Author Topic: Question about Error's Bulk optimization  (Read 2748 times)

0 Members and 1 Guest are viewing this topic.

Offline hieido

  • Heavy QuantumATK user
  • ***
  • Posts: 29
  • Country: kr
  • Reputation: 0
    • View Profile
Question about Error's Bulk optimization
« on: March 29, 2012, 03:31 »
 Dear All!
I made an idea device on the form of bulk Junction device
Electrode: Au on the both left and right
Central region: Silicon (were passivated by H) connected with Molecule in the absence of Sulfur in the both Left and Right sight.
Whole the central region I translated into the bulk form. And then I putted this one on the bulk Junction.
Then I going to optimization whole electrode with the condition like:
1.  In the new calculation: ATK: DFT. Electrode Temperature: 300K, Density mesh - cut off: 100 RyB, No SCF iteration, K-point sampling na: nb:nc = 1:1:100. Basic set: I chose Au and Si: single zeta polarize.
2. Optimization Geo: Maxim Stress, Maxin Force : 0.05, Maxim number test: 100. Maxim step size: 0.5. Constrain cell: I click on X, y and none with Z.
and then running.
And then got the messenger: "This application has requested the Runtime to terminate it in an unusual way.
Please contact the application's support team for more information."

Can You help me?

« Last Edit: March 29, 2012, 10:20 by hieido »

Offline Anders Blom

  • QuantumATK Staff
  • Supreme QuantumATK Wizard
  • *****
  • Posts: 5576
  • Country: dk
  • Reputation: 96
    • View Profile
    • QuantumATK at Synopsys
Re: Question about Error's Bulk optimization
« Reply #1 on: March 29, 2012, 09:48 »
The error message signals that you have run out of memory on the computer. You don't need 100 k-points in C to optimize the geometry as bulk; those k-points are only relevant for the electrodes. If you try k-points 1x1x1 you will see a drastic reduction in memory usage, no least because you can operate with real matrices instead of complex. Note that this only applies to the central region, i.e. if you try to optimize that separately, as bulk.

Another thing - I would imagine that a full optimization of this structure will take a very, very long time. You may want to go directly for a transport calculation just to see if it looks reasonable.
« Last Edit: March 29, 2012, 09:52 by Anders Blom »