[AMBER] DFTB3 level energies calculated by sqm and Gaussian are very different

From: Suitian Lai via AMBER <amber.ambermd.org>
Date: Sun, 8 Oct 2023 23:36:15 +0800

Hi,
I have a system containing 3 atoms as QM atoms and 3 atoms as point charges. I calculated the single point energy of this system in DFTB3 level by sqm and Gaussian, respectively.
The “SCF Energy” in the output of sqm was -0.38 a.u.. The “SCF Done: E(RDFT-SCTB)” in the output of Gaussian was -4.06 a.u..
Can some one tell me why are the two results so far apart ?

The sqm input is
single point energy calculation
 &qmmm
  qm_theory = 'DFTB3',
  qmcharge = 0,
  maxcyc = 0,
  qmmm_int = 1,
/
8 O 56.826 46.730 28.825
1 H 57.129 47.502 28.347
1 H 57.303 46.761 29.655
#EXCHARGES
8 O 56.665 44.677 27.123 -0.834
1 H 56.734 45.319 27.830 0.417
1 H 56.974 43.859 27.511 0.417
#END

The Gaussian input is
%chk=gau_job0.chk
%NProcShared=16
%mem=30GB
#P DFTB SCF=(Conver=8) NoSymm Force Charge Prop=(Field,Read)

Gaussian run using SANDER external interface.

0 1
O 56.826 46.730 28.825
H 57.129 47.502 28.347
H 57.303 46.761 29.655

56.665 44.677 27.123 -0.834
56.734 45.319 27.830 0.417
56.974 43.859 27.511 0.417

.H-H.prm
.O-O.prm
.O-H.prm
.H-O.prm

从 Windows 版邮件发送

_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Sun Oct 08 2023 - 09:00:02 PDT
Custom Search