From owner-chemistry@ccl.net Sun Nov 25 18:31:00 2012 From: "lars.goerigk.=-=.chem.usyd.edu.au" To: CCL Subject: CCL:G: DFT Methods For Intermolecular Pi-Pi Interactions Message-Id: <-47911-121125180725-13068-OPYBOhNxxiooABKVUCTQww=-=server.ccl.net> X-Original-From: lars.goerigk-$-chem.usyd.edu.au Content-Disposition: inline Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=ISO-8859-1; DelSp="Yes"; format="flowed" Date: Mon, 26 Nov 2012 10:07:10 +1100 MIME-Version: 1.0 Sent to CCL by: lars.goerigk{=}chem.usyd.edu.au Dear Adam, the reason why pure B3LYP is not a good choice for your problem is that common DFT approximations cannot describe London-dispersion interactions correctly. London-dispersion is an electron correlation effect, which is why you do not see any improvements with CAM-B3LYP (or other long-range corrected hybrid functionals). CAM-B3LYP only seperates the exchange into a short- and long-range part, but not the correlation. M05-2X and related methods were fitted to non-covalent interactions and empirically they may work for equilibrium distances, as shown in many applications. However, they do not have the correct asymptotic behaviour of the dispersion contribution and it was argued that those functionals also need so-called London-dispersion corrections, particularly when longer distances are involved: ChemPhysChem 2011, 12, 3421. Stefan Grimme's new DFT-D3 approach is such a correction and it has been tested thoroughly for energetics and geometries. The DFT-D3 correction is combined with your normal DFT result and basically does not cost you any additional computational effort. Currently it can be combined with about 50 DFT methods.It is available in the ORCA and Turbomole codes, but the DFT-D3 program itself can also be downloaded > from Grimme's webiste. Here are some links to DFT-D3 related papers: J. Chem. Phys. 2010, 132, 154104. J. Comput. Chem. 2011, 32, 1456. Phys. Chem. Chem. Phys. 2011, 13, 6670. In terms of the basis set you have to consider that a choice like 6-31G* may lead to wrong results, because of a so-called basis-set superposition error, which leads to an artificial overstabilisation of your complexes. In principle, you should use larger basis sets, but if you can't afford them because of system size, I can recommend another correction by Grimme, which has to be shown to yield very good results when combined with DFT-D3: J. Chem. Phys. 2012, 136, 154101. J. Org. Chem. 2012, DOI: 10.1021/jo302156p. If downloading the respective programs is no option, Grimme's website also offers a service which allows you to calculate the contributions of those corrections for your specific problem. I hope these references can help you. Cheers, Lars -- Dr. Lars Goerigk School of Chemistry (Building F11) The University of Sydney, NSW 2006 Australia Quoting "Adam Langlois Adam.Langlois23/11/12%USherbrooke.ca" : > > Sent to CCL by: "Adam Langlois" [Adam.Langlois||USherbrooke.ca] > I have been assigned a calculations project where I am asked to do a DFT > calculations to optimise the geometry of a porphyrin triamer and hexamer that > interact through Pi-Pi interactions with C60. The interaction is a "ball-in- > glove" type of interaction where there is an overlap of the Pi orbitals of > the porphyrin rings and the Pi orbitals of the C60. > > My question is what DFT method is best suited for this type of calculation? > > For most of the calculations that I have carried out in the past I have used > the B3LYP method. However, I have heard that this is not the best method for > long-distance or intermolecular interactions. > > I have been told about two other methods that may be useful in a situation > like this. The first is the CAM-B3LYP method which is supposed to take long > distance interactions into account. However, I have recently compared the > CAM-B3LYP method to the B3LYP method in calculations that involved a co- > facial special pair of porphyrin and found that the B3LYP method better > represented that experimental data. > > I have also heard of the a method known as the M05-2X functional, which is > supposed to another good method for long distance interactions such as > hydrogen bonding. However, I do not see this functional as an option in the > Gaussian calculation set-up window. > > A second question is with regard to the basis sets. I plan on using the > ONIOM method to save on computation time. I plan on applying the PM6 method > as a low level and a DFT method as a high level. However I am not sure how > large of a basis set I should use. Normally I carry out my calculations with > a 6-31G* basis set. Is this large enough to take into account the > intermolecular interactions or go I need to employ a more disperse basis set > such as 6-311g++. A paper that I found used a 6-311g++(2df,2dp) basis set to > examine intermolecular hydrogen bonding which is a large, and very disperse > basis set. However, the paper also modelled very small structures. Any > suggestions for basis sets here would be appreciated. > > Finally, once I have modelled the triamer and the hexamer without C60 I will > need to model them with C60. Where again I will employ the ONIOM method. My > question here is, do I include the C60 in my high layer or my low layer? > > I would like to thank you in advance for all of your help. > > Adam Langlois > Universite de Sherbooke > Quebec Canada > Adam.Langlois-,-USherbrooke.ca> > > ---------------------------------------------------------------- This message was sent using IMP, the Internet Messaging Program. From owner-chemistry@ccl.net Sun Nov 25 23:32:00 2012 From: "ABHISHEK SHAHI shahi.abhishek1984/a\gmail.com" To: CCL Subject: CCL: covalency and ionicity Message-Id: <-47912-121125231938-3417-K0tjHtO96vva6ELg95eDbw|,|server.ccl.net> X-Original-From: ABHISHEK SHAHI Content-Type: multipart/alternative; boundary=e89a8f2348cf192e2004cf5e3d4d Date: Mon, 26 Nov 2012 09:49:05 +0530 MIME-Version: 1.0 Sent to CCL by: ABHISHEK SHAHI [shahi.abhishek1984[#]gmail.com] --e89a8f2348cf192e2004cf5e3d4d Content-Type: text/plain; charset=ISO-8859-1 Dear All, Is there any computational method to calculate covalency of a particular bond ? In many article, It has been seen that this particular bond have some percentage of covalency and ionicity as well. Will energy decomposition method be helpful here ? What else a bond contains besides covalency and ionicity ? Please give your suggestion considering weak interaction along with covalent bond. Thanks a lot. * Wish you a Merry Christmas and may this festival bring abundant joy and happiness in your life!* Regards : *ABHISHEK SHAHI* *Research Scholar Inorganic and Physical Chemistry Indian Institute Of Science Bangalore-12* --e89a8f2348cf192e2004cf5e3d4d Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Dear All,

=A0=A0=A0=A0=A0 Is there any computational method to calc= ulate covalency of a particular bond ? In many article, It has been seen th= at this particular bond have some percentage of covalency and ionicity as w= ell. Will energy decomposition method be helpful here ? What else a bond co= ntains besides covalency and ionicity ? Please give your suggestion conside= ring weak interaction along with covalent bond. Thanks a lot.

=A0 Wish you a Merry Christmas and may this festi= val bring abundant joy and happiness in your life!=

Regards :
=A0ABHISHEK SHAHI
Research Scholar
Inorga= nic and Physical Chemistry
Indian Institute Of Science
Bangal= ore-12



--e89a8f2348cf192e2004cf5e3d4d--