Wrong basis set assigned in Gateway/Seward
Dear OpenMolcas team, the following input:
&GATEWAY
coord
3
Coords in angstrom
C 0.000000 0.000000 0.000000
F 1.379013 -1.379013 -1.379013
O 0. 0. 1.
Basis set
C.6-31G
Basis set
F.6-31G
Basis set
O.6-31G
produces the following output:
()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()
&GATEWAY
only a single process is used
available to each process: 500 MB of memory, 1 thread?
pid: 1500540
()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()()
Basis Set 1 Label: C.6-31G
Basis set is read from library:basis_library
Basis Set Reference(s):
W.J. Hehre, R. Ditchfield, J.A. Pople. J. Chem. Phys. 56 (1972) 2257-2261. doi:10.1063/1.1677527
CARBON (10s,4p) -> [3s,2p]
Basis Set 2 Label: F.ANO-S-MB
Basis set is read from library:basis_library
Basis Set Reference(s):
K. Pierloot, B. Dumez, P.-O. Widmark, B.O. Roos. Theor. Chim. Acta 90 (1995) 87-114. doi:10.1007/bf01113842
FLUORINE (10s,6p,3d) -> [7s,6p,3d]
Basis Set 3 Label: O.6-31G
Basis set is read from library:basis_library
Basis Set Reference(s):
W.J. Hehre, R. Ditchfield, J.A. Pople. J. Chem. Phys. 56 (1972) 2257-2261. doi:10.1063/1.1677527
OXYGEN (10s,4p) -> [3s,2p]
where the second element (F) is erroneously assigned the ANO-S-MB basis set. When increasing the print level, one can confirm that actually the exponents and coefficients of ANO-S-MB are loaded. When requesting ANO-R for all elements, then ANO-S-MB is still loaded for F, but X2C and finite nuclei are used for F.
Patrick Zobel and me checked that actually the correct basis set library is loaded and that the basis.tbl file is correct. The issue does not seem to be related to any particular basis set file, because it occurs with different basis sets used in the input file. Rather, it seems that the "default" basis for F is loaded because of some parsing error. If the order of elements is altered, the issue seems to affect always the second element.
The following equivalent input produces the expected result:
&GATEWAY
Basis set
C.6-31G
C 0.000000 0.000000 0.000000
end of basis
Basis set
F.6-31G
F 1.379013 -1.379013 -1.379013
end of basis
Basis set
O.6-31G
O 0. 0. 1.
end of basis
end of input
This issue appeared in OpenMolcas V21.10.
Best regards, Sebastian Mai