Greetings Francois!
Here is what I got using ANTECHAMBER to get the resp charges.
I used as you suggested, a higher level of theory(B3YLP and so on.)
obtaining a planar geometry for oxalate...
CHARGE -2.00
Formula: C2 O4
ATOM 1 C1 OXL 1 0.000 0.000 0.821 0.727276 c
ATOM 2 C2 OXL 1 0.000 0.000 -0.821 0.727276 c
ATOM 3 O1 OXL 1 -0.001 1.124 1.392 -0.863638 o
ATOM 4 O2 OXL 1 -0.001 -1.124 -1.392 -0.863638 o
ATOM 5 O3 OXL 1 0.001 -1.124 1.392 -0.863638 o
ATOM 6 O4 OXL 1 0.001 1.124 -1.392 -0.863638 o
BOND 1 1 2 0 C1 C2
BOND 2 1 3 0 C1 O1
BOND 3 1 5 0 C1 O3
BOND 4 2 4 0 C2 O2
BOND 5 2 6 0 C2 O4
The PC GAMESS version is broken :(, only got the one for chemoffice
...and the source code gave me a headache.
Oh my!.
Here it is the output from GAMESS for the RESP charges:
0.735501 0.732186 -0.867237 -0.866606 -0.867237 -0.866606
and here is the p2n input
REMARK
REMARK TITLE MOLECULE
REMARK CHARGE-VALUE -2
REMARK MULTIPLICITY-VALUE 1
REMARK
ATOM 1 C1 OXL 557 0.747 0.000 0.002 C1
ATOM 2 C2 OXL 557 -0.747 0.000 0.000 C2
ATOM 3 O3 OXL 557 1.385 -1.105 0.001 O1
ATOM 4 O4 OXL 557 -1.385 -1.105 -0.002 O2
ATOM 5 O5 OXL 557 1.385 1.105 -0.002 O3
ATOM 6 O6 OXL 557 -1.385 1.105 0.001 O4
CONECT 1 2 3 5
CONECT 2 1 4 6
CONECT 3 1
CONECT 4 2
CONECT 5 1
CONECT 6 2
END
I was wondering what is the mistake you said I had on the files . . .
Okay Francois, now I am a believer.
Happy Oxalates for everybody.
Eduardo
FyD wrote:
> Quoting Eduardo Mendez <pckboy.gmail.com>:
>
>> Geometry optimization(s) is/are being computed for molecule 1 ...
>> [ FAILED ]
>> See the file(s) "JOB1-gam_m1-1.log"
>>
>> Any idea why the "Geometry optimization" is not working?.
>
> I tested your GAMESS output file & it looks OK for me. However, there
> is something that looks strange to me at the beginning of your QM output:
>
> cp OXL.inp .//OXL.F05
> unset echo
> setenv ERICFMT .//ericfmt.dat
> setenv MCPPATH .//mcpdata
> [..]
>
> My understanding is that OXL.F05 should be in your scratch directory.
> So if you copy OXL.inp into .//OXL.F05. Does it mean that your
> 'working' directory is your 'scratch' directory ?
>
> Here is what I get when I run GAMESS on my machine:
> [fyd.localhost ~/GAMESS]$ rungms test > test.log
> cp test.inp /usr/local/0QM_SCR/test.F05
> unset echo
> setenv ERICFMT /usr/local/gamess/ericfmt.dat
> setenv EXTBAS /dev/null
> setenv IRCDATA /usr/local/0QM_SCR/test.irc
> [..]
>
> I run GAMESS from the "/home/fyd/GAMESS" working directory & my
> scratch directory is at "/usr/local/0QM_SCR/"
>
> Or does it mean that in the last GAMESS version the "rungms" script
> has changed ?
> I tried to get the last version of GAMESS without success since the
> GAMESS MySQL/PHP download system seems broken. Did you modify the
> "rungms" script, can you send me your "rungms" script ?
>
> BTW, you might consider using PC-GAMESS (if your machine is a PC)
> since it looks really faster than GAMESS-US.
>
>> Attached is my output and p2n file.
>
> Your P2N file is not correct. Please, see the one I sent you. Your 1st
> column of atom name is not correct: The four oxygen atoms should be
> chemically equivalent. The two carbons should also be equivalent.
>
> Please read also the tutorial .
> http://q4md-forcefieldtools.org/Tutorial/Tutorial-1.php#2
> Read in particular the sections:
> - General information about charge fitting
> - Rules followed by the R.E.D. program for automatic RESP input
> generation
> - Limitations of the R.E.D. versions I and II and creation of the P2N
> file format
> And, see an example of P2N file .
> http://q4md-forcefieldtools.org/Tutorial/Tutorial-1.php#ATOM-NAME2
>
> regards, Francois
>
>
> -----------------------------------------------------------------------
> The AMBER Mail Reflector
> To post, send mail to amber.scripps.edu
> To unsubscribe, send "unsubscribe amber" to majordomo.scripps.edu
>
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber.scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo.scripps.edu
Received on Sun Sep 23 2007 - 06:07:25 PDT
