Physics Letters B 778 (2018) 71–78

Contents lists available at ScienceDirect

Physics Letters B
www.elsevier.com/locate/physletb

On current contribution to Fronsdal equations

N.G. Misuna a,b,∗
a
I.E. Tamm Department of Theoretical Physics, Lebedev Physical Institute, Leninsky prospect 53, 119991, Moscow, Russia
b
Moscow Institute of Physics and Technology, Institutsky lane 9, 141700, Dolgoprudny, Moscow region, Russia

a r t i c l e i n f o a b s t r a c t

Article history: We explore a local form of second-order Vasiliev equations proposed in [arXiv:1706.03718] and obtain an
Received 20 September 2017 explicit expression for quadratic corrections to bosonic Fronsdal equations, generated by gauge-invariant
Received in revised form 3 January 2018 higher-spin currents. Our analysis is performed for general phase factor, and for the case of parity-
Accepted 9 January 2018
invariant theory we find the agreement with expressions for cubic vertices available in the literature.
Available online 17 January 2018
Editor: M. Cvetič
This provides an additional indication that local frame proposed in [arXiv:1706.03718] is the proper one.
© 2018 The Author. Published by Elsevier B.V. This is an open access article under the CC BY license
(http://creativecommons.org/licenses/by/4.0/). Funded by SCOAP3 .

1. Introduction
Linear equations that describe a free propagation of massless
higher-spin (HS) fields were found a long time ago by Fronsdal
and Fang [1,2]. But to build any consistent nonlinear deformation
of them turned out to be extremely nontrivial task. Up to date
the only available example of full nonlinear HS gauge theory is
provided by Vasiliev equations [3,4]. They represent an interact-
ing theory of massless fields of all spins over anti-de Sitter ( AdS)
background. As opposed to e.o.m. of standard field theory, Vasiliev
equations are so-called unfolded ones, i.e. they represent first-
order differential equations in terms of exterior (0- and 1-) forms.
Each field of given spin is described by an infinite number of un-
folded fields parametrising all its degrees of freedom. An infinite
number of vertices, describing HS interactions, are encoded into
the evolution over auxiliary twistor-like variables. In this regard
Vasiliev equations can be considered as generating ones (“equa-
tions for equations”).
A reconstruction of space–time dynamics from Vasiliev equa-
tions is a nontrivial problem, essentially because of the freedom
in the choice of resolution operator for twistor-like variables. As
usual, the resolution operator is determined up to an arbitrary so-
lution of homogeneous equation, that in terms of physical fields
amounts to the freedom in a field redefinition, which can affect
the form of e.o.m. For example, in [5] it was found that by non-
local field redefinitions one can get rid of interactions in 3d HS
equations (see also [6,7] for the proof of pseudolocal-triviality of
*Correspondence to: I.E. Tamm Department of Theoretical Physics, Lebedev Phys-
ical Institute, Leninsky prospect 53, 119991, Moscow, Russia.
E-mail address: misuna@phystech.edu.
any 3d HS currents). In [8] it was shown that the simplest choice
of the resolution operator lead to nonlocal expressions for 4d cubic
HS vertices. All that brings up a question of admissible functional
class of field redefinitions [9–12]. On the other hand, field redefi-
nitions, bringing quadratic equations to the local form, were found
in [13] for the sector of 0-forms and in [14] for the sector of
1-forms. These were tested in [15–17], where it was shown that
the resulting local HS equations properly reproduce holographic
correlators in accordance with Klebanov–Polyakov HS AdS /C F T
conjecture [18]. Later, in [19] it was shown how to construct a
proper resolution operator, enforcing the locality at the second
order and minimising nonlocality at higher orders. Formally this
operator can be considered as the resolution operator of [8], recti-
fied by non-local field redefinitions of [13,14].
In this note we provide a further analysis of unfolded lo-
cal quadratic equations of [14] and obtain an explicit form of
corrections to bosonic Fronsdal equations that are generated by
gauge-invariant HS currents. These should be compared with re-
sults of [20] where HS cubic couplings were found in flat space
in lightcone formulation, and [21] where they were restored via
AdS /C F T from correlators of boundary free scalar theory and later
in [22] shown to solve the bulk Noether procedure. Expressions for
quadratic corrections we found turn out to be in the full agree-
ment with these results, thus providing one more confirmation
that the local frame of [13,14] is the appropriate one. In addition,
we worked out the dependence of vertices on the phase factor en-
tering Vasiliev equations, thus extending previous results to parity-
breaking theories. It turns out that there is a specific value of the
phase ϕ = π4 , where leading-derivative vertex maximally breaks
parity, which may have interesting implications for dual boundary
theory.

https://doi.org/10.1016/j.physletb.2018.01.019
0370-2693/© 2018 The Author. Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). Funded by
SCOAP3 .
72 N.G. Misuna / Physics Letters B 778 (2018) 71–78

2. Higher-spin equations which leaves only one field of every integer spin and is reached by
setting
HS equations in four dimensions are [4]     
W Z ; Y | K |x|θ A , dxm → W Z ; Y |x|θ A , dxm 1 + kk̄ ,
dW + W ∗ ∧ W = −i θα ∧ θ α (1 + η B ∗  k)  
  B ( Z ; Y | K |x) → B ( Z ; Y |x) k + k̄ . (2.12)
− i θ̄α̇ ∧ θ̄ α̇ 1 + η̄ B ∗ ¯ k̄ , (2.1)
η in (2.1) is a free complex parameter of the theory which can
dB + W ∗ B − B ∗ W = 0. (2.2) be normalised to be unimodular1 ηη̄ = 1, hence representing the
Here d is the space–time de Rham differential, W and B are phase factor freedom. HS theory is parity-invariant in the two
master-fields of the theory (onwards we omit wedge symbol) cases of η = 1 (A-model) and η = i (B-model) [23].
dependent on space–time coordinates and twistor-like variables
    3. Perturbation theory
Y A = y α , ȳ α̇ , Z A = zα , z̄α̇ with two-valued spinor indices α
and α̇ . The Y and Z realise the HS algebra through the noncom- To start a perturbative expansion one has to fix some vacuum
mutative star product solution to (2.1), (2.2). Eq. (2.2) can be solved by setting the vac-
uum value of B to zero
( f ∗ g )( Z , Y )

A B 0 = 0. (3.1)
= d4 U d4 V e iU A V f ( Z + U , Y + U ) g ( Z − V , Y + V ), (2.3)
Then the solution for (2.1) can be chosen as
with the integration measure fixed so as 1 ∗ f = f ∗ 1 = 1. Spinor
W 0 = ω AdS + Z A θ A , (3.2)
indices are raised and lowered via sp (2)-metrics
with the space–time 1-form of sp (4)-connection ω AdS describing
v α =  αβ v β , v α = β α v β , v̄ α̇ =  α̇ β̇ v̄ β̇ , v̄ α̇ = β̇ α̇ v̄ β̇ . the AdS 4 background

(2.4) i  αβ 
ω AdS = − ωL y α y β + ω̄α̇L β̇ ȳ α̇ ȳ β̇ + 2λhα β̇ y α ȳ β̇ , (3.3)
4
sp (4)-indices are transformed by  A B built from α β and α̇ β̇ dω AdS + ω AdS ∗ ω AdS = 0, (3.4)

V A
= AB
V B, V A = B A V . B
(2.5) where λ is the cosmological parameter (inverse radius of AdS).
Performing an expansion of (2.1)–(2.2) around vacuum
 and ¯ in (2.1) are inner Klein operators, which are specific ele- (3.1)–(3.2) one gets at the linear order
ments of the star-product algebra
    Dad ω (Y | K |x) = L (C ) , (3.5)
 := exp izα y α , ¯ := exp i z̄α̇ ȳ α̇ , (2.6)
Dt w C (Y | K |x) = 0, (3.6)
having the distinguishing properties where
    iλ
 ∗  = 1,  ∗ f zα , y α = f − zα , − y α ∗ , (2.7) L (C ) := η H̄ α̇ β̇ ∂¯α̇ ∂¯β̇ C (0, ȳ | K |x) k
4
α
f ( z, y ) ∗  = f (− y , − z) e izα y , (2.8) iλ
+ η̄ H α β ∂α ∂β C ( y , 0| K |x) k̄, (3.7)
4
and analogously for ¯ .
Master-field B is a 0-form, while W is a 1-form in a space–time H α β := hα γ̇ hβ γ̇ , H̄ α̇ β̇ := hγ α̇ hγ β̇ (3.8)
differential dxm or in an auxiliary differential θ A dual to Z A . All ∂ ∂
differentials anticommute ∂α := , ∂¯α̇ := , (3.9)
∂ yα ∂ ȳ α̇
 

 
dxm , dxn = dxm , θ A = θ A , θ B = 0. (2.9) Dad f (Y | K |x) := D L f + λhα β̇ y α ∂¯β̇ + ∂α ȳ β̇ f , (3.10)
 
Besides the inner Klein
 operators
 there is also a pair of exterior Dt w f (Y | K |x) := D L f − i λhα β̇ y α ȳ β̇ − ∂α ∂¯β̇ f , (3.11)
Klein operators K = k, k̄ which have similar properties to (, )
¯  
D L f := d f + ω L y α ∂β + ω̄ L ȳ α̇ ∂¯β̇ f .
αβ α̇ β̇
(3.12)
kk = 1, kf ( zα ; y α ; θ α ) = f (− zα ; − y α ; −θ α )k, (2.10)
Eqs. (3.5)–(3.6) represent a so-called unfolded form of Fronsdal
(analogously for k̄), but k (k̄) in addition anticommute with θ (θ̄ ) equations, describing free propagation of HS fields over AdS 4 back-
differentials that does not permit to realise them as elements of ground. Let us expand HS fields as
the star-product algebra. ∞

Thus the full arguments of master-fields are ω (Y | K |x) = ωm,n (Y | K |x) ,
 
m,n=0
W =W Z ; Y | K |x|θ A , dxm , B = B ( Z ; Y | K |x) . (2.11)
∞

K -dependence of the fields leads to the splitting of the field spec- C (Y | K |x) = C m,n (Y | K |x) , (3.13)
trum into topological and physical sectors. The first one describes m,n=0

finite-dimensional modules and contains W linear in k or k̄ and B

depending on kk̄. We truncate it away. The physical sector describ- 1
In [4] it was conjectured that a different situation when (η) η̄ = 0 corresponds
ing relativistic fields contains W depending on kk̄ and B linear in to (anti)selfdual HS theory, allowing no nontrivial amplitudes. Here we do not con-
k or k̄. Moreover, in this note we consider a bosonic reduction, sider this case.
 N.G. Misuna / Physics Letters B 778 (2018) 71–78 73

 
where 
s
in n
× ∂γ1 ∂ 2γ
(s + n − 1)!
n =0
f m,n (Y ) := f α1 ...αm ,β̇1 ...β̇n y α1 ... y αm ȳ β̇1 ... ȳ β̇n . (3.14) 
 n     2  

We also introduce a decomposition into different helicity-sign sec- + ∂¯γ̇ ∂¯
1 2 γ̇ 1
C Y | K |x C Y | K |x  . (3.23)
Y 1 = Y 2 =0
tors as

ω = ω+ + ω− + ω0 , C = C + + C − + C 0 , (3.15) 4. Currents contribution to Fronsdal equations

where Our goal is to develop an explicit expression for quadratic cor-

   rections to Fronsdal equations that are generated by (3.21)–(3.22).
f+ = f m,n , f− = f m,n , f0 = f m,n (3.16) Double-traceless field of spin-s is described in terms of spinors
m>n m<n m=n as ωα (s−1),α̇ (s−1)|β β̇ . We make use of the fact that the currents in
are positive-helicity, negative-helicity and zero-helicity sectors re- question are conformal [24], so we can keep track only of totally
spectively. Then a submodule describing spin-s field consists of traceless (in Lorentz tensor language) components of the Frons-
ωm,n , n + m = 2 (s − 1) and C m,n , |m − n| = 2s. dal fields, that in spinor language corresponds to totally symmetric
At the second order one should make field redefinitions that spinor-tensors φα (s),α̇ (s)
brings equations to the local frame, removing infinite higher-
derivative tails. Such redefinitions were found in [13,14]. Applying φs,s := ωs−1,s−1|β β̇ y β ȳ β̇ . (4.1)
them one obtains Next, as we are on-shell we can take our fields to be transverse

Dad ω + [ω, ω]∗ = L (C ) + Q (C , ω) + s<s1 +s2 ( J ) + can ( J ) , D α β̇ ∂ α ∂¯ β̇ φs,s = 0. (4.2)

(3.17) Finally, an important fact is that although the full nonlinear HS
sum theory does not admit a flat limit, cubic couplings we are studying
Dt w C (Y | K |x) + [ω, C ]∗ = −Hη ( J ) − Hη̄ ( J ) + Dt w B ( J),
do admit it (for Fradkin–Vasiliev 2 − s − s vertex [25,26] this was
(3.18) shown in [27]; see also [28]). So we can take a flat limit in our
equations and consider derivatives to be commuting
where
 
      D α α̇ , D β β̇ = 0. (4.3)
1 2 1 2
J Y , Y | K |x := C Y | K |x C Y | K |x (3.19)
In order to do this we rescale HS fields as follows
is a bilinear HS current. The above-mentioned redefinitions serve |m−n| m+n
to make J -dependent terms local. We will analyse the first equa- ωm,n −→ λ− 2 ωm,n , C m,n −→ λ− 2 C m,n . (4.4)
tion (3.17) that comprise Fronsdal equations with quadratic correc-
tions. These corrections are of the four types: [ω, ω]∗ term which For rescaled fields the flat limit λ → 0 turn covariant derivatives to
is completely fixed by HS symmetry algebra; gauge-dependent
contribution Q (C , ω) which is local from the very beginning Dad ω (Y | K |x) −→ D L ω + hα β̇ y α ∂¯β̇ ω− + hα β̇ ∂α ȳ β̇ ω+ , (4.5)
because ω is a polynomial in Y of restricted degree for any Dt w C (Y | K |x) −→ D C + ihα β̇ ∂α ∂¯β̇ C ,
L
(4.6)
fixed spin; s<s1 +s2 ( J ) being the current deformation in gauge-
dependent sector inside the triangle inequality s < s1 + s2 ; can ( J ) where D L and hα β̇ are Lorentz-covariant derivative and vierbein
which is gauge-invariant current deformation outside the triangle of Minkowski space–time. Then one substitutes (4.5)–(4.6) into
inequality, s ≥ s1 + s2 . It is this last contribution that we are inter- (3.5)–(3.6) and gets linear equations for HS fields in flat space–
ested in.
time.
Now we convert all objects to 0-forms expanding them in terms
of vierbeins D L ω (Y | K |x) + hα β̇ y α ∂¯β̇ ω− (Y | K |x) + hα β̇ ∂α ȳ β̇ ω+ (Y | K |x)
i i
ωm,n = hα β̇ ωm,n|α β̇ , D L = hα β̇ D α β̇ . (3.20) = η H̄ α̇ β̇ ∂¯α̇ ∂¯β̇ C (0, ȳ | K |x) k + η̄ H α β ∂α ∂β C ( y , 0| K |x) k̄,
4 4
Then one can rewrite a relevant sector of (3.17) describing current (4.7)
contribution to spin-s field e.o.m. as [14]
D C (Y | K |x) + ihα β̇ ∂α ∂¯β̇ C (Y | K |x) = 0.
L
(4.8)
D α β̇ ωs−2,s|α = − ȳ β̇ ∂α ωs−1,s−1|α − y α ∂¯β̇ ωs−3,s+1|α β̇
β̇ β̇
Now the first step is to express C fields in (3.23) via deriva-
tives of Fronsdal fields. To this end one rewrites (4.7) in terms of
+ ∂ α ∂α J s , s , (3.21)
0-forms
D β α̇ ωs,s−2| α̇ = − y β ∂¯α̇ ωs−1,s−1| α̇ − ȳ α̇ ∂β ωs+1,s−3| α̇
β β β
D β α̇ ωn,m|β α̇ = − y β ∂¯α̇ (ω− )n−1,m+1|β α̇ − ∂ β ȳ α̇ (ω+ )n+1,m−1|β α̇
+ ∂¯α̇ ∂¯α̇ Js,s , (3.22)
i
+ ηδn,0 ∂¯α̇ ∂¯α̇ C 0,m+2k, (4.9)
where 2

(s − 2)!  (m + k)! (2s − m − k)!

s D α β̇ ωn,m|α β̇ = − y α ∂¯ β̇ (ω− )n−1,m+1|α β̇ − ∂α ȳ β̇ (ω+ )n+1,m−1|α β̇
Js,s = i
8 (2s)! (s − k)!k! (s − m)!m! i
+ η̄δm,0 ∂α ∂α C n+2,0k̄. (4.10)
k,m=0
2
 m  β 2 s−m  α̇ 1 s−k  β̇ 2 k
× y α ∂α1 − y ∂β ȳ ∂¯α̇ − ȳ ∂¯β̇ Contracting (4.9) with ȳ α̇ ȳ α̇ and (4.10) with y α y α yields
74 N.G. Misuna / Physics Letters B 778 (2018) 71–78

   
i
ȳ α̇ D α α̇ ∂ α φn,m = n · m (φ− )n−1,m+1 − ηδn,1m (m + 1) C 0,m+1k, + C d1 ,2s1 +d1 Y 1 | K |x C d2 ,2s2 +d2 Y 2 | K |x

2
  
(4.11) 
+ C d2 ,2s2 +d2 Y 1 | K |x C d1 ,2s1 +d1 Y 2 | K |x  .
i Y 1 = Y 2 =0
yα D α α̇ ∂¯ φn,m = n · m (φ+ )n+1,m−1 − η̄δm,1 n (n + 1) C n+1,0 k̄.
α̇
2 (4.18)
(4.12)
That spins s1 , s2 of constituent fields are fixed and all Y and Y 2 1

From this one finds are eventually put to zero reduces the fivefold sum in (4.18) to the

2i η  s single one:
C 2s,0 = y α D α α̇ ∂¯ α̇ φs,s k̄, (4.13)
(s − 2)! 
s
s · (2s)! (s + s1 − s2 )! (s − s1 + s2 )!
 s JsH−s1 −s2 = i
2i η̄ 8 (2s)! ( s − d )!d ! (−s1 + s2 + d)! (s + s1 − s2 − d)!
C 0,2s = ȳ α̇ D α α̇ ∂ α φs,s k . (4.14) d =0
s · (2s)!  s+s1 −s2 −d  β 2 −s1 +s2 +d  α̇ 1 s−d
× y α ∂α1 y ∂β ȳ ∂¯α̇
Then (4.8) gives
 d i s1 +s2 (−1)s+d 1 + (−1)s+s1 +s2 

2η · i d +1  d  s β̇ ¯ 2
× ȳ ∂β̇
C 2s+d,d = y β D β β̇ ȳ β̇ y α D α α̇ ∂¯ α̇ φs,s k̄, (s + s1 + s2 − 1)!
s · (2s + d)!d! 
 s1 +s2  
(4.15) × ∂γ1 ∂ 2γ C 2s1 +s−d,s−d Y 1 |x
2η̄ · i d +1  d  s
C d,2s+d = y β D β β̇ ȳ β̇ ȳ α̇ D α α̇ ∂ α φs,s k . 
s · (2s + d)!d!   
2 
× C 2s2 +d,d Y |x + h.c .  (4.19)
(4.16) Y 1 = Y 2 =0

Now one contracts (3.21) with y α y α (or (3.22) with ȳ α̇ ȳ α̇ ) and (here we resolved K -dependence using (2.10)), that after evaluat-
makes use of (4.12) (or (4.11)) to obtain ing derivatives from the second and third lines yields
(s − 2)! (s + s1 − s2 )! (s − s1 + s2 )! s1 +s2 +1
φs,s + ... = −s2 (s − 1) Js,s + ..., (4.17) JsH−s1 −s2 = i
8 (2s)! (s + s1 + s2 − 1)!
where  = 12 D α α̇ D α α̇ , ellipsis on the l.h.s. denotes other terms of  
× 1 + (−1)s+s1 +s2
Fronsdal kinetic operator besides the box and ellipsis on the r.h.s.
s 
denotes other (gauge-noninvariant) sources generated by Q (C , ω) s+d
 s1 +s2
and s<s1 +s2 ( J ) in (3.17). · (−1) ∂γ C 2s1 +s−d,s−d (Y |x)
Now let us consider the current (3.23). We want to extract d =0
 s +s  s +s
the term describing s − s1 − s2 vertex. A simple counting shows · ∂ γ 1 2 C 2s2 +d,d (Y |x) + ∂¯γ̇ 1 2
that two kind of terms are presented in (3.23): either two co-  s1 +s2 
directional helicities are coupled (C + C + or C − C − ), then the term ¯
× C 2s1 +s−d,s−d (Y |x) · ∂ γ̇ C 2s2 +d,d (Y |x) .
has (s + s1 + s2 ) derivatives, or two opposite ones (C + C − or
C − C + ), then total number of derivatives is (s + |s1 − s2 |) (let us (4.20)
remind that we are in s ≥ s1 + s2 sector). This corresponds to two  s+s1 +s2

Note that due to 1 + (−1) factor, (4.20) vanishes if the
types of 4d cubic HS vertices found in [29]. Altogether this means
total sum of spins is odd. In fact, this is because we have only
there are no higher-derivative improvements to vertices of [29],
one field of every spin, similarly to the electrodynamics where one
which could, for instance, affect locality issue in higher orders.
needs two copies of the matter fields to have a nonzero electric
(Note that lower-derivative improvements to (s + s1 + s2 )-term
current. So if one considers matrix-valued HS fields, the contribu-
cannot contribute to (s + |s1 − s2 |)-term because they have dif-
tion would be nonzero.
ferent helicity structure.) We will analyse two vertices separately.
Now let us analyse the spinorial expression in (4.20). Our goal
is to bring it to the form that can be simply re-expressed in terms
4.1. Maximal-derivative part
of Lorentz tensors.
First, we use (4.15) to rewrite it as
First, let us consider the part of (3.23) with (s + s1 + s2 ) deriva-
 s1 +s2  s1 +s2
tives. This looks as follows ∂γ C 2s1 +s−d,s−d (Y |x) · ∂ γ C 2s2 +d,d (Y |x)

(s − 2)! 
s
(m + k)! (2s − m − k)! 4η 2 i s
JsH−s1 −s2 = i =−
8 (2s)! (s − k)!k! (s − m)!m! (2s1 + s − d)! (2s2 + d)! (s − d)!d! · s1 · s2
 
k,m=0
 s1 +s2  α s−d  s1
 m  β 2 s−m  α̇ 1 s−k  β̇ 2 k × ∂γ y D α α̇ ȳ α̇ α
y D α α̇ ∂ ¯ α̇ φs1 ,s1
× y α ∂α1 − y ∂β ȳ ∂¯α̇ − ȳ ∂¯β̇  
 s +s  d  s2

s
in  n  n  × ∂ γ 1 2 y β D β β̇ ȳ β̇ y β D β β̇ ∂¯ β̇ φs2 ,s2 . (4.21)
× ∂γ1 ∂ 2γ + ∂¯γ̇1 ∂¯ 2γ̇
(s + n − 1)!
n =0 Evaluating spinorial derivatives gives
∞     
    s1 +s2  α s−d  s1
× C 2s1 +d1 ,d1 Y 1 | K |x C 2s2 +d2 ,d2 Y 2 | K |x ∂γ y D α α̇ ȳ α̇ y α D α α̇ ∂¯ α̇ φs1 ,s1
d1 ,d2 =0  
     γ s1 +s2  β d  s
β̇ 2
× ∂ y D β β̇ ȳ β̇ β
y D β β̇ ∂ ¯ φs2 ,s2
+ C 2s2 +d2 ,d2 Y 1 | K |x C 2s1 +d1 ,d1 Y 2 | K |x
 N.G. Misuna / Physics Letters B 778 (2018) 71–78 75

s1 !s2 ! (s + 2s1 − d)! (2s2 + d)! where at the penultimate step we used that
=
(s + s1 − s2 − d)! (−s1 + s2 + d)!

 s +s  s+s1 −s2 −d  μ̇ s−d  s−d D α γ̇ φ · D β γ̇ φ = D β γ̇ φ · D α γ̇ φ + α β D γ γ̇ φ · D γ γ̇ φ
× δγ μ 1 2 y μ ȳ D μμ̇
 = D β γ̇ φ · D α γ̇ φ + α β  (φ · φ)
 s1
× D μβ̇ φμ(s1 ), β̇(s1 ) − α β (φ · φ + φ · φ)
 ≈ D β γ̇ φ · D α γ̇ φ,
 s +s  −s +s +d  ν̇ d (4.29)
×  γ ν 1 2 yν 1 2 ȳ ( D ν ν̇ )d
and at the last step that

s2
× ( D ν α̇ ) φν (s2 ), α̇ (s2 )
. (4.22) D α α̇ D β β̇ φ α β̇ ≈ D β α̇ D α β̇ φ α β̇ ≈ 0. (4.30)

Thus (4.27) turns into

Due to symmetrisation
s +s s +μs (and over ν ), γ indices after apply-
 over
ing δγ μ 1 2 and  γ ν 1 2 will hang symmetrically on fields
s1 !s2 ! (s + 2s1 − d)! (2s2 + d)!  μ s  μ̇ s
and derivatives. But we are going to arrange gammas in some par- y ȳ
ticular order. To this end we establish some useful relations. The (s + s1 − s2 − d)! (−s1 + s2 + d)!
 
first is (we write down only relevant indices)  s−d  s1 −s2
× D μμ̇ D μβ̇ ( D δ α̇ )s2 φγ (s1 ), β̇(s1 )
D α γ̇ φβ γ̇ = D β γ̇ φα γ̇ + α β D γ γ̇ φ γ γ̇ ≈ D β γ̇ φα γ̇ , (4.23)  d−s1 +s2  γ s1 −s2  γ s2 δ(s ),α̇ (s )

where the approximate equality symbol means that we lopped off × D μμ̇ D μ̇ D β̇ φ 2 2
. (4.31)
a divergence of the field, as we neglect it in our problem. The sec-  s1 −s2
ond is Now we want to exchange β̇ in D μβ̇ in the second line
 s −s
D α α̇ D β β̇ φ ≈ D α β̇ D β α̇ φ, (4.24) with μ̇ in D γ μ̇ 1 2 in the third line. This can be done by virtue
of a relation, similar to (4.28):
which is modulo boxes (that can be redefined away) and terms
with D α β̇ D α β̇ (D β α̇ D β α̇ ), which are zeros in flat space. Using these D μβ̇ D μβ̇ φ γ β̇ β̇ · D γ μ̇ φ = D μβ̇ D μμ̇ φ γ β̇ β̇ · D γ β̇ φ
two relations, we obtain the third one
+ D μβ̇ D μα̇ φ γ β̇ μ̇ · D γ α̇ φ
D γ γ̇ D α β̇ φβ β̇ ≈ D β γ̇ D α β̇ φγ β̇ . (4.25)
Altogether they imply that we are free to put gammas on any
≈ D μβ̇ D μμ̇ φ γ β̇ β̇ · D γ β̇ φ
places instead of lower μ (ν ) indices in (4.22) as all combinations + D μβ̇ D γ α̇ φ γ β̇ μ̇ · D μ α̇ φ
are equivalent. So, assuming for definiteness s1 ≥ s2 , we rewrite
(4.22) as ≈ D μβ̇ D μμ̇ φ γ β̇ β̇ · D γ β̇ φ. (4.32)
s1 !s2 ! (s + 2s1 − d)! (2s2 + d)!  μ s  μ̇ s  s1 −s2
y ȳ This allows us to perform all necessary exchanges in D μβ̇
(s + s1 − s2 − d)! (−s1 + s2 + d)!
 
 s−d  s1 −s2  s2 except for the last one, because to use (4.32) we need at least two
× D μμ̇ D μβ̇ D γ β̇ φγ (s1 ), β̇(s1 ) D μβ̇ . So we have
 d−s1 +s2  γ s1 −s2  γ s2 γ (s ),α̇ (s )

× D μμ̇ D μ̇ D α̇ φ 2 2
, (4.26) s1 !s2 ! (s + 2s1 − d)! (2s2 + d)!  μ s  μ̇ s
y ȳ
(s + s1 − s2 − d)! (−s1 + s2 + d)!
and, using (4.23), further as  s+s1 −s2 −d−1

 μ s  μ̇ s × D μβ̇ D μμ̇ ( D α α̇ )s2 φβ(s1 ), β̇(s1 )

s1 !s2 ! (s + 2s1 − d)! (2s2 + d)!
y ȳ  
(s + s1 − s2 − d)! (−s1 + s2 + d)!  −s +s +d  β s1 −1 α (s ),α̇ (s )
  × D β μ̇ D μμ̇ 1 2 D β̇ φ 2 2
, (4.33)
 s−d  s1 −s2  s2
β̇(s1 )
× D μμ̇ D μβ̇ D δ β̇ φγ (s1 ),
 and the last exchange leads to the expression of the form
d−s1 +s2  γ s1 −s2  γ s2 δ(s ),α̇ (s )

× D μμ̇ D μ̇ D α̇ φ 2 2
. (4.27)
s1 !s2 ! (s + 2s1 − d)! (2s2 + d)!  μ s  μ̇ s
 s2 y ȳ (−1)s1
(s + s1 − s2 − d)! (−s1 + s2 + d)!
Now we want to replace all lower β̇ in D δ β̇ in the second 
 s+s1 −s2 −d
line with lower α̇ so as to make these derivatives to be entirely × D μμ̇ ( D α α̇ )s2 φ β(s1 ),β̇(s1 )
contracted with the spin-s2 field. We can perform this with the
help of D μβ̇ in the second line, because  −s +s +d  s1
· D μμ̇ 1 2 D β β̇ φ α (s2 ),α̇ (s2 )
D μβ̇ D δ β̇ φ γ β̇ β̇ · D γ α̇ φ δ α̇ = D μβ̇ D δ α̇ φ γ β̇ β̇ · D γ β̇ φ δ α̇  s+s1 −s2 −d−1
− D μμ̇ ( D α α̇ )s2 D β μ̇ φ β(s1 ),β̇(s1 )

+ D μβ̇ D δ γ̇ φ γ β̇ α̇ · D γ γ̇ φ δ α̇  −s +s +d   s 1 −1
· D μβ̇ D μμ̇ 1 2 D β β̇ φ α (s2 ),α̇ (s2 ) . (4.34)
≈ D μβ̇ D δ α̇ φ γ β̇ β̇ · D γ β̇ φ δ α̇
Now, substituting (4.34) for the last two lines in (4.21), then
+ D μβ̇ D γ γ̇ φ γ β̇ α̇ · D δ γ̇ φ δ α̇
(4.21) in (4.20), adding conjugate expression and simplifying, one
≈ D μβ̇ D δ α̇ φ γ β̇ β̇ · D γ β̇ φ δ α̇ , (4.28) gets
76 N.G. Misuna / Physics Letters B 778 (2018) 71–78

  ∞ 
  
(s − 2)! 
s
s + s1 − s2 s − s1 + s2  
JsH−s1 −s2 = −i × C 2s1 +d1 ,d1 Y 1 | K |x C d2 ,2s2 +d2 Y 2 | K |x
4 (2s)! d s−d
d =0 d1 ,d2 =0
   
× (s1 − 1)! (s2 − 1)!
  + C 2s2 +d2 ,d2 Y 1 | K |x C d1 ,2s1 +d1 Y 2 | K |x
i s+s1 +s2 (−1)s+d+s1 1 + (−1)s+s1 +s2  μ s  μ̇ s  
× y ȳ  
(s + s1 + s2 − 1)! + C d1 ,2s1 +d1 Y 1 | K |x C 2s2 +d2 ,d2 Y 2 | K |x
  s+s1 −s2 −d    
× η2 + η̄2 D μμ̇ ( D α α̇ )s2 φ β(s1 ),β̇(s1 ) 
+ C d2 ,2s2 +d2 Y 1 | K |x C 2s1 +d1 ,d1 Y 2 | K |x  ,
Y 1 = Y 2 =0
 −s +s +d  s1  
· D μμ̇ 1 2 D β β̇ φ α (s2 ),α̇ (s2 ) + η2 − η̄2 (4.38)

 s+s1 −s2 −d−1 practically repeats analysis of the maximal-derivative one.
× D μβ̇ D μμ̇ ( D α α̇ )s2 φ β(s1 ),β̇(s1 ) First, one evaluates derivatives from the second line of (4.38)
 −s +s +d   s 1 −1 and simplifies the expression to
· D β μ̇ D μμ̇ 1 2 D β β̇ φ α (s2 ),α̇ (s2 )
(s − 2)! (s + s1 + s2 )! (s − s1 − s2 )! s1 +s2 +1
 s+s1 −s2 −d−1 JsL−s1 −s2 = i
− D β μ̇ D μμ̇ ( D α α̇ )s2 φ β(s1 ),β̇(s1 ) 8 · (2s)! (s + s1 − s2 − 1)!
  
 −s +s +d   s 1 −1 × 1 + (−1)s+s1 +s2
· D μβ̇ D μμ̇ 1 2 D β β̇ φ α (s2 ),α̇ (s2 ) . 
s
 s −s
· (−1)s+d+s1 ∂γ 1 2 C 2s1 +d,d (Y |x)
(4.35)
d =0
Then, using Vandermonde’s identity  s −s
· ∂ γ 1 2 C s−2s2 −d,s−d (Y |x)
c 
    s −s
a b
=
a+b
(4.36) + ∂¯γ̇ 1 2 C d,2s1 +d (Y |x)
n c −n c  s1 −s2 
n =0
· ∂¯ γ̇ C s−d,s−2s2 −d (Y |x) . (4.39)
and integrating by parts one can perform a summation over d ex-
plicitly, reducing (4.35) to Then, using (4.15), one rewrites the first term in brackets in (4.39)
as
JsH−s1 −s2
   s1 −s2  s1 −s2
i s+s1 +s2 +1 (s − 2)! (s1 − 1)! (s2 − 1)! (−1)s1 1 + (−1)s+s1 +s2 ∂γ C 2s1 +d,d (Y |x) · ∂ γ C s−2s2 −d,s−d (Y |x)
=−
4 · s!s! (s + s1 + s2 − 1)! 4i (−1) (s1 − 1)! (s2 − 1)!
s s2

 μ s  μ̇ s  2  s =−
(s1 + s2 + d)! (s − s1 − s2 − d)! (s − d)!d!
× y ȳ η + η̄2 D μμ̇ ( D α α̇ )s2 φ β(s1 ),β̇(s1 ) 
 s −s  s +s +d  μ̇ d  d
 s1   · δγ μ 1 2 y μ 1 2 ȳ D μμ̇
· D β β̇ φ α (s2 ),α̇ (s2 ) + η2 − η̄2
 
  s −1  s
× D μβ̇ D μμ̇ ( D α α̇ )s2 φ β(s1 ),β̇(s1 ) × D μα̇ 1 φμ(s1 ), α̇ (s1 )

  s 1 −1  γ ν s1 −s2  ν s−s1 −s2 −d  ν̇ s−d s−2s2 −d
· D β μ̇ D β β̇ φ α (s2 ),α̇ (s2 ) ·  y ȳ (D ν ν̇ )
  s −1 
− D β μ̇ D μμ̇ ( D α α̇ )s2 φ β(s1 ),β̇(s1 ) × ( D α ν̇ ) s2
φ α (s2 ) (4.40)
 ν̇ (s2 ) .
  s 1 −1
· D μβ̇ D β β̇ φ α (s2 ),α̇ (s2 )
. (4.37)
As in Section 4.1, by means of (4.23)–(4.25) one hangs all gammas
in the third line of (4.40) on spin-s1 field
Here we reached our goal because this expression can easily be
translated into Lorentz tensors as we show below. Now we are  s1 −s2  s1 −s2
∂γ C 2s1 +d,d (Y |x) · ∂ γ C s−2s2 −d,s−d (Y |x)
going to process another part of the current (3.23) that contains
(s + |s1 − s2 |) derivatives. 4i (−1) (s1 − 1)! (s2 − 1)!
s s1
=−
(s1 + s2 + d)! (s − s1 − s2 − d)! (s − d)!d!
4.2. Minimal-derivative part  s  μ̇ s  d  s1

· yμ ȳ D μμ̇ D μβ̇ φμ(s2 ) γ (s1 −s2 ),β̇(s1 )

Analysis of the minimal-derivative part of (3.23) which has the  s −s  s−s1 −s2 −d  s

form × D γ μ̇ 1 2 D μμ̇ D α μ̇ 2 φ α (s2 ), μ̇(s2 ) ,

(s − 2)!  (m + k)! (2s − m − k)!

s (4.41)
JsL−s1 −s2 = i
8 (2s)! (s − k)!k! (s − m)!m! and exchanges (s1 − s2 ) pieces of β̇ of D μβ̇ in the first bracket with
k,m=0
 m  s−m  s−k  k μ̇ of D γ μ̇ from the second bracket
× y α ∂α1 − y β ∂β2 ȳ α̇ ∂¯α̇1 − ȳ β̇ ∂¯β̇2  d  s1

 n  n  D μμ̇ D μβ̇ φμ(s2 ) γ (s1 −s2 ),β̇(s1 )


s
in
× ∂γ1 ∂ 2γ + ∂¯γ̇1 ∂¯ 2γ̇  s1 −s2  s−s1 −s2 −d  s2 α (s ),

(s + n − 1)! × D γ μ̇ D μμ̇ D α μ̇ φ 2 μ̇(s2 )

n =0
 N.G. Misuna / Physics Letters B 778 (2018) 71–78 77

 s1 −s2 +d  s2

 (s1 − 1)! (s2 − 1)!
= D μμ̇ D μβ̇ φμ(s2 ) γ (s1 −s2 ),β̇(s1 ) φa(s) + ... =
  (s − 1)!
s1 −s2  s−s1 −s2 −d  s2 α (s ), s1 +s2 ≤s
×  
D γ β̇ D μμ̇ D α μ̇ φ 2 μ̇(s2 ) . × 1 + (−1)s+s1 +s2 i s+s1 +s2 +1 2s1 +s2
  
(4.42) (−1)s1 η2 + η̄2
· ( D a )s1 −s2 ( D b )s2 φ c (s1 )
Substituting all this into (4.39), adding conjugate term and allow-
4 · (s + s1 + s2 − 1)!
 
ing for s−s1 +s2 s1 b (s2 ) (−1)s1 i η2 − η̄2
· (Da) (Dc ) φ +
4 · (s + s1 + s2 − 1)!
D μβ̇ φμ β̇ · D α μ̇ φ α μ̇ ≈ D μμ̇ φ α β̇ · D α β̇ φμμ̇ − D μμ̇ φ α β̇ · D μμ̇ φα β̇
× af cg D f ( D a )s1 −s2 −1 ( D b )s2 φ c(s1 )
− D β μ̇ φ μ̇ · D μα̇ φμ α̇
β
(4.43)
· D g ( D a ) s − s 1 + s 2 ( D c ) s 1 −1 φ b ( s 2 )
leads to the following expression for the minimal-derivative part
s2 
2− s 2 s
of the current + φ b (s1 ) (−1)s2 +n 2
  (s + s1 − s2 − 1)! n
(s − 2)! i s+s1 +s2 1 + (−1)s+s1 +s2 n =0

JsL−s1 −s2 = −i
(2s)! (s + s1 − s2 − 1)! × ( D a )s−n ( D b )s1 −s2 +n φb(s2 −n)a(n) + ... (4.48)
× (s1 − 1)! (s2 − 1)!
 s   To simplify the form of this equation one can rescale fields as
s + s1 + s2 s − s1 − s2
× (−1)d
s−d d n
−2
d =0 2

 s μ̇ s   d φa(n) −→ φa(n) , (4.49)
× yμ ȳ D μμ̇ φ α (s1 ),α̇ (s1 ) (n − 1)!i (n+1)
 s  then (4.48) turns into
 2
s  s−d−n
· (−1)n 2 D μμ̇ ( D α α̇ )s1 −s2 +n    s+s1 +s2
n =0
n φa(s) + ... = 1 + (−1)s+s1 +s2 2 2
 s1 +s2 ≤s

× φα (s2 −n)μ(n),α̇ (s2 −n)μ̇(n) . (4.44) (−1)s2
× cos (2ϕ ) ( D a )s ( D b )s2 φ c (s1 )
2 ·  (s + s1 + s2 )
As in the Section 4.1, using Vandermonde’s identity (4.36) and in- (−1)s2
tegrating by parts one can evaluate the sum over d, obtaining · ( D c )s1 φ b(s2 ) − sin (2ϕ )
2 ·  (s + s1 + s2 )
(s − 2)! (s1 − 1)! (s2 − 1)! s+s1 +s2 +1 × af cg D f ( D a )s−1 ( D b )s2 φ c(s1 ) · D g ( D c )s1 −1 φ b(s2 )
JsL−s1 −s2 = − i
s!s! (s + s1 − s2 − 1)!  s
   s  μ̇ s α (s ),α̇ (s ) (−1)s 2

+ kn ( D a )s−n φ b(s1 )
× 1 + (−1)s+s1 +s2 y μ ȳ φ 1 1
 (s + s1 − s2 )
n =0
 s  
 2
s  s−n s1 −s2 +n
· (−1)n 2 D μμ̇ ( D α α̇ )s1 −s2 +n · (Db) φb(s2 −n)a(n) + ... (4.50)
n
n =0
 where
× φα (s2 −n)μ(n),α̇ (s2 −n)μ̇(n) . (4.45)  
s2
s2
kn := 2−s2 , kn = 1, (4.51)
n
This completes the analysis of minimal-derivative part of the cur- n =0
rent. and we introduced a ‘phase angle’ ϕ
4.3. Fronsdal equations with HS current corrections η = exp (i ϕ ) . (4.52)

Now we are ready to make a final step and write down a Let us discuss (4.50), which is the main result of the paper, in
some more details. First of all, let us remind that ellipsis on the
current contribution to quadratic HS equations in Lorentz tensor
l.h.s. denotes the rest of kinetic Fronsdal operator, while ellipsis
language. From (4.17) we have
on the r.h.s. denotes contributions in s < s1 + s2 domain and the
 s  μ̇ s contributions of HS currents outside the transverse-traceless (TT)
φμ(s),μ̇(s) y μ ȳ + ...
sector. The non-TT part is completely fixed by the TT one, which
  
we have found (the procedure of completion of TT part to the full
= − s 2 ( s − 1) JsH−s1 −s2 + JsL−s1 −s2 + ..., (4.46)
Lagrangian AdS HS cubic vertex were demonstrated in [21,30]).
s1 +s2 ≤s
Next, we see that minimal-derivative part of the current (the
where JsH−s1 −s2 and JsL−s1 −s2 are given in (4.37) and (4.45) respec- last term in brackets) is ϕ -independent, while maximal-derivative
tively. After removing twistor variables y and ȳ, tensor indices are part consists of two different ϕ -dependent terms. Term propor-
restored via σ -matrices, that gives by virtue of tional to sin (2ϕ ) contains Levi-Civita symbol and thus is parity-
  violating, so it expectedly vanishes in parity-invariant A- and
Tr {σa σ̄b } = 2ηab , σ a σ̄ b σ c − σ c σ̄ b σ a = 2i  abcd σd (4.47) B-models (ϕ = 0 and ϕ = π2 ). For parity-invariant models vertices
in (4.50) coincide up to a normalisation with the expressions avail-
the following result able in the literature [20,21], confirming the correctness of local
78 N.G. Misuna / Physics Letters B 778 (2018) 71–78
frame of Vasiliev equations found in [14]. Another peculiar situa-
tion is ϕ = π4 model. In this case the first term with cos (2ϕ ) is
absent, so the maximal-derivative part of the vertex is in whole
proportional to Levi-Civita symbol, being somewhat of ‘maximally
parity-breaking’. It would be interesting to see the implication of
that for dual theory, which is conjectured to be 3d Chern–Simons
theory coupled to scalar fields [31,32].
5. Conclusion
In the note we obtained quadratic corrections to bosonic Frons-
dal equations generated by gauge-invariant HS currents, starting
with the local second-order Vasiliev equations of [13,14]. The re-
sult agrees with previously known expressions [20,21] for HS cubic
vertices in case of parity-invariant models. This gives an additional
confirmation that the local frame of HS equations, proposed in [13,
14], is the appropriate one. For the case of ϕ = π4 model we found
that maximal-derivative part of the vertex is proportional to Levi-
Civita symbol, being maximally parity-breaking, that may have in-
teresting consequences for dual boundary Chern–Simons theory. It
would be interesting also to study the theories with fermions as
well as to find the contribution of gauge-dependent sector, that
would allow one to write down the full quadratic HS equations.
Acknowledgements
The author is grateful to O.A. Gelfond for important technical
remark, to V.E. Didenko for useful comments, to D. Francia and
M. Taronna for the correspondence, and especially to M.A. Vasiliev
for numerous valuable discussions and critical remarks. The work
was partially supported by the Russian Basic Research Founda-
tion Grant No. 17-02-00546 and by the Foundation for Theoretical
Physics Development “Basis”.
References
[1] C. Fronsdal, Phys. Rev. D 18 (1978) 3624;
C. Fronsdal, Phys. Rev. D 20 (1979) 848.
[2] J. Fang, C. Fronsdal, Phys. Rev. D 18 (1978) 3630;
J. Fang, C. Fronsdal, Phys. Rev. D 22 (1980) 1361.
[3] M.A. Vasiliev, Phys. Lett. B 243 (1990) 378.
[4] M.A. Vasiliev, Phys. Lett. B 285 (1992) 225.
[5] S.F. Prokushkin, M.A. Vasiliev, Nucl. Phys. B 545 (1999) 385, arXiv:hep-
th/9806236.
[6] S.F. Prokushkin, M.A. Vasiliev, Theor. Math. Phys. 123 (2000) 415–435, Teor.
Mat. Fiz. 123 (2000) 3–25, arXiv:hep-th/9907020.
[7] P. Kessel, G.L. Gómez, E. Skvortsov, M. Taronna, J. High Energy Phys. 1511
(2015) 104, arXiv:1505.05887.
[8] N. Boulanger, P. Kessel, E.D. Skvortsov, M. Taronna, J. Phys. A 49 (9) (2016)
095402, arXiv:1508.04139.
[9] M.A. Vasiliev, J. High Energy Phys. 1506 (2015) 031, arXiv:1502.02271.
[10] E.D. Skvortsov, M. Taronna, J. High Energy Phys. 1511 (2015) 044, arXiv:
1508.04764.
[11] M. Taronna, in: Proceedings, International Workshop on Higher Spin Gauge
Theories, Singapore, Singapore, November 4–6, 2015, 2017, pp. 59–84, arXiv:
1602.08566.
[12] M. Taronna, J. Phys. A 50 (7) (2017) 075401, arXiv:1607.04718.
[13] M.A. Vasiliev, arXiv:1605.02662.
[14] O.A. Gelfond, M.A. Vasiliev, arXiv:1706.03718.
[15] V.E. Didenko, M.A. Vasiliev, arXiv:1705.03440.
[16] M. Taronna, Higher-spin theories and locality, talk given at “Higher-spin theory
and holography-6”, http://www.hsth.lpi.ru/proceedings/taronna.pdf.
[17] E. Sezgin, E.D. Skvortsov, Y. Zhu, arXiv:1705.03197.
[18] I.R. Klebanov, A.M. Polyakov, Phys. Lett. B 550 (2002) 213, arXiv:hep-th/
0210114.
[19] M.A. Vasiliev, arXiv:1707.03735.
[20] R.R. Metsaev, Mod. Phys. Lett. A 6 (1991) 359.
[21] C. Sleight, M. Taronna, Phys. Rev. Lett. 116 (18) (2016) 181602, arXiv:
1603.00022.
[22] C. Sleight, M. Taronna, J. High Energy Phys. 1702 (2017) 095, arXiv:1609.00991.
[23] E. Sezgin, P. Sundell, J. High Energy Phys. 0507 (2005) 044, arXiv:hep-th/
0305040.
[24] O.A. Gelfond, M.A. Vasiliev, Theor. Math. Phys. 187 (3) (2016) 797, Teor. Mat.
Fiz. 187 (3) (2016) 401, arXiv:1510.03488.
[25] E.S. Fradkin, M.A. Vasiliev, Nucl. Phys. B 291 (1987) 141.
[26] E.S. Fradkin, M.A. Vasiliev, Phys. Lett. B 189 (1987) 89.
[27] N. Boulanger, S. Leclercq, P. Sundell, J. High Energy Phys. 0808 (2008) 056,
arXiv:0805.2764.
[28] E. Joung, M. Taronna, Nucl. Phys. B 861 (2012) 145–174, arXiv:1110.5918.
[29] R.R. Metsaev, Nucl. Phys. B 759 (2006) 147, arXiv:hep-th/0512342.
[30] D. Francia, G. Lo Monaco, K. Mkrtchyan, J. High Energy Phys. 1704 (2017) 068,
arXiv:1611.00292.
[31] S. Giombi, S. Minwalla, S. Prakash, S.P. Trivedi, S.R. Wadia, Xi Yin, Eur. Phys. J.
C 72 (2012) 2112, arXiv:1110.4386.
[32] O. Aharony, G. Gur-Ari, R. Yacoby, J. High Energy Phys. 1203 (2012) 037, arXiv:
1110.4382.