Excited-state populations- NAMD

[Valentina Andrea Rovasio]

Hi everyone,

I'm taking my first steps in non-adiabatic molecular dynamics simulations using NWChem. My main objective is to model electron-phonon interactions in semiconductor quantum dots, but to better understand both the theory and the software, I’m currently working with small molecules.

I'm stuck on one point: I’d like to analyze excited-state populations over a swarm of trajectories, similar to what's done in the paper "First Principles Nonadiabatic Excited-State Molecular Dynamics in NWChem" (J. Chem. Theory Comput. 2020, 16, 6418−6427), specifically in Figure 2.

I was able to reproduce an example trajectory showing internal conversion between states, but to study excited-state populations: should I run multiple trajectories? Or am I perhaps missing some output data in the files?

I've included my input file below. Any help would be greatly appreciated , thank you in advance!
Valentina 

-----------------------------
title "Benceno QMD-NAMD"
echo
scratch_dir /scratch
permanent_dir ./perm

start benzene
echo

geometry noautosym
 C  -1.299627   -0.506491   -0.000000
 C  -0.211073   -1.378740    0.000107
 C   1.088436   -0.872256   -0.000082
 C   1.299585    0.506588    0.000017
 C   0.211180    1.378720    0.000078
 C  -1.088501    0.872180   -0.000080
 H  -2.317618   -0.903354   -0.000069
 H  -0.376599   -2.458749    0.000060
 H   1.941158   -1.555376   -0.000184
 H   2.317669    0.903193   -0.000009
 H   0.376456    2.458759    0.000036
 H  -1.941066    1.555512   -0.000073
end

basis
 * library 6-31G
end

dft
 xc b3lyp
end

tddft
 nroots 10
 notriplet
 cis
 civecs
 grad
  root 1
 end
end

qmd
 nstep_nucl 5000
 dt_nucl 0.5
 targ_temp 300.0
 thermostat svr 500
 namd SH
   nstates 5    
   init_state 3    
   dt_elec 0.1
   deco .true.
 end
end

task tddft qmd

[Niri Govind]

Hi Valentina,

>> I was able to reproduce an example trajectory showing internal conversion between states, but to study excited-state populations: should I run multiple trajectories? Or am I perhaps missing some output data in the files?

There are a few steps to get the excited-state populations. 

1. You need to run multiple trajectories.

2. For each trajectory, grep out "Current State" from the output for each time step.

3. For each timestep, you need to go across each trajectory and bin the state for each time step. This will give you the average over trajectories.

Hope this helps.

Best,

-Niri

Niri Govind

PNNL

--
You received this message because you are subscribed to the Google Groups "NWChem Forum" group.
To unsubscribe from this group and stop receiving emails from it, send an email to nwchem-forum...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/nwchem-forum/241929f1-3ade-4d83-9492-efdbc0b577c0n%40googlegroups.com.

[Valentina Andrea Rovasio]

Hi Niri! 
Thank you for your reply, it is very helpful.
I would like to take this opportunity to ask you another question: did you study how many trajectories simulate in order to obtain an accurate result?
I understand that it should depend on the system and what you are trying to simulate. In our case, we want to study the interaction of electrons with vibrational modes in inorganic qdots. The electronic transitions we are interested in are those with the lowest energy. 
Thank you in advance.
Best,
Valentina 

To view this discussion visit https://groups.google.com/d/msgid/nwchem-forum/CADCCgPvKXq2qZoWRnBGMfmwXbrtzHQG7BLER1qfuCtfQGN01-Q%40mail.gmail.com.

[Niri Govind]

Hi Valentina,

>> did you study how many trajectories simulate in order to obtain an accurate result?
>> I understand that it should depend on the system and what you are trying to simulate.

Typically, at least 100 trajectories for ~10 ps for each trajectory. But this also depends on the cost of the calculations.

>>In our case, we want to study the interaction of electrons with vibrational modes in inorganic qdots. 

This could get expensive depending on the size of the qdots. An alternate strategy would be to simulate the vibrational modes on the ground state first with a frequency calculation. Then sample geometries along the vibrational modes you are interested in and explore how these modes affect the ground and excited state electronic structure by performing DFT and TDDFT calculations on these sampled structures.

Hope this helps.

Best,

-Niri

Niri Govind

PNNL

To view this discussion visit https://groups.google.com/d/msgid/nwchem-forum/CAEv9K3g5GCJxnKqFOGLqsRn9zQY34Q4pshSVnU%2BZhh8U3k8PQQ%40mail.gmail.com.

[Valentina Andrea Rovasio]

Hi Niri,
Perfect! Thanks for your reply.
In principle,  we’ll be working with a small qdot of 30 atoms, but to scale up to larger (and real) systems, I will use the alternative you suggested.  

Going back to the topic of trajectories, and after reading the NWchem bibliography, I was wondering:  is the input I shared above enough to map multiple trajectories?  I can see that the code assigns different initial velocities to each trajectory, but I'm not sure if that assignment is enough to create the set of initial conditions or if I will be mapping very similar structures, thereby losing information. Another alternative that comes to mind is to generate multiple initial conditions with different structures (using Newton-X, for example) and then running NAMD with NWchem.  

Thanks in advance!
Best,
Valentina