A study of time reversal symmetry of abelian anyons
HJE
A study of time reversal symmetry of abelian anyons
Yasunori Lee 0 1
Yuji Tachikawa 0 1
0 Kashiwa , Chiba 2778583 , Japan
1 Kavli Institute for the Physics and Mathematics of the Universe, University of Tokyo
We perform a study of time reversal symmetry of abelian anyons A in 2+1 dimensions, in the spin structure independent cases. We will find the importance of the group C of timereversalsymmetric anyons modulo anyons composed from an anyon and its time reversal. Possible choices of local Kramers degeneracy are given by quadratic refinements of the braiding phases of C, and the anomaly is then given by the Arf invariant of the chosen quadratic refinement. We also give a concrete study of the cases when A is
Anomalies in Field and String Theories; Anyons; Discrete Symmetries; Topo
1 Introduction and summary 2
Basics of abelian anyons and their time reversal
2.1
2.2
Defining data of abelian anyons
Examples of abelian anyons
2.2.1
2.2.2
2.2.3
2.2.4
Abelian ChernSimons
Nonabelian ChernSimons at level 1
Finite group gauge theories
Universality of abelian ChernSimons constructions
MooreSeiberg data of abelian anyons
The time reversal, the obstruction and the anomaly
2.3
2.4
3.1
3.2
3.3
5.1
5.2
5.3
3
Time reversal and the anomaly formula
General properties to be established
Derivations of the properties
A comment on the spin case
4
5
Case study I: when A is odd
Case study II: A = (Z2)
N
Classification of the time reversal action T
When the braid pairing B is symplectic
When the braid pairing B is orthogonal
5.3.1
5.3.2
Some simple cases
The general case
can be anomalous, in which case the system lives on the boundary of a symmetryprotected
topological phase (SPT) in the bulk, i.e. the anomalous TQFT provides a gapped boundary
of an SPT.
– 1 –
The basic formalism of symmetry actions on 2+1d TQFTs1 was laid out in [1], and
a detailed analysis for the timereversal symmetry was given in [2]. In these references
one can find the entire formalism together with various interesting examples, which mostly
involved nonabelian anyons.
What we aim to provide in this paper is a study of time reversal actions on abelian
anyon systems. Abelian anyons are far simpler than nonabelian anyons, and the group
Z2 generated by the time reversal is far easier than the general group of symmetries. This
makes many of the necessarily complicated equations in [1, 2] more accessible. Still, the
TQFT structure in the abelian anyons show many of the features of general nonabelian
anyons, and we can hope that the antiunitarity inherent in the timereversal might give
us an interesting twist in the analysis.
S = (i/2π)k R (ada − bdb) such that T exchanges a and b.
Some examples.
A major source of abelian anyon systems is the abelian ChernSimons
theories, whose action is given by S = (i/2π)KIJ R aI daJ , for N U(
1
) gauge fields aI ,
(I = 1, . . . , N ) and an integer matrix KIJ .2 The time reversal T can act on aI by a matrix
TIJ such that T2 = 1. Then the classical action is invariant under the time reversal if
and only if TKT = −K. Timereversalinvariant abelian anyon systems in this setup was
studied in detail in [3]. One obvious example is the U(
1
)k × U(
1
)−k theory with the action
There are however subtler examples of timereversalsymmetric abelian anyon systems
known in the literature. One example is the socalled semionfermion system, which is the
U(
1
)2 × U(
1
)−1 theory with the action S = (i/π) R ada − (i/2π) R bdb [4, 5]. In this case,
there is no integer matrix T acting on U(
1
) gauge fields a and b such that TKT = −K,
and therefore the timereversal action cannot even be implemented at the level of this
Lagrangian. One manifestation is that this timereversal action has an anomaly.
Methods and objectives. In order to cover these subtler cases as well, we use an
T2 = id, with the only constraint that it reverses the spin, θ(Ta) = θ(a).
anyon charges, such that for each anyon type a ∈ A its topological spin θ(a) = e2πih(a)
U(
1
) is given. We then consider an arbitrary time reversal action T : A → A such that
∈
Because of this generality, a general timereversal symmetry T can first have a
symmetry localization obstruction [2, 6]. When the obstruction is nonvanishing, the symmetry
of the anyon system is not Z2 = {id, T} but is a 2group obtained by extending this Z2 by
the 1form symmetry group A [7, 8]. When the obstruction vanishes, we can then study
the anomaly of the timereversal symmetry, which is known to be characterized by two
signs Zanomaly(RP4) = ±1 and Zanomaly(CP2) = ±1, which are the partition functions of
the corresponding 3+1d SPT characterizing the anomaly. In the following, we simply use
the word obstruction for the symmetry localization obstruction, and the word anomaly for
1In this paper, we restrict our analysis to 2+1d nonspin TQFTs, i.e. those which do not require any
2Some of the examples mentioned below will need the spin structure to be specified on the spacetime to
be welldefined, but this subtlety does not play a role in the rough discussion in this introduction.
– 2 –
the timereversal anomaly.3 In particular, there is a formula [2]
of as the local eigenvalue of T2 associated to the anyon a.
computing the anomaly from the topological spins θ(a) = ±1 and the local Kramers
degeneracy η(a) = ±1 for timereversalsymmetric anyons a = Ta: η(a) can loosely be thought
Many natural questions then arise, for example: i) What are the allowed form of the
timereversal actions T on abelian anyons A? ii) Are there cases where the obstruction is
nonvanishing? iii) What can be said about the anomalies, assuming that the obstruction
vanishes? This paper is our first step toward answering these questions.
We will see the importance of the group C defined as follows:
C = {a = Ta  a ∈ A}/{c + Tc  c ∈ A}.
In words, this is the group of timereversalsymmetric anyons modulo anyons composed
from an anyon and its time reversal.
When the obstruction vanishes, we will see that
different allowed choices of the local Kramers degeneracy η is classified by this group C.
Also, when the obstruction vanishes, we will see that the anomaly (1.1) can be rewritten as
simplifying a sum over timereversalinvariant anyons in A into a sum over C. We give
a general analysis showing that C = (Z2)2m, and the anomaly (1.3) is the associated Arf
invariant.
We will also analyze two classes of explicit examples in detail: one is when A is odd,
and another is when A = (Z2)N . Among them, we will not find any explicit example whose
timereversal symmetry is obstructed in this paper.
We also carry out a general analysis if it is possible or not to choose a linear function
η on timereversalinvariant anyons valued in {±1} such that the anomaly formula (1.1)
yields ±1. Nonexistence of such an assignment of η was used as a sufficient condition
for the existence of the obstruction in [6], where several nonabelian anyon models with
obstruction were found. We will see below that it is always possible to choose such an η
in the case of abelian anyon models. This supports, but does not prove, the idea that the
timereversal symmetry on abelian anyon models is not obstructed in general.
Finally, in a recent paper [8], it was shown using the anomaly inflow that any unitary
finite group symmetry on any abelian anyon system is not obstructed. These observations
strongly suggest that the timereversal symmetry of an abelian anyon system is never
obstructed. It would be interesting to further investigate if this conjecture holds or not.
3This is not the standard usage in the literature, where they are often both called obstructions or
anomalies. Hopefully our usage is clearer.
– 3 –
(1.1)
(1.2)
(1.3)
Organization of the paper.
The rest of the paper is organized as follows. We start in
section 2 by reviewing the formalism we need. We spell out the defining data of abelian
anyons, discuss three major sources of such systems, and recall the MooreSeiberg data
associated to them. We then explain how to express the timereversal action in this
formalism, and how to compute the obstruction and the anomaly.
In section 3, we study what can be said about general timereversalsymmetric abelian
anyon systems, without using the detailed features of the MooreSeiberg data. We will
see that the anomaly formula (1.1) can be rewritten in terms of a sum over the group C
defined as in (1.3), which is the Arf invariant of θ(a)η(a) on C.
anyon systems. As our preceding analysis will make clear, the situations differ drastically
depending on whether an anyon can be divided by two or not. We study two extreme
cases. Namely, in section 4 we study the case when A is odd. There, a straightforward
argument shows that the theory is necessarily a gauge theory for an abelian group A with
a trivial timereversal action on A, such that A = A2. The obstruction and the anomaly
will vanish automatically. In section 5 we study the case when every element of A is order
two, i.e. when A = (Z2)N . There, a recent mathematical result allows us to enumerate all
possible timereversal actions. We study the obstruction and the anomaly for each of these
cases by direct computations using a computer program. We will not find any case with
In section 4 and 5, we consider concrete cases of timereversal actions on abelian
HJEP07(218)9
obstructions.
2
2.1
Basics of abelian anyons and their time reversal
Defining data of abelian anyons
Let us first review the defining data of abelian anyons in 2+1 dimensions. In this paper
we restrict to the case where the system is nonspin, by which we mean that the system is
welldefined without specifying the spin structure on the manifold.
Three main sources of such theories are U(
1
)N ChernSimons theories, nonabelian
ChernSimons theories at level 1, and gauge theories of finite abelian groups. We prefer to
use a democratic formalism which treat the output of these distinct methods equally.
Following [9–11], we take the defining data of a system of abelian anyons to be the
triple (A, θ, c) where
• the group of charges of anyons A is finite and abelian,
• the topological spin θ : A → U(
1
) is a nondegenerate homogeneous quadratic
function,
• and the chiral central charge c is an integer satisfying the Gauss sum constraint.
Here, a function θ : A → U(
1
) is called quadratic if the braiding phase defined by
B(a, b) := θ(a + b)θ(a)−1θ(b)−1
(2.1)
– 4 –
The Gauss sum constraint is the condition
examples.
2.2
2.2.1
Examples of abelian anyons
Abelian ChernSimons
θ(na) = θ(a)n2 .
X θ(a) = p
Ae2πic/8.
A
is bilinear; it is called nondegenerate if B is nondegenerate; and it is called homogeneous if
This constraint determines c modulo 8. As this description is rather abstract, let us discuss
The first set of examples are U(
1
)N ChernSimons theories. The action is given in Euclidean
signature by the formula
S = i KIJ Z
2π
M
aI daJ
where aI for I = 1, . . . , N are U(
1
) gauge fields on the 3d manifold M and KIJ is a
symmetric matrix. For this Lagrangian to be welldefined on 3d oriented manifolds without
specifying the spin structure, KIJ needs to be an integral matrix such that the diagonal
entries are even. We call such matrix an even integral matrix. The analysis of abelian
ChernSimons theory using the matrix K is often called the Kmatrix formalism in the
condensedmatter literature.
The charges of the anyons are characterized by a finite abelian group
and an anyon a = (aI ) ∈ A has the topological spin
A = ZN /KZN ,
θ(a) = e2πi· 12 aI (K−1)IJ aJ .
The right hand side is a welldefined function of a + KZN thanks to the fact that K is even
and integral. The braiding phase between two anyons a, b ∈ A is given by
B(a, b) = e2πi·aK−1b
which is bilinear, symmetric and nondegenerate. The topological central charge c of the
system is the signature of K, namely the difference of the number of positive eigenvalues
and the number of negative eigenvalues of K, and it is a classic mathematical result that
the Gauss sum constraint (2.3) is satisfied.
2.2.2
Nonabelian ChernSimons at level 1
The second set of examples are nonabelian ChernSimons theories Gk with level k, when
G is simplylaced and k = 1. In fact such a theory is equivalent to a U(
1
)N ChernSimons
theory where N is the rank of G and the associated KIJ defines the root lattice of G.
In particular, when G = E8, the root lattice is equivalent to the weight lattice, and the
group of anyon is trivial, A = 0. This theory still has a nontrivial chiral central charge c = 8.
– 5 –
(2.2)
(2.3)
(2.4)
(2.5)
(2.6)
(2.7)
HJEP07(218)9
The third set of examples are gauge theories of finite abelian group A. For these theories,
the group of anyons is A = A ⊕ Aˆ where Aˆ is the Pontrjagin dual of A, namely the group
of 1dimensional representations of A. Physically, an anyon a ∈ A carries a magnetic flux
specified by a and an anyon χ ∈ Aˆ carries an electric charge specified by χ. The topological
spins are given by
θ(a + χ) = χ(a),
and the chiral central charge is zero.
This third set can be further generalized by introducing a nonzero DijkgraafWitten
action ω ∈ H3(A, U(
1
)); the anyons become nonabelian in general, but they remain abelian
when ω satisfies a certain simplifying condition [12, 13]. This simplifying condition is
automatically met when A = Zn.
(2.8)
HJEP07(218)9
2.2.4
Universality of abelian ChernSimons constructions
In [9] the quantization of U(
1
)N ChernSimons theories was analyzed carefully, and two
different Lagrangians leading to the same triple (A, θ, c) are shown to be dual, i.e. are
equivalent as quantum mechanical theories. Conversely, it is a classic mathematical
result [14–16] that any triple (A, θ, c) comes from an even integral lattice (ZN , KIJ ). Here
we note that the chiral central charge c is determined by θ by the Gauss sum relation (2.3)
mod 8, and the mod 8 part can be freely changed by tensoring the E8 level 1 theory or its
orientation reversal.
This means that we do not lose any generality by assuming that the anyon system we
consider in fact comes from an abelian ChernSimons theory. This point of view might be
mentally reassuring to some of the readers.
2.3
MooreSeiberg data of abelian anyons
As recalled above, an anyon system is characterized by the triple (A, θ, c). In order to
perform the computation as a 3d topological quantum field theory, we need the MooreSeiberg
data [17, 18], or equivalently, we need to describe anyons as a modular tensor category.
The MooreSeiberg data of abelian anyons forming a cyclic group Z
n were discussed in
appendix E of [17]; the data for the general case were discussed in detail e.g. in [10, 19].
We quickly recall salient features below.
In general, a 3d topological quantum field theory is specified by morphisms
(2.9)
(2.10)
describing the fusion and
describing the halfbraiding, where a, b, c are three arbitrary anyons, satisfying the
pentagon and hexagon relations.
F (a, b, c) : a ⊗ (b ⊗ c) −∼→ (a ⊗ b) ⊗ c
R(a, b) : a ⊗ b −→ b ⊗ a
∼
– 6 –
The halfbraiding R(a, b) and the data θ(a), B(a, b) are related by the formula
There are U (a, b) such that (F, R) = (U.F, U.R). This happens if and only if4 there is
and we can define
correspond to the same θ and B.
a function β : A → U(
1
) such that
(U.F )(a, b, c) := U (b, c)U (a + b, c)−1U (a, b + c)U (a, b)−1F (a, b, c),
(U.R)(a, b) := U (a, b)−1U (b, a)R(a, b).
The pairs (F, R) and (U.F, U.R) are considered physically equivalent. In particular, they
U (a, b) = β(a)β(b)/β(a + b).
(2.16)
Existence and essential uniqueness. It is known that for any θ where θ is a
homogeneous quadratic function on A, there is a unique equivalence class of (F, R). This can be
seen as follows.
First we show that an explicit representative can be constructed by giving an ordered
basis on A (see e.g. [10, 11, 20]). Namely, we fix a decomposition A ≃ Z
and fix generators gi of Zni . We call such a choice of an ordered basis by O. Then an
arbitrary element a ∈ A can be written as a = P aigi, where 0 ≤ ai < ni. We then define
n1 × n2 × · · · ×
Znk ,
Z
(2.11)
(2.12)
(2.13)
(2.14)
(2.15)
(2.17)
(2.18)
Required relations and equivalences.
For a system of abelian anyons, these
morphisms F (a, b, c) and R(a, b) can be thought of simply as phases ∈ U(
1
). Then the pentagon
relation is
F (a, b, c + d)F (a + b, c, d) = F (b, c, d)F (a, b + c, d)F (a, b, c)
and the hexagon relations are
R(a, b + c) = F (a, b, c)−1R(a, b)F (b, a, c)R(a, c)F (b, c, a)−1,
R(a + b, c) = F (a, b, c)R(b, c)F (a, c, b)−1R(a, c)F (c, a, b).
θ(a) = R(a, a),
B(a, b) = R(b, a)R(a, b).
The phases F (a, b, c) and R(a, b) are not basis independent in the following sense. For
each pair of anyons a, b, we can introduce phases
U(
1
) ∋ U (a, b) : a ⊗ b −→ a ⊗ b
∼
(FO)(a, b, c) :=
(RO)(a, b) :=
Y θ(gi)aibi Y B(gi, gj )aibj .
i<j
θ(gi)niai if bi + ci ≥ ni
if bi + ci < ni,
,
Y
(
1
i
i
– 7 –
4The if part can be checked by a simple computation. To show the only if part, suppose one is given
such a U . From F = U.F , U is a twococycle, and determines an extension 0 → U(
1
) → Aˆ →p A → 0, such
that for a section s : A → Aˆ we have s(a)s(b) = U (a, b)s(a + b). From R = U.R, we see U (a, b) = U (b, a).
Therefore Aˆ is abelian. We now construct another section t : A → Aˆ as follows. We pick an ordered basis,
and choose t(gi) such that p(t(gi)) = gi and t(gi)ni = 1. Then, for a = P aigi, we define t(a) = Qi t(gi)ai .
We can check t(a)t(b) = t(a + b). We finally find β(a) via the relation β(a)t(a) = s(a).
Next, given another pair (F, R) for a given (A, θ) with an ordered basis, there is an
explicit algorithm given in section 2.5 of [20] which produces an appropriate U such that
(U.F, U.R) = (FO, RO).
Other constructions of the MooreSeiberg data. The data (F, R) can also be given
in terms of KIJ for the abelian ChernSimons theory [19, chapter 12], as we describe
below. Since any finite abelian anyon system comes from an abelian ChernSimons theory
as reviewed above, this also provides the existence proof of the MooreSeiberg data for
arbitrary abelian anyon systems.
We denote by A = Λ∗/Λ where Λ∗ = ZN , Λ = KZN ⊂ Λ∗. We denote the inner
product on Λ∗ by hα, βi = αI (K−1)IJ βJ . Note that its restriction makes Λ an even
We now fix a function δ : L × L → U(
1
) satisfying the following condition:
δ(α, β)
δ(β, α)
= (−1)hα,βi
if α, β ∈ L0.
This is the socalled cocycle factor, which also appears in the careful definition of the vertex
operators of 2d chiral bosons on Rn/Λ, see e.g. [21, p.19]. An example of such a δ is given by
δ(α, β) := eπi PI<J (K−1α)IKIJ (K−1β)J ,
but any other choice will do.
For each anyon a ∈ A, we fix a lift αa ∈ Λ∗. We then define
and
Rδ(a, b) =
δ(αb, αa)
δ(αa, αb) eπihαa,αbi
Fδ(a, b, c) =
Rδ(a + b, c)
Rδ(a, c)Rδ(b, c)
.
The pair (Fδ, Rδ) defined above satisfies the required properties (2.11), (2.12) and (2.13).
There is yet another way to describe the MooreSeiberg data of an abelian anyon
system (A, θ), which is intermediate between the one using the ordered basis and the one
using the abelian ChernSimons system. This can be found in appendix D of [22].
Relation to the anomaly of 1form symmetries.
The discussion in the last
paragraph establishes that the set of the equivalence classes of the pair (F, R) is in onetoone
correspondence with the set of homogeneous possiblydegenerate quadratic functions on
A. This set is known to be equal to
H4(K(A, 2), U(
1
))
from an old work of Eilenberg and Mac Lane [23].
More recently [24, 25], it was noticed that the cohomology group Hd+2(K(A, k +
1), U(
1
)) characterizes the anomaly of a (d + 1)dimensional system with the kform
symmetry A. Therefore, the object (2.23) classifies the anomaly of 1form symmetry A in 2+1
– 8 –
(2.19)
(2.20)
(2.21)
(2.22)
(2.23)
dimensions. In our case, the point is to regard the group A of abelian anyons as giving the
1form symmetry of the system. Then, the worldlines of abelian anyons labeled by elements
of A define a 1cycle in Z1(M3, A) which acts as the background gauge field for the 1form
symmetry A, and the topological spins θ and the braiding phases B describe the change in
the phase of the partition function as we change the 1cycle in Z1(M3, A) keeping its
homology class in H1(M3, A). Therefore, they describe the anomaly of the 1form symmetry A.
S and T matrices. The discussions in this subsection up to this point did not require
the nondegeneracy of B; in particular, the identification of the equivalence classes of
(F, R) with (2.23) needs homogeneous quadratic functions θ which lead to degenerate B,
for example θ(a) ≡ 1. Therefore, the preceding discussions are more about the structure
of the oneform symmetry A in 2+1 dimensions.
For (A, q) and the associated data (F, R) to actually describe a topological quantum
field theory, we need the nondegeneracy of B. In this case, the modular matrices are given
by
Sab =
1
A1/2 B(a, b),
Tab = e−2πic/24δabθ(a).
(2.24)
S is invertible if and only if B is nondegenerate: the nondegeneracy means that B(a, b)
is a character table of the abelian group A.
2.4
The time reversal, the obstruction and the anomaly
Group actions on general, possibly nonabelian anyons were discussed in detail in [1, 26]
from mathematical and condensedmatter points of view. The equations discussed there
were rather cumbersome. Here we restrict our attention to the action of time reversal on
abelian anyons.
Time reversal T and associated objects U , β and Ω. We denote the action of
timereversal on the anyons by
which we require to satisfy T2 = id. We require
T : A → A
θ(Ta) = θ(a).
MooreSeiberg data (TF, TR) by the formula
We fix the MooreSeiberg data (F, R) for (A, θ). Let us now define the timereversed
TF (a, b, c) := F (Ta, Tb, Tc),
TR(a, b) := R(Ta, Tb).
The pair (TF, TR) also forms a MooreSeiberg data for (A, θ). Therefore, there are phases
U (a, b) such that
(TF, TR) = (U.F, U.R).
where we remind the reader that the right hand side is defined in (2.15).
Note that we trivially have (F, R) = (TTF, TTR). Computing the right hand side
using (2.28) twice, we have
(F, R) = (κ.F, κ.R)
where
κ(a, b) := U (Ta, Tb)U (a, b).
– 9 –
(2.25)
(2.26)
(2.27)
(2.28)
(2.29)
Therefore, there should be phases β(a) as in (2.16) such that
We now define
U (Ta, Tb)U (a, b) = β(a)β(b)/β(a + b).
Ω(a) := β(Ta)/β(a).
Ω˜ = Ω − ν + Tν.
[Ω] ∈ A/(1 −
T)A.
Ω = ν −
βˆ(a) := β(a)γ(Ta)γ(a),
Ωˆ = Ω.
β˜(a) := β(a)ν(a)
Therefore, the choiceindependent content is the equivalence class
Using (2.30), one finds that Ω is linear, i.e. Ω(a + b) = Ω(a)Ω(b).
Choices in the construction.
Recall that we started from T, from which we got U
via (2.28), from which we got β via (2.30), from which we got Ω via (2.31). There are
HJEP07(218)9
certain indeterminacies at each stage.
If U satisfies (2.28),
Uˆ (a, b) = U (a, b)γ(a)γ(b)/γ(a + b)
for any γ also satisfies the same equation, as discussed around (2.16). Correspondingly, β
is changed but Ω is unchanged:
(2.30)
(2.31)
(2.32)
(2.33)
(2.34)
(2.35)
(2.36)
(2.37)
(2.38)
equally solves the same equation if ν is linear, i.e. if ν(a + b) = ν(a)ν(b). This changes
Ω(a) to
Ω˜ (a) := Ω(a)ν(Ta)/ν(a).
The obstruction [Ω].
Now, we note that any linear map f : A → U(
1
) is realized as
f (a) = B(a, f ) for some f ∈ A, since B is assumed to be nondegenerate. Therefore, Ω(a)
corresponds to an element Ω ∈ A
. Similarly, ν appearing in (2.34) and (2.35) was also
assumed to be linear, and therefore we have a corresponding element ν ∈ A, and we have
We call this element the obstruction. In other words, the obstruction vanishes [Ω] = 0 if
and only if we can solve the following equation:
It is known that [7, 8, 27] when the obstruction is nonvanishing, the group Z2 = {1, T} is
not quite the group of symmetries of the system, but rather is nontrivially extended by
the 1form symmetry A
.
In passing, we mention that it is not at all clear whether the obstruction generally
vanishes in this description. Some subcases when it vanishes can be established. In
section 5.1.1 of [26] and in the appendix of [28], the obstruction was shown to vanish when
A is odd. Similarly, the obstruction can be shown to vanish when G is odd. Also, the
obstruction obviously vanishes when one can find an abelian ChernSimons realization such
that T is actually an order2 symmetry of KIJ which furthermore preserves the cocycle
factor (2.19).
The object η.
When the obstruction vanishes, the group G acts as a genuine symmetry.
In this case, there is a choice of β(a) such that Ω = 0 ∈ A
. To emphasize that this is a
special case, it is useful to denote such a choice of β(a) by a different letter η(a). More
explicitly, η(a) needs to satisfy
U (Ta, Tb)U (a, b) = η(a)η(b)/η(a + b),
η(a)η(Ta) = 1.
We immediately see that
η(a) = ±1,
η(a + b) = η(a)η(b)
if Ta = a and Tb = b. This quantity η(a) for Ta = a has the interpretation of the local
eigenvalue of T2, and sometimes called the local Kramers degeneracy [2].
Note that the choice of η is not unique. We can replace η following (2.34) as follows:
η˜(a) := η(a)ν(a) = η(a)B(ν, a).
This solves (2.39) if and only if
If we replace U by Uˆ in (2.32), η is replaced by
which satisfies the relations (2.39) automatically. In particular, when γ(a) = B(γ, a), the
change (2.43) corresponds to the change (2.41) with
We physically identify two choices of η different by this type of ν. In other words, physical
equivalence classes of allowed η are parameterized by ν satisfying (2.42) modulo ν given
by (2.44); two allowed η’s are different by an element in
C := Ker(1 −
T)/ Im(1 + T).
This is the group which we introduced in (1.2) in the introduction. Mathematically, we
say that the set of η is a torsor over C.
Note that changing η using ν does not change its value on Im(1 + T), as can be checked
easily. In fact η(c + Tc) can be written in terms of B. To see this, one first sets a = c,
b = Tc in the first equation of (2.39) to show
ν = Tν.
ηˆ(a) := η(a)γ(Ta)γ(a),
ν = γ + Tγ.
η(c + Tc) = U (Tc, c)U (c, Tc)−1,
(2.39)
(2.40)
(2.41)
(2.42)
(2.43)
(2.44)
(2.45)
(2.46)
where we used the second equation of (2.39). Now, the explicit form of the equation
TR = U.R in (2.28) is
Setting a = c, b = Tc again, we find
Combining with (2.46), we conclude that
R(Ta, Tb) = U (a, b)−1U (b, a)R(a, b).
U (Tc, c)U (c, Tc)−1 = R(c, Tc)−1R(Tc, c)−1 = B(c, Tc)−1.
η(c + Tc)B(c, Tc) = 1.
(2.47)
(2.48)
(2.49)
HJEP07(218)9
Anomalies. In general, the time reversal symmetry of nonspin (d + 1) dimensional
systems are characterized by the phase given by the symmetry protected topological phase
in (d + 2) dimensional unoriented spacetime, which is a homomorphism to U(
1
) from the
cobordism group Ωdu+no2riented [29–32]. In our case Ωunoriented = Z
4
2 × Z2 is generated by
RP4 and CP2, and therefore the anomaly is characterized by two signs Zanomaly(RP4) and
Zanomaly(CP2).
These two signs were computed in [2] for general 3d nonspin topological quantum field
theories; see also [33–35] for the spin case. For abelian anyons, the formulas of [2] becomes
1
Since Zanomaly(CP2) is uniquely fixed in terms of c, we will only be interested in
Zanomaly(RP4) below.
3
3.1
Time reversal and the anomaly formula
General properties to be established
In this section, we study the property of the time reversal T on general abelian anyon
systems. Since we do not have a good control of the MooreSeiberg data (F, R) in the
general case, we will use only the following information in this section, namely:
• The time reversal T : A → A satisfies T2 = id,
• θ(Ta) = θ(a)−1, and B(a, b) := θ(a + b)θ(a)−1θ(b)−1 is nondegenerate,
• η(a) = ±1 and η(a + b) = η(a)η(b) if a and b are T invariant, see (2.40).
We find the following general properties:
1. The group C = Ker(1 −
T)/ Im(1 + T) introduced in (2.45) is = (Z2)n for some n.
2. The nondegenerate pairing B : A × A → U(
1
) restricts to a nondegenerate pairing
on B : C × C → {±1}. Furthermore, n is even, n = 2m.
3. There is an obstruction if the summand q(a) := θ(a)η(a) on Ker(1 − T) does not
restrict to a function on C.
4. There is always a choice of η(a) such that q(a) = θ(a)η(a) restricts to a function on
5. If q(a) = θ(a)η(a) restricts to a function on C, then the anomaly Zanomaly(RP4) is
the Arf invariant of q : C → {±1}. In particular, there are 2m−1(2m + 1) choices of
η’s for which the anomaly vanishes, and 2m−1(2m
anomaly is nonvanishing.
− 1) choices of η’s for which the
In [6], the nonexistence of the assignment η so that Zanomaly(RP4) = ±1 was considered
as a simple sufficient condition to see if a group action on anyon systems is obstructed;
several obstructed nonabelian models were found there that way. Our Property 4 here
HJEP07(218)9
then means that we cannot find any obstructed abelian anyon systems with the technique
of this section 3.
Derivations of the properties
Let us show these properties.
Property 1.
We start from a trivial observation that
which simply follows from T2 = 1. We now consider the group
Im(1 + T) ⊂ Ker(1 −
T)
C = Ker(1 −
T)/ Im(1 + T),
B(a, Tb) = B(Ta, b)−1.
B((1 + T)a, b) = B(a, (1 −
T)b).
(1 −
T)a = 0. Therefore C = (Z2)n for some n.
Property 2.
We first note that
i.e. the group of timereversal invariant anyons modulo anyons which are composites of
an anyon and its time reversal. Every element in C is order two, since 2a = a + Ta if
Therefore, we have
Therefore
Ker(1 −
T) ⊂ [Im(1 + T)]⊥,
Ker(1 + T) ⊂ [Im(1 −
T)]⊥.
Using the nondegeneracy of B, we have
 Ker(1 −
T) ≤ A/ Im(1 + T),
 Ker(1 + T) ≤ A/ Im(1 −
T)

Now, we obviously have
ities:
A/ Ker(1 + T) =  Im(1 + T),
A/ Ker(1 −
T) =  Im(1 −
T) .

The relations (3.6) and (3.7) together shows that the inclusions in (3.5) are actually
equalKer(1 −
T) = [Im(1 + T)]⊥,
Ker(1 + T) = [Im(1 −
T)]⊥.
(3.1)
(3.2)
(3.3)
(3.4)
(3.5)
(3.6)
(3.7)
(3.8)
The relation (3.1) means that B(a, b) on A descends to bilinear forms on C and the
relation (3.8) means that B thus defined on C are actually nondegenerate.
We now recall the standard fact that any nondegenerate pairing B on (Z2)n can be
put into either of the following two forms:5
• Symplectic: there is a basis u1, v1,, u2, v2, . . . , um, vm ∈ A with n = 2m such that
B(ui, uj ) = B(vi, vj ) = 1,
B(ui, vj ) =
(
1
(i 6= j)
.
−1 (i = j)
(3.9)
(3.10)
(3.11)
HJEP07(218)9
• Orthogonal: there is a basis u1, . . . , un ∈ A such that
B(ui, uj ) =
(
1
(i 6= j)
.
−1 (i = j)
When n is odd, the orthogonal complement of the vector (1, 1, . . . , 1) has a symplectic
structure as given above.
In our case, B on C satisfies B(a, a) = +1 for all a ∈ C, since if we regard a ∈ Ker(1−T),
B(a, a) = θ(2a)/θ(a)2 = θ(a)2 = 1; the last equality follows since θ(a) = θ(Ta) = θ(a).
Therefore, B should be of the symplectic type, so that n is even: n = 2m. Therefore
A =  Im(1 + T)2 · 22m.
Property 3. Let us now try to evaluate the anomaly (2.50)
Let us perform the sum over Ker(1 −
T) by first summing within a coset a′ ∈ a + Im(1 + T)
and then over C. We would like to relate, then, θ(a + c + Tc)η(a + c + Tc) and θ(a)η(a).
5The proof goes as follows, see e.g. [36, Theorem 2.1]. Consider a nondegenerate pairing B on a
finitedimensional Z2vector space V . As a zeroth step, we note that any B(a, b) = ±1, because B(a, b)2 =
B(2a, b) = 1. Then, as a first step, we show that V is a direct sum of an orthogonal part and a symplectic
part. To see this, if there is a element x ∈ V such that B(x, x) = −1, one takes the orthogonal complement
of x, and repeat the process. Eventually, there is no x ∈ V such that B(x, x) = −1.
Then, pick a
nonzero x ∈ V randomly. From nondegeneracy, there is a y ∈ V such that B(x, y) = −1. Then we
take the orthogonal complement of x and y, and repeat the process. As a second step, one shows that
V3 = (Z2)
3 with the orthogonal B can in fact be split into a onedimensional orthogonal vector space plus
a twodimensional symplectic space. This can be done by taking the orthogonal complement of the vector
(
1, 1, 1
) ∈ V3. This completes the proof.
6Note that in this section we are analyzing the anomaly without actually using the MooreSeiberg data.
It is still useful to recall that we saw in (2.49) that B(c, Tc)η(c + Tc) = 1 when the obstruction can be
shown to vanish using the MooreSeiberg data.
over a′ ∈ a + Im(1 + T) simply vanishes, and we have
the right hand side is a homomorphism Im(1 + T) → {±1}. If this is nontrivial, the sum
Zanomaly(RP4) = 0,
(3.14)
which should not happen if there is no obstruction.
Property 4.
Resuming the discussion, let us ask if we can choose an η such that the
right hand side of (3.13) is a constant = 1. For this purpose, we regard f (c) := B(c, Tc)
to a onedimensional representation of f : Ker(1 − T) → U(
1
). Let q(a) = θ(a)f (a) for
a ∈ Ker(1 − T). By construction, this function on Ker(1 − T) is constant on a coset
a′ ∈ a + Im(1 + T), and therefore restricts to a function C → U(
1
) satisfying q(a + b) =
q(a)q(b)B(a, b) on C. Since B on C is of symplectic type, q(a) = ±1.7 Since θ(a) = ±1, we
conclude that f (a) = ±1. Therefore we can use this f (a) on Ker(1 −
satisfies every condition which should be satisfied by η.
T) as η(a), since f
T), this f can be extended (nonuniquely)
Property 5.
As discussed, without an obstruction, θ(a′)η(a′) should be constant on
a′ ∈ a + Im(1 + T). Let us denote this function by q(a) = θ(a)η(a) : C → Z2. Then we have
(3.15)
(3.16)
Zanomaly(RP4, η) =
1
We note that the function q(a) satisfies B(a, b) = q(a+b)q(a)−1q(b)−1 on C. Therefore,
q is a nondegenerate homogeneous quadratic function on C ≃ (Z2)2m. Such a function q
is known as a quadratic refinement of B, and for such a q, the right hand side of (3.15) is
known as its Arf invariant, which is known to take values in {±1}:8
Zanomaly(RP4, η) = Arf q.
It is a standard result that for C = (Z2)2m there are 2m−1(2m+1) choices of q’s for which
the Arf invariant is +1, and there are 2m−1(2m
−1) choices of q’s for which the Arf invariant
is −1. This in particular means that if the obstruction vanishes, there is at least one
assignment of η which makes the system free of the timereversal anomaly Zanomaly(RP4).
7This can be shown by actually constructing one quadratic refinement q for B in the standard form.
This turns out to take value in {±1}. Every other q is obtained by multiplying it by a homomorphism
C → U(
1
) which is necessarily valued in {±1}, the statement follows.
8Another place where the Arf invariant appears is in the description of the spin structure on a Riemann
surface [37, 38]. Briefly, for a Riemann surface Σ, we let C := H1(Σ, Z2), and B(a, b) = RΣ a ∪ b for a, b ∈ C.
This B is nondegenerate, and moreover, B(a, a) = +1. We define q : C → Z2 so that q(a) is +1 / −1 if the
spin structure is NeveuSchwarz / Ramond around a nonintersecting loop representing the Poincar´e dual
to a, respectively. This function q is known to satisfy q(a + b) = q(a)q(b)B(a, b), and its Arf invariant is
defined as the right hand side of (3.15). The spin structure is called even or odd depending on whether the
Arf invariant is +1 or −1.
3.3
time reversal T to be an order2 operation on A .
′
At this level of generality, it is not difficult to extend the analysis to abelian anyon systems
which depends on the spin structure. The main difference is that among the anyons there
is a special anyon, sometimes called the transparent fermion f ∈ A such that 2f = 0 and
θ(f ) = −1. Anyons in A such that B(f, a) = 1 are in the NeveuSchwarz sector ANS, while
those with B(f, a) = −1 are in the Ramond sector AR. Then, for any anyon a ∈ ANS, we
have af ∈ ANS and θ(af ) = −θ(a). It is known that η(af ) = −η(f ).
Since a and af always appear in pairs, we consider physically distinct anyons in the
NeveuSchwarz sector to be labeled by A′ := ANS/{0, f }; this is why f is called transparent.
The braiding B descends to a nondegenerate pairing on A′, and we then require that the
For systems which do not feel the spin structure, the anomaly Zanomaly(RP4) = ±1 as
we reviewed above. For systems which do feel the spin structure, the anomaly is in general
given by
Zanomaly(RP4) = e2πiν/16
for an integer ν modulo 16. A generalization of the anomaly formula for this was found
in [33, 35] and is given for abelian anyons by
Now the sum is over timereversal invariant anyons in A′.
Our analysis for the nonspin case can be repeated up to Property 3 without any change
except the replacement of A by A′ everywhere. The only additional change in Property 5
is that q(a) = θ(a)η(a) is now a function q : C → {±1, ±i}, and we still have (3.15). The
right hand side in this case is known as the BrownArf invariant = e2πik/8 for an integer
k modulo 8.9 Comparing with (3.17), we conclude ν = 2k, that is, we found that the
timereversal anomaly of abelian anyons is always an even integer modulo 16.
4
Case study I: when A is odd
reversal action T : A → A
Let us discuss the case when A is odd. We first determine the standard form of the time
, which satisfies T2 = id. A greatly simplifying feature is that
when A is odd, one can always divide an anyon by two, in the sense that for any a ∈ A
there is a unique b such that a = 2b. We denote such this element b by a/2. We note that
(Ta)/2 = T(a/2).
Then any a ∈ A is a sum a = a++a− such that Ta± = ±a±, since we can explicitly take
a± = (a ± Ta)/2.
(4.1)
This means that A is a direct sum A = A+ ⊕ A−.
9The BrownArf invariant appears in the description of the pin− structure on a possibly nonorientable
Riemann surface, see e.g. section 3 of [39] or the appendix of [40]. The sum (3.15) is also a special case of
the general Gauss sum (2.3), since (C, q) satisfies all the mathematical conditions to be an ordinary abelian
anyon system.
Now let us determine B on A compatible with this action of T. For elements a+, b+ ∈
A+, let c+ = a+/2. Then we have
B(c+, b+) = B(Tc+, Tb+)−1 = B(c+, b+)−1
and therefore B(a+, b+) = 1. We similarly have B(a−, b−) = 1 for arbitrary a−, b− ∈ A−.
In order for the braiding B to be nondegenerate on A = A+ ⊕ A−, this means that the
nontrivial pairing should happen between A+ and A−. Equivalently, A+ = Ad−, where Gˆ
for an abelian group G denotes its Pontrjagin dual. Therefore we have B(a+, b−) = a+(b−),
where a+ is now regarded as a homomorphism A− → U(
1
).
A compatible θ on A is then determined as follows: for any a± ∈ A±, we take c± =
a±/2 ∈ A±. We have
and therefore θ(a±) = 1. From this, we easily conclude that
θ(c±) = θ(Tc±)−1 = θ(±c±)−1 = θ(c±)−1
θ(a+ + b−) = a+(b−).
Comparing with the discussion in section 2.2.3, we find that this is a gauge theory with
by the time reversal T.
finite abelian gauge group A = A+ with a trivial action of the timereversal. The anyons
labeled by Aˆ = A− are Wilson lines. For consistency, an anyon a ∈ Aˆ = A− is sent to −a
We can easily see that there is no obstruction and there is no anomaly. To see that there
is no obstruction, we pick an ordered basis in A = A+ and then a corresponding ordered
basis in Aˆ = A−. From the explicit formulas (2.17) and (2.18) of the MooreSeiberg data
(FO, RO) in this ordered basis, we see that
(TFO, TRO) = (FO, RO).
Then U can be taken to be identically 1, and therefore the obstruction vanishes. Then
η(a) is a linear function on A which is = ±1 if a = Ta, i.e. if a ∈ A = A+. This
is identically = +1 since any a is divisible by 2. By the anomaly formula, we see that
Zanomaly(RP4) = +1, i.e. the system is nonanomalous.
The group C is trivial. This alone allows us to conclude that the anomaly vanishes,
using our general analysis given in the last section.
5
Case study II: A = (Z2)
N
A ≃ (Z2)N .
In the previous section we studied the case where A was odd. What made the analysis
straightforward was that we can always divide an anyon by two. In this section we consider
the opposite extreme case, where any anyon a ∈ A satisfies 2a = 0. This means that
(4.2)
(4.3)
(4.4)
(4.5)
We first study all possible actions of T, compatible with the braid pairing B. This was
recently carried out with a different motivation in [36], whose results we summarize below.
Recall that any nondegenerate pairing B on A = (Z2)N is given either by a symplectic
one or an orthogonal one, as we discussed around (3.9), (3.10). Second, all possible forms
of T were classified for both symplectic and orthogonal B in [36]. Note that, since B(a, b) =
±1, the condition for T is that B(Ta, Tb) = B(a, b)−1 = B(a, b).
When B is symplectic, T with respect to the standard basis is a direct sum of the three
matrices:
I =
"
When B is orthogonal, the situation is more complicated. Again, we always choose
the standard basis as in (3.10). When N is odd, any T action fixing B is known to fix
the vector ω = (1, · · · , 1), i.e. Tω = ω. This is because ω is uniquely characterized by
the condition B(a, a) = B(a, ω) for all a ∈ A
symplectic structure, and T on it is given by the direct sum of I, J and M as above.
. The orthogonal complement to ω carries a
Let us move on to the case when N = 2n is even. An important operation is a Z
operation on n × n matrices with entries ∈ Z2, defined by
m(M )ij = 1 − Mij .
One can show that when M 2 = 1, m(M )2 = 1. This operation m is called the mirror
in [36]. Any T can then be conjugated to exactly one of the following forms:
I⊕(n−k)
⊕ J ⊕k
m I⊕(n−k)
m I⊕(n−k−1)
⊕ J ⊕k
⊕ J ⊕k
⊕ J
and their mirrors (1 ≤ k ≤ n − 1),
and their mirrors (0 ≤ k ≤ n − 1),
(1 ≤ k ≤ n − 2).
The last cases are conjugate to mirrors of their own.
simply need to study the following cases:
Clearly, there is no need to study T which is given by a direct sum. Therefore, we
I,
J,
m(I⊕(n−k)
⊕ J ⊕k),
m(m(I⊕(n−k)) ⊕ J ⊕k).
J ⊕k) with even k.
We will see below that there are no compatible θ for m(J ⊕l) with odd l and m(m(I⊕n−k) ⊕
We will tabulate the results we obtained by explicit computations in the following
subsections. The computations are done as follows.
B and T. We fix an ordered basis g1, . . . , gn of A
We start from A with a specified
. We first classify all compatible θ.
For each θ, we compute (F, R) in the standard basis.
We then find U which satisfies
(TF, TR) = (U.F, U.R) using the algorithm given in section 2.5 of [20].
We then find
one choice of β by solving (2.30), which can be done by setting β(gi) = 1 for all basis
(5.1)
2
(5.2)
(5.3)
(5.4)
elements and finding β(a) for linear combinations; it is guaranteed that there is such a β.
From this β we can easily compute Ω, and the obstruction is checked by whether we can
solve (2.38). We implemented the algorithm we explained above in a computer program,
which is available upon request to the authors.
When the braid pairing B is symplectic
We start by analyzing the symplectic cases.
T = I. Our anyons are A = {0, u, v, u + v} with the pairing B(u, v) = −1, B(u, u) =
B(v, v) = +1. There are four choices of θ : A → U(
1
). Up to the relabeling of anyons, we
can choose either of the two cases:
a)
b)
θ(u) = θ(v) = 1,
θ(u) = θ(v) = θ(u + v) = −1.
θ(u + v) = −1,
(5.5)
!
The former corresponds to the standard Z2 gauge theory, also known as the toric code
theory, and is the U(
1
)2 ChernSimons theory with level matrix
whose action is
S = (i/π) R adb.
By an explicit computation, we find that the obstruction vanishes for both choices of θ,
and η is simply a linear function A → U(
1
). There are four choices of η for each case. For
the case (a), three choices give Zanomaly(RP4) = +1 and one choice Zanomaly(RP4) = −1;
this one choice is when η(u) = η(v) = −1. This last case is sometimes called the eTmT
phase in the condensedmatter literature.
For the case (b) too, three choices give Zanomaly(RP4) = +1 and one choice
Zanomaly(RP4) = −1; this choice is when η(u) = η(v) = +1. The cases (a) and (b)
can be distinguished by looking at Zanomaly(CP2) = +1 for (a) and Zanomaly(CP2) = −1
for (b).
T = J . Our anyons are still A = {0, u, v, u + v} with the pairing B(u, v) = −1,
B(u, u) = B(v, v) = +1. There are two choices of θ : A → U(
1
) compatible with u = Tv,
which is again given by (5.5). By an explicit computation, we find that the obstruction
anyons are 0 and u + v, and one finds Zanomaly(RP4) = +1.
vanishes for both choices of θ, and η(u + v) is forced to be −1. Timereversal invariant
T = M . Anyons are generated by u1,2 and v1,2. Up to relabeling u1 ↔ u2 and v1 ↔ v2,
there is only one allowed choice of θ, given by
θ(u1) = +1, θ(u2) = −1, θ(v1) = +1, θ(v2) = −1.
(5.6)
The timereversalinvariant anyons are generated by u1 + u2 and v1 + v2, and the group C
is trivial.
An explicit computation shows that the obstruction vanishes, and there is only one
allowed choice of η which is η(u1+u2) = η(v1+v2) = +1. One finds that Zanomaly(RP4) = +1.
T = m(I⊕2).
allowed choices of θ, given by
Zanomaly(RP4) = −1 may occur.
four allowed choices of θ, given by
(5.7)
(5.8)
(5.9)
θ(u1) = +i, θ(u2) = +i, θ(u3) = +i, θ(u4) = +i;
θ(u1) = +i, θ(u2) = +i, θ(u3) = −i, θ(u4) = −i;
θ(u1) = −i, θ(u2) = −i, θ(u3) = −i, θ(u4) = −i.
The timereversalinvariant anyons are generated by u1 + u2, u1 + u3 and u1 + u4,
and the group C = (Z2)2. An explicit computation shows that the obstruction vanishes,
and there exist various allowed choices of η. One finds that both Zanomaly(RP4) = +1 and
T = m(I ⊕ J ). Anyons are generated by u1,2 and v1,2. Up to relabeling, there are
Let us move on to the case where B is orthogonal. We start by analyzing a few simple
cases.
This means θ(a) = ±i. The time reversal action is Ta = a, and therefore we cannot have
θ(a) = θ(Ta). Therefore this is inconsistent as a nonspin theory.10 Since T = I is just two
We will start with the simplest case when A = Z2 = {0, a}, B(a, a) = −1.
copies of this system, this is also inconsistent.
T = J . Our anyons are A = {0, u, v, u + v} with the pairing B(u, u) = B(v, v) = −1
and B(u, v) = +1. Up to the relabeling of anyons, there is only one choice of θ compatible
with this T:
θ(u) = +i,
θ(v) = −i.
Timereversalinvariant anyons are 0 and u + v, with the group C being trivial. By an
explicit computation, we find that the obstruction vanishes, and η(u + v) = −1. We find
Anyons are generated by u1,2,3,4. Up to relabeling, there are three
θ(u1) = +i, θ(u2) = +i, θ(v1) = +i, θ(v2) = +i;
θ(u1) = +i, θ(u2) = +i, θ(v1) = −i, θ(v2) = −i;
θ(u1) = −i, θ(u2) = −i, θ(v1) = +i, θ(v2) = +i;
θ(u1) = −i, θ(u2) = −i, θ(v1) = −i, θ(v2) = −i.
The timereversalinvariant anyons are generated by u1 + u2 and v1 + v2, and the group
various allowed choices of η. One finds that Zanomaly(RP4) = +1.
C is trivial. An explicit computation shows that the obstruction vanishes, and there exist
10As a spin theory this is consistent as we discussed in section 3.3, and describes the semionfermion
system, which has Zanomaly(RP4) = e±2πi/8.
and one finds that Zanomaly(RP4) = +1.
interesting to study if this is the case or not.
see, for some choice of T there is no compatible θ.
As already mentioned, we implemented the algorithm in a program and studied all the
choices (5.4) from smaller n to larger n. We did not find any case where the timereversal
This leads us to suspect that the timereversal symmetry
→ A on an abelian anyon system might be always unobstructed. It would be
Below, we study various choices of T in (5.4) using the anomaly formula. As we will
T = m(I⊕k).
Anyons are generated by u1, · · · , u2k. A compatible θ is
for even k, and
for odd k.
θ(ul) = +i
(l = 1, · · · , 2k)
"
+i (l 6= 2k)
−i (l = 2k)
T = m(J ⊕2).
Completely the same argument as T = m(I ⊕ J ) case goes through,
(5.10)
(5.11)
(5.12)
±
Tinvariant anyons consist of the sum of even number of anyons, since Tui 6= ui and
T(ui +uj ) = ui +uj . The image of 1+T are either 0 or u1 +· · · u2k. Therefore C = (Z2)2k−2.
Let us see what the anomaly formula tells us. η is a {±1}valued linear function on
Tinvariant anyons. Let us arbitrarily extend it to a {±1}valued linear function on the
entire A
η˜(a)θ(a) is either real or purely imaginary, depending on whether a is Tinvariant or not.
. We denote the extension by η˜. We then have η˜(ul)θ(ul) = ±i, and furthermore,
Therefore we have
Zanomaly(RP4) = √22k
1
=
1
2k Re
= 0
X Re η˜(a)θ(a)
a∈A
h
1 + i m 1 − i 2k−mi
+1 k − m ≡ 0 mod 4 ,
k − m ≡ ±1 mod 4 ,
−1 k − m ≡ 2 mod 4
where m is the number of the basis anyon ul such that η˜(ul)θ(ul) = +i.
entire A
Each distinct choice of η on Tinvariant anyons corresponds to two choices of η˜ on the
. Therefore, by a short computation from (5.12), we see that there are 2k−2(2k−1
1) choices of η for which the anomaly formula gives ±1. This agrees with our general
discussion in section 3, see Property 5.
Anyons are generated by u1, v1, · · · , uk, vk. Odd k is inconsistent since
there is no compatible choice of θ. This can be seen as follows. We start from
Noting that both θ(ul) and θ(vl) are ±i, we see
θ(Tul) = θ(vl)−1 Y θ(um)θ(vm).
k
since um + vm is Tinvariant. We then have Qk
get (±1)k = ∓1, which runs into a contradiction when k is odd.
We now introduce cm := θ(um)θ(vm) = θ(um + vm) = ±1, where the last equality follows
m=1 cm = −cl for arbitrary l. Therefore, we
When k is even, a compatible θ is
θ(ul) = θ(vl) = +i (l = 1, · · · , k).
A short computation shows that Tinvariant anyons are the linear combinations of
ui + vi’s, and ui + vi itself is in the image of 1 + T. Therefore the group C is trivial. The
anomaly formula can be evaluated explicitly, using the signs η(ul + vl)θ(ul + vl) = ±1 for
"
+1 k = m ,
0
k 6= m .
Indeed, there is only a single allowed choice of η, which is simply given by η(ul + vl) =
(5.13)
(5.14)
(5.15)
(5.16)
(5.17)
(5.18)
(5.19)
for odd n. We do not repeat the analysis of the anomaly formula, since it is analogous to
Anyons are generated by u1, · · · , u2(n−k) and
w1, x1, · · · , wk, xk. This time, even k is inconsistent, since there is no compatible θ. This
can be seen as follows. We start from
θ(ul) = +i (l = 1, · · · , 2(n − k)),
θ(wm) = θ(xm) = +i (m = 1, · · · , k)
Anyons are generated by u1, · · · , u2(n−k) and
θ(wm) = θ(xm) = +i (m = 1, · · · , k)
θ(ul + vl).
T
m(I⊕(n−k) ⊕ J ⊕k).
w1, x1, · · · , wk, xk. An allowed θ is
for even n and
"
+i (l 6= 2(n − k)),
−i (l = 2(n − k)),
the ones we have already given above.
T
m m(I⊕(n−k)) ⊕ J ⊕k .
θ(Tul) = θ(ul) ·
Y θ(wm)θ(xm)
k
m=1
Using θ(ul) = ±i, we have
Next, we consider
which implies
For odd k, a compatible θ is
for even n, and
"
+i (l 6= 2(n − k))
−i (l = 2(n − k))
,
θ(Twj ) = θ(xj )−1 Y θ(wm)θ(xm) ·
2(n−k)
Y
l=1
θ(ul)
θ(ul) = θ(wj )θ(xj ),
for arbitrary j. In other words, θ(wj )θ(xj ) = ±1 for arbitrary j, and this sign is
independent of j. This contradicts with (5.20) if k is even.
θ(ul) = +i (l = 1, · · · , 2(n − k)),
θ(wm) = θ(xm) = +i (m = 1, · · · , k)
(5.20)
(5.21)
(5.22)
(5.23)
(5.24)
HJEP07(218)9
θ(wm) = θ(xm) = +i (m = 1, · · · , k)
for odd n. Again, we do not repeat the analysis of the anomaly formula.
Acknowledgments
The authors thank Francesco Benini and PoShen Hsin for explaining the content of [8].
Y.L. is partially supported by the Programs for Leading Graduate Schools, MEXT, Japan,
via the Leading Graduate Course for Frontiers of Mathematical Sciences and Physics.
Y.T. is partially supported by JSPS KAKENHI GrantinAid (WakateA), No.17H04837
and JSPS KAKENHI GrantinAid (KibanS), No.16H06335, and also by WPI Initiative,
MEXT, Japan at IPMU, the University of Tokyo.
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.
topological phases, arXiv:1410.4540 [INSPIRE].
[2] M. Barkeshli, P. Bonderson, C.M. Jian, M. Cheng and K. Walker, Reflection and time
reversal symmetry enriched topological phases of matter: path integrals, nonorientable
manifolds and anomalies, arXiv:1612.07792 [INSPIRE].
topological superconductors: surface topological order from vortex condensation, the 16 fold
way and fermionic Kramers doublets, arXiv:1406.3032 [INSPIRE].
[5] N. Seiberg and E. Witten, Gapped boundary phases of topological insulators via weak
coupling, PTEP 2016 (2016) 12C101 [arXiv:1602.04251] [INSPIRE].
[6] M. Barkeshli and M. Cheng, Timereversal and spatial reflection symmetry localization
anomalies in (2 + 1)D topological phases of matter, arXiv:1706.09464 [INSPIRE].
arXiv:1803.09336 [INSPIRE].
hepth/0505235 [INSPIRE].
[9] D. Belov and G.W. Moore, Classification of abelian spin ChernSimons theories,
[10] S.D. Stirling, Abelian ChernSimons theory with toral gauge group, modular tensor categories
and group categories, Ph.D. thesis, Math Dept., Texas U., Austin, U.S.A., (2008)
[arXiv:0807.2857] [INSPIRE].
[11] A. Kapustin and N. Saulina, Topological boundary conditions in abelian ChernSimons
theory, Nucl. Phys. B 845 (2011) 393 [arXiv:1008.0654] [INSPIRE].
[12] R. Dijkgraaf, C. Vafa, E.P. Verlinde and H.L. Verlinde, The operator algebra of orbifold
models, Commun. Math. Phys. 123 (1989) 485 [INSPIRE].
[13] A. Coste, T. Gannon and P. Ruelle, Finite group modular data,
Nucl. Phys. B 581 (2000) 679 [hepth/0001158] [INSPIRE].
[16] V.V. Nikulin, Integer symmetric bilinear forms and some of their geometric applications,
Math. USSRIzv 14 (1980) 103.
Commun. Math. Phys. 123 (1989) 177 [INSPIRE].
[17] G.W. Moore and N. Seiberg, Classical and quantum conformal field theory,
[18] G.W. Moore and N. Seiberg, Lectures on RCFT, in Strings ′89, Proceedings of the Trieste
Spring School on Superstrings,
[INSPIRE].
http://www.physics.rutgers.edu/∼gmoore/LecturesRCFT.pdf, World Scientific, (1990)
[19] C. Dong and J. Lepowsky, Generalized vertex algebras and relative vertex operators,
Prog. Math. 112, Birkh¨auser, Boston, MA, U.S.A., (1993).
[20] F. Quinn, Group categories and their field theories, Geom. Topol. Monographs 2 (1999) 407
[math/9811047] [INSPIRE].
[21] J. Polchinski, String theory, volume 2: superstring theory and beyond,
Cambridge University Press, Cambridge, U.K., (1998) [INSPIRE].
[22] V. Drinfeld, S. Gelaki, D. Nikshych and V. Ostrik, On braided fusion categories. I,
Selecta Math. 16 (2010) 1 [arXiv:0906.0620].
group cohomology, arXiv:1404.3230 [INSPIRE].
JHEP 02 (2015) 172 [arXiv:1412.5148] [INSPIRE].
HJEP07(218)9
arXiv:1802.04790 [INSPIRE].
group cohomology, arXiv:1403.1467 [INSPIRE].
topological phases and cobordisms, JHEP 12 (2015) 052 [arXiv:1406.7329] [INSPIRE].
arXiv:1604.06527 [INSPIRE].
arXiv:1803.10796 [INSPIRE].
PTEP 2017 (2017) 033B04 [arXiv:1610.07010] [INSPIRE].
[1] M. Barkeshli , P. Bonderson , M. Cheng and Z. Wang , Symmetry, defects and gauging of [3] A.P.O. Chan , J.C.Y. Teo and S. Ryu , Topological phases on nonorientable surfaces: twisting by parity symmetry, New J . Phys. 18 ( 2016 ) 035005 [arXiv: 1509 .03920] [INSPIRE]. [4] M.A. Metlitski , L. Fidkowski , X. Chen and A. Vishwanath , Interaction effects on 3D [7] Y. Tachikawa , On gauging finite subgroups , arXiv:1712 .09542 [INSPIRE]. [8] F. Benini , C. C´ordova and P.S. Hsin , On 2group global symmetries and their anomalies, [14] C.T.C. Wall , Quadratic forms on finite groups, and related topics , Topology 2 ( 1963 ) 281 . [15] C.T.C. Wall , Quadratic forms on finite groups. II, Bull. London Math. Soc. 4 ( 1972 ) 156 .
Ann. Math. 60 ( 1954 ) 49 . [23] S. Eilenberg and S. Mac Lane , On the groups H(Π, n). II. Methods of computation , [24] A. Kapustin and R. Thorngren , Anomalies of discrete symmetries in various dimensions and [25] D. Gaiotto , A. Kapustin , N. Seiberg and B. Willett , Generalized global symmetries, [26] S.X. Cui , C. Galindo , J.Y. Plavnik and Z. Wang , On gauging symmetry of modular categories , Commun. Math. Phys. 348 ( 2016 ) 1043 [arXiv: 1510 .03475] [INSPIRE]. [27] C. Co ´rdova, T.T. Dumitrescu and K. Intriligator , Exploring 2group global symmetries, [28] E. Pavel and C. Galindo , Reflection fusion categories, arXiv: 1803 .05568.
[30] A. Kapustin , R. Thorngren , A. Turzillo and Z. Wang , Fermionic symmetry protected