Simulate I-V Curve didn't converge

[HouChaojian]

Use ATK+DFT to simulate I-Vcurve.
The results was put in the next
Exchange coreelation:LDA+PZ
cut off:150Rydberg
maxmum steps:500
Tolerance:0.0001Hartree
Electrode's Basis Set: SingleZetaPolarzed
others Basis Set:DoubleZetaPolarzed
Poisson Solver:FFT2D
but the similute also didn't converge.How to deal with it.Are there need to increase the steps?

Warning:
+------------------------------------------------------------------------------+
| 499 E = -155.742 dE =  9.927973e-03 dH =  7.003176e-03                       |
+------------------------------------------------------------------------------+
################################################################################
#                                                                              #
# Warning: The calculation did not converge to the requested tolerance!        #
#                                                                              #
################################################################################

[Anders Blom]

How is anyone supposed to help with these details? We don't even know how many Uranium atoms you have in the system

[HouChaojian]

Hi
In my simulation,  The model has two nanowires and a  nanoparticle; nanowire is electrode and nanoparticle(Shape is a circle) is center zone;I want get I-V. when I set the bias 0-2V in very small model(total atoms is 60),I can get it;But  if I set up the bias 0-3V or 0-4v or 0-5V, The iteration is oscillation, can not converge;And I try a big model in the bias 0-2V, The iteration is also oscillation.
What should I do?
nanoparticle is metal
I need your help,Thank you!

[Anders Blom]

So it's the nanowire which is made out of Uranium, or the nanoparticle?
I'm joking of course - but the point remains, such a question cannot be answered without detailed information about your calculation.

[Anders Blom]

There are many things to think about, so it would be best to post the script (or email it), otherwise we can go on forever checking point by point (length of electrodes - not seen in picture, basis set, k-points, temperature, mixer settings, etc).

[HouChaojian]

What is email address

[Anders Blom]

Click email icon...

[HouChaojian]

get it,I will  sent the detailed infromation to you! thank you！