wiley & Sons, Inc.