set it for each atom one by one: %block DM.InitSpin 1 4.0 2 -4.0 3 ...... %endblock DM.InitSpin
From: toufik esssakhri Date: 2014-06-24 16:47 To: siesta code Subject: RE: [SIESTA-L] 4zGNR at the begening i set the species with block ChemicalSpeciesLabel NumberOfSpecies 2 NumberOfAtoms 2 %block ChemicalSpeciesLabel 1 26 Fe #magnetic 2 45 Rh #nonmagnetic %endblock ChemicalSpeciesLabel according to siesta manual i should do something like this: %block DM.InitSpin 1 + # Fe atom is magnetic %endblock DM.InitSpin in this case the system is ferromagnetic and i get Q(up) - Q(down) /=0 my question is : how to fix: Fe1 + Fe2 - how to modify the block ChemicalSpeciesLabel to be coherent with the block DM.InitSpin?? what i see that Fe1 and Fe2 not reconized! thaks taoufik Date: Tue, 24 Jun 2014 15:24:39 +0800 From: [email protected] To: [email protected] Subject: Re: [SIESTA-L] 4zGNR Do it in the "DM.InitSpin" block. From: toufik esssakhri Date: 2014-06-24 15:06 To: siesta code Subject: RE: [SIESTA-L] 4zGNR Dear, i am intrested on this subject >InitSpinAf only sets every second atom spin up, and every other second spin >down. this is true to configurate the hole system as antiferromagnetic. now for exemple i have one system with two atom, one antiferromagnetic and one nonmagnetic (for example FeRh), Fe up Fe down Rh nonmagnetic how to fix magnetization in this case? cheers taoufik From: [email protected] Date: Tue, 24 Jun 2014 08:56:19 +0200 To: [email protected] Subject: Re: [SIESTA-L] 4zGNR InitSpinAf only sets every second atom spin up, and every other second spin down. So maybe it depended on your sorting? 2014-06-24 3:16 GMT+02:00 Francisco Wellery Nunes Silva <[email protected]>: Hi Nick, Thanks a lot for you advice, your sugestion realy works, and I had forgot about such possibility, the curious thing is, if I perform a calculation for a bigger zGNR such as 6-zGNR by setting the flag "DM.InitSpinAF .true." and let it relaxing I will get the expected result for the band structure without building the block "DM.InitSpin". It should be probably due the narrowness of the ribbon, and the opposite edges could be interacting, once it is known the the spin polarization in zGNR is width dependent. 2014-06-23 16:41 GMT-04:00 Nick Papior Andersen <[email protected]>: If I recall correctly then spin polarized in siesta means that it initializes a ferromagnetic state, you need to initialize the spins to be anti-ferromagnetic, see DM.Initspin in the manual. 2014-06-23 19:18 GMT+00:00 Francisco Wellery Nunes Silva <[email protected]>: Dear SIESTA Users, I'm trying to compute the band structure of a 4-zGNR, using GGA-PBE as exchange functional, however at the end I got a FM ground state, in which I see no band gap in the plot. Since such result is unexpected does someone could explain me why this is happening??? I've attached my fdf and XV files (with relaxed positions) for better understanding of my question. Thanks for your time and attention, -- ******************************************* Francisco Wellery Nunes Silva. ******************************************* Skype >> silvafwn Ph.D. student at, Universidade Federal do Ceará (UFC), Block 928, Ground Floor, Room -- 02 Caixa Postal 6030 ( Mail-Box 6030 ) C.E.P -- 60.455-900 Fortaleza - Ceará – Brasil ******************************************* -- Kind regards Nick -- ******************************************* Francisco Wellery Nunes Silva. ******************************************* Skype >> silvafwn Ph.D. student at, Universidade Federal do Ceará (UFC), Block 928, Ground Floor, Room -- 02 Caixa Postal 6030 ( Mail-Box 6030 ) C.E.P -- 60.455-900 Fortaleza - Ceará – Brasil ******************************************* -- Kind regards Nick
