From dave@carbon.chem.csiro.au Mon Jul 19 12:38:22 1993 Message-Id: <199307190738.AA01713@shark.mel.dit.csiro.au> Date: Mon, 19 Jul 93 17:38:22 EST From: (Dr.) Dave Winkler Subject: DFT/ab initio calcs on tungsten To: chemistry@ccl.net Dear Netters, I have some inorganic compounds which contain tungsten and cobalt. I note that all of the codes in Unichem, including Dgauss, will only handle elements up to Xe. Do any semiempirical or ab-initio (especially DFT) codes handle tunsgten? Cheers, 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. "Life is what happens to you while you're making other plans" From shenkin@still3.chem.columbia.edu Mon Jul 19 05:18:19 1993 Date: Mon, 19 Jul 93 09:18:19 -0400 From: shenkin@still3.chem.columbia.edu (Peter Shenkin) Message-Id: <9307191318.AA05091@still3.chem.columbia.edu> To: Dongchul Lim , Subject: Re: SGI..HP..IBM Workstations > From: Dongchul Lim > Especially, [[ AIX's ]] virtual memory manager (VMM) is very different > from other UNIX > systems. For example, when the program requests an allocation of memory (using > malloc), the AIX returns a pointer to the memory location like other UNIX system. > But unlike ordinary UNIX, the memory is not actually allocated in the AIX until > the program accesses it. If the memory is not available at that time, the AIX > generates a SIGDANGER signal and finally kills the program to free the space. Readers who are interested in this ought to know that virtually all UNIXes are moving or have moved to the memory-allocation paradigm that Lim ascribes to AIX, mostly for reasons of compliance with the POSIX-mandated semantics of the fork() system call. SGI has used a similar scheme since IRIX 4.0; I believe HP has recently moved to this sort of scheme, and I'd be surprised if you do not see the same thing under the new Solaris. Can anyone speak authoritatively about DEC Alpha/OSF? If so, is the allocation scheme mandated by OSF? There are many arguments on both sides of this issue, but the new paradigm ("bind physical resources to the process on first use of a variable") makes better overall use of a system's total resources. The old paradigm might be described, "bind physical resources when a variable is declared." Conceptually, this is at startup time for programs which do not do dynamic storage allocation (e.g., standard Fortran programs). The old paradigm ties up large amounts of RAM/swap which may never be touched in a given run, which makes them unavailable for other processes. On the other hand, the old paradigm is deterministic for a single process: in the Fortran example, if the program succeeds in starting up, it is guaranteed that all memory declared will be available in an attempt is made to use it, since RAM/swap corresponding to all declared varaibles is bound to the process at start-up time. Analogous differences occur for programs which do perform dynamic memory allocation, such as standard C programs calling malloc(). Under the old paradigm, when malloc() was called, the system would return a valid pointer if the requested storage was available. At that point RAM/swap would have been allocated. The storage would therefore certainly be there if a subsequent attempt was made to use it. If there wasn't enough storage available at the moment malloc() was called, malloc() would return NULL. Under the new paradigm, however, malloc() will return a valid pointer even if the resources requested exceed the amount of RAM/swap available at the moment. The system will then kill the job (or some other job, such as the largest process running) as soon as some process tries to access (e.g., write to) a memory location that exceeds the available RAM/swap available then. -P. ************************f*u*cn*rd*ths*u*cn*gt*a*gd*jb************************ Peter S. Shenkin, Box 768 Havemeyer Hall, Dept. of Chemistry, Columbia Univ., New York, NY 10027; shenkin@still3.chem.columbia.edu; (212) 854-5143 ********************** Wagner, Beame, Screvane in '93! ********************** From WILLSD@conrad.appstate.edu Mon Jul 19 05:58:51 1993 Date: Mon, 19 Jul 1993 09:58:51 -0400 (EDT) From: WILLSD@conrad.appstate.edu Subject: vibrational mode animation To: CHEMISTRY@ccl.net Message-Id: <01H0Q0TFM2W299DH2X@conrad.appstate.edu> Some time ago there was a thread on ccl discussing tools for computational chemistry. I'd like to announce a somewhat specialized tool that may be of interest to some of you. There is a simple program available via anonymous ftp from cardinal.ncsc.org that will read the output from a GAUSSIAN92 frequency job and produce a file that xmol will read to display an animation of any selected vibrational mode. This is particularly handy for visualizing the reaction coordinate for a transition state. The program consists of two files, animate.c and animate.dat. To get them ftp to cardinal.ncsc.org, login as anonymous, and give your email as the password. then do the following: cd pub/utils/xmol mget an* quit The program may be compiled as follows: cc -o viban animate.c -lm The -lm may not be required on your system, but it is on dec 5000. The program will(viban) prompt you for the name of the gaussian file, the name of the xmol file (you end this file name in .xyz), the mode to be visualized, the desired number of animation steps, and a scale factor (start with .5). The program will produce an xmol file in the .xyz format that contains structures that step through the vibrational mode. This program was written as a class assignment by a computational chemistry student, Warren Harper, who will be entering the University of Kentucky graduate program in the fall. It has been tested with a number of smallish (less than 10 atoms) inorganic molecules calculated with g92, and has been tested on VAX and decstation platforms. The comments should explain how the program works. I hope you enjoy it.... Steve Williams Department of Chemistry Appalachian State University Boone, NC 28608 willsd@stat.appstate.edu From raman@bioc01.uthscsa.edu Mon Jul 19 03:54:40 1993 From: raman@bioc01.uthscsa.edu (C.S.RAMAN) Message-Id: <9307191354.AA00351@bioc01.uthscsa.edu> Subject: Message flood... :( To: lim@rani.chem.yale.edu (Dongchul Lim) Date: Mon, 19 Jul 1993 08:54:40 -0500 (CDT) Dear Netters Is it just me or everyone receiving the following post over and over again? So, far I have received the same message > 40 times! :( I am hoping that this problem will be rectified soon! Thanks -raman > > gerson@vnet.IBM.COM writes: > > Earlier in the long list of appends, Tom Pierce of Rohm and Haas reported that > > you need to "reallocate how memory is partitioned" on the IBM to improved > > performance for interactive tasks. Latter on, there were several appends > > to this bulletin board which discussed compiler options for improved > > performance > > > > The following may not belong to this thread, but it may be helpful to some readers. > There was a review on IBM RS/6000 system (SunWorld Page 53-58, February 1993 issue). > RISC workstations run AIX, IBM's coporate UNIX implementation. > The AIX has some innovative features, such as dynamically linked libraries at > the kernel level and the pagable kernel. This indicates that RISC is a good > choice for a real time work. However, the AIX is not a UNIX which we got used to. > Especially, its virtual memory manager (VMM) is very different from other UNIX > systems. For example, when the program requests an allocation of memory (using > malloc), the AIX returns a pointer to the memory location like other UNIX system. > But unlike ordinary UNIX, the memory is not actually allocated in the AIX until > the program accesses it. If the memory is not available at that time, the AIX > generates a SIGDANGER signal and finally kills the program to free the space. > What's the difference? In other UNIX systems, programmers know if the memory > is available when they try to allocate it. On AIX, they don't until they get > a SIGDANGER signal from the operating system. > Other than memory menaging problem, the review was mostly praising RISC's great > performance. > -DCL > > > > --- > 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 > --- > > -- C.S.Raman raman@bioc01.uthscsa.edu - Internet UNIX Programming & Administration 70412.2354@compuserve.com - CIS SPARC & SGI Systems raman@hermes.chpc.utexas.edu - CHPC Department of Biochemistry craman@launchpad.unc.edu UTHSCSA 7703 Floyd Curl Dr. (210) 567-6623 [Tel] San Antonio, TX 78284-7760 (210) 567-6595 [Fax] ****************************************************************************** If a man's wit be wandering, let him study the Mathematics -Francis Bacon ****************************************************************************** From DSMITH@uoft02.utoledo.edu Mon Jul 19 05:27:57 1993 Date: Mon, 19 Jul 1993 10:27:57 -0500 (EST) From: "DR. DOUGLAS A. SMITH, UNIVERSITY OF TOLEDO" Subject: azaaromatics and vibronic coupling To: chemistry@ccl.net Message-Id: <01H0Q16II80I00038J@UOFT02.UTOLEDO.EDU> In a recent paper (J. Org. Chem. 1993, 58, 6) we reported on the geomietric distortion of the triplet nitrenium ion derived from azepine (C6H6N+), although we could not distinguish between Hartree-Fock doublet instability and vibronic coupling-induced pseudo Jahn Teller distortion as the cause. In trying to sort this out, I have several questions to put to the readership of this mailing list. 1. In other azaaromatic systems such as pyridine and pyrimidine, vibronic coupling of low lying triplet states is invoked as the reason for geometric distortions. These distortions are out-of-plane deformations, as opposed to the in-plane distortion we observed in the azepine system. Is this because the two former systems are six-membered rings while azepine is a seven atom heterocycle? 2. Are there any examples of vibronic coupling-induced geometric distortions for other homoaromatic systems, e.g. tropylium? If so, are these in-plane or out-of-plane distortions? 3. Are there any examples of vibronic coupling-induced geometric distortions for three-membered ring aromatic systems, e.g. cyclopropenyl cation? This system would necessarily be planar, so any distortion should be (IMHO) in-plane. 4. In the absence of experimental data, how does one computationally determine the existence of vibronic coupling? From my reading of the literature (e.g. Buma, et al., J. Am. Chem. Soc. 1992, 114, 9544) I cannot see how coupling is unambiguously determined. In the paper cited, the authors calculated several low lying excited triplet states and examined the normal mode frequencies,then they correlated the direction of the geometric distortions in each state. This somehow led them to the conclusion that vibronic coupling was occuring. (I am only an organic chemist, their arguments just don't make enough sense to me.) Can someone help explain this to me? 5. Furthermore, the explanation by these authors based on electronic structure ("We argue that an azabenzene in an npi* state may well be susceptible to out-of-plane distortions because the pi-electron system in this kind of molecules [sic] becomes in first approximation a 7-pi-electron system and thus is no longer aromatic.") seems to me to be just handwaving of the type an organic chemist would use. The explanation seems empirically valid but does not really explain anything, especially in light of our results for the azepine system. On the other hand, if this explanation really is valid, then perhaps our azepine results are due to HF doublet instability rather than vibronic coupling. Help! 6. These authors also use energetic arguments that don't seem to me to "prove" vibronic coupling. For example, "The excitation to the 3A2(npi*) state... results in a large out-of-plane distortion o fhte C2-H7 fragment with a large stabilization energy, indicating that the vibronic coupling between the 3A2(npi*) and 3B2(pipi*) states is considerably stronger than the vibronic coupling between the 3B1(npi*) and 3A1(pipi*) states [where the stabilization energy was small]." This sounds like circular reasoning to me, i.e. vibronic coupling exists because there is a large stabilization energy due to vibronic coupling. Or am I missing the point? On a related topic, another question. When doing a CASSCF calculation on a triplet state of any molecule, is it absolutely necessary to include both unpaired electrons and both singly occupied orbitals (and their corresponding antibonding orbitals)? In particular, when calculating the triplet state of a nitrenium ion, should the two unpaired electrons formally on the nitrogen be tightly coupled or correlated? Please direct replies to the net, in order to foster discussion and further insights. Doug Douglas A. Smith Assistant Professor of Chemistry The University of Toledo Toledo, OH 43606-3390 voice 419-537-2116 fax 419-537-4033 email dsmith@uoft02.utoledo.edu From DSMITH@uoft02.utoledo.edu Mon Jul 19 06:51:26 1993 Date: Mon, 19 Jul 1993 11:51:26 -0500 (EST) From: "DR. DOUGLAS A. SMITH, UNIVERSITY OF TOLEDO" Subject: GB/SA solvation - what charges? To: mmodinfo@uoft02.utoledo.edu, chemistry@ccl.net Message-Id: <01H0Q4N1RF8800038J@UOFT02.UTOLEDO.EDU> When using the GB/SA (generalized Born solvent accessible surface area) continuum solvation method implemented in MacroModel, I have heard at various meetings and talks that the default atomic charges in the force fields are not good to use. 1. What force field is preferred for the GB/SA method? Is any one to be avoided? If so, why? 2. How should the charges be calculated? Ab initio or MNDO ESP, scaled or unscaled? CHELPG? Others? 3. Obviously, charges are dependant upon the conformation of the molecule as well as the local environment. But the force field parameters are valid only for the charges included in the ff - although I recognize that all the parameters are averages for a variety of structures and conformations. How much does using externally derived charges invalidate the force field parameters? In other words, will using externally derived charges with, say, the AMBER all atom parameters lead to spurious, invalid or "incorrect" structures and/or energies? Doug Douglas A. Smith Assistant Professor of Chemistry The University of Toledo Toledo, OH 43606-3390 voice 419-537-2116 fax 419-537-4033 email dsmith@uoft02.utoledo.edu From lim-dongchul@ycc.YALE.EDU Mon Jul 19 08:11:12 1993 Date: Mon, 19 Jul 1993 12:11:12 -0400 From: Dongchul Lim Message-Id: <199307191611.AA07838@MINERVA.CIS.YALE.EDU> To: chemistry@ccl.net Subject: apology Dear netters, I was very embarassed when I found my message has been posted millions of times. Our system doesn't have any problem with mail. So I guess there was (still is) a problem at OSU. I hope even this message won't be posted multiple times. I'm sorry about the multiple messages but I'm not to blame. Please don't send me any more hate/harass mail. I've been already harassed enough. -DCL From shenkin@still3.chem.columbia.edu Mon Jul 19 08:52:29 1993 Date: Mon, 19 Jul 93 12:52:29 -0400 From: shenkin@still3.chem.columbia.edu (Peter Shenkin) Message-Id: <9307191652.AA07925@still3.chem.columbia.edu> To: "DR. DOUGLAS A. SMITH, Subject: Re: GB/SA solvation - what charges? Doug Smith writes: > When using the GB/SA (generalized Born solvent accessible surface area) > continuum solvation method implemented in MacroModel, I have heard at various > meetings and talks that the default atomic charges in the force fields are > not good to use. Not necessarily; read on. > 1. What force field is preferred for the GB/SA method? Is any one to > be avoided? If so, why? I recently visited a large company where a detailed study is in progress comparing a variety of different approaches to the calculation of solvation free-energies with experiment. I won't identify the place/person because I don't want to violate confidentiality, but the study will eventually be published. Also, since I didn't do the work myself, bear in mind that I might have gotten a detail or two wrong. Based on results using a variety of typical organic molecules, the finding was that using MacroModel with the OPLS force-field gave very good agreement with experiment for DELTA-G's of hydration. AMBER* gave poor agreement with DELTA-G's of hydration, but very good agreement for DELTA-DELTA-G's; that is, if one were to consider, say, the DELTA-Ghyd of benzene (or anything else) as a reference, and measure the other DELTA-G's with regard to it, the deviations from the reference, but not the absolute values, were good for AMBER* GB/SA from MacroModel. In our lab, we have carefully reparameterized AMBER* charges for peptide and peptide-like substructures, and feel that AMBER* would probably give best results (better than OPLS, which might in addition have a number of parameters that are missing altogether) for peptides. So where do we stand with regard to the interaction of, say, proteins with non-peptide-like drugs? AMBER* will presumably give good DELTA-DELTA-G's of hydration (taking one particular structure as a reference), but there's probably not any one force-field properly parameterized to give good DELTA-G's of hydration for both drug-like and peptide-like stuctures. We will probably be reparameterizing AMBER* charges to improve its GB/SA performance for "typical" organic molecules. Oh -- MM2 charges give lousy values of both DELTA-G's and DELTA-DELTA-G's of hydration, in the current implementation, so stay away from it if you're interested in solvation energies. > 2. How should the charges be calculated? Ab initio or MNDO ESP, scaled > or unscaled? CHELPG? Others? I'll let Quentin McDonald address this, if he's listening in. > 3. Obviously, charges are dependant upon the conformation of the molecule > as well as the local environment. ....How > much does using externally derived charges invalidate the force field > parameters? In other words, will using externally derived charges with, say, > the AMBER all atom parameters lead to spurious, invalid or "incorrect" > structures and/or energies? Quentin might have more to say about this, but I'll mention three things. First, you have to be careful mainly about torsional parameters. After altering charges, you *must* check torsional barriers and adjust torsional parameters to reproduce the values the barriers had with the old charges. Ideally, one should do a full reparameterization with each change, but even we don't do this. Second, the study I mentioned above found that OPLS with newly calculated charges (from Gaussian-92) didn't do quite as well as Jorgensen's own "tweaked" OPLS parameters, which implies that one can do better than a simple-minded reparameterization if one is very careful to do a global reparameterizion. Third, a possible "wave of the future" is a paradigm which is capable of altering charge sets on the fly during a simulation in an intelligent way, as the conformation changes. Nobody does this -- or has parameters for doing this -- right now. -P. ************************f*u*cn*rd*ths*u*cn*gt*a*gd*jb************************ Peter S. Shenkin, Box 768 Havemeyer Hall, Dept. of Chemistry, Columbia Univ., New York, NY 10027; shenkin@still3.chem.columbia.edu; (212) 854-5143 ********************** Wagner, Beame, Screvane in '93! ********************** From st-amant@cgl.ucsf.EDU Mon Jul 19 04:42:32 1993 Date: Mon, 19 Jul 93 11:42:32 -0700 Message-Id: <9307191842.AA00254@socrates.ucsf.EDU> From: st-amant@cgl.ucsf.edu (Alain St-Amant) To: chemistry@ccl.net, dave@carbon.chem.csiro.au Subject: Re: DFT/ab initio calcs on tungsten > Dear Netters, > > I have some inorganic compounds which contain tungsten and cobalt. I note > that all of the codes in Unichem, including Dgauss, will only handle elements > up to Xe. Do any semiempirical or ab-initio (especially DFT) codes handle > tunsgten? > > Cheers, > > Dave I know that a DFT code I wrote called deMon can handle tungsten, but the catch is that to do it, you must keep it's f electrons frozen in a model core potential (I know that with gold, others got fairly good answers with such a scheme). I don't know if that option is supported anymore. The code is I think now being marketed commercially (if not, in the near future) by Biosym, but I am not sure. alain st-amant department of pharmaceutical chemistry university of california, san francisco From jas@medinah.atc.ucarb.com Mon Jul 19 09:48:10 1993 Message-Id: <9307191849.AA17395@medinah.atc.ucarb.com> Date: Mon, 19 Jul 1993 14:48:10 -0500 To: chemistry@ccl.net From: jas@medinah.atc.ucarb.com (Jack Smith) Subject: Re: DFT/ab initio calcs on tungsten David Wrinkler wrote: >I have some inorganic compounds which contain tungsten and cobalt. I note >that >all of the codes in Unichem, including Dgauss, will only handle elements up to >Xe. Do any semiempirical or ab-initio (especially DFT) codes handle tunsgten? The DMOL program (from Biosym) can handle elements beyond Xe if you're willing to go through some extra gyrations. They've intentionally made it difficult to go beyond Xe to force users to realize they're on thin ice regarding relativistic effects. The Biosym response center (at rcenter@biosym.com) can lead you through the necessary steps once they've expelled with all the appropriate caveats regarding the exclusion of relativistic effects. If you have any problems beyond that, they'll probably put you in direct contact with Mike Wrinn (but I wouldn't start there). -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= JACK A. SMITH ...................................................................... Union Carbide Corp. || Phone: (304) 747-5797 Catalyst Skill Center || FAX: (304) 747-5571 P.O. Box 8361 || S. Charleston, WV 25303 || Internet: jas@medinah.atc.ucarb.com -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= From bruce@dggpi2.chem.purdue.edu Mon Jul 19 09:09:59 1993 Date: Mon, 19 Jul 93 14:09:59 -0500 From: bruce@dggpi2.chem.purdue.edu (Bruce Luxon) Message-Id: <9307191909.AA15127@dggpi2.chem.purdue.edu> To: chemistry@ccl.net Subject: MORASS: NOESY CONSTRAINED MD PROGRAM ANNOUNCEMENT: MORASS 2.0 The program MORASS, (Multispin Overhauser Relaxation AnalysiS and Simulation) analyzes 2D NMR NOESY data of nucleic acids and proteins to evaluate cross-relaxation rates (sigma_ij) from which interproton distances are obtained. NOESY derived interproton distances are used to conduct constrained molecular dynamics and molecular mechanics studies on proteins and DNA/RNA. MORASS was written in the labs of Professor David Gorenstein at Purdue University in W. Lafayette, IN. MORASS calculations are carried out by matrix methods involving the relaxation rate matrix specified by the set of simultaneous Bloch equations. MORASS also calculates a NOESY spectrum given a model structure including proton coordinate positions. An important aspect of MORASS is that it has the capability to incrementally merge the experimental NOESY values into the simulated matrix. This has a great deal of value in that it permits the experimental values to be "massaged in" gradually allowing a reasonably continuous transformation of the model structure into the experimentally consistent one. This is particularly important for experimental structures that deviate significantly from model-built ones. These are often some of the most interesting ones in practice for DNA, RNA and Proteins. Development of MORASS was motivated in large part by a need to analyze NOESY data measured from proteins and oligonucleotides. Although application of MORASS is not limited to these macromolecules, certain features characteristic to them have been adopted in the standard operation of MORASS (e.g. residue and atom names). MORASS accepts cartesian files in AMBER, CHARMm and PDB format. Others are relatively easy to format into the source code. Output files necessary to conduct constrained molecular dynamics is provided in a variety of formats including AMBER (both harmonic and flatwell) and SANDER. Other formats are again, relatively easy to add. A multiple mixing time version is also available on specific request to me. The accuracy of the matrix method for determining interproton distances >from 2D NOESY data has been examined in the literature. These studies and other details of MORASS have been reported in: > Nikonowicz et al, Bull. Magn. Reson.,11,226(1989) > Post et al, JACS,112,6796(1990) > Powers et al, J. Biomol. Struct. Dyn.,8,253(1990) > Nikonowicz et al, Biochem.,29,4193(1990) > Nikonowicz et al, Biochem.,30,1323(1991) > Meadows et al, Bull. Magn. Reson.,13,22(1991) > etc... MORASS is written in standard FORTRAN77 for portability. We have compiled and executed the MORASS code succesfully on our uVAX, SGI IRIS 4D, SUN 4-110, SGI INDIGO and DECstation 3100 machines. An IBM RISC system 6000 version is also available. MORASS does not produce graphics files at this time. We in no way warrant the software MORASS or any of its results. Nor are we liable for any use which may be made of the software. The MORASS software is under copyright protection. We request that any and all versions of MORASS not be distributed except by Prof. Gorenstein's group. Publications of results obtained with MORASS should acknowledge its use by an appropriate citation. We would appreciate a copy of any publications using MORASS or at least a reference (email to me is fine). Obtain MORASS 2.0 by ftp anonymous: 1> ftp dggpi2.chem.purdue.edu 2> name: anonymous 3> password: your email address 4> cd pub/morass (For the RISC system 6000 version: cd pub/morass/r6000) 5> get morass.tar.Z 6> quit morass.tar.Z contains all the source code, documentation(in LaTex), a unix "help" script and test input files. Use the following unix commands to uncompress and "detar" the files: %> uncompress morass.tar.Z %> tar xvf morass.tar If you have difficulties please feel free to contact me anytime. ____________________________________________________________________________ | | | |"You can't always get what you want, | Dr. Bruce A. Luxon | | But if you try sometimes, | Chemistry Department | | You'll get what you need ..." | Purdue University | | | W. Lafayette, IN 47907 | | Mick Jagger | (317)494-5289; Fax (317)494-0239 | | | bruce@dggpi2.chem.purdue.edu | |______________________________________|___________________________________| From jtgolab@amoco.com Mon Jul 19 10:03:34 1993 Date: Mon, 19 Jul 1993 15:03:34 -0500 From: jtgolab@amoco.com Message-Id: <9307192003.AA15365@vacc47.nap.amoco.com> To: CHEMISTRY@ccl.net Subject: Re: G92 Animations In regards Steve Williams recent posting about animating G92 vibs in Xmol, I would like to add that QCPE has a program called VIBRATE that also reads G92 files and displays them on the workstation (SGI) directly. Furthermore, VIBRATE is easy to modify so that other program's vibrational output can also be read and animated. :Joe jtgolab@amoco.com From st-amant@cgl.ucsf.EDU Mon Jul 19 10:23:31 1993 Date: Mon, 19 Jul 93 17:23:31 -0700 Message-Id: <9307200023.AA24908@socrates.ucsf.EDU> From: st-amant@cgl.ucsf.edu (Alain St-Amant) To: chemistry@ccl.net, jas@medinah.atc.ucarb.com Subject: Re: DFT/ab initio calcs on tungsten Jack A. Smith (jas@medinah.atc.ucarb.com) wrote: > The DMOL program (from Biosym) can handle elements beyond Xe if you're > willing to go through some extra gyrations. They've intentionally made it > difficult to go beyond Xe to force users to realize they're on thin ice > regarding relativistic effects. The Biosym response center (at I forgot to mention that for the program deMon, we also did not include relativistic effects. However, the model core potentials (that you must use to do something like tungsten) are obtained from numerical, relativistic calculations on the atoms. So you get the indirect relativistic effects when you perform the molecular non-relativistic calculation. This seems to do a nice job, but my personal experience is limited to gold and platinum complexes. An all-electron non-relativistic calculation on any such system would have yielded much worse answers. alain st-amant department of pharmaceutical chemistry university of california, san francisco