Electric Field for Molecular Dynamic Calculation

[Joseph Page]

I am looking into a time-dependent electric field that affects the molecular dynamics of molecules. I have been using the &EFIELD section with the &RUN_TYPE being MD. 

Is this correct as the &EFIELD section of the documentation suggests that it can be used for RPT calculations (not MD)? (Basically, can I run an MD calc with the &EFIELD section and it be what I expect)

Thank you,

Joe

P.S. I have seen people suggest using the &PERIODIC_EFIELD section in similar questions, but I don't think that's what I need as I need my field to be Gaussian in the future.

[Matt Watkins]

Hi, no I'm afraid that only works with real time propagation TDDFT.

To approximately get a time dependent field I maybe suggest a normal MD run with the  PERIODIC_EFIELD. Restart at short intervals modulating field as you want. You can't expect a very rapid efield to be reasonably modelled in any case. 

This is something that maybe should be made available via a python or other scripting interface...()

Matt

[Joseph Page]

Hi Matt,

Thank you.

I do rather like your suggestion as to how it could work with periodic field, so I am going to be giving that a go, so thank you for that as well.

Joe

[Evelyn Navarro]

Hi, I have a similar question. I want to set up a constant electric field in MD and then study the effect of the electric field on the structure. I used the &EFIELD part, the aperiodic boundary. I wonder if this is the right thing to do.
thank you.

[André Borrfors]

Hi,

Have you looked in to using %RUN_TYPE EHRENFEST_DYN?

This utilizes real_time_propagation within DFT to propagate the electronic response to an electric field specified in the input, and the nuclear geometry is updated within the ehrenfest scheme.

Note that this is very expensive for most systems and will require fine tuning of the relevant input parameters and access to heavy computing power if more than one or two molecules are to be propagated.

Best,

André