what's wrong with the transmission spectrum?

[wot19920302]

Dear Quantumstaffs:
            I got transmission spectrums for pristine MoS2 nanoribbon(fig1) and vacancy MoS2 nanoribbon(fig2), respectly. These two transmission spectrums, I think, look strange. Considering 1D structures, I set k-points for calculators attached to both structures  1*1*100, and k-points for transmission calculating  1*1. Do I get correct transmission spectrums? If not, what's wrong with me?
           I'm looking for your reply
           Best regards

[Anders Blom]

Why do you think it's wrong?

[wot19920302]

    Well....It's my first time to  analyse transmission spectrum. I think the  curve of fig2 undulates much from -0.7eV to -2.0eV  and the fig1 is step-like. Is that reseaonable?

[Anders Blom]

Yes, it's reasonable (which doesn't mean I say your calculation is perfect, for that one would have to check a few other things).

[wot19920302]

Dear Anders Blom：
           Thanks for your kindness. In fact, I fear  the effect of a bad optimized structure on transmission spectrum, how can I check if the transmission spectrum is perfect?

[Anders Blom]

There is no simple way to see that from the transmission, but obviously you can compute the forces on the atoms, and if these are large then the structure is not well optimized.

[wot19920302]

      When we optimize structures using ATK, we set max_force and max _stress in optimization term. Does it mean we don't calculate the force of atoms in strcuture again after optimizing the structure?

[Anders Blom]

Huh?

[wot19920302]

Dear Anders Blom :
      You said :" obviously you can compute the forces on the atoms, and if these are large then the structure is not well optimized." .   So if we want to calculate transmssion spectrum after optimization(it means we have set max_torce in optimization term and forces on atoms in the structure satisfy the max_force), should I compute forces on atoms again?

[Anders Blom]

Eh, no... Computing the forces just tell you how large they are - but you know that from the optimization...

[wot19920302]

Eh, no... Computing the forces just tell you how large they are - but you know that from the optimization...

   
               So you mean that I don't need to calculate the forces on the atoms after optimizing the configuration...am I right?
       

[Anders Blom]

Right... Actually the forces are printed during the optimization.

[wot19920302]

thank you, sir!

[wot19920302]

Dear Anders Blom
           I want to study effects of defects in MoS2 nanoribbon on thermoelectric properties. I know I should calculate electron transmission spectrum to get conductance,seebeck coefficient,electrical thermal conductance and get phonon thermal conductance using phonon transmission spectrum. But when I get ZT as the function of E-Ef, I find that the transmission value of E-Ef corresponding to ZTmax close to Zero!  I don't know whether it's reasonable, and I also can't analyse conductance in different defect systems (because it's too small). I hope you can help me
           Best wishes
           Here are my scripts: