It is mentioned in the turbomol manual that, to get accurate b2-plyp single point enrgy, a rimp2 run must be performed after the dft step.
So, first, I performed a ridft single point energy calculation with the functional b2-plyp and also the keyword $disp in the control file(because I want to take care of the dispersion correlation). Then, I modified the control file with define. I kept the previous inputs in the control file intact and just change the mp2 parameters for running the mp2 single point energy calculation.
But, the problem arises here. The program stopped with the rimp2.out file:
------------------------------------------------------
calculation of (ti|Q)(Q|P)**(-1/2)
------------------------------------------------------
total memory allocated for calculation of (ti|Q)(Q|P)**(-1/2) : 644 MByte
record length scale factor = 8
starting loop 1 of direct evaluation of B(ti|P)
starting loop 2 of direct evaluation of B(ti|P)
total lp3mp2 cpu-time : 31.07 seconds
total lp3mp2 wall-time : 31.07 seconds
total trdrei cpu-time : 2 minutes and 6 seconds
total trdrei wall-time : 2 minutes and 6 seconds
total trifro cpu-time : 22.04 seconds
total mkbaoi cpu-time : 2 minutes and 38 seconds
total mkbaoi wall-time : 2 minutes and 38 seconds
-----------------------------------------------------
n**5-steps following now
-----------------------------------------------------
memory allocated for pij: 9410 real(
words
energy threshold for MO-block-neighbourships: 0.10E-02
total memory allocated for N**5 steps : 3916 MiB
this includes: 145924 real(
words for pab
31786220 real(
words for bai
31786220 real(
words for yai
7733972 real(
words for riajb
409900516 real(
words for tiajb
31786220 real(
words for bbj
1 real(
words for bgi
20246 real(
words for symscr
479 real(
words for symin1
Allocation failure for galoop in rimp2
Here is the restart.rimp2 file:
restart after mkbaoi
What is the problem here? Please give your suggestion.
Regards.