[Wien] Bad formation energies for the charged vacancies
I've started some tests after the first query and it seems we might miss a term in the total energy. I created a clmsum-file (density) which is constant and is normalized to one and put this into a cell with a single H nucleus. So it refers to the test case of a H+ ion in a lattice, where I do not add a constant background, but put the "background charge into case.clmsum. When one switches off the XC-terms, the resulting E-tot contains the integral (rho *V-coul) and since rho is constant (equal to Q/volume), we get the average potential in the unit cell (not only the interstital, where it is zero anyway) multiplied by the constant rho). This term is missing when I put a clmsum file with rho=zero, but add a "background charge" by case.inm, while the resulting potentials are identical for the two methods. However, for a charged bulk system there is still a big problem, because V-coul is determined only up to a constant and is shifted arbitrarily such that the potential in the interstital is zero. In "neutral" calculations such a shift does not matter, since it will be canceled by the sum of eigenvalues, but when adding the constant background it matters. Thus, this correction term depends on RMT ? At the moment I'm not sure how I should continue. I think in other codes such a correction is added, but as mentioned, I guess the correction depends on the arbitrary choice of V-zero. Laurence Marks schrieb: > Please see the next email on the > list:http://zeus.theochem.tuwien.ac.at/pipermail/wien/2007-January/008713.html > I think this is right and you take V0 from case.output0 (it is printedthere). > You should do an empty cell test (no electrons) to verify thisand the units > of V0, perhaps also looking at the code itself -- andremember to check the > limit as the distance between atoms gets large. > > On Wed, Feb 24, 2010 at 11:05 AM, Yurko Natanzon > wrote:> Dear Wien2k users and developers,> I'd like to refresh the discussion > about the total energies of the> charged cells which took place three years > ago:> > http://zeus.theochem.tuwien.ac.at/pipermail/wien/2007-January/008711.html>> > I'm trying to calculate the formation energy of the Hydrogen vacancy> in +/-1 > charge states and find that the results are bad (much differ> from the > literature) although the formation energy of the neutral> hydrogen vacancy is > good. So my question arises if we can trust the> values of the total energies > for the charged cells in the recent> version of Wien2k?>> To investigate this > issue further I have performed the following> tests: I've done the > calculations of the total energy of Mg, MgH2 and> GaN for three cases: > neutral cell, cell with one electron removed (+1> charge) and a cell with an > electron added (-1 charge). The results> were compared with the same calculati on with another plane-wave code> and are the following:> -> hcp Mg:> Wien2k:> E(+1)-E(0) = 0.245 Ry> E(-1)-E(0) = -0.199 Ry>> Plane-Wave code:> E(+1)-E(0) = -0.226 Ry> E(-1)-E(0) = 0.281 Ry>> bcc MgH2> Wien2k:> E(+1)-E(0) = 0.277 Ry> E(-1)-E(0) = 0.085 Ry>> Plane-Wave code:> E(+1)-E(0) = 0.024 Ry> E(-1)-E(0) = 0.326 Ry>> fcc GaN> Wien2k:> E(+1)-E(0) = 1.12 Ry> E(-1)-E(0) = -0.717 Ry>> Plane-Wave code:> E(+1)-E(0) = -0.151 Ry> E(-1)-E(0) = 0.443 Ry> ->> In wien2k the charged cell was created by changing the number of> electrons in case.in2 and adding the corresponding background charge> in case.inm. One can observe, that the energies have the same order of> magnitude, but the sequence of energies E(+1), E(0) and E(1) is> inverse. It seems, that the system with +1 charge (electron removed)> behaves like the system with -1 charge in the Plane-Wave code. Of> course, the results of tests are no t physical, because no supercell> was used and no geometric relaxation was performed (however, it is not> needed for Mg), but if one tries to do all the supercell and> relaxation stuff and tries to calculate the defect formation energy,> the result will be the same.>> I'd be grateful if you comment on this and suggest any corrections> which should be provided to the total energies for the charged cells.>> with kind regards,> Yurko>>> --> Yurko (aka Yuriy, Iurii, Jurij etc) Natanzon> PhD student> Department for Structural Research (NZ31)> Henryk Niewodnicza?ski Institute of Nuclear Physics> Polish Academy of Sciences> ul. Radzikowskiego 152,> 31-342 Krakow, Poland> E-mail: Yurii.Natanzon at ifj.edu.pl, yurko.natanzon at gmail.com> ___> Wien mailing list> Wien at zeus.theochem.tuwien.ac.at> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien> > > > -- Laurence MarksDepartment of Materials Science and EngineeringMSE Rm 2036 > Cook Hall2220 N Campus DriveNorthwestern UniversityEvanston, IL 60208, > USATel: (847) 491-3996 Fax: (847) 491-7820email: L-marks at northwe
[Wien] volume optimization and force minimization
Dear Laurence Marks Sir, Thank you very much for your reply. What we can do then.we optimize both volume and coordinates including spin polarization and GGA+U and take this as the optimized structure as far as the theory is concerned. Then we take that structure and run one scf including spin-orbit coupling too. If the energy is less than the minimum energy obtained in volume optimization we can go with spin-orbit coupling or otherwise not. But it may happen that the energy is less and the volume is different (I don't know how much effect spin-orbit coupling will have on the volume!!...atleast we have experience that spin polarization definitely effects volume) then which volume we should accept as our optimized volume? On Wed, Feb 24, 2010 at 7:26 PM, Laurence Marks wrote: > Unless something has changed, forces are not implemented for > spin-orbit correctly so you cannot do a force minimization with them > (unless you do some tricks, i.e. minimize the enegy by hand not using > the code). > > For spin/U the question is how large a difference does this make to > the electron density -- if it is large then the positions/optimal > volume will be very different. > > But...do not depair! If spin-polarized and U is a much better > description of the physics and you do not have multiple magnetic > states, convergence of a LDA+U (or GGA+U) calculation can be much > faster than a LDA calculation which does not describe the physics > well. This follows from the physics/math of mixing. > > N.B., this is also why using a small mixing term in a > Broyden/Multisecant method does not do what most people think it does, > conventional wisdom is incorrect or at most correct for Pratt mixing. > (Also, large mixing terms do not do what people think they do!) In > general this not the way to solve a badly convergent problem, the > correct (only) solution is to improve the physical model. > > 2010/2/24 shamik chakrabarti : > > Dear Wien2k users, > > I have a question regarding volume > optimization > > and force minimization. > > If I do volume optimization and force minimization for non magnetic > > calculation and then use that optimized volume and structural coordinates > as > > input for the calculation of volume optimization and force minimization > > including spin polarization, spin orbit coupling term and LDA+U term, > then > > whether I get the correct optimized values for both volume and > coordinates > > of the structure which really need those corrections? > > If I want to do volume and coordinates optimization including those > > correction terms from the begining then it will take a very long time. > But > > if we do the optimization using non magnetic calculation to reach closer > to > > the actual energy and force minima and then put those optimized structure > > for non magnetic calculation for the optimization of the structure having > > those corrections it may take lesser time to reach the actual minima. But > > whether we really get actual minima in this way? > > The above two questions actually are the same questions. I am very sorry > for > > that reapeating. Actually I want to make clear my intentions to the > wien2k > > users. > > Thanks in advance. > > regards, > > Shamik Chakrabarti > > ___ > > Wien mailing list > > Wien at zeus.theochem.tuwien.ac.at > > http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien > > > > > > > > -- > Laurence Marks > Department of Materials Science and Engineering > MSE Rm 2036 Cook Hall > 2220 N Campus Drive > Northwestern University > Evanston, IL 60208, USA > Tel: (847) 491-3996 Fax: (847) 491-7820 > email: L-marks at northwestern dot edu > Web: www.numis.northwestern.edu > Chair, Commission on Electron Crystallography of IUCR > www.numis.northwestern.edu/ > Electron crystallography is the branch of science that uses electron > scattering and imaging to study the structure of matter. > ___ > Wien mailing list > Wien at zeus.theochem.tuwien.ac.at > http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien > -- next part -- An HTML attachment was scrubbed... URL: <http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20100224/8c7beb20/attachment.htm>
[Wien] volume optimization and force minimization
Dear Wien2k users, I have a question regarding volume optimization and force minimization. If I do volume optimization and force minimization for non magnetic calculation and then use that optimized volume and structural coordinates as input for the calculation of volume optimization and force minimization including spin polarization, spin orbit coupling term and LDA+U term, then whether I get the correct optimized values for both volume and coordinates of the structure which really need those corrections? If I want to do volume and coordinates optimization including those correction terms from the begining then it will take a very long time. But if we do the optimization using non magnetic calculation to reach closer to the actual energy and force minima and then put those optimized structure for non magnetic calculation for the optimization of the structure having those corrections it may take lesser time to reach the actual minima. But whether we really get actual minima in this way? The above two questions actually are the same questions. I am very sorry for that reapeating. Actually I want to make clear my intentions to the wien2k users. Thanks in advance. regards, Shamik Chakrabarti -- next part -- An HTML attachment was scrubbed... URL: <http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20100224/768d5d5f/attachment.htm>
[Wien] Bad formation energies for the charged vacancies
Dear Wien2k users and developers, I'd like to refresh the discussion about the total energies of the charged cells which took place three years ago: http://zeus.theochem.tuwien.ac.at/pipermail/wien/2007-January/008711.html I'm trying to calculate the formation energy of the Hydrogen vacancy in +/-1 charge states and find that the results are bad (much differ from the literature) although the formation energy of the neutral hydrogen vacancy is good. So my question arises if we can trust the values of the total energies for the charged cells in the recent version of Wien2k? To investigate this issue further I have performed the following tests: I've done the calculations of the total energy of Mg, MgH2 and GaN for three cases: neutral cell, cell with one electron removed (+1 charge) and a cell with an electron added (-1 charge). The results were compared with the same calculation with another plane-wave code and are the following: - hcp Mg: Wien2k: E(+1)-E(0) = 0.245 Ry E(-1)-E(0) = -0.199 Ry Plane-Wave code: E(+1)-E(0) = -0.226 Ry E(-1)-E(0) = 0.281 Ry bcc MgH2 Wien2k: E(+1)-E(0) = 0.277 Ry E(-1)-E(0) = 0.085 Ry Plane-Wave code: E(+1)-E(0) = 0.024 Ry E(-1)-E(0) = 0.326 Ry fcc GaN Wien2k: E(+1)-E(0) = 1.12 Ry E(-1)-E(0) = -0.717 Ry Plane-Wave code: E(+1)-E(0) = -0.151 Ry E(-1)-E(0) = 0.443 Ry - In wien2k the charged cell was created by changing the number of electrons in case.in2 and adding the corresponding background charge in case.inm. One can observe, that the energies have the same order of magnitude, but the sequence of energies E(+1), E(0) and E(1) is inverse. It seems, that the system with +1 charge (electron removed) behaves like the system with -1 charge in the Plane-Wave code. Of course, the results of tests are not physical, because no supercell was used and no geometric relaxation was performed (however, it is not needed for Mg), but if one tries to do all the supercell and relaxation stuff and tries to calculate the defect formation energy, the result will be the same. I'd be grateful if you comment on this and suggest any corrections which should be provided to the total energies for the charged cells. with kind regards, Yurko -- Yurko (aka Yuriy, Iurii, Jurij etc) Natanzon PhD student Department for Structural Research (NZ31) Henryk Niewodnicza?ski Institute of Nuclear Physics Polish Academy of Sciences ul. Radzikowskiego 152, 31-342 Krakow, Poland E-mail: Yurii.Natanzon at ifj.edu.pl, yurko.natanzon at gmail.com
[Wien] parallel wien2k
Dear Laurence and All, Thank you very much for the information. It has been very helpful in clarifying some of the issues. Based on your input, I was able to prepare the .machines in the correct format, I believe: .machines: # lapw0: nx59:2 nx58:2 1:nx59 1:nx59 1:nx58 1:nx58 granularity:1 extrafine:1 and .machine0 nx59 nx59 nx58 nx58 However, I still got the same problem in TiC.vns, which presumably resulted in the crash in lapw1para. Are there any places in the output files that I can look for clues of problem? For the same calculation, I can run the lapw0 in serial mode and lapw1 in k-point parallel mode successfully. Thanks in advance, Zhiyong - Original Message - From: "Laurence Marks" To: "A Mailing list for WIEN2k users" Sent: Tuesday, February 23, 2010 4:55:19 AM GMT -08:00 US/Canada Pacific Subject: Re: [Wien] parallel wien2k Several points: 1) You only use "-fc X" for a structure with variable atomic positions, and the TiC example has none so it will report that there are no forces (but this should not stop the calculation). 2) The "NaN" in your case.vns file means that something went wrong in the lapw0 call, which is why lapw1 is crashing. It is safer to delete the case.vns file. 3) Are you using single core CPU's or multicore? The normal format for a parallel lapw0 call (using mpi) is lapw0: nx1:2 nx62:2 -- please note the space after the ":", it often matters To do this you have to have mpi installed and have compiled lapw0_mpi. If you do not have it you can use lapw0: nx1 This will run a serial lapw0 on nx1 4) All the above assumes that you have local control of what nodes you can use rather than this being controlled by a queuing system such as pbs. If you are using pbs or similar then you have to have a script to generate the .machines file since you do not know what machines to use (unless you are running interactively). 5) The script you have will run serial (i.e. not mpi) lapw1, 2 k-vectors on nx1 and 2 on nx62. If you want to have these run using the parallel versions (i.e. lapw1_mpi) you would need to use 1:nx1:2 1:nx62:2 (Note no space after the ":"). Whether it is faster to run with 2 processors on nx1, as against 2 different k-points will depend upon your cpu's. For a simple calculation such as TiC it will be hard to see much difference, but this can matter for larger ones. Be aware that if (for instance) you had 4 processors on nx1 it may be a bad idea to use 1:nx1:2 1:nx1:2 because some variants of mpi will launch both lapw1_mpi jobs on the same cores (CPU_AFFINITY is often the relevant flag, but this varies with mpi flavor). 2010/2/22 zyzhang : > Dear All, > > > > I am trying to test wien2k in parallel mode and I got into some problem. I > am using > > > > run_lapw -p -i 40 -fc 0.001 ?I > > > > If I use a number of 0.001 for the option fc above, I got the following > error: > > > > Force-convergence not possible. Forces not present. > > > > If I do not use a number for the ?fc option, but use ?run_lapw -p -i 40 -fc > ?I? instead > > > > Then lapw0 finishes without a problem but the program doesn?t branch to > lapw1. An error message is generated when doing the test > > > > ?if ($fcut == "0") goto lapw1 > > > > I was able to do ?run_lapw -p -i 40 ?I?, without the ?-fc? option at all and > was able to finish ?lapw0 ?p? and then start ?lapw1 ?p? but got into the > following error: > > > > error: command?? /home/zzhang/wien2k/lapw1para lapw1.def?? failed > > > > Does anybody have similar problems and know how to fix this? > > > > It does the following: > > > > running LAPW1 in parallel mode (using .machines) > > > > and the .machines file is as follows: > > > > # > > lapw0:nx1? nx1? nx62? nx62 > > lapw1:nx1? nx1? nx62? nx62 > > lapw2:nx1? nx1? nx62? nx62 > > 1:nx1 > > 1:nx1 > > 1:nx62 > > 1:nx62 > > granularity:1 > > extrafine:1 > > > > Thanks, > > Zhiyong > > > > ___ > Wien mailing list > Wien at zeus.theochem.tuwien.ac.at > http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien > > -- Laurence Marks Department of Materials Science and Engineering MSE Rm 2036 Cook Hall 2220 N Campus Drive Northwestern University Evanston, IL 60208, USA Tel: (847) 491-3996 Fax: (847) 491-7820 email: L-marks at northwestern dot edu Web: www.numis.northwestern.edu Chair, Commission on Electron Crystallography of IUCR www.numis.northwestern.edu/ Electron crystallography is the branch of science that uses electron scattering and imaging to study the structure of matter. ___ Wien mailing list Wien at zeus.theochem.tuwien.ac.at http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
[Wien] Bad formation energies for the charged vacancies
Dear Peter, Is the question regarding the computation of total energy per unit cell in an infinite crystal with non-neutral unit cells? If so, then the total energy diverges -- and so is not well-defined. (So neutralizing backgrounds must be added in such cases to obtain meaningful results, etc.) John On Feb 24, 2010, at 1:18 PM, Peter Blaha wrote: > I've started some tests after the first query and it seems we might > miss a term in the total > energy. > > I created a clmsum-file (density) which is constant and is > normalized to one and put this into > a cell with a single H nucleus. > So it refers to the test case of a H+ ion in a lattice, where I do > not add a constant background, but > put the "background charge into case.clmsum. > > When one switches off the XC-terms, the resulting E-tot contains the > integral (rho *V-coul) and since > rho is constant (equal to Q/volume), we get the average potential in > the unit cell (not only the > interstital, where it is zero anyway) multiplied by the constant rho). > > This term is missing when I put a clmsum file with rho=zero, but add > a "background charge" > by case.inm, while the resulting potentials are identical for the > two methods. > > However, for a charged bulk system there is still a big problem, > because V-coul is determined > only up to a constant and is shifted arbitrarily such that the > potential in the interstital is zero. > In "neutral" calculations such a shift does not matter, since it > will be canceled by the sum of > eigenvalues, but when adding the constant background it matters. > > Thus, this correction term depends on RMT ? > > At the moment I'm not sure how I should continue. I think in other > codes such a correction is > added, but as mentioned, I guess the correction depends on the > arbitrary choice of V-zero. > > > Laurence Marks schrieb: >> Please see the next email on the list:http:// >> *zeus.theochem.tuwien.ac.at/pipermail/wien/2007-January/008713.html >> I think this is right and you take V0 from case.output0 (it is >> printedthere). You should do an empty cell test (no electrons) to >> verify thisand the units of V0, perhaps also looking at the code >> itself -- andremember to check the limit as the distance between >> atoms gets large. >> On Wed, Feb 24, 2010 at 11:05 AM, Yurko Natanzon> > wrote:> Dear Wien2k users and developers,> I'd like to refresh >> the discussion about the total energies of the> charged cells which >> took place three years ago:> http://*zeus.theochem.tuwien.ac.at/ >> pipermail/wien/2007-January/008711.html>> I'm trying to calculate >> the formation energy of the Hydrogen vacancy> in +/-1 charge states >> and find that the results are bad (much differ> from the >> literature) although the formation energy of the neutral> hydrogen >> vacancy is good. So my question arises if we can trust the> values >> of the total energies for the charged cells in the recent> version >> of Wien2k?>> To investigate this issue further I have performed the >> following> tests: I've done the calculations of the total energy of >> Mg, MgH2 and> GaN for three cases: neutral cell, cell with one >> electron removed (+1> charge) and a cell with an electron added (-1 >> charge). The results> were compared with the same calculati > on with another plane-wave code> and are the following:> > -> hcp Mg:> Wien2k:> E(+1)- > E(0) = 0.245 Ry> E(-1)-E(0) = -0.199 Ry>> Plane-Wave code:> E(+1)- > E(0) = -0.226 Ry> E(-1)-E(0) = 0.281 Ry>> bcc MgH2> Wien2k:> E(+1)- > E(0) = 0.277 Ry> E(-1)-E(0) = 0.085 Ry>> Plane-Wave code:> E(+1)- > E(0) = 0.024 Ry> E(-1)-E(0) = 0.326 Ry>> fcc GaN> Wien2k:> E(+1)- > E(0) = 1.12 Ry> E(-1)-E(0) = -0.717 Ry>> Plane-Wave code:> E(+1)- > E(0) = -0.151 Ry> E(-1)-E(0) = 0.443 Ry> > ->> In wien2k the > charged cell was created by changing the number of> electrons in > case.in2 and adding the corresponding background charge> in > case.inm. One can observe, that the energies have the same order of> > magnitude, but the sequence of energies E(+1), E(0) and E(1) is> > inverse. It seems, that the system with +1 charge (electron > removed)> behaves like the system with -1 charge in the Plane-Wave > code. Of> course, the results of tests are no > t physical, because no supercell> was used and no geometric > relaxation was performed (however, it is not> needed for Mg), but if > one tries to do all the supercell and> relaxation stuff and tries to > calculate the defect formation energy,> the result will be the > same.>> I'd be grateful if you comment on this and suggest any > corrections> which should be provided to the total energies for the > charged cells.>> with kind regards,> Yurko>>> --> Yurko (aka Yuriy, > Iurii, Jurij etc) Natanzon> PhD student> Department for Structural > Research (NZ31)> Henryk Niewodnicza?ski
[Wien] Bad formation energies for the charged vacancies
Did you try (from one of Freeman's papers I believe): -Q*V_0(Q)/2 where Q is the excess charge of the cell (-ve for a negative cell) and V_0(Q) is the vacuum Coulomb potential in the calculation which is a function of Q which will depend upon the RMT. Note the factor of 2. I don't think this is in Wien2k at the moment, it needs to be added retroactively (or could in principle be added to the code). It is a long time ago and I seem to remember that this worked for the empty cell test, but no longer have the data, so On Wed, Feb 24, 2010 at 3:18 PM, Peter Blaha wrote: > I've started some tests after the first query and it seems we might miss a > term in the total > energy. > > I created a clmsum-file (density) which is constant and is normalized to one > and put this into > a cell with a single H nucleus. > So it refers to the test case of a H+ ion in a lattice, where I do not add a > constant background, but > put the "background charge into case.clmsum. > > When one switches off the XC-terms, the resulting E-tot contains the > integral (rho *V-coul) and since > rho is constant (equal to Q/volume), we get the average potential in the > unit cell (not only the > interstital, where it is zero anyway) multiplied by the constant rho). > > This term is missing when I put a clmsum file with rho=zero, but add a > "background charge" > by case.inm, while the resulting potentials are identical for the two > methods. > > However, for a charged bulk system there is still a big problem, because > V-coul is determined > only up to a constant and is shifted arbitrarily such that the potential in > the interstital is zero. > In "neutral" calculations such a shift does not matter, since it will be > canceled by the sum of > eigenvalues, but when adding the constant background it matters. > > Thus, this correction term depends on RMT ? > > At the moment I'm not sure how I should continue. I think in other codes > such a correction is > added, but as mentioned, I guess the correction depends on the arbitrary > choice of V-zero. > > > Laurence Marks schrieb: >> >> Please see the next email on the >> list:http://zeus.theochem.tuwien.ac.at/pipermail/wien/2007-January/008713.html >> I think this is right and you take V0 from case.output0 (it is >> printedthere). You should do an empty cell test (no electrons) to verify >> thisand the units of V0, perhaps also looking at the code itself -- >> andremember to check the limit as the distance between atoms gets large. >> >> On Wed, Feb 24, 2010 at 11:05 AM, Yurko Natanzon >> wrote:> Dear Wien2k users and developers,> I'd like to refresh the >> discussion about the total energies of the> charged cells which took place >> three years ago:> >> http://zeus.theochem.tuwien.ac.at/pipermail/wien/2007-January/008711.html>> >> I'm trying to calculate the formation energy of the Hydrogen vacancy> in >> +/-1 charge states and find that the results are bad (much differ> from the >> literature) although the formation energy of the neutral> hydrogen vacancy >> is good. So my question arises if we can trust the> values of the total >> energies for the charged cells in the recent> version of Wien2k?>> To >> investigate this issue further I have performed the following> tests: I've >> done the calculations of the total energy of Mg, MgH2 and> GaN for three >> cases: neutral cell, cell with one electron removed (+1> charge) and a cell >> with an electron added (-1 charge). The results> were compared with the same >> calculati > > on with another plane-wave code> and are the following:> > -> hcp Mg:> Wien2k:> E(+1)-E(0) = > 0.245 Ry> E(-1)-E(0) = -0.199 Ry>> Plane-Wave code:> E(+1)-E(0) = -0.226 Ry> > E(-1)-E(0) = 0.281 Ry>> bcc MgH2> Wien2k:> E(+1)-E(0) = 0.277 Ry> E(-1)-E(0) > = 0.085 Ry>> Plane-Wave code:> E(+1)-E(0) = 0.024 Ry> E(-1)-E(0) = 0.326 > Ry>> fcc GaN> Wien2k:> E(+1)-E(0) = 1.12 Ry> E(-1)-E(0) = -0.717 Ry>> > Plane-Wave code:> E(+1)-E(0) = -0.151 Ry> E(-1)-E(0) = 0.443 Ry> > ->> In wien2k the charged > cell was created by changing the number of> electrons in case.in2 and adding > the corresponding background charge> in case.inm. One can observe, that the > energies have the same order of> magnitude, but the sequence of energies > E(+1), E(0) and E(1) is> inverse. It seems, that the system with +1 charge > (electron removed)> behaves like the system with -1 charge in the Plane-Wave > code. Of> course, the results of tests are no > t physical, because no supercell> was used and no geometric relaxation was > performed (however, it is not> needed for Mg), but if one tries to do all > the supercell and> relaxation stuff and tries to calculate the defect > formation energy,> the result will be the same.>> I'd be grateful if you > comment on this and suggest any corrections> which should be provided to the > total energies for the charged cells.>> with kind regards,> Yurko>>> --> > Yurko (aka Yuriy, I
[Wien] volume optimization and force minimization
Sorry, I am not an expert on Fe although I think you only want spin-orbit for heavy atoms. 2010/2/24 shamik chakrabarti : > Dear Laurence Marks Sir, > ?? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? We are doing calculation on a system > which has Fe atom as the most atomic numbered atom. Other atoms have less > atomic no. than Fe. Then is it really necessary to include spin orbit > coupling term for this system?otherwise we can exclude it and do the > necessary optimization. > Another question isis this spin-orbit coupling term can have any effect > on volume or internal coordinates? > Sirtill now I have got lots of information from you. I am very greatfull > to you sir for all your replies. > Thanking You, > with regards, > Shamik Chakrabarti > > On Wed, Feb 24, 2010 at 10:34 PM, Laurence Marks > wrote: >> >> No, you cannot compare energies with/without spin-orbit, they are >> different physics. You could optimize the volume after adding it in >> (not internal co-ordinates) and this would be reasonable to do. >> >> 2010/2/24 shamik chakrabarti : >> > Dear?Laurence Marks Sir, >> > ?? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? Thank you very much for your reply. >> > What we can do then.we optimize both volume and coordinates >> > including >> > spin polarization and GGA+U and take this as the optimized structure as >> > far >> > as the theory is concerned. Then we take that structure and run one scf >> > including spin-orbit coupling too. If the energy is less than the >> > minimum >> > energy obtained in volume optimization we can go with spin-orbit >> > coupling or >> > otherwise not. But it may happen that the energy is less and the volume >> > is >> > different (I don't know how much effect spin-orbit coupling will have on >> > the >> > volume!!...atleast we have experience that spin polarization definitely >> > effects volume) then which volume we should accept as our optimized >> > volume? >> > >> > On Wed, Feb 24, 2010 at 7:26 PM, Laurence Marks >> > >> > wrote: >> >> >> >> Unless something has changed, forces are not implemented for >> >> spin-orbit correctly so you cannot do a force minimization with them >> >> (unless you do some tricks, i.e. minimize the enegy by hand not using >> >> the code). >> >> >> >> For spin/U the question is how large a difference does this make to >> >> the electron density -- if it is large then the positions/optimal >> >> volume will be very different. >> >> >> >> But...do not depair! If spin-polarized and U is a much better >> >> description of the physics and you do not have multiple magnetic >> >> states, convergence of a LDA+U (or GGA+U) calculation can be much >> >> faster than a LDA calculation which does not describe the physics >> >> well. This follows from the physics/math of mixing. >> >> >> >> N.B., this is also why using a small mixing term in a >> >> Broyden/Multisecant method does not do what most people think it does, >> >> conventional wisdom is incorrect or at most correct for Pratt mixing. >> >> (Also, large mixing terms do not do what people think they do!) In >> >> general this not the way to solve a badly convergent problem, the >> >> correct (only) solution is to improve the physical model. >> >> >> >> 2010/2/24 shamik chakrabarti : >> >> > Dear Wien2k users, >> >> > ?? ? ? ? ? ? ? ? ? ? ? ? ? ?I have a question regarding volume >> >> > optimization >> >> > and force minimization. >> >> > If I do volume optimization and force minimization for non magnetic >> >> > calculation and then use that optimized volume and structural >> >> > coordinates as >> >> > input for the calculation of volume optimization and force >> >> > minimization >> >> > including spin polarization, spin orbit coupling term and LDA+U term, >> >> > then >> >> > whether I get the correct optimized values for both volume and >> >> > coordinates >> >> > of the structure which really need those corrections? >> >> > If I want to do volume and coordinates optimization including those >> >> > correction terms from the begining then it will take a very long >> >> > time. >> >> > But >> >> > if we do the optimization using non magnetic calculation to reach >> >> > closer >> >> > to >> >> > the actual energy and force minima and then put those optimized >> >> > structure >> >> > for non magnetic calculation for the optimization of the structure >> >> > having >> >> > those corrections it may take lesser time to reach the actual minima. >> >> > But >> >> > whether we really get actual minima in this way? >> >> > The above two questions actually are the same questions. I am very >> >> > sorry >> >> > for >> >> > that reapeating. Actually I want to make clear my intentions to the >> >> > wien2k >> >> > users. >> >> > Thanks in advance. >> >> > regards, >> >> > Shamik Chakrabarti >> >> > ___ >> >> > Wien mailing list >> >> > Wien at zeus.theochem.tuwien.ac.at >> >> > http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien >> >> > >> >> > >> >> >> >
[Wien] Bad formation energies for the charged vacancies
Please see the next email on the list: http://zeus.theochem.tuwien.ac.at/pipermail/wien/2007-January/008713.html I think this is right and you take V0 from case.output0 (it is printed there). You should do an empty cell test (no electrons) to verify this and the units of V0, perhaps also looking at the code itself -- and remember to check the limit as the distance between atoms gets large. On Wed, Feb 24, 2010 at 11:05 AM, Yurko Natanzon wrote: > Dear Wien2k users and developers, > I'd like to refresh the discussion about the total energies of the > charged cells which took place three years ago: > http://zeus.theochem.tuwien.ac.at/pipermail/wien/2007-January/008711.html > > I'm trying to calculate the formation energy of the Hydrogen vacancy > in +/-1 charge states and find that the results are bad (much differ > from the literature) although the formation energy of the neutral > hydrogen vacancy is good. So my question arises if we can trust the > values of the total energies for the charged cells in the recent > version of Wien2k? > > To investigate this issue further I have performed the following > tests: I've done the calculations of the total energy of Mg, MgH2 and > GaN for three cases: neutral cell, cell with one electron removed (+1 > charge) and a cell with an electron added (-1 charge). The results > were compared with the same calculation with another plane-wave code > and are the following: > - > hcp Mg: > Wien2k: > E(+1)-E(0) = 0.245 Ry > E(-1)-E(0) = -0.199 Ry > > Plane-Wave code: > E(+1)-E(0) = -0.226 Ry > E(-1)-E(0) = 0.281 Ry > > bcc MgH2 > Wien2k: > E(+1)-E(0) = 0.277 Ry > E(-1)-E(0) = 0.085 Ry > > Plane-Wave code: > E(+1)-E(0) = 0.024 Ry > E(-1)-E(0) = 0.326 Ry > > fcc GaN > Wien2k: > E(+1)-E(0) = 1.12 Ry > E(-1)-E(0) = -0.717 Ry > > Plane-Wave code: > E(+1)-E(0) = -0.151 Ry > E(-1)-E(0) = 0.443 Ry > - > > In wien2k the charged cell was created by changing the number of > electrons in case.in2 and adding the corresponding background charge > in case.inm. One can observe, that the energies have the same order of > magnitude, but the sequence of energies E(+1), E(0) and E(1) is > inverse. It seems, that the system with +1 charge (electron removed) > behaves like the system with -1 charge in the Plane-Wave code. Of > course, the results of tests are not physical, because no supercell > was used and no geometric relaxation was performed (however, it is not > needed for Mg), but if one tries to do all the supercell and > relaxation stuff and tries to calculate the defect formation energy, > the result will be the same. > > I'd be grateful if you comment on this and suggest any corrections > which should be provided to the total energies for the charged cells. > > with kind regards, > Yurko > > > -- > Yurko (aka Yuriy, Iurii, Jurij etc) Natanzon > PhD student > Department for Structural Research (NZ31) > Henryk Niewodnicza?ski Institute of Nuclear Physics > Polish Academy of Sciences > ul. Radzikowskiego 152, > 31-342 Krakow, Poland > E-mail: Yurii.Natanzon at ifj.edu.pl, yurko.natanzon at gmail.com > ___ > Wien mailing list > Wien at zeus.theochem.tuwien.ac.at > http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien > -- Laurence Marks Department of Materials Science and Engineering MSE Rm 2036 Cook Hall 2220 N Campus Drive Northwestern University Evanston, IL 60208, USA Tel: (847) 491-3996 Fax: (847) 491-7820 email: L-marks at northwestern dot edu Web: www.numis.northwestern.edu Chair, Commission on Electron Crystallography of IUCR www.numis.northwestern.edu/ Electron crystallography is the branch of science that uses electron scattering and imaging to study the structure of matter.
[Wien] volume optimization and force minimization
No, you cannot compare energies with/without spin-orbit, they are different physics. You could optimize the volume after adding it in (not internal co-ordinates) and this would be reasonable to do. 2010/2/24 shamik chakrabarti : > Dear?Laurence Marks Sir, > ?? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? Thank you very much for your reply. > What we can do then.we optimize both volume and coordinates including > spin polarization and GGA+U and take this as the optimized structure as far > as the theory is concerned. Then we take that structure and run one scf > including spin-orbit coupling too. If the energy is less than the minimum > energy obtained in volume optimization we can go with spin-orbit coupling or > otherwise not. But it may happen that the energy is less and the volume is > different (I don't know how much effect spin-orbit coupling will have on the > volume!!...atleast we have experience that spin polarization definitely > effects volume) then which volume we should accept as our optimized volume? > > On Wed, Feb 24, 2010 at 7:26 PM, Laurence Marks > wrote: >> >> Unless something has changed, forces are not implemented for >> spin-orbit correctly so you cannot do a force minimization with them >> (unless you do some tricks, i.e. minimize the enegy by hand not using >> the code). >> >> For spin/U the question is how large a difference does this make to >> the electron density -- if it is large then the positions/optimal >> volume will be very different. >> >> But...do not depair! If spin-polarized and U is a much better >> description of the physics and you do not have multiple magnetic >> states, convergence of a LDA+U (or GGA+U) calculation can be much >> faster than a LDA calculation which does not describe the physics >> well. This follows from the physics/math of mixing. >> >> N.B., this is also why using a small mixing term in a >> Broyden/Multisecant method does not do what most people think it does, >> conventional wisdom is incorrect or at most correct for Pratt mixing. >> (Also, large mixing terms do not do what people think they do!) In >> general this not the way to solve a badly convergent problem, the >> correct (only) solution is to improve the physical model. >> >> 2010/2/24 shamik chakrabarti : >> > Dear Wien2k users, >> > ?? ? ? ? ? ? ? ? ? ? ? ? ? ?I have a question regarding volume >> > optimization >> > and force minimization. >> > If I do volume optimization and force minimization for non magnetic >> > calculation and then use that optimized volume and structural >> > coordinates as >> > input for the calculation of volume optimization and force minimization >> > including spin polarization, spin orbit coupling term and LDA+U term, >> > then >> > whether I get the correct optimized values for both volume and >> > coordinates >> > of the structure which really need those corrections? >> > If I want to do volume and coordinates optimization including those >> > correction terms from the begining then it will take a very long time. >> > But >> > if we do the optimization using non magnetic calculation to reach closer >> > to >> > the actual energy and force minima and then put those optimized >> > structure >> > for non magnetic calculation for the optimization of the structure >> > having >> > those corrections it may take lesser time to reach the actual minima. >> > But >> > whether we really get actual minima in this way? >> > The above two questions actually are the same questions. I am very sorry >> > for >> > that reapeating. Actually I want to make clear my intentions to the >> > wien2k >> > users. >> > Thanks in advance. >> > regards, >> > Shamik Chakrabarti >> > ___ >> > Wien mailing list >> > Wien at zeus.theochem.tuwien.ac.at >> > http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien >> > >> > >> >> >> >> -- >> Laurence Marks >> Department of Materials Science and Engineering >> MSE Rm 2036 Cook Hall >> 2220 N Campus Drive >> Northwestern University >> Evanston, IL 60208, USA >> Tel: (847) 491-3996 Fax: (847) 491-7820 >> email: L-marks at northwestern dot edu >> Web: www.numis.northwestern.edu >> Chair, Commission on Electron Crystallography of IUCR >> www.numis.northwestern.edu/ >> Electron crystallography is the branch of science that uses electron >> scattering and imaging to study the structure of matter. >> ___ >> Wien mailing list >> Wien at zeus.theochem.tuwien.ac.at >> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien > > > ___ > Wien mailing list > Wien at zeus.theochem.tuwien.ac.at > http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien > > -- Laurence Marks Department of Materials Science and Engineering MSE Rm 2036 Cook Hall 2220 N Campus Drive Northwestern University Evanston, IL 60208, USA Tel: (847) 491-3996 Fax: (847) 491-7820 email: L-marks at northwestern dot edu Web: www.numis.northwestern.edu Chair, Comm
[Wien] (no subject)
Dear users, I'm working on the supercell approch, for determine the electronic properties of In1-xSnxS alloy. When i have constructed the supercell of this alloy, i have found that the initial space group have changed from orthormbic structure to monoclinc one. My question is: I work with the new structure obtained from "sgroup" for initialize calculations and construct a new struct file? -- next part -- An HTML attachment was scrubbed... URL: <http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20100224/84d6ba13/attachment.htm>
[Wien] volume optimization and force minimization
Unless something has changed, forces are not implemented for spin-orbit correctly so you cannot do a force minimization with them (unless you do some tricks, i.e. minimize the enegy by hand not using the code). For spin/U the question is how large a difference does this make to the electron density -- if it is large then the positions/optimal volume will be very different. But...do not depair! If spin-polarized and U is a much better description of the physics and you do not have multiple magnetic states, convergence of a LDA+U (or GGA+U) calculation can be much faster than a LDA calculation which does not describe the physics well. This follows from the physics/math of mixing. N.B., this is also why using a small mixing term in a Broyden/Multisecant method does not do what most people think it does, conventional wisdom is incorrect or at most correct for Pratt mixing. (Also, large mixing terms do not do what people think they do!) In general this not the way to solve a badly convergent problem, the correct (only) solution is to improve the physical model. 2010/2/24 shamik chakrabarti : > Dear Wien2k users, > ?? ? ? ? ? ? ? ? ? ? ? ? ? ?I have a question regarding volume optimization > and force minimization. > If I do volume optimization and force minimization for non magnetic > calculation and then use that optimized volume and structural coordinates as > input for the calculation of volume optimization and force minimization > including spin polarization, spin orbit coupling term and LDA+U term, then > whether I get the correct optimized values for both volume and coordinates > of the structure which really need those corrections? > If I want to do volume and coordinates optimization including those > correction terms from the begining then it will take a very long time. But > if we do the optimization using non magnetic calculation to reach closer to > the actual energy and force minima and then put those optimized structure > for non magnetic calculation for the optimization of the structure having > those corrections it may take lesser time to reach the actual minima. But > whether we really get actual minima in this way? > The above two questions actually are the same questions. I am very sorry for > that reapeating. Actually I want to make clear my intentions to the wien2k > users. > Thanks in advance. > regards, > Shamik Chakrabarti > ___ > Wien mailing list > Wien at zeus.theochem.tuwien.ac.at > http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien > > -- Laurence Marks Department of Materials Science and Engineering MSE Rm 2036 Cook Hall 2220 N Campus Drive Northwestern University Evanston, IL 60208, USA Tel: (847) 491-3996 Fax: (847) 491-7820 email: L-marks at northwestern dot edu Web: www.numis.northwestern.edu Chair, Commission on Electron Crystallography of IUCR www.numis.northwestern.edu/ Electron crystallography is the branch of science that uses electron scattering and imaging to study the structure of matter.
