Scribd
Upload
46 views2 pages

Monte Carlo Simulation of 1D Ising Model: Bo Gazi Ci University Department of Physics Phys 496/68N Fall 2011 Project 2

This document describes a project to simulate a 1D Ising model using Monte Carlo methods. Students are asked to: 1) Simulate the 1D Ising model for systems of N=50 spins, sweeping the system 10x50 times for relaxation and making 20 measurements of N=50 sweeps separated by measurements. 2) Perform simulations for a range of magnetic field (B) values at fixed temperature (T=1K and T=2K) and interaction strength (J=1k). 3) Plot the simulation results alongside the expected analytical solution to compare magnetization (M/N) versus magnetic field (B/k).

Uploaded by

Hemanta Bhattarai
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
46 views2 pages

Monte Carlo Simulation of 1D Ising Model: Bo Gazi Ci University Department of Physics Phys 496/68N Fall 2011 Project 2

This document describes a project to simulate a 1D Ising model using Monte Carlo methods. Students are asked to: 1) Simulate the 1D Ising model for systems of N=50 spins, sweeping the system 10x50 times for relaxation and making 20 measurements of N=50 sweeps separated by measurements. 2) Perform simulations for a range of magnetic field (B) values at fixed temperature (T=1K and T=2K) and interaction strength (J=1k). 3) Plot the simulation results alongside the expected analytical solution to compare magnetization (M/N) versus magnetic field (B/k).

Uploaded by

Hemanta Bhattarai
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

Boğaziçi University

Department of Physics
Phys 496/68N Taylan Akdoğan Fall 2011
Project 2
Due on December 30th , 2011

Monte Carlo Simulation of 1D Ising Model

1D Ising Model

The energy of the system is given by:


N
X N
X
E(S) = −J sk sk+1 − B sk
k=1 k=1

Assume a periodic boundary condition; the last particle is next to the first particle: sN +1 ≡ s1 Write
a Monte Carlo code using Metropolis et. al. algorithm to simulate this system for a given N, J, B,
and T. Note that the probability of being in a state S = {s1 , s2 , ... , sN } is, according to statistical
mechanics, related as:
PS ∼ e−β·E(S)
Then the acceptance comparison is done using

r = exp(−β · ∆E)

where ∆E = E(Strial ) − E(S), and one can show that it is equal to

∆E = 2si [J(si−1 + si+1 ) + B]

for the case of flipping si to strial


i = −si . Note that, ∆E depends on only three particles, and there are
limited number of different ∆E values, thus one can pre-calculate r = exp(−β · ∆E) for all possible
cases and create a short table (possible 3D table) to speed up the simulation by not calculating the
same exponential over and over again.

The magnetization per spin of the system is simply the average of the spins.
N
1 1 X
M= sk
N N k=1

In class, we derived the analytical solution for the average value of magnetization at the thermodynamic
limit. The magnetization per particle for a 1D Ising system at given J, B, and T is calculated by:

1 sinh B/kT
M=q
N sinh2 B/kT + e−4J/kT

Project:

• Take N = 50,

• Take Nsweep = 50 for a measurement

• Start with hot configuration (randomly distributed spins)

• Sweep 10 × Nsweep for relaxation

• Make Nmeasure = 20 measurements with Nsweep = 50 sweeps separated with each measurements
• Make a series of measurements for J = 1k, T = 1 K for B = {−1, ..., 1}k with 0.1k steps.

• Make a series of measurements for J = 1k, T = 2 K for B = {−1, ..., 1}k with 0.1k steps.

• Make a plot showing your simulation results with their corresponding error bars for Nmeasure
measurements, and the expected curve calculated by the analytical solution in a single figure as
shown below:

1
T=1
0.8 T=2

0.6

0.4

0.2
M/N

−0.2

−0.4

−0.6

−0.8

−1
−1 −0.8 −0.6 −0.4 −0.2 0 0.2 0.4 0.6 0.8 1
B/k

You might also like