From chemistry-request@server.ccl.net Sat Mar 3 10:07:02 2001 Received: from compchem.dfh.dk (compchem.dfh.dk [130.225.225.51]) by server.ccl.net (8.11.0/8.11.0) with ESMTP id f23F71w19622 for ; Sat, 3 Mar 2001 10:07:01 -0500 Received: (from jeremy@localhost) by compchem.dfh.dk (SGI-8.9.3/8.9.3) id QAA57852 for chemistry@ccl.net; Sat, 3 Mar 2001 16:05:02 +0100 (MET) Date: Sat, 3 Mar 2001 16:05:02 +0100 From: Jeremy Greenwood To: chemistry@ccl.net Subject: Interpreting the quadrapole for aromatics Message-ID: <20010303160502.A157782@compchem.dfh.dk> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0i Dear all, I'm studying series of closely-related planar (halogen,oxo,methyl)-substituted azines. I'm interested in how I can interpret the multipoles, particularly the quadrapole. Unfortunately, they are anions. 1) Is the quadrapole as confounded as the dipole as regards choice of origin for an ion? The mass difference between F and Br will disturb the centre-of-mass origin of otherwise identical anions. 2) Using MP2/6-311+G(3df,2p) set I get quadrapoles like XX = -97 YY = -83 ZZ = -66 XY = -5.1 XZ = 0 YZ = 0 (orientation X, Y, 0) Can comparisons of e.g. ZZ and XY across the series give me an idea about the degree of pi-density/aromaticity, and an idea about electron density in within the ring vs. on the electronegative substituents? 3) Given the size, my options are HF, DFT or MP2, with large basis set including diffuse functions. Any reason to choose one over the other when comparing multipoles, electron density and aromaticity? I seem to recall B3LYP has a reputation for over-estimating aromaticity. 4) EPS-derived atomic charges from Gaussian (e.g. MKS) seem hard to interpret because fitting to the (origin dependent) dipole swamps the effects I am looking at. Would multipole fitting be a good way to go, or should I be looking at AIM or even Mullikan charges instead, bearing in mind that these are aromatics. Thanks for any enlightenment, Jeremy ---------------------------------------------------------------------- Jeremy Greenwood jeremy.greenwood@i.am Department of Medicinal Chemistry bh +45 35306117 Royal Danish School of Pharmacy fx +45 35306040 Universitetsparken 2, DK-2100 Copenhagen, Denmark ah +45 32598030 ---------------------------------------------------------------------- Never underestimate the bandwidth of a truck full of 1/4 inch tapes ---------------------------------------------------------------------- From chemistry-request@server.ccl.net Sat Mar 3 12:48:20 2001 Received: from mcmail.cis.mcmaster.ca (mattacf@[130.113.64.66]) by server.ccl.net (8.11.0/8.11.0) with ESMTP id f23HmJw20914 for ; Sat, 3 Mar 2001 12:48:19 -0500 Received: from localhost (mattacf@localhost) by mcmail.cis.mcmaster.ca with ESMTP id MAA09443 for ; Sat, 3 Mar 2001 12:48:04 -0500 (EST) Date: Sat, 3 Mar 2001 12:48:04 -0500 (EST) From: "C.F. Matta" To: CHEMISTRY@ccl.net Subject: SUMMARY: Volume of a Molecule within an Isodensity Surface Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Hi everyone, A couple of weeks ago I posted the following question: " I wonder is someone knows of a program that calculates the volume of a molecule contained within a particular density envelope (e.g. the 0.001 a.u. envelope) from a GAUSSIAN94 wavefunction or cube ? (The volume I want to calculate is the INTEGRATED volume within the envelope of choice AND NOT the volume computed from intersecting spheres). " Cherif F. Matta SUMMARY OF REPLIES: =================== Several very useful answers were received. Dr. Curt Breneman (Reply4) suggested a nice and simple algorithm, Dr. Hoefinger (Reply7) sent a useful awk script that I tried and obtained volumes that agree with those calculated from GAUSSIAN94 by invoking SCRF=IPCM as suggested by Dr. Doug Fox (Reply6). I would like to thank others who sent helpful comments and web pointers: Dr. N.O.J. Malcolm (Reply1: MORPHY program); Dr. Paulo Couto (Reply2: Moelekel program); Dr. Laurence Cuffe (Reply3: "volume" keyword); and Dr. TJ O'Donnell (Reply5: Codes to calculate molecular volumes and surfaces). INDIVIDUAL REPLIES: =================== Reply 1 ====== From: noj malcolm Our in-house code MORPHY (a general AIM program) does what you require, for details see our web-pages http://www.ch.umist.ac.uk/morphy/default.htm. Dr. N.O.J. Malcolm n.malcolm@umist.ac.uk Chemistry Department UMIST, Sackville Street Manchester, England. M60 1QD ------------------------------------------------------------------------------------------- Reply2 ====== From: Paulo Couto Moelekel visualization package allows you to plot isodensity surfaces from a Gaussian or Gamess output file. Furthermore, the volume enclosed by a given isodensity surface is computed. Please check http://www.cscs.ch/molekel/ ------------------------------------------------------------------------------------------- Reply3 ====== From: Laurence Cuffe I am probably mising something here but have you used the "volume" keyword? It seems to do what you want via a monte carlo integration procedure. ------------------------------------------------------------------------------------------- Reply4 ====== From: Curt M. Breneman One thing we used to do in the Wiberg group (back in my late 80's post-doc days) was to run the cube file through a little algorithm that counted the number of points with greater than or equal to 0.001 (or 0.002 e/au3) density values, then multiply that number by the cubic volume increment to get the volume. This works pretty well, particularly for small molecules and cubes that are around 81x81x81 points. More coarse grids also work, but with slightly lower accuracy. Curt Breneman RPI Chemistry ------------------------------------------------------------------------------------------- Reply5 ====== From: tj Reply-To: tjo@acm.org Cc: Jim Kress Most any 3D contouring software, for example those using the marching cubes algorithm, should output the volume (and area) of the "envelope". I have some old code that can do this. I could dig it up if you can't find anything else. It would most likely have to be modified to read G94 files. I have a vague recollection that I had added the feature to read G9x cube files. I had sent my code to Jim Kress who was using it with G94. I haven't heard from him in a while. I've copied him on this message and you might try getting in touch with him. and you might try getting in touch with him. My agreement with him was that he could use the code, but not distribute it. I have no problem with you using my code, as long as proper acknowledgement is given. TJ O'Donnell tjo@acm.org ------------------------------------------------------------------------------------------- Reply6 ====== From: Cust. Service Doug Reply-To: gaussian.com!help%gaussian.com@gaussian.com If you turn on #p running SCRF=IPCM you will get the integrated volume and surface area from the isodensity contour specified. Remember that this is supplied after the structure, dielectric and then the cutoff. The volume is printed after the first SCF when it gets into l117. Douglas J. Fox Technical Support Gaussian, Inc. help@gaussian.com ------------------------------------------------------------------------------------------- Reply7 ====== From: Siegfried Hoefinger If you've got a *.wfn file, then you could easily get a *.qub file with AIMPAC and the below attatched awk script should give you the volume. Note that the file that specifies the *.qub is called tmp.inf below and threshold becomes an isodensity value. ( example shown below ) Siegfried Hoefinger, M.S.,Ph.D. Universita' degli Studi di Bologna phone: [0039]/051 2099498 Dipartimento di Chimica "G. Ciamician" fax: [0039]/051 2099456 Via F. Selmi 2, I-40126, Bologna, Italy http://poseidon.gup.uni-linz.ac.at/sh/sh_frames.html The awk script should be cited as: ********************************** Hoefinger, S. (2000): Aimpacvol, a program to calculate molecular volumes, [unpublished, private communication]. Here is the script in full and notes on how to use it: ====================================================== 1.) Create a *.qub with AIMPAC from a *.wfn file: the qub specifying file is called 'tmp.inf' and the *.wfn file 'out.wfn' and the resulting *.qub fiel will be termed 'tmp.qub'. /SOME_PATH/AIMPAC/cubev tmp out *********** * tmp.inf * *********** TITLE: ch3co-/631+g INCRM: 10.0 0.3 0 CENTR: 0.0 0.0 0.0 PLANE: 0 1 2 NFUNC: 1 # # # Specfies a box of 10 Bohr x 10 Bohr x 10 Bohr, centered at 0/0/0 # with increments of 0.3 Bohr, computing the electron density ( NFUNC ) # considering no symmetry and leaving atoms 1 and 2 in the xy plane # of interest. The whole thing will result in a tmp.qub file. # NOTE: use increments not smaller than 1/100 of the boxlength to # avoid core dumps. # >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>> 2.) Estimate the volume with the help of awk script 'aimpacvol.awk' ********************* * aimpacvol.awk * ********************* # # PURPOSE: Counts the number of entries greater than a user defined # threshold value within a *.qub file produced from AIMPAC. # Based on this infinitesimal partial volume of one of these # grid cells, an estimation of the molecular volume inside # the specified threshold may be performed and the result # is presented in the end. In addition, all the coordinates # of these gridpoints having an electron density value # between exact the threshold and 15% above, are written # to another file called 'tmp.xyz' and may be used for # further processing, probably from awk-script 'fndclspnt.awk'. # USAGE: awk -f aimpacvol.awk tmp.qub tmp.inf threshold # AUTHOR: Siegfried Hoefinger, sh@mdy.univie.ac.at # DATE: 24.05.2000 # BEGIN{ # for( i=1; i 0 ) { for( i=1; i<=NF; i++ ) { CNT++ IZ=int( (CNT-1)/NXNY ) IY=int( (CNT-(IZ*NXNY)-1)/NX ) IX=CNT-(IZ*NXNY)-(IY*NX)-1 if( $i >= THRESH ) { SUCCESSCNT++ if( $i <= (THRESH*1.15) ) { X=STX+(IX*INCR) Y=STY+(IY*INCR) Z=STZ+(IZ*INCR) printf "%12.6lf %12.6lf %12.6lf\n", X, Y, Z > "tmp.xyz" } } } } } END{ printf "%10s %dx%dx%d %12s\n", "There were", NX, NY, NZ,"gridpoints" printf "%d %18s\n", CNT, "actually counted" printf "%d %18s\n", SUCCESSCNT, "inside threshold" printf "%12s %12.6lf %9s %12.6lf %7s\n", \ "gridvolume::", INCR*INCR*INCR, " Bohr^3 =", \ INCR*INCR*INCR*0.1481847435, " Angs^3" printf "%25s %12.6lf %9s %12.6lf %7s\n", "volume inside threshold::", SUCCESSCNT*INCR*INCR*INCR, " Bohr^3 =", \ SUCCESSCNT*INCR*INCR*INCR*0.1481847435, " Angs^3" } ------------------------------------------------------------------------------------------- END OF REPLIES ============== Cheers, Cherif ___________________________________________________________________________ Cherif F. Matta tel. (905) 525-9140 ext. 22502 Chemistry Department fax (905) 522-2509 McMaster University Hamilton, Ontario, CANADA L8S 4M1. ___________________________________________________________________________