Multiwfn forum / AdDNP dependent on basis set

Re: AdDNP dependent on basis set

dummy@example.com (sobereva) — Thu, 11 Apr 2019 00:06:32 +0000

In principle, the AdNDP result produced under the two basis set should not be so large...Note that the way of picking out AdNDP orbitals highly depends on the user's subjectivity. Without more information, I am difficult to say which result is more reliable; however, the ECP60MWB_SEG in the Stuttgart website should have better quality than the RSC basis set in EMSL, because the former is much newer and has larger size.

PS: The name of the method is AdNDP not AdDNP

Re: AdDNP dependent on basis set

dummy@example.com (iampcd) — Wed, 10 Apr 2019 02:06:09 +0000

The Stuttgart RSC 1997 ECP and associated basis set from https://bse.pnl.gov/bse/portal, it is also the SDD in Gaussian program.

Re: AdDNP dependent on basis set

dummy@example.com (sobereva) — Tue, 09 Apr 2019 12:52:42 +0000

What is the "RSC basis set"? In below page, I can only find ECP60MWB_SEG:
http://www.tc.uni-koeln.de/cgi-bin/pp.p … p=ECP60MWB

AdDNP dependent on basis set

dummy@example.com (iampcd) — Tue, 09 Apr 2019 02:10:59 +0000

Dear Multiwfn users:
Recently, I performed AdDNP analyses for Boron cluster with actinide atoms included, and when the ECP60MWB RECP with RSC basis set was used to describe the actinide atoms, the results of AdDNP analyses are: 36*3c-2e, 12*5c-2e, 8*6c-2e and 1*37-2e, on the other hand, when the ECP60MWB_SEG basis set was used, the results of AdDNP analyses are: 36*3c-2e, 24*5c-2e, 4*6c-2e. It is obviously that the results are very different. My question is which one is more correct?