From owner-chemistry #at# ccl.net Tue Sep 23 08:19:00 2014 From: "Henrique Junior henriquecsj[]gmail.com" To: CCL Subject: CCL:G: MOPAC - Problem with Iron complex and gradients Message-Id: <-50524-140923081542-3812-No4elLgUDRgLzjLKqbqcbg]_[server.ccl.net> X-Original-From: Henrique Junior Content-Type: multipart/alternative; boundary=089e0149c0b006131d0503ba860a Date: Tue, 23 Sep 2014 09:14:50 -0300 MIME-Version: 1.0 Sent to CCL by: Henrique Junior [henriquecsj|gmail.com] --089e0149c0b006131d0503ba860a Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Hi, I'm very sorry for my delay (I was having connection issues). I'll give it a try and share my results. Thank you. 2014-09-17 10:26 GMT-03:00 Henrique Castro Silva Junior henriquecsj|,| gmail.com : > > Sent to CCL by: "Henrique Castro Silva Junior" [henriquecsj**gmail.com] > Hi, I am an undergraduate in Chemistry from Brazil, writing my thesis as > a requirement to receive my graduation. Im working in thermodynamic (and > MO) descriptions that takes place in [Fe(OH2)6]2+ --> [Fe(OH2)6]3+, > mostly using computational chemistry softwares. > I have decided to use MOPAC2012 and Gabedit (as a GUI), but Im having > problems in getting realistic values. Here is my input: > > * =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D > * Input file for Mopac > * =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D > PM7 THERMO ROT=3D1 EPS=3D78.39 RSOLV=3D1.3 UHF CHARGE=3D2 Quintet BONDS= AUX > Mopac file generated by Gabedit > Fe -0.1136 1 -0.9091 1 0.0000 1 > O -0.1136 1 0.9109 1 0.0000 1 > H 0.5055 1 1.2122 1 -0.7127 1 > O 1.7064 1 -0.9091 1 0.0000 1 > H 2.0628 1 -1.3124 1 0.8307 1 > O -0.1136 1 -0.9091 1 -1.8200 1 > H -0.4817 1 -1.7682 1 -2.1477 1 > O -0.1136 1 -2.7291 1 0.0000 1 > H -0.8413 1 -3.0457 1 0.5936 1 > O -1.9336 1 -0.9091 1 0.0000 1 > H -2.2569 1 -1.5900 1 -0.6429 1 > O -0.1136 1 -0.9091 1 1.8200 1 > H -1.0440 1 -0.9414 1 2.1566 1 > H -2.2613 1 -0.0487 1 -0.3649 1 > H 0.7122 1 -3.0886 1 0.4108 1 > H -0.7542 1 -0.2257 1 -2.1427 1 > H 2.0192 1 -1.5032 1 -0.7289 1 > H 0.3089 1 -1.7285 1 2.1807 1 > H 0.3233 1 1.2180 1 0.8341 1 > > The result is this: > > GRADIENT NORM =3D 298.62520 > ** GRADIENT IS TOO LARGE TO ALLOW FORCE MATRIX TO BE CALCULATED, > (LIMIT=3D10) ** > EITHER ADD 'LET' OR REDUCE GRADIENT > USING 'TS' OR OTHER GEOMETRY OPTIMIZER > > If I add the keyword LET the calculation goes on, returning this: > HEAT OF FORMATION =3D -39.019942 KCALS/MOLE > > With bond distance of ~1,82 A (much less than what is expected, right?). > Then, I tried to perform an optimization first, to reduce the gradient > and do not use LET: > PM7 XYZ EPS=3D78.39 RSOLV=3D1.3 UHF CHARGE=3D2 Quintet BONDS AUX > > The result is a fragmentation, [Fe(OH)6] + 6H and the same result (the > fragmentation) goes with Gaussian09. > Any help will be very much appreciated and Im sorry if Im making some > obvious and stupid mistake here, but most of my knowledge on > computational chemistry Ive learned by myself. > > > > -=3D This is automatically added to each message by the mailing script = =3D-> > > --=20 *Henrique C. S. Junior* Qu=C3=ADmica Industrial - UFRRJ Centro de Processamento de Dados - PMP --089e0149c0b006131d0503ba860a Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable
Hi, I'm very sorry for my delay (I was having connecti= on issues).
I'll give it a try and share my results.

=
Thank you.

2014-09-17 10:26 GMT-03:00 Henrique Castro Silva Junior he= nriquecsj|,|gmail.com &l= t;owner-chemis= try#ccl.net>:

Sent to CCL by: "Henrique Castro Silva Junior" [henriquecsj**gmail.com]
Hi, I am an undergraduate in Chemistry from Brazil, writing my thesis as a requirement to receive my graduation. Im working in thermodynamic (and MO) descriptions that takes place in [Fe(OH2)6]2+ --> [Fe(OH2)6]3+,
mostly using computational chemistry softwares.
I have decided to use MOPAC2012 and Gabedit (as a GUI), but Im having
problems in getting realistic values. Here is my input:

* =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D
* Input file for Mopac
* =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D
PM7 THERMO ROT=3D1=C2=A0 EPS=3D78.39 RSOLV=3D1.3 UHF CHARGE=3D2 Quintet=C2= =A0 BONDS AUX
Mopac file generated by Gabedit
Fe=C2=A0 -0.1136=C2=A0 1 -0.9091=C2=A0 1 0.0000=C2=A0 1
O=C2=A0 -0.1136=C2=A0 1 0.9109=C2=A0 1 0.0000=C2=A0 1
H=C2=A0 0.5055=C2=A0 1 1.2122=C2=A0 1 -0.7127=C2=A0 1
O=C2=A0 1.7064=C2=A0 1 -0.9091=C2=A0 1 0.0000=C2=A0 1
H=C2=A0 2.0628=C2=A0 1 -1.3124=C2=A0 1 0.8307=C2=A0 1
O=C2=A0 -0.1136=C2=A0 1 -0.9091=C2=A0 1 -1.8200=C2=A0 1
H=C2=A0 -0.4817=C2=A0 1 -1.7682=C2=A0 1 -2.1477=C2=A0 1
O=C2=A0 -0.1136=C2=A0 1 -2.7291=C2=A0 1 0.0000=C2=A0 1
H=C2=A0 -0.8413=C2=A0 1 -3.0457=C2=A0 1 0.5936=C2=A0 1
O=C2=A0 -1.9336=C2=A0 1 -0.9091=C2=A0 1 0.0000=C2=A0 1
H=C2=A0 -2.2569=C2=A0 1 -1.5900=C2=A0 1 -0.6429=C2=A0 1
O=C2=A0 -0.1136=C2=A0 1 -0.9091=C2=A0 1 1.8200=C2=A0 1
H=C2=A0 -1.0440=C2=A0 1 -0.9414=C2=A0 1 2.1566=C2=A0 1
H=C2=A0 -2.2613=C2=A0 1 -0.0487=C2=A0 1 -0.3649=C2=A0 1
H=C2=A0 0.7122=C2=A0 1 -3.0886=C2=A0 1 0.4108=C2=A0 1
H=C2=A0 -0.7542=C2=A0 1 -0.2257=C2=A0 1 -2.1427=C2=A0 1
H=C2=A0 2.0192=C2=A0 1 -1.5032=C2=A0 1 -0.7289=C2=A0 1
H=C2=A0 0.3089=C2=A0 1 -1.7285=C2=A0 1 2.1807=C2=A0 1
H=C2=A0 0.3233=C2=A0 1 1.2180=C2=A0 1 0.8341=C2=A0 1

The result is this:

GRADIENT NORM =3D 298.62520
=C2=A0** GRADIENT IS TOO LARGE TO ALLOW FORCE MATRIX TO BE CALCULATED,
(LIMIT=3D10) **
=C2=A0EITHER ADD 'LET' OR REDUCE GRADIENT
=C2=A0USING 'TS' OR OTHER GEOMETRY OPTIMIZER

If I add the keyword LET the calculation goes on, returning this:
=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 HEAT OF FORMATION =3D=C2=A0 =C2=A0 =C2= =A0-39.019942 KCALS/MOLE

With bond distance of ~1,82 A (much less than what is expected, right?). Then, I tried to perform an optimization first, to reduce the gradient
and do not use LET:
PM7 XYZ=C2=A0 EPS=3D78.39 RSOLV=3D1.3 UHF CHARGE=3D2 Quintet=C2=A0 BONDS AU= X

The result is a fragmentation, [Fe(OH)6] + 6H and the same result (the
fragmentation) goes with Gaussian09.
Any help will be very much appreciated and Im sorry if Im making some
obvious and stupid mistake here, but most of my knowledge on
computational chemistry Ive learned by myself.



-=3D This is automatically added to each message by the mailing script =3D-=
E-mail to subscribers: CHEMISTRY#ccl.n= et or use:
=C2=A0 =C2=A0 =C2=A0 http://www.ccl.net/cgi-bin/ccl/send_ccl_message=

E-mail to administrators: CHEM= ISTRY-REQUEST#ccl.net or use
=C2=A0 =C2=A0 =C2=A0 http://www.ccl.net/cgi-bin/ccl/send_ccl_message=

Subscribe/Unsubscribe:
=C2=A0 =C2=A0 =C2=A0 http://www.ccl.net/chemistry/sub_unsub.shtml

Before posting, check wait time at: http://www.ccl.net

Job: http://www.ccl.n= et/jobs
Conferences: http://server.ccl.net/chemistry/announcements/co= nferences/

Search Messages: http://www.ccl.net/chemistry/searchccl/index.shtml
=C2=A0 =C2=A0 =C2=A0 http://www.ccl.net/spammers.txt

RTFI: http://www.ccl.net/chemistry/aboutccl/instructions/





--
Henr= ique C. S. Junior
Qu=C3=ADmica Industrial - UFRRJ
Centro d= e Processamento de Dados - PMP
--089e0149c0b006131d0503ba860a--