Hi Alex,
I calculated the p-type Seebeck coefficient of SnSe as shown in the following figures.
At 300K, these results are reasonable and similar to the experimental results.
However, at 750K, the Seebeck coefficient seems to undergo a p-n type transition.
Is this phenomenon reasonable?
Best
Wei Cao
After test, I find that the problem is the setting parameter ‘bandgap’.
At 300K, setting ‘bandgap’ or not does not matter.
However, setting ‘bandgap’ by a HSE06 value can get the right results at 750K.
But I still have a problem.
For a narrow gaped system (PBE gap 0.1eV and HSE06 gap 0.2eV), I set the ‘bandgap’ by a HSE06 value and I still get the p-n type transition of Seebeck coefficient at high temperature.
Is it reasonable? Or should I set a larger ‘bandgap’ value?
This effect is caused by bipolar conduction and will also occur in experiment. The best results are obtained when the band gap is set to the experimental value or a HSE band gap.
