From chemistry-request@server.ccl.net Tue Jun 12 07:20:06 2001 Received: from gatekeeper.itqb.unl.pt (qmailr@[193.136.177.1]) by server.ccl.net (8.11.0/8.11.0) with SMTP id f5CBK2531252 for ; Tue, 12 Jun 2001 07:20:03 -0400 Received: (qmail 15146 invoked from network); 12 Jun 2001 11:19:59 -0000 Received: from mistral.itqb.unl.pt (sgodinho@193.136.176.14) by gatekeeper.itqb.unl.pt with SMTP; 12 Jun 2001 11:19:59 -0000 Received: from localhost (sgodinho@localhost) by mistral.itqb.unl.pt (8.9.3/8.9.3) with ESMTP id MAA04455 for ; Tue, 12 Jun 2001 12:19:57 +0100 Date: Tue, 12 Jun 2001 12:19:57 +0100 (WEST) From: Sofia Godinho To: CCL Subject: Summary:nth-order saddle point, n>1 In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Dear CCLers, some days ago I posted a question regarding what would be a nth-order saddle point, n>1. Here you have the summary of the answers. Thanks are due for their help, to - "in order of arrival": Xavier Assfeld, James Robinson, Laurence Cuffe, David Shobe, Christoph van Wuellen, Andrus Metsala, Huub van Dam and Wai-To Chan. ________________ Original message: ---------------- Dear CCLers, I wonder what is the chemical/physical meaning of a nth-order saddle point (n>1)? That is...for instance, when I get four immaginary frequencies does it mean that I have to "pass through" four vibrational modes to arrive to a minimum? (Or to three to get to a transition state?) Or it simply has nothing to do with it and it is only a mathematical definition of a saddle point? Or...? May I thank you in advance for any help, Sofia _______ Summary: ------- From: Xavier ASSFELD A nth order saddle point correspond to the transition state a n chemical reaction occuring simultaneously. There was a discussion about it some months ago on CCL, and as far as I remember, there was experimental evidences that proved that 2nd order saddle points exist. Nth order saddle point do not belong to the minimum energy path, but are available configurations in highly energetic (kinetic) chemical system. Hope this helps. -- ...Xav ---------------------------------------------------------------------------- From: "Robinson, James" A transition state should have only ONE imaginary frequency. It would be odd if a single TS could lead to different products. How would the TS distinguish between paths to follow. I suspect that attempting to rationalise some mathematical result will be academic. I doubt if it has any physio/chemical meaning and is probably just the result of computation. Simple finding some computational result does not mean anything. Results should be reconciled with real-world problems. James ---------------------------------------------------------------------------- From: Laurence Cuffe For a bent triatomic molecule where the bent form is the most stable then a linear arrangement may appear as an order two saddle point, indicating that there is a two dimensional surface of acceptable trajectories to the minimum rather than a single path. (your on to of a hill and every way is down!) more generally where I have encountered higher order saddle points examination of the molecular motions involved in the imaginary frequencies have revealed that they corespond to rotations of separate groups in the molecule each of which was at a local saddle point. Hope this helps Laurence Cuffe ---------------------------------------------------------------------------- From: "Shobe, Dave" I don't think it means anything other than that there are four "independent" ways of distorting the molecule that lower the energy. (These can be symmetry-equivalent pairs or triples of distortions--"independent" really has to do with the orthogonality of vectors). You can go directly to the minimum from the fourth order saddle point without going through other stationary points. As an example, imagine you are at the top of a hill--a maximum, but also a 2nd-order saddle point. If you go in the right direction you will reach a first-order saddle point, but in most cases following the gradient downward takes you directly to a local minimum. --David Shobe ---------------------------------------------------------------------------- From: "Christoph van [iso-8859-1] Wuellen" >A transition state should have only ONE imaginary frequency. > the reason is that if it is a higher-order TS, there will be another one with lower energy, and molecules are sharp enough to take that one. --------------------------------------------------------------------------- From: Andrus Metsala A transition state may have more than one imaginary frequency. For example so-called monkey saddle has two imaginary frequencies. We may imagine transition states even with more than two imaginary frequencies. Andrus ----------------------------------------------------------------------------- From: Huub van Dam Lets restrict this problem to a 2 dimensional problem. Think of the potential energy surface as hilly country side. Down in the valleys you will find 2 positive frequencies. This means that moving either along the north-south axes or the east-west axes or any linear combination of the two, you will always have to go uphill. Passing from one valley to the next between to hill tops, you will find a point with 1 imaginary frequency. This means that there is one direction along which you can only go down hill from that point. Along another direction you will go uphill towards one of the hill tops. Finally, when you are on a hill top you will find 2 imaginary frequencies. This means that there are 2 linearly independent directions both leading downhill from the point where you are. In your example, where you have 4 imaginary frequencies this simply means that there are 4 linearly independents directions along which you can only go down in energy. That is all. I hope this helps, Huub ---------------------------------------------------------------------- From: Wai-To Chan You can easily pick up a higher-order saddle point structure. Let me give you an illustration. Suppose you are trying to locate the transition state (TS) structure of the hydrogen transfer reaction of a large hydrocarbon molecule: RH + X ----> R---H--X ----> R + HX You obtained a TS structure R---H---X with the 1st imaginary vibration corresponding to the H-transfer motion. But your TS optimization job could fail to minimize the energy of the structure with respect to some of the degrees of freedom other than the reaction coordiates. For instance, a methyl group in R is locked in a higher energy conformation and you obtain a second imaginary frequency corresponding to the torsion of that methyl group. If you obtained a TS structure with more than one imaginary frequencies you should first examine their motions with a program like MOLDEN or GAUSSVIEW. If you are lucky one of the vibrations would be the one corresponding to the motion along the reaction coordinate. You may then try a reoptimization of this higher order saddle point to a 1st order structure. For the RH+X illustration from above you may try partial optimization of the 2nd order structure by relaxing any constraints of the the motion of the problematic methyl group. You then repeat your TS search job using the reoptimized structure as the starting geometry. Wai-To Chan ---------------------------------------------------------------- Sofia Godinho Ph-D student ITQB-UNL-Portugal email: sgodinho@itqb.unl.pt From chemistry-request@server.ccl.net Tue Jun 12 04:37:13 2001 Received: from mel.ruc.dk ([130.225.220.27]) by server.ccl.net (8.11.0/8.11.0) with ESMTP id f5C8bB508933 for ; Tue, 12 Jun 2001 04:37:12 -0400 Received: from smtprelay.ruc.dk (smtprelay.ruc.dk [130.225.220.22]) by mel.ruc.dk (8.9.3/8.9.1) with ESMTP id KAA16947; Tue, 12 Jun 2001 10:37:11 +0200 (MET DST) Received: from virgil.ruc.dk (virgil.ruc.dk [130.225.220.110]) by smtprelay.ruc.dk (8.9.3+Sun/8.9.3) with ESMTP id KAA25332; Tue, 12 Jun 2001 10:17:28 +0200 (MEST) Received: from VIRGIL/SpoolDir by virgil.ruc.dk (Mercury 1.44); 12 Jun 01 10:37:11 +0100 Received: from SpoolDir by VIRGIL (Mercury 1.44); 12 Jun 01 10:37:05 +0100 From: "Jens Spanget-Larsen" Organization: Roskilde Universitetscenter To: "Dr. Richard L. Wood" , "chemistry@ccl.net" Date: Tue, 12 Jun 2001 10:36:46 +0100 Subject: Re: CCL:phase information/orbitals question Priority: normal X-mailer: Pegasus Mail for Windows (v2.23) Message-ID: <12CF3F8C3ABA@virgil.ruc.dk> Dr. Richard L. Wood: > I think he means that the idea of an orbital is NOT a mathematical > concept, but a "physical", i.e. from a physics point of view, > concept. Well, we apparently disagree. To me, the idea of "orbitals" in the case of a many electron system is basically a mathematical concept, since these "orbitals" have no physical reality. But they do of course have physical significance, in so far as they are excellent approximate models in the rationalization of many kinds of physical and chemical behaviour. Yours, Jens >--< =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= JENS SPANGET-LARSEN Phone: +45 4674 2000 (RUC) Department of Chemistry +45 4674 2710 (direct) Roskilde University (RUC) Fax: +45 4674 3011 P.O.Box 260 E-Mail: JSL@virgil.ruc.dk DK-4000 Roskilde, Denmark http://virgil.ruc.dk/~jsl/ =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= From chemistry-request@server.ccl.net Mon Jun 11 18:37:39 2001 Received: from web13404.mail.yahoo.com ([216.136.175.62]) by server.ccl.net (8.11.0/8.11.0) with SMTP id f5BMbd505171 for ; Mon, 11 Jun 2001 18:37:39 -0400 Message-ID: <20010611223739.41106.qmail@web13404.mail.yahoo.com> Received: from [169.232.140.246] by web13404.mail.yahoo.com; Mon, 11 Jun 2001 15:37:39 PDT Date: Mon, 11 Jun 2001 15:37:39 -0700 (PDT) From: eric hu Subject: protein folding and unfolding To: chemistry@ccl.net MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Hi, I'd like to have some input about the difference of protein folding and unfolding pathway. For the folding there are many pathways. In some books, they say the reversible unfolding process has less number of pathways... Thanks! Eric __________________________________________________ Do You Yahoo!? Get personalized email addresses from Yahoo! Mail - only $35 a year! http://personal.mail.yahoo.com/ From chemistry-request@server.ccl.net Tue Jun 12 16:43:43 2001 Received: from mail-d.bcc.ac.uk ([144.82.100.24]) by server.ccl.net (8.11.0/8.11.0) with ESMTP id f5CKhh511394; Tue, 12 Jun 2001 16:43:43 -0400 Received: from dns (actually host oganov-jpb98.geol.ucl.ac.uk) by mail-d.bcc.ac.uk with SMTP (Mailer); Tue, 12 Jun 2001 21:43:49 +0100 Reply-To: "a.oganov" From: "Artem R. Oganov" To: chemistry Cc: chemistry-request Subject: CCL: RE: Superconducting phase transitions Date: Tue, 12 Jun 2001 21:41:37 +0100 Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2911.0) Importance: Normal In-Reply-To: X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4133.2400 Dear all, Many thanks to Dirk Sayes and Adam Hixson for very useful replies. I have found a very good discussion of the superconducting phase transitions in the book by Rose-Innes, suggested by Dirk Sayes. Yours, Artem P.S. The replies are below: ------------------------------------------------- I'd do a search on the Ginzberg-Landau theory of phase transitions. Adam Hixson (chixson@chemdept.chem.ou.edu; hixsonc@yahoo.com) ------------------------------------------------- i got this mail from my brother for the theory of superconductivity i can recommend the book Fundamentels of The theory of Metals by A.A. Abrikosov. (p348) for the experimental part Rose Innes : introduction to superconductivity. Dirk Saeys TFVS UIA Antwerpen ------------------------------------------------- ----------------------------------------------------- Artem R. Oganov Department of Geological Sciences University College London Gower Street London WC1E 6BT tel: +44 (020)-7679-3344 fax: +44 (020)-7387-1612 email: a.oganov@ucl.ac.uk http://slamdunk.geol.ucl.ac.uk/~artem --------------------------------------------------- -----Original Message----- From: Computational Chemistry List [mailto:chemistry-request@ccl.net]On Behalf Of Artem R. Oganov Sent: 08 June 2001 09:55 PM To: chemistry Cc: chemistry-request Subject: CCL:Superconducting phase transitions Dear all, I would appreciate if you could give me any references to papers or books which would explain why the superconducting phase transitions are always second order. I'd also like to see any experimental papers confirming this conclusion (e.g., by showing the variation of the X-ray diffraction peak intensities). I'll summarise on request. Thanks a lot. Cheers, Artem ----------------------------------------------------- Artem R. Oganov Department of Geological Sciences University College London Gower Street London WC1E 6BT tel: +44 (020)-7679-3344 fax: +44 (020)-7387-1612 email: a.oganov@ucl.ac.uk http://slamdunk.geol.ucl.ac.uk/~artem --------------------------------------------------- -= This is automatically added to each message by mailing script =- CHEMISTRY@ccl.net -- To Everybody | CHEMISTRY-REQUEST@ccl.net -- To Admins MAILSERV@ccl.net -- HELP CHEMISTRY or HELP SEARCH CHEMISTRY-SEARCH@ccl.net -- archive search | Gopher: gopher.ccl.net 70 Ftp: ftp.ccl.net | WWW: http://www.ccl.net/chemistry/ | Jan: jkl@ccl.net From chemistry-request@server.ccl.net Tue Jun 12 11:06:48 2001 Received: from gate.syntem-sa.fr ([146.19.248.65]) by server.ccl.net (8.11.0/8.11.0) with ESMTP id f5CF6l504423 for ; Tue, 12 Jun 2001 11:06:47 -0400 Received: (from uucp@localhost) by gate.syntem-sa.fr (980427.SGI.8.8.8/980728.SGI.AUTOCF) id RAA54593; Tue, 12 Jun 2001 17:05:44 +0200 (MDT) Received: from syntem-sa.fr(10.0.0.2) by gate.syntem-sa.fr via smap (4.1) id xma570565; Tue, 12 Jun 01 17:05:19 +0200 Received: from SUCROSE ([10.0.0.127]) by laetitia.syntem-sa.fr (980427.SGI.8.8.8/980728.SGI.AUTOCF) via SMTP id RAA63259; Tue, 12 Jun 2001 17:02:09 +0200 (MDT) Reply-To: From: "David Turner" To: Cc: "'Kim Branson'" Subject: Summary: Docking of Covalently Bound ligands Date: Tue, 12 Jun 2001 17:04:18 +0200 Message-ID: <0196B8B2A850EF4F8D669EE51C21C5AA10DC54@digitaline.nimes.syntem.com> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_0000_01C0F361.B88009B0" X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook CWS, Build 9.0.2416 (9.0.2911.0) Importance: Normal In-Reply-To: X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4133.2400 C'est un message de format MIME en plusieurs parties. ------=_NextPart_000_0000_01C0F361.B88009B0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Dear all Original message: Can anybody point me to some overviews/discussions of the docking of ligands that are (or are potentially) covalently bound to the receptor site. I am looking at this mainly but not solely from the view-point of high throughput virtual screening. ############################################################################ ## From: Suzie Byun [byun@ludwig.chem.wesleyan.edu] Hi David, I saw your posting on ccl regarding docking of covalent ligands. I did a project on this very topic. I had a crystal structure of an enzyme-ligand complex and I had to dock ligands back into the active site. Of course the vdw contacts makes this prohibitive to dock back in non-covalent ligands. There are two programs I am aware of that can dock covalent ligands. The first I have only read about - GOLD which in the literature claims to be able to do the job. THe second program, I have had direct experience with and promising results. THe program is called Dockit, from Metaphorics Inc. in Northern California (I believe). I used Dockit to re-dock the original ligand back into the crystal structure and within the top 5 best scorers, the rmsd of the fits were around 1.5-2 Angstroms. The program works by specifying 1-2, 1-3, and 1-4 close contact distances. Anyway, there is a website you can check out if you are interested. Please let me know if you are. Good luck, Suzie ############################################################################ ## From: David Clark [David.Clark@argentadiscovery.com] Hi Dave Saw your question on CCL - the Agouron approach to this was described in: Gehlhaar, D.K., Bouzida, D. and Rejto, P.A. Reduced Dimensionality in Ligand-Protein Structure Prediction: Covalent Inhibitors of Serine Proteases and Design of Site-Directed Combinatorial Libraries. In Rational Drug Design: Novel Methodology and Practical Applications, ACS Symposium Series Vol. 719, Parrill, A.L., Reddy, M.R., Eds.; American Chemical Society: Washington DC; 1999, pp. 292-311. ############################################################################ ### From: Joe M Leonard [jle@world.std.com] - Tripos David, I have always felt that covalently-bound ligands are less a docking than a conformational analysis problem - and the tool set for the latter is more extensive. Are you trying to vHTS a set of compounds with a known positioning in the cleft? If so, QM/MM calculations seem way too slow to actually handle the electron/proton motion... We get a lot of people purchasing FlexX to vHTS arbitrary collections of molecules, but they usually don't know what the active site configuration would be. There are a few who want to specifically position individual ligands, but that's a bit more work and usually has to be done molecule by molecule (making any automation difficult). The DOCK literature might cover this, as there have been a number of articles over the last few years from Kuntz and folks. Joe ############################################################################ ### From: Dave Young dyoung@cytoclonal.com > > David, > > Most docking programs are designed for non-covalently bound ligands, > which have been the focus of structure based drug design. I am aware of > two commercial programs that can have the ligand attached to the active > site. One is GOLD (http://www.ccdc.cam.ac.uk/prods/gold/index.html) and > the other is DockVision (www.dockvision.com). Both of these allow the > user to define an attachment between the ligand and the active site. > However, it is a geometric constraint only...neither attempts to include > the covalent bond energy in the binding energy. This can be a problem > since you can find a large range of covalent binding energies depending > upon whether it is an irreversible bond formation, or a reversible > partially-covalent adduct formation. > > The only program I'm aware of that does a covalently bound docking > including the covalent bond energy is a QM/MM docking algorithm that I > wrote here at Cytoclonal Pharmaceutics. This is one piece of Quantum > Core Technology (QCT), which is a mechanism based drug design system > proprietary to Cytoclonal. This software is not available commercially, > but Cytoclonal is interested in business partnerships with other > corporations to utilize this technology. > > Let me know what solution you come up with for this problem. > ############################################################################ ### ------=_NextPart_000_0000_01C0F361.B88009B0 Content-Type: text/html; charset="us-ascii" Content-Transfer-Encoding: quoted-printable Summary: Docking of Covalently Bound ligands

Dear all

Original message:

Can anybody point me to some overviews/discussions of = the
docking of ligands that are (or are potentially) = covalently
bound to the receptor site. I am looking at this = mainly
but not solely from the view-point of high throughput = virtual
screening.

################################################################= ##############
From: Suzie Byun = [byun@ludwig.chem.wesleyan.edu]
Hi David,
I saw your posting on ccl regarding docking of = covalent ligands.  I
did a project on this very topic.  I had a = crystal structure of an
enzyme-ligand complex and I had to dock ligands back = into the active site.
Of course the vdw contacts makes this prohibitive to = dock back in
non-covalent ligands.  There are two programs I = am aware of that can
dock covalent ligands.  The first I have only = read about - GOLD which in
the literature claims to be able to do the job.  = THe second program, I
have had direct experience with and promising = results.  THe program is
called Dockit, from Metaphorics Inc. in Northern = California
(I believe).  I used Dockit to re-dock the = original ligand
back into the crystal structure and within the top 5 = best scorers, the
rmsd of the fits were around 1.5-2 Angstroms.  = The program works by
specifying 1-2, 1-3, and 1-4 close contact = distances. 

Anyway, there is a website you can check out if you = are interested. Please
let me know if you are.
Good luck, Suzie

################################################################= ##############
From: David Clark = [David.Clark@argentadiscovery.com]

Hi Dave
Saw your question on CCL - the Agouron approach to = this was described in:
Gehlhaar, D.K., Bouzida, D. and Rejto, P.A. Reduced = Dimensionality in
Ligand-Protein Structure Prediction: Covalent = Inhibitors of Serine Proteases
and Design of Site-Directed Combinatorial Libraries. = In Rational Drug
Design: Novel Methodology and Practical Applications, = ACS Symposium Series
Vol. 719, Parrill, A.L., Reddy, M.R., Eds.; American = Chemical Society:
Washington DC; 1999, pp. 292-311.

################################################################= ###############
From: Joe M Leonard [jle@world.std.com] - = Tripos
David, I have always felt that covalently-bound = ligands are
less a docking than a conformational analysis problem = - and
the tool set for the latter is more extensive.  = Are you trying
to vHTS a set of compounds with a known positioning = in the
cleft?  If so, QM/MM calculations seem way too = slow to actually
handle the electron/proton motion...

We get a lot of people purchasing FlexX to vHTS = arbitrary collections
of molecules, but they usually don't know what the = active site
configuration would be.  There are a few who = want to specifically
position individual ligands, but that's a bit more = work and usually
has to be done molecule by molecule (making any = automation difficult).
The DOCK literature might cover this, as there have = been a number
of articles over the last few years from Kuntz and = folks.

Joe

################################################################= ###############

From: Dave Young dyoung@cytoclonal.com
>
> David,
>
> Most docking programs are designed for = non-covalently bound ligands,
> which have been the focus of structure based = drug design.  I am aware of
> two commercial programs that can have the ligand = attached to the active
> site.  One is GOLD (http://www.ccdc.= cam.ac.uk/prods/gold/index.html) and
> the other is DockVision = (www.dockvision.com).  Both of these allow the
> user to define an attachment between the ligand = and the active site.
> However, it is a geometric constraint = only...neither attempts to include
> the covalent bond energy in the binding = energy.  This can be a problem
> since you can find a large range of covalent = binding energies depending
> upon whether it is an irreversible bond = formation, or a reversible
> partially-covalent adduct formation.
>
> The only program I'm aware of that does a = covalently bound docking
> including the covalent bond energy is a QM/MM = docking algorithm that I
> wrote here at Cytoclonal Pharmaceutics.  = This is one piece of Quantum
> Core Technology (QCT), which is a mechanism = based drug design system
> proprietary to Cytoclonal.  This software = is not available commercially,
> but Cytoclonal is interested in business = partnerships with other
> corporations to utilize this technology.
>
> Let me know what solution you come up with for = this problem.
>

################################################################= ###############

------=_NextPart_000_0000_01C0F361.B88009B0-- From chemistry-request@server.ccl.net Tue Jun 12 11:57:50 2001 Received: from hkusua.hku.hk ([147.8.2.91]) by server.ccl.net (8.11.0/8.11.0) with ESMTP id f5CFvm505600 for ; Tue, 12 Jun 2001 11:57:49 -0400 Received: from win95pc ([147.8.148.66]) by hkusua.hku.hk (8.11.4/8.11.4) with SMTP id f5CFvZh29465 for ; Tue, 12 Jun 2001 23:57:36 +0800 (HKT) Message-ID: <002101c0f358$8df09600$42940893@win95pc.hku.hk> From: "dwang" To: Subject: summerize: IRC question Date: Tue, 12 Jun 2001 23:58:41 +0800 MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_001E_01C0F39B.9B106FC0" X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 4.72.3110.5 X-MimeOLE: Produced By Microsoft MimeOLE V4.72.3110.3 This is a multi-part message in MIME format. ------=_NextPart_000_001E_01C0F39B.9B106FC0 Content-Type: text/plain; charset="gb2312" Content-Transfer-Encoding: quoted-printable Dear cclers, The following is the replies to my question which was posted a few = weeks ago. Thanks a lot to Dr. Matthias Hofmann and Dr.Douglas J. Fox. >>>>>>>>>>>original question<<<<<<<<<<<<<<<<=20 >=20 > Dear Sir/Madam > I am performing a IRC program, A->TS->D. By using opt=3D3Dqst2 , I = found =3D > the structure of TS. But the irc(result B ->C) didn't result in what I = =3D > want. So I think the path may be A ->TS1 ->B ->TS0 -> C -> TS2 ->D. I = =3D > repeat the calculation by using C and D as input file to find TS2, = then =3D > IRC. While the results indicate that this IRC included the IRC I have = =3D > performed first. Why? Any advice will be highly appreciated. >>>>>>>>>>>>>>>>>>>>>>>>><<<<<<<<<<<<<<<<<<<<<<<<<<<<< Hello,=20 Regarding your IRC problems:=20 You were not very specific how you did your calculations.=20 I recommend, to use internal coordinates (Z-Matrix) for the=20 IRC runs and to separate into a forward and a reverse job.=20 Sometimes the IRC stops very early and reports it has reached=20 a minimum while this is not true. (Just run a opt for the structure to confirm it as a minimum)=20 When trying to restart an IRC the output becomes completely mixed up unless you restrict the run to one direction.=20 Good luck,=20 Matthias Hofmann -------------------------------------------------------------------------= --- Dr. Matthias Hofmann Anorganisch-Chemisches Institut =20 Ruprecht-Karls Universitaet Heidelberg Tel.: (INT+) 6221 = 54-8451 In Neuenheimer Feld 270 Fax.: (INT+) 6221 = 54-8439 D-69120 Heidelberg e-mail: = matthias.hofmann@urz.uni-heidelberg.de Germany = http://www.uni-heidelberg.de/institute/fak12/AC/hofmann =20 -------------------------------------------------------------------------= --- Dr. Wang, Let me restate this my way just to make sure I understand. For this simple system you are mapping out the full potential energy surface including isomers for Cl attacking at C and S, the transition state for isomerization and possibly some other decomposition products. = At this point you have found a few local minima and a few transition state candidates. Is this accurate? Assuming it is I have a couple of questions. First, are you = monitoring the spin state here? For the Cs structures there is at least a doublet A' and A" state. Probably only two relevant states for the C2v = structure. If you are not staying on one of these surfaces then I could see some of the complexity here relating to having two surfaces. If all of these structures are not planar then you would have only one doublet state but it will connect to only one of the Cs/C2v surfaces. So watch that you = are not mixing states. Second, DFT is notorious for underestimating barrier heights. So once you have identified the principle paths you should look at how MP2 or some traditional correlation method affects this. Since you have IRC paths you might want to use IRCMax to cut down on the effort. It will let you trace MP2 energies along this path and get close to the MP2 TS structures. Then you can reoptimize the ones which show promise. Third, are you checking the FREQ results at each of the TS structures to verify that only one imaginary mode is present? Also that the=20 barrier heights are relevant wrt the the zero point energy if these are rather shallow maxima/minima. If you are having problems getting the optimization tight enough with DFT consider improving the = integration grid, INT=3DGRID=3DUltraFine for example, to make sure that the = optimizer is not being confused by numerical error. Finally, have you compared your surface with the CO2 analog? = Certainly the chemistry will be different but I would expect it to have a lot of similar features.=20 Douglas J. Fox Technical Support Gaussian, Inc. help@gaussian.com >>>>>>>>>>>>>>>>>>>>My letter to Dr. Fox<<<<<<<<<<<<<<<<<<<<<<<<<<=20 > Dear Dr. Fox, > Thank you for your help. > Now I am performing the reformation reaction between the carbon = isomer(A) > and sulfur isomer(B) of CS2Cl radical(B ->TS ->A)(method: DFT, basis = set: > 6-311G**)(the radicals come from CS2+chlorine atom radical). I have = supposed > the sulfur isomer(symmetry: Cs) is more unstable than the carbon > isomer(Symmetry: C2v), so A will be the reactant. I have found the = transient > state by using qst2 and two intermediates(C and D)by performing IRC. = Then I > finished the optimization of C and D(result in C1 and D1 = respectively), and > found that the structure of C1 is almost the same with A and D1 is the = same > with B(So the path may be A ->TS ->B?). > I think maybe D can transform into D1 through path D ->TS2 ->D1, so = I > performed TS2-searching by using qst2( between D and D1), then IRC(end = points: E and F). The result indicated that this IRC included almost the > whole path , and E is almost the same with C1. To F and D1, the bond > lengthes are almost the same , while there is a rather large energe = gap > (about 0.010 hatree)between F and D1, and the bond angle of (C-S-Cl) = of F is much less than D1( about 71.2 and 105.2 respectively). > Now I have found another transient state between F and D1 still by = qst2 > and the IRC is in progress. This time I added the keyword = Opt=3D(verytight, > reverse) to the path. I don't know if this is reasonable because I = have > little experience in IRC calculation. >=20 > This is a system only including four atoms and I have some bigger = systems > waiting for calculation. So I wish I can learn much from the IRC of = CS2Cl in > order to perform the coming work well. Could you give me some advice? > Thank you very much! >>>>>>>>>>>>>>>>>>>><<<<<<<<<<<<<<<<<<<<<<<<<<<<< Dr. Wang, The use of QST2 improves the chances that you will zero in on the transition state of interest however, there is always the possibility that either your expectations are wrong or the model chemistry you are using does not match reality well. So what I read here you have found a transition state which does not connect to either the product or reactants you expected. While your three-part reaction mechanism is possible it seems a bit involved if you actually see the results of this reaction. If this four part reaction sequence should also result in measurable amounts of these = other=20 products. Are they seen or at least plausible? I don't quite understand your last comment except it reminds me that perhaps you are confused that the IRC attempts to produce the actual optimized reactant/product structures as the end points. Rather it does a fixed number of steps and from that you can generally conclude that it is in the right product/reactant valley. If you then looked for the path from C to D it would coincide with the earlier path. Perhaps if you describe this more completely I could make a better=20 determination of what is up. What methods are you using? What are the structures A,B,C,D? Douglas J. Fox Director of Technical Support help@gaussian.com ------=_NextPart_000_001E_01C0F39B.9B106FC0 Content-Type: text/html; charset="gb2312" Content-Transfer-Encoding: quoted-printable
Dear = cclers,
 
  = The following is the replies to=20 my question which was posted a few weeks ago. Thanks a lot to Dr. = Matthias=20 Hofmann and Dr.Douglas J. Fox.
>>>>>>>>>>>original=20 question<<<<<<<<<<<<<<<<&= nbsp;
>=20
> Dear Sir/Madam
>   I am performing a IRC = program,=20 A->TS->D. By using opt=3D3Dqst2 , I found =3D
> the = structure of TS. But=20 the irc(result B ->C) didn't result in what I =3D
> want. So I = think the=20 path may be A ->TS1 ->B ->TS0 -> C -> TS2 ->D. I = =3D
>=20 repeat the calculation by using C and D as input file to find TS2, then=20 =3D
> IRC. While the results indicate that this IRC included the = IRC I have=20 =3D
> performed first. Why? Any advice will be highly=20 appreciated.
>>>>>>>>>>>>>>&= gt;>>>>>>>>>><<<<<<<&l= t;<<<<<<<<<<<<<<<<<<= ;<<<
Hello, =
Regarding your IRC problems:=20
You were not very specific how you did your calculations.
I = recommend,=20 to use internal coordinates (Z-Matrix) for the
IRC runs and to = separate into=20 a forward and a reverse job.
Sometimes the IRC stops very early and = reports=20 it has reached
a minimum while  this is not true. (Just run a = opt for=20 the structure
to confirm it as a minimum)
When trying to restart = an IRC=20 the output becomes completely mixed
up unless you restrict the run to = one=20 direction.

Good luck,
Matthias Hofmann
----------------------------------------------------------------= ------------
 =20 Dr. Matthias Hofmann
  Anorganisch-Chemisches=20 Institut           = ;   =20
  Ruprecht-Karls Universitaet=20 Heidelberg           = Tel.:=20 (INT+) 6221 54-8451
  In Neuenheimer Feld=20 270           &nbs= p;            = ; =20 Fax.: (INT+) 6221 54-8439
  D-69120=20 Heidelberg          e-mail: = matthias.hofmann@u= rz.uni-heidelberg.de
 =20 Germany           = http://w= ww.uni-heidelberg.de/institute/fak12/AC/hofmann =20
---------------------------------------------------------------------= -------
Dr. = Wang,
  Let me restate this=20 my way just to make sure I understand.

  For this simple = system you=20 are mapping out the full potential energy
surface including isomers = for Cl=20 attacking at C and S, the transition
state for isomerization and = possibly=20 some other decomposition products.  At
this point you have found = a few=20 local minima and a few transition
state candidates.  Is this=20 accurate?

   Assuming it is I have a couple of = questions. =20 First, are you monitoring
the spin state here?  For the Cs = structures=20 there is at least a doublet
A' and A" state.  Probably only two = relevant=20 states for the C2v structure.
If you are not staying on one of these = surfaces=20 then I could see some
of the complexity here relating to having two=20 surfaces.  If all of these
structures are not planar then you = would have=20 only one doublet state but
it will connect to only one of the Cs/C2v=20 surfaces. So watch that you are
not mixing = states.

  =20 Second, DFT is notorious for underestimating barrier heights.  = So
once=20 you have identified the principle paths you should look at how
MP2 or = some=20 traditional correlation method affects this.  Since you have
IRC = paths=20 you might want to use IRCMax to cut down on the effort.
It will let = you trace=20 MP2 energies along this path and get close to
the MP2 TS = structures. =20 Then you can reoptimize the ones which = show
promise.

  =20 Third, are you checking the FREQ results at each of the TS = structures
to=20 verify that only one imaginary mode is present?  Also that the =
barrier=20 heights are relevant wrt the the zero point energy if these
are = rather=20 shallow maxima/minima.  If you are having problems getting
the=20 optimization tight enough with DFT consider improving the = integration
grid,=20 INT=3DGRID=3DUltraFine for example, to make sure that the = optimizer
is not being=20 confused by numerical error.

   Finally, have you = compared your=20 surface with the CO2 analog?  Certainly
the chemistry will be = different=20 but I would expect it to have a lot of
similar = features. 

 =20 Douglas J. Fox
  Technical Support
  Gaussian, = Inc.
 =20 help@gaussian.com

 >>>>>>>>>>>>>>>>= >>>>My=20 letter to Dr.=20 Fox<<<<<<<<<<<<<<<<<&l= t;<<<<<<<< 
> = Dear Dr. Fox,
>  =20 Thank you for your help.
>   Now  I am performing = the=20 reformation reaction between the carbon isomer(A)
> and sulfur = isomer(B)=20 of CS2Cl radical(B ->TS ->A)(method: DFT, basis set:
> = 6-311G**)(the=20 radicals come from CS2+chlorine atom radical). I have supposed
> = the=20 sulfur isomer(symmetry: Cs) is more unstable than the carbon
>=20 isomer(Symmetry: C2v), so A will be the reactant. I have found the=20 transient
> state by using qst2 and two intermediates(C and D)by=20 performing IRC. Then I
> finished the optimization of C and = D(result in C1=20 and D1 respectively), and
> found that the structure of C1 is = almost the=20 same with A and D1 is the same
> with B(So the path may be A = ->TS=20 ->B?).
>   I think maybe D can transform into D1 = through path=20 D ->TS2 ->D1, so I
> performed TS2-searching by using qst2( = between=20 D and D1), then IRC(end points: E and F). The result indicated that this = IRC=20 included almost the
> whole path , and E is almost the same with = C1. To F=20 and D1, the bond
> lengthes are almost the same , while there is a = rather=20 large energe gap
> (about 0.010 hatree)between F and D1, and the = bond=20 angle of (C-S-Cl) of F is much less than D1( about 71.2 and 105.2=20 respectively).
>   Now I have found another transient = state=20 between F and D1 still by qst2
> and the IRC is in progress. This = time I=20 added the keyword Opt=3D(verytight,
> reverse) to the path. I = don't know if=20 this is reasonable because I have
> little experience in IRC=20 calculation.
>
>   This is a system only = including four=20 atoms and I have some bigger systems
> waiting for calculation. So = I wish=20 I can learn much from the IRC of CS2Cl in
> order to perform the = coming=20 work well. Could you give me some advice?
>   Thank you = very=20 much!
>>>>>>>>>>>>>>>>= >>>><<<<<<<<<<<<<<&= lt;<<<<<<<<<<<<<<=
 
Dr. = Wang,

   The use=20 of QST2 improves the chances that you will zero in on the
transition = state of=20 interest however, there is always the possibility
that either your=20 expectations are wrong or the model chemistry you are
using does not = match=20 reality well.

   So what I read here you have found a=20 transition state which does
not connect to either the product or = reactants=20 you expected.  While
your three-part reaction mechanism is = possible it=20 seems a bit involved
if you actually see the results of this = reaction. =20 If this four part
reaction sequence should also result in measurable = amounts=20 of these other
products.  Are they seen or at least=20 plausible?

   I don't quite understand your last = comment except=20 it reminds me that
perhaps you are confused that the IRC attempts to = produce=20 the actual
optimized reactant/product structures as the end = points. =20 Rather it does
a fixed number of steps and from that you can = generally=20 conclude that
it is in the right product/reactant valley.  If = you then=20 looked for the
path from C to D it would coincide with the earlier=20 path.

   Perhaps if you describe this more completely I = could=20 make a better
determination of what is up.  What methods are = you=20 using?  What are the
structures A,B,C,D?

  Douglas = J.=20 Fox
  Director of Technical Support
  help@gaussian.com
------=_NextPart_000_001E_01C0F39B.9B106FC0-- From chemistry-request@server.ccl.net Tue Jun 12 11:36:20 2001 Received: from spot-us.spotfire.com ([63.118.157.70]) by server.ccl.net (8.11.0/8.11.0) with ESMTP id f5CFaJ505116 for ; Tue, 12 Jun 2001 11:36:20 -0400 Received: from spotfire.com (spot-us.spotfire.com [172.16.1.10]) by spot-us.spotfire.com (Postfix) with ESMTP id 501577F4E2 for ; Tue, 12 Jun 2001 11:36:15 -0400 (EDT) Message-ID: <2140620016212153657410@spotfire.com> X-EM-Version: 5, 0, 0, 21 X-EM-Registration: #01B0530810E603002D00 X-Priority: 3 Reply-To: webcast@spotfire.com X-MSMail-Priority: Normal From: "Spotfire Webcast" To: chemistry@server.ccl.net Subject: Commercial: Spotfire Lead Discovery Webcast- Getting the Most from SAR Analysis Date: Tue, 12 Jun 2001 11:36:57 -0400 MIME-Version: 1.0 Content-type: text/plain; charset="US-ASCII" Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by server.ccl.net id f5CFaK505120 Spotfire Webcast Announcement: As a chemistry or life science researcher, you might be interestred in attending the next Spotfire Lead Discovery Webcast: Getting the Most from Structure Activity Relationships- Cluster Analysis, ISIS Structure Key-Based Clustering, Searching Current Drugs IDdb3 Database, and Developing and Exporting SAR Tables Join this free Webcast on Thursday, June 14th at 11am EDT and learn some of the latest tools and techniques used by leading pharmaceutical researchers to help speed drug discovery. This presentation of Spotfire DecisionSite for Lead Discovery will demonstrate analysis of Structure Activity Relationships (SAR), including: Cluster analysis, including ISIS key-based clustering; Incorporating results of clustering into the analysis; Developing and exporting a SAR table; Using the Investigational Drugs database (IDdb3) to retrieve additional information to supplement the analysis Date: Thursday, June 14, 2001 Time: 11:00 am EDT Presenter: Matt Anstett, Spotfire Application Specialist To register for the webcast, please visit http://www.spotfire.com/webcast. For more information, send an e-mail message to webcast@spotfire.com or contact us by phone: (617)702-1600 The webcast will be hoted by Centra, a leading provider of online meeting technology. Visit https://www.centranow.com/support/faq.asp#g4 to see complete system requirements, or click on https://www.centranow.com/systemcheck/launchsyscheck.asp to run a computer system compatability check. ----- If you would like to be excluded from future webcast or other Spotfire announcements, just reply to this message with the word "unsubscribe" in the subject line. From chemistry-request@server.ccl.net Tue Jun 12 19:11:37 2001 Received: from eefus.Colorado.EDU ([128.138.134.21]) by server.ccl.net (8.11.0/8.11.0) with ESMTP id f5CNBb514129 for ; Tue, 12 Jun 2001 19:11:37 -0400 Received: from localhost (horinek@localhost) by eefus.Colorado.EDU (8.9.3/8.9.3/UnixOps-Hesiod (PDH,JWD)) with ESMTP id RAA14550 for ; Tue, 12 Jun 2001 17:09:49 -0600 Date: Tue, 12 Jun 2001 17:09:49 -0600 (MDT) From: Dominik Horinek To: chemistry@ccl.net Subject: 3d glasses on pc Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Hi! Is anyone aware of molecular building programs on PCs that support use of 3D glasses. What graphics hardware do I need to run these programs, if there are any available? Thanks for any info, Dominik -- Dominik Horinek Dept. of Chemistry and Biochemistry University of Colorado at Boulder 80309-0215 Boulder From chemistry-request@server.ccl.net Tue Jun 12 22:04:55 2001 Received: from carbon.chem.ucla.edu ([128.97.35.55]) by server.ccl.net (8.11.0/8.11.0) with ESMTP id f5D24t517198 for ; Tue, 12 Jun 2001 22:04:55 -0400 Received: from [128.97.45.51] ([128.97.45.51]) by carbon.chem.ucla.edu (8.9.1a/8.9.1) with ESMTP id TAA15050 for ; Tue, 12 Jun 2001 19:04:55 -0700 (PDT) Mime-Version: 1.0 Message-Id: Date: Tue, 12 Jun 2001 19:04:52 -0700 To: chemistry@ccl.net From: Eric Scerri Subject: small correction to earlier posting Content-Type: text/plain; charset="us-ascii" ; format="flowed" Contrary to what I posted earlier. It occurs to me that even if phases are observed in a one-electron system this would still not be evidence of individual orbitals but perhaps only for the overall wavefunction of the system. eric scerri -- Dr. Eric Scerri , UCLA, Department of Chemistry & Biochemistry, 607 Charles E. Young Drive East, Los Angeles, CA 90095-1569 USA E-mail : scerri@chem.ucla.edu tel: 310 206 7443 fax: 310 206 2061 Web Page: http://www.chem.ucla.edu/dept/Faculty/scerri/index.html Editor of Foundations of Chemistry http://www.wkap.nl/journalhome.htm/1386-4238 Also see International Society for the Philosophy of Chemistry http://www.georgetown.edu/earleyj/ISPC.html