From chemistry-request@ccl.net Wed Jan 29 07:37:34 1992 Date: Wed, 29 Jan 92 13:16 N From: Subject: POSITION OFFERED To: chemistry@ccl.net Status: R =============================================================================== OFFER FOR A POST-DOCTORAL RESEARCH COLLABORATOR --------------------------------------------------------- SIMULATION OF NON-STATIONNARY PROCESSES AT REDOX POLYMER COATED ELECTRODES -------------------------------------------------------------------------- DESCRIPTION OF THE PROJECT: --------------------------- The objectives of the project are the application of numerical methods to the simulation of electrochemical mechanisms involved in non-stationnary process at redox polymer coated electrodes encountred in electrochemical measurments e.g cyclic voltammetry, chronoamperometry, ac-impedance, beam deflection, etc.... The computer program we have developed so far, calculates the concentrations profiles and fluxes of all species and potential profile as well. Within this model we yield the current and many experimental quantities as a function of the electrode potential and of the hydrodynamical diffusion layer at the different scan points. In the future this software should be developed into a userfriendly package suitable for a computer assisted moddeling of electrochemical data and finally to design of electrochemical cells with thin layer. DURATION OF THE PROJECT: ------------------------ Two years, starting in the spring 1992. Salary: Legal wages practiced at the University of Fribourg. ADDRESS: -------- Claude Auguste DAUL Institut de Chimie Inorganique et Analytique de Fribourg, Prolles CH-1700 FRIBOURG SWITZERLAND. Telephone: (41) 037/826436 (Office) or (41) 037/826422 (secretary). Electronic mail: CDAUL@CFRUNI52 Telefax: (41) (37) 826531 From chemistry-request@ccl.net Wed Jan 29 09:23:41 1992 Date: Wed, 29 Jan 92 08:22:53 EST From: m10!frisch@uunet.UU.NET (Michael Frisch) Subject: Re: To: chemistry@ccl.net Status: R uunet!OHSTVMA.ACS.OHIO-STATE.EDU!CUNDARIT%MEMSTVX1.BITNET commented about transition metal SCF calculations. I'd like to respond to one comment: He says: Although somewhat biased (having worked with the North Dakota State GAMESS bunch), my experience has been that Gaussian is more prone to convergence problems for transition metal complexes than GAMESS. Recently, new initial guess routines have been written for GAMESS which in our experience has virtually eliminated convergence problems even for notorious cases: first row metal complexes, coordinatively unsaturated complexes, high spin molecules. Sorry, if this sounds like an ad, but one "trick" would be to get a copy of GAMESS, hope for convergence and then read these MOs into your other favorite program as an initial guess. He doesn't say which version of Gaussian -- the last one acquired at North Dakota was Gaussian 86, so if that's what he refers to, he's comparing the latest changes to GAMESS with an ancient version of Gaussian. This is not a very helpful comparison; for example, not only is the extrapolation different in more recent versions, but most of the methods recommended for difficult cases (such as QCSCF) weren't even present in G86. I think a general discussion of SCF convergence issues in this this forum would be quite useful, especially if specific cases are presented. However, these general criticisms without specifying even what program is referred to (nor well what systems) are neither helpful nor fair. Michael Frisch Gaussian, Inc. ------- From chemistry-request@ccl.net Wed Jan 29 12:14:11 1992 Date: Wed, 29 Jan 92 10:37:58 CST From: Theresa Windus Organization: North Dakota Higher Education Computer Network Subject: NDSU To: CHEMISTRY@ccl.net Status: R Actually, we have G88 at North Dakota State University. What versions of Gaussian have the new convergence routines? Theresa Windus North Dakota State University Fargo, North Dakota 58104 NU131691@NDSUVM1 From chemistry-request@ccl.net Wed Jan 29 13:55:09 1992 From: fant@umbc3.umbc.edu (Andrew Fant) Subject: Saddle Points in MOPAC and AMPAC To: chemistry@ccl.net Date: Wed, 29 Jan 92 13:35:44 EST Status: R Greetings I have been trying to use the saddle point locator in MOPAC 6.0 to find a transition state and have come up against the dread "This message should never appear" error message. I found a rough geometry of the transition state of the reaction by a quick and dirty search, and then tightened up the criteria and went looking with the TS eigenvector-following geometry optimizer. Has anyone else ever seen this happen? Also, after the discussion of the chain method for TS location here, I am interested in adding the chain subroutine from ampac to mopac. I have the ampac source code, but have lost the documentation. Has anyone tried to do this, and have they had any success? I would compile ampac and use it, but I want to use PM3, and my compiler complains about the common blocks in ampac 2.1 and I don't relish going through and cleaning them up, if I can avoid it. Finally, are the docs for Ampac availible by ftp anywhere? I don't remember seeing them at the site at UT where ampac is ftp'able. Please let me know if you know where I can snarf them. Thank you very much. +-----------------------------------------------------------------------------+ Andy "The Fantom" Fant (fant@{chem3,umbc4,blt}.umbc.edu) Disclaimer: My postings are definitive, reality is often inaccurate "It is my supposition that the Universe is not only queerer than we imagine, It is queerer than we CAN imagine." - J.B.S. Haldane From chemistry-request@ccl.net Wed Jan 29 14:39:18 1992 Date: Wed, 29 Jan 1992 13:00:04 -0600 To: chemistry@ccl.net From: igorl@ncsa.uiuc.edu (Igor Livshits) Subject: Workshop announcement Status: R Visualization in the Chemistry Classroom ---------------------------------------- Workshop for High School Chemistry Teachers The National Center for Supercomputing Applications, with funding from the National Science Foundation (pending), is developing prototype classroom activities and materials designed to introduce high performance computing and data visualization into high school chemistry classes. The interactive activities are based on visualizations obtained by students and teachers using NCSA's Cray supercomputer connected to a local workstation through a network or a telephone connection. The general chemistry student population is our target audience. The project is being developed in conjunction with four chemistry teachers from local high schools and will expand nationwide this summer. As part of this expansion, NCSA will conduct two one-week long workshops for interested high school chemistry teachers during the summer of 1992; NSF will provide support for travel and living expenses. Successful candidates will have a chemistry background, will be committed to innovation in the classroom, and will demonstrate both interest and ability to collaborate with colleagues. Applications will be accepted through the end of April 1992. This program will result in a more intimate relationship between high school chemistry education and current trends in the professional use of available technology. NCSA will maintain contact with all participants. One of the modules nearing completion centers on the visualization of electron densities and of atomic bonding. Other modules will deal with solutions, gases, and crystallization. Interested teachers may develop additional modules. To obtain more information about this project and solicit participation in the workshops, please contact Igor Livshits at NCSA. Igor Livshits The National Center for Supercomputing Applications 152 Computer Applications Building 605 East Springfield Avenue Champaign, Illinois 61820 Telephone: (217) 244-5606 Electronic mail: igorl@ncsa.uiuc.edu From chemistry-request@ccl.net Wed Jan 29 17:20:44 1992 Date: Wed, 29 Jan 92 15:58:05 EST From: m10!frisch@uunet.UU.NET (Michael Frisch) Subject: SCF convergence To: chemistry@ccl.net Status: R Actually, we have G88 at North Dakota State University. What versions of Gaussian have the new convergence routines? Theresa Windus North Dakota State University Fargo, North Dakota 58104 NU131691@NDSUVM1 G88 has most of the basic new features for convergence, in particular the quadratic convergence SCF. However, the algorithms are considerably more robust in G90 and also have additional options for difficult cases. Incidentally, a good reference on the subject of SCF convergence is: H. B. Schlegel and J. J. McDouall, in "Computational Advances in Organic Chemistry" C. Ogretir and I. G. Csizmadia, eds., Kluwer Academic, the Netherlands, 1991, pg 167-185. This chapter discusses both theoretical issues such as SCF stability, and how to handle practical convergence problems. While there are occasional references to features unique to the Gaussian programs, the article gives a general background as well as many tactics for difficult cases which can be applied in most SCF programs. Michael Frisch Gaussian, Inc. -------