Electron Beam Driven Discrete Dipole Approximation

The following code is a modified verison of Draine's Discrete Dipole Approximation (DDA) code version 7.1. We have modified the code to to allow for an electron beam excitation source. The following code computes the electron energy loss spectroscopy (EELS) and angle-resolved cathodoluminescence (CL) for both aloof and internal geometries.

Instructions

General

• Delete all executables within source_code by typeing "make clean; make veryclean"
• Compile the fortran code my typing "make all"
• Running this code works nearly the same as DDA 7.1, however the input file, "ddscat.par" now has two new lines indicating the electron beam position and velocity
• See the folder "useful_scripts" in order to run spectra and spectrum images on the cluster.

Useful Scripts

Make spectrum images

This folder will launch many scattering calculations to form a 2-D spectrum image by raster scanning the electron beam.

• Replace shape.dat with your shape file. You can use the given shape.f90 file as a template to make your shape if you'd like.
• Update ddscat.par as you would for a normal e-dda calculation, except DO NOT change the electron beam position. It should always read: " 0.0 0.0 0.0 " exactly.
• Launch an interactive node ( type in command line: srun -p build --time=2:00:00 --mem=100G --pty /bin/bash )
• Define the extent of the window and the raster step size in create_folders.sh. These are set at the end of Line 4. extent extends the raster window beyond the shape points in each direction in the given number of dipole spacings. raster_ss sets the y and z raster step size.
• Once both numbers are updated, run bash creat_folders.sh.
• This will create the listed number of folders; one folder for each e-beam raster point calculation.
• Next, make the launch scripts to batch launch all the points. To do so, type in the command line: module load anaconda3_5.3; python makelaunchfiles.py
• This will create launch files where each job submits 20 EEL calculations. Feel free to change the batch size (Line 6) if the calculation is too big to complete within the set 4 hour time window.
• Submit all the jobs by typing in command line: bash submit_jobs.sh
• Lastly, kill your build node job.
• Once all the points have finished, run python collect_EEL.py. This will create spectrum_image.txt.

Citations

If you use e-DDA to compute EELS, we request you cite: https://doi.org/10.1021/nn302980u.

If you use e-DDA to compute CL, we request you additionally cite https://doi.org/10.1021/nn401161n.