Multiplicative Structures Subm 130626

MULTIPLICATIVE STRUCTURES ON THE TWISTED
K-THEORY OVER FINITE GROUPS

C
ESAR GALINDO, ISMAEL GUTI
ERREZ, AND BERNARDO URIBE

Abstract. Let K be a nite group and let G be a nite group acting over K
by automorphisms. In this paper we study two dierent but intimate related
subjects: on the one side we classify all possible multiplicative and associative
structures which one can endow the twisted G-equivariant K-theory over K,
and on the other, we classify all possible monodical structures which one en-
dow the category of twisted and G-equivariant bundles over K. We achieve
this classication by encoding the relevant information in the cochains of a
sub double complex of the double bar resolution associated to the semi-direct
product K G; we use known calculations of the cohomology of K, G and
K G to produce concrete examples of our classication.
In the case on which K = G and G acts by conjugation, the multiplication
map G G G is a homomorphism of groups and we dene a shue ho-
momorphism which realizes this map at the homological level. We show that
the categorical information that denes the Twisted Drinfeld Double can be
realized as the dual of the shue homomorphism applied to any 3-cocycle of
G. We use the pullback of the multiplication map in cohomology to classify
the possible ring structures that Grothendieck ring of representations of the
Twisted Drinfeld Double may have, and we include concrete examples of this
procedure.
Introduction
The purpose of this work is to investigate a relationship existing among certain
tensor categories attached to a semi-direct product of groups (they include the
Twisted Drinfeld Double of a discrete group), their fusion algebras and the coho-
mology groups of the semi-direct product. These tensor categories, as well as some
of our results, are closely related to abelian extensions of Hopf algebras and the
cohomological description of Opext(C
G
, CF)), given by Kac in [16], cf. [19].
The abelian extension theory of Hopf algebras was generalized to coquasi-Hopf
algebras by Masuoka in [20]. Some of our results and constructions can be framed in
the abelian extension theory of coquasi-Hopf algebras in the particular case where
the matched pair of groups is a semi-direct product. However, our approach to these
tensor categories does not follow Masuokas point of view, instead we use the con-
cept of pseudomonoids in a suitable 2-monoidal 2-category associated to a group.
2010 Mathematics Subject Classication. (primary) 19L50, 18D10, (secondary) 16L35, 20J06.
Key words and phrases. Twisted Equivariant K-theory, Twisted Drinfeld Double, Fusion Cat-
egory, Multiplicative Structure, Fusion Algebra.
C.G would like to thank the hospitality of the Mathematics Department at MIT where part
of this work was carried out. C.G. was partially supported by Vicerrectora de Investigaciones de
la Universidad de los Andes. I.G acknowledges and thanks the nancial support of the Deutscher
Akademischer Austausch Dienst. B.U. acknowledges and thanks the nancial support of the
Alexander Von Humboldt Foundation.
1
2 C

There are some reasons why we prefer to use the pseudomonoid approach: First, it
exhibits more clearly the relationship between the cohomology of semi-direct prod-
ucts and the multiplicative structures of the G-equivariant twisted K-theory over a
nite group K. Second, some terms of a spectral sequence associated to the double
complex have a direct interpretation in terms of obstructions and classication of
the possible pseudomonoid structures. Third, some of our constructions and re-
sults make sense in categories dierent from the category of sets; in particular, if
we change to the Cartesian categories of locally compact topological spaces and
change discrete group cohomology by the Borel-Moore cohomology [21], we have a
much general theory where the categories of coquasi-Hopf algebras may not capture
all the desired information.
There are two main reasons for our interest in the fusion algebra of the tensor
categories dened in this paper. On the one side, semisimple tensor categories can
be encoded in a combinatorial structure divided in two parts: the fusion algebra
(or the Gorthendick ring of the tensor category) and a non-abelian cohomological
information provided by the F-matrices of the 6j-symbols, [24].
And on the other, these fusion algebras are generalizations of

K([G/G]), the
w-twisted stringy K-theory of the groupoid [G/G] (see [25]).
In the case that the semi-direct product is nite, the associated tensor categories
are fusion categories, and they belong to a bigger family of fusion categories denoted
group-theoretical fusion categories for which many interesting results have been
established cf. [12, 22]. Since an explicit description of the fusion rules of the
tensor categories studied in this paper, via induction and restriction of projective
representations of certain subgroups of G already appear in [26, Theorem 4.8], our
approach focuses in determining the number of fusion category and fusion algebra
structures associated to a x semi-direct product. We accomplish this task in several
steps. First, we show that the information encoding a pseudomonoid with strict unit
in the 2-category of G-sets with twists over the group K is equivalent to a 3-cocycle
in Z
3
(Tot
(A
,
(K G, T))), where the double complex A
,
(K G, T) is the sub
double complex without the 0-th row of the double bar resolution C
,
(K G, T)
and whose total cohomology calculates the cohomology of KG. Second, we show
that the information encoded in a pseudomonoid with strict unit in the 2-category
of G-sets with twists /, is precisely the precise information required to endow
the category Bun
G
(/) of projective G-equivariant complex vector bundles with a
monodical structure; hence the isomorphism classes of bundles Groth(Bun
G
(/))
becomes a fusion algebra, and this fusion algebra structure could be alternatively
understood as a twisted G-equivariant K-theory ring over K. Third, we dene
the twisted G-equivariant K-theory over the group K and we show the conditions
under which this group could be endowed with a multiplicative structure making it
a ring; we dene the group of multiplicative structures by MS
G
(K) and we show
that this group could be calculated by the use of a spectral sequence associated
to the complex Tot
(A
,
(K G, T)). We study the canonical homomorphism
H
3
(Tot
(A
,
(K G, T)))

MS
G
(K) and we give an explicit description of its
kernel and its cokernel; a multiplicative structure on MS
G
(K) not appearing in the
image of endows the twisted G-equivariant K-theory of K with a ring structure,
which is not the fusion algebra of the tensor categories Bun
G
(/), or in other words
an algebra structure, which is not possible to categorify.
MULTIPLICATIVE STRUCTURES ON TWISTED EQUIVARIANT K-THEORY 3
Of particular interest is the case on which G = K and G acts on itself by
conjugation. In this case, the cohomological information which was used in [9] to
dene the Twisted Drinfeld Double D
w
(G) for w Z
3
(G, T) denes an element in
Z
3
(Tot
(A
,
(G G, T))). The formul dening this 3-cocycle were reminiscent
of the formul appearing in [11, Theorem 5.2] on the proof of the Eilenberg-Zilber
theorem, and we conjectured that there had to exist a way to dene for any cocycle
in Z
n
(G, T) a n-cocycle in Z
n
(Tot
(A
,
(G G, T))) having similar properties as
the ones dened for n = 3. We show in this paper that indeed this is the case and
its proof is based on two facts: rst that the multiplication map : G G G,
(k, g) = kg is a homomorphism of groups, and second, on a construction of an
explicit Shue homomorphism at the chain level, whose dual
: C
(G, T)
Tot
(C
,
(GG, T)) applied to w recovers the cocycle dened in [9], and moreover
that in cohomology equals the pullback of , e.i.
: H
(G, T) H
(G
G, T). Since the group GG is isomorphic to the product GG we get that the
map H
3
(Tot
(A
,
(K G, T)))

MS
G
(K) is surjective, and since we know that
the cohomology class of the 3-cocycle that is dened in [9] could be recovered from
the cohomology class of
w, we give a simple procedure to determine the fusion

algebras of Rep(D
w
(G)) which are isomorphic to the G-equivariant K-theory ring
KU
G
(K); at the end of this work this procedure exemplied in some interesting
cases.
This paper is organized as follows. In Section 1 we provide background material
on the semi-direct products and the double bar complex associated to its cohomol-
ogy. In Section 2 the Shue homomorphism of a trivializable semi-direct product
is dened and some of its properties are shown. In Section 3 the 2-monoidal 2-
category of twisted G-sets with strict unit and the 2-category of pseudomonoids in
this 2-category are dened and described using the complex Tot
(A
,
(KG, T)).
In Section 4 the tensor category of equivariant vector bundles over a group, ten-
sor functors and monoidal natural isomorphism associated to the 2-category of
pseudomonoids in the 2-category of twisted G-sets with strict unit are dened.
In Section 5 the obstruction to the existence of multiplicative structures over the
twisted G-equivariant K-theory over a group K is described using the spectral se-
quence associated to the ltration F
r
:= A
,>r
of the double complex A
,
. In
Section 6 several concrete examples are completely calculated. We nish with an
appendix in Section 7 on which we give the explicit relation of our tensor categories
and coquasi-bialgebras.
1. Preliminaries
1.1. Semi-direct products. Let K be a discrete group endowed with an action
of the discrete group G dened through a homomorphism : G Aut(K); for
simplicity, for g G and k K denote the action by g(k) := (g)(k). Denote
by K G the group dened by the semi-direct product of G with K; as a set
KG := KG and the product structure is dened by (a, g)(b, h) := (a g(b), gh).
The group K G ts in the short exact sequence
1 K K G
2
G 1
and we say that K G is isomorphic to another split extension
1 K E
p
G 1
4 C

whenever there is an isomorphism : K G
=
E such that
2
= p . In what
follows we will outline the conditions under which the semi-direct product K G
is isomorphic to the direct product K G as split extensions.
For this recall that Inn(K) is the group of inner automorphisms of the group K,
i.e. the automorphisms of K induced by conjugation, and that it ts in the short
exact sequences
1 Z(K) K

Inn(K) 1
1 Inn(K) Aut(K) Out(K) 1
where Z(K) denotes the center of K and Out(K) denotes the group of outer auto-
morphisms of K.
Proposition 1.1. The semi-direct product K G is isomorphic to the product
K G if and only if there is a homomorphism : G K such that the following
diagram commutes
G
Inn(K)
where the action of G on K is given by the inner automorphisms dened by .
Proof. Suppose that there is a homomorphism : G K with the desired prop-
erties. Dene the isomorphism of sets
: K G K G, (a, g) (a(g), g);
note that on the one hand
((a, g) (b, h)) = (a(g)(b), gh)
= (a(g)b(g)
1
, gh)
= (a(g)b(h), gh)
and on the other hand
(a, g)(b, h) = (a(g), g)(b(h), h)
= (a(g)b(h), gh);
therefore the map is the desired isomorphism of groups.
For the converse let us suppose that : K G K G is an isomorphism of
split extensions. In particular we have that
2
((a, g)) = g
since the projection on the second coordinate on both groups produce the anchor
map to G. Dene the map
: G K, g
1
((1, g))
and hence we have that (1, g) = ((g), g). Since is an isomorphism we have
that
((gh), gh) = (1, gh) = (1, g)(1, h) = ((g), g)((h), h) = ((g)(h), gh)
and therefore we obtain that the map is moreover a homomorphism of groups.
Denote by : K K the isomorphism of groups induced by when is dened
by the equation
(a, 1) = ((a), 1).
Applying to both sides of the equation
(1, g) (a, 1) = (g(a), g)
we obtain on the one side
(1, g)(a, 1) = ((g), g)((a), 1) = ((g)(a), g)
and on the other
(g(a), g) = (g(a), 1) (1, g) = ((g(a)), 1)((g), g) = ((g(a))(g), g),
implying that
(g)(a)(g)
1
= (g(a))
and applying
1
we get
1
((g))a
1
((g))
1
= g(a)
which implies the desired equation
(
1
((g))) = (g).
Hence the homomorphism
:=
1
: G K
ts into the diagram
G
Inn(K).
Corollary 1.2. The semi-direct product K Inn(K) is isomorphic as split exten-

sions to the direct product KInn(K) if and only if the natural short exact sequence
1 Z(K) K Inn(K) 1 splits.
Corollary 1.3. If the semi-direct product K G is isomorphic to K G then the
action of G on K is given by inner automorphisms of K, and there is a homomor-
phism : G K that realizes such inner automorphisms.
Example 1.4. Let K = G and consider the conjugation action of K on itself. In
this case = and we can take = Id
K
. Therefore the map
: K K K, (a, g) ag
is a homomorphism of groups and

2
: K K K K, (a, g) (ag, g)
is an isomorphism.
6 C

Remark 1.5. Note that whenever we have a homomorphism : G K such that
= then the map
: K G K, (k, g) k(g)
becomes a group homomorphism. Moreover, the map
K G K K, (k, g) (k, (g))
is a homomorphism of groups
1.2. Bar resolution. Let us nd an explicit model for the homology of the group
K G. For this, let us rst setup the notation for the explicit model for the bar
resolution that we will use.
Take H a discrete group and dene the complex C
(EH, Z) with
C
n
(EH, Z) := ZH
n+1
and with dierential
H
: C
n
(EH, Z) C
n1
(EH, Z) dened by the equation on
generators
H
(h
1
, h
2
, ..., h
n+1
) = (h
2
, h
3
, ..., h
n+1
) +
n
i=1
(1)
i
(h
1
, ..., h
i
h
i+1
, ..., h
n+1
).
The complex (C
(EH, Z),
H
) becomes a complex in the category of ZH-modules
if we endow each C
n
(EH, Z) with the left ZH-module structure dened by the
equation
h (h
1
, ..., h
n+1
) := (h
1
, ..., h
n+1
h
1
).
The augmentation map
: C
0
(EH, Z) Z, (h) = 1
is a map of ZH-modules and the complex C
(EH, Z) becomes a ZH-free resolution

of the trivial ZH-module Z,
: C
(EH, Z) Z.
The elements
(h
1
, h
2
, ..., h
n
, 1)
generate the ZH-module C
n
(EH, Z) and therefore we could write them using the
bar notation
[h
1
[h
2
[...[h
n
] := (h
1
, h
2
, ..., h
n
, 1);
the dierential
H
in this base becomes
H
[h
1
[h
2
...[h
n
] = [h
2
[h
3
[...[h
n
] +
n
i=1
(1)
i
[h
1
[...[h
i
h
i+1
[...[h
n
]
+ (1)
n+1
h
1
n
[h
1
[...[h
n1
].
For a left ZH-module W, the homology groups of H with coecients in W are
dened as
H
(H, W) := H
(C
(EH, Z)
ZH
W)
and the cohomology groups of H with coecients in W as
H
(H, W) := H
(Hom
ZH
(C
(EH, Z), W)).

Since we have a canonical isomorphism of Z-modules
Hom
ZH
(C
n
(EH, Z), W)

= Maps(H
n
, W),
the cohomological dierential
H
in terms of the bar notation becomes
(
H
f)[h
1
[h
2
...[h
n
] = f[h
2
[h
3
[...[h
n
] +
n
i=1
(1)
i
f[h
1
[...[h
i
h
i+1
[...[h
n
]
+ (1)
n+1
h
1
n
f[h
1
[...[h
n1
].
1.3. Cohomology of K G. Consider the double complex
C
(EG, Z)
Z
C
(EK, Z)
with dierentials
G
1 and 1
K
. Denote by (g
1
, ..., g
p+1
[[k
1
, ..., k
q+1
) a generator
in C
p
(EG, Z)
Z
C
q
(EK, Z) and dene the action of (k, g) KG by the equation
(k, g) (g
1
, ..., g
p+1
[[k
1
, ..., k
q+1
) := (g
1
, ..., g
p+1
g
1
[[g(k
1
), ..., g(k
q+1
)k
1
).
A straightforward computation shows that indeed it is an action and therefore
C
p
(EG, Z)
Z
C
q
(EK, Z) becomes a free Z(K G)-module. Since the dierentials
G
1 and 1
K
are also maps of Z(K G)-modules, we could take the total
complex
Tot
(C
(EG, Z)
Z
C
(EK, Z))
whose degree n-component is
Tot
n
(C
(EG, Z)
Z
C
(EK, Z)) :=
p+q=n
C
p
(EG, Z)
Z
C
q
(EK, Z)
and whose dierential is
G
1 (1)
p
1
K
thus obtaining a free Z(K G) complex. Since the homology of the total complex
of this double complex is just Z in degree 0,
H
(Tot
(C
(EG, Z)
Z
C
(EK, Z)),
G
1 (1)
p
1
K
) = Z,
we have that
Tot
(C
(EG, Z)
Z
C
(EK, Z))

Z
is a free Z(K G) resolution of Z.
Making use of the bar notation we take the elements
[g
1
[...[g
p
[[k
1
[...[k
q
] := (g
1
, ..., g
p
, 1[[k
1
, ..., k
q
, 1)
as a set of generators of C
p
(EG, Z)
Z
C
q
(EK, Z) as a Z(K G)-module; in this
base we have the equality
(g
1
, ..., g
p
, g[[k
1
, ..., k
q
, k) = (k
1
, g
1
) [g
1
[...[g
p
[[g(k
1
)[...[g(k
q
)].
This choice of base provides us with an isomorphism of Z-modules
Hom
Z(KG)
(C
p
(EG, Z)
Z
C
q
(EK, Z), Z)

= Maps(G
p
K
q
, Z)
that allows us to transport the dual of the dierentials
G
1 and 1
K
to
Maps(G
p
K
q
, Z); the induced dierentials will be denoteb by
G
and
K
. Therefore
we obtain:
Denition 1.6. Let C
p,q
(K G, Z), p, q 0, be the double complex
C
p,q
(K G, Z) := Maps(G
p
K
q
, Z)
8 C

with dierentials
G
: C
p,q
C
p+1,q
and
K
: C
p,q
C
p,q+1
dened by the
equations
(
G
F)[g
1
[...[g
p+1
[[k
1
[...[k
q
] = F[g
2
[...[g
p+1
[[k
1
[...[k
q
]
+
p
i=1
(1)
i
F[g
1
[...[g
i
g
i+1
[..[g
p+1
[[k
1
[...[k
q
]
+ (1)
p+1
F[g
1
[...[g
p
[[g
p+1
(k
1
)[...[g
p+1
(k
q
)]
(
K
F)[g
1
[...[g
p
[[k
1
[...[k
q+1
] = F[g
1
[...[g
p
[[k
2
[...[k
q+1
]
+
q
j=1
(1)
j
F[g
1
[...[g
p
[[k
1
[...[k
j
k
j+1
[...[k
q+1
]
+ (1)
q+1
F[g
1
[...[g
p
[[k
1
[...[k
q
],
where by convention G
0
K
0
is the set with one point.
Hence we have
Lemma 1.7. The cohomology of the total complex of the double complex C
,
(K
G, Z),
G
,
K
is the cohomology of the group K G, i.e.
H
_
Tot
(C
,
(K G, Z)),
G
(1)
p
K
_

= H
(K G, Z).
We can take a smaller double complex, more suited for our work, which is called
the normalized double complex. Let us dene it
Denition 1.8. The normalized double complex of C
,
(K G, Z),
G
,
K
is the
double complex
C
p,q
(K G, Z) C
p,q
(K G, Z)
consisting of maps F : G
p
K
q
Z such that F[g
1
[...[g
p
[[k
1
[..[k
q
] = 1 whenever
g
i
= 1 or k
j
= 1. The dierentials on C
p,q
are also
G
and
K
. We setup C
0,0
= Z.
It is known in homological algebra that the normalized complex of the bar res-
olution is quasi-isomorphic to the bar resolution (see page 215 in [13]). Therefore
we have
Lemma 1.9. The induced map on total complexes
Tot
(C
,
(K G, Z)) Tot
(C
,
(K G, Z))
is a quasi-isomorphisms. Then
H
(Tot
(C
,
(K G, Z)),
G
(1)
p
K
)

= H
(K G, Z).
We can generalize our denition of the double complex to other coecients.
Denote by T the group S
1
and consider the exact sequence of coecients
0 Z R T 0.
We will denote the complex C
p,q
(K G, T) as the complex generated by maps
F : G
p
K
q
T on which F is 1 if one of the entries is the identity, and the
dierentials
G
and
K
are the same ones as Denition 1.6 but changing the sums
by multiplications. Note also that we set up C
0,0
(K G, T) := T.
Then
H
_
Tot
(C
,
(K G, T)),
G

(1)
p
K
_

= H
(K G, T).
1.3.1. Decomposition of the cohomology of K G. Note that the 0-th row of the
double complex is isomorphic to the normalized bar cochain complex of G
(C
,0
(K G, Z),
G
)

= (C
(G, Z),
G
),
and the action of the dierential
K
on this row is trivial. Therefore, if we dene
Denition 1.10. Let
A
,
(K G, Z) := C
,>0
(K G, Z)
be the sub double complex of C
,
(KG, Z) with trivial 0-th row, and dierentials
G
and
K
.
Then we obtain
Lemma 1.11. There is a canonical isomorphism of double complexes
(C
,
(KG, Z),
G
(1)
p
K
)

= (A
,
(KG, Z),
G
(1)
p
K
)(C
(G, Z),
G
),
which induces a canonical isomorphism in cohomology
H
(K G, Z)

= H
(Tot
(A
,
(K G, Z))) H
(G, Z),
and the respective one with coecients in T,
H
(K G, T)

= H
(Tot
(A
,
(K G, T))) H
(G, T)
where A
,
(K G, T) = C
,>0
(K G, T)
2. The case of the trivializable semi-direct product K G
Whenever the semi-direct product K G is isomorphic to K G we know by
Proposition 1.1 and Remark 1.5 that there exists a homomorphism : G K such
that KG KK, (k, g) (k, (g)) is a homomorphism of groups. In this case
we have the homomorphisms
K G K K

K, (k, g) (k, (g)) k(g)
which induce the following homomorphism at the level of their homologies
H
(K G, Z) H
(K K, Z)

H
(K, Z).
Since the interesting information lies on the homomorphism
we will investigate
its properties.
2.1. The Shue homomorphism. In what follows we will describe how to obtain
the map
at the chain level using the explicit models for H
(K K, Z) and
H
(K, Z) dened previously. Its denition needs some preparation.

Take a base element [g
1
[...[g
p
[[k
1
[...[k
q
] in C
p
(EK, Z)
Z
C
q
(EK, Z) and think of
it as a way to represent
p
q
as the product of the simplices [g
1
[...[g
p
] [k
1
[...[k
q
]
in BK. For Shu(p, q) a (p, q)-shue, i.e. an element in the symmetric group
S
p+q
such that (i) < (j) whenever 1 i < j p or p + 1 i < j p + q, we
can dene an element
[g
1
[...[g
p
[k
1
[...[k
q
] C
p+q
(EK, Z)
such that [g
1
[...[g
p
[k
1
[...[k
q
] := [s
1
[...[s
p+q
] with
s
(i)
=
_
g
i
if i p
g
p+(i)i+1
. . . g
p1
g
p
k
ip
(g
p+(i)i+1
. . . g
p1
g
p
)
1
if i > p.
10 C

From the Eilenberg-Zilber theorem it follows that the set
[g
1
[...[g
p
[k
1
[...[k
q
] : Shu(p, q)
is a subdivision in simplices of dimension p + q of the product of the simplices
[g
1
[...[g
p
] [k
1
[...[k
q
].
An equivalent way to see the elements [g
1
[...[g
p
[k
1
[...[k
q
] is the following: a
(p, q)-shue can also be understood as a way to parameterize a lattice path of
minimum distance from the point (0, 0) to the point (p, q); one moves one unit to
the right in the steps (1), ..., (p) and one unit up in the steps (p+1), ..., (p+q).
We label the horizontal and vertical edges in the rectangular lattice dened by the
points (0, 0), (p, 0), (p, q), (0, q) by the following rule:
The horizontal path between (i 1, j) and (i, j) is labeled with g
i
indepen-
dent of j.
The vertical path between (p, j 1) and (p, j) is labeled with k
j
and all
the other vertical paths are labeled in such a way that the squares become
commutative squares (when thinking of the labels as maps). This implies
that the vertical edge from (i, j 1) to (i, j) is labeled with
g
i+1
. . . g
p1
g
p
k
j
(g
i+1
. . . g
p1
g
p
)
1
.
Since parameterizes a path in the lattice, the element [g
1
[...[g
p
[k
1
[...[k
q
] encodes
the labels that the path follow in order to go from (0, 0) to (p, q).
For example, consider the (3, 2)-shue
:=
_
1 2 3 4 5
1 3 5 2 4
_
then the element [g
1
[g
2
[g
3
[k
1
[k
2
] can be seen as the path
(2, 2)
g3
(3, 2)
(1, 1)
g2
(2, 1)
g3k2g
1
3
g3
(3, 1)
k2
(0, 0)
g1
(1, 0)
g2g3k1(g2g3)
1
g2
(2, 0)
g3
(3, 0)
k1
and therefore
[g
1
[g
2
[g
3
[k
1
[k
2
] = [g
1
[g
2
g
3
k
1
g
1
3
g
1
2
[g
2
[g
3
k
2
g
1
3
[g
3
].
We can now dene the map of complexes which in homology realizes
:
Denition 2.1. Let
: C
p
(EK, Z)
Z
C
q
(EK, Z) C
p+q
(EK, Z)
be the graded homomorphism dened on generators by the equation
(g
1
, ...g
p
, g[[k
1
, ..., k
q
, k) :=
Shu(p,q)
(gk)
1
(1)
||
[g
1
[...[g
p
[gk
1
g
1
[...[gk
q
g
1
]
where [[ denotes the sign of the permutation . This is usually called the Shue
homomorphism.
We claim that the Shue homomorphism is an admissible product in the sense
described in [7, IV-55, p. 118]
Theorem 2.2. The graded homomorphism
: Tot
(C
(EK, Z)
Z
C
(EK, Z)) C
(EK, Z)
is a chain map that satises the equations
_
(a
1
, h
1
) (g
1
, ...g
p
, g[[k
1
, ..., k
q
, k)
_
= (ha)
1
(g
1
, ...g
p
, g[[k
1
, ..., k
q
, k)
( )(k, g) = (kg).
Hence it induces a graded homomorphism in homology
: H
(K K, Z) H
(K, Z)
which is equal to the one induced by the pushforward
.
Proof. Let us start by proving the equations. For the rst one, we have on the one
side
((a
1
, h
1
) (g
1
, ...g
p
,g[[k
1
, ..., k
q
, k))
=
_
g
1
, ...g
p
, gh[[h
1
k
1
h, ..., h
1
k
q
h, h
1
kha
_
=
Shu(p,q)
(gkha)
1
(1)
||
[g
1
[...[g
p
[gk
1
g
1
[...[gk
q
g
1
]
and on the other
(ha)
1
(g
1
, ...g
p
,g[[k
1
, ..., k
q
, k)
= (ha)
1
Shu(p,q)
(gk)
1
(1)
||
[g
1
[...[g
p
[gk
1
g
1
[...[gk
q
g
1
]
=
Shu(p,q)
(gkha)
1
(1)
||
[g
1
[...[g
p
[gk
1
g
1
[...[gk
q
g
1
];
this in particular implies that is a homomorphism which preserves the respective
module structures, namely that
((a, h)(g
1
, ...g
p
, g[[k
1
, ..., k
q
, k)) = (a, h)(g
1
, ...g
p
, g[[k
1
, ..., k
q
, k).
For the second, simply note that
( )(k, g) = 1 1 = 1 = (kg)
which implies that (k, g)) = ( )(k, g).
The proof of the fact that is a chain map, namely that
K
= (
K
1 + (1)
p
1
K
),
is essentially included in the proof of [11, Theorem 5.2]; the decomposition in sim-
plices of dimension p+q of the product of the simplices [g
1
[...[g
p
] [k
1
[...[k
q
] in BK
is done by choosing appropriately p + q edges with the use of the (p, q)-shues as
follows
[g
1
[...[g
p
[[k
1
[...[k
q
] =
Shu(p,q)
(1)
[g
1
[...[g
p
[k
1
[...[k
q
].
With this decomposition of [g
1
[...[g
p
] [k
1
[...[k
q
], its boundary can be calculated
as (
K
1+(1)
p
1
K
)[g
1
[...[g
p
][k
1
[...[k
q
] or alternatively as
K
([g
1
[...[g
p
[[k
1
[...[k
q
]);
then it follows that is a chain map.
12 C

Now, since is a chain map which preserves the module structures, then it
induces a chain map at the level of the coinvariants
Tot
(C
(EK, Z)
Z
C
(EK, Z))
Z(KK)
Z C
(EK, Z)
ZK
Z
which denes a homomorphism
: H
(K K, Z) H
(K, Z).
This homomorphism
must be equal to the pushforward
since preserves the

module structures dened by .
If we have a homomorphism of groups : G K for which the map : KG
K G, (k, g) (k, (g)) induces an isomorphism of semi-direct products, then we

have that the map
( 1) : C
p
(EG, Z)
Z
C
q
(EK, Z) C
p+q
(EK, Z)
[g
1
[...[g
p
[[k
1
[...[k
q
] [(g
1
)[...[(g
p
)[[k
1
[...[k
q
]
induces a chain map
( 1) : Tot
(C
(EG, Z)
Z
C
(EK, Z)) C
(EK, Z)
preserving their respective module structures, and hence inducing a homomorphism
( ( 1))
: H
(K G, Z) H
(K, Z)
which is equal to the pushforward map of the composition ; i.e.
( )
= ( ( 1))
: H
(K G, Z) H
(K, Z).
2.2. The dual of the Shue homomorphism. Dualizing the map , we obtain
a homomorphism
: C
n
(K, Z))
p+q=n
C
p,q
(K K, Z)
(
F)[s
1
[...[s
n
]
p+q=n
F([s
1
[...[s
p
[[s
p+1
[...[s
p+q
])
which induces a cochain map
: (C
(K, Z),
K
) (Tot
(C
,
(K K, Z)),
G
(1)
p
K
);
here we have kept the notation of the dierentials as
G
and
K
in order to avoid
confussions. A straightforward calculation shows that the cochain map
preserves
normalized cochains
: (C
(K, Z),
K
) (Tot
(C
,
(K K, Z)),
G
(1)
p
K
)
and therefore it induces a homomorphism at the level of cohomologies
: H
(K; Z) H
(K K, Z)
which by Theorem 2.2 is equal to the pullback of the homomorphism of groups .
If we bundle up our previous discussion we have
Theorem 2.3. The homomorphism in cohomology
: H
(K, Z) H
(K K, Z)
dened by the cochain map
: C
n
(K, Z))
p+q=n
C
p,q
(K K, Z)
(
F)[s
1
[...[s
n
]
p+q=n
F([s
1
[...[s
p
[[s
p+1
[...[s
p+q
])
is equal to
: H
(K, Z) H
(K K, Z)
which is the pullback of the group homomorphism : K K K, (k, g) kg.
2.3. Further properties of the Shue homomorphism. Consider the homo-
morphisms
K
: K K K,
G
: K K K
x (x, 1
K
) g (1
K
, g).
and note that they t into the commutative diagram
K

K
K K
K
which induces the commutative diagram
H
(K, Z)
=
.
(K, Z) H
(K K, Z)
(K, Z).
(2.1)
From Lemma 1.11 we know that there is a canonical isomorphism
H
p
(K K, Z)

= H
p
(Tot
(A
,
(K K, Z))) H
p
(K, Z)
and the homomorphism
G
is precisely the projection on the second component of
the direct sum H
p
(Tot
(A
,
(K K, Z))) H
p
(K, Z).
Moreover, by dening the chain map
1
: C
n
(K, Z))
p+q=n, q>0
A
p,q
(K K, Z) (2.2)
(
1
F)[s
1
[...[s
n
]
p+q=n, q>0
F([s
1
[...[s
p
[[s
p+1
[...[s
p+q
])
we have that the chain map
: C
(K, Z)) Tot
(A
,
(K K, Z)) C
,0
(K K, Z)
is isomorphic to the chain map
1
1 : C
(K, Z)) Tot
(A
,
(K K, Z)) C
,0
(K K, Z)
14 C

and therefore we have that at the cohomological level we obtain the commutative
diagram
H
p
(K, Z)
H
p
(K, Z) H
p
(Tot
(A
,
(K K, Z)))
K
|
A
(2.3)
for all p > 0, where at the cochain level
K
[
A
is simply the projection map on the
0-th column
K
[
A
: Tot
(A
,
(K K, Z)) A
,0
(K K, Z)

= C
>0
(K, Z).
Therefore, if
Denition 2.4. Let
B
,
(K G, Z) := C
>0,>0
(K G, Z)
be sub double complex of C
,
(K G, Z) with trivial 0-th row and trivial 0-th
colum, and dierentials
G
and
K
.
We obtain
Lemma 2.5. The homomorphism
H
p
(K, Z) H
p
(Tot
(B
,
(K K, Z))) H
p
(Tot
(A
,
(K K, Z)))
x y
1
x +y
is an isomorphism for all p > 0.
Proof. The short exact sequence of complexes
0 Tot
(B
,
(K K, Z)) Tot
(A
,
(K K, Z)) C
>0
(K, Z) 0
induces a long exact sequence in cohomology groups
H
p
(Tot
(B
,
(K K, Z))) H
p
(Tot
(A
,
(K K, Z)))

K
|
A
H
p
(K, Z)
which splits by diagram (2.3).
Therefore we can conclude that there is a canonical splitting of the cohomology
of K K in terms of the cohomology of K and of the cohomology of the double
complex B
,
(K K, Z).
Proposition 2.6. The homomorphism
H
p
(K, Z) H
p
(Tot
(B
,
(K K, Z))) H
p
(K, Z)
=
H
p
(K K, Z)
x y z
1
x +y +
2
z
is an isomorphism for all p > 0. Here
2
: K K K, (a, g) g denotes
the homomorphism induced by the projection on the second coordinate satisfying
2

G
= 1.
3. Categorical definitions
3.1. 2-category of discrete G-sets with twist. We shall x a discrete group G.
We dene the 2-category of discrete G-sets with twist as follows:
(1) Objects will be called discrete G-sets with twist, and they are pairs (X, ),
where X is a discrete left G-set and Z
2
G
(X, T) is a normalized 2-cocycle
in the G-equivariant complex of X, i.e. a map
: GGX T,
such that
[[[[x] [[[[x]
1
[[[[x] [[[[x]
1
= 1
for all , , G, x X.
Note that the previous equation is equivalent to the equation
G
= 1,
when we see as element in C
2,1
(X G, T).
(2) Let (X,
X
), (Y,
Y
) be discrete G-sets with twist. A 1-cell from (X,
X
)
to (Y,
Y
), also called a G-equivariant map, is a pair (L, ),
(X,
X
)
(L,)
(Y,
Y
)
where
L : X Y is a morphism of G-sets,
C
1
G
(X, T) is a normalized cochain such that
G
() = L
(
Y
)(
X
)
1
,
i.e., a map
: GX T
such that
[[[x] [[x]
1
[[[x] =
Y
[[[[L(x)]
X
[[[[x]
1
,
for all , G, x X.
(3) Given two 1-cells (L, ), (L,
) : (X,
X
) (Y,
Y
), a 2-cell : (L, )
(L,
)
(X,
X
)
(L,)
(L,
(Y,
Y
)
is 0-cochain C
0
G
(X, T) such that
G
() =
1
, i.e., a map : X T,
such that
[x][x]
1
=
[[[x] [[[x]
1
for all G, x X.
Let us dene the composition on 1-cells and 2-cells in the following way. Let
(F,
F
) : (X,
X
) (Y,
Y
) and (G,
G
) : (Y,
Y
) (Z,
Z
) two 1-cells, dene
their composition as
(G,
G
) (F,
F
) = (G F, F
(
G
)
F
) : (X,
X
) (Z,
Z
),
16 C

(X,
X
)
(F,
F
)
(GF,F
(
G
)
F
)
(Y,
Y
)
(G,
G
)
(Z,
Z
)
and if : (L, ) (L,
) and
: (L,
) (L,
) are 2-cells, their composition

is the product of the maps, namely
=:
: (L, ) (L,
)
(X,
X
)
(L,
(L,
(L,)
(Y,
Y
) = (X,
X
)
(L,)
(L,
(Y,
Y
)
A straightforward calculation implies that

Lemma 3.1. The composition of 1-cells and 2-cells satisfy the axioms of a 2-
category.
3.2. Pseudomonoids. A strict 2-monoidal 2-category is a triple (B, , I) where B
is a 2-category, : B B B is a 2-functor and I is an object in B, such that
(Id
B
) = (Id
B
) and I X = X I = X for every object in B (for
more details see [6]).
Denition 3.2. Given a strict 2-monoidal 2-category (B, , I), a pseudomonoid
in B consists of:
an object C B,
together with:
a multiplication 1-cell m : CC C,
an identity-assigning 1-cell I : I C,
together with the following 2-isomorphisms:
the associator:
CCC
CC CC
C
idm
mid
a
(3.1)
the left and right unit laws:
I C CC CI
C
Iid
idI
=
.
(3.2)
such that the following diagrams commute:
the pentagon identity for the associator:
(3.3) m(mid)(midid)
m(mid)(idmid)
m(idm)(ididm)
m(idm)(idmid)
m(mm)
m(aid)
a(midid)
a(idmid)
m(ida)
a(ididm)
where we use the equalities

(id m) (mid id) = mm
(mid) (id id m) = mm
in order to compose the upper 1-cells a(midid) and a(ididm),
equalities which follow from the fact that is a 2-functor.
the triangle identity for the left and right unit laws:
(3.4) m(mid)(idIid)
a(idIid)
m(rid)
m(idm)(idIid)
m(id)
m
where we use the fact that CI = C = IC to make sense of the diagonal
arrows.
Remark 3.3. The 2-category of G-sets with twist has a 2-monoidal structure,
where the product of two objects (X,
X
), (Y,
Y
) is given by
(X,
X
) (Y,
Y
) = (X Y,
X

Y
),
where XY is a G-set with the diagonal G-action and
X
Y
:=
1
(
X
)
2
(
Y
).
In an analogous way we construct the product for 1-cells and 2-cells. A unit object
is any xed G-set with one element and the constant function 1 for 2-cocycle.
3.3. Pseudomonoids in the 2-category of G-sets with twist. We are inter-
ested in studying pseudomoniods in the 2-category of G-sets with twists, but in
order to get normalized cocycles we are forced to consider only pseudomonoids
where the identity-assigning 1-cell is strict in the sense that the cochain (the sec-
ond component of the 1-cell) is trivial, and furthermore, that the unit constraints
in diagram (3.2) are identities, namely that the diagram (3.2) commutes. We will
call these pseudomonoids with strict unit.
18 C

Proposition 3.4. A pseudomonoid with strict unit in the 2-category of G-sets with
twists is equivalent to the following data:
A monoid (K, m, 1), where K is a G-set, m is G-equivariant and 1 K is
a G-invariant element.
C
2,1
(K G, T), C
1,2
(K G, T), C
0,3
(K G, T) such that
is a three cocycle in
_
Tot(A
,
(K G, T)),
G

(1)
p
K
_
with
A
,
(K G, T) the double complex introduced in Denition 1.10.
Proof. A pseudomonoid in the 2-category of G-sets is:
i) An object C = (K, ) where K is a G-set and
: GGK T
such that is normalized in the components of the group G and that
G
= 1.
ii) A multiplication 1-cell
m = (m, ) : (K K, ) (K, )
such that m : K K K is a G-equivariant map, and a map
: GK K T
satisfying the equation
G
= m
( )
1
. (3.5)
iii) An identity-assigning 1-cell
I = (1
K
, ) : (, 1) (K, )
where 1
K
: K is a map choosing a G-invariant element 1
K
:= 1
K
()
in K, and : G T is the constant map 1 because we are only
considering pseudomonoids with strict unit. The cochain condition on
reads
(
G
)[g
1
[g
2
[[] = [g
1
[g
2
[[1
K
]
and since is the constant function it follows that [g
1
[g
2
[[1
K
] = 1, namely
that is normalized in the K variable.
iv) The left hand side of diagram (3.2) translates to the diagram
( K, 1 )
(2,1)
(1
K
id,1)
(K K, )
(m,)
(K, )
where
2
is the projection on the second component. The composition of
the 1 at the level of the G-sets implies the equation
m(1
K
, h) = h
for any element h K. Now, the composition of the 1-cells at the level of
the cochains implies the equation
[g[[] [g[[1
K
[h] = 1
with g G and h K. Since the cochain is equal to the constant
function 1, we have that [g[[1
K
[h] = 1. Applying the same arguments as
above we conclude that [g[[h[1
K
] = 1 and therefore the left and right unit
laws imply that is normalized on the K components, and moreover that
1
K
is a unit for the multiplication map m.
v) Diagram (3.1) translates to the diagram
(K K K, )
(mid,1)
(idm,1)
(K K, )
(m,)
(K K, )
(m,)
(K, )
whose commutativity at the level of G-sets implies that the multiplication
m : K K K is associative, and at the level of cochains the diagram
implies the equation
(
G
)[g[[h
1
[h
2
[h
3
] = (3.6)
[g[[h
2
[h
3
] [g[[h
1
h
2
[h
3
]
1
[g[[h
1
[h
2
h
3
] [g[[h
1
[h
2
]
1
.
vi) Diagram (3.3) translates into the equation
[k
1
k
2
[k
3
[k
4
] [k
1
[k
2
[k
3
k
4
] = [k
1
[k
2
[k
3
] [k
1
[k
2
k
3
[k
4
] [k
2
[k
3
[k
4
] (3.7)
for all elements k
1
, k
2
, k
3
, k
4
in K.
vii) Diagram (3.4) translates into the equality
(3.8) [k
1
[1
K
[k
2
] = 1
for all k
1
, k
2
K, because the unit constraints are trivial.
From ii) we have that the multiplication map m : KK K is a G-equivariant
map and v) tells us that the multiplication m is associative. From iii) we know
that K is provided with a G invariant element 1
K
and iv) tells us that this element
is a left and right unit for the multiplication m. We have then that (K, m, 1
K
) is a
monoid endowed with a G-action compatible with m and 1
K
.
Since K is a G-equivariant monoid with unit, we can use the notation of Deni-
tion 1.6 to see that equation (3.5) can be written as
G
()
K
() = 1,
equation (3.6) becomes
G
()
K
()
1
= 1
and equation (3.7) becomes
K
() = 1.
We furthermore have that
G
= 1 by the denition of an object in the category
of G-sets with twist.
Equation (3.8) implies that is normalized in the variable of the middle; this
fact, together with the fact that
2
= 1 implies that is normalized in all the
variables since
1 =
K
[1
K
[1
K
[k
1
[k
2
] = [1
K
[k
1
[k
2
]
1 =
K
[k
1
[k
2
[1
K
[1
K
] = [k
1
[k
2
[1
K
].
From iii) we know that is normalized in the K coordinate, and from iv) we
know that is normalized in the K coordinates. Therefore the maps , and
20 C

are normalized in all variables since their normalization on coordinates of G follow
from the denition of the 2-category of G-sets with twists.
Summarizing we have that C
2,1
(K G, T), C
1,2
(K G, T) and
C
0,3
(KG, T) such that is a three cocycle in
_
Tot(A
,
(K G, T)),
G

(1)
p
K
_
because we have that
(
G

(1)
p
K
)( ) =
G
()
K
()
G
()
K
()
1
G
()
K
()
=1 1 1 1,
or in a diagram
4 1
3 1
2 1
1 1
0 1 2 3
G
(
K
)
1
G
From the construction above, it is easy to see that if we are given the G equivari-
ant monoid (K, m, 1
K
) with unit, plus the cocycle then one can construct
in a unique way a pseudomonoid with strict unit in the 2-category of G-sets with
twists. This nishes the proof.
Remark 3.5. For a xed G equivariant monoid (K, m, 1
K
), the possible pseu-
domonoid structures with strict unit in the 2-category of G-sets associated over
(K, m, 1
K
) are classied by elements in
Z
3
(Tot(A
,
(K G, T))),
namely, 3-cocycles in the total complex of A
,
(K G, T).
Denition 3.6. Given pseudomonoids (C, m, I, a) and (C
, m
, I
, a
) in a 2-category
B, a morphism
T : (C, m, I, a) (C
, m
, I
, a
)
consists of:
1-cell F : C C
equipped with:
a 2-isomorphism
CC
C
C
C
FF
F2

a 2-isomorphism
C C
I
I
/
I
F0
(3.9)
such that diagrams expressing the following laws commute:
compatibility of F
2
with the associator:
m
(m
id) (F F F) m
(F F) (mid) F m (mid)
m
(id m
) (F F F) m
(F F) (id m) F m (id m)
m
(F2F)
F2(mid)
Fa
(FFF)
(FF2)
F2(idm)
compatibility of F
0
with the left unit law:
m
(I
F) F
m
(F F) (I id) F m (I id)
F

F
(F0F)
F2(Iid)
compatibility of F
0
with the right unit law:
m
(F I
) F
m
(F F) (id I) F m (id I)
Fr
Fr
(FF0)
F2(idI)
Denition 3.7. Given two pseudomonoids with strict unit / = (K, m, 1, , , )

and /
= (K
, m
, 1
) in the 2-category of G-sets with twists, a morphism

T : / /
is a morphism of pseudomonoids (as in Denition 3.6) such that the

2-isomorphism F
0
of diagram (3.9) is an identity.
Proposition 3.8. A morphism of pseudomonoids with strict unit in the 2-category
of G-sets with twists T : / /
consists of the triple T = (F, , ) with

F : K K
a G-equivariant morphism of monoids, and cochains C

1,1
(K G, T) and
C
0,2
(K G, T) such that
(
G

(1)
p
K
)(
1
) = F
/ F
/ F
/.
Proof. Following Denition 3.6 a morphism T : / /
consists of:
i) A 1-cell (F, ) : (K, ) (K
), i.e. a G-equivariant map F : K K
and a normalized cochain C

1
G
(K, T) such that
(3.10)
G
= F
/.
22 C

ii) A 2-cell C
0
G
(K K, T)
(K K, )
(m,)
(FF,)
(K, )
(F,)
(K
)
(m
(K
)
such that
G
= ( m
)/( F
).
Note that at the level of the G sets the diagram is commutative, therefore
F : K K
preserves the multiplication; and since
K
= (m
)
1
,
we can rewrite the equation above as:
G

K
= /F
. (3.11)
iii) The commutativity of the diagram (3.9)
(K, ) (K
)
(, 1)
I
/
I
(F,)
implies that the map F : K K
preserves the unit and that (g, 1) = 1

for any g G, namely that is normalized in the K variable and therefore
we could say that C
1,1
(K G, T).
iv) The commutativity of with the associator. This implies that F preserves
the associativity of the multiplications m and m
, and that the following

equation is satised
[F(a
1
)[F(a
2
)[F(a
3
)] [a
2
[a
3
] [a
1
[a
2
a
3
] = [a
1
[a
2
] [a
1
a
2
[a
3
] [a
1
[a
2
[a
3
]
for all a
1
, a
2
, a
3
K. Note that this last equation can be written as
K
= /F
. (3.12)
v) Compatibility with the left and right units, but as the 2-cells F
0
, l, r, l
and
r
are identities, then this implies that [1[a] = 1 = [a[1] and therefore
we have that is normalized in the K variables and we can assume that
C
0,2
(K G, T).
From the previous arguments it follows that F : K K
is a G-equivariant
morphism of monoids, and moreover, calculating the dierential, we have that
(
G

(1)
p
K
)(
1
) =
G
()
K
()
1
G
()
1
K
()
1
= F
/ F
/ F
/,
where the second equality follows from equations (3.10), (3.11) and (3.12). In a
diagram
3 F
/
2
1
F
/
1 F
/
0 1 2
G
(
K
)
1
G
This nishes the proof.
Denition 3.9. Given morphisms T = (F, F

0
, F
2
) and ( = (G, G
0
, G
2
) from
(C, m, I, a) to (C
, m
, I
, a
) pseudomonoids in B, a 2-morphism s : T ( is a
2-cell s : F G in B such that the following diagrams commute:
compatibility with F
2
and G
2
:
m
(F F)
m
(ss)
F2
(GG)
G2
F m
sm
G m
(3.13)
compatibility with F
0
and G
0
:
F I G I
I
F0
G0
sI
(3.14)
Proposition 3.10. Given morphisms of pseudomonoids with strict unit in the 2-
category of G-sets with twists T = (F, , ) and T
= (F,
), with T, T
: /
/
, a 2-morphism : T T
is a cochain : C
0,1
(K G, T) such that
(
G

(1)
p
K
)() = (
/, /
).
Proof. Let : (F, ) (F,
) be the 2-cell dened by the 2-morphism, then we

have that
G
() =
/.
Now, by diagram (3.13) we have that
K
() = /
,
and by diagram (3.14) we have that is a normalized cochain.
Remark 3.11. Propositions 3.4, 3.8 and 3.10 imply that the relevant information
encoded in cochains for the 2-category of pseudomonoids with strict unit of G-sets
with twists, is given by the cochains of the total complex
_
Tot(A
,
(K G, T)),
G

(1)
p
K
_
24 C

For a xed G-equivariant monoid (K, m, 1
K
), Proposition 3.4 tells us that the
3-cocycles Z
3
(Tot
(A
,
(KG, T))) are in one to one correspondence with the set
of possible pseudomonoid structures with strict unit in the 2-category of G-sets
with twist over K. If we only consider invertible morphisms of pseudomonoids as
dened in Proposition 3.8, we may dene a groupoid which encodes the equivalence
classes of pseudomonoid structures over K. Let us be more explicit.
3.4. Equivalence classes of pseudomonoid structures over a xed monoid.
Denition 3.12. Fix a G-equivariant monoid (K, m, 1
K
). Dene the groupoid
Psdmn
G
(K) whose set of objects is Z
3
(Tot
(A
,
(K G, T))) and whose mor-
phisms are invertible morphisms of pseudomonoids as dened in Proposition 3.8.
The groupoid Psdmn
G
(K) encodes the information of all pseudomonoid strucures
over K and its coarse moduli space [Psdmn
G
(K)[, i.e. the set of equivalence classes
dened by the morphisms, is the set of equivalence classes of pseudomonoid struc-
tures on K.
Note that a morphism in Psdmn
G
(K) consists of a triple (F, , ) where F :
K K must be a G-equivariant automorphism. If we denote by
Aut
G
(K) := f Aut(K): f(gk) = gf(k) for all g G
then we have that the group Aut
G
(K) is isomorphic to the subgroup of Aut(KG)
which leaves the G xed; for a G-equivariant automorphism f Aut
G
(K) we can
associate the automorphism f Aut(K G) by the equation
f(k, g) := (f(k), g).
Since the automorphism f leaves G xed, then the groups of automorphism act on
the double complex A
,
(K G, Z); we claim
Lemma 3.13. The set of equivalence classes of pseudomonoid structures on K is
isomorphic can be described by the quotient
[Psdmn
G
(K)[

= H
3
(Tot
(A
,
(K G, T)))/Aut
G
(K).
Proof. We rst perform the quotient with the morphisms (F, , ) where F is the
identity on K; this quotient is precisely H
3
(Tot
(A
,
(K G, T))). Then we see
that elements of H
3
(Tot
(A
,
(KG, T))) lying on the same orbit of the action of
Aut
G
(K) dene equivalent pseudomonoid structures.
In particular we may conclude that if H
3
(Tot
(A
,
(K G, T))) = 0, then all
pseudomonoid structures with strict unit in the 2-category of G-sets with twist over
K are isomorphic to the trivial one. Since we are interested in nding pseudomonoid
structures with strict unit in the 2-category of G-sets with twist over K non iso-
morphic to the trivial one, we will calculate the group H
3
(Tot
(A
,
(K G, T))),
the group Aut
G
(K) and the set [Psdmn
G
(K)[ for some particular examples.
The main tooll we will use in order to calculate the group H
3
(Tot
(A
,
(K
G, T))) will be the Lyndon-Hochschild-Serre spectral sequence. This spectral se-
quence can be obtained if the complex Tot
(A
,
(K G, T)) is ltered by the
complexes
F
n
:= Tot
(A
n,
(K G, T))
thus dening a spectral sequence whose second page becomes
E
p,q
2
= H
p
(G, H
q
(K, T))
where G acts on H
q
(K, T) through the induced action of G on K; note in particular
that
E
0,q
2
= H
q
(K, T)
G
and E
p,0
2
= H
p
(G, T).
On the other hand we have that
Aut
G
(K)

= C
Aut(K)
((G))
where : G Aut(K) determines the action of G, and
C
Aut(K)
((G)) := f Aut(K): [f, (g)] = 1 for all g G
is the centralizer of (G) on Aut(K), consisting of the automorphisms of K which
commute with the G-action. In particular when G = Aut(K) we have that Aut
G
(K) =
Z(Aut(K)). Let us see some examples:
3.4.1. K = Z/p for prime p > 2, and G = Aut(Z/p) = Z/(p 1). Here we have
that Aut
G
(K) = Z(Aut(K))

= Z/(p 1) and that
H
0
(G, H
3
(K, T)) = H
3
(K, T)
G

= (Z/p)
Z/(p1)
= 0
H
1
(G, H
2
(K, T)) = Hom(Z/(p 1), 0) = 0
H
2
(G, H
1
(K, T)) = H
2
(Z/(p 1), Z/p) = 0
where the last equality follows from the fact that H
n
(G, M) is anhilated by [G[ for
all n > 0 [7, III.10.2]. In this case H
3
(Tot
(A
,
(K G, T))) = 0 and therefore
all pseudomonoid structures on K = Z/p are equivalent to the trivial one. This
example could be generalized as follows:
3.4.2. Groups with order relatively prime. Let us recall two facts. First, if G is a
nite group of order m, r is a positive integer with (m, r) = 1 and A
r
= 0, then
H
n
(K, A) = 0 for all n and all subgroups K of G, see [17, Proposition 1.3.1]. And
second, if e is the exponent of H
2
(G, T) then e
2
divides the order of G, see [17,
Theorem 2.1.5].
Then for the case on which [G[ is relatively prime to [K[ we obtain that
H
1
(G, H
2
(K, T)) = 0 and H
2
(G, H
1
(K, T)) = 0.
Therefore we have that H
3
(Tot
(A
,
(K G, T)))

= H
3
(K, T)
G
and
[Psdmn
G
(K)[

= H
3
(K, T)
G
/Aut
G
(K).
3.4.3. The dihedral group D
n
as a semi-direct product. The dihedral group is iso-
morphic to Z/n Z/2 when Z/2 acts on Z/n by multiplication of 1. Since
the induced action of Z/2 on the cohomology ring H
(Z/n, Z)

= Z[x]/nx maps
x x, we have that x
2
x
2
and therefore H
4
(Z/n, Z)
Z/2
= Z/n and
H
2
(Z/2, H
2
(Z/n, Z)) = H
2
(Z/n, Z)
Z/2
=
_
Z/2 if n is even
0 if n is odd.
Since
H
4
(D
n
, Z) =
_
Z/2 Z/2 Z/n if n is even
Z/2 Z/n if n is odd
we know that
H
3
(Tot
(A
,
(Z/n Z/2, T))) =
_
Z/2 Z/n if n is even
Z/n if n is odd,
26 C

and therefore
[Psdmn
Z/2
(Z/n)[ =
_
(Z/2 Z/n)/Aut(Z/n) if n is even
(Z/n)/Aut(Z/n) if n is odd.
because in this case Aut
Z/2
(Z/n) = Aut(Z/n).
In particular, when n = 4 we have that Aut(Z/4) = Z/2. and therefore the
action of Aut
Z/2
(Z/4) on H
3
(Tot
(A
,
(Z/4 Z/2, T))) is trivial. Hence
[Psdmn
Z/2
(Z/4)[ = Z/2 Z/4.
3.5. The case of the group acting on itself by conjugation. Perhaps the
most known pseudomonoid with strict unit in the 2-category of G-sets with twist
was introduced by Dijkgraaf, Pasquier and Roche in [9, Section 3.2] while dening
the quasi Hopf algebra D
w
(G) with for w Z
3
(G; T). In the equations (3.2.5) and
(3.2.6) of [9] they dened a 3-cocycle
w

w

w
Z
3
(Tot(A
,
(GG, T))) by
the equations
1
w
[g[h[[x] :=
w[g[h[x] w[ghxh
1
g
1
[g[h]
w[g[hxh
1
[h]
w
[g[[x[y] :=
w[g[x[y] w[gxg
1
[gyg
1
[g]
w[gxg
1
[g[y]
w
[x[y[z] := w[x[y[z],
where
w
was used to dene the algebra law,
w
to dene the coalgebra law, and
w
encoded the fact that the coproduct is quasicoassociative (and other coherences).
This quasi Hopf algebra D
w
(G) is known as the Twisted Drinfeld Double of G
twisted by w (cf. [25]). Firstly we claim the following.
Lemma 3.14. The 3-cocycle
w

w

w
equals
1
w, the image of w under the
restricted shue homomorphis
1
w dened in (2.2).
Proof. From the following diagrams associated to (
1
w)[g[h[k]

g
h
k

g
hkh
1
h

ghkh
1
g
1
hkh
1
h

g
h
k
ghg
1
g
k

ghg
1
gkg
1
g
we obtain that
w

w

w
=
1
w.
Therefore by Lemma 1.11, Theorem 2.2 and Lemma 2.5 we get
Proposition 3.15. The cohomology class
[
w

w

w
] [w] H
3
(Tot
(A
,
(GG, T))) H
3
(G, T)
1
In order to get exactly the same formul it is necessary to change G by G
op
.
is equal to
[w] where : G G G, (a, g) ag is the multiplication map.

Moreover we obtain the isomorphism
H
3
(G, T) H
3
(Tot
(B
,
(GG, T))) H
3
(Tot
(A
,
(GG, T)))
[w] [x] [
w

w

w
] + [x].
Now, since GG

= GG, we have that the Lyndon-Hochschild-Serre spectral
sequence collapses at the second page. And since the action of G on G is given
by conjugation, then the action of G on H
(G, T) is trivial. Hence we have that

H
3
(Tot(B
,
(GG, T))) sits in the middle of the short exact sequence
0 H
2
(G, Hom(G, T)) H
3
(Tot(B
,
(GG, T))) Hom(G, H
2
(G, T)) 0.
Moreover, in the present situation we have (G) = Inn(G), and therefore
Aut
G
(G) = C
Aut(G)
(Inn(G)),
namely the group of automorphisms of G which commute with all inner automor-
phisms.
With the previous calculations at hand we can calculate the group H
3
(Tot(B
,
(G
G, T))) and Aut
G
(G) in some particular examples:
3.5.1. G simple and non abelian: When G is simple and nonabelian, its abelian-
ization G/[G, G] is trivial. Therefore Hom(G, T) and Hom(G, H
2
(G, T)) are trivial
and hence
H
3
(Tot
(B
,
(GG; T))) = 0.
So we have that
[Psdmn
G
(G)[

= H
3
(G, T)/Aut
G
(G).
When G is the alternating group A
n
we have that Aut
An
(A
n
) = C
Sn
(A
n
) = 1
and therefore
[Psdmn
An
(A
n
)[

= H
3
(A
n
, T), for n ,= 6 and n > 4.
In particular when n = 5 we have that H
3
(A
5
, T) = Z/120 and hence
[Psdmn
A5
(A
5
)[

= Z/120.
3.5.2. Binary icosahedral group. The binary icosahedral group A
5
is a subgroup of
SU(2) that can be obtained as the pullback of the diagram
A
5

SU(2)
A
5

SO(3)
where A
5
embeds in SO(3) as the group of isometries of an icosahedron. This group
satises H
1
(A
5
, Z) = 0, H
2
(A
5
, Z) = 0 and H
3
(A
5
, Z) = Z/120 (see [1, Page 279]),
and therefore H
1
(A
5
, Z) = H
2
(A
5
, Z) = H
3
(A
5
, Z) = 0 and H
4
(A
5
, Z) = Z/120.
Hence H
1
(A
5
, T) = H
2
(A
5
, T) = 0, H
3
(A
5
, T) = Z/120 and H
3
(Tot(B
,
(A
5

A
5
; T))) = 0.
In this case Inn(A
5
)

= A
5
and Aut(A
5
)

= S
5
, therefore Aut
A5
(G) = C
S5
(A
5
) =
1 and
[Psdmn
A5
(A
5
)[

= Z/120.
A similar argument applies to any superperfect group since by denition they
are the ones such that H
1
(G, Z) = 0 and H
2
(G, Z) = 0.
28 C

3.5.3. The dihedral group G = D
n
with n odd. In this case
Hom(D
n
, T) = Z/2, H
2
(D
n
; Z/2) = Z/2 and H
2
(D
n
; T) = 0,
hence
H
3
(Tot
(B
,
(D
n
D
n
; T))) = H
2
(D
n
, Hom(D
n
, T) = Z/2.
Since H
3
(D
n
; T) = Z/2 Z/n we have that the isomorphism classes of pseu-
domonoid structures coming from the Twisted Drinfeld Double construction are
Z/2 Z/n and that there is an independent pseudomonoid structure which comes
fromH
2
(D
n
, Hom(D
n
; T)) = Z/2. In this case we have that Aut
Dn
(D
n
) = Z(D
n
) =
1 and therefore
[Psdmn
Dn
(D
n
)[

= (Z/n Z/2 Z/2).
3.5.4. The symmetric group G = S
n
for n 4. From [1, VI-5] we know that
Hom(S
n
, T) = Z/2, H
2
(S
n
; Z/2) = Z/2
H
2
(S
n
; T) = Z/2 and H
3
(S
n
; Z/2) = Z/2 Z/2
hence we get the exact sequence
0 Z/2 H
3
(Tot
(B
,
(S
n
S
n
; T))) Z/2 0.
In particular we could say that the nontrivial element in H
2
(S
n
, Hom(S
n
, T)) =
Z/2 induces a pseudomonoid structure on S
n
which is not isomorphic to any struc-
ture coming from the construction of the Twisted Drinfeld Double. This follows
from the fact that for n ,= 2 and n ,= 6, Aut(S
n
) = Inn(S
n
) = S
n
and therefore
Aut
Sn
(S
n
) = Z(S
n
) = 1 is the trivial group. Whenever n = 6 we know that
Out(S
6
) = Z/2; nevertheless Aut
S6
(S
6
) = 1. Therefore we have that
[Psdmn
Sn
(S
n
)[ = H
3
(S
n
, T) H
3
(Tot
(B
,
(S
n
S
n
; T)))
with
2
H
3
(S
n
, T) =
_
Z/12 Z/2 if n = 4, 5
Z/12 Z/2 Z/2 if n 6
3.5.5. G cyclic group. When G = Z/n we get that
H
3
(Tot
(B
,
(Z/n Z/n; T))) = H
2
(Z/n, Hom(Z/n, T)) = Z/n
and H
3
(Z/n, T) = Z/n. The action of Aut
Z/n
(Z/n) = Z/n
, which is the multi-

plicative group of units in Z/n, on H
1
(Z/n, T) = Z/n is given by multiplication
and while on H
3
(Z/n, T) = Z/n is given by the square of the multiplication; hence
we get that
[Psdmn
Z/n
(Z/n)[

= (Z/n Z/n)/Z/n
where the action is given by

Z/n
(Z/n Z/n) (Z/n Z/n), (a, (x, y)) (ax, a

2
y).
For example when n = 4 we have that
[Psdmn
Z/4
(Z/4)[

= (Z/4)/(Z/4
) Z/4.
2
See Group Cohomology of Symmetric Groups in http://groupprops.subwiki.org
4. The monoidal category of equivariant vector bundles on a
pesudomonoid
Let G be a group and / = (K, m, 1, , , ) a pseudomonoid with strict unit in
the 2-category of G-sets with twists.
We dene the category Bun
G
(/) of G-equivariant nite dimensional bundles
over K as follows:
An object is a K-graded nite dimensional Hilbert space 1 =
kK
1
k
and a
twisted G-action
: G U(1)
such that
1
k
= 1
k
( h
k
) = [[[[k]() h
k
e h = h
for all , G, k K, h
k
1
k
. Morphisms in the category are linear maps that
preserve the grading and the twisted action, i.e., a linear map f : 1 1
is a
morphism if
f(1
k
) 1
k
,
f( h) = f(h)
for all G, k K and h 1.
We dene a monoidal structure on Bun
G
(/) as follows:
Proposition 4.1. Let 1 and 1
be objects in Bun
G
(/), then the tensor product
of Hilbert spaces 11
is a G-equivariant K-bundle with K-grading (11
)
k
=
x,yK:xy=k
1
x
1
y
and twisted G-action
(h
x
h
y
) := [[[x[y]( h
x
h
y
),
for all k K, G, h
x
1
x
and h
y
1
y
.
Proof. We just need to see that
( (h
x
h
y
)) = [[[[xy]() (h
x
h
y
)
for all , G, x, y K. In fact we have
((h
x
h
y
))
= [[[x[y]
_
( h
x
h
y
)
_
= [[[x[y] [[[x[y]
_
( h
x
) ( h
y
)
_
= [[[x[y] [[[x[y] [[[[x] [[[[y]
_
() h
x
() h
y
_
=
[[[x[y] [[[x[y] [[[[x] [[[[y]
[[[x[y]
_
() (h
x
h
y
)
_
= [[[[xy]
_
() (h
x
h
y
)
_
,
where the last equality follows from the equation
G
()
K
() = 1 obtained in (3.5)
in the proof of Proposition 3.4.
30 C

Now, for 1, 1
and 1
objects in Bun
G
(/) the associativity constraint
: (11
) 1
1(1
),
for the monoidal structure is dened by
((h
x
h
y
) h
z
) = [x[y[z]
1
h
x
(h
y
h
z
)
for all x, , y, z K, h
x
1
x
, h
y
1
y
and h
z
1
z
.
In order to see that is natural isomorphism in Bun
G
(/) we need to see that it
preserves the grading, which follows from the denition, and that is compatible
with the G-action. The compatibility follows from the equalities
( [(h
x
h
y
) h
z
]) = [[[xy[z]( (h
x
h
y
) h
z
)
= [[[xy[z][[[x[y](( h
x
h
y
) h
z
)
=
[[[xy[z][[[x[y]
[ x[ y[ z]
h
x
( h
y
h
z
)
=
[[[x[yz][[[y[z]
[x[y[z]
h
x
( h
y
h
z
)
=
[[[x[yz]
[x[y[z]
h
x
(h
y
h
z
)
= [x[y[z]
1
h
x
(h
y
h
z
)
= ((h
x
h
y
) h
z
),
where the equality between the third and the fourth line follows from the equation
1
()
2
()
1
= 1 obtained in (3.6) in the proof of Proposition 3.4.
The pentagonal axiom follows directly from the 3-cocycle condition of .
Finally we dene the unit object C as the one dimensional Hilbert space C
graded only at the unit element e K, endowed with trivial G-action. From the
normalization of all cochains, it follows that C is unit object in Bun
G
(/) with
respect to the tensor product. All in all, we have just proved that
Proposition 4.2. For / = (K, m, 1, , , ) a pseudomonoid with strict unit in
the 2-category of G-sets with twists, the triple (Bun
G
(/), , C), endowed with the
tensor product and the unit element C is a monoidal category (or tensor category).
Remark 4.3. Let G be a group acting by the right over an abelian group H. Then
G also acts by the left over

H the abelian group of all characters of H. In the case
that the 3-cocycle Z
3
(Tot(A
,
(

H G, T))) is trivial, then the tensor
category Bun
G
(1) is just Rep(H G) the tensor category of nite dimensional
unitary representation of the semi-direct product H G, where G acts by the left
on H as g a := ag
1
, for all g G, a H.
Remark 4.4. Consider the case of a group G acting on itself by conjugation
and the pseudomonoid /
w
G
= (G, m, 1,
w
,
w
,
w
) dened in section 3.5. In the
case that G is nite, the tensor category Bun
G
(/
w
G
) is exactly the category of
representations Rep(D
w
(G)) of the Twisted Drinfeld Double D
w
(G). Note that the
quasi-Hopf algebra D
w
(G) is dened only for G nite, but Bun
G
(/
w
G
) is dened for
an arbitrary discrete group. So, the tensor category Bun
G
(/
w
G
) is a generalization
of the Twisted Drinfeld Double of a nite group.
The Twisted Drinfeld Double D
w
(G) is important at least for the following
two reasons. First, D
w
(G) is a quasi-triangular quasi-Hopf algebra [9] and then
Rep(D
w
(G)) is a braided tensor category and it denes representations of the braid
group and knots and links invariants [24, Chapter 1]. Second, Rep(D
w
(G)) is a
Modular Tensor Category (MTC), so it denes a 3D-TQFT using Reshetikhin-
Turaev construction, see [3, Chapter 4] and [24, Chapter 4], .
For any group, the tensor category Bun
G
(/
w
G
) is a braided tensor category, so it
denes representations of the braid group and knots and links invariants. However
Bun
G
(/
G
) is not a MTC if G is innite. In fact, the category of nite dimensional
unitary representation of G is a full tensor subcategory of Bun
G
(/
w
G
) and it has
innite many isomorphism classes of irreducible representations, so in particular
Bun
G
(/
w
G
) would also have innitely many simple objects.
4.1. Morphism of pseudomonoids, monoidal functors and natural isomor-
phisms. A morphism in the 2-category of pseudomonoids in the 2-category G-sets
with twists induce a monoidal functor between the associated monoidal categories.
Proposition 4.5. Let T = (F, , ) : / /
be a morphism of pseudomonoids.
Then T induces a monoidal functor from the monoidal categories (Bun
G
(/), , C)
and (Bun
G
(/
),
, C
).
Proof. Let T = (F, , ) : / /
be a morphism of pseudomonoids as dened in

Proposition 3.8. We dene a functor F : Bun
G
(/) Bun
G
(/
) in the following
way: for 1 an object in Bun
G
(/), the K
-graded Hilbert space F(1) is the direct

sum
F(1)
y
=
{xK:F(x)=y}
1
x
.
The twisted G-action on F(1) is dened as follows: take h
y
F(1)
y
dened by
the element h
y
= h
x
for some vector h
x
1
x
with F(x) = y. Dene the twisted
G-action
on h
y
by

y
:= [[[x]( h
x
),
and note that

y
) =
([[x] h
x
)
= [[[x] [[x] ( h
x
)
= [[[x] [[x] [[[[x] h
x
=
[[[[y] [[[x] h
x
=
[[[[y]
y
where the equality between the third and the fourth lines follows from the equation
G
= F
/ obtained in (3.10) in the proof of Proposition 3.8. Therefore the

Hilbert space F(1) is an object in Bun
G
(/
).
Let us check that the functor F is monoidal: take 1, 1
two objects in Bun

G
(/)
and consider the map
R : F(1)
F(1
) F(11
)
h
x1

h
x2
(x
1
, x
2
)
1
h
x1
h
x2
where h
x1
1
x1
and h
x2
1
x2
but we see them both as elements in F(1)
F(x1)
and F(1
)
F(x2)
respectively, and the element h
x1
h
x2
we see it as an element in
32 C

F(1 1
)
F(x1x2)
. Let us show that the map R is a morphism in Bun
G
(/
), i.e.
that R(
(h
x1

h
x2
)) =
R(h
x1

h
x2
). For he left hand side we have:
R(
(h
x1

h
x2
))
=
[[[F(x
1
)[F(x
2
)] [[[x
1
] [[[x
2
]R( h
x1

h
x2
)
=
[[[F(x
1
)[F(x
2
)] [[[x
1
] [[[x
2
] [x
1
[x
2
]
1
( h
x1
h
x2
),
and for the right hand side we have:

R(h
x1

h
x2
) =
[x
1
[x
2
]
1
(h
x1
h
x2
)
= [[[x
1
x
2
] [[[x
1
[x
2
] [x
1
[x
2
]
1
( h
x1
h
x2
);
and note that the two expressions coincide due to the equation
G

K
= /F
that appeared in (3.11) under the proof of Proposition 3.8.

The hexagonal axiom of the monoidal functor follows directly from the equation
K
= /F
and we will not reproduce its proof.

Proposition 4.6. Given morphisms of pseudomonoids with strict unit in the 2-
category of G-sets with twists T = (F, , ) and T
= (F,
), with T, T
: /
/
, a 2-morphism : T T
, induces a monoidal natural isomorphism between

the monoidal functors F and F
.
Proof. Using the notation dened above, we dene the transformation between F
and F
as follows:
F(1) F
(1)
h
x
(x)
1
h
x
.
Equations
1
=
/ and
2
= /
shown in the proof of Proposition 3.10 imply

that the transformation is natural and monoidal, respectively.
4.2. Automorphisms of pseudomonoids and their action on the monoidal

category of equivariant vector bundles. Let us x / = (K, m, 1, , , ) a
pseudomonoid with strict unit in the 2-category of G-sets with twists and let
(Bun
G
(/), , C) be the monoidal category of G-equivariant bundles over K.
Take T = (F, , ) : / / an invertible morphism of the pseudomonoid / and
note that Proposition 4.5 tells us that the induced monoidal functor
F : (Bun
G
(/), , C) (Bun
G
(/), , C)
becomes an automorphism of the monoidal category (Bun
G
(/), , C).
If we denote by
Aut
Psmnd
(/)
the 2-group of automorphisms of the pseudomonoid /, whose morphisms are invert-
ible morphisms T : / / and whose 2-morphisms are the natural transformations
between functors : T T
, and
Aut
(Bun
G
(/))
the 2-group of automorphisms of the monoidal category (Bun
G
(/), , C), whose
morphisms are invertible monoidal functors and whose 2-morphisms are monoidal
natural transformations, then we have that Propositions 4.5 and 4.6 imply that
there is a 2-functor
Aut
Psmnd
(/) Aut
(Bun
G
(/))
: T T
: F F
from the 2-group of automorphisms of the pseudomonoid / to the 2-group of au-

tomorpshisms of Bun
G
(/).
To understand the previous action in more detail, let us start by studying the
category Aut
Psmnd
(/).
An automorphism T = (F, , ) : / / consists of a G-equivariant automor-
phism F Aut
G
(K), together with a degree 2 cochain
1
in Tot
(A
,
(K
G, T)) such that
(
G

(1)
p
K
)(
1
) = F
/ F
/ F
/.
The automorphism F lies on the image of the forgetful functor
Aut
Psmnd
(/) Aut
G
(K)
T = (F, , ) F;
if and only if the cohomology classes [ ] and F
[ ] are equal as
cohomology classes in H
2
(Tot
(A
,
(K G, T))). If we dene
Aut
G
(K; [ ]) := F Aut
G
(K)[F
[ ] = [ ]
we have that the 2-group of automorphisms of / sits in the exact sequence
0
Tot
1
(A
,
(K G, T)))
Z
2
(Tot
(A
,
(K G, T)))

Aut
Psmnd
(/)

Aut
G
(K; [ ])

0
where Z
2
(Tot
(A
,
(K G, T))) denotes the degree 2-cocycles and Tot
1
(A
,
(K
G, T)) parameterizes the 2-morphisms between the morphisms of Z
2
(Tot
(A
,
(K
G, T))).
If we take equivalence classes of automorphisms in Aut
Psmnd
(/) dened by the
2-morphisms, we obtain a group which is usually denoted by
1
(Aut
Psmnd
(/)) ;
this group sits in the middle of the short exact sequence
0 H
2
(Tot
(A
,
(K G, T)))
1
(Aut
Psmnd
(/)) Aut
G
(K; [ ]) 0;
and by the Lyndon-Hochschild-Serre spectral sequence we know that there is an
exact sequence
0 H
1
(G, Hom(K, T)) H
2
(Tot
(A
,
(KG, T))) H
2
(K, T)
G
d2
H
2
(G, Hom(K, T)),
where d
2
is the dierential of the second page.
34 C

Furthermore, if we take the group of 2-morphisms of the identity morphism in
Aut
Psmnd
(/), we obtain a group which is usually denoted by
2
(Aut
Psmnd
(/))
and is equal to H
1
(Tot(A
,
(K G, T))) = Hom(K, T)
G
.
4.3. The Grothendieck ring associated to the monoidal category. Consider
the isomorphism classes of objects in the monoidal category Bun
G
(/). Since the
objects could be understood as nite dimensional vector spaces which are K-graded
endowed with a projective G-action, we can add them up and moreover we can
multiply them by using the tensor product of the monoidal category. What we
obtain is a semi-ring which we can make into a ring by applying the standard
Grothendieck construction argument of K-theory. Denoting by Groth (Bun
G
(/))
the Gronthendieck ring constructed from the monoidal category Bun
G
(/), we have
a functor
Psmnd
G
Rings
/ Groth (Bun
G
(/))
from the 2-category of pseudomonoids with strict unit in the 2-category G-sets with
twists, to the category of rings.
The ring Groth (Bun
G
(/)) can also be understood as the -twisted G-equivariant
K-theory of the monoid K where the multiplication is induced by the pushforward
m
of the multiplication m : KK K. This twisted K-theory ring was the main

motivation of this work and is the subject of the next section.
In the case on which G = K and G acts on G by the left adjoint action
we have seen that any 3-cocycle w Z
3
(G; T) induces a 3-cocycle
w

w

w
Z
3
(Tot
(A
,
(G G, T))) that makes / := (G, m, 1,
w
,
w
,
w
) into a pseu-
domonoid with strict unit in the 2-category G-sets with twists. In this case the
Grothendieck ring Groth (Bun
G
(/)) is isomorphic to the Grothendieck ring of rep-
resentations
Groth(Rep(D
w
(G)))
of the Twisted Drinfeld Double D
w
(G) of the group G (see [25, Section 3]), which
is also isomorphic to the w-twisted stringy K-theory
w
K
st
([G/G])
of the groupoid [G/G] [25, Prop. 18] c.f. [2, 5, 14].
4.3.1. Automorphisms. Since the 2-functor
Aut
Psmnd
(/) Aut
(Bun
G
(/))
induces a homomorphism
1
(Aut
Psmnd
(/))
1
(Aut
(Bun
G
(/))),
we have that there is a homomorphism of groups
1
(Aut
Psmnd
(/)) Aut(Groth(Bun
G
(/)))
which composed with the inclusion
H
2
(Tot(A
,
(K G, T)))
1
(Aut
Psmnd
(/))
denes a homomorphism
H
2
(Tot(A
,
(K G, T))) Aut(Groth(Bun
G
(/))).
The previous morphism will be of interest when we compare it with the group
of automorphisms of the twisted equivariant K-theory ring in section 5.7.1.
5. The fusion product and the twisted G-equivariant K-theory ring
Whenever X is a nite G-CW complex with G a nite group, the elements in
H
3
G
(X; Z) classify the isomorphism classes of projective unitary stable and equivari-
ant bundles over X, and these bundles provide the required information to dene
equivariant Fredholm bundles over X; the homotopy groups of the space of section
of a such bundle is one way to dene the twisted G-equivariant K-theory groups of
X (see [4]). The homotopy classes of automorphisms of a projective unitary stable
and equivariant bundle over X are in one to one correspondence with H
2
G
(X; Z)
and this group acts on the twisted G-equivariant K-theory groups.
Whenever the space X is a discrete G-set, there is an equivalent but easier way
to dene the twisted G-equivariant K-theory groups of X. Let us review it.
5.1. Twisted G-equivariant K-theory. Take a normalized 2-cocycle : GG
X T and dene an -twisted G-vector bundle over X as a nite dimensional X-
graded complex vector space E, which can alternatively seen as a nite dimensional
complex vector bundle p : E X with nite support, endowed with a G action
such that p is G equivariant, the action of G on the bers is complex linear, and
such that the composition of the action on E satises the equation
g (h z) = (g, h[[p(z))(gh z)
for all z in E. Two -twisted G-vector bundles over X are isomorphic if there exists
a G equivariant map E E
of complex vector bundles which is an isomorphism

of vector bundles.
Denition 5.1. The -twisted G-equivariant K-theory of X is the Grothendieck
group
KU
G
(X; )
associated to the semi-group of isomorphism classes of -twisted G-vector bundles
over X endowed with the direct sum operation.
If we have a G-equivariant map F : Y X then the pullback of bundles induces
a group homomorphism
F
: KU
G
(X; ) KU
G
(Y ; F
).
5.1.1. For a normalized cochain C
1
G
(X; T) with
G
=
/ then there is an
induced isomorphism of groups
: KU
G
(X; )
=
KU
G
(X;
)
where (E) := E and the G-action
on z (E) is given by the equation:

h
z := [h[[p(z)](h z).
Since cohomologous twistings induce isomorphic twisted K-theory groups, we
have that H
2
G
(X; T) classies the isomorphism classes of twistings for the G-equivariant
K-theory of X. And since the isomorphisms and (
G
) are equal, we have
36 C

that the group H
1
G
(X; T) acts on the -twisted G-equivariant K-theory group
KU
G
(X; ) by automorphisms.
5.2. Pushforward. For a G-equivariant map F : Y X and Z
2
G
(X; T) there
is a pushforward map
F
: KU
G
(Y ; F
) KU
G
(X; )
dened at the level of vector bundles as follows
(F
E)
x
:=
{yY |F(y)=x}
E
y
where the G-action on F
E is the one induced by the G-action on Y and the

G-action on E.
5.3. External product. If we consider two G-sets with twist (X,
X
) and (Y,
Y
),
the external product is the homomorphism
KU
G
(X;
X
) KU
G
(Y ;
Y
)

KU
G
(X Y ;
Y
)
where (E F)
(x,y)
:= E
x
F
y
and
1
and
2
denote the projections of X Y on
the rst and the second coordinate respectively.
5.4. Multiplicative structures on Twisted Equivariant K-theory. In the
particular case on which the G-set X is endowed with the additional structure of a
G-equivariant multiplication map
m : X X X
and moreover that the cohomology class [] of the twisting is multiplicative i.e.
1
[]
2
[] = m
[], then the -twisted G-equivariant K-theory group can be en-

dowed with a product structure. This construction could be done for G-equivariant
H-spaces, but for clarity we will restrict ourselves to the case on which the G-set is
a G-equivariant monoid with unit.
Let K be a G-equivariant discrete monoid with unit and denote by m : K
K K the multiplication of the monoid. Take a twist Z
2
G
(K; T) that is
multiplicative, i.e. that there exist a cochain C
1
G
(K K; T) such that
G
=
m
or equivalently
G

K
= 1, then we can compose the following morphisms
KU
G
(K; ) KU
G
(K; )

KU
G
(K K;
2
)

KU
G
(K K;m
)
m
KU
G
(K; )
thus producing a product structure
: KU
G
(K; ) KU
G
(K; ) KU
G
(K; )
(E, F) m
(((E, F))).
It is a simple calculation to see that the product previously dened is associative
whenever the cohomology class [] H
1
G
(K K; T) satises the equation
K
[] = 1
as a cohomology class in H
1
G
(KKK; T). We therefore have that if there exists
a cochain C
0
G
(K K K; T) such that
G
=
K
then the product previously dened endows the group KU

G
(K; ) with a ring
structure. Summarizing:
Proposition 5.2. Consider Z
2
G
(K; T) and C
1,2
(K G; T) satisfying the
equations
G

K
= 1,
2
[] = 1.
Then the group KU
G
(K; ) endowed with the product structure
becomes a ring.
Let us denote this ring by
KU
G
(K; , ) := (KU
G
(K; ),
)
and let us call the pair (, ) a multiplicative structure for K.
5.4.1. Many of the features of the twisted G-equivariant K-theory rings are better
understood if we work with the notation introduced in section 1.
Recall that the double complex A
,
:= A
,
(K G; T) is the subcomplex of
C
,
(KG; T) disregarding the 0-th row. Consider the subcomplex A
,>3
of A
,
dened as subcomplex of C
,
(K G; T) where we disregard the rst four rows.
The double complex A
,
/A
,>3
consists of the second, third and fourth rows of
the double complex C
,
(K G; T), and we have that if KU
G
(K; ) can be made
into a ring is because there exists and such that the cochain becomes
a 3-cocycle in the complex Tot
(A
,
/A
,>3
) and the 3-cocycle can be seen in a
diagram as follows
4
3 1
2 1
1 1
0 1 2 3
G
(
K
)
1
G
Proposition 5.3. If the cocycle Z
2
G
(K; T) can be lifted to a 3-cocycle
in the complex Tot
(A
,
/A
,>3
) then the the group KU
G
(K; ) can be endowed
with the ring structure KU
G
(K; , ).
5.4.2. Let us see another way to understand the conditions under which the twist
can dene a multiplicative structure on KU
G
(K; ). Consider the ltration of
the double complex A
,
given by the subcomplexes F
r
:= A
,r
. The spectral
sequence that the ltration denes abuts to the cohomology of the total complex
of A
,
E
,
(Tot(A
,
))
and has for rst page
E
p,q
1
= H
p
G
(K
q
; T)
38 C

with dierential
d
1
: E
p,q
1
E
p,q+1
1
, d
1
[x] := [(
K
)
(1)
p
x].
If we have a twist , its cohomology class [] is an element in E
2,1
1
. The element
d
1
[] is the rst obstruction to lift to a 3-cocycle in Tot
(A
,
), that is, d
1
[] = 1
if and only if there exists C
1,2
(K G; T) such that
G

K
= 1. Note
furthermore that
d
1
[] =
m
[]
1
[]
2
[]
and therefore we recover what we knew, namely that the twist may induce a
product in KU
G
(K; ) if and only if the cohomology class [] is multiplicative.
If the cohomology class [] is multiplicative, then [] survives to the second page
of the spectral sequence with [] E
2,1
2
. The second dierential applied to [] is
d
2
[] = [(
2
)
1
]
3 (
K
)
1
2 1
1
0 1 2
G
(
K
)
1
3 [(
K
)
1
]
2
1 []
0 1 2
d2
and this is the second obstruction to lift to a 3-cocycle in Tot
(A
,
), i.e. d
2
[] = 1
if and only if there exists C
0,3
(K G, T) such that
G
(
K
)
1
= 1. Note
that d
2
[] measures the obstruction for the multiplication
in KU
G
(K; ) to be
associative.
We have then that the equations d
1
[] = 1 = d
2
[] are the equations that need
to be satised in order for the group KU
G
(K; ) to become a ring with respect to
the construction provided in this section.
If [] survives to the third page we have that d
3
[] = [
K
] and therefore d
3
[] =
1 implies that is a 3-cocycle in Tot
(A
,
). So if [] survives to the fourth
page, and hence the page at innity, then / = (K, m, 1, , , ) is a pseudomonoid
with strict unit in the 2-category of G-sets with twists. Summarizing:
Proposition 5.4. Consider Z
2
G
(K; T), then can be lifted to a 3-cocycle
in Tot
(A
,
/A
,>3
) if and only if d
1
[] = 1 = d
2
[], and this implies
that KU
G
(K; , ) becomes a ring. If furthermore d
3
[] = 1 then is a 3-
cocycle in Tot(A
,
) and this implies that / = (K, m, 1, , , ) is a pseudomonoid
with strict unit in the 2-category of G-sets with twists, and in this case
KU
G
(K; , )

= Groth (Bun
G
(/)) .
5.5. Isomorphism classes of Multiplicative structures. In this section we
want to determine some sucient conditions under which the twisted G-equivariant
K-theory rings KU
G
(K; , ) and KU
G
(K;
) induced by the multiplicative

structures (, ) and (
) become isomorphic.
Let us consider the double complex A
,
/A
,>2
which consists of the rst and
the second row of the double complex C
,
(K G, T). We claim
Lemma 5.5. Consider (, ) and (
) multiplicative structures on K. If
and
are cohomologous as 3-cocycles in Tot
(A
,
/A
,>2
), then the rings
KU
G
(K; , ) and KU
G
(K;
) are isomorphic.
Proof. If and
are cohomologous as 3-cocycles in Tot
(A
,
/A
,>2
),
then there exists a 2-cochain
1
with A
1,1
and A
0,2
such that
G

(1)
p
K
(
1
) =
/,
namely that
G
=
/ and (
K
)
1
(
G
)
1
=
/, or diagramatically
3
2
1
/
1
/
0 1 2
G
(
K
)
1
G
The isomorphism : KU
G
(K; )
=
KU
G
(K;
) induces an isomorphism of
rings
: KU
G
(K; , )
=
KU
G
(K;
)
where
:= (
K
)
1
. Since (
K
)
1
(
G
)
1
=
/, we have that
(
G
),
therefore the isomorphism
and
are equal and we obtain that KU

G
(K;
)

=
KU
G
(K;
).
The short exact sequence of complexes
0 A
,>2
/A
,>3
A
,
/A
,>3
A
,
/A
,>2
0
induces a long exact sequence in cohomology groups
H
3
(Tot
(A
,
/A
,>3
)) H
3
(Tot
(A
,
/A
,>2
)) H
4
(Tot
(A
,>2
/A
,>3
))
[ ] [ ] [(
2
)
1
]
and we see that Lemma 5.5 and Proposition 5.4 imply that the subgroup
MS
G
(K) := [ ] H
3
(Tot
(A
,
/A
,>2
))[[
K
] = 1
is precisely the group of equivalence classes of multiplicative structures associated
to the twisted G-equivariant K-theory of the monoid K. We dene
Denition 5.6. The group
MS
G
(K) := [ ] H
3
(Tot(A
,
/A
,>2
))[[
2
] = 1
will be called the group of multiplicative structures for the G-equivariant K-theory
of the monoid K.
And therefore we have that
Proposition 5.7. The elements of the group MS
G
(K) are in one to one corre-
spondence with the set of isomorphism classes of ring structures (in the sense of
Lemma 5.5) on the twisted G-equivariant K-theory to the monoid K.
In particular we have that there are at most #(MS
G
(K)) of dierent multiplica-
tive structures in the twisted G-equivariant K-theory groups of K.
40 C

5.6. Automorphisms of the twisted equivariant K-theory ring. From sec-
tion 5.1.1 we know that if C
1
G
(K; T) satises
G
= 1 then the map induces
an isomorphism of groups
: KU
G
(K; )
=
KU
G
(K; ).
Whenever (, ) is a multiplicative structure, it follows that the map induces an
isomorphism of rings whenever the homomorphism
K
is the identity map, namely
that [
K
] = 1 as a cohomology class in H
1
G
(K K; T). If we dene the group of
multiplicative elements by
H
1
G
(K; T)
mult
:=
_
[] H
1
G
(K; T)[
K
[] =
1
[]
2
[] m
[]
1
= 1
_
we have then that the group of automorphisms of the twisted G-equivariant K-
theory ring is equal to the multiplicative elements in H
1
G
(K; T), i.e.
Aut(KU
G
(K; , )) = H
1
G
(K; T)
mult
Using the spectral sequence of of section 5.4.2 we see that the multiplicative
terms appear in the second page of the spectral sequence
H
1
G
(K; T)
mult
= E
1,1
2
since d
1
[] = [
K
]
1
is the obstruction of being multiplicative. If furthermore a
multiplicative element satises d
2
[] = 1, then we have that [] can be lifted to an
element [ ] in H
2
(Tot(A
,
)). This means that we have the exact sequence
0 H
2
(C
(K; T)
G
) H
2
(Tot(A
,
)) H
1
G
(K; T)
mult
d2
H
3
(C
(K; T)
G
) (5.1)
where the G-invariant cochains come from the rst page of the spectral sequence,
i.e. E
0,q
1
= C
q
(K; T)
G
, and its cohomology appears in the second page, i.e. E
0,q
2
=
H
q
(C
(K; T)
G
,
K
).
5.7. Relation between the Grothendieck ring associated to a monoidal
category and the twisted equivariant K-theory ring. Let us consider a G-
equivariant monoid with unit K. We have seen that to a pseudomoinoid with strict
unit in the 2-category of G-sets with twist / = (K, m, 1, , , ) over K we can
associate the ring Groth(Bun
G
(/)) of isormpism classes of objects in the monoidal
category Bun
G
(/). At the same time, since Z
3
(Tot
(A
,
(KG, T))),
then (, ) is a multiplicative structure and we get that
Groth(Bun
G
(/))

= KU
G
(K; , )
as rings. We have therefore a canonical map
H
3
(Tot(A
,
)) MS
G
(K) (5.2)
[ ] [ ]
from the isomorphism classes of pseudomonoid structures with strict unit of G-sets
with twist over K, to the group of multiplicative structures of the G-equivariant
twisted K-theory groups of K. Let us understand this map in more detail.
Consider the projection homomorphism between the exact sequences of com-
plexes
0

A
,>2
p
A
,
A
,
/A
,>2

0
0

A
,>2
/A
,>3

A
,
/A
,>3

A
,
/A
,>2

0
and the homomorphism between the long exact sequences induced
H
3
(Tot(A
,
)
H
3
(Tot(A
,
/A
,>2
))
H
4
(Tot(A
,>2
)
p
H
3
(Tot(A
,
/A
,>3
))
H
3
(Tot(A
,
/A
,>2
))

H
4
(Tot(A
,>2
/A
,>3
))

where and are the connection homomorphisms with [ ] =

2
[].
By denition
MS
G
(K) = ker( )
and furthermore note that the map is injective since H
3
(Tot(A
,>3
)) = 0. There-
fore the canonical map dened in (5.2) is precisely the map
H
3
(Tot(A
,
))

MS
G
(K).
In complete generality it is dicult to give an explicit description of how the ker-
nel and the cokernel of the map H
3
(Tot(A
,
))

MS
G
(K) looks like, but for
calculations we can give the following description:
Theorem 5.8. The kernel of the homomorphism H
3
(Tot(A
,
))

MS
G
(K) is
isomorphic to
ker
_
H
3
(Tot(A
,
))

MS
G
(K)
_

=
_
H
3
(C
(K, T)
G
)
_
and the cokernel is isomorphic to
coker
_
H
3
(Tot(A
,
))

MS
G
(K)
_

= ker
_
H
4
(C
(K, T)
G
) H
4
(Tot(A
,
))
_
.
Proof. Let us start with the kernel of the map . From the long exact sequences of
cohomologies dened above we get that
ker
_
H
3
(Tot(A
,
))

MS
G
(K)
_

=
_
H
3
(Tot(A
,>2
))
_
where the group H
3
(Tot(A
,>2
)) consists of elements in C
0,3
that are closed under
the dierentials
G
and
K
and therefore
H
3
(Tot(A
,>2
)) = Z
3
(K, T)
G
.
Now, since the elements
K
(C
2
(K, T)
G
) are all zero in H
3
(Tot(A
,
)) we have that
_
Z
3
(K, T)
G
_
=
_
H
3
_
C
(K, T)
G
__
and therefore
ker
_
H
3
(Tot(A
,
))

MS
G
(K)
_

=
_
H
3
(C
(K, T)
G
)
_
.
42 C

For the cokernel we have that the long exact sequence in cohomologies dened
above implies that
coker
_
H
3
(Tot(A
,
))

MS
G
(K)
_

= H
4
(Tot(A
,>2
)) ker(p) ker().
Now, the projection map
p : H
4
(Tot(A
,>2
)) H
4
(Tot(A
,>2
/A
,>3
)) = H
1
G
(K
3
, T)
has for kernel the elements in the fourth cohomology of the G-invariant K-chains
ker(p) = H
4
(C
(K, T)
G
),
and this follows from the spectral sequence that converges H
(Tot(A
,>2
)) asso-
ciated to the ltration Tot(A
>q,
). Since the natural homomorphism
H
4
(C
(K, T)
G
) H
4
(Tot(A
,
))
coincides with the map , we have the desired isomorphism
coker
_
H
3
(Tot(A
,
))

MS
G
(K)
_

= ker
_
H
4
(C
(K, T)
G
) H
4
(Tot(A
,
))
_
.
For calculation purposes let us understand the kernel and the cokernel of the
map from the point of view of the spectral sequence of section 5.4.2.
Proposition 5.9. Consider the ltration of the complex Tot(A
,
) dened by the
subcomplexes Tot(A
,>q
) and consider the spectral sequence that it denes which
converges to H
(Tot(A
,
). Then we have the isomorphisms
ker
_
H
3
(Tot(A
,
))

MS
G
(K)
_

= E
0,3
4
coker
_
H
3
(Tot(A
,
))

MS
G
(K)
_

= d
3
(E
2,1
3
) +d
2
(E
1,2
2
).
In the particular case on which the spectral sequence collapses at the second page
we conclude that
0 H
3
(C
(K, T)
G
) H
3
(Tot(A
,
))

MS
G
(K) 0.
Proof. The rst page of the spectral sequence associated to the ltration Tot(A
,q>
)
is given by E
0,q
1
= C
q
(K, T)
G
and E
p,q
1
= H
p
G
(K
q
, T) whenever q > 0, and therefore
one has that the second page is given by
E
0,q
2
=H
q
(C
(K, T)
G
)
and for p > 0 and q > 0
E
p,q
2
=
ker
_
H
p
G
(K
q
, T)

K
H
p
G
(K
q+1
, T)
_
im
_
H
p
G
(K
q1
, T)

K
H
p
G
(K
q
, T)
_ .
The map
H
4
(C
(K, T)
G
) H
4
(Tot(A
,
))
coincides with the standard map E
0,4
2
H
4
(Tot(A
,
)) and its kernel consists of
the images of E
1,2
2
and E
2,1
3
under the dierentials d
2
and d
3
respectively, since we
know that the map
E
0,4
4
H
4
(Tot(A
,
)).
is injective. Therefore we have that
d
2
(E
1,2
2
) ker
_
H
4
(C
(K, T)
G
) H
4
(Tot(A
,
))
_
and the above inclusion is an equality whenever d
3
(E
2,1
3
) = 0. In the case that
d
3
(E
2,1
3
) ,= 0 we could abuse of the notation and say that
d
3
(E
2,1
3
) +d
2
(E
1,2
2
) = ker
_
H
4
(C
(K, T)
G
) H
4
(Tot(A
,
))
_
.
A similar argument could be used to calculate
_
H
3
(C
(K, T)
G
)
_
. Since E
0,3
2
=
H
3
(C
(K, T)
G
) we have that its image (E
0,3
2
) H
3
(Tot(A
,
)) is equal to the
image of the canonical map
E
0,3
2
H
3
(Tot(A
,
)).
Since the image is isomorphic to the group to which it converges, in this case E
0,4
4
,
then we can conclude that
_
H
3
(C
(K, T)
G
)
_

= E
0,3
4
and therefore
ker
_
H
3
(Tot(A
,
))

MS
G
(K)
_

= E
0,3
4
.
Finally, whenever the spectral sequence collapses at the second page we have
that d
2
= 0 = d
3
and therefore is surjective. Since we have in this case we have
that E
0,3
4
= E
0,3
2
= H
3
(C
(K, T)
G
), the proposition follows.
From Theorem 5.8 we can deduce two things.
If / = (K, m, 1, , , ) is a pseudomonoid with strict unit in the 2-category
of G-sets with twists such that [ ] lies in the image of , then the
Grothendieck ring Groth(Bun
G
(/)) is isomorphic to the untwisted ring
KU
G
(K).
Multiplicative structures (
) in MS
G
(K) such that
belongs to
Z
3
(Tot(A
,
/A
,>3
)) and
K
,= 0, dene ring structures KU

G
(K;
)
which cannot be obtained as the Grothendieck ring Groth(Bun
G
(/)) for
any pseudomonoid / with strict unit in the 2-category of G-sets with twists.
5.7.1. Relation between the automorphisms. We have seen that the isomorphism
classes of automorphisms of / that leave the monoid K xed is isomorphic to the
group H
2
(Tot(A
,
(K G, T))). Since the automorphism group of KU
G
(K; , )
is H
1
G
(K; T)
mult
we have that there is an induced map
H
2
(Tot(A
,
)) H
1
G
(K; T)
mult
which matches the homomorphism of that appears in the exact sequence (5.1)
0 H
2
(C
(K; T)
G
) H
2
(Tot(A
,
)) H
1
G
(K; T)
mult
d2
H
3
(C
(K; T)
G
).
Note that in the case that the spectral sequence collapses at the second page we
get the short exact sequence
0 H
2
(C
(K; T)
G
) H
2
(Tot(A
,
)) H
1
G
(K; T)
mult
0.
A more elaborate analysis of the homomorphisms
H
3
(Tot(A
,
))

MS
G
(K) and H
2
(Tot(A
,
)) H
1
G
(K; T)
mult
will depend on the choice of the group G and of the G-equivariant monoid K. In the
next chapter we will calculate explicitly the previous homomorphisms for several
44 C

examples, and from them we will deduce interesting information with regard to the
twisted equivariant K-theory rings.
6. Examples
The main objective of this section is to use Proposition 5.9 to calculate the kernel
and cokernel of the homomorphism
H
3
(Tot(A
,
))

MS
G
(K)
for dierent choices of G and K, in order to show the dierent twisted G-equivariant
K-theory rings over K that can appear.
6.1. Trivial action of G on K. In this case we have that the spectral sequence
dened in Proposition 5.9 collapses at the second page and moreover we have that
C
(K, T) = C
(K, T)
G
. Therefore we obtain the short exact sequence
0 H
3
(K, T) H
3
(Tot(A
,
))

MS
G
(K) 0,
thus implying that
MS
G
(K)

= H
3
(Tot
(B
,
(K G, T))),
and moreover that all multiplicative structures for the G-equivariant K-theory of
K can be obtained from the ring structures dened by the Grothendieck rings of
the monoidal categories Bun
G
(/). We also obtain the short exact sequence
0 H
2
(K, T) H
2
(Tot(A
,
)) Hom(K, Hom(G, T)) 0
where in this case H
1
G
(K, T)
mult
= Hom(K, Hom(G, T)).
Furthermore, if [] H
3
(K, T) is non trivial then we can dene a non-trivial
pseudomoinoid with strict unit in the 2-category of G-sets with twist / = (K, m, 1, 0, 0, )
with [0 0 ] non-zero in H
3
(Tot(A
,
)), such that
Groth(Bun
G
(/))

= R(G)
Z
Z[K]
where R(G) is the Grothendieck ring of nite dimensional complex representations
of G and Z[K] is the group ring of K, since we know that R(G)
Z
Z[K] is isomorphic
to the non-twisted ring structure on KU
G
(K).
6.1.1. G = Z/n and K = Z/m. In this case we have that H
3
(K, T) = Z/m and
H
3
(Tot(A
,
)) = Z/mZ/(n, m)
where (n, m) is the maximum common denominator of the pair n, m. Therefore
MS
G
(K) = Z/(n, m) and we have that all non trivial multiplicative structures
come from the group
Z/(n, m) = H
3
(Tot(B
,
)) H
3
(Tot(A
,
)).
6.2. Adjoint action of G on itself. From Lemma 2.5 we know that in this case
we have the split short exact sequence
0

H
(Tot(B
,
))

H
(Tot(A
,
))

H
(G, T))
0
induced by the short exact sequence of complexes
0 Tot(B
,
) Tot(A
,
) C
>0
(G, T) 0.
Since we have the inclusion H
(Tot(B
,
)) H
(Tot(A
,
)), we can deduce
that the groups E
,0
2
of the 0-th column of the second page of the spectral se-
quence converging to H
(Tot(A
,
)) dened in Proposition 5.9, are unaected by
the dierentials d
i
for i > 1; this follows from the injectivity between the spectral
sequences associated to the ltrations B
,>q
and A
,>q
of Tot(B
,
) and Tot(A
,
)
respectively.
Therefore we have that E
0,3
4
= E
0,3
2
= H
3
(C
(G, T)
G
) and d
2
(E
1,2
2
) = 0 =
d
3
(E
2,1
3
), and by Proposition 5.9 we get the short exact sequence
0 H
3
(C
(G, T)
G
) H
3
(Tot(A
,
))

MS
G
(G) 0. (6.1)
Moreover, the cokernel of the inclusion
E
0,q
H
q
(Tot(A
,
))
should match the cohomology group H
q
(Tot(B
,
)) since this piece is built from the
groups E
r,s
with r, s > 1 and r+s = p, therefore we have the canonical isomorphism

H
q
(C
(G, T)
G
) H
q
(Tot(B
,
))
=
H
q
(Tot(A
,
)), x y x +y
which in particular implies that
H
q
(C
(G, T)
G
)
=
H
q
(G, T).
Then we can conclude that the composition of the maps
H
3
(Tot(B
,
)) H
3
(Tot(A
,
))

MS
G
(G)
is an isomorphism, and therefore we obtain the canonical isomorphism
(6.2) H
3
(Tot(B
,
))
=
MS
G
(G), [ ] [ ].
Also we obtain the short exact sequence
0 H
2
(C
(G; T)
G
) H
2
(Tot(A
,
)) H
1
G
(G; T)
mult
0,
with H
1
G
(G; T)
mult

= H
2
(Tot(B
,
))

= Hom(G, Hom(G, T)).
Now we will study the multiplicative structures that dene the pseudomonoids
constructed via the formalism introduced in [9], and whose properties were outlined
in section 3.5. For this purpose we need to calculate the composition of the maps
H
3
(G, T)

1
H
3
(Tot(A
,
))

MS
G
(G)
where
1
is the induced map in cohomology which was dened at the chain level
in (2.2). This calculation will be carried out using the ring structure of the ring
H
(G, Z) together with the pullback map
: H
(G, Z) H
(GG, Z) induced
by the multiplication : G G G as we have in Theorem 2.3. Since we have
the isomorphism
H
3
(Tot(A
,
)) H
3
(G, T)
=
H
3
(GG, T), x y x +
2
y
46 C

where
2
: GG G, (a, g) g is the projection, then the short exact sequence
of (6.1) implies that the following is also a short exact sequence
0 H
3
(C
(K; T)
G
) H
3
(G, T) H
3
(GG, T) MS
G
(G) 0.
We conclude that the homomorphism

1
: H
3
(G, T) MS
G
(G)
is equivalent to the composition of homomorphisms
H
3
(G, T)

H
3
(GG, T)/(H
3
(C
(K; T)
G
) H
3
(G, T))
=
MS
G
(G).
Therefore we simply need to nd the restriction to H
3
(Tot(B
,
)) of the image
of
in H
3
(GG, T),
H
3
(G, T)

H
3
(GG, T)
res
H
3
(Tot(B
,
))

= MS
G
(G).
Let us see some examples.
6.2.1. Cyclic groups. Consider G = K = Z/n and note that as rings H
(G; Z) =
Z[x]/(nx) where [x[ = 2. By the Kunneth theorem we have that
H
2
(GG; Z)

= H
2
(G; Z) H
0
(G; Z) H
0
(G; Z) H
2
(G; Z)
and moreover we have that
x = x 1 + 1 x.
Since x
2
is the generator of H
4
(G; Z) we obtain
x
2
= x
2
1 + 2x x + 1 x
2
where
H
4
(Tot(B
,
(GG, Z))) = x x

= Z/n
and therefore we obtain that the map

1
: H
3
(G, T) MS
G
(G)
is equivalent to the map
H
4
(G; Z) H
4
(Tot(B
,
(GG, Z))), x
2
2x x.
Since in this case H
3
(G; T)

= MS
G
(G)

= Z/n we have that
1
: Z/n
2
Z/n.
Therefore, when n is odd, the map
1
is an isomorphism and therefore the
Grothendieck rings of representations Groth(Rep(D
w
(G))) of the Twisted Drinfeld
Doubles D
w
(G) for w H
3
(G, T) are all non-isomorphic.
Meanwhile when n is even, the cocycle w Z
3
(G, T) whose cohomology class is
n
2
Z/n

= H
3
(G, T) has for Grothendieck ring of representations Groth(Rep(D
w
(G)))
a ring isomorphic to Groth(Rep(D(G))), which is by denition the ring KU
G
(G) =
R(G) ZG. Moreover, the multiplicative structures dened by odd numbers in
Z/n

= MS
G
(G) dene Grothendieck rings of representations associated to the re-
spective pseudomonoids which cannot be recovered via the Grothendieck ring of
representations associated to the Twisted Drinfeld Double construction.
Note furthermore that in this case the automorphisms groups are isomorphic
H
2
(Tot(A
,
(GG, T)))

= H
1
G
(G, T)
mult

= Z/n.
6.2.2. Quaternionic group. Consider G = K = Q
8
the quaternionic group and
recall that Q
8
SU(2) and that it sits in the short exact sequence
0 Z/2 Q
8
Z/2 Z/2 0.
From the Serre vibration
SU(2)/Q
8
ESU(2)/Q
8
BSU(2)
one can deduce that the integral cohomology ring of Q
8
is
H
(Q
8
, Z) = Z[x, y, e]/(x
2
, y
2
, xy, 2x, 2y, xe, ye, 8e) with [x[ = [y[ = 2 and [e[ = 4.
Since the cohomology of Q
8
is all of even degree, by the Kunneth isomorphism
we have that
H
4
(Q
8
Q
8
, Z)

=
4
j=0
H
j
(Q
8
, Z) H
4j
(Q
8
, Z)
and since Q
8
Q
8

= Q
8
Q
8
we have that
H
(Q
8
Q
8
, Z)

= H
(Q
8
Q
8
, Z).
In this case we have that
MS
Q8
(Q
8
)

= H
4
(Tot(B
,
(Q
8
Q
8
, Z))) = x x, x y, y x, y y

= (Z/2)
4
and since
e = e 1 + 1 e
we can deduce that the map

1
: H
3
(Q
8
, T)
0
MS
Q8
(Q
8
)
is the trivial map.
A nice consequence of the triviality of the map
1
is that for all w Z
3
(Q
8
, T),
the Grothendieck ring of representations of the w-Twisted Drinfeld Double is iso-
morphic as rings to KU
Q8
(Q
8
), which is the Grothendieck ring of representations
of the untwisted Drinfeld double D(Q
8
).
Note also that the automorphisms groups are isomorphic
H
2
(Tot(A
,
(Q
8
Q
8
, T)))

= H
1
Q8
(Q
8
, T)
mult

= (Z/2)
4
.
6.2.3. G simple and non-abelian. In section 3.5.1 we showed that H
3
(Tot(B
,
(G
G, T))) = 0 and therefore we have that MS
G
(G) = 0. Hence all Grothendieck rings
of representations for all pseudomonoids are isomorphic to the ring KU
G
(G).
6.2.4. G binary icosahedral. The same result applies to the binary icosehedral group
since we showed in section 3.5.2 that H
3
(Tot(B
,
(G G, T))) = 0 and hence
MS
G
(G) = 0.
6.3. Z/n acted by Z/2. Consider the action of G = Z/2 on K = Z/n given by
multiplication of 1. The group Z/nZ/2 is isomorphic to the dihedral group D
n
of rigid symmetries of the regular polygon of n sides.
48 C

6.3.1. n odd. Let us suppose that n is odd and recall that in this case H
1
(D
n
, T) =
Z/2, H
2
(D
n
, T) = 0 and H
3
(D
n
, T) = Z/2 Z/n. Since H
3
(Z/2, T) = Z/2 we can
conclude that H
3
(Tot(A
,
)) = Z/n. Now, applying the spectral sequence dened
in Proposition 5.9 we have that E
1,q
1
= H
1
G
(K
q
, T) = Z/2, since the only xed point
of the Z/2 action is the p-tuple of zeroes, and therefore one obtains that E
1,q
2
= 0
for q > 0 since a simple calculation shows that the maps d
1
: E
1,2i1
1
E
1,2i
1
are
all isomorphisms. Moreover, since E
2,q
1
= H
2
G
(K
q
, T) = 0 because H
2
(Z/2, T) = 0
we have that E
2,1
3
= 0 and we can conclude that
coker
_
H
3
(Tot(A
,
))

MS
G
(K)
_
= 0;
hence we have that the map H
3
(Tot(A
,
))

MS
G
(K) is surjective.
It remains now to calculate ker
_
H
3
(Tot(A
,
))

MS
G
(K)
_
. Applying the
same argument as before we have that E
0,3
2
= H
3
(C
(K, T)
G
) and we already know
that E
1,2
2
= 0 = E
2,1
2
; therefore (H
3
(C
(K, T)
G
)) coincides with the image of the
canonical map E
0,3
2
H
3
(Tot(A
,
)) and this map must be surjective. Therefore
we have that
ker
_
H
3
(Tot(A
,
))

MS
G
(K)
_
= Z/n
and we can conclude that
MS
G
(K) = 0;
i.e. all multiplicative structures on the Z/2-equivariant K-theory of Z/n are trivial
and and all the Grothendieck rings Groth(Bun
Z/2
(/)) are isomorphic to the ring
KU
Z/2
(Z/n) for any / = (Z/n, m, 1, , , ). By Remark 4.3, KU
Z/2
(Z/n) is just
the ring of isomorphism classes of representations of the dihedral group D
n
.
In this case the automorphisms groups are both trivial
H
2
(Tot(A
,
(Z/n Z/2, T))) = 0 = H
1
Z/2
(Z/n, T)
mult
.
6.3.2. n even. Let us now suppose that n is even; in this case H
1
(D
n
, T) = Z/2
Z/2, H
2
(D
n
, T) = Z/2 and H
3
(D
n
, T) = Z/2Z/2Z/n. Since H
3
(Z/2, T) = Z/2
we have that H
3
(Tot(A
,
)) = Z/2 Z/n. Now, we also have that
E
1,q
1
= H
1
G
(K
q
, T)

= Maps((Z/2)
q
, Z/2)
since the xed points of the Z/2 action on K
q
consists of q-tuples of points with
either 0 or
n
2
for entries. It is a simple calculation to show that the dierential
d
1
: E
1,i
1
E
1,i+1
1
is precisely the dierential of the cohomology of the group Z/2
with coecients in Z/2 and therefore we get that
E
1,q
2
= H
q
(Z/2, Z/2)

= Z/2.
The groups E
2,q
1
are trivial because H
2
(Z/2, T) = 0.
Let us now calculate the group E
0,1
2
= H
1
(C
(K, T)
G
). This group consists
of the maps f : Z/n T such that f is invariant under the G-action, namely
f(x) = f(x), and that
1
f = 0, which means that f is a homomorphism. The
only G-invariant homomorphisms are the ones that take values in the subgroup
Z/2 T and therefore we have that
E
0,1
2
= H
1
(C
(K, T)
G
) = Z/2.
The information we have obtained so far on the cohomology groups of the second
page of the spectral sequence is the following
3 ?
2 ? Z/2
1 Z/2 Z/2 0
0 T Z/2 0 Z/2
0 1 2 3
and since we know that H
1
(D
n
, T) = Z/2 Z/2 and H
2
(D
n
, T) = Z/2 we can
deduce that E
0,2
= E
0,2
2
= 0 and therefore E
0,3
2
= E
4
0, 3 = Z/n. Hence we have
that
H
3
(C
(G, T)
G
) = Z/n, MS
G
(K) = Z/2 and
H
3
(Tot(A
,
))

MS
G
(K), Z/n Z/2 Z/2
is the canonical projection map.
In this case the automorphisms groups are isomorphic
H
2
(Tot(A
,
(Z/n Z/2, T)))

= H
1
Z/2
(Z/n, T)
mult

= Z/2.
7. Appendix: Relation with (coquasi) bialgebras
The classical Tannakian duality permits to reconstruct a compact group from its
nite dimensional representations and a generalization of this statement was done
for quantum groups (Hopf algebras); see [15] for a survey on classical and quantum
Tannakian duality. The Tannakian principle works in the following way. If ( is
a C-linear abelian tensor category with a faithful, exact and strong tensor functor
( Vec
C
(called ber functors), then there exists a unique bialgebra H that solves
certain universal problem and such that the category of corepresentations of H is
canonically equivalent to ( as a tensor category, see [15, Theorem 3 and Proposition
4].
The tensor category Bun
G
(/) is endowed with a natural forgetful functor U :
Bun
G
(/) Vec
C
, that is exact and faithful. However, unless Z
3
(K, T) has
trivial cohomology, the functor U is not monoidal. The functor U is only quasi-
monoidal (a quasi-ber functor) and a reconstruction of a Hopf algebra in general is
not possible. A generalized version of Tannakian duality using quasi-ber functors
was developed by Shahn Majid in [18], using a generalization of the notion of
Hopf algebra, called coquasi-Hopf algebra, previously dened (in a dual version) by
Drinfeld [10].
In what follows we will show how Bun
G
(/) can be understood as the tensor
category of corepresentations associated to an explicit coquasi-bialgebra, and for
this purpose we will show that the input necessary for dening such coquasi-Hopf
algebra is equivalent to the information encoded in a pseudomonoid with strict unit
in the 2-category of G-sets with twists.
7.1. Coquasi-bialgebras.
50 C

7.1.1. Coalgebras and comodules. Let k be a eld. A coalgebra over k is a vector
space over k together with two linear maps : C CC, : C k (called
comultiplication and counit respectively) such that (C ) = (C) and
(C) = ( C) = C. We shall use the Sweedlers notation omitting the sum
symbol, that is (c) = c
1
c
2
if c C.
If C is a coalgebra, a right C-comodule is a vector k-space M with a linear map
: M MC such that (C) = (M) and (M) = M. Again for the
comodule structure we shall use sweedlers notation omitting the sum symbol, i.e.,
(m) = m
0
m
1
, m
0
M, m
1
C. If M, N are C-comodules, a comodule map is
a linear map f : M N such that
N
f = (fC)
M
. We shall denote by /
C
the
category of right C-comodules.
If C, C
are coalgebras C C
is a coalgebra with comultiplication (cc
) =
(c
1
c
1
)(c
2
c
2
) and counit (cc
) = (c)(c
).
For a coalgebra C the space C
is an associative algebra with the convolution

product f g(c) = f(c
1
)f(c
2
) and unit .
7.1.2. Coquasi-bialgebras. Coquasi-bialgebras and coquasi-Hopf algebras are gener-
alizations of the notion of bialgebras and Hopf algebras and they were dened (in a
dual version) by Drinfeld in [10]. We shall recall the denition: a coquasi-bialgebra
is a ve-tuple (H, , m, 1
H
, ) where H is a coassociative coalgebra with counit,
m : HH H, hg hg is a coalgebra map, 1
H
is a grouplike element (i.e.
(1
H
) = 1
H
1
H
) which is a unit for m, and (HHH)
is a convolution
invertible map (called the coassociator), satisfying the identities
(g1
H
h) = (g)(h), (7.1)
m(mid
H
) = m(id
H
m), (7.2)
(d
1
f
1
g
1
h
1
)(d
2
f
2
g
2
h
2
) = (d
1
f
1
g
1
)(d
2
f
2
g
2
h
1
)(f
1
g
3
h
2
) (7.3)
for all f, g, h H.
Remark 7.1. The notion of quasi-bialgebra is closely related with the notion of
pseudomonoid, in fact a quasi-bialgebra is a pseudomonoid in CoAlg the 2-category
of coalgebras, see [23, Chapter 15].
The category of right H-comodules /
H
is monoidal, where the tensor product
is over the base eld and the comodule structure of the tensor product is induced
by the multiplication. The associator is given by
U,V,W
: (U V ) W U (V W)
U,V,W
((u v) w) =
1
(u
1
, v
1
, w
1
)u
0
(v
0
w
0
)
for u U, v V , w W and U, V, W /
H
.
Example 7.2. Let K be a discrete group and Z
3
(K, T) a normalized 3-cocycle.
The group algebra CK is a quasi-bialgebra with comultiplication (u
) = u
,
counit (u
) = 1 and coassociator
(u
, u
, u
) = (, , ),
for all , , K.
The coquasi-bialgebra (CK, , ,
) is a coquasi-Hopf algebra, with S(u
) =
u
1, 1 and (u
) = (,
1
, ) for all G.
7.2. Coquasi-bialgebras associated to pseudomonoids. Let G be a nite
group and / = (K, m, 1, , , ) a pseudomonoid with strict unit in the 2-category
of G-sets with twists; let us denote C
G
:= Maps(G, C), let
C
G
be the function
that assigns 1 to and 0 otherwise, and let
,
be the Diracs delta associated to
the pair , G, namely
,
is 1 whenever = and 0 otherwise.
Theorem 7.3. The vector space C
G
#K := C
G
CK with basis
#x[ G, x
K is a coquasi-bialgebra with product
(
#x)(
#y) =
,
[[[x[y]
#xy,
coproduct,
(
#x) =
a,bG:ab=
[a[b[[x]
a
#bx
b
#x,
associator
(
#x,
#y,
#z) =
,e
,e
,e
[x[y[z],
counit (
#x) =
,e
, and unit 1#e for all , , G, x, y, z K.
Proof. Let us check that is coassociative,
(id)(
#x) = (id)(
ab=
[a[b[[x]
a
#bx
b
#x)
=
abc=
[ab[c[[x][a[b[[cx]
a
#bcx
b
#cx
c
#x
and
(id )(
#x) = (id )(
ab=
[a[b[[x]
a
#bx
b
#x)
=
abc=
[a[bc[[x][b[c[[x]
a
#bcx
b
#cx
c
#x.
Since
G
() = 1, is coassociative. It is immediate to see that 1#e is a grouplike
element and the unit for the product.
Now we shall see that the product is a coalgebra map:
((
#x)(
#y)) = ([[[x[y]
#xy)
=
ab=
[ab[[x[y][a[b[[xy]
a
#b(xy)
b
#xy
and
(mm)
C
G
#KC
G
#K
(
#x
#y) =
mm(
a,b,r,sG:ab=sr=
[a[b[[x][r[s[[y]
a
#bx
r
#s
b
#x
s
#y) =
ab=
[a[b[[x][a[b[[y][a[[bx[by][b[[x[y]
a
#(bx)(by)
b
#xy.
Since
G
= m
( )
1
= (
K
)
1
we have
[ab[[x[y][a[b[[xy] = [a[b[[x][a[b[[y][a[[bx[by][b[[x[y]
for all a, b G, x, y K, so the multiplication is a coalgebra map.
52 C

Let us check axiom (7.2)
m(mid) (
#x,
#y,
#z) =
,
[[[x[y] [[[xy[z] [x[y[z]
#xyz
and
m(idm)(
#x,
#y,
#z) =
,
[x[y[z] [[[y[z] [[[x[yz]
#xyz.
Since
G
()
K
()
1
= 1 we have
[[[x[y] [[[xy[z] [x[y[z] = [x[y[z] [[[y[z] [[[x[yz]
for all G, x, y, z G, then m(mid) = m(idm).
Finally in order to prove (7.1) and (7.3), note that the projection : C
G
#K
CK,
#x
,e
u
x
is a coalgebra map and =
(see Example 7.2), then since
(CKCKCK)
is a coassociator for CK, is a coassociator for C

G
#K.
Remark 7.4. If is trivial, C
G
#K is a Hopf algebra.
Theorem 7.5. Let G be a nite group and / = (K, m, 1, , , ) a pseudomonoid
with strict unit in the 2-category of G-sets with twists. The tensor category of
right C
G
#CK-comodules is tensor isomorphic to the monoidal category Bun
G
(/)
of equivariant vector bundles on /.
Proof. Let V =
xK
V
x
be an object in Bun
G
(/), we dene a right C
G
#CK-
comodule structure over V by
(v
x
) =
G
v
x

#x,
for all v
x
V
x
, x K.
The coassociativity of follows from the equations
(id)(v
x
) =
G
( v
x
)
#x
=
,G
( v
x
)
#x
#x,
(id)(v
x
) =
G
v
x
(
#x)
=
a,bG:ab=
[a[b[[x] v
x

a
#bx
b
#x
=
,G
[[[[x]() v
x

#x
#x.
and the fact that ( v
k
) = [[[[k]() v
k
for all , G, k K, v
k
V
k
,
(id) = (id).
The counity axiom follows from
(id)(v
x
) =
G
v (
#x)
= e v
x
= v
x
Now, for f : V W a morphism in Bun
G
(/), we shall see that f is also a
morphism of comodules, and this follows from
(fid)
V
(v
x
) =
G
f( v)
#x
=
G
f(v
x
)
#x
=
W
(f(v
x
)),
where last equality follows because f(v
x
) W
x
.
Conversely, if (V, ) is a right C
G
#CK-comodule, using the coalgebra epimor-
phism : C
G
#K CK,
#x
,e
u
x
, V is a CK-comodule by (id). Recall
that CK-comodules are just K-graded vector spaces, then
V =
xK
V
x
, where V
x
= v V : (id) = v x.
Since
#x[ G, x K is a basis of C
G
#CK and by denition (V
x
)
V CG#x for all x G, the comodule map denes and is dened by unique linear
maps : CGV
x
V (x G) such that
(v
x
) =
G
v
x
#x,
for all v
x
V
x
.
First we shall see that V
x
V
x
for all G, x X:
G
( v
x
)
#x = (id)(v
x
)
= (id)(v
x
)
=
,G
[[[[x]() v
x

#x
#x,
so
( v
x
) =
G
[[[[x]() v
x

#x,
then
(id)( v
x
) =
G
[[[[x]() v
x
(
#x)
= [e[[[x] v
x
x
= v
x
x,
and therefore v
x
V
x
. Moreover we have
,G
( v
x
)
#x
#x = (id)(v
x
)
= (id)(v
x
)
=
,G
[[[[x]() v
x

#x
#x,
so
( v
x
) = [[[[x]() v
x
54 C

for all v
x
V
x
, x K, , G.
The counit axiom
v
x
= (id)(v
x
)
=
G
v (
#x)
= e v
x
,
for all v
x
V
x
.
Now, let (V,
V
), (W,
W
) be right C
G
#CK-comodules. Let us see that every
morphism of comodules f : V W denes a morphism between the equivariant
twisted vector bundles associated.
If v
x
V
x
,
(id )
W
(f(v
x
)) = (id )(fid)
V
(v
x
)
= (fid)(id )
V
(v
x
)
= (fid)(v
x
x)
= f(v
x
)x,
then f(v
x
) W
x
. Since f(v
x
) W
x
,
G
f(v
x
)
#x =
W
(f(v
x
))
= (fid)
V
(x)
=
G
f( v
x
)
#x,
then f(v
x
) = f( v
x
) for all x X, v
x
V
x
, G.
Finally we shall prove that the functors previously dened are monoidal. Let
V, W /
C
G
#CK
, then the grading of the twisted vector bundles associated to
V W is given by (V W)
k
=
x,yK:xy=k
V
x
W
y
, in fact, if v
x
V
x
, w
y
W
y
,
then
(id)
V W
(v
x
w
y
) = id
_

,G
[[[x[y] v
x
w
y

,
#xy
_
=
G
[[[x[y] v
x
w
y
(
#xy)
= v
x
w
y
xy,
then W
x
V
y
(V W)
xy
. The comodules V and W are nite dimensional, so
(V W)
k
=
x,yK:xy=k
V
x
W
y
and since
G
(v
x
w
y
)
#xy =
V W
(v
x
w
y
)
=
G
[[[x[y] v
x
w
y

#xy
the twisted action on V W is given by (v
x
w
y
) = [[[x[y] v
x
w
y
.
We shall check that the associator induced by the coquasi-biagebra structure
agrees with the associator of Bun
G
(/).
Let U, V, W be right comodules. If u
x
U
x
, v
y
V
y
and w
z
W
z
, then
U,V,W
((u
x
v
y
) w
z
) =
1
((u
x
)
1
, (v
y
)
1
, (w
z
)
1
)(u
x
)
0
((v
y
)
0
(w
z
)
0
)
=
,,G
1
(
#x,
#y,
#z) u
x
v
y
w
z
= [x[y[z]
1
u
x
v
y
w
z
.
Remark 7.6. Whenever G acts by conjugation on K = G and we consider a 3-

cocycle w Z
3
(G, T), the coquasi-bialgebra C
G
#CG dened by the pseudomonoid
(G, m, 1,
w
,
w
,
w
) as in section 3.5, is the dual of the Twisted Drinfeld Double
of the nite group G dened in [9, Section 3.2].
We nish with a corollary of the results of this appendix and the ones of section
6.2. A quasi-isomorphism between coquasi-Hopf algebras (H, ) and (H
, ) is a
pair (F, ) consisting of a coalgebra isomorphism F : H H
and a convolution
invertible map (H H)
such that (1 h) = (h 1) = (h) and

(g
1
h
1
)F(g
2
h
2
) = F(g
1
)F(h
2
)(g
2
h
2
),
for all g, h H. A quasi-isomorphism is called an isomorphism of quasi-Hopf
algebras if additionally
(f
1
g
2
)(f
2
g
2
h
1
)(f
3
g
3
h
3
) =
(F(f
1
)F(g
1
)F(h
1
))(g
2
h
2
)(f
2
g
3
h
3
).
Coquasi-isomorphisms and isomorphisms of coquasi-Hopf algebras have categor-
ical interpretations. Isomorphisms correspond to tensor equivalences and quasi-
isomorphism with quasi-monoidal equivalences (or non-coherent monoidal equiva-
lence). In the case of cosemisimple coquasi-Hopf algebras, it was proved in [8, Prop.
4], that two coquasi-Hopf algebras are coquasi-isomorphic if and only if they have
equivalent Grothendieck rings.
In general, the Twisted Drinfeld Double D
w
(G) of a nite group G, is not
isomorphic to a Hopf algebra, or equivalently, the tensor category of representa-
tions of D
w
(G) is not equivalent to the category of representations of any Hopf
algebra. However, by the isomorphism outlined in (6.2), if G is nite and acts
over itself by conjugation, the Grothendieck ring of Bun
G
(/) for any 3-cocycle
in Z
3
(Tot
(A
,
(G G, T))), is always equivalent to the Grothendieck ring of
Bun
G
(/
), where the 3-cocycle associated lives in Z

3
(Tot
(B
,
(G G, T))). By
Theorem 7.3 and Theorem 7.5, Bun
G
(/
) is the category of representation of a

Hopf algebra, so in particular we can conclude
Corollary 7.7. The Twisted Drinfeld Double of a nite group is always quasi-
isomorphic to a Hopf algebra.
References
[1] A. Adem and R. J. Milgram. Cohomology of nite groups, volume 309 of
Grundlehren der Mathematischen Wissenschaften [Fundamental Principles of
Mathematical Sciences]. Springer-Verlag, Berlin, 1994.
[2] A. Adem, Y. Ruan, and B. Zhang. A stringy product on twisted orbifold
K-theory. Morsmos, 11(2), 2007.
56 C

[3] B. Bakalov and A. Kirillov, Jr. Lectures on tensor categories and modular
functors, volume 21 of University Lecture Series. American Mathematical
Society, Providence, RI, 2001.
[4] N. Barcenas, J. Espinoza, M. Joachim, and B. Uribe. Classica-
tion of twists in equivariant K-theory for proper and discrete actions.
http://arxiv.org/abs/1202.1880, 2012.
[5] E. Becerra and B. Uribe. Stringy product on twisted orbifold K-theory for
abelian quotients. Trans. Amer. Math. Soc., 361(11):57815803, 2009.
[6] F. Borceux. Handbook of categorical algebra. 2, volume 51 of Encyclopedia of
Mathematics and its Applications. Cambridge University Press, Cambridge,
1994. Categories and structures.
[7] K. S. Brown. Cohomology of groups, volume 87 of Graduate Texts in Mathe-
matics. Springer-Verlag, New York, 1982.
[8] A. A. Davydov. Finite groups with the same character tables, Drinfel
d al-
gebras and Galois algebras. In Algebra (Moscow, 1998), pages 99111. de
Gruyter, Berlin, 2000.
[9] R. Dijkgraaf, V. Pasquier, and P. Roche. Quasi Hopf algebras, group coho-
mology and orbifold models. Nuclear Phys. B Proc. Suppl., 18B:6072 (1991),
1990. Recent advances in eld theory (Annecy-le-Vieux, 1990).
[10] V. G. Drinfel
d. Quasi-Hopf algebras. Algebra i Analiz, 1(6):114148, 1989.

[11] S. Eilenberg and S. Mac Lane. On the groups of H(, n). I. Ann. of Math.
(2), 58:55106, 1953.
[12] P. Etingof, D. Nikshych, and V. Ostrik. On fusion categories. Ann. of Math.
(2), 162(2):581642, 2005.
[13] P. J. Hilton and U. Stammbach. A course in homological algebra, volume 4 of
Graduate Texts in Mathematics. Springer-Verlag, New York, second edition,
1997.
[14] J. Hu and B.-L. Wang. Delocalized chern character for stringy orbifold K-
theory. http://arxiv.org/abs/1110.0953, 2011.
[15] A. Joyal and R. Street. An introduction to Tannaka duality and quantum
groups. In Category theory (Como, 1990), volume 1488 of Lecture Notes in
Math., pages 413492. Springer, Berlin, 1991.
[16] G. I. Kac. Group extensions which are ring groups. Mat. Sb. (N.S.), 76
(118):473496, 1968.
[17] G. Karpilovsky. The Schur multiplier, volume 2 of London Mathematical So-
ciety Monographs. New Series. The Clarendon Press Oxford University Press,
New York, 1987.
[18] S. Majid. Tannaka-Kren theorem for quasi-Hopf algebras and other results.
In Deformation theory and quantum groups with applications to mathematical
physics (Amherst, MA, 1990), volume 134 of Contemp. Math., pages 219232.
Amer. Math. Soc., Providence, RI, 1992.
[19] A. Masuoka. Hopf algebra extensions and cohomology. In New directions
in Hopf algebras, volume 43 of Math. Sci. Res. Inst. Publ., pages 167209.
Cambridge Univ. Press, Cambridge, 2002.
[20] A. Masuoka. Cohomology and coquasi-bialgebra extensions associated to a
matched pair of bialgebras. Adv. Math., 173(2):262315, 2003.
[21] C. C. Moore. Group extensions and cohomology for locally compact groups.
III. Trans. Amer. Math. Soc., 221(1):133, 1976.
[22] V. Ostrik. Module categories, weak Hopf algebras and modular invariants.
Transform. Groups, 8(2):177206, 2003.
[23] R. Street. Quantum groups, volume 19 of Australian Mathematical Society
Lecture Series. Cambridge University Press, Cambridge, 2007. A path to
current algebra.
[24] V. G. Turaev. Quantum invariants of knots and 3-manifolds, volume 18 of
de Gruyter Studies in Mathematics. Walter de Gruyter & Co., Berlin, revised
edition, 2010.
[25] S. Willerton. The twisted Drinfeld double of a nite group via gerbes and nite
groupoids. Algebr. Geom. Topol., 8(3):14191457, 2008.
[26] S. J. Witherspoon. Products in Hochschild cohomology and Grothendieck rings
of group crossed products. Adv. Math., 185(1):136158, 2004.
Departamento de Matem aticas, Universidad de los Andes, Carrera 1 N. 18A - 10,
Bogot a, Colombia.
E-mail address: cn.galindo1116@uniandes.edu.co, cesarneyit@gmail.com
Departamento de Matem aticas, Universidad del Norte, Km.5 Va Puerto Colombia,
Barranquilla, Colombia.
E-mail address: isgutier@uninorte.edu.co
Departamento de Matem aticas, Universidad de los Andes, Carrera 1 N. 18A - 10,
Bogot a, Colombia.
Current address: Rheinische Wilhelms-Universit at Bonn, Mathematisches Institut, En-
denicher Allee 62, 53115 Bonn, Germany
E-mail address: buribe@gmail.com

Multiplicative Structures Subm 130626

Enviado por

Dados do documento

Direitos autorais

Formatos disponíveis

Compartilhar este documento

Compartilhar ou incorporar documento

Opções de compartilhamento

Você considera este documento útil?

Este conteúdo é inapropriado?

Direitos autorais:

Formatos disponíveis

Multiplicative Structures Subm 130626

Enviado por

Direitos autorais:

Formatos disponíveis

MULTIPLICATIVE STRUCTURES ON THE TWISTED

K-THEORY OVER FINITE GROUPS

ESAR GALINDO, ISMAEL GUTI

ERREZ, AND BERNARDO URIBE

ESAR GALINDO, ISMAEL GUTI

ERREZ, AND BERNARDO URIBE

w, we give a simple procedure to determine the fusion

ESAR GALINDO, ISMAEL GUTI

ERREZ, AND BERNARDO URIBE

Corollary 1.2. The semi-direct product K Inn(K) is isomorphic as split exten-

ESAR GALINDO, ISMAEL GUTI

ERREZ, AND BERNARDO URIBE

(EH, Z) becomes a ZH-free resolution

(EH, Z), W)).

ESAR GALINDO, ISMAEL GUTI

ERREZ, AND BERNARDO URIBE

at the chain level using the explicit models for H

(K, Z) dened previously. Its denition needs some preparation.

ESAR GALINDO, ISMAEL GUTI

ERREZ, AND BERNARDO URIBE

ESAR GALINDO, ISMAEL GUTI

ERREZ, AND BERNARDO URIBE

must be equal to the pushforward

since preserves the

K G, (k, g) (k, (g)) induces an isomorphism of semi-direct products, then we

(K, Z)) Tot

(K, Z)) Tot

ESAR GALINDO, ISMAEL GUTI

ERREZ, AND BERNARDO URIBE

ESAR GALINDO, ISMAEL GUTI

ERREZ, AND BERNARDO URIBE

) are 2-cells, their composition

A straightforward calculation implies that

where we use the equalities

ESAR GALINDO, ISMAEL GUTI

ERREZ, AND BERNARDO URIBE

ESAR GALINDO, ISMAEL GUTI

ERREZ, AND BERNARDO URIBE

MULTIPLICATIVE STRUCTURES ON TWISTED EQUIVARIANT K-THEORY 21

Denition 3.7. Given two pseudomonoids with strict unit / = (K, m, 1, , , )

) in the 2-category of G-sets with twists, a morphism

is a morphism of pseudomonoids (as in Denition 3.6) such that the

consists of the triple T = (F, , ) with

a G-equivariant morphism of monoids, and cochains C

), i.e. a G-equivariant map F : K K

and a normalized cochain C

ESAR GALINDO, ISMAEL GUTI

ERREZ, AND BERNARDO URIBE

preserves the multiplication; and since

implies that the map F : K K

preserves the unit and that (g, 1) = 1

, and that the following

Denition 3.9. Given morphisms T = (F, F

) be the 2-cell dened by the 2-morphism, then we

ESAR GALINDO, ISMAEL GUTI

ERREZ, AND BERNARDO URIBE

ESAR GALINDO, ISMAEL GUTI

ERREZ, AND BERNARDO URIBE

[w] where : G G G, (a, g) ag is the multiplication map.

(G, T) is trivial. Hence we have that

ESAR GALINDO, ISMAEL GUTI

ERREZ, AND BERNARDO URIBE

, which is the multi-

where the action is given by