Modular Hamiltonians for deformed halfspaces and the averaged null energy condition
HJE
Modular Hamiltonians for deformed halfspaces and the averaged null energy condition
Thomas Faulkner 0 1 2
Robert G. Leigh 0 1 2
Onkar Parrikar 0 1 2
Huajia Wang 0 1 2
0 Department of Physics, University of Illinois , USA
1 the second term comes from @R
2 To see this, we focus on a particular term @
We study modular Hamiltonians corresponding to the vacuum state for deformed halfspaces in relativistic quantum eld theories on R1;d 1. We show that in addition to the usual boost generator, there is a contribution to the modular Hamiltonian at rst order in the shape deformation, proportional to the integral of the null components of the stress tensor along the Rindler horizon. We use this fact along with monotonicity of relative entropy to prove the averaged null energy condition in Minkowski spacetime. This subsequently gives a new proof of the HofmanMaldacena bounds on the parameters appearing in CFT threepoint functions. Our main technical advance involves adapting newly developed perturbative methods for calculating entanglement entropy to the problem at hand. These methods were recently used to prove certain results on the shape dependence of entanglement in CFTs and here we generalize these results to excited states and real time dynamics. We also discuss the AdS/CFT counterpart of this result, making connection with the recently proposed gravitational dual for modular Hamiltonians in holographic theories.
AdSCFT Correspondence; Field Theories in Higher Dimensions

1.1
Setup & summary of results
Modular Hamiltonian for deformed halfspace
Reduced density matrix
Modular Hamiltonian
1 Introduction
2
3
4
5
2.1
2.2
3.1
3.2
5.1
5.2
Averaged null energy condition
Positivity of KbA0
Computing hKbA0 i
KbA
hKbAi
Modular Hamiltonians in AdS/CFT
Discussion
Sharpening the argument
Generalizations
A Cuto
at the entangling surface
A is the reduced density matrix of the state
over the subregion A. The modular
Hamiltonian is, in general, a complicated, nonlocal operator and not of much practical use.
However, the situation greatly simpli es for the vacuum state in the case of certain special
symmetric subregions. For instance, the modular Hamiltonian for a halfspace in relativistic
quantum
eld theories takes a very simple form; it is the restriction to the half space of the
generator of boosts which preserve the entangling surface [1], and consequently generates
a local and geometric modular
ow. A similar construction is also possible for spherical
subregions in conformal eld theories [2], for null slabs in the case where the vacuum state
is de ned with respect to the generator of null translations on a null hypersurface [3, 4]
etc. Recently, it has been argued that the modular Hamiltonian for states with classical
was made in [7] (following previous work in [8{12]), where perturbative techniques were
used to study the shapedependence of entanglement entropy in conformal eld theories. In
the present paper, we adapt these techniques to study modular Hamiltonians for deformed
halfspaces in relativistic (not necessarily conformal) quantum
eld theories.
The study of shape dependence of entanglement is an important task for several
reasons. The entanglement structure of quantum systems is highly constrained by powerful
inequalities, such as strong subadditivity of entanglement entropy, positivity and
monotonicity of relative entropy, etc. In many situations, these entanglement inequalities further
imply fundamental constraints on the properties of quantum
eld theories. For instance,
the strong subadditivity property was used in [13, 14] to prove an entropic version of the
ctheorem for renormalization group
ows in two and three dimensions. Similarly, the
properties of relative entropy have been used to prove several interesting results such as
the Bekenstein bound [15], the generalized second law for causal horizons [16] and the
covariant entropy bound in the context of semiclassical gravity [3, 4]. Entanglement
inequalities have also been shown to constrain the bulk geometry in states with classical
gravity duals [17{20]. The entanglement inequality which will be relevant for our purpose
is that the full modular Hamiltonian for the vacuum state
KbA = KA
KAc
(1.2)
(i.e., the di erence between the modular Hamiltonian of the subregion A and that of the
complementary subregion Ac) satis es a \monotonicity" property under inclusion [21]. This
means that if we shrink the subregion A, then the corresponding change in the full modular
Hamiltonian
KbA is a negative semide nite operator. This property in fact follows from
the monotonicity of relative entropy, as was shown in [22]. In the present work, we will show
that this monotonicity property of the full modular Hamiltonian along with perturbative
results on the shape dependence of the modular Hamiltonian allow us to prove another
fundamental constraint, namely the averaged null energy condition (ANEC)
Z 1
1
1However, there are alternative proposals for pointwise quantum energy conditions.
See for
example [27{29].
{ 2 {
for many special cases such as free scalar and Maxwell elds in general dimension [30{32],
arbitrary quantum
eld theories in d = 2 with a mass gap and some assumptions on the
stress tensor [33], CFTs with classical gravitational duals in general dimension [34], etc.
Wall has also argued that the ANEC holds true for free or superrenormalizable eld theories
in general dimension [16]. In summary, there is substantial evidence so far to suggest that
the ANEC is satis ed by generic quantum
eld theories on Minkowski spacetime, but a
general proof has been missing hitherto (although see [35] for an argument involving certain
assumptions on the OPE of nonlocal operators)  in this paper, we will partially ll this
gap. On the other hand, the ANEC is known to be violated in general curved spacetimes,
but an alternative proposal called the selfconsistent achronal ANEC exists in this case 
see [36{38] and references therein for further discussion. While this is out of the scope
of the present paper, our results can nevertheless be extended to prove the ANEC along
static bifurcate Killing horizons even in curved spacetimes.
There is another motivation for trying to prove the ANEC in Minkowski spacetime.
In [39], Hofman and Maldacena (HM) showed that in a conformal eld theory the validity of
the ANEC in a certain class of states created by operator insertions implies bounds on the
coe cients appearing in the threepoint correlation functions of that CFT. For instance,
in d = 4 they used this to derive a bound on the ratio of central charges
1
3
a
c
31
18
(1.4)
where a and c are the coe cients of the Euler density term and the Weyl tensor squared
term in the conformal anomaly; tighter bounds can be obtained by imposing
supersymmetry. While the assumption of the ANEC was considered reasonable, in the original paper
no proof was given. Since then, there have been several attempts at a proof of the HM
bounds with varying levels of success [35, 40, 41].
In particular, using analytic bootstrap methods the HM bounds were proven for a class
of threepoint functions in [42], building on the work of [43, 44]. These methods take as an
input crossing symmetry and re ection positivity and apply these principles to various four
point functions in a lightcone limit to delicately extract the HM bounds. In particular in
this guise the HM bounds were related to causality properties of correlation functions in
a shockwave background [44].2 In contrast, we will show that the general HM constraints
on CFT threepoint functions can be extracted directly from the threepoint function itself
 when the threepoint function is interpreted as calculating some modular energy of the
CFT in an excited state. Overall it is satisfying to see the ANEC, and consequently the HofmanMaldacena bounds, arise as a natural consequence of the fundamental constraints satis ed by the entanglement structure of the vacuum.
1.1
Setup & summary of results
We now outline the calculation we are interested in, and present a brief summary of our
results. Consider the density matrix j ih j corresponding to a pure state de ned on the
2In theories with gravity duals, the HM bounds have also been shown to be related to bulk causality
constraints [34, 35, 45, 46].
{ 3 {
product h
= hA
density matrix
Cauchy surface . Let us partition
into two subregions A and its complement Ac. For
local quantum
eld theories, we expect the Hilbert space h
to factorize into the tensor
hAc . If this is the case, we can trace over hAc to obtain the reduced
A = TrAc (j ih j
)
which contains all the relevant information pertaining to the subregion A. The
entanglement entropy between A and Ac is de ned as the von Neumann entropy of A
In this context, the boundary @A of A is referred to as the entangling surface. The modular
Hamiltonian (also known as the entanglement Hamiltonian) KA is de ned as
Similarly, we can also de ne the modular Hamiltonian corresponding to the region Ac,
which we denote KAc . We can combine KA and KAc into another useful operator:
(1.5)
(1.6)
(1.7)
(1.8)
(1.9)
(1.10)
(1.11)
(1.12)
which we will refer to as the full modular Hamiltonian.
In this paper, we will primarily study the operators KA; KAc and KbA for the vacuum
state of a relativistic quantum
eld theory (as such we drop the label
from now on), with
the region A being a slightly deformed halfspace. To specify the geometry in more detail,
let us pick global coordinates x
= (x0; x1;
; xd 1) = (x0; x) on R1;d 1, where x0 is the
time coordinate, and x denotes spatial coordinates. Pick the Cauchy surface
x0 = 0, and consider the half space A0 given by
The vacuum modular Hamiltonian for the half space takes a particularly simple form [1]
i.e., it is the generator of boosts which preserve the entangling surface restricted to the
region A0, a result known as the BisognanoWichmann theorem [1]. Correspondingly, the
full modular Hamiltonian is given by the full boost generator
KbA0 = 2
Z
d
d 1x x1 T00(0; x)
Note that KbA0 is a conserved charge, and as such annihilates the vacuum KbA0 j0i = 0.
(This later property is true for more general regions as well.)
One can consider a small deformation of the region A0 to
KA =
ln A
KbA = KA
1Ac
1A
KAc
A0 = n
x 2 R1;d 1jx0 = 0; x1 > 0 :
o
Z
A0
KA0 = 2
d
d 1x x1 T00(0; x) + constant
SEE[ ; A] =
TrA
A ln A :
A = n
x 2 R1;d 1
x0 = 0; x1 > (~x)o
{ 4 {
c
c
H+
Also shown are the Rindler horizons H
directions ~x are implicit.)
line), such that D(A) (darker shaded region) is contained inside D(A0) (lighter shaded region).
corresponding to the regions A0 and Ac0. (The transverse
where (~x) is a smooth function of the (d
2) transverse spatial coordinates (parametrizing
the entangling surface), collectively denoted by ~x = (x2;
; xd 1). The deformation is
special in that it is restricted within the original Cauchy surface. We can generalize this
to also include timelike deformations (see gure 1)
A = n
x~ 2 R1;d 1
x~0 = 0(x1; ~x); x~1 = x1 + 1(x1; ~x); x1 > 0 :
(1.13)
o
We identify the in nitesimal as the deformation vector eld and pick to point inward,
i.e. D(A)
D(A0) (where D denotes the domain of dependence).
Our primary results in this paper are as follows:
(i) We will rst show that the modular Hamiltonian KA, up to rst order in the shape
deformation, is given by
Z
H+
Z
H
T
+ 2
Z
A0
U KAU y = KA0
2
where U : hA ! hA0 is a unitary transformation the details of which we will specify
later, H
are the future and past Rindler horizons of D(A0) shown in gure 1, and
appearing in the second and third terms above are components of the vector eld
on the entangling surface in lightcone coordinates x
= x0
x1. Similar expressions
can also be written for KAc and KbA.
(ii) We will then consider the expectation value h jKbAj i in states of the form
(1.14)
(1.15)
j
i = e
H
O (0; x)j0i
{ 5 {
and linear combinations thereof, where H is the Hamiltonian and O is an arbitrary
local operator whose quantum numbers (dimension, spin etc.) are collectively denoted
by .3 The factor of e
H is added to make these states normalizable. In a CFT this
class of states is a basis for the entire Hilbert space, via the stateoperator mapping.
For a general QFT similar statements should hold. In fact, there is no obstruction to
generalizing our argument to include states created by many local and even nonlocal
operators inserted throughout the lower half Euclidean plane. Further, we could also
insert the operators in real time. In the interest of simplifying our presentation we
choose to represent our state via a single operator insertion on the Euclidean section,
although we expect all our conclusions to go through even in the more general case.
We then show that equation (1.14), along with the positivity of the operator KbA0
KbA
(i.e. monotonicity under inclusion, which recall follows from the monotonicity of
relative entropy) implies the averaged null energy condition (ANEC)
Z 1
1
remarks below equation (1.15), our derivation of the ANEC also applies to these
states.
(iii) Finally, we also discuss the (vacuum) full modular Hamiltonian for deformed
halfspaces in CFTs with classical gravity duals, which allows us to make contact with the
recent proposal by Ja erisLewkowyzcMaldacenaSuh (JLMS) [6] for the holographic
dual to the modular Hamiltonian.
At this point we should mention that in continuum quantum
eld theory there are
signi cant ultraviolet (UV) issues associated with the de nition of the reduced density
matrix for a region, often resulting in divergences for entanglement entropy and modular
energy which are local to the entangling surface. These issues and associated divergences
are however not present for quantities like the relative entropy, and the full modular
Hamiltonian [47, 48]. Since this is ultimately what we are interested in, and in the interest of
simplicity of presentation, we will for the most part suppress the need for a UV cuto
at
the entangling surface. Indeed the answers we will nd will be nite, partly justifying this
approach. For further discussion on how to include such a UV cuto in our calculation, see
appendix A, where we will argue for the irrelevance of the details of such a cuto beyond
3In the case of tensor operators, we contract them with appropriate polarizations, for instance O (x) =
its existence.
1 2 s
In this section, we give an explicit formula for the modular Hamiltonian KA of the vacuum
state over a deformed halfspace, to rst order in the shape deformation.
interest of generality, let us instead consider a more general state rather than the vacuum4
j i =
X c j
i = e
H X c O (0; x)j0i;
( > 0)
(2.1)
HJEP09(216)38
This state can be constructed similarly as a sum over pathintegrals, but with the operator
O
in the term proportional to c . The reduced density matrix
below (x0E ! 0 ; x1 > 0) the region A0 (see gure 2)
corresponding to j i, associated with the undeformed halfspace A0 is constructed as a
Euclidean path integral with speci ed eld con gurations above (x0E ! 0+; x1 > 0) and
where we have collectively denoted all the elds integrated over in the path integral as ,
1
and j 0i; 0i 2 hA0 are eigenstates of the eld operator
restricted to A0. The prefactor
NA0; is added to ensure the normalization of the density matrix, i.e. TrA0 A0; = 1. We
have explicitly displayed the dependence of the reduced density matrix on the metric
5
through the path integral measure (which we assume is di eomorphism invariant), the
action and the normalization.
For convenience, we will henceforth use the notation (inside pathintegrals)
X =
X
and below A, and with a realtime fold around x0E = 0 in the case of timelike deformations
( 0 6= 0). We can deal with this path integral by performing a di eomorphism f : x
!
x
, which maps A to A0. We can take
to be nonvanishing (corresponding to
non0
trivial f ) only within a small region jxE j < ` (for some `
cuto ). Of course, such a di eomorphism has a nontrivial action on the background metric
, but much larger than the
4Later, we will also need to compute the expectation value hKAi = TrA
AKA; in the excited state
j i5; Hsoerweebyderive the reduced density matrix
section which is used in constructing the Euclidean path integral is
.
we are denoting the metric in real time. Of course the corresponding metric on the Euclidean
A along the way while setting up the calculation for KA.
g = (f 1
)
{ 7 {
(2.3)
(2.4)
x
†
x
6The unitarity follows from the di eomorphism invariance of the measure: (f ) [D ]
[D (f 0)] =
[D 0]g :
at x0E =
the state j i, over the original half space A0 (solid blue line). The operator insertions are marked
. The black dot is the entangling surface (with transverse directions ~x implicit).
where
denotes the pullback. We claim that the reduced density matrix over A (with the
metric ) is given by
A; = U y A0;g U
restricted to A by j i; j i
where U is a unitary transformation, and
undeformed halfspace, but with the deformed metric g.
A0;g is the reduced density matrix over the
We now give a quick formal proof of this claim. If we denote the eigenstates of
2 hA, then we can construct a unitary 6 operator U : hA ! hA0
U =
[D ] j(f 1
)
ih j
Then the claim (2.5) can be checked explicitly by a series of manipulations on the
pathintegral de nitions of the above density matrices [49]
1
1
Z
Z
h j A; j i = NA;
[D ] X e S[ ; ]
+=
=
Z (f ~)+=
(f ~) =
= NA;
= NA0;g ~ =(f 1)
1 Z ~+=(f 1)
= h(f 1
) j A0;gj(f 1
)
i
= h jU y A0;gU j i:
[D(f ~)] X e S[ ;(f ~)]
[D ~]g X e S[g; ~]
x0E = ⌧
x0E = ⌧
A; , the second equality is obtained by
changing variables
= f ~ inside the path integral, while the third equality follows from
the assumption that the measure is di eomorphism invariant. We have throughout used
the fact that the operator insertions (denoted by X, following the de nition (2.3)) are away
from the region where the di eomorphism f has nontrivial support, and so f acts trivially
on these operators.
In the case where f is an in nitesimal di eomorphism, we can obtain a perturbative
formula for A0;g. Writing the deformed metric on the Euclidean section as
where
is appropriately Wick rotated to Euclidean space, we obtain
U A; U y =
A0; +
ddx g (x) A0;
coordinate in the (x0E ; x1) plane, then
where g
), and T is the angularordering operator: if
2 (0; 2 ) is the angular
g
=
) + O(
2
)
(2.8)
where H is the Heaviside step function. For the special case j i = j0i, we then obtain
space for the vacuum state
We are now in a position to construct the modular Hamiltonian over the deformed
halfKA;
ln A; =
U y (ln A0;g) U = U yKA0;g U
In order to perturbatively expand the right hand side in powers of , we use the
resolvent trick
ln A0;g =
1 +
which together with equation (2.11) gives
KA0;g = KA0; +
KA0 + O(
2
)
In the interest of simplifying notation, we will henceforth drop the explicit reference to
the Minkowski metric on
A0; , and simply refer to it as A0 . It is possible to perform
{ 9 {
0
xE
x
1
Rb
and the region outside is Re. Also shown is the brachcut @Re .
the
integral by going to the spectral representation (for details see [7, 12], where similar
calculations were performed). The result is
ddx g (x) A0
is=2 : T
: (x) iAs0=2
(2.17)
Z
1
A0
Since the operator is=2
generates modular evolution in Rindler time s, we see that the
stress tensor is e ectively liberated from the Euclidean section and inserted in real time.
We now arti cially split the integration region over which the stress tensor is inserted
into two parts: a small solid cylinder Rb of radius b around the entangling surface, and
its complement Re. We will later show that the contribution from inside the cylindrical
neighborhood vanishes in the limit b ! 0. The region of integration is thus R = Rb [ Re
where we should remember that R contains a branch cut along the surface A0. We now
write g
and integrate by parts on the region Re
KA0 =
is=2
A0
Z
R
e
: T
:
d
+
Z
A0
is=2 +
Kb
(2.18)
The rst term involves the divergence of the stress tensor; in the absence of other operator
insertions in the region where
has support, we can drop this term. (Indeed, expectation
values in states of the form (1.15) which we will be interested in have precisely this property,
since
has no support at the location of the operators O .) The second term is integrated
and gets two types of contributions: (i) from the boundary
surface,7 and (ii)
above and below the region A0, which we will refer to as the branch cut (see
gure 3). Finally
Kb represents the contribution (iii) from inside the cylinder Rb.
7Not to be confused with the UV cuto surface that we discuss in appendix A.
(i) Imaginary cuto
surface: let us rst deal with the term supported on the surface
@Rb. It is convenient to switch to complex coordinates
z = x
1
ix0E; z =
(x1 + ix0E):
+
Tzz(xs)e i + e 2s+i Tzz(xs)
z
HJEP09(216)38
In these coordinates, we nd where,
is=2
A0
n T (x)
is=2
A0
=
e2s i Tzz(xs) + Tzz(xs)ei
z
xs = (b sin( + is); b cos( + is); ~x)
Further, n is the (inward pointing) unit normal to @Rb
and z and z are the components of the vector eld
close to the entangling surface
in holomorphic coordinates
We now proceed by shifting the s integration contour s ! s + i in order to remove
the
dependence from the stress tensor. We do this after switching the order of
integration so that the s integral comes before the
integral. This step assumes
analyticity in the complex s plane and that the contributions from s !
which can be justi ed in a spectral representation of (2.18). This gives
1 vanish,
(2.19)
(2.20)
(2.21)
(2.22)
(2.23)
(e2sTzz : Tzz :) zei + (: Tzz : e 2sTzz) ze i
(2.24)
where now these stresstensors are evaluated at x0E = ib sinh(s); x1 = b cosh(s). We
can now perform the
integral using
Z 2
0
Z
the limit b ! 0.8 So we get
d
b
Z
d
d 2 Z 2
~x
d
1
ds
1
ds(e sTzz
es : Tzz :) z
(2.26)
d
e i = 2 e s ( s)
2
(s)
(2.25)
The delta function term above can be dropped since this term does not contribute in
0
1
Stack = 2 b
Z dd 2~x T1
1
:
8Actually, rather than drop this term, let us add it to a stack:
We will update Stack everytime we nd a term of this type in our calculation.
coordinates are de ned as x
= x0
x1), we obtain
the vector eld back to real time, i.e. z !
+ and z !
Naively, it might seem that all the terms on the right hand side vanish in the b ! 0
limit. In fact, the terms involving Tzz do indeed vanish in this limit.9 However, the
terms involving Tzz and Tzz get an enhancement from the s integral, coming from the
s
ln b and s
ln b limits respectively. Taking the limit b ! 0 and Wick rotating
(where the lightcone
Z
d
~x
dx+ +T++(x+; x
0
+
(2.27)
where note that the rst term on the right hand side is integrated over the future
Rindler horizon H+, while the second term is integrated over the past Rindler horizon
H , shown in gure 1.
(ii) Branch cut : now we come to the second remaining term supported over @Re+ [ @Re .
Once again, deforming the s contours to get rid of the
dependence from the stress
tensors, we obtain
=
Z
d
~x
dx1
ds
where we have also shown the Wick rotated Rindler coordinates. The hole region
Rb corresponds to r < b and since we are again working in a region close to the
9By this we mean that the Tzz terms do not contribute to matrix elements in the class of states (1.15)
which are of interest here. On the other hand, if we were to evaluate matrix elements in Rindler eigenstates
we would
nd potential divergences in this limit. Note that since we used a spectral representation for
A0
at an intermediate stage, we were exactly evaluating this in Rindler eigenstates. So the order in which this
limit is taken is a somewhat delicate issue which is best ignored on a rst pass. In appendix A we confront
this issue explicitly.
where t = @x0 , and the stress tensor is evaluated on the region A0, i.e. T
E
T (x0E = 0; x1; ~x) above. The rst term inside the brackets comes from @Re+ while
(after the contour deformation s ! s + 2
It is clear from equation (2.28) that the s integral precisely picks out the doublepole
at s = 0. A straightforward application of the residue theorem gives
~x
dx1 t
b
T (0; x1; ~x); KA0
(iii) Inside the hole: we can follow the same methods as in (i). Pick coordinates close to
the entangling surface such that:
ds2 = dr2 + r2d 2 + d~x2
! dr2
r2ds2 + d~x2
b
1
4 sinh2 s+i
1
2
+
1
4 sinh2 s i
2
!
t
A0is : T
: iAs0
(2.28)
(2.29)
(2.30)
entangling surface we can take the di eomorphism at leading order to be independent
of r. After shifting the integration contour s ! s + i and Wick rotating the
eld we have:
is+
2
A0
Z
d Z
d x
ds
is
2
A0
1
The lightlike coordinates where the stress tensor on the right hand side above is
located are x
=
re s. We still have to integrate (2.31) over:
: : : =
Z
d
d 2 Z
~x
I
drr
d
r<b
1
ds
1
dependence is the same as in (2.25) and we can again do the
integral. After ignoring the (s) contribution which vanishes in the limit b ! 0,10 this
has exactly the e ect of switching the angular integral in the Euclidean calculation to
a real time integral localized near the Rindler horizon: 0 < r < b and
(see gure 4). The integrand is the stress tensor coupled to a real time di eomorphism
1 < s < 1
of the metric
for the following vector eld:
Kb = 2
Z
0<r<b
ddx T
( x0)
We have again ignored a contribution localized at x0 = 0, coming from the derivative
of the step functions above, which vanishes in the limit b ! 0.11
It is not hard to see that (2.33) should vanish in the limit b ! 0. However it is
somewhat enlightening to go another route and instead integrate by parts on (2.33). We
get two terms, one from the r = b boundary and the other from precisely the past and
future Rindler horizons on the boundary of the domain of dependence of A0. It turns
out the former term cancels (2.26) prior to taking the b ! 0 limit (although we always
need b small), and the later term is exactly the desired result given in (2.27) . So in
the end when we add all the terms together, no b ! 0 limit is necessary and the null
10Once again, we add this term to the stack de ned in footnote 8:
Stack ! Stack 2
Z dd 2~x Z b
0
11These terms go into the stack as well:
Stack ! Stack + 2
Z dd 2~x Z b
dx1(T10 0 + T00 1)
= 2
0
Z dd 2~x Z b
0
+ 2T10 0 + (T11 + T00) 1
where in the second equality we have integrated by parts; this is then exactly the extension of the x1 integral
in (2.29) so that it ranges from 0 to 1. Even though all these terms vanish as b ! 0, it is satisfying that
they add up like this.
energy operators in (2.27) simply emerge. This is perhaps not too surprising since the
r = b surface is imaginary, and there should be no dependence on b, however we nd the
detailed cancelations that occur and the form in (2.33) intriguing (including in the running
footnote Stack), perhaps hinting that there is a di erent way to do this calculation directly
in real times.
To summarize, putting everything together, we nd that the modular Hamiltonian
over the deformed halfspace is given by
HJEP09(216)38
U KAU y = KA0
2
which is the result claimed in (1.14).12 We emphasize once again that the
appearing in
the second and third terms above are de ned at the entangling surface and in particular
do not depend on the null coordinates x
along the Rindler horizons. We can also derive
a similar expression for the modular Hamiltonian corresponding to the complement Ac
V KAc V y = KAc0 + 2
+T++
2
+ 2
A0c
t
T ; KAc0 (2.36)
where H
c are the Rindler horizons corresponding to the complement Ac0, and V : hAc ! hAc0
is a unitary transformation. Finally, putting these together, we obtain the following formula
for the full modular Hamiltonian
U KbAU y = KbA0
2
where we have de ned the light sheets L
transformation given by U = U
12Roughly speaking, the \nullenergy" terms measure the amount of modular energy leaving the Rindler
wedge, while the commutator term comes from the action of the unitary transformations on the original
(undeformed) modular Hamiltonian.
Z
H+
Z
c
H+
Z
L+
Z
H
Z
H
c
Z
L
= H
A0
t
Z
Z
T
T
+ 2
t
h
T ; KbA0
i
(2.37)
[ Hc , and U : h
! h is a unitary
In this section, we will consider the expectation value h jKbAj i in the class of states (1.15).
We will then use the positivity of the operator KbA0
KbA to prove the averaged null energy
condition within this class.
For completeness, we begin with a brief review of the argument that KbA0
KbA is a positive
operator, following [22].13 Consider any two states, which we take here to be the vacuum
j0i and a nontrivial pure state j i. Given an entangling region A0 and the corresponding
reduced density matrices A0 and
A0 , one de nes the relative entropy
(3.1)
(3.2)
(3.3)
(3.4)
(3.5)
(3.6)
(3.7)
gion A0.
tive quantity
Relative entropy has a number of interesting properties. For instance, it is a
posiS( A0 jj A0 )
0
Further, if we pick another region A such that A
A0 (more precisely, if D(A)
D(A0), where D(A) is the domain of dependence of A) then the monotonicity of relative
entropy implies
S( Ajj A)
S( A0 jj A0 )
Intuitively, the relative entropy measures the distinguishability between two states. From
this point of view, the monotonicity property states that the distinguishability between
two states decreases as we consider their reduced density matrices over smaller and
smaller regions.14
From equations (3.2) and (3.4), we obtain
S( A0 jj A0 ) = TrA0
A0 ln A0
TrA0
A0 ln A0
=
hTrA0
A0 KA0
+ hTrA0
TrA0 ( A0 KA0 )
i
A0 ln A0
TrA0 ( A0 ln A0 )
i
hKA0 i
SEE[A0]:
where KA0 is the modular Hamiltonian corresponding to the vacuum state over the
rehKAi
KAc0
hKA0 i
hKAc i
SEE[A] +
SEE[Ac0] +
SEE[A0]
SEE[Ac]
0
0
where all modular Hamiltonians are de ned relative to the vacuum. Adding the two
inequalities we have
D
13Similar arguments have been used in [4, 50]. A rigorous proof of the positivity of this operator can also
be found in [21] which uses methods of algebraic QFT.
replica trick.
14See [51{53] for eld theoretic calculations of relative entropy in excited states, using a version of the
Now, since all vacuum contributions vanish, we can drop the
annihilates the vacuum for any region A). This implies
. (This is because KbA
D
where the last equality follows from the purity of j i. Since this is true for any pure state
j i, we deduce that KbA0
KbA is a positive operator.
We now explicitly compute the expectation value of KbA0
KbA in the state
where as before we take A0 to be the halfspace x1 > 0, and A to be the deformed halfspace.
Using the relation we nd where
h jKbAj i = TrA
AKA
TrAc
Ac KAc
h j(KbA0
KbA)j i = T(
1
) + T(
2
)
where note that the unitary transformations U and V have dropped out inside the trace.
The second term above is straightforward to evaluate from equations (2.35) and (2.36):
Z
L+
T(
2
) = 2
+hT++i
2
where we have de ned
h
T
i + 2
h jY j i
h j i
:
Z
t
h
h KbA0 ; T
i
i
The rst two terms in (3.14) are the appropriate null energy expectation values (albeit
integrated along the transverse ~x directions with arbitrary coe cients
(~x)) which enter
in the averaged null energy condition. We will see below that the last term in (3.14) is
precisely cancelled by a contribution coming from T(
1
).
Consider for instance, the rst term in T(
1
); from equations (2.9) and (3.12) we obtain
Rd, and recall the notation
On the right hand side, the correlators h
i indicate Euclidean correlation functions on
A0
Z
L
hY i
X =
X
(3.9)
(3.10)
(3.11)
(3.12)
(3.13)
(3.14)
(3.15)
(3.16)
(3.17)
Note that the Euclidean correlation function appears naturally from the path integral
construction of the deformed density matrix for the excited state  we need only add an
insertion of KA0 along A0 and trace  resulting in the above correlation function. Because
of the KA0 operator insertion, we should remove an in nitesimal cut running along A0 from
the region over which we integrate the di eomorphism R. Bearing this in mind, we can
integrate by parts in the region R to obtain only a contribution from above and below the
KA0 operator insertion, yielding a commutator:
TrA0
vanishes at the locations of the operators O , which allows us to
drop the divergence of the stress tensor. For the full modular Hamiltonian we then have:
As promised, this term precisely cancels the last term in equation (3.14). We therefore
conclude that
Since
(KbA0
hKbA0 i
hKbAi
= 2
+hT++i
2
h
T
+ > 0 and
< 0 by construction (see
gure 1), the positivity of the operator
KbA) leads to the averaged null energy conditions
1
Z
L+
Z 1
Z 1
1
1
This concludes our proof that the monotonicity property of KbA implies the ANEC. While
we presented the proof above in the context of halfspaces in Minkowski spacetime, the
above calculation can also be extended in an obvious way to general static bifurcate Killing
horizons. In this case we would be studying the modular Hamiltonian for small
deformations of the entangling cut away from the bifurcation point in the HartleHawking state.
The monotonicity constraint then leads to the ANEC for complete null generators of the
Killing horizon.
4
Modular Hamiltonians in AdS/CFT
In this section we make a connection between our results and the recently proposed JLMS
formula [6]:
KACFT =
4GN
+ O(GN )
(4.1)
where A denotes the boundary subsystem, and M denotes the bulk region enclosed by
the minimal/extremal RyuTakayanagi/HRT [54, 55] surface @MnA (which ends on the
entangling surface @A, and is homologous to A). Further, the
denote local terms on the
extremal surface, which will not be relevant in the following discussion. The result (4.1)
arises as a consequence of the formula for quantum corrections to the RyuTakaynagi
entropy [56, 57] (see also [58].)
Here, we want to perform a simple consistency check of our formula (2.37) for the
full modular Hamiltonian of deformed halfspaces against equation (4.1). In particular,
restricting to pure states so that the bulk area operator Area@MnA evaluates the same over
A and Ac, (4.1) allows us to equate the full modular energies between the boundary and
the bulk theories
KbACFT = Kb Mbulk + O(GN )
(4.2)
HJEP09(216)38
where note that the local terms on the extremal surface have also dropped out.
In order to make contact with our previous results, we take A to be a small deformation
of the boundary halfspace A0 = fx1 > 0; x0 = 0g. If we use the coordinates z; x0; x1; ~x
on the Poincare patch of AdS, with
(4.3)
(4.4)
(4.5)
gAdS =
dz2
(dx0)2 + (dx1)2 + (d~x)2
z2
;
then the corresponding (undeformed) extremal surface in the bulk is given by the
codimension two surface x
0 = x
1 = 0, and we have the corresponding undeformed region
M0 = fx1 > 0; x0 = 0g. To linear order in the CFT shape deformation , we then expect
h KbAC0FTi CFT = h bulkKb Mbu0lki bulk + O(GN );
where we have evaluated the deformations in the CFT and bulk modular Hamiltonians in
an excited boundary state j iCFT and the dual bulk state j ibulk respectively. Further,
bulk is the deformation in the bulk minimal surface as a consequence of the boundary shape
deformation, and approaches
in the limit z ! 0; it is xed away from the boundary by
the requirement that the deformed bulk surface remain extremal [55].
For simplicity, we focus on null deformations on the CFT side, i.e.
= 0. The
lefthand side of equation (4.4) has been computed previously; see equation (3.20).
Furthermore, if we view the bulk e ective theory as a weakly coupled quantum
eld theory in
background AdS geometry, then we expect that our at space arguments from the previous
sections can be extended to the bulk e ective theory  this is because we have a Killing
vector eld in AdS which generates a boost around the unperturbed extremal surface in the
(x0; x1) plane. The bulk modular Hamiltonian then satis es a covariant version of (3.20):
Z
p
h bulkKb Mbulk
i bulk =
h b+ulk(z; ~x)hT+bu+lk(x+; z; ~x)i bulk;
where hij is the induced metric on the undeformed minimal surface @M0:
h =
dz2 + (d~x)2
z2
:
A consequence of JLMS combined with our calculations is therefore:
Z
d
d 2~xdx+ +
h CFTjT+C+FTj CFTi
=
Z
hdd 2~xdx+dz b+ulkh bulkjT+bu+lkj bulki
(4.6)
Our goal now is to establish this equivalence using the usual rules of AdS/CFT [59{61]. In
particular in order to apply the JMLS argument the state under consideration must have
a perturbative in GN backreaction on the bulk AdS spacetime and so we will consider
boundary states of the form j CFTi = O(x )j0iCFT, where O is a single trace primary
scalar operator of dimension
= O(
1
) in terms of largeN counting. The operator is
= 0). In this case the dual bulk state can be
identi ed as j bulki = limz!0 z
(z; x )j0iAdS for the corresponding bulk
eld
. The
equality in (4.6) can already be read o from the work of Hofman and Maldacena [39] but
here, for completeness, we will give another derivation.
For a CFT with a weakly coupled Einstein gravity dual in the bulk, the l.h.s. can be
computed using Witten diagrams. In particular, we assume that the relevant part of the
tensor in O(1=N ). We thus have for the integrand:
hOy(x? )T+C+FT(x)O(x )i =
T bulk ( )
Z
Z
m2 2 is the leading order bulk stress
n
=
dzddy
p
GD++(z; x0; x)T bulk
D (z; x0; x )
p
1
2
(4.8)
(4.9)
o
(4.10)
(4.11)
bulk Lagrangian takes the form:
S
bulk
Z
AdSd+1
p
G
1
GN
The leading order Witten diagram contribution to the l.h.s. of (4.6) comes from the coupling
between the graviton (h
) and the scalar stress tensor:
(RG + ) +
G
where D
(z; x0; x) and D (z; x0; x ) are the bulktoboundary propagators for the graviton
and scalar respectively. We identify the products of scalar boundarytobulk propagators
as giving rise to the expectation value of the bulk stress tensor in the state j ibulk:
T bulk
D (z; x0; x ) = hT bulk(z; x0)i bulk
in T bulk( ). When viewed as a bulk
operator, its expectation value in the bulk state j bulki is given by:
h
(z; x0)i bulk = lim
; 0!0
0
(z; x0) ( ; x )
(4.12)
i
The leading order (disconnected) diagram of the bulk 4point function is given by products
of bulktobulk propagator:15
i
i
= @ Dbulktobulk( ; x? ; z; x0)@ Dbulktobulk(z; x0; 0; x )
The boundarytobulk and bulktobulk propagators are related by the limit:
!0
D (z; x0; x) = lim
Dbulktobulk(z; x0; ; x)
(4.13)
tions hold for the other two terms in T bulk and we conclude that:
(z; x0)i bulk . Similar
relaZ
hOy(x? )T+C+FT(x+; xi)O(x )i =
GD++(z; x0; x)hT bulk(z; x0)i bulk
(4.14)
Z
Z
We need to integrate this relation over R dx+dd 2~x +(xi) on the boundary. In particular,
since
+(~x) is independent of x+, we can take it out of the null integral, and replace
R dx+D++(z; x0; x+; ~x) =
++Dshock(z; ~x0; ~x) (x0 ), where Dshock(z; ~x0; ~x) is the
boundarytobulk propagator for the shock wave graviton mode: h++(z; x0 ; ~x) = f (z; ~x0) (x0 ).
In AdSd+1, Dshock(z; ~x0; ~x) is determined by solving Einstein's equations for this metric
uctuation giving the shockwave equation:16
Dshock(z; yi; xi) = 0; lim Dshock( ; yi; xi) !
yi)
The factor ++ (x0 ) localizes the bulk integral onto L+(@M0), and projects onto the
(++) component of bulk stress tensor:
+(xi)hT+C+FT(x+; xi)i CFT =
ph ~+(z; ~x0)hT+bu+lk(z; x0 = 0; x0+; ~x0)i bulk
~+(z; ~x0) = z 2
d
d 2
~x +(~x)Dshock(z; ~x0; ~x)
One can
nally check from (4.15) that ~+(z; ~x0) satis es the extremal bulk extension of
d
1
z
~+(z; ~x0) = 0; lim ~+( ; ~x0) !
+(~x0)
!0
which is precisely what de nes b+ulk(z; ~x), making (4.16) equivalent to (4.6), consistent
with JLMS formula.
15We analytically continue these propagators to real time such that the ordering is the appropriate one
for computing expectation values in the state j ibulk.
not used this fact.
16This shock wave metric is actually a full nonlinear solution to Einstein's equations although we have
(4.15)
(4.16)
(4.17)
In this paper, following the circle of ideas in [16, 22, 50], we established a relation between
the monotonicity of relative entropy and the averaged null energy condition in arbitrary
QFTs, and in so doing proved the most general HofmanMaldacena bounds on the data
in CFT threepoint functions.
We will now summarize the perturbative calculation we
performed to establish this connection and then conclude with possible future work.
The general goal was to study perturbatively the shape dependence of modular
Hamiltonians/energies. We did this by applying \perturbation theory for reduced density
matrices" which turns out to have some novel features which we describe now. Schematically
the important term in our calculation (2.15) came from expanding the log used to de ne
the modular Hamiltonian. Here we give an alternative description of this expansion:
ln A0 (1 +
1
A0
) = KA0
1
X(
1
)n Bn h
n=0
n!

h
KA0 ; KA0 ; : : : KA0 ; A0
h
n {tizmes
1 iii
}
+O(
2
) (5.1)
where Bn are the Bernoulli numbers. The right hand side comes about due to the non
commutativity of the two matrices in the log on the left hand side. That is, these are the
usual nested commutator terms in the BakerCampbellHausdor formula keeping only
terms to order O( ) (see also [62] for related discussion).
This set of nested commutators clearly has something to do with the evolution with
respect to KA0  or in other words modular ow. So it should come as no surprise that
these terms can be resummed into an integral over
is=2
A0
1
A0
is=2
A0
multiplied by
some kernel  a fact we used in (2.17). In fact, in going from equation (5.1) to (2.17), one
simply uses the following integral representation of the Bernoulli numbers [63, 64]:17
Bn =
( i)n Z 1
(
2
)n
1
ds
s
n
4 sinh2( s+2i )
(n 2 Z):
(5.2)
Surprisingly this integral and kernel as well have the e ect of switching the original
Euclidean di eomorphism contained in
and used to move around the entangling surface,
to a real time di eomorphism determined by the new vector eld given in (2.33). From
here the null energy operators involved in the ANEC just pop out as boundary terms when
integrating by parts over the real time di eomorphism. Of course in real times now a new
boundary has opened up; what previously was the codimension 2 entangling surface at
the origin in Euclidean space becomes a null hypersurface along the Rindler horizon where
the null energy operators are de ned.
The noncommutativity emphasized in (5.1) was of fundamental importance to our
calculation. We feel that we do not fully understand the magic behind this calculation and
that there are new surprises lurking if we go to higher orders in perturbation theory and try
to systematize this approach. Similar tools were applied in [7, 12] to entanglement entropy
17Note that we pick the convention where B1 = + 12 ; also recall that B2m+1 = 0 for m = 1; 2
The corresponding terms in the integral representation (5.2) pick out the residue at s = 0, which is only
nontrivial for n = 1.
where it was important to control these commutator terms in order to
nd agreement
between this perturbative approach and known results from AdS/CFT. Here we have also
established a similar agreement with AdS/CFT and in particular the recent proposal by
JLMS [6] for the modular Hamiltonian in AdS/CFT.
Apart from gaining a deeper understanding into the inner working of these calculations
we now give some detail of future work that we think would be valuable to pursue.
Sharpening the argument
In the main sections of the paper our derivation eschewed any issues related to the precise
de nition of entanglement and modular energy in quantum
eld theory. Indeed these
quantities are expected to be a icted by signi cant UV divergences, and possibly even
ambiguities related to how one splits the degrees of freedom between A and Ac. Thus
in order to calculate these quantities we must specify a regulator and a prescription for
splitting the Hilbert space. However it became clear to us that we never needed to do this,
and so any real discussion of a regulator was relegated to appendix A.
Ultimately this should not have come as a surprise, the
nal goal was to calculate
either relative entropy or the full modular Hamiltonian  both of which are expected to be
UV
nite quantities and both of which can actually be given a de nition directly in the
continuum [47, 48]. This de nition however was not convenient for our current calculation
so at an intermediate step we needed to calculate the expectation of the full modular
Hamiltonian in terms of the UV sensitive (half) modular Hamiltonian. Since we never
explicitly saw these UV divergences, our manipulations should be regarded as formal.18
Appendix A is an attempt to remedy this, by giving some details of a brick wall like
regulator [66] that renders the modular energy and associated quantities wellde ned. The
brick wall regulator introduces dependence on the boundary conditions one chooses for
elds at the wall close to the entangling surface.
The regulated version of relative entropy does not satisfy the property of monotonicity
(for a nite but small cuto scale) since the brick wall cuto is a rather drastic modi cation
to the theory that does not allow one to compare di erent spatial regions with the same
modi cation. So to claim a completely rigorous proof of the ANEC we still need to show
that when we remove the brick wall cuto the quantity we get is the continuum version of
relative entropy  which is then known to be monotonic [47]. This requires methods that
are beyond the scope of this paper, and we leave this to future investigations. Ultimately
we would like a mathematically rigorous derivation, perhaps without reference to density
matrices and using methods of algebraic quantum
eld theory [21, 48].
Finally we would like to understand if there are any restrictions on the state in which
we calculate the expectation value of the deformed modular Hamiltonian. For example
we formulated our state in terms of a local operator insertion at x
su ciently general for a CFT. More generally, say for relativistic theories, our argument
will go through relatively unmodi ed if we just insert a general state of the theory and
0
E =
, which is
18They might be regarded as about as formal as the usual derivation of the replica trick for Renyi entropies
in terms of a partition function on a singular surface [65].
its conjugate in
at space along the Euclidean time slices x
. However we are
required to separate the di eomorphism that moves around the entangling surface away
0E =
. We can make the region in which the di eomorphism acts small but we
should be limited by j j the size of the di eomorphism vector eld at the entangling surface.
This presumably puts some restriction on the state such that the expectation value of the
stress tensor cannot get arbitrarily large. For example if we work with the state created by
that the perturbative expansion converge and we can always arrange this to happen by
1 < j j 1. This is likely just the restriction
a local operator insertion j RH h
T
i j
taking a small enough spatial deformation.
One obvious generalization involves attempting to prove the ANEC in other spacetimes as
well as along more general complete achronal null geodesics.19 Along these lines it might
be an easier rst step to try to apply the methods of this paper to stationary but not
static black holes with the null generator lying along a bifurcate Killing horizon (like the
Kerr black hole). Since we used the framework of perturbation theory starting from a state
described by a known density matrix (the HartleHawking state) we are not very optimistic
this will succeed when we don't have such a starting point.
Instead perhaps a more fruitful direction to pursue would be to consider the generalized
second law (GSL) for quantum
elds outside of a black hole with a static bifurcate Killing
horizon. Here we are referring to the work of [16] where the GSL was proven for free as well
as super renormalizable QFTs.20 The GSL applies to the following generalized entropy:
Sgen =
Area(@A)
4GN
+ SEE( A
)
(5.3)
(5.4)
(5.5)
HJEP09(216)38
19These are geodesics where no two points on the curve are timelike separated. The ANEC is known to
fail in curved spacetimes where the null geodesic is chronal [37, 67].
20The Hawking area theorem proves the GSL when the area term dominates in the GN ! 0 limit  that
is for a classical dynamical background where the classical matter satis es the NEC. As discussed in [16],
what remains, is to prove the GSL when classically the area does not increase  for quantum
elds on a
stationary black hole background plus free gravitons. For obvious reasons here we then focus on the static
case, and leave out gravitons for simplicity.
where Area(@A) refers to the area of a codimension1 slice of the Killing horizon where
the spatial region A ends (@A) and SEE is the entanglement entropy of the quantum
elds outside this horizon slice. Applying the monotonicty of relative entropy to SEE( A
Sgen
Area(@A)
4GN
+
where now
is a
nite null deformation ( x+ =
+(~y)) of the entangling surface @A
to the future of the bifurcation surface @A0. The change in the area is simply due to the
perturbative back reaction of the quantum elds on the spacetime via Einstein's equations:
d
d 2
~x
Z 1
+
dx+x+ hT++i
+
where we have made use of the Raychaudhuri equation with the correct future boundary
condition appropriate for a causal horizon.
To make further progress we need some handle on KA for general null deformations
away from A0. This does not sound very promising for our perturbative approach, however
it does seem like we can carry out our calculations to arbitrary orders in
+ [68]. Thus
with some luck we might be able to prove a statement about KA and get a handle on (5.5)
and possibly show the GSL in this case,
Sgen
0. A further hint comes actually from
AdS/CFT. For a Rindler space cut we have carried out a more detailed calculation21
than that outlined in section 4 where we previously showed the equality between the null
energy operators in the bulk and boundary. More generally one can show for nite null
deformations, but small perturbations to the state (in the 1=N sense):
2
Z
4GN
+
+ 2
T+C+FT
Z
d 2 p Z 1
~x h
+
bulk
dx+ x
+
+
bulk
T+m+atter
(5.6)
and our notation is the same as that in section 4, where for example bulk is the bulk
HRT extremal surface corresponding to the deformation + on the boundary and M is the
spatial region between this extremal surface and A on the boundary. Here the area term
is the change in the area of the extremal surface due to the backreaction on the metric g
in the state
(via Einstein's equations.) Note that the extremal surface condition in pure
AdS for
nite null deformations remains a linear di erential equation that matches with
the in nitesimal version (4.17) and so b+ulk is the same extension as that used in section 4.
Now comparing this statement with that of JLMS [6] we could consistently identify the
+
modular Hamiltonian for nite null deformed regions as:
KACFT =? 2
Z
d
~x
dx+ x
+
+
+ T+C+FT
(5.7)
M
up to an additive constant, with a similar equation holding for the bulk region modular
Hamiltonian Kbulk. This is certainly not a proof. We have made two guesses (for the
bulk and the boundary) and shown them to be selfconsistent. And in particular this only
works for a special class of states that don't have a large back reaction on the bulk. Note
that this last issue also plagued our comparison between the bulk and boundary for small
deformations. We simply note here that our perturbative approach, when considered at
higher orders, can possibly prove such a statement.22 Of course if (5.7) is true then the
GSL follows trivially since the right hand side of the inequality in (5.5) just vanishes.
21This calculation has some overlap with [69] and the details will be reported elsewhere.
22Actually a simpler argument is to take the perturbative result we have derived for null shape
deformations and then use the QFT boost generator around the original undeformed Rindler cut to amplify the
deformation. This boost will then act on the state. However if we work in a su ciently general state this
should not matter. This process seems to work, and agrees with (5.7), when trying to construct the full
modular Hamiltonian and we leave the details of how this works for the half space modular Hamiltonian
for the future. We thank Aron Wall for suggesting this argument to us.
Finally we point out that in some sense these calculations have already been pushed to
higher orders. Rather than consider the excited state modular energy, if we just calculate
the modular Hamiltonian in the original vacuum state it should reproduce the entropy of
the vacuum. At rst order this vanishes but the second order variation of entropy in a
CFT was calculated in [7] using similar methods to this paper. This quantity is sometimes
referred to as entanglement density [70]. Although it was not realized at the time the answer
in [7] can be related to a correlation function of two \null energy operators"  the same null
energy operators that appear in the (half sided) modular Hamiltonian in this work. This
will be the subject of a forthcoming paper [71]. Taken together this hints at a unifying
picture for vacuum entanglement in CFTs related to null energy operators that may even
pave the way to a new understanding and proof of the RyuTakayanagi [54] and HRT [55]
proposals for calculating entanglement entropy in the vacuum state of holographic CFTs.
Acknowledgments
It is a pleasure to thank Xi Dong, Gary Horowitz, Veronika Hubeny, Aitor Lewkowycz,
Don Marolf, Mukund Rangamani, David SimmonsDu n and Aron Wall for discussions
and suggestions.
Work supported in part by the U.S. Department of Energy contract
DEFG0213ER42001 and DARPA YFA contract D15AP00108.
A
Cuto
at the entangling surface
In this appendix we would like to give a prescription for regulating the modular energy
that we calculate in the main part of the paper. We go through this in some depth because
the arguments we gave previously were somewhat formal. Although the quantity in which
we are ultimately interested  the full modular Hamiltonian  is UV
nite [48], at
intermediate steps we encountered quantities which are not. In particular the modular energy
of some state is expected to have the same UV divergences as the entanglement entropy of
that state because the di erence between them is the relative entropy which is UV
nite.23
Thus the issues here are the same as the usual issues of de ning entanglement entropy
in the continuum.24 There are several ways to de ne a regulated version of entanglement
entropy, but the most convenient for us will be a \brick wall" regulator [66]. This is so we
can still use Euclidean path integral methods to construct the density matrices in question.
Apart from possible IR issues the entropies are now
nite  the IR issues do not concern
us and cancel when evaluating the di erences between excited and vacuum states, at least
for states that are su ciently close to the vacuum near the boundaries of space.
23There are still several reasons to expect some of our intermediate steps to be nite. Any divergences
should be local to the entangling surface, and assuming our regulator is geometric[72{74] no such term
which respects the S(A) = S(Ac) purity condition can generate a divergence for rst order spatial shape
deformations. Similarly there is a general expectation that any such divergences cancel in the di erence
S( A) although we will nd evidence that this cancellation might not always occur. Of course
these variations and di erences are still calculated in terms of divergent quantities so we proceed.
24For a recent discussion of some of the issues involved see [72, 75]. When the QFT in question is a gauge
theory there are even questions about how the degrees of freedom are split between two spatial regions [76].
r = a
0
xE
x
1
state. We cut out a cylindrical region of radius r = a around the entangling surface, with
brickwalllike boundary conditions. Also shown is the ctitious cuto surface of radius r = b.
Roughly speaking we can simply go through our calculation in sections 2 and 3.2 with
the reduced density matrices de ned via path integrals on manifolds with a cylindrical
region of radius a cut out from around the entangling surface:
A0;g !
A0;g(a) (see
gure 2). In order to to do this consistently we should impose boundary conditions on the
cuto surface  we will assume that the boundary conditions at the cuto surface decouple
in the limit a ! 0. We might also need to add new degrees of freedom here [77, 78] and
there are good reasons to believe these should decouple when calculating such things as
relative entropy [76]. We also require the following:
Rotation/Boost invariance for the undeformed Rindler region. This is so that KA0
still has the interpretation as the generator of rotations/boosts around the cuto
surface. For example this will require that the stress tensor at the cuto
surface is
constrained to have zero rotation
should be required as part of the boundary conditions.25
ux T rjr!a ! 0 into the cuto
cylinder. This
For a more general region  we cut out a cylinder in Gaussian normal coordinates.
Here of course we do not have rotation invariance. We use normal coordinates so we
can still use the relation (2.5) derived in the main text. One way to do this is to
pick the di eomorphism to map the deformed entangling surface to Gaussian normal
coordinates  where the regulator is then picked to be a metric distance a orthogonal
to the surface away from A. For us this amounts to the choices:
2
2
g dx dx =
dzdz +
(A.1)
(A.2)
25Note however that this can fail in the case of chiral theories, in which case the boost symmetry is
anomalous [79{82].
where we have expanded the di eomorphism and the metric close to the entangling
surface. We then cuto the path integral which de nes A0;g(a) in (2.5) at r = jzj = a
supplying some appropriate yet unspeci ed boundary conditions. After making this
slight modi cations the di eomorphism acts the same way as in the bulk of the text
in particular there is no boundary term due to the displacement of the cuto surface
(although of course the new stress tensor could have delta function contributions on
the cuto surface.)
At the very minimum we require that for some local operator inserted in the path
integral that de ned A0;g(a) we should have:
a!0
lim TrA0 A0;g(a)O(x) = hO(x)i
(A.3)
(A.4)
(A.5)
Following the steps below (2.18) in section 2.2 for the change in modular Hamiltonian
due to the stress tensor deformation, the di erences are due to a slightly modi ed di
eomorphism and a di erent region of integration for the stress tensor R0 in the Euclidean
plane which is cuto
by r > a; see gure 2. This cuto is distinct from the imaginary
cuto
surface de ned in section 2.2 with r > b and we will take a
b. Indeed splitting
the contribution from the stress tensor integral into the three regions as we did previously,
there is only one term which is sensitive to the cuto in the limit a
b and this is the
contribution from inside the hole a < r < b which we call
Kb(a). The other two
contributions from the branch cut and the imaginary cuto surface give identical results as in the
main text. With the same set of manipulations we can write the potentially problematic
term as:
Kb(a) = 2
Z
a<r<b
ddx : T
: ge
where the resulting real time metric deformation is:
e
g dx dx =
1
2
dxidxj
Note the integration region is now a section of a solid hyperboloid. This is slightly modi ed
from (2.33) and (2.34) since we are now working in Gaussian normal coordinates. Of course
to analyze the limiting properties of (A.4) we should trace it against our state Tr A0 ( ).
At this point it is possible to remove the brick wall regulator a ! 0 from (A.4).
Dependence on a appears in the integration region R0 as well as implicitly in T
since this
is the appropriate eld theory stress tensor in the presence of a boundary. Note that the
boundary conditions on the brick wall in Euclidean space have naturally been mapped to
Rindler space in real times along the hyperbola r = a,
1 < s < 1. The claim is that we
can remove the regulator from the integrand using the requirement (A.3). Of course the
remaining ddx integral may still be divergent, however we found this not to be the case in
the main text. It is this sense in which we expect the boundary conditions on the surface
r = a to decouple. To show this rigorously we would have to show the integrand converges
su ciently uniformly to the a = 0 limit. Note that the metric deformation gij
in Rindler coordinates and so this is a mild condition on the behavior of the stress tensor
re jsj
close to the brick wall.26 To say anything further we would need to specify more about
the boundary conditions than we are willing to. However since the details of the boundary
conditions are not important for de ning relative entropy or the full modular Hamiltonian
in the continuum of a QFT, it must be the case that any divergences we might see here
should cancel when calculating these nal quantities. Instead we can turn this condition
around and demand that this should be true for any brick wall regulator that is supposed
to be a good regulator for calculating modular energy.
We now turn to the second contribution to the deformed modular energy (the rst term
in (3.12).) Compared to our previously obtained expressions we now
nd a contribution
from the boundary of the cuto region at r = a which looks like:
= a
I
TrA0
A0 KA0
n
h
T
(x)KA0 i
h
T
(x)i hKA0 i
(A.6)
If we instead calculate this contribution to the shape deformation of the full modular
hamiltonian (this was de ned as T(
1
) in (3.12)) we get a term coming from the complement
Ac0 which adds to give the total contribution to T(
1
) coming from the cuto surface @Ra:
T(
1
)
I
(x)i hKbA0 i
(A.7)
where we remind the reader that KbA0 is the undeformed full modular Hamiltonian. In the
limit a ! 0, the above term appears to be linearly suppressed; however one might worry
that there are potential enhancements from the stress tensor coming close to KbA0 in the
rst term above. To see that this does not happen, recall that KbA0 is a conserved charge,
namely the generator of rotations around the entangling surface. Consequently, we can
freely move it away from cuto surface as well as the other stress tensor inside the above
correlator. Here we have to take into account the fact that the boundary condition on
we could move KbA0 o to x0E =
the cuto
surface should not allow for any KbA0
ux into the cuto surface Tr
! 0. If
1, then the corresponding term would vanish, because
KbA0 annihilates the vacuum. However, as we keep moving KbA0 away from the stress
tensor, we will eventually cross the operator insertions O
or O
y (depending on whether
we move KbA0 towards x0
E !
contribution of the form h
T
1 or x0
[KbA0 ; O m
]
E ! +1). Every such crossing gives a nontrivial
i, where
denotes the remaining operator
insertions. However, it should now be clear that these remaining terms are correlation
functions between wellseparated operators (as long as
a), and we do not get any
enhancement to cancel the factor of a. Therefore, we conclude that the contribution from
the cuto surface to T(
1
) vanishes in the limit a ! 0. We claim victory.
Before moving on we note that if we did not do the subtraction that de ned the full
modular energy, the term (A.6) might still be divergent. We can give the following crude
estimate for any such divergence. Note that the half sided modular Hamiltonian, as an
26Note because of the :: vacuum subtraction for the stress tensor any divergence that might appear exactly
at the cuto surface when a is xed (say due to an image charge) is state independent and will cancel. The
potential divergence we are worried about is in the subsequent limit a ! 0.
integral over the stress tensor, can still be moved around but now it is always anchored to
the cuto surface. We can use this freedom to move the two stress tensors in (A.7) as far
apart as possible  on opposite sides of the hole. To get an estimate we now replace the
correlation function in the rst term of (A.6) with the CFT correlation function without
surface  on
at Euclidean space. We need to consider the OPE of two stress
tensors schematically of the from:
T T
X C ( x2) d+
(A.8)
where x
2
(a + r)2 + (~x
~x0)2 and where r > a refers to the location of the modular
Hamiltonian stress tensor. Here O
we can expand
(~x) and hO (~x0)i
possibly contribute a divergence. Close to the entangling surface for any divergent term
are some local primary operators and only scalars can
by taking ~x0
~x. We get some leading term from
the unit operator. But there are good reasons this term should vanish. Firstly it is state
independent and so should occur for the vacuum state, but there is simply no term we
can write down at linear order in
which is local to the entangling surface and has the
required rotation/boost invariance around the entangling surface. However now consider a
non unit operator, we no longer have rotation around the entangling surface. Then using
scale invariance we only expect a divergent contribution to the modular energy of the form:
C a
Z
d
d 2 Z
a
drr( x2) d+ =2+1=2 @ O (~x0)
2 d+
C
d
d 2
(A.9)
Naively one would have expected that such a contribution is not possible for uniform
deformations
independent of ~x  since in that case we can write an expression for KA
and KA0 in the absence of the cuto surface and there is seemingly no divergence. However
since these are half sided modular Hamiltonians it seems we should have allowed for the
possibility that even the undeformed KA0 has a local divergence:
KA0
2 d+
C
+ nite
Z
(A.10)
At least this seems to be required if we want the answer to be consistent with our UV
regulator and di eomorphism invariance. This calculation is far too crude to be trusted, but
it does suggest that any kind of brick wall cuto leaves one susceptible to state dependent
divergences in the half sided modular energy of the above nature. In the main text we
could have taken such divergences into account simply by adding (A.10) and this would
have generated the term in (A.9) without any need for a brick wall.
This new divergent contribution occurs if there is a very relevant
d
2 scalar
operator appearing in the T T OPE. For example it cannot be charged under any
symmetries. Symmetries would also disallow (A.10). It is not clear a theory with such a scalar
operator can exist  see [4] for a related appearance of such operators in state dependent
divergences for entanglement entropy. In recent proofs of the HofmanMaldacena bounds
from bootstrap methods, these operators also make an appearance [43]. In our work they
are always harmless and cancel when we calculate the full modular Hamiltonian.
Open Access.
This article is distributed under the terms of the Creative Commons
Attribution License (CCBY 4.0), which permits any use, distribution and reproduction in
any medium, provided the original author(s) and source are credited.
Phys. 17 (1976) 303 [INSPIRE].
entropy, JHEP 05 (2011) 036 [arXiv:1102.0440] [INSPIRE].
HJEP09(216)38
Rev. D 90 (2014) 044002 [arXiv:1404.5635] [INSPIRE].
eld theories and the Bousso bound, Phys. Rev. D 91 (2015) 084030
[arXiv:1406.4545] [INSPIRE].
[5] D.L. Ja eris and S.J. Suh, The gravity duals of modular hamiltonians, arXiv:1412.8465
[6] D.L. Ja eris, A. Lewkowycz, J. Maldacena and S.J. Suh, Relative entropy equals bulk relative
entropy, JHEP 06 (2016) 004 [arXiv:1512.06431] [INSPIRE].
[7] T. Faulkner, R.G. Leigh and O. Parrikar, Shape dependence of entanglement entropy in
conformal eld theories, JHEP 04 (2016) 088 [arXiv:1511.05179] [INSPIRE].
[8] V. Rosenhaus and M. Smolkin, Entanglement entropy: a perturbative calculation, JHEP 12
(2014) 179 [arXiv:1403.3733] [INSPIRE].
[9] V. Rosenhaus and M. Smolkin, Entanglement entropy for relevant and geometric
perturbations, JHEP 02 (2015) 015 [arXiv:1410.6530] [INSPIRE].
[10] A. Allais and M. Mezei, Some results on the shape dependence of entanglement and Renyi
entropies, Phys. Rev. D 91 (2015) 046002 [arXiv:1407.7249] [INSPIRE].
[11] M. Mezei, Entanglement entropy across a deformed sphere, Phys. Rev. D 91 (2015) 045038
[12] T. Faulkner, Bulk emergence and the RG ow of entanglement entropy, JHEP 05 (2015) 033
[13] H. Casini and M. Huerta, A
nite entanglement entropy and the ctheorem, Phys. Lett. B
600 (2004) 142 [hepth/0405111] [INSPIRE].
Rev. D 85 (2012) 125016 [arXiv:1202.5650] [INSPIRE].
[arXiv:0804.2182] [INSPIRE].
[14] H. Casini and M. Huerta, On the RG running of the entanglement entropy of a circle, Phys.
[15] H. Casini, Relative entropy and the Bekenstein bound, Class. Quant. Grav. 25 (2008) 205021
[16] A.C. Wall, A proof of the generalized second law for rapidly changing elds and arbitrary
horizon slices, Phys. Rev. D 85 (2012) 104049 [arXiv:1105.3445] [INSPIRE].
[17] N. Lashkari, C. Rabideau, P. SabellaGarnier and M. Van Raamsdonk, Inviolable energy
conditions from entanglement inequalities, JHEP 06 (2015) 067 [arXiv:1412.3514]
energy theorems from information inequalities, arXiv:1605.01075 [INSPIRE].
anomaly and causality, JHEP 07 (2016) 121 [arXiv:1605.02553] [INSPIRE].
Poincare, Phys. Theor. 63 (1995) 331.
[22] D.D. Blanco and H. Casini, Localization of negative energy and the Bekenstein bound, Phys.
Rev. Lett. 111 (2013) 221601 [arXiv:1309.1121] [INSPIRE].
[23] A. Borde, Geodesic focusing, energy conditions and singularities, Class. Quant. Grav. 4
(1987) 343 [INSPIRE].
[INSPIRE].
(1993) 1486 [Erratum ibid. 75 (1995) 1872] [grqc/9305017] [INSPIRE].
[26] C.J. Fewster, Lectures on quantum energy inequalities, arXiv:1208.5399 [INSPIRE].
[27] R. Bousso, Z. Fisher, S. Leichenauer and A.C. Wall, Quantum focusing conjecture, Phys.
Rev. D 93 (2016) 064044 [arXiv:1506.02669] [INSPIRE].
[28] R. Bousso, Z. Fisher, J. Koeller, S. Leichenauer and A.C. Wall, Proof of the quantum null
energy condition, Phys. Rev. D 93 (2016) 024017 [arXiv:1509.02542] [INSPIRE].
[29] J. Koeller and S. Leichenauer, Holographic proof of the quantum null energy condition, Phys.
Rev. D 94 (2016) 024026 [arXiv:1512.06109] [INSPIRE].
[30] G. Klinkhammer, Averaged energy conditions for free scalar elds in at spacetimes, Phys.
Rev. D 43 (1991) 2542 [INSPIRE].
[31] L.H. Ford and T.A. Roman, Averaged energy conditions and quantum inequalities, Phys.
Rev. D 51 (1995) 4277 [grqc/9410043] [INSPIRE].
Phys. Rev. D 46 (1992) 2726 [INSPIRE].
[32] A. Folacci, Averaged null energy condition for electromagnetism in Minkowski spacetime,
[33] R. Verch, The averaged null energy condition for general quantum
eld theories in
twodimensions, J. Math. Phys. 41 (2000) 206 [mathph/9904036] [INSPIRE].
[34] W.R. Kelly and A.C. Wall, Holographic proof of the averaged null energy condition, Phys.
Rev. D 90 (2014) 106003 [arXiv:1408.3566] [INSPIRE].
[35] D.M. Hofman, Higher derivative gravity, causality and positivity of energy in a UV complete
QFT, Nucl. Phys. B 823 (2009) 174 [arXiv:0907.1625] [INSPIRE].
[36] E.E. Flanagan and R.M. Wald, Does back reaction enforce the averaged null energy condition
in semiclassical gravity?, Phys. Rev. D 54 (1996) 6233 [grqc/9602052] [INSPIRE].
[37] N. Graham and K.D. Olum, Achronal averaged null energy condition, Phys. Rev. D 76
(2007) 064001 [arXiv:0705.3193] [INSPIRE].
[38] E.A. Kontou and K.D. Olum, Proof of the averaged null energy condition in a classical
curved spacetime using a nullprojected quantum inequality, Phys. Rev. D 92 (2015) 124009
[arXiv:1507.00297] [INSPIRE].
[39] D.M. Hofman and J. Maldacena, Conformal collider physics: energy and charge correlations,
JHEP 05 (2008) 012 [arXiv:0803.1467] [INSPIRE].
106 (2011) 011601 [arXiv:1007.0553] [INSPIRE].
deep inelastic scattering, arXiv:1601.05453 [INSPIRE].
[40] M. Kulaxizi and A. Parnachev, Energy ux positivity and unitarity in CFTs, Phys. Rev. Lett.
[41] Z. Komargodski, M. Kulaxizi, A. Parnachev and A. Zhiboedov, Conformal eld theories and
collider bounds, JHEP 06 (2016) 111 [arXiv:1603.03771] [INSPIRE].
[INSPIRE].
[45] X.O. Camanho and J.D. Edelstein, Causality constraints in AdS/CFT from conformal
collider physics and GaussBonnet gravity, JHEP 04 (2010) 007 [arXiv:0911.3160]
[46] A. Buchel, J. Escobedo, R.C. Myers, M.F. Paulos, A. Sinha and M. Smolkin, Holographic GB
gravity in arbitrary dimensions, JHEP 03 (2010) 111 [arXiv:0911.4257] [INSPIRE].
[47] H. Araki, Relative entropy of states of von neumann algebras, Publ. Res. Inst. Math. Sci. 11
109 (2012) 010402 [arXiv:1109.5672] [INSPIRE].
[50] A.C. Wall, Proving the achronal averaged null energy condition from the generalized second
law, Phys. Rev. D 81 (2010) 024038 [arXiv:0910.5751] [INSPIRE].
[51] N. Lashkari, Relative entropies in conformal eld theory, Phys. Rev. Lett. 113 (2014) 051602
[arXiv:1404.3216] [INSPIRE].
Lett. 117 (2016) 041601 [arXiv:1508.03506] [INSPIRE].
[52] N. Lashkari, Modular Hamiltonian for excited states in conformal eld theory, Phys. Rev.
[53] G. Sarosi and T. Ugajin, Relative entropy of excited states in two dimensional conformal eld
theories, JHEP 07 (2016) 114 [arXiv:1603.03057] [INSPIRE].
Phys. Rev. Lett. 96 (2006) 181602 [hepth/0603001] [INSPIRE].
[54] S. Ryu and T. Takayanagi, Holographic derivation of entanglement entropy from AdS/CFT,
[55] V.E. Hubeny, M. Rangamani and T. Takayanagi, A covariant holographic entanglement
entropy proposal, JHEP 07 (2007) 062 [arXiv:0705.0016] [INSPIRE].
[56] T. Faulkner, A. Lewkowycz and J. Maldacena, Quantum corrections to holographic
entanglement entropy, JHEP 11 (2013) 074 [arXiv:1307.2892] [INSPIRE].
[57] N. Engelhardt and A.C. Wall, Quantum extremal surfaces: holographic entanglement entropy
beyond the classical regime, JHEP 01 (2015) 073 [arXiv:1408.3203].
[58] X. Dong, D. Harlow and A.C. Wall, Reconstruction of bulk operators within the entanglement
[INSPIRE].
[hepth/9802150] [INSPIRE].
[59] J. Maldacena, The largeN limit of superconformal eld theories and supergravity, Int. J.
[61] S.S. Gubser, I.R. Klebanov and A.M. Polyakov, Gauge theory correlators from noncritical
string theory, Phys. Lett. B 428 (1998) 105 [hepth/9802109] [INSPIRE].
[62] W.R. Kelly, K. Kuns and D. Marolf, 't Hooft suppression and holographic entropy, JHEP 10
HJEP09(216)38
(2015) 059 [arXiv:1507.03654] [INSPIRE].
[64] F.W.J. Olver eds., NIST handbook of mathematical functions, Cambridge University Press,
Cambridge, U.K. (2010).
0406 (2004) P06002 [hepth/0405152] [INSPIRE].
[65] P. Calabrese and J.L. Cardy, Entanglement entropy and quantum eld theory, J. Stat. Mech.
[67] M. Visser, Gravitational vacuum polarization. 2: energy conditions in the Boulware vacuum,
Phys. Rev. D 54 (1996) 5116 [grqc/9604008] [INSPIRE].
[68] T. Faulkner and O. Parrikar, Entanglement entropy and shape perturbation theory, in
[69] W. Bunting, Z. Fu and D. Marolf, A coarsegrained generalized second law for holographic
conformal eld theories, Class. Quant. Grav. 33 (2016) 055008 [arXiv:1509.00074]
[INSPIRE].
[70] J. Bhattacharya, V.E. Hubeny, M. Rangamani and T. Takayanagi, Entanglement density and
gravitational thermodynamics, Phys. Rev. D 91 (2015) 106009 [arXiv:1412.5472] [INSPIRE].
[71] S. Balakrishnan and T. Faulkner, Entanglement density via null energy correlators and
[72] H. Casini, M. Huerta, R.C. Myers and A. Yale, Mutual information and the Ftheorem,
gravitational shockwaves, in preparation.
JHEP 10 (2015) 003 [arXiv:1506.06195] [INSPIRE].
freedom, JHEP 04 (2013) 062 [arXiv:1202.2070] [INSPIRE].
[73] H. Liu and M. Mezei, A re nement of entanglement entropy and the number of degrees of
[74] T. Grover, A.M. Turner and A. Vishwanath, Entanglement entropy of gapped phases and
topological order in three dimensions, Phys. Rev. B 84 (2011) 195120 [arXiv:1108.4038]
[75] K. Ohmori and Y. Tachikawa, Physics at the entangling surface, J. Stat. Mech. 1504 (2015)
[76] H. Casini, M. Huerta and J.A. Rosabal, Remarks on entanglement entropy for gauge elds,
Phys. Rev. D 89 (2014) 085012 [arXiv:1312.1183] [INSPIRE].
Rev. Lett. 114 (2015) 111603 [arXiv:1412.1895] [INSPIRE].
[77] W. Donnelly and A.C. Wall, Entanglement entropy of electromagnetic edge modes, Phys.
arXiv:1509.04325 [INSPIRE].
[arXiv:1509.04288] [INSPIRE].
gravitational anomalies, JHEP 07 (2014) 114 [arXiv:1405.2792] [INSPIRE].
HJEP09(216)38
[1] J.J. Bisognano and E.H. Wichmann , On the duality condition for quantum elds , J. Math.
[2] H. Casini , M. Huerta and R.C. Myers , Towards a derivation of holographic entanglement [3 ] R. Bousso , H. Casini , Z. Fisher and J. Maldacena , Proof of a quantum Bousso bound , Phys.
[4] R. Bousso , H. Casini , Z. Fisher and J. Maldacena, Entropy on a null surface for interacting [18] N. Lashkari , J. Lin , H. Ooguri , B. Stoica and M. Van Raamsdonk , Gravitational positive [19] S. Banerjee , A. Bhattacharyya , A. Kaviraj , K. Sen and A. Sinha , Constraining gravity using entanglement in AdS/CFT , JHEP 05 ( 2014 ) 029 [arXiv: 1401 .5089] [INSPIRE].
[20] A. Bhattacharyya , L. Cheng and L. Y. Hung , Relative entropy, mixed gaugegravitational [21] H. Borchers , On the use of modular groups in quantum eld theory ., Ann. Inst. Henri [24] F.J. Tipler , Energy conditions and spacetime singularities , Phys. Rev. D 17 ( 1978 ) 2521 [25] J.L. Friedman , K. Schleich and D.M. Witt , Topological censorship, Phys. Rev. Lett . 71 [42] D.M. Hofman , D. Li , D. Meltzer , D. Poland and F. RejonBarrera , A proof of the conformal [43] T. Hartman , S. Jain and S. Kundu , A new spin on causality constraints , arXiv:1601 .07904 [44] T. Hartman , S. Jain and S. Kundu , Causality constraints in conformal eld theory , JHEP 05 [48] R. Haag , Local quantum physics: elds, particles, algebras, Springer, Germany ( 2012 ).
[49] S. Banerjee , WessZumino consistency condition for entanglement entropy , Phys. Rev. Lett.
wedge in gaugegravity duality , Phys. Rev. Lett . 117 ( 2016 ) 021601 [arXiv: 1601 .05416] [60] E. Witten , Antide Sitter space and holography, Adv. Theor. Math. Phys. 2 ( 1998 ) 253 [63] NIST Digital Library of Mathematical Functions , http://dlmf.nist.gov/, release 1.0 . 10 ( 2015 ).
[66] G. Hooft , On the quantum structure of a black hole , Nucl. Phys. B 256 ( 1985 ) 727 .
[78] W. Donnelly and A.C. Wall , Geometric entropy and edge modes of the electromagnetic eld , [79] A. Castro , S. Detournay , N. Iqbal and E. Perlmutter , Holographic entanglement entropy and [80] N. Iqbal and A.C. Wall , Anomalies of the entanglement entropy in chiral theories , [81] T. Nishioka and A. Yarom , Anomalies and entanglement entropy , JHEP 03 ( 2016 ) 077 [82] T.L. Hughes , R.G. Leigh , O. Parrikar and S.T. Ramamurthy , Entanglement entropy and