Dear Fazle, I don't have profound knowledge on the Transiesta. Probably you are using the same .TSHS file for left and right electrodes.. Since, the change in coordinates of corresponding atoms in your left and right electrodes are not same upon relaxation, final coordinate information in the left and right electrode atoms are not identical.
I have faced the same problem.... The steps that I am following currently to avoid the problem is as follows: ################# Within SIESTA ################################ Step 1: Draw the unit cell of Graphene Nanoribbon (GNR). Step 2: Relax the unit cell structure (using VC relax but apply constrain in out of plane direction if you want ). Step 3: Repeat the relaxed unit cell of AGNR to make whole device structure (Left Electrode + Scattering Region + Right Electrode). Step 4: Rigid the Left and Right electrodes while VC relaxing the entire structure. ################ Within TRANSIESTA ############################# Step 5: *Electrodes*: For symmetrical Left and Right electrodes, calculate self-energies of the Left electrode. Use .XV file of the Step 4 to construct relaxed Left electrode. Step 6: *SCF of Junctions*: Use the .TSHS files for Left and Right electrodes for calculating self energies of the scattering region employing Green's function method. Step 7: *Transmission*: For TBT calculation, use relatively denser K-Point grid of say, *100 X 100 X 1*. It is expected to get the Current Values on the output script at each bias point if the system has .netcdf files installed properly. Step 7: *I-V Curve*: Change the bias value of the Transiesta run at steps 5-7 to get new value of current at this particular bias point. Repeat the procedure to get the complete set of I-V values. Please let me know if you can solve your problem.. Best Regards.. Bibhas On Tue, Jan 9, 2018 at 8:02 AM, Fazle Subhan <[email protected]> wrote: > Dear SIESTA/TRANSIESTA users > > > I need to relax only scattering region (SR) from the whole ribbon. Because > I create a vacancy in the SR and when I relax the structure (Electrodes + > SR) by siesta using many CG steps. After relaxation I apply simple diagonal > method (using transiesta executable) on the whole structure it was working > well but when I start to apply the Green Function there is an error occur > (electrodes are not in the same co-ordinates). Therefore now I want to > relax only the SR. Please help me in this regard, Thanks in advance > > > > Best Regards > Fazle Subhan > > >
