Topmod

This is a program for generating an ELF (Electron localization function] from a WFN file, and analysing this function for synaptic basins, together with the total integrated electron density bounded by those basins.

The program source code is available. It should be unzipped into a folder. CD into the folder. Using a text editor, ensure that the file Makefile has the appropriate path to the install directory (by default /usr/local/bin ). The Fortran compiler defined is g77, which should be already installed on your system. Invoke the following

make clean (removes any old object files, and ensure that all the source code is compiled)
make (compiles and links the code)
make install (does what it says on the tin). These three operations should not take more than 1 minute.

The Makefile has preset -O3 on the Fortran compiler. This does seem to produce code which runs about 25% faster than with -O2 (using the g77 Fortran compiler. For installation of this compiler, see here.

To run the program (assuming it has been installed in a directory with a pre-defined path to it, ie /usr/local/bin), simply type

The wfn file. Make sure the top line of this file is edited to a sensible string. The resulting sbf files will then be prefixed with this string, and hence avoid potential overwriting of an earlier file.
grid09 (which takes a WFN file produced using the Gaussian keyword output=wfn) and outputs temporary files. This can take a few minutes, depending on the resolution of the requested grid. Do not ask for more than about 150 intervals in each of the directions, or everything will take a long time
bas09 (which defines ELF and RHO basins). this step can take hours, depending on the size of the molecule. Run this program twice, once for ELF and once for RHO (the latter only if you want to integrate the basins).
1. This will write a file elf09_bas.gjf which contains the coordinate of the basins. The file needs a little editing to make it viewable using eg Gaussview. At this point, you need to decide if you are going to integrate all the basins for the electron density, or just a select few. If you have more than about 150 basins, you should probably opt for selecting the ones you want. In Gaussview, if you view labels, you can get the index number of your desired basins.
pop09 (which takes these basins and integrates them for electron density). This can take hours, depending on the size of the molecule.
1. The program will prompt for the number of ELF and AIM basins to be integrated. 0 0 integrates them all. For large molecules, 8 1 will integrate 8 ELF basins. The index numbers of these (obtained from Gaussview as above) are then typed. The minimum no of AIM basins is 1. For a large molecule (>150 atoms) it make take an hour per basin. The values of the integrations will be appended to elf09_bas.gjf
sbf_to_cube is useful to convert any of the three sbf files (rho, lap or elf) to a Gaussian cube file.This can the be converted to a .jvxl compressed file using this utility

Manual available.