Hi all,

It seems to solve this problem by increase TS.biasContour.NumPoints to 10.

But one more question: How to select this value with different bais? or
this value is important to the results or not?


On Mon, Aug 13, 2012 at 5:15 PM, Longhua Li <[email protected]> wrote:

> Dear all,
>
> I have strange question to ask. this question is about transiesta under
> bais.
>
> I have successfully performed a transiesta calculation with zero bais
> (TS.Voltage = 0.0 eV)
>
> Then I continue the calculation with TS.Voltage = 0.1 eV started from the
> .TSHS obtain previously. However, the transiesta stop for the following
> error:
>  " Real-axis: From EFermiR to EFermiL
>  Gaussian fermi contour, ngau=          -1
>  ERROR: No. points=           5  not valid for real axis gaussian
> quadrature, mi
>  n          10
> ERROR: Contour: too few points for real axis int
> "
>
> My Transiesta options are:
> "SolutionMethod      transiesta
> # TS options
> TS.Voltage      0.1 eV #default=0
> TS.MixH         true #default=false
> #TS.TriDiag      true #default=false
> # ELECTRODE options
> TS.HSFileLeft           ./Au-LEL.TSHS
> TS.NumUsedAtomsLeft      147
> TS.BufferAtomsLeft       0
> TS.HSFileRight           ./Au-LEL.TSHS
> TS.NumUsedAtomsRight     147
> TS.BufferAtomsRight       0
> # GF options
> #TS.CalcGF                 false
> TS.ComplexContour.Emin       -30.0 eV
> TS.ComplexContour.NPoles       6
> TS.ComplexContour.NCircle      30
> TS.ComplexContour.NLine        10
> # BIAS OPTIONS
> TS.biasContour.NumPoints       5
> "
>
> Does anybody has experiences on such a problem?
>
> Thanks in advance,
> arlonne
>



-- 

----------------------------------------

Longhua Li

Prof. Arthur J. Freeman's Group

Department of Physics and Astronomy

Northwestern University

F263 Tech Institute

2145 Sheridan Road

Phone: 847 491 8640

Responder a