From owner-chemistry@ccl.net Sun Jan 11 05:28:01 2015 From: "Stefan Grimme grimme:_:thch.uni-bonn.de" To: CCL Subject: CCL: Molecular dynamics for metal complexes Message-Id: <-50886-150111042534-12882-irNF9CJ5flXG6+syao41SQ-$-server.ccl.net> X-Original-From: "Stefan Grimme" Date: Sun, 11 Jan 2015 04:25:32 -0500 Sent to CCL by: "Stefan Grimme" [grimme++thch.uni-bonn.de] Dear Eli, I have recently developed a classical but molecule-specific force-field which can be generated automatically also for metal complexes and which can be used for MD (A General Quantum Mechanically Derived Force Field (QMDFF) for Molecules and Condensed Phase Simulations, J. Chem. Theory Comput., (2014), 10, 4497-4514. DOI: 10.1021/ct500573f). As input you need a QM optimized structure and corresponding Hessian as obtained from standard electronic structure packages (e.g. ORCA). Cheers! Stefan From owner-chemistry@ccl.net Sun Jan 11 17:46:00 2015 From: "uekstrom.:.gmail.com uekstrom.:.gmail.com" To: CCL Subject: CCL: Any regrets open-sourcing your programs? Message-Id: <-50887-150111051713-15403-SPTDDxf7W1E5dPaDvJpjkw*server.ccl.net> X-Original-From: "uekstrom---gmail.com" Content-Type: multipart/alternative; boundary=001a11c381be3955ce050c5db1ea Date: Sun, 11 Jan 2015 11:17:08 +0100 MIME-Version: 1.0 Sent to CCL by: "uekstrom]^[gmail.com" [uekstrom]^[gmail.com] --001a11c381be3955ce050c5db1ea Content-Type: text/plain; charset=UTF-8 Thanks to everyone that replied to the open source thread. In particular the RasMol case was interesting. Our situation is this: We have programs that are already distributed in source form free of charge, but with a special license agreement. We do not expect any serious contributions from people outside the current community just because we open source the program. (However, if the program is accepted into a major Linux distribution you get automatically free high quality testing and bug reports on many platforms). What we do expect is that open sourcing will make release management much easier. This is both because we can use excellent free hosting services (such as github), and because we don't have to bother with faxing licences etc. It is hard to estimate how big the savings would be, but I think it amounts to some man-months per release+patches. > From the replies in this thread I don't see any really bad examples of the "dangers of open source". My own worst example is this: I released a software library under the LGPL license and took contributions from many people. Then it turned out that LGPL was not appropriate for various reasons, and I wanted to relicense to MIT. Of course I then had to get permission from all the contributors. Luckily there were no problems, but this is not something that you can count on. If I would have wanted to commercialize the program I would probably have to rewrite it from scratch (which is fair since it is now owned by all the contributors). Regards, Ulf --001a11c381be3955ce050c5db1ea Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable
Thanks to everyone that replied to the open source thread.= In particular the RasMol case was interesting.

Our situ= ation is this: We have programs that are already distributed in source form= free of charge, but with a special license agreement. We do not expect any= serious contributions from people outside the current community just becau= se we open source the program. (However, if the program is accepted into a = major Linux distribution you get automatically free high quality testing an= d bug reports on many platforms).

What we do expec= t is that open sourcing will make release management much easier. This is b= oth because we can use excellent free hosting services (such as github), an= d because we don't have to bother with faxing licences etc. It is hard = to estimate how big the savings would be, but I think it amounts to some ma= n-months per release+patches.

From the replies in = this thread I don't see any really bad examples of the "dangers of= open source". My own worst example is this: I released a software lib= rary under the LGPL license and took contributions from many people. Then i= t turned out that LGPL was not appropriate for various reasons, and I wante= d to relicense to MIT. Of course I then had to get permission from all the = contributors. Luckily there were no problems, but this is not something tha= t you can count on. If I would have wanted to commercialize the program I w= ould probably have to rewrite it from scratch (which is fair since it is no= w owned by all the contributors).=C2=A0

Regards,
Ulf
--001a11c381be3955ce050c5db1ea-- From owner-chemistry@ccl.net Sun Jan 11 21:52:01 2015 From: "Brian Skinn bskinn[]alum.mit.edu" To: CCL Subject: CCL: Molecular dynamics for metal complexes Message-Id: <-50888-150111214801-13549-bJqwRee76lP5ewXCQoG5ug]*[server.ccl.net> X-Original-From: Brian Skinn Content-Type: multipart/alternative; boundary=f46d043bdf7c7d7892050c6b88b2 Date: Sun, 11 Jan 2015 21:47:34 -0500 MIME-Version: 1.0 Sent to CCL by: Brian Skinn [bskinn,,alum.mit.edu] --f46d043bdf7c7d7892050c6b88b2 Content-Type: text/plain; charset=UTF-8 Eli, My experience with MM/MD and force field construction is limited, but I believe the main challenge is a lack of sufficiently universal underlying trends in the PES'es of transition metal complexes. Whereas for organic compounds the types/geometries of bonding are quite few and the valences of atoms singular or few (e.g., oxygen nearly always 2; sulfur most often 2, 4, or 6), in transition metal complexes not even the geometries can necessarily be trusted to remain consistent from complex to complex (all sorts of gradations are possible between square-planar and tetrahedral for 4-coordination, for example), and thus not just the force constants but also the equilibrium geometry parameters are uncertain. One thus encounters a sort of chicken-and-egg problem: for practically relevant systems, one must often have higher-level (often quite computationally expensive) theoretical computation results available in order to assemble the force field since experimental data (I think?) are typically insufficient for force field construction. (I welcome correction from experts in the field!) That being said, one of the most prominent names I know of in MM/MD of transition metal species is Peter Comba. As a starting point, I'd look at his publications and the citations therein, including: - Comba, P & Hambley, T.W. "Molecular Modeling of Inorganic Compounds." Wiley, 1995. ISBN 3527290761 (newer edition possibly available) - Comba P & Remenyi R. *Coord Chem Rev* 238-239 (2003) 9-20 doi: 10.1016/S0010-8545(02)00286-2 - Comba P & Schiek W. *Coord Chem Rev* 238-239 (2003) 21-29 doi: 10.1016/S0010-8545(02)00294-1 - Comba P & Kerscher M *Coord Chem Rev* 253 (2009) 564-574 doi: 10.1016/j.ccr.2008.05.019 Another big name is Robert Deeth: - Deeth RJ et al. *Coord Chem Rev *253 (2009) 795-816 doi: 10.1016/j.ccr.2008.06.018 - Below are two application-specific papers, which deal with MM treatment of Jahn-Teller distortions in Cu(II) amine ligand complexes (the details of the nature of Cu(II) J-T distortions had caught my interest for a time): - Burton VJ & Deeth RJ. *J Chem Soc Chem Commun* (1995) 573-574 10.1039/C39950000573 - Deeth RJ & Hearnshaw LJA. *Dalton Transactions* (2006) 1092-1100 10.1016/j.ccr.2008.06.018 I'm sure both of these researchers have more recent publications as well, and they are by no means alone in the effort to develop MM/MD for transition metal-containing systems. Also, the two issues of *Coord Chem Rev* cited above have much more to offer in terms of background on computational transition metal chemistry. Although they are aging somewhat (a new *CCR* issue devoted to computational/theoretical coordination chemistry would be most welcome!), and in particular provide little information on, e.g., empirical dispersion corrections and range-corrected and other non-local density functionals, in my opinion they provide a superb starting point. Best of luck, Brian On Wed, Jan 7, 2015 at 4:25 AM, Eli Lam eli421^hku.hk < owner-chemistry~~ccl.net> wrote: > > Sent to CCL by: "Eli Lam" [eli421[*]hku.hk] > Dear CCLers, > > I am totally new to molecular dynamics simulations. Yet I have heard from > my > seniors commenting that MD simulations is unrealistic towards transition > metal > complexes especially in non-convalent interactions. May I ask about the > reasons? I would be grateful if CCLers can introduce me some materials > about > the basics on MD. > > Thank you so much in advance! > Eli> > > --f46d043bdf7c7d7892050c6b88b2 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable
Eli,

My experience with MM/MD and force= field construction is limited, but I believe the main challenge is a lack = of sufficiently universal underlying trends in the PES'es of transition= metal complexes.=C2=A0 Whereas for organic compounds the types/geometries = of bonding are quite few and the valences of atoms singular or few (e.g., o= xygen nearly always 2; sulfur most often 2, 4, or 6), in transition metal c= omplexes not even the geometries can necessarily be trusted to remain consi= stent from complex to complex (all sorts of gradations are possible between= square-planar and tetrahedral for 4-coordination, for example), and thus n= ot just the force constants but also the equilibrium geometry parameters ar= e uncertain.=C2=A0 One thus encounters a sort of chicken-and-egg problem: f= or practically relevant systems, one must often have higher-level (often qu= ite computationally expensive)=C2=A0theoretical computation results availab= le in order to assemble the force field since experimental data (I think?) = are typically insufficient for force field construction.

(I welcome correction from experts in the field!)

That being said, one of the most prominent names I know of in MM/MD = of transition metal species is Peter Comba.=C2=A0 As a starting point, I= 9;d look at his publications and the citations therein, including:

Another big name is Robert Deeth:
    =
  • Deeth RJ et al. Coord Chem Rev 253 (2009) 795-816 doi:=C2=A010.1016/j.ccr.2008.06.0= 18
  • Below are two application-specific papers, which deal with M= M treatment of Jahn-Teller distortions in Cu(II) amine ligand complexes (th= e details of the nature of Cu(II) J-T distortions had caught my interest fo= r a time):=C2=A0

I'm sure both of these researchers hav= e more recent publications as well, and they are by no means alone in the e= ffort to develop MM/MD for transition metal-containing systems.
<= br>
Also, the two issues of Coord Chem Rev=C2=A0cited abov= e have much more to offer in terms of background on computational transitio= n metal chemistry.=C2=A0 Although they are aging somewhat (a new CCR= =C2=A0issue devoted to computational/theoretical coordination chemistry wou= ld be most welcome!), and in particular provide little information on, e.g.= , empirical dispersion corrections and range-corrected and other non-local = density functionals, in my opinion they provide a superb starting point.


Best of luck,
Bri= an


On Wed, Jan 7, 2015 at 4:25 AM, Eli Lam eli421^hku.hk <owner-chemistry~~ccl.net> wrote:<= br>

Sent to CCL by: "Eli=C2=A0 Lam" [eli421[*]hku.hk]
Dear CCLers,

I am totally new to molecular dynamics simulations. Yet I have heard from m= y
seniors commenting that MD simulations is unrealistic towards transition me= tal
complexes especially in non-convalent interactions. May I ask about the
reasons? I would be grateful if CCLers can introduce me some materials abou= t
the basics on MD.

Thank you so much in advance!
Eli



-=3D This is automatically added to each message by the mailing script =3D-=
E-mail to subscribers: CHEMISTRY~~ccl.n= et or use:
=C2=A0 =C2=A0 =C2=A0 http://www.ccl.net/cgi-bin/ccl/send_ccl_message=

E-mail to administrators: CHEM= ISTRY-REQUEST~~ccl.net or use
=C2=A0 =C2=A0 =C2=A0 http://www.ccl.net/cgi-bin/ccl/send_ccl_message=

Subscribe/Unsubscribe:
=C2=A0 =C2=A0 =C2=A0 http://www.ccl.net/chemistry/sub_unsub.shtml

Before posting, check wait time at: http://www.ccl.net

Job: http://www.ccl.n= et/jobs
Conferences: http://server.ccl.net/chemistry/announcements/co= nferences/

Search Messages: http://www.ccl.net/chemistry/searchccl/index.shtml
=C2=A0 =C2=A0 =C2=A0 http://www.ccl.net/spammers.txt

RTFI: http://www.ccl.net/chemistry/aboutccl/instructions/



--f46d043bdf7c7d7892050c6b88b2-- From owner-chemistry@ccl.net Sun Jan 11 22:26:01 2015 From: "Joseph E Maxwell jaymax36^^^gmail.com" To: CCL Subject: CCL: [CCL] Molecular dynamics for metal complexes Message-Id: <-50889-150111202129-22284-np682dmazBtaH+HZ5HzWuw*server.ccl.net> X-Original-From: Joseph E Maxwell Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=windows-1252; format=flowed Date: Sun, 11 Jan 2015 17:21:19 -0800 MIME-Version: 1.0 Sent to CCL by: Joseph E Maxwell [jaymax36(~)gmail.com] Dear Stefan, An interesting approach- but if you already have a QM optimized structure and the corresponding Hessian what is the real benefit in seemingly 'reverting' to a FF, except for perhaps solving for a derivative model. Some clarification (my re-education) will surely be appreciated. Thanks! Joseph On 1/11/2015 1:25 AM, Stefan Grimme grimme:_:thch.uni-bonn.de wrote: > Sent to CCL by: "Stefan Grimme" [grimme++thch.uni-bonn.de] > Dear Eli, > I have recently developed a classical but molecule-specific force-field which can be > generated automatically also for metal complexes and which can be used for MD > (A General Quantum Mechanically Derived Force Field (QMDFF) for Molecules and > Condensed Phase Simulations, J. Chem. Theory Comput., (2014), 10, 4497-4514. > DOI: 10.1021/ct500573f). > As input you need a QM optimized structure and corresponding Hessian as > obtained from standard electronic structure packages (e.g. ORCA). > Cheers! > Stefan> > >