an error in the calculation of IV curve

[huangshenjie]

Dear all,
I made a MTJ by ATK, but the I-Vcurve shows that when the bias is increasing, firstly, the current also increases, but then the current quickly declines to zero.

I don't know how to explain it, since I think that the current should always increases with the increase of the bias.

Waiting for your answer.

[mads.engelund]

I am not sure if your MTJ should have this feature, but it is possible for the current to go down when the bias is increased.
http://en.wikipedia.org/wiki/Negative_differential_conductivity
Check if others have observed this feature, if they have not, return.

[Anders Blom]

I don't think if one can say anything about your results. The graph looks like it has 5-6 points only, and the current is exactly zero for the higher bias. However, that can also be caused by bad convergence, esp. if you are using older versions of ATK that could happen (the whole calculation converged to zero electron density). You need smaller steps in bias to really get a proper smooth curve.