In Orca we can do the relaxed or unrelaxed scanning of any particular internal coordinates (bonds, angled, dihedral). But in output after the optimization we get the energy of the total system. How can I derive ab initio potential for a specific type of bond/angle/dihedral from an orca output? So that I shall be able to use it for potential fitting case. I have attached an image for reference.
You can use GULP for such purpose (http://gulp.curtin.edu.au/gulp/). Check Help > GULP Examples > Example 22. All you need to do is to write the structure and energy from different points of your potential energy surface (generated by changing one internal coordinate at a time). You can substract the total energy of the optimum structure to the total energy of each point if you want your energy to be 0 at the equilibrium length/angle, but this will have no effect on your fitted potential (except for the energy shift, of course). If you intend to use the fitted potential for an MD simulation, the important thing is not the total energy but the *fluctuations* of the total energy.
If you simply need to represent the potential as harmonic potential, then sobtop (http://sobereva.com/soft/Sobtop/) is very suitable. Boot up the code, load .mol2 file of present system, then enter function 5, and input path of .hess file produced by frequency analysis task of ORCA. After that, if you input e.g. 3,5, then force constant between atoms 3 and 5 will be immediately printed. Sobtop supports different ways to derive force contant based on Hessian, such as modified Seminario method, diagonal Hessian elements.
- Badges
- Science topic
- Similar topics
- Physical Sciences
- Atomic, Molecular and Optical Physics
- Atoms