Distribute Absorption Spectrum
This workflow computes the absorption spectra for a given molecular system and returns a set of files in TXT format. The principle of distribution workflow is dividing the work in multiple, separated, instances (chunks), in order to be able to split time-consuming jobs into smaller, quicker ones.
In this tutorial, we want to compute the excited states at each point of a pre-computed MD trajectory for the guanine system. Please note that this trajectory has already been used in the advanced example of the absorption_spectrum tutorial, where the spectrum analysis was performed on 1 out of 4 points only. Here we take advantage of the distribution workflow to increase by four times the accuracy of sampling with no substantial variation of the computational cost in terms of time by dividing the job in five chunks (each taking charge of 4 points out of 20).
Preparing the input
The input is described in YAML format as showed in the following example:
workflow:
distribute_absorption_spectrum
project_name: guanine_distribution
active_space: [20, 20]
dt: 1
path_hdf5: "guanine.hdf5"
path_traj_xyz: "guanine_twentyPoints.xyz"
scratch_path: "/tmp/distribute_absorption_spectrum"
calculate_guesses: "first"
xc_dft: pbe
tddft: stda
stride: 1
blocks: 5
workdir: "."
job_scheduler:
free_format: "
#! /bin/bash\n
#SBATCH --job-name=guanine_distribution\n
#SBATCH -N 1\n
#SBATCH -t 1:00:00\n
#SBATCH -p short\n
source activate qmflows\n
module load CP2K/7.1.0\n\n"
cp2k_general_settings:
basis: "DZVP-MOLOPT-SR-GTH"
potential: "GTH-PBE"
cell_parameters: 25.0
periodic: none
executable: cp2k.popt
cp2k_settings_main:
specific:
template: pbe_main
cp2k_settings_guess:
specific:
template: pbe_guess
In your working directory, create an input_test_distribute_absorption_spectrum.yml file and copy the previous input inside it (remember to respect the indentation). Also copy locally the small pre-computed MD trajectory of the guanine system, guanine_twentyPoints.xyz.
In the input file, pay particular attention to the following parameters that have been added/modified with respect to the previous tutorial (advanced example):
stride: Controls the accuracy of sampling of geometries contained in the MD trajectory of reference. Here, a value of stride: 1 indicates that the spectrum analysis will be performed on each point in the reference trajectory. Two important things have to be pointed out:
The workflow will perform SCF calculations for each point in the trajectory; only afterwards it will sample the number of structures on which the spectrum analysis will be performed
Down-sampling issues might arise from the number of points that are actually printed during the MD calculations. Some programs, indeed, offer the possibility to print (in the output file) only one point out of ten (or more) calculated. For example, applying a stride: 10 would in practice mean that you are sampling 1 point out of 100 points in the trajectory.
blocks: Indicates into how many blocks has the job to be split. This will generate as many chunks’ folders in your working directory.
workdir: Path to the chunks’ folders.
The job_scheduler can also be found below these parameters. Customize these settings according to the system and environment you are using to perform the calculations.
Setting up the calculation
Once all settings in input_test_distribute_absorption_spectrum.yml have been customized, you will need to create the different chunks.
First, activate QMFlows:
conda activate qmflowsUse the command distribute_jobs.py to create the different chunks:
distribute_jobs.py -i input_test_distribute_absorption_spectrum.yml
A number of new folders are created. In each folder you will find a submission file, launch.sh, a sub-trajectory file (containing the points assigned to that chunk), chunk_xyz, and an input.yml file. In the input.yml file, you will find all your settings. Check for any possible manual errors.
If you are satisfied with the inputs, submit each of your jobs for calculation.
You can keep track of the calculations by going to your scratch path. The location where all points of the chunks are calculated is your assigned scratch path plus project name plus a number.
Results
Once the calculations are completed, you will find multiple output_n_stda.txt files in your scratch directories (with n being the index of the geometry at which the spectrum analysis has been performed). The first two lines of the file output_0_stda.txt, found in /tmp/distribute_absorption_spectrum/scratch_chunk_0/ are reported below.
# state energy f t_dip_x t_dip_y t_dip_y weight from energy to energy delta_E
1 4.566 0.03832 -0.51792 -0.25870 0.08573 0.50158 20 -5.175 21 -1.261 3.914
For each excited state (line), the first six columns contain, from left to right:
# state: Assigned index, in ascending order of energy. Here, the lowest excitation is reported and corresponds to # state 1.
energy: Transition energy, in eV.
f: Oscillator strength, dimensionless.
t_dip_x, t_dip_y, t_dip_z: Transition dipole moment components along x, y and z.
The next six columns report some useful information about the dominant single orbital transition for the excited state under examination:
weight: Weight in the overall transition. Always 1.0000 in the Single Orbital approximation.
from: Index of the initial occupied orbital in the active space.
energy: Energy of the initial occupied orbital.
to: Index of the final virtual orbital in the active space.
energy: Energy of the final virtual orbital.
delta_E:Energy of the dominant single orbital transition. Corresponds to the excited state energy in the Single Orbital approximation.
Copy all the output files to your working directory and plot the absorption spectrum (averaged over all sampled structures) using the script convolution.py:
convolution.py -nm True
To plot the absorption spectrum of a specific sample, for example our point 0, use the -n option.
convolution.py -n 0 -nm True
Reporting a bug or requesting a feature
To report an issue or request a new feature you can use the github issues tracker.