Basic array size parameters:
  | Number of species                 :        2
  | Number of atoms                   :        2
  | Number of lattice vectors         :        3
  | Max. basis fn. angular momentum   :        1
  | Max. atomic/ionic basis occupied n:        3
  | Max. number of basis fn. types    :        1
  | Max. radial fns per species/type  :        4
  | Max. logarithmic grid size        :     1334
  | Max. radial integration grid size :       40
  | Max. angular integration grid size:      302
  | Max. angular grid division number :        8
  | Radial grid for Hartree potential :     1334
  | Number of spin channels           :        1

------------------------------------------------------------
          Reading file control.in.
------------------------------------------------------------
  XC: Using LibXC. Please refer to LibXC output for further details.
  Using libxc - Initializing
  Libxc version: 5.1.7


  ** PLEASE ** be careful using LibXC. Whilst most of the LDA, GGA and MGGA functionals,
  and hybrid variations therein, should be stable (though we always recommend testing!),
  there are several fancy functionals with e.g. range-separation, like HSE03/HSE06/LC-PBEh,
  and also MGGA functionals that depend on the laplacian of the density, like BR89,
  that are not fully implemented due to the infrastrucutre for these functionals being incomplete or non-generic.
  If you want to use these functionals please contact the development team and we will work on them with you!

  You have selected libxc functional GGA_X_PBE_SOL (116)
    - Description: Perdew, Burke & Ernzerhof SOL
    - Type: GGA(Exchange)
    - Reference to Literature:
      [1] J. P. Perdew, A. Ruzsinszky, G. I. Csonka, O. A. Vydrov, G. E. Scuseria, L. A. Constantin, X. Zhou, and K. Burke, Phys. Rev. Lett. 100, 136406 (2008)

  You have selected libxc functional GGA_C_PBE_SOL (133)
    - Description: Perdew, Burke & Ernzerhof SOL
    - Type: GGA(Correlation)
    - Reference to Literature:
      [1] J. P. Perdew, A. Ruzsinszky, G. I. Csonka, O. A. Vydrov, G. E. Scuseria, L. A. Constantin, X. Zhou, and K. Burke, Phys. Rev. Lett. 100, 136406 (2008)

  Scalar relativistic treatment of kinetic energy: on-site free-atom approximation to ZORA.
  Found k-point grid:         4         4         4
  Requested output level: MD_light
  DFPT calculation will begin after the DFT/HF calculation.

  Reading configuration options for species O                   .
  | Found nuclear charge :   8.0000
  | Found atomic mass :    15.999400000000000      amu
  | Found l_max for Hartree potential  :   4
  | Found cutoff potl. onset [A], width [A], scale factor :    3.50000    1.50000    1.00000
  | Threshold for basis-dependent cutoff potential is   0.100000E-03
  | Found data for basic radial integration grid :    36 points, outermost radius =    5.000 A
  | Found multiplier for basic radial grid :   1
  | Found angular grid specification: user-specified.
  | Specified grid contains     5 separate shells.
  | Check grid settings after all constraints further below.
  | Found free-atom valence shell :  2 s   2.000
  | Found free-atom valence shell :  2 p   4.000
  | No ionic wave fns used. Skipping ion_occ.
  | No ionic wave fns used. Skipping ion_occ.
  Species O                   : Missing cutoff potential type.
  Defaulting to exp(1/x)/(1-x)^2 type cutoff potential.
  Species O : No 'logarithmic' tag. Using default grid for free atom:
  | Default logarithmic grid data [bohr] : 0.1000E-03 0.1000E+03 0.1012E+01
  Species O : On-site basis accuracy parameter (for Gram-Schmidt orthonormalisation) not specified.
  Using default value basis_acc =  0.1000000E-03.
  Species O                   : Using default innermost maximum threshold i_radial=  2 for radial functions.
  Species O                   : Default cutoff onset for free atom density etc. : 0.35000000E+01 AA.
  Species O                   : Basic radial grid will be enhanced according to radial_multiplier =   1, to contain    36 grid points.

  Reading configuration options for species Mg                  .
  | Found nuclear charge :  12.0000
  | Found atomic mass :    24.305000000000000      amu
  | Found l_max for Hartree potential  :   4
  | Found cutoff potl. onset [A], width [A], scale factor :    4.00000    1.50000    1.00000
  | Threshold for basis-dependent cutoff potential is   0.100000E-03
  | Found data for basic radial integration grid :    40 points, outermost radius =    5.500 A
  | Found multiplier for basic radial grid :   1
  | Found angular grid specification: user-specified.
  | Specified grid contains     5 separate shells.
  | Check grid settings after all constraints further below.
  | Found free-atom valence shell :  3 s   2.000
  | Found free-atom valence shell :  2 p   6.000
  | No ionic wave fns used. Skipping ion_occ.
  | No ionic wave fns used. Skipping ion_occ.
  Species Mg                  : Missing cutoff potential type.
  Defaulting to exp(1/x)/(1-x)^2 type cutoff potential.
  Species Mg: No 'logarithmic' tag. Using default grid for free atom:
  | Default logarithmic grid data [bohr] : 0.1000E-03 0.1000E+03 0.1012E+01
  Species Mg: On-site basis accuracy parameter (for Gram-Schmidt orthonormalisation) not specified.
  Using default value basis_acc =  0.1000000E-03.
  Species Mg                  : Using default innermost maximum threshold i_radial=  2 for radial functions.
  Species Mg                  : Default cutoff onset for free atom density etc. : 0.40000000E+01 AA.
  Species Mg                  : Basic radial grid will be enhanced according to radial_multiplier =   1, to contain    40 grid points.

  Finished reading input file 'control.in'.

------------------------------------------------------------


------------------------------------------------------------
          Reading geometry description geometry.in.
------------------------------------------------------------
  | The smallest distance between any two atoms is         2.11309971 AA.
  | The first atom of this pair is atom number                      1 .
  | The second atom of this pair is atom number                     2 .
  | Wigner-Seitz cell of the first atom image           0     0     0 .
  | (The Wigner-Seitz cell of the second atom is 0 0 0  by definition.)

  Symmetry information by spglib:
  | Precision set to  0.1E-04
  | Number of Operations  : 48
  | Space group           : 225
  | International         : Fm-3m
  | Schoenflies           : Oh^5
  Input structure read successfully.
  The structure contains        2 atoms,  and a total of         20.000 electrons.

  Input geometry:
  | Unit cell:
  |        0.00000000        2.11309971        2.11309971
  |        2.11309971        0.00000000        2.11309971
  |        2.11309971        2.11309971        0.00000000
  | Atomic structure:
  |       Atom                x [A]            y [A]            z [A]
  |    1: Species Mg            0.00000000        0.00000000        0.00000000
  |    2: Species O             2.11309971        2.11309971        2.11309971

  Lattice parameters for 3D lattice (in Angstroms) :     2.988374    2.988374    2.988374
  Angle(s) between unit vectors (in degrees)       :    60.000000   60.000000   60.000000


  Quantities derived from the lattice vectors (in Angstrom^-1):
  | Reciprocal lattice vector 1: -1.486722  1.486722  1.486722
  | Reciprocal lattice vector 2:  1.486722 -1.486722  1.486722
  | Reciprocal lattice vector 3:  1.486722  1.486722 -1.486722
  | Unit cell volume                               :   0.188708E+02  Angstrom^3


  Finished reading input file 'geometry.in'.



  Consistency checks for stacksize environment parameter are next.

  | Maximum stacksize for task 0: 63 [Mb]
  | Maximum stacksize for task 1: 63 [Mb]
  | Maximum stacksize for task 2: 63 [Mb]
  | Maximum stacksize for task 3: 63 [Mb]
  | Maximum stacksize for task 4: 63 [Mb]
  | Maximum stacksize for task 5: 63 [Mb]
  | Maximum stacksize for task 6: 63 [Mb]
  | Maximum stacksize for task 7: 63 [Mb]
   *** Stacksize is not set to unlimited. This might cause problems...
  | Current stacksize for task 0: 63 [Mb]
  | Current stacksize for task 1: 63 [Mb]
  | Current stacksize for task 2: 63 [Mb]
  | Current stacksize for task 3: 63 [Mb]
  | Current stacksize for task 4: 63 [Mb]
  | Current stacksize for task 5: 63 [Mb]
  | Current stacksize for task 6: 63 [Mb]
  | Current stacksize for task 7: 63 [Mb]

  Consistency checks for the contents of control.in are next.

  MPI_IN_PLACE appears to work with this MPI implementation.
  | Keeping use_mpi_in_place .true. (see manual).
  Target number of points in a grid batch is not set. Defaulting to  100
  Method for grid partitioning is not set. Defaulting to parallel hash+maxmin partitioning.
  Batch size limit is not set. Defaulting to    200
  By default, will store active basis functions for each batch.
  If in need of memory, prune_basis_once .false. can be used to disable this option.
  communication_type for Hartree potential was not specified.
  Defaulting to calc_hartree .
  Defaulting to Pulay charge density mixer.
  Pulay mixer: Number of relevant iterations not set.
  Defaulting to    8 iterations.
  Pulay mixer: Number of initial linear mixing iterations not set.
  Defaulting to    0 iterations.
  Work space size for distributed Hartree potential not set.
  Defaulting to   0.200000E+03 MB.
  Mixing parameter for charge density mixing has not been set.
  Using default: charge_mix_param =     0.0500.
  The mixing parameter will be adjusted in iteration number     2 of the first full s.c.f. cycle only.
  Algorithm-dependent basis array size parameters:
  | n_max_pulay                         :        8
  Maximum number of self-consistency iterations not provided.
  Presetting  1000 iterations.
  Presetting      1001 iterations before the initial mixing cycle
  is restarted anyway using the sc_init_iter criterion / keyword.
  Presetting a factor      1.000 between actual scf density residual
  and density convergence criterion sc_accuracy_rho below which sc_init_iter
  takes no effect.
 * No s.c.f. convergence criteria (sc_accuracy_*) were provided in control.in.
 * The run will proceed with a reasonable default guess, but please check whether.
 * the s.c.f. cycles seem to take too much or too little time.
  Calculation of forces was not defined in control.in. No forces will be calculated.
  Geometry relaxation not requested: no relaxation will be performed.
 * Notice: The s.c.f. convergence criterion sc_accuracy_rho was not provided.
 * We used to stop in this case, and ask the user to provide reasonable
 * scf convergence criteria. However, this led some users to employ criteria
 * that led to extremely long run times, e.g., for simple relaxation.
 * We now preset a default value for sc_accuracy_rho if it is not set.
 * You may still wish to check if this setting is too tight or too loose for your needs.
 * Based on n_atoms and forces and force-correction status, FHI-aims chose sc_accuracy_rho =  0.400000E-05 .
  No accuracy limit for integral partition fn. given. Defaulting to  0.1000E-14.
  No threshold value for u(r) in integrations given. Defaulting to  0.1000E-05.
  No occupation type (smearing scheme) given. Defaulting to Gaussian broadening, width =  0.1000E-01 eV.
  The width will be adjusted in iteration number     2 of the first full s.c.f. cycle only.
  S.C.F. convergence parameters will be adjusted in iteration number     2 of the first full s.c.f. cycle only.
  No accuracy for occupation numbers given. Defaulting to  0.1000E-12.
  No threshold value for occupation numbers given. Defaulting to  0.0000E+00.
  No accuracy for fermi level given. Defaulting to  0.1000E-19.
  Maximum # of iterations to find E_F not set. Defaulting to  200.
  Preferred method for the eigenvalue solver ('KS_method') not specified in 'control.in'.
  Defaulting to serial version, LAPACK (via ELSI), since more k-points than CPUs available.
  Will not use alltoall communication since running on < 1024 CPUs.
  Threshold for basis singularities not set.
  Default threshold for basis singularities:  0.1000E-04
  partition_type (choice of integration weights) for integrals was not specified.
  | Using a version of the partition function of Stratmann and coworkers ('stratmann_sparse').
  | At each grid point, the set of atoms used to build the partition table is smoothly restricted to
  | only those atoms whose free-atom density would be non-zero at that grid point.
  Partitioning for Hartree potential was not defined. Using partition_type for integrals.
  | Adjusted default value of keyword multip_moments_threshold to:       0.10000000E-11
  | This value may affect high angular momentum components of the Hartree potential in periodic systems.
  Spin handling was not defined in control.in. Defaulting to unpolarized case.
  Angular momentum expansion for Kerker preconditioner not set explicitly.
  | Using default value of   0
  No explicit requirement for turning off preconditioner.
  | By default, it will be turned off when the charge convergence reaches
  | sc_accuracy_rho  =   0.400000E-05
  No special mixing parameter while Kerker preconditioner is on.
  Using default: charge_mix_param =     0.0500.
  No q(lm)/r^(l+1) cutoff set for long-range Hartree potential.
  | Using default value of  0.100000E-09 .
  | Verify using the multipole_threshold keyword.
  Defaulting to new monopole extrapolation.
  Density update method: automatic selection selected.
  Using density matrix based charge density update.
  Using density matrix based charge density update.
  Using packed matrix style: index .
  Defaulting to use time-reversal symmetry for k-point grid.
  WARNING: Switching off the compensation of charge integration errors
  on the 3D integration grid, since you have requested a DFPT calculation.
  This is unfortunately necessary since "compensate_multipole_errors" is
  not yet implemented for DFPT.
  Use the "compensate_multipole_errors" flag to overwrite this behaviour.
  Set 'collect_eigenvectors' to be '.true.' for all serial calculations. This is mandatory.
  Set 'collect_eigenvectors' to be '.true.' for KS_method lapack_fast and serial.

  Consistency checks for the contents of geometry.in are next.


  Range separation radius for Ewald summation (hartree_convergence_parameter):      2.90568064 bohr.
  Number of empty states per atom not set in control.in .
  | Since you are using a method that relies on the unoccupied spectrum
  | (MP2,GW,RPA et al.), will use the full Hamiltonian size (see below)
  | as the max. possible number of states (occupied plus empty).

  Structure-dependent array size parameters: 
  | Maximum number of distinct radial functions  :        7
  | Maximum number of basis functions            :       11
  | Number of Kohn-Sham states (occupied + empty):       11
------------------------------------------------------------

------------------------------------------------------------
          Preparing all fixed parts of the calculation.
------------------------------------------------------------
  Determining machine precision:
     2.2250738585072014E-308
  Setting up grids for atomic and cluster calculations.

  Creating wave function, potential, and density for free atoms.
  Runtime choices for atomic solver:
  | atomic solver xc        : Libxc (as specified by xc flag)
  | compute density gradient: 1
  | compute kinetic density : F

  Species: O

  List of occupied orbitals and eigenvalues:
    n    l              occ      energy [Ha]    energy [eV]
    1    0           2.0000       -18.856196      -513.1032
    2    0           2.0000        -0.874632       -23.8000
    2    1           4.0000        -0.329448        -8.9647


  Species: Mg

  List of occupied orbitals and eigenvalues:
    n    l              occ      energy [Ha]    energy [eV]
    1    0           2.0000       -46.248371     -1258.4822
    2    0           2.0000        -2.918339       -79.4120
    3    0           2.0000        -0.167132        -4.5479
    2    1           6.0000        -1.703839       -46.3638


  Adding cutoff potential to free-atom effective potential.
  Creating fixed part of basis set: Ionic, confined, hydrogenic.
  Creating atomic-like basis functions for current effective potential.

  Species O                   :

  List of atomic basis orbitals and eigenvalues:
    n    l      energy [Ha]    energy [eV]    outer radius [A]
    1    0       -18.856196      -513.1032       1.415765
    2    0        -0.874632       -23.8000       4.413171
    2    1        -0.329448        -8.9647       4.522403


  Species Mg                  :

  List of atomic basis orbitals and eigenvalues:
    n    l      energy [Ha]    energy [eV]    outer radius [A]
    1    0       -46.248371     -1258.4822       0.932375
    2    0        -2.918339       -79.4120       3.621735
    3    0        -0.167132        -4.5479       5.100062
    2    1        -1.703839       -46.3638       4.404175

  Assembling full basis from fixed parts.
  | Species O :   atomic orbital   1 s accepted.
  | Species O :   atomic orbital   2 s accepted.
  | Species O :   atomic orbital   2 p accepted.
  | Species Mg :   atomic orbital   1 s accepted.
  | Species Mg :   atomic orbital   2 s accepted.
  | Species Mg :   atomic orbital   3 s accepted.
  | Species Mg :   atomic orbital   2 p accepted.

  Basis size parameters after reduction:
  | Total number of radial functions:        7
  | Total number of basis functions :       11

  Per-task memory consumption for arrays in subroutine allocate_ext:
  |           2.391276MB.
  Testing on-site integration grid accuracy.
  |  Species  Function  <phi|h_atom|phi> (log., in eV)  <phi|h_atom|phi> (rad., in eV)
           1        1               -513.1031890709               -513.1030840604
           1        2                -23.7999535917                -23.7999544099
           1        3                 -8.9648862342                 -8.9650414469
           2        4              -1258.4822164247              -1258.4816211506
           2        5                -79.4120407249                -79.4120382024
           2        6                 -4.5534193756                 -4.5532263455
           2        7                -46.3638162945                -46.3638162098

  Preparing densities etc. for the partition functions (integrals / Hartree potential).

  Preparations completed.
  max(cpu_time)          :      0.104 s.
  Wall clock time (cpu1) :      0.167 s.
------------------------------------------------------------

  Initializing index lists of integration centers etc. from given atomic structure:
  Mapping all atomic coordinates to central unit cell.

  Initializing the k-points
  Using symmetry for reducing the k-points
  | k-points reduced from:       64 to       36
  | Number of k-points                             :        36
  | Consuming         36 KiB for k_phase_exx.
  The eigenvectors in the calculations are COMPLEX.
  | K-points in task   0:         4
  | K-points in task   1:         5
  | K-points in task   2:         5
  | K-points in task   3:         5
  | K-points in task   4:         5
  | K-points in task   5:         4
  | K-points in task   6:         4
  | K-points in task   7:         4
  | Number of basis functions in the Hamiltonian integrals :       662
  | Number of basis functions in a single unit cell        :        11
  | Number of centers in hartree potential         :      1097
  | Number of centers in hartree multipole         :       800
  | Number of centers in electron density summation:       573
  | Number of centers in basis integrals           :       633
  | Number of centers in integrals                 :       203
  | Number of centers in hamiltonian               :       612
  | Consuming        204 KiB for k_phase.
  | Number of super-cells (origin) [n_cells]                     :        2197
  | Number of super-cells (after PM_index) [n_cells]             :         364
  | Number of super-cells in hamiltonian [n_cells_in_hamiltonian]:         364
  | Size of matrix packed + index [n_hamiltonian_matrix_size] :        9157
  | Size of matrix packed + index (no_symmetry):       16652

------------------------------------------------------------
          Begin self-consistency loop: Initialization.

          Date     :  20220804, Time     :  103747.654
------------------------------------------------------------
  | Estimated reciprocal-space cutoff momentum G_max:         3.25746748 bohr^-1 .
  | Reciprocal lattice points for long-range Hartree potential:      64
  Partitioning the integration grid into batches with parallel hashing+maxmin method.
  | Number of batches:      161
  | Maximal batch size:     101
  | Minimal batch size:      49
  | Average batch size:      74.137
  | Standard deviation of batch sizes:      14.502

  Integration load balanced across     8 MPI tasks.
  Work distribution over tasks is as follows:
  Task     0 has       1531 integration points.
  Task     1 has       1478 integration points.
  Task     2 has       1477 integration points.
  Task     3 has       1507 integration points.
  Task     4 has       1493 integration points.
  Task     5 has       1507 integration points.
  Task     6 has       1491 integration points.
  Task     7 has       1452 integration points.
  Initializing partition tables, free-atom densities, potentials, etc. across the integration grid (initialize_grid_storage).
  | initialize_grid_storage: Actual outermost partition radius vs. multipole_radius_free
  | (-- VB: in principle, multipole_radius_free should be larger, hence this output)
  | Species        1: Confinement radius =              4.999999999999999 AA, multipole_radius_free =              5.048384829883283 AA.
  | Species        1: outer_partition_radius set to              5.048384829883283 AA .
  | Species        2: Confinement radius =              5.500000000000000 AA, multipole_radius_free =              5.555717568450569 AA.
  | Species        2: outer_partition_radius set to              5.555717568450569 AA .
  | The sparse table of interatomic distances needs       2198.75 kbyte instead of      3205.51 kbyte of memory.
  | Net number of integration points:    11936
  | of which are non-zero points    :     9174
  | Numerical average free-atom electrostatic potential    :    -17.36907272 eV
  Renormalizing the initial density to the exact electron count on the 3D integration grid.
  | Initial density: Formal number of electrons (from input files) :      20.0000000000
  | Integrated number of electrons on 3D grid     :      20.0033153124
  | Charge integration error                      :       0.0033153124
  | Normalization factor for density and gradient :       0.9998342619
  Obtaining max. number of non-zero basis functions in each batch (get_n_compute_maxes).
  | Maximal number of non-zero basis functions:      279 in task     0
  | Maximal number of non-zero basis functions:      280 in task     1
  | Maximal number of non-zero basis functions:      262 in task     2
  | Maximal number of non-zero basis functions:      271 in task     3
  | Maximal number of non-zero basis functions:      266 in task     4
  | Maximal number of non-zero basis functions:      270 in task     5
  | Maximal number of non-zero basis functions:      263 in task     6
  | Maximal number of non-zero basis functions:      273 in task     7
  Allocating        0.007 MB for KS_eigenvector_complex
  Integrating Hamiltonian matrix: batch-based integration.
  Time summed over all CPUs for integration: real work        0.267 s, elapsed        0.353 s => Consider using load balancing!
  Integrating overlap matrix.
  Time summed over all CPUs for integration: real work        0.161 s, elapsed        0.205 s

  Updating Kohn-Sham eigenvalues and eigenvectors using ELSI and the (modified) LAPACK eigensolver.

  Obtaining occupation numbers and electronic chemical potential using ELSI.
  | Note that, for insulating systems, the printed 'chemical potential' value is not uniquely defined.
  | It can be anywhere in the energy gap, as long as it correctly separates occupied and unoccupied states.
  | In systems with a gap, the physically relevant chemical potential is the VBM or HOMO.

  | Chemical potential (Fermi level):    -7.80102073 eV
  Writing Kohn-Sham eigenvalues.
  K-point:       1 at    0.000000    0.000000    0.000000 (in units of recip. lattice)

  State    Occupation    Eigenvalue [Ha]    Eigenvalue [eV]
      1       2.00000         -46.266954        -1258.98788
      2       2.00000         -18.964771         -516.05766
      3       2.00000          -2.944987          -80.13717
      4       2.00000          -1.730355          -47.08535
      5       2.00000          -1.730355          -47.08535
      6       2.00000          -1.730355          -47.08535
      7       2.00000          -1.076952          -29.30535
      8       2.00000          -0.456949          -12.43420
      9       2.00000          -0.456949          -12.43420
     10       2.00000          -0.456949          -12.43420
     11       0.00000          -0.216497           -5.89118

  What follows are estimated values for band gap, HOMO, LUMO, etc.
  | They are estimated on a discrete k-point grid and not necessarily exact.
  | For converged numbers, create a DOS and/or band structure plot on a denser k-grid.

  Highest occupied state (VBM) at    -12.43420491 eV (relative to internal zero)
  | Occupation number:      2.00000000
  | K-point:       1 at    0.000000    0.000000    0.000000 (in units of recip. lattice)

  Lowest unoccupied state (CBM) at    -5.89118204 eV (relative to internal zero)
  | Occupation number:      0.00000000
  | K-point:       1 at    0.000000    0.000000    0.000000 (in units of recip. lattice)

  ESTIMATED overall HOMO-LUMO gap:      6.54302287 eV between HOMO at k-point 1 and LUMO at k-point 1
  | This appears to be a direct band gap.
  The gap value is above 0.2 eV. Unless you are using a very sparse k-point grid,
  this system is most likely an insulator or a semiconductor.
  Calculating total energy contributions from superposition of free atom densities.

  Total energy components:
  | Sum of eigenvalues            :        -151.86548755 Ha       -4132.47017409 eV
  | XC energy correction          :         -24.31313237 Ha        -661.59399343 eV
  | XC potential correction       :          31.66909171 Ha         861.75983125 eV
  | Free-atom electrostatic energy:        -130.74002132 Ha       -3557.61698976 eV
  | Hartree energy correction     :          -0.00000000 Ha          -0.00000000 eV
  | Entropy correction            :          -0.00000000 Ha          -0.00000000 eV
  | ---------------------------
  | Total energy                  :        -275.24954954 Ha       -7489.92132603 eV
  | Total energy, T -> 0          :        -275.24954954 Ha       -7489.92132603 eV  <-- do not rely on this value for anything but (periodic) metals
  | Electronic free energy        :        -275.24954954 Ha       -7489.92132603 eV

  Derived energy quantities:
  | Kinetic energy                :         275.29901364 Ha        7491.26731262 eV
  | Electrostatic energy          :        -526.23543081 Ha      -14319.59464522 eV
  | Energy correction for multipole
  | error in Hartree potential    :           0.00000000 Ha           0.00000000 eV
  | Sum of eigenvalues per atom                           :       -2066.23508704 eV
  | Total energy (T->0) per atom                          :       -3744.96066302 eV  <-- do not rely on this value for anything but (periodic) metals
  | Electronic free energy per atom                       :       -3744.96066302 eV
  Initialize hartree_potential_storage
  Max. number of atoms included in rho_multipole:            2

  End scf initialization - timings             :  max(cpu_time)    wall_clock(cpu1)
  | Time for scf. initialization                :        0.296 s           0.359 s
  | Boundary condition initialization           :        0.070 s           0.076 s
  | Integration                                 :        0.097 s           0.107 s
  | Solution of K.-S. eqns.                     :        0.018 s           0.018 s
  | Grid partitioning                           :        0.018 s           0.018 s
  | Preloading free-atom quantities on grid     :        0.135 s           0.148 s
  | Free-atom superposition energy              :        0.035 s           0.046 s
  | Total energy evaluation                     :        0.000 s           0.002 s

  Partial memory accounting:
  | Current value for overall tracked memory usage:
  |   Minimum:        0.182 MB (on task 0)
  |   Maximum:        0.184 MB (on task 1)
  |   Average:        0.183 MB
  | Peak value for overall tracked memory usage:
  |   Minimum:        2.139 MB (on task 0 after allocating min_atom_atom_tab)
  |   Maximum:        2.139 MB (on task 0 after allocating min_atom_atom_tab)
  |   Average:        2.139 MB
  | Largest tracked array allocation so far:
  |   Minimum:        1.396 MB (atom_atom_dist_list on task 0)
  |   Maximum:        1.396 MB (atom_atom_dist_list on task 0)
  |   Average:        1.396 MB
  Note:  These values currently only include a subset of arrays which are explicitly tracked.
  The "true" memory usage will be greater.
------------------------------------------------------------
  Time for density update prior                :  max(cpu_time)    wall_clock(cpu1)
  | self-consistency iterative process          :        0.072 s           0.081 s
------------------------------------------------------------
Convergence:    q app. |  density  | eigen (eV) | Etot (eV) |             . |       CPU time |     Clock time
  SCF    1 :  0.35E-02 |  0.44E+00 |   0.32E+01 |  0.63E+00 |             . |        0.324 s |        0.369 s
  SCF    2 :  0.63E-02 |  0.41E+00 |   0.64E+02 |  0.53E+01 |             . |        0.287 s |        0.327 s
  SCF    3 :  0.57E-02 |  0.34E-01 |  -0.41E+01 | -0.34E-01 |             . |        0.263 s |        0.276 s
  SCF    4 :  0.45E-02 |  0.33E-01 |   0.13E+01 |  0.17E-01 |             . |        0.263 s |        0.305 s
  SCF    5 :  0.39E-02 |  0.20E-01 |   0.23E+01 |  0.11E-01 |             . |        0.260 s |        0.284 s
  SCF    6 :  0.32E-02 |  0.79E-02 |  -0.21E+01 | -0.17E-02 |             . |        0.262 s |        0.292 s
  SCF    7 :  0.29E-02 |  0.97E-02 |   0.77E+00 |  0.24E-02 |             . |        0.280 s |        0.297 s
  SCF    8 :  0.31E-02 |  0.23E-03 |   0.12E+00 | -0.74E-04 |             . |        0.265 s |        0.283 s
  SCF    9 :  0.33E-02 |  0.49E-03 |  -0.84E-01 |  0.55E-04 |             . |        0.268 s |        0.279 s
  SCF   10 :  0.33E-02 |  0.74E-04 |   0.11E-02 |  0.36E-05 |             . |        0.270 s |        0.276 s
  SCF   11 :  0.33E-02 |  0.33E-05 |  -0.11E-02 |  0.47E-06 |             . |        0.218 s |        0.226 s

  Total energy components:
  | Sum of eigenvalues            :        -149.47398716 Ha       -4067.39413740 eV
  | XC energy correction          :         -24.72714077 Ha        -672.85973521 eV
  | XC potential correction       :          32.21199198 Ha         876.53289938 eV
  | Free-atom electrostatic energy:        -130.74002132 Ha       -3557.61698976 eV
  | Hartree energy correction     :          -2.30273417 Ha         -62.66058488 eV
  | Entropy correction            :          -0.00000000 Ha          -0.00000000 eV
  | ---------------------------
  | Total energy                  :        -275.03189144 Ha       -7483.99854786 eV
  | Total energy, T -> 0          :        -275.03189144 Ha       -7483.99854786 eV  <-- do not rely on this value for anything but (periodic) metals
  | Electronic free energy        :        -275.03189144 Ha       -7483.99854786 eV

  Derived energy quantities:
  | Kinetic energy                :         277.71222779 Ha        7556.93421084 eV
  | Electrostatic energy          :        -528.01697846 Ha      -14368.07302350 eV
  | Energy correction for multipole
  | error in Hartree potential    :           0.00400912 Ha           0.10909358 eV
  | Sum of eigenvalues per atom                           :       -2033.69706870 eV
  | Total energy (T->0) per atom                          :       -3741.99927393 eV  <-- do not rely on this value for anything but (periodic) metals
  | Electronic free energy per atom                       :       -3741.99927393 eV
  What follows are estimated values for band gap, HOMO, LUMO, etc.
  | They are estimated on a discrete k-point grid and not necessarily exact.
  | For converged numbers, create a DOS and/or band structure plot on a denser k-grid.

  Highest occupied state (VBM) at     -4.54350493 eV (relative to internal zero)
  | Occupation number:      2.00000000
  | K-point:       1 at    0.000000    0.000000    0.000000 (in units of recip. lattice)

  Lowest unoccupied state (CBM) at    -2.84957554 eV (relative to internal zero)
  | Occupation number:      0.00000000
  | K-point:       1 at    0.000000    0.000000    0.000000 (in units of recip. lattice)

  ESTIMATED overall HOMO-LUMO gap:      1.69392939 eV between HOMO at k-point 1 and LUMO at k-point 1
  | This appears to be a direct band gap.
  The gap value is above 0.2 eV. Unless you are using a very sparse k-point grid,
  this system is most likely an insulator or a semiconductor.

  | Chemical Potential                          :    -4.13539083 eV
  | Note that, for insulating systems, the printed 'chemical potential' value is not uniquely defined.
  | It can be anywhere in the energy gap, as long as it correctly separates occupied and unoccupied states.
  | In systems with a gap, the physically relevant chemical potential is the VBM or HOMO.

  Self-consistency cycle converged.


  Energy and forces in a compact form:
  | Total energy uncorrected      :         -0.748399854786232E+04 eV
  | Total energy corrected        :         -0.748399854786232E+04 eV  <-- do not rely on this value for anything but (periodic) metals
  | Electronic free energy        :         -0.748399854786232E+04 eV

 -------------------------------------------------------------------------------
 |              ENTERING DFPT_DIELECTRIC (PERIODIC CALCULATION)                |
 |                                                                             |
 |  Details on the implementation can be found in the following reference:     |
 |                                                                             |
 |    Honghui Shang, Nathaniel Raimbault, Patrick Rinke,                       |
 |     Matthias Scheffler,  Mariana Rossi and Christian Carbogno,              |
 |    'All-Electron, Real-Space Perturbation Theory for Homogeneous            |
 |    Electric Fields: Theory, Implementation, and Application within dft'     |
 |                                                                             |
 |  Please cite New Journal of Physics, 20(7):073040, 2018                     |
 -------------------------------------------------------------------------------


===========================================================================
  before CPSCF, get Omega_MO = <i(k)|-r|j(k)> in MO basis

==========================================================================
  Integrating momentum_matrix_sparse matrix: batch-based integration.
  Time summed over all CPUs for integration: real work        0.338 s, elapsed        0.417 s
  Integrating momentum_matrix_sparse matrix: batch-based integration.
  Time summed over all CPUs for integration: real work        0.345 s, elapsed        0.440 s
  Integrating momentum_matrix_sparse matrix: batch-based integration.
  Time summed over all CPUs for integration: real work        0.348 s, elapsed        0.398 s


===========================================================================
  CPSCF working for j_coord =    1

==========================================================================


--------------------- CPSCF ---------------------------------------
          Begin CP-self-consistency iteration #    1
...
------------------------------------------------------------
dP_dE (Bohr^3) at every cycle:--->
         (62.704458381858579,4.26886443673872329E-016) (-1.18863252573930822E-014,-1.44937593336985743E-015)  (1.94289029309402395E-013,-1.27131239373201648E-014)
  (-1.14006026841195762E-014,2.84945178701606841E-015)         (62.704460371362202,2.08441602063711008E-015)   (2.62570173936715889E-012,1.48872993768335917E-014)
   (1.99618099827603146E-013,1.59983825851750464E-015)  (2.61556748482050239E-012,-1.59218361357929929E-014)        (62.704460371199445,-7.44850800293927954E-015)
DFPT polarizability (Bohr^3)        xx        yy        zz        xy        xz        yz
  | Polarizability:--->             62.704    62.704    62.704    -0.000     0.000     0.000

DFPT for dielectric_constant:--->  # PARSE DFPT_dielectric_tensor
               7.187592646886194            -0.000000000000001             0.000000000000019
              -0.000000000000001             7.187592843207769             0.000000000000259
               0.000000000000020             0.000000000000258             7.187592843191707

                                               :  max(cpu_time)    wall_clock(cpu1)
  | Time for  dielectric calculation            :        0.000 s           0.001 s
===========================================================================

------------------------------------------------------------------------------
  Final output of selected total energy values:

  The following output summarizes some interesting total energy values
  at the end of a run (AFTER all relaxation, molecular dynamics, etc.).

  | Total energy of the DFT / Hartree-Fock s.c.f. calculation      :          -7483.998547862 eV
  | Final zero-broadening corrected energy (caution - metals only) :          -7483.998547862 eV
  | For reference only, the value of 1 Hartree used in FHI-aims is :             27.211384500 eV
  | For reference only, the overall average (free atom contribution
  | + realspace contribution) of the electrostatic potential after
  | s.c.f. convergence is                                          :            -16.001261159 eV

  Before relying on these values, please be sure to understand exactly which
  total energy value is referred to by a given number. Different objects may
  all carry the same name 'total energy'. Definitions:

  Total energy of the DFT / Hartree-Fock s.c.f. calculation:
  | Note that this energy does not include ANY quantities calculated after the
  | s.c.f. cycle, in particular not ANY RPA, MP2, etc. many-body perturbation terms.

  Final zero-broadening corrected energy:
  | For metallic systems only, a broadening of the occupation numbers at the Fermi
  | level can be extrapolated back to zero broadening by an electron-gas inspired
  | formula. For all systems that are not real metals, this value can be
  | meaningless and should be avoided.

------------------------------------------------------------------------------
  Methods described in the following list of references were used in this FHI-aims run.
  If you publish the results, please make sure to cite these reference if they apply.
  FHI-aims is an academic code, and for our developers (often, Ph.D. students
  and postdocs), scientific credit in the community is essential.
  Thank you for helping us!

  For any use of FHI-aims, please cite:

    Volker Blum, Ralf Gehrke, Felix Hanke, Paula Havu, Ville Havu,
    Xinguo Ren, Karsten Reuter, and Matthias Scheffler
    'Ab initio molecular simulations with numeric atom-centered orbitals'
    Computer Physics Communications 180, 2175-2196 (2009)
    http://doi.org/10.1016/j.cpc.2009.06.022


  The provided symmetry information was generated with SPGlib:

    Atsushi Togo, Yusuke Seto, Dimitar Pashov
    SPGlib 1.7.3 obtained from http://spglib.sourceforge.net
    Copyright (C) 2008 Atsushi Togo


  The DFPT (perturbation by an electric field) part of the code was used.
  Here is the corresponding reference:

    Honghui Shang, Nathaniel Raimbault, Mariana Rossi,
    Christian Carbogno, Patrick Rinke and Matthias Scheffler,
    'All-Electron, Real-Space Perturbation Theory for Homogeneous
    Electric Fields: Theory, Implementation, and Application within dft'
    New Journal of Physics, 20(7):073040, 2018


  The ELSI infrastructure was used in your run to solve the Kohn-Sham electronic structure.
  Please check out http://elsi-interchange.org to learn more.
  If scalability is important for your project, please acknowledge ELSI by citing:

    V. W-z. Yu, F. Corsetti, A. Garcia, W. P. Huhn, M. Jacquelin, W. Jia,
    B. Lange, L. Lin, J. Lu, W. Mi, A. Seifitokaldani, A. Vazquez-Mayagoitia,
    C. Yang, H. Yang, and V. Blum
    'ELSI: A unified software interface for Kohn-Sham electronic structure solvers'
    Computer Physics Communications 222, 267-285 (2018).
    http://doi.org/10.1016/j.cpc.2017.09.007


  For the real-space grid partitioning and parallelization used in this calculation, please cite:

    Ville Havu, Volker Blum, Paula Havu, and Matthias Scheffler,
    'Efficient O(N) integration for all-electron electronic structure calculation'
    'using numerically tabulated basis functions'
    Journal of Computational Physics 228, 8367-8379 (2009).
    http://doi.org/10.1016/j.jcp.2009.08.008

  Of course, there are many other important community references, e.g., those cited in the
  above references. Our list is limited to references that describe implementations in the
  FHI-aims code. The reason is purely practical (length of this list) - please credit others as well.

------------------------------------------------------------
          Leaving FHI-aims.
          Date     :  20220804, Time     :  103756.708

          Computational steps:
          | Number of self-consistency cycles          :           26
          | Number of SCF (re)initializations          :            1
          | Number of coupled perturbed self-consistency cycles          :            0
          | Number of CPSCF (re)initializations          :            0

          Have a nice day.
------------------------------------------------------------