Mailman 3 March 2023 - Kwant-discuss

Realistic values of hopping parameters
by Yu Li 18 Oct '23

18 Oct '23

Dear all, Now I’m studying some transport property based on multilayered system, with nonmagnetic layer/ferromagnetic layer (NM/FM) stack. May I ask when creating a tight binding system, and setting hoppings by the following function: syst[(lat(1, 0), lat(1, 1))] = t, how to associate the hopping with a realistic system? Such as hoppings between FM and FM atoms, or between FM and NM atoms? I found in many literatures people just simply put t=1, but I assume the value should depend on the type of atoms. My another question is, is it possible to define more than one type of hopping between two sites? For the tight-binding approximation of p electrons, there are two types of hopping integrals: Vppσ, and Vppπ, so I’m wondering if it’s possible to simulate the effect of both hoppings? Thanks a lot for reading my questions! Cheers, Yu Li

4 3

A two sites impurity on an infinite one-dimensional chain (double quantum dot)
by Gabriel Garcia 01 Apr '23

01 Apr '23

Dear Author, I am trying to reproduce the 2.7 section of Tkwant tutorial: Green Function. When i try to calculate the lesser and retard greens's function occurred this error: green_lesser = [] green_retard = [] for time in times: green.evolve(time, 0) green.refine_intervals() green_lesser.append(green.lesser(site_0, site_0)) green_retard.append(green.retarded(site_0, site_0)) /home/gabriel/.local/lib/python3.9/site-packages/kwantspectrum/kwant_spectrum.py:361: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray. pvec = np.array([b * b / 3, 0, a * a]) --------------------------------------------------------------------------- ImportError Traceback (most recent call last) /tmp/ipykernel_158862/3957660185.py in <module> 7 green.refine_intervals() 8 ----> 9 green_lesser.append(green.lesser(site_0, site_0)) 10 green_retard.append(green.retarded(site_0, site_0)) ~/.local/lib/python3.9/site-packages/tkwant/greenfunctions.py in lesser(self, i, j, root) 375 def lesser(self, i, j, root=0): 376 if self._green_lesser is None: --> 377 self._init_lesser() 378 self._do_jobs(self._jobs_lesser) 379 return self._green_lesser.evaluate(i, j, root) ~/.local/lib/python3.9/site-packages/tkwant/greenfunctions.py in _init_lesser(self) 430 def _init_lesser(self): 431 logger.debug('init G^<') --> 432 self._green_lesser = _GreenGeneric(*self._args, **self._kwargs) 433 434 def _init_retarded(self): ~/.local/lib/python3.9/site-packages/tkwant/greenfunctions.py in __init__(self, refine, *args, **kwargs) 36 37 # no refine on individual wave function based on density ---> 38 self._state0 = manybody.State(*args, **kwargs, refine=False) 39 self._state1 = manybody.State(*args, **kwargs, refine=False) 40 ~/.local/lib/python3.9/site-packages/tkwant/manybody.py in __init__(self, syst, tmax, occupations, params, spectra, boundaries, intervals, refine, combine, error_op, scattering_state_type, manybody_wavefunction_type, mpi_distribute, comm) 1511 tasks.update(tasks_) 1512 -> 1513 self.comm = mpi.get_communicator(comm) 1514 1515 psi_init = self._calc_initial_state(tasks, self.comm) ~/.local/lib/python3.9/site-packages/tkwant/mpi.py in get_communicator(comm) 66 if comm is None: 67 if _COMM is None: ---> 68 communicator_init() 69 comm = _COMM 70 return comm ~/.local/lib/python3.9/site-packages/tkwant/mpi.py in communicator_init(comm) 48 if _COMM is None: 49 if comm is None: ---> 50 from mpi4py import MPI 51 _COMM = MPI.COMM_WORLD.Dup() 52 else: ImportError: libmpi.so.12: cannot open shared object file: No such file or directory Someone knows what should i do? I am beginner in python and kwant.

2 1

tkwant-green function
by 1833356804＠qq.com 01 Apr '23

01 Apr '23

Dear author, I am writing to seek your help regarding an issue I encountered while studying tkwant. Specifically, when trying to calculate the Green functions, I got an error message "AttributeError: module 'tkwant.manybody' has no attribute 'GreenFunction'". I was wondering if you could kindly explain to me what may be the cause of this error. Thank you very much for your time and attention. I greatly appreciate any guidance you could provide.

2 1

tkwant-bandstructures and conductances
by 1833356804＠qq.com 29 Mar '23

29 Mar '23

Dear Tkwant experts, I'm a newcomer to Tkwant and I wonder if it is possible to use Tkwant to calculate Floquet bandstructures and conductances? If yes, what functions should I use? Could you also provide some code examples for me to learn from? Thank you very much for your help.

1 0

discretize_landau attribute
by mgarcia.o＠usal.es 29 Mar '23

29 Mar '23

Hi! I'm a first year PhD student and I know kwant basics, I run python with conda interpreter so I have python 3.10.9 and kwant 1.4.3. My problem came when I try to reproduces the 2.11 Adding magnetic field tutorial cause when I try to run kwant.continuum.discretize_landau(hamiltonian, N=10) this error took place: module 'kwant.continuum' has no attribute 'discretize_landau'. I looked for a solution and I found an email of this list with the same problem but he didn't say how he fixed it. Sorry if this sounds repetitive. I take a look at your gitlab page and I download the kwant package from there and try to copy-paste the kwant carpet with the landau_levels.py (I don't know if this is enough or I have to install independently with the set up or make some connections, please let me know if so). When I did that and I called kwant.continuum.landau_levels.discretize_landau(hamiltonian ...) and it gave me another error because kwant.system don't have NoSymmetry attribute but if I open system.py there is a NoSymmetry Class so I don't understand what is happening. Could you shed some light on this matter? thanks in advance and thank you for the Kwant package, it's great! Sorry if this question is not relevant but I don't find how can I fix it to work with landau_levels.

3 2

TKwant Installation procedure
by Gabriel Garcia 27 Mar '23

27 Mar '23

Hello, I installed Tkwant package by conda command on Linux: conda create -n env-tkwant python=3.7 conda activate env-tkwant conda install tkwant -c conda-forge The installation was success. But, when i tried to use the function tkwant.manybody.GreenFunction an error has occurred: green = tkwant.manybody.GreenFunction(syst, max(times), occupations) -------------------------------------------------------------------------- AttributeError Traceback (most recent call last) /tmp/ipykernel_43617/232427282.py in <module> ----> 1 green = tkwant.manybody.GreenFunction(syst, max(times), occupations) AttributeError: module 'tkwant.manybody' has no attribute 'GreenFunction' What is happening? What did i wrong? Can someone help me? Thanks.

2 2

the bandstructures of Kane-Mele model
by 1833356804＠qq.com 26 Mar '23

26 Mar '23

Dear professionals, I am a beginner and I have encountered a problem with my Kane-Mele model. The band structure I obtained with this parameter is different from the ones obtained by others. I am not sure where the error lies and I would appreciate it if you could help me look into it. Thank you for your time and assistance. I am looking forward to hearing from you. My codes are as follows: import kwant import numpy as np import cmath from matplotlib import pyplot # Pauli matrices s0 = np.identity(2) sz = np.diag([1, -1]) graphene = kwant.lattice.honeycomb() a, b = graphene.sublattices nnn_hoppings_a = (((-1, 0), a, a), ((0, 1), a, a), ((1, -1), a, a)) nnn_hoppings_b = (((1, 0), b, b), ((0, -1), b, b), ((-1, 1), b, b)) nnn_hoppings = nnn_hoppings_a + nnn_hoppings_b W=10 M=0 t1=1 t2=-0.06 phi=np.pi/2 spin_orbit = 1j *t2*cmath.exp(1j*phi)* sz def graphene_shape(pos): x, y = pos return (0<y<W) def onsite(site): return s0 * M bulk_kane_mele = kwant.Builder(kwant.TranslationalSymmetry([1, 0])) bulk_kane_mele[graphene.shape(graphene_shape, (0, 0))] = onsite bulk_kane_mele[graphene.neighbors(1)] = s0*t1 bulk_kane_mele[[kwant.builder.HoppingKind(*hopping) for hopping in nnn_hoppings]] = spin_orbit syst = bulk_kane_mele.finalized() bands = kwant.physics.Bands(syst) j=np.linspace(0, 6, 101) energies = [bands(k) for k in j] pyplot.figure() pyplot.plot(j, energies) pyplot.xlabel("momentum [(lattice constant)^-1]") pyplot.ylabel("energy [t]") pyplot.show()

1 0

the quasi-one-dimensional Haldane model energy band structure
by 1833356804＠qq.com 24 Mar '23

24 Mar '23

Dear professionals, I am a beginner and I have encountered a problem with my Haldane model. The band structure I obtained with this parameter is different from the ones obtained by others. I am not sure where the error lies and I would appreciate it if you could help me look into it. Thank you for your time and assistance. I am looking forward to hearing from you. my codes are as follows: import kwant import numpy as np from math import pi, sqrt, tanh from matplotlib import pyplot #HIDDEN_BEGIN_makesyst Haldane = kwant.lattice.honeycomb() a, b = Haldane.sublattices L=10 W=20 M=2/3 t=1 t2=1/3 phi=np.pi/4 def rec(pos): x, y = pos return 0<=x<=L and 0<=y<=W syst = kwant.Builder() syst[Haldane.shape(rec, (0, 0))] = M syst[Haldane.neighbors()] = t # add second neighbours hoppings syst[Haldane.a.neighbors()] = t2*np.exp(1j*phi) syst[Haldane.b.neighbors()] = t2*np.exp(-1j*phi) syst.eradicate_dangling() sym0 = kwant.TranslationalSymmetry(Haldane.vec((-1, 0))) def lead0_shape(pos): x, y = pos return (0<y<W) sym0.add_site_family(Haldane.sublattices[0] , other_vectors=[(1,-2)]) sym0.add_site_family(Haldane.sublattices[1] , other_vectors=[(1,-2)]) lead0 = kwant.Builder(sym0) lead0[Haldane.shape(lead0_shape, (0, 0))] = M lead0[Haldane.neighbors()] = t # add second neighbours hoppings lead0[Haldane.a.neighbors()] = t2*np.exp(-1j*phi) lead0[Haldane.b.neighbors()] = t2*np.exp(1j*phi) lead0.eradicate_dangling() syst.attach_lead(lead0) syst.attach_lead(lead0.reversed()) #kwant.plot(syst) syst = syst.finalized() bands = kwant.physics.Bands(syst.leads[0]) j=np.linspace(0, 5, 101) energies = [bands(k) for k in j] pyplot.figure() pyplot.plot(j, energies) pyplot.xlabel("momentum [(lattice constant)^-1]") pyplot.ylabel("energy [t]") pyplot.show()

2 1

Re: How does the grid spacing in kwant.continuum.discretize effect the system?
by Clarke, Isobel 24 Mar '23

24 Mar '23

Dear Christoph, Thank you for your help. I will have a look at the discretization for a simpler case and then scale up the system to what I have currently. Best wishes, Isobel ________________________________ From: Christoph Groth Sent: Wednesday, March 22, 2023 16:32 To: Clarke, Isobel Cc: kwant-discuss(a)python.org Subject: Re: [Kwant] Re: How does the grid spacing in kwant.continuum.discretize effect the system? Clarke, Isobel wrote: > I've had a look at the Kwant tutorial about the > discretization. However, the response I am seeing isn't consistent > with the idea of an energy limit being reached. Sorry, I have no more specific idea about what the problem could be in your case. When seeing such inconsistent behavior within a sequence of discretizations, I would try to verify the consistency of that sequence in simpler circumstances. For example, one could verify that the bandstructure evolves as expected as one changes the discretization length. A second possible check would be to do some time-independent transport computation. If you find an inconsistency already with these simpler checks, it will be easier to understand. If you do not, you have narrowed down the problem already.

1 0

Re: How does the grid spacing in kwant.continuum.discretize effect the system?
by Clarke, Isobel 22 Mar '23

22 Mar '23

Dear Christoph, Thank you so much for your reply. I've had a look at the Kwant tutorial about the discretization. However, the response I am seeing isn't consistent with the idea of an energy limit being reached. It appears that as I change the number of lattice points, therefore the grid spacing, the response of the system changes as shown in the attached image. The graphs show the state being evolved in Tkwant with the expectation values of the Pauli spin matrices plotted. For 20, 40 and 120 lattice points the graphs show an expected and similar response. For 60, 70 and 100 the graphs show no evolving of the state or an unphysical response. It appears to be very inconsistent with the grid spacing value. Could there be another cause of this result, for example how the system is built rather than a continuum limit? Thank you for any more help with this. Best wishes, Isobel ________________________________ From: Christoph Groth Sent: Thursday, March 16, 2023 15:05 To: Clarke, Isobel Cc: kwant-discuss(a)python.org Subject: Re: [Kwant] How does the grid spacing in kwant.continuum.discretize effect the system? Hi Isobel, As you decrease the discretization grid spacing you should approach the continuum limit. For example, the band structure should become more and more accurate also for larger momenta. Did you check the Kwant tutorial on the discretizer? In particular the example comparing a=1.0 and a=0.25: https://kwant-project.org/doc/1/tutorial/discretize#limitations-of-discreti… Tkwant should not really matter here. I suggest that you first check whether your discretization behaves as it should without even using tkwant, for example by looking at the band structure. Cheers Christoph

2 1