Physica A 418 (2015) 78–87

Contents lists available at ScienceDirect

Physica A
journal homepage: www.elsevier.com/locate/physa

An introduction to Monte Carlo methods

J.-C. Walter a,∗ , G.T. Barkema b,c
a
Laboratoire Charles Coulomb UMR 5221 & CNRS, Université Montpellier 2, 34095 Montpellier, France
b
Institute for Theoretical Physics, Utrecht University, The Netherlands
c
Instituut-Lorentz, Universiteit Leiden, P.O. Box 9506, 2300 RA Leiden, The Netherlands

highlights
• We describe the main principles of Monte Carlo simulations.
• We illustrate the method with algorithms for the Ising model.
• We discuss dynamical effects such as thermalization and correlation time.

article info abstract

Article history: Monte Carlo simulations are methods for simulating statistical systems. The aim is to gen-
Available online 25 June 2014 erate a representative ensemble of configurations to access thermodynamical quantities
without the need to solve the system analytically or to perform an exact enumeration. The
Keywords:
main principles of Monte Carlo simulations are ergodicity and detailed balance. The Ising
Monte Carlo simulations
model is a lattice spin system with nearest neighbor interactions that is appropriate to il-
Ising model
Algorithms lustrate different examples of Monte Carlo simulations. It displays a second order phase
transition between disordered (high temperature) and ordered (low temperature) phases,
leading to different strategies of simulations. The Metropolis algorithm and the Glauber
dynamics are efficient at high temperature. Close to the critical temperature, where the
spins display long range correlations, cluster algorithms are more efficient. We introduce
the rejection free (or continuous time) algorithm and describe in details an interesting al-
ternative representation of the Ising model using graphs instead of spins with the so-called
Worm algorithm. We conclude with an important discussion of the dynamical effects such
as thermalization and correlation time.
© 2014 Elsevier B.V. All rights reserved.

1. Introduction

Most models in statistical physics are not solvable analytically, and therefore an alternative way is needed to determine
thermodynamical quantities. Numerical simulations help in this task, but introduce another challenge: it is not possible,
in most cases, to enumerate all the possible configurations of a system; one therefore has to create a set of configurations
that are representative for the entire ensemble. In this section, we will illustrate our purpose with the Ising model. This is a
renowned model because of its simplicity and success in the description of critical phenomena [1]. The degrees of freedom
are spins Si = ±1 placed at the vertex i of a lattice. This lattice will be square or cubic for simplicity, with edge size L and

∗ Corresponding author. Tel.: +33 673985696.

E-mail addresses: [email protected] (J.-C. Walter), [email protected] (G.T. Barkema).

http://dx.doi.org/10.1016/j.physa.2014.06.014
0378-4371/© 2014 Elsevier B.V. All rights reserved.
 J.-C. Walter, G.T. Barkema / Physica A 418 (2015) 78–87 79

dimension D. Thus, the system contains N = LD spins. The Hamiltonian of the Ising model is:

H = −J Si Sj , (1)
⟨ij⟩

where the summation runs over all pairs of nearest-neighbor spins ⟨ij⟩ of the lattice and J is the strength of the interaction.
The statistical properties of the system are obtained from the partition function:

e−β E (C ) ,

Z = (2)
C

where the summation runs over all the configurations C . The energy of a configuration is denoted by E (C ). Here, β ≡ 1/(kB T )
is the inverse temperature (temperature T and Boltzmann constant kB ). The Ising model displays a second-order phase tran-
sition at the temperature Tc , characterized by a high temperature phase with an average magnetization zero (disordered
phase) and a low temperature phase with a non-zero average magnetization (ordered phase). The system is exactly solvable
in one and two dimensions. For D ≥ 4, the critical properties are easily obtained by the renormalization group. In three
dimensions no exact solution is available. Even a 3D cubic lattice of very modest size 10 × 10 × 10 generates 21000 ≈ 10301
configurations in the partition function. If we want to obtain e.g. critical exponents with a sufficient accuracy, we need sizes
that are at least an order of magnitude larger. An exact enumeration is a hopeless effort. Monte Carlo simulations are one
of the possible ways to perform a sampling of configurations. This sampling is made out of a set of configurations of the
configuration space that contributes the most to the averages, without the need of generating every single configuration.
This is referred to as importance sampling. In this sampling of the configuration space, it is important to choose the appro-
priate Monte Carlo scheme to reduce the computational time. In that respect, the Ising model is interesting because the
different regimes in temperature lead to the development of new algorithms that reduce tremendously the computational
time, specifically close to the critical temperature.
We will start these notes by introducing two important principles of Monte Carlo simulations: detailed balance and
ergodicity. Then we will review different examples of Monte Carlo methods applied to the Ising model: local and cluster
algorithms, the rejection free (or continuous time) algorithm, and another kind of Monte Carlo simulations based on an
alternative representation of the spin system, namely the so-called Worm algorithm. We continue with discussing dynamical
quantities, such as the thermalization and correlation times.

2. Principles of MC simulations: ergodicity and detailed balance condition

The basic idea of most Monte Carlo simulations is to iteratively propose a small random change in a configuration Ci , re-
sulting in the trial configuration Cit+1 (the index ‘‘t’’ stands for trial). Next, the trial configuration is either accepted, i.e. Ci+1 =
Cit+1 , or rejected, i.e. Ci+1 = Ci . The resulting set of configurations for i = 1 . . . M is known as a Markov chain in the config-
uration space of the system. We define PA (t ) as the probability to find the system in the configuration A at the time t and
W (A → B) the transition rate from the state A to the state B. This Markov process can be described by the master equation:
dPA (t ) 
= [PB (t )W (B → A) − PA (t )W (A → B)] , (3)
dt A̸=B

with the condition W (A → B) ≥ 0 and B W (A → B) = 1 for all A and B. The transition probability W (A → B) can be fur-

ther decomposed into a trial proposition probability T (A → B) and an acceptance probability A(A → B) so that W (A → B)
= T (A → B) · A(A → B). A proposed change in the configuration is usually referred to as a Monte Carlo move. Conven-
tionally, the time scale in Monte Carlo simulations is chosen such that each degree of freedom of the system is proposed to
change once per unit time, statistically.
The first constraint on this Markov chain is called ergodicity: starting from any configuration C0 with nonzero Boltzmann
weight, any other configuration with nonzero Boltzmann weight should be reachable through a finite number of Monte
Carlo moves. This constraint is necessary for a proper sampling of the configuration space, as otherwise the Markov chain
will be unable to access a part of configuration space with a nonzero contribution to the partition sum.
Apart from a very small number of peculiar algorithms, a second constraint is known as the condition of detailed balance.
For every pair of states A and B, the probability to move from A to B, as well as the probability for the reverse move, are
related via:
PA · T (A → B) · A(A → B) = PB · T (B → A) · A(B → A). (4)
The meaning of this condition can be seen in Eq. (3): a stationary probability (i.e. dPA /dt = 0) is reached if each individual
term in the summation on the right hand side cancels. This prevents the Markov chain to be trapped in a limit cycle [2]. This
is a strong, but not necessary, condition. We mention that generalizations of Monte Carlo process that do not satisfy detailed
balance exist. The combination of ergodicity and detailed balance assures a correct algorithm, i.e., given a long enough time,
the desired distribution probability is sampled.
The key question in Monte Carlo algorithms is which small changes one should propose, and what acceptance probabil-
ities one should choose. The trial proposition and acceptance probabilities have to be well chosen so that the probability of
80 J.-C. Walter, G.T. Barkema / Physica A 418 (2015) 78–87

sampling of a configuration A (after thermalization) is equal to the Boltzmann weight:

e−β EA
PA = , (5)
Z
in which EA is the energy of configuration A. The knowledge of the partition function Z is not necessary because the transition
probabilities are constructed with the ratio of probabilities. The detailed balance condition (4), using (5), can be rewritten as:

T (B → A) · A(B → A) PA
= = e−β(EA −EB ) . (6)
T (A → B) · A(A → B) PB

3. Local MC algorithms: Metropolis and Glauber

One often-used approach to realize detailed balance is to propose randomly a small change in state A, resulting in an-
other state B, in such a way that the reverse process (starting in B and then proposing a small change that results in A) is
equally likely. More formally, a process in which the condition T (A → B) = T (B → A) holds for all pairs of states {A, B}.
For example, taking the example of an Ising model containing N spins, it corresponds to chose randomly one of the spins on
the lattice, therefore T (A → B) = T (B → A) = 1/N. Detailed balance allows for a common scale factor in the acceptance
probabilities for the forward and reverse Monte Carlo moves, but being probabilities, they cannot exceed 1. Simulations are
then fastest if the largest of the two acceptance probabilities is equal to 1, i.e. either A(A → B) or A(B → A) is equal to
1. These conditions (including detailed balance) are realized by the so-called Metropolis algorithm, in which the acceptance
probability is given by:

Amet (A → B) = min [1, PB /PA ] = min [1, exp(−β(EB − EA ))] . (7)

Thus, a proposed move that does not raise the total energy is always accepted, but a proposed move which results in higher
energy is accepted with a probability that decreases exponentially with the increase of the energy difference. For the sake
of illustration, let us describe how a simulation of the Ising model looks like:
1. initialize all spins (either random or all up);
2. perform N random trial moves (N = LD ):
(a) randomly select a site
(b) compute the energy difference 1E = EB − EA if the trial (here a spin flip) induces a change in energy
(c) generate a random number rn uniformly distributed in [0, 1]
(d) if 1E < 0 or if rn < exp(−1E ): flip the spin;
3. perform measurement of some observables.
The step 2 corresponds to one unit time step of the Monte Carlo simulation. An alternative to the Metropolis algorithm is
the Glauber dynamics [3]. The trial probability is the same as Metropolis i.e. T (A → B) = T (B → A) = 1/N. However the
acceptance probability is now:

e−β(EB −EA )
Agla (A → B) = , (8)
1 + e−β(EB −EA )
which also satisfies the detailed balance condition Eq. (6).

4. Cluster algorithms: the example of the Wolff algorithm

Many models encounter phase transitions at some critical temperature. The paradigmatic example for the second order
phase transitions is the Ising model defined in Eq. (1). In the vicinity of the critical temperature, the spins display critical
fluctuations. As shown in Fig. 1 (middle), large aligned spin domains appear. This phenomenon is associated with (i) the
divergence of the correlation length ξ of the connected spin–spin correlation function C (|i − j|) = ⟨Si · Sj ⟩ − ⟨Si ⟩2 (ii) the
divergence of the correlation time of the autocorrelation function C (|t − t ′ |) = ⟨Si (t ) · Si (t ′ )⟩ − ⟨Si (t )⟩⟨Si (t ′ )⟩. Moreover, the
correlation time increases with the size of the system like τ ∼ Lzc where zc is the critical dynamical exponent. For the 2D Ising
model simulated with the Metropolis algorithm, zc = 2.1665(12) [5]. This phenomenon of critical slowing down reflects
the difficulty to change the magnetization of a correlated spin cluster. Take again the example of a 2D spin system where
one spin has four nearest neighbors. If this spin is surrounded by aligned spins, its contribution to the energy is EA = −4J
and after the reversal of this spin, this becomes EB = 4J. Right at Tc ≈ 2.269, the acceptance probability is low for the
Metropolis algorithm: A(A → B) = e−8βc J = 0.0294 . . . . Thus, most of the flipping attempts are rejected. Making matters
worse, even if such a spin with aligned neighbors is flipped, the next time it is selected, it will surely flip back. Only spin
flips at the edge of a cluster have a significant effect over a longer time; but their fraction becomes vanishingly small when
the critical temperature is approached and the cluster size diverges.
 J.-C. Walter, G.T. Barkema / Physica A 418 (2015) 78–87 81

Fig. 1. Snapshots of the 2D Ising model defined in Eq. (1) at three different temperatures: from left to right, T ≪ Tc , T ≈ Tc and T ≫ Tc where Tc is the
critical temperature. White and blacks dots denote spins up and down. The system size is 200 × 200. At Tc (middle), we observe large clusters of correlated
spins: these are critical fluctuations that slow down Monte Carlo simulations when local algorithms such as Metropolis or Glauber are used. This critical
slowing down is reduced by non-local (or cluster) algorithms like the Wolff algorithm [4].

A B

Fig. 2. Sketch of one iteration of the non-local algorithm introduced by Wolff [4] between two spin configurations A and B. The white and black dots stand
for spins of opposite signs. The spins within the loop (dashed line) belong to the same cluster. The steps to form the cluster are: (i) choose randomly a
seed spin; (ii) add aligned spins with the probability Padd (see text); (iii) add iteratively aligned neighbors of newly added spins with the probability Padd ;
(iv) flip all the spins in the cluster at once when the cluster is completed. This is an efficient algorithm for the Ising model at criticality.

One remedy is to develop a non-local algorithm that flips a whole cluster of spins at once. Such an algorithm has been
designed for the Ising model by Wolff [4], following the idea of Swendsen and Wang [6] for more general spin systems. A
sketch of this procedure is shown in Fig. 2.
The procedure consists of first choosing a random initial site (seed site). Then, we add each neighboring spin, provided it
is aligned, with the probability Padd . If it is not aligned, it cannot belong to the cluster. This step is iteratively repeated with
each neighbor added to the cluster. When no neighbor can be added to the cluster anymore, all the spins in the cluster flip at
once. The probability to form a certain cluster of spins in state A before the Wolff move is the same as that in state B after the
Wolff move, except for the aligned spins that have not been added to the cluster at the boundaries. The probability not to add
an aligned spin is 1 − Padd . If m and n stand for non-added aligned spins to the cluster for A and B, T (A → B)/T (B → A) =
(1 − Padd )m−n and the detailed balance condition (6) can be rewritten as:
T (A → B) · A(A → B) A(A → B)
= (1 − Padd )m−n = e−β(EB −EA ) . (9)
T (B → A) · A(B → A) A(B → A)
Noticing that EA − EB = 2J (n − m), it follows that:

A(A → B)
= (1 − Padd )e2β J
n−m
.

(10)
A(B → A)

Therefore, choosing Padd = 1−e−2β J , the acceptance probabilities simplifies: A(A → B) = A(B → A) = 1. For this reason the
spins can be automatically flipped when the cluster is formed. In the vicinity of the critical point, the Wolff algorithm signifi-
cantly reduces the autocorrelation time and the critical dynamical exponent compared to a local algorithm (such as Metropo-
lis or Glauber). We notice that the time τ̃W measured in units of Wolff iterations involves a subset of spins corresponding to
the averaged size ⟨p⟩ of a cluster. On the other hand a time τM measured in units of Metropolis iterations involves all spins of
the network i.e. N = LD spins. To compare the efficiency of both algorithms fairly, it is therefore necessary to define a rescaled
Wolff autocorrelation time τW = τ̃W ⟨p⟩/LD . Moreover, ⟨p⟩ and the magnetic susceptibility χ are related via χ = β⟨p⟩ [2].
It is well known that χ behaves at criticality as χ ∼ Lγ /ν where γ and ν are critical exponents [1]. We also define critical
82 J.-C. Walter, G.T. Barkema / Physica A 418 (2015) 78–87

dynamical exponents z̃cW and zcW for τ̃W and τW , respectively. Assuming τ̃W ∼ Lz̃c , we obtain τW ∼ Lz̃c +γ /ν−D . It leads to the
W W

definition of the dynamical critical exponent zcW = z̃cW + γ /ν − D. In 2D for example, remarkably, the dynamical exponent
is zcW ≈ 0 for the Wolff algorithm (see e.g. Refs. [7,8] and references therein) whereas zcM = 2.1665(12) [5] for Metropolis.

5. Continuous-time or rejection free algorithm

As we have seen in the previous subsection, with a local algorithm (like e.g. Metropolis), a spin flip of the Ising model at
criticality has a high probability to be rejected, and this holds even more in the ferromagnetic phase. A significant amount
of the computational time will therefore be spent without making the system evolve. An alternative way has been proposed
by Gillespie [9] in the context of chemical reactions and afterwards applied by Bortz, Kalos and Lebowitz in the context of
spin systems [10].
This algorithm lists all possible Monte Carlo moves that can be performed in the system in its current configuration.
One of these moves is chosen randomly according to its probability, and the system is forced to move into this state. The
time step of evolution during such a move can be estimated rigorously. This time will change from each configuration and
cannot be set to unity as in the Metropolis algorithm: it takes a continuous value. This is why this algorithm is sometimes
called continuous time algorithm. On the one hand, this algorithm has to maintain a list of all possible moves, which requires
a relatively heavy administrative task, on the other hand, the new configuration is always accepted and this saves a lot of
time when the probability of rejection would otherwise be high. It is also sometimes called the rejection free algorithm. The
efficiency of this algorithm will be maximized for T ≤ Tc . In detail, one iteration of the continuous time algorithm looks like:
1. list all possible moves from the current configuration. Each ofthese n moves has an associated probability Pi ;
n
2. calculate the integrated probability that a move occurs Q = i=1 Pi ;
3. generate a random number rn1 uniformly distributed in [0, Q ]. This selects the chosen move with probability Pi /Q ;
4. estimate the time elapsed during the move: 1t = Q −1 ln (1 − rn2 ) where rn2 is a random number uniformly distributed
in [0, 1[.1
Implementation of this algorithm becomes easier if the probabilities Pi can only take a small number of values. In that
case, lists can be made of all moves with the same probability Pi . The selection process is then first to select one of the
lists, with the appropriate probability, after which randomly one move is selected from that list. This is the case e.g. in Ising
simulations on a square (2D) or cubic (3D) lattice, when the probability Pi is limited to the values 1, e−4β J , e−8β J , or e−12β J
(the latter occurring only in 3D).

6. The Worm algorithm

We present here another example of a local algorithm, the so-called Worm algorithm introduced by Prokof’ev, Svistunov
and Tupitsyn [15,16]. The difference with the algorithms presented above is an alternative representation of the system,
in terms of graphs instead of spins. The Markov chain is therefore performed along graph configurations rather than spin
configurations, but always with Metropolis acceptance rates. The principle is based on the high temperature expansion of
the partition function. Suppose that we want to sample the magnetic susceptibility of the Ising model. We can access it via
the correlation function using the (discrete) fluctuation–dissipation theorem:
β
χ= G(i − j), (11)
N i ,j

where G(i − j) = ⟨Si · Sj ⟩ − ⟨Si ⟩2 is the connected correlation function between sites i and j. In the high temperature phase,
the average value of the spin cancels and G(i − j) = ⟨Si · Sj ⟩. The first step is to write the correlation function G(i − j) of the
Ising model in the following form:
1 βJ

G(i − j) = Si · Sj e ⟨k,l⟩ Sk ·Sl , (12)
Z {S }

1  
= cosh(β J )DN Si · Sj (1 + Sk · Sl tanh(β J )) . (13)
Z {S } ⟨k,l⟩

The configurations that contribute to the sum in (13) contain an even number of spins in the product at any given site.
Other products involving an odd number of spins in the product contribute zero. Each term can be associated with a path
determined by the sites involved in it. A contribution to the sum is made of a (open) path joining sites i and j and closed
loops. The sum over the configurations can be replaced by a sum over such graphs. Fig. 3 sketches such a contribution for
a given couple of source sites i and j. The importance sampling is no longer made over spin configurations but over graphs

1 The probability of a spin flip is exponential versus Q : P (1t ) = exp(−Q 1t ).

 J.-C. Walter, G.T. Barkema / Physica A 418 (2015) 78–87 83

i i

A B

Fig. 3. Illustration of a move with the Worm algorithm. Thick lines stand for one example of graph contributing to the correlation function: one path
joining the sites i et j (the sources) and possibly closed loops. These two graphs differ in one iteration of the Worm algorithm. According to (13), the graph
on the left and on the right have respective equilibrium probabilities PA ∝ tanh5 β J and PB ∝ tanh6 β J (we neglect loops that are not relevant for this
purpose). From the detailed balance condition (4), the Metropolis acceptance rates are A(A → B) = min(1, tanh β J ) and A(B → A) = min(1, 1/ tanh β J ).
In both cases T (A → B) = T (B → A) = 1/2D where D is the dimension of the (hypercubic) lattice.

that are generated as follows. One of the two sources, say i, is mobile. At every steps, it moves randomly to a neighboring
site. Any nearest-neighbor site can be chosen with the trial probability T (A → B) = 1/2D, where D is the dimension of the
(hypercubic) lattice. If no link is present between the two sites, then a link is created with the acceptance probability:
A(A → B) = min(1, tanh β J ). (14)
If a link is already present, it is erased with the acceptance probability:
A(A → B) = min(1, 1/ tanh β J ). (15)
Since 0 ≤ tanh x < 1 for all values x > 0, the probability (15) is equal to unity and the link is always erased. These proba-
bilities are obtained considering the Metropolis acceptance rate Eq. (7) and the expression of the correlation function (13).
The procedure is illustrated in Fig. 3.
The open paths in the two graphs of Fig. 3 are made of 5 (graph on the left) and 6 (graph on the right) lattice spacings (we
neglect the loop that does not contribute in this example). According to (13), the graphs on the left and on the right have
equilibrium probabilities PA ∝ tanh5 (β J ) and PB ∝ tanh6 (β J ), respectively. The transition probability (7) is thus written
W (A → B) = 1/2D × min(1, tanh β J ) and W (B → A) = 1/2D × min(1, 1/ tanh β J ), in agreement with (14) and (15).
If the two sources meet, they can move together on another random site with a freely chosen transition probability. When
the two sources move together, they leave a closed loop behind that justifies the simultaneous presence of open path and
closed loop in Fig. 3. These loops may disappear if the head of the worm meets them. Compared to the Swendsen–Wang algo-
rithm, the Worm algorithm has a dynamical exponent slightly higher in 2D but significantly lower in 3D [11]. The efficiency
of this algorithm can be improved with the use of a continuous time implementation [12]. The formalism of the Worm algo-
rithm is suitable for high temperature. In the critical region, the number of graphs that contribute to the correlation function
increases exponentially. In order to check the convergence of the algorithm, we can compare it with the Wolff algorithm for
the 5D Ising model with different lattice sizes in Fig. 4 (data from Ref. [12]). The two algorithms give results in good agree-
ment, except in the critical region where the convergence of the Worm algorithm is slower as the lattice size L increases.

7. Dynamical aspects: thermalization and correlation time

In order to perform a sampling of thermodynamical quantities at a given temperature, one has to first thermalize the
system. Usually, it is possible to set up the system either at infinite temperature (all spins random) or in the ground state
(all spins up or down). Let us start from an initial random configuration. If the thermalization takes place above the critical
temperature, then the relaxation is exponential. As we come closer to the critical point, the equilibrium correlation length
becomes larger and the relaxation becomes much slower and eventually algebraic right at Tc . An example of such process for
the 2D Ising model is given in Fig. 5. Initially (t = 0), the system is prepared at infinite temperature, all the spins are random.
Then the Glauber dynamics is applied at the critical temperature. We see the nucleation and the evolution of correlated spin
domains in time (t =0, 10, 100, 1000 from left to right, measured in terms of MC iterations). It is possible to show that in
such a quench, the correlation length grows with time like [13]:

ξ (t ) ∼ t 1/zc , (16)
where zc is the critical dynamical exponent (zc = 2.1665(12) [5] for the 2D Ising model). The time needed to complete
thermalization at criticality is therefore τth ∼ Lzc . In the case of a subcritical quench, the system has to choose between
two ferromagnetic states of opposite magnetization. Again, the relaxation is slow because there is nucleation and growth
84 J.-C. Walter, G.T. Barkema / Physica A 418 (2015) 78–87

Fig. 4. (color online) Comparison of the magnetic susceptibility χ obtained with the Worm algorithm and the Wolff algorithm for the Ising model in 5D
with different lattice sizes. The results are in excellent agreement for both algorithms in the high temperature phase. As we move closer to Tc i.e. in the
critical regime (dashed line ellipse), the convergence of the Worm algorithm is slower as the lattice size increases (keeping all other parameters fixed).
This is explained by an exponential increase in the number of graphs.
Source: Data from Ref. [12].

Fig. 5. Snapshots of the evolution of the 2D Ising of size 200 × 200 after a quench from a disordered state until equilibration at the critical temperature Tc
with the Glauber dynamics. We see the nucleation and the growth of correlated domains. From left to right, t =0, 10, 100, 1000 (expressed in Monte Carlo
unit time after the quench). The thermalization is completed when the correlation length reaches its static value ξ ∼ L. Using Eq. (16), the thermalization
time at criticality behaves like τth ∼ Lzc (where zc ≈ 2.1665(12) [5] is the critical dynamical exponent) and therefore diverges with system size.

of domain of opposite magnetization. We define the typical size of a domain at a time t by Ld (t ). The thermalization pro-
cess involves the growth (coarsening) of these domains, until eventually one domain spans the whole system. Only then,
the equilibrium is reached (in the low temperature phase, the expectation of the absolute value of the magnetization is
nonzero). The motion of the domain walls is mostly diffusive, i.e., Ld (t ) ∼ t 1/z with a dynamical exponent z = 2 [13]. The
walls have to cover a distance ∼L, so that the thermalization time scales as τth ∼ L2 . This time diverges again with system
size. Starting from an ordered state does not help for the critical quench (but it does help to start in the ground state to
thermalize the system at T < Tc ).
Once the system is thermalized, one has to be aware of another dynamical effect: the correlation time. This is the time
needed to perform sampling between statistically uncorrelated configurations. In the high temperature phase, the correla-
tion time is equal to the thermalization time, up to some factor close to unity. This is not surprising, as proper thermalization
requires the configuration to become uncorrelated from the initial state. Practically, in all Monte Carlo simulations, one has
to estimate τ at the temperature of sampling to treat properly the error bars.
In the low temperature phase, after thermalization, the magnetization is either positive or negative, and stays like
that over prolonged periods of time. So-called magnetization reversals do occur now and then, but the characteristic time
between those increases exponentially with system size. Because of the strict symmetry between the parts of configuration
space with positive and negative magnetization, in practice one is not so much interested in the time of magnetization
reversals, but rather in the correlation time τ within the up- or down-phase; and this time is some temperature-dependent
constant, irrespective of the system size provided it is significantly larger than the correlation length.
Let us consider now the two-time spin–spin correlation functions at equilibrium in the framework of dynamical scal-
ing [14]. We will use for this purpose a continuous space, so that the spin Si on the site i is now denoted by S⃗r where ⃗ r is
the position vector. Upon a dilatation with a scale factor b, the equilibrium correlation C (⃗ r , t , |T − Tc |) = ⟨S0 (0) · S⃗r (t )⟩ is
assumed to satisfy the homogeneity relation:

r , t , |T − Tc |) = b−2xσ C r /b, t /bzc , |T − Tc |b1/ν ,

C (⃗
 
(17)
 J.-C. Walter, G.T. Barkema / Physica A 418 (2015) 78–87 85

where xσ is the scaling dimension of magnetization density with 2xσ = η for two-dimensional systems. The motivation for
the last two arguments of the scaling function in Eq. (17) comes from the behavior of the correlation length either with time,
ξ ∼ t 1/zc , or with temperature, ξ ∼ |T − Tc |−1/ν . Setting b = t 1/zc in Eq. (17), we obtain:

r , t ) = t −η/zc C r /t 1/zc , |T − Tc |t 1/(ν zc ) .

C (⃗
 
(18)

The algebraic prefactor corresponds to the critical behavior while the scaling function includes all corrections to it. The
characteristic time:

τ ∼ ξ zc ∼ |T − Tc |−ν zc , (19)

appears as the relaxation time of the system. Here we are interested only in autocorrelation functions for which r = 0.
Moreover, we expect an exponential decay of the scaling function C (t /τ ) in the paramagnetic phase. Therefore, the auto-
correlation function can generally be written at equilibrium as:

e−t /τ
C (t , T ) ∼ . (20)
t η/zc
The spin–spin autocorrelation function C (t , T ) versus time is plotted in Fig. 6 (left) for the 2D Ising model of size N =
√ curves correspond to different inverse temperatures β = 0.35 to 0.39 (the critical inverse temperature
50 × 50. The different
is βc = 1/2 ln(1 + 2) ≈ 0.44). We observe an increase in the autocorrelation time as the temperature comes closer to Tc .
The autocorrelation time can be obtained from a fit of the curve in the main graph, assuming Eq. (20). The result is plotted
versus |T − Tc | in the inset. The numerics tend to the behavior of Eq. (19) as T goes to Tc .
Some other aspects of critical dynamics are interesting to study, for instance, the time evolution of the equilibrium mean-
square displacement of the magnetization. It is defined as:

h(t ) = (M (t ) − M (0))2 .
 
(21)

At small time differences (t < 1), the dynamics consists of sparsely distributed proposed spin flips, each of which has a
nonzero acceptance probability. The difference between the number of flipped up-spins and down-spins is on average zero,
but shows fluctuations. Since these spin flips are uncorrelated and their number scales as LD t, the squared difference scales
also as LD t. Hence, in the short-time regime (t ≈ 1), h(t ) behaves as:

h(t ) ∼ LD t . (22)

The magnetization at long time t > τ (τ ∼ L ) is no longer correlated i.e. ⟨M (t ) · M (0)⟩ ≈ 0. Moreover,
zc
  the expectation
value of the squared magnetization is directly related to the magnetic susceptibility like χ ≡ β/N M 2 where N = LD (for
T ≥ Tc , ⟨M ⟩ ≈ 0). It diverges at the critical temperature with system size as ∼Lγ /ν , implying:

h(t ) = M (t )2 + M (0)2 − 2M (t ) · M (0) ≈ 2 M 2 ∼ LD+γ /ν (t > τ ).

   
(23)
D+γ /ν
Therefore h(t ) has to grow from h(t ≈ 1) ∼ L to h(t ∼ L ) ∼ L D zc
. Assuming a power law behavior, it follows that
h(t ) ∼ t γ /(ν zc ) . Therefore, in this regime, we can assume the following form for h(t ):

h(t ) ∼ LD+γ /ν F (t /Lzc ), (24)

γ /(ν zc )
where F is a scaling function with the limit F (x) = constant when x ≫ 1 and F (x) ∼ x at intermediate times. We
measured the function h(t ) as defined in Eq. (21) in simulations of the 2D Ising model at the critical temperature for various
systems sizes. The scaling function F is plotted in Fig. 6 (right), using the exponents γ = 1.75, ν = 1 and zc = 2.1665. With
increasing system size, the data become increasingly consistent with a simple power law behavior for intermediate times
between the early-time behavior given by Eq. (22) and the time of saturation behaving like ∼Lzc . This power law behavior
corresponds to an instance of anomalous diffusion i.e. a mean-square deviation growing as a power law with an exponent
̸= 1 compatible with:

h(t ) ∼ t γ /(ν zc ) ∼ t 0.81 . (25)

Assuming Eq. (25), the magnetization autocorrelation function for intermediate times (i.e. between times of order unity
and the correlation time, thus spanning many decades) is compatible with the first terms of the Taylor expansion of a
stretched exponential, which might lead to an unusual interpretation of the autocorrelation function:

CM (t ) = ⟨|M (t )| · |M (0)|⟩ ∼ exp −(t /τ )γ /(ν zc ) .

 
(26)

This conjecture compares well with the numerics for the correlation function in Fig. 6 (right, in the inset) and gives evidences
that the dynamical critical exponent zc appears at relatively small times 1 ≪ t ≪ Lzc . This yet unpublished result would be
interesting to investigate in the future.
86 J.-C. Walter, G.T. Barkema / Physica A 418 (2015) 78–87

Fig. 6. (color online) Simulation results for the 2D Ising model. (Left) Spin–spin autocorrelation function C (t ) at different β = J /(kB T ) = 0.35, 0.36, 0.37,
0.38 and 0.39 (N = 50 × 50). The decay is slower as we come closer to βc ≈ 0.44. Inset: assuming the expression Eq. (20), the correlation time τ is plotted
versus |T − Tc | and follows the trend of Eq. (19). (Right) Scaling function h(t )/L2+γ /ν versus t /Lzc at Tc for different lattice sizes. At intermediate times,
it displays an anomalous diffusion compatible with h(t ) ∼ t γ /(ν zc ) = t 0.81 . Inset: the autocorrelation CM (t ) = ⟨|M (0)| · |M (t )|⟩ is compatible with the
stretched exponential Eq. (26), deduced from the behavior of h(t ).

8. Conclusion

In these lecture notes, we provide an introduction to Monte Carlo simulations that are a way to produce a set of represen-
tative configurations of a statistical system. We start with the basic principles: ergodicity and detailed balance. In the next
parts, we present several Monte Carlo algorithms. To illustrate their functioning, we use the example of the Ising model. This
model is defined by scalar spins on a lattice that interact via nearest-neighbor interactions. This is the paradigmatic system
for second order phase transitions: a critical temperature separates a disordered phase at high temperature and an ordered
phase at low temperature. These different regimes induce different strategies for the Monte Carlo simulations. In the disor-
dered phase, local algorithms such as Metropolis or Glauber are efficient. In the critical region, the appearance of long range
correlations have set a computational challenge. It has been solved by the use of cluster algorithms such as the Wolff algo-
rithm that flips a whole cluster of correlated spins. Below the critical temperature, when the probability of spin flip is low,
it is a gain of computational time to program the continuous time algorithm. It forces the system into new configurations,
with a jump in time according to the probability of transition. The last description is an interesting algorithm based on an
alternative representation of the model in terms of graphs instead of spins. We conclude with important considerations on
the dynamics with the thermalization and the correlation time.

Acknowledgments

J-CW thanks Christophe Chatelain for his careful reading of the manuscript and the various collaborations that have
largely inspired these notes. We also thank Raoul Schram for stimulating discussions and the reading of the manuscript.
J-CW is supported by the Laboratory of Excellence Initiative (Labex) NUMEV, OD by the Scientific Council of the University
of Montpellier 2. This work is part of the D-ITP consortium, a program of the Netherlands Organisation for Scientific Research
(NWO) that is funded by the Dutch Ministry of Education, Culture and Science (OCW).

References

[1] J.J. Binney, N.J. Dowrick, A.J. Fisher, M.E. Newman, The Theory of Critical Phenomena, Clarendon Press, Oxford, 1995.
[2] M.E.J. Newman, G.T. Barkema, Monte Carlo Methods in Statistical Physics, Oxford University Press, 1999.
[3] R.J. Glauber, Time-dependent statistics of the Ising model, J. Math. Phys. 4 (1963) 294.
[4] U. Wolff, Collective Monte Carlo updating for spin systems, Phys. Rev. Lett. 62 (1989) 361.
[5] M.P. Nightingale, H.W.J. Blöte, Dynamic exponent of the two-dimensional Ising model and Monte Carlo computation of the subdominant eigenvalue
of the stochastic matrix, Phys. Rev. Lett. 76 (1996) 4548.
[6] R.H. Swendsen, J.S. Wang, Nonuniversal critical dynamics in Monte Carlo simulations, Phys. Rev. Lett. 58 (1987) 86–88.
[7] S. Gündüç, M. Dilaver, M. Aydin, Y. Gündüç, A study of dynamic finite size scaling behavior of the scaling functions-calculation of dynamic critical
index of Wolff algorithm, Comput. Phys. Comm. 166 (2005) 1.
[8] J. Du, B. Zheng, J.-S. Wang, Dynamic critical exponents for Swendsen–Wang and Wolff algorithms obtained by a nonequilibrium relaxation method,
J. Stat. Mech. Theory Exp. (2006) P05004.
[9] D.T. Gillespie, A general method for numerically simulating the stochastic time evolution of coupled chemical reactions, J. Comput. Phys. 22 (1976)
403–434;
Exact stochastic simulation of coupled chemical reactions, J. Phys. Chem. 81 (1977) 2340–2361.
[10] A.B. Bortz, H.M. Kalos, J.L. Lebowitz, A new algorithm for Monte Carlo simulation of Ising spin systems, J. Comput. Phys. 17 (1975) 10–18.
[11] Y. Deng, T.M. Garoni, A.D. Sokal, Dynamic critical behavior of the worm algorithm for the Ising model, Phys. Rev. Lett. 99 (2007) 110601.
[12] B. Berche, C. Chatelain, C. Dhall, R. Kenna, R. Low, J.-C. Walter, Extended scaling in high dimensions, J. Stat. Mech. (2008) P11010.
[13] A.J. Bray, Theory of phase-ordering kinetics, Adv. Phys. 43 (1994) 357.
 J.-C. Walter, G.T. Barkema / Physica A 418 (2015) 78–87 87

[14] P.C. Hohenberg, B.I. Halperin, Theory of dynamic critical phenomena, Rev. Modern Phys. 49 (1977) 435.
[15] N.V. Prokof’ev, B.V. Svistunov, I.S. Tupitsyn, ‘‘Worm’’ algorithm in quantum Monte Carlo simulations, Phys. Lett. A 238 (1998) 253–257.
[16] N. Prokof’ev, B. Svistunov, Worm algorithms for classical statistical models, Phys. Rev. Lett. 87 (2001) 160601.