How to Calculate Adsorption Energy using Quantum ESPRESSO and DFT? [TUTORIAL]

Summary

In this Phys Whiz tutorial, Manas Sharma demonstrates how to calculate the adsorption energy of a water molecule on the LiH (001) surface using Density Functional Theory (DFT) with Quantum ESPRESSO. The process involves three separate DFT calculations for the total system, the isolated molecule, and the surface. Sharma guides viewers through the use of BURAI, a GUI for Quantum ESPRESSO, to set up and run simulations, aiming to reproduce a study's adsorption energy value. The tutorial concludes with a comparison of the calculated energy to a reference value, highlighting the importance of pseudopotentials and energy cutoffs in achieving accurate results.

Takeaways

• 👨‍🏫 The tutorial is presented by Manas Sharma from Phys Whiz, focusing on calculating the adsorption energy of a molecule on a surface using density functional theory (DFT) and Quantum ESPRESSO.
• 💧 The specific molecule and surface discussed are H2O (water) and the LiH (001) surface, respectively.
• 📚 Adsorption energy is calculated using a formula that involves the total energy of the system and the energies of the isolated components.
• 📉 A negative adsorption energy indicates favorable adsorption, although some literature uses a reverse convention with a positive value indicating favorability.
• 📘 The tutorial references a 2017 paper and uses its supplementary material for the H2O-LiH system structure.
• 🧩 The structure provided includes various supercell sizes, but the tutorial uses the one with 16 Li and 16 H atoms, totaling 32 atoms in the surface.
• 📝 The structure is converted into a CIF file for use in Quantum ESPRESSO, as opposed to the original VASP format (POSCAR file).
• 🔬 The DFT calculation uses the PBE exchange-correlation functional aiming to reproduce the adsorption energy value reported in the paper.
• 🔄 Three separate calculations are needed: one for the total periodic system, one for the isolated H2O molecule, and one for the isolated LiH surface.
• 🛠️ BURAI, a GUI for Quantum ESPRESSO, is used to run the simulations and manage the project.
• 🔧 Pseudopotentials are carefully selected for each atom to ensure consistency and accuracy in the simulations.
• ⚙️ Energy cutoffs and charge cutoffs are set, with occupations fixed due to the likely semiconductor or insulator nature of the materials.
• 🔢 The tutorial demonstrates how to extract and calculate the final adsorption energy, comparing it to the value obtained in the referenced paper.
• 🔗 Links to the paper, simulation files, and a web app for unit conversion are provided in the description for further reference.

Q & A

What is the main topic of the tutorial presented by Manas Sharma in the Phys Whiz video?

The main topic of the tutorial is calculating the adsorption energy of a molecule, specifically H2O on a LiH (001) surface, using density functional theory and Quantum ESPRESSO.

What is the general formula for calculating adsorption energy as described in the script?

The general formula for calculating adsorption energy is the total DFT energy of the system minus the sum of the energies of the isolated molecule and the surface or slab.

What does a negative adsorption energy value indicate according to the convention used in the script?

A negative adsorption energy value indicates that the adsorption is favorable, according to the convention used in the script.

What is the significance of the paper from 2017 mentioned in the script?

The 2017 paper is significant because it provides the method and formula used for calculating the adsorption energy, and the script aims to reproduce the results from this paper using Quantum ESPRESSO.

How many atoms are in the surface structure used for the tutorial, and what is the total number of atoms including the H2O molecule?

The surface structure used in the tutorial contains 32 atoms (16 Li and 16 H atoms), and including the H2O molecule, the total number of atoms is 35.

What is the purpose of using the same pseudopotentials for all atoms in the calculations?

Using the same pseudopotentials for all atoms ensures consistency and accuracy in the calculations, preventing discrepancies that could arise from using different types of pseudopotentials.

Why is the energy cutoff for the wave function set to 50 Rydbergs and the charge cutoff to 500 Rydbergs in the calculations?

The energy cutoff is set to 50 Rydbergs and the charge cutoff to 500 Rydbergs to ensure a balance between accuracy and computational speed, as a smaller cutoff speeds up simulations.

What is the role of BURAI in the calculations described in the script?

BURAI is a GUI for Quantum ESPRESSO, and it is used to set up, run, and manage the SCF calculations for the total system, the isolated H2O molecule, and the isolated LiH surface.

What is the importance of ensuring that the water molecule does not interact with its neighboring periodic images in the isolated H2O calculation?

Ensuring that the water molecule does not interact with its neighboring periodic images is important to accurately calculate the energy of an isolated molecule, avoiding any artificial interactions that could skew the results.

How does the script address the difference in conventions between a positive and a negative adsorption energy value?

The script acknowledges that some papers use a reverse convention where a positive adsorption energy value indicates favorable adsorption. It clarifies that the results should be compared based on the magnitude, regardless of the sign.

What is the final calculated adsorption energy value in milli electron volts, and how does it compare to the value from the 2017 paper?

The final calculated adsorption energy value is -223 milli electron volts, which is close to the value of 219 milli electron volts obtained in the 2017 paper, indicating good agreement despite slight differences.

What could be the reasons for the slight difference between the calculated adsorption energy and the value from the 2017 paper?

The slight difference could be due to different choices of pseudopotentials or the energy cutoff not being fully converged, suggesting that a larger or smaller value might have been used in the paper.

Outlines

00:00

🔬 Calculating Adsorption Energy with DFT and Quantum Espresso

In this tutorial, Manas Sharma introduces viewers to the process of calculating the adsorption energy of a molecule on a surface using Density Functional Theory (DFT) and the Quantum Espresso software. The specific example given is the adsorption of a water molecule (H2O) on the LiH (001) surface. The adsorption energy is determined by calculating the total energy of the system and subtracting the energies of the isolated components. The tutorial emphasizes the importance of understanding different conventions for calculating adsorption energy, where a negative value can indicate favorable adsorption in one convention, while a positive value does in another. The structure of the H2O-LiH system is sourced from a 2017 paper, and the tutorial aims to reproduce the adsorption energy value reported there using the PBE exchange-correlation functional.

05:05

🛠 Preparing Quantum Espresso Simulations with BURAI GUI

The second part of the tutorial focuses on setting up simulations using the BURAI graphical user interface (GUI) for Quantum Espresso. The process involves assigning pseudopotentials to each atom in the system, ensuring consistency and choosing pseudopotentials with a smaller energy cutoff to speed up simulations. The tutorial explains the importance of selecting the appropriate energy and charge cutoffs and using fixed occupations for the material, which is likely a semiconductor or insulator. The calculations for the total system, the isolated LiH surface, and the isolated H2O molecule are prepared, each with specific settings for wave function and charge cutoffs, and occupations. The tutorial also details the steps to run the self-consistent field (SCF) calculations for each system.

10:06

📊 Analyzing Results and Comparing with Literature Values

In the final segment, the tutorial demonstrates how to analyze the results from the Quantum Espresso simulations. The energies obtained from the calculations for the total system, the isolated LiH surface, and the H2O molecule are extracted and used to calculate the adsorption energy. The process includes using an online calculator to convert the final adsorption energy from Rydbergs to electron volts. The calculated adsorption energy is then compared with the value reported in the literature, noting that the results are in good agreement despite minor differences that could be attributed to different pseudopotential choices or energy cutoff values. The tutorial concludes by summarizing the steps taken to calculate the adsorption energy and encourages viewers to like, subscribe, and comment with any questions or doubts.

Keywords

💡Adsorption Energy

Adsorption energy refers to the energy change when a molecule adheres to a surface. In the context of the video, it is calculated for a water molecule (H2O) on a LiH (001) surface using density functional theory (DFT). The script explains that a negative adsorption energy indicates favorable adsorption, aligning with the convention used in the referenced paper.

💡Density Functional Theory (DFT)

Density functional theory is a computational quantum mechanical modeling method used in physics, chemistry, and materials science to investigate the electronic structure of many-body systems. In the video, DFT is employed to calculate the adsorption energy of a water molecule on a surface, which is a central theme of the tutorial.

💡Quantum ESPRESSO

Quantum ESPRESSO is an open-source suite of computer codes for quantum mechanics, molecular dynamics, plane waves, and more. The script details a tutorial on using Quantum ESPRESSO to calculate the adsorption energy, showcasing its application in simulating the interaction between a molecule and a surface.

💡Pseudopotentials

Pseudopotentials are a method used in quantum chemistry and solid-state physics to simplify the calculation of the electronic structure of a system by replacing the core electrons with an effective potential. The video script discusses the importance of choosing consistent pseudopotentials for accurate simulation results.

💡SCF Calculation

SCF stands for Self-Consistent Field, a type of calculation used in quantum chemistry to determine the lowest energy state of a system. The script mentions running three SCF calculations for the total system, the isolated molecule, and the isolated surface to determine the adsorption energy.

💡BURAI

BURAI is a graphical user interface (GUI) for Quantum ESPRESSO, which simplifies the process of setting up and running simulations. The script uses BURAI to demonstrate how to perform the necessary calculations for determining adsorption energy.

💡CIF File

A CIF (Crystallographic Information File) is a text file format used by crystallographers to store and exchange data. In the script, the structure of the H2O-LiH system is converted into a CIF file for use in the Quantum ESPRESSO simulations.

💡Supercell

In computational materials science, a supercell is a conceptual unit cell that is repeated in all three spatial dimensions to model a crystal lattice. The video script mentions various supercell sizes provided for the H2O-LiH system, with the tutorial focusing on a 32-atom supercell.

💡Gamma Point

The gamma point refers to the point (0,0,0) in the reciprocal space of a crystal lattice, which is often used in simulations to simplify calculations. The script notes that a gamma point calculation was used to reproduce the results from the referenced paper.

💡Occupations

In the context of DFT, occupations refer to the distribution of electrons in energy levels. The script mentions setting occupations to 'fixed' for the semiconductor or insulator being studied, which is an important parameter in the SCF calculation.

💡Rydberg

The Rydberg is a unit of energy used in atomic physics, equal to the energy of a photon whose wavelength is 1.09677582 × 10^7 meters. The script discusses energy cutoffs in terms of Rydbergs to ensure accurate and efficient simulations.

Highlights

Introduction to the tutorial on calculating adsorption energy using density functional theory and Quantum ESPRESSO.

Explanation of adsorption energy and its significance in understanding the interaction between a molecule and a surface.

The formula for calculating adsorption energy and the implications of a negative value indicating favorable adsorption.

Different conventions for calculating adsorption energy and their impact on the interpretation of results.

Selection of the H2O-LiH (001) system for the tutorial and the importance of choosing an appropriate system.

Use of supplementary material from a 2017 paper to obtain the structure of the H2O-LiH system.

Conversion of the structure into a CIF file for use with Quantum ESPRESSO, emphasizing the importance of file format compatibility.

The choice of the PBE exchange correlation functional for the DFT calculations.

Targeting to reproduce a specific adsorption energy value from the literature using Quantum ESPRESSO.

Introduction of BURAI, a GUI for Quantum ESPRESSO, and its role in simplifying the simulation process.

The necessity of running three separate SCF calculations for the total system, isolated molecule, and isolated surface.

Importance of selecting appropriate pseudopotentials for accurate simulation results.

Setting energy and charge cutoffs in BURAI to optimize simulation speed and accuracy.

Preparation of inputs for the isolated LiH surface calculation, emphasizing the need to remove the H2O molecule from the structure.

Ensuring the isolated H2O molecule does not interact with its periodic images by adjusting the unit cell size.

Execution of SCF calculations for each system component and the importance of using the same pseudopotentials across calculations.

Extraction of total energy values from the results and the process of calculating the final adsorption energy.

Comparison of the calculated adsorption energy with the literature value and discussion of potential reasons for discrepancies.

Conclusion of the tutorial with a summary of the steps involved and the outcome of the adsorption energy calculation.

Encouragement for viewers to like, subscribe, and ask questions for further engagement with the content.

Transcripts

Hey guys, how's it going? I'm Manas  Sharma and welcome back to Phys Whiz.

In this tutorial, I'll be showing you guys  how to calculate the adsorption energy of a  

molecule on top of a surface using density  functional theory and quantum espresso.

Now for this tutorial, we'll be  calculating the adsorption energy  

of the H2O or the water molecule  on top of the LiH (001) surface.

And the adsorption energy is usually given  by a formula that looks something like this.

So you basically calculate the energy, the DFT  energy of the total system, total periodic system  

and then you subtract the energies of the  isolated molecule or in this case the water  

molecule as well as the slab or the surface,  that is in this case the LiH (001) surface.

So you calculate the total energy  and then you subtract the energy  

of the components from it and this  gives you the adsorption energy.

And please bear in mind that if you  use this formula or this convention,  

then a negative value of the adsorption energy  indicates that the adsorption is favorable.

However, some you know scientists or papers also  use a reverse formula or reverse convention.

For example, if we go to this paper here, then  they use a formula where they subtract the  

energy of the total periodic system from the  sum of the components and in this convention  

a positive value of the adsorption energy  indicates that the adsorption is favorable.

So just keep this in mind whenever you're  calculating or comparing your results with  

literature and umm, so we will be following  this paper, umm for this particular tutorial.

This paper was written back in 2017 and we  will obtain the structure of the H2O-LiH  

surface from the supplementary  information of this material.

Now I have already downloaded the  supplementary information and they  

have provided the coordinates of the H2O  LIH system of various supercell sizes.

So they provide, you know, systems  where you have 16 atoms in the surface,  

32 atoms in the surface, 64 atoms  and 128 atoms in the surface.

Now for this tutorial to keep it quick, we'll be  using the second structure that contains 16 Li  

and 16 H atoms, giving us a total of 32 atoms in  the surface and three atoms of the H2O molecule.

And I have already converted this structure  into a CIF file as you can see over here.

So they provided it in the VASP  format that is a poscar file.

But here is a CIF file and for  this tutorial what we will be  

doing is we will be calculating the DFT  adsorption energy using the PBE exchange  

correlation functional and trying  to reproduce this particular value.

So in this table two of this paper  they calculate the adsorption energy  

using a gamma point and 32 atoms in the  surface which is what we are also using.

And with PBE they get an asorption  energy of 219 milli electron volts.

And in our convention, it would actually be -219  electronvolts, but that really doesn't matter.

What matters is the magnitude.

So let's see if we can reproduce this value  with Quantum Espresso in our simulations.

Now to run the Quantum Espresso simulations,  

I'll be using BURAI, which is GUI for  Quantum Espresso that I really like.

And in order to calculate the adsorption energy,  

we'll need to run 3 calculations,  3 SCF calculations actually.

So first will be the energy SCF  calculation on the total periodic system.

The other would be for the isolated H2O molecule,  

and the last would be for  the isolated LiH surface.

So let's come back to BURAI  and import this CIF file that  

I already have and I'll put it  in the description down below.

And in fact, I'll put the link to the  paper as well as all the simulation  

files that we create in this tutorial  will be in the description down below.

So please be sure to check it out.

So let's import this structure into BURAI.

Now it looks really good and so nothing seems odd.

However, what we'll have to check is if  the pseudo potentials are correct or not.

Now you can see that BURAI  has assigned pseudopotentials  

of different categories to each atom.

So let's be consistent and  assign the same category.

Or you know the type of the pseudo potential to  be PAW for each of the atoms, so I'll choose.

Let's say this one.

Which is a PAW pseudo of potential I I  am not choosing this one because here the  

cutoff of the wave function is really  high, that is 102 oryeah, 2 rydbergs.

But I'll use this one because here the  cutoff, the energy cutoff is quite small.

So this actually, you know,  

speeds up your simulations if you  have a smaller cutoff of energy.

So let's choose this pseudo potential for Li now.

Similarly, let's choose a PAW  potential for hydrogen atom.

And again, for oxygen, let's.

Choose PAW pseudo potentials such as this one.

Now you can see in all these pseudo  potentials that we have chosen the  

cutoff of the recommended cutoff for the  energy is never more than 49 Rydberg.

So this should keep our simulations fast.

Now let's go ahead and save this project  and call it H2O-Li16H16 and save it and  

then come to the SCF tab of BURAI and here we  will write the cutoff to be 50 rydbergs and.

To be safe, we will set the cutoff  for charge to be 500 rybergs.

That is 10 times of 50.

And we will use fixed occupations because this  is probably a semiconductor or an insulator.

So let's just go ahead and save this project once  again and run the SCF calculation on 4 threads.

OK, now while this is running, let's also prepare  the inputs for the other two calculations.

That is the isolated LiH surface.

So once again, just import this CIF file.

Into BURAI and this time we  will delete these three atoms  

corresponding to the water molecule,  so they are at the end of our file.

So we will just go ahead  and delete all these three.

Atoms and this just gives us the  LiH-001 surface. Now once again  

we will come to the elements tab to set  our pseudopotentials and we will again  

choose the same paw pseudopotentials  that we chose for the total system.

And let's now save this  project by the name Li16H16.

Again, come to the SCF tab and set the  cutoff for wave function to be 50 rydbergs.

Charge to be 500 rydbergs;  occupations to be fixed.

And by the way, I didn't mention  it but we are using just a gamma  

point here because we are trying to reproduce this  

value in the paper and this was also  obtained using a gamma point calculation.

So we are just using that.

So let us save it again and run the SCF  calculation for this particular system as well.

Using 4 threads again, so we we can  see that the previous calculation is  

still running while the newer  calculation is in the queue.

And in the meanwhile, let's prepare the.

Calculation for the third energy SCF calculation.

That is for the isolated H2O molecule.

So let's import the CIF file into a BURAI again  

and this time we will get  rid of all the LiH atoms.

So just go ahead and select  all these and get rid of these.

And then upload this to the GUI.

OK so that works.

However this time I'm going  to make one more change so.

Since this calculation has to be for an  isolated water molecule, we need to make  

sure that the water molecule doesn't interact  with its neighbouring periodic images and I'm  

not sure if you know this particular unit  cell or supercell with would do the job.

So let me switch to a cubic unit cell with you  know lattice parameter of 15 angstrom and this  

would definitely make sure that the water molecule  is not interacting with its periodic images.

Again, we will set the same  pseudo potential that we used  

before because this is really important,  otherwise your results won't Make sense.

So let's use the paw again and save this project  by the name H2O and come to the SCF tab again.

And this time again, we will choose the  same energy cutoff and charge cutoff.

And this time again this would  be gamma point calculation,  

because for molecules the K  points don't really make sense.

Set the occupations to fixed for molecule,  

save it again and run the SCF  calculation for this system as well.

OK, so now let's see.

OK, so now we can see that the previous  two calculations have already finished,  

so let's go ahead and extract the energy.

So here we will go to results.

And see the log file.

Search for the final total energy and here it is.

So the final total energy at the end is this.

So let's go ahead and copy this.

Copy and then come to our browser,  let's say or calculator, whatever.

I'll be using Wolfram Alpha actually.

And I'll paste this value over here.

Sorry, something went wrong, OK.

Copy and paste this value over  here and so this would be the  

value for the energy value for the total system.

Then we will subtract the  energy of the components.

So let us now go ahead and pick  up the energy for the slab.

So again in the results log file  search for the total energy.

That is over here.

Go ahead and copy this.

And again, paste it over here and I believe  

that the calculation for the H2O  molecule is also converged now.

So let's go ahead and check that out.

Log file search for this exclamation mark.

And actually this is still running,  so let's just wait for a few minutes.

OK? So now it's done.

So, OK, so now it is finished and we'll just  

go ahead and copy this energy  again and paste it over here.

And let's see what we get.

So we get an adsorption energy that is negative so  that it negates that the absorption is favorable.

Do not worry if it is not positive because  this paper was using a reverse convention.

So basically our results are equivalent.

Right now both results indicate that the  

adsorption is favorable and we  get a value of 0.0164 Rydberg.

So let's calculate that and come to this web app  

that I have created and convert this  value in Rydbergs To electron volts.

So paste in this value in  Rydbergs and here is what we get.

So our result is that the adsorption  energy is -223 milli electron volts.  

And don't worry about the minus sign. As I  already mentioned the arc convention is the  

opposite so these two results are equivalent  and our results results of 223 Milli electron  

volts is actually quite close to the result  obtained in this paper of 219 Milli electron  

volts using VASP and the remaining differences,  In our results could be due to various facts.

So one reason could be a different  choice of pseudo potential.

I'm not sure what pseudopodential did they use.

Another reason could be I didn't really converge  my results with respect to the energy cut off.

So maybe 50 rydbergs wasn't enough.

Maybe I should have chosen a larger  value, maybe they chose a smaller value.

So these could be the reasons  for this very slight difference.

But overall I think this is a really  good agreement with the reference.

So yeah, so that is it.

In today today's tutorial, you have learned  how to calculate the asorption energy of a  

molecule on top of the surface  using Quantum Espresso and DFT.

I hope you guys really enjoyed  this tutorial and found it useful.

In case you did, then don't forget to hit  

the like button and subscribe to my  channel for more videos like this.

If you have any questions or doubts, leave  them in the comment section down below.

And thanks for watching.

Have a great day.