Among all the schemes available to calculate charges (inside and outside SIESTA), the Mulliken is among the worses.
Try the other two schemes. []'s, @mps On Fri, Aug 14, 2015 at 6:50 AM, Radhika Narayanan < [email protected]> wrote: > Dear Camps, > > Thank you for your reply. Following is the Mulliken charge analysis of the > carboxyl group in an ester > > C1 3.445 > *C3* 4.213 > *O4* 5.983 > O2 5.870 > > where C3 and O4 are the carbonyl carbon and oxygen atoms. Why is the > mulliken charge positive for the oxygen atoms? In a similar calculation > using Gaussian 09, the oxygen atoms are having negative charge. > > Further, the carbonyl carbon is having higher value (C3) than the other > carbon atom (C1). Does it mean that mulliken charge analysis is wrong in my > molecule? > > Thanks > Radhika > > > On Fri, Aug 14, 2015 at 6:12 PM, I. Camps <[email protected]> wrote: > >> SIESTA has implemented various method to calculate the charges. There are: >> - *Mulliken* (section 6.15.1 of the manual for version siesta-trunk-462) >> - *Voronoi and Hirshfeld* (section 6.15.2 of the manual) >> >> You should calculate the charges for both systems using the same >> parameters (type of pseudo, basis set, mesh cutoff, etc.) >> >> >> >> >> []'s, >> >> @mps >> >> On Fri, Aug 14, 2015 at 6:02 AM, Radhika Narayanan < >> [email protected]> wrote: >> >>> Dear all, >>> >>> Could anyone suggest how to compare the charge distribution on an >>> isolated molecule and a molecule in a crystal using the Siesta software? >>> >>> Thanks >>> Radhika >>> >>> -- >>> Research fellow, >>> Theoretical and Computational Chemistry Lab, >>> Department of Chemistry, >>> Tokyo Metropolitan University, >>> 1-1 Minami-Osawa, Hachioji-shi, Tokyo,Japan,192-0397 >>> E-mail:[email protected] >>> >> >> > > > -- > Research fellow, > Theoretical and Computational Chemistry Lab, > Department of Chemistry, > Tokyo Metropolitan University, > 1-1 Minami-Osawa, Hachioji-shi, Tokyo,Japan,192-0397 > E-mail:[email protected] >
