Starter Kit for Modeling a GaAs-Si interface with Quantum Espresso

Starter Kit for Modeling a GaAs-Si interface

with Quantum Espresso

A.Martí and I. Artacho

Instituto de Energía Solar – Universidad Politécnica de Madrid, Spain

The motivation behind this upload to Zenodo is to provide the interested visitor with an initial set of files (“starter kit”) so they can model, visualize and perhaps, even carry out further research on the physical and electronic properties of GaAs/Si interfaces. Behind this bundle is the idea of gaining progressive insight into the potential growth of III-Vs on silicon by techniques such as molecular beam epitaxy (MBE): how atoms rearrange, how defects originate, etc. To this end we provide:

• The necessary input files to be executed in Quantum Espresso (see Notes below). These are listed in Table I (see pdf file included), showing also how they can be executed together with some comments and illustrative results. For the calculations we have used 12 cores (indicated by the use of “mpirun -np 12”). This can be omitted for single core calculations. In our case, using a Lenovo ThinkPad 21KVS0DM00 Laptop, calculation times have been around 2 days for the examples provided.
• The Blender file (GaAs_on_Si.blend  within the folder blender) that allows visualizing in 3D the atomic positions used in the calculations (Fig. 3 in pdf file included) as well as automatically creating a video to visualize the relaxation of the atoms from their initial positions to their final positions. By using this file, the reader can visualize the relaxation from different visual perspectives on their own. Several examples of the videos that can be created are included in this bundle as examples: bottom.mkv, lateral.mkv and front.mkv. The Blender file includes a script in Python where the atomic positions should be inserted in order to automatically generate new videos. For inserting the initial atomic positions, please search into the script for the block commented as : # Parameters: Initial positions of the atoms are inserted here; These positions are inserted in frame 1 of the video. For inserting the final positions, please search for the lines commented as # Insert here the final atomic positions obtained after relaxation. These positions will be inserted in frame 250 and Blender will interpolate in-between. Once the Blender composition is created after running the script, the reader is assumed to have some knowledge about Blender to create their own videos with the perspective they wish. It is also possible to change the color, size of the atoms, etc. We recommend this is done in the python script provided within Blender.  If you create your own video, note you might need to change the location where the created video file is stored (in our case, in a tmp folder). Use EEVEE (instead of Cycles) for faster rendering.

NOTES:

• Quantum Espresso [1], [2] is an open-source suite of codes for electronic-structure calculations and materials modeling at the nanoscale, based on density functional theory, plane waves, and pseudopotentials. It can be downloaded from www.quantum-espresso.org together with the instructions for installing it. Quantum Espresso can be installed both in Windows as in Linux. The example we provide has been tested on Quantum Espresso 7.3.1 installed under Ubuntu 22.04.5 LTS and tested in a Lenovo ThinkPad 21KVS0DM00 Laptop.
• The input files provided contain several instructions, but we make no aim in this document to explain what each instruction does since the reader can easily find this information in the Quantum Espresso manual or, nowadays, they can simply ask AI tools such as ChatGPT  (https://openai.com/es-ES/chatgpt/overview/) to obtain a very good summary of what these instructions do adapted to their needs. ChatGPT has assisted us in creating the script for the Blender file and the input files for Quantum Espresso. However, we have not used ChatGPT to review this text.
• The output files have not been included to save space. They will be created again once you run the input files.
• The unit supercell we use, extends 50 A in the z direction to simulate only one mixed layer of GaAs and Si atoms:

CELL_PARAMETERS angstrom

5.43  0.00   0.00

0.00  5.43   0.00

0.00  0.00   50.00

• The value of 5.43 A corresponds to the lattice constant of silicon since in our simulation we wanted to put the Ga and As atoms as if initially they were also located preserving this lattice constant. GaAs has a larger lattice constant (5.65 A)  and, in fact, the simulation (best appreciated in the videos) reveals an expansion of the position of the atoms as expected from the difference in the value between the two lattice constants.
•   The pseudopotentials for As, Ga and Si have been downloaded from http://www.quantum-simulation.org/potentials/sg15_oncv/upf/  . They are included within the folder “pseudo”.
• The crystal structure of Si and GaAs have been obtained from Wikipedia.
• For plotting the actual band diagrams, we shall assume “gnuplot” (www.gnuplot.info/) is  installed in your system.
• Blender can be downloaded from https://www.blender.org/download/
• Files provided must be understood, as mentioned, as starter kits to initiate research on the electronic properties of the GaAs/Si interface using Quantum Espresso. No rigorous physical significance of the results here presented is claimed at this stage. Results are only illustrative. We think further work should continue by, for example: a) enlarging the supercell; b) carrying out cell relaxation (using vc-relax); c) exploring other crystal orientations.
• Some times, the following error is generated: “The following floating-point exceptions are signaling: IEEE_DENORMAL”. This error is considered non-critical and has been ignored.

References:

[1]        P. Giannozzi et al., “QUANTUM ESPRESSO: a modular and open-source software project for quantum simulations of materials,” J. Phys. Condens. Matter, vol. 21, no. 39, p. 395502, 2009, doi: 10.1088/0953-8984/21/39/395502.

[2]        P. Giannozzi et al., “Advanced capabilities for materials modelling with Quantum ESPRESSO,” J. Phys. Condens. Matter, vol. 29, no. 46, p. 465901, 2017, doi: 10.1088/1361-648X/aa8f79.

Acknowledgments:

 This action has been funded from grant PID2021-124193OB-C21 (PVBooster Project) funded by Spanish Ministerio de Ciencia e Innovación with European Funds.

This action has been financed through the R+D activities program with reference TEC-2024/ECO-72 and acronym 4EVERPV-CM granted by the Community of Madrid through the General Directorate of Research and Technological Innovation through Order 3177/2024