DPLOT in RT-TDDFT


Just Got Here
Can someone help me to with the input to visualise the electron density during rt-tddft run? I used the visualisation block as mentioned in the example of RT-TDDFT with water molecule. A error was displayed that indicated that there was no DPLOT block in the input file. Can someone describe me how to generate the cube files for the same?

Forum Regular
Below I've posted an example input for using the visualization feature of the RT-TDDFT code. A dplot input block is needed, but only a minimal setup is needed since the RT-TDDFT code will take care of naming the input and output files for dplot, so you just need to specify the bounds in each direction along with the number of points, and the keyword gaussian produces cube files. The other important point to notice is that there are some quirks (i.e. bugs) with the RT-TDDFT code when it comes to getting properties at the end of the trajectory. In the case of visualization, you need to specify a propagation time (at least) 2 time-steps longer than you actually want cube files for, and you need to specify tend to be (at least) 1 time-step longer. In the example below, I wanted cube files from time 0 to time 100 so I needed to specify tmax as 100.4 (0.2 time-step) and tend as 100.2. Also note that the code will most likely crash at the end rather than ending normally. While somewhat disconcerting, this isn't actually a problem because the RT-TDDFT propagation was finished as well as all of the cube file generations and the code is just crashing because of another bug related to cleaning up the integrals.



echo

start water

geometry "system" units angstroms nocenter noautoz noautosym
 O     0.00000000    -0.00001441    -0.34824012
H -0.00000000 0.76001092 -0.93285191
H 0.00000000 -0.75999650 -0.93290797
end

set geometry "system"

basis
* library 6-31G
end

dft
xc pbe0
end

task dft energy

dplot
gaussian
limitxyz
-5 5 10
-5 5 10
-5 5 10
end

rt_tddft
 tmax 100.4
dt 0.2

 field "driver"
type gaussian
polarization z
frequency 0.3768 # = 10.25 eV
center 393.3
width 64.8
max 0.0001
end

 visualization
tstart 0.0
tend 100.2
treference 0.0
dplot
end

 excite "system" with "driver"
end
task dft rt_tddft

Just Got Here
Thanks, Sean. That was of great help.

Clicked A Few Times
hi
Dear Sean

would you please advice me how could plot the result data??? (how plot this cubic file)? I try VMD but wasn't suitable!

Forum Regular
Jmol can render cube files http://jmol.sourceforge.net

Clicked A Few Times
dear sean,

I read about the property of generating difference density (cube file) in Dplot e.g:

vectors dft-cation.movec dftneutral.movecs #computes the density difference (cation - neutral)

but how would I be able use this property in every time step of
RT-TDDFT block---> with visualization part?

I mean a way of calulation difference density in every time step but from another fix state out of rttddft block,
sth like " treference 0.0 " but with another vectors.

I appreciate your guiding me regarding this.

Forum Regular
The code is not setup to be able to do this in a simple way. You could potentially accomplish it by first running the propagation according to the template I provided above. This will generate density matrix files for every step you are interested in (along with cube files but you can disregard those if you are not interested in have a density difference with one of the steps in the propagation as a reference). You could then use those density matrix files to run a series of simple dplot calculations where you provide two density matrix files as input using the densmat keyword. To get the density matrix for your reference, assuming your reference is another ground state, you will probably have to run a couple step propagation with the RT module with all of the visualization keywords to generate a density matrix like you did for the full propagation before.


Forum >> NWChem's corner >> Running NWChem