Skip to content


Subversion checkout URL

You can clone with
Download ZIP
Tools to build and diagonalize many body Hamiltonian for actinide systems.
Branch: master
Pull request Compare This branch is even with witzclassic:master.

Fetching latest commit…

Cannot retrieve the latest commit at this time

Failed to load latest commit information.
scripts Initial commit to git.


This repository contains three subdirectories:
1. pythonSource contains all of the python code necessary to build a many body Hamiltonian derived from the application Density Functional Theory to actinide molecular complexes, An(H_2O)_8 and AnO_2(H_2O)_5 where An can be the elements U, Np, Pu and Am.  Although care was taken to make the code as independent as possible, some dependencies exist with respect to these molecules (see e.g.) and to using ADF's output format in their TAPE15 and TAPE21 files.
2. scripts contains python and bash scripts used to help automate the building process.
3. SparseMatrix contains the project code to build the utility that diagonalizes the many body Hamiltonian. 

Note, scripts may generate a lot of output to the screen.  It is good practice to pipe the output to a log file.

1. Create working directory

2. In working directory:
   2a. Link or copy ADF TAPE21 and TAPE15 files to <file>.t21 and <file>.t15, respectively. 
   2b. Copy script ... edit parameters in script and execute, output is Ham.plist.  An should equal the actinide atom name used by ADF, withO is true for AnO2 molecule, false otherwise, withD should be false (is a hook for future work to include 6d orbitals), and soCC should be 3 (for f orbitals), followed by the spin-orbit coupling constant and energy shift.
   2c. Execute, output is Nks.pks, Uks.pks, Occ.pkl and Hdft.pkl.
   2d. Execute script
   2e. Last thing printed by is Nks = <integer>.<fraction>, create subdirectories n<integer> and n<integer+1> (actual subdir name is not important).

3. In working subdirectory:
   3a. Execute script, output Ham.plist, Hopping.pkl, Uo.pkl, Hks.pks, Occ.pkl links.
   3b. Create state tables: edit and execute python script with proper n representing the number of particles in the many body system (see suggestion for subdir names above).  Output mbDimension.pkl, nParticles.pkl, stateTable.pkl and stateTableDict.pkl.  These take a lot of space; they can be soft linked from a central directory.  
   3c. Execute script ... alter F2, F4 and F6 as needed.
   3d. Execute script

4. Diagonalize ... run sparseMatrix ... execute script or modify as needed.

5. Output of sparseMatrix can be analyzed ... execute script

Something went wrong with that request. Please try again.