next up previous contents
Next: File IO Up: BIOTR Previous: Introduction   Contents

Program Structure

Data is initialized in initr() and factr(), then the user selects the initial and final state, type of calculation (relativistic, or non-relativistic) and the type of transition E1, E2, M1, M2, .. or *'. The program then proceeds with the following steps:

  1. cfgin2() reads the two sets of configuration lists and determines the orthogonality conditions between the two sets. Next, rasin() checks if the CI expansions used for $\Psi_1$ and $\Psi_2$ satisfy the closure under de-excitation property.

  2. Calculate the CSF-coupling coefficient ${A^{\mu\nu}_{ij}}$ in the expression: ${a^{\dagger}_{i}\hat{a}_{j}\mid\Phi_{nu}\rangle =
\sum_{\nu}{A^{\mu\nu}_{ij}}\mid\Phi_{mu}}$, which are needed to perform the transformation of the CI-coefficients.

  3. Calculate the one-electron orbital overlap matrix and decompose it into block-triangular factors to find the new radial functions ${\{R^A_{\it nl}(r)\}}$ and ${\{R^B_{\it nl}(r)\}}$ which are biorthonormal.

  4. transform the configuration interaction matrix ${C^A_{\mu}}$ and ${C^B_{\mu}}$ by a sequence of single orbital replacements. main routine for employing biorthogonal rotations for RAS type wave functions, allowing the calculation of transition moments between two RAS states. The task of this part of the code is to change two sets of orbitals into biorthogonal orbitals and counter-rotate the CI coefficients. transform the configuration interaction matrix ${C^A_{\mu}}$ and ${C^B_{\mu}}$ by a sequence of single orbital replacements.

  5. Apply the orthogonal Racah algebra to transform the many electron amplitude into a sum of one-electron reduced matrix elements. the l.h.s and r.h.s orbital indices of the se now refer to the two different orbital biorthonormal sets.

Figure 6.35: Program structure.
\begin{figure}\begin{center}
\centerline{\psfig{figure=tex/fig/biotr_main.epsi}}\end{center}\end{figure}

The user is expected to provide information about which two sets of input files should be used and the type of calculation (E1, E2,..., M1, M2, ...). Note in the input below that E1 and O1 supplied by the user for Name of Initial State and Name of Final State, imply the existence of: E1.c E1.w E1.j and O1.c O1.w O1.j, if a relativistic calculation is being selected:

#  ........A parity changing transition calculation........
>biotr 
  Name of Initial State
E1
  Name of Final State
O1
  intermediate printing (y or n) ?
n 
  Relativistic calculation ? (y/n)
y
  Type of transition ? (E1, E2, M1, M2, .. or *)
E1
------------------------------------------------------------------------

The angular data is not saved and therefore, for calculations along an iso-electronic sequence, the angular calculations are repeated for each atom.


next up previous contents
Next: File IO Up: BIOTR Previous: Introduction   Contents
2001-10-11