In the general input section of the MOLECULE input file, we will consider such information as molecular symmetry , number of symmetry distinct atoms , generators of a given molecular point group , and so on. This information usually constitutes the four/five first lines of the input, and is formatted.
The input is best described by an example.
The following is the first lines of an input for
tetrahedrane , treated in
symmetry, with a 4-31G** basis. The line numbers are for
convenience in the subsequent input description and should
not appear in the actual input. Note also that in order to fit
the example across the page some liberties have been taken with
column spacings.
1:INTGRL 2: Tetrahedrane, Td_symmetric geometry 3: 4-31G** basis 4: 2 2 X Y Z 1.00D-15
We now define the input line-by-line. The FORMAT is given in parentheses.
INTGRL
(A6).
CRT,NONTYP,KCHARG,SYMTXT,((KASYM(I,J),I=1,3),J=1,3), ID3, THRS
(A1,I4,I3,A2,10A1,D10.2).
Note that if one wants to use the basis set library , there are two options. One option is to use a common basis set for the entire molecule in which the first line should be replaced by two lines, which for a calculation using the 4-31G** basis would look like:
1:BASIS 2:4-31G**This option will not be active with customizable basis sets like the ANO or NQvD sets.
Alternatively you may specify different basis sets for different atoms, in which case the first line should read
1:ATOMBASIS
The fourth line (fifth in a calculation using the basis set library) looks a bit devastating. However, for ordinary Hartree-Fock or MP2 calculations, only the number of different atom types and the charge need to be given (if the molecule is charged), as symmetry and Hartree-Fock occupation will be taken care of by the program. Thus this line could in the above example be reduced to
4: 2
Let us finally give some remarks about the symmetry detection routines. These routines will detect any symmetry of a molecule by explicit testing for the occurrence of rotation axes, mirror planes and center of inversion. The occurrence of a symmetry element is tested in the program against a threshold which may be adjusted by the keyword .SYMTHR in the *READIN input section. By default, the program will require geometries that are correct to the sixth decimal place in order to detect all symmetry elements.
The program will translate and rotate the molecule into a suitable
reference geometry before testing for the occurrence of symmetry
operations. The program will not, due to the handling of symmetry in the
program, transform the molecule back to original input
coordinates. Furthermore, if there are
symmetry equivalent nuclei, these will be removed from the input, and a
new, standardized molecule input file will be generated and used in
subsequent iterations of for instance a geometry optimization. This
standarized input file (including basis set) can be printed to the
file DALTON.BAS
by using the keyword .MOLPRI
in the
*READIN
submodule.
DALTON can only take advantage of point groups that are subgroups
of D . If symmetry higher than that is detected, the program
will use the highest common subgroup of the symmetry group
detected and D
.
We recommend that the automatic symmetry detection feature is not used when doing MCSCF calculations, as symmetry generators and their order in the input determines the order of the irreproducible representations needed when specifying active spaces. Thus, for MCSCF calculations we recommend that the symmetry is explicitly specified through the appropriate symmetry generators, as well as the explicit Hartree-Fock occupation numbers.