When I try to do ASCI calculations for larger active spaces
the program stops with: Q-Chem cannot handle an active space of this size.
Could someone tell me how this estimated by the program?
Using the formula for the number of configurations based on n electrons (nel) and n orbitals (n orb), the program stops even when the number of configurations drops. Is there a fixed limit for nel,norb?strong text
Hello Peter,
Could you provide the output file and send it to support@q-chem.com? If the system is confidential, you may build a smallest structure that can reproduce the error.
Thanks
Hello. Running into the same issue. Here’s my input file:
$comment
Plain old SCF on the ground state. Only activate
stability analysis for the known trouble-makers - Fe(CO)2.
$end
$molecule
0 1
Fe
C 1 rFeC_Short
C 1 rFeC_Short 2 a_CC_Short
C 1 rFeC_Long 3 a_CC_LS 2 d_CL_Fe_CS_CS
C 1 rFeC_Long 2 a_CC_LS 3 d_CL_Fe_CS_CS
O 1 rFeO_Short 2 a_CO_Short 4 d_OS_Fe_CS_CL
O 1 rFeO_Short 3 a_CO_Short 5 d_OS_Fe_CS_CL
O 1 rFeO_Long 4 a_CO_Long 2 d_OL_Fe_CL_CS
O 1 rFeO_Long 5 a_CO_Long 3 d_OL_Fe_CL_CS
rFeC_Short = 1.79664
rFeO_Short = 2.94322
rFeC_Long = 1.85939
rFeO_Long = 3.00332
a_CC_Short = 122.56975
a_CC_LS = 93.69502
a_CO_Short = 119.45518
a_CO_Long = 163.25567
d_CL_Fe_CS_CS = 96.77000
d_OL_Fe_CL_CS = 118.49600
d_OS_Fe_CS_CL = 96.77000
$end
$rem
METHOD HF
UNRESTRICTED FALSE
BASIS GEN
PURECART 11
SYMMETRY FALSE
SYM_IGNORE TRUE
GEN_SCFMAN TRUE
SCF_ALGORITHM DIIS_GDM
SCF_CONVERGENCE 9
MAX_SCF_CYCLES 200
!INTERNAL_STABILITY TRUE
!INTERNAL_STABILITY_ITER 10
!INTERNAL_STABILITY_DAVIDSON_ITER 250
MEM_TOTAL 128000
MEM_STATIC 1000
THRESH 14
$end
$basis
C 0
cc-pVDZ
****
O 0
cc-pVDZ
****
Fe 0
cc-pVDZ
****
$end
@@@
$comment
Several CASSCF calculations with increasing number of
NDETS for the ASCI solver... No need to spin-purify as the states
states are nearly spin-pure anyways. Similarly, avoid PT2 correction
for now. These two will be done in the final CASCI step.
$end
$molecule
0 1
Fe
C 1 rFeC_Short
C 1 rFeC_Short 2 a_CC_Short
C 1 rFeC_Long 3 a_CC_LS 2 d_CL_Fe_CS_CS
C 1 rFeC_Long 2 a_CC_LS 3 d_CL_Fe_CS_CS
O 1 rFeO_Short 2 a_CO_Short 4 d_OS_Fe_CS_CL
O 1 rFeO_Short 3 a_CO_Short 5 d_OS_Fe_CS_CL
O 1 rFeO_Long 4 a_CO_Long 2 d_OL_Fe_CL_CS
O 1 rFeO_Long 5 a_CO_Long 3 d_OL_Fe_CL_CS
rFeC_Short = 1.79664
rFeO_Short = 2.94322
rFeC_Long = 1.85939
rFeO_Long = 3.00332
a_CC_Short = 122.56975
a_CC_LS = 93.69502
a_CO_Short = 119.45518
a_CO_Long = 163.25567
d_CL_Fe_CS_CS = 96.77000
d_OL_Fe_CL_CS = 118.49600
d_OS_Fe_CS_CL = 96.77000
$end
$rem
METHOD HF
UNRESTRICTED FALSE
BASIS GEN
PURECART 11
SYMMETRY FALSE
SYM_IGNORE TRUE
GEN_SCFMAN TRUE
SCF_GUESS READ
SCF_ALGORITHM GDM
SCF_CONVERGENCE 8
MAX_SCF_CYCLES 0
MAX_CASSCF_CYCLES 2000
CAS_METHOD 2 ! 1: CASCI, 2: CASSCF
CAS_SOLVER 2 ! 2: ASCI
CAS_M_S 0
CAS_N_ELEC 32
CAS_N_ORB 38
CAS_N_ROOTS 5
CAS_SAVE_NAT_ORBS TRUE
ASCI_SPIN_PURIFY FALSE
ASCI_NDETS 50000
ASCI_SKIP_PT2 TRUE
MEM_TOTAL 128000
MEM_STATIC 500
THRESH 14
$end
$basis
C 0
cc-pVDZ
****
O 0
cc-pVDZ
****
Fe 0
cc-pVDZ
****
$end
@@@
$comment
Several CASSCF calculations with increasing number of
NDETS for the ASCI solver... No need to spin-purify as the states
states are nearly spin-pure anyways. Similarly, avoid PT2 correction
for now. These two will be done in the final CASCI step.
$end
$molecule
0 1
Fe
C 1 rFeC_Short
C 1 rFeC_Short 2 a_CC_Short
C 1 rFeC_Long 3 a_CC_LS 2 d_CL_Fe_CS_CS
C 1 rFeC_Long 2 a_CC_LS 3 d_CL_Fe_CS_CS
O 1 rFeO_Short 2 a_CO_Short 4 d_OS_Fe_CS_CL
O 1 rFeO_Short 3 a_CO_Short 5 d_OS_Fe_CS_CL
O 1 rFeO_Long 4 a_CO_Long 2 d_OL_Fe_CL_CS
O 1 rFeO_Long 5 a_CO_Long 3 d_OL_Fe_CL_CS
rFeC_Short = 1.79664
rFeO_Short = 2.94322
rFeC_Long = 1.85939
rFeO_Long = 3.00332
a_CC_Short = 122.56975
a_CC_LS = 93.69502
a_CO_Short = 119.45518
a_CO_Long = 163.25567
d_CL_Fe_CS_CS = 96.77000
d_OL_Fe_CL_CS = 118.49600
d_OS_Fe_CS_CL = 96.77000
$end
$rem
METHOD HF
UNRESTRICTED FALSE
BASIS GEN
PURECART 11
SYMMETRY FALSE
SYM_IGNORE TRUE
GEN_SCFMAN TRUE
SCF_GUESS READ
SCF_ALGORITHM GDM
SCF_CONVERGENCE 8
MAX_SCF_CYCLES 0
MAX_CASSCF_CYCLES 2000
CAS_METHOD 2 ! 1: CASCI, 2: CASSCF
CAS_SOLVER 2 ! 2: ASCI
CAS_M_S 0
CAS_N_ELEC 32
CAS_N_ORB 38
CAS_N_ROOTS 5
CAS_SAVE_NAT_ORBS TRUE
ASCI_SPIN_PURIFY FALSE
ASCI_NDETS 100000
ASCI_SKIP_PT2 TRUE
MEM_TOTAL 128000
MEM_STATIC 500
THRESH 14
$end
$basis
C 0
cc-pVDZ
****
O 0
cc-pVDZ
****
Fe 0
cc-pVDZ
****
$end
@@@
$comment
The final CASCI step with spin purification and PT2
correction, using the largest number of NDETS for ASCI. We use
the refined CASSCF orbitals with smaller NDETS.
$end
$molecule
0 1
Fe
C 1 rFeC_Short
C 1 rFeC_Short 2 a_CC_Short
C 1 rFeC_Long 3 a_CC_LS 2 d_CL_Fe_CS_CS
C 1 rFeC_Long 2 a_CC_LS 3 d_CL_Fe_CS_CS
O 1 rFeO_Short 2 a_CO_Short 4 d_OS_Fe_CS_CL
O 1 rFeO_Short 3 a_CO_Short 5 d_OS_Fe_CS_CL
O 1 rFeO_Long 4 a_CO_Long 2 d_OL_Fe_CL_CS
O 1 rFeO_Long 5 a_CO_Long 3 d_OL_Fe_CL_CS
rFeC_Short = 1.79664
rFeO_Short = 2.94322
rFeC_Long = 1.85939
rFeO_Long = 3.00332
a_CC_Short = 122.56975
a_CC_LS = 93.69502
a_CO_Short = 119.45518
a_CO_Long = 163.25567
d_CL_Fe_CS_CS = 96.77000
d_OL_Fe_CL_CS = 118.49600
d_OS_Fe_CS_CL = 96.77000
$end
$rem
METHOD HF
UNRESTRICTED FALSE
BASIS GEN
PURECART 11
SYMMETRY FALSE
SYM_IGNORE TRUE
GEN_SCFMAN TRUE
SCF_GUESS READ
MAX_SCF_CYCLES 0
CAS_METHOD 1 ! 1: CASCI, 2: CASSCF
CAS_SOLVER 2 ! 2: ASCI
CAS_M_S 0
CAS_N_ELEC 32
CAS_N_ORB 38
CAS_N_ROOTS 5
ASCI_SPIN_PURIFY TRUE
ASCI_NDETS 200000
ASCI_SKIP_PT2 FALSE
MEM_TOTAL 128000
MEM_STATIC 500
THRESH 14
$end
$basis
C 0
cc-pVDZ
****
O 0
cc-pVDZ
****
Fe 0
cc-pVDZ
****
$end
Terminates in the first CASSCF job.
-----------------------------------------------------------------------
General SCF calculation program by
Eric Jon Sundstrom, Paul Horn, Yuezhi Mao, Dmitri Zuev, Alec White,
David Stuck, Shaama M.S., Shane Yost, Joonho Lee, David Small,
Daniel Levine, Susi Lehtola, Hugh Burton, Evgeny Epifanovsky,
Bang C. Huynh
-----------------------------------------------------------------------
Complete Active-Space and Truncated CI program by
Daniel Levine, Alec White, Susi Lehtola
-----------------------------------------------------------------------
Using mrci option set 2
Option set summary: CASSCF.
Reading orbital from previous CASSCF step
Orbitals are orthogonal
using current memeber C for guess
Orthonormalizing orbitals
Skip SCF calculation as requested
<S^2> = 0.000000000
The disked orbital energies are calculated using alpha and beta Focks
Building Core-Active Integrals...done (47.000 seconds)
Building Active-Active Integrals...done (109.000 seconds)
-----------------------------------------------------------------------
Adaptive Sampling Configuration Interaction (ASCISCF) program by
Daniel Levine, Norm Tubman, Diptarka Hait, Susi Lehtola
-----------------------------------------------------------------------
Q-Chem fatal error occurred in module libalmo/libalmo/mrci/asci_wf.C, line 638:
Q-Chem cannot handle an active space of this size
Please submit a crash report at q-chem.com/reporter