From owner-chemistry@ccl.net Sun Aug 12 01:49:01 2007 From: "Aleksandra Rudnitskaya aleksandra.rudnit001#umb.edu" To: CCL Subject: CCL: solvent calc. Message-Id: <-34919-070812012857-15627-QhgOWzbkhsGzNuVJOPltqw^^server.ccl.net> X-Original-From: "Aleksandra Rudnitskaya" Date: Sun, 12 Aug 2007 01:28:53 -0400 Sent to CCL by: "Aleksandra Rudnitskaya" [aleksandra.rudnit001##umb.edu] Thanks a lot to everyone who sent his answer. I realized I took a wrong example. What if I choose the other system with benzene as a solvent? The questions are the same - what should I write in input file for optimization, and TS? Thank you Aleksandra >Dear CCL members, > >Ive just started calculations with solvent. I decided to start from reaction of neutralization NaOH+HCl=H2O+NaCl. As I understand I should calculate >1) optimization sodium hydroxide in solvent (water); >2) optimization hydrochloric acid in solvent; >3) optimization sodium chloride in solvent; >4) transition state again in solvent. >Idea is very similar to that when I calculate rxns in gas phase. Is it correct? >When I calculate stable species without any solvent I use > opt freq=noraman RHF/6-31G(d) >For TS calculations I use > RHF/6-31G(d) Opt(TS,CalcFC,noeigentest) Freq=noraman > >and then charge and multiplicity, and simple Z-matrix in specification part. How should input files look for optimization and searching TS? I know I have to use keyword SCRF, and one of those (SCRF=Dipole, PCM, DPCM, CPCM or IEFPCM), then specify solvent, and probably, dielectric constant somewhere in specification part after matrix. If you may guide me how to create input file itll help a lot. > >Thank you very much. From owner-chemistry@ccl.net Sun Aug 12 10:01:01 2007 From: "Ol Ga eurisco1**pochta.ru" To: CCL Subject: CCL:G: Subject: CCL: Re No analytic 2nd derivatives for this method Message-Id: <-34920-070812053623-12343-KGPirDs7pYFYj/qRh6cP8g(0)server.ccl.net> X-Original-From: "Ol Ga" Date: Sun, 12 Aug 2007 05:36:18 -0400 Sent to CCL by: "Ol Ga" [eurisco1{}pochta.ru] Dear Alvyn Liang! Keyword CalcAll demands calculation 2nd-derivatives in each point for which energy has been calculated. Gaussian does attempt to calculate 2nd-derivatives analytically. But for the given method it appears it is impossible. In this case it is possible to calculate 2nd-derivatives numerically, using a corresponding method of optimization: Link Function L103 Berny optimizations to minima and TS, STQN transition state searches L106 Numerical differentiation of forces/dipoles to obtain polarizability/hyperpolarizability L110 Double numerical differentiation of energies to produce frequencies L111 Double num. diff. of energies to compute polarizabilities and hyperpolarizabilities L113 EF optimization using analytic gradients L114 EF numerical optimization (using only energies)!!!!!!!! If the keyword is absent CalcAll 2nd-derivatives in the first point it is evaluated, and for the subsequent points hessian is not updated. Basis set has no relation to process of optimization and removal + cannot change anything (try attentively still time and you are convinced of this). Send My e-mail for another discussion. Best Regards. --------- Can anyone tell me why I cannot calculate diffuse function in rob3lyp with keyword 'opt=calcall'.