That's (as always) no= t an easy question. Some things you have to take into account:

1. PCM models were not parametrized = considering that you'll add explicit solvent molecules.

2. But it was also not parametrized considering that there may be expli= cit interactions of the H-bond types.

3. In case you= add explicit solvent molecules, you'll have to think very thoroughly on ho= w to add thermal corrections. Entropy may not be a straightforward value to= add.

4. It is generally not recommended to optimize= using PCM, but just to add the solvent in a single point calculation over = a gas phase geometry.

5. However, if there are big c= harge separations, then better optimize with solvent.

6. Sometimes adding only one molecule of w= ater per electronegative atom is not enough. Water tends to form H-bonded c= lusters around the first solvation shell. Water is a nasty molecule that do= esn't respect quantum chemists.

7. I'm sure ther= e are other issues that I didn't think about.

But if you want a straight answer, and taking into ac= count I know almost nothing of your system, I can tell you what I would do:=

Geometry optimization in gas phase with two explici= t water molecules (one if the model gets to big) per electronegative atom o= f your molecule. Then single point with PCM. And then explain in the paper = that the results are more qualitative than quant= itative.

I believe that you can obtai= n more accurate results than the experimental ones only if your system has a maximum five atoms. For the rest, our work is to gain insight= into reactions, not accuracy. So better have fun with our limitations.

Sebastian.

From: Haya Kornweitz hayak**ariel.ac.il <owner-chemistry(-)ccl.net>
= To: "Kozuch, Sebastian -i= d#3qn-" <kozuchs(-)yahoo.com>
Subject: CCL: solvatation and hydrogen bonds
=

=0A
Sent to CCL by: "Haya Kornweitz" [hayak%x%ariel.ac= .il]
Dear CCls
I have a question about solvatation. I refer to organo= -metallic molecules with N, O atoms solvated in water.
I can think= about 3 methods of calculations:
1.    Optimizing the ge= ometry of the molecule and than using PCM model.
2.    Ad= ding H2O molecules next to each O or N atom in the molecule and than optimi= zing the geometry.
3.    Adding H2O molecules next to eac= h O or N atom in the molecule than optimizing the geometry, and finally usi= ng PCM model.
Please let me know your recommendation concerning these me= thods, or another one.
Thanks
Haya

-=3D This is automa= tically added to each message by the mailing script =3D-
To recover the = email address of the author of the message, please change
the strange ch= aracters on the top line to the (-) sign. You can also
look up the X-Origi= nal-From: line in the mail header.

E-mail to subscribers: CHEMISTRY(-)ccl.net or use:
http://www.ccl.net/cgi-bin/ccl/send_ccl_m= essage

E-mail to administrators: CHEMISTRY-REQUEST(-)cc= l.net or use
http://www.ccl.net/cgi-bin/ccl/= send_ccl_message
=

Before posting, check = wait time at: http://www.ccl.net
Co= nferences: http://server.ccl.net/chemistry/announcements/conferences/
&= nbsp;

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

--1975645123-72540174-1330439399=:12783-- From owner-chemistry@ccl.net Tue Feb 28 11:38:00 2012 From: "Andreas Klamt klamt(a)cosmologic.de" To: CCL Subject: CCL: solvatation and hydrogen bonds Message-Id: <-46399-120228113711-30802-VnEjJzOG14A2t2oK8VzzHQ^server.ccl.net> X-Original-From: Andreas Klamt Content-Transfer-Encoding: 8bit Content-Type: text/plain; charset=ISO-8859-15; format=flowed Date: Tue, 28 Feb 2012 17:37:03 +0100 MIME-Version: 1.0 Sent to CCL by: Andreas Klamt [klamt[]cosmologic.de] Dear Haya, although not completely answering your questions, I guess it should be interesting for you to have a look to our recent extensive study on ~ 2500 HB complexes in COSMO (which is largely equivalent to PCM, especially to C-PCM and IEFPCM). See: Polarization charge densities provide a predictive quantification of hydrogen bond energies Andreas Klamt , Jens Reinisch , Frank Eckert , Arnim Hellweg and Michael Diedenhofen Phys. Chem. Chem. Phys., 2012, 14, 955-963; DOI: 10.1039/C1CP22640A Best regards Andreas Am 28.02.2012 11:52, schrieb Haya Kornweitz hayak**ariel.ac.il: > Sent to CCL by: "Haya Kornweitz" [hayak%x%ariel.ac.il] > Dear CCls > I have a question about solvatation. I refer to organo-metallic molecules with N, O atoms solvated in water. > I can think about 3 methods of calculations: > 1. Optimizing the geometry of the molecule and than using PCM model. > 2. Adding H2O molecules next to each O or N atom in the molecule and than optimizing the geometry. > 3. Adding H2O molecules next to each O or N atom in the molecule than optimizing the geometry, and finally using PCM model. > Please let me know your recommendation concerning these methods, or another one. > Thanks > Haya> > > -- Prof. Dr. Andreas Klamt CEO / Gesch�ftsf�hrer COSMOlogic GmbH& Co. KG Burscheider Strasse 515 D-51381 Leverkusen, Germany phone +49-2171-731681 fax +49-2171-731689 e-mail klamt,cosmologic.de web www.cosmologic.de HRA 20653 Amtsgericht Koeln, GF: Prof. Dr. Andreas Klamt Komplementaer: COSMOlogic Verwaltungs GmbH HRB 49501 Amtsgericht Koeln, GF: Prof. Dr. Andreas Klamt Honarary-Professor at Inst. of Physical and Theoretical Chemistry, University of Regensburg From owner-chemistry@ccl.net Tue Feb 28 12:35:00 2012 From: "Arne Dieckmann adieckma,+,googlemail.com" To: CCL Subject: CCL: solvatation and hydrogen bonds Message-Id: <-46400-120228123340-7535-v6xXTkHzr5CdUy2eDZ1Kiw/a\server.ccl.net> X-Original-From: Arne Dieckmann Content-Transfer-Encoding: 8bit Content-Type: text/plain; charset=us-ascii Date: Tue, 28 Feb 2012 09:33:26 -0800 Mime-Version: 1.0 (Apple Message framework v1257) Sent to CCL by: Arne Dieckmann [adieckma _ googlemail.com] This is indeed a difficult question and there is probably not a single correct answer. Sebastian already mentioned a few very good and important points and I agree with most of those. However, I do not agree with the recommendation of optimizing in the gas phase using one or two explicit water molecules for each electronegative atom. There is no physical reason for using only one or two water molecules and once your number of explicit solvent molecules overcomes a certain threshold you have to sample their configurations/conformations. I would optimize using a PCM for water and compare those structures to gas phase geometries. Afterwards, you can still add a few explicit solvent molecules and study their individual impact on the geometry of your system. As Sebastian also mentioned, this will not result in quantitative, but in qualitative data. For sure, you will not be able to discuss any free energy changes, but only electronic energies or enthalpies. I believe it is always a good idea to compare the results of different strategies for problematic cases like this one. Cheers, Arne - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Dr. Arne Dieckmann Houk Research Lab | University of California, Los Angeles On Feb 28, 2012, at 2:52 AM, Haya Kornweitz hayak**ariel.ac.il wrote: > > Sent to CCL by: "Haya Kornweitz" [hayak%x%ariel.ac.il] > Dear CCls > I have a question about solvatation. I refer to organo-metallic molecules with N, O atoms solvated in water. > I can think about 3 methods of calculations: > 1. Optimizing the geometry of the molecule and than using PCM model. > 2. Adding H2O molecules next to each O or N atom in the molecule and than optimizing the geometry. > 3. Adding H2O molecules next to each O or N atom in the molecule than optimizing the geometry, and finally using PCM model. > Please let me know your recommendation concerning these methods, or another one. > Thanks > Haya> > From owner-chemistry@ccl.net Tue Feb 28 15:49:00 2012 From: "Ron Bakus rbakus===chem.ucsb.edu" To: CCL Subject: CCL:G: Recovery of converged wavefunction/states in Gaussian Message-Id: <-46401-120228154742-14540-bZMeoVZlwuoNRB7XLCznmw,+,server.ccl.net> X-Original-From: "Ron Bakus" Date: Tue, 28 Feb 2012 15:47:40 -0500 Sent to CCL by: "Ron Bakus" [rbakus{:}chem.ucsb.edu] I have a question about about recovery of the converged wavefunction in Gaussian in order to preform additional calculations. Is it possible, after a geometry optimization, to read in from the checkpoint everything you would need to preform a TDDFT without having to go through another SCF cycle at the beginning of the TDDFT? I know that you can use --Link1-- to chain up jobs (eg. OPT then TDDFT), but usually when you do this, the first thing that Gaussian does is to go through an SCF cycle after the link, and that just seems like a waste of time (or is that me being naive). I tried: #p td geom=allcheck guess=tcheck density=checkpoint (with and without specifying b3lyp/6-31g [density should cover this]) But that failed, with output like below: [Geom read in] [Guess read in and translate] [Density read in] [MO printout (note proper energy levels printer here)] 12 initial guesses have been made. ***WARNING States NOT converged!! Guess Only. *********************************************************************** Excited states from singles matrix: *********************************************************************** 1PDM for each excited state written to RWF 633 Ground to excited state transition densities written to RWF 633 Ground to excited state transition electric dipole moments (Au): state X Y Z Dip. S. Osc. 1 8.3701 0.0000 -0.0498 70.0604 0.0000 2 0.0000 3.0076 0.0000 9.0459 0.0000 3 0.0000 4.1215 0.0000 16.9866 0.0000 ____________________ Excited State 1: 1.000-A 0.0000 eV 0.00 nm f=0.0000 =0.000 218 ->219 0.35355 218 <-219 0.35356 This state for optimization and/or second-order correction. Total Energy, E(TD-HF/TD-KS) = 0.000000000000E+00 What am I doing wrong? Thanks, Ron