why the electron difference density is not consistant with previous study

[njuxyh]

Dear All:

i consists a MoS2-graphene hetero structures using the Interface tool in the Builder option.The parameters chosen are shown in figure 1. the relaxed geometry is a=6.321,b=6.321,c=24.804. the distance between Gr and Mos2 is 3.34 angstrom.
the relaxed configuration  py file are attached.
and we noticed that in our configurations, one C atoms of Gr is located uopn the S atom in mos2. so we though it is very close to TS configuration presents in the Nanoscale, 2011, 3, 3883–3887.

but we plot the electron difference density, and found it is NOT consistent with that in eNanoscale, 2011, 3, 3883–3887. the different details are show in figure 2 and figure 3

so i am wondering what is the reason make the electron difference density different, especially the C layer lost electrons while in our calculation C  get electrons.
 
i need your help.

 
thanks a lot

[Petr Khomyakov]

Your supercell looks much smaller than that corresponding to the TS configuration. In addition, you have to adopt similar computational settings (density functional, k-point sampling) to do relaxation as done in that paper; actually, you have not posted the entire script with calculator and optimization sections included. 

But it seems that the main issue now is the cell size. In the Interface Builder, you may go through different interface structures found by the Interface Builder to see which one matches the one in the paper.  You may also adjust search settings if you do not find the right cell, using the default settings.