Import Error in transition metal ADC(2) calculation - Help requested

Greetings,
for my master thesis project I’m trying to calculate the excited states of an Iron(IV)oxo complex with ADC(2) but I’m getting the fatal error “Import AO data … FAILED!” after entering the ADCMAN routine. Could anyone help me understand what causes this error and how to fix it? Thanks in advance :slight_smile:

The last couple lines of the output files are:

Import AO data … FAILED!
[ERROR] libvmm::evmm::allocate(size_t), /export/data/drehn/compile_qchem_5.2/qchem/libvmm/libvmm/evmm/evmm.h (411), bad_size
bad_size

Q-Chem fatal error occurred in module /export/data/drehn/compile_qchem_5.2/qchem/adcman/adcman/adcman_main.C, line 46:

libvmm::evmm::allocate(size_t), /export/data/drehn/compile_qchem_5.2/qchem/libvmm/libvmm/evmm/evmm.h (411), bad_size
bad_size

Please submit a crash report at Q-Chem Crash Reporter

The input file is:

$molecule
1 3
26 0.000002 -1.043435 0.328612
7 -1.967125 -0.900545 0.027142
6 -2.845386 -1.843929 0.379337
6 -4.172692 -1.779120 -0.010899
6 -4.588129 -0.715586 -0.799864
6 -3.666730 0.256177 -1.171385
6 -2.357408 0.136783 -0.738029
6 -1.287919 1.171307 -1.013974
7 0.000003 0.432425 -1.210020
6 0.000007 -0.149704 -2.580077
6 1.287922 1.171309 -1.013968
7 -0.000011 0.694941 1.722323
6 -1.199332 1.555496 1.540669
6 -1.239764 2.197378 0.147935
6 -0.000002 3.044229 -0.030257
6 1.239759 2.197377 0.147944
6 1.199314 1.555493 1.540677
6 2.357412 0.136786 -0.738024
6 3.666736 0.256185 -1.171374
6 4.588136 -0.715577 -0.799853
6 4.172698 -1.779116 -0.010894
6 2.845391 -1.843929 0.379336
7 1.967130 -0.900545 0.027143
6 -0.000016 0.196682 3.121777
1 -2.457353 -2.651403 0.983026
1 -4.858794 -2.555055 0.296207
1 -5.615876 -0.643927 -1.128986
1 -3.955889 1.093163 -1.791393
1 -1.535050 1.721662 -1.924353
1 0.000008 0.657926 -3.315823
1 0.877673 -0.772541 -2.714807
1 1.535055 1.721667 -1.924345
1 -1.182405 2.349793 2.293905
1 -2.091691 0.962801 1.722224
8 -0.000000 4.224849 -0.279285
1 1.182383 2.349789 2.293914
1 2.091670 0.962795 1.722238
1 3.955896 1.093175 -1.791377
1 5.615884 -0.643913 -1.128970
1 4.858801 -2.555050 0.296211
1 2.457358 -2.651406 0.983022
1 0.882338 -0.411871 3.291579
1 -0.000017 1.045925 3.810441
1 -0.882373 -0.411868 3.291573
1 -0.877656 -0.772543 -2.714812
8 0.000001 -2.075675 1.584915
1 2.126295 2.825871 0.076491
1 -2.126300 2.825872 0.076473
9 0.000002 -2.353146 -0.970218
$end

$rem
METHOD adc(2)
BASIS General
! MEM_TOTAL 200000
MEM_STATIC 2000
! THREADS 16
CC_SYMMETRY false
EE_STATES 10
ADC_DO_DIIS true
max_scf_cycles 10000
scf_convergence 6
$end

$basis
Fe 0
def2-TZVP


F 0
def2-TZVP


O 0
6-31G(d)


C 0
6-31G(d)


N 0
6-31G(d)


H 0
6-31G(d)


$end

When I run this with your input file, which corresponds to using SCF_ALGORITHM=DIIS (default value), the calculation takes 8050 SCF cycles to converge but then the ADC calculation fails with a memory error. With SCF_ALGORITHM=DIIS_GDM, the SCF converges in 211 cycles but the final SCF energy is ‘nan’. Then the ADC fails with the error that you note, however I don’t trust the information obtained at the SCF level. Have you examined the electronic structure of your SCF solution to see that everything looks okay?