QuantumATK Forum

QuantumATK => General Questions and Answers => Topic started by: kaypu on December 7, 2010, 17:56

Title: questions about the convergence
Post by: kaypu on December 7, 2010, 17:56

Dear blom:
     i've some questions about the convergence
① in my two probe system,when the bias voltage is set at -0.9V, it doesn't converge at all,  but in 0.9V it works well. why? because of the asymmetry of the molecular or the position of the molecular in the two probe?              
② i increase the temperature of the electrode from 300k to 1000k in the -0.9V,but it doesn't work!! what should i do ?
 the electrode iteration mixing parameters are set as follows:
     iterationMixingParameters
    algorithm = IterationMixing.Pulay,
    diagonal_mixing_parameter = 0.01,
    quantity = IterationMixing.Hamiltonian,
    history_steps = 15
and the k point is 5*5*100,

Title: Re: questions about the convergence
Post by: Nordland on December 7, 2010, 19:00

Based on your script posted, I can see that you are using ATK 2008.10 version.

All the newer version of ATK has a double contour implemented making it much more stable for high voltages.
I would recommend you trying a new version.

Title: Re: questions about the convergence
Post by: kaypu on December 8, 2010, 15:02

thank you very much Nordland
if i insiste on the 2008.10 canyou give me some advices on the problem
     best wishes

Title: Re: questions about the convergence
Post by: Nordland on December 8, 2010, 19:48

There used to be a guide on convergence on Fe-MgO-Fe, which had a tons of tricks in it. Let me see if I can find it for you.

Title: Re: questions about the convergence
Post by: kaypu on December 9, 2010, 00:54

i'm waiting for you
 :)thank you  very much