From owner-chemistry@ccl.net Thu Apr 7 10:51:00 2011 From: "Dr. Xiaofeng (Frank) Duan duanx.ctr++afrl.hpc.mil" To: CCL Subject: CCL: FW: smearing with NWChem Message-Id: <-44309-110407104218-6884-95JepoXYlM8UaMVAoewMww#,#server.ccl.net> X-Original-From: "Dr. Xiaofeng (Frank) Duan" Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=ISO-8859-1; format=flowed Date: Thu, 07 Apr 2011 10:42:08 -0400 MIME-Version: 1.0 Sent to CCL by: "Dr. Xiaofeng (Frank) Duan" [duanx.ctr+*+afrl.hpc.mil] Hi Huub and Niri, Thanks much for your comments and suggestions. In my case, I did see the energy increase with smearing and did not see much spin state change. Take one case as example: 1. With default smear value (0.001): .... sigma= .10D-02 eFermi= -0.31098 T*S= 0.00E+00 nel(1)= 53.00 nel(2)= 52.00 .... Total DFT energy = -756.386904455476 .... = 0.7658 (Exact = 0.7500) 2. Without smear: Total DFT energy = -756.388015008368 ..... = 0.7666 (Exact = 0.7500) In above 1, is eFermi= -0.31098 the correction to the total energy in unit au? Anyway, I will take your suggestions to re-run single point calculations with the largest smear value and multiplicity preservation for all the structures. One more question, I also did vibrational frequencies to get thermal corrections. Do you think I need re-do frequencies with the same smear value to get consistent results, or it is OK to keep these thermal corrections? Thanks, Frank On 04/06/11 19:32, Van Dam, Hubertus J HubertusJJ.vanDam. ~~ .pnl.gov wrote: > > Sent to CCL by: "Van Dam, Hubertus J" [HubertusJJ.vanDam[-]pnl.gov] > Hi Frank, > > In addition to my earlier post I am forwarding some additional comments from Niri Govind. Please have a look below. > > Best wishes, > > Huub > > -----Original Message----- >> From: Govind, Niri > Sent: Wednesday, April 06, 2011 4:25 PM > To: Van Dam, Hubertus J; CCL Subscribers > Cc: nwchem-users^emsl.pnl.gov > Subject: RE: smearing with NWChem > > Hi Frank, > > Try the following if you want to preserve the multiplicity with smear. > > dft > mult 3 > ... > ... > end > set dft:spinset t > task dft > > Best regards, > -Niri > > > > -----Original Message----- >> From: owner-nwchem-users^emsl.pnl.gov [mailto:owner-nwchem-users^emsl.pnl.gov] On Behalf Of Van Dam, Hubertus J > Sent: Wednesday, April 06, 2011 4:08 PM > To: CCL Subscribers > Cc: nwchem-users^emsl.pnl.gov > Subject: [NWCHEM] RE: smearing with NWChem > > Hi Frank, > > This is an interesting question. There are a few specific points here: > 1. The energy that is associated with the smearing is normally supposed to be negative, so the total energies with smearing should be lower than the ones without smearing. > 2. The smearing functionality may change the spin state of your system. As you are working on transition metal complexes I am guessing you might have open shell systems. I can provide some special directives to control this if needed. > 3. The code will actually print the energy that it adds on as a result of the smearing. Search for "eFermi=" in the output. (see also 4.) > 4. However, if you subtract the energy from 3. you will not get the energy you are looking for. The problem is that the DFT energy for a system with fractionally occupied orbitals is higher than the energy of the same system with integer occupations. So the eFermi term that is added on corrects for that difference and adds a bit more on top of that. > 5. Concluding it seems to me that your best bet is to run all calculations with the same smearing parameter for the final energies and compare the resulting energies. > I am sorry if this is not the answer you were looking for. If you want any more specific information please feel free to contact me. > > Best wishes, > > Huub > ________________________________________ >> From: owner-chemistry+hubertus.vandam==pnl.gov^ccl.net [owner-chemistry+hubertus.vandam==pnl.gov^ccl.net] On Behalf Of Xiaofeng Frank Duan duanx:_:asc.hpc.mil [owner-chemistry^ccl.net] > Sent: Wednesday, April 06, 2011 1:12 PM > To: Van Dam, Hubertus J > Subject: CCL: smearing with NWChem > > Sent to CCL by: "Xiaofeng Frank Duan" [duanx^asc.hpc.mil] > Hello, > > I optimized some transitional metal compounds with DFT in NWChem. For some compounds, I had to use "smear" to get SCF converged even though I decreased "smear" values gradually to the minimum. Since 'smear' adds an additional energy term, the total energy is higher than that of the "zero smear" for the same structure. > My question is: How to correct or scale the "smeared" total energy to the "zero smeared" one so that I can compare energies among different compounds. > > Thanks, > Frankhttp://www.ccl.net/cgi-bin/ccl/send_ccl_messagehttp://www.ccl.net/chemistry/sub_unsub.shtmlhttp://www.ccl.net/spammers.txt> > From owner-chemistry@ccl.net Thu Apr 7 11:26:01 2011 From: "lavanya chebrolu lavanyadevi.kada^^^gmail.com" To: CCL Subject: CCL: about two photon absorption calculation in tddft in g09 Message-Id: <-44310-110407104002-5735-XaFf6Z8WQoRk3hqJkLYt4w#%#server.ccl.net> X-Original-From: "lavanya chebrolu" Date: Thu, 7 Apr 2011 10:40:00 -0400 Sent to CCL by: "lavanya chebrolu" [lavanyadevi.kada a gmail.com] thanq you in advance hai i am lavanya doing my phd in iict, hyderabad, as I am working on two photon absorption, I would like to know how I can get transition moments for the ground and excited states in TD-DFT. If anybody help me in this regard I am very thankfull to them. thanking you, C.Lavanyadevi c/0 Dr. V.J.Rao iict hyderabad mail.Id. lavanyadevi.kada : gmail.com From owner-chemistry@ccl.net Thu Apr 7 13:33:00 2011 From: "Phil Hasnip phil.hasnip]^[york.ac.uk" To: CCL Subject: CCL:G: smearing with NWChem Message-Id: <-44311-110407133152-19628-MKqyVDvBb2Ze/bMvKgkMDw++server.ccl.net> X-Original-From: Phil Hasnip Content-Type: text/plain; format=flowed; charset=ISO-8859-1 Date: 07 Apr 2011 18:31:39 +0100 Mime-Version: 1.0 Sent to CCL by: Phil Hasnip [phil.hasnip{=}york.ac.uk] Dear Frank, > Since 'smear' adds an additional energy term, the total energy is higher > than that of the "zero smear" for the same structure. My question is: How > to correct or scale the "smeared" total energy to the "zero smeared" one I don't use NWChem, but the basic answer is that it depends on the smearing scheme. With Fermi level smearing the free energy is E-TS, where T is the smearing "temperature" and S is the electronic entropy due to the fractional occupancies. As the smearing is increased, E increases and E-TS decreases. How you estimate what the "zero smearing" energy would be depends on the smearing scheme you use; if you use Gaussians, for example, then the "zero smearing" energy is E-0.5TS to 1st order in the smearing. In Castep this estimate of the zero temperature energy is printed out along with E and E-TS -- perhaps NWChem does the same somewhere? Even if it doesn't, all you need are E and E-TS and you can correct it yourself. There are other smearing schemes that allow a better correction (i.e. higher order) back to the "zero smearing" result, and these are known as "cold smearing" schemes -- see for example the work of Methfessel and Paxton (Phys. Rev. B 40, 3616-3621 (1989)). However you should bear in mind that these corrections only give you an estimate of the energy, and do not correct other quantities that may be important such as energy derivatives. Yours, Phil Hasnip ------------------------------------------------------- Dr Phil Hasnip Email: pjh503]-[york.ac.uk Dept of Physics University of York Tel: +44 (0)1904 434624 York YO10 5DD From owner-chemistry@ccl.net Thu Apr 7 18:21:01 2011 From: "Rajarshi Guha rajarshi.guha:+:gmail.com" To: CCL Subject: CCL: =?windows-1252?Q?Last_Call_for_Papers=3A_High_Content_Screening=3A_Explor?= =?windows-1252?Q?ing_Relationships_Between_Small_Molecules_and_Phenotypic_Res?= =?windows-1252?Q?ults=2E_242nd_ACS_National_Meeting_Denver=2C_Aug_28_=96_Sept_1=2C_?= =?windows-1252?Q?2011?= Message-Id: <-44312-110407102831-15498-okD5Yr3S8UJDuB29pcPWNQ * server.ccl.net> X-Original-From: Rajarshi Guha Content-Transfer-Encoding: 8bit Content-Type: text/plain; charset=windows-1252 Date: Thu, 7 Apr 2011 10:28:23 -0400 MIME-Version: 1.0 Sent to CCL by: Rajarshi Guha [rajarshi.guha(~)gmail.com] High Content Screening: Exploring Relationships Between Small Molecules and Phenotypic Results. 242nd ACS National Meeting Denver, Aug 28 – Sept 1, 2011 CINF Division Dear Colleagues, we are organizing an ACS symposium, focusing on the use of High Content Screening (HCS) for small molecule applications. High content screens, while resource intensive, are capable of providing a detailed view of the phenotypic effects of small molecules. Traditional reporter based screens are characterized by a one-dimensional signal. In contrast, high content screens generate rich, multi-dimensional datasets that allow for wide-ranging and in-depth analysis of various aspects of chemical biology including mechanisms of action, target identification and so on. Recent developments in high-throughput HCS pose significant challenges throughout the screening pipeline ranging from assay design and miniaturization to data management and analysis. Underlying all of this is the desire to connect chemical structure to phenotypic effects. We invite you to submit contributions highlighting novel work and new developments in High Content Screening (HCS), High Content Analysis (HCA), and data exploration as it relates to the field of small molecules. Topics of interest include but are not limited to * Compound & in silico screening for drug discovery * Compound profiling by high content analysis * Chemistry & probes in imaging * Lead discovery strategies – one size fits all or horses for courses? * Application of HCA in discovering toxicology screening strategies * Novel data mining approaches for HCS data that link phenotypes to chemical structures * Software & informatics for HCS data management and integration In addition to these topics special consideration will be given to contributions that present contributions in in-silico exploration based on HCS data. We would also like to point out that sponsorship opportunities are available. The deadline for abstract submissions is April 8, 2011. All abstracts should be submitted via PACS at http://abstracts.acs.org. If you have any questions feel free to contact Tim or myself. Tim Moran Accelrys tmoran||accelrys.com +1 858 799 5609 Rajarshi Guha NIH Chemical Genomics Center guhar||mail.nih.gov +1 814 404 5449 -- Rajarshi Guha NIH Chemical Genomics Center From owner-chemistry@ccl.net Thu Apr 7 18:56:00 2011 From: "Van Dam, Hubertus J HubertusJJ.vanDam!^!pnl.gov" To: CCL Subject: CCL: FW: smearing with NWChem Message-Id: <-44313-110407183640-29170-AwKcxH3OSp4XWX7OOMCMCg-*-server.ccl.net> X-Original-From: "Van Dam, Hubertus J" Content-Language: en-US Content-Transfer-Encoding: 8bit Content-Type: text/plain; charset="us-ascii" Date: Thu, 7 Apr 2011 15:36:28 -0700 MIME-Version: 1.0 Sent to CCL by: "Van Dam, Hubertus J" [HubertusJJ.vanDam{}pnl.gov] Hi Frank, I agree with you on the spin state. The S2 value does not seem to be affected much by the smearing so that is probably OK. The units of the eFermi energy are indeed Hartrees (i.e. atomic units). With respect to rerunning any calculations did you also see Phil Hasnip's comments on "cold smearing" total energy estimates (Phil thanks for those). You might be able to use those to estimate the zero temperature total energies (rather than rerunning your single point energies). If you have missed Phil's message then please ping me directly and I will forward the message to you. The one thing that does worry me is that the energy of the calculations with smearing come out higher than the ones without. At present I don't understand why that happens which makes me feel uneasy about this (which is why I am not sure about rerunning calculations either). However, I have received a test case from Adam Pelzer (thanks for providing that Adam) which I am running at the moment and I hope to have a handle on this soon. As far as the zero-point corrections are concerned I simply don't know how much of a difference that will make. If the temperature is small you would think it should not make a big difference but I don't know. So I would recommend just trying it out on a single molecule and see what you find. Best wishes, Huub -----Original Message----- > From: owner-chemistry+hubertus.vandam==pnl.gov^-^ccl.net [mailto:owner-chemistry+hubertus.vandam==pnl.gov^-^ccl.net] On Behalf Of Dr. Xiaofeng (Frank) Duan duanx.ctr++afrl.hpc.mil Sent: Thursday, April 07, 2011 7:42 AM To: Van Dam, Hubertus J Subject: CCL: FW: smearing with NWChem Sent to CCL by: "Dr. Xiaofeng (Frank) Duan" [duanx.ctr+*+afrl.hpc.mil] Hi Huub and Niri, Thanks much for your comments and suggestions. In my case, I did see the energy increase with smearing and did not see much spin state change. Take one case as example: 1. With default smear value (0.001): .... sigma= .10D-02 eFermi= -0.31098 T*S= 0.00E+00 nel(1)= 53.00 nel(2)= 52.00 .... Total DFT energy = -756.386904455476 .... = 0.7658 (Exact = 0.7500) 2. Without smear: Total DFT energy = -756.388015008368 ..... = 0.7666 (Exact = 0.7500) In above 1, is eFermi= -0.31098 the correction to the total energy in unit au? Anyway, I will take your suggestions to re-run single point calculations with the largest smear value and multiplicity preservation for all the structures. One more question, I also did vibrational frequencies to get thermal corrections. Do you think I need re-do frequencies with the same smear value to get consistent results, or it is OK to keep these thermal corrections? Thanks, Frank On 04/06/11 19:32, Van Dam, Hubertus J HubertusJJ.vanDam.[].pnl.gov wrote: > > Sent to CCL by: "Van Dam, Hubertus J" [HubertusJJ.vanDam[-]pnl.gov] > Hi Frank, > > In addition to my earlier post I am forwarding some additional comments from Niri Govind. Please have a look below. > > Best wishes, > > Huub > > -----Original Message----- >> From: Govind, Niri > Sent: Wednesday, April 06, 2011 4:25 PM > To: Van Dam, Hubertus J; CCL Subscribers > Cc: nwchem-users^emsl.pnl.gov > Subject: RE: smearing with NWChem > > Hi Frank, > > Try the following if you want to preserve the multiplicity with smear. > > dft > mult 3 > ... > ... > end > set dft:spinset t > task dft > > Best regards, > -Niri > > > > -----Original Message----- >> From: owner-nwchem-users^emsl.pnl.gov [mailto:owner-nwchem-users^emsl.pnl.gov] On Behalf Of Van Dam, Hubertus J > Sent: Wednesday, April 06, 2011 4:08 PM > To: CCL Subscribers > Cc: nwchem-users^emsl.pnl.gov > Subject: [NWCHEM] RE: smearing with NWChem > > Hi Frank, > > This is an interesting question. There are a few specific points here: > 1. The energy that is associated with the smearing is normally supposed to be negative, so the total energies with smearing should be lower than the ones without smearing. > 2. The smearing functionality may change the spin state of your system. As you are working on transition metal complexes I am guessing you might have open shell systems. I can provide some special directives to control this if needed. > 3. The code will actually print the energy that it adds on as a result of the smearing. Search for "eFermi=" in the output. (see also 4.) > 4. However, if you subtract the energy from 3. you will not get the energy you are looking for. The problem is that the DFT energy for a system with fractionally occupied orbitals is higher than the energy of the same system with integer occupations. So the eFermi term that is added on corrects for that difference and adds a bit more on top of that. > 5. Concluding it seems to me that your best bet is to run all calculations with the same smearing parameter for the final energies and compare the resulting energies. > I am sorry if this is not the answer you were looking for. If you want any more specific information please feel free to contact me. > > Best wishes, > > Huub > ________________________________________ >> From: owner-chemistry+hubertus.vandam==pnl.gov^ccl.net [owner-chemistry+hubertus.vandam==pnl.gov^ccl.net] On Behalf Of Xiaofeng Frank Duan duanx:_:asc.hpc.mil [owner-chemistry^ccl.net] > Sent: Wednesday, April 06, 2011 1:12 PM > To: Van Dam, Hubertus J > Subject: CCL: smearing with NWChem > > Sent to CCL by: "Xiaofeng Frank Duan" [duanx^asc.hpc.mil] > Hello, > > I optimized some transitional metal compounds with DFT in NWChem. For some compounds, I had to use "smear" to get SCF converged even though I decreased "smear" values gradually to the minimum. Since 'smear' adds an additional energy term, the total energy is higher than that of the "zero smear" for the same structure. > My question is: How to correct or scale the "smeared" total energy to the "zero smeared" one so that I can compare energies among different compounds. > > Thanks, > Frankhttp://www.ccl.net/cgi-bin/ccl/send_ccl_messagehttp://www.ccl.net/chemistry/sub_unsub.shtmlhttp://www.ccl.net/spammers.txthttp://www.ccl.net/cgi-bin/ccl/send_ccl_messagehttp://www.ccl.net/chemistry/sub_unsub.shtmlhttp://www.ccl.net/spammers.txt From owner-chemistry@ccl.net Thu Apr 7 19:30:00 2011 From: "Van Dam, Hubertus J HubertusJJ.vanDam!=!pnl.gov" To: CCL Subject: CCL: CORRECTION: smearing with NWChem Message-Id: <-44314-110407185904-21173-2Dd60tiueh6OGtTv/TBOtA- -server.ccl.net> X-Original-From: "Van Dam, Hubertus J" Content-Language: en-US Content-Transfer-Encoding: 8bit Content-Type: text/plain; charset="us-ascii" Date: Thu, 7 Apr 2011 15:58:55 -0700 MIME-Version: 1.0 Sent to CCL by: "Van Dam, Hubertus J" [HubertusJJ.vanDam|a|pnl.gov] Hi Frank, I am sorry to add to the confusion but I have just realized that I interpreted the eFermi energy wrong. The eFermi energy is not the Mermin term but the actual Fermi energy. The Mermin term can be found in the output by looking for "T*S=". Strangely enough this seems to come out as 0.0e+0 most of the time. This might be right for low temperatures but then you are essentially doing a calculation without smearing. I apologize once again for the confusion. Best wishes, Huub