Part III — Schramm–Loewner Evolutions

Based on lectures by J. Miller

Notes taken by Dexter Chua

Lent 2018

These notes are not endorsed by the lecturers, and I have modified them (often
significantly) after lectures. They are nowhere near accurate representations of what
was actually lectured, and in particular, all errors are almost surely mine.

Schramm–Loewner Evolution (SLE) is a family of random curves in the plane, indexed

by a parameter κ ≥ 0. These non-crossing curves are the fundamental tool used to
describe the scaling limits of a host of natural probabilistic processes in two dimensions,
such as critical percolation interfaces and random spanning trees. Their introduction
by Oded Schramm in 1999 was a milestone of modern probability theory.

The course will focus on the definition and basic properties of SLE. The key ideas are
conformal invariance and a certain spatial Markov property, which make it possible
to use Itô calculus for the analysis. In particular we will show that, almost surely, for
κ ≤ 4 the curves are simple, for 4 ≤ κ < 8 they have double points but are non-crossing,
and for κ ≥ 8 they are space-filling. We will then explore the properties of the curves
for a number of special values of κ (locality, restriction properties) which will allow us
to relate the curves to other conformally invariant structures.

The fundamentals of conformal mapping will be needed, though most of this will be
developed as required. A basic familiarity with Brownian motion and Itô calculus will
be assumed but recalled.

1
Contents III Schramm–Loewner Evolutions

Contents
0 Introduction 3

1 Conformal transformations 4
1.1 Conformal transformations . . . . . . . . . . . . . . . . . . . . . 4
1.2 Brownian motion and harmonic functions . . . . . . . . . . . . . 5
1.3 Distortion estimates for conformal maps . . . . . . . . . . . . . . 6
1.4 Half-plane capacity . . . . . . . . . . . . . . . . . . . . . . . . . . 10

2 Loewner’s theorem 14
2.1 Key estimates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
2.2 Schramm–Loewner evolution . . . . . . . . . . . . . . . . . . . . 19

3 Review of stochastic calculus 21

4 Phases of SLE 24

5 Scaling limit of critical percolation 31

6 Scaling limit of self-avoiding walks 34

7 The Gaussian free field 41

Index 47

2
0 Introduction III Schramm–Loewner Evolutions

0 Introduction
Schramm–Loewner evolution is a random curve in a complex domain D (which
we will often take to be H for convenience), parametrized by a positive real
number κ. This was introduced by Schramm in 1999 to describe various scaling
limits that arise in probability theory and statistical physics.
Recall that if we take a random walk on the integer lattice Zd , and take the
scaling limit as the grid size tends to 0, we converge towards a Brownian motion.
There are other discrete models that admit a scaling limit, and the limit is often
something that is not Brownian motion.
Schramm showed that if the scaling limit satisfies certain conformal invariance
properties, which many models do, then it must be SLEκ for some κ. If we can
identify exactly which κ it belongs to, then this will completely determine what
the scaling limit is, and often conveys a lot of extra information such as the
critical exponents. This has been done for several discrete models:

– The scaling limit of loop-erased random walk is SLE2

– The scaling limit of hexagonal percolation is SLE6 (see chapter 5)
– It is conjectured that the scaling limit of self-avoiding random walk is
SLE8/3 (see chapter 6)

– The “level sets” of a Gaussian free field are SLE4 ’s (see chapter7).
In this course, we will prove some basic properties of SLEκ , and then establish
the last three results/conjectures.

3
1 Conformal transformations III Schramm–Loewner Evolutions

1 Conformal transformations
1.1 Conformal transformations
Definition (Conformal map). Let U, V be domains in C. We say a holomorphic
function f : U → V is conformal if it is a bijection.

We will write D for the open unit disk, and H for the upper half plane. An
important theorem about conformal maps is the following:
Theorem (Riemann mapping theorem). Let U be a simply connected domain
with U =6 C and z ∈ U be any point. Then there exists a unique conformal
transformation f : D → U such that f (0) = z, and f 0 (0) is real and positive.

We shall not prove this theorem, as it is a standard result. An immediate

corollary is that any two simply connected domains that are distinct from C are
conformally equivalent.
Example. Take U = D. Then for z ∈ D, the map promised by the Riemann
mapping theorem is
w+z
f (w) = .
1 + z̄w
In general, every conformal transformation f : D → D is of the form
w+z
f (w) = λ
1 + z̄w
for some |λ| = 1 and z ∈ D.
Example. The map f : H → D given by
z−i
f (z) = ,
z+i
is a conformal transformation, with inverse

i(1 + w)
g(w) = .
1−w
In general, the conformal transformations H → H consist of maps of the form
az + b
f (z) =
cz + d
with a, b, c, d ∈ R and ad − bc 6= 0.
√
Example. For t ≥ 0, we let Ht = H \ [0, 2 ti]. The map Ht → H given by
p
z 7→ z 2 + 4t

is a conformal transformation. Observe that this map satisfies

p
|gt (z) − z| = | z 2 + 4t − z| → 0 as z → ∞.

So gt (z) ∼ z for large z.

4
1 Conformal transformations III Schramm–Loewner Evolutions

Observe also that the family gt (z) satisfies the ODE

∂g 2
= , g0 (z) = z.
∂t gt (z)
√
We can think of these functions gt as being generated by the curve γ(t) = 2 ti,
where for each t, the function gt is the conformal transformation that sends
H \ γ([0, t]) to H (satisfying |gt (z) − z| → ∞).
Given the set of functions gt , we can recover the curve γ as follows — for
each z ∈ H, we can ask what is the minimum t such that gt is not defined on
z. By ODE theorems, there is a solution up till the point when gt (z) = 0, in
which case the denominator of the right hand side blows up. Call this time τ (z).
We then see that for each t, there is a unique z such that τ (z) = t, and we have
γ(t) = z.
More generally, suppose γ is any simple (i.e. non-self-intersecting) curve
in H starting from 0. Then for each t, we can let gt be the unique conformal
transformation that maps H\γ([0, t]) to H with |gt (z)−z| → ∞ as above (we will
later see such a map exists). Then Loewner’s theorem says there is a continuous,
real-valued function W such that
∂gt 2
= , g0 (z) = z.
∂t gt (z) − Wt
This is the chordal Loewner equation. We can turn this around — given √a
function Wt , what is the corresponding
√ curve γ(t)? If W = 0, then γ(t) = 2 ti.
More excitingly, if we take Wt = κBt , where Bt is a standard Brownian motion,
and interpret this equation as a stochastic differential equation, we obtain SLEκ .

1.2 Brownian motion and harmonic functions

Recall that a function f = u + iv is holomorphic iff it satisfies the Cauchy–
Riemann equations
∂u ∂v ∂u ∂v
= , =− .
∂x ∂y ∂y ∂x
This in particular implies that both u and v are harmonic.
Definition (Harmonic function). A function f : Rk → R is harmonic if it is C 2
and  2
∂2

∂
∆f = + ··· + f = 0.
∂x21 ∂x2k
Indeed, we can calculate

∂2u ∂2u
 
∂ ∂v ∂ ∂v
2
+ 2 = + − = 0.
∂x ∂y ∂x ∂y ∂y ∂x

In Advanced Probability, we saw that Brownian motion is very closely related

to harmonic functions.
Definition (Complex Brownian motion). We say a process B = B 1 + iB 2 is a
complex Brownian motion if (B 1 , B 2 ) is a standard Brownian motion in R2 .
Recall some results from Advanced Probability:

5
1 Conformal transformations III Schramm–Loewner Evolutions

Theorem. Let u be a harmonic function on a bounded domain D which is

continuous on D̄. For z ∈ D, let Pz be the law of a complex Brownian motion
starting from z, and let τ be the first hitting time of D. Then

u(z) = Ez [u(Bτ )].

Corollary (Mean value property). If u is a harmonic function, then, whenever

it makes sense, we have
Z 2π
1
u(z) = u(z + reiθ ) dθ.
2π 0

Corollary (Maximum principle). Let u be harmonic in a domain D. If u attains

its maximum at an interior point in D, then u is constant.
Corollary (Maximum modulus principle). Let D be a domain and let f : D → C
be holomorphic. If |f | attains its maximum in the interior of D, then f is constant.
Proof. Observe that if f is holomorphic, then log |f | is harmonic and attains its
maximum in the interior of D. It then follows that |f | is constant (if |f | vanishes
somewhere, then consider log |f + M | for some large M , and do some patching if
necessary). It is then a standard result that a holomorphic function of constant
modulus is constant.
The following lemma should be familiar from the old complex analysis days
as well:
Lemma (Schwarz lemma). Let f : D → D be a holomorphic map with f (0) = 0.
Then |f (z)| ≤ |z| for all z ∈ D. If |f (z)| = |z| for some non-zero z ∈ D, then
f (w) = λw for some λ ∈ C with |λ| = 1.
Proof. Consider the map
(
f (z)/z z 6= 0
g(z) = .
f 0 (0) z=0

Then one sees that g is holomorphic and |g(z)| ≤ 1 for all z ∈ ∂D, hence for all
z ∈ D by the maximum modulus principle. If |f (z0 )| = |z0 | for some z0 ∈ D \ {0},
then g must be constant, so f is linear.

1.3 Distortion estimates for conformal maps

We write U for the collection of conformal transformations f : D → D, where
D is any simply connected domain with 0 ∈ D and D 6= C, with f (0) = 0 and
f 0 (0) = 1. Thus, it must be of the form
∞
X
f (z) = z + an z n .
n=2

Theorem (Koebe-1/4 theorem). If f ∈ U and 0 < r ≤ 1, then B(0, r/4) ⊆

f (rD).
This says f cannot look like this:

6
1 Conformal transformations III Schramm–Loewner Evolutions

By scaling it suffices to prove this for the case r = 1. This follows from the
following result:
Theorem. If f ∈ U, then |a2 | ≤ 2.
The proof of this proposition will involve quite some work. So let us just
conclude the theorem from this.
Proof of Koebe-1/4 theorem. Suppose f : D → D is in U, and z0 6∈ D. We shall
show that |z0 | ≥ 41 . Consider the function

z0 f (z)
f˜(z) = .
z0 − f (z)

Since f˜ is composition of conformal transformations, it is itself conformal, and a

direct computation shows f˜ ∈ U. Moreover, if

f (z) = z + a2 z 2 + · · · ,

then  
˜ 1
f (z) = z + a2 + z2 + · · · .
z0
So we obtain the bounds
1
|a2 |, a2 + ≤ 2.
z0
By the triangle inequality, we must have |z0−1 | ≤ 4, hence |z0 | ≥ 14 .
The 1/4 theorem bounds the distortion in terms of the value of f 0 (0). Con-
versely, if we know how much the distortion is, we can use the theorem to derive
bounds on f 0 (0).
Corollary. Let D, D̃ be domains and z ∈ D, z̃ ∈ D̃. If f : D → D̃ is a conformal
transformation with f (z) = z̃, then

d˜ 4d˜
≤ |f 0 (z)| ≤ ,
4d d
where d = dist(z, ∂D) and d˜ = dist(z̃, ∂ D̃).
Proof. By translation, scaling and rotation, we may assume that z = z̃ = 0,
d = 1 and f 0 (0) = 1. Then we have

D̃ = f (D) ⊇ f (D) ⊇ B(0, 1/4).

So d˜ ≥ 14 , as desired. The other bound follows by considering f −1 .

7
1 Conformal transformations III Schramm–Loewner Evolutions

We now proceed to prove the theorem. We first obtain a bound on area,

using IA Vector Calculus:
Proposition. Let f ∈ U. Then
∞
X
area(f (D)) = π n|an |2 .
n=1

Proof. In the ideal world, we will have something that helps us directly compute
area(f (D)). However, for the derivation to work, we need to talk about what
f does to the boundary of D, but that is not necessarily well-defined. So we
compute area(f (rD)) for r < 1 and then take the limit r → 1.
So fix r ∈ (0, 1), and define the curve γ(θ) = f (reiθ ) for θ ∈ [0, 2π]. Then we
can compute
Z Z
1 1
z̄ dz = (x − iy)(dx + i dy)
2i γ 2i γ
Z
1
= (x − iy) dx + (ix + y) dy
2i γ
ZZ
1
= 2i dx dy
2i f (rD)

= area(f (rD)),
using Green’s theorem. We can also compute the left-hand integral directly as
1 2π
Z Z
1
z̄ dz = f (reiθ )f 0 (reiθ )ireiθ dθ
2i γ 2i 0
∞
! ∞ !
1 2π X
Z X
= ān rn e−iθn an nrn−1 eiθ(n−1) ireiθ dθ
2i 0 n=1 n=1
∞
X
=π r2n |an |2 n.
n=1

Definition (Compact hull). A compact hull is a connected compact set K ⊆ C

with more than one point such that C \ K is connected.
We want to obtain a similar area estimate for compact hulls. By the Riemann
mapping theorem, if K is a compact hull, then there exists a unique conformal
transformation F : C \ D̄ to C \ K that fixes infinity and has positive derivative
at ∞, i.e. lim F (z)/z > 0. Let H be the set of compact hulls containing 0 with
z→∞
lim F (z)/z = 1. If K ∈ H, then the Laurent expansion of the corresponding F
z→∞
is
∞
X bn
F (z) = z + b0 + .
n=1
zn
Observe there is a correspondence between the space of such F and U by sending
F to 1/F (1/z), and vice versa.
Proposition. If K ∈ H, then
∞
!
X
2
area(K) = π 1 − n|bn | .
n=1

8
1 Conformal transformations III Schramm–Loewner Evolutions

Observe that the area is always non-negative. So in particular, we obtain the

bound
X∞
n|bn |2 ≤ 1.
n=1

In particular, |b1 | ≤ 1. This will be how we ultimately bound |a2 |.

Proof. The proof is essentially the same as last time. Let r > 1, and let
Kr = F (rD̄) (or, if you wish, C \ F (C \ rD̄)), and γ(θ) = F (reiθ ). As in the
previous proposition, we have
Z
1
area(Kr ) = z̄ dz
2i γ
1 2π
Z
= F (reiθ )F 0 (reiθ )ireiθ dθ
2i 0
∞
!
X
2 2 −2n
=π r − n|bn | r .
n=1

Then take the limit as r → 1.

By the correspondence we previously noted, this gives us some estimates on
the coefficients of any f ∈ U. It turns out applying this estimate directly is not
good enough. Instead, we want to take the square root of f (z 2 ).
Lemma. Let f ∈ U. Then there exists an odd function h ∈ U with h(z)2 =
f (z 2 ).
Proof. Note that f (0) = 0 by assumption, so taking the square root can poten-
tially be problematic, since 0 is a branch point. To get rid of the problem, define
the function (
f (z)
z 6= 0
f˜(z) = z .
f 0 (0) z = 0

Then f˜ is non-zero and conformal in D, and so there is a function g with

g(z)2 = f˜(z). We then set
h(z) = zg(z 2 ).
Then h is odd and h2 = z 2 g(z 2 )2 = f (z 2 ). It is also immediate that h(0) = 0 and
h0 (0) = 1. We need to show h is injective on D. If z1 , z2 ∈ D with h(z1 ) = h(z2 ),
then
z1 g(z12 ) = z2 g(z22 ). (∗)
By squaring, we know
z12 f˜(z12 ) = z22 f˜(z2 )2 .
Thus, f (z12 ) = f (z22 ) and so z12 = z22 . But then (∗) implies z1 = z2 . So h is
injective, and hence h ∈ U.

We can now prove the desired theorem.

9
1 Conformal transformations III Schramm–Loewner Evolutions

Proof of theorem. We can Taylor expand

h(z) = z + c3 z 3 + c5 z 5 + · · · .
Then comparing h(z)2 = f (z 2 ) implies
a2
c3 = .
2
1
Setting g(z) = h(1/z) , we find that the z −1 coefficient of g is − a22 , and as we
previously noted, this must be ≤ 1.

1.4 Half-plane capacity

Definition (Compact H-hull). A set A ⊆ H is called a compact H-hull if A
is compact, A = H ∩ Ā and H \ A is simply connected. We write Q for the
collection of compact H-hulls.
A large part of the course will be about studying families of compact H-hulls,
and in particular random families of compact H-hulls. In this chapter, we will
establish some basic results about compact H-hulls, in preparation for Loewner’s
theorem in the next chapter. We begin with the following important proposition:
Proposition. For each A ∈ Q, there exists a unique conformal transformation
gA : H \ A → H with |gA (z) − z| → 0 as z → ∞.
This conformal transformation gA will be key to understanding compact
H-hulls. The proof of this requires the following theorem:
Theorem (Schwarz reflection principle). Let D ⊆ H be a simply connected
domain, and let φ : D → H be a conformal transformation which is bounded on
bounded sets and sends R ∩ D to R. Then φ extends by reflection to a conformal
transformation on
D∗ = D ∪ {z̄ : z ∈ D} = D ∪ D̄
by setting φ(z̄) = φ(z).
Proof of proposition. The Riemann mapping theorem implies that there exists
a conformal transformation g : H \ A → H. Then g(z) → ∞ as z → ∞. By the
Schwarz reflection principle, extend g to a conformal transformation defined on
C \ (A ∪ Ā).
By Laurent expanding g at ∞, we can write
N
X
g(z) = b−N z N .
n=−∞

Since g maps the real line to the real line, all bi must be real. Moreover, by
injectivity, considering large z shows that N = 1. In other words, we can write
∞
X bn
g(z) = b−1 z + b0 + ,
n=1
zn

with b−1 6= 0. We can then define

g(z) − b0
gA (z) = .
b−1

10
1 Conformal transformations III Schramm–Loewner Evolutions

Since b0 and b−1 are both real, this is still a conformal transformation, and
|gA (z) − z| → 0 as z → ∞.
0 0 −1
To show uniqueness, suppose gA , gA are two such functions. Then gA ◦ gA :
H → H is such a function for A = ∅. Thus, it suffices to show that if g : H → H
is a conformal mapping such that g(z) − z → 0 as z → ∞, then in fact g = z.
But we can expand g(z) − z as
∞
X cn
g(z) − z = ,
n=1
zn

and this has to be holomorphic at 0. So cn = 0 for all n, and we are done.

Definition (Half-plane capacity). Let A ∈ Q. Then the half-plane capacity of
A is defined to be
hcap(A) = lim z(gA (z) − z).
z→∞
Thus, we have
∞
hcap(A) X bn
gA (z) = z + + .
z n=2
zn
To justify the name “capacity”, hcap(A) had better be non-negative and
increasing in A. These are both true, as we will soon prove. However, let us first
look at some examples.
√ √
Example.
√ z 7→ z 2 + 4t is the unique conformal transformation H\[0, ti] → H
with | z 2 + 4z − z| → 0 as z → ∞. We can expand
p 2t
z 2 + 4t = z + + ··· .
z
√
Thus, we know that hcap([0,
√ 2 ti]) = 2t. This justifies our previous funny
parametrization of [0, ti].
Example. The map z 7→ z+ z1 maps H\ D̄ → H. Again, we have z + 1
z −z →0
as z → ∞, and so hcap(H ∩ D̄) = 1.
Proposition.
(i) Scaling: If r > 0 and A ∈ Q, then hcap(rA) = r2 hcap(A).
(ii) Translation invariance: If x ∈ R and a ∈ Q, then hcap(A + x) = hcap(A).
(iii) Monotonicity: If A, Ã ∈ Q are such that A ⊆ Ã. Then hcap(A) ≤ hcap(Ã).
Proof.
(i) We have grA (z) = rgA (z/r).
(ii) Observe gA+x (z) = gA (z − x) + x.
(iii) We can write
gà = ggA (Ã\A) ◦ gA .
Thus, expanding out tells us
hcap(Ã) = hcap(A) + hcap(gA (Ã \ A)).
So the desired result follows if we know that the half-plane capacity is
non-negative, which we will prove next.

11
1 Conformal transformations III Schramm–Loewner Evolutions

Observe that these results together imply that if A ∈ Q and A ⊆ r(D̄ ∩ H),
then
hcap(A) ≤ hcap(r(D̄ ∩ H)) ≤ r2 hcap(D̄ ∩ H) = r2 .
So we know that hcap(A) ≤ diam(A)2 .
Compared to the above proofs, it seems much less straightforward to prove
non-negativity of the half-plane capacity. Our strategy for doing so is to relate
the half-plane capacity to Brownian motion! This is something we will see a lot
in this course.
Proposition. Let A ∈ Q and Bt be complex Brownian motion. Define the
stopping time
τ = inf{t ≥ 0 : Bt 6∈ H \ A}.
Then
(i) For all z ∈ H \ A, we have

im(z − gA (z)) = Ez [im(Bτ )]

(ii)
hcap(A) = lim y Ey [im(Bτ )].
y→∞

In particular, hcap(A) ≥ 0.

(iii) If A ⊆ D̄ ∩ H, then
Z π
2
hcap(A) = Eeiθ [im(Bτ )] sin θ dθ.
π 0

Proof.
(i) Let φ(z) = im(z − gA (z)). Since z − gA (z) is holomorphic, we know φ is
harmonic. Moreover, ϕ is continuous and bounded, as it → 0 at infinity.
These are exactly the conditions needed to solve the Dirichlet problem
using Brownian motion.
Since im(gA (z)) = 0 when z ∈ ∂(H \ A), we know im(Bτ ) = im(Bt −
gA (Bτ )). So the result follows.
(ii) We have

hcap(A) = lim z(gA (z) − z)

z→∞
= lim (iy)(gA (iy) − iy)
y→∞

= lim y im(iy − gA (iy))

y→∞

= lim y Eiy [im(Bτ )]

y→∞

where we use the fact that hcap(A) is real, so we can take the limit of the
real part instead.

(iii) See example sheet.

12
1 Conformal transformations III Schramm–Loewner Evolutions

In some sense, what we used above is that Brownian motion is conformally

invariant, where we used a conformal mapping to transfer between our knowledge
of Brownian motion on H to that on H \ D. Informally, this says the conformal
image of a Brownian motion is a Brownian motion.
We have in fact seen this before. We know that rotations preserve Brownian
motion, and so does scaling, except when we scale we need to scale our time
as well. In general, a conformal mapping is locally a rotation and scaling. So
we would indeed expect that the conformal image of a Brownian motion is a
Brownian motion, up to some time change. Since we are performing different
transformations all over the domain, the time change will be random, but that
is still not hard to formalize. For most purposes, the key point is that the image
of the path is unchanged.
Theorem. Let D, D̃ ⊆ C be domains, and f : D → D̃ a conformal transforma-
tion. Let B, B̃ be Brownian motions starting from z ∈ D, z̃ ∈ D̃ respectively,
with f (z) = z̃. Let

τ = inf{t ≥ 0 : Bt 6∈ D}
τ̃ = inf{t ≥ 0 : B̃t 6∈ D̃}

Set
Z τ
τ0 = |f 0 (Bs )|2 ds
0
 Z s 
0 2
σ(t) = inf s ≥ 0 : |f (Br )| dr = t
0
Bt0 = f (Bσ(t) ).

Then (Bt0 : t < τ̃ 0 ) has the same distribution as (B̃t : t < τ̃ ).

Proof. See Stochastic Calculus.
Example. We can use conformal invariance to deduce the first exit distribution
of a Brownian motion from different domains. First of all, observe that if we
take our domain to be D and start our Brownian motion from the origin, the
first exit distribution is uniform. Thus, applying a conformal transformation, we
see that the exit distribution starting from any point z ∈ D is

1 − |z|2
 
iθ 1
f (e ) = .
2π |eiθ − z|
Similarly, on H, starting from z = x + iy, the exit distribution is
1 y
f (u) = .
π (x − u)2 + y 2
Note that if x = 0, y = 1, then this is just
 
1 1
f (u) = .
π u2 + 1
This is the Cauchy distribution!

13
2 Loewner’s theorem III Schramm–Loewner Evolutions

2 Loewner’s theorem
2.1 Key estimates
Before we prove Loewner’s theorem, we establish some key identities and esti-
mates. As before, a useful thing to do is to translate everything in terms of
Brownian motion.
Proposition. Let A ∈ Q and B be a complex Brownian motion. Set

τ = inf{t ≥ 0 : Bt 6∈ H \ A}.

Then
– If x > Rad(A), then
 
1
gA (x) = lim πy − Piy [Bτ ∈ [x, ∞)] .
y→∞ 2

– If x < −Rad(A), then

 
1
gA (x) = lim πy Piy [Bτ ∈ (−∞, x]] − .
y→∞ 2

Proof. First consider the case A = ∅ and, by symmetry, x > 0. Then

 
1
lim πy − Piy [Bτ ∈ [x, ∞)]
y→∞ 2
= lim πy Piy [Bτ ∈ [0, x)]
y→∞
Z x
y
= lim πy 2 + y2 )
ds
y→∞ 0 π(s
= x,

where the first equality follows from the fact that Brownian motion exits through
the positive reals with probability 12 ; the second equality follows from the
previously computed exit distribution; and the last follows from dominated
convergence.
Now suppose A 6= ∅. We will use conformal invariance to reduce this to the
case above. We write gA = uA + ivA . We let

σ = inf{t > 0 : Bt 6∈ H}.

Then we know

Piy [Bτ ∈ [x, ∞)] = PgA (iy) [Bσ ∈ [gA (x), ∞)]
= PivA (iy) [Bσ ∈ [gA (x) − uA (iy), ∞)].

Since gA (z) − z → 0 as z → ∞, it follows that vAy(iy) → 1 and uA (iy) → 0 as

y → ∞. So we have

PivA (iy) Bσ ∈ [gA (x) − uA (iy), ∞) − Piy Bσ ∈ [gA (x), ∞) = o(y −1 )

   

as y → ∞. Combining with the case A = ∅, the the proposition follows.

14
2 Loewner’s theorem III Schramm–Loewner Evolutions

Corollary. If A ∈ Q, Rad(A) ≤ 1, then

1
x ≤ gA (x) ≤ x + if x > 1
x
1
x+ ≤ gA (x) ≤ x if x < −1.
x
Moreover, for all A ∈ Q, we have

|gA (z) − z| ≤ 3 Rad(A).

1
Proof. Exercise on the first example sheet. Note that z 7→ z + z sends H \ D̄ to
H.

This corollary gives us a “zeroth order” bound on gA (z) − z. We can obtain

a better first-order bound as follows:
Proposition. There is a constant c > 0 so that for every A ∈ Q and |z| >
2 Rad(A), we have
 
hcap(A) Rad(A) · hcap(A)
ga (z) − z + ≤c .
z |z|2

Proof. Performing a scaling if necessary, we can assume that Rad(A) = 1. Let

hcap(A)
h(z) = z + − gA (z).
z
We aim to control the imaginary part of this, and then use the Cauchy–Riemann
equations to control the real part as well. We let

im(z)
v(z) = im(h(z)) = im(z − gA (z)) = hcap(A).
|z|2

Let B be a complex Brownian motion, and let

σ = inf{t ≥ 0 : Bt 6∈ H \ D̄}
τ = inf{t ≥ 0 : Bt 6∈ H}.

Let p(z, eiθ ) be the density with respect to the Lebesgue measure at eiθ for Bσ .
Then by the strong Markov property at the time σ, we have
Z π
im(z − gA (z)) = Eeiθ [im(Bτ )]p(z, eiθ ) dθ.
0

In the first example sheet Q3, we show that

  
2 im(z) 1
p(z, eiθ ) = 2
sin θ 1 + O . (∗)
π |z| |z|

We also have Z π
2
hcap(A) = Eeiθ [im(Bτ )] sin θ dθ.
π 0

15
2 Loewner’s theorem III Schramm–Loewner Evolutions

So
im(z)
|v(z)| = im(z − gA (z)) − hcap(A)
|z|2
Z π
im(z) 2 π
Z
= Eeiθ [im(Bτ )]p(z, eiθ ) dθ − · E iθ [im(Bτ )] sin θ dθ .
0 |z|2 π 0 e
By applying (∗), we get
c hcap(A) im(z)
|v(z)| ≤ c .
|z|3
where c is a constant.
Recall that v is harmonic as it is the imaginary part of a holomorphic function.
By example sheet 1 Q9, we have
c hcap(A) c hcap(A)
|∂x v(z)| ≤ , |∂y v(z)| ≤ .
|z|3 |z|3
By the Cauchy–Riemann equations, re(h(z)) satsifies the same bounds. So we
know that
c hcap(A)
|h0 (z)| ≤ .
|z|3
Then Z ∞
h(iy) = h0 (is) ds,
y

since h(iy) → 0 as y → ∞. Taking absolute values, we have

Z ∞
|h(iy) = |h0 (is)| ds
y
Z ∞
≤ c hcap(A) s−3 ds
y
≤ c0 hcap(A)y −z
.

To get the desired bound for a general z, integrate h0 along the boundary of the
circle of radius |z| to get
c hcap(A)
|h(z)| = |h(reiθ )| ≤ + h(iz).
|z|2
The following is a very useful fact about Brownian motion:
Theorem (Beurling estimate). There exists a constant c > 0 so that the
following holds. Let B be a complex Brownian motion, and A ⊆ D̄ be connected,
0 ∈ A, and A ∩ ∂ D̄ 6= ∅. Then for z ∈ D, we have

Pz [B[0, τ ] ∩ A = ∅] ≤ c|z|1/2 ,

where τ = inf{t ≥ 0 : Bt 6∈ D}.

We will not prove this, since it is quite tricky to prove. The worst case
behaviour is obtained when A = [−i, 0].
To the theorem, we need one more estimate, which uses the Beurling estimate.

16
2 Loewner’s theorem III Schramm–Loewner Evolutions

Proposition. There exists a constant c > 0 so that the following is true:

Suppose A, à ∈ Q with A ⊆ à and à \ A is coonnected. Then
(
(dr)1/2 d ≤ r
diam(gA (Ã \ A)) ≤ c ,
Rad(Ã) d > r

where
d = diam(Ã \ A), r = sup{im(z) : z ∈ Ã}.
Proof. By scaling, we can assume that r = 1.
– If d ≥ 1, then the result follows since

|gA (z) − z| ≤ 3Rad(A),

and so

diam(gA (Ã \ A)) ≤ diam(A) + 6Rad(A) ≤ 8 diam(Ã).

– If d < 1, fix z ∈ H so that U = B(z, d) ⊇ Ã \ A. It then suffices to bound

the size of gA (U ) (or, to be precise, gA (U \ A)).
Let B be a complex Brownian motion starting from iy with y ≥ 2. Let

τ = inf{t ≥ 0 : Bt 6∈ H \ A}.

For B[0, τ ] to reach U , it must

(i) Reach B(z, 1) without leaving H \ A, which occurs with probability
at most c/y for some constant c, by example sheet.
(ii) It must then hit U before leaving H \ A. By the Beurling estimate,
this occurs with probability ≤ cd1/2 .
Combining the two, we see that

lim sup Piy [B[0, τ ] ∩ U 6= ∅] ≤ cd1/2 .

y→∞

By the conformal invariance of Brownian motion, if σ = inf{t ≥ 0 : Bt 6∈

H}, this implies

lim sup y Py [B[0, σ] ∩ gA (Ã \ A) 6= ∅] ≤ cd1/2 .

y→∞

Since gA (Ã \ A) is connected, by Q10 of example sheet 1, we have

diam(gA (Ã \ A)) ≤ cd1/2 .

We can finally get to the key content of Loewner’s theorem. An important

definition is the following:
Definition. Suppose (At )t≥0 is a family of compact H-hulls. We say that (At )
is
(i) non-decreasing if s ≤ t implies As ⊆ At .

17
2 Loewner’s theorem III Schramm–Loewner Evolutions

(ii) locally growing if for all T > 0 and ε > 0, there exists δ > 0 such that
whenever 0 ≤ s ≤ t ≤ s + δ ≤ T , we have
diam(gAs (At \ As )) ≤ ε.
This is a continuity condition.
(iii) parametrized by half-plane capacity if hcap(At ) = 2t for all t ≥ 0.
We write A be the set of families of compact H-hulls which satisfy (i) to (iii).
We write AT for the set of such families defined on [0, T ].
Example. If γ is is a simple curve in H starting from 0 and At = γ[0, t]. This
clearly satisfies (i), and the previous proposition tells us this satsifies (ii). On the
first example sheet, we show that we can reparametrize γ so that hcap(At ) = 2t
for all t ≥ 0. Upon doing so, we have At ∈ A.
Theorem. Suppose that (At ) ∈ A. Let gt = gAt . Then there exists a continuous
function U : [0, ∞) → R so that
2
∂t gt (z) = , g0 (z) = z.
gt (z) − Ut
This is known as the chordal Loewner ODE , and U is called the driving
function.
Proof.
T First note that since the hulls are locally growing, the intersection
g
s≥t (As ) consists of exactly one point. Call this point Ut . Again by the
t
locally growing property, Ut is in fact a continuous function in t.
Recall that if A ∈ Q, then
 
hcap(A) hcap(A)Rad(A)
gA (z) = z + +O .
z |z|2
If x ∈ R, then as gA+x (z) − x = gA (z − x), we have
 
hcap(z) hcap(A)Rad(A + x)
gA (z) = gA (z + x) − x = z + +O . (∗)
z+x |z + x|2
Fix ε > 0. For 0 ≤ s ≤ t, let
gs,t = gt ◦ gs−1 .
Note that
hcap(gT (At+ε \ At )) = 2ε.
Apply (∗) with A = gt (At+ε \ At ), x = −Ut , and use that Rad(A + x) =
Rad(A − Ut ) ≤ diam(A) to see that
 
2ε 1
gA (z) = gt,t+ε (z) = z + + 2ε diam(gt (At+ε \ At ))O .
z − Ut |z − Ut |2
So
gt+ε (z) − gt (z) = (gt,t+ε − gt,t ) ◦ gt (z)
 
2ε 1
= + 2ε diam(gt (At+ε \ At ))O .
gt (z) − Ut |gt (z) − Ut |2
Dividing both sides by ε and taking the limit as ε → 0, the desired result follows
since diam(gt (At+ε \ At )) → 0.

18
2 Loewner’s theorem III Schramm–Loewner Evolutions

We can ask about the reverse procedure — given a continuous function Ut ,

can we find a collection At ∈ A whose driving function is Ut ? We can simply let
gt be the solution to this differential equation. We can then let

At = H \ domain(gt ).

On the example sheet, we show that this is indeed a family in A, and gt are
indeed conformal transformations.

2.2 Schramm–Loewner evolution

We now take Ut to be a random process adapted to Ft = σ(Us : s ≤ t). We can
then define gt (z) as the solution to the Loewner equation for each z, and define
At accordingly. We then obtain a random family At in A.
Definition (Conformal Markov property). We say that (At ) satisfy the confor-
mal Markov property if
d
(i) Given Ft , (gt (At+s ) − Ut )s≥0 = (As )s≥0 .
d
(ii) Scale-invariance: (rAt/r2 )t≥0 = (At ).
Here (i) is the Markov property, and the second part is the conformal part,
since the only conformal maps H → H that fix 0 and ∞ are the rescalings.
Schramm was interested in such processes because they provide us with
potential scaling limits of random processes, such as self-avoiding walk or perco-
lation. If we start with a process that has the Markov property, then the scaling
limit, if exists, ought to be scale invariant, and thus have the conformal Markov
property. Schramm classified the list of all possible such scaling limits:
Theorem (Schramm). If (A √ t ) satisfy the conformal Markov property, then there
exists κ ≥ 0 so that Ut = κBt , where B is a standard Brownian motion.
Proof. The first property is exactly the same thing as saying that given Ft , we
have
d
(Ut+s − Ut )s≥0 = (Us ).
So Ut is a continuous process with stationary, independent increments. This
implies there exists κ ≥ 0 and a ∈ R such that
√
Ut = κBt + at,

where B is a standard Brownian motion. Then the second part says

d
(rUt/r2 )t≥0 = (Ut )t≥0 .

So U satisfies Brownian scaling. Plugging this in, we know

√ d √
r κBt/r2 + at/r = r κBt + at.
√ d √
Since r κBt/r2 = r κBt , we must have a = 0.
This finally gives us the definition of SLE.

19
2 Loewner’s theorem III Schramm–Loewner Evolutions

Definition (Schramm–Loewner√evolution). For κ > 0, SLEκ is the random

family of hulls encoded by Ut = κBt , where B is a standard Brownian motion.
√
When κ = 0, we have Ut = 0. This corresponds to the curve γ(t) = 2 ti.
Usually, when talking about SLEκ , we are instead thinking about a curve.
Theorem (Rhode–Schramm, 2005). If (At ) is an SLEκ with flow gt and driving
function Ut , then gt−1 : H → At extends to a map on H̄ for all t ≥ 0 almost surely.
Moreover, if we set γ(t) = gt−1 (Ut ), then H \ At is the unbounded component of
H \ γ([0, t]).
We will not prove this. From now on, SLEκ will often refer to this curve
instead.
In the remainder of the course, we will first discuss some properties of SLEκ ,
and then “prove” that the scaling limit of certain objects are SLEκ ’s. We will not
be providing anything near a full proof of the identification of the scaling limit.
Instead, we will look at certain random processes, and based on the properties of
the random process, we deduce that the scaling limit “ought to” have a certain
analogous property. We will then show that for a very particular choice of κ,
the curve SLEκ has that property. We can then deduce that the scaling limit is
SLEκ for this κ.
What we will do is to prove properly that the SLEκ does indeed have
the desired property. What we will not do is to prove that the scaling limit
exists, satisfies the conformal Markov property and actually satisfies the desired
properties (though the latter two should be clear once we can prove the scaling
limit exists).

20
3 Review of stochastic calculus III Schramm–Loewner Evolutions

3 Review of stochastic calculus

Before we start studying SLE, we do some review of stochastic calculus. In
stochastic calculus, the basic object is a continuous semi-martingale. We write
these as
Xt = Mt + At ,
where M is a continuous local martingale and A is continuous with bounded
variation.
The important concepts we will review are
(i) Stochastic integrals
(ii) Quadratic variation

(iii) Itô’s formula

(iv) Lévy characterization of Brownian motion
(v) Stochastic differential equations

The general setting is that we have a probability space (Ω, F, P) with a

continuous time filtration Ft , which satisfies the “usual conditions”, namely F0
contains all P-null sets, and Ft is right continuous, i.e.
\
Ft = Fs .
s>t

Stochastic integral
If Xt = Mt + At is a continuous semi-martingale, and Ht is a previsible process,
we set Z t Z t Z t
Hs dXs = Hs dMs + Hs dAs ,
0 0 0
where the first integral is the Itô integral, and the second is the Lebesgue–Stieltjes
integral. The first term is a continuous local martingale, and the second is a
continuous bounded variation process.
The Itô integral is defined in the same spirit as the Riemann integral. The
key thing that makes it work is that there is “extra cancellation” in the definition,
from the fact that we are integrating against a martingale, and so makes the
integral converge even though the process we are integrating against is not of
bounded variation.

Quadratic variation
If M is a continuous local martingale, then the quadratic variation is
d2n te−1
X
[M ]t = lim (M(k+1)2−n − Mk2−n )2 ,
n→∞
k=0

and is the unique continuous non-decreasing process of bounded variation such

that Mt2 − [M ]t is a continuous local martingale. Applying the same definition

21
3 Review of stochastic calculus III Schramm–Loewner Evolutions

to a bounded variation process always gives zero, so if X is a semi-martingale, it

makes sense to define
[X]t = [M + A]t = [M ]t .
Also, we have "Z #
(·) Z t
Hs dMs = Hs2 d[M ]s ,
0 0
t
where the integral on the right is the Lebesgue–Stieltjes integral.

Itô’s formula
Itô’s formula is the stochastic calculus’ analogue of the fundamental theorem of
calculus. It takes a bit of work to prove Itô’s formula, but it is easy to understand
what the intuition is. If f ∈ C 2 , then we have
n
X
f (t) = f (0) + (f (tk ) − f (tk−1 ))
k=1

for any partition 0 = t0 < · · · < tn = t. If we were to do a regular fundamental

theorem of calculus, then we can perform a Taylor approximation
n
X   Z t
f (t) = f (0) + f 0 (tk−1 ) (tk − tk−1 ) + o(tk − tk−1 ) → f (0) + f 0 (s) ds
k=1 0

as max |tk − tk−1 | → 0.

In Itô’s formula, we want to do the same thing but with Brownian motion.
Suppose B is a Brownian motion. Then
n 
X 
f (Bt ) = f (B0 ) + f (Btk ) − f (Btk−1 ) .
k=1

When we do a Taylor expansion of this, we have to be a bit more careful than

before, as we cannot ignore all the higher order terms. The point is that the
quadratic variation of Bt is non-zero. We write this as
n h
X 1 i
f (0)+ f 0 (Btk−1 )(Btk −Btk−1 )+ f 00 (Btk−1 )(Btk −Btk−1 )2 +o((Btk +Btk−1 )2 ) .
2
k=1

Taking the limit, and using that E[(Btk − Btk−1 )2 ] = tk − tk−1 , we get
Z t Z t
1
f (Bt ) = f (0) + f 0 (Bs ) dBs + f 00 (Bs ) ds.
0 2 0

More generally, if Xt = Mt + At is a continuous semi-martingale, and f ∈

C 1,2 (Rt × R), so that it is C 1 in the first variable and C 2 in the second, then
Z t Z t
f (t, Xt ) = f (0, X0 ) + ∂s f (s, Xs ) ds + ∂x f (s, Xs ) dAs
0 0
Z t
1 t 2
Z
+ ∂x f (s, Xs ) dMs + ∂ f (s, Xs ) d[M ]s .
0 2 0 x

22
3 Review of stochastic calculus III Schramm–Loewner Evolutions

Lévy characterization of Brownian motion

Theorem (Lévy characterization). Let Mt is a continuous local martingale with
[M ]t = t for t ≥ 0, then Mt is a standard Brownian motion.
Proof sketch. Use Itô’s formula with the exponential moment generating function
2
eiθMt +θ /2[M ]t .

Stochastic differential equations

Suppose (Ω, F, P) and (Ft ) are as before, and (Bt ) is a Brownian motion adapted
to (Ft ). We say a process Xt satisfies satisfies the stochastic differential equation

dXt = b(Xt ) dt + σ(Xt ) dBt

for functions b, σ iff

Z t Z t
Xt = b(Xs ) ds + σ(Xs ) dBs + X0
0 0

for all t ≥ 0. At the end of the stochastic calculus course, we will see that there
is a unique solution to this equation provided b, σ are Lipschitz functions.
This in particular implies we can solve
2
∂t gt (z) = , g0 (z) = z
gt (z) − Ut
√
for Ut = κBt , where Bt is a standard Brownian motion.

23
4 Phases of SLE III Schramm–Loewner Evolutions

4 Phases of SLE
We now study some basic properties of SLEκ . Our first goal will be to prove the
following theorem:

Theorem. SLEκ is a simple curve if κ ≤ 4, and is self-intersecting if κ > 4.

When proving this theorem, we will encounter the Bessel stochastic differential
equation, and so we shall begin by understanding this SDE and the associated
Bessel process.

Definition (Square Bessel process). Let X = (B 1 , . . . , B d ) be a d-dimensional

standard Brownian motion. Then

Zt = kXt k2 = (Bt1 )2 + (Bt2 )2 + · · · + (Btd )2

is a square Bessel process of dimension d.

The square Bessel process satisfies a stochastic differential equation, which

we can obtain from Itô’s formula. Itô’s formula tells us

dZt = 2Bt1 dBt1 + · · · + 2Btd dBtd + d · dt.

For reasons that will become clear soon, we define

Z t 1
Bs dBs1 + · · · + Bsd dBsd
Yt = 1/2
.
0 Zs
Then tautologically, we have
1/2
dZt = 2Zt dYt + d · dt.

Observe that Yt is a continuous local martingale, and the quadratic variation is

equal to Z t
(Bs1 )2 + · · · + (Bsd )2
[Y ]t = ds = t.
0 Zs
By the Lévy characterization, YT is a standard Brownian motion. So we can
write
Lemma.
1/2
dZt = 2Zt dB̃t + d · dt.
where B̃ is a standard Brownian motion.
This is called the square Bessel stochastic differential equation of dimension
d.
Ultimately, we are interested in the Bessel process, i.e. the square root of the
square Bessel process.
Definition (Bessel process). The Bessel process of dimension d, written BESd ,
is
1/2
Ut = Zt .

24
4 Phases of SLE III Schramm–Loewner Evolutions

Applying Itô’s formula, we have

1 −1/2 1 −3/2
dUt = Zt dZt − Zt d[Z]t
2 8
The square Bessel SDE lets us understand the first term, and we can calculate

[Z]t = 4Zt dt

Simplifying, this becomes

Lemma.  
d−1
dUt = Ut−1 dt + dB̃t .
2
This is the Bessel stochastic differential equation of dimension d.
Observe that we can make sense of this equation for any d ∈ R, not necessarily
integral. However, the equation only makes sense when Ut 6= 0, and Ut−1 fails
to be Lipschitz as Ut → 0. To avoid this problem, we say this is defined up to
when it hits 0.
Proposition. Let d ∈ R, and Ut a BESd .

(i) If d < 2, then Ut hits 0 almost surely.

(ii) If d ≥ 2, then Ut doesn’t hit 0 almost surely.
This agrees with what we expect from the recurrence of Brownian motion,
for integral d. In particular, for d = 2, a Brownian motion gets arbitrarily close
to 0 infinitely often, so we expect that if we are just a bit lower than d = 2, we
would hit 0 almost surely.
Proof. The proof is similar to the proof of recurrence of Brownian motion. For
a ∈ R>0 , we define
τa = inf{t ≥ 0 : Ut = a}.
We then consider P[τb < τa ], and take the limit a → 0 and b → ∞. To do so, we
claim
Claim. Ut2−d is a continuous local martingale.
To see this, we simply compute using Itô’s formula to get
1
dUt2−d = (2 − d)Ut1−d dUt + (2 − d)(1 − d)Ut−d d[U ]t
2
(2 − d)(d − 1) 1−d 1
= (2 − d)Ut1−d
dB̃t + Ut dt + (2 − d)(1 − d)Ut−d dt
2Ut 2
= (2 − d)Ut1−d dB̃t .

Therefore Ut is a continuous local martingale. Since Ut∧τa ∧τb is bounded, it is a

true martingale, and so optional stopping tells us

U02−d = E[Uτ2−d
a ∧τb
] = a2−d P[τa < τb ] + b2−d P[τb < τa ].

25
4 Phases of SLE III Schramm–Loewner Evolutions

– If d < 2, we set a = 0, and then

U02−d = b2−d P[τb < τ0 ].

Dividing both sides as b, we find that

 2−d
U0
= P[τb < τ0 ].
b

Taking the limit b → ∞, we see that Ut hits 0 almost surely.

– If d > 2, then we have
 2−d  2−d
U0 b
P[τa < τb ] = − P[τb < τa ] → 0
a a

as a → 0 for any b and U0 > 0. So we are done in this case.

– If d = 2, then our martingale is just constant, and this analysis is useless.

In this case, we consider log Ut and perform the same analysis to obtain
the desired conclusion.
Now take an SLEκ curve. For x ∈ R, consider the process

Vtx = gt (x) − Ut = gt (x) − gt (γ(t)).

Essentially by definition, x starts being in At when Vtx = 0 (formally, we need

to take limits, since x is not in H). We thus define

τx = inf{t ≥ 0 : Vtx = 0}.

This is then the time γ cuts x off from ∞. We thus want to understand P[τx < ∞].
We can calculate
2 √ 2 √
dVtx = d(gt (x) − Ut ) = dt − κ dBt = x dt − κ dBt .
gt (x) − Ut Vt

This looks almost like the Bessel SDE, but we need to do some rescaling to write
this as  x
V 2/κ
d √t = × √ dt + dB̃t , B̃t = −Bt .
κ Vt / κ
√
So we get that Vt× / κ ∼ BESd , with d = 1 + 4/κ. Thus, our previous result
implies (
1 κ>4
P[τx < ∞] = .
0 κ≤4

Theorem. SLEκ is a simple curve if κ ≤ 4, and is self-intersecting if κ > 4.

Proof. If κ ≤ 4, consider the probability that γ(t + n) hits γ([0, n]). This is
equivalently the probability that γ(t + n) hits ∂An . This is bounded above by
the probably that gn (γ(t + n)) − Un hits ∂H. But by the conformal Markov
property, gn (γ(t + n)) − Un is an SLEκ . So this probability is 0.

26
4 Phases of SLE III Schramm–Loewner Evolutions

If κ > 4, we want to reverse the above argument, but we need to be a bit

careful in taking limits. We have

lim P[γ(t) ∈ [−n, n] for some t] = 1.

n→∞

On the other hand, for any fixed n, we have

lim P[gm (Am ) − Um ⊇ [−n, n]] = 1.

m→∞

The probability that SLEκ self-intersects is

≥ P[gm (Am ) − Um ⊇ [−n, n] and gm (γ(t + m)) − Um ∈ [−n, n] for some t].

By the conformal Markov property, gm (γ(t + m)) − Um is another SLEκ given

Fm . So this factors as

P[gm (Am ) − Um ⊇ [−n, n]]P[γ(t) ∈ [−n, n] for some t].

Since this is true for all m and n, we can take the limit m → ∞, then the limit
n → ∞ to obtain the desired result.
It turns out there are two further regimes when κ > 4.
Theorem. If κ ≥ 8, then SLEκ is space-filling, but not if κ ∈ (4, 8).
This will be shown in the second example sheet. Here we prove a weaker
result:
Proposition. SLEκ fills ∂H iff κ ≥ 8.
To phrase this in terms of what we already have, we fix κ > 4. Then for any
0 < x < y, we want to understand the probability

g(x, y) = P[τx = τy ].

where we assume 0 < x < y. If τx = τy , this means the curve γ cuts x and y from
∞ ft the same time, and hence doesn’t hit [x, y]. Conversely, if P[τx = τy ] = 0,
then with probability one, γ hits (x, y), and this is true for all x and y. Thus,
we want to show that g(x, y) > 0 for κ ∈ (4, 8), and g(x, y) = 0 for κ ≥ 8.
To simplify notation, observe that g(x, y) = g(1, y/x) as SLEκ is scale-
invariant. So we may assume x = 1. Moreover, we know

g(1, r) → 0 as r → ∞,

since P[τ1 < t] → 1 as t → ∞, but P[τr < t] → 0 as r → ∞ for t fixed.

Definition (Equivalent events). We say two events A, B are equivalent if

P[A \ B] = P[B \ A] = 0,

Proposition. For r > 1, the event {τr = τ1 } is equivalent to the event

Vtr − Vt1
sup < ∞. (∗)
t<τ1 Vt1

27
4 Phases of SLE III Schramm–Loewner Evolutions

Proof. If (∗) happens, then we cannot have that τ1 < τr , or else the supremum
is infinite. So (∗) ⊆ {τ1 = τr }. The prove the proposition, we have to show that

Vtr − Vt1
 
P τ1 = τr , sup = ∞ = 0.
t<τ1 Vt1

For M > 0, we define

Vr −V1
 
σM = inf t ≥ 0 : t 1 t ≥ M .
Vt

It then suffices to show that

Vtr − Vt1
 
PM ≡ P τ1 = τr sup ≥ M = P[τ1 = τr | σM < τ1 ] → 0 as M → ∞.
t<τ1 Vt1

But at time σM , we have Vtr = (M + 1)Vt1 , and τ1 = τr if these are cut off at
the same time. Thus, the conformal Markov property implies

PM = g(1, M + 1).

So we are done.
So we need to show that
(
Vtr − Vt1

> 0 κ ∈ (4, 8)
P sup 1 <∞ .
t<τ1 Vt =0 k≥8

We will prove this using stochastic calculus techniques. Since ratios are hard, we
instead consider  r
Vt − Vt1

Zt = log .
Vt1
We can then compute

d − 1 Vtr − Vt1
    
3 1 1
dZt = −d + dt − 1 dBt , Z0 = log(r − 1).
2 (Vt1 )2 2 (Vt1 )2 Vtr Vt

Here we can already see why κ = 8 is special, since it corresponds to d = 32 .

When faced with such complex equations, it is usually wise to perform a
time change. We define σ(t) by the equation

dσ(t)
σ(0) = 0, dt = 1 )2 .
(Vσ(t)

So the map
Z σ(t)
1
t 7→ B̃t = − dBs
0 Vs1
is a continuous local martingale, with
" Z #
σ(t) Z σ(t)
1 1
[B̃]t = − 1
dBs = ds = t.
0 Vs 0 (Vs1 )2
t

28
4 Phases of SLE III Schramm–Loewner Evolutions

So by the Lévy characterization, we know that B̃t is a Brownian motion.

Now let
Z̃t = Zσ(t) .
Then we have
"  r 1
#
d − 1 Vσ(t) − Vσ(t)
 
3
dZ̃t = −d + r dt + dB̃t .
2 2 Vσ(t)

In integral form, we get

r 1
d − 1 t Vσ(s) − Vσ(s)
  Z
3
Z̃t = Z̃0 + B̃t + −d t+ r ds
2 2 0 Vσ(s)
 
3
≥ Z̃0 + B̃t + − d t.
2
4
Now if κ ≥ 8, then d = 1 + κ ≤ 23 . So we have

Z̃t ≥ Z̃0 + B̃t .

So we find that
sup Z̃t ≥ Z̃0 + sup B̃t = +∞.
t t

So we find that
Vtr − Vt1
sup = ∞.
t<τ1 Vt1
So g(x, y) = 0 for all 0 < x < y if κ ≥ 8.
Now if κ ∈ (4, 8), we pick ε > 0 and set
ε
r =1+ .
2
Then we have Z̃0 = log(r − 1) = log(ε/2). We will show that for small ε, we have
g(1, r) > 0. In fact, it is always positive, which is to be shown on the example
sheet.
We let
τ = inf{t > 0 : Z̃t = log ε}.
Then
r 1
d − 1 t∧τ Vσ(s) − Vσ(s)
  Z
3
Z̃t∧τ = Z̃0 + B̃t∧τ + −d t∧τ + r ds
2 2 0 Vσ(s)
d − 1 t∧τ Z̃s
  Z
3
≤ Z̃0 + B̃t∧τ + −d t∧τ + e ds
2 2 0
   
3 d−1
≤ Z̃0 + B̃t∧τ + −d t∧τ − (t ∧ τ )ε
2 2
    
3 d−1
= Z̃0 t + B̃t∧τ + −d + ε (t ∧ τ ).
2 2
We let  
3 d−1
Zt∗ = Z̃0 t + B̃t + −d+ ε t.
2 2

29
4 Phases of SLE III Schramm–Loewner Evolutions

We then have shown that

∗
Zt∧τ ≥ Z̃t∧τ .
∗
We assume that ε > 0 is such that 32 − d + d−1


2 ε < 0. Then Zt is a Brownian
ε
motion with a negative drift, starting from log 2 , so (by the second example
sheet), we have  
P sup Zt∗ < log ε > 0.
t≥0

So we know that  
P sup Z̃t < log ε > 0.
t≥0

So we have  ε
g 1, 1 + > 0.
2
This concludes the proof of the theorem.

SLE on other domains

So far, we only defined SLEκ in the upper half plane H from 0 to ∞. But we
can define SLE in really any simply connected domain. If D ⊆ C is a simply
connected domain, and x, y ∈ ∂D are distinct, then there exists a conformal
transformation ϕ : H → D with ϕ(0) = x and ϕ(∞) = y.
An SLEκ in D from x to y is defined by setting it to be γ = ϕ(γ̃), where γ̃
is an SLEκ in H from 0 to ∞. On the second example sheet, we check that this
definition is well-defined, i.e. it doesn’t depend on the choice of the conformal
transformation.

30
5 Scaling limit of critical percolation III Schramm–Loewner Evolutions

5 Scaling limit of critical percolation

The reason we care about SLE is that they come from scaling limits of certain
discrete models. In this chapter, we will “show” that SLE6 corresponds to the
scaling limit of critical percolation on the hexagonal lattice.
Let D ⊆ C be simply connected, and x, y ∈ ∂D distinct. Pick a hexagonal
lattice in D with hexagons of size ε such that x and y are lattice points. We
perform critical percolation on the hexagonal lattice, i.e. for each hexagon, we
colour it black or white with probability p = 12 . We enforce the condition that
the hexagons that intersect the clockwise arc of ∂D from x to y to all be black,
and those along the counterclockwise arc must be white.
y

Then there exists a unique interface γε that connects x to y with the property
that the black hexagons on its left and white on its right. It was conjectured
(and now proved by Smirnov) that the limit of the law of γε exists in distribution
and is conformally invariant.
This means that if D̃ is another simply connected domain, and x̃, ỹ ∈ ∂ D̃ are
distinct, then ϕ : D → D̃ is a conformal transformation with ϕ(x) = x̃, ϕ(y) = ỹ,
then ϕ(γ) is equal in distribution of the scaling limit of percolation in D̃ from x̃
to ỹ.
Also, percolation also satisfies a natural Markov property: if we condition on
γε up to a given time t, then the rest of γε is a percolation exploration in the
remaining domain. The reason for this is very simple — the path only determines
the colours of the hexagons right next to it, and the others are still randomly
distributed independently.
If we assume these two properties, then the limiting path γ satisfies the
conformal Markov property, and so must be an SLEκ . So the question is, what
is κ?
To figure out this κ, we observe that the scaling limit of percolation has a
locality property, and we will later see that SLE6 is the only SLE that satisfies
this locality property.
To explain the locality property, fix a simply-connected domain D in H (for
simplicity), and assume that 0 ∈ ∂D. Fixing a point y ∈ ∂D, we perform the
percolation exploration as before. Then the resulting path would look exactly
the same as if we performed percolation on H (with black boundary on R<0 and
white boundary on R>0 ), up to the point we hit ∂D \ ∂H. In other words, γ

31
5 Scaling limit of critical percolation III Schramm–Loewner Evolutions

doesn’t feel the boundary conditions until it hits the boundary. This is known
as locality.
It should then be true that the corresponding SLEκ should have an analogous
property. To be precise, we want to find a value of κ so that the following is
true: If γ is an SLEκ in H from 0 to ∞, run up until it first hits ∂D \ ∂H, then
ψ(γ) is a (stopped) SLEκ in H from 0 to ∞ where ψ : D → H is a conformal
transformation with ψ(0) = 0, ψ(y) = ∞. This is the locality property. We will
show that locality holds precisely for κ = 6.
Suppose that (At ) ∈ A has Loewner driving function Ut . We define

Ãt = ψ(At ).

Then Ãt is a family of compact H-hulls that are non-decreasing, locally growing,
and A0 = ∅. However, in general, this is not going to be parametrized by capacity.
On the second example sheet, we show that this has half plane capacity
Z t
ã(t) = hcap(Ãt ) = γ(ψt0 (Us ))2 ds,
0

which should be somewhat believable, given that hcap(rA) = r2 hcap(A).

Ãt has a “driving function” Ũt given by

Ũt = ψt (Ut ), ψt = g̃t ◦ ψ ◦ gt−1 , gt = gAt .

We then have
∂t ãt
∂t g̃t (z) = , g̃0 (z),
g̃t (z) − Ũt
To see this, simply recall that if At is γ([0, t]) for a curve γ, then Ut = gt (γ(t)).
To understand Ũt , it is convenient to know something about ψt :

Proposition. The maps (ψt ) satisfy

(ψt0 (Ut ))2 ψ 0 (Ut )

 
∂t ψt (z) = 2 − t .
ψt (z) − ψt (Ut ) z − Ut

In particular, at z = Ut , we have

∂t ψt (Ut ) = lim ∂t ψt (z) = −3ψt00 (Ut ).

z→Ut

Proof. These are essentially basic calculus computations.

To get Loewner’s equation in the right shape, we perform a time change

 Z u 
0 2
σ(t) = inf u ≥ 0 : (ψs (Us )) ds = t .
0

Then we have
2
∂t g̃σ(t) (z) = , g̃0 (z) = z.
g̃σ(t) − Ũσ(t)

32
5 Scaling limit of critical percolation III Schramm–Loewner Evolutions

It then remains to try to understand √ dŨσ(t) . Note that so far what we said
works for general Ut . If we put Ut = κBt , where B is a standard Brownian
motion, then Itô’s formula tells us before the time change, we have

dŨt = dψt (Ut )

 κ  √
= ∂t ψt (Ut ) + ψt00 (Ut ) dt + κψt0 (Ut ) dBt
2
κ − 6 00 √
= ψt (Ut ) dt + κψt0 (Ut ) dBt .
2
After the time change, we get
00
κ − 6 ψσ(t) (Uσ(t) ) √
dŨσ(t) = 0 2
dt + κ dB̃t ,
2 ψσ(t) (Uσ(t) )

where Z σ(t)
B̃t = ψs0 (Us ) dBs
0
is a standard Brownian motion by the Lévy characterization and the definition
of σ(t).
The point is that when κ = 6, the drift term goes away, and so
√
Ũσ(t) = 6B̃t .

So (Ãσ(t) ) is an SLE6 . Thus, we have proved that

Theorem. If γ is an SLEκ , then ψ(γ) is an SLEκ up until hitting ψ(∂D \ ∂H)
if and only if κ = 6.
So κ = 6 is the only possible SLEκ which could be the limit of percolation.

33
6 Scaling limit of self-avoiding walks III Schramm–Loewner Evolutions

6 Scaling limit of self-avoiding walks

We next think about self-avoiding walk (SAW).
Definition (Self-avoiding walk). Let G = (V, E) be a graph with uniformly
bounded degree, and pick x ∈ V and n ∈ N. The self-avoiding in G starting
from x of length n is the uniform measure on simple paths in G starting from x
of length n.
Self-avoiding walk tends to be difficult to understand, since they usually
don’t have the Markov property. However, in the scaling limit, it tends to be
more well-behaved. Restricting to the case of G = Zd , it has been shown that
for d ≥ 5, the scaling limit is just Brownian motion (Hara and Slade).
In d = 4, the same is conjectured to be true. In d = 3, there is no conjecture
for the continuous object which should describe its scaling limit. In this course,
we are only interested in d = 2. In this case, the self-avoiding walk is conjectured
to converge to SLE8/3 (Lawler–Schramm–Werner).
In percolation, the property that allowed us to conclude that the scaling limit
was SLE6 is the locality property. In self-avoiding walks, the special property
is restriction. This is a very simple concept — if G0 = (V 0 , E 0 ) is a subgraph
of G with x ∈ V 0 , then the self-avoiding walk on G conditioned to stay in G0
is a self-avoiding walk on G0 . Indeed, uniform measures always restricts to the
uniform measure on a smaller subset. It turns out SLE8/3 is the only SLE which
satisfies a continuum version of restriction. We will only show one direction,
namely that SLE8/3 satisfies the restriction property, but the other direction is
also known.

Brownian excursion
When studying the restriction property, we will encounter another type of process,
known as Brownian excursion. Fix a simply connected domain D ⊆ C and
points x, y ∈ ∂D distinct. Then roughly speaking, a Brownian excursion is a
Brownian motion starting at x conditioned to leave D at ∂D. Of course, we
cannot interpret these words literally, since we want to condition on a zero
probability event.
To make this rigorous, we first use conformal invariance to say we only have
to define this for H with x = 0, y = ∞.
To construct it in H, we start with a complex Brownian motion B = B 1 +iB 2 ,
with B01 = 0 and B02 = ε > 0. We then condition B on the event that B 2 hits
R  0 before hitting 0. This is a positive probability event. We then take limits
R → ∞ and ε → 0. On the second example sheet, we will show that this makes
sense, and the result is called Brownian excursion.
It turns out the limiting object is pretty simple. It is given by
B̂ = (B̂ 1 , B̂ 2 )
where B̂ 1 is a standard Brownian motion and B̂ 2 ∼ BES3 , and the two are
independent.
The key property about Brownian excursion we will need is the following:
Proposition. Suppose A be a compact H-hull and gA is as usual. If x ∈ R \ A,
then
0
Px [B̂[0, ∞) ∩ A = ∅] = gA (x).

34
6 Scaling limit of self-avoiding walks III Schramm–Loewner Evolutions

Proof. This is a straightforward computation. Take z = x + iε with ε > 0, and

let Bt be a Brownian motion. Define

σR = inf{t ≥ 0 : im(Bt ) = R}.

Then the desired probability is

lim lim Pz [B[0, σR ] ∩ A = ∅ | B[0, σR ] ∩ R = ∅].

ε→0 R→∞

By Bayes’ theorem, this is equal to

Pz [B[0, σR ] ∩ (A ∪ R) = ∅]
lim lim .
ε→0 R→∞ P[B[0, σR ] ∩ R = ∅]

We understand the numerator and denominator separately. The gambler’s ruin

estimate says the denominator is just ε/R, and to bound the numerator, recall
that for z ∈ H \ A, we have

|gA (z) − z| ≤ 3Rad(A).

Thus, using conformal invariance, we can bound

PgA (z) [B[0, σR+3Rad(A) ] ∩ R = ∅] ≤ Pz [B[0, σR ] ∩ (A ∪ R) = ∅]

≤ PgA (z) [B[0, σR−3Rad(A) ] ∩ R = ∅].

So we get
im(gA (z)) im(gA (z))
≤ numerator ≤ .
R + 3Rad(A) R − 3Rad(Z)
Combining, we find that the desired probability is

im(gA (x + iε)) 0
lim = gA (x).
ε→0 ε

The restriction property

We now return to understanding the restriction property. We assume that κ ≤ 4,
since this is the range of κ-values so that SLEκ is simple.
Recall that Q is the set of all compact H-hulls. We define

Q+ = {A ∈ Q : Ā ∩ (−∞, 0] = ∅}
Q− = {A ∈ Q : Ā ∩ [0, ∞) = ∅}

For A ∈ Q± = Q+ ∪ Q− , we define ψA : H → A → H by

ψA (z) = gA (z) − gA (0).

This is the version of gA that fixes 0, which is what we need. This is the unique
conformal transformation with
ψA (z)
ψA (0) = 0, lim = 1.
z→∞ z
We will need the following fact about SLE:

35
6 Scaling limit of self-avoiding walks III Schramm–Loewner Evolutions

Fact. SLEκ is transient, i.e. if γ is an SLEκ in H from 0 to ∞, then

lim γ(t) = ∞ almost surely.

t→∞

This is not very difficult to prove, but the proof is uninteresting and we have
limited time. Since SLEκ is simple for κ ≤ 4 and is also transient, it follows that
for all A ∈ Q± ,
0 < P[γ[0, ∞) ∩ A = ∅] < 1.
This is useful because we want to condition on this event. Write

VA = {γ[0, ∞) ∩ A = ∅}.

Definition (Restriction property). We say an SLEκ satisfies the restriction

property if whenever γ is an SLEκ , for any A ∈ Q± , the law of ψA (γ) conditional
on VA is that of an SLEκ curve (for the same κ).
Observe that the law of γ is determined by the probabilities A 7→ P[VA ] for
all A ∈ Q± .
Lemma. Suppose there exists α > 0 so that
0
P[VA ] = (ψA (0))α

for all A ∈ Q± , then SLEκ satisfies restriction.

Proof. Suppose the assertion in the lemma is true. Suppose that A, B ∈ Q± .
Then we have that
−1
P[γ[0, ∞) ∩ (ψA (B) ∪ A) = ∅]
P[ψA (γ[0, ∞)) ∩ B = ∅ | VA ] =
P[γ[0, ∞) ∩ A = ∅]
0
(ψ(ψ −1
(B)∪A)
(0))α
A
= 0 (0))α
(ψA
0 0
(ψB (0))α (ψA (0))α
= 0 (0))α
(ψA
0
= (ψB (0))α
= P[VB ],

where we used that

ψψ−1 (B)∪A = ψB ◦ ψA .
A

So the law of ψA (γ) given VA is the law of γ.

We now have to show that √ SLE8/3 satisfies the condition in the lemma. Let
Ft be the filtration of Ut = κBt . Then

M̃t = P[VA | Ft ]

is a bounded martingale with M̃0 = P[VA ]. Also,

M̃t → 1VA

36
6 Scaling limit of self-avoiding walks III Schramm–Loewner Evolutions

by the martingale convergence theorem. Also, if we define the stopping time

τ = inf{t ≥ 0 : γ(t) ∈ A},

then we get

M̃T = P[VA | Ft ] = P[VA | Ft ]1{t<τ } = P[Vgt (A)−gt (0) ]1{t<τ }

by the conformal Markov property.

Observe that if Mt is another bounded Ft -martingale with the property that
Mt → 1VA as t → ∞, then Mt = M̃t for all t ≥ 0, since

Mt = E[1VA | Ft ] = M̃t .

Given what we were aiming for, we consider

Mt = (ψg0 t (A)−gt (0) (0))α 1{t<τ }

Lemma. Mt∧τ is a continuous martingale if

8 5
κ= , α= .
3 8
These numbers are just what comes out when we do the computations.
0
Proof. Recall that we showed that gA is a probability involving Brownian ex-
0
cursion, and in particular is bounded in [0, 1]. So the same is true for ψA , and
hence Mt∧τ . So it suffices to show that Mt∧τ is a continuous local martingale.
Observe that
Mt∧τ = (ψg0 t∧τ (A)−gt∧τ (0) (0))α
So if we define
Nt = (ψg0 t (A)−gt (0) (0))α ,
then it suffices to show that Nt is a continuous local martingale by optional
stopping. We write
ψt = g̃t ◦ ψA ◦ gt−1 ,
where g̃t = gψA (γ(0,t]) . We then have

Nt = (ψt0 (Ut ))α .

In the example sheet, we show that

ψt00 (Ut )2 4
∂t ψt0 (Ut ) = − ψt000 (Ut ).
2ψt0 (Ut ) 3
By Itô’s formula, we get

(α − 1)κ + 1 ψt00 (Ut )2 ψt000 (Ut )

   
κ 4
dNt = αNt + − dt
2 ψt0 (Ut )2 2 3 ψt0 (Ut )
ψt00 (Ut ) √
+ αNt · κ dBt .
ψt0 (Ut )

Picking κ = 83 ensures the second dt term vanishes, and then setting α = 5

8 kills
the first dt term as well, and we are done.

37
6 Scaling limit of self-avoiding walks III Schramm–Loewner Evolutions

We next want to establish that Mt → 1VA as t → ∞.

Recall that Mt is the probability that a Brownian excursion on H \ γ[0, t]
from γ(t) to infinity does not hit A. So we would expect this to be true, since
on VA , as t increases, the tip of the SLE gets further and further away from
A, and so it is difficult for a Brownian excursion to hit A; conversely on VAC , it
eventually gets to A, and then we are dead.
By scaling, we can assume that

sup{im(ω) : ω ∈ A} = 1.

It is convenient to define the stopping times

σ(r) = inf{t ≥ 0 : im(γ(t)) = r}.

Note that σr < ∞ almost surely for all r > 0 since SLE8/3 is transient.
Lemma. Mt∧τ → 1 on VA as t → ∞.
Proof. Let B̂ be a Brownian excursion in H \ γ[0, σr ] from γ(σr ) to ∞. Let B
be a complex Brownian motion, and

τR = inf{t ≥ 0 : im(Bt ) = R}, z = γ(σr ) + iε.

Then the probability that B̂ hits A is

Pz [B[0, τR ] ⊆ H \ γ[0, σr ], B[0, τR ] ∩ A 6= ∅]

1 − ψσ0 r (Uσr ) = lim lim , (∗)
ε→0 R→∞ Pz [B[0, τR ] ⊆ H \ γ[0, σr ]]

We will show that this expression is ≤ Cr−1/2 for some constant C > 0. Then we
know that Mσr ∧τ → 1 as r → ∞ on VA . This is convergence along a subsequence,
but since we already know that Mt∧τ converges this is enough.
We first tackle the denominator, which we want to bound from below. The
idea is to bound the probability that the Brownian motion reaches the lime
im(z) = r + 1 without hitting R ∪ γ[0, σr ]. Afterwards, the gambler’s ruin
estimate tells us the probability of reaching im(z) = R without going below the
1
im(z) = r line is R−r .
In fact, we shall consider the box S = [−1, 1]2 +γ(σr ) of side length 2 centered
at γ(σr ). Let η be the first time B leaves S, and we want this to leave via the
top edge `. By symmetry, if we started right at γ(σr ), then the probability of
leaving at ` is exactly 14 . Thus, if we are at z = γ(σr ) + iε, then the probability
of leaving via ` is > 41 .
What we would want to show is that
1
Pz [B(η) ∈ ` | B[0, η] ∩ γ[0, σr ] = ∅] > . (†)
4
We then have the estimate
1 1
denominator ≥ · Pz [B[0, η] ∩ γ[0, σr ] = ∅] · .
4 R−r
Intuitively, (†) must be true, because γ[0, σr ] lies below im(z) = r, and so if
B[0, η] doesn’t hit γ[0, σr ], then it is more likely to go upwards. To make this

38
6 Scaling limit of self-avoiding walks III Schramm–Loewner Evolutions

rigorous, we write
1
< Pz [B(η) ∈ `]
4
= Pz [B(η) ∈ ` | B[0, η] ∩ γ[0, σr ] = ∅] P[B[0, η] ∩ γ[0, σr ] = ∅]
+ Pz [B(η) ∈ ` | B[0, η] ∩ γ[0, σr ] 6= ∅] P[B[0, η] ∩ γ[0, σr ] 6= ∅]
To prove (†), it suffices to observe that the first factor of the second term is is
≤ 14 , which follows from the strong Markov property, since Pw [B(η) ∈ `] ≤ 41
whenever im(w) ≤ r, which in particular is the case when w ∈ γ[0, σr ].
To bound the numerator, we use the strong Markov property and the Beurling
estimate to get
Pz [B hits A without hitting R ∪ γ[0, σr ]] ≤ Pz [B[0, η] ∩ γ[0, σr ]] · Cr−1/2 .
Combining, we know the numerator in (∗) is
1
≤ C · r−1/2 · P[B[0, η] ∩ γ[0, σr ] = ∅].
R
These together give the result.
Lemma. Mt∧τ → 0 as t → ∞ on VAc .
This has a “shorter” proof, because we outsource a lot of the work to the
second example sheet.
Proof. By the example sheet, we may assume that A is bounded by a smooth,
simple curve β : (0, 1) → H.
Note that γ(τ ) = β(s) for some s ∈ (0, 1). We need to show that
lim ψt0 (Ut ) = 0.
t→τ

For m ∈ N, let  
1
tm = inf t ≥ 0 : |γ(t) − β(s)| =
m
Since β is smooth, there exists δ > 0 so that
` = [β(s), β(s) + δn] ⊆ A,
where n is the unit inward pointing normal at β(s). Let
Lt = gt (`) − Ut .
Note that a Brownian motion starting from a point on ` has a uniformly positive
chance of exiting H \ γ[0, tm ] on the left side of γ[0, tm ] and on the right side as
well.
On the second example sheet, we see that this implies that
Ltm ⊆ {w : im(w) ≥ a| Re(w)|}
for some a > 0, using the conformal invariance of Brownian motion. Intuitively,
this is because after applying gt − Ut , we have uniformly positive probability of
exiting via the positive or real axis, and so we cannot be too far away in one
directionn.
Again by the second example sheet, the Brownian excursion in H from 0 to
∞ hits Ltm with probability → 1 as m → ∞.

39
6 Scaling limit of self-avoiding walks III Schramm–Loewner Evolutions

We thus conclude

Theorem. SLE8/3 satisfies the restriction property. Moreover, if γ ∼ SLE8/3 ,

then
0
P[γ[0, ∞) ∩ A = ∅] = (ψA (0))5/8 .
There is a rather surprising consequence of this computation. Take γ1 , . . . , γ8
to be independent SLE8/3 ’s. Then we have
0
P[γj [0, ∞) ∩ A = ∅ for all j] = (ψA (0))5 .

Note that this is the same as the probability that the hull of γ1 , . . . , γ8 des not
intersect A, where Sthe hull is the union of the γj ’s together with the bounded
components of H \ j γj .
In the same manner, if B̂1 , . . . , B̂5 are independent Brownian excursions,
then
0
P[B̂j [0, ∞) ∩ A = ∅ for all j] = (ψA (0))5 .
Thus, the hull of γ1 , . . . , γ8 has the same distribution as the hull of B̂1 , . . . , B̂5 .
Moreover, if we take a boundary point of the hull of, say, γ1 , . . . , γ8 , then
we would expect it to belong to just one of the SLE8/3 ’s. So the boundary of
the hull of γ1 , . . . , γ8 looks locally like an SLE8/3 , and the same can be said for
the hull fo B̂1 , . . . , B̂5 . Thus, we conclude that the boundary of a Brownian
excursion looks “locally” like SLE8/3 .

40
7 The Gaussian free field III Schramm–Loewner Evolutions

7 The Gaussian free field

We end by discussing the Gaussian free field, which we can think of as a two-
dimensional analogue of Brownian motion, i.e. a random surface. We will show
that the level curves of the Gaussian free field are SLE4 s.
To define the Gaussian free field, we have to do some analysis.
Notation.
– C ∞ is the space of infinitely differentiable functions on C
– C0∞ is the space of functions in C ∞ with compact support.
– If D is a domain, C0∞ (D) is the functions in C0∞ supported in D.
Definition (Dirichlet inner product). Let f, g ∈ C0∞ . The Dirichlet inner
product of f, g is Z
1
(f, g)∇ = ∇f (x) · ∇g(x) dx.
2π
This defines an inner product function on C0∞ . If D ⊆ C is a non-trivial
simply-connected domain, i.e. not ∅ or C, we can define
Definition (H01 (D)). We write H01 (D) for the Hilbert space completion of
C0∞ (D) with respect to ( · , · )∇ .
Elements of H01 (D) can be thought of as functions well-defined up to a null
set. These functions need not be continuous (i.e. need not have a continuous
representatives), and in particular need not be genuinely differentiable, but they
have “weak derivatives”.
We will need the following key properties of H01 (D):
Proposition.
(i) Conformal invariance: Suppose ϕ : D → D̃ is a conformal transformation,
and f, g ∈ C0∞ (D). Then

(f, g)∇ = (f ◦ ϕ−1 , g ◦ ϕ−1 )∇

In other words, the Dirichlet inner product is conformally invariant.

In other words, ϕ∗ : H01 (D) → H01 (D̃) given by f 7→ f ◦ ϕ−1 is an
isomorphism of Hilbert spaces.
(ii) Inclusion: Suppose U ⊆ D is open. If f ∈ C0∞ (U ), then f ∈ C0∞ (D).
Therefore the inclusion map i : H01 (U ) → H01 (D) is well-defined and
associates H01 (U ) with a subspace of H01 (D). We write the image as
Hsupp (U ).
(iii) Orthogonal decomposition: If U ⊆ D, let

Hharm (U ) = {f ∈ H01 (D) : f is harmonic on U }.

Then
H01 (D) = Hsupp (U ) ⊕ Hharm (U )
is an orthogonal decomposition of H01 (D). This is going to translate to a
Markov property of the Gaussian free field.

41
7 The Gaussian free field III Schramm–Loewner Evolutions

Proof. Conformal invariance is a routine calculation, and inclusion does not

require proof. To prove orthogonality, suppose f ∈ Hsupp (U ) and g ∈ Hharm (U ).
Then
Z Z
1 1
(f, g) = ∇f (x) · ∇g(x) dx = − f (x)∆g(x) dx = 0.
2π 2π
since f is supported on U and ∆g is supported outside of U .
To prove that they span, suppose f ∈ H01 (D), and f0 the orthogonal projec-
tion of f onto Hsupp (U ). Let g0 = f − f0 . We want to show that g0 is harmonic.
It would be a straightforward manipulation if we can take ∆, but there is no
guarantee that f0 is smooth.
We shall show that g0 is weakly harmonic, and then it is a standard analysis
result (which is also on the example sheet) that g0 is in fact genuinely harmonic.
Suppose ϕ ∈ C0∞ (U ). Then since g0 ⊥ Hsupp (U ), we have
Z Z
1 1
0 = (g0 , ϕ) = ∇g0 (x) · ∇ϕ(x) dx = − g0 (x)∆ϕ(x) dx.
2π 2π
This implies g0 is C ∞ on U and harmonic.
We will define the Gaussian free field to be a Gaussian taking values in H01 .
To make this precise, we first understand how Gaussian random variables on Rn
work.
Observe that if α1 , . . . , αn are iid N (0, 1) variables, and e1 , . . . , en is a stan-
dard basis of Rn , then
h = α1 e1 + · · · + αn en
is a standard
P n-dimensional Gaussian random variable.
If x = xj ej ∈ Rn , then
n
X
(h, x) = αj xj ∼ N (0, kxk).
j=1

Moreover, if x, y ∈ Rn , then (h, x) and (h, y) are jointly Gaussian with covariance
(x, y).
Thus, we can associate with h a family of Gaussian random variables (h, x)
indexed by x ∈ Rn with mean zero and covariance given by the inner product
on Rn . This is an example of a Gaussian Hilbert space.
We now just do the same for the infinite-dimensional vector space H01 (D).
One can show that this is separable, and so we can pick an orthonormal basis
(fn ). Then the Gaussian free field h on D is defined by
∞
X
h= αj fj ,
j=1

where the αj ’s are iid N (0, 1) random variables. Thus, if f ∈ H01 (D), then

(h, f )∇ ∼ N (0, kf k2∇ ).

More generally, if f, g ∈ H01 (D), then (h, f )∇ and (h, g)∇ are jointly Gaussian
with covariance (f, g)∇ . Thus, the Gaussian free field is a family of Gaussian
variables (H, f )∇ indexed by f ∈ H01 (D) with mean zero and covariance ( · , · )∇ .

42
7 The Gaussian free field III Schramm–Loewner Evolutions

P∞
We can’t actually quite make this definition, because the sum h = j=1 αj fj
does not converge. So h is not really a function, but a distribution. However,
this difference usually does not matter.
We can translate the properties of H01 into analogous properties of the
Gaussian free field.
Proposition.
(i) If ϕ : D → D̃ is a conformal transformation and h is a Gaussian free field
on D, then h ◦ ϕ−1 is a Gaussian free field on D̃.
(ii) Markov property: If U ⊆ D is open, then we can write h = h1 + h2 with
h1 and h2 independent where h1 is a Gaussian free field on U1 and h2 is
harmonic on U .
Proof.
(i) Clear.
(ii) Take h1 to be the projection onto Hsupp (U ). This works since we can take
the orthonormal basis (fn ) to be the union of an orthonormal basis of
Hsupp (U ) plus an orthonormal basis of Hharm (U ).
Often, we would rather think about the L2 inner product of h with something
else. Observe that integration by parts tells us
−1
(h, f )∇ =
(h, ∆f )L2 .
2π
Thus, we would be happy if we can invert ∆, which we can by IB methods.
Recall that
∆(− log |x − y|) = −2πδ(y − x),
where ∆ acts on x, and so − log |x − y| is a Green’s function for ∆. Given a
domain D, we wish to obtain a version of the Green’s function that vanishes on
the boundary. To do so, we solve ∆G̃x = 0 on D with the boundary conditions
G̃x (y) = − log |x − y| if y ∈ ∂D.
We can then set
G(x, y) = − log |x − y| − G̃x (y).
With this definition, we can define
Z
−1 1
∆ ϕ(x) = − G(x, y)ϕ(y) dy.
2π
Then ∆∆−1 ϕ(x) = ϕ(x), and so
(h, ϕ) ≡ (h, ϕ)L2 = −2π(h, ∆−1 ϕ)∇ .
Then (h, ϕ) is a mean-zero Gaussian with variance
(2π)2 k∆−1 ϕk2∇ = (2π)2 (∆−1 ϕ, ∆−1 ϕ)∇
= −2π(∆−1 ϕ, ∆∆−1 ϕ)
= (−2π∆−1 ϕ, ϕ)
ZZ
= ϕ(x)G(x, y)ϕ(y) dx dy.

43
7 The Gaussian free field III Schramm–Loewner Evolutions

More generally, if ϕ, ψ ∈ C0∞ (D), then

ZZ
cov((h, ϕ), (h, ψ)) = ϕ(x)G(x, y)ψ(y) dx dy.

On the upper half plane, we have a very explicit Green’s function

G(x, y) = GH (x, y) = − log |x − y| + log |x − ȳ|.

It is not hard to show that the Green’s function is in fact conformally invariant:
Proposition. Let D, D̃ be domains in C and ϕ is a conformal transformation
D → D̃. Then GD (x, y) = GD̃ (ϕ(x), ϕ(y)).
One way to prove this is to use the conformal invariance of Brownian motion,
but a direct computation suffices as well.
Schramm and Sheffield showed that the level sets of h, i.e. {x : h(x) = 0}
are SLE4 ’s. It takes some work to make this precise, since h is not a genuine
function, but the formal statement is as follows:
Theorem (Schramm–Sheffield). Let λ = π2 . Let γ ∼ SLE √4 in H from 0 to ∞.
Let gt its Loewner evolution with driving function Ut = κBt = 2Bt , and set
ft = gt − Ut . Fix W ⊆ H open and let

τ = inf{t ≥ 0 : γ(t) ∈ W }.

Let h be a Gaussian free field on H, λ > 0, and 𝒽 be the unique harmonic

function on H with boundary values λ on R>0 and −λ on R<0 . Explicitly, it is
given by
2λ
𝒽=λ− arg( · ).
π
Then
d
h + 𝒽 = (h + 𝒽) ◦ ft∧τ ,
where both sides are restricted to W .
A few words should be said about why we should think of this as saying the
level curves of h are SLE4 ’s. First, observe that since 𝒽 is harmonic, adding 𝒽
to h doesn’t change how h pairs with other functions in H01 (H). All it does is to
change the boundary conditions of h. So we should think of h + 𝒽 as a version
of the Gaussian free field with boundary conditions

(h + 𝒽)(x) = sgn(x)λ.

It is comforting to know the fact that the boundary conditions of h is well-defined

as an element of L2 .
Similarly, (h + 𝒽) ◦ ft∧τ is a Gaussian free field with the boundary condition
that it takes λ to the right of γ ∼ SLE4 and −λ to the left of it. Thus, we can
think of this as taking value 0 along γ. The theorem says this has the same
distribution just h + 𝒽. So we interpret this as saying the Gaussian free field has
the same distribution as a Gaussian free field forced to vanish along an SLE4 .
What we stated here is a simplified version of the theorem by Schramm and
Sheffield, because we are not going to deal with what happens Γ hits W .
The proof is not difficult. The difficult part is to realize that this is the thing
we want to prove.

44
7 The Gaussian free field III Schramm–Loewner Evolutions

Proof. We want to show that if ϕ ∈ C0∞ (W ),

d
((h + 𝒽) ◦ ft∧τ , ϕ) = (h + 𝒽, ϕ).

In other words, writing

ZZ
mt (ϕ) = (𝒽 ◦ ft , ϕ), σ02 (ϕ) = ϕ(x)GH (ft (x), ft (y))ϕ(y) dx dy,

we want to show that

((h + 𝒽) ◦ ft∧τ , ϕ) ∼ N (m0 (ϕ), σ02 (ϕ)).

This is the same as proving that

θ2
h i  
E eiθ((h+𝒽)◦ft∧τ ,ϕ) = exp iθm0 (ϕ) − σ02 (ϕ) .
2

Let Ft = σ(Us : s ≤ t) be the filtration of Ut . Then

h i h i
E eiθ((h+𝒽)◦ft∧τ ,ϕ) Ft∧τ = E eiθ(h◦ft∧τ ,ϕ) | Ft∧τ eiθmt∧τ (ϕ)
θ2 2
 
= exp iθmt∧τ (ϕ) − σt∧τ (ϕ) ,
2

If we knew that
θ2 2
 
exp iθmt (ϕ) − σt (ϕ)
2
is a martingale, then taking the expectation of the above equation yields the
desired results.
Note that this looks exactly like the form of an exponential martingale, which
in particular is a martingale. So it suffices to show that mt (ϕ) is a martingale
with
[m· (ϕ)]t = σ02 (ϕ) − σt2 (ϕ).
To check that mt (ϕ) is a martingale, we expand it as

2λ 2λ
𝒽 ◦ ft (z) = λ − arg(ft (z)) = λ − im(log(gt (z) − Ut )).
π π
So it suffices to check that log(gt (z) − Ut ) is a martingale. We apply Itô’s formula
to get

1 2 1 κ/2
d log(gt (z)−Ut ) = · dt− dUt − dt,
gt (z) − Ut gt (z) − Ut gt (z) − Ut (gt (z) − Ut )2

and so this is a continuous local martingale when κ = 4. Since mt (ϕ) is bounded,

it is a genuine martingale.
We then compute the derivative of the quadratic variation
Z    
2 2
d[m· (ϕ)]t = ϕ(x) im im ϕ(y) dx dy dt.
gt (x) − Ut gt (y) − Ut

45
7 The Gaussian free field III Schramm–Loewner Evolutions

To finish the proof, we need to show that dσt2 (ϕ) takes the same form. Recall
that the Green’s function can be written as

GH (x, y) = − log |x − y| + log |x − ȳ| = − Re(log(x − y) − log(x − ȳ)).

Since we have
log(ft (x) − ft (y)) = log(gt (x) − gt (y)),
we can compute
 
1 2 2
d log(gt (x) − gt (y)) = − dt
gt (x) − gt (y) gt (x) − Ut gt (y) − Ut
−2
= dt.
(gt (x) − Ut )(gt (y) − Ut )

Similarly, we have
−2
d log(gt (x) − gt (y)) = dt.
(gt (x) − Ut )(gt (y) − Ut )

So we have
   
2 2
dGt (x, y) = − im im dt.
gt (x) − Ut gt (y) − Ut

This is exactly what we wanted it to be.

46
Index III Schramm–Loewner Evolutions

Index

H01 (D), 41 harmonic, 5

D, 4 harmonic function, 5
H, 4
Q, 10 Koebe-1/4 theorem, 6
U, 6
Lévy characterization, 23
BESd , 24 locality, 32
Bessel process, 24 locality property, 32
Bessel stochastic differential locally growing, 18
equation, 25
Beurling estimate, 16 maximum modulus principle, 6
Brownian excursion, 34 maximum principle, 6
mean value property, 6
Cauchy–Riemann equations, 5
chordal Loewner equation, 5 non-decreasing, 17
chordal Loewner ODE, 18
compact H-hull, 10 parametrized by half-plane capacity,
compact hull, 8 18
complex Brownian motion, 5
restriction property, 36
conformal map, 4
conformal Markov property, 19 Schramm–Loewner evolution, 20
Schwarz lemma, 6
Dirichlet inner product, 41
Schwarz reflection principle, 10
driving function, 18
self-avoiding walk, 34
equivalent events, 27 SLE, 20
square Bessel process, 24
Gaussian Hilbert space, 42 square Bessel stochastic differential
equation, 24
half-plane capacity, 11 stochastic differential equation, 23

47