The LMFIT program reads the database created by LMTRANS and performs a fit of Taylor expansions to the values of the elements of the diabatic dipole matrix and polarisability tensor contained in the database using a simple optimiser.

An incomplete fourth-order Taylor series containing linear, quadratic, bilinear, cubic and quartic terms is used to fit to each element.

The group theoretical arguments detailed here are used to determine whether a given parameter of the Taylor expansions is zero by symmetry.

The output of the LMFIT program consists of a single ascii file containing the values of the fitted parameters in a human-readable format. For an input file named file.inp, the output file is named file.lmi.

To run the program type
lmfit input
or
lmfit input.inp
where input (or input.inp) denotes the input file. The input format uses, like all the Quantics package input files, keywords that are for the most part free format and case insensitive. See Quantics input file structure

for further information on the general use of keywords, noting that there are no sections in the VCHAM input files. The input file ends with the keyword end-input


Specifying the Fit to be Made

Fit Specification
Keyword Description
infofile = A The database written by LMTRANS to be used is called A
fit_select = m1,m2,... The optimisation will be of terms defined with respect to the modes m1,m2,...
dipole_fit Flags that the parameters entering into the expansion of the dipole matrix are to be optimised
polar_fit Flags that the parameters entering into the expansion of the polarisability tensor are to be optimised


Specification of symmetry labels

The symmetries of the normal modes are read from the specified LMTRANS database file.

The point group is specified using the syntax

point_group = G

where G is the name of the point group (Cs, C2v, etc). Only Abelian point groups are supported. If the point group of the molecule is non-Abelian, the highest Abelian subgroup should be used.

The symmetries of the states and vectors pointing along the x-, y-, and x-axes have to be given in the input as follows.

(1) State symmetries
The symmetries of the states is specified between the lines

state_symmetry
and
end-state_symmetry

For example, for the C2v point group

state_symmetry
1 A1
2 B1
.
.
.
end-state_symmetry

would specify that the state 1 has A1 symmetry, state two has B1 symmetry, and so on.

(2) Axis symmetries
The symmetries of the axes is specified between the lines

axis_symmetry
and
end-axis_symmetry

For example, for the C2v point group

state_symmetry
x B1
y B2
z A1
end-state_symmetry

would specify that a vector lying along the x-axis generates the B1 representation, and so on.