Dear Mohit, I see the formula you are mentioning using the matsubara frequencies.
To use that expression, you need to use imaginary energies. The steps are; Sites1=[lat(5,j) for j in range(W)] # the sites where to mount lead 2 Sites2=[lat(6,j) for j in range(W)] # the sites where to mount lead 3 mount_vlead(sys, Sites1, 4) # number of orbitals =4. Here we mount lead number 2 mount_vlead(sys, Sites2, 4) # number of orbitals =4. Here we mount lead number 3 syst=sys.finalized() G12=greens_function(syst, energy=-1.8*1j, in_leads=[3],out_leads=[2], check_hermiticity=False).data G21=greens_function(syst, ,energy=-1.8*1j, in_leads=[2],out_leads=[3], check_hermiticity=False).data H12=syst.hamiltonian_submatrix(to_sites=Sites1, from_sites=Sites2) H21=syst.hamiltonian_submatrix(to_sites=Sites2, from_sites=Sites1) Notices that, we had to use "check_hermiticity=False" because for these complex energies, the system is not hermitian Trace(np.dot(H21, G12)-np.dot(H12,G21) )) # check if the order is correct. I hope this helps. Adel On Wed, Jun 17, 2020 at 6:51 PM <mohit.gupta9...@gmail.com> wrote: > Dear Abbout, > I apologize for not being clear in my question. My system is infinite > (that is there are leads attached to the system) and I mounted virtual > leads at points where I wanted kwant to evaluate green's function. Sorry > about that. > > The formula that I am using is from the supplementary material of the > following paper: > https://journals.aps.org/prl/abstract/10.1103/PhysRevLett.120.047702 > It is possible that I am misinterpreting the formula as given in the paper > as am very new to the subject. > > I want to evaluate critical current in the Josephson Junction by > maximizing current-phase dependency, for that I need total current in the > system as a function of phase difference phi, I am not sure how to do that > with local current operator in kwant. > > Thank you for clarifying the output of green's function I am now sure that > I was reading the output wrong which must have been the primary reason for > getting zero. > > I again apologize if I am not clear about what I am saying. > -- Abbout Adel
