ABINIT Discussion Forums

Hi all,

it's very strange that the hdf small version is not backwards compatible. I have encountered the same thing on an old cluster of ours with the (new) hdf 1.12 and netcdf 4.7.4. They should ensure API and binary compatibility at least until version 2!

Hello Nedhal, So, the general idea is ok. However you need to worry about the following things: - abinit is a periodic plane wave code, so you need PBC with vacuum perpendicular to the wire and to find a periodic slice of the wire along the axis. You will need to ensure the atom positions and "c" la...

Hello all, my new version of gfortran (10.2.0) complains with an error when I compile abinit, due to simultaneous casts that it recognizes passing the same variable (real or complex) to a routine that only expects one of the 2 types (even if the latter is outside its scope, like a lapack/blas call)....

PS: see https://iopscience.iop.org/article/10.1 ... /14/13/302 for instance for the spontaneous polarization of GaN

Hi Daosheng, this gets a bit technical. The electrical displacement D you probably know about (berryopt 16), and you'll have to read the supplementary of Stengel's paper to grasp the difference for the reduced stuff. Or just trust them and use berryopt 14 (which is what I would do)! The issue is wit...

Hi again Doasheng, ok, so the warning is not the problem, the output is. It's very strange that the dielectric constant comes out ok (you get it from the P(E) curve?) but the Z* not... Can you compare the P(E) in the ncpp and PAW cases, as well as the P(R) ones? Could it be you are not in the linear...

Hello Daosheng, 1) a few checks first: have you done DFPT calculations of the Born charges? And have you tried norm conserving psp (see pseudo-dojo.org) for the Berry phase version? The former will give you a baseline and is less prone to instabilities, and the latter will tell you if it's a PAW pro...

Also, 12 12 4 is a bit coarse. Even if your system has a large unit cell, unfortunately, the k grid convergence can be quite slow

Hello, we have seen similar effects with some compounds. I think it is due to a rough interpolation of the EPC at low q and omega, which makes the a2F function too large at low w. This unbalances the calculation of omega_log, which is the large lambda limiting maximal value of Tc. Could you post ome...

Brilliant! Thank you - we were about to start debugging this. I will test it immediately
It should be in v9.2 if everything goes smoothly

Best

Matthieu

Hello Erichmond, prt1dm simply does a sum over grid points along 2 directions, divides by the number of points per plane, and prints the result. If your lattice vectors are not orthogonal this is probably not the projection you would expect, but still useful to get an idea. If they are orthogonal to...

Use spinmagtarget, it's easiest.

Do you confirm your system is insulating? Otherwise there is no point...

Hello again (been a while), 1) I hope you are looking for the electronic part of kappa, that is all the epc and anaddb will get you 2) for Si, there is a gap, so nothing will come out, idem for undoped graphene as the DOS is 0 3) your files are identical but contain huge numbers 1.e309! This seems a...

NL/Raman will only work for insulators. If your FM system is really insulating, you should be able to run with occopt 1. Recent versions of abinit should allow for metallic occopt as well (presuming the user knows what she is doing), but I'm not sure this was merged into the public branch yet. Which...

Hello, the easiest is in the log and main output file, if you look for the lambda averaged isotropic ep coupling strength. The A2F files contain the Eliashberg spectral function, which is frequency resolved, to show which phonons contribute to the epc, and lambda is the first inverse momend of a2F. ...

Dear mverstra Thank you very much. I konw the renson of the error.I don't have storage space for caculating.I want to get the ddb.file to compute the elastic constanr and piezoelectric constant.There are two warning in my out.file --------------------------------------------------------------------...

I used the ddb.file to compute elastic constant and piezoelectric constant.The warning appear ddb_piezo : WARNING - Acoustic sum rule violation met : the eigenvalues of accoustic mode are too large at Gamma point Increase cutoff energy or k-points sampling. The three eigenvalues are: 2.438948E-05 -...

hello again, the input file is a good start, but it is still not clear to me exactly what you are calculating and especially when (which dataset, at which step) it is stopping. You are using PAW? I am not sure if all of the alchemical stuff is implemented, and in particular for DFPT. Is it crashing ...

which two datasets? In general the elastic constants and all DFPT is more sensitive to ecut and nkpt than the ground state is. You have to re-check in detail. The ease of convergence depends more on having a delicate system which is magnetic or close to a metal-insulator transition, or very anisotro...

Hello, you are converging, but badly and slowly. You can add more iterations, more empty bands, and you can try to use metallic occupations (occopt 7 for example) with a very small smearing width. This makes the convergence easier, even for insulators. Your post is not clear about which dataset is c...

It runs for me... Check that * all of the GKK files are there and complete, then * that they are included in the merge for mrggkk, and that * you do _not_ include the GKK files from the DDK (following?) in the main GKK file with the epc matrix elements. These files should be listed (just the names) ...

I don't understand what you mean by parameter.

The formula cited calculates the value of the LO-TO phonon frequency split, and is calculated in anaddb for each q vector direction (q->0) given in the "second list of q2ph" (see the tutorials on response functions).

This is copied from Wofferman - please make separate posts for each topic. The short answer is, this is not a bug, simply the error message is not easy to see or find if you run in parallel. This is inevitable - the test is done at the right time and the error message is clear, but in parallel on a ...

Regarding the output units, you are correct that they are not implemented systematically. Some files, but mainly the output+log were implemented a long time ago, and more recent developments (probably the PHDOS file as well) are only in Hartree a.u. This is not a big deal to post-process, and the ma...

See, e.g. eq 61 in

http://iffwww.iff.kfa-juelich.de/~wortm ... l45/C2.pdf

There is a sum evaluated over the different born charges projected on the TO mode you want to renormalize. The change is non-analytical in the sense that it depends on which direction your q vector points when you take the limit q -> 0 (discontinuous at Gamma).