Author Topic: ATKError: hard to convergence  (Read 7259 times)

0 Members and 1 Guest are viewing this topic.

Offline zh

  • QuantumATK Support
  • Supreme QuantumATK Wizard
  • *****
  • Posts: 1141
  • Reputation: 24
    • View Profile
Re: ATKError: hard to convergence
« Reply #15 on: January 6, 2011, 02:22 »
I did some test calculations for your configuration with some modification and got a converged solution. The main modification made in my test calculations:
1) As pointed out in my last replies, the atomic configuration in the electrode and surface scattering regions should be adjusted.
2) The k-mesh increases up to 6x6x100.
3) integral_lower_bound = 8.0*Rydberg,
4) electrode_constraint = ElectrodeConstraints.RealSpaceDensity
The last point seems to a key step to solve this specific problem.