Restart the nonadiabatic dynamics

[John Zhang]

Dear all

I tried to restart the nonadiabatic dynamics calculation. Following the manual's instructions, I replaced control.dyn by the one in INFO_RESTART, and ran the calculation as usual. The  dynamics runs but with a wrong nondiabatic coupling term. The coupling term at the beginning of the restart run is zero, which is different from the value at the last stopped trajectory. The computed energy and gradient at these two points (first point in the restarting run and last point in the stopped trajectory) are the same, and the time is also correct. I used Gaussian TDDFT with the OD approximation for the NAC calculation. 

Could anyone provide some suggestion to fix this problem? Thanks!

Best wishes

Peng