From owner-chemistry@ccl.net Sat Mar 16 13:49:00 2013 From: "Sergio Manzetti sergio.manzetti#%#gmx.com" To: CCL Subject: CCL:G: Divergence in metal cluster runs Message-Id: <-48416-130316134545-22381-rQYIQSi8imYFWOIGX7I61g . server.ccl.net> X-Original-From: "Sergio Manzetti" Content-Type: multipart/alternative; boundary="========GMXBoundary213551363455937499875" Date: Sat, 16 Mar 2013 18:45:37 +0100 MIME-Version: 1.0 Sent to CCL by: "Sergio Manzetti" [sergio.manzetti.:.gmx.com] --========GMXBoundary213551363455937499875 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 8bit Dear all, running two cluster types in gaussian, it is nearly impossible to reach convergence. Even though these are only 8 and 5 atoms, it does not converge eitehr with BLYP, MPWPW91, PWPW91 or B3LYP. Can it help in adding PBC? Are there other approaches to reach convergence for difficult metal clusters? Sergio --========GMXBoundary213551363455937499875 Content-Type: text/html; charset="utf-8" Content-Transfer-Encoding: quoted-printable Dear all= , running two cluster types in gaussian, it is nearly impossible to reach c= onvergence. Even though these are only 8 and 5 atoms, it does not converge = eitehr with BLYP, MPWPW91, PWPW91 or B3LYP.

Can it help in addin= g PBC? Are there other approaches to reach convergence for difficult metal = clusters?

Sergio --========GMXBoundary213551363455937499875-- From owner-chemistry@ccl.net Sat Mar 16 16:56:00 2013 From: "Reinaldo Pis Diez reinaldo.pisdiez*o*gmail.com" To: CCL Subject: CCL:G: Divergence in metal cluster runs Message-Id: <-48417-130316162859-3329-IAgnd2A2Q0yBfuQHGnanKQ(-)server.ccl.net> X-Original-From: Reinaldo Pis Diez Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=UTF-8; format=flowed Date: Sat, 16 Mar 2013 17:28:45 -0300 MIME-Version: 1.0 Sent to CCL by: Reinaldo Pis Diez [reinaldo.pisdiez{:}gmail.com] Dear Sergio, I don't know which basis sets you're using in your calculations. You didn't mention if you're dealing with transition or main group metals. The former are much more difficult to treat due to the large amount of electronic states found in a relatively small energy range. You should give a try to the local M06-L functional and a pseudopotential like LANL08 (or LANL2TZ, they are the same basis set, differing only in some contraction schemes). I've found very good SCF convergence with the above method. Hope this helps. Regards, Reinaldo On 03/16/2013 02:45 PM, Sergio Manzetti sergio.manzetti#%#gmx.com wrote: > Dear all, running two cluster types in gaussian, it is nearly > impossible to reach convergence. Even though these are only 8 and > 5 atoms, it does not converge eitehr with BLYP, MPWPW91, PWPW91 or > B3LYP. > > Can it help in adding PBC? Are there other approaches to reach > convergence for difficult metal clusters? > > Sergio From owner-chemistry@ccl.net Sat Mar 16 21:17:00 2013 From: "Vadim Debinsky tdebinsky^gmail.com" To: CCL Subject: CCL: Fundamental Question - Raman vs IR for Crystals Message-Id: <-48418-130316184502-15142-EDGcmBdDGcxuOgjYbhqXxw%x%server.ccl.net> X-Original-From: "Vadim Debinsky" Date: Sat, 16 Mar 2013 18:45:01 -0400 Sent to CCL by: "Vadim Debinsky" [tdebinsky.###.gmail.com] Dear CCl'ers I have some fundamental question regarding optical vibrational spectroscopy which I cannot answer to myself. I am wondering what is the reason that the theoretically predicted Raman spectra stay always in better agreement with solid-state experimental data than IR? I have never ever faced any opposite case. Particullary, please look at some inorganic systems. For example at any vibrational spectra of perovskites or any other case. - Obviously, in general, the IR spectra are more rich in bands than RS. - IR spectroscopy as well as RS (first order) are limited to the zone centre phonons, hence we probe only gamma point with no dispersion. Hence there is no principal difference (as in the case of Neutron Spectroscopy, where the modes are averaged over the Brillouin Zone). So what is the answer? Why even for the simplest systems we cannot find as good agreement as in the case of RS? - RS probe only one molecule per few millions, so it is more 'local technique', while in IR experiment all the system is absorbing simultaneously. Hence, is the experimental spectrum somehow more disturbed? - Another question, or maybe the same question is - why the IR bands are generally visibly broader that RS? What may be the reason? Some long-range transition dipole couplings which are nor present in RS? Do you know any other case where RS bands are also so broad? Obviously we are talking about crystals, not amorphic systems or molecules in vacuo. Could anyone help me please? Regards, Vadim