• Home
  • Textbooks
  • A Kinetic View of Statistical Physics
  • Coarsening

A Kinetic View of Statistical Physics

Pavel L. Krapivsky, Sidney Redner, Eli Ben-Naim

Chapter 9

Coarsening - all with Video Answers

Educators


Chapter Questions

01:25

Problem 1

Examine the equation $m_{t}=\nabla^{4} m$ with the "wrong" positive sign on the righthand side. Analyze solutions of the form $m(x, t=0)=\sin k x$. Show that shortwavelength perturbations quickly grow with time.

Ajay Singhal
Ajay Singhal
Numerade Educator
08:51

Problem 2

Complete the derivations of the autocorrelation function $(9.18)$ and the two-body correlation function (9.19).

Chris Trentman
Chris Trentman
Numerade Educator
07:59

Problem 3

Verify that the kinks for the model with the potential $V=\frac{1}{2} m^{2}\left(1-m^{2}\right)^{2}$ are given by Eq. (9.26).

Sam Stansfield
Sam Stansfield
Numerade Educator
15:01

Problem 4

Show that $\psi_{0}$ from (9.33) corresponds to the lowest energy state by the following reasoning:
(a) Verify that $m_{0}(x)$ is a monotonic function of $x$.
(b) Verify that $\psi_{0}$ has no nodes (i.e. zeros) due to (a).
(c) Apply a theorem that, for the one-dimensional Schrödinger equation with an arbitrary potential, the eigenfunction with no nodes has the lowest energy.

Nathan Silvano
Nathan Silvano
Numerade Educator
01:22

Problem 5

Study the stability of the ground state $m_{0}=1$ for the model with potential $V=$ $\frac{1}{2}\left(1-m^{2}\right)^{2}$. Show that the long-time decay of small magnetization perturbations is $\phi \sim e^{-4 t}$

James Kiss
James Kiss
Numerade Educator
01:35

Problem 6

Study the stability of the kink $m_{0}=\left(1+e^{-2 x}\right)^{-1 / 2}$ for the model with potential $V=\frac{1}{2} m^{2}\left(1-m^{2}\right)^{2}$ by the following steps:
(a) Write the corresponding Schrödinger equation.
(b) Sketch the potential of this Schrödinger equation and show that it monotonically increases from 1 at $x=-\infty$ to 4 at $x=\infty$
(c) Using (b) and basic results about the one-dimensional Schrödinger equation, show that there is:
(i) only one discrete level $E_{0}=0$,
(ii) a continuous spectrum in the interval $1<E<4$, with non-degenerate eigenvalues (each eigenfunction is oscillatory when $x \rightarrow-\infty$ and decays as $e^{-x \sqrt{4-E}}$ when $x \rightarrow \infty$ ), and
(iii) finally a continuous spectrum for $E \geq 4$ with doubly degenerate eigenvalues.

Nick Johnson
Nick Johnson
Numerade Educator
02:16

Problem 7

Verify the dependence on parameter $a$ for the terms in the free energy in Eq. (9.51).

James Kiss
James Kiss
Numerade Educator
02:59

Problem 8

The homotopy group $\pi_{i}\left(\mathrm{~S}^{j}\right)$ classifies mappings of the $i$-dimensional sphere $\mathrm{S}^{i}$ into the $j$-dimensional sphere $\mathrm{S}^{j}$. This classification does not distinguish mappings that can be continuously deformed one into the other. When $i=j$, every map from $\mathbb{S}^{j}$ to itself has an integer topological charge measuring how many times the sphere is wrapped around itself. Thus $\pi_{j}\left(\mathrm{~S}^{j}\right)=\mathbb{Z}$ for all $j \geq 1$. Let us analyze the simplest case of mappings of a circle into a circle. The $S^{1} \rightarrow \mathbb{S}^{1}$ mapping can be written as $\theta \rightarrow f(\theta) .$ The angle $\theta$ is defined modulo $2 \pi$
(a) Show that $f(2 \pi)-f(0)=2 \pi n$, where $n$ is an integer called the winding number, $w(f)=n$. The winding number plays a role of a topological charge in the case of the circle. This winding number can be rewritten in an integral form
$$
n=\frac{1}{2 \pi} \int_{0}^{2 \pi} d \theta \frac{d f}{d \theta}
$$This expression is the analog of Eq. (9.62) which gives the topological charge for the mapping of the usual sphere $S^{2}$ into itself.
(b) Show that the mapping
$$
f(\theta)= \begin{cases}\theta, & \text { when } 0 \leq \theta \leq \pi \\ 2 \pi-\theta, & \text { when } \pi \leq \theta \leq 2 \pi\end{cases}
$$
can be deformed into a trivial mapping $f_{0}(\theta)=0$.
(c) Show that the mapping $f_{0}(\theta)=0$ cannot be deformed into the mapping $f_{1}(\theta)=\theta$
(d) Show that the mapping $f_{n}(\theta)=n \theta$ cannot be deformed into the mapping $f_{m}(\theta)=m \theta$ if $n \neq m$
(e) Consider two mappings $f(\theta)$ and $g(\theta)$ and define their product $f * g$ as the mapping that is identical to $f$ for $0<\theta<\pi$, but advancing with twice the speed, and then equal to $g$ but advancing at twice the speed. In equations:
$$
f * g(\theta)= \begin{cases}f(2 \theta), & \text { when } 0 \leq \theta \leq \pi \\ f(2 \pi)+g(2 \theta-2 \pi), & \text { when } \pi \leq \theta \leq 2 \pi\end{cases}
$$
Show that $f_{n} * f_{m}=f_{n+m}$ and that in general the winding numbers add: $w(f * g)=w(f)+w(g) .$ This explains that $\pi_{1}\left(\mathbb{S}^{1}\right)=\mathbb{Z}$ has the structure of the additive group of integers.

Hafiz Shahzaib
Hafiz Shahzaib
Numerade Educator
02:27

Problem 9

Show that any magnetization that satisfies $(9.67)$, and of course the constraint (9.56), is also a solution of the governing equation $(9.58)$.

Harshita Goel
Harshita Goel
Numerade Educator
12:07

Problem 10

Consider the vortex (9.72).
(a) Verify that it is a solution of Eq. (9.67).
(b) Verify that the topological charge of the vortex (9.72) is $q=1$.
(c) Compute the free energy of the vortex $(9.72)$ and show that it is equal to $4 \pi$.
(d) Construct an explicit expression (similar to (9.72)) for the magnetization distribution of a vortex with charge $q=2$

Linda Winkler
Linda Winkler
Numerade Educator
14:01

Problem 11

Consider the $X Y$ model in two dimensions.
(a) Show that the constraint (9.56) can be eliminated by employing the representation $\mathbf{m}=(\cos \phi, \sin \phi)$
(b) Express the free energy in terms of $\phi$ and show that the minimization of this free energy yields $\nabla^{2} \phi=0$
(c) Seek solutions of the equation $\nabla^{2} \phi=0$ in the polar coordinates $(r, \theta) .$ Show that solutions depend only on the angular coordinate, $\phi=\phi(\theta)$, and have the form $\phi=n \theta+$ const., with $n$ being an integer (the winding number).
(d) Compute the free energy of the vortex $\phi_{n}=n \theta$ and show that it is given by
$$
F=\frac{1}{2} \int d \mathbf{x}\left(\nabla \phi_{n}\right)^{2}=\pi n^{2} \int_{0}^{\infty} \frac{d r}{r}
$$This free energy diverges logarithmically (in agreement with our general result that the $X Y$ model does not admit stationary solutions with finite free energy). ${ }^{21}$ The most important vortices are again those with $n=\pm 1$; in the original variables they can be represented as $\mathbf{m}=\pm \mathbf{x} /|\mathbf{x}| .$ In hydrodynamic language, the first vortex is a source and the second is a sink. These vortices are singular, not only in the sense that they have diverging energy - there is also a singularity at the origin where the magnetization is ill-defined.

BL
Blake Lee
Numerade Educator
15:39

Problem 12

Consider the "Mexican hat" (or "wine bottle") potential $V=\left(1-\mathbf{m}^{2}\right)^{2}$, which is a generalization of the double-well potential $V=\left(1-m^{2}\right)^{2} .$ Show that there are no non-trivial stationary solutions in any spatial dimension $d \geq 2$, independent of the dimension of the internal space of the order parameter.

Brian Ketelobeter
Brian Ketelobeter
Numerade Educator
03:08

Problem 13

The quasi-static approximation provided the leading asymptotic behavior of the growth law in Eq. (9.81). Estimate the sub-leading correction to the $t^{1 / 3}$ growth in three dimensions.

Eric Mockensturm
Eric Mockensturm
Numerade Educator
01:40

Problem 14

Verify that for the function $g(z)$ defined in $(9.82)$, equation $g(z)=0$ has a double root (at $z=3 / 2$ ) when $\gamma=\gamma^{*}=4 / 27$

AG
Ankit Gupta
Numerade Educator
01:02

Problem 15

Generalize the LSW theory to arbitrary dimension $d>2$. Verify that the $t^{1 / 3}$ scaling holds independent of $d$. Show that the scaling function becomes
$$
\phi=C_{d} z^{2}(z+3)^{-1-4 d / 9}(3-2 z)^{-2-5 d / 9} e^{-d /(3-2 z)}
$$
for $z<3 / 2$ and $\phi(z)=0$ for $z \geq 3 / 2$

Hast Aggarwal
Hast Aggarwal
Numerade Educator
01:47

Problem 16

Consider LSW coarsening for $d=2$. The absence of an isotropic stationary solution of the Laplace equation that vanishes at infinity leads to a failure of the standard theory. Use the approach developed in Section $2.7$ for the reaction rate and show that the result is the slightly slower growth $R_{c} \sim(t / \ln t)^{1 / 3}$

Hast Aggarwal
Hast Aggarwal
Numerade Educator
01:45

Problem 17

Complete the derivation of Eq. (9.88).

Manik Pulyani
Manik Pulyani
Numerade Educator
01:18

Problem 18

This problem concerns the scaling function $\Phi(z)$ for extremal dynamics that appears in Eq. $(9.92)$
(a) Compute $\Phi(z)$ in the range $3 \leq z \leq 5$
(b) Compute $\Phi(z)$ in the range $5 \leq z \leq 7$
(c) Show that $\Phi(z)$ has discontinuities at the odd integers $z=1,3,5,7, \ldots$ that become progressively weaker: a discontinuity in $\Phi$ at $z=1$, a discontinuity in $\Phi^{\prime}$ at $z=3$, a discontinuity in $\Phi^{\prime \prime}$ at $z=5$, etc.
(d) Show that $\Phi(z)$ decays exponentially at the $z \rightarrow \infty$ limit, $\Phi(z) \sim A e^{-a z}$. Compute $A$ and $a$.

Varsha Aggarwal
Varsha Aggarwal
Numerade Educator
05:01

Problem 19

Consider the one-dimensional nucleation and growth process.
(a) Let $h(x, t)$ be the density of holes of length $x$ at time $t$. Show that this hole length distribution evolves according to
$$
\left(\frac{\partial}{\partial t}-2 v \frac{\partial}{\partial x}\right) h(x, t)=2 \gamma(t) \int_{x}^{\infty} h(y, t) d y-\gamma(t) x h(x, t)
$$(b) Show that the density of holes (or islands) $n(t)$ and the fraction $1-\rho(t)$ of uncovered space are given by
$$
n(t)=\int_{0}^{\infty} h(x, t) d x, \quad 1-\rho(t)=\int_{0}^{\infty} x h(x, t) d x
$$
(c) Show that for instantaneous nucleation, $\gamma(t)=\sigma \delta(t)$, the governing equation turns into the wave equation $h_{t}-2 v h_{x}=0$ whose general solution is $h(x, t)=$ $h_{0}(x+2 v t) .$ Show that the appropriate initial condition is $h_{0}(x)=\sigma^{2} e^{-\sigma x}$ and therefore $h(x, t)=\sigma^{2} e^{-\sigma x} e^{-2 \sigma v t}$
(d) Show that for homogeneous nucleation, $\gamma(t)=\gamma$ when $t>0$, the hole length distribution becomes
$$
h(x, t)=(\gamma t)^{2} e^{-\gamma x t} e^{-\gamma v t^{2}}
$$
(e) Let $g(x, t)$ be the density of holes of length $x$ at time $t$. Show that this distribution obeys
$$
\begin{aligned}
\left(\frac{\partial}{\partial t}+2 v \frac{\partial}{\partial x}\right) g(x, t)=& \gamma(t)[1-\rho(t)] \delta(x)+2 v \frac{h(0, t)}{[n(t)]^{2}} \\
& \times\left[\int_{0}^{x} g(y, t) g(x-y, t) d y-2 n(t) g(x, t)\right]
\end{aligned}
$$

Khoobchandra Agrawal
Khoobchandra Agrawal
Numerade Educator
01:12

Problem 20

Consider nucleation and growth in one dimension with initial seeding.
(a) Derive the two-point correlation function (9.99).
(b) Write the correlation function in terms of $z=x /(2 v t)$ and show that the correlation function depends on coverage and spatial dimension only as in (9.102b).
(c) Obtain the overlap function $F_{1}(z)$.

Christopher Stanley
Christopher Stanley
Numerade Educator
01:02

Problem 21

Compute the normalized overlap function $F_{d}(z)$ for the case of initial seeding of nucleation sites.

Tyler Moulton
Tyler Moulton
Numerade Educator
11:10

Problem 22

Obtain a formal expression for the $n$-point correlation function
$$
g_{n}\left(\mathbf{x}_{1}, \mathbf{x}_{2}, \ldots, \mathbf{x}_{n}\right)=\left\langle\varphi\left(\mathbf{x}_{1}\right) \varphi\left(\mathbf{x}_{2}\right) \cdots \varphi\left(\mathbf{x}_{n}\right)\right\rangle
$$
in $d$ dimensions for nucleation and growth with continuous nucleation at constant rate $\gamma$.

Abhirup Pal
Abhirup Pal
Numerade Educator
02:08

Problem 23

The three-point correlation function is governed by a union of three exclusion zones. Express this union of exclusion zones as a linear combination of single-point exclusion volume, and the volume of overlap between two and three exclusion zones.

Harshita Goel
Harshita Goel
Numerade Educator