Hi everyone! Good morning or good day for those that have just joined!

Good morning from Sydney

Good evening from Jodhpur

Good morning from Seoul, South Korea

We are having a short break before the tutorial session

If you want to try the tutorials on your computers

the material is in the website. You would need to have virtual box installed and a GitHub account

http://www.quantum-multiscale.org/school-materials.html

The session will start in 20 minutes

The celldm(1) is one dimensional ?

Cool

it is a cubic cell, so one parameter is all you need

Some geometries are given without indicating the symmetry (just the cartesian coordinates), is it possible that QE identify the symmetry?

Let me stress here that the parameters you see now in Davide’s input are all the main parameters that you need to take care of in normal simulations

celldm, ecutwfc, ecutrho, conv_thr, mixing_beta K_Points

What is the d0 at the end of the mass

👌👌

for the virtual machines, this would be ~/Software/quantum-espresso

But in the VM the executables have been moved to /usr/local/bin

The last line. Those 111 what do those mean?

=

there is a typo in mpirun -np

-np

-np

That should be -

Thanks!

Could you explain what -nk 2 < means

-np

the resolution of the overhead screen is bad and we did not notice it there

Select portion of screen for sharing

is the total electronic energy the energy required to move all electrons infinitely far away from each other and from the nuclei?

The total energy corresponds to electronic part?

It depends on the pseudo-potentials and on the density functional that you selected, so it makes little sense per se

oh i see. thank you!

energy differences are what we look for

I have QE on my ubuntu, for others(PAOFLOW, Environ,..)should I install VM?

we have little problem with install of V M

You can download a zip file and do it “manually”

Maybe you can change the resolution of VM by: View-> Virtual Screen-> 100%

Yes !

I was able to download the repository

it is better download from gitHub site?

can we check git clone --bare followed by https://

I think you have to check the boxes when generating the key

yes, I generated the key and clone It successfully

To generate valid tokens, check repo box under select scopes

If you have issues with the git clone command, you can always download the zip of the repository from the website

So this script will over write the .out files

QE_BIN should be changed to /usr/local/bin if you use VM

What is the purpose of the eos calculation? Is this to relax the cell?

Optimization of cell is done with vc-relax

For lattice relaxation, should we run the geometry optimization first and implement the eos calculation later?

ok

Ok

ok

How do you choose the number of empty bands?

This is a periodic wave planes calculation. Where are included the atomic orbitals?. When you are projecting over atoms, what are you doing computationally?. Can you comment?

What are the units of the band energies in the xml file? Are the band energies shifted such that the fermi energy is 0?

Why did you use an "unshifted" mesh for nscf compared to scf? I missed that part.

I want Gamma (0 0 0) point

There is an echo

Sorry, I did not get which would be the solution in case the DOS and total pDOS do not match properly

How far can we trust the values beyond the HOMO ?

Not very much, GGA functional will systemically underestimate the gap due to self interaction errors

👌👌

It gives a nice approximation though to the topology of the bands , metaGgA and hybrids help out with reducing that self interaction error and yield better gaps

Thanks a lot @Etienne

You implemented the calculation of bands with HSE directly from QE?

What about HSE calculations for GaAs? We do alone?

What are we defining in smearing?

I couldn’t see a keyword for temperature

There is a very useful pw.x input file description page

Thank you David

Thank you !

Thank you so much

Thank you!

Thank you

thank you