• Home
  • Textbooks
  • Condensed Matter Physics
  • Microscopic Theories of Conduction

Condensed Matter Physics

Michael P. Marder

Chapter 18

Microscopic Theories of Conduction - all with Video Answers

Educators


Chapter Questions

01:49

Problem 1

Resistance of liquid aluminum: The structure function $S(q)$ has been measured for molten aluminum at $700^{\circ} \mathrm{C}$, and is shown below.
(FIGURE CAN'T COPY)
Use these data, and the pseudopotential of Eq. (10.38), with the values listed after it, to compute the resistivity of aluminum at $700^{\circ} \mathrm{C}$, and compare with the experimental result of $24.7 \mu \Omega \cdot \mathrm{~cm}$. Note that the potential $U(\vec{q})$ in Eq. (18.16) differs from the pseudopotential $U_{\vec{q}}$ in Eq. (10.38) by a factor of $\Omega$, the volume per atom.

Narayan Hari
Narayan Hari
Numerade Educator

Problem 2

Electron in an incommensurate potential: Consider the Hamiltonian
$$
\hat{\mathcal{H}}=\mathrm{t} \sum_l\left\{\frac{1}{2}(|l\rangle(l+1|+| l\rangle\langle l-1|)+\cos (2 \pi l \tau)|l\rangle\langle l|\right\} .
$$
The problem becomes particularly interesting when the potential $\cos (2 \pi l \tau)$ is incommensurate with the lattice sites $|l\rangle$, which happens when $\tau$ is irrational. For example, one might take $\tau$ to be the golden mean,
$$\tau=\frac{\sqrt{5}+1}{2} .$$
(a) Suppose that instead of taking $\tau$ to be exactly the golden mean, one replaces it by rational approximants to the golden mean:
$$\tau_n=\frac{F_{n+1}}{F_n},$$
where $F_n$ is the $n$th Fibonacci number, $F_n=F_{n-1}+F_{n-2}$,
$$F_0=1, F_1=1, F_2=2, F_3=3, F_4=5, F_5=8 \ldots$$
How many bands does (18.149) have when one uses $\tau_n$ for $\tau$ (compare with Problem 7 in Chapter 8).
(b) For $\tau$ now given by Eq. (18.150), the wave functions of the Hamiltonian are a curious intermediate between localized and extended. Demonstrate this fact by assuming that there is an eigenstate at $\mathcal{E}=0$, taking $\psi_0=\langle 0 \mid \psi\rangle=1$ and $\psi_1=\langle 1 \mid \psi\rangle=1$. Then use Eq. (18.149) to compute $\psi_m=\langle m \mid \psi\rangle$ for $m$ on the order of $10^4$ and observe how the magnitude of $\psi_m$ scales with $m$.
One way to do this is to plot the participation ratio
$$
P(m)=\frac{\sum_{l=1}^m \psi_l^4}{\left(\sum_{l=1}^m \psi_l^2\right)^2}
$$
In order to calculate $P(m)$, make use of quantities calculated in order to find $P(m-1)$; do not carry out a sum starting at 0 and going up to $m$ for each individual $P(m)$.
Compare the behavior of the participation ratio with what would be expected for localized states, and what would be expected for extended states.

Check back soon!

Problem 3

Introduction to transfer matrices: Consider a tight-binding model for a one-dimensional chain of atoms with random impurities scattered along it:
$$
\hat{\mathcal{H}}=\sum_l|l\rangle U_l\langle l|+\mathrm{t}|l\rangle\langle l+1|+\mathrm{t}|l+1\rangle\langle l| .
$$
Take $U_l$ to be a random variable occupying all values between $-W / 2$ and $W / 2$.
(a) Show that a solution $|\psi\rangle$ of Schrödinger's equation with energy $\mathcal{E}$ satisfies
$$
\binom{\psi_{l+1}}{\psi_l}=\mathbf{T}\binom{\psi_l}{\psi_{l-1}}
$$
and find the $2 \times 2$ transfer matrix $\mathbf{T}$ making Eq. (18.155) true.
(b) Suppose that $\psi_0=0$ and $\psi_1=1$. Using Eq. (18.155) in a numerical routine, find how $\left|\psi_l\right|^2$ behaves for large $l$ for $\mathcal{E}=0, W / \mathrm{t}=10$, and $W / \mathrm{t}=1$. Plot $\left|\psi_l\right|^2$ versus $l$ on a linear-log plot.
(c) What conclusion can one draw about solutions of Schrödinger's equation? It is helpful to suppose that $U_l=U_{-l}$.

Check back soon!

Problem 4

Tight-binding Hamiltonian in two dimensions:
(a) Consider a square lattice in two dimensions. Write down the tight-binding Hamiltonian, with nearest-neighbor hopping $\mathbf{t}$ and onsite energies $U_l$.
(b) Prepare a numerical routine that takes a wave function $\psi$ on an $8 \times 8$ square lattice with periodic boundary conditions and computes $\hat{\mathcal{H}} \psi$. $\hat{\mathcal{H}}$ is a $64 \times 64$ matrix, and $\psi$ is a 64 -component vector. Set up the routine so that it is easy to change the size of the lattice. The components of $\psi$ must be allowed to be complex numbers.
(c) Set the onsite energy $U_l$ to zero. Set $\psi_{(0,0)}=1$ and all other components of $\psi$ to zero. Set $\mathrm{t} d t / h=0.01$ Compute the time evolution of $\psi$ by
$$
\psi^{(n)}=\left[1-\frac{i d t}{h} \hat{\mathcal{H}}\right] \psi^{(n-1)} .
$$
Actually, this is not a good way to solve Schrödinger's equation, because it does not preserve normalization and is not accurate to high order , but it will do for the present. Find $\psi_{0,0}^{(n)}$ for 4096 values of $n$, and prepare a plot of the real and imaginary parts of $\psi_{(0,0)}$ as a function of time.
(d) Multiply $\psi(t)$ by $\exp [-0.1 t /(\hbar / t)]$, take the fast Fourier transform of the result, and plot real and imaginary parts. To what function computed in this chapter should the result correspond?

Check back soon!

Problem 5

Scaling theory of localization I:
(a) Consider a three-dimensional tight-binding model on a square lattice. Show that solutions $\psi$ of Schrödinger's equation must obey the equation

$$
\binom{\psi_{l+1}}{\psi_l}=\mathbf{T}\binom{\psi_l}{\psi_{l-1}},
$$
where
$$
\mathbf{T}=\left(\begin{array}{cc}
-\hat{\mathcal{H}}_2 & -1 \\
1 & 0
\end{array}\right) .
$$
Here $\hat{\mathcal{H}}_2$ is the tight-binding Hamiltonian for the two-dimensional square lattice, and $\psi_l$ is condensed notation for $\psi_{j, k, l}$. The three indices on $\psi$ label the three-dimensional sites of the tight-binding model. $\psi_i$ is a vector (whose values can be indexed by $j$ and $k$ ) with as many components as a two-dimensional tight-binding model and where the final index $l$ is being treated separately for use with the transfer matrix $\mathbf{T}$.
(b) The transfer matrix for a $2 \times 2$ two-dimensional Hamiltonian with periodic boundary conditions in which all the diagonal elements vanish is
$$
\left(\begin{array}{cccccccc}
0 & -1 & -1 & 0 & -1 & 0 & 0 & 0 \\
-1 & 0 & 0 & -1 & 0 & -1 & 0 & 0 \\
-1 & 0 & 0 & -1 & 0 & 0 & -1 & 0 \\
0 & -1 & -1 & 0 & 0 & 0 & 0 & -1 \\
1 & 0 & 0 & 0 & 0 & 0 & 0 & 0 \\
0 & 1 & 0 & 0 & 0 & 0 & 0 & 0 \\
0 & 0 & 1 & 0 & 0 & 0 & 0 & 0 \\
0 & 0 & 0 & 1 & 0 & 0 & 0 & 0
\end{array}\right) .
$$
Generate an automatic procedure to create this matrix.
(c) Print the matrix for a $3 \times 3$ two-dimensional system.

Check back soon!

Problem 7

Scaling theory of localization II: Let $\mathbf{T}$ be the transfer matrix for a tightbinding Hamiltonian on an $L \times L$ square lattice,
$$
\mathbf{T}_l=\left(\begin{array}{cc}
\mathcal{E}-\hat{\mathcal{H}}_l & -1 \\
1 & 0
\end{array}\right) .
$$
The subscript $l$ is needed because the matrices $\hat{\mathcal{H}}_l$ have random diagonal elements, chosen with equal probability to lie within [ $-W / 2, W / 2$ ]. Let
$$\mathbf{Q}=\prod_{l=1}^L \mathbf{T}_l$$
Then according to Pichard and André (1986), the conductance $G$ of an $L \times L \times L$ cube, in units of $e^2 / h$, is
$$G=\frac{1}{R}=\operatorname{Tr}\left[\frac{2}{\mathbf{Q Q}^*+\left(\mathbf{Q Q}^*\right)^{-1}+2}\right] .$$
(a) Rewrite the expression for $G$ in terms of the eigenvalues of the matrix $\mathbf{Q Q}^*$.
(b) Calculate the resistance of a $3 \times 3 \times 3$ block at $W=0$ and $\mathcal{E}=0$ and find In $R=1.4818$.
(c) Find the resistance of a single $3 \times 3 \times 3$ block at $W=10$ and $\mathcal{E}=0$.
(d) The resistance found in the previous part will depend greatly upon the particular values of the random disorder. To obtain a more meaningful result, carry out averages over many realizations of the randomness. That is, compute $\ln R$ for one $3 \times 3 \times 3$ block, obtaining $\ln R_1$. Without initializing the random number generator, find the resistance of a second $3 \times 3 \times 3$ block, obtaining $\ln R_2$. Continue in this way, generating the series of resistances $\ln R_l$. The fluctuations in resistance are so large that $\bar{R}$ is not well defined, but $\overline{\ln } \bar{R}$ will average well. Let $\overline{\ln R_l}$ be the average of $\ln R$ after $l$ computations. Keep computing until the fluctuations in $\ln R_l$ settle down to within around $4 \%$.

Check back soon!
22:57

Problem 7

Scaling theory of localization III:
Use the numerical routine of problem 6 to compute the conductance $G$, defined in Eq. (18.162), and therefore find the scaling function for localization in three dimensions. That is, reproduce Figure 18.9. This figure was produced by using tens of thousands of cubes of size up to $7 \times 7 \times 7$, but if systems this size are too time-consuming, an adequate figure can be produced with cubes of size up to $5 \times 5 \times 5$.

Robert Schnibbe
Robert Schnibbe
Numerade Educator

Problem 8

Weak localization: The goal of this problem is to carry out the calculations leading from Eq. (18.101) to Eq. (18.103). Specialize to the center of the band, $\mathcal{E}=0$, and write $\mathcal{F}(g, \mathcal{E}=0)=\mathcal{F}(g)$.
(a) Show that
$$\lambda^{-1}=-\int d g \mathcal{F}(g) \ln |g| .$$
(b) Show that if $\mathcal{P}(U)$ is very narrow and centered on zero, then to leading order in $W /$ t one obtains
$$\mathcal{F}(g)=\frac{1}{g^2} \mathcal{F}(-1 / g)$$
(c) Verify that Eq. (18.164) has solution
$$\mathcal{F}=\frac{C}{\sqrt{1+g^4}},$$
where $C$ is a constant.
(d) Show that to normalize $\mathcal{F}, C=0.2696 \ldots$.
(e) Equation (18.165) is completely independent of the probability distribution $\mathcal{P}$. From Eq. (18.101), an approximate form for $\mathcal{F}$ that involves $\mathcal{P}$ is
$$
\mathcal{F}(g)=\frac{\mathrm{t}}{g^2} \int d g^{\prime} \mathcal{P}\left(-\frac{\mathrm{t}}{g}-\mathrm{tg}^{\prime}\right) \frac{C}{\sqrt{1+g^{\prime 4}}}
$$
Iterating Eq. (18.101) further would produce even better approximations for $\mathcal{F}$, but to leading order in $W / \mathrm{t}$ it is not necessary.
(f) Inserting Eqs. (18.166) and (18.81) into Eq. (18.163), and working to leading order in $W / t$, verify Eq. (18.103).

Check back soon!

Problem 9

Luttinger Liquids:
(a) Verify that
$$
\begin{aligned}
{\left[\hat{\rho}_l(-q), \hat{\Psi}\right] } & =-\sum_{q^{\prime}} \frac{2 \pi}{q^{\prime} L} e^{-i q^{\prime} x} \hat{\Psi}\left[\hat{\rho}_l(-q), \hat{\rho}_l\left(q^{\prime}\right)\right] \\
& =-e^{i q x} \hat{\Psi}
\end{aligned}
$$
(b) Verify Eq. (18.145). Use Eq. (13.119) to treat averages of exponentials of Bose operators. Keep only terms that survive the expectation value in the ground state.
(c) Verify Eq. (18.146).

Check back soon!