Mailman 3 September 2017 - Kwant-discuss

Gate-defined Nanowire Quantum Dot
by Jonathan Fields 15 Sep '22

15 Sep '22

Dear Reader, I am new to KWANT and trying to figure out if it is suitable for calculating transport properties of gate-defined quantum dots made out of nanowires, along the lines of the structure used in https://arxiv.org/abs/cond-mat/0609463. Now, I understand that the mailing list is not a place to ask people to do my work for me, and my question is also not for someone to do this. What I am looking for is some insights into if this is possible (and not all that difficult) with the package. At first sight (and going through the tutorials as well as the APS March Meeting material) this does not seem straightforward; what I struggle with is how to define this type of dot in KWANT. Now, one way would of course be to built the entire 3D geometry of the system, but this will be computationally expensive, and probably also rather tricky to do in terms of defining everything, from the hexagonal wire itself to all of the gates and oxide layers and such. Some abstraction would be preferred, perhaps even reducing the dimensionality and making a 2D system, such as often done in the tutorial. Is this a good idea? My problem is that if one does this, then I run into the problem of how to model the 'half wrap-around' type gates typically used in these devices. In the transport through barrier section of the APS March Meeting material there is a section on how to model realistic potentials, but this would not work all that well for these wrap around type gates. On the other hand, in a way they are perhaps not so different from the QPC in that section. One could model the three gates as something like https://i.imgur.com/WZC983c.png as they do essentially form channels in the wire. Apart from if this sacrifices too much of the nanowire nature in favor of a 2DEG system, I do suppose such a system should in principle be able to be treated as a quantum dot. I haven't been able to confirm this as I am not yet sure how one can apply a source-drain voltage in KWANT; I first thought that this would probably be related to the energy of the modes, but then again I have not been able to produce Coulomb diamonds in this way. The question is getting rather lenghty, and also a bit unclear at this point. Perhaps I should finish up by stating it in a concise form; would you think that it is possible to simulate the transport of such a gate defined nanowire quantum dot device with KWANT, and if so, is the approach I am suggesting above a viable one, or would you go about it very differently? Kind regards Jonathan <http://aka.ms/weboutlook>

4 4

How the spin is positioned in the Smatrix?
by Qingtian Zhang 06 Mar '22

06 Mar '22

Dear all, I am trying square lattice with spin-orbit coupling and Zeeman splititng just like the example of Tutorial 2.3.1. Now I want to get the spin-dependent conductance：Guu,Gud,Gdu,and Gdd, where "u" indicates up spin, d indicates down spin. We can have the scattering matrix through Smatrix=kwant.smatrix(sys, energy), but how the spin is positioned in the scattering matrix? I have checked some simple cases, it seems the Smatrix is organized like this: | r t | S= | t' r' | but how the spin is included? Thanks in advance. Qingtiian Zhang The code is: import kwant from matplotlib import pyplot from numpy import matrix import tinyarray sigma_0 = tinyarray.array([[1, 0], [0, 1]]) sigma_x = tinyarray.array([[0, 1], [1, 0]]) sigma_y = tinyarray.array([[0, -1j], [1j, 0]]) sigma_z = tinyarray.array([[1, 0], [0, -1]]) sys=kwant.Builder() lat=kwant.lattice.square() a=1 t=1.0 alpha=0.5 e_z=0.08 W=2 L=2 sys[(lat(x, y) for x in range(L) for y in range(W))] = \ 4 * t * sigma_0 + e_z * sigma_z # hoppings in x-direction sys[kwant.builder.HoppingKind((1, 0), lat, lat)] = \ -t * sigma_0 - 1j * alpha * sigma_y # hoppings in y-directions sys[kwant.builder.HoppingKind((0, 1), lat, lat)] = \ -t * sigma_0 + 1j * alpha * sigma_x #### Define the left lead. #### lead = kwant.Builder(kwant.TranslationalSymmetry((-a, 0))) lead[(lat(0, j) for j in xrange(W))] = 4 * t * sigma_0 + e_z * sigma_z # hoppings in x-direction lead[kwant.builder.HoppingKind((1, 0), lat, lat)] = \ -t * sigma_0 - 1j * alpha * sigma_y # hoppings in y-directions lead[kwant.builder.HoppingKind((0, 1), lat, lat)] = \ -t * sigma_0 + 1j * alpha * sigma_x #### Attach the leads and return the finalized system. #### sys.attach_lead(lead) sys.attach_lead(lead.reversed()) kwant.plot(sys) sys=sys.finalized() print "Hamiltonian=" print sys.hamiltonian_submatrix() Smatrix=kwant.smatrix(sys, energy=3.) print "The scattering matrix=" print matrix.round((Smatrix.data),2) print "T=",(Smatrix.transmission(1, 0))

3 3

Spin polarized transport calculation
by L.M J 06 Mar '22

06 Mar '22

Dear Kwant users and developers I have two questions related to the spin calculation using kwant. 1. I'm wondering if we can do spin-polarized transport in kwant? For example, can we can get the separate Conductance Vs Energy diagram for spin up and spin down terms. I'm thinking one way of doing this is to set up the Rashba Hamiltonian as a whole. And then manually extract the spin up and spin down hamiltonian and do the transport calculation separately. But this will eliminate some extra terms that I don't know whether this is correct anymore. Any suggestions? 2. How can we set up the spin polarization for a particular site? For example, in some of the topological insulator, can we set up spin polarization on the edge as up and the other side of the edge as down, and then do the transport? Or this question is completely wrong? Thanks for advance. Sincerely, Liming

3 3

How to caculate hall resistance
by ZHANG Bing 12 Sep '21

12 Sep '21

Dear Sir, I am a PhD student of Hong Kong University of Science and Technology. I want to use KWANT to caculate Hall resistance of a Hall bar structure.We can get the conductance between 6 electrodes, but how to get hall resistance? Can you give me some help? Thank you very much. Best Regards, Zhang Bing

4 3

beyond effective mass (limiting kwant to a range of k)
by Maurer, Leon 30 Oct '17

30 Oct '17

Hello, I’m interested in using kwant to look at transport beyond the effective mass approximation. To that end, I’ve entered a Hamiltonian that reproduces silicon’s band structure [specifically, the k.p Hamiltonian from M. Cardona and F. H. Pollak, Phys. Rev. 142, 530 (1966)] into a 1D kwant lattice. When I plot the bands in the leads using kwant.plotter.bands, at first it looks nothing like Si’s band structure (see lead_bands.pdf, attached). However, when zoomed in to an appropriate k range for Si, Si’s band structure is there as expected (see lead_bands_zoom.pdf, attached). To be more specific, this is Si’s band structure in the (100) direction, which is what I was aiming for. However, this is still useless for transport because kwant calculates transmission as a function of energy for all k values – including k values that are meaningless for Si and need to be excluded from the calculation. So, I think that my question boils down to: is there a reasonably simple way to restrict the range of k values that kwant considers? If not, can you think of another way to hack a full band structure into kwant? Thanks. -Leon PS. Just to preempt some non-helpful answers: I am not interested in replies along the lines of “You couldn’t possibly need to include the full band structure. Just use an effective mass.” I have good reasons to want to include the full band structure.

4 7

Current in a closed system
by Abbout Adel 26 Oct '17

26 Oct '17

Dear All, I was consulting the example of the current in a *closed system* and I am a little bit confused: The figure showing the current flowing inside a circle [1] is obtained for a magnetic field B=0 ! But we know that for B=0, the Hamiltonian is a real symmetric matrix and thus diagonalized by a real orthogonal matrix, so the eigenvectors are real (any phase is just artificial.) By using the formula giving the current we should find zero ! moreover we can ask why we have a current flowing clockwise and not anticlockwise (I remind that B=0) ! When we put B=0.05, we obtain the figures below for the current and the wave function. the wave function has circular symmetry but not the current. Moreover, x and y play exactly the same role so the current should be the same on x and on y (independently from the used gauge) Am I missing something ? Thank you in advance. Adel [1] https://kwant-project.org/doc/1/tutorial/spectrum

3 3

data_voltage
by kamal ghosh 28 Sep '17

28 Sep '17

Dear all,        I want to know the voltage at each site of my device. I have set the terminal voltage and segmented that total voltage in my choice. But, I cannot get the output. Please help me in giving a way out.My code is appended herewith.Regards.K.K.Ghoshkk_ghosh(a)rediffmail.com

1 0

Finding the position of a site in the system
by Luan Vieira de Castro 26 Sep '17

26 Sep '17

Dear all, I am trying to write the current density in a file. In order to write the current in an external file, I have to know the coordinates associated with the current as well. In Kwant < 1.3, I used *sys**.site(i).pos* to know the position of a given tag *i*. In the current version, this is not available anymore. How can I do this now? That is, given *current = J(psi)* (As in the example on the website), how can I know the position associated with *i* in *current[i]*? Best Regards, Luan

2 1

Hamiltonian with tight-binding model
by Alex Currie 20 Sep '17

20 Sep '17

Dear All, I have a Hamaltonian (BHZ hamiltonian) and i'm trying to introduce a magnetic field. I believe I have to use the Peierls phase, however i'm not sure how to implement this after I have finalized the BHZ hamiltonian. Any assistance is appreciated. I will paste the code i have to generate my hamiltonian. You can find more information about the BHZ hamiltonian here: arxiv:0801.0901 <https://arxiv.org/abs/0801.0901> """ import kwant import numpy as np import scipy.linalg as la import scipy.sparse.linalg as sla import sympy import matplotlib.pyplot as plt from ipywidgets import interact Gamma_so = [[0, 0, 0, -1], [0, 0, +1, 0], [0, +1, 0, 0], [-1, 0, 0, 0]] hamiltonian = """ + M * kron(sigma_0, sigma_z) - B * (k_x**2 + k_y**2) * kron(sigma_0, sigma_z) - D * (k_x**2 + k_y**2) * kron(sigma_0, sigma_0) + A * k_x * kron(sigma_z, sigma_x) - A * k_y * kron(sigma_0, sigma_y) + Delta * Gamma_so """ params_bhz=dict(A=364.5, B=-686, D=-512, M=-10, Delta=1.6) hamiltonian = kwant.continuum.sympify(hamiltonian, locals=dict(Gamma_so=Gamma_so)) hamiltonian grid_spacing = 5 template = kwant.continuum.discretize(hamiltonian, grid_spacing=grid_spacing) syst = kwant.wraparound.wraparound(template) syst = syst.finalized() """ Best, Alex

3 4

lead_info
by Prof. CHAN Kwok Sum 20 Sep '17

20 Sep '17

Dear Kwant developer, Thank you for your effort in developing Kwant. I have a question about the definition of the direction of momentum in the smatrix.lead_info. Assume a lead lies along the y direction with primitive vector (0,1). The symmetry operation of a lead is (0,1) if the lead goes towards the positive direction and is (0,-1) if the lead goes towards the negative direction. For the momenta contained in smatrix.lead_info for these two leads , do they use the same coordinate system? Positive momenta mean the direction (0,1), negative momenta mean the direction (0,-1). For velocity, in smatrix.lead_info, it has nothing to do with the coordinate system. Positive means outgoing, negative means incoming in the lead. Thank you KS Chan Disclaimer: This email (including any attachments) is for the use of the intended recipient only and may contain confidential information and/or copyright material. If you are not the intended recipient, please notify the sender immediately and delete this email and all copies from your system. Any unauthorized use, disclosure, reproduction, copying, distribution, or other form of unauthorized dissemination of the contents is expressly prohibited.

3 3