From wolf@tut.ac.jp Tue Feb 2 00:13:14 1993 Date: Mon, 1 Feb 93 15:13:14 +0900 From: wolf@tut.ac.jp (Ihlenfeldt Wolf-Dietrich) Message-Id: <9302010613.AA15676@tutuser.tut.ac.jp> To: CHEMISTRY@ccl.net Subject: Ghostview Ok, Ross Mair (rwm@stan.xx.swin.oz.au) writes: > > >Ghostview-1.3 is available via anonymous ftp from: > > .... > > Ghostview-1.3 is outdated. Version 1.4 seems to be the current one. So - is Vers 1.4 available by anon ftp from anywhere?? Ross Yes, it is. Actually, version 1.4.1 is already out. A quick check (with xarchie1.3, I can't understand how I could live without it) reveals: aenas.mit.edu /pub/gnu aix370.rrz.uni-koeln.de /.disk1/gnu ascwide.ascii.co.jp pub/GNU ashley.cs.widener.edu /pub/src/gnu askhp.ask.uni-karlsruhe.de /pub/hpux/Gnu clover.csv.warwick.ac.uk /pub/gnu colonsea.dcs.ed.ac.uk /export/postscript and /pub/postscript cs.ubc.ca /mirror2/gnu and so on ... (136 sites in total). In Australia, try for example plaza.aarnet.edu.au /gnu My advice (I know, nobody asked for it :-)): Get archie or xarchie and dive into the wonderful universe of the internet. If only chemistry packages would be as easily available as GNU software - the distribution methods of QCPE are obviously no longer state of the art. WDI From radomski@mond1.ccrc.uga.edu Sun Jan 31 22:26:34 1993 Message-Id: <9302010826.AA10861@mond1.ccrc.uga.edu> To: chemistry@ccl.net Subject: Re: Postscript file viewer Date: Mon, 01 Feb 93 03:26:34 -0500 From: radomski@mond1.ccrc.uga.edu >> Ghostview-1.3 is outdated. Version 1.4 seems to be the current one. >So - is Vers 1.4 available by anon ftp from anywhere?? >Ross Ver. 1.4.1 is available from the same site as v. 1.3... i.e. prep.ai.mit.edu:/pub/gnu/ghostview-1.4.1.tar.Z; you'll need to get ghostscript-2.5.2.tar.Z as well, which is also updated version. And it is really worth all your effort to make it running; nifty, quick, well designed layout - and a lot more. Quite a [nice] surprise, as the time for public domain programs of such quality is apparently moving fast into our past, commercialization rear- ing its ugly... Kudos for the AUTHOR[s]. Jan ------------ Jan P. Radomski Complex Carbohydr Research Center, Univ. of Georgia, Athens, GA 30602. Voice (706) 542-4452; Fax (706) 542-4412 From cfat1019@servus.rus.uni-stuttgart.de Mon Feb 1 13:04:28 1993 Date: Mon, 1 Feb 1993 12:04:28 +0100 From: cfat1019@servus.rus.uni-stuttgart.de Message-Id: <9302011104.AA30645@servus08.rus.uni-stuttgart.de> To: chemistry@ccl.net Subject: Conference Announcement Subject: Conference on "Relativistic Effects in Heavy-Element Chemistry and Physics" The following is a Tex document of the first announcement: \documentstyle[din_a4]{article} \input{prepictex} \input{pictex} \input{postpictex} \pagestyle{empty} \parindent0cm \addtolength{\leftmargin}{2mm} \addtolength{\topmargin}{20mm} \begin{document} \newfont{\smpt}{cmr5 scaled 500} \begin{center} \large RESEARCH CONFERENCE ON \\[25mm] {\beginpicture \setshadesymbol ( {\smpt .} ) \setshadegrid span <1.2pt> \shaderectangleson \frame <4mm> { \lines{ \Large\bf RELATIVISTIC EFFECTS IN \rule[-2mm]{0mm}{2mm}\cr \Large\bf HEAVY-ELEMENT CHEMISTRY AND PHYSICS: \rule[-4mm]{0mm}{2mm}\cr \large\bf Approximate Relativistic Methods in Electronic Structure Calculations } } \endpicture} \\ \vspace{15mm} \xipt\em De Haan, near Oostende, Belgium, 10-14 June 1993 \rm \\[5mm] \parbox{97mm}{\makebox[4cm]{\bf Chairmen:} E.J. Baerends (Amsterdam) \\ \makebox[4cm]{} H. Stoll (Stuttgart)} \\[10mm] \large\bf \underline{Preliminary Programme} \\[4mm] \end{center} \xipt \rm The purpose of this first European Research Conference on Relativistic Effects in Heavy-Element Chemistry and Physics is to describe, compare and evaluate approximate methods for determining relativistic contributions to the properties of atoms, molecules and solids. Such methods include relativistic semi-empirical approaches, relativistic pseudopotentials to be used in formally non-relativistic valence-only calculations, relativistic density-functional schemes as well as one- and two-component all-electron ab-initio techniques. A mutual checking and validation of results obtained by the various methods should lead to better understanding of their range of applicability. A critical comparison against more accurate (four-component) relativistic treatments should lead to a better understanding of the fundamentals on which these methods are based. \\[5mm] The following scientists have been invited to give presentations: \begin{tabbing} \makebox[75mm]{} \= \kill G.B. Bachelet (Roma) \> W. Kutzelnigg (Bochum) \\ E.J. Baerends (Amsterdam) \> C.M. Marian (Bonn) \\ N.E. Christensen (Aarhus) \> P. Pyykk\"{o} (Helsinki) \\ J.P. Daudey (Toulouse) \> N. R\"{o}sch (M\"{u}nchen) \\ M. Dolg (Stuttgart) \> A.J. Sadlej (Lund) \\ W.C. Ermler (Hoboken, NJ) \> K. Schwarz (Wien) \\ B. Fricke (Kassel) \> W.H.E. Schwarz (Siegen) \\ O. Gropen (Troms\o) \> P. Schwerdtfeger (Auckland) \\ B.A. Hess (Bonn) \> L. Seijo (Madrid) \\ J. Karwowski (Toru\a'{n}) \> U. Wahlgren (Stockholm) \end{tabbing} A discussion session (I.P. Grant, Oxford) as well as poster sessions will be organized. \end{document} From marc@david.saclay.cea.fr Mon Feb 1 14:26:49 1993 Date: 01 Feb 93 19:26:49-0500 From: SBPM Marc GINGOLD Message-Id: <9302011826.AA05181(a)david.saclay.cea.fr> To: chemistry@ccl.net Subject: GAUSSIAN 90 doc How can I see the Gaussian 90 documentation. Is it on line somewhere? Can I get a copy (email or paper)? Thanks in advance Marc Gingold ================================================ COMMISSARIAT A l'ENERGIE Atomique Centre d'Etudes de Saclay (CEA CE Saclay) DSV/DBCM/SBPM F-91191 Gif-sur-Yvette Cedex FRANCE ------------------------------------------------ Phone:(33) 1 6908 2965 marc@david.saclay.cea.fr ================================================ From WXIIBR@WEIZMANN.WEIZMANN.AC.IL Mon Feb 1 16:27:10 1993 Message-Id: <199302011229.AA13096@oscsunb.ccl.net> Date: Mon, 01 Feb 93 14:27:10 +0200 From: Amram Golombek Subject: PICT from SG to MAC To: chemistry@ccl.net Dear Netters, how do we most efficiently migrate .PICT files fron Silicon Graphics environment to the Mac, thanks, Amram. From zottola@chemvax.chem.duke.edu Mon Feb 1 03:38:37 1993 Date: Mon, 01 Feb 1993 08:38:37 EST From: zottola@chemvax.chem.duke.edu To: CHEMISTRY@ccl.net Message-Id: <00967794.BAC90600.105@chemvax.chem.duke.edu> Subject: LST in Gaussian92 I have run into a problem with using the LST option in gaussian92. I have followed the example in the gaussian manual, specifying both a reactant and a product geometry. However, instead of using a Z-matrix, I defined both geometries in cartesian coordinates. Does the LST option require a z-matrix? Secondly, unlike the example in the manual, the system I am studying is under- going more than an isomerization, it is actually losing a ligand. Will this major structural transformation cause LST to bomb? Thanks for any and all help! -mark zottola From gene@eastrg2.cray.com Mon Feb 1 06:13:16 1993 Date: Mon, 1 Feb 93 11:13:16 EST From: gene@eastrg2.cray.com (Eugene Fleischmann) Message-Id: <9302011613.AA13607@sodium.cray.com> To: chemistry@ccl.net Subject: Re: Postscript file viewer > From chemistry-request@ccl.net Mon Feb 1 10:52:26 1993 > To: chemistry@ccl.net > Cc: rwm@stan.xx.swin.oz.au (Ross Mair) > Cc: radomski@mond1.ccrc.uga.edu > Subject: Re: Postscript file viewer > X-Mts: smtp > Sender: chemistry-request@ccl.net > > >> Ghostview-1.3 is outdated. Version 1.4 seems to be the current one. > > >So - is Vers 1.4 available by anon ftp from anywhere?? > > >Ross > > Ver. 1.4.1 is available from the same site as v. 1.3... i.e. > prep.ai.mit.edu:/pub/gnu/ghostview-1.4.1.tar.Z; > you'll need to get ghostscript-2.5.2.tar.Z as well, which > is also updated version. > > And it is really worth all your effort to make it running; nifty, > quick, well designed layout - and a lot more. Quite a [nice] > surprise, as the time for public domain programs of such quality > is apparently moving fast into our past, commercialization rear- > ing its ugly... > Kudos for the AUTHOR[s]. > > Jan > ------------ > Jan P. Radomski > Complex Carbohydr Research Center, > Univ. of Georgia, > Athens, GA 30602. > Voice (706) 542-4452; Fax (706) 542-4412 > > > > --- > Administrivia: This message is automatically appended by the mail exploder: > CHEMISTRY@ccl.net --- everyone CHEMISTRY-REQUEST@ccl.net --- coordinator > OSCPOST@ccl.net send help from chemistry Anon. ftp www.ccl.net > CHEMISTRY-SEARCH@ccl.net --- search the archives, read help.search file first > --- > > I am confused. FTP session to prep.ai.mit.edu follows: Connected to prep.ai.mit.edu. 220 aeneas FTP server (Version 6.2 Mon Oct 19 22:51:51 EDT 1992) ready. Name (prep.ai.mit.edu:gene): anonymous 331 Guest login ok, send e-mail address as password. Password: 230-Welcome, archive user! This is an experimental FTP server. If have any 230-unusual problems, please report them via e-mail to . 230-If you do have problems, please try using a dash (-) as the first character 230-of your password -- this will turn off the continuation messages that may 230-be confusing your ftp client. 230- 230-Please read the file README.trademark 230- it was last modified on Fri Apr 3 18:05:15 1992 - 303 days ago 230 Guest login ok, access restrictions apply. ftp> ls -l 200 PORT command successful. 150 Opening ASCII mode data connection for /bin/ls. total 12 -rw-rw-r-- 1 3009 1898 Dec 4 1991 Index -rw-rw-r-- 48 15806 263 Apr 3 1992 README.trademark drwxrwx--- 5 daemon 512 Dec 4 1991 archive drwxrwx--x 2 14640 512 Dec 4 1991 bin -rwxr-xr-x 1 root 0 Nov 3 1991 copyright drwxrwx--x 3 14640 512 Nov 1 15:08 etc drwxrwxr-x 2 root 512 Dec 4 1991 interop88 drwxrwx--x 7 root 1024 May 19 1992 private drwxrwxr-x 17 15806 1024 Nov 3 10:35 pub lrwxrwxrwx 1 root 2 Apr 2 1989 u -> u1 drwxrwxr-x 2 root 512 Dec 4 1991 u1 lrwxrwxrwx 1 root 2 Apr 2 1989 u2 -> u1 226 Transfer complete. remote: -l 653 bytes received in 0.38 seconds (1.7 Kbytes/s) ftp> get /pub/gnu/ghostview-1.4.1.tar.Z /pub/gnu/ghostview-1.4.1.tar.Z: No such file or directory ftp> get /pub/gnu/ghostscript-2.5.2.tar.Z /pub/gnu/ghostscript-2.5.2.tar.Z: No such file or directory What am I doing wrong? Gene Fleischmann From WILLSD@CONRAD.APPSTATE.EDU Mon Feb 1 06:25:52 1993 Date: 01 Feb 1993 11:25:52 -0500 (EST) From: WILLSD@CONRAD.APPSTATE.EDU Subject: anion structure? To: CHEMISTRY@ccl.net Message-Id: <01GU7EZDNZ828ZEJEK@conrad.appstate.edu> Dear Netters I have a fundamental question regarding calculations of structures. It applies specificially to the use of MO methods (esp GAUSSIAN), but applies in principle to any method that can give structures. How is the structure calculated for an ion to be interpreted? For example, suppose that a minimum energy tetrahedral structure for tetrafluoroborate (BF4-) gives BF distances that are say, 0.05 A longer than the distances known from Xray crystallography. Is it possible to determine if this discrepancy is due to some type of defect in the method used to find the structure, or if it is due to comparing a gas phase calculation to a crystal structure (ie neglect of clearly strong interactions in the solid between anions and cations, versus defects in the basis set, etc.) Another way of asking this question is: Does anyone out there know of ANY experimental measurement of the structure of an ion (esp. an anion) in the gas phase? Clearly, there are severe experimental problems with such measurements, but they may have been done perhaps flame emission spectroscopy, or some sort of elctron capture / electron diffraction method. I would appreciate hearing of any insight you might have on this, and I will summarize the responses to the list. Thanks, Steve Williams Chemistry ASU, Boone NC 28608 willsd@stat.appstate.edu From FOX@CMCHEM.CHEM.CMU.EDU Mon Feb 1 09:17:31 1993 Date: Mon, 1 Feb 1993 14:17:31 EST From: FOX@CMCHEM.CHEM.CMU.EDU Message-Id: <930201141731.212024c3@CMCHEM.CHEM.CMU.EDU> Subject: Re: LST in Gaussian92 To: chemistry@ccl.net You need two Z-matrices to run the LST code in Gaussian 92. Use the NewZmat utility to produce these since you have the product and reactant geometries in cartesians, newzmat -ixyz -ozmat prod.xyz prod.com The LST does best when the reactant and product geometries are "close" and the concept of a linear path is likely to lie close to the transition state. You simply need strucutures on either side of the TS so even with a dissociation reaction you can specify a structure where the departing ligand is not separated to improve the performance of the LST method. Doug Fox help@gaussian.com From vazquez@iqm.unicamp.br Mon Feb 1 14:18:30 1993 Date: Mon, 1 Feb 1993 17:18:30 -0300 From: vazquez@iqm.unicamp.br Message-Id: <9302012018.AA16239@karibdys.iqm.unicamp.br> To: CHEMISTRY@ccl.net Subject: Re: Postscript file viewerIn-Reply-To: <9302011613.AA13607@sodium.cray.com>; from "Eugene Fleischmann" at Feb 1, 93 4:52 pm Stuff deleted > > I am confused. FTP session to prep.ai.mit.edu follows: > > Connected to prep.ai.mit.edu. > 220 aeneas FTP server (Version 6.2 Mon Oct 19 22:51:51 EDT 1992) ready. > Name (prep.ai.mit.edu:gene): anonymous > 331 Guest login ok, send e-mail address as password. > Password: stuff deleted > ftp> get /pub/gnu/ghostview-1.4.1.tar.Z > /pub/gnu/ghostview-1.4.1.tar.Z: No such file or directory > ftp> get /pub/gnu/ghostscript-2.5.2.tar.Z > /pub/gnu/ghostscript-2.5.2.tar.Z: No such file or directory > > What am I doing wrong? You omited the destination file name. If your disk does not have a /pub/gnu dir you must specify the destination file: ftp> get /pub/gnu/ghostscript-2.5.2.tar.Z ghostscript-2.5.2.tar.Z ^^^^^^^^^^^^^^^^^^^^^^^ or ftp> cd pub/gnu From youkha@iris75.biosym.com Mon Feb 1 06:47:41 1993 Date: Mon, 1 Feb 93 14:47:41 -0800 From: youkha@biosym.com Message-Id: <9302012247.AA20137@iris75.biosym.com> To: chemistry@ccl.net Subject: Summary: Internals & Non-Bonds SUMMARY OF DISCUSSION ON NOMENCLATURE FOR INTERNALS & NON-BONDS: Ten days ago, I posted a question on the net (see original posting later): I asked for a word that would describe both INTERNALS AND NON-BONDS. I received mail from Peter Shenkin, Jack Smith, Wolfgang Sauer, Jeff Nauss, and a few scientists at Biosym. It follows, together with my answer to Peter on "coordinate systems". In fact, I would have expected more mail on that topic of "redundant coordinates", since it is a difficult and controversial problem. PROPOSED TERMS: "force-field variables" (Peter Shenkin) 'energy term' (Kottalam) 'energy components' (Marv Waldman) 'diagonals' ( Steve Koerber) "force field" or "interaction" coordinates (Jack Smith) "interaction supercoordinates" (Jack Smith) "Proximal" or "distal" coordinates (Jack Smith) "bonded and non-bonded interactions" (Wolfgang Sauer) "topological" coordinates "atom-atom interactions" (or substitute "interaction" with "distance")(W. Sauer) "distance matrix" (Wolfgang Sauer) specifically for non-bonds: "extra" or "external" coordinates (Jack Smith) ORIGINAL POSTING: ******************************************************************************* I am looking for a good word that describes BOTH : - Internal Coordinates or simply "INTERNALS": (bond, valence angle, out-of-plane angle, dihedral displacements) representing 1-2, 1-3, 1-4 interactions. - Non-Bond Distances or simply "NON-BONDS" representing 1-4...n interactions. Since we use them both in the analytical form of the Potental Energy, I would like to use one word only to refer to that coordinate system as a whole. One solution is to use simply the word Internals, in adopting a wider meaning, since after non-bond interactions are both intra-molecular and inter-molecular. The problem is that there is no convention here and confusion may arise. There is not even a convention about internal coordinates themselves ! Any thoughts ? ****************************************************************************** DETAILED ANSWERS: *************************************************************************** Peter S. Shenkin, shenkin@still3.chem.columbia.edu; (212) 854-5143 *************************************************************************** picked on the following sentence in my original posting: I would like to use one word only to refer to that coordinate system ^^^^^^^^^^^^^^^^^ as a whole. ^^^^^^^^^^ >Philippe, I think you want to use a term like "force-field variables" >to refer to this set of variables. What defines them is not that they >correspond to a set of coordinates -- they do not -- but rather that >they are the variables in terms of which the energy and forces are most >simply expressed; computationally, they constitute the input to the >function that calculates the energy and forces. > >For example, whether you choose to represent the molecule in Cartesian >or in internal coordinates, the distances between all pairs of atoms are >well defined once a complete set of coordinates has been assigned to the >molecule. However, none of the non-bonded distances is equal to the value >of any coordinate in either system. The non-bonded distances are redundant >geometric variables which can be calculated from the coordinates; they >are not coordinates themselves, unless you are working in distance space. >Therefore I'd use a term like "force-field variables". (Note the >hyphen. :-) ) ************************************************************************** JACK A. SMITH jas@medinah.atc.ucarb.com (304) 747-5797 ************************************************************************** Philippe: >I can see what you're after, but I think you need to be careful, especially >with the word "coordinates" - to lump the non-bonded distances in the with >"internal" coordinates is going beyond an already complete coordinate >system. Also, you're speaking strictly from a "molecular mechanics" >viewpoint and this can just lead to more confusion in the use of internal >coordinates for quantum mechanical computations. But even with this >caveat... > >Perhaps you could refer to the non-bonded distances as "external" >coordinates - implying that they are "outside" the normal(?) coordinate >system, or even "extra-coordinates". The combined set could then be called >the "topological" coordinates since it implies connectivity. "Proximal" or >"distal" coordinates might imply the distance-orientation of both bonded >and non-bonded interactions. > >How about "force field" or "interaction" coordinates? - I think most could >identify with that. But then how do you deal with variations like those >sometimes used in treating organometallic complexes, where angles and >dihedrals involving the central metal atom are dropped in favor of 1-3 and >1-4 interactions, or where pi bonds are complexed through bond or ring >centroids? And what about H-bonds, ionic bonds, dative bonds, etc. -- >perhaps there's too many ways to define a force field to have a single name >for its inherent (overcomplete) "coordinate" system. How would apply this >terminology to ionic crystals or metals? Should you distinguish between >intermolecular and intramolecular terms? > >I would vote for "interaction supercoordinates" (although I'll probably >change my mind after I send this!) *************************************************************************** Jeffrey L. Nauss nauss@wrair-emh1.army.mil 202-576-3485 *************************************************************************** >Your idea of referring to both internal coordinates and non-bonded interactions >with the same term may be confusing. I think of the internal coordinates as >simple geometric terms. On the other hand, I think of the non-bonded terms as >involving "knowledge" of the three-dimensional structure of the molecule in >question, i.e. the non-bonded terms are a higher order of complexity. > >So, I think it might be confusing to refer to both as "Internals." > >Now to completely waffle on my answer, whatever you decide to call your set of >terms, please, be sure and clearly and repeatedly emphasis what you mean. Be >very clear in your definition. > >Those are my thoughts but on only one cup of coffee :-). *************************************************************************** Wolfgang Sauer sauer@organik.uni-erlangen.de 49/0 - 9131 - 85 - 2952 *************************************************************************** >to me - by default - the term "internal coord." is so tightly bound to >Z-matrices, which are only the minimal sets of your first definition, that >yes, you would confuse me, if you were using it in the wider meaning. > >The same, but only to a lesser degree with Peter's suggestion "force-field >variables". Yes, I know the difference between variables and parameters, but if >you are doing parameterisations most of your time, the parameters are far more >variable than the coordinates; esp. if you consider that the coordinates for any >given conformation will be fixed (while calculating the energy). > >I guess terms like "bonded and non-bonded interactions" or "atom-atom >interactions" ( or substitute "interaction" with "distance") would be quite >accurate, but also veeery boring to read several times in one paragraph. If you >are referring mainly to the distances, why not call it the "distance matrix" of >that molecule ? Confused caused should be minimal, since DM methods do employ >exactly that: lists of distances between any pair of atoms, whether they are >bound or not ... > >... just my thoughts ... **************************************************************************** Marvin Waldman marvin@biosym.com **************************************************************************** >How about energy components? **************************************************************************** Steven Koerber steven@biosym.com **************************************************************************** >Just on first view, because both non-bonds and internals could be considered >as diagonal (as opposed to cross) energy terms, how about DIAGONALS? **************************************************************************** Kottalam at Biosym jk@biosym.com (619) 546 5366 **************************************************************************** >They are really two different things. Internals are coordinates and >non-bonds are 'interactions'. To us they appear together because of >their contribution to the MM energy. So, in your context can you call >them 'energy term'? Later he added to some of my remarks on the Urey-Bradley Force Field for the treatment of 1-3 interactions using non-bond distances: >Yes, I see that they are closely related. >Internal coordinates form a complete set of coordinate system >to describe the structure of the molecule, i.e., except position >and orientation. That is why they are called 'coordinates'. >U-B 'coordinate' comes into being only due to the energy term >it contributes. So do the 'nonbond' coordinates; although, >nonbond distances can be viewed as a subset of the distance matrix. >I am trying to explain why there is not a common name in the >literature. So whatever we invent at Biosym will need to be >assimilated by the community in the future. **************************************************************************** Philippe Youkharibache youkha@biosym.com **************************************************************************** Peter ... I like "Force Field variables", but does it mean you relate to a parametric space ? What I want is to collect enough independent suggestions to understand what do people understand/mean by internals and non-bonds, if they do relate at all to the configuration space ? the distance space ? the real space ? a parametric Energy space ? what is the space(s) that we are talking about ? what is its mathematical structure ? In fact most of previous suggestions at Biosym related to Energy "Energy Term", "Energy components" etc... It is becoming increasingly clear that "Energy" or "Force-Field" is what comes to mind with internals and non-bonds, but the Kinetic Energy is expressed in terms of the cartesian coords (an verify 6 conditions of invariance by translation and rotation). So the Kinetic Energy is a quadratic form of cartesian velocities and is clearly the metric tensor of the configuration space. The parametric Energy space has to relate mathematically to the configuration space or does it not ? What about if I just want structural information expressed in terms of proximity (through space) and topological (bonded) / topographical (values). Some people think of non-bonds as "interactions", like an afterthought an attribute of a configuration. Well when part of the Potential Energy Function, these are causes. I wrote a paper about 10 years ago on the "Internal Space of Molecules" (Chem Phys Lett,87, p 165 (1982)) because I relate to the configuration space, but that was before I even relate to "consistent" force fields and the "non-bonds" in the description of the Potential Energy. Now really I wonder what are the mathematical "spaces" we are talking about. Even if i might be right the words are anchored in people's mind what I am trying to do is to relate to people's meaning. There is a very difficult problem ... that you raised. The term coordinate system does not mean necessarily that these are independent and form a basis of the configuration space. You mentionned Distance space, well if all distances are coordinates then they certainly do not form a basis set Neither do so called internal coordinates when refering to the configuration space, but people use the terms "internal coordinate space" In fact when people refer to internal coords they mean valence coordinates usually, but it is not neceessarily the case. Urey Bradley Force Field uses Non bond distances for 1-3 interactions. In some models you can scale down 1-4 non bond interactions because they are correlated with torsions (another representation of 1-4 interactions). In simulated annealing (a la XPlor) you remove completely torsional terms and let non-bond (different functional form) play the role. It is very clear that our descriptions are overdescriptive. For quantum chemists what they mean by internal coords is a set of 3N-6 of them (a basis set) but the problem with that is that you loose the inherent symmetry of a molecule, if you want to re-establish it and still use "internal coords" you have to introduce virtual atoms, and therefore virtual degrees of freedom ... nothing really different from redundant internal coordinates There has been an incredible amount of litterature on redundancies in the past. Now with consistent force fields, the problem is even wider, since there are explicit non-bond terms in the empirical form of the energy, in addition to an already redundant set of internal energy terms. It might be that all these fields and their understanding of "Energy" and "coordinates" are different, but I see so many messages on this net where people are confused (specially newcomers) by Force Fields, internal coordinates (a la Wilson), Hessians computed ab-initio (or should I say (ab-initio Hessian) expressed in cartesian coords, or in "internal" coordinates, and that want to take these results and inject them in some way to run molecular dynamics simulations. Well confusion is largely already here. I am just trying to have some convention. ************************************************************************** Philippe Youkharibache youkha@biosym.com From baell@syd.dah.csiro.au Tue Feb 2 05:12:51 1993 Date: Tue, 2 Feb 93 11:12:51 CST From: (baell) Message-Id: <40383.baell@syd.dah.csiro.au> To: chemistry@ccl.net Subject: DMF dielectric Does anyone know (?reference) the dielectric constant for N,N- dimethylformamide (DMF)? It is mysteriously missing from all tables of collated values that I've looked at. Thanks in advance Jonathan Baell Dr. Jonathan BAELL CSIRO Div. of Animal Health Pvt. Bag No 1, P O Glebe 2037 Phone : 61 2 5667943 Fax : 61 2 6928561 Email : baell@syd.dah.csiro.au From rose@duck.agouron.com Mon Feb 1 09:57:38 1993 Date: Mon, 1 Feb 93 17:57:38 -0800 From: rose@duck.agouron.com (Peter Rose) Message-Id: <9302020157.AA26946@duck> To: chemistry@ccl.net Subject: Transition Structure Search, LST Mark, I developed a LST routine for the SPARTAN program that allows you to specify your reactant and product in cartesian coordinates. You simply draw the reactant and product using the BUILDER in SPARTAN. SPARTAN is available from Wavefunction Inc. (Dr. Warren J. Hehre) (714) 955-2120, FAX (714) 955-2118. The guess transition structures generated by the original LST method described by Halgren and Lipscomb (included in GAUSSIAN and SPARTAN) have two serious defects: (1) bond lengths of forming and breaking bonds are much too long, (2) the hybridization at the reaction center is improper. I developed a new algorithm (XLST) which uses different interpolation formulas and produces reasonable bond length for forming and breaking bonds. Using the XLST structure, the transition structure is than located by the following procedure: (1) Fix the bond length of the forming and breaking bonds at the bond length calculated by XLST and perform 10 steps of constrained minimization in cartesian coordinates (This relaxes the improper hybridizations). (2) Calculate the full HESSIAN matrix. (3) Start the transition structure search using the HESSIAN matrix. This works very well even for complex isomerization, dissociation and addition reactions. Peter W. Rose Agouron Pharmaceuticals, Inc. Computational Chemistry 3565 General Atomics Court San Diego, CA 92121 From chjaim@unix1.sncc.lsu.edu Mon Feb 1 15:31:35 1993 Date: Mon, 1 Feb 1993 21:31:35 -0600 From: chjaim@unix1.sncc.lsu.edu (Jaime Combariza) Message-Id: <9302020331.AA19850@unix1.sncc.lsu.edu> To: CHEMISTRY@ccl.net Subject: 9,10-dicyanoanthracene ground and excited states Posted for someone else: Hi. Does anybody out there know of any type of calculation done on 9,10-dicyanoanthracene. Thanks