From owner-chemistry@ccl.net Sat Dec 5 08:44:01 2009 From: "jaleel uc jaleel.uc%x%gmail.com" To: CCL Subject: CCL: computational chemistry curriculum Message-Id: <-40868-091203114020-21547-eDAy5arUVRUdMo0s1jdFiw]~[server.ccl.net> X-Original-From: jaleel uc Content-Type: multipart/alternative; boundary=00504502cb1ded93d40479d5a49d Date: Thu, 3 Dec 2009 22:10:08 +0530 MIME-Version: 1.0 Sent to CCL by: jaleel uc [jaleel.uc##gmail.com] --00504502cb1ded93d40479d5a49d Content-Type: text/plain; charset=ISO-8859-1 *UNIT I -* Introduction to computational chemistry -*Force fields and molecular representations of matter Intramolecular* (bonding) interactions-Non-bonded interactions -Hydrogen bonds-Constraints and Restraints-United atom and other coarse-grained approaches-Non-pairwise interactions*Methods for Simulating Large systems*:*Non-bonded Cutoffs *Shifted potential and shifted force-Switching functions-Neighbor lists*Boundaries -*Periodic Boundary conditions-Stochastic forces at spherical boundary*Long-range interactions -*The Ewald Sum-The Reaction field method *UNIT II -* *Energy Minimization and related analysis techniques* Steepest descent-Conjugate gradient-Newton-Raphson-Comparison of methods-Advanced techniques: Simulated Annealing, Branch-and-bound, simplex-What's the big deal about the minimum anyway? *Free Energy* :Perturbation methods,Thermodynamic integration On 12/3/09, Richard Harper drrwharper,gmail.com wrote: > > I am contemplating introduction of a module on computational chemistry into > chemistry courses; possibly both graduate and undergraduate. Is there any > sort of curriculum standard to guide the content of a course? What should > be the expectations for teaching and learning? > > Thanks in advance, > Dick > -- Dr Jaleel.U.C Asst professor in cheminformatics Malabar christian college calicut --00504502cb1ded93d40479d5a49d Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable

UNIT I - Introduct= ion to computational chemistry -F= orce fields and molecular representations of matter Intramolecular (bon= ding) interactions-Non-bonded interactions -Hydrogen bonds-Constraints and = Restraints-United atom and other coarse-grained approaches-Non-pairwise int= eractionsMethods for Simulating L= arge systems:Non-bonded Cutof= fs Shifted potential and shifted force-Switching functions-Neighbor lis= tsBoundaries -Periodic Bounda= ry conditions-Stochastic forces at spherical boundaryLong-range interactions -The Ewald Sum-The Reaction= field method

=A0

=A0

UNIT II - Energy Minimization and related analysis = techniques Steepest descent-Conjugate gradient-Newton-Raphson-Compariso= n of methods-Advanced techniques: Simulated Annealing, Branch-and-bound, si= mplex-What's the big deal about the minimum anyway? Free Energy :Perturbation methods,Thermodynamic = integration



On 12/3/09, = Richard Harper drrwharper,gmail.com &l= t;owner-chemistry() ccl.net>= ; wrote:
I am contemplating introduction = of a module on computational chemistry into chemistry courses; possibly bot= h graduate and undergraduate.=A0 Is there any sort of curriculum standard t= o guide the content of a course?=A0 What should be the expectations for tea= ching and learning?

Thanks in advance,
Dick


=
--
Dr Jaleel.U.C
Asst=A0=A0professor in cheminformatics
Mala= bar christian college
calicut=20 --00504502cb1ded93d40479d5a49d-- From owner-chemistry@ccl.net Sat Dec 5 16:53:00 2009 From: "Soren Eustis soreneustis[*]gmail.com" To: CCL Subject: CCL:G: Global minima Message-Id: <-40869-091205164131-2035-0Etum3y5MAX41u1rXSmW1A]^[server.ccl.net> X-Original-From: Soren Eustis Content-transfer-encoding: quoted-printable Content-type: text/plain; charset="ISO-8859-1" Date: Sat, 05 Dec 2009 22:41:15 +0100 Mime-version: 1.0 Sent to CCL by: Soren Eustis [soreneustis.[a].gmail.com] Peter, There really is no way to set up a global optimization in Gaussian - believ= e me I have tried. Your converged geometries will be local minima which may or may not be the global minimum. You just need to try as many reasonable geometries as are reasonable for your level of theory. Another option is t= o use Avogadro, or a similar program to perform a simulated annealing at the MM or semi-empirical level, then feed that geometry into Gaussian as a starting point. Quantum annealing, Monte Carlo, or GA methods are all being used in various ways in the literature, but global optimization codes are not at the turn-key level (i.e. implemented in Gaussian). I am always on the lookout for a good program for this or some way to implement it in Gaussian. I have some ideas, but nothing worth mentioning as of yet. Vshift is one method for convergence problems. However, I use that usually as a last resort. Further, increasing the SCF iteration limit beyond 128 i= s rarely helpful. It usually just wastes CPU time. I would try this method (in this order). 1) Change the guess: (guess=3Dindo or guess=3Dhuckel). This works well especially if the SCF if 'confused'. 2) scf=3Dxqc 3) Try getting scf convergence with a smaller basis set. Run a sp calculation, save the output and use guess=3Dread for the real run. Further ideas can be found here: http://www.ccl.net/cca/documents/dyoung/topics-orig/converge.html Regards, Soren --=20 Dr. Soren N. Eustis ETH =AD Zurich Institute for Biogeochemistry and Pollutant Dynamics Universitatstrasse 16 8092 Zurich On 12/4/09 8:21 PM, "Ol Ga eurisco1]~[pochta.ru" wrote: >=20 > Sent to CCL by: "Ol Ga" [eurisco1**pochta.ru] > Dear Peter Khlyabich, >=20 >=20 > 1) First, optimization procedure implemented in Gaussian searches minima = which > can be or can not to be global. You should compare energies of structures= . To > try to prevent convergence to TS (other type of stationary points) you sh= ould > use route section similar to > NoSymm Opt(tight,calcfc,GDIIS) >=20 > Anyway, it is probable that you will find TS point in this case. And you = can > _guess_ that structure is global minima because full scan of PES is impos= sible > now. >=20 > 2) Are your charged structures ion-radicals? If "Yes", use keyword vshift= =3D150 > in SCF group. > SCF(XQC,vshift=3D150) >=20 > Sincerely, >=20 > Ol Ga >=20 >=20 > **************************** > Sent to CCL by: "Peter Khlyabich" [khlyabic() usc.edu] > Hey, >=20 > I have two questions: >=20 > 1) I'm trying to run calculations using Gaussian. In order to check if th= e > obtained structures are the global minima I run frequency calculations. V= ery > often the optimized structures are not in the minima but in saddle point.= Is > there any way to run calculations the way that they won't stop until glob= al > minima is found? >=20 > 2) Also I study charge molecules. This means I study anions and cations. = When > I run calculation very often I get the error that convergence criteria no= t met > and SCF is confused. To avoid this I use SCF=3Dqc or SCF=3Dxqc,MaxCyc=3D1000 bu= t non > of this methods helps. The only way i found to overcome this problem is u= se > IOp(5/6=3D4). But I don't consider this as the best method. Does any one ha= ve > any ideas how to overcome this message error? >=20 > Appreciate your help in advance >=20 > Peter >=20 >=20 >=20 > -=3D This is automatically added to each message by the mailing script =3D->=20>=20>=20>=20>=20>=20>=20>=20>=20 >=20 From owner-chemistry@ccl.net Sat Dec 5 20:07:00 2009 From: "Ryan Kopreski rpkopreski]_[gmail.com" To: CCL Subject: CCL: TINKER MM2.prm Sulfur Params Message-Id: <-40870-091205104930-24452-lsgLLWhuMBXk7xIY/CuGYg(_)server.ccl.net> X-Original-From: "Ryan Kopreski" Date: Sat, 5 Dec 2009 10:49:27 -0500 Sent to CCL by: "Ryan Kopreski" [rpkopreski(!)gmail.com] I am a new user of the TINKER package and would like to optimize a structure containing a Csp2-S-Csp3 moiety using the mm2 parameter set, but I get an error due to missing parameters for the atoms about sulfur. I do not understand some of the columns in the mm2.prm file, but thought I might be able to add the missing params from data I have obtained optimizing similar structures with Chem3D. Could someone help me understand this file? Or better, does anyone have a modified mm2.prm file for this C-S-C atom set? Example interpretation from mm2.prm: atom types ? bond length (A) bond 1 15 3.213 1.815 As you can see, I do not know what is the meaning of column 4 in the "bond" parameter list, and I have not found any information from the web on this yet. I appreciate any clues/suggestions/assistance! Ryan Kopreski rpkopreski(at)gmail(dot)com