[Wannier] Semi-infinite slab in the Wannier basis

Tue Sep 18 12:05:30 CEST 2018

Dear Alex,

My general advice would be to use a k-point grid at which the property/properties you are interested in are converged to the tolerance level that you require.

My comment about your k-point grid was more to do with the ratio between the number of grid points in the out-of-plane direction vs the in-plane direction. Apologies if this is all very obvious:

Your lattice vectors (let’s call them A1, A2 and A3) are orthogonal, with A2 perpendicular to the slab and being almost 10 times longer than A1 and A3. Your Brillouin zone is defined by a set of reciprocal lattice vectors (let’s call them B1, B2 and B3). In your case, the reciprocal lattice vectors also form an orthogonal set that are co-aligned with A1, A2 and A3. But B2 is now shorter than B1 and B3 by almost 10 times. This is a simply consequence of the standard definition of a reciprocal lattice. As a zeroth-order approach, one would want to use a k-point sampling of the Brillouin zone so that the sampling density is roughly the same in all three reciprocal lattice vector directions (unless there are specific reasons related to your system not to). In your case this means, as a first try, I would expect a k-point sampling of N x M x N, where N and M are integers and N is larger than M by about a factor of 10. In your input file and your post below, you have it the other way around. You would need to do some convergence testing to determine what actual values of N and M you would need for your problem.

Best wishes,

Arash

—
Arash Mostofi — www.mostofigroup.org<http://www.mostofigroup.org>
Director, Thomas Young Centre @Imperial
Imperial College London

On 18 Sep 2018, at 10:02, Alex.Durie <alex.durie at open.ac.uk<mailto:alex.durie at open.ac.uk>> wrote:

Dear Arash,

Many thanks for your response, despite getting the error Wannier90 seems to terminate successfully, as I have observed in a previous posting (except when use_ws_distance = true).

Judging from your response, do you believe the issues stem from poor convergence due to lack of out-of-plane k-points, rather than k-points being incorrectly placed within the vacuum?

I had set this low as I wished for a relatively fast runtime, just during testing.

When I have access to the University's supercomputer I intend to use a 6x90x6 grid for the scf run and a 3x45x3 grid thereafter.

Best Wishes

Alex

From: Alex.Durie
Sent: Tuesday, 18 September, 09:55
Subject: Re: Wannier Digest, Vol 128, Issue 5
To: wannier at lists.quantum-espresso.org<mailto:wannier at lists.quantum-espresso.org>

Dear Arash,

Many thanks for your response, despite getting the error Wannier90 seems to terminate successfully, as I have observed in a previous posting (except when use_ws_distance = true).

Judging from your response, do you believe the issues stem from poor convergence due to lack of out-of-plane k-points, rather than k-points being incorrectly placed within the vacuum?

I had set this low as I wished for a relatively fast runtime, just during testing.

When I have access to the University's supercomputer I intend to use a 6x90x6 grid for the scf run and a 3x45x3 grid thereafter.

Best Wishes

Alex

From: wannier-request at lists.quantum-espresso.org<mailto:wannier-request at lists.quantum-espresso.org>
Sent: Sunday, 16 September, 20:29
Subject: Wannier Digest, Vol 128, Issue 5
To: wannier at lists.quantum-espresso.org<mailto:wannier at lists.quantum-espresso.org>

Send Wannier mailing list submissions to wannier at lists.quantum-espresso.org<mailto:wannier at lists.quantum-espresso.org> To subscribe or unsubscribe via the World Wide Web, visit https://lists.quantum-espresso.org/mailman/listinfo/wannier or, via email, send a message with subject or body 'help' to wannier-request at lists.quantum-espresso.org<mailto:wannier-request at lists.quantum-espresso.org> You can reach the person managing the list at wannier-owner at lists.quantum-espresso.org<mailto:wannier-owner at lists.quantum-espresso.org> When replying, please edit your Subject line so it is more specific than "Re: Contents of Wannier digest..." Today's Topics: 1. Re: Semi-infinite slab in the Wannier basis (Mostofi, Arash) ---------------------------------------------------------------------- Message: 1 Date: Sun, 16 Sep 2018 19:28:25 +0000 From: "Mostofi, Arash" To: "wannier at lists.quantum-espresso.org<mailto:wannier at lists.quantum-espresso.org>" Subject: Re: [Wannier] Semi-infinite slab in the Wannier basis Message-ID: Content-Type: text/plain; charset="utf-8" Dear Alex, Please have a look at this earlier post and see whether it helps. http://lists.quantum-espresso.org/pipermail/wannier/2018-February/001484.html Also, it seems that you are using a lower density of k-points (larger number of k-points) along the direction normal to the surface (y-direction in your case) than along the directions parallel to the surface. I would have expected the opposite. Best wishes, Arash ? Arash Mostofi ? www.mostofigroup.org<http://www.mostofigroup.org> Imperial College London On 16 Sep 2018, at 14:55, Alex.Durie > wrote: Dear experts, I am attempting to make a slab of FCC ferromagnetic cobalt enclosed by a vacuum. In order to reduce the number of atoms in the supercell, it is grown in the 011 direction in-plane so that there is one atom in-plane, and five atoms out-of-plane. I am using a 3x9x3 Monkhorst-Pack mesh as I require an in-plane bandstructure. (Here I have chosen the y-axis to be out-of-plane, as Quantum ESPRESSO defaults the magnetization to be along the z-axis). Ultimately I wish to model a dielectric and make use of the real space Hamiltonian't that Wannier90 produces, but as a check I wish to make sure I am generating the slab correctly in the first instance. I have used kmesh.pl to generate the list of k-points. Whilst I know this generates the correct set of k-points to be used for QE's nscf run, as indicated in it's output, I am unsure as to whether this creates the correct set of points for use with Wannier90. My concerns are; When wannier90 terminates, it outputs 'wrong irdist_ws' in seedname.wout, I get the following output; +----------------------------------------------------------------------------+ | b_k Vectors (Ang^-1) and Weights (Ang^2) | | ---------------------------------------- | | No. b_k(x) b_k(y) b_k(z) w_b | | --- -------------------------------- -------- | | 1 0.000000 0.028788 0.000000 603.306159 | | 2 0.000000 -0.028788 0.000000 603.306159 | | 3 0.000000 0.000000 0.820467 0.742759 | | 4 0.820467 0.000000 0.000000 0.742759 | | 5 0.000000 0.000000 -0.820467 0.742759 | | 6 -0.820467 0.000000 0.000000 0.742759 | +----------------------------------------------------------------------------+ Whilst I don't fully understand what this means, the weight of the first two b_k vectors concerns me. Finally, my spread doesn't seem very good. At 8000 iterations, my final spread is 454, with the 20th WF representing a spread of 113. My input files follow; Many thanks, Alex Durie PhD Student Open University United Kingdom cobalt_up.win ------------------------------------------------------------------------------------------- num_bands = 45 num_wann = 45 num_iter = 8000 dis_win_max = 30.0 dis_froz_max = 12.5 dis_num_iter = 45000 dis_mix_ratio = 0.5 write_hr = true translate_home_cell = true guiding_centres = true search_shells = 60 Begin Kpoint_Path G 0.00 0.00 0.00 X 0.50 0.00 0.50 X 0.50 0.00 0.500 L 0.00 0.00 0.50 L 0.00 0.00 0.50 G 0.00 0.00 0.00 End Kpoint_Path # restart = plot bands_plot = .true. begin unit_cell_cart bohr 4.82388 0.000000 0.00000 0.00000 45.82686 0.00000 0.00000 0.000000 4.82388 end unit_cell_cart begin atoms_frac Co 0.0000000 -0.148865 0.0000000 Co 0.5000000 -0.074432 0.5000000 Co 0.0000000 0.000000 0.0000000 Co 0.5000000 0.074432 0.5000000 Co 0.0000000 0.148865 0.0000000 end atoms_frac begin projections Co:s;p;d end projections mp_grid : 3 9 3 begin kpoints 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 0.33333333 0.00000000 0.00000000 0.66666667 0.00000000 0.11111111 0.00000000 0.00000000 0.11111111 0.33333333 0.00000000 0.11111111 0.66666667 0.00000000 0.22222222 0.00000000 0.00000000 0.22222222 0.33333333 0.00000000 0.22222222 0.66666667 0.00000000 0.33333333 0.00000000 0.00000000 0.33333333 0.33333333 0.00000000 0.33333333 0.66666667 0.00000000 0.44444444 0.00000000 0.00000000 0.44444444 0.33333333 0.00000000 0.44444444 0.66666667 0.00000000 0.55555556 0.00000000 0.00000000 0.55555556 0.33333333 0.00000000 0.55555556 0.66666667 0.00000000 0.66666667 0.00000000 0.00000000 0.66666667 0.33333333 0.00000000 0.66666667 0.66666667 0.00000000 0.77777778 0.00000000 0.00000000 0.77777778 0.33333333 0.00000000 0.77777778 0.66666667 0.00000000 0.88888889 0.00000000 0.00000000 0.88888889 0.33333333 0.00000000 0.88888889 0.66666667 0.33333333 0.00000000 0.00000000 0.33333333 0.00000000 0.33333333 0.33333333 0.00000000 0.66666667 0.33333333 0.11111111 0.00000000 0.33333333 0.11111111 0.33333333 0.33333333 0.11111111 0.66666667 0.33333333 0.22222222 0.00000000 0.33333333 0.22222222 0.33333333 0.33333333 0.22222222 0.66666667 0.33333333 0.33333333 0.00000000 0.33333333 0.33333333 0.33333333 0.33333333 0.33333333 0.66666667 0.33333333 0.44444444 0.00000000 0.33333333 0.44444444 0.33333333 0.33333333 0.44444444 0.66666667 0.33333333 0.55555556 0.00000000 0.33333333 0.55555556 0.33333333 0.33333333 0.55555556 0.66666667 0.33333333 0.66666667 0.00000000 0.33333333 0.66666667 0.33333333 0.33333333 0.66666667 0.66666667 0.33333333 0.77777778 0.00000000 0.33333333 0.77777778 0.33333333 0.33333333 0.77777778 0.66666667 0.33333333 0.88888889 0.00000000 0.33333333 0.88888889 0.33333333 0.33333333 0.88888889 0.66666667 0.66666667 0.00000000 0.00000000 0.66666667 0.00000000 0.33333333 0.66666667 0.00000000 0.66666667 0.66666667 0.11111111 0.00000000 0.66666667 0.11111111 0.33333333 0.66666667 0.11111111 0.66666667 0.66666667 0.22222222 0.00000000 0.66666667 0.22222222 0.33333333 0.66666667 0.22222222 0.66666667 0.66666667 0.33333333 0.00000000 0.66666667 0.33333333 0.33333333 0.66666667 0.33333333 0.66666667 0.66666667 0.44444444 0.00000000 0.66666667 0.44444444 0.33333333 0.66666667 0.44444444 0.66666667 0.66666667 0.55555556 0.00000000 0.66666667 0.55555556 0.33333333 0.66666667 0.55555556 0.66666667 0.66666667 0.66666667 0.00000000 0.66666667 0.66666667 0.33333333 0.66666667 0.66666667 0.66666667 0.66666667 0.77777778 0.00000000 0.66666667 0.77777778 0.33333333 0.66666667 0.77777778 0.66666667 0.66666667 0.88888889 0.00000000 0.66666667 0.88888889 0.33333333 0.66666667 0.88888889 0.66666667 end kpoints ------------------------------------------------------------------------------------------- cobalt.scf ------------------------------------------------------------------------------------------- #note finite in y rather than usual z. This is since magnetization #is in z direction. Otherwise would be simpler to use fractional #co-ordinates and ibrav = 6 &control calculation='scf' restart_mode='from_scratch', pseudo_dir = '../../../../pslibrary.1.0.0/pz/PSEUDOPOTENTIALS/', outdir='./' prefix='co' / &system ibrav = 8, celldm(1) =4.82388, celldm(2) =9.5, celldm(3) =1., nat= 5, ntyp= 1, ecutwfc = 30.0 nspin = 2 occupations='smearing', smearing='cold', degauss=0.02 starting_magnetization = -1 nbnd = 45 / &electrons electron_maxstep = 500 diagonalization='cg' conv_thr = 1.0e-6 mixing_beta = 0.3 / ATOMIC_SPECIES Co 58.933195 Co.pz-n-kjpaw_psl.1.0.0.UPF ATOMIC_POSITIONS {bohr} Co 0.0000000 -6.8220000 0.000000 Co 2.4119400 -3.4110000 2.411940 Co 0.0000000 0.0000000 0.000000 Co 2.4119400 3.4110000 2.411940 Co 0.0000000 6.8220000 0.000000 K_POINTS {automatic} 6 18 6 0 0 0 ------------------------------------------------------------------------------------------- cobalt.nscf ------------------------------------------------------------------------------------------- &control calculation='nscf' restart_mode='from_scratch', pseudo_dir = '../../../../pslibrary.1.0.0/pz/PSEUDOPOTENTIALS/', outdir='./' prefix='co' / &system ibrav = 8, celldm(1) =4.82388, celldm(2) =9.5, celldm(3) =1., nat= 5, ntyp= 1, ecutwfc = 30.0 nspin = 2 occupations='smearing', smearing='cold', degauss=0.02 starting_magnetization = -1 nosym = .true. nbnd = 45 / &electrons electron_maxstep = 500 diagonalization='cg' conv_thr = 1.0e-6 / ATOMIC_SPECIES Co 58.933195 Co.pz-n-kjpaw_psl.1.0.0.UPF ATOMIC_POSITIONS {bohr} Co 0.0000000 -6.8220000 0.000000 Co 2.4119400 -3.4110000 2.411940 Co 0.0000000 0.0000000 0.000000 Co 2.4119400 3.4110000 2.411940 Co 0.0000000 6.8220000 0.000000 K_POINTS crystal 81 0.00000000 0.00000000 0.00000000 1.234568e-02 0.00000000 0.00000000 0.33333333 1.234568e-02 0.00000000 0.00000000 0.66666667 1.234568e-02 0.00000000 0.11111111 0.00000000 1.234568e-02 0.00000000 0.11111111 0.33333333 1.234568e-02 0.00000000 0.11111111 0.66666667 1.234568e-02 0.00000000 0.22222222 0.00000000 1.234568e-02 0.00000000 0.22222222 0.33333333 1.234568e-02 0.00000000 0.22222222 0.66666667 1.234568e-02 0.00000000 0.33333333 0.00000000 1.234568e-02 0.00000000 0.33333333 0.33333333 1.234568e-02 0.00000000 0.33333333 0.66666667 1.234568e-02 0.00000000 0.44444444 0.00000000 1.234568e-02 0.00000000 0.44444444 0.33333333 1.234568e-02 0.00000000 0.44444444 0.66666667 1.234568e-02 0.00000000 0.55555556 0.00000000 1.234568e-02 0.00000000 0.55555556 0.33333333 1.234568e-02 0.00000000 0.55555556 0.66666667 1.234568e-02 0.00000000 0.66666667 0.00000000 1.234568e-02 0.00000000 0.66666667 0.33333333 1.234568e-02 0.00000000 0.66666667 0.66666667 1.234568e-02 0.00000000 0.77777778 0.00000000 1.234568e-02 0.00000000 0.77777778 0.33333333 1.234568e-02 0.00000000 0.77777778 0.66666667 1.234568e-02 0.00000000 0.88888889 0.00000000 1.234568e-02 0.00000000 0.88888889 0.33333333 1.234568e-02 0.00000000 0.88888889 0.66666667 1.234568e-02 0.33333333 0.00000000 0.00000000 1.234568e-02 0.33333333 0.00000000 0.33333333 1.234568e-02 0.33333333 0.00000000 0.66666667 1.234568e-02 0.33333333 0.11111111 0.00000000 1.234568e-02 0.33333333 0.11111111 0.33333333 1.234568e-02 0.33333333 0.11111111 0.66666667 1.234568e-02 0.33333333 0.22222222 0.00000000 1.234568e-02 0.33333333 0.22222222 0.33333333 1.234568e-02 0.33333333 0.22222222 0.66666667 1.234568e-02 0.33333333 0.33333333 0.00000000 1.234568e-02 0.33333333 0.33333333 0.33333333 1.234568e-02 0.33333333 0.33333333 0.66666667 1.234568e-02 0.33333333 0.44444444 0.00000000 1.234568e-02 0.33333333 0.44444444 0.33333333 1.234568e-02 0.33333333 0.44444444 0.66666667 1.234568e-02 0.33333333 0.55555556 0.00000000 1.234568e-02 0.33333333 0.55555556 0.33333333 1.234568e-02 0.33333333 0.55555556 0.66666667 1.234568e-02 0.33333333 0.66666667 0.00000000 1.234568e-02 0.33333333 0.66666667 0.33333333 1.234568e-02 0.33333333 0.66666667 0.66666667 1.234568e-02 0.33333333 0.77777778 0.00000000 1.234568e-02 0.33333333 0.77777778 0.33333333 1.234568e-02 0.33333333 0.77777778 0.66666667 1.234568e-02 0.33333333 0.88888889 0.00000000 1.234568e-02 0.33333333 0.88888889 0.33333333 1.234568e-02 0.33333333 0.88888889 0.66666667 1.234568e-02 0.66666667 0.00000000 0.00000000 1.234568e-02 0.66666667 0.00000000 0.33333333 1.234568e-02 0.66666667 0.00000000 0.66666667 1.234568e-02 0.66666667 0.11111111 0.00000000 1.234568e-02 0.66666667 0.11111111 0.33333333 1.234568e-02 0.66666667 0.11111111 0.66666667 1.234568e-02 0.66666667 0.22222222 0.00000000 1.234568e-02 0.66666667 0.22222222 0.33333333 1.234568e-02 0.66666667 0.22222222 0.66666667 1.234568e-02 0.66666667 0.33333333 0.00000000 1.234568e-02 0.66666667 0.33333333 0.33333333 1.234568e-02 0.66666667 0.33333333 0.66666667 1.234568e-02 0.66666667 0.44444444 0.00000000 1.234568e-02 0.66666667 0.44444444 0.33333333 1.234568e-02 0.66666667 0.44444444 0.66666667 1.234568e-02 0.66666667 0.55555556 0.00000000 1.234568e-02 0.66666667 0.55555556 0.33333333 1.234568e-02 0.66666667 0.55555556 0.66666667 1.234568e-02 0.66666667 0.66666667 0.00000000 1.234568e-02 0.66666667 0.66666667 0.33333333 1.234568e-02 0.66666667 0.66666667 0.66666667 1.234568e-02 0.66666667 0.77777778 0.00000000 1.234568e-02 0.66666667 0.77777778 0.33333333 1.234568e-02 0.66666667 0.77777778 0.66666667 1.234568e-02 0.66666667 0.88888889 0.00000000 1.234568e-02 0.66666667 0.88888889 0.33333333 1.234568e-02 0.66666667 0.88888889 0.66666667 1.234568e-02 ------------------------------------------------------------------------------------------- _______________________________________________ Wannier mailing list Wannier at lists.quantum-espresso.org<mailto:Wannier at lists.quantum-espresso.org> https://lists.quantum-espresso.org/mailman/listinfo/wannier -------------- next part -------------- An HTML attachment was scrubbed... URL: ------------------------------ Subject: Digest Footer _______________________________________________ Wannier mailing list Wannier at lists.quantum-espresso.org<mailto:Wannier at lists.quantum-espresso.org> https://lists.quantum-espresso.org/mailman/listinfo/wannier ------------------------------ End of Wannier Digest, Vol 128, Issue 5 ***************************************

_______________________________________________
Wannier mailing list
Wannier at lists.quantum-espresso.org<mailto:Wannier at lists.quantum-espresso.org>
https://lists.quantum-espresso.org/mailman/listinfo/wannier

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/wannier/attachments/20180918/a24e8e09/attachment-0001.html>