Degenerate Series Representations of the q-Deformed Algebra so'q(r,s)
The q-deformed algebra so'q(r,s) is a real form of the q-deformed algebra Uq'(so(n,C)), n = r + s, which differs from the quantum algebra Uq(so(n,C)) of Drinfeld and Jimbo. We study representations of the most degenerate series of the algebra so'q(r,s). The formulas of action of opera...
Gespeichert in:
| Datum: | 2007 |
|---|---|
| 1. Verfasser: | |
| Format: | Artikel |
| Sprache: | English |
| Veröffentlicht: |
Інститут математики НАН України
2007
|
| Schriftenreihe: | Symmetry, Integrability and Geometry: Methods and Applications |
| Online Zugang: | http://dspace.nbuv.gov.ua/handle/123456789/147377 |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| Назва журналу: | Digital Library of Periodicals of National Academy of Sciences of Ukraine |
| Zitieren: | Degenerate Series Representations of the q-Deformed Algebra so'q(r,s) / V.A. Groza // Symmetry, Integrability and Geometry: Methods and Applications. — 2007. — Т. 3. — Бібліогр.: 21 назв. — англ. |
Institution
Digital Library of Periodicals of National Academy of Sciences of Ukraine| id |
irk-123456789-147377 |
|---|---|
| record_format |
dspace |
| spelling |
irk-123456789-1473772019-02-15T01:24:59Z Degenerate Series Representations of the q-Deformed Algebra so'q(r,s) Groza, V.A. The q-deformed algebra so'q(r,s) is a real form of the q-deformed algebra Uq'(so(n,C)), n = r + s, which differs from the quantum algebra Uq(so(n,C)) of Drinfeld and Jimbo. We study representations of the most degenerate series of the algebra so'q(r,s). The formulas of action of operators of these representations upon the basis corresponding to restriction of representations onto the subalgebra so'q(r) × so'q(s) are given. Most of these representations are irreducible. Reducible representations appear under some conditions for the parameters determining the representations. All irreducible constituents which appear in reducible representations of the degenerate series are found. All *-representations of so'q(r,s) are separated in the set of irreducible representations obtained in the paper. 2007 Article Degenerate Series Representations of the q-Deformed Algebra so'q(r,s) / V.A. Groza // Symmetry, Integrability and Geometry: Methods and Applications. — 2007. — Т. 3. — Бібліогр.: 21 назв. — англ. 1815-0659 2000 Mathematics Subject Classification: 16B35; 16B37; 81R50 http://dspace.nbuv.gov.ua/handle/123456789/147377 en Symmetry, Integrability and Geometry: Methods and Applications Інститут математики НАН України |
| institution |
Digital Library of Periodicals of National Academy of Sciences of Ukraine |
| collection |
DSpace DC |
| language |
English |
| description |
The q-deformed algebra so'q(r,s) is a real form of the q-deformed algebra Uq'(so(n,C)), n = r + s, which differs from the quantum algebra Uq(so(n,C)) of Drinfeld and Jimbo. We study representations of the most degenerate series of the algebra so'q(r,s). The formulas of action of operators of these representations upon the basis corresponding to restriction of representations onto the subalgebra so'q(r) × so'q(s) are given. Most of these representations are irreducible. Reducible representations appear under some conditions for the parameters determining the representations. All irreducible constituents which appear in reducible representations of the degenerate series are found. All *-representations of so'q(r,s) are separated in the set of irreducible representations obtained in the paper. |
| format |
Article |
| author |
Groza, V.A. |
| spellingShingle |
Groza, V.A. Degenerate Series Representations of the q-Deformed Algebra so'q(r,s) Symmetry, Integrability and Geometry: Methods and Applications |
| author_facet |
Groza, V.A. |
| author_sort |
Groza, V.A. |
| title |
Degenerate Series Representations of the q-Deformed Algebra so'q(r,s) |
| title_short |
Degenerate Series Representations of the q-Deformed Algebra so'q(r,s) |
| title_full |
Degenerate Series Representations of the q-Deformed Algebra so'q(r,s) |
| title_fullStr |
Degenerate Series Representations of the q-Deformed Algebra so'q(r,s) |
| title_full_unstemmed |
Degenerate Series Representations of the q-Deformed Algebra so'q(r,s) |
| title_sort |
degenerate series representations of the q-deformed algebra so'q(r,s) |
| publisher |
Інститут математики НАН України |
| publishDate |
2007 |
| url |
http://dspace.nbuv.gov.ua/handle/123456789/147377 |
| citation_txt |
Degenerate Series Representations of the q-Deformed Algebra so'q(r,s) / V.A. Groza // Symmetry, Integrability and Geometry: Methods and Applications. — 2007. — Т. 3. — Бібліогр.: 21 назв. — англ. |
| series |
Symmetry, Integrability and Geometry: Methods and Applications |
| work_keys_str_mv |
AT grozava degenerateseriesrepresentationsoftheqdeformedalgebrasoqrs |
| first_indexed |
2025-07-11T01:59:03Z |
| last_indexed |
2025-07-11T01:59:03Z |
| _version_ |
1837314192546201600 |
| fulltext |
Symmetry, Integrability and Geometry: Methods and Applications SIGMA 3 (2007), 064, 12 pages
Degenerate Series Representations
of the q-Deformed Algebra so′
q(r, s)
Valentyna A. GROZA
National Aviation University, 1 Komarov Ave., 03058 Kyiv, Ukraine
E-mail: groza@i.com.ua
Received January 26, 2007, in final form April 18, 2007; Published online May 02, 2007
Original article is available at http://www.emis.de/journals/SIGMA/2007/064/
Abstract. The q-deformed algebra so′
q(r, s) is a real form of the q-deformed algebra
U ′
q(so(n, C)), n = r + s, which differs from the quantum algebra Uq(so(n, C)) of Drinfeld
and Jimbo. We study representations of the most degenerate series of the algebra so′
q(r, s).
The formulas of action of operators of these representations upon the basis corresponding
to restriction of representations onto the subalgebra so′
q(r)× so′
q(s) are given. Most of these
representations are irreducible. Reducible representations appear under some conditions for
the parameters determining the representations. All irreducible constituents which appear
in reducible representations of the degenerate series are found. All ∗-representations of
so′
q(r, s) are separated in the set of irreducible representations obtained in the paper.
Key words: q-deformed algebras; irreducible representations; reducible representations
2000 Mathematics Subject Classification: 16B35; 16B37; 81R50
1 Introduction
In this paper we consider most degenerate series representations of the q-deformed algebra
so′q(r, s), which is a real form of the complex q-deformed algebra U ′
q(so(n, C)) defined in [1].
The algebra U ′
q(so(n, C)) differs from the quantum algebra Uq(so(n, C)) defined by Drinfeld [2]
and Jimbo [3] (see also [4]). Drinfeld and Jimbo defined Uq(so(n, C)) by means of Cartan
subalgebras and root subspaces of the Lie algebra so(n, C). However, the Lie algebra so(n, C)
has a different structure based on the basis elements Ik,k−1 = Ek,k−1 −Ek−1,k (where Eis is the
matrix with elements (Eis)rt = δirδst). The q-deformation of this structure leads to the algebra
U ′
q(so(n, C)), determined in [1]. Later on it was shown that this q-deformation of so(n, C) is very
useful in many directions of contemporary mathematics. Namely, representations of the algebra
U ′
q(so(n, C)) and of its real forms are closely related to the theory of q-orthogonal polynomials
and q-special functions. Some ideas of such applications can be found in [5].
The algebra U ′
q(so(n, C)) (especially its particular case U ′
q(so(3, C))) is related to the algebra
of observables in 2+1 quantum gravity on the Riemmanian surfaces (see, for example, [6]).
A quantum analogue of the Riemannian symmetric space SU(n)/SO(n) is constructed by means
of the algebra U ′
q(so(n, C)) [7]. It is clear that a construction of a quantum analogue of some
pseudo-Riemannian symmetric spaces is connected with the q-deformed algebra so′q(r, s).
A q-analogue of the theory of harmonic polynomials (q-harmonic polynomials on quantum
vector space Rn
q ) is constructed by using the algebra U ′
q(so(n, C)). In particular, a q-analogue of
separations of variables for the q-Laplace operator on Rn
q is given by means of this algebra and
its subalgebras (see [8, 9]). The algebra U ′
q(so(n, C)) also appears in the theory of links in the
algebraic topology [10].
The representation theory of the q-deformed algebra U ′
q(so(n, C)) differs from that for the
Drinfeld–Jimbo algebra Uq(so(n, C)). One of these differences is related to the fact that the
mailto:groza@i.com.ua
http://www.emis.de/journals/SIGMA/2007/064/
2 V.A. Groza
Drinfeld–Jimbo algebra Uq(so(n, C)) admits the inclusion
Uq(so(n, C)) ⊃ Uq(so(n− 2, C))
and does not admit the inclusion
Uq(so(n, C)) ⊃ Uq(so(n− 1, C)).
The algebra U ′
q(so(n, C)) admits such an inclusion. This allows to construct Gel’fand–Tsetlin
bases for finite dimensional representations of U ′
q(so(n, C)) (see [1]).
As in the case of real forms of Drinfeld–Jimbo quantum algebras (see [11, 12, 13]) the real form
so′q(r, s) of U ′
q(so(r+s, C)) is defined by means of introducing a ∗-operation into U ′
q(so(r+s, C)).
When q → 1 then the q-deformed algebra so′q(r, s) turns into the universal enveloping algebra
U(sor,s) of the Lie algebra sor,s which corresponds to the pseudo–orthogonal Lie group SO0(r, s).
Representations of the algebra so′q(r, s), considered in this paper, are given by one continuous
parameter. These representations are q-deformations of the representations of the classical
Lie group SO0(r, s) considered in [14, 15] (see also [16, 17]). We derive several series of ∗-
representations of the algebra so′q(r, s). As in the case of the quantum algebra Uq(su1,1), the
algebra so′q(r, s) has the so-called strange series of ∗-representations, which is absent in the case
of the Lie group SO0(r, s). When q → 1 then matrix elements of operators of the strange series
representations tend to the infinity and representations become senseless.
Everywhere below we consider that q is a positive number. We also suppose that r > 2
and s > 2. Representations of degenerate series of the algebra so′q(r, 1) were considered in [18].
Representations of the algebra so′q(r, 2) were studied in [19]. In fact, we generalize a part of the
results of papers [18, 19].
It is well-known that the algebra U ′
q(so(n, C)) has finite dimensional irreducible representa-
tions of two types: representations of the classical type (at q → 1 they tend to the corresponding
representations of the Lie algebra so(n, C)) and representations of the non-classical type (there
exists no analogue of these representations in the case of so(n, C)). The algebra so′q(r, s) has no
degenerate series representations of the non-classical type. The reason is that the “compact”
algebras so′q(n) have no degenerate irreducible representations of the non-classical type.
2 The q-deformed algebra so′
q(r, s)
The algebra so′q(r, s) is a real form of the q-deformed algebra U ′
q(so(r + s, C)) which is separated
by the ∗-operation. The algebra U ′
q(so(r + s, C)) is defined in [1].
The classical universal enveloping algebra U(so(n, C)) is generated by the elements Ii,i−1,
i = 2, 3, . . . , n, that satisfy the relations
Ii,i−1I
2
i+1,i − 2Ii+1,iIi,i−1Ii+1,i + I2
i+1,iIi,i−1 = −Ii,i−1, (1)
I2
i,i−1Ii+1,i − 2Ii,i−1Ii+1,iIi,i−1 + Ii+1,iI
2
i,i−1 = −Ii+1,i, (2)
[Ii,i−1, Ij,j−1] = 0, |i− j| > 1 (3)
(they follow from the well-known commutation relations for the generators Iij of the Lie algebra
so(n, C)). In the approach to the q-deformed orthogonal algebra of paper [1], a q-deformation
of the associative algebra U(so(n, C)) is defined by deforming the relations (1)–(3). These
q-deformed relations are of the form
Ii,i−1I
2
i+1,i − aIi+1,iIi,i−1Ii+1,i + I2
i+1,iIi,i−1 = −Ii,i−1, (4)
I2
i,i−1Ii+1,i − aIi,i−1Ii+1,iIi,i−1 + Ii+1,iI
2
i,i−1 = −Ii+1,i, (5)
Degenerate Series Representations of the q-Deformed Algebra so′q(r, s) 3
[Ii,i−1, Ij,j−1] = 0, |i− j| > 1, (6)
where a = q1/2+q−1/2 and [·, ·] denotes the usual commutator. Obviously, in the limit q→1 for-
mulas (4)–(6) give relations (1)–(3). Remark that relations (4) and (5) differ from the q-deformed
Serre relations in the approach of Jimbo and Drinfeld to the quantum algebras Uq(so(n, C)) by
appearance of nonzero right hand sides and by possibility of reduction
U ′
q(so(n, C)) ⊃ U ′
q(so(n− 1, C)).
Below, by the algebra U ′
q(so(n, C)) we mean the q-deformed algebra defined by formulas (4)–(6).
The “compact” real form so′q(n) of the algebra U ′
q(so(n, C)) is defined by the involution
given as
I∗i,i−1 = −Ii,i−1, i = 2, 3, . . . , n. (7)
The “noncompact” real form so′q(r, n− r) of U ′
q(so(n, C)) is determined by the involution
I∗i,i−1 = −Ii,i−1, i 6= r + 1, I∗r+1,r = Ir+1,r. (8)
It would be more correct to use the notation U ′
q(soq(r, n− r)) for so′q(r, n− r). We use the last
notation since it is simpler.
The q-deformed algebra so′q(n) contains the subalgebra so′q(n − 1). This fact allows us to
consider Gel’fand–Tsetlin bases of carrier spaces of representations of so′q(n) [1]. The q-deformed
algebra so′q(n− r, r) contains the subalgebra so′q(n− r)× so′q(r).
3 Representations of so′
q(n)
Irreducible representations of the algebra so′q(r, s) are described by means of finite dimensional
irreducible representations of the subalgebras so′q(r) and so′q(s). Therefore, we describe repre-
sentations of so′q(n) which will be used below.
Irreducible finite dimensional representations of the algebra so′q(3) (belonging to the classical
type) are given by integral or half-integral nonnegative number l. We denote these representa-
tions by Tl. The carrier space of the representation Tl has the orthonormal basis
|m〉, m = −l,−l + 1, . . . , l,
and the operators Tl(I21) and Tl(I32) act upon this basis as (see [18])
Tl(I21)|m〉 = i[m]q|m〉, i =
√
−1, (9)
T (I32)|m〉 = d(m) ([l −m]q[l + m + 1]q)
1/2 |m + 1〉
− d(m− 1) ([l + m]q[l −m + 1]q)
1/2 |m− 1〉, (10)
where
d(m) =
(
[m]q[m + 1]q
[2m]q[2m + 2]q
)1/2
and [b]q is a q-number defined by the formula
[b]q =
qb/2 − q−b/2
q1/2 − q−1/2
.
4 V.A. Groza
Let us describe finite dimensional irreducible representations of the algebra so′q(n), n > 3,
which are of class 1 with respect to the subalgebra so′q(n− 1) [9]. As in the classical case, these
representations are given by highest weights (mn, 0, . . . , 0), where mn is a nonnegative integer.
We denote these representations by Tmn . Under restriction to the subalgebra so′q(n−1), the
representation Tmn contains (with unit multiplicity) those and only those irreducible represen-
tations Tmn−1 of this subalgebra for which we have
mn ≥ mn−1 ≥ 0.
Exactly in the same way as in the case of the classical group SO(n), we introduce the Gel’fand–
Tsetlin basis of the carrier space of the representation Tmn by using the successive reduction
so′q(n) ⊃ so′q(n− 1) ⊃ so′q(n− 2) ⊃ · · · ⊃ so′q(3) ⊃ so′q(2).
We denote the basis elements of this space by
|mn,mn−1,mn−2, . . . ,m3,m2〉,
where mn ≥ mn−1 ≥ mn−2 ≥ · · · ≥ |m2| and mn−i determines the representation Tmn−i of
so′q(n− i). With respect to this basis the operator T (In,n−1) of the representation Tmn of so′q(n)
is given by the formula
Tmn(In,n−1)|mn,mn−1, . . . ,m2〉
= ([mn + mn−1 + n− 2]q[mn −mn−1]q)1/2R(mn−1)|mn,mn−1 + 1, . . . ,m2〉 (11)
− ([mn + mn−1 + n− 3]q[mn −mn−1 + 1]q)1/2R(mn−1 − 1)|mn,mn−1 − 1, . . . ,m2〉,
where
R(mn−1) =
(
[mn−1 + mn−2 + n− 3]q[mn−1 −mn−2 + 1]q
[2mn−1 + n− 3]q[2mn−1 + n− 1]q
)1/2
.
The other operators T (Ii,i−1) are given by the same formulas with the corresponding change for
mn and mn−1 or at i = 3, 2 by the formulas for irreducible representations of the algebra so′q(3),
described above.
The representations Tmn are characterized by the property that under restriction to the
subalgebra so′q(n − 1) the restricted representations contain a trivial irreducible representation
of so′q(n−1) (that is, a representation with highest weight (0, 0, . . . , 0)). This is why one says that
the representation Tmn is of class 1 with respect to so′q(n−1). The irreducible representations Tmn
exhaust all irreducible representations of class 1 of the algebra so′q(n).
4 Representations of the degenerate principal series
We shall consider infinite dimensional representations of the associative algebra so′q(r, s). More-
over, we admit representations by unbounded operators. There exist non-equivalent definitions
of representations of associative algebras by unbounded or bounded operators (see [20, 21]).
In order to have a natural definition of a representation of so′q(r, s) we take into account the
following items:
(I) We shall deal also with ∗-representations (it is well-known that these representations
are an analogue of unitary representations of Lie groups). Therefore, for each representation
operator there should exist an adjoint operator. This means that a representation space have
to be defined on a Hilbert space.
Degenerate Series Representations of the q-Deformed Algebra so′q(r, s) 5
(II) Unbounded operators cannot be defined on the whole Hilbert space. However, existence
of an adjoint operator A∗ to an unbounded operator A means that the operator A must be
defined on an everywhere dense subspace in the Hilbert space.
(III) In order to be able to consider products of representation operators, there must exist
an everywhere dense subspace of the Hilbert space which enter to a domain of definition of each
representation operator.
Therefore, we give the following definition of a representation of so′q(r, s). A representation T
of the associative algebra so′q(r, s) is an algebraic homomorphism from so′q(r, s) into an algebra of
linear (bounded or unbounded) operators on a Hilbert spaceH for which the following conditions
are fulfilled:
a) the restriction of T onto the “compact” subalgebra so′q(r)×so′q(s) decomposes into a direct
sum of its finite dimensional irreducible representations (given by highest weights) with finite
multiplicities;
(b) operators of a representation T are determined on an everywhere dense subspace W of H,
containing all subspaces which are carrier spaces of irreducible finite dimensional representations
of so′q(r)× so′q(s) from the restriction of T .
In other words, our representations of so′q(r, s) are Harish-Chandra modules of so′q(r, s) with
respect to so′q(r)× so′q(s).
There exist different non-equivalent definitions of irreducibility of representations of associa-
tive algebras by unbounded operators (see [20]). Since unbounded representation operators are
not defined on all elements of the Hilbert space, then we cannot define irreducibility as in the
finite dimensional case. A natural definition is the following one. A representation T of so′q(r, s)
on H is called irreducible if H has no non-trivial invariant subspaces such that its closure does
not coincide with H. If operators of a representation T obey the relations
T (Ii,i−1)∗ = −T (Ii,i−1), i 6= r + 1, T (Ir+1,r)∗ = T (Ir+1,r) (12)
(compare with formulas (8)) on a common domain W of definition, then T is called a ∗-
representation.
To define a representation T of so′q(r, s) it is sufficient to give the operators T (Ii,i−1), i =
2, 3, . . . , r + s, satisfying relations (4)–(6) on a common domain of definition. Let us define
representations of so′q(r, s) belonging to the degenerate principal series. They are given by
a complex number λ and a number ε ∈ {0, 1}. We denote the corresponding representation
by Tελ. The space H(Tελ) of the representation Tελ is an orthogonal sum of the subspaces
V(m,0;m′,0), which are the carrier spaces of the finite dimensional representations of so′q(r)×
so′q(s) with highest weights (m,0;m′,0) such that
m + m′ ≡ ε (mod 2).
Here (m,0) and (m′,0), m ≥ 0, m′ ≥ 0, are highest weights of irreducible representations of
the subalgebras so′q(r) and so′q(r), respectively, and 0 denotes the set of zero coordinates (in the
case of the subalgebra so′q(3), 0 must be omitted). We assume that the basis vectors from (11)
are orthonormal in V(m,0;m′,0). Therefore, we have
H(Tελ) =
⊕
m+m′≡ε (mod 2)
V(m,0;m′,0), (13)
where we suppose that the sum means a closure of the corresponding linear span. A linear span
of the subspaces V(m,0;m′,0) determines an everywhere dense subspace on which all operators
of the representation Tελ are defined. Recall that we suppose that r > 2 and s > 2.
6 V.A. Groza
We choose in the subspaces V(m,0;m′,0) the orthonormal bases which are products of the
bases corresponding to irreducible representations of the subalgebras so′q(r) and so′q(s) intro-
duced in Section 3 (Gel’fand–Tsetlin bases). Elements of such bases are labeled by double
Gel’fand–Tsetlin patterns which will be denoted as
|M〉 = |m, k, j, . . . ;m′, k′, j′, . . . 〉. (14)
It is clear that entries of patterns (14) obey the following conditions:
m ≥ k ≥ j ≥ · · · , m′ ≥ k′ ≥ j′ ≥ · · · (15)
if r > 3 and s > 3. Thus, elements of the orthonormal basis of the carrier space of the
representation Tελ are labeled by all patterns (14) satisfying betweenness conditions (15) and
the equality m + m′ ≡ ε (mod 2).
In order to define the representations Tελ we give explicit formulas for the operators Tελ(Ii,i−1),
i = 2, 3, . . . , r + s. The operators Tελ(Ii,i−1), i = 2, 3, . . . , r, act upon basis elements (14) by
the formulas of Section 3 as operators of the corresponding irreducible representations of the
subalgebra so′q(r). It is clear that these operators act only upon entries k, j, . . . of vectors (14)
and do not change entries m, m′, k′, j′, . . . . The operators Tελ(Ii,i−1), i = r +2, r +3, . . . , r + s,
act upon basis elements (14) by formulas of Section 3 as operators of corresponding irreducible
representations of the subalgebra so′q(s). The operator Tελ(Ir+1,r) acts upon vectors (14) by the
formula
Tελ(Ir+1,r)|m, k, j, . . . ;m′, k′, j′, . . . 〉
= KmLm′ [λ + m + m′]q|m + 1, k, j, . . . ;m′ + 1, k′, j′, . . . 〉
−KmLm′−1[λ + m−m′ − s + 2]q|m + 1, k, j, . . . ;m′ − 1, k′, j′, . . . 〉
+ Km−1Lm′ [λ−m + m′ − r + 2]q|m− 1, k, j, . . . ;m′ + 1, k′, j′, . . . 〉
−Km−1Lm′−1[λ−m−m′ − r − s + 4]q|m− 1, k, j, . . . ;m′ − 1, k′, j′, . . . 〉, (16)
where
Km =
(
[m− k + 1]q[m + k + r − 2]q
[2m + r]q[2m + r − 2]q
)1/2
, (17)
Lm′ =
(
[m′ − k′ + 1]q[m′ + k′ + s− 2]q
[2m′ + s]q[2m′ + s− 2]q
)1/2
. (18)
So, this operator changes only the entries m and m′ in vectors (14).
To prove that these operators really give a representation of the algebra so′q(r, s) we have to
verify that the relations
Tελ(Ir+1,r)2Tελ(Ir,r−1)−
(
q
1
2 + q−
1
2
)
Tελ(Ir+1,r)Tελ(Ir,r−1)Tελ(Ir+1,r)
+ Tελ(Ir,r−1)Tελ(Ir+1,r)2 = −Tελ(Ir,r−1), (19)
Tελ(Ir+1,r)Tελ(Ir,r−1)2 −
(
q
1
2 + q−
1
2
)
Tελ(Ir,r−1)Tελ(Ir+1,r)Tελ(Ir,r−1)
+ Tελ(Ir,r−1)2Tελ(Ir+1,r) = −Tελ(Ir+1,r), (20)
as well as relations (19) and (20), in which Tελ(Ir,r−1) is replaced by Tελ(Ir+2,r+1), and the
relations
[Tελ(Ii,i−1), Tελ(Ij,j−1)] = 0, |i− j| > 1, (21)
where [·, ·] is the usual commutator, are fulfilled.
Degenerate Series Representations of the q-Deformed Algebra so′q(r, s) 7
Fulfilment of these relations can be shown by a direct calculation. Namely, we act by both
their parts upon vector (14), then collect coefficients at the same resulting basis vectors and
show that the relations obtained are correct. We do not give here these direct calculations.
Irreducibility of representations Tελ is studied by means of the following proposition.
Proposition 1. The representation Tελ is irreducible if each of the numbers [λ + m + m′]q,
[λ + m−m′ − s + 2]q, [λ−m + m′ − r + 2]q, [λ−m−m′ − r − s + 4]q in (16) vanishes only if
the vector on the right hand side of (16) with the coef f icient, containing this number, does not
belong to the Hilbert space H(Tελ).
This proposition is proved in the same way as the corresponding proposition in the classical
case (see Proposition 8.5 in Section 8.5 of [16]). Thus, the main role under studying irreducibility
of the representations Tελ is played by the coefficients at vectors in (16) numerated in Proposi-
tion 1.
5 Irreducibility
To study the representations Tελ we take into account properties of the function
w(z) = [z]q =
qz/2 − q−z/2
q1/2 − q−1/2
=
ehz/2 − e−hz/2
eh/2 − e−h/2
.
where q = eh. Namely, we have
w(z + 4πki/h) = w(z), w(z + 2πki/h) = −w(z), k are odd integers.
These relations mean that the following proposition holds:
Proposition 2. For arbitrary λ the pairs of representations Tελ and Tε,λ+4πki/h are coinciding
and the pairs of representations Tελ and Tε,λ+2πki/h are equivalent.
Therefore, we may consider only representations Tελ with 0 ≤ Im λ < 2π/h.
Most of the representations Tελ are irreducible. Nevertheless, some of them are reducible. As
in the classical case (see [14, 15, 16, 17]), reducibility appears because of vanishing of some of
the coefficients [λ+m+m′]q, [λ+m−m′−s+2]q, [λ−m+m′−r+2]q, [λ−m−m′−r−s+4]q
in formula (16). Using this fact we derive (in the same way as in the classical case; see [16]) the
following theorem:
Theorem 1. If r and s are both even, then the representation Tελ is irreducible if and only if λ
is not an integer such that λ ≡ ε (mod 2). If one of the numbers r and s is even and the other
one is odd, then the representation Tελ is irreducible if and only if λ is not an integer. If r and s
are both odd, then the representation Tελ is irreducible if and only if λ is not an integer or if λ
is an integer such that
λ ≡ ε (mod 2), 0 < λ < 1
2(r + s)− 2.
Irreducible representations Tελ admit additional equivalence relations.
Proposition 3. The pairs of irreducible representations Tελ and Tε,r+s−λ−2 are equivalent.
This equivalence is proved by constructing an explicit form of intertwining operators. These
operators are diagonal in the basis (14) and can be evaluated exactly in the same way as in the
classical case.
8 V.A. Groza
Sometimes it is useful to have the representations Tελ in somewhat different basis, namely,
in the basis
|m, k, j, . . . ; m′, k′, j′, . . . 〉′
which is related to the basis (14) by the formulas
|m0 + ε + i, k, j, . . . ;m0 − i, k′, j′, . . . 〉
=
m0∏
t=1
[−λ + ε + r + s + 2t− 4]1/2
q
m0∏
t=1
[λ + ε + 2t− 2]1/2
q
i∏
t=1
[−λ + ε + r + 2t− 2]1/2
q
i∏
t=1
[λ + ε− s + 2t]1/2
q
× |m0 + ε + i, k, j, . . . ;m0 − i, k′, j′, . . . 〉′,
|m0 + ε− i, k, j, . . . ;m0 + i, k′, j′, . . . 〉
=
m0∏
t=1
[−λ + ε + r + s + 2t− 4]1/2
q
m0∏
t=1
[λ + ε + 2t− 2]1/2
q
i∏
t=1
[λ + ε− s− 2t + 2]1/2
q
i∏
t=1
[−λ + ε + r − 2t]1/2
q
× |m0 + ε− i, k, j, . . . ;m0 + i, k′, j′, . . . 〉′.
In the new basis the operators Tελ(Ii,i−1), i 6= r + 1, have the same form as in the basis (14),
and the operator Tελ(Ir+1,r) is of the form
Tελ(Ir+1,r)|m, k, j, . . . ;m′, k′, j′, . . . 〉′
= KmLm′{[λ + m + m′]q[−λ + m + m′ + r + s− 2]q}1/2
× |m + 1, k, j, . . . ;m′ + 1, k′, j′, . . . 〉′
−KmLm′−1{[λ + m−m′ − s + 2]q[−λ + m−m′ + r]q}1/2
× |m + 1, k, j, . . . ;m′ − 1, k′, j′, . . . 〉′
+ Km−1Lm′{[λ−m + m′ − r + 2]q[−λ−m + m′ + s]q}1/2
× |m− 1, k, j, . . . ;m′ + 1, k′, j′, . . . 〉′
−Km−1Lm′−1{[λ−m−m′ − r − s + 4]q[−λ−m−m′ + 2]q}1/2
× |m− 1, k, j, . . . ;m′ − 1, k′, j′, . . . 〉′. (22)
Formulas (16) and (22) are used to select ∗-representations in the set of all irreducible rep-
resentations Tελ of the algebra so′q(r, s). This selection is fulfilled in the same way as in the
case of the q-deformed algebras so′q(2, 1) and so′q(3, 1) in [18], that is, by a direct check that the
relations (8) are satisfied. This selection leads to the following theorem.
Theorem 2. All irreducible representations Tελ of so′q(r, s) with λ = −λ + r + s− 2 are ∗-rep-
resentations (the principal degenerate series of ∗-representations). All irreducible representa-
tions Tελ with Im λ = π/h are ∗-representations (the strange series). If r and s are both even
or both odd, then all irreducible representations T0λ, 1
2(r + s) − 1 < λ < 1
2(r + s), for even
1
2(r + s) and all irreducible representations T1λ, 1
2(r + s) − 1 < λ < 1
2(r + s), for odd 1
2(r + s)
are ∗-representations (the supplementary series). If among the integers r and s one is odd and
another is even, then all irreducible representations Tελ, 1
2(r + s) − 1 < λ < 1
2(r + s − 1) are
∗-representations (supplementary series).
This theorem describes all ∗-representations in the set of irreducible representations Tελ.
However, there are equivalent representations in the formulation of Theorem 2. All possible
equivalences are given by equivalence relations described above or by their combinations (pro-
ducts).
Degenerate Series Representations of the q-Deformed Algebra so′q(r, s) 9
6 Reducible representations Tελ
Let us study a structure of reducible representations Tελ of the algebra so′q(r, s). Vanishing
of some coefficients in formula (16) or (22) leads to appearing of invariant subspaces in the
carrier space of the representation Tελ. Analysis of reducibility and finding of all irreducible
constituents in Tελ are done in the same way as in the classical case [15, 16, 17]. For this reason,
we shall formulate the results of such analysis without detailed proof.
Let us also note that, as in the classical case, reducible representations Tελ and Tε,−λ+r+s−2
contain the same (equivalent) irreducible constituents. This is easily seen from formula (22).
Studying the representations Tελ, we have to distinguish the cases of odd and even r and s since
in different cases a structure of reducible representations Tελ is different. Below, we investigate
all reducible representations Tελ (which are excluded in Theorem 1).
6.1 The case of even r and s
Let λ be an even integer in T0λ and an odd integer in T1λ. If λ ≤ 0 then in the carrier space
H(Tελ) of the representation Tελ there exist invariant subspaces
HF
λ =
⊕
m+m′≤−λ
V(m,0;m′,0), H0
λ =
⊕
λ−r+2≤m−m′≤−λ+s−2
V(m,0;m′,0),
where V(m,0;m′,0), is the subspace of H(Tελ), on which the irreducible representation of the
subalgebra so′q(r)× so′q(s) with highest weight (m,0;m′,0) is realized. On the subspace HF
λ the
finite dimensional irreducible representation of the algebra so′q(r, s) with highest weight (−λ,0)
is realized. An irreducible representation of so′q(r, s) is realized on the quotient space H0
λ/HF
λ .
We denote it by T 0
λ . A direct sum of two irreducible representations of so′q(r, s) is realized on the
quotient space H(Tελ)/H0
λ. One of them acts on the direct sum of the subspaces V(m,0;m′,0)
for which m −m′ > −λ − s − 2 (we denote it by T−
λ ). The second one acts on the direct sum
of the subspaces V(m,0;m′,0) for which m′ −m > −λ + r − 2 (we denote it by T+
λ ). Figures
showing distribution of subspaces V(m,0;m′,0) between the subspaces HF
λ , H0
λ, H+
λ , H−
λ (and
also for other cases, considered below) are the same as in the classical case and can be found
in [15], Chapter 8.
Let now λ be even in the representation T0λ and odd in the representation T1λ, and let
0 < λ ≤ 1
2(r + s) − 2. Then on the space H(Tελ) there exists only one invariant subspace H0
λ,
which is the orthogonal sum of the subspaces V(m,0 ;m′,0) for which
m−m′ ≤ −λ + s− 2, m′ −m ≤ −λ + r − 2.
The representation of so′q(r, s) on this subspace is irreducible (we denote it by T 0
λ ). The direct
sum of two irreducible representations of so′q(r, s) is realized on the quotient space H(Tελ)/H0
λ.
For one of them we have m−m′ > −λ+s−2 (we denote this irreducible representation by T−
λ ),
and for another one m′ − m > −λ + r − 2 (we denote it by T+
λ ). For λ = 1
2(r + s) − 2 the
range of values of m and m′ lies on one line in the coordinate space (m,m′). Physicists call such
subrepresentations ladder representations.
For λ = 1
2(r + s) − 1, the representation T0λ of so′q(r, s), if this number λ is even, and the
representation T1λ, if this number is odd, is a direct sum of two irreducible representations T−
λ
and T+
λ : for the first one we have m′−m ≤ −λ+r−2 and for the second one m−m′ ≤ −λ+s−2.
Since the reducible representations Tελ and Tε,−λ+r+s−2 contain the same irreducible con-
stituents, a structure of other reducible representations in this case is determined by that of the
representations considered above.
In the same way as in the classical case, it is easy to verify that the following irreducible
representations, considered here, are ∗-representations:
10 V.A. Groza
(a) all the representations T+
λ and T−
λ (the discrete series);
(b) the representation T 0
(r+s−4)/2.
Proposition 4. Irreducible representations Tελ and the irreducible representations T+
λ , T−
λ ,
T 0
λ of this subsection exhaust all infinite dimensional irreducible representations of the algebra
so′q(r, s) with even r and s which consist under restriction to so′q(r) × so′q(s) of irreducible rep-
resentations of this subalgebra only with highest weights of the form (m, 0, . . . , 0)(m′, 0, . . . , 0).
Moreover, these representations are pairwise non-equivalent.
This proposition is proved in the same way as in the case of the group SO0(r, s) (see Chapter 8
in [16]).
6.2 The case of even r and odd s
Let λ be a non-positive integer of the same evenness as m + m′ does. Then in the space H(Tελ)
of the reducible representation Tελ there exist two invariant subspaces
HF
λ =
⊕
m+m′≤−λ
V(m,0;m′,0), H0
λ =
⊕
m′−m≤−λ+r−2
V(m,0;m′,0).
The irreducible finite dimensional representation TF
λ of soq(r, s) with the highest weight (−λ,0)
is realized in the first subspace. In the quotient spaces H0
λ/HF
λ and H(Tελ)/H0
λ the irreducible
representations of so′q(r, s) are realized which will be denoted by T 1
λ and T+
λ respectively. So, in
this case the representation Tελ consists of three irreducible constituents.
If 0 < λ < 1
2(r+s)−2 and, besides, λ is an integer of the same evenness as m+m′ does, then
in H(Tελ) there exists only one invariant subspace H0
λ which is a direct sum of the subspaces
V(m,0;m′,0) for which m′ − m ≤ −λ + r − 2. The irreducible representations of the algebra
soq(r, s) are realized on H0
λ and H(Tελ)/H0
λ. We denote them by T 1
λ and T+
λ respectively.
If λ < 1
2(r + s) − 2 and, besides, λ is an integer such that λ ≡ (m + m′ + 1) (mod 2), then
only one invariant subspace exists in H(Tελ). This subspace is
H0
λ =
⊕
m−m′≤−λ+s−2
V(m,0;m′,0).
The irreducible representations of the algebra so′q(r, s) are realized on H0
λ and H(Tελ)/H0
λ. We
denote them by T 2
λ and T−
λ , respectively.
Since the reducible representations Tελ and Tε,−λ+r+s−2 contain the same irreducible con-
stituents, a structure of other reducible representations in this case is determined by that of the
representations considered above.
In the set of irreducible representations, considered here, only the representations T+
λ and T+
λ
are ∗-representations.
The case of odd r and even s is considered absolutely in the same way, interchanging the
roles of r and s as well as of m and m′. For this reason, we omit consideration of this case.
Proposition 5. Irreducible representations Tελ and the irreducible representations T+
λ , T−
λ , T 1
λ ,
T 2
λ of this subsection exhaust all infinite dimensional irreducible representations of the algebra
so′q(r, s) with even r and odd s which consist under restriction to so′q(r) × so′q(s) of irreducible
representations of this subalgebra only with highest weights of the form (m, 0, . . . , 0)(m′, 0, . . . , 0).
Moreover, these representations are pairwise non-equivalent.
Degenerate Series Representations of the q-Deformed Algebra so′q(r, s) 11
6.3 The case of odd r and s
If λ is a non-positive integer such that λ ≡ (m + m′) (mod 2), then in the space H(Tελ) of the
representation Tελ there exists only one invariant subspace HF
λ containing all those subspaces
V(m,0;m′,0) for which m + m′ ≤ −λ. On this invariant subspace the irreducible finite dimen-
sional representation of so′q(r, s) with the highest weight (−λ,0) is realized. On the quotient
space H(Tελ)/HF
λ the irreducible representation of so′q(r, s) is realized which is denoted by T 3
λ .
If λ ≤ 1
2(r + s) − 2 and, besides, λ is an integer such that λ ≡ (m + m′ + 1) (mod 2), then
in the space H(Tελ) there exists only one invariant subspace H0
λ containing all the subspaces
V(m,0;m′,0) for which
m′ −m ≤ −λ + r − 2, m−m′ ≤ −λ + s− 2.
An irreducible representation of so′q(r, s) is realized on H0
λ (we denote it by T 0
λ ). On the quotient
space H(Tελ)/H0
λ a direct sum of two irreducible representations of so′q(r, s) acts. For one of
them we have m′ −m ≤ −λ + r − 2, and for the other m−m′ ≤ −λ + s− 2. We denote these
irreducible representations by T−
λ and T+
λ , respectively.
If λ = 1
2(r+s)−1, then the representation T0λ for odd 1
2(r+s)−1 and the representation T1λ
for even 1
2(r + s)− 1 decompose into a direct sum of two irreducible representations of so′q(r, s)
(we denote them by T−
λ and T+
λ ). For the first representation we have m′ − m ≤ −λ + r − 2
and for the second one m−m′ ≤ −λ + s− 2.
Since the reducible representations Tελ and Tε,−λ+r+s−2 contain the same irreducible con-
stituents, a structure of other reducible representations in this case is determined by that of the
representations considered above.
In the set of irreducible representations, considered in this subsection, only the following ones
are ∗-representations:
(a) all the representations T+
λ and T−
λ (the discrete series);
(b) the representation T 0
(r+s−4)/2.
Proposition 6. Irreducible representations Tελ and the irreducible representations T+
λ , T−
λ , T 0
λ ,
T 3
λ of this subsection exhaust all infinite dimensional irreducible representations of the algebra
so′q(r, s) with odd r and s which consist under restriction to so′q(r) × so′q(s) of irreducible rep-
resentations of this subalgebra only with highest weights of the form (m, 0, . . . , 0)(m′, 0, . . . , 0).
Moreover, these representations are pairwise non-equivalent.
References
[1] Gavrilik A.M., Klimyk A.U., q-deformed orthogonal and pseudo-orthogonal algebras and their representa-
tions, Lett. Math. Phys. 21 (1991), 215–220.
[2] Drinfeld V.G., Hopf algebras and quantum Yang–Baxter equation, Sov. Math. Dokl. 32 (1985), 254–258.
[3] Jimbo M., A q-difference analogue of Uq(gl(N + 1)) and the Yang–Baxter equations, Lett. Math. Phys. 10
(1985), 63–69.
[4] Klimyk A.U., Schmüdgen K., Quantum groups and their representations, Springer, Berlin, 1997.
[5] Klimyk A.U., Kachurik I.I., Spectra, eigenvectors and overlap functions for representation operators of
q-deformed algebras, Comm. Math. Phys. 175 (1996), 89–111.
[6] Nelson J., Regge T., 2+1 gravity for genus s > 1, Comm. Math. Phys. 141 (1991), 211–223.
[7] Noumi M., Macdonald’s symmetric polynomials as zonal spherical functions on quantum homogeneous
spaces, Adv. Math. 123 (1996), 16–77.
[8] Noumi M., Umeda T., Wakayama M., Dual pairs, spherical harmonics and a Capelli identity in quantum
group theory, Compos. Math. 104 (1996), 227–277.
12 V.A. Groza
[9] Iorgov N.Z., Klimyk A.U., The q-Laplace operator and q-harmonic polynomials on the quantum vector
space, J. Math. Phys. 42 (2001), 1326–1345.
[10] Bullock D., Przytycki J.H., Multiplicative structure of Kauffman bracket skein module quantization,
math.QA/9902117.
[11] Twietmeyer E., Real forms of Uq(g), Lett. Math. Phys. 49 (1992), 49–58.
[12] Dobrev V.K., Canonical q-deformation of noncompact Lie (super)algebras, J. Phys. A: Math. Gen. 26
(1993), 1317–1329.
[13] Celegini E., Giachetti R., Reyman A., Sorace E., Tarlini M., SOq(n+1, n− 1) as a real form of SOq(2n, C),
Lett. Math. Phys. 23 (1991), 45–44.
[14] Raczka R., Limic N., Niederle J., Discrete degenerate representations of the noncompact rotation groups,
J. Math. Phys. 7 (1966), 1861–1876.
[15] Molchanov V.F., Representations of pseudo-orthogonal groups associated with a cone, Math. USSR Sbornik
10 (1970), 353–347.
[16] Klimyk A.U., Matrix elements and Clebsch–Gordan coefficients of group representations, Naukova Dumka,
Kiev, 1979.
[17] Howe R.E., Tan E.C., Homogeneous functions on light cone: the infinitesimal structure of some degenerate
principal series representations, Bull. Amer. Math. Soc. 28 (1993), 1–74.
[18] Gavrilik A.M., Klimyk A.U., Representations of q-deformed algebras Uq(so2,1) and Uq(so3,1), J. Math. Phys.
35 (1994), 3670–3686.
[19] Kachurik I.I., Klimyk A.U., Representations of the q-deformed algebra Uq(sor,2), Dokl. Akad. Nauk Ukrainy,
Ser. A (1995), no. 9, 18–20.
[20] Schmüdgen K., Unbounded operator algebras and representation theory, Birkhäuser, Basel, 1990.
[21] Ostrovskyi V., Samoilenko Yu., Introduction to the theory of representations of finitely presented ∗-algebras,
Reviers in Math. and Math. Phys. 11 (1999), 1–261.
http://arxiv.org/abs/math.QA/9902117
1 Introduction
2 The q-deformed algebra so'q(r,s)
3 Representations of so'q(n)
4 Representations of the degenerate principal series
5 Irreducibility
6 Reducible representations T
6.1 The case of even r and s
6.2 The case of even r and odd s
6.3 The case of odd r and s
References
|