From jabs@chemie.uni-halle.dbp.de Fri Dec 4 14:58:28 1992 Date: Fri, 4 Dec 1992 13:58:28 +0100 From: jabs@chemie.uni-halle.dbp.de Message-Id: <921204135827*/S=jabs/OU=chemie/PRMD=UNI-HALLE/ADMD=DBP/C=DE/@MHS> To: chemistry@ccl.net Subject: IR_MOPAC Hi, has anyboby experiences in the AM1/PM3 calculation of oligopeptides and the correlation of calculated spectra with experimental data especially for intra- and intermolecular hydrogen bonding ? Andreas Jabs M.-Luther-University Department of Chemistry Theoretical Chemistry Group Weinbergweg 16 O-4020 Halle/Saale PF 8 e-mail: c=de;a=dbp;p=uni-halle;o=chemie;s=jabs From ross@cgl.ucsf.EDU Fri Dec 4 01:23:45 1992 Date: Thu, 3 Dec 92 22:23:34 -0800 Message-Id: <9212040623.AA29850@socrates.ucsf.EDU> From: ross@cgl.ucsf.edu (Bill Ross) To: chemistry@ccl.net Subject: Online Phys. Desk Ref.? Is there anything like a Physician's Desk Reference for drugs on the net? Bill Ross From khillig@Chem.LSA.umich.edu Fri Dec 4 04:55:34 1992 Date: Fri, 4 Dec 92 09:55:34 -0500 From: khillig@Chem.LSA.umich.edu Message-Id: <9212041455.AA09095@Be> To: Lawrence.Lohr@UM.CC.UMich.EDU, chemistry@ccl.net Subject: What's the best method for calculating carbodiimides? I've been asked by one of our students about how to calculate energies of some carbodiimides, as an aid to understanding why he hasn't been able to make some of these while others can be synthesized. As I'm not an expert in this kind of molecule, I'd appreciate any pointers from you. The molecules in question are of the form R1-N=C=N-R2, where he's interested in the following: R1 R2 CH3 CH3 C6H5 C6H5 CH3 C6H5 p-C6H4NO2 p-C6H4NO2 -CN -CN (the last is NC-N=C=N-CN) Tools available are Mopac 6.0 and Gaussian 90 on an SGI 4D/360. Thanx! From lim@rani.chem.yale.edu Fri Dec 4 07:01:49 1992 From: Dongchul Lim Message-Id: <9212041701.AA20242@rani.chem.yale.edu> Subject: Random Distribution of Points on a Sphere To: chemistry@ccl.net (Computational Chemistry) Date: Fri, 4 Dec 92 12:01:49 EST Hi fellow chemists, This may fit better in comp.graphics, but someone in this group may have experienced it already. So here it goes. I want to generate points on the surface of a sphere as randomly as possible. These points may be used for drawing van der Waals surface of an atom. I tried with symmetrically distributed points (generated by varying phi and theta, etc), but due to symmetry, the points looked like marching ants at certain angles. Since many of symmetrically distributed points are likely to be overlapped when they are placed in 2-D screen, increasing the density of VDW points would require unnecessarily large number of sphere points. Random distribution of points would help minimizing the number of points required and will give a better appearance. Any ideas? -DCL lim@rani.chem.yale.edu From COFFEY@bert.chem.wisc.edu Fri Dec 4 06:55:00 1992 Message-Id: <22120412552802@bert.chem.wisc.edu> Date: Fri, 4 Dec 92 12:55 CST From: "Martin J. Coffey" Subject: G92 stumper... To: chemistry@ccl.net Gaussian Jocks, This is the situation. The molecule is acetylene. The ground state is 1-AG and the excited state is 1-AU. I want to calculate equilibrium geometries and extract some vibrational frequencies. The ground state I have calculated using several methods (RHF,MP2,CAS(4,4)). The excited state is a problem. The best I have done so far is to start off with a 1-AG wavefuntion used as a guess for the UHF calculation. I alter the LU and HO in the alpha orbitals and wind up with a wavefunction of 1-AU symmetry. The initial guess has an =1.000 and after optimization the wavefunction still has and = 1.015. This indicates a lot more triplet contamination than I would be happy with. It manages to converge to a geometry that is not bad when compared to the experimental geometry, but it could be better. UMP2 calculations do not improve the situation much. I have tried to get CAS(UNO,...) to use the UHF wavefunction and optimize from there, but I am not having much luck. The questions that I would appreciate some help with are: 1) Is there a way to control the amount of triplet contamination that the UHF calculation allows to come in, or is there a trick one can use to reduce it by starting from another initial guess? 2) Is it possible to have G92 converge to a 1-AU solution for a molecule such as this where it has 14 electrons and it wants to try to doubly occupy all the orbitals? This is particularly the problem with the CAS(UNO,...) calculation that I am doing. It seems to read in the UHF wavefunction, mix the alpha and beta orbitals and, then it converges it back to 1-AG symmetry. I would appreciate any suggestions. Replies can be mailed directly to me. Martin J Coffey coffey@chem.wisc.edu UW-Madison From topper@haydn.chm.uri.edu Fri Dec 4 09:11:45 1992 Date: Fri, 4 Dec 92 14:11:45 -0500 From: topper@haydn.chm.uri.edu (Robert Q. Topper) Message-Id: <9212041911.AA14051@haydn.chm.uri.edu> To: chemistry@ccl.net Subject: Re: Random Distribution of Points on a Sphere In response to Dongchul's question, > Hi fellow chemists, > This may fit better in comp.graphics, but someone in this group > may have experienced it already. So here it goes. > I want to generate points on the surface of a sphere as randomly > as possible. These points may be used for drawing van der Waals > surface of an atom. I tried with symmetrically distributed points > (generated by varying phi and theta, etc), but due to symmetry, > the points looked like marching ants at certain angles. > Since many of symmetrically distributed points are likely to be > overlapped when they are placed in 2-D screen, increasing the > density of VDW points would require unnecessarily large number > of sphere points. Random distribution of points would help > minimizing the number of points required and will give a better > appearance. > Any ideas? > -DCL > lim@rani.chem.yale.edu Check Appendix G.4 of Allen and Tildesley's wonderful book "Computer Simulation of Liquids" (Clarendon Press, Oxford, 1987) for an algorithm for uniformly sampling the surface of a sphere. -RQT ******************************** * Robert Q. Topper * * Department of Chemistry * * University of Rhode Island * * Kingston, RI 02881 * ******************************** * rtopper@chm.uri.edu OR * * topper@haydn.chm.uri.edu * * (401) 792-2597 [office] * * (401) 792-5072 [FAX] * ******************************** From HACR0002@SMUVM1.bitnet Fri Dec 4 08:01:41 1992 Message-Id: <199212042017.AA23355@oscsunb.ccl.net> Date: Fri, 04 Dec 92 14:01:41 CST From: HACR0002%SMUVM1.BITNET@OHSTVMA.ACS.OHIO-STATE.EDU Organization: Southern Methodist University Subject: distribution of points on a sphere To: CHEMISTRY@ccl.net The original question is: >From: Dongchul Lim >Subject: Random Distribution of Points on a Sphere > Hi fellow chemists, >This may fit better in comp.graphics, but someone in this group >may have experienced it already. So here it goes. >I want to generate points on the surface of a sphere as randomly >as possible. These points may be used for drawing van der Waals >surface of an atom. I tried with symmetrically distributed points >(generated by varying phi and theta, etc), but due to symmetry, >the points looked like marching ants at certain angles. >Since many of symmetrically distributed points are likely to be >overlapped when they are placed in 2-D screen, increasing the >density of VDW points would require unnecessarily large number >of sphere points. Random distribution of points would help >minimizing the number of points required and will give a better >appearance. >any ideas? >DCL >lim@rani.chem.yale.edu To Lim: There are several ways to get points on a sphere with even distribution. For example, vertics and (projects of) face-centers of pentakisdodecahedron are 92 points on the sphere. In this way, you can have more or less. Please see J.Chem.Phys. 97(1992)4162 for 92 points, see J.Compt.Chem.8(1987)778 for 60 points. Bingze Wang From dan@omega.chem.yale.edu Fri Dec 4 11:12:11 1992 From: Dan Severance Message-Id: <9212042112.AA08114@omega.chem.yale.edu> Subject: Re: Solvation calculation To: chemistry@ccl.net Date: Fri, 4 Dec 92 16:12:11 EST Organization: Laboratory for Computational Chemistry Hi, > > I have a question concerning the solvation calculation of reaction path. > Let's say we want to study CO2 and HO- interaction at long distance (9 > angstrom). There are two ways to do this: > > a). Do two separate calculations. > b). Do one calculation treat CO2 + HO- as a supermolecule. > Can these two calculations give same results? In principle, they should > since at 9 anbstrom the interaction should be almost nothing. > These are not likely to give the same solvation energies for a couple of reasons: The assumption that the interaction energy is nil at 9 Angs. is not a particularly good one, especially with charged molecules. Coulombic Energies fall off quite slowly with distance. Even if it the Solute-Solute interaction were nil, you need to remember that you are interested in the solvation energy. The solvation shells about each of the two molecules is greatly perturbed in the presence of the other. You can think of them as overlapping spheres. A solvent molecule in the area between the two molecules is very much perturbed since it is only around 4-5A from each solute, even though the solutes themselves are 9A apart... In my study of benzene dimers in water, (Jorgensen and Severance, JACS 112, p. 4768, 1992), The PMF had had not leveled out completely even after 10A. The solute-solute interactions by that point were indeed quite small (<0.2 kcal/mol if memory serves), but the solvation spheres still overlapped severely (12 A solute-solvent cut-off). Dan Severance dan@omega.chem.yale.edu From rupley!local@cs.arizona.edu Sat Dec 5 02:18:13 1992 Date: Sat, 5 Dec 92 02:18:13 GMT From: rupley!local@cs.arizona.edu Message-Id: <9212050218.AA24053@rupley> To: arizona!ccl.net!chemistry@cs.arizona.edu Subject: Re: distribution of points on a sphere Bingze Wang writes: # The original question is: # >From: Dongchul Lim # >Subject: Random Distribution of Points on a Sphere # > Hi fellow chemists, # >This may fit better in comp.graphics, but someone in this group ^^^^^^^^^^^^^ Yes, if you want a probably overkill number of responses (:-). The question is nearly an faq. And it doesn't lack interest. # >may have experienced it already. So here it goes. # >I want to generate points on the surface of a sphere as randomly # >as possible. These points may be used for drawing van der Waals # >surface of an atom. I tried with symmetrically distributed points # >(generated by varying phi and theta, etc), but due to symmetry, # >the points looked like marching ants at certain angles. # # To Lim: # There are several ways to get points on a sphere with even distribution. # For example, vertics and (projects of) face-centers of # pentakisdodecahedron are 92 points on the sphere. In this way, # you can have more or less. Please see J.Chem.Phys. 97(1992)4162 for # 92 points, see J.Compt.Chem.8(1987)778 for 60 points. # Bingze Wang There are various and different questions regarding the distribution of points on the surface of a sphere. Among these are: 1. How do you get a "uniform" or "even" distribution? The question was addressed by Bingze Wang. 2. How do you get a random distribution? Where each element of the surface has equal probability of having a point, but probability being what it is, the points may not be equidistant. 3. How do you get a distribution that is approximately random and uniform? I suspect this is what Dongchul Lim was asking about. One algorithm for doing this is, put point charges randomly on the surface of the sphere and minimize the energy. I will be happy to send interested people a _totally unedited_ collection of postings from comp.graphics on this thread. John Rupley rupley@cs.arizona.edu