Author Topic: problems on thermoelectric calculations  (Read 2547 times)

0 Members and 1 Guest are viewing this topic.

Offline wot19920302

  • QuantumATK Guru
  • ****
  • Posts: 124
  • Country: cn
  • Reputation: 0
    • View Profile
problems on thermoelectric calculations
« on: March 25, 2016, 19:06 »
  Dear staffs, I meet some problems when studying thermoelectric materials using ATK:
      1.what does T mean in ZT(figure of merit)? environment temperature or difference between hot and cold?
      2.when I use DFT to calculate dynamic matrix to get phonon transmission spetrum so that I can get thermal conductance, I find one term in calculatior setting is to define electrode temperature(top picture,the default is 300K). And when I open the Thermolelctric Coefficients  plugin, I find I can get various thermal conductance by changing temperature T(buttom picture). If I want to get the  thermal conductance in 500K, for example, I can use DFT with 300K, and then set 500K in Thermolelctric Coefficients  plugin; or  I can use DFT with 500K and get thermal conductance directly. which one correct?
   Best regards.
     

Offline Jess Wellendorff

  • QuantumATK Staff
  • Supreme QuantumATK Wizard
  • *****
  • Posts: 933
  • Country: dk
  • Reputation: 29
    • View Profile
Re: problems on thermoelectric calculations
« Reply #1 on: March 29, 2016, 11:37 »
1) Please have a look through the literature on thermoelectrics to understand the meaning of the ZT "figure of merit", or simply browse through the wikipedia entry on the subject: https://en.wikipedia.org/wiki/Thermoelectric_materials#Materials_selection_criteria. The ZT is a dimensionless number that indicates the ability of a given material to efficiently produce thermoelectric power. ZT should therefore be as high as possible. This tutorial shows how to calculate it manually with ATK (without the plugin): http://docs.quantumwise.com/tutorials/thermoelectrics_cnt_isotope.html#thermoelectrics-cnt-isotope

2) The electronic temperature in the Calculator Settings is only used for getting the DFT calculation to converge. It is not a "real" temperature" but rather a smearing of the distribution of electronic occupations around the Fermi level. It is therefore something very different from the temperatures given in the Thermoelectric Coefficients plugin. Use the first method you mention: Only change the temperature in the Thermoelectric Coefficients plugin.

Offline wot19920302

  • QuantumATK Guru
  • ****
  • Posts: 124
  • Country: cn
  • Reputation: 0
    • View Profile
Re: problems on thermoelectric calculations
« Reply #2 on: March 30, 2016, 16:49 »
1) Please have a look through the literature on thermoelectrics to understand the meaning of the ZT "figure of merit", or simply browse through the wikipedia entry on the subject: https://en.wikipedia.org/wiki/Thermoelectric_materials#Materials_selection_criteria. The ZT is a dimensionless number that indicates the ability of a given material to efficiently produce thermoelectric power. ZT should therefore be as high as possible. This tutorial shows how to calculate it manually with ATK (without the plugin): http://docs.quantumwise.com/tutorials/thermoelectrics_cnt_isotope.html#thermoelectrics-cnt-isotope

2) The electronic temperature in the Calculator Settings is only used for getting the DFT calculation to converge. It is not a "real" temperature" but rather a smearing of the distribution of electronic occupations around the Fermi level. It is therefore something very different from the temperatures given in the Thermoelectric Coefficients plugin. Use the first method you mention: Only change the temperature in the Thermoelectric Coefficients plugin.
Does it mean that "elecrode temperature" in the calculator setting also couldn't affect electronic transmisson spetrum when calculating electronic conductance态seebeck coefficient in various temperature?

Offline Jess Wellendorff

  • QuantumATK Staff
  • Supreme QuantumATK Wizard
  • *****
  • Posts: 933
  • Country: dk
  • Reputation: 29
    • View Profile
Re: problems on thermoelectric calculations
« Reply #3 on: March 31, 2016, 08:55 »
Yes, the electronic temperature should not affect electronic transmission spectra, provided you do set it to something excessively high, like 10000 K.

Offline wot19920302

  • QuantumATK Guru
  • ****
  • Posts: 124
  • Country: cn
  • Reputation: 0
    • View Profile
Re: problems on thermoelectric calculations
« Reply #4 on: March 31, 2016, 09:03 »
thanks for your help!