From pensak@wizards.dupont.com Thu Aug 20 08:00:39 1992 Date: Thu, 20 Aug 92 07:56:55 EDT From: pensak@wizards.dupont.com (Dave Pensak) To: chemistry@ccl.net Subject: Polymer modelling Status: R Can anyone point me towards code which, given the coordinates of a monomer and the desired helicity parameters, will generate the appropriate dihedral angles and the ultimate coordinates ? Thanks David A. Pensak E. I. DuPont de Nemours, Inc. Wilmington, Del From ernst@bkfug.kfunigraz.ac.at Thu Aug 20 10:28:13 1992 Date: Thu, 20 Aug 1992 16:28:11 -0100 From: ernst@bkfug.kfunigraz.ac.at To: chemistry@ccl.net Subject: motecc Status: R I would like to obtain ANY information about the program package MOTECC, Modern Techniques in Computational Chemistry. What is in the package, where and how can a licence be obtained, etc... Please reply directly to ernst@bkfug.kfunigraz.ac.at I will summarize if the response is sufficient. Thanks, Margot From jie@qtp.ufl.edu Thu Aug 20 12:48:37 1992 Date: Thu, 20 Aug 92 12:54:16 EDT From: jie@qtp.ufl.edu To: CHEMISTRY@ccl.net Subject: Re: semi-empirical methodology Status: R The followings are our points about semi-empirical methodology disscused in this net recently by some people: 1) It is always the attitude of Prof. Dewar: 'to keep the semiempirical approach under control'. If someone wanted to reparamize some or all parameters in AM1, for example, he/she may give hundreds set of new parameters. There would be no any SE if such hundred set of parameters of same method would have been used. From MINDO to MNDO to AM1 was not simply by reparameterizing the parameters, they are different procedures or having important modification to the preceding one. The only exception was PM3 which was created by reparamizing the existed method of MNDO (or AM1?) by Dr. Stewart. 2) Why using Gaussians in AM1? The reason is quite simple, as I understand. Because MNDO-type method uses non-splitted wavefunction and the function used there, exp(-aR), as core-core repulsion function was too simple to mimic the behavior of such interaction covering a large area from bonding distance to van der Waals, some kind of modification to this form of function was necessary. By linear combination of a few (ussuly 2 to 3) very small(compare to exp(-aR)) Gaussians to add to the function is not an uncepted way. Actually, from the point of view of mathmatics it is similar to the way of linear combination of GTOs to get STO which used widely in ab intio method. The correction to core-core repulsion might be, of course, too brute if these gaussians are too big or too sharp or located on inadequte positions. The way of using Gaussians has some advantiges, however. It can effectively correct some tendentious errors in energy calculations and the parameters in those Gaussions can be easily optimized. If those gaussions were selected properly they would not cause trouble. We can think this as a special kind of pseudo-potential people employed in other places. I can not see any reason why those gaussions will cause the wavefunction less ab initio-like. 3)Since AM1-type SE methods are so simple to use, people are using them everywhere for very wide chemical problems. Some incorrect results have been discovered from time to time. It is unfair to kill or accuse them only by such isolated and specific errors. NO method would do every thing correct today even some very high level ab initio methods which people could not perform easily. Parameterization for AM1-type method was often not a happy work. Millions calculations might be performed and as many experiment data as we could find may need to be checked before the parameters be published. *** The following paragraph was written by Prof. Dewar: *** 4) SAM1 is a new theoretical treatment which we have developed and which will be published in the near future. Preliminary results for CHON were announced at the Sanibel Symposium in March and later results were included in a progress report to AFOSR and have been circulated privately to various colleagues, not icluding Dr. Stewart. Final results for halogen-containing molecules have not been released. Dr. Stewart's statements concerning SAM1 should therefore be disregarded. Caoxian Jie Jianguo Yu Chemistry Department University of Florida From AHOLDER@VAX1.UMKC.EDU Thu Aug 20 14:30:27 1992 Date: 20 Aug 1992 13:29:59 -0500 (CDT) From: AHOLDER@VAX1.UMKC.EDU Subject: More semiempirical stuff.... To: CHEMISTRY@ccl.net Status: R Fellow netters, More semiempirical comments. Joe Leonard at Wavefunction asked about the "consensus" on what to do with identified problems in semiempirical parameter sets. There is no really good answer to this dilema. One of my colleagues pointed out to me that all parameterization schemes are compromises, and some things are treated well at the expense of others. That's why PEOPLE must be involved in the process. People can make informed decisions as to the proper course to follow to get the most chemistry. Computer algorithms are generally not that good. And yes, Joe, the parameterizers had better get it right the first time or the world will suffer. On the subject of Gaussians, some clarification is probably needed. When quantum jocks hear "Gaussian" they translate "orbitals". In AM1/PM3 protocol, this is not correct. The Gaussian terms used in these models are adjustments to the core/core repulsion function that describes the core electrons and nucleus. They don't "alter the wavefunction", but change the interaction energy between nuclei. This can have the INDIRECT effect of altering the wavefunction by driving the geometry optimization to a different point on the surface. All of this Gaussian talk brings up a good example of compromise. When the AM1 parameters for phosphorous were developed (Dewar, M. J. S.; Jie, C. J. Mol. Struct. (THEOCHEM) 1989, 187, 1) a repulsive Gaussian was placed at around three angstroms to handle some specific prob- lems that Prof. Jie encountered. This results in a "lump" in the energy potential when two phosphorous atoms are at a range of 3 angstroms. (Rather than being a regular hexagon, the P congener of benzene is lozenge shaped.) This Gaussian still remains because it solved other problems that were considered more important. The behaivor of these parameters as well as all of the others is quite complex and almost inextricably intertwined. I feel that Gaussian functions should be left out of the parameterization scheme until the QM has had its chance to do its best. They can then be added on a case by case basis (each for a reason) as needed. Others have a differing viewpoint on this. As to the parameterization strategies of MM and SE, I think that they are quite different. In MM, for instance, if a description of amides is not correct, you can examine the specific parameters for the bond and other potentials in the molecule and adjust them independently. If one is dealing with a problem in an SE model, you are dealing with sets of parameters that describe the WHOLE RANGE of nitrogen, hydrogen, carbon and oxygen chemistry. A horse of a different color.... My understanding on the implementation of MNDO/D is that Walter has ex- tended the multipole expansion to d-orbitals and this is the basis of his treatment rather than explicit d gaussians. If you'd like to actually read about it, he just published (Thiel, W.; Voityuk, A. A. Theor. Chim. Acta 1992, 81, 391.) See y'all at ACS! Andy =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= DR. ANDREW HOLDER Assistant Professor of Computational/Organic Chemistry Department of Chemistry || BITNET Addr: AHOLDER@UMKCVAX1 University of Missouri - Kansas City || Internet Addr: aholder@vax1.umkc.edu Spencer Chemistry, Room 315 || Phone Number: (816) 235-2293 Kansas City, Missouri 64110 || FAX Number: (816) 235-1717 =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= From jle@world.std.com Thu Aug 20 15:42:47 1992 Date: Thu, 20 Aug 92 15:42:11 -0400 From: jle@world.std.com (Joe M Leonard) To: chemistry@ccl.net Subject: MM3 atom type for Sulfur... Status: RO This questions applies for MM2 as well - what is the corect atom type for Sulfur in H2C=S? I recall using 2-15 as the C=S bond with MM2 in the good old days, but is this correct? Looking at the list of types, I don't see =S as an option... Joe Leonard jle@world.std.com From J_BROWN@uvmvax.uvm.edu Thu Aug 20 16:10:59 1992 Date: Thu, 20 Aug 1992 16:11 EST From: J_BROWN@uvmvax.uvm.edu Subject: REQUESTING HELP WITH ZR METAL MODELING PARAMETERS To: CHEMISTRY@ccl.net Status: RO Does any one know where I can get information on modeling parameters (force constants etc.) for octahedral Zr phosphates. Our software only has tetrahedral parameters. We are using Insight II on SGI. please send any information to J_BROWN@UVMVAX.BITNET THANK YOU From dave@carbon.chem.csiro.au Thu Aug 20 19:56:58 1992 Date: Fri, 21 Aug 92 09:55:50 EST From: (Dr.) Dave Winkler Subject: Re: Organic chemistry mailing list To: TIMMERMA@rulgl.LeidenUniv.nl, chminf-l@iubvm.ucs.indiana.edu, Status: RO Dear Johan and netters, > Via the quantum chemistry E-mail list we picked up your request about an > organic chemistry mailing list. From then on, we lost track, but are very > interested in taking part in one. Do you know more about how to get on to > > > such a list? > Thanks in advance, > yours sincerely, > Johan Timmermans > Dept. of Chemistry > Leiden University > The Netherlands After posting the message about the ORGCHE-L mailing list, I received quite a few messages back from people who had tried to subscribe. Like me they got nowhere. I think it is safe to assume that the ORGCHE-L list is not functioning. Everyone felt that such a list would be very useful. What we need is for someone to set one up! I cannot do it as I'm not a synthetic chemist and our machine is not able to function as a list server. Do you have any ideas how we can persuade someone to set one up? A bit of medicinal organic chemistry appears on the ANCHODD mailing list but I think there would be a tremendous response to a mailing list dedicated to organic (synthetic) chemistry. Is there anyone out there in a strong organic chemistry school who can help everyone out? Dave __________________________________________________________________________ Dr. David A. Winkler Voice: 61-3-542-2244 Principal Research Scientist Fax: 61-3-543-8160 CSIRO Division of Chemicals and Polymers Private Bag 10 Clayton, Australia.