From owner-chemistry@ccl.net Thu Aug 30 03:26:00 2007 From: "Dr. Alexander Hofmann ah#,#chemie.hu-berlin.de" To: CCL Subject: CCL: Heterogeneous catalysis in Gaussian Message-Id: <-35055-070830015136-478-iuhTQn2FpfQCL0g+OMqxjw:server.ccl.net> X-Original-From: "Dr. Alexander Hofmann" Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=ISO-8859-1; format=flowed Date: Thu, 30 Aug 2007 06:51:36 +0200 MIME-Version: 1.0 Sent to CCL by: "Dr. Alexander Hofmann" [ah:-:chemie.hu-berlin.de] Hello Kass, welcome to the 3D world! Kaci Tizi_Ouzou kaci.tiziouzou~!~gmail.com wrote: > Greetings all, > > I am currently working on a problem in which I intend to model a > chemical reaction (oxydation of CO) catalysed by TiO2. At this point I > wonder what is the usual procedure to model the catalyst. This is what I > think to do, but I am willing to explore other routes: > Just go for some publications in journals like Phys. Rev. B or Surface Science. You'll get a good overview. > [1] Optimize CO alone > > [2] Optimize TiO2 alone. In this case, I will consider a single cell of > TiO2. The problem will of course be how to include the crystal effect. I > hear that VASP or ABINIT use plane wave basis sets which takes care of > the crystal effect. ANY COMMENTS? Here is a compilation: http://www.psi-k.org/codes.shtml (CP2K is missing) There are many different flavours around, like plane wave or LCAO, pseudopotentials or not. I'd suggest to start with plane waves and pseudo potentials in your case. I personally prefer VASP (licence fee). If you want to go away with free programs, CPMD, CP2K or DACAPO (CAMPOS) are surely good choices. > > [3] Approach a CO molecule and use a scan to determine the TS. In this > case, I use the cluster (super cluster) methodology. "CO + TiO2" is my > cluster. > > [4] Question: Once I have the TS, can I say that E(TS) = E(CO) + E(TiO2) > + E(Ads). Here I am quite lost. > Yep. Just the way you calculate interaction energies. Have fun Alexander -- Dr. Alexander Hofmann Humboldt-Universitaet zu Berlin Institut fuer Chemie Arbeitsgruppe Quantenchemie Post: Unter den Linden 6 10099 Berlin Visitors: Brook-Taylor-Strasse 2 12489 Berlin ah chemie.hu-berlin.de ]_[ Tel.: +49-30-2093-7138 Fax.: +49-30-2093-7136 http://www.chemie.hu-berlin.de/ag_sauer/index.html PGP-Key: wwwkeys.de.pgp.net ID: D9D62D35 From owner-chemistry@ccl.net Thu Aug 30 04:28:01 2007 From: "Ol Ga eurisco1:+:pochta.ru" To: CCL Subject: CCL:G: To: Thermochemical calculations from Gaussian output file Message-Id: <-35056-070830042124-10319-c16cNucpV4ro9HKO4ohh7A:server.ccl.net> X-Original-From: "Ol Ga" Date: Thu, 30 Aug 2007 04:21:20 -0400 Sent to CCL by: "Ol Ga" [eurisco1 _ pochta.ru] Moltran - a program for inspection of results after quantum chemical calculations performed with different quantum chemical packages including Gaussian and GAMESS. It also allows editing of a molecular structure working as a molecular builder. During the molecular editing, you can save or retrieve the molecular fragments to a database and use them later to create more complicated structures. A special feature of MOLTRAN is a possibility to perform various thermodynamical calculations (with a broad variation of model parameters) and taking into account the internal rotations if they are present in the molecule. Have a look at the Moltran sample screenshots (sample data files creating the screenshots can also be found here). Moltran Manual http://ichem.unn.ru/Moltran/Moltran.pdf Download the current version of MOLTRAN (v.2.5, build 820) with Short Manual and UnInstaller (MoltranSetup.exe, zipped to avoid the server restrictions) http://ichem.unn.ru/Moltran/MoltranSetup.zip From owner-chemistry@ccl.net Thu Aug 30 07:25:01 2007 From: "Yangsoo Kim vsmember(~)gmail.com" To: CCL Subject: CCL:G: Is it possible to install G03 D.02 on the Linux Machine with Multi-Core & Multi-CPU? Message-Id: <-35057-070830045226-23997-+SdYsbBu/bN+poyWNO8Zyg[-]server.ccl.net> X-Original-From: "Yangsoo Kim" Content-Type: multipart/alternative; boundary="----=_Part_2426_16353221.1188460514046" Date: Thu, 30 Aug 2007 16:55:14 +0900 MIME-Version: 1.0 Sent to CCL by: "Yangsoo Kim" [vsmember===gmail.com] ------=_Part_2426_16353221.1188460514046 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline Dear CCL members I'm running G03 on the Linux 64bit Machine (AMD X2 4200+ DualCore CPU & M/B with Single CPU). I want to upgrade my computer spec. and construct SMP linux machine(multi-core). However I'm wondering It is possible to install G03 D.02 on the Linux machine with QuadCore CPU. My computer spec. candicates for upgrading are below. 1) AMD(opteron) Quadcore CPU + M/B with Dual CPU = Total 8 cores 2) AMD(opteron) Quadcore CPU + M/B with Single CPU = Total 4 cores 3) AMD(opteron) Dualcore CPU + M/B with Dual CPU = Total 4 cores 4) INTEL(xeon) Core2quad CPU + M/B with Dual CPU = Total 8 cores 5) INTEL(xeon) Core2quad CPU + M/B with Single CPU = Total 4 cores 6) INTEL(xeon) Core2duo CPU + M/B with Dual CPU = Totla 4 cores Which combination of CPU and M/B could I select? If tested and possble, please tell me candidate number... Or not, recommand me another combination of CPU and M/B. Great thanks, Yangsoo Kim ------=_Part_2426_16353221.1188460514046 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline

Dear CCL members

I'm running G03 on the Linux 64bit Machine (AMD X2 4200+ DualCore CPU & M/B with Single CPU).
I want to upgrade my computer spec. and construct SMP linux machine(multi-core).
However I'm wondering It is possible to install G03 D.02 on the Linux machine with QuadCore CPU.

My computer spec. candicates for upgrading are below.
1) AMD(opteron) Quadcore CPU + M/B with Dual CPU = Total 8 cores
2) AMD(opteron) Quadcore CPU + M/B with Single CPU = Total 4 cores
3) AMD(opteron) Dualcore CPU + M/B with Dual CPU = Total 4 cores
4) INTEL(xeon) Core2quad CPU + M/B with Dual CPU = Total 8 cores
5) INTEL(xeon) Core2quad CPU + M/B with Single CPU = Total 4 cores
6) INTEL(xeon) Core2duo CPU + M/B with Dual CPU = Totla 4 cores

Which combination of CPU and M/B could I select?
If tested and possble, please tell me candidate number...
Or not, recommand me another combination of CPU and M/B.

Great thanks,

Yangsoo Kim

------=_Part_2426_16353221.1188460514046-- From owner-chemistry@ccl.net Thu Aug 30 08:37:00 2007 From: "Xavier Gallet xgallet{:}nautilusbiotech.com" To: CCL Subject: CCL: Tinker compilation Message-Id: <-35058-070830083456-12807-KKrAk8pAnP9lCZoZ9UnCWw-*-server.ccl.net> X-Original-From: "Xavier Gallet" Date: Thu, 30 Aug 2007 08:34:53 -0400 Sent to CCL by: "Xavier Gallet" [xgallet|*|nautilusbiotech.com] Dear CCLers, I'd like to recompile the Tinker suite under Windows after modification of the maxatm value. I used compile.make, library.make and link.make files, respectively using the g77 Fortran compiler of Cygwin. The first file runs correctly. But the library.make file gave the following error: "ar: libtinker.a: File format not recognized" Please, can you suggest me some solutions ? Thank you in advance, Xavier From owner-chemistry@ccl.net Thu Aug 30 14:02:01 2007 From: "Shrinwantu Pal paul|,|jncasr.ac.in" To: CCL Subject: CCL: Heterogeneous catalysis in Gaussian Message-Id: <-35059-070830050751-30133-iOp7WLDunAWG1bU8y4Dvig^^server.ccl.net> X-Original-From: "Shrinwantu Pal" Content-Transfer-Encoding: 8bit Content-Type: text/plain;charset=iso-8859-1 Date: Thu, 30 Aug 2007 14:45:21 +0530 (IST) MIME-Version: 1.0 Sent to CCL by: "Shrinwantu Pal" [paul!=!jncasr.ac.in] Hi , ab init and vasp sure do plane wave basis set calculations for infinite systems. however u can use siesta also. it is freely downloadable. also to do this u need to download (or generate) pseudo potential files for Ti and O. See documentation. Once u have the structure, import it to do a local(small) calculation as to how the systems docks onto the catlyst, as in ur case is TiO2. Best of luck.. S.P JNCASR India From owner-chemistry@ccl.net Thu Aug 30 14:37:00 2007 From: "Alfredo Mayall Simas simas###ufpe.br" To: CCL Subject: CCL: Replacing PM3 with RM1 Message-Id: <-35060-070830143023-16664-cMw7IeFldX/VvJnlzQbL5w:server.ccl.net> X-Original-From: "Alfredo Mayall Simas" Content-Type: multipart/alternative; boundary="----=_NextPart_000_00C1_01C7EB1A.ADEC2580" Date: Thu, 30 Aug 2007 15:30:22 -0300 MIME-Version: 1.0 Sent to CCL by: "Alfredo Mayall Simas" [simas(_)ufpe.br] This is a multi-part message in MIME format. ------=_NextPart_000_00C1_01C7EB1A.ADEC2580 Content-Type: text/plain; charset="ISO-8859-9" Content-Transfer-Encoding: quoted-printable Dear Sina, Thank you for your interest in the RM1 model. Answering your questions, please note that parameters are not = interchangeable among different semiempirical methods. Mixing parameters = > from different methods will only lead to unstable results, mostly very = innacurate. Hence, if you use parameters designed for AM1 in RM1, the results will = likely be unreliable, at times random - seldom good, most times bad. Thus, calculations with mixed parameters cannot be trusted. The advice = by John McKelvey, reinforced by Gustavo Seabra, is truly valid. When you say that "However all the replacements I have seen are between = AM1 and RM1", actually, what you have probably been seeing is = implementation of RM1 in a software which already has AM1.=20 That is possible (and is very easy to be carried out) because the = equations in both AM1 and RM1 are identical: RM1 differs from AM1 only = in the values of the parameters. But after replacing the AM1 parameters with the RM1 ones, the = calculations are no longer AM1. Indeed they then become RM1 calculations = only, with no mixing of parameters involved. What you need, for your specific work on chlorophylls, are RM1 = parameters for Mg. However, I am sorry to say that they are not yet = available - but we are carrying out research in that direction. I am very glad to hear from you that RM1 has been shown to perform = better for the types of biomolecules you are working on. More details on the RM1 model can be found at = http://www.rm1.sparkle.pro.br Sincerely yours, Alfredo Mayall Simas ---------------------------------------------------------- Departamento de Qu=EDmica Fundamental Universidade Federal de Pernambuco Recife, PE, Brasil ---------------------------------------------------------- But I think the difference is (as someone told me) the equations in RM1 = and AM1 are the same where is in PM3 there are some differences. Onyl = problem here is that those MG parameters I found were generated with AM1 = parameters which also include interractions with the "old" N parameters. = But still substitutin those MG parameters into RM1 seems to give good = results.=20 On 8/29/07, Gustavo Seabra gustavo.seabra*_*gmail.com < = owner-chemistry_._ccl.net> wrote:=20 Sent to CCL by: "Gustavo Seabra" [gustavo.seabra()gmail.com] On 8/29/07, Sina T=FCreli sinatureli..gmail.com wrote: > Thanks for the advice, I have found this=20 > > http://gepard.bioinformatik.uni-saarland.de/people/hutter/mgam1.html = > > which is a work to place a MG parameter in AM1 (so I suppose it = should work=20 > with RM1). I have found the files that are used for giving the = paramteres in > spartan. I am now replacing them all to see if they give good = results. Just a note... RM1 is a "re-parametrization" of AM1, as is PM3. In the = end, it is a different semi-empirical method and, in principle, John McKelvey's advice is still valid here. > On the other hand can mopac do geometry optimizations? Thanks... It should... Gustavo. ----- Original Message -----=20 From: Sina T=FCreli sinatureli..gmail.com=20 Sent: Wednesday, August 29, 2007 12:53 PM Subject: CCL: Placing RM1 parameters to PM3 Thanks for the advice, I have found this http://gepard.bioinformatik.uni-saarland.de/people/hutter/mgam1.html which is a work to place a MG parameter in AM1 (so I suppose it should = work with RM1). I have found the files that are used for giving the = paramteres in spartan. I am now replacing them all to see if they give = good results.=20 =20 On the other hand can mopac do geometry optimizations? Thanks... ----- Original Message -----=20 From: John McKelvey jmmckel-$-gmail.com=20 Sent: Wednesday, August 29, 2007 10:58 AM Subject: CCL: Placing RM1 parameters to PM3 One should be careful replacing a single element's parameters from one = parameter set with that from another. I do not know if it is possible = to replace parameetrs in Spartan. =20 On the other hand, academeic researchers can get the latest MOPAC from = Jimmy Stewart for no cost, and this contains RM1, and is _extremely_ = efficient. [google "MRMOPAC" ]=20 Cheers, John McKelvey Or in another point of view, will adding MG parameters into a RM1 or AM1 = parameters file work? On 8/29/07, Sina T reli sinatureli%x%gmail.com < = owner-chemistry]^[ccl.net> wrote:=20 Sent to CCL by: "Sina T reli" [sinatureli],[ gmail.com] Greetings, I am working with biomolecules using spartan 04 and I would like to = update my se methods to RM1 which has been shown to perform better for = certain types of biomolecules (one is partially which I am working on). = I have seen people replacing parameters in AM1 with those in RM1. But = since the molecule I am working with contains MG, I need to use the PM3 = method. I am wondering, will just replacing the paramters in PM3 with = those in RM1 is likely to produce a good result? Because all the = replacements I have seen is between AM1-RM1=20 Thanks, Sina ------=_NextPart_000_00C1_01C7EB1A.ADEC2580 Content-Type: text/html; charset="ISO-8859-9" Content-Transfer-Encoding: quoted-printable
Dear Sina,
 
Thank you for your interest in the RM1=20 model.
 
Answering your questions, please note = that=20 parameters are not interchangeable among different semiempirical = methods.=20 Mixing parameters from different = methods will=20 only lead to unstable results, mostly very innacurate.
 
Hence, if you use parameters = designed for AM1=20 in RM1, the results will likely be unreliable, at = times random=20 - seldom good, most times bad.
 
Thus, calculations with mixed = parameters cannot be=20 trusted. The advice by John McKelvey, reinforced by Gustavo=20 Seabra, is truly valid.
 
When you say that "However all the = replacements I=20 have seen are between AM1 and RM1", actually, what you have = probably=20 been seeing is implementation of RM1 in a software which already = has AM1.=20
That is possible (and is very easy to = be carried=20 out) because the equations in both AM1 and RM1 are identical: RM1 = differs from=20 AM1 only in the values of the parameters.
But after replacing the AM1 parameters = with the RM1=20 ones, the calculations are no longer AM1. Indeed they then=20 become RM1 calculations only, with no mixing of parameters=20 involved.
 
What you need, for your specific work on chlorophylls, are RM1 = parameters=20 for Mg. However, I am sorry to say that they are not yet available = - but we=20 are carrying out research in that direction.
 
I am very glad to hear from you that = RM1 has been=20 shown to perform better for the types of biomolecules you are = working=20 on.
 
More details on the RM1 model can be = found at http://www.rm1.sparkle.pro.br<= /FONT>
 
Sincerely yours,
Alfredo Mayall Simas
 
----------------------------------------------------------
Departamento de Qu=EDmica = Fundamental
Universidade Federal de = Pernambuco
Recife, PE, Brasil
----------------------------------------------------------
 
But I think the=20 difference is (as someone told me) the equations in RM1 and AM1 are the = same=20 where is in PM3 there are some differences. Onyl problem here is that = those MG=20 parameters I found were generated with AM1 parameters which also include = interractions with the "old" N parameters. But still substitutin those = MG=20 parameters into RM1 seems to give good results.

On 8/29/07, Gustavo=20 Seabra gustavo.seabra*_*gmail.com <=20 owner-chemistry_._ccl.net> wrote:=20

Sent=20 to CCL by: "Gustavo Seabra" [gustavo.seabra()gmail.com]
On 8/29/07, = Sina=20 T=FCreli sinatureli..gmail.com <owner-chemistry:_:=20 ccl.net> wrote:
> Thanks for the advice, I have found = this=20
>
> http://gepard.bioinformatik.uni-saarland.de/people/hutter= /mgam1.html=20
>
> which is a work to place a MG parameter in AM1 = (so I=20 suppose it should work
> with RM1). I have found the files that = are=20 used for giving the paramteres in
> spartan. I am now replacing = them all=20 to see if they give good results.

Just a note... RM1 is a=20 "re-parametrization" of AM1, as is PM3. In the
end, it is a = different=20 semi-empirical method and, in principle, John
McKelvey's advice is = still=20 valid here.

> On the other hand can mopac do geometry = optimizations?=20 Thanks...

It should...

Gustavo.
----- Original Message -----=20
From: Sina=20 T=FCreli sinatureli..gmail.com
Sent: Wednesday, August 29, 2007 12:53 PM
Subject: CCL: Placing RM1 parameters to PM3

Thanks for the advice, I have found this

http://gepard.bioinformatik.uni-saarland.de/people/hutter/mgam1.html=

which=20 is a work to place a MG parameter in AM1 (so I suppose it should work = with=20 RM1). I have found the files that are used for giving the paramteres = in=20 spartan. I am now replacing them all to see if they give good results. =
 

On the other hand can mopac do geometry = optimizations?=20 Thanks...

 
----- Original Message -----=20
From: John=20 McKelvey jmmckel-$-gmail.com
Sent: Wednesday, August 29, 2007 10:58 AM
Subject: CCL: Placing RM1 parameters to PM3

One should be careful replacing a single element's = parameters=20 from one parameter set with that from another.  I do not know if = it is=20 possible to replace parameetrs in Spartan. 

On the other = hand,=20 academeic researchers can get the latest MOPAC from Jimmy Stewart for = no cost,=20 and this contains RM1, and is _extremely_ efficient. [google "MRMOPAC" = ]=20

Cheers,

John=20 McKelvey

Or in another point of view, will = adding MG=20 parameters into a RM1 or AM1 parameters file work?
On 8/29/07, Sina T reli=20 sinatureli%x%gmail.com < = owner-chemistry]^[ccl.net>=20 wrote:=20

Sent=20 to CCL by: "Sina  T  reli" [sinatureli],[ gmail.com]
Greetings,

I am working with = biomolecules=20 using spartan 04 and I would like to update my se methods to RM1 which = has=20 been shown to perform better for certain types of biomolecules (one is = partially which I am working on). I have seen people replacing = parameters in=20 AM1 with those in RM1. But since the molecule I am working with = contains MG, I=20 need to use the PM3 method. I am wondering, will just replacing the = paramters=20 in PM3 with those in RM1 is likely to produce a good result? Because = all the=20 replacements I have seen is between AM1-RM1=20

Thanks,
Sina
 ------=_NextPart_000_00C1_01C7EB1A.ADEC2580-- From owner-chemistry@ccl.net Thu Aug 30 19:53:01 2007 From: "Gustavo A Mercier gustavo.mercier+/-bmc.org" To: CCL Subject: CCL: Computational Chemistry and Molecular Imaging Message-Id: <-35061-070830134129-1140-nAxgjRwRFgqwAVWT8rYd0Q() server.ccl.net> X-Original-From: "Gustavo A Mercier" Date: Thu, 30 Aug 2007 13:41:25 -0400 Sent to CCL by: "Gustavo A Mercier" [gustavo.mercier===bmc.org] Hi! I am interested in the application of computatinal chemistry in the development of molecular imaging agents, and contrast media in general. In the past I did work on a putative MRI contrast agent based on a Manganese Metalloporphyrin. With the recent introduction of a new journal, 'Contrast Media', database, MICAD, and interest group by the Society of Nuclear Medicine (Molecular Imaging), the time seems right to identify those members of the computational chemistry community whose interest is in the development of molecular imaging agents and contrast media. My interest is to exchange ideas, and see how we could promote computational chemistry as a means to expedite the development of new molcular imaging agents and contrast media. It seems to me that even within the chemistry community, this endeavor remains a small fraction of the work done by computational chemists. It also seems relatively unknown among the larger radiology/nuclear medicine departments that work on developing such agents. I am sending this e-mail to the CCL community in the hopes of compiling a list of individuals with interest in this field. If you would like to start a special interest group (for now simple exchange of e-mail addresses), please let me know. Thank you for your time! -- Gustavo A. Mercier, Jr. MD,PhD Boston Medical Center Radiology - Nuclear Medicine gamercier::yahoo.com (preferred e-mail address) Gustavo.Mercier::bmc.org gumercie::bu.edu From owner-chemistry@ccl.net Thu Aug 30 20:28:01 2007 From: "John McKelvey jmmckel|gmail.com" To: CCL Subject: CCL: Replacing PM3 with RM1 Message-Id: <-35062-070830191003-19568-Dg3LNOHK+IFLRCrSp3G7HQ\a/server.ccl.net> X-Original-From: "John McKelvey" Content-Type: multipart/alternative; boundary="----=_Part_8043_27776061.1188506885217" Date: Thu, 30 Aug 2007 16:48:05 -0400 MIME-Version: 1.0 Sent to CCL by: "John McKelvey" [jmmckel.\a/.gmail.com] ------=_Part_8043_27776061.1188506885217 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Content-Disposition: inline Why not try PM6 in the latest MOPAC [V7.2] ? I believe in has parameters for H through I, probably with the exception on the rare fases. Cheers, John McKelvey On 8/30/07, Alfredo Mayall Simas simas###ufpe.br wrote: > > Dear Sina, > > Thank you for your interest in the RM1 model. > > Answering your questions, please note that parameters are not > interchangeable among different semiempirical methods. Mixing parameters > from different methods will only lead to unstable results, mostly very > innacurate. > > Hence, if you use parameters designed for AM1 in RM1, the results will > likely be unreliable, at times random - seldom good, most times bad. > > Thus, calculations with mixed parameters cannot be trusted. The advice by > John McKelvey, reinforced by Gustavo Seabra, is truly valid. > > When you say that "However all the replacements I have seen are between > AM1 and RM1", actually, what you have probably been seeing is implementat= ion > of RM1 in a software which already has AM1. > That is possible (and is very easy to be carried out) because the > equations in both AM1 and RM1 are identical: RM1 differs from AM1 only in > the values of the parameters. > But after replacing the AM1 parameters with the RM1 ones, the calculation= s > are no longer AM1. Indeed they then become RM1 calculations only, with no > mixing of parameters involved. > > What you need, for your specific work on chlorophylls, are RM1 parameters > for Mg. However, I am sorry to say that they are not yet available - but = we > are carrying out research in that direction. > > I am very glad to hear from you that RM1 has been shown to perform bette= r > for the types of biomolecules you are working on. > > More details on the RM1 model can be found at > http://www.rm1.sparkle.pro.br > > Sincerely yours, > Alfredo Mayall Simas > > ---------------------------------------------------------- > Departamento de Qu=EDmica Fundamental > Universidade Federal de Pernambuco > Recife, PE, Brasil > ---------------------------------------------------------- > > But I think the difference is (as someone told me) the equations in RM1 > and AM1 are the same where is in PM3 there are some differences. Onyl > problem here is that those MG parameters I found were generated with AM1 > parameters which also include interractions with the "old" N parameters. = But > still substitutin those MG parameters into RM1 seems to give good results= . > > On 8/29/07, Gustavo Seabra gustavo.seabra*_*gmail.com > wrote: > > > > > > Sent to CCL by: "Gustavo Seabra" [gustavo.seabra()gmail.com] > > On 8/29/07, Sina T=FCreli sinatureli..gmail.com > > wrote: > > > Thanks for the advice, I have found this > > > > > > http://gepard.bioinformatik.uni-saarland.de/people/hutter/mgam1.html > > > > > > which is a work to place a MG parameter in AM1 (so I suppose it shoul= d > > work > > > with RM1). I have found the files that are used for giving the > > paramteres in > > > spartan. I am now replacing them all to see if they give good results= . > > > > Just a note... RM1 is a "re-parametrization" of AM1, as is PM3. In the > > end, it is a different semi-empirical method and, in principle, John > > McKelvey's advice is still valid here. > > > > > On the other hand can mopac do geometry optimizations? Thanks... > > > > It should... > > > > Gustavo. > > ----- Original Message ----- *From:* Sina T=FCreli sinatureli..gmail.co= m > > *Sent:* Wednesday, August 29, 2007 12:53 PM > > *Subject:* CCL: Placing RM1 parameters to PM3 > > > > Thanks for the advice, I have found this > > > > http://gepard.bioinformatik.uni-saarland.de/people/hutter/mgam1.html > > > > which is a work to place a MG parameter in AM1 (so I suppose it should > > work with RM1). I have found the files that are used for giving the > > paramteres in spartan. I am now replacing them all to see if they give = good > > results. > > > > > > On the other hand can mopac do geometry optimizations? Thanks... > > > > > > ----- Original Message ----- *From:* John McKelvey jmmckel-$-gmail.com= > > *Sent:* Wednesday, August 29, 2007 10:58 AM > > *Subject:* CCL: Placing RM1 parameters to PM3 > > > > One should be careful replacing a single element's parameters from one > > parameter set with that from another. I do not know if it is possible = to > > replace parameetrs in Spartan. > > > > On the other hand, academeic researchers can get the latest MOPAC from > > Jimmy Stewart for no cost, and this contains RM1, and is _extremely_ > > efficient. [google "MRMOPAC" ] > > > > Cheers, > > > > John McKelvey > > > > Or in another point of view, will adding MG parameters into a RM1 or AM= 1 > parameters file work? > > On 8/29/07, Sina T reli sinatureli%x%gmail.com < owner-chemistry]^[ccl.ne= t> > wrote: > > > > > > Sent to CCL by: "Sina T reli" [sinatureli],[ gmail.com] > > Greetings, > > > > I am working with biomolecules using spartan 04 and I would like to > > update my se methods to RM1 which has been shown to perform better for > > certain types of biomolecules (one is partially which I am working on).= I > > have seen people replacing parameters in AM1 with those in RM1. But sin= ce > > the molecule I am working with contains MG, I need to use the PM3 metho= d. I > > am wondering, will just replacing the paramters in PM3 with those in RM= 1 is > > likely to produce a good result? Because all the replacements I have se= en is > > between AM1-RM1 > > > > Thanks, > > Sina > > > ------=_Part_8043_27776061.1188506885217 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Content-Disposition: inline Why not try PM6 in the latest MOPAC [V7.2]  ?  I believe in has p= arameters for H through I, probably with the exception on the rare fases.
Cheers,

John McKelvey

= On 8/30/07,=20 Alfredo Mayall Simas simas###ufpe.br <= owner-chemistry\a/ccl.net> = wrote:
Dear Sina,
 
Thank you for your interest in the RM1= =20 model.
 
Answering your questions, please note = that=20 parameters are not interchangeable among different semiempirical metho= ds.=20 Mixing parameters from different met= hods will=20 only lead to unstable results, mostly very innacurate.
 
Hence, if you use parameters des= igned for AM1=20 in RM1, the results will likely be unreliable, at times rand= om=20 - seldom good, most times bad.
 
Thus, calculations with mixed paramete= rs cannot be=20 trusted. The advice by John McKelvey, reinforced by Gustavo=20 Seabra, is truly valid.
 
When you say that "However all th= e replacements I=20 have seen are between AM1 and RM1", actually, what you have = probably=20 been seeing is implementation of RM1 in a software which already has A= M1.=20
That is possible (and is very easy to = be carried=20 out) because the equations in both AM1 and RM1 are identical: RM1 differs f= rom=20 AM1 only in the values of the parameters.
But after replacing the AM1 parameters= with the RM1=20 ones, the calculations are no longer AM1. Indeed they then=20 become RM1 calculations only, with no mixing of parameters=20 involved.
 
What you need, for your specific work on chlorophylls, are RM1 paramet= ers=20 for Mg. However, I am sorry to say that they are not yet available - b= ut we=20 are carrying out research in that direction.
 
I am very glad to hear from you that R= M1 has been=20 shown to perform better for the types of biomolecules you are wor= king=20 on.
 
More details on the RM1 model can be f= ound at http://www.rm1.sparkle.pr= o.br
 
Sincerely yours,
Alfredo Mayall Simas
 
--------------------------------------= --------------------
Departamento de Qu=EDmica Fundamental<= /font>
Universidade Federal de Pernambuco
Recife, PE, Brasil
--------------------------------------= --------------------
 
But I think the=20 difference is (as someone told me) the equations in RM1 and AM1 are the sam= e=20 where is in PM3 there are some differences. Onyl problem here is that those= MG=20 parameters I found were generated with AM1 parameters which also include=20 interractions with the "old" N parameters. But still substitutin = those MG=20 parameters into RM1 seems to give good results.

On = 8/29/07, Gustavo=20 Seabra gustavo.seabra*_*gmail.com <=20 owner-chemistry_._ccl.net> wrote:=20

Sent=20 to CCL by: "Gustavo Seabra" [gustavo.seabra()gmail.com]
On 8= /29/07, Sina=20 T=FCreli sinatureli..gmail.com <owner-chemistry:_:=20 ccl.net> wrote:
> Thanks for the advice, I have found this= =20
>
> http://gepard.bioinformatik.uni-saarland.de/people/hu= tter/mgam1.html=20
>
> which is a work to place a MG parameter in AM1 (so I= =20 suppose it should work
> with RM1). I have found the files that ar= e=20 used for giving the paramteres in
> spartan. I am now replacing the= m all=20 to see if they give good results.

Just a note... RM1 is a=20 "re-parametrization" of AM1, as is PM3. In the
end, it is a= different=20 semi-empirical method and, in principle, John
McKelvey's advice is= still=20 valid here.

> On the other hand can mopac do geometry optimizat= ions?=20 Thanks...

It should...

Gustavo.
--= --- Original Message -----=20
Sent: Wednesday, August 29, 2007 12:53 PM
Subject: CCL: Placing RM1 parameters to PM3

Thanks for the advice, = I have found this

http://gepard.bioinformatik.uni-saarland.de/people/hutter/mgam1.html
which=20 is a work to place a MG parameter in AM1 (so I suppose it should work wit= h=20 RM1). I have found the files that are used for giving the paramteres in= =20 spartan. I am now replacing them all to see if they give good results.=20
 

On the other hand can mopac do geometry optimizations?= =20 Thanks...

 
----- Original Message -----=20
Sent: Wednesday, August 29, 2007 10:58 AM
Subject: CCL: Placing RM1 parameters to PM3

One should be careful r= eplacing a single element's parameters=20 from one parameter set with that from another.  I do not know if it = is=20 possible to replace parameetrs in Spartan. 

On the other han= d,=20 academeic researchers can get the latest MOPAC from Jimmy Stewart for no = cost,=20 and this contains RM1, and is _extremely_ efficient. [google "MRMOPA= C" ]=20

Cheers,

John=20 McKelvey

Or in another point of= view, will adding MG=20 parameters into a RM1 or AM1 parameters file work?
On = 8/29/07, Sina T reli=20 sinatureli%x%gmail.com < owner-chemistry]^[ccl.net>=20 wrote:=20

Sent=20 to CCL by: "Sina  T  reli" [sinatureli],[ <= a>gmail.com]
Greetings,

I am working with biomolecules=20 using spartan 04 and I would like to update my se methods to RM1 which ha= s=20 been shown to perform better for certain types of biomolecules (one is=20 partially which I am working on). I have seen people replacing parameters= in=20 AM1 with those in RM1. But since the molecule I am working with contains = MG, I=20 need to use the PM3 method. I am wondering, will just replacing the param= ters=20 in PM3 with those in RM1 is likely to produce a good result? Because all = the=20 replacements I have seen is between AM1-RM1=20

Thanks,
Sina

------=_Part_8043_27776061.1188506885217-- From owner-chemistry@ccl.net Thu Aug 30 21:50:00 2007 From: "Raji Raji raji_._anal.chem.tohoku.ac.jp" To: CCL Subject: CCL:G: Error termination via Lnk1e in /usr/ap/g03/l914.exe Message-Id: <-35063-070830214730-20735-np682dmazBtaH+HZ5HzWuw!A!server.ccl.net> X-Original-From: "Raji Raji" Date: Thu, 30 Aug 2007 21:47:26 -0400 Sent to CCL by: "Raji Raji" [raji..anal.chem.tohoku.ac.jp] Dear CCL Members, My TDDFT calculation on an organic molecule is terminated with an error. Please do find the part of the input and out put files below. It seems the error in the Link commands. I checked in the gaussian manual, it says that the L914.exe corresponds to CI-Singles, RPA and Zindo excited states; SCF stability. But dont know how to solve this problem. Any help will be thankfully acknowledged. INPUT ------------------------------------------------------------ %Chk=molecule-ex.chk %NProc=16 %Mem=128Gb TD(NStates=15, Singlets, Direct) B3LYP/6-311+G(d) SCRF=(CPCM, Solvent=Water) AmDMPT Excited Singlet state calculation (CPCM-Water) 0 1 ------------------------------------------------------------ OUTPUT ------------------------------------------------------------ 60 initial guesses have been made. Convergence on wavefunction: 0.001000000000000 Iteration 1 Dimension 60 NMult 60 CISAX will form 60 AO SS matrices at one time. Unable to match L and R vectors in BiOrth. Error termination via Lnk1e in /usr/ap/g03/l914.exe at Thu Aug 30 12:17:23 2007. Job cpu time: 0 days 2 hours 10 minutes 30.8 seconds. File lengths (MBytes): RWF= 599 Int= 0 D2E= 0 Chk= 5 Scr= 1 ------------------------------------------------------------