In Periodic problems and plane-wave discretisations we already summarised the net effect of Bloch's theorem. In this notebook, we will explore some basic facts about periodic systems, starting from the very simplest model, a tight-binding monoatomic chain. The solutions to the hands-on exercises are given at the bottom of the page.
In this model, each site of an infinite 1D chain is a degree of freedom, and the Hilbert space is ℓ2(Z), the space of square-summable biinfinite sequences (ψn)n∈Z.
Each site interacts by a "hopping term" with its neighbors, and the Hamiltonian is H=(………………010……101……010………………)
Exercise 1
Find the eigenstates and eigenvalues of this Hamiltonian by solving the second-order recurrence relation.
Exercise 2
Do the same when the system is truncated to a finite number of N sites with periodic boundary conditions.
We are now going to code this:
function build_monoatomic_hamiltonian(N::Integer, t)
H = zeros(N, N)
for n = 1:N-1
H[n, n+1] = H[n+1, n] = t
end
H[1, N] = H[N, 1] = t # Periodic boundary conditions
H
end
build_monoatomic_hamiltonian (generic function with 1 method)
Exercise 3
Compute the eigenvalues and eigenvectors of this Hamiltonian. Plot them, and check whether they agree with theory.
Now we are going to consider a diatomic chain A B A B ...
, where the coupling
A<->B
(t1) is different from the coupling B<->A
(t2). We will use a new
index α to denote the A
and B
sites, so that wavefunctions are now
sequences (ψαn)α∈{1,2},n∈Z.
Exercise 4
Show that eigenstates of this system can be looked for in the form ψαn=uαeikn
Exercise 5
Show that, if ψ is of the form above (Hψ)αn=(Hku)αeikn,
whereH_k = \left(\begin{array}{cc} 0 & t_1 + t_2 e^{-ik}\\ t_1 + t_2 e^{ik} & 0 \end{array}\right)
Let's now check all this numerically:
function build_diatomic_hamiltonian(N::Integer, t1, t2)
# Build diatomic Hamiltonian with the two couplings
# ... <-t2-> A <-t1-> B <-t2-> A <-t1-> B <-t2-> ...
# We introduce unit cells as such:
# ... <-t2-> | A <-t1-> B <-t2-> | A <-t1-> B <-t2-> | ...
# Thus within a cell the A<->B coupling is t1 and across cell boundaries t2
H = zeros(2, N, 2, N)
A, B = 1, 2
for n = 1:N
H[A, n, B, n] = H[B, n, A, n] = t1 # Coupling within cell
end
for n = 1:N-1
H[B, n, A, n+1] = H[A, n+1, B, n] = t2 # Coupling across cells
end
H[A, 1, B, N] = H[B, N, A, 1] = t2 # Periodic BCs (A in cell1 with B in cell N)
reshape(H, 2N, 2N)
end
function build_diatomic_Hk(k::Integer, t1, t2)
# Returns Hk such that H (u e^ikn) = (Hk u) e^ikn
#
# intra-cell AB hopping of t1, plus inter-cell hopping t2 between
# site B (no phase shift) and site A (phase shift e^ik)
[0 t1 + t2*exp(-im*k);
t1 + t2*exp(im*k) 0 ]
end
using Plots
function plot_wavefunction(ψ)
p = plot(real(ψ[1:2:end]), label="Re A")
plot!(p, real(ψ[2:2:end]), label="Re B")
end
plot_wavefunction (generic function with 1 method)
Exercise 6
Check the above assertions. Use a k of the form 2πlN in order to have a ψ that has the periodicity of the supercell (N).
Exercise 7
Plot the band structure, i.e. the eigenvalues of Hk as a function of k Use the function
build_diatomic_Hk
to build the Hamiltonians. Compare with the eigenvalues of the ("supercell") Hamiltonian frombuild_diatomic_hamiltonian
. In the case t1=t2, how do the bands follow from the previous study of the monoatomic chain?
Exercise 8
Repeat the above analysis in the case of a finite-difference discretization of a continuous Hamiltonian H=−12Δ+V(x) where V is periodic Hint: It is advisable to work through Comparing discretization techniques before tackling this question.
TODO
This solution has not yet been written. Any help with a PR is appreciated.
TODO
This solution has not yet been written. Any help with a PR is appreciated.
TODO
This solution has not yet been written. Any help with a PR is appreciated.
TODO
This solution has not yet been written. Any help with a PR is appreciated.
TODO
This solution has not yet been written. Any help with a PR is appreciated.
TODO
This solution has not yet been written. Any help with a PR is appreciated.
TODO
This solution has not yet been written. Any help with a PR is appreciated.
TODO
This solution has not yet been written. Any help with a PR is appreciated.