BoltzTraP2 Tutorial | Doping Semiconductors

In this video we go over how to dope semiconductors in BoltzTraP2 and subsequently plot the Onsager coefficients. The relevant scripts to the video can be found here, please star the repo if you find it useful.
github.com/alg...

[@tikarestianingsih9055]

Hello Sir, for plotting ZT value, Do we need to divided the value of PF and kappa_el with relaxation time? I have kappa_lattice data (not kappa_lattice/\tau). if I want to plot ZT, I need to multiply the kappa_lattice with the constant relaxation time used in the calculation, isn't it?And how do we know the value of constant relaxation time in this calculation? Thank you

[@algebraiccontinuation9291]

You can just write out the units on paper and do the math there are a few separate equations and quantities you're asking about. Anything in terms of 1/Tau needs to have that multiplied out, so not k_lattice. For how to get the relaxation time, you can either get it from literature, approximation from experiment, or deformation potential there with an initio calculations.

[@tikarestianingsih9055]

@@algebraiccontinuation9291 Okay, Thank you. At first, I didn't understand. But by watching your video over and over and learning boltztrap2 now, I finally understand I have more question, Do you know How to plot onsager coefficient as a function of temperature without specifying the chemical potential? If I plot using "btp2 plot -T -c '["xx", "zz"]' -s 100 interpolation.btj S", there are many lines with different chemical potentials. I also want to plot average "S", not every Sxx,Syy, and Szz. your code only provided for doping

[@algebraiccontinuation9291]

I'm confused because those last two things are different issues. Is your issue that you want a single line S, or that you want a single line S per chemical potential? There is no way to just remove the chemical potential. It has to be selected to plot against temperature.

[@tikarestianingsih9055]

@@algebraiccontinuation9291 oh, I see. Is there a spesific command to select particular chemical potential?

[@algebraiccontinuation9291]

It's more typical to pick a carrier concentration like in this video. You can always look in your condtens files though, and look for the chemical potential you're desiring. So if you don't know what carrier concentration you want, probably using the interpolate approach is better than dope. I forget if there is a way to pin the chemical potential, you can run help on the dope command though to double check. btp2 dope -h (or maybe it's --h?)

[@markedwincleofe5589]

Good day, Sir. I just wanna ask why is the plot show carrier concentration 1e18 even though you set it to 1e19 in target_conc? Thanks in advance!

[@algebraiccontinuation9291]

That is the scale for the plot of the conductivity, not the carrier concentration. So in general there’s no reason to expect those to be the same.

[@markedwincleofe5589]

I understand, Sir. Thank you for your reply. However in that case, what is the purpose of 1e18 on the plot? Why is it being shown there? Thank you in advance! :))

[@algebraiccontinuation9291]

@markedwincleofe5589 Hmm I'm not sure I see what the misunderstanding/confusion is? That's the scale on the plot of conductivities.

[@ComicBook-vf7wq]

@@algebraiccontinuation9291 That is very interetsing, the number (scaling 1e18) appears due to the approximation used in Boltztrap2 code RTA. THe electrical conductivity in the plot is Sigma/Tau where Tau is the relaxation time which is system dependent. If you want get the actual conductivity you have to multiply the conductivity with Tau (relation time) which upcourse is system dependent. Hope this helps

[@wien2khelp459]

hello sir, whenever I run doping.py I got the gpaw error. I tried to install the gpaw but i failed. Is there any alternate option for running doping.py without gpaw. waiting for your respose. I am using wien2k code.

[@algebraiccontinuation9291]

GPAW is installable with conda so there shouldn’t be an issue there. What is the specific problem you’re getting? You can use whatever code you want to do doping.py Be sure to read the script and it’s comments carefully, there should be something in there about the volume if you’re not using GPAW.

[@burhanahmed6701]

Thanks your your response sir. run_doping.py is working fine for me, but when I run doping.py I got the error message that there is no module named 'gapw' .@@algebraiccontinuation9291

[@wien2khelp459]

@@algebraiccontinuation9291 Thanks for your response sir. I think I have a problem in installing GPAW with both conda and pip. Actually I have installed intel python compiler which includes both conda and python but I am not be able to install gpaw using conda. Is there any alternate way to install gpaw??

[@algebraiccontinuation9291]

I think both of you are running into the same issue likely. If you aren't using GPAW for the DFT, then you don't need GPAW at all. You will have to modify doping.py to replace the things that GPAW does though, see the comments in the script for this, I believe it's just manually inputting the volume.

[@wien2khelp459]

@@algebraiccontinuation9291 Thanks a lot sir. It works fine for me now. One more question, while putting cell volume, it is mentioned that it should in ang but in my result I got the volume in ang^3. So should I convert it into ang or use the direct value of ang^3. One more thing if possible make such a script for temp vs power factor and temp vs ZT value.