QuantumATK Forum

QuantumATK => General Questions and Answers => Topic started by: kaypu on June 27, 2011, 14:35

Title: Graphene convergence
Post by: kaypu on June 27, 2011, 14:35

Dear everyone:
     I've got a problem about Graphene convergence, this is my py script.
     In the Electrodes Calculation and Equivalent Bulk Calculation (Initial Density for TwoProbe,
     it converge well, but in two-probe calculation, it doesn't work well
     this is the calculation process:
      sc  1 : q =  482.55087 e  dRho =  5.4297E-01
      sc  2 : q =  465.94638 e  dRho =  8.9014E+02
      sc  3 : q =  488.77028 e  dRho =  8.9009E+02
      sc  4 : q =  491.19750 e  dRho =  6.2499E-01
      sc  5 : q =  497.45857 e  dRho =  2.9714E+00
      sc  6 : q =  495.12513 e  dRho =  1.2585E+00
      sc  7 : q =  493.84228 e  dRho =  5.4206E+02
      sc  8 : q =  491.72195 e  dRho =  3.3873E+02
      sc  9 : q =  494.09409 e  dRho =  6.1353E+02
      ......
      what' wrong with the system?

Title: Re: Graphene convergence
Post by: Anders Blom on June 28, 2011, 07:41

Although we can only offer limited support for version 2008.10 these days, I suggest you have a look at the k-point sampling. You should probably have more points in Z (50 or 100), you only need 1 in Y, and you need more in X, perhaps 9 or 12.