(HNO3 + graphene) Metadynamics tutorial

67 views
Skip to first unread message

Oliver

unread,
Sep 23, 2022, 4:19:24 AM9/23/22
to cp2k
Hi cp2k users,
I read carefully the (HNO3+graphene) tutorial related to Metadynamics and came through several questions:

1. the CVs are based on the definition (1/N0) Sum_i Sum_j (1-Rij/R0)^n/(1-Rij/R0)^m which represents a coodination number. N0 is not clearly defined in the text: is it a normalisation factor and if yes how is it defined, or is it the total number of atoms which could be involved in the CV definition? 

Moreover, this CV definition is not quite the same as the one reported in litterature where there is no N0 at all. This point needs clarification in relation with cp2k usage and  its reference manual. 

2. from the files included in ../gr_2Hno3/dftb_mtd_3cv/; i tried to apply the graph.popt tool to calculate the curves FES=f(CVs). The abscissa ranges calculated for the three possible FES  are: [-0.596: 1.532], [1.247: 4.315] and  [-0.599: 3.975], but the CVs in the COLVAR file give something very different: [~0.04: ~0.98], [~2.18: ~3.45] and [~0.005 : ~3.52]. I can't understand why the ranges are not the same in both calculations. Moreover, all the CVs should be always positive. Therefore, how the FES abscissa are calculated in graph.popt tool?

3. In the FORCE_EVAL module appears three &EXTERNAL_POTENTIAL submodules with different definitions (one related to x coordinate, one for y coordinate and one for z coordinate). But in the user's manual it is written that the function definition cannot be repeated. Therefore, what is calculated exactly and isn't there a bug somewhere?

Moreover, if i well understand, the external potential is added to the hamiltonian in the form:  Sum_i p_i^2/2m_i+U+V_ext. Therefore, the field potential is modified and so the CVs. The barrier height which appears in the FES of the order of -0.35 Ha is unfortunately biased and cannot be corrected. Am i right? 
Finally, All the FES curves have a U shape with several inflection points (in particular FES=f(CV1) and FES=f(CV3)). Do these special points have a particular meaning from the structural point of view?

I'm looking forward to reading your comments.
Thank you very much for your help.
Best regards
Oliver

Reply all
Reply to author
Forward
0 new messages