ASE Tutorials
Getting Started with DFT Calculations
In the first exercise, we will be studying MXenes and how to determine their lattice constants, then we will be studying the interaction between the MXene and an adsorbate. For Homework 5, everyone will be studying the same system (Ti2C). For the Final Project, you will use the same structure but with different chemical composition (e.g., Mo2N instead of Ti2C).
Contents
A Typical ASE Script
ASE scripts can be run directly in the terminal (in the login node) or submitting to external nodes. Generally, you will be submitting jobs to external nodes and only small scripts will be run on the login node. By default, all output from any submitted script will be written from the directory where the submission command was executed, so make sure you are inside the calculation folder before running the submission command.
There are two files that are necessary to run jobs on the Stampede cluster. The first is spede_esp.sub; this is the file that tells the scheduler how much time the job is allowed, how many processors it requires, and other pertinent information. First, notice the comments in the beginning. These include information such as how much time to allocate, the number of nodes required, what the names of the output and error files are, what the name of the job should be, and what your email is.
#!/bin/bash
#SBATCH -J test # Job Name
#SBATCH -A TG-CHE160084 # Allocation number: Do not change this
#SBATCH -o ll_out # Output file name
#SBATCH -e ll_err # Error file name
#SBATCH -n 16 # Total number of cores requested
#SBATCH -p development # which queue to run in
#SBATCH -t 00:30:00 # Run time (hh:mm:ss)
#SBATCH --mail-user=youremail@whatever.com # Make sure to change this!!!!
#SBATCH --mail-type=end # Emails you at the end of the job
which pw.x
cd $SLURM_SUBMIT_DIR
export TMPDIR=$SLURM_SUBMIT_DIR
echo $EPS_PSP_PATH
export OMP_NUM_THREADS=1
python Energy.py
Finally, the last line python Energy.py picks the script you want to run. Therefore, you need to change the name of the file depending on which script you are running.
Let’s look at how a typical ASE script is written. Open the Energy.py script. We import all the relevant ASE modules in for this calculation
from espresso import espresso
from ase.io import read, write
from espresso import espresso imports the Quantum ESPRESSO calculator for the ASE interface, and from ase.io import read, write imports the read and write commands for trajectory files.
An existing trajectory can be read in:
# read in the slab
slab = read('Ti2C.traj')
Then, the Quantum ESPRESSO calculator is set up. All parameters related to the electronic structure calculation are included here. The following example shows typical parameters that we use in the group for MXene calculations.
calc = espresso(pw=700, #plane-wave cutoff
dw=7000, #density cutoff
xc='BEEF-vdW', #exchange-correlation functional
kpts=(8,8,1), #k-point sampling;
nbands=-20, #20 extra bands besides the bands needed to hold
#the valence electrons
sigma=0.1,
convergence= {'energy':1e-6,
'mixing':0.1,
'nmix':10,
'mix':4,
'maxsteps':500,
'diag':'david'
}, #convergence parameters
output={'removesave':True},
dipole={'status':False}, #dipole correction to account for periodicity in z
spinpol=False,
outdir='calcdir') #output directory for Quantum Espresso files
Finally, the Quantum ESPRESSO calculator is attached to the slab Atoms object, the energy calculation is ran, and the total energy of the system is output in the log file (defined in the spede_esp.sub file above).
To submit the job, use:
sbatch -J $PWD spede_esp.sub
The -J $PWD gives the name of the job as the current directory. Make sure this calculations runs correctly before proceeding. You should get a total energy of -3353.960 eV.
Lattice Constant Determination
Find the Lattice_Constant.py script in the lattice folder. This script calculates the different energies of the system as a function of the lattice constant. Before you run this job, make sure you read the comments within to understand what it does. You will later need to modify this file for the Final Project.
Remember to change the script name to Lattice_Constant.py in the spede_esp.sub file! Submit the script by running:
sbatch --job-name=$PWD spede_esp.sub
Here, --job-name=$PWD sets the current working directory as the job name.
The output states the energy with respect to the given lattice constant. Take this data and plot it however you choose. Fit ~5 points near the minimum of this function with a quadratic function. Then, use calculus to find the minimum energy, and thus the DFT lattice constant.
HW 5: Plot the energies as listed above, and report the DFT lattice constant.
The two-dimensional bulk modulus B describes the compressibility of a two-dimensional sheet (how difficult it is to stretch or compress). Take the lattice script given before, change the given value to the DFT lattice constant, and change the strain value to run from 0.98 to 1.02, with five steps of 0.1. Fit the energies with a quadratic function. Then, calculuate B via:
where S is the surface area of the sheet (a variable) and S0 is the true surface area. Note that in this case we are fitting the surface area S, not the lattice constant! The surface area of the MXenes is given by:
HW 5: Report the two-dimensional bulk modulus of Ti2C.
Convergence with k-Points
Next, we will determine how well-converged the total energy is with respect to the number of k-points in each direction. Modify the Lattice_Constant.py script using the lattice parameter obtained from the previous section. Instead of looping over strain values as above, modify the script to keep the same lattice constant and loop over k-points instead. Try using k = (2,2,1), (4,4,1), (6,6,1), (8,8,1), and (10,10,1), and plot the energy as a function of k-points. Pick one and try to justify why it would be a reasonable choice. The relevant k-points will usually be known, since we have consistent settings that we use throughout the group. In principle, one should always check for convergence when working with a new system. (Side note: Think about why the last k-point is always 1).
HW 5: Show the k-point convergence plot, your pick for the k-points, and your rationale.
For the Final Project, we will be using (8x8x1) k-points for the (1x1) MXene surfaces, and (4x4x1) k-points for the (2x2) MXene surfaces.
We have also provided the Lattice_Resize.py script that reads in a .traj file, and changes the lattice to the correct version resized with the lattice constant provided. In the script, you need to change the lattice constant you want manually. Unlike the rest of the scripts given to you, this script can be run directly from the command line, without using the submission system, by using:
python Lattice_Resize.py
Next: move on to Adsorption to learn about how to add adsorbates on your surface.