Generate UV-Vis absorption spectra from TDHF/TDDFT data in ADF, GAMESS, Gaussian, and Jaguar output files.
The discrete spectrum obtained from an electronic structure TDHF/TDDFT calculation is not the most intuitive method for visualizing a simulated UV-Vis absorption spectrum. Experimentally, such spectra have broad peaks and are described by their line shape. This program extracts the excited state energies and oscillator strengths from the output files of TDHF/TDDFT calculations performed by several widely-used electronic structure programs and generates a line shape function by summing together Gaussian functions fit to each peak.
This package includes the uvspecgen script for quickly processing
electronic structure TDHF/TDDFT output files to produce a file containing
the discrete spectrum and line shape function for plotting. This script uses
the included uvspec module, which can be imported into your own Python
scripts for use of the AbsorptionSpectrum class, which parses a TDHF/TDDFT
output file and stores the excited state energies, oscillator strengths, and
line shape function as attributes of the class.
This program uses the cclib Python library (ref. 1) for parsing and
interpreting the results of computational chemistry packages. It currently
supports parsing the results of TDHF/TDDFT calculations for the following
electronic structure programs
- ADF
- GAMESS
- Gaussian03
- Gaussian09
- Jaguar
The uvspecgen script requires the argparse and configparser modules,
which are installed using the setup.py script as described below. For
plotting spectra, the matplotlib package is required. For information on
how to install matplotlib, visit http://www.matplotlib.org.
The uvspec module uses the cclib parsing library to extract the
TDHF/TDDFT excited state energies and oscillator strengths, which is installed
when using the setup.py script as described below. The cclib requires
the numpy package, which is not installed using the setup.py script.
For information on how to install numpy, visit http://www.numpy.org.
Installations are best performed using the setuptools Python package via
the included setup.py file. To learn more about custom installations, visit
the Installing Python Modules documentation. Standard installations are
described below.
Install as Root
If you have root permissions on the target system, run the following commands at the command prompt:
tar xzvf uvspec-$VERSION.tar.gz cd uvspec-$VERSION sudo python setup.py install
Install a Local Copy
For users that do not have root privileges, a local installation can be performed. The installation location is system-specific, but can quickly determined by running:
./setup.py install --help
and look for the text accompanying the --user option. More information on
local installations can be found at
http://docs.python.org/2/install/#alternate-installation-the-user-scheme.
The installation should then be performed as follows:
tar xzf uvspec-$VERSION.tar.gz cd uvspec-$VERSION python setup.py install --user
The source files uvspec-$VERSION/ and uvspec-$VERSION.tar.gz can be
deleted after installation.
The AbsorptionSpectrum class and methods can be used in your own scripts.
An example of its usage is provided below:
from uvspec.spectrum import AbsorptionSpectrum
AS = AbsorptionSpectrum()
AS.excited_state_energy = [1,2,3]
AS.oscillator_strength = [0.1,0.2,0.3]
AS.generate()
AS.write('api-test')
AS.plot()
Rather than manually assigning the attributes excited_state_energy and
oscillator_strength, you can call the extract(logfile) method with a
logfile name provided as a string. This will extract the excited state energies
and oscillator strengths into these attributes automatically. The plot()
method only needs to be called if you want to generate a matplotlib plot of
the line shape function.
- N. M. O'Boyle, A. L. Tenderholt, K. M. Langner, cclib: a library for package-independent computational chemistry algorithms, J. Comp. Chem. 29 (5), pp. 839-845, 2008.