This is a bug in Gaussian and it exists there for a long time. The radius
information is lost once it is used to calculate ESP. To solve this problem,
you can prepare two-step Gaussian jobs. In the first step, minimization is
performed and the second step single point calculation is performed at the
same level and write out ESP. Of course, the radius information is given in
the second step.
Best
Junmei
On Sat, Oct 31, 2009 at 7:10 AM, megha bajaj <micro.megha.gmail.com> wrote:
> Hi,
>
> I used antechamber to generate input file (.com file) for gaussian using:
>
> $ antechamber -i 16.mol2 -fi mol2 -o mol16.com -fo gcrt
>
> Then manually changed the basis set to B3LYP/LANL2DZ and added
> Pop=(MK,ReadRadii) to set the Hg radius in Gaussian.
>
> --Link1--
> %chk=mol16
> #B3LYP/LANL2DZ SCF=tight Test Pop=(MK,ReadRadii) iop(6/33=2) opt
> remark line goes here
> 0 1
> C -0.0169 1.3921 0.0097
> C 1.1701 2.0948 0.0021
> C 2.3776 1.4193 -0.0135
> C 2.4036 0.0359 -0.0216
> C 1.2241 -0.6793 -0.0137
> C 0.0021 -0.0041 0.0020
> C -1.2651 -0.7633 0.0110
> O -1.2455 -1.9779 0.0047
> O -2.4424 -0.1088 0.0262
> Hg 4.1790 2.4986 -0.0256
> O 5.8517 3.5009 -0.0368
> H -0.9586 1.9205 0.0260
> H 1.1570 3.1747 0.0081
> H 3.3496 -0.4851 -0.0345
> H 1.2458 -1.7591 -0.0205
> H -3.2719 -0.6058 0.0321
> H 6.5671 2.8505 -0.0474
> Hg 2.00
>
> Then I used this com file to generate log file (attached herewith) and
> error
> came out as:
>
> No NMR shielding tensors so no spin-rotation constants.
> Leave Link 601 at Sat Oct 31 17:44:44 2009, MaxMem= 6291456 cpu:
> 0.2
> Merz-Kollman atomic radii used.
> Read replacement radii for each type of atom:
> Atom Element Radius
> 1 6 1.50
> 2 6 1.50
> 3 6 1.50
> 4 6 1.50
> 5 6 1.50
> 6 6 1.50
> 7 6 1.50
> 8 8 1.40
> 9 8 1.40
> 10 80 0.00
> GetVDW: no radius for atom 10 atomic number 80.
> Error termination via Lnk1e in /opt/n1ge6/gaussian/tar/g03/l602.exe at Sat
> Oct 31 17:44:44 2009.
> Job cpu time: 0 days 0 hours 23 minutes 38.7 seconds.
> File lengths (MBytes): RWF= 21 Int= 0 D2E= 0 Chk= 7
> Scr= 1
>
> Can anyone help me out??
>
> Thanks,
> Megha
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Nov 02 2009 - 08:00:02 PST
