A serious error: distance between some ions is very small

[bhagyashridevaru_bhat]

Hi,

I was trying to do structural optimization for Ca2MnIrO6 with the experimental lattice parameters. So I tried with IBRION =2 and ISIF =2 and different MAGMOM options. but the relaxed structure was completely distorted, it didn't retain the experimental geometry. Hence I just did IBRION = -1 and ISIF =2, with different MAGMOM options. Out of 10 structures (depending on MAGMOM options) I have selected the least energy structure as the magnetic ground state. Using this structure I need to do the OER (Oxygen evalution reaction) calculations. So I took a supercell of that and I need to do O, OH and OOH adsorption on surface active Mn and Ir atoms. First I am trying with Ir active site: O adsorption on Ir. It's a slab calculation. This O-atom, I have kept at a bond length of 2.05 Ang (even tried from 0.97 to 3 Ang). When I run the calculation (IBRION = -1 (to retain the experimental structure), ISIF = 2 (to keep the experimental lattice parameter)), I am getting an error

Code: Select all

WARNING: PSMAXN for non-local potential too small
 LDA part: xc-table for (Slater+PW92), standard interpolation
 -----------------------------------------------------------------------------
|                                                                             |
|           W    W    AA    RRRRR   N    N  II  N    N   GGGG   !!!           |
|           W    W   A  A   R    R  NN   N  II  NN   N  G    G  !!!           |
|           W    W  A    A  R    R  N N  N  II  N N  N  G       !!!           |
|           W WW W  AAAAAA  RRRRR   N  N N  II  N  N N  G  GGG   !            |
|           WW  WW  A    A  R   R   N   NN  II  N   NN  G    G                |
|           W    W  A    A  R    R  N    N  II  N    N   GGGG   !!!           |
|                                                                             |
|     The distance between some ions is very small. Please check the          |
|     nearest-neighbor list in the OUTCAR file.                               |
|     I HOPE YOU KNOW WHAT YOU ARE DOING!                                     |
|                                                                             |
 -----------------------------------------------------------------------------



POSCAR, INCAR and KPOINTS ok, starting setup
 FFT: planning ... GRIDC
 FFT: planning ... GRID_SOFT
 FFT: planning ... GRID
 WAVECAR not read
 -----------------------------------------------------------------------------
|                                                                             |
|     EEEEEEE  RRRRRR   RRRRRR   OOOOOOO  RRRRRR      ###     ###     ###     |
|     E        R     R  R     R  O     O  R     R     ###     ###     ###     |
|     E        R     R  R     R  O     O  R     R     ###     ###     ###     |
|     EEEEE    RRRRRR   RRRRRR   O     O  RRRRRR       #       #       #      |
|     E        R   R    R   R    O     O  R   R                               |
|     E        R    R   R    R   O     O  R    R      ###     ###     ###     |
|     EEEEEEE  R     R  R     R  OOOOOOO  R     R     ###     ###     ###     |
|                                                                             |
|     Orbital orthonormalization failed in the inversion of matrix            |
|     LAPACK: Routine ZPOTRF failed! kpoint: 1 spin: 1                        |
|                                                                             |
|     [b][b][b]Possible solutions for this issue are:                                  |
|     - Please check whether the atoms are too close to each other.           |
|     - Decreasing POTIM might help during relaxations or MDs.                |
|     - Deleting CHGCAR or WAVECAR if incompatible with the structure.        |
|     - Try calculating without symmetrizing (ISYM = 0 or ISYM = -1).         |
|     - As last resort: using a different PAW for some of the atoms.          |
|                                                                             |
|       ---->  I REFUSE TO CONTINUE WITH THIS SICK JOB ... BYE!!! <----     

|                                                                             |
"

I even removed the adsorbate O atom and run the calculation, still getting the error. Though the structure looks perfectly fine in VESTA, why VASP writes " The distance between some ions is very small."??
I tried all these Possible solutions for this issue are:
- Please check whether the atoms are too close to each other.------->Checked the POSCAR structure in VESTA, looks fine
- Decreasing POTIM might help during relaxations or MDs. ----->tried with 0.1, 0.03, 0.01, 0.3
- Deleting CHGCAR or WAVECAR if incompatible with the structure. -------->Fresh calculation no CHGCAR and WAVECAR

• Try calculating without symmetrizing (ISYM = 0 or ISYM = -1). ---------> Tried with ISYM = 0
  • As last resort: using a different PAW for some of the atoms. ---------> PAW_PBE Ca_sv 06Sep2000 , PAW_PBE Mn 06Sep2000, PAW_PBE Ir 06Sep2000, PAW_PBE O 08Apr2002

I even tried with LSCALAPACK = .FALSE. NSW = 0, NSW = 80, by changing ALGO = 38 to 48. But none of these worked.

I have uploaded all the input and output files. Please let me know where I am wrong.

Thank you very much in advance
Sincerely,
Bhagyashri

[jonathan_lahnsteiner2]

Dear Bhagyashridevaru Bhat,

As you are telling in your post the distance between some atoms is very small. I was checking the neighbor list output in your OUTCAR file. Here are the neighbor lists of the first 5 atoms:

Code: Select all

  ion  position               nearest neighbor table
    1  0.954  0.008  0.554-   5 0.00  26 1.93  32 1.93  31 1.93  25 1.93  24 1.93  30 1.93  29 1.93
                             23 1.93  37 2.30  36 2.30  54 2.30  55 2.34  56 2.34  87 2.34  91 2.34
    2  0.046  0.992  0.446-   6 0.00  19 1.93  29 1.93  20 1.93  30 1.93  21 1.93  31 1.93  22 1.93
                             32 1.93  34 2.30  35 2.30  61 2.30  58 2.34  57 2.34  80 2.34  88 2.34
    3  0.546  0.508  0.446-  27 1.93  33 1.93  62 2.30  39 2.30  63 2.34  76 2.34  45 2.34  40 2.34
                             82 2.34  90 2.34  72 2.34  79 2.34  35 2.50  61 2.50  34 2.50  18 3.11
    4  0.454  0.492  0.554-  28 1.93  33 1.93  51 2.30  38 2.30  69 2.34  77 2.34  44 2.34  41 2.34
                             85 2.34  89 2.34  73 2.34  78 2.34  36 2.50  54 2.50  37 2.50  17 3.11
    5  0.954  0.008  0.554-   1 0.00  26 1.93  32 1.93  31 1.93  25 1.93  24 1.93  30 1.93  29 1.93
                             23 1.93  37 2.30  36 2.30  54 2.30  55 2.34  56 2.34  87 2.34  91 2.34

From this list you can see for example that the distance between atom 1 and atom 5 is equal to zero. Which means the atoms are sitting on top of each other. The same is for example true for atom pairs 2 and 6. When considering your POSCAR file and taking the lines 9 (position of atom 1) and line 13 (position of atom 5) you will get the following values:

Code: Select all

position 1 5.1630459999999996  5.3903809999999996 19.5825999915999986
position 2 5.1630459999999996  0.0427810000000000 19.5825999915999986

Now if you compare the y position of atom 1 with the lattice parameter in y direction (5.3475999832000003) you will see the atom is sticking out of the box. Therefore periodic boundary conditions have to be applied. When doing so the y position of atom 1 becomes 5.390380-5.3475999=0.0427810 which is the same as the value of atom 5. And so the atoms are occupying the same site.
To get rid of your errors you will have to correct the positions in the POSCAR file. You can use the neighbor list from the OUTCAR file to help with this.

All the best

Jonathan