Superconformal blocks for SCFTs with eight supercharges
Field Theories in
Superconformal blocks for SCFTs with eight
Nikolay Bobev 0 1 3
Edoardo Lauria 0 1 3
Dalimil Mazáč 0 1 2
Gauge Theory 0 1
0 31 Caroline Street North , ON N2L 2Y5 , Canada
1 Celestijnenlaan 200D , B3001 Leuven , Belgium
2 Perimeter Institute for Theoretical Physics
3 Instituut voor Theoretische Fysica, KU Leuven
We show how to treat the superconformal algebras with eight Poincaré supercharges in a unified manner for spacetime dimension 2 < d suited for analyzing the quadratic Casimir operator of the superconformal algebra and its use in deriving superconformal blocks. We illustrate this by an explicit construction of the superconformal blocks, for any value of the spacetime dimension, for external protected scalar operators which are the lowest component of flavor current multiplets.
Conformal and W Symmetry; Conformal Field Theory; Supersymmetric

2.1
2.2
3.1
3.2
3.3
3.4
4.1
4.2
4.3
1 Introduction 2 Superconformal algebras with eight supercharges
Setup
Lowering the dimension
3
The superconformal Casimir equations
4 Superconformal blocks for moment map operators
The fourpoint function of moment map operators
The superconformal Casimir operator
The h'+'+' ' i Casimir equation
The h'+' '+' i Casimir equation
The general logic of our derivation
Results
5
Discussion
A Free hypermultiplet check
A.1 F = 0 channel
A.2 F = 1 channel
A.3 F = 2 channel
1
Introduction
The conformal bootstrap programme has evolved from a general set of constraints that
consistent CFTs should obey [1–7] to an increasingly important tool to obtain quantitative
information about strongly interacting CFTs [8, 9].1 The basic idea is to implement the
constraints imposed by unitarity and crossing symmetry on the fourpoint functions in the
CFT, to constrain the spectrum of local operators and OPE coefficients of the theory. To
this end, one expands a fourpoint function in terms of the conformal blocks, which are
functions capturing the contribution of a given conformal family exchanged in a fixed OPE
channel. It is thus clear that conformal blocks are an important technical ingredient for
the success of the conformal bootstrap programme. Despite the fact that conformal blocks
are kinematical quantities, i.e. their functional form is entirely determined by the conformal
1See also [10–12] for a review and a more comprehensive list of references.
the conformal block and one can treat (at least formally) CFTs in noninteger dimensions.
The explicit form of these scalar conformal blocks are known in two, four and six dimensions
while for other values of d one needs to resort to a series expansion, see for example [14–18].
It is reasonable to expect that supersymmetry constrains further the space of consistent
CFTs and it is thus natural to apply the conformal bootstrap methods to supersymmetric
CFTs. This program has had a lot of success recently with a plethora of explicit quantitative
analytical and numerical results for SCFTs in various dimensions, see for example [19–40].
A prerequisite for these bootstrap studies is the explicit construction of the socalled
superconformal blocks, i.e. the analog of conformal blocks for SCFTs. The goal of our work
is to address the construction of superconformal blocks for theories with eight Poincaré
supercharges2 in general spacetime dimension in the range 2 < d
6. Our choice for the
upper bound on d can be attributed to the fact that there are no superconformal algebras in
more than six dimensions [41]. As explained in section 2, the reason to restrict to theories
in more than two dimensions is more technical and is related to the existence of a family
of superconformal algebras with eight supercharges in twodimensions.
The main motivation in constructing explicitly superconformal blocks for SCFTs with
eight supercharges is to understand the dynamics of these theories using conformal
bootstrap methods. SCFTs with eight supercharges posses a rich mathematical structure and
have proven to be very useful theoretical laboratories for understanding conformal field
theories and RG flows. Such theories arise naturally in string and Mtheory through various
brane and geometric constructions. In particular SCFTs with eight (or more) supercharges
provide the only known examples of unitary interacting CFTs in more than four
spacetime dimensions, see for example [42, 43]. In addition, SCFTs with eight supercharges in
d = 3; 5; 6 are isolated [44, 45] (see also [46]), i.e. they do not posses exactly marginal
supersymmetric deformations. This fact makes these SCFTs particularly amenable to analysis
using algebraic techniques like the conformal bootstrap.
SCFTs with eight supercharges necessarily posses at least an SU(2) Rsymmetry group.
In addition to that, almost all known examples of these theories have a continuous flavor
symmetry group.3 The conserved current associated to this flavor symmetry belongs to
a short superconformal multiplet, the lowest component of which is a real scalar operator
transforming in the spin1 representation of SU(2)R and in the adjoint representation of the
flavor group, see for example [45, 47]. In a slight abuse of notation, we will refer to these
scalar operators as “moment map” operators. The fourpoint function of these operators is
the main object of interest in this paper. In particular, we show how to expand this
fourpoint function into superconformal blocks, which we explicitly compute in any spacetime
dimension in the range 2 < d
6. To achieve this, we need a unified language to discuss
superconformal algebras with eight supercharges and their representations. A convenient
way to approach this is to start with the (1; 0) superconformal algebra in six dimensions,
which has an SU(2)R Rsymmetry, and then obtain the lowerdimensional superconformal
algebras as a formal dimensional reduction. One then finds the following familiar list of
2The closure of the superconformal algebra implies that these theories also posses eight conformal
supercharges.
3We call all global symmetries that commute with the supercharges of the SCFT flavor symmetries.
– 2 –
Rsymmetry groups in integer dimensions.
from the rotation group in the “transverse” 2 and 3 dimensions respectively.4 We can use
this pattern as a suggestive hint and formulate, at least formally, the superconformal algebra
with eight supercharges in any value of the spacetime dimension. This approach is similar
to the one employed in [48] for superconformal algebras with four supercharges. Using
this formal construction, we can easily study the quadratic Casimir of the superconformal
algebra for any value of d. This operator is of particular importance for superconformal
blocks since under certain conditions, these are eigenfunctions of the quadratic
superconformal Casimir. For the fourpoint function of moment map operators, we are able to exploit
this fact and derive differential equations for the corresponding superconformal blocks and
demonstrate how to solve them explicitly.
This method for constructing superconformal blocks based on the quadratic
superconformal Casimir operator follows closely the approach employed in [48, 49]. We want to stress
that this is different from an explicit analysis of supersymmetric Ward identities using
superspace or other methods [50, 51]. The results and methods of [50, 51] are the ones usually
employed in the literature on superconformal blocks for threedimensional N = 4 [52, 53],
N = 6 [54] and N = 8 [55] as well as fourdimensional5 N = 2 [57], N = 3 [58] and N = 4
CFTs [22, 59] SCFTs. Our approach can be viewed as a supersymmetric extension of the
work of DolanOsborn who employed the fact that conformal blocks are eigenfunctions of
the quadratic Casimir operator of the conformal algebra in nonsupersymmetric CFTs [14].
We continue our story in the next section with a discussion on superconformal algebras
with eight supercharges in general spacetime dimensions. This sets the stage for a discussion
of the quadratic superconformal Casimir operator in section 3. In section 4, we use the
quadratic Casimir operator to derive and solve differential equations in general spacetime
dimensions obeyed by superconformal blocks for a particular class of external protected
scalar operators. In section 5, we conclude with some comments and list a number of
possible avenues for further developments. In appendix A, we show how the fourpoint
function of the moment map operators in the theory of a free hypermultiplet is decomposed
explicitly in terms of our superconformal blocks.
Note added.
During the final stages of writing this manuscript, we became aware of the
recent work [60] which has some overlap with our results. In particular, the authors of [60]
derive superconformal blocks for the fourpoint function of scalar moment map operators
using a method based on the results of [51]. Our results agree with the ones in the third
version of [60] on the arXiv.6
4In d = 2 the small superconformal algebra has an SO(4) Rsymmetry which can be fully accounted for
by the rotation group in the (3; 4; 5; 6) directions. The “universal” SU(2)R is thus not present in d = 2.
5See also [56] for another method to derive superconformal blocks for fourdimensional SCFTs.
6There were some typographical errors in the expressions for the short multiplet blocks in the first two
versions of [60].
– 3 –
2.1
Setup
Superconformal algebras with eight supercharges
To study superconformal algebras with eight supercharges for general values of the
spacetime dimensions d, we follow the approach outlined in [48], where a similar problem was
addressed for superconformal algebras with four supercharges. We would like to stress that
superconformal algebras are welldefined only in integer dimensions. Therefore, many of the
formulae below should be considered as a collection of formal manipulations which reduce
to the wellknown superconformal algebras when d is an integer.
The main idea is to start from the d = 6 superconformal algebra with (1; 0)
supersymHJEP07(21)6
metry and obtain the algebras for smaller values of d by a formal dimensional reduction.
We work in the Euclidean signature and impose reality conditions consistent with
unitarity in the Lorentzian signature as usual. Our notation is such that Latin indices run over
the unreduced spacetime directions i = 1; : : : ; d, while with hatted indices we denote the
reduced directions ^i = d + 1; : : : ; 6. The bosonic generators of the superconformal algebra
include the momenta Pi, special conformal Ki and dilation D generators. In addition, we
have the rotations in the unreduced dimensions Mij as well as the rotations in the reduced
dimensions M^i^j. As emphasized in [48], it is important to formally keep the reduced
rotations M^i^j for any value of d, although there are no such generators for integer values of
d > 4. The explicit bosonic commutation relations are
[Mij ; Mkl] =
[M^i^j; Mk^^l] =
[Mij ; Pk] =
[Mij ; Kk] =
[D; Pi] =
i( ilMjk + jkMil
i( ^i^lM^jk^ + ^jk^M^i^l
ikMjl
^ik^M^j^l
jlMik) ;
^j^lM^ik^) ;
i( jkPi
i( jkKi
iPi ;
ikPj ) ;
ikKj ) ;
[D; Ki] = iKi ;
[Pi; Kj ] =
2i( ij D + Mij ) ;
with all other commutators vanishing. In addition to the generators of the conformal
algebra, there is also the omnipresent SU(2)R symmetry, denoted in red in (1.1), whose
generators are
where ( A)ab are the usual Pauli matrices. They commute with all conformal generators
and obey the following algebra
Rab = ( A)abRA ;
[RA; RB] = i"ABC RC :
The extra factors in the Rsymmetry algebra for d
4 in (1.1) can be thought of as arising
from the rotations in the reduced dimensions generated by M^i^j.
We adopt the following Hermitian conjugation rules
Dy =
D ;
RAy = RA ;
Miyj = Mij ;
M^iy^j = M^i^j ;
Piy = Ki :
(2.4)
– 4 –
(2.1)
(2.2)
(2.3)
Note that in our conventions, the action of the dilation generator D on an operator O is
SU(2)R, and spinor indices respectively. The anticommutator of these fermionic generators
takes the form
fQa ; Qb g = ab
i Pi ;
(2.5)
where i are a set of antisymmetric matrices satisfying extra conditions as discussed later.
The conformal supercharges transform in the conjugate Weyl representation, denoted with
an upper index. We also make use of the following conjugation rule
With these definitions, index contraction makes sense since both a and
correspond to
unitary representations. Using the notation ~
= ( i ) , the anticommutator of the S
i
generators takes the form
The QQK Jacobi identity determines
and hence
identity
It follows from the P KQ Jacobi identity that the
matrices should obey the following
This identity is of course obeyed if we choose i to be the usual Weyl matrices. The action
of the rotation generators on the supercharges can be written as
S
a
= (Qa )y :
fSa ; Sb
g = ab ~
i Ki :
[Ki; Qa ] = ab
i Sb ;
[Pi; Sa ] =
ab ~
i Qb :
~
i j + ~j i = 2 ij ;
[Mij ; Qa ] = (mij )
Qa ;
[Mij ; Sa ] =
(mij ) S
a :
mij =
4
i (~i j
~
j i) :
[RA; Qa ] =
[RA; Sa ] =
( A)baQb ;
( A)abSb ;
1
2
1
2
– 5 –
(2.6)
(2.7)
(2.8)
(2.9)
(2.10)
(2.11)
(2.12)
(2.13)
The P KQ Jacobi identity then leads to the relation
As pointed out earlier, the supercharges are in the doublet representation of SU(2)R and
thus obey the following relations
Employing various Jacobi identities one can then determine the following anticommutator
between the Poincaré and the conformal supercharges
fSa ; Qb g = i ab
D
4
Rab + ab(mij )
Mij :
(2.14)
satisfied in the above algebra when we let the spacetime vector indices, i; j; : : :, run from 1
to any d
6, using only the Clifford algebra (2.10). In addition, we will take the relations
in (2.10), (2.11), and (2.12) to hold also for the hatted indices, ^i; ^j; : : : which label the
reduced dimensions. This action ultimately defines the action of the extra Rsymmetry
factors in (1.1) for integer values of d.
The SQQ Jacobi identity requires a more careful treatment. To obey it, one has to
modify the SQ anticommutator relation in (2.14) by making the coefficient of the SU(2)R
Rsymmetry dimensiondependent, and include the rotations in the reduced dimensions.
The result is the following anticommutation relation
fSa ; Qb g = i ab
D
(d
2)
Rab + ab(mij )
Mij
ab(m^i^j)
M^i^j :
(2.15)
As explained in the beginning of this section, one has to take the unhatted spacetime vector
indices, i; j, to run from 1 to d, and the hatted ones, ^i; ^j, from d + 1 to 6. Note the negative
sign in front of the term involving M^i^j on the right hand side of (2.15). This ensures
the correct action of the extra Rsymmetry factors in (1.1). It can be checked that this
ddependent modification of the anticommutator in (2.14) does not spoil any of the other
Jacobi identities. To obey the SQQ Jacobi identity, the Weyl matrices have to obey the
following two quartic relations
(mij ) (mij )
~
i
i
=
d
1
2
(m^i^j) (m^i^j)
+ (
$
) =
d
3
2
(d
2)
+ (mij ) (mij )
+ (
$
) ;
(m^i^j) (m^i^j) :
(2.16)
(2.17)
Remarkably, these relations can be checked to hold for any d = 1; : : : ; 6.
We do not
know if they can be derived in a dimensionindependent language but we will assume that
they hold in the discussions below. Note however that the constant d
2 in (2.15) in
front of the original SU(2)R Rsymmetry generators, Rab, can be derived in a
dimensionindependent language by taking traces of the quartic relations, using the Clifford algebra,
and the identities i i = d, ^i^ = 6
i
We can thus conclude that using the approach summarized above, we have a unified
way to describe the superconformal algebras with eight Poincaré supercharges in any integer
dimension. In addition, these formulae can be used for other purposes, e.g. for calculations
involving the quadratic Casimir operator, for noninteger values of d.
We would like to emphasize that for d
2 the discussion above is not entirely valid
since some generators decouple from the superconfomal algebra. In particular, the SU(2)R
– 6 –
symmetry, denoted in red in (1.1), is not present and the M^i^j generators in the four reduced
dimensions produce the SO(4) Rsymmetry of the twodimensional “small” superconformal
algebra.7 Due to this subtlety, we will restrict ourselves to the range 2 < d
6 in the rest of this note.
The fourpoint function of moment map operators
In this note, we focus on the fourpoint function of the socalled moment map operators.
They are the superconformal primaries of the socalled D[0; 1] multiplet. These operators
are spacetime scalars of scaling dimension
= d
2, transforming in the vector repre
sentation of SU(2)R and in the adjoint representation of the flavor group. The notation
D[0; 1] refers to ` = 0, R = 1 of the lowest component. Upon acting on the superconformal
primary with two Q supercharges, one obtains a flavor current.
We will denote the superconformal primaries by 'A, where A = 1; 2; 3 is the SU(2)R
vector index. Since the flavor group commutes with the superconformal generators, it does
not play a role in the construction of superconformal blocks and we will supress the adjoint
flavor indices. The D[0; 1] multiplet in general d is the dimensional reduction of the D[0; 1]
multiplet in d = 6. As a consequence of this fact, 'A are neutral under the SO(6
d)
Rsymmetry coming from rotations in the reduced dimensions.
Conformal symmetry implies that the fourpoint function of moment map operators
takes the following form
h
'A(x1)'B(x2)'C (x3)'D(x4)i =
1
(jx12jjx34j)2(d 2) F
where z and z are defined by
zz
x212x2324 ;
x123x24
(1
z)(1
z)
x214x2223 :
x123x24
The SU(2)R symmetry ensures that the operators exchanged in the schannel OPE
must transform in either R = 0, R = 1, or R = 2 representations of SU(2)R. The function
ABCD(z; z) can be decomposed accordingly as
F
where Y ABCD are the SU(2) eigentensors, taking the following form
7We are not able to incorporate the D(2; 1; ) family of “large” superconformal algebras in our formalism.
The schannel OPE leads to the following decomposition of each F
where the sum runs over conformal primary operators transforming as the symmetric
traceless tensors of SO(d), and g ;`(z; z) are the corresponding conformal blocks. Note that all
of the primaries appearing in this OPE expansion transform trivially under the SO(6
d)
Rsymmetry since the same holds true for the moment maps.
Superconformal symmetry further relates the coefficients c2''P in (3.5) of different
conformal primaries from the same superconformal multiplet. This means that F
can be expanded in terms of the superconformal blocks GO
ABCD(z; z)
F
where the sum runs over superconformal primaries. Each superconformal block is a sum
over Rsymmetry components as follows
GO
terms of the coefficient of its corresponding superconformal primary. This implies that the
ABCD(z; z) are fully fixed by the superconformal symmetry. We will be able to
functions GO
use the superconformal Casimir equation to find them in a closed form. To this end, let us
first derive the form of the superconformal Casimir operator.
3.2
The superconformal Casimir operator
The quadratic superconformal Casimir operator, C, must be a linear combination of the
quadratic Casimir, Cb, of the conformal subalgebra, the quadratic Casimir of the SO(6
d) group of “transverse” rotations, 12 M^ ij M^ ij , and Rsymmetry RARA, as well as terms
quadratic in the fermionic generators. The form of this operator is completely fixed by the
requirement that C commutes with all generators of the superconformal algebra. We find
O
2
R=0
C = Cb + [Sa ; Qa ]
(d
2)RARA +
1 ^
2
Mij M^ ij ;
Cb =
Mij Mij
D2
(PiKi + KiPi) ;
1
2
where the quadratic Casimir operator of the conformal algebra is given by
and as usual we have assumed summation over repeated indices.
Let us now consider a superconformal primary with dimension
, transforming as a
symmetric traceless tensor of spin ` under Mij (recall that by construction this primary
operator is a singlet under M^ ij ) and with SU(2)R charge R. Using the superconformal
algebra, it is easy to check that
1
2
1
2
[C; O ;`;R] = cO ;`;R ;
– 8 –
where
and
c
Cb
Cb + 4
(d
2)R(R + 1) ;
(
d) + `(` + d
2) ;
(3.11)
(3.12)
is the eigenvalue of Cb.
It was emphasized in [14] that the conformal blocks, g ;` in (3.5), are eigenfunctions
of the quadratic Casimir operator of the conformal algebra, Cb, with eigenvalue
Cb given
by (3.12). This fact was then used in [14] to derive differential equations for the
functions g ;`(z; z). The same logic can be applied to the quadratic Casimir operator of the
superconformal algebra in order to find differential equations for the superconformal blocks
in [48] and we will apply the same method for the case of eight supercharges below.
GOR(z; z). This procedure was successfully implemented for theories with four supercharges
In order to arrive at differential equations satisfied by the superconformal blocks, we
need to act with the Casimir operator, C, on the operators at positions x1 and x2 in the
fourpoint function (3.1). This action will in general mix different fourpoint functions. However,
we can get decoupled differential equations by making special choices of the external SU(2)R
indices. Let us first introduce the following convenient basis for the SU(2)R vector indices
'+
1
p ('1
2
i'2) ;
'
1
2
p ('1 + i'2) ;
'
0
'3 :
(3.13)
h
Considering the action of the superconformal Casimir operator on the h'+'+' ' i and
'+' '+' i correlators leads to two independent differential equations, discussed in the
next subsections, which allow us to fix the superconformal blocks completely.
3.3
The h'+'+' ' i Casimir equation
After specializing the external SU(2)R indices to h'+'+' ' i, we find from (3.3) that
only the R = 2 component contributes
F
++
(z; z) = 6F 2(z; z) :
Let us first understand how the various terms in the superconformal Casimir (3.8)
act on the h'+'+' ' i of the fourpoint function. The conformal Casimir operator Cb
acts as the usual nonsupersymmetric differential operator, DDO, employed by Dolan and
Osborn [14]. The action of the second term in (3.8), containing the fermionic generators,
can be simplified by the following equations
[Sa ; 'A(0)] = 0 ; [Q1 ; '+(x)] = [Q2 ; ' (x)] = 0 :
The first identity above is a consequence of the fact that 'A is a superconformal primary,
while the latter two are special cases of the BPS condition satisfied by 'A. A short
computation shows that 12 [Sa ; Qa ] then acts as a scalar multiplication by 8 ' = 8(d
The third term in (3.8), including the minus sign, acts as a scalar multiplication by
2).
(d
2)RP (RP + 1) =
6(d
2) since F 2(z; z) only receives contributions from conformal
primaries P with R
P = 2. The last term in (3.8) gives zero.
(3.14)
(3.15)
– 9 –
When we restrict F 2(z; z) to the contributions coming from a fixed superconformal
family, the Casimir must act by scalar multiplication by
C . Hence, we find the following
differential equation for the G2(z; z) component of the superconformal block
The differential operator DDO is the same as the one found in [14].
DDO
2z2(1
+ 2(d
dimension
P and spin `P .
between
, `, R,
P and `P
Since equation (3.16) takes the form of the usual differential equation satisfied by
nonsupersymmetric conformal blocks, we can conclude that any nonzero solution is a single
conformal block corresponding to a conformal primary P with Rcharge R
P = 2, conformal
Since P must be a symmetric traceless tensor, (3.16) imposes the following constraint
P
( P
d)+`P (`P +d 2)+2(d 2) =
(
d+4)+`(`+d 2) (d 2)R(R+1): (3.18)
We proceed in the next subsection, where we obtain a differential equation involving also
the G0 and G1 components of the superconformal blocks.
3.4
The h'+' '+' i Casimir equation
Let us now turn our attention to the h'+' '+' i component of the superconformal
Casimir equation. First, it follows from (3.3) that
F
+ + (z; z) = F 0(z; z) + F 1(z; z) + F 2(z; z) :
(3.19)
Next, we have to analyze how the various terms in the superconformal Casimir (3.8) act
in this case. As previously, the first term of (3.8) acts as the DolanOsborn differential
operator DDO. With some work, following the same logic as detailed in section 3.2 of [48],
the second term in (3.8) can be written as a differential operator
[Sa ; Qa ] 7!
DSQ
4 z(1
(3.20)
1
2
2
R=0
The third term in (3.8) becomes a multiplication by the SU(2)R Casimir eigenvalue times
(d 2), and the last term vanishes. Hence, the final equation obeyed by the superconformal
blocks in this channel takes the form
(DDO + DSQ) X
GR(z; z) = (d
2)[2G1(z; z) + 6G2(z; z)] + C
GR(z; z) ;
(3.21)
with C defined in (3.11).
One may wonder about the meaning of the remaining Casimir equation involving only
A =
components, namely the equation for the h'+' ' '+i correlator. It can be
2
X
R=0
conformal primaries P with uniform (
1
)RP +`P , i.e.
obtained from the equation for h'+' '+' i by simultaneous application of (
1
)R and
the swap of coordinates x1 $ x2. Consequently, a solution of equation (3.21) will solve
the equation following from the h'+' ' '+i correlator if the expansion only involves
(
1
)RP +`P = (
1
)RO+`O ;
(3.22)
where O is the superconformal primary. In fact, this “parity” constraint (3.22) will be
fundamental to fix the parameters in our superconformal blocks. Note that the same “parity”
constraint for d = 4 was discussed in section 3.1.1 of [57].
Superconformal blocks for moment map operators
In order to derive the superconformal blocks for moment map fourpoint functions, it is
instructive to collect some wellknown facts about the structure of unitary multiplets of
the superconformal algebras with eight supercharges for 2 < d
6. The unitary repre
sentations of superconformal algebras with extended supersymmetry have been studied by
many authors beginning with the pioneering work in [61–63]. For the results summarized
below, we have also made use of the more recent work in [44, 45, 47, 50, 64, 65]. A cursory
look at these references makes it clear that the structure of superconformal multiplets of
SCFTs with eight supercharges depends heavily on the dimension of spacetime and on the
Rsymmetry groups summarized in (1.1). Thus one may worry that our attempt to derive
the superconformal blocks in a dimensionindependent way is bound to fail. Fortunately,
as we summarize below, there is a way around this apparent impasse.
The key observation is that the structure of unitary superconformal multiplets that
can in principle appear in the OPE of two moment map operators is fairly uniform across
spacetime dimensions. In particular, it was shown in [50, 51, 57] that whenever a conformal
primary appears in the OPE of two moment map operators, then also its corresponding
superconformal primary appears. It follows that the superconformal primary is a symmetric
traceless tensor of SO(d). Moreover, the moment maps are neutral under the SO(6
d)
Rsymmetry coming from rotations in the reduced dimensions, and so any operator appearing
in their OPE is also neutral under these transformations. The list of unitary superconformal
multiplets satisfying the above properties follows. With
, ` and R we denote the scaling
dimension, spin and Rcharge of the superconformal primary respectively.
4
d
6.
L
(4.1)
2 < d
[ ; `; R];
The first lines in (4.1) and (4.2) correspond to long unitary multiplets. The second
lines correspond to the short multiplet that emerges when the long multiplet reaches the
unitarity bound. We call these regular short multiplets. The remaining lines correspond to
isolated short multiplets. Note that the Btype multiplet is isolated for d > 4 but becomes
regular in d
4. For R = ` = 0, this multiplet contains the Rsymmetry current and the
energymomentum tensor.
A few comments are in order. The expressions in (4.1) and (4.2) have been derived
rigorously in the respective integer dimensions but we have written them in a suggestive way
such that the dimension, d, appears as a parameter. One should note that for d = 6, there
are the somewhat special C[0; R] multiplets which are due to the presence of selfdual tensor
in six dimensions. In particular, C[0; 0] is the (1; 0) free tensor multiplet. This multiplet
and many others on the above list are in fact ruled out from appearing in the moment map
OPE by our Casimir equation, as explained in the following section.
4.2
The general logic of our derivation
Crucially, not all multiplets listed in the previous subsection actually appear in the OPE of
two moment map operators. For example, we can clearly restrict the Rcharge to R = 0; 1; 2.
In addition, superconformal Ward identities further restrict the allowed set. In fact, our
superconformal Casimir equations are powerful enough to sidestep the use of superconformal
Ward identities. Indeed, the equations in (3.16) and (3.21) admit nonzero solutions only
for the multiplets allowed by the Ward identities. Moreover, in the cases when a solution
exists, it is unique and thus equal to the sought superconformal block.
G
R is a finite linear combination of ordinary conformal blocks
Let us spell out our procedure in more detail. To determine a superconformal block
means to find the functions G0, G
1 and G2 for each allowed superconformal family. Each
X
n;m2Z
GOR(z; z) =
fnR;mg O+n;`O+m(z; z) ;
where O is the superconformal primary. Each element of the list of unitary multiplets with
a symmetric traceless superconformal primary, presented in section 4.1, provides an Ansatz
for the set of conformal primaries that appear on the r.h.s. of (4.3). We can then apply the
superconformal Casimir equations in (3.16) and (3.21) and fix the undetermined coefficients
fn;m. More specifically, we use the following representation of ordinary conformal blocks as
a power series in s = pzz, discussed in [18] (see in particular equation (2.24) of [18]).
g ;`(z; z) =
1
n=0
X hn( )s +n ;
(4.3)
(4.4)
where C ( ) are the Gegenbauer polynomials as used in [18]. We can then attempt to solve
the superconformal Casimir equations order by order in s. Sometimes, the only allowed
solution vanishes identically, meaning that the corresponding superconformal multiplet is
not allowed to appear in the OPE by the superconformal Ward identities. The following
subsection lists the nonzero solutions of the Casimir equation, corresponding to all allowed
= (z + z)=2jzj and hn( ) can be determined recursively, starting from the initial
After a careful study of the superconformal Casimir equations in (3.16) and (3.21), we
find that the set of multiplets that can appear in our OPE is completely uniform across
dimensions (with 2 < d
6). The long multiplets L[ ; `; R] are allowed to appear if and
only if R = 0. In addition, there are certain types of short multiplets, namely B[`; 0],
B[`; 1], D[0; 0], D[0; 1], and D[0; 2]. Let us now present the superconformal blocks for these
multiplets. We recall that g ;l denotes the ordinary conformal block with normalization
specified by equations (4.4), (4.5).
The L[ ; `; 0] multiplet. The structure of the L[ ; `; 0] multiplet leads to the following
where
condition
multiplets.
After using the superconformal Casimir equations (3.16) and (3.21) and the expansion for
conformal blocks discussed around (4.4), one can find the explicit form of the coefficient
For fn0;m, we find
2)(
d
` + 5)(
d
d
2 2d(4`+5) 8(` 3)(`+1) +
+ (d
2)
d`2
(d
2)d` + 2(d
(
d + 5)(
For fn1;m, the result is
Finally, for f22;0 we have
For unitary SCFTs, the coefficients fnR;m in the expansion (4.3) of superconformal blocks
in terms of the ordinary blocks have to be positive real numbers. This is simply due to the
fact that these coefficients are related to the square of certain OPE coefficients. Using the
unitarity bounds for the L[ ; `; 0] multiplet presented in (4.1) and (4.2), one can show that
indeed all coefficients in (4.7), (4.8), and (4.9) are positive real numbers. This constitutes a
nontrivial consistency check of our results. When d = 4, we can compare our results with
a perfect agreement with the results presented in the appendix B of their paper.8
the discussion on superconformal blocks in fourdimensional N = 2 SCFTs in [57]. We find
In addition to that, we observe that when d = 4, the coefficients fnR;m simplify
dramatically and one finds the following curious relation between the superconformal blocks and
the ordinary nonsupersymmetric conformal blocks
G ;`
+ + (z; z) = G0 ;`(z; z) + G1 ;`(z; z) + G2 ;`(z; z) = (zz) 1g +2;`(z; z) :
(4.10)
This type of relation between superconformal blocks and nonsupersymmetric conformal
blocks with shifted arguments exists also for SCFTs with four supercharges for any value
of d, as pointed out in [48]. For theories with eight supercharges, the relation (4.10) holds
only for d = 4. It will be curious to understand better the reason for the existence of this
type of relations.
B[`; R] multiplets.
Due to SU(2)R selection rules and the superconformal Casimir
equations (3.16) and (3.21), the B[`; R] multiplets can appear in the superconformal block
expansion only for R = 0 and R = 1.
8In comparing the two sets of results, one should note that our ordinary blocks g ;` are related to the
ordinary blocks G(l) of reference [57] by G(l) = `2+`1 g ;`.
For the type B[`; 0] short multiplet, one has
= l + d
2 and the following Ansatz for the superconformal blocks
G`0 = gl+d 2;` + f20;2gl+d;`+2 ;
G`1 = f11;1gl+d 1;`+1 ;
G`2 = 0 :
The superconformal Casimir equations determine uniquely the coefficients above
As is familiar by now, the superconformal Casimir equations determines all coefficients
in (4.13)
superconformal blocks
D[`; R] multiplets.
Due to the SU(2)R selection rules and the superconformal Casimir
equations (3.16) and (3.21), the D[`; R] multiplets can appear in the superconformal block
expansion only for R = 0, R = 1 and R = 2. The R = 0 multiplet contains only the identity
For the type D[0; 1] multiplet, we have
= d
2 and the following Ansatz for the
G
G
G
0 = f10;1gd 1;1 ;
1 = gd 2;0 ;
2 = 0 :
(4.11)
(4.12)
(4.13)
(4.14)
(4.15)
(4.16)
(4.17)
(4.18)
For the type D[0; 2] multiplet, we have
superconformal blocks
= 2d
This completes the derivation of the superconformal blocks for the fourpoint function
of moment map operators. Since the derivation and the final result are quite lengthy, it is
important to perform some consistency checks. When d = 4, superconformal blocks for long
and short multiplets in fourdimensional N = 2 SCFTs were presented explicitly in [57]. Our
results above agree with the ones in [57] upon setting d = 4. Another consistency check can
be made by considering the moment map operators in the theory of a free hypermultiplet.
In appendix A, we show explicitly how to decompose the fourpoint function of moment
map operators in this theory in terms of our superconformal blocks for any value of d.
5
The main focus of our work has been the explicit construction of the superconformal blocks
for external moment map operators in SCFTs with eight supercharges. To this end, we
have adopted a procedure similar to the one in [48] to treat (at least formally)
superconformal algebras and the superconformal quadratic Casimir operator in continuous dimensions
2 < d
our results.
6. There are many interesting topics for further studies that could build upon
First, it is clear that the general method for constructing superconformal blocks
outlined in this work should be applicable to other external scalar operators, most directly to
superconformal primaries of the D[0; R] multiplet with R > 1. One of the most important
open problems in the theory of superconformal blocks is the construction of the latter when
the external operators are the superconformal primaries of the multiplet containing the
stress tensor, namely B[0; 0]. Substantial progress on this question was made in [33, 66],
but a full formula for the superconformal blocks is still missing. We hope that the
superconformal Casimir operator that we derive in this note will prove useful for this problem.
Another interesting extension is to consider external unprotected scalar operators. It
has been recently pointed out that one can also make use of the cubic Casimir operator
of the superconformal algebra, in addition to the quadratic Casimir operator used in our
approach, to derive conformal blocks for external nonprotected operators [40]. It will be
interesting to explore this method for SCFTs with eight supercharges. The construction of
superconformal blocks for external operators of nonvanishing spin can also be addressed,
although we expect that the results will be significantly more involved.
It is also intriguing to understand better the structure of our superconformal blocks.
Recently, it was emphasized that there is a connection between conformal and
superconformal blocks and integrability [67–69]. This relation has not been explored for superconformal
blocks with eight supercharges and our results may shed some light on this story. One
particular curiosity that emerged from our calculations is that in d = 4, we can write the
superconformal blocks of long multiplets as ordinary nonsupersymmetric conformal blocks
with shifted arguments, see (4.10). This is reminiscent of the similar situation for SCFTs
with four supercharges where for any value of d
4, one can write the superconformal
blocks in terms of shifted nonsupersymmetric blocks [48]. It will be interesting to
understand the reasons behind this phenomenon and why this structure fails for SCFTs with
eight supercharges in d 6= 4.
The results of this paper set the stage for a numerical exploration of the space of SCFTs
with eight supercharges in various dimensions. It will be certainly desirable to study the
constraints on such theories imposed by unitarity and crossing symmetry using numerical
bootstrap methods. This has been quite successful for fourdimensional N = 2 [35, 57, 70]
as well as threedimensional N = 4 SCFTs [52]. A particular fruitful avenue for further
progress should be the study of theories in five and six dimensions with exceptional flavor
symmetry groups since these arise naturally in string and Mtheory.9 The advantage offered
by our results is that one can perform the numerical analysis for any value of the spacetime
dimension d. This has proven instructive in the analysis of SCFTs with four supercharges
via numerical bootstrap methods [48, 49].
A beautiful algebraic structure spanned by some of the protected operators in
fourdimensional N = 2, sixdimensional N = (2; 0) and threedimensional N = 4 SCFTs was
uncovered in [21]. An important open question is whether there is a generalization of this
structure for fivedimensional N = 1 and sixdimensional N = (1; 0) SCFTs. We hope that
the explicit results for short and long superconformal blocks presented in this work will
shed some light on this problem.
Acknowledgments
We would like to thank Marco Baggio, Chris Beem, Sheer ElShowk, Davide Gaiotto, Friðrik
Gautason, Pedro Liendo, Gabriele TartaglinoMazzucchelli, Marco Meineri, Miguel Paulos,
Silviu Pufu, Emilio Trevisani, and Balt van Rees for useful discussions. The work of NB
and EL is supported in part by the starting grant BOF/STG/14/032 from KU Leuven,
by an Odysseus grant G0F9516N from the FWO, by the KU Leuven C1 grant ZKD1118
C16/16/005, by the Belgian Federal Science Policy Office through the InterUniversity
Attraction Pole P7/37, and by the COST Action MP1210 The String Theory Universe.
EL is additionally supported by the European Research Council grant no. ERC2013CoG
616732 HoloQosmos, as well as the FWO Odysseus grants G.001.12 and G.0.E52.14N. The
9This strategy was implemented recently in [60] for sixdimensional (1; 0) SCFTs.
research of DM was supported by Perimeter Institute for Theoretical Physics. Research at
Perimeter Institute is supported by the Government of Canada through Industry Canada
and by the Province of Ontario through the Ministry of Research and Innovation. DM is
grateful to the ITF at KU Leuven for hospitality during various stages of the long gestation
period of this project. EL and DM are grateful for the hospitality of the ICTPSAIFR São
Paulo during the completion of this work.
A
Free hypermultiplet check
In this appendix, we find the decomposition of the fourpoint function of the moment
map operators in the theory of the free hypermultiplet into our superconformal blocks in
general spacetime dimension. The fact that this is possible, and the fact that the resulting
coefficients have appropriate positivity properties is a nice consistency check of our formulae
for the superconformal blocks.
In the notation of section 4.1, the hypermultiplet is denoted as D[0; 1=2]. Its bottom
component consists of two free complex scalars in the doublet of SU(2)R. For our purposes,
it is better to think of these as four real scalars p, p = 1; : : : ; 4, thus manifesting the full
SO(4) = SU(2)R
SU(2)F symmetry group. SU(2)R is the familiar Rsymmetry, while
SU(2)F is a genuine flavor symmetry. We can organize the four real scalars in a 2
2
matrix
where 1;2;3 = i 1;2;3, p being the usual Pauli matrices, and 4 the identity matrix. The
undotted and dotted indices on
aa_ transform as doublets under SU(2)R and SU(2)F
respectively. The twopoint function of aa_ is, up to normalization,
aa_ =
p pa a_ ;
h
_
aa_ (x) bb(0)i =
ab a_ b_
jxj2 ;
(A.1)
(A.2)
(A.3)
(A.4)
R0(u; v),
where
= (d
2)=2.
take the form
We would like to study the moment map operators for the flavor symmetry SU(2)F ,
denoted 'AA_ . The capital undotted and dotted indices transform in the adjoint
representation of SU(2)R and SU(2)F respectively. Up to normalization, the moment map operators
'AA_ =
ab a_ b_
A A_ aa_ bb_ ;
'AA_ can be computed using Wick contractions and by the virtue of the SU(2)R
where
aAb = ac( A)cb with ( A)cb the usual Pauli matrices. The fourpoint function of
SU(2)F
admits the decomposition
h
'AA_ (x1)'BB_ (x2)'CC_ (x3)'DD_ (x4)i =
YRABCDYFA_B_ C_ D_ F RF (u; v) ;
1
(jx12jjx34j)4
2
X
R;F =0
F
R1(u; v), F
with Y ABCD defined in (3.4). Here F stands for the charge under the Cartan of SU(2)F .
R
Since SU(2)F does not mix with the superconformal symmetry, the functions F
R2(u; v) should each admit a decomposition into our superconformal blocks.
We now turn to finding this decomposition for each of these functions.
In the normalization where the identity contributes as 1, the functions F
R0(u; v) take the
u +
+ u +
1
6
1
9
2
v
u
2
v
u
2
:
;
+ u2 +
;
u
2
v
To find the decomposition of this collection of functions into superconformal blocks means
to find an expansion of the following form
with
2
R=0
X YRABCDF R0(u; v) =
where GOR(u; v) were presented in section 4. We find the following expansion
the superconformal block expansion (A.6). We find the following explicit formulae
where the coefficients , ,
multiplying each multiplet stand for the c2''O
coefficients in
`
(0) =
`
(0) =
n(0;`) =
2 `=2
n2 ` +
n2 ` +1
(
1
) n2 + 2` (n=2 +
2 2` 1 `+1 + 2
2
In the above and in the following, we use the notation
Note that all coefficients are positive, as required by unitarity in our conventions.
(a)b
(a + b)
(a)
:
(A.5)
(A.6)
(A.7)
(A.8)
(A.9)
(A.10)
In the normalization where the identity contributes as 1, the functions F
R1(u; v) take the
F
F
F
01(u; v) =
11(u; v) =
21(u; v) =
u
v
u
v
1
12
u +
+ u +
1
6
1
4
u
2
v
:
u
2
;
+ u2
This collection of functions can be decomposed into our superconformal blocks with the
2D[0; 1] +
X
+ `+1
2 `=2
(2 )`2(2 )`+1
4`!( )`+1(` + 4 )`
:
;
(` + ) (` + 2 ) (n + 4 + 1) ( `=2 + n=2 + )
( `=2 + n=2 + 2 )
4 (` + 1) ( ) (2 )3 ( `=2 + n=2 + 1) (n + 3 + 1) ( ` + n + 2 )
2
2
(`=2 + n=2 + 2 )
(`=2 + n=2 + 3 )
(`=2 + n=2 +
+ 1) (` + n + 4 )
:
Again, all coefficients are positive as they should be.
A.3
F = 2 channel
following form
In the normalization where the identity contributes as 1, the functions F
R2(u; v) take the
F
F
F
02(u; v) =
12(u; v) =
22(u; v) =
1
18
1
12
1
36
v
v
+ u2
u
2
;
2
+ u2 + 4
u
2
v
u
2
v
;
:
following result
where
`
(
1
) =
`
(
1
) =
n(1;`) =
This collection of functions can be decomposed into our superconformal blocks with the
following result
1
6 D[0; 2] +
X
`(2)B[`; 1] +
X
n(2;`)L[4 + n; `; 0] ;
(A.16)
(A.17)
(n + 4 )
2
n ` + 2
2
n+` + 3
(3 + n + 1) (n
` + 2 ) (n + ` + 4 )
n2 +
n2 + 2
+ 1 !
:
HJEP07(21)6
where
`
(` + )
4 ` `+1 + 2
2
2
2
n ` +
+ n+` + 1
2
(
1
) 2` + n2 +1
2
2` + 1
`+1
2
2
2` +
( )2
1
A
Once again, all coefficients are positive as they should be.
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.
Fiz. 66 (1974) 23 [INSPIRE].
[2] S. Ferrara, R. Gatto and A.F. Grillo, Properties of Partial Wave Amplitudes in Conformal
Invariant Field Theories, Nuovo Cim. A 26 (1975) 226 [INSPIRE].
[3] S. Ferrara, R. Gatto and A.F. Grillo, Positivity Restrictions on Anomalous Dimensions,
Phys. Rev. D9 (1974) 3564 [INSPIRE].
[4] S. Ferrara, A.F. Grillo and R. Gatto, Tensor representations of conformal algebra and
conformally covariant operator product expansion, Annals Phys. 76 (1973) 161 [INSPIRE].
[5] S. Ferrara, A.F. Grillo and R. Gatto, Manifestly conformal covariant operatorproduct
expansion, Lett. Nuovo Cim. 2S2 (1971) 1363 [INSPIRE].
[6] S. Ferrara, A.F. Grillo, R. Gatto and G. Parisi, Analyticity properties and asymptotic
expansions of conformal covariant green’s functions, Nuovo Cim. A 19 (1974) 667 [INSPIRE].
[7] S. Ferrara, A.F. Grillo, G. Parisi and R. Gatto, Covariant expansion of the conformal
fourpoint function, Nucl. Phys. B 49 (1972) 77 [Erratum ibid. B 53 (1973) 643] [INSPIRE].
[8] R. Rattazzi, V.S. Rychkov, E. Tonni and A. Vichi, Bounding scalar operator dimensions in
4D CFT, JHEP 12 (2008) 031 [arXiv:0807.0004] [INSPIRE].
[9] S. ElShowk, M.F. Paulos, D. Poland, S. Rychkov, D. SimmonsDuffin and A. Vichi, Solving
the 3D Ising Model with the Conformal Bootstrap, Phys. Rev. D 86 (2012) 025022
[arXiv:1203.6064] [INSPIRE].
[10] S. Rychkov, EPFL Lectures on Conformal Field Theory in D >= 3 Dimensions,
SpringerBriefs in Physics (2016). [arXiv:1601.05000].
[11] D. Poland and D. SimmonsDuffin, The conformal bootstrap, Nature Phys. 12 (2016) 535.
[12] D. SimmonsDuffin, The Conformal Bootstrap, arXiv:1602.07982 [INSPIRE].
[13] F.A. Dolan and H. Osborn, Conformal four point functions and the operator product
expansion, Nucl. Phys. B 599 (2001) 459 [hepth/0011040] [INSPIRE].
[14] F.A. Dolan and H. Osborn, Conformal partial waves and the operator product expansion,
Nucl. Phys. B 678 (2004) 491 [hepth/0309180] [INSPIRE].
JHEP 09 (2016) 070 [arXiv:1509.00428] [INSPIRE].
[15] J. Penedones, E. Trevisani and M. Yamazaki, Recursion Relations for Conformal Blocks,
conformal blocks, JHEP 07 (2016) 057 [arXiv:1603.05552] [INSPIRE].
[17] F.A. Dolan and H. Osborn, Conformal Partial Waves: Further Mathematical Results,
arXiv:1108.6194 [INSPIRE].
(2013) 106004 [arXiv:1303.1111] [INSPIRE].
[18] M. Hogervorst and S. Rychkov, Radial Coordinates for Conformal Blocks, Phys. Rev. D 87
[19] D. Poland and D. SimmonsDuffin, Bounds on 4D Conformal and Superconformal Field
Theories, JHEP 05 (2011) 017 [arXiv:1009.2087] [INSPIRE].
[20] D. Bashkirov, Bootstrapping the N = 1 SCFT in three dimensions, arXiv:1310.8255
[INSPIRE].
Lett. 111 (2013) 071601 [arXiv:1304.1803] [INSPIRE].
[23] L.F. Alday and A. Bissi, The superconformal bootstrap for structure constants, JHEP 09
(2014) 144 [arXiv:1310.3757] [INSPIRE].
Covariant Approaches to Superconformal Blocks, JHEP 08 (2014) 129 [arXiv:1402.1167]
[25] M. Berkooz, R. Yacoby and A. Zait, Bounds on N = 1 superconformal theories with global
symmetries, JHEP 08 (2014) 008 [Erratum ibid. 1501 (2015) 132] [arXiv:1402.6068]
general scalar operators, JHEP 08 (2014) 049 [arXiv:1404.5300] [INSPIRE].
superconformal theories, JHEP 10 (2014) 37 [arXiv:1407.6354] [INSPIRE].
[30] S.M. Chester, S. Giombi, L.V. Iliesiu, I.R. Klebanov, S.S. Pufu and R. Yacoby, Accidental
Symmetries and the Conformal Bootstrap, JHEP 01 (2016) 110 [arXiv:1507.04424]
[31] C. Beem, M. Lemos, L. Rastelli and B.C. van Rees, The (2; 0) superconformal bootstrap,
Phys. Rev. D 93 (2016) 025016 [arXiv:1507.05637] [INSPIRE].
[32] A. Bissi and T. Łukowski, Revisiting N = 4 superconformal blocks, JHEP 02 (2016) 115
[arXiv:1508.02391] [INSPIRE].
[33] P. Liendo, I. Ramirez and J. Seo, Stresstensor OPE in N = 2 superconformal theories,
JHEP 02 (2016) 019 [arXiv:1509.00033] [INSPIRE].
[arXiv:1509.06368] [INSPIRE].
bootstrap of the K3 CFT, JHEP 05 (2017) 126 [arXiv:1511.04065] [INSPIRE].
[37] Z. Li and N. Su, The Most General 4D N = 1 Superconformal Blocks for Scalar Operators,
JHEP 05 (2016) 163 [arXiv:1602.07097] [INSPIRE].
dimensions, JHEP 05 (2017) 112 [arXiv:1610.05371] [INSPIRE].
arXiv:1702.00404 [INSPIRE].
[INSPIRE].
B 390 (1997) 169 [hepth/9609161] [INSPIRE].
JHEP 11 (2016) 135 [arXiv:1602.01217] [INSPIRE].
JHEP 11 (2016) 091 [arXiv:1606.00810] [INSPIRE].
[44] C. Cordova, T.T. Dumitrescu and K. Intriligator, Deformations of Superconformal Theories,
[45] M. Buican, J. Hayling and C. Papageorgakis, Aspects of Superconformal Multiplets in D > 4,
[46] J. Louis and S. L¨st, Supersymmetric AdS7 backgrounds in halfmaximal supergravity and
marginal operators of (1; 0) SCFTs, JHEP 10 (2015) 120 [arXiv:1506.08040] [INSPIRE].
[47] C. Cordova, T.T. Dumitrescu and K. Intriligator, Multiplets of Superconformal Symmetry in
Diverse Dimensions, arXiv:1612.00809 [INSPIRE].
[48] N. Bobev, S. ElShowk, D. Mazac and M.F. Paulos, Bootstrapping SCFTs with Four
Supercharges, JHEP 08 (2015) 142 [arXiv:1503.02081] [INSPIRE].
[49] N. Bobev, S. ElShowk, D. Mazac and M.F. Paulos, Bootstrapping the ThreeDimensional
Supersymmetric Ising Model, Phys. Rev. Lett. 115 (2015) 051601 [arXiv:1502.04124]
[50] F.A. Dolan and H. Osborn, Superconformal symmetry, correlation functions and the operator
product expansion, Nucl. Phys. B 629 (2002) 3 [hepth/0112251] [INSPIRE].
[51] F.A. Dolan, L. Gallot and E. Sokatchev, On fourpoint functions of 1=2BPS operators in
general dimensions, JHEP 09 (2004) 056 [hepth/0405180] [INSPIRE].
[52] S.M. Chester, J. Lee, S.S. Pufu and R. Yacoby, Exact Correlators of BPS Operators from the
3d Superconformal Bootstrap, JHEP 03 (2015) 130 [arXiv:1412.0334] [INSPIRE].
dimensions, JHEP 09 (2014) 143 [arXiv:1406.4814] [INSPIRE].
JHEP 12 (2015) 159 [arXiv:1508.03611] [INSPIRE].
bootstrap, JHEP 03 (2016) 183 [arXiv:1412.7541] [INSPIRE].
theories, JHEP 04 (2017) 032 [arXiv:1612.01536] [INSPIRE].
arXiv:1612.02363 [INSPIRE].
in Six Dimensions), arXiv:1705.05392 [INSPIRE].
Extended Conformal Supersymmetry, Phys. Lett. B 162 (1985) 127 [INSPIRE].
Supersymmetry: Function Space Realizations and Invariant Differential Operators, Fortsch.
Phys. 35 (1987) 537 [INSPIRE].
[arXiv:1602.07269] [INSPIRE].
[1] A.M. Polyakov , Nonhamiltonian approach to conformal quantum field theory, Zh . Eksp. Teor.