Menu
▾
▴
Re: [cclib-devel] [cclib-users] FMO properties
|
From: Noel O'B. <bao...@gm...> - 2013-09-23 15:51:57
|
I no longer have access to any of these I'm afraid... On 19 September 2013 05:18, Karol M. Langner <kar...@gm...> wrote: > I looked at the atomcharges code, and fixed the GAMESS-US parser a bit. > > One more thing, though. Our current Jaguar and Molpro log files for unit testing do not contain any partial charges, so there is nothing to test at this point. I could update the Molpro files, but does anyone have access to Jaguar now? > > - Karol > > On Sep 18 2013, Karol M. Langner wrote: >> Yeah, forwarding to the dev list. >> >> So... any thoughts for/against these candidates: >> 1. aopop >> 2. aopops >> 3. aopopulation >> 4. aopopulations >> 5. populations >> >> I think I prefer the first one, for its brevity. >> >> Cheers, >> Karol >> >> On Sep 18 2013, Adam Tenderholt wrote: >> > Hi Karol, >> > >> > Thanks for taking care of this. I forgot that an atomcharges attribute was >> > added. As far as aopop or aopopuations, perhaps that discussion should be >> > moved to cclib-devel? >> > >> > Adam >> > >> > >> > >> > On Wed, Sep 18, 2013 at 3:24 PM, Karol M. Langner >> > <kar...@gm...>wrote: >> > >> > > Hi, >> > > >> > > Some time ago we actually created the 'atomcharges' attribute: >> > > http://cclib.sourceforge.net/wiki/index.php/Atomcharges >> > > (see that page for the data description) >> > > >> > > So I went ahead and extended that to NPA charges and just commited >> > > code that parses them for Gaussian09. And the newest trunk code >> > > should parse the output file you sent, Fedor. >> > > >> > > We don't have atomcharges coded yet for Turbomole or Jaguar, but it >> > > should not be hard and this would be a good time to do it. >> > > >> > > As far as populations go, I believe we would need to create a new attribute >> > > for that, since it doesn't fit anywhere else. It should probably also >> > > be a dictionary. >> > > >> > > To keep with our current naming scheme, how about 'aopop' or >> > > 'aopopulation'? >> > > >> > > Cheers, >> > > Karol >> > > >> > > On Sep 13 2013, Fedor Zhuravlev wrote: >> > > > Hi Adam and Karol, >> > > > >> > > > Frontier Molecular Orbitals: This is exactly what I was looking for >> > > folks, thanks! >> > > > >> > > > For the NPA analysis: I have included 3 log files from g09, Jaguar 7 >> > > and Turbomole 6.1, all containing the output from the NPA calculations (at >> > > the end of the file, they look very similar). The values of interest could >> > > be the Natural Charge and Natural Electron Population. Those would be nice >> > > to extract. >> > > > >> > > > Fedor >> > > > >> > > > From: Adam Tenderholt [mailto:ate...@gm...] >> > > > Sent: 12 September 2013 18:08 >> > > > To: Fedor Zhuravlev >> > > > Cc: Karol M. Langner; ccl...@li... >> > > > Subject: Re: [cclib-users] FMO properties >> > > > >> > > > Hi Fedor, >> > > > >> > > > Getting the energies of the HOMO and LUMO is pretty straight-forward: >> > > > >> > > > from cclib.parser import ccopen >> > > > >> > > > parser = ccopen(logfile) >> > > > data = parser.parser() >> > > > homo = data.homos[0] # assuming spin-restricted >> > > > lumo = homo + 1 >> > > > >> > > > homo_energy = data.moenergies[0][homo] >> > > > lumo_energy = data.moenergies[0][lumo] >> > > > >> > > > As far as parsing population analyses, Noel, Karol, and I will have to >> > > discuss the potential implementation details. Do you have a favorite QM >> > > package? >> > > > >> > > > Adam >> > > > >> > > > >> > > > On Thu, Sep 12, 2013 at 12:45 AM, Fedor Zhuravlev <fe...@dt...<mailto: >> > > fe...@dt...>> wrote: >> > > > Hi Karol and Adam, >> > > > >> > > > Thanks much for the prompt reply. I was looking for the energies of HOMO >> > > and LUMO. From there, using the Koopman's theorem one gets the ionization >> > > energy (HOMO=IE). (HOMO + LUMO)/2 is chemical potential (CP); LUMO-HOMO is >> > > chemical hardness (H), while CP^2/2H is Parr's electrophilicity (JACS, >> > > 1999, 121, 1922). 4 very useful, easily chemically interpretable chemical >> > > descriptors for the price of two frontier MO! Would be very nice to have! >> > > > >> > > > As for the natural population analysis the QM code would normally do it >> > > itself (at least Gaussian, Turbomole, and Jaguar), so it is just the >> > > question of proper parsing the data. Most people would normally be >> > > interested in the natural charge and natural population. >> > > > >> > > > That was my wish list for Christmas :) >> > > > >> > > > Fedor Zhuravlev >> > > > Senior Researcher >> > > > DTU Nutech >> > > > Technical University of Denmark >> > > > Center for Nuclear Technologies >> > > > Hevesy Laboratory >> > > > DTU Risø Campus >> > > > >> > > > Frederiksborgvej 399 Building 202 >> > > > Building 202 >> > > > 4000 Roskilde >> > > > fe...@dt...<mailto:fe...@dt...> >> > > > www.nutech.dtu.dk<http://www.nutech.dtu.dk> >> > > > >> > > > >> > > > >> > > > >> > > > >> > > > >> > > > -----Original Message----- >> > > > From: Karol M. Langner [mailto:kar...@gm...<mailto: >> > > kar...@gm...>] >> > > > Sent: 12 September 2013 00:31 >> > > > To: Adam Tenderholt >> > > > Cc: Fedor Zhuravlev; ccl...@li...<mailto: >> > > ccl...@li...> >> > > > Subject: Re: [cclib-users] FMO properties >> > > > >> > > > I assumed fragment energies in the fragment molecular orbital method in >> > > GAMESS-US... >> > > > >> > > > On Sep 11 2013, Adam Tenderholt wrote: >> > > > > Hi Fedor, >> > > > > >> > > > > I wanted to follow up on Karol's email. I'm not familiar with the >> > > > > details of NPA, and I briefly looked up the Reed/Weinstock/Weinhold >> > > > > paper. Adding it as a method in cclib is likely possible, but it would >> > > > > probably take several hours to fully understand the algorithm, and >> > > > > then implement and test it. This does interest me, although I'm not >> > > > > sure when I could get around to it. >> > > > > >> > > > > Also, what exactly do you mean by FMO energies? To me, FMO could stand >> > > > > for either fragment molecular orbitals or frontier molecular orbitals. >> > > > > Do you have a concrete example? >> > > > > >> > > > > Adam >> > > > > >> > > > > >> > > > > >> > > > > On Wed, Sep 11, 2013 at 8:01 AM, Karol M. Langner >> > > > > <kar...@gm...<mailto:kar...@gm...>>wrote: >> > > > > >> > > > > > On Sep 11 2013, Fedor Zhuravlev wrote: >> > > > > > > Dear cclib development team, >> > > > > > > >> > > > > > > Just had a first look at cclib and here's my questions: >> > > > > > > >> > > > > > > >> > > > > > > 1. Is there a single-liner to get FMO energies only? >> > > > > > > >> > > > > > > 2. What about natural population analysis? >> > > > > > > >> > > > > > > Thanks much, >> > > > > > > >> > > > > > > Fedor >> > > > > > >> > > > > > Hi Fedor, >> > > > > > >> > > > > > Nice to hear from you. I've used cclib to parse FMO output from >> > > > > > GAMESS-US in the past, but it involved a quick custom hack, which I >> > > > > > still may have somewhere I guess (I did this outside of version >> > > > > > control). However, this was some years ago, and I think the format >> > > > > > has changed significantly since then. >> > > > > > >> > > > > > As far as population analyses are concerned, we do not parse them, >> > > > > > although there was a discussion about that in the past. We have >> > > > > > implemented several population analysis algorithms alongside cclib, >> > > though: >> > > > > > http://cclib.sourceforge.net/wiki/index.php/Calculation_methods >> > > > > > ... although NAO is not among them. If you're thinking about the >> > > > > > original approach published by Reed/Weinstock/Weinhold, I think it's >> > > > > > a modification of Lowdin's orthoganolization, which we do. So you >> > > > > > could try to extend it to get what you want. Maybe Noel or Adam has >> > > more thoughts about this? >> > > > > > >> > > > > > If you want something specific parsed, you will generally need to >> > > > > > provide us with an example output file that we can use for testing. >> > > > > > >> > > > > > Cheers, >> > > > > > Karol >> > > > > > >> > > > > > -- >> > > > > > written by Karol M. Langner >> > > > > > Wed Sep 11 10:51:11 EDT 2013 >> > > > > > >> > > > > > >> > > > > > -------------------------------------------------------------------- >> > > > > > ---------- How ServiceNow helps IT people transform IT departments: >> > > > > > 1. Consolidate legacy IT systems to a single system of record for IT >> > > > > > 2. Standardize and globalize service processes across IT 3. >> > > > > > Implement zero-touch automation to replace manual, redundant tasks >> > > > > > http://pubads.g.doubleclick.net/gampad/clk?id=51271111&iu=/4140/ostg >> > > > > > .clktrk _______________________________________________ >> > > > > > cclib-users mailing list >> > > > > > ccl...@li...<mailto: >> > > ccl...@li...> >> > > > > > https://lists.sourceforge.net/lists/listinfo/cclib-users >> > > > > > >> > > > >> > > > -- >> > > > written by Karol M. Langner >> > > > Wed Sep 11 18:30:58 EDT 2013 >> > > > >> > > >> > > >> > > >> > > >> > > >> > > -- >> > > written by Karol M. Langner >> > > Wed Sep 18 18:11:05 EDT 2013 >> > > >> >> -- >> written by Karol M. Langner >> Wed Sep 18 22:29:09 EDT 2013 > > -- > written by Karol M. Langner > Thu Sep 19 00:14:43 EDT 2013 > > ------------------------------------------------------------------------------ > LIMITED TIME SALE - Full Year of Microsoft Training For Just $49.99! > 1,500+ hours of tutorials including VisualStudio 2012, Windows 8, SharePoint > 2013, SQL 2012, MVC 4, more. BEST VALUE: New Multi-Library Power Pack includes > Mobile, Cloud, Java, and UX Design. Lowest price ever! Ends 9/20/13. > http://pubads.g.doubleclick.net/gampad/clk?id=58041151&iu=/4140/ostg.clktrk > _______________________________________________ > cclib-devel mailing list > ccl...@li... > https://lists.sourceforge.net/lists/listinfo/cclib-devel |
