From oosten@rulhm1.LeidenUniv.nl Thu Oct 15 11:36:42 1992 Date: Thu, 15 Oct 92 10:36:42 +0100 From: oosten@rulhm1.LeidenUniv.nl (A.B. van Oosten) To: chemistry@ccl.net Subject: J-T distortion I relation to Doug Smith's posting about a possible spontaneous symmetry lowering of C2v symmetric azepine, C6H7N : There is no orbital degeneracy in C2v symmetry, so how can a C2v molecule like azepine be a J-T system ? +---------------------------------------+-----------------------------------+ | Tony van Oosten | Huygens Laboratory | | | University of Leiden | | Ab initio cluster calculations | PO Box 9504, 2300 RA Leiden, | | on Defects in Semiconductors | The Netherlands | | | Phone: (+31) 71 275908 | | | Fax: (+31) 71 275819 | +---------------------------------------+-----------------------------------+ | Network : | +---------------------------------------------------------------------------+ From e701pm@FRCCSC21.bitnet Thu Oct 15 06:48:59 1992 Date: Thu, 15 Oct 92 11:11 FWT From: "Jean-Marie Wurtz, Biostructure" To: "Ohio Supercomputer Center, bulk mailing" Subject: Neural nets and small molecules Dear Netters, I am currently looking for references using Neural Networks applied to the small molecule area. Please mail your answers at the following email adress e701pm@frccsc21.bitnet Thanks Jean-marie WURTZ From e701pm@FRCCSC21.bitnet Thu Oct 15 06:59:56 1992 Date: Thu, 15 Oct 92 11:24 FWT From: "Jean-Marie Wurtz, Biostructure" To: "Ohio Supercomputer Center, bulk mailing" Subject: Neural nets and small molecules Dear Netters, I am currently looking for references using Neural Networks applied to the small molecule area. Please mail your answers at the following email adress e701pm@frccsc21.bitnet Thanks Jean-marie WURTZ From oosten@rulhm1.LeidenUniv.nl Thu Oct 15 12:57:29 1992 Date: Thu, 15 Oct 92 11:57:29 +0100 From: oosten@rulhm1.LeidenUniv.nl (A.B. van Oosten) To: chemistry@ccl.net Subject: maple I am using maple to solve a set of 8 nonlinear equations in 8 variables. After some 140 CP seconds on an RS6000 Maple complains : " Error, (in fsolve/gensys) case not implemented, # of equations <> # of variables " Is there anything wrong in my input or is Maple unable to handle this ? Here comes the input : Sb12 := -0.021011683569 ; Sb16 := 0.113883677873 ; Sb26 := -0.119037729749 ; Sb25 := 0.135073160847 ; Sb67 := 0.005152610204 ; Sb27 := 0.001744297114 ; Sb23 := -0.018341767483 ; Sb24 := 0.034250575028 ; eq12 := Sb12 = s12 + s12*s23 + s12*s24 + s12*s25 + s16*s26 + s16*s27 ; eq26 := Sb26 = s26 + s12*s16 + s23*s27 + s24*s27 + s25*s26 + s27*s67 ; eq16 := Sb16 = s16 + s12*s26 + s12*s27 + s12*s27 + s12*s26 + s16*s67 ; eq25 := Sb25 = s25 + s12*s12 + s23*s24 + s24*s23 + s26*s26 + s27*s27 ; eq23 := Sb23 = s23 + s12*s12 + s24*s25 + s25*s24 + s26*s27 + s27*s26 ; eq24 := Sb24 = s24 + s12*s12 + s23*s25 + s25*s23 + s26*s27 + s27*s26 ; eq27 := Sb27 = s27 + s12*s16 + s23*s26 + s24*s26 + s25*s27 + s26*s67 ; eq67 := Sb67 = s67 + s16*s16 + s26*s27 + s27*s26 + s27*s26 + s26*s27 ; fsolve ({eq12,eq26,eq16,eq25,eq23,eq24,eq27,eq67}, { s12, s26, s16, s25, s23, s24, s27, s67}, { s12=0.0..0.3, s26=0.0..0.3, s16=0.0..0.3, s25=0.0..0.3, s23=0.0..0.3, s24=0.0..0.3, s27=0.0..0.3, s67=0.0..0.3}); From mail Thu Oct 15 10:19:39 1992 From: hyper!slee (Thomas Slee) Subject: Nobels and "computational chemistry" To: chemistry@ccl.net Date: Thu, 15 Oct 1992 10:17:59 -0400 Seems like some comment is called for on the list on the occasion (about time) of another theory Nobel (R. A. Marcus), and I would like to try to get some discussion started here. One thing that strikes me is that the only theory winners in the last couple of decades (Hoffmann, Fukui and now Marcus) have won for things that are fairly distinct from the mainly quantitative stuff that gets discussed here on OSC, and I wonder why this is? (Polanyi knows and does plenty of theory of course, but I am counting his stuff as chemical physics for the moment). My own theory is that the "mainstream" of computational chemistry--the development of new methods--has been hugely collaborative, and although there are clear "leaders in the field" there is no single person with whom the innovations can be identified. Progress has been essentially incremental. Nobels, of course, are awarded for discoveries, and the elucidation of fragment/frontier orbital models, and Marcus's electron transfer stuff (I assume this is what it was for; I just saw one newspaper article about it) are just that. There is another possibility, however: the awards have been for those who have really got deeply involved in the phenomonology of chemistry, and who have not been afraid to sacrifice rigour to tackle complicated systems and problems, rather than simply wait for computers to get faster. Perhaps there is a lesson that we can learn from "real chemists" here: there is certainly still an anti-theory attitude among many chemists (compare the ratio of theory to experiment in any chemistry and physics departments near you: quite a difference!), but the Nobels show that they do listen when theory really has an impact on their world. As a final comment, with Marcus being Canadian born and educated, and with the Blue Jays reaching the world series, Thurs 14 Oct. 1992 was a good day at Hypercube! Tom -- Tom Slee Hypercube, Inc., #7-419 Phillip St., Waterloo, Ont. N2L 3X2 Internet: slee@hyper.com Tel. (519) 725-4040 From urquhart@mcmail.cis.mcmaster.ca Thu Oct 15 10:28:52 1992 Date: Thu, 15 Oct 1992 14:28:52 -0400 (EDT) From: Stephen Urquhart Subject: Re: Nobels and "computational chemistry" To: Thomas Slee Hi Tom! > As a final comment, with Marcus being Canadian born and educated, and > with the Blue Jays reaching the world series, Thurs 14 Oct. 1992 was a good > day at Hypercube! > and here at Mac!! excellent game, eh? Mac sorta shut down yesterday afternoon. Stephen From jle@world.std.com Thu Oct 15 13:06:23 1992 Date: Thu, 15 Oct 1992 17:06:23 -0400 From: jle@world.std.com (Joe M Leonard) To: chemistry@ccl.net Subject: MM3 correction factors mean what? As folks know, I've been trying to understand the MM3 forcefield, and have been posting regarding this from time to time. I've got another one, which is a bit different... When looking at the output from MM3, there are miscellaneous corrections included (provided below) BOND LENGTH R(0) K(S) ENERGY E-NEG MISCEL C( 1)- C( 2) 1.5100 1.5090 4.8000 0.0003 B C( 1)- H( 3) 1.0000 1.1190 4.7400 6.5533 0.0070 CO C( 1)- H( 4) 1.0000 1.1138 4.7400 5.9126 0.0018 CO C( 1)- H( 5) 1.0000 1.1138 4.7400 5.9126 0.0018 CO C( 2)- O( 6) 1.2300 1.2080 10.1000 0.3326 C( 2)- H( 7) 1.0000 1.1180 4.3700 5.9253 B The structure is acetaldehyde. What does the code CO mean, and what are these corrections. Finally, where are they described...? Joe Leonard jle@world.std.com From hogue@mosquito.den.mmc.com Thu Oct 15 10:08:05 1992 Date: Thu, 15 Oct 92 16:08:05 MDT From: hogue@mosquito.den.mmc.com (Pat Hogue 1-2183) To: chemistry@ccl.net Subject: Solvent on plastics Greetings netters: We need suggestions on the "best" program to model the interaction of hot plymers with various solvents in order to predict dissolution or reaction. For example hot polypropylene with cyclohexane-will it dissolve? Thanks to all who respond. Pat Hogue From U6124673@ucsvc.ucs.unimelb.edu.au Tue Oct 16 20:42:06 1992 Date: 16 Oct 1992 10:42:06 +1000 From: U6124673@ucsvc.ucs.unimelb.edu.au Subject: Re: Nobel Prizes To: CHEMISTRY@ccl.net Dear CompList readers, I have just read Tom Slee's interesting comment about theory Nobel prizes in chemistry - it does seem amazing that only three recent recipients were theoreticians. Perhaps this reflects the current trends in computational chemistry, ie., tackling problems with the aim of obtaining results as good as experiment, rather than using computational methods to gain an understanding of a problem which can be applied by many other chemists. An example is Marcus' theory for describing electron-transfer rates by mainly thermodynamic considerations. The result is a non-numerical framework which has been applied by numerous other workers to explain results or make predictions. Perhaps there is still a place for simplistic quantum-chemical calculations which give insight into a problem rather than the exact result? Of course there are many situations in which this is not appropriate, but does anyone have comments on this philosophy? Greg Scholes School of Chemistry The University of Melbourne AUSTRALIA