Hi every one

I model a collision between two pyrene molecules in Forcite.And I am interested in energy transferring process of this collision.What I did was

(1)doing a geometry optimization for two molecules in the same .xsd file

(2)heating them at 1600k by a NVT dynamics with Berendsen

(3)making they equilibrium with a NVE dynamics(I am sure that they have reached equilibration state,and this is also confirmed by Simon https://community.accelrys.com/message/22321#22321)

(4)making them move toward each other in NVE by script made by Renier https://community.accelrys.com/message/22117#22117 ,which will result a collision between two molecules

(5)calculating the bond energy of the dimer(I think there is a Van der Waals force when they are close to each other) used the script made by Stephen

As far as I know,bonding energy can be calculated by E=E(close)-E(molecule1)-E(molecule2),if the energy is negative which represents they are bonded.

but the value of the bonding energy calculated  by the script is positive or negative alternative.It is can be understood if this situation occurs when they are close to each other(the negative indicates bonded,while the positive suggests separation),because the Van der Waals is formed and broken alternative.However,this situation lasts over the whole process(except the early frames),even when they are far away,which is impossible in my opinion.I know I should find out the reason myself.But I am total new for MD simulation,I can not figure it out although I  tried my best to do so.

I post necessary files here.

Any help will be appreciated.

Sincerely

Li