Geometry optimization of benzene with PM6

[Qiang Zhu]

Hi, everyone. I'm a new user of CP2K. Just met some problem in doing
relaxation on Benzene with PM6 method in CP2K.
Here is the input. (4 benzene molecules's cell)

&GLOBAL
  PROJECT USPEX
  RUN_TYPE GEO_OPT
&END GLOBAL

&MOTION
  &GEO_OPT
    MINIMIZER CG
    MAX_ITER 100
    MAX_FORCE 5.00000000E-02
  &END GEO_OPT
   &PRINT
     &CELL
       &EACH
          GEO_OPT 1
       &END EACH
     &END CELL
   &END PRINT
&END MOTION

&FORCE_EVAL
  METHOD Quickstep
  &DFT
    &QS
      METHOD PM6
      &SE
       ANALYTICAL_GRADIENTS F
      &END
    &END QS
    &SCF
      MAX_SCF 50
      SCF_GUESS ATOMIC
    &END SCF
 &END DFT
STRESS_TENSOR DIAGONAL_NUMERICAL
&SUBSYS
&CELL
ABC [angstrom] 9.7348 7.2468 6.8024
ALPHA_BETA_GAMMA [deg] 90.0568 89.9078 97.0491
&END
&COORD

SCALED
C 0.445 0.082632 0.411988
C 0.587051 0.061656 0.40965
C 0.63521 -0.084611 0.515571
C 0.54101 -0.209069 0.623282
C 0.399219 -0.186529 0.62442
C 0.35245 -0.039658 0.517829
C 0.13395 0.917039 0.019528
C 0.040751 0.791438 -0.088236
C -0.10124 0.812788 -0.090298
C -0.14924 0.95955 0.01558
C -0.054938 1.0842 0.122859
C 0.08679 1.0611 0.123569
C 0.58689 0.560767 0.12428
C 0.44518 0.583808 0.123389
C 0.35095 0.459568 0.015697
C 0.398909 0.312946 -0.090473
C 0.540899 0.291515 -0.088337
C 0.634049 0.416756 0.019965
C 0.085989 0.562436 0.40981
C 0.1347 0.416236 0.515129
C 0.040969 0.291097 0.6228
C -0.10092 0.312878 0.624511
C -0.14828 0.459718 0.518551
C -0.053453 0.583887 0.411501
H 0.406731 0.197748 0.328657
H 0.660681 0.160268 0.3244
H 0.74706 -0.10152 0.514012
H 0.578619 -0.324468 0.706913
H 0.324739 -0.284358 0.70914
H 0.2408 -0.021515 0.518428
H 0.2459 0.900619 0.021414
H 0.079333 0.676167 -0.171368
H -0.17493 0.714077 -0.175088
H -0.26104 0.97669 0.014273
H -0.092423 1.1999 0.206388
H 0.16133 1.1591 0.207968
H 0.66141 0.658438 0.20899
H 0.40811 0.69961 0.207389
H 0.23921 0.477299 0.014687
H 0.324909 0.214806 -0.17545
H 0.579069 0.176204 -0.171828
H 0.745979 0.400225 0.02195
H 0.15957 0.661275 0.324839
H 0.24667 0.399896 0.513098
H 0.079096 0.175668 0.70593
H -0.17503 0.214519 0.709192
H -0.26004 0.477279 0.519622
H -0.090445 0.699717 0.327871
&END
&END SUBSYS
&END FORCE_EVAL

However, the calculation failed to end within several hours. Is it
normal for such size at the level of semi empirical calculation? Also
I noticed that the SCF run could not converge even I put 50 on
MAX_SCF.

Could anyone help me?

Best,
Qiang Zhu