Monoidal Categories
(Spine title: Hopf Cyclic Cohomology In Braided Monoidal Categories)
(Thesis format: Monograph)
by
Arash Pourkia
Graduate Program
in
Mathematics
A thesis submitted in partial fulﬁllment
of the requirements for the degree of
Doctor of Philosophy
School of Graduate and Postdoctoral Studies
The University of Western Ontario
London, Ontario, Canada
c Arash Pourkia 2009
i
Certiﬁcate of Examination
THE UNIVERSITY OF WESTERN ONTARIO
SCHOOL OF GRADUATE AND POSTDOCTORAL STUDIES
Chief Adviser: Examining Board:
Professor Masoud Khalkhali Professor Tatyana Foth
Advisory Committee: Professor Jan Minac
Professor John Bell
Professor Piotr M. Hajac
The thesis by
Arash Pourkia
entitled:
Hopf Cyclic Cohomology In Braided Monoidal Categories
is accepted in partial fulﬁllment of the
requirements for the degree of
Doctor of Philosophy
Date:
Chair of Examining Board
Firstname Lastname
ii
Abstract
In the ﬁrst three chapters of this thesis we recall basics of Hopf algebras, cyclic
cohomology and braided monoidal categories.
Chapters four and ﬁve form the heart of this thesis. In Chapter four, we ex
tend the whole theory of Hopf cyclic cohomology with coeﬃcients [18, 19, 25, 26], to
symmetric braided monoidal abelian categories. We also obtain a braided version of
ConnesMoscovici’s Hopf cyclic cohomology [9, 10, 11] in any (not necessarily sym
metric) braided monoidal abelian category. We use our theory to deﬁne a Hopf cyclic
cohomology for super Hopf algebras and for quasitriangular quasiHopf algebras.
In Chapter ﬁve, we deﬁne a super version of ConnesMoscovici Hopf algebra H
1
[9]. For that we deﬁne a superbicrossproduct Hopf algebra k[G
s
2
]U(g
s
1
), analogous
to the non super case [9, 17]. We call this superbicrossproduct Hopf algebra the
super version of H
1
and denote it by H
s
1
.
Keywords: Noncommutative geometry, Hopf algebra, braided monoidal categories,
Hopf cyclic cohomology, super mathematics.
iii
Acknowledgments
First and foremost, my great gratitude goes to my supervisor, Professor Masoud
Khalkhali, for his invaluable guidance and his constant supports and encouragement.
In the absence of his direction and assistance, I could not imagine myself even starting
this work.
I want to thank my external examiner Professor Piotr M. Hajac and examiners
Professors Jan Minac, Tatyana Foth and John Bell for carefully reading my thesis
and for their valuable suggestions.
I am also grateful for all supports and helps I have constantly received from the
chair of graduate studies of the Department of Mathematics, Professor Andre Boivin.
In addition, I want to thank Professor Stuart A. Rankin for helping me creating
my thesis ﬁle and for his valuable technical supports.
I would also like to thank the depaerment’s administrative staﬀ, Janet Williams,
Debbie Mayea, and Terry Slivinski.
Last, but not least, there are those who have supported me through the hardest
times, with their love and spirit, and I cannot express what I owe them: my parents,
my sisters, my brother and his family.
iv
To my family
Specially to my beautiful sisters Sanaz,
Haleh, Sima and to my lovely niece Zahra
v
Table of Contents
Certiﬁcate of Examination . . . . . . . . . . . . . . . . . . . . . . . . . ii
Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iii
Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iv
Dedication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v
1 Hopf Algebra Basics . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1 Hopf algebras . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Hopf module algebras and coalgebras . . . . . . . . . . . . . . . . . . 5
1.3 Quasitriangular quasiHopf algebras . . . . . . . . . . . . . . . . . . . 6
2 Cyclic Cohomology Basics . . . . . . . . . . . . . . . . . . . . . . . . 13
2.1 Cyclic cohomology of algebras . . . . . . . . . . . . . . . . . . . . . . 13
2.2 Cyclic and cocyclic modules . . . . . . . . . . . . . . . . . . . . . . . 17
2.3 Paracyclic and cocyclic modules . . . . . . . . . . . . . . . . . . . . 22
2.4 Cyclic (co)homology of (co)cyclic objects . . . . . . . . . . . . . . . 25
2.5 Periodic cyclic cohomology of cocyclic objects . . . . . . . . . . . . . 29
2.6 Hopf cyclic cohomology . . . . . . . . . . . . . . . . . . . . . . . . . . 30
2.7 Dual Hopf cyclic cohomology . . . . . . . . . . . . . . . . . . . . . . 37
2.8 YetterDrinfeld modules . . . . . . . . . . . . . . . . . . . . . . . . . 40
2.9 Stable anti YetterDrinfeld modules . . . . . . . . . . . . . . . . . . . 42
2.10 Hopf cyclic cohomology with coeﬃcients . . . . . . . . . . . . . . . . 45
3 Braided Monoidal Categories . . . . . . . . . . . . . . . . . . . . . 49
3.1 Braided monoidal categories . . . . . . . . . . . . . . . . . . . . . . . 49
3.2 Examples of braided monoidal categories . . . . . . . . . . . . . . . . 55
3.3 Braided algebras, coalgebras and Hopf algebras . . . . . . . . . . . . 59
4 Braided Hopf Cyclic Cohomology . . . . . . . . . . . . . . . . . . . 65
4.1 The cocyclic module of a braided triple (H, C, M) . . . . . . . . . . . 66
4.2 The braided version of ConnesMoscovici’s Hopf cyclic cohomology . . 80
4.3 Hopf cyclic cohomology for super Hopf algebras . . . . . . . . . . . . 85
4.4 Hopf cyclic cohomology of the enveloping algebra of a super Lie algebra 86
4.5 Hopf cyclic cohomology in nonsymmetric monoidal categories . . . . 94
4.6 A Hopf cyclic theory for quasitriangular quasiHopf algebras . . . . . 106
vi
5 A super version of the ConnesMoscovici Hopf algebra . . . . . . . 110
5.1 The ConnesMoscovici Hopf algebra H
1
. . . . . . . . . . . . . . . . . 111
5.2 The super group G
s
= Diff
+
(R
1,1
) and its factorisation . . . . . . . 116
5.3 Two super Hopf algebras U(g
s
1
) and F(G
s
2
) . . . . . . . . . . . . . . . 120
5.3.1 The super Hopf algebra U(g
s
1
) . . . . . . . . . . . . . . . . . . 120
5.3.2 The super Hopf algebra F(G
s
2
) . . . . . . . . . . . . . . . . . 122
5.4 Actions and coactions . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
5.4.1 Actions of X . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
5.4.2 Actions of Y . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
5.4.3 Actions of Z . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
5.4.4 Actions of U . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
5.4.5 Actions of V . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
5.4.6 Actions of W . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
5.4.7 Coactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
5.5 Compatibilities and the super Hopf algebra H
s
1
. . . . . . . . . . . . 152
Vita . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
vii
Chapter 1
Hopf Algebra Basics
In this chapter we review the basics about Hopf algebras and quasitriangular quasi
Hopf algebra. We also recall the notions of Hopf module algebra and coalgebra.
1.1 Hopf algebras
We assume basic notions of algebra and coalgebra in the category of vector spaces
over a ﬁeld k, as explained in [22, 31, 37]. In this section we recall the notions of
bialgebra and Hopf algebra over a ﬁeld k, and give some examples.
A bialgebra is simultaneously an algebra and a coalgebra satisfying some com
patibility conditions. More precisely:
Deﬁnition 1.1.1. A bialgebra is a quadruple (H, m, η, ∆, ε) where H is a vector space
over a ﬁeld k, m : H⊗H →H, η : k →H, ∆ : H →H⊗H and ε : H →k are linear
maps called multiplication, unit, comultiplication and counit, respectively, such that
(H, m, η) is a unital associative algebra, (H, ∆, ε) is a counital coassociative coalgebra,
and the two structures are compatible in the sense that ∆ and ε are morphisms of
algebras or, equivalently, m and η are morphisms of coalgebras.
Compatibility conditions can be expressed in terms of commutative diagrams
as below:
1
2
The map ∆ is a morphism of algebras:
H ⊗H
∆m
∆⊗∆
H ⊗H
H ⊗H ⊗H ⊗H
id⊗τ⊗id
H ⊗H ⊗H ⊗H
m⊗m
¸¸
where τ : H ⊗H →H ⊗H is the ﬂip map deﬁned by τ(g ⊗h) = h ⊗g, for all g and
h in H, and
H
∆
H ⊗H
k ⊗k = k
η
¸¸
η⊗η
k
k
k
k
k
k
k
k
k
k
k
k
k
k
k
The map ε is a morphism of algebras:
H ⊗H
m
ε⊗ε
H
ε
.l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
k ⊗k = k
k
id
η
k
H
ε
o
o
o
o
o
o
o
o
o
o
o
o
o
o
Throughout this thesis, (except for slight changes in the last chapter), we use
the Sweedler’s notation, with summation understood, i.e., we write ∆h = h
(1)
⊗h
(2)
to denote the comultiplication of bialgebras. Similarly for higher comultiplications
we write:
∆
n
h = h
(1)
⊗h
(2)
⊗h
(3)
⊗. . . ⊗h
(n+1)
.
Deﬁnition 1.1.2. A Hopf algebra (H, m, η, ∆, ε, S) consists of a bialgebra
3
(H, m, η, ∆, ε) and a linear map S : H →H, called antipode, satisfying:
S(h
(1)
)h
(2)
= h
(1)
S(h
(2)
) = ε(h)1, ∀h ∈ H,
or equivalently,
m(S ⊗id)∆ = m(id ⊗S)∆ = ηε . (1.1.1)
The antipode axioms (1.1.1) can be expressed by commutative diagrams as
below:
H ⊗H
S⊗id
H ⊗H
m
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
H
ε
∆
¸¸
∆
k
η
H
H ⊗H
id⊗S
H ⊗H
m
m
m
m
m
m
m
m
m
m
m
m
m
m
m
m
In this thesis H always denotes a Hopf algebra. A Hopf algebra (H, m, η, ∆, ε, S)
is called commutative if it is commutative as an algebra, i.e.,
ab = ba ∀a, b ∈ H,
or equivalently
mτ = m.
H is called cocommutative if it is a cocommutative coalgebra, i.e.,
h
(1)
⊗h
(2)
= h
(2)
⊗h
(1)
∀h ∈ H,
or equivalently
τ∆ = ∆.
Proposition 1.1.1. [37] The antipode S of a Hopfalgebra H is unique. It is an
4
antialgebra map and an anticoalgebra map. More precisely, for all g and h in H:
S(gh) = S(h)S(g), S(1) = 1,
S(h)
(1)
⊗S(h)
(2)
= S(h
(2)
) ⊗S(h
(1)
),
εS(h) = ε(h).
Proposition 1.1.2. [37] If H is commutative or cocommutative then S
2
= id.
Example 1.1.1. Let G be a discrete group and H = kG the group algebra of G over
the ﬁeld k. Let
∆(g) = g ⊗g, S(g) = g
−1
, and ε(g) = 1,
for all g ∈ G and extend them by linearity to kG. Then (H, ∆, ε, S) is a cocommutative
Hopf algebra. It is commutative if and only if G is commutative.
Example 1.1.2. Let g be a Lie algebra over the ﬁeld k and H = U(g) be the universal
enveloping algebra of g. Using the universal property of U(g) one checks that there
are uniquely deﬁned algebra homomorphisms ∆ : U(g) →U(g) ⊗U(g), ε : U(g) →k
and an antialgebra map S : U(g) →U(g), determined by
∆(X) = X ⊗1 + 1 ⊗X, ε(X) = 0, and S(X) = −X,
for all X ∈ g. Then (U(g), ∆, ε, S) is a cocommutative Hopf algebra. It is commu
tative if and only if g is an abelian Lie algebra, in which case U(g) = S(g) is the
symmetric algebra of g.
5
1.2 Hopf module algebras and coalgebras
The idea of symmetry in noncommutative geometry is encoded by action or coaction
of a Hopf algebra on an algebra or a coalgebra. There are four possibilities of which
we recall only two in this section.
Deﬁnition 1.2.1. A left Hmodule algebra is an algebra (A, m
A
, η
A
) equipped with
a left Hmodule structure via the left action
φ
A
: H ⊗A →A, φ
A
(h ⊗a) = ha
such that , m
A
and η
A
are Hmodule maps, i.e.,
h(ab) = h
(1)
ah
(2)
b,
or equivalently,
φ
A
(id
H
⊗m
A
) = m
A
(φ
A
, φ
A
)(id
H
⊗τ ⊗id
A
)(∆
H
⊗id
A
⊗id
A
),
and
h1
A
= ε(h)1
A
,
or equivalently,
φ
A
(id
H
⊗η
A
) = η
A
ε
H
.
Deﬁnition 1.2.2. A left Hmodule coalgebra is a coalgebra (C, ∆
C
, ε
C
) equipped with
a left Hmodule structure via the left action
φ
C
: H ⊗C →C, φ
C
(h ⊗c) = hc
6
such that ∆
C
and ε
C
are Hmodule maps i.e. :
(hc)
(1)
⊗(hc)
(2)
= h
(1)
c
(1)
⊗h
(2)
c
(2)
,
or equivalently,
∆
C
φ
C
= (φ
C
⊗φ
C
)(id
H
⊗τ ⊗id
C
)(∆
H
⊗∆
C
),
and
ε
C
(hc) = ε
H
(h)ε
C
(c),
or equivalently,
ε
C
φ
C
= ε
H
⊗ε
C
.
Example 1.2.1. The conjugation action H ⊗ H → H, g ⊗ h → g
(1)
hS(g
(2)
) gives
H the structure of a left Hmodule algebra.
Example 1.2.2. The multiplication m
H
: H⊗H →H turns H into a left Hmodule
coalgebra.
1.3 Quasitriangular quasiHopf algebras
In this section we recall the deﬁnitions of a quasitriangular Hopf algebra, quasiHopf
algebra, quasitriangular quasiHopf algebra, and provide some examples.
Deﬁnition 1.3.1. [22, 31] A quasitriangular Hopf algebra consists of a Hopf algebra
H and an invertible element R = R
1
⊗ R
2
of H ⊗ H, with the inverse R
−1
=
(R
−1
)
1
⊗ (R
−1
)
2
, satisfying the following relations: (note that we have used the
Sweedler’s notation, therefore R and R
−1
are not necessarily simple tensors in H⊗H)
(∆⊗id)(R) = R
13
R
23
,
7
(id ⊗∆)(R) = R
13
R
12
,
τ∆(h) = R(∆(h))R
−1
, ∀h ∈ H, (1.3.2)
where
R
13
= R
1
⊗1 ⊗R
2
, R
23
= 1 ⊗R
1
⊗R
2
, R
12
= R
1
⊗R
2
⊗1.
Deﬁnition 1.3.2. In Deﬁnition (1.3.1), if R
−1
= τ(R), i.e., if (R
−1
)
1
⊗ (R
−1
)
2
=
R
2
⊗R
1
, then (H, R) is called a triangular Hopf algebra.
Remark 1.3.1. One can deﬁne a coquasitriangular Hopf algebra in a dual fashion
[22, 31].
Proposition 1.3.1. If (H, R) is a quasitrangular Hopf algebra, then:
(ε ⊗id)(R) = (id ⊗ε)(R) = 1,
(S ⊗id)(R) = R
−1
, (id ⊗S)(R) = R,
R
12
R
13
R
23
= R
23
R
13
R
12
.
Example 1.3.1. Every cocommutative Hopf algebra is quasitriangular with the trivial
quasitriangular structure R = 1 ⊗1.
We note that every commutative quasitrangular Hopf algebra is cocommutative.
Example 1.3.2. [31] Let CZ
n
be the group algebra of the ﬁnite cyclic group of order
n. By Example (1.1.1) it is a commutative and cocommutative Hopf algebra. One can
deﬁne a nontrivial quasitriangular structure on CZ
n
by:
R = (1/n)
n−1
a,b=0
e
(−2πiab)/n
g
a
⊗g
b
,
8
where g is the generator of Z
n
.
Example 1.3.3. If in Example (1.3.2) we let n = 2 then we have H = CZ
2
with the
nontrivial quasitriangular structure R = R
1
⊗R
2
deﬁned by
R := (
1
2
)(1 ⊗1 + 1 ⊗g + g ⊗1 −g ⊗g),
where g is the generator of the cyclic group Z
2
.
If in the deﬁnition of a bialgebra we weaken the coassociativity of the comulti
plication in a suitable way, we obtain the notion of a quasibialgebra.
Deﬁnition 1.3.3. A quasibialgebra consists of a quadruple (H, m, η, ∆, ε) where
(H, m, η) is an unital associative algebra and (H, ∆, ε) is a counital coalgebra in which
the coassociativity property is replaced by a weaker version:
(id ⊗∆)∆ = φ[(∆⊗id)∆]φ
−1
. (1.3.3)
Here φ = φ
1
⊗φ
2
⊗φ
3
, called an associator, is an invertible element of H ⊗H ⊗H
with inverse φ
−1
= (φ
−1
)
1
⊗(φ
−1
)
2
⊗(φ
−1
)
3
, and satisfying :
[(id ⊗id ⊗∆)(φ)] [(∆⊗id ⊗id)(φ)] = (1 ⊗φ) [(id ⊗∆⊗id)(φ)] (φ ⊗1), (1.3.4)
(id ⊗ε ⊗id)(φ) = 1. (1.3.5)
As we shall see in Chapter (3), the above properties (1.3.3)(1.3.5) are equivalent
to saying that the category of representations of the algebra (H, m, η) is a monoidal
category [22].
Deﬁnition 1.3.4. A quasiHopf agebra consists of a quasibialgebra (H, m, η, ∆, ε, φ),
a bijective linear map S : H →H and elements α and β of H such that:
S(h
(1)
)αh
(2)
= ε(h)α, ∀h ∈ H,
9
h
(1)
βS(h
(2)
) = ε(h)β, ∀h ∈ H,
φ
1
βS(φ
2
)αφ
3
= 1,
S((φ
−1
)
1
)α(φ
−1
)
2
βS((φ
−1
)
3
) = 1,
where φ
−1
= (φ
−1
)
1
⊗(φ
−1
)
2
⊗(φ
−1
)
3
is the inverse of φ in H ⊗H ⊗H.
Remark 1.3.2. By dualizing the above two deﬁnitions one can deﬁne a coquasi
bialgebra or a coquasiHopf algebra [22, 31].
Deﬁnition 1.3.5. A quasitriangular quasiHopf algebra is a quasiHopf algebra
(H, m, η, ∆, ε, φ, S, α, β) equipped with an invertible element R = R
(1)
⊗R
(2)
of H⊗H
satisfying :
(∆⊗id)(R) = φ
312
R
13
φ
−1
132
R
23
φ,
(id ⊗∆)(R) = φ
−1
231
R
13
φ
213
R
12
φ
−1
,
τ∆(h) = R(∆(h))R
−1
, ∀h ∈ H.
As we shall see in Chapter (3), the above properties are equivalent to saying
that that the monoidal category of Hmodules is a braided monoidal category [22].
Recall that a pair (X, Y ) of Hopf algebras (bialgebras) is called a matched pair
if X is a left Y module coalgebra via : Y ⊗ X → X and Y is a right Xmodule
coalgebra via : Y ⊗X →Y such that:
y (xx
/
) = (y
(1)
x
(1)
)((y
(2)
x
(2)
) x
/
)), y 1 = ε(y)1,
(yy
/
) x = (y (y
/(1)
x
(1)
))(y
/(2)
x
(2)
), 1 x = ε(x)1,
y
(1)
x
(1)
⊗y
(2)
x
(2)
= y
(2)
x
(2)
⊗y
(1)
x
(1)
,
for all x, x
/
in X and y, y
/
in Y . For example for any ﬁnitedimensional Hopf algebra
(H, m, η, ∆, ε, S) with S, the pair ((H
op
)
∗
, H) is a matched pair. Here (H
op
)
∗
=
10
(H
∗
, (m
op
)
∗
, η, ∆
∗
, ε, S
∗
). For any matched pair (X, Y ) of Hopf algebras there exist
a unique Hopf algebra structure on the vector space X ⊗Y deﬁned by:
(x ⊗y)(x
/
⊗y
/
) = x(y
(1)
x
/(1)
) ⊗(y
(2)
x
/(2)
)y
/
,
∆(x ⊗y) = (x
(1)
⊗y
(1)
) ⊗(x
(2)
⊗y
(2)
),
1
X⊗Y
= 1 ⊗1, ε(x ⊗y) = ε(x)ε(y),
S(x ⊗y) = S(y
(2)
) S(x
(2)
) ⊗S(y
(1)
) S(x
(1)
).
This Hopf algebra is called the bicrossed product of X and Y and denoted by
XY [22]. Using this structure for the matched pair ((H
op
)
∗
, H), when H is ﬁnite
dimensional, one obtains the Drinfeld’s quantum double of H, D(H) = (H
op
)
∗
H.
Its structure is explicitly given in the next example.
Example 1.3.4. [22] Let H be a ﬁnitedimensional Hopf algebra with an invertible
anipode, and (H
op
)
∗
be its dual Hopf algebra as mentioned above. The quantum double
of D(H) = (H
op
)
∗
H is the vector space (H
op
)
∗
⊗H with the structure:
(f ⊗a)(g ⊗b) = fg
(S
−1
(a
(3)
)a
(1)
)
⊗a
(2)
b, 1
D(H)
= 1 ⊗1,
∆(f ⊗a) = (f
(1)
⊗a
(1)
) ⊗(f
(2)
⊗a
(2)
), ε(f ⊗a) = ε(a)f(1).
In the ﬁrst formula the map g
(S
−1
(a
(3)
)a
(1)
)
is deﬁned by:
g
(S
−1
(a
(3)
)a
(1)
)
(x) = g(S
−1
(a
(3)
)xa
(1)
).
The quantum double D(H) is a quasitriangular Hopf algebra via R ∈ D(H) ⊗
D(H) deﬁned by:
R =
i∈I
(1 ⊗e
i
) ⊗(e
∗
i
⊗1),
11
where ¦e
i
¦
i∈I
is a basis for H, and ¦e
∗
i
¦
i∈I
is the dual basis for H
∗
.
Example 1.3.5. [22] Let G be a ﬁnite group, the quantum double of G is deﬁned to
be D(kG) = ((kG)
op
)
∗
⊗ kG. It is usually denoted by D(G), and has the following
structure:
(g
∗
⊗a)(h
∗
⊗b) = g
∗
(aha
−1
)
∗
⊗ab = δ
g,aha
−1
(g
∗
⊗ab),
∆(g
∗
⊗a) =
st=g
s
∗
⊗a ⊗t
∗
⊗a,
ε(g
∗
⊗a) = δ
g,1
,
S(g
∗
⊗a) = (a
−1
g
−1
a)
∗
⊗a
−1
,
R =
g∈G
(1 ⊗g) ⊗(g
∗
⊗1).
Example 1.3.6. [22] Let G be a ﬁnite group and let ω : G G G → k
be a
normalized 3cocycl, i.e, for all a, b, c and d in G :
ω(a, b, c)ω(d, ab, c)ω(d, a, b)
ω(da, b, c)ω(d, a, bc)
= 1,
and
ω(a, b, c) = 1 if a, b, or c = 1.
For any a, b, g ∈ G, we deﬁne
θ
g
(a, b) =
ω(g, a, b)ω(a, b, (ab)
−1
gab)
ω(a, a
−1
ga, b)
,
γ
g
(a, b) =
ω(a, b, g)ω(g, g
−1
ag, g
−1
bg)
ω(a, g, g
−1
bg)
.
The twisted quantum double D
ω
(G) of G with respect to ω is (kG)
∗
⊗kG as a vector
12
space with the further structures given by:
(g
∗
⊗a)(h
∗
⊗b) = θ
g
(a, b)g
∗
(aha
−1
)
∗
⊗ab,
∆(g
∗
⊗a) =
st=g
γ
a
(s, t)s
∗
⊗a ⊗t
∗
⊗a,
ε(g
∗
⊗a) = δ
g,1
,
S(g
∗
⊗a) = θ
g
−1
(a, a
−1
)
−1
γ
a
(g, g
−1
)
−1
(a
−1
g
−1
a)
∗
⊗a
−1
,
R =
g∈G
(g
∗
⊗1) ⊗((
h
h
∗
) ⊗g),
φ =
g,s,t∈G
ω(g, s, t)
−1
g
∗
⊗1 ⊗s
∗
⊗1 ⊗t
∗
⊗1,
α = 1
D
ω
(G)
= 1
∗
⊗1 , β =
g∈G
ω(g, g
−1
, g)g
∗
⊗1,
where ¦g
∗
[g ∈ G¦ is the dual basis of the canonical basis of kG and δ
g,1
is the
Kronecker delta.
The above structure deﬁnes D
ω
(G) = (kG)
∗
⊗ kG as a quasitriangular quasi
Hopf algebra, and is isomorphic to the quantum double D(G) if ω is trivial, i.e, if
ω(a, b, c) = 1, for all a, b, c in G.
Chapter 2
Cyclic Cohomology Basics
In this chapter we review basics of cyclic (co)homology theory. We start with the
cyclic cohomology of algebras, then proceed to more general concepts of (co)cyclic and
para(co)cyclic modules and their cyclic (co)homology. Next we recall the Hopf cyclic
cohomology, and Hopf cyclic cohomology with coeﬃcients. For this latter theory we
recall the notion of stable anti YetterDrinfeld modules.
2.1 Cyclic cohomology of algebras
In this section we recall three equivalent ways to deﬁne the cyclic cohomology of
an algebra A. They are based on Connes’ complex, (b, b
/
)bicomplex, and (b, B)
bicomplex. One can also deﬁne the cyclic homology of algebras, in a dual fashion.
In the next sections we give the fourth possible deﬁnition of cyclic (co)homology for
algebras, by deﬁning the notion of (co)cyclic module. All algebras and coalgebras are
unital and counital respectively.
Let us ﬁrst recall the Hochschild cohomology of an algebra A. The Hochschild
cochain complex of A is deﬁned by
C
0
(A)
b
−−−→ C
1
(A)
b
−−−→ C
2
(A)
b
−−−→ C
3
(A) . . . ,
(2.1.1)
where,
C
n
(A) = Hom
C
(A
⊗(n+1)
, C), n = 0, 1, 2, ...
13
14
and the Hochschild coboundary b : C
n
(A) →C
n+1
(A) is deﬁned by
bϕ(a
0
, ..., a
n+1
) =
n
i=0
(−1)
i
ϕ(a
0
, ..., a
i
a
i+1
, ...a
n+1
) + (−1)
n+1
ϕ(a
n+1
a
0
, ..., a
n
).
(2.1.2)
The Hochschild cohomology of A denoted by HH
∗
(A) is deﬁned to be the cohomology
of the cochain complex (2.1.1) [28].
We now turn to the deﬁnition of cyclic cohomology of algebras.
Deﬁnition 2.1.1. Let A be a unital algebra. The cyclic cohomology of A denoted by
HC
∗
(A) is the cohomology of the the following cochain complex known as Connes’
complex
C
0
λ
(A)
b
−−−→ C
1
λ
(A)
b
−−−→ C
2
λ
(A)
b
−−−→ C
3
λ
(A) . . .
where C
n
λ
(A) is the subcomplex of C
n
(A) containing those ϕ ∈ C
n
(A) that satisfy
ϕ(a
0
, ..., a
n
) = (−1)
n
ϕ(a
n
, a
0
, ..., a
n−1
), n = 0, 1, 2, ...
for all a
0
, ..., a
n
in A, and b : C
n
λ
(A) → C
n+1
λ
(A) is the Hochschild coboundary as
deﬁned in formula (2.1.2).
There are alternative ways to deﬁne cyclic cohomology. Consider the Hochschild
complex C
n
(A) = Hom
C
(A
⊗(n+1)
, C). Deﬁne the so called faces δ
i
: C
n−1
(A) →
C
n
(A) and cyclic maps τ
n
: C
n
(A) →C
n
(A) by:
δ
i
ϕ(a
0
, ..., a
n
) =
_
¸
¸
_
¸
¸
_
ϕ(a
0
, ..., a
i
a
i+1
, ..., a
n
) 0 ≤ i < n
ϕ(a
n
a
0
, a
1
, ..., a
n−1
) i = n
and
τ
n
ϕ(a
0
, ..., a
n
) = ϕ(a
n
, a
0
, ..., a
n−1
).
15
Let
b =
n
0
(−1)
i
δ
i
,
b
/
=
n−1
0
(−1)
i
δ
i
,
λ
n
= (−1)
n
τ
n
,
and denoting λ
n
by λ,
N = 1 + λ + λ
2
+ ... + λ
n
.
The relations
b
2
= b
/2
= 0, (1 −λ)b = b
/
(1 −λ), and (1 −λ)N = N(1 −λ) = 0,
can be veriﬁed [6, 28]. Therefore one can construct the so called (b, b
/
)bicomplex of
A, denoted by c(A), as:
.
.
.
.
.
.
.
.
.
C
2
(A)
1−λ
−−−→ C
2
(A)
N
−−−→ C
2
(A)
1−λ
−−−→
¸
¸
¸b
¸
¸
¸−b
¸
¸
¸b
C
1
(A)
1−λ
−−−→ C
1
(A)
N
−−−→ C
1
(A)
1−λ
−−−→
¸
¸
¸b
¸
¸
¸−b
¸
¸
¸b
C
0
(A)
1−λ
−−−→ C
0
(A)
N
−−−→ C
0
(A)
1−λ
−−−→
It can be shown that the cyclic cohomology of A is isomorphic to the cohomology of
the total complex Totc(A) [28].
Let us deﬁne degeneracies σ
i
: C
n+1
(A) →C
n
(A) by,
σ
i
ϕ(a
0
, ..., a
n
) = ϕ(a
0
, ...a
i
, 1, a
i+1
, ..., a
n
), 0 ≤ i ≤ n.
16
Now let
B = Ns(1 −λ) : C
n+1
(A) →C
n
(A),
where the operator s : C
n+1
(A) →C
n
(A) is deﬁned by
sϕ(a
0
, ..., a
n
) = ϕ(1, a
0
, ..., a
n
).
Then we have the relations: [7]
b
2
= B
2
= 0, and bB + Bb = 0.
We construct the (b, B)bicomplex of A, denoted by B(A), as:
.
.
.
.
.
.
.
.
.
C
2
(A)
B
−−−→ C
1
(A)
B
−−−→ C
0
(A)
b
¸
¸
¸ b
¸
¸
¸
C
1
(A)
B
−−−→ C
0
(A)
b
¸
¸
¸
C
0
(A)
Again it can be shown that the cyclic cohomology of A is isomorphic to the cohomol
ogy of the total complex TotB(A) [7, 28].
Remark 2.1.1. One can use the (b, b
/
)bicomplex to drive the long exact sequence of
Connes as in [28]. Alternatively, consider the short exact sequence:
0 →C
λ
(A) →C(A)
π
→C(A)/C
λ
(A) →0
One can show that the cohomology of the quotient complex C(A)/C
λ
(A) is isomorphic
to the cyclic cohomology of A with a shift in dimension by one. The resulting long
17
exact sequence is Connes’ long exact sequence:
−→HC
n
(A)
I
−→HH
n
(A)
B
−→HC
n−1
(A)
S
−→HC
n+1
(A) −→
The operator S is known as Connes’ periodicity operator.
2.2 Cyclic and cocyclic modules
Yet another approach to cyclic (co)homology is based on the notion of (co)cyclic
modules [6].
Deﬁnition 2.2.1. The simplicial category ∆ is a small category whose objects are
the totally ordered sets (cf. e.g. [28])
[n] = ¦0 < 1 < < n¦, n = 0, 1, 2, . . . .
A morphism f : [n] →[m] of ∆ is an order preserving, i.e. monotone nondecreasing,
map f : ¦0, 1, . . . , n¦ →¦0, 1, . . . , m¦.
Of particular interest among the morphisms of ∆ are faces δ
i
and degeneracies
σ
j
,
δ
i
: [n −1] →[n], i = 0, 1, . . . , n
σ
j
: [n + 1] →[n], j = 0, 1, . . . , n.
By deﬁnition, δ
i
is the unique injective morphism missing i, and σ
j
is the unique
surjective morphism identifying j with j + 1. It can be checked that they satisfy the
18
following simplicial identities:
δ
j
δ
i
= δ
i
δ
j−1
if i < j,
σ
j
σ
i
= σ
i
σ
j+1
if i ≤ j,
σ
j
δ
i
=
_
¸
¸
¸
¸
¸
_
¸
¸
¸
¸
¸
_
δ
i
σ
j−1
if i < j
id if i = j or i = j + 1
δ
i−1
σ
j
if i > j + 1.
(2.2.3)
Every morphism of ∆ can be uniquely decomposed as a product of faces followed by
a product of degeneracies [28].
Deﬁnition 2.2.2. The cyclic category Λ has the same set of objects as ∆ and in fact
contains ∆ as a subcategory. Morphisms of Λ are generated by simplicial morphisms
δ
i
and σ
j
as above and new morphisms τ
n
: [n] →[n] for n ≥ 0 deﬁned by
τ
n
(i) =
_
¸
¸
_
¸
¸
_
n i = 0
i −1 i ,= 0
In other words τ
n
is the following cyclic permutation
τ
n
=
_
_
0, 1, 2, ..., n
n, 0, 1, ..., n −1
_
_
.
The following extra relations hold between δ
i
, σ
i
and τ
n
in Λ:
τ
n
δ
i
= δ
i−1
τ
n−1
1 ≤ i ≤ n
τ
n
δ
0
= δ
n
τ
n
σ
i
= σ
i−1
τ
n+1
1 ≤ i ≤ n
τ
n
σ
0
= σ
n
τ
2
n+1
(2.2.4)
τ
n+1
n
= id.
19
Let Λ
op
denote the opposite category of the category Λ.
Deﬁnition 2.2.3. Let c be a category. A cyclic object in c is a functor Λ
op
→ c.
Equivalently, a cyclic object in c is given by the data
X = (X
n
, δ
i
, σ
i
, τ
n
), n ≥ 0,
where X
n
, n ≥ 0, is a family of objects of c, δ
i
: X
n
→ X
n−1
, 0 ≤ i ≤ n,
σ
i
: X
n
→ X
n+1
, 0 ≤ i ≤ n, and τ
n
: X
n
→ X
n
called faces, degeneracies and
cyclic maps, respectively, are morphisms of c satisfying the relations, dual to the
relations (2.2.3) and (2.2.4):
δ
i
δ
j
= δ
j−1
δ
i
if i < j,
σ
i
σ
j
= σ
j+1
σ
i
if i ≤ j,
δ
i
σ
j
=
_
¸
¸
¸
¸
¸
_
¸
¸
¸
¸
¸
_
σ
j−1
δ
i
if i < j
id if i = j or i = j + 1
σ
j
δ
i−1
if i > j + 1
(2.2.5)
and
δ
i
τ
n
= τ
n−1
δ
i−1
1 ≤ i ≤ n,
δ
0
τ
n
= δ
n
,
σ
i
τ
n
= τ
n+1
σ
i−1
1 ≤ i ≤ n,
σ
0
τ
n
= τ
2
n+1
σ
n
,
τ
n+1
n
= id. (2.2.6)
Deﬁnition 2.2.4. Let k be a commutative ring. A cyclic object in the category of
kmodules is called a cyclic kmodule or just a cyclic module.
20
Deﬁnition 2.2.5. A cocyclic object in c is a functor Λ →c. Equivalently, a cocyclic
object in c is given by the data
X = (X
n
, δ
i
, σ
i
, τ
n
), n ≥ 0,
where X
n
, n ≥ 0, are objects of c, and faces δ
i
: X
n−1
→ X
n
, 0 ≤ i ≤ n,
degeneracies σ
i
: X
n+1
→ X
n
, 0 ≤ i ≤ n, and cyclic maps τ
n
: X
n
→ X
n
are
morphisms of c satisfying all the relations (2.2.3) and (2.2.4).
Deﬁnition 2.2.6. Let k be a commutative ring. A cocyclic object in the category of
kmodules is called a cocyclic kmodule or just a cocyclic module.
For any commutative ring k, we denote the category of cyclic kmodules by
Λ
k
. A morphism of cyclic kmodules is a natural transformation between the cor
responding functors. Equivalently, a morphism f : X → Y consists of a sequence
of klinear maps f
n
: X
n
→ Y
n
compatible with faces, degeneracies, and cyclic op
erators. One can of course talk about Λ
k
the category of cocyclic modules in the
same manner. It is clear that Λ
k
is an abelian category. The kernel and coker
nel of a morphism f are deﬁned pointwise: (Ker f)
n
= Ker f
n
: X
n
→ Y
n
and
(Coker f)
n
= Coker f
n
: X
n
→Y
n
. More generally, if / is any abelian category then
the category Λ
/
of cyclic objects in / is itself an additive category.
Let Alg
k
denote the category of unital kalgebras and unital algebra homomor
phisms. Note that Alg
k
is not an additive category.
Example 2.2.1. To an algebra A in Alg
k
, we associate the cyclic kmodule A
deﬁned
by
A
n
= A
⊗(n+1)
, n ≥ 0,
21
with face, degeneracy and cyclic operators given by
δ
i
(a
0
, ..., a
n
) =
_
¸
¸
_
¸
¸
_
(a
0
, ..., a
i
a
i+1
, ..., a
n
) 0 ≤ i < n
(a
n
a
0
, ..., a
n−1
) i = n
σ
i
(a
0
, ..., a
n
) = (a
0
, ...a
i
, 1, a
i+1
, ..., a
n
), 0 ≤ i ≤ n,
τ
n
(a
0
, ..., a
n
) = (a
n
, a
0
, ..., a
n−1
).
In above formulas by (a
0
, ..., a
n
) we mean (a
0
⊗ a
1
⊗ ... ⊗ a
n
) ∈ A
⊗(n+1)
. We
will use this convention in sequel when there is no confusion. From the above cyclic
module one deﬁnes the cyclic homology of the algebra A (cf. Section 2.4).
A unital algebra map f : A → B induces a morphism of cyclic modules f
:
A
→B
by f
(a
0
, ..., a
n
) = (f(a
0
), , f(a
n
)), and this deﬁnes a functor .
: Alg
k
−→Λ
k
,
Theorem 2.2.1. [6] For any unital kalgebra A, there is a canonical isomorphism:
HC
n
(A) · Ext
n
Λ
k
(A
, k
), for all n ≥ 0. (2.2.7)
In fact the isomorphism (2.2.7) can be made explicit for n = 0 [6]. Given a
trace t : A →k, one deﬁnes a map of cyclic modules t
: A
→k
by:
t
(a
0
, a
1
, ..., a
n
) := t(a
0
a
1
...a
n
).
Example 2.2.2. Let (C, ∆, ε) be a coalgebra. To C, we associate the cocyclic module
C
deﬁned by
C
n
= C
⊗(n+1)
, n ≥ 0,
22
with face, degeneracy and cyclic operators deﬁned by
δ
i
(c
0
, ..., c
n−1
) =
_
¸
¸
_
¸
¸
_
(c
0
, ..., c
(1)
i
, c
(2)
i
, ..., c
n−1
) 0 ≤ i < n
(c
(2)
0
, c
1
, ..., c
n−1
, c
(1)
0
) i = n
σ
i
(c
0
, ..., c
n+1
) = (c
0
, ...c
i
, ε(c
i+1
), ..., c
n+1
), 0 ≤ i ≤ n,
τ
n
(c
0
, ..., c
n
) = (c
1
, ..., c
n
, c
0
).
In the next sections we shall see more examples of cyclic and cocyclic modules.
2.3 Paracyclic and cocyclic modules
One can deﬁne the notion of para(co)cyclic object by eliminating the last relation,
τ
n+1
n
= id in the deﬁnition of a (co)cyclic object.
Deﬁnition 2.3.1. A paracocyclic object in a category c is given by the data
X = (X
n
, δ
i
, σ
i
, τ
n
), n ≥ 0,
where X
n
, n ≥ 0, are objects of c, faces δ
i
: X
n−1
→X
n
, 0 ≤ i ≤ n, degeneracies
σ
i
: X
n+1
→ X
n
, 0 ≤ i ≤ n, and cyclic maps τ
n
: X
n
→ X
n
are morphisms of c
satisfying relations (2.2.3) and:
τ
n
δ
i
= δ
i−1
τ
n−1
1 ≤ i ≤ n
τ
n
δ
0
= δ
n
τ
n
σ
i
= σ
i−1
τ
n+1
1 ≤ i ≤ n
τ
n
σ
0
= σ
n
τ
2
n+1
(2.3.8)
23
One also deﬁnes the notion of paracyclic object in the same manner, i.e., by
eliminating the condition τ
n+1
n
= id from the relations (2.2.5) and (2.2.6).
Let k be a commutative ring. A para(co)cyclic object in the category of k
modules is called a para(co)cyclic kmodule.
Example 2.3.1. Let A be a unital algebra and σ : A →A be an automorphism of A.
One can associate to (A, σ) the paracyclic module A
σ
deﬁned by
(A
σ
)
n
= A
⊗(n+1)
, n ≥ 0,
with face, degeneracy and cyclic operators given by
δ
i
(a
0
, ..., a
n
) =
_
¸
¸
_
¸
¸
_
(a
0
, ..., a
i
a
i+1
, ..., a
n
) 0 ≤ i < n
(σ(a
n
)a
0
, ..., a
n−1
) i = n
σ
i
(a
0
, ..., a
n
) = (a
0
, ...a
i
, 1, a
i+1
, ..., a
n
), 0 ≤ i ≤ n,
τ
n
(a
0
, ..., a
n
) = (σ(a
n
), a
0
, ..., a
n−1
).
Note that in this example
τ
n+1
n
(a
0
, ..., a
n
) = (σ(a
0
), σ(a
1
), ..., σ(a
n
)).
Therefore τ
n+1
n
,= id except when σ = id. In fact for σ = id we get the cyclic module
of Example (2.2.1).
Example (2.3.1) is an special case of the following more general one.
Example 2.3.2. Let A be a unital algebra and G be a group acting on A by auto
morphisms. We can associate to (A, G) the paracyclic module A
G
deﬁned by [16]
(A
G
)
n
= CG⊗A
⊗(n+1)
, n ≥ 0,
24
with face, degeneracy and cyclic operators given by
δ
i
(g, a
0
, ..., a
n
) =
_
¸
¸
_
¸
¸
_
(g, a
0
, ..., a
i
a
i+1
, ..., a
n
) 0 ≤ i < n
(g, (g
−1
a
n
)a
0
, ..., a
n−1
) i = n
σ
i
(a
0
, ..., a
n
) = (a
0
, ...a
i
, 1, a
i+1
, ..., a
n
), 0 ≤ i ≤ n,
τ
n
(g, a
0
, ..., a
n
) = (g, g
−1
a
n
, a
0
, ..., a
n−1
).
Example 2.3.3. The above example can be further generalized as follows [1]. Let H
be a Hopf algebra and A be an Hmodule algebra. One can introduce the paracyclic
module
(A
H
)
n
= H ⊗A
⊗(n+1)
,
by deﬁning face, degeneracy and cyclic operators as follows:
δ
i
(g, a
0
, ..., a
n
) =
_
¸
¸
_
¸
¸
_
(g, a
0
, ..., a
i
a
i+1
, ..., a
n
) 0 ≤ i < n
(g
(1)
, (S(g
(2)
)a
n
)a
0
, ..., a
n−1
) i = n
σ
i
(a
0
, ..., a
n
) = (a
0
, ...a
i
, 1, a
i+1
, ..., a
n
), 0 ≤ i ≤ n,
τ
n
(g, a
0
, ..., a
n
) = (g
(1)
, (S(g
(2)
a
n
), a
0
, ..., a
n−1
).
Given a paracocyclic object (X
n
, δ
i
, σ
i
, τ
n
), n ≥ 0, in an abelian category c,
we can always deﬁne a cocyclic object by considering
X
n
:= ker (id −τ
n+1
n
), (2.3.9)
and restricting the faces, degeneracies, and cyclic operators to these subspaces.
25
In a dual fashion given a paracyclic object (X
n
, δ
i
, σ
i
, τ
n
), n ≥ 0, in an abelian
category c, we deﬁne a cyclic object by
X
n
:=
X
n
Im(id −τ
n+1
n
)
, (2.3.10)
and the faces, degeneracies, and cyclic operators are naturally induced by the given
δ
i
, σ
i
and τ
n
to these quotients.
2.4 Cyclic (co)homology of (co)cyclic objects
To any (co)cyclic object in an abelian category, one can assign a cyclic (co)homology
in diﬀerent, but equivalent, ways, analogous to what we deﬁned for a unital algebra
in section (2.1).
Deﬁnition 2.4.1. Let c be an abelian category and X = (X
n
, δ
i
, σ
i
, τ
n
), n ≥ 0
be a cocyclic object in c. The cyclic cohomology of X denoted by HC
∗
(X) is the
cohomology of the cochain complex
X
0
λ
b
−−−→ X
1
λ
b
−−−→ X
2
λ
b
−−−→ X
3
λ
. . . ,
Here
X
n
λ
= Ker (1 −λ
n
),
where
λ
n
= (−1)
n
τ
n
,
and b : X
n−1
→X
n
is deﬁned by
b =
n
i=0
(−1)
i
δ
i
.
26
Deﬁnition 2.4.2. Let X = (X
n
, δ
i
, σ
i
, τ
n
), n ≥ 0 be a cyclic object in an abelian
category c. The cyclic homology of X denoted by HC
∗
(X) is the homology of the
chain complex
X
λ
0
b
←−−− X
λ
1
b
←−−− X
λ
2
b
←−−− X
λ
3
,
where
X
λ
n
=
X
n
Im(1 −λ
n
)
,
and b : X
n
→X
n−1
is deﬁned by
b =
n
i=0
(−1)
i
δ
i
.
When the ambient categry is the category of vector spaces over a ﬁeld k of
characteristic zero, to obtain the cyclic cohomology of the cocyclic object X =
(X
n
, δ
i
, σ
i
, τ
n
) of c = V ect
k
one can alternatively construct the (b, b
/
)bicomplex,
denoted by c(X), as:
.
.
.
.
.
.
.
.
.
X
2
1−λ
−−−→ X
2
N
−−−→ X
2
1−λ
−−−→
¸
¸
¸b
¸
¸
¸−b
¸
¸
¸b
X
1
1−λ
−−−→ X
1
N
−−−→ X
1
1−λ
−−−→
¸
¸
¸b
¸
¸
¸−b
¸
¸
¸b
X
0
1−λ
−−−→ X
0
N
−−−→ X
0
1−λ
−−−→
where
b
/
=
n−1
i=0
(−1)
i
δ
i
λ
n
= (−1)
n
τ
n
,
N = 1 + λ + λ
2
+ ... + λ
n
.
27
The relations
b
2
= b
/2
= 0, (1 −λ)b = b
/
(1 −λ), and (1 −λ)N = N(1 −λ) = 0,
are veriﬁed. It can be shown that the cyclic cohomology of X is isomorphic to the
cohomology of the total complex Totc(X).
When c = V ect
k
where k is a ﬁeld of characteristic zero, one can also construct
the (b, B)bicomplex of X, denoted by B(X), as
.
.
.
.
.
.
.
.
.
X
2
B
−−−→ X
1
B
−−−→ X
0
b
¸
¸
¸ b
¸
¸
¸
X
1
B
−−−→ X
0
b
¸
¸
¸
X
0
with B deﬁned by
B = Ns(1 −λ),
where the operator s called the extra degeneracy is given by
s = σ
n
τ
n+1
: X
n+1
→X
n
.
Note that the relations:
b
2
= B
2
= 0, and bB + Bb = 0.
are satisﬁed. Again it can be shown that the cyclic cohomology of X is isomorphic
to the cohomology of the total complex TotB(X).
Remark 2.4.1. The same statements are true for the cyclic homology of a cyclic
28
object X = (X
n
, δ
i
, σ
i
, τ
n
) in c = V ect
k
with dualized diagrams as follows.
.
.
.
.
.
.
.
.
.
X
2
1−λ
←−−− X
2
N
←−−− X
2
1−λ
←−−−
¸
¸
_b
¸
¸
_−b
¸
¸
_b
X
1
1−λ
←−−− X
1
N
←−−− X
1
1−λ
←−−−
¸
¸
_b
¸
¸
_−b
¸
¸
_b
X
0
1−λ
←−−− X
0
N
←−−− X
0
1−λ
←−−−
.
.
.
.
.
.
.
.
.
X
2
B
←−−− X
1
B
←−−− X
0
b
¸
¸
_ b
¸
¸
_
X
1
B
←−−− X
0
b
¸
¸
_
X
0
and
B = (1 −λ)sN,
where
s = τ
n+1
σ
n
: X
n
→X
n+1
.
Example 2.4.1. Following Sections (2.1) and (2.2), the cyclic cohomology of a unital
kalgebra A is an example of the cyclic cohomology of a cocyclic module. Also Example
(2.2.1) gives a cyclic object in the category of kmodules, and from that one deﬁnes
the cyclic homology of a unital kalgebra A.
Example 2.4.2. Considering the cocyclic object given in Example (2.2.2) and using
methods of this section one can deﬁne the cyclic cohomology of a coalgebra C.
29
2.5 Periodic cyclic cohomology of cocyclic objects
Deﬁnition 2.5.1. Let c be an abelian category and X = (X
n
, δ
i
, σ
i
, τ
n
), n ≥ 0, be a
cocyclic object in c. The periodic cyclic cohomology of X denoted by HP
∗
(X), where
∗ = 0, 1, is deﬁned as the following direct limit:
HP
i
(X) = lim
→
n
HC
i+2n
(X), i = 0, 1,
where the direct limit, lim
→
n
, is taken by using the Connes’ periodicity operator S :
HC
n
(X) →HC
n+2
(X).
When c = V ect
k
where k is a ﬁeld of characteristic zero, alternatively, HP
∗
(X)
can be deﬁned as the cohomology of the total complex of either of the following two
bicomplexes:
The bicomplex
¯
c(X)
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
X
2
1−λ
−−−→ X
2
N
−−−→ X
2
1−λ
−−−→ X
2
1−λ
−−−→ X
2
N
−−−→
¸
¸
¸−b
¸
¸
¸b
¸
¸
¸b
¸
¸
¸−b
¸
¸
¸b
X
1
1−λ
−−−→ X
1
N
−−−→ X
1
1−λ
−−−→ X
1
1−λ
−−−→ X
1
N
−−−→
¸
¸
¸−b
¸
¸
¸b
¸
¸
¸b
¸
¸
¸−b
¸
¸
¸b
X
0
1−λ
−−−→ X
0
N
−−−→ X
0
1−λ
−−−→ X
0
1−λ
−−−→ X
0
N
−−−→
30
Or the bicomplex
¯
B(X)
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
X
4
B
−−−→ X
3
B
−−−→ X
2
B
−−−→ X
1
B
−−−→ X
0
b
¸
¸
¸ b
¸
¸
¸ b
¸
¸
¸ b
¸
¸
¸
X
3
B
−−−→ X
2
B
−−−→ X
1
B
−−−→ X
0
b
¸
¸
¸ b
¸
¸
¸ b
¸
¸
¸
X
2
B
−−−→ X
1
B
−−−→ X
0
b
¸
¸
¸ b
¸
¸
¸
X
1
B
−−−→ X
0
b
¸
¸
¸
X
0
Note that the the total complexes Tot
¯
c(X) and Tot
¯
B(X) are as follows, respec
tively:
i≥0
X
i
−−−→
i≥0
X
i
−−−→
i≥0
X
i
i≥0
X
2i
(b+B)
−−−−→
i≥0
X
2i+1
(b+B)
−−−−→
i≥0
X
2i
(b+B)
−−−−→
i≥0
X
2i+1
We shall provide some examples in the next section.
2.6 Hopf cyclic cohomology
Hopf cyclic cohomology is a cohomology theory of cyclic type for any Hopf algebra
equipped with a modular pair in involution. This theory was introduced by Connes
and Moscovici in [9, 10, 11]. Their ﬁrst motivation came from the computation of
the index of transversally elliptic operators on foliations. For that they constructed,
31
for any n ≥ 1, a Hopf algebra H(n), and computed the Hopf cyclic cohomology of
H(n) [9]. They also developed a theory of characteristic classes for actions of Hopf
algebras on algebras. Hopf cyclic cohomology can in fact be regarded as the right
noncommutative analogue of both group and Lie algebra homology. Here is a quick
recall:
Let (H, ∆, ε, S) be a Hopf algebra. A character of H is a unital algebra map
δ : H → k. A grouplike element of H is a nonzero element σ in H such that
∆(σ) = σ ⊗ σ. Let δ be a character of H and σ a group like element in H. We say
(δ, σ) is a modular pair if δ(σ) = 1. We can deﬁne a δtwisted antipode by
¯
S = δ ∗ S,
i.e.
¯
S(h) = δ(h
(1)
)S(h
(2)
), ∀h ∈ H.
A modular pair (δ, σ) is called a modular pair in involution (MPI), [9, 10, 11] if
σ
−1
¯
S
2
σ = id, that is:
σ
−1
¯
S
2
(h)σ = h, ∀h ∈ H.
Example 2.6.1. If H is commutative or cocommutative then S
2
= id which implies
that (ε, 1) is an MPI for H. Conversely if (ε, 1) is an MPI for H, then S
2
= id. Note
that the condition S
2
= id does not imply that H is commutative or cocommutative.
Example 2.6.2. The quantum universal enveloping algebra U
q
(sl(2, k)) is a kHopf
algebra which is generated as a k algebra by symbols σ, σ
−1
, x, y subject to the
following relations
σσ
−1
= σ
−1
σ = 1, σx = q
2
xσ, σy = q
−2
yσ, xy −yx =
σ −σ
−1
q −q
−1
.
The coproduct, counit and antipode of U
q
(sl(2, k)) are deﬁned by:
∆(x) = x ⊗σ + 1 ⊗x, ∆(y) = y ⊗1 + σ
−1
⊗y, ∆(σ) = σ ⊗σ,
S(σ) = σ
−1
, S(x) = −xσ
−1
, S(y) = −σy,
32
(σ) = 1, (x) = (y) = 0.
It is easy to check that S
2
(h) = σhσ
−1
for all h in H. Therefore (ε, σ
−1
) is a modular
pair in involution for U
q
(sl(2, k)).
Example 2.6.3. Let k be a ﬁeld of characteristic zero and q ∈ k, q ,= 0 and q not a
root of unity. The Hopf algebra H = A(SL
q
(2, k)) is deﬁned as follows [24]. As an
algebra it is generated by symbols a, b, c, d, with the following relations:
ab = qba, ac = qca, ad −da = (q −q
−1
)bc,
bc = cb, bd = qdb, cd = qdc,
ad −qbc = da −q
−1
bc = 1.
Comultlipication, counit, and antipode on H are deﬁned by:
∆(a) = a ⊗a + b ⊗c, ∆(b) = a ⊗b + b ⊗d,
∆(c) = c ⊗a + d ⊗c, ∆(d) = c ⊗b + d ⊗d,
ε(a) = ε(d) = 1, ε(b) = ε(c) = 0,
S(a) = d, S(b) = −q
−1
b,
S(c) = −qc, S(d) = a.
Note that S
2
,= id. A modular pair (δ, σ) for H is deﬁned by:
δ(a) = q, δ(b) = δ(c) = 0, δ(d) = q
−1
,
and σ = 1. Then one can check that
¯
S
2
= id, and hence (δ, 1) is modular pair in
involution for H.
33
Theorem 2.6.1. [9, 10, 11] Let (H, ∆, ε, S) be a Hopf algebra endowed with a modu
lar pair in involution (δ, σ). Set C
n
(H) = H
⊗n
, n ≥ 0, and deﬁne faces, degeneracies,
and cyclic maps by:
δ
i
(h
1
, ..., h
n−1
) =
_
¸
¸
¸
¸
¸
_
¸
¸
¸
¸
¸
_
(1, h
1
, ..., h
n−1
) i = 0
(h
1
, ..., ∆h
i
, ..., h
n−1
) 1 ≤ i ≤ n −1
(h
1
, ..., h
n−1
, σ) i = n
σ
i
(h
1
, ..., h
n+1
) = ε(h
i+1
)(h
1
, ..., h
i
, h
i+2
, ..., h
n+1
), 0 ≤ i ≤ n
τ
n
(h
1
, ..., h
n
) = ∆
n−1
¯
S(h
1
) (h
2
, ..., h
n
, σ). (2.6.11)
Then (C
∗
(H), δ
i
, σ
i
, τ) is a cocyclic module.
The action in Formula (2.6.11) is the diagonal action of H
⊗n
on itself deﬁned
by:
(g
1
, g
2
, ..., g
n
).(h
1
, h
2
, ..., h
n
) := (g
1
h
1
, g
2
h
2
, ..., g
n
h
n
).
Therefore
∆
n−1
¯
S(h
1
) (h
2
, ..., h
n
, σ) =
_
S(h
(n)
1
)h
2
, ..., S(h
(2)
1
)h
n
,
¯
S(h
(1)
1
)σ
_
Deﬁnition 2.6.1. The cohomology of the cocyclic module deﬁned in Theorem (2.6.1)
is called the Hopf cyclic cohomology of H and is denoted by HC
∗
(δ,σ)
(H), or sometimes
just HC
∗
(H), if there is no confusion.
Example 2.6.4. [9, 10, 11, 12] Let G be a discrete group and CG be its group algebra
(cf. Example 1.1.1). We recall from Example (2.6.1) that (ε, 1) is an MPI for CG.
34
It can be shown that:
HC
n
(CG) =
_
¸
¸
_
¸
¸
_
C if n = even
0 if n = odd
This implies that:
HP
0
(CG) = C, HP
1
(CG) = 0.
Example 2.6.5. [9, 10, 11] Let g be a Lie algebra and U(g) be its universal enveloping
algebra (cf. Example 1.1.2). For any character δ : U(g) →C, one can show that (δ, 1)
is an MPI for U(g), and we have:
HP
0
(U(g)) =
i≥0
H
Lie
2i
(g, C
δ
), HP
1
(U(g)) =
i≥0
H
Lie
2i+1
(g, C
δ
).
Proof. [9, 10, 11, 12] One shows that the antisymmetrization map
A :
n
g →U(g)
⊗n
,
A(x
1
∧ ... ∧ x
n
) = (
σ∈S
n
sign(σ)(x
σ(1)
, ..., x
σ(n)
))/n!,
35
induces an quasiisomorphism between the following two bicomplexes;
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
_
4
g
d
Lie
−−−→
_
3
g
d
Lie
−−−→
_
2
g
d
Lie
−−−→
_
1
g
d
Lie
−−−→ C
0
¸
¸
¸ 0
¸
¸
¸ b
¸
¸
¸ 0
¸
¸
¸
_
3
g
d
Lie
−−−→
_
2
g
d
Lie
−−−→
_
1
g
d
Lie
−−−→ C
0
¸
¸
¸ 0
¸
¸
¸ 0
¸
¸
¸
_
2
g
d
Lie
−−−→
_
1
g
d
Lie
−−−→ C
0
¸
¸
¸ 0
¸
¸
¸
_
1
g
d
Lie
−−−→ C
0
¸
¸
¸
C
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
U(g)
⊗4
B
−−−→ U(g)
⊗3
B
−−−→ U(g)
⊗2
B
−−−→ U(g)
B
−−−→ C
b
¸
¸
¸ b
¸
¸
¸ b
¸
¸
¸ b
¸
¸
¸
U(g)
⊗3
B
−−−→ U(g)
⊗2
B
−−−→ U(g)
B
−−−→ C
b
¸
¸
¸ b
¸
¸
¸ b
¸
¸
¸
U(g)
⊗2
B
−−−→ U(g)
B
−−−→ C
b
¸
¸
¸ b
¸
¸
¸
U(g)
B
−−−→ C
b
¸
¸
¸
C
The rows in the ﬁrst bicomplex are the ChevalleyEilenberg complex
C
g
δ
¸
_
2
g
d
Lie
¸
_
3
g
d
Lie
¸ ...
d
Lie
¸
36
d
Lie
(x
1
∧ ... ∧ x
n
) = (
n
i=1
(−1)
i+1
δ(x
i
)x
1
∧ ... ∧ ´ x
i
∧ ... ∧ x
n
)
+ (
i<j
(−1)
i+j
[x
i
, x
j
] ∧ x
1
∧ ... ∧ ´ x
i
∧ ... ∧ ´ x
j
∧ ... ∧ x
n
),
which computes the Lie algebra homology H
Lie
∗
(g, C
δ
) of the Lie algebra g with
coeﬃcients in C
δ
.
Example 2.6.6. For any Hopf algebra H equipped with an MPI (δ, σ), consider the
(b, B)bicomplex B(H)
.
.
.
.
.
.
.
.
.
H
⊗2
B
−−−→ H
B
−−−→ k
b
¸
¸
¸ b
¸
¸
¸
H
B
−−−→ k
b
¸
¸
¸
k
The total comples TotB(H) is:
k
b
−−−→ H
B+b
−−−→ k ⊕H
⊗2
B+b
−−−→ ,
where, for all r in k and h in H,
b(r) = r(1 −σ), b(h) = 1 ⊗h −∆(h) + h ⊗σ, B(h) = δ(h) + ε(h).
Therefore we have:
HC
0
(H) = Ker(b) =
_
¸
¸
_
¸
¸
_
k if σ = 1
0 if σ ,= 1
37
We also have:
HC
1
(H) =
Ker(B) ∩ Ker(b)
Im(b)
.
In the special case when σ = 1 we will have:
HC
1
(H) = Ker(B) ∩ Ker(b) = Ker(δ + ε) ∩ P(H),
where P(H) is the space of primitive elemnts of H, i.e. those elements h in H such
that ∆(h) = 1 ⊗h + h ⊗1. In the very special case when δ = ε and σ = 1, we have:
HC
1
(H) = Ker(ε) ∩ P(H) = P(H).
2.7 Dual Hopf cyclic cohomology
In [25] a dual Hopf cyclic theory for Hopf algebras is introduced. Their motivations
include the study of the coactions of Hopf algebras, and the fact that for group
algebras and in general for Hopf algebras with a normalized Haar integral the Hopf
cyclic cohomology is trivial [12]. They deﬁne a cyclic module for Hopf algebras wich
is a dual of the cocyclic module introduced in [9, 10, 11] by Connes and Moscovici
(cf. Section 2.6). In [25] they compute their dual theory for group algebras and some
quantum groups. In this subsection we, shortly, review this dual theory.
Let (H, m, η, ∆, ε, S) be a Hopf algebra over a commutative ring k, σ be a
nonzero grouplike element of H and δ : H −→k be a character for H. The pair (δ, σ)
is called a modular pair if δ(σ) = 1, and a modular pair in involution, in the dual
sense, if
¯
S
2
σ
= id, (2.7.12)
where
¯
S
σ
(h) = σ
(h)
δ(h
(2)
)S(h
(1)
). (2.7.13)
38
One can associate a cyclic module to (H, δ, σ) which is a dual of the Connes
Moscovici’s cocyclic module. First, one deﬁnes S
σ
by S
σ
(h) = σS(h), and proves the
following properties:
S
σ
(h
1
h
2
) = S
σ
(h
2
)S(h
1
),
S
σ
(1) = σ,
∆S
σ
(h) =
(h)
S
σ
(h
(2)
) ⊗S
σ
(h
(1)
),
ε(S
σ
(h)) = ε(h).
Next, using ε and δ one can endow k with an Hbimodule structure via:
H ⊗k −→k hr = δ(h)r,
and
k ⊗H −→k, rh = rε(h),
For all h in H and r in k. The desired cyclic module as a simplicial module turnes
out to be exactly the Hochschild complex of the algebra H with coeﬃcients in the
Hbimodule k.
This cyclic module is deﬁned as follows. The complex is
¯
C
(δ,σ)
n
(H) = H
⊗n
, n ≥
0, with faces, degeneracies, and cyclic maps given by:
δ
0
(h
1
, h
2
, ..., h
n
) = ε(h
1
)(h
2
, h
3
, ..., h
n
),
δ
i
(h
1
, h
2
, ..., h
n
) = (h
1
, h
2
, ..., h
i
h
i+1
, ..., h
n
), 1 ≤ i ≤ n −1,
δ
n
(h
1
, h
2
, ..., h
n
) = δ(h
n
)(h
1
, h
2
, ..., h
n−1
),
σ
0
(h
1
, h
2
, ..., h
n
) = (1, h
1
, ..., h
n
),
σ
i
(h
1
, h
2
, ..., h
n
) = (h
1
, h
2
, ..., h
i
, 1, h
i+1
, ..., h
n
), 1 ≤ i ≤ n −1,
σ
n
(h
1
, h
2
, ..., h
n
) = (h
1
, h
2
, ..., h
n
, 1),
39
τ
n
(h
1
, h
2
, ..., h
n
) = δ(h
(2)
n
)
_
S
σ
(h
(1)
1
h
(1)
2
...h
(1)
n
), h
(2)
1
, ..., h
(2)
n−1
_
.
The cyclic homology obtained from this cyclic module is denoted by
¯
HC
(δ,σ)
∗
(H), or
in the perodic case by
¯
HP
(δ,σ)
∗
(H).
Example 2.7.1. (Compare with Example 2.6.4) Let G be a discrete group and kG
be its group algebra over k. . It is obvious that kG
⊗n
can be identiﬁed with kG
n
, the
free k module generated by G
n
. One can check that
δ
i
(g
1
, ..., g
n
) =
_
¸
¸
¸
_
¸
¸
¸
_
(g
2
, ..., g
n
) if i = 0
(g
1
, ..., g
i
g
i+1
, ...g
n
) 1 ≤ i < n
(g
1
, ..., g
n−1
) if i = n
σ
i
(g
1
, ..., g
n
) =
_
¸
¸
¸
_
¸
¸
¸
_
(1, g
1
, ..., g
n
) if i = 0
(g
1
, ..., g
i
, 1, g
i+1
, ...g
n
) 1 ≤ i ≤ n −1
(g
1
, ..., g
n−1
, g
n
, 1) if i = n
τ(g
1
, g
2
, ..., g
n
) = ((g
1
g
2
...g
n
)
−1
, g
1
, ..., g
n−1
),
and
¯
HP
(,1)
n
(kG) =
_
¸
¸
_
¸
¸
_
i≥0
H
2i
(G; k) n = 0
i≥0
H
2i+1
(G; k) n = 1
Example 2.7.2. (Compare with Example 2.6.5) Let g be a Lie algebra over k and
U(g) be its enveloping algebra. For any group like element σ the pair (ε, σ) is an MPI
for U(g). One can prove that:
¯
HC
(,σ)
n
(U(g)) =
k≥0
H
Lie
n−2k
(g; k).
40
Example 2.7.3. If H is a cocommutative Hopf algebra then,
¯
HC
(,1)
n
(H) =
k≥0
H
n−2k
(H, k),
where in the right hand side is the Hochschild homology of H with trivial coeﬃcients
via ε for both left and right action of H
Example 2.7.4. Let V be a kmodule and T(V ) be its tensor algebra .Then T(V ) is
a cocommutative Hopf algebra. One knows that H
0
(T(V ), k) = k, H
1
(T(V ), k) = V
and the other homology groups are zero. Therefore
¯
HC
(,1)
n
(T(V )) = k if n is even
¯
HC
(,1)
n
(T(V )) = V if n is odd.
2.8 YetterDrinfeld modules
One of the motivations to deﬁne the YetterDrinfeld modules was to deﬁne a braiding
on the monoidal category HMod of all Hmodules where H is a not necessarily
commutative or cocommutative, Hopf algebra. To deﬁne such a braiding one should
either restrict to special classes of Hopf algebras called quasitriangular Hopf algebras,
or, to special classes of modules known as YetterDrinfeld modules (cf. also Example
3.2.8).
Deﬁnition 2.8.1. [33, 40] Let H be a Hopf algebra with a bijective antipode S. A left
left YetterDrinfeld Hmodule consists of a vector space M, a left Hmodule structure
on M :
H ⊗M →M, h ⊗m →hm,
41
and a left Hcomodule structure on M:
M →H ⊗M, m →m
(−1)
⊗m
(0)
.
The left action and coaction are supposed to satisfy the YetterDrinfeld compatibility
condition:
(hm)
(−1)
⊗(hm)
(0)
= h
(1)
m
(−1)
S(h
(3)
) ⊗h
(2)
m
(0)
,
for all h ∈ H and m ∈ M.
Deﬁnition 2.8.2. A rightleft YetterDrinfeld Hmodule is a right Hmodule and left
Hcomodule M such that
(mh)
(−1)
⊗(mh)
(0)
= S
−1
(h
(3)
)m
(−1)
h
(1)
⊗m
(0)
h
(2)
,
for all h ∈ H and m ∈ M.
There are of course analogous deﬁnitions for leftright, and rightright Yetter
Drinfeld modules.
The category
H
H
¸T of all leftleft Yetterdrinfeld Hmodules is the center of
the monoidal category H − Mod. Recall that the (left) center Zc of a monoidal
category is a category whose objects are pairs (X, σ
X,−
), where X is an object of c
and σ
X,−
: X⊗− →−⊗X is a natural isomorphism satisfying certain compatibility
conditions with the associativity and unit constraints of c. It can be shown that the
center of a monoidal category is a braided monoidal category, and: [22]
Z(H −Mod) =
H
H
¸T.
Example 2.8.1. Let H = kG be the group algebra of a discrete group G. A left
42
kGcomodule is simply a Ggraded vector space
M =
g∈G
M
g
where the coaction M → kG ⊗ M is deﬁned by m → g ⊗ m, for all m in M
g
. An
action kG⊗M →M, (g, m) →gm, deﬁnes a YetterDrinfeld module structure on
M iﬀ for all g, h ∈ G and m ∈ M
g
, hm ∈ M
hgh
−1
.
2.9 Stable anti YetterDrinfeld modules
Stable anti YetterDrinfeld modules were introduced in [18]. Motivation was to ﬁnd
a right deﬁnition for Hopf cyclic cohomology with coeﬃcients [19] (cf. Section 2.10).
This class of Hmodules is the right choice for the coeﬃcients in Hopf cyclic coho
mology. It turns out that one dimensional stable anti YetterDrinfeld modules are
exactly modular pairs in involution.
Deﬁnition 2.9.1. A leftleft antiYetterDrinfeld Hmodule is a left Hmodule and
left Hcomodule M such that
(hm)
(−1)
⊗(hm)
(0)
= h
(1)
m
(−1)
S
−1
(h
(3)
) ⊗h
(2)
m
(0)
, (2.9.14)
for all h ∈ H and m ∈ M. We say that M is stable if in addition we have
m
(−1)
m
(0)
= m,
for all m ∈ M.
Deﬁnition 2.9.2. A rightleft antiYetterDrinfeld Hmodule is a right Hmodule
and left Hcomodule M such that
(mh)
(−1)
⊗(mh)
(0)
= S(h
(3)
)m
(−1)
h
(1)
⊗m
(0)
h
(2)
, (2.9.15)
43
for all h ∈ H and m ∈ M. We say that M is stable if in addition we have
m
(0)
m
(−1)
= m,
for all m ∈ M.
Remark 2.9.1. Notice that if one changes S
−1
to S in the formula (2.9.14), or, S
to S
−1
in (2.9.15), one gets the relations for leftleft and rightleft YetterDrinfeld
modules as in Deﬁnitions (2.8.1) amd (2.8.2). Therefore if S
2
= id the the notions
of YetterDrinfeld and anti YetterDrinfeld modules coincide.
There are of course analogous deﬁnitions for leftright and rightright stable
antiYetterDrinfeld (SAYD) modules [18].
The following lemma shows that 1dimensional rightleft SAYD modules corre
spond to ConnesMoscovici’s modular pairs in involution:
Lemma 2.9.1. [18] There is a oneone correspondence between modular pairs in
involution (MPI) (δ, σ) on H and rightleft SAYD module structure on M = k, deﬁned
by
rh = δ(h)r, r →σ ⊗r,
for all h ∈ H and r ∈ k. We denote this module by M =
σ
k
δ
.
Technically, we want to show that the two conditions
¯
S
2
(h) = σhσ
−1
, ∀h ∈ H, (2.9.16)
and
(rh)
(−1)
⊗(rh)
(0)
= S(h
(3)
)r
(−1)
h
(1)
⊗r
(0)
h
(2)
, ∀h ∈ H, r ∈ k, (2.9.17)
are equivalent.
44
Proof. (2.9.16) =⇒(2.9.17): we have
δ(h)1 =
¯
S(δ(h)) =
¯
S(
¯
S(h
(1)
)h
(2)
)
=
¯
S(h
(2)
)
¯
S
2
(h
(1)
) = δ(h
(2)
)S(h
(3)
)
¯
S
2
(h
(1)
)
(2.9.16)
= δ(h
(2)
)S(h
(3)
)σh
(1)
σ
−1
⇐⇒
rδ(h)σ = rδ(h
(2)
)S(h
(3)
)σh
(1)
⇐⇒
(rh)
(−1)
⊗(rh)
(0)
= S(h
(3)
)r
(−1)
h
(1)
⊗r
(0)
h
(2)
.
(2.9.17) =⇒ (2.9.16):
¯
S
2
(h) =
¯
S(
¯
S(h))) =
¯
S(δ(h
(1)
)S(h
(2)
))
= δ(h
(1)
)
¯
S(S(h
(2)
)) = δ(h
(1)
)δ((S(h
(2)
))
(1)
)S((S(h
(2)
))
(2)
)
= δ(h
(1)
)δ(S(h
(2)(2)
))S(S(h
(2)(1)
))
= δ(h
(1)
)δ(S(h
(3)
))S
2
(h
(2)
) = δ(S(h
(3)
))S
2
(h
(2)
)δ(h
(1)
)
(2.9.17)
= δ(S(h
(3)
))S
2
(h
(2)
)S(h
(1)(3)
)σh
(1)(1)
σ
−1
δ(h
(1)(2)
)
= δ(S(h
(5)
))S
2
(h
(4)
)S(h
(3)
)σh
(1)
σ
−1
δ(h
(2)
)
= δ(S(h
(5)
))S(h
(3)
S(h
(4)
))σh
(1)
σ
−1
δ(h
(2)
)
= δ(S(h
(4)
))ε(h
(3)
)σh
(1)
σ
−1
δ(h
(2)
)
= δ(S(h
(3)
))σh
(1)
σ
−1
δ(h
(2)
)
= δ(h
(2)
S(h
(3)
))σh
(1)
σ
−1
= ε(h
(2)
)σh
(1)
σ
−1
= σhσ
−1
.
45
Example 2.9.1. The above lemma gives all onedimensional rightleft SAYD H
modules as modular pairs in involution.
Example 2.9.2. In the special case where S
2
= id (e.g., when H is (co)commutative),
(ε, 1) is an MPI.
2.10 Hopf cyclic cohomology with coeﬃcients
Hopf cyclic cohomology was generalized to Hopf cyclic cohomology with coeﬃcients in
[18, 19] by introducing the notion of stable anti YetterDrinfeld modules (SAYD). They
assigned a cocyclic module to a triple (H, C, M) where C is a Hmodule coalgebra
and M is a SAYD Hmodule:
Theorem 2.10.1. [18, 19] Let H be a Hopf algebra, C an Hmodule coalgebra and
M a rightleft SAYD Hmodule. Then the complex c
n
H
(C, M) := M⊗
H
C
⊗(n+1)
with
the following faces, degeneracies and cyclic maps, deﬁnes a cocyclic module attached
to the triple (H, C, M):
δ
i
(m⊗
H
c
0
⊗ ⊗c
n−1
) =
_
¸
¸
_
¸
¸
_
m⊗
H
c
0
⊗ ⊗c
(1)
i
⊗c
(2)
i
⊗c
n−1
0 ≤ i < n
m
(0)
⊗
H
c
(2)
0
⊗c
1
⊗ ⊗c
n−1
⊗m
(−1)
c
(1)
0
i = n
σ
i
(m⊗
H
c
0
⊗ ⊗c
n+1
) = m⊗
H
c
0
⊗ ⊗ε(c
i+1
) ⊗ ⊗c
n+1
, 0 ≤ i ≤ n
τ
n
(m⊗
H
c
0
⊗ ⊗c
n
) = m
(0)
⊗
H
c
1
⊗ ⊗c
n
⊗m
(−1)
c
0
.
46
Example 2.10.1. Consider C = H as a Hmodule coalgebra via m
H
, and M =
σ
k
δ
as in Lemma (2.9.1). Then the above cocyclic module reduces to the cocyclic module
of Connes and Moscovici mentioned in Theorem (2.6.1).
Proof. First we notice that the map f : H
⊗(n+1)
→H
⊗(n+1)
, deﬁned by
f(h
0
⊗ ⊗h
n
) := h
(1)
0
⊗S(h
(2)
0
)(h
1
⊗ ⊗h
n
)
= h
(1)
0
⊗S(h
(2)
0
)
(1)
h
1
⊗ ⊗S(h
(2)
0
)
(n)
h
n
,
= h
(1)
0
⊗S(h
(n+1)
0
)h
1
⊗ ⊗S(h
(2)
0
)h
n
,
deﬁnes an Hmodule isomorphism, where H
⊗(n+1)
, on the left, is considered as an H
module via diagonal action and H
⊗(n+1)
, on the right, is considered as an Hmodule
via multiplication by the ﬁrst term. The inverse of f is deﬁned by
f
−1
(h
0
⊗ ⊗h
n
) := h
(1)
0
⊗h
(2)
0
(h
1
⊗ ⊗h
n
)
= h
(1)
0
⊗h
(2)(1)
0
h
1
⊗ ⊗h
(2)(n)
0
h
n
,
= h
(1)
0
⊗h
(2)
0
h
1
⊗ ⊗h
(n+1)
0
h
n
.
For any M, the map f induces the isomorphism
¯
f : M ⊗
H
H
⊗(n+1)
∼
= M ⊗
H
H
⊗(n+1)
= (M ⊗
H
H) ⊗H
⊗(n)
.
If we combine this map with the natural isomorphism M ⊗
H
H
∼
= M, we have the
isomorphism
φ : M ⊗
H
H
⊗(n+1)
∼
= M ⊗H
⊗(n)
,
φ(m⊗
H
h
0
⊗ ⊗h
n
) = mh
(1)
0
⊗S(h
(n+1)
0
)h
1
⊗ ⊗S(h
(2)
0
)h
n
. (2.10.18)
47
It is also easy to see that φ
−1
: M ⊗ H
⊗(n)
∼
= M ⊗
H
H
⊗(n+1)
, composed of the
natural isomorphism M
∼
= M ⊗
H
H, and f
−1
, is deﬁned by:
φ
−1
(m⊗h
1
⊗h
n
) = m⊗
H
1 ⊗h
1
⊗ ⊗h
n
.
Now if M =
σ
k
δ
then φ : k ⊗
H
H
⊗(n+1)
∼
= k ⊗H
⊗(n)
∼
= H
⊗(n)
is deﬁned by
φ(r ⊗
H
h
0
⊗ ⊗h
n
) = rδ(h
(1)
0
)S(h
(n+1)
0
)h
1
⊗ ⊗S(h
(2)
0
)h
n
= rS(h
(n)
0
)h
1
⊗ ⊗
¯
S(h
(1)
0
)h
n
= r
¯
S(h
0
)(h
1
⊗ ⊗h
n
), (2.10.19)
and
φ
−1
: H
⊗(n)
∼
= k ⊗
H
H
⊗(n+1)
,
φ
−1
(h
1
⊗h
n
) = 1
k
⊗
H
1 ⊗h
1
⊗ ⊗h
n
.
Next it is easy to check that the map φ in (2.10.19), is an isomorphism of cocyclic
modules between the cocyclic module c
n
H
(H, k) := k ⊗
H
H
⊗(n+1)
deﬁned in Theo
rem (2.10.1), for M = k and C = H, and the ConnesMoscovici’s cocyclic module
C
n
(H) = H
⊗n
, n ≥ 0, deﬁned in Theorem (2.6.1).
Example 2.10.2. More generally, for any SAYD Hmodule M and for C = H, the
isomorphism φ in (2.10.18), reduces the cocyclic module deﬁned in Theorem (2.10.1)
to the following simpler one. The complex c
n
H
(H, M) := M ⊗ H
⊗n
with the faces,
degeneracies and cyclic maps as follows:
δ
i
(m⊗h
1
⊗ ⊗h
n−1
) =
_
¸
¸
¸
¸
¸
_
¸
¸
¸
¸
¸
_
m⊗1 ⊗h
1
⊗ ⊗h
n−1
i = 0
m⊗h
1
⊗ ⊗h
(1)
i
⊗h
(2)
i
⊗h
n−1
1 ≤ i < n
m
(0)
⊗h
1
⊗ ⊗h
n−1
⊗m
(−1)
i = n
48
σ
i
(m⊗h
1
⊗ ⊗h
n+1
) = m⊗h
1
⊗ ⊗ε(h
i+1
) ⊗ ⊗h
n+1
, 0 ≤ i ≤ n
τ
n
(m⊗h
1
⊗ ⊗h
n
) = m
(0)
h
(1)
1
⊗S(h
(2)
1
)(h
2
⊗ ⊗h
n
⊗m
(−1)
).
Deﬁnition 2.10.1. The cyclic cohomology of this cocyclic module is by deﬁnition the
Hopf cyclic cohomology of H with coeﬃcients in M.
Chapter 3
Braided Monoidal Categories
In this chapter we recall basic notions of braided monoidal categories and review the
notions of algebras, coalgebras, bialgebras, and Hopf algebras in those categories.
We also recall modules and comodules over a Hopf algebra in a braided monoidal
category. We provide some examples in each case.
3.1 Braided monoidal categories
Braided monoidal categories are the proper underlying context to deﬁne notions like
braided algebra, coalgebra and Hopf algebra and many more (cf. Section 3.3). When
it comes to practicing the homological algebra, one also needs the braided monoidal
category to be abelian (cf. Chapter 4). In this section we give a quick review of the
concept of braided monoidal category. In next section we provide some examples.
Basic references are [20, 21, 22, 29, 30, 31].
Deﬁnition 3.1.1. A monoidal, or tensor, category (c, ⊗, I, a, l, r) consists of a cat
egory c, a functor ⊗ : c c →c (called tensor product), an object I ∈ c (called unit
object), and three natural isomorphisms, deﬁned for all objects A, B, C, of c,
a = a
A,B,C
: A ⊗(B ⊗C) →(A ⊗B) ⊗C,
l = l
A
: I ⊗A →A, r = r
A
: A ⊗I →A,
49
50
called the associativity isomorphism, the left unit isomorphism, and the right unit
isomorphism, respectively, such that the following pentagon and triangle diagrams
commute [29, 31]:
((A ⊗B) ⊗C) ⊗D
.i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
U
U
U
U
U
U
U
U
U
U
U
U
U
U
U
U
(A ⊗(B ⊗C)) ⊗D
(A ⊗B) ⊗(C ⊗D)
A ⊗((B ⊗C) ⊗D)
A ⊗(B ⊗(C ⊗D))
(A ⊗I) ⊗B
N
N
N
N
N
N
N
N
N
N
N
A ⊗(I ⊗B)
p
p
p
p
p
p
p
p
p
p
p
A ⊗B
The coherence theorem of Mac Lane [29] asserts that all diagrams formed by
a, l, r by tensoring and composing, commute. More precisely, it asserts that any two
natural transformations deﬁned by a, l, r between any two functors deﬁned by ⊗ and
I are equal.
Deﬁnition 3.1.2. A braided monoidal category is a monoidal category c endowed
with a natural family of isomorphisms
ψ
A,B
: A ⊗B →B ⊗A,
called braiding such that for all objects A, B, C of c the following hexagon diagrams
commute :
A ⊗(B ⊗C)
ψ
(B ⊗C) ⊗A
a
−1
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
(A ⊗B) ⊗C
a
−1
m
m
m
m
m
m
m
m
m
m
m
m
ψ⊗id
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
B ⊗(C ⊗A)
(B ⊗A) ⊗C
a
−1
B ⊗(A ⊗C)
id⊗ψ
m
m
m
m
m
m
m
m
m
m
m
m
51
(A ⊗B) ⊗C
ψ
C ⊗(A ⊗B)
a
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
A ⊗(B ⊗C)
a
m
m
m
m
m
m
m
m
m
m
m
m
id⊗ψ
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
(C ⊗A) ⊗B
A ⊗(C ⊗B)
a
(A ⊗C) ⊗B
ψ⊗id
m
m
m
m
m
m
m
m
m
m
m
m
Deﬁnition 3.1.3. A braiding is called a symmetry if we have
ψ
A,B
◦ ψ
B,A
= id
B⊗A
,
for all objects A and B of c. Sometimes we just write ψ
2
= id to signify the symmetry
condition. A symmetric monoidal category is a monoidal category endowed with a
symmetry.
Deﬁnition 3.1.4. A monoidal category c is called strict if its associativity and unit
isomorphisms, a, l, r, are equalities, i.e., for all objects A, B, C of c,
(A ⊗B) ⊗C = A ⊗(B ⊗C),
I ⊗A = A ⊗I = A.
By a theorem of Mac Lane [29] (cf. also [31]), any braided monoidal abelian
category is monoidal equivalent to a braided strict monoidal abelian category in which
the hexagon commuting diagrams, in Deﬁnition (3.1.2), are reduced to the following
equalities:
ψ
A,B⊗C
= (id
B
⊗ψ
A,C
)(ψ
A,B
⊗id
C
),
ψ
A⊗B,C
= (ψ
A,C
⊗id
B
)(id
A
⊗ψ
B,C
),
for all objects A, B, C of c.
Here we, roughly, review the procedure of strictiﬁcation by recalling the notion
of monoidal equivalence. We show how one can go back and forth between a given
52
nonstrict category, c, and its strictiﬁed version, c
st
. This guarantees that one always
can work in a strict braided monoidal category without losing any information [22,
29, 30, 31].
Let us ﬁrst recall that a monoidal functor between two monoidal category
(c, ⊗, I, a, l, r) and (c
/
, ⊗
/
, I
/
, a
/
, l
/
, r
/
) consists of following data [29] :
(i) A functor F : c →c
/
,
(ii)A morphism F
tensor
(x, y) : F(x) ⊗
/
F(y) →F(x ⊗y) in c
/
, for any two objects x
and y in c,
(iii) A morphism F
unit
: I
/
→F(I) in c
/
,
all subject to certain commutative diagrams, which guarantee that the functor F pre
serves the monoidal structure. A monoidal functor (F, F
tensor
, F
unit
), from c to c
/
,
is called a monoidal equivalence between c and c
/
, if there exist a monoidal functor
(G, G
tensor
, G
unit
) : c
/
→c,
such that FG and GF are naturally isomorphic to the identity. In such a case, we
say, c and c
/
are monoidal equivalent or gauge equivalent.
Starting from a braided monoidal (abelian) category (c, ⊗, I, a, l, r, ψ) one can
produce a strict braided monoidal (abelian) category (c
st
, ⊗
st
, I
st
, ψ
st
) monoidal
equivalent to c [29]. c
st
is called the strictiﬁcation of c.
Roughly speaking, objects of c
st
are all ﬁnite strings of objects of c shown by
[A
1
, ..., A
n
], where n = 1, 2, ..., and if n > 1 then A
i
,= I for all 1 ≤ i ≤ n. We shall
denote [A
1
, ..., A
n
] by A
st
. Let us deﬁne a map
LR : c
st
→c
LR[A
1
, ..., A
n
] := (A
1
⊗(A
2
⊗(A
3
...(A
n−1
⊗A
n
)))...).
The object (A
1
⊗(A
2
⊗(A
3
...(A
n−1
⊗A
n
)))...))) is called the left to right representation
53
of the string [A
1
, ..., A
n
] in c.
The morphisms between two strings A
st
= [A
1
, ..., A
n
] and B
st
= [B
1
, ..., B
m
]
of c
st
are deﬁned to be the morphisms between LR[A
1
, ..., A
n
] and LR[B
1
, ..., B
m
] in
c, with compositions just as in c.
The tensor product ⊗
st
in c
st
is deﬁned as follows.
[I] ⊗
st
A
st
= A
st
⊗
st
[I] = A
st
, ∀ A
st
∈ c
st
,
and for any other two strings [A
1
, ..., A
n
] and [B
1
, ..., B
m
] is just their concatenation
:
[A
1
, ..., A
n
] ⊗
st
[B
1
, ..., B
m
] := [A
1
, ..., A
n
, B
1
, ..., B
m
].
By above deﬁnition it is clear that [I] is the unit object of c
st
, i.e., I
st
:= [I].
Let f : [A
1
, ..., A
n
] → [B
1
, ..., B
m
] and g : [C
1
, ..., C
s
] → [D
1
, ..., D
t
] be two
morphisms in c
st
. As we mentioned above they are actually the morphisms f :
LR[A
1
, ..., A
n
] → LR[B
1
, ..., B
m
] and g : LR[C
1
, ..., C
s
] → LR[D
1
, ..., D
t
] in c. The
tensor product
f ⊗
st
g : [A
1
, ..., A
n
, C
1
, ..., C
s
] →[B
1
, ..., B
m
, D
1
, ..., D
t
],
is deﬁned by:
LR[A
1
, ..., A
n
, C
1
, ..., C
s
]
∼
= LR[A
1
, ..., A
n
] ⊗LR[C
1
, ..., C
s
]
f⊗g
→ LR[B
1
, ..., B
m
] ⊗LR[D
1
, ..., D
t
]
∼
= LR[B
1
, ..., B
m
, D
1
, ..., D
t
].
The ﬁrst and last isomorphisms in the above formula are the natural isomorphisms
in c generated by the associativity map a.
It is easy to see that c
st
is a strict braided monoidal category where a
st
, l
st
and
r
st
are equalities. Following the deﬁnition of morphisms in c
st
, for any two strings
54
A
st
, B
st
, the braiding map ψ
st
A
st
,B
st
in c
st
is deﬁned by:
ψ
st
A
st
,B
st
: LR[A
1
, ..., A
n
, B
1
, ..., B
m
]
∼
= LR[A
1
, ..., A
n
] ⊗LR[B
1
, ..., B
m
]
ψ
→LR[B
1
, ..., B
m
] ⊗[A
1
, ..., A
n
]
∼
= LR[B
1
, ..., B
m
, A
1
, ..., A
n
].
Again, the ﬁrst and last isomorphisms in the above formula are the natural isomor
phisms in c generated by the associativity map a. It can also be checked that if c is
an abelian category then c
st
is abelian too.
Now one can deﬁne a monoidal functor (F, F
tensor
, F
unit
) : c
st
→c, where the
functor F is given by
F(A
st
) = LR(A
st
), F(f) = f,
for all objects A
st
and morphisms f in c
st
. The morphism
F
tensor
(A
st
, B
st
) : F(A
st
) ⊗F(B
st
) →F(A
st
⊗
st
B
st
),
is the following natural isomorphisms in c generated by the associativity map a:
LR[A
1
, ..., A
n
] ⊗LR[B
1
, ..., B
m
]
∼
= LR[A
1
, ..., A
n
, B
1
, ..., B
m
],
and ﬁnally the morphism F
unit
: F[I] = I →I is just the identity, id
I
.
One can also check that (F, F
tensor
, F
unit
) is a monoidal equivalence between
c
st
and c with the inverse monoidal functor
(G, G
tensor
, G
unit
) : c →c
st
,
where the functor G is deﬁned by
G(A) = [A], G(f) = f,
55
for all objects A and morphisms f in c. The morphism G
tensor
(A, B) : G(A) ⊗
st
G(B) → G(A ⊗ B) is just the identity map id
A⊗B
: A ⊗ B → A ⊗ B, and G
unit
:
G(I) = [I] →[I] is also the identity, id
[I]
= id
I
.
Remark 3.1.1. We remark that if the original category is symmetric then its stric
tiﬁcation is symmetric as well. This plays an important role in our approach. In
fact, using this result, we can safely assume that our symmetric monoidal categories
are strict and symmetric. Working with strict categories drastically simpliﬁes the
formalism and that is what we shall do in this thesis.
3.2 Examples of braided monoidal categories
Example 3.2.1. The category V ect
k
of vector spaces over a ﬁeld k with the usual
tensor product ⊗
k
and with the braiding ψ = usual ﬂip, is a nonstrict symmetric
braided monoidal abelian category. More generaly for any unital ring R, the category
of (left) Rmodules is a symmetric monoidal abelian category.
Example 3.2.2. The category (Set, , I) with the Cartesian product as its tensor,
the one point set I = ¦1¦ as the unit, and with the usual ﬂip map deﬁned by, ψ
A,B
(a⊗
b) := b⊗a, for any two sets A and B, as the braiding map, is a nonabelian, nonstrict,
symmetric braided monoidal category.
Most of the categories that we mention or work with in this thesis are abelian.
Here we have to recall the notion of braid groups which we need to deﬁne the braid
category in the next example. The braid group B
n
, for n ≥ 3, is deﬁned to be the
group generated by n−1 generators b
i
, 1 ≤ i ≤ n−1, called braids, with relations:
b
i
b
j
= b
j
b
i
, [i −j[ ,= 1,
b
i
b
i+1
b
i
= b
i+1
b
i
b
i+1
, 1 ≤ i ≤ n −2.
56
The braid group B
2
is deﬁned to be the inﬁnite cyclic group generated by one gener
ator b
1
, and B
1
is deﬁned to be the identity group.
It is helpful and instructive to have the following intuition about the braids,
for a ﬁxed n. Suppose there are n identical strings hanging down from the ceiling
with their top ends all tightened up to an straight line (e.g., a bar) attached to the
ceiling. Let assign to those end points numbers 1 to n, in order. The braid b
i
can be
visualized by crossing the i
th
string over the (i + 1)
th
string, leaving all other strings
untouched, and keeping all the bottom end points ﬁxed. It is important to notice
that, the braid b
i
actually changes the position of the i
th
string with the position of
the (i + 1)
th
string. The braid b
−1
i
then would be produce in a similar way but by
crossing the (i + 1)
th
string over the i
th
string. It is again important to notice that,
the braid b
−1
i
, like b
i
, changes the position of the i
th
string with the position of the
(i + 1)
th
string, but in a reverse fashion. The identity braid id
n
is to simply leave all
the strings hanging down, untouched. The multiplication b
i
b
j
of two braids means to
do the procedure b
i
followed by b
j
, with the change of positions understood. [23, 29]
Example 3.2.3. (Braid category [23, 29])
The braid category B is deﬁned as follows. Objects of B are natural numbers, 0, 1, 2, ....
For any n ≥ 1 in B, Hom
B
(n, n) := B
n
. The only morphism from 0 to 0 is deﬁned to
be the identity, and is called the empty braid. There is no morphism between n and
m if n ,= m. Tensor product of two objects n and m is deﬁned to be n + m.
Having the above intuition in mind, n + m means to simply put the set of m
strings beside the set of n strings. Now it is easy to see that, any pairs of braids
f : n →n and g : m →m, uniquely deﬁne a braid called f ⊗g : n + m →n + m, by
just putting them side by side. One can also deﬁne the braiding map
ψ
n,m
: n + m →m + n,
by crossing the whole set of n strings over the set of m strings.
57
The braid category B deﬁned as above, is a braided monoidal category, with the
empty braid as its unit. It is clearly strict, but is not symmetric, i.e.,
ψ
m,n
ψ
n,m
,= id
n+m
.
Example 3.2.4. Let (H, R = R
1
⊗ R
2
) be a quasitriangular Hopf algebra and H
Mod be the category of all left Hmodules. Then HMod is a braided monoidal abelian
category. It is symmetric if and only if R
−1
= R
2
⊗R
1
[31]. In this latter case, (H, R)
is actually called a triangular Hopf algebra. The monoidal structure on HMod is
deﬁned by
h (v ⊗w) = h
(1)
v ⊗h
(2)
w,
and the braiding map ψ
V ⊗W
acts by
ψ
V ⊗W
(v ⊗w) := (R
2
w ⊗R
1
v),
for any V and W in c, where denotes the action of H.
Example 3.2.5. In a dual manner if we consider a coquasitriangular Hopf algebra
(H, R), then the category of left Hcomodules is a braided monoidal abelian category.
Example 3.2.6. As a very special case of Example (3.2.4), let H = CZ
2
with the
nontrivial quasitriangular structure R = R
1
⊗R
2
deﬁned by
R := (
1
2
)(1 ⊗1 + 1 ⊗g + g ⊗1 −g ⊗g),
where g is the generator of the cyclic group Z
2
. Then the category c = Z
2
 Mod is the
category of super vector spaces with even morphisms [31]. The braiding map ψ
V ⊗W
58
for any V = V
0
⊕V
1
and W = W
0
⊕W
1
in c acts as below:
ψ
V ⊗W
(v ⊗w) = (−1)
[v[[w[
(w ⊗v). (3.2.1)
This is a symmetric, braided, monoidal abelian category, with the unit C = C ⊕0.
There is also a category of diﬀerential graded (DG) super vector spaces whose
objects are super complexes V
0
d
d
V
1
, and its morphisms are even chain maps. It is
a braided monoidal category with the same braiding map as (3.2.1).
Example 3.2.7. One can extend Example (3.2.6) to CZ
n
for any n > 2 with non
trivial quasitriangular structure on CZ
n
given by:
R = (1/n)
n−1
a,b=0
e
(−2πiab)/n
g
a
⊗g
b
,
where g is the generator of Z
n
. Notice that the braided monoidal abelian category
CZ
n
Mod is not symmetric for any n > 2. Therefore, this provides a good source of
nonsymmetric braided monoidal categories[31].
Example 3.2.8. Let H be a Hopf algebra over a ﬁeld k with comultiplication ∆h =
h
(1)
⊗h
(2)
and a bijective antipode S. We recall from Deﬁnition (2.8.1), that a left
left YetterDrinfeld (YD) Hmodule consists of a vector space V , a left Hmodule
structure on V [33, 40]:
H ⊗V →V, h ⊗v →hv,
and a left Hcomodule structure on V:
V →H ⊗V, v →v
(−1)
⊗v
(0)
.
The left action and coaction are supposed to satisfy the YetterDrinfeld (YD) com
59
patibility condition:
(hv)
(−1)
⊗(hv)
(0)
= h
(1)
v
(−1)
S(h
(3)
) ⊗h
(2)
v
(0)
,
for all h ∈ H and v ∈ V . The category of all YD Hmodules is called the Yetter
Drinfeld category of H, and is usually denoted by
H
H
¸T. It is a braided monoidal
abelian category with the braiding map:
ψ
V ⊗W
(v ⊗w) = v
(−1)
w ⊗v
(0)
.
This category is in general not symmetric. In fact the inverse of the braiding is given
by :
ψ
−1
V ⊗W
(w ⊗v) = v
(0)
⊗S
−1
(v
(−1)
)w.
3.3 Braided algebras, coalgebras and Hopf
algebras
In this section we recall the notions of algebra, coalgebra, bialgebra and Hopf algebra
in a braided monoidal category. By a braided algebra we mean an algebra in a braided
monoidal category. We use the similar convention for braided coalgebras, bialgebras
and Hopf algebras. We recall some basic properties of braided Hopf algebras and
provide some important examples of them. We also give the deﬁnitions of Hmodules
and comodules for a Hopf algebra H in a braided monoidal category c. We ﬁx a
strict braided monoidal category (c, ⊗, I, ψ) throughout this section.
Deﬁnition 3.3.1. An algebra (H, m, η) in c consists of an object H ∈ objc and
morphisms m : H ⊗ H → H and η : I → H called multiplication and unit maps
satisfying the commutation relations:
60
H ⊗H
m
H
H ⊗H ⊗H
id⊗m
¸¸
m⊗id
H ⊗H
m
¸¸
associativity
H ⊗H
m
H H ⊗H
m
¸
I ⊗H = H
η⊗id
¸¸
id
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
H = H ⊗I
id⊗η
¸¸
id
¸¸S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
unit
Deﬁnition 3.3.2. A coalgebra in c is a triple (H, ∆, ε) consisting of an object H ∈
objc and morphisms ∆ : H → H ⊗ H and ε : H → I called comultiplication and
counit maps satisfying the commutation relations:
H
∆
∆
H ⊗H
id⊗∆
H ⊗H
∆⊗id
H ⊗H ⊗H
coassociativity
H ⊗H
ε⊗id
H
∆
∆
¸
id
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
id
.l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
H ⊗H
id⊗ε
I ⊗H = H H = H ⊗I
counit
Deﬁnition 3.3.3. A bialgebra (H, m, η, ∆, ε) in c is an algebra and a coalgebra si
multaneously, satisfying the compatibility conditions:
H ⊗H
∆m
∆⊗∆
H ⊗H
H ⊗H ⊗H ⊗H
id⊗ψ⊗id
H ⊗H ⊗H ⊗H
m⊗m
¸¸
61
H
∆
H ⊗H
I ⊗I = I
η
¸¸
η⊗η
k
k
k
k
k
k
k
k
k
k
k
k
k
k
k
H ⊗H
m
ε⊗ε
H
ε
.l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
I ⊗I = I
I
id
η
I
H
ε
o
o
o
o
o
o
o
o
o
o
o
o
o
o
Deﬁnition 3.3.4. A Hopf algebra (H, m, η, ∆, ε, S) in c consists of a bialgebra plus
a morphism S : H →H called the antipode map satisfying the relations:
m(S ⊗id)∆ = m(id ⊗S)∆ = ηε. (3.3.2)
The relations (3.3.2), in terms of commutative diagrams, are as follows:
H ⊗H
S⊗id
H ⊗H
m
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
H
ε
∆
¸¸
∆
I
η
H
H ⊗H
id⊗S
H ⊗H
m
m
m
m
m
m
m
m
m
m
m
m
m
m
m
m
The following proposition shows that the standard properties of Hopf algebras
hold for braided Hopf algebras.
Proposition 3.3.1. If (H, m, η, ∆, ε, S) is a braided Hopf algebra (in c), then:
Sm = mψ(S ⊗S) = m(S ⊗S)ψ, (3.3.3)
62
Sη = η,
∆S = ψ(S ⊗S)∆ = (S ⊗S)ψ∆, (3.3.4)
εS = ε.
Example 3.3.1. A Hopf algebra in the category V ect
k
is a Hopf algebra in the stan
dard sense.
Example 3.3.2. For any V in
H
H
¸T of Example (3.2.8), the tensor algebra T(V ) is
a braided Hopf algebra in
H
H
¸T. Its comultiplication, counit, and antipode are deﬁned
by ∆(v) = 1 ⊗v + v ⊗1, ε(v) = 0, and S(v) = −v, for all v in V .
Hopf algebras in
H
H
¸T are called YetterDrinfeld Hopf algebras. They are stud
ied in [36].
We recall from Example (3.2.6) that a super vector space V = V
0
⊕ V
1
is an
object of Z
2
Mod. The degree of a homogeneous element a in V will be denoted by
[a[.
Example 3.3.3. A super Hopf algebra is a Hopf algebra H in Z
2
Mod. Thus a super
Hopf algebra is a super vector space H = H
0
⊕ H
1
which is simultaneously a super
algebra and a super coalgebra, the two structures are compatible, and H also has a
degree preserving antipode map. More precisely:
• H is a Z
2
graded algebra (or a super algebra), i.e.,
[ab[ = [a[ +[b[,
for all homogeneous elements a and b of H.
• H is a super coalgebra, i.e.,
[a[ = [a
(1)
[ +[a
(2)
[,
63
for all homogeneous elements a of H.
• H is a super bialgebra, i.e., we have the following compatibility condition between
the above two structures:
∆(ab) = (ab)
(1)
⊗(ab)
(2)
= (−1)
[a
(2)
[[b
(1)
[
(a
(1)
b
(1)
⊗a
(2)
b
(2)
) (3.3.5)
• Furthermore, there is an even map S : H →H, the antipode, such that:
S(h
(1)
)h
(2)
= h
(1)
S(h
(2)
) = ε(h)1, for all h in H.
Remark 3.3.1. We emphasize that the compatibility condition (3.3.5), shows that a
super Hopf algebra is, in general, not a Hopf algebra in the category of vector spaces.
Proposition 3.3.2. One easily checks that formulas (3.3.3) and (3.3.4)of Proposition
(3.3.1) reduce to the following ones for a super Hopf algebra H:
S(ab) = (−1)
[a[[b[
S(b)S(a),
∆S(a) = S(a)
(1)
⊗S(a)
(2)
= (−1)
[a
(1)
[[a
(2)
[
S(a
(2)
) ⊗S(a
(1)
).
Example 3.3.4. [31] Recall from Example (3.2.4) that for any quasitriangular Hopf
algebra (H, R = R
1
⊗R
2
), the category, HMod, of all (left) Hmodules is a braided
monoidal abelian category. One can prove that, every (H, R) can be turned into a
braided Hopf algebra H in HMod as follows. As a vector space H = H, with H
module structure given by conjugation
a h = a
(1)
hS(a
(2)
).
It has the same multiplication, unit, and counit as H, but the comultiplication and
64
antipode are replaced by:
∆(h) = h
(1)
⊗h
(2)
= h
(1)
S(R
2
) ⊗R
1
h
(2)
,
S(h) = R
2
S(R
1
h).
Deﬁnition 3.3.5. Let H be a braided Hopf algebra in c. A right Hmodule is an
object M in c equipped with a morphism φ
M
: M ⊗ H → M, called H action, such
that:
(φ)(id
M
⊗m
H
) = (φ)(φ ⊗id
H
),
(φ)(id
M
⊗η
H
) = id
M
.
Deﬁnition 3.3.6. A left Hcomodule is an object M in c equipped with a morphism
ρ
M
: M →H ⊗M, called H coaction, such that:
(∆
H
⊗id
M
)(ρ) = (id
M
⊗ρ)(ρ),
(
H
⊗id
M
)(ρ) = (id
H
⊗
H
)(ψ
H,M
)(ρ) = id
M
.
Left Hmodules and right Hcomodules can be deﬁned in a similar manner.
Chapter 4
Braided Hopf Cyclic Cohomology
This chapter and the next one form the heart of this thesis. In this chapter we work
in an arbitrary braided monoidal abelian category and extend the notion of a stable
antiYetterDrinfeld module over a Hopf algebra in such a category. We show that to
any braided Hopf algebra endowed with a braided modular pair in involution one can
associate a paracocyclic object. This paracocyclic object is cocyclic if the ambient
category is symmetric. In fact Theorem (4.5.2) (cf. also Remark (4.5.1)), shows that
this paracocyclic object is almost never cocyclic if the category is not symmetric. Of
course, as with any paracocyclic object, by restricting to an appropriate subspace
we obtain a cocyclic object (cf. section (2.3)).
In the symmetric case one can go much further. In this case and for an arbitrary
braided stable antiYetterDrinfeld module we obtain a cocyclic object. In fact in this
case it is no longer needed to restrict to Hopf algebras and one can work with module
coalgebras.
A good motivation for this work is as follows. There are many examples of
Hopf algebralike objects that are close to being a Hopf algebra but are not a Hopf
algebra in the usual sense. Examples include (diﬀerential graded) super Hopf al
gebras, quasiHopf algebras, multiplier Hopf algebras, Hopf algebroids, and locally
compact quantum groups. In some cases, but certainly not always, these objects are
Hopf algebras in an appropriate braided monoidal category. Diﬀerential graded super
Hopf algebras and quasiHopf algebras are examples of this. Study of Hopf algebras
in symmetric monoidal categories goes back to [35]. Recent work on braided Hopf
algebras is mostly motivated by low dimensional topology [30, 31, 38].
65
66
As a special case of the theory that we provide, we will deﬁne a Hopf cyclic
cohomology for a (diﬀerential graded) super Hopf algebra and relate it to the co
homology of super Lie algebras by considering the enveloping algebra of a super Lie
algebra. We also provide a Hopf cyclic theory for quasitriangular quasiHopf algebras.
We should mention that the cyclic cohomology of (ribbon)algebras in braided
monoidal abelian categories has been introduced and studied in [2], motivated by
nonassociative geometry.
Throughout this chapter we shall use the following conventions to denote ob
jects and morphisms of a braided monoidal category (c, ⊗, I, ψ):
• A
n
for A
⊗n
,
• 1 for id, e.g. we write 1
A
or just 1 for id
A
,
• (f, g) for (f ⊗g),
• id
n
or just 1
n
for id
A
n,
• 1
A,B
for 1
A
⊗1
B
,
• ψ for ψ
A,A
.
For example instead of writing ∆
H
m
H
= (m
H
⊗m
H
)(id
H
⊗ψ
H,H
⊗id
H
)(∆
H
⊗∆
H
),
which expresses the fact that the comultiplication of a Hopf algebra is an anti
algebra map, we just write ∆m = (m, m)(1, ψ, 1)(∆, ∆); or instead of ψ
U,U⊗U
=
(id
U
⊗ ψ
U,U
)(ψ
U,U
⊗ id
U
), we simply write ψ
U,U
2
= (1, ψ)(ψ, 1) when there is no
chance of confusion, and so on.
4.1 The cocyclic module of a braided triple
(H, C, M)
In this section we extend the notion of a stable antiYetterDrinfeld (SAYD) module
[18, 19] (cf. Section 2.9) to braided monoidal categories and assign a cocyclic object
to a braided triple (H, C, M) in a symmetric braided monoidal abelian category c.
67
In the last section of this chapter we treat the general nonsymmetric case which is
much more subtle. Recall from section (2.3) that, by deﬁnition, in a paracocyclic
object all axioms of a cocyclic object are satisﬁed except the relations τ
n+1
n
= id.
Given a paracocyclic object X
n
, n ≥ 0, in an abelian category, we can always deﬁne
a cocyclic object by considering
X
n
:= ker (id −τ
n+1
n
), (4.1.1)
and restricting the faces, degeneracies, and cyclic operators to these subspaces. For
general notion of (co)cyclic objects and (co)cyclic modules we also refer to sections
(2.2)(2.4).
We ﬁx a strict, braided monoidal category (c, ⊗, I, ψ), and a Hopf algebra
(H, m, η, ∆, ε, S) in c. For the following deﬁnitions c need not be symmetric or
additive.
Deﬁnition 4.1.1. (compare with Deﬁnition 2.9.2) A rightleft braided stable anti
YetterDrinfeld Hmodule in c is an object M in c such that:
(i) M is a right Hmodule via an action φ
M
: M ⊗H →M,
(ii) M is a left Hcomodule via a coaction ρ
M
: M →H ⊗M,
(iii) M satisﬁes the braided antiYetterDrinfeld condition, i.e. :
(ρ)(φ) = [(m)(S ⊗m) ⊗φ][(ψ
H
⊗2
,H
⊗id
M
⊗id
H
)(id
H
⊗2
⊗ψ
M,H
⊗id
H
)
(id
H
⊗2
⊗id
M
⊗ψ
H,H
)(id
H
⊗ψ
M,H
⊗id
H
⊗2
)][ρ ⊗∆
2
]. (4.1.2)
(iv) M is stable, i.e. :
(φ)(ψ
H,M
)(ρ) = id
M
.
Remark 4.1.1. To deal with large expressions like (4.1.2) we break them into two
lines.
68
Deﬁnition 4.1.2. A quadruple (C, ∆
C
,
C
, φ
C
) is called a left (braided) Hmodule
coalgebra in c if (C, ∆
C
,
C
) is a coalgebra in c, and C is a left Hmodule via an
action φ
C
: H ⊗C →C such that φ
C
is a coalgebra map in c i.e. we have:
∆
C
φ
C
= (φ
C
⊗φ
C
)(id
H
⊗ψ
H,C
⊗id
C
)(∆
H
⊗∆
C
),
ε
C
φ
C
= ε
H
⊗ε
C
.
Deﬁnition 4.1.3. Let (C, φ
C
) be a left Hmodule. The diagonal action of H on
C
n+1
:= C
⊗(n+1)
is deﬁned by:
φ
C
n+1
: H ⊗C
n+1
→C
n+1
φ
C
n+1
:= (φ
C
, φ
C
, ..., φ
C
)
. ¸¸ .
n+1 times
(T(ψ
H,C
))(∆
n
H
⊗1
C
n+1
),
where,
T(ψ
H,C
) :=
n
i=1
(id
H
i
, ψ
H,C
, ψ
H,C
, ..., ψ
H,C
. ¸¸ .
n+1−i times
, id
C
i
).
Now we are going to associate a paracocyclic object to any triple (H, C, M),
where H is a Hopf algebra, C is an Hmodule coalgebra and M is a SAYD H
module, all in a symmetric monoidal category c. Notice that c need not be additive.
For n ≥ 0, let
C
n
= C
n
(C, M) := M ⊗C
n+1
.
We deﬁne faces δ
i
: C
n−1
→ C
n
, degeneracies σ
i
: C
n+1
→ C
n
and cyclic maps
τ
n
: C
n
→C
n
by:
69
δ
i
=
_
¸
¸
_
¸
¸
_
(1
M
, 1
C
i
, ∆
C
, 1
C
n−i−1
) 0 ≤ i < n
(1
M
, ψ
C,C
n)(1
M
, φ
C
, 1
C
n)(ψ
H,M
, 1
C
n+1
)(ρ
M
, ∆
C
, 1
C
n−1
) i = n
σ
i
= (1
M
, 1
C
i+1
, ε
C
, 1
C
n−i
), 0 ≤ i ≤ n
τ
n
= (1
M
, ψ
C,C
n)(1
M
, φ
C
, 1
C
n)(ψ
H,M
, 1
C
n+1
)(ρ
M
, 1
C
n+1
)
Proposition 4.1.1. If c is a symmetric monoidal category, then (C
∗
, δ
i
, σ
i
, τ) is a
paracocyclic object in c.
Remark 4.1.2. Note that by just using the faces we can construct the Hochschild
cohomology of the coalgebra C with coeﬃcients in the Cbicomodule M ⊗C with the
left and right coactions deﬁned as:
lρ
M,C
= (φ
C
, 1
M
, 1
C
)(1
M
, ψ
M,C
, 1
C
)(ρ
M
, ∆
C
) : (M ⊗C) →C ⊗(M ⊗C),
rρ
M,C
= (1
M
, ∆
C
) : (M ⊗C) →(M ⊗C) ⊗C.
For instance in a special case, if we put M = I with ρ
M
= (η
H
, 1
I
) the trivial
coaction, since I ⊗C = C, we get the Hochschild cohomology of the coalgebra C with
the coeﬃcients in the Cbicomodule C with the left and right coaction deﬁned by
∆
C
.
Now let us assume in addition that c is an abelian category. Then we can form
the balanced tensor products
C
n
H
= C
n
H
(C, M) := M ⊗
H
C
n+1
, n ≥ 0,
70
with induced faces, degeneracies and cyclic maps denoted by the same letters δ
i
, σ
i
and τ
n
.
Remark 4.1.3. Note that by just using the faces we can construct the Hochschild
cohomology of the coalgebra C with coeﬃcients in the Cbicomodule M ⊗
H
C with
the left and right coactions induced by lρ and rρ deﬁned in Remark (4.1.2).
For example, as a special case, if we put (C, ∆
C
) = (H, ∆
H
), since M⊗
H
H =
M, we get the Hochschild cohomology of the coalgebra H with coeﬃcients in the
Hbicomodule M with the left coaction deﬁned by ρ
M
and the right trivial coaction
(1
M
, η
H
) : M →M ⊗H.
Now we proceed to the main theorem of this thesis.
Theorem 4.1.1. If c is a symmetric monoidal abelian category, then (C
∗
H
, δ
i
, σ
i
,
τ) is a cocyclic object in c.
Proof. The most diﬃcult part is to show that the cyclic map τ
n
is well deﬁned
on C
n
H
(C, M) for all n. For this, we use the following diagram:
M ⊗H ⊗C
n+1
f
M,C
n+1
−−−−−−→ M ⊗C
n+1
coker(f
M,C
n+1
)
−−−−−−−−−−−→ M ⊗
H
C
n+1
¸
¸
_
τ
¸
¸
_
τ
M ⊗H ⊗C
n+1
f
M,C
n+1
−−−−−−→ M ⊗C
n+1
coker(f
M,C
n+1
)
−−−−−−−−−−−→ M ⊗
H
C
n+1
where f
M,C
n+1
= ((φ
M
⊗1
C
n+1
) −(1
M
⊗φ
C
n+1
)). So we have to show that
coker(f
M,C
n+1
) (τ) (f
M,C
n+1
) = 0,
i.e.
coker(f
M,C
n+1
) [(1
M
, ψ
C,C
n)(1
M
, φ
C
, 1
C
n)(ψ
H,M
, 1
C
n+1
)
(ρ
M
, 1
C
n+1
)] (φ
M
⊗1
C
n+1
−1
M
⊗φ
C
n+1
) = 0. (4.1.3)
71
It is not hard to see that the equality (4.1.3) is equivalent to:
coker(f
M,H
2
) [ [(1
M,H
, m
H
)(1
M
, ψ
H
2
,H
)(ψ
H,M
, 1
H
2
)(ρ
M
, ∆
H
)]
−[(1
M
, η
H
, 1
H
)(ψ
H,M
)(ρ
M
)(φ
M
)] ] = 0,
i.e. if we put:
α = [(1
M,H
, m
H
)(1
M
, ψ
H
2
,H
)(ψ
H,M
, 1
H
2
)(ρ
M
, ∆
H
)],
and
β = [(1
M
, η
H
, 1
H
)(ψ
H,M
)(ρ
M
)(φ
M
)],
then:
coker(f
M,H
2
) (α −β) = 0.
To prove this equality, we will deﬁne an isomorphism
¯ ϕ : M ⊗
H
H
2
→M ⊗
H
H
2
∼
= M ⊗H
and will show that,
( ¯ ϕ) coker(f
M,H
2
) (α −β) = 0.
More explicitly:
• Step 1:
We claim that ϕ deﬁned as below is a Hlinear isomorphism, where the domain
H
2
is considered as a Hmodule via diagonal action (φ
H
2
), and the codomain H
2
is
considered as a Hmodule via multiplication in the ﬁrst factor (φ
/
H
2
).
ϕ := (1
H
, m
H
)(1
H
, S
H
, 1
H
))(∆
H
, 1
H
) : H
2
→H
2
.
72
It is easy to see that ϕ is an isomorphism and in fact its inverse map is
ϕ
−1
:= (1, m)(∆, 1).
To see that ϕ is Hlinear, we have to show that the following diagram commutes:
H
2
ϕ
B
H ⊗H
2
1⊗ϕ
φ
H
2
¸¸
H ⊗H
2
φ
H
2
¸¸
i.e
(ϕ)(φ
H
2
) = (φ
/
H
2
)(1, ϕ), (4.1.4)
where φ
H
2
:= (m, m)(1, ψ, 1)(∆, 1
H
2
) is the diagonal action of H on the domain H
2
,
and φ
/
H
2
:= (m, 1) is the other action of H on the codomain H
2
. To prove that
(4.1.4) we see that:
RHS = (m, 1) (1, 1, m)(1, 1, S, 1)(1, ∆, 1)
= (m, m)(1, 1, S, 1)(1, ∆, 1),
and
LHS
(1)
= (1, m)(1, S, 1)(∆, 1) (m, m)(1, ψ, 1)(∆, 1, 1)
(2)
= (1, m)(1, S, 1)(1, 1, m)(∆m, 1, 1)(1, ψ, 1)(∆, 1, 1)
(3)
= (1, m)(1, 1, m)(1, S, 1, 1)(m, m, 1, 1)(1, ψ, 1, 1, 1)(∆, ∆, 1, 1)
(1, ψ, 1)(∆, 1, 1)
(4)
= (1, m)(1, 1, m)(m, 1, 1, 1)(1, 1, Sm, 1, 1)(1, ψ, 1, 1, 1)
(∆, ∆, 1, 1)(1, ψ, 1)(∆, 1, 1)
(5)
= (1, m)(1, 1, m)(m, 1, 1, 1)(1, 1, m, 1, 1)(1, 1, S, S, 1, 1)
(1, 1, ψ, 1, 1)(1, ψ, 1, 1, 1)(∆, ∆, 1, 1)(1, ψ, 1)(∆, 1, 1)
(6)
= (1, m)(1, 1, m)(m, 1, 1, 1)(1, 1, m, 1, 1)(1, 1, S, S, 1, 1)
(1, ψ
H,H
2
, 1, 1)(1, 1, ∆, 1, 1)(∆, 1, 1, 1)(1, ψ, 1)(∆, 1, 1)
(7)
= (1, m)(1, 1, m)(m, 1, 1, 1)(1, 1, m, 1, 1)(1, 1, S, S, 1, 1)
73
(1, ∆, 1, 1, 1)(1, ψ, 1, 1)(1, 1, ψ, 1)(∆, 1, 1, 1)(∆, 1, 1)
(8)
= (1, m)(1, 1, m)(m, 1, 1, 1)(1, 1, m, 1, 1)(1, 1, S, S, 1, 1)
(1, ∆, 1, 1, 1)(1, ψ
H
2
,H
)(1, ∆, 1, 1)(∆, 1, 1)
(9)
= (1, m)(1, 1, m)(m, 1, 1, 1)(1, 1, m, 1, 1)(1, 1, S, S, 1, 1)
(1, ∆, 1, 1, 1)(1, 1, ∆, 1)(1, ψ, 1)(∆, 1, 1)
(10)
= (1, m)(m, 1, 1)(1, 1, 1, m)(1, 1, 1, m, 1)(1, 1, S, S, 1, 1)
(1, 1, 1, ∆, 1)(1, ∆, 1, 1)(1, ψ, 1)(∆, 1, 1)
(11)
= (1, m)(m, 1, 1)(1, 1, 1, m)(1, 1, S, m(S, 1)∆, 1)
(1, ∆, 1, 1)(1, ψ, 1)(∆, 1, 1)
(12)
= (1, m)(m, 1, 1)(1, 1, 1, m)(1, 1, S, η, 1)(1, 1, 1, ε, 1)
(1, ∆, 1, 1)(1, ψ, 1)(∆, 1, 1)
(13)
= (1, m)(m, 1, 1)(1, 1, S, m(η, 1))(1, ∆, 1)(1, 1, ε, 1)
(1, ψ, 1)(∆, 1, 1)
(14)
= (m, m)(1, 1, S, 1)(1, ∆, 1)(1, ε, 1, 1)(∆, 1, 1)
= (m, m)(1, 1, S, 1)(1, ∆, 1)((1, ε)∆, 1, 1)
(15)
= (m, m)(1, 1, S, 1)(1, ∆, 1) = RHS.
Stages (1) to (15) are explained below:
(1) We use deﬁnitions ϕ = (1, m)(1, S, 1))(∆, 1) and φ
H
2
= (m, m)(1, ψ, 1)(∆, 1
H
2
).
(2) We use (m, m) = (1, m)(m, 1, 1), commute (1, m) and (∆, 1), i.e. (∆, 1)(1, m) =
(1, 1, m)(∆, 1, 1), and then use (∆, 1, 1)(m, 1, 1) = (∆m, 1, 1).
(3) We commute (1, S, 1) and (1, 1, m), and use ∆m = (m⊗m)(id ⊗ψ ⊗id)(∆⊗∆).
(4) We use (m, m, 1, 1) = (m, 1, 1, 1)(1, 1, m, 1, 1), commute (m, 1, 1, 1) and (1, S, 1, 1),
and then compose (1, 1, m, 1, 1) with (1, 1, S, 1, 1).
(5) We use Sm = m(S, S)ψ.
(6) We use ψ
H,H
2
= (1, ψ)(ψ, 1) and
(∆, ∆, 1, 1) = (1, 1, ∆, 1, 1)(∆, 1, 1, 1).
(7) We use the naturality of ψ to commute (1, ψ
H,H
2
, 1, 1) and (1, 1, ∆, 1, 1), more
74
explicitly we use the following commuting diagram:
H
2
(1,∆)
ψ
H ⊗H
2
ψ
H,H
2
H
2
(∆,1)
H
2
⊗H
also we commute (∆, 1, 1, 1) and (1, ψ, 1), i.e.
(∆, 1, 1, 1)(1, ψ, 1) = (1, 1, ψ, 1)(∆, 1, 1, 1).
(8) We use ψ
H
2
,H
= (ψ, 1)(1, ψ) and the coassociativity (∆, 1)∆ = (1, ∆)∆.
(9) We commute (1, ψ
H
2
,H
) and (1, ∆, 1, 1) using the naturality of ψ as in step (7).
(10) We use simple commutations and the associativity m(1, m) = m(m, 1), in four
ﬁrst parenthesis, also commute (1, ∆, 1, 1, 1) and (1, 1, ∆, 1).
(11) We use
(1, 1, 1, m, 1)(1, 1, S, S, 1, 1)(1, 1, 1, ∆, 1) = (1, 1, S, m(S, 1)∆, 1).
(12) We use m(S, 1)∆ = ηε.
(13) We use (1, 1, 1, ε, 1)(1, ∆, 1, 1) = (1, ∆, 1)(1, 1, ε, 1).
(14) We use m(η, 1) = 1, also commute (1, 1, ε, 1) and (1, ψ, 1) using naturality of ψ.
In fact we use this commuting diagram:
H
2
(ε,1)
ψ
I ⊗H
ψ
I,H
=id
I
H
2
(1,ε)
H ⊗I
(15) We use (1, ε)∆ = 1.
• Step 2:
Considering 1
M
: M → M as a Hlinear isomorphism we have, 1
M
⊗ ϕ :
75
M ⊗H
2
∼
= M ⊗H
2
and so we have:
¯ ϕ := 1
M
⊗
H
ϕ : M ⊗
H
H
2
∼
= M ⊗
H
H
2
∼
= (M ⊗
H
H) ⊗H
∼
= M ⊗H.
• Step 3:
Now we prove that ( ¯ ϕ) coker(f
M,H
2
) (α − β) = 0. For that, we look at this
commuting diagram
M ⊗H ⊗H
2
f
M,H
2
−−−−−→ M ⊗H
2
coker(f
M,H
2
)
−−−−−−−−−→ M ⊗
H
H
2
¸
¸
_
1
M
⊗ϕ
¸
¸
_¯ ϕ
M ⊗H ⊗H
2
f
M,H
2
−−−−−→ M ⊗H
2
coker(f
M,H
2
)
−−−−−−−−−→ M ⊗
H
H
2
where f
M,H
2
= (φ
M
⊗1
H
2
−1
M
⊗φ
H
2
) and f
/
M,H
2
= (φ
M
⊗1
H
2
−1
M
⊗φ
/
H
2
), which
shows, ( ¯ ϕ)(coker(f)) = (coker(f
/
))(1
M
⊗ϕ), so we instead will prove (coker(f
/
))(1
M
⊗
ϕ) (α −β) = 0, i.e.
(coker(f
/
))(1
M
⊗ϕ) (α) = (coker(f
/
))(1
M
⊗ϕ) β). (4.1.5)
The LHS of (4.1.5) is equal to:
(1)
= (coker(f
/
)) (1
M
, 1, m)(1
M
, 1, S, 1)(1
M
, ∆, 1) (1
M
, 1, m)(1
M
, ψ
H
2
,H
)
(ψ
H,M
, 1, 1)(ρ, ∆)
(2)
= (coker(f
/
)) (1
M
, 1, m)(1
M
, 1, S, 1)(1
M
, ∆, 1)(1
M
, ψ)(1
M
, m, 1)(ψ
H,M
, 1, 1)(ρ, ∆)
(3)
= (coker(f
/
)) (1
M
, 1, m)(1
M
, 1, S, 1)(1
M
, ψ
H,H
2
)(1
M
, 1, ∆)(1
M
, m, 1)
(ψ
H,M
, 1, 1)(ρ, ∆)
(4)
= (coker(f
/
)) (1
M
, 1, m)(1
M
, 1, S, 1)(1
M
, ψ
H,H
2
)(1
M
, m, 1, 1)(ψ
H,M
, 1, 1, 1)
(1, 1
M
, 1, ∆)(ρ, ∆)
(5)
= (coker(f
/
)) (1
M
, 1, m)(1
M
, 1, S, 1)(1
M
, 1, 1, m)(1
M
, ψ
H
2
,H
2
)
(ψ
H,M
, 1, 1, 1)(ρ, ∆
2
)
(6)
= (coker(f
/
)) (1
M
, 1, m(S, m))(1
M
, ψ
H
2
,H
2
)(ψ
H,M
, 1, 1, 1)(ρ, ∆
2
)
76
(7)
= (coker(f
/
)) (1
M
, m(1, η), m(S, m))(1
M
, ψ
H
2
,H
2
)(ψ
H,M
, 1, 1, 1)(ρ, ∆
2
)
(8)
= (coker(f
/
)) (1
M
, m, 1)(1
M
, 1, η, 1)(1
M
, 1, m(S, m))(1
M
, ψ
H
2
,H
2
)
(ψ
H,M
, 1, 1, 1)(ρ, ∆
2
)
(9)
= (coker(f
/
)) (1
M
, φ
/
H
2
)(1
M
, 1, η, 1)(1
M
, 1, m(S, m))(1
M
, ψ
H
2
,H
2
)
(ψ
H,M
, 1, 1, 1)(ρ, ∆
2
)
(10)
= (coker(f
/
)) (φ
M
, 1, 1)(1
M
, 1, η, 1)(1
M
, 1, m(S, m))
(1
M
, ψ
H
2
,H
2
)(ψ
H,M
, 1, 1, 1)
. ¸¸ .
(ρ, ∆
2
)
Stages (1) to (10) are explained below:
(1) We just use deﬁnitions of ϕ = (1, m)(1, S, 1))(∆, 1) and
α = [(1
M,H
, m
H
)(1
M
, ψ
H
2
,H
)(ψ
H,M
, 1
H
2
)(ρ, ∆)].
(2) We use the naturality of ψ to commute (1
M
, 1, m) and (1
M
, ψ
H
2
,H
), i.e. we use
the diagram:
H
2
⊗H
(m,1)
ψ
H,H
2
H
2
ψ
H ⊗H
2
(1,m)
H
2
(3) We commute (1
M
, ∆, 1) and (1
M
, ψ), again using naturality of ψ :
H
2
(1,∆)
ψ
H ⊗H
2
ψ
H,H
2
H
2
(∆,1)
H
2
⊗H
(4) We commute (1
M
, 1, ∆) ﬁrst with (1
M
, m, 1) and next with (ψ
H,M
, 1, 1).
(5) We commute (1
M
, ψ
H,H
2
) and (1
M
, m, 1, 1) using the diagram:
H
2
⊗H
2
(m,1,1)
ψ
H
2
,H
2
H ⊗H
2
ψ
H,H
2
H
2
⊗H
2
(1,1,m)
H
2
⊗H
77
Also we use (1, 1
M
, 1, ∆)(ρ, ∆) = (ρ, ∆
2
) and use the deﬁnition ∆
2
:= (1, ∆)∆.
(6) We use (1
M
, 1, m)(1
M
, 1, S, 1)(1
M
, 1, 1, m) = (1
M
, 1, m(S, m)).
(7) We use 1 = m(1, η).
(8) We use (1
M
, m(1, η), m(S, m)) = (1
M
, m, 1)(1
M
, 1, η, 1)(1
M
, 1, m(S, m)).
(9) We use the deﬁnition φ
/
H
2
= (m, 1).
(10) We use coker(f
/
) (1
M
, φ
/
H
2
) = coker(f
/
) (φ
M
, 1, 1), which comes from the deﬁ
nition of cokernel and this diagram:
M ⊗H ⊗H
2
f
M,H
2
=(φ
M
⊗1
H
2
−1
M
⊗φ
H
2
)
−−−−−−−−−−−−−−−−−−−−−→ M ⊗H
2
coker(f
)
−−−−−−→ M ⊗
H
H
2
The RHS of (4.1.5) is equal to:
(1)
= (coker(f
/
)) (1
M
, 1, m)(1
M
, 1, S, 1)(1
M
, ∆, 1) (1
M
, η, 1)(ψ
H,M
)(ρ)(φ)
(2)
= (coker(f
/
)) (1
M
, 1, m)(1
M
, 1, S, 1)(1
M
, η, η, 1)(ψ
H,M
)(ρ)(φ)
(3)
= (coker(f
/
)) (1
M
, 1, m)(1
M
, η, η, 1)(ψ
H,M
)(ρ)(φ)
(4)
= (coker(f
/
)) (1
M
, η, 1)(ψ
H,M
)(ρ)(φ)
(5)
= (coker(f
/
)) (1
M
, η, 1)(ψ
H,M
) (m(S, m), φ)(ψ
H
2
,H
, 1
M
, 1)(1, 1, ψ
M,H
, 1)
(1, 1, 1
M
, ψ)(1, ψ
M,H
, 1, 1)(ρ, ∆
2
)
(6)
= (coker(f
/
)) (1
M
, η, 1)(ψ
H,M
) (1, φ)(m(S, m), 1
M
, 1)(ψ
H
2
,H
, 1
M
, 1)
(1, 1, ψ
M,H
, 1)(1, 1, 1
M
, ψ)(1, ψ
M,H
, 1, 1)(ρ, ∆
2
)
(7)
= (coker(f
/
)) (1
M
, η, 1)(φ, 1)(ψ
H,M⊗H
)(m(S, m), 1
M
, 1)(ψ
H
2
,H
, 1
M
, 1)
(1, 1, ψ
M,H
, 1)(1, 1, 1
M
, ψ)(1, ψ
M,H
, 1, 1)(ρ, ∆
2
)
(8)
= (coker(f
/
)) (1
M
, η, 1)(φ, 1)(1
M
, 1, m(S, m))(ψ
H
3
,M⊗H
)(ψ
H
2
,H
, 1
M
, 1)
(1, 1, ψ
M,H
, 1)(1, 1, 1
M
, ψ)(1, ψ
M,H
, 1, 1)(ρ, ∆
2
)
(9)
= (coker(f
/
)) (φ, 1, 1)(1
M
, 1, η, 1)(1
M
, 1, m(S, m))
(ψ
H
3
,M⊗H
)(ψ
H
2
,H
, 1
M
, 1)(1, 1, ψ
M,H
, 1)(1, 1, 1
M
, ψ)(1, ψ
M,H
, 1, 1)
. ¸¸ .
(ρ, ∆
2
)
Stages (1) to (9) are explained below:
(1) We just use the deﬁnitions of ϕ = (1, m)(1, S, 1))(∆, 1) and,
β = [(1
M
, η, 1)(ψ
H,M
)(ρ)(φ)].
78
(2) We use (1
M
, ∆, 1)(1
M
, η, 1) = (1
M
, ∆η, 1) and ∆η = η ⊗η .
(3) We use (1
M
, 1, S, 1)(1
M
, η, η, 1) = (1
M
, η, Sη, 1) and Sη = η .
(4) We use (1
M
, 1, m)(1
M
, η, η, 1) = (1
M
, η, m(η, 1)) and m(η, 1) = 1 .
(5) We use the AYD formula (ρ)(φ) = (m(S, m), φ)(ψ
H
2
,H
, 1
M
, 1)(1, 1, ψ
M,H
, 1)
(1, 1, 1
M
, ψ)(1, ψ
M,H
, 1, 1)(ρ, ∆
2
).
(6) We use (m(S, m), φ) = (1, φ)(m(S, m), 1
M
, 1).
(7) We commute (ψ
H,M
) and (1, φ) using:
H ⊗(M ⊗H)
(1,φ)
ψ
H,M⊗H
H ⊗M
ψ
H,M
(M ⊗H) ⊗H
(φ,1)
M ⊗H
(8) We commute (ψ
H,M⊗H
) and (m(S, m), 1
M
, 1) using:
H
3
⊗(M ⊗H)
(m(S,m),1
M
,1)
−−−−−−−−−−→ H ⊗(M ⊗H)
ψ
H
3
,M⊗H
¸
¸
_
¸
¸
_
ψ
H,M⊗H
(M ⊗H) ⊗H
3
−−−−−−−−−−→
(1
M
,1,m(S,m))
(M ⊗H) ⊗H
(9) We commute (1
M
, η, 1) and (φ, 1).
Thus, LHS = RHS in (4.1.5) if:
(ψ
H
3
,M⊗H
)(ψ
H
2
,H
, 1
M
, 1)(1, 1, ψ
M,H
, 1)(1, 1, 1
M
, ψ)(1, ψ
M,H
, 1, 1)
= (1
M
, ψ
H
2
,H
2
)(ψ
H,M
, 1, 1, 1),
which is true since:
(ψ
H
3
,M⊗H
)(ψ
H
2
,H
, 1
M
, 1)(1, 1, ψ
M,H
, 1)(1, 1, 1
M
, ψ)(1, ψ
M,H
, 1, 1)
(1)
= (1
M
, 1, ψ
H
2
,H
)(ψ
H
3
,M⊗H
)(1, 1, ψ
M,H
, 1)(1, 1, 1
M
, ψ)(1, ψ
M,H
, 1, 1)
(2)
= (1
M
, 1, ψ
H
2
,H
) (1
M
, ψ, 1, 1)(1
M
, 1, ψ, 1)(1
M
, 1, 1, ψ)(ψ
H,M
, 1, 1, 1)(1, ψ
H,M
, 1, 1)
(1, 1, ψ
H,M
, 1)(1, 1, ψ
M,H
, 1)
. ¸¸ .
(1, 1, 1
M
, ψ)(1, ψ
M,H
, 1, 1)
(3)
= (1
M
, 1, ψ
H
2
,H
) (1
M
, ψ, 1, 1)(1
M
, 1, ψ, 1)(1
M
, 1, 1, ψ)(ψ
H,M
, 1, 1, 1)(1, ψ
H,M
, 1, 1)
(1, 1, 1
M
, ψ)(1, ψ
M,H
, 1, 1)
79
(4)
= (1
M
, 1, ψ
H
2
,H
) (1
M
, ψ, 1, 1)(1
M
, 1, ψ, 1)(1
M
, 1, 1, ψ)(ψ
H,M
, 1, 1, 1)(1, 1
M
, 1, ψ)
(1, ψ
H,M
, 1, 1)(1, ψ
M,H
, 1, 1)
. ¸¸ .
(5)
= (1
M
, 1, ψ
H
2
,H
) (1
M
, ψ, 1, 1)(1
M
, 1, ψ, 1)(1
M
, 1, 1, ψ)(ψ
H,M
, 1, 1, 1)(1, 1
M
, 1, ψ)
(6)
= (1
M
, 1, ψ
H
2
,H
) (1
M
, ψ, 1, 1)(1
M
, 1, ψ, 1) (1
M
, 1, 1, ψ)(1
M
, 1, 1, ψ)
. ¸¸ .
(ψ
H,M
, 1, 1, 1)
(7)
= (1
M
, 1, ψ
H
2
,H
) (1
M
, ψ, 1, 1)(1
M
, 1, ψ, 1)(ψ
H,M
, 1, 1, 1)
(8)
= (1
M
, ψ
H
2
,H
2
)(ψ
H,M
, 1, 1, 1)
Stages (1) to (8) are explained below:
(1) We commute (ψ
H
3
,M⊗H
) and (ψ
H
2
,H
, 1
M
, 1) using:
H
3
⊗(M ⊗H)
(ψ
H
2
,H
,1
M
,1)
−−−−−−−−−−→ H
3
⊗(M ⊗H)
ψ
H
3
,M⊗H
¸
¸
_
¸
¸
_
ψ
H
3
,M⊗H
(M ⊗H) ⊗H
3
−−−−−−−−−−→
(1
M
,1,ψ
H
2
,H
)
(M ⊗H) ⊗H
3
(2) We use (ψ
H
3
,M⊗H
) = (1
M
, ψ, 1, 1)(1
M
, 1, ψ, 1)(1
M
, 1, 1, ψ)(ψ
H,M
, 1, 1, 1).
(1, ψ
H,M
, 1, 1)(1, 1, ψ
H,M
, 1)
(3) We use (1, 1, ψ
H,M
, 1)(1, 1, ψ
M,H
, 1) = 1, which comes from the symmetric prop
erty of ψ, i.e. ψ
H,M
ψ
M,H
= 1.
(4) We commute (1, ψ
H,M
, 1, 1) and (1, 1, 1
M
, ψ).
(5) We again use the symmetric property of ψ to put (1, ψ
H,M
, 1, 1)(1, ψ
M,H
, 1, 1) =
1.
(6) We commute (ψ
H,M
, 1, 1, 1) and (1, 1
M
, 1, ψ).
(7) We again use the symmetric property of ψ to put (1
M
, 1, 1, ψ)(1
M
, 1, 1, ψ) = 1.
(8) We use (1
M
, ψ
H
2
,H
2
) = (1
M
, 1, ψ
H
2
,H
) (1
M
, ψ, 1, 1)(1
M
, 1, ψ, 1).
Using the stability property of M, (φ)(ψ
H,M
)(ρ) = id
M
, and with almost the
80
same procedure as above, one proves:
(τ
n
)
n+1
= id.
Also the other properties of a cocyclic object can be easily checked. This ﬁnishes the
proof of Theorem (4.1.1).
Example 4.1.1. As a special case, if we put C = H as a Hmodule coalgebra over
itself, and M = I, we obtain a braided version of ConnesMoscovici’s Hopf cyclic
theory [9, 10, 11] in any symmetric monoidal abelian category. We shall explain this
example in more details in the next section.
Example 4.1.2. More generally, for C = H and any SAYD M, we can obtain a
braided version of the Hopf cyclic cohomology of H with coeﬃcients in M.
4.2 The braided version of ConnesMoscovici’s
Hopf cyclic cohomology
In this section we show that when C = H and M =
σ
I
δ
, the Hopf cyclic complex of the
triple (H, H,
σ
I
δ
), simpliﬁes and reduces to a braided version of ConnesMoscovici’s
Hopf cyclic theory [9, 10, 11] in any symmetric monoidal abelian category.
Let c be a strict braided monoidal category and (H, ∆, ε, m, η, S) be a Hopf al
gebra in c. Notice that, except for Theorem (4.2.1), c is not assumed to be symmetric
or additive.
Deﬁnition 4.2.1. A character for H is a morphism δ : H → I in c which is an
algebra map, i.e :
δm = δ ⊗δ and δη = id
I
.
81
A cocharacter for H is a morphism σ : I →H which is a coalgebra map, i.e :
∆σ = σ ⊗σ and εσ = id
I
.
A pair (δ, σ) consisting of a character and a cocharacter is called a braided modular
pair if:
δσ = id
I
.
Deﬁnition 4.2.2. If δ is a character for H, the corresponding δtwisted antipode
¯
S
is deﬁned by:
¯
S := (δ ⊗S)∆.
Proposition 4.2.1. If
¯
S is a δtwisted antipode for H then we have:
¯
Sm = mψ(
¯
S ⊗
¯
S), (4.2.1)
¯
Sη = η,
∆
¯
S = ψ(
¯
S ⊗S)∆, (4.2.2)
ε
¯
S = δ, δ
¯
S = ε,
¯
Sσ = Sσ,
m(Sσ ⊗σ) = m(
¯
Sσ ⊗σ) = η. (4.2.3)
Deﬁnition 4.2.3. A modular pair (δ, σ) for H is called a braided modular pair in
involution (BMPI) if:
m((id ⊗m)(
¯
Sσ ⊗
¯
S
2
⊗σ)) = id.
It is easy to see that
m((id ⊗m)(
¯
Sσ ⊗
¯
S
2
⊗σ)) = m(
¯
Sσ ⊗m(
¯
S
2
⊗σ)).
82
Example 4.2.1. If H is a Hopf algebra in V ect
C
, then the above deﬁnitions reduce
to those of ConnesMoscovici [9, 10, 11].
Example 4.2.2. One can easily check that, if I is considered as a right Hmodule
via a character δ:
φ
I
= δ : I ⊗H = H →I,
and as a left Hcomodule via a cocharacter σ:
ρ
I
= σ : I →H ⊗I = I,
then I is a braided SAYD module over H if and only if (δ, σ) is a braided MPI. We
denote this SAYD module by
σ
I
δ
.
Lemma 4.2.1. If H is commutative in the sense that mψ = m or cocommutative in
the sense that ∆ = ψ∆, then S
2
= id and thus (ε, 1) is an MPI for H.
Proof. We have:
1 = id
H
(1)
= m(η, 1)(ε, 1)∆ = m(ηε, 1)∆
(2)
= m(Sηε, 1)∆ = m(S(m(1, S)∆, 1)∆
(3)
= m(m(S, S)ψ(1, S)∆, 1)∆ = m(m(S
2
, S)ψ∆, 1)∆
(4)
= m(m(S
2
, S)∆, 1)∆ = m(m, 1)(S
2
, S, 1)(∆, 1)∆
(5)
= m(1, m)(S
2
, S, 1)(1, ∆)∆ = m(S
2
, m(S, 1)∆)∆ = m(S
2
, ηε)∆
= m(1, η)(S
2
, 1
I
)(1, ε)∆ = (S
2
, 1
I
)
(6)
= S
2
Stages (1) to (6) are explained below:
(1) We use (ε, 1)∆ = m(η, 1) = 1.
(2) We use Sη = η and m(1, S)∆ = ηε.
(3) We use Sm = m(S, S)ψ , commute ψ and (1, S) using the naturality of ψ, and
83
use (S, S)(S, 1) = (S
2
, S).
(4) In the commutative case we ﬁrst use (S, S)ψ = ψ(S, S) and then use mψ = m. In
the cocommutative case we use ∆ = ψ∆, and then (m(S
2
, S)∆, 1) = (m, 1)(S
2
, S, 1)(∆, 1).
(5) We use m(m, 1) = m(1, m), (∆, 1)∆ = (1, ∆)∆, and m(S, 1)∆ = ηε.
(6) We use (ε, 1)∆ = m(η, 1) = 1 again.
Remark 4.2.1. As one can see, in the proof of the above theorem we didn’t need the
symmetric property ψ
2
= id for c.
Now we are ready to give a braided version of ConnesMoscovici’s Hopf cyclic
theory.
Theorem 4.2.1. Suppose (H, (δ, σ)) is a braided Hopf algebra in a symmetric braided
monoidal abelian category c, where (δ, σ) is a braided MPI. If we put (C; φ
C
, ∆
C
) =
(H; m
H
, ∆
H
), and M =
σ
I
δ
, then the cocyclic object of Theorem (4.1.1) reduces to
the following one which is a braided version of ConnesMoscovici’s Hopf cyclic theory:
C
0
(H) = I and C
n
(H) = H
n
, n ≥ 1,
with faces, degeneracies, and cyclic maps given by:
δ
i
=
_
¸
¸
¸
¸
_
¸
¸
¸
¸
_
(η, 1, 1, ..., 1) i = 0
(1, 1, ..., 1, ∆
i−th
, 1, 1, ...1) 1 ≤ i ≤ n −1
(1, 1, ..., 1, σ) i = n
σ
i
= (1, 1, ..., ε
(i+1)−th
, 1, 1..., 1), 0 ≤ i ≤ n
τ
n
=
_
_
_
id
I
n = 0
(m
n
)(∆
n−1
¯
S, 1
H
n−1
, σ) n ,= 0
84
Here by m
n
we mean, m
1
= m, and for n ≥ 2:
m
n
= m
H
n = (m, m, ..., m
. ¸¸ .
n times
)T
n
(ψ),
where
T
n
(ψ) :=
n−1
j=1
(1
H
j
, ψ, ψ, ..., ψ
. ¸¸ .
n−j times
, 1
H
j
).
Remark 4.2.2. In section (4.5) we discuss the analogue of this theorem in a non
symmetric braided monoidal abelian category.
Remark 4.2.3. One can deﬁne the notions of Hmodule algebra and δinvariant
σtrace and deﬁne a characteristic map as in [9, 10, 11].
Example 4.2.3. (ConnesMoscovici’s theory) If one puts c = V ect
C
and I = C,
then the above formulas reduce to those in [9, 10, 11].
As another example of the above theory, we devote Section (4.3) to provide a
Hopf cyclic theory for super Hopf algebras. But before that we give one more result
here:
The following braided version of Theorem (4.2) in [25], can be proved along the
same lines.
Theorem 4.2.2. If H is commutative, then
HC
n
(ε,1)
(H) =
i=n(mod 2)
HH
i
(H, I), ∀ n ≥ 0, (4.2.4)
where the left hand side is the cyclic cohomology of the braided Hopf algebra H with
MPI (ε, 1) and the right hand side is the Hochschild cohomology of coalgebra H with
coeﬃcients in Hbimodule I.
85
4.3 Hopf cyclic cohomology for super Hopf
algebras
In this section we give explicit formulas for the Hopf cyclic complex in the special
case of super Hopf algebras and compute it in the super commutative case.
Remark 4.3.1. Let H = H
0
⊕H
1
be a super Hopf algebra and (δ, σ) be a super MPI
for H. One checks that formulas (4.2.1), (4.2.2), and (4.2.3) in proposition (4.2.1)
reduce to:
¯
S(ab) = (−1)
[a[[b[
¯
S(b)
¯
S(a),
∆
¯
S(a) =
¯
S(a)
(1)
⊗
¯
S(a)
(2)
= (−1)
[a
(1)
[[a
(2)
[
S(a
(2)
) ⊗
¯
S(a
(1)
),
S(σ) =
¯
S(σ) = σ
−1
.
Proposition 4.3.1. Let H = H
0
⊕H
1
be a super Hopf algebra endowed with a braided
modular pair in involution (δ, σ). Then the complex, faces, degeneracies and cyclic
maps of Theorem (4.2.1) can be written as :
C
0
(H) = C and C
n
(H) = H
n
, n ≥ 1
δ
i
(h
1
, ..., h
n−1
) =
_
¸
¸
¸
_
¸
¸
¸
_
(1, h
1
, h
2
, ..., h
n−1
) i = 0
(h
1
, h
2
, ..., h
(1)
i
, h
(2)
i
, ..., h
n−1
) 1 ≤ i ≤ n −1
(h
1
, h
2
, ..., h
n−1
, σ) i = n
σ
i
(h
1
, h
2
, ..., h
n+1
) = ε(h
i+1
) (h
1
, h
2
, ..., h
i
, h
i+2
, ..., h
n+1
), 0 ≤ i ≤ n
τ
n
(h
1
, h
2
, ..., h
n
) = αβ (S(h
(n)
1
)h
2
, S(h
(n−1)
1
)h
3
, ..., S(h
(2)
1
)h
n
,
¯
S(h
(1)
1
)σ),
86
where h
i
’s are homogeneous elements and:
α =
n−1
i=1
(−1)
([h
(1)
1
[+...+[h
(i)
1
[)([h
(i+1)
1
[)
,
β =
n−1
j=1
(−1)
[h
(j)
1
[([h
2
[+[h
3
[+...+[h
n−j+1
[)
.
The next lemma is a corollary of Lemma (4.2.1).
Lemma 4.3.1. If a super Hopf algebra H = H
0
⊕H
1
is super commutative or super
cocommutative, then S
2
= 1 and thus (ε, 1) ia a super MPI for H.
The next proposition is a corollary of Theorem (4.2.2)
Proposition 4.3.2. If a super Hopf algebra H = H
0
⊕ H
1
is super commutative,
then we have a decomposition:
HC
n
(ε,1)
(H) =
i=n(mod 2)
HH
i
(H, C), ∀ n ≥ 0.
Remark 4.3.2. All the results of this section easily extend to the case of diﬀerential
graded super Hopf algebras.
4.4 Hopf cyclic cohomology of the enveloping
algebra of a super Lie algebra
A good reference for super Lie algebras, their enveloping algebras, and a super ana
logue of the Poincar´eBirkhoﬀWitt theorem is [13]. This latter result is specially
important for the proof of Lemma (4.4.2) . Let g = g
0
⊕ g
1
be a super Lie algebra,
let
g :=
T(g)
(a ⊗b + (−1)
[a[[b[
b ⊗a)
,
87
be the exterior algebra of g and let
H = U(g) :=
T(g)
([a, b] −a ⊗b + (−1)
[a[[b[
b ⊗a)
,
be the enveloping algebra of g. Here T(g) is the tensor algebra of g. All these
constructions are done in the category of super vector spaces [13]. U(g) is a super
cocommutative super Hopf algebra.
Our goal in this section is to show that, analogous to the nongraded case
[9, 10, 11], the relation
HP
∗
(δ,1)
(U(g)) =
i=∗(mod 2)
H
i
(g; C
δ
),
holds, where δ is a character for g. Here HP
∗
(δ,1)
(U(g)) is the periodic Hopf cyclic
cohomology of the super Hopf algebra H = U(g), and H
i
(g; C
δ
) is the (super) Lie
algebra homology of g with coeﬃcient in the gmodule C
δ
.
Let H be any super Hopf algebra with a(super) modular pair in involution
(δ, σ). Deﬁne H
by H
n+1
in degree n with maps:
¯
δ
i
(h
0
, h
1
, ..., h
n
) =
_
_
_
(h
0
, h
1
, h
2
, ..., h
(1)
i
, h
(2)
i
, ..., h
n
) 0 ≤ i ≤ n
µ (h
(2)
0
, h
1
, h
2
, ..., h
n
, σh
(1)
0
) i = n + 1
where µ = (−1)
[h
(1)
0
[([h
(2)
0
[+[h
1
[+...+[h
n
[)
,
¯ σ
i
(h
0
, h
1
, ..., h
n+1
) = ε(h
i+1
) (h
0
, h
1
, ..., h
i
, h
i+2
, ..., h
n+1
), 0 ≤ i ≤ n
¯ τ
n
(h
0
, h
1
, h
2
, ..., h
n
) = ξ (h
1
, h
2
, ..., h
n
, σh
0
),
where ξ = (−1)
[h
0
[([h
1
[+...+[h
n
[)
,
¯
λ
n
= (−1)
n
¯ τ
n
(h
0
, h
1
, h
2
, ..., h
n
) = (−1)
n
ξ (h
1
, h
2
, ..., h
n
, σh
0
),
88
¯ s(h
0
, h
1
, h
2
, ..., h
n
) = ε(h
0
)(h
1
, h
2
, ..., h
n
),
¯
N = 1 +
¯
λ +
¯
λ
2
+ ... +
¯
λ
n
,
¯
B =
¯
N¯ s(1 −
¯
λ).
After forming C
δ
⊗
H
H
we get the complex H
δ,σ
which is H
n
in degree n. Deﬁne
the projection:
π : H
→ H
δ,σ
π(h
0
, ..., h
n
) =
¯
S(h
0
) (h
1
, ..., h
n
) = αβ(S(h
(n)
0
)h
1
, ..., S(h
(2)
0
)h
n−1
,
¯
S(h
(1)
0
)h
n
),
where
α =
n−1
i=1
(−1)
([h
(1)
0
[+...+[h
(i)
0
[)([h
(i+1)
0
[)
and β =
n−1
i=1
(−1)
([h
(i)
0
[)([h
1
[+...+[h
n−i
[)
,
After applying π, the maps on H
δ,σ
will coincide with those in Proposition (4.3.1),
i.e.
δ
i
(h
1
, ..., h
n−1
) =
_
¸
¸
¸
_
¸
¸
¸
_
(1, h
1
, h
2
, ..., h
n−1
) i = 0
(h
1
, h
2
, ..., h
(1)
i
, h
(2)
i
, ..., h
n−1
) 1 ≤ i ≤ n −1
(h
1
, h
2
, ..., h
n−1
, σ) i = n
σ
i
(h
1
, h
2
, ..., h
n+1
) = ε(h
i+1
) (h
1
, h
2
, ..., h
i
, h
i+2
, ..., h
n+1
), 0 ≤ i ≤ n
τ
n
(h
1
, h
2
, ..., h
n
) =
¯
S(h
1
) (h
2
, ..., h
n
, σ)
= α
/
β
/
(S(h
(n)
1
)h
2
, S(h
(n−1)
1
)h
3
, ..., S(h
(2)
1
)h
n
,
¯
S(h
(1)
1
)σ),
where h
i
’s are homogeneous elements and:
α
/
=
n−1
i=1
(−1)
([h
(1)
1
[+...+[h
(i)
1
[)([h
(i+1)
1
[)
,
β
/
=
n−1
j=1
(−1)
[h
(j)
1
[([h
2
[+[h
3
[+...+[h
n−j+1
[)
,
89
and
λ
n
= (−1)
n
τ
n
,
s(h
1
, h
2
, ..., h
n+1
) =
¯
S(h
1
) (h
2
, ..., h
n+1
) =
α
/
β
/
(S(h
(n)
1
)h
2
, S(h
(n−1)
1
)h
3
, ..., S(h
(2)
1
)h
n
,
¯
S(h
(1)
1
)h
n+1
),
N = 1 + λ + λ
2
+ ... + λ
n
,
B = Ns(1 −λ).
Note that the projection π is a map of cocyclic modules, so we have for example
Bπ = π
¯
B.
Next we notice that the Hochschild cohomology HH
∗
(H, C
σ
) depends only on
the coalgebra structure of H and the grouplike elements σ and 1. In fact we have:
Lemma 4.4.1. HH
∗
(H, C
σ
) = Cotor
∗
H
(C, C
σ
).
Let
S(g) :=
T(g)
(a ⊗b −(−1)
[a[[b[
b ⊗a)
denote the symmetric algebra of the super vector space g. It is a super cocommutative
super Hopf algebra with the comultiplication deﬁned by ∆(x) = x ⊗ 1 + 1 ⊗ x for
homogeneous elements x of g.
Lemma 4.4.2. The antisymmetrization map
A :
_
n
g →U(g)
n
deﬁned by
A(x
1
∧ ... ∧ x
n
) = (
σ∈S
n
(−1)
α
σ
sign(σ)(x
σ(1)
, ..., x
σ(n)
))/n!. (4.4.5)
90
induces an isomorphism HH
∗
(U(g), C) =
_
∗
(g). Here
α
σ
=
n
i=1
[x
σ(i)
[([x
1
[ +[x
2
[ + ...
[x
σ(j)
[ + ... +[x
σ(i)−1
[),
where
[x
σ(j)
[ means that if there are any of [x
σ(j)
[’s for all j < i they should be
omitted. Simply α
σ
contains [a[[b[ for any two elements a and b of x
i
’s, if they cross
each other.
Proof. Since by the super Poincar´eBirkhoﬀWitt theorem [13], U(g) = S(g) as
super coalgebras, it is enough to prove the lemma just for S(g), i.e. HH
∗
(S(g), C) =
_
∗
(g).
Let f
1
, ..., f
n
=f
e
1
, ..., f
e
k
, f
o
1
..., f
o
l
be a graded basis for g = g
0
⊕ g
1
of dimen
sion n = k + l and f
∗
1
, ..., f
∗
n
be the dual basis for g
∗
= g
∗
0
⊕ g
∗
1
. Extend f
∗
i
’s to
superderivations f
i
: S(g) →S(g); 1 ≤ i ≤ n = k +l. Consider the free resolution of
left S(g)comodules:
0 −−−→ C −−−→ S(g) ⊗
_
0
g
∂
−−−→ S(g) ⊗
_
1
g
∂
−−−→ S(g) ⊗
_
2
g
∂
−−−→ ...
where ∂ is deﬁned on homogeneous elements by:
∂(a ⊗x
1
∧ ... ∧ x
n
) =
f
i
(a) ⊗f
i
∧ x
1
∧ ... ∧ x
n
.
Let denote the cotensor product. By Lemma (4.4.1), HH
∗
(S(g), C) is com
puted by the following complex:
C
S(g)
S(g) ⊗
_
0
g −−−→ C
S(g)
S(g) ⊗
_
1
g −−−→ C
S(g)
S(g) ⊗
_
2
g −−−→ ...
where diﬀerentials are C
S(g)
∂. It is easy to see that this complex is nothing but the
following one:
91
_
0
g
0
−−−→
_
1
g
0
−−−→
_
2
g
0
−−−→ ...
so we have: HH
∗
(S(g)) =
_
∗
(g).
Comparing the above resolution with the standard bar resolution 0 → C →
B(S(g), C) we will show that this isomorphism is induced by A. First we recall that
the standard bar resolution is:
0 −−−→ C −−−→ S(g)
b
−−−→ S(g)
2
b
−−−→ S(g)
3
b
−−−→ ...
b
/
(x
0
, ..., x
n
) = (
n
i=0
(−1)
i
(x
0
, ..., ∆(x
i
), ...x
n
)) + (−1)
n+1
(x
0
, ..., x
n
, 1).
To compare the two resolutions, we deﬁne the map:
T : S(g)
n+1
→S(g) ⊗
n
g
T(x
0
, ..., x
n
) = x
0
⊗p(x
1
) ∧ ... ∧ p(x
n
),
where p is the projection map from S(g) to g. One checks that T is a chain map
between two free resolutions. Therefore the map
ˆ
T = C
S(g)
T : S(g)
n
→
n
g
ˆ
T(x
1
, ..., x
n
) = p(x
1
) ∧ ... ∧ p(x
n
),
induces our isomorphism HH
∗
(S(g), C) =
_
∗
(g). Now we check the relation
ˆ
TA =
id, which shows that the isomorphism is also induced by A. We have:
ˆ
TA(x
1
∧ ... ∧ x
n
)
92
=
ˆ
T((
σ∈S
n
(−1)
α
σ
sign(σ)(x
σ(1)
, ..., x
σ(n)
))/n!)
= (
σ∈S
n
(−1)
α
σ
sign(σ)
ˆ
T(x
σ(1)
, ..., x
σ(n)
))/n!
= (
σ∈S
n
(−1)
α
σ
sign(σ)(x
σ(1)
∧ ... ∧ x
σ(n)
))/n!
= (
σ∈S
n
(−1)
α
σ
sign(σ)(−1)
α
σ
sign(σ)(x
1
∧ ... ∧ x
n
))/n!
= (
σ∈S
n
(x
1
∧ ... ∧ x
n
))/n!
= (x
1
∧ ... ∧ x
n
).
The following complex is the super analogue of the ChevalleyEilenberg complex
to compute the Lie algebra homology H
∗
(g, C
δ
) of the super Lie algebra g:
_
0
g
_
1
g
δ
¸
_
2
g
d
¸
_
3
g
d
¸ ...
d
¸
d(x
1
, ..., x
n
) = (
n
i=1
(−1)
i+1+α
i
δ(x
i
)x
1
∧ ... ∧ ´ x
i
∧ ... ∧ x
n
)
+(
i<j
(−1)
i+j+α
i
+α
j
−[x
i
[[x
j
[
[x
i
, x
j
]∧x
1
∧...∧´ x
i
∧...∧´ x
j
∧...∧x
n
),
where α
1
:= 0 and α
i
= [x
i
[([x
1
[ + ... +[x
i−1
[) ; i > 1.
Clearly we have a bicomplex
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
_
4
g
d
−−−→
_
3
g
d
−−−→
_
2
g
d
−−−→
_
1
g
d
−−−→ C
0
¸
¸
¸ 0
¸
¸
¸ b
¸
¸
¸ 0
¸
¸
¸
_
3
g
d
−−−→
_
2
g
d
−−−→
_
1
g
d
−−−→ C
0
¸
¸
¸ 0
¸
¸
¸ 0
¸
¸
¸
_
2
g
d
−−−→
_
1
g
d
−−−→ C
0
¸
¸
¸ 0
¸
¸
¸
_
1
g
d
−−−→ C
0
¸
¸
¸
C
93
whose total homology is
i=∗(mod 2)
H
i
(g; C
δ
). On the other hand we have the
(b, B)bicomplex
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
U(g)
⊗4
B
−−−→ U(g)
⊗3
B
−−−→ U(g)
⊗2
B
−−−→ U(g)
B
−−−→ C
b
¸
¸
¸ b
¸
¸
¸ b
¸
¸
¸ b
¸
¸
¸
U(g)
⊗3
B
−−−→ U(g)
⊗2
B
−−−→ U(g)
B
−−−→ C
b
¸
¸
¸ b
¸
¸
¸ b
¸
¸
¸
U(g)
⊗2
B
−−−→ U(g)
B
−−−→ C
b
¸
¸
¸ b
¸
¸
¸
U(g)
B
−−−→ C
b
¸
¸
¸
C
whose total homology is HP
∗
(δ,1)
(U(g)).
Let us check that the antisymmetrization map Acommutes with the Boperator:
BA(x
1
∧ ... ∧ x
n
)
= Bπ(
σ∈S
n
(−1)
α
σ
sign(σ)(1 ⊗x
σ(1)
, ..., x
σ(n)
))/n!
= π
¯
B(
σ∈S
n
(−1)
α
σ
sign(σ)(1 ⊗x
σ(1)
, ..., x
σ(n)
))/n!
= π
¯
N¯ s(1 −
¯
λ)(
σ∈S
n
(−1)
α
σ
sign(σ)(1 ⊗x
σ(1)
, ..., x
σ(n)
))/n!
= π
¯
N¯ s(
σ∈S
n
(−1)
α
σ
sign(σ)(1 ⊗x
σ(1)
, ..., x
σ(n)
) −(−1)
n
(x
σ(1)
, ..., x
σ(n)
⊗1))/n!
= π
¯
N(
σ∈S
n
(−1)
α
σ
sign(σ)(x
σ(1)
, ..., x
σ(n)
))/n!
= π(
σ∈S
n
(−1)
α
σ
sign(σ)(x
σ(1)
, ..., x
σ(n)
))/(n −1)!
= (
σ∈S
n
(−1)
α
σ
sign(σ)π(x
σ(1)
, ..., x
σ(n)
))/(n −1)!
= (
σ∈S
n
(−1)
α
σ
sign(σ)(δ(x
σ(1)
)(x
σ(2)
, ..., x
σ(n)
)−
n
i=2
(−1)
[x
σ(1)
[([x
σ(2)
[+...+[x
σ(i−1)
[)
(x
σ(2)
, ..., x
σ(1)
x
σ(i)
, ..., x
σ(n)
)))/(n−1)!
94
= A((
n
i=1
(−1)
i+1+α
i
δ(x
i
)x
1
∧ ... ∧ ˆ x
i
∧ ... ∧ x
n
)
+ (
i<j
(−1)
i+j+α
i
+α
j
−[x
i
[[x
j
[
[x
i
, x
j
] ∧ x
1
∧ ... ∧ ˆ x
i
∧ ... ∧ ˆ x
j
∧ ... ∧ x
n
))
= Ad(x
1
, ..., x
n
).
Now using Lemma (4.4.2) we conclude that the antisymmetrization map
A :
_
n
g →U(g)
n
,
deﬁnes a quasiisomorphism between the above bicomplexes. Summarizing everything
we have:
Theorem 4.4.1.
HP
∗
(δ,1)
(U(g)) =
i=∗(mod 2)
H
i
(g; C
δ
).
4.5 Hopf cyclic cohomology in nonsymmetric
monoidal categories
In Theorem (4.2.1) we obtained a braided version of ConnesMoscovici’s Hopf cyclic
theory in a symmetric monoidal category. This was obtained as a special case of a
more general result in Section (4.1), for braided triples (H, C, M) in a symmetric
monoidal category. In this section we proceed to eliminate the restrictive symmetry
condition ψ
2
= id. For this, we will directly show that the cocyclic object of Theorem
(4.2.1) remains paracocyclic in any braided abelian monoidal category c, without
any symmetry condition on the part of c (Theorem (4.5.1) below). We shall also
indicate how the symmetry condition on the braiding is related to the cyclic condition
τ
n+1
n
= id (Theorem (4.5.2) below). The upshot is that to obtain a cocyclic object and
a Hopf cyclic cohomology in nonsymmetric monoidal categories one must inevitably
restrict to the subcomplex ker(1 − τ
n+1
n
) (cf. formula (4.1.1)). This procedure can,
95
to some extent, be generalized to braided triples as in Theorem (4.1.1). This more
general case however needs some additional structure and will be dealt with elsewhere.
Theorem 4.5.1. Let (H, (δ, σ)) be a braided Hopf algebra in a braided abelian monoidal
category c, where (δ, σ) is a BMPI. The following deﬁnes a paracocyclic object in c:
C
0
(H) = I and C
n
(H) = H
n
, n ≥ 1
δ
i
=
_
¸
¸
¸
¸
_
¸
¸
¸
¸
_
(η, 1, 1, ..., 1) i = 0
(1, 1, ..., 1, ∆
i−th
, 1, 1, ...1) 1 ≤ i ≤ n −1
(1, 1, ..., 1, σ) i = n
σ
i
= (1, 1, ..., ε
(i+1)−th
, 1, 1..., 1), 0 ≤ i ≤ n
τ
n
=
_
_
_
id
I
n = 0
(m
n
)(∆
n−1
¯
S, 1
H
n−1
, σ) n ,= 0
Here by m
n
we mean, m
1
= m, and for n ≥ 2:
m
n
= m
H
n = (m, m, ..., m
. ¸¸ .
n times
)T
n
(ψ),
where
T
n
(ψ) :=
n−1
j=1
(1
H
j
, ψ, ψ, ..., ψ
. ¸¸ .
n−j times
, 1
H
j
).
Proof. Among all relations in a paracocyclic object, only the relations
τ
n
σ
0
= σ
n
τ
2
n+1
, τ
n
σ
i
= σ
i−1
τ
n+1
,
and others involving the cyclic operator τ
n
are not obvious because the braiding map
is involved. Here we give a detailed proof of the ﬁrst formula in degree n = 2, i.e. we
prove that τ
2
σ
0
= σ
2
τ
2
3
.
96
In the following, in addition to our previous conventions, we shall write (3, ψ, 2)
for (1, 1, 1, ψ, 1, 1), ψ
23
for ψ
H
2
⊗H
3
, and so on. We have:
σ
2
τ
2
3
= (1, 1, ε)(m
3
)(∆
2
¯
S, 1, 1, σ)(m
3
)(∆
2
¯
S, 1, 1, σ)
= (1, 1, ε)(m, m, m)(1, ψ, 3)(3, ψ, 1)(2, ψ, 2)((S, S,
¯
S)(1, ψ)(ψ, 1)(1, ψ)∆
2
, 1, 1, σ)
(m, m, m)(1, ψ, 3)(3, ψ, 1)(2, ψ, 2)((S, S,
¯
S)(1, ψ)(ψ, 1)(1, ψ)∆
2
, 1, 1, σ)
= (m, m, εm)(1, ψ, 3)(3, ψ, 1)(2, ψ, 2)(S, S,
¯
S, 1, 1, σ)(1, ψ, 2)(ψ, 3)(1, ψ, 2)
(∆
2
m, m, m)(S, 1, S, 1,
¯
S, σ)(1, ψ, 2)(3, ψ)(2, ψ, 1)(1, ψ, 2)(ψ, 3)(1, ψ, 2)(∆
2
, 2)
= (m, m, ε, ε)(S, 1, S, 1,
¯
S, σ)(1, ψ, 2)(3, ψ)(2, ψ, 1)(1, ψ, 2)(ψ, 3)(1, ψ, 2)
((m, m, m)(1, ψ, 3)(3, ψ, 1)(2, ψ, 2)(∆
2
, ∆
2
), m, m)(S, 1, S, 1,
¯
S, σ)
(1, ψ, 2)(3, ψ)(2, ψ, 1)(1, ψ, 2)(ψ, 3)(1, ψ, 2)(∆
2
, 2)
= (m, m)(S, 1, S, 1, ε
¯
S, εσ)(1, ψ, 2)(3, ψ)(2, ψ, 1)(1, ψ, 2)(ψ, 3)(1, ψ, 2)
(m, m, m, m, m)(1, ψ, 7)(3, ψ, 5)(2, ψ, 6)(∆
2
, ∆
2
, 4)(S, 1, S, 1,
¯
S, σ)(1, ψ, 2)(3, ψ)
(2, ψ, 1)(1, ψ, 2)(ψ, 3)(1, ψ, 2)(∆
2
, 2)
= (m, m)(S, 1, S, 1, δ)(1, ψ, 2)(3, ψ)(2, ψ, 1)(1, ψ, 2)(ψ, 3)(1, ψ, 2)
(m, m, m, m, m)(1, ψ, 7)(3, ψ, 5)(2, ψ, 6)(∆
2
S, ∆
2
, S, 1,
¯
S, σ)(1, ψ, 2)(3, ψ)
(2, ψ, 1)(1, ψ, 2)(ψ, 3)(1, ψ, 2)(∆
2
, 2)
= (m, m)(S, 1, S, 1, δ)(m, m, m, m, m)(2, ψ
2,2
, 4)(6, ψ
2,2
)(4, ψ
2,2
, 2)(2, ψ
2,2
, 4)
(ψ
2,2
, 6)(2, ψ
2,2
, 4)(1, ψ, 7)(3, ψ, 5)(2, ψ, 6)(S, S, S, 3, S, 1,
¯
S, σ)(1, ψ, 6)
(ψ, 7)(1, ψ, 6)(∆
2
, ∆
2
, 3)(1, ψ, 2)(3, ψ)(2, ψ, 1)(1, ψ, 2)(ψ, 3)(1, ψ, 2)(∆
2
, 2)
= (m, m)(Sm, m, Sm, m, δm)(2, ψ
2,2
, 4)(6, ψ
2,2
)(4, ψ
2,2
, 2)(2, ψ
2,2
, 4)(ψ
2,2
, 6)
(2, ψ
2,2
, 4)(S, 1, S, 1, S, 1, S, 1,
¯
S, σ)(1, ψ, 6)(3, ψ, 4)(2, ψ, 5)(1, ψ, 6)
(ψ, 7)(1, ψ, 6)(∆
2
, ∆
2
, 3)(1, ψ, 2)(3, ψ)(2, ψ, 1)(1, ψ, 2)(ψ, 3)(1, ψ, 2)(∆
2
, 2)
= (m, m)(mψ, m, mψ, m)(S, S, 1, 1, S, S, 1, 1, δ, δ)(2, ψ
2,2
, 4)(6, ψ
2,2
)(4, ψ
2,2
, 2)
(2, ψ
2,2
, 4)(ψ
2,2
, 6)(2, ψ
2,2
, 4)(S, 1, S, 1, S, 1, S, 1,
¯
S, σ)(1, ψ, 6)(3, ψ, 4)(2, ψ, 5)
(1, ψ, 6)(ψ, 7)(1, ψ, 6)(∆
2
, ∆
2
, 3)(1, ψ, 2)(3, ψ)(2, ψ, 1)(1, ψ, 2)(ψ, 3)(1, ψ, 2)(∆
2
, 2)
= (m, m)(mψ, m, mψ, m)(2, ψ
2,2
, 2)(ψ
2,2
, 4)(δ, δ, S, S, S, S, 4)
97
(S, 1, S, 1, S, 1, S, 1,
¯
S, σ)(1, ψ, 6)(3, ψ, 4)(2, ψ, 5)(1, ψ, 6)(ψ, 7)(1, ψ, 6)
(∆
2
, ∆
2
, 3)(1, ψ, 2)(3, ψ)(2, ψ, 1)(1, ψ, 2)(ψ, 3)(1, ψ, 2)(∆
2
, 2)
= (m, m)(1, ψ, 1)(ψ, 2)(mψ, mψ, m, m)(δS, δ, S
2
, S, S
2
, S, S, 1,
¯
S, σ)
(1, ψ, 6)(3, ψ, 4)(2, ψ, 5)(1, ψ, 6)(ψ, 7)(1, ψ, 6)(∆
2
, ∆
2
, 3)(1, ψ, 2)(3, ψ)(2, ψ, 1)
(1, ψ, 2)(ψ, 3)(1, ψ, 2)(∆
2
, 2)
= (m, m)(1, ψ, 1)(ψ, 2)(mψ, mψ, m, m)(1, ψ, 5)(ψ, 6)(S
2
, S
2
, δS, δ, S, S, S, 1,
¯
S, σ)
(∆
2
, ∆
2
, 3)(1, ψ, 2)(3, ψ)(2, ψ, 1)(1, ψ, 2)(ψ, 3)(1, ψ, 2)(∆
2
, 2)
= (m, m)(1, ψ, 1)(ψ, 2)(mψ, mψ, m, m)(1, ψ, 5)(ψ, 6)(S
2
, S
2
, δS, δ, S, S, S, 1,
¯
S, σ)
(3, ∆
2
, 3)(3, ψ, 2)(5, ψ)(4, ψ, 1)(3, ψ, 2)(∆
2
, 4)(ψ, 3)(1, ψ, 2)(∆
2
, 2)
= (m, m)(1, ψ, 1)(ψ, 2)(mψ, mψ, m, m)(1, ψ, 5)(ψ, 6)(S
2
, S
2
, δS, δ, S, S, S, 1,
¯
S, σ)
(3, ∆
2
, 3)(3, ψ, 2)(5, ψ)(4, ψ, 1)(3, ψ, 2)(ψ
13
, 3)(1, ψ
13
, 2)(2, ∆
2
, 2)(∆
2
, 2)
= (m, m)(1, ψ, 1)(ψ, 2)(mψ, mψ, m, m)(1, ψ, 5)(ψ, 6)(S
2
, S
2
, δS, δ, S, S, S, 1,
¯
S, σ)
(3, ψ
13
, 2)(7, ψ)(4, ψ
13
, 1)(3, ψ, 4)(ψ
13
, 3)(1, ψ
13
, 2)(5, ∆
2
, 1)(2, ∆
2
, 2)(∆
2
, 2)
= (m, m)(1, ψ, 1)(ψ, 2)(mψ, mψ, m, m)(1, ψ, 5)(ψ, 6)(2, ψ
12
, 3)(5, ψ, 3)(3, ψ
12
, 2)
(2, ψ, 4)(ψ
12
, 5)(1, ψ
12
, 4)(
¯
S, S, S
2
, S
2
, δS, δ, S, S, 1, σ)(5, ∆
2
, 1)(2, ∆
2
, 2)(∆
2
, 2)
= (m, m)(1, ψ, 1)(ψ, 2)(mψ, mψ, m, m)(1, ψ, 5)(ψ, 6)(S
2
, S
2
, δ, S, S, S, 1,
¯
S, δS, σ)
(2, ψ
13
, 3)(6, ψ
21
)(3, ψ
23
, 1)(2, ψ, 5)(ψ
12
, 6)(1, ψ
12
, 5)(5, ∆
2
, 1)(2, ∆
2
, 2)(∆
2
, 2)
= (m, m)(1, ψ, 1)(ψ, 2)(mψ, mψ, m, m)(1, ψ, 5)(ψ, 6)(S
2
, S
2
, δ, S, S, S, 1,
¯
S, σ, δS)
(4, ψ, 3)(3, ψ, 4)(2, ψ, 5)(6, ψ, 1)(7, ψ)(5, ψ, 2)(6, ψ, 1)(4, ψ, 3)(5, ψ, 2)
(3, ψ, 4)(4, ψ, 3)(2, ψ, 5)(1, ψ, 6)(ψ, 7)(2, ψ, 5)(1, ψ, 6)(5, ∆
2
, 1)(2, ∆
2
, 2)(∆
2
, 2)
= (m, m)(1, ψ, 1)(ψ, 2)(mψ, mψ, m, m)(1, ψ, 5)(ψ, 6)(S
2
, S
2
, δ, S, S, S, 1,
¯
S, δS, σ)
(4, ψ, 3)(3, ψ, 4)(2, ψ, 5)(6, ψ, 1)(5, ψ, 2)(4, ψ, 3)(3, ψ, 4)(7, ψ)(6, ψ, 1)(5, ψ, 2)
(4, ψ, 3)(2, ψ, 5)(1, ψ, 6)(ψ, 7)(2, ψ, 5)(1, ψ, 6)(5, ∆
2
, 1)(2, ∆
2
, 2)(∆
2
, 2)
= (m, m)(1, ψ, 1)(ψ, 2)(mψ, mψ, m, m)(1, ψ, 5)(ψ, 6)(S
2
, S
2
, δ, S, S, S, 1,
¯
S, δS, σ)
(4, ψ, 3)(3, ψ, 4)(2, ψ, 5)(6, ψ, 1)(5, ψ, 2)(4, ψ, 3)(3, ψ, 4)(2, ψ, 5)(1, ψ, 6)(ψ, 7)
(2, ψ, 5)(1, ψ, 6)(7, ψ)(6, ψ, 1)(5, ψ, 2)(4, ψ, 3)(5, ∆
2
, 1)(2, ∆
2
, 2)(∆
2
, 2)
= (m, m)(1, ψ, 1)(ψ, 2)(m, m, m, m)(ψ, 6)(2, ψ, 4)(1, ψ, 5)(ψ, 6)(3, ψ, 3)(2, ψ, 4)
(5, ψ, 1)(4, ψ, 2)(3, ψ, 3)(2, ψ, 4)(1, ψ, 5)(S
2
,
¯
S, S
2
, S, δ, S, S, 1, σ, δS)(ψ, 7)(2, ψ, 5)
98
(1, ψ, 6)(7, ψ)(6, ψ, 1)(5, ψ, 2)(4, ψ, 3)(5, ∆
2
, 1)(2, ∆
2
, 2)(∆
2
, 2)
= (m, m)(1, ψ, 1)(ψ, 2)(m, m, m, m)(ψ, 6)(2, ψ, 4)(1, ψ, 5)(3, ψ, 3)(2, ψ, 4)(5, ψ, 1)
(4, ψ, 2)(3, ψ, 3)(2, ψ, 4)(ψ, 6)(1, ψ, 5)(S
2
, (δ, S)∆, S
2
, S, δ, S, S, 1, σ, δS)
(ψ, 7)(2, ψ, 5)(1, ψ, 6)(7, ψ)(6, ψ, 1)(5, ψ, 2)(4, ψ, 3)(5, ∆
2
, 1)(2, ∆
2
, 2)(∆
2
, 2)
= (m, m)(1, ψ, 1)(ψ, 2)(m, m, m, m)(ψ, 6)(2, ψ, 4)(1, ψ, 5)(3, ψ, 3)(2, ψ, 4)(5, ψ, 1)
(4, ψ, 2)(3, ψ, 3)(2, ψ, 4)(ψ, 6)(1, ψ, 5)(S
2
, δ, S, S
2
, S, δ, S, S, 1, σ, δS)(1, ∆, 7)
(ψ, 7)(2, ψ, 5)(1, ψ, 6)(7, ψ)(6, ψ, 1)(5, ψ, 2)(4, ψ, 3)(5, ∆
2
, 1)(2, ∆
2
, 2)(∆
2
, 2)
= (1, m)(m, 2)(ψ
12
, 1)(m, m, m, m)(ψ, 6)(2, ψ, 4)(1, ψ, 5)(3, ψ, 3)(2, ψ, 4)
(5, ψ, 1)(4, ψ, 2)(3, ψ, 3)(2, ψ, 4)(ψ, 6)(1, ψ, 5)(S
2
, δ, S, S
2
, S, δ, S, S, 1, σ, δS)
(ψ
21
, 7)(3, ψ, 5)(2, ψ, 6)(8, ψ)(7, ψ, 1)(6, ψ, 2)(5, ψ, 3)(∆, 8)
(5, ∆, 2)(5, ∆, 1)(2, ∆, 3)(2, ∆, 2)(∆, 3)(∆, 2)
= (1, m)(ψ, 1)(1, m, 1)(m, 2, m)(2, m, 3)(4, m, 2)(ψ, 6)(2, ψ, 4)(1, ψ, 5)(3, ψ, 3)
(2, ψ, 4)(5, ψ, 1)(4, ψ, 2)(3, ψ, 3)(2, ψ, 4)(ψ, 6)(1, ψ, 5)(S
2
, δ, S, S
2
, S, δ, S, S, 1, σ, δS)
(ψ
21
, 7)(3, ψ, 5)(2, ψ, 6)(8, ψ)(7, ψ, 1)(5, ψ
12
, 2)(6, ∆, 2)(∆, 7)(5, ∆, 1)
(2, ∆, 3)(2, ∆, 2)(∆, 3)(∆, 2)
= (1, m)(ψ, 1)(m, 1, m)(2, m, 2)(2, m, 3)(4, m, 2)(ψ, 6)(2, ψ, 4)(1, ψ, 5)(3, ψ, 3)
(2, ψ, 4)(5, ψ, 1)(4, ψ, 2)(3, ψ, 3)(2, ψ, 4)(ψ, 6)(1, ψ, 5)
(S
2
, δ, S, S
2
, S, δ, S, S, 1, σ, δS)(ψ
21
, 7)(3, ψ, 5)(2, ψ, 6)(8, ψ)(7, ψ, 1)(5, ψ
12
, 2)
(6, ∆, 2)(∆, 7)(5, ∆, 1)(2, ∆, 3)(2, ∆, 2)(∆, 3)(∆, 2)
= (1, m)(ψ, 1)(m, 1, m)(2, m, 2)(3, m, 2)(4, m, 2)(ψ, 6)(2, ψ, 4)(1, ψ, 5)(3, ψ, 3)
(2, ψ, 4)(5, ψ, 1)(4, ψ, 2)(3, ψ, 3)(2, ψ, 4)(ψ, 6)(1, ψ, 5)(S
2
, δ, S, S
2
, S, δ, S, S, 1, σ, δS)
(ψ, 8)(5, ∆, 3)(1, ψ, 6)(3, ψ, 4)(2, ψ, 5)(7, ψ)(6, ψ, 1)(5, ψ, 2)(∆, 7)
(5, ∆, 1)(2, ∆, 3)(2, ∆, 2)(∆, 3)(∆, 2)
= (1, m)(ψ, 1)(m, 1, m)(2, m, 2)(3, m, 2)(3, m, 3)(ψ, 6)(2, ψ, 4)(1, ψ, 5)(3, ψ, 3)
(2, ψ, 4)(5, ψ, 1)(4, ψ, 2)(3, ψ, 3)(2, ψ, 4)(ψ, 6)(1, ψ, 5)
(δ, S
2
, S, S
2
, S, (δ, S)∆, S, 1, σ, δS)(1, ψ, 6)(3, ψ, 4)(2, ψ, 5)(7, ψ)(6, ψ, 1)(5, ψ, 2)
(∆, 7)(5, ∆, 1)(2, ∆, 3)(2, ∆, 2)(∆, 3)(∆, 2)
= (1, m)(ψ, 1)(m, 1, m)(2, m, 2)(3, m, 2)(3, m, 3)(ψ, 6)(2, ψ, 4)(1, ψ, 5)(3, ψ, 3)
99
(2, ψ, 4)(5, ψ, 1)(4, ψ, 2)(3, ψ, 3)(2, ψ, 4)(ψ, 6)(1, ψ, 5)(ψ, 6)
(δ, S, S
2
, S
2
, S,
¯
S, S, 1, σ, δS)(3, ψ, 4)(2, ψ, 5)(7, ψ)(6, ψ, 1)(5, ψ, 2)
(∆, 7)(5, ∆, 1)(2, ∆, 3)(2, ∆, 2)(∆, 3)(∆, 2)
= (1, m)(ψ, 1)(m, 1, m)(2, m, 2)(3, m, 2)(ψ, 5)(3, m, 3)(2, ψ, )(3, ψ, 3)(1, ψ, 5)
(2, ψ, 4)(5, ψ, 1)(4, ψ, 2)(3, ψ, 3)(2, ψ, 4)(1, ψ, 5)(ψ, 6)(1, ψ, 5)
(δ, S, S
2
, S
2
, S,
¯
S, S, 1, σ, δS)(3, ψ, 4)(2, ψ, 5)(7, ψ)(6, ψ, 1)(5, ψ, 2)
(∆, 7)(5, ∆, 1)(2, ∆, 3)(2, ∆, 2)(∆, 3)(∆, 2)
= (1, m)(ψ, 1)(m, 1, m)(2, m, 2)(3, m, 2)(ψ, 5)(3, m, 3)(2, ψ
21
, 3)(1, ψ
21
, 4)(5, ψ, 1)
(4, ψ, 2)(3, ψ, 3)(1, ψ
12
, 4)(ψ, 6)(δ, S, S
2
, S
2
, S,
¯
S, S, 1, σ, δS)(2, ψ, 5)(3, ψ, 4)
(2, ψ, 5)(7, ψ)(6, ψ, 1)(5, ψ, 2)(∆, 7)(5, ∆, 1)(2, ∆, 3)(2, ∆, 2)(∆, 3)(∆, 2)
= (1, m)(ψ, 1)(m, 1, m)(2, m, 2)(3, m, 2)(ψ, 5)(2, ψ, 3)(1, ψ, 4)(1, m, 5)(5, ψ, 1)(4, ψ, 2)
(3, ψ, 3)(1, ψ
12
, 4)(ψ, 6)(δ, S, S
2
, S
2
, S,
¯
S, S, 1, σ, δS)(3, ψ, 4)(2, ψ, 5)(3, ψ, 4)(7, ψ)
(6, ψ, 1)(5, ψ, 2)(∆, 7)(5, ∆, 1)(2, ∆, 3)(2, ∆, 2)(∆, 3)(∆, 2)
= (1, m)(ψ, 1)(m, 1, m)(2, m, 2)(3, m, 2)(ψ, 5)(2, ψ, 3)(1, ψ, 4)(4, ψ, 1)(3, ψ, 2)(2, ψ, 3)
(1, m, 5)(1, ψ
12
, 4)(ψ, 6)(δ, S, S
2
, S
2
, S,
¯
S, S, 1, σ, δS)(3, ψ, 4)(2, ψ, 5)(3, ψ, 4)(7, ψ)
(6, ψ, 1)(5, ψ, 2)(∆, 7)(5, ∆, 1)(2, ∆, 3)(2, ∆, 2)(∆, 3)(∆, 2)
= (1, m)(ψ, 1)(m, 1, m)(2, m, 2)(3, m, 2)(ψ, 5)(2, ψ, 3)(1, ψ, 4)(4, ψ, 1)(3, ψ, 2)(2, ψ, 3)
(1, ψ, 4)(2, m, 4)(ψ, 6)(δ, S, S
2
, S
2
, S,
¯
S, S, 1, σ, δS)(3, ψ, 4)(2, ψ
21
, 4)(7, ψ)
(6, ψ, 1)(5, ψ, 2)(∆, 7)(5, ∆, 1)(2, ∆, 3)(2, ∆, 2)(1, ∆, 2)(∆, 2)
= (1, m)(ψ, 1)(m, 1, m)(2, m, 2)(3, m, 2)(ψ, 5)(2, ψ, 3)(1, ψ, 4)(4, ψ, 1)(3, ψ, 2)(2, ψ, 3)
(1, ψ, 4)(ψ, 5)(2, m, 4)(δ, S, S
2
, S
2
, S,
¯
S, S, 1, σ, δS)(3, ψ, 4)(2, ψ
21
, 4)(7, ψ)
(6, ψ, 1)(5, ψ, 2)(∆, 7)(5, ∆, 1)(1, ∆, 4)(1, ∆, 3)(1, ∆, 2)(∆, 2)
= (1, m)(ψ, 1)(m, 1, m)(2, m, 2)(3, m, 2)(ψ, 5)(2, ψ, 3)(1, ψ, 4)(4, ψ, 1)(3, ψ, 2)(2, ψ, 3)
(1, ψ, 4)(ψ, 5)(2, m, 4)(δ, S, S
2
, S
2
, S,
¯
S, S, 1, σ, δS)(3, ψ, 4)(2, ψ
21
, 4)(2, ∆, 5)
(6, ψ)(5, ψ, 1)(4, ψ, 2)(∆, 6)(4, ∆, 1)(1, ∆, 3)(1, ∆, 2)(∆, 2)
= (1, m)(ψ, 1)(m, 1, m)(2, m, 2)(3, m, 2)(ψ, 5)(2, ψ, 3)(1, ψ, 4)(4, ψ, 1)(3, ψ, 2)(2, ψ, 3)
(1, ψ, 4)(ψ, 5)(2, m, 4)(δ, S, S
2
, S
2
, S,
¯
S, S, 1, σ, δS)(3, ψ, 4)(3, ∆, 4)(2, ψ, 4)
(6, ψ)(5, ψ, 1)(4, ψ, 2)(∆, 6)(4, ∆, 1)(1, ∆, 3)(1, ∆, 2)(∆, 2)
100
= (1, m)(ψ, 1)(m, 1, m)(2, m, 2)(3, m, 2)(ψ, 5)(2, ψ, 3)(1, ψ, 4)(4, ψ, 1)(3, ψ, 2)(2, ψ, 3)
(1, ψ, 4)(ψ, 5)(δ, S, S
2
, m(S
2
, S)ψ∆,
¯
S, S, 1, σ, δS)(2, ψ, 4)
(6, ψ)(5, ψ, 1)(4, ψ, 2)(∆, 6)(4, ∆, 1)(1, ∆, 3)(1, ∆, 2)(∆, 2)
= (1, m)(ψ, 1)(m, 1, m)(2, m, 2)(3, m, 2)(ψ, 5)(2, ψ, 3)(1, ψ, 4)(4, ψ, 1)(3, ψ, 2)(2, ψ, 3)
(1, ψ, 4)(ψ, 5)(δ, S, S
2
, ηε,
¯
S, S, 1, σ, δS)(2, ψ, 4)
(6, ψ)(5, ψ, 1)(4, ψ, 2)(5, ∆, 1)(2, ∆, 3)(2, ∆, 2)(∆, 3)(∆, 2)
= (1, m)(ψ, 1)(m, 1, m)(2, m, 2)(3, m, 2)(ψ, 5)(2, ψ, 3)(1, ψ, 4)(4, ψ, 1)(3, ψ, 2)(2, ψ, 3)
(1, ψ, 4)(ψ, 5)(2, η, 4)(δ, S, S
2
,
¯
S, S, 1, σ, δS)(3, ε, 4)(2, ψ, 4)
(6, ψ)(5, ψ, 1)(4, ψ, 2)(5, ∆, 1)(2, ∆, 3)(2, ∆, 2)(∆, 3)(∆, 2)
= (1, m)(ψ, 1)(m, 1, m)(2, m, 2)(3, m, 2)(3, η, 3)(ψ, 4)(3, ψ, 1)(2, ψ, 2)(1, ψ, 3)(ψ, 4)
(δ, S, S
2
,
¯
S, S, 1, σ, δS)(5, ψ)(4, ψ, 1)(3, ψ, 2)(2, ε, 5)(5, ∆, 1)
(2, ∆, 3)(2, ∆, 2)(∆, 3)(∆, 2)
= (1, m)(ψ, 1)(m, 1, m)(2, m, 2)(3, m(η, 1), 2)(ψ, 4)(3, ψ, 1)(2, ψ, 2)(1, ψ, 3)(ψ, 4)
(δ, S, S
2
,
¯
S, S, 1, σ, δS)(5, ψ)(4, ψ, 1)(3, ψ, 2)(4, ∆, 1)
(2, (ε, 1)∆, 3)(2, ∆, 2)(∆, 3)(∆, 2)
= (1, m)(ψ, 1)(m, 1, m)(2, m, 2)(ψ, 4)(3, ψ, 1)(2, ψ, 2)(1, ψ, 3)(ψ, 4)
(δ, S, S
2
,
¯
S, S, 1, σ, δS)(5, ψ)(4, ψ, 1)(3, ψ, 2)(4, ∆, 1)(2, ∆, 2)(∆, 3)(∆, 2)
= (1, m)(1, m, m)(ψ
21
, 1)(2, m, 2)(3, ψ, 1)(2, ψ, 2)(ψ, 4)(1, ψ, 3)(ψ, 4)
(δ, S, S
2
,
¯
S, S, 1, σ, δS)(5, ψ)(4, ψ, 1)(3, ψ, 2)(4, ∆, 1)(2, ∆, 2)(1, ∆, 2)(∆, 2)
= (1, m)(1, m, m)(m, 4)(ψ
22
, 1)(3, ψ, 1)(2, ψ, 2)(δ,
¯
S, S
2
, S, S, 1, σ, δS)
(1, ψ, 4)(2, ψ, 3)(1, ψ, 4)(5, ψ)(4, ψ, 1)(3, ψ, 2)(4, ∆, 1)(1, ∆, 3)(1, ∆, 2)(∆, 2)
= (m, 1)(2, m)(2, m, m)(1, ψ, 3)(2, ψ, 2)(ψ, 4)(1, ψ, 3)(3, ψ, 1)(2, ψ, 2)
(δ,
¯
S, S
2
, S, S, 1, σ, δS)(1, ψ, 4)(2, ψ, 3)(1, ψ, 4)(5, ψ)(4, ψ, 1)(3, ψ, 2)
(1, ∆, 4)(3, ∆, 1)(1, ∆, 2)(∆, 2)
= (m, 1)(2, m)(2, m, 1)(4, m)(1, ψ, 3)(2, ψ, 2)(3, ψ, 1)(ψ, 4)(1, ψ, 3)(2, ψ, 2)
(δ,
¯
S, S
2
, S, S, 1, σ, δS)(1, ψ, 4)(2, ψ, 3)(1, ψ, 4)(5, ψ)(4, ψ, 1)(3, ψ, 2)(1, ∆, 4)
(3, ∆, 1)(1, ∆, 2)(∆, 2)
= (m, 1)(2, m)(3, m)(3, m, 1)(1, ψ
31
, 1)(ψ, 4)(1, ψ
21
, 3)(δ,
¯
S, S
2
, S, S, 1, σ, δS)
101
(2, ψ, 3)(1, ψ, 4)(2, ψ, 3)(1, ∆, 4)(4, ψ)(3, ψ, 1)(2, ψ, 2)(3, ∆, 1)(1, ∆, 2)(∆, 2)
= (m, 1)(2, m)(3, m)(1, ψ
21
, 1)(2, m, 2)(ψ, 4)(1, ψ
21
, 3)(δ,
¯
S, S
2
, S, S, 1, σ, δS)
(2, ψ, 3)(1, ψ
21
, 3)(1, ∆, 4)(4, ψ)(3, ψ, 1)(2, ψ, 2)(3, ∆, 1)(1, ∆, 2)(∆, 2)
= (m, 1)(2, m)(3, m)(1, ψ
21
, 1)(ψ, 3)(1, ψ, 3)(1, m, 3)(δ,
¯
S, S
2
, S, S, 1, σ, δS)
(2, ψ, 3)(2, ∆, 3)(1, ψ, 3)(4, ψ)(3, ψ, 1)(2, ψ, 2)(3, ∆, 1)(1, ∆, 2)(∆, 2)
= (m, 1)(2, m)(3, m)(1, ψ
21
, 1)(ψ, 3)(1, ψ, 3)(δ,
¯
S, m(S
2
, S)ψ∆, S, 1, σ, δS)
(1, ψ, 3)(4, ψ)(3, ψ, 1)(2, ψ, 2)(3, ∆, 1)(1, ∆, 2)(∆, 2)
= (m, 1)(2, m)(3, m)(1, ψ, 2)(2, ψ, 1)(ψ, 3)(1, ψ, 3)(δ,
¯
S, ηε, S, 1, σ, δS)
(1, ψ, 3)(4, ψ)(3, ψ, 1)(2, ψ, 2)(3, ∆, 1)(1, ∆, 2)(∆, 2)
= (m, 1)(2, m)(3, m)(1, ψ, 2)(2, ψ, 1)(ψ, 3)(1, ψ, 3)(1, η, 3)(δ,
¯
S, S, 1, σ, δS)
(2, ε, 3)(1, ψ, 3)(4, ψ)(3, ψ, 1)(2, ψ, 2)(3, ∆, 1)(1, ∆, 2)(∆, 2)
= (m, 1)(2, m)(3, m)(3, η, 1)(1, ψ, 1)(ψ, 2)(δ,
¯
S, S, 1, σ, δS)
(3, ψ)(2, ψ, 1)(1, ψ, 2)(2, ∆, 1)(1, ε, 3)(1, ∆, 2)(∆, 2)
= (m, 1)(2, m)(3, m(η, 1))(1, ψ, 1)(ψ, 2)(δ,
¯
S, S, 1, σ, δS)
(3, ψ)(2, ψ, 1)(1, ψ, 2)(2, ∆, 1)(1, (ε, 1)∆, 2)(∆, 2)
= (m, 1)(2, m)(1, ψ, 1)(ψ, 2)(δ,
¯
S, S, 1, δS, σ)(3, ψ)(2, ψ, 1)(1, ψ, 2)(2, ∆, 1)(∆, 2)
= (m, 1)(2, m)(1, ψ, 1)(ψ, 2)(δ, δS,
¯
S, S, 1, σ)(2, ∆, 1)(∆, 2)
= (m, 1)(2, m)(1, ψ, 1)(ψ, 2)(δ, δ, 4)(1, S,
¯
S, S, 1, σ)(2, ∆, 1)(∆, 2)
= (m, 1)(2, m)(1, ψ, 1)(ψ, 2)(δm, 4)(1, S,
¯
S, S, 1, σ)(∆, 3)(1, ∆, 1)
= (m, 1)(2, m)(1, ψ, 1)(ψ, 2)(δm(1, S)∆,
¯
S, S, 1, σ)(1, ∆, 1)
= (m, 1)(2, m)(1, ψ, 1)(ψ, 2)(δηε,
¯
S, S, 1, σ)(1, ∆, 1)
= (m, 1)(2, m)(1, ψ, 1)(ψ, 2)(ε,
¯
S, S, 1, σ)(1, ∆, 1)
= (m, 1)(2, m)(1, ψ, 1)(ψ, 2)(
¯
S, S, 1, σ)(ε, 3)(1, ∆, 1)
= (m, m)(1, ψ, 1)(ψ, 2)(
¯
S, S, 1, σ)(∆, 1)(ε, 2)
= (m, m)(1, ψ, 1)(ψ(
¯
S, S)∆, 1, σ)(ε, 2)
= (m, m)(1, ψ, 1)(∆
¯
S, 1, σ)(ε, 2)
= τ
2
σ
0
102
Theorem 4.5.2. Under the conditions of Theorem (4.5.1),
τ
3
2
= ψ
2
H,H
.
Proof. We have:
τ
3
2
= τ
2
τ
2
2
= τ
2
(m
2
)(∆
¯
S, 1, σ)(m
2
)(∆
¯
S, 1, σ)
= τ
2
(m, m)(1, ψ, 1)(ψ(
¯
S, S)∆, 1, σ)(m, m)(1, ψ, 1)(ψ(
¯
S, S)∆, 1, σ)
= τ
2
(m, m)(1, ψ, 1)(ψ, 2)(
¯
S, S, 1, σ)(∆, 1)(m, m)(1, ψ, 1)(ψ, 2)(
¯
S, S, 1, σ)(∆, 1)
= τ
2
(m, m)(1, ψ, 1)(ψ, 2)(
¯
S, S, 1, σ)(m, m, m)(1, ψ, 3)(∆, ∆, 2)(S, 1,
¯
S, σ)(1, ψ)
(ψ, 1)(∆, 1)
= τ
2
(m, m)(1, ψ, 1)(ψ, 2)(
¯
Sm, Sm, m, σ)(1, ψ, 3)(∆S, ∆,
¯
S, σ)(1, ψ)(ψ, 1)(∆, 1)
= τ
2
(m, m)(1, ψ, 1)(ψ, 2)(m(
¯
S,
¯
S)ψ, m(S, S)ψ, m, σ)(1, ψ, 3)((S, S)ψ∆, ∆,
¯
S, σ)
(1, ψ)(ψ, 1)(∆, 1)
= τ
2
(m, m)(1, ψ, 1)(ψ, 2)(m, m, m, σ)(
¯
S,
¯
S, S, S, 1, 1)(ψ, 4)(2, ψ, 2)(1, ψ, 3)
(S, S, 1, 1,
¯
S, σ)(ψ, 3)(∆, ∆, 1)(1, ψ)(ψ, 1)(∆, 1)
= τ
2
(m, m)(1, ψ, 1)(ψ, 2)(m, m, m, σ)(
¯
S,
¯
S, S, S, 1, 1)(ψ, 4)(2, ψ, 2)(1, ψ, 3)
(S, S, 1, 1,
¯
S, σ)(ψ, 3)(∆, ∆, 1)ψ
12
(∆, 1)
= τ
2
(m, m)(1, ψ, 1)(ψ, 2)(m, m, m, σ)(
¯
S,
¯
S, S, S, 1, 1)(ψ, 4)(2, ψ, 2)(1, ψ, 3)
(S, S, 1, 1,
¯
S, σ)(ψ, 3)ψ
14
(1, ∆, ∆)(∆, 1)
= τ
2
(m, m)(1, ψ, 1)(ψ, 2)(m, m, m, σ)(
¯
S,
¯
S, S, S, 1, 1)(1, S, 1, S,
¯
S, σ)
(ψ, 3)(2, ψ, 1)(1, ψ, 2)(ψ, 3)ψ
14
((1, ∆)∆, ∆)
= τ
2
(1, m)(m, 2)(ψ
12
, 1)(m, m, m, σ)(
¯
S,
¯
SS, S, S
2
,
¯
S, σ)(ψ, 3)(2, ψ, 1)
(1, ψ, 2)(ψ, 3)ψ
14
((∆, 1)∆, ∆)
= τ
2
(1, m)(ψ, 1)(1, m, 1)(m, m, m, σ)(
¯
S,
¯
SS, S, S
2
,
¯
S, σ)(ψ, 3)(2, ψ, 1)
(1, ψ, 2)(ψ, 3)ψ
14
(∆, 3)(∆, 2)(1, ∆)
103
= τ
2
(1, m)(ψ, 1)(m, m(m, m), σ)(
¯
S,
¯
SS, S, S
2
,
¯
S, σ)(ψ, 3)(2, ψ, 1)(1, ψ, 2)
(ψ, 3)(3, ψ)(2, ψ, 1)(1, ψ, 2)(ψ, 3)(∆, 3)(∆, 2)(1, ∆)
= τ
2
(1, m)(ψ, 1)(m, (m(m, 1)(1, m, 1)), σ)(
¯
S,
¯
SS, S, S
2
,
¯
S, σ)(ψ, 3)(2, ψ, 1)
(3, ψ)(1, ψ, 2)(ψ, 3)(2, ψ, 1)(1, ψ, 2)(ψ, 3)(∆, 3)(∆, 2)(1, ∆)
= τ
2
(1, m)(ψ, 1)(m, 1, σ)(2, m(m, 1))(3, m, 1)(
¯
S,
¯
SS, S, S
2
,
¯
S, σ)(ψ, 3)
(2, ψ
21
)(ψ
22
, 1)(ψ∆, 3)(∆, 2)(1, ∆)
= τ
2
(1, m)(ψ, 1)(m, 1, σ)(2, m(m, 1))(3, m, 1)(
¯
S,
¯
SS, S, S
2
,
¯
S, σ)(3, ψ∆)
(ψ, 2)(2, ψ)(ψ
12
, 1)(∆, 2)(1, ∆)
= τ
2
(1, m)(ψ, 1)(m, 1, σ)(2, m(m, 1))(
¯
S,
¯
SS, S, m(S
2
,
¯
S)ψ∆, σ)(ψ, 2)(2, ψ)
(ψ
12
, 1)(∆, 2)(1, ∆)
= τ
2
(1, m)(ψ, 1)(m, 1, σ)(2, m(m, 1))(
¯
S,
¯
SS, S, ηδ, σ)(ψ, 2)
(2, ψ)(ψ
12
, 1)(∆, 2)(1, ∆)
= τ
2
(1, m)(ψ, 1)(m, 1, σ)(2, m(m, 1))(3, η, 1)(
¯
S,
¯
SS, S, σ)(3, δ)(ψ, 2)(2, ψ)
(ψ
12
, 1)(∆, 2)(1, ∆)
= τ
2
(1, m)(ψ, 1)(m, 1, σ)(2, m(m(1, η), 1))(
¯
S,
¯
SS, S, σ)(ψ, 1)(δ, 3)(∆, 2)(1, ∆)
= τ
2
(1, m)(ψ, 1)(m, 1, σ)(2, m)(ψ, 2)(
¯
SS,
¯
S, S, σ)(δ, 3)(∆, 2)(1, ∆)
= τ
2
(1, m)(ψ, 1)(m, m, σ)(ψ, 2)(
¯
S(δ, S)∆,
¯
S, S, σ)(1, ∆)
= τ
2
(1, m)(m, m, σ)(ψ
22
)(ψ, 2)(
¯
S
2
,
¯
S, S, σ)(1, ∆)
= τ
2
(1, m)(m, m, 1)(2, ψ, 1)(ψ
22
, 1)(
¯
S
2
,
¯
S, S, σ, σ)(1, ∆)
= τ
2
(m, m(m, 1))(2, ψ, 1)(1, ψ, 2)(ψ, 3)(2, ψ, 1)(1, ψ, 2)(
¯
S
2
,
¯
S, S, σ, σ)(1, ∆)
= τ
2
(m, m(m, 1))(ψ
13
, 1)(2, ψ, 1)(1, ψ, 2)(
¯
S
2
,
¯
S, S, σ, σ)(1, ∆)
= (m, m)(1, ψ, 1)(ψ, 2)(
¯
S, S, 1, σ)(∆, 1)(m, m(m, 1))(ψ
13
, 1)(2, ψ, 1)(1, ψ, 2)
(
¯
S
2
,
¯
S, S, σ, σ)(1, ∆)
= (m, m)(1, ψ, 1)(ψ, 2)(
¯
S, S, 1, σ)(∆m, m(m, 1))(ψ
13
, 1)(2, ψ, 1)(1, ψ, 2)
(
¯
S
2
,
¯
S, S, σ, σ)(1, ∆)
= (m, m)(1, ψ, 1)(ψ, 2)(
¯
S, S, 1, σ)((m, m)(1, ψ, 1)(∆, ∆), m(m, 1))
(ψ
13
, 1)(2, ψ, 1)(1, ψ, 2)(
¯
S
2
,
¯
S, S, σ, σ)(1, ∆)
= (m, m)(1, ψ, 1)(ψ, 2)(
¯
S, S, 1, σ)(m, m, m(m, 1))(1, ψ, 4)(∆, ∆, 3)
104
(ψ
13
, 1)(2, ψ, 1)(1, ψ, 2)(
¯
S
2
,
¯
S, S, σ, σ)(1, ∆)
= (m, m)(1, ψ, 1)(ψ, 2)(
¯
Sm, Sm, m(m, 1), σ)(1, ψ, 4)(ψ
15
, 1)(1, ∆, ∆, 2)
(2, ψ, 1)(1, ψ, 2)(
¯
S
2
,
¯
S, S, σ, σ)(1, ∆)
= (m, m)(1, ψ, 1)(ψ, 2)(mψ(
¯
S,
¯
S), mψ(S, S), m(m, 1), σ)(1, ψ, 4)(ψ
15
, 1)
(3, ψ
12
, 1)(1, ∆, 1, ∆, 1)(1, ψ, 2)(
¯
S
2
,
¯
S, S, σ, σ)(1, ∆)
= (m, m)(1, ψ, 1)(ψ, 2)(mψ, mψ, m(m, 1), 1)(
¯
S,
¯
S, S, S, 3, σ)(1, ψ, 4)(ψ
15
, 1)
(3, ψ
12
, 1)(1, ψ
12
, 3)(1, 1, ∆, ∆, 1)(
¯
S
2
,
¯
S, S, σ, σ)(1, ∆)
= (m, m)(1, ψ, 1)(ψ, 2)(mψ, mψ, m(m, 1), 1)(
¯
S,
¯
S, S, S, 3, σ)(1, ψ, 4)(ψ
15
, 1)
(3, ψ
12
, 1)(1, ψ
12
, 3)(
¯
S
2
,
¯
S, ∆S, ∆σ, σ)(1, ∆)
= (m, m)(1, ψ, 1)(ψ, 2)(mψ, mψ, m(m, 1), 1)(
¯
S,
¯
S, S, S, 3, σ)(1, ψ, 4)(ψ
15
, 1)
(3, ψ
12
, 1)(1, ψ
12
, 3)(
¯
S
2
,
¯
S, (S, S)ψ∆, σ, σ, σ)(1, ∆)
= (m, m)(1, ψ, 1)(ψ, 2)(mψ, mψ, m(m, 1), 1)(
¯
S,
¯
S, S, S, 3, σ)(1, ψ, 4)(ψ
15
, 1)
(3, ψ
12
, 1)(1, ψ
12
, 3)(
¯
S
2
,
¯
S, S, S, σ, σ, σ)(2, ψ)(2, ∆)(1, ∆)
= (m, m)(1, ψ, 1)(ψ, 2)(mψ, mψ, m(m, 1), 1)(
¯
S,
¯
S, S, S, 3, σ)(1, ψ, 4)(ψ
15
, 1)
(3, ψ
12
, 1)(1, ψ
12
, 3)(
¯
S
2
,
¯
S, S, S, σ, σ, σ)(2, ψ)(2, ∆)(1, ∆)
= (m, m)(1, ψ, 1)(ψ, 2)(mψ, mψ, m(m, 1), 1)(
¯
S,
¯
S, S, S, 3, σ)(1, ψ, 4)(ψ
15
, 1)
(3, ψ
12
, 1)(
¯
S
2
, S, S,
¯
S, σ, σ, σ)(1, ψ
12
)(2, ψ)(2, ∆)(1, ∆)
= (m, m)(1, ψ, 1)(ψ, 2)(mψ, mψ, m(m, 1), 1)(
¯
S,
¯
S, S, S, 3, σ)(1, ψ, 4)
(ψ
15
, 1)(
¯
S
2
, S, S, σ, σ,
¯
S, σ)(1, ψ
12
)(2, ψ)(2, ∆)(1, ∆)
= (m, m)(1, ψ, 1)(ψ, 2)(mψ, mψ, m(m, 1), 1)(
¯
S,
¯
S, S, S, 3, σ)
(1, ψ, 4)(S, S, σ, σ,
¯
S,
¯
S
2
, σ)(ψ
13
, 1)(1, ψ
12
)(2, ψ)(2, ∆)(1, ∆)
= (m, m)(1, ψ, 1)(ψ, 2)(mψ, mψ, m(m, 1), 1)(
¯
S,
¯
S, S, S, 3, σ)(S, σ, S, σ,
¯
S,
¯
S
2
, σ)
(ψ
13
, 1)(1, ψ
12
)(2, ψ)(2, ∆)(1, ∆)
= (m, m)(1, ψ, 1)(ψ, 2)(m, m, m(m, 1), 1)(ψ, ψ, 4)(
¯
SS,
¯
Sσ, S
2
, Sσ,
¯
S,
¯
S
2
, σ, σ)
(ψ
13
, 1)(1, ψ
12
)(2, ψ)(2, ∆)(1, ∆)
= (m, m)(m, m(m, 1), m, 1)(2, ψ
23
, 1)(ψ
22
, 4)(
¯
Sσ,
¯
SS, Sσ, S
2
,
¯
S,
¯
S
2
, σ, σ)
(ψ
13
, 1)(1, ψ
12
)(2, ψ)(2, ∆)(1, ∆)
= (m, m)(m, m(m, 1), m, 1)(2, ψ
23
, 1)(Sσ, S
2
,
¯
Sσ,
¯
SS,
¯
S,
¯
S
2
, σ, σ)
105
(ψ, 2)(ψ
13
, 1)(1, ψ
12
)(2, ψ)(2, ∆)(1, ∆)
= (m, m)(m, m(m, 1), m, 1)(Sσ, S
2
,
¯
S,
¯
S
2
, σ,
¯
Sσ,
¯
SS, σ)
(1, ψ
12
)(ψ, 2)(ψ
13
, 1)(1, ψ
12
)(2, ψ)(2, ∆)(1, ∆)
= (m, m)(m, m(m, 1), m, 1)(Sσ, S
2
,
¯
S,
¯
S
2
, σ,
¯
Sσ,
¯
SS, σ)
(2, ψ)(1, ψ, 1)(ψ, 2)(2, ψ)(1, ψ, 1)(ψ, 2)(2, ψ)(1, ψ, 1)(2, ψ)(2, ∆)(1, ∆)
= (m, m)(m, m(m, 1), m, 1)(Sσ, S
2
,
¯
S,
¯
S
2
, σ,
¯
Sσ,
¯
SS, σ)
(2, ψ)(1, ψ, 1)(ψ, 2)(2, ψ)(1, ψ, 1)(2, ψ)(ψ, 2)(1, ψ
21
)(2, ∆)(1, ∆)
= (m, m)(m, m(m, 1), m, 1)(Sσ, S
2
,
¯
S,
¯
S
2
, σ,
¯
Sσ,
¯
SS, σ)
(2, ψ)(1, ψ, 1)(ψ, 2)(1, ψ, 1)(2, ψ)(1, ψ, 1)(ψ, 2)(1, ψ
21
)(2, ∆)(1, ∆)
= (m, m)(m, m(m, 1), m, 1)(Sσ, S
2
,
¯
S,
¯
S
2
, σ,
¯
Sσ,
¯
SS, σ)
(2, ψ)(ψ, 2)(ψ
12
, 1)(1, ψ
12
)(ψ, 2)(1, ψ
21
)(1, ∆, 1)(1, ∆)
= (m, m)(m, m(m, 1), m, 1)(Sσ, S
2
,
¯
S,
¯
S
2
, σ,
¯
Sσ,
¯
SS, σ)
(2, ψ)(ψ, 2)(ψ
12
, 1)(1, ψ
12
)(ψ, 2)(2, ∆)(1, ψ)(1, ∆)
= (m, m)(m, m(m, 1), m, 1)(Sσ, S
2
,
¯
S,
¯
S
2
, σ,
¯
Sσ,
¯
SS, σ)
(2, ψ)(ψ, 2)(ψ
12
, 1)(1, ∆, 1)(1, ψ)(ψ, 1)(1, ψ)(1, ∆)
= (m, m)(m, m(m, 1), m, 1)(Sσ, S
2
,
¯
S,
¯
S
2
, σ,
¯
Sσ,
¯
SS, σ)
(2, ψ)(ψ, 2)(∆, 2)(ψ, 1)(1, ψ)(ψ, 1)(1, ψ)(1, ∆)
= (m, m)(m, m(m, 1), m, 1)(Sσ, S
2
,
¯
S,
¯
S
2
, σ,
¯
Sσ,
¯
SS, σ)
(ψ∆, 2)(1, ψ)(ψ, 1)(1, ψ)(ψ, 1)(1, ψ)(1, ∆)
= (m(m, 1), m(m, 1))(2, m, 3)(1, m, 5)(Sσ, S
2
,
¯
S,
¯
S
2
, σ,
¯
Sσ,
¯
SS, σ)
(ψ∆, 2)(1, ψ)(ψ, 1)(1, ψ)(ψ, 1)(1, ψ)(1, ∆)
= (m(m, 1), m(m, 1))(2, m, 3)(Sσ, m(S
2
,
¯
S)ψ∆,
¯
S
2
, σ,
¯
Sσ,
¯
SS, σ)
(1, ψ)(ψ, 1)(1, ψ)(ψ, 1)(1, ψ)(1, ∆)
= (m(m, 1), m(m, 1))(2, m, 3)(Sσ, ηδ,
¯
S
2
, σ,
¯
Sσ,
¯
SS, σ)
(1, ψ)(ψ, 1)(1, ψ)(ψ, 1)(1, ψ)(1, ∆)
= (m(m, 1), m(m, 1))(1, η, 4)(1, m, 3)(Sσ,
¯
S
2
, σ,
¯
Sσ,
¯
SS, σ)
(ψ)(δ, 2)(ψ, 1)(1, ψ)(ψ, 1)(1, ψ)(1, ∆)
= (m(m(1, η), 1), m(m, 1))(1, m, 3)(Sσ,
¯
S
2
, σ,
¯
Sσ,
¯
SS, σ)(ψ)(2, δ)(ψ, 1)(1, ψ)(1, ∆)
106
= (m, m(m, 1))(1, m, 3)(Sσ,
¯
S
2
, σ,
¯
Sσ,
¯
SS, σ)(ψ)(ψ)(1, δ, 1)(1, ∆)
= (m, m(m, 1))(1, m, 3)(Sσ,
¯
S
2
, σ,
¯
Sσ,
¯
S, σ)(1, S)ψ
2
(1, δ, 1)(1, ∆)
= (m, m(m, 1))(1, m, 3)(Sσ,
¯
S
2
, σ,
¯
Sσ,
¯
S, σ)ψ
2
(1, S)(1, δ, 1)(1, ∆)
= (m, m(m, 1))(1, m, 3)(Sσ,
¯
S
2
, σ,
¯
Sσ,
¯
S, σ)ψ
2
(1, (δ, S)∆)
= (m, m(m, 1))(1, m, 3)(Sσ,
¯
S
2
, σ,
¯
Sσ,
¯
S, σ)ψ
2
(1,
¯
S)
= (m(1, m), m(m, 1))(Sσ,
¯
S
2
, σ,
¯
Sσ,
¯
S, σ)(1,
¯
S)ψ
2
= (m(1, m)(Sσ,
¯
S
2
, σ), m(m, 1)(
¯
Sσ,
¯
S
2
, σ))ψ
2
= (1, 1)ψ
2
= ψ
2
Remark 4.5.1. With the same method one can prove that more generally, for all
n ,= 0 :
τ
n+1
n
= ψ
n
H
(n−1)
,H
which is equal to id iﬀ ψ
2
= id.
4.6 A Hopf cyclic theory for quasitriangular
quasiHopf algebras
Our goal in this section is to use Theorem (4.5.1), to provide a Hopf cyclic theory for
quasitriangular quasiHopf algebras. For this, we shall consider any quasitriangular
quasiHopf algebra, H, as a braided Hopf algebra, H, in the category of Hmodules.
Recall from Example (3.2.4) that for any quasitriangular Hopf algebra (H, R =
R
1
⊗R
2
), the category, HMod, of all (left) Hmodules is a braided monoidal abelian
category. It is symmetric if and only if R
−1
= R
−1
1
⊗R
−1
2
= R
2
⊗R
1
. In this latter
case, (H, R) is actually called a triangular Hopf algebra. The monoidal structure on
107
HMod is deﬁned by
h (v ⊗w) = h
(1)
v ⊗h
(2)
w,
and the braiding map ψ
V ⊗W
acts by
ψ
V ⊗W
(v ⊗w) := (R
2
w ⊗R
1
v),
for any V and W in c, where denotes the action of H. The associativity morphisms
are the usual associativities, of V ect
k
, deﬁned by:
a
A,B,C
((a ⊗b) ⊗c) := (a ⊗(b ⊗c)),
for all A, B and C in HMod [31].
This result is true for any quasitriangular quasiHopf algebra (H, R, φ, α, β) as
follows [3, 31]. Let us put φ = X
1
⊗ X
2
⊗ X
3
∈ H
3
. The category, HMod is a
braided monoidal abelian category as before by the associativity morphisms deﬁned
by:
a
A,B,C
((a ⊗b) ⊗c) := (X
1
a ⊗(X
2
b ⊗X
3
c)),
for all A, B and C in HMod.
As we mentioned in Example (3.3.4), every quasitriangular Hopf algebra H can
be turned into a braided Hopf algebra H in HMod as follows [31]. As a vector space
H = H, with Hmodule structure given by conjugation
a h = a
(1)
hS(a
(2)
).
It has the same multiplication, unit, and counit as H, but the comultiplication and
108
antipode are replaced by:
∆(h) = h
(1)
⊗h
(2)
= h
(1)
S(R
2
) ⊗R
1
h
(2)
,
S(h) = R
2
S(R
1
h).
This result is generalized, in, [3], for any quasitriangular quasiHopf algebra
(H, R, φ, α, β). As before, as a vector space H = H, with Hmodule structure given
by conjugation
a h = a
(1)
hS(a
(2)
).
The algebra structure on H is given by:
m(a, b) = a.b = X
1
aS(x
1
X
2
)αx
2
X
(1)
3
bS(x
3
X
(2)
3
),
with unit β. The coalgebra structure of H is given by:
∆(h) = h
(1)
⊗h
(2)
=
x
1
X
1
h
(1)
g
1
S(x
2
R
2
y
3
X
(2)
3
) ⊗x
3
R
1
y
1
X
2
h
(2)
g
2
S(y
2
X
(1)
3
),
with counit ε = ε. Finally the antipode for H is deﬁned by:
S(h) = x
1
R
2
p
2
S(q
1
(X
2
R
1
p
1
h)S(q
2
)X
3
).
Here the following notations have been used:
φ = X
1
⊗X
2
⊗X
3
,
φ
−1
= x
1
⊗x
2
⊗x
3
= y
1
⊗y
2
⊗y
3
,
p
1
⊗p
2
= x
1
⊗x
2
βS(x
3
),
109
q
1
⊗q
2
= X
1
⊗S
−1
(αX
3
)X
2
,
and
g
1
⊗g
2
= ∆(S(x
1
)αx
2
)ξ(S ⊗S)(∆
op
(x
3
)),
where,
ξ = B
1
βS(B
4
) ⊗B
2
βS(B
3
)
B
1
⊗B
2
⊗B
3
⊗B
4
= (∆⊗id ⊗id)(Φ)(Φ
−1
⊗1).
Now, let (δ, σ) be a braided modular pair in involution for H, in HMod. Using
Theorem (4.5.1) we can associate a paracocyclic object to H. By passing to subspaces
ker(1 − τ
n+1
n
) we obtain a cocyclic object. In this way we can deﬁne a Hopf cyclic
cohomology HC
∗
(δ,σ)
(H, R, φ, α, β) for any quasitriangular quasiHopf algebra H
endowed with a braided modular pair in involution.
Chapter 5
A super version of the ConnesMoscovici
Hopf algebra
In [9] Connes and Moscovici deﬁned a Hopf algebra H(n),for any n ≥ 1, and computed
the periodic Hopf cyclic cohomology of H(n). For n = 1, the ConnesMoscovici Hopf
algebra H(1), denoted here by H
1
, is isomorphic to a bicrossproduct Hopf algebra
F(G
2
)U(g
1
) [9, 17]. The actions and coactions involved in this bicrossproduct
can be understood by using the factorisation of the group of orientation preserving
diﬀeomorphisms of the real line, Diff
+
(R).
In this chapter our goal is to deﬁne a super version of H
1
which we will denote by
H
s
1
. For that we deﬁne a super version of the group G = Diff
+
(R), namely the super
group G
s
= Diff
+
(R
1,1
) of orientation preserving diﬀeomorphisms of the super line
R
1,1
. We deﬁne two (super) subgroups G
s
1
and G
s
2
of G
s
where G
s
1
is the group of
aﬃne transformations. We show that the factorisation G
s
= G
s
1
G
s
2
holds. We will
use this factorisation to deﬁne a bicrossproduct super Hopf algebra F(G
s
2
)U(g
s
1
),
analogous to the non super case. We will call this superbicrossproduct Hopf algebra
the super version of H
1
and denote it by H
s
1
.
In this chapter for notations to be more consistent with non super case, we
use the following conventions. To denote the comultlipication ∆ : B → B ⊗ B of a
bialgebra B we use the Sweedler’s notation ∆
n
(b) = b
(1)
⊗ b
(2)
⊗ ⊗ b
(n+1)
. Also
for a coaction A →B ⊗A of B on an algebra A we use ∇(a) = a
(1)
⊗a
(2)
.
110
111
5.1 The ConnesMoscovici Hopf algebra H
1
In this section we recall the deﬁnition of the standard (non super) ConnesMoscovici
Hopf algebra H
1
and its description in terms of a bicrossproduct Hopf algebra [9, 17].
The following deﬁnition gives a description of H
1
by generators and relations.
Deﬁnition 5.1.1. [9, 17] The ConnesMoscovici Hopf algebra H
1
is generated by
elements X, Y , δ
n
, n ≥ 1 with relations:
[Y, X] = X, [X, δ
n
] = δ
n+1
, [Y, δ
n
] = nδ
n
, [δ
m
, δ
n
] = 0, ∀m, n
∆(X) = X ⊗1 + 1 ⊗X +Y ⊗δ
1
, ∆(Y ) = Y ⊗1 + 1 ⊗Y, ∆(δ
1
) = δ
1
⊗1 + 1 ⊗δ
1
,
ε(X) = ε(Y ) = ε(δ
n
) = 0, ∀n,
S(X) = Y δ
1
−X, S(Y ) = −Y, S(δ
1
) = −δ
1
.
Remark 5.1.1. The above deﬁnition follows the righthanded notattion as in [17], in
the sense that in the deﬁnition of ∆(X), the term δ
1
⊗Y in [9] is replaced by Y ⊗δ
1
in [17].
Lemma 5.1.1. [17, 31] Let A and H be two Hopf algebras such that A is a left H
module algebra, and H is right Acomodule coalgebra. Let furthermore these structures
satisfy the following compatibility conditions:
∆(h a) = h
(1)
(1)
a
(1)
⊗h
(1)
(2)
(h
(2)
a
(2)
),
ε(h a) = ε(h)ε(a),
∇
r
(gh) = g
(1)
(1)
h
(1)
⊗g
(1)
(2)
(g
(2)
h
(2)
), ∇
r
(1) = 1 ⊗1,
h
(2)
(1)
⊗(h
(1)
a)h
(2)
(2)
= h
(1)
(1)
⊗h
(1)
(2)
(h
(2)
a),
112
for any a, b in A and g, h in H, where we have denoted the actions by h a and the
coactions by ∇
r
(h) = h
(1)
⊗h
(2)
. Then the vector space A⊗H can be equipped with
a Hopf algebra structure as follows:
(a ⊗h)(b ⊗g) = a(h
(1)
b) ⊗h
(2)
g,
∆(a ⊗h) = a
(1)
⊗h
(1)
(1)
⊗a
(2)
h
(1)
(2)
⊗h
(2)
,
ε(a ⊗h) = ε(a)ε(h),
S(a ⊗h) = (1 ⊗S(h
(1)
))(S(ah
(2)
) ⊗1),
for any a, b in A and g, h in H. It is called the leftright bicrossproduct Hopf algebra
and is denoted by AH.
In [9] a Hopf subalgebra of H
1
is deﬁned as the unital commutative subalgebra
of H
1
generated by ¦δ
n
, n ≥ 1¦. This Hopf algebra, which we denote by F(G
2
), is
isomorphic to the so called comeasuring Hopf algebra of the real line, generated by
¦a
n
, n ≥ 1¦ with a
1
= 1 and with the following structure [14, 15, 17]:
∆(a
n
) =
n
k=1
(
i
1
++i
k
=n
a
i
1
a
i
k
) ⊗a
k
,
ε(a
n
) = δ
n,1
, (5.1.1)
S(a
n+1
) =
(c
1
, ,c
n+1
)∈Λ
(−1)
n−c
1
(2n −c
1
)!c
1
! a
c
1
1
a
c
2
2
a
c
n+1
n+1
(n + 1)! c
1
!c
2
! c
n+1
!
,
where
Λ = ¦(c
1
, , c
n+1
) [
n+1
j=1
c
j
= n,
n+1
j=1
jc
j
= 2n¦.
If instead of generators a
n
we work with n!a
n
we get the so called Faa di Bruno Hopf
algebra again isomorphic to F(G
2
) [14, 15, 17].
One can also deﬁne another Hopf algebra, denoted by U(g
1
), as follows. let g
1
113
be the Lie algebra generated by two elements X and Y as in H
1
, (i.e., [Y, X] = X),
then U(g
1
) is the universal enveloping algebra of g
1
.
Lemma 5.1.2. F(G
2
) is a left U(g
1
)module algebra via the actions:
X a
n
= (n + 1)a
n+1
−2a
2
a
n
, Y a
n
= (n −1)a
n
,
and U(g
1
) is a right F(G
2
)comodule coalgebra via the coactions:
∇
r
(X) = X ⊗1 + Y ⊗2a
2
, ∇
r
(Y ) = Y ⊗1,
and these actions and coactions are compatible in the sense of Lemma (5.1.1).
Theorem 5.1.1. [9, 17] H
1
is isomorphic to the bicrossproduct Hopf algebra
F(G
2
)U(g
1
).
A good way to understand the actions, coactions and even the notations, intro
duced above, is to look at the factorisation of the group Diff
+
(R) [9, 17]. Let
G = Diff
+
(R) = ¦Φ ∈ Diff(R) [
˙
Φ(x) > 0, ∀x ∈ R¦,
be the group of positively oriented diﬀeomorphisms of the real line and
G
1
= ¦ψ = (a, b) ∈ G [ ψ(x) = ax + b, a, b ∈ R, a > 0¦,
be the aﬃne subgroup of G. The following representattion of G
1
as a subgroup of
GL(2) is very useful:
G
1
=
_
_
_
(a, b) =
_
_
a b
0 1
_
_
∈ GL(2) [ a > 0
_
_
_
. (5.1.2)
Let also
G
2
= ¦φ ∈ G [ φ(0) = 0,
˙
φ(0) = 1¦.
114
The factorisation Diff
+
(R) = G
1
G
2
is as follows. For any Φ in G we have Φ = ψφ
with ψ ∈ G
1
and φ ∈ G
2
, where
ψ = (
˙
Φ(0), Φ(0)), φ(x) =
Φ(x) −Φ(0)
˙
Φ(0)
, ∀x ∈ R. (5.1.3)
Here
˙
Φ(x) =
d
dx
(Φ(x)). We will use this notation in the sequel. We will also use
¨
Φ(x) =
d
2
dx
2
(Φ(x)).
Let us recall that, more generally, given any group G with two subgroups G
1
and G
2
, we say we have a group factorisation G = G
1
G
2
if for any g ∈ G there
is a unique decomposition g = ab where a ∈ G
1
and b ∈ G
2
. Given any group
factorisation G = G
1
G
2
, one has always a left action of G
2
on G
1
and a right action
of G
1
on G
2
deﬁned in the folloing way. First one deﬁnes two maps π
1
: G → G
1
,
π
2
: G → G
2
by π
1
(g) = a and π
2
(g) = b, for any g in G. Next one can deﬁne the
aforementioned actions by g
2
g
1
= π
1
(g
2
g
1
) and g
2
g
1
= π
2
(g
2
g
1
) for any g
1
in
G
1
and g
2
in G
2
.
Now considering the above factorisation of the group Diff
+
(R) in (5.1.3), we
have the following left action of G
2
on G
1
and a right action of G
1
on G
2
.
φ ψ = (a
˙
φ(b), φ(b)), (φ ψ)(x) =
φ(ax + b) −φ(b)
a
˙
φ(b)
,
for any ψ = (a, b) in G
1
and φ in G
2
.
Next, using the matrix representation of G
1
in GL(2) as in (5.1.2), we can deﬁne
the Lie algebra g
1
= Lie(G
1
). It turns out that g
1
is generated by two elements
X =
_
_
0 1
0 0
_
_
, Y =
_
_
1 0
0 0
_
_
,
which clearly satisfy the relation [Y, X] = X. Thus we have the universal enveloping
algebra U(g
1
). On the other hand in the Hopf algebra F(G
2
) deﬁned by relations
115
(5.1.1), the generators a
n
can be considered as functions on G
2
as follows:
a
n
(φ) =
1
n!
φ
(n)
(0), ∀φ ∈ G
2
.
The actions deﬁned in Lemma (5.1.2), can be realised in the following way.
Recall ﬁrst that from X =
_
_
0 1
0 0
_
_
, Y =
_
_
1 0
0 0
_
_
in g
1
we can deﬁne
elements
e
tX
= (1, t), e
tY
= (e
t
, 0),
of G
1
, for any t in R. The action X a
n
can be realised by:
(X a
n
)(φ) =
d
dt
[
t=0
a
n
(φ e
tX
)
=
d
dt
[
t=0
a
n
(
φ(x + t) −φ(t)
˙
φ(t)
)
=
d
dt
[
t=0
(
φ
(n)
(t)
n!
˙
φ(t)
)
=
d
dt
[
t=0
(
φ
(n)
(0) + tφ
(n+1)
(0)
n!(
˙
φ(0) + t
¨
φ(0))
)
=
d
dt
[
t=0
(1/n!)(φ
(n)
(0) −t
¨
φ(0)φ
(n)
(0) + tφ
(n+1)
(0))
= (1/n!)(−
¨
φ(0)φ
(n)
(0) + φ
(n+1)
(0))
= [(n + 1)a
n+1
−2a
2
a
n
](φ),
for any φ in G
2
. This impies
X a
n
= (n + 1)a
n+1
−2a
2
a
n
,
as in Lemma (5.1.2). A similar computation will give the action Y a
n
as in Lemma
(5.1.2). The coactions deﬁned in Lemma (5.1.2) can also be realised using the fac
torisation G = G
1
G
2
[17].
Note that the above realisation of actions and coactions is not needed for the
116
proof of Lemma (5.1.2). It rather gives a good intuition about where those formulas
for actions and coactions come from.
In the rest of this chapter, we pursue the goal of deﬁning a super version of
H
1
denoted by H
s
1
. For that we replace R by R
1,1
in the deﬁnition of the group
G = Diff
+
(R) to obtain a supergroup G
s
= Diff
+
(R
1,1
). Then along the same way
as in this section we factorise G
s
= Diff
+
(R
1,1
) into two super subgroups denoted
by G
s
1
and G
s
2
. We deﬁne two super Hopf algebras F(G
s
2
) and U(g
s
1
) and give the
action and coaction between them. Eventually, we deﬁne a super bicrossproduct Hopf
algebra F(G
s
2
)U(g
s
1
). This latter super Hopf algebra denoted by H
s
1
, will be the
super version of H
1
.
5.2 The super group G
s
= Diff
+
(R
1,1
) and its
factorisation
In this section, by replacing R by the supermanifold R
1,1
, we deﬁne a super version of
the group G = Diff
+
(R) namely the super group G
s
= Diff
+
(R
1,1
). Analogous to
the non super case, we consider two sub super groups G
s
1
and G
s
2
of G
s
where G
s
1
is
called the aﬃne part of G
s
. We show the factorisation G
s
= G
s
1
G
s
2
. This factorisation
will result to a left action of G
s
2
on G
s
1
and a right action of G
s
1
on G
s
2
.
For the general theory of supermanifolds we refer to [4, 34, 39]. The superman
ifold R
1,1
, also called the superline, is a super ringed space S = (R
1
, O
S
), where O
S
is a sheaf of super commutative Ralgebras over R deﬁned as follows. Let U ⊂ R be
open, then:
O
S
(U) = C
∞
(U) ⊗∧
1
(R) = C
∞
R
(U)[θ],
where θ, the generator of ∧
1
(R) is called the odd generator. We also denote the
even generator normally by x. Supermanifolds form a category where the morphism
f : S
1
→S
2
is a morphism of the underlying super spaces [4, 34, 39].
117
A super Lie group is a group object in the category of supermanifolds. Alterna
tively a super Lie group can be deﬁned as a representable functor from the category
of supermanifolds to the category of groups [4, 34, 39]. For example GL(p, q) =
GL(R
p,q
) is the super general linear group of automorphisms of R
p,q
. This super
group has a matrix representation as follows. It is formed by matrices
_
_
A B
C D
_
_
,
where A, D are, respectively, p p and q q invertible matrices with even entries,
and C, D are, respectively, p q and q p matrices with odd entries. As an special
case GL(2, 1) is formed by matrices
_
_
_
_
_
_
a b x
c d y
z w e
_
_
_
_
_
_
,
where a, b, c, d, e are even elements, ad−bc ,= 0, e ,= 0, and x, y, z, w are odd elements.
Deﬁnition 5.2.1. The group of positively oriented diﬀeomorphisms of the super real
line, R
1,1
, is deﬁned as follows:
G
s
= Diff
+
(R
1,1
) = ¦Φ(x, θ) = (A(x) + B(x)θ , C(x) + D(x)θ)¦,
such that A(x), D(x) are even, A(x), D(x) ∈ Diff(R),
˙
A(x) > 0,
˙
D(x) > 0, and
B(x), C(x) are odd.
Deﬁnition 5.2.2. The aﬃne part of G
s
, denoted by G
s
1
and also denoted by Aff(R
1,1
)
is deﬁned by:
G
s
1
= ¦ψ(x, θ) = (ax + bθ + e, cx + dθ + f) ∈ G
s
¦,
118
such that a, e, d are even, a, d > 0, and b, c, f are odd. An element ψ(x, θ) = (ax +
bθ + e, cx + dθ + f) of G
s
1
can also be represented in the following way:
ψ(x, θ) ≡
_
_
a b
c d
_
_
_
_
x
θ
_
_
+
_
_
e
f
_
_
.
Therefore there exist the following representation of G
s
1
in GL(2, 1) which will be very
useful for our purpose.
G
s
1
=
_
¸
¸
¸
¸
_
¸
¸
¸
¸
_
ψ =
_
_
_
_
_
_
a e b
0 1 0
c f d
_
_
_
_
_
_
[ a, e, d are even, b, c, f are odd, and a, d > 0
_
¸
¸
¸
¸
_
¸
¸
¸
¸
_
.
Deﬁnition 5.2.3. A second super subgroup of G
s
is deﬁned by;
G
2
= ¦φ = (A(x) + B(x)θ, C(x) + D(x)θ) ∈ G
s
[ φ(0, 0) = 0, Jφ(0, 0) = 1¦.
In other words
G
s
2
= ¦φ = (A(x) + B(x)θ, C(x) + D(x)θ) ∈ G
s
¦, (5.2.4)
where A(0) = B(0) = C(0) =
˙
C(0) = 0, and
˙
A(0) = D(0) = 1.
In order to proceed to the factorisation G
s
= G
s
1
G
s
2
, for any Φ ∈ Diff
+
(R
1,1
)
we deﬁne:
π
1
(Φ) = (JΦ(0, 0), Φ(0, 0))
≡
_
_
˙
A(0) B(0)
˙
C(0) D(0)
_
_
_
_
x
θ
_
_
+
_
_
A(0)
C(0)
_
_
∈ G
s
1
,
119
and
π
2
(Φ) = (JΦ(0, 0))
−1
(Φ(x, θ)−Φ(0, 0)) =
_
_
˙
A(0) B(0)
˙
C(0) D(0)
_
_
−1
(Φ(x, θ)−Φ(0, 0)) ∈ G
s
2
.
Here we have used the deﬁnition
JΦ(x, θ) :=
_
_
∂Φ
1
∂x
−
∂Φ
1
∂θ
∂Φ
2
∂x
∂Φ
2
∂θ
_
_
=
_
_
˙
A(x) +
˙
B(x)θ B(x)
˙
C(x) +
˙
D(x)θ D(x)
_
_
,
where Φ
1
= A(x) + B(x)θ and Φ
2
= C(x) + D(x)θ are, respectively, even and odd
components of Φ ∈ Diff
+
(R
1,1
). The operator
∂
∂x
is even and the operator
∂
∂θ
is
odd. Also the formula for the inverse super matrix is
_
_
a b
c d
_
_
−1
=
1
da
_
_
d +
bc
a
−b
−c a +
cb
d
_
_
.
Now if we let ψ(x, θ) = π
1
(Φ) and φ(x, θ) = π
2
(Φ), it is clear that for any
Φ ∈ Diff
+
(R
1,1
) :
Φ = ψφ,
which proves the factorisation
G
s
= G
s
1
G
s
2
.
Therefore, analogous to the non super case, we have the following two natural actions:
The left action of G
s
2
on G
S
1
:
G
s
2
G
s
1
→G
s
1
φ ψ = π
1
(φψ),
and the right action of G
s
1
on G
s
2
:
G
s
2
G
s
1
→G
s
2
120
φ ψ = π
2
(φψ).
5.3 Two super Hopf algebras U(g
s
1
) and F(G
s
2
)
In this section we assign to super groups G
s
1
and G
s
2
, two super Hopf algebras U(g
s
1
)
and F(G
s
2
). The super Hopf algebra U(g
s
1
) is just the universal enveloping algebra
of the super Lie algebra g
s
1
= Lie(G
s
1
) which we will introduce. As for the F(G
s
2
) we
actually extend the theory presented in [14, 15] to the super real line R
1,1
. This means
we consider G
s
2
as the group of certain diﬀeomorphisms on R
1,1
( local representatives
of orientationpreserving diﬀeomorphisms of R
1,1
leaving the origin ﬁxed) and deﬁne
the coordinate functions a
n
, b
n
, c
n
and d
n
on G
s
2
. Then F(G
s
2
) would be the super
Hopf algebra generated, as a super commutative super algebra, by a
n
, b
n
, c
n
and d
n
for which we will deﬁne the Hopf algebra structure.
5.3.1 The super Hopf algebra U(g
s
1
)
Recall from Deﬁnition (5.2.2) that the super group G
s
1
is:
G
s
1
= Aff(R
1,1
) ≡
_
¸
¸
¸
¸
_
¸
¸
¸
¸
_
_
_
_
_
_
_
a e b
0 1 0
c f d
_
_
_
_
_
_
∈ GL(2, 1)
_
¸
¸
¸
¸
_
¸
¸
¸
¸
_
,
such that a, e, d are even, b, c, f are odd, and a, d > 0.
The super Lie algebra g
s
1
 where G
s
1
= Lie(g
s
1
) is generated by three even
generators:
X =
_
_
_
_
_
_
1 0 0
0 0 0
0 0 0
_
_
_
_
_
_
, Y =
_
_
_
_
_
_
0 1 0
0 0 0
0 0 0
_
_
_
_
_
_
, Z =
_
_
_
_
_
_
0 0 0
0 0 0
0 0 1
_
_
_
_
_
_
,
121
and three odd generators
U =
_
_
_
_
_
_
0 0 1
0 0 0
0 0 0
_
_
_
_
_
_
, V =
_
_
_
_
_
_
0 0 0
0 0 0
−1 0 0
_
_
_
_
_
_
, W =
_
_
_
_
_
_
0 0 0
0 0 0
0 −1 0
_
_
_
_
_
_
.
Note that, in contrast with non super case, we have exchanged the role of generators
X and Y . This was just for our notations to be more consistent.
It is easy to check that the following super bracket relations hold:
[X, Y ] = XY −Y X = Y, [X, Z] = 0, [Y, Z] = 0,
[X, U] = U, [X, V ] = −V, [X, W] = 0,
[Y, U] = 0, [Y, V ] = −W, [Y, W] = 0,
[Z, U] = −U, [Z, V ] = V, [Z, W] = W,
[U, V ] = UV + V U = −(X + Z), [U, W] = −Y, [V, W] = 0.
[X, X] = [Y, Y ] = [Z, Z] = [U, U] = [V, V ] = [W, W] = 0. (5.3.5)
The super Hopf algebra U(g
s
1
) is the universal enveloping algebra of g
s
1
. Recall
that the super Hopf algebra structure of U(g
s
1
) is deﬁned by the ﬂlowing relations:
[h, g] = hg −(−1)
[h[[g[
gh,
∆(h) = 1 ⊗h + h ⊗1, ε(h) = 0, S(h) = −h, (5.3.6)
for all h, g in g
s
1
⊂ U(g
s
1
).
122
5.3.2 The super Hopf algebra F(G
s
2
)
Recall from Deﬁnition (5.2.3) that the super group G
s
2
is:
G
s
2
= ¦φ = (A(x) + B(x)θ, C(x) + D(x)θ) ∈ G
s
[ φ(0, 0) = 0, Jφ(0, 0) = 1¦.
The super Hopf algebra F(G
s
2
), as a super commutative super algebra (the
super polynomial algebra R
1,1
[a
n
, b
n
, c
n
, d
n
]) , is generated by:
two sets of even generators
a
n
, d
n
, n ≥ 1
and two sets of odd generators
b
n
, c
n
, n ≥ 1
where for any φ(x, θ) = (A(x) + B(x)θ, C(x) + D(x)θ) in G
s
2
a
n
(φ) = (1/n!)A
(n)
(0),
b
n
(φ) = (1/n!)B
(n)
(0),
c
n
(φ) = (1/n!)C
(n)
(0),
d
n
(φ) = (1/n!)D
(n)
(0). (5.3.7)
To deﬁne the coproduct on F(G
s
2
), analogous to non super case [14, 15], we use
the following formalism:
m∆(a
n
)(φ ⊗φ
/
) = a
n
(φφ
/
), φ, φ
/
∈ G
s
2
, (5.3.8)
and the same formulas for b
n
, c
n
and d
n
.
we need to study the composition of two elements of G
s
2
. Let φ(x, θ) = (A(x) +
123
B(x)θ , C(x)+D(x)θ) and φ
/
(x, θ) = (A
/
(x)+B
/
(x)θ , C
/
(x)+D
/
(x)θ) be two elements
of G
s
2
. We want to ﬁnd the formula for the composition φφ
/
(x, θ). Let f = A
/
(x) +
B
/
(x)θ and g = C
/
(x) + D
/
(x)θ, then clearly we have the following formula for the
composition:
φφ
/
(x, θ) = (A(f) + B(f)g , C(f) + D(f)g). (5.3.9)
Since A(x) = a
0
+ a
1
x + a
2
x
2
+ and since
f
n
= A
/
(x)
n
+ nA
/
(x)
n−1
B
/
(x)θ, n ≥ 0,
we have
A(f) = A(A
/
(x)) +
˙
A(A
/
(x))B
/
(x)θ,
and in the same way
B(f) = B(A
/
(x)) +
˙
B(A
/
(x))B
/
(x)θ,
C(f) = C(A
/
(x)) +
˙
C(A
/
(x))B
/
(x)θ,
D(f) = D(A
/
(x)) +
˙
D(A
/
(x))B
/
(x)θ.
These imply that
B(f)g = B(A
/
(x))C
/
(x) + B(A
/
(x))D
/
(x)θ +
˙
B(A
/
(x))B
/
(x)θC
/
(x)
= B(A
/
(x))C
/
(x) + B(A
/
(x))D
/
(x)θ −
˙
B(A
/
(x))B
/
(x)C
/
(x)θ.
Therefore
A(f) + B(f)g = [A(A
/
(x) + B(A
/
(x))C
/
(x)] + (5.3.10)
[
˙
A(A
/
(x))B
/
(x) + B(A
/
(x))D
/
(x) −
˙
B(A
/
(x))B
/
(x)C
/
(x)]θ,
124
and
C(f) + D(f)g = [C(A
/
(x) + D(A
/
(x))C
/
(x)] + (5.3.11)
[
˙
C(A
/
(x))B
/
(x) + D(A
/
(x))D
/
(x) −
˙
D(A
/
(x))B
/
(x)C
/
(x)]θ.
By applying these two formulas to Formula (5.3.9), we will get the desired formula
for the composition.
Now we proceed to comultiplications, starting with ∆(a
n
). First of all, it is
convenient for us to ﬁx the following notation for any polynomial P(x):
P(x) =
∞
n=0
λ
P
n
x
n
= p
0
+ p
1
x + p
2
x
2
+ p
3
x
3
+ . (5.3.12)
Therefore from the deﬁnition of G
s
2
we have:
A(x) =
∞
n=0
λ
A
n
x
n
= x + a
2
x
2
+ a
3
x
3
+ ,
B(x) =
∞
n=0
λ
B
n
x
n
= b
1
x + b
2
x
2
+ b
3
x
3
+ ,
C(x) =
∞
n=0
λ
C
n
x
n
= c
2
x
2
+ c
3
x
3
+ ,
D(x) =
∞
n=0
λ
D
n
x
n
= 1 + d
1
x + d
2
x
2
+ d
3
x
3
+ ,
where a
i
and d
i
are even and b
i
and c
i
are odd.
We also recall from [14, 15, 17] that:
P(A(x)) =
∞
n=0
λ
P(A)
n
x
n
,
125
with
λ
P(A)
0
= p
0
, λ
P(A)
n
=
n
k=1
p
k
(
l
1
+l
2
++l
k
=n
a
l
1
a
l
2
a
l
k
), n ≥ 1. (5.3.13)
Therefore:
A(A
/
(x)) =
∞
n=0
λ
A(A
)
n
x
n
=
∞
n=1
(
n
k=1
a
k
l
1
+l
2
++l
k
=n
a
/
l
1
a
/
l
2
a
/
l
k
)x
n
= x + (a
/
2
+ a
2
)x
2
+ (a
/
3
+ 2a
2
a
/
2
+ a
3
)x
3
+ (5.3.14)
Similarly we have
B(A
/
(x)) =
∞
n=0
λ
B(A
)
n
x
n
=
∞
n=1
(
n
k=1
b
k
l
1
+l
2
++l
k
=n
a
/
l
1
a
/
l
2
a
/
l
k
)x
n
= b
1
x + (b
1
a
/
2
+ b
2
)x
2
+ (b
1
a
/
3
+ 2b
2
a
/
2
+ b
3
)x
3
+ (5.3.15)
C(A
/
(x)) =
∞
n=0
λ
C(A
)
n
x
n
=
∞
n=1
(
n
k=1
c
k
l
1
+l
2
++l
k
=n
a
/
l
1
a
/
l
2
a
/
l
k
)x
n
= c
2
x
2
+ (2c
2
a
/
2
+ c
3
)x
3
+ (5.3.16)
D(A
/
(x)) =
∞
n=0
λ
D(A
)
n
x
n
= 1 +
∞
n=1
(
n
k=1
d
k
l
1
+l
2
++l
k
=n
a
/
l
1
a
/
l
2
a
/
l
k
)x
n
= 1 + d
1
x + (d
1
a
/
2
+ d
2
)x
2
+ (d
1
a
/
3
+ 2d
2
a
/
2
+ d
3
)x
3
+ (5.3.17)
Now back to ∆(a
n
). We know that for any tow polynomials
f(x) =
∞
n=0
λ
f
n
x
n
, and g(x) =
∞
n=0
λ
g
n
x
n
,
126
one has:
fg(x) =
∞
n=0
λ
fg
n
x
n
, with λ
fg
n
=
n
i=0
λ
f
i
λ
g
n−i
. (5.3.18)
From (5.3.15) and (5.3.18) we get:
B(A
/
(x))C
/
(x) =
∞
n=0
λ
B(A
)C
n
x
n
,
with λ
B(A
)C
0
= 0 and for n ≥ 1
λ
B(A
)C
n
=
n
i=1
λ
B(A
)
i
λ
C
n−i
=
n
i=1
(
i
k=1
b
k
l
1
+l
2
++l
k
=i
a
/
l
1
a
/
l
2
a
/
l
k
)c
/
n−i
.
Therefore up to degree three:
B(A
/
(x))C
/
(x) = b
1
c
/
2
x
3
+
All these imply that:
A(A
/
(x)) + B(A
/
(x))C
/
(x) =
∞
n=1
_
λ
A(A
)
n
+ λ
B(A
)C
n
_
x
n
,
where
λ
A(A
)
n
+ λ
B(A
)C
n
=
n
k=1
a
k
l
1
+l
2
++l
k
=n
a
/
l
1
a
/
l
2
a
/
l
k
+
n
i=1
(
i
k=1
b
k
l
1
+l
2
++l
k
=i
a
/
l
1
a
/
l
2
a
/
l
k
)c
/
n−i
.
e.g., up to degree three:
A(A
/
(x)) + B(A
/
(x))C
/
(x) = x + (a
/
2
+ a
2
)x
2
+ (a
/
3
+ 2a
2
a
/
2
+ a
3
+ b
1
c
/
2
)x
3
+
127
Now since by (5.3.7), (5.3.9) and (5.3.10) we have:
a
n
(φφ
/
) = (λ
A(A
)
n
+ λ
B(A
)C
n
),
We can use Formalism (5.3.8) to drive the formula for ∆(a
n
). For example:
a
1
(φφ
/
) = 1 = m(1 ⊗1)(φ ⊗φ
/
),
a
2
(φφ
/
) = a
/
2
+ a
2
= m(1 ⊗a
2
+ a
2
⊗1)(φ ⊗φ
/
),
a
3
(φφ
/
) = a
/
3
+ 2a
2
a
/
2
+a
3
+b
1
c
/
2
= m(1 ⊗a
3
+ 2a
2
⊗a
2
+a
3
⊗1 +b
1
⊗c
2
)(φ ⊗φ
/
),
which imply that:
∆(a
1
) = ∆(1) = 1 ⊗1,
∆(a
2
) = 1 ⊗a
2
+ a
2
⊗1,
∆(a
3
) = 1 ⊗a
3
+ a
3
⊗1 + 2a
2
⊗a
2
+ b
1
⊗c
2
,
and more generaly for n ≥ 1
∆(a
n
) =
n
k=1
a
k
⊗
l
1
+l
2
++l
k
=n
a
l
1
a
l
2
a
l
k
+
n
i=1
i
k=1
b
k
⊗(
l
1
+l
2
++l
k
=i
a
l
1
a
l
2
a
l
k
)c
n−i
.
Let us now procees to ∆(b
n
). Since
˙
A(x) =
∞
n=0
(n + 1)a
n+1
x
n
= 1 + 2a
2
x + 3a
3
x
2
+ 4a
4
x
3
+ ,
128
by (5.3.13) we have
˙
A(A
/
(x)) =
∞
n=0
λ
˙
A(A
)
n
x
n
= 1 +
∞
n=1
_
_
n
k=1
(k + 1)a
k+1
(
l
1
+l
2
++l
k
=n
a
/
l
1
a
/
l
2
a
/
l
k
)
_
_
x
n
= 1 + 2a
2
x + (2a
2
a
/
2
+ 3a
3
)x
2
+ (2a
2
a
/
3
+ 6a
3
a
/
2
+ 4a
4
)x
3
+ .
So again by using (5.3.18) we get:
˙
A(A
/
(x))B
/
(x) =
∞
n=0
λ
˙
A(A
)B
n
x
n
,
where λ
˙
A(A
)B
0
= 0 and for n ≥ 1
λ
˙
A(A
)B
n
=
n
i=0
λ
˙
A(A
)
i
λ
B
n−i
= b
/
n
+
n
i=1
i
k=1
(k + 1)a
k+1
(
l
1
+l
2
++l
k
=i
a
/
l
1
a
/
l
2
a
/
l
k
)b
/
n−i
(5.3.19)
E.g., up to dehree three:
˙
A(A
/
(x))B
/
(x) = b
/
1
x + (b
/
2
+ 2a
2
b
/
1
)x
2
+ (b
/
3
+ 2a
2
b
/
2
+ 2a
2
a
/
2
b
/
1
+ 3a
3
b
/
1
)x
3
+
Also again by (5.3.15) and (5.3.18) we have:
B(A
/
(x))D
/
(x) =
∞
n=0
λ
B(A
)D
n
x
n
,
129
with λ
B(A
)D
0
= 0 and for n ≥ 1
λ
B(A
)D
n
=
n
i=1
λ
B(A
)
i
λ
D
n−i
=
n
i=1
(
i
k=1
b
k
l
1
+l
2
++l
k
=i
a
/
l
1
a
/
l
2
a
/
l
k
)d
/
n−i
. (5.3.20)
For example up to degree three:
B(A
/
(x))D
/
(x) = b
1
x + (b
1
d
/
1
+ b
1
a
/
2
+ b
2
)x
2
+ (b
1
d
/
2
+ b
1
a
/
2
d
/
1
+ b
2
d
/
1
+ b
1
a
/
3
+ 2b
2
a
/
2
+ b
3
)x
3
+
Next, since
˙
B(x) =
∞
n=0
(n + 1)b
n+1
x
n
= b
1
+ 2b
2
x + 3b
3
x
2
+ 4b
3
x
+ ,
by (5.3.13) we have
˙
B(A
/
(x)) =
∞
n=0
λ
˙
B(A
)
n
x
n
= b
1
+
∞
n=1
_
_
n
k=1
(k + 1)b
k+1
(
l
1
+l
2
++l
k
=n
a
/
l
1
a
/
l
2
a
/
l
k
)
_
_
x
n
= b
1
+ 2b
2
x + (2b
2
a
/
2
+ 3b
3
)x
2
+ (2b
2
a
/
3
+ 6b
3
a
/
2
+ 4b
4
)x
3
+
By using (5.3.18) twice we have:
˙
B(A
/
(x))B
/
(x) =
∞
n=0
λ
˙
B(A
)B
n
x
n
,
130
where λ
˙
B(A
)B
0
= 0 and for n ≥ 1
λ
˙
B(A
)B
n
=
n
i=0
λ
˙
B(A
)
i
λ
B
n−i
= b
1
b
/
n
+
n
i=1
i
k=1
(k + 1)b
k+1
(
l
1
+l
2
++l
k
=i
a
/
l
1
a
/
l
2
a
/
l
k
)b
/
n−i
,
which implies, up to degree three:
˙
B(A
/
(x))B
/
(x) = b
1
b
/
1
x + (b
1
b
/
2
+ 2b
2
b
/
1
)x
2
+ (b
1
b
/
3
+ 2b
2
b
/
2
+ 2b
2
a
/
2
b
/
1
+ 3b
3
b
/
1
)x
3
+ ,
and
˙
B(A
/
(x))B
/
(x)C
/
(x) =
∞
n=0
λ
˙
B(A
)B
C
n
x
n
,
where λ
˙
B(A
)B
C
0
= 0 and for n ≥ 1
λ
˙
B(A
)B
C
n
=
n
i=0
λ
˙
B(A
)B
i
λ
C
n−i
(5.3.21)
=
n
i=1
_
_
b
1
b
/
i
c
/
n−i
+
i
j=1
j
k=1
(k + 1)b
k+1
(
l
1
+l
2
++l
k
=j
a
/
l
1
a
/
l
2
a
/
l
k
)b
/
i−j
c
/
n−i
_
_
,
which implies up todegree three:
˙
B(A
/
(x))B
/
(x)C
/
(x) = b
1
b
/
1
c
/
2
x
3
+ .
131
Thus by (5.3.19), (5.3.20) and (5.3.21) we have:
˙
A(A
/
(x))B
/
(x) + B(A
/
(x))D
/
(x) −
˙
B(A
/
(x))B
/
(x)C
/
(x)
=
∞
n=1
_
λ
˙
A(A
)B
n
+ λ
B(A
)D
n
−λ
˙
B(A
)B
C
n
_
x
n
= (b
/
1
+ b
1
)x + (b
/
2
+ 2a
2
b
/
1
+ b
1
d
/
1
+ b
1
a
/
2
+ b
2
)x
2
+(b
/
3
+ 2a
2
b
/
2
+ 2a
2
a
/
2
b
/
1
+ 3a
3
b
/
1
+b
1
d
/
2
+ b
1
a
/
2
d
/
1
+ b
2
d
/
1
+ b
1
a
/
3
+ 2b
2
a
/
2
+ b
3
−b
1
b
/
1
c
/
2
)x
3
+
Now as in the case of ∆(a
n
) above, we use (5.3.7),(5.3.8), (5.3.9) and (5.3.10)
to drive the formula for ∆(b
n
). For example:
b
1
(ΦΦ
/
) = b
/
1
+ b
1
= (1 ⊗b
1
+ b
1
⊗1)(φ ⊗φ
/
),
b
2
(ΦΦ
/
) = b
/
2
+ 2a
2
b
/
1
+ b
1
d
/
1
+ b
1
a
/
2
+ b
2
,
= (1 ⊗b
2
+ b
2
⊗1 + 2a
2
⊗b
1
+ b
1
⊗d
1
+ b
1
⊗a
2
)(φ ⊗φ
/
),
b
3
(ΦΦ
/
) = b
/
3
+ 2a
2
b
/
2
+ 2a
2
a
/
2
b
/
1
+ 3a
3
b
/
1
+
b
1
d
/
2
+ b
1
a
/
2
d
/
1
+ b
2
d
/
1
+ b
1
a
/
3
+ 2b
2
a
/
2
+ b
3
−b
1
b
/
1
c
/
2
.
= (1 ⊗b
3
+ b
3
⊗1 + 2a
2
⊗b
2
+ 2a
2
⊗a
2
b
1
+ 3a
3
⊗b
1
+ b
1
⊗d
2
+b
1
⊗a
2
d
1
+ b
2
⊗d
1
+ b
1
⊗a
3
+ 2b
2
⊗a
2
−b
1
⊗b
1
c
2
)(φ ⊗φ
/
)
These imply that
∆(b
1
) = 1 ⊗b
1
+ b
1
⊗1,
∆(b
2
) = 1 ⊗b
2
+ b
2
⊗1 + 2a
2
⊗b
1
+ b
1
⊗d
1
+ b
1
⊗a
2
,
132
∆(b
3
) = 1 ⊗b
3
+ b
3
⊗1 + 2a
2
⊗b
2
+ 2a
2
⊗a
2
b
1
+ 3a
3
⊗b
1
+ b
1
⊗d
2
+b
1
⊗a
2
d
1
+ b
2
⊗d
1
+ b
1
⊗a
3
+ 2b
2
⊗a
2
−b
1
⊗b
1
c
2
.
In general since
b
n
(φφ
/
) = λ
˙
A(A
)B
n
+ λ
B(A
)D
n
−λ
˙
B(A
)B
C
n
,
by using Formalism (5.3.8) we get the following formula for ∆(b
n
), for n ≥ 1:
∆(b
n
) = 1 ⊗b
n
+
n
i=1
i
k=1
(k + 1)a
k+1
⊗(
l
1
+l
2
++l
k
=i
a
l
1
a
l
2
a
l
k
)b
n−i
+
n
i=1
i
k=1
b
k
⊗(
l
1
+l
2
++l
k
=i
a
l
1
a
l
2
a
l
k
)d
n−i
−
n
i=1
_
_
b
1
⊗b
i
c
n−i
+
i
j=1
j
k=1
(k + 1)b
k+1
⊗(
l
1
+l
2
++l
k
=j
a
l
1
a
l
2
a
l
k
)b
i−j
c
n−i
_
_
With a similar computation for ∆(c
n
), n ≥ 1 we have:
∆(c
n
) = 1 ⊗c
n
+
n
k=1
c
k
⊗
l
1
+l
2
++l
k
=n
a
l
1
a
l
2
a
l
k
+
n
i=1
i
k=1
d
k
⊗(
l
1
+l
2
++l
k
=i
a
l
1
a
l
2
a
l
k
)c
n−i
.
In particular for n = 2, 3:
∆(c
2
) = 1 ⊗c
2
+ c
2
⊗1,
∆(c
3
) = 1 ⊗c
3
+ c
3
⊗1 + 2c
2
⊗a
2
+ d
1
⊗c
2
.
133
Also for ∆(d
n
), n ≥ 1 we have:
∆(d
n
) = 1 ⊗d
n
+
n
i=1
i
k=1
(k + 1)c
k+1
⊗(
l
1
+l
2
++l
k
=i
a
l
1
a
l
2
a
l
k
)b
n−i
+
n
i=1
i
k=1
d
k
⊗(
l
1
+l
2
++l
k
=i
a
l
1
a
l
2
a
l
k
)d
n−i
−
n
i=1
_
_
d
1
⊗b
i
c
n−i
+
i
j=1
j
k=1
(k + 1)d
k+1
⊗(
l
1
+l
2
++l
k
=j
a
l
1
a
l
2
a
l
k
)b
i−j
c
n−i
_
_
In particular for n = 1, 2, 3:
∆(d
1
) = 1 ⊗d
1
+ d
1
⊗1,
∆(d
2
) = 1 ⊗d
2
+ d
2
⊗1 + 2c
2
⊗b
1
+ d
1
⊗d
1
+ d
1
⊗a
2
,
∆(d
3
) = 1 ⊗d
3
+ d
3
⊗1 + 2c
2
⊗b
2
+ 2c2 ⊗a
2
b
1
+ 3c
3
⊗b
1
+ d
1
⊗d
2
+d
1
⊗a
2
d
1
+ d
2
⊗d
1
+ d
1
⊗a
3
+ 2d
2
⊗a
2
−d
1
⊗b
1
c
2
.
5.4 Actions and coactions
As mentioned before, we want to construct a bicrossproduct super Hopf algebra
F(G
s
2
)U(g
s
1
) analogous to the non super case [9, 17]. For that of course we need
a (left) action of U(g
s
1
) on F(G
s
2
) and a (right) coaction of F(G
s
2
) on U(g
s
1
). In this
section we provide these action and coaction in terms of generators. This means
that, for the action we provide the actions of generators X, Y, Z, U, V, W of U(g
s
1
) on
generators a
n
, b
n
, c
n
, d
n
of F(G
s
2
), and also for the coaction we provide the coactions
of F(G
s
2
) on generators X, Y, Z, U, V, W of U(g
s
1
). We proceed to actions ﬁrst.
134
5.4.1 Actions of X
We recall that as in the non super case [9, 17], for X =
_
_
_
_
_
_
1 0 0
0 0 0
0 0 0
_
_
_
_
_
_
∈ g
s
1
we have
e
tX
=
_
_
_
_
_
_
e
t
0 0
0 1 0
0 0 1
_
_
_
_
_
_
≡ (e
t
x, θ) ∈ G
s
1
, (t even).
And the action Y a
n
can be realised by:
(X a
n
)(φ) =
d
dt
[
t=0
a
n
(φ e
tX
),
for any φ = (A(x) + B(x)θ, C(x) + D(x)θ) ∈ G
s
2
. So, we compute
φ e
tX
= π
2
(φe
tX
)
=
_
_
e
t
0
0 1
_
_
−1
(A(e
t
x) + B(e
t
x)θ, C(e
t
x) + D(e
t
x)θ) −(A(0), C(0))
= e
−t
_
_
1 0
0 e
t
_
_
_
_
A(e
t
x) + B(e
t
x)θ
C(e
t
x) + D(e
t
x)θ
_
_
= (e
−t
A(e
t
x) + e
−t
B(e
t
x)θ, C(e
t
x) + D(e
t
x)θ).
Then we have
a
n
(φ e
tX
) = (1/n!)e
(n−1)t
A
(n)
(0),
(X a
n
)(φ) =
d
dt
[
t=0
a
n
(φ e
tY
) = (n −1)a
n
(φ).
This implies that
X a
n
= (n −1)a
n
.
In the same way we have
X b
n
= (n −1)b
n
,
135
X c
n
= nc
n
,
X d
n
= nd
n
.
5.4.2 Actions of Y
As before, for Y =
_
_
_
_
_
_
0 1 0
0 0 0
0 0 0
_
_
_
_
_
_
∈ g
s
1
we have:
e
tY
=
_
_
_
_
_
_
1 t 0
0 1 0
0 0 1
_
_
_
_
_
_
≡ (x + t, θ) ∈ G
s
1
, (t even).
And the action Y a
n
can be realised by:
(Y a
n
)(φ) =
d
dt
[
t=0
a
n
(φ e
tY
),
for any φ = (A(x) + B(x)θ, C(x) + D(x)θ) ∈ G
s
2
. So, if we compute
φ e
tY
= π
2
(φe
tY
)
=
_
_
˙
A(t) B(t)
˙
C(t) D(t)
_
_
−1
(A(x + t) + B(x + t)θ, C(x + t) + D(x + t) −(A(t), C(t))
= (
1
˙
A(t)D(t)
)
_
_
_
D(t) + (
B(t)
˙
C(t)
˙
A(t)
) −B(t)
−
˙
C(t)
˙
A(t) + (
˙
C(t)B(t)
D(t)
)
_
_
_
_
_
A(x + t) −A(t) + B(x + t)θ
C(x + t) −C(t) + D(x + t)θ
_
_
= (
1
˙
A(t)D(t)
)((
˙
A(t)D(t) + B(t)
˙
C(t))(
A(x+t)
˙
A(t)
−
A(t)
˙
A(t)
+
B(x+t)
˙
A(t)
θ)
−B(t)C(x + t) + B(t)C(t) −B(t)D(x + t)θ ,
−
˙
C(t)A(x + t) +
˙
C(t)A(t) −
˙
C(t)B(x + t)
136
+ ((
˙
A(t)D(t) +
˙
C(t)B(t))(
C(x+t)
D(t)
−
C(t)
D(t)
+
D(x+t)
D(t)
θ)
= ([(1 +
B(t)
˙
C(t)
˙
A(t)D(t)
)(
A(x+t)
˙
A(t)
−
A(t)
˙
A(t)
) −
B(t)C(x+t)
˙
A(t)D(t)
+
B(t)C(t)
˙
A(t)D(t)
]
+ [(1 +
B(t)
˙
C(t)
˙
A(t)D(t)
)
B(x+t)
˙
A(t)
−
B(t)D(x+t)
˙
A(t)D(t)
]θ ,
[−
˙
C(t)A(x+t)
˙
A(t)D(t)
+
˙
C(t)A(t)
˙
A(t)D(t)
+ (1 +
˙
C(t)B(t)
˙
A(t)D(t)
)(
C(x+t)
D(t)
−
C(t)
D(t)
)]
+ [
−
˙
C(t)B(x+t)
˙
A(t)D(t)
+ (1 +
˙
C(t)B(t)
˙
A(t)D(t)
)
D(x+t)
D(t)
]θ),
then we have
a
n
(φ e
tY
) = (1/n!)[(1 +
B(t)
˙
C(t)
˙
A(t)D(t)
)(
A
(n)
(t)
˙
A(t)
) −
B(t)C
(n)
(t)
˙
A(t)D(t)
]
(1/n!)[(1 +
(
˙
B(0)t)(
¨
C(0)t)
(1 +
¨
A(0)t)(1 +
˙
D(0)t)
)(
A
(n)
(0) + A
(n+1)
(0)t
1 +
¨
A(0)t
)
−
(
˙
B(0)t)(C
(n)
(0) + C
(n+1)
(0)t)
(1 +
¨
A(0)t)(1 +
˙
D(0)t)
]
(1/n!)[A
(n)
(0) + (A
(n+1)
(0) −A
(n)
(0)
¨
A(0))t −(
˙
B(0)C
(n)
(0))t],
where means equality up to terms of order t
2
. Also
(Y a
n
)(φ) =
d
dt
[
t=0
a
n
(φ e
tY
)
= (1/n!)[(A
(n+1)
(0) −A
(n)
(0)
¨
A(0)) −(
˙
B(0)C
(n)
(0))]
= [(n + 1)a
n+1
−2a
n
a
2
−b
1
c
n
](φ).
This implies that
Y a
n
= (n + 1)a
n+1
−2a
n
a
2
−b
1
c
n
.
137
In the same way we obtain:
Y b
n
= (n + 1)b
n+1
−2b
n
a
2
−b
1
d
n
,
Y c
n
= −2c
2
a
n
+ (n + 1)c
n+1
−c
n
d
1
,
Y d
n
= −2c
2
b
n
+ (n + 1)d
n+1
−d
n
d
1
.
5.4.3 Actions of Z
Again for Z =
_
_
_
_
_
_
0 0 0
0 0 0
0 0 1
_
_
_
_
_
_
∈ g
s
1
we have
e
tZ
=
_
_
_
_
_
_
1 0 0
0 1 0
0 0 e
t
_
_
_
_
_
_
≡ (x, e
t
θ) ∈ G
s
1
, (t even).
And for any φ = (A(x) + B(x)θ, C(x) + D(x)θ) ∈ G
s
2
,
(Z a
n
)(φ) =
d
dt
[
t=0
a
n
(φ e
tZ
).
138
So, we ﬁrst compute
φ e
tZ
= π
2
(φe
tZ
)
=
_
_
1 0
0 e
t
_
_
−1
(A(x) + B(x)e
t
θ, C(x) + D(x)e
t
θ) −(A(0), C(0))
= e
−t
_
_
e
t
0
0 1
_
_
_
_
A(x) + B(x)e
t
θ
C(x) + D(x)e
t
θ
_
_
= (A(x) + e
t
B(x)θ, e
−t
C(x) + D(x)θ).
Then we have
a
n
(φ e
tZ
) = (1/n!)A
(n)
(0),
(Z a
n
)(φ) =
d
dt
[
t=0
a
n
(φ e
tZ
) = 0.
This implies that
Z a
n
= 0.
In the same way we have
Z b
n
= b
n
,
Z c
n
= −c
n
,
Z d
n
= 0.
139
5.4.4 Actions of U
Again we have:
U =
_
_
_
_
_
_
0 0 1
0 0 0
0 0 0
_
_
_
_
_
_
∈ g
s
1
, e
tU
=
_
_
_
_
_
_
1 0 t
0 1 0
0 0 1
_
_
_
_
_
_
≡ (x + tθ, θ) ∈ G
s
1
, (t odd).
And for any φ = (A(x) + B(x)θ, C(x) + D(x)θ) ∈ G
s
2
,
(U a
n
)(φ) =
d
dt
[
t=0
a
n
(φ e
tU
).
For that, we ﬁrst compute
φ e
tU
= π
2
(φe
tU
)
=
_
_
1 t
0 1
_
_
−1
(A(x + tθ) + B(x + tθ)θ, C(x + tθ) + D(x + tθ)θ −(A(0), C(0))
=
_
_
1 −t
0 1
_
_
_
_
A(x + tθ) + B(x + tθ)θ
C(x + tθ) + D(x + tθ)θ
_
_
= (A(x + tθ) + B(x + tθ)θ −tC(x + tθ) −tD(x + tθ)θ, C(x + tθ) + D(x + tθ)θ)
= (A(x) +
˙
A(x)tθ + B(x)θ −tC(x) −tD(x)θ, C(x) +
˙
C(x)tθ + D(x)θ)
= ([A(x) −tC(x] + [t
˙
A(x) + B(x) −tD(x)]θ, [C(x)] + [−t
˙
C(x) + D(x)]θ)
And then
a
n
(φ e
tU
) = (1/n!)A
(n)
(0) −t(1/n!)C
n
(0),
(U a
n
)(φ) =
d
dt
[
t=0
a
n
(φ e
tU
) = −(1/n!)C
n
(0) = −c
n
(φ).
This implies that (after reversing the signs, c.f. Remark 5.4.1)
U a
n
= c
n
.
140
In the same way we have
U b
n
= −(n + 1)a
n+1
+ d
n
,
U c
n
= 0,
U d
n
= (n + 1)c
n+1
.
Remark 5.4.1. As one notices we have reversed the signs in actions of U. This
change of sign is crucial for the actions of U to be compatible with the other actions
and structures. We will see this in, for example, the proof of Lemma (5.4.1).
5.4.5 Actions of V
We have
V =
_
_
_
_
_
_
0 0 0
0 0 0
−1 0 0
_
_
_
_
_
_
∈ g
s
1
, e
tV
=
_
_
_
_
_
_
1 0 0
0 1 0
−t 0 1
_
_
_
_
_
_
≡ (x, −tx + θ) ∈ G
s
1
, (t odd).
And for any φ = (A(x) + B(x)θ, C(x) + D(x)θ) ∈ G
s
2
,
(V a
n
)(φ) =
d
dt
[
t=0
a
n
(φ e
tV
).
For that, we ﬁrst compute
φ e
tV
= π
2
(φe
tV
)
=
_
_
1 0
−t 1
_
_
−1
(A(x) + B(x)(−tx + θ), C(x) + D(x)(−tx + θ) −(A(0), C(0))
=
_
_
1 0
t 1
_
_
_
_
A(x) + B(x)(−tx + θ)
C(x) + D(x)(−tx + θ)
_
_
141
=
_
_
1 0
t 1
_
_
_
_
A(x) + txB(x) + B(x)θ
C(x) −txD(x) + D(x)θ
_
_
= (A(x) + txB(x) + B(x)θ, tA(x) + t
2
xB(x) + tB(x)θ + C(x) −txD(x) + D(x)θ)
= ([A(x) + txB(x)] + [B(x)]θ, [tA(x) + C(x) −txD(x)] + [+tB(x) + D(x)]θ)
And then
a
n
(φ e
tV
) = (1/n!)A
(n)
(0) + t(1/(n −1)!)B
(n−1)
(0),
(V a
n
)(φ) =
d
dt
[
t=0
a
n
(φ e
tV
) = b
n−1
(φ).
This implies that
V a
n
= b
n−1
.
In the same way we have
V b
n
= 0,
V c
n
= a
n
−d
n−1
,
V d
n
= +b
n
.
5.4.6 Actions of W
We have
W =
_
_
_
_
_
_
0 0 0
0 0 0
0 −1 0
_
_
_
_
_
_
∈ g
s
1
, e
tW
=
_
_
_
_
_
_
1 0 0
0 1 0
0 −t 1
_
_
_
_
_
_
≡ (x, θ −t) ∈ G
s
1
, (t odd).
And for any φ = (A(x) + B(x)θ, C(x) + D(x)θ) ∈ G
s
2
,
(W a
n
)(φ) =
d
dt
[
t=0
a
n
(φ e
tW
).
So, if we compute
φ e
tW
= π
2
(φe
tW
)
142
=
_
_
1 −
˙
B(0)t 0
−
˙
D(0)t 1
_
_
−1
(A(x) + B(x)(θ −t), C(x) + D(x)(θ −t)
−(A(0) −B(0)t, C(0) −D(0)t)
= (
1
1−
˙
B(0)t
)
_
_
1 0
t
˙
D(0) 1 + t
˙
B(0)
_
_
_
_
A(x) + B(x)θ + tB(x)
C(x) + D(x)θ −tD(x) + t
_
_
= (
1
1−
˙
B(0)t
)(A(x) −tB(x) + B(x)θ,
t
˙
D(0)A(x) + t
˙
D(0)B(x)θ +
˙
D(0)t
2
B(x) + C(x) + t
˙
B(0)C(x)
+ D(x)θ + t
˙
B(0)D(x)θ −tD(x) −t
˙
B(0)tD(x) + t + t
˙
B(x)t)
= (1/(1 + t
˙
B(0)))([A(x) −tB(x)] + [B(x)]θ,
[t
˙
D(0)A(x) + (1 + t
˙
B(0))C(x) −tD(x) + t] + [t
˙
D(0)B(x) + (1 + t
˙
B(0))D(x)]θ)
= ([
A(x)
(1+t
˙
B(0))
+
−tB(x)
(1+t
˙
B(0))
] + [
B(x)
(1+t
˙
B(0))
]θ,
[
t
˙
D(0)A(x)
(1+t
˙
B(0))
+ C(x) +
−tD(x)
(1+t
˙
B(0))
+
t
(1+t
˙
B(0))
] + [
t
˙
D(0)B(x)
(1+t
˙
B(0))
+ D(x)]θ),
then
a
n
(φ e
tW
) = (1 −t
˙
B(0))(1/n!)A
(n)
(0) + (1 −t
˙
B(0))t(1/n!)B
n
(0),
(Wa
n
)(φ) =
d
dt
[
t=0
a
n
(φe
tW
) = −
˙
B(0)(1/n!)A
(n)
(0)+(1/n!)B
n
(0) = b
1
a
n
−b
n
(φ).
This implies that
W a
n
= −b
1
a
n
+ b
n
.
In the same way we have
W b
n
= −b
1
b
n
,
W c
n
= d
1
a
n
−d
n
,
W d
n
= d
1
b
n
.
143
Using all above actions we can actually prove the following lemma.
Lemma 5.4.1. Let us deﬁne the actions of X, Y, Z, U, V, W on a
n
, b
n
, c
n
, d
n
by the
following relations:
X a
n
= (n −1)a
n
, X b
n
= (n −1)b
n
, X c
n
= nc
n
, X d
n
= nd
n
,
Y a
n
= (n + 1)a
n+1
−2a
n
a
2
−b
1
c
n
, Y b
n
= (n + 1)b
n+1
−2b
n
a
2
−b
1
d
n
,
Y c
n
= −2c
2
a
n
+ (n + 1)c
n+1
−c
n
d
1
, Y d
n
= −2c
2
b
n
+ (n + 1)d
n+1
−d
n
d
1
,
Z a
n
= 0, Z b
n
= b
n
, Z c
n
= −c
n
, Z d
n
= 0,
U a
n
= c
n
, U b
n
= −(n +1)a
n+1
+d
n
, U c
n
= 0, U d
n
= (n +1)c
n+1
,
V a
n
= b
n−1
, V b
n
= 0, V c
n
= a
n
−d
n−1
, V d
n
= +b
n
,
W a
n
= −b
1
a
n
+b
n
, W b
n
= −b
1
b
n
, W c
n
= d
1
a
n
−d
n
, W d
n
= d
1
b
n
,
(5.4.22)
and extend those actions linearly to an action of U(g
s
1
) on F(G
s
2
) by:
(gh) a = g (h a), (5.4.23)
h (ab) := (−1)
[a[[h
(2)
[
(h
(1)
a)(h
(2)
b), (5.4.24)
for a, b in F(G
s
2
) and g, h in U(g
s
1
). Then, F(G
s
2
) is a left U(g
s
1
)module algebra.
Proof. It is enough to show that this action is well deﬁned on basis in the
sense that it is consistent with bracket relations (5.3.5) and (5.3.6). We check this
for some of the basis elements. The rest would be the same:
(XY ) a
n
(5.4.23)
= X (Y a
n
)
(5.4.22)
= X ((n + 1)a
n+1
−2a
n
a
2
−b
1
c
n
)
= (n + 1)(X a
n+1
) −2X (a
n
a
2
) −X (b
1
c
n
)
144
(5.3.6),(5.4.24)
= n(n+1)a
n+1
−2(a
n
(Xa
2
) +(Xa
n
)a
2
) −(b
1
(Xc
n
) +(Xb
1
)c
n
)
= (n
2
+ n)a
n+1
−2na
n
a
2
−nb
1
c
n
,
and
(Y X) a
n
= Y (X a
n
) = Y ((n −1)a
n
)
= (n −1)((n + 1)a
n+1
−2a
n
a
2
−b
1
c
n
)
= (n
2
−1)a
n+1
−2(n −1)a
n
a
2
−(n −1)b
1
c
n
.
Therefore
[X, Y ] a
n
(5.3.6)
= (XY −Y X) a
n
= (n + 1)a
n+1
−2a
n
a
2
−b
1
c
n
(5.4.22)
= Y a
n
,
which is consistent with the relation [X, Y ] = Y of (5.3.5). Let us check some more
examples:
(XV ) c
n
= X (V c
n
) = X (a
n
−d
n−1
) = (X a
n
) −(X d
n−1
)
= (n −1)a
n
−(n −1)d
n−1
,
and
(V X) c
n
= V (X c
n
) = V (nc
n
) = na
n
−nd
n−1
.
Therefore
[X, V ] c
n
= (XV −V X) c
n
= −(a
n
−d
n−1
) = (−V ) c
n
,
which is consistent with the relation [X, V ] = −V of (5.3.5).
(Y V ) c
n
= Y (V c
n
) = Y (a
n
−d
n−1
) = (Y a
n
) −(Y d
n−1
)
= ((n + 1)a
n+1
−2a
n
a
2
−b
1
c
n
) −(−2c
2
b
n−1
+ nd
n
−d
n−1
d
1
)
= (n + 1)a
n+1
−2a
n
a
2
−b
1
c
n
+ 2c
2
b
n−1
−nd
n
+ d
n−1
d
1
,
145
(V Y ) c
n
= V (Y c
n
) = V (−2c
2
a
n
+ (n + 1)c
n+1
−c
n
d
1
)
= −2(−c
2
(V a
n
) + (V c
2
)a
n
) + (n + 1)(a
n+1
−d
n
)
−(−c
n
(V d
1
) + (V c
n
)d
1
)
= 2c
2
b
n−1
−2a
2
a
n
+ 2d
1
a
n
+ (n + 1)a
n+1
−(n + 1)d
n
+ c
n
b
1
−a
n
d
1
+ d
n−1
d
1
.
Therefore
[Y, V ] c
n
= (Y V −V Y ) c
n
= d
n
−a
n
d
1
= −(d
1
a
n
+ d
n
) = (−W) c
n
,
which is consistent with the relation [Y, V ] = −W.
(UV ) b
n
= U (V b
n
) = U (0) = 0,
(V U) b
n
= V (U b
n
) = V (−(n + 1)a
n+1
+ d
n
)
= −(n + 1)(V a
n+1
) + V d
n
= −(n + 1)b
n
+ b
n
= −nb
n
.
Therefore
[U, V ] b
n
= (UV + V U) b
n
= −nb
n
= (−(X + Z)) b
n
,
which is consistent with the relation [U, V ] = −(X + Z).
(UV ) c
n
= U (V c
n
) = U (a
n
−d
n−1
) = c
n
−nc
n
= (1 −n)c
n
,
(V U) c
n
= V (U c
n
) = V (0) = 0.
Therefore
[U, V ] c
n
= (UV + V U) c
n
= −(n −1)c
n
= (−(X + Z)) c
n
,
which agrees with the relation [U, V ] = −(X + Z).
146
(UW) c
n
= U (W c
n
) = U (d
1
a
n
−d
n
)
= (d
1
(U a
n
) + (U d
1
)a
n
) −(n + 1)c
n+1
= d
1
c
n
+ 2c
2
a
n
−(n + 1)c
n+1
,
(WU) c
n
= W (U c
n
) = W (0) = 0.
Therefore
[U, W] c
n
= (UW + WU) c
n
= d
1
c
n
+ 2c
2
a
n
−(n + 1)c
n+1
= (−Y ) c
n
,
which agrees with the relation [U, W] = −Y .
(V W) a
n
= V (W a
n
) = V (−b
1
a
n
+ b
n
)
= −(−b
1
(V a
n
) + (V b
1
)a
n
) + 0 = b
1
b
n−1
,
(WV ) a
n
= W (V a
n
) = W b
n−1
= −b
1
b
n−1
.
Therefore
[V, W] a
n
= (V W + WV ) a
n
= 0 = (0) a
n
,
which agrees with the relation [V, W] = 0.
(V W) d
n
= V (W d
n
) = V (d
1
b
n
)
= (d
1
(V b
n
) + (V d
1
)b
n
) = d
1
(0) + b
1
b
n
= b
1
b
n
,
(WV ) d
n
= W (V d
n
) = W b
n
= −b
1
b
n
.
Therefore
[V, W] d
n
= (V W + WV ) d
n
= b
1
b
n
−b
1
b
n
= 0 = (0) d
n
,
147
which agrees with the relation [V, W] = 0.
(UW) d
n
= U (W d
n
) = U (d
1
b
n
)
= (d
1
(U b
n
) + (U d
1
)b
n
) = d
1
(−(n + 1)a
n+1
+d
n
) + 2c
2
b
n
= −(n + 1)d
1
a
n+1
+
d
1
d
n
+ 2c
2
b
n
,
(WU)d
n
= W(Ud
n
) = W((n+1)c
n+1
) = (n+1)d
1
a
n+1
−(n+1)d
n+1
.
Therefore
[U, W] d
n
= (UW + WU) d
n
= 2c
2
b
n
−(n + 1)d
n+1
+ d
1
d
n
= (−Y ) d
n
,
which agrees with the relation [U, W] = −Y .
5.4.7 Coactions
By using almost the same method we can ﬁnd the coactions and prove the following
lemma. We denote the right coaction of F(G
s
2
) on U(g
s
1
), by ∇
r
: U(g
s
1
) → U(g
s
1
) ⊗
F(G
s
2
).
Lemma 5.4.2. Let us deﬁne the coactions of F(G
s
2
) on generators X, Y, Z, U, V, W
of U(g
s
1
) by:
∇
r
(X) = X ⊗1, ∇
r
(Y ) = 2X ⊗a
2
+ Y ⊗1 + Z ⊗d
1
+ U ⊗b
1
+ 2V ⊗c
2
,
∇
r
(Z) = Z ⊗1, ∇
r
(U) = U ⊗1, ∇
r
(V ) = V ⊗1,
∇
r
(W) = X ⊗b
1
+ V ⊗d
1
+ W ⊗1, (5.4.25)
148
and extend them linearly to U(g
s
1
) by:
∇
r
(gh) = (−1)
[h
(1)
[([g
(1)
(2)
[+[g
(2)
[)
g
(1)
(1)
h
(1)
⊗g
(1)
(2)
(g
(2)
h
(2)
), (5.4.26)
for all g and h in U(g
s
1
). Then U(g
s
1
) is a right F(G
s
2
)comodule coalgebra.
Proof. It is very easy and straightforward to check the coaction property,
(id ⊗∆)∇
r
(h) = (∇
r
⊗id)∇
r
(h), for all h in U(g
s
1
), in other words:
h
(1)
⊗h
(2)
(1)
⊗h
(2)
(2)
= h
(1)(1)
⊗h
(1)(2)
⊗h
(2)
.
We leave that to the reader. We prove that this coaction is well deﬁned in the sense
that it is consistent with bracket relations (5.3.5) and (5.3.6). We show this for a few
cases, the rest are the same. By formula (5.4.26) we have:
∇
r
(Y W) = (−1)
[W
(1)
[([Y
(1)
(2)
[+[Y
(2)
[)
Y
(1)
(1)
W
(1)
⊗Y
(1)
(2)
(Y
(2)
W
(2)
)
= [(−1)
[W
(1)
[[Y [
W
(1)
⊗(Y W
(2)
)] ∆(Y ) = 1 ⊗Y +Y ⊗1
+[(−1)
[W
(1)
[[Y
(2)
[
Y
(1)
W
(1)
⊗Y
(2)
W
(2)
] ∆(Y ) = 1⊗Y +Y ⊗1
= [X ⊗(Y b
1
) + V ⊗(Y d
1
) + W ⊗(Y 1)]
+ [2X
2
⊗a
2
b
1
+ 2XV ⊗a
2
d
1
+ 2XW ⊗a
2
+ Y X ⊗b
1
+ Y V ⊗d
1
+ Y W ⊗1
+ ZX ⊗d
1
b
1
+ ZV ⊗d
1
d
1
+ ZW ⊗d
1
+ UX ⊗b
1
b
1
−UV ⊗b
1
d
1
−UW ⊗b
1
+ 2V X ⊗c
2
b
1
−2V
2
⊗c
2
d
1
−2V W ⊗c
2
]
= [X ⊗(2b
2
−2b
1
a
2
−b
1
d
1
) + V ⊗(−2c
2
b
1
+ 2d
2
−d
1
d
1
) + W ⊗(0)]
+ [2X
2
⊗a
2
b
1
+ 2XV ⊗a
2
d
1
+ 2XW ⊗a
2
149
+ Y X ⊗b
1
+ Y V ⊗d
1
+ Y W ⊗1
+ ZX ⊗d
1
b
1
+ ZV ⊗d
1
d
1
+ ZW ⊗d
1
+ UX ⊗b
1
b
1
−UV ⊗b
1
d
1
−UW ⊗b
1
+ 2V X ⊗c
2
b
1
−2V
2
⊗c
2
d
1
−2V W ⊗c
2
]
= [X ⊗2b
2
−X ⊗2b
1
a
2
−X ⊗b
1
d
1
−V ⊗2c
2
b
1
+ V ⊗2d
2
−V ⊗d
1
d
1
]
+ [2X
2
⊗a
2
b
1
+ 2XV ⊗a
2
d
1
+ 2XW ⊗a
2
+ Y X ⊗b
1
+ Y V ⊗d
1
+ Y W ⊗1
+ ZX ⊗d
1
b
1
+ ZV ⊗d
1
d
1
+ ZW ⊗d
1
+ UX ⊗b
1
b
1
−UV ⊗b
1
d
1
−UW ⊗b
1
+ 2V X ⊗c
2
b
1
−2V
2
⊗c
2
d
1
−2V W ⊗c
2
]
From the other hand:
∇
r
(WY ) = (−1)
[Y
(1)
[([W
(1)
(2)
[+[W
(2)
[)
W
(1)
(1)
Y
(1)
⊗W
(1)
(2)
(W
(2)
Y
(2)
)
= [(−1)
[Y
(1)
[[W[
Y
(1)
⊗(W Y
(2)
)] ∆(W) = 1 ⊗W +W ⊗1
+[(−1)
[Y
(1)
[[W
(2)
[
W
(1)
Y
(1)
⊗W
(2)
Y
(2)
] ∆(W) = 1⊗W +W ⊗1
= [2X⊗(Wa
2
)+Y ⊗(W1)+Z⊗(Wd
1
)−U ⊗(Wb
1
)−2V ⊗(Wc
2
)]
+ [2X
2
⊗a
2
b
1
+ 2V X ⊗a
2
d
1
+ 2WX ⊗a
2
+ XY ⊗b
1
+ V Y ⊗d
1
+ WY ⊗1
+ XZ ⊗d
1
b
1
+ V Z ⊗d
1
d
1
+ WZ ⊗d
1
−XU ⊗b
1
b
1
+ V U ⊗b
1
d
1
+ WU ⊗b
1
−2XV ⊗(−c
2
b
1
) + 2V
2
⊗c
2
d
1
+ 2WV ⊗c
2
]
= [2X ⊗(−b
1
a
2
+b
2
) +Y ⊗(0) +Z ⊗(d
1
b
1
) −U ⊗(b
1
b
1
) −2V ⊗(d
1
a
2
−d
2
)]
+ [2X
2
⊗a
2
b
1
+ 2V X ⊗a
2
d
1
+ 2WX ⊗a
2
+ XY ⊗b
1
+ V Y ⊗d
1
+ WY ⊗1
150
+ XZ ⊗d
1
b
1
+ V Z ⊗d
1
d
1
+ WZ ⊗d
1
−XU ⊗b
1
b
1
+ V U ⊗b
1
d
1
+ WU ⊗b
1
−2XV ⊗(−c
2
b
1
) + 2V
2
⊗c
2
d
1
+ 2WV ⊗c
2
]
= [−2X ⊗b
1
a
2
+ 2X ⊗b
2
+ Z ⊗d
1
b
1
−2V ⊗d
1
a
2
+ 2V ⊗d
2
)]
+ [2X
2
⊗a
2
b
1
+ 2V X ⊗a
2
d
1
+ 2WX ⊗a
2
+ XY ⊗b
1
+ V Y ⊗d
1
+ WY ⊗1
+ XZ ⊗d
1
b
1
+ V Z ⊗d
1
d
1
+ WZ ⊗d
1
−XU ⊗b
1
b
1
+ V U ⊗b
1
d
1
+ WU ⊗b
1
−2XV ⊗(−c
2
b
1
) + 2V
2
⊗c
2
d
1
+ 2WV ⊗c
2
]
Thus
∇
r
([Y, W])
(5.3.6)
= ∇
r
(Y W −WY ) = ∇
r
(Y W) −∇
r
(WY )
(5.3.6)
= 2[X, X] ⊗a
2
b
1
+ 2[X, V ] ⊗a
2
d
1
+ 2[X, W] ⊗a
2
+[Y, X] ⊗b
1
+[Y, V ] ⊗d
1
+[Y, W] ⊗1+[Z, X] ⊗d
1
b
1
+[Z, V ] ⊗d
1
d
1
+[Z, W] ⊗d
1
+(UX +XU) ⊗b
1
b
1
−[U, V ] ⊗b
1
d
1
−[U, W] ⊗b
1
+2[V, X] ⊗c
2
b
1
−2[V, V ] ⊗
c
2
d
1
−2[V, W] ⊗c
2
−X ⊗b
1
d
1
−V ⊗2c
2
b
1
−V ⊗d
1
d
1
−Z ⊗b
1
d
1
+ 2V ⊗d
1
a
2
(5.3.5)
= 0.
This is consistent with the relation [Y, W] = 0 from (5.3.5). With the same method
we have:
∇
r
(V W) = (−1)
[W
(1)
[([V
(1)
(2)
[+[V
(2)
[)
V
(1)
(1)
W
(1)
⊗V
(1)
(2)
(V
(2)
W
(2)
)
= [(−1)
[W
(1)
[[V [)
W
(1)
⊗(V W
(2)
)] ∆(V ) = 1 ⊗V +V ⊗1
+[(−1)
[W
(1)
[[V
(2)
[
V
(1)
W
(1)
⊗V
(2)
W
(2)
] ∆(V ) = 1⊗V +V ⊗1
= [(−1)
[W
(1)
[
W
(1)
⊗(V W
(2)
)]
151
+ [V
(1)
W
(1)
⊗W
(2)
]
= [X ⊗(V b
1
) −V ⊗(V d
1
) −W ⊗(V 1)]
+ [V X ⊗b
1
+ V
2
⊗d
1
+ V W ⊗1
= [−V ⊗b
1
]
+ [V X ⊗b
1
+ V
2
⊗d
1
+ V W ⊗1,
and
∇
r
(WV ) = (−1)
[V
(1)
[([W
(1)
(2)
[+[W
(2)
[)
W
(1)
(1)
V
(1)
⊗W
(1)
(2)
(W
(2)
V
(2)
)
= [(−1)
[V
(1)
[[W[)
V
(1)
⊗(WV
(2)
)] ∆(W) = 1 ⊗W +W⊗1
+[(−1)
[V
(1)
[[W
(2)
[
W
(1)
V
(1)
⊗W
(2)
V
(2)
] ∆(W) = 1⊗W +W ⊗1
= [−V ⊗(0)]
+ [(−1)
[W
(2)
[
W
(1)
V ⊗W
(2)
]
= −XV ⊗b
1
+ V
2
⊗d
1
+ WV ⊗1.
Thus
∇
r
([V, W]) = ∇
r
(V W + WV ) = ∇
r
(V W) +∇
r
(WV )
= [V, X] ⊗b
1
+ [V, V ] ⊗d
1
+ [V, W] ⊗1 −v ⊗b
1
= 0,
which agrees with [V, W] = 0 of relations (5.3.5).
We also leave it to the reader to check that U(g
s
1
) is a right F(G
s
2
)comodule
coalgebra, i.e., for all h in U(g
s
1
) :
h
(1)
(1)
⊗h
(1)
(2)
⊗h
(2)
= (−1)
[h
(2)
(1)
[[h
(1)
(2)
[
h
(1)
(1)
⊗h
(2)
(1)
⊗h
(1)
(2)
h
(2)
(2)
.
152
5.5 Compatibilities and the super Hopf algebra
H
s
1
To complete the construction of the bicrossproduct super Hopf algebra F(G
s
2
)U(g
s
1
)
we need to check a certain compatibility conditions between the actions and coactions
introduced in the last section. These compatibility conditions are in fact a super
version of those mentioned in Lemma (5.1.1). In this section we establish a super
version of Lemma (5.1.1) and show that the actions and coactions deﬁned in the
last section satisfy the compatibility conditions. Therefore we have a bicrossproduct
super Hopf algebra F(G
s
2
)U(g
s
1
) that we will denote it by H
s
1
, and call it the super
version of H
1
.
Lemma 5.5.1. (c.f. Lemma (5.1.1)) Let A and H be two super Hopf algebras such
that A is a left Hmodule algebra, and H is right Acomodule coalgebra. Let further
more these structures satisfy the following compatibility conditions:
∆(h a) = (−1)
[a
(1)
[([h
(1)
(2)
[+[h
(2)
[)
h
(1)
(1)
a
(1)
⊗h
(1)
(2)
(h
(2)
a
(2)
), (5.5.27)
ε(h a) = ε(h)ε(a),
∇
r
(gh) = (−1)
[h
(1)
[([g
(1)
(2)
[+[g
(2)
[)
g
(1)
(1)
h
(1)
⊗g
(1)
(2)
(g
(2)
h
(2)
), (5.5.28)
∇
r
(1) = 1 ⊗1,
(−1)
[h
(1)
[[h
(2)
(1)
[+[a[[h
(2)
(2)
[
h
(2)
(1)
⊗(h
(1)
a)h
(2)
(2)
= h
(1)
(1)
⊗h
(1)
(2)
(h
(2)
a),
(5.5.29)
for any a, b in A and g, h in H, where we have denoted the actions by, h a and the
coactions by ∇
r
(h) = h
(0)
⊗h
(1)
. Then the super vector space A⊗H can be equipped
153
with a super Hopf algebra structure as follows:
(a ⊗h)(b ⊗g) = (−1)
[h
(2)
[[b[
a(h
(1)
b) ⊗h
(2)
g, (5.5.30)
∆(a ⊗h) = (−1)
[h
(1)
(1)
[[a
(2)
[
a
(1)
⊗h
(1)
(1)
⊗a
(2)
h
(1)
(2)
⊗h
(2)
, (5.5.31)
ε(a ⊗h) = ε(a)ε(h),
S(a ⊗h) = (−1)
[h
(1)
[[a[
(1 ⊗S(h
(1)
))(S(ah
(2)
) ⊗1),
for any a, b in A and g, h in H. We call this super Hopf algebra the leftright bi
crossproduct super Hopf algebra AH.
Remark 5.5.1. Before starting the proof, we have to give a very important remark
concerning the signs when working with super Hopf algebras. In the following compu
tation whenever we use the higher coproduct of ∆
H
, i.e.,
∆
n
(h) = h
(1)
⊗h
(2)
⊗ ⊗h
(n+1)
, ∀h ∈ H, (5.5.32)
to keep track of signs in a correct way we have to ﬁrst use diﬀerent copies of h like h
/
,
h
//
, h
///
just in terms of notation, and after ﬁnding the right signs we rename those
copies by h again. Of course in the nonsuper case there is no need for this.
Proof. We prove that ∆, deﬁned in (5.5.31), is an algebra map.
First we have:
∆(a ⊗h) ∆(b ⊗g)
(5.5.31)
= (−1)
β
0
(a
(1)
⊗h
(1)
(1)
⊗a
(2)
h
(1)
(2)
⊗h
(2)
) (b
(1)
⊗g
(1)
(1)
⊗b
(2)
g
(1)
(2)
⊗g
(2)
),
(5.5.41)
= (−1)
β
0
+β
1
(a
(1)
⊗h
(1)
(1)
) (b
(1)
⊗g
(1)
(1)
) ⊗(a
(2)
h
(1)
(2)
⊗h
(2)
) (b
(2)
g
(1)
(2)
⊗
g
(2)
),
(5.5.30)
= (−1)
β
0
+β
1
+β
2
a
(1)
(h
(1)
(1)
(1)
b
(1)
) ⊗h
(1)
(1)
(2)
g
(1)
(1)
⊗a
(2)
h
(1)
(2)
(h
(2)(1)
154
b
(2)
g
(1)
(2)
) ⊗h
(2)(2)
g
(2)
coassociativity of ∆
H
= (−1)
β
0
+β
1
+β
2
a
(1)
(h
/
(1)
(1)
(1)
b
(1)
) ⊗h
/
(1)
(1)
(2)
g
(1)
(1)
⊗
a
(2)
h
/
(1)
(2)
(h
/
(2)
b
(2)
g
(1)
(2)
) ⊗h
/
(3)
g
(2)
(∗1)
(5.5.38)
= (−1)
β
0
+β
1
+β
2
+β
3
a
(1)
(h
/
(1)
(1)
(1)
b
(1)
) ⊗h
/
(1)
(1)
(2)
g
(1)
(1)
⊗
a
(2)
h
/
(1)
(2)
(h
/
(2)(1)
b
(2)
)(h
/
(2)(2)
g
(1)
(2)
) ⊗h
/
(3)
g
(2)
coassociativity of ∆
H
= (−1)
β
0
+β
1
+β
2
+β
3
a
(1)
(h
//
(1)
(1)
(1)
b
(1)
) ⊗h
//
(1)
(1)
(2)
g
(1)
(1)
⊗
a
(2)
h
//
(1)
(2)
(h
//
(2)
b
(2)
)(h
//
(3)
g
(1)
(2)
) ⊗h
//
(4)
g
(2)
(∗2)
(5.5.41)
= (−1)
β
0
+β
1
+β
2
+β
3
−β
4
(a
(1)
⊗1⊗a
(2)
⊗1)(h
//
(1)
(1)
(1)
b
(1)
⊗h
//
(1)
(1)
(2)
g
(1)
(1)
⊗
h
//
(1)
(2)
(h
//
(2)
b
(2)
) ⊗1) (1 ⊗1 ⊗h
//
(3)
g
(1)
(2)
⊗h
//
(4)
g
(2)
)
(5.5.41)
= (−1)
β
0
+β
1
+β
2
+β
3
−β
4
−β
5
(a
(1)
⊗1 ⊗a
(2)
⊗1) (h
//
(1)
(1)
(1)
b
(1)
⊗h
//
(1)
(1)
(2)
⊗
h
//
(1)
(2)
(h
//
(2)
b
(2)
) ⊗1) (1 ⊗g
(1)
(1)
⊗h
//
(3)
g
(1)
(2)
⊗h
//
(4)
g
(2)
)
(5.5.40)
= (−1)
β
0
+β
1
+β
2
+β
3
−β
4
−β
5
−β
6
(a
(1)
⊗1 ⊗a
(2)
⊗1)
_
(h
//
(1)
(1)
(1)
⊗h
//
(1)
(1)
(2)
⊗h
//
(1)
(2)
(h
//
(2)
b
(2)
) ⊗1) (b
(1)
⊗1 ⊗1 ⊗1)
_
(1⊗g
(1)
(1)
⊗
h
//
(3)
g
(1)
(2)
⊗h
//
(4)
g
(2)
)
(5.5.29)
= (−1)
β
0
+β
1
+β
2
+β
3
−β
4
−β
5
−β
6
+β
7
(a
(1)
⊗1 ⊗a
(2)
⊗1)
_
(h
//
(2)
(1)
(1)
⊗h
//
(2)
(1)
(2)
⊗(h
//
(1)
b
(2)
)h
//
(2)
(2)
⊗1) (b
(1)
⊗1 ⊗1 ⊗1)
_
(1⊗g
(1)
(1)
⊗
h
//
(3)
g
(1)
(2)
⊗h
//
(4)
g
(2)
)
(5.5.41)
= (−1)
β
0
+β
1
+β
2
+β
3
−β
4
−β
5
−β
6
+β
7
−β
8
(a
(1)
⊗1 ⊗a
(2)
(h
//
(1)
b
(2)
) ⊗1)
_
(h
//
(2)
(1)
(1)
⊗h
//
(2)
(1)
(2)
⊗h
//
(2)
(2)
⊗1) (b
(1)
⊗1 ⊗1 ⊗1)
_
(1⊗g
(1)
(1)
⊗h
//
(3)
g
(1)
(2)
⊗
155
h
//
(4)
g
(2)
)
(5.5.39)
= (−1)
β
0
+β
1
+β
2
+β
3
−β
4
−β
5
−β
6
+β
7
−β
8
+β
9
(a
(1)
⊗1 ⊗a
(2)
(h
//
(1)
b
(2)
) ⊗1)
_
(h
//
(2)(1)
(1)
⊗h
//
(2)(2)
(1)
⊗h
//
(2)(1)
(2)
h
//
(2)(2)
(2)
⊗1) (b
(1)
⊗1 ⊗1 ⊗1)
_
(1 ⊗g
(1)
(1)
⊗
h
//
(3)
g
(1)
(2)
⊗h
//
(4)
g
(2)
)
coassociativity of ∆
H
= (−1)
β
0
+β
1
+β
2
+β
3
−β
4
−β
5
−β
6
+β
7
−β
8
+β
9
(a
(1)
⊗ 1 ⊗ a
(2)
(h
///
(1)
b
(2)
) ⊗1)
_
(h
///
(2)
(1)
⊗h
///
(3)
(1)
⊗h
///
(2)
(2)
h
///
(3)
(2)
⊗1) (b
(1)
⊗1 ⊗1 ⊗1)
_
(1⊗g
(1)
(1)
⊗
h
///
(4)
g
(1)
(2)
⊗h
///
(5)
g
(2)
) (∗3)
after rename h
by h
= (−1)
β
0
+β
1
+β
2
+β
3
−β
4
−β
5
−β
6
+β
7
−β
8
+β
9
(a
(1)
⊗ 1 ⊗ a
(2)
(h
(1)
b
(2)
) ⊗1)
_
(h
(2)
(1)
⊗h
(3)
(1)
⊗h
(2)
(2)
h
(3)
(2)
⊗1) (b
(1)
⊗1 ⊗1 ⊗1)
_
(1⊗g
(1)
(1)
⊗
h
(4)
g
(1)
(2)
⊗h
(5)
g
(2)
) (∗4)
(5.5.40),(5.5.41)
= (−1)
β
0
+β
1
+β
2
+β
3
−β
4
−β
5
−β
6
+β
7
−β
8
+β
9
−β
10
(a
(1)
⊗1⊗a
(2)
(h
(1)
b
(2)
)⊗
1)
_
(h
(2)
(1)
⊗1 ⊗h
(2)
(2)
⊗1) (b
(1)
⊗1 ⊗1 ⊗1)
_
(1⊗h
(3)
(1)
g
(1)
(1)
⊗h
(3)
(2)
(h
(4)
g
(1)
(2)
) ⊗h
(5)
g
(2)
)
(5.5.40),(5.5.41)
= (−1)
β
0
+β
1
+β
2
+β
3
−β
4
−β
5
−β
6
+β
7
−β
8
+β
9
−β
10
+β
11
(a
(1)
⊗ 1 ⊗ a
(2)
⊗ 1)
_
(h
(2)
(1)
⊗1 ⊗(h
(1)
b
(2)
)h
(2)
(2)
⊗1) (b
(1)
⊗1 ⊗1 ⊗1)
_
(1 ⊗h
(3)
(1)
g
(1)
(1)
⊗h
(3)
(2)
(h
(4)
g
(1)
(2)
) ⊗h
(5)
g
(2)
)
(5.5.29)
= (−1)
β
0
+β
1
+β
2
+β
3
−β
4
−β
5
−β
6
+β
7
−β
8
+β
9
−β
10
+β
11
−β
12
(a
(1)
⊗ 1 ⊗ a
(2)
⊗ 1)
_
(h
(1)
(1)
⊗1 ⊗h
(1)
(2)
(h
(2)
b
(2)
) ⊗1) (b
(1)
⊗1 ⊗1 ⊗1)
_
(1 ⊗h
(3)
(1)
g
(1)
(1)
⊗h
(3)
(2)
(h
(4)
g
(1)
(2)
) ⊗h
(5)
g
(2)
)
(5.5.40)
= (−1)
β
0
+β
1
+β
2
+β
3
−β
4
−β
5
−β
6
+β
7
−β
8
+β
9
−β
10
+β
11
−β
12
+β
13
(a
(1)
⊗1 ⊗a
(2)
⊗1) (h
(1)
(1)
b
(1)
⊗1 ⊗h
(1)
(2)
(h
(2)
b
(2)
) ⊗1) (1 ⊗h
(3)
(1)
g
(1)
(1)
⊗
156
h
(3)
(2)
(h
(4)
g
(1)
(2)
) ⊗h
(5)
g
(2)
)
(5.5.41)
= (−1)
β
0
+β
1
+β
2
+β
3
−β
4
−β
5
−β
6
+β
7
−β
8
+β
9
−β
10
+β
11
−β
12
+β
13
+β
14
(a
(1)
h
(1)
(1)
b
(1)
⊗1⊗a
(2)
h
(1)
(2)
(h
(2)
b
(2)
) ⊗1) (1⊗h
(3)
(1)
g
(1)
(1)
⊗h
(3)
(2)
(h
(4)
g
(1)
(2)
) ⊗h
(5)
g
(2)
)
(5.5.41)
= (−1)
β
0
+β
1
+β
2
+β
3
−β
4
−β
5
−β
6
+β
7
−β
8
+β
9
−β
10
+β
11
−β
12
+β
13
+β
14
+β
15
(a
(1)
h
(1)
(1)
b
(1)
⊗ h
(3)
(1)
g
(1)
(1)
⊗ a
(2)
h
(1)
(2)
(h
(2)
b
(2)
)h
(3)
(2)
(h
(4)
g
(1)
(2)
) ⊗
h
(5)
g
(2)
)
(5.5.33)
where at the end
h
(1)
⊗h
(2)
⊗h
(3)
⊗h
(4)
⊗h
(5)
= h
(1)
⊗h
(2)(1)(1)(1)
⊗h
(2)(1)(1)(2)
⊗h
(2)(1)(2)
⊗h
(2)(2)
,
and
β
0
= [h
(1)
(1)
[[a
(2)
[ +[g
(1)
(1)
[[b
(2)
[
after (∗1)
= [h
/
(1)
(1)
[[a
(2)
[ +[g
(1)
(1)
[[b
(2)
[
after (∗2)
= [h
//
(1)
(1)
[[a
(2)
[ +[g
(1)
(1)
[[b
(2)
[
after (∗3)
= [h
///
(1)
(1)
[[a
(2)
[ +[g
(1)
(1)
[[b
(2)
[
after (∗4)
= [h
(1)
(1)
[[a
(2)
[ +[g
(1)
(1)
[[b
(2)
[
β
1
= [(b
(1)
⊗g
(1)
(1)
)[[(a
(2)
h
(1)
(2)
⊗h
(2)
)[
(5.5.42),(5.5.43)
= ([(b
(1)
[ +[g
(1)
(1)
[)([(a
(2)
[ +[h
(1)
(2)
[ +[h
(2)
[)
after (∗1)
= ([(b
(1)
[ +[g
(1)
(1)
[)([(a
(2)
[ +[h
/
(1)
(2)
[ + ([h
/
(2)
[ +[h
/
(3)
[))
after (∗2)
= ([(b
(1)
[ +[g
(1)
(1)
[)([(a
(2)
[ +[h
//
(1)
(2)
[ + ([h
//
(2)
[ +[h
//
(3)
[ +[h
//
(4)
[))
after (∗3)
= ([(b
(1)
[ +[g
(1)
(1)
[)([(a
(2)
[ +[h
///
(1)
(2)
[ + ([h
///
(2)
[ +[h
///
(3)
[ +[h
///
(4)
[ +[h
///
(5)
[))
after (∗4)
= ([(b
(1)
[ +[g
(1)
(1)
[)([(a
(2)
[ +[h
(1)
(2)
[ + ([h
(2)
[ +[h
(3)
[ +[h
(4)
[ +[h
(5)
[))
= [b
(1)
[[a
(2)
[ +[b
(1)
[[h
(1)
(2)
[ +[b
(1)
[([h
(2)
[ +[h
(3)
[ +[h
(4)
[ +[h
(5)
[) +[g
(1)
(1)
[[a
(2)
[
+[g
(1)
(1)
[[h
(1)
(2)
[ +[g
(1)
(1)
[[h
(2)
[ +[g
(1)
(1)
[[h
(3)
[ +[g
(1)
(1)
[[h
(4)
[ +[g
(1)
(1)
[[h
(5)
[
157
β
2
= [b
(1)
[[h
(1)
(1)
(2)
[ +[b
(2)
g
(1)
(2)
[[h
(2)(2)
[
(5.5.42),(5.5.43)
= [b
(1)
[[h
(1)
(1)
(2)
[ + ([b
(2)
[ +[g
(1)
(2)
[)([h
(2)(2)
[)
after (∗1)
= [b
(1)
[[h
/
(1)
(1)
(2)
[ + ([b
(2)
[ +[g
(1)
(2)
[)([h
/
(3)
[)
after (∗2)
= [b
(1)
[[h
//
(1)
(1)
(2)
[ + ([b
(2)
[ +[g
(1)
(2)
[)([h
//
(4)
[)
after (∗3)
= [b
(1)
[[h
///
(1)
(1)
(2)
[ + ([b
(2)
[ +[g
(1)
(2)
[)([h
///
(5)
[)
after (∗4)
= [b
(1)
[[h
(1)
(1)
(2)
[ + ([b
(2)
[ +[g
(1)
(2)
[)([h
(5)
[)
= [b
(1)
[[h
(1)
(1)
(2)
[ +[b
(2)
[[h
(5)
[ +[g
(1)
(2)
[[h
(5)
[
β
3
= [b
(2)
[[h
/
(2)(2)
[
after (∗2)
= [b
(2)
[[h
//
(3)
[
after (∗3)
= [b
(2)
[[h
///
(4)
[
after (∗4)
= [b
(2)
[[h
(4)
[
β
4
= [a
(2)
[[(h
//
(1)
(1)
(1)
b
(1)
⊗h
//
(1)
(1)
(2)
g
(1)
(1)
)[
(5.5.42),(5.5.43)
= [a
(2)
[([(h
//
(1)
(1)
(1)
[ +[b
(1)
[ +[h
//
(1)
(1)
(2)
[ +[g
(1)
(1)
[)
after (∗3)
= [a
(2)
[([(h
///
(1)
(1)
(1)
[ +[b
(1)
[ +[h
///
(1)
(1)
(2)
[ +[g
(1)
(1)
[)
after (∗4)
= [a
(2)
[([(h
(1)
(1)
(1)
[ +[b
(1)
[ +[h
(1)
(1)
(2)
[ +[g
(1)
(1)
[)
= [a
(2)
[[h
(1)
(1)
(1)
[ +[a
(2)
[[b
(1)
[ +[a
(2)
[[h
(1)
(1)
(2)
[ +[a
(2)
[[g
(1)
(1)
[
β
5
= [g
(1)
(1)
[[(h
//
(1)
(2)
(h
//
(2)
b
(2)
) ⊗1)[
(5.5.42),(5.5.43)
= [g
(1)
(1)
[([(h
//
(1)
(2)
[ +[h
//
(2)
[ +[b
(2)
)[)
after (∗3)
= [g
(1)
(1)
[([(h
///
(1)
(2)
[ + ([h
///
(2)
[ +[h
///
(3)
[) +[b
(2)
)[)
after (∗4)
= [g
(1)
(1)
[([(h
(1)
(2)
[ +[h
(2)
[ +[h
(3)
[ +[b
(2)
)[)
= [g
(1)
(1)
[[(h
(1)
(2)
[ +[g
(1)
(1)
[[h
(2)
[ +[g
(1)
(1)
[[h
(3)
[ +[g
(1)
(1)
[[b
(2)
)[
158
β
6
= [b
(1)
[[(h
//
(1)
(1)
(2)
⊗h
//
(1)
(2)
(h
//
(2)
b
(2)
) ⊗1)[
(5.5.42),(5.5.43)
= [b
(1)
[([h
//
(1)
(1)
(2)
[ +[h
//
(1)
(2)
[ +[h
//
(2)
[ +[b
(2)
[)
after (∗3)
= [b
(1)
[([h
///
(1)
(1)
(2)
[ +[h
///
(1)
(2)
[ + ([h
///
(2)
[ +[h
///
(3)
[) +[b
(2)
[)
after (∗4)
= [b
(1)
[([h
(1)
(1)
(2)
[ +[h
(1)
(2)
[ +[h
(2)
[ +[h
(3)
[ +[b
(2)
[)
= [b
(1)
[[h
(1)
(1)
(2)
[ +[b
(1)
[[h
(1)
(2)
[ +[b
(1)
[[h
(2)
[ +[b
(1)
[[h
(3)
[ +[b
(1)
[[b
(2)
[
β
7
= [h
//
(2)
(1)
[[(h
//
(1)
[ +[b
(2)
[[h
//
(2)
(2)
[
. ¸¸ .
β
8
= [(h
//
(2)
(1)
(1)
⊗h
//
(2)
(1)
(2)
)[[(h
//
(1)
b
(2)
)[
(5.5.42),(5.5.43)
= ([h
//
(2)
(1)
(1)
[ +[h
//
(2)
(1)
(2)
[)([h
//
(1)
[ +[b
(2)
[)
= ([h
//
(2)
(1)
[)([h
//
(1)
[ +[[b
(2)
[) =
¸ .. ¸
[h
//
(2)
(1)
[[h
//
(1)
[ +[h
//
(2)
(1)
[[b
(2)
[
And so far we have:
β
0
+ β
1
+ β
2
+ β
3
−β
4
−β
5
−β
6
+ (β
7
−β
8
)
= β
0
+ β
1
+ β
2
+ β
3
−β
4
−β
5
−β
6
+
_
[h
//
(2)
(1)
[[(h
//
(1)
[ +[b
(2)
[[h
//
(2)
(2)
[ −[h
//
(2)
(1)
[[h
//
(1)
[ −[h
//
(2)
(1)
[[b
(2)
[
_
(5.5.43),(5.5.44)
= β
0
+ β
1
+ β
2
+ β
3
−β
4
−β
5
−β
6
+[h
//
(2)
[[b
(2)
[
after (∗3)
= β
0
+ β
1
+ β
2
+ β
3
−β
4
−β
5
−β
6
+ ([h
///
(2)
[ +[h
///
(3)
[)[b
(2)
[
after (∗4)
= β
0
+ β
1
+ β
2
+ β
3
−β
4
−β
5
−β
6
+ ([h
(2)
[ +[h
(3)
[)[b
(2)
[
β
9
= [h
//
(2)(2)
(1)
[[h
//
(2)(1)
(2)
[
after (∗3)
= [h
///
(3)
(1)
[[h
///
(2)
(2)
[
after (∗4)
= [h
(3)
(1)
[[h
(2)
(2)
[
β
10
= [h
(3)
(2)
[[b
(1)
[ +[h
(3)
(2)
[[g
(1)
(1)
[ +[h
(3)
(1)
[[h
(2)
(2)
[ +[h
(3)
(1)
[[b
(1)
[
β
11
= y[(h
(1)
b
(2)
)[[h
(2)
(1)
[
(5.5.42),(5.5.43)
= ([(h
(1)
[ +[b
(2)
[)[h
(2)
(1)
[
= [(h
(1)
[[h
(2)
(1)
[ +[b
(2)
[[h
(2)
(1)
[
159
β
12
= [h
(1)
[[h
(2)
(1)
[ +[b
(2)
[[h
(2)
(2)
[
β
13
= [(h
(1)
(2)
(h
(2)
b
(2)
)[[b
(1)
[
(5.5.42),(5.5.43)
= ([h
(1)
(2)
[ +[(h
(2)
[ +[b
(2)
[)[b
(1)
[
= [h
(1)
(2)
[[b
(1)
[ +[(h
(2)
[[b
(1)
[ +[b
(2)
[[b
(1)
[
β
14
= [a
(2)
[[(h
(1)
(1)
b
(1)
)[
(5.5.42),(5.5.43)
= [a
(2)
[[h
(1)
(1)
[ +[a
(2)
[[b
(1)
[
β
15
= [(a
(2)
h
(1)
(2)
(h
(2)
b
(2)
))[[(h
(3)
(1)
g
(1)
(1)
)[
(5.5.42),(5.5.43)
= ([a
(2)
[ +[h
(1)
(2)
[ +[h
(2)
[ +[b
(2)
[)([h
(3)
(1)
[ +[g
(1)
(1)
[)
= [a
(2)
[[h
(3)
(1)
[ +[h
(1)
(2)
[[h
(3)
(1)
[ +[h
(2)
[[h
(3)
(1)
[ +[b
(2)
[[h
(3)
(1)
[ +[a
(2)
[[g
(1)
(1)
[
+[h
(1)
(2)
[[g
(1)
(1)
[ +[h
(2)
[[g
(1)
(1)
[ +[b
(2)
[[g
(1)
(1)
[
After all we have:
β
0
+β
1
+β
2
+β
3
−β
4
−β
5
−β
6
+β
7
−β
8
+β
9
−β
10
+β
11
−β
12
+β
13
+β
14
+β
15
= β
0
+ β
1
+ β
2
+ β
3
−β
4
−β
5
−β
6
+ ([h
(2)
[ +[h
(3)
[)[b
(2)
[
+β
9
−β
10
+ β
11
−β
12
+ β
13
+ β
14
+ β
15
= [h
(1)
(1)
[[a
(2)
[ +[g
(1)
(1)
[[b
(2)
[
+[b
(1)
[[a
(2)
[ +[b
(1)
[[h
(1)
(2)
[ +[b
(1)
[([h
(2)
[ +[h
(3)
[ +[h
(4)
[ +[h
(5)
[) +[g
(1)
(1)
[[a
(2)
[
+[g
(1)
(1)
[[h
(1)
(2)
[ +[g
(1)
(1)
[[h
(2)
[ +[g
(1)
(1)
[[h
(3)
[ +[g
(1)
(1)
[[h
(4)
[ +[g
(1)
(1)
[[h
(5)
[
[b
(1)
[[h
(1)
(1)
(2)
[ +[b
(2)
[[h
(5)
[ +[g
(1)
(2)
[[h
(5)
[
[a
(2)
[[h
(1)
(1)
(1)
[ +[a
(2)
[[b
(1)
[ +[a
(2)
[[h
(1)
(1)
(2)
[ +[a
(2)
[[g
(1)
(1)
[
[g
(1)
(1)
[[(h
(1)
(2)
[ +[g
(1)
(1)
[[h
(2)
[ +[g
(1)
(1)
[[h
(3)
[ +[g
(1)
(1)
[[b
(2)
)[
[b
(1)
[[h
(1)
(1)
(2)
[ +[b
(1)
[[h
(1)
(2)
[ +[b
(1)
[[h
(2)
[ +[b
(1)
[[h
(3)
[ +[b
(1)
[[b
(2)
[
[h
(3)
(1)
[[h
(2)
(2)
[ +[h
(3)
[)[b
(2)
[
160
[h
(3)
(2)
[[b
(1)
[ +[h
(3)
(2)
[[g
(1)
(1)
[ +[h
(3)
(1)
[[h
(2)
(2)
[ +[h
(3)
(1)
[[b
(1)
[
[(h
(1)
[[h
(2)
(1)
[ +[b
(2)
[[h
(2)
(1)
[
[h
(1)
[[h
(2)
(1)
[ +[b
(2)
[[h
(2)
(2)
[
[h
(1)
(2)
[[b
(1)
[ +[(h
(2)
[[b
(1)
[ +[b
(2)
[[b
(1)
[
[a
(2)
[[h
(1)
(1)
[ +[a
(2)
[[b
(1)
[
[a
(2)
[[h
(3)
(1)
[ +[h
(1)
(2)
[[h
(3)
(1)
[ +[h
(2)
[[h
(3)
(1)
[ +[b
(2)
[[h
(3)
(1)
[ +[a
(2)
[[g
(1)
(1)
[
+[h
(1)
(2)
[[g
(1)
(1)
[ +[h
(2)
[[g
(1)
(1)
[ +[b
(2)
[[g
(1)
(1)
[
(5.5.42),(5.5.43)(5.5.44)
=
([h
(3)
[ +[[h
(4)
[ +[[h
(5)
[)([b
(1)
[ +[b
(2)
[)
+[g
(1)
(1)
[[a
(2)
[ +[g
(1)
(1)
[[h
(1)
(2)
[ +[g
(1)
(1)
[[h
(2)
[ +[g
(1)
(1)
[[b
(2)
[
+[h
(3)
(1)
[[a
(2)
[ +[h
(3)
(1)
[[h
(1)
(2)
[ +[h
(3)
(1)
[[h
(2)
[ +[h
(3)
(1)
[[b
(2)
[
+[a
(2)
[[h
(1)
(1)
[ +[a
(2)
[[b
(1)
[
+[h
(5)
[[g
(1)
[
+[b
(1)
[[(h
(2)
[ +[b
(1)
[[h
(1)
(2)
[
+[g
(1)
(1)
[[h
(4)
[ +[g
(1)
(1)
[[h
(3)
(2)
[
(5.5.34)
From the other hand we have:
∆((a ⊗h) (b ⊗g))
(5.5.30)
= (−1)
α
0
∆(a(h
(1)
b) ⊗h
(2)
g)
(5.5.31)
= (−1)
α
0
+α
1
(a(h
(1)
b))
(1)
⊗ (h
(2)
g)
(1)
(1)
⊗ (a(h
(1)
b))
(2)
(h
(2)
g)
(1)
(2)
⊗
(h
(2)
g)
(2)
(5.5.37)
= (−1)
α
0
+α
1
+α
2
+α
3
a
(1)
(h
(1)
b)
(1)
⊗(h
(2)(1)
g
(1)
)
(1)
⊗
a
(2)
(h
(1)
b)
(2)
(h
(2)(1)
g
(1)
)
(2)
⊗h
(2)(2)
g
(2)
161
(5.5.27),(5.5.28)
= (−1)
α
0
+α
1
+α
2
+α
3
+α
4
+α
5
a
(1)
(h
(1)(1)
(1)
b
(1)
)⊗(h
(2)(1)(1)
(1)
g
(1)
(1)
)⊗
a
(2)
h
(1)(1)
(2)
(h
(1)(2)
b
(2)
)(h
(2)(1)(1)
(2)
(h
(2)(1)(2)
g
(1)
(2)
)) ⊗h
(2)(2)
g
(2)
,
coassociativity of ∆
H
= (−1)
α
0
+α
1
+α
2
+α
3
+α
4
+α
5
a
(1)
(h
/
(1)
(1)
b
(1)
)⊗(h
/
(3)
(1)
g
(1)
(1)
)⊗
a
(2)
h
/
(1)
(2)
(h
/
(2)
b
(2)
)(h
/
(3)
(2)
(h
/
(4)
g
(1)
(2)
)) ⊗h
/
(5)
g
(2)
(∗1),
after rename h
by h
= (−1)
α
0
+α
1
+α
2
+α
3
+α
4
+α
5
a
(1)
(h
(1)
(1)
b
(1)
) ⊗(h
(3)
(1)
g
(1)
(1)
) ⊗
a
(2)
h
(1)
(2)
(h
(2)
b
(2)
)(h
(3)
(2)
(h
(4)
g
(1)
(2)
)) ⊗h
(5)
g
(2)
(∗2),
(5.5.35)
where at the end
h
(1)
⊗h
(2)
⊗h
(3)
⊗h
(4)
⊗h
(5)
= h
(1)(1)
⊗h
(1)(2)
⊗h
(2)(1)(1)
⊗h
(2)(1)(2)
⊗h
(2)(2)
,
and
α
0
= [h
(2)
[[b[ = ([h
(2)(1)(1)
[ +[[h
(2)(1)(2)
[ +[[h
(2)(2)
[)([b
(1)
[ +[b
(2)
[)
after (∗1) and (∗2)
= ([h
(3)
[ +[[h
(4)
[ +[[h
(5)
[)([b
(1)
[ +[b
(2)
[)
α
1
= [(h
(2)
g)
(1)
(1)
[[(a(h
(1)
b)
(2)
[
= [(h
(2)(1)(1)
(1)
g
(1)
(1)
)[[a
(2)
h
(1)(1)
(2)
(h
(1)(2)
b
(2)
)[
= ([h
(2)(1)(1)
(1)
[ +[g
(1)
(1)
[)([a
(2)
[ +[h
(1)(1)
(2)
[ +[h
(1)(2)
[ +[b
(2)
[)
after (∗1) and (∗2)
= ([h
(3)
(1)
[ +[g
(1)
(1)
[)([a
(2)
[ +[h
(1)
(2)
[ +[h
(2)
[ +[b
(2)
[)
= [g
(1)
(1)
[[a
(2)
[ +[g
(1)
(1)
[[h
(1)
(2)
[ +[g
(1)
(1)
[[h
(2)
[ +[g
(1)
(1)
[[b
(2)
[
+[h
(3)
(1)
[[a
(2)
[ +[h
(3)
(1)
[[h
(1)
(2)
[ +[h
(3)
(1)
[[h
(2)
[ +[h
(3)
(1)
[[b
(2)
[
α
2
= [a
(2)
[[(h
(1)
b)
(1)
[ = [a
(2)
[[(h
(1)(1)
(1)
b
(1)
)[
= [a
(2)
[([h
(1)(1)
(1)
[ +[b
(1)
[)
162
after (∗1) and (∗2)
= [a
(2)
[([h
(1)
(1)
[ +[b
(1)
[) = [a
(2)
[[h
(1)
(1)
[ +[a
(2)
[[b
(1)
[
α
3
= [h
(2)(2)
[[g
(1)
[
after (∗1) and (∗2)
= [h
(5)
[[g
(1)
[
α
4
= [b
(1)
[([(h
(1)(2)
[ +[h
(1)(1)
(2)
[)
after (∗1) and (∗2)
= [b
(1)
[([(h
(2)
[ +[h
(1)
(2)
[) = [b
(1)
[[(h
(2)
[ +[b
(1)
[[h
(1)
(2)
[
α
5
= [g
(1)
(1)
[([h
(2)(1)(2)
[ +[h
(2)(1)(1)
(2)
[)
after (∗1) and (∗2)
= [g
(1)
(1)
[([h
(4)
[ +[h
(3)
(2)
[) = [g
(1)
(1)
[[h
(4)
[ +[g
(1)
(1)
[[h
(3)
(2)
[
Therefore by (5.5.34) we see that:
α
0
+ α
1
+ α
2
+ α
3
+ α
4
+ α
5
=
β
0
+β
1
+β
2
+β
3
−β
4
−β
5
−β
6
+β
7
−β
8
+β
9
−β
10
+β
11
−β
12
+β
13
+β
14
+β
15
. (5.5.36)
By (5.5.33) and (5.5.35) and (5.5.36) we have
∆((a ⊗h) (b ⊗g)) = ∆(a ⊗h) ∆(b ⊗g).
The rest of the proof can be done by a similar method.
In the above proof we have also used the following formulas and properties
related to super (Hopf) algebras. For any super Hopf algebra H, the comultiplication
∆ being an algebra map means for all g and h in H,
∆(gh) = (gh)
(1)
⊗(gh)
(2)
= (−1)
h
(1)
g
(2)

g
(1)
h
(1)
⊗g
(2)
h
(2)
= ∆(g)∆(h). (5.5.37)
The algebra A being a left Hmodule algebra means that for all h in H and a and b
163
in A:
h (ab) = (−1)
[a[[h
(2)
[
(h
(1)
a)(h
(2)
b). (5.5.38)
The coalgebra H being a right Acomodule coalgebra, H →H ⊗A; h →h
(1)
⊗h
(2)
,
means that for all h in H :
h
(1)
(1)
⊗h
(1)
(2)
⊗h
(2)
= (−1)
[h
(2)
(1)
[[h
(1)
(2)
[
h
(1)
(1)
⊗h
(2)
(1)
⊗h
(1)
(2)
h
(2)
(2)
. (5.5.39)
If A and B are two left Hmodule (algebras), then A ⊗ B is a left H ⊗ Hmodule
(algebra) with action deﬁned, for all h in H, a in A, and b in B, by:
(h ⊗g) (a ⊗b) = (−1)
[a[[g[
(h a) ⊗(g b). (5.5.40)
For any two super (Hopf)algebra A and B, A ⊗ B is an algebra with the following
multiplication:
(a ⊗b) (a
/
⊗b
/
) := (−1)
[b[[a
[
aa
/
⊗bb
/
. (5.5.41)
For any three (Hopf) algebras A, B, and C, if f : A ⊗B →C is a degree preserving
map then, for all homogeneous elements a and b we have:
[f(a ⊗b)[ = [a[ +[b[. (5.5.42)
Also if g : C → A ⊗ B, g(c) =
i
a
i
⊗ b
i
, is a degree preserving map then for all
homogeneous elements c:
[a
i
[ +[b
i
[ = [c[, ∀i, . (5.5.43)
It is obvious but very important for us that (−1)
−α
= (−1)
α
therefore
(−1)
β−α
= (−1)
β+α
, (5.5.44)
for any two natural numbers α and β.
164
Now one can use Lemma (5.5.1) for the super Hopf algebras F(G
s
2
) and U(g
s
1
),
and show that the actions and coactions deﬁned in Lemmas (5.4.1) and (5.4.2) satisfy
the conditions (5.5.27)(5.5.29) in this lemma. This means that we have the following
theorem which is the main result of this chapter.
Theorem 5.5.1. The actions and coactions deﬁned in Lemmas (5.4.1) and (5.4.2)
satisfy the conditions (5.5.27)(5.5.29) in Lemma (5.5.1). Therefore we have a bi
crossproduct super Hopf algebra F(G
s
2
)U(g
s
1
) with the following structures;
(a ⊗h)(b ⊗g) = (−1)
[h
(2)
[[b[
a(h
(1)
b) ⊗h
(2)
g,
∆(a ⊗h) = (−1)
[h
(1)
(1)
[[a
(2)
[
a
(1)
⊗h
(1)
(1)
⊗a
(2)
h
(1)
(2)
⊗h
(2)
,
ε(a ⊗h) = ε(a)ε(h),
S(a ⊗h) = (−1)
[h
(1)
[[a[
(1 ⊗S(h
(1)
))(S(ah
(2)
) ⊗1),
for any a, b in F(G
s
2
) and g, h in U(g
s
1
).
Deﬁnition 5.5.1. The bicrossproduct super Hopf algebra F(G
s
2
)U(g
s
1
) is called the
super version of ConnesMoscovici Hopf algebra H
1
and is denoted by H
s
1
.
Bibliography
[1] R. Akbarpour, and M. Khalkhali, Hopf algebra equivariant cyclic homology and
cyclic homology of crossed product algebras. J. Reine Angew. Math. 559 (2003),
137–152.
[2] S. E. Akrami, and S. Majid, Braided cyclic cocycles and nonassociative geometry.
J. Math. Phys. 45 (2004), no. 10, 3883–3911.
[3] D. Bulacu, and E. Nauwelaerts, Radford’s biproduct for quasiHopf algebras and
bosonization. J. Pure Appl. Algebra 174 (2002), no. 1, 1–42.
[4] L. Caston, and R. Fioresi, MATHEMATICAL FOUNDATIONS OF SUPER
SYMMETRY. http://arxiv.org/pdf/0710.5742v1, (2007).
[5] A. Connes, Spectral sequence and homology of currents for operator algebras.
41/81, Funkiionalanalysis and C
∗
algebras, 279/310, 1981.
[6] A. Connes, Cohomologie cyclique et foncteurs Ext
n
. (French) (Cyclic cohomol
ogy and functors Ext
n
) C. R. Acad. Sci. Paris Sr. I Math. 296 (1983), no. 23,
953–958.
[7] A. Connes, Noncommutative diﬀerential geometry. Inst. Hautes
´
Etudes Sci. Publ.
Math. No. 62 (1985), 257–360.
[8] A. Connes, Noncommutative Geometry, Academic Press, San Diego, CA, 1994,
661 p., available online at http://alainconnes.org/.
[9] A. Connes and H. Moscovici, Hopf algebras, Cyclic Cohomology and the trans
verse index theorem, Comm. Math. Phys. 198 (1998), no. 1, 199–246.
165
166
[10] A. Connes and H. Moscovici, Cyclic cohomology and Hopf algebras. Lett. Math.
Phys. 48 (1999), no. 1, 97–108.
[11] A. Connes, and H. Moscovici, Cyclic cohomology and Hopf algebra symmetry.
Conference Mosh´e Flato 1999 (Dijon). Lett. Math. Phys. 52 (2000), no. 1, 1–28.
[12] M. Crainic, Cyclic cohomology of Hopf algebras. J. Pure Appl. Algebra 166
(2002), no. 12, 29–66.
[13] P. Deligne, and J. W. Morgan, Notes on supersymmetry (following Joseph Bern
stein) Quantum ﬁelds and strings: a course for mathematicians 1: 4197, Amer
ican Mathematical Society, Providence, RI; 1999.
[14] H. Figueroa, and J. GraciaBond´ıa, Combinatorial Hopf algebras in quantum
ﬁeld theory, I. Rev. Math. Phys. 17 (2005), no. 8, 881–976.
[15] H. Figueroa, J. GraciaBonda, and J. C. V´arilly, Fa`a di Bruno Hopf algebras.
http://arxiv.org/pdf/math/0508337, (2005).
[16] E. Getzler, and D. S. Jones, The cyclic homology of crossed product algebras. J.
Reine Angew. Math. 445 (1993), 161–174.
[17] T. Hadﬁeld and S. Majid, Bicrossproduct Approach to the ConnesMoscovici
Hopf Algebra, J. Algebra 312 (2007) 228256.
[18] P. M. Hajac, M. Khalkhali, B. Rangipour, and Y. Sommerh¨auser, Stable anti
YetterDrinfeld modules. C. R. Math. Acad. Sci. Paris 338 (2004), no. 8, 587–590.
[19] P. M. Hajac, M. Khalkhali, B. Rangipour, and Y. Sommerh¨auser, Hopfcyclic
homology and cohomology with coeﬃcients. C. R. Math. Acad. Sci. Paris 338
(2004), no. 9, 667–672.
[20] A. Joyal and R. Street, Braided tensor categories. Adv. Math. 102 (1993), no. 1,
20–78.
167
[21] A. Joyal, and R. Street, The geometry of tensor calculus. I. Adv. Math. 88
(1991), no. 1, 55–112.
[22] Ch. Kassel, Quantum groups. Graduate Texts in Mathematics, 155. Springer
Verlag, New York, 1995.
[23] Ch. Kassel, and V. Turaev, Braid groups. With the graphical assistance of Olivier
Dodane . Graduate Texts in Mathematics, 247. Springer, New York, 2008.
[24] A. Klimyk, and K. Schm¨ udgen, Quantum groups and their representations, Texts
and Monographs in Physics. SpringerVerlag, Berlin, (1997).
[25] M. Khalkhali, and B. Rangipour, A new cyclic module for Hopf algebras. K
Theory 27 (2) (2002), 111131.
[26] M. Khalkhali, and B. Rangipour, Invariant cyclic homology. KTheory 28 (2)
(2003), 183205.
[27] M. Khalkhali, and B. Rangipour,Introduction to Hopfcyclic cohomology. Non
commutative geometry and number theory, 155–178, Aspects Math., E37,
Vieweg, Wiesbaden, 2006. (Reviewer: Jinhyun Park)
[28] J. L. Loday, Cyclic homology. SpringerVerlag, Berlin, 1992.
[29] S. Mac Lane, Categories for the working mathematician. Second edition. Grad
uate Texts in Mathematics, 5. SpringerVerlag, New York, 1998.
[30] S. Majid, Algebras and Hopf algebras in braided categories. Advances in Hopf
algebras, 55–105, Lecture Notes in Pure and Appl. Math., 158, Dekker, New
York, 1994.
[31] S. Majid, Foundations of quantum group theory. Cambridge University Press,
Cambridge, 1995.
168
[32] G. Mason, and S.H. Ng, Group cohomology and gauge equivalence of some
twisted quantum doubles. Trans. Amer. Math. Soc. 353 (2001), no. 9, 3465–3509
(electronic).
[33] D. Radford, and J. Towber, YetterDrinfeld categories associated to an arbitrary
bialgebra. J. Pure Appl. Algebra 87 (1993), 259279.
[34] A. Rogers, Supermanifolds. Theory and applications. World Scientiﬁc Publishing
Co. Pte. Ltd., Hackensack, NJ, 2007.
[35] N. Saavedra Rivano, Cat´egories Tannakiennes,, Lecture Notes in Mathematics,
no.265, SpringerVerlag, Berlin, 1972.
[36] Y. Sommerh¨auser, YetterDrinfel’d Hopf algebras over groups of prime order.
Lecture Notes in Mathematics, 1789. SpringerVerlag, Berlin, 2002.
[37] M. E. Sweedler, Hopf algebras. Mathematics Lecture Note Series W. A. Ben
jamin, Inc., New York 1969 vii+336 pp.
[38] V. G. Turaev, Quantum invariants of knots and 3manifolds, de Gruyter studies
in Mathematics, no. 18, W de Gruyter, Berlin, 1994.
[39] V. S. Varadarajan, Supersymmetry for mathematicians: an introduction.
Courant Lecture Notes in Mathematics, 11. New York University, Courant In
stitute of Mathematical Sciences, New York; American Mathematical Society,
Providence, RI, 2004.
[40] D. Yetter, Quantum groups and representations of monoidal categories. Math.
Proc. Cambridge Philos. Soc. 108 (1990), no. 2, 261–290.
169
Vita
Name: Arash Pourkia
Postsecondary Kerman University
Education and Kerman, Iran
Degrees: 19861993, B. Sc.
Kerman University
Kerman, Iran
19931996, M.Sc.
The University of Western Ontario
London, Ontario, Canada
20052009, Ph.D.
Honors and Iranian Higher Education Scholarship
Awards Ministry of Higher Education
of Iran, 20052009
Related Work Teaching Assistant
Experience The University of Western Ontario, 20052009
Lecturer
Persian Gulf University
Boushehr, Iran, 19982005
Publications:
[1] Masoud Khalkhali and Arash Pourkia, “Hopf Cyclic Cohomology in Braided
Monoidal Categories”, ( preprint http://xxx.lanl.gov/pdf/0807.3890v1)
[2] Masoud Khalkhali and Arash Pourkia, “A super version of ConnesMoscovici
Hopf algebra”, ( preprint )